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
BMC Med EthicsBMC Medical Ethics1472-6939BioMed Central London 1472-6939-6-81608350510.1186/1472-6939-6-8Research ArticleNeed for enforcement of ethicolegal education – an analysis of the survey of postgraduate clinical trainees Mayeda Mayumi [email protected] Kozo [email protected] Department of Health Science Policies Division of Research Development, Graduate School, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan2005 6 8 2005 6 8 8 4 10 2004 6 8 2005 Copyright © 2005 Mayeda and Takase; licensee BioMed Central Ltd.2005Mayeda and Takase; 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 number of medical lawsuits in Japan was between 14 and 21 each year before 1998, but increased to 24 to 35 per year after 1999. There were 210 lawsuits during this 10-year period. There is a need for skills and knowledge related to ethics, which is as fundamental to the practice of medicine as basic sciences or clinical skills. in Japan education in ethics is relatively rare and its importance is not yet recognized. Establishing ethics education using legal precedents, which has already been achieved in Western countries, will be a very important issue in Japan. In the present study, a questionnaire survey was conducted among graduate intern doctors, in order to investigate whether ethics education using precedents might have a positive effect in Japan.
Methods
In 2002, a questionnaire survey entitled Physicians' Clinical Ethics was carried out in a compulsory orientation lecture given to trainees before they started clinical practice in our hospital. The attendees at this lecture were trainees who came from colleges in various districts of Japan. During the lecture, 102 questionnaires were distributed, completed by attendees and collected. The recovery rate was 100%. The questionnaire consisted of 22 questions (in three categories), of which 20 were answered by multiple choices, and the other two were answered by description. The time required to complete the questionnaire was about 10 minutes.
Results
The recovered questionnaires were analyzed using statistical analysis software (SPSS for Windows, Release 10.07J-1/June/2000), in addition to simple statistical analysis. answers using multiple choices for the 20 questions in the questionnaire were input into SPSS. The principal component analysis was performed for each question. As a result, the item that came to the fore was "legal precedent". Since many intern doctors were interested in understanding laws and precedents, learning about ethical considerations through education using precedents might better meet with their needs and interests.
Conclusion
We applied a new method in which the results of principal component analysis and frequencies of answers to other questions were combined. From this we deduced that the precedent education used in Western countries was useful to help doctors acquire ethical sensitivity and was not against their will. A relationship was found between reading precedents and the influence of lawsuits, and it was thought that student participation-type precedent education would be useful for doctors in order to acquire ethical sensitivity.
==== Body
Background
The conditions for acquiring a physician's license in Japan are considered to be less strict than those in other advanced countries [1]. Consequently, only technical aspects have been emphasized in medical education, and paternalistic treatment has continued in clinical practice [2]. However, as the incidence of patient requests to physicians has increased following establishment of the self-decision-making rights of patients [3], contradictions that cannot be addressed adequately by conventional medical education or clinical settings have arisen [4]. In medical education, the amount of information learned by students has been increasing year by year, in line with scientific advances. Clinical medicine is thought to be shifting toward a patient-oriented contract, and in this model, a patient's right to autonomy as expressed by the term "informed decision" [5], and a physician's right to exercise his/her professional discretion are two of the main concepts [6].
Therefore, the six-year study period for medical students is not adequate, and medical education that allows sufficient time for learning knowledge and techniques is needed. The importance of culture as a part of medical education has been raised [7] and physicians' ethical views have become an issue. There is a need for skills and knowledge related to ethics, which is as fundamental to the practice of medicine as basic sciences or clinical skills [8].
Given this situation, the Medical Education Model Core Curriculum – Educational Contents Guideline (Core Curriculum) [9] – was announced by the Ministry of Education, Culture, Sports, Science and Technology when the postgraduate clinical trainee system became obligatory (in the 2004 financial year). The core curriculum, which is a standardized medical education guideline presented by the Ministry of Education, Culture, Sports, Science and Technology in Japan, shows the basic principles of educational reform [10] and promotes the guidelines as an educational form that does not have the borders which usually exist in colleges between culture, basic medicine and clinical medicine [11]. In particular, it deserves special attention that "medical ethics" was first established as a general education subject in the core curriculum [12]. To change the conventional cramming education to problem solving-type education [13], the core curriculum is expected to act as a catalyst for each college in introducing ethics (including law and law cases) as an elective.
Against this background, there has recently been an increase in medical lawsuits in Japan. The number of medical lawsuits in Japan was between 14 and 21 each year before 1998, but increased to 24 to 35 per year after 1999. There were 210 lawsuits during this 10-year period [14]. Possible factors associated with this increase in lawsuits are loss of rapport between the physician and patient [15,16], neglect of the physician's duty to explain [17], manipulation of medical records [18] and secretive behavior [19]. As medical technology becomes more highly developed, minor mistakes by physicians may result in serious harm to patients [20]. In addition, consistent differences in malpractice experience exist among medical schools [21]. To reduce these lawsuits, acquiring a balance in training in both technique and ethics at the stage of medical education might be necessary [22].
As mentioned above, if ensuring that medical students acquire a sense of ethics results in a decrease in medical lawsuits, then ethics education for medical students using legal precedents as subject matter might be the best option. Courses in medical ethics are becoming an integral part of the curricula for many medical schools in Europe. However, in Japan education in ethics is relatively rare and its importance is not yet recognized. Establishing ethics education using legal precedents, which has already been achieved in Western countries, will be a very important issue in Japan.
In the present study, a questionnaire survey was conducted among graduate intern doctors, in order to investigate whether ethics education using precedents might have a positive effect in Japan.
The trainee period is a critical time for fostering ethical reasoning [23]. The survey of awareness among intern doctors during this period of transition from medical student to physician may provide key points for pre-graduate and postgraduate education. There have been no previous reports about questionnaire surveys of postgraduate clinical trainees in Japan. The present study was conducted in order to gain a better understanding of the ethical sensitivity of postgraduate clinical trainees in Japan and reflect obtained results in medical education.
Methods
In 2002, a questionnaire survey entitled Physicians' Clinical Ethics was carried out in a compulsory orientation lecture given to trainees before they started clinical practice in our hospital (Appendix 1) [see Additional file 1]. "Clinical ethics" indicates "the personal (confidential) relations between doctors and patients", "clinical orders that doctors must carry out" and "morality to maintain these relations and orders". The attendees at this lecture were trainees who came from colleges in various districts of Japan. The study was approved by the Center for Postgraduate Education of Tokyo Medical and Dental University, in compliance with the internal regulations of the hospital.
During the lecture, 102 questionnaires were distributed, completed by attendees and collected. The recovery rate was 100%. The questionnaire consisted of 22 questions (in three categories), of which 20 were answered by multiple choices, and the other two were answered by description. The time required to complete the questionnaire was about 10 minutes. The recovered questionnaires were analyzed using statistical analysis software (SPSS for Windows, Release 10.07J-1/June/2000), in addition to simple statistical analysis. Correlation analysis was performed to evaluate the similarity among the items. In particular, principal component analysis was performed to find factors in the background of physicians' awareness. The principal component analysis was used to comprehend entire trends that could not be obtained through comparison of frequencies within each answer or through one-to-one correlation analysis. Only one component to indicate the entire trend was not calculated, but plural components were worked out in descending order of degrees to explain factors.
First, answers using multiple choices for the 20 questions in the questionnaire were input into SPSS. The principal component analysis was performed for each question. There were many questions that had various dispersed principal components and whose answering trend could not be explained. Interestingly, in respect of Question A ("Disclosure of medical record"), the entire trend could be explained by the top two extracted components. Consequently, the item that came to the fore was "legal precedent".
Results
Disclosure of medical records and law case education
As a first step, principal component analysis was performed as for the intern doctors' answers (multiple choice) to the question "Why do you think disclosure of medical records is required?"
The characteristics of the three components extracted from the answers were elicited in reference to question items. Component 1 was one of the top two components (with a cumulative variance of 49%) and was considered to reflect "eagerness to heal" because the two most frequent choices were "reflect self-helping efforts" and "to summarize medical records". Component 2 may reflect "distrust of physicians" because the frequencies of "distrust physicians" and "considering lawsuits" were high (see Figure 1).
Figure 1 Principle component analysis of the awareness of the reasons for disclosure of medical records. Principal component analysis was performed as for the intern doctors' answers (multiple choice) to the question "Why do you think disclosure of medical records is required?" To evaluate the relationship between components 1 and 2, they were placed in a 2 × 2 matrix, with attention paid to each component's strengths and weaknesses, and classified into (A), (B), (C) and (D).
In addition, to evaluate the relationship between components 1 and 2, they were placed in a 2 × 2 matrix, with attention paid to each component's strengths and weaknesses, and classified into (A), (B), (C) and (D) (see Figure 2). (A) indicated that "the disclosure was required since patients themselves had willingness to be cured or patients had distrust for doctors"; (B) indicated that "the disclosure was required only for patient's willingness to be cured"; (C) indicated that "the disclosure was required due to simple interests"; and (D) indicated that "the disclosure was required since patients had distrust for doctors".
Figure 2 Law cases of medical malpractice and results of principle component analysis of disclosure of medical records. A particularly strong trend could be observed in relation to answers (A), (B), (C) and (D) with respect to the answers to the question "Have you read literature concerning medical malpractice precedents?" Answers were: (A) in 22 subjects, (B) in 21, (C) in 24, and (D) in 35. In (A) to (D), the percentages of answers to the above question were expressed as pie graphs.
The following analysis was performed by combining the above results for (A), (B), (C) and (D) with answers to other questions. As a result, a particularly strong trend could be observed in relation to answers (A), (B), (C) and (D) with respect to the answers to the question "Have you read literature concerning medical malpractice precedents?" Answers were: (A) in 22 subjects, (B) in 21, (C) in 24, and (D) in 35. In (A) to (D), the percentages of answers to the above question were expressed as pie graphs. As a result, the interesting fact emerged that doctors who selected (D) (followed by (C), (B) and (A)) had more experience of reading legal precedents. This indicated that there was some relation between interns' reading legal precedents and their reason given for "Why do patients require the disclosure of medical records?" An arrow directed from "patients" to "doctors" in Figure 1 shows how doctors perceive patients' evaluations of doctors.
Patients' awareness
Subsequently, which answer ((A), (B), (C) or (D)) best reflects the thoughts of patients was investigated with reference to the reasons actually given by patients for requesting medical records disclosure. At 3-year intervals, The Japanese Ministry of Health, Labour, and Welfare produces a "patients' awareness survey", based on the contents/results of a study of treatment-receiving behavior.
I compared the 1999 version of the ministry's survey with "eagerness to heal" in this questionnaire survey (see Figure 3) [24]. With respect to the reason that outpatients wish to know the contents of their medical records, about 50% of patients selected "to deepen the understanding of treatment I receive". The second highest percentage of patients selected "to know the true disease name, disease condition, and treatment contents". Thus, patients' "eagerness to heal" was the reason given by most patients. Based on these results, the reasons given by patients differed from those given by doctors who selected (C) and (D) (i.e., the disclosure was required due to simple interests or because patients had distrust for doctors), and was close to those of doctors who selected (A) and (B) (i.e., the disclosure was required for patients' willingness to be cured). Therefore, in terms of mutual understanding between doctors and patients, answers (A) and (B) were preferable to (C) and (D).
Figure 3 Outpatient survey: Reasons for the request for disclosure of medical records (Treatment receiving behavior survey, fiscal 1999: Ministry of Health, Labour, and Welfare). With respect to the reason that outpatients wish to know the contents of their medical records, about 50% of patients selected "to deepen the understanding of treatment I receive". The second highest percentage of patients selected "to know the true disease name, disease condition, and treatment contents". Thus, patients' "eagerness to heal" was the reason given by most patients.
Lawsuits and eagerness to study
Subsequently, on the assumption that precedent education (i.e., ethical education based on precedents) was useful to educate intern doctors, we investigated which of answers (A) and (B) was preferable. Answers to the question "What influences do you think medical lawsuits have on physicians? [- An increase in eagerness to study]" were used (see Figure 4). Respondents had the following two choices: Medical lawsuits "do not increase physicians' eagerness to study" and "do increase physician's eagerness to study". These answers were combined with Figure 1 and further classified according to "trust or distrust of physicians" with cluster analysis.
Figure 4 Influences of medical lawsuits and law case education. Answers to the question "What influences do you think medical lawsuits have on physicians? The percentage of "do not increase eagerness to study" responses was higher for "trust of physicians", and that of "do increase eagerness to study" was higher for "undecided" and "distrust of physicians". Therefore, answer (A) was more desirable than (B) in terms of to "increase eagerness to study".
In cluster analysis, common points are extracted and grouped. As shown in the scatter diagram of factor scores (see Figure 4) for disclosure of medical records, the following three definite clusters emerged: "trust of physicians", "undecided", and "distrust of physicians". The percentage of "do not increase eagerness to study" responses was higher for "trust of physicians", and that of "do increase eagerness to study" was higher for "undecided" and "distrust of physicians". Therefore, answer (A) was more desirable than (B) in terms of to "increase eagerness to study". (A) can be defined as the following balanced state: as physicians read more cases, they increase their awareness of patients' eagerness to heal and distrust of physicians, and increase their eagerness to study and aim at reliable medical treatment to eliminate this distrust. Thus, law case education for trainees has favorable effects, and its incorporation into medical education should be useful.
Effects of law case education
The general effects of law case education were evaluated (see Figure 5). The survey showed that only a low percentage of the trainees had ever read "provisions of the medical practitioners law/medical service law" and "law cases of medical malpractice" in their entirety and with complete understanding. This may represent trainees' perceptions that reading law cases is difficult due to complex technical terms, long sentences, and obscure expressions. However, many trainees had some interest in "medical practice and laws" and "law cases of medical malpractice". Therefore, high-level law and law case education may not always be necessary for trainees who rarely read laws or cases. Since many intern doctors were interested in understanding laws and precedents, learning about ethical considerations through education using precedents might better meet with their needs and interests.
Figure 5 Trainees' awareness of laws and law cases. The general effects of law case education were evaluated (see Figure 5). The survey showed that only a low percentage of the trainees had ever read "provisions of the medical practitioners law/medical service law" and "law cases of medical malpractice" in their entirety and with complete understanding.
Law case education and the influences of medical lawsuits
We next considered what matters would be appropriate for precedent education. The purpose of precedent education is to strengthen the ethical sensitivity of doctors toward patients. As a first step, principal component analysis of trainee's answers to the question "What influences do you think medical lawsuits have on physicians?" was performed (see Figure 6). The analysis method was the same as that shown in Figure 1. However, since the subject was the attitude of doctors toward patients, an arrow is directed from "doctors" to "patients".
Figure 6 Principle component analysis of the influences of medical lawsuits on physicians. Principal component analysis of trainee's answers to the question "What influences do you think medical lawsuits have on physicians?" was performed. As a result of principal component analysis, two components were extracted. Component 1 was defined as "eagerness to perform treatment" and Component 2 as "distrust of patients".
As a result of principal component analysis, two components were extracted. Component 1 was defined as "eagerness to perform treatment" and Component 2 as "distrust of patients". To evaluate the relationship between components 1 and 2, they were classified according to their strength into the following four items: (a) eagerness to perform treatment and distrust of patients; (b) eagerness to perform treatment and trust of patients; (c) lack of interest in treatment but trust of patients; and (d) lack of interest in treatment and distrust of patients. When (a) to (d) were combined with the answers to the question "Have you ever read law cases of medical malpractice?", experience in reading cases increased in the order of (d), (c), (b), and (a) (see Figure 7). The results of the survey indicated that doctors who had read more precedents showed more distrust toward patients.
Figure 7 Principle component analysis of the influences of medical lawsuits on physician. When (a) to (d) were combined with the answers to the question "Have you ever read law cases of medical malpractice?", experience in reading cases increased in the order of (d), (c), (b), and (a). The results of the survey indicated that doctors who had read more precedents showed more distrust toward patients.
Discussion
We applied a new method in which the results of principal component analysis and frequencies of answers to other questions were combined. From this we deduced that the precedent education used in Western countries was useful to help doctors acquire ethical sensitivity and was not against their will. However, since precedent education could increase doctors' distrust toward patients, it might be necessary to be selective about the subjects of precedent education. (Figures 2 and 7 seem to indicate the same results at the first glance. However, doctors' awareness of patients' distrust toward doctors and doctors' aggravating distrust toward patients greatly differed in quality.)
If Western-style, precedent-based education, is introduced in Japan, the issues listed below need to be addressed, while Japanese precedents might also be helpful when considering educational content.
Complex and obscure expressions and technical terms are used in law cases in Japan. People (other than those who have a law education) cannot read and understand the cases' provisions or issues. This suggests that simply reading law in medical education is a problem. Student participation-type education, in which students learn not by simply reading and remembering the precedents and articles, but by comprehending particular cases, would be a useful way to introduce precedent education into medical education. We also believe that teaching using a Socratic method based on notable medical malpractice cases would be appropriate.
In Japan, paternalistic-like treatment continues to occur in clinical practice. It could be a relatively simple process to help doctors to understand the mental pain that patients experience and their distrust toward doctors, by carefully teaching them the processes of medical malpractice suits using past precedents. This might lead doctors to develop a deeper understanding of patients' standpoints.
Cases about accountability violation, which becomes problematic in many malpractice suits in Japan, might be suitable subject matter for education using precedents. For instance, in 2001, the Supreme Court decided (in a case about mammary amputation) that doctors have a responsibility to disclose even a treatment that has not yet been technically established in order to achieve accountability [25]. In the case, a patient with breast cancer had pleaded with a doctor not to remove the breast, and to perform breast conservation therapy. However, the doctor removed the entire breast and was found to have been derelict in accountability. In education using precedents it will be necessary for students to learn the importance of accountability.
Many patients feel dissatisfied with the treatment they receive from doctors. The facts and issues shown in precedent cases reflect this distrust of patients toward doctors, and can increase the distrust that doctors feel toward patients if they are read without appropriate instructions. Doctors can have deep distrust toward patients (see Figure 7a) even when patients have a strong willingness to be cured. Education needs to be directed at helping doctors gain the trust of patients (see Figure 2A) while achieving a balance between doctors' willingness to cure and their trust toward patients (see Figure 7b). That is, the education must increase the mutual trust between doctors and patients [26,27]. Illustrating the standpoints of each party, through the facts and issues shown in the precedents, would be the first step toward developing the trust that doctors feel toward patients.
Japanese students have difficulty with student participation-type lessons. To change the conventional cramming-type education to a more problem solving-type education is not contrary to trainees' wishes, based on the results of the questionnaire, and does not require high-level knowledge of law cases.
Another possible subject matter could be precedents about medical record alteration and concealment, which are often seen in Japan. In 2001, there was a case in which the medical records of a patient who died due to malpractice during an operation at the University Hospital (the case involved Tokyo Women's Medical University [28]) were altered. The alteration was systematically organized by the entire medical team, and the case had a significant impact on society. When an error is made in medical practice in Japan, the evidential facts tend to be hidden. It is necessary for students to learn that understanding the patient's viewpoint and adopting an honest attitude to malpractice would be in their own best interests.
Student participation-type education (i.e., education involving simulation, problem-based learning [29,30], role-playing [31] and skills laboratories) is rarely used in Japan. An education method for law and ethics in which a small group of intern doctors and teachers of ethics (who can play an important role in modeling the very nature of ethics) [32], maintain smooth communication and foster mutual understanding from the standpoints of doctors and patients, corresponds to the traditional custom of Asian countries including Japan, where ethical sensitivity is established in families and communities. This method might be a highly effective way for doctors to acquire ethical sensitivity.
Lectures in which "mock trials" are performed and discussed, with students playing the parts of the accuser, accused and judge, should be useful. The jury system has not yet been introduced in Japan, although it will be in the near future (although approximately 70% of citizens do not want to be part of a jury). Ideally, any simulated trials should incorporate juries.
Conclusion
The analysis was performed following an increasing incidence of medical lawsuits in Japan. The results of our survey indicated that precedent education performed in Western countries would also be useful in Japan, and would not go against the attitude of doctors toward precedents and ethics. The relationship between precedents and ethics came to the forefront in a questionnaire survey of graduate intern doctors, which had not previously been conducted in Japan, and the principal component analysis developed from a new viewpoint with [a narrower focus]. A relationship was found between reading precedents and the influence of lawsuits, and it was thought that student participation-type precedent education would be useful for doctors in order to acquire ethical sensitivity. From this study, certain educational devices using materials and reminders based on Japanese precedents could be proposed. These devices might be necessary when introducing precedent-based education in Japan.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
MM: Whole of the study and writing the manuscript.
KT: Review of questionnaire plan and manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Supplementary Material
Additional file 1
Questionnaire of physicians' clinical ethics
Click here for file
Acknowledgements
I thank Dr. Y. Tanaka and Dr. K. Fushimi (Tokyo Medical and Dental University) for helpful suggestions and caring out questionnaire and its coordination.
==== Refs
Raffel MW Comparative Health Systems 1984 Pennsylvania, U.S.A: The Pennsylvania State University Press
Ikezaki S Advocacy of a Mutual Participation Model that Values Doctor-patient Communication Medical Education (Japan) 2003 34 224
Morioka K A problem of a doctor The Japanese Red Cross Medical Journal 2002 54 51 (In Japanese)
Fukaya T Kakita A A countermeasure & a point of medical suit Medical risk management: countermeasures for preventing medical accidents 2003 Tokyo: Ishiyaku Publishers, Inc 156 (In Japanese)
Braddock CH 3rdEdwards KA Hasenberg NM Laidley TL Levinson W Informed decision making in outpatient practice: time to get back to basics JAMA 1999 282 2313 2320 10612318 10.1001/jama.282.24.2313
Tanaka T Definition of patient's right for autonomy and doctor's right for discretion Journal of Japan Association for Bioethics 2001 12 115 (In Japanese)
Kurokawa K Education for "professional" physicians toward the internationalization age in the 21st century Academic studies; Journal of Health Care and Society 2001 10 45 (In Japanese)
Braunack-Mayer AJ Gillam LH Vance EF Gillett GR Kerridge IH McPhee J Saul P Smith DE Wellsmore HM Koczwara B Rogers WA McNeill PM Newell CJ Parker MH Walton M Whitehall JS An ethics core curriculum for Australasian medical schools Med J Aust 2001 20; 175 205 10 11587281
The Ministry of Education, Culture, Sports, Science and Technology in Japan. Medical education model core curriculum (In Japanese)
Sato T What the core curriculum aims at Gendai Iryo 2002 34 86 (In Japanese)
Ito H Medical education and facing medical problems University Bulletin 2002 51 11 (In Japanese)
Gross ML Ethics education and physician morality Soc Sci Med 1999 49 329 342 10414818 10.1016/S0277-9536(99)00113-6
Imai M Life ethical education in medical faculties Kyouiku to Igaku 2002 50 27 34 (In Japanese)
Medical lawsuits in Japan (In Japanese)
Vincent C Young M Phillips A Why do people sue doctors? A study of patieits and relatives taking legal action Lancet 1994 343 1609 1613 7911925 10.1016/S0140-6736(94)93062-7
Shapiro RS Simpson DE Lawrence SL Talsky AM Sobocinski KA Schiedermayer DL A survey of sued and nonsked physicians and suing patients Arch Intern Med 1989 149 2190 2196 2802885 10.1001/archinte.149.10.2190
Faden RR Beauchamp TL A history and theory of informed 1986 New York: Oxford University Press
Mizunuma H Medical record- Description contents and probative value, production orders The Law Times Report 1989 686 104 (In Japanese)
Kaibara S Medical record disclosure and medical treatment side problem Jurist 1998 1142 37 41 (In Japanese)
Grady MF Why are people negligent? Technology, nondurable precautions, and the medical malpractice explosion Northwestern University Law Review 1988 82 293 333
Holm S Nielsen GH Norup M Vegner A Guldmann F Andreasen PH Changes in moral reasoning and the teaching of medical ethics Med Educ 1995 29 420 3 8594405
Fox RC Albrecht GL, Fitzpatrick R, Scrimshaw SC Medical Uncertainty Revisited Handbook of Social Studies in Health and Medicine 2000 London: SAGE Publications 409 425
Koh Y Residents' preparation for and ability to manage ethical conflicts in Korean residency programs Academic Medicine 2001 76 297 300 11242587
The Ministry of Health, Labour and Welfare, Japan. Treatment receiving behavior survey, fiscal 1999 (In Japanese)
Supreme Court Decision of 27 Dec 2001. The 576th (O) in 1998
Ong LM de Haes JC Hoos AM Lammes FB Doctor-patient communication, A review of the literature Soc Sci Med 1995 40 903 918 7792630 10.1016/0277-9536(94)00155-M
Henzen-Klemens I Lapinska E Doctor-patient interaction, patient's health behavior and effects of treatment Social Science & Medicine 1984 19 9 18 6474225 10.1016/0277-9536(84)90132-1
Tokyo District Court Decision of 15 July 2003. The 27357th (Wa) in 2000
Ito K Problem based learning tutorial Gendai Iryo 2002 34 87 (In Japanese)
Fujisaki K Medical interview and communication Gendai Iryo 2002 34 113 (In Japanese)
Otaki J Basic clinical technical education and its evaluation Gendai Iryo 2002 34 119 (In Japanese)
Sidney Bloch Teacher-clinicians are not always adequate role models MJA 2003 178 167 169 12580743
|
16083505
|
PMC1192799
|
CC BY
|
2021-01-04 16:31:58
|
no
|
BMC Med Ethics. 2005 Aug 6; 6:8
|
utf-8
|
BMC Med Ethics
| 2,005 |
10.1186/1472-6939-6-8
|
oa_comm
|
==== Front
BMC MicrobiolBMC Microbiology1471-2180BioMed Central London 1471-2180-5-441604579410.1186/1471-2180-5-44Research ArticleGenotyping of Mycobacterium tuberculosis clinical isolates in two cities of Turkey: Description of a new family of genotypes that is phylogeographically specific for Asia Minor Zozio Thierry [email protected] Caroline [email protected] Selami [email protected] Zeynep [email protected] Alpaslan [email protected] Riza [email protected] Maryse [email protected] Nalin [email protected] Christophe [email protected] Unité de la Tuberculose et des Mycobactéries, Institut Pasteur de Guadeloupe2 Laboratoire de la Tuberculose, Institut Pasteur de Bruxelles3 Department of Clinical Microbiology, Faculty of Medicine, Inonu University, Malatya, Turkey4 Department of Microbiology and Clinical Microbiology, Faculty of Medicine, Hacettepe University, Ankara, Turkey2005 26 7 2005 5 44 44 14 4 2005 26 7 2005 Copyright © 2005 Zozio 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
Population-based bacterial genetics using repeated DNA loci is an efficient approach to study the biodiversity and phylogeographical structure of human pathogens, such as Mycobacterium tuberculosis, the agent of tuberculosis. Indeed large genetic diversity databases are available for this pathogen and are regularly updated. No population-based polymorphism data were yet available for M. tuberculosis in Turkey, at the crossroads of Eurasia.
Results
A total of 245 DNAs from Mycobacterium tuberculosis clinical isolates from tuberculosis patients residing in Turkey (Malatya n = 147 or Ankara n = 98) were genotyped by spoligotyping, a high-throughput genotyping method based on the polymorphism of the Direct Repeat locus. Thirty-three spoligotyping-defined clusters including 206 patients and 39 unique patterns were found. The ST41 cluster, as designated according to the international SpolDB3 database project, represented one fourth and when gathered to three genotypes, ST53, ST50 and ST284, one half of all the isolates. Out of 34 clinical isolates harboring ST41 which were further genotyped by IS6110 and by MIRU-VNTR typing, a typical 2-copy IS6110-RFLP pattern and a "215125113322" MIRU-VNTR pattern were observed among 21 clinical isolates. Further search in various databases confirms the likely Turkish-phylogeographical specificity of this clonal complex.
Conclusion
We described a new phylogeographically-specific clone of M. tuberculosis, designated LAM7-TUR. Further investigations to assess its frequency within all regions of Turkey and its phylogeographical origin and phylogenetic position within the global M. tuberculosis phylogenetic tree will shed new light on its endemicity in Asia Minor.
==== Body
Background
Turkey is a large and densely populated country (area.: 780 576 km2, population: around 71 millions). In Turkey, tuberculosis remains an important public health concern with a case notification rate of 26.2/100.000 inhabitants in 2002 [1]. In many of the studies, social aspect of TB is also underestimated. The men/women sex ratio shows a proportion much more important of men (3x) than women which suffers from the disease. Besides, the global drug resistance rates in men are twice higher than that of women. The 17–39 year group accounts for 80% of the studied strains, which shows that a young male population is specifically concerned by tuberculosis [1]. Contrary to the picture of most European countries, where the last years have seen a tremendous amount of studies describing the population-based genetic structure of Mycobacterium tuberculosis, no data of this kind are yet available for Turkey, with the exception of a recent paper describing the diversity of phospholipases genes in 106 clinical isolates of M. tuberculosis [2].
From a geographical, historical and anthropological aspect, Turkey is a link between Europe and Asia, an early region of human settlement, located in the Western part of Eurasia [3]. Various cultural and anthropological influences of early civilizations have left complex scars, making from Turkey an anthropologically rich area. More evidence about the unique nature and complexity of human genetics in Anatolia has been discovered, whether based on mitochondrial or on Y chromosome diversity [4-7]. The existence of a hypothetical proto-Indo-European language, whose link to Anatolia appears likely, is another issue that reinforces the importance of Turkey in human history, especially of the "Indo-European" lineages [8].
Molecular characterization studies on M. tuberculosis in Turkey until recently focused on the use of the "gold-standard" IS6110-RFLP method and on the description and characterization of multi-drug-resistance rates and their mechanisms, two issues of great importance for Public-Health [9-14]. However, to the best of our knowledge, no population-based data were available on the genetic diversity of M. tuberculosis in Turkey. The goal of this study was to get an initial insight into the biodiversity of M. tuberculosis in Turkey by studying a set of 245 DNAs from Anatolia by spoligotyping, a high throughput technique for which large polymorphism databases have been created [15]. These DNAs originated from as many clinical isolates from tuberculosis patients resident in Turkey. We show in this paper that a single genotype or "clonal complex" (ST41) accounts for one fourth of the total TB cases. A second genotype, ST284, may also bear some Eastern Mediterranean or Asian specificity but remains to be more exhaustively investigated. The predominance of a specific clonal complex in Turkey may argue in favour of a long-lasting presence of tuberculosis in this country and open new avenues of investigation to better understand why and how such a clone became predominant.
Results
Result of the genotyping analysis by spoligotyping
The global structure of the Mycobacterium tuberculosis population by spoligotyping is shown on the Figure 1. Our results provide a first raw information on the distribution of the spoligotypes within two cities of Turkey. One spoligotype, ST41, is highly frequent. This spoligotype had already been detected in SpolDB3 and designated as LAM7, given the absence of spacer 21–24 in this superfamily of genotypes [16]. Another genotype, ST284 (frequent MIRU-VNTR : 223323153322), which remains undesignated for the time-being, is also frequent. We identified 37 different STs in the set of DNAs originating from Malatya (n = 147) and 42 genotypes in the set of DNAs originating in Ankara (n = 98). These results suggest that the genetic diversity in Ankara is superior to the one found in Malatya, which would be logical given that Ankara is a larger city and the administrative capital of Turkey. The data were further compared to the spolDB4 database [17]. Nine spoligotypes from Malatya were truly unique ("orphans") whereas 3 were in cluster with other clinical isolates found in the database: ST1936 was created by match with a clinical isolate from Sweden; ST 1937 (internal cluster) was created by a match between two clinical isolates T53 and T92, which are presumed to be linked to ST41; ST1938 was created by a match between two clinical isolates from Turkey (N500, N502) and one from Indonesia. The study shows that more than 58% of the patients in Malatya and 38% in Ankara, were gathered in only four different STs which are ST41 (LAM7-TUR family, 50/245 : 21%), ST53 (ill-defined T1 superfamily, 40/245 : 16.3%), ST50 (Haarlem 3 family, 13/245 : 5.3%), and ST284 (undesignated, 14/245 : 5.7%). It is likely that these tuberculosis genotypes are representative of strains that were introduced in Turkey at a distant past. If we compare the distribution of these genotypes between the rest of the world and Turkey, we note that ST41 prevails largely in Turkey (21% versus 0.35% in SpolDB4) whereas ST53 and ST50 are distributed equally world-wide and in Turkey. ST284 is poorly prevalent in the world database (0.1%) whereas highly frequent in our study, which also suggests a strong local phylogeographical specificity.
Result of the genotypic analysis by IS6110-RFLP typing
The full results describing the complete molecular epidemiological analysis of the 145 clinical isolates, representative of as many patients from Malatya will be reported elsewhere. However, when we looked at the IS6110-RFLP results on the isolates bearing ST41, we confirmed that most of them harbored a similar IS6110 pattern based on a 2-copy signature, at around 2.1 kb and 2.8 kb respectively (Figure 2). Another subgroup of 5 clinical isolates harbored a slightly different 2-banded profile (2.1 and 4.8 kb; Figure 2). The description of TB genotypes harboring two copies of IS6110 was already reported in previous studies [13,18].
Result of the genotypic analysis by MIRU-VNTR typing
A total of 33 clinical isolates belonging to the ST41 were further genotyped by MIRU-VNTR typing. Our aim was to apply this highly discriminant method to further distinguish between isolates with ST41 spoligotypes. Results are shown in Table 1. A main pattern, found in 19 clinical isolates was observed (215125113322). Another well represented variant pattern was found in 8 clinical isolates (214125113322). When a minimum parsimonious tree was built by minimum spanning tree, a unique clonal complex was observed (Figure 2).
Result of other databases search
The 215125113322 pattern was introduced into the SITVIT1 database (spoligo-MIRU-VNTR international database, to be described elsewhere). Two isolates previously reported by L. Cowan (USA08096990178) and J. Driscoll (USA012004S00232) bearing the same ST41 spoligo and the same MIRU-VNTR pattern were identified. This MIRU-VNTR pattern was designated as VNTR-international-type (VIT) number 310. The second most represented allele 214125133322 bore the VIT number 194. When the clinical isolates bearing this MIRU-VNTR type value were compared to the IS6110-RFLP pattern of the same isolates (done in blind), an homogenous cluster of four fully identical IS6110-RFLPs with a double band at 2.2–2.8 kb was detected (T25, T18, T49, T66, cf. Figure 1 boxed), thus suggesting the full linkage between the two markers.
When a search was made against another database, the New-York state spoligotyping database, (version 2005 March 1st) maintained at the Wadworth Center in Albany, NY, we observed that ST41 (spoligopattern NYS_00232 in the NY database) was also found in 28 more clinical isolates in the USA, among which 22 were from Turkish clinical isolates from the city of Samsun, on the border of the Black Sea (J Driscoll and A Sanic, personal communication).
Discussion
Tuberculosis remains a great public health concern in Turkey. The resistance to antituberculosis drugs, which represents a specific threat and as such deserves much attention, was recently the focus of many investigations in Turkey [9-14], however, molecular epidemiology is of more recent interest [13,18]. Indeed, the unraveling of the effect of genetic variability of M. tuberculosis on the presentation of the disease remains a challenging poorly investigated issue, which consists in understanding why a strain may become prevalent in certain communities [19]. Recent results obtained on the polymorphims of genes known to be involved in pathogenicity and virulence (phospholipases) may create a bridge between pathogenicity and population genetics studies [2]. However, the population-based genetic landscape of tuberculosis biodiversity was, to the best of our knowledge, unknown before this study and as such deserved such an attempt to define which genotypes are responsible of the TB cases in Turkey, a subject which, given the highly complex anthropological structure of Turkey, is of great interest for clinical scientists, bacteriologists, and evolutionary biologists.
A total of 245 DNAs extracted from M. tuberculosis clinical isolates from TB Turkish patients were genotyped by spoligotyping. A major genotype, as revealed by spoligotyping ST41, which misses spacer 20–24, 26–27 and 33–36, and had been previously described under the designation of "LAM7", represented up to one fourth of all TB isolates. When these genotypes were further investigated by IS6110-RFLP or by the highly discriminant MIRU-VNTR technique, highly similar profiles were obtained suggesting that these strains define a true genotype family or clonal complex. This genotype is likely to be identical to the one described in Table 2 by IS6110-RFLP and pTBN12 in a recent paper [13]. However, given the highly discriminative power of pTBN12 as a second genotyping method, and the difficulties to compare these patterns, a total of 7 subclusters were described initially (Ia to Ig). How these previous results correlate to MIRU-VNTR-based or spoligotyping-based clustering remains to be further investigated.
The finding of another genotype ST284 that was already detected in SpolDB3 but without origin of potential phylogeographical specificity is intriguing. This genotype is currently under investigation and is also found to be present in Bulgaria (T. Zozio et al., unpublished obervations). Whether this genotype also bears a larger Eastern-Mediterranean or Middle Eastern phylogeographical specificity remains speculative for the time-being.
The incidence of tuberculosis in Turkey was recently estimated around 26.6 new cases per 100.000 inhabitants [1]. For a city such as Malatya (853.658 inhabitants), which has a slightly superior incidence (32/100.000), the total estimated number of new cases per year would be n = 272. Thus, our sampling (n = 145) represents the equivalent of a quarter of a two-year recruitment, which, we assume, is fully representative of the genetic diversity in Malatya. In Ankara, for which the recruitment was less important, similar genotyping results were obtained. In a third city from the border of the Black Sea (Samsun), similar results were also obtained on 100 DNAs by an independent team (A. Sanic and J. Driscoll, personal communication). Thus the genotyping results obtained in Ankara and Malatya seems to be quite representative of Anatolia, suggesting that the ST41/VIT310 and ST41/VIT196 could represent traces of a contemporary and/or historically endemic/epidemic clone in Anatolia. If further investigations on isolates from the Aegean, Mediterranean and eastern sides of Turkey confirm the prevalence of the ST41 genotype, and provided that it is really ancient, one may expect that its distribution will vary depending on the human population structure in Turkey. Thus, the observed geographical variations in the frequency distribution of ST41 may allow to precisely define its presumed origin. However one should be extremely cautious with such historical inferences. Indeed, one should not forget that epidemics by their bursting nature, may rapidly promote the replacement of genotypes by others and that recent human migrations do complexify the issue [20]. Turkey is a country where ancient Central Asian and European civilizations can be seen. Preliminary phylogenetical results (not shown) suggest that the LAM7-TUR genotype family of M. tuberculosis may be related to the large LAM9 superfamily of genotypes; however, another spoligotype, ST353, could also be the ancestor type of ST41 and cannot be excluded at this step as a potential ancestor of ST41. Further studies using combined MIRU-VNTR-spoligotyping will facilitate the finding of the ancestor clone of the LAM7-TUR family.
Conclusion
We described a new phylogeographically-specific clone of M. tuberculosis, designated LAM7-TUR. Further investigations to assess its frequency within all regions of Turkey and its phylogeographical origin and phylogenetic position within the global M. tuberculosis phylogenetic tree will shed new light on its endemicity in Asia Minor.
Abbreviations
ST : Shared type, LAM : Latino-American and Mediterranean, MIRU-VNTR : Mycobacterial Interspersed Repetitive Unit-Variable Number of DNA Tandem Repeats, TUR : Turkey.
Methods
Studied Population
Turkey counted in 2003 more than 71 million inhabitants leaving in an urban zone (64,9 %) twice more than those leaving in a rural setting (35,1 %). In 2000 incidence of tuberculosis in Turkey was about 26,6 cases for 100000 inhabitants [2]. Sixty-four percent of the patients within this study were men (mean age = 35) in between 21 to 64 years (patient data from Malatya only, n = 147). The diagnostic of pulmonary tuberculosis was done for the 147 patients for whom 147 clinical isolates were identified. All the patients were resident in Malatya except for 2 individuals who lived in Adiyaman. The recruitment covers the 1998–2004 period.
Spoligotyping
The Direct Repeat (DR) locus, which consists in multiple direct variable repeats (DVRs) [21] was assessed by the PCR-based reverse-line-blot nucleic acid hybridization method called "spacer oligonucleotide typing" or spoligotyping [22,23]. Spoligotyping was done at the Pasteur Institute of Guadeloupe using home-made membranes and results entered in Excel spreadsheets, Bionumerics (Applied Maths, Sint-Martens-Latem, Belgium), and Taxotron (PAD Grimont, Taxolab, Institut Pasteur, Paris).
IS6110-RFLP
IS6110-Restriction-Fragment-Length-Polymorphism was done using the standardized method [24]. The genotyping was done on the Malatya's isolates only in Dr. Durmaz's laboratory in Turkey. Results were analyzed using H37Rv as international standard and comparison was done using Bionumerics (Applied Maths, Sint-Martens-Latem, Belgium). Strain H37Rv was used as the reference standard for IS6110-RFLP. Results were exported to Taxotron as a Molecular weight text file, a pairwise distance matrix was built using the Dice Index, and this file was summed and averaged to a similar pairwise distance matrix of the spoligotyping results built using the Jaccard Index, to produce the results shown in Figure 2 (combined numerical analysis, Taxotron's manual).
MIRU-VNTR-typing (Mycobacterial-Interspersed-Repetitive-Units-Variable-Number of Tandem-Repeats typing)
MIRU-VNTRs were amplified from 12 genomic loci using 4 different multiplex PCRs with the previously described fluorescent primers, except that Hex labeling was replaced by Vic labeling [25]. Amplification was performed with HotStartTaq polymerase (Qiagen) using the same cycling conditions as in [25], except that 30 cycles were used instead of 40. Two μL of PCR products were mixed with 10 μL of formamide and 0.2 μL of MapMarker1000 ladder (bioventures). DNA fragments were separated by capillary electrophoresis using the ABI Prism 3100- Avant Genetic Analyzer (Applied Biosystems) as described in [26]. Sizing of the PCR fragments and assignment of the various MIRU-VNTR alleles were done using the GeneScan and customized Genotyper software packages (PE Applied Biosystem). MIRU-VNTR typing was done at the Pasteur Institute in Brussels.
Phylogeny reconstruction: Taxotron (numerical taxonomy) and Bionumerics, Minimum Spanning Tree (population modeling)
The Pairwise distance between clinical isolates was computed using the 1-Jaccard (1-Sj) index for the spoligotyping method [27] and using the Dice Index for IS6110-RFLP results [28]. The UPGMA algorithm (unweighted pair-group method using arithmetic averages) was used for clustering [29]. Distance-based methods are indeed fairly popular methods that have proved to be very useful to define some major phylogeographical clades within the M. tuberculosis complex [30,31]. The Bionumerics software (version 3.5) (Applied Maths, Sint Martens-Latem, Belgium) was used to reconstruct the hypothetical evolution of the ST41VIT310 clonal complex, following the user's manual (Figure 3).
Database Search
A search was done using two Polymorphisms databases. SITVIT1 (which was designed and maintained at the Pasteur Institute of Guadeloupe, query done on March 1st, to be described elsewhere) and the New York State spoligotyping database (version 2005 March 1st), which was developed and maintained by Dr. J. Driscoll at the Wadsworth Centers.
Authors' contributions
CS contacted ZS in Ankara and suggested a collaborative work. AA worked in collaboration with ZS to isolate and characterize the M. tuberculosis clinical isolates from the Ankara region, prepared quality DNA and send it to CS. To increase recruitment and representativity of the sampling, another collaborative work was simultaneously launched between NR and RD. RD and SG have isolated the strains from Malatya, characterized it, extracted the DNAs, done IS6110-RFLP typing in RD's laboratory and send DNAs to NR. TZ did the spoligotyping experiments of all the data shown here in NR's laboratory with auditing of the results by CS, at the Institut Pasteur of Guadeloupe. CA did the MIRU-VNTR-typing experiments in MF's laboratory in Brussels on DNAs aliquoted and sent by TZ. CS conceived the study, participated to its design, coordinated the action, interpreted the results with TZ and CA, built the figures and tables, wrote the paper with TZ.
Acknowledgements
We are very grateful to Dr. A. Sanic and Dr. J. Driscoll for communication of unpublished results on genotyping of Mycobacterium tuberculosis clinical isolates from the region of Samsun (n = 100 spoligotypes). Dr. J. Driscoll and the network of co-investigators of the multicentric SpolDB4 project are also warmly acknowledged for continuous participation to the various database construction projects. This work was supported by the "Réseau International des Instituts Pasteur et Instituts Associés", Institut Pasteur, Paris.
Figures and Tables
Figure 1 UPGMA tree constructed by numerical analysis of spoligotyping data using the Taxotron software (P.A.D. Grimont, Taxolab, Institut Pasteur, Paris). Remarkable genotypes are shown in bold at the extreme right of the picture (ST n°). The size of the characters is proportional to the frequency of genotypes. Dotted lines and some families (Haarlem1 and 3 as H1 and H3 and the newly "LAM7-TUR" family) are shown at the left of the dendrogram. Unique patterns are designated either as "orphan" or with their ST n° if already described in the SpolDB4 database.
Figure 2 Partial UPGMA tree constructed by combined numerical analysis of IS6110-RFLP and spoligotyping data using Taxotron, and built on isolates bearing ST41 spoligotypes and variants. A: unrooted dendrogram with dissimilarity index scale, B: IS6110-RFLP patterns, C: spoligotyping pattern (binary representation), D: MIRU genotyping values for ST41 clinical isolates. Note the association of value 4 of MIRU10 and the IS6110-RFLP cluster defined by isolates T25, T18, T49, T66 and the observation of a double band at 2.2 kb.
Figure 3 Minimum spanning tree built using the Bionumerics sofware on ST41-Mirutyped DNAs (Version 3.5, Applied Maths, Sint-Marteen-Latem, Belgium). 3A : tree panel window. 3B: complex panel window showing the MIRU value of the presumed origin of the tree. Genotypes are arbitrarily designated by letters. B : 215125113322 is suggested as the original genotype. For each designation, the number of cases is indicated (ex: 22 cases for genotype B). The number on the branch indicates the change required to go from one allele to another : exemple: a change on the 3rd MIRU on genotype B from an allelic value = 5 to a value = 4 provides genotype F.
Table 1 MIRU results for 36 isolates with ST41
origin clinical isolate n° MIRU typing result Octal spoligo result
Malatya N2 215125113322 777777404760771
Malatya N9 215125113322 777777404760771
Malatya N16 215125113322 777777404760771
Malatya N30 214125113322 777777404760771
Malatya N44 215125113322 777777404760771
Malatya N45 215125113422 777777404760771
Malatya N49 214125113322 777777404760771
Malatya N50 215125113522 777777404760771
Malatya N83 215125113322 777777404760771
Malatya N88 215125113322 777777404760771
Malatya N154 215125113322 777777404760771
Malatya N271 215125113422 777777404760771
Malatya N503 214125113322 777777404760771
Malatya N505 214125113322 777777404760771
Malatya T6 214125113222 777777404760771
Malatya T9 215125113322 777777404760771
Malatya T10 215126113322 777777404760771
Malatya T13 215125113322 777777404760771
Malatya T18 214125113322 777777404760771
Malatya T22 215125113322 777777404760771
Malatya T25 214125113322 777777404760771
Malatya T29 215125113322 777777404760771
Malatya T33 215125113322 777777404760771
Malatya T37 213125113322 777777404760771
Malatya T49 214125113322 777777404760771
Malatya T62 215125113322 777777404760771
Malatya T66 214125113322 777777404760771
Malatya T72 215125113322 777777404760771
Malatya T82 215125113322 777777404760771
Malatya T83 215125113322 777777404760771
Malatya T84 215125113322 777777404760771
Malatya T85 215125113322 777777404760771
Malatya T91 215125113322 777777404760771
Ankara 1 215125113322 777777404760771
Ankara 7 215125113322 777777404760771
Ankara 8 215125113322 777777404760771
==== Refs
Three years in Health services. From May 1999 to May 2002 : Ministry of Health of Turkey, Ankara 2002
Talarico S Durmaz R Yang Z Insertion- and deletion-associated genetic diversity of Mycobacterium tuberculosis phospholipase C-encoding genes among 106 clinical isolates from Turkey J Clin Microbiol 2005 43 533 8 15695641 10.1128/JCM.43.2.533-538.2005
Cavalli-Sforza LL Menozzi P Piazza A Luca Cavalli-Sforza L, Menozzi P, Piazza A Chapter 4: Asia, 4.1: General Introduction, Geography and Environment The History and Geography of Human genes 1994 Abridged paperback Princeton, New Jersey: Princeton University Press 195 197
Comas D Calafell F Mateu E Perez-Lezaun A Bertranpetit J Geographic variation in human mitochondrial DNA control region sequence: the population history of Turkey and its relationship to the European populations Mol Biol Evol 1996 13 1067 77 8865661
Di Benedetto G Erguven A Stenico M Castri L Bertorelle G Togan I Barbujani G DNA diversity and population admixture in Anatolia Am J Phys Anthropol 2001 115 144 56 11385601 10.1002/ajpa.1064
Mergen H Oner R Oner C Mitochondrial DNA sequence variation in the Anatolian Peninsula (Turkey) J Genet 2004 83 39 47 15240908
Cinnioglu C King R Kivisild T Kalfoglu E Atasoy S Cavalleri GL Lillie AS Roseman CC Lin AA Prince K Excavating Y-chromosome haplotype strata in Anatolia Hum Genet 2004 114 127 48 Epub 2003 Oct 29 14586639 10.1007/s00439-003-1031-4
Gray RD Atkinson QD Language-tree divergence times support the Anatolian theory of Indo-European origin Nature 2003 426 435 9 14647380 10.1038/nature02029
Bengisun JS Karnak D Palabiyikoglu I Saygun N Mycobacterium tuberculosis drug resistance in Turkey, 1976–97 Scand J Infect Dis 2000 32 507 10 11055655 10.1080/003655400458785
Cavusoglu C Hilmioglu S Guneri S Bilgic A Characterization of rpoB mutations in rifampin-resistant clinical isolates of Mycobacterium tuberculosis from Turkey by DNA sequencing and line probe assay J Clin Microbiol 2002 40 4435 8 12454132 10.1128/JCM.40.12.4435-4438.2002
Kart L Altin R Tor M Gulmez I Oymak SF Atmaca HM Erdem F Antituberculosis drug resistance patterns in two regions of Turkey: a retrospective analysis Ann Clin Microbiol Antimicrob 2002 1 6 12537590 10.1186/1476-0711-1-6
Saribas Z Kocagoz T Alp A Gunalp A Rapid detection of rifampin resistance in Mycobacterium tuberculosis isolates by heteroduplex analysis and determination of rifamycin cross-resistance in rifampin-resistant isolates J Clin Microbiol 2003 41 816 8 12574290 10.1128/JCM.41.2.816-818.2003
Durmaz R Gunal S Yang Z Ozerol IH Cave MD Molecular epidemiology of tuberculosis in Turkey Clin Microbiol Infect 2003 9 873 7 14616712 10.1046/j.1469-0691.2003.00654.x
Ozturk CE Balbay OA Kaya D Ceyhan I Bulut I Sahin I The Resistance to Major Antituberculous Drugs of Mycobacterium tuberculosis Strains Isolated from the Respiratory System Specimens of Tuberculosis Patients in Duzce, Turkey Jpn J Infect Dis 2005 58 47 9 15728994
Filliol I Driscoll JR van Soolingen D Kreiswirth BN Kremer K Valetudie G Dang DA Barlow R Banerjee D Bifani PJ Snapshot of moving and expanding clones of Mycobacterium tuberculosis and their global distribution assessed by spoligotyping in an international study J Clin Microbiol 2003 41 1963 70 12734235 10.1128/JCM.41.5.1963-1970.2003
Filliol I Driscoll JR van Soolingen D Kreiswirth BN Kremer K Valétudie G Anh DD Barlow R Banerjee D Bifani PJ Global distribution of Mycobacterium tuberculosis spoligotypes Emerg Inf Dis 2002 8 1347 1350
Brudey K An Appraisal of the geographic prevalence of Major Genotypic Families of Mycobacterium tuberculosis complex through the updated SpolDB4 database Submitted for publication to the Emerg Inf Dis journal (+65 co-authors)
Durmaz R Ozerol IH Durmaz B Gunal S Senoglu A Evliyaoglu E Primary drug resistance and molecular epidemiology of Mycobacterium tuberculosis isolates from patients in a population with high tuberculosis incidence in Turkey Microb Drug Resist 2003 9 361 6 15000742 10.1089/107662903322762798
Malik AN Godfrey-Faussett P Effects of genetic variability of Mycobacterium tuberculosis strains on the presentation of disease Lancet Infect Dis 2005 5 174 83 15766652
Vasileva D Bulgarian Turkish emigration and return Int Migr Rev 1992 26 342 52 12285857
Groenen PMA Bunschoten AE van Soolingen D van Embden JDA Nature of DNA polymorphism in the direct repeat cluster of Mycobacterium tuberculosis; application for strain differentiation by a novel typing method Mol Microbiol 1993 10 1057 1065 7934856
Kamerbeek J Schouls L Kolk A van Agterveld M van Soolingen D Kuijper S Bunschoten A Molhuizen H Shaw R Goyal M Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology J Clin Microbiol 1997 35 907 914 9157152
van Embden JDA van Gorkom T Kremer K Jansen R van der Zeijst BAM Schouls LM Genetic variation and evolutionary origin of the Direct repeat locus of Mycobacterium tuberculosis complex bacteria J Bacteriol 2000 182 2393 2401 10762237 10.1128/JB.182.9.2393-2401.2000
van Embden JDA Cave MD Crawford JT Dale JW Eisenach KD Gicquel B Hermans P Martin C McAdam R Shinnick TM Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: recommendations for a standardized methodology J Clin Microbiol 1993 31 406 409 8381814
Supply P Lesjean S Savine E Kremer K van Soolingen D Locht C Automated high-throughput genotyping for study of global epidemiology of Mycobacterium tuberculosis based on mycobacterial interspersed repetitive units J Clin Microbiol 2001 39 3563 71 11574573 10.1128/JCM.39.10.3563-3571.2001
Allix C Supply P Fauville-Dufaux M Utility of fast mycobacterial interspersed repetitive unit-variable number tandem repeat genotyping in clinical mycobacteriological analysis Clin Infect Dis 2004 39 783 9 Epub 2004 Aug 27 15472808 10.1086/423383
Jaccard P Nouvelles recherches sur la distribution florale Bull Soc Vaud Sci Nat 1908 44 223 270
Dice LR Measures of the amount of ecologic association between species Ecology 1945 26 297 302
Sneath PHA Sokal R Numerical taxonomy: the principles and practices of classification 1973 San Francisco: WH Freeman and Co
van Soolingen D Qian L de Haas PEW Douglas JT Traore H Portaels F Qing HZ Enkhsaikan D Nymadawa P van Embden JDA Predominance of a Single Genotype of Mycobacterium tuberculosis in Countries of East Asia J Clin Microbiol 1995 33 3234 3238 8586708
Sola C Filliol I Legrand E Mokrousov I Rastogi N Mycobacterium tuberculosis phylogeny reconstruction based on combined numerical analysis with IS1081, IS6110, VNTR and DR-based spoligotyping suggests the existence of two new phylogeographical clades J Mol Evol 2001 53 680 689 11677628 10.1007/s002390010255
|
16045794
|
PMC1192800
|
CC BY
|
2021-01-04 16:03:40
|
no
|
BMC Microbiol. 2005 Jul 26; 5:44
|
utf-8
|
BMC Microbiol
| 2,005 |
10.1186/1471-2180-5-44
|
oa_comm
|
==== Front
BMC MicrobiolBMC Microbiology1471-2180BioMed Central London 1471-2180-5-461607639210.1186/1471-2180-5-46Research ArticleExpressed sequence tags from the oomycete fish pathogen Saprolegnia parasitica reveal putative virulence factors Torto-Alalibo Trudy [email protected] Miaoying [email protected] Kamal [email protected] Mark E [email protected] West Pieter [email protected] Sophien [email protected] Department of Plant Pathology, The Ohio State University, Ohio Agricultural Research and Development Center, Wooster, Ohio, USA2 National Center for Genome Resources, Santa Fe, New Mexico, USA3 Aberdeen Oomycete Group, College of Life Sciences and Medicine, University of Aberdeen, Foresterhill, Scotland, United Kingdom2005 2 8 2005 5 46 46 6 6 2005 2 8 2005 Copyright © 2005 Torto-Alalibo 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 oomycete Saprolegnia parasitica is one of the most economically important fish pathogens. There is a dramatic recrudescence of Saprolegnia infections in aquaculture since the use of the toxic organic dye malachite green was banned in 2002. Little is known about the molecular mechanisms underlying pathogenicity in S. parasitica and other animal pathogenic oomycetes. In this study we used a genomics approach to gain a first insight into the transcriptome of S. parasitica.
Results
We generated 1510 expressed sequence tags (ESTs) from a mycelial cDNA library of S. parasitica. A total of 1279 consensus sequences corresponding to 525944 base pairs were assembled. About half of the unigenes showed similarities to known protein sequences or motifs. The S. parasitica sequences tended to be relatively divergent from Phytophthora sequences. Based on the sequence alignments of 18 conserved proteins, the average amino acid identity between S. parasitica and three Phytophthora species was 77% compared to 93% within Phytophthora. Several S. parasitica cDNAs, such as those with similarity to fungal type I cellulose binding domain proteins, PAN/Apple module proteins, glycosyl hydrolases, proteases, as well as serine and cysteine protease inhibitors, were predicted to encode secreted proteins that could function in virulence. Some of these cDNAs were more similar to fungal proteins than to other eukaryotic proteins confirming that oomycetes and fungi share some virulence components despite their evolutionary distance
Conclusion
We provide a first glimpse into the gene content of S. parasitica, a reemerging oomycete fish pathogen. These resources will greatly accelerate research on this important pathogen. The data is available online through the Oomycete Genomics Database [1].
==== Body
Background
Water molds such as Saprolegnia and Aphanomyces species are responsible for devastating infections on fish in aquaculture, fish farms and hobby fish tanks [2,3]. Members of the genus Saprolegnia cause saprolegniosis, a disease that is characterized by visible white or grey patches of filamentous mycelium on the body or fins of freshwater fish [2]. The oomycete Saprolegnia parasitica is economically one of the most important fish pathogens, especially on salmon and trout species. It causes tens of million dollar losses to aquaculture business worldwide, notably in Scotland, Scandinavia, Chile, Japan, Canada, and the USA [4,5]. S. parasitica infections are second only to bacterial diseases. In Japan, there is an annual mortality rate of 50% in coho salmon and elver due to S. parasitica infections [5-8]. In the United States, "winter kill" in catfish caused by Saprolegnia results in financial loses of up to 50%, which represents an economic loss of $40 million [6]. In Scotland, saprolegniosis also causes significant losses with the main problem occurring in salmon hatcheries.
Previously, Saprolegnia infections were kept under control with malachite green, an organic dye that is very efficient at killing the pathogen. However, since 2002 the use of malachite green has been banned around the world, due to its carcinogenic and toxicological effects. This has resulted in dramatic recrudescence of Saprolegnia infections. Therefore, there is an urgent need for novel alternative methods of management of Saprolegniosis.
Saprolegnia is often considered an opportunistic pathogen that is saprotrophic and necrotrophic [6]. However, it has become apparent that some S. parasitica strains are highly virulent and able to cause primary infections on salmon [3,9,10]. Infections occur on both eggs and fish. On eggs the disease is manifested by profuse mycelial growth on the egg surface resulting in rapid death. On fish, Saprolegnia invades epidermal tissues and can infect the entire surface of the body [11]. It causes cellular necrosis as well as dermal and epidermal damage, which ultimately leads to death by heamodilution [5,12]. Severe Saprolegnia infections result in lethargic behaviour, loss of equilibrium and commonly death of the fish [12,13].
Oomycete species can be pathogenic on plants, insects, crustaceans, fish, vertebrate animals, and various microorganisms [14,15]. Oomycetes, including Saprolegnia, have many fungus-like characteristics, but are not true fungi. A number of studies have indicated that they should be classified with the golden-brown algae and diatoms as stramenopiles [16-18]. This implies that oomycetes evolved genetic and biochemical mechanisms for interaction with animals and plants that are different from those of true fungi [14]. Indeed, oomycetes have several clearly defined developmental stages that are not found in fungal pathogens. For example, Saprolegnia species have a complex life cycle that includes both sexual and asexual reproduction. The asexual spore or sporangium is formed at the end of hyphal cells and can release many motile primary zoospores [6]. The primary zoospores swim only for a short time before they encyst and release a secondary zoospore. Secondary zoospores are motile for a longer period and are the main infection spore [5,11]. Secondary zoospores are able to encyst and release new zoospores several times. This process is called "polyplanetism" [19], and may have evolved to allow the zoospores to have several attempts to locate and infect a host [6]. Uniquely within the class of oomycetes, secondary zoospores of Saprolegnia can possess hairs that are thought to be required for attachment to the host [11,19]. For example zoospores of S. parasitica have long hooked hairs that are believed to increase efficiency of the attachment to the fish hosts [12,19].
Although different in their selection of host organisms, plant and fish pathogenic oomycetes have many features in common. Evidently, the formation of specialised spore structures including zoospores, sporangia and oospores are similar. Also, infection strategies are comparable to some extent, involving encystment and attachment of zoospores on host surfaces, and penetration of host tissues. Furthermore, it is hypothesized that similar to biotrophic plant pathogenic oomycetes, such as Peronospora and several Phytophthora species, suppression of host defenses is likely to play a critical role in Saprolegnia pathogenesis. Host defense suppression by oomycetes remains poorly understood and only a few pathogen molecules that suppress host defenses have been identified in pathogenic oomycetes [20-22]. There is intriguing evidence that Saprolegnia-infected fish appear to be immuno-compromised [23,24]. Possibly, virulence factors secreted by the pathogen might account for the immuno-suppression and the lack of an effective response to pathogen infection.
Despite the huge economic importance of animal pathogenic oomycetes, such as S. parasitica, very little is known about the fundamental molecular mechanisms underlying development, pathogenicity and host specificity [14]. A thorough understanding of the basic molecular processes in Saprolegnia, the nature of the interactions with its hosts, and the identification of genes and proteins involved in these processes, could lead to novel control strategies that increase fish health, reduce disease losses and increase profits. In this study we used a genomics approach to gain a first insight into the transcriptome of S. parasitica. We generated random cDNA sequences (expressed sequence tags or ESTs) from a cDNA library of S. parasitica to identify genes that inform us about the biology of this organism and that could be involved in pathogenicity. We provide an overview of the identified sequences as well as a detailed description of a number of notable cDNAs. The data is available through a publicly accessible website as part of the Oomycete Genomics Database (OGD) [1].
Results
cDNA library and sequencing
We constructed a unidirectional cDNA library using mRNA isolated from mycelium of ATCC90214, a S. parasitica strain isolated from diseased salmon [25]. Mycelium was obtained from nutrient deprived 29-day-old in vitro culture. In other oomycetes, such as P. infestans, similar treatments promote the expression of stress-related genes and possibly mimic infection conditions [26,27]. A total of 2296 sequencing reactions corresponding to the 5' end of the cDNA insert were performed. Of these, 2102 gave readable sequences. The sequences and the quality (phred) scores were fed into NCGR's X Genome Initiative (XGI) [28] annotation pipeline and subjected to further quality controls [29]. 1510 ESTs remained after vector and low quality sequences were removed. Of these, 5% were assessed to be in reverse orientation based on the occurrence of at least eight consecutive A residues within the first 38 bp. Following additional quality screening and assembly, 1279 consensus sequences (so-called unigenes) were obtained consisting of 1146 singletons and 133 consensus with two or more ESTs. In total, 525,944 bp of assembled sequences were obtained corresponding to an average consensus sequence length of 411 bp. At 61%, the GC content of the assembled sequences was relatively high and similar to the GC content reported for Phytophthora spp. (57–58%) [14,30].
Sequence annotation
The 1279 consensus sequences were annotated using the methods implemented in the XGI pipeline (see methods). A total of 609 sequences (48%) showed significant similarities to known protein sequences (E value < 10-5) based on BLASTX searches, 398 (31%) gave significant hits to protein motifs in the BLOCKS+ database, and 600 (47%) gave hits to the InterPro protein motif database with at least one of the 12 algorithms implemented in InterProScan. Among these, InterPro database searches with HmmPfam revealed 340 hits. In total, 585 consensus sequences (46%) could be assigned identities based on the Gene Ontology (GO) Consortium (see methods). The differences between the different analyses are expected for such bioinformatics annotations and reflect, among other things, differences in sensitivity between the various programs. A total of 70 sequences were positive with the PexFinder algorithm and are candidate for carrying signal peptides and encoding extracellular proteins.
Taxonomic identity of the homologs of S. parasitica cDNAs
Considering the classification of oomycetes as stramenopiles, it was interesting to systematically examine the taxonomic identity of the homologs of S. parasitica cDNAs. To this end, we took advantage of the availability of several eukaryotic genomes to compile a data set of 270334 proteins covering six major phyla of eukaryotes: fungi, animals, plants, alveolates, discicristates, and heterokonts. The data included the complete proteomes of at least one species for all phyla except for discicristates (see methods). We used BLASTX to compare the 1279 S. parasitica unigenes to these eukaryotic proteins. In total, 715 sequences showed no significant hits (E value > e-5). Of the 582 sequences that showed significant hits, 32% (185) had the top hit to a diatom protein confirming the affinity between diatoms and oomycetes as stramenopiles. In contrast, only 56 (about 10%) sequences had a fungal protein as a top hit.
Phylogenetic analyses
We also exploited the sequence data to examine phylogenetic affinities between S. parasitica, three Phytophthora spp., and the diatom Thalassiosira pseudonana. We performed reciprocal BLAST searches to identify a common set of conserved protein sequences between the five species. Multiple alignments of the conserved portions of 18 different proteins covering 2533 amino acids were concatenated and used in phylogenetic analyses. The obtained tree clearly supported a monophyletic relationship between the four oomycetes, and consistent with published phylogenies of Phytophthora [31,32] suggested that P. sojae and P. ramorum are more closely related (Fig. 1). Average amino acid identity among sequences of the Phytophthora spp. was 93% (range, 91.1–95.0%). In contrast, average amino acid identity between S. parasitica and the three Phytophthora spp. was 77% (range, 76.4–77.9%). Average amino acid identity between the diatom and the four oomycetes was 66.7% (range, 66.2–67.5%).
Most represented protein domains
As described above, searches of the InterPro database with HmmPfam revealed 340 hits among the 1297 consensus sequences. We classified the InterPro domains based on their abundance (Table 1). The 340 hits corresponded to 198 different InterPro domains. Of these, 29 domains were represented three times or more (range, 3–22). By far the most abundant domain was IPR000719 for eukaryotic protein kinase, which was represented 22 times. Other domains that function in signal transduction, such as ankyrin-repeat, G-protein, Ras GTPase, myb DNA binding, were also relatively abundant. Other well-represented domains comprised fungal type I cellulose-binding domain (IPR000254), PAN domain (IPR003014), and papain cysteine protease family (IPR000668).
Fungal type I cellulose binding domain
Four sequences showed similarities to fungal-type I cellulose binding domain (CBD) (InterPro domain IPB000254). Three of these occurred in cDNAs predicted to encode extracellular proteins. Two cDNAs, Sp_002_00594 and Sp_001_01439, encoded putative proteins with two CBDs. The full length sequence of cDNA SPM5F8 (Sp_001_01439) was obtained (GenBank accession number DQ143887). This cDNA contained an ORF of 306 bp corresponding to a protein of 101 amino acids. SignalP [33] analysis of the predicted protein identified a 20-amino acid signal peptide with a significant mean S value of 0.93. Domain IPB000254 has been mainly reported in fungi [34], but also occurs in one protein from Ectocarpus siliculosus Virus EsV-1 (Phycodnaviridae), a viral pathogen of brown algae [35]. To determine the extent to which the CBD occurs in oomycetes, we performed iterative BLAST searches of all publicly available Phytophthora sequences using the S. parasitica CBD sequences. In total, 12 different sequences similar to the S. parasitica CBDs were recovered from P. infestans (3), P. sojae (4), and P. ramorum (5). The 18 oomycete CBD sequences aligned perfectly over a 34 amino acid region. Multiple alignments of the oomycete CBDs with the well-studied CBD of cellobiohydrolase I (Cel6A) of Trichoderma resei [36,37] suggested that the major features of the domain are conserved in oomycetes (Fig. 2). Amino acid residues defining the Cel6A domain including the four cysteine backbone, as well as a glutamine (Gln32) and the three tyrosines (Tyr3, Tyr29 and Tyr30) that are important for binding to cellulose, were frequently conserved. Nonetheless, tyrosines, particularly Tyr3 and Tyr29, were often replaced by other aromatic residues, such as tryptophane and phenylalanine, as observed for various fungal CBDs, such as endoglucanase I of T. resei [38].
CBEL-like sequences
Seven distinct cDNA sequences showed similarity to the Cellulose Binding, Elicitor and Lectin-like protein (CBEL), a 34-kDa extracellular glycoprotein that was first isolated from Phytophthora parasitica [39,40]. Five of these CBEL-like cDNAs were also predicted to encode extracellular proteins based on PexFinder analyses. CBEL has a dual function: (1) it is required for attachment to plant surfaces, (2) it elicits necrosis and defense gene expression in tobacco plants [39,40]. CBEL contains two regions with similarity to the PAN module (InterPro: IPR000177), a conserved domain that includes the Apple domain and functions in protein-protein or protein-carbohydrate interactions [41]. The similarity to CBEL in the identified S. parasitica cDNAs centered on the PAN module regions. Three of the seven S. parasitica cDNAs contained two PAN-like domains while the other four had a single domain. We used these sequences to survey the available Phytophthora sequences for additional PAN/CBEL-like domains using iterative BLAST searches. In total, 42 PAN/CBEL-like domains were identified in 28 putative proteins of P. infestans, P. sojae, P. ramorum, and P. parasitica suggesting that this domain is widely distributed in oomycetes (Fig. 3). Multiple alignment of the 52 oomycete CBEL-like domains revealed a conserved pattern centered around a conserved core of six cysteines (Fig. 3B).
Glycosyl hydrolases
Six cDNAs showed similarity to various classes of glucanases. One of these, Sp_001_01488, showed significant similarity to microbial endo-1,3-β-glucanases (glycosyl hydrolase family 17), as well as high similarity to the recently described gene, piendo1, from P. infestans [42]. The 1258 bp insert of a full-length cDNA (SPM16A2) corresponding to this endo-1,3-β-glucanase was fully sequenced (GenBank accession number AY974332). The sequence revealed a single ORF of 1197 bp encoding a predicted protein of 398 amino acids with 38% identity to PIENDO1. SignalP [33] analysis of the predicted protein identified a 19-amino acid signal peptide with a significant mean S value of 0.94. Previously, phylogenetic analyses of several Phytophthora hydrolytic enzymes, including PIENDO1, revealed unexpected affinity to fungal proteins [27,42-44]. Indeed, BLASTP searches of the protein encoded by SPM16A2 showed significant similarities to fungal and bacterial proteins (top hits with E value = 2e-21 and 3e-22, respectively) but none to other eukaryotic proteins.
Proteases
We found a set of 12 cDNAs with similarity to aspartyl (2), serine (3), and cysteine (7) proteases among the annotated sequences of S. parasitica. The sequence of SPM3B2, a full length cDNA corresponding to unigene Sp_004_00851 was obtained (GenBank accession number AY974331). This cDNA encoded a putative protein of 379 amino acids. BLASTP searches of the MEROPS database [45] revealed significant similarity to pepsin aspartic proteases such as cathepsin D (MEROPS Family A01, E value = 1e-72 for best hit). SignalP [33] analysis of the predicted protein identified a 17-amino acid signal peptide with a significant mean S value of 0.73. We also determined the full length sequence of cDNA SPM9F1 (Sp_001_01152) (GenBank accession number AY974330). This cDNA encoded a putative protein of 524 amino acids with significant similarity to papain cysteine proteases (MEROPS Family C01A, E value = 1e-58 for best hit). SignalP [33] analysis of the predicted protein identified a 22-amino acid signal peptide with a significant mean S value of 0.75. BLASTP searches against GenBank NR and the Phytophthora data sets revealed that both proteases are widely distributed among eukaryotes and oomycetes.
Protease inhibitors
We have also identified two S. parasitica cDNAs with similarity to protease inhibitor domains of two structural classes: (1) Kazal-like serine protease inhibitor (InterPro domain IPR002350, MEROPS family I1), (2) cysteine protease inhibitor (InterPro IPR000010, MEROPS family I25). We further analyzed these two cDNAs by aligning their putative inhibitor domains to those of known protease inhibitors (Fig. 4). Sp_001_01027 showed significant similarity to the Kazal-like inhibitors recently described by Tian et al. [21] from P. infestans and other plant pathogenic oomycetes. Amino acid residues defining the Kazal motif, including the six cysteine backbone, tyrosine and asparagine residues, were conserved in Sp_001_01027 (Fig. 4A). The predicted active site P1, which is central to the specificity of Kazal inhibitors [46,47], consisted of a proline, and therefore differed from all previously reported oomycete Kazal domains [21]. Sp_001_01374 is predicted to encode a secreted protein that bears the hallmark of the cystatin class of cysteine protease inhibitors including the highly conserved QXVXG motif in the first binding loop (L1) [48] (Fig. 4B). These findings suggest that secretion of protease inhibitors is a common feature of oomycetes.
Thiamine biosynthetic enzyme
We identified one sequence (Sp_001_00801) with significant similarity to a thiamine biosynthetic enzyme from plants (top hit to protein AAV92556 from the conifer Pseudotsuga menziesii, E value = 2e-63) and fungi (Schizosaccharomyces pombe protein CAA21093, E value = 3e-53). Unlike S. parasitica and other oomycetes, members of the genus Phytophthora are thiamine auxotrophs, they require exogenous sources of thiamine for growth [49,50]. Interestingly, BLAST searches of the genome sequence reads of P. sojae and P. ramorum, as well as all available sequences of P. infestans, failed to reveal sequences with similarity to Sp_001_00801 or to the plant and fungal enzymes. These findings suggest that this thiamine biosynthetic enzyme may have been lost in the Phytophthora lineage and could be related to thiamine auxotrophy in this genus.
Discussion
In this study we generated 1510 high quality ESTs from S. parasitica, an economically important and reemerging oomycete pathogen that causes multimillion dollar losses in the aquaculture industry. The ESTs were generated from a cDNA library constructed from one-month old nutrient deprived mycelium cultures. So far significant data sets of oomycete ESTs have been described for three plant pathogenic species, P. infestans [26,27], P. sojae [30] and P. parasitica [51]. Therefore, the S. parasitica ESTs offer some insights into the transcriptome of animal pathogenic oomycetes, which have been extremely understudied. Prior to this work, only 13 nucleotide and 2 protein sequences of S. parasitica could be retrieved from GenBank (March 2005 release). The sequence data and the corresponding annotations described in this study are accessible through an interactive public resource, the Oomycete Genomics Database (OGD). We hope that this pilot genomics project will accelerate research on this important pathogen and lays the foundation for more significant genome and cDNA sequencing initiatives of animal pathogenic oomycetes.
We used the S. parasitica ESTs to confirm the phylogenetic affinities between oomycetes and diatoms [16-18]. About 32% of the S. parasitica sequences that showed significant similarities to eukaryotic proteins matched a protein of the diatom T. pseudonana as a top hit. Within the oomycetes, S. parasitica is classified with other water molds, such as Achlya and Aphanomyces, in the order Saprolegniales [52-54]. These species are morphologically very distinct from the great majority of plant pathogens, such as the Peronosporales Phytophthora and downy mildews, or the Pythiales Pythium [55]. Indeed, the S. parasitica sequences tended to be relatively divergent from Phytophthora sequences. For example, based on the sequence alignments of 18 different conserved proteins, the average amino acid identity between S. parasitica and three Phytophthora spp., P. infestans, P. sojae, and P. ramorum, was 77% compared to 93% within Phytophthora. Differences in transcript content were also noted. cDNAs with similarity to elicitins, a group of 10-kDa proteins that occur in all Phytophthora and some Pythium species [56-61] and form 1–2% of mycelial ESTs in Phytophthora [26,27,62], were not found in the S. parasitica dataset. Although, elicitin-like genes could very well occur in the genome of S. parasitica, they do not seem to be abundantly expressed in mycelium. Elicitins were shown to function as sterol carriers [63,64]. Among the oomycetes, members of the Saprolegniales are able to synthesize sterols de novo whereas Phytophthora and Pythium spp. are sterol auxotrophs [50]. Possibly, S. parasitica may not require sterol carriers, such as elicitins, for optimal hyphal growth. Another difference between Phytophthora and Saprolegnia involves thiamine metabolism. Members of the genus Phytophthora are thiamine auxotrophs and require exogenous sources of thiamine for growth [49,50]. We identified one sequence in S. parasitica (Sp_001_00801) that shows significant similarity to a thiamine biosynthetic enzyme from plants and fungi but that is absent in the draft genome sequences of P. sojae and P. ramorum. This finding suggests that this thiamine biosynthetic enzyme may have been lost in the Phytophthora lineage and could be related to thiamine auxotrophy in this genus.
We searched the annotated data set for S. parasitica sequences that show similarities to known proteins and protein motifs that could inform us about the biology and pathology of this microbe. About half of the unigenes showed similarities to known protein sequences and could be assigned a putative function. A number of sequences showed particularly interesting similarities. cDNAs with similarity to signal transduction proteins, such as kinases and transcription factors, were particularly abundant. In total, 70 cDNAs encoded proteins with a putative signal peptide that are potentially secreted to the extracellular space. Secretion is an essential mechanism for delivery of virulence factors by eukaryotic pathogens to their appropriate site in infected host tissue. Therefore, several putative secreted proteins of S. parasitica, such as CBD proteins, CBEL-like proteins, glycosyl hydrolases, proteases, and protease inhibitors could function in virulence and will be worthy of additional studies.
Phylogenetic analyses indicated that several Phytophthora proteins, particularly hydrolytic enzymes such endopolygalacturonases, pectate lyases, exo-1,3-beta-glucanases, and an endo-1,3-beta-glucanase, are more similar to fungal proteins than to their counterparts in other eukaryotes [27,42-44]. These observations are in sharp contrast with phylogenies constructed from ribosomal sequences or compiled protein sequences from mitochondrial and housekeeping chromosomal genes, which indicate considerable evolutionary distance between oomycetes and fungi [16-18,65,66]. The apparent discrepancies between these phylogenies could reflect convergent evolution in the arsenal of hydrolytic enzymes between these pathogens, perhaps as a result of common mechanisms of infection among filamentous microbes [27,67]. The S. parasitica sequences allowed us to evaluate whether the similarity to fungal proteins extends to oomycetes other than Phytophthora and to animal pathogenic oomycetes. Although no S. parasitica sequences similar to the cell wall degrading enzymes endopolygalacturonases and pectate lyases were found, we identified a cDNA, SPM9F1, that encodes a 524 amino acid protein with high similarity to endo-1,3-β-glucanases, including the recently described PIENDO1 of P. infestans [42]. Similar to PIENDO1, SPM9F1 was most similar to fungal glucanases and no significant BLASTP hits (E value > 0.01) were observed to non-fungal eukaryotic proteins. Therefore, conservation in the arsenal of hydrolytic enzymes appears to extend beyond Phytophthora spp. to the Saprolegniales and animal pathogenic oomycetes.
Domain annotation of the S. parasitica sequences revealed the occurrence of a protein domain typically associated with fungi. Type I CBDs (InterPro domain IPB000254) are thought to be unique to fungi [34], although a related domain also occurs in the brown algae viral pathogen Ectocarpus siliculosus Virus EsV-1 (Phycodnaviridae) [35,68]. In this study, we found that this domain is widespread and diverse in S. parasitica and other oomycetes. A total of 18 domains from four oomycete species were found to share a 34 amino acid region that aligns perfectly with the canonical T. resei Cel6A CBD highlighting a core of conserved four cysteines and aromatic residues known to bind the cellulosic substrate [36,38]. Interestingly, the occurrence of this CBD in a virus of brown algae, which are related to oomycetes, suggests that type I CBDs might be more widespread in stramenopiles although we did not detect them in the draft genome sequence of the diatom T. pseudonana. In fungi, the CBDs are usually located in the N- or C-terminal regions of hydrolytic enzymes, such as cellulases and xylanases, and function by concentrating the catalytic domains on the surface of the insoluble cellulose substrate [34]. One of the S. parasitica cDNAs, SPM5F8, encodes a small 101 amino acid protein with two CBDs. Such a protein could function as a scaffolding component of the multienzyme complex known as cellulosome [34]. The function of this and other CBD proteins in S. parasitica may relate to attachment to organic debris on the host surface or during saprophytic growth. Alternatively, since cellulose is a major component of the cell wall of oomycetes, these proteins may play endogenous function in cell wall biogenesis.
Seven S. parasitica cDNAs showed similarity to CBEL, a 34-kDa cell wall glycoprotein of P. parasitica that binds to cellulose and host surfaces, functions in the agglutination of red blood cells, and elicits necrosis and defense gene expression in tobacco [39,40]. The similarity centered mainly on two regions of CBEL that match the PAN module/Apple domain (InterPro IPR000177). The CBEL-like PAN module, which is thought to function in protein-protein or protein-carbohydrate interactions [41], appeared to be particularly diverse in oomycetes with 52 different sequences identified in five species. The PAN module was found in proteins with diverse functions, such as the blood coagulation factor XI and the plasma protein pre-kallikrein [41]. Recently, several secreted proteins from apicomplexan mammalian parasites were found to contain Apple-like domains and are thought to play a role during parasite attachment and invasion of host cells [69-72]. For example, MIC4, an adhesin secreted by the apicomplexan Toxoplasma gondi, contains six Apple domains [69]. It remains to be determined the extent to which the secreted PAN/CBEL-like proteins of S. parasitica play a role in attachment and invasion during interaction with the fish host. Nonetheless, it appears that in oomycetes, similar to the apicomplexan parasites, some adhesins are secreted PAN module proteins.
Proteolytic enzymes are considered important virulence factors that aid in host colonization and release of nutrients by animal pathogenic microbes. It has long been known that the Saprolegnia spp. pathogenic on fish exhibit significant extracellular protease activity and it was postulated that this enzymatic activity contributes to pathogenesis [73]. A serine protease gene, AaSP2 from the related crayfish pathogen Aphanomyces astacus, was recently characterized and shown to be highly expressed during in vivo growth [74]. However, besides AaSP2, genes for secreted proteases of animal pathogenic oomycetes have not been reported. In this study, we identified a diverse set of 12 cDNAs of S. parasitica with similarity to the major catalytic classes of proteases. A number of the identified proteases had a signal peptide that would predict them to be localised at the interface between pathogen and host and suggests that they are candidate virulence factors.
Tian et al. [21] recently reported that plant pathogenic oomycetes secrete a diverse family of Kazal-like serine protease inhibitors with at least 35 members identified from P. infestans, P. sojae, P. ramorum, P. brassicae, and the downy mildew Plasmopara halstedii. Among these, the two-domain EPI1 protein and the three domain EPI10 of P. infestans were found to inhibit and interact with P69B, a defense subtilase of tomato, and were suggested to play a role in counterdefense [21]. Inhibitors of serine protease might be ubiquitous among eukaryotic parasites. For instance, the apicomplexan obligate parasite Toxoplasma gondii secretes TgPI-1 and TgPI-2, four-domain serine protease inhibitors of the Kazal family [75-77], and the intestinal hookworm Ancylostoma ceylanicum secretes an 8-kDa broad spectrum serine protease inhibitor of the Kunitz family [78]. Here we found that Kazal-like motifs also occur in Saprolegniales proteins.
In addition to Kazal-like motifs, we also discovered a cDNA that encodes a secreted protein with similarity to the cystatin class of cysteine protease inhibitors [79]. Cysteine protease inhibitors, such as chagasin, have been reported in animal parasites, mainly trypanosomids, and are thought to target proteases of the insect vector or the mammalian host [80-82]. Perhaps, inhibition of host proteases is a widespread counterdefense strategy in animal and plant pathogenic eukaryotes. Future studies will help to address whether the discovered protease inhibitors play a role in S. parasitica-fish interactions.
Conclusion
This pilot cDNA sequencing project provides a first look into the gene content of S. parasitica and sets the basis for genomics research in this reemerging animal pathogen. Annotation of the ESTs revealed a number of genes that could function in virulence. Future work will focus on developing molecular tools for functional analysis of S. parasitica genes. In this regards, stable transformation of Saprolegnia monoica has been reported [83], and the RNAi protocol recently developed for P. infestans [84] should be adaptable to S. parasitica. Gene expression profiling will also be applied to investigate transcriptome changes during S. parasitica-fish interactions. Overall, these resources will greatly accelerate research on this important pathogen and could lead to novel perspectives for controlling saprolegniosis.
Methods
Strains and growth conditions
Saprolegnia parasitica ATCC90214, an isolate from lesions on coho salmon (Oncorhynchus kisutch) [25], was used in this study. Working stocks of this strain were routinely maintained on cornmeal agar (Difco Lab. Detroit, MI) at 18°C. To obtain axenically prepared mycelium, ATCC90214 was grown in GY broth (5 g glucose, 2.5 g yeast extract/L) for 29 days, which corresponds to stationary phase. Mycelium was harvested by filtration and immediately frozen prior to RNA extraction.
cDNA construction
Total RNA from S. parasitica mycelium was isolated using the phenol-guanidine isothiocyanate based reagent Trizol, (Life Technologies Carlsbad, CA) according to the manufacturer's instructions. PolyA+ mRNA was isolated using the oligotex mRNA purification kit (Qiagen, Valencia, CA). The cDNA library was synthesized and cloned in plasmid pSPORT1 using the Superscript™ plasmid system for cDNA synthesis and cloning (Invitrogen Life Technologies, Carlsbad, CA). Polyadenylated mRNA was used to synthesize oligo (dT) primed cDNAs, which were cloned unidirectionally in NotI/SalI digested vector pSPORT1. Plasmid ligations were transformed into Escherichia coli ElectroMax-DH10B ™ cells (Invitrogen Life Technologies, Carsbad, CA). Selection was done on Luria-Bertani (LB) agar plates containing ampicillin (50 mg/L) [85]. Individual colonies were picked randomly with the Qpix robot (Genetix, Hampshire, UK) into 384 well plates containing LB freezing buffer (36 mM K2HPO4, 13.2 mM K2HPO4, 1.7 mM citrate, 0.4 mM MgSO4, 6.8 mM (NH4)2SO4, 4.4 % v/v glycerol in 1 × LB), incubated overnight without shaking, and stored at -80°C. Subsequently clones were transferred from the 384 well plate to 96 well plate for shipment to the Genomics Technology Support Facility (GTSF) at the Michigan State University where they were sequenced following manufacturer's recommendations using an ABI Prism 3700 DNA Analyzer. Identification codes for the cDNAs/ESTs were derived from the position of the corresponding cDNA clone in the microtiter plates preceded by SPM (for Saprolegnia parasitica mycelial) and the successive number of the microtiter plate.
DNA sequencing
For the ESTs, DNA from bacterial cultures was purified at GTSF using Qiagen 3000 or Autogen 850 robots. Fluorescently labeled sequencing products were generated using the universal T7 primer resulting in 5' cDNA sequences. The sequencing products were separated by capillary electrophoresis on an ABI Prism 3700 DNA Analyzer (PE Applied Biosystems). A dataset representing 2296 EST sequences and the corresponding electropherograms were then made available through the Geospiza Finch web interface of GTSF. The complete inserts of selected cDNAs were sequenced by primer walking at the OARDC Molecular and Cellular Imaging Center (MCIC), Wooster, Ohio, using an ABI Prism 377 automated sequencer (PE Applied Biosystems).
Bioinformatics
The sequences were processed using the XGI pipeline [28]. The assembly described in this paper is known as the May 2004 assembly. The consensus sequences (unigenes) were named Sp_N1_N2_May04 with N1 referring to the number of ESTs in the contig, and N2 the contig number. The consensus sequences were annotated using the methods implemented in the XGI pipeline [29]. These include BLASTX [86] searches against NCBI non-redundant (nr) protein library; BLIMPS search against Blocks+ protein motif database [87,88]; searches with the 12 algorithms of InterProScan [89] against the InterPro database [90]; and identification of signal peptides for extracellular secretion with PexFinder [91], an algorithm based on SignalP 2.0 [33,92]. Automated post-analysis annotation links BLAST and Blocks+ hits to their cognate Gene Ontology entries [93,94], whereas InterPro hits are automatically linked to GO annotations.
Additional similarity searches using BLAST [86] and other bioinformatics analyses were also performed locally on Mac OSX G4/G5 workstations. BLAST E-value lower than 0.01 were retained, and searches were conducted with the low-complexity filter on. Local databases were compiled from GenBank nonredundant (NR), dBEST, and TraceDB databases [95] and the Broad Institute [96]. They included "darwin_270334.faa" a curated dataset of 270334 eukaryotic proteins that we compiled. The data covers six major phyla: fungi, animals, plants, alveolates, discicristates, and heterokonts. It includes the complete proteomes of 17 species and at least the complete proteome of one species for all phyla except for discicristates. The MEROPS database of proteases and protease inhibitors was also queried [45]. Multiple alignments were conducted using the program Clustal-X [97], adjusted manually as necessary, and visualized with BOXSHADE [98]. Consensus sequences were visualized with weblogo [99]. A cDNA was deemed likely to be full length when it was the most 5' proximal EST among assemblies and gave hits to the N-terminal portion of known proteins following similarity searches.
Phylogenetic analysis
A data set of concatenated protein sequences was developed to perform phylogenetic comparisons of four oomycete species and the diatom Thalassosira pseudonana [100]. First, BLASTX searches of the S. parasitica unigenes against the diatom proteome were performed. Matching sequences with E value < 1e-20 were extracted and then used to search WGS and EST reads of P. infestans, P. sojae, and P. ramorum. A total of 18 sequences that were conserved among all five species were identified and were aligned individually with Clustal-X. Poorly aligned edges were then trimmed and the alignments were concatenated. PAUP v4.0b8 (Sinauer Associates Inc., Sunderland, MA) was used to reconstruct phylogenetic trees using the neighbor joining method and maximum parsimony with 1000 bootstrap replications.
Data dissemination
The DNA sequences, assemblies, and annotations are publicly available through the Oomycete Genomics Database (OGD) [1]. The 1510 high quality ESTs were also deposited in NCBI's GenBank under accession numbers DN615772-DN617281.
Authors' contributions
TTA, performance of majority of wet lab experiments including construction of cDNA library, annotation and analyses of specific sequences, writing of manuscript. MT, sequencing of select cDNAs, annotation and analyses of specific sequences. KG, performance of bioinformatics analyses including OGD pipeline. MEW, performance of bioinformatics analyses including OGD pipeline, writing of manuscript. PvW, annotation and analyses of specific sequences, writing of manuscript. SK, supervision of experimental work, annotation and analyses of specific sequences, writing of manuscript.
Acknowledgements
We thank Annette Thelen and her team at the GTSF, Michigan State University, as well as Tea Meulia and the staff of the OARDC-MCIC, Wooster, Ohio, for DNA sequencing. We also thank Shujing Dong and Diane Kinney for technical assistance. P.v.W. was supported by the Royal Society and the BBSRC. Salaries and research support were provided by State and Federal Funds appropriated to the Ohio Agricultural Research and Development Center, the Ohio State University.
Figures and Tables
Figure 1 Phylogenetic relationships between Saprolegnia parasitica and four other stramenopiles. The phylogenetic tree was constructed using the neighbor joining method based on concatenated alignments from 18 conserved proteins (2533 amino acids). Percentile bootstrap values based on 1000 replications and obtained with the neighbor joining methods/ maximum parsimony methods are indicated at the nodes. The scale bar represents 5% weighted amino acid sequence divergence.
Figure 2 Fungal type I cellulose binding domain (CBD) is widespread in oomycetes. (A) Multiple sequence alignment of 18 oomycete type I CBDs with the domain of Trichoderma resei Cel6A (Tr_Cel6A). The four conserved cysteine residues are marked with asterisks. The glutamine and three aromatic residues that are known to be important for binding the carbohydrate substrate are shown by arrows. Sequence names refer to the Saprolegnia parasitica unigene (Sp_), Phytophthora infestans (Pi_), Phytophthora sojae (Ps_), and Phytophthora ramorum (Pr_) followed by the OGD accession number, GenBank accession number, or NCBI Trace Archive identifier (Ti number). Multiple domains originating from the same sequence are marked with the letters a or b at the end of the sequence name. (B) Consensus sequence pattern of the oomycete type I CBD. Consensus sequence was calculated using WebLogo. The bigger the letter, the more conserved the amino acid site. The positions of amino acids in the consensus sequence correspond to the positions in the sequence alignment in panel A.
Figure 3 PAN module/Apple domain is widespread in oomycetes. (A) Multiple sequence alignment of 52 oomycete PAN module/Apple domain including the two domains of the previously described CBEL protein of Phytophthora parasitica (Pp_CBELa and Pp_CBELb). The six conserved cysteine residues are marked with asterisks. Sequence names refer to the Saprolegnia parasitica unigene (Sp_), Phytophthora infestans (Pi_), Phytophthora sojae (Ps_), and Phytophthora ramorum (Pr_) followed by the OGD accession number, GenBank accession number, or NCBI Trace Archive identifier (Ti number). Multiple domains originating from the same sequence are marked with the letters a or b at the end of the sequence name. (B) Consensus sequence pattern of the oomycete oomycete PAN module/Apple domain. Consensus sequence was calculated using WebLogo. The bigger the letter, the more conserved the amino acid site. The positions of amino acids in the consensus sequence correspond to the positions in the sequence alignment in panel A.
Figure 4 Protease inhibitors in Saprolegnia parasitica. (A) Sequence alignment of Sp_001_01027 predicted amino acid sequence with representative Kazal family inhibitor domains. Protein names correspond to protease inhibitors of Saprolegnia parasitica Sp_001_01027, Phytophthora infestans EPI1 (EPI1a-b, AY586273) and EPI10 (EPI10a-c, AY586282), the crayfish Pacifastacus leniusculus (PAPI-1a-d, CAA56043), and the apicomplexan Toxoplasma gondii (TgPI-1a-d, AF121778). The conserved cysteine residues that define the Kazal family protease inhibitor domain are marked with asterisks. The putative disulfide linkages formed by cysteine residues within the predicted Kazal domains are shown. The position of the predicted P1 residues is shown by an arrow. (B) Sequence alignment of Saprolegnia parasitica Sp_001_01374 (N-terminal fragment of the mature protein), chicken egg white cystatin (CHKCYS, P01038, mature protein), rice oryzacystatin-I (Oryzacystatin-I, P09229), human cystatin A (CYTA_human, P01040) and cystatin B (CYTB_human, P04080). The proposed active-site residues in cystatins, forming the N-terminal trunk (NT) and first binding loop (L1), are indicated.
Table 1 Most represented protein domains in the Saprolegnia parasitica EST contigs
InterPro ID Description Number of contigs
IPR000719 Eukaryotic protein kinase 22
IPR004000 Actin and actin-like 11
IPR001993 Mitochondrial energy transfer proteins (carrier protein) 9
IPR000626 Ubiquitin domain 6
IPR002110 Ankyrin-repeat 5
IPR001680 G-protein beta WD-40 repeats 5
IPR000795 GTP-binding elongation factor 5
IPR001353 Multispecific proteasome proteases 5
IPR003008 Tubulin/FtsZ family 5
IPR000873 AMP-dependent synthetase and ligase 4
IPR000254 Cellulose-binding domain, fungal type 4
IPR002130 Cyclophilin-type peptidyl-prolyl cis-trans isomerase 4
IPR002048 EF-hand 4
IPR000173 Glyceraldehyde 3-phosphate dehydrogenase 4
IPR000232 Heat shock factor (HSF)-type DNA-binding domain 4
IPR001806 Ras GTPase superfamily 4
IPR000504 RNA-binding region RNP-1 (RNA recognition motif) 4
IPR001464 Annexin family 3
IPR001623 DnaJ N-terminal domain 3
IPR001753 Enoyl-CoA hydratase/isomerase 3
IPR001179 FKBP-type peptidyl-prolyl cis-trans isomerase (PPIase) 3
IPR001005 Myb DNA binding domain 3
IPR003014 PAN domain 3
IPR000668 Papain cysteine protease (C1) family 3
IPR001272 Phosphoenolpyruvate carboxykinase (ATP) 3
IPR001849 Pleckstrin homology (PH) domain 3
IPR001232 SKP1-like 3
IPR000063 Thioredoxin 3
IPR001440 TPR repeat 3
==== Refs
Oomycete Genomics Database (OGD)
Neish GA Hughes GC Fungal diseases of fishes, Book 6. 1980 Neptune, New Jersey, USA, T.W.F. Publications
Willoughby LG Pickering AD Viable Saprolegniaceae spores on the epidermis of the salmonid fish Salmo trutta and Salmo alpinus. Trans Br Mycol Soc 1977 68 91 95
Bly JE Lawson LA Abdel-Aziz ES Clem LW Channel catfish, Ictalurus punctatus, immunity to Saprolegnia sp. J Appl Aquacult 1994 3 35 50
Hatai K Hoshiai GI Mueller GJ Pathogenicity of Saprolegnia parasitica Coker. Salmon Saprolegniasis 1994 Portland, Oregon, U.S. Department of Energy, Bonneville Power Administration 87 98
Bruno DW Wood BP Woo PTK and Bruno DW Saprolegnia and other oomycetes Fish Diseases and Disorders 1999 3 Wallingford, UK, CAB International 599 659
Hussein MM Hatai K Nomura T Saprolegniosis in salmonids and their eggs in Japan J Wildl Dis 2001 37 204 7 11272500
Hatai K Hoshiai G Mass mortality in cultured coho salmon (Oncorhynchus kisutch) due to Saprolegnia parasitica coker J Wildl Dis 1992 28 532 6 1474649
Neish GA Observations on saprolegniasis of adult sockeye salmon, Oncorhynchus nerka (Walbaum) J Fish Biol 1977 10 513 522
Whisler HC Identification of Saprolegnia spp. Pathogenic in Chinook Salmon 1996 Washington, D.C., US Department of Energy 43 p.
Willoughby LG Fungi and Fish Diseases 1994 Stirling, Scotland, Pisces Press 57 p.
Pickering AD Willoughby LG Roberts RJ Saprolegnia infections of salmonid fish Microbial Diseases of Fish 1982 London, England, Academic Press 271 297
Bruno DW Poppe TT A Color Atlas of Salmonid Diseases 1996 London, England, Academic Press 189 p.
Kamoun S Molecular genetics of pathogenic oomycetes Eukaryot Cell 2003 2 191 9 12684368 10.1128/EC.2.2.191-199.2003
Margulis L Schwartz KV Five Kingdoms: An Illustrated Guide to the Phyla of Life on Earth 2000 New York, W.H. Freeman and Co.
Sogin ML Siiberman JD Evolution of the protists and protistan parasites from the perspective of molecular systematics Int J Parasitol 1998 28 11 20 9504331 10.1016/S0020-7519(97)00181-1
Baldauf SL Roger AJ Wenk-Siefert I Doolittle WF A kingdom-level phylogeny of eukaryotes based on combined protein data Science 2000 290 972 7 11062127 10.1126/science.290.5493.972
Baldauf SL The deep roots of eukaryotes Science 2003 300 1703 6 12805537 10.1126/science.1085544
Beakes G A comparative account of cyst coat ontogeny in saprophytic and fish-lesion (pathogenic) isolates of the Saprolegnia declina-parasitica complex Can J Bot 1982 61 603 625
Rose JK Ham KS Darvill AG Albersheim P Molecular cloning and characterization of glucanase inhibitor proteins: coevolution of a counterdefense mechanism by plant pathogens Plant Cell 2002 14 1329 45 12084830 10.1105/tpc.002253
Tian M Huitema E da Cunha L Torto-Alalibo T Kamoun S A Kazal-like extracellular serine protease inhibitor from Phytophthora infestans targets the tomato pathogenesis-related protease P69B J Biol Chem 2004 279 26370 7 15096512 10.1074/jbc.M400941200
Sanchez LM Ohno Y Miura Y Kawakita K Doke N Host selective suppression by water-soluble glucans from Phytophthora spp. of hypersensitive cell death of suspension-cultured cells from some solanaceous plants caused by hyphal wall elicitors of the fungi Ann Phytopathol Soc Japan 1992 58 664 670
Alvarez F Villena A Zapata A Razquin B Histopathology of the thymus in Saprolegnia-infected wild brown trout, Salmo trutta L Vet Immunol Immunopathol 1995 47 163 72 8533294 10.1016/0165-2427(94)05384-5
Alvarez F Razquin B Villena A Lopez Fierro P Zapata A Alterations in the peripheral lymphoid organs and differential leukocyte counts in Saprolegnia-infected brown trout, Salmo trutta fario Vet Immunol Immunopathol 1988 18 181 193 3388763 10.1016/0165-2427(88)90060-8
Hatai K Willoughby LG Beakes GW Some characteristics of Saprolegnia obtained from fish hatcheries in Japan Mycol Res 1990 94 182 190
Kamoun S Hraber P Sobral B Nuss D Govers F Initial assessment of gene diversity for the oomycete pathogen Phytophthora infestans based on expressed sequences Fungal Genet Biol 1999 28 94 106 10587472 10.1006/fgbi.1999.1166
Randall TA Dwyer RA Huitema E Beyer K Cvitanich C Kelkar H Fong AM Gates K Roberts S Yatzkan E Gaffney T Law M Testa A Torto-Alalibo T Zhang M Zheng L Mueller E Windass J Binder A Birch PR Gisi U Govers F Gow NA Mauch F van West P Waugh ME Yu J Boller T Kamoun S Lam ST Judelson HS Large-scale gene discovery in the oomycete Phytophthora infestans reveals likely components of phytopathogenicity shared with true fungi Mol Plant Microbe Interact 2005 18 229 243 15782637
X Genome Initiative (XGI)
Waugh M Hraber P Weller J Wu Y Chen G Inman J Kiphart D Sobral B The Phytophthora genome initiative database: informatics and analysis for distributed pathogenomic research Nucleic Acids Res 2000 28 87 90 10592189 10.1093/nar/28.1.87
Qutob D Hraber PT Sobral BW Gijzen M Comparative analysis of expressed sequences in Phytophthora sojae Plant Physiol 2000 123 243 54 10806241 10.1104/pp.123.1.243
Cooke DE Drenth A Duncan JM Wagels G Brasier CM A molecular phylogeny of Phytophthora and related oomycetes Fungal Genet Biol 2000 30 17 32 10955905 10.1006/fgbi.2000.1202
Kroon LP Bakker FT van den Bosch GB Bonants PJ Flier WG Phylogenetic analysis of Phytophthora species based on mitochondrial and nuclear DNA sequences Fungal Genet Biol 2004 41 766 82 15219561 10.1016/j.fgb.2004.03.007
Nielsen H Engelbrecht J Brunak S von Heijne G A neural network method for identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites Int J Neural Syst 1997 8 581 99 10065837 10.1142/S0129065797000537
Levy I Shani Z Shoseyov O Modification of polysaccharides and plant cell wall by endo-1,4-beta-glucanase and cellulose-binding domains Biomol Eng 2002 19 17 30 12103362 10.1016/S1389-0344(02)00007-2
Delaroque N Muller DG Bothe G Pohl T Knippers R Boland W The complete DNA sequence of the Ectocarpus siliculosus Virus EsV-1 genome Virology 2001 287 112 32 11504547 10.1006/viro.2001.1028
Mattinen ML Kontteli M Kerovuo J Linder M Annila A Lindeberg G Reinikainen T Drakenberg T Three-dimensional structures of three engineered cellulose-binding domains of cellobiohydrolase I from Trichoderma reesei Protein Sci 1997 6 294 303 9041630
Srisodsuk M Lehtio J Linder M Margolles-Clark E Reinikainen T Teeri TT Trichoderma reesei cellobiohydrolase I with an endoglucanase cellulose-binding domain: action on bacterial microcrystalline cellulose J Biotechnol 1997 57 49 57 9335165 10.1016/S0168-1656(97)00088-6
Mattinen ML Linder M Drakenberg T Annila A Solution structure of the cellulose-binding domain of endoglucanase I from Trichoderma reesei and its interaction with cello-oligosaccharides Eur J Biochem 1998 256 279 286 9760165 10.1046/j.1432-1327.1998.2560279.x
Gaulin E Jauneau A Villalba F Rickauer M Esquerré-Tugayé MT Bottin A The CBEL glycoprotein of Phytophthora parasitica var. nicotianae is involved in cell wall deposition and adhesion to cellulosic substrates J Cell Sci 2002 115 4565 4575 12415001 10.1242/jcs.00138
Villalba Mateos F Rickauer M Esquerre-Tugaye MT Cloning and characterization of a cDNA encoding an elicitor of Phytophthora parasitica var. nicotianae that shows cellulose-binding and lectin-like activities Mol Plant-Microbe Interact 1997 10 1045 1053 9390419
Tordai H Banyai L Patthy L The PAN module: the N-terminal domains of plasminogen and hepatocyte growth factor are homologous with the apple domains of the prekallikrein family and with a novel domain found in numerous nematode proteins FEBS Lett 1999 461 63 67 10561497 10.1016/S0014-5793(99)01416-7
McLeod A Smart CD Fry WE Characterization of 1,3-b glucanase and 1,3;1,4-b glucanase genes from Phytophthora infestans Fungal Genet Biol 2002 38 250 263 10.1016/S1087-1845(02)00523-6
Gotesson A Marshall JS Jones DA Hardham AR Characterization and evolutionary analysis of a large polygalacturonase gene family in the oomycete plant pathogen Phytophthora cinnamomi Mol Plant-Microbe Interact 2002 15 907 921 12236597
Torto TA Rauser L Kamoun S The pipg1 gene of the oomycete Phytophthora infestans encodes a fungal- like endopolygalacturonase Curr Genet 2002 40 385 390 11919677 10.1007/s00294-002-0272-4
Rawlings ND Tolle DP Barrett AJ MEROPS: the peptidase database Nucleic Acids Res 2004 32 D160 D164 14681384 10.1093/nar/gkh071
Lu W Apostol I Qasim MA Warne N Wynn R Zhang WL Anderson S Chiang YW Ogin E Rothberg I Ryan K Laskowski MJ Binding of amino acid side-chains to S1 cavities of serine proteinases J Mol Biol 1997 266 441 461 9047374 10.1006/jmbi.1996.0781
Lu SM Lu W Qasim MA Anderson S Apostol I Ardelt W Bigler T Chiang YW Cook J James MN Kato I Kelly C Kohr W Komiyama T Lin TY Ogawa M Otlewski J Park SJ Qasim S Ranjbar M Tashiro M Warne N Whatley H Wieczorek A Wieczorek M Wilusz T Wynn R Zhang W Laskowski MJ Predicting the reactivity of proteins from their sequence alone: Kazal family of protein inhibitors of serine proteinases Proc Natl Acad Sci U S A 2001 98 1410 1415 11171964 10.1073/pnas.031581398
Margis R Reis EM Villeret V Structural and Phylogenetic Relationships among Plant and Animal Cystatins Arch Biochem Biophys 1998 359 24–30 9799556 10.1006/abbi.1998.0875
Leonian LH Lilly VG Studies on the nutrition of fungi. I. Thiamin, its constituents, and source of nitrogen Phytopathol 1938 28 531 548
Erwin DC Ribeiro OK Phytophthora Diseases Worldwide 1996 St. Paul, Minnesota, APS Press
Panabieres F Amselem J Galiana E Le Berre JY Gene identification in the oomycete pathogen Phytophthora parasitica during in vitro vegetative growth through expressed sequence tags Fungal Genet Biol 2005 42 611 623 15950158 10.1016/j.fgb.2005.03.002
Dick MW Margulis L, Corliss JO, Melkonian M and Chapman DJ Phylum Oomycota Handbook of Protoctista 1990 Boston, Jones & Bartlett 661 685
Leclerc MC Guillot J Deville M Taxonomic and phylogenetic analysis of Saprolegniaceae (Oomycetes) inferred from LSU rDNA and ITS sequence and ITS sequence comparisons Antonie van Leuwenhoek 2000 77 369 377 10.1023/A:1002601211295
Riethmuller A Weiss M Oberwinkler F Phylogenetic studies of Saprolegniomycetidae and related groups based on nuclear large subunit ribosomal DNA sequences Can J Bot 1999 77 1790 1800 10.1139/cjb-77-12-1790
Riethmuller A Voglmayr H Goker M Weiss M Oberwinkler F Phylogenetic relationships of the downy mildews (Peronosporales) and related groups based on nuclear large subunit ribosomal DNA sequences Mycologia 2002 94 834 849
Kamoun S Young M Glascock C Tyler BM Extracellular protein elicitors from Phytophthora: Host-specificity and induction of resistance to fungal and bacterial phytopathogens Mol Plant Microbe Interact 1993 6 15 25
Kamoun S Klucher KM Coffey MD Tyler BM A gene encoding a host-specific elicitor protein of Phytophthora parasitica Mol Plant Microbe Interact 1993 6 573 581 8274771
Kamoun S Young M Forster H Coffey MD Tyler BM Potential role of elicitins in the interaction between Phytophthora: species and tobacco App Env Microbiol 1994 60 1593 1598
Kamoun S van West P de Jong AJ de Groot K Vleeshouwers V Govers F A gene encoding a protein elicitor of Phytophthora infestans is down-regulated during infection of potato Mol Plant-Microbe Interact 1997 10 13 20 9002268
Kamoun S Lindqvist H Govers F A novel class of elicitin-like genes from Phytophthora infestans Mol Plant-Microbe Interact 1997 10 1028 1030 9353948
Ponchet M Panabieres F Milat ML Mikes V Montillet JL Suty L Triantaphylides C Tirilly Y Blein JP Are elicitins cryptograms in plant-Oomycete communications? Cell Mol Life Sci 1999 56 1020 1047 11212320 10.1007/s000180050491
Qutob D Huitema E Gijzen M Kamoun S Variation in structure and activity among elicitins from Phytophthora sojae Mol Plant Pathol 2003 4 119 124 10.1046/j.1364-3703.2003.00158.x
Mikes V Milat ML Ponchet M Ricci P Blein JP The fungal elicitor cryptogein is a sterol carrier protein FEBS Lett 1997 416 190 192 9369212 10.1016/S0014-5793(97)01193-9
Mikes V Milat ML Ponchet M Panabieres F Ricci P Blein JP Elicitins, proteinaceous elicitors of plant defense, are a new class of sterol carrier proteins Biochem Biophys Res Commun 1998 245 133 139 9535796 10.1006/bbrc.1998.8341
Paquin B Laforest MJ Forget L Roewer I Wang Z Longcore J Lang BF The fungal mitochondrial genome project: evolution of fungal mitochondrial genomes and their gene expression Curr Genet 1997 31 380 395 9162109 10.1007/s002940050220
Lang BF Seif E Gray MW O'Kelly CJ Burger G A comparative genomics approach to the evolution of eukaryotes and their mitochondria J Eukaryot Microbiol 1999 46 320 326 10461380
Latijnhouwers M de Wit PJGM Govers F Oomycetes and fungi: similar weaponry to attack plants Trends Microbiol 2003 11 462 469 14557029 10.1016/j.tim.2003.08.002
Van Etten JL Graves MV Muller DG Boland W Delaroque N Phycodnaviridae-large DNA algal viruses Arch Virol 2002 147 1479 1516 12181671 10.1007/s00705-002-0822-6
Brecht S Carruthers VB Ferguson DJ Giddings OK Wang G Jakle U Harper JM Sibley LD Soldati D The toxoplasma micronemal protein MIC4 is an adhesin composed of six conserved apple domains J Biol Chem 2001 276 4119 4127 11053441 10.1074/jbc.M008294200
Brown PJ Billington KJ Bumstead JM Clark JD Tomley FM A microneme protein from Eimeria tenella with homology to the Apple domains of coagulation factor XI and plasma pre-kallikrein Mol Biochem Parasitol 2000 107 91 102 10717305 10.1016/S0166-6851(00)00179-1
Brown PJ Gill AC Nugent PG McVey JH Tomley FM Domains of invasion organelle proteins from apicomplexan parasites are homologous with the Apple domains of blood coagulation factor XI and plasma pre-kallikrein and are members of the PAN module superfamily FEBS Lett 2001 497 31 38 11376658 10.1016/S0014-5793(01)02424-3
Deng M Templeton TJ London NR Bauer C Schroeder AA Abrahamsen MS Cryptosporidium parvum genes containing thrombospondin type 1 domains Infect Immun 2002 70 6987 6995 12438378 10.1128/IAI.70.12.6987-6995.2002
Peduzzi R Bizzozero S Immunochemical investigation of four Saprolegnia species with parasitic activity in fish: serological and kinetic charcterization of a chymotrypsin-like activity Microbial Ecol 1977 3 107 119 10.1007/BF02010400
Bangyeekhun E Cerenius L Söderhäll K Molecular cloning and characterization of two serine proteinase genes from the crayfish plague fungus, Aphanomyces astaci J Invert Pathol 2001 77 206 216 10.1006/jipa.2001.5019
Morris MT Coppin A Tomavo S Carruthers VB Functional analysis of Toxoplasma gondii protease inhibitor 1 J Biol Chem 2002 277 45259 45266 12228242 10.1074/jbc.M205517200
Morris MT Carruthers VB Identification and partial characterization of a second Kazal inhibitor in Toxoplasma gondii Mol Biochem Parasitol 2003 128 119 122 12706808 10.1016/S0166-6851(03)00051-3
Pszenny V Angel SO Duschak VG Paulino M Ledesma B Yabo MI Guarnera E Ruiz AM Bontempi EJ Molecular cloning, sequencing and expression of a serine proteinase inhibitor gene from Toxoplasma gondii Mol Biochem Parasitol 2000 107 241 249 10779600 10.1016/S0166-6851(00)00202-4
Milstone AM Harrison LM Bungiro RD Kuzmic P Cappello M A broad spectrum Kunitz type serine protease inhibitor secreted by the hookworm Ancylostoma ceylanicum J Biol Chem 2000 275 29391 29399 10893410 10.1074/jbc.M002715200
Rawlings ND Tolle DP Barrett AJ Evolutionary families of peptidase inhibitors Biochem J 2004 378 705 716 14705960 10.1042/BJ20031825
Santos CC Sant'anna C Terres A Cunha ESNL Scharfstein J de ALAP Chagasin, the endogenous cysteine-protease inhibitor of Trypanosoma cruzi, modulates parasite differentiation and invasion of mammalian cells J Cell Sci 2005 118 901 915 15713748 10.1242/jcs.01677
Besteiro S Coombs GH Mottram JC A potential role for ICP, a Leishmanial inhibitor of cysteine peptidases, in the interaction between host and parasite Mol Microbiol 2004 54 1224 1236 15554964 10.1111/j.1365-2958.2004.04355.x
Rigden DJ Mosolov VV Galperin MY Sequence conservation in the chagasin family suggests a common trend in cysteine proteinase binding by unrelated protein inhibitors Protein Sci 2002 11 1971 1977 12142451 10.1110/ps.0207202
Mort-Bontemps M Fevre M Transformation of the oomycete Saprolegnia monoica to hygromycin-B resistance Curr Genet 1997 31 272 275 9065391 10.1007/s002940050205
Whisson SC Avrova AO van West P Jones JT A method for double-stranded RNA mediated transient gene silencing in Phytophthora infestans Mol Plant Pathol 2005 6 153 163 10.1111/j.1364-3703.2005.00272.x
Sambrook J Fritsch EF Maniatis T Molecular Cloning: A Laboratory Manual 1989 2nd Cold Spring Harbor, N.Y., Cold Spring Harbor Laboratory
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 Nuc Acids Res 1997 17 3389 3402 10.1093/nar/25.17.3389
Henikoff JG Greene EA Pietrokovski S Henikoff S Increased coverage of protein families with the blocks database servers Nucleic Acids Res 2000 28 228 230 10592233 10.1093/nar/28.1.228
Henikoff JG Pietrokovski S McCallum CM Henikoff S Blocks-based methods for detecting protein homology Electrophoresis 2000 21 1700 1706 10870957 10.1002/(SICI)1522-2683(20000501)21:9<1700::AID-ELPS1700>3.0.CO;2-V
Zdobnov EM Apweiler R InterProScan--an integration platform for the signature-recognition methods in InterPro Bioinformatics 2001 17 847 848 11590104 10.1093/bioinformatics/17.9.847
Apweiler R Attwood TK Bairoch A Bateman A Birney E Biswas M Bucher P Cerutti L Corpet F Croning MD Durbin R Falquet L Fleischmann W Gouzy J Hermjakob H Hulo N Jonassen I Kahn D Kanapin A Karavidopoulou Y Lopez R Marx B Mulder NJ Oinn TM Pagni M Servant F The InterPro database, an integrated documentation resource for protein families, domains and functional sites Nucleic Acids Res 2001 29 37 40 11125043 10.1093/nar/29.1.37
Torto T Li S Styer A Huitema E Testa A Gow NA van West P Kamoun S EST mining and functional expression assays identify extracellular effector proteins from plant pathogen Phytophthora Genome Res 2003 13 1675 1685 12840044 10.1101/gr.910003
Nielsen H Brunak S von Heijne G Machine learning approaches for the prediction of signal peptides and other protein sorting signals Protein Eng 1999 12 3 9 10065704 10.1093/protein/12.1.3
The Gene Ontology
Ashburner M Ball CA Blake JA Botstein D Butler H Cherry JM Davis AP Dolinski K Dwight SS Eppig JT Harris MA Hill DP Issel-Tarver L Kasarskis A Lewis S Matese JC Richardson JE Ringwald M Rubin GM Sherlock G Gene ontology: tool for the unification of biology. The Gene Ontology Consortium Nat Genet 2000 25 25 29 10802651 10.1038/75556
TraceDB databases
Broad Institute
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 25 4876 4882 9396791 10.1093/nar/25.24.4876
BOXSHADE
WebLogo
Armbrust EV Berges JA Bowler C Green BR Martinez D Putnam NH Zhou S Allen AE Apt KE Bechner M Brzezinski MA Chaal BK Chiovitti A Davis AK Demarest MS Detter JC Glavina T Goodstein D Hadi MZ Hellsten U Hildebrand M Jenkins BD Jurka J Kapitonov VV Kroger N Lau WW Lane TW Larimer FW Lippmeier JC Lucas S Medina M Montsant A Obornik M Parker MS Palenik B Pazour GJ Richardson PM Rynearson TA Saito MA Schwartz DC Thamatrakoln K Valentin K Vardi A Wilkerson FP Rokhsar DS The genome of the diatom Thalassiosira pseudonana: ecology, evolution, and metabolism Science 2004 306 79 86 15459382 10.1126/science.1101156
|
16076392
|
PMC1192801
|
CC BY
|
2021-01-04 16:03:40
|
no
|
BMC Microbiol. 2005 Aug 2; 5:46
|
utf-8
|
BMC Microbiol
| 2,005 |
10.1186/1471-2180-5-46
|
oa_comm
|
==== Front
BMC MicrobiolBMC Microbiology1471-2180BioMed Central London 1471-2180-5-471607640010.1186/1471-2180-5-47Research ArticleThe influence of acetyl phosphate on DspA signalling in the Cyanobacterium Synechocystis sp. PCC6803 Morrison S Shawn [email protected] Conrad W [email protected] Mark K [email protected] Department of Basic Medical Sciences, Biochemistry Section, the University of the West Indies, Mona Campus, Kingston 7, Jamaica2 Department of Biology, University College London, Darwin Building, Gower Street, London WC1E 6BT, UK3 School of Biological and Chemical Sciences, Queen Mary, University of London, Mile End Road, London E1 4NS, UK2005 2 8 2005 5 47 47 22 3 2005 2 8 2005 Copyright © 2005 Morrison 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 dspA (hik33) gene, coding for a putative sensory histidine kinase, is conserved in plastids (ycf26) and cyanobacteria. It has been linked with a number of different stress responses in cyanobacteria.
Results
We constructed an insertional mutant of dspA (ycf26) in Synechocystis 6803. We found little phenotypic effect during nitrogen starvation. However, when the mutation was combined with deletion of the pta gene coding for phosphotransacetylase, a more significant phenotype was observed. Under nitrogen starvation, the pta/dspA double mutant degrades its phycobilisomes less than the wild type and still has about half of its chlorophyll-protein complexes.
Conclusion
Our data indicates that acetyl-phosphate-dependent phosphorylation of response regulator(s) overlaps with DspA-dependent signalling of the degradation of chlorophyll-protein complexes (and to a lesser extent phycobilisomes) in Synechocystis 6803.
==== Body
Background
The assembly, composition and maintenance of photosynthetic pigment-protein complexes are regulated by many different factors including light quality, light intensity, temperature, water availability and nutrient status [1]. Microorganisms have diverse mechanisms for sensing and acclimating to environmental parameters. In bacteria, two-component signal transduction pathways form a major part of the signalling machinery, mediating adaptive responses to a broad range of environmental stimuli [2-4]. In response to such stimuli, a sensor kinase autophosphorylates an internal histidine residue, and then transfers the phosphate to a conserved aspartate residue on the receiver domain of its cognate response regulator. The phosphorylation of the response regulator often leads to changes in gene transcription [5].
Response regulators are generally phosphorylated by their cognate sensor kinases, but phosphorylation can also be mediated through low molecular mass phosphate donors, such as acetyl phosphate, phosphoramidate (NH2PO3) and carbamyl phosphate [6-9]. The pta gene encodes phosphotransacetylase which catalyses the production of acetyl phosphate, the activated acetate intermediate that occurs during conversion of acetyl-CoA to acetate [10]. Chamnongpol & Groisman [11] identified acetyl phosphate as the phosphate donor for a mutant version of the response regulator PhoP (PhoP*) and showed that inactivation of the pta gene abolished PhoP*-mediated transcription. Most response regulators can be phosphorylated by acetyl phosphate in vitro [8], but at physiological concentration this may not be significant depending on the Km for each individual response regulator for acetyl phosphate. It may be that acetyl phosphate maintains a low level of phosphorylation for most response regulators. The pta gene has been identified in Synechocystis 6803 (slr2132) and Crocosphaera watsonii by homology to other bacterial pta sequences (up to 47% sequence identity by BLASTP), however pta has not been identified in the genome sequence of the other cyanobacteria.
The gene dspA (ycf26) codes for a putative sensory histidine kinase that is conserved in plastids and cyanobacteria. The deduced amino acid sequence of DspA contains 663 amino acids in Synechocystis sp. PCC 6803 (Synechocystis 6803). DspA has a HAMP [12] and a PAS sensor domain [13]. We have also identified a possible additional sensory domain in Ycf26, which will be discussed below. All cyanobacterial genomes sequenced to date contain an ORF encoding a Ycf26 orthologue and ycf26 is also the only histidine kinase gene present in the plastid genomes of the red algae Porphyra purpurea, Gracilaria tenuistipitata and Cyanidium caldarium [14]. This suggests that it performs a key role in the regulation of the level of photosynthetic pigment-protein complexes.
The probable orthologue for Ycf26 in the cyanobacterium Synechocystis 6803 is an ORF designated sll0698 in the CyanoBase genome database which has also been called Hik33 [15] and DspA [16]. It was identified as a factor that confers resistance to a variety of herbicides and was suggested to be a histidine kinase involved in chemical sensing [16]. It was found to be involved in gene expression (including photosynthesis-related genes), demonstrating characteristics of a sensor that can detect a decrease in membrane fluidity, and is involved in the perception and transduction of salt stress and low temperature signals in Synechocystis 6803 [15,17,18]. Deletion of this open reading frame leads to global changes in gene expression [19,20] and in addition a potential link has been identified between Hik33 and Rre31 (RpaA). In another cyanobacterium, Synechococcus sp. PCC7942 (Synechococcus 7942), the probable orthologue for Ycf26 was shown to regulate phycobilisome degradation during nutrient starvation and in high light. The gene was therefore designated nblS (non-bleaching mutant sensor) [21]. DspA is also important for the survival of Synechocystis 6803 when grown in high-light [22]. One possible cognate response regulator (NblR) has been identified in Synechococcus 7942 for NblS, however, no orthologue for NblR has been identified in Synechocystis 6803 (MK Ashby & J Houmard, unpublished) and it is likely that NblS also interacts with other response regulators [21]. NblS has been proposed to integrate redox and light signals and may influence other signalling pathways involved in acclimation responses [21]. Thus it appears that the orthologues of DspA in other cyanobacteria have different but overlapping functions.
A role for the sll0698 gene product in controlling the degradation of the photosynthetic apparatus during nutrient stress has not yet been demonstrated in Synechocystis 6803. We have constructed an insertional deletion mutant in this ORF, which we will refer to as dspA. We show that the inactivation of dspA has little phenotypic effect under nitrogen starvation, unless the mutation is combined with deletion of the pta gene. In the double mutant there is a dramatic effect on the degradation of chlorophyll-protein complexes under nitrogen starvation, which is not seen in either of the single mutants, however phycobilisomes are degraded, though not to the same level as in wild type. We conclude that acetyl phosphate masks DspA signalling under our laboratory conditions. The results indicate that the degradation of photosynthetic complexes during nutrient stress is controlled by multiple inputs, including acetyl phosphate level.
Results
Analysis of cyanobacterial and plastid deduced Ycf26 protein sequences
The deduced amino acid sequences for Ycf26 from 15 available cyanobacterial annotated genome sequences and the red algal species Porphyra purpurea, Gracilaria tenuistipitata and Cyanidium caldarium were analysed by Pfam and SMART. This identified a HAMP, PAS (low confidence in Cyanidium), HisKA and HATPase_c domain (Figure 1). SMART and Interpro also identified two potential transmembrane helical regions upstream of the HAMP domain (33–55 and 201–223 in 6803). The identification of transmembrane sequences by SMART is with an accuracy of 97–98% [23]. Because the 145 amino acid region that is flanked by the two transmembrane regions is highly conserved in all 15 cyanobacterial sequences and Porphyra (see additional figure online and Prodom Family PD339187 [24]) and it is adjacent and upstream of a HAMP domain (which is known to in most cases be actively involved in transmission of signals from a periplasmic signalling domain to the cytoplasmic portion of the protein [12,25,26]), we think it probably functions as a periplasmic sensor. We have not been able to detect homologous sequences in any other protein sequence, suggesting that it is unique to Ycf26 (DspA/NblS). The PAS sensor domain may bind a redox sensitive cofactor [13] that could potentially report on the metabolic state of the cell. Phylogenetic analysis (Ashby, MK and Houmard, J unpublished) and ClustalW alignments of these sequences and the alignment scores indicate that all these proteins (except Cyanidium caldarium and Gracilaria tenuistipitata) are probably orthologues (see supplementary data online). This is supported by Tu et al., [19] who were able to rescue their dspA null mutant in Synechocystis 6803 with the nblS gene from Synechococcus 7942.
Phenotype of the pta, dspA and pta/dspA mutants under nutrient-replete conditions
Mutants were routinely characterised by whole cell absorption spectra, which can be used to give an estimate of the amount of chlorophyll and phycocyanin per cell [27]. Mutants were further characterised by 77 K fluorescence emission spectra. These spectra give an indication of changes in the PSII:PSI ratio [28].
When cells were grown under low light (white light at 10 μmol m-2 s-1) the absorption spectra were similar, indicating little difference in the chlorophyll and phycocyanin content of the cells (data not shown). When cells were grown under higher light (80 μmol m-2 s-1), only slight differences in absorption spectra and Chl/PC ratios were seen (Table 2). However low-temperature fluorescence spectra recorded with chlorophyll excitation revealed changes in the ratio of PSII/PSI (Figure 2). These spectra show peaks at 685 nm and 695 nm (both from PSII) and 725 nm (PSI) [28,29]. Spectra have been normalised to the PSI fluorescence emission, as their absolute amplitudes are unreliable [28]. When cells were grown at 80 μmol m-2 s-1, the dspA and pta single mutants show a lower PSII/PSI ratio than the wild-type, and this difference was more pronounced in the pta/dspA double mutant (Figure 2). A similar pattern was seen in cells grown at 10 μmol m-2 s-1, but the differences were less pronounced (not shown).
Phenotype of the pta, dspA and pta/dspA mutants after nitrogen starvation
An nblS mutant of Synechococcus 7942 shows a pronounced phenotype under nitrogen starvation. In the wild-type, cells gradually lose their photosynthetic pigments under these conditions. However, an nblS mutant is bleached to a much lesser extent, retaining its phycobilisomes in particular [21]. We therefore looked at the pigmentation of our Synechocystis 6803 mutants during growth in modified BG11 medium supplemented with glucose (to help maintain the viability of the cells), but containing no NaNO3.
After three days nitrogen starvation in moderate light (80 μmol m-2 s-1), the wild-type and the dspA and pta single mutants were almost completely bleached (Table 2). We could detect no significant differences (less than 5%) in the rate of bleaching during nitrogen starvation in the three strains (data not shown). By contrast, the pta/dspA double mutant retained significant levels of pigment (Table 2). Both the chlorophyll and phycocyanin per cell dropped somewhat compared to the start of the treatment, but loss of phycocyanin was much greater (Table 2). Thus only the pta/dspA double mutant shows a 'non-bleaching like' phenotype: Chlorophyll concentration decreases by a factor of about two and that of phycocyanin by about five. Figure 3 shows low-temperature fluorescence emission spectra for the pta/dspA double mutant grown after 3 days nitrogen starvation. Whether with or without nitrate, it exhibits a similar spectrum meaning that the starved cells retain both PSI and PSII. The starved pta-/dspA- cells also retained significant levels of oxygen evolution (data not shown). Thus the pta/dspA double mutant retains its photosynthetic apparatus during nitrogen starvation, in contrast to the wild-type and both single mutants, where nearly all pigment is lost.
Discussion
A potential new sensor domain in DspA/Ycf26
Ycf26 may have a periplasmic sensor domain, which could play an important role in sensing stress either directly or indirectly (Figure 1). The putative periplasmic sensor and PAS domain may combine to enable Ycf26 to sense a wide variety of stress conditions [[15,17-19,21,22] & this work]. It remains to be determined whether DspA senses any of these signals directly or via other proteins or phosphorelays and the precise way in which these sensor domains interact. This apparent central role of Ycf26 in the regulation of gene expression in response to a number of different stresses is supported by its universal occurrence in the 15 available cyanobacterial genome sequences, particularly the 3 Prochlorococcus species which only have 4–6 histidine kinases in their genome ([30], MK Ashby & J Houmard, unpublished).
Roles of DspA and acetyl phosphate in cell signalling
We have constructed Synechocystis 6803 mutants deficient in the pta gene coding for phosphotransacetylase and the dspA gene coding for the DspA sensory histidine kinase (otherwise known as Ycf26 or NblS), and a double mutant deficient in both genes. We have used the mutants to examine the possible roles of DspA and acetyl phosphate in controlling cell responses to nitrogen starvation. We find strong evidence that DspA and acetyl phosphate act cooperatively as signalling inputs into pathways controlling photosystem stoichiometry under nutrient-replete conditions, and photosystem degradation during nitrogen starvation. Under nutrient-replete conditions, the dspA and pta single mutants both show a decreased PSII/PSI ratio (Figure 2). The PSII/PSI ratio becomes even lower in the pta/dspA double mutant, indicating additive effects of the two mutations. Since acetyl phosphate is known to act as a phosphodonor to response regulators [11], we propose that PSII/PSI ratio is influenced by gene expression controlled by one or more response regulators (or hybrid kinases) whose phosphorylation level is controlled by phosphotransfer both from DspA and from acetyl phosphate. This scheme (in its simplest form) is illustrated in Figure 4.
The effects seen under nitrogen starvation are even more striking. After 3 days' nitrogen starvation under moderate light, wild-type cells are virtually devoid of pigment, indicating almost complete loss of the phycobilisomes and the chlorophyll-protein complexes (Table 2). In this case, the dspA and pta single mutants both behaved in the same way as the wild-type, but the pta/dspA double mutant retained a significant proportion of its photosynthetic complexes, in particular the chlorophyll-protein complexes. Therefore the pta/dspA double mutant is clearly deficient in a signalling pathway required to initiate degradation of these complexes under these conditions. Again, we suggest that this pathway involves gene expression controlled by one or more response regulators whose phosphorylation level is determined by phosphotransfer from both DspA and acetyl phosphate (Figure 4). In this case the double mutation is needed to produce a detectable phenotypic effect, suggesting that either DspA or acetyl phosphate alone are sufficient to maintain high enough levels of phosphorylation of the relevant response regulator(s) (Figure 4). The fact that the phenotypic difference is seen only in the double mutant very strongly suggests that DspA and acetyl phosphate are acting as phosphodonors to the same response regulator(s). Again, it appears that a high level of response regulator phosphorylation promotes the degradation of the photosynthetic complexes.
Acetyl phosphate as a potential global regulator of gene expression in Synechocystis 6803
In E. coli the concentration of acetyl phosphate has been shown to be dependent on the metabolic state of the cell as well as the growth phase, carbon source, pH and temperature [8,31,32], therefore acetyl phosphate could potentially serve as a global sensor of the physiological state of the cell in E. coli. Acetyl phosphate is known to phosphorylate several RR's, including CheY, NRI, PhoB and OmpR [8]in vivo, but its ability to phosphorylate response regulators in vivo depends on the Km of the response regulator for acetyl phosphate, which in many cases is too high to allow phosphorylation [33]. In the four systems studied in one investigation of crosstalk between two component systems, one, the Uhp system was shown to be activated by acetyl phosphate when E. coli is grown on pyruvate [34]. There have been no comparable studies in cyanobacteria, but it seems likely that the level of acetyl phosphate would change in response to physiological stress, in particular nutrient deprivation. If this is the case, it seems likely that the Km of at least some response regulators could be tuned to the physiological concentrations of acetyl phosphate in Synechocystis 6803, allowing the level of acetyl phosphate to be part of the signalling process. The level of pigment complexes in the pta mutant was close to normal in nutrient replete conditions (table 2), indicating that the level of acetyl phosphate is not having an effect on overall expression of photosynthetic gene expression via a number of response regulators, but we cannot rule out that the cumulative effect of pta- and dspA- that we observe in nutrient depleted conditions may be influenced by a reduced phosphorylation of response regulator(s) other than those phosphorylated by DspA.
Role of DspA in initiating degradation of photosynthetic complexes during nutrient stress in Synechocystis 6803
NblS (DspA homologue) was originally identified in Synechococcus 7942 as a factor important for initiating the degradation of the photosynthetic complexes (in particular the phycobilisomes) during nitrogen starvation [21]. In Synechocystis 6803 there is strong evidence for DspA as a global regulator controlling the expression of many genes involved in photosynthesis, carbon metabolism and stress responses [15,17-19,21,22]. However, dspA mutations in Synechocystis 6803 have not previously been shown to induce a non-bleaching phenotype like that seen in Synechococcus 7942. Here we have shown that such a phenotype can be seen in the dspA mutant of Synechocystis 6803, provided that the pta gene is also deleted. The phenotype is still not identical to that seen in Synechococcus 7942, however, since in Synechocystis 6803 it is not only the degradation of the phycobilisomes which is affected but also the degradation of chlorophyll-protein complexes (Table 2).
Conclusion
It is clear that, under our conditions, DspA is a sensor for nitrogen starvation in association with the level of acetyl phosphate, leading to photosystem degradation, since this response occurs similarly in the wild-type and in the dspA single mutant (Table 2). However, it is unlikely to be coincidental that non-bleaching phenotypes are associated with nblS/dspA mutations both in Synechococcus 7942 and in Synechocystis 6803. We suggest that there must be other, as yet undetermined, conditions in which acetyl phosphate levels in wild-type cells are low. Under such conditions, DspA activity will become a physiologically important factor controlling the degradation of the photosynthetic complexes. The presence of Pta in Synechocystis 6803 may reflect its more flexible metabolic lifestyle compared to other cyanobacteria.
Methods
Strain and culture conditions
All strains used are derivatives of the glucose-tolerant strain of Synechocystis 6803 [35] and are listed in Table 1. All strains were grown in BG-11 medium [36] supplemented with 10 mM sodium bicarbonate and 12 mM sodium thiosulphate at 30°C in an illuminated shaking incubator (New Brunswick) under white light. For nutrient deprivation studies, BG-11 medium lacking NaNO3 but supplemented with 5 mM glucose was used, this also served as the wash medium. Illumination was at 80 μEm-2s-1 unless otherwise indicated. Cultures for nitrogen starvation experiments were grown in BG-11 with glucose at a low cell density for at least 3 days prior to nitrogen starvation. Erythromycin and kanamycin were added to media or plates when required at a final concentration 50 μg ml-1.
Sequence analysis
Deduced amino acid sequences were obtained form Cyanobase [37], the National Centre for Biotechnology Information (NCBI) [38] and the DOE Joint Genome Institute [39]. Protein to protein BLASTP 2.2.6/3 searches were performed at NCBI and Cyanobase websites. Protein sequences were analysed by Pfam [40,41], SMART [23,42] and Interpro [43]. CLUSTALW alignments [44] were performed at the European Bioinformatics Institute [45].
Molecular biology
Routine DNA manipulations were performed as in [46] using reagents purchased from QIAGEN, New England Biolabs, Inc., Promega and Sigma. Genomic DNA from Synechocystis 6803 was prepared as described by Porter [47]. The primers used for PCR amplification of the dspA gene had the sequences 5'-gcgagctcttctgtgtccaatccaacg and 5'-gcggtaccatggattgatacacggccag. The same primers were used to monitor segregation in the transformant. The primers used for PCR amplification of the pta gene had the sequences 5'-GCGAGCTCCCTTTATTTAAGCACCACC and 5'-GCGGTACCTTGCAAAGCTGTAATTACCACC and were also used to monitor segregation in the transformants. Primers were purchased from Invitrogen Technologies.
Construction of dspA and pta mutants of Synechocystis 6803
The pta open reading frame (slr2132 in the Cyanobase database) and the dspA open reading frame (sll0698) were amplified by PCR from Synechocystis 6803 DNA, using primers shown above. The PCR products were cloned into pBluescriptII KS +. The slr2132 sequence was cut at a PstI site at position 872 in the coding sequence, and an erythromycin resistance cassette (pBSEmEco1) was ligated in. The sll0698 sequence was cut at an EcoRI site at position 1384 in the coding sequence, and a kanamycin resistance cassette (pUC4K) was ligated in [48,49]. Slr2132 is 265 bp downstream of slr1888 and it is probably transcribed on an mRNA encoding only Pta, so its interruption should not have any polar effects. Sll0698 is upstream of two putative transposases sll0699 and sll0700. The glucose-tolerant strain of Synechocystis 6803 was transformed with the slr2132::EmR construct as described by Porter [47]. After several rounds of growth on erythromycin plates, PCR was used to check the insertion. This confirmed that the wild-type slr2132 locus had been completely replaced by the slr2132::EmR locus, and hence that the pta gene had been completely inactivated generating a fully-segregated pta mutant. A similar procedure was followed with the sll0698::KmR construct, which was used to transform both the wild-type and the pta mutant, to generate a sll0698 single insertional mutant (dspA) and a sll0698:slr2132 double insertional mutant (pta/dspA). These mutants were shown to have segregated after six rounds of streaking to single colonies on kanamycin (plus erythromycin for pta double mutants) plates as has been observed by Hsiao et al. [22], though without glucose in this work.
Procedure for initiating nutrient deprivation
50 ml cultures were grown to mid-log phase in BG11 medium. Each culture was then harvested by centrifugation, resuspended in 30 ml of wash medium, harvested a second time and resuspended in 10 ml of BG11 medium supplemented with glucose but lacking NaNO3. This was then divided into two 5 ml portions, which were then used to inoculate parallel 20 ml cultures.
Pigment content analysis
Fluorescence emission spectra were measured at 77 K in a Perkin-Elmer LS50 luminescence spectrometer equipped with a liquid-nitrogen sample holder. Cells were harvested by centrifugation and resuspended in growth medium to a chlorophyll a concentration of 5 μM [28]. Re-absorption of emitted fluorescence was negligible at the cell concentrations used. The samples were not mixed with glycerol, as this alters the low temperature fluorescence emission spectrum [29]. The samples were then injected into silica tubes (2 mm internal diameter) and dark adapted for 5 minutes. Samples were rapidly frozen by immersion in liquid nitrogen. Excitation and emission slit widths were 5 nm. Chlorophyll a was excited at 435 nm: light at this wavelength is not significantly absorbed by the phycobilisomes [50]. Chlorophyll a concentration was estimated from the absorption of methanol extracts at 665 nm [51]. Spectra from different samples were normalised to the PSI peak (emission at 725 nm). Cell absorption spectra were recorded with an Aminco DW2000 spectrophotometer. The concentrations of chlorophyll and phycocyanin were estimated from the spectra using the equations of Myers et al. [27]. Cell numbers were estimated from the optical density at 750 nm, calibrated using a hemocytometer.
Abbreviations
aa – amino acids; Dsp – drug sensory protein, HAMP – histidine kinases, adenyl cyclases, methyl binding proteins and phosphatases; NblS – non-bleaching mutant sensor, PAS – PER-ARNT-SIM; RR – Response regulator.
Authors' contributions
SSM performed the spectroscopy, analysed the segregation of the mutants and drafted the manuscript. CWM supervised SSM at UCL and drafted the manuscript. MKA conceived the study, constructed the interposon mutants and drafted the manuscript.
Supplementary Material
Additional File 1
ClustalW alignment of NblS (DspA, Ycf26) putative amino acid sequences.
Click here for file
Acknowledgements
SSM is supported by a studentship from the Office of Graduate Studies and Research, University of the West Indies. This work was funded by a New Initiative Grant to MKA from the Office of Planning and Institutional Research, University of the West Indies. CWM acknowledges financial support from the Biotechnology and Biological Sciences Research Council.
Figures and Tables
Figure 1 Cartoon of the domain structure of Ycf26 (NblS, DspA). HAMP – histidine kinase and methyl-accepting proteins; HisKA and HATpase – catalytic domains of histidine kinase; PAS – period protein, aryl hydrocarbon receptor nuclear translocator protein and single-minded protein; PSD – Putative sensor domain; TMH – Trans membrane helix;
Figure 2 77 K fluorescence emission spectra for cells grown under nutrient-replete conditions. Synechocystis 6803 wild-type and mutants. Excitation is at 435 nm (chlorophyll excitation) and spectra are normalised to the PSI fluorescence emission peak at 725 nm. Cells grown under moderately high light (80 μmol m-2 s-1).
Figure 3 77 K fluorescence emission spectra for cells after growth in nitrate-depleted medium. Synechocystis 6803 wild-type and mutants grown for 3 days in these conditions. Excitation is at 435 nm.
Figure 4 Simplest model that fits the data for information flow during DspA signalling. Acetyl phosphate and DspA can both act as phosphodonors to one or more response regulators. Under nitrogen starvation, phosphorylation by either acetyl phosphate or DspA alone is sufficient to saturate the response. Thus pta or dspA single mutants behave like the wild-type, and only the pta-dspA- double mutant has a distinctive phenotype. Under other conditions the steady-state phosphorylation level of the response regulators (or hybrid kinases) is not saturated either by acetyl phosphate or by DspA alone. Then pta and dspA deletions have additive effects. Both single mutants show a phenotypic change, but a stronger effect is seen in the double mutant. RR: response regulator; AK: acetate kinase; PTA: phosphotransacetylase.
Table 1 Bacterial strains and plasmids used
Strain Description Reference/Source
Synechocystis 6803 Glucose-tolerant 'wild-type' strain [35]
pta ORF slr2132 inactivated by insertion of EmR gene into PstI site at position 1,251,313 of the genome (872 bp into the coding seq of slr2132). This study
dspA ORF sll0698 inactivated by insertion of KmR gene into EcoR1site at position 125,060 of the genome (1384 bp into the coding seq of sll0698). This study
ptadspA ORF sll0698 inactivated in pta This study
pBluescriptII KS + Multipurpose cloning vector with AmpR. Stratagene
pBSEmEco1 Cloned EmR gene cassette. [48,49]
pUC4K Cloned KmR gene cassette Stratagene
Table 2 Pigment contents of Synechocystis 6803 wild-type and mutants. Data before and after nitrogen starvation for 3 days at 80 μmol m-2s-1 as described in Experimental Procedures. Pigment concentrations and cell densities were estimated from absorption spectra. Values given are the means of three replicates, with standard deviations. Asterisks indicate pigment concentrations that were too low to be accurately determined. All media contained glucose.
Strain Chlorophyll/Cell (molecules × 106) Phycocyanin/Cell (molecules × 106) Chlorophyll/phycocyanin
+nitrate -nitrate +nitrate -nitrate +nitrate -nitrate
Wild-type 13.6 ± 2.1 * 5.6 ± 1.4 * 2.4 ± 0.1 *
pta 12.6 ± 0.4 * 4.6 ± 1.5 * 2.7 ± 0.1 *
dspA 13.8 ± 0.1 * 5.9 ± 0.5 * 2.4 ± 0.1 *
pta/dspA 15.8 ± 0.9 7.8 ± 0.5 5.9 ± 0.5 1.0 ± 0.2 2.7 ± 0.2 7.8 ± 2.3
==== Refs
Niyogi KK Photoprotection Revised: Genetic and molecular approaches Annu Rev Plant Physiol Plant Mol Biol 1999 50 333 359 15012213 10.1146/annurev.arplant.50.1.333
Parkinson JS Kofoid EC Communication modules in bacterial signaling proteins Annu Rev Genet 1992 26 71 112 1482126 10.1146/annurev.ge.26.120192.000443
Parkinson JS Signal transduction schemes of bacteria Cell 1993 73 857 871 8098993 10.1016/0092-8674(93)90267-T
Stock AM Robinson VL Goudreau PN Two-Component Signal Transduction Annu Rev Biochem 2000 69 183 215 10966457 10.1146/annurev.biochem.69.1.183
Robinson VL Buckler DR Stock AM A tale of two components: a novel kinase and a regulatory switch Nature Struct Biol 2000 7 626 633 10932244 10.1038/77915
Feng J Atkinson MR McCleary W Stock JB Wanner BL Ninfa AJ Role of phosphorylated metabolic intermediates in the regulation of glutamine synthetase synthesis in Escherichia coli J Bacteriol 1992 174 6061 6070 1356964
McCleary WR Stock JB Ninfa AJ Is acetyl phosphate a global signal in Escherichia coli? J Bacteriol 1993 175 2793 2798 8491699
McCleary WR Stock JB Acetyl phosphate and the activation of two-component response regulators J Biol Chem 1994 269 31567 31572 7989325
Wolfe AJ The Acetate Switch Microbiol Mol Biol Rev 2005 69 12 50 15755952 10.1128/MMBR.69.1.12-50.2005
Brown TDK Jones-Mortimer MC Kornberg HI The enzymatic interconversion of acetate and acetyl-coenzyme A in Escherichia coli J Gen Microbiol 1977 102 327 336 21941
Chamnongpol S Groisman EA Acetyl Phosphate-dependent Activation of a Mutant PhoP Response Regulator that Functions Independently of its Cognate Sensor Kinase J Mol Biol 2000 300 291 305 10873466 10.1006/jmbi.2000.3848
Appleman JA Stewart V Mutational analysis of a conserved signal-transduction element: the HAMP linker of the Escherichia coli nitrate sensor NarX J Bacteriol 2003 185 89 97 12486044 10.1128/JB.185.1.89-97.2003
Gilles-Gonzales MA Gonzales G Signal transduction by heme-containing PAS-domain proteins J Appl Physiol 2004 96 774 783 14715687 10.1152/japplphysiol.00941.2003
Ashby MK Houmard J Mullineaux CW The ycf27 genes from cyanobacteria and eukaryotic algae: distribution and implications for chloroplast evolution FEMS Microbiol Lett 2002 214 25 30 12204368 10.1016/S0378-1097(02)00834-0
Suzuki I Los DA Kanesaki Y Mikami K Murata N The pathway for perception and transduction of low-temperature signals in Synechocystis EMBO J 2000 19 1327 1334 10716932 10.1093/emboj/19.6.1327
Bartsevich VV Shestakov SV The dspA gene product of the cyanobacterium Synechocystis sp. strain PCC 6803 influences sensitivity to chemically different growth inhibitors and has amino acid similarity to histidine protein kinases Microbiology 1995 141 2915 2920 8535519
Suzuki I Kanesaki Y Mikami K Kanehisa M Murata N Cold-regulated genes under the control of the cold sensor Hik33 in Synechocystis Mol Microbiol 2001 40 235 244 11298290 10.1046/j.1365-2958.2001.02379.x
Marin K Suzuki I Yamaguchi K Ribbeck K Yamamoto H Kanesaki Y Hagemann M Murata N Identification of histidine kinases that act as sensors in the perception of salt stress in Synechocystis sp. PCC 6803 Proc Natl Acad Sci USA 2003 100 9061 9066 12853569 10.1073/pnas.1532302100
Tu C Shrager J Burnap RL Postier BL Grossman AR Consequences of a deletion in dsp A on transcript accumulation in Synechocystis sp. strain PCC 6803 J Bacteriol 2004 186 3889 3902 15175303 10.1128/JB.186.12.3889-3902.2004
Paithoonrangsarid K Shoumskaya MA Kanesaki Y Satoh S Tabata S Los DA Zinchenko VV Hayashi H Tanticharoen M Suzuki I Murata N Five histidine kinases perceive osmotic stress and regulate distinct sets of genes in Synechocystis J Biol Chem 2004 279 53078 53086 15471853 10.1074/jbc.M410162200
van Waasbergen LG Dolganov N Grossman AR nblS, a gene involved in controlling photosynthesis-related gene expression during high light and nutrient stress in Synechococcus elongatus PCC 7942 J Bacteriol 2002 184 2481 2490 11948163 10.1128/JB.184.9.2481-2490.2002
Hsiao H He Q van Waasbergen LG Grossman AR Control of photosynthetic and high-light responsive genes by the histidine kinase DspA: negative and positive regulation and interactions between signal transduction pathways J Bacteriol 2004 186 3882 3888 15175302 10.1128/JB.186.12.3882-3888.2004
Letunic I Goodstadt L Dickens NJ Doerks T Schultz J Mott R Ciccarelli F Copley RR Ponting CP Bork P Recent improvements to the SMART domain-based sequence annotation resource Nucleic Acids Res 2002 30 242 244 11752305 10.1093/nar/30.1.242
Prodom
Zhu Y Inouye M The HAMP Linker in Histidine Kinase Dimeric Receptors is Critical for Symmetric Transmembrane Signal Transduction J Biol Chem 2004 279 48152 48158 15316026 10.1074/jbc.M401024200
Kristich CJ Ordal GW Analysis of Chimeric Chemoreceptors in Bacillus subtilis Reveals a Role for CheD in the Function of the McpC HAMP Domain J Bacteriol 2004 186 5950 5955 15317802 10.1128/JB.186.17.5950-5955.2004
Myers J Graham JR Wang RT Light harvesting in Anacystis nidulans studied in pigment mutants Plant Physiol 1980 66 1144 1149
Mullineaux CW Excitation energy transfer from phycobilisomes to Photosystem I in a cyanobacterial mutant lacking Photosystem II Biochim Biophys Acta 1994 1184 71 77
Mao H Li G Li D Wu Q Gong Y Zhang X Zhao N Effects of glycerol and high temperatures on the structure and function of phycobilisomes in Synechocystis sp. PCC 6803 FEBS Lett 2003 553 68 72 14550548 10.1016/S0014-5793(03)00973-6
Rocap G Larimer FW Lamerdin J Malfatti S Chain P Ahlgren NA Arellano A Coleman M Hauser L Hess WR Johnson ZI Land M Lindell D Post AF Regala W Shah M Shaw SL Steglich C Sullivan MB Ting CS Tolonen A Webb EA Zinser ER Chisholm SW Genome divergence in two Prochlorococcus ecotypes reflects oceanic niche differentiation Nature 2003 424 1042 1047 12917642 10.1038/nature01947
Heyde M Laloi P Portalier R Involvement of carbon source and acetyl phosphate in the external-PH-dependant expression of porin genes in Escherichia coli J Bacteriol 2000 182 198 202 10613880
Pruss BM Wolfe AJ Regulation of acetyl phosphate synthesis and degradation, and the control of flagellar expression in Escherichia coli Mol Microbiol 1994 12 973 984 7934904
McCleary WR The activation of PhoB by acetylphosphate Mol Microbiol 1996 20 1155 1163 8809768
Verhamme DT Arents JC Postma PW Crielaard W Hellingwerf K Investigation of in vivo cross-talk between key two-component systems of Escherichia coli Microbiology 2002 148 69 78 11782500
Williams JGK Packer, L, Glazer AN Construction of specific mutations in Photosystem II photosynthetic reaction center by genetic engineering methods in Synechocystis 6803 Methods in Enzymology 1988 167 San Diego: Academic Press 766 778
Castenholz RW Packer L, Glazer AN Culturing methods for cyanobacteria Methods in Enzymology 1988 167 San Diego: Academic Press 68 93
Cyanobase
National Centre for Biotechnology Information
DOE Joint Genome Institute
Protein families database of alignments and HMMs
Bateman A Birney E Cerruti L Durbin R Etwiller L Eddy SR Griffiths-Jones S Howe KL Marshall M Sonnhammer EL The Pfam protein families database Nucleic Acids Res 2002 30 276 280 11752314 10.1093/nar/30.1.276
Simple Modular Architecture Research Tool
Interpro TMHMM Server
Thompson JD Higgins DG Gibson TJ CLUSTALW: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice Nucleic Acids Res 1994 22 4673 4680 7984417
European Bioinformatics Institute
Sambrook J Fritsch EF Maniatis T Molecular cloning: a laboratory manual 1989 Second Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
Porter RD Packer L, Glazer AN DNA Transformation Methods in Enzymology 1988 167 San Diego: Academic Press 703 712 3148838
Ashby MK Mullineaux CW Cyanobacterial ycf 27 gene products regulate energy transfer from phycobilisomes to photosystems I and II FEMS Microbiol Lett 1999 181 253 260 10585546 10.1016/S0378-1097(99)00547-9
Ashby MK Mullineaux CW The role of ApcD and ApcF in energy transfer from phycobilisomes to PS I and PS II in a cyanobacterium Photosynth Res 1999 61 169 179 10.1023/A:1006217201666
Li Y Zhang J Xie J Zhao J Jiang L Temperature-induced decoupling of phycobilisomes from reaction centers Biochim Biophys Acta 2001 1504 229 234 11245787
Porra RJ Thompson WA Kriedemann PE Determination of accurate excitation coefficients and simultaneous equations for assaying chlororphylls a and b extracted with four different solvents: Verification of the concentration of chlorophyll standards by atomic absorption spectroscopy Biochim Biophys Acta 1989 975 384 394
|
16076400
|
PMC1192802
|
CC BY
|
2021-01-04 16:03:40
|
no
|
BMC Microbiol. 2005 Aug 2; 5:47
|
utf-8
|
BMC Microbiol
| 2,005 |
10.1186/1471-2180-5-47
|
oa_comm
|
==== Front
BMC NeurosciBMC Neuroscience1471-2202BioMed Central London 1471-2202-6-501610217110.1186/1471-2202-6-50Research ArticleSpatiotemporal receptive field properties of epiretinally recorded spikes and local electroretinograms in cats Wilms Marcus [email protected] Reinhard [email protected] Institute of Neurophysics, Philipps-University Marburg, Renthof 7, 35032 Marburg, Germany2 Institute of Medicine, Research Centre Jülich, 52425 Jülich, Germany2005 15 8 2005 6 50 50 30 3 2005 15 8 2005 Copyright © 2005 Wilms and Eckhorn; 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
Receptive fields of retinal neural signals of different origin can be determined from extracellular microelectrode recordings at the inner retinal surface. However, locations and types of neural processes generating the different signal components are difficult to separate and identify. We here report epiretinal receptive fields (RFs) from simultaneously recorded spikes and local electroretinograms (LERGs) using a semi-chronic multi-electrode in vivo recording technique in cats. Broadband recordings were filtered to yield LERG and multi unit as well as single unit spike signals. RFs were calculated from responses to multifocal pseudo-random spatiotemporal visual stimuli registered at the retinal surface by a 7-electrode array.
Results
LERGs exhibit spatially unimodal RFs always centered at the location of the electrode tip. Spike-RFs are either congruent with LERG-RFs (N = 26/61) or shifted distally (N = 35/61) but never proximally with respect to the optic disk. LERG-RFs appear at shorter latencies (11.9 ms ± 0.5 ms, N = 18) than those of spikes (18.6 ms ± 0.4 ms, N = 53). Furthermore, OFF-center spike-RFs precede and have shorter response rise times than ON-center spike-RFs. Our results indicate that displaced spike-RFs result from action potentials of ganglion cell axons passing the recording electrode en route to the optic disk while LERG-RFs are related to superimposed postsynaptic potentials of cells near the electrode tip.
Conclusion
Besides contributing to the understanding of retinal function we demonstrate the caveats that come with recordings from the retinal surface, i.e., the likelihood of recordings from mixed sets of retinal neurons. Implications for the design of an epiretinal visual implant are discussed.
==== Body
Background
The intact retina of animals and humans provides several sources of electrical responses to visual stimulation. Among them, the receptors, horizontal, bipolar, and to some extent the amacrine cells generate graded local potentials [1]. Their different superimposed neuronal contributions have extensively been studied in the electroretinogram (ERG) [2-6]. The retinal output comprises ganglion cells that possess functional specificity in that they preferentially respond to visual stimuli with contrasting center and surround structure flashed onto a restricted area of the retinal surface called the receptive field (RF) [7]. Ganglion cell spikes travel along axons beneath the inner retinal surface to the cortical centers for further visual processing. Interestingly, the axonal trajectory [8] and layering in the retina is highly organized. Radius & Anderson [9] reported that in primates axons from more central ganglion cells lie more superficially than axons arising from more peripheral retinal sites. Minckler [10] added to this that axonal bundles are consistently organized in a way that axons of peripapillary origin lie most superficially. It is because of this superficial arrangement of axons that the interpretation of epiretinal electrical recordings should meet necessary caveats. Thus, when Kuffler [7] performed his pioneering epiretinal electrical recordings in the cat retina, he distinguished axon from ganglion cell potentials by the distinct polyphasic waveforms and shifted RFs of axon potentials. He did not, however, further investigate the spatial relationships between the RFs of different spike sources and he did not record the local ERG response. Other studies used intracellular recordings from optic tract (e.g., [11]) or the cornea (e.g., [12]) lack the spatial resolution needed for revealing contributions of distinct retinal signal sources. Chichilnisky & Kalmar [13] registered spikes in vitro in macaque monkey retina but constrained the angular extent of the visual stimulus for the sake of spatial resolution. This may have obscured the contributions of shifted axonal spike-RFs. Recently, Segev and coworkers [14] used multi-electrode array and spike-sorting techniques to record retinal signals from a large fraction of ganglion cells in a patch of in vitro salamander retina. They mostly recorded somatic spikes and reported only rare observations of axonal spikes presumably due to the arrangement of axons in tight bundles. The few axonal RFs they observed were displaced with respect to the somatic RFs. We here present data of broadband epiretinal recordings in in vivo cat retina that allowed us to compare the spatio-temporal RFs of distinct spike sources and LERG signals measured simultaneously with the same microelectrode.
The superficial arrangement of ganglion cell axons is also expected to play a role in the design of an epiretinal visual implant [15,16]. Epiretinally applied electrical stimuli can activate axons as well as somata which should potentially lead to non-retinotopic cortical activations. This poses the question whether a useful epiretinal implant is feasible. One way to test this is to intracortically record evoked potentials in response to epiretinal electrical stimuli [17]. However, this approach is not optimal as it is not known in advance where exactly the cortical activation following an epiretinal axon stimulation is to be expected, making it difficult to access the right cortical site with a microelectrode. Another way of approaching the issue of retinotopic mapping is from the perspective of the receptive field concept. The latter is fundamental for the understanding of how the outside visual world retinotopically maps to neurons of the visual pathway. The analysis of retinal rather than cortical receptive fields allows one to specifically ask which retinal neurons might be activated by epiretinal stimuli. This makes sense assuming that a similar group of retinal neurons can be stimulated as well as recorded from by the same epiretinal microelectrode. We thus can avoid electrical stimulation tests in human patients or animals by detecting the neuronal structures a recording electrode is sensitive to, rather than the neuronal structures the same electrode excites when being switched to stimulation mode. We argue why this indirect approach is reasonable and discuss possible implications for the design and resolution of a visual prosthesis that is based on epiretinal electrical stimulation [17].
Results
We present data from experiments in four adult cats (experiments 140, 302, and 033: right eye; 142: left eye). Each set of data was recorded using seven epiretinal microelectrodes simultaneously in one or more electrode array positions per experiment. RF characteristics for each electrode position were typically confirmed more than once but they only once contribute to the present study resulting in a total of 44 different retinal recording positions.
Spatial RF aspects
The receptive fields estimated for epiretinally recorded signals often exhibited multiple, spatially segregated peaks (e.g., Fig. 1C, Fig. 2). We found LERG-RFs at 34/44 (77 %) of the recording positions. In the remaining 10/44 (23 %) we could not evaluate LERG-RFs either due to low signal-to-noise ratio or due to electrode dysfunction. LERG-RFs were always spatially unimodal and located at the actual position of the electrode as verified by back-projecting the electrode tip to visual space. Figure 2A exemplifies this finding: Each RF map corresponds to one retinal electrode and is arranged according to the hexagonal electrode array configuration. For example, electrode 2 was located at a retinal position corresponding to the upper left in visual space. Accordingly, the back-projection of the electrode tip (white crosshair) as well as the automated RF mapping (colored blob) points at an upper left position within the RF map. We were able to estimate RFs for six of the seven retinal LERG signals in this example, all being congruent with the projected electrode positions. The location of electrode 5 was only mapped using our back-projection protocol together with manual RF mapping based on spike activity (cf. Methods).
Multi unit activity (MUA, cf. Methods) comprises spikes from several neurons and exhibits multiple RFs that were found to be aligned along the estimated course of fiber bundles at the respective retinal eccentricity. This can be seen in Fig. 2B where we show RFs of MUA recordings that were simultaneously recorded with the LERGs in Fig. 2A. Note the spatially segregated peaks of positive or negative polarity that are all shifted to the left from the projected electrode position. To test whether such peaks correspond to ON- or OFF-RFs of single units we separated the retinal spikes from the raw broadband data (cf. Methods). We found that spikes separated by an amplitude threshold criterion belong to functionally distinct ganglion cells that differ in RF center locations, polarities, and/or time courses. We detected 81 spike- (or MUA-) RFs at 34/44 (77 %) of the recording positions. Thus, on average, we found about 1.8 spike-RFs per retinal recording position.
When LERG- and spike-RFs were detected simultaneously (in 28/44 (64 %) recording positions, 61 spike-RFs) we compared the relative position of RFs: Spike-RFs were either congruent with LERG-RFs (local RFs, i.e., located at the projected recording position; N = 26/61 (43 %)) or shifted distally (N = 35/61 (57 %)) but never proximally with respect to the optic disk. This strongly indicates that displaced spike-RFs result from the recording of spikes from passing axons en route to the optic disk. This is illustrated in Fig. 3 where we superimposed the LERG- and MUA-RFs from Figure 2 in addition to the estimated trajectories of ganglion cell axon bundles (reconstructed from [8]). MUA-RFs are either congruent with, or shifted distally from, the LERG-RFs with respect to the optic disk. Note that the positions of the LERG-RFs roughly resemble the hexagonal geometry of the electrode array. The geometry appears slightly distorted due to the angle at which the electrodes approached the retina and minor imperfections in the electrode array geometry.
Spike-RFs were found at all distal retinal separations from the retinal electrode tip that were mapped by the visual stimulus. They were either of ON-type (53/81; 65 %) or OFF-type (28/81; 35 %), whereas LERG-RFs were always of OFF-type (N = 34).
Temporal RF aspects
Apart from the spatial aspects of retinal RFs we also analyzed their temporal characteristics to further elucidate the origin of LERG- and shifted spike-RFs. For this we plotted the averaged time course of the retinal response to stimulation within the respective RF center (this is the stimulus-response cross-correlation) and calculated the response latency (cf. Methods). Only the broadband recordings were appropriate (N = 3 experiments) as they provided sufficient temporal resolution for the detection of small latency differences. In one experiment we lost the exact temporal information about LERG due to wrong filter settings reducing the number of experiments with broadband recorded LERG to two (N = 18 LERG-RFs). Broadband recorded spikes were available in three experiments (N = 53 spike-RFs).
Typical data from two experiments are presented in Fig. 4, where LERG, spike-ON, and spike-OFF RFs are plotted along with the corresponding averaged time courses of the retinal responses to stimulation in the RF center. Figure 4A (upper row) shows RF maps for two retinal spike trains that were separated from the same and simultaneous electrode recording (experiment 033, right eye). Both spike signals comprise two components as is evident from the two RFs in each map. The spatial shift of the spike-RFs with respect to the projected electrode tip position (crosshair) is to the left as the recording was made paracentrally in the right eye. The averaged time courses of the retinal signals in response to stimulation in their respective RF center (stimulus grid coordinates given in the lower right corners) are shown in Figure 4A (a-d). The average LERG response (leftmost graph) exhibits a smooth negative deflection with a short latency of 12.0 ms. For spike recordings ON-responses can be recognized by a peak of positive, OFF-responses by a peak of negative correlation with stimulus onsets. Interestingly, the spike ON-responses in the left and middle graphs evolve smoothly whereas the OFF-response depicted in the rightmost graph evolves in a stepwise fashion. In quantitative terms the ON-response latency and rise time is 18.1 and 2.7 ms in the left graph, and 15.4 and 2.6 ms in the middle graph. The longer latency in the left graph can partly be attributed to the longer conduction time for spikes that need to travel from their site of initiation (center of RF) to the recording position. The latency and rise time for the OFF-response is 15.5 and 1.0 ms.
Similar results can be derived from Figure 4B. Again RF maps are plotted for two separated spike populations of the same electrode recording (experiment 142, left eye). Here the left spike-RF map comprises three components (two OFF and one ON) and the right map only one (ON) as is evident from the peaks in the correlation maps. The spatial shift of spike-RFs with respect to the projected actual electrode position (crosshair) is now to the right as the recording was made paracentrally in the left eye. The latencies are 15.9 and 16.5 ms for the OFF-responses and 17.4 and 16.9 ms for the ON-responses. OFF-responses again evolve in a stepwise fashion with small rise times, whereas ON-responses evolve more smoothly with longer rise times. Moreover, the rightmost ON-response has two ON-correlation peaks. We found such double-ON-peaks in 5–7 of 53 spike-RFs (9–13 %; the uncertainty is due to low signal-to-noise ratio in two cases).
LERG responses generally had shorter latencies (11.9 ms ± 0.5 ms; N = 18 LERG-RFs in two experiments, standard error) when compared to spike-responses with local RFs (17.3 ms ± 0.7 ms, N = 17) or pooled local and distal RFs (18.6 ms ± 0.4 ms, N = 53). This is shown in Figure 5A, where we plotted frequency distributions for LERG (yellow), pooled ON- and OFF-spike, ON-spike, and OFF-spike RF latencies (blue). Note that the LERG latency distribution hardly overlaps the spike latency distributions. Superimposed on the latency distributions we plotted latency histograms for spike responses with local RFs (brown). Spike responses with distal RFs tended to peak later than those with local RFs most probably due to the additional axonal conduction time. Therefore, LERG latencies can best be compared with latencies of local spikes with latency differences of 5.4 ms ± 0.9 ms (two-tailed test, t(33) = 6.23, p < 10-6).
We further compared the response latencies for ON- and OFF-spikes. OFF-spike latencies were shorter (16.3 ms ± 1.0 ms, N = 8 local RFs) than ON-spike latencies (18.2 ms ± 1.0 ms, N = 9 local RFs). Latencies differed insignificantly by 1.9 ms ± 1.4 ms (t (15) = 1.35, p < 0.20). We estimated congruent ON- and OFF-spike RFs simultaneously with the same electrode at four recording positions. The mean latency difference in these cases was 2.4 ms ± 0.8 ms (1.6, 2.0, 2.5, 3.5 ms). The OFF-spike RFs temporally preceded the ON-spike RFs in all cases.
Stepwise OFF-responses
In two of the three experiments with broadband recordings (142, 033), the averaged time courses of spike OFF-responses to stimulation in the respective RF center appeared stepwise (Fig. 4). We therefore analyzed the steepness of the rising and falling phases of the time course for averaged ON and OFF responses, respectively (Fig. 5B). In the two experiments with stepwise signal kinetics, OFF-rise times were 1.6 ms ± 0.3 ms (N = 8) and ON rise times were 2.5 ms ± 0.1 ms (N = 15). The difference between the rise times was 1.0 ms ± 0.3 ms (t(21) = 3.63, p < 0.002). In the other experiment (302) OFF-responses appeared barely stepwise with rise times not significantly different (ON: 3.3 ms ± 0.1 ms (N = 25), OFF: 3.7 ms ± 0.5 ms (N = 5)).
Discussion
RFs of local electroretinograms
A general prerequisite for the generation of LERGs is the constructive superposition of extracellular field currents of individual neurons and glia cells [2,18]. Particularly retinal cells generating or relaying field currents in a roughly radial direction contribute to a field potential that can be registered with an epiretinal electrode. We recorded the locally generated field potentials (LERG) with the indifferent electrode placed in the vitreous body and the active electrode in close proximity to the inner limiting membrane (often somewhat impinging on it). This arrangement minimized the contribution of diffuse ERG from other sources [19,20].
The intact retina as used in our study provides several sources of graded local potentials, namely the receptor (outer nuclear layer), horizontal, bipolar, and to some extent the amacrine cells (inner nuclear layer) [1]. It is widely accepted that the graded potentials of the radially oriented receptors are responsible for the ERG's first negative deflection in response to a flash stimulus (a-wave, [3]). OFF-bipolar cells [21] and perhaps other more proximal cells [22] add to the shape of the a-wave. The a-wave can be observed with bright stimuli as used in our experiment. Under scotopic conditions the positive b-wave ERG component becomes more prominent. It originates mainly in ON-bipolar cells but is further influenced by OFF-center bipolar cells [23]. Thus, as the LERG originates in retinal processing that hierarchically precedes ganglion cell activity it makes sense that its latency is shorter than that of subsequent spike responses of ganglion cells as measured in this study. Given the polarity and short latencies (< 20 ms) we identify the first LERG component in our recordings as the ERG a-wave. We want to stress, however, that the LERG waveforms strongly depend on the electrode configuration and the stimulus used. Rodieck and Ford [18] argue that with small visual stimuli the distribution of extracellular currents from single cells must be considered as the lateral components do not cancel each other as with full-field stimuli. LERG responses to full-field and small visual stimuli can thus not easily be compared. Consequently, multi-focally evoked LERGs cannot be interpreted in a one-to-one manner in terms of ERG components measured with Ganzfeld stimuli [12,19].
All LERG-RF time courses showed negative initial peaks that were sometimes followed by a positive deflection (N = 34). Thus, field potentials were negative-going after ON-stimuli ("1" in the stimulus time course) and/or positive-going after OFF-stimuli ("-1" in the stimulus time course). This is typical of recordings with the recording electrode positioned intraretinally rather than in the vitreous humour [19,24].
The LERG yielded RFs that were always located at the position of the electrode tips projected to visual space. This strongly suggests that LERG are generated near the electrode tip. Interestingly, the localized nature of the LERG suggests that their RFs can be used as spatial reference points for the analysis of spike-RF positions. This can be very useful in experiments involving many retinal electrodes where the back-projection technique for the analysis of electrode positions would be cumbersome. However, a prerequisite is that the electrodes do not occlude the incoming light from the retina. The seven fiber electrodes in our setup were optically diffused as they were very thin and lying in a different focal plane than the photoreceptors. Their presence therefore had only minimal influence on the retinal light intensity distribution of the stimulus.
Another slow negative potential that can be invoked in the dark-adapted state with very dim light stimuli is the scotopic threshold response (STR) [25]. However, we used bright test stimuli that were not appropriate to elicit the STR. Moreover, STRs have longer latencies (more than 40 ms) than the a-wave evoked with brighter stimuli (less than 20 ms) [26]. The major (negative) response component that we observed clearly falls into the range for a-wave latencies. We therefore conclude that we did not misinterpret LERG a-wave/b-wave complexes with the STR.
We can further exclude that LERGs are merely artifacts due to impulse responses of the hardware filter to large action potentials because firstly, due to the causality of the used filters, LERG signals then needed to follow the spikes, instead LERG signals have shorter latencies than spikes. Secondly, LERG-RFs would have to be located at the same positions as the RFs of (large) spikes, which often was not the case.
RFs of single unit spikes
The RFs of single unit spikes simultaneously registered by the same electrode were either located at the site of the corresponding LERG-RF of this electrode (local RFs, 43 % of 61 spike-RFs) or tended to line up in a way that correlated well with the estimated fiber trajectories (distal RFs, 57 % of 61 spike-RFs). This strongly indicates that displaced spike-RFs result from the recording of axonal spikes originating at more distal locations along the fiber bundle and passing the recording electrode en route to the optic disk. It is not surprising that we registered axonal spikes since axons were in much closer proximity to the recording electrodes than the ganglion cell somata in situations where the electrode tip just touched the inner limiting membrane [27]. In addition to that we need to recall that the axonal trajectory [8] and layering [9,10] is highly organized in a way that the distance between the recording electrode and an axon correlates with the retinal eccentricity of the axon's origin. The amplitude of a recorded spike should therefore correlate with the retinal distance to its site of initiation (the ganglion cell soma). However, we could not find a correlation like this, possibly due to the uncertainty of the exact retinal recording depth. Moreover, axons originating in more peripheral retinal sites have larger calibers and hence produce larger voltage drops at the recording electrode than axons originating from peripapillary locations.
Given the limited angular extent of the visual stimulus in our setup (max. 17.5° monitor diagonal), it was a challenge to map distal RFs which could easily be separated by ten or more degrees of visual angle. Reducing the distance between eye and stimulus monitor increased the viewing angle and therefore the probability of mapping the RFs but at the expense of angular resolution. Additional RFs beyond the monitor's stimulation field were manually confirmed on the distal but not the proximal side of the LERG-RFs in all experiments. On average, we found 1.8 spike-RFs per retinal recording electrode. This number surely gives a lower limit as additional, more distal RFs were not detected, and more sophisticated event separation algorithms might filter out even more single neuronal units (e.g., [28]). Higher spatial stimulus resolutions can probably help to further separate neurons with strongly overlapping RFs. It has to be considered, though, that stimulus energy decreases with smaller pixels. The signal-to-noise ratio of the evoked responses will therefore decrease due to a lower spike activation probability.
Temporal aspects of retinal RFs
Spikes and LERGs simultaneously registered with the same electrode often had congruent RFs (local RFs). In these cases, however, LERG-RFs appeared at shorter latencies than spike-RFs. Since for local RFs optimal stimulation was done with the same stimulus pixels (i.e., in the RF center), we can exclude stimulus timing issues as an explanation for the latency difference.
Chichilnisky & Kalmar [13] performed in vitro multi-electrode recordings from flat mounts of macaque retinae and detected functional asymmetries in ON and OFF ganglion cells (probably parasol (magno) cells, comparable to cat Y-cells) including response kinetics. In particular they found that ON cells had a 10–20% shorter time-to-peak, trough and zero-crossing in the biphasic temporal impulse response than OFF cells. However, they also note that the response latency which they estimated as the time to 5% of the peak response was on average 1–2 ms shorter for OFF cells than for ON cells. The latter is in good accord with our estimation of a 1.9 ms latency difference between ON and OFF responses (time to 70% of peak response). The reason why we did not analyze the response peak times in more detail is that we found the OFF peaks to exhibit stepwise time courses that deemed it inappropriate to define a peak time because of their flat peak plateau (e.g., Fig. 4). Along with Chichilnisky & Kalmar [13] we assume that the biochemical steps involved in the generation of the ON-bipolar response yield a longer latency for ganglion cell spike initiation than the directly gated ionic currents underlying the OFF-bipolar response. This is because ON-center bipolar cells have metabotropic glutamate receptors that mediate a (time-consuming) response polarity reversal leading to membrane depolarization. On the other hand OFF-center bipolar cells have ionotropic glutamate receptors and respond to light like the photoreceptors with a membrane hyperpolarization which is less time consuming [29,30].
This asymmetry in the retinal ON/OFF pathways may also be accountable for the stepwise time course of cross-correlation functions between stimulus and OFF-responses in two of the three experiments with broadband recordings (Fig. 4). Here, OFF rise times were about 0.9 ms shorter than ON rise times (1.6 and 2.5 ms, respectively). This observation did not depend on the animals' medication nor the electrode recording depth in the retina since we saw smooth ON- and stepwise OFF-responses among simultaneously recorded single unit spikes from the same electrode (Fig. 4). As probably less retinal stages are involved in the initiation of OFF-center ganglion cell spikes the average spike onset time might be more reliably encoded and consequently yields a steeper rise in the cross-correlation function between stimulus and OFF-response. OFF-responses appeared barely stepwise in one other experiment (302). Here, rise times for ON- and OFF-responses were similar and both longer than for stepwise OFF responses. From the limited number of observations we cannot deduce why OFF-responses were stepwise in some recordings and not in others.
Based on the knowledge of spike responses recorded by the same electrode with local and distal RFs one can in principle deduce the spike conduction time as the difference of their response latencies. This is only true for RF-pairs with the same polarity as we argued that the signal transduction kinetics for ON- and OFF-responses differ and thus add different temporal delays to the overall signal latency. Response latencies varied considerably (Fig. 5A), though, sometimes leading to a distal RF temporally preceding a local one recorded with the same electrode despite the extra conduction time. This may indicate that single unit spike responses are sometimes recorded from different types retinal ganglion cells or amacrine cells. For example, Cleland and Levick [31] found that latencies differ for X and Y cells and depend on the retinal eccentricity. We did therefore not correct the latencies of responses with distal RFs for the additional conduction time. With a spike conduction velocity of about 1 m/s [32], the observed retinal RF separations of 0–2 mm correspond to about 0–2 ms extra conduction time. This is smaller than the latency difference between LERG and spike responses (6.7 ms, cf. Results) and further indicates that the source of the LERG response to visual stimulation must be precedent to spike initiation.
Consequences for the design of an epiretinal implant
Our results may have important implications for the design of epiretinal implants that aim at substituting retinal function by eliciting localized visual percepts ("phosphenes") via focal electrical stimulation [15,16,33]. Rather simplifying we assume that a similar group of retinal neurons may be stimulated as well as recorded from by the same epiretinal microelectrode. We thus imply the reciprocity of some properties of epiretinal recording and stimulation. This is reasonable as the electrode-retina interface is the same in both cases. The electrical field seen by a recording electrode is spatially similar to (but not the same as) the electrical stimulation field seen by retinal neurons since both depend on the distance between the electrode and the neuronal structure. Since we used microelectrodes with relatively large tips, the probability of recordings of spikes from fibers with large diameter (Y-cells) was high [27]. As the fibers of large diameter also have lower stimulation thresholds [34] it seems probable that the fibers preferentially recorded from were also those that would be preferentially activated by a short impulse via that electrode. However, neuronal elements can be stimulated indirectly via synapses of other retinal neurons. Moreover, the spatial structure of the neuronal sources and sinks, i.e., the shape of the individual cell recorded from must be known to predict which parts of the cell can be stimulated [34]. Also, degenerating retinae found in blind patients suffering from macular degeneration or retinitis pigmentosa undergo an extensive retinal remodelling that can further complicate the issue of controlled stimulation of retinal targets [35].
Our data demonstrate multi-site epiretinal RFs suggesting that epiretinal electrical stimuli might unspecifically activate retinal axons that originate in an elongated area of the retina distal to the stimulation site, potentially resulting in multi-site or elongated visual percepts. One way of dealing with multiple axonal stimulation would be to reduce the distance between electrode and ganglion cell somata by advancing small-tipped cone electrodes to the cell body layer. This would increase the probability of activating ganglion cell somata rather than axons. One could in principle test RFs of neurons from "quasi-intraretinal" recordings for contributions of multiple sources and adjust the recording depth until only a local component remains. We therefore tested the retinal RFs' dependence on the recording depth in one pilot measurement: We measured four LERG-RFs and one corresponding spike-RF. The LERG-RFs exhibited a slight latency decrease with penetration depth (appr. 5–10 ms/100 μm). Additionally, the polarity of LERG-RFs abruptly reversed at a certain recording depth (appr. 250 μm from the inner limiting membrane) which was accompanied by an increase in LERG-RF size. This is a typical situation if the choroidal side of Bruch's membrane is penetrated [19]. In contrast to the LERG, the latency of the averaged spike-response did not change with recording depth. Rather, the signal-to-noise ratio of the spike-RF in this example degraded with recording depth so that the spike-RF was not recognizable from background noise when the LERG-RF polarity reversal was complete. All of these effects were reversible when the recording depth was reduced. We assume that the suppression of the spike-RF indicates that quasi-intraretinal electrical stimuli can avoid axonal activation and be rather localized. It must be stressed that these results are from few recordings in one cat only which need to be further investigated in subsequent measurements.
McIntyre & Grill [36] approached the issue of the selectivity of neuronal activation studying a computer-based integrated field-neuron model. They found that the stimulus waveform can be a means of selective activation of targeted neuronal populations. Asymmetrical stimulus waveforms with a long-duration low-amplitude prepulse followed a shorter and high-amplitude compensating main stimulus pulse were among the most effective in selectively activating local cells rather than fibers of passage. If only one, displaced, perceived phosphene remained, this could be compensated for by an electronic preprocessor that maps the camera input to the retinal electrode grid [16].
Whether or not epiretinal electrical stimulation yields multi-site visual sensations must ultimately be tested in alert subjects that can report evoked visual percepts. Very promising data have been collected by Humayun and coworkers [37] who implanted an 4×4 epiretinal electrode array with disk shaped electrodes of 520 μm diameter into a blind subject's eye. Interestingly, they did not report displaced phosphene percepts and argue that a biphasic current pulse of 1 ms/phase and 1 ms interphase delay can activate a local pool of bipolar cells. As their stimulation electrodes were rather large, it remains an open issue whether retinotopic phosphene percepts can be guaranteed for much more densely packed electrodes as they would be needed for useful vision [38,39]. Indeed, contrary to Humayun and coworkers, Rizzo et al. [40] often observed elongated or multi-site visual percepts in human patients when they electrically stimulated their patients with single epiretinal electrodes. When stimulated with multiple electrodes, the perception of either elongated or multi-site phosphenes depended on the stimulation array's orientation relative to the ganglion cell axon trajectory underneath the electrodes. This supports our hypothesis that epiretinal electrical stimuli can lead to ambiguous visual percepts and that the analysis of epiretinal visual RFs can help to optimize the design of an epiretinal implant for substituting basic visual function.
Conclusion
Our data demonstrate the feasibility of obtaining highly specific receptive field data from different neural sources by epiretinal broadband recordings with microelectrodes. We could separate distinct retinal signals from the epiretinal broadband recordings which enabled us to study the RF properties of simultaneously recorded signal components from different retinal sources. The caveats that come with recordings from the retinal surface, i.e., the likelihood of mixed contributions from different retinal neurons, may also have implications for the design of an epiretinal visual implant.
Methods
Anesthesia, surgery, and animal care
For semichronic preparation adult cats (N = 4; 3.0 to 4.5 kg) received atropine sulphate (0.03 – 0.04 mg/kg) to reduce salivation. Anaesthesia was induced by intramuscular injection of a mixture of ketamine hydrochloride (Ketanest, 10–15 mg/kg) and xylazine hydrochloride (Rompun, 0.7 – 1.0 mg/kg). After orotracheal intubation, anaesthesia was maintained by ventilation with N2O / O2 (70 % / 30 %) and isofluorane (0.5 – 1.5 %). The level of anaesthesia was continuously controlled by monitoring the rectal temperature (38°C), endexpiratory CO2 (3.8 – 4.2 %), ECG, EEG, and muscle reflexes. For head fixation we implanted two bolts in cavities of the forehead by dental acrylic cement that enabled positioning in a standard Horsley-Clark support without pressure points typically induced by the use of ear bars. The operated eye was stabilized with the help of the electrode array and two sutures that were pierced through the conjunctiva and fixed to the Horsley-Clark support. Refraction of the eyes was corrected for the eye-to-monitor distance of 1.3 m using contact lenses. Besides during the ophthalmic surgery it was not necessary to give further analgesia between recording sessions as was assured by monitoring the cat's vital functions. One of the four cats was not sacrificed at the end of the recording sessions. In this case, sclera and conjunctiva were closed with polyglactin sutures and the animal prophylactically received penicillin. This semi-chronic preparation enabled repeated experiments in the same eye after allowing 2–3 weeks for convalescence. At the end of the other three experiments the animals were deeply anesthetized with isofluorane (3 %) followed by an intravenous injection of a lethal dose of T61 (Hoechst Roussel Vet, 2 ml). All animal experiments were conducted in accordance with the German animal welfare law, the guidelines of the European Community Council Directives (86/609/EEC) and the NIH Principles of Laboratory Animal Care (Publication 86-23, revised 1985).
Multiple microelectrode retinal recordings
Epiretinal recording was performed with fiber microelectrodes [41] with cone-shaped tips of 20–25 μm diameter and 200–500 k impedance at 1 kHz. We used a 7-electrode array (modified from the manipulator of Eckhorn & Thomas [42]) to insert the electrodes through a 1.1 mm scleral opening approximately 4 mm posterior to the temporal limbus. Electrode separations were either 0.3 mm (N = 2) or 0.25 mm (N = 2). In the latter case, three additional stainless steel capillaries similar to those guiding the electrodes were inserted through the same scleral opening as a means for adjusting the inner eye pressure via an elevated support of Ringer's solution (experiment 302 and 033). The electrodes were hexagonally arranged to provide for a two-dimensional recording and to minimize the scleral opening by a dense packing of electrodes (0.9 mm diameter of total bundle of guide capillaries). Retinal electrodes could be individually moved in axial direction under computer control. This allowed for precise epiretinal positioning of the electrode tips under visual inspection with an ophthalmoscope. Additionally, the whole electrode manipulator could be spherically moved around the electrodes' point of insertion into the eye so that a wide range of retinal positions could be approached without mechanical stress to the eye [43]. The tips of the electrodes were positioned 4° to 9° paracentrally with very low pressure on the retina. This resulted in a slight dimpling of the retinal surface that could precisely be adjusted under visual inspection with an ophthalmoscope and double-checked by listening to the onset of spike activity via an audio monitor. The epiretinal positions of the electrode tips were marked on a tangent screen with a custom-built back-projection device [44].
Data recording and pre-processing
Vital functions of the cat, including ECG, EEG, body temperature, and expiratory CO2 were continuously monitored during the experiments. Data were recorded and considered for analysis only when vital functions were normal and stable. Between the measurement sessions, all electrodes were individually retreated to a safe distance in the vitreous body, the electrode array was moved to another retinal site, and the electrodes individually positioned on the retinal surface again. This took at least 30 minutes. The typical interval between sessions was 60 minutes.
In our first experiment (140) we hardware-filtered the retinal signals to obtain local electroretinograms (LERG: 1–140 Hz, -3 dB at 12 dB/oct) and multi unit activity (MUA: 0.5–10 kHz, full wave rectified, then low pass filtered to 1–140 Hz, -3 dB at 30 dB/oct). These filtered signals were then sampled at 500 Hz (CED 1401plus, Cambridge Electronic Design) and stored for offline data analysis. Since we wanted to separate single unit spike trains contained in the raw signal we recorded broadband signals (1–4000 Hz) at 20 kHz sampling rate (Multi Channel Systems, Tübingen) in all subsequent experiments (142, 302, 033). For the sake of comparison with our first data set (140), LERGs were then extracted off-line by convolving the broadband signals with the impulse response of our laboratory hardware filter. Using an amplitude thresholding approach enabled us to extract spikes from high-pass filtered (0.5–1 kHz) broadband data: First, local signal peaks of a certain range of widths were detected by the analysis of the first and second derivatives of the signal time series. In a second step, the peaks were sorted into ranges of amplitudes and stored into separate event data files. However, it cannot be ruled out that one sorted spike population comprised more than one spiking neuron.
Visual stimulation and receptive field evaluations
Monocular RFs were qualitatively evaluated with a hand-held projector by stimulating with an adjustable light bar on a tangent screen 1.3 m in front of the cat. This distance also left enough room to quickly access the eyes for medical treatments between the recording sessions. For the quantitative characterization of the position, size, subfields, and orientation of RFs we used an automated mapping technique in which the luminance of a computer monitor was randomly spatio-temporally modulated covering the RF positions of all retinal recording locations simultaneously (28×28 grid of 0.45°×0.45° stimulus pixels, total range 12.5°×12.5° visual angle, maximum contrast of a computer monitor at 101 Hz frame rate in a dark room, the luminance of the white pixels was 75 cd/m2 and the luminance of the black pixels 0.1 cd/m2). The multi-focal pseudo-random stimuli were based on binary m-sequences with 4095 steps [45,46]. These were generated using a feedback shift register algorithm [47,48] with subsequent testing for m-sequence properties. On average 50 % of all grid positions were bright in each stimulus frame and each pixel was bright in 50 % of the stimulus frames. Each stimulus time course was corrected for its temporal delay due to the row-by-row cathode ray scanning of the computer monitor. The spatio-temporal RFs were determined by cross-correlation among the binary luminance time course of each stimulus pixel ("1" for bright and "-1" for dark stimuli) and the retinal LERG- and spike-responses, respectively (Fig. 1B). Thus, a positive cross-correlation can either result from a signal increase in response to a bright stimulus or a signal decrease in response to a dark stimulus. Likewise, a negative cross-correlation can result from signal decrease in response to a bright stimulus or signal increase in response to a dark stimulus. Each multi-focal m-sequence was presented 45 times (total duration 30 min, experiment 140) or 15 times (total duration 10 min, experiments 302, 033, 142) to improve averaged cross-correlations between stimulus and response. The temporal resolution of the stimulus was limited by the monitor frame rate (9.9 ms frame duration). We therefore resampled the stimulus and signal time courses at 50 times the stimulus' temporal resolution (i.e., 0.2 ms) prior to calculating the cross-correlation functions in order to provide for sufficient temporal interpolation.
In a further step we plotted color-coded 2-dimensional maps of the correlation values between each stimulus time course and retinal response at certain temporal delays (Fig. 1C). For ease of interpretation we eightfold interpolated the correlation maps by zero-padding in the frequency domain [49]. Color coding was adjusted for each signal separately in such a way that the lowest negative correlation value was plotted in dark blue and the highest positive correlation value in light yellow. A sequence of such correlation maps for successive temporal delays ("RF time course") can thus be interpreted as the averaged retinal response amplitude or spike probability over time for all visual stimulus positions or as the mean visual stimulus given before a retinal response is recorded [50].
With our multi-electrode approach we were able to map up to seven epiretinal sites at once. From each electrode signal we separated one LERG signal and often several single unit spike trains which were then tested for the center position, time course, and polarity of their RFs (ON- or OFF-center). The RF center positions were estimated from the peak maximum of 2-dimensional RF maps (low-pass-filtered to half of the stimulus grid's spatial frequency). Once the RF center was located, we analyzed the time course of the RF as given by the cross-correlation function among the time course of the optimal stimulus pixel and the retinal signal. In order to improve the signal-to-noise ratio of the estimated RF time course, we averaged the time courses for the optimal stimulus pixel and its adjacent eight stimulus pixels. This was possible since most RFs extended over more than one stimulus pixel width. Visual RF center positions were compared by superposition in a common frame of reference in visual and retinal space (Fig. 3).
The polarity of a RF (ON- vs OFF-center neuron) was defined as the polarity of the first major signal deflection in the cross-correlation function. Moreover, we calculated the time elapsed from the stimulus onset to 70% of the RF peak amplitude as a measure for the retinal response latency. We also analyzed the steepness of the rising and falling phase of the time course for averaged ON and OFF responses, respectively. As a reliable measure for this we used the signal rise time from 30% to 70% of the RF peak amplitude.
The calculated RF center positions were compared to the back-projected electrode tip positions. This way we were able to check for any deviations of functional RF positions from the geometrically expected ones. Based on the work of Stone & Holländer [8], we further identified the approximate trajectory of ganglion cell axon bundles in the vicinity of the electrode position by back-projecting the optic disk and pattern of major retinal blood vessels. All data analyses were performed using the program language IDL (Research Systems Inc., Boulder, CO, U.S.A.).
Authors' contributions
MW performed the experiments, analyzed the data, and drafted the manuscript. RE contributed to experiment conception, design, and the critical revision of the article.
Acknowledgements
We thank Dr. M. Eger and Dr. Th. Schanze for their dedicated assistance with the animal experiments, including data recordings and eye surgeries, respectively. The authors also gratefully thank A. Rentzos, W. Gerber, and M. C. Wilms for their excellent technical and secretarial support. This work was supported by grants from the German Federal Ministry of Education, Science, Research, and Technology (BMFT, grant 01 IN 501 F and KP 0006 to R.E.).
Figures and Tables
Figure 1 Setup for the analysis of retinal RFs. A: The dynamic multifocal visual stimulus was presented on a computer monitor 1.3 m in front of the cat. Retinal signals were recorded by epiretinally positioned electrodes. B: Spikes were extracted from high-pass filtered (0.5–1 kHz) broadband data using an amplitude thresholding approach. LERG and spike signals were cross-correlated with the binary stimulus sequences at all pixel positions (N = 784). C: Example RF maps for the four signals separated from one electrode recording shown in B. White crosshairs indicate the back-projected epiretinal electrode positions that served as reference points for the RF centers. Crosscorrelation values for a 25 ms time delay between stimulus and response are color-coded for each signal separately (dark blue: weak or negative correlation, light yellow: strong positive correlation). The polarity of the LERG-RF is reversed to ease its comparison to the spike-RFs. Note that in this example the neurons responsible for LERG and the medium sized spikes both have RF centers that are congruent with the position of the electrode tip. The neurons responsible for small and large sized spikes, however, have RF centers that are clearly shifted by several degrees of visual angle. All RFs are of ON-type in this example. (Data from experiment 302, right eye)
Figure 2 Retinal RFs of simultaneously recorded LERG and MUA. Simultaneously recorded retinal LERG- (A) and MUA-RFs (B). Individual RF maps are interpolated and arranged according to the hexagonal electrode array configuration. A: All LERG-RFs are congruent with the projected electrode tip positions (white crosshairs). The LERG from electrode 5 was not available in this recording due to an amplifier failure. B: The MUA-RF maps exhibit multiple spatially segregated peaks of positive or negative polarity that are shifted to the left from the projected electrode position. The optical disk is located in the upper right. Color coding as explained in Fig. 1. (Data from experiment 140, right eye)
Figure 3 Superimposed RF-centers for simultaneously recorded LERG and MUA. Superimposed RF centers of the simultaneously recorded retinal LERG- (open circles) and MUA-RFs (filled circles) in Fig. 2. The black cross denotes the area centralis position. Gray arrows approximate the trajectories of ganglion cell axon bundles (reconstructed from [8]). Each color codes one retinal electrode. Note that MUA-RFs are shifted distally from the corresponding LERG-RFs with respect to the optic disk.
Figure 4 Example RF time courses. Examples for the different kinds of retinal RF time courses. A, upper row: Example spike-RFs for one retinal recording position in the right eye in experiment 033. The two RF maps were calculated from spike trains derived from the same broadband recording. Note the ON- and OFF-center RFs that are slightly shifted with respect to each other and shifted distally from the projected electrode position (crosshair). a-d: Time courses of the retinal responses to stimulation with the optimal stimulus (i.e., in the center of the particular RF). From left to right: LERG (a), left RF of first spike train (b), right RF of first spike train (c), left RF of second spike train (d). Response rise times from 30%–70% of the peak amplitude (red) and latencies (blue) are indicated. The baseline (dark blue) is calculated from the average correlation strength between the retinal signal and all pixel luminance time courses. B: Similar results for recordings from the left eye in another experiment (142). Four spike-RFs and corresponding time courses are shown (e-h).
Figure 5 Summary statistics for response latencies and rise times. A: Latencies of LERG-RFs (yellow), pooled spike ON- and OFF-RFs, spike ON-RFs, and spike OFF-RFs (blue histograms with Gaussian fits, N = 18 LERG-RFs in two experiments, N = 53 spike-RFs in three experiments). Brown histograms of data restricted to local spike-RFs. B: Rise times for all spike-RFs (blue) and spike OFF-RFs (black). The upper plot shows pooled data for all three experiments. The lower plot shows data from the two experiments with stepwise response kinetics (experiments 142 and 033). Note here, that the rise times for spike OFF-responses are shorter than for spike ON-responses.
==== Refs
Rodieck RW The vertebrate retina Principles of structure and function 1973 1 W. H. Freeman and Company, San Francisco, U.S.A
Brown KT The electroretinogram: Its components and their origin Vision Res 1968 8 633 677 4978009 10.1016/0042-6989(68)90041-2
Penn RD Hagins WA Signal transduction along retinal rods and the origin of the electroretinographic a-wave Nature 1969 223 201 204 4307228
Miller RF Dowling JE Intracellular responses of the Müller (glial) cells of mudpuppy retina: Their relation to b-wave of the electroretinogram J Neurophysiol 1970 33 323 341 5439340
Newman EA Current source-density analysis of the b-wave of frog retina J Neurophysiol 1980 43 1355 1366 6246222
Karwoski CJ Xu X Current source-density analysis of light-evoked field potentials in rabbit retina Vis Neurosci 1999 16 369 377 10367970 10.1017/S0952523899162163
Kuffler SW Discharge patterns and functional organization of mammalian retina J Neurophysiol 1953 16 37 68 13035466
Stone J Holländer H Optic nerve axon diameters measured in the cat retina: some functional considerations Exp Brain Res 1971 13 498 503 5137299
Radius RL Anderson DR The course of axons through the retina and optic nerve head Arch Ophthalmol 1979 97 1154 1158 109071
Minckler DS The organization of nerve fiber bundles in the primate optic nerve head Arch Ophthalmol 1980 98 1630 1636 6158937
Hammond P Cat retinal ganglion cells: Size and shape of receptive field centers J Physiol 1974 242 99 118 4436829
Hood DC Seiple W Holopigian K Greenstein V A comparison of the components of the multifocal and full-field ERGs Vis Neurosci 1997 14 533 544 9194320
Chichilnisky EJ Kalmar RS Functional asymmetries in on and off ganglion cells of primate retina J Neurosci 2002 22 2737 2747 11923439
Segev R Goodhouse J Puchalla J Berry II MJ Recording spikes from a large fraction of the ganglion cells in a retinal patch Nature Neurosci 2004 7 1155 1162 10.1038/nn1323
Wyatt J Rizzo J Ocular implants for the blind IEEE Spectrum 1996 33 47 53 10.1109/6.490056
Eckmiller R Learning retina implants with epiretinal contacts Ophthalmic Res 1997 29 281 289 9323719
Wilms M Eger M Schanze T Eckhorn R Visual resolution with epi-retinal electrical stimulation estimated from activation profiles in cat visual cortex Vis Neurosci 2003 20 543 555 14977333 10.1017/S0952523803205083
Rodieck RW Ford RW The cat local electroretinogram to incremental stimuli Vis Res 1969 9 1 24 5778041 10.1016/0042-6989(69)90028-5
Brown KT Wiesel TN Analysis of the intraretinal electroretinogram in the intact cat eye J Physiol 1961 158 229 256
Boynton RM Stray light and the human electroretinogram J Opt Soc Am 1953 43 442 449 13070104
Bush RA Sieving PA A proximal retinal component in the primate photopic ERG a-wave Invest Ophthalmol Vis Sci 1994 35 635 645 8113014
Robson JG Saszik SM Ahmed J Frishman LJ Rod and cone contributions to the a-wave of the electroretinogram of the dark-adapted macaque J Physiol 2003 547 509 530 12562933 10.1113/jphysiol.2002.030304
Sieving PA Murayama K Naarendorp F Push-pull model of the primate photopic electroretinogram: a role for hyperpolarizing neurons in shaping the b-wave Vis Neurosci 1994 11 519 532 8038126
Brown KT Wiesel TN Localization of origins of electroretinogram components by intraretinal recording in the intact cat eye J Physiol 1961 158 257 280
Sieving PA Frishman LJ Steinberg RH Scotopic threshold response of proximal retina in cat J Neurophysiol 1986 56 1049 1061 3783228
Wakabayashi K Gieser J Sieving PA Aspartate separation of the scotopic threshold response (STR) from the photoreceptor a-wave of the cat and monkey ERG Invest Ophthalmol Vis Sci 1988 29 1615 1622 3182196
Stone J Sampling properties of microelectrodes assessed in the cats retina Journal of Neurophysiology 1973 36 1071 1079 4761721
Abeles M Goldstein MH Multispike train analysis Proc IEEE 1977 65 762 773
The organization of the retina and visual system
Nelson R Kolb H Synaptic patterns and response properties of bipolar and ganglion cells in the cat retina Vis Res 1983 23 1183 1195 6649437 10.1016/0042-6989(83)90032-9
Cleland BG Levick WR Brisk and sluggish concentrically organized ganglion cells in the cat's retina Journal of Physiology 1974 240 421 56 4421622
Stanford LR Conduction velocity variations minimize conduction time differences among retinal ganglion cell axons Science 1987 238 358 360 3659918
Zrenner E Will retinal implants restore vision? Science 2002 295 1022 1025 11834821 10.1126/science.1067996
Ranck JB Which elements are excited in electrical stimulation of mammalian central nervous system: A review Brain Res 1975 98 417 440 1102064 10.1016/0006-8993(75)90364-9
Marc RE Jones BW Watt CB Strettoi E Neural remodeling in retinal degeneration Progr Ret Eye Res 2003 22 607 655 10.1016/S1350-9462(03)00039-9
McIntyre CC Grill WM Extracellular stimulation of central neurons: Influence of stimulus waveform and frequency on neuronal output J Neurophysiol 2002 88 1592 1604 12364490
Humayun MS Weiland JD Fujii GY Greenberg R Williamson R Little J Mech B Cimmarusti V VanBoemel G Dagnelie G de Juan E Jr Visual perception in a blind subject with a chronic microelectronic retinal prosthesis Vis Res 2003 43 2573 2581 13129543 10.1016/S0042-6989(03)00457-7
Cha K Horch KW Normann RA Mobility performance with a pixelized vision system Vis Res 1992 32 1367 1372 1455709 10.1016/0042-6989(92)90229-C
Sommerhalder J Oueghlani E Bagnoud M Leonards U Safran AB Pelizzone M Simulation of artificial vision: I. Eccentric reading of isolated words and perceptual learning Vis Res 2003 43 269 283 12535986 10.1016/S0042-6989(02)00481-9
Rizzo JF Wyatt J Loewenstein J Kelly S Shire D Perceptual efficacy of electrical stimulation of human retina with a microelectrode array during short-term surgical trials IOVS 2003 44 5362 5369
Reitböck HJ Fiber microelectrodes for electrophysiological recordings J Neurosci Meth 1983 8 249 262 10.1016/0165-0270(83)90038-9
Eckhorn R Thomas U A new method for the insertion of multiple microprobes into neural and muscular tissue, including fiber electrodes, fine wires, needles and microsensors J Neurosci Meth 1993 49 175 179 10.1016/0165-0270(93)90121-7
Schanze T Eckhorn R Hesse L Eger M Wilms M Kossler R Nebeling B Experimental setup for assessing the efficacy and quality of retina implant stimulations by retinal and cortical recording in cat [abstract] Göttingen Neurobiology Report 1998 2 s476
Eger M Information theoretical methods for the functional adjustment of retina implant parameters PhD thesis, Philipps-University Marburg, Germany, 2001 2002 VDI Verlag, Düsseldorf
Reid RC Victor JD Shapley RM The use of m-sequences in the analysis of visual neurons: Linear receptive field properties Vis Neurosci 1997 14 1015 1027 9447685
Sutter EE Marmarelis VZ A practical non-stochastic approach to nonlinear time-domain analysis Advanced methods in physiological system modeling 1987 1 Los Angeles, California, University of Southern California 303 315
Borish J Angell JB An efficient algorithm for measuring the impulse response using pseudorandom noise J Audio Eng Soc 1983 31 478 488
Mac Williams FJ Sloane NJA Pseudo-random sequences and arrays Proc IEEE 1976 64 1715 1729
Elliott DF Rao KR Fast transforms Algorithms, analyses, applications 1982 Academic Press
Eckhorn R Krause F Nelson JI The RF-cinematogram. A cross-correlation technique for mapping several visual receptive fields at once Biol Cybernetics 1993 69 37 55 10.1007/BF00201407
|
16102171
|
PMC1192803
|
CC BY
|
2021-01-04 16:03:48
|
no
|
BMC Neurosci. 2005 Aug 15; 6:50
|
utf-8
|
BMC Neurosci
| 2,005 |
10.1186/1471-2202-6-50
|
oa_comm
|
==== Front
BMC PediatrBMC Pediatrics1471-2431BioMed Central London 1471-2431-5-191598516910.1186/1471-2431-5-19Research ArticleMethacholine bronchial provocation measured by spirometry versus wheeze detection in preschool children Bentur Lea [email protected] Raphael [email protected] Nael [email protected] Asher [email protected] Ori [email protected] Yaacov [email protected] Daphna [email protected] Pediatric Pulmonary Unit, Meyer Children's Hospital, Rambam Medical Center, and the Rappaport Faculty of Medicine, Technion – Israel Institute of Technology, Haifa, Israel2 Pediatric Pulmonary Unit, The Edmond and Lili Safra Children's Hospital, Chaim Sheba Medical Center, Tel-HaShomer, Ramat-Gan, Israel2005 28 6 2005 5 19 19 16 1 2005 28 6 2005 Copyright © 2005 Bentur et al; licensee BioMed Central Ltd.2005Bentur 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
Determination of PC20-FEV1 during Methacholine bronchial provocation test (MCT) is considered to be impossible in preschool children, as it requires repetitive spirometry sets. The aim of this study was to assess the feasibility of determining PC20-FEV1 in preschool age children and compares the results to the wheeze detection (PCW) method.
Methods
55 preschool children (ages 2.8–6.4 years) with recurrent respiratory symptoms were recruited. Baseline spirometry and MCT were performed according to ATS/ERS guidelines and the following parameters were determined at baseline and after each inhalation: spirometry-indices, lung auscultation at tidal breathing, oxygen saturation, respiratory and heart rate. Comparison between PCW and PC20-FEV1 and clinical parameters at these end-points was done by paired Student's t-tests.
Results and discussion
Thirty-six of 55 children (65.4%) successfully performed spirometry-sets up to the point of PCW. PC20-FEV1 occurred at a mean concentration of 1.70+/-2.01 while PCW occurred at a mean concentration of 4.37+/-3.40 mg/ml (p < 0.05). At PCW, all spirometry-parameters were markedly reduced: FVC by 41.3+/-16.4% (mean +/-SD); FEV1 by 44.7+/-14.5%; PEFR by 40.5+/-14.5 and FEF25–75 by 54.7+/-14.4% (P < 0.01 for all parameters). This reduction was accompanied by de-saturation, hyperpnoea, tachycardia and a response to bronchodilators.
Conclusion
Determination of PC20-FEV1 by spirometry is feasible in many preschool children. PC20-FEV1 often appears at lower provocation dose than PCW. The lower dose may shorten the test and encourage participation. Significant decrease in spirometry indices at PCW suggests that PC20-FEV1 determination may be safer.
==== Body
Background
Measurements of bronchial hyper-reactivity (BHR) have provided insight into the physiological basis of asthma, and provide a tool for asthma diagnosis, assessment of asthma severity and response to treatment [1,2]. The bronchial provocation tests require an objective outcome measurement that reflects airway function. Forced expiratory volume in 1 second (FEV1) has been standardized to measure changes in airway caliber that occur with bronchial provocation [3]. In the Methacholine challenge test (MCT), the provocative concentration reducing FEV1 by 20% from baseline (PC20-FEV1) is considered the end point of the test. Traditionally, spirometry in young children has been difficult to achieve. Therefore, techniques that do not require cooperation (i.e., detection of wheeze during normal breathing, a fall of 5% in O2-saturation (SaO2), or an increase of 50% in respiratory rate and/or heart rate) have been used as alternative end points in bronchial provocation tests in the preschool age [4-7]. Recently it has been shown that young children can be taught to perform reliable forced expiratory maneuvers [8-11]. Yet, it is unclear whether these young children have the drive to perform and tolerate repetitive reproducible spirometry-sets that are measured during the interval between inhalations. Concentration of methacholine (MCH) causing wheeze, a fall of 5% in O2- Saturation, an increase of 50% in respiratory rate and/or heart rate (PCW) and PC20-FEV1 were compared in school children and a good correlation was found between the two methods [7,12-14].
This study assesses the ability of young asthmatic preschool children to cooperate with repetitive spirometry-sets during MCT, and thereby allow determination of PC20-FEV1 in comparison with PCW.
Methods
Subjects
Consecutive preschool children referred to the Pediatric Pulmonary Clinic, Meyer Children's Hospital, Rambam Medical Center, Haifa, over a 6-month period were recruited. Of 62 families offered participation in the study, seven refused. None of the children had experienced spirometry previously. Inclusion criteria were: 2.5–6.5 year-old children who were asthmatic according to GINA guidelines [15] with recurrent episodes of wheeze, cough and/or shortness of breath with clinical response to bronchodilator; normal chest auscultation and FEV1 >75% of predicted for healthy preschool children [9] after saline inhalation. Exclusion criteria were: presence of other chronic respiratory conditions; emergency room visit in the past three months; respiratory infection in the past month; oral or inhaled steroids or other anti-inflammatory medication taken in the last week; bronchodilator taken within 24 hours prior to the test.
The Rambam Medical Center Ethics Board approved the study. Parental consent was obtained for each child.
Methacholine challenge
Tests were performed in a designated room at the Pediatric Pulmonary Unit, Meyer Children's Hospital, Haifa, Israel. A parent and the investigating team (a pediatric pulmonary physician, respiratory physiologist and technician) were present throughout the test. MCT was performed according to published guidelines, [3], with doubling doses of fresh Methacholine solutions (0.06 to 8.00 mg/ml) dissolved in saline. Solutions were driven by compressed air of 5 l/min flow (giving a mean output of 0.4 ml/min), and nebulized using a Hudson nebulizer (Hudson RCI, Temecula, CA, USA). Inhalations were performed using a facemask while the child was sitting up straight and breathing normally. Nebulized Methacholine was inhaled for 2 minutes, with 5-minute intervals between doses, until the maximal concentration or the end point was reached. To ensure safety in light of the risk of airway closer, the MCH increment was only half the usual amount when transient wheeze or cough was noted, keeping in mind that the accumulative dose is affected by this manipulation. Oxygen saturation and heart rate were monitored continuously by pulse oximetry (Biox 3700e; Ohmeda). A single observer (LB) performed auscultation for 20 seconds over the trachea and two zones of both lungs (upper front and lower back) according to Springer et al. [7]
The following indices were considered "end of test": appearance of audible wheeze, a fall of ≥5% in O2-saturation, or an increase of ≥50% in respiratory rate and/or heart rate [7]. At the "end of test", spirometric measurements were performed, followed by administration of nebulized Albuterol (2.5 mg).
Spirometry
Forced expiratory flow volume (FEFV) curves were measured with a ZAN100 commercial spirometer (ZAN Messgeraete GmbH, Oberthulba, Germany). Calibration was performed before the testing sessions. The curves were monitored on the computer screen to ensure best effort. Results were corrected to BTPS conditions. The software included an interactive animated computer game (SpiroGame®) set by targets of the FEFV maneuver, combining forced inhalation preceding forced expiration, peak expiratory flow rate (PEFR) and forced vital capacity (FVC) with emphasis on prolonged expiration. [8] The targets were the extrapolated values derived from comparative data from older children, corrected for height. [16] An experienced pulmonary technician instructed each child how to operate the game. Teaching time was limited to 15 minutes. On-line rejection of curves was based on visual inspection for "non-cooperation" errors and included: poor effort; incomplete expiration; cough; glottis closure. Curves had to show a rapid rise to peak flow, and gradual, smooth decline of flow down to residual volume. Baseline maneuvers were repeated to visually obtain best possible efforts on at least 3 technically acceptable FEFV curves. After obtaining baseline spirometry, MCT was performed. A duplicate spirometry set was performed immediately after auscultation. PC20-FEV1 was determined off line by the provocative concentration that reduced FEV1 by 20% from baseline. PC values were log-transformed before statistical analyses. Spirometry indices included FVC, FEV1, PEFR, forced expiratory flow at 50% FVC (FEF50), FEV1/FVC ratio.
Analysis and statistics
Three baseline spirometry curves were analyzed for acceptability criteria according to ATS/ERS guidelines [17,18] and in comparison with similar data for preschool children [11,19]. These included: a) "Start of test" criteria: time to peak expiratory flow and backward extrapolated volume (Vbe) b) "End of test criteria": described by "total expiratory time" and the ratio of "no change in expiratory volume" to "total expiratory time" c) reproducibility (coefficient of variation) of the three baseline curves, calculated as SD/mean*100.
After inhalations, the curves were inspected visually online, and were analyzed offline in relation to baseline using paired t-test. Differences were considered significant when p < 0.05. The level of agreement between the dose at end of test and the dose of PC20 were compared by Bland and Altman analysis (20).
Results
A total of 55 children (28F/27M, age range 2.8–6.4 years) were recruited. Eleven children failed spirometry and underwent MCT by auscultation only. Failure to perform spirometry was due to lack of comprehension (4 children) or failure to repeat spirometry after baseline measurements (7 children). Failure was not age dependent. Eight children refused to cooperate with either test. Thirty-six of 55 (65.5%) children performed the MCT with spirometry tests and with auscultation. Of these 36 children, eleven were 2.5–3.9 years old, 15 were 4–5 years old, and 10 were >5 years old. Three children failed to produce FEV1 on the baseline measurements but were able to produce it after saline administration. In these children, post saline FEV1 measurements were considered as baseline. FEV1 at that point was >75% predicted. The anthropometric data and baseline lung function of the 36 patients are presented in Table 1 and clinical characteristics in Table 2.
Table 1 Anthropometric data and lung function. The results are expressed as mean ± SD.
Anthropometric data Baseline lung function %predicted [16]
N Height (cm) Weight (kg) Sex (M/F) FVC FEV1 FEV1/FVC PEFR FEF50
36 104 ± 7 18 ± 3 20/16 95 ± 15 91 ± 14 96 ± 3 99 ± 14 101 ± 16
Table 2 Clinical Characteristics
N = 36 Recurrent cough Recurrent lung infiltrates Shortness of breath Wheezing Atopy Family history of allergy
N 35 24 24 16 16 23
The 36 children participating in both tests had a previous response to bronchodilators as judged by clinical observation. The average duration of respiratory symptoms was 18 ± 14 weeks. Five children were not receiving any medication for a period of weeks. Nine children were receiving bronchodilators as needed, and 22 were using both inhaled steroids and bronchodilators as needed.
Quality of baseline maneuvers: Start of test: Peak expiratory flow rates were reached within a mean of 98 ± 7 ms (range 89–115 ms) and mean Vbe was 3.4 ± 1.5% of FVC (range 1.2–5.7). Intra-subject reproducibility for the baseline triple maneuvers was: for FVC, 4.1 ± 2.3% (range 1.8–6.3); for FEV1, 3.8 ± 2.3% (range 0.4–7.3); for PEFR, 4.4 ± 2.8% (range 0.3–8.6) and for FEF25–75, 7.9 ± 3.5% (range 2.7–13.2). End of test: Mean expiratory time was 1.48 ± 0.47 seconds and the ratio of "no change in expiratory-volume" to "total expiration time" was 0.20 ± 0.06.
MCT test
Children's response to MCT (n = 36) is summarized in Figure 1 and Table 3. Average test time to reach PC20-FEV1 was 29 ± 11 minutes, while for PCW it was 41 ± 10 minutes (not including bronchodilator administration) (p < 0.001). The end point of the challenge was determined by the pediatric pulmonologist as positive in 35/36 children. One child did not display any of the determined criteria for PCW up to 8 mg/ml and was considered to have no BHR. The mean (± SD) concentration at PCW for the 36 children was 4.26± 3.31 mg/ml. Wheezing at the end point was observed in 26/36 children and in 9/36 the test was ended before the appearance of wheeze due to either oxygen desaturation or tachypnea accompanied by audible long expiration. Mean increase in heart rate at PCW was 25.5 ± 11% (range 10–42%); respiratory rate increased by 30.0 ± 21.1% (range 0–42%) and SaO2 decreased by 6.3 ± 2.7% (range 2.3–10.3%).
Table 3 Appearance of respiratory distress signs at PCW and PC20-FEV1
Symptom Cough Wheeze Prolonged Audible Expiration Decrease SaO2 Increased HR Increased RR
# Children at PCW 32 26 24 33 28 25
# Children at PC20-FEV1 28 2 7 15 3 7
Figure 1 Number of children responding to each MCH concentration (mg/ml) at PCW and at PC20-FEV1
PC20-FEV1 occurred at a mean concentration value of 1.96- ± 1.83 mg/ml. The one child who did not respond to MCH of up to 8 mg/ml by PCW (negative BHR) did not show a fall of 20% from baseline FEV1 value either. The other 35 children exhibited a fall of 20% in FEV1 from baseline values in response to MCH ≤8 mg/ml (Figure 1 and Table 3). A representative set of FEFV curves from a single patient that includes the predicted curve, baseline, PC20-FEV1 and end of test curves is shown in Figure 2.
Figure 2 A representative example of forced expiratory flow-volume curves from one child. Predicted, Baseline, PC20-FEV1 and PCW curves are presented
At PC20-FEV1 there was a mean increase in heart rate of 13.5 ± 11.0%, respiratory rate increased by 15.4 ± 15.8% and SaO2 decreased by 2.4 ± 2.1% from baseline level. These changes were significantly lower than those found at PCW (p < 0.01 for three parameters). The appearance of PC20-FEV1 occurred 2 concentrations earlier than PCW in 5 children, 1.5-concentrations earlier in 3 children, one concentration earlier in 17 children, 0.5 concentrations earlier in 3 children and at the same concentration as PCW in 7 children (Figure 1). The effects of MCH on the spirometry parameters are presented in Table 4. At PC20-FEV1, parameters were moderately decreased, while at end point, test parameters were markedly reduced. The severity of FEV1 reduction at PCW was variable, ranging from 30.8 to 68.2% of baseline. The level of agreement between the dose at end of test (PCW) and the dose at PC20[20] is presented in Figure 3. Dotted lines represent 95% coefficient of variation values.
Table 4 Changes in respiratory indices at PCW and at PC20-FEV1. The results are expressed as mean ± SD. (n = 35/36, as one child did not respond to MCH and his spirometry did not change throughout the test).
Parameter End of test PC20-FEV1
FVC - 41.3 ± 15.5 - 18.4 ± 10.0 *
FEV1 - 44.7 ± 14.5 - 24.6 ± 6.4 *
FEV1/FVC - 6.09 ± 6.8 - 4.1 ± 3.8 *
PEFR - 44.2 ± 13.2 - 21.4 ± 10.6 *
FEF50 - 61.2 ± 14.2 - 38.6 ± 16.9 *
Expiratory time (sec) +2.8 ± 0.4 +2.2 ± 0.4 *
* Changes at PC20-FEV1 are significantly lower than at "end of test", p < 0.01
Figure 3 Analysis of the difference in dose values at end of test (PCW) and the dose at PC20, as compared with mean Dose values of the two, in a Bland and Altman analysis (20). Dotted lines represent 95% coefficient of variation values.
Bronchodilators improved FEV1 by 43 ± 29% from PCW values and all respiratory symptoms disappeared shortly after bronchodilator administration.
Discussion
In this study we assessed the feasibility of determining PC20-FEV1 during Methacholine bronchial provocation testing in asthmatic preschool children. We found MCT was feasible in 65% of this group of wheezy preschool children. Children as young as 3 years old complied and cooperated with what seems to be a most fatiguing procedure. Baseline measurements met most of the ATS criteria for older children and adults [17,18] and quality control studies on spirometry in preschool children [11,19]. We found that PC20-FEV1 correlates with PCW. However, PC20-FEV1 frequently precedes PCW. All spirometry parameters at PC20-FEV1 were significantly higher than those measured at PCW.
In this study, we used interactive spirometry games [8] with multiple spirometry targets, since single targeted games (usually peak expiratory flow targeted) have not fulfilled expectations [21,22]. Our teaching method is supported by the findings that 65% of the children fully cooperated not only with baseline measurements but also with spirometry sets. Of note, 26 of the 36 children were younger than 5 years. Conforming quality control was necessary to proceed with the test. The quality control of baseline spirometry in our study met most ATS/ERS criteria concerning reproducibility and start of test criteria [17,18] and matched those reported for preschool children [11,19], encouraging us to continue with the MCT test. Vbe = 5%FVC found in our study is narrower than reported [11], as we have rejected in advance curves with Vbe >5%FVC at the expense of success rate. It should be stressed that our work did not compare verbal coaching [9] or other spirometry games [11,23] as the preferable methodology for keeping the child going and performing repetitive spirometry sets.
The mean PC20-FEV1 of 1.96 ± 1.83 mg/ml found in our group reflects a mild degree of BHR, as we recruited children with mild asthmatic symptoms. Our findings for PC20-FEV1 are comparable to those of Hayden et al [13], who found a mean PC20-FEV1 at FEV0.5 of 2.49 ± 2.55 mg/ml in infants. Adinoff et al [24] reported a mean provocative dose of 3.0 mg/ml Methacholine in their preschool children and infants. Tepper [25] et al. reported that infants with asthma-like respiratory symptoms might respond to MCH concentrations as low as 1.25 mg/ml. In that respect we found that PC20-FEV0.5 occurred at a mean concentration value of 1.29- ± 1.47 mg/ml, meaning that the responsiveness of the airways in the preschool age may be similar to that of infants, despite differences in the measurement techniques. It is important to note that PC20-FEV0.5 occurred at a significant mean lower concentration than PC20-FEV1 (1.96- ± 1.83 mg/ml; p < 0.01), however, standardization is needed to accept the PC20-FEV0.5 value for the determination of hyper-reactive airways.
PCW
We found that PCW occurred in our group at a mean concentration of 4.26 ± 3.31 mg/ml. PCW values in our study were much higher than the PCW (0.4 mg/ml) reported by Springer et al [7]. The difference may be attributed to inclusion of more severe asthmatics in their study group.
Spirometry at PCW
We found that PC20-FEV1 occurred at a lower concentration than PCW in most subjects. This finding is in agreement with several other studies comparing PCW detection to PC20-FEV1 in school age children. [4-7,14]. However, in none of these studies were spirometry measurements carried out to the point of wheeze. We expected to find a good correlation between the two tests (PCW = 1.2195*PC20-FEV1 + 0.0288; R2 = 0.9733; p < 0.005), yet the Blant and Altman analysis revealed that in children with higher mean provocation PCW dose (≥6 mg), the level of agreement between the methods was low, reflecting higher sensitivity of the PC20 method, especially in mild airway reactivity (Figure 3).
We found that at PCW, FEFV curves visually seemed to be smaller and all parameters were reduced simultaneously (Figure 2), with a highly significant reduction in flows and volume parameters. The reduction in curve was gradual in most children, accompanied by an increase in respiratory symptoms (Table 2), and responded to bronchodilators, and hence was not considered to reflect fatigue. To further strengthen this point a representative curve of one child illustrating, a poor effort performed at teaching process vs. end of test curve is shown in figure 4. The poor-effort curve did not fulfill start of test criteria and is round while the "end of test curve" has an obstructed shape.
Figure 4 A representative example of poor-effort forced expiratory flow-volume curves from one child. Baseline, Post challenge and poor effort during teaching process are presented.
The reduced FVC and flows are most likely due to a severe degree of airway narrowing involving small to medium airways that may be accompanied by air trapping, partial closure of airways and elevation in FRC. Reduced FVC may also be due to increased glottic narrowing due to MCH irritation [26,27], but the flow volume curve was not suggestive of upper airway obstruction (trimmed PEFR). Alternatively, the upper airways response to methacholine may contribute to the increase in total respiratory resistance [27]. This pattern occurred in some cases before appearance of wheeze or other clinical end-points. Indeed, in 9/36 subjects, the test was terminated due to oxygen desaturation or tachypnea rather than wheeze. Similar to our results, Sprikkelman et al [28] reported that wheeze was detected in only 33% of 15 school-age asthmatic children at PC20-FEV1, and Springer et al [7] terminated the test without the presence of wheeze in 19.2% of young children. In this respect we would argue that FEV1 does make a contribution beyond simply asking the subject if they wheeze. Novitzki et al (4) found in 5–8 year-old children that FEV1 is decreased by 33.3 ± 7.4% at PCW. Spence et al [29] reported a mean fall of 51 ± 14% from baseline FEV1 when wheeze appeared in their asthmatic older subjects. Our results strengthen these prior findings, and suggest that spirometric PC20-FEV1 may be achieved with inhalation of lower MCH concentrations than those used to achieve wheeze.
Measuring PCW during tidal volume breathing has the advantage that no active cooperation on the child's part is needed. Therefore the success rate of PCW is higher than spirometry (44/55 children). However, using PC20-FEV1 (or PC20-FEV0.5) can preclude inhalation of higher concentrations of MCH used to achieve wheeze, leading to alarmingly diminished flows found at PCW and a significant shortening of test time relative to PCW.
Conclusion
We conclude that PC20-FEV1 is feasible in preschool asthmatic children when using respiratory games teaching techniques and that the children tolerate repetitive duplicate sets of spirometry maneuvers. PC20-FEV1 in preschool children appears to be as sensitive as in adults and school children. Yet, many questions remain open as to the usefulness of this test in a random sample of young children and/or how discriminating this test is as a diagnostic tool. It would also be necessary to assess the sensitivity of this test to various severities of disease. Further studies are needed for standardization and definition of methodological criteria.
Competing interests
The author(s) declare that they have no competing interest.
The SpiroGame program is privately patented in USA, granted to Dr. Vilozni.
Dr. Vilozni does not foresee any financial gain or loss, now or in the future from publishing this manuscript. The patent is not commercialized.
Authors' contributions
Dr. Lea Bentur and Dr. Daphna Vilozni had primary responsibility for protocol development, outcome assessment, data analysis and writing of the manuscript.
Dr. Raphael Beck, Dr. Nael Elias, Dr. Asher Barak, Dr. Ori Efrati and Prof. Yaacov Yahav contributed to this study by patients screening, patient enrollment, analysis of the data and quality control of the data.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
The Study was funded by the Israel Lung Association, Tel-Aviv, Israel
==== Refs
Cockcroft DW Bronchoprovocation methods: direct challenges Clin Rev Allergy Immunol 2003 24 19 26 12644716 10.1385/CRIAI:24:1:19
Spiropoulos K Stevens J Eigen H Spiropoulos A Specificity and sensitivity of methacholine challenge test in children with normal and hyper-reactive airways Acta Paediatr Scand 1986 75 737 743 3564942
Guidelines for Methacholine and exercise challenge testing - 1999. The Official Statement of the American Thoracic Society Am J Respir Crit Care Med 2000 161 309 329 10619836
Noviski N Cohen L Springer C Bar-Yishay E Avital A Godfrey S Bronchial provocation determined by breath sounds compared with lung function Arch Dis Child 1991 66 952 955 1929491
Wilson NM Bridge P Silverman M The measurement of methacholine responsiveness in 5-year-old children: three methods compared Eur Respir J 1995 8 364 370 7789478 10.1183/09031936.95.08030364
Wilts M Hop WC van der Heyden GH Kerrebijn KF de Jongste JC Measurement of bronchial responsiveness in young children: comparison of transcutaneous oxygen tension and functional residual capacity during induced bronchoconstriction and dilatation Pediatr Pulmonol 1992 12 181 185 1641275
Springer C Godfrey S Picard E Uwyyed K Rotschild M Hanania S Noviski N Avital A Efficacy and Safety of Methacholine Bronchial Challenge performed by auscultation in young asthmatic children Am J Respir Crit Care Med 2000 162 857 860 10988095
Vilozni D Barker M Jellouschek H Heimann G Blau H An interactive computer-animated system (SpiroGame) facilitates spirometry in pre-school children Am J Respir Crit Care Med 2001 164 2200 2205 11751188
Eigen H Bieler H Grant D Christoph K Terrill D Heilman DK Ambrosius WT Tepper RS Spirometric pulmonary function in healthy preschool children Am J Respir Crit Care Med 2001 163 619 623 11254514
Zapletal A Chalupova J Forced expiratory parameters in healthy preschool children (3–6 years of age) Pediatr Pulmono 2003 35 200 207 10.1002/ppul.10265
Aurora P Stocks J Oliver C Saunders C Castle R Chaziparasidis G Bush A Quality control for spirometry in preschool children with and without lung disease Am J Respir Crit Care Med 2004 169 1152 1159 15028561 10.1164/rccm.200310-1453OC
Sprikkelman AB Grol MH Lourens MS Gerritsen J Heymans HS van Aalderen WM Use of tracheal auscultation for the assessment of bronchial responsiveness in asthmatic children Thorax 1996 51 317 319 8779140
Hayden MJ Devadason SG Sly PD Wildhaber JH LeSouef PN Methacholine responsiveness using the raised volume forced expiration technique in infants Am J Respir Crit Care Med 1997 155 1670 1675 9154874
Martinez FC Yarza GPE Ruiz AA Knorr EJI Blecua CM Aramburu MJ Agreement between tracheal auscultation and pulmonary function in methacholine bronchial inhalation challenge in asthmatic children An Esp Pediatr 2002 56 304 309 11927097
Global Strategy for Asthma Management and Prevention 1995 National Institutes of Health, National Heart, Lung and Blood Institute, Bethesda
Zapletal A Samanek M Prague TP Lung function in children and adolescents: Methods, reference values Basel, Karger 1987
Standardization of spirometry-1994 update. Statement of the American Thoracic Society Am J Respir Crit Care Med 1995 153 1107 1236
Quanjer PH Tammeling GJ Cotes JE Pedersen OF Peslin R Yernault JC Lung volumes and forced ventilatory flows. Report Working Party Standardization of Lung Function Tests, European Community for Steel and Coal. Official Statement of the European Respiratory Society Eur Respir J 1993 5 40 8381090
Crenesse D Berlioz M Bourrier T Albertini M Spirometry in children aged 3 to 5 years: reliability of forced expiratory maneuvers Pediatr Pulmonol 2001 32 56 61 11416877 10.1002/ppul.1089
Bland JM Altman DG Statistical methods for assessing agreement between two methods of clinical measurement Lancet 1986 1 307 310 2868172
Nystad W Samuelsen SO Nafstad P Edvardsen E Stensrud T Jaakkola JJ Feasibility of measuring lung function in preschool children Thorax 2002 57 1021 1027 12454295 10.1136/thorax.57.12.1021
Gracchi V Boel M Van der Laag J Van der Ent CK Spirometry in young children: should computer-animation programs be used during testing? Eur Respir J 2003 2 872 875 12765436 10.1183/09031936.03.00059902
Kozlowska W Aurora P Stocks J The use of computer-animation programs during spirometry in preschool children Eur Respir J 2004 2 494 495 15065846
Adinoff AD Schlosberg RT Strunk RC Methacholine inhalation challenge in young children: results of testing and follow-up Ann Allergy 1988 61 282 286 3177971
Tepper RS Airway reactivity in infants: a positive response to methacholine and metaproterenol J Appl Physiol 1987 62 1155 1159 3553140
England SJ Ho V Zamel N Laryngeal constriction in normal humans during experimentally induced bronchoconstriction J Appl Physiol 1985 58 352 356 3884571
Marchal F Loos N Monin P Peslin R Methacholine-induced volume dependence of respiratory resistance in preschool children Eur Respir J 1999 14 1167 74 10596708 10.1183/09031936.99.14511679
Sprikkelman AB Schouten JP Lourens MS Heymans HS van Aalderen WM Agreement between spirometry and tracheal auscultation in assessing bronchial responsiveness in asthmatic children Respir Med 1999 93 102 107 10464860 10.1016/S0954-6111(99)90298-6
Spence DP Graham DR Jamieson G Cheetham BM Calverley PM Earis JE The relationship between wheezing and lung mechanics during methacholine induced broncho-constriction in asthmatic subjects Am J Respir Crit Care Med 1996 154 290 294 8756796
|
15985169
|
PMC1192804
|
CC BY
|
2021-01-04 16:31:07
|
no
|
BMC Pediatr. 2005 Jun 28; 5:19
|
utf-8
|
BMC Pediatr
| 2,005 |
10.1186/1471-2431-5-19
|
oa_comm
|
==== Front
BMC PediatrBMC Pediatrics1471-2431BioMed Central London 1471-2431-5-301609295710.1186/1471-2431-5-30Research ArticleA survey of transcutaneous blood gas monitoring among European neonatal intensive care units Rüdiger Mario [email protected]öpfer Kerstin [email protected] Hannes [email protected] Gerd [email protected] Roland R [email protected] Clinic of Neonatology; Universitätsmedizin Berlin, Charité-Mitte; 10098 Berlin; Germany2 Department for Neonatology, Medical University Innsbruck, Department for Neonatology, 6020 Innsbruck, Austria2005 10 8 2005 5 30 30 17 12 2004 10 8 2005 Copyright © 2005 Rüdiger et al; licensee BioMed Central Ltd.2005Rüdiger 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
PCO2 and PO2 are important monitoring parameters in neonatal intensive care units (NICU). Compared to conventional blood gas measurements that cause significant blood loss in preterms, transcutaneous (tc) measurements allow continuous, non-invasive monitoring of blood gas levels. The aim of the study was to survey the usage and opinions among German speaking NICUs concerning tc blood gas monitoring.
Methods
A questionnaire was developed and sent to 56 head nurses of different NICUs in Germany, Switzerland and Austria.
Results
A completely answered questionnaire was obtained from 41 NICUs. In two of these units tc measurements are not performed. In most NICUs (77%), both PtcO2 and PtcCO2 are measured simultaneously. Most units change the sensors every 3 hours; however, the recommended temperature of 44°C is used in only 15% of units. In only 8% of units are arterial blood gases obtained to validate tc values. Large variations were found concerning the targeted level of oxygen saturation [median upper limit: 95% (range 80–100%); median lower limit: 86% (range 75–93%)] and PO2 [median upper limit: 70 mmHg (range 45–90 mmHg); median lower limit: 44 mmHg (range 30–60 mmHg)].
Conclusion
Our survey shows that the use of tc monitors remains widespread among German speaking NICUs, despite earlier data suggesting that their use had been abandoned in many NICUs worldwide. In addition, we suggest that the current method of monitoring oxygenation may not prevent hyperoxemia in preterm infants.
==== Body
Background
Preterm infants are vulnerable to alterations in arterial oxygen or carbon dioxide tension [1]. Changes in oxygen supply contribute to the subsequent development of retinopathy of prematurity or bronchopulmonary dysplasia [2]. Hypocarbia has been associated with the subsequent development of periventricular leucomalacia [3] and cerebral palsy [4], and while hypercarbia may protect the perinatal brain from hypoxemic-ischemic damage [5,6], it could also cause retardation of retinal vascularization [7].
Despite an ongoing discussion concerning the optimal values of blood gas levels, there is consensus that the partial pressures of arterial oxygen and carbon dioxide (PaO2 and PaCO2) should be kept within a narrow range. Thus, intermittent or continuous determination of blood gases is required. However, the repeated arterial blood sampling that is required for the correct measurement of PaO2 and PaCO2 is difficult to perform in preterm infants because the usage of indwelling catheters is associated with complications and significant blood loss. Capillary blood samples, which are painful but easier to obtain, provide satisfactory values for PaCO2 but tend to underestimate PaO2 [8].
Transcutaneous (tc) measurement of oxygen (PtcO2) and carbon dioxide (PtcCO2) tension is a non-invasive method that has recently offered some promise [1]. Several studies have shown a good correlation between tc and arterial values [9-11]. However, during routine clinical treatment, several problems – such as burns – appear [1]. Furthermore, a poor correlation between PaO2 and PtcO2 was found under routine clinical conditions [12]. On the basis of these reports, alert letters on the subject of tc PO2 measurements were published by Canadian and British health authorities [13,14], and after clinical introduction of pulse oximetry, the interest in tc oxygen monitoring decreased was abandoned altogether in many neonatal intensive care units (NICUs) around the world [15]. However, the actual status of tc blood gas monitoring in German speaking NICUs remains unknown.
The present observational study was performed to answer the following questions:
1.) To what extent is tc blood gas monitoring performed in German speaking NICUs?
2.) Given reports that nurses are reluctant to perform tc monitoring because they question its reliability [15], what are the opinions of nurses concerning the reliability of tc values?
3.) Are there any differences between NICUs concerning technical aspects of tc sensor application?
4.) What methods are used to detect hypo- or hyperoxia and what are the upper and lower limits for oxygen saturation and partial pressure in different NICUs?
Methods
The questionnaire consisted of four main parts and is described below. A pre-test of the questionnaire was performed at the authors' institution. Twenty nurses were asked to answer the questions. Four of the questions in the original version were found to be misleading and were reworded for the final version of the questionnaire.
The final questionnaire was sent to NICUs by ordinary mail. To avoid any bias due to differences in national medical regulations, the questionnaire was distributed only in German speaking countries. From a list of 168 university hospitals in Germany, Austria and Switzerland every third unit was chosen (n = 56 units). Because nurses are mainly responsible for the usage of tc equipment, head nurses were asked to answer the questionnaire according to their institutional guidelines.
Usage of tc measurements
The first part of the questionnaire was designed to obtain information concerning the usage of tc measurements. The following questions were included:
1. What aged patients do you mainly care for (only preterm infants / preterms and neonates / neonates and older infants)?
2. On which patients do you perform tc measurements (conventional mechanically ventilated patients / CPAP patients / only supplemental oxygen patients / every patient)?
3. Which parameters do you measure (tc PO2 / tc PCO2 / both)?
4. What manufacturer does the monitoring system come from?
Nursing practice
The second part of the questionnaire consisted of questions concerning nursing practice during tc measurements:
1. How often do you change the site of the sensor (every 1 / 2 / 3 hours / more / less frequently)?
2. Do you think the changes violate "minimal handling" practices?
3. Do you use a special treatment for erythematous skin areas?
Technical details of tc monitoring
The third section investigated technical details of tc usage and included the following questions:
1. How often do you change the sensor site?
2. What is the temperature of the sensor?
Accuracy of transcutaneous measurements
The final part of the questionnaire was dedicated to the correlation of tc and invasive blood gas measurements. The following questions were asked:
1. What is your impression concerning the accuracy of tc measurements (good / moderate / poor)?
2. On average, how often do you compare tc values with blood gases (routinely / depending on the values)?
3. What source of blood do you use for validation (capillary / arterial / venous)?
4. Concerning monitoring of oxygenation in preterm infants, which value is more important when estimating hypoxia (saturation / tc PO2) and hyperoxia (saturation / tc PO2)? What are the lower and upper limit values?
Statistics
Data were analyzed with descriptive statistics using Excel (Microsoft) software. Data are presented as median and range or as relative percentages where appropriate.
Results
The questionnaire was completed by 41 of the 56 NICUs (73%). Among the 41 units with completed questionnaires, 2 did not perform tc measurements and hence were excluded from the subsequent analysis.
The head nurses of the 15 non-responding units were contacted by telephone by which it was confirmed that no tc measurements were performed in 8 units, tc measurements were performed but no further information was voluntarily offered in 4 units, and no information at all was offered in 3 units.
Usage of tc measurements
Most of the evaluated NICUs (28/39) mainly care for preterm and term new-borns. In the remaining 11 NICUs, both new-borns and infants are treated. Most of the units perform tc measurements on mechanically ventilated infants or on infants on continuous positive pressure (CPAP) support (Fig. 1A). About 30% of units also use tc monitoring for infants on supplemental oxygen.
Figure 1 Shown are the number of NICUs. (A) that use tc monitoring on patients on conventional mechanical ventilation (CMV), continuous positive airway pressure support (CPAP) or supplemental oxygen (suppl. oxygen) [multiple answers were possible]; (B) that use a combination of tc PO2 and tc PCO2 sensors (Combination), or a single sensor; (C) that use a sensor temperature of 42°, 42.5°, 43°, 43.5° or 44°C; (D) that compare tc values with blood gases routinely every 6, 8, 12 or 24 h or do not have a specified routine.
The majority of the answering units use a combination of tc PO2 and tc PCO2 sensors. Some units use either tc PO2 or tc PCO2 and two units use both sensors separately (Fig. 1B).
Devices for tc blood gas measurements were from multiple suppliers; however, Radiometer was the most commonly used manufacturer, followed by Hewlett Packard and Hellige.
Nursing practice
Analysis of handling showed that most units change the site of the sensor every 3 hours or even more frequently, 6 of 39 units change the sensor every 4 hours, and 3 of 39 less often than every 4 hours. There was no correlation between frequency of changes and manufacturer.
In 60% of the participating units, nurses considered the changing of the sensor as a discomfort for the patient and a violation of the minimal handling policy.
About one third of the units do not have a special treatment of erythematous sensor areas, whereas the remaining units use various ointment therapies.
Technical details of tc measurements
Large differences were found concerning the technical aspects of tc blood gas monitoring. In 17 of 39 units, sensors are calibrated after each change of sensor site. A routine calibration of the sensor is performed every 4 hours in 8 units and once daily in 11 units.
Sensor temperature mainly depends on the age of the patient; however, in most units the sensor works at a temperature of 43°C. In some units temperatures between 42 and 44°C are used (Fig. 1C).
Individual opinion concerning the accuracy of transcutaneous values
Invasive blood gas measurements are routinely performed for comparison with tc values in 14 of 39 units (Fig. 1D). Blood gases are mainly obtained from capillary blood, with only 8% of units obtaining arterial samples (Fig. 2).
Figure 2 Distribution of blood sample type used to compare tc values.
The majority of respondent nurses considered the tc measurements as a good (29/39) or intermediate (9/39) estimate of arterial blood gases, while only one NICU nurse stated that tc measurements lead to poor estimates.
The question concerning hyperoxia and hypoxia detection was answered by only 35 units, but these provided some interesting data. To detect hyperoxia in preterm infants, 16 of the 35 NICUs use only oxygen saturation, 10 use only PtcO2, 8 use both and 1 uses neither method. The median upper limit for saturation was 95% (range 80–100%) and the median upper limit for PtcO2 was 70 mmHg (range 45–90 mmHg).
To detect hypoxia, the majority (24/35) of units use only saturation, whereas 9/35 units use both methods (PtcO2 and saturation) and two units use only invasive blood gas methods. The median lower limit for saturation was 86% (range 75–93%) and the median lower limit for PtcO2 was 44 mmHg (range 30–60 mmHg).
Discussion
To prevent acute or chronic damage, blood gases must be monitored in preterm infants [1]. Transcutaneous (tc) measurement of blood gases represents a valuable tool for continuous, non-invasive monitoring.
Tc monitoring is associated with several problems and it has been reported that this type of monitoring has been abandoned in many NICUs around the world [15,16]. Up until now, no data were available concerning the usage of tc monitors in German speaking neonatal units. In our observational study, we received answers from 41 of 56 NICUs approached. Of these, 39 responded that they currently use tc monitoring. Four of the 15 non-responding units also performed tc measurements but did not provide any further information. Thus, our representative survey suggests that at least 43 of 56 NICUs (77%) use tc blood gas monitoring. In six units tc monitoring had been completely abandoned. In contrast to data that suggest a reluctance of nurses to use tc monitoring [15], our study shows wide acceptance of the technique among NICU nurses. The majority of surveyed nurses stated that the accuracy of the tc readings is mostly reliable. However, the need for frequent changes in sensor sites was considered a violation of the minimal handling policy.
Tc monitoring and pulse oximetry are useful techniques for the non-invasive monitoring of oxygenation in new-borns that require supplemental oxygen. Whereas capillary blood gases and pulsoximetry are sufficient to detect hypoxia, it is not sufficient to use either method to prevent hyperoxia. Nevertheless, in the present survey 16 of 35 NICUs used only saturation to detect hyperoxia. Since pulse oximetry values cannot be used to detect hyperoxia, arterial PO2 should also be measured intermittently. About half of all answering units stated that they perform blood gas analysis exclusively from capillary blood samples, but capillary PO2 estimations can only exclude hypoxia and are insufficient for detecting hyperoxia [17]. Thus, it could be speculated that the current oxygen-monitoring policy of some units exposes infants requiring supplemental oxygen to a higher risk of subsequent development of oxygen associated damage, such as retinopathy [18].
Large variations were found among the different NICUs with regard to the definition of hypoxia and hyperoxia. The upper limit for oxygen saturation ranged between 80% and 100%, and if tc measurements were used, the upper limit ranged between 45 and 90 mmHg (median 70 mmHg). A similarly wide range was found for the detection of hypoxia with a lower saturation border between 75% and 93% (mean 86%). These differences are substantial and could explain some of the described differences in outcomes of preterm infants [19]. These differences require further investigation and specification. The present data do not allow a differentiation between the target values for infants with supplemental oxygen or those with respiratory support.
The present study included some limitations that are partially associated with the chosen method of obtaining information. First, the study sample is based on the return of completed questionnaires. We achieved a return rate of 73%, which is considered a good result and allows reliable interpretation. Secondly, the questionnaire was not designed to identify an association between the monitoring policy at the institution and the clinical outcome parameters; however, the present study does provide sufficient data to plan an appropriate study protocol to address that question. Finally, the questionnaire was only designed to receive information consistent with institutional guidelines. In some cases, the unique situation of an individual patient could lead to deviations from the general policy. A follow-up study could further specify the use of tc monitoring under different clinical conditions (ventilation, oxygen supply, CPAP) and in different populations (preterm, term infants), and could also include the primary reason for the use of tc monitoring. However, surveys of these factors should be mainly performed among the attending neonatologists.
Conclusion
The present survey provides valuable data concerning the current situation of routine clinical blood gas monitoring in German speaking NICUs and has produced the following conclusions: 1) Transcutaneous blood gas monitoring is frequently used in neonatal intensive care units; 2) large variations exist concerning the targeted range of oxygen saturation or PO2; and 3) in infants requiring supplemental oxygen, the current method of monitoring oxygen may not be sufficient to prevent hyperoxia.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
MR developed the study design and drafted the manuscript. KT conceived the study, developed the study design, distributed the questionnaire and collected the data. GS was significantly involved in drafting and revising the article and contributed to analysis and interpretation of data. HH helped to coordinate the study and to develop the questionnaire. RW contributed to the conception and design of the study and the interpretation of data.
Pre-publication history
The pre-publication history for this paper can be accessed here:
==== Refs
Brouillette RT Waxman DH Evaluation of the newborn's blood gas status Clin Chem 1997 43 215 221 8990256
Saugstad OD Oxygen toxicity in the neonatal period-a review Acta Paediatr Scand 1990 79 881 892 2264459
Ambalavanan N Carlo W Hypocapnia and hypercarbia in respiratory management of newborn infants Clin Perinatol 2002 28 517 531 11570152 10.1016/S0095-5108(05)70104-4
Collins MP Lorenz JM Jetton JR Paneth N Hypocapnia and other ventilation related risk factors for cerebral palsy in low birth weight infants Pediatr Res 2001 50 712 719 11726729
Vannucci RC Towfighi J Heitjan DF Brucklacher RM Carbon dioxide protects the perinatal brain from hypoxic-ischemic damage: an experimental study in the immature rat Pediatrics 1995 95 868 874 7761212
Mariani G Cifuentes J Carlo W Randomized trial of permissive hypercapnia in preterm infants Pediatrics 1999 104 1082 1088 10545551 10.1542/peds.104.5.1082
Holmes JM Zhang S Leske DA Lanier WL Carbon dioxide-induced retinopathy in the neonatal rat Curr Eye Res 1998 17 608 616 9663850 10.1076/ceyr.17.6.608.5176
Eaton T Rudkin S Garret JE The clinical utility of arterialized earlobe capillary blood in the assessment of patients for long-term oxygen therapy Respir Med 2001 95 655 660 11530953 10.1053/rmed.2001.1118
Binder N Atherton H Thorkelsson T Hoath SB Measurement of transcutaneous carbon dioxide in low birthweight infants during the first two weeks of life Am J Perinatol 1994 11 237 241 8048993
Geven WB Nagler E deBoo T Lemmens W Combined transcutaneous oxygen, carbon dioxide tensions and end-expired CO2 levels in severely ill newborns Adv Exp Med Biol 1987 220 115 120 3118653
Lofgren O Henriksson P Jacobson L Johansson O Transcutaneous PO2 monitoring in neonatal intensive care Acta Paediatr Scand 1978 67 693 697 716868
Yip WC Tay JS Wong HB Ho TF Reliability of transcutaneous oxygen monitoring of critical ill children in a general pediatric unit Clin Pediatr (Phila) 1983 22 431 435 6839623
Health and Welfare Canada HPB Transcutaneous infant PO2 monitors, Medical Devices Alert No73 Canada 1985
Department of Health and Social Security Transcutaneous oxygen monitors, Hazard Notice R/M 1069/4493 New York 1980
Miké V Krauss A Ross GS Doctors and the health industry: a case study of transcutaneous oxygen monitoring in neonatal intensive care Soc Sci Med 1996 42 1247 1258 8733195 10.1016/0277-9536(95)00222-7
Miké V Krauss A Ross GS Responsibility for clinical innovation. A case study in neonatal medicine Eval Health Prof 1998 21 3 26 10183338
McLain BI Evans J Dear PR Comparison of capillary and arterial blood gas measurements on neonates Arch Dis Child 1988 63 743 747 3137897
Flynn JT Bancalari E Snyder F Goldberg R Feuer W Cassady J Schiffman J Feldman HI Bachynski B Buckley E A cohort study of transcutaneous oxygen tension and the incidence and severity of retinopathy of prematurity N Engl J Med 1992 326 1050 1054 1549150
Vohr BR Wright LL Dusick AM Perritt R Poole WK Tyson JE Steichen JJ Bauer CR Wilson-Costello DE Mayes LC Center differences and outcomes of extremely low birth weight infants Pediatrics 2004 113 781 789 15060228 10.1542/peds.113.4.781
|
16092957
|
PMC1192805
|
CC BY
|
2021-01-04 16:31:06
|
no
|
BMC Pediatr. 2005 Aug 10; 5:30
|
utf-8
|
BMC Pediatr
| 2,005 |
10.1186/1471-2431-5-30
|
oa_comm
|
==== Front
BMC Public HealthBMC Public Health1471-2458BioMed Central London 1471-2458-5-771602949510.1186/1471-2458-5-77Study ProtocolRationale, design and conduct of a comprehensive evaluation of a primary care based intervention to improve the quality of life of osteoarthritis patients. The PraxArt-project: a cluster randomized controlled trial [ISRCTN87252339] Rosemann Thomas [email protected]örner Thorsten [email protected] Michel [email protected] Jochen [email protected] Christiane [email protected] Stefanie [email protected] Joachim [email protected] Department. of General Practice and Health Services Research, University of Heidelberg, Voßstr. 2, 69115 Heidelberg, Germany2 Centre for Quality of Care Research, Radboud University Medical Centre Nijmegen, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands3 Institute for General Practice, Chronic Care and Health Services Research Unit, University of Frankfurt, Theodor-Stern-Kai 7, 60590 Frankfurt a.M. Germany2005 19 7 2005 5 77 77 8 6 2005 19 7 2005 Copyright © 2005 Rosemann et al; licensee BioMed Central Ltd.2005Rosemann 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
Osteoarthritis (OA) has a high prevalence in primary care. Conservative, guideline orientated approaches aiming at improving pain treatment and increasing physical activity, have been proven to be effective in several contexts outside the primary care setting, as for instance the Arthritis Self management Programs (ASMPs). But it remains unclear if these comprehensive evidence based approaches can improve patients' quality of life if they are provided in a primary care setting.
Methods/Design
PraxArt is a cluster randomised controlled trial with GPs as the unit of randomisation. The aim of the study is to evaluate the impact of a comprehensive evidence based medical education of GPs on individual care and patients' quality of life. 75 GPs were randomised either to intervention group I or II or to a control group. Each GP will include 15 patients suffering from osteoarthritis according to the criteria of ACR.
In intervention group I GPs will receive medical education and patient education leaflets including a physical exercise program. In intervention group II the same is provided, but in addition a practice nurse will be trained to monitor via monthly telephone calls adherence to GPs prescriptions and advices and ask about increasing pain and possible side effects of medication.
In the control group no intervention will be applied at all. Main outcome measurement for patients' QoL is the GERMAN-AIMS2-SF questionnaire. In addition data about patients' satisfaction (using a modified EUROPEP-tool), medication, health care utilization, comorbidity, physical activity and depression (using PHQ-9) will be retrieved.
Measurements (pre data collection) will take place in months I-III, starting in June 2005. Post data collection will be performed after 6 months.
Discussion
Despite the high prevalence and increasing incidence, comprehensive and evidence based treatment approaches for OA in a primary care setting are neither established nor evaluated in Germany. If the evaluation of the presented approach reveals a clear benefit it is planned to provide this GP-centred interventions on a much larger scale.
==== Body
Background
Arthritis is a most frequent affection of joints and a common condition in general practice (roughly 60–80 patients per thousand cases with an average of 5.7 GP contacts per quarter) [1,2]. The GP is the primary contact for arthritis patients and the main care provider for most patients. Previous studies, including our own qualitative pilot study, have shown that arthritis related pain and fear of increasingly reduced mobility represent the most important burden for arthritis patients [3]. There was a large general need for information among patients concerning as for instance individual options to influence the course of the disease [3,4]. GPs approaches to OA varied widely and patient education concerning life style and motivation for physical activity was mostly vague and unsighted [5]. An important need for information about evidence based pain management according to WHO-recommendations was detected among GPs. Several studies underlined the effectiveness of complex interventions with active patient involvement such as the "Arthritis-self-management-programs" (ASMPs) in the US and Canada. However, these programs generally take place outside of medical care settings [6-9]. But even there is quite good evidence for these interventions, the implementation of these approaches in a primary care setting seems to be accompanied with additional problems, while in these setting less positive results were revealed in former studies [10-12].
However, it still remains unclear what approach is the best to implement evidence based treatment approaches into daily practice [13]. Without a doubt different settings and cultures of implementing knowledge have to be considered. In Germany quality circles are a well established concept and several studies have proven their impact on different outcome parameters as for instance on prescriptions [14]. But previous studies revealed that the improvement in care is mostly moderate if no additional strategies are provided to improve the impact of the meetings for GPs [13]. Regarding the field of chronic care and especially degenerative joint diseases, the involvement of practice nurses, as for instance to perform frequent telephone calls, has shown to increase the patients' quality of life [15]. It could be assumed that the impact on patients may increase if the induced implementations on the patients' level are frequently monitored by practice nurses.
Methods
Aim of the study
The study examines whether a multifaceted intervention with evidence based medical education for GPs can improve the quality of life of arthritis patients.
Scientific hypothesis
A targeted evidence based medical education for GPs on osteoarthritis has no effect on the quality of life of patients with degenerative joint diseases and their prescribed medication. Monitoring GPs' prescriptions and advices for lifestyle changes by monthly telephone calls of practice assistants with a structured form is not superior.
The study content is guided by internationally available evidence for arthritis therapy in General Practice. Due to the lack of a German arthritis guideline an evidence based review for arthritis care in General Practice will be compiled from European guidelines [7,9,16]. Subsequently a preliminary guideline will be elaborated based on this material. Additionally motivational strategies and communication skills will be taught to GPs in order to improve the implementation of life style changes.
Study design
The study is a (prospective) cluster-randomized, open, three-armed intervention study. The design of a cluster randomized study was chosen because this has optimal internal validity (absence of confounders) while avoiding contamination of interventions associated with patient randomization.
Sample size
Sample size calculations for cluster randomized trials differ completely from sample size calculations for common RCTs [17-19]. Based on the main outcome parameter (QoL) and the main outcome-assessment instrument (GERMAN-AIMS2-SF) [20], we performed a power calculation with the Cluster Randomization Sample Size Calculator ver.1.02 of the University of Aberdeen. Assuming an effect size of 30 % (according to recommendations of Guillemin et al. [21]), an ICC of 0.03 (based on previous studies and on data available at the website of the University of Aberdeen [22]) a power of 90 % a mean of 2.7 and a minimal difference to detect of 0.9 and a significance level of 0.05, we have to include 14 patients in each of the 25 practices.
Recruitment of GPs and randomization
The GPs are the unit of randomization. They were eligible for randomization if their practice had a contract with all German insurances, so it is assured that patients of all social levels have unlimited admission to the practice. If they were working in a non-single-handed practice, it was important that they had their own patients which could clearly be allocated to them. In Germany most of GPs work in single handed practices, but even if not, patients frequently are treated by one specific GP in a practice, so that these inclusion criteria will not represent a source of bias, because only an absolute minority of practices will not be eligible for inclusion due to these criteria. About 500 GPs in the area of Baden-Wuerttemberg and Bavaria, fulfilling inclusion criteria, were invited by a formal letter of the Department of General Practice and Health services Research of the University of Heidelberg, to participate in the study. 120 GPs gave their written consent to participate in the study. Based on detailed information about the practice and the GP, the inclusion criteria were checked. No GP or no practice had to be excluded due to the inclusion criteria. The 120 GPs were invited to information meetings were the aim and the procedure of the study were explained in detail. After the meeting, the 120 eligible GPs were put on a list with numbers from 1 to 120. Out of this list, 75 GPs were randomized with SPSS version 11.0 to one of the intervention groups or the control group by an independent assistant who is not familiar to one of the participating doctors.
Patient inclusion criteria
To be eligible for inclusion patients have to be adult and diagnosed with gonarthritis or coxarthritis according to the ACR criteria [23]. They will be identified by the following ICD-10 code in patients file: M 16.0–16.9 and M 17.0–17.5. Based on this process, participating practices keep an alphabetic record of their patients. Patients from this list are contacted in consecutive order of appearance in the practice and informed about the option to participate in the study. After giving their written informed consent they receive the questionnaire and a stamped envelope with the postal address of the university. The patients are asked to return this questionnaire in the envelope to the university. Neither the GP nor the practice team has any possibility to get knowledge of the patients' answers.
Patient exclusion criteria
1. Insufficient German language skills.
2. Patients, who contacted the practice for emergencies only or as a substitute practice.
Data collection
After giving their informed and written consent to participate in the study patients will receive a questionnaire which is based on arthritis-related indicators and include the GERMAN-AIMS2-SF [20], a revised version of the EUROPEP-questionnaire[24], as well as items that assess secondary outcome parameters as shown in table 1. The envelopes are opened at the university by an independent research assistant and immediately scanned with the "eyes and hands ™ FORMS"-Software (Version 5) of Read Soft. A TIF-file is generated out of each questionnaire to avoid any data-manipulation and to ease data storage. The data are transferred into the SPSS program (version 12.0). Patients' information on medication and health care utilization will be checked by three research assistants, visiting each practice.
Table 1 Outcome-parameters and instruments of the study
Outcome-Parameter (Patient) Instrument
Primary Outcome
Quality of life GERMAN-AIMS2-SF
Secondary outcome
Health care utilization questonnaire, retrospective chart review
Patient satisfaction modified EUROPEP
Physical activity 6 minutes walking, CDC-criteria, specific questions
Medication questionnaire, retrospective chart review
Confounder control
Mental comorbidity PHQ-9
Outcome-Parameter
Table 1 displays the outcome parameters and associated used instruments. The primary outcome is quality of life assessed by the AIMS2-SF questionnaire, an internationally validated instrument for the assessment of quality of life among arthritis patients [21]. We have validated this instrument for German general practice in a previous study [20].
Secondary outcomes include
• Medication (evidence based use of NSAR, application of WHO-recommendations); data retrieved from patients chart
• Health Care utilization (referrals to orthopedists, imaging, inpatient care, physiotherapy); data retrieved from patients chart
• Physical activity (percentage of patients meeting CDC criteria)
• Patient satisfaction (modified EUROPEP-questionnaire) [25]
• Potential confounders are being detected (concurrent depression may influence the potential motivational change for more physical activity) by means of PHQ-9 [26].
These data will be compiled from patient questionnaires and patients chart review. All instruments represent well established and validated instruments. Measurements and analysis will take place before intervention (pre-data-collection) and 6 months later (post-data collection).
Intervention
1. Implementation strategy – aiming at the GP
The implementation strategy consists of two interactive quality circle meetings of 3 hours including 12–13 participating GPs (Intervention group I). These meetings have three main contents: evidenced based treatment of osteoarthritis in a primary care setting, optimizing pain treatment according to the WHO recommendations, providing advanced motivation skills. Intervention group II represents an "add-on" approach. GPs will participate in meetings with the same content as GPs of intervention group I. In addition to these meetings, practice assistants will also participate in a course. During this course the assistance is trained to call patients and provide an OA specific telephone questionnaire which addresses to three main topics: Side effects of the prescribed drugs, adherence to recommended physical activity and changes in pain.
In both intervention groups, GPs will receive a summary of evidence based treatments of OA in a primary care setting. These information contains the recommendations of the EULAR group for the treatment of OA and the information provided by the German Medical Association [7,9,16]. GPs will also receive two written patient leaflets. Leaflet one provides information about the cause and the treatment possibilities as well as coping strategies and contact addresses of self help groups for the patients. Leaflet two provides a detailed exercise program, developed by a German self help group, the "Deutsche Rheumaliga". This leaflet contains pictures and a step by step exercise program, even patients with severe OA can perform.
2. Clinical intervention
In intervention group II frequent telephone calls will be provided by each practice. For this purpose an osteoarthritis specific telephone questionnaire – the "ArthMol" tool – has been developed in cooperation with the Department of General Practice at Johann-Wolfgang Goethe-University Clinic, Frankfurt am Main. GPs' assistants will contact OA patients via telephone every four weeks and complete a structured telephone form during the conversation. The form contains items referring to pain, adherence to prescriptions, the exercise program and possible side effects of the medication. According to the urgency of the information it is either directly reported to the GP or transmitted during the following day.
There is no implementation strategy in the control group (group III).
Timeframe of the study
The study team has already randomized the 75 out of the initial recruited 120 GPs who have declared their willingness to participate in the study and to accepted random assignment to the different groups. The patient inclusion and pre data collection will take place in months I-III, the intervention (quality circles and telephone calls in group II) will take place in months IV-X. Post data collection will be performed in months X-XII.
Description of risks
Serious risks or undesired effects of questionnaires have not been described in the literature. There are no specific risks related to the study.
Ethical and legal aspects
Ethical principles
The study is being conducted in accordance with medical professional codex and the Helsinki Declaration as of 1996 as well as the German Federal Data Security Law (BDSG).
Study participation of patients is voluntary and can be cancelled at any time without provision of reasons and without negative consequences for their future medical care.
Patient informed consent
Previous to study participation patients receive written and spoken information about the content and extent of the planned study; for instance about potential benefits for their health and potential risks. In case of acceptance they sign the informed consent form.
In case of study discontinuation all material will be destroyed or the patient will be asked if he/she accepts that existing material can be analyzed in the study.
Legal principles
Vote of the ethics committee
The study protocol was approved by the ethics committee of the University of Heidelberg previous to the start of the study in January 2005. Inclusion of patients/ participants did not start unless there was a written and unrestricted positive vote of the ethics committee. This vote was received in March 2005 (approval number 021/2005).
Data security/ disclosure of original documents
The patient names and all other confidential information fall under medical confidentiality rules and are treated according to German Federal Data Security Law (BDSG). The results of the patient questionnaires are not accessible to the GPs. Questionnaires are directly mailed to the study center by the patient.
All study related data and documents are stored on a protected central server of the Heidelberg University Clinic. Only direct members of the internal study team can access the respective files.
Intermediate and final reports are stored in the office of the Department of General Practice and Health Services Research at the Heidelberg University Clinic.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
TR, TK and MW conceived and performed the study and draft the manuscript. JG and CM developed the "ArthMoL"-tool. SJ and JS participated in the study design. All authors read and approved the final manuscript.
Figure 1 Flow diagram of the progress through the phases of the study.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
This study is part of the PRAXART project that aims to improve the quality of life of patients suffering from OA. The project is financed by the German Ministry of Education and Research (BMBF), grant-number 01GK0301.
==== Refs
Woolf AD Pfleger B Burden of major musculoskeletal conditions Bull World Health Organ 2003 81 646 656 14710506
Felson DT Lawrence RC Dieppe PA Hirsch R Helmick CG Jordan JM Kington RS Lane NE Nevitt MC Zhang Y Sowers M McAlindon T Spector TD Poole AR Yanovski SZ Ateshian G Sharma L Buckwalter JA Brandt KD Fries JF Osteoarthritis: new insights. Part 1: the disease and its risk factors Ann Intern Med 2000 133 635 646 11033593
Neville C Fortin PR Fitzcharles MA Baron M Abrahamowitz M Du BR Esdaile JM The needs of patients with arthritis: the patient's perspective Arthritis Care Res 1999 12 85 95 10513496 10.1002/1529-0131(199904)12:2<85::AID-ART3>3.0.CO;2-W
Buckley LM Vacek P Cooper SM Educational and psychosocial needs of patients with chronic disease. A survey of preferences of patients with rheumatoid arthritis Arthritis Care Res 1990 3 5 10 2285739
de Bock GH Kaptein AA Mulder JD Dutch general practitioners' management of patients with distal osteoarthritic symptoms Scand J Prim Health Care 1992 10 42 46 1589663
Felson DT Lawrence RC Hochberg MC McAlindon T Dieppe PA Minor MA Blair SN Berman BM Fries JF Weinberger M Lorig KR Jacobs JJ Goldberg V Osteoarthritis: new insights. Part 2: treatment approaches Ann Intern Med 2000 133 726 737 11074906
Mazieres B Bannwarth B Dougados M Lequesne M EULAR recommendations for the management of knee osteoarthritis. Report of a task force of the Standing Committee for International Clinical Studies Including Therapeutic Trials Joint Bone Spine 2001 68 231 240 11394623 10.1016/S1297-319X(01)00271-8
Warsi A LaValley MP Wang PS Avorn J Solomon DH Arthritis self-management education programs: a meta-analysis of the effect on pain and disability Arthritis Rheum 2003 48 2207 2213 12905474 10.1002/art.11210
Zhang W Doherty M Arden N Bannwarth B Bijlsma J Gunther KP Hauselmann HJ Herrero-Beaumont G Jordan K Kaklamanis P Leeb B Lequesne M Lohmander S Mazieres B Martin-Mola E Pavelka K Pendleton A Punzi L Swoboda B Varatojo R Verbruggen G Zimmermann-Gorska I Dougados M EULAR evidence based recommendations for the management of hip osteoarthritis: report of a task force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT) Ann Rheum Dis 2005 64 669 681 15471891 10.1136/ard.2004.028886
Solomon DH Warsi A Brown-Stevenson T Farrell M Gauthier S Mikels D Lee TH Does self-management education benefit all populations with arthritis? A randomized controlled trial in a primary care physician network J Rheumatol 2002 29 362 368 11838857
Grol R Grimshaw J From best evidence to best practice: effective implementation of change in patients' care Lancet 2003 362 1225 1230 14568747 10.1016/S0140-6736(03)14546-1
Grol R Grimshaw J Evidence-based implementation of evidence-based medicine Jt Comm J Qual Improv 1999 25 503 513 10522231
Grimshaw JM Thomas RE MacLennan G Fraser C Ramsay CR Vale L Whitty P Eccles MP Matowe L Shirran L Wensing M Dijkstra R Donaldson C Effectiveness and efficiency of guideline dissemination and implementation strategies Health Technol Assess 2004 8 iii 72 14960256
Wensing M Broge B Kaufmann-Kolle P Andres E Szecsenyi J Quality circles to improve prescribing patterns in primary medical care: what is their actual impact? J Eval Clin Pract 2004 10 457 466 15304146 10.1111/j.1365-2753.2004.00517.x
Rene J Weinberger M Mazzuca SA Brandt KD Katz BP Reduction of joint pain in patients with knee osteoarthritis who have received monthly telephone calls from lay personnel and whose medical treatment regimens have remained stable Arthritis Rheum 1992 35 511 515 1575787
R. L B. MO R. L and B. MO Empfehlungen zur Therapie von degenerativen Gelenkerkrankungen Therapieempfehlungen Ärzteschaft SA 2001 Lasek R.; Mueller-Oerlinghausen B.
Campbell M Grimshaw J Steen N Sample size calculations for cluster randomised trials. Changing Professional Practice in Europe Group (EU BIOMED II Concerted Action) J Health Serv Res Policy 2000 5 12 16 10787581
Campbell MK Mollison J Grimshaw JM Cluster trials in implementation research: estimation of intracluster correlation coefficients and sample size Stat Med 2001 20 391 399 11180309 10.1002/1097-0258(20010215)20:3<391::AID-SIM800>3.0.CO;2-Z
Rosemann T Korner T Wensing M Schneider A Szecsenyi J Evaluation and cultural adaptation of a German version of the AIMS2-SF questionnaire (German AIMS2-SF) Rheumatology (Oxford) 2005
Guillemin F Coste J Pouchot J Ghezail M Bregeon C Sany J The AIMS2-SF: a short form of the Arthritis Impact Measurement Scales 2. French Quality of Life in Rheumatology Group Arthritis Rheum 1997 40 1267 1274 9214427
Aberdeen U Empirical etsimates of ICCs from changing professional practice studies http //www abdn ac uk/hsru/epp/cluster shtml 2005
Altman R Alarcon G Appelrouth D Bloch D Borenstein D Brandt K Brown C Cooke TD Daniel W Feldman D . The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hip Arthritis Rheum 1991 34 505 514 2025304
Grol R Wensing M Mainz J Jung HP Ferreira P Hearnshaw H Hjortdahl P Olesen F Reis S Ribacke M Szecsenyi J Patients in Europe evaluate general practice care: an international comparison Br J Gen Pract 2000 50 882 887 11141874
Grol R Wensing M Mainz J Ferreira P Hearnshaw H Hjortdahl P Olesen F Ribacke M Spenser T Szecsenyi J Patients' priorities with respect to general practice care: an international comparison. European Task Force on Patient Evaluations of General Practice (EUROPEP) Fam Pract 1999 16 4 11 10321388 10.1093/fampra/16.1.4
Lowe B Kroenke K Herzog W Grafe K Measuring depression outcome with a brief self-report instrument: sensitivity to change of the Patient Health Questionnaire (PHQ-9) J Affect Disord 2004 81 61 66 15183601 10.1016/S0165-0327(03)00198-8
|
16029495
|
PMC1192806
|
CC BY
|
2021-01-04 16:28:58
|
no
|
BMC Public Health. 2005 Jul 19; 5:77
|
utf-8
|
BMC Public Health
| 2,005 |
10.1186/1471-2458-5-77
|
oa_comm
|
==== Front
BMC PharmacolBMC Pharmacology1471-2210BioMed Central London 1471-2210-5-131610916810.1186/1471-2210-5-13Research ArticleIn vivo antimuscarinic actions of the third generation antihistaminergic agent, desloratadine Howell G [email protected] L [email protected] C [email protected] B [email protected] D [email protected] R [email protected] Department of Pharmacology and Toxicology, University of Mississippi Medical Center, Jackson, MS 39216, USA2 Tougaloo College, Tougaloo, MS, USA2005 18 8 2005 5 13 13 6 10 2004 18 8 2005 Copyright © 2005 Howell et al; licensee BioMed Central Ltd.2005Howell 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
Muscarinic receptor mediated adverse effects, such as sedation and xerostomia, significantly hinder the therapeutic usefulness of first generation antihistamines. Therefore, second and third generation antihistamines which effectively antagonize the H1 receptor without significant affinity for muscarinic receptors have been developed. However, both in vitro and in vivo experimentation indicates that the third generation antihistamine, desloratadine, antagonizes muscarinic receptors. To fully examine the in vivo antimuscarinic efficacy of desloratadine, two murine and two rat models were utilized. The murine models sought to determine the efficacy of desloratadine to antagonize muscarinic agonist induced salivation, lacrimation, and tremor. Desloratadine's effect on the cardiovascular system was explored in both rodent models.
Results
In the pithed rat, both desloratadine (1.0 mg/kg, i.v.) and the muscarinic M2 selective antagonist, methoctramine (0.5 mg/kg, i.v.), inhibited negative inotropic (left ventricular dP/dt) effects caused by oxotremorine, a nonselective muscarinic agonist (p < 0.05). Negative chronotropic effects caused by oxotremorine were inhibited by desloratadine, methoctramine, and the muscarinic M3 selective antagonist, 4-DAMP (1.0 mg/kg, i.v.). A late positive inotropic event observed after the initial decrease was inhibited by all three test compounds with desloratadine and 4-DAMP being the most efficacious. In the conscious animal, inhibition of baroreflex-mediated bradycardia was evaluated. Unlike atropine (0.5 mg/kg, i.v.), desloratadine did not alter this bradycardia. The antimuscarinic action of desloratadine on salivation, lacrimation, and tremor was also explored. In urethane-anesthetized (1.5 g/kg, i.p.) male ICR mice (25–35 g) desloratadine (1.0, 5.0 mg/kg) did not inhibit oxotremorine-induced (0.5 mg/kg, s.c.) salivation, unlike atropine (0.5 mg/kg) and 4-DAMP (1.0 mg/kg). In conscious mice, desloratadine failed to inhibit oxotremorine-induced (0.5 mg/kg, s.c.) salivation, lacrimation, and tremor. However, desloratadine did inhibit oxotremorine-induced tremor in phenylephrine pretreated animals.
Conclusion
The presented data demonstrate that the third generation antihistamine, desloratadine, does not significantly antagonize peripheral muscarinic receptors mediating salivation and lacrimation, therefore, xerostomia and dry eyes should not be observed with therapeutic use of desloratadine. Our data also indicate when administered to a patient with a compromised blood-brain barrier, desloratadine may cause sedation. Patients with compromised cardiovascular systems should be closely monitored when administered desloratadine based on our results that desloratadine has the ability to interfere with normal cardiovascular function mediated by muscarinic receptors.
==== Body
Background
Antihistaminergic drugs are commonly classified into three generations. First generation antihistamines, such as diphenhydramine, effectively block the H1 receptor subtype but their use is limited due to significant central (sedation) and peripheral (tachycardia, xerostomia) antimuscarinic side effects. Second generation antihistamines, such as loratadine, retain a high selectivity for the H1 receptor and have fewer centrally mediated side effects than the first generation compounds because second generation compounds do not readily enter the central nervous system (CNS) [1]. However, two second generation antihistamines, astemizole and terfenadine, cause prolongation of the QT interval resulting in torsades de pointes. This adverse effect prompted the removal of terfenadine from the drug market [2]. The most recent, third generation compounds, include fexofenadine and desloratadine. These compounds are active metabolites of the second generation antihistamines, terfenadine and loratadine, respectively, and generally retain or surpass the H1 receptor selectivity of their parent compounds. For instance, desloratadine displays a higher affinity for the H1 receptor than does loratadine and antagonizes the human H1 receptor in a pseudoirreversible manner [3,4].
Questions remain concerning the potential for antimuscarinic adverse effects with desloratadine since both in vitro and in vivo experimentation indicates that desloratadine has the ability to block muscarinic receptors. Desloratadine demonstrated in vitro IC50 values of 48 nM and 125 nM against cloned human M1 and M3 muscarinic receptor subtypes, respectively [4]. In vivo muscarinic receptor blockade has been demonstrated in that desloratadine has been shown to inhibit pilocarpine induced salivation in mice and inhibit contractions of isolated rabbit and guinea pig iris smooth muscle [5,6]. Therefore, these data present the need to more definitively ascertain the potential antimuscarinic activity of desloratadine, in vivo. In the present study, several in vivo models were used to further assess antimuscarinic activity of desloratadine as well as the potential for penetration of the blood-brain barrier.
Results
Oxotremorine-induced tremor
Intraperitoneal injection of oxotremorine (0.5 mg/kg) induced tremor in conscious mice. The only dose of desloratadine causing inhibition of oxotremorine-induced tremor was 5.0 mg/kg (Figure 1). Desloratadine (1.0, 0.1, and 0.01 mg/kg) did not significantly inhibit generation of tremor. Unlike atropine sulfate (0.5 mg/kg), atropine methyl nitrate (0.5 mg/kg) did not inhibit tremors which confirms the central locus for oxotremorine-induced tremors. Diphenhydramine (1.0 mg/kg) significantly inhibited the generation of tremor by oxotremorine as did administration of both 4-DAMP (1.0 mg/kg) and methoctramine (0.5 mg/kg) prior to administration of oxotremorine.
Figure 1 Inhibition of oxotremorine-induced tremors. Mice were treated with a single i.p. injection of one of the test agents (atropine sulfate, AT; atropine methyl nitrate, AMN; diphenhydramine, DPH; methoctramine, MOT; 1,1-dimethyl-4-diphenylacetoxypiperidinium iodide, 4-DAMP; desloratadine, DL) and placed in individual shoebox cages for observation. Fifteen minutes later, each mouse received a single s.c. injection of oxotremorine sesquifumarate (0.5 mg/kg) at the nape of the neck. At 5, 10 and 15 minutes following the oxotremorine injection, mice were observed for severity of tremor and for the presence of salivation and lacrimation. The sum of the scores for the three time points for tremor is presented as Total Tremor Score. Numbers in parenthesis represent the number of animals in each group and asterisk denotes statistical significance (P < 0.05) vs. control.
Oxotremorine-induced tremor with phenylephrine pretreatment
Pretreatment with the vasopressor agent, phenylephrine (10 μg/kg; PE), functions to disrupt the blood-brain barrier by inducing acute hypertension [7,8]. Blood-brain barrier disruption resulted in significant inhibition of oxotremorine-induced tremors by desloratadine (1.0 mg/kg) compared to the 1% DMSO vehicle as reflected by their respective total tremor scores (Figure 2). Pretreatment with PE followed by administration of desloratadine (1.0 mg/kg; i.p.) elicited an oxotremorine-induced total tremor score of 2.0 ± 0.7 (mean ± SD) whereas PE pretreatment followed by vehicle elicited an oxotremorine-induced total tremor score of 5.2 ± 1.3 (mean ± SD).
Figure 2 Inhibition of oxotremorine-induced tremors with phenylephrine pretreatment. Male ICR mice (30–35 g) were treated with a single i.p. injection of phenylephrine hydrochloride (10 mg/kg; 10 μL/g). Each mouse was then placed in an individual shoebox cage for observation. Five minutes after this, each mouse was given a second i.p. injection of either vehicle (control; 10 μL/g) or desloratadine (free base) at a dose of 1.0 mg/kg (DL). Fifteen minutes later, each mouse received a single s.c. injection of oxotremorine sesquifumarate (0.5 mg/kg) at the nape of the neck. At 5, 10 and 15 minutes following the oxotremorine injection, mice were observed for severity of tremor and for the presence of salivation and lacrimation. The sum of the scores for the three time points is presented as Total Tremor Score. Numbers in parenthesis represent the number of animals in each group and asterisk denotes statistical significance (P < 0.05) vs. control.
Oxotremorine-induced salivation and lacrimation
Administration of oxotremorine (0.5 mg/kg, i.p.) elicited salivation in conscious and urethane-anesthetized mice. In conscious mice, oxotremorine-induced salivation was not significantly inhibited by pretreatment with desloratadine (0.01, 0.1, 1.0, and 5.0 mg/kg) (Figure 3). However, salivation was significantly inhibited by pretreatment with atropine (0.5 mg/kg), atropine methyl nitrate (0.5 mg/kg), and 4-DAMP (1.0 mg/kg). Pretreatment with methoctramine (0.5 mg/kg) and diphenhydramine (1.0 mg/kg) failed to inhibit oxotremorine-induced salivation.
Figure 3 Inhibition of lacrimation during oxotremorine-induced tremors. Mice were treated with a single i.p. injection of one of the test agents (atropine sulfate, AT; atropine methyl nitrate, AMN; diphenhydramine, DPH; methoctramine, MOT; 1,1-dimethyl-4-diphenylacetoxypiperidinium iodide, 4-DAMP; desloratadine, DL) and placed in individual shoebox cages for observation. Fifteen minutes later, each mouse received a single s.c. injection of oxotremorine sesquifumarate (0.5 mg/kg) at the nape of the neck. At 5, 10 and 15 minutes following the oxotremorine injection, mice were observed for severity of tremor and for the presence of salivation and lacrimation. The sum of the scores for the three time points for lacrimation is presented as Total Lacrimation Score. Numbers in parenthesis represent the number of animals in each group and asterisk denotes statistical significance (P < 0.05) vs. control.
Inhibition of oxotremorine-induced salivation in urethane-anesthetized mice yielded results similar to those obtained in conscious male ICR mice. Pretreatment with desloratadine (1.0 mg/kg) failed to significantly inhibit oxotremorine-induced salivation (Figure 4). As in the conscious animal, pretreatment with atropine (0.5 mg/kg) and 4-DAMP (1.0 mg/kg) significantly inhibited oxotremorine-induced salivation. Also, administration of diphenhydramine (1.0 mg/kg) and methoctramine (0.5 mg/kg) failed to significantly inhibit saliva production.
Figure 4 Inhibition of salivation during oxotremorine-induced tremors. Mice were treated with a single i.p. injection of one of the test agents (atropine sulfate, AT; atropine methyl nitrate, AMN; diphenhydramine, DPH; methoctramine, MOT; 1, 1-dimethyl-4-diphenylacetoxypiperidinium iodide, 4-DAMP; desloratadine, DL) and placed in individual shoebox cages for observation. Fifteen minutes later, each mouse received a single s.c. injection of oxotremorine sesquifumarate (0.5 mg/kg) at the nape of the neck. At 5, 10 and 15 minutes following the oxotremorine injection, mice were observed for severity of tremor and for the presence of salivation and lacrimation. The sum of the scores for the three time points for salivation is presented as Total Salivation Score. Numbers in parenthesis represent the number of animals in each group and asterisk denotes statistical significance (P < 0.05) vs. control.
Desloratadine (0.01, 0.1, 1.0, and 5.0 mg/kg) pretreatment had no significant effect on oxotremorine-induced lacrimation in conscious mice (Figure 5). As with inhibition of salivation in the conscious animal, pretreatment with atropine (0.5 mg/kg) and 4-DAMP (1.0 mg/kg) significantly inhibited oxotremorine-induced lacrimation. Pretreatment with either methoctramine (0.5 mg/kg) or diphenhydramine (1.0 mg/kg) did not significantly inhibit oxotremorine-induced lacrimation.
Figure 5 Inhibition of oxotremorine-induced salivation. Mice were anesthetized with urethane (1.5 g/kg, i.p.; 1 g/ml). Each mouse was then treated with a single i.p. injection of one of the test agents (atropine sulfate, AT; diphenhydramine, DPH; methoctramine, MOT; 1,1-dimethyl-4-diphenylacetoxypiperidinium iodide, 4-DAMP; desloratadine, DL). Each mouse was then placed prone and head-down on a plexiglass plate inclined at 10° and covered with a sheet of Whatman no. 3 MM filter paper. Fifteen minutes after administration of the test agent, a 0.5 mg/kg dose of oxotremorine was administered in a volume of 1 μl/g. Every five minutes for 30 minutes, each mouse was moved up the incline. Salivation production was measured immediately following the move at the end of each collection period by measurement of the circumference of the moist area of filter paper immediately beneath each mouse's mouth. The sum of the 6 collection periods is listed as Total Salivation Score. Numbers in parenthesis represent the number of animals in each group and asterisk denotes statistical significance (P < 0.05) vs. control.
Oxotremorine-induced changes in left ventricular contractility
Intravenous injections of oxotremorine (0.00125-0.02 mg/kg) elicited biphasic inotropic responses. The initial phase consisted of a dose-dependent decrease in dP/dt. This decrease in inotropy began approximately 30–60 seconds after the beginning of oxotremorine injection and lasted for approximately 60 seconds. Both desloratadine and methoctramine treatments effectively blocked the negative inotropic effect (Figure 6). Administration of desloratadine (1.0 mg/kg) significantly inhibited oxotremorine-induced (0.00125, 0.0025, and 0.02 mg/kg) decreases in dP/dt indicated by a shift in the dose-response curve to the right. Also, administration of methoctramine (0.5 mg/kg) significantly inhibited oxotremorine-induced (0.0025, 0.01, 0.02 mg/kg) decreases in dP/dt indicated by a shift in the dose-response curve to the right. After administration of test agents, an additional dose of oxotremorine (0.04 mg/kg) was administered causing percentage decreases in left ventricular (LV) contractility of -14.2 ± 2.7, -38.3 ± 7.6, and -20.7 ± 1.5 for desloratadine, 4-DAMP, and methoctramine treatments, respectively (data not shown). Treatment with 4-DAMP had little antagonistic effect on the negative inotropic response to oxotremorine.
Figure 6 Effect of desloratadine (DL), 4-DAMP, or methoctramine on oxotremorine-induced decrease in left ventricular contractility in the pithed rat. Isoflurane-anesthetized animals (n = 6) were pithed after insertion of femoral and carotid arterial catheters. The carotid catheter was advanced into the LV to enable recording of contractility, which was expressed as the change in pressure over the change in time (dP/dt). Anesthesia was then discontinued. Following administration of atenolol (1.0 mg/kg, i.v.), oxotremorine was administered in random (n = 3) or ascending (n = 3) order of doses. The third generation antihistamine, DL (1.0 mg/kg, i.v.), the muscarinic M3 receptor antagonist, 4-DAMP (1.0 mg/kg, i.v.), or the muscarinic M2 receptor antagonist, methoctramine (0.5 mg/kg, i.v.), was then administered and the oxotremorine doses were repeated. No statistically significant differences were found between animals in which oxotremorine was given in random vs. ascending order of doses and both sets of data were pooled. Data are representative of the maximal percent fall in dP/dt compared to control following the administration of each dose of oxotremorine. The highest dose tested for oxotremorine (0.04 mg/kg, i.v.) could not be given prior to treatment with 4-DAMP. Statistical analysis was done using the paired t-test with P < 0.05 denoting a statistically significant difference versus control as indicated by an asterisk.
The second phase of the inotropic response to oxotremorine consisted of a dose-dependent increase in dP/dt. This increase immediately followed the initial decrease and had a duration of 2–5 minutes. Both desloratadine and 4-DAMP antagonized the oxotremorine-induced positive inotropic effect (Figure 7). Administration of desloratadine (1.0 mg/kg) significantly inhibited oxotremorine-induced (0.01 and 0.005 mg/kg) increases in dP/dt indicated by a shift in the dose-response curve to the right. Also, administration of 4-DAMP (1.0 mg/kg) significantly inhibited oxotremorine-induced (0.005, 0.01, and 0.02 mg/kg) increases in dP/dt indicated by a shift in the dose-response curve to the right. In contrast to desloratadine and 4-DAMP, methoctramine treatment (0.5 mg/kg) resulted in a statistically significant (P < 0.05) increases in dP/dt after oxotremorine (0.01 and 0.02 mg/kg) administration compared to control values recorded prior to methoctramine treatment. After administration of test agents, an additional dose of oxotremorine (0.04 mg/kg) was administered causing percentage increases in LV contractility of 32.2 ± 7.1, 20.5 ± 5.4, and 52.9 ± 9.5 for desloratadine, 4-DAMP, and methoctramine treatments, respectively (data not shown).
Figure 7 Effect of desloratadine (DL), 4-DAMP, or methoctramine on oxotremorine-induced increase of left ventricular contractility in the pithed rat. Isoflurane-anesthetized animals (n = 6) were pithed after insertion of femoral and carotid arterial catheters. The carotid catheter was advanced into the LV to enable recording of contractility, which was expressed as the change in pressure over the change in time (dP/dt). Anesthesia was then discontinued. Following administration of atenolol (1.0 mg/kg, i.v.), oxotremorine was administered in random (n = 3) and ascending (n = 3) order of doses. The third generation antihistamine, DL (1.0 mg/kg, i.v.), the muscarinic M3 receptor antagonist, 4-DAMP (1.0 mg/kg, i.v.), or the muscarinic M2 receptor antagonist, methoctramine (0.5 mg/kg, i.v.) was then administered and the oxotremorine doses were repeated. No statistically significant differences were found between animals in which oxotremorine was given in random vs. ascending order of doses and both sets of data were pooled. Data are representative of the maximal percent increase in left ventricular contractility compared to control following the administration of each dose of oxotremorine. The highest dose tested for oxotremorine (0.04 mg/kg, i.v.) could not be given prior to treatment with 4-DAMP. Statistical analysis was done using the paired t-test with P < 0.05 denoting a statistically significant difference versus control as indicated by an asterisk.
Oxotremorine-induced bradycardia
Administration of oxotremorine caused a dose-dependent decrease in heart rate. All three of the test agents antagonized this decrease as indicated by a shift in the dose-response curve to the right (Figure 8). After administration of all three test agents, the negative chronotropic effects of oxotremorine (0.005, 0.01, and 0.02 mg/kg) were significantly inhibited. Treatment with desloratadine (1.0 mg/kg) or 4-DAMP (1.0 mg/kg) also significantly inhibited the negative chronotropic response to oxotremorine (0.0025 mg/kg) while methoctramine (0.5 mg/kg) treatment inhibited the response to oxotremorine (0.00125 mg/kg). After administration of test agents, an additional dose of oxotremorine (0.04 mg/kg) was administered causing percentage decreases in heart rate of -18.8 ± 3.5, -32.1 ± 7, and -19.7 ± 4.2 for desloratadine, 4-DAMP, and methoctramine treatments, respectively (data not shown).
Figure 8 Effect of desloratadine (DL), 4-DAMP, or methoctramine on oxotremorine-induced decrease of heart rate (HR) in the pithed rat. Isoflurane-anesthetized animals (n = 6) were pithed after insertion of femoral and carotid arterial catheters. The femoral catheter was inserted approximately four centimeters into the femoral artery to enable recording of heart rate. Anesthesia was then discontinued. Following administration of atenolol (1.0 mg/kg, i.v.), oxotremorine was administered in random (n = 3) and ascending (n = 3) order of doses. The third generation antihistamine, DL (1.0 mg/kg, i.v.), the muscarinic M3 receptor antagonist, 4-DAMP (1.0 mg/kg, i.v.), or the muscarinic M2 receptor antagonist, methoctramine (0.5 mg/kg, i.v.), was then administered and the oxotremorine doses were repeated. No statistically significant differences were found between animals in which oxotremorine was given in random vs. ascending order of doses and both sets of data were pooled. Data are representative of the maximal percent fall in heart rate compared to control following the administration of each dose of oxotremorine. The highest dose tested for oxotremorine (0.04 mg/kg, i.v.) could not be given prior to treatment with 4-DAMP. Statistical analysis was done using the paired t-test with P < 0.05 denoting a statistically significant difference versus control as indicated by an asterisk.
Inhibition of baroreceptor reflex
The ability of desloratadine to significantly alter the baroreceptor reflex was assessed in the conscious rat. Data were expressed as the percent change from corresponding control values of blood pressure and heart rate and subsequently analyzed by linear regression. The mean slope values were then analyzed for significant differences (data not shown). Administration of desloratadine (1.0 mg/kg) prior to stimulation of the baroreceptor reflex resulted in a slope value of -0.708 ± 0.03 (mean ± SE; n = 6) with the corresponding control slope value of -0.795 ± 0.03 (n = 6) which was not a statistically significant difference. Unlike desloratadine, administration of atropine (0.5 mg/kg) prior to stimulation of the baroreceptor reflex resulted in a slope value of -0.548 ± 0.03 (n = 6) with the corresponding control slope value of -0.670 ± 0.02 (n = 6) which was a statistically significant difference (P < 0.05).
Discussion
The focus of the present experiments was to determine the degree of antimuscarinic effects exerted by desloratadine at M2 and M3 receptors, in vivo. The non-selective muscarinic receptor agonist, oxotremorine [9], was employed as the challenge agent in the murine and rat models. Relatively selective antagonists at the M2 and M3 receptors, methoctramine [10,11] and 4-DAMP [12,13], respectively, were used for comparison. Our results indicate the third generation antihistaminergic agent, desloratadine, possesses a significant degree of antimuscarinic activity, primarily against cardiac M2 and M3 receptor subtypes, using in vivo whole animal preparations. However, the doses at which these activities are demonstrated exceed those normally utilized for therapeutic antihistaminergic effects. In addition, while penetration of the blood-brain barrier by desloratadine is unlikely to occur at therapeutic doses [14], evidence has been obtained suggesting penetration can be achieved and result in significant central antimuscarinic effects if the blood-brain barrier is compromised by administration of a vasopressor agent.
Oxotremorine-induced tremor, salivation, and lacrimation in the mouse have been used by others to evaluate the presence of antimuscarinic actions of drugs of interest [13,15,16]. The elicitation of tremor by oxotremorine is centrally mediated [17,18] and blockade of this response gauges penetration of an antimuscarinic agent across the blood-brain barrier. Thus, blockade of oxotremorine-induced tremor is indirectly indicative of the potential for an antimuscarinic agent to exert central actions, such as sedation, following peripheral administration. In the presence of an intact blood-brain barrier, desloratadine did not exert significant blockade of oxotremorine-induced tremor, except at the highest dose tested (5.0 mg/kg) which caused roughly 30% reduction in tremor severity. In contrast, following treatment with the vasopressor agent, phenylephrine, to open the blood-brain barrier, a previously ineffective dose of desloratadine (1.0 mg/kg) caused a 60% reduction in tremor severity. These data suggest that while desloratadine is unlikely to exert central antimuscarinic effects at therapeutic dosages (5.0 mg recommended dose) in normal adults, considerably greater CNS penetration may occur when the blood-brain barrier is compromised. The significance of this when desloratadine is combined with a vasopressor decongestant or when infection may compromise the blood-brain barrier [19,20] remains for further study. The present results showing blockade by pretreatment with either methoctramine or 4-DAMP, indicate that oxotremorine-induced tremor is mediated by both M2 and M3 receptors in the mouse as has been previously demonstrated by others [13,21].
Both oxotremorine-induced lacrimation [22] and salivation [23] have been shown to be mediated selectively through the M3 receptor subtype, a mediation confirmed by the present study. Thus, while methoctramine pretreatment had no effect on either variable, 4-DAMP pretreatment was capable of reducing both lacrimation and salivation by 60–80% below control responses. In direct contrast, desloratadine inhibited neither lacrimation nor salivation at doses as high as 5 mg/kg.
The pithed, atenolol-treated rat provides a useful acute model with which to examine antimuscarinic drug action on the circulatory system in the absence of both basal and phasic sympathetic nervous system influences. The administration of oxotremorine, in this model elicits dose-dependent bradycardia, and biphasic effects on cardiac inotropy. Oxotremorine causes an initial decline in contractility, as determined by ventricular dP/dt, followed by a more prolonged positive inotropic phase. This biphasic inotropic response to a muscarinic agonist has been previously reported for acetylcholine, bethanechol, and carbachol in a variety of experimental species [24-27].
The rat heart contains multiple muscarinic receptors, including the M1 [28], M2 [21], and M3 [29] subtypes. Of these, the M2 subtype predominates based on reverse-transcriptase polymerase chain reaction (rt-PCR) data indicating the M2 subtype constitutes more than 90% of the total muscarinic receptor mRNA, therefore, supporting its role as the major mediator of muscarinic influence over the functional state of the myocardium [30]. However, Krejci and Tucek also demonstrated the presence of mRNA for M1 and M3 subtypes, each constituting less than 1% and 3%, respectively, of the total muscarinic receptor mRNA in the rat heart [30]. M2 receptor agonists elicit bradycardia and a negative inotropic response through inhibition of cardiac adenylyl cyclase and/or an increase in potassium conductance via the muscarinic potassium channel [31,32]. In contrast, effects mediated through the M1 and/or M3 receptors may lead to increased contractile strength, through enhanced activity of phospholipase C and subsequent downstream events leading to increased intracellular free calcium availability [29,33]. Wang et al. [34] have recently reviewed the existence of multiple muscarinic receptors in the mammalian myocardium and have emphasized the presence of and physiological functions exerted by M3 receptors. A lesser body of data supports functional actions of the M1 subtype.
Desloratadine, at a dose of 1.0 mg/kg, effectively antagonized bradycardia and both negative and positive inotropic responses elicited by oxotremorine. Assuming adequate selectivity between cardiac muscarinic receptor subtypes, our data suggest the ability of methoctramine to blunt oxotremorine-induced negative inotropic event and the ability of 4-DAMP to blunt oxotremorine-induced positive inotropic event to be indicative of M2 and M3 receptor mediation of these phenomena, respectively. In contrast, however, both methoctramine (0.5 mg/kg) and 4-DAMP (1.0 mg/kg) blunted oxotremorine-induced bradycardia. Therefore, the possibility exists that oxotremorine-induced bradycardia is mediated by both M2 and M3 receptor subtypes.
The context in which the present results are taken is worthy of discussion. Both in vitro receptor binding data [35-37] and results from prior in vivo studies [5,6,37] demonstrate a considerably greater affinity of desloratadine for histaminergic than muscarinic receptors (for reviews see, [38,39]). Desloratadine has been found to exhibit a peak plasma concentration of approximately 28 ng/ml in healthy volunteers following a therapeutic antihistaminic dose of its parent compound, loratadine [35]. Single oral doses of desloratadine of 5, 7.5, 10, and 20 mg yielded peak plasma concentrations of 2.18, 3.03, 3.80, and 8.08 ng/L in human volunteers [36]. In mice, desloratadine exhibits an ED50 of 0.15 mg/kg in reduction of histamine induced paw edema [37]. Cardelus et al. [5] noted local antimuscarinic effects following topical ocular administration of 1–10 mg/ml of desloratadine. However, it is unlikely that systemic concentrations of desloratadine would rise to levels approaching those in the present study following normal therapeutic dosages of desloratadine, a fact which has been emphasized by others [37]. Thus, the antimuscarinic actions of desloratadine demonstrated in the present study would, most probably, be of significance only in overdose situations.
Conclusion
Our findings indicate that, at doses greater than those recommended for antihistaminergic therapy, desloratadine causes significant blockade of cardiac M2 and possibly cardiac M3 receptors, in vivo. This was demonstrated by significant inhibition of oxotremorine-mediated positive and negative inotropic events and bradycardia by desloratadine in the pithed rat. In contrast, desloratadine does not significantly antagonize the M3 receptor subtype responsible for salivation and lacrimation as demonstrated by the compound's inability to inhibit oxotremorine-mediated salivation and lacrimation in the conscious mouse and lacrimation in the anesthetized mouse. Also, under normal physiological conditions, desloratadine does not effectively cross the blood-brain barrier. However, upon disruption of this barrier, desloratadine has the potential for CNS penetration and muscarinic receptor blockade.
Methods
Drugs and solutions
Test agents included atropine sulfate, atropine methyl nitrate, diphenhydramine hydrochloride, methoctramine hydrochloride, 1,1-dimethyl-4-diphenylacetoxypiperidinium iodide (4-DAMP), and desloratadine. All were reconstituted in 1% DMSO / PBS, aliquoted into separate vials, and stored at -20°C until used. With the exception of desloratadine and 4-DAMP, all test agent concentrations were calculated using the salt weights. Atropine sulfate, atropine methyl nitrate, diphenhydramine hydrochloride, DMSO, oxotremorine sesquifumarate, atenolol, halothane, and urethane were all purchased from Sigma Chemical Co. (St. Louis, MO, USA). Other purchased agents were isoflurane (Abbott Laboratories; North Chicago, IL, USA), methoctramine hydrochloride (ICN Biochemicals, Inc.; Aurora, OH, USA), and 4-DAMP (Tocris; Ellisville, MO, USA). Desloratadine was provided by Aventis Pharmaceuticals (Bridgewater, NJ, USA).
Animal experiments
Male Sprague Dawley rats (275–325 g) and male ICR mice (25–35 g) were purchased from Harlan Sprague Dawley and housed in plastic group shoebox cages in an AAALAC-approved Laboratory Animal Facility. Animals were housed under a twelve hour light-dark cycle with food and water ad libitum. Food was withheld twelve hours prior to experimentation or surgical procedures. All animal use protocols were approved by the University of Mississippi Medical Center Institutional Animal Care and Use Committee.
Inhibition of oxotremorine-induced tremor, salivation, and lacrimation
A murine model was used to test the ability of desloratadine to antagonize muscarinic actions induced by administration of the muscarinic agonist, oxotremorine [15]. On the afternoon of an experiment, each mouse was weighed and placed in a clear shoebox cage for observation 15 minutes prior to any drug administration. Test agents for this experiment were vehicle, atropine sulfate (0.5 mg/kg), atropine methyl nitrate (0.5 mg/kg), diphenhydramine hydrochloride (1.0 mg/kg), methoctramine hydrochloride (0.5 mg/kg), 4-DAMP (1.0 mg/kg), and desloratadine (5.0, 1.0, 0.1, and 0.01 mg/kg). Each mouse was given a single intraperitoneal (i.p.) injection in a volume of 1 μl/g of one of the test agents. Fifteen minutes later, each mouse received a single subcutaneous (s.c.) injection of oxotremorine sesquifumarate (0.5 mg/kg), a non-selective muscarinic agonist, at the nape of the neck. At 5, 10, and 15 minutes following oxotremorine injection, each mouse was assessed for the degree of tremor and for the presence or absence of salivation and lacrimation. A modified five-point grading scale was used to evaluate tremor: 0 = no observable tremor; 0.5 = limb tremor observable when mouse is held by the tail with all feet off the cage bottom for 15 seconds; 1 = intermittent tremor, with bouts lasting from 3–5 seconds; 2 = intermittent tremor, with bouts lasting more than 5 seconds; or continuous, fine tremor noticeable on tail and ears; 3 = severe, continuous, whole-body tremor. Salivation and lacrimation were separately evaluated on a two-point scale: 0 = no observable salivation/lacrimation; 1 = salivation/lacrimation present. All responses were assessed by each of two observers with no knowledge of the pretreatment given each mouse. The grade for each mouse reflects the sum of the three consecutive observations as either Total Tremor, Total Salivation, or Total Lacrimation Score.
Inhibition of oxotremorine-induced salivation
A second paradigm, using mice (25–35 g), anesthetized with ethyl carbamate (urethane, 1.5 g/kg, i.p., 1 g/ml) was used to independently evaluate putative M3 receptor blockade inhibition of oxotremorine-induced salivation. This was modified after similar methods described for use in the rat by Lavy and Mulder [16]. All mice were denied access to food, but not to water, for 16 hours prior to anesthetization. Test agents for this experiment were vehicle, atropine sulfate (0.5 mg/kg), diphenhydramine hydrochloride (1.0 mg/kg), methoctramine hydrochloride (0.5 mg/kg), 4-DAMP (1.0 mg/kg; free base) and desloratadine (1.0 mg/kg; free base). A single i.p. injection of one of the test agents or vehicle was administered in a volume of 1.0 μl/g five minutes following injection of urethane. Each mouse was then placed prone and head-down on a plexiglass plate inclined at 10° and covered with a sheet of Whatman no. 3 MM filter paper. Fifteen minutes after test agent administration, a 0.5 mg/kg (i.p.) dose of oxotremorine was given in a volume of 1.0 μl/g body weight. Each mouse was moved up the incline every five minutes for thirty minutes. Saliva production was quantitated at the end of each five-minute collection period by measurement of the circumference of the moist area of filter paper immediately beneath each mouse's mouth. The sum of values from the six collection periods was recorded as Total Salivation Score (TSS).
Inhibition of oxotremorine-induced tremor with phenylephrine pretreatment
Mice (30–35 g) were treated with a single i.p. injection of phenylephrine hydrochloride (10 μg/kg; 10 μL/g) to elevate systemic blood pressure and open the blood-brain barrier [7,8]. Each mouse was then placed in an individual shoebox cage for observation. Five minutes after this, each mouse was given a second i.p. injection of either vehicle or desloratadine (free base) at a dose of 1.0 mg/kg. Fifteen minutes later, each mouse received a single s.c. injection of oxotremorine (0.5 mg/kg) at the nape of the neck. At 5, 10, and 15 minutes post oxotremorine injection, mice were observed for severity of tremor. The sum of the scores for the three time points is presented as Total Tremor Score.
Inhibition of oxotremorine-induced changes in cardiac function
The influence of oxotremorine over cardiac function in a pithed rat model was employed to evaluate muscarinic receptor antagonistic properties of desloratadine [15]. Male rats (275–325 g) were acutely anesthetized with 2–4% isoflurane in medical grade oxygen. Polyethylene arterial (PE-50) and venous (PE-10) catheters and a tracheal cannula (PE-240) were surgically implanted to permit monitoring of arterial blood pressure and chronotropy, i.v. drug administration, and maintenance of respiration by means of a Harvard rodent respirator, respectively. A catheter (PE-50) was passed via the right carotid artery into the left cardiac ventricle for measurement of left ventricular dP/dt as an index of inotropy. Responses were obtained using either a Grass Model 7P20G differentiator and recorded on a Grass Model 7D polygraph (Grass Instrument Co.; Quincy, MASS, USA) or with PowerLab/16 SP data acquisition system using Chart for Windows v4.0 recording software (ADInstruments; Colorado Springs, CO, USA). Each rat was then pithed by insertion of a blunt stainless steel rod, 2 millimeters in diameter, through the orbit of the eye and passed through the brain and spinal column, thus destroying the central nervous system (CNS) from forebrain to the terminus of the spinal cord. Atenolol (10 mg/kg, 1.0 ml/kg, i.v.) was administered to obviate peripheral catecholamine-induced increases in cardiac function. After a 15 minute stabilization period, doses of oxotremorine (0.00125, 0.0025, 0.005, 0.01, 0.02, 0.1 mg/kg), were administered randomly and flushed with 0.1 ml of heparinized 0.9% saline. Subsequently, a single i.v. injection of one of three test agents was administered over a two minute period. The test agents were desloratadine (1.0 mg/kg, 1.0 ml/kg, i.v.), the selective M3 muscarinic receptor antagonist, 4-DAMP (1.0 mg/kg, 1.0 ml/kg, i.v.), or the selective M2 muscarinic receptor antagonist, methoctramine (0.5 mg/kg, 1.0 ml/kg, i.v.). All doses of oxotremorine were then repeated in the order they were given prior to administration of the test compound. Maximal changes in chronotropy and inotropy were measured with each injection. Values are expressed as percent of the control value taken immediately before injection of each dose of oxotremorine.
Inhibition of baroreceptor reflex
Rats (300–325 g) were used to determine the ability of desloratadine to block the vagally-mediated bradycardic component of the baroreceptor reflex. Catheters were inserted into the femoral artery (PE-50) and femoral vein (PE-10) and exteriorized between the animal's shoulders. Animals were allowed to recover for a minimum of three days. The experiment lasted two days per animal. Before the baroreceptor reflex of each animal was measured, the animal was allowed an acclimation period. The first day consisted of control baroreceptor reflex measurement. On the second day, either desloratadine (1.0 mg/kg, i.v.) or atropine (0.5 mg/kg, i.v.) was given prior to baroreceptor challenge. After an acclimation period, baroreceptor reflex measurement was repeated. The baroreceptor reflex was initiated by increasing doses of both phenylephrine and sodium nitroprusside to increase or decrease blood pressure, respectively. Dosing was discontinued when a maximal change of 50 mmHg was achieved. Blood pressure was recorded with Powerlab Data Acquisition system via a transducer attached to the arterial line. The linear regression feature in Origin 6.0 (OriginLab Corp.; Northhampton, MA) analyzed data and the slopes compared with SigmaStat 2.0 (Jandel Scientific Software; San Rafael, CA).
Statistical analysis
Data obtained from the murine models of oxotremorine-induced salivation, lacrimation, and tremor were analyzed using a one-way repeated measures ANOVA with a Dunn's post hoc test. Data obtained from animals pretreated with either phenylephrine or vehicle was analyzed via the paired t-test. Inhibition of oxotremorine-induced alterations in cardiac function before and after either desloratadine, methoctramine, or 4-DAMP administration was analyzed via the paired t-test with changes in cardiac function at each dose of oxotremorine being compared. In all statistical comparisons, P ≤ 0.05 was deemed statistically significant.
Authors' contributions
G. Howell performed all of the in vivo cardiovascular testing, drug preparation, data analysis on cardiovascular results, and drafted the manuscript. L. West performed in vivo salivation, lacrimation, and tremor testing, drug preparation, and data analysis on salivation and lacrimation results. C. Jenkins, B. Lineberry, and D. Yokum assisted L. West with oxotremorine-induced tremor experimentation on conscious mice. R. Rockhold is the corresponding author and principal investigator.
Acknowledgements
This work was supported by a grant from Aventis Pharmaceuticals and in part by an award from the Howard Hughes Medical Institute.
==== Refs
Kay GG Harris AG Loratadine: a non-sedating antihistamine. Review of its effects on cognition, psychomotor performance, mood and sedation Clin Exp Allergy 1999 29 Suppl 3 147 150 10444229 10.1046/j.1365-2222.1999.0290S3147.x
DuBuske LM Second-generation antihistamines: the risk of ventricular arrhythmias Clin Ther 1999 21 281 295 10211532 10.1016/S0149-2918(00)88286-7
Anthes JC Gilchrest H Richard C Eckel S Hesk D West REJ Williams SM Greenfeder S Billah M Kreutner W Egan RE Biochemical characterization of desloratadine, a potent antagonist of the human histamine H(1) receptor Eur J Pharmacol 2002 449 229 237 12167464 10.1016/S0014-2999(02)02049-6
Handley DA McCullough JR Fang Y Wright SE Smith ER Descarboethoxyloratadine, a metabolite of loratadine, is a superior antihistamine Ann Allergy Asthma Immunol 1997 78 P164
Cardelus I Puig J Bou J Jauregui J Llenas J Fernandez AG Palacios JM Xerostomia and mydriasis, two muscarinic peripheral side effects associated with descarboethoxyloratadine, the main metabolite of loratadine Br J Pharmacol 1998 123 267P
Cardelus I Anton F Beleta J Palacios JM Anticholinergic effects of desloratadine, the major metabolite of loratadine, in rabbit and guinea-pig iris smooth muscle Eur J Pharmacol 1999 374 249 254 10422766 10.1016/S0014-2999(99)00310-6
Mayhan WG Disruption of blood-brain barrier during acute hypertension in adult and aged rats Am J Physiol 1990 258 H1735 H1738 2113773
Kramer JM Aragones A Waldrop TG Reflex cardiovascular responses originating in exercising muscles of mice J Appl Physiol 2001 90 579 585 11160056
Ringdahl B Jenden DJ Pharmacological properties of oxotremorine and its analogs Life Sci 1983 32 2401 2413 6343743 10.1016/0024-3205(83)90365-X
Giraldo E Micheletti R Montagna E Giachetti A Vigano MA Ladinsky H Melchiorre C Binding and functional characterization of the cardioselective muscarinic antagonist methoctramine J Pharmacol Exp Ther 1988 244 1016 1020 3252019
Wess J Angeli P Melchiorre C Moser U Mutschler E Lambrecht G Methoctramine selectively blocks cardiac muscarinic M2 receptors in vivo Naunyn Schmiedebergs Arch Pharmacol 1988 338 246 249 3057387 10.1007/BF00173395
Moriya H Takagi Y Nakanishi T Hayashi M Tani T Hirotsu I Affinity profiles of various muscarinic antagonists for cloned human muscarinic acetylcholine receptor (mAChR) subtypes and mAChRs in rat heart and submandibular gland Life Sci 1999 64 2351 2358 10374898 10.1016/S0024-3205(99)00188-5
Sanchez C Lembol HL The involvement of muscarinic receptor subtypes in the mediation of hypothermia, tremor, and salivation in male mice Pharmacol Toxicol 1994 74 35 39 8159635
Clissold SP Sorkin EM Goa KL Loratadine. A preliminary review of its pharmacodynamic properties and therapeutic efficacy Drugs 1989 37 42 57 2523301
Watanabe T Kakefuda A Tanaka A Takizawa K Hirano S Shibata H Yamagiwa Y Yanagisawa I Synthesis and biological evaluation of phenylacetyl derivatives having low central nervous system permeability as potent and selective M2 muscarinic receptor antagonists Chem Pharm Bull (Tokyo) 1998 46 53 68 9468637
Lavy UI Mulder D Salivary inhibition in mice and rabbits by a number of anticholinergics. A methodological investigation Arch Int Pharmacodyn Ther 1969 178 437 445 5355569
Espinola EB Oliveira MG Carlini EA Differences in central and peripheral responses to oxotremorine in young and aged rats Pharmacol Biochem Behav 1999 62 419 423 10080232 10.1016/S0091-3057(98)00192-0
Sanchez C Meier E Central and peripheral mediation of hypothermia, tremor and salivation induced by muscarinic agonists in mice Pharmacol Toxicol 1993 72 262 267 8372044
Boje KM Inhibition of nitric oxide synthase attenuates blood-brain barrier disruption during experimental meningitis Brain Res 1996 720 75 83 8782899 10.1016/0006-8993(96)00142-4
Mayhan WG Effect of lipopolysaccharide on the permeability and reactivity of the cerebral microcirculation: role of inducible nitric oxide synthase Brain Res 1998 792 353 357 9593993 10.1016/S0006-8993(98)00259-5
Bymaster FP Carter PA Zhang L Falcone JF Stengel PW Cohen ML Shannon HE Gomeza J Wess J Felder CC Investigations into the physiological role of muscarinic M2 and M4 muscarinic and M4 receptor subtypes using receptor knockout mice Life Sci 2001 68 2473 2479 11392615 10.1016/S0024-3205(01)01041-4
Nakamura M Tada Y Akaishi T Nakata K M3 muscarinic receptor mediates regulation of protein secretion in rabbit lacrimal gland Curr Eye Res 1997 16 614 619 9192172 10.1076/ceyr.16.6.614.5077
Schiavone A Brambilla A Muscarinic M3 receptors mediate secretion from sweat glands in the rat Pharmacol Res 1991 23 233 239 2068048
Du XY Schoemaker RG Bos E Saxena PR Characterization of the positive and negative inotropic effects of acetylcholine in the human myocardium Eur J Pharmacol 1995 284 119 127 8549614 10.1016/0014-2999(95)00384-W
Eglen RM Montgomery WW Whiting RL Negative and positive inotropic responses to muscarinic agonists in guinea pig and rat atria in vitro J Pharmacol Exp Ther 1988 247 911 917 2849676
Nishimaru K Tanaka Y Tanaka H Shigenobu K Positive and negative inotropic effects of muscarinic receptor stimulation in mouse left atria Life Sci 2000 66 607 615 10794516 10.1016/S0024-3205(99)00633-5
Yang JM Chung KT Yang SN Muscarinic activation causes biphasic inotropic response and decreases cellular Na+ activity in canine cardiac Purkinje fibers J Biomed Sci 1999 6 176 182 10343166 10.1159/000025385
Sharma VK Colecraft HM Wang DX Levey AI Grigorenko EV Yeh HH Sheu SS Molecular and functional identification of m1 muscarinic acetylcholine receptors in rat ventricular myocytes Circ Res 1996 79 86 93 8925573
Ponicke K Heinroth-Hoffmann I Brodde OE Demonstration of functional M3-muscarinic receptors in ventricular cardiomyocytes of adult rats Br J Pharmacol 2003 138 156 160 12522085 10.1038/sj.bjp.0704997
Krejci A Tucek S Quantitation of mRNAs for M(1) to M(5) subtypes of muscarinic receptors in rat heart and brain cortex Mol Pharmacol 2002 61 1267 1272 12021386 10.1124/mol.61.6.1267
Dhein S van Koppen CJ Brodde OE Muscarinic receptors in the mammalian heart Pharmacol Res 2001 44 161 182 11529684 10.1006/phrs.2001.0835
McMorn SO Harrison SM Zang WJ Yu XJ Boyett MR A direct negative inotropic effect of acetylcholine on rat ventricular myocytes Am J Physiol 1993 265 H1393 H1400 8238427
Gallo MP Alloatti G Eva C Oberto A Levi RC M1 muscarinic receptors increase calcium current and phosphoinositide turnover in guinea-pig ventricular cardiocytes J Physiol 1993 471 41 60 8120813
Wang Z Shi H Wang H Functional M3 muscarinic acetylcholine receptors in mammalian hearts Br J Pharmacol 2004 142 395 408 15148264 10.1038/sj.bjp.0705787
Haria M Fitton A Peters DH Loratadine. A reappraisal of its pharmacological properties and therapeutic use in allergic disorders Drugs 1994 48 617 637 7528133
Gupta S Banfield C Affrime M Marco A Cayen M Herron J Padhi D Desloratadine demonstrates dose proportionality in healthy adults after single doses Clin Pharmacokinet 2002 41 Suppl 1 1 6 12169040
Kreutner W Hey JA Anthes J Barnett A Young S Tozzi S Preclinical pharmacology of desloratadine, a selective and nonsedating histamine H1 receptor antagonist. 1st communication: receptor selectivity, antihistaminic activity, and antiallergenic effects Arzneimittelforschung 2000 50 345 352 10800633
Geha RS Meltzer EO Desloratadine: A new, nonsedating, oral antihistamine J Allergy Clin Immunol 2001 107 751 762 11295678 10.1067/mai.2001.114239
Henz BM The pharmacologic profile of desloratadine: a review Allergy 2001 56 Suppl 65 7 13 11243504 10.1034/j.1398-9995.2001.00101.x
|
16109168
|
PMC1192807
|
CC BY
|
2021-01-04 16:33:00
|
no
|
BMC Pharmacol. 2005 Aug 18; 5:13
|
utf-8
|
BMC Pharmacol
| 2,005 |
10.1186/1471-2210-5-13
|
oa_comm
|
==== Front
BMC Struct BiolBMC Structural Biology1472-6807BioMed Central London 1472-6807-5-101598516710.1186/1472-6807-5-10Research ArticleA model of the ternary complex of interleukin-10 with its soluble receptors Pletnev Sergei [email protected] Eugenia [email protected] Alexander [email protected] Alexander [email protected] Macromolecular Crystallography Laboratory, Center for Cancer Research, National Cancer Institute at Frederick, Frederick, MD21702-1201, USA2 Basic Research Program, Science Application International Corporation-Frederick, National Cancer Institute at Frederick, Frederick, MD21702-1201, USA2005 28 6 2005 5 10 10 1 4 2005 28 6 2005 Copyright © 2005 Pletnev 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
Interleukin-10 (IL-10) is a cytokine whose main biological function is to suppress the immune response by induction of a signal(s) leading to inhibition of synthesis of a number of cytokines and their cellular receptors. Signal transduction is initiated upon formation of a ternary complex of IL-10 with two of its receptor chains, IL-10R1 and IL-10R2, expressed on the cell membrane. The affinity of IL-10R1 toward IL-10 is very high, which allowed determination of the crystal structure of IL-10 complexed with the extracellular/soluble domain of IL-10R1, while the affinity of IL-10R2 toward either IL-10 or IL-10/sIL-10R1 complex is quite low. This so far has prevented any attempts to obtain structural information about the ternary complex of IL-10 with its receptor chains.
Results
Structures of the second soluble receptor chain of interleukin-10 (sIL-10R2) and the ternary complex of IL-10/sIL-10R1/sIL-10R2 have been generated by homology modeling, which allowed us to identify residues involved in ligand-receptor and receptor-receptor interactions.
Conclusion
The previously experimentally determined structure of the intermediate/binary complex IL-10/sIL-10R1 is the same in the ternary complex. There are two binding sites for the second receptor chain on the surface of the IL-10/sIL-10R1 complex, involving both IL-10 and sIL-10R1. Most of the interactions are hydrophilic in nature, although each interface includes two internal hydrophobic clusters. The distance between C-termini of the receptor chains is 25 Å, which is common for known structures of ternary complexes of other cytokines. The structure is likely to represent the biologically active signaling complex of IL-10 with its receptor on the surface of the cell membrane.
==== Body
Background
IL-10 is a pleiotropic cytokine (reviewed in [1,2]) that suppresses immune system response by inhibiting synthesis of proinflammatory cytokines and their cellular receptors [3,4]. In addition to immunosuppressive functions, IL-10 also possesses immunostimulatory activities, including stimulation of growth and proliferation of thymocytes, cytokine-activated T-cells, mast cells, and B-cells [5-9].
Signal transduction is initiated when IL-10 binds to its cellular receptor, which consists of two chains, IL-10R1 and IL-10R2. IL-10 is an intercalated dimer made of two identical polypeptides, each 160 amino acids long [10-12], packed in two compact, six-helix bundle domains (reviewed in [2,13-15]). The domains are related by a twofold symmetry axis passing through the interdomain interface. Both receptor chains have about a 200 amino acids-long extracellular domain, a 20 amino acids-long transmembrane helix, and an intracellular/cytoplasmic domain, which contains 322 or 62 amino acids for IL-10R1 and IL-10R2, respectively. When the ternary complex IL-10/IL-10R1/IL-10R2 is formed, tyrosine kinases Tyk2 and Jak1 become activated and phosphorylate specific tyrosine residues on signal transducers and activators of transcription, which eventually leads to activation of the transcription of corresponding genes [16,17]. Kinetic binding data on soluble IL-10 receptor chains (sIL-10R) showed that sIL-10R1 exhibits nanomolar binding constants for IL-10, while sIL-10R2 does not display any significant affinity for IL-10, although binding of sIL-10R2 to a preformed binary IL-10/sIL-10R1 complex is noticeably better [18]. It has been commonly accepted that at first IL-10 binds to its high affinity receptor IL-10R1, forming a binding site for IL-10R2, and then the binding of the latter receptor finalizes the ternary signaling complex. It has been also shown for the IFN-(/receptor complex, which is likely to be similar to the IL-10 complex, that under physiological conditions the receptor chains are already preassembled together, even before the ligand binds. Thus, binding of the ligand is necessary only for opening the intracellular domain regions to accommodate intracellular components of the downstream signal transduction pathway [19]. It was shown for IL-19 and IL-20, which also belong to the IL-10 family, that their final ternary complexes with the soluble receptors are very much alike, no matter what was the order of binding components in solution [20]. Therefore, the structure of the ternary IL-10/sIL-10R1/sIL-10R2 complex can be predicted on the basis of the structures of the intermediate binary complex IL-10/sIL-10R1 and the structure of sIL-10R2. The former, stable complex was crystallized [21] and its structure was determined [22]; however, all attempts to obtain stable ternary complexes suitable for crystallization have so far failed.
The known structures of binary and ternary cytokine-receptor complexes show that although they differ from each other in details, their overall organization is somewhat similar [22-29]. What is even more important, both ligand/receptor and receptor/receptor binding sites are located in topologically similar areas on the surface of the interacting proteins and involve similar structural elements of the molecules (Table 1). For example, in the case of the ligands, only two major receptor binding sites were found, and both included topologically conserved N- and C-terminal helices that form a left-handed four-helix bundle [30], which is a signature element of all helical cytokines, as well as loop AB. The difference was that in some cases site I was used as a high-affinity receptor binding site, while in other cases site II was the one where the high-affinity receptor was found to bind (Table 1). The crystal structure of a binary complex of IL-10 with sIL-10R1 showed that the high-affinity receptor binding site is site I, and it is reasonable to assume that sIL-10R2 will bind IL-10 in site II, making receptor/receptor contacts similar to the already published structures of ternary complexes.
We generated a structural model of the ternary complex of IL-10 with its two soluble receptors, sIL-10R1 and sIL-10R2. The model has been built on the assumption that the structure of the binary IL-10/sIL-10R1 complex must be preserved upon the binding of the second receptor chain, the complex must possess twofold symmetry, and that the structure of the complex of one domain of IL-10 with sIL-10R1 and sIL-10R2 should be somewhat similar to the already known structures of ternary complexes having two different receptor chains. To fulfill the latter requirement, we took the structure of IL-6/IL-6Rα /gp130 [29] as an example.
Results
Structure of sIL-10R2
The structures of sIL-10R2 and sIL-10R1 appear to be very similar (Fig. 1). They are both L-shaped molecules consisting of two elongated N- and C-terminal domains (D1 and D2), oriented at about 90° with respect to each other, each having FBN-III-like topology [31-33]. Each domain consists of seven β-strands organized in two antiparallel β-sheets – A, B, E and C, C', F, and G – that pack one against another in the shape of a sandwich. Loops L2-L4 of D1 and L5-L6 of D2 (Fig. 2) are involved in ligand binding. Domains D1 and D2 of sIL-10R2 comprise 96 and 94 amino acid residues, respectively (residues 1–103 and 109–206), and are connected to each other by a 5 amino acid-long linker (residues 104–108).
Deletions in the amino acid sequences of loops L2, L3 and β-strand C' in domain D1 of sIL-10R2 (Fig. 3) make its structure more compact than sIL-10R1. As a result, Tyr43, which is a conserved residue in class II receptors [34], is located a little higher than its counterpart in sIL-10R1. Although in sIL-10R1 Tyr43 plays a very important role in the interaction with the ligand, in sIL-10R2 it is located at the very edge of the receptor/ligand interface. Cysteines 54 and 62 of sIL-10R2 form a disulfide bond that is conserved in class II receptors [32], linking β-strand C' with the loop between the strand E and helix B. An equivalent disulfide bridge in sIL-10R1 is formed between cysteines 54 and 35 [12,22], linking together β-strands C' and C (Fig. 1).
Loop L3 of sIL-10R2 is two amino acids shorter in sIL-10R1 and does not protrude out of the D1 domain as much as in sIL-10R1. This alters the shape of the N-terminal part of the interdomain region and makes it complementary to the surface of the ligand-binding site II. Tyr70 is shifted ~2.5 Å inside of the D1 domain to compensate for the shortage of loop L3. In both receptors, the side chain of Tyr70 participates in the formation of a hydrophobic core of the N-terminal domain. Deletions in strand C' and loop L3 make the N-terminal domain of sIL-10R2 about 3 Å shorter than in sIL-10R1.
β-strands F (residues 77–81) of both receptors contain a conserved sequence (A/L)RVRA. Together with the sequences HSDWV in sIL-10R2 and HSNWT in sIL-10R1 (residues 87–91), this motif makes an extended π-cation system [35,36], usually found in the C-terminal domains of the type I family of cytokine receptors [37]. In sIL-10R1, a stacking structure is formed by His87, Arg80, Trp90, Arg78, Trp48, and Leu41; and is about 19 Å in length, which corresponds to roughly half of the domain length (~ 36 Å). It connects β-strands C, C', F, and G, and stabilizes the structure of the domain. In sIL-10R2, β-strand C' is located away from the β-strand F, which makes it impossible for the side chains of Phe48 and Leu41 to participate in the formation of a π-cation system, consisting in this case of only four residues – His87, Arg80, Trp90 and Arg78 – that keep together β-strands G and F. Although the π-cation system of sIL-10R2 is only ~11 Å long, which is about one-third of the total length of the domain, it still may be important for the stability of the structure of the domain. It is also likely that the π-cation system may play a dual role during the lifetime of a cytokine receptor. Mutagenesis studies of the erythropoietin receptor revealed the importance of the π-cation system for the transport of the receptor from endoplasmic reticulum to the Golgi apparatus [38,39].
In comparison to D2 of sIL-10R1, D2 of sIL-10R2 has three single amino acid deletions: Arg144, Pro154, and Glu167, as well as a single insertion of Pro176A (Fig. 3). The first deletion makes the connection between loop L5 and β-strand C shorter and moves the residues of loop L5 about 0.5 Å closer to the main body of the domain. Similarly, deletions of Pro154 and Glu167 make the loops connecting strands C and C' and strands C' and E slightly shorter (Fig. 1). β-strands F and G are held together by a disulfide bridge between Cys181 and Cys202, which is a common feature for class II cytokine receptors. Insertion of Pro176A in sIL-10R2 extends the loop between β-strands E and F by about 3 Å, bringing this part of the D2 domain closer to the cell membrane.
Hydrophobic patches on the surface and intermolecular hydrophobic clusters of the ternary IL-10/sIL-10R1/sIL-10R2 complex
Both the ligand and receptor chains have several hydrophobic patches located on the surface of the molecules (Table 2). Although the surface of IL-10 is mostly hydrophilic, with 86% of hydrophobic residues involved in the formation of the internal hydrophobic core [13], it also has three solvent-exposed hydrophobic patches (Table 2). Patch 1 is involved in the binding of sIL-10R1; patch 2 interacts with both sIL-10R1 and sIL-10R2, while patch 3 is located on the surface of helices A and D of IL-10 and participates in interactions with sIL-10R2.
sIL-10R1 contains six hydrophobic patches, three in the N-terminal domain and three in the C-terminal domain (Table 2). Patches 1 and 2 are similar in size and located on the opposite sides of the domain. Patch 1 is formed by five residues coming from β-strands C' and E and helix B, as well as three residues of the C-terminal domain flexible loop connecting β-strand B and helix A. This is the largest hydrophobic region in the first domain sIL-10R1, covering the surface of 475 Å2. Patch 2 is formed by seven residues coming from β-strands A, B, and E, and covers the surface of 448 Å2. Thr25 is at the center of the patch and although it is not a hydrophobic residue, its side chain is oriented in such a way that Oγ 1 points inside the protein molecule and forms a hydrogen bond with the main chain nitrogen of Ser10, whereas Cβ and Cγ 2 atoms are exposed to solvent. Patch 3 consists of three residues of loop L2 and covers the surface of 207 Å2 in the interdomain region of sIL-10R1. Patch 4 consists of eight residues covering the surface of 546 Å2 and is located close to the C-terminus, near the area which is likely to be in contact with the cellular membrane in vivo (Table 2). Patch 5 includes six residues coming from β-turn CC' and the first half of β-strand C'. It covers the surface of 409 Å2 and has an elongated shape. Patch 6 is formed by three residues of loop L5 and covers 253 Å2 of the sIL-10R1 interdomain region.
The pattern of hydrophobic patches of sIL-10R2 is similar, although not identical, to that of sIL-10R1. sIL-10R2 has seven patches, three in the N-terminal domain and four in the C-terminal domain. Patch 1 lies at the top of the sIL-10R2 N-terminal domain and is composed of 10 residues covering the surface of 635 Å2. This is the largest hydrophobic solvent-exposed area on the surface of sIL-10R2. Patches 2 and 3 (Table 2) cover 284 Å2 and 154 Å2, respectively, and are located in the proximity of the ligand/receptor interface. Patch 4 is formed by seven residues of β-strands C' and E, filling the gap between the strands and shielding the internal hydrophobic core of the molecule from the solvent. The area of the patch is 572 Å2. Patch 5 is formed by four residues coming from the β-turn connecting strands A and B, and is 274 Å2 in size. Patches 4 and 5 are located at the very bottom of the C-terminal domain and are likely to be in contact with the cellular membrane. Patch 6 covers 203 Å2 and is located in the central part of the receptor-receptor interface. Patch 7 consists of only two residues that belong to loops L5 and L6. It is 178 Å2 in size and is a part of the IL-10/sIL-10R2 interface.
Upon formation of the ternary complex, hydrophobic patches from different molecules compensate each other, creating intermolecular hydrophobic clusters inside of ligand-receptor and receptor-receptor interfaces (Table 2, Fig. 4). Thus, one IL-10 domain complexed with two receptor chains has two such clusters (1a and 2a, Fig. 4) in the IL-10/sIL-10R1 interface, two clusters (1b and 2b) in the IL-10/sIL-10R2 interface, and two clusters (1c and 2c) in the sIL-10R1/sIL-10R2 interface. Intermolecular hydrophobic clusters found in the ternary IL-10/sIL-10R1/sIL-10R2 complex are important elements that hold molecules together after they have attracted each other by long-range ionic interactions. It is worthwhile to note that patch 1 and 2 of sIL-10R1 and patches 1, 3 and 4 of sIL-10R2 remain exposed to solvent on the surface of the ternary complex.
The structure of the IL-10/sIL-10R1/sIL-10R2 complex
The ternary complex of IL-10/sIL-10R1/sIL-10R2 is a twofold symmetrical molecule in which each of the domains of IL-10 is bound with two receptor chains (Fig. 5). The receptor chains are bound to adjacent sides on the surface of a single IL-10 domain (Table 1). The sIL-10R1 binding site (site I) is formed by helix A, loop AB, and helix F', while the sIL-10R2 binding site (site II) is made by helices A and D (Fig. 6). sIL-10R1 and sIL-10R2, bound to the same IL-10 domain, interact with each other by their D2 domains, forming a receptor-receptor binding site (site III, Fig. 6). The distance between C-termini of sIL-10R1 and sIL-10R2 is 25.1 Å, which lies within the 25–32 Å range found in the structures of other ternary complexes [23-25,29]. The structure of the binary/intermediate complex of IL-10/sIL-10R1 [22] is the same in the ternary complex. We found only minor movements of some side chains involved in the interactions with sIL-10R2 in sites II and III. Both sIL-10R1 and sIL-10R2 receptor chains interact with ligand loops L2-L6. Site III involves interactions between residues coming from β-strands C, C', E, F and loop L6 of sIL-10R1, and residues from β-strands A, B, E and loop L5 of sIL-10R2 (Figs. 2, 6).
Site I interface (IL-10/sIL-10R1)
The interaction of IL-10 with sIL-10R1 in the ternary complex is the same as that found in the crystal structure of the intermediate/binary complex [22]. Briefly, the interface is formed by the residues of the second half (middle and C-terminal) of helix A, loop AB, and helix F' of IL-10, as well as loops L2-L6 of sIL-10R1. Twenty-seven residues of the ligand interact with 23 residues of the receptor chain, creating an extensive network of hydrogen bonds (70%) (Table 3, Figs. 6A, 6B) and hydrophobic interactions (30%) (Table 2). The total change in the accessible surface for both IL-10 and sIL-10R1, calculated using a spherical probe of the radius 1.40 Å, is about 2116 Å2. Thus, the contact area between IL-10 and sIL-10R1 is about 1058 Å2. The most important residues involved in receptor-ligand binding are Pro20, Arg24, Arg27, Gln38, Glu42, Lys138, Ser141, Asp144, Glu151, and Ile158 of IL-10 and Tyr43, Arg76, Arg96, Phe143, Ser190 and Arg191 of the sIL-10R1 [22]. The IL-10/sIL-10R1 interface includes two hydrophobic clusters located at the top and at the bottom of the interface (Table 2, Fig. 4).
Site II interface (IL-10/sIL-10R2)
The interface of site II is formed by helix A, loop CD, and helix D of IL-10, as well as by loops L2, L3, L6, and α-helix A of the sIL-10R2 (Table 2, Figs. 6A, 6C). Helix A interacts with loops L3, L6, and helix A, whereas helix D interacts with loop L2 of sIL-10R2. Site II is smaller than site I; its total surface is 568 Å2, which is only 54% of the surface of site I. IL-10 and sIL-10R2 each donate 13 residues to the interface. Interacting residues are mostly polar (~80%) and link the ligand to the receptor via hydrogen bonds (Table 3) and hydrophobic interactions (Table 2). The most important interface residues are Pro16, Asn18, Asn21, Arg24, Asp28, Arg32, Glu81, Ala89, His90, and Asn92 of IL-10 and Arg46, Ile47, Ser68, Lys69, Thr133, Asn138, Ser142, Trp143, and Arg191 of sIL-10R2. Most interactions occur between IL-10 helix A and loops L3 and L6 of sIL-10R2, with only a few contacts between IL-10 helix D and loop L2 of the sIL-10R2. Arg24 of IL-10 occupies a unique position in the complex. Its side chain projects down towards the cell membrane along the receptor-receptor interface and forms hydrogen bonds with the carbonyl oxygen of Val188 of sIL-10R1 and the side-chain atom Oδ 1 of Asn138 of sIL-10R2, linking both receptor chains together. The IL-10/sIL-10R2 interface also includes a hydrophobic cluster located at the bottom part of the interface, formed by Pro16 of the ligand and Trp143 of the receptor. The upper part of the IL-10/sIL-10R2 interface also includes a hydrophobic cluster formed by Ala89 and His90 of IL-10, and I47 of sIL-10R2. Unlike in the cluster 2c (Fig. 4), the hydrophobic area 1c is weaker but still is an important element of the IL-10/sIL-10R2 interface. A smaller interface area and a reduced amount of interacting residues reflect the low affinity of the sIL-10R2 towards its binding partners.
Site III interface (sIL-10R1/sIL-10R2)
The interface between sIL-10R1 and sIL-10R2 is formed by residues that belong to β-strands C, C', E, and F of sIL-10R1 and by residues of β-strands A, B, E, and loop L5 of sIL-10R2 (Tables 2, 3; Fig. 6D). The surfaces of C-terminal domains of the receptors are complementary. Within 3.8 Å distance cutoff, receptors 1 and 2 donate 15 and 13 residues, respectively, to form a contact area that is composed of 65% hydrophilic and 35% hydrophobic residues and is about 803Å2 in size (Tables 2, 3). The most important residues are Gly116, Phe117, Glu145, Glu147, His161, Lys162, Lys165, His166, Ser170, Leu172, Gly175, and Lys185 of sIL-10R1, and Gln110, Leu114, His119, Arg121, Leu123, Ala124, Lys126, Glu132, Asn138, Asn141, and Tyr166 of sIL-10R2. The contact area between two receptors can be divided into three parts. The upper part of site III is hydrophilic and contains a wide network of hydrogen bonds. Almost all residues that are located in this area are hydrophilic. The middle part of the interface has a roughly equal number of charged and uncharged residues, while the lower part of the receptor/receptor interface is almost solely hydrophobic, with an intermolecular hydrophobic cluster at the bottom of C-terminal domains of the receptors (Figs. 4, 6).
Glycosylation sites
The receptors sIL-10R1 and sIL-10R2 contain six and four potential N-linked glycosylation sites, respectively (Fig. 4). In the complex, all sites are fully exposed to the solvent and are not part of receptor-ligand or receptor-receptor interfaces. Moreover, each particular potential carbohydrate-binding site is located at a sufficient distance from the interfaces, which eliminates the possibility of a sugar chain clashing with a neighboring protein molecule. In sIL-10R1, Asn29 flanks the top end of the receptor N-terminal domain, whereas Asn53 and Asn89 are in the central part of the domain. They are located on loop BC, β-strand C', and in the region between β-strands G1 and G2, respectively. The N-terminal domain of sIL-10R2 also has three potential glycosylation sites: Asn33, Asn56, and Asn92. Asn33 is located close to the N-terminal end of the domain, whereas Asn56 and Asn92 are found in the central part of the domain, on the opposite sides of a β-sandwich.
The C-terminal domain of sIL-10R1 contains three potential glycosylation sites at Asn133, Asn156, and Asn168, located near the α-helix A, on β-turn CC', and at the beginning of β-strand E (Figs. 2, 4), respectively. The site at Asn133 is located near the N-terminal end of the domain, close to the interdomain region of sIL-10R1. Asn156 is at the bottom of the C-terminal domain, while the oligosaccharide attached to Asn168 is in the central part of domain D2. In the ternary complex, this residue is likely to be in contact with the cellular membrane. The C-terminal domain of sIL-10R2 has only one potential carbohydrate site, Asn153, which is close to the C-terminus of the receptor. This residue is a part of β-turn CC' and is very close to position Asn156 of sIL-10R1 upon superposition of the receptor chains.
Discussion
It is clear that the quality of any theoretical model should be assessed based on its agreement with experimental data such as, for example, mutagenesis. Unfortunately, we were unable to find any such data for the IL-10/IL-10R1/IL-10R2 complex, either in literature or as personal communications. Therefore, to evaluate the correctness of the model, we can only rely on the commonly accepted criteria: the model is in the global minimum of energy; general similarity to other ternary complexes of cytokines; usage of similar, well defined receptor binding sites on the surface of the ligand (Table 1); and correlation between the available kinetic binding data and intermolecular binding surfaces and contacts. As we have already mentioned, a general similarity of the ternary IL-10 complex to known structures of other ternary complexes and involvement of the similar ligand binding sites have been postulated from the beginning; thus, these conditions have been necessarily satisfied. The model of the ternary complex does correspond to the minimum of the energy, and the pattern of the interactions of sIL-10R2 with the binary IL-10/sIL-10R1 complex agrees well with its low affinity nature. Recent study of binding of IL-10 to IL-10R2 by peptide scans [40] showed that, although IL-10R2 did not form a binary complex with IL-10, it recognized regions of helix A and loop AB of IL-10 (amino acid residues 19–43) when they were presented as 15-mer peptides. In our model, the IL-10/IL-10R2 interface includes residues 18–32 of IL-10 (Table 3). Among the residues involved in creating the IL-10R1/IL-10R2 interface, seven residues of IL-10R1 have the same residue type as residues 33–43 of IL-10 (Table 2, Table 3). This may explain the necessity of formation of IL-10/IL-10R1 binary complex before IL-10R2 can join and complete the ternary signaling complex.
IL-10 binds to the receptor chains via two sites located on the adjacent sides of its four-helix bundle [30]. Site I is formed by helices A and F', whereas site II is formed by helices A and D (Table 1). Both receptor chains interact with IL-10 through their cytokine recognition motifs, which consist of the loops and β-turns of the receptor interdomain region connecting consecutive antiparallel β-strands [22-29]. Typically, the low-affinity receptor cannot bind to a ligand by itself. Its association occurs via a cooperative binding site that is formed by both the ligand and the high-affinity receptor. With the exception of the prolactin ternary complex, C-terminal domains of the receptors contact each other in the vicinity of the cellular membrane (Table 1). Such interaction brings the intracellular parts of the receptors together, resulting in activation of signal transduction.
Receptor chains of the ternary IL-10 complex interact with each other via β-strands C, C', E, and F of sIL-10R1 and β-strands E, B, A, and loop L5 of sIL-10R2, forming an interface between the sides of their C-terminal domains Tables (2, 3; Figs. 4, 5, 6). The distance between the Cα atoms of the C-termini of IL-10R1 and IL-10R2 is 25.1 Å. In the known structures of other ternary complexes, the distances between the Cα atoms of the last residues of the extracellular parts of the receptor chains vary between 25 and 32 Å.
The surfaces of IL-10 and of both the sIL-10R1 and sIL-10R2 receptors contain several hydrophobic patches (Table 2). In the ternary complex, most of these areas face each other to form intermolecular hydrophobic clusters. Each interface contains two such areas that flank opposite sides of contact regions (Fig. 4). The bottom clusters of IL-10/sIL-10R1 and IL-10/sIL-10R2 touch each other at the point where the ligand and both receptors join, creating a larger hydrophobic area formed by all three molecules. Such a cluster is not observed in other structures of ternary complexes and is, therefore, unique for the ternary IL-10 complex.
Both receptors have several potential glycosylation sites, which are located on protein surfaces in places where they do not conflict with the three-dimensional organization of the IL-10/sIL-10R1/sIL-10R2 complex (Fig. 4). The role of carbohydrates in forming and maintaining cytokine/receptor complexes is not well understood. It was shown that the granulocyte-macrophage colony-stimulating factor (GM-CSF) receptor α subunit requires N-glycosylation for binding and signaling [41]. Tunicamycin treatment inhibited GM-CSF binding in a dose-dependent manner, with a maximum of 85% inhibition at 3 μg/ml. Treatment of human leukemia HL-60 cells with tunicamycin completely blocked GM-CSF-induced tyrosine phosphorylation, which suggests that N-glycosylation of the receptor is necessary for intracellular signaling [41]. However, the non-glycosylated mutant of sIL-10R1 is capable of forming a stable binary complex with IL-10 [22]. GM-CSF is a glycoprotein in which glycosylation is not required for its biological activity. GM-CSF that is expressed in E. coli and thus, not glycosylated, retains full activity [42]. By contrast, the IL-10R2 binding region on IL-22 contains an N-linked carbohydrate that is likely important for binding [43]. Similarly, N-linked glycosylation of viral IL-6 is shown to increase gp130 binding and biological activity [44]. Taking into account that no carbohydrates are involved in the intermolecular interactions in the ternary IL-10/sIL-10R1/sIL-10R2 complex, as well as the fact that deglycosylated sIL-10R1 can form a quite stable binary complex with the IL-10, we may conclude that N-linked oligosaccharides are not essential for the formation of the biologically active ternary IL-10/IL-10R1/IL-10R2 complex. It is clear that more structural data are required to address the question of why some cytokines and their receptors depend upon the presence of carbohydrates, whereas others do not.
It was shown that a series of anti-human IL-10 antibodies can efficiently neutralize this cytokine [45]. Based on the recognition epitope, antibodies can be divided into three groups, A, B and C. The epitopes were identified using IL-10-derived, overlapping peptide scans prepared by spot synthesis. Antibodies of group A inhibit biological activity of IL-10 in an approximately equimolar ratio, at concentrations as low as 10 pM [45]. It has been shown that monoclonal antibody CB/RS/2 recognizes two binding regions of a discontinuous epitope on the surface of the molecule that comprises the N-terminal half of helix A and helix D [46]. Thus, this antibody binds to site II of the ligand and prevents IL-10 from association with its low-affinity receptor. An overlapping peptide scan technique has also been used for mapping IL-10 residues that bind to sIL-10R1, as well as mapping the residues of sIL-10R1 that interact with the ligand [47]. It was shown that residues of helices A and C of the ligand interact with the residues of loops L3-L6 of the high-affinity receptor. The structure of the IL-10/sIL-10R1 complex confirmed those results [22]. However, no mapping data exist for the residues that form the IL-10/sIL-10R2 interface. Such data would be very valuable to guide the design of biochemical experiments.
In the absence of sIL-10R1, IL-10 cannot bind to sIL-10R2, and the two receptor chains cannot interact with each other without a ligand. Therefore, assembly of the ternary complex consists of two steps. In the first step, IL-10 binds to sIL-10R1, forming a binary complex; subsequently, sIL-10R2 binds to the preformed IL-10/sIL-10R1 complex, completing the ternary complex. Such differences in binding abilities could be linked to the physical characteristics of the corresponding interfaces. The area of site I (1058 Å2) is about 25–45% larger than the areas of site II (568 Å2) and site III (803 Å2). The number of residues involved in interface formation is also larger for the IL-10/sIL-10R1 contact region. Site I is formed by a total of 50 residues, whereas sites II and III are composed of 26 and 28 residues, respectively. However, when combined together, sites II and III have an area and number of charged and hydrophobic interactions comparable to those of site I, which may explain the ability of the low-affinity receptor to bind to a preformed ligand-high-affinity receptor complex. This is also true if we compare the areas of binding sites I, II and III in other ternary complexes of cytokines such as GH, PL, EPO, and IL-6; there, the area of site I is usually larger than each of the areas of site II or III, whereas combined area of sites II and III are roughly equal to or larger than the area of site I [23,25,24,29].
Methods
The model of sIL-10R2, the low-affinity receptor, was generated using the published crystal structure of sIL-10R1 as a template (pdb code 1J7V ([22]). The original amino acid sequence of sIL-10R1 was mutated to that of sIL-10R2 in accordance with the alignment of their sequences shown in Figure 3 (20.4% identity and 53.1% similarity) [34,48]. Model building was followed by energy minimization, which included 500 cycles of positional refinement to optimize stereochemistry of newly built parts of the structure, followed by simulated annealing slow cool protocol and another 500 cycles of positional refinement. To preserve the overall fold, a 2.8 kcal/mole restraint has been imposed on all Cα atoms of the model. All other main and side-chain atoms were allowed to move freely within the appropriate stereochemical parameters.
A model of the ternary complex of one domain of IL-10 with sIL-10R1 and sIL-10R2 was generated on the assumption that the structure of the intermediate binary complex IL-10/sIL-10R1 does not change much upon binding of the second receptor chain, sIL-10R2, and that mutual arrangement of ligand and receptor chains is similar to what was found in the crystal structure of IL-6/IL-6-Rα /gp130 [29]. At the first step, the Cα atoms of the structure of one domain of IL-10 complexed with sIL-10R1 were superimposed onto Cα atoms of IL-6/sIL-6Rα, and then the structure of sIL-10R2 was superimposed onto gp130. Subsequently, the best fit of charge complementarity of sIL-10R2 to the intermediate binary complex was achieved upon superposition of the C-terminal domain of sIL-10R2 with the C-terminal domain of gp130, followed by rotation of sIL-10R2 as a whole around the axis perpendicular to the cell surface and passing through the center of mass of the IL-10 domain, followed again by translation along the surface of helices A and D. The second half of the hexameric IL-10/sIL-10R1/sIL-10R2 complex was generated by applying twofold symmetry (Fig. 5). Model building was followed by the energy minimization procedure under conditions described above until the drop in total energy between consecutive refinement steps was less than 1%. Root mean square deviations in positions of Cα atoms for IL-10, sIL-10R1 and sIL-10R2 in the ternary complex before and after energy minimization were 0.38 Å, 0.40 Å and 0.45 Å, respectively.
Sequence alignment was performed with program CLUSTALW 1.74 [48], model building with program CHAIN [49], and energy minimization with XPLOR 3.1 [50]; all superpositions were performed with program LSQKAB from CCP4 [51]. Figures 1, 4, and 6 were generated with program SETOR [52], while Figure 5 was generated with the program pyMOL (DeLano W.L. The PyMOL Molecular Graphics System 2002, DeLano Scientific, San Carlos, CA, USA).
Authors' contributions
SP and EM performed amino acid sequence alignment, generated the model and participated in writing the manuscript. AW participated in discussions and writing the manuscript. AZ conceived the study and participated in all stages of the work. All authors read and approved the final manuscript.
Acknowledgements
This publication has been funded in part with Federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. NO1-CO-12400. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade name, commercial products or organizations imply endorsement by the U.S. Government.
Figures and Tables
Figure 1 Stereo diagram of the superposition of sIL-10R1 and sIL-10R2. sIL-10R1 is green, sIL-10R2 is magenta, the disulfide bonds are shown in yellow.
Figure 2 Topology diagram of a single ligand domain associated with high- and low-affinity receptors. β-strands of the receptor molecules are shown as arrows and the helices are shown as rectangles. Binding loops of the receptor D1 domain are red and the loops of the D2 domain are blue. α-helices of single ligand domain are shown as circles. Helices A, B, C, and D (red) belong to one polypeptide chain of the ligand and helices E' and F' (green) belong to another polypeptide chain of the ligand. Highlighted helices (A, C, D, F') constitute the four-helix bundle that is involved in receptor binding.
Figure 3 Structure-based sequence alignment of soluble receptors sIL-10R1 and sIL-10R2. Aligned molecules have 20.4% of identical residues and 53.1% of homologous residues (red). Secondary structure elements are shown in green. Cysteines involved in disulfide bonds are highlighted in yellow, disulfide bonds are shown by black lines.
Figure 4 Stereo diagram of intermolecular hydrophobic clusters of the ternary IL-10/sIL-10R1/sIL-10R2 complex. Polypeptide chains of IL-10 are shown in orange and cyan, sIL-10R1 is green, and sIL-10R2 is magenta. Side chains of the residues that constitute hydrophobic clusters are shown in blue. Clusters 1a and 2a are the "top" and the "bottom" hydrophobic regions of IL-10/sIL-10R1 interface, clusters 1b and 2b are the "top" and the "bottom" hydrophobic regions of IL-10/sIL-10R2 interface and clusters 1c and 2c are the "top" and the "bottom" hydrophobic regions of sIL-10R1/sIL-10R2 interface. Disulfide bonds are shown in yellow. The potential glycosylation sites are shown in red.
Figure 5 Stereo diagram of the ternary complex of human IL-10/sIL-10R1/sIL-10R2. Polypeptide chains of IL-10 are shown in orange and cyan, sIL-10R1 is green, sIL-10R2 is magenta. The hypothetical cell membrane is perpendicular to the plane of the figure on the right and the twofold symmetry axis (not shown) is horizontal.
Figure 6 Stereo diagram of each interface within a single IL-10/sIL-10R1/sIL-10R2 signaling unit (panel A). Polypeptide chains of IL-10 are shown in orange and cyan. sIL-10R1 is green and sIL-10R2 is magenta. Panels B, C and D represent close-up view of each interface, including contact residues. (B)- an interface between IL-10 and sIL-10R1, (C)- an interface between IL-10 and sIL-10R2, (D)- an interface between sIL-10R1 and sIL-10R2. Intermolecular hydrogen bonds calculated with in 3.2 Å distance cutoff are shown as blue dotted lines.
Table 1 Positions of the cytokine-receptor binding sites in known structures of ternary and binary complexes.
Cytokine-receptor complexes (PDB code/reference) Site I (Ligand/Receptor) Site II (Ligand/Receptor) Site III (Recept/Recept)
SSE* Ligand SSE* Receptor SSE* Ligand SSE* Receptor Contacting Domains
Ternary complexes
hGH/GHbp/GHbp (3HHR [23]) A, loop AB, D L1, L3-L6 A, C L1, L3-L6 D2-D2
hEPO/sEPOR/sEPOR (1EER [24]) A, loop AB, D L1-L6 A, C L1-L3, L5, L6 no interaction
oPL/srPRLR/srPRLR (1F6F [25]) A, loop AB, D L1-L6 A, C L1-L4 D2-D2
hIL6/sIL6Rα /gp130 (D1-D3) (1P9M [29]) A, D L2-L6 A, C L1-L3, L5 D2-D3
Binary complexes
hINF-γ /sINF-γ Rα ** (1FG9 [53]) A, loop AB, F' L2, L3, L5, L6
hLIF/gp130 (D2-D3) (1PVH [54]) A, C L2, L3, L5
G-CSF/G-CSFR (BN-BC) (1CD9 [27]) A, C L2-L6
hIL4/IL4-R1 (1IAR [26]) A, C L1-L6
hIL-10/sIL-10R1** (1J7V [22]) A, loop AB, F' L2-L6
cmvIL-10/sIL-10R1** (1LQS [55]) A, loop AB, F' L2-L6
vIL-6/gp130 (1I1R [56]) A, C L2, L3, L5
Ternary hIL-10 complex
hIL10/sIL10R1/sIL10R2**,*** A, loop AB, F' L2-L6 A, D L2, L3, L5, L6 D2-D2
(*)- SSE: Secondary structure elements.
(**)- In INF-γ and IL-10 structures helix F' is a topological equivalent of helix D of other helical cytokines.
(***)- Interface areas were calculated with program Surface (CCP4) using spherical probe of the radius 1.40 Å2.
Table 2 Hydrophobic patches of IL-10, sIL-10R1 and sIL-10R2
IL-10
Interaction site
Patch 1
Residue
Access. (%) Surface area 189 Å2
L46 L53 I145 A152
50 25 34 44 IL-10/sIL-10R1 interface,
with patch 3 of sIL-10R1
Patch 2
Residue
Access. (%) Surface area 294 Å2
H14 P16 P20 I158
73 66 71 25 IL-10/sIL10R1 interface,
with patch 6 of sIL-10R1
IL-10/sIL10R2 interface,
with Patch 7 of sIL-10R2
Patch 3
Residue
Access. (%) Surface area 183 Å2
F36 A89 H90
33 86 45 IL-10/sIL-10R2 interface
with Patch 2 (sIL-10R2)
IL-10R1 Domain D1
Patch 1
Residue
Access. (%) Surface area 475 Å2
I51 Y60 A64 V65 L67
18 39 90 34 46
M129 P131 A132
60 38 100 No interaction
Patch 2
Residue
Access. (%) Surface area 448 Å2
P6 P9 W12 T25 P26 P28 L58
45 45 45 39 56 65 56 No interaction
Patch 3
Residue
Access. (%) Surface area 207 Å2
Y43 G44 I45
44 66 83 IL-10/sIL10R1 interface,
with patch 1 of IL-10
Domain D2
Patch 4
Residue
Access. (%) Surface area 546 Å2
I113 H114 G116 F117
60 56 47 58
L119 L172 G175 V177
41 84 61 69 sIL-10R1/sIL10R2 interface,
with patch 5 of sIL-10R2
Patch 5
Residue
Access. (%) Surface area 409 Å2
V153 P154 G155 F157 F159 H161
45 56 55 73 43 56 sIL-10R1/sIL10R2 interface,
with patch 6 of sIL-10R2
Patch 6
Residue
Access. (%) Surface area 253 Å2
H142 F143 A189
77 62 35 IL-10/sIL10R1 interface,
with patch 2 of IL-10
IL-10R2 Domain D1
Patch 1
Residue
Access. (%) Surface area 635 Å2
L2 G3 M4 P6 P7 A28
77 78 45 56 40 91
F29 G32 F83 A84
97 93 34 85 No interaction
Patch 2
Residue
Access. (%) Surface area 284 Å2
L41 I47 F48
29 88 62 IL-10/sIL10R2 interface,
with patch 3 of IL-10
Patch 3
Residue
Access. (%) Surface area 154 Å2
C98 P98A V99
52 48 42 No interaction
Domain D2
Patch 4
Residue
Access. (%) Surface area 572 Å2
F160 I162 F169 L172
73 26 12 71
L175 P176A W177
49 73 57 No interaction
Patch 5
Residue
Access. (%) Surface area 274 Å2
V113 L114 A115 H119
47 62 100 40 sIL-10R1/sIL10R2 interface,
with patch 4 of sIL-10R1
Patch 6
Residue
Access. (%) Surface area 203 Å2
P107 L123 A124
69 65 54 sIL-10R1/sIL10R2 interface,
with patch 5 of sIL-10R1
Patch 7
Residue
Access. (%) Surface area 178 Å2
W143 P189
57 48 IL-10/sIL10R2 interface,
with patch 2 of IL-10
Table 3 Intermolecular hydrogen bonds of the IL10/sIL10R1/sIL10R2 complex
IL-10-IL-10R1 interface
IL-10 SSE IL-10R1 SSE Distance Å
R24L NE A R191R O L6 3.0
R24L NH1 A V188R O L6 2.9
R24L NH1 A R191R O L6 3.0
R27L NE A S190R O L6 2.9
R27L NH1 A D100R OD2 L4 3.2
K34L NZ A D100R OD1 L4 2.9
Q38L OE1 A R96R N L4 3.0
D41L O loop AB R76R NH1 L3 2.9
Q42L O loop AB R76R NH2 L3 3.0
D44L OD1 loop AB R76R NH2 L3 2.8
D44L O loop AB G44R N L2 2.9
N45L N loop AB E46R OE2 L2 3.0
K138S NZ F' Y43R OH L2 2.8
S141S O F' R96R NH2 L4 2.9
D144S OD1 F' R96R NH2 L4 2.8
D144S OD2 F' R96R NH2 L4 3.0
E151S OE1 F' R191R NH1 L6 2.9
E151S OE2 F' R191R NH1 L6 2.8
E151S OE1 F' R191R NH2 L6 2.9
IL-10-IL-10R2 interface
N18L OD1 A R191A NH1 L6 2.7
N18L ND2 A S142A OG1 L6 2.7
N21L ND2 A N138A O α A 2.8
R24L NH2 A N138A OD1 α A 2.7
R24L NH2 A T133A O loop Bα A 3.1
D28L OD1 A K69A NZ L3 2.8
D28L OD2 A K69A NZ L3 3.0
D28L O A K69A NZ L3 3.2
R32L NH1 A S68A OG L3 3.3
E81L OE2 loop CD R46A NH2 L2 3.3
N92L OD1 D R46A NH1 L2 2.8
IL-10R1-IL-10R2 interface
E145R OE1 C N138A OD1 L5 3.2
E145R OE2 C N141A ND2 L5 2.7
E147R OE2 C K126A NZ loop Bα A 3.2
H161R NE2 C' A124A O loop Bα A 2.9
K162R NZ C' Q110A OE1 A 2.6
K165R NZ loop C'E N141A OD1 L5 2.7
H166R NE2 loop C'E Y166A OH loop C'E 3.0
S170R OG E R121A NH1 B 3.1
S170R O E R121A NH1 B 2.6
K185R NZ F E132A OE2 loop Bα A 3.2
(*)- SSE: Secondary structure elements. (**)- In INF-γ and IL-10 structures helix F' is a topological equivalent of helix D of other helical cytokines. (***)- Interface areas were calculated with program Surface (CCP4) using spherical probe of the radius 1.40 Å2.
==== Refs
Moore KW de Waal MR Coffman RL O'Garra A Interleukin-10 and the interleukin-10 receptor Annu Rev Immunol 2001 19 683 765 11244051 10.1146/annurev.immunol.19.1.683
Pestka S Krause CD Sarkar D Walter MR Shi Y Fisher PB Interleukin-10 and Related Cytokines and Receptors Annu Rev Immunol 2004 22 929 979 15032600 10.1146/annurev.immunol.22.012703.104622
de Waal Malefyt R Abrams J Bennett B Figdor CG de Vries J Interleukin-10 (IL-10) inhibits cytokine synthesis by human monocytes: an autoregulatory role of IL-10 produced by monocytes J Exp Med 1991 174 1209 1220 1940799 10.1084/jem.174.5.1209
D'Andrea A Aste-Amezaga M Valiante NM Ma X Kubin M Trinchieri G Interleukin 10 (IL-10) inhibits human lymphocyte interferon gamma-production by suppressing natural killer cell stimulatory factor/IL-12 synthesis in accessory cells J Exp Med 1993 178 1041 1048 8102388 10.1084/jem.178.3.1041
Defrance T Vanbervliet B Briere F Durand I Rousset F Banchereau J Interleukin 10 and transforming growth factor beta cooperate to induce anti-CD40-activated naive human B cells to secrete immunoglobulin A J Exp Med 1992 175 671 682 1371300 10.1084/jem.175.3.671
Rousset F Garcia E Defrance T Peronne C Vezzio N Hsu DH Kastelein R Moore KW Banchereau J Interleukin 10 is a potent growth and differentiation factor for activated human B lymphocytes Proc Natl Acad Sci USA 1992 89 1890 1893 1371884
Thompson-Snipes L Dhar V Bond MW Mosmann TR Moore KW Rennick DM Interleukin 10: a novel stimulatory factor for mast cells and their progenitors J Exp Med 1991 173 507 510 1899106 10.1084/jem.173.2.507
MacNeil IA Suda T Moore KW Mosmann TR Zlotnik A IL-10, a novel growth cofactor for mature and immature T cells J Immunol 1990 145 4167 4173 2124236
Suda T O'Garra A MacNeil I Fischer M Bond MW Zlotnik A Identification of a novel thymocyte growth-promoting factor derived from B cell lymphomas Cell Immunol 1990 129 228 240 2194678 10.1016/0008-8749(90)90200-B
Zdanov A Schalk-Hihi C Gustchina A Tsang M Weatherbee J Wlodawer A Crystal structure of interleukin-10 reveals the functional dimer with an unexpected topological similarity to interferon g. Structure 1995 3 591 601 8590020 10.1016/S0969-2126(01)00193-9
Walter MR Nagabhushan TL Crystal structure of interleukin 10 reveals an interferon gamma- like fold Biochemistry 1995 34 12118 12125 7547951 10.1021/bi00038a004
Zdanov A Schalk-Hihi C Wlodawer A Crystal structure of human interleukin-10 at 1.6 Å resolution and a model of a complex with its soluble receptor Protein Sci 1996 5 1955 1962 8897595
Zdanov A Structural features of the interleukin-10 family of cytokines Curr Pharm Des 2004 10 3873 3884 15579076 10.2174/1381612043382602
Walter MR Strucure of interleukin-10/interleukin-10R1 complex: a paradigm for class 2 cytokine activation Immunol Res 2002 26 303 308 12403368 10.1385/IR:26:1-3:303
Walter MR Crystal structures of alpha-helical cytokine-receptor complexes: we've only scratched the surface BioTechniques 2002 Suppl 46 57 12395927
Liu Y Wei SHY Ho ASY Malefyt RW Moore KW Expression cloning and characterization of a human IL-10 receptor J Immunol 1994 152 1821 1829 8120391
Kotenko SV Krause CD Izotova LS Pollack BP Wu W Pestka S Identification and functional characterization of a second chain of the interleukin-10 receptor complex EMBO J 1997 16 5894 5903 9312047 10.1093/emboj/16.19.5894
Logsdon NJ Jones BC Josephson K Cook J Walter MR Comparison of interleukin-22 and interleukin-10 soluble receptor complexes J Interferon Cytokine Res 2002 22 1099 1112 12513909 10.1089/10799900260442520
Krause CD Mei E Xie J Jia Y Bopp MA Hochstrasser RM Pestka S Seeing the light: preassembly and ligand-induced changes of the interferon gamma receptor complex in cells Mol Cell Proteomics 2002 1 805 815 12438563 10.1074/mcp.M200065-MCP200
Pletnev S Magracheva E Kozlov S Tobin G Kotenko SV Wlodawer A Zdanov A Characterization of the recombinant extracellular domains of human interleukin-20 receptors and their complexes with interleukin-19 and interleukin-20 Biochemistry 2003 42 12617 12624 14580208 10.1021/bi0354583
Josephson K McPherson DT Walter MR Purification, crystallization and preliminary X-ray diffraction of a complex between IL-10 and soluble IL-10R1 Acta Crystallogr D Biol Crystallogr 2001 57 1908 1911 11717514 10.1107/S0907444901016249
Josephson K Logsdon NJ Walter MR Crystal structure of the IL-10/IL-10R1 complex reveals a shared receptor binding site Immunity 2001 15 35 46 11485736 10.1016/S1074-7613(01)00169-8
de Vos AM Ultsch M Kossiakoff AA Human growth hormone and extracellular domain of its receptor: Crystal structure of the complex Science 1992 255 306 312 1549776
Syed RS Reid SW Li C Cheetham JC Aoki KH Liu B Zhan H Osslund TD Chirino AJ Zhang J Finer-Moore J Elliott S Sitney K Katz BA Matthews DJ Wendoloski JJ Egrie J Stroud RM Efficiency of signalling through cytokine receptors depends critically on receptor orientation Nature 1998 395 511 516 9774108 10.1038/26773
Elkins PA Christinger HW Sandowski Y Sakal E Gertler A de Vos AM Kossiakoff AA Ternary complex between placental lactogen and the extracellular domain of the prolactin receptor Nat Struct Biol 2000 7 808 815 10966654 10.1038/79047
Hage T Sebald W Reinemer P Crystal structure of the interleukin-4/receptor alpha chain complex reveals a mosaic binding interface Cell 1999 97 271 281 10219247 10.1016/S0092-8674(00)80736-9
Aritomi M Kunishima N Okamoto T Kuroki R Ota Y Morikawa K Atomic structure of the GCSF-receptor complex showing a new cytokine-receptor recognition scheme Nature 1999 401 713 717 10537111 10.1038/44394
Walter MR Windsor WT Nagabhushan TL Lundell DJ Lunn CA Zauodny PJ Narula SK Crystal structure of a complex between interferon-gamma and its soluble high-affinity receptor Nature 1995 376 230 235 7617032 10.1038/376230a0
Boulanger MJ Chow DC Brevnova EE Garcia KC Hexameric structure and assembly of the interleukin-6/IL-6 alpha-receptor/gp130 complex Science 2003 300 2101 2104 12829785 10.1126/science.1083901
Presnell SR Cohen FE Topological distribution of four-a-helix bundles Proc Natl Acad Sci USA 1989 86 6592 6596 2771946
Bork P Holm L Sander C The Immunoglobulin Fold. Structural classification, sequence patterns and common core J Mol Biol 1994 242 309 320 7932691 10.1006/jmbi.1994.1582
Bazan JF Structural design and molecular evolution of a cytokine receptor superfamily Proc Natl Acad Sci USA 1990 87 6934 6938 2169613
Bazan JF Shared architecture of hormone binding domains in type I and II interferon receptors Cell 1990 61 753 754 2140527 10.1016/0092-8674(90)90182-E
Kotenko SV The family of IL10 related cytokines and their receptors: related, but to what extent? Cytokine Growth Factor Rev 2002 13 223 240 12486876 10.1016/S1359-6101(02)00012-6
Dougherty DA Cation-pi interactions in chemistry and biology: a new view of benzene, Phe, Tyr, and Trp Science 1996 271 163 168 8539615
Ma JC Dougherty DA The Cationminus signpi Interaction Chem Rev 1997 97 1303 1324 11851453 10.1021/cr9603744
Scrutton NS Raine AR Cation-pi bonding and amino-aromatic interactions in the biomolecular recognition of substituted ammonium ligands Biochem J 1996 319 ( Pt 1) 1 8 8870640
Hilton DJ Watowich SS Katz L Lodish HF Saturation mutagenesis of the WSXWS motif of the erythropoietin receptor J Biol Chem 1996 271 4699 4708 8617735 10.1074/jbc.271.23.13754
Hilton DJ Watowich SS Murray PJ Lodish HF Increased cell surface expression and enhanced folding in the endoplasmic reticulum of a mutant erythropoietin receptor Proc Natl Acad Sci U S A 1995 92 190 194 7816815
Wolk K Witte E Reineke U Witte K Friedrich M Sterry W Asadullah K Volk HD Sabat R Is there an interaction between interleukin-10 and interleukin-22? Genes Immun 2005 6 8 18 15526001 10.1038/sj.gene.6364224
Ding DX Vera JC Heaney ML Golde DW N-glycosylation of the human granulocyte-macrophage colony-stimulating factor receptor alpha subunit is essential for ligand binding and signal transduction J Biol Chem 1995 270 24580 24584 7592677 10.1074/jbc.270.41.24580
Kaushansky K O'Hara PJ Hart CE Forstrom JW Hagen FS Role of carbohydrate in the function of human granulocyte-macrophage colony-stimulating factor Biochemistry 1987 26 4861 4867 3499177 10.1021/bi00389a038
Logsdon NJ Jones BC Allman JC Izotova L Schwartz B Pestka S Walter MR The IL-10R2 binding hot spot on IL-22 is located on the N-terminal helix and is dependent on N-linked glycosylation J Mol Biol 2004 342 503 514 15327950 10.1016/j.jmb.2004.07.069
Dela Cruz CS Lee Y Viswanathan SR El Guindy AS Gerlach J Nikiforow S Shedd D Gradoville L Miller G N-linked glycosylation is required for optimal function of Kaposi's sarcoma herpesvirus-encoded, but not cellular, interleukin 6 J Exp Med 2004 199 503 514 14970177 10.1084/jem.20031205
Sabat R Seifert M Volk HD Glaser RW Neutralizing murine monoclonal anti-interleukin-10 antibodies enhance binding of antibodies against a different epitope Mol Immunol 1996 33 1103 1111 9047376 10.1016/S0161-5890(96)00072-7
Reineke U Schneider-Mergener J Glaser RW Stigler RD Seifert M Volk HD Sabat R Evidence for conformationally different states of interleukin-10: binding of a neutralizing antibody enhances accessibility of a hidden epitope J Mol Recognit 1999 12 242 248 10440995 10.1002/(SICI)1099-1352(199907/08)12:4<242::AID-JMR461>3.0.CO;2-1
Reineke U Sabat R Volk HD Schneider-Mergener J Mapping of the interleukin-10/interleukin-10 receptor combining site Protein Sci 1998 7 951 960 9568901
Combet C Blanchet C Geourjon C Deleage G NPS@: network protein sequence analysis Trends Biochem Sci 2000 25 147 150 10694887 10.1016/S0968-0004(99)01540-6
Sack JS CHAIN-a crystallographic modeling program J Mol Graph 1988 6 244 245
Brünger AT X-PLOR Version 31 A System for X-ray Crystallography and NMR 1992 New Haven, Yale University Press
CCP4 Collaborative Computational Project, Number 4, 1994. The CCP4 Suite: Programs for Protein Crystallography. Acta Crystallogr 1994 D50 760 763
Evans SV SETOR: hardware-lighted three-dimensional solid model representations of macromolecules J Mol Graph 1993 11 134 138 8347566 10.1016/0263-7855(93)87009-T
Thiel DJ le Du MH Walter RL D'Arcy A Chene C Fountoulakis M Garotta G Winkler FK Ealick SE Observation of an unexpected third receptor molecule in the crystal structure of human interferon-gamma receptor complex Structure Fold Des 2000 8 927 936 10986460 10.1016/S0969-2126(00)00184-2
Boulanger MJ Bankovich AJ Kortemme T Baker D Garcia KC Convergent mechanisms for recognition of divergent cytokines by the shared signaling receptor gp130 Mol Cell 2003 12 577 589 14527405 10.1016/S1097-2765(03)00365-4
Jones BC Logsdon NJ Josephson K Cook J Barry PA Walter MR Crystal structure of human cytomegalovirus IL-10 bound to soluble human IL-10R1 Proc Natl Acad Sci USA 2002 99 9404 9409 12093920 10.1073/pnas.152147499
Chow D He X Snow AL Rose-John S Garcia KC Structure of an extracellular gp130 cytokine receptor signaling complex Science 2001 291 2150 2155 11251120 10.1126/science.1058308
|
15985167
|
PMC1192808
|
CC BY
|
2021-01-04 16:03:52
|
no
|
BMC Struct Biol. 2005 Jun 28; 5:10
|
utf-8
|
BMC Struct Biol
| 2,005 |
10.1186/1472-6807-5-10
|
oa_comm
|
==== Front
BMC SurgBMC Surgery1471-2482BioMed Central London 1471-2482-5-171608350910.1186/1471-2482-5-17Research ArticleComplications after cryosurgery with new miniature cryoprobes in long hollow bones: An animal trial Popken Frank [email protected] Peter [email protected] Heike [email protected] Timmo [email protected] Marfalda [email protected] Jürgen H [email protected] Peer [email protected] Department of Orthopaedic Surgery, University of Cologne, Josef-Stelzmann-Str. 9, 50931 Cologne, Germany2 Institute of Pathology, University of Cologne, Josef-Stelzmann-Str. 9, 50931 Cologne, Germany3 Institute of Experimental Medicine, University of Cologne, Robert-Koch-Str. 10, 50931 Köln, Germany2005 7 8 2005 5 17 17 29 6 2004 7 8 2005 Copyright © 2005 Popken et al; licensee BioMed Central Ltd.2005Popken 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 vitro studies show that new miniature cryoprobes are suitable for cryoablation of bone tissue. The aim of this animal trial on 24 sheep was to examine the perioperative complications, particularly the danger of embolism, of cryoablation when using miniature cryoprobes.
Methods
Cryoablations with 2 freeze-thaw cycles each were carried out in the epiphysis of the right tibia and the metaphysis of the left femur. Pulmonary artery pressure (PAP) and central venous pressure (CVP) were measured. Throughout the intra- and perioperative phase, heart rate and oxygen saturation by pulse oxymetry, blood gas and electrolytes were monitored regularly. Postoperative complications were examined up to 24 weeks postoperativ.
Results
As result, no significant increase of PAP, CVP or heart rate were observed. Blood gases were unremarkable, with pO2 and pCO2 remaining constant throughout the operation. Regarding pH, standard bicarbonate and base excess, only a non-significant shift towards a slight acidosis was seen. There was a mean hemoglobin decrease of 0.5 g/dl. One animal showed postoperative wound infection and wound edge necrosis. No major peri- and postoperative complications associated with cryosurgery of bone were observed, especially regarding clinically relevant pulmonary embolism.
Conclusion
Surgery with new types of miniature cryoprobes appears to be a safe alternative to or a complement to conventional resection of abnormal bone tissue.
==== Body
Background
Surgical treatment of bone tumours often requires generous resection of bone, leaving defects which are difficult to span. Freezing tumours with liquid nitrogen was introduced in the late 1960's as an adjuvant treatment to extend the surgical margin of excision for intralesional resection or for curettage by pouring or spraying the nitrogen directly into the bone cavity [1-9]. Animal trials by Gage et al. [10] have shown that devitalised bone matrix can serve as a framework for new periostal and endostal growth, and hence that the former tumour space can be bridged with autologous, healthy bone tissue. However, the freezing procedure is difficult to control, and therefore harbours risks of injury for the patient [11,12] and the surgical team, as well as of gas embolisms caused by evaporation [13] and spread of the liquid nitrogen.
Aside from the open use of liquid nitrogen, closed systems for treating bone tumours have also been used [14], although they never became popular because the cooling power of the cryoprobes then used was low compared to their diameter [15]. Recent technical advances [16] made it possible for us to develop new probes for cryoablation of bone tissue and test these in animal trials [17,18]. The efficiency of these procedure and the extent of tissue distruction is well documented in a former study with a comparable setup [19].
Various complications have been reported, ranging from soft tissue wound infection and fractures [20] to bone marrow and fat embolism caused by the spread of the ice front due to an increase in intramedullar pressure [21]. These miniature cryoprobes with a minimised diameter allow precise control of the freezing process, thus avoiding uncontrolled freezing of soft parts and healthy bone tissue, as well as a sudden expansion of the ice front. Aim of this animal trial was to determine whether the use of modern miniature cryoprobes can avoid the above described complications.
Methods
A commercially available cryotherapy system (Erbokryo CS-6-System, Erbe, Tübingen, Germany) was used for cryoablation. This system consists of a casing with a control board and up to 6 vacuum-isolated flexible tubes with a cryoprobe (diameter 3.2 mm) at the end. A cryoprobe creates a cold zone 3 cm long. The Erbokryo is also fitted with computer-controlled temperature sensors (diameter 1.2 mm) witch allow 6 simultaneous measurements.
The cryoprobes were introduced via an access hole 3.6 mm in diameter drilled perpendicular to the cortical substance. Temperature was measured inside the cortical substance. Two 15-minute freezing cycles were done with the probe at full power, with a 6-minute thaw in between. Prior to starting the cryoablation, the position of the cryoprobe was checked radiologically and recorded.
24 sheep with a mean weight of 61 kg (range 39–78 kg) were placed under general anaesthesia and, using a single cryoprobe introduced through a lateral access hole, one cryoablation was done in the distal diametaphyseal transitional region of the head of the medial left tibia and the right femur of each sheep. For the tibial head, a medial access hole was drilled and the cryoprobe introduced centrally 1.5 cm below the joint and pushed to the other side of the cortical substance.
For the femoral cryoablation, the cryoprobe was introduced through a distal, posterolateral access hole in the area of the linea aspera at the diametaphyseal transition. In addition, 4 temperature sensors were introduced through access holes drilled radially at 1 cm from the bore hole. The cryoprobes were only introduced 1 cm into the bone to avoid freezing the cortical substance. Thus, the necrosis zone (which roughly corresponds to the -10°C isotherm [24]) only comprised an area of 2.4 × 2.4 in the outer cortical substance. Control holes and access holes were drilled on the contralateral sides (left femur, head of the right tibia).
All operations were done under 600 mg clindamycin i.v. and 12 hours prior to surgery, each animal also received thrombosis prophylaxis (0.3 ml nadroparin calcium [Fraxiparin®] s.c.). There was no postop thrombosis prophylaxis since all animals were fully mobile after anaesthesia. The pulmonary-arterial pressure (PAP) and the central venous pressure (CVP) were measured via a pulmonary catheter inserted into the jugular vein. Measurements were done before the first cryoablation (1, Fig. 1, 2, 3), after the first cryoablation on the right femur (2, Fig. 1, 2, 3) and the head of the left tibia (3, Fig. 1, 2, 3), as well as after drilling the control holes immediately after shifting the animals from the right back to the left unilateral recumbent position (4, Fig. 1, 2, 3). Intra- and perioperative monitoring was complemented by measurements of heart rate (Fig. 3) and oxygen saturation via pulse oxymetry, as well as blood counts (Fig. 4), deep body temperature (Fig. 5), blood gases and electrolytes after each cryoablation. Postoperative complications were monitored clinically. 8 animals were sacrificed at 8, 16 and 32 weeks postop and tissue samples taken from the lungs and from blood vessels in the areas of cryoablation were examined for signs of embolism. Samples were taken from each lobe and from the femoral vein, fixed in formaline and stained with hematoxylin and eosin (HE). Furthermore, the ablation sites were examined histologically for infection of the soft tissue respectively osteomyelitis. X-rays were taken after the animals were sacrificed.
Figure 1 Mean central venous pressure (CVP): Measurements were done before (point 1) and after the two cryoablations (point 2, 3) and after drilling the controll holes (point 4).
Figure 2 Mean pulmonary artery pressure (PAP) before (point 1) and after the two cryoablations (point 2, 3) and after drilling the controll holes (point 4).
Figure 3 Mean heartrate measured at different times of the operation (see fig. 1, 2).
Figure 4 Mean hemoglobin (HB) measured before and directly after the operation.
Figure 5 Mean deep temperature measured before and after the operation.
Results
Mean duration of operation was 4 h 22 min, mean duration of anaesthesia 5 h 13 min. In 18 of 24 cases, the mean pulmonary-arterial pressure PAP and the mean central venous pressure (CVP) showed no significant changes at any of the 4 sampling times (Figs. 1 and 2). The PAP rose in 2 animals while in 4 other cases, the haemodynamics could not be determined for technical reasons. There was only a temporary increase of the mean PAP and CVP while the animals were in the right unilateral recumbent position, but all values normalised when the animals were shifted to their left side. The heart increased marginally during the operation (Fig. 3).
Blood gas analysis showed an increase in pCO2 at constant oxygen saturation after the animals were shifted to their right side. In 5 of the animals, the pCO2 remained slightly higher even after they were shifted to their left side. PH, base excess, electrolytes and lactate showed no changes at any sampling time. Hemoglobin decreased marginally (Fig. 4). The deep body temperature (taken central-venously) before and after cryosurgery was not obviously related to any single cryoablation, but it did show a mean decrease of 1.25°C over the course of the entire operation (Fig. 5). Clinical follow-up observations yielded no remarkable findings. The mean time of recovery – i.e., from the end of anaesthesia until the animals were able to stand up again – was 86 minutes. None of the 24 animals showed any clinically significant respiratory insufficiency. One animal developed a severe wound infection which required treatment. The wound was excised and showed secondary healing under antibiotics (Fig. 6).
Figure 6 Superficial wound infection 2 weeks after cryosurgery.
Autopsy revealed clean scars in all animals. There were macroscopically and microscopically no signs of ongoing or resolved osteomyelitis. All vein samples were free of old or recent thrombi. The lungs were grossly unremarkable. Histology of the bronchopulmonary arteries showed no evidence of acute or cronic embolism or thrombus. Thus, none of the animals had suffered an embolism even at the segment level. Some did show a marked bronchial pneumonia, although it followed an asymptomatic course and was therefore of no consequence. No animal showed spontaneous fractures, and X-rays taken at the end of the trial revealed no evidence of healed fractures.
Discussion
Intravasation of adipose tissue or bone marrow can lead to acute blockage of pulmonary microcirculation, with increased resistance in the arterial pulmonary capillaries and a secondary increase of pressure in the pulmonary artery and the right atrium [22]. This causes the pCO2 to rise and oxygen saturation to decrease, with tachycardia resulting from left ventricular volume deficiency. To date, no comparable trials on large animals have been done examining cryoablation in bone tissue with modern cryoprobes. Kerschbaumer et al. [23] found no lung embolisms in rabbits after cryosurgery. On the other hand, Oeseburg also used rabbits and observed a large number of bone marrow embolisms in the extraosseous veins immediately after cryoablation [21].
One of the aims of this study was the detection of larger, clinically significant lung embolisms, we restricted ourselves to venously measurable hemodynamic parameters so as not to cause additional, iatrogenic complications. Likewise, we decided against a transesophageal ultrasound probe for detecting microemboli [24,25] since the animals already were administered a large transesophageal tube to aspirate gastric juices during the operation, thus leaving no room for an ultrasound probe.
None of the animals showed histological evidence of lung embolisms, nor were any embolism-specific hemodynamic phenomena or blood gas changes observed. The rise in pCO2 seen in all animals at the end of the operation is likely to be due to the impaired pulmonary gas exchange and decreased venous flow resulting from the increased intraabdominal pressure, which in turn can probably be attributed to the distended rumen during while the animals were lying on their right side. The rise in pCO2 explains the slight acidosis which the animals showed towards the end of the operation, and is therefore not to be seen as pathological.
In our opinion, the absence of significant embolisms is due to the small diameter of the probe, which prevents the intramedullar pressure from rising when the probe is introduced. A further reason could be the controlled expansion of the ice front in the bone, which prevents intramedullar pressure from peaking, and hence bone marrow or adipose tissue from being pressed out of the marrow cavity.
As expected, we did not observe any decrease in body temperature after cryosurgery as was reported for small animals such as mice and rats [26]. The minimal reduction in body temperature which we did see towards the end of the operation can be explained by the normal cooling of the body despite a heating pad. We believe that the more pronounced decrease in body temperature in small animals is due to their smaller body volume, next to which the cryoprobe is comparatively much larger. Hence, the results of our trials with large animals can be extrapolated more readily to human patients than can results from similar trials with small animals, and a significant decrease in body temperature is not to be expected in human patients.
Except for hemoglobin, all blood chemistry values remained essentially unchanged during the operation. The drop in hemoglobin by on the average 0.5 g/dl is not clinically significant.
Clinical follow-up revealed one serious wound infection, which underscores the tendency these wounds have for infection [20]. Even so, it would seem to us that the risk of infection can be controlled with perioperative administration of antibiotics, as is also evident from the histology of the treated bone sections, none of which developed acute inflammation. Nevertheless, it must be admitted that the tissue treated here was healthy bone in animals with an intact immune system, and a higher infection rate must be expected when applying this method to patients with advanced malignancy.
Convalescence after cryosurgery is associated with changes in bone stability, a topic which few studies have addressed so far. Gage et al. (1967) reported 11 spontaneous fractures in 20 dogs, where the entire cross-section of the femur was frozen over a length of 4.5–7 cm [9]. Further studies report a maximum reduction in bone stability some 8 weeks after cryosurgery [27,20]. The absence of fractures in our trial shows that limiting bone necrosis by controlled freezing and minimum tissue loss when introducing the cryoprobe helps minimise the reduction in bone stability, and hence prevent fractures. To be sure, the size of a tumour dictates the extent of the freezing zone, so that stabilising measures may be necessary.
Conclusion
In conclusion the use of modern miniature cryoprobes for cryoablation of bone tissue seems to be a gentle method, and the complications reported for earlier systems did not occur in this study. Therefore, this cryosurgical technique has a pontential in human subjects and could be usefull either to complement conventional resections, or else as a minimally invasive alternative procedure.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
FP and PM carried out the operations and drafted the manuscript. MB and TK carried out the hole perioperativ treatment of the animals and the monitoring throughout the intra- and perioperative phase. HE performed the histological investigations. JHF participated in the design of the study and performed the statistical analysis. PE conceived of the study, and participated in its design and coordination. All authors read and approved the final manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
This study was performed with approval of the local animal care committee and supported by the Köln Fortune Programm/Faculty of Medicine, University of Cologne, Germany.
==== Refs
Malawer MM Bickels J Meller I Buch RG Henshaw RW Kollender Y Cryosurgery in the treatment of giant cell tumor. A long-term followup study Clin Orthop 1999 359 176 188 10078141 10.1097/00003086-199902000-00019
Marcove RC Miller TR Cahan WC The treatment of primary and metastatic bone tumors by repetitive freezing Bull N Y Acad Med 1968 44 532 544 5239935
Marcove RC Miller TR Treatment of primary and metastatic bone tumors by cryosurgery JAMA 1969 207 1890 1894 5304523 10.1001/jama.207.10.1890
Marcove RC Weis LD Vaghaiwalla MRT Pearson R Huvos AG Cryosurgery in the Treatment of Giant Cell Tumors of the bone Cancer 1978 41 957 969 638982
Marcove RC Weis LD Vaghaiwalla MR Pearson R Cryosurgery in the treatment of giant cell tumors of bone: a report of 52 consecutive cases Clin Orthop 1978 134 275 289 729255
Marcove RC Sheth DS Brien EW Huvos AG Healey JH Conservative surgery for giant cell tumors of the sacrum. The role of cryosurgery as a supplement to curettage and partial excision Cancer 1994 74 1253 60 8055446
Schreuder HW Conrad EU 3rdBruckner JD Howlett AT Sorensen LS Treatment of simple bone cysts in children with curettage and cryosurgery J Pediatr Orthop 1997 17 814 820 9591989 10.1097/00004694-199711000-00022
Schreuder HW van Egmond J van Beem HB Veth RP Monitoring during cryosurgery of bone tumors J Surg Oncol 1997 65 40 45 9179266 10.1002/(SICI)1096-9098(199705)65:1<40::AID-JSO8>3.0.CO;2-O
Schreuder HW Pruszczynski M Veth RP Lemmens JA Treatment of benign and low-grade malignant intramedullary chondroid tumours with curettage and cryosurgery Eur J Surg Oncol 1998 24 120 6 9591027 10.1016/S0748-7983(98)91459-7
Gage AA Greene GW Neiders ME Emmings FG Freezing bone without excision. An experimental study of bone-cell destruction and manner of regrowth in dogs JAMA 1966 196 770 774 5952306 10.1001/jama.196.9.770
Marcove RC Miller TR Cahan WC The treatment of primary and metastatic bone tumors by repetitive freezing Bull N Y Acad Med 1968 44 532 544 5239935
Marcove RC Weis LD Vaghaiwalla MRT Pearson R Huvos AG Cryosurgery in the Treatment of Giant Cell Tumors of the bone Cancer 1978 41 957 969 638982
Schreuder HW van Beem HB Veth RP Venous gas embolism during cryosurgery for bone tumors J Surg Oncol 1995 60 196 200 7475071
Russe W Kerschbaumer F Bauer R Kryochirurgie in der Orthopädie Orthopäde 1984 13 142 150
Hewitt PM Zhao J Akhter J Morris DL A Comparative Laboratory Study of Liquid Nitrogen and Argon Gas Cryosurgery Systems Cryobiology 1997 35 303 308 9425653 10.1006/cryo.1997.2039
Marcove RC Miller TR Cahan WC The treatment of primary and metastatic bone tumors by repetitive freezing Bull N Y Acad Med 1968 44 532 544 5239935
Popken F Bertram C Land M König DP Bilgic M Jeschkeit S Hackenbroch MH Fischer JH The cryosurgical ablation of bone tissue by means of a new miniature cryoprobe – adaption of the method for an in-vitro and in-vivo application to bone Z Orthop Ihre Grenzgeb 2001 139 64 69 11253524 10.1055/s-2001-11872
Popken F Bertram C König DP Rütt J Land M Hackenbroch MH The cryosurgical ablation of bone tissue by means of a new miniture cryoprobe – adaption of the method for an in-vitro and in-vivo application to bone Arch Orthop Trauma Surg 2002 122 129 133 11927992 10.1007/s00402-001-0371-6
Popken F Land M Bosse M Erberich H Meschede P König DP Fischer JH Eysel P Cryosurgery in long bones by means of a new miniature cryoprobe: An experimental in vivo study of the cryosurgical temperature field in sheep Eur J Surg Oncol 2003 29 542 547 12875863 10.1016/S0748-7983(03)00069-6
Kerschbaumer F Weiser G Neuerer G Russe W Bauer R Cryolesions of Bone. An Experimental Study. Part I: Examinations in technique of controlled cryolesion in bone Arch Orthop Traumat Surg 1980 96 5 9 10.1007/BF01246134
Oeseburg HB Cryochirurgische behandeling van enkele beentumoren Proschrift 1977 Groningen
Schlag G Schliep HJ Dingeldein E Grieben A Ringsdorf W Does methylmethacrylate induce cardiovascular complications during alloarthroplastic surgery of the hip joint? Anaesthesist 1976 25 60 7 1083699
Kerschbaumer F Krösel P Schlag G Cryogenic surgery of the bone – examinations of hemodynamics and tissue Beitr Orthop u Traumatol 1982 29 134 139
Popovic AD Milovanovic B Neskovic AN Pavlovski K Putnikovic B Hadzagic I Detection of massive pulmonary embolism by transesophageal echocardiography Cardiology 1992 80 94 99 1611638
van der Wouw PA Bax M Images in clinical medicine. Massive pulmonary embolism N Engl J Med 1997 6, 336 416 9010148 10.1056/NEJM199702063360605
Myers RS Hammond WG Ketcham AS A method for cryosurgical investigation of mouse tumors Int Surg 1969 52 232 233 5804096
Fisher AD Williams DF Bradley PF The effect of cryosurgery on the strength of bone Br J Oral Surg 1978 15 215 222 272917 10.1016/0007-117X(78)90003-3
|
16083509
|
PMC1192809
|
CC BY
|
2021-01-04 16:28:04
|
no
|
BMC Surg. 2005 Aug 7; 5:17
|
utf-8
|
BMC Surg
| 2,005 |
10.1186/1471-2482-5-17
|
oa_comm
|
==== Front
BMC SurgBMC Surgery1471-2482BioMed Central London 1471-2482-5-171608350910.1186/1471-2482-5-17Research ArticleComplications after cryosurgery with new miniature cryoprobes in long hollow bones: An animal trial Popken Frank [email protected] Peter [email protected] Heike [email protected] Timmo [email protected] Marfalda [email protected] Jürgen H [email protected] Peer [email protected] Department of Orthopaedic Surgery, University of Cologne, Josef-Stelzmann-Str. 9, 50931 Cologne, Germany2 Institute of Pathology, University of Cologne, Josef-Stelzmann-Str. 9, 50931 Cologne, Germany3 Institute of Experimental Medicine, University of Cologne, Robert-Koch-Str. 10, 50931 Köln, Germany2005 7 8 2005 5 17 17 29 6 2004 7 8 2005 Copyright © 2005 Popken et al; licensee BioMed Central Ltd.2005Popken 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 vitro studies show that new miniature cryoprobes are suitable for cryoablation of bone tissue. The aim of this animal trial on 24 sheep was to examine the perioperative complications, particularly the danger of embolism, of cryoablation when using miniature cryoprobes.
Methods
Cryoablations with 2 freeze-thaw cycles each were carried out in the epiphysis of the right tibia and the metaphysis of the left femur. Pulmonary artery pressure (PAP) and central venous pressure (CVP) were measured. Throughout the intra- and perioperative phase, heart rate and oxygen saturation by pulse oxymetry, blood gas and electrolytes were monitored regularly. Postoperative complications were examined up to 24 weeks postoperativ.
Results
As result, no significant increase of PAP, CVP or heart rate were observed. Blood gases were unremarkable, with pO2 and pCO2 remaining constant throughout the operation. Regarding pH, standard bicarbonate and base excess, only a non-significant shift towards a slight acidosis was seen. There was a mean hemoglobin decrease of 0.5 g/dl. One animal showed postoperative wound infection and wound edge necrosis. No major peri- and postoperative complications associated with cryosurgery of bone were observed, especially regarding clinically relevant pulmonary embolism.
Conclusion
Surgery with new types of miniature cryoprobes appears to be a safe alternative to or a complement to conventional resection of abnormal bone tissue.
==== Body
Background
Surgical treatment of bone tumours often requires generous resection of bone, leaving defects which are difficult to span. Freezing tumours with liquid nitrogen was introduced in the late 1960's as an adjuvant treatment to extend the surgical margin of excision for intralesional resection or for curettage by pouring or spraying the nitrogen directly into the bone cavity [1-9]. Animal trials by Gage et al. [10] have shown that devitalised bone matrix can serve as a framework for new periostal and endostal growth, and hence that the former tumour space can be bridged with autologous, healthy bone tissue. However, the freezing procedure is difficult to control, and therefore harbours risks of injury for the patient [11,12] and the surgical team, as well as of gas embolisms caused by evaporation [13] and spread of the liquid nitrogen.
Aside from the open use of liquid nitrogen, closed systems for treating bone tumours have also been used [14], although they never became popular because the cooling power of the cryoprobes then used was low compared to their diameter [15]. Recent technical advances [16] made it possible for us to develop new probes for cryoablation of bone tissue and test these in animal trials [17,18]. The efficiency of these procedure and the extent of tissue distruction is well documented in a former study with a comparable setup [19].
Various complications have been reported, ranging from soft tissue wound infection and fractures [20] to bone marrow and fat embolism caused by the spread of the ice front due to an increase in intramedullar pressure [21]. These miniature cryoprobes with a minimised diameter allow precise control of the freezing process, thus avoiding uncontrolled freezing of soft parts and healthy bone tissue, as well as a sudden expansion of the ice front. Aim of this animal trial was to determine whether the use of modern miniature cryoprobes can avoid the above described complications.
Methods
A commercially available cryotherapy system (Erbokryo CS-6-System, Erbe, Tübingen, Germany) was used for cryoablation. This system consists of a casing with a control board and up to 6 vacuum-isolated flexible tubes with a cryoprobe (diameter 3.2 mm) at the end. A cryoprobe creates a cold zone 3 cm long. The Erbokryo is also fitted with computer-controlled temperature sensors (diameter 1.2 mm) witch allow 6 simultaneous measurements.
The cryoprobes were introduced via an access hole 3.6 mm in diameter drilled perpendicular to the cortical substance. Temperature was measured inside the cortical substance. Two 15-minute freezing cycles were done with the probe at full power, with a 6-minute thaw in between. Prior to starting the cryoablation, the position of the cryoprobe was checked radiologically and recorded.
24 sheep with a mean weight of 61 kg (range 39–78 kg) were placed under general anaesthesia and, using a single cryoprobe introduced through a lateral access hole, one cryoablation was done in the distal diametaphyseal transitional region of the head of the medial left tibia and the right femur of each sheep. For the tibial head, a medial access hole was drilled and the cryoprobe introduced centrally 1.5 cm below the joint and pushed to the other side of the cortical substance.
For the femoral cryoablation, the cryoprobe was introduced through a distal, posterolateral access hole in the area of the linea aspera at the diametaphyseal transition. In addition, 4 temperature sensors were introduced through access holes drilled radially at 1 cm from the bore hole. The cryoprobes were only introduced 1 cm into the bone to avoid freezing the cortical substance. Thus, the necrosis zone (which roughly corresponds to the -10°C isotherm [24]) only comprised an area of 2.4 × 2.4 in the outer cortical substance. Control holes and access holes were drilled on the contralateral sides (left femur, head of the right tibia).
All operations were done under 600 mg clindamycin i.v. and 12 hours prior to surgery, each animal also received thrombosis prophylaxis (0.3 ml nadroparin calcium [Fraxiparin®] s.c.). There was no postop thrombosis prophylaxis since all animals were fully mobile after anaesthesia. The pulmonary-arterial pressure (PAP) and the central venous pressure (CVP) were measured via a pulmonary catheter inserted into the jugular vein. Measurements were done before the first cryoablation (1, Fig. 1, 2, 3), after the first cryoablation on the right femur (2, Fig. 1, 2, 3) and the head of the left tibia (3, Fig. 1, 2, 3), as well as after drilling the control holes immediately after shifting the animals from the right back to the left unilateral recumbent position (4, Fig. 1, 2, 3). Intra- and perioperative monitoring was complemented by measurements of heart rate (Fig. 3) and oxygen saturation via pulse oxymetry, as well as blood counts (Fig. 4), deep body temperature (Fig. 5), blood gases and electrolytes after each cryoablation. Postoperative complications were monitored clinically. 8 animals were sacrificed at 8, 16 and 32 weeks postop and tissue samples taken from the lungs and from blood vessels in the areas of cryoablation were examined for signs of embolism. Samples were taken from each lobe and from the femoral vein, fixed in formaline and stained with hematoxylin and eosin (HE). Furthermore, the ablation sites were examined histologically for infection of the soft tissue respectively osteomyelitis. X-rays were taken after the animals were sacrificed.
Figure 1 Mean central venous pressure (CVP): Measurements were done before (point 1) and after the two cryoablations (point 2, 3) and after drilling the controll holes (point 4).
Figure 2 Mean pulmonary artery pressure (PAP) before (point 1) and after the two cryoablations (point 2, 3) and after drilling the controll holes (point 4).
Figure 3 Mean heartrate measured at different times of the operation (see fig. 1, 2).
Figure 4 Mean hemoglobin (HB) measured before and directly after the operation.
Figure 5 Mean deep temperature measured before and after the operation.
Results
Mean duration of operation was 4 h 22 min, mean duration of anaesthesia 5 h 13 min. In 18 of 24 cases, the mean pulmonary-arterial pressure PAP and the mean central venous pressure (CVP) showed no significant changes at any of the 4 sampling times (Figs. 1 and 2). The PAP rose in 2 animals while in 4 other cases, the haemodynamics could not be determined for technical reasons. There was only a temporary increase of the mean PAP and CVP while the animals were in the right unilateral recumbent position, but all values normalised when the animals were shifted to their left side. The heart increased marginally during the operation (Fig. 3).
Blood gas analysis showed an increase in pCO2 at constant oxygen saturation after the animals were shifted to their right side. In 5 of the animals, the pCO2 remained slightly higher even after they were shifted to their left side. PH, base excess, electrolytes and lactate showed no changes at any sampling time. Hemoglobin decreased marginally (Fig. 4). The deep body temperature (taken central-venously) before and after cryosurgery was not obviously related to any single cryoablation, but it did show a mean decrease of 1.25°C over the course of the entire operation (Fig. 5). Clinical follow-up observations yielded no remarkable findings. The mean time of recovery – i.e., from the end of anaesthesia until the animals were able to stand up again – was 86 minutes. None of the 24 animals showed any clinically significant respiratory insufficiency. One animal developed a severe wound infection which required treatment. The wound was excised and showed secondary healing under antibiotics (Fig. 6).
Figure 6 Superficial wound infection 2 weeks after cryosurgery.
Autopsy revealed clean scars in all animals. There were macroscopically and microscopically no signs of ongoing or resolved osteomyelitis. All vein samples were free of old or recent thrombi. The lungs were grossly unremarkable. Histology of the bronchopulmonary arteries showed no evidence of acute or cronic embolism or thrombus. Thus, none of the animals had suffered an embolism even at the segment level. Some did show a marked bronchial pneumonia, although it followed an asymptomatic course and was therefore of no consequence. No animal showed spontaneous fractures, and X-rays taken at the end of the trial revealed no evidence of healed fractures.
Discussion
Intravasation of adipose tissue or bone marrow can lead to acute blockage of pulmonary microcirculation, with increased resistance in the arterial pulmonary capillaries and a secondary increase of pressure in the pulmonary artery and the right atrium [22]. This causes the pCO2 to rise and oxygen saturation to decrease, with tachycardia resulting from left ventricular volume deficiency. To date, no comparable trials on large animals have been done examining cryoablation in bone tissue with modern cryoprobes. Kerschbaumer et al. [23] found no lung embolisms in rabbits after cryosurgery. On the other hand, Oeseburg also used rabbits and observed a large number of bone marrow embolisms in the extraosseous veins immediately after cryoablation [21].
One of the aims of this study was the detection of larger, clinically significant lung embolisms, we restricted ourselves to venously measurable hemodynamic parameters so as not to cause additional, iatrogenic complications. Likewise, we decided against a transesophageal ultrasound probe for detecting microemboli [24,25] since the animals already were administered a large transesophageal tube to aspirate gastric juices during the operation, thus leaving no room for an ultrasound probe.
None of the animals showed histological evidence of lung embolisms, nor were any embolism-specific hemodynamic phenomena or blood gas changes observed. The rise in pCO2 seen in all animals at the end of the operation is likely to be due to the impaired pulmonary gas exchange and decreased venous flow resulting from the increased intraabdominal pressure, which in turn can probably be attributed to the distended rumen during while the animals were lying on their right side. The rise in pCO2 explains the slight acidosis which the animals showed towards the end of the operation, and is therefore not to be seen as pathological.
In our opinion, the absence of significant embolisms is due to the small diameter of the probe, which prevents the intramedullar pressure from rising when the probe is introduced. A further reason could be the controlled expansion of the ice front in the bone, which prevents intramedullar pressure from peaking, and hence bone marrow or adipose tissue from being pressed out of the marrow cavity.
As expected, we did not observe any decrease in body temperature after cryosurgery as was reported for small animals such as mice and rats [26]. The minimal reduction in body temperature which we did see towards the end of the operation can be explained by the normal cooling of the body despite a heating pad. We believe that the more pronounced decrease in body temperature in small animals is due to their smaller body volume, next to which the cryoprobe is comparatively much larger. Hence, the results of our trials with large animals can be extrapolated more readily to human patients than can results from similar trials with small animals, and a significant decrease in body temperature is not to be expected in human patients.
Except for hemoglobin, all blood chemistry values remained essentially unchanged during the operation. The drop in hemoglobin by on the average 0.5 g/dl is not clinically significant.
Clinical follow-up revealed one serious wound infection, which underscores the tendency these wounds have for infection [20]. Even so, it would seem to us that the risk of infection can be controlled with perioperative administration of antibiotics, as is also evident from the histology of the treated bone sections, none of which developed acute inflammation. Nevertheless, it must be admitted that the tissue treated here was healthy bone in animals with an intact immune system, and a higher infection rate must be expected when applying this method to patients with advanced malignancy.
Convalescence after cryosurgery is associated with changes in bone stability, a topic which few studies have addressed so far. Gage et al. (1967) reported 11 spontaneous fractures in 20 dogs, where the entire cross-section of the femur was frozen over a length of 4.5–7 cm [9]. Further studies report a maximum reduction in bone stability some 8 weeks after cryosurgery [27,20]. The absence of fractures in our trial shows that limiting bone necrosis by controlled freezing and minimum tissue loss when introducing the cryoprobe helps minimise the reduction in bone stability, and hence prevent fractures. To be sure, the size of a tumour dictates the extent of the freezing zone, so that stabilising measures may be necessary.
Conclusion
In conclusion the use of modern miniature cryoprobes for cryoablation of bone tissue seems to be a gentle method, and the complications reported for earlier systems did not occur in this study. Therefore, this cryosurgical technique has a pontential in human subjects and could be usefull either to complement conventional resections, or else as a minimally invasive alternative procedure.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
FP and PM carried out the operations and drafted the manuscript. MB and TK carried out the hole perioperativ treatment of the animals and the monitoring throughout the intra- and perioperative phase. HE performed the histological investigations. JHF participated in the design of the study and performed the statistical analysis. PE conceived of the study, and participated in its design and coordination. All authors read and approved the final manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
This study was performed with approval of the local animal care committee and supported by the Köln Fortune Programm/Faculty of Medicine, University of Cologne, Germany.
==== Refs
Malawer MM Bickels J Meller I Buch RG Henshaw RW Kollender Y Cryosurgery in the treatment of giant cell tumor. A long-term followup study Clin Orthop 1999 359 176 188 10078141 10.1097/00003086-199902000-00019
Marcove RC Miller TR Cahan WC The treatment of primary and metastatic bone tumors by repetitive freezing Bull N Y Acad Med 1968 44 532 544 5239935
Marcove RC Miller TR Treatment of primary and metastatic bone tumors by cryosurgery JAMA 1969 207 1890 1894 5304523 10.1001/jama.207.10.1890
Marcove RC Weis LD Vaghaiwalla MRT Pearson R Huvos AG Cryosurgery in the Treatment of Giant Cell Tumors of the bone Cancer 1978 41 957 969 638982
Marcove RC Weis LD Vaghaiwalla MR Pearson R Cryosurgery in the treatment of giant cell tumors of bone: a report of 52 consecutive cases Clin Orthop 1978 134 275 289 729255
Marcove RC Sheth DS Brien EW Huvos AG Healey JH Conservative surgery for giant cell tumors of the sacrum. The role of cryosurgery as a supplement to curettage and partial excision Cancer 1994 74 1253 60 8055446
Schreuder HW Conrad EU 3rdBruckner JD Howlett AT Sorensen LS Treatment of simple bone cysts in children with curettage and cryosurgery J Pediatr Orthop 1997 17 814 820 9591989 10.1097/00004694-199711000-00022
Schreuder HW van Egmond J van Beem HB Veth RP Monitoring during cryosurgery of bone tumors J Surg Oncol 1997 65 40 45 9179266 10.1002/(SICI)1096-9098(199705)65:1<40::AID-JSO8>3.0.CO;2-O
Schreuder HW Pruszczynski M Veth RP Lemmens JA Treatment of benign and low-grade malignant intramedullary chondroid tumours with curettage and cryosurgery Eur J Surg Oncol 1998 24 120 6 9591027 10.1016/S0748-7983(98)91459-7
Gage AA Greene GW Neiders ME Emmings FG Freezing bone without excision. An experimental study of bone-cell destruction and manner of regrowth in dogs JAMA 1966 196 770 774 5952306 10.1001/jama.196.9.770
Marcove RC Miller TR Cahan WC The treatment of primary and metastatic bone tumors by repetitive freezing Bull N Y Acad Med 1968 44 532 544 5239935
Marcove RC Weis LD Vaghaiwalla MRT Pearson R Huvos AG Cryosurgery in the Treatment of Giant Cell Tumors of the bone Cancer 1978 41 957 969 638982
Schreuder HW van Beem HB Veth RP Venous gas embolism during cryosurgery for bone tumors J Surg Oncol 1995 60 196 200 7475071
Russe W Kerschbaumer F Bauer R Kryochirurgie in der Orthopädie Orthopäde 1984 13 142 150
Hewitt PM Zhao J Akhter J Morris DL A Comparative Laboratory Study of Liquid Nitrogen and Argon Gas Cryosurgery Systems Cryobiology 1997 35 303 308 9425653 10.1006/cryo.1997.2039
Marcove RC Miller TR Cahan WC The treatment of primary and metastatic bone tumors by repetitive freezing Bull N Y Acad Med 1968 44 532 544 5239935
Popken F Bertram C Land M König DP Bilgic M Jeschkeit S Hackenbroch MH Fischer JH The cryosurgical ablation of bone tissue by means of a new miniature cryoprobe – adaption of the method for an in-vitro and in-vivo application to bone Z Orthop Ihre Grenzgeb 2001 139 64 69 11253524 10.1055/s-2001-11872
Popken F Bertram C König DP Rütt J Land M Hackenbroch MH The cryosurgical ablation of bone tissue by means of a new miniture cryoprobe – adaption of the method for an in-vitro and in-vivo application to bone Arch Orthop Trauma Surg 2002 122 129 133 11927992 10.1007/s00402-001-0371-6
Popken F Land M Bosse M Erberich H Meschede P König DP Fischer JH Eysel P Cryosurgery in long bones by means of a new miniature cryoprobe: An experimental in vivo study of the cryosurgical temperature field in sheep Eur J Surg Oncol 2003 29 542 547 12875863 10.1016/S0748-7983(03)00069-6
Kerschbaumer F Weiser G Neuerer G Russe W Bauer R Cryolesions of Bone. An Experimental Study. Part I: Examinations in technique of controlled cryolesion in bone Arch Orthop Traumat Surg 1980 96 5 9 10.1007/BF01246134
Oeseburg HB Cryochirurgische behandeling van enkele beentumoren Proschrift 1977 Groningen
Schlag G Schliep HJ Dingeldein E Grieben A Ringsdorf W Does methylmethacrylate induce cardiovascular complications during alloarthroplastic surgery of the hip joint? Anaesthesist 1976 25 60 7 1083699
Kerschbaumer F Krösel P Schlag G Cryogenic surgery of the bone – examinations of hemodynamics and tissue Beitr Orthop u Traumatol 1982 29 134 139
Popovic AD Milovanovic B Neskovic AN Pavlovski K Putnikovic B Hadzagic I Detection of massive pulmonary embolism by transesophageal echocardiography Cardiology 1992 80 94 99 1611638
van der Wouw PA Bax M Images in clinical medicine. Massive pulmonary embolism N Engl J Med 1997 6, 336 416 9010148 10.1056/NEJM199702063360605
Myers RS Hammond WG Ketcham AS A method for cryosurgical investigation of mouse tumors Int Surg 1969 52 232 233 5804096
Fisher AD Williams DF Bradley PF The effect of cryosurgery on the strength of bone Br J Oral Surg 1978 15 215 222 272917 10.1016/0007-117X(78)90003-3
|
16045807
|
PMC1192810
|
CC BY
|
2021-01-04 16:37:34
|
no
|
Biomed Eng Online. 2005 Jul 27; 4:46
|
latin-1
|
Biomed Eng Online
| 2,005 |
10.1186/1475-925X-4-46
|
oa_comm
|
==== Front
Cancer Cell IntCancer Cell International1475-2867BioMed Central London 1475-2867-5-241605352610.1186/1475-2867-5-24Primary ResearchP53 and Beta-Catenin Activity during Estrogen treatment of Osteoblasts Chandar Nalini [email protected] Rasleen [email protected] Peter C [email protected] Kevin [email protected] Walter C [email protected] Department of Biochemistry, Chicago College of Osteopathic Medicine, Midwestern University, 555, 31st Street, Downers Grove, IL 60515, USA2 Department of Pharmacology, Chicago College of Osteopathic Medicine, Midwestern University, 555, 31st Street, Downers Grove, IL 60515, USA2005 29 7 2005 5 24 24 21 4 2005 29 7 2005 Copyright © 2005 Chandar et al; licensee BioMed Central Ltd.2005Chandar 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 was undertaken to examine the relationship between the tumor suppressor gene p53 and the nuclear signaling protein beta-catenin during bone differentiation. Cross talk between p53 and beta-catenin pathways has been demonstrated and is important during tumorigenesis and DNA damage, where deregulation of beta catenin activates p53. In this study, we used estrogen treatment of osteoblasts as a paradigm to study the relationship between the two proteins during osteoblast differentiation.
Results
We exposed osteoblast-like ROS17/2.8 cells to 17-beta estradiol (E2), in a short term assay, and studied the cellular distribution and expression of beta-catenin. We found beta-catenin to be up regulated several fold following E2 treatment. Levels of p53 and its functional activity mirrored the quantitative changes seen in beta-catenin. Alkaline phosphatase, an early marker of osteoblast differentiation, was increased in a manner similar to beta-catenin and p53. In order to determine if there was a direct relationship between alkaline phosphatase expression and beta-catenin, we used two different approaches. In the first approach, treatment with LiCl, which is known to activate beta-catenin, caused a several fold increase in alkaline phosphatase activity. In the second approach, transient transfection of wild type beta-catenin into osteoblasts increased alkaline phosphatase activity two fold over basal levels, showing that beta catenin expression can directly affect alkaline phosphatase expression. However increase in beta catenin activity was not associated with an increase in its signaling activity through TCF/LEF mediated transcription. Immunofluorescence analyses of p53 and beta-catenin localization showed that E2 first caused an increase in cytosolic beta-catenin followed by the accumulation of beta-catenin in the nucleus. Nuclear p53 localization was detected in several cells.
Expression of p53 was accompanied by distribution of beta-catenin to the cytoplasm and cell borders. A sub population of cells staining strongly for both proteins appeared to be apoptotic.
Conclusion
These results suggest that interactions between p53 and beta-catenin signaling pathways may play a key role in osteoblast differentiation and maintenance of tissue homeostasis.
==== Body
Introduction
The organization of cells in tissues and organs is controlled by molecular control mechanisms that allow cells to interact with their neighboring cells and the extra cellular matrix. Cell-cell recognition and adhesion are critical processes in development, differentiation and the maintenance of tissue architecture. The cadherins family of Ca2+-dependent cells and their associated molecules such as beta-catenin are major components of the cellular adhesion machinery and play central roles in these various processes [1]. The cadherins are trans-membrane proteins that mediate Ca2+ dependent cell-cell adhesion. Beta catenin is a multifunctional protein which associates with the intracellular domain of cadherins. In addition to providing a physical link between cells, these adherent junctional proteins influence various signaling pathways. Beta-catenin is an important component of the Wnt / Wingless signaling pathway and can act as a transcription factor in the nucleus by serving as a co activator of the lymphoid enhancer factor (LEF)/TCF family of DNA-binding proteins [2].
The p53 tumor suppressor gene acts as a guardian of the genome and a loss of its function is seen in a wider variety of cancers [3]. P53 acts by sensing DNA damage and directing the cell to arrest or undergo apoptosis [1,3]. In this way, p53 is thought to prevent the excessive accumulation of mutations that could give rise to malignancies. However, p53 activities may not be limited to tumor suppressor functions. Accumulating evidence suggests that p53 function may be critical during differentiation of various tissues and organs [4-7]. Defects in p53-null embryos have been reported, suggesting that p53 may have a role in tissue organization during development [8,9]. We have, in previous studies, demonstrated a role for p53 in osteoblast differentiation and expression of the bone specific protein osteocalcin [10]. In studies with p53 null and heterozygous mice, we have also shown that a decrease in p53 expression interferes with the ability of osteoblasts to express osteocalcin [11]. During in vitro osteoblast differentiation, proliferation is followed by matrix deposition and mineralization. Alkaline phosphatase is generally seen as an early marker of osteoblast differentiation, while osteocalcin is considered a late marker. In our studies with estrogen, we have shown p53 to be up regulated and its activity to be associated with cell cycle arrest and expression of osteoblast differentiation markers rather than apoptosis [12,13].
Cross talk between p53 and beta-catenin pathways has been demonstrated and appears to be especially important during tumorigenesis and DNA damage, where deregulation of beta catenin is known to activate p53 [14,15]. Because of the importance of the cadherins and beta-catenin in tissue differentiation, we wanted to determine if this type of cross talk with p53 exists in osteoblasts under physiological conditions. We observed expression of several apoptosis-related and cell cycle arrest proteins during short term treatment of bone cells with estrogen [13]. Expression of several caspases have been shown to be required for expression of bone markers during osteoblast differentiation [16]. Treatment with 17-beta estradiol did not result in any appreciable apoptotic cell death [12]. In studies reported here, we investigated if 17-beta estradiol could modulate the expression and subcellular distribution of beta catenin and how it might relate to p53 expression.
Results
17-Beta estradiol (E2) up regulates expression of beta-catenin in osteoblastic osteosarcoma cells
ROS17/2.8 cells stably expressing 13 copies of a p53 binding sequence (ROS-PG13CAT) fused to a chloramphenicol acetyl transferase (CAT) gene were used to study effects of estrogen on changes in endogenous p53 functional activity. Binding of endogenous p53 to the PG-13CAT sequence and subsequent activation of gene expression was studied by analyzing CAT activity as described in previous studies [17]. In all other aspects this cell line is representative of ROS 17/2.8 cells an osteoblastic osteosarcoma line that is used extensively to study osteoblast differentiation. These cells were treated with E2 for different lengths of time as described under Methods and the resultant protein was separated on SDS PAGE and analyzed by western blotting. As may be seen in Figure 1A, an increase in beta-catenin expression occurred within 6 h of treatment and peaked at 16 h of E2 treatment followed by a drop and a second peak during 48 h after E2 treatment. The first increase was less dramatic than the second increase in beta-catenin (10 vs 30 fold increase).
Figure 1 (A) 17-beta-estradiol treatment increases the amount of beta-catenin in ROS-PG13 cells: Osteoblasts were treated with 10-11 M E2 for the times indicated and harvested for protein. The resulting protein lysates were subjected to western blot analysis of beta-catenin protein and beta-tubulin to monitor equal loading. The intensity of the bands was analyzed using a phosphorimager and results were normalized to beta-tubulin (loading control). A representative blot is shown. (B) Increase in p53 transactivating activity mirrors changes in beta-catenin. Equal amounts of E2 treated protein lysates described above were also subjected to a CAT assay to measure functional activity of endogenous p53 as described under methods. The resulting activity was plotted as fold change when compared to zero time (no treatment). The results represent mean ± SEM of 3 independent experiments in triplicates. (C) Alkaline phosphatase activity during E2 treatment: Enzyme activity was measured using a colorimetric assay as described under methods using the protein lysate described above. Values represent fold change when compared to control. Values represent mean ± SEM with an n of 3.
P53 functional activity parallels changes in beta-catenin expression during E2 treatment
P53 function was monitored by measuring CAT activity in ROS-PG-13 cells. As may be seen in Figure 1B, p53 transcription activating activity was increased about 4-fold 16 h after E2 treatment followed by a drop and an increase corresponding to the change seen in beta-catenin at 48 h interval (about 17-fold). P53 expression is known to accompany beta-catenin activation and is also thought to be critical in the regulation of beta catenin function [15]. P53 expression was also measured by western blot analysis and was found to be high after 16 h and remained high until 48 h of E2 treatment (not shown).
Figure 2 LiCl treatment of ROS-PG13 cells leads to activation of alkaline phosphatase expression. Protein lysates were prepared from cells treated with 10 mM LiCl or 10 mM NaCl and harvested after the times indicated. Equal amounts of protein extracts were used to measure alkaline phosphatase (AP) activity using p-nitro phenol phosphate as substrate at 420 nm and the relative activity of LiCl treated samples to NaCl treated controls is shown. The experiment represents mean ± SEM. n = 3 per treatment group.
Alkaline Phosphatase, an early marker of bone differentiation is increased during treatment with 17-B-estradiol
Alkaline phosphatase activity was measured during the same time intervals using a colorimetric assay. While alkaline phosphatase increased steadily with E2 treatment, the enzyme activity showed a clear spike during the 48 h interval (Figure 1C). While initial induction of alkaline phosphatase activity occurred with an increase in beta-catenin activity, the subsequent boost to its activity was seen during 48 h corresponding to the large increase in beta-catenin activity.
Figure 3 (A) Transient transfection of wild type beta-catenin into ROS-PG-13 cells increases alkaline phosphatase and p53 functional activity. Following transfection with the beta-catenin or control DNA, cells were lysed and equal amount of protein lysates were used to measure alkaline phosphatase activity. Enzyme activity was normalized to the mock transfected controls and reported as fold change. Data represents mean ± SEM of n = 4. *p < 0.05 versus control. (B) Transient transfection of wild type beta-catenin into ROS-PG-13 cells increases alkaline phosphatase and p53 functional activity. Lysates from the above experiment was also used to measure p53 activity by CAT assay. Equal amounts of protein were used and the resulting CAT activity is reported as a fold change over control levels. These experiments represent mean ± SEM of n = 4 ** p < 0.05 versus control.
Is there a direct relationship between beta-catenin expression and alkaline phosphatase activity?
In order to determine if an increase in beta-catenin nuclear signaling activity is associated with increased alkaline phosphatase activity, we used a LiCl treatment as a model for beta-catenin activation. Treatment with LiCl is known to inhibit GSK activity, which is critical for phosphorylation and inactivation of beta-catenin function [18]. Immunofluorescent staining for beta-catenin revealed a transient increase in beta-catenin expression in the nuclei of ROS-PG-13 in 24 h 10 mM LiCl treated cells but not in the control NaCl treated cells (not shown). Protein lysates from the cells similarly treated with either LiCl or NaCl were tested for alkaline phosphatase activity. As may be seen in Figure 2, LiCl treated cells showed an increase in alkaline phosphatase activity 24 h after treatment, compared to a less than 2-fold activation in the NaCl treated cells (Figure 2).
Figure 4 Beta-catenin expression during E2 treatment: Immunostaining of E2 treated ROS-PG13 with anti-beta-catenin antibody demonstrates its localization within the cells at 24 and 48 h after treatment. Cells were grown on cover slips and treated with 10-11 M 17-beta estradiol in 2% charcoal treated serum containing media for the different lengths of time indicated. Control cells were grown in 2% charcoal treated serum containing media. Other details are as described under methods.
Transient overexpression of wild type beta-catenin in ROS-PG13 cells increases alkaline phosphatase activity as well as p53 transcriptional activity
In order to determine if over-expression of beta-catenin produced similar effects on alkaline phosphatase, we transiently transfected a wild type beta-catenin plasmid into ROS-PG13 cells. Control cells were transfected with non-specific DNA. Alkaline phosphatase activity was measured in the control, mock-transfected and beta-catenin-transfected cells 24 h later. There was a small but statistically significant increase in alkaline phosphatase activity in beta catenin transfected cells when compared to cells that received non-specific DNA (Figure 3A). The same experiment was also repeated with a constitutively active beta-catenin (B-catenin S33Y) and similar results were obtained (not shown) suggesting that beta-catenin expression facilitates alkaline phosphatase expression in rat osteoblasts.
Figure 5 Increase in beta-catenin expression during E2 treatment does not result in beta-catenin signaling through TCF/LEF response elements in ROS-PG13 cells. Cells were transfected with TopFlash (wild type promoter) and FopFlash (mutant promoter) luciferase reporters in 2% media, and E2 treatment was started three hours later. E2 treatments were staggered during the 48 h interval and all cells were harvested after 48 h after the indicated time of exposure to the hormone. Mutant FopFlash activity was unchanged during the treatment and is not shown. LiCl treatment was carried out to demonstrate the validity of the assay (Inset). Cells were exposed to LiCl or NaCl 24 h after transfection for 16 h. In both these experiments luciferase activity was measured in cell lysates using equal amounts of protein. Experiments represent average ± SEM of triplicate measurements.
Protein lysates from the transiently transfected cells were subjected to CAT assay for determination of p53 functional activity during the same time period. P53 activity was 5 fold higher in cells transfected with wild type beta-catenin when compared to control cells (Figure 3B), showing that a parallel increase in p53 activity may not be limited to conditions of DNA damage but also occurs under physiological conditions.
Figure 6 Immunohistochemical staining of beta-catenin and p53. Cells were treated with 10-11 M 17-beta estradiol for 48 h as described under methods and stained for p53 (green) and beta-catenin (red) using specific antibodies. Panel A shows cells strongly staining for both proteins in the nucleus (broken arrow), strong p53 staining in the nucleus and beta-catenin at the plasma membrane (solid arrows) and strong staining of p53 in the nucleus with beta-catenin relegated to the cytoplasm (arrow head). B and C show apoptotic cells with strong staining of both proteins.
Subcellular distribution of beta-catenin during treatment
In order to determine the localization of beta-catenin during the treatment protocol, we conducted immunofluorescence analyses of estrogen treated cells (Figure 4). Cells were grown to confluency and switched to 2% charcoal treated media for 24 h before exposure to 17-beta estradiol. At the start of experiment (0 time), beta-catenin staining was only seen at the adherent junctions between cells and was undetectable intracellularly. 24 h after treatment with 17-beta estradiol, there was a dramatic increase in the amount of beta-catenin within the cells; most of the beta-catenin appeared to be in the cytoplasm and peri nuclear region. By 48 h strong staining for beta-catenin could be detected within the nucleus of a significant number of cells.
No change in beta-catenin transcriptional activity during E2 treatment
Since we observed nuclear staining of beta-catenin, experiments were carried out to determine if beta-catenin signaling through TCF/LEF family of transcriptional factors was activated. We transiently transfected the wild type TCF/LEF response elements (TOPFLASH) or the mutant sequence (FOPFLASH) followed by treatment with E2 treatment. No significant change in luciferase activity was noted during E2 treatment (Figure 5). The validity of the assay was checked using LiCL treatments (Figure 5 insert). These results indicate that endogenous beta-catenin signaling is not activated during E2 treatment even though the expression of beta-catenin was observed in the nuclei of treated cells.
p53 expression during 17-beta estradiol treatment
The patterns of p53 distribution were also monitored by immunostaining. Immunofluorescence staining for p53 also showed a heterogeneous pattern. P53 expression was high within the nucleus in a number of isolated cells. Among the cells that stained strongly for p53, some of them were apoptotic and counter staining with Hoescht reagent showed a pyknotic nucleus. In other cases strong staining was evident in nuclei that looked morphologically normal. P53's presence in the nucleus was also confirmed with western blots of cytoplasmic and nuclear fractions (not shown). Its presence in the nucleus correlated with its functional activity as measured by the CAT assay.
A better understanding of the relationship between the two proteins was evident when we stained simultaneously for both proteins and a representative field is shown in figure 6. Three types of association were evident. Strong staining of nuclear p53 was accompanied by beta-catenin in the cell borders (arrows). When both proteins were present in the nucleus, the cell was generally apoptotic (broken arrows). When intracellular staining for beta catenin was strong it was mostly contained in the cytoplasm when p53 decorated the nucleus (arrow head).
Discussion
In previous studies, we have shown the tumor suppressor gene p53 to be up regulated by estrogen and to be important for differentiative functions in bone [12,13]. In the studies reported here, we show that beta-catenin expression is increased during estrogen treatment of osteoblasts. This large increase in beta-catenin expression that we observed may be the result of either a direct increase in gene expression, or from stabilization of cytosolic beta-catenin. With regard to the latter possibility it is worth noting that in other cell types, estrogen has been shown to inhibit GSK activity which results in the stabilization of beta-catenin [18].
The association of beta catenin activation with increases in alkaline phosphatase expression is also very interesting, but not completely new. This association has been recently detected in several cell types where alkaline phosphatase plays a role in differentiated behavior of the cell [19-21]. Recent studies have implicated the wnt signaling pathway and beta-catenin in the regulation of alkaline phosphase expression in osteoblasts [21]. It appears that beta-catenin is able to increase alkaline phosphatase albeit indirectly, because no TCF binding sites have been detected within the alkaline phosphatase gene [22].
The role of p53 in the regulation of beta-catenin is best understood under conditions of DNA damage and tumorigenesis [15]. Stabilization of beta-catenin has been observed to cause stabilization of p53 through inhibition of its degradation [14,23,24]. While it is possible that beta-catenin results in the stabilization of p53, the resulting increase in p53 is not responsible for apoptosis, an activity that is regulated by p53 during DNA damage. Instead, under physiological conditions, p53 appears to monitor the environment such that an abnormal increase in beta catenin within the nucleus results in apoptosis, while in other cells the presence of p53 in the nucleus prevents the accumulation of beta-catenin. Beta catenin under these conditions appears to be relegated to the plasma membrane.
In the studies reported here we show treatment with 17-beta estradiol increases expression of beta-catenin and cause its migration in to the nucleus. Estrogen may mediate this effect by its action on GSK activity as seen in other tissues [25]. However, beta-catenin expression in the nucleus does not result in the activation of its signaling through TCF/LEF transcription factor binding sites. There are several likely reasons for this observation. As has been noted earlier, the level of signaling through the canonical pathway may be low and below detection limits using TCF/LEF reporter constructs [26]. It is also possible that beta-catenin may not directly act through the Wnt canonical pathway, but crosstalk with other pathways to generate a response. It has been shown that beta-catenin signaling does not function independently but synergizes with morphogens like BMP-2 to induce the early bone phenotypes in undifferentiated cells [21,22]. In a similar manner, estrogen treatment has been observed to enhance the binding of beta-catenin to estrogen receptors alpha and beta in human colon and breast cancer cells [27] and also participate in the transactivation of estrogen responsive genes. This suggests that beta-catenin may function as a common mediator of different bone specific agents to induce early bone phenotype. In this context it is interesting that beta-catenin and LEF1 repress expression of the osteocalcin gene, a late marker of the bone phenotype [28].
While the role of estrogen as bone-protective anabolic agent is well established, the mechanism of action is only now being understood at the molecular level [29,30]. Estrogen affects osteoblasts by non genotropic mechanisms that go to increase the life span of the osteoblasts by its action on plasma membrane signaling proteins [31]. Antiapoptotic mechanism by estrogen is transient in osteoblasts and it is not clear if p53 plays a role in this process. In a manner similar to estrogen receptors, p53 has been shown to bind beta-catenin resulting in its stabilization and transcriptional activation [23]. P53 is also able to inhibit expression of TCF-4 by directly binding to the promoter of the gene [32]. This type of regulation may be important to maintain cell-cell interactions and prevent apoptosis. These types of cross signaling may be relevant and important for osteoblast differentiation as opposed to osteoblast proliferation and may critically depend on the cellular environment. P53 is known to interact with a plethora of proteins [33] and these interactions may determine the final outcome for the cell. P53's ability to sense the environment allows for cell cycle arrest and differentiation under some circumstances and apoptosis in other instances. Expression of alkaline phosphatase a differentiation marker in bone may be facilitated by beta-catenin nuclear activity. However once alkaline phosphatase is increased, p53 activity may be critical to maintain the differentiated behavior of the cell by making sure beta-catenin is retained at cell borders rather than within the nucleus. Further studies are required to understand how the interactions between estrogen receptors, beta-catenin, p53 and related proteins facilitate the differentiation process.
Conclusion
Our data shows that beta-catenin activity is modulated during estrogen induced osteoblast differentiation and its increase is associated with an increase in p53 and alkaline phosphatase. The cellular localization of endogenous p53 and beta-catenin appears be mutually exclusive during estrogen treatment and reflects the role of p53 in regulating growth and differentiation.
Methods
Establishment of cell lines
The cell line ROS 17/2.8, a rat osteosarcoma cell line, was kindly provided by Dr. G. Rodan (Merck, Research Laboratory, West Point, PA). Cells were grown in minimal essential medium with α-F12 with 10% fetal bovine serum in a modified atmosphere of 95% air and 5% CO2 at 37°C. This cell line contains a wild type endogenous p53 [17] and can be induced to mineralize in culture and express genes associated with advanced stages of differentiation. The ROS17/2.8 cells were stably transfected with the plasmid PG-13-CAT (a kind gift of Dr. B. Vogelstein, Johns Hopkins University, Baltimore, MD). This plasmid encodes 13 copies of a p53 binding DNA sequence fused to a CAT reporter gene)[17]. In the present studies cells transfected with this plasmid (referred to as ROS-PG13) cells were used to monitor transcriptional activity of endogenous p53.
Cell Culture conditions & Treatment with 17β-Estradiol
Cells for E2 treatment were exposed to phenol red free media before and during treatment with E2. The water-soluble form, 17β-estradiol (Sigma, St. Louis, MO) was used at the concentration of 10-11 M. Cells used for E2 treatment were exposed to 2% charcoal-treated serum containing phenol red free media for 24 hours before treatment with E2. For experiments requiring E2 for longer than 24 hours, fresh media with E2 was maintained on cells. Unless otherwise mentioned, all experiments were done using E2 at a final concentration of 10-11 M. This concentration is based on results obtained with our previous studies, where we saw maximal induction of p53 at 10-11 M – 10-12 M [12]. Cells were treated for different lengths of time ranging from 0–72 h.
Transient Transfections
For beta-catenin transfections, we used HA-β-catenin (WT beta catenin) and S33Y β-catenin (a constitutively active mutant), a kind gift of Dr. Ben-Ze'ev, Weizmann Institute, Rehovot, Israel. Cells were transfected with Superfect (Qiagen) in 10-cm plates for 24–48 h followed by protein lysis. The total amount of DNA used was maintained equally in these experiments. Equal amount of protein was used for measurement of alkaline phosphatase and CAT activity.
Measurement of CAT Activity
CAT activity of ROS-PG13 cells after treatment was used as a measure of p53 DNA binding activity and reflected p53 function at any time point. Harvested cells were suspended in buffered saline and then in a 0.25 M Tris buffer pH 7.8, disrupted by 3 freeze-thaw cycles. The supernatants were collected after centrifugation and heated at 65°C for 10 minutes to inactivate cellular acetylase activity. Protein concentrations were measured with the Bradford method and equal amounts of protein were used in the assays. CAT activity was determined by means of liquid scintillation counting, and was measured over a linear range of chloramphenicol acetylation such that the fraction acetylated was proportional to actual activity. All measurements were carried out on triplicate samples. Other details are as described earlier.
Measurement of Luciferase Activity
For reporter assays, cells were transfected with the beta-catenin responsive firefly luciferase reporter plasmids TopFlash (wild type promoter) or FopFlash (Mutant promoter) (Upstate Biotechnologies) for 48 h. Three hours after transfection, cells received 17-beta estradiol to a concentration of 10–11 M for the times indicated. Cells were exposed to LiCl for 16 hours, lysed and equal amount of protein was used for measuring luciferase activity. All measurements were carried out on triplicate samples and experiments were repeated at least thrice.
Immunofluorescence staining
Beta-catenin and p53 were visualized by indirect immunocytochemistry using a rabbit anti beta catenin (Zymed Laboratories Inc., San Francisco, CA.) or a mouse anti-p53 (1C12) (Cell Signaling Technology, Beverly, MA.) as the primary antibodies. ROS-PG13 cells were plated on coverslips and treated with E2 as described above. Cells were fixed in ice cold methanol and permeabilized for ten minutes. Cells were then blocked with 10% goat serum for 10 minutes room temperature. Samples were incubated for 1 hour with primary antibody followed by a-30 minute incubation with a goat, anti-rabbit TRITC-conjugate or goat, anti-mouse FITC-conjugate. Cells were then viewed with a Nikon Eclipse 400 fluorescence microscope using 40× and 100× objectives. Digital images were captured with a Spot digital camera (Diagnostic Instruments, Sterling Heights, MI) using automated exposure times and gain settings for the bright-field images. Dark-field fluorescence images were captured using a gain setting of 16 and exposure times of 3 s for green and 1 s for red and blue. The digital images were processed using the Image-Pro Plus images analysis software package (Media Cybernetics, Silver Spring, MD).
Negative controls consisted of samples that were incubated without the primary antibodies. All labeling experiments were repeated at least three times and were highly reproducible.
Immuno Blotting
Protein lysates were prepared using M-PER Reagent (Pierce, Rockford, IL) combined with a protease inhibitor cocktail, Complete Mini (Roche, Mannheim, Germany). Twenty-five micrograms of each protein lysate was subjected to 10% SDS-PAGE, and transferred to immun-Blot PVDF membrane (Bio-Rad, Hercules, CA). Expression was determined using rabbit anti beta catenin (Cell Signaling Technology, MA) and HRP-goat anti rabbit conjugate (Zymed Laboratories Inc.CA). Membranes were then developed using enhanced chemiluminescence (Amersham International, Amersham, Bucks, UK).
Alkaline Phosphastase
Alkaline phosphatase activity was measured using a quantitative colorimetric assay with para-nitrophenol phosphate as substrate using a commercially available kit (Sigma Chemical Company, St. Louis, MO).
Statistical Analyses
The differences in the means of experimental results were analyzed for their statistical significance with the one-way ANOVA combined with a multiple comparison procedure (Tukey Kramer multiple comparisons test).
Acknowledgements
This work was supported, in part by Grant R15 CA098113 to NC from National Cancer Institute and in part by Grant R01 ES006478 (to W.C.P) from National Institute of Environmental Health Sciences and by funds from Midwestern University to NC. The authors gratefully thank Drs. Rodan, Vogelstein and Ben Ze'ev for their gifts of reagents and Vicki Sears for her help with the manuscript.
==== Refs
Gooding JM Yap KL Ikura M The cadherin-catenin complex as a focal point of cell adhesion and signalling: new insights from three-dimensional structures Bioessays 2004 26 497 511 15112230 10.1002/bies.20033
Westendorf JJ Kahler RA Schroeder TM Wnt signaling in osteoblasts and bone diseases Gene 2004 341 19 39 15474285 10.1016/j.gene.2004.06.044
Vogelstein B Kinzler KW Cancer genes and the pathways they control Nat Med 2004 10 789 799 15286780 10.1038/nm1087
Cordenonsi M Dupont S Maretto S Insinga A Imbriano C Piccolo S Links between tumor suppressors: p53 is required for TGF-beta gene responses by cooperating with Smads Cell 2003 113 301 314 12732139 10.1016/S0092-8674(03)00308-8
Peller S Frenkel J Lapidot T Kahn J Rahimi-Levene N Yona R Nissim L Goldfinger N Sherman DJ Rotter V The onset of p53-dependent apoptosis plays a role in terminal differentiation of human normoblasts Oncogene 2003 22 4648 4655 12879009 10.1038/sj.onc.1206541
Billon N Terrinoni A Jolicoeur C McCarthy A Richardson WD Melino G Raff M Roles for p53 and p73 during oligodendrocyte development Development 2004 131 1211 1220 14960496 10.1242/dev.01035
Matas D Milyavsky M Shats I Nissim L Goldfinger N Rotter V p53 is a regulator of macrophage differentiation Cell Death Differ 2004 11 458 467 14713961 10.1038/sj.cdd.4401379
Armstrong JF Kaufman MH Harrison DJ Clarke AR High-frequency developmental abnormalities in p53-deficient mice Curr Biol 1995 5 931 936 7583151 10.1016/S0960-9822(95)00183-7
Ohyama K Chung CH Chen E Gibson CW Misof K Fratzl P Shapiro IM p53 influences mice skeletal development J Craniofac Genet Dev Biol 1997 17 161 171 9493073
Chandar N Campbell P Novak J Smith M Dependence of induction of osteocalcin gene expression on the presence of wild-type p53 in a murine osteosarcoma cell line Mol Carcinog 1993 8 299 305 8280378
Chandar N Donehower L Lanciloti N Reduction in p53 gene dosage diminishes differentiation capacity of osteoblasts Anticancer Res 2000 20 2553 2559 10953326
Bovenkerk S Lanciloti N Chandar N Induction of p53 expression and function by estrogen in osteoblasts Calcif Tissue Int 2003 73 274 280 14667141 10.1007/s00223-002-1041-6
Chandar N Logan D Szajkovics A Harmston W Gene expression changes accompanying p53 activity during estrogen treatment of osteoblasts Life Sci 2004 75 2045 2055 15312749 10.1016/j.lfs.2004.03.028
Damalas A Kahan S Shtutman M Ben-Ze'ev A Oren M Deregulated beta-catenin induces a p53- and ARF-dependent growth arrest and cooperates with Ras in transformation Embo J 2001 20 4912 4922 11532955 10.1093/emboj/20.17.4912
Oren M Decision making by p53: life, death and cancer Cell Death Differ 2003 10 431 442 12719720 10.1038/sj.cdd.4401183
Mogi M Togari A Activation of caspases is required for osteoblastic differentiation J Biol Chem 2003 278 47477 47482 12954609 10.1074/jbc.M307055200
Schwartz KA Lanciloti NJ Moore MK Campione AL Chandar N p53 transactivity during in vitro osteoblast differentiation in a rat osteosarcoma cell line Mol Carcinog 1999 25 132 138 10365915 10.1002/(SICI)1098-2744(199906)25:2<132::AID-MC8>3.0.CO;2-2
Cardona-Gomez P Perez M Avila J Garcia-Segura LM Wandosell F Estradiol inhibits GSK3 and regulates interaction of estrogen receptors, GSK3, and beta-catenin in the hippocampus Mol Cell Neurosci 2004 25 363 373 15033165 10.1016/j.mcn.2003.10.008
Vincan E Leet CS Reyes NI Dilley RJ Thomas RJ Phillips WA Sodium butyrate-induced differentiation of human LIM2537 colon cancer cells decreases GSK-3beta activity and increases levels of both membrane-bound and Apc/axin/GSK-3beta complex-associated pools of beta-catenin Oncol Res 2000 12 193 201 11341469
Mariadason JM Bordonaro M Aslam F Shi L Kuraguchi M Velcich A Augenlicht LH Down-regulation of beta-catenin TCF signaling is linked to colonic epithelial cell differentiation Cancer Res 2001 61 3465 3471 11309309
Rawadi G Vayssiere B Dunn F Baron R Roman-Roman S BMP-2 controls alkaline phosphatase expression and osteoblast mineralization by a Wnt autocrine loop J Bone Miner Res 2003 18 1842 1853 14584895
Bain G Muller T Wang X Papkoff J Activated beta-catenin induces osteoblast differentiation of C3H10T1/2 cells and participates in BMP2 mediated signal transduction Biochem Biophys Res Commun 2003 301 84 91 12535644 10.1016/S0006-291X(02)02951-0
Damalas A Ben-Ze'ev A Simcha I Shtutman M Leal JF Zhurinsky J Geiger B Oren M Excess beta-catenin promotes accumulation of transcriptionally active p53 Embo J 1999 18 3054 3063 10357817 10.1093/emboj/18.11.3054
Sadot E Geiger B Oren M Ben-Ze'ev A Down-regulation of beta-catenin by activated p53 Mol Cell Biol 2001 21 6768 6781 11564862 10.1128/MCB.21.20.6768-6781.2001
Goodenough S Schafer M Behl C Estrogen-induced cell signalling in a cellular model of Alzheimer's disease J Steroid Biochem Mol Biol 2003 84 301 305 12711016 10.1016/S0960-0760(03)00043-8
Hu H Hilton MJ Tu X Yu K Ornitz DM Long F Sequential roles of Hedgehog and Wnt signaling in osteoblast development Development 2005 132 49 60 15576404 10.1242/dev.01564
Kouzmenko AP Takeyama K Ito S Furutani T Sawatsubashi S Maki A Suzuki E Kawasaki Y Akiyama T Tabata T Kato S Wnt/beta-catenin and estrogen signaling converge in vivo J Biol Chem 2004 279 40255 40258 15304487 10.1074/jbc.C400331200
Kahler RA Westendorf JJ Lymphoid enhancer factor-1 and beta-catenin inhibit Runx2-dependent transcriptional activation of the osteocalcin promoter J Biol Chem 2003 278 11937 11944 12551949 10.1074/jbc.M211443200
Manolagas SC Kousteni S Jilka RL Sex steroids and bone Recent Prog Horm Res 2002 57 385 409 12017554 10.1210/rp.57.1.385
Syed F Khosla S Mechanisms of sex steroid effects on bone Biochem Biophys Res Commun 2005 328 688 696 15694402 10.1016/j.bbrc.2004.11.097
Kousteni S Chen JR Bellido T Han L Ali AA O'Brien CA Plotkin L Fu Q Mancino AT Wen Y Vertino AM Powers CC Stewart SA Ebert R Parfitt AM Weinstein RS Jilka RL Manolagas SC Reversal of bone loss in mice by nongenotropic signaling of sex steroids Science 2002 298 843 846 12399595 10.1126/science.1074935
Rother K Johne C Spiesbach K Haugwitz U Tschop K Wasner M Klein-Hitpass L Moroy T Mossner J Engeland K Identification of Tcf-4 as a transcriptional target of p53 signalling Oncogene 2004 23 3376 3384 14990988 10.1038/sj.onc.1207464
Coutts AS La Thangue NB The p53 response: emerging levels of co-factor complexity Biochem Biophys Res Commun 2005 331 778 785 15865933 10.1016/j.bbrc.2005.03.150
|
16053526
|
PMC1192811
|
CC BY
|
2021-01-04 16:24:16
|
no
|
Cancer Cell Int. 2005 Jul 29; 5:24
|
utf-8
|
Cancer Cell Int
| 2,005 |
10.1186/1475-2867-5-24
|
oa_comm
|
==== Front
J CarcinogJournal of Carcinogenesis1477-3163BioMed Central London 1477-3163-4-111609114910.1186/1477-3163-4-11ResearchBystander effects in UV-induced genomic instability: Antioxidants inhibit delayed mutagenesis induced by ultraviolet A and B radiation Dahle Jostein [email protected] Egil [email protected] Trond [email protected] Department of Radiation Biology, The Norwegian Radium Hospital, Montebello, 0310 OSLO, Norway2005 9 8 2005 4 11 11 9 11 2004 9 8 2005 Copyright © 2005 Dahle et al; licensee BioMed Central Ltd.2005Dahle 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
Genomic instability is characteristic of many types of human cancer. Recently, we reported that ultraviolet radiation induced elevated mutation rates and chromosomal instability for many cell generations after ultraviolet irradiation. The increased mutation rates of unstable cells may allow them to accumulate aberrations that subsequently lead to cancer. Ultraviolet A radiation, which primarily acts by oxidative stress, and ultraviolet B radiation, which initially acts by absorption in DNA and direct damage to DNA, both produced genomically unstable cell clones. In this study, we have determined the effect of antioxidants on induction of delayed mutations by ultraviolet radiation. Delayed mutations are indicative of genomic instability.
Methods
Delayed mutations in the hypoxanthine phosphoribosyl transferase (hprt) gene were detected by incubating the cells in medium selectively killing hprt mutants for 8 days after irradiation, followed by a 5 day period in normal medium before determining mutation frequencies.
Results
The UVB-induced delayed hprt mutations were strongly inhibited by the antioxidants catalase, reduced glutathione and superoxide dismutase, while only reduced glutathione had a significant effect on UVA-induced delayed mutations. Treatment with antioxidants had only minor effects on early mutation frequenies, except that reduced glutathione decreased the UVB-induced early mutation frequency by 24 %. Incubation with reduced glutathione was shown to significantly increase the intracellular amount of reduced glutathione.
Conclusion
The strong effects of these antioxidants indicate that genomic instability, which is induced by the fundamentally different ultraviolet A and ultraviolet B radiation, is mediated by reactive oxygen species, including hydrogen peroxide and downstream products. However, cells take up neither catalase nor SOD, while incubation with glutathione resulted in increased intracellular levels of glutathione. Previously, we have shown that ultraviolet induced delayed mutations may be induced via a bystander effect and that this effect is 5-fold higher for UVB radiation than for UVA radiation. Therefore, we propose that the antioxidants inhibit an ultraviolet radiation-induced bystander effect and that the effect is transmitted via the medium and via an internal transfer between cells, like gap junctional intercellular communication, for UVB radiation and only by the latter mechanism for UVA radiation.
==== Body
Background
Genomic instability is a hallmark of multistage carcinogenesis [1]. The instability is likely to accelerate the mutation rate of unstable cells and give them growth advantage over normal cells. It is established that ionizing radiation causes genomic instability in mammalian cells in the form of a persistent increase in various types of genetic changes such as mutations, sister chromatid exchanges and chromosome aberrations that last for many cell generations after exposure [2]. The mechanisms of induction and maintenance of these persistent changes are largely unknown. However, it is likely that ionizing radiation-induced genomic instability is maintained for many cell generations by factors such as a persistent induction of DNA damage by increased levels of oxidative stress, mutations in DNA repair enzymes or radiation-induced up regulation of error-prone DNA repair mechanisms [3].
Ionizing radiation causes only a minor fraction of human cancers, and the types and locations of cellular damage by ionizing radiation are different from those of other environmental stress factors or endogenous stress [4,5]. We have chosen to use induction of genomic instability by ultraviolet A (UVA radiation, 320 – 400 nm) and ultraviolet B radiation (UVB radiation, 290–320 nm) as our model system [6]. UVB radiation is an established complete carcinogen that is the main cause of non-melanoma skin cancers such as squamous cell carcinomas [7]. UVA radiation is suspected to play a role in induction of genomically unstable malignant melanoma [8,9]. Both UVA and UVB radiation induce genomic instability in mammalian cells in the form of delayed gene mutations that occur 10–20 cell generations after exposure [6]. The unstable cell clones with delayed mutations were also chromosomally unstable.
UVB radiation acts initially by absorption in DNA and production of direct DNA damage in the form of pyrimidine dimers and other photoproducts [10]. In contrast, UVA radiation acts primarily in the presence of oxygen and endogenous photosensitizers by generating reactive oxygen species (ROS) in exposed cells [11]. Thus, UVA radiation induces a type of oxidative stress that resembles endogenous oxidative stress by inflammation or by other endogenous sources [12]. However, UVB radiation subsequently also induces oxidative stress to some degree, especially in the form of hydrogen peroxide and lipid peroxides [13,14], and UVA radiation also induce pyrimidine dimers [15]. Nevertheless, the spectra and time courses of UVA-induced ROS and DNA damage are very different from those of UVB [16].
ROS are involved in many aspects of biology, such as atherosclerosis, aging, mutagenesis and carcinogenesis [17]. Increased levels of ROS have been found in genomically unstable cell clones induced by ionizing radiation [18], and the antioxidant catalase has been shown to inhibit UVA-induced delayed formation of micronuclei [19]. Antioxidants reduce oxidation of biomolecules by ROS and other oxidants and may be a factor in reduction of the incidence of many types of human cancer, including skin cancer, by high consumption of tea, fruits and vegetables [20-22].
Reduced glutathione (GSH), catalase, and superoxide dismutase (SOD) are water-soluble antioxidants [23]. GSH reduces a broad spectrum of oxidants, whereas catalase and SOD specifically reacts with H2O2 and superoxide, respectively [23]. Catalase and SOD are not taken up by cells, but extracellular GSH can be broken down by the plasma membrane enzyme γ-glutamyltransferase (GGT), which removes the γ-glutamyl moiety, and the resulting products can be taken up by cells and stimulate intracellular GSH synthesis [31]. H2O2 is a weak oxidant and poorly reactive. However, it can be converted to the highly reactive hydroxyl radical via the iron catalysed Fenton reaction. H2O2might be produced by a number of different non-enzymatic or enzymatic (SOD, oxidases, peroxidases and hydrogenases) processes [23]. UV radiation can activate flavin-containing NAD(P)H oxidases which subsequently lead to production of H2O2 [24].
Recently it was shown that UV-induced delayed mutations might be induced via a bystander effect [25]. The bystander effect was 5-fold higher for UVB radiation than for UVA radiation. The factors mediating the bystander effect are unknown but reactive oxygen species are likely candidates. We hypothesize that antioxidants added after UVA and UVB radiation are potent inhibitors of genomic instability manifested as delayed mutations.
Materials and methods
Cell line and culture conditions
Chinese hamster fibroblasts (V79 cells) were cultured in RPMI 1640 medium (PAA, Austria) supplemented with 10 % FCS (PAA, Austria), penicillin (100 U/ml), streptomycin (100 μg/ml), L-glutamine (2 mM) and the following antioxidants (Sigma, Saint Louis, MO, USA): 1000 IU/ml catalase from a 20.000 IU/ml stock solution in PBS, 5 mM GSH from a stock solution of 0.1 M in PBS (pH adjusted to 7.4) or 500 IU/ml SOD from a stock solution of 7500 IU/ml in PBS.
UV- measurements
The irradiance from the lamps was routinely measured with a PMA 2200 photometer (Solar light Co, Philadelphia, PA, USA) before each treatment. The lamp spectra (Figure 1) and absolute irradiance from the lamps were measured by an irradiance-calibrated USB 2000 spectrometer (Avantes, Eerbeek, Holland) as described previously [6].
Figure 1 Lamp spectra. A) The UVA lamp has a peak at 375 nm. Total UVA- irradiance (320–400 nm) was 436 W/m2. B) The UVB lamp has a peak at 312 nm. Total UVB-irradiance (290–320 nm) was 23.3 W/m2.
UV-irradiation protocol
Cells were seeded in 25 cm2 flasks (Nalge Nunc, Naperville, IL, USA) the day before irradiation. The cells were irradiated from above in PBS. The UVA lamp contains a 3 kW Hg-Xe light bulb and filters to remove unwanted UVB radiation and visible light (Sellamed 3000 lamp, Sellamed, Gevelsberg, Germany). The irradiance from the UVA lamp was 436 W/m2. The flasks transmitted 392 W/m2. The UVB lamp contains a bank of six fluorescing tubes (TL-20W/01RS, Philips, Amsterdam, Holland). The irradiance from the UVB lamp was 23.3 W/m2. The flasks transmitted 18.6 W/m2.
Measurement of survival
Eight hundred cells were seeded 18 h before exposure with UVA or UVB radiation. Antioxidants were added immediately after irradiation. The cells were incubated for 6–7 days after irradiation in cell medium, washed with 9 mg/ml NaCl, fixed with absolute alcohol and stained with a saturated methylene blue solution. Colonies larger than 50 cells were counted manually.
Measurement of Early HPRT Mutations
Cells were irradiated with UVA or UVB radiation. Antioxidants were added immediately after irradiation. The cells were subcultured on the third and sixth day after irradiation. On the eight day after irradiation one million cells were seeded in 4–6 100 mm dishes in medium with 5 μg/ml 6-thioguanine (6TG) (Sigma, Saint Louis, MO, USA) and 200 cells in triplicate 60 mm dishes in medium without 6TG. Seven days after seeding the cells were washed, fixed and stained. Images were obtained of the dishes with a flatbed scanner and homemade software was used to count the colonies [26]. The mutation frequency, m, (number of mutants per million cells) was calculated by equation 1:
where M is the total number of mutants counted, C is the total number of cells seeded (usually 1 million) and c.e. is the cloning efficiency.
Measurement of Delayed HPRT Mutations
The procedure used to determine delayed mutations has been described before [25]. Briefly, however, Cells were irradiated with UVA radiation or UVB radiation. Immediately after irradiation the PBS was replaced by medium containing HAT (0.2 mM hypoxanthine, 0.4 μM aminopterin and 75 μM thymidine) (Sigma, Saint Louis, MO, USA). Aminopterin (Sigma, Saint Louis, MO, USA) blocks de novo synthesis of DNA precursors, killing cells that lack the purine salvage pathway (HPRT mutants) whereas the wild type cells survive [27]. Four days after irradiation the cells were subcultured and again seeded in medium with HAT. Four days later the cells were subcultured again, but now incubated in ordinary medium (without HAT). After five days in ordinary medium, one million cells were seeded in 3–6 100 mm dishes per dose in medium with 5 μg/ml 6TG and 200 cells in triplicate 60 mm dishes per dose in medium without 6TG. Antioxidants were added immediately after irradiation and used until seeding in medium with 6TG. Six days after seeding the cells were washed, fixed, stained and counted as described above. The mutation frequency was calculated by equation 1. It may be conceivable that the delayed HPRT mutants detected were early mutations that were resistant to aminopterin treatment. However, when 13 different UV-and X radiation-induced HPRT- clones were tested for growth in HAT medium, only four cells in 25 million cells plated formed colonies after 7 days of growth. Thus, the assay used was valid for detecting delayed mutations.
Flow cytometry
Cells were washed two times with PBS and incubated with 50 μM monochlorobimane (MCB) for 45 minutes as described previously [28]. Subsequently the cells were trypsinsed and stained with 1 μg/ml propidium iodide (PI). The samples were analyzed in a FACSDIVA flow cytometer (Becton Dickinson, USA) equipped with one argon (Spectra Physics, USA) and one krypton laser (Coherent, USA) tuned to 488 nm and UV, respectively. Forward scatter (FSC), side scatter (SSC) and PI fluorescence (>630 nm) were measured at the first focus (488 nm, 200 mW). The acquisition was triggered on the FSC signal, and the area of the PI signal was measured and used to gate away dead cells. MCB fluorescence (465–505 nm) was excited with 50 mW of UV light (351/356 nm) at the second laser intercept.
Statistics
Kruskal-Wallis one-way analysis of variance on ranks was used to test for differences among the treatment groups. Mann Whitney rank sum test, Dunn's method and student t-test were used for comparison of treated groups versus control groups. A significance level of 0.05 was used unless otherwise noticed.
Results
To investigate the effect of reactive oxygen species produced after UVA or UVB radiation on early and delayed hprt mutations as well as cell survival, we treated irradiated cells with three different antioxidants. The cells were cultured in medium with antioxidants from immediately after irradiation and until seeding in selective medium for determination of mutant frequencies and until fixation for determination of cell survival.
Survival of cells after UV radiation
High mutation frequencies were advantageous in order to get high sensitivity of the antioxidant treatment. Therefore, doses of 321 kJ/m2UVA-radiation and 8.1 kJ/m2 UVB-radiation were chosen in the present study. This choice is based on figure 2 of ref [25], which shows that the fraction of early hprt mutants increases with the dose up to 321 kJ/m2 for UVA radiation and up to 8.1 kJ/m2 for UVB radiation, above which the mutant fraction reaches a plateau. For delayed mutations, 321 kJ/m2 of UVA radiation and 11.3 kJ/m2 of UVB radiation results in maximum delayed mutant frequency (Figure 2 of ref [25]). These doses resulted in around 36 % cell survival for UVA radiation and 33 % cell survival for UVB radiation (Table 1). The doses were physiologically relevant as they correspond to midday sun exposure times in midsummer of 2–5 hours in Finland and half an hour in Italy [29,30]
Figure 2 Effect of antioxidants on UVA- and UVB-induced early mutagenesis. Cells were exposed to 321 kJ/m2 UVA- (A) or 8.1 kJ/m2 UVB radiation (B). The cells were grown with or without antioxidants in ordinary medium for 8 days after ultraviolet irradiation. Subsequently, the cells were seeded in medium containing 5 μg/ml 6-thioguanine and no antioxidants. 0: medium without antioxidants, CAT: catalase, GSH: glutathione, SOD: superoxide dismutase. Error bars = standard error from 3–10 independent experiments each with 3–4 parallel dishes. *: significantly different from unirradiated cells (Mann-Whitney rank sum test, p < 0.05). ◆: significantly different from cells not treated with antioxidants (Mann-Whitney rank sum test, p < 0.05). Mutant frequency = number of mutant colonies per million cells seeded.
Table 1 Clonogenic survival (%) of cells incubated with antioxidants after exposure to 321 kJ/m2 UVA- or 8.1 kJ/m2 UVB-radiation
Antioxidant Cells seeded before radiationa Cells seeded 13 days after radiationa
UVA UVB UVA UVB
Noneb 36 ± 12 33 ± 10 85 ± 7 80 ± 5
Catalase 21 ± 6 30 ± 5 84 ± 5 127 ± 12d
GSHc 19 ± 8 41 ± 13 127 ± 15 163 ± 26d
SODc 28 ± 6 29 ± 4 135 ± 12 105 ± 9
aMean values ± standard error from 3–10 independent experiments.
bThe cloning efficiency of unirradiated cells treated with antioxidants for 6 days after seeding were not significantly different from untreated cells.
cThe cloning efficiency for unirradiated cells treated with GSH or SOD for 13 days before seeding was significantly lower than for unirradiated cells not treated with antioxidants (t-test, p < 0.05).
dSignificantly higher survival than irradiated cells not treated with antioxidants (t-test, p < 0.05).
UV radiation decreased the clonogenic survival of the cells significantly (t-test, p < 0.01), but none of the antioxidants increased the survival of the irradiated V79 cells significantly (Table 1). There were also no significant differences between the sizes of the colonies of cells that had been treated with antioxidants or untreated (data not shown). Both irradiated and unirradiated cells divided more often than every 24 hours [6]. However, UV radiation resulted in significantly smaller colonies than the unirradiated cells when measuring survival shortly after radiation ("Cells seeded before radiation"-columns of Table 1), which may indicate that irradiated cells grew slower than the unirradiated for a while after radiation. The clonogenic survival of UVA- and UVB-treated cells was slightly decreased compared to control cells 13 days after radiation, but not significantly, suggesting only a limited amount of delayed cell death. There was apparently no effect of the HAT treatment on clonogenic survival since there was no significant difference between the unirradiated cells seeded before radiation and unirradiated HAT treated cells seeded 13 days after radiation. However, there were some long-term effects of the antioxidants. For UVB radiation, catalase and GSH significantly increased survival. For unirradiated cells GSH and SOD significantly decreased the cloning efficiency, which suggest that long-term incubation with GSH and SOD may be toxic.
Early UV-induced mutations
Early hprt mutations were measured after an expression period of 8 days in normal medium. Both 321 kJ/m2 of UVA radiation and 8.1 kJ/m2 of UVB radiation significantly increased the early mutant frequencies above control frequency (Figure 2). Treatment with antioxidants after irradiation had only minor effects on the early mutagenic effect of UVA and UVB radiation (Figure 2). However, GSH significantly reduced the early UVB-induced mutation frequency by 24 %, suggesting an oxidative component for early UVB-induced mutations. The antioxidants seemed to increase the background mutation frequency. However, the changes were not significant.
Delayed UV-induced mutations
Delayed hprt mutations were measured after 8 days in HAT medium to eliminate early mutations, followed by 5 days in normal medium. Both 321 kJ/m2 of UVA radiation and 8.1 kJ/m2 of UVB radiation significantly increased the delayed mutant frequencies above control frequency (Figure 3). Figure 3A shows that GSH significantly reduced the formation of delayed hprt mutations after UVA radiation by 74 % as compared with UVA radiation alone (p < 0.05). SOD and catalase had no significant effect on UVA induced delayed mutagenesis. Antioxidants had a greater effect on UVB-induced delayed mutations (Fig. 3B): catalase, GSH and SOD reduced the mutation frequency by 85 %, 94 % and 77 % as compared with UVB radiation alone, respectively (p < 0.05). Both catalase and GSH decompose H2O2. Thus, the results suggest that a persistent increase in H2O2 was involved in the formation of delayed mutations.
Figure 3 Antioxidants inhibit UVA- and UVB-induced delayed mutagenesis. Cells were exposed to 321 kJ/m2 UVA- (A) or 8.1 kJ/m2 UVB radiation (B). The cells were grown with or without antioxidants in HAT-medium for 8 days after ultraviolet irradiation to kill early HPRT mutants. Subsequently, the cells were subcultured and grown with antioxidants in ordinary medium for 5 days before seeding in medium containing 5 μg/ml 6-thioguanine and no antioxidants. 0: medium without antioxidants, CAT: catalase, GSH: glutathione, SOD: superoxide dismutase. Error bars = standard error from 3–10 independent experiments each with 3–4 parallel dishes. *: significantly different from unirradiated cells (Mann-Whitney rank sum test, p < 0.05). Mutant frequency = number of mutant colonies per million cells seeded.
Measurement of intracellular level of GSH
The fluorochrome monochlorobimane (MCB) was used to measure whether incubation with GSH could increase intracellular levels of GSH. Cells were treated with 5 mM GSH for 4 days, washed and stained with 50 μM MCB, which resulted in a significant increase in fluorescence. The fluorescence ratio between cells treated with GSH and untreated cells was 2.0 ± 0.4 (Mean ± SD).
Discussion
We here report that both UVA and UVB radiation induced delayed hprt mutations in V79 Chinese hamster fibroblasts. Mutations induced several cell generations after UV radiation are an indication of genomic instability. The UVB-induced delayed hprt mutations were strongly inhibited by adding the antioxidants catalase, superoxide dismutase and GSH after irradiation, while only GSH had a significant effect on UVA-induced delayed mutations. The strong effects of these antioxidants indicate that oxidative stress has a major role in maintaining increased mutation rates long after exposure to UV radiation. The increased mutation rates of unstable cells may allow them to accumulate mutations that subsequently lead to cancer.
Recently, we showed that a much higher degree of bystander effect was involved in UVB-induced than in UVA-induced delayed mutagenesis [25]. UV-induced delayed mutations were measured with the same method as in the present paper, which allows for a high degree of cell-cell contact, and with a cloning method in which delayed mutations were measured in individual cell clones [6]. UVA- and UVB-induced delayed mutagenesis was 4 and 19-fold higher, respectively, when using the present method as compared with the cloning method. The difference in degree of bystander-induced delayed mutations between UVA and UVB radiation might be explained by the present results. If it is assumed that the factors mediating the bystander-induced delayed mutations are long-lived reactive oxygen species like H2O2, then antioxidants should inhibit the bystander effect. Only GSH inhibited UVA-induced delayed mutagenesis and GSH, SOD and catalase inhibited UVB-induced delayed mutagenesis (Fig. 3). Cells probably take up neither GSH, SOD nor catalase. However, breakdown of extracellular GSH can be initiated by the plasma membrane enzyme γ-glutamyltransferase (GGT), which removes the γ-glutamyl moiety, and the resulting products can be taken up by cells and stimulate intracellular GSH synthesis [31]. This hypothesis is supported by a two-fold increase in monochlorobimane-fluorescence from cells incubated with 5 mM GSH for 4 days as compared with control cells. Thus, to inhibit UVA-induced delayed mutagenesis the intracellular level of antioxidants has to be increased and the bystander effect may be assumed to be an effect that is transferred between generations of cells or via gap junctions. In fact, inhibition of gap junctional intercellular communication by dieldrin significantly decreased the induction of delayed mutations [25]. On the other hand, catalase and SOD, which are only effective outside the cells, inhibited UVB-induced delayed mutagenesis, but not as effective as GSH. Therefore, much of the UVB-induced bystander effect is suggested to be an effect that is transferred to neighbouring cells via the medium. This may be a mechanism of bystander induced delayed mutations only relevant for UVB radiation and that comes in addition to the bystander effect between related cells or via gap junctions. In conclusion, induction of delayed mutations may be divided into three categories: 1) Directly induced delayed mutations due to genomic instability caused by mutations in gatekeeper or caretaker genes [32]; 2) Induced by a bystander effect transmitted from generation to generation of related cells or via gap junctions; and 3) Induced by a bystander effect transmitted via the medium.
Catalase decomposes H2O2 to water and oxygen while GSH reduces a broader spectrum of oxidants, including H2O2. GSH decomposes H2O2 in a reaction catalysed by glutathione peroxidase and reacts with H2O2 downstream products. H2O2 is a weak oxidant and is poorly reactive [23]. Nevertheless, hydrogen peroxide and its more reactive downstream products, such as lipid peroxides and hydroxyl radical, may act as clastogenic factors [33] and thus produce delayed chromosome aberrations and mutations. Therefore, inhibition of UVB-induced delayed mutagenesis by catalase and GSH may indicate that H2O2 was produced in mutagenic concentrations in the period 8 to 13 days after irradiation. This conclusion is consistent with those of previous reports showing that hydrogen peroxide may induce genomic instability [34,35]. UVB-induced delayed mutagenesis was also inhibited by SOD. It has been shown that superoxide, which is converted to hydrogen peroxide by SOD, may act as a clastogenic factor and induce chromosome damage [37]. Thus, superoxide may also be involved in UVB-induced delayed mutagenesis. Furthermore, it has been demonstrated that medium harvested from V79 cells 22 hours after ultraviolet C exposure contains factors that can increase the mutation frequency of cells [36]. The identification of such clastogenic factors is clearly a key challenge for radiation research.
Using dihydrorhodamine 123 that become fluorescent when oxidized by ROS, including H2O2, genomically unstable cell clones exhibited significantly elevated levels of oxidative stress as compared to V79 control cells [38]. In accordance, ionizing radiation induced genomically unstable clones also had significantly increased levels of oxidative stress [39]. Interestingly, the ionizing radiation induced genomically unstable clones had elevated numbers of dysfunctional mitochondria with increased ROS production as compared to normal cells. Thus, it is tempting to speculate that UV-induced genomic instability also occurs via damage to mitochondria and a subsequent increase in oxidative stress. However, we have not been able to detect UV-induced persistent increase in oxidative stress in the whole population of V79 cells (J. Dahle, unpublished results), only in genomically unstable clones. Thus, it may be that only a small fraction of the irradiated cells gets an increase in the production of ROS, and that these cells are the same cells that have an unstable genome and get delayed mutations.
The antioxidants catalase, GSH and SOD decreased the UVB-induced delayed mutation frequency significantly but had almost no effect on the early UVB induced mutation frequency, except for a slight but significant effect of GSH on UVB-induced early mutations. The reason for this discrepancy may be differences in the types of early and delayed UV-induced DNA damage. Pyrimidine dimers, the main UVB-induced DNA lesion, are caused by direct absorption of UVB photons in DNA [10]. Thus, this type of damage cannot be inhibited by antioxidants added after irradiation, indicating that early and delayed UVB-induced DNA damage are different. This hypothesis is supported by the finding that UVB-induced delayed mutations, but not early mutations, are accounted for by large deletions [40], which are similar to the mutations produced by oxidative stress [41,42].
Pyrimidine dimer formation, oxidation of guanine and deletions are potential origins of early mutations after UVA-radiation [15,40,43,44]. The mechanism of oxidation of guanine probably involves singlet oxygen produced by UVA-activated photosensitizers [43]. Pyrimidine dimer formation by UVA radiation was reported after very high doses and the mechanism of induction may involve DNA damage via photosensitization [45]. The UVA absorbing chromophores have not been identified. Thus, because of the short life time of photosensitized reactions, the early DNA damage probably cannot be inhibited using antioxidants added up to 5 minutes after irradiation. In a multiplex PCR study it was shown that the majority of both early and delayed UVA-induced hprt mutants exhibited either total gene deletion or deletion of some of the exons of the gene [40]. Sequencing of delayed mutations might reveal further insight into the mechanisms of UV-induced genomic instability.
Conclusion
In conclusion, UVB-induced delayed hprt mutations were strongly inhibited by antioxidants that eliminate H2O2 and superoxide anion, suggesting that a persistent increase of these reactive oxygen species or downstream products are involved in a bystander-induced formation of delayed mutations. This bystander effect may be mediated both via the medium and via gap junctional intercellular communication or transmitted from mother to daughter cells since CAT and SOD were not effective inside the cells, while GSH probably was effective both inside and outside the cells. The mechanism of UVA-induced delayed mutations may involve bystander-induced formation of delayed mutations mediated by gap junctional intercellular communication or transmitted from mother to daughter cells. UVA and UVB both induce delayed mutations – but are fundamentally different in how they interact with cells. Therefore, oxidative stress may also have a role in induction of delayed mutations induced by other types of stress than ultraviolet radiation.
Abbreviations
GSH – glutathione
hprt – hypoxantine phosphoribosyl transferase
MCB – monochlorobimane
ROS – reactive oxygen species
SOD – superoxide dismutase
UVA – ultraviolet A (320 – 400 nm)
UVB – ultraviolet B (290 – 320 nm)
6TG – 6-thioguanine
Authors' contributions
JD conceived the study, carried out the experimental work, performed the statistical analysis and wrote the manuscript. EK and TS participated in the design and coordination of the study and with writing of the manuscript. All authors read and approved the final manuscript.
Acknowledgements
We are grateful to Bjørn Høvik and Mali Strand Ellefsen for technical assistance, and to professor R. B. Setlow (Brookhaven National Laboratory) and professor Rune Blomhoff (Department of Nutrition, Faculty of Medicine, University of Oslo) for valuable comments. Supported by the Norwegian Research Council (JD).
==== Refs
Hanahan D Weinberg RA The hallmarks of cancer Cell 2000 100 57 70 10647931 10.1016/S0092-8674(00)81683-9
Limoli CL Kaplan MI Corcoran J Meyers M Boothman DA Morgan WF Chromosomal instability and its relationship to other end points of genomic instability Cancer Res 1997 57 5557 5563 9407967
Morgan WF Non-targeted and delayed effects of exposure to ionizing radiation: I. Radiation-induced genomic instability and bystander effects in vitro Radiat Res 2003 159 567 580 12710868
Khanna KK Jackson SP DNA double-strand breaks: signaling, repair and the cancer connection Nat Genet 2001 27 247 254 11242102 10.1038/85798
Kiefer J Schreiber A Gutermuth F Koch S Schmidt P Mutation induction by different types of radiation at the Hprt locus Mutat Res 1999 431 429 448 10636006
Dahle J Kvam E Induction of delayed mutations and chromosomal instability in fibroblasts after UVA-, UVB- and X-radiation Cancer Res 2003 63 1464 1469 12670891
De Gruijl FR Sterenborg HJ Forbes PD Davies RE Cole C Kelfkens G Wavelength dependence of skin cancer induction by ultraviolet irradiation of albino hairless mice Cancer Res 1993 53 53 60 8416751
Setlow RB Grist E Thompson K Woodhead AD Wavelengths effective in induction of malignant melanoma Proc Natl Acad Sci U S A 1993 90 6666 6670 8341684
Ley RD Ultraviolet radiation A-induced precursors of cutaneous melanoma in Monodelphis domestica Cancer Res 1997 57 3682 3684 9288772
Moan J Peak MJ Effects of UV radiation of cells J Photochem Photobiol B 1989 4 21 34 2509659 10.1016/1011-1344(89)80099-5
Arimoto-Kobayashi S Ando Y Horai Y Okamoto K Hayatsu H Lowe JE Mutation, DNA strand cleavage and nitric oxide formation caused by N- nitrosoproline with sunlight: a possible mechanism of UVA carcinogenicity Carcinogenesis 2002 23 1537 1540 12189198 10.1093/carcin/23.9.1537
Janssen YM Van Houten B Borm PJ Mossman BT Cell and tissue responses to oxidative damage Lab Invest 1993 69 261 274 8377469
Masaki H Atsumi T Sakurai H Detection of hydrogen peroxide and hydroxyl radicals in murine skin fibroblasts under UVB irradiation Biochem Biophys Res Commun 1995 206 474 479 7826364 10.1006/bbrc.1995.1067
Morliere P Moysan A Tirache I Action spectrum for UV-induced lipid peroxidation in cultured human skin fibroblasts Free Radic Biol Med 1995 19 365 371 7557551 10.1016/0891-5849(95)00043-W
Rochette PJ Therrien JP Drouin R Perdiz D Bastien N Drobetsky EA UVA-induced cyclobutane pyrimidine dimers form predominantly at thymine- thymine dipyrimidines and correlate with the mutation spectrum in rodent cells Nucleic Acids Res 2003 31 2786 2794 12771205 10.1093/nar/gkg402
De Gruijl FR Photocarcinogenesis: UVA vs UVB Methods Enzymol 2000 319 359 366 10907526
Davies KJ Oxidative stress: the paradox of aerobic life Biochem Soc Symp 1995 61 1 31 8660387
Limoli CL Hartmann A Shephard L Yang CR Boothman DA Bartholomew J Apoptosis, reproductive failure, and oxidative stress in Chinese hamster ovary cells with compromised genomic integrity Cancer Res 1998 58 3712 3718 9721883
Phillipson RP Tobi SE Morris JA McMillan TJ UV-A induces persistent genomic instability in human keratinocytes through an oxidative stress mechanism Free Radic Biol Med 2002 32 474 480 11864787 10.1016/S0891-5849(01)00829-2
Kris-Etherton PM Hecker KD Bonanome A Coval SM Binkoski AE Hilpert KF Bioactive compounds in foods: their role in the prevention of cardiovascular disease and cancer Am J Med 2002 113 71S 88S 12566142 10.1016/S0002-9343(01)00995-0
Longtin R Selenium for prevention: eating your way to better DNA repair? J Natl Cancer Inst 2003 95 98 100 12529339
Kazi A Smith DM Daniel K Zhong S Gupta P Bosley ME Potential molecular targets of tea polyphenols in human tumor cells: significance in cancer prevention In Vivo 2002 16 397 403 12494882
Halliwell B Gutteridge JM Free Radicals in Biology and Medicine 1999 3 New York: Oxford University Press
Hockberger PE Skimina TA Centonze VE Lavin C Chu S Dadras S Activation of flavin-containing oxidases underlies light-induced production of of H2O2 in mammalian cells Proc Natl Acad Sci U S A 1999 96 6255 6260 10339574 10.1073/pnas.96.11.6255
Dahle J Stokke T Kaalhus O Kvam E Bystander effects may modulate ultraviolet A and B induced delayed mutagenesis Radiation Res 2005 163 289 295 15733036
Dahle J Kakar M Kaalhus O Steen HB Automated counting of mammalian cell colonies by means of a flat bed scanner and image processing Cytometry Part A 2004 60A 182 188 10.1002/cyto.a.20038
Fujimoto WY Subak-Sharpe JH Seegmiller JE Hypoxanthine-guanine phosphoribosyltransferase deficiency: chemical agents selective for mutant or normal cultured fibroblasts in mixed and heterozygote cultures Proc Natl Acad Sci U S A 1971 68 1516 1519 5283941
Nair S Singh SV Krishan A Flow cytometric monitoring of glutathione content and anthracycline retention in tumor cells Cytometry 1991 12 336 342 1648468 10.1002/cyto.990120408
Kolari PJ Lauharanta J Hoikkala M Midsummer solar UV-radiation in Finland compared with the UV-radiation from phototherapeutic devices measured by different techniques Photodermatol 1986 3 340 345 3588354
Meloni D Casale GR Siani AM Palmieri S Cappellani F Solar UV dose patterns in Italy Photochem Photobiol 2000 71 681 690 10857363 10.1562/0031-8655(2000)071<0681:SUDPII>2.0.CO;2
Shi M Gozal E Choy HA Forman HJ Extracellular glutathione and gamma-glutamyl transpeptidase prevent H2O2-induced injury by 2,3-dimethoxy-1,4-naphthoquinone Free Radic Biol Med 1993 15 57 67 8103030 10.1016/0891-5849(93)90125-E
Michor F Iwasa Y Nowak MA Dynamics of cancer progression Nat Rev Cancer 2004 4 197 205 14993901 10.1038/nrc1295
Emerit I Reactive oxygen species, chromosome mutation, and cancer: possible role of clastogenic factors in carcinogenesis Free Radic Biol Med 1994 16 99 109 8300000 10.1016/0891-5849(94)90246-1
Li CY Little JB Hu K Zhang W Zhang L Dewhirst MW Persistent genetic instability in cancer cells induced by non-DNA-damaging stress exposures Cancer Res 2001 61 428 432 11212225
Yamamura E Nunoshiba T Nohmi T Yamamoto K Hydrogen peroxide-induced microsatellite instability in the Escherichia coli K-12 endogenous tonB gene Biochem Biophys Res Commun 2003 306 570 576 12804603 10.1016/S0006-291X(03)01027-1
Ghosh R Bhaumik G Supernatant medium from UV-irradiated cells influences the cytotoxicity and mutagenicity of V79 cells Mutat Res 1995 335 129 135 7477043
Emerit I Garban F Vassy J Levy A Filipe P Freitas J Superoxide-mediated clastogenesis and anticlastogenic effects of exogenous superoxide dismutase Proc Natl Acad Sci U S A 1996 93 12799 12804 8917499 10.1073/pnas.93.23.12799
Dahle J Kvam E Increased level of oxidative stress in genomically unstable cell clones J Photochem Photobiol B:Biol 2004 74 23 28 10.1016/j.jphotobiol.2004.01.004
Limoli CL Giedzinski E Morgan WF Swarts SG Jones GD Hyun W Persistent oxidative stress in chromosomally unstable cells Cancer Res 2003 63 3107 3111 12810636
Dahle J Noordhuis P Stokke T Svendsrud DH Kvam E Multiplex PCR analysis of ultraviolet A and B induced delayed and early mutations in V79 Chinese hamster cells Photochem Photobiol 2005 81 114 119 15453821 10.1562/2004-05-19-RA-174.1
Oller AR Thilly WG Mutational spectra in human B-cells. Spontaneous, oxygen and hydrogen peroxide-induced mutations at the hprt gene J Mol Biol 1992 228 813 826 1469715 10.1016/0022-2836(92)90866-I
Turner DR Dreimanis M Holt D Firgaira FA Morley AA Mitotic recombination is an important mutational event following oxidative damage Mutat Res 2003 522 21 26 12517408
Kvam E Tyrrell RM Induction of oxidative DNA base damage in human skin cells by UV and near visible radiation Carcinogenesis 1997 18 2379 2384 9450485 10.1093/carcin/18.12.2379
Besaratinia A Synold TW Xi B Pfeifer GP G-to-T transversions and small tandem base deletions are the hallmark of mutations induced by ultraviolet a radiation in mammalian cells Biochemistry 2004 43 8169 8177 15209513 10.1021/bi049761v
Douki T Reynaud-Angelin A Cadet J Sage E Bipyrimidine photoproducts rather than oxidative lesions are the main type of DNA damage involved in the genotoxic effect of solar UVA radiation Biochemistry 2003 42 9221 9226 12885257 10.1021/bi034593c
|
16091149
|
PMC1192812
|
CC BY
|
2021-01-04 16:39:19
|
no
|
J Carcinog. 2005 Aug 9; 4:11
|
utf-8
|
J Carcinog
| 2,005 |
10.1186/1477-3163-4-11
|
oa_comm
|
==== Front
Cardiovasc DiabetolCardiovascular Diabetology1475-2840BioMed Central London 1475-2840-4-121608684410.1186/1475-2840-4-12Original InvestigationEndothelial cell injury by high glucose and heparanase is prevented by insulin, heparin and basic fibroblast growth factor Han Juying [email protected] Anil K [email protected] Linda M [email protected] Department of Veterinary Biomedical Sciences, University of Saskatchewan, Saskatoon, Saskatchewan, S7N 5B4, Canada2 Department of Medicine, University of Florida, Jacksonville, Florida, 32086, USA2005 9 8 2005 4 12 12 23 6 2005 9 8 2005 Copyright © 2005 Han et al; licensee BioMed Central Ltd.2005Han 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
Uncontrolled hyperglycemia is the main risk factor in the development of diabetic vascular complications. The endothelial cells are the first cells targeted by hyperglycemia. The mechanism of endothelial injury by high glucose is still poorly understood. Heparanase production, induced by hyperglycemia, and subsequent degradation of heparan sulfate may contribute to endothelial injury. Little is known about endothelial injury by heparanase and possible means of preventing this injury.
Objectives
To determine if high glucose as well as heparanase cause endothelial cell injury and if insulin, heparin and bFGF protect cells from this injury.
Methods
Cultured porcine aortic endothelial cells were treated with high glucose (30 mM) and/or insulin (1 U/ml) and/or heparin (0.5 μg/ml) and /or basic fibroblast growth factor (bFGF) (1 ng/ml) for seven days. Cells were also treated with heparinase I (0.3 U/ml, the in vitro surrogate heparanase), plus insulin, heparin and bFGF for two days in serum free medium. Endothelial cell injury was evaluated by determining the number of live cells per culture and lactate dehydrogenase (LDH) release into medium expressed as percentage of control.
Results
A significant decrease in live cell number and increase in LDH release was found in endothelial cells treated with high glucose or heparinase I. Insulin and/or heparin and/or bFGF prevented these changes and thus protected cells from injury by high glucose or heparinase I. The protective ability of heparin and bFGF alone or in combination was more evident in cells damaged with heparinase I than high glucose.
Conclusion
Endothelial cells injured by high glucose or heparinase I are protected by a combination of insulin, heparin and bFGF, although protection by heparin and/or bFGF was variable.
==== Body
Background
Diabetes mellitus is characterized by hyperglycemia and vascular complications including microangiopathy and macroangiopathy [1,2]. The hallmarks of diabetic microangiopathy are retinopathy and nephropathy leading to blindness and renal failure respectively [3,4]. Macroangiopathy in diabetes, includes coronary artery disease, peripheral vascular disease, and cerebrovascular disease, and results from an acceleration of atherosclerosis and increased thrombosis thus increasing the risk of myocardial infarction, stroke and ischaemia [5,6]. It is reported that under better glycemic control fewer patients develop eye and/or renal complications [7].
Since the initial injury by hyperglycemia occurs in the blood vessel, endothelial cells (ECs) are considered to be the first target. Heparan sulfate proteoglycans (HSPGs), an important EC component, are synthesized by ECs and incorporated into the plasma membrane and extracellular matrix (ECM) [8,9]. In the ECM, HSPGs interact with fibronectin, laminin, collagen and growth factors such as basic fibroblast growth factor (bFGF) and help maintain vascular integrity [10,11]. HSPGs with their negative charged sulfate and carboxylate residues, create a "charge barrier", which decrease the permeability of anionic plasma proteins [12]. Thus degradation of HSPGs could lead to an increase in vascular permeability, decrease in vascular integrity, and changes in growth factor activity. Depletion of heparan sulfate (HS) and/or abnormal glycosaminoglycan (GAG) metabolism appears to be a pivotal mechanism associated with diabetic EC injury. HSPG or HS were decreased in the glomerular basement membrane (GBM) of patients with overt diabetic nephropathy which correlated with the degree of proteinuria [13,14]. A similar decrease in HS content was observed in the aortic intima of diabetic patients [15]. Skin basement membrane thickness was significantly reduced in patients with diabetic nephropathy compared to those without nephropathy. As well, HSPG synthesis was decreased in aorta, liver and intestinal epithelium of diabetic rats [16-18]. Thus in the diabetic condition, changes in HS metabolism may occur in any tissue suggesting the link between HS abnormalities and vascular complications in both large and small vessels.
Heparanase is an endo-β-D-glucuronidase that cleaves HS at specific interchain sites. Under normal physiological conditions, heparanase is expressed in platelets, cytotrophoblasts, mast cells, neutrophils, macrophages, and the placenta [19]. Heparanase activity was found in the urine of some diabetic patients and heparanase protein was expressed in both the glomerular mesangial and epithelial cell lysates, but not in intact cells [20]. HSPG degradation by heparanase upregulation may contribute to EC injury by hyperglycemia. Thus we wished to determine if heparanase as well as high glucose injured ECs.
Insulin and heparin alone, or in combination, prevented the intercellular gaps formed in ECs cultured in high glucose [21]. Several previous studies have shown that insulin increases nitric oxide (NO) production in cultured ECs and ensures normal vascular function [22,23]. Heparin can accumulate in ECs at a greater concentration than in plasma, increase HS on the EC surface, and prevent ECs from free radical injury [24-26]. Therefore, we postulated that insulin and/or heparin would protect ECs from high glucose or heparanase injury.
bFGF has a high affinity for heparin and HS which is required for interaction with its receptor (bFGFR). However, HS in the ECM also limits bFGF release into interstitial spaces [27,28]. The high affinity of bFGF for heparin and HS, together with the EC proliferation potential of bFGF [29], may protect ECs. Previous studies have shown that both heparanase and plasmin degrade HSPG and decrease the stability of the bFGF/HS/bFGFR complex resulting in loss of bFGF which was corrected by exogenous heparin [30,31]. Thus, the stabilization of bFGF/HS/bFGFR complex, by supplying heparin and bFGF, may protect ECs from injury by high glucose or heparanase.
Therefore, the purposes of our study were to determine if high glucose or heparanase induced cultured EC injury and if supplementation of insulin, heparin and bFGF would protect ECs from this damage.
Methods
Culture of Porcine Aortic Endothelial Cells (PAECs)
PAECs were cultured according to the method of Gotlieb and Spector [32]. Porcine aortic segments were rinsed in three changes of calcium- and magnesium-free Dulbecco's phosphate-buffered saline (CMF-DPBS), while the adventitial connective tissue and end pieces were trimmed. One end of the aorta was clamped with two hemostats (ensuring no leakage from bottom or branch points) and the lumen was rinsed three times with CMF-DPBS and then filled with collagenase solution (Type IV, SIGMA, St. Louis, MO, USA; 1 mg/ml in CMF-DPBS at 37°C). After 6 minutes the collagenase solution was removed and the endothelial surface was gently rinsed with M199 (GibcoBRL, Life Technologies, Inc., Grand Island, NY, USA) containing 5% fetal bovine serum (FBS, GibcoBRL), 50 μg/ml penicillin (SIGMA) and 10 μg/ml streptomycin (SIGMA). The medium was removed and plated onto 60 mm culture dishes. The volume was made up to 2 ml medium/dish. Cells were incubated at 37°C with 5% CO2 / 95% air in a humidified environment. ECs were identified by their morphological appearance of cobblestone-like flattened cells and the presence of von Willebrand Factor (vWF) in initial cultures. Non-endothelial-like cells, such as smooth muscle cells and fibroblasts, were destroyed by mechanical suction before the first passage.
To subculture, confluent 60 mm dishes of PAECs were washed twice with sterile CMF-DPBS, followed by exposure to a sterile 0.025% trypsin solution for two or three minutes at room temperature. The cells were then resuspended in 6 ml of culture medium and seeded onto three 60 mm dishes (2 ml/dish). Confluent cultures at passage 4, in 35 mm dishes, were used in experiments.
Reagents
Reagents were first prepared as stock solutions in CMF-DPBS at the following concentrations: glucose (D-Glucose, BDH Inc. Toronto, Canada) 3M, unfractionated bovine lung heparin (151 USP U/mg Upjohn Pharmaceuticals, Kalamazoo, MI, USA) 0.1 mg/ml, insulin (Humulin® N) 100 U/ml. Stock solutions of heparinase I (SIGMA) 10 and 1 U/ml and bFGF (SIGMA) 0.1 ng/μl were prepared in M199 without serum.
Cell Treatment
Cell medium was changed just prior to addition of reagents to 1 ml of medium per dish. For high glucose, 10 μl of 3 M stock solution was added to give a final added concentration of 30 mM. For heparin, 5 μl of 0.1 mg/ml of stock solution was added to give a final concentration 0.5 μg/ml. For insulin, 10 μl of 100 U/ml stock solution was added to give a final concentration of 1 U/ml. For bFGF, 10 μl of 0.1 ng/μl stock solution was added to give a final concentration of 1 ng/ml. Cell medium was changed and reagents were added fresh every other day for seven days. Cells were harvested 48 hours after the last addition.
In order to determine the culture conditions and damaging doses of heparinase I in PAECs, heparinase I was added to cultures at concentrations of 0.01, 0.05, 0.1, 0.3 and 0.5 U/ml in medium, which was produced by adding 10 μl of 1 U/ml and 5, 10, 30 and 50 μl of 10 U/ml of heparinase I to 1 ml medium respectively. Cells were cultured either in medium with serum for six or ten days, where cell medium was changed and fresh heparinase I was added every other day; or in serum free medium for two days by adding heparinase I once.
To determine the effect of heparin, insulin and bFGF in the presence heparinase I, cell medium was changed to M199 without serum. Then 30 μl of 10 U/ml of heparinase I was added to give a final concentration of 0.3 U/ml. Then heparin, insulin or bFGF was added to medium as described above. Cells were harvested two days later.
Assessment of Cell Injury
Trypan blue exclusion
Number of viable cells was determined by using trypan blue dye. The cells were washed with CMF-DPBS, detached from the culture dishes using 0.5 ml of 0.025% trypsin for 2–3 minutes and then were suspended in 1 ml of culture medium. An aliquot of 100 μl of the cell suspension was mixed with 10 μl of 0.4% of trypan blue solution (SIGMA) for 2–3 minutes. Cells were counted in a hemocytometer chamber with the light microscope. The number of live cells, those excluding trypan blue, and dead cells, those taking up trypan blue, were counted and calculated per dish. The number of live cells in experimental cultures was expressed as percent of live cells in control cultures.
Lactate dehydrogenase (LDH) assay
Cell medium (100 μl) from each culture was saved in a microcentrifuge tube for LDH determination using Sigma Diagnostic Kit No. 228-UV. A 50 μl sample was added to 500 μl of reagent, and LDH was quantified spectrophotometrically by the rate of change in absorbance at 340 nm at room temperature. The increased absorbance at 340 nm is the result of reduction of NAD+ to NADH as LDH catalyzes the conversion of lactate to pyruvate, thus the rate of NADH production is directly proportional to the LDH activity in the sample.
Statistical Analysis
All data was expressed as mean +/- standard error (SE) from three dishes/group. A one-way ANOVA was used to determine significant differences between groups. Values of P < 0.05 were considered to be statistically significant.
Results
Effect of Heparin and Insulin on PAECs Injured by High Glucose
PAECs were exposed to high glucose (30 mM) alone, insulin (1 U/ml) alone, heparin (0.5 μg/ml) alone and glucose plus heparin plus insulin for seven days. PAECs treated with high glucose showed a significant decrease in live cell number and increase in LDH release compared to control cells. Compared to control cells, there were significant changes in live cell number and LDH release in insulin alone treated cells, but not in heparin alone treated cells. Live cell number was significantly greater in heparin or insulin alone versus high glucose treated cultures. The combination of heparin and insulin in the presence of high glucose significantly increased live cell number and decreased LDH release compared to cells injured by high glucose alone (Figure 1).
Figure 1 PAECs Injured by High Glucose were Protected by a Combination of Heparin and Insulin. PAECs were exposed to glucose (30 mM), insulin (1 U/ml), heparin (0.5 μg/ml) and glucose plus heparin plus insulin for seven days. Cell medium was changed and fresh reagents were added every other day. Cells were counted and LDH release in medium was determined 48 hrs after the last addition of reagents. Results are expressed as mean +/- SE of three dishes per group. Significantly different than a, control; b, glucose; c, glucose+heparin+insulin (P < 0.01) (one-way ANOVA).
To determine if insulin and /or heparin protect PAECs from high glucose injury, PAECs were treated with high glucose (30 mM), glucose plus insulin (1 U/ml), glucose plus heparin (0.5 μg/ml), and glucose plus insulin plus heparin for seven days (Figure 2). A trend towards a decrease in live cell number and a significant increase in LDH release were seen in PAECs treated with high glucose compared to control cultures. A significant increase in live cell number and decrease in LDH release was seen in PAECs treated with high glucose and a combination of insulin and heparin compared to high glucose treatment alone, similar to results shown in Figure 1. A significant increase in live cell number and decrease in LDH release was seen when insulin was added to high glucose injured cells. High glucose plus heparin treated cultures showed a trend towards an increase in live cell number, and a significant decrease in LDH release compared to high glucose treatment alone (Figure 2).
Figure 2 The Protective Effect of Insulin and/or Heparin on PAECs Injured by High Glucose. PAECs were treated with glucose (30 mM), glucose plus insulin (1 U/ml), glucose plus heparin (0.5 μg/ml) and glucose plus insulin plus heparin for seven days. Cell medium was changed and fresh reagents were added every other day. Cells were counted and LDH release to medium was determined 48 hrs after the last addition of reagents. Results are expressed as mean +/- SE of three dishes per group. Significantly different than a, control; b, glucose; c, glucose + heparin; d, glucose + insulin (P < 0.01) (one-way ANOVA).
Effect of Insulin and/or Heparin on PAECs Injured by High Glucose in the Presence of bFGF
PAECs were treated with high glucose (30 mM), glucose plus bFGF (1 ng/ml), glucose plus bFGF plus insulin (1 U/ml), glucose plus bFGF plus heparin (0.5 μg/ml) and glucose plus bFGF plus insulin plus heparin for seven days. A significant decrease in live cell number and increase in LDH release was shown in high glucose treated versus control cells. When bFGF was present in cell medium, the combination of insulin and heparin had a protective effect on high glucose injured cells as shown by a significant increase in live cell number and decrease in LDH release in cells treated with glucose plus bFGF plus insulin plus heparin versus high glucose alone. In addition, a significant increase in live cell number and decrease in LDH release was shown in cells treated with high glucose plus insulin plus bFGF versus glucose alone. Heparin with bFGF or bFGF added to high glucose treated cells showed a significant increase in live cell number versus high glucose treatment alone. Although LDH release was less in high glucose plus heparin plus bFGF and high glucose plus bFGF versus the high glucose alone treated cells, this difference did not reach significance. The combination of insulin plus bFGF, and insulin plus heparin plus bFGF was more protective than bFGF or bFGF plus heparin on high glucose treated cultures, when numbers of live cells were considered. The combination of insulin plus heparin plus bFGF was more protective than bFGF plus heparin when LDH was considered (Figure 3).
Figure 3 Insulin and/or Heparin Protected PAECs from High Glucose Injury when bFGF was Present in Cell Medium. PAECs were treated with glucose (30 mM), glucose plus bFGF (1 ng/ml), glucose plus bFGF plus insulin (1 U/ml), glucose plus bFGF plus heparin (0.5 μg/ml) and glucose plus bFGF plus insulin plus heparin for seven days. Cell medium was changed and fresh reagents were added every other day. Cells were counted and medium LDH was determined 48 hrs after the last addition of reagents. Results are expressed as mean +/- SE of three dishes per group. Significantly different than a, control; b, glucose; c, glucose+bFGF; d, glucose+bFGF+heparin (P < 0.05) (one-way ANOVA).
Damaging Effect of Heparinase I on PAECs
PAECs were exposed to different doses of heparinase I (0.01, 0.05, 0.1 and 0.3 U/ml) for six or ten days in M199 with 5% serum. Cells were harvested 24 hours after the last addition of heparinase I. There were no significant differences in live cell number and LDH release in control cultures compared to those treated with different doses of heparinase I. PAECs exposed to heparinase I (0.05, 0.1, 0.3 and 0.5 U/ml) for 48 hours in serum free M199 showed a significant decrease in cell viability and increase in LDH release compared to the control group. Cell injury was dose dependent since there was a significant decrease in cell viability, with heparinase I 0.5 U/ml compared to 0.05 U/ml (data not shown). Doses of 0.3 U/ml heparinase I in serum free media conditions were chosen for the following experiments.
Effect of Insulin, Heparin and bFGF on Heparinase I Induced PAEC Injury
PAECs were treated with heparinase I (0.3 U/ml) and/or insulin (1 U/ml) and/or heparin (0.5 μg/ml) for 48 hours in serum free M199. Treatment with heparinase I showed a significant decrease in live cell number and increase in LDH release compared to control cells. Addition of insulin or heparin to heparinase I treated cells showed a significant increase in live cell number and decrease in LDH release compared to heparinase I treatment alone. Furthermore, the combination of insulin and heparin showed a significant increase in live cell number compared to all other groups, the LDH levels were also the lowest in this group (Figure 4).
Figure 4 Heparinase I Induced PAECs Injury was Prevented by Insulin and/or Heparin. PAECs were treated with heparinase I (0.3 U/ml) and/or insulin (1 U/ml) and/or heparin (0.5 μg/ml) for 48 hrs in serum free medium 199, then cells were counted and media LDH was determined. Insulin and/or heparin were added immediately after heparinase I addition. All reagents were only added once. Results are expressed as mean +/- SE of three dishes per group. Significantly different than a, control; b, heparinase I; c, heparinase I+heparin; d, heparinase I+insulin (P < 0.005) (one-way ANOVA).
To determine the protective effect of insulin and/or heparin on PAECs injured by heparinase I when bFGF was present in cell medium, PAECs were treated with heparinase I, heparinase I plus bFGF (1 ng/ml), heparinase I plus insulin plus bFGF, heparinase I plus heparin plus bFGF and heparinase I plus insulin plus heparin plus bFGF for 48 hours in serum free M199. Heparinase I treated PAECs showed a significant decrease in live cell number and increase in LDH release compared control cells. Cells treated with heparinase I and bFGF showed a significant decrease in LDH release, but not an increase in live cell number when compared to heparinase I treated cells. A significant increase in live cell number and decrease in LDH release was seen in cultures treated with bFGF plus insulin, bFGF plus heparin and bFGF plus insulin plus heparin in the presence of heparinase I versus heparinase I treatment alone. Furthermore, when compared to bFGF, bFGF plus insulin plus heparin showed a significant increase in live cell number in the presence of heparinase I (Figure 5).
Figure 5 The Protective Effect of Insulin and/or Heparin on PAECs Injured by Heparinase I when bFGF was Present in Cell Medium. PAECs were treated with heparinase I (0.3 U/ml), heparinase I plus bFGF (1 ng/ml), heparinase I plus insulin (1 U/ml) plus bFGF, heparinase I plus heparin (0.5 μg/ml) plus bFGF and heparinase I plus insulin plus heparin plus bFGF for 48 hrs in serum free medium 199. After 48 hrs, cells were counted and media LDH was determined. Insulin, heparin and bFGF were added immediately after heparinase I addition. Results are expressed as mean +/- SE of three culture dishes per group. Significantly different than a, control ; b, heparinase I; c, heparinase I+bFGF (P < 0.05) (one-way ANOVA).
Discussion
Vascular complications are the main causes of morbidity and mortality in diabetes mellitus. ECs play a pivotal role in the regulation of vascular tone, as well as in the maintenance of vascular integrity, blood fluidity and homeostasis. EC injury is the initial step leading to irreversible structural abnormalities, followed by progressive microvascular occlusion in the eye and kidney as well as intimal proliferation in large vessels [33-35]. The exact cause of EC injury is still unclear.
In this study, PAECs were used as an in vitro model to study human vascular disease associated with EC injury since there is a similarity between human and porcine tissue [36]. ECs of both micro and macro vascular origin present similar pathological features in diabetic complications. Thus ECs from porcine aorta (macro vessel) injured by high glucose could mimic EC injury associated with uncontrolled hyperglycemia in diabetes. High glucose injury in our model agrees with previous observations of ECs grown under hyperglycemic conditions showing decreased proliferation and fibrinolytic potential and increased programmed cell death [37,38]. It was previously reported that normal human umbilical vein ECs showed increased proliferation when cultured in medium with high glucose (30 mM) for ten to twelve days [39], a longer period compared to the seven day treatment in our study. Cell proliferation increased similarly when umbilical ECs were obtained from pregnant diabetic women [40]. PAECs treated with high glucose for seven days in the present study may represent early forms of injury and showed reduced live cell number indicating decreased cell proliferation. Some variation exists in the response of ECs to high glucose conditions. Cell conditions such as variation between animals, subtle differences in medium, CO2 levels, humidity, and other unidentified factors may be responsible for this variation.
Depletion and abnormalities in HS and HSPG have been found in the kidney, skin and aortic intima of diabetic patients with nephropathy [13-15,41]. The degradation of HSPG may play a role in EC injury leading to diabetic vascular complications. Heparanase cleaves HS chains at specific sites and may be responsible for HSPG degradation contributing to EC injury. Heparanase has been found in the kidney and urine of diabetic patients [20]. In order to determine if heparanase as well as high glucose damage ECs, PAECs were treated with heparinase I.
Several heparanases have been purified and characterized from platelets, placenta, and Chinese Hamster Ovary (CHO) cells including connective tissue activating peptide III (CTAP-III), Hpa I, Hpa II and CHO cell heparanases [42]. Heparinase I, from Flacobacterium heparinum (Cytophagia heparinia), the commercially available heparanase, was chosen for PAEC treatment, and cleaves HS [43]. Heparinase I did not cause injury when PAECs were cultured in M199 with serum for six to ten days, but showed a dose effect when cultured in serum-free medium for two days. These findings suggest that a serum constituent inhibits heparanase activity. A recently discovered cell surface protein, HS/heparin-interacting protein (HIP), was shown to prevent heparanase access to its substrate HS by competing with the same binding recognition site as in the HS chain [44,45]. Thus in our experiments, serum may contain HIP so that heparanase was active only in serum free medium.
Our findings showing cell injury with both high glucose and heparinase I treatment suggest that high glucose may induce heparanase upregulation which degrades HS causing cell injury. This injury occurs in the presence of serum which would contain HIP that may interact with heparanase at the cell surface. This suggests that with high glucose, heparanase may be produced within the cell. Heparanase activity is optimal between pH 5.0 and 6.5, with much less activity above pH 7.0 [46]. Glucose (30 mM) added for seven days to ECs lowers the medium pH (medium color become yellow) and may further stimulate heparanase activity.
Exogenous heparin significantly reduces proteinuria in diabetic patients and animals [47,48]. Heparin promotes antioxidant and barrier properties of blood vessels, prevents the formation of occlusive vascular thrombi, protects against proteolytic or oxidative damage, and lowers blood pressure [26,49,50]. Heparin and HS, similar in chemical structure, possess common physiological and biological features important in the vasculature. Heparin modifies the synthesis and the structure of HSPG [51,52]. In our study, addition of heparin to heparinase I treated ECs significantly increased live cell number and decreased LDH release compared to ECs treated with heparinase I alone suggesting that heparin has the ability to prevent cell injury by heparanase. A significant decrease in LDH release and a trend towards an increase in live cell number (close to control levels) seen in high glucose and heparin treated cells compared to high glucose alone also indicate the potential of heparin to protect ECs injured by high glucose.
HS and heparin have high affinity for bFGF and are part of the bFGF/bFGFR complex that affects the growth, differentiation and migration of many cell types [53]. Thus, bFGF function is protected by HS synthesis and perturbed by its degradation. Our results showed a significant increase in live cell number and a trend towards a decrease in LDH release both in cells treated with high glucose plus bFGF and high glucose plus bFGF plus heparin when compared to high glucose alone indicating some protective effects of bFGF. However, the live cell number in controls is significantly greater than high glucose plus bFGF indicating bFGF alone dose not eliminate high glucose injury. Since high glucose produces many metabolic and biochemical abnormalities through several cellular pathways, normal bFGF function may be altered by its interaction with abnormal metabolites. Previous studies showed that in hyperglycemia, nonenzymatic glycosylation of bFGF decreased bFGF activity [54] and could explain our observations here. Moreover, we observed similar results when ECs were damaged by heparinase I. The protective effect of bFGF on heparinase I injury is shown by a significantly decreased LDH release and a trend towards an increase in cell number compared to heparinase I injury alone. This protective ability of bFGF is consistent with that seen in high glucose plus bFGF treated ECs.
The binding of heparin to bFGF depends on the molecular mass, degree of sulfation and the disaccharide composition. Unfractionated bovine lung heparin used here is highly sulfated and of high molecular weight and has previously been shown to protect bFGF from tryptic cleavage. This capacity was reduced by N-desulfation and N-acetylation of the bovine lung heparin [55]. Previous studies have also suggested that heparin first needs to bind to the cell surface to fulfill the role of heparan sulfate in bFGF receptor interactions [56]. We have observed that bovine lung heparin binds to the surface of cultured porcine endothelial cells and thus would be able to interact with bFGF [57].
When heparin was added to ECs treated with heparinase I and bFGF, live cell number increased and LDH release decreased significantly compared to heparinase I treatment alone and showed a more pronounced increase in live cells and decrease in LDH than addition of bFGF alone suggesting that bFGF and heparin bind together to prevent HSPG from degradation by heparinase I. These findings cause us to speculate that heparin may exert its protective effect in two steps: firstly, heparin increases EC synthesis of HS; secondly, newly synthesized HS with exogenous bFGF and heparin form the bFGF/HS or heparin/ bFGFR complex which allows bFGF to play its physiological role in cell growth, differentiation, proliferation. As well, in the case of heparanase injury, heparin in the medium may compete with HS for heparanase and thus may prevent the degradation of HS.
Insulin not only stimulates cells to utilize glucose, but also promotes DNA synthesis and cell growth. The latter effect was supported in this study when ECs, treated with insulin alone, significantly increased live cell number compared to controls (Figure 1). Insulin protection of high glucose or heparinase I treated ECs was shown in all treatment combinations including insulin alone, insulin plus heparin, insulin plus bFGF and insulin plus heparin plus bFGF. The mechanism by which insulin protects ECs from high glucose injury is not entirely understood. Other vasoprotective actions of insulin are its ability to increase NO production [58], act as an antioxidant and prevent atherosclerosis by reducing oxygen consumption [59]. Our present study suggests that the combination of insulin, heparin, and bFGF may have additive effects with significantly increased live cell number and decreased LDH compared to high glucose or heparinase I alone. With high glucose injury the three combined treatments were more effective than bFGF and bFGF plus heparin and suggested increased effectiveness compared to insulin plus bFGF when medium LDH levels were considered. With heparinase I injury combined treatments were significantly more effective than bFGF with a tendency towards increased effectiveness with heparin or insulin plus bFGF when live cell number was considered.
Conclusion
This study demonstrates that both high glucose and heparinase I cause EC injury and suggests a link between hyperglycemia and heparanase induction in diabetic complications. Exogenous heparinase I damages ECs only in serum free conditions. The mechanism of EC injury by high glucose is a complicated process during which a variety of metabolic abnormalities occur, and the induction of heparanase may be one of them. The protective effect of heparin and bFGF alone or in combination was more evident in heparinase I versus high glucose injury indicating the limited damage induced by heparinase I and the complexity of glucose-induced cell injury. Cell injury by heparinase I further confirms that degradation of HSPG on the EC surface or the ECM contributes to the diabetic vasculopathy consistent with previous observations, both in vivo and in vitro. Our findings are the first to show the protective effects of heparin and/or insulin and/or bFGF on cells injured by high glucose or heparinase I. Interestingly, we found that the protective effects of bFGF in the presence of heparin or insulin in cell medium were more evident when cells were treated with heparinase I compared to high glucose. Regardless of the interaction between heparin, insulin and bFGF, this study demonstrated that these three compounds in combination protect cells from high glucose or heparanase injury. These findings provide the basis for further studies in the understanding and treatment of diabetic vascular complications.
List of abbreviations
bFGF-basic fibroblast growth factor
bFGFR-basic fibroblast growth factor receptor
CHO-Chinese Hamster Ovary
CMF-DPBS-calcium- and magnesium-free Dulbecco's phosphate-buffered saline
CTAP-III-connective tissue activating peptide III
ECM-extracellular matrix
EC(s)-endothelial cell(s)
GAG-glycosaminoglycan
GBM-glomerular basement membrane
HIP-HS/heparin-interacting proteinHS- heparan sulfate
HSPG(s)-Heparan sulfate proteoglycan(s)
LDH-lactate dehydrogenase
NO-nitric oxide
PAEC(s)-Porcine Aortic Endothelial Cell(s)
SE-standard error
vWF-von Willebrand Factor
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
JH participated in the design of the study carried out the experiments and drafted the manuscript. AM conceived the study and participated in the design. LH conceived the study, participated in the design and co-ordination and contributed to the writing of the manuscript.
Acknowledgements
This study was supported by a grant from the Canadian Diabetes Association in honor of Mildred I. Wright and from private sources. We are indebted to Tilly Ping and Dr. Adrienne Woytowich for technical assistance.
==== Refs
Ruderman NB Williamson JR Brownlee M Glucose and diabetic vascular disease FASEB J 1992 6 2905 2914 1644256
Lee TS Saltsman KA Ohashi H King GL Activation of protein kinase C by elevation of glucose concentration: proposal for a mechanism in the development of diabetic vascular complications Proc Natl Acad Sci U S A 1989 86 5141 5145 2740348
Kamata K Miyata N Abiru T Kasuya Y Functional changes in vascular smooth muscle and endothelium of arteries during diabetes mellitus Life Sci 1992 50 1379 1387 1374144 10.1016/0024-3205(92)90256-O
Aiello LP Avery RL Arrigg PG Keyt BA Jampel HD Shah ST Pasquale LR Thieme H Iwamoto MA Park JE . Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders
2 N Engl J Med 1994 331 1480 1487 7526212 10.1056/NEJM199412013312203
Mogensen CE Schmitz A Christensen CK Comparative renal pathophysiology relevant to IDDM and NIDDM patients
5 Diabetes Metab Rev 1988 4 453 483 3061756
Jarrett RJ McCartney P Keen H The Bedford survey: ten year mortality rates in newly diagnosed diabetics, borderline diabetics and normoglycaemic controls and risk indices for coronary heart disease in borderline diabetics Diabetologia 1982 22 79 84 7060853
Pirart J [Diabetes mellitus and its degenerative complications: a prospective study of 4,400 patients observed between 1947 and 1973 (3rd and last part) (author's transl)] Diabete Metab 1977 3 245 256 598565
Muir H Structure and function of proteoglycans of cartilage and cell-matrix interactions Soc Gen Physiol Ser 1977 32 87 99 333597
Hook M Kjellen L Johansson S Cell-surface glycosaminoglycans Annu Rev Biochem 1984 53 847 869 6433783 10.1146/annurev.bi.53.070184.004215
Kraemer PM Heparan sulfates of cultured cells. I. Membrane-associated and cell-sap species in Chinese hamster cells Biochemistry 1971 10 1437 1445 4253009 10.1021/bi00784a026
Hedman K Johansson S Vartio T Kjellen L Vaheri A Hook M Structure of the pericellular matrix: association of heparan and chondroitin sulfates with fibronectin-procollagen fibers Cell 1982 28 663 671 6210440 10.1016/0092-8674(82)90221-5
Hardebo JE Kahrstrom J Endothelial negative surface charge areas and blood-brain barrier function Acta Physiol Scand 1985 125 495 499 4083048
van den BJ van den Heuvel LP Bakker MA Veerkamp JH Assmann KJ Berden JH A monoclonal antibody against GBM heparan sulfate induces an acute selective proteinuria in rats Kidney Int 1992 41 115 123 1593846
van den BJ van den Heuvel LP Bakker MA Veerkamp JH Assmann KJ Weening JJ Berden JH Distribution of GBM heparan sulfate proteoglycan core protein and side chains in human glomerular diseases Kidney Int 1993 43 454 463 8441243
Wasty F Alavi MZ Moore S Distribution of glycosaminoglycans in the intima of human aortas: changes in atherosclerosis and diabetes mellitus Diabetologia 1993 36 316 322 8477876 10.1007/BF00400234
Brown DM Klein DJ Michael AF Oegema TR 35S-glycosaminoglycan and 35S-glycopeptide metabolism by diabetic glomeruli and aorta Diabetes 1982 31 418 425 7152134
Kjellen L Bielefeld D Hook M Reduced sulfation of liver heparan sulfate in experimentally diabetic rats Diabetes 1983 32 337 342 6219905
Levy P Picard J Bruel A Evidence for diabetes-induced alterations in the sulfation of heparin sulfate intestinal epithelial cells Life Sci 1984 35 2613 2620 6239964 10.1016/0024-3205(84)90029-8
Freeman C Browne AM Parish CR Evidence that platelet and tumour heparanases are similar enzymes Biochem J 1999 342 ( Pt 2) 361 368 10455023 10.1042/0264-6021:3420361
Katz A Van Dijk DJ Aingorn H Erman A Davies M Darmon D Hurvitz H Vlodavsky I Involvement of human heparanase in the pathogenesis of diabetic nephropathy Isr Med Assoc J 2002 4 996 1002 12489489
Mandal AK Puchalski JT Lemley-Gillespie S Taylor CA Kohno M Effect of insulin and heparin on glucose-induced vascular damage in cell culture
15 Kidney Int 2000 57 2492 2501 10844618 10.1046/j.1523-1755.2000.00108.x
Zeng G Quon MJ Insulin-stimulated production of nitric oxide is inhibited by wortmannin. Direct measurement in vascular endothelial cells
8 J Clin Invest 1996 98 894 898 8770859
Vallance P Collier J Moncada S Effects of endothelium-derived nitric oxide on peripheral arteriolar tone in man
20 Lancet 1989 2 997 1000 2572793 10.1016/S0140-6736(89)91013-1
Hiebert LM Wice SM McDuffie NM Jaques LB The heparin target organ--the endothelium. Studies in a rat model Q J Med 1993 86 341 348 8327652
Nader HB Toma L Pinhal MA Buonassisi V Colburn P Dietrich CP Effect of heparin and dextran sulfate on the synthesis and structure of heparan sulfate from cultured endothelial cells Semin Thromb Hemost 1991 17 Suppl 1 47 56 2068571
Hiebert LM Liu JM Heparin protects cultured arterial endothelial cells from damage by toxic oxygen metabolites Atherosclerosis 1990 83 47 51 2390135
Moscatelli D High and low affinity binding sites for basic fibroblast growth factor on cultured cells: absence of a role for low affinity binding in the stimulation of plasminogen activator production by bovine capillary endothelial cells J Cell Physiol 1987 131 123 130 3032990 10.1002/jcp.1041310118
Flaumenhaft R Moscatelli D Rifkin DB Heparin and heparan sulfate increase the radius of diffusion and action of basic fibroblast growth factor J Cell Biol 1990 111 1651 1659 2170425 10.1083/jcb.111.4.1651
Broadley KN Aquino AM Woodward SC Buckley-Sturrock A Sato Y Rifkin DB Davidson JM Monospecific antibodies implicate basic fibroblast growth factor in normal wound repair Lab Invest 1989 61 571 575 2811305
Yayon A Klagsbrun M Esko JD Leder P Ornitz DM Cell surface, heparin-like molecules are required for binding of basic fibroblast growth factor to its high affinity receptor Cell 1991 64 841 848 1847668 10.1016/0092-8674(91)90512-W
Nissen NN Shankar R Gamelli RL Singh A DiPietro LA Heparin and heparan sulphate protect basic fibroblast growth factor from non-enzymic glycosylation Biochem J 1999 338 ( Pt 3) 637 642 10051433 10.1042/0264-6021:3380637
Gotlieb AI Spector W Migration into an in vitro experimental wound: a comparison of porcine aortic endothelial and smooth muscle cells and the effect of culture irradiation Am J Pathol 1981 103 271 282 7195151
Bresnick GH Davis MD Myers FL de Venecia G Clinicopathologic correlations in diabetic retinopathy. II. Clinical and histologic appearances of retinal capillary microaneurysms Arch Ophthalmol 1977 95 1215 1220 880082
Steffes MW Osterby R Chavers B Mauer SM Mesangial expansion as a central mechanism for loss of kidney function in diabetic patients Diabetes 1989 38 1077 1081 2670639
Baumgartner-Parzer SM Wagner L Pettermann M Grillari J Gessl A Waldhausl W High-glucose--triggered apoptosis in cultured endothelial cells
2 Diabetes 1995 44 1323 1327 7589831
Lee KT Tumbleson ME Swine as animal models in cardiovascular research Swine in biomedical research 1986 New York, Plenum 1481 1496
Lorenzi M Glucose toxicity in the vascular complications of diabetes: the cellular perspective Diabetes Metab Rev 1992 8 85 103 1425126
Cagliero E Roth T Roy S Lorenzi M Characteristics and mechanisms of high-glucose-induced overexpression of basement membrane components in cultured human endothelial cells Diabetes 1991 40 102 110 2015965
Sank A Wei D Reid J Ertl D Nimni M Weaver F Yellin A Tuan TL Human endothelial cells are defective in diabetic vascular disease J Surg Res 1994 57 647 653 7996840 10.1006/jsre.1994.1195
Tamsma JT van den BJ Bruijn JA Assmann KJ Weening JJ Berden JH Wieslander J Schrama E Hermans J Veerkamp JH Expression of glomerular extracellular matrix components in human diabetic nephropathy: decrease of heparan sulphate in the glomerular basement membrane Diabetologia 1994 37 313 320 8174847 10.1007/BF00398060
Bame KJ Heparanases: endoglycosidases that degrade heparan sulfate proteoglycans
2 Glycobiology 2001 11 91R 98R 11445547 10.1093/glycob/11.6.91R
Desai UR Wang HM Linhardt RJ Substrate specificity of the heparin lyases from Flavobacterium heparinum Arch Biochem Biophys 1993 306 461 468 8215450 10.1006/abbi.1993.1538
Liu S Zhou F Hook M Carson DD A heparin-binding synthetic peptide of heparin/heparan sulfate-interacting protein modulates blood coagulation activities Proc Natl Acad Sci U S A 1997 94 1739 1744 9050848 10.1073/pnas.94.5.1739
Marchetti D Liu S Spohn WC Carson DD Heparanase and a synthetic peptide of heparan sulfate-interacting protein recognize common sites on cell surface and extracellular matrix heparan sulfate J Biol Chem 1997 272 15891 15897 9188488 10.1074/jbc.272.25.15891
Ihrcke NS Parker W Reissner KJ Platt JL Regulation of platelet heparanase during inflammation: role of pH and proteinases J Cell Physiol 1998 175 255 267 9572470 10.1002/(SICI)1097-4652(199806)175:3<255::AID-JCP3>3.0.CO;2-N
Tamsma JT van der Woude FJ Lemkes HH Effect of sulphated glycosaminoglycans on albuminuria in patients with overt diabetic (type 1) nephropathy Nephrol Dial Transplant 1996 11 182 185 8649631
Gambaro G Cavazzana AO Luzi P Piccoli A Borsatti A Crepaldi G Marchi E Venturini AP Baggio B Glycosaminoglycans prevent morphological renal alterations and albuminuria in diabetic rats Kidney Int 1992 42 285 291 1328749
Fairman RP Sessler CN Bierman M Glauser FL Protamine sulfate causes pulmonary hypertension and edema in isolated rat lungs J Appl Physiol 1987 62 1363 1367 3597212 10.1063/1.339639
Mandal AK Lyden TW Fazel A Saklayen MG Mehrotra B Mehling B Taylor CA Yokokawa K Colvin RV Heparin-induced endothelial cell cytoskeletal reorganization: a potential mechanism for vascular relaxation Kidney Int 1995 48 1508 1516 8544408
Colburn P Buonassisi V Anti-clotting activity of endothelial cell cultures and heparan sulfate proteoglycans Biochem Biophys Res Commun 1982 104 220 227 6462138 10.1016/0006-291X(82)91962-3
Colburn P Buonassisi V Dietrich CP Nader HB N-glycansulfated fibronectin: one of the several sulfated glycoproteins synthesized by endothelial cells in culture Biochem Biophys Res Commun 1987 147 920 926 3663221 10.1016/S0006-291X(87)80158-4
Folkman J Klagsbrun M Angiogenic factors Science 1987 235 442 447 2432664
Giardino I Edelstein D Brownlee M Nonenzymatic glycosylation in vitro and in bovine endothelial cells alters basic fibroblast growth factor activity. A model for intracellular glycosylation in diabetes J Clin Invest 1994 94 110 117 8040253
Coltrini D Rusnati M Zoppetti G Oreste P Grazioli G Naggi A Presta M Different effects of mucosal, bovine lung and chemically modified heparin on selected biological properties of basic fibroblast growth factor
1 Biochem J 1994 303 ( Pt 2) 583 590 7980421
Fannon M Forsten KE Nugent MA Potentiation and inhibition of bFGF binding by heparin: a model for regulation of cellular response
3 Biochemistry 2000 39 1434 1445 10684625 10.1021/bi991895z
Hiebert LM McDuffie NM The internalization and release of heparins by cultured endothelial cells: the process is cell source, heparin source, time and concentration dependent
3 Artery 1990 17 107 118 2155600
Shen W Xu X Ochoa M Zhao G Wolin MS Hintze TH Role of nitric oxide in the regulation of oxygen consumption in conscious dogs Circ Res 1994 75 1086 1095 7525103
|
16086844
|
PMC1192813
|
CC BY
|
2021-01-04 16:25:03
|
no
|
Cardiovasc Diabetol. 2005 Aug 9; 4:12
|
utf-8
|
Cardiovasc Diabetol
| 2,005 |
10.1186/1475-2840-4-12
|
oa_comm
|
==== Front
Comp HepatolComparative Hepatology1476-5926BioMed Central London 1476-5926-4-61600883310.1186/1476-5926-4-6ResearchAccuracy of hyaluronic acid level for predicting liver fibrosis stages in patients with hepatitis C virus Halfon Philippe [email protected]ère Marc [email protected]énaranda Guillaume [email protected] Romaric [email protected] Christophe [email protected] Danielle [email protected] Albert [email protected] Isabelle [email protected] Isabelle [email protected] Alessandra [email protected] Denis [email protected] Department of virology, Alphabio Laboratory, Marseille, France2 Department of Hepato-Gastroenterology, Saint-Joseh Hospital, Marseille, France3 Department of Hepato-Gastroenterology, Hyères Hospital, Hyères, France4 Department of Hepato-Gastroenterology, La Conception Hospital, Marseille, France5 Department of Hepato-Gastroenterology, Archet Hospital, Nice, France6 Department of Hepato-Gastroenterology, Arnault Tzanck Institute, Saint Laurent du Var, France2005 11 7 2005 4 6 6 7 4 2005 11 7 2005 Copyright © 2005 Halfon et al; licensee BioMed Central Ltd.2005Halfon 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 patients with chronic hepatitis C virus, liver biopsy is the gold standard for assessing liver disease stage; nevertheless, it is prone to complications, some of them serious. Non-invasive methods have been proposed as surrogate markers for liver fibrosis. It was shown that serum hyaluronic acid (HA) level increases with the development for liver fibrosis. The aim of this study was to evaluate the diagnostic value of HA as well as to determine the HA level cut-off for predicting the presence or absence of fibrosis, severe fibrosis, and cirrhosis.
Results
405 patients with chronic hepatitis C were prospectively included with biomarker measurement and liver biopsy done the same day: 151 in the training set (only biopsy lengths of 25 mm or more) and 254 in the validation set. For the discrimination of significant fibrosis, severe fibrosis, and cirrhosis in the training set, the areas under curve (AUCs) were 0.75 ± 0.03, 0.82 ± 0.02, and 0.89 ± 0.03, respectively. Absence of significant fibrosis, severe fibrosis, and cirrhosis can be predicted by HA levels of 16, 25, and 50 μg/l, respectively (with negative predictive values of 82%, 89%, and 100%, in the same order). Presence of significant fibrosis, severe fibrosis, and cirrhosis can be predicted by HA levels of 121, 160, and 237 μg/l, respectively (with positive predictive values of 94%, 100%, and 57%, in the same order).
Conclusion
In the validation set, HA was accurate in predicting significant fibrosis, severe fibrosis, and cirrhosis with AUCs of 0.73, 0.77, and 0.97, respectively. Moreover, accurate HA level cut-offs were defined for predicting significant fibrosis, severe fibrosis, and cirrhosis. Thus, the study supports that HA level may be clinically useful as a non-invasive marker for liver fibrosis and/or cirrhosis.
==== Body
Background
Liver biopsy is currently recommended as the gold standard method of staging fibrosis in patients with chronic hepatitis C [1,2]. The risk of developing cirrhosis depends on the stage (degree of fibrosis) and the grade (degree of inflammation and necrosis) observed in the initial liver biopsy [3,4]. This procedure, however, is invasive and has potential complications [5,6]. Non-invasive approaches developed to assess histological samples include clinical symptoms, routine laboratory tests, and radiolographic imaging [7-10]. Several clinical studies have attempted to identify serum markers that correlate with the degree of fibrosis and thus could be used, with feasibility, in conjunction with or in place of a liver biopsy [2-4,6,8,9]. The serum markers of fibrogenesis include platelet count [11], prothrombin time [12], the ratio of alanine aminotransferase and aspartate aminotransferase levels [8], gamma-glutamyl transferase level [13], and serum albumin level [14]. Fibrotest (FT) is a simple non-invasive panel of biochemical markers for fibrosis and activity [15].
Another non-invasive approach relies on the measurement of substances that regulate fibrosis or participate in the generation of the liver extracellular matrix. The most applicable include hyaluronic acid (HA) [16], type IV collagen [17], N-terminal propeptide of type III procollagen [18], metalloproteinases [19], inhibitors of metalloproteinases [19], and growth-transforming factor beta [20]. HA is a high molecular weight glycosaminoglycan, which is an essential component of extracellular matrix in virtually every tissue in the body [21]. In the liver, HA is mostly synthesized by the hepatic stellate cells and degraded by the sinusoidal endothelial cells [6]. HA levels are increased in chronic liver diseases [6]. In patients with chronic hepatitis C virus (HCV), HA levels increase with the development of liver fibrosis. Moreover, in patients with cirrhosis, HA levels correlate with clinical severity [7,10,11].
The first aim of this study was to evaluate the diagnostic value of HA for significant fibrosis (F2-F4), severe fibrosis, (F3F4) and cirrhosis (F4), in patients with HCV infection. The second aim was to determine the serum HA level cut-off to predict both presence and absence of F2-F4, F3F4, and F4.
Results
Patients
The cohort included 405 patients. Table 1 shows the patient characteristics at the time of liver biopsy. The training and validation sets did not significantly differ in any of the assessed variables. Of the patients, 47% (190/405) had significant fibrosis (F2-F4), 24% (99/405) had severe fibrosis (F3F4), and 6% (25/405) had cirrhosis.
Table 1 Characteristics of the 405 patients at the time of liver biopsy (comparison between the training and the validation sets).
Characteristics Training set (n = 151) Validation set (n = 254) All patients (n = 405)
Age (Mean ± SD) 51 ± 14 47 ± 12 49 ± 13
Male (n (%)) 82 (54) 133 (52) 215 (53)
HA (μg/l) (Mean {95% CI}) 63 {47;79} 53 {41;65} 57 {47;67}
Stage of fibrosis (n (%))
0 28 (19) 33 (13) 61 (15)
1 51 (34) 103 (41) 154 (38)
2 33 (22) 58 (23) 91 (23)
3 27 (18) 47 (18) 74 (18)
4 12 (7) 13 (5) 25 (6)
Hyaluronic acid and fibrosis in the training set
Figure 1 shows HA levels and stages of fibrosis are well correlated (Spearman r = 0.55 – p < .0001). Although there is an overlap between HA levels and fibrosis determined by liver biopsy, there is a significant increase in HA levels when considering F0 to F4 scores (Kruskal-Wallis – p < 0.0001).
Figure 1 Box & Whisker plot representing the relation between the stage of fibrosis and HA level. The line through the box is the median; the top and bottom edges of each box represent the 25th and 75th percentiles, giving the interquartile range; and the cross in the box is the mean. The vertical lines at each side of the box represent distribution from the quartile to the farthest observation. The curve represents the HA median value of each fibrosis stage (F0: 20 μg/l, F1: 25 μg/l, F2: 30 μg/l, F3: 58 μg/l, and F4: 180 μg/l). The relation between the stages of fibrosis and HA level was statistically significant (Kruskal-Wallis – p < 0.0001). Spearman rank correlation coefficient (r) between the stage of fibrosis and HA level was 0.55.
Fibrosis, severe fibrosis, and cirrhosis diagnosis
Figure 2 and 3 shows receiver operating characteristic curves of discriminatory values of HA according to the severity of liver fibrosis in the training and validation sets. For the discrimination of fibrosis, severe fibrosis, and cirrhosis in the training set, areas under curve (AUCs) were (Mean ± SE) 0.75 ± 0.03, 0.82 ± 0.02, and 0.89 ± 0.03, respectively. In the validation set, AUCs were 0.73 ± 0.03, 0.77 ± 0.04, and 0.97 ± 0.04, respectively.
Figure 2 Receiver operating characteristic curves of HA for the prediction of significant fibrosis (F2-F4), severe fibrosis (F3F4), and cirrhosis (F4) in the training set.
Figure 3 Receiver operating characteristic curves of HA for the prediction of significant fibrosis (F2-F4), severe fibrosis (F3F4), and cirrhosis (F4) in the validation set.
Fibrosis
Two cut-off values were chosen for identifying absence (less than 16 μg/l) and presence (greater than 121 μg/l) of significant fibrosis (F2-F4). Applying the lower cut-off, the presence of significant fibrosis could be excluded with a high certainty as only 3 (17%) of the 18 patients with an HA level below 16 μg/l had significant fibrosis, with a negative predictive value (NPV) of 83%. In the validation set, 11 (18%) of 60 patients with a score below 16 μg/l had significant fibrosis (NPV of 82%). Applying the high cut-off to the training group (HA greater than 121 μg/l), only 2 (13%) of the 15 patients with HA greater than 121 μg/l had no fibrosis, with a positive predictive value (PPV) of 87%. In the validation set, 16 of the 17 patients with HA greater than 121 μg/l had significant fibrosis (PPV of 94%) (Table 2).
Table 2 Diagnostic performance of HA in the validation set.
HA cut-off Sensitivity (%) Specificity (%) NPV (%) PPV (%) Population involved (%) Interpretation
<16 91 36 82 55 24 Absence of fibrosis (F0F1) (82% certainty)
>121 14 99 57 94 7 Presence of fibrosis (F2) (94% certainty)
≤ 25 78 53 89 34 46 Absence of severe fibrosis (F0F1F2) (89% certainty)
>160 22 100 81 100 5 Presence of severe fibrosis (F3) (100% certainty)
≤ 50 100 79 100 20 75 Absence of cirrhosis (F0F1F2F3) (100% certainty)
>237 31 99 96 57 3 Presence of cirrhosis (F4) (57% certainty)
Note 1: Accuracy = Sensitivity + Specificity + NPV + PPV. Note 2: Population involved stands for the proportion of patients who fall in the corresponding cut-off.
Severe fibrosis
As for significant fibrosis diagnosis, 2 cut-off values were chosen for identifying absence (less than 25 μg/l) and presence (greater than 160 μg/l) of severe fibrosis (F3F4). Applying the lower cut-off, only 3 (5%) of the 64 patients with HA lower than 25 μg/l had severe fibrosis (NPV of 95%). In the validation set, only 13 (11%) of the 123 patients with HA lower than 25 μg/l had severe fibrosis (NPV of 89%). Applying the higher cut-off (HA greater than 160 μg/l) to the training set, 10 (NPV of 91%) out of 11 patients with HA greater than 160 μg/l had severe fibrosis, and none of the 13 patients with HA greater than 160 μg/l from the validation set had severe fibrosis (PPV of 100%) (Table 2).
Cirrhosis
Two cut-off values were chosen for identifying absence (less than 50 μg/l) and presence (greater than 237 μg/l) of cirrhosis (F4). Applying the lower cut-off, 100 (NPV of 99%) of the 101 patients with HA lower than 264 μg/l had no cirrhosis. In the validation set, none of the patients with HA lower than 50 μg/l had cirrhosis (NPV of 100%). When applying the higher cut-off (237 μg/l) to the training set, a fair PPV of 71% was obtained for predicting the presence of cirrhosis. The PPV was lower when applying the cut-off to the validation set (PPV of 57%) (Table 2).
Discussion
Combining HA level with other serum markers for assessing liver fibrosis has been considered in some other studies [22,23]. One of the interests of our study was to focus on the diagnostic accuracy of HA alone in predicting fibrosis and cirrhosis in HCV-infected patients.
In the present study, HA level was accurate in predicting significant fibrosis, severe fibrosis, and cirrhosis, with AUCs of 0.75, 0.82, and 0.89, respectively, in the training set; and of 0.73, 0.77, and 0.97, respectively, in the validation set. Using values below the lower cut-off level or above the higher cut-off level, one could predict absence or presence of significant fibrosis, severe fibrosis, and cirrhosis in 31%, 51%, and 78%, respectively, in the validation set patients. Therefore, it is likely that the association of HA level with the degree of hepatic fibrosis represents an indirect one, expressing the functional correlation between fibrosis and both the concomitant capillarization and hepatic hemodynamic changes.
Significant fibrosis can be predicted by a HA level of <16 μg/l for its absence (NPV of 82%) and of >121 μg/l for its presence (PPV of 94%) in the validation set. Severe fibrosis can be predicted by a HA level of ≤ 25 μg/l for its absence (NPV of 89%) and of >160 μg/l for its presence (PPV of 100%). Cirrhosis can be predicted by an HA level of ≤ 50 μg/l for its absence (NPV of 100%) and of > 237 μg/l for its presence (PPV of 57%). Considering a serum HA cut-off of 60 μg/l for absence of cirrhosis diagnosis, our data are in the same range as those of other studies [3,24].
A cut-off value of 110 μg/l was suggested for separating patients with and without cirrhosis [18]. Taking in consideration this cut-off in our cohort of patients, a misclassification of cirrhotic patients was observed in 23% (3/13) (sensitivity of 77%) of patients with proven cirrhosis. This difference may be due to the use of a different HA assay.
Our study included a sufficient proportion of patients with significant fibrosis: 47% in the training set and 46% in the validation set; however, the proportion of patients with cirrhosis was low in the two sets: 7% and 5%, respectively. The second limitation of this study is the number of unclassified patients (between 22% and 69%).
Conclusion
This study showed that significant fibrosis, severe fibrosis, and cirrhosis can be predicted by serum HA levels in patients with HCV infection. The notion of routinely measuring a marker that reflects the function of the sinusoidal endothelial cells, rather than the hepatocytes themselves, is an exciting concept. Serum HA would be clinically useful as a non-invasive marker of liver fibrosis or cirrhosis in HCV-infected patients. It suffers from the need to limit, as much as possible, potential confounding variables such as the effects of exercise and eating.
Further studies conducted in a large cohort of cirrhosis patients are needed to corroborate this study, namely because few of our patients had cirrhosis and the cut-off levels must be considered in an independent study. Moreover, a comparison of HA levels with other non-invasive markers and scores of liver fibrosis (FT, APRI, Forns, age-platelets index, platelet count, prothrombin time, etc.) would be of interest.
Methods
Study participants
The cohort included 405 patients (mean age 49 ± 13 years, 53% men) with HCV infection between November 2002 and December 2003, in five centers in southern France [Conception Hospital and Saint-Joseph Hospital (Marseille), Archet Hospital (Nice), Hyères Hospital (Hyères), and Arnault Tzanck Institute (St Laurent du Var)]. HA assessment was evaluated with data from 151 patients (training group) and was validated in the remaining 254 patients (validation group). Only biopsy lengths of 25 mm or more were included in the training group, in accordance with a recent study [25]; the remaining patients were included in the validation set. None of the patients had joint injuries, based on clinical examinations and inflammatory markers estimated at the time of inclusion in the study. None of the patients had renal impairment, based on normal creatinine clearance (Cockroft calculation).
Liver biopsies
Liver biopsy examination was performed in each center by evaluating the stage of fibrosis and grade of activity according to the METAVIR scoring system [26,27]. Liver biopsies were histologically assessed in the local centers and a second assessment of each biopsy was done in a reference laboratory. The second assessments were done by the same pathologist. In this study, all liver biopsies were re-staged by the central reference laboratory (n = 405). No liver biopsies were found with more than one-stage difference between the local pathologist and the reference pathologist. One-stage discordance is considered pathology-dependent and thus not considered significant. Fibrosis was staged on a scale of 0 to 4: F0 = no fibrosis, F1 = portal fibrosis without septa, F2 = portal fibrosis and few septa, F3 = numerous septa without cirrhosis, F4 = cirrhosis.
Hyaluronic acid
HA levels were measured by Alphabio Laboratories, Marseille, with the Corgenix Hyaluronic Acid Test Kit, Corgenix Inc., CO, following the manufacturer's instructions (patients in fasting conditions, no physical effort). Each HA level was measured in duplicate (range 1 to 871 μg/l) and a pool control set was used. All serum samples were obtained on the day of liver biopsy.
Statistical analysis
The association between HA levels and liver biopsy staging was measured with the Kruskal-Wallis multiple comparison test and with the Spearman rank correlation coefficient. A p < 0.05 was considered significant. Respective diagnostic values were reported by the area under the receiver operating characteristic curves (AUC) (± standard error mean), NPV, PPV, sensitivity, and specificity.
Authors' contributions
PH and MB conceived and wrote the manuscript. RD, CR, DBF, AT, IP, IA, ARA and DO were responsible for the patient drafting, carried out biochemical analysis, participated in the coordination of the study, and drafted the paper. GP performed the statistical analysis and participated in the writing of the paper. All authors read and approved the final manuscript
Acknowledgements
We thank Denice Taylor (Corgenix Inc., CO) for kindly providing hyaluronic acid test kit.
==== Refs
Parillo R Long term effect of antiviral therapy on liver histology in chronic hepatitis C 2004 EASL
Pares A Deulofeu R Gimenez A Caballeria L Bruguera M Caballeria J Ballesta AM Rodes J Serum hyaluronate reflects hepatic fibrogenesis in alcoholic liver disease and is useful as a marker of fibrosis Hepatology 1996 24 1399 1403 8938169
Oberti F Valsesia E Pilette C Rousselet MC Bedossa P Aube C Gallois Y Rifflet H Maiga MY Penneau-Fontbonne D Cales P Noninvasive diagnosis of hepatic fibrosis or cirrhosis Gastroenterology 1997 113 1609 1616 9352863 10.1053/gast.1997.v113.pm9352863
Wai CT Greenson JK Fontana RJ Kalbfleisch JD Marrero JA Conjeevaram HS Lok AS A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C Hepatology 2003 38 518 526 12883497 10.1053/jhep.2003.50346
Cadranel JF [Good clinical practice guidelines for fine needle aspiration biopsy of the liver: past, present and future] Gastroenterol Clin Biol 2002 26 823 824 12434092
Guechot J Loria A Serfaty L Giral P Giboudeau J Poupon R Serum hyaluronan as a marker of liver fibrosis in chronic viral hepatitis C: effect of alpha-interferon therapy J Hepatol 1995 22 22 26 7751583 10.1016/0168-8278(95)80255-X
Gibson PR Fraser JR Brown TJ Finch CF Jones PA Colman JC Dudley FJ Hemodynamic and liver function predictors of serum hyaluronan in alcoholic liver disease Hepatology 1992 15 1054 1059 1592343
Forns X Ampurdanes S Llovet JM Aponte J Quinto L Martinez-Bauer E Bruguera M Sanchez-Tapias JM Rodes J Identification of chronic hepatitis C patients without hepatic fibrosis by a simple predictive model Hepatology 2002 36 986 992 12297848
Imbert-Bismut F Ratziu V Pieroni L Charlotte F Benhamou Y Poynard T Biochemical markers of liver fibrosis in patients with hepatitis C virus infection: a prospective study Lancet 2001 357 1069 1075 11297957 10.1016/S0140-6736(00)04258-6
Poupon RE Balkau B Guechot J Heintzmann F Predictive factors in ursodeoxycholic acid-treated patients with primary biliary cirrhosis: role of serum markers of connective tissue Hepatology 1994 19 635 640 8119688
Korner T Kropf J Gressner AM Serum laminin and hyaluronan in liver cirrhosis: markers of progression with high prognostic value J Hepatol 1996 25 684 688 8938546 10.1016/S0168-8278(96)80239-X
Benlloch S Berenguer M Prieto M Rayon JM Aguilera V Berenguer J Prediction of fibrosis in HCV-infected liver transplant recipients with a simple noninvasive index Liver Transpl 2005 11 456 462 15776403 10.1002/lt.20381
Silva IS Ferraz ML Perez RM Lanzoni VP Figueiredo VM Silva AE Role of gamma-glutamyl transferase activity in patients with chronic hepatitis C virus infection J Gastroenterol Hepatol 2004 19 314 318 14748879 10.1111/j.1440-1746.2003.03256.x
Lu LG Zeng MD Wan MB Li CZ Mao YM Li JQ Qiu DK Cao AP Ye J Cai X Chen CW Wang JY Wu SM Zhu JS Zhou XQ Grading and staging of hepatic fibrosis, and its relationship with noninvasive diagnostic parameters World J Gastroenterol 2003 9 2574 2578 14606100
Poynard T Imbert-Bismut F Munteanu M Messous D Myers RP Thabut D Ratziu V Mercadier A Benhamou Y Hainque B Overview of the diagnostic value of biochemical markers of liver fibrosis (FibroTest, HCV FibroSure) and necrosis (ActiTest) in patients with chronic hepatitis C Comp Hepatol 2004 3 8 15387887 10.1186/1476-5926-3-8
Guechot J Poupon RE Giral P Balkau B Giboudeau J Poupon R Relationship between procollagen III aminoterminal propeptide and hyaluronan serum levels and histological fibrosis in primary biliary cirrhosis and chronic viral hepatitis C J Hepatol 1994 20 388 393 8014451 10.1016/S0168-8278(94)80013-8
Santos VN Leite-Mor MM Kondo M Martins JR Nader H Lanzoni VP Parise ER Serum laminin, type IV collagen and hyaluronan as fibrosis markers in non-alcoholic fatty liver disease Braz J Med Biol Res 2005 38 747 753 15917956
Guechot J Laudat A Loria A Serfaty L Poupon R Giboudeau J Diagnostic accuracy of hyaluronan and type III procollagen amino-terminal peptide serum assays as markers of liver fibrosis in chronic viral hepatitis C evaluated by ROC curve analysis Clin Chem 1996 42 558 563 8605673
El-Gindy I El Rahman AT El-Alim MA Zaki SS Diagnostic potential of serum matrix metalloproteinase-2 and tissue inhibitor of metalloproteinase-1 as non-invasive markers of hepatic fibrosis in patients with HCV related chronic liver disease Egypt J Immunol 2003 10 27 35 15719620
Wang H Mengsteab S Tag CG Gao CF Hellerbrand C Lammert F Gressner AM Weiskirchen R Transforming growth factor-beta1 gene polymorphisms are associated with progression of liver fibrosis in Caucasians with chronic hepatitis C infection World J Gastroenterol 2005 11 1929 1936 15800982
Lindqvist U Laurent TC Serum hyaluronan and aminoterminal propeptide of type III procollagen: variation with age Scand J Clin Lab Invest 1992 52 613 621 1455153
Patel K Gordon SC Jacobson I Hezode C Oh E Smith KM Pawlotsky JM McHutchison JG Evaluation of a panel of non-invasive serum markers to differentiate mild from moderate-to-advanced liver fibrosis in chronic hepatitis C patients J Hepatol 2004 41 935 942 15582126 10.1016/j.jhep.2004.08.008
Rosenberg WM Voelker M Thiel R Becka M Burt A Schuppan D Hubscher S Roskams T Pinzani M Arthur MJ Serum markers detect the presence of liver fibrosis: a cohort study Gastroenterology 2004 127 1704 1713 15578508 10.1053/j.gastro.2004.08.052
McHutchison JG Blatt LM de Medina M Craig JR Conrad A Schiff ER Tong MJ Measurement of serum hyaluronic acid in patients with chronic hepatitis C and its relationship to liver histology. Consensus Interferon Study Group J Gastroenterol Hepatol 2000 15 945 951 11022838 10.1046/j.1440-1746.2000.02233.x
Bedossa P Dargere D Paradis V Sampling variability of liver fibrosis in chronic hepatitis C Hepatology 2003 38 1449 1457 14647056
Intraobserver and interobserver variations in liver biopsy interpretation in patients with chronic hepatitis C. The French METAVIR Cooperative Study Group Hepatology 1994 20 15 20 8020885 10.1016/0270-9139(94)90128-7
Bedossa P Poynard T An algorithm for the grading of activity in chronic hepatitis C. The METAVIR Cooperative Study Group Hepatology 1996 24 289 293 8690394
|
16008833
|
PMC1192814
|
CC BY
|
2021-01-04 16:38:24
|
no
|
Comp Hepatol. 2005 Jul 11; 4:6
|
utf-8
|
Comp Hepatol
| 2,005 |
10.1186/1476-5926-4-6
|
oa_comm
|
==== Front
Epidemiol Perspect InnovEpidemiologic perspectives & innovations : EP+I1742-5573BioMed Central London 1742-5573-2-61595339310.1186/1742-5573-2-6Analytic PerspectiveTeaching: the role of active manipulation of three-dimensional scatter plots in understanding the concept of confounding Busstra Cora MC [email protected] Rob [email protected] 't Veer Pieter [email protected] Division of Human Nutrition, Wageningen University, P. O. Box 8129, 6700 EV Wageningen, The Netherlands2 Wageningen Multi Media Research Centre, Wageningen University, Dreijenplein 2, 6703 HB Wageningen, The Netherlands2005 14 6 2005 2 6 6 3 12 2004 14 6 2005 Copyright © 2005 Busstra et al; licensee BioMed Central Ltd.2005Busstra 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.
In teaching epidemiology, confounding is a difficult topic. The authors designed active learning objects (LO) based on manipulable three-dimensional (3D) plots to facilitate understanding of confounding. The 3D LOs help illustrate of how confounding can occur, how it generates bias and how to adjust for it. For the development of the LOs, guidelines were formulated based on epidemiology and theories of instructional design. These included integrating the conceptual and empirical aspects: the causal relationships believed to be operating in the study population (conceptual aspect) and data-oriented associations (empirical aspect). Other guidelines based on theories of instructional design included: actively engage the students, use visual methods when possible, and motivate the students about the importance of the topic. Students gave the method strong positive evaluations. Experts in epidemiology agreed that the 3D LOs apply generally accepted scientific views on confounding. Based on their experiences, the authors think that the 3D plots can be useful addition in the teaching of confounding. The article includes links and a downloadable file that provide a demonstration of the 3D LO-based teaching materials.
==== Body
Introduction
A major goal in teaching epidemiology is that students master the concept of confounding. They should understand when confounding may occur, how it can result in bias, and how to assess the presence of confounding and adjust for it.
As described by Rothman [1], "on the simplest level, confounding may be considered a confusion of effects. Specifically, the apparent effect of the exposure of interest is distorted because the effect of an extraneous factor is mistaken for or mixed with the actual exposure effect". (See Newman or Greenland for more fundamental definitions of confounding [2,3].) A confounding factor therefore must be: (1) a risk factor of the disease (in the unexposed), based on biological and epidemiological evidence, which requires information not included in the data; and (2) imbalanced between the exposure groups, which depends on the study design and population. In a dataset, these two criteria imply that a confounding factor must be associated with the disease and exposure. The third criterion for confounding is based on the causal relations between exposure, disease and confounding factor; this also requires information not included in the data. Rothman describes this third criterion as follows: (3) "A confounding factor must not be affected by the exposure or disease. In particular, it cannot be an intermediate step in the causal pathway between the exposure and the disease" [1].
Despite theoretical and practical work in our courses, problems in understanding confounding become clear when, in one of our courses, students analyze a dataset of a cross-sectional study. To do this, first the biological background of the exposure-outcome relation and potential confounding factors are presented. Next the students evaluate confounding using three plots: (a) of the crude association between exposure and outcome, (b) of the association between the potential confounding factor and the outcome and (c) of the association between the potential confounding factor and the exposure. Based on this information, the student must conclude whether confounding is present in the data and whether the crude association seen in the first plot provides a valid representation of the causal relationship between exposure and outcome in which the student is interested.
Communication with students indicated that knowledge of the criteria and their application to the dataset is not sufficient for understanding confounding. For example, it appeared difficult to imagine that confounding can invert the apparent direction of the effect of exposure. Several explanations of the unsatisfactory level of understanding can be put forward. One explanation is that students have to study the joint (three-dimensional) distribution of the exposure, outcome and confounding factor, but they have to use three separate (two-dimensional) plots instead of one three-dimensional plot. Obviously, simultaneously conceptualizing the three graphs requires complex cognitive processing and this could lead to cognitive overload. Another possible explanation is that most epidemiological textbooks tend to distinguish two aspects of confounding: In all textbooks, there is emphasis on a priori (prior to data collection) criteria for confounding (conceptual aspect) and on the evaluation of confounding by comparing crude and adjusted estimates (empirical aspect). The conceptual aspect focuses on background knowledge about the causal network that links exposure, outcome and potential confounders, which corresponds to the classical definition of confounding. The empirical aspect focuses on statistical associations within the data and corresponds to the collapsibility definition of confounding [2,3]. For students it seems difficult to understand how these two aspects are related.
To facilitate understanding of confounding, we developed digital learning objects (LOs) based on three-dimensional (3D) scatter plots. In the following, we describe the guidelines and requirements for the design of the 3D LOs, describe the 3D LOs and provide a hands-on example for the reader, and evaluate the results.
Analysis
Design process
Three-dimensional learning objects were designed for two courses: a BSc course (6 ECTS: European Credit Transfer System) which gives an introduction on study designs and the biases and an MSc course (6 ECTS), which focuses on data-analysis.
To direct the design process, guidelines were formulated, based on theories of instructional design (learning and teaching) and subject matter (content issues and learning goals). Students, teachers, and experts in epidemiology evaluated whether the requirements were fulfilled. In the next section, the guidelines and requirements that played a major role in the design of the 3D LOs are described. Emphasis is put on guidelines based on subject matter. Table 1 summarizes the guidelines, the requirements and the evaluators.
Table 1 Description of guidelines and requirements
Guidelines. Requirements for the 3D LOs Evaluation by
Based on subject matter and learning goals
Use rotatable 3D plots. - Students and experts perceive the 3D LOs as a valuable addition to the textbook. Students and Experts
Integrate the conceptual and empirical aspect of confounding. - Teachers confirm that the 3D LOs support the learning goals for confounding. Experts
- Experts in epidemiology confirm that the 3D LOs apply accepted scientific views on confounding. Experts
- Experts in epidemiology confirm that it is useful to use the 3D LOs in addition to epidemiological textbooks and lectures. Experts
- 80% of the students are able to answer exam questions (which integrate the conceptual and empirical approach) correctly. Evaluation of exams
Based on learning and instruction theories
Actively engage the students [26]. - Students feel that the elements in the 3D LOs that require them to become active learners help them to understand confounding. Students
Visualize important concepts. [10,11]. - Students perceive the plots in the 3D LOs as a valuable addition to the textbook. Students
- Students feel that actively manipulating the 3D plots helps them to understand confounding. Students
Motivate the students (based on ARCS model [27]): the LOs should: - Students feel that the elements that require them to become active learners motivate them to study. Students
- capture the Attention of the student, - Students judge the material with at least a 4 (on a five-point scale). Students
- be received as Relevant - Students feel they learned from the 3D LOs. Students
- induce Confidence and Satisfaction by students. - The student is able to solve the exercises. Students
Design guidelines based on subject matter
Guideline: Use rotatable 3D plots
Proving an appropriate 3D illustration of the underlying 3D relationship, to help students to understand the concept of confounding, was the primary goal of this effort. Because epidemiological analyses usually deal with higher dimensional datasets, higher dimensional visualization techniques are used to design the 3D plots. These techniques aim at viewing several variables in the same representation, using computer-supported, interactive, visual representations of abstract data, to amplify cognition [4]. Several statistical software packages (such as SAS/insight and SPSS) offer three-dimensional visualization tools, like 3D scatter plots.
Some authors have recommended 3D scatter plots as tool for understanding statistical concepts [5] and as a tool for analyzing data [6,7]. Fox et al. stated that 3D scatter plots could be potentially useful when two-dimensional plots fail to reveal structure in the data, e.g. in case of certain kinds of clustering and non-linearity [8]. In addition, Yu found that subjects performed better in detecting outliers and examination of non-linear relationship using 3D plots than using 2D plots [9]. However, in these studies non-linear functions were used, so the conclusions should not be over-generalized to linear functions. In general, the use of a 3D plot instead of three 2D plots is helpful because a relationship between three variables may not be visible in 2D plots. A 3D plot, which can be rotated by the student, provides a better view of the distribution of the three variables in the 3D space. Furthermore, by projecting three-dimensional data on a two-dimensional plane it is possible to produce 2D plots to evaluate the criteria for confounding.
Furthermore, Larkin and Sweller suggest that, when images accompany text, understanding and retention of knowledge will generally improve [10,11]. Given our experience in teaching confounding, we expect that 3D data representation may also facilitate the understanding of confounding.
Guideline: Integrate the conceptual and empirical aspect of confounding
Some epidemiological textbooks distinguish the (a priori) conceptual and (data-based) empirical aspect explicitly [1,2,12-16] while others do so implicitly [17-23]. The conceptual aspect is usually illustrated by examples of exposures, diseases, confounding factors, and non-confounding covariates. Some textbooks summarize the criteria for confounding using causal path diagrams [12,14,20,21,23-25]. The empirical aspect is usually illustrated by examples of crude and adjusted data presented in tables [1,15,20,21] or graphs [22]. In this context, stratification and regression analysis are used as tools to assess the presence of confounding and to adjust for it. None of the examples we found in epidemiological textbooks illustrates how confounding can cause reversal of the apparent effect (i.e. the reversal of the sign of the association, the side of the null on which the effect lies) although some books do mention that it is a possibility.
Many students have trouble in connecting the two aspects of confounding when confronted with a real dataset. Therefore, we consider it important to integrate the two aspects of confounding in our teaching. This is achieved, in the 3D LOs, by visualizing that both aspects originate from the same 3D representation of the data. Our method integrates these aspects by illustrating that manipulating the association between the exposure and the confounder results in different crude associations (empirical aspect), although they are derived from the same underlying relationships (conceptual aspect).
Design guidelines based on learning and instruction theories
The most important guidelines for the development of the 3D LOs, based on theories about learning and instruction, are summarized in this section.
Guidelines: Actively engage the student in studying confounding
The first guideline is to actively involve the student, because practice is believed to strengthen understanding [11,26]. In the 3D LOs, we will involve students in studying confounding with activities that include answering questions, performing simulations, and projecting data on one surface of the plot. In later applications of these methods we used self-tests to help clarify for students what was most important in the 3D LOs. Using these self-tests, the student could verify whether he understand the meaning of the different characteristics of the 3D LOs by interpreting some other examples of epidemiological data visualized in 3D plots.
Guidelines: Use visual methods when possible
A second guideline is to visualize important concepts. Besides visualizing the concept of confounding by using 3D plots, other visual methods are also used in the exercises that accompanied the 3D plots. For example, in the exercises, causal path diagrams are used to emphasize the causal relation between fiber intake, blood pressure and bodyweight.
Guidelines: Motivate the students
The last guideline is to motivate the students. Motivation is essential to learning. According to the ARCS model, four factors are essential to motivate the students: Instruction should capture the Attention of the student, it should be perceived as Relevant, and it should induce Confidence and Satisfaction [27]. From this principle, guidelines for the design of digital learning material were derived (see Table 1). The attention of the student is drawn by providing novelty (e.g., the 3D plots and several pictures). The relevance of the subject matter is shown by emphasizing the importance of the concept of confounding: the example used in the LOs illustrates the case where failure to adjust for confounding could lead to the conclusion that the effect of an exposure is in the opposite direction of the true relationship. Providing hints and gradually building up the difficulty of the exercises enhances students' confidence and satisfaction in understanding the concepts. For example, in the first 3D LO, several questions with hints are provided while in the third LO students are expected to explore the 3D plot by themselves. This third LO gives also the possibility to test skills that are attained in the first LOs.
Requirements and evaluation
Students evaluated how well the teaching method fulfilled these guidelines in the BSc and MSc courses at our university, and in an international PhD course organized by our university. At our university students' perception of the quality of courses, course material and teachers was assessed with standard evaluation forms using agree-disagree questions on a five-point Likert scale. An average appreciation score of 3 on these evaluation forms is considered satisfactory while an average higher than 4 is considered excellent. The 3D LOs were specifically evaluated using such evaluation forms. In addition, exam results of students were analyzed to get an indication of their understanding of confounding.
For the evaluation with experts, evaluation forms with disagree-agree questions on a five-point Likert scale and free response questions were used. The experts worked through the 3D LOs and the exercises as if they were students. They were also asked to focus particularly on whether they think the 3D LOs apply accepted scientific views on confounding. Before this formal evaluation, three of our PhD students and two teachers evaluated the 3D LOs. This resulted in some minor improvements
Description of the 3D LOs
The following is a description of one of the 3D LO-based lessons we used in our courses. It is based on data from (hypothetical) studies on the relation between fiber intake and blood pressure conducted in three different populations. Body weight is chosen as the potential confounding factor, because it is known to be a risk factor for high blood pressure. We constructed the example so that body weight is not an effect modifier. Each 3D LO starts with a rotatable 3D plot with the outcome (blood pressure) on the y-axis, exposure (fiber intake) on the x-axis, and the possible confounding factor (body weight) on the z-axis. In all the 3D LOs, the values of blood pressure, fiber intake and body weight are chosen so that body weight is a risk factor for high blood pressure and fiber intake is negatively associated with blood pressure. Only the association between fiber intake and body weight differs between the three plots.
In all plots the data can be projected on one side (plane) of the plot, so each plot illustrates:
1. The joint distribution of the three variables together: In all plots visualized by the linear plane fitted to the data (BP = β0 + β1 * fiber intake + β2 * body weight + error) (Figure 1),
2. That body weight is a risk factor for high blood pressure (β2) (Figure 2),
3. The adjusted association between fiber intake and blood pressure (β1),
4. The association between fiber intake and body weight (differs between the LOs) (Figure 3),
5. The crude association between fiber intake and blood pressure, illustrated by a regression line through the projection of the data on the fiber-blood pressure side of the plot (Figure 4),
6. The association between fiber intake and blood pressure stratified for body weight (a slider can be used to highlight only data within a certain stratum of body weight).
The learning material consists of three parts, containing a 3D plot and some exercises. Figure 2 shows the main characteristics of the 3D plot as visualized in the second part of the learning material (the second LO).
Figure 1 Illustrations of results from the example exercise (see text for instructions for running the exercise). Joint distribution of exposure (fiber intake), effect (high blood pressure), and potential confounder (body weight).
Figure 2 Illustration of results from the example exercise (see text for instructions on running the exercise). Projection of the data on the weight-blood pressure plane: weight risk is a risk factor for high blood pressure.
Figure 3 Illustration of results from the example exercise (see text for instructions on running the exercise). Projection of the data on the fiber intake–weight plane: fiber intake and weight are negatively associated.
Figure 4 Illustrations of results from the example exercise (see text for instructions for running the exercise). Projection of the data on the fiber intake–blood pressure plane: the crude association (the
slope of the line) differs from the adjusted association (the slope of the plane).
Figure 5 Examples of questions used to help students explore the characteristics of the 3D LOs
The 3D plot in the first LO represents data from a study in which fiber intake is independent of body weight. This LO illustrates the case where the apparent association between fiber and blood pressure is not confounded by the blood-pressure-increasing effect of body weight. In all LOs we assume that the effect of fiber intake on blood pressure is not mediated by body weight (criterion 3 for confounding [1]).
The second LO (Figure 1,2,3,4) and the third LO show that confounding arises when fiber intake and body weight are associated positively or negatively. For the second 3D LO, subjects with high fiber intake tend to have a lower body weight, perhaps because they are more health conscious. In the second 3D LO, the crude association (the slope of the line resulting from projecting the data to the fiber-blood pressure plane) differs from the adjusted association (the slope of the regression plane, β1) so body weight is a confounding factor (Figure 4). The reader can access the second 3D LO presented in this paper, as well as other examples, at our website [28]. (See endnote 1 for more information about the website and instructions on how to use the file published with this article which contains a version of what is on the website.)
In the third 3D LO, results of another (hypothetical) study shows how body weight reverses the apparent effect of fiber intake on blood pressure, when fiber intake and body weight are strongly positively associated.
Practical experiences with the 3D LOs and results of evaluations
Evaluation by students
The 3D LOs are used in our BSc course (104 students, from which 100 filled out the evaluation forms), MSc course (in two subsequent years, in total 44 students) and an international PhD course organized by our university (19 students). Evaluation forms were used to assess the judgments of the students. As indicated in Table 2 the students judged the 3D LOs with a 3.7, 4.5 and 4.2 (on a five-point scale). The value of these student evaluations are limited by the lack of validation of the instrument, a clear definition of what the scores mean, and most importantly, the fact that few of these students had experience learning the material using other teaching tools, so they had nothing to compare this method to. Nevertheless, we interpret the scores as support for the value of this teaching method.
Table 2 Results of evaluation with students
Mean score (% with a score of 4 or 5)
Evaluation question* BSc course (n = 100) MSc course (n = 44) International PhD course (n = 19)
1. The 3D plots help me to understand confounding. 3.6 (60) 4.4 (92) 4.2 (89)
2. It was useful to work with the 3D plots in addition to the lectures and textbook. 3.7 (68) -† -†
3. I enjoyed studying confounding using the 3D plots. 3.4 (53) 4.6 (100) 4.7 (100)
4. Active handling the 3D plots helps me to understand confounding. 3.5 (52) 4.5 (100) 4.2 (100)
5. The self-tests were useful. - ‡ 4.6 (100) - ‡
6. Overall rating of the 3D plots (1 = poor to 5 = excellent). 3.7 (64) 4.5 (100) 4.2 (95)
*All questions were Disagree – Agree questions with a five-point Likert scale. As indicated an average score of 3 is considered satisfactory while an average higher than 4 is considered excellent.
† In the MSc and PhD course this question was not included on the evaluation form because there was no additional learning material provided about confounding.
‡ Self-tests were only available in the MSc and PhD course.
To get an indication of the level of competence attained by the students, exam results were analyzed. The exam questions were different for the BSc and MSc course. As indicated in Figure 6 the students scored well for the exam; for each question in the BSc course 66% or more of the students gave the right answer. The questions about the integration of the conceptual and empirical aspect of confounding appear the most difficult ones (question 6 and 7). In the MSc course, in two multiple-choice questions descriptions of epidemiological studies must be combined with plots that show the data of the studies. On these questions, respectively 83 and 75% of the students gave the correct answer. Although the same exam questions were not asked in the past, this rating is considerably better than the results from similar exam questions on the same topic that were asked in the past.
Figure 6 Example of exam question and summary of exam results
Illustration of the usefulness of the method to the students came in the MSc course, where students further practiced with 3D plots during the analysis of a cross-sectional study. Most of the students took advantage of the opportunity to consult the 3D LOs again during the data-analysis. From our experiences in previous years, it seems that during this MSc course students who were taught using the 3D LOs had a better understanding the concept of confounding and multiple regression as a method to adjust for confounding than previous years (though we concede that this evaluation suffers from the usual problems of non-blinded evaluators who are invested in the outcome). Students asked questions that are more advanced. For instance, many students extrapolated the method to effect modification by describing how a 3D plot would look like in the presence of effect modification.
Since the courses in which the 3D LOs were used and similar courses in which they were not used differ from year to year with respect to specific topics, learning material, form of the exam, number of students, prior knowledge of students, etc., it is not possible to investigate precisely the effect of the 3D LOs (as it would had we been able to do a clean and large scale randomized study). This is a well-known challenge in educational research [29]. Therefore, rather than relying too much on the students' demonstrated learning and own evaluations of the methods, we base much of our evaluation on the more indirect method of assessing how well 3D LOs fulfilled the above guidelines and how experts evaluated them.
Evaluation by experts in epidemiology
Eight experts in epidemiology reviewed the 3D LOs; seven were teachers at Dutch universities and one at a non-Dutch university. Six of them filled in the evaluation form while two only responded by giving a general opinion about the 3D LOs. The experts were not involved in the design of or teaching using the 3D LOs. Table 4 summarizes the scores on the evaluation questions. In addition, the experts responded to some open-ended questions. The results suggest that the experts agree that the 3D LOs apply generally accepted scientific views on confounding and should enhance understanding of confounding. However, two experts expressed concern that the 3D LOs would not be helpful for some students who have difficulties with interpreting 3D objects. Three experts suggested that we develop additional learning material explaining the difference between confounding and effect modification. There were also suggestions that the issue of causality in relation to the third criterion [1] for confounding needed further explanation, which we have added (though this change came subsequent to the students' experience with the learning material).
Table 4 Evaluation of the 3D LOs by experts in epidemiology
Evaluation question* Mean Score (n = 6)
1. I think the students like the module. 4.3
2. The questions in this modules where clear and understandable 4.8
3. It is useful that the 3D plots are rotatable 3.0
4. The questions in this module are useful 4.8
5. I think that this module applies general accepted scientific views on confounding 4.5
6. I think that the use of 3D plots enhanced understanding of confounding by students 4.0
7. I think that this modules provides a useful addition to epidemiological textbooks and lectures 4.2
8. I think that this module stimulated the student to study confounding 3.8
9. I think that this module is useful in my own course. 3.8
10. Overall rating of the module. 3.8
*All questions were Disagree – Agree questions with a five-point Likert scale.
Conclusion
Recently, other graphical approaches to teaching confounding have been described [30,31]. Unlike our 3D LOs, these approaches address confounding without the use of multivariate regression techniques. Therefore, the approaches could be useful to introduce the concept of confounding and to make the students aware of the importance of considering possible confounders. These approaches do not directly address the relation between the criteria for confounding (conceptual aspect) and the effect of the confounder on the studied exposure-outcome relation (empirical aspect), as do the 3D LOs. Thus, the 3D LOs seem to be more useful at an intermediate level, preparing the students for epidemiological data analysis. Therefore, we think the approaches could complement each other.
Teaching tools using 3D plots are potentially useful in illustrating effect modification, non-linearity in datasets [8], and other relationships of three variables. We plan to design additional learning material contrasting confounding and effect modification. In addition, 3D plots can be useful in teaching other epidemiological principles. For example, how measurement errors in the confounding factor, exposure variable, or outcome variable can lead to, respectively, residual confounding, bias toward the null, or decrease of precision. We will make revisions of the current method and additions of other concepts in our 3D LOs available at our website [28].
Our first experience with the 3D LOs indicate that the integration of the conceptual and the empirical aspect of confounding stimulate the student to think beyond confounding. Although it might be possible that the 3D LOs will not be helpful for some students (e.g. students who have difficulties with interpreting 3D objects) we think that, based on our experiences, the 3D LOs can provide a valuable addition to standard epidemiological textbooks and other graphical presentations of confounding for most students.
Endnotes
1. To ensure the existence of a permanent archive, the website that contains the example emphasized in this article has been published with the article as an additional file (however, the website is easier to use, more extensive, and will contain subsequent versions of the software, and thus we recommend readers access it at if possible rather than using the additional file). To use the additional file, download the .zip file, unzip it to a folder, and run (double click on) index.html.
Note that to run either the web or local version of this demo requires the Macromedia Flash player browser plug-in, which you probably have, as well as a plug-in for viewing 3D images (Cortona from Parallel Graphics) that you will likely need to install. These are free and the index page contains links that will let you install them. We apologize that in its present form, our software will not work with all browsers, security configurations, etc. We recommend the use of Microsoft Internet Explorer and it will be necessary to turn off pop-up blockers. The index page contains a link to check your system's compatibility.
List of abbreviations
2D, Two-dimensional.
3D, Three-dimensional.
BSc, Bachelor of Science.
ECTS, European Credit Transfer System.
LO, learning object.
MSc, Master of Science.
PhD, Postdoctoral.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
MCB designed, developed and evaluated the 3D LOs and led the writing of the manuscript, but all three authors contributed to editing and revision. RH initiated the project and provided the initial arguments for investing in the development of some form of 3D visualization; he reviewed the LOs from an educational point of view. PvtV contributed in the design of the 3D LOs and reviewed the epidemiological content of the 3D LOs.
Acknowledgements
We would like to thank H van der Schaaf for technical implementation of the 3D LOs, E Kampman, E G Schouten and J Burema for a critical discussion of the 3D LOs during the early stages of the design process and assistance during the evaluation of the 3D LOs. In addition, we would like to thank teachers and experts in epidemiology from outside Wageningen University for critical reviewing the 3D LOs.
==== Refs
Rothman KJ Greenland S Modern Epidemiology Philadelphia 1998 Philadelphia, Lippincott Williams & Wilkins
Newman SC Commonalities in the classical, collapsibility and counterfactual concepts of confounding. Journal of Clinical Epidemiology 2004 57 325 329 15135831 10.1016/j.jclinepi.2003.07.014
Greenland S Robins JM Pearl J Confounding and collapsibility in causal inference. Statistical Science 1999 14 29 46 10.1214/ss/1009211805
Card SK Mackinlay JD Shneiderman B Card SK, Mackinlay JD and Shneiderman B Information visualization Readings in information visualization: using vision to think 1999 San Francisco, Morgan Kaufmann Publishers 1 34
Monette G Fox J and Long JS Geometry of multiple regression and interactive 3-D graphics Modern methods of data analysis 1990 Newbury Park, Sage 209 256
Huber PJ Experiences with three-dimensional scatter plots Journal of the American Statistical Association 1987 82 448 453
Cook RD Regression Graphics: ideas for studying regression through graphics. 1998 New York, John Wiley & Sons
Fox J Stine R Monette G Vohra N Detecting clusters and nonlinearity in three-dimensional dynamic graphs Journal of computational and graphical statistics 2002 11 875 895 10.1198/106186002321018849
Yu C The interaction of research goal, data type, & graphical format in multivariate visualization 1995 Tempe, Arizona state university
Larkin JH Simon HA Why a diagram is (sometimes) worth ten thousand words Cognitive Science 1987 11 65 99 10.1016/S0364-0213(87)80026-5
Sweller J van Merriënboer JJG Paas FGWC Cognitive architecture and instructional design Educational Psychology Review 1998 10 251 296 10.1023/A:1022193728205
Hernán MA Hernández-Díaz S Werler MM Mitchell AA Causal knowledge as a prerequisite for confounding evaluation: An application to birth defects epidemiology American Journal of Epidemiology 2002 155 176 184 11790682 10.1093/aje/155.2.176
Rothman KJ Epidemiology, an introduction 2002 New York, Oxford University Press
Szklo M Nieto FJ Epidemiology: Beyond the basics 2000 Gaithersburg, Aspen Publishers
Kleinbaum DG Kupper LL Morgenstern H Epidemiologic research: principles and quantitative methods. 1982 New York, Van Nostrand Reinhold
Kleinbaum DG Whyte D ActivEpi 2002 New York, Springer Verlag
Kelsey JL Thompson WD Evans AS Methods in observational epidemiology 1986 Oxford, Oxford University Press
Breslow NE Day NE Statistical methods in cancer research: Vol 1- the analysis of case-control studies. IARC scientific publications No 32 1980 Lyon, International agency for research on cancer
Miettinen OS Theoretical epidemiology: principles of occurence research in medicine. 1985 New York, John Wiley & Sons
Breslow NE Day NE Statistical methods in cancer research: Vol II- the design and analysis of cohort studies IARC scientific publications No 32 1987 Lyon, International agency for research on cancer
Schlesselman JJ Case-Control Studies: design, conduct, analysis 1982 New York, Oxford University press
Margetts BM Nelson M Design concepts in nutritional epidemiology 1997 2 Oxford/New York, Oxford University Press
Hennekens CH Buring JE Epidemiology in medicine 1987 Boston/Toronto, Little, Brown and Company
Beaglehole R Bonita R Kjellström T Basic Epidemiology 1993 Geneva, World Health Organization
Ahlbom A Norell S Introduction to modern epidemiology 1984 2 Chestnut Hill, Epidemiology resources
Anderson JR Learning and memory: An integrated approach 1995 New York, Wiley
Keller JM Development and use of the ARCS model of motivational design Journal of instructional development 1987 10 2 10
Demo Version of the 3D LO's.
Collis B Moonen J Flexible Learning in a Digital World: Experiences and Expectations 2001 London, Kogan Page Limited
Vander Stoep A A didactic device for teaching epidemiology students how to anticipate the effect of a third factor on an exposure-outcome relation American journal of epidemiology 1999 150 221 10412971
Wainer H The BK-plot: making the Simpsons's paradox clear to the masses. Chance magazin 2002 15 60 62
|
15953393
|
PMC1192815
|
CC BY
|
2021-01-04 16:36:38
|
no
|
Epidemiol Perspect Innov. 2005 Jun 14; 2:6
|
utf-8
|
Epidemiol Perspect Innov
| 2,005 |
10.1186/1742-5573-2-6
|
oa_comm
|
==== Front
Int Semin Surg OncolInternational seminars in surgical oncology : ISSO1477-7800BioMed Central London 1477-7800-2-141611148310.1186/1477-7800-2-14Case ReportPneumothorax after a clinical breast fine-needle aspiration of a lump in a patient with Poland's syndrome Salhab M [email protected] Sarakbi W [email protected] N [email protected] K [email protected] The Princess Grace Hospital, London, UK2005 19 8 2005 2 14 14 20 7 2005 19 8 2005 Copyright © 2005 Salhab et al; licensee BioMed Central Ltd.2005Salhab et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
We report the first case in the medical literature of a pneumothorax complicating fine needle aspiration cytology (FNAC) of a breast lump in a woman with a mild form of Poland's syndrome. The pneumothorax was treated conservatively. This is the first case of breast FNA-related pneumothorax seen in our clinical practice. We believe that the absence of pectoral muscles has increased the risk of this complication. We have also diagnosed an incidental screen-detected breast cancer affecting the ipsilateral breast in the same patient. We conclude that caution should be exercised when performing FNAC of breast lesions in patients with Poland's syndrome. The procedure should be preferably performed under image guidance in such patients in order to minimise the risk of this complication.
PneumothoraxFNACbreast cancer and Poland's syndrome.
==== Body
Introduction
Fine needle aspiration cytology (FNAC) of the breast is a minimally invasive, safe, fast and cost effective technique which provides diagnostic information as to the nature of a breast lesion and can be used in appropriate settings to allow rapid management planning.
Pneumothorax occurring after FNAC of the breast is a recognized complication that has been reported in the literature. It was first described by Orr and Margarey in 1978 [1]. An Italian study of more than 200 000 FNAC procedures of the breast showed that pneumothorax occurred in 1 in 10000 cases (0.01%), however, the authors conceded that this figure might be underestimated due to unrecognized and asymptomatic cases of pneumothoraces [2,3].
Other studies have reported higher incidence of this complication; one in 417 by Kaufman [4] and one in 1000 by Gateley [5].
In this case we report a pneumothorax after FNAC of a breast lump in a woman with Poland's syndrome.
Case report
A 52 years old female attended the breast clinic with a cystic mass superior to the left nipple. Clinical FNAC of this mass was performed (using a 23 gauge needle and 10 cc syringe) yielding 5 cc of straw-coloured fluid. Subsequently the patient had digital mammography. Thereafter, she complained of a difficulty in breathing and a subsequent chest X ray confirmed the presence of left-sided pneumothorax (Figures 1, 2) which treated with per-cutaneous aspiration. The patient was admitted overnight for observation and a follow up chest X-ray showed resolution of the pneumothorax. The post-procedure mammogram showed an area of architectural distortion and irregularity in the medial aspect of the left breast which was regarded as suspicious. Three days later, this was subjected to an ultrasound guided core biopsy of the lesion and histology showed severe atypical hyperplasia. Ultrasound guided excision of the lesion was performed and the final histology showed radial scar, a 7 mm grade I infiltrating ductal carcinoma associated with ductal carcinoma in situ (DCIS). The surgical margins were clear. The patient subsequently had a sentinel node biopsy using the dual localization technique under local anaesthesia.
Figure 1 CXR in inspiration showing left sided small pneumothorax.
Figure 2 CXR in expiration showing obvious left sided pneumothorax.
Digital mammography failed to adequately demonstrate the left pectoralis major muscle due to hypoplasia (Figure 3). Clinical examination confirmed the absence of the costo-sternal portion of the pectoralis major on the left side.
Figure 3 Digital mammogram (Mediolateral view) showing absence of the pectoralis major muscle and architectural distortion on the left side and normal right breast.
Discussion
Pneumothorax as a complication of fine needle aspiration of the breast has been reported regularly in the medical literature over the last three Decades [1-10]. Pnemothorax is a relatively rare complication of breast FNAC partly due to the fact that the ribs form a much larger area of the chest wall surface than the intercostals muscles. Furthermore, the pectoral muscles provide additional protective structures. The reported incidence of this complication varies between 3 in 100 and 1 in 10000 [11]. Such an iatrogenic pneumothorax is usually treated conservatively without thoracostomy and drainage [12].
It is essential to perform the FNAC correctly in order to obtain a representative cytological sample from breast masses and avoid complications. To minimise the risk of developing pneumothorax after breast FNAC, it was suggested that the needle should be introduced parallel rather than perpendicular whenever possible to the chest wall therefore, avoiding overshooting of the needle tip which might pierce the intercostal muscles and injure the pleura and the underlying lung [13]. Patients should be encouraged to breathe normally while the procedure is being performed [4]. Furthermore it is advisable that patients with deep, centrally located and small or nonpalpable lesions have their FNAC under ultrasound-guidance. This is a safe and highly accurate procedure [14,15]. Pneumothorax after breast FNAC is seen more frequently in patients who have thin bodies and small sized breasts [3]. Furthermore, patients with deeply-located central lesions [4,8], or lesions in the tail of the breast [5] are also at higher risk for developing this complication.
In our case, the patient has an absence of the costo-sternal portion of the pectoralis major muscle on the same side where the breast FNA was performed, therefore the thickness of the chest wall behind the breast was much smaller than the contra-lateral side, therefore increasing the risk of inadvertent pneumothorax. Our patient has no obvious deformities. However, subsequent mammogram and clinical examination showed an absence of the pectoralis major; a case which is related to a rare condition called Poland's syndrome.
In 1841, Alfred Poland described an autopsy report of a 27 years old male with absence of the sternocostal portion of the pectoralis major, complete absence of pectoralis minor, hypoplasia of the serratus anterior and external oblique and defects in the middle phalanges [16]. In 1967 Baudinne et al coined the term Poland's syndrome [17]. This condition is very rare with a reported incidence of one in 20 000 – 32 000 live births and a 3:1 male predominance [18,19].
The clinical manifestations of Poland's syndrome vary, but typically it is characterized by a combination of hypoplasia or absence of the breast, nipple -areola complex and or the subcutaneous tissue [20], hypoplasia or absence of the costosternal portion of the pectoralis major muscle, serratus anterior or external oblique; absence of pectoralis minor muscle and absence of the costal cartilages of ribs 2,3,4 or 3,4,5. In addition, some patients have deformities of the hand and the upper limb [21].
Our patient has a minor manifestation of Poland's syndrome; the absence of the costo-sternal portion of the pectoralis major is a characteristic feature of this syndrome and is found in 100% of cases [19].
The association between Poland's syndrome and malignancy has been previously reported. Leukaemia, non-Hodgkin lymphoma, cervical cancer, leiomyocarcome, Wilms tumour and lung cancer have been reported in association with Poland's syndrome [22-27]. Furthermore, breast cancer has also been observed in females with this syndrome despite having mammary hypoplasia [28-34]. Interestingly, hypoplastic breasts may develop breast cancer similarly to normal breasts as observed by Havlik et al [28]. This observation raises important questions regarding a possible association between the two entities.
Pneumothorax in patients with Poland's syndrome has been previously reported but not in association with breast FNAC. Luh et al described two patients with Poland's syndrome anomalies presented with spontaneous pneumothorax. To the best of our knowledge our case is the first report of pneumothorax after FNAC of a breast lump in a woman with Poland's syndrome. We have also diagnosed an incidental screen-detected breast cancer affecting the ipsilateral breast.
Reconstructive surgery after mastectomy for breast cancer in patients with Poland's syndrome is challenging due to the absence or hypoplasia of the chest wall muscles thus necessitating the need to use myocutaneous flaps such as the latismus dorsi flap if implants are used. The TRAM, GAP and DEIP flaps represent alternative reconstructive techniques. Furthermore the subsequent use of radiotherapy in these patients carries a higher risk of lung complications due to the decreased protection by chest wall muscles.
In conclusion Poland's syndrome may be associated with an increased risk pneumothorax complicating FNAC of the breast. Therefore, this procedure should be carefully performed in such patients and preferably with image guidance in order to minimise the risk.
==== Refs
Orr KB Margarey CJ Pneumothorax after aspiration of breast cysts Med J Aust 1978 1 101 651697
Catania S Boccato B Bono A Di Pietro S Pilotti S Ciatto S Pneumothorax: a rare complication of fine needle aspiration of the breast Acta Cytol 1989 33 140 Letter 2916363
Catania S Veronesi P Marassi A Pluchinotta A Bono A Zurrida S Risk of pneumothorax after fine needle aspiration of the breast. Italian experience of more than 200 000 aspirations Breast 1993 2 246 7 10.1016/0960-9776(93)90008-4
Kaufman Z Shpitz B Shapiro M Dinbar A Pneumothorax. A complication of fine needle aspiration of breast tumours Acta Cytol 1994 38 737 8 8091907
Gateley CA Maddox PR Mansel RE Pneumothorax: a complication of fine needle aspiration of the breast BMJ 1991 303 627 8 1932905
Chen KT Tschang TP Pneumothorax: a complication of fine needle aspiration of breast tumors Acta Cytol 1994 38 737 8 8091907
Dixon JM Pneumothorax after fine needle aspiration of the breast BMJ 1991 303 924 Letter
Arisio R Carbone G Maina A Donvito V Pneumothorax as a complication of fine needle aspiration of the breast Panminerva Med 1988 33 140
Berthiot G Perotin D Quinquenel MC Body G Baudouin R IatrogenicPneumothorax after puncture of the breast Rev Pneumol Clin 1985 41 201 3 4048751
Stewart LH Pneumothorax following breast aspiration Ulster Med J 1990 59 211 12 2278121
Bates T Davidson T Mansel RE Litigation for pneumothorax as a complication of fine-needle aspiration of the breast Br J Surg 2002 89 34 7
Brown KT Brody LA Getrajdman GI Napp TE Outpatient treatmentof iatrogenic pneumothorax after needle biopsy Radiology 1997 205 249 52 9314993
Quercidella Rovere G Benson JR Childs P Hastings L Johri A Is the tangential or parallel approach to FNA cytology of breast lesions always possible and compatible with reliable sampling Breast 2001 10 352 5 10.1054/brst.2001.0288
Broderick LS Kopecky KK Cramer H image-guided coaxial fine needle aspiration biopsy with a side existing guide J comput Assisst Tomogr 2002 26 292 7 10.1097/00004728-200203000-00023
Liao J Davey DD Warren G Davis J Moore AR Samayoa LM Ultrasound-guided fine-needle aspiration biopsy remains a valid approach in the evaluation of nonpalpable breast lesions Diagn Cytopathol 2004 30 325 31 15108230 10.1002/dc.20068
Poland A Deficiency of the pectoral muscles Guy's Hosp, Bull 1841 6 191 3
Baudinne P Bovy GI Wasterlain A A case report of Poland's syndrome Acta Paediat Belgica 1967 21 407 10
Freire-Maia N Chautard EA Opitz JM The Poland syndrome: clinical and genealogical data, dermatoglyphic analysis, and incidence Hum Hered 1973 23 97 4356989
Mace JW Kaplan JM Schanberger JE Gorlin RW Poland's syndrome: report of seven cases and review of literature Clin Paediatr 1972 11 98 102
Perlyn C Edmiston J Tunnessen WW Jr Picture of the month, Unilateral amastia (Poland Syndrome) Arch Paediatr Adolesc Med 1999 135 1305 6
Gausewitz Sh Meals RA Setoguchi Y severe limb deficiency in Poland's syndrome Clin Orthop 1984 185 9 13 6323086
Enzenauer RW Hasting CP Leukaemia and absent of pectoralis major. No association? Am J Dis Child 1981 135 763 5 6943931
Esquembre C Ferris J Verdeguer A Prieto F Badia L Castel V Poland syndrome and leukaemia Eur J Paediatr 1987 146 444 10.1007/BF00444964
Hershatter BW Montana GS Poland's syndrome and lymphoma Am J Dis Child 1983 137 1211 2 6314810
Shaham D Ramu N Bar-Ziv J Leiomyosarcoma in Poland's syndrome. A case report Act Radiol 1992 33 444 6
Athale UH Warrier R Poland's syndrome and Wilms tumor: an unusualassociation Med Paediatr Oncol 1998 30 67 8 10.1002/(SICI)1096-911X(199801)30:1<67::AID-MPO16>3.0.CO;2-5
Ahn MI Park SH Park YH Poland's syndrome with lung cancer. A case report Acta Radiol 2000 41 432 4 11016761 10.1080/028418500127345875
Havlik RJ Sian KU Wagner JD Breast cancer in Poland's syndrome Plast Reconstr Surg 1999 104 180 2 10597692
Katz SC Hazen A Colen SR Roses DF Poland's syndrome and carcinoma of the breast: A case report Breast J 2001 7 56 9 11348417 10.1046/j.1524-4741.2001.007001056.x
Khandelwal A O'Hea BJ Garguilo G Breast cancer in a patient withPoland's syndrome Am surg 2004 70 491 5 15212400
Tamiolakis D Venizelos D Antoniou C Tsiminikakis N Alifieris E Papadopoulos N Breast cancer development in a female with breast cancer syndrome Onkologie 2004 27 569 71 15591718 10.1159/000081341
Wong TC Lim J Lim TC A case of ductal carcinoma in situ of breast with Poland syndrome Ann Acad Med Singapore 2004 33 382 4 15175787
Okamo H Miura K Yamane T Fujii H Matsumoto Y Invasive ductal carcinoma of the breast associated with Poland's syndrome: report of a case Surg Today 2002 32 257 60 11991512 10.1007/s005950200030
Fukushima T Otake T Yashima R Nehei M Takeuchi S Kimijima II Tsuchiya A Breast cancer in two patients with Poland's syndrome Breast cancer 1999 6 127 130 11091704
|
16111483
|
PMC1192816
|
CC BY
|
2021-01-04 16:38:37
|
no
|
Int Semin Surg Oncol. 2005 Aug 19; 2:14
|
utf-8
|
Int Semin Surg Oncol
| 2,005 |
10.1186/1477-7800-2-14
|
oa_comm
|
==== Front
Malar JMalaria Journal1475-2875BioMed Central London 1475-2875-4-281598750810.1186/1475-2875-4-28ResearchThe quality of antimalarials available in Yemen Abdo-Rabbo Ahmed [email protected] Amal [email protected] Hoda [email protected] Department of Pharmacology and Therapeutics, Faculty of Medicine and Health Sciences, University of Sana'a, Republic of Yemen2 Division of Communicable Diseases, Eastern Mediterranean Regional Office of the World Health Organization, Abdul Razzak Al Sanhouri Street, P.O.Box 7608, Nasr City, Cairo 11371, Egypt2005 29 6 2005 4 28 28 27 1 2005 29 6 2005 Copyright © 2005 Abdo-Rabbo et al; licensee BioMed Central Ltd.2005Abdo-Rabbo 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
Malaria has always been a major public health problem in Yemen. Several studies in developing countries have demonstrated ineffective and poor quality drugs including antimalarials. Therefore, quality assessment of antimalarial drugs is of crucial importance. This study aimed to assess the quality of antimalarials (chloroquine and sulfadoxine/pyrimethamine) available in Yemen and to determine whether the quality of these products was related to the level of the distribution chain at which the samples were collected or related to the manufacturers.
Methods
Four samples from each antimalarial product were collected from each of the various levels of the distribution chain. One sample was kept with the research team. Two were tested at Sana'a and Aden Drug Quality Control Laboratories. The fourth was sent to the Centre for Quality Assurance of Medicines in Potchefstroom, South Africa, for analysis. Quality indicators measured were the content of the active ingredient and dissolution rate (for tablets only) in comparison to standard specifications for these products in the relevant pharmacopoeia.
Results
The results identified several problems of sub-standard products within the drug distribution chain. They included high and low failures in ingredient content for chloroquine tablets and chloroquine syrup. There was some dissolution failure for chloroquine tablets, and high sulfadoxine/pyrimethamine tablets dissolution failures. Failures with the dissolution of the pyrimethamine were found at most of the collection points. No clear relationship neither between the quality products and the level of the distribution chain, nor between locally manufactured and imported products was observed.
Conclusion
There are sub-standard antimalarial products circulating within the drug distribution chains in the country, which will have serious implications on the reduced therapeutic effectiveness and on the development of drug resistance. This appears to be due to non-compliance with Good Manufacturing Practice guidelines by manufacturers in the production of the antimalarials.
==== Body
Background
Malaria has always been a major public health problem in Yemen. About half of the population are at risk of malaria and there are several hundred deaths every year [1].
A preliminary, exploratory meeting of experts and other concerned parties was held in Geneva in October 1998. The meeting concluded, that poor quality antimalarials (AMs) posed a problem for malaria control, and therefore to public health, especially in countries where there is little or no drug regulatory infrastructure.
Treatment failure, associated with drug resistance may also be due to poor quality products, instability of products, or the use of counterfeit products. Several WHO-sponsored studies have demonstrated significantly the instability of products such as ergometrine, [2-4] and other essential medicines, during transport by sea, and also during road transport inland [5,6]. A WHO document on accelerated stability studies under simulated tropical conditions provides some indication of intrinsic stability of commonly used drugs [7]. A study in Sudan on stability of drugs in the tropics showed a considerable reduction in the concentration of the liquid dosage forms of certain preparations, whereas solid dosage forms were relatively stable [8]. A recent study in selected African countries identified several significant problems of sub-standard AM products within the drug distribution chain [9].
The problem of counterfeit drugs is well recognized in different parts of the developing world [10]. The use of ineffective and products of poor quality will not only endanger therapeutic treatment but also erode public confidence in national health programmes.
In most countries the quality of AMs is rarely independently evaluated and the local capacity for independent drug-quality assurance is worst where the disease burden is the highest. AMs of poor quality might contribute to the emergence of resistance. It is therefore, important to consider product quality when dealing with the problem of antimalarial resistance.
In Yemen, drug registration is overseen by the Supreme Board of Drugs and Medical Appliances (SBD&MA). There are two national reference laboratories for drug quality control, one in Sana'a and the other in Aden.
The Drug Fund (DF) tender for public sector drugs centrally. Suppliers deliver drugs to the four public regional stores in the country. Lahej Governorate obtain their medical supplies from Aden regional medical stores (RMS) through Lahej central medical stores (CMS). The existing district medical stores receive supplies from Lahej CMS and distribute them to the health facilities within their district. Sometimes drugs may be supplied directly from Lahej CMS to the health facilities.
In view of the need to clarify the nature and magnitude of the problem of AM quality in the country, it was decided to carry out a study in a selected Governorate to evaluate the extent of the problem and guide designing suitable intervention strategies.
This study aimed at assessing the quality of the commonly used AMs in Yemen, particularly, chloroquine and sulfadoxine/pyrimethamine and to determine whether the sub-standard quality of these products was related to the level of the distribution chain at which the samples were collected or to the manufacturer. It also aimed at evaluating the performance of the national reference laboratories, by comparing their results to those obtained from an international reference laboratory in South Africa.
Methods
Chloroquine phosphate 250 mg tablets (CQT), chloroquine base 50 mg/5 ml syrup (CQS), and sulfadoxine/pyrimethamine 500 mg/25 mg tablets (SPT) were sampled and evaluated. They were chosen because they are the most widely used AMs in Yemen and included in Yemen Treatment Guidelines and Essential Drugs List [11].
The study adopted the same methods used by the WHO to assess the quality of antimalarials in malaria endemic countries [9]. Therefore, samples were collected at various levels of the drug distribution chain in the public and private sector in Lahej Governorate, which is one of the highest malaria-endemic areas in Yemen. Samples were collected from 3 medical stores, the Governorate hospital, 4 district hospitals/health centres, 9 health units and 5 private sector pharmacies. The total samples collected from all levels of the drug distribution were 15 CQS, 25 CQT, and 10 SPT, for details see Table 1.
Table 1 Summary sample collection grid for quality study
Facilities # CQT CQS SPT Total
RMS 1 1 1 1 3
GMS 1 1 1 1 3
DMS 1 2 - - 2
GH 1 1 1 1 3
RH/HC 4 6 (1 exp) 3 (2 exp) 2 11
HU 9 9 4 (1 exp) - 13
PH 5 5 5 5 15
Total 22 25 15 10 50
RMS = Regional Medical Store, Aden;
GMS = Central Medical Stores, Lahej Governorate DMS = District Medical Stores; GH = Governorate General Hospital; RH/HC = Rural Hospital/Health Centre; HU = Health Unit; PP = Private Pharmacy/Drug Store; exp = Expired CQT = Chloroquine tablets; CQS = Chloroquine syrup; SPT= Sulfadoxine Pyrimethamine tablets.
The sampling was performed as follows. Four districts, in addition to the capital of the governorate, were selected by simple random sampling. Samples from the Governorate hospital and the hospitals of the four districts were studied (there is only one hospital or health centre in each district). Nine health units which are working and easy to reach were chosen (most health units are difficult to reach because of long distance or bad roads). Samples were also collected from only one district medical stores because there are no medical stores in the other districts as the hospital/health centre store functioning as a hospital store, pharmacy and also supply the district health units.
Samples were also collected from 5 private pharmacies in the governorate.
One sample was collected from the governorate capital and one sample from each of the studied districts. The pharmacies were selected by systematic random sampling.
Chloroquine syrup in both public and private sector was available in bottles of 100 or 120 ml. Chloroquine tablets were available in strips of 10 tablets or tins/bottles of 1000 tablets. Sulfadoxine/pyrimethamine tablets were present in strips of 3 or 10 tablets each.
Four bottles of chloroquine syrup, four samples of chloroquine tablets (20 tablets/sample) and four samples of sulfadoxine/pyrimethamine tablets (20 tablets/sample) were taken by convenience sampling. Each preparation was from the same batch.
Four samples from each batch were collected from each private community pharmacy/drug store and each level of the public drug distribution chain. These samples were divided as follow:
One was kept with the research team; one was tested at Sana'a Drug Quality Control Laboratory (Sana'a DQCL); one was tested at Aden Drug Quality Control Laboratory (Aden DQCL); one was sent to Eastern Mediterranean Regional office of the World Health Organization (WHO/EMRO) and then forwarded to the Centre for Quality Assurance of Medicine (CENQAM) in Potchefstroom, in South Africa for analysis.
The following information was recorded during field survey for each drug sample on a serially numbered data collection form: location of sample collection point (name of the facility); facility type; name of drug (brand or generic); strength and dosage form; date of manufacture; date of expiry; description of packaging material and any remarks on storage; date of sample collection; name of person in charge of institution and name of person collecting the sample
The collected samples of tablets were transferred from their original containers into amber glass bottles with screw cap seals. Syrups were left in their original containers. The containers were carefully labeled at the point of collection with sticker containing the same information on the original container, particularly, drug name, batch number, facility name, sampling date and sample form number. Each sticker was given a unique sampling identification code.
All testing was carried out at Sana'a and Aden DQCLs. For verification the same samples were tested in CENQAM Laboratories. Samples were analysed according to pharmacopoeial specifications to assess the quality of the products. The quality indicators measured were the content of the active ingredient for syrup, and dissolution and content for tablets.
There is no analytical monograph in the United States Pharmacopoeia (USP) or British Pharmacopoeia (BP) for the assay of CQS. CENQAM used a high pressure liquid chromatography (HPLC) method which was developed and validated in-house [9]. Sana'a and Aden DQCLs used the method of PHARCO Pharmaceuticals-an Egyptian pharmaceutical manufacturer. In this method the absorbance of both standard and test solutions was measured by UV-spectroscopy at 329 nm.
The content of CQ in tablets was determined by UV-spectrophotometry in accordance with an adapted method for CQT of the USP 2003:424. The method consists of recording the UV absorbance of the sample at a wavelength of 343 nm and comparing it to the absorbance of a reference standard in a concentration range representing a 100% label claim at the same wavelength.
The content of SP was determined by means of the HPLC method for the assay of SPT described in the USP 2003:1734. The mobile phase was adjusted to satisfy the system suitability criteria of the USP for chromatographic systems.
The general method and apparatus of the USP were used for testing the dissolution characteristics of both CQT and SPT according to their respective dissolution monographs. The dissolution parameters for CQT in the USP 2003:424 and for SPT in the USP 2003:1734 were followed.
Results
Table 2 shows the laboratory results of CQS tested in the three reference laboratories. As CQS has no official monograph in the USP or BP, an acceptance criteria of 90–110 of the label claim was adopted. Only one product did not comply with the adapted criteria and tested too high i.e. had some "high" content failure. This high percentage ingredient failure (116.2%) was recorded at Tor-Albeh rural hospital/health centre and manufactured by YEDCO, Republic of Yemen.
Table 2 The laboratory results of chloroquine syrup tested at the three reference laboratories
CHLOROQUINE SYRUP (Limit: 90–110%)
Code CENQAM DQCL-Sana'a DQCL-Aden
Assay (%) % RSD Assay (%) % RSD Assay (%) % RSD
CQS/RMS 104.1 3.20 96.20 0.60 102.0 0.36
CQS/SMS 105.7 1.10 100.5 1.01 100.5 3.71
CQS/LGH 109.0 3.20 99.4 1.94 101.2 1.77
CQS/MRH-1 91.1 3.50 101.2 2.65 102.7 3.80
CQS/HRH 100.0 0.85 95.3 0.65 107.7 0.71
CQS/TRH-EXP 116.2 3.00 98.5 1.22 99.4 0.75
CQS/HU-1 108.3 4.90 102.2 0.78 102.7 4.11
CQS/HU-7 107.8 4.60 98.5 0.3 97.6 6.33
CQS/HU-8 107.4 4.30 102.6 2.01 100.8 1.30
CQS/HU-9 109.9 1.20 98.1 0.83 95.9 9.15
CQS/PP-1 92.6 2.70 97.2 3.82 94.8 4.56
CQS/PP-2 90.7 2.50 102.6 2.10 96.4 2.10
CQS/PP-3 107.0 4.90 97.5 3.10 100.7 2.8
CQS/PP-4 98.0 4.30 101.3 2.10 99.7 1.45
CQS/PP-5 100.3 4.60 101.6 1.14 102.4 1.36
CENQAM = Centre for Quality Assurance of Medicine, South Africa
RSD = Relative standard deviation
DQCL = Drug Quality Control Laboratory
RMS = Regional Medical Store, Aden; GMS = Central Medical Stores, Lahej Governorate; DMS = District Medical Stores; GH = Governorate General Hospital; RH/HC = Rural Hospital/Health Centre; HU = Health Unit; PP = Private Pharmacy/Drug Store; exp = Expired; CQS = Chloroquine syrup.
The acceptance criteria for CQT (phosphate) in the USP are 93–107% of the stated amount per unit. Most failures were "low" failures i.e. sample contents below the minimum recommended levels for the products The lowest percentage failures in ingredient content were recorded for 3 products at Al-Raga health unit in Tor-Albaheh district, central medical stores in Lahej and regional medical stores in Aden. The three products were manufactured by YEDCO, Republic of Yemen. The highest percentage ingredient contents failure were found at Al-Mosimir rural hospital and Akan health unit in Al-Mosimir district. They were manufactured by YEDCO, Republic of Yemen, and PHARMED, the Netherlands respectively.
Dissolution testing was done on 6 units and % relative standard deviation (% RSD) of < 5 % was observed for most of the products tested. The exceptions were CQT/HU-7 (5.3%) and CQT/HU-9 (7.6%).
The dissolution acceptance criteria for CQT of the USP were applied. CQT have a Q-value (the quantity of dissolved active specified in the monograph, expressed as a percentage of the labeled content) of 75% in 45 minutes and, according to the acceptance table of USP 2003:2161, the dissolution of each of the six units tested should not be less than 80% (Q+5). The product found at Al-Meghafah health unit, Toben district (CQT/HU-7) and the product found at Al-Fiosh health unit, Toben district (CQT/HU-9) did not comply with the criteria. These two products are manufactured by ALKALOIDA Chemical Co. Ltd., Hungary (See Additional file 1, showing the results of chloroquine tablets testing in the three reference laboratories).
The content criteria for SPT in the USP 2003:1734 are 90–110 of the label claims for both actives. The percentage of SP contents in tablets for all studied SPT samples complied with the criteria with respect to both actives.
Dissolution testing was done on 6 units and a high % RSD was observed for pyrimethamine in some of the products tested. In general the % RSD for the pyrimethamine was higher than for the sulfadoxine content in most of the products.
The dissolution acceptance criteria for SPT of the USP were applied. SPT have a Q-value of 60% in 30 minutes for both actives and according to the acceptance table of USP 2003:2161, the dissolution of each 6 units tested should not be less than 65% (Q+5) for both the actives.
However, all products complied with the criteria with regard to sulfadoxine, only two products complied with the criteria for both actives.
The first product was found at Eskander private drug store in Toben district (SPT/PP-1), which manufactured by Intas Pharmaceuticals Ltd., India. The other product was found at Al-Ikhlas private drug store in Al-Mosimir district (SPT/PP-2), and manufactured by Roche, Switzerland.
The other products did not comply with the criteria for pyrimethamine. Also only 1 unit of product SPT/CMS [67.1% – 3.3 (% RSD) = 63.8%] did not comply with the criteria for pyrimethamine(ç).
There was no significant difference between the national DQC and CENQAM reference laboratories regarding the active ingredients for CQS, CQT, and SPT, or CQT dissolution rate. On the other hand, there was a significant difference between them regarding the SP dissolution whereby only 30% of SP tested were fulfilling the USP criteria in CENQAM laboratory compared to all tested products in the national DQCLs (Table 3).
Table 3 Comparison between national reference laboratories and CENQAM Reference laboratory regarding the quality screening of AM drugs
CENQAM % fulfilling the USP content criteria Sana'a DQCL % fulfilling the USP content criteria P1-value Aden DQCL % fulfilling the USP content criteria P2-value P3-value
CQS Assay 93.3 (14/15) 100 (15/15) 1 100 (15/15) 1 1
CQT Assay 80 (20/25) 92 (23/25) 0.41 84 (21/25) 1 0.66
CQT dissolution 92 (23/25) 100 (25/25) 0.47 96 (24/25) 1 1
SP Assay 100 (10/10) 100 (10/10) 1 100 (10/10) 0 1
SPT dissolution 20 (2/10) 100 (10/10) 0.005* 100 (10/10) 0.005* 1
P1: p-value of Chi-Square test for comparing the proportion of tested products fulfilling the USP content criteria in CENQAM versus Sana'a reference laboratory.
P2: p-value of Chi-Square test for comparing the proportion of tested products fulfilling the USP content criteria in CENQAM versus Aden reference laboratory.
P3: p-value of Chi-Square test for comparing the proportion of tested products fulfilling the USP content criteria in Sana'a versus Aden reference laboratory.
Table 4 shows a comparison between the quality of antimalarial drugs obtained from this study and that of 7 malaria endemic countries.
Table 4 Comparison between country results: percentage failure of samples
Yemen Sudan Gabon Ghana Mali Kenya Mozambique Zimbabwe
CQS Assay 6.7 26.6 0 5.0 66.7 25.0 25.0 13.3
CQT assay 20 5.2 29.0 66.7 47.3 42.8 20.0 57.1
CQT dissolution 8 12.5 5.8 20.0 5.2 28.6 6.7 7.1
SP assay 0 0 0 37.5 0 0 5.5 10.0
SP dissolution 70 80 - 75.0 100 91.7 100 100
Discussion
The quality control performed in this study was utilized to ensure that the batch of tested AMs complies with its specification and is fit for its intended in terms of efficacy, safety and acceptability.
CQS had content failure rate of average 6.7%. The failure was "high" failure. Content failure for CQT was very significant and averaged at 20%. Failures were "low" and "high" failures. SPT content complied with criteria with respect to both actives in the 3 reference laboratories.
Dissolution failure rates for CQT were 8%. SPT had problems mostly with the dissolution of the pyrimethamine component of the formulation, and averaged at 80% for the SPT samples analysed.
The data presented in this report indicates there is a problem of sub-standard AMs available at the public health facilities and circulating in the market. The main problem seems to be samples below the lower limit of specification. The most significant results were the low content failures for CQT content and SPT dissolution. The failures detected in this study indicate a serious problem that warrants further investigation and intervention.
Quality problems were not limited to a particular distribution level but recorded at different levels. Also there were failures among locally made products as well as foreign products. Further investigation of this phenomenon will be important since it is easier for national drug regulatory authorities to act and correct problems that involve domestic manufacturers.
The discrepancy between the national reference laboratories and CENQAM results in ingredient content failure for CQ and SPT dissolution, can only be taken as indicative, but not conclusive and should be carefully investigated to identify the causes and rectify them. As the three laboratories use the same methodology for quality screening of drugs, the possible reasons for such discrepancy could be personnel errors and/or inadequate calibration of the instruments used.
In comparison with other malaria endemic countries that adopted the same methodology in evaluating the quality of antimalarial drugs [9], the following comments could be drawn from the study:
• CQS content failure in Yemen is relatively low and is almost comparable to Ghana's. On the other hand, it is significantly lower than other countries such as: Sudan, Mali, Kenya, Mozambique and Zimbabwe.
• CQT content failure in Yemen is higher than Sudan, comparable to Gabon, and Mozambique, but significantly lower than Ghana, Mali, Kenya and Zimbabwe.
• CQT dissolution failure in Yemen is relatively low. It is lower than Sudan, Ghana, Kenya, but comparable to Gabon, Mali and Mozambique.
• SP content showed no failure, as in most other malaria endemic countries. Similarly, SP dissolution failure was as high as other malaria endemic countries.
Conclusion
Problems of sub-standard AMs exist in different studied districts, within the drug distribution chain in Lahej, Yemen and produced by both domestic and foreign manufacturers.
Percentage failure of samples based on ingredient content is 6.7% for CQS and 20% for CQT and in dissolution failure 8% for CQT and 70% for SPT, cannot be ignored. This will have serious implications not only on the reduced therapeutic effectiveness of AMs but also on the development of drug resistance.
In view of potential danger that sub-standard AMs could already be posing in the fight against malaria, an intervention plan should be developed immediately. This could involve setting up quality surveillance systems within drug regulatory authorities in the country, supporting manufacturers to improve Good Manufacturing Practice (GMP) compliance and promote effective managing drug supply.
The lessons learnt from this pilot study could be valuable in any future investigations, and interventions to improve the quality of AMs and other medicines in use in domestic and international markets.
Recommendations
• Promoting good procurement practices in the public sector by registration or pre-qualification of suppliers, i.e. purchasing from reliable and approved sources.
• More emphasis should be given to testing for initial quality, the routine testing of new supplies, ensuring adherence of manufacturers and suppliers to GMP.
• Antimalarial drugs should be stored and distributed in appropriate conditions in harbours, stores and health facilities as well as at the level of the end users.
• Measures should be taken to stop the illegal importing or smuggling of drugs.
• Ensure that people in charge of drug management supply and quality control are competent and receive training, a measure which would be cost-effective.
• Although the cost of antimalarial drugs is important, good quality antimalarial drugs are more important than cheaper poor drugs.
Authors' contributions
Dr Ahmed Abdo-Rabbo: protocol development, field work implementation, data collection and analysis and interpretation of data; drafted the article.
Dr Amal Bassili assisted in protocol development, supervision on the progress of the project, assistance in data analysis and interpretation; revision of the manuscript.
Dr Hoda Atta assisted in protocol development, supervision on the progress of the project; revision of the manuscript. All authors have given final approval to the version to be submitted for publication.
Supplementary Material
Additional File 1
Results of chloroquine tablets testing in the three reference laboratories
Click here for file
Additional File 1
Results of chloroquine tablets testing in the three reference laboratories
Click here for file
Acknowledgements
We would like to thank the DG and staff members of the health office and health facilities visited in Lahej Governorate, also the directors and analysts of Sana'a and Aden Drug Quality Control Laboratories as well as the Centre for Quality Assurance of Medicines laboratories, Potchefstroom, South Africa.
This investigation received technical and financial support from the joint WHO Eastern Mediterranean Region (EMRO), Division of Communicable Diseases (DCD) and the WHO Special Programme for Research and Training in Tropical Diseases (TDR): The EMRO/ DCD/TDR Small Grants Scheme for Operational Research in Tropical and other Communicable Diseases and Essential Drugs and Biological programme of the World Health Organization.
==== Refs
National Malaria Control Programme Annual Report (Arabic), Ministry of Public Health and Population, Sana'a, Republic of Yemen
Hogerzeil HV Battersby A Srdanovic V Hansen LV Boye O Lindgren B Everitt G Stjernstrom NE WHO/UNICEF study on the stability of drugs during international transport WHO/DAP/91.1
Walker GJA Hogerzeil HV Potency of ergometrine in tropical countries Lancet 1988 332 393 2899794 10.1016/S0140-6736(88)92858-9
Hogerzeil HV Walker GJA Instability of (methyl) ergometrine in tropical climates: an overview Eur J Obstet Gynecol Reprod Biol 1996 69 25 29 8909953 10.1016/0301-2115(95)02530-8
Hogerzeil HV Battersby A Srdanovic V Stjernstrom NE Stability of essential drugs during shipment to the tropics BMJ 1992 304 210 212 1739795
Nazerali H Hogerzeil HV The quality and stability of essential drugs in rural Zimbabwe: controlled longitudinal study BMJ 1998 317 512 513 9712597
World Health Organization Accelerated stability studies of widely used pharmaceutical substances under simulated hospital conditions Geneva: World Health Organization 1998 WHO/PHARM/86.529
Abu Reid IO El-Samani SA Hag Omer AI Khalil NY Mahgoub KM Everitt G Grundstrom K Lindgren B Stjernstrom NE Stability of drugs in the tropics. A study in Sudan Int Pharm J 1990 4 6 10
Maponga C Ondari C The quality of antimalarials, a study in selected African countries Geneva: World Health Organization WHO/EDM/PAR/20034 2003
Wondemagegnehu E Counterfeit and sub-standard drugs in Myanmar and Vietnam Geneva, World Health Organization 1999 WHO/EDM/QSM/99.3
Yemen Treatment Guidelines and Essential Drugs List. Pharmacy Sector, Ministry of Public Health and Population, Sana'a, Yemen
|
15987508
|
PMC1192817
|
CC BY
|
2021-01-04 16:24:13
|
no
|
Malar J. 2005 Jun 29; 4:28
|
utf-8
|
Malar J
| 2,005 |
10.1186/1475-2875-4-28
|
oa_comm
|
==== Front
Popul Health MetrPopulation Health Metrics1478-7954BioMed Central London 1478-7954-3-71601417410.1186/1478-7954-3-7ResearchAssessment of the health of Americans: the average health-related quality of life and its inequality across individuals and groups Asada Yukiko [email protected] Department of Community Health and Epidemiology, Faculty of Medicine, Dalhousie, University, 5790 University Avenue, Halifax, Nova Scotia, B3H 1V7, Canada2005 13 7 2005 3 7 7 3 1 2005 13 7 2005 Copyright © 2005 Asada; licensee BioMed Central Ltd.2005Asada; 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 assessment of population health has traditionally relied on the population's average health measured by mortality related indicators. Researchers have increasingly recognized the importance of including information on health inequality and health-related quality of life (HRQL) in the assessment of population health. The objective of this study is to assess the health of Americans in the 1990s by describing the average HRQL and its inequality across individuals and groups.
Methods
This study uses the 1990 and 1995 National Health Interview Survey from the United States. The measure of HRQL is the Health and Activity Limitation Index (HALex). The measure of health inequality across individuals is the Gini coefficient. This study provides confidence intervals (CI) for the Gini coefficient by a bootstrap method. To describe health inequality by group, this study decomposes the overall Gini coefficient into the between-group, within-group, and overlap Gini coefficient using race (White, Black, and other) as an example. This study looks at how much contribution the overlap Gini coefficient makes to the overall Gini coefficient, in addition to the absolute mean differences between groups.
Results
The average HALex was the same in 1990 (0.87, 95% CI: 0.87, 0.88) and 1995 (0.87, 95% CI: 0.86, 0.87). The Gini coefficient for the HALex distribution across individuals was greater in 1995 (0.097, 95% CI: 0.096, 0.099) than 1990 (0.092, 95% CI: 0.091, 0.094). Differences in the average HALex between all racial groups were the same in 1995 as 1990. The contribution of the overlap to the overall Gini coefficient was greater in 1995 than in 1990 by 2.4%. In both years, inequality between racial groups accounted only for 4–5% of overall inequality.
Conclusion
The average HRQL of Americans was the same in 1990 and 1995, but inequality in HRQL across individuals was greater in 1995 than 1990. Inequality in HRQL by race was smaller in 1995 than 1990 because race had smaller effect on the way health was distributed in 1995 than 1990. Analysis of the average HRQL and its inequality provides information on the health of a population invisible in the traditional analysis of population health.
==== Body
Introduction
To assess the health of a population, we have traditionally relied on the average or overall level of health in a population. For example, 77.2 years of life expectancy for Americans in 2001 [1] or an infant mortality rate of 6.8 per 1,000 in the United States in 2001 [2] provide some information about the health of Americans. But the average or overall health is, by definition, one number from a population and arguably artificial. Whose level of health does the average or overall health really represent?
Researchers and policy-makers have increasingly paid attention to health inequality as an indicator of population health [3-5]. They believe that a traditional average health of a population does not provide enough information as a population health measure, and investigation of the distribution of health within a population is necessary. Thus, such policy documents as Healthy People 2010 [5], the national health plan for the decade in the US, and The World Health Report 2000 [4] clearly state two goals of improving population health: the increase in the average or overall level of health and the decrease in health inequality or disparity.
Concurrently, researchers and policy-makers have increasingly recognized the importance of health-related quality of life (HRQL) in the assessment of population health. We have traditionally measured population health with indicators of life years or mortality. These are the most robust measures of health for their objectivity and availability. Yet we value both living long and living well [6], and researchers have developed various HRQL measures to capture the value for living well [7]. Healthy People 2010, for example, states the importance of looking at HRQL in the assessment of how healthy Americans are [5].
Little research has incorporated both of these two interests in health inequality and HRQL into the assessment of population health [8-10]. This study assesses the health of Americans in terms of the average HRQL and its inequality, using the 1990 and 1995 National Health Interview Survey (NHIS). This study uses the Health and Activity Limitation Index (HALex) as a measure of HRQL. The use of the HALex for this present study is particularly suitable, since the HALex was developed to monitor the health of Americans during the 1990s [11]. One of the three goals of Healthy People 2000 is to increase the span of healthy life for Americans [12]. To assist this goal, Erickson and her colleagues created a new health measure, Years of Healthy Life (YHL) [11]. The YHL combines information on HRQL and mortality. To obtain HRQL information, Erickson and her colleagues developed the HALex based on two questions from the NHIS, activity limitation and self-perceived health. Although the HALex is one component of the YHL, researchers have used it independently as a useful measure of HRQL [13-15].
This study measures inequality in HRQL across groups as well as individuals. Researchers and policy-makers have traditionally measured health inequality across groups, for example, by income, education, occupation, race, or geographic location. Recently researchers at the World Health Organization (WHO) proposed to measure health inequality across individuals, irrespective of individuals' group affiliations, in much the same way as measuring income inequality [16]. The group and individual approaches measure different dimensions of health inequality and could yield different results [17]. The group and individual approaches can complement each other and strengthen the assessment of population health [18].
Methodologically, this study explores two recent advancements in empirical health inequality research. First, it provides confidence intervals for the degree of health inequality estimated. Although a few pioneer studies exist [4,18,19], statistical inference has yet to become a standard practice in health inequality analysis. Without statistical inference, we cannot conclude with confidence whether health inequality increased or decreased. Statisticians have developed bootstrap methods to overcome difficulties in estimating the standard error of the degree of inequality using data with a complex survey design, like the NHIS [20-22]. This study employs such methods.
Second, this study examines health inequality across groups by decomposing overall inequality into inequality within each group, inequality between groups, and inequality overlapping groups. This decomposition technique is common in analysis of income inequality and poverty [23-27]. It provides richer information than the conventional analysis of comparing the average health between groups, as recent health studies show [18,28-30]. To explore the decomposition technique, this study uses race as an example. Although this study does not intend to undertake a full investigation of health inequality by race in the US, the focus on race is compatible with the recent extension of the attention from inequalities in health care by race [31,32] to inequalities in health outcomes by race [33,34].
In this study, "health distribution" is a way in which health is spread among individuals or groups of people in a population. "Health equality" suggests the health distribution in which health is spread equally to every chosen unit of analysis. "Health inequality" means all health distributions that are otherwise. I synonymously use such terms as "inequality," "disparity," and "difference."
Methods
Sample and Data
Data come from the 1990 and 1995 National Health Interview Survey (NHIS) [35,36]. I select these study years because the questionnaire design of the 1990 and 1995 NHIS permits the construction of the Health and Activity Limitation Index (HALex), the health variable of this present study, exactly as proposed by its developers (see below) [11]. The NHIS uses a stratified multistage probability design, yielding a nationally representative sample of the civilian non-institutionalized US population. The method of data collection is face-to-face household interview. The interviewers obtain surrogate information for children younger than 17 years of age and persons absent at the time of the interview. The response rate is over 95%. I exclude observations missing an answer to a question necessary to construct the HALex (0.5% missing in 1990, 1.2% missing in 1995). The sample size for this study is 119,003 (1990) and 101,277 (1995). Table 1 shows the unweighted number of observations by age group (0–14, 15–24, 25–44, 45–64, 65+ years of age), sex, and race (White, Black, and other racial groups (Aleut, Eskimo or American Indian, Asian or Pacific Islander, and any other race not listed separately)) in 1990 and 1995.
Table 1 Description of Sample
1990 1995
N % N %
All Ages 119003 100 101277 100
Age, y
0–14 27822 23.4 24661 24.4
15–24 16289 13.7 13510 13.3
25–44 37886 31.8 31435 31.0
45–64 22487 18.9 19834 19.6
65+ 14519 12.2 11837 11.7
Men 56830 47.8 48266 47.7
Race
White 97290 81.8 83527 82.5
Black 17886 15.0 13629 13.5
Other 3827 3.2 4121 4.1
Measure of Health: the Health and Activity Limitation Index (HALex)
The HALex combines two types of questions collected in the NHIS, one assessing activity limitation and the other measuring self-perceived health [11]. The activity limitation questions create six categories: (1) not limited, (2) limited in other activities, (3) limited in major activity, (4) unable to perform major activity, (5) unable to perform instrumental activities of daily living, and (6) unable to perform activities of daily living. Self-perceived health is in five categories: excellent, very good, good, fair, and poor. These two items together make up a matrix of 30 combinations (Table 2).
Table 2 The Health and Activity Limitation Index (HALex)
Perceived health status
Excellent Very good Good Fair Poor Dead
Activity limitation Single attribute score 1.00 0.85 0.70 0.30 0.00
Not limited 1.00 1.00 0.92 0.84 0.63 0.47
Limited in performing other activities 0.75 0.87 0.79 0.72 0.52 0.38
Limited in performing major acitivities 0.65 0.81 0.74 0.67 0.48 0.34
Unable to perform major activity 0.40 0.68 0.62 0.55 0.38 0.25
Limited in instrumental activities of daily living (IADL) 0.20 0.57 0.51 0.45 0.29 0.17
Limited in activities of daily living (ADL) 0.00 0.47 0.41 0.36 0.21 0.10
Dead 0.00
Source: Erickson, Wilson, Shannon (1995)
Assignment of a score to each of these 30 combinations took three steps. Developers of the HALex first assigned a score for each of the six levels of activity limitation and the five levels of self-perceived health ("Single attribute score" in Table 2), using a mathematical technique called correspondence analysis. Correspondence analysis belongs to a family of multidimensional scaling, a technique creating a scale for a concept with multiple dimensions, for example, health consisting of mobility, sensory, cognition, emotion, and pain, or social support consisting of informational, emotional, and practical support. Correspondence analysis finds the best simultaneous representation of two domains, activity limitation and self-perceived health in the case of the HALex, by maximizing the correlation between them.
The simplest correspondence analysis applies to a two-way crosstabulation, as in the case for the HALex, activity limitation and self-perceived health. One can assign a score for each of the six levels of activity limitation by weighted least-squares where each of the six levels of activity limitation is weighted by its frequency divided by the total frequency of the six levels, and distances between each of the six levels are measured by the chi-square distance. To measure the distance between "not limited" and "limited in performing other activities" in activity limitation, for example, correspondence analysis uses the chi-square distance between these two categories by examining how people in these two categories differ with respect to the five levels of self-perceived health. Developers of the HALex conducted separate correspondence analysis for each of several different 5-year age groups and different years of the NHIS. Based on the analyses, they assigned single attribute scores for each of the two domains as listed in Table 2 that maximize the correlation between activity limitation and self-perceived health in all age groups. Please refer to Greenacre [37,38] for detail explanation of correspondence analysis.
Next, the developers of the HALex made the following assumptions. They assumed that the score for the health state with no activity limitation and excellent self-perceived health is 1.00, and the score for the health state with limited activities of daily living and poor self-perceived health is 0.10. In addition, they assumed that a health state with limited activities of daily living and excellent self-perceived health is equally bad as the health state with no activity limitation and poor self-perceived health. Based on another HRQL measure, the Health Utilities Index Mark I, they assigned the score of 0.47 for these two health states.
Finally, using the scores assigned for each level of activity limitation and self-perceived health, and the four scores based on the assumptions described above, the developers of the HALex calculated scores for the rest of the 26 health states. The formula of this calculation is based on multiattribute utility theory. Multiattribute utility theory extends the traditional expected utility theory, a theory of rational decision making under uncertainty, by adding an independence assumption. The developers of the HALex, in particular, assumed mutual utility independence, that is, health domains other than self-perceived health and activity limitation (for example, pain, emotion, or hearing) have no effect on the HALex score. For example, the HALex score for the health state with limited activities of daily living and excellent self-perceived health is 0.47 regardless of the existence of pain or emotional or hearing problems. Due to this mutual utility independence assumption, the developers of the HALex used a multiplicative function for calculating the HALex scores. Drummond et al. [39] gives detail explanation of multiattribute utility theory, and technical notes of the YHL [11] provide the further detail of the HALex construction.
Erickson has later evaluated and confirmed the construct validity of the HALex [40]. For the following health inequality analysis, I assign a HALex score to each observation in the 1990 and 1995 data.
Measure of Health Inequality: the Gini Coefficient
A measure of health inequality summarizes a health distribution into one number. This facilitates comparison and examination by quantifying a degree of health inequality. This study uses the Gini coefficient as the measure of health inequality. The Gini coefficient has most frequently been applied to income distribution, but it is possible to apply it to health distribution as previous studies demonstrated [41,42].
Figure 1 explains the Gini coefficient and the Lorenz curve (see Figure 1). Imagine that we horizontally line up individuals in a population from the sickest to the healthiest and vertically line up these individuals' health share, in the case of this present study, the cumulative percentage of the HALex. The resulting dotted curve AC is called the Lorenz curve. When the population is perfectly equal, the Lorenz curve is the diagonal line, AC. When the population is most unequal, that is, in the case of this present study, one person has a HALex score equal to or greater than 0.1 and the HALex of all others is zero (dead), the Lorenz curve follows AB and BC. The Gini coefficient is the shaded area in the graph divided by the triangle, ABC. It can take a value between zero when the Lorenz curve is diagonal, thus, perfectly equal, and one when the Lorenz curve follows AB and BC, the most unequal.
Figure 1 The Lorenz curve.
Arithmetically, the Gini coefficient (G) is expressed as:
Where the target population holds n people, yi is the HALex score of individual i, yj is the HALex score of individual j, and the average HALex in the population is μ.
Subgroup Decomposition of the Gini Coefficient
Customarily, the measurement of health inequality by group is the difference between the average health of groups (Figure 2a, see Figure 2). But the use of averages is questionable, especially when a health distribution does not follow a normal distribution. Figures 2b and 2c schematically illustrate that the same mean difference in health by group in Figure 2a could come from different distributions. The degree of overlap between the two groups in Figure 2b is smaller than that of Figure 2c. Despite the same absolute mean difference, the extent of group stratification or isolation with respect to health is greater in Figure 2b than Figure 2c. This present study adds this overlap information to the conventional absolute mean difference in analyzing health inequality by race.
Figure 2 Mean difference (a), small overlap (b), and big overlap (c). Suppose we are here interested in two groups. Conventionally we compare the average health of these two groups (Figure 2a). But the same average health could come from different distributions (Figures 2b and 2c). Although Figures 2b and 2c have the same average health, the overlap between groups is greater in Figure 2c than Figure 2b. A greater overlap indicates that the group characteristic does not have much effect on the way health is distributed. These figures are not based on actual distributions and used only for illustrative purposes.
Suppose we have subpopulation k = 1,2,.., n. Decomposition of the Gini coefficient (G) by subpopulation can be expressed as follows [24]:
G = GB + ∑ ak Gk +GO
Where GB is the between-group Gini coefficient, calculated under the assumption that everybody's health in subpopulation k is the average health of subpopulation k. Gk is the Gini coefficient within subpopulation k. Each of this within-group Gini coefficient is weighted by population share and health share of subpopulation k, and its sum for all subpopulations ∑ak Gk is the total within-group Gini coefficient. GO is a residual, which can be interpreted as the overlap Gini coefficient. When subpopulations do not overlap, GO equals to zero. When subpopulations are identical, that is, subpopulations perfectly overlap, GO also equals to zero. Unless subpopulations are perfectly identical, a greater value of GO suggests a higher degree of overlap of subpopulations. A higher degree of subpopulation overlap indicates that the group characteristic does not have much effect on the way health is distributed. The analysis of health inequality by race in this present study looks at how much contribution (expressed in percentage) this overlap term makes to the overall Gini coefficient, in addition to the absolute mean difference between groups.
Analysis
This study consists of three parts: (1) analysis of the average HALex, (2) analysis of inequality in the HALex across individuals, and (3) analysis of inequality in the HALex by race. All three parts use both 1990 and 1995 data. I provide 95% confidence intervals (CI) for the average HALex using linearization (Taylor Series) methods [43,44].
Providing 95% CI for the Gini coefficient in this present study faces two difficulties. First, the Gini coefficient is a non-linear function and bounded between zero and one, which makes it difficult to use asymptotic theory. Second, the NHIS uses a complex survey design involving stratification, clustering, and multistage sampling. To overcome these two difficulties, I use a bootstrap method modified for survey data with the complex design: the two-stage with-replacement bootstrap [20-22]. Bootstrap is a simulation method only using data at hand. With a few assumptions, it can estimate the standard error for any statistic. The original bootstrap proposed by Efron and Tibshirani [45] assumes independence of observations, thus, without modification, cannot be legitimately applied to data using a complex survey design. Modification of the original bootstrap has been suggested for variance estimation of a complex survey design [46,47]. I use one of the modified versions of bootstrap, the two-stage with-replacement bootstrap, where a bootstrap sample with the sampling weight is randomly selected with replacement in two stages. McCarthy and Snowden showed that the with-replacement bootstrap yields more favourable variance estimation and CIs in stratified cluster sampling designs than the without-replacement bootstrap, another bootstrap method modified for complex surveys, where one creates an artificial population from the sample and repeatedly and randomly draws samples without replacement [21]. I repeat the simulation process 2000 times and use the percentile method to obtain 95%CI.
All analyses use weighted data. I use Stata software to conduct all analyses [48].
Results
The Average HALex
The assessment of the health of Americans in 1990 and 1995 differs in terms of the average HALex and life expectancy. Table 3 presents the average HALex, its 95% CI, and life expectancy [49,50] of the US population in 1990 and 1995 by sex and age group. The average HALex for both sexes (0.87 in both years), men (0.88 in 1990, 0.87 in 1995), and women (0.87 in 1990, 0.86 in 1995) of these two years were not statistically significantly different at the 5% level. Life expectancy, on the other hand, was higher in 1995 than 1990 by 0.4-year for both sexes, 0.7-year for men, and 0.1-year for women.
Table 3 The Average HALex and Life Expectancy in the US, 1990 and 1995
Both sexes (95% CI) Male (95% CI) Female (95% CI)
Life expectancy, y
1990 75.4 71.8 78.8
1995 75.8 72.5 78.9
The Average HALex
All ages
1990 0.87 (0.87, 0.88) 0.88 (0.88, 0.88) 0.87 (0.86, 0.87)
1995 0.87 (0.86, 0.87) 0.87 (0.87, 0.88) 0.86 (0.86, 0.86)
0–14 years
1990 0.93 (0.93, 0.93) 0.93 (0.93, 0.93) 0.94 (0.93, 0.94)
1995 0.93 (0.93, 0.93) 0.93 (0.92, 0.93) 0.94 (0.93, 0.94)
15–24 years
1990 0.92 (0.92, 0.92) 0.93 (0.92, 0.93) 0.91 (0.91, 0.91)
1995 0.91 (0.91, 0.91) 0.92 (0.92, 0.92) 0.91 (0.90, 0.91)
25–44 years
1990 0.90 (0.90, 0.90) 0.90 (0.90, 0.91) 0.89 (0.89, 0.89)
1995 0.89 (0.88, 0.89) 0.9 (0.89, 0.90) 0.88 (0.88, 0.88)
45–64 years
1990 0.82 (0.82, 0.83) 0.83 (0.82, 0.83) 0.82 (0.81, 0.82)
1995 0.81 (0.81, 0.82) 0.82 (0.82, 0.83) 0.81 (0.80, 0.81)
65+ years
1990 0.73 (0.72, 0.74) 0.74 (0.74, 0.75) 0.72 (0.71, 0.73)
1995 0.73 (0.72, 0.73) 0.73 (0.73, 0.74) 0.72 (0.71, 0.73)
Figure 3 shows that the difference between the HALex in 1990 and 1995 was not consistent at every age (see Figure 3). The average HALex for both sexes was lower among 15–24 and 25–44 year olds in 1995 than 1990 (p < 0.05). Differences between the average HALex in 1990 and 1995 within other age groups were not statistically significant at the 5% level.
Figure 3 The average HALex by age in 1990 and 1995.
In both 1990 and 1995, overall women's HALex was lower than men's (0.01 difference, p < 0.05 both in 1990 and 1995). This was true in all age groups, except among 0–14 year olds both in 1990 and 1995, 45–64 year olds in 1990, and 65 year olds and older in 1995.
Inequality in the HALex
Inequality in the HALex across individuals was greater in 1995 than 1990. Table 4 presents the Gini coefficient and its 95% CI for the US population in these years by sex and age group. The Gini coefficient was slightly greater for both sexes in 1995 than 1990 (0.005 increase for both sexes combined, p < 0.05, 0.005 increase for men, p < 0.05, and 0.004 increase for women, p < 0.05). Stratified by age group, the Gini coefficient was greater in 1995 than 1990 only for 25–44 year olds (0.007 increase for both sexes combined and both for male and female young adults, p < 0.05).
Table 4 The Gini Coefficient in the US in 1990 and 1995
Both sexes (95% CI) Male (95% CI) Female (95% CI)
All ages
1990 0.092 (0.091, 0.094) 0.087 (0.086, 0.088) 0.097 (0.096, 0.099)
1995 0.097 (0.096, 0.099) 0.092 (0.090, 0.095) 0.101 (0.100, 0.103)
0–14 years
1990 0.048 (0.047, 0.049) 0.049 (0.048, 0.051) 0.046 (0.044, 0.048)
1995 0.049 (0.048, 0.050) 0.052 (0.050, 0.054) 0.046 (0.044, 0.048)
15–24 years
1990 0.056 (0.054, 0.059) 0.053 (0.050, 0.056) 0.059 (0.056, 0.062)
1995 0.060 (0.058, 0.062) 0.056 (0.053, 0.059) 0.063 (0.060, 0.066)
25–44 years
1990 0.072 (0.071, 0.074) 0.069 (0.067, 0.071) 0.075 (0.073, 0.077)
1995 0.079 (0.077, 0.081) 0.076 (0.074, 0.079) 0.082 (0.079, 0.084)
45–64 years
1990 0.127 (0.124, 0.130) 0.126 (0.122, 0.131) 0.127 (0.124, 0.131)
1995 0.132 (0.128, 0.138) 0.130 (0.123, 0.137) 0.134 (0.130, 0.140)
65+ years
1990 0.183 (0.178, 0.188) 0.172 (0.166, 0.178) 0.190 (0.184, 0.196)
1995 0.183 (0.174, 0.188) 0.174 (0.166, 0.183) 0.189 (0.177, 0.197)
Inequality in the HALex by Race
Inequality in the HALex by race was smaller in 1995 than 1990 because race had smaller effect on the way health was distributed in 1995 than 1990 while the absolute mean differences between racial groups were the same between these years. Table 5 summarizes the average HALex, the Gini coefficient, and their 95% CI, and the Gini coefficient decomposed for Whites, Blacks, and other racial groups in these years. The average HALex was lower in 1995 than 1990 in all three racial groups, although only the difference among Whites was statistically significant (p < 0.05). Differences in the average HALex between all racial groups were the same in 1995 and 1990. Despite no difference in the average HALex between racial groups, the contribution of the overlap to the overall Gini coefficient was greater in 1995 than in 1990 by 2.4%.
Table 5 The Average HALex, the Gini Coefficient by Race in 1990 and 1995
1990 1995
Average HALex (95% CI)
All 0.87 (0.87, 0.88) 0.87 (0.86, 0.87)
White 0.88 (0.88, 0.88) 0.87 (0.87, 0.87)
Black 0.845 (0.84, 0.85) 0.84 (0.84, 0.85)
Other 0.89 (0.88, 0.90) 0.88 (0.87, 0.89)
Average HALex difference (* p < 0.05)
White – Black 0.03* 0.03*
Other – Black 0.04* 0.04*
Other – White 0.01 0.010
The Gini coefficient (95% CI)
All 0.092 (0.091, 0.094) 0.097 (0.096, 0.099)
White 0.090 (0.089, 0.092) 0.095 (0.093, 0.098)
Black 0.109 (0.104, 0.115) 0.112 (0.109, 0.116)
Other 0.077 (0.072, 0.083) 0.085 (0.080, 0.092)
Decomposition of the Gini coefficient (Contribution, %)
Overall 0.092 (100) 0.097 (100)
Between-group 0.004 (4.72) 0.004 (4.17)
Within-group 0.066 (71.47) 0.068 (69.69)
Overlap 0.022 (23.81) 0.025 (26.15)
Health inequality between racial groups only minimally explains overall health inequality. The between-group Gini coefficient explains only 4.7% (in 1990) and 4.2% (in 1995) of the overall Gini coefficient.
Discussion
This study showed that the average HALex of Americans in 1990 and 1995 were the same (0.87), but inequality in the HALex across individuals was slightly greater in 1995 (the Gini coefficient: 0.097) than 1990 (0.092) (p < 0.05). This study explored decomposition of the Gini coefficient as a tool to examine health inequality by group using race as an example. The decomposition analysis showed that inequality in the HALex by race was smaller in 1995 than 1990 because race had smaller effect on the way health was distributed in 1995 than 1990. Moreover, the decomposition analysis suggested that inequality in the HALex between racial groups explains only 4.7% (in 1990) and 4.2% (in 1995) of overall inequality in the HALex.
This study confirmed that one obtains different pictures of the health of a population when measuring it by life years and HRQL [4,51]. According to this study, the average HALex of Americans was the same in 1990 and 1995, although their life expectancy was higher in 1995 than 1990. This study only compared the average HALex of two years, 1990 and 1995, thus, it does not provide information of a trend of the HALex. Routine collection of the average HALex of Americans along with measures of mortality or life year will enable a richer assessment of the health of Americans. Moreover, reporting of an age-standardized HALex can show changes in the HALex independent from the age structure of the populations.
For the wider use of the HALex in the assessment of the health of Americans, future work should acknowledge that the HALex is derived from self-reported activity limitation and self-perceived health questions. Should we assess population health based on a self-reported measure of health such as the HALex or on an "objective" measure of health such as medical diagnosis? Observation of the differences in the HALex between men and women in this study suggests the importance of this question. This study showed that in both 1990 and 1995, women's HALex was lower than men's in all age groups, except among children, 0–14 years old. All of these differences in the HALex by sex, except among 45–64 year olds in 1990 and 65 year olds and older in 1995, are statistically significant at the 5% level. In contrast, life expectancy was 7 years higher for women than men in 1990, and 6.4 years higher for women than men in 1995. In addition, the WHO reports that healthy life expectancy, which combines life expectancy and HRQL, was 4.1 years higher for women (71.3 years) than men (67.2 years) in 2002 [52]. Is women's health status "objectively" lower than men's, or do women perceive their health status lower than men's? What if we discovered that women perceive the same, "objective" health conditions lower than men – should we consider low perception as a health problem? The issue of perception is not only limited to sex but also applies to socioeconomic status, racial groups, or geographic location. The future work needs to investigate how much of the difference in the HALex is due to the difference in perception and identify appropriateness of using the HALex or any other self-reported measure of health in the assessment of population health.
This study was the first to describe inequality in the HRQL among a nationally representative sample of Americans. It showed that inequality in the HALex across individuals in the US was greater in 1995 than 1990. Although this descriptive study cannot explain reasons for this difference, one possible speculation for this finding is that the increasing income inequality due to uneven distribution of the economic growth [53,54] might have an effect on health inequality.
This study used the Gini coefficient as the measurement of health inequality. Following the previous studies that applied the Gini coefficient to health distribution [41,42], this study reported three-decimal Gini coefficients. However, an appropriate level of precision of the Gini coefficient used in health distribution, especially distribution of the HALex, is unknown. Future work needs to investigate this point.
Age group analysis uncovered the worrisome health of young Americans, especially 25–44 year olds. The average HALex was lower and its inequality was greater in 1995 than 1990 in this age group. A possible etiology for this finding is the spread of HIV/AIDS. In 1995, HIV/AIDS was the leading cause of death among the young [55]. Another factor might be the general trend that the onset of chronic diseases has shifted to younger ages. Even without these epidemiologic trends, one might call the age group of 25–44 year olds the "forgotten" age group. Health policy tends to focus on infants, adolescents, and the elderly. Young adulthood is often considered as the most resilient stage of life in terms of human biology. It is, however, this period of life in which the proportion of the uninsured is the second highest (after 18–24 year olds) [56], and young families struggle to establish themselves.
This study explored the subgroup decomposition technique as a tool for analysis of health inequality by group. A striking finding from the decomposition analysis was that only 4–5% reduction of overall inequality would be possible even if differences in the mean HALex between all racial groups disappeared. The importance of race for social justice considerations does not depend on the magnitude of the between-group inequality. Nonetheless, such information can be useful in examining policy implications of health inequalities by different group characteristics. In other words, the decomposition analysis helps us investigate health inequality from a broader perspective: given that differences in the HALex between racial groups only explains 4–5% of overall inequality in the HALex, what factors are accountable for the rest of 95–96% of overall inequality?
Application of the subpopulation decomposition technique to health inequality analysis is admittedly still in its infancy. This study used the subgroup decomposition technique only for one group characteristic (race) at a time. The literature on income inequality and poverty and pioneering health studies point to three possible extensions of this unidimensional subgroup decomposition. First, adjusting for a number of groups in subgroup characteristics (for example, three groups for race, and five groups for income), one can compare results of the subgroup decomposition applied to different subgroups [25,26]. This means that one can identify how much of the overall health inequality comes from, for example, health inequalities by gender, income, education, and geographic location. Second, overall health inequality can be decomposed not only by subgroup but by component [30]. An overall health state assessed by the HALex consists of two components, self-perceived health and activity limitation. By decomposing overall health inequality by component, one can identify inequality in which of these components contribute more to inequality in overall health. Finally, multidimensional decomposition is possible. One can decompose overall health inequality either jointly by subgroup and component [57] or by multiple subgroups at once (for example, race and income) [58]. The multidimensional decomposition has proven to be of great policy value in income inequality and poverty fields. With these developments, the decomposition technique is promising not only for summarizing diverse health inequality information but also for identifying determinants of health inequalities.
Although this study did not aim to investigate fully health inequality by race, the contrast between the greater health inequality across individuals and the smaller health inequality by race in 1995 than 1990 is worth emphasizing. One must receive this welcome finding with a caution. Differences in the average HALex between the three racial groups were the same in 1990 and 1995. Moreover, the greater overlap of the HALex distribution between racial groups came with the lower average HALex of all racial groups in 1995 than 1990. These findings of the HALex contrast with life years. Life expectancy both for Blacks and Whites were longer in 1995 than 1990 (Blacks: 69.1 in 1990, 76.1 in 1995, Whites: 69.6 in 1990 and 76.5 in 1995). The difference in life expectancy between Blacks and Whites was slightly smaller (0.1 year) in 1995 than 1990 because the difference in life expectancy between 1990 and 1995 was marginally greater for Blacks than Whites [49,50]. A follow-up of Healthy People 2000 found that all racial or ethnic groups, except American Indian or Alaska Native, reduced all-cause morality rates between 1990 and 1995 [33,34]. The categorization of racial groups is a major limitation of this study. The "other" group limits further meaningful comparisons of this study to other studies. Given that this study showed the average HALex was highest and its inequality was lowest in the "other" racial group both in 1990 and 1995, the further classification of this group would provide useful information on the health of different racial groups in the US.
Conclusion
Healthy People 2010 [5] aims to improve the average or overall health of Americans and to reduce health inequality among them. According to these two goals of population health, the assessment of population health is incomplete without analyzing health inequality. This present study is one example of health inequality analyses for the assessment of the health of Americans. Just as national health statistics routinely report the average or overall health of a population, health inequality needs to be routinely reported as an indicator of the health of Americans. Future work must determine which health inequalities can best serve as national health statistics.
Healthy People 2010 [5] also emphasizes the importance of paying attention to HRQL in the assessment of population health. As this study showed, inclusion of HRQL in the assessment of the health of Americans enriches our knowledge of population health and stimulate health policy debate.
Although the focus of the present study is primarily American, these key messages are internationally generalizable. This study should be of interest for health researchers and policy-makers in the US and elsewhere who wish to advance the assessment of population health.
List of Abbreviations
CI (confidence intervals)
HALex (Health and Activity Limitation Index)
HIV/AIDS (Human Immunodeficiency Virus / Acquired Immune Deficiency Syndrome)
HRQL (health-related quality of life)
NHIS (National Health Interview Survey)
US (United States of America)
WHO (World Health Organization)
Competing Interests
The author(s) declare that they have no competing interests.
Authors' Contributions
YA was solely responsible for this work.
Acknowledgements
I am indebted to Drs. David Kindig, John Mullahy, Patrick Remington, Alberto Palloni, Daniel Hausman, and Daniel Wikler for their general assistance for my dissertation, from which this present study was derived. I am also grateful to Drs. Nuala Kenny and George Kephart and reviewers for their comments on an earlier version of the manuscript. All remaining errors are mine. This project was supported by grant number 1 R03 HS 13116 from the Agency for Healthcare Research and Quality, and the Canadian Institute of Health Research Training Program for Ethics and Health Policy and Research.
==== Refs
Arias E Anderson RN Kung H Murphy SL Kochanek KD Deaths: final data for 2001 National Vital Statistics Reports 2003 52 1 115 14570230
National Center for Health Statistics Health, United States, with chartbook on trends in the health of Americans 2004 Hyattsville, Maryland
Health Canada Taking action on population health: a position paper for health promotion and programs branch staff 1999 Ottawa
World Health Organization The World Health Report 2000: Health systems: Improving performance 2000 Geneva
U.S. Department of Health and Human Services Healthy people 2010: Understanding and improving health 2000 Washington, DC
Mullahy J Live long, live well: Quantifying the health of heterogeneous populations Health Economics 2001 10 429 440 11466804 10.1002/hec.633
McDowell I Newll C Measuring health: A guide to rating scales and questionnaires 1996 2 New York: Oxford University Press
Gerdtham U-G Johannesson M Income-related inequality in life-years and quality-adjusted life-years Journal of Health Economics 2000 19 1007 1026 11186842 10.1016/S0167-6296(00)00055-2
Williams A Intergenerational equity: An exploration of the 'fair innings' argument Health Economics 1997 6 117 132 9158965 10.1002/(SICI)1099-1050(199703)6:2<117::AID-HEC256>3.0.CO;2-B
Asada Y Ohkusa Y Analysis of health-related quality of life (HRQL), its distribution, and its distribution by income in Japan, 1989 and 1998 Social Science and Medicine 2004 59 1423 1433 15246171 10.1016/j.socscimed.2004.01.025
Erickson P Wilson R Shannon I Years of Health Life Healthy People 2000, Statistical Notes 1995 7 1 14 11762385
U.S. Department of Health and Human Services Healthy people 2000: National health promotion and disease prevention objectives 1991 Washington, DC
Ko CY Maggard M Livingston EH Evaluating health utility in patients with melanoma, breast cancer, colon cancer, and lung cancer: A nationwide population-based assessment Journal of Surgical Research 2003 114 1 5 13678691 10.1016/S0022-4804(03)00167-7
Livingston EH Ko CY Use of the Health and Activity Limitation Index as a measure of quality of life in obesity Obesity Research 2002 10 824 832 12181392
Bradley CJ Kroll J Holmes-Rovner M The Health and Activities Limitation Index in patients with acute myocardial infarction Journal of Clinical Epidemiology 2000 53 555 562 10880773 10.1016/S0895-4356(99)00219-X
Murray CJL Gakidou EE Frenk J Health inequalities and social group differences: What should we measure? Bulletin of the World Health Organization 1999 77 537 543 10444876
Houweling TAJ Kunst AE Mackenbach JP World Health Report 2000: Inequality index and socioeconomic inequalities in mortality Lancet 2001 357 1671 1672 11425373 10.1016/S0140-6736(00)04829-7
Gakidou E King G Measuring total health inequality: Adding individual variation to group-level differences International Journal for Equity in Health 2002 1 1 12 12234390 10.1186/1475-9276-1-3
Kakwani N Wagstaff A van Doorslaer E Socioeconomic inequalities in health: Measurement, computation, and statistical inference Journal of Econometrics 1997 77 87 103 10.1016/S0304-4076(96)01807-6
Sitter RR Comparing three bootstrap methods for survey data The Canadian Journal of Statistics 1992 20 135 154
McCarthy PJ Snowden CB The bootstrap and finite population sampling Vital and Health Statistics Volume 2 1985 95 1 23
Jolliffe D Krushelnytskyy B Bootstrap standard errors for indices of inequality: INEQERR Stata Technical Bulletin 1999 51 1 6
Dagum C A new approach to the decomposition of the Gini income inequality ratio Empirical Economics 1997 22 515 531 10.1007/BF01205777
Lambert PJ Aronson JR Inequality decomposition analysis and the Gini coefficient revisited The Economic Journal 1993 103 1221 1227
Warner G A Lorenz Curve based index of income stratification The Review of Black Political Economy 2001 Winter 41 57
Yitzhaki S Lerman RI Income stratification and income inequality Review of Income and Wealth 1991 37 313 329 10.1111/j.1475-4991.1991.tb00374.x
Yitzhaki S Economic distance and overlapping of distributions Journal of Econometrics 1994 61 147 159 10.1016/0304-4076(94)90081-7
Pradhan M Sahn DE Younger SD Decomposing world health inequality Journal of Health Economics 2003 22 271 293 12606146 10.1016/S0167-6296(02)00123-6
Wagstaff A van Doorslaer E Overall versus socioeconomic health inequality: a measurement framework and two empirical illustrations Health Economics 2004 13 297 301 14981654 10.1002/hec.822
Clarke PM Gerdtham U-G Connelly LB A note on the decomposition of the health concentration index Health Economics 2003 12 511 516 12759920 10.1002/hec.767
Weinick RM Zuvekas SH Racial and ethnic differences in access to and use of health care services, 1977 to 1996 Medical Care Research and Review 2000 57 36 54 11092157 10.1177/107755800773743592
Committee on Understanding and Eliminating Racial and Ethnic Disparities in Health Care Unequal treatment: Confronting racial and ethnic disparities in health 2002 Washington, DC
Keppel KG Pearcy JN Wagener DK Trends in racial and ethnic-specific rates for the health status indicators: United States, 1990–98 Healthy People 2000: Statistical Notes 2002 23 1 16 11808619
Pearcy JN Keppel KG A summary measure of health disparity Public Health Reports 2002 117 273 280 12432138
National Center for Health Statistics Design and estimation for the National Health Interview Survey, 1985–94 Vital and Health Statistics, Series 2: Data Evaluation and Methods Research 1989 130 1 27
National Center for Health Statistics Design and estimation for the National Health Interview Survey, 1995–2004 Vital and Health Statistics, Series 2: Data Evaluation and Methods Research 2000 130 1 31
Greenacre M Correspondence analysis of the Spanish National Health Survey Gac Sanit 2002 16 160 170 11958753
Greenacre M Correspondence analysis in medical research Statistical Methods in Medical Research 1992 1 97 117 1341654
Drummond MF O'Brien B Stoddart GL Torrance GW Methods for the economic evaluation of health care programmes 1997 2 Oxford: Oxford University Press
Erickson P Evaluation of a population-based measure of quality of life – The Health and Activity Limitation Index (Halex) Quality of Life Research 1998 7 101 114 9523491 10.1023/A:1008897107977
Illsley R Le Grand J Williams A Measurement of inequality in health Health in economics 1987 London: Macmillan 12 36
Le Grand J Inequality in health: Some international comparison European Economic Review 1987 31 182 191 10.1016/0014-2921(87)90030-4
StataCorp Manual for stata statistical software: Release 60 1999 College Station, Texas
Lohr SL Sampling: Design and analysis 1999 Pacific Grove: Duxbury Press
Efron B Tibshirani RJ An introduction to the bootstrap 1993 Boca Raton: Chapman & Hall/CRC
Rao JNK Wu CFJ Resampling inference with complex survey data Journal of the American Statistical Association 1988 83 231 241
Rust KF Rao JNK Variance estimation for complex surveys using replication techniques Statistical Methods in Medical Research 1996 5 283 310 8931197
StateCorp Stata statistical software: Release 60 1999 College Station, Texas
National Center for Health Statistics Vital statistics of the United States, 1990, sec 6 life tables 1994 II Washington, DC
National Center for Health Statistics Vital statistics of the United States, 1995, preprint of mortality, part A sec 6 life tables 1998 II Hyattsville, Maryland
REVES (Réseau sur l'espérance de vie en santé) Year book
World Health Organization The World Health Report 2004: Changing history 2004 Geneva
Luxembourg Income Study Luxembourg Income Study key figures: Income inequality measures
U.S. Census Bureau Household shares of aggregate income by fifths of the income distribution: 1967 to 2001
Anderson RN Kochanek KD Murphy SL Report of final mortality statistics, 1995 Monthly Vital Statistics Report 1997 1 80
Families USA Going without health insurance: Nearly one in three non-elderly Americans
Mussard S Xu K A note on the multidimensional decomposition of Sen's Index
Wondon QT Between group inequality and targeted transfers Review of Income and Wealth 1999 45 21 39 10.1111/j.1475-4991.1999.tb00310.x
|
16014174
|
PMC1192818
|
CC BY
|
2021-01-04 16:37:42
|
no
|
Popul Health Metr. 2005 Jul 13; 3:7
|
utf-8
|
Popul Health Metr
| 2,005 |
10.1186/1478-7954-3-7
|
oa_comm
|
==== Front
Reprod Biol EndocrinolReproductive biology and endocrinology : RB&E1477-7827BioMed Central London 1477-7827-3-371610721110.1186/1477-7827-3-37ResearchMolecular cloning and characterization of a nuclear androgen receptor activated by 11-ketotestosterone Olsson Per-Erik [email protected] A Håkan [email protected] Hofsten Jonas [email protected] Birgitta [email protected] Anna [email protected] Anders [email protected] Johnny [email protected] Carina [email protected] Bertil [email protected] Peter [email protected] Department of Natural Science, Unit of Molecular Biology, Örebro University, SE-701 82 Örebro, Sweden2 Department of Marine Science, University of Texas Marine Science Institute, University of Texas, Port Aransas, Texas 78373, USA3 Department of Molecular Biology, Umeå University, SE-901 87 Umeå, Sweden4 Department of Zoology, Stockholm University, SE-106 91 Stockholm, Sweden2005 17 8 2005 3 37 37 26 5 2005 17 8 2005 Copyright © 2005 Olsson et al; licensee BioMed Central Ltd.2005Olsson 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 11-ketotestosterone is a potent androgen and induces male secondary sex characteristics in many teleosts, androgen receptors with high binding affinity for 11-ketotestosterone or preferential activation by 11-ketotestosterone have not been identified. So, the mechanism by which 11-ketotestosterone exhibits such high potency remains unclear. Recently we cloned the cDNA of an 11-ketotestosterone regulated protein, spiggin, from three-spined stickleback renal tissue. As spiggin is the only identified gene product regulated by 11-ketotestosterone, the stickleback kidney is ideal for determination of the mechanism of 11-ketotestosterone gene regulation. A single androgen receptor gene with two splicing variants, belonging to the androgen receptor-β subfamily was cloned from stickleback kidney. A high affinity, saturable, single class of androgen specific binding sites, with the characteristics of an androgen receptor, was identified in renal cytosolic and nuclear fractions. Measurement of ligand binding moieties in the cytosolic and nuclear fractions as well as to the recombinant receptor revealed lower affinity for 11-ketotestosterone than for dihydrotestosterone. Treatment with different androgens did not up-regulate androgen receptor mRNA level or increase receptor abundance, suggesting that auto-regulation is not involved in differential ligand activation. However, comparison of the trans-activation potential of the stickleback androgen receptor with the human androgen receptor, in both human HepG2 cells and zebrafish ZFL cells, revealed preferential activation by 11-ketotestosterone of the stickleback receptor, but not of the human receptor. These findings demonstrate the presence of a receptor preferentially activated by 11-ketotestosterone in the three-spined stickleback, so far the only one known in any animal.
==== Body
Introduction
Androgens have critical physiological roles in male sexual differentiation and in the development of male secondary sex characteristics. Androgens primarily mediate their actions through interactions with receptors belonging to the steroid hormone receptor super-family. Androgen receptors (AR) have been identified in many vertebrates and demonstrate similar binding characteristics in the various species investigated. Most vertebrates have one AR with high specificity for 5α-dihydrotestosterone (DHT). However, some teleosts have two AR with high binding affinities for either testosterone (T) or DHT [1,2]. 11-ketotestosterone (KT) is a major androgen in many teleosts and it often occurs at higher levels in the circulation and has a higher potency in inducing male reproductive functions than other androgens [3]. Although several AR have been isolated and characterized from teleost species, none of these have characteristics expected in a specific KT receptor. Consequently, an explanation for the high androgenic potency of KT in teleosts is currently not available.
AR have been cloned and characterized from several teleosts, including Japanese eel (Anguilla japonica), rainbow trout (Oncorhynchus mykiss), tilapia (Oreochromis niloticus) and red seabream (Pagrus major) [4-7]. Rainbow trout, tilapia and Japanese eel have two AR isoforms and while only one isoform (ARα) is a functional AR in rainbow trout, both Japanese eel AR are functional receptors [6,7]. Following transfection of human embryonic kidney 293 cells with either of the Japanese eel AR cDNAs it was observed that KT and DHT were equally potent activators of both AR isoforms, while T was significantly less potent in activating an MMTV-LTR driven luciferase reporter vector [4,6]. In contrast, the red sea bream AR was equally activated by T and KT via an MMTV promoter system in transfected COS-7 cells [5]. Similarly, a rainbow trout ARα reporter system in a carp EPC cell line was equally activated by DHT, KT and T [8]. Thus, none of the previously cloned AR genes show any preference for KT in trans-activation assays. A lack of specificity for KT in androgen signaling pathways was also suggested by androgen binding studies on tissues from Atlantic croaker (Micropogonias undulatus), kelp bass (Paralabrax clathratus), rainbow trout and goldfish (Carassius auratus) [1,2,8-10]. Based on tissue specific binding profiles there are two forms of AR in Atlantic croaker and kelp bass. One isoform, AR1, shows the highest binding affinity for T followed by DHT and KT while the other, AR2, shows the highest affinity for DHT followed by T and KT [1]. A limitation of the previous studies was that no KT-regulated gene had been identified in any of the studied species, and therefore no definite conclusion regarding endogenous gene regulation by KT could be drawn.
Hypertrophy of the kidneys and the production of a glue, spiggin, used in nest building by the kidneys of male three-spined sticklebacks (Gasterosteus aculeatus) during the breeding season, is currently the clearest example of a male reproductive process induced by KT [3,11]. Although treatment of stickleback with a variety of androgens can induce the kidney hypertrophy normally observed during the breeding season, KT is by far the most potent inducer of kidney hypertrophy and spiggin production [12,13]. These findings are of physiological importance because KT is the major plasma androgen during the breeding season in male stickleback [14]. However, a previous study showing that [3H]-KT was equally displaced from stickleback kidney tissue fragments by unlabelled KT or DHT provided no evidence for the presence of a specific KT AR in this species [15]. Thus, even in sticklebacks, where a specific KT-regulated gene product has been identified, the mechanism by which KT exhibits high androgenic potency remains unclear.
Due to the well-known regulation of spiggin by KT in the three-spined stickleback, the stickleback kidney is an ideal system in which to determine the mechanism by which KT regulates gene expression in teleosts and to determine the possible involvement of specific KT receptors in this signaling pathway. The binding characteristics of stickleback kidney cytosolic and nuclear preparations were comprehensively investigated in the present study to determine the androgen specificity of the AR(s) in this species. Furthermore, AR was cloned from stickleback kidney RNA and the molecular structures of two splicing variants, ARβ1 and ARβ2, were characterized. While the investigations of the binding affinities of androgens for the stickleback AR indicated that DHT bound better than KT to the receptor, comparison of the trans-activation potency of the stickleback AR with the human AR showed that KT was more potent at down-stream gene activation than DHT, in transfected HepG2 and ZFL cells, in the presence of the stickleback AR but not of the human AR.
Experimental
Materials
[3H]-DHT (92.84 Ci/mmol) and [3H]-KT (98.05 Ci/mmol) were purchased from New England Nuclear (Boston, MA, USA) and stored at -20°C. The unlabeled steroids were purchased from either Steraloids, Inc. (Wilton, NH, USA) or from Sigma Chemical Company (St. Louis, MO, USA). All radiolabelled steroids were stored in 95% ethanol at -20°C. Chemicals and salts used for making the buffers were purchased from Sigma and from Fisher Scientific (Pittsburgh, PA, USA). The scintillation cocktail was a mixture of 4 L toluene, 16 g PPO (7,5-diphenyl-oxazole), and 0.4 g POPOP (1,4-bis [5-phenyl-2-oxazolyl]-benzene).
Fish Maintenance
Adult stickleback were caught in Öresund and routinely housed in large aquaria containing brackish water (0.5% salinity) at 20°C under a photoperiod of 16:8 h light:dark (LD). A total of 300 male and female sticklebacks were used in the present study. The fish ranged in weight between 1 and 3 g. The water was aerated and filtered and the bottom was covered with sand. The fish were fed red midge larvae daily. Kidneys from males that had been brought into breeding condition by exposure to 20°C under a photoperiod of 16:8 h LD were used for the characterization of the AR. For the experimental treatments, the fish were initially in non-breeding condition. For the spiggin-experiment post-breeding females were used, these had been exposed to LD 16:8 and 17°C for a few months. For the other experimental treatments, fish were caught in non-reproductive seasons and housed under low temperature (9°C) and a photoperiod of LD 8:16 until a few days before treatment when they were put into high temperature and long photoperiod in order to acclimate. These investigations were approved by the Stockholm Northern Animal Experiment Ethical Committee (permit N 185/00).
Tissue sampling and steroid treatments
Steroid treatments were performed using females, as their kidneys do not undergo hypertrophy under natural conditions, and castrated males. The fish were anesthetized with 0.1% (v/v) 2-phenoxyethanol (Sigma, St. Louis, MO, USA) and implanted intraperitoneally with Silclear silicone tubing (10-mm length, 0.6-mm inner diameter, 1.2-mm outer diameter) containing steroids dissolved in cocoa butter. The steroids used were 11-ketoandrostenedione (KA), DHT, T and 17β-estradiol (E2) (1 or 25 μg μl-1). KA was used as this steroid is converted into KT by the fish [14]. Tubes containing cocoa butter alone were used as controls. Following implantation the fish were maintained in 50-liter aquaria containing brackish water (0.5% salinity) at 20°C under LD 16:8 h. At sampling fish were decapitated and the kidneys were excised, frozen using liquid nitrogen, and stored at -70°C. A number of experiments were performed as follows. Spiggin mRNA was measured in post-breeding females that were dissected 2 days following implantation of tubes. Cytosolic androgen-receptors were measured in females and in castrated males. Males were castrated under anesthesia as above, by making a 1.5 mm long incision into the abdominal cavity on each side and removing the testes using forceps. The females and the castrated males were implanted with control and high dose tubes and dissected after 6 days. AR mRNA was studied in females implanted for 12 h, 2 and 16 days (separate experiments) with control and high dose tubes and in males castrated as above and implanted for 10 days with control tubes and high doses of KA or E2.
Reverse transcriptase-Polymerase Chain Reaction
Total RNA was extracted from a pooled sample of five mature male kidneys using Tri Reagent™ (Sigma). The cDNA was synthesized from 1 μg of total RNA using the First Strand cDNA Synthesis Kit (Amersham Pharmacia Biotech, Buckinghamshire, UK). Amplification reactions were assembled using oligonucleotides based upon conserved regions in teleost AR (GeneBank accession numbers: AB012095, AB012096, AB017158, AB023960, AF121257, AB025361 and AF326200). These oligonucleotides were: forward (5'-GGGAAACAGAAATACCTGTGTG-3') and reverse (5'-CTCTGCAATCATCTCTGGAAAG-3'). Amplification was conducted for 40 cycles at 94°C for 30 s, at 40°C for 1 min and at 72°C for 1 min using a PTC-200 Thermal Cycler (MJ Research, Waltham, MA, USA). Amplified products were ligated into pGEM®-T (Promega, Madison, WI, USA) and recombinant plasmids were isolated using the Wizard® Plus SV Miniprep System (Promega). Cycle sequencing was performed using the DYEnamic ET Terminator Cycle Sequencing Premix Kit (Amersham Pharmacia Biotech, Piscataway, NJ, USA). The reactions were resolved on an ABI Prism™ 377 DNA Sequencer (Perkin-Elmer, Milano, Italy) and the data obtained were analyzed using EditView (Version 1.01) (Perkin-Elmer).
cDNA library screening
A unidirectional cDNA library, against mature male kidney cDNA, constructed in Lambda ZAP Express® (Stratagene, La Jolla, CA, USA) was used to screen for AR cDNA using a DIG-labeled AR cRNA probe. The screening was conducted according to the Lambda ZAP Express® manual (Stratagene). DIG labeled anti-sense RNA probes were generated using the DIG RNA Labeling Kit (Roche, Mannheim, Germany). Hybridization was performed at 45°C overnight (O/N) in hybridization buffer (5 × SSC, 50% formamid, 0.02% SDS (w/v), 0,1% N-laurylsarcosine (w/v) and 2% blocking solution (w/v)) (Roche). Membranes were washed for 2 × 5 min in 2 × SSC and 0.1% (w/v) SDS at room temperature, and for 2 × 15 min in 0.2 × SSC, 0.1% (w/v) SDS at 68°C. Signals were detected using CSPD (Roche) and exposure of Hyperfilm™-MP film (Amersham Pharmacia Biotech, Buckinghamshire, England) and hybridization signals were visualized using a CURIX 60 Film Developer (AGFA-Gevaert AB, Kista, Sweden). Positive plaques were purified through four successive hybridization rounds, and individual clones were isolated by phagemid excision. Following sequence identification of the clones as AR, they were sequenced to completion by Cybergene AB (Huddinge, Sweden).
Slot Blot Analysis
Total RNA was extracted using Tri Reagent™ (Sigma). Aliquots of 10 μg of total RNA were mixed with denaturing solution (6 × SSC, 7% (v/v) formaldehyde) and transferred onto a nylon membrane (Amersham Pharmacia Biotech) using a Minifold II Slot Blot Apparatus (Schleicher and Schuell, Keene, NH, USA). Membranes were probed using either a randomly primed [α-32P]-dCTP radiolabelled AR cDNA fragment (918 base pairs) that was isolated by reverse transcriptase-polymerase chain reaction and sequenced as above, or a randomly primed [α-32P]-dCTP radiolabelled spiggin-α cDNA. Hybridizations were performed at 65°C overnight (6 × SCC, 0.1% (w/v) SDS, 100 μg ml-1 tRNA, and 5 × Denhardt's solution). The membranes were washed for 2 × 30 minutes at 42°C and 65°C in 0.1 × SCC, 0.1% (w/v) SDS whereafter Hyperfilm™-MP film was exposed at -70°C. The films were visualized using Curix 60 Film Developer. Following AR and spiggin mRNA determination the membranes were stripped and analyzed for 18S rRNA using a cDNA fragment as internal control. Each AR or spiggin transcript was then semi-quantified using Quantity ONE 4.2.3 (Bio-Rad, Laboratories, Inc, Hercules, CA, USA) in relation to its internal 18S transcript.
Southern analyses of stickleback genomic DNA
Southern analyses were performed using 20 μg aliquots of male or female stickleback genomic DNA digested with 10 units of either SacI, BamHI or EcoRI at 37°C for 8 h according to Sambrook et al. [16]. Membranes were probed using [α32P]-dCTP radiolabelled partial cDNA encoding stickleback AR, which was isolated and sequenced as described above. Hybridizations were performed at 60°C O/N in 5 × SSC, 0.02% SDS (w/v), 0,1% N-laurylsarcosine (w/v) and 1% blocking solution (w/v) (Roche). Membranes were washed for 2 × 30 min periods at 42°C and 60°C in 0.1 × SSC, 0.1% (w/v) SDS. Membranes were exposed to Hyperfilm™-MP and were visualized as described above.
Sequence similarity analysis
The amino acid sequence alignments were made using sequences obtained from the GenBank sequence data bank with the following accession numbers: Human, Homo sapiens AR (M34233); Mouse, Mus musculus AR (X53779); Rat, Rattus norvegicus AR (M20133); Zebra finch, Taeniopygia guttata AR (AF532914); African clawed toad, Xenopus leavis AR (U67129); Tilapia ARα (AB045211); Tilapia ARβ (AB045212); Burton's mouthbreeder AR, Haplochromis burtoni AR (AF121257); Burton's mouthbreeder ARβ (AY082342); Japanese eel ARα (AB023960); Japanese eel ARβ (AB025361); Red seabream AR (AB017158); Three-spined stickleback ARβ (AY247207); Goldfish AR (AY090897); rainbow trout ARα (AB012095); rainbow trout ARβ (AB012096); Japanese medaka, Oryzias latipes ARα (AB076399). The sequence alignments were performed using Omiga 2.0 (Oxford Molecular Ltd). A phylogenetic tree was constructed using Tree view (Version 1.6.2) [17] following the alignment using the Clustal W algorithm (Version 1.7) [18] based on the GenBank sequences used for the sequence alignment.
Expression vector cloning
The excised AR pBK CMV clones were digested with XbaI and EcoRI and the inserts were gel purified prior to ligation into the pCMV-TNT vector (Promega) between XbaI and EcoRI. The obtained AR pCMV-TNT constructs were confirmed by sequencing. AR was expressed by introducing the pCMV-TNT into the TNT coupled reticulocyte lysate system (Promega) according to the manufacturer's instructions. The size of the expressed products was determined by addition of S35-methionin to the reticulocyte lysate system. The obtained protein suspension was analyzed on a denaturing 8% SDS PAGE and the sizes were determined using prestained SDS-PAGE broad range protein molecular weight standard (Bio-Rad) and a control luciferase protein of 62 kDa included in the reticulocyte lysate system (Promega). For use in binding assays AR was expressed in the absence of radioactive S35-methionein and stored at 4°C until further used in recombinant binding assay experiments.
Sample preparation for AR assays
The excised kidney was weighed and thereafter frozen in liquid nitrogen prior to storage at -80°C. Following thawing the samples were homogenized in 10 volumes ice cold homogenization buffer (50 mM Tris-HCl, 10 mM sodium molybdate, 1 mM EDTA, 12 mM monothioglycerol and 10% glycerol (v/v), pH 7.4) by using a Polytron Tissuemizer (Tekmar, Cincinnati, OH, USA) a setting 7 for 10 seconds followed by five passes in a glass homogenizer. The homogenate was centrifuged at 2,500 × g for 15 min. The supernatant was transferred to an ultracentrifuge tube and spun at 150,000 × g for 60 min. The supernatant was charcoal stripped by mixing 10 ml of the supernatant with 2.5 ml Dextran-coated charcoal buffer (DCC) (50 mM Tris-HCl, 1 mM EDTA, 10% glycerol (v/v), 1% Norit-A charcoal (w/v) and 0,1% Dextran T-70 (w/v), pH 7.5) and centrifuging for 10 min at 6,000 × g to pellet the charcoal and obtain the cytosolic fraction. The cytosolic extract was further diluted 5 × with homogenization buffer prior to binding assay analysis. The crude nuclear pellet obtained from the initial 2,500 × g spin was washed three times in ice-cold wash buffer (10 mM Tris-HCl, 2 mM MgCl2, 2 mM monothioglycerol, 250 mM sucrose and 10% glycerol (v/v), pH 7.5). The resulting pellet was resuspended in 5 ml extraction buffer (homogenization buffer supplemented with 0.7 M KCl), and incubated for 60 min at 4°C with vortexing at 15 min intervals. The preparation was centrifuged at 150,000 × g for 60 min and the supernatant provided the nuclear fraction. The preparations were either used immediately or kept frozen at -80°C until assayed.
Binding Assay
[3H]-DHT or [3H]-KT was dried under nitrogen, re-dissolved in 50 μl ice-cold homogenization buffer and added to assay tubes with or without a 100-fold excess of unlabeled DHT or KT. Aliquots (250 μl) of the nuclear or cytosolic preparations were added to each tube, the tubes were vortexed and incubated over night at 4°C. To stop the binding reaction, 250 μl DCC was added to each tube. The tubes were vortexed and the suspensions were incubated for 5 min at 4°C, before spinning at 3,000 × g at 4°C for 5 min. The supernatants were decanted into 7 ml scintillation vials, 5 ml of standard scintillation cocktail was added and the radioactivity in each sample was measured in a liquid scintillation counter (Beckman LS 6000SC, Beckman Instruments, Fullerton, CA). The procedure used for the expressed recombinant ARβ2 was modified as follows. [3H]-DHT was dried down under nitrogen, re-dissolved in 100 μl ice-cold homogenization buffer and added to assay tubes with or without a 1,000-fold excess of unlabeled DHT. The expressed proteins were diluted 1:100 and aliquots (100 μl) were added to each reaction tube. The tubes were vortexed and incubated over night at 4°C. To stop the binding reaction, 200 μl DCC was added to each tube. The tubes were vortexed and the suspension were incubated for 5 min at 4°C, before spinning at 3,000 × g at 4°C for 5 min. 250 μl of the supernatant was transferred into 5 ml scintillation vials and 4 ml of standard scintillation cocktail was added before the radioactivity in each sample was measured in a liquid scintillation counter (Wallac 1409, Wallac Oy, Turku, Finland).
Saturation and Scatchard analysis
[3H]-DHT (0.42 – 18.2 nM) or [3H]-KT (0.42 – 15.4 nM) was added to each reaction tube with or without a 100-fold excess of unlabeled steroid. For the expressed AR, [3H]-DHT (2.5 – 42 nM) was added to each reaction tube with or without a 1000-fold excess of unlabeled steroid. Nuclear or cytosolic suspensions were incubated with steroids over night at 4°C. The reactions were terminated and the radioactivity measured as described above. Non-linear curve fitting procedures (GraphPad Prism, version 3.03, GraphPad Software Inc) were used to calculate the dissociation constant (Kd) and to estimate the concentration of binding sites (Bmax) in the different suspensions.
Association and Dissociation kinetics
To investigate the time needed to reach binding equilibrium, cytosolic preparations were incubated in 4 nM [3H]-DHT with or without 400 nM unlabeled steroid. The reaction was terminated at different time-points ranging between 1 min and 24 h. The specific binding at each time-point was determined as described above. In order to investigate the receptor dissociation time, the preparation were incubated over night with 4 nM [3H]-DHT with or without 100-fold excess of unlabelled DHT for determining non-specific binding, before adding 100-fold excess of unlabelled DHT, and binding was subsequently measured between 5 min and 28 h.
Steroid specificity
The steroid specificity of receptor binding was examined using a competitive binding assay. Diluted protein suspension (1:100) were incubated over night at 4°C in tubes containing 1 nM [3H]-DHT with or without unlabelled steroid at end concentrations ranging between 1 pM and 100 μM. Steroids used were: KT; DHT; T; E2; 17,20β-dihydroxy-4-pregnen-3-one (17,20β-P); 17,20β,21-trihydroxy-4-pregnen-3-one, (20β-S) and cortisol (F). Following incubation, the reaction was terminated by addition of DCC and centrifugation and the specific binding in each tube was determined as described above.
Cell culture and transfection assays
Two liver epithelial cell lines were used to test receptor activity. HepG2 cells (ATCC, Manassas, VA) were maintained in DMEM supplemented with 10% FCS and non-essential amino acids. ZFL cells (ATCC) were maintained in 50% L-15 Leibovitz, 15% H-12 Ham and 35% DMEM supplemented with 10% FCS (Gibco, Paisley, Scotland, UK) and 50 ng ml-1 EGF (Sigma). Cells were cultured at 37°C (HepG2) or 28°C (ZFL) in 5% CO2. The transfection experiments were initiated by replacing the tissue culture medium with fresh phenol-free medium supplemented with charcoal stripped FCS and the cells were seeded on 24-well plates. Transfections were performed at 90–95% confluence using 1,5 μl Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) and 0,6 μg DNA per well. Transfected DNA contained 60 ng pRL (Promega), 270 ng androgen response element (ARE) luciferase reporter vector, 270 ng stickleback ARβ2 expression vector or 270 ng human AR expression vector (pCMVhAR) while 270 ng empty TNT cloning vector was used to control background luciferase levels following transfections. Two different ARE containing promoter constructs were tested, one containing 3 separate ARE (HRE; slp-ARU) and a second one construct (slp-HRE2) containing 4 copies of the ARE with high affinity for R1881, a synthetic androgen, and low affinity for dexamethasone, a synthetic glucocorticoid [19].
After 16 h the transfection medium was replaced with fresh phenol-free medium, supplemented with charcoal stripped FCS, containing different steroid hormones. The cells were exposed to the hormones for 40 h and luciferase levels were detected using the Dual Luciferase Assay kit (Promega) in a TD 20/20 Luminometer (Turner Designs, Sunnyvale, CA). For each cell line, transfection assay was performed for each concentration with n = 4 and the luciferase value of each assay were normalized to its corresponding Renilla luciferase activity. Each experiment was repeated a minimum of 3 times. The fold-induction is presented as luciferase values normalized against the control. The control levels were arbitrarily set to 1.0 for each cell line.
Statistical analyses
All data from semi-quantitative analyses of mRNA are shown as AR in relation to 18S expression and presented as mean ± SD. Significance were determined using ANOVA and Student t test.
Results
Spiggin induction
Implantation of females with 25 μg KA caused a 30-fold increase in spiggin mRNA levels. At this dose, KT was 10 times more effective than DHT and 50 times more effective than T at increasing kidney spiggin mRNA levels (Fig. 1). When implanted with 1 μg steroid a more pronounced difference was observed with KT being 23 times more potent than DHT and T not inducing spiggin production.
Figure 1 Induction of spiggin mRNA in adult female stickleback kidney by different steroids. KA, 11-ketoandrostenedione; DHT, 5α-dihydrotestosterone; T, testosterone; E2, 17β-estradiol (1 or 25 μg/μl in implants). All values represent the mean (±SD) values from 6 fishes. The mean value of the control group (C) was arbitrarily set to 1.00. Significant difference from the control is indicated by an asterisk (P < 0.01).
AR cloning and characterization
A 918 base pair internal fragment of the three-spined stickleback AR was isolated by RT-PCR and used to screen a mature male three-spined stickleback kidney lambda ZAP Express cDNA library. Several positive clones were obtained and two different size transcripts were identified. One longer sequence (3168 bp) coded for a partial AR, ARβ2 (Gene Bank Accession no AY247206), and one shorter sequence (2515 bp) coded for a full length AR, ARβ1 (Gene Bank Accession no AY247207). The ARβ1 sequence contained a 322 bp 5'-UTR (untranslated region) including an in frame stop codon, a 2046 bp coding sequence and a 147 bp 3'-UTR, and the gene coded for a 682 amino acid AR (ARβ1) protein. The ARβ2 sequence differed from ARβ1 in that it contained additional amino acid sequence in the N-terminal region, coding for 735 amino acids and also contained a longer (961 bp) 3'-UTR. The 3'-UTR of both transcripts was identical up to the poly-A tail of the shorter UTR (147 bp), indicating the presence of two poly-adenylation sites in the stickleback AR gene. The sequence of both transcripts exhibited a conserved identity (100%) in overlapping regions at both the nucleotide and protein levels, thus indicating that both were derived from a single locus by alternative splicing. Southern analysis of genomic material from both male and female stickleback demonstrated a hybridization pattern that was compatible with the existence of a single ARβ gene (Fig. 2).
Figure 2 Southern analysis of the AR locus. Stickleback genomic DNA (20 μg), from one male (lanes 1–3) and one female (lanes 4–6) kidney, digested with EcoRI (lane 1 and 4), SacI (lane 2 and 5) and BamHI (lane 3 and 6). The position of molecular weight markers (kb) is given in the right margin.
Sequence comparison analysis defined the three-spined stickleback AR as an ARβ isotype clustered with other teleost ARβ isotypes (Fig. 3). The closest overall similarity was found with the red seabream AR (76.1%) and the tilapia ARβ (70.9%) (Fig. 4). Low similarity was observed with ARα isoforms and with the mammalian AR. Determination of sequence similarities between different domains showed a high conservation of the DNA-binding domain (DBD) and the ligand binding domain (LBD), while the N-terminal trans-activation domains (AF1 and AF5) were less conserved (Fig. 4). In both AF1 and AF5, the stickleback ARβ showed the highest similarity to red seabream AR and tilapia ARβ.
Figure 3 Comparative sequence analysis of selected vertebrate AR proteins. The tree was constructed using Tree View (Version 1.6.2) following alignment of the protein sequences by the Clustal W algorithm (Version 1.7). GenBank Accession Nos.: human AR (M34233); mouse AR (X53779); rat AR (M20133); African clawed toad AR (U67129); tilapia ARα (AB045211); tilapia ARβ (AB045212); Burton's mouthbreeder AR (AF121257); Burton's mouthbreeder ARβ (AY082342); Japanese eel ARα (AB023960); Japanese eel ARβ (AB025361); red seabream AR (AB017158); three-spined stickleback ARβ (AY247207); goldfish AR (AY090897); rainbow trout ARα (AB012095); rainbow trout ARβ (AB012096); Japanese medaka ARα (AB076399). The numbers at the base of each clade division represent bootstrap values after 1000 repeats. Scale bar represents 0.1 amino acid replacements per amino acid site.
Figure 4 Percentage similarity of AR and specific AR domains. A, Schematic representation of the stickleback AR and the localization of specific domains. The AF1 (activator function 1) domain corresponds to amino acids (aa) 102–370 in the human AR. The AF5 (activator function 5) domain corresponds to aa 360–385 in the human AR. The DNA binding domain (DBD) corresponds to aa 550–635 in the human AR. The ligand binding domain (LBD) corresponds to aa 672–919 in the human AR. B, Percentage similarity to selected AR from different species.
Alignment of the three-spined stickleback AR LBD with other AR showed that the sequences were highly conserved and that the four amino acids thought to be involved in direct ligand-interactions with DHT in human AR (N705, Q711, R752 and T877), or amino acids with possible close contact to DHT (L704, M745 and F764), were conserved in the stickleback [20]. An arginine, located at position 779 in human AR that has been suggested to be of importance for ligand binding pocket architecture [21], has been substituted with a threonine in the stickleback sequence (Fig 5).
Figure 5 Sequence alignment of the ligand-binding region of AR. The alignment was performed using Omiga 2.0 (Oxford Molecular Ltd). Amino acid identity between sequences was illustrated with grey boxes and amino acid match with black boxes. Numbering is according to the human AR sequence. Amino acids (N705, Q711, R752, T877) with possible interaction with DHT in human AR are indicated by arrows from above. Amino acids with possible close contact to DHT (L704, M745 and F764) and of importance for ligand binding pocket architecture (R779) are indicated with arrows from below. GenBank Accession Nos.: Human AR (hAR) (M34233); Mouse AR (mAR) (X53779); Zebra finch AR (zAR) (AF532914); African clawed toad AR (aAR) (U67129); Tilapia ARα (tARα) (AB045211); Tilapia ARβ (t ARβ) (AB045212); Japanese eel ARα (jARα) (AB023960); Japanese eel ARβ (jARβ) (AB025361); Red seabream AR (rAR) (AB017158); Three-spined stickleback ARβ (sAR) (AY247207).
Both AR splicing variants were re-cloned into the pCMV-TNT expression vector in order to produce AR in vitro for binding assays and for expression in HepG2 and ZFL cells. S35-Met labeled AR was separated on a SDS-polyacrylamide gel to determine the size of the expressed proteins. From the nucleotide sequence, ARβ1 was estimated to code for a 76.6 kD protein while ARβ2 was estimated to code for a 82.2 kD protein. Both AR migrated as double bands following electrophoresis and the transcript sizes were 52 and 61 kD for ARβ1 and 75 and 87 kD for ARβ2 (Fig. 6). These results indicate that ARβ2 was properly translated in the reticulocyte system, while ARβ1 appeared as a truncated translation product.
Figure 6 In vitro translation products of three-spined stickleback ARβ1 and β2. AR was expressed by introducing the pCMV-TNT into the TNT Coupled Reticulocyte Lysate System. The size of the expressed products was determined by addition of S35-methionin to the reticulocyte lysate system. The protein suspensions were analyzed on a 8% SDS PAGE. Deduction of protein size from sequencing data yielded a expected size of 76.6 kDa for ARβ1 and 82.2 kDa for ARβ2. Estimation of translation product sizes resulted in 61 kDa and 52 kDa for ARβ1 and 87 kDa and 75 kDa for ARβ2.
AR binding characteristics
As a first step in the characterization of the AR binding characteristics in stickleback kidney, we determined the nature of interaction between ligand and receptor. Saturation ligand binding assays were performed on cytosolic, nuclear and membrane fractions of kidney extracts using either [3H]-DHT or -KT as tracer. In the initial studies we used [3H]-DHT to determine the presence of ARs in stickleback kidneys. High affinity and saturable binding was observed in both the cytosolic and nuclear fraction (Fig. 7A and 7E) while no specific binding to the membrane fraction was observed (data not shown). Next, using [3H]-KT we also observed high affinity and saturable binding sites in the kidney cytosol (Fig. 7C). Finally we used AR produced in the reticulocyte system to determine the characteristics of the cloned AR (Fig. 7G).
Figure 7 Representative saturation curves and Scatchard analyses. Saturation curves describing [3H]-DHT binding to AR in cytosolic fraction (A), [3H]-KT binding in cytosolic fraction (C), [3H]-DHT binding to AR in nuclear fraction (E) and [3H]-DHT binding to AR in reticulocyte lysate samples (G). Specific binding (▼) was determined by subtracting non-specific binding (■) from total binding (▲). Scatchard analyses of the specific binding of [3H]-DHT to cytosolic fraction (B), [3H]-KT to cytosolic fraction (D), of [3H]-DHT to nuclear fraction (F) and of [3H]-DHT to reticulocyte samples (H).
The saturation assay analysis, using [3H]-DHT as a tracer, was consistent with a single class of high affinity cytosolic AR with a Kd of 18.7 ± 3.21 nM and a Bmax of 5.61 ± 0.63 pmol/g tissue (Fig. 7B). Specific binding of [3H]-DHT was also present in the nuclear fraction where a Kd of 3,82 ± 0,26 nM and a Bmax of 0.67 ± 0.17 pmol/g tissue was observed (Fig. 7F). Saturation assay analysis, using [3H]-KT as a tracer, was also consistent with a single class of high affinity cytosolic AR with a Kd of 4.44 ± 1.61 nM and a Bmax of 1.06 ± 0.15 pmol/g tissue (Fig. 7D). Analysis of the recombinant ARβ2 using [3H]-DHT as a tracer also showed high affinity specific binding to reticulocyte extract with a Kd of 15.33 ± 2.50 nM and a Bmax of 0.324 ± 0.04 pmol/mg protein (Fig. 7H). The truncated translation product of ARβ1 did not give any measurable binding using the reticulocyte system and was therefore not further used in the present study.
Determinations of the binding kinetics in the kidney cytosolic fraction demonstrated rapid association between the ligand and receptor at 4°C. The binding of [3H]-DHT had a t1/2 of 3.3 min and reached equilibrium after 20 minutes (Fig. 8). The complete dissociation of [3H]-DHT bound to the receptor occurred within 8 hours with a t1/2 of 80 min (Fig. 8 insert).
Figure 8 Association and dissociation (inset) kinetics. [3H]-DHT binding to AR in three-spined stickleback kidney cytosolic extracts was determined at 4°C. The reactions were terminated at time-points ranging between 1 min and 28 h.
AR steroid specificity
Ligand competition assays were performed using 1 nM [3H]-DHT as a tracer to determine the relative affinity of steroid hormones to AR (Fig. 9). The binding curves were parallel, indicating competitive binding between the various unlabelled steroids and [3H]-DHT, allowing EC50 values to be determined. Using kidney extracts, the highest affinity binding was observed for DHT (EC50 = 1.31 ± 0.18 nM) while both KT and T had lower affinity (42.75 ± 3.42 nM and 68.81 ± 14.45 nM respectively). Binding was also observed for E2 (85.10 ± 17.87 nM), but not for 17,20β-dihydroxy-4-pregnen-3-one (17,20β-P), 17,20β,21-trihydroxy-4-pregnen-3-one (20β-S) or F (Fig 9A). Using ARβ2 containing reticulocyte extract, the highest affinity again was observed for DHT (EC50 = 0.67 ± 0.19 nM), followed by KT and T (2.18 ± 0.71 nM and 3.03 ± 0.43 nM), while E2 showed lower binding affinity (64.61 ± 0.28 nM) and F showed poor binding (Fig. 9B). Thus, in both cases DHT was found to have higher affinity than KT or T for the stickleback AR.
Figure 9 Competition curves for the binding of various natural steroids to AR in kidney. A, cytosolic extracts, and B, reticulocyte lysate samples. Varying concentrations of unlabelled steroids were incubated with 1 nM [3H]-DHT over night. Each data point represents the average of two assays. Steroids used were: KT, 11-ketotestosterone; DHT, 5α-dihydrotestosterone; T, testosterone; E2, 17β-estradiol; 17,20β-P, 17,20β-dihydroxy-4-pregnen-3-one; 20β-S, 17,20β,21-trihydroxy-4-pregnen-3-one; F, cortisol.
AR regulation
Determination of hormone specific regulation was performed by exposing three-spined sticklebacks for 6 days to DHT, KA, T or E2. None of the three tested androgens regulated AR protein as determined by radioreceptor assay, or AR mRNA as determined by slot blot. However, E2 was found to be a repressor of both AR protein and AR mRNA (data not shown).
Ligand specific AR gene activation
The ability of ARβ2 to regulate gene expression through an ARE-regulated luciferase vector was determined using both human HepG2 cells and zebrafish ZFL cells. Both DHT and KT showed dose-dependent activation via the stickleback AR in both HepG2 and ZFL cells (Fig. 10A) The results showed that KT induced luciferase activity 12 fold in ZFL cell and 10 in HepG2 cells, while DHT induced luciferase activity 4.5-fold in ZFL cells and 4.1-fold in HepG2 cells (Fig. 10A). The higher activation obtained with KT was significantly different from the activity obtained with DHT using both 10-8 M (P < 0.001 for both cell lines) and 10-6 M (p < 0.05 for both cell lines) steroid. Furthermore, comparison of the stickleback AR with the human AR, in the ZFL cell line, showed that the stickleback AR was preferentially activated by KT while the human AR did not discriminate between androgens (Fig. 10B).
Figure 10 Activation of stickleback ARβ2 and human AR in transfected cells. A, HepG2 (black) and ZFL (grey) cells were co-transfected with the ARE-luciferase vector, stickleback ARβ2 expression vector and the Renilla (pRL) control vector. The cells were treated during 40 hours with increasing concentrations (10-14 M to 10-6 M) of KT (circles) or DHT (triangles). Exposure to the two highest doses (10-8 M and 10-6 M) of both steroids was significantly different (p < 0.01) from the 10-10 M exposure and the control levels. B, ZFL cells were cotransfected with the ARE-luciferase vector, the Renilla (pRL) control vector, and the stickleback ARβ2 expression vector (grey) or the human AR expression vector (white). The cells were treated with 10-8 M of each steroid. The luciferase levels obtained with KT exposure were significantly different (p < 0.01) from DHT and T when using the stickleback AR but not the human AR expression vector. Statistically significant differences from control levels are indicated with an asterisk (p < 0.01). In both experiments the cells were treated with steroids for 40 h. Data were normalized against the untreated control for each cell line. The results are shown as mean ± SD (n = 4).
Discussion
The aim of the present study was to characterize key components of the AR signaling pathway through which KT mediates its induction of spiggin in the stickleback. Central to androgenic signaling is the interaction between the hormone and its receptor. So far no AR, which is preferentially activated by KT, has been identified in any animal. The only KT-induced gene product clearly identified to date is the underwater adhesive protein, spiggin [11,13], which is synthesized in the kidneys of male three-spined sticklebacks and used to construct a nest. While KT is the main inducer of spiggin production, other androgens will also increase spiggin synthesis in a dose-dependent fashion, although less efficiently. Measurements of androgen levels in stickleback have shown that KT is a prominent androgen in males during the breeding period [14,22]. It has been shown that the KT levels increased from 2 ng/ml during the prespawning period to peak at 40 ng/ml during spawning and to thereafter return to lower levels (≤1 ng/ml) in post spawning male sticklebacks [14,22]. Through these stages the T levels remain at about 1–4 ng/ml plasma. In spawning female stickleback the KT levels have been determined to be 1 ng/ml while the T levels are 20 ng/ml [22]. Circulating DHT levels have not been measured in the stickleback, but are likely to be low since neither DHT nor 5α-androstane-3,17-dione were formed in significant amounts (≤0.5%), if at all, when nesting stickleback testes tissue were incubated with tritiated androstenedione (reexamination of data on which [23] are based). While there are also no measurements of DHT levels in the stickleback kidney, 5α-reductase has been identified in stickleback kidney samples [24], thus suggesting that DHT may be locally formed and functional in stickleback. However, these studies suggest that KT is the main androgen in spawning male sticklebacks. In the present study KA was 10 times more potent than DHT and 50 times more potent than T at inducing spiggin mRNA (Fig. 1). In a previous study, using longer exposures (16 days) spiggin mRNA induction was only observed with KA [13]. These results clearly show that 11-ketoandrogens are by far the most effective inducers of spiggin mRNA. Furthermore, androgen induction of spiggin protein [25] and mRNA synthesis (Olsson, P.-E., Berg, A.H., Lindberg, C., unpublished results) can be blocked by the anti-androgen flutamide. This suggests that spiggin is induced, preferentially by KT, via a traditional AR, belonging to the nuclear steroid hormone receptor family.
Cloning and characterization of the stickleback AR demonstrated that there is a single AR gene locus that codes for a single receptor and that the ligand-binding characteristics of both the cloned and the endogenous receptor are indicative of a single class of high affinity AR (Fig. 7) with higher affinity for DHT than for KT or T (Fig 9). The binding kinetics shows that the AR present in the stickleback kidney has ligand-binding characteristics similar to those of previously characterized nuclear AR from other teleosts [1,2,9].
The stickleback AR shows the greatest similarity to AR from seabream and tilapia, fishes that like the stickleback, but unlike the rainbow trout and eel, belong to Series Percomorpha among the Acanthypterygii [26]. The receptor was found to have 2 splicing variants and 2 poly adenylation sites, and therefore could result in 4 possible different transcripts. The ligand-binding domain of the AR was found to be highly similar to other AR and contains the amino acids that are thought to be important for DHT interaction with the human AR [20,21]. The crystallographic structure of human AR indicates that N705, Q711, R752 and T877 have possible hydrogen bonds with DHT [20]. These 4 amino acids are conserved in teleost AR with one exception; in tilapia ARβ has a R752K substitution (GenBank AB045212). However, no data is available on the function of that receptor. Amino acids L704, M745 and F764 that are considered to have close contact with the ligand in human AR [20], are also conserved in teleost AR. An additional amino acid considered to be important for the architecture of AR is R779 [21]. Three mutations (R779A, R779Q and R779S) of the human AR were tested for trans-activation in COS cells and were found to lead to complete inactivation of AR [21]. It is interesting to note that this amino acid is not conserved among teleosts where both R779S and R779T are found. Neither of these alternative amino acids results in silencing of AR in teleosts as observed by transcriptional activation studies [5,6]. These results demonstrate that the AR ligand binding domains of teleosts and mammals are highly similar, although not identical.
Several previous studies have shown the presence of DHT and T receptors in teleosts, but so far no specific KT receptor has been identified [1,2,5-7]. Comparison of trans-activation efficacy between human AR and Japanese eel ARα and ARβ, using 293 cells, showed that KT was equally effective at activating gene transcription via either the human AR or Japanese eel ARβ [27], and that all three AR failed to distinguish between KT, DHT and T. Although several studies have determined either the ligand binding affinity of AR or the trans-activation of AR, there are few studies dealing with both aspects. However, in one study [8], the rainbow trout ARα ligand binding properties were determined in a heterologous system, the COS-1 cell line. The potency of binding was highest (IC50: 6 × 10-10 M), for DHT followed by T, while the poorest binding (IC50: 8 × 10-9 M) was observed for KT. In contrast to this, determination of the trans-activation potential using the carp epithelioma papillosum cyprini cell line indicated that the efficiency of activation of the rainbow trout ARα was equal for all three natural androgens [8]. Thus, there was an apparent discrepancy between the binding studies, showing that rainbow trout ARα had higher affinities for DHT and T than KT, and the reporter gene studies, showing that this difference was not reflected in activation potential [8]. In the present study we observed that DHT bound with higher affinity than KT to the native stickleback AR (Fig. 9A) and the in vitro translated recombinant stickleback ARβ2 (Fig. 9B). However, following transfection of the stickleback ARβ2-CMVTNT expression vector into either human HepG2 or zebrafish ZFL cells, together with either the slp-ARU or the slp-ARE, KT was in both cases approximately 2 to 3-fold more potent than DHT in activating the luciferase reporter gene (Fig. 10). In contrast to this the human AR was equally activated by DHT, KT and T (Fig. 10), which is in agreement with earlier studies [27]. Taken together the results are consistent with the data obtained on rainbow trout [8] and suggest that KT preferentially activate the stickleback ARβ2. The N-terminal domain has been suggested to be important for both trans-activation and protein-protein interactions and it is conceivable that there are ligand specific interactions between the LBD and the N-terminal trans-activation domain [28,29]. Evidence that ligand-dependent conformational changes may occur with the AR has been obtained with mesterolone, a synthetic androgen with an additional methyl group on carbon 1 of the A ring. Although mesterolone was found to have a similar binding affinity as DHT and T for both the wild type human AR and a mutant AR found in subfertile men, only mesterolone was able to restore mutant AR function to normal [30].
The 3-fold difference in trans-activation potential probably does not entirely explain the observed potency differences between KT and DHT in inducing spiggin in stickleback in vivo [13]. This result may partly be due to the different in the binding affinities of various androgens to the sex steroid binding proteins in fish blood. While KT has low affinity for binding to sex steroid binding proteins [31-36], both DHT and T show high affinities for these binding proteins in both mammals and teleosts [33,35]. The lower binding affinity of 11-ketoandrogens for sex steroid binding proteins in the plasma, and the resulting higher circulating levels of free steroid, may also contribute to the higher effectiveness of KT in inducing spiggin and kidney epithelium hypertrophy as well as male reproductive behavior in three-spined stickleback [12,13,37].
In contrast to estrogen receptors that are auto-regulated by estrogens, AR is not generally up-regulated by androgens and no ARE have been found in the promoter or 5'-flanking region of cloned AR. While AR is generally transcriptionally down regulated by androgens, the protein half-life appears to be increased by androgens [38,39]. Auto-regulation has, however, been observed in some tissues, including rat ventral prostate [40], the Harderian gland located in the orbital cavity of the golden hamster [41], male rat forebrain [42], human bone cells [43] and Atlantic croaker brain [44]. ARE have been identified in the coding region of AR cDNA from rat and shown to be functional, but require interaction with Myc family protein [45]. The ARE sequences (5'-TGTCCT-3') and (5'AGTACTCC-3') are separated by 182 bp in the rat AR cDNA and highly similar sequences are also found in other species, including the three-spined stickleback. However, despite the presence of possible AREs in the AR cDNA we did not observe any auto-regulation of kidney AR by androgens. E2 was the only steroid tested that altered AR mRNA and protein levels. Down-regulation of AR by estrogens is a common feature of AR, and E2 was shown to reduce both the AR mRNA and protein levels of stickleback kidney. This could be a component in the mechanism by which estrogens can suppress kidney hypertrophy in the stickleback [46]. In the present study we found no evidence that upregulation of AR mRNA or protein stability contributes to the KT specificity of spiggin induction. Thus, the stickleback AR is the first example of an AR preferentially activated by KT in any animal. Our results indicate that auto-regulation is not involved in this phenomenon. However, the lower binding affinity of KT than either DHT or T to sex steroid binding proteins in the circulation may be a contributing factor in vivo. As these proteins are not present in the in vitro systems the presently observed differences in ligand dependent transcriptional activation cannot be due to transport proteins.
Summary
The present study indicates the presence of a single gene coding for a nuclear AR in the three-spined stickleback. Furthermore, while the results show that the receptor has the highest binding affinity for DHT, it is preferentially activated by KT. While the present study represents the first identification of an AR preferentially regulated by KT, the elucidation of the KT signaling mechanism in teleosts clearly requires further research.
Acknowledgements
The ARE-luciferase (ARU-slp-Luc) reporter was a kind gift from Dr. Guy Verrijdt at the Division of Biochemistry, Faculty of Medicine, Campus Gasthuisberg, University of Leuven, Belgium. The human AR expression vector (pCMVhAR) was a kind gift from Dr. Elisabeth Wilson, USA. The present study was financed by grants from the Natural Science Research Council, Sweden (PEO), Swedish Animal Welfare Agency (PEO), and US Environmental Protection Agency STAR grant R827399 (PT).
==== Refs
Sperry TS Thomas P Identification of two nuclear androgen receptors in kelp bass (Paralabrax clathratus) and their binding affinities for xenobiotics: comparison with Atlantic croaker (Micropogonias undulatus) androgen receptors Biol Reprod 1999 61 1152 1161 10491657
Sperry TS Thomas P Characterization of two nuclear androgen receptors in Atlantic croaker: comparison of their biochemical properties and binding specificities Endocrinology 1999 140 1602 1611 10098494 10.1210/en.140.4.1602
Borg B Androgens in teleost fishes Comp Biochem Physiol 1994 109C 219 245
Todo T Ikeuchi T Kobayashi T Nagahama Y Fish androgen receptor: cDNA cloning, steroid activation of transcription in transfected mammalian cells, and tissue mRNA levels Biochem Biophys Res Commun 1999 254 378 383 9918846 10.1006/bbrc.1998.9919
Touhata K Kinoshita M Tokuda Y Toyohara H Sakaguchi M Yokoyama Y Yamashita S Sequence and expression of a cDNA encoding the red seabream androgen receptor Biochim Biophys Acta 1999 1450 481 485 10395960 10.1016/S0167-4889(99)00055-5
Ikeuchi T Todo T Kobayashi T Nagahama Y cDNA cloning of a novel androgen receptor subtype J Biol Chem 1999 274 25205 25209 10464240 10.1074/jbc.274.36.25205
Takeo J Yamashita S Two distinct isoforms of cDNA encoding rainbow trout androgen receptors J Biol Chem 1999 274 5674 5680 10026186 10.1074/jbc.274.9.5674
Takeo J Yamashita S Rainbow trout androgen receptor-alpha fails to distinguish between any of the natural androgens tested in transactivation assay, not just 11-ketotestosterone and testosterone Gen Comp Endocrinol 2000 117 200 206 10642442 10.1006/gcen.1999.7398
Sperry TS Thomas P Androgen binding profiles of two distinct nuclear androgen receptors in Atlantic croaker (Micropogonias undulatus) J Steroid Biochem Mol Biol 2000 73 93 103 10925208 10.1016/S0960-0760(00)00069-8
Wells KL Van der Kraak G Differential binding of endogenous steroids and chemicals to androgen receptors in rainbow trout and goldfish Environ Toxicol Chem 2000 19 2059 2065 10.1897/1551-5028(2000)019<2059:DBOESA>2.3.CO;2
Jakobsson S Borg B Haux C Hyllner SJ An 11-ketotestosterone induced kidney-secreted protein: the nest building glue from male three-spined stickleback, Gasterosteus aculeatus Fish Physiol Biochem 1999 20 79 85 10.1023/A:1007776016610
Borg B Antonopoulou E Andersson E Carlberg T Mayer I Effectiveness of several androgens in stimulating kidney hypertrophy, a secondary sexual character, in castrated male three-spined sticklebacks, Gasterosteus aculeatus Can J Zool 1993 71 2327 2329
Jones I Lindberg C Jakobsson S Hellqvist A Hellman U Borg B Olsson P-E Molecular cloning and characterization of spiggin. An androgen-regulated extraorganismal adhesive with structural similarities to von Willebrand factor-related proteins J Biol Chem 2001 276 17857 17863 11279178 10.1074/jbc.M101142200
Mayer I Borg BV Schulz R Seasonal changes in and effects of castration-androgen replacement on the plasma levels of five androgens in the male three-spined stickleback Gasterosteus acculeatus L Gen Comp Endocrinol 1990 79 23 30 2354778 10.1016/0016-6480(90)90084-Y
Jakobsson S Mayer I Schulz RW Blankenstein MA Borg B Specific binding of 11-ketotestosterone in an androgen target organ, the kidney of the male three-spined stickleback, Gasterosteus aculeatus Fish Physiol Biochem 1996 15 459 467 10.1007/BF01874920
Sambrook J Fritsch EF Maniatis T Molecular Cloning: A Laboratory Manual 1989 Cold Spring Harbor Laboratory, Cold Spring Harbor, NY
Page RDM An application to display phylogenetic trees on personal computers Comput Appl Biosci 1996 12 357 358 8902363
Thompson JD Higgins DG Gibson TJ CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice Nucleic Acids Res 1994 22 4673 4680 7984417
Verrijdt G Schauwaers K Haelens A Rombauts W Claessens F Functional interplay between two response elements with distinct binding characteristics dictates androgen specificity of the mouse sex-limited protein enhancer J Biol Chem 2002 277 5191 35201 10.1074/jbc.M205928200
Sack JS Kish KF Wang C Attar RM Kiefer SE An Y Wu GY Scheffler JE Salvati ME Krystek SR JrWeinmann R Einspahr HM Chrystallographic structures of the ligand-binding domains of the androgen receptor and its T877A mutant complexed with the natural agonist dihydrotestosterone Proc Natl Acad Sci USA 2001 98 4904 4909 11320241 10.1073/pnas.081565498
Poujol N Wurtz J-M Tahiri B Lumbroso S Nicolas J-C Moras D Sultan C Specific recognition of androgens by their nuclear receptor. A structure-function study J Biol Chem 2000 275 24022 24031 10787411 10.1074/jbc.M001999200
Borg B Mayer I Androgens and behaviour in the three-spined stickleback Behaviour 1995 132 1025 1035
Borg B Schoonen WGEJ Lambert JGD Steroid metabolism in the testes of the breeding and nonbreeding three-spined stickleback, Gasterosteus aculeatus Gen Comp Endocrinol 1989 73 40 45 2920898 10.1016/0016-6480(89)90053-1
Borg B Mayer I Lambert JGD Granneman J Schulz R Metabolism of androstenedione and 11-ketotestosterone in the kidney of the three-spined stickleback, Gasterosteus aculeatus Gen Comp Endocrinol 1992 86 248 256 1601274 10.1016/0016-6480(92)90108-V
Katsiadaki I Scott AP Mayer I he potential of the three-spined stickleback (Gasterosteus aculeatus L.) as a combined biomarker for oestrogens and androgens in European waters Mar Environ Res 2002 54 725 728 12408643 10.1016/S0141-1136(02)00110-1
Nelson JS Fishes of the World 1994 3 John Wiley & Sons, Inc. New York, Chichester, Brisbane, Toronto, Singapore
Ikeuchi T Todo T Kobayashi T Nagahama Y Two subtypes of androgen and progestogen receptors in fish testes Comp Biochem Physiol 2001 129B 449 455
Lee H-J Chang C Recent advances in androgen receptor action Cell Mol Life Sci 2003 60 1613 1622 14504652 10.1007/s00018-003-2309-3
Reid J Betney R Watt K McEwan IJ The androgen receptor transactivation domain: the interplay between protein conformation and protein-protein interactions Biochem Soc Transact 2003 31 1042 1046
Lim J Ghadessy FJ Abdullah AA Pinsky L Trifiro M Yong EL Human androgen receptor mutation disrupts ternary interactions between ligand, receptor domains, and the coactivator TIF2 (transcription intermediary factor 2) Mol Endocrinol 2000 14 1187 1197 10935543 10.1210/me.14.8.1187
Fostier A Breton B Binding of steroids by plasma of a teleost: the rainbow trout, Salmo gairdnerii J Steroid Biochem 1975 3–4 345 351 10.1016/0022-4731(75)90154-5
Chang C-F Lee Y-H Yoshida T Sun L-T Characterization of the plasma sex steroid-binding protein in eel (Anguilla japonica) Comp Biochem Physiol 1994 108B 189 197
Pasmanik M Callard G Characteristics of a testosterone-estradiol binding globulin (TEBG) in goldfish serum Biol Reprod 1986 35 838 845 3814697
Chang C-F Lee Y-H Purification of the sex steroid-binding protein from common carp (Cyprinus carpio) plasma Comp Biochem Physiol 1992 101B 587 590
Laidley CW Thomas P Partial characterization of a sex-steroid binding protein in the spotted seatrout (Cynoscion nebulosus) Biol Reprod 1994 51 982 992 7849199
Øvrevik J Stenersen J Nilssen K Tollefsen K-E Partial characterization of a sex steroid-binding protein in plasma from arctic charr (Salvelinus alpinus L.) Gen Comp Endocrinol 1994 122 31 39 10.1006/gcen.2001.7602
Borg B Stimulation of reproductive behaviour by aromatizable and non-aromatizable androgens in the male three-spined stickleback, Gasterosteus aculeatus Proc V Congr Europ Ichthyol, Stockholm 1985 1987 269 271
Kemppainen JA Lane MV Sar M Wilson EM Androgen receptor phosphorylation, turnover, nuclear transport, and transcriptional activation. Specificity for steroids and antihormones J Biol Chem 1992 267 968 974 1730684
Zhou ZX Lane MV Kemppainen JA French FS Wilson EM Specificity of ligand-dependent androgen receptor stabilization: receptor domain interactions influence ligand dissociation and receptor stability Mol Endocrinol 1995 9 208 218 7776971 10.1210/me.9.2.208
Mora GR Mahesh VB Autoregulation of androgen receptor in rat ventral prostate: involvement of c-fos as a negative regulator. Regulation of androgen receptor mRNA expression in primary culture of Harderian gland cells: cross-talk between steroid hormones Mol Cell Endocrinol 1996 124 111 120 9027330 10.1016/S0303-7207(96)03939-1
Esposito T Astore E Cardone A Angelini F Varriale B Regulation of androgen receptor mRNA expression in primary culture of Harderian gland cells: cross-talk between steroid hormones Comp Biochem Physiol 2002 132B 97 105
McAbee MD DonCarlos LL Estrogen, but not androgens, regulates androgen receptor messenger ribonucleic acid expression in the developing male rat forebrain Endocrinology 1999 140 3674 3681 10433226 10.1210/en.140.8.3674
Wiren KM Zhang X Chang C Keenan E Orwoll ES Transcriptional up-regulation of the human androgen receptor by androgen in bone cells Endocrinology 1997 138 2291 2300 9165014 10.1210/en.138.6.2291
Larsson DGJ Sperry TS Thomas P Regulation of androgen receptors in Atlantic croaker brains by testosterone and estradiol Gen Comp Endocrinol 2002 128 224 230 12392696 10.1016/S0016-6480(02)00503-8
Grad JM Dai JL Wu S Burnstein KL Multiple androgen response elements and a Myc consensus site in the androgen receptor (AR) coding region are involved in androgen-mediated up-regulation of AR messenger RNA Mol Endocrinol 1999 13 1896 1911 10551783 10.1210/me.13.11.1896
Oguro C Notes on the change in the kidney of Gasterosteus aculeatus aculeatus (L.) caused by estrogen administration J Fac Sci Hokkaido Univ Ser VI Zool 1957 13 404 407
|
16107211
|
PMC1192819
|
CC BY
|
2021-01-04 16:37:12
|
no
|
Reprod Biol Endocrinol. 2005 Aug 17; 3:37
|
utf-8
|
Reprod Biol Endocrinol
| 2,005 |
10.1186/1477-7827-3-37
|
oa_comm
|
==== Front
Thromb JThrombosis Journal1477-9560BioMed Central London 1477-9560-3-101604580410.1186/1477-9560-3-10Original Clinical InvestigationResistance to aspirin is increased by ST-elevation myocardial infarction and correlates with adenosine diphosphate levels Borna Catharina [email protected] Eduardo [email protected] Heusden Catharina [email protected]Öhlin Hans [email protected] David [email protected] Department of Cardiology, Heart & Lung Division, Lund University Hospital, Sweden2 Department of Medicine, University of North Carolina, School of Medicine, Chapel Hill, USA2005 26 7 2005 3 10 10 6 6 2005 26 7 2005 Copyright © 2005 Borna et al; licensee BioMed Central Ltd.2005Borna 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
To be fully activated platelets are dependent on two positive feedback loops; the formation of thromboxane A2 by cyclooxygenase in the platelets and the release of ADP. We wanted to evaluate the effect of aspirin on platelet function in patients with acute coronary syndromes and we hypothesized that increased levels of ADP in patients with acute coronary syndromes could contribute to aspirin resistance.
Methods
Platelet activity in 135 patients admitted for chest pain was assessed with PFA-100. An epinephrine-collagen cartridge (EPI-COLL) was used for the detection of aspirin resistance together with an ADP-collagen cartridge (ADP-COLL). ADP was measured with hplc from antecubital vein samples. Three subgroups were compared: chest pain with no sign of cardiac disease (NCD), NonST-elevation myocardial infarction (NSTEMI) and STEMI.
Results
Platelet activation was increased for the STEMI group compared NCD. Aspirin resistance defined as <193 sec in EPI-COLL was 9.7 % in NCD, and increased to 26.0 % (n.s.) in NSTEMI and 83.3 % (p < 0.001) in STEMI. Chronic aspirin treatment significantly reduced platelet aggregation in NCD and NSTEMI, but it had no effect in STEMI. Plasma levels of ADP were markedly increased in STEMI (905 ± 721 nmol/l, p < 0.01), but not in NSTEMI (317 ± 245), compared to NCD (334 ± 271, mean ± SD). ADP levels correlated with increased platelet activity measured with ADP-COLL (r = -0.30, p < 0.05). Aspirin resistant patients (EPI-COLL < 193 sec) had higher ADP levels compared to aspirin responders (734 ± 807 vs. 282 ± 187 nmol/l, mean ± SD, p < 0.05).
Conclusion
Platelets are activated and aspirin resistance is more frequent in STEMI, probably due to a general activation of platelets. ADP levels are increased in STEMI and correlates with platelet activation. Increased levels of ADP could be one reason for increased platelet activity and aspirin resistance.
aspirinacute coronary syndromesplateletsADP
==== Body
Background
To be fully activated platelets are dependent on two positive feedback loops; the formation of thromboxane A2 by cyclooxygenase in the platelets and the release of ADP from dense platelet granules. Thromboxane A2 and ADP then activates specific receptors on the extracellular side of the platelet membrane. Therapeutic intervention aimed at the first positive feedback loop by inhibiting cyclooxygenase with aspirin is highly efficient in reducing death and cardiovascular events by approximately 25% [1]. However, ADP may be even more important as evidenced by the CAPRIE-study, in which the ADP receptor antagonist clopidogrel was more beneficial than aspirin in reducing cardiovascular events [2]. Furthermore, the CURE and CREDO studies have established clopidogrel in combination with aspirin as a valuable treatment for acute coronary syndromes [3,4].
The platelet inhibitory effect of aspirin varies and aspirin resistance has been found in 9–45% of patients [5-7]. Little is known about the clinical consequences of aspirin resistance but lately two different studies indicate that aspirin resistance could be associated with an increased number of cardiovascular events [8,9].
Platelet activation is difficult to assess. Laboratory tests available are either not sufficiently reliable or rather complicated and therefore ineligible for clinical routine use. In this study we used a novel platelet function test, PFA-100. PFA-100 is an ex vivo assay of shear stress induced platelet adhesion and aggregation in whole blood. It simulates an injured blood vessel by a collagen-coated membrane together with either epinephrine or ADP. It has been found to be a sensitive test of aspirin resistance [6]. Aspirin resistance has previously been studied in healthy controls and in stable patients with a previous myocardial infarction.
In this study we wanted to evaluate the effect of aspirin on platelet function in patients with acute coronary syndromes. Furthermore, we hypothesized that increased levels of ADP in patients with acute coronary syndromes could contribute to aspirin resistance.
Methods
Patients
135 patients were enrolled from patients admitted for chest pain to the emergency ward, Lund University. Hospital between 2001–2003. Patients with chest pain within the last hour before admittance were eligible for inclusion. Patients were defined as aspirin users or patients not using aspirin the last three weeks (and not receiving aspirin during transport to hospital). The use of aspirin was defined as daily intake of aspirin for at least one week before admittance. Most patients were on aspirin 75 mg once daily, but a few (< 10%) were on 320 mg once daily. Exclusion criteria were: ingestion of clopidogrel, dipyridamole, nonsteroidal antiinflammatory drugs, heparin, low molecular heparin, warfarin, receiving bolus dose of aspirin on their way to hospital, platelet count <140 × 109/l, hemoglobin <90 g/l, renal failure (creatinine >140 μmol/l) or hemolysis in blood sample. The Human Ethics Committee of Lund University approved the project. All participants gave informed written consent before enrolment.
Based on the diagnosis at discharge, three prespecified subgroups were compared: chest pain with no sign of cardiac disease (NCD), Non ST-elevation myocardial infarction (NSTEMI) and STEMI. The NCD group presented no recent ECG changes and normal values of TNT. Exercise test where appropriate before discharge were negative. A minor number of patients in the NSTEMI and STEMI group were on beta-blocker, ACE-inhibitors and Ca-channel blockers. However, the groups were to small for subgroup analysis.
PFA-100 system testing
The PFA-100 system has been described in detail by Kundu and co-workers [10]. The PFA-100 uses a disposable test cartridge that simulates an injured blood vessel. The PFA-100 simulates primary haemostasis by flowing whole blood at a high shear rate through an aperture (147 μm diameter) cut into a collagen-coated membrane coated with either ADP (50 μg) or epinephrine (10 μg), where it comes into contact with the membrane surface and aggregate. A platelet plug forms, with occlusion of the aperture and cessation of blood flow. The closure time reflects platelet function in the sample evaluated. Shorter closure times indicate increased platelet aggregation. Testing was done in whole blood from antecubital vein samples anticoagulated with 3.8% sodium citrate. Samples were obtained at admission in 135 patients. PFA-100 tests were performed within 30 min after blood sampling, within 1 hour after admission to hospital. The epinephrine-collagen cartridge (EPI-COLL) is sensitive to aspirin and can be used for the detection of aspirin resistance [6]. The ADP-collagen cartridge (ADP-COLL) is only weakly sensitive to aspirin. Aspirin resistance was defined as normal EPI-COLL closure times (<193 sec) based on the 90% central interval in a normal population [6].
Nucleotide measurements
Nucleotides were measured in a total of 64 patients (16 patients with STEMI, 16 patients with NSTEMI and 32 patients with NCD). ADP was measured with hplc from antecubital vein samples. Sampling was done at admission. 5 ml blood was added to tubes containing citrate and immediately centrifuged for 10 min at 1200 G, 4°C. Platelet contamination was excluded by Burker chamber examination. The plasma was aspirated and mixed with an equal amount of 10% TCA to precipitate all proteins and inactivate ectonucleotideases. After centrifugation the protein free supernatant was frozen at -80°. Samples were sent on dry ice by courier, to Department of Medicine, University of North Carolina, School of Medicine, Chapel Hill, USA for analysis.
Samples were thawed at room temperature and TCA was extracted three times with six volumes of ethyl ether. Ethyl ether was removed by gassing N2, and the resulting samples diluted in the corresponding nucleotide assay buffer as indicated below.
Luciferin-luciferase assay
This assay has been previously described in detail elsewhere [11]. Typically, extracts were diluted 1:20 in HEPES-buffered Hanks Balanced Salt Solution (HBSS, pH. 7.4) and a 30-μl sample was added to a test tube and the volume adjusted to 300 μl with HPLC-grade H2O. The luciferin-luciferase reaction mix (100 μl) was added to tubes with a built-in injector into the light protected chamber of an Auto-Lumat LB953 luminometer. Luminescence was subsequently recorded during 10 seconds and compared against an ATP standard curve performed in parallel. Luminescence was linear with slope of one between 0.1 and 1000 nM ATP.
Derivatization of adenosine and adenine nucleotides
We have adopted and slightly modified a derivatization protocol originally described by Levitt and co-workers [12]. Sample extracts were incubated for 30 min at 72°C in the presence of 1.0 M chloroacetaldehyde and 25 mM Na2HPO4 (pH 4.0) in a 200-μl final volume. Samples were transferred to ice, alkalinized with 50 μl 0.5 M NH4HCO3, and analyzed by HPLC within 24 h. Identification and quantification of ethenylated species were performed with an automated Waters HPLC apparatus equipped with a fluorescence detector. Derivatized samples were transferred to 0.7 ml plastic shell vials and kept at 4°C in the sample injector rack. A 100 μl sample aliquot was injected into a 250-mm, 10 μm Hamilton PRP-X100 anion exchange column. The mobile phase (2 ml/min, 30 % methanol) developed linearly from 0.250 to 0.275 M NH4HCO3 (pH 8.5) during the first 8 min, remaining isocratic at 0.275 M NH4HCO3for additional 4 min. The column was subsequently rinsed for 3 min with 0.425 M NH4HCO3 in 30% methanol, and re-equilibrated to the initial conditions for 15 min. Elution times (in min) were: ε-ADO, 3.2; ε-AMP, 5.9; ε-ADP, 7.6, and ε-ATP, 9.4.
Reagents
[14C]glucose-1P (300 mCi/mmol) and molecular biology grade ATP and UTP were purchased from Amersham Pharmacia Biotech (Piscataway, NJ). ADP, AMP, and adenosine were from Roche Molecular Biochemicals (Indianapolis, IN). Etheno-adenyl standards were from Sigma (St. Louis, MO). Firefly luciferase and luciferin were purchased from PharMingen International (San Diego, CA). Other chemicals were of the highest purity available.
Calculation and statistics
Calculations and statistics were performed using the GraphPad Prism 3.02 software. Values are presented as mean ± SD. Statistical significance was accepted when P < 0.05 (two-tailed test). For continuous variables Kruskal-Wallis test followed by Dunnett multiple comparisons test was used. Spearman's rank correlation coefficient test was used for regression analysis. Categorical variables were compared using the Fisher exact test.
Ethics
The Ethics Committee of Lund University approved the project. The study complies with the Declaration of Helsinki. All patients gave written consent to participate in the study.
Results
Patient characteristics
The clinical characteristics of the NCD, NSTEMI and STEMI groups are shown in Table 1.
Table 1 The clinical characteristics of the NCD (no sign of cardiac disease), NSTEMI (non ST elevation myocardial infarction) and STEMI (ST elevation myocardial infarction) groups. Values are expressed as mean ± SD or numbers.
Characteristics Controls n = 67 NSTEMI n = 38 STEMI n = 30
Age 66 ± 12 72 ± 15 72 ± 13
M/F 46/21 29/9 21/9
Diabetes Mellitus 8 8 4
Prior IHD 29 18 12
Hemoglobine, g/l 135 ± 15 132 ± 15 132 ± 16
Platelet count, ×109/l 224 ± 74 218 ± 51 222 ± 66
Cholesterol mmol/l - 4,67 ± 0,94 4,60 ± 0,79
Triglycerides mmol/l - 1,19 ± 0,46 1,28 ± 0,94
BMI - 24 ± 2,8 27 ± 3,2
PFA-100
In patients without aspirin therapy there was an increased platelet activation in the STEMI group compared to NCD in the EPI-coll. NCD: 139 ± 44, NSTEMI: 121 ± 23 (n.s.), STEMI: 99 ± 28 sec (p < 0.001, mean ± SD). (Fig 1a.) These differences were also seen in patients on aspirin. NCD: 280 ± 41, NSTEMI: 243 ± 72 (n.s.), STEMI: 116 ± 56 sec (p < 0.001, Fig 2a). Lower values indicate increased platelet activation. Similar results were observed with ADP-COLL measurements (Fig 1b and 2b).
Figure 1 (a) Closure time measurements (epi-collagen) in NCD (no sign of cardiac disease), NSTEMI (non ST elevation myocardial infarction) and STEMI (ST elevation myocardial infarction) groups. (b) Closure time measurements (ADP-collagen) in the NCD, NSTEMI and STEMI groups. *** p < 0.001, compared to NCD. Lower values indicate increased platelet activation.
Figure 2 (a) Effect of chronic aspirin treament in the NCD (no sign of cardiac disease), NSTEMI (non ST elevation myocardial infarction) and STEMI (ST elevation myocardial infarction) groups measured as closure time with the EPI-collagen cartridge. (b) Effects of aspirin in the NCD, NSTEMI and STEMI groups measured as closure time with the ADP-collagen cartridge. White bars: no aspirin treatment, black bars: aspirin treated patients. Values are expressed as mean values ± SD, *** p < 0.001, * p < 0.05, n.s. = not significant, compared to NCD. Lower values indicate increased platelet activation.
Chronic aspirin treatment (at least one week before admission) significantly reduced platelet aggregation in NCD and NSTEMI as seen by increased PFA-100 times for EPI-COLL, but only in NCD for ADP-COLL (Fig 2). However, aspirin had no effect in either EPI-COLL or ADP-COLL in patients with STEMI (Fig 2). Aspirin resistance defined as <193 sec in EPI-COLL was 9.7 % in NCD, and increased to 26.0 % in NSTEMI (n.s.) and 83.3 % in STEMI(p < 0,001).
For PFA-100 measurements there were no significant correlation with age, diabetes, hemoglobin, CKMB, troponin T or platelet levels.
Nucleotide release
Plasma levels of ADP were markedly increased in STEMI (905 ± 721 nmol/l, p < 0.01), but not in NSTEMI (317 ± 245), compared to NCD (334 ± 271, mean ± SD) (Fig 3). Similar findings were found for other purines (ATP and AMP). ADP levels correlated with increased platelet activity measured with ADP-COLL in the whole material (r = -0.30, p < 0.05, Fig 4a). Similar results were seen for total purines (r = -0.30, p < 0.05, Fig 4b). There was a non-significant trend for EPI-COLL to correlate with both ADP and total purines in patients with aspirin treatment (ADP: r = -0.31, p = 0.09, total purines: r = -0.32, p = 0.09).
Figure 3 Plasma concentrations of extracellular purines in NCD (no sign of cardiac disease), NSTEMI (non ST elevation myocardial infarction) and STEMI (ST elevation myocardial infarction) groups expressed as mean values ± SD. * p < 0.05, ** p < 0.01, compared to NCD.
Figure 4 (a) Correlation between extracellular total purine levels and platelet activity measured as closure time with the ADP-COLL cartridge. (b) Correlation between extracellular ADP levels and platelet activity measured as closure time with the ADP-COLL cartridge.
Aspirin resistant patients (EPI-COLL < 193 sec) had higher ADP levels compared to aspirin responders (734 ± 807 vs. 282 ± 187 nmol/l, mean ± SD, p < 0.05), and increased levels of total purines (1615 ± 1493 vs. 737 ± 408 nmol/l, mean ± SD, p < 0.05).
For purine measurements there were no significant correlation with age, diabetes, haemoglobin, CKMB or troponin T levels. Platelet contamination was excluded by cell counting and we did not see any correlations between platelet counts and purine levels.
Nucleotide turnover is fast in whole blood due to ectonucleotidases. ATP degradation was evaluated both in samples with endogenous ATP and in samples were ATP had been added. The degradation was rapid in samples where ATP was added with a T 1/2 of 5.2 min. Endogenous ATP levels had a slower degradation rate, with a T 1/2 of approximately 30 min.
Baseline levels of adenosine were markedly lower than those of its nucleotides and barely detectable, most likely due to both rapid uptake into the red blood cells and degradation. This was because we have not included adenosine deaminase and nucleoside transport inhibitors in the perfused solution. The magnitude of changes in adenosine levels could therefore not be studied but this was not the aim of the study.
Discussion
In agreement with previous studies we found that platelets are activated in acute coronary syndromes [13,14]. Furthermore, for the first time we could demonstrate a rise in systemic levels of ADP and a decreased platelet inhibitory effect of aspirin in patients with STEMI. It is possible that the raised ADP level contributes to the increased platelet activity and the reduced effect of aspirin
There is growing evidence that a significant number of patients do not benefit from therapy with standard doses of aspirin. Aspirin resistance however, is a poorly defined term describing a number of conditions including the inability of aspirin to protect individuals from cardiovascular events, to inhibit platelet aggregation measured with a number of different methods and to inhibit thromboxane A2 formation [15]. We found aspirin resistance levels in controls (NCD) of 9.7%, which is in agreement with a previous study that found 9.5% aspirin resistance using PFA-100 [6]. It is also in agreement or even lower than other methods that found frequencies of 9–45%. [6-8]. interestingly, we found increased frequency of aspirin resistance in acute coronary syndromes rising from 9.7% in controls to 26.0 % in NSTEMI (n.s.) and 83.3 % in STEMI (p < 0,001). In fact, we could not see any significant effect of aspirin on platelet aggregation measured with PFA-100 in acute STEMI. Thus, in the situations where the patients need the platelet inhibitory effect of aspirin the most, the aspirin resistance is most frequent
The causative factors of aspirin resistance are still unclear. Altman et al., (2004) described several possible mechanisms behind aspirin resistance and the difficulty to interpret and compare the results of different studies [16]. Both the possibilities of inadequate doses of aspirin and thromboxane independent platelet aggregation mechanisms have been discussed. In some patients aspirin resistance could simply reflect non-compliance with drug therapy. Weber et al., (2002) suggested several possible types of aspirin resistance where one is linked to inability to inhibit thromboxane formation [15]. Another possible mechanism is classified as "pseudo resistance" since aspirin exerted the expected pharmacodynamic effect of inhibiting thromboxane formation but platelet aggregation was not inhibited. Hillarp et al., (2003) found that in a series of 200 aspirin treated patients, none was found to have unblocked cyclooxygenase activity [17]. Aspirin resistance despite blocked cyclooxygenase activity has been suggested to be explained by increased platelet sensivity to ADP and collagen [5,18].
Our data show that aspirin resistance increases in STEMI. This is probably due to a general activation of the platelets, because platelet aggregation was also increased in patients not treated with aspirin. The increased aspirin resistance could be explained by increased activity of important positive feedback systems such as ADP and thromboxane [19], but also by an increase in vW factor that are released from endothelial cells under high shear stress [20].
Several studies have shown sufficient inhibition of platelet aggregation with the 75 mg dose [1,20]. Recently though, a randomized study of 60 patients with stable coronary artery disease showed that the effect of aspirin was dose-dependent and the conclusion was that doses of less than 100 mg of aspirin was less effective inhibiting platelet aggregation than doses greater than 100 mg [22]. This could indicate that higher doses of aspirin could be necessary to inhibit increased activity in the thromboxane positive feedback system, but for the time being there is no clear evidence saying that higher doses of aspirin improve clinical outcome [1].
However, aspirin resistance could also be dependent on increased activity of the ADP positive feedback system. ADP is released by activated platelets, but also in the heart from cardiac myocytes during ischemia, or from endothelial cells, red blood cells and sympathetic nerves [22,23]. Erythrocytes are known to contain large amounts of ADP, which may increase the platelet activity and modulate the effect of aspirin [24]. It is known that nucleotides are released from numerous cells during stress and exercise [23]. The control group in this study was therefore chosen to present pain and stress and it is our belief that this control group is more relevant for comparison than unstressed healthy individuals to prove ischemia induced increases in ADP levels.
For the first time we have demonstrated in man that ADP levels are increased during myocardial infarction. It is possible that increased ADP levels during acute myocardial ischemia could contribute to the increased frequency of aspirin resistance. There are at least two subtypes of ADP receptors on the platelet. The P2Y12 receptor is linked via Gi protein to adenylate cyclase. The P2Y12 receptor stimulates platelet aggregation and has a high expression in platelets [25,26]. Clopidogrel blocks the P2Y12 receptor irreversibly and the value of this treatment has been established by the CURE and CAPRIE studies. P2Y1 is a Gq protein linked ADP receptor expressed in platelets that mediates shape change [25,26]. P2Y12 and thromboxane receptors act via different intracellular second messenger mechanisms, cAMP and inositol triphosphate (IP3), respectively. This explains the additive clinical effect of clopidogrel when it is combined with aspirin [3,4]. However, the P2Y1 receptor acts via the same second messenger system as thromboxane (IP3). Thus, high levels of ADP could replace thromboxane as stimulator of IP3 by activation of P2Y1 receptors. If ADP contributes to aspirin resistance, it may not be sufficient to block the P2Y12 receptor. It is possible that P2Y1 antagonists also will be necessary to achieve inhibition of both the important intracellular second messenger systems in the platelet.
Several studies have shown a higher incidence of cardiovascular events in patients reported to be aspirin resistant. Gum et al found that aspirin resistance was associated with an increased risk of death, myocardial infarction or cerebrovascular accident compared to aspirin sensitive patients (24% vs. 10%) [8]. Grotemeyer and colleagues described in 1993 a 40% risk for major events (stroke, myocardial infarction or vascular death) for aspirin resistant patients compared to a 4% risk in the aspirin responder group [7]. Eikelboom et al., (2002) recently reported that elevated urine concentrations of 11-dehydro thromboxane B2 could predict the risk of myocardial infarction [9]. Further prospective studies will be needed to evaluate if ADP levels could be of importance in the identification of patients with increased risk of myocardial infarction in the future. Why did the patients with aspirin resistance have higher ADP levels? It is possible that this was the result of a more pronounced cardiac ischemia. However, ADP levels did not correlate with either CKMB or troponin T levels. Another explanation could be that subgroups of patients are more dependent on the ADP positive feedback system than on the thromboxane system, and therefore more aspirin resistant. Then our finding of increased ADP levels may reflect an increased ADP release from platelets. These patents may benefit more from inhibitors of ADP mediated platelet activation than aspirin.
Limitations of the study
Patients found to be aspirin resistant were not all confirmed by a second test with the PFA-100. However, this was not possible in the NSTEMI and STEMI groups since they were treated early with clopidogrel, enoxaparin or GPIIb/IIIa-blockers that influence the PFA-100 measurements. ADP measurements are difficult because of the rapid degradation by ectonucleotidases being present predominantly on endothelial cells. We found a half-life of 5.2 min when ATP was added to our blood samples in vitro, however endogenous levels of ATP levels were more stable. The baseline ATP levels were in the micromolar range, which is similar to previously reported levels of circulating ATP in man [27]. Our sampling of venous blood in the antecubital vein is clearly not optimal to detect purine release in the heart. The released purine has passed both the lung and systemic circulation resulting in a degradation chain from ATP, ADP, AMP to adenosine. Adenosine is then rapidly taking up by the red blood cells. Thus the adenosine levels were barely detectable. It is our belief that blood sampling directly from heart veins would have resulted in markedly increased purine levels in the NSTEMI and STEMI groups and probably also better correlations with platelet activity. Another limitation is that the correlation analysis had to be done on the whole material since the subgroups were too small for separate correlation analysis.
Conclusion
Platelets are activated and aspirin resistance is more frequent in STEMI, probably due to a general activation of platelets. ADP levels are increased in STEMI and correlates with platelet activation. Increased levels of ADP could be one reason for increased platelet activity and aspirin resistance.
Authors' contributions
CB was the principal investigator for the study, responsible for recruitment of patients, aspirin resistance analysis, study design and wrote the manuscript.
CvH performed the purine analytical assays.
EL performed the purine analytical assays and participated in writing the manuscript.
HÖ participated in study design, helped in recruiting patients and wrote the manuscript.
DE conceived the study, guided throughout the study and wrote the manuscript.
All authors read and approved the final manuscript.
Acknowledgements
The study has been supported by the Swedish Heart and Lung Foundation, the Franke and Margareta Bergqvist Foundation, the Wiberg Foundation, the Bergwall Foundation, the Zoegas Foundation, the Westergren Foundation, the Swedish Medical Society, and the Swedish Medical Research Council, Grant 13130.
==== Refs
Antithrombotic Trialists' Collaboration Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients BMJ 2002 324 71 86 11786451 10.1136/bmj.324.7336.S71
CAPRIE Steering Committee A randomised, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE) Lancet 1996 348 1329 39 8918275 10.1016/S0140-6736(96)09457-3
The Clopidogrel in Unstable angina to prevent Recurrent Events (CURE) trial investigators Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation New Engl J Med 2001 345 494 502 11519503 10.1056/NEJMoa010746
Steinhubl SR Berger PB Mann JT 3rdFry ET DeLago A Wilmer C Topol EJ Early and sustained dual oral antiplatelet therapy following percutaneous coronary intervention: a randomized controlled trial JAMA 2002 288 2411 20 12435254 10.1001/jama.288.19.2411
Macchi L Christiaens L Brabant S Sorel N Allal J Mauco G Brizard A Resistance to aspirin in vitro is associated with increased platelet sensitivity to adenosine diphosphate Thromb Res 2002 107 45 49 12413588 10.1016/S0049-3848(02)00210-4
Gum PA Kottke-Marchant K Poggio ED Gurm H Welsh PA Brooks L Sapp SK Topol EJ Profile and prevalence of aspirin resistance in patients with cardiovascular disease Am J Cardiol 2001 88 230 235 11472699 10.1016/S0002-9149(01)01631-9
Grotemeyer KH Scharafinski HW Husstedt IW Two-year follow-up of aspirin responder and aspirin non responder. A pilot-study including 180 post-stroke patients Thromb Res 1993 71 397 403 8236166 10.1016/0049-3848(93)90164-J
Gum PA Kottke-Marchant K Welsh PA White J Topol E A prospective, blinded determination of the natural history of aspirin resistance among stable patients with cardiovascular disease J Am Coll Cardiol 2003 41 961 965 12651041 10.1016/S0735-1097(02)03014-0
Eikelboom JW Hirsh J Weitz JI Johnston M Yi Q Yusuf S Aspirin-resistant thromboxane biosynthesis and the risk of myocardial infarction, stroke, or cardiovascular death in patients at high risk for cardiovascular events Circulation 2002 105 1650 5 11940542 10.1161/01.CIR.0000013777.21160.07
Kundu SK Heilmann EJ Sio R Garcia C Davidson RM Ostgaard RA Description of an in vitro platelet function analyzer-PFA-100 Semin Thromb Hemost 1995 21 106 12 7660150
Lazarowski ER Boucher RC Harden TK Constitutive release of ATP and evidence for major contribution of ecto-nucleotide pyrophosphatase and nucleoside diphosphokinase to extracellular nucleotide concentrations J Biol Chem 2000 275 31061 31068 10913128 10.1074/jbc.M003255200
Levitt B Head RJ Westfall DP High-pressure liquid chromatographic fluorometric detection of adenosine and adenine nucleotides: application to endogenous content and electrically induced release of adenyl purines in guinea pig vas deferens Anal Biochem 1984 137 93 100 6731811 10.1016/0003-2697(84)90352-X
Kristensen SD Bath PM Martin JF Differences in bleeding time, aspirin sensitivity and adrenaline between acute myocardial infarction and unstable angina Cardiovasc Res 1990 24 19 23 2328510
Milner PC Martin JF Shortened bleeding time in acute myocardial infarction and its relation to platelet mass Br Med J (Clin Res Ed) 1985 290 1767 70 3924246
Weber AA Przytulski B Schanz A Hohlfeld T Schror K Towards a definition of aspirin resistance: a typological approach Platelets 2002 1 37 40 11918835 10.1080/09537100120104890
Altman R Luciardi HL Muntaner J Herrera RN The antithrombotic profile of aspirin. Aspirin resistance, or simply failure? Thromb J 2004 2 1 14723795 10.1186/1477-9560-2-1
Hillarp A Lethagen S Mattiasson I Aspirin resistance is not a common biochemical phenotype explained by unblocked cyclooxygenase-1 activity J Thromb Haemost 2003 1 196 197 12871563 10.1046/j.1538-7836.2003.00012.x
Kawasaki T Ozeki Y Igawa T Kambayashi J Increased platelet sensitivity to collagen in individuals resistant to low-dose aspirin Stroke 2000 31 591 5 10700490
Rasmanis G Vesterqvist O Green K Edhag O Henriksson P Effects of intermittent treatment with aspirin on thromboxane and prostacyclin formation in patients with acute myocardial infarction Lancet 1988 2 245 7 2899236 10.1016/S0140-6736(88)92537-8
Heper G Bayraktaroglu M The importance of von Willebrand factor level and heart rate changes in acute coronary syndromes: a comparison with chronic ischemic conditions Angiology 2003 54 287 99 12785021
Buerke M Pittroff W Meyer J Darius H Aspirin therapy: optimized platelet inhibition with different loading and maintenance doses Am Heart J 1995 130 465 72 7661062 10.1016/0002-8703(95)90353-4
Malhotra S Sharma YP Grover A Majumdar S Hanif SM Bhargava VK Bhatnagar A Pandhi P Effect of different aspirin doses on platelet aggregation in patients with stable coronary artery disease Intern Med J 2003 33 350 4 12895165 10.1046/j.1445-5994.2003.00360.x
Gordon JL Extracellular ATP: effects, sources and fate Biochem J 1986 233 309 19 3006665
Vassort G Adenosine 5'-Triphosphate: a P2-purinergic agonist in the myocardium Physiological Reviews 2001 767 806 11274344
Valles J Santos MT Aznar J Osa A Lago A Cosin J Sanchez E Broekman MJ Marcus AJ Erythrocyte promotion of platelet reactivity decreases the effectiveness of aspirin as an antithrombotic therapeutic modality: the effect of low-dose aspirin is less than optimal in patients with vascular disease due to prothrombotic effects of erythrocytes on platelet reactivity Circulation 1998 97 350 355 9468208
Storey RF Newby LJ Heptinstall S Effects of P2Y(1) and P2Y(12) receptor antagonists on platelet aggregation induced by different agonists in human whole blood Platelets 2001 12 443 7 11674863 10.1080/09537100120085450
Wang L Östberg O Wihlborg A-K Brogren H Jern S Erlinge D Quantification of ADP and ATP Receptor Expression in Human Platelets J Thromb & Haem 2003 1 330 336 10.1046/j.1538-7836.2003.00070.x
González-Alonso J Olsen DB Saltin B Erythrocyte and the regulation of human skeletal muscle blood flow and oxygen delivery. Role of circulating ATP Circ Res 2002 91 1046 1055 12456491 10.1161/01.RES.0000044939.73286.E2
|
16045804
|
PMC1192820
|
CC BY
|
2021-01-04 16:36:24
|
no
|
Thromb J. 2005 Jul 26; 3:10
|
utf-8
|
Thromb J
| 2,005 |
10.1186/1477-9560-3-10
|
oa_comm
|
==== Front
Thromb JThrombosis Journal1477-9560BioMed Central London 1477-9560-3-91604281010.1186/1477-9560-3-9Case ReportIntracardiac thrombus in Behçet's disease: Two case reports Hammami Sonia [email protected] Silvia [email protected] Khaldoun [email protected] Radhia [email protected] Habib [email protected] Farhat Mohamed [email protected] Department of Internal Medicine, F Bourguiba University Hospital, Monastir, Tunisia2 Department of Cardiology, F Bourguiba University Hospital, Monastir, Tunisia3 Department of Radiology, F Bourguiba University Hospital, Monastir, Tunisia2005 25 7 2005 3 9 9 27 5 2005 25 7 2005 Copyright © 2005 Hammami et al; licensee BioMed Central Ltd.2005Hammami 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.
Intracardiac thrombus in Behçet's disease is an extremely rare manifestation. We report two such cases. A 20-year-old man presented with dyspnoea, cough and haemoptysis. Right heart thrombus associated with pulmonary artery aneurysm and thromboembolism was identified by helical CT and transoesophageal echocardiography. The second case was a 29-year-old male admitted for fever and chest pain. A diagnosis of right atrial thrombosis associated with pulmonary embolism and hyperhomocysteinemia was made. Due to the absence of haemodynamic compromise, medical management consisting of immunosupressive and anticoagulation therapy was adopted which resulted in complete dissolution of the thrombus with dramatic clinical improvement in both cases of clinical status.
Conclusion: intracardiac thrombus is a rare complication of Behçet's disease. As shown in our patients, medical treatment should be considered as the first line.
Intracardiac thrombusBehcet's disease
==== Body
Background
Behcet's disease (BD) is a multisystemic inflammatory disease with a clinical spectrum that has greatly expanded since it was first described in 1937 as a triple complex of recurrent oral genital ulcers and uveitis. Cardiac involvement is extremely rare and often associated with poor prognosis [1]. We report two patients with a large intracardiac thrombus (ICTs) out of a series of 130 patients with BD.
Case reports
Case 1
A 20-year-old Tunisian man with a three weeks history of dyspnoea, cough and haemoptysis was admitted. At the age of 18, he had suffered from painful oral and genital ulcerations and polyarthralgias. At that time, examination revealed bilateral papillary oedema and brain magnetic resonance imaging showed superior sagittal and left lateral thromboses. The patient was given oral prednisone at the dose of 1 mg/kg/day that was tapered gradually and colchicine 1 mg/day in addition to acenocoumarol to maintain International Normalized Ratio between 2 and 3. These medications were discontinued per the patient 7 months later which resulted in the recurrence of aphtous ulcerations and papulopustular eruptions episodes.
On physical examination the patient had fever, face and neck oedema, prominent superficial thoracic venous collaterals, and pseudofolliculitis lesions. There was evidence of penile and scrotal scarring and minor aphthae on the buccal mucosa. Blood pressure was 110/80 mm Hg and pulse rate 100/min.
As to laboratory tests, haemoglobin concentration was of 9 g/dl, erythrocyte sedimentation rate of 60 mm/hr and C reactive protein concentration of 19 mg/l (normal < 2 mg/l). Renal and liver function tests were normal. Levels of serum IgG and IgM anticardiolipin, protein C, protein S, antithrombine III and total homocysteine were within the normal range. Electrocardiogram showed sinus rhythm tachycardia with no other abnormalities. Chest X-ray demonstrated hilar enlargement. Transthoracic and transoesophageal echocardiography showed multiple cardiac masses, one in the right atrium protruding through the tricuspid valve and two in the right ventricle (Figure 1 – see Additional file 1). Helical computed tomography (CT) showed multiple thrombi in both right atrium and ventricle extending into the superior vena cava. There was innominate and brachiocephalic vein occlusion, it also showed bilateral pulmonary embolism and multiple pulmonary infarcts in the lower lobe of the lungs. A large (14 mm in diameter) aneurysm located in the right basal segmental arteries (Figure 2 – see Additional file 2). We opted for a treatment based on low molecular weight heparin twice daily then oral anticoagulant, 1 gr of methyl prednisolone per day for 3 days, 1 mg/kg/day of oral, then tapered over 3 weeks and 1 g pulse cyslophosphamide monthly associated with colchicine 1 mg/day.
Figure 1 Transthoracic echocardiography: in apical four chamber view. (VD: right ventricle, VG: left ventricle, OD: right atrium, OG: left atrium) image of one thrombus in the right atrium, and two thrombi in the right ventricle (a), transoesophageal echocardiography showing a large thrombus (b), after treatment, complete resolution of the thrombus (c).
Figure 2 Chest helical computed tomography demonstrating a single (14 mm) right main pulmonary artery aneurysm and image of thrombi in the right heart (a), diffuse venous collateral vessels and superior vena cava thrombosis (b).
Two weeks later, oedema of the chest and neck has completely resolved. Thrombus size has substantially decreased. Nine months after discharge, no cardiac masses were detected by echocardiography and CT scan showed no evidence of previously mentioned thromboses with a complete disappearance of pulmonary aneurysm.
Case 2
A 29-year-old Tunisian man with a two-month history of fever of unknown origin, weight loss and inspiratory thoracic pain was admitted. He had suffered from both genital and oral ulcers over five months. The initial physical examination revealed a temperature of 38°C, multiple pseudofolliculitis, oral and scrotal ulcerations. Laboratory tests on admission revealed: haemoglobin of 13 g/dl, erythrocyte sedimentation of 105 mm/hr and C reactive protein concentration of 204 mg/l. The tests looking for antiphospholipid antibodies, protein C, protein S and antithrombine III deficiencies were negative. The plasma total homocysteine level was 27 μmol/l (normal < 10 μmol/l). HLAB5 and pathergy tests were positive. Electrocardiogram showed sinus rhythm tachycardia, and chest X-ray was normal. Transoesophageal echocardiography revealed a cardiac mass in the right atrium of 20/23 mm size attached into atrial septum protruding through the tricuspid valve into the right ventricle (Figure 3 – see Additional file 3). These findings were confirmed by CT scan that also showed a partial obstruction of the terminal portion of the inferior vena cava and thrombosis of the left lobar pulmonary artery with multiple pulmonary infarcts. We started treatment with intravenous heparin then oral anticoagulant, 1 gr of methyl prednisolone per day for 3 days, 1 mg/kg/day of oral, which was tapered gradually and 1 g pulse of cycslophosphamide monthly associated with colchicine 1 mg/day. Consequently, the thrombus in the right atrium has substantially decreased in size. At five months follow-up, a complete resolution of the thrombi in the right atrium, vena cava and pulmonary artery tree was observed.
Figure 3 Transthoracic echocardiography: image of the thrombus in the right atrium (a), after treatment complete resolution of the thrombus (b).
Discussion
Cardiac manifestations in BD which are indication of poor prognosis were reported to occur in about 1 – 5% of cases [1]. They consist of cardiomegaly, endocarditis or pericarditis and less commonly of myocardial infarction and myocarditis [2,3]. Association with intracardiac thrombus which is a serious complication, is even more rare, up to this date, less than 50 cases have been reported so far [4-8].
Our patients fulfilled the proposed criteria of the international study group for BD [9], with active disease (two or more active clinical features related to BD). Two interesting issues about these cases should be emphasized: first, the unusual presentation of BD with ICTs and pulmonary thromboembolism, second, the favourable response to medical management. The association of intracardiac thrombosis, with the less uncommon pulmonary arteritis and vena cava thrombosis was described for the first time by Houman [10] and reported in only few cases so far [4]. In our patients, we can reasonably exclude retrospectively myxoma and endocarditis, in view of the mass resolution on immunosuppressive and anticoagulation therapy.
Biopsy carries an excessive risk [2], but has the advantage of providing material for histological examination. The organized thrombus usually contained an inflammatory cell infiltrate composed of a mixture of granulocytes and mononuclear inflammatory cells or predominantly lymphocytes. The histologic descriptions of the thrombi may be dependent on the biopsy timing [4]. We did not perform a right ventricular biopsy in our patients.
The pathogenic mechanism underlying thrombotic tendency in patients with BD is not well known. It is however believed to be due to endothelial cell ischemia or disruption that leads to enhancement of platelet aggregation [11]. Also decreased release of vascular tissue plasminogen activator has been reported in systemic and cutaneous vasculitis [12]. Another possible pathogenic mechanism of thrombosis in BD is attributed to the presence of anti phospholipid antibodies which is reported to be present in 18% of cases [13,14]. Elevated Von Willebrand factor antigen levels have recently been demonstrated [15]. Hyperhomocysteinemia was reported to be present in patients with BD and was associated with increased risk of vascular thrombosis [16], which may have contributed to cardiac thrombus formation in our second patient. To the best of our knowledge this association has not been previously reported.
Diagnosis may be confirmed either at necropsy or after surgery. In our cases, this complication of BD was diagnosed by echocardiography. Other imaging modalities including CT scan and MRI can show vascular complications and give information concerning the lung parenchyma. The management of ICTs is still controversial [4]. Surgical removal has the advantage of providing material for histological examination; medical management however was associated with a better outcome. Mogulkoc's review about 24 cases of ICTs, surgery was unsuccessful in 4/12 cases, whereas complete resolution of the thrombus on medical therapy was observed in 7/8 [4].
Anticoagulant or thrombolytic therapy was the first line treatment of intracardiac thrombus [4]. In the presence of pulmonary aneurysms this therapy could lead to fatal haemoptysis especially in bilateral and large aneurysms. In our first case ICTs was associated with multiple venous thrombi and only a small aneurysm with no haemodynamic compromise. Considering the risk of surgical treatment we preferred a conservative approach with immunosuppressive and anticoagulant therapy.
We conclude that thrombi especially in the right heart cavities can be present in BD without causing specific symptoms but can lead to pulmonary embolism. Early echocardiography seems advisable to detect the presence of cardiac involvement and medical therapy can be effective in the resolution of ICTs in the setting of BD.
List of abbreviations used
BD: Behcet's disease
ICTs: IntraCardiac Thrombus
CT: Helical Computed Tomography
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
S.H.: conceived of the study, tacked out the responsibility for diagnosis of patients, their evolution and elaborated in the design of these 2 cases and draft the manuscript; S.M. carried out the therapy evolution; K.B.H.: interested in the cardiac aspects and echocardiography diagnosis; R.B.: tacked out X-ray, MRI and CT examinations; H.G and M.B.H: 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 patients and their relative for publication of study. We would like to thank Pr M. Hammami for his biochemistry assistance to homocysteinemia analysis.
==== Refs
Wechsler B Du LT Kieffer E Cardiovascular manifestations of Behçet's disease Ann Med Interne 1999 150 542 54
BayKan M Celik S Erdol C Baykan EC Durmus I Bahadir S Erdol H Orem C Cakirbay H Behçet's disease with a large intracardiac thrombus: a case report Heart 2001 85 E7 11250984 10.1136/heart.85.4.e7
Le Thi Huong D Wechsler B Papo T Zuette D Bletry O Hernigou A Delcourt A Godeau P Piette JC Endomyocardial fibrosis in Behçet's disease Ann Reum Dis 1997 56 205 8
Mogulkoc N Burgess MI Bishop PW Intracardiac thrombus in Behçet's disease. A systematic review Chest 2000 118 479 87 10936144 10.1378/chest.118.2.479
Yochida S Fujimori K Hareyama M Nakata T Cardiac thrombus in Behçet's disease Chest 2001 120 688 9 10.1378/chest.120.2.688
Duzgun N Anil C Ozer F Acican T The disappearance of pulmonary artery aneurysms and intracardiac thrombus with immunosuppressive treatment in a patient with Behçet's disease Clin Exp Rheumatol 2002 20 S56 7 12371638
Kaya A Ertan C Gurkan OU Fitoz S Atasoy C Kilickap M Numanoglu N Behçet's disease with right ventricle thrombus and bilateral artery aneurysms: a case report Angiology 2004 55 573 5 15378123
Ben Ghorbel I Ibn Elhadj Z Khanfir M Braham A Fekih M Drissa H Houman MH Intracardiac thrombus in Behçet's disease. A report of three cases J Mal Vasc 2004 29 159 61 15343111
International Study Group for Behçet's disease Criteria for diagnosis of Behçet's disease Lancet 1990 335 1078 80 1970380
Houman M Ksontini I Ben Ghorbel I Lamloum M Braham A Mnif E Miled M Association of right heart thrombosis, endomyocardial fibrosis and pulmonary artery aneurysm in Behçet's disease Eur J Inter Med 2002 13 455 7 10.1016/S0953-6205(02)00134-6
Schmitz-Huebner U Knop J Evidence for an endothelial cell dysfunction in association with Behçet's disease Thromb Res 1984 34 277 85 6429888 10.1016/0049-3848(84)90384-0
Jordan JM Allen NB Pizzo SV Defective release of tissue plasminogen activator in systemic and cutaneous vasculitis Am J Med 1987 82 397 400 3103439 10.1016/0002-9343(87)90436-0
Hull RG Harris EN Gharavi AE Tincani A Asherson RA Valesini G Denman AM Froude G Hughes GR Anticardiolipin antibodies: occurence in Behçet's syndrome Ann Rheum Dis 1984 43 746 8 6497467
Harmouche H Tazi Mezalek Z Adnaoui M Aouni M Mohattane A Maaouni A Berbich A Association of pulmonary artery aneurysm, right heart thromboses and antiphospholipid antibodies in Behçet's disease Rev Med Interne 1998 19 512 5 9775202
Direskeneli H Keser G D'cruz D Khamashta MA Akoglu T Yazini H Yurdakul S Hamuryudan V Ozgun S Goral AJ Antiendothelial cell antibodies, endothelial proliferation and Von Willebrand factor antigen in Behçet's disease Clin Rheumatol 1995 14 55 61 7743745
Ates A Aydintug O Olmez U Duzgun N Duman M Serum homocystein level is higher in Behçet's disease with vascular involvement Rheumatol Int 2005 25 42 4 14586553 10.1007/s00296-003-0398-9
|
16042810
|
PMC1192821
|
CC BY
|
2021-01-04 16:36:24
|
no
|
Thromb J. 2005 Jul 25; 3:9
|
utf-8
|
Thromb J
| 2,005 |
10.1186/1477-9560-3-9
|
oa_comm
|
==== Front
World J Surg OncolWorld Journal of Surgical Oncology1477-7819BioMed Central London 1477-7819-3-411598750910.1186/1477-7819-3-41Case ReportSentinel node biopsy as an adjunct to limb salvage surgery for epithelioid sarcoma of the hand Seal Alex [email protected] Raymond [email protected] Bret [email protected] Alex [email protected] Claire LF [email protected] University of British Columbia, Vancouver, British Columbia, Canada2 Division of Plastic Surgery, University of Western Ontario, London, Ontario, Canada3 Department of Pathology, University of Western Ontario, London, Ontario, Canada4 Department of Radiation Oncology, London Region Cancer Centre, London, Ontario, Canada2005 29 6 2005 3 41 41 22 4 2005 29 6 2005 Copyright © 2005 Seal et al; licensee BioMed Central Ltd.2005Seal 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
Epithelioid sarcomas of the hand are rare, high-grade tumors with a propensity for regional lymphatic spread approaching 40%.
Case presentation
A 54-year-old male with an epithelioid sarcoma of the palm was treated with neoadjuvant radiation, wide excision, and two-stage reconstruction. Sentinel lymph node biopsy was used to stage the patient's axilla. Sentinel node biopsy results were negative. The patient has remained free of local, regional and distant disease for the follow-up time of 16 months.
Conclusion
The rarity of this tumor makes definitive conclusions difficult but SLN biopsy appears to be a useful adjunct in the treatment of these sarcomas.
==== Body
Background
Epithelioid sarcoma is a rare, high-grade, soft tissue sarcoma. These tumors typically present on the extremities, in males who are 20 to 30 years of age. Overall 5 and 10-year survival rates are 70% and 42% respectively [1]. Epithelioid sarcoma is among a group of sarcomas with a propensity for regional lymphatic spread, with lymph node metastasis rates reported between 17%-80% [2-7]. Due to the risk of regional spread, sentinel lymph node biopsy (SLN) may be useful in the management of this tumor.
The success of SLN biopsy is based on the principle that the primary tumor drains to one or a few lymph nodes in the regional basin. Histopathological analysis of these sentinel nodes has been shown to reflect the histology of the entire lymphatic basin [8,9]. This approach is currently the least invasive and most accurate nodal staging procedure for breast cancer and melanoma [10,11]. However, SLN biopsy has not been thoroughly investigated for sarcoma.
We report a case of SLN biopsy in conjunction with limb salvage surgery and complex soft tissue, neurovascular, and staged tendon reconstruction for the management of an epithelioid sarcoma of the hand.
Case presentation
An otherwise healthy 54-year-old right-handed laborer presented with an eight-year history of a slowly enlarging "callus" in the palm of his right hand. He underwent excision at an outside institution. The operative note suggested that the tumor had the appearance of a sebaceous cyst with a large amount of surrounding tissue reaction. The neurovascular bundles were identified and preserved to the 4th web space. An epithelioid sarcoma was identified upon histological analysis. Review of surgical pathology at our institution confirmed this diagnosis (Figure 1a and 1b). The excision was incomplete with tumor extending to several margins.
Figure 1 Epithelioid sarcoma a) Conglomerate of tumor nodules with central necrosis mimicking a necrotizing granulomatous process at low magnification (hematoxylin and eosin ×4) b) Epithelioid cells with abundant eosinophilic cytoplasm and prominent nuclear atypia, appreciated at high magnification (hematoxylin and eosin ×40), distinguishes epithelioid sarcoma from its benign mimics. Note the presence of a mitotic figure.
On examination, the patient had a transverse scar just proximal to the 4th web space, with no palpable tumor. Neurovascular exam was normal, with normal range of motion of the associated digits. There were several small, palpable ipsilateral axillary nodes.
MRI of the right hand showed an ill-defined signal change within the palmar subcutaneous fat just deep to the surgical incision and distal to the 4th and 5th metacarpal-phalangeal joints, consistent with post-surgical changes and scarring. No discrete soft tissue mass was seen to suggest gross tumor. Edema signal was seen in the distal aspect of the lumbrical muscle between the ring and small flexor tendons. The interosseous muscles appeared uninvolved. No osseous or articular abnormalities were identified.
Preoperative computerized tomography (CT) of the chest was negative. The axilla was reported as having benign appearing lymph nodes, fatty in nature with no evidence of necrosis.
The patient received preoperative radiation of 50 Gy in 25 fractions. CT simulation was used for planning the gross tumor volume (GTV) to ascertain the depth to be treated using electrons. A customized lead cutout was designed to avoid treating the full width of the hand and to avoid normal tissues ulnarly, radially and at depth. Bolus was placed over the palm to ensure adequate superficial skin and scar dose whilst ensuring the dose at depth covered the tumor and previous operative bed. A daily dose of 2 Gy was delivered to a total of 50 Gy over a five week period. Moderate erythema of the palm occurred which healed well post-treatment. There was no significant edema post-irradiation. Dysethesias were reported on the radial aspect of the small finger; however, two point discrimination remained normal.
Surgery was performed 6 weeks following completion of radiotherapy. Preoperative lymphoscintography (figure 2) identified a single "hot" ipsilateral axillary lymph node, which was successfully removed. Vital blue dye was not used, as it was felt that the blue stained tissues would worsen visibility for the wide excision.
Figure 2 Preoperative lymphoscintigraphy identifies uptake of the radiolabelled tracer in a single axillary lymph node.
Wide, en bloc excision was carried out including palmar skin, subcutaneous fat, palmar fascia, flexor digitorum superficialis and profundus tendons (including their associated lumbricals) to the small and ring finger, and the neurovascular bundles to the 3rd and 4th web space. The excision plane was carried down to the level of the fascia of the interossei, along the volar shafts of the metacarpals (Figure 3).
Figure 3 The appearance of the hand is shown following wide excision of skin, palmar fascia, flexor tendons, lumbricals, and neurovascular bundles. The small and ring fingers are postured in extension due to the absence of flexor tendons. The resection specimen is shown above.
Reconstruction included the placement of silicone rods to the small and ring fingers for the first part of a 2-stage flexor tendon reconstruction (Figure 4). Sural nerve grafts were placed to the small, ring and long finger. The small and ring fingers were revascularized from the superficial palmar arch with a Y-shaped vein graft harvested from the volar forearm.
Figure 4 Sural nerve grafts, reversed Y-vein graft for digital revascularization of the small and ring fingers, and first-stage tendon reconstruction with silicone rod insertion is seen here. The construct was then covered with a contralateral free radial forearm flap.
The wound was then covered with a contralateral free radial forearm flap anastomosed to the radial artery and vena comitantes. The flap was innervated by neurorrhaphy of the lateral antebrachial cutaneous nerve in the flap to the palmar cutaneous branch of the median nerve. A small skin graft was used to cover the proximal pedicle to avoid compression.
Surgical pathology of the en bloc excision was negative for residual malignancy. Metastatic tumor was not identified within the sentinel lymph node following examination of multiple tissue levels of the node using both standard hematoxylin-eosin staining and immunohistochemical staining with antibodies against multiple cytokeratins and CD34, immunomarkers (frequently positive in epithelioid sarcoma).
At six months post-procedure, the radial forearm flap and donor site were well healed (Figure 5). The second-stage tendon reconstruction was undertaken by exchanging the silicone rods with extensor digitorum longus grafts from his 3rd and 4th toes for restoration of active finger flexion.
Figure 5 Prior to second-stage tendon grafting, the patient had a well-healed radial forearm flap on his right palm. The donor site on the left forearm was satisfactory.
At 16 months post treatment, the patient remains free of local, regional and distant disease. He has regained acceptable hand function, with small and ring finger individual joint range of motion of 90 degrees at the metacarpal-phalangeal joints, 30 degrees at the proximal interphalyngeal joints, and 30 degrees at the distal interphalangeal joints (Figure 6). His moving two-point discrimination ranges from 5 to 7 mm.
Figure 6 The final appearance of both hands after staged tendon grafting to the right ring and small fingers. There is functional restoration of composite finger flexion.
Discussion
Soft tissue sarcomas generally have a low incidence of regional lymph node metastasis (3–10%) [2] and regional lymph node recurrence (4–10%) [12,13]. Standard treatment includes wide local excision with pre-or postoperative radiotherapy. Limb salvage surgery provides acceptable local control comparable to amputation, with no difference in survival. [13-20] Multivariate analysis has demonstrated that the presence of metastasis at presentation is the single most important risk factor for local recurrence. [4] This likely reflects the more aggressive biologic potential of tumors that metastasize early and are more likely to fail local treatment.
Treatment of sarcomas of the hand is particularly challenging due to the concentrated and intricate anatomy, which makes sparing of critical structures difficult. Furthermore, the majority of these tumors are extra-compartmental, violating multiple tissue planes. Microsurgical skill for complex vascular and neural repair is an integral part of the overall planning of these cases, since without sophisticated reconstruction, limb salvage for hand sarcomas is unlikely to be useful. Free flaps are commonly required to restore function as well as to facilitate primary healing. [14-18]
The propensity of epithelioid sarcoma for regional spread supports the role of minimally invasive regional node staging procedures for prognosis and treatment. SLN biopsy has dramatically changed the management of melanoma and breast cancer. It has been investigated in the mapping of other tumors including penile [21], lung [22], colon [23,24], upper GI tumors [25], gynecologic cancer [26,27], thyroid cancer [28,29], and squamous cell carcinoma of the head and neck [7,30,31].
Given the success of the technique in other malignancies, it seems reasonable to apply SLN biopsy to soft tissue sarcomas of the extremity [8]. The role of SLN biopsy has not been extensively investigated in the treatment of sarcoma. In fact, there is only a single published report on its use in a child with rhabdomyosarcoma [9]. With the improved survival advantage of radical lymphadenectomy for clinically evident lymph node metastases from sarcoma, [4] accurate early detection of micrometastases may be important. Furthermore, patients with a negative sentinel lymph node biopsy for micrometastasis would be spared the morbidity of formal lymphadenectomy.
Identification of the sentinel node in sarcomas is more challenging than in breast and melanoma patients. Although successful identification of sentinel nodes exceeds 95% when using both a vital blue dye and a nuclear tracer [32,33], we avoided blue dye because it stain tissues and obscure planes. For resection in the hand, it is paramount to maintain precise visibility and a dye-free and bloodless field. Furthermore, when neoadjuvant radiotherapy is used, radiation-associated scarring of lymphatics could alter the accuracy of lymphatic mapping. Therefore, despite SLN biopsy, close follow-up of regional nodal basins is required. This includes assessment with clinical examination as well as with imaging such as high-resolution ultrasound. Our particular patient required serial chest computed-tomography scans for follow-up of unrelated, non-specific lung nodules. This provided a concurrent, detailed, and serial assessment of the benign appearance of his operated axillary bed.
Conclusion
We report a case of epithelioid sarcoma of the hand successfully managed with a multi-disciplinary approach including neoadjuvant radiation, sentinel node biopsy and wide surgical excision. The rarity of this tumor makes definitive conclusions difficult but SLN biopsy appears to be a useful adjunct in the treatment of these sarcomas.
Competing interests
In the past five years we have not received reimbursements, fees, funding, or salary from an organization that may in any way gain or lose financially from the publication of this manuscript, either now or in the future. No such an organization financed this manuscript (including the article-processing charge).
• We do not hold any stocks or shares in an organization that may in any way gain or lose financially from the publication of this manuscript, either now or in the future.
• We do not hold nor are applying for any patents relating to the content of the manuscript. We have not received reimbursements, fees, funding, or salary from an organization that holds or has applied for patents relating to the content of the manuscript.
• We have no other financial competing interests.
We do not have any non-financial competing interests (political, personal, religious, academic, intellectual, commercial or any other) to declare in relation to this manuscript.
Authors' contributions
CT: Primary surgeon; report conception, writing, preparation and revision of manuscript, response to reviewers' questions, submission of manuscript, photographs
RT: First assistant surgeon; literature review, chart review, data collection, writing and preparation of manuscript
AS: Second assistant surgeon; literature review, chart review, data collection, writing and preparation of manuscript
AH: Radiation oncologist; chart review, writing and preparation of manuscript
BW: Pathologist; writing, preparation and revision of manuscript, histologic slide review, photographs
Acknowledgements
Patient consent was obtained for publication of this case report.
==== Refs
Spillane AJ Thomas JM Fisher C Epithelioid sarcoma: the clinicopathological complexities of this rare soft tissue sarcoma Ann Surg Oncol 2000 7 218 225 10791853
Skinner KA Eilber FR Soft tissue sarcoma nodal metastases: biologic significance and therapeutic considerations Surg Oncol Clin N Am 1996 5 121 127 8789497
Chase DR Enzinger FM Epithelioid sarcoma: diagnosis, prognostic indicators and treatment Am J Surg Path 1985 9 241 263 4014539
Fong Y Coit DG Woodruff JM Brennan MF Lymph node metastases from soft tissue sarcoma in adults. Analysis of data from a prospective database of 1772 sarcoma patients Ann Surg 1993 217 72 77 8424704
Mazeron JJ Suit HD Lymph node as sites of metastases from sarcomas of soft tissue Cancer 1987 60 1800 1808 3308055
Weingrad DN Rosenberg SA Early lymphatic spread of osteogenic and soft-tissue sarcomas Surgery 1978 84 231 240 278229
Ross HM Lewis JJ Woodruff JM Brennan MF Epithelioid sarcoma: clinical behavior and prognostic factors of survival Ann Surg Oncol 1997 4 491 495 9309338
Blazer DG Sabel MS Sondak VK Is there a role for sentinel lymph node biopsy in the management of sarcoma? Surg Oncol 2003 12 201 206 12957624 10.1016/S0960-7404(03)00030-6
McMulkin HM Yanchar NL Fernandez CV Giacomantonio C Sentinel lymph node mapping and biopsy: a potentially valuable tool in the management of childhood extremity rhabdomyosarcoma Pediatr Surg Int 2003 19 453 456 12740706 10.1007/s00383-003-0956-y
Giuliano AE Dale PS Turner RR Morton DL Evans SW Krasne DL Improved axillary staging of breast cancer with sentinel lymphadenectomy Ann Surg 1995 222 394 399 7677468
Clary BM Brady MS Lewis JJ Coit DG Sentinel lymph node biopsy in the management of patients with primary cutaneous melanoma; review of a large single institutional experience with an emphasis on recurrence Ann Surg 2001 233 250 258 11176132 10.1097/00000658-200102000-00015
Potter DA Glenn J Kinsella T Glatstein E Lack EE Restrepo C White DE Seipp CA Wesley R Rosenberg SA Patterns of recurrence in patients with high grade soft tissue sarcomas J Clin Oncol 1985 3 353 366 3973646
Vezeridis MP Moore R Karakousis CP Metastatic patterns in soft tissue sarcomas Arch Surg 1983 118 915 918 6307217
Popov P Tukiainen E Asko-Seljavaara S Huuhtanen R Virolainen M Virkkunen P Blomqvist C Soft-tissue sarcomas of the upper extremity: surgical treatment and outcome Plast Reconstr Surg 2004 113 222 230 discussion 14707640 10.1097/01.PRS.0000095946.90511.1D
Lin PP Guzel VB Pisters PW Zagars GK Weber KL Feig BW Pollock RE Yasko AW Surgical management of soft tissue sarcomas of the hand and foot Cancer 2002 95 852 861 12209730 10.1002/cncr.10750
Lohman RF Nabawi AS Reece GP Pollock RE Evans GR Soft tissue sarcoma of the upper extremity: A 5-year experience at two institutions emphasizing the role of soft tissue flap reconstruction Cancer 2002 94 2256 2264 12001125 10.1002/cncr.10419
Doi K Kuwata N Kawakami F Hattori Y Otsuka K Ihara K Limb-sparing surgery with reinnervated free-muscle transfer following radical excision of soft-tissue sarcoma in the extremity Plast Reconstr Surg 1999 104 1679 1687 10541169
Athanasian EA Malignant bone and soft-tissue sarcomas of the hand J Am Society Surg Hand 2004 4 60 72 10.1016/j.jassh.2004.02.005
Whitworth PW Pollock RE Mansfield PF Couture J Romsdahl MM Extremity epithelioid sarcoma Arch Surg 1991 126 1485 1489 1842177
Bray PW Bell RS Bowen CV Davis A O'Sullivan B Limb salvage surgery and adjuvant radiotherapy for soft tissue sarcomas of the forearm and hand J Hand Surg [Am] 1997 22 495 503 9195461
Cabanas RM An approach to the treatment of penile carcinoma Cancer 1977 39 456 466 837331
Liptay MJ Grondin SC Fry WA Pozdol C Carson D Knop C Masters GA Perlman RM Watkin W Intraoperative sentinel lymph node mapping in non-small-cell lung cancer improves detection of micrometastases J Clin Oncol 2002 20 1984 1988 11956256 10.1200/JCO.2002.08.041
Esser S Reilly WT Riley LB Eyvazzadeh C Arcona S The role of sentinel lymph node mapping in staging of colon and rectal cancer Dis Colon Rectum 2001 44 850 854 discussion 854–856 11391147
Bendavid Y Latulippe JF Younan RJ Leclerc YE Dube S Heyen F Morin M Girard R Bastien E Ferreira J Cerino M Dube P Phase I study on sentinel lymph node mapping in colon cancer: a preliminary report J Surg Oncol 2002 79 81 84 Discussion 85 11815993 10.1002/jso.10052
Aikou T Higashi H Natsugoe S Hokita S Baba M Tako S Can sentinel node navigation surgery reduce the extent of lymph node dissection in gastric cancer? Ann Surg Oncol 2001 8 82S 85S 11599909
Holub Z Jabor A Kliment L Comparison of two procedures for sentinel lymph node detection in patients with endometrial cancer: a pilot study Eur J Gynaecol Oncol 2002 23 53 57 11876394
Levenback C Coleman RL Burke TW Bodurka-Bevers D Wolf JK Gershenson DM Intraoperative lymphatic mapping and sentinel node identification with blue dye in patients with vulvar cancer Gynecol Oncol 2002 84 449 452 11855886 10.1006/gyno.2001.6572
Fukui Y Yamakawa T Taniki T Numoto S Miki H Monden Y Sentinel lymph node biopsy in patients with papillary thyroid carcinoma Cancer 2001 84 2868 2874 11753960 10.1002/1097-0142(20011201)92:11<2868::AID-CNCR10129>3.0.CO;2-I
Gallowitsch HJ Mikosch P Kresnik E Starlinger M Lind P Lymphoscintigraphy and gamma probe-guided surgery in papillary thyroid carcinoma Clin Nuc Med 1999 24 744 746 10.1097/00003072-199910000-00002
Altinyoller H Bergeroglu U Celen O Lymphatic mapping and sentinel lymph node biopsy in squamous cell carcinoma of the lower lip Eur J Surg Oncol 2002 28 72 74 11869018 10.1053/ejso.2001.1206
Wiseman SM Loree TR Hicks WL JrRigual NR Sentinel lymph node biopsy in SCC of the head and neck: a major advance in staging the N0 neck Ear Nose Throat 2002 81 156 160
Clary BM Brady MS Lewis JJ Coit DG Sentinel lymph node biopsy in the management of patients with primary cutaneous melanoma; review of a large single institutional experience with an emphasis on recurrence Ann Surg 2001 233 250 258 11176132 10.1097/00000658-200102000-00015
Jacobs IA Chevinsky AH Swayne LC Magidson JG Britto EJ Smith TJ Gamma probe-directed lymphatic mapping and sentinel lymphadenectomy in primary melanoma: reliability of the procedure and analysis of failures after long-term follow-up J Surg Oncol 2001 77 157 164 11455551 10.1002/jso.1088
|
15987509
|
PMC1192822
|
CC BY
|
2021-01-04 16:39:04
|
no
|
World J Surg Oncol. 2005 Jun 29; 3:41
|
utf-8
|
World J Surg Oncol
| 2,005 |
10.1186/1477-7819-3-41
|
oa_comm
|
==== Front
World J Surg OncolWorld Journal of Surgical Oncology1477-7819BioMed Central London 1477-7819-3-411598750910.1186/1477-7819-3-41Case ReportSentinel node biopsy as an adjunct to limb salvage surgery for epithelioid sarcoma of the hand Seal Alex [email protected] Raymond [email protected] Bret [email protected] Alex [email protected] Claire LF [email protected] University of British Columbia, Vancouver, British Columbia, Canada2 Division of Plastic Surgery, University of Western Ontario, London, Ontario, Canada3 Department of Pathology, University of Western Ontario, London, Ontario, Canada4 Department of Radiation Oncology, London Region Cancer Centre, London, Ontario, Canada2005 29 6 2005 3 41 41 22 4 2005 29 6 2005 Copyright © 2005 Seal et al; licensee BioMed Central Ltd.2005Seal 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
Epithelioid sarcomas of the hand are rare, high-grade tumors with a propensity for regional lymphatic spread approaching 40%.
Case presentation
A 54-year-old male with an epithelioid sarcoma of the palm was treated with neoadjuvant radiation, wide excision, and two-stage reconstruction. Sentinel lymph node biopsy was used to stage the patient's axilla. Sentinel node biopsy results were negative. The patient has remained free of local, regional and distant disease for the follow-up time of 16 months.
Conclusion
The rarity of this tumor makes definitive conclusions difficult but SLN biopsy appears to be a useful adjunct in the treatment of these sarcomas.
==== Body
Background
Epithelioid sarcoma is a rare, high-grade, soft tissue sarcoma. These tumors typically present on the extremities, in males who are 20 to 30 years of age. Overall 5 and 10-year survival rates are 70% and 42% respectively [1]. Epithelioid sarcoma is among a group of sarcomas with a propensity for regional lymphatic spread, with lymph node metastasis rates reported between 17%-80% [2-7]. Due to the risk of regional spread, sentinel lymph node biopsy (SLN) may be useful in the management of this tumor.
The success of SLN biopsy is based on the principle that the primary tumor drains to one or a few lymph nodes in the regional basin. Histopathological analysis of these sentinel nodes has been shown to reflect the histology of the entire lymphatic basin [8,9]. This approach is currently the least invasive and most accurate nodal staging procedure for breast cancer and melanoma [10,11]. However, SLN biopsy has not been thoroughly investigated for sarcoma.
We report a case of SLN biopsy in conjunction with limb salvage surgery and complex soft tissue, neurovascular, and staged tendon reconstruction for the management of an epithelioid sarcoma of the hand.
Case presentation
An otherwise healthy 54-year-old right-handed laborer presented with an eight-year history of a slowly enlarging "callus" in the palm of his right hand. He underwent excision at an outside institution. The operative note suggested that the tumor had the appearance of a sebaceous cyst with a large amount of surrounding tissue reaction. The neurovascular bundles were identified and preserved to the 4th web space. An epithelioid sarcoma was identified upon histological analysis. Review of surgical pathology at our institution confirmed this diagnosis (Figure 1a and 1b). The excision was incomplete with tumor extending to several margins.
Figure 1 Epithelioid sarcoma a) Conglomerate of tumor nodules with central necrosis mimicking a necrotizing granulomatous process at low magnification (hematoxylin and eosin ×4) b) Epithelioid cells with abundant eosinophilic cytoplasm and prominent nuclear atypia, appreciated at high magnification (hematoxylin and eosin ×40), distinguishes epithelioid sarcoma from its benign mimics. Note the presence of a mitotic figure.
On examination, the patient had a transverse scar just proximal to the 4th web space, with no palpable tumor. Neurovascular exam was normal, with normal range of motion of the associated digits. There were several small, palpable ipsilateral axillary nodes.
MRI of the right hand showed an ill-defined signal change within the palmar subcutaneous fat just deep to the surgical incision and distal to the 4th and 5th metacarpal-phalangeal joints, consistent with post-surgical changes and scarring. No discrete soft tissue mass was seen to suggest gross tumor. Edema signal was seen in the distal aspect of the lumbrical muscle between the ring and small flexor tendons. The interosseous muscles appeared uninvolved. No osseous or articular abnormalities were identified.
Preoperative computerized tomography (CT) of the chest was negative. The axilla was reported as having benign appearing lymph nodes, fatty in nature with no evidence of necrosis.
The patient received preoperative radiation of 50 Gy in 25 fractions. CT simulation was used for planning the gross tumor volume (GTV) to ascertain the depth to be treated using electrons. A customized lead cutout was designed to avoid treating the full width of the hand and to avoid normal tissues ulnarly, radially and at depth. Bolus was placed over the palm to ensure adequate superficial skin and scar dose whilst ensuring the dose at depth covered the tumor and previous operative bed. A daily dose of 2 Gy was delivered to a total of 50 Gy over a five week period. Moderate erythema of the palm occurred which healed well post-treatment. There was no significant edema post-irradiation. Dysethesias were reported on the radial aspect of the small finger; however, two point discrimination remained normal.
Surgery was performed 6 weeks following completion of radiotherapy. Preoperative lymphoscintography (figure 2) identified a single "hot" ipsilateral axillary lymph node, which was successfully removed. Vital blue dye was not used, as it was felt that the blue stained tissues would worsen visibility for the wide excision.
Figure 2 Preoperative lymphoscintigraphy identifies uptake of the radiolabelled tracer in a single axillary lymph node.
Wide, en bloc excision was carried out including palmar skin, subcutaneous fat, palmar fascia, flexor digitorum superficialis and profundus tendons (including their associated lumbricals) to the small and ring finger, and the neurovascular bundles to the 3rd and 4th web space. The excision plane was carried down to the level of the fascia of the interossei, along the volar shafts of the metacarpals (Figure 3).
Figure 3 The appearance of the hand is shown following wide excision of skin, palmar fascia, flexor tendons, lumbricals, and neurovascular bundles. The small and ring fingers are postured in extension due to the absence of flexor tendons. The resection specimen is shown above.
Reconstruction included the placement of silicone rods to the small and ring fingers for the first part of a 2-stage flexor tendon reconstruction (Figure 4). Sural nerve grafts were placed to the small, ring and long finger. The small and ring fingers were revascularized from the superficial palmar arch with a Y-shaped vein graft harvested from the volar forearm.
Figure 4 Sural nerve grafts, reversed Y-vein graft for digital revascularization of the small and ring fingers, and first-stage tendon reconstruction with silicone rod insertion is seen here. The construct was then covered with a contralateral free radial forearm flap.
The wound was then covered with a contralateral free radial forearm flap anastomosed to the radial artery and vena comitantes. The flap was innervated by neurorrhaphy of the lateral antebrachial cutaneous nerve in the flap to the palmar cutaneous branch of the median nerve. A small skin graft was used to cover the proximal pedicle to avoid compression.
Surgical pathology of the en bloc excision was negative for residual malignancy. Metastatic tumor was not identified within the sentinel lymph node following examination of multiple tissue levels of the node using both standard hematoxylin-eosin staining and immunohistochemical staining with antibodies against multiple cytokeratins and CD34, immunomarkers (frequently positive in epithelioid sarcoma).
At six months post-procedure, the radial forearm flap and donor site were well healed (Figure 5). The second-stage tendon reconstruction was undertaken by exchanging the silicone rods with extensor digitorum longus grafts from his 3rd and 4th toes for restoration of active finger flexion.
Figure 5 Prior to second-stage tendon grafting, the patient had a well-healed radial forearm flap on his right palm. The donor site on the left forearm was satisfactory.
At 16 months post treatment, the patient remains free of local, regional and distant disease. He has regained acceptable hand function, with small and ring finger individual joint range of motion of 90 degrees at the metacarpal-phalangeal joints, 30 degrees at the proximal interphalyngeal joints, and 30 degrees at the distal interphalangeal joints (Figure 6). His moving two-point discrimination ranges from 5 to 7 mm.
Figure 6 The final appearance of both hands after staged tendon grafting to the right ring and small fingers. There is functional restoration of composite finger flexion.
Discussion
Soft tissue sarcomas generally have a low incidence of regional lymph node metastasis (3–10%) [2] and regional lymph node recurrence (4–10%) [12,13]. Standard treatment includes wide local excision with pre-or postoperative radiotherapy. Limb salvage surgery provides acceptable local control comparable to amputation, with no difference in survival. [13-20] Multivariate analysis has demonstrated that the presence of metastasis at presentation is the single most important risk factor for local recurrence. [4] This likely reflects the more aggressive biologic potential of tumors that metastasize early and are more likely to fail local treatment.
Treatment of sarcomas of the hand is particularly challenging due to the concentrated and intricate anatomy, which makes sparing of critical structures difficult. Furthermore, the majority of these tumors are extra-compartmental, violating multiple tissue planes. Microsurgical skill for complex vascular and neural repair is an integral part of the overall planning of these cases, since without sophisticated reconstruction, limb salvage for hand sarcomas is unlikely to be useful. Free flaps are commonly required to restore function as well as to facilitate primary healing. [14-18]
The propensity of epithelioid sarcoma for regional spread supports the role of minimally invasive regional node staging procedures for prognosis and treatment. SLN biopsy has dramatically changed the management of melanoma and breast cancer. It has been investigated in the mapping of other tumors including penile [21], lung [22], colon [23,24], upper GI tumors [25], gynecologic cancer [26,27], thyroid cancer [28,29], and squamous cell carcinoma of the head and neck [7,30,31].
Given the success of the technique in other malignancies, it seems reasonable to apply SLN biopsy to soft tissue sarcomas of the extremity [8]. The role of SLN biopsy has not been extensively investigated in the treatment of sarcoma. In fact, there is only a single published report on its use in a child with rhabdomyosarcoma [9]. With the improved survival advantage of radical lymphadenectomy for clinically evident lymph node metastases from sarcoma, [4] accurate early detection of micrometastases may be important. Furthermore, patients with a negative sentinel lymph node biopsy for micrometastasis would be spared the morbidity of formal lymphadenectomy.
Identification of the sentinel node in sarcomas is more challenging than in breast and melanoma patients. Although successful identification of sentinel nodes exceeds 95% when using both a vital blue dye and a nuclear tracer [32,33], we avoided blue dye because it stain tissues and obscure planes. For resection in the hand, it is paramount to maintain precise visibility and a dye-free and bloodless field. Furthermore, when neoadjuvant radiotherapy is used, radiation-associated scarring of lymphatics could alter the accuracy of lymphatic mapping. Therefore, despite SLN biopsy, close follow-up of regional nodal basins is required. This includes assessment with clinical examination as well as with imaging such as high-resolution ultrasound. Our particular patient required serial chest computed-tomography scans for follow-up of unrelated, non-specific lung nodules. This provided a concurrent, detailed, and serial assessment of the benign appearance of his operated axillary bed.
Conclusion
We report a case of epithelioid sarcoma of the hand successfully managed with a multi-disciplinary approach including neoadjuvant radiation, sentinel node biopsy and wide surgical excision. The rarity of this tumor makes definitive conclusions difficult but SLN biopsy appears to be a useful adjunct in the treatment of these sarcomas.
Competing interests
In the past five years we have not received reimbursements, fees, funding, or salary from an organization that may in any way gain or lose financially from the publication of this manuscript, either now or in the future. No such an organization financed this manuscript (including the article-processing charge).
• We do not hold any stocks or shares in an organization that may in any way gain or lose financially from the publication of this manuscript, either now or in the future.
• We do not hold nor are applying for any patents relating to the content of the manuscript. We have not received reimbursements, fees, funding, or salary from an organization that holds or has applied for patents relating to the content of the manuscript.
• We have no other financial competing interests.
We do not have any non-financial competing interests (political, personal, religious, academic, intellectual, commercial or any other) to declare in relation to this manuscript.
Authors' contributions
CT: Primary surgeon; report conception, writing, preparation and revision of manuscript, response to reviewers' questions, submission of manuscript, photographs
RT: First assistant surgeon; literature review, chart review, data collection, writing and preparation of manuscript
AS: Second assistant surgeon; literature review, chart review, data collection, writing and preparation of manuscript
AH: Radiation oncologist; chart review, writing and preparation of manuscript
BW: Pathologist; writing, preparation and revision of manuscript, histologic slide review, photographs
Acknowledgements
Patient consent was obtained for publication of this case report.
==== Refs
Spillane AJ Thomas JM Fisher C Epithelioid sarcoma: the clinicopathological complexities of this rare soft tissue sarcoma Ann Surg Oncol 2000 7 218 225 10791853
Skinner KA Eilber FR Soft tissue sarcoma nodal metastases: biologic significance and therapeutic considerations Surg Oncol Clin N Am 1996 5 121 127 8789497
Chase DR Enzinger FM Epithelioid sarcoma: diagnosis, prognostic indicators and treatment Am J Surg Path 1985 9 241 263 4014539
Fong Y Coit DG Woodruff JM Brennan MF Lymph node metastases from soft tissue sarcoma in adults. Analysis of data from a prospective database of 1772 sarcoma patients Ann Surg 1993 217 72 77 8424704
Mazeron JJ Suit HD Lymph node as sites of metastases from sarcomas of soft tissue Cancer 1987 60 1800 1808 3308055
Weingrad DN Rosenberg SA Early lymphatic spread of osteogenic and soft-tissue sarcomas Surgery 1978 84 231 240 278229
Ross HM Lewis JJ Woodruff JM Brennan MF Epithelioid sarcoma: clinical behavior and prognostic factors of survival Ann Surg Oncol 1997 4 491 495 9309338
Blazer DG Sabel MS Sondak VK Is there a role for sentinel lymph node biopsy in the management of sarcoma? Surg Oncol 2003 12 201 206 12957624 10.1016/S0960-7404(03)00030-6
McMulkin HM Yanchar NL Fernandez CV Giacomantonio C Sentinel lymph node mapping and biopsy: a potentially valuable tool in the management of childhood extremity rhabdomyosarcoma Pediatr Surg Int 2003 19 453 456 12740706 10.1007/s00383-003-0956-y
Giuliano AE Dale PS Turner RR Morton DL Evans SW Krasne DL Improved axillary staging of breast cancer with sentinel lymphadenectomy Ann Surg 1995 222 394 399 7677468
Clary BM Brady MS Lewis JJ Coit DG Sentinel lymph node biopsy in the management of patients with primary cutaneous melanoma; review of a large single institutional experience with an emphasis on recurrence Ann Surg 2001 233 250 258 11176132 10.1097/00000658-200102000-00015
Potter DA Glenn J Kinsella T Glatstein E Lack EE Restrepo C White DE Seipp CA Wesley R Rosenberg SA Patterns of recurrence in patients with high grade soft tissue sarcomas J Clin Oncol 1985 3 353 366 3973646
Vezeridis MP Moore R Karakousis CP Metastatic patterns in soft tissue sarcomas Arch Surg 1983 118 915 918 6307217
Popov P Tukiainen E Asko-Seljavaara S Huuhtanen R Virolainen M Virkkunen P Blomqvist C Soft-tissue sarcomas of the upper extremity: surgical treatment and outcome Plast Reconstr Surg 2004 113 222 230 discussion 14707640 10.1097/01.PRS.0000095946.90511.1D
Lin PP Guzel VB Pisters PW Zagars GK Weber KL Feig BW Pollock RE Yasko AW Surgical management of soft tissue sarcomas of the hand and foot Cancer 2002 95 852 861 12209730 10.1002/cncr.10750
Lohman RF Nabawi AS Reece GP Pollock RE Evans GR Soft tissue sarcoma of the upper extremity: A 5-year experience at two institutions emphasizing the role of soft tissue flap reconstruction Cancer 2002 94 2256 2264 12001125 10.1002/cncr.10419
Doi K Kuwata N Kawakami F Hattori Y Otsuka K Ihara K Limb-sparing surgery with reinnervated free-muscle transfer following radical excision of soft-tissue sarcoma in the extremity Plast Reconstr Surg 1999 104 1679 1687 10541169
Athanasian EA Malignant bone and soft-tissue sarcomas of the hand J Am Society Surg Hand 2004 4 60 72 10.1016/j.jassh.2004.02.005
Whitworth PW Pollock RE Mansfield PF Couture J Romsdahl MM Extremity epithelioid sarcoma Arch Surg 1991 126 1485 1489 1842177
Bray PW Bell RS Bowen CV Davis A O'Sullivan B Limb salvage surgery and adjuvant radiotherapy for soft tissue sarcomas of the forearm and hand J Hand Surg [Am] 1997 22 495 503 9195461
Cabanas RM An approach to the treatment of penile carcinoma Cancer 1977 39 456 466 837331
Liptay MJ Grondin SC Fry WA Pozdol C Carson D Knop C Masters GA Perlman RM Watkin W Intraoperative sentinel lymph node mapping in non-small-cell lung cancer improves detection of micrometastases J Clin Oncol 2002 20 1984 1988 11956256 10.1200/JCO.2002.08.041
Esser S Reilly WT Riley LB Eyvazzadeh C Arcona S The role of sentinel lymph node mapping in staging of colon and rectal cancer Dis Colon Rectum 2001 44 850 854 discussion 854–856 11391147
Bendavid Y Latulippe JF Younan RJ Leclerc YE Dube S Heyen F Morin M Girard R Bastien E Ferreira J Cerino M Dube P Phase I study on sentinel lymph node mapping in colon cancer: a preliminary report J Surg Oncol 2002 79 81 84 Discussion 85 11815993 10.1002/jso.10052
Aikou T Higashi H Natsugoe S Hokita S Baba M Tako S Can sentinel node navigation surgery reduce the extent of lymph node dissection in gastric cancer? Ann Surg Oncol 2001 8 82S 85S 11599909
Holub Z Jabor A Kliment L Comparison of two procedures for sentinel lymph node detection in patients with endometrial cancer: a pilot study Eur J Gynaecol Oncol 2002 23 53 57 11876394
Levenback C Coleman RL Burke TW Bodurka-Bevers D Wolf JK Gershenson DM Intraoperative lymphatic mapping and sentinel node identification with blue dye in patients with vulvar cancer Gynecol Oncol 2002 84 449 452 11855886 10.1006/gyno.2001.6572
Fukui Y Yamakawa T Taniki T Numoto S Miki H Monden Y Sentinel lymph node biopsy in patients with papillary thyroid carcinoma Cancer 2001 84 2868 2874 11753960 10.1002/1097-0142(20011201)92:11<2868::AID-CNCR10129>3.0.CO;2-I
Gallowitsch HJ Mikosch P Kresnik E Starlinger M Lind P Lymphoscintigraphy and gamma probe-guided surgery in papillary thyroid carcinoma Clin Nuc Med 1999 24 744 746 10.1097/00003072-199910000-00002
Altinyoller H Bergeroglu U Celen O Lymphatic mapping and sentinel lymph node biopsy in squamous cell carcinoma of the lower lip Eur J Surg Oncol 2002 28 72 74 11869018 10.1053/ejso.2001.1206
Wiseman SM Loree TR Hicks WL JrRigual NR Sentinel lymph node biopsy in SCC of the head and neck: a major advance in staging the N0 neck Ear Nose Throat 2002 81 156 160
Clary BM Brady MS Lewis JJ Coit DG Sentinel lymph node biopsy in the management of patients with primary cutaneous melanoma; review of a large single institutional experience with an emphasis on recurrence Ann Surg 2001 233 250 258 11176132 10.1097/00000658-200102000-00015
Jacobs IA Chevinsky AH Swayne LC Magidson JG Britto EJ Smith TJ Gamma probe-directed lymphatic mapping and sentinel lymphadenectomy in primary melanoma: reliability of the procedure and analysis of failures after long-term follow-up J Surg Oncol 2001 77 157 164 11455551 10.1002/jso.1088
|
16045797
|
PMC1192828
|
CC BY
|
2021-01-04 16:29:46
|
no
|
BMC Vet Res. 2005 Jul 26; 1:1
|
latin-1
|
BMC Vet Res
| 2,005 |
10.1186/1746-6148-1-1
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 1612234910.1371/journal.pbio.0030305Research ArticleAnimal BehaviorNeuroscienceNutritionDrosophilaCandidate Gustatory Interneurons Modulating Feeding Behavior in the Drosophila Brain Brain Interneurons Modulating FeedingMelcher Christoph
1
Pankratz Michael J [email protected]
1
1Institut für Genetik, Forschungszentrum Karlsruhe, Karlsruhe, GermanyBate Michael Academic EditorUniversity of CambridgeUnited Kingdom9 2005 30 8 2005 30 8 2005 3 9 e30524 9 2004 30 6 2005 Copyright: © 2005 Melcher 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.
How Fruitflies Know It's Time for Lunch
Feeding is a fundamental activity of all animals that can be regulated by internal energy status or external sensory signals. We have characterized a zinc finger transcription factor, klumpfuss
(klu), which is required for food intake in Drosophila larvae. Microarray analysis indicates that expression of the neuropeptide gene hugin
(hug) in the brain is altered in klu mutants and that hug itself is regulated by food signals. Neuroanatomical analysis demonstrates that hug-expressing neurons project axons to the pharyngeal muscles, to the central neuroendocrine organ, and to the higher brain centers, whereas hug dendrites are innervated by external gustatory receptor-expressing neurons, as well as by internal pharyngeal chemosensory organs. The use of tetanus toxin to block synaptic transmission of hug neurons results in alteration of food intake initiation, which is dependent on previous nutrient condition. Our results provide evidence that hug neurons function within a neural circuit that modulates taste-mediated feeding behavior.
Neuropeptide expressing neurons interconnecting taste input with higher brain centers modulate feeding behavior in Drosophila
==== Body
Introduction
All animals must be able to evaluate their nutrient requirement, as well as the nutrient supply offered by the environment, and translate the resulting information into appropriate behavioral responses. These can range from deciding to stop or continue feeding, or to look for alternate food sources. The nutrient signals can derive internally, reflecting the body's energy state and metabolic need, or through external sensory inputs, such as olfactory and gustatory signals. The sensory modalities further provide the basis for many types of higher brain functions, such as learning and memory. Feeding behavior, in turn, decisively influences almost all aspects of animal growth and reproduction. The role of the central nervous system (CNS) in integrating an animal's feeding behavior with sensory signals on the availability and quality of nutrients is, although undisputed, insufficiently understood [1].
Drosophila provides a genetically accessible system to study the molecular mechanisms that coordinate feeding behavior with sensory signals. This organism has an array of feeding characteristics that can be exploited for behavioral analysis, and insects in general have been used extensively as models for a wide range of behavioral and physiological studies [2,3]. In this context, the identification of genes encoding chemosensory receptors in Drosophila has provided a major impetus in understanding sensory signal transduction [4–8]. These genes have been broadly divided as encoding olfactory or gustatory receptors (ORs and GRs, respectively). Olfactory sensory neurons expressing specific ORs in the external mouth region project axons to distinct glomeruli of the antennal lobe [8–12]. Projection neurons then connect the antennal lobe to the mushroom body, where central processing of olfactory information occurs [13–15]. Gustatory sensory neurons are located not only in the external mouth region, but also internally in the pharynx; both types project to the subesophageal ganglion (SOG), a region implicated in feeding and taste response [8,16–18]. As compared with the antennal lobe, much less is known about the organization of the SOG—for example, whether it is also organized in glomerular structure. The neurons that connect the SOG to higher brain centers, in a manner analogous to the olfactory projection neurons, have also not been identified.
In both olfactory and gustatory cases, the knowledge is even sparser concerning the identity of interneurons that act between the sensory neurons and motor or neuroendocrine outputs and how they might influence feeding behavior. Studies in different insects have shown that various parts of the CNS are interconnected with the neuroendocrine organs and the enteric (stomatogastric) nervous system, which innervates the feeding apparatus [19,20]. The mouth parts have also been shown to be innervated by nerves from the SOG [21]. Nevertheless, an integrated map of the neurons comprising these circuits and their function in mediating a behavioral response has been lacking.
We have previously identified a gene, pumpless
(ppl), that is required for food intake behavior in the Drosophila larvae [22]. It encodes a subunit of the glycine cleavage system and is expressed exclusively in the fat body. Although not feeding, ppl mutant larvae do not show characteristics of starving larvae, as assayed both by molecular markers and behavioral characteristics; furthermore, feeding high levels of amino acids can phenocopy several aspects of the ppl feeding phenotype. These observations led to a model in which amino acid-dependent signals from the fat body to the brain can signal cessation of feeding. In this study, we characterize another mutant, klumpfuss
(klu), with a phenotype very similar to ppl mutants. Through microarray analysis, we identified the neuropeptide gene hugin (hug) as being deregulated in klu mutants. hug is expressed in a small assembly of about 20 neurons in the SOG. Connectivity mapping and behavioral studies suggest that hug-expressing neurons function in a neural circuitry in the brain that modulates chemosensory signal-dependent feeding behavior.
Results
Molecular Characterization of Larval Mutant Defective in Feeding Behavior
In a screen for Drosophila mutant larvae defective in feeding, we identified the P-element line P(9036). These animals fail to pump food from the pharynx into the esophagus (Figure 1A), which is not due to a morphological block in the esophagus. The failure to feed is also not due to a general illness of the animal or global locomotory defects, because they can move around with the same vigor as wild-type or heterozygote siblings. P(9036) larvae also display wandering-like behavior, in which they move away from the food (Figure 1B and 1C). During this wandering-like phase, P(9036) larvae move about with food lodged in their pharynx, further supporting the view that the feeding defect is not due to a general body movement defect. Wandering behavior is observed in wild-type larvae when they stop feeding and move away from food shortly before pupariation [23]. These feeding behavior defects have also been observed for ppl mutants [22]. ppl encodes an amino acid catabolizing enzyme that is expressed exclusively in the fat body, an organ analogous to the vertebrate liver. Thus, P(9036) and ppl mutants, as immature first instar larvae, display feeding behaviors characteristic of sated, full-grown, third instar larvae. We characterized the gene corresponding to P(9036) and found it to be klu, a zinc finger protein-encoding gene that is expressed specifically in the developing nervous system [24,25]. P(9036) fails to complement the lethality of all klu alleles tested, and trans-heterozygotes also show the characteristic feeding defect (Figure 1D).
Figure 1 Phenotypic Characterization and Expression Analysis of P(9036) Mutants
(A) P(9036)-mutant larvae show feeding and growth defects as compared to wild-type (wt) controls. In wild-type, red food fills the gut, whereas it accumulates in the pharynx of P(9036) mutants. Physically, the food can pass the mutant pharynx (see arrow).
(B) P(9036) mutants show a wandering behavior, in which they leave the food source (depicted in a schematic snapshot drawing; dots represent larvae outside the food source, which is shown in red).
(C) Quantification of the P(9036) wandering phenotype shows that from 50 individuals, 23 ± 4 mutants can be found outside the food source, in contrast to only 2 ± 1 for the wild-type control at a given time point (n = 8 time points).
(D) The known klu alleles kluXR19, kluR51C, and P(1741) fail to complement P(9036), and trans-heterozygous larvae display the feeding phenotype.
(E) Microarray analysis of P(9036) revealed several neuropeptide genes as being deregulated in P(9036) mutants compared to wild-type controls (×-fold changes in P(9036) are listed; genes with more than 2-fold change are boxed).
(F–H) Semi-quantitative in situ hybridizations to first instar larval brains verify the upregulation of the neuropeptide gene hug in P(9036) klu mutants in comparison to wild-type controls, and show an upregulation of hug in ppl feeding mutants as well. The expression of hug is in the SOG. Anterior points to the upper left corner; ventral views are shown.
(I–K) Semi-quantitative in situ hybridizations to late second instar larval brains demonstrate a downregulation of hug in wild-type under starvation (PBS) and sugar conditions (20% sucrose), in comparison to normal feeding condition (yeast). Anterior points to the upper left corner; ventral views are shown.
The Neuropeptide Gene hug Expression is Altered in klu Mutants and in Amino Acid-Deficient Conditions
To study the central control process that could underlie the feeding defect of klu mutants, we performed microarray analysis of klu mutants with a focus on neuropeptide genes. We reasoned that their expression patterns in the brain would be specific enough for analysis at single-cell resolution. Furthermore, neuropeptides have been shown to influence food intake in different organisms, including mammals [1]. RNA from klu mutant larvae and wild-type larvae were isolated and hybridized to three Affymetrix chips each and compared. Figure 1E lists Drosophila neuropeptide genes and their expression profile in klu mutants, relative to wild-type first instar larvae. We then performed in situ hybridizations on wild-type larval brains with the six highest upregulated genes. We decided to focus our efforts on hug because it had the most specific expression pattern in the larval brain. While all others showed staining in different parts of the brain or in the ventral nerve cord (VNC) (unpublished data), hug showed staining in only a cluster of about 20 cells in the SOG of the larval brain, with no staining anywhere else (Figures 1F and 2). Its expression in embryos is also highly restricted in the brain [26]. hug encodes a prepropeptide capable of generating at least two neuropeptides, Drm-PK2 and hug-γ. The former encodes a myostimulatory peptide while the latter shows homology to ecdysis-triggering hormone-1, ETH-1 [26]. Both can activate a G-protein–coupled receptor belonging to the vertebrate neuromedin U group [27]. A hug homolog is also found in Anopheles
gambiae [28].
Figure 2 Neuroanatomical Analysis of hug Expression and Neuronal Projection Patterns in Larvae
(A) A schematic drawing of the Drosophila larval CNS, showing the relative positions of the two brain hemispheres (BH) and the VNC, with the neuroendocrine ring gland (RG) located dorsoanterior to the CNS. The esophagus goes through the brain, and we have termed the hole through which the esophagus passes as the foramen (F). Positions of the SOG, the larval antennal lobes (AL), the mushroom bodies (MB), and the median neurosecretory cells (mNSC) are depicted.
(B) hug expression (shown in green) is restricted to the SOG. adipokinetic hormone (akh, shown in red) serves as a ring gland marker; the CNS is false-colored in blue.
(C) Detection of marker gene expression under hug promoter-Gal4 (hugS3) construct (shown in green) labels hug cell bodies and neuronal projections relative to the neuropil (22C10, shown in red) and nuclei of post-mitotic neurons (elav, shown in blue).
(D–F) The endogenous hug expression pattern (shown in red in [D]) is reproduced by the hugS3 expression pattern (shown in [E]). Note double-positive cell bodies in (F). akh serves as ring gland marker; CNS, ring gland, and foramen are outlined.
(G and H) Immunofluorescent (G) as well as direct (H) detection of GFP expressed under the hug promoter-Ga14 construct reveals putative hug dendrites dorsoanterior of the hug cell bodies, where spherical structures are innervated. The hug cell bodies in the depicted blow-up of SOG region are out of focus in (H).
(I) hug axons innervating the protocerebrum cross the midline, showing ipsilateral and contralateral innervation of mushroom body region.
(J–L) Expression of eGFP under hugS3 shows hug axons leaving the SOG laterally (see arrows). These can be followed to the cephalopharyngeal complex of the larvae (K and L) onto the pharyngeal muscles (axons shown in green overlayed with transmission light picture; note the mouth hook apparatus in black). Dorsal (K) and lateral views (L) are shown, anterior points to upper left corner.
All fluorescence images in this and subsequent figures are Z-stack projections, and all scales bars are 50 μm, unless otherwise noted.
To confirm the microarray data, we performed semi-quantitative in situ hybridization in wild-type and klu mutant larval brains. hug is upregulated in klu mutants (Figure 1F and 1G). We then investigated whether hug expression is also regulated in ppl larvae, which display a similar feeding defect as klu. There is also an upregulation of hug in ppl mutants (Figure 1H). We next investigated whether hug expression is regulated by different nutrient signals. We therefore placed wild-type larvae in starvation and sugar conditions (that is, both being amino acid-deficient diets) and monitored hug expression. hug was downregulated in both conditions (Figure 1I, 1J, and 1K), indicating a response to nutrient signals distinct from simple lack of energy. As hug is upregulated in klu and in ppl mutants, both of which do not feed and wander about, a higher hug level correlates with decrease of food intake and food-seeking behavior. For hug downregulation under starvation and sugar conditions, a lower hug level correlates with increased food-seeking behavior since Drosophila larvae become hyperactive and disperse when food is removed.
hug-Expressing Neurons Project to the Ring Gland, the Pharyngeal Muscles, and the Protocerebrum
As mentioned above, hug is expressed specifically in a small group of neurons in the SOG (Figure 2A and 2B). To gain insight into the physiological processes that hug-expressing neurons (referred to as hug neurons) could be involved in, we wanted to determine their connectivity pattern. Therefore, we constructed a hug promoter-Gal4 line (hugS3) in order to express different versions of green fluorescent protein (GFP) marker genes for neuroanatomical studies. This approach revealed hug neuron projection to the ring gland (Figure 2C). The ring gland, as the master neuroendocrine organ of Drosophila larvae, controls metabolism and growth. For example, median neurosecretory cells of the pars intercerebralis that express Drosophila insulin like peptides (dilps), also project to the ring gland, whereas adipokinetic hormone (akh), thought to be a glucagon homolog, is produced by the ring gland [29–31].
In addition to the ring gland, we also observed hug neuron projection to the protocerebrum, near the median neurosecretory cells and the mushroom bodies (Figure 2D–2F), which comprise the center for olfactory learning and memory [13,32]. The axons projecting to the protocerebrum also cross at the midline just above the foramen (Figure 2I). We also noticed an intriguing glomerular-like structure of what are most likely hug dendrites in the SOG, just dorsoanterior to the hug cell bodies (Figure 2G and 2H). Singh [17] has described glomerular organization in the SOG of adult Drosophila that relays gustatory information. Such a glomerular organization has not been previously recognized for the larval SOG, but it would be analogous to the glomerular organization in the antennal lobes that relay olfactory information.
Strikingly, there is also projection of hug neurons to the pharyngeal muscles (Figure 2J, 2K, and 2L), which pump food into the mouth atrium. These arise from axons that leave the brain (Figure 2J) and project anteriorly along each side of the dorsal pharyngeal muscles, and terminate near the anterior end of the pharynx (Figures 2K, 2L, and 4). There has been no previous case of identified neurons in the larval SOG that project to motor outputs. At this point, we do not know whether the pharyngeal muscles are actually innervated by these axons. Taken together, these results demonstrate that hug neurons in the larvae project to key organs regulating feeding and growth—namely, the pharynx and the ring gland—as well as to higher brain centers.
hug Dendrites Innervate GR-Expressing Sensory Organs and Chemosensory Organs of the Pharynx
The projection of hug neurons to the mushroom body region, together with the fact that hug is expressed specifically in the SOG, which relays gustatory information, suggested that hug neurons could be involved in mediating chemosensory signals. Therefore, we investigated whether hug neurons receive direct input from the chemosensory organs (Figure 3A–3D). It has been demonstrated that sensory organs in the larval head that express ORs or GRs send their axons either to the antennal lobe or the SOG [8] (Figure 3A and 3B). Recently, an enhancer trap line MJ94 was used to label putative chemosensory organs of the internal pharynx [18]. As internal pharyngeal sensory organs are good candidates for transducing gustatory signals, we wondered if these sensory organs terminate at hug dendrites. As shown in Figure 3E and 3F, they indeed terminate in the contact region of hug dendrites.
Figure 3 hug Neurons Receive Gustatory Input
(A) Schematic drawing of the head region of a Drosophila larva with external as well as internal chemosensory neurons in antennomaxillary complex and internal mouth region innervating the larval CNS (depicted in green). hug neurons in the SOG (shown in red) project to the ring gland (RG), pharyngeal muscle (PM) region, and the protocerebrum (PC). Relative positions of external chemosensory sensillae (dorsal organ [do] and terminal organ [to]), internal chemosensory sensillae (ventral pharyngeal sense organ [vps], dorsal pharyngeal sense organ [dps], dorsal pharyngeal organ [dpo], and posterior pharyngeal sense organ [pps]), and major projections to the CNS are shown.
(B) GR21D1-positive sensory neurons (shown by X-Gal staining) in the dorsal organ project axons to the CNS (see arrow).
(C and D) Optical section through median CNS (composed of ten confocal 1-μm sections) shows hug arborizations in the SOG and mushroom body region (shown in green) relative to general neuropil (red) and cortical (DNA marker Draq5, blue) landmarks. Note the labeling of corpora allata cells in the ring gland (arrow; see Materials and Methods and Figure S2). Boxed area is shown at higher magnification in (D), revealing spherically organized neuropil regions lateral to foramen (partially outlined).
(E) Expression of nSyb-GFP under MJ94 enhancer trap construct labels axon terminals of internal gustatory sensory neurons (shown in green), innervating SOG and VNC.
(F) Close-up of SOG region shows co-localization of spherically organized axon terminals of MJ94 positive gustatory neurons (green) and hug neuronal arborizations (red).
(G) Axon terminals of GR66C1-positive chemosensory neurons (shown in green) can be detected in the vicinity of hug cell bodies (shown in red). dilp3 staining (blue) serves as morphological landmark.
(H) Optical section (composed of ten confocal 1-μm sections) containing the GR66C1 axon terminals (shown in green) also include the spherical hug arborizations. hug cell bodies are out of focus in (H); a close-up of the SOG is depicted.
(I and J) Axon terminals of GR21D1 positive chemosensory neurons (shown in green) also project to the vicinity of hug cell bodies (shown in red in [I]), but the optical section comprising these terminals does not contain the hug arborizations (note the absence of spherical hug dendrites in [J] as compared to [H]).
(K–M) Comparison of axon tracts used by olfactory projection neurons, which can be labeled by enhancer trap line GH146 (shown by GFP real color in [K] and in green in [L], marked by asterisks), with those used by hug neurons (shown by YFP real color in [K] and in red in [L], marked by arrows), indicate that hug neurons are distinct from olfactory projection neurons (OPN). These differences are summarized in (M). LAN, larval antennal nerve; LMN, larval maxillary nerve; LPN, larval pharyngeal nerve.
To further test this, we checked to see if chemosensory neurons that express specific GRs also project to hug dendrites. For example, it has been shown that GR66C1-positive neurons project to the SOG, whereas GR21D1-positive neurons project to the antennal lobe [8]. To see if these sensory projections terminate at or near hug neurons, we first performed staining of GR66C1- and GR21D1-positive axon terminals with hug in situ hybridization. GR66C1 receptor neurons indeed project to the vicinity of hug-expressing cells (Figure 3G). To see if these axons may potentially make synaptic contacts with hug dendrites, we used a hug promoter–yellow fluorescent protein (YFP) line (in which YFP was placed directly under the hug promoter), in combination with GR promoters driving nSyb-GFP [8,33], thus allowing simultaneous visualization of GR axon terminals and hug dendrites. As shown in Figure 3H, GR66C1-positive neurons project to the glomerular-like SOG region contacted by hug dendrites. GR21D1-positive neurons also project near the hug cells (Figure 3I), but by contrast to GR66C1-positive neurons, do not contact hug dendrites (Figure 3J). Rather, they terminate dorsoanterior to the hug dendrites, where the antennal lobes are located [8]. Taken together, these results suggested that hug neurons may act as second-order interneurons that relay gustatory information.
To further distinguish the relationship between hug neurons and the olfactory or gustatory systems, we determined whether hug neurons share the same axon tracts to the mushroom bodies as the second-order neurons that relay olfactory sensory input. The dendrites of these olfactory projection neurons underlie the glomerular structure of the antennal lobes and vertically transduce olfactory information for processing to the mushroom bodies. These projections can be visualized by the enhancer trap line GH146 [14]. As shown in Figure 3K and 3L, the axon projections of hug neurons are distinct from olfactory projection neurons: hug neurons project to a more dorsomedial region in the protocerebrum than olfactory projection neurons, and they use different axon tracts. These results essentially rule out hug neurons being olfactory projection neurons. Projection neurons transducing gustatory signals to higher brain centers have not yet been identified. In this context, hug neurons could act as gustatory projection neurons that connect gustatory sensory neurons via SOG with the protocerebrum (Figure 3M).
Subpopulation of hug Neurons Project to Distinct Targets
We have also noticed a difference in the projection specificity among the hug neurons. A series of enhancer trap lines have been isolated that label cells projecting their axons to the ring gland [34], one of which (Okt30) is co-expressed with hug (Figure 4A and 4B). When we use our hug promoter-YFP line (to distinguish it from GFP reporter constructs) in combination with the Okt30 ring gland enhancer trap line, we find that a distinct set of hug neurons project to only the ring gland and not to the protocerebrum, the pharynx, or the ventral cord (Figure 4C).
Figure 4 Subpopulations of hug Neurons Innervate Distinct Targets
(A and B) Enhancer trap line Okt30 (shown in green) labels SOG neurons projecting to ring gland. Okt30 expression pattern colocalizes with hug expression (shown in red in [B]). There are four double positive cells in (B). dilp3 staining (blue) serves as morphological landmark.
(C) Direct detection and false colorization of GFP expressed in Okt30 positive cells (green) and YFP expressed under hug promoter (red) reveals only the axons to ring gland as double positive.
(D) TH promoter construct labels dopaminergic CNS neurons (shown in green). Four TH-positive SOG neurons also express hug (shown in red). capa staining (blue) serves as morphological landmark.
(E) Combination of projection patterns of TH positive cells (green) with hug cells (red) reveals the axons innervating the pharyngeal muscles as the only double positive ones.
(F) Schematic summary of (A–E) showing distinct subpopulations of hug neurons projecting to the ring gland only (green), pharynx only (blue), and the remaining targets (red).
(G–I) Projection pattern of hug is unaffected in klu mutants. Direct detection and false coloring of hug promoter-YFP (shown in green) in the klu background reveals the pharyngeal muscles (PM), the ring gland (RG), and the VNC as being targeted in feeding mutants. Trachea are false-colored (red) in composite figure (I).
Another subpopulation of hug neurons is revealed by using the TH-Gal4 line. This drives reporter gene expression under the promoter of tyrosine hydroxylase (TH), a key enzyme in dopamine synthesis [35]. As shown in Figure 4D, specific hug neurons express TH-Gal4 reporter gene, indicating that a subset of hug neurons might be dopaminergic. When TH-Gal4-driven lacZ is used in combination with hug promoter-YFP, we observe that TH-positive hug neurons project to only the pharynx and not to the protocerebrum, the ring gland, or the VNC (Figure 4E). These results indicate that at least three distinct subpopulations of hug neurons exist: those projecting to only the ring gland, those projecting to only the pharyngeal muscles, and those projecting to the protocerebrum and/or the VNC (Figure 4F). The distinct target specificity suggests differences in the function of the hug subpopulations. In the honeybee Apis, the subesophageal-calycal tract neurons are located in the SOG, send axons to the protocerebrum, and receive input from the sensory neurons of the proboscis; these neurons are thought to transduce gustatory information [36]. Some of the hug neurons could act similarly to these honeybee neurons. Based on the connectivity map of the hug neurons, we also wondered if the global targeting of these neurons was altered in klu mutant larvae. Therefore, we crossed the hug promoter-YFP construct into klu mutant background (Figures 4G and 4H). Although we cannot rule out subtle local differences, the basic connectivity pattern is retained in the mutants (Figure 4I).
hug Neuron Connectivity Pattern Is Similar in Larvae and Adults
To see if hug neurons might also have a function in the adults, we determined the connectivity pattern in adult animals. There are some noticeable morphological differences in the feeding apparatus and neuroendocrine organs between adults and larvae (Figure 5A). One is the presence of the crop in the adult but not in the larva. The crop is a food storage organ, and its absence in the larvae most likely reflects a difference in the feeding habits; whereas adults are intermittent feeders, the larvae feed continuously. Another is the relocation of the neuroendocrine organs. The corpora cardiaca/corpora allata (CC/CA) complex, which comprises part of the ring gland in the larvae, is located right above the proventriculus in the adults, at the junction between the gut and the crop. This is in contrast to the larvae, where it is located on top of the brain hemispheres. Monitoring hugS3 expression in adults, we observe axon projections to the protocerebrum, the CC/CA complex, and the ventral cord (Figure 5B–5H). A subpopulation of hug neurons may also be dopaminergic, as in the larvae (Figure 5I). To further characterize the projections to the protocerebrum, we used the OK107 enhancer trap Gal4 line [37] together with hug promoter-YFP. These stainings indicate that hug axons traverse along the median neurosecretory cells in the pars intercerebralis, and terminate near the mushroom bodies (Figure 5J and 5K). A similar pattern is observed in the larvae (Figure 5L and 5M). The precise targets of hug neurons projecting to the protocerebrum remain to be determined. Taken together, despite the morphological differences, the connectivity pattern of hug neurons is remarkably similar between larvae and adults.
Figure 5 Neuroanatomical Analysis of hug Interneurons in Adults
(A) Relative positions of adult CNS composed of brain hemispheres (BH) and VNC, as well as esophagus passing the brain through the foramen (F), proventriculus (P), crop (C), and gut (G) are depicted schematically. The neuroendocrine CC/CA complex is located on top of the proventriculus (arrow).
(B) Localization of hug neurons and projections in adult CNS relative to nuclear marker. Note projections to the protocerebrum (top of the head), VNC, and CC/CA (arrow).
(C) hug neuronal projections relative to general neuropil marker (22c10, shown in red).
(D–F) Close-ups of different optical sections (composed of confocal 1- to 2.5-μm sections) show spherical hug arborizations in median SOG region (green in [D]), arborizations in lateral SOG region (green in [E], arrow) and in protocerebrum (green in [F]).
(G) Transmission light image of CC/CA complex (arrow) on top of proventriculus (P).
(H) Immunofluorescent close-up of the area boxed in (G). Nuclei of elav positive CC cells (blue) and of CA (green, see arrow), hug axon terminals (green, see arrowhead).
(I) TH positive SOG neurons (green) co-express hug (red); capa staining serves as morphological landmark.
(J and K) Different optical sections (composed of confocal 1- to 2.5-mm sections) show hug (green) projections above the mushroom bodies and adjacent to the median neurosecretory cells marked by OK107 (mushroom body and median neurosecretory cell marker, red); elav marker, blue. (K) Close-up of the region in (J), showing the medial neurosecretory cells. Scale bars equal 20 μ.
(L and M) Different optical sections (composed of confocal 1-to 2.5-mm sections) showing OK107 (green) and hug (red) in larva. (M) Close-up of the median neurosecretory cell region. Scale bars equal 20 μ.
Blocking Synaptic Transmission of hug Neurons Alters Food Intake Behavior
Based on the connectivity map of hug neurons and the alteration in hug expression under different nutrient and feeding conditions, we initiated a series of experiments to explore the role of hug in regulating feeding. As hug mutants have not yet been identified, we tested the effects of overexpressing hug in the larvae. We first used hugS3 to drive hug expression but did not observe any phenotype (unpublished data). This is most likely because using an endogenous promoter does not result in high enough overexpression of hug in cells that already express physiological levels of hug. We therefore used a strong ubiquitous promoter (tubulin-gal4). There was a strong reduction in growth (Figure 6A), with no larvae surviving to pupal stage; we also observed defects in food intake, although not to the same strong degree as with klu mutants (Figure 6A). This is consistent with the view outlined earlier that high hug levels correlate with decreased food intake.
Figure 6 Overexpression of hug and Blocking hug Synaptic Transmission Causes Feeding Phenotypes
(A) Overexpression of hug neuropeptide gene under ubiquitous promoter leads to reduced feeding and growth (compare size of UAS-hug larvae and controls of same age) as well as to larval lethality. Note the individual phenocopying the klu feeding phenotype (arrow).
(B) Partial rescue of klu feeding defect by blocking hug neuronal activity. n= 5; error bars represent standard deviation; see Materials and Methods for details.
(C) Schematic of adult internal morphology with proventriculus, crop, and gut. Crop is depicted in full and empty state.
(D) Feeding behavior of adult flies monitored by the presence of red food in the midgut and crop (marked by arrows). Starting from the same feeding status (empty crop at time point 0 min), experimental flies expressing TeTxLC under hugS3 construct initiate uptake of red food immediately (notice large amount of red food in crop and gut after 5 min) when confronted with red yeast paste after feeding on standard fly food overnight (overnight normal). In contrast, control flies (hugS3 flies and TeTxLC flies crossed with wild-type) initiate food uptake after 15–35 min (notice traces of red food in midgut) when confronted with red yeast paste. The same feeding status is detected in experimental as well as in control flies after long feeding period on red yeast (overnight normal, 180 min). Overnight starvation equalizes feeding behavior when flies are confronted with red yeast paste for 15 min (notice same amount of red food in all cases at overnight starvation 15 min). Two representative samples of each time point and genotype are displayed. Two independent experiments were carried out, and for each experimental set ten individuals were taken out randomly and dissected. From 20 total flies, 16 ± 2 showed the phenotype displayed here o/n, overnight.
In order to gain further information on the function of hug neurons, we then blocked synaptic transmission in these cells using tetanus toxin light chain (TeTxLC) [38]. We first carried out the experiments in the larvae but did not see any difference (unpublished data). However, we reasoned that any potential increase in feeding response may not be readily detectable in the larvae because they feed continuously, already at a maximal rate. Therefore, we tested whether blocking synaptic transmission of hug neurons could suppress the feeding defect of klu mutants. We indeed observed a significant rescue of klu mutant feeding phenotype (Figure 6B).
We then carried out behavioral analysis on adults, since they are discontinuous feeders and thus may display an increased feeding behavior. Furthermore, one can visualize the quantity of food eaten by the size of the crop (Figure 6C). Experimental and control flies were placed in food vials containing standard fly food for several days. They were then transferred to yeast paste containing red dye. A striking result was observed. After 5 min, the experimental flies had a completely filled crop (Figure 6D, left column), whereas the control lines (Figure 6D, middle and right columns) had very little food in the crop. Even after 30 min, the control flies had very little food in their crops and only traces of red food were detectable in the midgut. By 180 min, both experimental and control flies showed the same degree of feeding. These results suggested that hug neurons are involved in regulating the initiation phase of feeding: control flies wait for a certain period before initiating feeding on the new food source, whereas decreasing hug neuronal signaling results in flies initiating their feeding immediately. When flies were transferred from yeast to colored yeast, or normal food to colored normal food, no difference was seen between experimental and control flies (unpublished data), indicating that hug neurons are not simply affecting the rate of feeding per se; we also did not observe a difference when transferring from yeast to normal food, indicating that the hug neuron-dependent behavioral effect is also not due to a simple fact of changing food sources. When flies were transferred from normal food into yeast containing 1M quinine (quinine is an aversive tastant), experimental flies again filled their crops earlier than controls (Figure S1). However, when flies were kept on yeast containing 1M quinine, and then transferred to yeast without the quinine, both experimental and control flies filled their crops with the new yeast within 5 min (Figure S1); analogously, when flies were starved before placing them on red yeast, both control and experimental flies filled their crops at about the same rate (Figure 6D, bottom row). These results suggest that the quality of previous food condition plays a role in defining hug neuronal function. Taken together, our studies support the view that hug neurons act within a neural circuitry in the brain that modulates feeding behavior based on chemosensory and nutrient signals.
Discussion
Central Relay of Gustatory Information
The identification of candidate chemosensory receptors in mammals and invertebrates has provided major insights into the molecular mechanisms underlying sensory information processing. In the Drosophila olfactory system, projections of OR-expressing sensory organs terminate at specific glomerular structures in the antennal lobe. The olfactory projection neurons then act in a second relay to convey the information to the mushroom bodies in the higher brain region. The gustatory organs, expressing specific GRs, project to a different brain region, the SOG, which has been implicated in gustatory signal transduction and feeding response in different insects. Our results indicate that the neurons that express the hug neuropeptide gene are likely candidates for acting as interneurons that transduce gustatory information. These comprise an assembly of about 20 neurons in the SOG. The close proximity of their dendrites with the axon terminals of gustatory sensory organs of the external head, and chemosensory organs of the internal pharynx, suggests a synaptic contact, but this requires functional verification. Whether the SOG is also organized into glomerular structure, like the antennal lobe, is not known. Such an organization has been suggested in adult Drosophila [17], although data on larvae have been lacking. Our results on the dendritic pattern of hug neurons also suggest a glomerular structure of the larval SOG, but this remains an open issue.
The hug neurons, in turn, send axons to at least three distinct targets: the ring gland, the pharyngeal muscles, and the protocerebrum. The projections to the ring gland and the pharyngeal muscles suggest that hug neurons coordinate sensory information with growth, metabolism, and food intake; the axon tracts to the protocerebrum suggest a role of hug neurons in transducing sensory signals for processing in the higher brain centers. These axon tracts are distinct from those of the olfactory projection neurons, projecting to a more dorsomedial region of the mushroom body, and adjacent to the median neurosecretory cells of the pars intercerebralis. Thus, hug neurons are ideally connected to undertake the role of integrating gustatory sensory signals with higher brain functions and feeding behavior.
Chemosensory Adaptation, Nutrient Status, and Food Intake Response
The chemosensory systems of all animals play critical roles in modulating feeding behavioral response. Feeding behavior can have diverse aspects, including locating a food source, evaluating food for nutritional appropriateness, choosing between different food sources, and deciding to initiate or terminate feeding. Blocking synaptic transmission by tetanus toxin in the hug neurons alters a specific aspect of the feeding behavioral response. When transferred to a certain new food source, the control flies wait for a period before initiating feeding, whereas experimental flies start feeding almost immediately. In both cases, the size of the crop after a longer feeding period does not change, meaning that no difference is seen in the termination phase of feeding. It is interesting to note that GR66C1 (also named GR66a) neurons, which project to hug dendrites, have recently been shown to mediate aversive taste response [39,40]. This is consistent with the behavior of flies in which hug signaling is decreased, since they lose their “aversive” response, as manifested in the elimination of a wait period before feeding. This behavior is dependent on internal nutrient status, as well as food quality, since if animals are starved or given food with an aversive tastant beforehand (such as yeast with quinine), control flies also start feeding immediately on the new yeast source.
Insects have evolved a wide variety of feeding behaviors based on food identity, quality, and availability. Some of these are innate, whereas others are acquired through experience. For example, food preference in the tobacco hornworm is dependent on what they initially encounter after hatching. They are capable of growing on a wide variety of sources, but once they have fed on a particular food type, they will maintain this food preference [41,42]. In this context, a possible scenario is that Drosophila associate feeding with a particular food source with which they become familiar. When they encounter a different food source, they must first re-evaluate it, perhaps for nutrient content, or adapt to it, before initiating feeding. Therefore, hug neurons appear to regulate the decision to initiate feeding based on previous food experience.
Central Integration of Feeding Behavior and Growth
In animals with a developed endocrine system, there is an intricate interdependence among feeding, growth, and neuroendocrine activity. Drosophila larvae are characterized by continuous feeding and a huge increase in organismal growth; in the adult, although no growth at organismal level takes place, a large cellular growth is required in the female for egg production. Both are highly dependent on feeding and the quality of food, such as protein content, and are under neuroendocrine control [2]. klu and ppl represent two genes that are required for food intake and growth in Drosophila. Mutations in both genes result in reduced food intake and growth. In addition, as young larvae, mutants display a wandering-like behavior, which is reminiscent of full-grown wild-type larvae, which stop feeding and move away from the food source just prior to pupariation, a process dependent on the neuroendocrine system [23]. Mutations in either of the genes lead to an upregulation of hug neuropeptide gene expression in the brain, whereas hug expression is downregulated in the absence of food signals.
What could be the function of the hug neuropeptides? hug encodes at least two distinct neuropeptides [26]. One (hug-γ) has homology to an ecdysone triggering hormone, while the second (Drm-PK-2) is a pyrokinin with myostimulatory activity. hug-γ could be involved in controlling growth and metabolism. This view is supported by projection of hug neurons to the ring gland, the master neuroendocrine organ. In addition, overexpression of hug has been shown to cause molting defects [26]. Drm-PK-2, on the other hand, may play a role in modifying the mechanical aspect of food intake, which is supported by the projection of hug neurons to the pharyngeal muscles. One interesting possibility is that the different neuropeptides are translated or trafficked to different targets in subset of hug neurons. In this case, a common gene expression pattern can be utilized to send out different signals to the different targets, such as to the higher brain center, feeding apparatus, and neuroendocrine organ. This would be a mechanism for coordinating different growth-dependent processes with a common input signal, for example, from a particular food signal. In this context, one way to explain the upregulation of the hug gene in klu and ppl mutants would be that the level of hug gene differentially correlates with the degree of food-seeking response. High levels, as in the mutants that do not feed, would reflect lower feeding and food-seeking response, whereas low levels, as in the absence of food sensory input, would reflect increased food-seeking response. This would also be consistent with hug overexpression studies and with the correlation seen between decreasing hug neuronal activity and increased feeding (Figure 7). Further functional studies, including imaging analysis [43,44], should increase our understanding of how the hug neural circuit coordinates sensory perception, feeding behavior, and growth.
Figure 7 Model of hug as Modulator of Feeding Behavior
The hug neurons, which express the neuropeptide gene hug and which interconnect gustatory sensillae via the SOG to the pharyngeal muscles, the protocerebrum, and the neuroendocrine organ, modulate chemosensory dependent feeding behavior. Increased hug signaling correlates with decreased feeding, whereas decreased hug signaling correlates with increased feeding (see Discussion section for details).
Materials and Methods
Feeding behavior assay
The larval feeding behavior assay was done as described previously [22]. Flies were allowed to lay eggs on apple juice agar plates containing colored yeast paste (150 mg Carmen Red, Sigma-Aldrich [St. Louis, Missouri, United States] per 100 g yeast paste). Given numbers of larvae from overnight egg collections were allowed to develop for 24 h at 25 °C and subsequently (2-h intervals) monitored for feeding and wandering phenotypes under a dissection microscope. For starvation experiments, wild-type larvae of late second instar were placed in petri dishes, containing filter paper that was soaked with either PBS (for complete starvation) or PBS containing 20% sucrose. For normal feeding conditions, fresh yeast paste was given. All feeding experiments were done at room temperature for 6 h. Overexpression studies were done using UAS-hug and tub-Gal4/TM3 fly lines [26], and heterozygote siblings were used as controls.
For larval rescue experiments, lines used were P(9036)/TM3 (parental line 1), UAS-TeTxLC; P(9036)/TM3 (parental line 2), and hugS3, P(9036)/TM3 (parental line 3). The genotypes assayed were +/UAS-TeTxLC; and P(9036)/P(9036) for control 1 (C1) + hugS3, P(9036)/P(9036) for control 2 (C2) and +/UAS-TeTxLC/+; hugS3, P(9036) for experimental (see Figure 6B). Feeding phenotypes were counted from 100 eggs per collection. Five independent collections per genotype were carried out.
For the adult feeding assay, experimental flies (hug promoter construct driving UAS-TeTxLC expression) and control flies (hugS3-Gal4 flies and UAS-TeTxLC flies crossed with wild-type) derived from 0- to 4-h egg collections were allowed to develop on indicated food for several days. After overnight feeding (for example, on standard fly food or PBS only), flies were allowed to feed on apple juice agar plates containing red-colored food being assayed. At given time points, randomly chosen individuals per genotype were removed and dissected under a dissection microscope. Preparations were fixed and mounted in Mowiol (see below) for microscopic analysis.
Molecular and microarray analyses
The hug construct was made by cloning a 1.5-kb PCR fragment containing the hug regulatory region (amplified from genomic DNA using 5′–
CTTCAGGGCCTTGGCTG and 5′–
GGGACAACTGATGCCACG as primers) into a pCaSpeR-AUG-Gal4 vector [10]. The direct hug promoter YFP construct was made by replacing AUG-Gal4 of the hug-pCaSpeR construct, with a YFP fragment derived from YFP-pCS2+ vector (Clontech, Palo Alto, California, United States). Transgenic flies were obtained following standard injection protocols.
Microarray experiments were, in principle, done as described previously [45] using Affymetrix (Santa Clara, California, United States) GeneChips representing some 13,500 genes. Egg collections (0–4 h) of the P(9036)/TM3-GFP line were allowed to develop for an additional 22 h at 25 °C. Homozygous P(9036) larvae were hand-picked under a fluorescence microscope, and total RNA was isolated using the NucleoSpin RNA II Kit (Macherey-Nagel, Düren, Germany). GeneChip hybridization and data analysis was done as described [45].
Histochemistry and fluorescence microscopy
Histochemical in situ hybridizations were done following standard protocols, with the slight modification of replacing the proteinase K digest with an overnight incubation of the dissected and fixed larval brains in methanol at −20 °C prior to hybridization. Samples were mounted either in Canada balsam or in Mowiol (12 ml glycerol, 4.8 g Mowiol 40–88, 12 ml H2O, and 24 ml 200 mM Tris [pH 8.5]), and images were taken using a Zeiss (Oberkochen, Germany) LSM 510 META in transmission mode. Fluorescence in situ hybridizations were done using the Tyramide Signal Amplification Kit (PerkinElmer, Wellesley, California, United States) and following the manufacturer's instructions. Overnight incubation with digoxygenin- and/or fluorescein-labeled riboprobes was followed by post-hybridization for additional 2 h. Detection of the first riboprobe with peroxidase (POD)-coupled antibody was performed by overnight incubation at 4 °C, followed by the first staining reaction using fluorescein-tyramide at 1:150 dilution and allowing the reaction to run for 10 min at room temperature. After inactivation of POD by incubation with 10 mM HCl in Drosophila Ringer's solution for 10 min, the second riboprobe was detected by overnight incubation with POD-coupled antibody at 4 °C. The second staining reaction was performed by applying Cy3-tyramide at 1:150 dilution for 10 min at room temperature. In cases of dual marker protein detection, primary antibodies (α–ßGal, Cappel, or α–GFP [Abcam, Cambridge, United Kingdom], used at 1:1,000) were applied together with first POD-antibody and secondary fluorescent antibody (Cy5-coupled α-rabbit, diluted at 1:200 [Jackson Immunoresearch, West Grove, Pennsylvania, United States]) was applied together with second POD-antibody. Samples were mounted in Mowiol and evaluated using a Zeiss LSM 510 META in confocal multitracking mode, generating optical 1- to 1.5-μm sections (using a Zeiss 40×/1.2W C-Apochromat lens) or 2.5-μm sections (using a Zeiss 25×/0.8Imm Plan-Neofluar lens). For direct detection and unmixing of GFP/YFP fluorescence, larval brains of appropriate genotype were dissected in chilled Drosophila Ringer's solution on ice, and mounted without fixation in PBS, using coverslips as spacers and nail polish as sealant. Immediate analysis was performed by emission fingerprinting, using a Zeiss LSM 510 META in confocal lambda mode. Other antibodies used for immunofluorescence were 22C10 diluted 1:100 (Developmental Studies Hybridoma Bank, Iowa City, Iowa, United States) and α–elav, diluted 1:300 (Developmental Studies Hybridoma Bank), as well as Alexa488-coupled α-rat and α-mouse antibodies, each diluted 1:200 (Molecular Probes, Eugene, Oregon, United States) and Cy3-coupled α-rat and α-rabbit antibodies diluted 1:200 (Jackson Immunoresearch, West Grove, Pennsylvania, United States). Nuclear counterstaining was performed using Draq5 (Biostatus Ltd., Leicestershire, United Kingdom), diluted 1:1,000 together with secondary antibodies. The GFP antibody co-labeling the corpora allata nuclei (Torrey Pines Biolabs, Houston, Texas, United States) was used at 1:1,000 dilution. The 3D reconstruction of optical sections and figure post-processing were done using Volocity 2.6 (Improvision, Lexington, Massachusetts, United States) and Photoshop 7.0 (Adobe Systems, San Jose, California, United States) on a Mac G4 computer (Apple Computer, Sunnyvale, California, United States).
Supporting Information
Figure S1 Qualitative Graphical Representation of Feeding Analysis
Flies were scored (none, traces, or full) based on amount of red food color in the gut and crop. Overnight treatment was for 12 h. Arrows represent transfer to fresh food, either of the same type or different. For each feeding regimen, two independent experiments were carried out, and for each experimental set and time point, ten individuals were taken out randomly and dissected. The top two graphs are graphical representations of the results depicted in Figure 6D. For experiments with quinine yeast as test food, 12 ± 2 individuals showed displayed phenotypes.
(3.1 MB TIF).
Click here for additional data file.
Figure S2 Fortuitous Staining of Corpora Allatum
Fortuitous staining of corpora allatum nuclei (shown in green, arrows) by α-GFP antibody from Torrey Pines Biolabs in wild-type larvae (A) and adults (B and C) relative to the neuropile (22C10, shown in red) and the cortex (DNA marker Draq5, shown in blue). We do not know the reason for this, as α-GFP antibodies from other sources do not show this cross-reactivity.
(9.7 MB TIF).
Click here for additional data file.
We thank G. Wahlström, R. F. Stocker, K. Scott, D. Schmucker, L. B. Vosshall, B. Gerber, S. Noselli, T. Klein, C. O'Kane, and T. Siegmund for fly lines and vectors, and T. Kastilan for help with the transgenics. This work was supported by Forschungszentrum Karlsruhe and by grants from DFG (Deutsche Forschungsgemeinschaft) to MJP.
Competing interests: The authors have declared that no competing interests exist.
Author contributions: CM and MJP conceived, designed, and performed the experiments, analyzed the data, contributed reagents/materials/analysis tools, and wrote the paper.
Citation: Melcher C, Pankratz MJ (2005) Candidate gustatory interneurons modulating feeding behavior in the Drosophila brain. PLoS Biol 3(9): e305.
Abbreviations
CC/CAcorpora cardiaca/corpora allata
CNScentral nervous system
GFPgreen fluorescent protein
GRgustatory receptor
hug
hugin
klu
klumpfuss
ORolfactory receptor
PODperoxidase
ppl
pumpless
SOGsubesophageal ganglion
THtyrosine hydroxylase
VNCventral nerve cord
YFPyellow fluorescent protein
==== Refs
References
Schwartz MW Woods SC Porte D Seeley RJ Baskin DG Central nervous system control of food intake Nature 2000 404 661 671 10766253
Pflugfelder O Entwicklungsphysiologie der Insekten 1958 Leipzig (Germany) Akademische Verlagsgesellschaft Geest & Portig 490
Dethier VG The hungry fly 1976 Cambridge (Harvard University Press) 489
Clyne PJ Warr CG Freeman MR Lessing D Kim J A novel family of divergent seven-transmembrane proteins: Candidate odorant receptors in Drosophila
Neuron 1999 22 327 338 10069338
Gao Q Chess A Identification of candidate Drosophila olfactory receptors from genomic DNA sequence Genomics 1999 60 31 39 10458908
Vosshall LB Amrein H Morozov PS Rzhetsky A Axel R A spatial map of olfactory receptor expression in the Drosophila antenna Cell 1999 96 725 736 10089887
Clyne PJ Warr CG Carlson JR Candidate taste receptors in Drosophila
Science 2000 287 1830 1834 10710312
Scott K Brady R Cravchik A Morozov P Rzhetsky A A chemosensory gene family encoding candidate gustatory and olfactory receptors in Drosophila
Cell 2001 104 661 673 11257221
Gao Q Yuan B Chess A Convergent projections of Drosophila olfactory neurons to specific glomeruli in the antennal lobe Nat Neurosci 2000 3 780 785 10903570
Vosshall LB Wong AM Axel R An olfactory sensory map in the fly brain Cell 2000 102 147 159 10943836
Marin EC Jefferis GS Komiyama T Zhu H Luo L Representation of the glomerular olfactory map in the Drosophila brain Cell 2002 109 243 255 12007410
Wong AM Wang JW Axel R Spatial representation of the glomerular map in the Drosophila protocerebrum Cell 2002 109 229 241 12007409
Heisenberg M Borst A Wagner S Byers D
Drosophila mushroom body mutants are deficient in olfactory learning J Neurogenet 1985 2 1 30 4020527
Stocker RF Heimbeck G Gendre N de Belle JS Neuroblast ablation in Drosophila P[GAL4] lines reveals origins of olfactory interneurons J Neurobiol 1997 32 443 456 9110257
Stocker RF The organization of the chemosensory system in Drosophila melanogaster: A review Cell Tissue Res 1994 275 3 26 8118845
Stocker RF Schorderet M Cobalt filling of sensory projections from internal and external mouthparts in Drosophila
Cell Tissue Res 1981 216 513 523 6786751
Singh NR Neurobiology of the gustatory systems in Drosophila and some terrestrial insects Microsc Res Tech 1997 275 3 26
Gendre N Luer K Friche S Grillenzoni N Ramaekers A Integration of complex larval chemosensory organs into the adult nervous system of Drosophila
Development 2004 131 83 92 14645122
Penzlin H Kerkut GA Gilbert LI Stomatogastric nervous system Comprehensive insect physiology 1985 Oxford Pergamon Press 371 406
Hartenstein V Tepass U Gruszynski-Defeo E Embryonic development of the stomatogastric nervous system in Drosophila
J Comp Neurol 1994 350 367 381 7884047
Aubele E Klemm N Origin, destination and mapping of tritocerebral neurons of locust Cell Tissue Res 1977 178 199 219 66098
Zinke I Kirchner C Chao LC Tetzlaff MT Pankratz MJ Suppression of food intake and growth by amino acids in Drosophila : The role of pumpless, a fat body expressed gene with homology to vertebrate glycine cleavage system Development 1999 126 5275 5284 10556053
Riddiford L Bate M Arias AM Hormones and Drosophila development The development of Drosophila 1993 Cold Spring Harbor (New York) Cold Spring Harbor Laboratory Press 899 939
Klein T Campos-Ortega JA Klumpfuss, a Drosophila gene encoding a member of the EGR family of transcription factors, is involved in bristle and leg development Development 1997 124 3123 3134 9272953
Yang X Bahri S Klein T Chia W Klumpfuss, a putative Drosophila zinc finger transcription factor, acts to differentiate between the identities of two secondary precursor cells within one neuroblast lineage Genes Dev 1997 11 1396 1408 9192868
Meng X Wahlstrom G Immonen T Kolmer M Tirronen M The Drosophila hugin gene codes for myostimulatory and ecdysis-modifying neuropeptides Mech Dev 2002 117 5 13 12204246
Park Y Kim YJ Adams ME Identification of G protein-coupled receptors for Drosophila PRXamide peptides, CCAP, corazonin, and AKH supports a theory of ligand-receptor co-evolution Proc Natl Acad Sci U S A 2002 99 11423 11428 12177421
Riehle MA Garczynski SF Crim JW Hill CA Brown MR Neuropeptides and peptide hormones in Anopheles gambiae
Science 2002 298 172 175 12364794
Brogiolo W Stocker H Ikeya T Rintelen F Fernandez R An evolutionarily conserved function of the Drosophila insulin receptor and insulin-like peptides in growth control Curr Biol 2001 11 213 221 11250149
Ikeya T Galic M Belawat P Nairz K Hafen E Nutrient-dependent expression of insulin-like peptides from neuroendocrine cells in the CNS contributes to growth regulation in Drosophila
Curr Biol 2002 12 1293 1300 12176357
Rulifson EJ Kim SK Nusse R Ablation of insulin-producing neurons in flies: Growth and diabetic phenotypes Science 2002 296 1118 1120 12004130
Zars T Fischer M Schulz R Heisenberg M Localization of a short-term memory in Drosophila
Science 2000 288 672 675 10784450
Estes PS Ho GL Narayanan R Ramaswami M Synaptic localization and restricted diffusion of a Drosophila neuronal synaptobrevin–green fluorescent protein chimera in vivo J Neurogenet 2000 13 233 255 10858822
Siegmund T Korge G Innervation of the ring gland of Drosophila
melanogaster
J Comp Neurol 2001 431 481 491 11223816
Friggi-Grelin F Coulom H Meller M Gomez D Hirsh J Targeted gene expression in Drosophila dopaminergic cells using regulatory sequences from tyrosine hydroxylase J Neurobiol 2003 54 618 627 12555273
Schroter U Menzel R A new ascending sensory tract to the calyces of the honeybee mushroom body, the subesophageal-calycal tract J Comp Neurol 2003 465 168 178 12949779
Connolly JB Roberts IJ Armstrong JD Kaiser K Forte M Associative learning disrupted by impaired Gs signaling in Drosophila mushroom bodies Science 1996 274 2104 2107 8953046
Sweeney ST Broadie K Keane J Niemann H O'Kane CJ Targeted expression of tetanus toxin light chain in Drosophila specifically eliminates synaptic transmission and causes behavioral defects Neuron 1995 14 341 351 7857643
Thorne N Chromey C Bray S Amrein H Taste perception and coding in Drosophila
Curr Biol 2004 14 1065 1079 15202999
Wang Z Singhvi A Kong P Scott K Taste representations in the Drosophila brain Cell 2004 117 981 991 15210117
del Campo ML Miles CI Schroeder FC Mueller C Booker R Host recognition by the tobacco hornworm is mediated by a host plant compound Nature 2001 411 186 189 11346793
del Campo ML Miles CI Chemosensory tuning to a host recognition cue in the facultative specialist larvae of the moth Manduca sexta
J Exp Biol 2003 206 3979 3990 14555738
Fiala A Spall T Diegelmann S Eisermann B Sachse S Genetically expressed chameleon in Drosophila
melanogaster is used to visualize olfactory information in projection neurons Curr Biol 2002 12 1877 1884 12419190
Wang JW Wong AM Flores J Vosshall LB Axel R Two-photon calcium imaging reveals an odor-evoked map of activity in the fly brain Cell 2003 112 271 282 12553914
Zinke I Schutz CS Katzenberger JD Bauer M Pankratz MJ Nutrient control of gene expression in Drosophila : Microarray analysis of starvation and sugar-dependent response Embo J 2002 21 6162 6173 12426388
|
16122349
|
PMC1193519
|
CC BY
|
2021-01-05 08:21:26
|
no
|
PLoS Biol. 2005 Sep 30; 3(9):e305
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030305
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 1612235010.1371/journal.pbio.0030315Research ArticleGenetics/Genomics/Gene TherapyDiabetes/Endocrinology/MetabolismHomo (Human)Lack of Support for the Association between GAD2 Polymorphisms and Severe Human Obesity GAD2 Polymorphisms and ObesitySwarbrick Michael M
1
Waldenmaier Björn
2
Pennacchio Len A
3
Lind Denise L
4
Cavazos Martha M
1
Geller Frank
5
Merriman Raphael
6
Ustaszewska Anna
3
Malloy Mary
4
Scherag André
5
Hsueh Wen-Chi
1
Rief Winfried
7
Mauvais-Jarvis Franck
8
Pullinger Clive R
4
Kane John P
4
Dent Robert
9
McPherson Ruth
10
Kwok Pui-Yan
4
Hinney Anke
2
Hebebrand Johannes
2
Vaisse Christian [email protected]
1
1Diabetes Center, University of California, San Francisco, California, United States of America,2Department of Child and Adolescent Psychiatry, University of Duisburg-Essen, Essen, Germany,3Department of Energy Joint Genome Institute, Walnut Creek, California, United States of America,4Cardiovascular Research Institute, University of California, San Francisco, California, United States of America,5Institute of Medical Biometry and Epidemiology, Phillips-University of Marburg, Marburg, Germany,6Department of Medicine, University of California, San Francisco, California, United States of America,7Department of Psychology, University of Marburg, Marburg, Germany,8Division of Diabetes, Endocrinology and Metabolism, Baylor College of Medicine, Houston, Texas, United States of America,9Ottawa Health Research Institute, Ottawa, Ontario, Canada,10University of Ottawa Heart Institute, Ottawa, Ontario, CanadaCardon Lon Academic EditorUniversity of OxfordUnited Kingdom9 2005 30 8 2005 30 8 2005 3 9 e3151 12 2004 11 7 2005 Copyright: © 2005 Swarbrick et al.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
A Genetic Link to Obesity: The Numbers Don't Add Up for GAD2
The demonstration of association between common genetic variants and chronic human diseases such as obesity could have profound implications for the prediction, prevention, and treatment of these conditions. Unequivocal proof of such an association, however, requires independent replication of initial positive findings. Recently, three (−243 A>G, +61450 C>A, and +83897 T>A) single nucleotide polymorphisms (SNPs) within glutamate decarboxylase 2 (GAD2) were found to be associated with class III obesity (body mass index > 40 kg/m2). The association was observed among 188 families (612 individuals) segregating the condition, and a case-control study of 575 cases and 646 lean controls. Functional data supporting a pathophysiological role for one of the SNPs (−243 A>G) were also presented. The gene GAD2 encodes the 65-kDa subunit of glutamic acid decarboxylase—GAD65. In the present study, we attempted to replicate this association in larger groups of individuals, and to extend the functional studies of the −243 A>G SNP. Among 2,359 individuals comprising 693 German nuclear families with severe, early-onset obesity, we found no evidence for a relationship between the three GAD2 SNPs and obesity, whether SNPs were studied individually or as haplotypes. In two independent case-control studies (a total of 680 class III obesity cases and 1,186 lean controls), there was no significant relationship between the −243 A>G SNP and obesity (OR = 0.99, 95% CI 0.83–1.18, p = 0.89) in the pooled sample. These negative findings were recapitulated in a meta-analysis, incorporating all published data for the association between the −243G allele and class III obesity, which yielded an OR of 1.11 (95% CI 0.90–1.36, p = 0.28) in a total sample of 1,252 class III obese cases and 1,800 lean controls. Moreover, analysis of common haplotypes encompassing the GAD2 locus revealed no association with severe obesity in families with the condition. We also obtained functional data for the −243 A>G SNP that does not support a pathophysiological role for this variant in obesity. Potential confounding variables in association studies involving common variants and complex diseases (low power to detect modest genetic effects, overinterpretation of marginal data, population stratification, and biological plausibility) are also discussed in the context of GAD2 and severe obesity.
A large genetic study involving multiple populations is not able to replicate previous findings linking variation in the GAD2 gene to susceptibility to obesity.
==== Body
Introduction
By dramatically increasing mortality [1] and morbidity [2] from cardiovascular disease, obesity has emerged as a major public health issue for the 21st century. Obesity is strongly associated with type 2 diabetes, hypertension, dyslipidemia, heart failure, and stroke [3]. This burden of disease is particularly high in individuals with class III obesity (body mass index [BMI] > 40 kg/m2), as they are more likely to develop at least one of these co-morbidities [4].
The importance of genetic factors in determining susceptibility to obesity has been well established elsewhere, by studies of twins [5], and adoptees [6]. At present, there is support for a model in which the propensity to become obese is determined largely by genetic factors, with environmental factors determining the expression of the condition [7]. These genetic influences are likely to be particularly powerful in individuals with severe or early-onset forms of obesity [8]. While several rare monogenic forms of non-syndromic obesity have been described to date [9–13], efforts aimed at identifying common susceptibility alleles for the condition have been much less successful [14].
The Chromosome 10p12 region has previously demonstrated significant linkage with severe human obesity [15]. In the initial study [15] involving individuals ascertained by a proband with class III obesity (BMI > 40 kg/m2) and at least one sibling with BMI > 27 kg/m2, strong evidence for linkage (maximum logarithm of odds score 4.85) was obtained at the marker D10S197. The linkage peak encompassed a region of approximately 15 centimorgans. Confirmation of this linkage, albeit at lower levels of significance, was obtained in German Caucasians [16] and a combined sample of Caucasian Americans and African Americans [17]. The marker D10S197 is located within intron 7 of the glutamate decarboxylase 2 (GAD2) gene, which encodes the 65-kDa subunit of glutamic acid decarboxylase—GAD65.
Recently, Boutin et al. [18] obtained evidence to implicate GAD2 as a candidate gene for human obesity. In a case-control study for class III obesity, the authors identified both a haplotype (consisting of the most frequent alleles of single nucleotide polymorphisms +61450 C>A and +83897 T>A), and a SNP (−243 A>G) within GAD2 that differed in frequency between cases and controls. In family-based tests of association involving 612 individuals from 188 nuclear families, the +61450 C>A and +83897 T>A SNPs were associated with class III obesity. The “protective” wild-type (WT) haplotype (+61450 C and +83897 T) identified in the case-control study was found to be in excess in unaffected offspring.
As the GAD2 variant allele −243 G was in the 5′ region of the gene (the other two SNPs were located in intronic regions), displayed the strongest association with class III obesity in the case-control study, and was in linkage disequilibrium with the +61450 C>A and +83897 T>A SNPs, functional studies were performed to test its effects on transcription and nuclear protein binding. In a luciferase reporter gene containing the GAD2 promoter, the −243 G allele increased the transcriptional activity 6-fold relative to an equivalent reporter gene containing the WT (−243 A) allele in βTC3 murine insulinoma cells. Also, relative to the WT (A) allele, oligonucleotide probes containing the variant (G) allele had a higher affinity for an unidentified nuclear protein from βTC3 cells. Overall, their results suggested that the GAD2 −243 G allele might not only constitute a genetic marker for class III human obesity, but may also exert a significant physiological effect.
In recent years, many unreplicated associations have been reported between common genetic polymorphisms and measures of adiposity [19,20]. Indeed, one of the significant challenges in genetic association studies is the presence of statistical trends towards susceptibility (with the suspected allele being neither sufficient nor necessary for disease expression), rather than clear cause-and-effect relationships. This factor reduces the power of individual studies; consequently, it has become critical to develop larger multi-center studies to confirm positive associations in other populations, and to perform meta-analyses to more accurately estimate the magnitude of the genetic effect [21]. In the present study, we attempted to replicate the recent findings of Boutin et al. [18] by performing family-based tests of association and case-control studies in three Caucasian populations.
Results
Family-Based Tests of Association
In the previous report [18], an excess of WT alleles was observed in unaffected offspring for the +61450 C>A and +83897 T>A SNPs (p = 0.03 for each), and the haplotype consisting of the WT alleles at these SNPs was found to be in excess in unaffected offspring (p = 0.05). However, excess transmission of the G allele of the −243 A>G SNP to affected offspring was not observed (p = 0.06).
To further assess these initial findings in a much larger cohort, we performed family-based tests of association in 2,359 German Caucasian individuals from 693 nuclear families. Nuclear families were composed of obese children (mean BMI percentile = 98.6 ± 2.3, range 90th–100th percentile), their obese siblings, and both of their parents. The clinical characteristics of the nuclear families segregating obesity are shown in Table 1. This group of individuals included the 89 families that had previously displayed suggestive linkage for obesity (maximum likelihood binomial logarithm of odds score of 2.24) at D10S197 [22].
Table 1 Clinical Characteristics of Participants Used for Family-Based Association and Case-Control Studies
Using the pedigree disequilibrium test (PDT) [23], we found no evidence for excess transmission of any GAD2 alleles to obese children in the 89 families displaying prior linkage of obesity to Chromosome 10p (Table 2). This finding alone suggested that the original linkage signal in this region might be due to different SNPs than the ones under study. We next included samples that were not previously tested for linkage, bringing the total group to 693 nuclear families. As in the linked families, we found no association between any of the GAD2 SNPs and obesity (Table 2). Similarly, studies of haplotype transmission in the entire group using the transmission disequilibrium test (TDT) did not provide any evidence for a “protective” haplotype consisting of the +61450 C and +83897 T alleles, or any other two- or three-allele GAD2 haplotypes (Table 3).
Table 2 PDT Results for GAD2 SNPs in 693 German Families Segregating Severe, Early-Onset Obesity
Table 3 Analysis of Haplotype Transmission in German Obesity Trios Using the TDT
Case-Control Studies
Previously [18], the −243 A>G, +61450 C>A, and +83897 T>A SNPs were associated with class III obesity in one case-control group (349 obese cases and 376 nonobese controls), whereas the association between the −243 A>G SNP and obesity was significant in the pooled sample of 575 obese cases and 646 controls (odds ratio [OR] of 1.3, 95% confidence interval [CI] 1.053–1.585, p = 0.014). We attempted to replicate these associations by performing case-control studies of class III obesity in two groups of North American Caucasians, one from the United States and the other from Canada. The clinical characteristics of the participants used in the case-control studies are shown in Table 1.
US case-control study
Each of the three GAD2 polymorphisms (−243 A>G, +61450 C>A, and +83897 T>A) was found to be in Hardy-Weinberg equilibrium in 302 class III obese (BMI > 40 kg/m2) cases and 427 lean controls (Table 4). None of the three variant alleles were associated with class III obesity (−243 G allele and class III obesity [OR = 1.11, 95% CI 0.84–1.46, p = 0.45]; the +61450 A allele [OR = 1.25, 95% CI 0.99–1.57, p = 0.058]; the +83897 A allele [OR = 1.14, 95% CI 0.87–1.50, p = 0.33]). Moreover, within the obese and lean groups, there was no association between GAD2 genotype and BMI for any of the three SNPs studied (unpublished data).
Table 4 Genotype Results for US Case-Control Study
Canadian case-control study
The Canadian participants were also genotyped for the GAD2 −243 A>G, +61450 C>A, and +83897 T>A SNPs. Genotypes of both cases and controls conformed to Hardy-Weinberg equilibrium, and the allele frequencies were similar to those observed in US Caucasians (Table 5).
Table 5 Genotype Results for Canadian Case-Control Study
As for US participants, the frequency of the −243 G allele did not differ between a group of 378 class III obese Canadian participants (frequency = 0.160) and a group of 759 lean controls (frequency = 0.175). The −243 G allele was not associated with class III obesity in this case-control study (OR = 0.90, 95% CI 0.71–1.14, p = 0.39). Pooling the results from the US and Canadian studies (680 class III obese cases and 1,186 lean controls) did not provide significant evidence for an association between the −243 G allele and class III obesity (OR = 0.99, 95% CI 0.83–1.18, p = 0.89).
Meta-analysis for the −243 A>G variant
It has been proposed elsewhere [24] that the interpretation of results from association studies may be aided by meta-analysis of all similar studies. We compiled all available genotyping data (ours and that of the previous GAD2 study [18]) pertaining to the relationship between the −243 A>G polymorphism and class III obesity, and performed a meta-analysis (Figure 1). Inclusion of the data from the original study and our two case-control studies (a total of 1,252 cases and 1,800 controls) yielded an OR of 1.11 (95% CI 0.90–1.36) for the association between the −243 G allele and class III obesity.
Figure 1 Meta-Analysis for the Association between the GAD2 −243 A>G Polymorphism and Class III Obesity
French groups 1 and 2 refer to the genotype results from two sets of Caucasian class III obese cases and controls studied by Boutin et al. [18]. US and Canadian groups were from the present study. The meta-analysis for the association between the −243 G allele and class III obesity yielded a summary OR of 1.11 (95% CI 0.90–1.36), obtained in a total sample of 1,252 class III obese cases and 1,800 controls using a Mantel-Haenszel method and a fixed effects model.
Further Investigation of GAD2 as a Candidate Gene for Severe Obesity
In order to evaluate the potential relationship between other common SNPs in GAD2 and severe obesity, we conducted a comprehensive investigation of haplotype structure in this region using the data from the International HapMap Project (http://www.hapmap.org/cgi-perl/gbrowse/gbrowse/hapmap) and Haploview (http://www.broad.mit.edu/mpg/haploview/index.php) [25]. As described earlier by Boutin et al., GAD2 (and 2 kilobases of its promoter) lies within a 90-kilobase block of linkage disequilibrium on Chromosome 10p12 (Figure S1). Two of the three SNPs used in this study, −243 A>G (rs2236418) and +83897 T>A (rs928197), were used in the construction of HapMap. To incorporate the data from the third SNP, +61450 C>A (rs992990), into this framework, we genotyped the same CEPH (Centre D'étude du Polymorphisme Humain) samples (Utah residents with ancestry from northern and western Europe) that were used in the creation of the map. When the genotype results from the +61450 C>A SNP were integrated with those from HapMap, the overall structure of the haplotype block did not change. To capture at least 95% of the haplotype diversity within this haplotype block, we then determined that genotyping of three more SNPs was required: rs3781117 (intron 4), rs3781118 (intron 4), and rs1330581 (intron 7) (Figure S2).
The German families and case-control participants from the US and Canada were genotyped for the SNPs rs3781117, rs3781118, and rs1330581. None of these three SNPs were transmitted to affected children more frequently than expected by chance (Table S1). Inclusion of these GAD2 SNPs with the original three did not yield significant results for association between any GAD2 haplotype and obesity (Table 6). Moreover, rs3781117, rs3781118, and rs1330581 were not independently associated with class III obesity in either the US or Canadian case-control study groups (Table S2).
Table 6 TDT Results for Six SNPs Spanning the GAD2 Haplotype Block
Ethnic differences in GAD2 allele frequency
The presence of an underlying population substructure, resulting from ethnic admixture, is a common bias in association studies [26]. We were therefore interested in determining whether different ethnic groups could display significant differences in GAD2 allele frequency.
In samples obtained from the Human Variation Collection (Coriell Institute for Medical Research, Camden, New Jersey, United States), the frequencies of GAD2 alleles in Caucasians were comparable with those observed for the Caucasian groups tested in previous studies (Table S3). However, we observed marked and highly significant differences in allele frequency for the −243 A>G SNP between Caucasians and populations of West African origin represented by samples collected in the US or in France. North African populations presented an intermediate allelic distribution.
Reporter Gene Assay for GAD2 −243 G Promoter Variant
In addition to the aforementioned genetic results, Boutin et al. [18] also found that luciferase reporter genes containing the −243 G allele in the GAD2 promoter (from −1710 to −4, relative to the transcriptional start site) displayed a 6-fold higher activity compared to reporter genes containing the −243A allele in βTC3 murine insulinoma cells (Figure 3 in [18]). We were interested in investigating the nature of this allele-specific difference in GAD2 promoter activity, with the goal of identifying the specific cis-acting elements responsible. To accomplish this, we also tested the effect of the −243 A>G SNP on transcription of a luciferase reporter gene in βTC3 cells.
We found that introduction of the −243 G allele into the −1710/−4 reporter gene did not elicit detectable effects on luciferase transcription relative to the WT reporter gene (Figure S3). Similarly, we could not detect any allele-specific effects of the −243 A>G polymorphism in two smaller reporter genes containing the GAD2 promoter (from −501 to −4 and from −1,234 to −4). However, the transcriptional activity of our WT −1710/−4 reporter gene was appreciably higher than that of pGL3Basic in βTC3 cells, suggesting that the GAD2 promoter does exhibit some basal transcriptional activity in this cell line.
Electrophoretic Mobility Shift Assay
In the previous study [18], oligonucleotide probes containing either of the −243 A>G alleles were tested for their affinity for nuclear extract prepared from βTC3 murine insulinoma cells. The probe containing the −243 G allele was found to have a 6-fold higher affinity for an unidentified nuclear protein (Figure 4 in [18]). However, the DNA–protein complex was also present to some extent in the negative control lanes (lacking nuclear extract), and the oligonucleotide probes differed with respect to their specific activity.
We obtained the sequences of oligonucleotide probes used by the authors, and utilized the electrophoretic mobility shift assay (EMSA) to confirm this allele-specific difference in binding affinity. We were also interested in determining whether this effect was specific for neuronal or β cells relative to other cell lines. Our experiments indicated that the −243 A allele had a greater affinity for an unidentified protein from βTC3 nuclear extracts relative to the −243 G allele (Figures S4 and S5). Our results were not consistent with those previously described by Boutin et al. [18].
Discussion
In the present study, we attempted to replicate the important recent findings of Boutin et al. [18], which implicated three SNPs in GAD2 (the −243 A>G allele and a haplotype of the +61450 C>A and +83897 T>A SNPs) in the predisposition to class III human obesity. To replicate their findings, we first performed family-based tests of association for all three SNPs in 693 nuclear families segregating severe obesity (2,359 participants, nearly four times as many participants as in the original report). This group of individuals included 89 families found to have linkage of severe obesity to Chromosome 10p12 [16,22]. No evidence for excess transmission of any GAD2 alleles or haplotypes from parents to affected offspring was obtained. Next, we conducted an adequately powered case-control study to test the association between class III obesity and the GAD2 −243 A>G variant in Caucasians. Consistent with the family-based association results, we did not observe any association between the −243 G variant and class III obesity in 680 cases and 1,186 lean controls. These findings were also obtained in a meta-analysis for the association between the −243 A>G SNP and class III obesity. Lastly, we obtained results from the reporter gene and DNA binding experiments for the −243 A>G variant that were inconsistent with the original report. Overall, we found that (i) a haplotype consisting of the WT alleles at SNPs +61450 C>A and +83897 T>A does not appear to protect against severe, early-onset obesity, (ii) the −243 A>G SNP is not associated with class III obesity in adults, (iii) other haplotypes in the region of GAD2 are not associated with severe obesity, and (iv) the −243 A>G SNP does not elicit detectable effects on transcription of a luciferase reporter gene in βTC3 murine insulinoma cells.
Irreproducibility of positive findings has been a common criticism leveled at association studies investigating the common genetic basis of complex diseases [19,24]. The reasons cited are numerous, and include a lack of statistical power to detect small to moderate effects, lack of control over the Type I error rate, overinterpretation of marginal data, population stratification, and poor biological plausibility [27,28]. Regarding the conflicting results obtained by Boutin et al. [18] and the current study, it is likely that the lack of replication could be ascribed to any of these causes, which are discussed below. The inconsistencies between association studies may also reflect the complex interactions between multiple population-specific genetic and environmental factors.
The lack of statistical power to detect alleles of minor effect is likely to have contributed to the differences between the study by Boutin et al. [18] and the current investigation. Based on the findings of the initial report, we conducted an adequately powered, ethnically matched, case-control study. Although our results overlapped with the size of the initial effect, they did not show a significant association between the −243 G allele and class III obesity (Figure 1). We estimate that we had 60% power to detect a significant difference (α of 0.05) in allele frequency between our pooled groups of cases and controls, assuming that the −243 G allele (frequency of 0.18) was the disease allele, a genotype relative risk of 1.25, and a prevalence of class III obesity in the general population of 5% [29]. The family-based association tests had a similar amount of power (∼60%), given the same assumptions. Under these conditions, the original study [18] may have been underpowered. Moreover, it must be pointed out that the marginally significant association (p = 0.04) they observed between the −243 G allele and class III obesity was observed in only one of their two groups of participants, and did not reach nominal significance in their family-based analysis (p = 0.06). Although the lack of statistical significance does not exclude the possibility of an association (as we cannot rule out smaller effects), the data do not support a relationship between this SNP and class III obesity.
The interpretation of results from genetic association studies is frequently complicated by other statistical issues, such as a failure to control for multiple hypothesis testing, overinterpretation of marginal data as positive trends, and the well-documented tendency for initial positive findings to overestimate the strength of the association [21]. This “jackpot” phenomenon [24] can be readily observed in our meta-analysis (Figure 1).
Population stratification may also account for some of the inconsistencies observed between association studies, though its importance may have been overestimated [19,26]. Population stratification is usually controlled for by careful matching of cases and controls by ethnicity, using family-based tests of association (such as the TDT) or studying multiple case-control populations [30]. Considering the marked differences in allele frequency that we observed between ethnic groups for the GAD2 SNPs (the −243 A>G and +61450 C>A SNPs in particular), as well as the known differences in the prevalence of class III obesity between Caucasian Americans and African Americans [31], it is plausible that a small difference in ancestry between cases and controls could lead to spurious claims of association. Naturally, future studies of the GAD2 gene should carefully take this into consideration.
There is no obvious explanation for the differences in results obtained for the EMSA and reporter gene assays. Regarding the EMSA, a major problem with these experiments is that most random DNA sequences will be bound by a nuclear extract from any cell line (Figures S4 and S5 and Figure 4 in [18]). It is likely that the introduction of single base-pair differences into this DNA sequence will interfere with the binding pattern observed. Moreover, while an allele-specific difference in the binding of βTC3 cell nuclear extract definitely occurs for the −243 A>G polymorphism, this observation is of limited physiological significance, because: (i) it appears to be restricted to this cell type (and there is no apparent difference in allele-specific binding for nuclear extract derived from a neuronal cell line); and (ii) the binding of this nuclear protein does not appear to affect transcription of a luciferase reporter gene in βTC3 cells. Finally, even if the −243 A>G SNP did affect transcription of the reporter gene in this context, there is no prior biological evidence to suggest that perturbation of GAD2 expression in β cells could exert detectable effects on long-term energy homeostasis.
This latest point raises the issue of biological plausibility. GAD2 encodes the 65-kDa isoform of the enzyme glutamate decarboxylase, which catalyzes the production of γ-aminobutyric acid (GABA), a major inhibitory neurotransmitter, from glutamic acid. The biological evidence implicating GAD2 as a candidate gene (and by extension, hypothalamic GABA levels as causative) in severe obesity is as follows: GAD2 mRNA is co-expressed with neuropeptide Y in neurons of the hypothalamic arcuate nucleus that act in the nearby paraventricular nucleus and other hypothalamic areas to stimulate food intake [32]. Concomitantly, these arcuate neuropeptide Y neurons inhibit the parallel and opposing effects of neighboring pro-opiomelanocortin/cocaine- and amphetamine-regulated transcript neurons via GABA-ergic collateral inputs [33]. In rats, administration of muscimol, a GABAA receptor agonist, into either the third ventricle or the hypothalamic paraventricular nucleus stimulates feeding in a dose-dependent manner [34]. Similarly, inhibition of GABA synthesis in the ventromedial hypothalamus, by injection of antisense GAD-65 and GAD-67 oligonucleotides, has been shown to suppress food intake [35].
However, enthusiasm for GAD2 as a candidate gene for severe obesity is dampened somewhat by the observation that GAD2-deficient mice appear normal with respect to behavior, locomotion, reproduction, and glucose homeostasis, but suffer from epileptic seizures [36]. Also, levels of GAD2 mRNA in the arcuate nucleus of the rat do not change in response to 48 hr of food deprivation, as do levels of prepro–neuropeptide Y mRNA [37]. Furthermore, the notion that hypothalamic GABA levels are proportional to food intake may be an oversimplification; although microinjection of GABA into the paraventricular nucleus and ventromedial hypothalamus stimulates feeding [38], injection of GABA, muscimol [39], or an adenovirus expressing GAD2 [40], into the lateral hypothalamus of rats has been observed to have the opposite effect.
While these experimental results do not exclude GAD2 as a candidate gene for human obesity, it remains possible that the linkage signal could be due to variation in a neighboring gene. Certainly GAD2 is the leading candidate in this region, due to some of the biological evidence presented above and the location of D10S197 within one of its introns. However, in light of the large number of genes involved in energy homeostasis (recently reviewed in [41] and [42]), the multiple tissue-specific roles of each gene, and the readily available information regarding the homology and expression pattern of uncharacterized genes, it now seems possible to make a tenuous case for almost any single gene in the regulation of body weight. For example, a preliminary glance at the Chromosome 10p12 region yielded several interesting genes: TPRT (trans-prenyltransferase), the enzyme that elongates the prenyl side-chain of coenzyme Q, one of the key elements of the respiratory chain within mitochondria; GPR158, which encodes a metabotropic glutamate, GABAB–like G-protein-coupled receptor; and PTF1A, which encodes pancreas-specific transcription factor 1a. Although only a little is known about each of these genes, it is possible to speculate on the potential role of each in obesity. GAD2 is no exception. At present, however, there is insufficient genetic or biological evidence to implicate genetic variation in GAD2 in the predisposition to severe obesity in humans.
Materials and Methods
Participants
The ascertainment strategy for these participants has been described previously [43]. BMI was calculated as weight in kg/(height in m)2. For the PDT and TDT analyses, we genotyped 973 (extremely) obese children and adolescents (693 probands and 280 of their siblings: mean age 14.0 ± 3.7 y, mean BMI 31.0 ± 6.0 kg/m2) and both of their parents (mean age 42.7 ± 5.9 y, mean BMI 30.4 ± 6.1 kg/m2). Written informed consent was given by all participants and, in the case of minors, their parents. The Ethics Committee of the University of Marburg approved the study.
Participants were selected from the Cardiovascular Research Institute Genomic Resource in Arteriosclerosis, a population-based investigation of dyslipidemia and atherosclerotic heart disease established at the University of California, San Francisco (UCSF) in California, United States. This population includes patients from the Lipid Clinic of UCSF [44,45], from the UCSF Interventional Cardiology Service, and from collaborating cardiology clinics throughout California. The UCSF Committee on Human Research approved the protocols, and informed written consent was obtained from all patients. From this study group, we selected class III obese (BMI ≥ 40 kg/m2) non-Hispanic Caucasian individuals, as well as a group of lean individuals (mean BMI = 22.9 kg/m2, range 20.0–25.6 kg/m2) with a similar age and sex distribution. Genomic DNA was extracted from buffy coats using a Puregene DNA Purification Kit from Gentra Systems (Minneapolis, Minnesota, United States).
Obese Caucasian individuals with a mean BMI of 48 kg/m2 (range 36–81 kg/m2, with 91% of individuals having a BMI > 40 kg/m2) were recruited from the Ottawa Hospital Weight Management Clinic. Age- and sex-matched lean Caucasian individuals with a BMI below the 10th percentile for age and sex were recruited as controls from the Ottawa region. The Human Ethics Research Boards of the Ottawa Hospital and the University of Ottawa Heart Institute approved the study. Informed written consent was obtained from all participants.
Genotyping
Three GAD2 SNPs were genotyped by PCR-based restriction fragment length polymorphism analysis or by tetra-amplification refractory mutation system PCR. To detect rs2236418 (−243 A>G), a PCR-amplicon of 636 base pairs (bp) (primers: GAD2–243-F: 5′-
GGAGCCAGACCTCAAACAAA-3′ and GAD2–243-R: 5′-
TTTGGAGACTGGAGCAGGTC-3′) was digested by DraI ([NEB GmbH, Frankfurt am Main, Germany]; 2 h at 37 °C; A-allele: 395 bp and 241 bp; G-allele: undigested). To detect rs928197 (+83897 T>A), a PCR-amplicon of 242 bp (primers: GAD2–83897-F: 5′-
GTGGCAGGCAGCTGATAGTC-3′ and GAD2–83897-R: 5′-
CACCTGTGGGACAGACCATA-3′), was digested by AluI ([Fermentas GmbH, St. Leon Rot, Germany]; 2 h at 37 °C; T-allele: 146 bp and 96 bp; A-allele: undigested). PCR products of all SNPs were electrophoresed in 2.5% agarose gels stained with ethidium bromide.
SNP rs992990 (+61450 C>A) was genotyped by tetra-amplification refractory mutation system PCR [46]. Primers were as follows: GAD-61450-FiC 5′-
ATTCTTACTGACAAAGCTGAGTTTACCC-3′ and GAD-61450-Ro 5′-
TATTTAGGTGAAGTGCTTAGAACTGTGC-3′ 199-bp amplicon detects the C-allele; GAD-61450-RiA 5′-
TCATGTTCTATGGCTAGATGTCTAATCCT-3′ and GAD-61450-Fo 5′-
GGCAGCTTCTCTTCTAAAAAGACAAATA-3′ 151-bp amplicon detects the A-allele. The amplicon length of the two outer primers (GAD-61450-Fo and GAD-61450-Ro) was 294 bp.
Positive controls of variant genotypes were run on each gel. To test validity of the genotypes, allele determinations were rated independently by at least two experienced persons. Discrepancies were resolved unambiguously, either by reaching consensus or by retyping.
Genotyping of the US participants for the GAD2 SNPs rs2236418, rs992990, and rs928197 was performed using fluorescently labeled allele-specific primer extension, assayed by fluorescence polarization template-directed dye incorporation [47]. The primers used to amplify the region around each SNP were as follows: rs2236418p1
CCTCCCTCTCTCGTGTTTTT, rs2236418p2
GTGTCACGCAGGAACAGAAA, rs928197p1
CCTCTTATCACTTGCAGGATCT, rs928197p2
GTGGTTCCATACTCCATCATTC, rs992990p1
GGGACAGAGAATTCAGTGACAG, and rs992990p2
GTCATTTGTGAGCTTGGTGAC. Single-base extension reactions for each SNP were performed using the primers rs2236418p4
TTGGAAGCCGGGGAGC, rs928197p3
AAACAATAAGGTTCTGACTGTTGAGC, and rs992990p4
CATGTTCTATGGCTAGATGTCTAATTC.
To test for ethnicity-specific differences in allele frequency, we also genotyped 99 Caucasian-American and 99 African-American individuals from the Human Variation Collection (Coriell Institute for Medical Research, Camden, New Jersey, United States), as well as 60 West Africans and 36 North Africans living in Paris, France.
To capture more than 95% of the haplotype diversity in the GAD2 region, we also genotyped the SNPs rs3781117, rs3781118, and rs1330581 (Tables 6, S1, and S2) using fluorescence polarization template-directed dye incorporation. The primers used were as follows: rs3781117p1, rs3781117p2, rs3781118p1, rs3781118p2, rs1330581p1, and rs1330581p2 (sequences available from the authors on request). For each of these SNPs, single-base extension reactions were performed with the primers rs3781117p3, rs3781118p3, and rs1330581p3 (as above).
Genotyping of the Canadian participants for the −243 A>G polymorphism was performed by PCR amplification with the primers GAD2_243A>G_ALU_F (
GGCTCCCTTTCCCTCAAAT) and GAD2_243A>G_ALU_R (
ATAACGTGTGTGTATGCGAGCTGGAGA) followed by digestion with AluI. When separated by agarose gel electrophoresis, this produced a unique set of bands corresponding to each genotype: AA (20, 47, and 92 bp), AG (20, 47, 92, and 139 bp), and GG (20 and 139bp).
For the SNPs (−243 A>G, +61450 C>A, and +83897 T>A), 94 DNA samples from each laboratory (Marburg, Germany; and San Francisco, California, United States) were exchanged and genotyped according to the other laboratory's method. Three discrepancies (out of 564 genotypes) were observed, resulting in a between-laboratory error rate of < 1%.
Family-based tests of association for single markers were carried out using the PDT, which accounts for the dependency between the sibs [23]. Haplotype TDTs for two and three markers were performed using the program GeneHunter, version 2.0 beta [48] (available at http://helix.nih.gov/apps/bioinfo/genehunter.html). Here, transmissions were counted only when phase could be determined unambiguously. TDT analysis of all six GAD2 SNPs (comprising the haplotype block) was performed using the program UNPHASED [49] implementing the EM (expectation-maximization) algorithm (available at http://www.mrc-bsu.cam.ac.uk/personal/frank/). For the US and Canadian participants, comparisons between cases and controls for allele frequency were performed using a two-tailed χ2 test, and p-values were calculated using the program GraphPad Prism version 3.0a for Macintosh (GraphPad Software, San Diego California, United States). For the meta-analysis, we used the Mantel-Haenszel method to calculate stratified summary effects using a fixed effect model. Power calculations were performed using the Genetic Power Calculator [50] provided at http://statgen.iop.kcl.ac.uk/gpc/.
Reporter gene constructs
The effects of the −243 A>G polymorphism on GAD2 transcriptional activity were tested in the context of a luciferase reporter gene. The oligonucleotides used to amplify a 2,200-bp fragment of the GAD2 promoter (sense-
CGGGTCTCTGCTTTGTTAGC and antisense-
TTTGGAGACTGGAGCAGGTC) were incorrectly specified in the original report [18], as sequence comparisons between them and the GAD2 promoter sequence suggested that they would have yielded a 1,706-bp PCR product. Moreover, neither oligonucleotide contained restriction sites for BglII or HindIII, as stated.
After communication with the author, P. Boutin, we amplified the −1710/−4 region using the primers GAD2PROMF3 (our designation)
CGGGGTACCCGGGTCTCTGCTTTGTTAGC and GAD2PROMR3
CAAGCTTTGGAGACTGGAGCAGG, digested the PCR product with KpnI and HindIII, and inserted it into the KpnI and HindIII sites in front of the firefly luciferase coding sequence, contained in the vector pGL3Basic (Promega, Madison, Wisconsin, United States). This vector was referred to hereafter as the GAD2 −1710/−4 construct (numbers refer to the regions of the GAD2 promoter, relative to the transcriptional start site). The −243 G variant allele was introduced into this construct by PCR amplification of the above fragment from a homozygous patient, digestion of this PCR product with NotI and HindIII, and substitution of this fragment into the NotI-HindIII sites of the WT construct. PCR was performed using TaKaRa LA Taq according to the manufacturer's instructions (TaKaRa Biomedicals, Otsu, Shiga, Japan). Using restriction enzyme digestion of the WT and variant GAD2 −1710/−4 constructs, we also made shorter GAD2 promoter reporter genes, referred to as −501/−4 and −1234/−4. All reporter genes were sequenced prior to transfection, and corresponded exactly with the human Chromosome 10 sequence provided on the UCSC Genome Bioinformatics site (http://genome.ucsc.edu).
EMSA
Sequences for EMSA probes were obtained from P. Boutin. The oligonucleotides used included GAD2A−243AF (
CTCTTTTAAA
GCTCCCCGGCTTCC), GAD2A−243AR (
GGAAGCCGGGGAGCT
TTAAAAGAG), GAD2−243GF (
CTCTTTTAAG
GCTCCCCGGCTTCCC
TTAAAAGAG). Bases in bold type indicate the differences introduced to reflect the −243 A>G polymorphism. Five hundred nanograms of each forward (F) oligonucleotide were end-labeled with γ-32P ATP (Perkin-Elmer, Boston, Massachusetts, United States) using T4 polynucleotide kinase (Promega) at 37 °C for 30 min. Subsequently, 1.5 μg of the corresponding unlabelled reverse (R) oligonucleotide and 50 μl of annealing buffer (100 mM NaCl in TE buffer) were added to each labeled (F) oligonucleotide, and the mixture was incubated for 10 min at 95 °C before being cooled slowly for 1–2 h. The resulting labeled, double-stranded probe was then column-purified (Stratagene NucTrap, La Jolla, California, United States), and the concentration of the probe in the eluate was approximately 10 ng/μl. Unlabeled double-stranded probes for competition experiments were also prepared in a similar manner.
All EMSA experiments were performed in a 20-μl reaction volume containing binding buffer (10 mM HEPES [pH 7.9], 75 mM KCl, 2.5 mM MgCl2, 0.1 mM EDTA, 1 mM DTT, 3% Ficoll), polydI/dC (final concentration 50 mg/l in TE), 1 μl nuclear extract, and approximately 3 ng of labeled probe. Nuclear extracts from βTC3, Neuro2A, T98G, HepG2, and HEK293 cells were prepared using the method of Schreiber et al. [51]. Cells from which nuclear extracts were prepared were maintained as described below. After addition of the probe, the mixture was incubated for 10 min at room temperature before loading onto a 5% nondenaturing acrylamide gel containing 0.5 × TBE (1× Tris-Borate EDTA buffer is 0.09 M Tris-borate, 2 mM EDTA). Gels were electrophoresed for approximately 2 h, dried, and exposed to autoradiographic film for 1–2 d.
All cells were maintained in a water-jacketed incubator set to 37 °C with 5% carbon dioxide. The murine insulinoma βTC3 cells were grown in DMEM supplemented with 15% horse serum (Hyclone, Logan, Utah, United States), 2.5% fetal bovine serum (Hyclone), and penicillin/streptomycin. The neuro2A cells were maintained in MEM supplemented with 10% horse serum, 5% fetal bovine serum, 2 mM L-glutamine, 1% non-essential amino acids, and antibiotics. The T98G glioblastoma cells were grown in MEM containing Earle's Balanced Salt Solution (EBSS), 10% fetal bovine serum, 1% non-essential amino acids, sodium pyruvate, and antibiotics. The HepG2 cells and human embryonic kidney (HEK293) cells were maintained separately in MEM containing EBSS, 10% fetal bovine serum, 2mM L-glutamine, 1% non-essential amino acids, and antibiotics.
Supporting Information
Figure S1 Haploview of the GAD2 Region on Chromosome 10
This figure was generated using data from the International HapMap Project (http://www.hapmap.org/cgi-perl/gbrowse/gbrowse/hapmap) and using the program Haploview (http://www.broad.mit.edu/mpg/haploview/index.php). The SNPs studied are indicated on the diagram by the following numbers: rs2236418 (#2); rs3781118 (#3), rs3781117 (#4), rs1330581 (#12), and rs928197 (#34). On the diagram, the blue squares indicate missing data and unfilled red squares indicate a high degree of linkage disequilibrium (linkage disequilibrium coefficient, D′ = 1) between pairs of markers. Lesser degrees of linkage disequilibrium are indicated by the lighter red shading.
(95 KB PPT).
Click here for additional data file.
Figure S2 Haplotype Tag SNPs Required to Capture > 95% of the Haplotype Diversity within the GAD2 Region
The SNPs genotyped in the initial phase of the study (−243 A>G/rs2236418, +61450 C>A/rs992990, and +83897 T>A/rs928197) are indicated on the upper part of the diagram as markers 2, 24, and 35, respectively. Haplotypes are depicted as rows, with their population frequency shown at the right side of each row. The SNPs that are in complete linkage disequilibrium with each other are shaded the same color. In order to determine > 95% of the haplotype information within the GAD2 region, genotypes at each of the SNPs indicated by the arrowheads (markers 1, 3, 4, 12, and 14, or a marker in perfect linkage disequilibrium with each) were required. To accomplish this, markers rs3781118 (#3 on diagram), rs3781117 (#4), and rs1330581 (#12) were genotyped in the second phase of the study.
(81 KB PPT).
Click here for additional data file.
Figure S3 Results from Transient Transfection of GAD2 Reporter Genes in βTC3 Cells Containing the −243 A>G Polymorphism
Three different sizes of luciferase reporter gene were constructed from the GAD2 promoter (−1710/−4, −501/−4 and −1234/−4) for transfection into βTC3 murine insulinoma cells. Each WT reporter construct contains the −243 A allele, and the corresponding mutant reporter construct is identical to the WT except for the introduction of the −243 G allele. Twenty-four h before transfection, βTC3 cells were seeded in 6-well plates at a density of 250,000 cells/well containing DMEM supplemented with 10% fetal calf serum (Hyclone, Logan, Utah, United States), 2 mM L-glutamine, and penicillin/streptomycin. On the day of the experiment, each well was transfected with pGL3Basic or a GAD2 promoter construct (0.4 μg) as well as 20 ng of the plasmid pRL-RSV (Promega), which encodes Renilla luciferase, to control for transfection efficiency. Transfections were performed in triplicate using Effectene reagent (Qiagen, Valencia, California, United States). Forty-eight h after transfection, cells were lysed, and firefly and Renilla luciferase assays were performed on the lysate using the Dual Luciferase Reporter Assay System (Promega), according to the manufacturer's standard protocol. Each experiment was repeated three times. We observed no significant difference in luciferase activity between each pair of WT and mutant GAD2 promoter constructs.
(36 KB PPT).
Click here for additional data file.
Figure S4 EMSA
Radiolabeled double-stranded oligonucleotide probes for each of the −243 A>G alleles were incubated with various nuclear extracts and electrophoresed in a 5% nondenaturing polyacrylamide gel. The arrow on the left side of the figure indicates the DNA-nuclear protein complex formed with βTC3 nuclear extract. Allele-specific differences in binding to βTC3 nuclear extract are seen in lanes 2 and 8.
(2.5 MB PPT).
Click here for additional data file.
Figure S5 Competitive EMSA Using Nuclear Extract from βTC3 Cells
The complex formed by the interaction between the radiolabeled double-stranded probe containing the −243 A allele and βTC3 nuclear extract (indicated by the arrow) was competed away by the addition of excess amounts of unlabeled −243 A probe (lanes 3–6), but not by the addition of the same amount of unlabeled −243 G probe (lanes 7–10).
(101 KB PPT).
Click here for additional data file.
Table S1 PDT Results for rs3781117, rs3781118, and rs1330581
(32 KB DOC).
Click here for additional data file.
Table S2 Association Study Results for rs3781117, rs3781118, and rs1330581 in US and Canadian Groups
None of the variant alleles at any of these three SNPs were associated with class III obesity in either of the two case-control groups or when pooled (for rs3781117, C allele: OR = 0.92, 95% CI 0.75–1.13, p = 0.43; for rs3781118, G allele: OR = 0.94, 95% CI 0.75–1.19, p = 0.62; for rs1330581, G allele: OR = 1.03, 95% CI 0.89–1.20, p = 0.65).
Can. Cases, Canadian cases.
(64 KB DOC).
Click here for additional data file.
Table S3 Genotype results for GAD2 SNPs in US Caucasians, African Americans, and Africans
Differences in allele frequency between ethnic groups were assessed by χ2 test. For the −243 A>G and +61450 C>A SNPs, the differences between Caucasian Americans (CA) and African Americans were highly significant (p < 0.001). For the −243 A>G SNP, the differences between CA, West Africans (WA), and North Africans (NA) were significant (CA vs. WA, p < 0.001; CA vs. NA, p = 0.014; WA vs. NA, p < 0.001). The frequency of the +83897 T>A SNP was also significantly different between CA and WA (p = 0.013). Other comparisons either yielded non-significant results, or were not conducted, as samples did not conform to Hardy-Weinberg equilibrium (indicated by an asterisk).
(53 KB DOC).
Click here for additional data file.
Accession Number
The GenBank (http://www.ncbi.nlm.nih.gov/Genbank) accession number for GAD2 is AY340073.
This research was supported by the National Institutes of Health (RO1 DK60540) and an American Diabetes Association Career Development Award to CV. These studies were carried out in part in the General Clinical Research Center, Moffitt Hospital, University of California, San Francisco (UCSF), with funds provided by the National Center for Research Resources (5 M01 RR-00079, US Public Health Service). This research was also supported by the UCSF Diabetes and Endocrinology Research Center (P30 DK63720). Portions of this research were also conducted at the US Department Of Energy's Joint Genome Institute under contract DE-AC0378SF00098 administered by the University of California (UC). FG, AS, AH and JH are supported by the BMBF (Bundesministerium für Bildung und Forschung) (NGFN1: 01GS0118 and NGFN2: 01GS0482). JPK is supported by the Mildred V. Strouss Charitable Trust, the Joseph Drown Foundation, and UC Discovery Grants.
The authors would like to thank John Todd (Cambridge Institute for Medical Research), Neil Risch (Center for Human Genetics, UCSF), Michael German, and Stuart Smith (UCSF Diabetes Center) for helpful discussions regarding the manuscript. We would also like to acknowledge the efforts of Marco Patti, Karen Bagatelos (Department of Surgery, UCSF), and James Ostroff (Department of Medicine, UCSF) in the recruitment of severely obese participants, and Trang Nguyen (Phillips University, Marburg, Germany) for additional statistical analysis.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. MMS, BW, LAP, DLL, AH, JH, and CV conceived and designed the experiments. MMS, BW, DLL, and AU performed the experiments. MMS, BW, DLL, MMC, FG, AS, and WCH analyzed the data. LAP, MMC, RM, MM, WR, FMJ, CRP, JPK, RD, RM, PYK, AH, and JH contributed reagents/materials/analysis tools. MMS, LAP, AH, JH, and CV wrote the paper.
Citation: Swarbrick MM, Waldenmaier B, Pennacchio LA, Lind DL, Cavazos MM, et al. (2005) Lack of support for the association between GAD2 polymorphisms and severe human obesity. PLoS Biol 3(9): e315.
Abbreviations
BMIbody mass index
bpbase pair
CIconfidence interval
EMSAelectrophoretic mobility shift assay
GABAγ-aminobutyric acid
ORodds ratio
PDTpedigree disequilibrium test
SNPsingle nucleotide polymorphism
WTwild-type
==== Refs
References
Calle EE Thun MJ Petrelli JM Rodriguez C Heath CW Body-mass index and mortality in a prospective cohort of U.S. adults N Engl J Med 1999 341 1097 1105 10511607
Eckel RH Krauss RM American Heart Association call to action: Obesity as a major risk factor for coronary heart disease. AHA Nutrition Committee Circulation 1998 97 2099 2100 9626167
Kopelman PG Obesity as a medical problem Nature 2000 404 635 643 10766250
Must A Spadano J Coakley EH Field AE Colditz G The disease burden associated with overweight and obesity JAMA 1999 282 1523 1529 10546691
Stunkard AJ Foch TT Hrubec Z A twin study of human obesity JAMA 1986 256 51 54 3712713
Maes HH Neale MC Eaves LJ Genetic and environmental factors in relative body weight and human adiposity Behav Genet 1997 27 325 351 9519560
Barsh GS Farooqi IS O'Rahilly S Genetics of body-weight regulation Nature 2000 404 644 651 10766251
O'Rahilly S Farooqi IS Yeo GS Challis BG Minireview: Human obesity-lessons from monogenic disorders Endocrinology 2003 144 3757 3764 12933645
Montague CT Farooqi IS Whitehead JP Soos MA Rau H Congenital leptin deficiency is associated with severe early-onset obesity in humans Nature 1997 387 903 908 9202122
Strobel A Issad T Camoin L Ozata M Strosberg AD A leptin missense mutation associated with hypogonadism and morbid obesity Nat Genet 1998 18 213 215 9500540
Clement K Vaisse C Lahlou N Cabrol S Pelloux V A mutation in the human leptin receptor gene causes obesity and pituitary dysfunction Nature 1998 392 398 401 9537324
Jackson RS Creemers JW Ohagi S Raffin-Sanson ML Sanders L Obesity and impaired prohormone processing associated with mutations in the human prohormone convertase 1 gene Nat Genet 1997 16 303 306 9207799
Krude H Biebermann H Luck W Horn R Brabant G Severe early-onset obesity, adrenal insufficiency and red hair pigmentation caused by POMC mutations in humans Nat Genet 1998 19 155 157 9620771
Swarbrick MM Vaisse C Emerging trends in the search for genetic variants predisposing to human obesity Curr Opin Clin Nutr Metab Care 2003 6 369 375 12806208
Hager J Dina C Francke S Dubois S Houari M A genome-wide scan for human obesity genes reveals a major susceptibility locus on chromosome 10 Nat Genet 1998 20 304 308 9806554
Hinney A Ziegler A Oeffner F Wedewardt C Vogel M Independent confirmation of a major locus for obesity on chromosome 10 J Clin Endocrinol Metab 2000 85 2962 2965 10946912
Price RA Li WD Bernstein A Crystal A Golding EM A locus affecting obesity in human chromosome region 10p12 Diabetologia 2001 44 363 366 11317669
Boutin P Dina C Vasseur F Dubois S Corset L GAD2 on chromosome 10p12 is a candidate gene for human obesity PLoS Biol 2003 1 e68 10.1371/journal.pbio.0000068 14691540
Redden DT Allison DB Nonreplication in genetic association studies of obesity and diabetes research J Nutr 2003 133 3323 3326 14608039
Perusse L Rankinen T Zuberi A Chagnon YC Weisnagel SJ The human obesity gene map: The 2004 update Obes Res 2005 13 381 490 15833932
Ioannidis JP Ntzani EE Trikalinos TA Contopoulos-Ioannidis DG Replication validity of genetic association studies Nat Genet 2001 29 306 309 11600885
Saar K Geller F Ruschendorf F Reis A Friedel S Genome scan for childhood and adolescent obesity in German families Pediatrics 2003 111 321 327 12563058
Martin ER Monks SA Warren LL Kaplan NL A test for linkage and association in general pedigrees: The pedigree disequilibrium test Am J Hum Genet 2000 67 146 154 10825280
Hirschhorn JN Lohmueller K Byrne E Hirschhorn K A comprehensive review of genetic association studies Genet Med 2002 4 45 61 11882781
Barrett JC Fry B Maller J Daly MJ Haploview: Analysis and visualization of LD and haplotype maps Bioinformatics 2005 21 263 265 15297300
Cardon LR Palmer LJ Population stratification and spurious allelic association Lancet 2003 361 598 604 12598158
Editorial Freely associating Nat Genet 1999 22 1 2 10319845
Cardon LR Bell JI Association study designs for complex diseases Nat Rev Genet 2001 2 91 99 11253062
Flegal KM Carroll MD Ogden CL Johnson CL Prevalence and trends in obesity among US adults, 1999–2000 JAMA 2002 288 1723 1727 12365955
Hirschhorn JN Altshuler D Once and again—Issues surrounding replication in genetic association studies J Clin Endocrinol Metab 2002 87 4438 4441 12364414
Freedman DS Khan LK Serdula MK Galuska DA Dietz WH Trends and correlates of class 3 obesity in the United States from 1990 through 2000 JAMA 2002 288 1758 1761 12365960
Ovesjo ML Gamstedt M Collin M Meister B GABAergic nature of hypothalamic leptin target neurones in the ventromedial arcuate nucleus J Neuroendocrinol 2001 13 505 516 11412337
Cowley MA Smart JL Rubinstein M Cerdan MG Diano S Leptin activates anorexigenic POMC neurons through a neural network in the arcuate nucleus Nature 2001 411 480 484 11373681
Pu S Jain MR Horvath TL Diano S Kalra PS Interactions between neuropeptide Y and gamma-aminobutyric acid in stimulation of feeding: A morphological and pharmacological analysis Endocrinology 1999 140 933 940 9927326
Bannai M Ichikawa M Nishihara M Takahashi M Effect of injection of antisense oligodeoxynucleotides of GAD isozymes into rat ventromedial hypothalamus on food intake and locomotor activity Brain Res 1998 784 305 315 9518663
Kash SF Johnson RS Tecott LH Noebels JL Mayfield RD Epilepsy in mice deficient in the 65-kDa isoform of glutamic acid decarboxylase Proc Natl Acad Sci U S A 1997 94 14060 14065 9391152
Schwartz MW Sipols AJ Grubin CE Baskin DG Differential effect of fasting on hypothalamic expression of genes encoding neuropeptide Y, galanin, and glutamic acid decarboxylase Brain Res Bull 1993 31 361 367 7683962
Leibowitz SF Brain monoamines and peptides: Role in the control of eating behavior Fed Proc 1986 45 1396 1403 2869977
Rattan AK Mangat HK Electrical activity and feeding correlates of intracranial hypothalamic injection of GABA, muscimol and picrotoxin in the rats Acta Neurobiol Exp (Wars) 1990 50 23 36 2220435
Noordmans AJ Song DK Noordmans CJ Garrity-Moses M During MJ Adeno-associated viral glutamate decarboxylase expression in the lateral nucleus of the rat hypothalamus reduces feeding behavior Gene Ther 2004 11 797 804 14961066
Flier JS Obesity wars: Molecular progress confronts an expanding epidemic Cell 2004 116 337 350 14744442
Saper CB Chou TC Elmquist JK The need to feed: Homeostatic and hedonic control of eating Neuron 2002 36 199 211 12383777
Hinney A Bornscheuer A Depenbusch M Mierke B Tolle A Absence of leptin deficiency mutation in extremely obese German children and adolescents Int J Obes Relat Metab Disord 1997 21 1190 9426388
Pullinger CR Hennessy LK Chatterton JE Liu W Love JA Familial ligand-defective apolipoprotein B. Identification of a new mutation that decreases LDL receptor binding affinity J Clin Invest 1995 95 1225 1234 7883971
Pullinger CR Eng C Salen G Shefer S Batta AK Human cholesterol 7alpha-hydroxylase (CYP7A1) deficiency has a hypercholesterolemic phenotype J Clin Invest 2002 110 109 117 12093894
Ye S Dhillon S Ke X Collins AR Day IN An efficient procedure for genotyping single nucleotide polymorphisms Nucleic Acids Res 2001 29 E88 11522844
Chen X Levine L Kwok PY Fluorescence polarization in homogeneous nucleic acid analysis Genome Res 1999 9 492 498 10330129
Spielman RS Ewens WJ A sibship test for linkage in the presence of association: The sib transmission/disequilibrium test Am J Hum Genet 1998 62 450 458 9463321
Dudbridge F Pedigree disequilibrium tests for multilocus haplotypes Genet Epidemiol 2003 25 115 121 12916020
Purcell S Cherny SS Sham PC Genetic Power Calculator: Design of linkage and association genetic mapping studies of complex traits Bioinformatics 2003 19 149 150 12499305
Schreiber E Matthias P Muller MM Schaffner W Rapid detection of octamer binding proteins with ‘mini-extracts,' prepared from a small number of cells Nucleic Acids Res 1989 17 6419 2771659
|
16122350
|
PMC1193520
|
CC BY
|
2021-01-05 08:21:26
|
no
|
PLoS Biol. 2005 Sep 30; 3(9):e315
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030315
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 10.1371/journal.pbio.0030321SynopsisGenetics/Genomics/Gene TherapyDiabetes/Endocrinology/MetabolismPathologyHomo (Human)A Genetic Link to Obesity: The Numbers Don't Add Up for GAD2
Synopsis9 2005 30 8 2005 30 8 2005 3 9 e321Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Lack of Support for the Association between GAD2 Polymorphisms and Severe Human Obesity
==== Body
Obesity is a leading cause of preventable death and is often linked to type II diabetes and heart disease. Being a complex trait, obesity is likely caused by the interplay of multiple environmental factors and many genes. Common genetic differences between individuals within a region of Chromosome 10 have previously been associated with obesity. This region contains several genes with the potential to be directly involved in the disease. One of these genes, GAD2, has been the subject of many studies. A new study by Michael Swarbrick, Björn Waldenmaier, Christian Vaisse, and their colleagues takes a new look at GAD2 and provides strong evidence that the gene might not be as relevant to obesity as previously thought.
GAD2 encodes a protein (called GAD-65) involved in the production of GABA, a neurotransmitter involved in a variety of brain functions, including appetite stimulation and energy consumption. Studies in mice have shown that increased levels of GABA result in hunger and overeating. In healthy mice, the levels of GAD2, and hence, GABA, are controlled, making sure that the balance between weight gain and loss is maintained. A 2003 study of a French population found that three genetic mutations in and around the GAD2 gene occurred at a high level in individuals with obesity. The 2003 study, conducted by different researchers, was also published in PLoS Biology. When Swarbrick et al. surveyed German, Caucasian-American, and Canadian populations for this genetic correlation, however, they found no statistically significant link between obesity and any of the mutations.
Scientists believe that genetic mutations in a specific region in Chromosome 10 play a role in obesity and have studied one gene, GAD2, intensively. But a new study finds no evidence linking GAD2 mutations with obesity
There are many possible reasons why different studies may show different results: ethnic differences between populations, as well as behavioral and dietary differences, could account for varying results when it comes to studying a trait as complex as obesity. Also, studies that seek to show an association between genetic differences and complex diseases rely heavily on the statistical power of their tests, which depends on the number of subjects involved. Swarbrick et al. have not only studied 2,359 German, 729 US, and 1,137 Canadian subjects, but also conducted a “meta-analysis”—a statistical analysis of a collection of individual studies—of their data and the previously published data from 1,221 French subjects. Meta-analyses help identify patterns from multiple individual studies that may not be visible in any one study alone, and also help rule out chance differences that may be apparent in one single study. In this case, the meta-analysis showed that when the results from French subjects are put together with the results from other ethnic populations, there is no evidence for a link between changes in GAD2 and obesity.
Although GAD2's role in controlling appetite made it an exciting candidate for a link to obesity-related conditions, Swarbrick et al. show that the numbers simply don't add up. The search for serious obesity gene contenders in this region of Chromosome 10 is all set to continue—and attention can now turn to several other potential gene candidates located nearby.
|
0
|
PMC1193521
|
CC BY
|
2021-01-05 08:21:27
|
no
|
PLoS Biol. 2005 Sep 30; 3(9):e321
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030321
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 10.1371/journal.pbio.0030332SynopsisAnimal BehaviorNeuroscienceNutritionHow Fruitflies Know It's Time for Lunch Synopsis9 2005 30 8 2005 30 8 2005 3 9 e332Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Candidate Gustatory Interneurons Modulating Feeding Behavior in the Drosophila Brain
==== Body
To control what you eat and when, your nervous system must coordinate a laundry list of signals: internal signals contain information about energy level, food preferences, and metabolic need, while external signals relay information about the quality of available food, determined by its smell and taste. Scientists studying the fruitfly Drosophila have traced the path of olfactory signals beginning with chemical receptors in the mouth, which set off neurons that signal the antennal lobe of the central nervous system. From here, the electrical stimulation zooms toward the so-called mushroom body, a mushroom-shaped cluster of neurons involved in olfactory processing. Less is known about the gustatory signals, which begin both in the mouth and in the pharynx and aim toward the subesophageal ganglion region of the fly's brain. How olfactory and gustatory signals influence feeding patterns remains murky.
In a new study, Michael Pankratz and Christoph Melcher used genetic analysis to gain insight into the adult and larval neural networks that use taste information to regulate eating. Specifically, they found that several types of neurons responsible for coordinating taste signals express the gene hugin (hug), a gene linked to abnormal eating activity and expressed in only the subesophageal ganglion. By altering hug expression, the researchers uncovered the gene's behavioral influence: hug-expressing neurons influence a fly's decision to sample new food sources. The researchers also proposed that hug proteins play a role in hormone-triggered growth, an important consequence of adequate feeding.
To begin their investigation, Melcher and Pankratz analyzed the DNA from flies with abnormal eating behavior. One group of these flies shared a mutant klumpfuss (klu) gene, normally responsible for encoding a protein transcription factor. Because neural transcription factors control production levels of other neural proteins, the researchers used DNA microarrays to compare gene expression in normal flies to that in klu mutants. Any klu-controlled genes expressed at different levels in klu mutants might contain clues about the neural circuitry modulating feeding behavior.
Using microarrays, Melcher and Pankratz discovered that mutant fly larvae overexpress the hug gene, which is known to encode at least two neural proteins related to growth signaling. The researchers then investigated which signals influence hug expression by exposing larvae to either high or low food levels. Because both starved and sugar-fed flies express little hug, the researchers inferred that hug levels do not solely signal internal energy requirements but respond to internal and external signals carrying information about the quality of food. The researchers also noted that the finicky pumpless (ppl) mutants, which have a feeding defect similar to klu, overexpress hug.
Behavioral studies confirmed that too much hug reduces food intake and leads to stunted growth, while too little stimulates eating. Melcher and Pankratz selected a group of flies and blocked the synapses of their hug neurons to inhibit the neurons' activity. In contrast to control flies, which start feeding on a novel food source only after an evaluation phase (they wait a while before initiating feeding), the experimental flies started eating new food right away. These hug neurons may help flies decide whether or not to eat a new food source.
Microarray, neuroanatomical, and biochemical analyses identified taste-sensitive neurons that help regulate feeding behavior in frutifly larvae
Larvae express hug in only about 20 neurons, all located in the subesophageal ganglion. The axons of some of these hug neurons extend into the ring gland, a crucial metabolism and growth organ in flies. Other axons contact the protocerebrum, a structure close to brain centers for learning and remembering odors. A third set of these axons extend to throat muscles—which is surprising because most subesophageal ganglion neurons have no connection to motor function. All together, these few hug neurons can signal structures controlling growth, feeding, and learning and memory.
Besides linking hug neurons to brain centers that regulate taste-related feeding behavior, the study also raises questions about how the nervous system prioritizes internal and external signals. How hungry must flies be to overcome taste aversion? How do the competing neural networks of taste and hunger signals decide whether the fly will eat? Future studies pairing behavioral and genetic analysis may begin to reveal answers to these open questions.
|
0
|
PMC1193522
|
CC BY
|
2021-01-05 08:21:27
|
no
|
PLoS Biol. 2005 Sep 30; 3(9):e332
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030332
|
oa_comm
|
==== Front
PLoS GenetPLoS GenetpgenplgeplosgenPLoS Genetics1553-73901553-7404Public Library of Science San Francisco, USA 1613208110.1371/journal.pgen.001001505-PLGE-RA-0065R2plge-01–02–09rResearch ArticleEvolutionMicrobiologyGenetics/Comparative GenomicsGenetics/Chromosome BiologyEubacteriaComparative and Evolutionary Analysis of the Bacterial Homologous Recombination Systems Recombination Systems in BacteriaRocha Eduardo P. C 12*Cornet Emmanuel 2Michel Bénédicte 31 Unité Génétique des Génomes Bactériens, Institut Pasteur, Paris, France
2 Atelier de Bioinformatique, Université Pierre et Marie Curie, Paris, France
3 Laboratoire de Génétique Microbienne, Institut National de la Recherche Agronomique, Jouy en Josas, France
Bell Stephen D EditorMRC Cancer Cell Unit, United Kingdom*To whom correspondence should be addressed. E-mail: [email protected] 2005 26 8 2005 1 2 e1531 3 2005 9 6 2005 Copyright: © 2005 Rocha 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.Homologous recombination is a housekeeping process involved in the maintenance of chromosome integrity and generation of genetic variability. Although detailed biochemical studies have described the mechanism of action of its components in model organisms, there is no recent extensive assessment of this knowledge, using comparative genomics and taking advantage of available experimental data on recombination. Using comparative genomics, we assessed the diversity of recombination processes among bacteria, and simulations suggest that we missed very few homologs. The work included the identification of orthologs and the analysis of their evolutionary history and genomic context. Some genes, for proteins such as RecA, the resolvases, and RecR, were found to be nearly ubiquitous, suggesting that the large majority of bacterial genomes are capable of homologous recombination. Yet many genomes show incomplete sets of presynaptic systems, with RecFOR being more frequent than RecBCD/AddAB. There is a significant pattern of co-occurrence between these systems and antirecombinant proteins such as the ones of mismatch repair and SbcB, but no significant association with nonhomologous end joining, which seems rare in bacteria. Surprisingly, a large number of genomes in which homologous recombination has been reported lack many of the enzymes involved in the presynaptic systems. The lack of obvious correlation between the presence of characterized presynaptic genes and experimental data on the frequency of recombination suggests the existence of still-unknown presynaptic mechanisms in bacteria. It also indicates that, at the moment, the assessment of the intrinsic stability or recombination isolation of bacteria in most cases cannot be inferred from the identification of known recombination proteins in the genomes.
Synopsis
Genomes evolve mostly by modifications involving large pieces of genetic material (DNA). Exchanges of chromosome pieces between different organisms as well as intragenomic movements of DNA regions are the result of a process named homologous recombination. The central actor of this process, the RecA protein, is amazingly conserved from bacteria to human. In addition to its role in the generation of genetic variability, homologous recombination is also the guardian of genome integrity, as it acts to repair DNA damage. RecA-catalyzed DNA exchange (synapse) is facilitated by the action of presynaptic enzymes and completed by postsynaptic enzymes (resolvases). In addition, some enzymes counteract RecA. Here, the researchers assess the diversity of recombination proteins among 117 different bacterial species. They find that resolvases are nearly as ubiquitous and as well conserved at the sequence level as RecA. This suggests that the large majority of bacterial genomes are capable of homologous recombination. Presynaptic systems are less ubiquitous, and there is no obvious correlation between their presence and experimental data on the frequency of recombination. However, there is a significant pattern of co-occurrence between these systems and antirecombinant proteins.
Citation:Rocha EPC, Cornet E, Michel B (2005) Comparative and evolutionary analysis of the bacterial homologous recombination systems. PLoS Genet 1(2): e15.
==== Body
Introduction
Homologous recombination was originally described as being the result of the sexual process—in bacteria as in eukaryotes—and was later identified as a major DNA repair process. Both genetic and biochemical studies revealed the crucial role of homologous recombination in all organisms for the repair of a variety of DNA damage of exogenous and endogenous origin [1,2]. Indeed, in all organisms in which it has been tested, inactivation of RecA causes a dramatic increase of sensitivity to all DNA-damaging agents used in laboratories. In addition to its housekeeping role in repair, recombination is fundamental for the genetic diversification of bacterial genomes. First, in bacteria it allows the integration of homologous alien DNA, arising from transformation or conjugation [3,4]. Second, by allowing allelic recombination between closely related strains [5], it assorts adaptive mutations and purges deleterious mutations hitchhiking with them [6]. Third, recombination between homologous segments in the genomes leads to chromosomal instability [7,8], and among bacteria, the rate of chromosome rearrangements correlates with the number of repeated sequences in the genomes [9]. Fourth, intrachromosomal homologous recombination between large repeated regions is often adaptive, allowing the generation of genotypic diversity, e.g., in pathogens [10–12].
The general outline of homologous recombination is common to all organisms studied to date. It involves a central step of strand-invasion and strand-exchange catalyzed by RecA or a RecA homolog. RecA is ubiquitous and highly conserved in sequence. Strand exchange is preceded by the action of enzymes called presynaptic enzymes. These enzymes act on DNA to render it accessible to RecA and thus allow the formation of a RecA filament, which is single-stranded DNA (ssDNA) coated with RecA molecules. The steps that follow strand exchange and result in the formation of a viable recombinant molecule are termed postsynaptic and are mainly the resolution of the recombination intermediate made by RecA. The entire process and the enzymes involved have been originally defined and extensively characterized in Escherichia coli, which has become a paradigm for homologous recombination [1,13,14]. For this reason, the E. coli genes were used in this work to search for homologs in other bacteria. Genes of Bacillus subtilis, the second model bacteria, were used for enzymes absent from E. coli.
The initiation of homologous recombination in E. coli may follow the RecBCD or the RecFOR pathway (Figure 1). Both pathways work to provide a ssDNA molecule coated with RecA to allow the invasion of a homologous molecule [13,15]. RecBCD promotes the repair of double-stranded DNA (dsDNA) breaks, whereas RecFOR is involved in the repair of ssDNA gaps. In the RecBCD pathway, all the required functions—helicase, nuclease, and RecA loading—are assembled in a single holoenzyme [16]. RecBCD binds to dsDNA ends, unwinds, and degrades DNA until it encounters a χ site. The activity of RecBCD is modified at χ, where it starts producing ssDNA and loading RecA [17]. RecF, RecO, and RecR bind gapped ssDNA and displace the SSB proteins to allow RecA coating. There is evidence for interactions between RecR and either RecF or RecO, but not for the existence of a tricomponent complex [18,19]. The RecJ ssDNA exonuclease acts in concert with RecFOR to enlarge the ssDNA region when needed. Strand exchange is catalyzed by RecA [20], a multifunctional protein also involved in the regulation of the SOS response and in the activity of polymerases that facilitate replication across DNA lesions [21]. In E. coli, the joint molecules formed by RecA are resolved by either the RuvABC complex or, in an unknown way, by the action of the RecG helicase. The RuvAB and RuvC proteins catalyze the branch migration and the resolution of Holliday junction recombination intermediates, respectively. These three proteins are thought to interact in a resolvasome complex, in which a RuvABC-junction complex tracks along DNA, with RuvC able to scan for cleavable sequences as the DNA passes through (Figure 1). Finally, replication is directly linked to the recombination process during double-strand break repair, as a viable recombinant is only obtained if the recombination intermediate is used to initiate replication, via the action of the PriA protein [22,23]. Conversely, recombination proteins participate in replication progression as, for example, RecFOR and RecA are required for the resumption of a normal replication rate after treatment with a DNA-damaging agent, and RecBC is required for the viability of several replication mutants [2].
Figure 1 Models for the Mode of Action of the Main Homologous Recombination Proteins in E. coli at ssDNA Gaps or dsDNA Ends
Enzymes of known biochemical activities are shown. The presynaptic steps result in the formation of a RecA filament. At gaps, this step requires RecJ, RecF, RecO, and RecR: the 5′ ssDNA exonuclease RecJ enlarges the ssDNA region (possibly with the help of various helicases, as no specific helicase is required for gap repair); RecF, RecO, and RecR promote RecA binding to SSB-coated DNA. At dsDNA ends, RecBCD (AddAB in B. subtilis) degrades DNA until it encounters a χ site; its helicase-nuclease activity is then modified to produce a 3′-ended ssDNA, to which it loads RecA. The synaptic step (homology search and strand exchange) is always performed by RecA and results in the formation of a Holliday junction (X structure). The postsynaptic steps are the migration and the resolution of Holliday junctions. Migration can be performed by RuvAB or by RecG, and resolution is made by RuvC (RecU in B. subtlis; RuvC forms a complex with RuvAB in E. coli). In addition, RecBCD-mediated recombination is always coupled with PriA-dependent replication restart. Antirecombinases are not shown: UvrD and MutLS prevent by different means the strand exchange reaction. In recBC mutants, the presynaptic steps of dsDNA end repair can be catalyzed by the helicase RecQ and the gap repair proteins RecJ and RecFOR, a reaction that is prevented by SbcB (and SbcCD) nucleases.
Evidence is accumulating that other bacteria use different proteins for some recombination steps. For example, in firmicutes, RecBCD is replaced by the analogous complex AddAB (named RexAB in streptococci and lactococci) [24,25], and there is evidence indicating that a functional χ site is present in these genomes, albeit variable in size and composition [26]. In these genomes, RecU also replaces RuvC [27]. The frequency of homologous recombination is diminished by the action of other proteins. The general mismatch repair system (MutS1LH in E. coli) antagonizes homologous recombination between nonidentical DNA sequences by blocking the RecA-mediated strand exchange process if mismatches are present [28]. Hence, the mismatch repair system prevents recombination between homeologous sequences and has an important role in defining bacterial species barriers [29]. The helicase II, UvrD, also acts as an antirecombinant, possibly by unwinding the paired DNA recombinant intermediates [30], or by displacing RecA from ssDNA [31]. On the other hand, UvrD can stimulate RecA-driven branch migration and can participate in the RecFOR pathway [32]. Finally, in recBC mutant cells, RecFOR can initiate recombination from DNA double-strand ends that have a single-strand extension, but only when SbcB, a ssDNA-specific 3′ → 5′ exonuclease, is inactivated. When present, this nuclease prevents RecFOR action by removing the 3′ extremity on which RecA could be loaded; in addition, the growth of recBC sbcB mutants requires the inactivation of the SbcCD proteins for unknown reasons [1,33]. Antirecombinant proteins must be taken into account when assessing the potential recombination machinery of bacteria, as it can be evaluated from genome sequences.
An extended assessment of the proteins involved in DNA repair followed the publication of the first genome sequences [34]. This pioneering work showed that genes implicated in homologous recombination are not homogeneously distributed among bacterial species. Unfortunately, no equivalent extensive work has been done recently that focuses precisely on homologous recombination and takes advantage of the nearly 200 completely sequenced genomes. Yet different sets of recombination-related genes have been found among some bacterial groups [35–38]. We have thus tried to assess the distribution of homologous recombination genes in complete genomes, using a large set of tools involving sequence and phylogenetic analysis [34,39], as well as colocalization data. This type of analysis presupposes that recombination proteins are ancient enough to have diverged from one or a few proteins for which we know the function for at least one element in the family. Although recombination is probably a very old process, our data suggest that some genes may have been missed because they are not yet functionally characterized. A further assumption of our analysis is that sequence similarity will remain strong enough to allow finding these genes by sequence similarity. We make some simulations that suggest that few genes are likely to have been lost if sequence divergence follows the pattern (but not necessarily the rate) of RecA. The use of genomic context should also reduce this problem. Finally, this analysis also supposes that orthologs have similar functions. Although this is usually assumed, proteins with multiple functions may have gained or lost part of them during evolution. For example, the role of RecA in SOS is unused—possibly lost—in the bacteria that lack this response. After establishing the repertoire of genes, we evaluate their co-occurrence, evolutionary rate, and colocalization, taking into account their functional association in known pathways. This was then put into relation with the evolutionary history of genes and the assessment of the experimental evidence for recombination.
Results/Discussion
Introductory Remarks
As described in Materials and Methods, we first applied an automatic methodology to find candidate orthologs of genes implicated in homologous recombination. The analysis started from genomes for which experimental evidence was available for the function of the genes. This typically included not only E. coli and B. subtilis but also much less studied bacteria such as mollicutes (for RuvAB [40]), actinobacteria (for Ku [41]), or others. Naturally, when an ortholog was found in a phylogenetic group, it was used to search for further orthologs within the group. Second, we made a more detailed analysis by searching for InterPro domains and making FASTA searches; and by taking into account phylogenetic analyses and information on gene colocalization. Using these diverse sources of information, we were able to list candidate homologous recombination genes in 117 genomes (Figure 2). Some genes are highly conserved in sequence and nearly ubiquitous. For these genes, the methods we used are very reliable and provide uniformly consistent results. However, for some less ubiquitous, fast-evolving, or poorly characterized genes, we found sometimes either inconsistent similarity or weak hits, e.g., similarity smaller than 40%, FASTA hits with E ~10−5, matches with a nonspecific motif or with large variation in protein length. Under these conditions, and when no reliable close ortholog is available, it is hazardous to confidently predict orthology. Hence, we conservatively regard these genes as “putative” orthologs. For some proteins, e.g., RecO and RecX, the list of putatives is relatively large.
Figure 2 Probable Presence, Putative, and Unlikely Presence of Recombination-Associated Genes in the Studied Genomes
Black indicates presence is probably, grey indicates putative presence, and white indicates presence is unlikely. F indicates that the gene is present in the genome but contains frame shifts (genes with known programmed frame shifts, introns, and inteins are indicated, as regular genes, in black).
RecA and Resolvases Are Nearly Ubiquitous Genes
No homologous recombination gene is present in all bacterial genomes. However, many genes are widespread among all or nearly all groups and are extremely frequent within each group (Figure 2). RecA is absent only in the several genomes of Buchnera and Blochmania and presents frame shifts in Onion Yellows (OY) phytoplasma. The near ubiquity of RecA matches well with its preeminent role in homologous recombination and has been previously noticed [34,42,43]. Its absence among intracellular bacteria has also been widely documented [36,44–46]. Unsurprisingly, bacteria lacking RecA have very few other recombination proteins. Several proteins are almost as frequent as RecA. The genes coding for the RuvAB Holliday junction branch migration complex always co-occur and are absent from the genomes that lack RecA and from only two genomes where RecA is present, Wigglesworthia Gb and Aquifex aeolicus (Figures 2 and 3). Although they lack RuvAB, these two genomes contain a RecG ortholog—another Holliday junction branch migrating helicase. The gene for RecG is also very frequent, absent only from all mollicutes and all Chlamydiacea as well as from Desulfovibrio vulgaris.
Figure 3 Distribution of Some Recombination Genes in a Phylogenetic Tree of Bacteria
Tree adapted from [99]. The position of Pirellula and Fusobacterium are still unclear [100].
Some proteins are believed to be functional analogs, although they apparently lack a common evolutionary history (i.e., they are not orthologs). RuvAB in E. coli forms a complex with the resolvase RuvC. RuvC is less ubiquitous than RuvAB, which is explained by its functional replacement by the analog RecU in firmicutes and mollicutes [27]. Our data indicate that only ten genomes lack both RuvC and RecU (this includes the genomes that lack RecA; Table 1). In these rare cases, the resolvase function may be provided by YqgF [47], which is only absent from seven genomes. However, our data suggest that RuvC/RecU and YqgF are not simple functional analogs because they co-occur in the large majority of genomes. In addition, a resolvase activity of the YqgF proteins has not yet been demonstrated either in vitro or in vivo. The function performed by resolution proteins may also be carried out by prophage-encoded proteins [48].
Table 1 Genomes Missing Key Components of the Homologous Recombination
Listed in each column are the genomes missing the components given at the top of each column.
aPresence of putative prophage resolvase [47].
bPresence of putative ubiquitous resolvase YqgF.
PriA is nearly ubiquitous and is only absent in genomes of some intracellular endosymbionts, Deinococcus radiodurans, A. aeolicus, and from most genomes of mollicutes. Among actinobacteria, there is a putative ortholog of PriA that is smaller and very divergent. With the exception of Candidatus Blochmania floridanus (which lacks RecA), all genomes with AddAB or RecBCD (the presynaptic proteins that act at double-strand ends) have PriA. In conclusion, RecA, branch migration systems, and resolvases, and to a lesser extent the protein that couples recombination and replication PriA, are present in nearly all the bacterial genomes (Table 1).
The RecBCD and AddAB Presynaptic Recombination Proteins
RecBCD provides another example of complementary distribution of similar but nonorthologous systems. The AddAB proteins (and their orthologs RexAB) replace RecBCD in firmicutes and in most β- and α-proteobacteria. AddAB is almost ubiquitous among these groups, as it is missing only in Bacillus halodurans, Neisseria meningitidis, and Chromobacterium violaceum—these having RecBCD instead. A recent work analyzed a homolog of AddA in proteobacteria and confirmed its role in the repair of double-strand breaks [49]. Although AddA and AddB closely co-occur in most genomes, the AddB gene of B. subtilis has no significant similarity with the ones of proteobacteria (E > 0.01 for FASTA hits, <25% identity on a global alignment). Because AddB is slightly more conserved than AddA among firmicutes (see following), one would expect the AddB protein of proteobacteria to have significant similarity with the AddB protein of firmicutes if it shared a common evolutionary history. Hence, the AddB proteins of the two clades may be functional analogs but not othologs. This is consistent with recent data indicating that AddA shares stronger resemblance with RecB than AddB does with RecC, reflecting a more central role for the function of RecB/AddA in the complex (M. El Karoui, personal communication).
Genes coding for proteins that participate in complexes tend to systematically co-occur in genomes. This is the case for AddAB, RuvAB, RuvAB/RuvC(RecU), SbcCD, and MutS1L (see following). A major exception to this trend is the frequent presence of a RecD protein when RecBC is absent, in mollicutes, firmicutes, D. radiodurans, both Streptomyces, and Des. vulgaris. The phylogenetic tree of this protein (Figure 4) shows a clear separation between RecD1 (a protein systematically associated with RecBC) and RecD2 (a protein present in genomes lacking RecBC). Within each RecD group, one can identify most of the major phylogenetic groups of bacteria. For example, among actinobacteria, the Mycobacterium (with RecBC) and the two Streptomyces (without) are on opposite sides of the tree, and a similar contrast is found in δ-proteobacteria, where Geobacter sulfurreducens has RecBC and Des. vulgaris does not. In some genomes, such as Chlamydiacea, there are multiple copies of RecD, typically one in each side of the tree. The analysis of the protein sequences of the two groups of RecD shows a major difference between them. RecD2 contains an N-terminus extension including a domain identified as RuvA domain 2–like in InterPro that is absent from RecD1. This domain is also present in UvrC and is essential for the 5′ incision in the prokaryotic nucleotide excision repair process [50]. The RecD2 protein of D. radiodurans, the only one biochemically studied, is a DNA helicase with a low processivity and a yet-unidentified role [51].
Figure 4 Unrooted Phylogenetic Tree of the RecD Protein
The dotted line separates genomes containing RecBCD from the ones containing only RecD. The tree was constructed using Tree-Puzzle, using the JTT+Γ model with eight classes [74]. Bootstraps were made using SEQBOOT and CONSENSE from the PHYLIP package [101]. C. acetobutylicum, Clostridium acetobutylicum; C. tepidum, Chlorobium tepidum; C. violaceum, Chromobacterium violaceum; D. vulgaris, Desulfovibrio vulgaris; E. carotovora, Erwinia carotovora; E. faecalis, Enterococcus faecalis; L. plantarum, Lactobacillus plantarum; L. lactis, Lactococcus lactis; P. multocida, Pasteurella multocida; M. mobile, Mycoplasma mobile; M. florum, Mesoplasma florum; M. mycoides, Mycoplasma mycoides; M. pulmonis, Mycoplasma pulmonis; P. maritima, Procholorcoccus maritima; S. enterica, Salmonella enterica; S. oneidensis, Shewanella oneidensis; X. fastidiosa, Xylella fastidiosa.
Finally, some bacteria have a functional nonhomologous end-joining mechanism (NHEJ), allowing the repair of dsDNA breaks [52]. Contrary to homologous recombination, NHEJ does not require sequence homology—only complementary ends. The key factors of NHEJ are a Ku protein that binds to the termini of the double-strand breaks and has the bridging activity, and a ligase that ligates the termini. Our results indicate that NHEJ genes are present in few bacteria (Ku is present in 24 genomes out of the 117), with no particular phylogenetic trend, as they are found in firmicutes, actinobacteria, and several groups of proteobacteria (see Figure 2). As indicated previously [53,54], the two genes tend to co-occur contiguously in genomes, probably constituting an operon. In some bacteria, we found many copies of the Ku/ligase genes. For example, Agrobacterium tumefaciens contains six copies of the Ku gene and eight copies of the ligase, and Bradyrhizobium japonicum contains four copies of the Ku gene and two copies of the ligase. Thus, in these genomes, Ku has probably a very important role. We then tested the patterns of co-occurrence of NHEJ and RecBCD/AddAB to see whether the presence of one could compensate for the absence of the other (as both act to repair double-strand breaks). We found these systems to co-occur independently (p = 0.6, χ2 test). NHEJ is the major pathway for repairing DNA double-strand breaks in mammalian cells, whereas homologous recombination is so in yeast [55]. Because most bacterial genomes lack NHEJ, homologous recombination also appears to be the major repair pathway acting on such lesions in bacteria.
The RecFOR Presynaptic Proteins
Whereas the RecB, RecC, and RecD polypeptides form a stable active complex, in the RecFOR pathway, there are interactions between some of the elements but no stable complexes between the three proteins. Interestingly, the RecBCD/AddAB and RecFOR proteins, instead of showing a complementary pattern of co-occurrence, tend to co-occur more frequently than expected (p < 0.001, χ2 test). This means that if RecBCD/AddAB is present (absent), then RecFOR is more likely to be present (absent), which probably reflects the specificity of these two systems on complementary types of lesions (see Figure 1).
Although RecF historically served as a reference for this pathway, it is absent from 29 genomes and is the least frequent protein in the set (see Figure 2). At the other extreme, RecR is the most frequent, being absent from only ten genomes, followed by RecO, which, counting putative orthologs, is only absent from 19 genomes. In agreement with RecR being present in the two active complexes RecOR and RecFR [18,19], there is no single occurrence of RecO or RecF when RecR is absent.
In E. coli, the RecJ exonuclease acts during gap repair to enlarge the ssDNA region for RecFOR binding [56]. RecJ is absent from the species that lack RecA and from the mollicutes and the mycobacteria, which may use an alternative exonuclease. RecQ is absent from 48 genomes, in agreement with the observation that the RecQ helicase is required in E. coli for RecFOR-mediated recombination only in a recBC sbcB sbcCD mutant [57].
Recombination without Presynaptic Recombination Proteins?
Our analysis indicates that certain bacterial genomes lack most presynaptic recombination proteins (see Figure 2). One possibility is that these genomes lack homologous recombination altogether. This may be the case for some species lacking nearly all homologous recombination proteins, such as all Buchnera, or the OY phytoplasma (Table 1). However, for the genomes containing RecA and resolvases, this is most unlikely. We therefore made an extensive analysis of the literature and selected genomes lacking most presynaptic proteins but for which there is evidence for homologous recombination (Table 2). Such evidence comes from experimental studies of the homologous recombination processes or experimental studies that have used homologous recombination to engineer/inactivate genes, and from multilocus sequence typing data that indicate a population structure driven by frequent recombination. One also typically assumes that natural transformation is used for recombination repair or gene acquisition, which suggests that competent bacteria should have some type of homologous recombination [4,58]. It is surprising that highly recombining genomes, such as Helicobacter pylori [59–61] or Streptomyces coelicolor [62] lack a large fraction of the presynaptic proteins. One should note that with the exception of both Streptomyces, these genomes also lack NHEJ, and many also code for antirecombinants, such as MutS2. This suggests that either presynaptic proteins are dispensable for efficient homologous recombination in some genomes or other, unknown systems, exist in these genomes. The first hypothesis is supported by data indicating that some E. coli recA mutations (RecA P67W, RecA441, RecA730, and RecA803) can displace SSB proteins much more efficiently than the wild-type, and thus function in the absence of presynaptic proteins [63]. However, if some genomes lack presynaptic functions because their RecA protein is able to efficiently bind SSB-covered DNA, it is not through one of the studied RecA mutations in E. coli, because we did not find any of these mutations in natural genomes. Furthermore, it remains to be understood how organisms lacking presynaptic functions could control RecA activity to avoid its improper fixation to any ssDNA (e.g., on the template of the replicating lagging strand). Yet-unidentified presynaptic systems may exist in these genomes. Recombination presynaptic functions are fulfilled in eukaryotes by proteins that have no homology with E. coli proteins, in spite of their capacity to facilitate the binding to DNA of their cognate RecA homolog [64].
Table 2 Bacteria for Which Some Evidence Exists of Homologous Recombination, but That Lack Elements of Both RecBCD/AddAB and RecFOR in the Genome
The genes present for each group are indicated. Types of evidence: natural transformation (NT); population is panmitic as observed from MLST data (MLST); genetic engineering is performed using homologous recombination (GE); recombination studies (RS).
aNot all genomes have all proteins.
bPutative.
Proteins That Antagonize Homologous Recombination
Another way of increasing the frequency of homologous recombination without making changes in the recombination machinery is to eliminate the function of antirecombinant proteins. We tested whether there are associations between the losses of presynaptic systems and the losses of antirecombinant proteins, such as UvrD, MutS1L, MutS2, and SbcB genes. UvrD is nearly ubiquitous. The presence of MutS1L correlates with the presence of RecBCD/AddAB and RecFOR (RecBCD/AddAB: observed 102, expected 69; RecFOR: observed 91, expected 80; both p < 0.005, Pearson's exact test). This suggests that a lower activity of RecA in the absence of presynaptic systems can be compensated for by the loss of the mismatch repair system. Contrary to MutS1, MutS2 is not involved in mismatch repair and suppresses homologous recombination between identical sequences, in addition to homeologous recombination, in H. pylori [60]. However, no significant association was found between the presence or absence of MutS2 and that of the presynaptic systems. As the H. pylori enzyme is the only MutS2 that has been studied in detail so far, it is possible that the antirecombination property of this MutS2 protein is specific for this species.
SbcB, which in RecBC− backgrounds prevents the repair of double-strand breaks by RecFOR, has a statistically significant pattern of co-occurrence and co-omission with RecBCD/AddAB (observed 63, expected 53, p < 0.01, Pearson's exact test), but not with RecFOR (p > 0.1, same test). In fact, only one of the bacteria lacking RecBC/AddAB contains SbcB. This indicates that the absence (presence) of RecBCD/AddAB is correlated with the absence (presence) of this antirecombinant gene, which may allow RecFOR to efficiently repair double-strand breaks in RecBCD−/AddAB− backgrounds. SbcCD is much more frequent than SbcB and also co-occurs with RecBCD/AddAB (observed 64, expected 52, p < 0.01, Pearson's exact test). However, the role of SbcCD in homologous recombination is unclear.
Colocalization of Genes
Genes involved in a common mechanism tend to be tightly coregulated and, for this reason, clustered in the genome [65]. We have therefore searched for the colocalization of these genes among our set of genomes. With few exceptions, we found that only the recombination genes that are part of stable complexes are systematically clustered. The addAB genes colocalize in 20 of 21 co-occurrences among firmicutes, the exception being Clostridium tetani. Among proteobacteria these genes are together in 13 of 13 genomes. The three genes for RecBCD were found to colocalize in 28 of their 31 co-occurrences. RuvA and RuvB colocalized in 77 of 111 co-occurrences, with exceptions including all chlamydiacea, all cyanobacteria, all ɛ-proteobacteria, all streptococci, all bacteroides, and most spirochetes, as well as a few phylogenetically dispersed genomes. RuvA, RuvB, and RuvC colocalized in 45 of 78 co-occurrences of the three genes. In firmicutes and mollicutes, RuvC is replaced by RecU, but this gene only colocalizes with RuvAB in two genomes (Mycoplasma genitalium and M. pneumoniae). Thus, RecU and RuvC are very different in this respect. YqgF was rarely found close to other recombination genes. The two key genes for NHEJ (Ku and the ligase) were found together in 19 of 24 genomes. Naturally, as for the co-occurrence of genes in genomes, the closeness of their co-occurrence is influenced by the phylogenetic distribution of the available genomes. Close occurrence of genes in highly sampled clades, e.g., firmicutes or proteobacteria, will be more preeminent than in clades with few available sequences.
RecA and RecX are close in many genomes and are partly coexpressed in E. coli [66]. In some bacteria, the overexpression of RecA is toxic in the absence of RecX, and in vitro, RecX modulates the action of RecA by blocking the extension of the RecA filament [67]. However, although in E. coli RecX inhibits the action of RecA [68], in Neisseria gonorrhoeae its inactivation leads to a decrease in homologous recombination [66]. Expanding previous observations [69], we found that 35 of the 37 co-occurrences of bona fide orthologs of recX colocalize with recA. The exceptions are N. meningitidis and Photorhabdus luminescens. In contrast, very few genes among the more distantly related, putative recX orthologs are physically close to recA genes. In particular, the putative recX of firmicutes are systematically far in the chromosome from recA. The proteins coded by these genes are larger and less than 40% similar to the RecX from E. coli and from actinobacteria. It is thus uncertain whether they perform the same function. However, RecX also shows large relative variations in length among well-characterized orthologs (e.g., among γ-proteobacteria the E. coli protein has 166 residues, whereas in Yersinia pestis it has 188, and in Shewanella oneidensis it has 123). It has been suggested that the uncoupling between recA and the putative recX in N. gonorrhoeae and B. subtilis could be associated with their competence for natural transformation [66]. However, such uncoupling is a characteristic of all firmicutes, not specifically of the competent ones, and it is not found in other competent bacteria such as Haemophilus influenzae or H. pylori (which lacks RecX).
Although recF, recR, and recO do not colocalize, both recF and recR often colocalize with genes coding for replication proteins. Many genomes have an operon close to the replication origin containing four genes: dnaA (involved in replication initiation), dnaN (β-clamp of the DNA polymerase III), recF, and gyrB (DNA gyrase) [70]. Among the 86 occurrences of recF, it is close to dnaA in 54, close to dnaN in 58, and close to gyrB in 52. The four genes are together in 40 genomes. Finally, the dnaX gene, which encodes both the τ and γ subunits of E. coli DNA polymerase III, is close to recR in E. coli, and the genes are partially cotranscribed [71]. Among the 97 genomes containing dnaX and recR, the genes colocalize in 65. These results indicate that instead of clustering together, recombination genes that are not part of stable complexes are often colocalized with genes involved in replication. The linkage between genes of these two cellular processes is certainly associated with the role of homologous recombination in repairing DNA lesions that block DNA synthesis [72,73].
Relative Evolutionary Rates of the Proteins
The substitution rate of proteins is the result of the interplay between mutation and functional constraints. Hence, if one discounts horizontal gene transfer, the differences in substitution rates between proteins should reflect their relative tolerance to change (i.e., they should be associated with the fraction of changes that allows maintaining the function). To assess the relative tolerance of each recombination protein to changes, we computed evolutionary distances within the sets of all bona fide orthologs, using Tree-Puzzle [74]. We then used RecA as the reference protein because of its near ubiquity and slow evolutionary rate [42]. The regression analyses of the substitution rates of each protein as a function of the substitution rate of RecA showed one single group in which RecA evolves faster—the mollicutes (data not shown). We have thus not used these points in the regressions. All other proteins were then compared to RecA, and we found a considerable diversity among the different proteins in terms of substitution rates (Figure 5). A more developed version of this method has recently been proposed to find horizontal gene transfer between distant taxa [75]. Using our data, we found very little evidence of such events (data not shown). RuvB has evolved almost as slowly as RecA (16% faster), whereas some proteins have evolved a little faster, such as RecR (+68%) and RecU (+100%). However, most proteins have evolved much faster than RecA. Among these, there is a group of proteins that has evolved between 4.0 and 4.5 times faster than RecA and that includes RecB, RecD, RecX, AddA, AddB, YqgF, and RecO. Because RecD is divided in two groups, these data only include the RecD proteins that are in the group of genomes containing RecBC (i.e., RecD1).
Figure 5 Regression Lines of the Substitution Rates of the Recombination Proteins Plotted against the Substitution Rates of RecA
RecA is the slowest and most ubiquitous of these proteins and its substitution rates are the x-axis of the plot (dashed lines indicate the RecA identity line). The regression was forced to pass through zero at intercept, and the slopes of the lines indicate the relative rapidity of the protein's evolution relative to that of RecA (varies from 1.3 for RuvB to 4.4 to RecB); the points associated with mollicutes were removed because we found these to evolve proportionally faster in RecA than in the other proteins.
The proteins of the RecFOR pathway have a peculiar evolutionary pattern. In addition to being present with very different frequency, with RecF being more frequently absent than RecR or RecO, they also show remarkably different substitution rates, with high conservation for RecR, lower conservation for RecF, and among the lowest conservation for RecO (Figure 5). This may be the result of the double participation of RecR in interactions with RecO and RecF, which would increase the constraints on its evolution. The crystal structure of the D. radiodurans RecR protein reveals the existence of a ring-shaped tetramer, theoretically able to encircle dsDNA [76]. This particular clamp-like structure may also have contributed to the high level of conservation of the protein.
It's interesting to note that among the fastest-evolving proteins, some are nearly ubiquitous (RecD and YqgF), and some are much rarer (RecB and AddAB). This suggests that few proteins have been missed in the analysis as a result of excessive sequence divergence. We made a set of simulations to assess this problem more precisely. We allowed protein sequences to evolve according to the evolutionary model of RecA, but at a different relative rates (see Materials and Methods). This analysis showed that only proteins evolving more than four times faster than RecA are expected to be missed in our similarity searches at this evolutionary distance and using our 40% similarity criterion (Figure 6). Even for proteins evolving 5.5 times faster than RecA, in none of our 100 simulations would we miss more than six orthologs. These orthologs were systematically in the fast-evolving mollicutes clade. Naturally, this is an oversimplification of the evolution of proteins, because proteins evolve in a changing context, and this may change their relative rates of evolution. In addition, these analyses do not take into account that insertions and deletions may be more frequent in some proteins than in others. Yet they indicate that few homologous genes are expected to have been lost in the present analysis as a result of excessive sequence divergence.
Figure 6 Results of the Simulations of Protein Evolution following the Phylogenetic Tree of RecA Using the JTT+Γ Model with Eight Classes and an α of 0.59
We used Seq-Gen to evolve protein sequences with regularly spaced scaling factors (100 experiments for each scaling factor) and analyzed for each experience which sequences showed less than 40% similarity.
Conclusion
The presynaptic role of RecBCD and RecFOR and the branch migration activity of RuvAB and RecG suggest functional redundancy, whereas, in contrast, the patterns of co-occurrence of these systems agree with the experimental works indicating complementary, and not redundant, roles for these proteins. Interestingly, this work also indicates that the RecFOR pathway may be more conspicuously important among bacteria than RecBCD, as it is significantly more frequent. RecR is the most conserved of the three proteins, and understanding how recombination is promoted in the organisms that encode a RecR homolog but do not have RecF or RecO would help understand the functioning of these recombination mediator proteins. The associations of recR and recF with genes involved in replication are often conserved, suggesting that the close association between replication and recombination observed in E. coli is common to most bacteria.
A central tenet of current genomic studies is the possibility of associating gene content with phenotype variation. Because the abundance of repeats in genomes correlates well with rearrangement rates and with the capacity of generating genetic variation [8,9], and because repeats are cause and consequence of recombination processes, one could expect an association between the repertoire of recombination genes and the number of repeats. We were unable to observe such a correlation. Indeed, except for genomes lacking RecA and resolvases (which are stable, have few repeats, and possibly lack homologous recombination), bacteria known to recombine frequently may either have a complete repertoire of known recombination genes or lack a substantial part of it. A striking example of the latter is provided by H. pylori [77], which is highly recombinogenic, although it lacks most presynaptic proteins and has antirecombinants such as UvrD and MutS2. In addition, at the intraspecies level, the differences in the population structure do not correlate with the genome content in recombination proteins. For example, serogroup A of N. meningitidis is mostly clonal, contrary to the majority of the others [78]. However, we found that both serotypes A [79] and B [80] have the same almost complete repertoire of homologous recombination proteins. Hence, associations between stability of a genome and the lack of some recombination proteins, as was proposed for Bifidobacterium longum [81] and Corynebacterium species [38], must be viewed with exceptional care before experimental confirmation.
The reasons for this lack of simple association between genotype and phenotype are probably multiple. Orthologs do not necessarily have the same exact functionalities and are likely to have different levels of activity. For example, presynaptic systems may be less necessary if the affinity of RecA for ssDNA is higher. The frequency of recombination events may also depend on the implication of recombination proteins in different cellular processes. For example, the coupling of recombination and replication may depend on the replication machinery and on the frequency of replication arrest. Specific genetic regulatory systems may also lead to different rates of recombination. For example, the onset of competence may be differently related in various organisms with cell growth and with the level of expression of recombination enzymes. Also, equivalent cellular processes may be associated with different enzymatic systems. For example, in neisserial species and E. coli, transformation-associated recombination takes place through the RecBCD pathway, whereas in B. subtilis, chromosomal transformation decreases 2.5-fold in a recO mutant [82], and in streptococci, AddAB is not involved in chromosome transformation [83], possibly because in competent firmicutes only ssDNA enters the cell. In contrast, in the competent Helicobacter and Campylobacter species, all these genes but RecR are absent. One could also expect that recombination activity is also constrained by ecological factors. Endosymbionts live in very protected environments, and this, associated with reductive genome evolution, has led to the loss of recombination functions [36,37]. However, apart from this case, we could not find any other obvious association between lifestyle and the presence or absence of recombination proteins, which once again is in agreement with the inherent housekeeping role of homologous recombination.
This housekeeping role of homologous recombination is probably also why we found little evidence of horizontal transfer among these genes. Genes implicated in the generation of genetic variation tend to be frequently horizontally transferred [84,85], but not housekeeping genes involved in managing genetic information [86]. Interestingly, multilocus sequence data also indicate that RecA rarely recombines among strains of the same species [87,88]. This does not mean that horizontal transfer is altogether absent. Such events are the most parsimonious explanation for the existence of some analogous replacements, such as AddAB among proteobacteria or RuvC in Thermoanaerobacter tengcongensis. They are also probably responsible for the sporadic occurrence of NHEJ in different phylogenetic groups. In addition, given the frequency of prophage sequences in bacterial genomes [89], and the many phage-encoded recombination systems, recombination genes of known phage origin, which have not been included in this study, may also play a role in the variations of recombination mechanisms.
Our study defines a core of recombination genes coding for proteins nearly ubiquitous in bacterial species. These include the genes that encode RecA (which has a homolog among eukaryotes), RuvAB, RecR, RuvC/RecU, and to a minor extent RecG, RecN, RecJ, and PriA. These genes are present in nearly all bacterial groups and show little horizontal transfer. This justifies the use of such proteins as phylogenetic markers [43]. Their widespread distribution demonstrates their importance in bacteria and justifies the emphasis on their detailed biochemical and functional study.
Materials and Methods
Data.
We analyzed the genomes of 117 different bacterial species (see Figure 2), taken from GenBank Genomes (ftp://ftp.ncbi.nih.gov/genomes/Bacteria/). The list of proteins related to homologous recombination was taken from the literature [13,24] and included RecA, RecB, RecC, RecD, RecF, RecG, RecJ, RecN, RecO, RecQ, RecR, RecU, RecX, RuvA, RuvB, RuvC, AddA (RexA), AddB (RexB), and PriA. Their function is summarized in Figure 1. Proteins such as RecE, RecT, and RusA were not analyzed because they were found to be very rare in bacterial genomes and are associated with prophages [13]. In addition, we included the antirecombination proteins SbcB, SbcC, SbcD, MutS1, MutS2, MutL, and UvrD; the putative resolvase YqgF [47]; and the Ku and ligase genes responsible for nonhomologous end joining in some bacteria [52].
Assignment of orthology.
One should note that many recombination genes belong to large protein families, such as helicases [90] or nucleases [47]. Hence, simple sequence similarity is not an indication of orthology. Assignment of orthology followed an automated step and then manual curation. The automatic method was the following. We started from the protein in E. coli (except for AddAB, MutS2, Ku, and RecU, where we started from B. subtilis) and searched for orthology in all other genomes. Genes were regarded as potential orthologs if they were bidirectional best hits with at least 40% similarity in sequence and their sequences were less than 30% different in length. The alignments were done using an adapted version of the Neddleman-Wusch algorithm (global alignment), in which the nonaligned edges of the largest sequence are not penalized [91], using the matrix BLOSUM60 and typical gap penalties. For comparison, we also made FASTA searches, because they allow for the detection of more local similarities [92]. Then we took the less similar protein hit, respecting the previously cited conditions as a query, and relaunched the analysis on the entire set of genomes with the same parameters. The proteins resulting from the intersection of these lists were temporarily regarded as bona fide orthologs. The other proteins were put together with the ones showing significant FASTA hits (E < 10−5) on the other genomes, as well as the ones originally annotated as orthologs (but not respecting the above conditions). We then searched for significant motifs in this set of proteins, using the InterPro database (http://www.ebi.ac.uk/interpro/) and visually analyzed and corrected multiple alignments. The proteins showing alignments with more than 40% similarity with bona fide orthologs were kept. When the alignments were within the range of 37%–40% similarity and did not show excessive gaps, and the proteins respected the 30% difference in length criterion or had significant InterPro motifs, the proteins were classed as putative. The bona fide orthologs were then aligned and phylogenetic distances computed as described below. The final list of “bona fide orthologs” took into account not only sequence similarity searches but also the phylogenetic information and colocalization data, as recommended [93].
Phylogenetic analyses and simulations of protein evolution.
Orthologs were aligned using ClustalW [94] and checked with Seaview [95]. Phylogenetic distances between the orthologous proteins were computed using Tree-Puzzle [74], with the JTT+Γ model with eight classes. For this analysis, and because we wanted to assess evolutionary rates, we removed only the regions with extended gaps from the multiple alignments. Phylogenetic trees were built using the same model with Phyml [96]. We used Seq-Gen [97] to generate 1,000 proteins with 1,000 residues, having the average amino acid composition of the JTT substitution matrix. The sequences were made to evolve along the RecA phylogenetic tree (which is largely congruent with the 16S rDNA tree [42]), using scaling factors in the range 0.5 to 6 (the fastest protein was found to evolve at less than 4.5 times the rate of RecA), and with the evolutionary model used to build the RecA tree. Each time, we used the evolved sequences to make global alignments and compute the similarity. For each experience, we counted how many genes had more than 40% and more than 37% similarity with the E. coli gene. This allowed the assessment of the number of orthologs that may be missed by the automatic similarity search part of the methods as a result of excessive sequence divergence.
Colocalization analysis.
Two genes were considered to closely co-occur if they were fewer than five genes away in a genome. A third gene is in close co-occurrence with the latter two if it is less than five genes away from at least one of the two genes. One should note that the average operon in E. coli and B. subtilis has fewer than five genes [98]. We started by analyzing the co-occurrence of the orthologs of the E. coli recombination genes. Then we did the same with the orthologs of B. subtilis genes that have no orthologue in E. coli. Finally, we analyzed particular cases described in the literature: the occurrence of recF in the dnaA region [70] and the co-occurrence of recR with dnaX [71], and recX with recA [69].
Meriem El Karoui, Ivan Matic, Vincent Daubin, and two anonymous reviewers provided important comments and criticisms on this manuscript. Alain Blanchard and Pascal Sirand-Pugnet provided important input and thoughts on recombination in mollicutes.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. EPCR conceived and designed the experiments. EPCR and EC performed the experiments. EPCR, EC, and BM analyzed the data. EPCR and BM wrote the paper.
Abbreviations
dsDNAdouble-stranded DNA
NHEJnonhomologous end-joining mechanism
OYOnion Yellows
ssDNAsingle-stranded DNA
==== Refs
References
Kuzminov A 1999 Recombinational repair of DNA damage in Escherichia coli and bacteriophage lambda Microbiol Mol Biol Rev 63 751 813 10585965
Michel B Grompone G Florès MJ Bidnenko V 2004 Multiple pathways process stalled replication forks Proc Natl Acad Sci U S A 101 12783 12788 15328417
Smith GR 1991 Conjugational recombination in E. coli: Myths and mechanisms Cell 64 19 27 1986865
Lorenz MG Wackernagel W 1994 Bacterial gene transfer by natural genetic transformation in the environment Microbiol Rev 58 563 602 7968924
Feil EJ 2004 Small change: Keeping pace with microevolution Nat Rev Microbiol 2 483 495 15152204
Otto SP Michalakis Y 1998 The evolution of recombination in changing environments Trends Ecol Evol 13 145 151 21238235
Hughes D 1999 Impact of homologous recombination on genome organization and stability Charlebois RL Organization of the prokaryotic genome Washington (DC) ASM Press 109 128
Rocha EPC 2004 Order and disorder in bacterial genomes Curr Opin Microbiol 7 519 527 15451508
Rocha EPC 2003 DNA repeats lead to the accelerated loss of gene order in bacteria Trends Genet 19 600 604 14585609
Finlay BB Falkow S 1997 Common themes in microbial pathogenicity revisited Microbiol Mol Biol Rev 61 136 169 9184008
Mehr IJ Seifert HS 1998 Differential roles of homologous recombination pathways in Neisseria gonorrhoeae pilin antigenic variation, DNA transformation, and DNA repair Mol Microbiol 30 697 710 10094619
Rocha EPC Blanchard A 2002 Genomic repeats, genome plasticity and the dynamics of Mycoplasma evolution Nucleic Acids Res 30 2031 2042 11972343
Kowalczykowski SC Dixon DA Eggleston AK Lauder SD Rehrauer WM 1994 Biochemistry of homologous recombination in Escherichia coli
Microbiol Rev 58 401 465 7968921
Cox MM 2002 Historical overview: Searching for replication help in all the rec places Proc Natl Acad Sci U S A 98 8173 8180
Konforti BB Davis RW 1990 The preference for a 3′ homologous end is intrinsic to RecA-promoted strand exchange J Biol Chem 265 6916 6920 2182632
Singleton MR Dillingham MS Gaudier M Kowalczykowski SC Wigley DB 2004 Crystal structure of RecBCD enzyme reveals a machine for processing DNA breaks Nature 432 187 193 15538360
Kowalczykowski SC 2000 Initiation of genetic recombination and recombination-dependent replication Trends Biochem Sci 25 156 165 10754547
Webb BL Cox MM Inman RB 1997 Recombinational DNA repair: The RecF and RecR proteins limit the extension of RecA filaments beyond single-strand DNA gaps Cell 91 347 356 9363943
Morimatsu K Kowalczykowski SC 2003 RecFOR proteins load RecA protein onto gapped DNA to accelerate DNA strand exchange: A universal step of recombinational repair Mol Cell 11 1337 1347 12769856
Lusetti SL Cox MM 2002 The bacterial RecA protein and the recombinational DNA repair of stalled replication forks Annu Rev Biochem 2002 71 100
Goodman MF Tippin B 2000 Sloppier copier DNA polymerases involved in genome repair Curr Opin Genet Dev 10 162 168 10753775
Marians KJ 2000 PriA-directed replication fork restart in Escherichia coli
Trends Biochem Sci 25 185 189 10754552
Polard P Marsin S McGovern S Velten M Wigley DB 2002 Restart of DNA replication in Gram-positive bacteria: Functional characterisation of the Bacillus subtilis PriA initiator Nucleic Acids Res 30 1593 1605 11917020
Dubnau D Lovett CM 2002 Transformation and recombination Sonenshein AL Hoch JA Losick R Bacillus subtilis and its closest relatives Washington (DC) ASM Press 453 471
el Karoui M Ehrlich D Gruss A 1998 Identification of the lactococcal exonuclease/recombinase and its modulation by the putative Chi sequence Proc Natl Acad Sci U S A 95 626 631 9435243
El Karoui M Biaudet V Schbath S Gruss A 1999 Characteristics of Chi distribution on different bacterial genomes Res Microbiol 150 579 587 10672998
Ayora S Carrasco B Doncel E Lurz R Alonso JC 2004
Bacillus subtilis RecU protein cleaves Holliday junctions and anneals single-stranded DNA Proc Natl Acad Sci U S A 101 452 457 14701911
Feinstein SI Low KB 1986 Hyper-recombining recipient strains in bacterial conjugation Genetics 113 13 33 3519362
Vulic M Dionisio F Taddei F Radman M 1997 Molecular keys to speciation: DNA polymorphism and the control of genetic exchange in enterobacteria Proc Natl Acad Sci U S A 94 9763 9767 9275198
Morel P Hejna JA Ehrlich SD Cassuto E 1993 Antipairing and strand transferase activities of E. coli helicase II (UvrD) Nucleic Acids Res 21 3205 3209 8341594
Veaute X Delmas S Selva M Jeusset J Le Cam E 2005 UvrD helicase, unlike Rep helicase, dismantles RecA nucleoprotein filaments in Escherichia coli
EMBO J 24 180 189 15565170
Mendonca VM Matson SW 1995 Genetic analysis of delta helD and delta uvrD mutations in combination with other genes in the RecF recombination pathway in Escherichia coli: Suppression of a ruvB mutation by a uvrD deletion Genetics 141 443 452 8647383
Gibson FP Leach DR Lloyd RG 1992 Identification of sbcD mutations as cosuppressors of recBC that allow propagation of DNA palindromes in Escherichia coli K-12 J Bacteriol 174 1222 1228 1531222
Eisen JA Hanawalt PC 1999 A phylogenomic study of DNA repair genes, proteins, and processes Mutat Res 435 171 213 10606811
Fraser CM Norris SJ Weinstock GM White O Sutton GG 1998 Complete genome sequence of Treponema pallidum the syphilis spirochete Science 281 375 388 9665876
Silva FJ Latorre A Moya A 2003 Why are the genomes of endosymbiotic bacteria so stable? Trends Genet 19 176 180 12683967
Dale C Wang B Moran N Ochman H 2003 Loss of DNA recombinational repair enzymes in the initial stages of genome degeneration Mol Biol Evol 20 1188 1194 12777518
Nakamura Y Nishio Y Ikeo K Gojobori T 2003 The genome stability in Corynebacterium species due to lack of the recombinational repair system Gene 317 149 155 14604803
Sicheritz-Ponten T Andersson SG 2001 A phylogenomic approach to microbial evolution Nucleic Acids Res 29 545 552 11139625
Ingleston SM Dickman MJ Grasby JA Hornby DP Sharples GJ 2002 Holliday junction binding and processing by the RuvA protein of Mycoplasma pneumoniae
Eur J Biochem 269 1525 1533 11874468
Della M Palmbos PL Tseng HM Tonkin LM Daley JM 2004 Mycobacterial Ku and ligase proteins constitute a two-component NHEJ repair machine Science 306 683 685 15499016
Eisen JA 1995 The RecA protein as a model molecule for molecular systematic studies of bacteria: Comparison of trees of RecAs and 16S rRNAs from the same species J Mol Evol 41 1105 1123 8587109
Santos SR Ochman H 2004 Identification and phylogenetic sorting of bacterial lineages with universally conserved genes and proteins Environ Microbiol 6 754 759 15186354
Shigenobu S Watanabe H Hattori M Sakaki Y Ishikawa H 2000 Genome sequence of the endocellular bacterial symbiont of aphids Buchnera sp. APS Nature 407 81 86 10993077
Tamas I Klasson L Canback B Naslund AK Eriksson AS 2002 50 million years of genomic stasis in endosymbiotic bacteria Science 296 2376 2379 12089438
Oshima K Kakizawa S Nishigawa H Jung HY Wei W 2004 Reductive evolution suggested from the complete genome sequence of a plant-pathogenic phytoplasma Nat Genet 36 27 29 14661021
Aravind L Makarova KS Koonin EV 2000 Survey and summary: Holliday junction resolvases and related nucleases: Identification of new families, phyletic distribution, and evolutionary trajectories Nucleic Acids Res 28 3417 3432 10982859
Sharples GJ Ingleston SM Lloyd RG 1999 Holliday junction processing in bacteria: Insights from the evolutionary conservation of RuvABC, RecG, and RusA J Bacteriol 181 5543 5550 10482492
Zuniga-Castillo J Romero D Martinez-Salazar JM 2004 The recombination genes addAB are not restricted to gram-positive bacteria: Genetic analysis of the recombination initiation enzymes RecF and AddAB in Rhizobium etli
J Bacteriol 186 7905 7913 15547262
Singh S Folkers GE Bonvin AM Boelens R Wechselberger R 2002 Solution structure and DNA-binding properties of the C-terminal domain of UvrC from E. coli
EMBO J 21 6257 6266 12426397
Wang J Julin DA 2004 DNA helicase activity of the RecD protein from Deinococcus radiodurans
J Biol Chem 279 52024 52032 15466873
Weller GR Kysela B Roy R Tonkin LM Scanlan E 2002 Identification of a DNA nonhomologous end-joining complex in bacteria Science 297 1686 1689 12215643
Aravind L Koonin EV 2001 Prokaryotic homologs of the eukaryotic DNA-end-binding protein Ku, novel domains in the Ku protein and prediction of a prokaryotic double-strand break repair system Genome Res 11 1365 1374 11483577
Doherty AJ Jackson SP Weller GR 2001 Identification of bacterial homologues of the Ku DNA repair proteins FEBS Lett 500 186 188 11445083
Krejci L Chen L Van Komen S Sung P Tomkinson A 2003 Mending the break: Two DNA double-strand break repair machines in eukaryotes Prog Nucleic Acid Res Mol Biol 74 159 201 14510076
Lovett ST Clark AJ 1984 Genetic analysis of the recJ gene of Escherichia coli K-12 J Bacteriol 157 190 196 6317649
Mendonca VM Klepin HD Matson SW 1995 DNA helicases in recombination and repair: Construction of a delta uvrD delta helD delta recQ mutant deficient in recombination and repair J Bacteriol 177 1326 1335 7868608
Wojciechowski MF Hoelzer MA Michod RE 1989 DNA repair and the evolution of transformation in Bacillus subtilis . II. Role of inducible repair Genetics 121 411 422 2497048
Falush D Kraft C Taylor NS Correa P Fox JG 2001 Recombination and mutation during long-term gastric colonization by Helicobacter pylori: Estimates of clock rates, recombination size, and minimal age Proc Natl Acad Sci U S A 98 15056 15061 11742075
Pinto AV Mathieu A Marsin S Veaute X Ielpi L 2005 Suppression of homologous and homeologous recombination by the bacterial MutS2 protein Mol Cell 17 113 120 15629722
Aras RA Kang J Tschumi AI Harasaki Y Blaser MJ 2003 Extensive repetitive DNA facilitates prokaryotic genome plasticity Proc Natl Acad Sci U S A 100 13579 13584 14593200
Chen CW Huang CH Lee HH Tsai HH Kirby R 2002 Once the circle has been broken: Dynamics and evolution of Streptomyces chromosomes Trends Genet 18 522 529 12350342
Kowalczykowski SC 1991 Biochemical and biological function of Escherichia coli RecA protein: Behavior of mutant RecA proteins Biochimie 73 289 304 1883888
Aylon Y Kupiec M 2004 New insights into the mechanism of homologous recombination in yeast Mutat Res 566 231 248 15082239
Hershberg R Yeger-Lotem E Margalit H 2005 Chromosomal organization is shaped by the transcription regulatory network Trends Genet 21 138 142 15734572
Stohl EA Seifert HS 2001 The recX gene potentiates homologous recombination in Neisseria gonorrhoeae
Mol Microbiol 40 1301 1310 11442829
Drees JC Lusetti SL Chitteni-Pattu S Inman RB Cox MM 2004 A RecA filament capping mechanism for RecX protein Mol Cell 15 789 798 15350222
Lusetti SL Drees JC Stohl EA Seifert HS Cox MM 2004 The DinI and RecX proteins are competing modulators of RecA function J Biol Chem 279 55073 55079 15489505
De Mot R Schoofs G Vanderleyden J 1994 A putative regulatory gene downstream of recA is conserved in gram-negative and gram-positive bacteria Nucleic Acids Res 22 1313 1314 8165147
Yoshikawa H Ogasawara N 1991 Structure and function of DnaA and the DnaA-box in eubacteria: Evolutionary relationships of bacterial replication origins Mol Microbiol 5 2589 2597 1779750
Chen KS Saxena P Walker JR 1993 Expression of the Escherichia coli dnaX gene J Bacteriol 175 6663 6670 7691799
Courcelle J Carswell-Crumpton C Hanawalt PC 1997
recF and recR are required for the resumption of replication at DNA replication forks in Escherichia coli
Proc Natl Acad Sci U S A 94 3714 3719 9108043
Chow KH Courcelle J 2004 RecO acts with RecF and RecR to protect and maintain replication forks blocked by UV-induced DNA damage in Escherichia coli
J Biol Chem 279 3492 3496 14625283
Schmidt HA Strimmer K Vingron M von Haeseler A 2002 TREE-PUZZLE: Maximum likelihood phylogenetic analysis using quartets and parallel computing Bioinformatics 18 502 504 11934758
Novichkov PS Omelchenko MV Gelfand MS Mironov AA Wolf YI 2004 Genome-wide molecular clock and horizontal gene transfer in bacterial evolution J Bacteriol 186 6575 6585 15375139
Lee BI Kim KH Park SJ Eom SH Song HK 2004 Ring-shaped architecture of RecR: Implications for its role in homologous recombinational DNA repair EMBO J 23 2029 2038 15116069
Tomb JF White O Kerlavage AR Clayton RA Sutton GG 1997 The complete genome sequence of the gastric pathogen Helicobacter pylori
Nature 388 539 547 9252185
Bart A Barnabe C Achtman M Dankert J van der Ende A 2001 The population structure of Neisseria meningitidis serogroup A fits the predictions for clonality Infect Genet Evol 1 117 122 12798026
Parkhill J Achtman M James KD Bentley SD Churcher C 2000 Complete DNA sequence of a serogroup A strain of Neisseria meningitidis Z2491
Nature 404 502 506 10761919
Tettelin H Saunders NJ Heidelberg J Jeffries AC Nelson KE 2000 Complete genome sequence of Neisseria meningitidis serogroup B strain MC58
Science 287 1809 1815 10710307
Schell MA Karmirantzou M Snel B Vilanova D Berger B 2002 The genome sequence of Bifidobacterium longum reflects its adaptation to the human gastrointestinal tract Proc Natl Acad Sci U S A 99 14422 14427 12381787
Fernandez S Kobayashi Y Ogasawara N Alonso JC 1999 Analysis of the Bacillus subtilis recO gene: RecO forms part of the RecFLOR function Mol Gen Genet 261 567 573 10323239
Halpern D Gruss A Claverys JP El-Karoui M 2004 rexAB mutants in Streptococcus pneumoniae
Microbiology 150 2409 2414 15256582
Kobayashi I 2001 Behavior of restriction-modification systems as selfish mobile elements and their impact on genome evolution Nucleic Acids Res 29 3742 3756 11557807
Denamur E Lecointre G Darlu P Tenaillon O Acquaviva C 2000 Evolutionary implications of the frequent horizontal transfer of mismatch repair genes Cell 103 711 721 11114328
Rivera MC Rain R Moore JE Lake JA 1998 Genomic evidence for two functionally distinct gene classes Proc Natl Acad Sci U S A 95 6239 6244 9600949
Feil E Zhou J Maynard Smith J Spratt BG 1996 A comparison of the nucleotide sequences of the adk and recA genes of pathogenic and commensal Neisseria species: Evidence for extensive interspecies recombination within adk
J Mol Evol 43 631 640 8995060
Tu ZC Dewhirst FE Blaser MJ 2001 Evidence that the Campylobacter fetus sap locus is an ancient genomic constituent with origins before mammals and reptiles diverged Infect Immun 69 2237 2244 11254579
Casjens S 2003 Prophages and bacterial genomics: What have we learned so far? Mol Microbiol 49 277 300 12886937
Tuteja N Tuteja R 2004 Unraveling DNA helicases. Motif, structure, mechanism, and function Eur J Biochem 271 1849 1863 15128295
Erickson BW Sellers PH 1983 Recognition of patterns in genetic sequences Sankoff D Kruskal JB Time warps, string edits, and macromolecules: The theory and practice of sequence comparison Boston Addison-Wesley 55 91
Pearson WR 1990 Rapid and sensitive sequence comparison with FASTP and FASTA Methods Enzymol 183 63 98 2156132
Eisen JA 1998 Phylogenomics: Improving functional predictions for uncharacterized genes by evolutionary analysis Genome Res 8 163 167 9521918
Thomson JD Higgins DG Gibson TJ 1994 Clustal W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice Nucleic Acids Res 22 4673 4680 7984417
Galtier N Gouy M Gautier C 1996 SEAVIEW and PHYLO_WIN: Two graphic tools for sequence alignment and molecular phylogeny Comput Appl Biosci 12 543 548 9021275
Guindon S Gascuel O 2003 A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood Syst Biol 52 696 704 14530136
Rambaut A Grassly NC 1997 Seq-Gen: An application for the Monte Carlo simulation of DNA sequence evolution along phylogenetic trees Comput Appl Biosci 13 235 238 9183526
Zheng Y Szustakowski JD Fortnow L Roberts RJ Kasif S 2002 Computational identification of operons in microbial genomes Genome Res 12 1221 1230 12176930
Brochier C Bapteste E Moreira D Philippe H 2002 Eubacterial phylogeny based on translational apparatus proteins Trends Genet 18 1 5 11750686
Glockner FO Kube M Bauer M Teeling H Lombardot T 2003 Complete genome sequence of the marine planctomycete Pirellula sp. strain 1 Proc Natl Acad Sci U S A 100 8298 8303 12835416
Felsenstein J 1993 PHYLIP. Phylogeny Inference Package, version 3.6a [computer program] Seattle University of Washington Department of Genetics
Dybvig K Woodard A 1992 Construction of recA mutants of Acholeplasma laidlawii by insertional inactivation with a homologous DNA fragment Plasmid 28 262 266 1461941
Sogaard IZ Boesen T Mygind T Melkova R Birkelund S 2002 Recombination in Mycoplasma hominis
Infect Genet Evol 1 277 285 12798006
Cordova CM Lartigue C Sirand-Pugnet P Renaudin J Cunha RA 2002 Identification of the origin of replication of the Mycoplasma pulmonis chromosome and its use in oriC replicative plasmids J Bacteriol 184 5426 5435 12218031
Kuipers EJ Israel DA Kusters JG Blaser MJ 1998 Evidence for a conjugation-like mechanism of DNA transfer in Helicobacter pylori
J Bacteriol 180 2901 2905 9603879
Suerbaum S Maynard-Smith J Bapumia K Morelli G Smith NH 1998 Free recombination within Helicobacter pylori
Proc Natl Acad Sci U S A 95 12619 12624 9770535
de Jonge R Bakker D van Vliet AH Kuipers EJ Vandenbroucke-Grauls CM 2003 Direct random insertion mutagenesis of Helicobacter pylori
J Microbiol Methods 52 93 100 12401231
Wilson DL Bell JA Young VB Wilder SR Mansfield LS 2003 Variation of the natural transformation frequency of Campylobacter jejuni in liquid shake culture Microbiology 149 3603 3615 14663092
Schouls LM Reulen S Duim B Wagenaar JA Willems RJ 2003 Comparative genotyping of Campylobacter jejuni by amplified fragment length polymorphism, multilocus sequence typing, and short repeat sequencing: Strain diversity, host range, and recombination J Clin Microbiol 41 15 26 12517820
Boer P Wagenaar JA Achterberg RP Putten JP Schouls LM 2002 Generation of Campylobacter jejuni genetic diversity in vivo Mol Microbiol 44 351 359 11972775
Fu R Voordouw G 1997 Targeted gene-replacement mutagenesis of dcrA, encoding an oxygen sensor of the sulfate-reducing bacterium Desulfovibrio vulgaris Hildenborough Microbiology 143 1815 1826 9202456
Bertolla F Van Gijsegem F Nesme X Simonet P 1997 Conditions for natural transformation of Ralstonia solanacearum
Appl Environ Microbiol 63 4965 4968 9406418
Cotter TW Thomashow MF 1992 A conjugation procedure for Bdellovibrio bacteriovorus and its use to identify DNA sequences that enhance the plaque-forming ability of a spontaneous host-independent mutant J Bacteriol 174 6011 6017 1400153
Hommes NG Sayavedra-Soto LA Arp DJ 1996 Mutagenesis of hydroxylamine oxidoreductase in Nitrosomonas europaea by transformation and recombination J Bacteriol 178 3710 3714 8682770
Cooper AJ Kalinowski AP Shoemaker NB Salyers AA 1997 Construction and characterization of a Bacteroides thetaiotaomicron recA mutant: Transfer of Bacteroides integrated conjugative elements is RecA independent J Bacteriol 179 6221 6227 9335266
Krafft T Gross R Kroger A 1995 The function of Wolinella succinogenes psr genes in electron transport with polysulphide as the terminal electron acceptor Eur J Biochem 230 601 606 7607234
Park SY Kim HK Yoo SK Oh TK Lee JK 2000 Characterization of glk, a gene coding for glucose kinase of Corynebacterium glutamicum
FEMS Microbiol Lett 188 209 215 10913707
|
16132081
|
PMC1193525
|
CC BY
|
2021-01-05 08:00:24
|
no
|
PLoS Genet. 2005 Aug 26; 1(2):e15
|
utf-8
|
PLoS Genet
| 2,005 |
10.1371/journal.pgen.0010015
|
oa_comm
|
==== Front
PLoS GenetPLoS GenetpgenplgeplosgenPLoS Genetics1553-73901553-7404Public Library of Science San Francisco, USA 10.1371/journal.pgen.001002705-PLGE-RV-0074R2plge-01-02-13ReviewEvolutionImmunologyGenetics/Comparative GenomicsHomo (Human)Mus (Mouse)PrimatesRattus (Rat)VertebratesComparative Genomics of Natural Killer Cell Receptor Gene Clusters ReviewKelley James Walter Lutz Trowsdale John *James Kelley and John Trowsdale are in the Department of Pathology, University of Cambridge, Cambridge, United Kingdom. Lutz Walter is in the Department of Primate Genetics, German Primate Center, Göttingen, Germany.
*To whom correspondence should be addressed. E-mail: [email protected] 2005 26 8 2005 1 2 e27Copyright: © 2005 Kelley 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.Many receptors on natural killer (NK) cells recognize major histocompatibility complex class I molecules in order to monitor unhealthy tissues, such as cells infected with viruses, and some tumors. Genes encoding families of NK receptors and related sequences are organized into two main clusters in humans: the natural killer complex on Chromosome 12p13.1, which encodes C-type lectin molecules, and the leukocyte receptor complex on Chromosome 19q13.4, which encodes immunoglobulin superfamily molecules. The composition of these gene clusters differs markedly between closely related species, providing evidence for rapid, lineage-specific expansions or contractions of sets of loci. The choice of NK receptor genes is polarized in the two species most studied, mouse and human. In mouse, the C-type lectin-related Ly49 gene family predominates. Conversely, the single Ly49 sequence is a pseudogene in humans, and the immunoglobulin superfamily KIR gene family is extensive. These different gene sets encode proteins that are comparable in function and genetic diversity, even though they have undergone species-specific expansions. Understanding the biological significance of this curious situation may be aided by studying which NK receptor genes are used in other vertebrates, especially in relation to species-specific differences in genes for major histocompatibility complex class I molecules.
Citation:Kelley J, Walter L, Trowsdale J (2005) Comparative genomics of natural killer cell receptor gene clusters. PLoS Genet 1(2): e27.
==== Body
Introduction
Natural killer (NK) cells destroy cells infected with certain viruses and other intracellular pathogens. They also influence immune responses through the release of cytokines. A key aspect of recognition of appropriate target cells is the ubiquitously expressed major histocompatibility complex (MHC) class I molecule, a ligand for which NK cells generally have multiple receptors [1]. All living jawed vertebrates studied to date have an adaptive immune system, which, among other features, rearranges T and B cell receptor gene segments and exploits antigen presentation provided by MHC molecules [2,3]. Evolutionarily intermediate species, which lacked these features, presumably became extinct through competition with other species possessing a viable adaptive immune system [4]. A recognition-based MHC system may have been pivotal to survival of gnathostome species [3]. NK receptor gene complexes are intimately associated, both genetically and functionally, with MHC recognition, and interactions of different combinations of NK receptors and MHC class I molecules may contribute significantly to selection and disease resistance [4,5].
NK cell receptors come in two forms: inhibitory and activating. Inhibitory receptors regulate NK actions by interrupting intracellular activation signals when MHC class I molecules are correctly expressed [6]. Activating receptors, some of which bind ligands other than MHC class I molecules, trigger NK responses to cells with viral, bacterial, or parasitic infections or to some tumor cells with downregulated MHC class I molecules [7]. The effector function of each receptor molecule is determined by the sequence of its transmembrane region and cytoplasmic tail [8]. Generally, inhibitory receptors possess an immunoreceptor tyrosine-based inhibitory motif (ITIM) in their cytoplasmic tails [6], which decreases activation [9]. Upon stimulation, the ITIM becomes tyrosine phosphorylated and associates with intracellular phosphatases such as Src homology 2 (SH2) domain–containing protein tyrosine phosphatase 1 (SHP1) or SHP2. SHP1 then dephosphorylates the actin cytoskeleton regulator, Vav, which blocks actin-dependent activation signals [9]. In contrast, activating receptors lack ITIMs in their cytoplasmic tails and often contain charged residues that facilitate association with adaptor molecules containing immunoreceptor tyrosine-based activation motifs (ITAMs) such as DAP12 [10,11]. When activating receptors are associated with ITAM-containing adaptor molecules, the adaptor molecules become tyrosine phosphorylated and bind to kinases, which then interact with Vav and Rac1. The subsequent molecular cascade leads to actin polymerization and, consequently, cytotoxicity and/or cytokine release [11]. The ITIM/ITAM paradigm of inhibition/activation is a feature shared by NK receptors in all species, but it does not apply to all receptors. The human KIR2DL4 and KIR3DL3 molecules, for example, have unconventional cytoplasmic tails, and their mode of action, inhibition or activation, has not been clearly established [12]. Examples of inhibitory, activating, and co-stimulatory NK receptors, as well as the proposed pathways they initiate after ligand binding, are illustrated in Figure 1.
Figure 1 Examples of Inhibitory and Activating NK Cell Receptors and the Consequences of Their Binding
This figure shows the association of NK receptors with MHC class I molecules and portions of the resulting signaling pathway, as described in the text. “Src Kinase” represents various kinases capable of participating in this interaction, such as Syk and ZAP70. SHP2 (PTPN11) is used in place of SHP1 (PTPN6) in some circumstances, such as associating with KIR2DL4. Symbols show the continuation of the pathway where arrows are not drawn. −, negatively charged residue; +, positively charged residue; AKT1, v-akt murine thymoma viral oncogene homolog 1 (also called RAC); DAG1, dystroglycan 1; FcɛRI-γ, receptor for Fc fragment of IgE, high affinity I, gamma polypeptide; GRB2, growth factor receptor-bound protein 2; INPP5D, inositol polyphosphate-5-phosphatase, 145kDa (also called SHIP); ITPKB, inositol 1,4,5-trisphosphate 3-kinase B (also called IP3KB); LAT, linker for activation of T cells; LCP2, lymphocyte cytosolic protein 2 (also called SLP76); P, phosphate group; PAK1, p21/Cdc42/Rac1-activated kinase 1 (STE20 homolog, yeast); PIK3CG, phosphoinositide-3-kinase, catalytic, gamma polypeptide (also called phosphatidylinositol 3-kinase); PLCG1, phospholipase C, gamma 1; PTEN, phosphatase and tensin homolog; PTPN6, protein tyrosine phosphatase, nonreceptor type 6 (also called SHP1); RAF, raf protein; RAS, rat sarcoma viral oncogene homolog; SH3BP2, SH3 domain–binding protein 2; SHC1, SHC (SH2 domain–containing) transforming protein 1; SOS1, son of sevenless homolog 1; VAV1, vav 1 oncogene; YINM, tyrosine-containing motif (YINM).
Genes that encode NK receptors are arranged in two main clusters: the leukocyte receptor complex (LRC) and the natural killer complex (NKC). The LRC encodes members of the immunoglobulin superfamily (IgSF), while the NKC encodes type II transmembrane, C-type lectin-like proteins [13]. The extent that receptors from each complex are expressed and utilized varies markedly among species. Comparing the genomic composition of NK receptor gene clusters in different species may provide clues to their evolution. The goal from these studies is to understand the selective forces, particularly in relation to disease, that have driven such extreme genomic differences. The diverse arrangements of genes encoding the MHC ligands of NK receptors [3] have to be taken into consideration, as NK receptors and their ligands coevolve.
Human NK Receptor Gene Complexes
The main NK receptors for MHC class I molecules in humans belong to the IgSF and are encoded in the LRC on Chromosome 19q13.4 [14]. Of the 45 genes in the LRC, the 30 IgSF receptors can be grouped into several related gene families based on gene organization, phylogeny, and structure [15]. These families include the killer cell immunoglobulin-like receptors (KIRs), leukocyte Ig-like receptors (LILRs; also called LIRs and ILTs), and the leukocyte-associated Ig-like receptors (LAIRs). The most centromeric end of the human LRC contains genes for the platelet glycoprotein VI (GP6), the natural cytotoxicity-triggering receptor 1 (NCR1; also called NKp46), and the receptor of the IgA Fc fragment (FCAR; also called CD89). These proteins are structurally similar to KIR genes but differ from other loci in the LRC by interacting with ligands other than MHC class I molecules [15]. The organization of the human LRC is shown in the comparative maps of Figure 2.
Figure 2 Comparative Genomics of Natural Killer Cell Receptor Complexes
This figure, showing LRCs and NKCs in various species, is not drawn to scale; however, the linear arrangement of genes is correct and is aligned vertically by homology within each region, where possible. Colors indicate genes related by gene organization, structure, and phylogeny. Gray indicates genes that are not considered NK receptors. White boxes mark pseudogenes. Slash marks represent large distances in the genomic sequence. Some non-NK receptor genes that are located within this region may not be represented in this figure. A question mark indicates that the gene has been mapped to the corresponding chromosome, but the specific chromosomal position is not known. An “X” indicates that the gene is not homologous with genes sharing its vertical alignment. The linear organization of these genes was taken from literature sources as described in the text and from NCBI's MapViewer [123]. Similar to in human, only one Ly49 gene has been found in baboon, orangutan, dog, cat, cow, and pig. However, the horse has multiple Ly49 genes, as does the rodent lineage [96]. The chicken contains two sequences similar to NKC-encoded genes within its MHC [102], again suggesting an evolutionary relationship between MHC and NK receptor genes. The chicken also possesses multiple genes for Ig domain–containing NK receptors: the CHIR genes [100]; however, the chicken arrangement is from data assembled from whole genome sequence and will be refined as more accurate annotation is applied. While the zebrafish contains NK receptors containing Ig or C-type lectin domains, the genomic organization does not resemble the mammalian pattern of clustering into two main regions [104].
The key LRC genes directing NK cell recognition of MHC class I are the KIRs. The KIR gene cluster exhibits marked differences in gene content and allelic polymorphism between individual haplotypes [16,17], with up to 17 KIR genes and pseudogenes arranged tandemly over approximately 150 kilobases [4]. KIR genes share high sequence similarity (most are approximately 97% identical), exhibit marked levels of polymorphism, and evolve rapidly [18]. This last characteristic may be facilitated by nonreciprocal crossovers in the tandemly arranged genes, occasionally generating hybrid loci or expansion/contraction of gene numbers [18,19]. The KIR gene family in the human/primate lineage appears to have expanded between 31 and 44 million years ago [13], as dated by the amplification of retroelements of the Alu S subfamily found in the KIR genomic sequence [17]. KIRs can possess either two (KIR2D) or three (KIR3D) extracellular Ig domains [15] and comprise long forms, which contain inhibitory ITIMs, or short forms, which lack ITIMs and are mostly activating [18]. KIR2DL4, however, is unique, having both putative inhibitory and activating properties [12]. In addition to possessing ITIMs in its cytoplasmic tail, KIR2DL4 has a charged arginine residue in the transmembrane region that allows association with the activating accessory protein FcɛRI-γ (FCER1G) [20].
Centromeric to the KIR family is another extensive IgSF gene family, the LILRs. This family consists of two inverted, duplicated clusters of six and seven loci, respectively [21]. The LILR genes show greater intergenic diversity, but less allelic variation than the KIR family [15]. Unlike KIR genes, there is little variation in the number of LILR genes on different haplotypes. Separating the LILR clusters are two LAIR loci, which sandwich members of the LRC-encoded novel gene (LENG) family. It has been proposed that the LAIR locus was duplicated and inverted with the LILR expansion, but the LRC-encoded novel genes are structurally unrelated to other Ig-like receptors in the LRC and have a different origin [15].
Additional genes related to those in the LRC are dispersed further centromeric of the LILR and KIR gene families in the “extended LRC” region. This region includes the receptor and transporter of the Fc fragment of IgG (FCGRT), which interestingly may have a common origin with MHC class I molecules [22], the sialic-acid-binding immunoglobulin-like lectin (SIGLEC) gene family [23], and the CD66-related gene family [24]—all of which are related to LILR and KIR sequences [13].
The other major cluster of NK receptor genes, the human NKC, is located on Chromosome 12p13.1 [25] and encodes mainly disulphide-linked dimeric, type II transmembrane molecules with homology to C-type lectins [26]. NKC-encoded genes have highly related genomic structures and are organized into distinct clusters of related genes [26]. Comparative maps of NKCs are shown in Figure 2.
The NKC contains a variety of C-type lectin genes, some of which are expressed specifically on NK cells. There is only one member of the Ly49 gene family, KLRA1 (also called Ly49L), in humans, as opposed to the multiple homologous genes encoding MHC class I ligands in rodents [27]. While KLRA1 is transcribed, a point mutation causes the production of a nonfunctional molecule [27]. Another gene family expanded in rodents but with only one human homolog is KLRB1A (also called NKRP1A). NKRP1A is expressed only on a subset of NK cells and T cells, as opposed to all NK cells in rodents [28].
Members of the NKG2 (KLRC) family of molecules dimerize with the linked, invariant CD94 molecule (KLRD1) on the cell surface, which as a partner chain provides the appropriate signaling motifs [29]. Some KLRC-encoded molecules, namely NKG2A (KLRC1) and NKG2C (KLRC2), signal as a result of binding to the nonclassical MHC class I molecule HLA-E [30]. This family encodes both inhibitory (NKG2A, NKG2B, and KLRL1) and activating (NKG2C, NKG2E [KLRC3], and NKG2H) receptors, some of which, such as NKG2B, NKG2E, and NKG2H, are products of alternative splicing [26]. When both inhibitory and activating NKG2 receptors are coexpressed on the cell surface, the inhibitory molecules appear to be functionally dominant, which may relate to the fact that they have a higher-affinity binding [31]. The NKG2 family also contains a unique receptor, NKG2F (KLRC4), that has a charged residue in the transmembrane region, an ITIM-like domain, and no C-type lectin–like domain [32]. The function of this molecule is not known.
NKG2D (KLRK1) shows limited sequence identity to other NKG2 molecules and is expressed on NK cells, T cells, and macrophages as a homodimer [26]. For activation, NKG2D signals by associating with DAP10, a molecule whose Src homology 2 domain recruits the p85 subunit of phosphatidylinositol 3-kinase (PI3K) [33,34]. Human NKG2D binds MHC class I chain-related protein (MIC) A, MICB, and the UL-16 binding protein (ULBP) family [35].
KLRF1, which is present in humans but not mice [26], stimulates NK cells upon cross-linking. The inhibitory-receptor-encoding KLRG1, also called mast cell function–associated antigen [36], is more centrally located in the human NKC than in that of the mouse and may have arisen by gene duplication and chromosomal inversion events [26]. Other NK surface molecules encoded in the human NKC include the activation-induced C-type lectin (AICL) [37], lectin-like transcript 1 (CLEC2D; also called OCIL and LLT1) [38], and CD69 [26]. Genes that encode products not found on NK cells, which are also located within the NKC, include alpha-2-macroglobulin (A2M) [26], oxidized low-density lipoprotein (lectin-like) receptor 1 (OLR1) [39], CLECSF12, CLEC1, and CLEC2 [40].
Nonhuman Primate NK Receptor Gene Complexes
KIR genes have diverged dramatically between different primate species, consistent with rapid, species-specific expansion of the gene family [41]. In chimpanzees, the seven KIR genes, of which only three (KIR2DL4, KIR2DL5, and KIR2DS4) are best reciprocal human orthologs, cover 106 kilobases [41,42]. In the gorilla, 11 KIR genes have been identified with two genes being orthologous to human KIRs [43]. Orangutans, which diverged earlier from humans in the primate lineage than chimpanzees and gorillas, also have a species-specific expanded KIR repertoire [44]. The rhesus macaque only has five KIR genes [45,46]. The African green monkey Chlorocebus sabaeus, another Old World primate more closely related to rhesus monkeys than to the apes, possesses multiple KIR genes, namely KIR3DL, KIR2DL4, KIR2DL5, and KIR3DH, a KIR form found in rhesus and African green monkeys but not apes, and a hybrid of KIR2DL5 and KIR3DH [47]. KIR2DL4 is the only orthologous KIR gene found in humans, chimpanzees, gorillas, rhesus macaques, and African green monkeys. Interestingly, not all gorillas appear to have KIR2DL4 [43], and it is nonfunctional as a receptor in orangutans [44]. Table 1 shows, where known, the differences in gene content of NK receptor genes among various species.
Table 1 Presence of NKC and LRC Encoded Genes in Different Species
aHUGO gene abbreviation.
bCommon, alternative protein name.
cShows activating (A), inhibitory (I), either (A/I), or co-stimulatory (C) function. An empty box indicates information is not available for that species. The number indicates the approximate number of genes present in that species for a particular gene family. Ψ indicates a pseudogene.
Asterisk indicates the presence of multiple KIR-ILT hybrid genes in horses [96].
These differences in KIR gene presence accord with differences in MHC class I gene content among primate species [48]. For example, HLA-C and its equivalent in other primate species differentiated from an HLA-B-like ancestor after divergence of the hominoid and monkey lineages [4,49]; therefore, it is only present in primate species more closely related to humans. While HLA-C is consistently present in humans and chimpanzees (the closest relative of humans), it is absent in approximately half of the orangutan haplotypes, and when it does occur in orangutans, it resembles an evolutionary intermediate of human HLA-C [4,44]. Furthermore, Popy-C, the orangutan HLA-C equivalent, provides only one of the two MHC-C motifs that are used to control human and chimpanzee NK cells [42,50]. Dimorphism at amino acid 80 defines these two groups, where MHC-C1 has Asn80 and MHC-C2 has Lys80. The receptors for the C1 and C2 groups are the inhibitory KIR2DL molecules: KIR2DL2 and KIR2DL3 interact with C1, while KIR2DL1 interacts with C2 at a higher affinity. From these primate evolutionary studies, we have learned that the weaker C1–KIR2DL2/3 interaction arose first, and the stronger C2–KIR2DL1 interaction arose later. Since both C1 and C2 allotypes are represented in all human populations, these molecules may have complementary functions that enable the strength of HLA-C-mediated inhibition to be varied. This situation has implications for MHC/KIR combinations in disease, as has been evidenced by several recent papers [51–55]. The association of combinations of C1/C2 types and KIR alleles with predisposition to preeclampsia suggests other selection mechanisms [56]. The reciprocal levels of HLA-C1/2 allotypes and their respective KIR ligands in different human populations is consistent with a balance of advantages and disadvantages in level of response, presumably depending on local parasites and environmental conditions.
LILR genes are present in all primates studied, and although the gene content of this family shows greater differences between primate species than is observed throughout the genome, LILR genes are less variable in gene number between haplotypes than the KIRs. In chimpanzee, the nine LILR genes border the KIR sequences and are arranged in two duplicated clusters similar to human [41,57]. Four of these genes are orthologs of human LILR sequences [57]. The rhesus macaque has five LILR sequences [41]. Other NKC genes known to be present in the chimpanzee and rhesus macaque are orthologs of FCAR and NCR1 [41].
There is limited information on NKC genes in primates. Members of the NKG2 family and CD94 have been found in chimpanzees [42], including orthologs for all members except KLRC2, for which there are two paralogs [58]. Rhesus monkeys possess NKG2A, NKG2B, NKG2C, NKG2D, and several splice variants [45]. In the orangutan, orthologs for human CD94, NKG2A, NKG2D, and NKG2F are present, along with a hybrid gene combining NKG2C and NKG2E midway between their expected syntenic positions [44]. Orangutans also possess a functional homolog of KLRA1 (Ly49L) [44], which is a pseudogene in humans [27,59]. A single Ly49 sequence, which appears to be functional, has also been reported in baboons [60].
Rodent NK Receptor Gene Complexes
The mouse LRC is located on Chromosome 7 [15], although it contains none of the KIR loci that form a cornerstone of MHC class I recognition in humans. However, two murine KIR-like sequences have been detected outside the LRC on the X chromosome [61,62]. There are conflicting reports on the function of one of these murine KIR-like sequences: one study found that Kir3dl1 (Kirl1) lacks an ITIM and any residue capable of binding to an activating adaptor molecule [61], while another reported the presence of two ITIMs in Kir3dl1 [62]. Surprisingly, the other KIR-like sequence, Kirl2, is selectively expressed in defined areas of the mouse brain [63]. The murine LRC contains orthologs of human GP6 [13], NCR1, RPS9, and LAIR1 [15]. The Pir gene family members (including Pira1 to Pira11 and Pirb), which share sequence identity with the human LILRs, are found between Ncr1 and Rps9, [64]. In addition to sequence similarity, the Pir genes are arranged into two clusters similar to the LILRs and encode products that may interact with certain MHC class I molecules [65]. Genes of the murine LRC broadly display syntenic homology with the human arrangement apart from the absence of KIR loci.
In the rat, the LRC is located on Chromosome 1 and includes an ortholog of murine Pirb, Ncr1 [15], and one KIR sequence (Kir3dl1) with a potential ITIM [62]. The rat also has Fcar, which, while also present in humans, was lost in the mouse lineage [66].
Orthologs of human NKC genes are reported in a syntenic region of murine Chromosome 6. The gene families are arranged in a similar order to the human NKC, except for expansions and interchanges of the C-type lectin related (Clr) (ortholog of human OCIL) and Nkrp1 (Klrb1) gene families [26]. Interestingly, the products of these two genetically intertwined families in mouse interact, suggesting potential functional advantages for their adjacent gene locations [67]. There are several Klrb1 (Nkrp1) genes in mouse, with Klrb1a, Klrb1c, and Klrb1f taking activating forms and Klrb1b and Klrb1d being inhibitory [26]. Members of the NKC-encoded C-type lectin related gene family have been found as ligands for Klrb1d and Klrb1f [67].
In the rat, the NKC is located on Chromosome 4. In fact, the first NKC gene discovered in any species was rat Nkrp1 [68,69]. Human, mouse, and rat all share orthologs of CD69 and KLRD1 (CD94); however, there are still many differences between human and rodent in the gene content of the NKC [70] (see Figure 2; Table 1).
In contrast to the single human Ly49 locus, the polymorphic Ly49 gene family in mice contains at least 16 genes and pseudogenes [71], named Ly49a to Ly49q, although the gene content can differ significantly in different mouse strains [26]. While framework Ly49 genes are present in all murine haplotypes, the Ly49 gene content between these framework genes varies at some loci [72], in a similar manner to what is observed in the human KIR family. These framework/strain-specific differences are shown in Figure 3.
Figure 3 Differences in Organization of KIR Genes in Human Haplotypes and Ly49 Genes in Various Mouse Strains
This figure is not drawn to scale, but it illustrates the presence of framework genes and the variable gene content in Ly49 genes in different mouse strains [72] and KIR sequences in human haplotypes. The vertical arrangement of genes within a family shows allelic relationships. White boxes indicate known pseudogenes. Please note that while many KIR haplotypes are possible, only a small selection has been shown here.
The Ly49 molecules have a variety of functions and expression patterns. Ly49a, Ly49c, Ly49g, and Ly49i contain ITIMs and act as NK inhibitory receptors when recognizing MHC class I molecules. Conversely, Ly49d and Ly49h exhibit activating properties. Ly49e functions in fetal NK cell development, when other Ly49 molecules are not present [26]. Ly49q is found on plasmacytoid dendritic cells and not NK cells [73,74].
In the rat, this family has expanded even further. The most recent report lists 19 functional rat Ly49 genes and 15 pseudogenes [75]. The rapid expansion of rat Ly49 genes appears to result from repeated tandem and block gene duplications, which occurred after species divergence from the mouse [76], and might be related to the higher number of MHC class I genes in the rat genome than in the mouse genome [77].
The Ly49 family in mouse parallels the human KIR both functionally and genetically [78]. Insight into the evolutionary drive behind gain or loss of Ly49 loci was provided by studying the relationship between Ly49 genes and murine cytomegalovirus (mCMV) [79–82]. When the Ly49h receptor in mice is not present, is blocked by monoclonal antibody, or has its pathway interrupted through mutation of the DAP12 molecule, uncontrolled viral replication of mCMV occurs [80,83], demonstrating that the function of this activating receptor is essential for mCMV resistance [26]. Ly49h interacts specifically with mCMV protein m157 to counteract inhibition imposed by the viral protein engaging Ly49i [84]—an example of the biological arms race [85].
The KLR gene family in mouse contains several members, including Klrc1, Klrc2, Klrc3, Klrd1, Klri1, and Klri2 [26,86]. Similar to recognition of HLA-E in humans [30], members of this family recognize the mouse HLA-E functional homolog H2-T23 [86]. Nkg2d binds minor histocompatibility molecule H60, the retinoic acid early transcript 1 (Rae-1) family, and mouse UL-16 binding protein–like transcript 1 (Mult1). Interestingly, the Mill gene family, the murine functional equivalent of human MIC genes, which are encoded in the human MHC class I region and bind NKG2D, are found near the murine LRC [87]. In mice, Nkg2d is proposed to have activating properties through association with Dap12 and Dap10 [34,88]. Human NKG2D cannot associate with DAP12 to produce an activating signal [89]. A direct ortholog of KLRF1 is not present in the mouse. The murine NKC also contains Klrg1 and Cd69, and genes other than those encoding NK receptors—genes such as alpha-2-macroglobulin, Clecsf12, and Clec2 [26]. Genes encoding three Klrc molecules, Nkg2d, CD94, Klrb1, and Klrh1 (an inhibitory receptor) have been reported in rat.
NK Receptor Gene Complexes in Other Species
Since the genomic organization of human and mouse class I receptors is so polarized, in terms of KIR or Ly49 gene content, respectively, it is informative to study other species. The bovine LRC, located on Chromosome 18, includes activating and inhibitory KIR sequences and an ortholog of NCR1 [90]. Sequence similarity comparisons between cows and primates indicate that the multigenic KIR families expanded independently in the two lineages [91]. Known genes in the bovine NKC include KLRK1, NKRP1 [92], KLRJ1 [91], and one Ly49 gene [93]. There is a CD69 transcript in cows as well, but its genetic location is not known [94].
Like humans, the single Ly49 sequence is recapitulated in cattle [93], domesticated cats, dogs, and pigs [95]. Like rodents [76], horses show multiple Ly49 genes: five that encode ITIMs and one with potentially activating properties [96]. The horse also has an expansion of a family of KIR-ILT hybrid genes [96]. These species with both multiple Ly49 and multiple KIR sequences indicate that the functions of the two sets of MHC class I ligands can be coordinated. The polarized arrangement in humans (all KIR) and mice (all Ly49) suggest that either (1) the existence of both sets within a single individual poses logistical problems, which have been solved by disabling one set, or (2) the human and mouse arrangements are outriders and in most other vertebrates the functions of both sets of genes synchronize well with each other. Clearly, the MHC class I/NK ligand gene arrangements of many other species need to be evaluated.
A gene complex encoding NK receptors in pig is located on Chromosome 5, with reports of direct orthologs for human CD69 and KLRK1 [97]. Chicken is a key species because of its “minimal essential” MHC [98]. A family of chicken Ig-like receptors (CHIRs) are related to Pirs and Fc receptors and are arranged in similar genomic clusters to human KIR and LILR genes [99]. Recent work has shown the presence of numerous CHIR genes in the chicken genome, with many receptors possessing both activating and inhibitory properties [100]. The multiple CHIR genes are thought to have emerged from a common ancestor with humans before the mammalian radiation and then expanded in a lineage-specific manner [100,101]. There are examples of chicken NK receptors, such as B-NK and N-lec, which contain C-type lectin domains, similar to the NKC-encoded NKRP1 and LLT1 in humans. It is interesting, though, that the genes encoding these proteins are located within the chicken MHC region, consistent with an ancient genetic relationship of these MHC and NK receptors [102].
The zebrafish, a teleost model organism, contains a cluster of putative activating and inhibitory NK receptors called the novel immune-type receptors (NITRs). While these genes encode an Ig domain, there is a cluster of C-type lectins encoded within this larger cluster [103,104]. Another bony fish, Oreochromis niloticus, possesses a KLR region, containing 26 genes, though more compact than its human counterpart [105]. Other teleosts have NKC genes, including Paralabidochromis chilotes [106] and rainbow trout [107], demonstrating that some of these genes arose in an ancestor common to both humans and bony fish earlier in the gnathostome lineage.
Conclusions
As explained above, features of NK receptor genes both within and between species are consistent with rapid evolutionary change. Inevitably, studies focus on human KIR and mouse Ly49 genes, but there are indications from the few studies of other vertebrates that variable C-type lectin and IgSF receptors for MHC class I molecules may co-exist in a species, such as horse [96], or may be functionally replaced by another divergent family of genes, as may be the case in chicken [100,101]. The marked expansion of the Ly49 gene family [76] and the large differences in gene number and content between closely related species, such as the mouse and rat [108], attest to rapid evolution of NK receptor genes. The existence of only one Ly49 pseudogene in humans [27] while a homolog remains functional in other primates [44,60] also indicates rapid evolutionary change since the divergence of a common ancestor. It is possible that the presence of gene families such as those found in the NK receptor complexes facilitates rapid evolution through recombination, subfunctionalization of duplicated genes, and conservation of essential sequence [109].
Emerging data are also consistent with rapid evolution of NKC-encoded genes. These include comparisons of KIR haplotypes in chimpanzee, rhesus macaque, and human, where, in addition to gene content differences, repeat elements in intronic regions suggest rapid evolution [41]. Another recent report showed that the KIR and Ly49 gene families have among the highest expansion rates in the genome, with the human KIRs expanding by 0.52 genes per million years and rat Ly49 genes expanding by 0.54 genes per million years since duplicating from a single common ancestor [76]. A more conservative study to determine the genome-wide average for this rate found it to be 0.001 to 0.03 genes duplicated per million years [110]. Recent data indicate that inhibitory KIRs are ancestral and that their activating counterparts have evolved from them by mutation [111]. It appears that the development of activating versions of polymorphic receptors takes place in both KIR and Ly49 loci. Thus, in different species, convergent evolution results in activating genes with similar signaling domains. This mechanism is necessitated so that the receptors can couple with appropriate signaling adaptors (see Figure 1), which are ancient and much more conserved than both KIR and Ly49. The mechanism is streamlined, in that tails associated with activating adapters become associated with different receptors by nonhomologous recombination [19]. Interestingly, activating receptors appear to evolve recurrently, presumably in line with selection associated with resistance to disease.
Perhaps because of this “response mode” fluctuation in activating versions of inhibitory receptors, it has proved difficult to identify their role and indeed their ligands. An interesting exception is mouse Ly49h, which recognizes mCMV-infected cells by a direct interaction with the m157 mCMV gene product [84]. So far, no other activating receptors appear to be dedicated to pathogen-specific products. However, there is evidence for epistasis between MHC class I loci and Ly49 in resistance to mCMV, which may explain the difficulty in identifying ligands for activating receptors [112]. Ly49p is an activating receptor that specifically recognizes mCMV-infected cells but only in the context of H-2Dk. Accordingly, binding of Ly49p was blocked by antibodies to H-2Dk but not by those to H-2Kk. It is not known what lies behind the epistasis. It is possible that Ly49p recognizes H-2Dk only when certain viral peptides are present. It has been proposed that in the case of another receptor, Ly49c, certain peptides might exert interactions through the floor of the H-2Kb binding groove, which are transmitted to the NK receptor by β2 microglobulin [113]. Alternatively, NK receptors such as Ly49c and Ly49p might respond to an H-2 molecule in the presence of a host-encoded protein that is upregulated upon viral infection. Intriguingly, similar mechanisms could be put forward to explain the role of activating KIRs such as KIR2DS1 being upregulated during Epstein-Barr virus infection [114].
Why do some species expand KIR genes while others expand Ly49 genes? While these two gene families produce proteins of analogous function, they do not share a common ancestor. These expansions vary greatly even within lineages, such as between human and chimpanzee (diverged approximately 5 million years ago) or mouse and rat (diverged approximately 20 million years ago) [115]. It is tempting to assume that different life spans, environments, sizes, and, specifically, pathogen interactions influence fixation of different NK receptor repertoires in different species. NK receptors are important components of antiviral immune responses and are an essential bridge between early innate responses, such as the release of virally induced interferon-alpha (IFNA) and interferon-beta (IFNB), and the later T cell and antibody adaptive responses [116]. Viruses evolve rapidly to environmental conditions, crossing species boundaries and mutating quickly to allow success in specific host species [117]. It is possible that the rapid evolution and host-species-specific adaptations of viral pathogens, which are common targets of NK-mediated immune responses, could influence the rapid expansion of different NK receptor repertoire combinations among species. Selective forces other than infection may also be entertained, including autoimmunity and reproduction, illustrated by the link between preeclampsia and combinations of HLA-C in the fetus and KIR in the mother [56].
NK receptor gene clusters coevolve with MHC genes [3,15], and there are clues to genetic and functional relationships between them [1]. MHC class I–like molecules have directed the development of different lymphocytes throughout evolution, as demonstrated both by the presence of common cell surface markers on NK cells, γδ T cells, and CD8+ αβ T cells and by distinct receptors present exclusively on each class [4]. The large differences in genomic organization of NK receptor gene complexes between species, and between populations of humans, are likely driven by resistance to infection and exposure to different local pathogens [118,119], although different mechanisms are possible [120]. Some MHC-encoded genes maintain a large number of low-frequency alleles within the population [118], produce molecules that interact to control NK cell function, and evolve rapidly to maintain their epistatic interactions [42]. Ig domains provide an example of the way coevolution of these interactions may occur. The Ig domains encoded by LRC genes are IgC2 or vIg-like [121]. IgC2 domains appear to have evolved to recognize different Ig-like receptors [15], such as those found in the MHC, consistent with coevolution of some LRC and MHC genes. Another example of coevolution between NK receptor genes and MHC genes is provided by the Mill gene family in mice and rats. The Mill family members are functional homologs of human MIC genes, which are found in the human MHC class I region. However, in mice and rats, the Mill genes have translocated to a chromosomal area near the LRC [87], which, through linkage with NK receptor genes, could facilitate coevolution of polymorphisms affecting their epistatic interactions and enhance their transcriptional regulation. Furthermore, the coevolution of NK receptors and HLA-C molecules, observed in primates [44], could have implications for diseases [4,5] and for the interactions between HLA-C molecules and decidual NK receptors in the placenta [56].
Additional insights into the rapid coevolution of the NK receptor genes and genes encoding their MHC class I ligands can be gained by studying their levels of polymorphism. Publicly available polymorphism data, evaluated in a previous study from our laboratory [122], showed that MHC class I molecules possess extremely high levels of polymorphism, while the numbers of polymorphisms per kilobase for NK receptor genes are nearer to values for the rest of the genome. Although some NK receptor gene families have noticeably varied gene contents between haplotypes, NK receptor genes are not as highly polymorphic as MHC class I ligands. Therefore, we could assume that selective pressures, such as exposure to pathogen, drive the generation of genetic variation primarily on MHC class I genes. The variation in NK receptor gene content both within and between species, such as preference towards rapid expansion of KIR, Ly49, or both, could be a mechanism for coevolving with the rapidly evolving MHC class I molecules. This would explain how NK receptor gene complexes exhibit rapid evolution, measured by parameters such as gene gain and loss [76], while having moderate levels of sequence polymorphism. Given the essential interactions of MHC and NK receptor gene clusters, the high levels of polymorphism, and association of the MHC with disease, studies of NK receptor gene complexes will have to be interpreted in relation to their MHC ligands.
As more configurations of NK receptor genes are determined for different species, it will become possible to track the way groups of KIR and Ly49 loci have followed different species lineages. Were rodents the only species to lose KIR function? Are primates unique in losing Ly49? What are the intermediates on the way to “all KIR” and “all Ly49” models? And what selective advantages drove the specialization towards the KIR or Ly49 model in different species?
JT is funded by Wellcome Trust and Medical Research Council.
Abbreviations
CHIRchicken Ig-like receptor
IgSFimmunoglobulin superfamily
ITAMimmunoreceptor tyrosine-based activation motif
ITIMimmunoreceptor tyrosine-based inhibitory motif
KIRkiller cell immunoglobulin-like receptor
LAIRleukocyte-associated Ig-like receptor
LILRleukocyte Ig-like receptor
LRCleukocyte receptor complex
mCMVmurine cytomegalovirus
MHCmajor histocompatibility complex
MICmajor histocompatibility complex class I chain-related protein
NKnatural killer
NKCnatural killer complex
SH2Src homology 2
SHP[number]Src homology 2 domain–containing protein tyrosine phosphatase [number]
==== Refs
References
Trowsdale J 2001 Genetic and functional relationships between MHC and NK receptor genes Immunity 15 363 374 11567627
Kasahara M Suzuki T Du Pasquier L 2004 On the origins of the adaptive immune system: Novel insights from invertebrates and cold-blooded vertebrates Trends Immunol 25 105 111 15102370
Kelley J Walter L Trowsdale J 2005 Comparative genomics of major histocompatibility complexes Immunogenetics 56 683 695 15605248
Parham P 2005 MHC class I molecules and KIRs in human history, health and survival Nat Rev Immunol 5 201 214 15719024
Kelley JM Trowsdale J 2005 Features of MHC and NK gene clusters Transpl Immunol In press.
Long EO 1999 Regulation of immune responses through inhibitory receptors Ann Rev Immunol 17 875 904 10358776
Natarajan K Dimasi N Wang J Mariuzza RA Margulies DH 2002 Structure and function of natural killer cell receptors: Multiple molecular solutions to self, nonself discrimination Ann Rev Immunol 20 853 885 11861620
Vely F Vivier E 1997 Conservation of structural features reveals the existence of a large family of inhibitory cell surface receptors and noninhibitory/activatory counterparts J Immunol 159 2075 2077 9278290
Stebbins CC Watzl C Billadeau DD Leibson PJ Burshtyn DN 2003 Vav1 dephosphorylation by the tyrosine phosphatase SHP-1 as a mechanism for inhibition of cellular cytotoxicity Mol Cell Biol 23 6291 6299 12917349
Lanier LL 2001 On Guard—Activating NK cell receptors Nat Immunol 2 23 27 11135574
Vivier E Nunes JA Vely F 2004 Natural killer cell signaling pathways Science 306 1517 1519 15567854
Kikuchi-Maki A Yusa S Catina TL Campbell KS 2003 KIR2DL4 is an IL-2-regulated NK cell receptor that exhibits limited expression in humans but triggers strong IFN-gamma production J Immunol 171 3415 3425 14500636
Trowsdale J Barten R Haude A Stewart CA Beck S 2001 The genomic context of natural killer receptor extended gene families Immunol Rev 181 20 38 11513141
Wende H Colonna M Ziegler A Volz A 1999 Organization of the leukocyte receptor cluster (LRC) on human chromosome 19q13.4 Mamm Genome 10 154 160 9922396
Martin AM Kulski JK Witt C Pontarotti P Christiansen FT 2002 Leukocyte Ig-like receptor complex (LRC) in mice and men Trends Immunol 23 81 88 11929131
Wilson MJ Torkar M Haude A Milne S Jones T 2000 Plasticity in the organization and sequences of human KIR/ILT gene families Proc Natl Acad Sci U S A 97 4778 4783 10781084
Martin AM Freitas EM Witt CS Christiansen FT 2000 The genomic organization and evolution of the natural killer immunoglobulin-like receptor (KIR) gene cluster Immunogenetics 51 268 280 10803839
Vilches C Parham P 2002 KIR: Diverse, rapidly evolving receptors of innate and adaptive immunity Ann Rev Immunol 20 217 251 11861603
Martin MP Bashirova A Traherne J Trowsdale J Carrington M 2003 Cutting edge: Expansion of the KIR locus by unequal crossing over J Immunol 171 2192 2195 12928362
Kikuchi-Maki A Catina TL Campbell KS 2005 Cutting edge: KIR2DL4 transduces signals into human NK cells through association with the Fc receptor {gamma} protein J Immunol 174 3859 3863 15778339
Torkar M Haude A Milne S Beck S Trowsdale J 2000 Arrangement of the ILT gene cluster: A common null allele of the ILT6 gene results from a 6.7-kbp deletion Eur J Immunol 30 3655 3662 11169408
Kandil E Egashira M Miyoshi O Niikawa N Ishibashi T 1996 The human gene encoding the heavy chain of the major histocompatibility complex class I-like Fc receptor (FCGRT) maps to 19q13.3 Cytogenet Cell Genet 73 97 98 8646894
Munday J Floyd H Crocker PR 1999 Sialic acid binding receptors (Siglecs) expressed by macrophages J Leukoc Biol 66 705 711 10577497
Hammarstrom S 1999 The carcinoembryonic antigen (CEA) family: Structures, suggested functions and expression in normal and malignant tissues Semin Cancer Biol 9 67 81 10202129
Renedo M Arce I Rodriguez A Carretero M Lanier LL 1997 The human natural killer gene complex is located on chromosome 12p12-p13 Immunogenetics 46 307 311 9218532
Yokoyama WM Plougastel BF 2003 Immune functions encoded by the natural killer gene complex Nat Rev Immunol 3 304 316 12669021
Westgaard IH Berg SF Orstavik S Fossum S Dissen E 1998 Identification of a human member of the Ly-49 multigene family Eur J Immunol 28 1839 1846 9645365
Lanier LL Chang C Phillips JH 1994 Human NKR-P1A. A disulfide-linked homodimer of the C-type lectin superfamily expressed by a subset of NK and T lymphocytes J Immunol 153 2417 2428 8077657
Lazetic S Chang C Houchins JP Lanier LL Phillips JH 1996 Human natural killer cell receptors involved in MHC class I recognition are disulfide-linked heterodimers of CD94 and NKG2 Subunits J Immunol 157 4741 4745 8943374
Braud VM Allan DS O'Callaghan CA Soderstrom K D'Andrea A 1998 HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C Nature 391 795 799 9486650
Vales-Gomez M Reyburn HT Erskine RA Lopez-Botet M Strominger JL 1999 Kinetics and peptide dependency of the binding of the inhibitory NK receptor CD94/NKG2-A and the activating receptor CD94/NKG2-C to HLA-E EMBO J 18 4250 4260 10428963
Plougastel B Trowsdale J 1997 Cloning of NKG2-F, a new member of the NKG2 family of human natural killer cell receptor genes Eur J Immunol 27 2835 2839 9394807
Wu J Song Y Bakker AB Bauer S Spies T 1999 An activating immunoreceptor complex formed by NKG2D and DAP10 Science 285 730 732 10426994
Billadeau DD Upshaw JL Schoon RA Dick CJ Leibson PJ 2003 NKG2D-DAP10 triggers human NK cell-mediated killing via a Syk-independent regulatory pathway Nat Immunol 4 557 564 12740575
Cosman D Mullberg J Sutherland CL Chin W Armitage R 2001 ULBPs, novel MHC class I- related molecules, bind to CMV glycoprotein UL16 and stimulate NK cytotoxicity through the NKG2D receptor Immunity 14 123 133 11239445
Guthmann MD Tal M Pecht I 1995 A new member of the C-type lectin family is a modulator of the mast cell secretory response Int ArchAllergy Immunol 107 82 86
Hamann J Montgomery KT Lau S Kucherlapati R van Lier RA 1997 AICL: A new activation-induced antigen encoded by the human NK gene complex Immunogenetics 45 295 300 9038101
Boles KS Barten R Kumaresan PR Trowsdale J Mathew PA 1999 Cloning of a new lectin-like receptor expressed on human NK cells Immunogenetics 50 1 7 10541800
Bull C Sobanov Y Rohrdanz B O'Brien J Lehrach H 2000 The centromeric part of the human NK gene complex: Linkage of LOX-1 and LY49L with the CD94/NKG2 region Genes Immun 1 280 287 11196705
Sobanov Y Bernreiter A Derdak S Mechtcheriakova D Schweighofer B 2001 A novel cluster of lectin-like receptor genes expressed in monocytic, dendritic and endothelial cells maps close to the NK receptor genes in the human NK gene complex Eur J Immunol 31 3493 3503 11745369
Sambrook JG Bashirova A Palmer S Sims S Trowsdale J 2005 Single haplotype analysis demonstrates rapid evolution of the killer immunoglobulin-like receptor (KIR) loci in primates Genome Res 15 25 35 15632087
Khakoo SI Rajalingam R Shum BP Weidenbach K Flodin L 2000 Rapid evolution of NK cell receptor systems demonstrated by comparison of chimpanzees and humans Immunity 12 687 698 10894168
Rajalingam R Parham P Abi-Rached L 2004 Domain shuffling has been the main mechanism forming new hominoid killer cell Ig-like receptors J Immunol 172 356 369 14688344
Guethlein LA Flodin LR Adams EJ Parham P 2002 NK cell receptors of the orangutan (Pongo pygmaeus): A pivotal species for tracking the coevolution of killer cell Ig-like receptors with MHC-C J Immunol 169 220 229 12077248
LaBonte ML Hershberger KL Korber B Letvin NL 2001 The KIR and CD94/NKG2 families of molecules in the rhesus monkey Immunol Rev 183 25 40 11782245
Hershberger KL Shyam R Miura A Letvin NL 2001 Diversity of the killer cell Ig-like receptors of rhesus monkeys J Immunol 166 4380 4390 11254692
Hershberger KL Kurian J Korber BT Letvin NL 2005 Killer cell immunoglobulin-like receptors (KIR) of the African-origin sabaeus monkey: Evidence for recombination events in the evolution of KIR Eur J Immunol 35 922 935 15714591
Adams EJ Parham P 2001 Species-specific evolution of MHC class I genes in the higher primates Immunol Rev 183 41 64 11782246
Adams EJ Thomson G Parham P 1999 Evidence for an HLA-C-like locus in the orangutan Pongo pygmaeus
Immunogenetics 49 865 871 10436180
Colonna M Borsellino G Falco M Ferrara GB Strominger JL 1993 HLA-C is the inhibitory ligand that determines dominant resistance to Lysis by NK1- and NK2-specific natural killer cells Procs Nat Acad Sci U S A 90 12000 12004
Khakoo SI Thio CL Martin MP Brooks CR Gao X 2004 HLA and NK cell inhibitory receptor genes in resolving hepatitis C virus infection Science 305 872 874 15297676
Martin MP Nelson G Lee JH Pellett F Gao X 2002 Cutting edge: Susceptibility to psoriatic arthritis: Influence of activating killer Ig-like receptor genes in the absence of specific HLA-C alleles J Immunol 169 2818 2822 12218090
van der Slik AR Koeleman BP Verduijn W Bruining GJ Roep BO 2003 KIR in type 1 diabetes: Disparate distribution of activating and inhibitory natural killer cell receptors in patients versus HLA-matched control subjects Diabetes 52 2639 2642 14514651
Yen JH Moore BE Nakajima T Scholl D Schaid DJ 2001 Major histocompatibility complex class I-recognizing receptors are disease risk genes in rheumatoid arthritis J Exp Med 193 1159 1167 11369787
Cook MA Milligan DW Fegan CD Darbyshire PJ Mahendra P 2004 The impact of donor KIR and patient HLA-C genotypes on outcome following HLA-identical sibling hematopoietic stem cell transplantation for myeloid leukemia Blood 103 1521 1526 14504099
King A Hiby SE Gardner L Joseph S Bowen JM 2000 Recognition of trophoblast HLA class I molecules by decidual NK cell receptors—A review Placenta 2 S81 S85
Canavez F Young NT Guethlein LA Rajalingam R Khakoo SI 2001 Comparison of chimpanzee and human leukocyte Ig-like receptor genes reveals framework and rapidly evolving genes J Immunol 167 5786 5794 11698452
Shum BP Flodin LR Muir DG Rajalingam R Khakoo SI 2002 Conservation and variation in human and common chimpanzee CD94 and NKG2 genes J Immunol 168 240 252 11751968
Barten R Trowsdale J 1999 The human Ly-49L gene Immunogenetics 49 731 734 10369937
Mager DL McQueen KL Wee V Freeman JD 2001 Evolution of natural killer cell receptors: Coexistence of functional Ly49 and KIR genes in baboons Curr Biol 11 626 630 11369209
Welch AY Kasahara M Spain LM 2003 Identification of the mouse killer immunoglobulin-like receptor-like (Kirl) gene family mapping to chromosome X Immunogenetics 54 782 790 12618911
Hoelsbrekken SE Nylenna O Saether PC Slettedal IO Ryan JC 2003 Cutting edge: Molecular cloning of a killer cell Ig-like receptor in the mouse and rat J Immunol 170 2259 2263 12594244
Bryceson YT Foster JA Kuppusamy SP Herkenham M Long EO 2005 Expression of a killer cell receptor-like gene in plastic regions of the central nervous system J Neuroimmunol 161 177 182 15748957
Kubagawa H Cooper MD Chen CC Ho LH Alley TL 1999 Paired immunoglobulin-like receptors of activating and inhibitory types Curr Top Microbiol Immunol 244 137 149 10453656
Liang S Baibakov B Horuzsko A 2002 HLA-G inhibits the functions of murine dendritic cells via the PIR-B immune inhibitory receptor Eur J Immunol 32 2418 2426 12207326
Maruoka T Nagata T Kasahara M 2004 Identification of the rat IgA Fc receptor encoded in the leukocyte receptor complex Immunogenetics 55 712 716 14647992
Iizuka K Naidenko OV Plougastel BF Fremont DH Yokoyama WM 2003 Genetically linked C-type lectin-related ligands for the NKRP1 family of natural killer cell receptors Nature Immunology 4 801 807 12858173
Chambers WH Vujanovic NL DeLeo AB Olszowy MW Herberman RB 1989 Monoclonal antibody to a triggering structure expressed on rat natural killer cells and adherent lymphokine-activated killer cells J Exp Med 169 1373 1389 2466943
Giorda R Rudert WA Vavassori C Chambers WH Hiserodt JC 1990 NKRP1, a signal transduction molecule on natural killer cells Science 249 1298 1300 2399464
Barten R Torkar M Haude A Trowsdale J Wilson MJ 2001 Divergent and convergent evolution of NK-cell receptors Trends Immunol 22 52 57 11286693
Wilhelm BT Gagnier L Mager DL 2002 Sequence analysis of the Ly49 cluster in C57BL/6 mice: A rapidly evolving multigene family in the immune system Genomics 80 646 661 12504856
Proteau MF Rousselle E Makrigiannis AP 2004 Mapping of the BALB/c Ly49 cluster defines a minimal natural killer cell receptor gene repertoire Genomics 84 669 677 15475244
Kamogawa-Schifter Y Ohkawa J Namiki S Arai N Arai KI 2004 Ly49Q defines two pDC subsets in mice Blood 105 2787 2792 15598811
Toyama-Sorimachi N Tsujimura Y Maruya M Onoda A Kubota T 2004 Ly49Q, a member of the Ly49 family that is selectively expressed on myeloid lineage cells and involved in regulation of cytoskeletal architecture Proc Natl Acad Sci U S A 101 1016 1021 14732700
Nylenna O Naper C Vaage JT Woon PY Gauguier D 2005 The genes and gene organization of the Ly49 region of the rat natural killer cell gene complex Eur J Immunol 35 261 272 15593300
Hao L Nei M 2004 Genomic organization and evolutionary analysis of Ly49 genes encoding the rodent natural killer cell receptors: Rapid evolution by repeated gene duplication Immunogenetics 56 343 354 15316687
Hurt P Walter L Sudbrak R Klages S Müller I 2004 The genomic sequence and comparative analysis of the rat major histocompatibility complex Genome Res 14 631 639 15060004
Hamerman JA Ogasawara K Lanier LL 2005 NK cells in innate immunity Curr Opin Immunol 17 29 35 15653307
Scalzo AA Lyons PA Fitzgerald NA Forbes CA Yokoyama WM 1995 Genetic mapping of Cmv1 in the region of mouse chromosome 6 encoding the NK gene complex-associated loci Ly49 and musNKR-P1 Genomics 27 435 441 7558024
Brown MG Dokun AO Heusel JW Smith HR Beckman DL 2001 Vital involvement of a natural killer cell activation receptor in resistance to viral infection Science 292 934 937 11340207
Daniels KA Devora G Lai WC O'Donnell CL Bennett M 2001 Murine cytomegalovirus is regulated by a discrete subset of natural killer cells reactive with monoclonal antibody to Ly49h J Exp Med 194 29 44 11435470
Lee SH Girard S Macina D Busa M Zafer A 2001 Susceptibility to mouse cytomegalovirus is associated with deletion of an activating natural killer cell receptor of the C-type lectin superfamily Nat Genetics 28 42 45 11326273
Sjolin H Tomasello E Mousavi-Jazi M Bartolazzi A Karre K 2002 Pivotal role of KARAP/DAP12 adaptor molecule in the natural killer cell-mediated resistance to murine cytomegalovirus infection J Expl Med 195 825 834
Arase H Mocarski ES Campbell AE Hill AB Lanier LL 2002 Direct recognition of cytomegalovirus by activating and inhibitory NK cell receptors Science 296 1323 1326 11950999
Lopez-Botet M Angulo A Guma M 2004 Natural killer cell receptors for major histocompatibility complex class I and related molecules in cytomegalovirus infection Tissue Antigens 63 195 203 14989708
Vance RE Kraft JR Altman JD Jensen PE Raulet DH 1998 Mouse CD94/NKG2A is a natural killer cell receptor for the nonclassical major histocompatibility complex (MHC) Class I molecule Qa-1(b) J Exp Med 188 1841 1848 9815261
Watanabe Y Maruoka T Walter L Kasahara M 2004 Comparative genomics of the mill family: A rapidly evolving MHC class I gene family Eur J Immunol 34 1597 1607 15162429
Ehrlich LI Ogasawara K Hamerman JA Takaki R Zingoni A 2005 Engagement of NKG2D by cognate ligand or antibody alone is insufficient to mediate costimulation of human and mouse CD8+ T cells J Immunol 174 1922 1931 15699119
Rosen DB Araki M Hamerman JA Chen T Yamamura T 2004 A structural basis for the association of DAP12 with mouse, but not human, NKG2D J Immunol 173 2470 2478 15294961
Storset AK Kulberg S Berg I Boysen P Hope JC 2004 NKp46 defines a subset of bovine leukocytes with natural killer cell characteristics Eur J Immunol 34 669 676 14991596
Storset AK Slettedal IO Williams JL Law A Dissen E 2003 Natural killer cell receptors in cattle: A bovine killer cell immunoglobulin-like receptor multigene family contains members with divergent signaling motifs Eur J Immunol 33 980 990 12672064
Govaerts MM Goddeeris BM 2001 Homologues of natural killer cell receptors NKG2-D and NKR-P1 expressed in cattle Vet Immunol Immunopathol 80 339 344 11457486
McQueen KL Freeman JD Takei F Mager DL 1998 Localization of five new Ly49 genes, including three closely related to Ly49c Immunogenetics 48 174 183 9683662
Ahn JS Hamilton MJ Davis WC Park YH 2002 Cloning of bovine CD69 Vet Immunol Immunopathol 88 43 48 12088643
Gagnier L Wilhelm BT Mager DL 2003 Ly49 genes in non-rodent mammals Immunogenetics 55 109 115 12712264
Takahashi T Yawata M Raudsepp T Lear TL Chowdhary BP 2004 Natural killer cell receptors in the horse: Evidence for the existence of multiple transcribed LY49 genes Eur J Immunol 34 773 784 14991607
Yim D Sotiriadis J Kim KS Shin SC Jie HB 2002 Molecular cloning, expression pattern and chromosomal mapping of Pig CD69 Immunogenetics 54 276 281 12136339
Kaufman J Milne S Gobel TWF Walker BA Jacob JP 1999 The chicken B locus is a minimal major histocompatibility complex Nature 401 923 925 10553909
Viertlboeck BC Crooijmans RP Groenen MA Gobel TW 2004 Chicken Ig-like receptor B2, a member of a multigene family, is mainly expressed on B lymphocytes, recruits both Src homology 2 domain containing protein tyrosine phosphatase (SHP)-1 and SHP-2, and inhibits proliferation J Immunol 173 7385 7393 15585863
Nikolaidis N Makalowska I Chalkia D Makalowski W Klein J 2005 Origin and evolution of the chicken leukocyte receptor complex Proc Natl Acad Sci U S A 102 4057 4062 15753291
Nikolaidis N Klein J Nei M 2005 >Origin and evolution of the Ig-like domains present in mammalian leukocyte receptors: Insights from chicken, frog, and fish homologues Immunogenetics Epub 2005 March 7.
Rogers SL Gobel TW Viertlboeck BC Milne S Beck S 2005 Characterization of the chicken C-type lectin-like receptors B-NK and B-lec suggests that the NK complex and the MHC share a common ancestral region J Immunol 174 3475 3483 15749883
Yoder JA Mueller MG Wei S Corliss BC Prather DM 2001 Immune-type receptor genes in zebrafish share genetic and functional properties with genes encoded by the mammalian leukocyte receptor cluster Proc Natl Acad Sci U S A 98 6771 6776 11381126
Yoder JA Litman RT Mueller MG Desai S Dobrinski KP 2004 Resolution of the novel immune-type receptor gene cluster in zebrafish Proc Natl Acad Sci U S A 101 15706 15711 15496470
Kikuno R Sato A Mayer WE Shintani S Aoki T 2004 Clustering of C-type lectin natural killer receptor-like loci in the bony fish Oreochromis niloticus
Scand J Immunol 59 133 142 14871289
Sato A Mayer WE Overath P Klein J 2003 Genes encoding putative natural killer cell C-type lectin receptors in teleostean fishes Proc Natl Acad Sci U S A 100 7779 7784 12802013
Zimmerman AM Evenhuis JP Thorgaard GH Ristow SS 2004 A single major chromosomal region controls natural killer cell-like activity in rainbow trout Immunogenetics 55 825 835 14968267
Wilhelm BT Mager DL 2004 Rapid expansion of the Ly49 gene cluster in rat Genomics 84 218 221 15203220
Trowsdale J Parham P 2004 Defense strategies and immune-related genes Eur J Immunol 34 7 17 14971025
Lynch M Conery JS 2003 The evolutionary demography of duplicate genes J Struct Funct Genomics 3 35 44 12836683
Abi-Rached L Parham P 2005 Natural selection drives recurrent formation of activating killer cell immunoglobulin-like receptor and Ly49 from inhibitory homologues J Exp Med 201 1319 1332 15837816
Desrosiers MP Kielczewska A Loredo-Osti JC Adam SG Makrigiannis AP 2005 Epistasis between mouse Klra and major histocompatibility complex class I loci is associated with a new mechanism of natural killer cell-mediated innate resistance to cytomegalovirus infection Nat Genetics 37 593 599 15895081
Dam J Guan R Natarajan K Dimasi N Chlewicki LK 2003 Variable MHC class I engagement by Ly49 natural killer cell receptors demonstrated by the crystal structure of Ly49C bound to H-2K(b) Nat Immunol 4 1213 1222 14595439
Stewart CA Laugier-Anfossi F Vely F Saulquin X Riedmuller J 2005 Recognition of peptide-MHC class I complexes by activating killer immunoglobulin-like receptors Proc Natl Acad Sci U S A In press.
Springer MS Murphy WJ Eizirik E O'Brien SJ 2003 Placental mammal diversification and the cretaceous-tertiary boundary Proc Natl Acad Sci U S A 100 1056 1061 12552136
Biron CA Nguyen KB Pien GC Cousens LP Salazar-Mather TP 1999 Natural killer cells in antiviral defense: Function and regulation by innate cytokines Ann Rev Immunol 17 189 220 10358757
Ludwig B Kraus FB Allwinn R Doerr HW Preiser W 2003 Viral zoonoses—A threat under control? Intervirology 46 71 78 12684545
Parham P Ohta T 1996 Population Biology of Antigen Presentation by MHC Class I Molecules Science 272 67 74 8600539
Hill AV 1999 Immunogenetics. Defence by diversity Nature 398 668 669 10227287
Hiby SE Walker JJ O'Shaughnessy KM Redman CW Carrington M 2004 Combinations of maternal KIR and fetal HLA-C genes influence the risk of preeclampsia and reproductive success J Exp Med 200 957 965 15477349
Henry J Miller MM Pontarotti P 1999 Structure and evolution of the extended B7 family Immunol Today 20 285 288 10354554
Kelley J de Bono B Trowsdale J 2005 IRIS: A database surveying known human immune system genes Genomics 85 803 833
Wheeler DL Barrett T Benson DA Bryant SH Canese K 2005 Database resources of the national center for biotechnology information Nucleic Acids Res 33 D39 D45 15608222
|
16132082
|
PMC1193534
|
CC BY
|
2021-01-05 08:00:58
|
no
|
PLoS Genet. 2005 Aug 26; 1(2):e27
|
utf-8
|
PLoS Genet
| 2,005 |
10.1371/journal.pgen.0010027
|
oa_comm
|
==== Front
PLoS GenetPLoS GenetpgenplgeplosgenPLoS Genetics1553-73901553-7404Public Library of Science San Francisco, USA 10.1371/journal.pgen.001002905-PLGE-RA-0001R4plge-01-02-12Research ArticleDevelopmentSystems BiologyGenetics/GenomicsGenetics/Gene ExpressionDanio (Zebrafish)Transcriptome Analysis of Zebrafish Embryogenesis Using Microarrays Zebrafish TranscriptomeMathavan Sinnakaruppan 1Lee Serene G. P 1Mak Alicia 1Miller Lance D 1Murthy Karuturi Radha Krishna 1Govindarajan Kunde R 1Tong Yan 2Wu Yi Lian 2Lam Siew Hong 2Yang Henry 4Ruan Yijun 1Korzh Vladimir 3Gong Zhiyuan 2Liu Edison T 1Lufkin Thomas 1*1 Genome Institute of Singapore, Singapore
2 Department of Biological Sciences, National University of Singapore, Singapore
3 Institute of Molecular and Cell Biology, Singapore
4 Bioinformatics Institute, Singapore
Mullins Mary EditorUniversity of Pennsylvania School of Medicine, United States of America* To whom correspondence should be addressed. E-mail: [email protected] 2005 26 8 2005 1 2 e2910 1 2005 14 7 2005 Copyright: © 2005 Mathavan 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.Zebrafish (Danio rerio) is a well-recognized model for the study of vertebrate developmental genetics, yet at the same time little is known about the transcriptional events that underlie zebrafish embryogenesis. Here we have employed microarray analysis to study the temporal activity of developmentally regulated genes during zebrafish embryogenesis. Transcriptome analysis at 12 different embryonic time points covering five different developmental stages (maternal, blastula, gastrula, segmentation, and pharyngula) revealed a highly dynamic transcriptional profile. Hierarchical clustering, stage-specific clustering, and algorithms to detect onset and peak of gene expression revealed clearly demarcated transcript clusters with maximum gene activity at distinct developmental stages as well as co-regulated expression of gene groups involved in dedicated functions such as organogenesis. Our study also revealed a previously unidentified cohort of genes that are transcribed prior to the mid-blastula transition, a time point earlier than when the zygotic genome was traditionally thought to become active. Here we provide, for the first time to our knowledge, a comprehensive list of developmentally regulated zebrafish genes and their expression profiles during embryogenesis, including novel information on the temporal expression of several thousand previously uncharacterized genes. The expression data generated from this study are accessible to all interested scientists from our institute resource database (http://giscompute.gis.a-star.edu.sg/~govind/zebrafish/data_download.html).
Synopsis
In an effort to shed light on the mysterious molecular process of embryogenesis, scientists have employed the powerful technological approach of high-density microarrays to analyze for the first time (to our knowledge) the expression patterns of over 16,000 zebrafish genes during the major stages of embryonic development. They determined that developmentally regulated genes in zebrafish could be grouped based upon their time of transcript accumulation onset (activation) and their peak of activity (peak accumulation). For example, genes encoding transcriptional regulators are activated en masse earlier than the target genes that they control. Thus, the developmental gene expression profile in zebrafish can be grouped into successive waves of gene activation followed by peak activity. One very novel group of genes that was discovered in this study is a set of about 100 genes that are activated even earlier than what was previously thought to be the earliest time of embryonic transcriptional activation. These genes are likely to be important for preparing the embryonic cells for subsequent high-level genome activation and normal development.
Citation:Mathavan S, Lee SGP, Mak A, Miller LD, Murthy KRK, et al. (2005) Transcriptome analysis of zebrafish embryogenesis using microarrays. PLoS Genet 1(2): e29.
==== Body
Introduction
Zebrafish is an important vertebrate model for the analysis of developmentally regulated genes. Its advantages are rapidly developing transparent embryos, a short generation time, and amenability to genetic manipulation. In recent years, large-scale mutagenesis has been undertaken by both chemical mutagens [1–3] and proviral insertions [4–6]. Genetic tools such as transgenesis and morpholino knock-down assays have also been established for zebrafish [7,8]. In effect, a unique combination of these features makes zebrafish a strong model to study vertebrate developmental disorders and human hereditary disease [1,9–11]. It is believed that the zebrafish genome may contain about 30% more genes than the human genome, owing to an additional round of genome duplication about 450 million years ago, followed by extensive gene loss [12]. Several expressed sequence tag (EST) projects were launched to characterize genes expressed during different developmental stages [13–15] and in the various tissues of zebrafish [16]. The zebrafish database contains more than 400,000 ESTs grouped into about 18,000 clusters, but only a few thousand ESTs have been analyzed by whole embryo in-situ hybridization (WISH) (http://zfin.org). Systematic analysis of the temporal expression of the ESTs by WISH may take several more years to complete. A high-throughput expression genomics approach would provide complementary information to the single-gene approaches currently under way. Microarrays are currently the strongest technology platform for large-scale analysis of gene expression profiles during embryogenesis. They provide an opportunity to simultaneously monitor the expression of thousands of genes at various developmental stages, thus providing an opportunity to analyze temporal and spatial patterns of gene expression [17–23]. Preliminary analysis of the expression of a subset of zebrafish genes has been undertaken using cDNA microarrays [14,24,25]. Global evaluation of gene expression during the life cycle and organogenesis of Drosophila and mouse has been reported [20,23,26–29]. In contrast, a high-density microarray for genome-wide gene expression (transcriptome) analysis has not previously been attempted for zebrafish. In the present study, we used high-density oligonucleotide microarrays containing 16,416 genes to map the transcriptome profiles at 12 different developmental time points (from unfertilized eggs to the hatching stage). RNA was collected at closely spaced time points during blastula and gastrula stages—the dynamic development stages—and thus provides a dense coverage and a comprehensive analysis of the expression patterns of developmentally regulated genes during embryogenesis.
Traditionally, it is believed that the zygotic genome becomes active at the mid-blastula transition (MBT) [30]. However, recent evidence in Xenopus has shown the commencement of zygotic genome transcription to occur prior to MBT [31]. We speculated that there might be a subset of genes activated with regards to transcript accumulation prior to MBT that in turn could play a subsequent role in genome-wide transcriptional activation but had escaped detection owing to their low numbers relative to the entire genome. To test our speculation, we made a systematic analysis of gene expression profiles preceding MBT, which revealed for the first time to our knowledge a striking cohort of genes that actively increase transcript levels prior to MBT. These new data form the starting point for determining the role of these genes in the control of zebrafish genome activation and embryogenesis. To make the data we have generated here easily available to the public, we have created a GIS-sponsored Web site (http://giscompute.gis.a-star.edu.sg/~govind/zebrafish/index.html) that includes annotation data and all expression data for all the genes in the zebrafish microarray.
Results/Discussion
Validation of Zebrafish Microarray Data
The microarray data obtained in this work were validated using the following criteria: Firstly, 172 copies of the beta-actin gene were printed on the oligonucleotide array as calibration spots. Biological and experimental replicates of array hybridizations showed that, at the selected time points, the signals of all the beta-actin calibration spots were identical (Figure S1), indicating the reproducibility and reliability of the array data. Secondly, candidate genes with known expression patterns were compared with the array-derived expression patterns, and the patterns were found to be similar to one another (Table S1). In addition, real-time PCR analysis was performed for two genes (beta-actin and myosin light chain, mlc2f) at all selected time points. The quantity of transcripts detected could not be compared because of different methods of estimations. However, the pattern of transcript abundance detected for these genes in the array and in real-time PCR showed nearly identical expression profiles (Figure S2). Further, expression data of ribosomal protein (RP) genes of zebrafish obtained from the arrays showed temporally coordinated expression (see results section of RP gene expression). In order to confirm the array data, a Northern blot analysis was performed for selected genes (rpl7a-, rps3a-, rps12-, and rps10-) using the RNA samples collected at 14 different time points during embryogenesis. The pattern of expression of RP genes obtained in the array and Northern blot were essentially identical, validating the array data (Figure S3A). Intensity of hybridization of RP genes in the Northern blot were digitized and measured. Using the digitized data, the pattern of transcript accumulation was compared with array data, and the pattern of expression was similar in both the experiments (Figure S3B). These analyses demonstrated the validity and reliability of the array data presented in this work.
Annotation of the Zebrafish Oligoarray
The annotation given by Compugen (oligonucleotide supplier) is based on Entrez Nucleotide, BLAST, and Gene Ontology (GO) (http://www.labonweb.com/oligo) and is outdated. For example, the company database gives a description based on GO terms for about 1,152 probes, and the remaining are indicated as “GO unknown.” From our recent annotation, we have found GO terms for more than twice those reported by the company. We made critical analysis of the clones and identified the duplicates to determine the actual number of non-redundant clones in the array. The array represented 16,416 GenBank entries. Of this, 16,177 probes had non-redundant GenBank entries and the rest of the 239 GenBank entries were duplicated (including 171 copies of beta-actin) in the array. We mapped the non-redundant GenBank entries to the zebrafish UniGene database (UniGene build 79) and found that 2,751 GenBank entries did not have UniGene IDs and the rest of the 13,426 entries matched to UniGenes. Of these clones with UniGene entries, 12,153 clones were non-redundant, while the rest (1,273 UniGenes) were duplicates. On the whole, the probes in the array represented 14,904 non-redundant genes (12,153 non-redundant UniGene IDs plus 2,751 non-redundant GenBank entries). We have included here the most recent putative annotations for the clones in the array by obtaining the annotations from various database sources. The annotations can be obtained from our institute resource Web site (http://giscompute.gis.a-star.edu.sg/~govind/zebrafish/index.html). We have also included examples of annotations in the Datasets. The resources used to build the zebrafish annotation database, strategies followed, and the query methods for annotation retrieval are described in the Materials and Methods section.
Transcriptome Analysis Identified Developmentally Regulated Genes
The gene expression pattern of the zebrafish genome during embryogenesis was investigated using high-density oligonucleotide arrays representing 16,416 zebrafish genes. High-resolution time-course expression data were generated during embryogenesis. Considering the sensitivity of the array data and the number of replicates of each sample, it is very likely that most of the polyadenylated transcripts were detected for the genes represented in the array for each time point. Total RNA was extracted at 12 different developmental time points (see Materials and Methods for details), and for each time point analysis two to three independent biological replicates were made. Reference RNA was prepared by pooling equal concentration of total RNA from various embryonic stages and also adult male and female fish. The intensity of value of each spot and each side was normalized using the intensity-based log ratio median method [32]. Differentially expressed genes were selected using modified t-statistic (SAM) [33] with 15% of the standard deviation percentile as the fudge constant and log2 ratios. Stage-specific genes were selected following paired t-test (see Materials and Methods for details). By this procedure, we have identified 3,657 genes that showed significant levels of differential expression with a single peak during the course of development. This subset of developmentally regulated genes was used for further analysis of gene activity during embryogenesis. For the benefit of other scientists, the entire dataset generated from this study has been placed in our institute resource database, which can be accessed at http://giscompute.gis.a-star.edu.sg/~govind/zebrafish/data_download.html.
Hierarchical clustering of 3,657 genes demonstrated diverse temporal expression profiles during zebrafish embryogenesis (Figure 1 and Dataset S1). The complexity of the gene cluster clearly displayed the significant modulations in the temporal activity of developmentally regulated genes. The cluster also revealed the identical expression profile of all the copies of beta-actin genes. Of the 3,657 genes extracted for the analysis, 622 genes showed a maximum level of transcript accumulation at the maternal stage (unfertilized egg), and the rest of the 3,035 genes showed an onset and peak of transcript abundance during one or more stages of embryonic development (Figure 2).
Figure 1 Hierarchical Clustering Analysis of Differentially Expressed Zebrafish Genes
Of the 16,416 genes in the array, 3,657 genes, which showed a significant level of differential expression at least at one time point, were clustered hierarchically into groups on the basis of similarity of their expression profile, following Eisen's clustering method. The horizontal lines indicate the expression pattern of each gene, and the vertical rows indicate the embryonic developmental stages. For each gene, the ratio of transcript abundance in the developing stages of the embryo to its abundance in the reference RNA is represented by color intensities (red color indicates the higher expression, and green color indicates the lower expression of the gene in the embryos). Coordinated expression of the beta-actin gene (control) is indicated.
Figure 2 Analysis of Expression Patterns of the Zebrafish Embryonic Transcriptome
(A) The onset and peak of transcript accumulation were calculated using an algorithm for each of the differentially expressed genes. An average performance of the gene in each group is presented in the clustergram. Of the set of 1,972 genes commencing at the onset of the blastula stage, subsets of genes displayed peak activity at blastula, gastrula, segmentation, and pharyngula stages. Average performance of these genes in each subset is represented in the clustergram. Similar analysis was made for the genes commencing transcript accumulation at gastrula, segmentation, and pharyngula stages, and the data are presented.
(B) Patterns of differential degradation of maternally loaded transcripts. (a) Transcripts level of the genes in this group degraded drastically prior to blastula. Groups of genes whose transcripts persisted till blastula (b) and subsequent stages (c) of development are represented graphically.
(C) Cumulative analysis of onset and peak of expression of zygotic genes. A number of zygotic genes commenced their transcript accumulation steadily during early development. About 99% of the genes initiated transcript accumulation by the end of the segmentation stage. The figure inserted in Figure 2C indicates that the percentage of genes showing peak of activity also increased steadily and about 90% of the zygotic genes displayed their peak of expression within 24 hpf (prior to the pharyngula stage). The dynamics of zygotic genome activation during early development is evident from the cumulative expression analysis.
(D and E) Frequency distribution of onset of transcript accumulation (D) and peak of expression (E). Of the 3,657 genes analyzed, about 54% (> 2,000 genes) commenced transcript accumulation during the blastula stage. Considering the peak of expression, the majority of genes (> 1,000 genes) displayed a peak of expression during the gastrula stage.
(F) Transcriptome analysis during blastula and gastrula stages. Of the total zygotic genes (3,035 genes; 100%; red circle), percentage of genes commencing transcript accumulation (80.5%; yellow circle) and showing peak of expression (53%; light-blue circle) during blastula and gastrula stages are indicated pictorially.
(G) Total period of embryonic development of zebrafish (48 h; yellow bar) and the time required for each stage of development is indicated (color-coded for each stage). Considering all the developmentally regulated genes (3,657), the percentage of genes showing peak of activity at the designated developmental stages are indicated in the top bar. The bottom bar indicates the percentage of zygotic genes (3,035 genes) commencing transcript accumulation at the designated developmental stages.
Differential Degradation of Maternally Loaded Gene Transcripts
The dataset of 622 gene transcripts displaying maximum abundance at the unfertilized egg stage (maternally loaded transcripts; (Figure 2B and Dataset S2) was divided into three groups based on the pattern of transcript degradation. The first group represented 221 genes, and the transcripts of these genes persisted only at the unfertilized egg stage and were degraded prior to the commencement of the blastula stage (Figure 2B, line a). This group contained the known maternal genes (examples are those encoding proteins of the zona pellucida, zp2, zp2.4, and zp3 [34]) and the chorion protein component. A number of genes with putative annotations also displayed an expression pattern identical to previously characterized maternal genes, suggesting a possible maternal function. The clones with putative annotation of rhamnose-binding lectins (STL) of other fish are included in this group of genes. This family of lectins is expressed in the ovary and egg of the steelhead trout (Oncorhynchus mykiss). Furthermore, a few members of the lectin family (STL2 and STL3) were found to be abundant in the ovary and the levels dramatically decreased after fertilization [35,36], indicating the maternal specificity of this gene. We have identified here, for the first time to our knowledge, a large number of previously uncharacterized genes and ESTs that appear to have maternal-specific activity. The maternal function of these clones and their control of zygotic genome activation will be an important area for future investigations.
The second group of transcripts (259 genes) persisted to a greater extent during maternal and blastula stages and subsequently degraded (Figure 2B, line b). These genes may be involved in cleavage and other early embryonic functions. One example is the maternally derived gene cyclin B1, which is expressed up to the early blastula stage. It is known that zebrafish embryos undergo synchronous cell divisions during the cleavage stage prior to their entry into the MBT [30]. cyclin B1 has been shown to be involved in the synchronous cell division phase [37], consistent with its maximum activity during these stages. Similar to cyclin B1, there are a number of genes that display activity at maternal and early zygotic stages. The gene transcripts that persisted during the blastula stage are involved, in one way or the other, in early developmental functions. Some of the genes functionally relevant during maternal and blastula stages are given here as examples: calreticulin (a chaperone involved in folding of newly synthesized glycoproteins), catalase (catalytic functions in cell growth), birc5b/survivin2 (cell death regulator), mcl1a (cellular apoptosis function), claudin-like gene (tight junction function during growth), lmnl3 (maternal lamin L3), oocyte and blastula nucleoskeletal proteins of the nuclear membrane during cell division. These gene transcripts are degraded slowly (by the end of the blastula stage). The third group (Figure 2B, line c) had 142 genes with a maximum level of transcripts at the maternal stage; this group of genes underwent a very slow rate of transcript degradation. Some of the transcripts persisted up to the segmentation stage, and the average performance of this group is presented.
These observations revealed that the maternally deposited gene transcripts are not degraded at the same rate and there is a definite gene-specific differential degradation. The differential degradation and the extended stability of specific maternal genes implicated a potential relevance of these genes in later zygotic stages. It is known that after fertilization the earliest developmental processes in the egg are programmed by maternally deposited gene transcripts [38]. Subsequently, embryos initiate zygotic genome activation for the continuation of embryonic development. The maternal control of zebrafish development before and after MBT has recently been reported following extensive analysis of maternal mutants [2,3]. These authors showed that mutations in some of the maternal genes have generated embryos with defects in zygotic gene functions and developmental alterations during pre-MBT, MBT, and also beyond these stages. In this study, we have identified maternal stage-specific expression for a number of previously uncharacterized ESTs, which may play a vital role in modulating normal development. It has been reported in Drosophila that maternal RNA is degraded during the course of early development either by maternally derived factors (to degrade transient transcripts) or zygotically derived factors (to degrade moderately stable transcripts), and that the combined action of maternal and zygotic factors is required to degrade the highly stable maternal transcripts [39–42]. Zygotic regulation of maternal cyclin A1 and cyclin B2 mRNAs has been demonstrated in Xenopus laevis [43,44]. We presume that similar mechanisms of maternal RNA degradation reported for Drosophila and Xenopus may be acting in zebrafish to differentially degrade the maternally loaded messages.
Zygotic Transcriptome Analysis and an Overview of Stage-Specific Clusters
The time point at which the gene transcripts commenced accumulation from their basal expression level is considered as the time of onset, and it is determined using an algorithm (see Materials and Methods and Protocol S1). In the present study, stage-wise developmental onset of gene transcription was determined for 3,052 zygotic genes expressed during MBT and beyond (Figure 2A). Of these, transcript accumulation of 1,967 genes began during the blastula stage, which represents 64.8% of the zygotic genes. Though these genes commenced transcript accumulation at the blastula stage, a subset of these genes showed expression peaks at subsequent stages of development. It was also observed that 15.7% (477genes), 15.1% (460 genes), and 4.4 % (131 genes) of the zygotic genes began their onset of transcript accumulation at gastrula, segmentation, and pharyngula stages, respectively. Average performance of the genes in these groups is presented as a clustergram (Figure 2A and Datasets S3–S12). Onset of temporally expressing genes aligns with their required developmental functions. For example, some of the genes exhibiting early transcript accumulation are those involved in early developmental functions such as the cell cycle (cyclins), cell regulation and cell adhesion (claudins, connexins), embryonic apoptotic functions (survivins), transcriptional regulation (smad, pou, sox, wnt), and other early developmental processes. The genes that began transcription during later stages of development (segmentation and pharyngula) are related more to organogenesis and encode, to a large extent, structural proteins (e.g., collagens, pro-collagens, skeletal muscle proteins, selenoproteins, ceruloplasmin).
Embryonic development in zebrafish (from fertilization to hatching) extends for 48 h. Within the first 5 h post-fertilization (hpf), cleavage and blastula are completed, and the process of gastrulation takes another 5 h. Thus, within 10 hpf (which is equal to about 21% of the total embryonic period), the early developmental processes are completed. Segmentation and pharyngula stages are completed in about 14 and 24 h, respectively (Figure 2G). About 80.5% of the zygotic genes initiated their transcript accumulation during the blastula (64.8%) and gastrula (15.7%) stages. Considering the peak of expression of zygotic genes, about 53% (Figure 2F) of the genes attained their peak during blastula and gastrula stages. Such high level of activity occurred within about 21% of the developmental period. Cumulative analysis of onset and peak of expression showed a sigmoid pattern: the onset of transcript accumulation sharply increased, displaying about 99% of the activity by 12 hpf (Figure 2C). Similarly, most of the developmentally regulated genes attained their peak by about 24 hpf (Figure 2C insert). Analysis of the frequency of distribution indicated the dominance of transcriptome dynamics during blastula and gastrula stages (Figure 2D and 2E). It is interesting that such a high percentage of genes would be activated within the relatively short window of the developmental period.
To obtain an overview of embryonic gene expression, we sought to specifically address the temporal expression patterns in the following developmental stages: (1) maternal (unfertilized eggs); (2) blastula; (3) gastrula; (4) segmentation; and (5) pharyngula/hatching stages (Figure 3A). From the expression data, we selected the genes that showed peak of activity at each stage and presented the data in clustergrams (Figure 3B). The average performance of the genes in each developmental stage is also graphically presented (Figure 3C). This analysis illustrates that different set of genes have peaks of expression at different developmental stages suggestive of their relevance for stage-specific developmental functions. Intriguingly, the clusters of the developmentally regulated genes grouped into specific or adjacent clusters, forming a wave of gene activity moving from maternal to blastula and through each subsequent stage of embryogenesis. The clustering of developmentally related genes indicated that the process of embryogenesis is continuously regulated and displayed dynamic gene activity that, in turn, directs the evolving program of development.
Figure 3 Overview of the Expression Patterns of Genes Peaking at Selected Developmental Stages
(A,B) Differentially expressed genes were clustered based on the peak of expression at the selected developmental stages and presented in the clustergram. Of the total number of 3,657 genes analyzed, 622, 609, 1,006, 688, and 732 genes showed peak expression at maternal, blastula, gastrula, segmentation, and pharyngula stages, respectively.
(C) General trend and average performance of the genes at each developmental stage are graphically represented (red, high expression; green, low expression). It is clear from this analysis that different sets of genes displayed their maximum at different stages of development, indicating the temporal/stage-specific maximum activity of the developmentally regulated genes.
The following observations can be made from this analysis. Of the total number of 3,657 genes analyzed, 622 genes (17%; Dataset S13) showed maximum activity at the maternal stage. Of the remaining 3,035 (zygotic) genes, 609 (16.6%), 1,006 (27.5%), 688 (18.9%), and 732 (20.0%) genes showed maximum transcript levels at blastula (Dataset S14), gastrula (Dataset S15), segmentation (Dataset S16), and pharyngula (Dataset S17) stages, respectively (Figure 3B). Analysis of our results indicated that some of the genes have short spans of activity while others have extended periods of expression, suggestive of functional requirements. As an example, we analyzed the array data and found abundant transcripts for birc5b (baculoviral IAP repeat-containing 5B, also called survivin2) and cell-death regulator mcl1a (myeloid cell leukemia sequence 1a) up to the blastula stage. These gene families have been implicated in anti-apoptotic functions [45–47]. It has also been shown that survivin is abundantly expressed in Xenopus during oogenesis and early embryogenesis functioning as an inhibitor of apoptosis and a positive regulator of the cell cycle [48]. The transcript abundance of birc5b and mcl1a, negative regulators of apoptosis, during maternal/blastula stages, indicated the importance of regulating apoptotic process during early embryonic development. It has been reported that embryonic apoptosis is activated in the zebrafish during the gastrula stage, and it is most likely that down-regulation of the maternally derived negative regulators may activate the embryonic apoptotic process in zebrafish during the gastrula stage [49].
Genes displaying a peak of expression during the pharyngula stage continued their expression beyond the pharyngula stage, suggesting a functional requirement for these genes during embryonic and post-embryonic stages of development. For example, ceruloplasmin (cp), which has been implicated in liver development [50], displayed maximum expression during the pharyngula stage. Expression of this gene did not terminate at the end of the pharyngula stage, suggesting that its continued expression during post-embryonic stages could be playing a role in liver function. The genes that displayed similar expression are involved in parallel embryonic and post-embryonic developmental processes. For example, muscle-specific genes and collagens also show similar patterns of expression, as they are involved in the progressive formation of muscles and skeletal development in zebrafish. It is known that almost all the primordial processes of organogenesis commence at gastrulation and continue during segmentation and post-segmentation stages. Thus, an overview of genome-wide expression analysis demonstrated a temporally demarked expression pattern of developmentally regulated genes and showed that the gene activities were well correlated with the developmental events at the respective stages. Prior to our analysis, only 7% of the differentially expressed genes were fully annotated (with GO), and temporal expression patterns for the majority of the clones in the array were not established. Our analysis, for the first time to our knowledge, identified the temporal patterns of expression for about 80%–90% of the clones, and for most of them the patterns of expression are presented here for the first time, to our knowledge.
Transcription Analysis of Selected Clusters of Functionally Related Genes
From the array data, expression patterns of a number of functionally related genes could be identified based on their GO and/or putative functions. However, we selected three groups of genes involved in (1) cell cycle, (2) ubiquitin functions (ubiquitins and proteasomes), and (3) somitogenesis as examples and discuss here the expression patterns of these genes in relation to their biological significance.
Cell cycle.
Expression of cell cycle related genes encoding cyclins and cyclin-dependent kinases were analyzed (Figure 4A; Dataset S18). Most of the cyclin genes involved in the cell cycle commenced their expression maternally, maintained expression through the blastula stage, and were subsequently down-regulated. The down-regulation of maternally derived cyclin transcripts was shown to be regulated by zygotically derived factors [43,44,51]. From the array data, it is clear that cyclinD1 alone commenced expression zygotically (at the blastula stage) and exhibited its peak after MBT, an observation that is in agreement with the earlier analysis of cyclinD1 gene transcription [52]. Thus, cyclinD1 is a non-maternally supplied G1 cyclin, and the onset of expression indicates the G1 phase in the developing embryo [52]. In contrast to cyclinD1, the maximum level of cyclinB1 and cyclinE transcripts were observed during the phase of synchronous cell divisions consisting of S and M phases (prior to MBT) [37,53]. However, the continued presence of transcripts of cyclinE during asynchronous cell divisions and post-MBT development indicated that the expression profiles differ from the transcripts of cyclinB1. Cyclin-dependent kinases, their regulatory subunits, activators, and associated proteins showed almost continuous transcript abundance with a maximum during blastula, gastrula, and segmentation stages (Figure 4A), indicating their continuous involvement in the regulation of one or other cyclin genes during the cell cycle.
Figure 4 Expression of Genes Involved in Specific Functions
(A) Expression patterns of genes involved in the cell cycle. Genes involved in the cell cycle, namely cyclin and cyclin-dependent kinase genes, were identified based on GO and clustered separately. Most of the cyclin genes commenced their expression maternally, and the cyclin-dependent kinase genes showed expression throughout embryogenesis.
(B and C) Genes involved in ubiquitin function (proteasomes and ubiquitins) are active during most of the developmental stages, and the peak of activity is between the gastrula and segmentation stages.
(D) Gene expression during somitogenesis. Genes involved in somitogenesis were selected based on the GO list. Expression of MSP genes (acta 1, actc1, tpma, ckm, tnnt1, mibp2, myhz1, myhz2, mylz2, mylz3, tnnc, and tnnt3a) are clustered. Most of the MSP genes commenced their expression around 12 hpf and displayed their maximum during the pharyngula stage.
(E) Expression pattern of somitogenic (myotome-specific) transcription factors (mespa, mespb, mef2a, mef2c, med2d, myf5, myog, foxc1a, her4, her6, her7, and her9). Most of the transcription factors commenced transcript accumulation from the blastula stage onwards well in advance of commencement of MSP gene transcription.
(F) Coordinated expression of RP genes. From the array data, the expression profile of RP genes was clustered. Most of the genes showed an identical pattern of expression.
Ubiquitins and proteasomes.
Genes involved in protein degradation by ubiquitin-mediated proteolytic function displayed low levels of transcripts during early development (up to the blastula stage). From the beginning of the early gastrula stage, these genes increased their transcript levels (Figure 4B; Dataset S19). Most of the genes encoding the proteasome subunits are involved in ubiquitin-mediated proteolysis (Figure 4C). Ubiquitin-mediated proteolysis and regulation of cell differentiation through the notch signaling pathway has been described previously [54]. In zebrafish, an involvement of the ubiquitin pathway in the degradation of paired-like homeobox gene Vsx1 has been reported [55]. From the present array data, significant expression of proteasome subunits could be detected from the late blastula stage. The expression patterns of proteasome subunits and ubiquitin conjugating enzymes present a global view of transcriptional regulation of genes behind the ubiquitin-dependent proteolytic pathway during zebrafish development.
Analysis of Gene Expression during Somitogenesis
Initiation of organogenesis in the embryo is accomplished by a set of structural genes and the transcription factors that modulate their expression. We have taken somitogenesis as an example and analyzed a subset of genes involved in this process. Broadly, somitogenesis can be divided into an early phase where the paraxial mesoderm is subdivided in a rostrocaudal pattern into blocks of cells called somites and a subsequent phase of cell differentiation within the somites where somatic cells acquire a variety of distinct dorsoventral and mediolateral fates. In zebrafish, the dorsolateral somitic mesodermal cells contribute to muscle development. Consistent with this, most of the muscle-specific protein (MSP) genes commenced expression during early somitogenesis, and they steadily increased their transcript abundance in positive correlation with somite formation. From the array expression data, the transcripts for MSP genes were first detected around 11 hpf (Figure 4D; Dataset S20), and, subsequently, their transcript levels increased in accordance with the progress of somitogenesis. We compared the expression of selected MSP genes from arrays with Northern blot analysis of the same genes reported earlier [56], and the results were identical. Therefore, array analysis faithfully recapitulated the data obtained from single gene analysis (Northern or WISH).
Distinct classes of genes encoding transcriptional regulators are expressed dynamically during somitogenesis in all vertebrates and precede the onset of MSP gene expression. The activities of members of the bHLH transcription factors, namely hairy (h), Enhancer of split (E spl), and hairy-related, have been shown to play critical roles during somitogenesis in vertebrates. Analysis of the zebrafish bHLH family of h/E (spl)-related genes has shown the involvement of hairy related 9 (her9-) [57], her1-, her4-, her6-, and her7- [58,59] in somitogenesis. Our array data clearly revealed that most of these transcription factors (mespa-, mespb-, mef2a-, mef2c-, mef2d-, foxc1a, and a number of her genes) commenced their expression prior to the onset of muscle-specific gene expression (Figure 4E). These results revealed that global analysis of gene expression using microarrays could detect a cascade of gene activity involved in somitogenesis in the zebrafish. Similar analyses could be extended to explore the expression patterns of genes involved in other organ systems.
RP Genes Are Coordinately Expressed during Zebrafish Embryogenesis
From the array expression data, 35 RP genes of zebrafish (similar to known RP genes) showed temporally coordinated expression (Figure 4F; Dataset S21). From the clustergram, it is evident that the level of transcripts of almost all the RP genes changed in concert except for a few, which differed slightly in their expression profile. On average, the onset of RP gene transcript accumulation commenced at the blastula stage and the expression increased continuously. Assembly of ribosomes requires coordinated expression of genes coding for their structural components, and are represented by several rRNA molecules and about 80 RPs. In eukaryotes, genes encoding rRNA and 5S RNA are amplified and presented by multiple copies, whereas the genes for RPs are present in only one or two copies per haploid genome. It has been reported that after fertilization and during the period of synchronous cell division, no transcripts were detected for zygotic genes [60]. This transcriptional block is terminated at MBT in concert with desynchronization of the cell cycle. It was shown that transcripts for some of the RPs (rps3, rpl17, and rpl31) [61] and rps24 [62] were detected at the early blastula stage prior to the detection of new zygotic transcripts in the sea urchin, suggesting their maternal origin. Published data on the expression of fish RP genes have not presented a comprehensive view [63,64]. However, our results presented here indicate that almost all the RP genes were coordinately expressed in a global and continuous fashion from the onset (beginning of MBT) to hatching.
Accumulation of Gene Transcripts Do Occur prior to the MBT in Zebrafish
Gene expression patterns during early development (pre-MBT and post-MBT stages) were studied by looking at the profiles of expression at selected time points during early cleavage and post-cleavage stages. The expression data obtained during these stages were selected following the statistical method described in the Materials and Methods section. Groups of genes showing distinct expression patterns were clustered based on the onset of transcript accumulation (Figure 5; Dataset S23–S25). Four distinct expression patterns were recorded. The first cluster represented maternally loaded transcripts (Dataset S22), and almost all the transcripts in this cluster degraded rapidly prior to blastula (4 hpf). Cluster 3 (Dataset S24) and Cluster 4 (Dataset S25) represented the expression of zygotic genes commencing the transcript accumulation after MBT stage (blastula and gastrula). This pattern of expression is in accordance with the pattern described earlier [30]. Cluster 2 contained a cohort of genes (Dataset S23) that commenced transcript accumulation prior to the onset of MBT. This expression pattern is a new discovery, demonstrating pre-MBT accumulation of zygotic transcripts. As transcript accumulation can take place only in the presence of gene transcription, these genes represent a group of pre-MBT transcribed genes. From a total of 16,416 genes analyzed in the array, 125 genes showed evidence for pre-MBT transcript accumulation. These genes did not show transcript accumulation during the one-cell, four-cell, and eight-cell stages; however, transcripts for this subset of 125 genes increased from the 64-/128-cell stage onwards. This pattern of pre-MBT zygotic gene expression has not been previously reported in zebrafish. We were the first to detect transcript accumulation at these early cleavage stages, owing to the unbiased and global nature of high-density microarrays.
Figure 5 Analysis of the Zebrafish Transcriptome during Pre-MBT and Post-MBT Stages of Embryonic Development
Transcriptional profiles of the genes during pre-MBT: one- to four-cell, 64-/128-cell, and post-MBT: blastula (4 hpf) and gastrula (6 hpf) stages were analyzed. This analysis revealed four different gene clusters. Clusters 1, 3, and 4 represent gene expression patterns, which have been identified earlier. Cluster 2 represents a novel pattern of gene expression commencing transcript accumulation at the 64-/128-cell stage onwards that represents the pre-MBT stage.
Since these novel gene transcripts were the earliest activated in zebrafish development, they are likely to be involved in specific early embryonic functions and possibly in subsequent genome-wide activation. We analyzed several of the genes in this cluster based on the predicted putative functions, which are as follows. Several proteasome subunits in this cluster are known to be involved in protein degradation by the ubiquitin-mediated proteolytic pathway [54,55]. The proteasomes may be required at this stage for the proteolytic degradation of maternal gene products. A number of proteasome subunits have been shown to be functional in the proliferating ARPE19 retinal pigment epithelium cells. Since the cleavage stage is similar to the proliferating cells stage, it is reasonable to find the proteasome transcripts at this stage. This cluster also contains the putative gene for ubiquitin conjugating enzyme, and it is known that ubiquitin conjugating enzymes mediate ubiquitination and degradation of specific substrates by the ubiquitin-dependent degradation pathway [65]. This enzyme has also been shown to promote cyclin proteolysis during mitosis [66]. We have reported in the present study that transcript levels of cyclins involved in cell cycle regulation (cyclin A2, cyclin B, cyclin B1, cyclin E) are highly abundant in unfertilized eggs and during cleavage stages [37,53]. For the degradation of these gene products, proteasomes and ubiquitin conjugating enzyme may be required at the synchronous cell cycles stage (pre-MBT) of zebrafish development.
Another gene in this cluster is a putative RNA helicase. RNA helicases are implicated in a number of cellular processes involving RNA secondary structures such as translation initiation and ribosome and spliceosome assembly. Some of the members of this family have been active in embryogenesis and in embryonic stem cells. RNA helicase has been shown to be essential for normal gastrulation in the mouse [67]. It has been shown that RNA helicase II is necessary for cell growth and cell cycle progression [68], and a role for RNA helicases in ribosomal RNA biogenesis in Xenopus oocytes has been observed [69]. In addition, it has been shown that down-regulation of RNA helicase II results in depletion of ribosomal RNA production in Xenopus oocytes and cell culture [69,70]. Thus, the expression of putative RNA helicase at the pre-MBT stage may be essential for cell growth and/or cell cycling during the cleavage stages of embryogenesis and also early commitment to normal gastrulation. Furthermore, it appears that the presence of RNA helicase II during early embryogenesis may facilitate ribosomal RNA production and the positioning of the translational machinery so as to meet the demand at MBT.
Other interesting genes in this cluster are alanyl-tRNA synthetase and nascent-polypeptide associated complex alpha (DNA primase small subunit). Transcripts for these genes have also been found in ES cells. DNA primase is involved in nascent polypeptide production resulting in protein biosynthesis, and tRNA synthetase is involved in protein synthesis related processes such as tRNA binding and tRNA ligase activity. Thus, all the genes described above are related to cell growth, cell proliferation, cell adhesion, nascent RNA synthesis and stability, and other cellular functions. It may be that these genes need to commence their function during pre-MBT stages in order to facilitate normal progression through MBT and subsequent embryonic development. GO analysis of the genes in this group reveals that more than 50% of the genes show putative functions such as DNA/RNA binding, development, protein folding, and protein biosynthesis (Figure S4). Human and mouse orthologs of some of the known genes in this group are given in Table S2. Clusters 3 and 4 displayed the cohort of zygotic genes displaying maximum transcript accumulation at blastula (4 hpf) and gastrula (6 hpf) stages, respectively. The genes that have been shown to be expressed in blastula- (for example, cyclin D1) and gastrula- (for example, cathepsin L) stage embryos were identified with Clusters 3 and 4, respectively, as expected. Thus, the microarray detected the expression of known genes at the expected time points, once again confirming the reliability of the array detection.
It has been shown that transcription can occur prior to MBT in Xenopus embryos [31]; namely, these authors showed that beta-catenin and tcf regulated pre-MBT transcription of nodal genes xnr5 and xnr6 does take place. However, in this study, they restricted the investigation to single gene expression analyses. Most of the earlier studies on the functional analysis of zebrafish genes were restricted to a handful of genes using WISH or Northern blot analysis with embryos at MBT or beyond. Parallel analysis of the expression of thousands of genes at selected closely spaced time points and early cleavage stages has not been previously attempted, to our knowledge. Thus, we are the first to our knowledge to undertake parallel analysis of expression of early genes using high-density zebrafish arrays. This approach revealed a cohort of genes that actively accumulate zygotic transcripts prior to the MBT stage, and the data presented here will serve as a starting point for the functional investigations of the roles of these genes during zebrafish embryogenesis.
To validate the expression profiles, real-time PCR was done for selected genes from each cluster. Real-time PCR was done for 16 samples from Cluster 2 (Figures 5 and 6) using the same total RNA used for array hybridization. The pattern of expression was similar in both analyses. Subsequently, we extracted new batches of RNA from one-cell, two-/four-cell, 64-/128-cell, 256-/512-cell, and 4-hpf embryos and performed real-time PCR for eight genes from Cluster 2 and compared the array data with RT-PCR data (Figure S5). Of these eight genes, real-time PCR was done for four genes using independently isolated batches of RNA, and the results are comparable. Quantity of transcripts for the selected genes at 64-/128-cell and 256-/512-cell (pre-MBT stages) is more than that at one- to four-cell stages indicating the accumulation of new transcripts prior to MBT. Similarly, real-time PCR was also done for selected genes from Clusters 3 and 4 (unpublished data). The results show that the pattern of expression obtained via real-time PCR analysis parallels the observations made in the array. Furthermore, whole-embryo RNA in situ hybridization was performed for eight different clones from the early genes of which real-time PCR was also done. Hybridization intensity was significantly higher at 64-/128-cell and 256-/512-cell compared to one- to four-cell stages, indicating transcript accumulation during the pre-MBT stages of development (Figure 7). The pattern of in situ data is similar to the real-time PCR profile. All these observations clearly support the presence of a cohort of early genes in zebrafish exhibiting pre-MBT transcript accumulation.
Figure 6 Validation of Zebrafish Transcriptome Pre-MBT and Post-MBT Results
Comparison between RT-PCR and microarray results for selected genes. Real-time PCR was done for 16 clones from Cluster 2 using the same RNA used for array hybridization. The RT-PCR profile closely parallels the microarray data, cross-validating both techniques as quantitative estimates.
Figure 7 WISH Was Performed for Selected Clones
All the clones subjected for in situ analysis showed a lower intensity of hybridization signal at one- to four-cell stages and increased intensity at 64-/128-cell and 256-/512-cell stages. These observations support the data obtained in the real-time PCR and microarray data, indicating the pre-MBT accumulation of transcripts.
In summary, we described here the first genome-wide microarray analysis of the embryonic zebrafish transcriptome. This study revealed a remarkable undulating developmental expression program and temporal clustering of gene groups never before characterized (to our knowledge) during embryogenesis. Furthermore, a previously unknown cohort of very early genes transcribed prior to MBT was discovered in this study, which now represents the earliest known (to our knowledge) transcribed and accumulating RNAs in zebrafish and revised our earlier thinking that MBT marked the onset of all zygotic gene transcription. We are the first to create a comprehensive integrated database by incorporating resources from different databases for the annotation of the clones in the zebrafish oligoarray (Compugen) that will serve as a significant resource for the zebrafish community.
Materials and Methods
Embryo collection.
Wild-type zebrafish (Singapore local stock) embryos were obtained from the zebrafish facility of the Institute of Molecular and Cell Biology. Embryos were collected immediately after fertilization, maintained at 28.5 °C, and staged by hpf using standard morphological criteria [71]. Unfertilized eggs were collected by squeezing the abdomen of spawning females. Embryos were collected at 12 time points (unfertilized egg, 3, 4.5, 6, 7.7, 9, 10.7, 12, 15, 24, 30, and 48 hpf) (Table S3), snap-frozen in liquid nitrogen, and stored at −80 °C. The 12 time points were assigned to the following developmental stages: (1) maternal (unfertilized egg), (2) blastula (3.0 and 4.5 hpf), (3) gastrula (6.0, 7.7, and 9.0 hpf), (4) segmentation (10.7, 12.0, and 15.0 hpf), and (5) pharyngula (24.0, 30.0, and 48.0 hpf). In order to maintain a uniform genetic background, all embryos were collected from the same batches of fish stock.
RNA isolation and reference RNA.
Total RNA was extracted from all the frozen embryos using Trizol reagent (Gibco BRL, Gaithersburg, Maryland, United States). RNA quality was evaluated by gel electrophoresis, and the concentration was measured with a UV spectrophotometer. To prepare reference RNA, total RNA was collected from the following stages (embryos/adults) and mixed in equal concentration: 1 hpf, 4 hpf, 24 hpf, 48 hpf, 3-wk-old fry, adult of male and female. Sufficient amount of reference RNA required for the entire project was prepared at one time, and the aliquots were stored at −80 °C.
Zebrafish oligonucleotide probe design and microarray construction.
Zebrafish oligonucleotide probes for this array were designed by Compugen and synthesized by Sigma-Genosys (The Woodlands, Texas, United States). For each gene, one 65-mer oligonucleotide probe was designed from the 3′ region of the sequence. Each probe was selected from a sequence segment that is common to a maximum number of splice variants predicted for each gene. The arrays contained 16,416 probes representing oligonucleotides of selected genes. Oligonucleotide probes were re-suspended in 3XSSC at 20 μM concentration and spotted onto poly-L-lysine-coated microscope slides using a custom-built DNA microarrayer. Printed arrays were post-processed following the standard procedure described for cDNA arrays [72].
Target labeling and hybridization strategy.
For fluorescence labeling of target cDNAs, 20 μg of total RNA from the reference and experimental samples was reverse transcribed in the presence of Cy3-dUTP and Cy5-dUTP (Amersham Biosciences, Little Chalfont, United Kingdom), respectively. Labeled target cDNAs were pooled, concentrated, and resuspended in DIG EasyHyb (Roche, Basel, Switzerland) buffer for hybridization. Hybridization strategies were as described by [72] with the following modifications: (1) usage of DIG EasyHyb buffer (Roche) and (2) hybridization at 42 °C. For each developmental stage, two to four independent replicate hybridizations were performed using a minimum of two biological replicates. For dye bias control, we employed a “dye-swapping” strategy for microarray hybridization whereby the Cy dye labeling scheme was reversed for alternating replicate hybridizations [73].
Data acquisition and statistical analysis.
The arrays were scanned using the GenePix 4000B microarray scanner (Axon Instruments, USA) to generate 16-bit TIFF images of Cy3 and Cy5 signal intensities. GenePix Pro 4.0 image analysis software (Axon Instruments, Union City, California, United States) was employed to measure the fluorescence signal intensities of the array features and local background.
Normalization of the two channels (sample and reference) was done for each slide using the intensity-based log ratio median method [32]. For selection of differentially expressed genes, which are supposed to be specific for a developmental stage, a two-step filtering process was used. First, only genes that differentially expressed at least at one time point (in all the arrays tested) were retained. Modified t-statistic (SAM) [33] with 15% of the standard deviation percentile as the fudge constant was used for identification of differentially expressed genes, and log2 ratios were used in the modified t-statistic. Predicted false discovery rate of 0.05 was used as the threshold for differential expression. At the next step, those differentially expressed genes were further filtered to determine whether they were specific for a particular stage. Paired t-test was used to check whether the means of log2 ratios of their expression levels (over the replicates) in a stage was significantly (at 95% confidence level) larger than the means at other stage(s). The above methods of analysis have been used for all the experimental data that are presented in this paper. The log ratios from replicate arrays were averaged, and the average expression values were used in subsequent analyses. The extracted datasets were hierarchically clustered and visualized (Cluster and Tree View; [74]).
Onset and offset finding algorithm.
The objective of the algorithm was to detect the onset (beginning) and peak time of transcript accumulation for a given gene using its time-course expression profile. Peak of the expression is defined as the time at which the mRNA abundance attains its highest level. The onset of expression is defined as the earliest time of the peak in the profile. This algorithm was used for the single peak genes selected as differentially expressed for this study. Prior to applying the algorithm, the dataset was selected based on statistical methods (SAM and FDR) described above, which removed the noisy profiles. Apart from that, the algorithm contains a Gaussian smoothing step that also removes the noisy profiles. Both onset and peak of expression are the onset and peak of relative abundance rather than absolute abundance. Manual verification of the profiles derived by using the algorithm showed that less than 0.1% of the data deviated from onset time by just one time point. The prediction of peak of expression did not vary from the observed data. Details of the algorithm are given in Protocol S1.
Annotation of the zebrafish oligoarray.
In order to obtain the most recent annotation, we tried to obtain the full-length sequences or assembled sequences for the clones in the array from different sources: Zebrafish gene collection (ZGC: http://zgc.nci.nih.gov/ZGC; ftp://ftp1.nci.nih.gov/pub/MGC/fasta); Zebrafish UniGene (UniGene build 79: ftp://ftp.ncbi.nih.gov/repository/UniGene/) and Zebrafish Gene Index ([ZGI] release 16: http://www.tigr.org/tigr-scripts/tgi/tc_ann.pl?db=zest
ftp://ftp.tigr.org/pub/data/tgi/Danio_rerio/). If the full-length sequence was not available, we obtained the assembled sequences from the zebrafish UniGene assembly (unique, longest sequences in the UniGene assembly) and from the ZGI (release 16), which provides current tentative consensus sequences generating assemblies of ESTs of particular gene. However, some of the clones in the array are represented as ESTs. Thus, we downloaded the sequences for the full-length clones, assembled sequences, and ESTs for all the probes in the array.
Putative annotations available for the clones in the array were downloaded from the following databases: (1) Compugen description (provided by oligo supplier: http://www.labonweb.com/oligo (2) Entrez Gene (for gene ID, gene description, and gene symbol: ftp://ftp.ncbi.nlm.nih.gov/gene/; http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=gene); (3) Zebrafish UniGene cluster ( for UniGene ID, UniGene description: ftp://ftp.ncbi.nlm.nih.gov/repository/UniGene/; http://www.ncbi.nlm.nih.gov/UniGene/clust); (4) GO http://www.geneontology.org release 21 November 2004; ftp://ftp.geneontology.org/pub/go/); (5) Locus link (ftp://ftp.ncbi.nih.gov/refseq/LocusLink/; release 21 November 2004); (6) UniGene protein similarity and description (for mouse and human protein similarity ID, percentage of similarity and description: ftp://ftp.ncbi.nlm.nih.gov/repository/UniGene/; http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=protein&dopt=GenPept&list); (7) ZGI (build 16) (for gene description and tentative consensus sequence number; ftp://ftp.tigr.org/pub/data/tgi/Danio_rerio/; http://www.tigr.org/tigr-scripts/tgi/tc_ann.pl?db=zest); (8) HomoloGene (mouse and human homolog gene ID and description; http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?DB=homologene; ftp://ftp.ncbi.nih.gov/pub/HomoloGene/); (9) Chromosomal location (ftp://ftp.ncbi.nih.gov/repository/UniGene/).
All the sequence data and annotation information are stored in a MySqL database and can be accessed at http://giscompute.gis.a-star.edu.sg/~govind/zebrafish/index.html details of retrieval are given in the help file. The database can be queried using the GenBank accession number of the probes in Compugen Zebrafish array and the following information can be obtained: (1) Compugen description, (2) Zebrafish UniGene ID (build 79), (3) Zebrafish UniGene description (build 79), (4) Entrez Gene description, (5) Entrez Gene ID and Gene symbol, (6) GO term, (7) Locus link, (8) UniGene protein similarity and description (mouse and human), (9) full-length or assembled sequences, (10) HomoloGene (human and mouse, build 38.1), and (11) chromosomal location of the gene. The annotation for all the GenBank entries in the Compugen array is available from the GIS-sponsored Web site designed for zebrafish microarray annotation (http://giscompute.gis.a-star.edu.sg/~govind/zebrafish/index.html). To our knowledge, we are the first to create such an integrated database and search facility for the zebrafish genes present in the array. This GIS zebrafish database will be updated periodically by integrating the latest release from the resource database.
From our expression analysis during embryogenesis, 3,657 differentially expressed genes were identified and annotated. There were no duplicates in the differentially expressed genes selected for the analysis. To obtain the annotation of these genes, the following procedures were adopted. Annotations/putative gene descriptions were obtained from the following databases by using GenBank accession number as query information: Entrez Gene, UniGene descriptions, protein similarity to human/mouse, and ZGI (build 16). From the above sources, we obtained annotation for about 75% of the clones. (Entrez Gene: 31%; UniGene, protein similarity, and ZGI: 44%). To annotate the rest of the clones, we BLASTed the assembled DNA sequence of the clones against the non-redundant database (ftp://ftp.ncbi.nlm.nih.gov/blast/db/ protein; ftp://www.expasy.ch/databases/sp_tr_nrdb/). Putative annotations were obtained for about 20% of clones from the non-redundant database. For the rest of the clones (5%), we got the description obtained from BLAST search (ftp://ftp.ncbi.nlm.nih.gov/blast/db/). By this process, we have obtained putative annotation for most of the differentially expressed genes. All of the resource data used in the annotations are available via our database (http://giscompute.gis.a-star.edu.sg/~govind/zebrafish/index.html
http://giscompute.gis.a-star.edu.sg/~govind/zebrafish/data_download.html) (Datasets S26 and S27). Based on the gene description and GO, the genes involved in selected functions were identified using appropriate keywords, such as “cell cycle,” “proteasomes,” “RPs,” and so on.
Northern blot hybridization and WISH analysis.
Total RNA isolated from embryos at various stages of development were fractionated on 1.2% formaldehyde-agarose gels and transferred to GeneScreen membranes following standard procedures. The blotted membranes were prehybridized and hybridized following standard protocols. Selected RP cDNA clones (l7a, s3a, s12, and s10) were used as templates for probe preparation and labeled with 32P using Random Primer DNA labeling system (Gibco BRL). Whole mount in situ hybridization using RNA probes labeled with digoxigenin (Roche) was carried out as previously reported [75].
Quantification by real-time PCR.
In order to verify the microarray results, two randomly chosen differentially expressed genes were tested in real-time RT-PCR analysis for all the time points. The SYBR Green I (Roche) RNA amplification kit was used on the LightCycler according to the manufacturer's instructions. The beta-actin and myosin light chain 2 (mlc2f) transcripts were amplified using the following primers (beta-actin: LC1–5′ CCGTGACATCAAGGAGAAGCT-3′; LC2 5′-TCGTGGATACCGCAAGATTCC-3′; mlc2f: LC1 5′-TCTCACTCATTACCCACAA-3′; LC2 5′-ACTCCATCGTGCTTCTTTC-3′). Prior to quantification, the optimal concentrations of template, primers, and magnesium were determined. Serially diluted plasmid DNA samples were used to construct a standard curve to quantify the test samples as well as the amplification efficiency. The products from real-time RT-PCR were also analyzed by agarose gel electrophoresis for single bands of the predicted size (unpublished data). The same protocol was followed for all real-time PCR analysis.
Analysis of early gene expression.
For the analysis of gene expression during early developmental time points, embryos were collected at the following cleavage and post-cleavage stages: one-cell, four-/eight-cell (mostly four-cell), 64-/128-cell (mostly 64-cell), 4 hpf, and 6 hpf. Total RNA extracted from unfertilized eggs was used as reference RNA in this analysis of gene expression. To perform the real-time PCR using the new set of RNA, total RNA was extracted from one-cell, two-/four-cell, 64-/128-cell, 256-/512-cell, and 4-hpf embryos and used for the experiment. To calculate ratio of expression for the real-time PCR data, real-time data of one-cell stage was used as reference. Normalization statistical analyses were done as described in the Materials and Methods.
Supporting Information
Dataset S1 Developmentally Regulated Genes with Expression Data
This dataset contains all the genes selected for subsequent analysis.
(688 KB DOC)
Click here for additional data file.
Dataset S2 Maternal Genes
Genes showing maximum transcript levels in unfertilized eggs and different patterns of degradation.
(137 KB DOC)
Click here for additional data file.
Dataset S3 List of Genes with Onset of Transcript Accumulation at the Blastula Stage and Peak of Expression at the Blastula Stage
(130 KB DOC)
Click here for additional data file.
Dataset S4 List of Genes with Onset of Transcript Accumulation at the Blastula Stage and Peak of Expression at the Gastrula Stage
(191 KB DOC)
Click here for additional data file.
Dataset S5 List of Genes with Onset of Transcript Accumulation at the Blastula Stage and Peak of Expression at the Segmentation Stage
(66 KB DOC)
Click here for additional data file.
Dataset S6 List of Genes with Onset of Transcript Accumulation at the Blastula Stage and Peak of Expression at the Pharyngula Stage
(44 KB DOC)
Click here for additional data file.
Dataset S7 List of Genes with Onset of Transcript Accumulation at Gastrula and Peak of Expression at Gastrula Stages
(27 KB DOC)
Click here for additional data file.
Dataset S8 List of Genes with Onset of Transcript Accumulation at Gastrula and Peak of Expression at Segmentation Stages
(51 KB DOC)
Click here for additional data file.
Dataset S9 List of Genes with Onset of Transcript Accumulation at Gastrula and Peak of Expression at Pharyngula Stages
(29 KB DOC)
Click here for additional data file.
Dataset S10 List of Genes with Onset of Transcript and Peak of Expression at the Segmentation Stage
(36 KB DOC)
Click here for additional data file.
Dataset S11 List of Genes with Onset of Transcript Accumulation at the Segmentation Stage and Peak of Expression at the Pharyngula Stage
(88 KB DOC)
Click here for additional data file.
Dataset S12 List of Genes with Onset of Transcript Accumulation and Peak of Expression at the Pharyngula Stage
(43 KB DOC)
Click here for additional data file.
Dataset S13 Genes Exhibiting Peak of Expression in the Unfertilized Egg (Maternally Loaded RNA)
(137 KB DOC)
Click here for additional data file.
Dataset S14 Genes Exhibiting Peak of Expression at the Blastula Stage
(130 KB DOC)
Click here for additional data file.
Dataset S15 Genes Exhibiting Peak of Expression at the Gastrula Stage
(199 KB DOC)
Click here for additional data file.
Dataset S16 Genes Exhibiting Peak of Expression at the Segmentation Stage
(142 KB DOC)
Click here for additional data file.
Dataset S17 Genes Exhibiting Peak of Expression at the Pharyngula Stage
(156 KB DOC)
Click here for additional data file.
Dataset S18 Cell-Cycle-Related Genes
(21 KB DOC)
Click here for additional data file.
Dataset S19 Proteasomes, Ubiquitins
(24 KB DOC)
Click here for additional data file.
Dataset S20 Somitogenesis-Related Genes
(24 KB DOC)
Click here for additional data file.
Dataset S21 RP Genes
(25 KB DOC)
Click here for additional data file.
Dataset S22 Gene Expression Dataset of Pre-MBT and Post-MBT Stages—Maternal Dominant Genes
(22 KB DOC)
Click here for additional data file.
Dataset S23 Gene Expression Dataset of Pre-MBT and Post-MBT Stages with Onset of Expression from 64 Cells Onwards
(39 KB DOC)
Click here for additional data file.
Dataset S24 Gene Expression Dataset of Pre-MBT and Post-MBT Stages with Onset of Expression from 4 hpf Onwards
(40 KB DOC)
Click here for additional data file.
Dataset S25 Gene Expression Dataset of Pre-MBT and Post-MBT Stages with Onset of Expression from 6 hpf Onwards
(100 KB DOC)
Click here for additional data file.
Dataset S26 Putative Annotation Obtained for the Selected 3,675 Genes Using the Zebrafish Annotation Database Generated in the Work also from the NR DB Blast Search
(3.5 MB XLS)
Click here for additional data file.
Dataset S27 Putative Annotation Obtained for the Selected Pre-MBT and Post-MBT Genes using the Zebrafish Annotation Database Generated in this Work
(525 MB XLS)
Click here for additional data file.
Figure S1 Expression Profile of the Beta-Actin Gene in the Array
All the copies of the oligos in the array showed an almost identical expression pattern, indicating the reproducibility and homogeneity of the array data.
(410 KB DOC)
Click here for additional data file.
Figure S2 Comparison of Expression Profiles Obtained for Myosin Light Chain (mlc2f) and Beta-Actin Genes Using Microarray Analysis and Real-Time PCR
The patterns of expression detected in both methods are almost identical.
(204 KB DOC)
Click here for additional data file.
Figure S3 Northern Blot Analysis of RP Genes
(A) Northern blot analysis to detect the transcript abundance for selected RP genes (l7a, s3a, s12, and s10) during embryogenesis. The expression pattern obtained by Northern blot analysis is similar to the data obtained from the array analysis.
(B) (I). Intensity of hybridization in the Northern blots for the selected RP gene was digitized and plotted. (II). Expression data obtained from the arrays for the selected RP genes used in Northern blotting are presented. Both analyses show similarity in the pattern of transcript accumulation.
(463 KB DOC)
Click here for additional data file.
Figure S4 Distribution of Early Gene Expression at Later Stages of Embryogenesis Based upon Peak Expression and Classification of Early Genes into Functional Groups Based on GO Terminology
(227 KB PDF)
Click here for additional data file.
Figure S5 Real-Time PCR Analysis of Early Genes
Real-time PCR was done using total RNA extracted from early stages of embryonic development (one-cell, two-/four-cell, 64-/128-cell, 256-/512-cell, and 4-hpf) for eight genes from Cluster 2 (pre-MBT transcribed genes). The expression pattern of these genes by real-time PCR and microarray are compared. The expression pattern is similar in both types of analysis.
(182 KB PDF)
Click here for additional data file.
Table S1 Comparison of the Expression Profile of Selected Genes in the Microarray Analysis and Other Methods of Transcript Detection
(11 KB PDF)
Click here for additional data file.
Table S2 Zebrafish Pre-MBT Transcribed (Early Gene) Orthologs in Mouse and Human
(9 KB PDF)
Click here for additional data file.
Table S3 Characteristics of Developmental Stages of Embryos That Were Collected for the Expression Profile Analysis in this Work
(24 KB PDF)
Click here for additional data file.
Protocol S1 Onset and Offset Finding Algorithm
(71 KB PDF)
Click here for additional data file.
Accession Numbers
The GenBank (http://www.ncbi.nlm.nih.gov/Genbank) accession numbers for the genes and gene products discussed in this paper are alanyl-tRNA synthetase (AI793498), beta-actin (AF025305), birc5b survivin2 (AY057058), calreticulin (BM101539), catalase (AF170069), cathepsin L (BI865754), ceruloplasmin (BC048037), cyclin B1 (BC055553), cyclin D1 (BI888360 and BC075743), cyclinE (BC045842), foxc1a (AF219949), her1- (X97329), her4- (AI794276), her6- (X97333), her7- (AF240772), her9- (AF301264), lmnl3 (AF397015), mcl1a (AF302805), mef2a- (BI880399, mef2c- (U66569), mef2d- (U66570), mespa- (AB037939), mespb- (AB037940), mlc2f (AF081462), nascent-polypeptide associated complex alpha (AI626587), RNA helicase (AW154620), rpl7a- (AI964218), rps10- (BI842921), rps12- (BM070699), rps3a- (AI558833), Vsx1 (AF025348), zp2 (AW170860), zp2.4 (AF331967), zp3 (AF095457), and BM184007.
We thank Aradhana Rani for help in data analysis, and Ben Zion Caveri for help with the Northern analysis. The financial support received from the Biomedical Research Council of Singapore is gratefully acknowledged.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. SM, LDM, VK, ZG, and ETL conceived and designed the experiments. SM, SGPL, AM, YT, YLW, and SHL performed the experiments. KRG performed the annotations. SM, LDM, KRKM, VK, ZG, ETL, and TL analyzed the data. KRKM, YR, VK, ZG, ETL, and TL contributed reagents/materials/analysis tools. SM, LDM, VK, ZG, ETL, and TL wrote the paper.
Abbreviations
ESTexpressed sequence tag
GOGene Ontology
hpfhours post-fertilization
MBTmid-blastula transition
MSPmuscle-specific protein
RPribosomal protein
WISHwhole embryo in-situ hybridization
ZGIZebrafish Gene Index
==== Refs
References
Dooley K Zon LI 2000 Zebrafish: A model system for the study of human disease Curr Opin Genet Dev 10 252 256 10826982
Dosch R Wagner DS Mintzer KA Runke G Wiemelt AP 2004 Maternal control of vertebrate development before the midblastula transition: Mutants from the zebrafish I Dev Cell 6 771 780 15177026
Wagner DS Dosch R Mintzer KA Wiemelt AP Mullins MC 2004 Maternal control of development at the midblastula transition and beyond: Mutants from the zebrafish II Dev Cell 6 781 790 15177027
Gaiano N Allende M Amsterdam A Kawakami K Hopkins N 1996 Highly efficient germ-line transmission of proviral insertions in zebrafish Proc Natl Acad Sci U S A 93 7777 7782 8755552
Amsterdam A Burgess S Golling G Chen W Sun Z 1999 A large-scale insertional mutagenesis screen in zebrafish Genes Dev 13 2713 2724 10541557
Golling G Amsterdam A Sun Z Antonelli M Maldonado E 2002 Insertional mutagenesis in zebrafish rapidly identifies genes essential for early vertebrate development Nat Genet 31 135 140 12006978
Long Q Meng A Wang H Jessen JR Farrell MJ 1997 GATA-1 expression pattern can be recapitulated in living transgenic zebrafish using GFP reporter gene Development 124 4105 4111 9374406
Nasevicius A Ekker SC 2000 Effective targeted gene ‘knockdown' in zebrafish Nat Genet 26 216 220 11017081
Zon LI 1999 Zebrafish: A new model for human disease Genome Res 9 99 100 10022974
Davis JM Clay H Lewis JL Ghori N Herbomel P 2002 Real-time visualization of mycobacterium-macrophage interactions leading to initiation of granuloma formation in zebrafish embryos Immunity 17 693 702 12479816
Fishman MC 2001 Genomics. Zebrafish—The canonical vertebrate Science 294 1290 1291 11701913
Amores A Force A Yan YL Joly L Amemiya C 1998 Zebrafish hox clusters and vertebrate genome evolution Science 282 1711 1714 9831563
Gong Z Yan T Liao J Lee SE He J 1997 Rapid identification and isolation of zebrafish cDNA clones Gene 201 87 98 9409775
Lo J Lee S Xu M Liu F Ruan H 2003 15000 unique zebrafish EST clusters and their future use in microarray for profiling gene expression patterns during embryogenesis Genome Res 13 455 466 12618376
Ton C Hwang DM Dempsey AA Tang HC Yoon J 2000 Identification, characterization, and mapping of expressed sequence tags from an embryonic zebrafish heart cDNA library Genome Res 10 1915 1927 11116087
Zeng S Gong Z 2002 Expressed sequence tag analysis of expression profiles of zebrafish testis and ovary Gene 294 45 53 12234666
Wang QT Piotrowska K Ciemerych MA Milenkovic L Scott MP 2004 A genome-wide study of gene activity reveals developmental signaling pathways in the preimplantation mouse embryo Dev Cell 6 133 144 14723853
Hamatani T Carter MG Sharov AA Ko MS 2004 Dynamics of global gene expression changes during mouse preimplantation development Dev Cell 6 117 131 14723852
Schena M Shalon D Heller R Chai A Brown PO 1996 Parallel human genome analysis: Microarray-based expression monitoring of 1000 genes Proc Natl Acad Sci U S A 93 10614 10619 8855227
White KP Rifkin SA Hurban P Hogness DS 1999 Microarray analysis of Drosophila development during metamorphosis Science 286 2179 2184 10591654
Kim JH Kim HY Lee YS 2001 A novel method using edge detection for signal extraction from cDNA microarray image analysis Exp Mol Med 33 83 88 11460886
DeRisi JL Iyer VR Brown PO 1997 Exploring the metabolic and genetic control of gene expression on a genomic scale Science 278 680 686 9381177
Arbeitman MN Furlong EE Imam F Johnson E Null BH 2002 Gene expression during the life cycle of Drosophila melanogaster
Science 297 2270 2275 12351791
Ton C Stamatiou D Dzau VJ Liew CC 2002 Construction of a zebrafish cDNA microarray: Gene expression profiling of the zebrafish during development Biochem Biophys Res Commun 296 1134 1142 12207891
Linney E Dobbs-McAuliffe B Sajadi H Malek RL 2004 Microarray gene expression profiling during the segmentation phase of zebrafish development Comp Biochem Physiol C Toxicol Pharmacol 138 351 362 15533793
Tomancak P Beaton A Weiszmann R Kwan E Shu S 2002 Systematic determination of patterns of gene expression during Drosophila embryogenesis Genome Biol 3 RESEARCH0088 12537577
Li TR White KP 2003 Tissue-specific gene expression and ecdysone-regulated genomic networks in Drosophila
Dev Cell 5 59 72 12852852
Klebes A Biehs B Cifuentes F Kornberg TB 2002 Expression profiling of Drosophila imaginal discs Genome Biol 3 RESEARCH0038 12186645
Furlong EE Andersen EC Null B White KP Scott MP 2001 Patterns of gene expression during Drosophila mesoderm development Science 293 1629 1633 11486054
Kane DA Kimmel CB 1993 The zebrafish midblastula transition Development 119 447 456 8287796
Yang J Tan C Darken RS Wilson PA Klein PS 2002 Beta-catenin/Tcf-regulated transcription prior to the midblastula transition Development 129 5743 5752 12421713
Yang H Haddad H Tomas C Alsaker K Papoutsakis ET 2003 A segmental nearest neighbor normalization and gene identification method gives superior results for DNA-array analysis Proc Natl Acad Sci U S A 100 1122 1127 12529501
Tusher VG Tibshirani R Chu G 2001 Significance analysis of microarrays applied to the ionizing radiation response Proc Natl Acad Sci U S A 98 5116 5121 11309499
Wang H Gong Z 1999 Characterization of two zebrafish cDNA clones encoding egg envelope proteins ZP2 and ZP3 Biochim Biophys Acta 1446 156 160 10395930
Tateno H Ogawa T Muramoto K Kamiya H Hirai T 2001 A novel rhamnose-binding lectin family from eggs of steelhead trout (Oncorhynchus mykiss) with different structures and tissue distribution Biosci Biotechnol Biochem 65 1328 1338 11471732
Tateno H Shibata Y Nagahama Y Hirai T Saneyoshi M 2002 Tissue-specific expression of rhamnose-binding lectins in the steelhead trout (Oncorhynchus mykiss)
Biosci Biotechnol Biochem 66 1427 1430 12162575
Kondo T Kotani T Yamashita M 2001 Dispersion of cyclin B mRNA aggregation is coupled with translational activation of the mRNA during zebrafish oocyte maturation Dev Biol 229 421 431 11150242
Davidson EH 1999 A view from the genome: Spatial control of transcription in sea urchin development Curr Opin Genet Dev 9 530 541 10508696
Edgar BA Datar SA 1996 Zygotic degradation of two maternal Cdc25 mRNAs terminates Drosophila's early cell cycle program Genes Dev 10 1966 1977 8756353
Edgar BA O'Farrell PH 1990 The three postblastoderm cell cycles of Drosophila embryogenesis are regulated in G2 by string Cell 62 469 480 2199063
Bashirullah A Halsell SR Cooperstock RL Kloc M Karaiskakis A 1999 Joint action of two RNA degradation pathways controls the timing of maternal transcript elimination at the midblastula transition in Drosophila melanogaster
Embo J 18 2610 2620 10228172
Tadros W Houston SA Bashirullah A Cooperstock RL Semotok JL 2003 Regulation of maternal transcript destabilization during egg activation in Drosophila
Genetics 164 989 1001 12871909
Audic Y Anderson C Bhatty R Hartley RS 2001 Zygotic regulation of maternal cyclin A1 and B2 mRNAs Mol Cell Biol 21 1662 1671 11238903
Audic Y Garbrecht M Fritz B Sheets MD Hartley RS 2002 Zygotic control of maternal cyclin A1 translation and mRNA stability Dev Dyn 225 511 521 12454927
Chen MC Gong HY Cheng CY Wang JP Hong JR 2000 Cloning and characterization of a novel nuclear Bcl-2 family protein, zfMcl-1a, in zebrafish embryo Biochem Biophys Res Commun 279 725 731 11118352
Kawamura K Sato N Fukuda J Kodama H Kumagai J 2003 Survivin acts as an antiapoptotic factor during the development of mouse preimplantation embryos Dev Biol 256 331 341 12679106
Yamashita M 2003 Apoptosis in zebrafish development Comp Biochem Physiol B Biochem Mol Biol 136 731 742 14662298
Murphy CR Sabel JL Sandler AD Dagle JM 2002 Survivin mRNA is down-regulated during early Xenopus laevis embryogenesis Dev Dyn 225 597 601 12454937
Ikegami R Hunter P Yager TD 1999 Developmental activation of the capability to undergo checkpoint-induced apoptosis in the early zebrafish embryo Dev Biol 209 409 433 10328930
Korzh S Emelyanov A Korzh V 2001 Developmental analysis of ceruloplasmin gene and liver formation in zebrafish Mech Dev 103 137 139 11335121
Rott NN Sheveleva GA 1968 Changes in the rate of cell divisions in the course of early development of diploid and haploid loach embryos J Embryol Exp Morphol 20 141 150 5699955
Yarden A Salomon D Geiger B 1995 Zebrafish cyclin D1 is differentially expressed during early embryogenesis Biochim Biophys Acta 1264 257 260 8547308
Yarden A Geiger B 1996 Zebrafish cyclin E regulation during early embryogenesis Dev Dyn 206 1 11 9019240
Itoh M Kim CH Palardy G Oda T Jiang YJ 2003 Mind bomb is a ubiquitin ligase that is essential for efficient activation of Notch signaling by Delta Dev Cell 4 67 82 12530964
Kurtzman AL Gregori L Haas AL Schechter N 2000 Ubiquitination and degradation of the zebrafish paired-like homeobox protein VSX-1 J Neurochem 75 48 55 10854246
Xu Y He J Wang X Lim TM Gong Z 2000 Asynchronous activation of 10 muscle-specific protein (MSP) genes during zebrafish somitogenesis Dev Dyn 219 201 215 11002340
Gajewski M Voolstra C 2002 Comparative analysis of somitogenesis related genes of the hairy/Enhancer of split class in Fugu and zebrafish BMC Genomics 3 21 12160468
Muller M v Weizsacker E Campos-Ortega JA 1996 Expression domains of a zebrafish homologue of the Drosophila pair-rule gene hairy correspond to primordia of alternating somites Development 122 2071 2078 8681788
Leve C Gajewski M Rohr KB Tautz D 2001 Homologues of c-hairy1 (her9) and lunatic fringe in zebrafish are expressed in the developing central nervous system, but not in the presomitic mesoderm Dev Genes Evol 211 493 500 11702199
Amaldi F Bozzoni I Beccari E Pierandrei-Amaldi P 1989 Expression of ribosomal protein genes and regulation of ribosome biosynthesis in Xenopus development Trends Biochem Sci 14 175 178 2672437
Pierandrei-Amaldi P Campioni N Beccari E Bozzoni I Amaldi F 1982 Expression of ribosomal-protein genes in Xenopus laevis development Cell 30 163 171 6889922
Angerer LM Yang Q Liesveld J Kingsley PD Angerer RC 1992 Tissue-restricted accumulation of a ribosomal protein mRNA is not coordinated with rRNA transcription and precedes growth of the sea urchin pluteus larva Dev Biol 149 27 40 1728593
Shemer R Eibschitz I Cavari B 2000 Isolation and characterization of medaka ribosomal protein S3a (fte-1) cDNA and gene Gene 250 209 217 10854794
Karsi A Patterson A Feng J Liu Z 2002 Translational machinery of channel catfish: I. A transcriptomic approach to the analysis of 32 40S ribosomal protein genes and their expression Gene 291 177 186 12095691
Haldeman MT Xia G Kasperek EM Pickart CM 1997 Structure and function of ubiquitin conjugating enzyme E2-25K: The tail is a core-dependent activity element Biochemistry 36 10526 10537 9265633
Yu H King RW Peters JM Kirschner MW 1996 Identification of a novel ubiquitin-conjugating enzyme involved in mitotic cyclin degradation Curr Biol 6 455 466 8723350
Lee CG da Costa Soares V Newberger C Manova K Lacy E 1998 RNA helicase A is essential for normal gastrulation Proc Natl Acad Sci U S A 95 13709 13713 9811865
Ou Y Fritzler MJ Valdez BC Rattner JB 1999 Mapping and characterization of the functional domains of the nucleolar protein RNA helicase II/Gu Exp Cell Res 247 389 398 10066367
Yang H Zhou J Ochs RL Henning D Jin R 2003 Down-regulation of RNA helicase II/Gu results in the depletion of 18 and 28 S rRNAs in Xenopus oocyte J Biol Chem 278 38847 38859 12851405
Henning D So RB Jin R Lau LF Valdez BC 2003 Silencing of RNA helicase II/Gualpha inhibits mammalian ribosomal RNA production J Biol Chem 278 52307 52314 14559904
Kimmel CB Ballard WW Kimmel SR Ullmann B Schilling TF 1995 Stages of embryonic development of the zebrafish Dev Dyn 203 253 310 8589427
Eisen MB Brown PO 1999 DNA arrays for analysis of gene expression Methods Enzymol 303 179 205 10349646
Miller LD Long PM Wong L Mukherjee S McShane LM 2002 Optimal gene expression analysis by microarrays Cancer Cell 2 353 361 12450790
Eisen MB Spellman PT Brown PO Botstein D 1998 Cluster analysis and display of genome-wide expression patterns Proc Natl Acad Sci U S A 95 14863 14868 9843981
Oxtoby E Jowett T 1993 Cloning of the zebrafish krox-20 gene (krx-20) and its expression during hindbrain development Nucleic Acids Res 21 1087 1095 8464695
|
16132083
|
PMC1193535
|
CC BY
|
2021-01-05 07:59:36
|
no
|
PLoS Genet. 2005 Aug 26; 1(2):e29
|
utf-8
|
PLoS Genet
| 2,005 |
10.1371/journal.pgen.0010029
|
oa_comm
|
==== Front
Malar JMalaria Journal1475-2875BioMed Central London 1475-2875-4-331603365310.1186/1475-2875-4-33ReviewThe promise and potential challenges of intermittent preventive treatment for malaria in infants (IPTi) O'Meara Wendy Prudhomme [email protected] Joel G [email protected] F Ellis [email protected] Division of Epidemiology and Population Studies, Fogarty International Center, National Institutes of Health, Bethesda, MD 20892 USA2005 20 7 2005 4 33 33 14 6 2005 20 7 2005 Copyright © 2005 O'Meara et al; licensee BioMed Central Ltd.2005O'Meara 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.
Intermittent preventive treatment (IPT) administers a full therapeutic course of an anti-malarial drug at predetermined intervals, regardless of infection or disease status. It is recommended by the World Health Organization (WHO) for protecting pregnant women from the adverse effects of malaria (IPTp) and shows great potential as a strategy for reducing illness from malaria during infancy (IPTi). Administered concurrently with standard immunizations, IPTi is expected to reduce the frequency of clinical disease, but to allow blood-stage infections to occur between treatments, thus allowing parasite-specific immunity to develop. While wide deployment of IPTi is being considered, it is important to assess other potential effects. Transmission conditions, drug choice and administration schedule will likely affect the possibility of post-treatment rebound in child morbidity and mortality and the increased spread of parasite drug resistance and should be considered when implementing IPTi.
==== Body
Background
Intermittent preventive treatment (IPT) protects pregnant women from malaria and placental parasitaemia and their newborns from malaria-associated low birth weight [1-6]. This strategy is being proposed to reduce malaria morbidity and mortality in infants (IPTi) in sub-saharan Africa by administration of a therapeutic course of an anti-malarial drug at predetermined intervals, regardless of infection, during routine vaccinations of infants in the Expanded Programme on Immunization (EPI). The intention is to clear any current infection and protect against new infection until the drug level decays in the bloodstream. Intervals between doses are typically longer than the time to clear the drug from the bloodstream (Figure 1), allowing the possibility of infection between doses. In infants, IPTi may simultaneously reduce the frequency of infection and life-threatening disease while allowing immunity to develop.
Figure 1 Concentration of anti-malarial drug in bloodstream during IPT. Drug concentration initially rises after administration of a therapeutic drug regimen. The concentration declines gradually as the drug is cleared from the bloodstream. Parasites with increasing sensitivity to the drug will be able to grow as the drug concentration declines. Parasites are divided roughly into three categories; fully sensitive, partially resistant and fully resistant. The interval between the time at which the concentration falls below the threshold for partially sensitive parasites and the time at which it falls below the threshold for sensitive parasites is a window of selection for resistant parasites. The dotted curve represents the target blood concentrations during chemoprophylaxis.
The context of IPTi is fundamentally distinct from that of IPTp: recipients of IPTp have already acquired some parasite-specific immunity, whereas infants are immunologically naïve and may comprise a disproportionately large infectious reservoir in the community [7-9]. A theoretical evaluation of the effect of drug pharmacokinetics on IPTi and IPTp efficacy in the face of rising drug resistance has recently been presented [10]. This paper discusses concerns related primarily to IPTi, specifically, whether: 1) immunity developed to malaria will be as robust in treated as in untreated individuals, forestalling a post-treatment increase in disease and mortality (rebound effect), 2) differences in transmission intensity should inform the timing of IPTi administration, 3) IPTi is likely to promote the spread of drug resistance, and 4) regional differences should inform the choice of IPTi drug.
Published IPTi studies
Two published studies have examined the effect of IPTi on malaria morbidity. Both were conducted in Tanzania and showed significant decreases in anaemia and clinical episodes of malaria in the treated group relative to the placebo group. In the study with sulfadoxine-pyrimethamine (SP), infants were given either a placebo or a drug three times with vaccinations (at two, three and nine months of age) and were followed until 22 months of age [11]. IPTi reduced the expected number of clinical cases of malaria in the first year of life by 59%. However, the effect declined significantly with time after the final dose. One month after the last dose – roughly the time to clear the final dose of SP from the bloodstream – the clinical case rates among infants in the control group and those in the treated group who had not previously contracted clinical malaria were indistinguishable and remained so until the end of the follow-up period. The follow-up to this study [12] reported that the protective effect of IPTi with SP extended well into the second year of life; however, only those children who contracted malaria during IPTi treatments experienced this long-term benefit. Those who did not contract malaria during IPTi treatments had malaria as frequently as placebo recipients who did not contract malaria. In the study with amodiaquine [13], infants three to four months old were given a full course of the drug every two months for six months and followed for four months after the last dose. The protective efficacy against clinical attacks of malaria was 65% during the six months of treatment. Some protection persisted throughout the follow-up period; by the end of the study, 35% of treated infants remained free from a first attack of malaria compared to 13% in the control group.
Thus, the results of the two published IPTi studies are encouraging. Two other studies [14,15] using strategies similar to IPTi failed to demonstrate statistically significant reductions in clinical episodes of malaria. Both administered SP at monthly intervals three times to anaemic children in Kenya. These children were older than those in the two IPTi studies (mean age at enrollment = 11 [14] and 20 months [15], versus 0 [11] and three [13] months), but still within the age range most at risk for malaria-related morbidity and mortality in these regions [16].
Immunity and rebound
An effective immune response to malaria is acquired slowly and cumulatively, requiring boosting; each subsequent infection generates increasing protection from severe illness, although the exact number of infections required to achieve a protected state is unknown. Of concern is whether administering anti-malarial drugs interrupts development of immunity and leads to decreased malaria-specific immunity and/or a rebound of increased morbidity and mortality after termination of treatment.
Rebound effects have been observed in several studies of chemoprophylaxis in children [17-21]. Using pyrimethamine-dapsone (P-D), Menendez et al [17] found that children given weekly doses during their first year had fewer malaria episodes during that year than did controls; in the year after termination of chemoprophylaxis, the frequency of cases in the treated group was twice that of the placebo group (Table 1). In a longer-term study, children were given P-D or placebo every two weeks from three months to five years of age or until the end of the five-year study, whichever was first [18,22]; at its conclusion, the study included children between one and ten years of age who had been treated for up to five years. The number of malaria episodes in the post-treatment group was greater than in the placebo group, and the probability of death from malaria between the age of five and six, immediately after termination of treatment, was slightly higher in the group that had received P-D for the full five years than in the placebo group. However, there was a 15% overall reduction in mortality in the treated group from age 0 to 10 years. Geerligs et al [23] review malaria chemoprophylaxis in children, including studies with and without rebound effects. Von Seidlein and Greenwood [24] offer a more general review of mass drug administration strategies including chemoprophylaxis and IPT.
Table 1 Comparison of studies during which a rebound effect was observed upon termination of drug treatment. Studies are in the order they are cited in the text.
Drug used (country) Ages of study group Duration of treatment Duration of post-intervention follow-up Effect on malaria morbidity Rebound effect Reference
P-D1 (Tanzania) 2 mo. at start of study Weekly for one year One year after termination of treatment Reduced incidence of clinical malaria by 40% during treatment period 80% higher incidence of clinical episodes in treated group during the year following termination of treatment [17]
P-D (The Gambia) 3 mo. at start of study Every 2 weeks for maximum of 5 years 5 years 65% reduction in malaria episodes after 3 years of chemoprophylaxis 52% more cases in treated group during the year following termination of treatment [18, 22]
SP2 + artesunate (The Gambia) Entire villages, all ages MDA3 1 dose 20 weeks Reduced rate of malaria attacks in children <11 yr by 60% Rate of clinical malaria was 69% higher in treated groups 3 months after treatment [25]
SP (Mali) 3 mo. to 20 years MDA 1 dose 24 weeks Reduced incidence of first malaria episode from 26% to 3% during first month Incidence of first malaria episodes in treated group rose to 42% compared to 17% (untreated group) during the third month after treatment [26]
1P-D: pyrimethamine-dapsone
2SP: sulfadoxine-pyrimethamine
3MDA: mass drug administration
Among the implicit assumptions of the IPTi strategy are that both the number of serious clinical episodes and the development of malaria-specific immunity are directly proportional to the frequency of infection. If these assumptions are true, intermittent doses of drug could reduce the frequency of serious malaria episodes, while the intervals between doses of drug could permit infection and the development of immunity. Thus, if suppressed development of immunity is responsible for any rebound effects from chemoprophylaxis, the IPTi strategy could reduce the likelihood of rebound, provided that infectious bites occur frequently enough to ensure exposure between doses. There is reason to doubt whether rebound effects are proportional to the frequency of infection and immunity in infants and children, however. In the five-year study with P-D cited above, the group that had received the drug for the full five years had levels of malaria-specific antibodies equivalent to those in the control group [18]. Nonetheless, they experienced more disease episodes and mortality during the first year post-chemoprophylaxis than those with lower antibody titres. Parasite-specific antibody titres often correlate with exposure, but not protection from disease.
The relationship between frequency of infection, immunity and rebound is further complicated by evidence that very short-term drug administration can lead to significant rebound effects (Table 1). During a one-time mass administration of SP-plus-artesunate to entire villages in The Gambia at the start of the rainy season [25], the frequency of malaria attacks among children under 11 years of age in treated villages was 60% lower than in control villages during the first six weeks. During the third month post-treatment, disease incidence doubled in children in the treated villages relative to the placebo villages. In another study [26], two groups of individuals from the same town in Mali, aged three months to 20 years, were treated with a single dose of SP or placebo. In the first four weeks, 5% of the treated group had at least one malaria episode compared to 25% of the control group. During weeks nine to 12, the incidence was twice as high in the SP group. These last examples included much wider age ranges than would IPTi, but they indicate that rebound can happen over very short time periods, with only a single dose of drug, signaling the possibility that IPTi could show similar effects.
Neither of the published IPTi studies reported a rebound after termination of treatment, even after extended follow-up. However, a significant rebound of malaria morbidity was observed in an IPTi trial conducted in Navrongo, Ghana (D. Chandramohan, personal communication). In this study, the number of rebound episodes apparently did not negate the benefit of IPTi. Even so, it is imperative that these effects be measured and reported prior to community-wide implementation of IPTi.
Transmission effects: timing of IPTi administration
It is important that the conditions for which the current conception of IPTi within the EPI delivery system will be beneficial be defined as clearly as possible so that resources for malaria control are allocated to the most effective programmes in each region. Under conditions of intense, perennial transmission, infants may be susceptible to severe clinical attacks from a few months after birth, and antimalarials given during routine EPI visits at two, three and nine months of age are likely to be effective in reducing episodes of malaria. Where transmission is less intense or unstable, children are at great risk until 5 years of age or older. As transmission rates decline with increasing urbanization and bednet usage in sub-Saharan Africa [27,28], a larger proportion of the population will be at greatest risk after one year of age. For example, the average age of first malaria infection in Prampram, Ghana [29], was estimated to be 10.5 months; here, IPTi within EPI (final dose at nine months) is unlikely to impact malaria morbidity. Thus, implementing IPTi uniformly and exclusively at the time of EPI vaccinations may be inadequate in many circumstances.
If the aim is to administer IPTi during the time of greatest risk of severe disease from unprotected exposure to infection, age-incidence data for severe malaria will be needed to target those most likely to benefit in a given location (Figure 2). In low and moderate transmission areas, IPT may be most effective if the schedule is extended into early childhood. Highly seasonal malaria may require unique strategies which target the appropriate age groups during the transmission season. If mechanisms for delivery can be established that allow greater flexibility in the window of administration, IPTi could have a much broader impact. A study of IPT in children under five in an area of highly seasonal transmission administered doses of SP at monthly intervals exclusively during the high transmission season. This approach reduced the incidence of clinical malaria by more than 80% in the treated group compared to the control group, indicating that adapting the IPT schedule can be accomplished with good results [30].
Figure 2 Effect of transmission intensity on outcome of IPTi. Qualitative depiction of the effect of transmission intensity on the age-incidence pattern of malaria morbidity and mortality. In areas of high transmission (solid line), malaria mortality is concentrated in the first two years of life. In areas of low or unstable transmission (dashed line), children may be at risk until much later in life. The arrows represent the window of greatest risk or the age interval during which roughly 75% of childhood malaria episodes are experienced.
The relationship between the age interval during which IPTi is administered and that of greatest risk will determine the overall benefit of IPTi, i.e. reduction in the total number of malaria cases. In the published SP study [11], prevalence among 12-month old infants was only 4% (annual Entomological Inoculation Rate = 29 [31]) whereas in the amodiaquine study the prevalence in 3.5 month-olds was 24–34% (EIR = 400). Although protective efficacy, or the percentage reduction in first clinical episodes, was similar between the two studies (59% and 65%), only 0.26 cases per person-year at risk were prevented in the SP study versus 1.49 cases in the amodiaquine study.
The frequency and clustering of infectious bites determines whether the time interval between doses will permit the development of immunity, and whether IPTi will act primarily to protect against infection or treat existing infections (Figure 3). Exposure to infectious bites must be sufficiently frequent, or IPT doses sufficiently spaced, to allow infection between periods of protection. Although the relationship between infection and development of clinical immunity remains poorly understood, it is clear that if no infectious bites are received between treatments, then IPTi will be functionally equivalent to chemoprophylaxis.
Figure 3 Effect of transmission intensity on outcome of IPTi. Effect of biting frequency (arrows represent infectious bites) on the development of malaria-specific immunity during IPTi. The frequency of infectious bites determines whether or not an infant is exposed to malaria parasites between doses of drug during IPTi.
Transmission intensity may vary considerably over very short distances and short time periods. While it will often be impractical to vary drug administration schedules in response, broader assessments of local transmission and disease patterns should guide IPTi schedules.
Drug resistance
IPTi has the potential to both affect and be affected by parasite resistance to antimalarial drugs. The possible relationship between drug resistance and the efficacy of IPTi has been discussed elsewhere [10]. Paradoxically, truncating an infection may interrupt the development of protective immunity leading to speculation about whether drugs which do not completely clear an infection may be more effective for IPTi. Data about frequencies of resistance at or near IPTi study sites (see below) will shed light on this relationship. The impact of IPTi on the spread of drug resistance will be difficult to assess, particularly where the study group is a small fraction of the community and the drug studied is also used for treatment of malaria episodes.
Drug pharmacokinetics
After administration of an anti-malarial drug, the concentration in the bloodstream initially peaks and then gradually declines. During successful treatment, parasites are damaged or killed by the drug and cleared by the immune system. An infection may include a mixture of genetically heterogeneous parasites with varying drug sensitivities. Drug-sensitive parasites are completely cleared by drug treatment and cannot grow in the presence of low concentrations of the drug. Fully resistant parasites survive "therapeutic" doses of a drug, whereas partially resistant parasites can grow in the presence of a drug, but only when the concentration drops below a critical threshold level (Figure 1). While the concentration remains above the critical threshold for sensitive parasites, resistant parasites are selected and competition from sensitive parasites is removed [32]. For instance, Watkins et al [33] showed that infections detected during the clearance of SP, but before SP concentration drops to zero (15–52 days post-treatment), are more likely to be pyrimethamine resistant. The spread of drug resistance is promoted by increased transmission of resistant strains relative to sensitive strains in the interval following drug treatment and before the immune system succeeds in clearing parasites [34] (Figure 4a).
Figure 4 Drug treatment and immune mechanisms act synergistically to eliminate parasites. (a) A patient with some pre-existing immunity clears infection and reduces the number and transmission of the parasites that survive treatment. (b) Without pre-existing immunity, the window for proliferation and differential transmission of resistant parasites is extended. White and black circles represent sensitive and resistant parasites, respectively.
During chemoprophylaxis, the intention is to administer the drug often enough to prevent plasma concentrations from falling below the survival threshold for resistant parasites (Figure 1). Lack of compliance resulting in fluctuations in drug concentrations increases parasite exposure to sub-preventive/sub-curative levels and promotes the development and spread of drug-resistant parasites. Compared to chemoprophylaxis, IPTi offers the advantage of decreased drug pressure through supervised doses that target a specific, relatively small age group. It has the disadvantage of causing cyclical fluctuations of drug titres ideal for repeated selection of resistant parasites.
IPTi studies have chosen drugs with long half-lives to maximize the prophylactic window of each dose. However, long clearance times also lead to long periods during which selective levels of drug are present in the bloodstream, and, when these drugs are administered to infants, the parasites are exposed to selective concentrations in an immunologically naive individual. With little or no pre-existing immunity, the host is less able to control an infection, so the potential for growth and transmission of resistant parasites is enhanced (Figure 4b). Preliminary results from two studies with SP in children under five years of age show increased frequencies of infections with SP-resistant parasites among children receiving IPT [30,35]. It is not known how much this increase will increase the community-wide transmission and prevalence of drug resistance, but several studies indicate that infants and young children are the most important infectious reservoir in a population, despite their minority [7-9]. Prospective, community-wide, cross-sectional surveys of molecular markers of parasite resistance before and after implementation of IPTi could address this question. Of course, in order to differentiate between selective pressure due to conventional treatment of malaria in the community and that due to IPTi, a drug not routinely given for febrile episodes must be chosen for IPTi in such a study.
Genetics of resistance
In infants, chemoprophylaxis decreases the multiplicity of infections (MOI), the number of distinguishable genotypes based on a single locus, present at any time in a single host [36]. Assuming proportional transmission, with fewer genotypes per host there is less potential for mating between parasites with different alleles at resistance loci. If resistance is conferred by alleles at two or more loci, inbreeding reduces the opportunity for the resistance alleles to be separated, and thus the probability that offspring will be susceptible. Parasite inbreeding increases the survival and prevalence of resistance genes [34].
Alternatively, IPT could increase antigenic diversification of drug-resistant strains (Figure 5). A treatment dose will kill all the sensitive parasites but may leave a few resistant parasites to multiply as the drug concentration declines. When the drug is completely cleared, the host becomes vulnerable to infection by drug-sensitive parasites and so may acquire multiple strains which can cross with resistant parasites. The total effect is to select for resistant parasites, allow them to multiply without competition and provide material for antigenic diversity in the next generation, via cross-breeding in the mosquito, before administering the drug again and once more selecting for the resistant parasites.
Figure 5 Effect of IPT on parasite diversity. Blood concentration of drug during two treatments of IPT is shown. All of the sensitive parasites (white circles) initially present in the host at the start of IPT are killed by the drug, leaving only resistant parasites (black circles). While the drug concentration is high, new infections are prevented and the resistant parasites proliferate without competition. When the drug concentration falls below the threshold for sensitive parasites, an infectious bite can produce an infection with sensitive parasites. A subsequent bloodmeal may take up gametocytes from both sensitive (white crescents) and resistant strains (black crescents), thereby allowing out-crossing between them. At the next treatment, the sensitive parasites are killed and the resistant parasites continue to proliferate and may acquire new genetic material at the next cycle. Black arrows at the top represent continual biting by infectious mosquitoes.
It is not known how IPTi would affect the dynamics of multi-strain infections, or whether IPTi would increase parasite inbreeding, and the prevalence of resistance genes, or increase out-crossing and the antigenic diversity of resistant populations. Transmission intensity will determine which, if any, of these outcomes is observed. Because the spread of resistant parasites could be significantly affected in either case, these potential outcomes of IPTi should be considered in designing implementation schemes.
Drug choice
Whether or not a particular drug is effective may be determined by the mechanism of protection of IPTi. If the prophylactic period is the primary protective mechanism, then drugs which are cleared very quickly, such as artemisinin, are not likely to be useful, either alone or in combination with other drugs. If the primary mechanism is clearance of existing infections, then drugs with very short half-lives should be very effective and would decrease the probability of accelerating the spread of resistance. Studies in mice have shown that suppression of blood-stage infection enhances development of protective immunity against liver-stage infection [37,38]. If the same is true in human hosts and if infectious bites occur between doses of drug, IPTi could protect by enhancing liver-stage immunity. In this case, drugs active against the liver stage of the parasite would not be effective. Several drug candidates with both long and intermediate persistence times, such as chlorproguanil/dapsone (Lapdap), SP with artesunate, amodiaquine with artesunate and mefloquine, are being tested in ongoing IPTi trials .
When choosing from among the effective drug candidates, the context in which IPTi will be implemented should also be considered, including the regional level of resistance to drugs available for IPTi and the community use of drugs for treatment of symptomatic malaria. If infants who present with clinical malaria during IPTi are treated with the same drug that is used for IPTi, the spread of drug-resistant parasites may be accelerated: an infection that appears during IPTi is more likely to be resistant to some level of that drug, so dual usage could lead to an increased selection and spread of highly resistant parasites.
The studies which will be completed in the near-term are using SP and it seems likely that SP will be the drug chosen for initial implementation of IPTi. However, an anti-folate drug closely related to SP, cotrimoxazole (trimethoprim-sulfamethoxazole), is currently recommended for prophylaxis against opportunistic infections in HIV-positive individuals [39] and is widely used in infants and children in malaria-endemic areas where IPTi could be implemented. Cross-resistance to SP and cotrimoxazole has been reported in Plasmodium falciparum [40] and may have serious consequences for malaria-infected patients undergoing chemoprophylaxis with cotrimoxazole. Where there is a high prevalence of HIV among infants at risk for malaria, the effect of cross-resistance to cotrimoxazole on efficacy of IPTi with SP must be evaluated and vice-versa.
Studies underway
There are several clinical trials of IPTi underway, six of which are being conducted within the IPTi Consortium . The annual EIRs among these six sites range from five to 200 infectious bites per person, allowing comparison of efficacy data from a relatively broad range of transmission conditions. However, if IPTi is uniformly administered at two, three and nine months of age, no information about the broader applicability of IPTi will be available from these studies. For example, if IPTi delivered through EPI offers no significant protection where the EIR is less than 50, it will be desirable to know if different schedules might be effective. New studies would need to be designed to answer this question.
Since an increase in drug resistant infections [30,35] and a rebound in malaria episodes (D. Chandramohan, personal communication) have been observed in IPTi studies, these possible adverse outcomes should be measured in further studies. Concerns have also been raised about the effects of IPTi on post-vaccination seroconversion [41]. If markers of immunity, EPI seroconversion and drug resistance data can be integrated in a meta analysis across these diverse trials, this would provide considerable evidence for evaluating the safety of this scheme of IPTi implementation. This effort would require extensive cooperation and standardized methods, but the result would be an unprecedented ability to effectively compare an intervention across trials in diverse malaria transmission zones. Information available online indicates that the IPTi Consortium is coordinating its cost-effectiveness and drug safety studies in precisely this manner.
Conclusion
IPTi is a promising new intervention strategy which aims to combine the short-term protective mechanisms of chemoprophylaxis with the long-term protection of naturally acquired immunity. It is certain that for some age interval, over some time interval, IPTi would reduce the clinical malaria case rate, whether the drug is acting as prophylaxis, or to clear sub-clinical infections, or both. However, the price to be paid in the broader community, for example in immunity and drug resistance, will be greatly affected by how well IPTi deployment addresses the local context of transmission and drug efficacy. While it is likely that IPTi linked to EPI immunization will be salutary in areas with intense stable transmission, much of Africa is moving toward reduced or unstable endemicity through urbanization and use of insecticide impregnated bednets and other vector control strategies. If bednet use prevents infections during the period of IPTi treatment (i.e. the first year of life), then IPTi will not add protection. Furthermore, if, in fact, the extended protective effect of IPTi depends on becoming infected during that period, IPTi programmes in areas of high bednet usage may not show the expected effects on malaria morbidity. The efficacy and practicality of IPTi under different schedules of administration should be explored.
To help verify that the intention of IPTi is realized, the effects of IPTi on malaria-specific acquired immunity should be measured and follow-up periods should include sufficient time for rebound to be evaluated. More thorough assessment of the potential impact of IPTi on the spread of drug resistance is also necessary. Amplification of drug resistance may be exacerbated by special factors connected with regimented, symptom-blind drug use in immunologically naive infants. Site-sensitive choices of drug and scheduling would help to minimize the amplification and maximize the chance that it will eventually be counter-balanced by decreased drug use later in life.
Authors' contributions
The manuscript was written by W.P.O. All authors contributed to the concepts and ideas presented and to the editing of the final manuscript.
Conflict of interest statement
The author(s) declare that they have no competing interests.
Acknowledgements
The authors would like to thank Brian Greenwood for helpful comments and feedback and David L. Smith for helpful discussion and critical review of this manuscript. No external funding sources contributed to this work.
==== Refs
Verhoeff FH Brabin BJ Chimsuku L Kazembe P Russell WB Broadhead RL An evaluation of the effects of intermittent sulfadoxine-pyrimethamine treatment in pregnancy on parasite clearance and risk of low birthweight in rural Malawi Ann Trop Med Parasitol 1998 92 141 150 9625909 10.1080/00034989859979
Parise ME Ayisi JG Nahlen BL Schultz LJ Roberts JM Misore A Muga R Oloo AJ Steketee RW Efficacy of sulfadoxlne-pyrimethamine for prevention of placental malaria in an area of Kenya with a high prevalence of malaria and human immunodeficiency virus infection Am J Trop Med Hyg 1998 59 813 822 9840604
Shulman CE Dorman EK Cutts F Kawuondo K Bulmer JN Peshu N Marsh K Intermittent sulphadoxine-pyrimethamine to prevent severe anaemia secondary to malaria in pregnancy: a randomised placebo-controlled trial Lancet 1999 353 632 636 10030329 10.1016/S0140-6736(98)07318-8
Rogerson SJ Chaluluka E Kanjala M Mkundika P Mhango C Molyneux ME Intermittent sulfadoxine-pyrimethamine in pregnancy: effectiveness against malaria morbidity in Blantyre, Malawi, in 1997–99 Trans R Soc Trop Med Hyg 2000 94 549 553 11132387 10.1016/S0035-9203(00)90083-X
van Eijk AM Ayisi JG ter Kuile FO Otieno JA Misore AO Odondi JO Rosen DH Kager PA Steketee RW Nahlen BL Effectiveness of intermittent preventive treatment with sulphadoxine-pyrimethamine for control of malaria in pregnancy in western Kenya: a hospital-based study Trop Med Int Health 2004 9 351 360 14996364 10.1111/j.1365-3156.2004.01196.x
Kayentao K Kodio M Newman RD Maiga H Doumtabe D Ongoiba A Coulibaly D Keita AS Maiga B Mungai M Parise ME Doumbo O Comparison of intermittent preventive treatment with chemoprophylaxis for the prevention of malaria during pregnancy in Mali J Infect Dis 2005 191 109 116 15593011 10.1086/426400
Bonnet S Gouagna LC Paul RE Safeukui I Meunier JY Boudin C Estimation of malaria transmission from humans to mosquitoes in two neighbouring villages in south Cameroon: evaluation and comparison of several indices Trans R Soc Trop Med Hyg 2003 97 53 59 12886806 10.1016/S0035-9203(03)90022-8
Akim NIJ Drakeley C Kingo T Simon B Senkoro K Sauerwein RW Dynamics of P. falciparum gametocytemia in symptomatic patients in an area of intense perennial transmission in Tanzania Am J Trop Med Hyg 2000 63 199 203 11388515
Drakeley CJ Akim NIJ Sauerwein RW Greenwood BM Targett GAT Estimates of the infectious reservoir of Plasmodium falciparum malaria in The Gambia and in Tanzania Trans R Soc Trop Med Hyg 2000 94 472 476 11132369 10.1016/S0035-9203(00)90056-7
White NJ Intermittent Presumptive Treatment for Malaria PLoS Med 2005 2 e3 15696210 10.1371/journal.pmed.0020003
Schellenberg D Menendez C Kahigwa E Aponte J Vidal J Tanner M Mshinda H Alonso P Intermittent treatment for malaria and anaemia control at time of routine vaccinations in Tanzanian infants: a randomised, placebo-controlled trial Lancet 2001 357 1471 1477 11377597 10.1016/S0140-6736(00)04643-2
Schellenberg D Menendez C Aponte JJ Kahigwa E Tanner M Mshinda H Alonso P Intermittent preventive antimalarial treatment for Tanzanian infants: Follow-up to age 2 years of a randomised, placebo-controlled trial Lancet 2005 365 1481 1483 15850632 10.1016/S0140-6736(05)66418-5
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
Verhoef H West CE Nzyuko SM de Vogel S van der Valk R Wanga MA Kuijsten A Veenemans J Kok FJ Intermittent administration of iron and sulfadoxine-pyrimethamine to control anaemia in Kenyan children: a randomised controlled trial Lancet 2002 360 908 914 12354473 10.1016/S0140-6736(02)11027-0
Desai MR Mei JV Kariuki SK Wannemuehler KA Phillips-Howard PA Nahlen BL Kager PA Vulule JM ter Kuile FO Randomized, controlled trial of daily iron supplementation and intermittent sulfadoxine-pyrimethamine for the treatment of mild childhood anemia in western Kenya J Infect Dis 2003 187 658 666 12599083 10.1086/367986
Bloland PB Boriga DA Ruebush TK McCormick JB Roberts JM Oloo AJ Hawley W Lal A Nahlen B Campbell CC Longitudinal cohort study of the epidemiology of malaria infections in an area of intense malaria transmission II. Descriptive epidemiology of malaria infection and disease among children Am J Trop Med Hyg 1999 60 641 648 10348241
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
Greenwood BM David PH OtooForbes LN Allen SJ Alonso PL Schellenberg JRA Byass P Hurwitz M Menon A Snow RW Mortality and morbidity from malaria after stopping malaria chemoprophylaxis Trans R Soc Trop Med Hyg 1995 89 629 633 8594677 10.1016/0035-9203(95)90419-0
Saarinen M Thoren E Iyambo N Carlstedt A Shinyafa L Fernanda M Paajanen H Paajanen K Indongo I Rombo L Malaria prophylaxis with proguanil to Namibian refugee children in Angola Trop Med Parasitol 1988 39 40 42 3387827
Bjorkman A Brohult J Pehrson PO Willcox M Rombo L Hedman P Kollie E Alestig K Hanson A Bengtsson E Monthly antimalarial chemotherapy to children in a holoendemic area of Liberia Ann Trop Med Parasitol 1986 80 155 167 3530156
Oyediran AB Tropley E Osunkoya BO Bamgboye A Williams AIO Ogunba EO Severe morbidity among children in a trial of malaria chemoprophylaxis with pyrimethamine or chloroquine in Ibarapa, Nigeria Afr J Med Sci 1993 22 55 63
Greenwood BM Bradley AK Byass P Greenwood AM Snow RW Hayes RJ Njie ABH Comparison of two strategies for control of malaria within a primary health-care programme in the Gambia Lancet 1988 1 1121 1127 2896957 10.1016/S0140-6736(88)91949-6
Geerligs PDP Brabin BJ Eggelte TA Analysis of the effects of malaria chemoprophylaxis in children on haematological responses, morbidity and mortality Bull World Health Organ 2003 81 205 216 12764517
von Seidlein L Greenwood BM Mass administrations of antimalarial drugs Trends in Parasitology 2003 19 452 460 14519583 10.1016/j.pt.2003.08.003
von Seidlein L Walraven G Milligan PJM Alexander N Manneh F Deen JL Coleman R Jawara M Lindsay SW Drakeley C De Martin S Olliaro P Bennett S van der Loeff MS Okunoye K Targett GAT McAdam K Doherty JF Greenwood BM Pinder M The effect of mass administration of sulfadoxine-pyrimethamine combined with artesunate on malaria incidence: a double-blind, community-randomized, placebo-controlled trial in The Gambia Trans R Soc Trop Med Hyg 2003 97 217 225 14584381 10.1016/S0035-9203(03)90125-8
Coulibaly D Diallo DA Thera MA Dicko A Guindo AB Kone AK Cissoko Y Coulibaly S Djimde A Lyke K Doumbo OK Plowe CV Impact of preseason treatment on incidence of falciparum malaria and parasite density at a site for testing malaria vaccines in Bandiagara, Mali Am J Trop Med Hyg 2002 67 604 610 12518850
Omumbo JA Guerra CA Hay SI Snow RW The influence of urbanisation on measures of Plasmodium falciparum infection prevalence in East Africa Acta Tropica 2005 93 11 21 15589793 10.1016/j.actatropica.2004.08.010
Hay SI Guerra CA Tatem AJ Atkinson PM Snow RW Urbanization, malaria transmission and disease burden in Africa Nature Reviews Microbiology 2005 3 81 90 15608702 10.1038/nrmicro1069
Wagner G Koram K McGuiness D Bennett S Nkrumah F Riley E High incidence of asymptomatic malaria infections in a birth cohort of children less than one year of age in Ghana, detected by multicopy gene polymerase chain reaction Am J Trop Med Hyg 1998 59 115 123 9684638
Cisse B Sokhna C Alexander N Lines J Trape J-F Greenwood B Intermittent preventive treatment (IPT) with artesunate plus sulfadoxine pyrimethamine to reduce under 5 malaria morbidiy in the Sahel Am J Trop Med Hyg 2004 71 1 79
Drakeley C Schellenberg D Kihonda J Sousa CA Arez AP Lopes D Lines J Mshinda H Lengeler C Schellenberg JA Tanner M Alonso P An estimation of the entomological inoculation rate for Ifakara: a semi-urban area in a region of intense malaria transmission in Tanzania Trop Med Int Health 2003 8 767 774 12950662 10.1046/j.1365-3156.2003.01100.x
Kun JFJ Lehman LG Lell B Schmidt-Ott R Kremsner PG Low-dose treatment with sulfadoxine-pyrimethamine combinations selects for drug-resistant Plasmodium falciparum strains Antimicrob Agents Chemother 1999 43 2205 2208 10471565
Watkins WM Mosobo M Treatment of Plasmodium-falciparum Malaria with Pyrimethamine-Sulfadoxine – Selective Pressure for Resistance Is a Function of Long Elimination Half-Life Trans R Soc Trop Med Hyg 1993 87 75 78 8465404 10.1016/0035-9203(93)90431-O
Hastings IM Modelling parasite drug resistance: lessons for management and control strategies Trop Med Int Health 2001 6 883 890 11703842 10.1046/j.1365-3156.2001.00800.x
Dicko A Sagara S Sissoko MS Guindo O Diallo ABI Kone M Thera MA Sacko M Doumbo OK Impact of intermittent preventive treatment with sulfadoxine pyrimethamine targetting the transmission season on the incidence of clinical malaria in children of 6 months to 10 years in Kambila, Mali Am J Trop Med Hyg 2004 71 1 79
Beck HP Felger I Vounatsou P Hirt R Tanner M Alonso P Menendez C The epidemiology of multiple Plasmodium falciparum infections – 8. Effect of iron supplementation and malaria prophylaxis in infants on Plasmodium falciparum genotypes and multiplicity of infection Trans R Soc Trop Med Hyg 1999 93 Suppl 1 41 45 10450425 10.1016/S0035-9203(99)90326-7
Ocana-Morgner C Mota MM Rodriguez A Malaria blood stage suppression of liver stage immunity by dendritic cells J Exp Med 2003 197 143 151 12538654 10.1084/jem.20021072
Belnoue E Costa FTM Frankenberg T Vigario AM Voza T Leroy N Rodrigues MM Landau I Snounou G Renia L Protective T cell immunity against malaria liver stage after vaccination with live sporozoites under chloroquine treatment J Immunol 2004 172 2487 2495 14764721
Gill CJ Sabin LL Tham J Hamer DH Reconsidering empirical cotrimoxazole prophylaxis for infants exposed to HIV infection Bull World Health Organ 2004 82 290 297 15259258
Iyer JK Milhous WK Cortese JF Kublin JG Plowe CV Plasmodium falciparum cross-resistance between trimethoprim and pyrimethamine Lancet 2001 358 1066 1067 11589941 10.1016/S0140-6736(01)06201-8
Rosen JB Breman JG Malaria intermittent preventive treatment in infants, chemoprophylaxis, and childhood vaccinations Lancet 2004 363 1386 1388 15110499 10.1016/S0140-6736(04)16052-2
|
16033653
|
PMC1193983
|
CC BY
|
2021-01-04 16:24:13
|
no
|
Malar J. 2005 Jul 20; 4:33
|
utf-8
|
Malar J
| 2,005 |
10.1186/1475-2875-4-33
|
oa_comm
|
==== Front
PLoS Comput BiolPLoS Comput. BiolpcbiplcbploscompPLoS Computational Biology1553-734X1553-7358Public Library of Science San Francisco, USA 10.1371/journal.pcbi.001003005-PLCB-RA-0103R2plcb-01-03-06Research ArticleBiochemistryMolecular Biology - Structural BiologyNeuroscienceNoneMolecular Origin of Polyglutamine Aggregation in Neurodegenerative Diseases A Molecular Basis of PolyQ AggregationKhare Sagar D 1Ding Feng 1Gwanmesia Kenneth N 12Dokholyan Nikolay V 1*1 Department of Biochemistry and Biophysics, University of North Carolina, Chapel Hill, North Carolina, United States of America
2 Department of Physics and Pre-Engineering, Delaware State University, Dover, Delaware, United States of America
Lai Luhua EditorPeking University, China*To whom correspondence should be addressed. E-mail: [email protected] 2005 26 8 2005 1 3 e3010 3 2005 13 7 2005 Copyright: © 2005 Khare 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.Expansion of polyglutamine (polyQ) tracts in proteins results in protein aggregation and is associated with cell death in at least nine neurodegenerative diseases. Disease age of onset is correlated with the polyQ insert length above a critical value of 35–40 glutamines. The aggregation kinetics of isolated polyQ peptides in vitro also shows a similar critical-length dependence. While recent experimental work has provided considerable insights into polyQ aggregation, the molecular mechanism of aggregation is not well understood. Here, using computer simulations of isolated polyQ peptides, we show that a mechanism of aggregation is the conformational transition in a single polyQ peptide chain from random coil to a parallel β-helix. This transition occurs selectively in peptides longer than 37 glutamines. In the β-helices observed in simulations, all residues adopt β-strand backbone dihedral angles, and the polypeptide chain coils around a central helical axis with 18.5 ± 2 residues per turn. We also find that mutant polyQ peptides with proline-glycine inserts show formation of antiparallel β-hairpins in their ground state, in agreement with experiments. The lower stability of mutant β-helices explains their lower aggregation rates compared to wild type. Our results provide a molecular mechanism for polyQ-mediated aggregation.
Synopsis
Nine human diseases, including Huntington's disease, are associated with an expanded trinucleotide sequence CAG in genes. Since CAG codes for the amino acid glutamine, these disorders are collectively known as polyglutamine diseases. Although the genes (and proteins) involved in different polyglutamine diseases have little in common, the disorders they cause follow a strikingly similar course: If the length of the expansion exceeds a critical value of 35–40, the greater the number of glutamine repeats in a protein, the earlier the onset of disease and the more severe the symptoms. This fact suggests that abnormally long glutamine tracts render their host protein toxic to nerve cells, and all polyglutamine diseases are hypothesized to progress via common molecular mechanisms. One possible mechanism of cell death is that the abnormally long sequence of glutamines acquires a shape that prevents the host protein from folding into its proper shape. What is the structure acquired by polyglutamine and what is the molecular basis of the observed threshold repeat length? Using computer models of polyglutamine, the authors show that if, and only if, the length of polyglutamine repeats is longer than the critical value found in disease, it acquires a specific shape called a β-helix. The longer the glutamine tract length, the higher is the propensity to form β-helices. This length-dependent formation of β-helices by polyglutamine stretches may provide a unified molecular framework for understanding the structural basis of different trinucleotide repeat-associated diseases.
Citation:Khare SD, Ding F, Gwanmesia KN, Dokholyan NV (2005) Molecular origin of polyglutamine aggregation in neurodegenerative diseases. PLoS Comp Biol 1(3): e30.
==== Body
Introduction
The appearance of polyglutamine (polyQ)-containing aggregates [1–3] is a hallmark of disease progression in all diseases in which CAG-expansions occur in genes [2]. Intranuclear inclusion bodies containing polyQ aggregates have been found in vitro [4,5], in cell cultures, animal models, and affected patients [6,7]. The aggregates are known to have a characteristic amyloid topology [8]. The inhibition of oligomerization by the azo-dye Congo red, or by the Hsp70/Hsp40 chaperone system, exerts marked protective effects in vivo and in vitro [9,10]. Aggregation and disease are observed if the number of glutamines in the expansion, n, exceeds a critical value, n
C (i.e., n
C = 35–40) [3]. The nearly universal existence of this criticality in all polyQ-related diseases (except in spinocerebellar ataxia 6) suggests that when the polyQ insert length exceeds a critical value (n > n
C), a pathological change, largely independent of the host protein, occurs in the polyQ insert itself. Therefore, isolated polyQ peptides (Qn) have been used as model systems for studying polyQ aggregation [4,11,12], and it is known that: (a) The nuclear uptake of polyQ peptide aggregates prepared in vitro is cytotoxic in cell cultures [13], (b) isolated polyQ peptides have in vitro aggregation properties similar to the corresponding full-length proteins containing the polyQ insert [4,14], (c) peptide aggregation follows a nucleated mechanism showing characteristic lag and growth phases [5,11], and (d) the glutamine tract-length dependence of the lag-time interval correlates well with the age of onset of disease [11]. Peptides of subcritical lengths (n < 35–40) have long lag times of aggregation and a corresponding (predicted) age of onset later than the typical life span of a person. Longer peptides (n > 35–40) have progressively smaller lag times of aggregation, and a correspondingly early age of onset of the disease [11].
Unaggregated polyQ peptides form random coil structures, whereas aggregates are composed of amyloid-like β-strands [15]. The conversion of random coil to β-strand occurs in an individual polyQ chain [11], and fibril formation occurs by addition of other polyQ chains to these monomeric β-strand nuclei. Therefore, the conformational dynamics of an individual polyQ chain determines both its aggregation mechanism and the structure of the final aggregates. The details of the conformational dynamics of polyQ and the length dependence of the dynamics are not well understood [6].
Results/Discussion
To elucidate the structural dynamics of single polyQ chains, we performed molecular dynamics (MD) simulations of simplified models of polyQ. An atomic-level representation of polyQ limits sampling efficiency in simulations, making it unsuitable to study the dynamics of aggregation. Therefore, we used a simplified pseudo-atom representation to capture the relevant degrees of freedom for aggregation. We introduced three types of nonbonded interactions between the glutamine pseudo-atoms: hydrophobic interactions between the glutamine methylene groups, geometrically determined hydrogen bonds between backbone NH and O atoms, and sidechain-backbone polar interactions between the sidechain carboxylamine group and the backbone NH or O atoms. Protein-solvent interactions play an important role in protein folding and aggregation. However, in simplified models of protein folding and aggregation the solvent interactions are considered implicitly [16]. In the interaction models employed in our study, the solvent effects were captured by the effective hydrophobic interactions between the methylene groups in the sidechain. We used the discrete molecular dynamics (DMD) algorithm [17] to study polyQ dynamics.
The first question that we address is the following: What is the underlying minimal set of interactions responsible for the experimentally observed conformational transitions in polyQ aggregation? Since the conformational transition from random-coil polyQ to β-strand is known to be a nucleated process [11], we expect that an energy barrier is crossed during β-strand formation. Barrier crossing is enhanced as the system temperature is increased. Therefore, we study the dynamics of model polyQ peptides as a function of temperature. We hypothesized that the physical basis of the conformational change from random coil to β-strands is the presence of unique sidechain-backbone hydrogen bonding interactions in polyQ. To test this hypothesis, we performed simulations of a 37-mer (Q
37) polyQ peptide with and without sidechain-backbone hydrogen bonding interactions. It is known that homopolymeric peptides with no sidechain-backbone interactions, e.g., polyalanine, form α-helices in their ground state [18,19], and at higher temperatures the helices melt to form a random coil [20] that then aggregates into a β-rich structure. We found that in the absence of sidechain-backbone interactions, polyQ dynamics are similar to polyalanine: It forms α-helices at low temperature (Figure 1A), and a random coil as the temperature is increased. A monomer peptide in this polyQ model does not form β-strands. In contrast, when sidechain-backbone hydrogen bonding is present, polyQ is a random coil at low temperatures, adopts a β-strand conformation in an intermediate range of temperatures, and is again a random coil at higher temperatures (Figure 1B).Thus, sidechain-backbone interactions lead to the formation of β-strands by a single polyQ peptide, which is the nucleating structural transition observed in polyQ-peptide aggregation.
Figure 1 Dynamics of PolyQ Peptides
(A, B) The fraction of secondary structure as a function of temperature for model polyQ peptides (A) without sidechain-backbone interactions, and (B) with sidechain-backbone interactions. Only helices and β-strands are shown; at each temperature the remainder is random coil.
(C) A β-helix formed by a Q
37 at the temperature T = 0.74. The width of the helix is 12–13 Å, and the distance between strands is 4–5 Å.
(D) The network of possible sidechain-backbone (green) and sidechain-sidechain (orange) hydrogen bonding interactions in the core of the β-helix. Hydrogen bonds are shown as dashed lines, and bifurcations indicate the multiple possibilities for interaction partners.
(E, F) Side (E) and top (F) views of the β-helix structure showing the well-packed core and outward-pointing sidechains.
Strikingly, the conformation adopted by a Q
37 chain under conditions in which it adopts β-strand topologies (T = 0.72 to T = 0.78, in units of ɛ/k
B, where ɛ is the energy unit and k
B is the Boltzmann's constant) is a parallel β-helix (Figure 1C). In these β-helices, all residues adopt β-strand backbone dihedral angles, and the polypeptide chain coils around a central helical axis. Several examples of such parallel β-helices (reviewed by Wetzel [21]) are found in the Protein Databank (http://www.rcsb.org/pdb/). For amyloid fibrils formed by polyQ, Perutz previously proposed a β-helix model based on X-ray diffraction data [8]. However, in contrast with Perutz's model, which has a central aqueous pore, the β-helices observed in our simulations are well packed, exclude the solvent, and are stabilized by buried sidechain-backbone and sidechain-sidechain hydrogen bonds (Figure 1D–1F). To evaluate whether these β-helix structures, once formed, have residence times long enough to propagate further aggregation, and to obtain better-defined thermodynamic ensembles, we evaluated the stability of β-helices at 300 Kelvin using all-atom MD simulations. As shown in Figure 2A, the polyQ structure remains stable on the nanosecond time scale accessible in all-atom simulations. If the formation of β-helices corresponds to the nucleation step in the aggregation reaction [5], and, assuming that the further elongation of the aggregate is diffusion-limited, the average time between protein collisions at a concentration of 100 μM is expected to be about 10 ns. Therefore, the observed stability of the polyQ β-helix on the nanosecond time scale is expected to be sufficient for further propagation of the aggregate.
Figure 2 Formation and Stability of β-Helices
(A) RMSD versus time in all-atom MD simulations of polyQ peptide β-helices for wild-type β-helix (red) and PG-Q9 β-helix (black). The PG-Q9 β-helix was constructed by replacing six glutamines in Q
37 with the proline-glycine sequence inserted after every nine glutamines.
(B) The fraction of β-helix content as a function of temperature for a 25-mer (diamonds), 37-mer (unfilled circles), and 45-mer (squares). The fraction of β-helix content was calculated by measuring the frequency of contacts between a residue i and residues i + 16 to i + 20, for all i.
Apart from sidechain-backbone interactions, a number of sidechain-sidechain interactions persisted throughout the MD simulation. We did not use sidechain-sidechain hydrogen bonding interactions in the DMD simulations used to generate β-helices. Thus, even though sidechain-sidechain interactions were not required for the formation of β-helices, they were nevertheless formed in β-helices. Further, these interactions were persistent throughout the length of the all-atom MD simulations, suggesting that they do play a significant role in the stabilization of the β-helices.
By characterizing the length dependence of β-helix formation, we uncovered the molecular basis of the observed glutamine length dependence of polyQ aggregation. Since the average number of residues per turn of the β-helix in our simulations is 18.5 ± 2 residues, we expected that about 33–40 glutamines would be required for its formation. To test this hypothesized length dependence of β-helix formation, we studied the conformational dynamics of 25-mer and 45-mer polyQ peptides and found that β-helices are absent at all temperatures when the repeat length was 25. Moreover, the β-helix topology was stable in a broader range of temperatures for the 45-mer (Figure 2B), demonstrating that β-helix formation increases with the length of polyQ. The formation of a β-helix from a random coil was accompanied by entropy loss, leading to a free energy barrier. This barrier results in the lag times observed in experiments of polyQ peptide aggregation [11]. Barrier crossing is enhanced for longer peptides, because the enthalpy gain upon β-helix formation compensates for the entropy loss in the transition. Therefore, β-helices are formed only by peptides longer than a critical length (n > n
C). Recently, Stork et al. [22] found that the dimerization of two Q
37 β-helices resulted in a stabilization of the (preformed, in their study) β-helix conformation. The stabilization of the β-helix upon dimerization shows that the dimerization is a downhill process (i.e., there is no energy barrier) on the free-energy landscape. Thus, we propose that length-dependent β-helix formation may be the molecular origin of polyglutamine-mediated aggregation. We also propose that once a β-helix is formed by a monomer, the elongation of the aggregate may involve the conversion of other chains to β-helices induced by the β-helix nucleus. A fibril may be formed by stacking of multiple β-helices, as suggested by Stork et al. [22], and these fibrils may arrange further to form larger fibers.
The formation of β-helices by a single polyQ chain can be used to rationalize the aggregation of experimentally characterized mutant polyQ peptides. Previously, Thakur and Wetzel [5] found that mutant polyQ peptides, in which the turn-inducing amino acid sequence proline-glycine was inserted at different sequence intervals—e.g., (Q9-PG-Q9)3 (PG-Q9) and (Q10-PG-Q10)3 (PG-Q10)—modulated the aggregation kinetics of polyQ peptides. The mutant PG-Q9 was found to aggregate at a marginally smaller rate than the wild-type polyQ, and the critical nucleus size for aggregation, as for the wild type, was one. The existence of nucleated aggregation kinetics of the mutant suggests that, similar to the wild type, barrier crossing from the ground state to an aggregation-prone state occurs. Therefore, we hypothesized that, similar to wild-type polyQ, a thermodynamically unfavorable nucleating conformational transition occurs in a single β-hairpin forming PGQ9 peptide. We studied the conformational dynamics of the PGQ9-mutant peptide and found that, in agreement with Thakur and Wetzel's prediction, this peptide forms a four-stranded antiparallel β-sheet. The antiparallel structures are formed at low temperatures in our simulation (Figure 3A and 3B). Do these mutants aggregate through antiparallel β-hairpin structures or by wild-type-like β-helix formation? Thakur and Wetzel's data [5] is compatible with either scenario. If β-hairpin formation is rate-limiting (i.e., nucleates aggregation), since β-hairpin formation by mutants is more thermodynamically favorable than wild type, the aggregation rates of mutants should be higher than wild type. In contrast, if wild type-like β-helix formation in the mutants nucleates aggregation, their aggregation rate compared to the wild type is expected to be determined by the relative stabilities of the metastable mutant and wild-type β-helices. To understand the nucleating conformational transition in these mutant peptides, we performed all-atom MD simulations of a PGQ9 sequence in a β-helix conformation (Figure 3C). We found that, similar to the wild type β-helix, the β-helix formed by PGQ9 remained stable on the nanosecond time scale, but showed a greater root mean-square distance (RMSD) compared to a wild type β-helix of identical length (see Figure 2A). We compared the RMSD per residue of the wild-type and mutant structures (unpublished data) during MD simulations and found that the destabilization induced by the proline-glycine is not limited to the proline-glycine residues—it is transduced across the whole peptide, leading to an overall higher RMSD. Thus, we propose that the differential stability of the transiently formed β-helix by the mutant peptide compared to wild-type polyQ may underlie the experimentally observed slower rate of aggregation of the mutant.
Figure 3 Characterization of PolyQ Mutants
(A) Secondary structure formed by the PG-Q9 mutant at T = 0.74. The glutamines formed β-sheets (solid line) with high frequency, whereas the proline-glycine sequence formed turns (dashed line).
(B) The β-hairpin formed by the PG-Q9 peptide in its ground state are shown; the proline and glycine residues are highlighted in green.
(C) The β-helix conformation of PG-Q9 after 3 ns in MD simulations is shown; the turns in the β-helix (green) are formed at the proline-glycine insertion.
Mechanisms of protein aggregation [23] are increasingly being sought as a framework for understanding and, importantly, therapeutically interfering with, the fundamental events that underlie misfolding diseases [24]. The common underlying basis of protein aggregation has been demonstrated by the discovery of antibodies can cross-react with early aggregates of different peptides and proteins [25]. Further, the early oligomers themselves, rather than the final fibrils, have been shown to be toxic [26]. Thus, the conversion to specific β-strand topologies is a common central feature associated with cytotoxicity in all aggregation-linked diseases. The structural basis of the mechanism of polyQ peptide aggregation that we present here may thus aid the understanding and development of rational therapies to modulate protein aggregation in these debilitating neurodegenerative diseases.
Materials and Methods
We modeled the polyQ chain as “beads on a string,” where each glutamine is represented by six pseudo-atoms—four corresponding to the backbone NH, C′, Cα, and O atoms, and two side chain atoms, one for the methylene (-CH2-CH2-) groups and another for the carboxylamine (-CONH2) group. Neighboring residues in this peptide representation are covalently constrained to mimic the peptide flexibility in real proteins.
To study the conformational dynamics of polyQ, we introduce simplified amino acid interactions: hydrophobic interactions between the methylene groups, polar interactions between sidechains and backbone NH or O beads, and nonspecific backbone hydrogen bonds as described in [16]. The interaction strengths of the hydrophobic interactions, sidechain-backbone polar interactions and nonspecific backbone hydrogen bonds are assigned as 0.7ɛ, 5ɛ, and 5ɛ, respectively. These interaction strengths were used to successfully fold a miniprotein, the Trp-cage, to within 1 Å of its native structure [16]. Interactions in proline and glycine were also modeled as in [16]. We used the rapid DMD algorithm to perform simulations on our model proteins [17,27,28].
We used the snapshots collected from DMD simulations to perform all-atom MD simulations using standard MD protocols. Using a three-step algorithm [29], we reconstructed all atoms of the polyQ chain from the snapshots taken from simulations of coarse-grained protein models. All-atom MD simulations were performed using the package AMBER 7, with the AMBER force-field of parm99 [30,31] at a temperature of 300 Kelvin and pressure of one atmosphere, in a octahedral periodic box of water. The protocol for MD simulations involved equilibration of the solvent and the peptide, and production as described by Urbanc et al. [29]. The trajectory was recorded for 3 ns after equilibration.
Sidechain-backbone interactions have been identified as playing important roles in the formation of protein structures. It has been pointed out that the hydrogen bonds between the polar side chain and backbones are important for the starting and ending of α-helices [32,33] and for the formation of turns in proteins [34]. To evaluate how sensitive our results were to the relative strengths of sidechain and backbone hydrogen bonds, we performed DMD simulations with varying relative strengths of sidechain and backbone hydrogen bonds (ɛsidechain/ɛbackbone). We found that for weaker sidechain hydrogen bonds compared to backbone hydrogen bonds, i.e., ɛsidechain/ɛbackbone < 1, polyglutamine (polyQ) formed α-helices at low temperatures as opposed to random coil structures observed at ɛsidechain/ɛbackbone = 1. The observation of random coils at low temperatures is in agreement with experiments in [35], and therefore we chose ɛsidechain/ɛbackbone = 1.
We thank R. Wetzel, F. U. Hartl, and B. Kuhlman for helpful discussions. This work was supported in part by Muscular Dystrophy Association grant MDA3720, Research Grant 5-FY03–155 from the March of Dimes Birth Defect Foundation, and the UNC/IBM Junior Investigator Award.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. SDK, FD, and NVD conceived and designed the experiments. SDK, FD, and KNG performed the experiments. SDK, FD, and NVD analyzed the data. SDK and NVD wrote the paper.
A previous version of this article appeared as an Early Online Release on July 14, 2005 (DOI: 10.1371/journal.pcbi.0010030.eor).
Abbreviations
DMDdiscrete molecular dynamics
MDmolecular dynamics
RMSDroot mean-square distance
==== Refs
References
Ross CA Wood JD Schilling G Peters MF Nucifora FC 1999 Polyglutamine pathogenesis Philos Trans R Soc Lond B Biol Sci 354 1005 1011 10434299
Ross CA Poirier MA 2004 Protein aggregation and neurodegenerative disease Nat Rev Neurosci 10 S10 17
Ross CA 2002 Polyglutamine pathogenesis: Emergence of unifying mechanisms for Huntington's disease and related disorders Neuron 35 819 822 12372277
Chen S Berthelier V Yang W Wetzel R 2001 Polyglutamine aggregation behavior in vitro supports a recruitment mechanism of cytotoxicity J Mol Bol 311 173 182
Thakur AK Wetzel R 2002 Mutational analysis of the structural organization of polyglutamine aggregates Proc Natl Acad Sci U S A 99 17014 17019 12444250
Masino L Pastore A 2002 Glutarnine repeats: Structural hypotheses and neurodegeneration Biochem Soc Trans 30 548 551 12196134
Temussi PA Masino L Pastore A 2003 From Alzheimer to Huntington: Why is a structural understanding so difficult? EMBO J 22 355 361 12554637
Perutz MF Finch JT Berriman J Lesk A 2002 Amyloid fibers are water-filled nanotubes Proc Natl Acad Sci U S A 99 5591 5595 11960014
Sanchez I Mahlke C Yuan JY 2003 Pivotal role of oligomerization in expanded polyglutamine neurodegenerative disorders Nature 421 373 379 12540902
Muchowski PJ Schaffar G Sittler A Wanker EE Hayer-Hartl MK 2000 Hsp70 and Hsp40 chaperones can inhibit self-assembly of polyglutamine proteins into amyloid-like fibrils Proc Natl Acad Sci U S A 97 7841 7846 10859365
Chen SM Ferrone FA Wetzel R 2002 Huntington's disease age-of-onset linked to polyglutamine aggregation nucleation Proc Natl Acad Sci U S A 99 11884 11889 12186976
Chen SM Berthelier V Hamilton JB O'Nuallain B Wetzel R 2002 Amyloid-like features of polyglutamine aggregates and their assembly kinetics Biochemistry 41 7391 7399 12044172
Yang W Dunlap JR Andrews RB Wetzel R 2002 Aggregated polyglutamine peptides delivered to nuclei are toxic to mammalian cells Hum Mole Genet 11 2905 2917
Scherzinger E Sittler A Schweiger K Heiser V Lurz R 1999 Self-assembly of polyglutamine-containing huntingtin fragments into amyloid-like fibrils: Implications for Huntington's disease pathology Proc Natl Acad Sci U S A 96 4604 4609 10200309
Altschuler EL Hud NV Mazrimas JA Rupp B 2000 Structure of polyglutamine FEBS Lett 472 166 167 10781826
Ding F Buldyrev SV Dokholyan NV 2005 Folding Trp-cage to NMR resolution native structure using a coarse-grained protein model Biophys J 88 147 155 15533926
Dokholyan NV Buldyrev SV Stanley HE Shakhnovich EI 1998 Discrete molecular dynamics studies of the folding of a protein-like model Fold Des 3 577 587 9889167
Marqusee S Robbins VH Baldwin RL 1989 Unusually stable helix formation in short alanine-based peptides Proc Natl Acad Sci U S A 86 5286 5290 2748584
Chakrabartty A Kortemme T Baldwin RL 1994 Helix propensities of the amino-acids measured in alanine-based peptides without helix-stabilizing side-chain interactions Protein Sci 3 843 852 8061613
Blondelle SE Forood B Houghten RA PerezPaya E 1997 Polyalanine-based peptides as models for self-associated beta-pleated-sheet complexes Biochemistry 36 8393 8400 9204887
Wetzel R 2002 Ideas of order for amyloid fibril structure Structure 10 1031 1036 12176381
Stork M Giese A Kretzschmar HA Tavan P 2005 Molecular dynamics simulations indicate a possible role of parallel beta- helices in seeded aggregation of poly-Gln Biophys J 88 2442 2451 15665127
Dobson CM 2003 Protein folding and misfolding Nature 426 884 890 14685248
Hammarstrom P Wiseman RL Powers ET Kelly JW 2003 Prevention of transthyretin amyloid disease by changing protein misfolding energetics Science 299 713 716 12560553
Kayed R Head E Thompson JL McIntire TM Milton SC 2003 Common structure of soluble amyloid oligomers implies common mechanism of pathogenesis Science 300 486 489 12702875
Kayed R, Sokolov Y, Edmonds B, McIntire TM, Milton SC, at al. 2004 Permeabilization of lipid bilayers is a common conformation-dependent activity of soluble amyloid oligomers in protein misfolding diseases J Biol Chem 279 46363 46366 15385542
Alder BJ Wainwright TE 1959 Studies in molecular dynamics. I. General method Journal of Chemical Physics 31 459 466
Zhou YQ Hall CK Karplus M 1996 First-order disorder-to-order transition in an isolated homopolymer model Physical Review Letters 77 2822 2825 10062054
Urbanc B Cruz L Ding F Sammond D Khare S 2004 Molecular dynamics simulation of amyloid beta dimer formation Biophys J 87 2310 2321 15454432
Cornell WD Cieplak P Bayly CI Gould IR Merz KM 1995 A 2nd generation force-field for the simulation of proteins, nucleic-acids, and organic-molecules J Am Chem Soc 117 5179 5197
Wang JM Cieplak P Kollman PA 2000 How well does a restrained electrostatic potential (RESP) model perform in calculating conformational energies of organic and biological molecules? J Comput Chem 21 1049 1074
Aurora R Rose GD 1998 Helix capping Protein Sci 7 21 38 9514257
Presta LG Rose GD 1988 Helix signals in proteins Sci 240 1632 1641
Stickle DF Presta LG Dill KA Rose GD 1992 Hydrogen-bonding in globular-proteins J Mol Biol 226 1143 1159 1518048
Altschuler EL Hud NV Mazrimas JA Rupp B 1997 Random coil conformation for extended polyglutamine stretches in aqueous soluble monomeric peptides J Pept Res 50 73 75 9273890
|
16158094
|
PMC1193989
|
CC BY
|
2021-01-05 09:18:22
|
no
|
PLoS Comput Biol. 2005 Aug 26; 1(3):e30
|
utf-8
|
PLoS Comput Biol
| 2,005 |
10.1371/journal.pcbi.0010030
|
oa_comm
|
==== Front
PLoS Comput BiolPLoS Comput. BiolpcbiplcbploscompPLoS Computational Biology1553-734X1553-7358Public Library of Science San Francisco, USA 10.1371/journal.pcbi.001003205-PLCB-RA-0076R3plcb-01-03-07Research ArticleBioinformatics - Computational BiologyEvolutionGenetics/Population GeneticsGenetics/EvolutionNoneEvolution of Genetic Potential Evolution of Genetic PotentialAncel Meyers Lauren 12*Ancel Fredric D 3Lachmann Michael 41 Section of Integrative Biology, Institute for Cellular and Molecular Biology, University of Texas, Austin, Texas, United States of America
2 Santa Fe Institute, Santa Fe, New Mexico, United States of America
3 Department of Mathematical Sciences, University of Wisconsin, Milwaukee, Wisconsin, United States of America
4 Max Planck Institute for Evolutionary Anthropology, Leipzig, Germany
Holmes Eddie EditorPennsylvania State University, United States of America*To whom correspondence should be addressed. E-mail: [email protected] 2005 26 8 2005 1 3 e3215 4 2005 22 7 2005 Copyright: © 2005 Meyers 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.Organisms employ a multitude of strategies to cope with the dynamical environments in which they live. Homeostasis and physiological plasticity buffer changes within the lifetime of an organism, while stochastic developmental programs and hypermutability track changes on longer timescales. An alternative long-term mechanism is “genetic potential”—a heightened sensitivity to the effects of mutation that facilitates rapid evolution to novel states. Using a transparent mathematical model, we illustrate the concept of genetic potential and show that as environmental variability decreases, the evolving population reaches three distinct steady state conditions: (1) organismal flexibility, (2) genetic potential, and (3) genetic robustness. As a specific example of this concept we examine fluctuating selection for hydrophobicity in a single amino acid. We see the same three stages, suggesting that environmental fluctuations can produce allele distributions that are distinct not only from those found under constant conditions, but also from the transient allele distributions that arise under isolated selective sweeps.
Synopsis
Variation is the fuel of natural selection. Understanding the mutational processes that underlie evolution has long been a central objective of population genetics. Today, amidst a computational revolution in biology, such understanding is pivotal to progress in many biological disciplines. For example, neutral mutations make the molecular clock tick, and this clock is fundamental to reconstructing phylogenies, measuring recombination rates, and detecting genetic functionality. In this manuscript, the researchers provide an original perspective on a long-standing question in evolutionary biology: to what extent do mutation rates evolve? They argue that to cope with environmental fluctuation, populations can evolve their phenotypic mutation rate without changing their genetic mutation rate. That is, populations can evolve “genetic potential”—a heightened sensitivity to the effects of mutation. The researchers use a simple mathematical model of amino acid evolution to illustrate the evolution of genetic potential, and show that as environmental variability decreases, evolving populations reach three distinct states. In a rapidly fluctuating environment, organisms evolve the flexibility to cope with variation within an individual lifetime; in moderately variable environments, populations evolve the ability to evolve rapidly; and in fairly constant environments, populations evolve robustness against the adverse effects of mutation.
Citation:Meyers LA, Ancel FD, Lachmann M (2005) Evolution of genetic potential. PLoS Comp Biol 1(3): e32.
==== Body
Introduction
Recent work in evolutionary biology has highlighted the degeneracy of the relationship between genes and traits [1]. For any particular trait value, there will exist a large set of genotypes that give rise to that value. A mutation from one such genotype to another will be neutral, having no noticeable impact on the physiology, behavior, or fitness of organisms. Metaphorically, one can imagine a population moving via mutation through a region of genotype space that maps to a neutral plateau in phenotype space. Near the periphery, mutations are likely to produce different (usually worse and occasionally better) phenotypes, whereas near the center of the neutral plateau, mutations have little impact on the phenotype. Evolutionary theory suggests that populations can harness this variation to achieve phenotypic stability under steady conditions through either mutational insensitivity [2,3] or mutational hypersensitivity [4], or to facilitate phenotypic exploration during adaptation [5,6].
A separate body of evolutionary theory addresses adaptation under fluctuating conditions [7,8]. The rate of the fluctuations will influence the resulting response. If the environment changes rapidly relative to the average generation time, populations may evolve mechanisms such as physiological plasticity and learning by which individual organisms can respond to their conditions [9,10]. As environmental change slows down, viable strategies include stochastic or directed heterogeneity in developmental pathways that give rise to phenotypic variation on the order of once per generation [11]. For even slower rates of change, mutations may produce novel phenotypes at a sufficiently high rate. Hypermutable lineages can produce novelty every few generations, as has been observed in viruses and mutator strains of bacteria [12,13]. When environmental fluctuations are rare, populations may experience extended epochs of directional selection and thus have sufficient time to achieve genetic robustness for any given state. Immediately following an environmental shift, however, such populations may pass through transitional periods of within-individual or between-generation plasticity before completely losing the previously favored phenotype in favor of a currently favored phenotype. This evolutionary transformation—from a trait that is acquired through phenotypic plasticity to a genetically determined version of the same trait—is known as the Baldwin Effect [9,14].
In this paper we show that genetic degeneracy may give rise to an alternative outcome under fluctuating conditions: the evolution of genotypes with heightened sensitivity to mutation. We introduce the term “genetic potential” to describe this state. Metaphorically, populations with genetic potential lie near the edge of neutral plateaus. Although the rate of mutation is unchanged, the likelihood that mutations produce beneficial variation increases. Heightened sensitivity to mutations has been recognized as a critical and transient phase of adaptive evolution [5,15,16]. Here we argue that genetic potential can be a stable condition for a population evolving under changing selection pressures. Using a simple mathematical model, we show that as environmental variability increases, natural selection at first moves populations between genetically robust states, then increasingly favors genetic potential, and ultimately produces mechanisms for environmental robustness within individual organisms.
We then present a more biological example of this phenomenon using a model of amino acid evolution. There is evidence that, within viral pathogens, the physiochemical properties of amino acids found within epitopes—regions of proteins that directly interact with the host immune system—can rapidly evolve [17,18]. Likewise, highly evolvable codons have been identified in bacteriophage experiencing shifting hosts [19] and in enzymes experiencing shifting substrates [20]. Motivated by these observations, we model codon evolution at a single amino acid site under fluctuating selection for hydrophobicity. As in the first model, natural selection produces three distinct outcomes with increasing environmental variability. Each outcome corresponds to distinct expectations about the distribution of amino acids and their codons at selected sites.
Under infrequent environmental change, populations evolve from one mutationally robust phenotype to another, briefly passing through genotypes that can easily mutate to either state. One might therefore be tempted to equate genetic potential with confinement to the intermediate steps on a path from robustness for one phenotype to robustness for another (Figure 1). While this is true in our simple model, the codon model illustrates that fluctuating environments may drive populations towards significantly greater genetic potential than found during these transient stages of isolated selective sweeps.
Figure 1 Evolution of Genetic Potential
The gray regions represent neutral networks—sets of genotypes that give rise to each phenotype. The degree of shading indicates the likelihood that mutations will impact phenotype, where darker regions are robust to mutations. Under constant conditions, populations evolve toward the most robust regions of neutral networks. Under variable conditions, populations may evolve toward genotypes that easily mutate from one phenotype to the other. These regions of genetic potential do not always lie on the evolutionary path between the equilibrium states for constant environments (arrow).
Results
Description of Models
The simple model.
We consider the evolution of a trait in an environment that alternates between two states (EA and EB), spending exactly λ generations per state between shifts. The simple model includes three phenotypes—one optimal phenotype for each of the two environments (A and B) and a third that has intermediate quality in both environments (V)—and a minimal amount of degeneracy in the relationship between the genotype and the phenotype. In particular, there is a single genetic locus, and five allelic possibilities at that locus (Figure 2A). Three of the alleles, g
0, g
1, and g
2, give rise to phenotype A, the fourth, g
3, gives rise to phenotype V, and the fifth, g
4, gives rise to phenotype B. The mutational structure is a pentagon in which gi can mutate to g
(i − 1) mod 5 or g
(i + 1) mod 5 for i ∈ {0,1,2,3,4}.
Figure 2 Mutational Networks
(A) Five alleles lie on a mutational pentagon with genetic degeneracy for the A phenotype. Colors indicate phenotypes with blue for A, yellow for B, and gray for V. Edges indicate that an allele on one side can mutate to the allele on the other side. Arrows illustrate the dynamics in equation 2.
(B) Each vertex represents an amino acid. The size of the vertex indicates the number of codons coding for the amino acid. Edges indicate point mutations between hydrophobicity classes. Mutations that preserve hydrophobicity class, including those that preserve the amino acid, are included in the model but not depicted here. The color of the vertex corresponds to the hydrophobicity class: blue indicates hydrophobic, yellow indicates hydrophilic, green indicates intermediate, and red indicates stop codons [21]. This network was drawn with PAJEK [50].
The fitness function changes with the environment such that
where w
A and w
B are the fitnesses in environments EA and EB, respectively, s > 0 is the fitness advantage for the specialized phenotype (A or B) in its preferred environment, and 0 ≤ k ≤ 1 determines the intermediacy of the V phenotype.
We can write the full model as a set of difference equations
for i ∈ {0,1,2,3,4}, where μ is the mutation rate and wt denotes the fitness in the current environment (Figure 2A). The number of individuals with genotype gi at time t is denoted by gi,t. The changing environment is governed by the following rule:
To simplify the analysis, this model tracks changes in the absolute population sizes of the various genotypes rather than their relative frequencies. Since the dynamics scale linearly with the total population size, one can achieve the same population dynamics by replacing the absolute sizes with relative frequencies and normalizing appropriately.
Variations on the simple model.
There are exactly 14 unique mutational networks consisting of five alleles on a pentagon, with at least one encoding A and at least one encoding B (see Materials and Methods). These include, for example, the pentagon with four consecutive alleles coding for A and one for B and the pentagon with alleles alternating in phenotype-A-B-A-V-B-. We are presenting analysis of the -A-A-A-V-B- model because it gives rise to some of the most interesting and generic dynamics found among these 14 models.
The codon model.
The previous model offers a transparent illustration of evolutionary dynamics under different rates of environmental change. Although somewhat simplistic, we believe that the qualitative predictions of the model will hold for a wide range of more plausible genotype–phenotype maps. To demonstrate this, we consider the evolution of a single amino acid site under fluctuating conditions. In this model, the genotypes
are the 64 codons in the standard genetic code and the phenotypes are hydrophobicities of the corresponding amino acids [21]. The environment alternately favors hydrophobic and hydrophilic amino acids. There are three classes of amino acids—hydrophobic, intermediate, and hydrophilic—and all amino acids in a class share the same fitness. The fitnesses are determined as in equation 1, with the fitnesses of all three stop codons equal to zero.
Each codon is mutationally connected to the nine others to which it can mutate via point mutation. This gives rise to the genetic network depicted in Figure 2B and the dynamics given by
for 1 ≤ i ≤ 64, where μ is the overall mutation rate, β is the transition/transversion ratio (2β is the transition/transversion rate ratio), Φi is the set of three transition point mutations of codon i, and Γi is the set of six transversion point mutations of codon i.
Analysis of the Simple Model
We provide an intuitive perspective on evolution in fluctuating environments using the simple model and then demonstrate the generality of the results in the codon model. The first results assume a mutation rate μ = 0.01, and fitnesses 1, 1.5, and 2 for the unfavored, intermediate, and favored phenotypes, respectively. In a constant environment, a population will equilibrate on genotypes that encode the optimal phenotype. In environment EA, the equilibrium relative frequencies of g
0, g
1, g
2, g
3, and g
4 are 0.291, 0.412, 0.292, 0.003, and 0.002, respectively, and in environment EB, they are 0.005, 0.000, 0.000, 0.010, and 0.985, respectively. When there is degeneracy, as there is for phenotype A, the populations evolve genetic robustness, that is, more mutationally protected genotypes appear in higher frequency. In particular, g
1, which lies in the center of the three genotypes that code for A, appears in higher frequency than either genotype on the edge of the neutral network for A (g
0 and g
2) at equilibrium in EA. In the absence of degeneracy (phenotype B), we observe a mutation–selection balance around the single optimal genotype. These findings are consistent with and provide a transparent example of the extensive theory on mutation–selection balance, quasi-species, and the evolution of genetic robustness in neutral networks [2,22–24].
Under rapid environmental fluctuations, populations do not have time to reach a stable allele distribution. As the environment becomes more variable, the distributions of alleles go through three distinct phases. Figure 3 shows the frequency of every allele averaged over each environmental condition after the population has reached steady oscillations. For relatively stable environments, the populations swing back and forth between near equilibrium conditions for EA and EB, thereby alternating between genetic robustness for A and a mutation–selection balance around the single allele for B. At intermediate rates of fluctuation, populations hover near g
4 and g
0, where the genotypes for A abut the genotype for B. Thus, mutation between the two phenotypes occurs frequently. We call this outcome genetic potential because of the enhanced potential for mutations to give rise to novel (beneficial) phenotypes. Finally, for highly variable environments, the populations converge on the phenotype V, which has unchanging, intermediate fitness in both environments. Phenotype V corresponds to organismal flexibility—individual organisms tolerate both conditions, but neither one exceptionally well. There are a variety of mechanisms that can give rise to an intermediate phenotype including homeostasis, somatic evolution, physiological plasticity, and behavioral plasticity [7,8]. As originally predicted by Dempster [25], the ascent of V under rapid fluctuations only occurs if the fitness of V is greater than the geometric mean fitness over time for either A or B.
Figure 3 Allele Distributions under Environmental Fluctuations
The graphs show the stationary allele distributions averaged over an EA epoch (top) and an EB epoch (bottom) as a function of the variability of the environment. As environmental variability decreases, the population moves from the intermediate phenotype to the genetic boundary between the A and B phenotypes, and eventually to an oscillation between the center of the network for A and the gene for B. Diagrams above the graphs illustrate the frequency distributions in each of the three phases. Vertex areas are proportional to the average frequencies for each allele. (For the data depicted in this figure, s = 1, k = 0.5, and μ = 0.01.)
Anaylsis of the Codon Model
The codon model gives rise to similar oscillations (Figure 4). Here we have assumed a transition/transversion ratio β = 2, mutation rate μ = 10−5, and fitnesses 1, 1.5, and 2 for the unfavored, intermediate, and favored phenotypes, respectively. (We address the impact of mutation rate in the Discussion.) Whereas in the simple model only one of the three phenotypes had multiple genotypes, in this model all three phenotypic classes have genetic degeneracy, and thus can evolve genetic robustness (Figure 4A). For highly variable environments, codons for amino acids with intermediate hydrophobicity dominate, and in particular, those that are least likely to mutate to one of the other two classes (Figure 4B). In a moderately variable environment, the populations exhibit genetic potential, hovering near the edges of the neutral networks for the two extreme classes, thereby enabling rapid evolution upon environmental transitions (Figure 4C). In relatively constant environments, we find alternating genetic robustness for the two extreme classes (Figure 4D).
Figure 4 Codon Distributions under Environmental Fluctuations
(A) gives the robustness for each codon, that is, the fraction of all possible point mutations that leave the hydrophobicity class unchanged. The codons have been ordered to reflect roughly the mutational adjacency of the hydrophobicity classes.
(B–D) show the average codon frequency distribution for each epoch type after the population has reached stationary oscillation. These show frequencies for environmental epochs of exactly λ generations (thick lines) and epochs of random duration—Poisson distributed with mean λ (thin lines). Black corresponds to epochs favoring hydrophobicity and gray corresponds to epochs favoring hydrophilicity. The rate of environmental fluctuations is decreasing from (B) to (D) (λ = 10, 102, and 106, respectively).
The genetic potential of a population can be estimated by the probability that a currently favored codon in the population will mutate to a currently unfavored or intermediate codon. This indicates the capacity to bounce back (via mutation and selection) if and when the environment reverts. For populations that have equilibrated in a constant environment and have recently experienced an environmental shift, genetic potential will decrease as the population becomes increasingly robust to the effects of mutation (Figure 5). For populations that have evolved under moderately fluctuating conditions, genetic potential remains noticeably higher. This suggests that the regular oscillations of such populations involve distributions of codons that are quite different (more mutable) than those found during the early stages of adaptation in an isolated selective sweep.
Figure 5 Faster Environmental Fluctuations Yield Greater Genetic Potential
Genetic potential is the likelihood that a mutation to a gene coding for the currently favored phenotype will produce the intermediate or unfavored phenotype. Thick lines correspond to populations that have reached stable oscillations when λ = 100, and thin lines correspond to populations that experience a single environmental shift after having equilibrated in a constant environment. The maximum genetic potential after a single shift is significantly less than the minimum under persistent fluctuations.
This difference also appears in the distributions of amino acids. We calculated the genetic potential in each generation of a population experiencing fluctuations every λ = 102 generations. Figure 6 (left) depicts the amino acid distributions for the generations that have the highest genetic potential in EA and EB. We then compared these two distributions to the evolving amino acid distribution in a population that equilibrates in one of the two environments and then faces an environmental shift. Figure 6 (right) shows the steady state distributions for this population and the transitional distributions that are most similar (i.e., smallest average squared difference in relative frequencies) to those depicted in Figure 6 (left). The distributions of amino acids in regions of genetic potential are strikingly different than those realized in populations evolving after an isolated environmental shift.
Figure 6 Amino Acid Distributions Reflect Genetic Potential
The left figure illustrates amino acid distribution in the generations with greatest genetic potential during each of the two epochs for λ = 100. Vertex area is proportional to the relative frequency of an amino acid. The right figure gives the amino acid distributions at equilibrium in the two environments (far left and right networks), and the transitional amino acid distributions that are most similar to those depicted for λ = 100 (left). Similarity is measured as mean squared difference in frequencies across all amino acids. The amino acid networks were drawn with PAJEK [50].
Discussion
We have provided an intuitive framework for studying the evolutionary implications of heterogeneous environments. Although much is known independently about the evolution of genetic robustness [3] and organismal flexibility [7,8], this model demonstrates that the extent of environmental variability may determine which of these two states evolves, and suggests the possibility of an intermediate state of heightened mutability. The transition points among the three states will be functions of both the environment and the mutation rate. In particular, increasing (decreasing) the mutation rate (within a moderate range) has the same qualitative effect as increasing (decreasing) the duration of an environmental epoch. As the mutation rate decreases, populations take longer to achieve genetic robustness, and therefore evolve genetic potential (rather than robustness) over large ranges of environmental variability. For example, at a mutation rate of μ = 10−5 in the codon model, populations evolve genetic potential when environment varies at rates of 101 < λ < 106 generations, approximately (Figure 4). If the mutation rate increases to μ = 10−2, the qualitative results are similar, with populations evolving genetic potential when the environmental variability is in the more limited range of 100 < λ < 103 generations, approximately. If, instead, the mutation rate decreases to μ = 10−9, then adaptation to genetic robustness proceeds at an exceedingly slow pace, yielding genetic potential throughout the extended range of 102 < λ < 1010 generations, approximately. To understand the comparable roles of mutation and environmental variability, note that the model includes three time-dependent processes—mutation, environmental change, and population growth. If one of these rates is changed, the other two can be modified to achieve identical system behavior on a shifted time scale. Since the dynamics only weakly depend on the force of selection, we can change the mutation rate and then scale the rate of environmental change to produce the original qualitative results. The connection between environmental variability and mutation has been noted before, with theory predicting that the optimal mutation rate under fluctuating environmental conditions is μ = 1/λ [26,27].
Our results suggest an alternative perspective on the evolution of mutation rates. Theory suggests that the optimal mutation rate should correspond to the rate of environmental change [26,28], yet the extent to which mutation rate can evolve is unclear [12,13,29]. Here we suggest that the genotypic mutation rate need not evolve as long as the phenotypic or effective mutation rate evolves. By evolving toward genotypes with higher genetic potential, populations increase the rate of phenotypically consequential mutations without modifications to the underlying genetic mutational processes.
We would like to emphasize that our second model is intended as one possible example of fluctuating selection among many thought to exist in nature. Whether or not one has much confidence in the particular evolutionary scenario, the qualitatively similar outcomes for the simple and complex models presented here suggest that the results may hold for a large class of systems in which there is redundancy in the relationship between genotype and phenotype. Hydrophobicity is just one of several physicochemical properties thought to play a role in the shifting functional demands on amino acids [17–20]. Another example is phase-shifting bacteria that have mutational mechanisms, for example, inversions in promoter regions [30] and slip-stranded mispairing within microsatellites [12], that lead to variation in functionally important phenotypes. The remarkable suitability of the phase-shifting variants to the diverse conditions experienced by the bacteria suggests that phase shifting may have evolved as a mechanism for genetic potential. We hypothesize that the major histocompatibility complex (MHC), which is the component of the immune system responsible for recognizing and binding foreign particles, may also have evolved genetic potential as a by-product of the flucuations arising out of coevolution with pathogens [31]. Studies suggest that several components of the immune system exhibit high overall rates of genetic change. In particular, there are specific amino acid sites within the MHC complex that seem to have experienced rapid evolutionary change [32]. One possible explanation is that each MHC gene as a whole, and these sites in particular, have a history of rapid adaptation to changing distributions of potential antigens. We therefore predict that such sites may have evolved genetic potential.
Evolvability has been defined as a population's ability to respond to selection [6,33]. Although the term has only recently taken root, ideas concerning the evolution of evolvability itself date back to the Fisher–Wright debate over the evolution of dominance [34,35] and include the large body of theory on the evolution of mutation rates and recombination [36,37]. Developmental biologists have begun to identify genetic architectures that promote diversification [38] and buffering mechanisms, such as heat shock proteins, that allow the accumulation of cryptic variation [39]. Although one can think of genetic potential as an abstraction of all mechanisms that increase the likelihood that a mutation will have a phenotypic effect, the genetic potential that evolves in our models is a very simple form of evolvability that exploits redundancy in the map from genotype to phenotype.
Genetic potential evolves in our models because prior and future environments are identical. If, instead, the environment continually shifts to completely novel states, the evolutionary history of a population may not prepare it for future adaptation. We speculate that some degree of genetic potential may still evolve if there exist genotypes on the periphery of neutral networks with broad phenotypic lability.
Biologists often refer to phenotypic plasticity, learning, and other forms of organismal flexibility as “adaptations” for coping with environmental heterogeneity [7,8]. Should genetic potential be seen as an alternative “solution,” or should it be viewed as simply a product of fluctuating selection? Although we remain agnostic, we note that this question might be asked of all forms of adaptive variation. Whether or not genetic potential should be viewed as an evolved strategy, we emphasize that it is not simply the truncation of the adaptive path a population follows from the equilibrium state in one constant environment to the equilibrium state in the other. In the codon model, intermediate rates of environmental fluctuations push the population into regions of the codon network where genetic potential is consistently higher than the regions of network through which a population crosses after an isolated environmental shift (Figures 1, 5, and 6).
A long-standing technique for identifying selected genes is to compare the frequencies of nonsynonymous and synonymous substitutions (Ka/Ks) [40]. Genes experiencing frequent selective sweeps should have relatively large amounts of variation in sites that modify amino acids. Such genes might be in the process of evolving a new function or, more likely, involved in an evolutionary arms race, for example, epitopes in human pathogens [31,41] or genes involved in sperm competition [42]. In the latter case, our model suggests that, in addition to an elevated Ka/Ks, such genes should employ a distinct set of codons with high genetic potential. Note that this type of genetic potential is not equivalent to codon bias, but rather implies changes in the actual distribution of amino acids.
A similar argument also underlies the recent use of codon distributions for detecting genetic loci under directional selection [43]. Codon volatility—the probability that a codon will mutate to a different amino acid class, relative to that probability for all codons in the same amino acid class—is a measure of genetic potential. Genes with significantly heightened volatility will be more sensitive to mutation. Our model suggests a different explanation for codon volatility than that presented in [43]: volatility may indicate a history of fluctuating selection rather than an isolated evolutionary event. If true, then we would not expect the codon distribution to reflect a transient out-of-equilibrium distribution as the population is moving from one constant environment to another [16]. Instead, we expect the distribution to reflect the stationary level of genetic potential that corresponds to variability in the selective environment for that gene. On a practical level, therefore, the isolated selective sweep model assumed in [43] may misestimate the expected volatility at such sites. Codon volatility, however, can arise as a by-product of processes other than positive (or fluctuating) selection. It has been noted that codon volatility may instead reflect selection for translation efficiency, relaxed negative selection, strong frequency-dependent selection, an abundance of repetitive DNA, or simple amino acid biases [44–48]. Therefore, the presence of codon volatility by itself may not be a reliable indicator of either recent directional selection or fluctuating selection.
We would like to emphasize that the goal of this study was not to develop a new method for detecting positive (or fluctuating) selection, but rather to develop a theoretical framework for considering the multiple outcomes of evolution under fluctuating conditions. We conclude by suggesting an empirical method to identify loci that have evolved genetic potential under such conditions as distinct from those that have experienced a recent selective sweep. Suppose that a gene experiences fluctuations at a characteristic rate across many species. Furthermore, suppose that multiple sites within the gene are influenced by such fluctuations. For example, there may be fluctuating selection for molecular hydropathy, charge, size, or polarity, and several sites within the gene may contribute to these properties. Such sites should evolve in tandem and equilibrate on similar levels of genetic potential, and thus exhibit similar codon (and amino acid) distributions across species. In contrast, if a gene experiences isolated selective sweeps, then the variation at all sites should correspond to both the history of selective events and the species phylogeny, and the amino acid distributions at sites should correlate only when sites functionally mirror each other. Thus, one can seek evidence for the evolution of genetic potential as follows. First, identify genes that are rapidly evolving, perhaps by calculating Ka/Ks ratios. Such sites have been identified, for example, in human class I MHC genes, the HIV envelop gene, and a gene from a human T cell lymphotropic virus (HTLV-1) [31,32]. Within these genes, search for sites for which there is minimal correlation between the species tree and the amino acid distribution. Our model predicts that some of these sites should share similar distributions of amino acids across species.
Materials and Methods
Mathematical analysis of models.
For the two models, we calculate the deterministic, infinite population allele frequency distributions in constant and fluctuating environments. Let M
A and M
B be the normalized transition matrices that govern changes in the allele frequencies in EA and EB epochs, respectively. The entries in these matrices are defined by equations 2 and 4. The left leading eigenvectors for M
A and M
B give the equilibrium frequency distributions of alleles in each of the two constant environments, respectively. Under fluctuating conditions with epoch duration of λ generations, we iteratively apply the matrices, and then compute the left leading eigenvector of
. This vector, which we call v
B, gives the allele frequency distribution at the end of an EA epoch followed by an EB epoch.
We are interested not only in the final allele distributions, but also in the dynamics throughout each epoch. Thus, we calculate the average frequency of each allele across a single EA epoch by
where G is the total number of alleles in the model (G = 5 for the simple model and G = 64 for the codon model) and the subscript k indicates the kth entry in the vector. Similarly, the average distribution across an EB epoch is given by
where v
A is the allele frequency distribution at the end of an EB epoch followed by an EA epoch and is equal to the left leading eigenvalue of
For the codon model, we compare these calculations that assume a regularly fluctuating environment to numerical simulations that assume a Poisson distribution of epoch lengths. In each generation of the simulations, the environmental state switches with probability 1/λ and the codon frequencies are then multiplied by the appropriate transition matrix.
Proof of 14 unique pentagonal networks.
We use an elementary group theoretic result known as Burnside's Lemma [49] to prove that there are 14 distinct mutational networks consisting of five alleles on a pentagon that map to the set of phenotypes {A, B, V} and contain at least one of each specialist phenotype (A and B) (Figure 7). We assume that all rotations and reflections of a network are equivalent to the original network, and that A and B are interchangeable. For example, the six networks with phenotypes -A-A-A-B-B-, -B-A-A-A-B-, -B-B-A-A-A-, -B-B-B-A-A-, -A-B-B-B-A-, and -A-A-B-B-B- are equivalent.
Figure 7 Pentagonal Mutational Networks
These are the 14 possible pentagonal mutational networks consisting of five alleles producing phenotypes A, B, or V, with at least one encoding A and one encoding B.
Let X be the set of all pentagons with vertices labeled {A, B, V} having at least one A vertex and at least one B vertex. The size of X is the number of all pentagons with labels {A, B, V} minus the number of pentagons with labels {A, V} or {B, V}, that is, |X| = 35 − (2 · 25 − 1) = 180.
We define the group G of all actions on X that produce equivalent pentagons (as specified above). G is made up of (1) the identity, (2) the four rotations and five reflections of the pentagon, (3) interchanging all As and Bs, and (4) all the combinations of the above actions. Thus G is equal to the 20-member group {ι, ρ, ρ2, ρ3, ρ4, σ0, σ1, σ2, σ3, σ4, α, αρ, αρ2, αρ3, αρ4, ασ0, ασ1, ασ2, ασ3, ασ4} where ι is the identity, ρ is a single (72°) rotation, σi is a reflection through vertex i, and α is replacement of all As with Bs and all Bs with As. (Note that the reflections are rotations of each other, for example, ρ2σ0 = σ1.)
The number of distinct mutational networks is equal to the number of orbits of G on X. Burnside's Lemma tells us that this number is
where F(g) = {x ∈ X | gx = x} is the set of fixed points of g. For each of the twenty elements of G, we exhaustively count F(g).
The identity fixes all elements of X, that is, F(ι) = X. Each of the various rotations of a pentagon (through 72°, 144°, 216°, and 288°) has the property that its iterations move a given vertex to every other vertex of the pentagon without changing the letter assigned to that vertex. The same is true of the square of the product of any rotation and an A–B flip. Hence, any fixed point of one of these elements of the group G would necessarily have the same label at each vertex of the pentagon. Since every labeled pentagon in X has at least one A label and at least one B label, then no element of X has the same label at each vertex. Thus, the fixed point set of every rotation and of every product of a rotation and an A–B flip must be empty, that is, F(ρn) = F(αρn) = ∅︀ for all n. By a similar argument, the simple A–B flip also has no fixed points. Every reflection fixes 12 elements of X, for example,
and every product of a reflection and an A–B flip fixes eight elements of X, for example,
In sum, all eight group elements that involve rotations fix no elements of X, all five reflections fix 12 elements of X, and all five combinations of a reflection and an A–B exchange fix eight elements of X. Thus,
We thank Carl Bergstrom and Jim Bull for their valuable insights and comments on the manuscript.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. LAM and ML conceived and designed the experiments. LAM performed the experiments. LAM, FDA, and ML analyzed the data and contributed reagents/materials/analysis tools. LAM and ML wrote the paper.
Abbreviation
MHCmajor histocompatibility
==== Refs
References
Huynen MA Stadler PF Fontana W 1996 Smoothness within ruggedness: The role of neutrality in adaptation Proc Natl Acad Sci U S A 93 397 401 8552647
van Nimwegen E Crutchfield JP Huynen MA 1999 Neutral evolution of mutational robustness Proc Natl Acad Sci U S A 96 9716 9720 10449760
De Visser JAGM Hermisson J Wagner GP Meyers LA 2003 Perspective: Evolution and detection of genetic robustness Evolution 57 1959 1972 14575319
Krakauer DC Plotkin JB 2002 Redundancy, antiredundancy, and the robustness of genomes Proc Natl Acad Sci U S A 99 1405 1409 11818563
Ancel LW Fontana W 2000 Plasticity, evolvability, and modularity in RNA J Exp Zool 288 242 283 11069142
Schlichting C Murren C 2004 Evolvability and the raw materials for adaptation Taylor I Plant adaptation: Molecular biology and ecology Vancouver NRC Canada Research Press 18 29
Meyers LA Bull JJ 2002 Fighting change with change: Adaptive variation in an uncertain world Trends Ecol Evol 17 551 557
Schlichting CD Pigliucci M 1998 Phenotypic evolution—A reaction norm perspective Sunderland (Massachusetts) Sinauer Associates 387 p.
Ancel LW 1999 A quantitative model of the Simpson-Baldwin effect J Theor Biol 196 197 209 9990740
Kawecki TJ 2000 The evolution of genetic canalization under fluctuating selection Evolution 54 1 12 10937177
Bull JJ 1987 Evolution of phenotypic variance Evolution 41 303 315
Moxon ER Rainey PB Nowak MA Lenski RE 1994 Adaptive evolution of highly mutable loci in pathogenic bacteria Curr Biol 4 24 33 7922307
Miller JH 1998 Mutators in Escherichia coli
Mutat Res 409 99 106 9875286
Baldwin JM 1896 A new factor in evolution Am Nat 30 441 451
Fontana W Schuster P 1998 Continuity in evolution: On the nature of transitions Science 280 1451 1455 9603737
Plotkin J Dushoff J Deasai M Fraser H 2004 Synonymous codon usage and selection on proteins. Arxiv.org E-Print Archives Available: http://arxiv.org/PS_cache/q-bio/pdf/0410/0410013.pdf . Accessed 3 August 2005.
Yang W Bielawski JP Yang Z 2003 Widespread adaptive evolution in the human immunodeficiency virus type 1 genome J Mol Evol 57 212 221 14562964
Bush R Bender C Subbarao K Cox N Fitch W 1999 Predicting the evolution of human influenza A Science 286 1921 1925 10583948
Crill WD Wichman HA Bull JJ 2000 Evolutionary reversals during viral adaptation to alternating hosts Genetics 154 27 37 10628966
Matsumura I Ellington AD 2001 In vitro evolution of beta-glucuronidase into a beta-galactosidase proceeds through non-specific intermediates J Mol Biol 305 331 339 11124909
Kyte J Doolittle RF 1982 A simple method for displaying the hydropathic character of a protein J Mol Biol 157 105 132 7108955
Eigen M McCaskill JS Schuster P 1989 The molecular quasispecies Adv Chem Phys 75 149 263
Wagner GP Booth G Bagheri-Chaichian H 1997 A population genetic theory of canalization Evolution 51 329 347
Wagner A Stadler PF 1999 Viral RNA and evolved mutational robustness J Exp Zool 285 119 127 10440723
Dempster E 1955 Maintenance of genetic heterogeneity Cold Spring Harb Symp Quant Biol 20 25 32 13433552
Lachmann M Jablonka E 1996 The inheritance of phenotypes: An adaptation to fluctuating environments J Theor Biol 181 1 9 8796186
Leigh EG 1973 The evolution of mutation rates Genetics 73 1 18 4265784
Meyers LA Levin BR Richardson AR Stojiljkovic I 2003 Epidemiology, hypermutation, within-host evolution, and the virulence of Neisseria meningitidis
Proc R Soc Lond B Biol Sci 270 1667 1677
Drake JW Charlesworth B Charlesworth D Crow JF 1998 Rates of spontaneous mutation Genetics 148 1667 1686 9560386
Lederberg J Iino T 1956 Phase variation in salmonella Genetics 41 743 757 17247660
Nielsen R Yang Z 1998 Likelihood models for detecting positively selected amino acid sites and applications to the HIV-1 envelope gene Genetics 148 929 936 9539414
Yang Z Wong WSW Nielsen R 2005 Bayes empirical Bayes inference of amino acid sites under positive selection Mol Biol Evol 22 1107 1118 15689528
Wagner GP Altenberg L 1996 Perspective: Complex adaptations and the evolution of evolvability Evolution 50 967 976
Fisher RA 1922 On the dominance ratio Proc R Soc Edinb 42 321 341
Wright S 1934 Physiological and evolutionary theories of dominance Am Nat 68 24 53
Sniegowski PD Gerrish PJ Johnson T Shaver A 2000 The evolution of mutation rates: Separating causes from consequences Bioessays 22 1057 1066 11084621
Feldman MW Otto SP Christiansen FB 1997 Population genetic perspectives on the evolution of recombinations Annu Rev Genet 30 261 295
Schlosser G Wagner GP 2004 Modularity in development and evolution Chicago University of Chicago Press 600 p.
Rutherford SL Lindquist S 1998 Hsp90 as a capacitor for morphological evolution Nature 396 336 342 9845070
Yang Z Bielawski J 2000 Statistical methods for detecting molecular adaptation Trends Ecol Evol 15 496 503 11114436
Endo T Ikeo K Gojobori T 1996 Large-scale search for genes on which positive selection may operate Mol Biol Evol 13 685 690 8676743
Torgerson DG Kulathinal RJ Singh RS 2002 Mammalian sperm proteins are rapidly evolving: Evidence of positive selection in functionally diverse genes Mol Biol Evol 19 1973 1980 12411606
Plotkin JB Dushoff J Fraser HB 2004 Detecting selection using a single genome sequence of M. tuberculosis and P. falciparum
Nature 428 942 945 15118727
Dagan T Graur D 2005 The comparative method rules! Codon volatility cannot detect positive Darwinian selection using a single genome sequence Mol Biol Evol 22 496 500 15525696
Hahn MW Mezey JG Begun DJ Gillespie JH Kern AD 2005 Evolutionary genomics: Codon bias and selection on single genomes Nature 433 E5 E6 15662370
Nielsen R Hubisz MJ 2005 Evolutionary genomics: Detecting selection needs comparative data Nature 433 E6
Sharp PM 2005 Gene “volatility” is most unlikely to reveal adaptation Mol Biol Evol 22 807 809 15616138
Zhang J 2005 On the evolution of codon volatility Genetics 169 495 501 15466418
Martin G 2001 Counting: The art of enumerative combinatorics New York Springer-Verlag
Batagelj V Mrvar A 1998 PAJEK—Program for large network analysis Connections 21 47 57
|
16158095
|
PMC1193991
|
CC BY
|
2021-01-05 09:18:22
|
no
|
PLoS Comput Biol. 2005 Aug 26; 1(3):e32
|
utf-8
|
PLoS Comput Biol
| 2,005 |
10.1371/journal.pcbi.0010032
|
oa_comm
|
==== Front
PLoS Comput BiolPLoS Comput. BiolpcbiplcbploscompPLoS Computational Biology1553-734X1553-7358Public Library of Science San Francisco, USA 1675800310.1371/journal.pcbi.001003305-PLCB-MI-0180plcb-01-03-08Message from ISCBISMB 2005 Conference Report Message from ISCBMorrison McKay B. J B. J. Morrison McKay is at the ISCB, La Jolla, California, United States of America. E-mail: [email protected]
8 2005 26 8 2005 1 3 e33Copyright: © 2005 B. J. Morrison McKay.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.Citation:Morrison McKay BJ (2005) ISMB 2005 conference report. PLoS Comp Biol 1(3): e33.
==== Body
The ISCB held its 13th annual Intelligent Systems for Molecular Biology Conference (ISMB) in Detroit, Michigan, June 25–29, 2005. Considering this conference has moved to many points on the globe in its 13 years, this location was not so far away geographically from the first ISMB held in Bethesda, Maryland, in 1993, but the increase in numbers in all categories—attendance, presentations, and special intrest group meetings and tutorials—put it in a separate stratosphere altogether (see Table 1).
Table 1 ISMB Growth since Inception
ISMB/ECCB 2004, held in Scotland last year as a joint conference with the European Conference on Computational Biology (ECCB), resulted in the highest attendance of any ISMB or any ECCB on record, with 2,136 total attendees. Given a variety of challenges, and given that the meeting in Detroit was not co-located with ECCB or any other conference in our field, ISMB 2005 drew attendees surprisingly well. In fact, ISMB 2005 had the second-highest attendance ever, and was the most highly attended North American ISMB to date.
As in years past, seven special interest group (SIG) meetings preceded this year's conference, enabling registrants to come for both the SIGs and the main conference, or attend just the SIGs. The SIGs are fast becoming one of the main attractions of ISMB, with topics as varied as the “Bioinformatics Open Source Conference,” “Alternative Splicing,” and “Bioinformatics and Disease.” Some 653 delegates arrived in Detroit early for these one- or two- day focused sessions—485 of whom then stayed on for ISMB. The “Alternative Splicing” SIG had the highest number of attendees, but the beauty of this year's SIG arrangements was the ability for all registrants to move among SIGs to capture more of a knowledge base on the varied topics. If you'd like to view the full list of options that were available this year, the preconference SIG abstracts are available at http://www.iscb.org/ismb2005/sigs.html.
Also serving as a precursor to the main conference were 14 half-day tutorial sessions that were attended by 548 delegates. New and interesting tutorials are sought for each ISMB, and this year's tutorial abstracts on the conference Web site (http://www.iscb.org/ismb2005/tutorials.html) include a link to the original proposal selected for one of the limited tutorial slots. If you are interested in presenting a tutorial proposal for 2006, we encourage you to follow the links at http://www.iscb.org/ismb2005/tutorials.html to gather ideas on preparation and submission of tutorials. Then watch for the call for tutorials via E-mail and on the ISMB 2006 Web site.
A very special few were recognized for their outstanding papers with the GlaxoSmithKline Bioinformatics Prizes for best paper. ISCB congratulates the winners (see Box 1). ISCB also bestowed its two highest awards of recognition during the conference. The Overton Prize, for a scientist in the early-to-mid-career stage, went to Ewan Birney, and the Senior Scientist Accomplishment Award went to Janet Thornton, both from the European Bioinformatics Institute (EBI). This was the first time both awards had gone to scientists at the same institution—a remarkable statement about the quality of science being generated from the EBI. Both award winners gave keynote lectures on the final day of ISMB—a very fitting way to end a successful conference.
Please commit to joining us for ISMB 2006 in Fortaleza, Cearà, Brazil, August 6–10 (http://www.iscb.org/ismb2006). Program co-chairs Phil Bourne, Editor-in-Chief of PLoS Computational Biology and Professor of Pharmacology at University of California at San Diego, and Søren Brunak, Director of the Center for Biological Sequence Analysis of the Technical University of Denmark, are working together with conference chair Goran Neshich of Empresa Brasileira de Pesquisa Agropecuária to develop one of the most interesting and innovative ISMBs yet. Additionally, the 20th Anniversary Swiss-Prot conference is co-locating to take place just prior to ISMB, and several Nobel laureates have already committed to speaking at each conference. Brazil has a strong lure, with its endless coastline of white-sand beaches, but the lure will be all about the science as we prepare to roll out an exceptional ISMB in the coming year. As 97% of this year's postconference-survey respondents indicated they will attend an ISMB again, we're counting on seeing you in Fortaleza!
Box 1. GlaxoSmithKline Bioinformatics Winners
Best Papers:
Matthew Dimmic Dimmic MW, Hubisz MJ, Bustamante CD, Nielsen R (2005) Detecting coevolving amino acid sites using bayesian mutational mapping. Bioinformatics 21: il26–il35.
John Spouge
Tharakaraman K, Mariño-Ramirez L, Sheetlin S, Landsman DL, Spouge J (2005) Genomic landmarks can aid in the identification of regulatory elements. Bioinformatics 21: i440–i448.
Best Student Papers:
Elena Nabieva Nabieva E, Jim K, Agarwal A, Chazelle B, Singh M (2005) Whole-proteome prediction of protein function via graph-theoretic analysis of interaction maps. Bioinformatics 21: i302–i310.
Alena Shmygelska Shmygelska A (2005) Search for folding nuclei in native protein structures. Bioinformatics 21: i394–i402.
Honorable Mention Paper:
Charlotte Deane Winstanley WF, Abeln S, Deane CM (2005) How old is your fold? Bioinformatics 21: i449–i458.
Honorable Mention Student Paper:
Shaun Mahony Mahony S, Golden A, Smith TJ, Benos PV (2005) Improved detection of DNA motifs using a self-organized clustering of familial binding profiles. Bioinformatics 21: i283–i291.
|
16758003
|
PMC1193992
|
CC BY
|
2021-01-05 09:18:22
|
no
|
PLoS Comput Biol. 2005 Aug 26; 1(3):e33
|
utf-8
|
PLoS Comput Biol
| 2,005 |
10.1371/journal.pcbi.0010033
|
oa_comm
|
==== Front
PLoS Comput BiolPLoS Comput. BiolpcbiplcbploscompPLoS Computational Biology1553-734X1553-7358Public Library of Science San Francisco, USA 10.1371/journal.pcbi.001003405-PLCB-PV-0128R2plcb-01-03-09PerspectivesBioinformatics - Computational BiologyPerspectiveWill a Biological Database Be Different from a Biological Journal? PerspectiveBourne Philip Philip E. Bourne is Editor-in-Chief of PLoS Computational Biology and is Codirector of the Protein Data Bank. E-mail: [email protected] 2005 26 8 2005 1 3 e34Copyright: © 2005 Philip Bourne.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.Citation:Bourne P (2005) Will a biological database be different from a biological journal? PLoS Comp Biol 1(3): e34.
==== Body
The differences, or otherwise, between biological databases and journals is an important question to consider as we ponder the future dissemination and impact of science. If databases and journals remain discrete, our methods of assimilating information will change relatively little in the years to come. On the other hand, if databases and journals become more integrated, the way we do science could change significantly. As both Editor-in-Chief of PLoS Computational Biology and Codirector of the Protein Data Bank (PDB), one of the oldest and widely used data resources in molecular biology, the question is particularly pertinent. Here, I give my perspective on what could and, I believe, should happen in the future.
My vision is that a traditional biological journal will become just one part of various biological data resources as the scientific knowledge in published papers is stored and used more like a database. Conversely, the scientific literature will seamlessly provide annotation of records in the biological databases. Imagine reading a description of an active site of a biological molecule in a paper, being able to access immediately the atomic coordinates specifically for that active site, and then using a tool to explore the intricate set of hydrogen-bonding interactions described in the paper. Not only are the data generated by the experiment immediately available within the context of what you are reading, but specific tools for interpreting these data are provided by the journal. Alternatively, if you are starting with the data, for example, viewing the chromosome location of a human single-nucleotide polymorphism associated with a neurological disorder, you can immediately access a variety of papers ranked in order of relevance to your profile, not just through links to abstracts but also by pinpointing the reference to the single-nucleotide polymorphism in the full-text article. The type and order of articles displayed could be different, depending on whether you are, for example, a molecular biologist or a neurosurgeon. At this point, whatever your user profile, the distinction between a database and a journal article disappears. How could this happen? To answer this question, we must think about the parallels that exist today between biological databases and biological journals.
The daily work of any high-throughput scientific journal or biological database consists of information input, information processing, and information output. Consider the parallels between a journal and a database for each of these three steps. On a daily basis, the journal accepts manuscripts; once these have been checked for format compliance and completeness, they undergo review, either by an internal group of scientific editors or, as is the case for PLoS Computational Biology, through peer review by the scientific community. Likewise, a biological database such as the PDB accepts submissions from the community, which are checked for format compliance and reviewed internally by experienced annotators. There are even parallel presubmission steps in journals and databases. For example, potential authors in PLoS Computational Biology may make presubmission inquiries to confirm the suitability of their paper, and depositors to the PDB may run their entries against a validation server to determine whether the data are in compliance, prior to having the same tests run by a PDB annotator.
Once registered with the corresponding online submission system, a journal manuscript receives a permanent manuscript number, while a database entry receives a unique identifier. Subsequent revisions can be mapped to these respective numbers, so that both journals and databases can provide an accurate audit trail of journal manuscripts and database entries, respectively. Once a manuscript or entry is accepted as compliant, both undergo review processes involving one or more iterative steps between institution and author, as the manuscript or the entry is refined and finally released. Release cycles of journals and databases have also become similar—journals such PLoS Computational Biology have an option for early online release as soon as the manuscript is accepted, and biological databases typically release entries on a daily or weekly basis, as soon as they have been processed.
Not only are the daily operations of databases and journals similar, but the business models also have parallels (I will not dwell on them here though). Certainly from a consumer's perspective, in terms of accessibility, there is no difference between a paper in a PLoS journal and an entry in the PDB database—they are freely available to all. In the case of open-access journals and open archives like the PDB, the parallels, from the perspective of the consumer, are even more profound than just free access yet are frequently overlooked. PLoS articles are published under a Creative Commons Attribution License, which means that the contents (text and images) of all PLoS journals can be used as the consumer sees fit, provided original attribution is given to the appropriate authors and source. So it is with the contents of many biological databases, including the PDB. Consumers are free to take and analyze the contents by any means they see fit, but are expected to attribute information to the authors of the original material, as appropriate. Finally, in the case of PLoS journals, the copyright of the material is not signed over to the publisher but remains with the original author, which is also true of information provided to most biological databases. In both forms of open access—journals and databases—the only requirement is to provide an immutable reference to the material. In the case of an online journal article, this reference most often takes the form of a digital object identifier (DOI), and for a database entry, it is usually a unique accession number. Like the contents of manuscripts and database entries, I expect these two forms of immutable identifiers to become indistinguishable from each other, as I will outline subsequently.
Given these parallels, at this point in time, what is the difference between an entry in a database and an article in a journal? Currently the difference can be characterized as a mix of perception and content. Clearly, no one perceives a database entry of, say, a sequence, or a specimen in a museum collection, as being as valuable as the journal paper that describes it. But, ironically, to the consumer, at least by one measure, the database entry may indeed be more valuable. The structure of human deoxyhemoglobin is one of the most downloaded structures in the PDB—in one year, it has been downloaded more times than the original paper has ever been cited thus far. Yet from the authors' perspective, the Nobel Prize does not come from constructing the PDB database entry, but from an eloquent description of the relationship between structure and function that was presented most completely in the literature. A tenure committee does not award tenure based on the number of deposits a faculty member has made to a biological database, but rather the number of papers they have published in leading journals.
Those of you who have made it this far might be thinking it is ridiculous that I should regard the content of a database entry in the same way that I regard the content of a scientific paper, given these differences in perception and content. It is possible, though, that you are thinking this way based on traditional perceptions of content and not the way things should be, going forward, given current technologies and social practices. To set the stage for the subsequent discussion, I will highlight three current observations that are relevant to this assertion.
First, publishers have embraced the Internet as a distribution medium but, for the most part, have not used the medium beyond that, simply distributing material in the same way as in printed form. Hyperlinks in documents and citation indexes are exceptions, but compared to what many biological database developers have achieved in terms of information integration and comprehension through novel display techniques, such added functionality is minimal. Second, online journals have greatly reduced the necessity for page limits on papers, since the costs of supporting a long versus short paper are much less online than in the printed form. Journals publishing both online and in print solve this size problem by having short articles in print and placing additional material as supplements in an online form only. This practice has increased dramatically in the past few years: consider the amount of supplementary material in one issue of the Proceedings of the National Academy of Sciences of the United States of America today versus five years ago. Supplementary material can be a valuable addition or, alternatively, can make for a disjointed piece of work. Moreover, the supplemental material is ad hoc and cannot be readily queried across all articles, even though a small amount of it is already tagged and comes directly from a database. Third, the perceived value of both a database entry and a journal article has changed over the years. As high-throughput techniques have become more prevalent, data are produced at an ever-increasing rate, so the value of a unit of data, for example, a sequence or structure, has diminished. Data producers hoard their data less than they did in past years. Similarly, the rate of publication has increased dramatically, this increase being brought about by accelerated technologies for manuscript production, large collaborative studies, and increased emphasis on the notion of “publish or perish.” In short, journal content is already becoming more like database content and vice versa.
Can this trend continue? Consider how the respective content of journals and databases is organized. Both have varying degrees of content organization. Papers have structure, but the organization of their content is less detailed than that found in a database, although this is changing with formal document type definitions being applied, from which database schema can be generated. Typically a paper has an introduction, a materials and methods section, a results section, and a discussion section; it possibly uses consistent terms for genes, enzymes, and diseases; and in a post-production step, keywords and/or medical subject headings for indexing the content of the article are added. Databases, on the other hand, frequently have a high level of organization, where data are granular and each granule is described in exquisite detail. The advantage of a paper is that it is relatively easy to input and maintain, but it requires human recall. Machine-based recall of meaningful information is poor, a problem being addressed but certainly not solved by the discipline of natural-language processing. A database, on the other hand, has excellent recall but requires much effort to organize and is best suited to quantitative data, not free text. I would contend that the future offers some middle ground for content organization.
We have taken the first steps toward a middle ground by making both the combined contents of biological databases and biological literature freely available in electronic form. Is the technology available to support the next steps in integration and is the scientific community ready for such a change? I believe that the answer to the technology part of the question is yes. I do not know the answer to the second part, but I think it's time for some preliminary experiments to find out. I would be most interested in hearing views on the matter and any suggestions for potential experiments. In the interim, here are a few experiments I am proposing.
As mentioned above, DOIs provide an immutable reference to a scientific document that exists online. The way I think about DOIs is the same way I think about addresses used to identify computers on the internet, each address possesses a unique identifier that in a seamless way can be resolved to access that specific computer. So it is with DOIs, which can be resolved not only to find the material referenced by the DOI but, through reverse searching, can also be used to find material that references the DOI. Think of what could happen if such DOIs were not only assigned to papers as they are now, but also to items of content within biological databases—protein structures, species distributions, neuroimaging datasets, and so on—and if these DOIs were referenced when that content was used or discussed elsewhere. An immediate outcome would be the ability to find all papers that reference a particular sequence motif, for example: a level of detail that is not currently available to someone accessing a sequence database. Conversely, accessing a paper would immediately provide a resolvable list of the sources of data used in the experiments, which could be accessed and further analyzed—a step toward achieving true reproducibility of an experiment, where the paper has become the interface to the data. Unfortunately, DOIs cost money, and providing a fine level of granularity, such as all sequence motifs for every sequence in the Protein Families Database of Alignments and HMMs, would be prohibitively expensive. Publishers should collaborate with the major database providers, so that database providers provide the appropriate immutable references and published articles reference them.
As another experiment, what if the data in an online paper became more alive? Some databases let you download data into spreadsheets or other client-side applications that render and analyze data. Papers could be treated this way, too. The technology is there to create these ubiquitous clients that are independent of operating systems and hardware and that are downloadable on demand. New levels of comprehension might be achievable. The first step would be to provide tools that better visualize specific types of biological data, without the need for specialized knowledge in using an esoteric tool. Later would come tools for basic analysis, for example, simple statistical tests or principle-component analysis.
Consider one final experiment, what if papers were made to show a higher level of organization than is possible today? Clearly, too much additional work by the author would be resisted, unless it bought clear rewards. Nevertheless, tools can be envisaged that, with minimal work by the author, would further classify the text such that, for example, annotation associated with a particular gene or set of genes is identified, or a set of keywords is generated to be associated with the paper as metadata, and all the author would have to do is confirm their validity. Recent benchmarks indicated that 80% of terms such as gene names could be identified automatically and hence associated with systematic annotation, which could simply be accepted or rejected by the author [1]. Would an author do it, if it led to more rapid citations? I would say so! This type of experiment has already proved to be successful in the community engaged in small-molecule structure determination, although without the data being publicly accessible in an easy way. With the incentive for more citations, the author would review the proposed systematic nomenclature, and we would then have the potential for a new association between the text of a paper and, say, a gene and the description of that gene in a database. If the connection is transparent to the reader, the paper has thus become a detailed entry point to the database and the database has become a detailed entry point to the literature.
These experiments, if successful, would go a long way in answering the question posed here—Is a biological database any different than a biological journal? I am working toward reaching an answer of, no, there is no difference. If you want to help answer this question, I would welcome hearing from you; after all, journals, like databases, should be community resources.
Thanks to the PLoS team of Johanna McEntyre, Catherine Nancarrow, Mark Patterson, Steven Brenner, and Michael Eisen for useful input.
Abbreviations
DOIdigital object identifier
PDBProtein Data Bank
==== Refs
Reference
Hirschman L Yeh A Blaschke C Valencia A 2005 Overview of BioCreAtIvE: Critical assessment of information extraction for biology BMC Bioinformatics 6 S1
|
16158097
|
PMC1193993
|
CC BY
|
2021-01-05 09:18:22
|
no
|
PLoS Comput Biol. 2005 Aug 26; 1(3):e34
|
utf-8
|
PLoS Comput Biol
| 2,005 |
10.1371/journal.pcbi.0010034
|
oa_comm
|
==== Front
PLoS Comput BiolPLoS Comput. BiolpcbiplcbploscompPLoS Computational Biology1553-734X1553-7358Public Library of Science San Francisco, USA 10.1371/journal.pcbi.0010038plcb-01-03-10CorrectionCorrection: Improving the Precision of the Structure–Function Relationship by Considering Phylogenetic Context CorrectionShakhnovich Boris E 8 2005 26 8 2005 1 3 e38Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Improving the precision of the structure-function relationship by considering phylogenetic context
==== Body
DOI: 10.1371/journal.pcbi.0010009
In PLoS Computational Biology, vol 1, issue 1.
The Editor for this research article was incorrectly given as Philip Bourne. The Editor was Burkhard Rost, Columbia University, United States of America.
This correction note may be found online at DOI: 10.1371/journal.pcbi.0010038.
Published August 26, 2005
Citation: (2005) Correction: Improving the precision of the structure–function relationship by considering phylogenetic context. PLoS Comp Biol 1(3): e38.
|
0
|
PMC1193994
|
CC BY
|
2021-01-05 09:18:22
|
no
|
PLoS Comput Biol. 2005 Aug 26; 1(3):e38
|
utf-8
|
PLoS Comput Biol
| 2,005 |
10.1371/journal.pcbi.0010038
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 1612235210.1371/journal.pbio.0030328Research ArticleBioinformatics/Computational BiologyCell BiologyGenetics/Genomics/Gene TherapyMolecular Biology/Structural BiologySystems BiologySaccharomycesSingle-Nucleosome Mapping of Histone Modifications in S. cerevisiae
Global Histone Modification PatternsLiu Chih Long
1
2
Kaplan Tommy
3
4
Kim Minkyu
5
Buratowski Stephen
5
Schreiber Stuart L
2
Friedman Nir
3
Rando Oliver J [email protected]
1
1Bauer Center for Genomics Research, Harvard University, Cambridge, Massachusetts, United States of America,2Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts, United States of America,3School of Computer Science and Engineering, The Hebrew University, Jerusalem, Israel,4Department of Molecular Genetics and Biotechnology, The Hebrew University, Jerusalem, Israel,5Department of Biological Chemistry and Molecular Pharmacology, Harvard University, Boston, Massachusetts, United States of AmericaBecker Peter Academic EditorAdolf Butenandt InstituteGermany10 2005 30 8 2005 30 8 2005 3 10 e32810 5 2005 16 7 2005 Copyright: © 2005 Liu et al.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
A Global View of DNA-Packing Proteins Cracks the Histone Code
Covalent modification of histone proteins plays a role in virtually every process on eukaryotic DNA, from transcription to DNA repair. Many different residues can be covalently modified, and it has been suggested that these modifications occur in a great number of independent, meaningful combinations. Published low-resolution microarray studies on the combinatorial complexity of histone modification patterns suffer from confounding effects caused by the averaging of modification levels over multiple nucleosomes. To overcome this problem, we used a high-resolution tiled microarray with single-nucleosome resolution to investigate the occurrence of combinations of 12 histone modifications on thousands of nucleosomes in actively growing S. cerevisiae. We found that histone modifications do not occur independently; there are roughly two groups of co-occurring modifications. One group of lysine acetylations shows a sharply defined domain of two hypo-acetylated nucleosomes, adjacent to the transcriptional start site, whose occurrence does not correlate with transcription levels. The other group consists of modifications occurring in gradients through the coding regions of genes in a pattern associated with transcription. We found no evidence for a deterministic code of many discrete states, but instead we saw blended, continuous patterns that distinguish nucleosomes at one location (e.g., promoter nucleosomes) from those at another location (e.g., over the 3′ ends of coding regions). These results are consistent with the idea of a simple, redundant histone code, in which multiple modifications share the same role.
High-resolution microarrays were used to investigate 12 histone modifications across thousands of yeast nucelosomes in vivo. Two main groups co-occurred, consistent with the redundant histone code hypothesis.
==== Body
Introduction
Nucleosomes play many roles in transcriptional regulation, ranging from repression through occlusion of binding sites for transcription factors [1], to activation through spatial juxtaposition of transcription factor-binding sites [2]. There are two main ways in which cells modulate nucleosomal influences on gene expression. One way is through chromatin remodelling, using the energy of adenosine triphosphate hydrolysis to modulate nucleosomal structure, often resulting in changed nucleosomal location [3]. Alternatively, covalent histone modifications have many effects on transcription. Histone proteins have highly conserved tails, which are subject to multiple types of covalent modification, including acetylation, methylation, phosphorylation, ubiquitination, sumoylation, and adenosine-diphosphate ribosylation [4–9].
Histone acetylation has been the subject of decades of research, whereas histone methylation has come under intense scrutiny more recently. Lysine acetylation neutralizes lysine's positive charge, and can influence gene expression in at least two ways. Firstly, charge neutralization can affect contacts between the positively charged histone tail and negatively charged neighbouring molecules, such as adjacent linker DNA [10], or acidic patches on histones in nucleosomes [11]. Alternatively, acetyl-lysine is bound by the bromodomain, a protein domain found in many transcriptional regulators; thus, acetylation might affect recruitment of protein complexes [12]. Histone acetylation is rapidly reversible, and acetyl groups turn over rapidly in vivo, with half-lives on the order of minutes [13], allowing for rapid gene expression changes in response to signals [14]. Acetylation of histone lysines has been associated with both transcriptional activation and transcriptional repression [15–17]. The outcome of acetylation depends on which lysine is acetylated and the location of the modified nucleosome. A recent genome-scale study of histone acetylation in yeast revealed a complicated relationship between histone modification and transcriptional output [18].
Histone methylation has been best characterized by histone 3-lysine 4 (H3K4), wherein methylation is associated with active transcription in multiple organisms, ranging from Saccharomyces cerevisiae to mammals. Lysine can be mono-, di-, or tri-methylated, and none of these methylation states will alter lysine's positive charge (under conditions of standard lysine pKa and physiological pH). As a result, it is unlikely that charge–charge interactions are modulated by methylation, which appears instead to affect cellular processes through binding of methyl-lysine–binding proteins. Indeed, methyl-lysine is bound by at least one domain type—the chromodomain [19,20]. In contrast to histone acetylation, histone methylation is long-lived. Although a histone-lysine demethylase (termed LSD1) was recently identified in metazoans. S. cerevisiae does not have a homolog of this protein. Even in metazoans, the proposed enzymatic mechanism allows for demethylation of mono- and di-methylated lysine, but not of tri-methylated lysine [21]. Whether or not enzymatic demethylation of tri-methyl-lysine occurs, and whatever other mechanisms allow for replacement of tri-methylated histones (such as histone replacement—[22]), in yeast, H3K4 tri-methylation is associated with active transcription. The histone tri-methylation persists for over an hour after transcription ceases, providing a memory of recent transcription [23].
The discovery of multiple modification types and modified residues suggested that different combinations of histone modifications might lead to distinctive transcriptional outcomes. According to the “histone code” hypothesis, “distinct histone modifications, on one or more tails, act sequentially or in combination to form a ‘histone code' that is read by other proteins to bring about distinct downstream events” [6].
This hypothesis has been the subject of much debate, much of it concerning the requirements for histone modifications to form a “code” [4–9]. In this study, we focused on the combinatorial complexity of histone modification patterns. Insights into this complexity require an understanding of which combinations of modifications occur in vivo, and the functional consequences of these combinations. Mutagenesis of histone tails has demonstrated that not all combinations of histone modifications lead to distinct transcriptional states [24]. In addition, genome-wide localization studies of histone modifications in yeast, flies, and mammals have demonstrated that not all possible histone-modification patterns occur in vivo [18,25,26].
A major confounding effect in the interpretation of previous genome-wide studies of histone modifications in vivo is the low resolution of the measurements (~500–1,000 base pairs [bp]) relative to the size of the nucleosome (~146 bp). Thus, the measured ratio for a given spot represents an aggregate that is actually an average of information from several nucleosomes, which complicates analysis. Furthermore, in some studies, acetylation patterns at intergenic and coding regions were measured using different microarrays, precluding a common reference point. Finally, whole genomic DNA has typically been used as the reference DNA in these microarray studies, thereby confounding the measurements of histone modification with underlying variation in nucleosome density [27,28].
To overcome these limitations, we made use of a recently developed, high-density oligonucleotide microarray with ∼20-bp resolution. We recently used this microarray to map nucleosome positions across almost half a megabase of the yeast genome [29]. In this study, we use this microarray to measure the levels of 12 different histone modifications in individual nucleosomes. We find that modifications do not occur independently of each other and that a small number of distinct combinations occur in vivo. Different modification patterns are enriched at specific locations in gene or promoter regions, and these patterns are predictive of the transcription level of the underlying gene. Sharp transitions in histone modifications mostly occur near the transcription start site (TSS). Together these results provide a simpler view of histone modification, and suggest that there is little combinatorial information encoded in the histone tails.
Results
High-Resolution Measurement of Histone Modifications Using Tiled Microarrays
Chromatin immunoprecipitation (ChIP) using modification-specific antibodies [30,31] was used to map histone modifications in actively growing yeast cultures. We used a standard ChIP protocol, with one major modification (Figure 1A). In our protocol, formaldehyde-fixed yeast were lysed gently by spheroplasting and osmotic lysis rather than by glass beads, and DNA was digested to mononucleosomes using micrococcal nuclease (rather than sheared to ~500 bp by sonication) (Figure S1). This allowed us to map modifications at nucleosomal resolution. We used antibodies specific to 12 individual modifications, including mono-, di-, and tri-methylation of histone H3K4, as well as acetylation of various lysines on all four histones. Immunoprecipitated DNA was isolated, linearly amplified [32], and labelled with Cy5 fluorescent dye, while mononucleosomal DNA treated under identical conditions was used as the “input” and labelled with Cy3. This choice of input served to control for nucleosomal occupancy differences (to prevent highly modified, low-occupancy nucleosomes from appearing to be poorly modified nucleosomes), as it has been shown that nucleosomes are not always present in every cell in a population [33,34]. Mixtures were hybridized to a tiled microarray covering half a megabase of yeast genomic sequence, including almost all of Chromosome III as well as 230 additional 1-kb promoter regions [29]. This represents approximately 4% of the yeast genome, and includes a total of 356 promoter regions. Finally, to measure active transcription (while avoiding effects of mRNA instability that influence mRNA abundance measurements), we also immunoprecipitated DNA associated with RNA polymerase II (this DNA was sheared by sonication rather than cut with micrococcal nuclease) [35].
Figure 1 Overview
(A) Nucleosomes are first cross-linked to DNA using formaldehyde. Cross-linked chromatin is digested to mononucleosomes with micrococcal nuclease. Mononucleosomal digests are immunoprecipitated using an antibody specific to a particular histone modification, and immunoprecipitated DNA is isolated and labelled with Cy5. DNA is also isolated from the same nuclease titration step prior to immunoprecipitation, labelled with Cy3, and mixed with Cy5-labeled immunoprecipitated DNA. Labelled DNA is then hybridized to a tiled microarray covering half a megabase of yeast genome.
(B) Example of raw data. Data are shown for all modifications tested, along with PolII data. Red (green) indicates enrichment (depletion), while grey indicates missing data. Data from probes found in linker regions are not shown. Each row represents median data from multiple replicates with one antibody, as indicated (PanAc refers to a nonspecific antibody to acetyl-lysine, which we used to measure bulk acetylation). “Nucleosomes” shows positions of nucleosomes previously described [29], with dark brown for well-positioned nucleosomes, very light brown for linkers, and intermediate brown for delocalized nucleosomes. “ORFs” shows locations of annotated genes. Data shown are for Chromosome III coordinates 58,900 to 72,100.
A Chromosomal View of Histone Modifications
The resulting data provide a rich view of histone modification over half a megabase of yeast sequence, demonstrating several prominent features (Figure 1B shows a sample stretch). First, histone modifications generally occur in broad domains, and there are few examples of nucleosomes whose modification pattern was significantly different from that of their adjacent nucleosomes. This was not due to limitations in the experimental technique, as we did find multiple examples of punctate nucleosomes that occurred in expected locations (see below). Second, modifications were generally homogeneous for all the probes within a given nucleosome. Third, correlations could be observed between a nucleosome's position relative to coding regions and its modification pattern. For example, most of the open reading frames shown in Figure 1B exhibit a striking pattern of histone H3K4 methylation, with tri-methylation occurring at the 5′ end of the coding region, shifting to di-methylation, and then to mono-methylation. This pattern is clear over most expressed open reading frames on Chromosome III, and is consistent with reports that Set1 association with RNA polymerase is responsible for methylation of this lysine [23,36]. Finally, we noticed broad domains of low acetylation occurring over heterochromatic regions on our array—subtelomeric sequences and the silent mating type loci [37] (Figure S2).
Coupling of Modifications to Organization of Transcriptional Units
To analyze the relationship of different modifications to the underlying sequence, we aligned all genes (and their promoters) by their start codon. For example, Figure 2A shows data for histone H4K16 acetylation on aligned genes that were clustered to highlight patterns (see Materials and Methods). Clearly notable in this representation is a hypo-acetylated domain adjacent to most start codons. We have recently discovered that TSSs are found in long nucleosome-free regions [29]. By aligning genes by the location of the first nucleosome following the TSS, a clear domain of two hypo-acetylated nucleosomes can be observed at most PolII promoters (Figure 2B). This alignment, therefore, provides a highly informative view of the relationship of histone modifications to the underlying structure of the genome (see Figure S3 for the remaining modifications).
Figure 2 Broad Patterns of Histone Modifications
(A) H4K16Ac aligned by ATG. In this representation, the horizontal axis represents location relative to the downstream gene's start codon, and each horizontal line represents one PolII-driven gene. Each cell in the resulting matrix corresponds to the acetylation level at a given microarray probe for one tail position. Red (green) cells mark hyper-acetylated (hypo-acetylated) probes. Non-nucleosomal probes are blackened. We clustered the promoters using a probabilistic agglomerative clustering algorithm (see Materials and Methods). Arrow indicates annotated ATG.
(B) H4K16 aligned by transcriptional start site, as in (A), except that arrow indicates TSS (identified in [29]) and data before and after the TSS are aligned by the first nucleosome in that direction.
(C) Relationship of histone modification patterns to transcription level. Genes were split into three groups based on PolII enrichment, and averaged data for these groups are shown as indicated, aligned as in (B). Transcription level is indicated by red triangles to the left of each set of three rows.
To explore the relationship of these modifications to transcription, we separated genes into “bins” of varying transcriptional activity (see Materials and Methods) and averaged the enrichment data for all aligned genes in each bin (Figures 2C and S4). Several previously identified features of yeast chromatin are apparent. First, histone H3K4 methylation enrichment correlates with transcription levels, and occurs in a 5′ to 3′ gradient (as also seen in Figure 1B) with tri-methyl enrichment at the 5′ end of genes, shifting to di-methyl and then mono-methyl. Histone H3K4 is methylated by Set1, which is associated with elongating RNA polymerase [23,36], and, as noted above, this gradient presumably reflects the kinetics of dissociation of Set1 from the polymerase, convoluted with the ensemble-average location of polymerase. Second, we reproduced previous observations that histone H3K9/K14 acetylation is enriched over the 5′ ends of coding regions [26,38].
Figure 2C also reveals novel locations of particular histone modification patterns. In particular, the two-nucleosome hypo-acetylation domain described above for H4K16 acetylation is surprisingly general, and a nearly identical pattern is also seen for acetylation of H4K8 and of H2B K16 (Figures S3 and 2C). This hypo-acetyl domain does not correlate with transcription levels (as measured by either PolII occupancy or by mRNA abundance [Figures 2C and S4]). Also, the acetylation of these residues at the middle and 3′ ends of coding regions is either uncorrelated (H2BK16) or anticorrelated (H4K8 and K16) with transcription (Figure 2C). We will therefore refer to this group of modifications as the transcription-independent modifications, for convenience (and to emphasize the stereotyped promoter-deacetyl domain). A two-nucleosome hypo-acetylation domain is also present at a smaller subset of promoters for the remaining acetylation states, and is generally found preferentially in poorly expressed genes (Figures S3 and 2C). However, the acetylation of these lysines is found at the 5′ end of coding regions, whereas acetylation of the transcription-independent group is largely excluded from 5′ coding regions. We will refer to this 5′-directed group of modifications as the transcription-dependent modifications. Acetylation of H2A K7 is an interesting case, as its pattern appears to be a mixture of the two types of patterns described. However, we have recently found that the H2A isoform Htz1 is enriched in a pattern that dramatically parallels the hypo-acetylation domain observed for the transcription-independent modifications (unpublished data), so H2A is expected to be depleted in this region. This, coupled with the 5′-enrichment of acetylation seen for H2A K7, in highly transcribed genes, leads us to include this modification in the transcription-dependent group.
Low Dimensionality of Nucleosome Modification Patterns
The analysis presented above is highly informative, but is based on aggregated data for many promoters, and thus may obscure interesting underlying phenomena. A more informative approach would be to examine the distinct modification patterns at individual nucleosomes. We defined the modification pattern of each nucleosome as the median hybridization value, for each measured antibody, of the probes associated with the nucleosome (usually between six and 15 probes; see Materials and Methods). In addition, we classified nucleosomes according to their positions relative to genome annotations (Figure 3A; see Materials and Methods). We used nine annotation categories that represent nucleosomes in promoter regions, transcribed regions, and other regions (tRNA genes and autonomously replicating sequences (ARSs). These classifications are discussed further below.
Figure 3 Nucleosome Modification Patterns
(A) Schematic of annotation scheme for nucleosomes based on their position relative to transcribed units. Intergenic nucleosomes were assigned to the following categories: promoter region (anything upstream of a coding region), nucleosome immediately upstream to the TSS (“distal”), and the nucleosome immediately downstream of the TSS (“proximal”). Transcribed regions were separated into 5′, middle, and 3′ CDSs. Finally, to capture features of chromatin not associated with PolII genes, we independently classified nucleosomes associated with ARS sequences, tRNA genes, and Null (any other intergenic region).
(B) Hierarchical clustering of 2,288 nucleosomes. Left panel: each row corresponds to a single nucleosome, and each column to a particular modification. Red (green) denotes hyper-acetylation (hypo-acetylation) in the first nine columns and relative level of methylation in the last three columns. Rows are sorted according to the dendogram built during clustering. PolII shows the PolII occupancy of the gene associated with the nucleosome in question. Right panel: each row corresponds to a nucleosome (matching the left panel), and each column corresponds to an annotation of the nucleosome according to the scheme of (A). A blue cell denotes a positive annotation of the nucleosome with the appropriate column label. Numbers indicate examples of clusters, as follows: (1) nucleosomes enriched for H3K9Ac, H3K14Ac, and H3K4Me3 that are mostly upstream of transcribed regions; (2) strongly hypo-acetylated nucleosomes, mostly at upstream regions or 3′ of coding regions; (3) nucleosomes acetylated at H4K8 and K16, and H2B K16 that are almost exclusively at the middle and 3′-ends of coding regions; and (4) hyper-acetylated and methylated nucleosomes that are mostly found at the 5′-end of coding regions.
(C) The Pearson correlations of the 12 modification levels between different probes show that there are two tightly correlated groups of acetylations at specific residues. The first group consists of H2A K7; H3K9, K14, and K18; and H4K5 and K12. The second group consists of H2B K16; and H4K8 and K16. Mono- and di-methylation of H3K4 are correlated with the second group, while tri-methylation of H3K4 is correlated with the first group.
(D) The percent of variance captured by using different number of components. The x-axis denotes the number of components, and the y-axis denotes the percent of the variance in the data explained by each components (blue bars) as well as the cumulative percentage explained (red bars).
(E) Representation of all nucleosomes in two-dimensional modification space. In the left panel, each point represents a nucleosome plotted according to the relative level of the first principal component (x-axis) and second principal component (y-axis) for the modification pattern. The right panel is a three-dimensional plot showing density of points along the plane.
Nucleosomes were clustered by modification pattern, using a probabilistic hierarchical agglomerative clustering procedure (see Materials and Methods). As is readily apparent from this clustering (Figure 3B), histone modification patterns span the full possible range of overall modification level, from hypo-acetylated to hyper-acetylated. Nevertheless, a striking aspect of this clustering is the limited range of observed modification patterns. Visual inspection suggests that, as previously noted [18], histone modifications are not independent of each other. Indeed, the matrix of correlations between the 12 modifications shows that there are two groups of strongly correlated acetylations (Figure 3C).
To better understand the effective number of degrees of freedom among the 12 dimensions available, we performed a principal component analysis (see Materials and Methods). Principal component analysis is a technique used to transform a large number of possibly correlated variables to a smaller number of uncorrelated variables, and thereby identify the number of independent dimensions in a dataset. As suggested by the observation above, 81% of the variance in histone modification patterns is captured by the first two principal components (Figure 3D). Moreover, if we examine only the nine acetylations, we can explain 90% of the variance using two components (unpublished data). The first principal component corresponds to overall level of histone modification (Figure S5). The second principal component corresponds to the relative levels of the two groups of histone modifications—the transcription-associated modifications that occur in 5′ to 3′ gradients over coding regions, and the group of acetylations characterized by short hypo-acetyl domains surrounding TSS (Figure S5). By projecting each nucleosome to a point in the plane spanned by the first two principal components (Figure 3E), we can visualize the range of observed modifications. There is a large region of allowable modifications that is spanned continuously by different nucleosomes. These results suggest that, at the level of cell populations, there are no discrete states for nucleosome modifications. Instead, nucleosome modification patterns occur continuously over a large range of possible space, though this two-dimensional space is dramatically simplified compared to the 12 dimensions available. In other words, nucleosomes have continuous variation, both in the total level of acetylation, and in the relative ratio of the two groups of modifications, but they do not show much complexity beyond these two axes.
Specific Chromosomal Locations Are Associated with Characteristic Histone Modifications
Notable in Figure 3B is an association of particular modification patterns with specific genomic locations. For example, Cluster 2 consists of hypo-acetylated nucleosomes that are predominantly located within promoter regions and at the 3′ ends of coding regions. We systematically explored these correlations by testing the modification data for statistically significant, location-specific differences in the levels of each modification type (Figure 4A). For example, promoter nucleosomes are globally hypo-acetylated in residues H2A K7 (presumably due to the enrichment of Htz1), H2B K16, and H4K8 and K16 (and, to a lesser extent, H3K18), and are depleted of mono- and di-methylated H3K4. Nucleosomes at 5′ ends of coding regions are enriched for H3K4Me3, as well as H3K18Ac, H4K12Ac, H3K9Ac, H3K14Ac, H4K5Ac, and H2AK7Ac. When we examine the modification patterns of individual nucleosomes in the two-dimensional principal component plot, we can clearly distinguish nucleosomes in promoter regions from those in transcribed regions (Figure 4B). Moreover, of the nucleosomes in transcribed regions, we can distinguish among nucleosomes in the 5′ end, the middle, and the 3′ end of the transcribed region (Figures 4C and S6).
Figure 4 Nucleosome Modifications Relate to Nucleosome Position
(A) Analysis of differential modification for each class of nucleosomes. Rows correspond to specific modifications, and columns correspond to genomic locations. Each cell is coloured by the average modification level of nucleosomes with this annotation. Non-significant (using false discovery rate of 95% on t-test p-values) cells are blackened.
(B) Promoter nucleosomes (orange) significantly differ from coding region nucleosomes (pink) in their histone modifications pattern. The left panel shows the two types of nucleosomes as points in the plane, where the x-axis represents the level of the first principal component, and the y-axis represents the second principal component. The right panel shows the density within each class.
(C) Distinction between nucleosomes in transcribed regions. Colours denote 5′-end (red), middle (green), and 3′- end (blue) nucleosomes. Visualization is as described in (B).
These results show that specific genomic regions are characterized by distinct modification patterns, with little overlap in modification types between the different regions. We conclude that the histone modification patterns are highly informative about the location of nucleosomes along the chromosome, and suggest that, in yeast, nucleosome modification patterns, like nucleosome positioning, exhibit local variation around a basic stereotype that is determined by the chromosomal location.
Variation in Modifications Occurring over Transcribed Regions is Predictive of Transcription Levels
While nucleosomes at different locations are associated with statistically different modification patterns, the correlations are imperfect, as a given nucleosome modification pattern can clearly be found in multiple locations (Figure 4B and 4C). This imperfect association might be due to differences in expression level of the coding regions examined. We therefore separated nucleosome locations (5′ coding, etc.) into bins according to the PolII activity level of the associated transcription unit. Figure 5A shows the modification pattern of each of five nucleosomes (defined by position) for highly PolII-enriched genes, while Figure 5B shows this pattern for PolII-depleted genes. This view emphasizes both the distinction between nucleosomes at various genomic locations (as seen in aggregate in Figure 4) and the transcription-associated variation in the modification pattern at a given location. Figure 5C shows a cartoon of the chromatin structure of an arbitrary yeast gene.
Figure 5 Nucleosome Modifications Partitioned by Location and by Transcription Level
(A) Modification patterns of nucleosomes associated with actively transcribed genes. Genes with high levels of PolII occupancy were grouped, and the modification data for the indicated nucleosome types were averaged.
(B) Modification patterns of nucleosomes associated with poorly transcribed genes, grouped as in (A), except that genes with low levels of PolII were selected.
(C) Schematic view of yeast chromatin architecture. Cartoon view showing chromatin structure of an arbitrary yeast gene. Yeast genes are typically characterized by an upstream nucleosome-free region, which serves as the transcriptional start site [29]. Surrounding this nucleosome-free region are two nucleosomes that exhibit low levels of acetylation at H2BK16, H4K8, and H4K16, and that carry Htz1 in place of the canonical H2A (unpublished data). The remaining acetylations occur in a gradient from 5′ to 3′ over actively transcribed genes. Similarly, actively transcribed genes exhibit a gradient of H3K4 methylation, with trimethylation occurring at the 5′- ends of genes, and di- and mono-methylation occurring over the middle of the coding region. Nucleosomes are coloured to emphasize the different average modification patterns at each indicated location.
To further explore the relationship between transcription activity and modification pattern at a given location, we tested each location for modifications that were significantly associated with high or low transcription. For example, we consider the nucleosomes near the 5′ ends of those genes with extreme levels of PolII enrichment or depletion (Figure 6A). Consistent with results shown in Figures 2C and 5A and 5B, we see that levels of mono- and tri-methylation of H3K4, as well as the acetylation level of H3K9, H3K14, H2A K7, H4K5, and H4K12 have significant differences between these two classes of 5′ coding region nucleosomes (p < 0.01 using t-test). We trained a classification method that examines these modifications and predicts whether the nucleosome is part of an expressed coding region or not. We evaluated this classifier using leave-one-out cross-validation (see Materials and Methods) to estimate its accuracy on unseen examples. This evaluation shows that the classifier is correct on 75.4% of the nucleosomes in the training set (compared to 60.1% when nucleosomes labels are randomly permuted; p < 0.0001). Thus, although expression values are not perfectly encoded by histone modifications, they are clearly reflected in them. We see a similar pattern if we examine nucleosomes in the middle of coding regions (Figure S7). In this case the accuracy is 82.7% (compared to 61.3% by chance; p < 0.0001). Notably, the set of significant modifications in this case is different, and in fact two of the transcription-independent modifications, H4K8 and K16, are both slightly anticorrelated with transcription here.
Figure 6 Nucleosome Modifications Relate to Transcription Level
(A) Classification plot of nucleosomes in 5′-coding regions according to PolII occupancy. A classifier was trained to distinguish between nucleosomes with high and low PolII occupancy, and evaluated using leave-one-out cross-validation. Each row corresponds to one nucleosome. Nucleosomes are split into three groups associated with genes corresponding to high, intermediate, and low PolII occupancy level (from top to bottom, respectively). The left 12 columns denote modification patterns of each nucleosome. Modifications with significant differences between high and low nucleosomes are marked with the p-value determined by t-test. Colours denote relative acetylation/methylation levels. The rightmost three columns correspond to the classifier's prediction of transcription, the expression level (mRNA abundance; see Materials and Methods) and the PolII occupancy of genes. The average accuracy of random classification was 60.71%, with a standard deviation of 4.3%. Accuracy of classifier was 75.38% (p < 0.0001).
(B) Classification plot of TSS proximal nucleosomes, labelled as in (A). The average accuracy of random classification was 62.45%, with a standard deviation of 4.75%. Accuracy of classifier was 72.8% (p = 0.0004).
(C) Classification plot of TSS distal nucleosomes; as in (A). The average accuracy of random classification was 65.79%, with a standard deviation of 4.22%. Accuracy of classifier was 58.4% (p = 0.9333).
These results indicate that over coding regions, variation in histone modification patterns is associated with transcription level. For example, the transcription-associated modifications are globally enriched at the 5′ ends of genes, and the level of these modifications is correlated with transcription level. To explore whether these results hold true for nucleosomes that are not found over transcribed regions, and to thereby test the idea that upstream histone modifications control gene expression, we repeated the classification analysis for nucleosomes surrounding the TSS (Figure 6B and 6C), which are modified in similar ways (Figure 4A) with the exception that the gene-proximal nucleosome is associated with DNA passaged by RNA polymerase, while the gene-distal nucleosome is not. Here, we found that the gene-proximal nucleosome indeed carries information about transcription level—a classification method tested using this nucleosome correctly identified 72.8% of gene expression patterns (as compared with 62.4% by chance; p = 0.0004). In contrast, the gene-distal nucleosome, which is not subjected to the passage of RNA polymerase and associated modifying enzymes, fails to accurately classify transcription levels (58.4%, as compared with 65.7% expected by chance), demonstrating that modification patterns associated with transcribed regions provide a much better predictor of transcription levels than do upstream modification patterns.
Modifications Associated with Transcriptional Regulators
The observed modifications at the two TSS nucleosomes might be either a prerequisite for PolII recruitment or a consequence of this step. Since we measure modification in a single condition, we cannot directly resolve this question. However, we can gain additional insight by examining nucleosomes in promoters reported to be bound by specific chromatin remodelers or by specific transcription factors. Using the results of several recent ChIP studies [39–41], we compiled a set of target promoters for each factor (see Materials and Methods). We then tested for distinct patterns in the promoter nucleosomes. In addition, we analyzed nucleosomes around putative transcription factor binding sites [42] (see Materials and Methods). Our results highlight specific factors that are significantly associated with specific modifications (Figure 7). For instance, we see that promoters of genes bound by the repressor Ume6 are significantly hypo-acetylated at most positions. This finding correlates with previous observations demonstrating recruitment of the HDAC Rpd3 by Ume6 [43,44]. Another interesting example is the significant hyper-acetylation of several positions among the targets of the Rsc remodeling complex. These include H3K9 and, to a lesser extent, H4K12, H3K14, and H4K5. Recently, mutants in the Rsc complex were shown to interact genetically with K14 mutations, a finding supported by binding of the complex to K14-acetylated H3-tail peptides [45].
Figure 7 Histone Modifiers
Analysis of differential modification of nucleosomes associated with various transcriptional regulators. Promoter nucleosomes located near binding sites of the indicated factors were tested for enrichment of all modifications relative to the overall promoter modification pattern. Each cell is coloured by the average modification level of nucleosomes with this annotation. Non-significant cells (using false discovery rate of 95% on t-test p-values) are blackened. Localization data are taken from the indicated studies [39–42].
Modification Boundaries Occur Near Transcriptional Start Sites
The availability of histone modification data at single nucleosome resolution allows analysis of the extent to which modification patterns occur discretely or in broad domains. As noted above and previously reported [44], histones can be deacetylated in a localized manner. However, visual inspection reveals that at locations farther away from the TSS, most histone modifications occur in broad domains. To further investigate this, we searched for sharp boundaries to histone modification domains by identifying pairs of nucleosomes between which a dramatic change occurs (increase or decrease of two standard deviations at one of the tail positions). We found ~100 boundaries for each modification (from 82 to 108). We then examined the locations of these boundaries, finding that most were located adjacent to TSSs. For example, boundaries for modifications associated with transcription, such as H3K4 tri-methyl, occurred across the TSS. This is visualized in Figure 8A, a scatterplot of K4 tri-methylation for adjacent nucleosomes (x-axis shows tri-methylation for nucleosome N, y-axis shows tri-methylation of N-1). The majority of nucleosomes show high correlation for this modification between adjacent nucleosomes, though there are two small groups of anticorrelated nucleosomes, indicating methylation boundaries. Pairs of nucleosomes that fall to either side of the TSS were plotted separately (grouped according to which strand the gene falls on), showing that most of the K4 tri-methyl boundaries occur at the TSSs, as expected.
Figure 8 Modification Boundaries
(A) H3K4Me3 boundaries occur across TSSs. The x-axis represents the level of H3K4Me3 for a given nucleosome, and the y-axis represents the level of this modification for the preceding nucleosome. Pairs of nucleosomes flanking the TSS for a gene on the W strand are plotted as blue squares, and pairs flanking TSSs for genes on the C strand are plotted as red squares. Remaining nucleosome pairs are plotted as grey circles.
(B) Example of a punctate nucleosome. Histone modification plotted as in Figure 1B for a subset of histone modifications. Arrow indicates a nucleosome whose modification pattern differs significantly for H3K4Me3 from nucleosomes to either side. Gene names are as labelled.
(C) Example of a punctate nucleosome, labelled as in (B).
We also examined “punctate” nucleosomes—those differing significantly in modification type from the two nucleosomes to either side. We found 44 nucleosomes with a punctate pattern of at least one of the 12 modifications in this study. Examples of punctate nucleosome are shown in Figure 8B and 8C. Most nucleosomes that exhibit this characteristic are found upstream of the TSS. In many cases, this is clearly due to the location of the nucleosome between two TSSs, leading to a single nucleosome exhibiting no transcription-associated modifications, surrounded by nucleosomes with the characteristic transcriptional modifications.
Discussion
Profiling Histone Modification at the Mononucleosome Level
We have mapped, at single-nucleosome resolution, 12 histone modifications in actively dividing cultures of S. cerevisiae. This, along with the translational positioning of nucleosomes described previously [29] and location studies on the H2A isoform Htz1 (unpublished data), provides a draft sequence (see below) of the primary structure of half a megabase of yeast chromatin. We wish to stress the importance of the high resolution of our method for deconvoluting the results of previous studies on histone modification. The use of ~1-kb intergenic and coding probes in standard microarray studies reports on mixtures of multiple nucleosomes. For example, we show that the two nucleosomes immediately adjacent to the TSS are generally deacetylated at H4K16, whereas surrounding nucleosomes are often highly acetylated (Figure 2B). As a result, the acetylation level measured in standard microarray studies will depend on the length of the 5′ untranslated region (which is especially confounding, as this correlates with functional classifications of the encoded genes [46]); the length of the entire intergenic region probed; and the nature of the intergenic region (divergent or parallel genes), as the deacetyl signals from the TSS will be diluted by these additional nucleosomes in a complicated way. Furthermore, the ~300–500-bp standard shear size used in microarray studies results in some sampling of additional nearby nucleosomes outside the borders of the microarray spot. Our methodology eliminates all these confounding variables and also controls for local variation in nucleosome density, thus dramatically simplifying modification mapping.
We note, however, that our study is subject to the same issues with antibody specificity that remain a crucial limitation of ChIP studies—the epitope accuracy of any ChIP study is determined by the specificity of the antibodies used. We used the state-of-the-art in antibodies (see Materials and Methods), but improvements in antibody specificity may improve the fidelity of these experiments. In addition, ensemble measurements such as those presented here necessarily provide population averages, and we cannot rule out the possibility that small subpopulations of cells in different phases of the cell cycle, or in different epigenetic states, might be characterized by modification patterns that are obscured in the population average. Finally, this study does not provide a complete sequence of chromatin's primary structure in our tiled region. A complete view of the primary structure requires the addition of all additional modifications, including core domain modifications, and, ideally, the conformations of the nucleosomes studied.
Histone Tail Modifications Occur in Two Groups that Vary Quantitatively
This mapping has allowed us to investigate combinatorial questions raised by the framing of histone modifications as a “code.” Most importantly, we have shown that many histone modifications are highly correlated with one another, resulting in few discrete histone modification patterns. However, we cannot say whether these modifications occur in the same nucleosome or whether the correlations are due to a mixture of partially modified nucleosomes at a given location. Some modified residues may be correlated because histone-modifying enzymes are not strongly residue-specific [8,47], whereas other correlations may be due to histone-modifying enzymes that are either recruited to chromatin by association with other types of modification, or preferentially act on tails carrying another modification [48–50]. Still other modifications may be correlated because the relevant modifying enzymes may be targeted by association with similar complexes, such as RNA polymerase [23,51]. These correlations suggest a high level of redundancy in yeast histone modification, implying that the code is extremely simple, carrying only a tiny fraction of the maximum possible amount of information. Indeed, as principal component analysis shows, we can compress the 12-dimensional space of possible modification patterns onto two main axes, with only a minor loss of accuracy.
This raises the important question of why so many different modifications occur in the cell, yet such a small subset of combinations is used. We suggest only a few possible answers. First, the loss of a positive charge that occurs with lysine acetylation should reduce the free energy of interaction with a negative charge by approximately 1–3 kcal/mol. Thus, loss of multiple positive charges could lead to much greater free energy changes in an interaction, and to a much more pronounced change in interactions than would be caused by a single acetylation. Furthermore, we note that at any given nucleosome location the quantitative level of acetylation varies, allowing for the possibility of “rheostat”-like control of transcription levels. This is consistent with recent mutagenesis studies showing that transcriptional response to H4K→R mutations is largely continuous and analogue, rather than discrete and digital [24]. Second, it is possible that multiple modifications occur together in order to cause several distinct required events to occur, whether they be co-occurring structural changes in the nucleosome or the 30-nm fibre, or recruitment of protein complexes that function together. This has been observed at the human interferon-β promoter, wherein activation of the promoter causes Gcn5-dependent acetylation of H3K9/14 and H4K8, whose acetylation recruits TFIID and hSWI/SNF, respectively [52]. If these protein complexes tend to function together, then the recruiting modifications will be correlated. Third, if modifications that occur together at steady-state do not occur simultaneously, but rather in a temporal cascade [6], this enables the possibility of complex signal filtering behaviour. For example, if one histone acetylase were to acetylate a single lysine, and that acetyl-lysine were to recruit a distinct histone acetylase that acetylated another lysine, then a requirement for both acetylations for transcription to occur would produce a low-pass filter. This filter would reject transient spikes in signalling pathways and allow transcriptional outcomes only in response to sustained signalling. A careful examination of the temporal response of histone modifications to signalling will help determine if this might occur for the correlated modifications. Finally, if one modification recruits enzymes that modify the remaining residues, then having multiple modifications allows for switch-like behaviour [53,54].
Stereotyped Promoter Architecture
One of the two groups of histone modifications exhibits a striking, stereotyped pattern in promoter regions. Nucleosomes immediately adjacent to the TSS are hypo-acetylated at H2BK16, H4K8, and H4K16. This hypo-acetylation does not correlate with transcription levels, and the inability of the histone modification pattern at the gene-distal TSS-adjacent nucleosome to accurately reflect transcriptional activity of the associated gene (Figure 6C) does not support the idea that upstream modifications are causal for transcription.
In separate work, we have identified this di-nucleosomal domain that flanks the TSS as highly enriched for the H2A isoform Htz1 (demonstrating that these nucleosomes do not appear deacetylated due to some artifactual difficulty with immunoprecipitation). Also, this enrichment is independent of transcription (unpublished data). In other words, the majority of promoter nucleosome-free regions in yeast are surrounded on either side by nucleosomes with hypo-acetylated H2BK16, hypo-acetylated H4K8 and K16, and Htz1 in place of H2A. These results raise two questions: how does this domain arise, and what is its functional role in transcription?
Previous reports have shown that Rpd3 deacetylates one to three nucleosomes when recruited to promoters [44], consistent with the width of this deacetylation domain. However, the generality of the pattern observed here suggests that multiple distinct deacetylases function in this localized manner, because Rpd3 is present at only a subset of the promoters analyzed [31,43]. Alternatively, it is possible that these nucleosomes turn over rapidly (due to the presence of some assembly of chromatin-remodelling activities at promoters), and that the histone isoform and modification pattern exhibited reflects the composition of free histones in the nucleoplasm. In either case, the function of this domain remains elusive at present.
Relationship of Histone Modifications to Transcription
We have described a group of histone modifications that co-occur, and that are preferentially found at the 5′ ends of actively transcribed genes. This relationship between histone modification patterns, location relative to coding regions, and transcript abundance, would be expected if histone modification played a largely passive, rather than instructive, role in transcription, with nucleosomes being modified by various enzymes associated with RNA polymerase. This is clearly the case, for example, for PolII-associated Set1, which is responsible for the correlation between H3K4 tri-methylation over the 5′ end of coding regions and corresponding transcription levels. A similar type of mechanism appears to hold for the Set2-mediated tri-methylation of H3K36, which occurs over transcribed genes [55]. However, mutant studies have shown abundant transcriptional defects associated with mutations in histone-modifying enzymes [56,57]. These studies cannot determine whether histone modification is instructive or permissive for transcription—in other words, whether histone modifications initiate a chain of events that result in transcription, or whether that gene is associated with a non-permissive chromatin structure that must be antagonized using the modification in question. We suggest that the transcription-associated modifications play a permissive role in gene expression, and that the transcriptional defects in histone-modification mutants result from a partial inability of RNA polymerase to transit unmodified nucleosomes [58,59], or to a failure to recruit factors required for efficient transcription [60]. However, we do not rule out the possibility that histone modifications play both roles, with an initial mark that is causal for a transcription pattern subsequently “erased” by modifications occurring with the resultant transcription.
The Histone Code
Taken together, these results do not support a model for the histone code in which a vast set of widely varying modification combinations play complicated instructive roles in transcriptional regulation. Instead, these results further extend genome-wide studies in Drosophila, which show that histone modifications occur in few independent combinations [25], and suggest that these patterns are often the result, rather than the cause, of transcription. These results therefore emphasize a role for modifications of the histone tails as facilitators of transcription. It will be of great interest in future studies to assay the dynamic nature of histone modifications during changes in transcription, and the establishment of histone modification patterns during DNA replication.
Materials and Methods
Yeast culture
An aliquot of 450 ml of BY4741 bar1Δ cells was grown to an A600 OD of 0.9 in 2-L flasks shaking at 200 rpm in a 28 °C water bath. Formaldehyde (37%) was added to a 1% final concentration, and the cells were incubated for 15 min at 25 °C, shaking, at 90 rpm. Then, 2.5 M glycine was added to a final concentration of 125 mM, to quench the formaldehyde. The cells were inverted and let to stand at 25 °C for 5 min. The cells were spun down at 3,000 × g for 5 min at 4 °C and washed twice, each time with an equal volume of ice-cold sterile water.
Micrococcal nuclease digestion
The cell pellets were resuspended in 39 ml Buffer Z (1 M sorbitol, 50 mM Tris-Cl [pH 7.4]), 28 μl of β-ME (14.3 M, final concentration 10 mM) was added, and cells were vortexed to resuspend. Then, 1 ml of zymolyase solution (10 mg/ml in Buffer Z; Seikagaku America, Falmouth, Massachusetts, United States) was added, and the cells were incubated at 28 °C, shaking at 200 rpm, in 50-ml conical tubes, to digest cell walls. Spheroplasts were then spun at 3,000 × g, 10 min, at 4 °C. Spheroplast pellets were resuspended and split into aliquots of 600 μl of NP-S buffer (0.5 mM spermidine, 1 mM β-ME, 0.075% NP-40, 50 mM NaCl, 10 mM Tris [pH 7.4], 5 mM MgCl2, 1 mM CaCl2) per 90-ml cell culture equivalent. Forty units of micrococcal nuclease (Worthington Biochemical, Lakewood, New Jersey, United States) were added, and the spheroplasts were incubated at 37 °C for 20 min—this was determined in initial titrations to yield > 80% mononucleosomal DNA (see Figure S1), but to repeat these results an independent titration should be carried out as a preliminary study. The digestion was halted by shifting the reactions to 4 °C and adding 0.5 M EDTA to a final concentration of 10 mM.
ChIP
All steps were done at 4 °C unless otherwise indicated. For each aliquot, Buffer L (50 mM Hepes-KOH [pH 7.5], 140 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate) components were added from concentrated stocks (10–20×) for a total volume of 800 μl per aliquot. Each aliquot was incubated with 80–100 μl of 50% Sepharose Protein A Fast-Flow bead slurry (Sigma, St. Louis, Missouri, United States) equilibrated in Buffer L for 1 h on a tube rotisserie rotator. The beads were pelleted with a 1-min spin at 3,000 × g, and approximately 2.5%–5% of the supernatant was set aside as ChIP input material. With the remainder, antibodies were added to each aliquot (20% of a 450-ml cell culture) in the following volumes: 25 μl anti-H3K4Me1 Ab (affinity purified; Abcam, Cambridge, Massachusetts, United States), 6 μl anti-H3K4Me2 Ab (affinity purified; Abcam), 6 μl anti-H3K4Me3 Ab (affinity purified; Abcam), 4 μl anti-H4K16Ac Ab (whole antiserum; Abcam), 9 μl anti-H4K5Ac Ab (whole antiserum; Abcam), 3 μl anti-H3K14Ac Ab (whole antiserum; Upstate Cell Signaling Solutions, Charlottesville, Virginia, United States), 3 μl anti-H2AK7Ac Ab (whole antiserum; Upstate), 2 μl, anti-H4K8Ac Ab (whole antiserum; Abcam), 15 μl, anti-H4K12Ac Ab (whole antiserum; Abcam), 25 μl anti-Ac Ab (whole antiserum; Abcam), 16 μl anti-H3K9Ac Ab (affinity purified; Abcam), 25 μl anti-H2BK16Ac (L) (whole antiserum; Abcam), and 3 μl anti-H3K18Ac Ab (whole antiserum; gift of M. Grunstein). We also used 3 μl of a distinct antibody to H4K16Ac (whole antiserum; gift of M. Grunstein) to assess specificity of different sources of antibody. Replicates using this antibody were as correlated with each other as they were with replicates using the Abcam antibody.
These were incubated, rotating, overnight (∼16 h), after which the sample was transferred to a tube containing 80–100 μl of 50% Protein A bead slurry. The sample was incubated with the beads for 1 h for the immunoprecipitation, after which the beads were pelleted by a 1-min spin at 3,000 × g. After removal of the supernatant, the beads were washed with a series of buffers in the following manner: 1 ml of the buffer would be added, and the sample rotated on the tube rotisserie for 5 min, after which the beads would be pelleted in a 30-s spin at 3,000 × g and the supernatant removed. The washes were performed twice for each buffer in the following order: Buffer L, Buffer W1 (Buffer L with 500 mM NaCl), Buffer W2 (10 mM Tris-HCl [pH 8.0], 250 mM LiCl, 0.5% NP-40, 0.5% sodium deoxycholate, 1mM EDTA), and 1× TE (10 mM Tris, 1 mM EDTA [pH 8.0]). After the last wash, 125 μl of elution buffer (TE [pH 8.0] with 1% SDS, 150 mM NaCl, and 5 mM dithiothreitol) was added to each sample, and the beads were incubated at 65 °C for 10 min, with frequent mixing. The beads were spun for 2 min at 10,000 × g, and the supernatant was removed and retained. The elution process was repeated once for a total volume of 250 μl of eluate. For the ChIP input material set aside, elution buffer was added for a total volume of 250 μl. After overlaying the samples with mineral oil, the samples were incubated overnight at 65 °C to reverse cross-links.
Antibody specificity
A significant concern with ChIP studies is the epitope specificity of the antibodies used. High correlations between different modifications could arise if two antibodies cross-reacted. We note four reasons that this is unlikely to be a major problem for this study. First, if antibodies did indeed cross-react, then the resulting profiles should look like some weighted average (depending on relative affinities of the two antibodies) of the two “pure” profiles. If there were a third modification pattern (besides what we term the transcription-dependent and transcription-independent patterns), then the two antibodies in question would be expected to show a third mixed pattern, distinct from the two patterns described, and this was not observed. On the other hand, if only two true patterns do exist but there is cross-reactivity for antibodies, the mixed profile is expected to show a 5′ gradient of acetylation, along with two deacetyl nucleosomes adjacent to the TSS. This pattern was seen for H2AK7, but, as we note, this is likely due to the replacement of H2A with Htz1 at the TSS-adjacent nucleosomes. Furthermore, this pattern was not seen for the H3K14 antibody, which recognizes lysine in the context of a similar site to that of H2AK7 (GGKA). So we do not believe that these antibodies are cross-reacting.
Second, we repeated experiments for one of the epitopes in this study (H4K16) with two distinct antibodies, and the results were indistinguishable. One of these antibodies, from the Grunstein lab, was previously tested for cross-reactivity by attempting ChIP from strains carrying the H4K16R mutation [37].
Third, there are two pairs of antibodies for which cross-reaction is most likely to be a concern: H4K5 and K12 (both lysines occur in the context of GKGG), and H2AK7 and H3K14 (both occur in the context of GGKA). However, within each pair, the two antibodies are more highly correlated with other antibodies in their group than with the other antibody with a similar recognition site (see Figure 3C). If these antibodies had cross-reacted, then their profiles should be the most highly correlated. In addition, technical literature from Upstate shows that both the H2AK7 and H3K14 acetylation antibodies fail to immunoprecipitate DNA from yeast strains carrying the appropriately mutated recognition site.
Finally, it is worth noting that even if a pair or two of antibodies cross-reacted, the point that histone modifications occur at reduced dimensionality would still hold. Instead of 12 dimensions reducing to two dimensions, we would say, for example, that 10 dimensions reduce to two. This is not, to our thinking, a significant change in the central message of this study. In addition, it would not challenge the other main points of the manuscript, that the two TSS-adjacent nucleosomes exhibit a stereotyped modification pattern and that most of the histone modification that correlates with transcription levels occurs over coding regions.
Protein degradation and DNA purification
After cooling the samples down to room temperature, each sample was incubated with an equal volume of proteinase K solution (1× TE with 0.4 mg/ml glycogen, and 1 mg/ml proteinase K) at 37 °C for 2 h. Each sample was then extracted twice with an equal volume of phenol and once with an equal volume of 25:1 chloroform:isoamyl alcohol. Phase-lock gel tubes were used to separate the phases (light gel for phenol, heavy gel for chloroform:isoamyl alcohol). Afterwards, 0.1 volume 3.0 M sodium acetate [pH 5.3] and 2.5 volumes of 100% ice-cold ethanol were added, and the DNA was allowed to precipitate overnight at −20 °C. The DNA was pelleted by centrifugation at 14,000 × g for 15 min at 4 °C, washed once with cold 70% ethanol, and spun at 14,000 × g for 5 min at 4 °C. After removing the supernatant, the pellets were allowed to dry and then were resuspended in 20 μl 10 mM Tris-Cl, 1 mM EDTA [pH 8.0], and 0.5 μg of RNase A was added. The samples were incubated at 37 °C for 1 h, and then treated with 7.5 units of calf intestinal alkaline phosphatase in a 30-μl volume supplemented with NEB Buffer 3 (10× concentration of 100 mM NaCl, 50 mM Tris-HCl [pH 7.9], 10 mM MgCl2, 1 mM dithiothreitol). The samples were then incubated for a further 1 h at 37 °C and then cleaned up with the Qiagen MinElute Reaction Cleanup Kit (Qiagen, Valencia, California, United States), following manufacturer's directions, except with an elution volume of 20 μl.
Linear amplification of DNA
The samples were amplified, with a starting amount of 125 ng for ChIP input materials and up to 75 ng for ChIP samples, using the DNA linear amplification method described in BMC Genomics 4:19 [32].
Microarray hybridization
RNA produced from the linear amplification (3 μg) was used to label probe via the amino-allyl method as described at http://www.microarrays.org. Labelled probes were hybridized onto a yeast tiled oligonucleotide microarray [29] at 65 °C for 16 h, and washed as described at http://www.microarrays.org. The arrays were scanned at 5-μm resolution with an Axon Laboratories (Sunnyvale, California, United States) GenePix 4000B scanner running GenePix 5.1.
Image analysis and data processing
Array features were filtered using the autoflagging feature of GenePix 5.1 with the following criteria defining features to be discarded: [Flags] = [Bad] Or[Flags] = [Absent] Or[Flags] = [Not Found] OrLCase([ID]) = “empty” OrLCase([ID]) = “blank” Or([SNR 635] < 3 And [SNR 532] < 3) Or[F Pixels] < 100 Or([F Pixels] < 150 And [Circularity] < 75).
The remaining features for each array were then block-normalized by calculating the average net signal intensity for each channel in a given block, and then taking the product of this average and the net signal intensity for each filtered array feature in the block. Afterwards, all block-normalized array features were normalized using a global average net signal intensity as the normalization factor.
Each histone tail modification epitope was chromatin-immunoprecipitated in three to six biological replicates, with additional technical replicates of the microarray hybridizations. Outlying replicates were removed (with a minimum remainder of three replicates), and the median was calculated and used for subsequent data analysis.
Normalization of modification and PolII data
Each assay was repeated three to six times, and median values per probe were calculated. Measurements for each antibody were first log (base 2) transformed and then normalized (to mean of zero and variance of one).
Data availability
Data can be viewed at http://compbio.cs.huji.ac.il/Nucs. Data are downloadable at http://www.cgr.harvard.edu/chromatin, and have been deposited in GEO.
Clustering of aligned genes
The genes were clustered using PCluster, a probabilistic hierarchical clustering algorithm [61]. Probes at locations relative to gene reference point, either beginning of coding sequence (CDS) (Figure 2A) or TSS (Figure 2B), are used as attributes of the gene. Linker probes (based on the nucleosome locations of [29]) were discarded and treated as missing values.
Splitting genes into transcriptional groups
Each gene was assigned a transcription activity value based on the average enrichment of PolII along CDS probes. Genes with less than five CDS probes were removed to reduce noise. We then used thresholds of 0.75 and −0.75 to classify genes as highly, mid-, and untranscribed. This resulted in 75 highly transcribed genes, 192 intermediate genes, and 57 poorly transcribed genes. We also repeated the analysis presented in Figure 2C using mRNA abundance rather than PolII occupancy to bin genes (Figure S4), and the results were qualitatively indistinguishable.
Averaging probes into nucleosomal-based data
A total of 24,947 probes were assigned to 2,288 nucleosomes using a four-probe minimum size cutoff [29]. We used the hand-called set of nucleosome positions (these were generated by inspection and adjustment of the automated hidden Markov model calls; these positions are provided in the dataset associated with [29]), as that set covered a slightly greater fraction of the genome. Results are qualitatively unchanged when only HMM calls are used (unpublished data). For each antibody, the nucleosomal values were set by the median levels of relevant probes.
Genomic classification of nucleosomes
Nucleosomes were annotated based on their relative position to nearby genes. Nucleosomes in the first (or last) 500 bp of annotated genes were annotated as 5′ CDS (or 3′ CDS) nucleosome. Other CDS nucleosomes were annotated as mid-CDS. The two TSS adjacent nucleosomes were annotated as TSS distal (5′) and proximal (3′) nucleosomes. Nucleosomes upstream (up to 1 kb or closer to non-dubious CDSs) were annotated as promoter nucleosomes. Nucleosomes around tRNA genes (200 bp from each side) or ARS elements (200 bp from each side) were annotated as tRNA or ARS nucleosomes. Other nucleosomes were annotated as null. In certain cases, we allowed more than one annotation per nucleosome; for instance, a nucleosome between two divergent genes can be annotated as TSS-proximal for one gene, and a promoter nucleosome for another one.
Single nucleosome clustering
Nucleosomes were clustered using PCluster [61], treating each nucleosome as a vector of 12 values.
Principal component analysis
Principal component analysis was applied to the nucleosomal modification data of 2,288 nucleosomes versus 12 modifications using MATLAB 6.5 (rel 13) procedure “princomp.” Density visualization was done using Parzen windows density estimator with Gaussian kernels (with standard deviation of 0.3) .
Genomic enrichment of modifications
We compared the modifications of nucleosomes affiliated with each genomic location (promoter, TSS distal, etc.) to all other nucleosomes, using a standard two-tail t-test. To correct for multiple hypotheses, we used a 5% false discovery rate procedure [62]. The average change was then calculated for < modification, genomic location > pairs with significant p-values.
Transcription-specific modifications
To identify specific modifications at genomic locations with significant correlations to expression levels of nearby genes, we trained a classification method to predict whether a nucleosome was associated with genes enriched or depleted for PolII. To prevent biased results, we applied a leave-one-out cross-validation procedure in which the tested nucleosome was removed from the training set, and a classifier was trained on the rest of the nucleosomes and used to predict the held-out nucleosome label. We used a Naive Bayes classifier [63] using the implementation described [64]. We then classified the held-out nucleosome, based on the probability of its modification pattern under each of the classes. We computed the overall accuracy of classification and a p-value by repeating the same leave-one-out procedure with randomly reshuffled nucleosome labels.
Functional classification of nucleosomes
We used recent genomic studies [39–41] and compiled a set of target promoters for each factor. We then tested the promoter and TSS-distal and TSS-proximal nucleosomes of these genes for enrichment of specific modifications. In addition, we created a subset of the target nucleosomes of Harbison et al., by restricting the nucleosomes to those up to 100 bp away from putative binding sites bound in rich growth conditions [42]. As described earlier, we compared the “bound” nucleosomes to all other promoter/TSS nucleosomes, and used a false discovery rate-corrected two-tail t-test.
Supporting Information
Dataset S1 Complete Dataset
Individual worksheets contain data for all individual replicates before range normalization, for combined median data organized by epitope, and for combined median data after range normalization.
(48 MB XLS).
Click here for additional data file.
Dataset S2 Replicate Reproducibility
Data contain correlations between individual experiments for each antibody.
(24 KB XLS).
Click here for additional data file.
Figure S1 Digestion of Chromatin to Mononucleosomes before Immunoprecipitation
Gels show micrococcal nuclease-digested DNA from multiple independent cultures used for the immunoprecipitations reported here. Molecular markers are as indicated. Blue dots indicate nucleosomal DNA used for immunoprecipitations, while green dots show sonicated DNA from the same culture. Digested DNA used for immunoprecipitation was typically > 80% mononucleosome.
(674 KB PDF).
Click here for additional data file.
Figure S2 Low Levels of Histone Modification over Heterochromatin
Data are plotted as in Figure 1B. Chromosome III coordinates are shown above the modification data. Three panels show data for a portion of (from left to right) TelIIIL, HML, and TelIIIR. Only partial regions of the three are shown, as the remainder was not tiled due to cross-hybridization concerns [29].
(551 KB PDF).
Click here for additional data file.
Figure S3 Broad Patterns of Histone Modifications
Data are aligned by the TSS, and plotted as in Figure 2B for all remaining modifications, as indicated.
(1.8 MB PDF).
Click here for additional data file.
Figure S4 Relationship of Histone Modifications to mRNA Abundance
Genes were grouped into low, medium, and high mRNA abundance classes using data from competitive hybridizations of mRNA versus genomic DNA on cDNA microarrays (CLL and SLS, unpublished data). Low-abundance mRNAs were defined as those with log(2) ratios less than −1, while high-abundance mRNAs were defined as those exhibiting log(2) ratios greater than 1. Histone modification data are averaged and displayed as in Figure 2C, and results are qualitatively indistinguishable from those generated using PolII occupancy to classify genes.
(676 KB PDF).
Click here for additional data file.
Figure S5 Representation of the First Two Principal Components
The first component (left panel) consists of all positive coefficients (plotted on the y-axis), and therefore captures the global magnitude of modification (both acetylation and methylation). The second component differentiates between the two groups of correlated modifications (see Figure 3C). Bars indicate different epitopes as indicated.
(512 KB PDF).
Click here for additional data file.
Figure S6 Principal Component Analysis of Nucleosome Modifications
Data plotted as in Figure 4B and 4C, right panels.
(580 KB PDF).
Click here for additional data file.
Figure S7 Nucleosome Modifications Relate to Transcription Level
Classification plot as described in Figure 5, using mid-CDS nucleosomes. The average accuracy of random classification was 61.27%, with a standard deviation of 5.76%. Accuracy of classifier was 82.65% (p < 0.0001).
(397 KB PDF).
Click here for additional data file.
Accession Numbers
The Gene Expression Omnibus (GEO) (http://www.ncbi.nlm.nih.gov/geo) accession numbers for the experiments described here are GSM64526–GSM64587, GSM64591, and GSM64592, and are part of series accession number GSE2954.
We would like to thank N. Francis, L. Garwin, H. Madhani, T. Maniatis, H. Margalit, A. Murray, A. Regev, B. Suter, and K. Thorn for critical reading of the manuscript and/or helpful discussions. We thank S. Kurdistani and M. Grunstein for their generous gifts of antibodies to H3K18Ac and H4K16Ac. TK is supported by the Yeshaya Horowitz Foundation through the Center for Complexity Science. NF is supported by the Harry and Abe Sherman Senior Lectureship in Computer Science. This research was supported by grants to SB, SLS, NF, and OJR from the National Institute of General Medical Sciences, and to NF from the Israeli Science Foundation.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. CLL, SB, SLS, and OJR conceived and designed the experiments. CLL and MK performed the experiments. CLL, TK, NF, and OJR analyzed the data. NF and OJR wrote the paper.
Citation: Liu CL, Kaplan T, Kim M, Buratowski S, Schreiber SL, et al. (2005) Single-nucleosome mapping of histone modifications in S. cerevisiae. PLoS Biol 3(10): e328.
Abbreviations
ARSautonomously replicating sequence
bpbase pairs
CDScoding sequences
ChIPchromatin immunoprecipitation
TSStranscription start site
==== Refs
References
Venter U Svaren J Schmitz J Schmid A Horz W A nucleosome precludes binding of the transcription factor Pho4 in vivo to a critical target site in the PHO5 promoter Embo J 1994 13 4848 4855 7957054
Stunkel W Kober I Seifart KH A nucleosome positioned in the distal promoter region activates transcription of the human U6 gene Mol Cell Biol 1997 17 4397 4405 9234698
Langst G Becker PB Nucleosome remodeling: One mechanism, many phenomena? Biochim Biophys Acta 2004 1677 58 63 15020046
Turner BM Histone acetylation and an epigenetic code Bioessays 2000 22 836 845 10944586
Turner BM Cellular memory and the histone code Cell 2002 111 285 291 12419240
Strahl BD Allis CD The language of covalent histone modifications Nature 2000 403 41 45 10638745
Schreiber SL Bernstein BE Signaling network model of chromatin Cell 2002 111 771 778 12526804
Kurdistani SK Grunstein M Histone acetylation and deacetylation in yeast Nat Rev Mol Cell Biol 2003 4 276 284 12671650
Berger SL Histone modifications in transcriptional regulation Curr Opin Genet Dev 2002 12 142 148 11893486
Hong L Schroth GP Matthews HR Yau P Bradbury EM Studies of the DNA binding properties of histone H4 amino terminus. Thermal denaturation studies reveal that acetylation markedly reduces the binding constant of the H4 “tail” to DNA J Biol Chem 1993 268 305 314 8416938
Luger K Mader AW Richmond RK Sargent DF Richmond TJ Crystal structure of the nucleosome core particle at 2.8 A resolution Nature 1997 389 251 260 9305837
Dhalluin C Carlson JE Zeng L He C Aggarwal AK Structure and ligand of a histone acetyltransferase bromodomain Nature 1999 399 491 496 10365964
Waterborg JH Dynamics of histone acetylation in Saccharomyces cerevisiae
Biochemistry 2001 40 2599 2605 11327883
Vogelauer M Wu J Suka N Grunstein M Global histone acetylation and deacetylation in yeast Nature 2000 408 495 498 11100734
Hebbes TR Thorne AW Crane-Robinson C A direct link between core histone acetylation and transcriptionally active chromatin Embo J 1988 7 1395 1402 3409869
De Nadal E Zapater M Alepuz PM Sumoy L Mas G The MAPK Hog1 recruits Rpd3 histone deacetylase to activate osmoresponsive genes Nature 2004 427 370 374 14737171
Wang A Kurdistani SK Grunstein M Requirement of Hos2 histone deacetylase for gene activity in yeast Science 2002 298 1412 1414 12434058
Kurdistani SK Tavazoie S Grunstein M Mapping global histone acetylation patterns to gene expression Cell 2004 117 721 733 15186774
Bannister AJ Zegerman P Partridge JF Miska EA Thomas JO Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain Nature 2001 410 120 124 11242054
Lachner M O'Carroll D Rea S Mechtler K Jenuwein T Methylation of histone H3 lysine 9 creates a binding site for HP1 proteins Nature 2001 410 116 120 11242053
Shi Y Lan F Matson C Mulligan P Whetstine JR Histone demethylation mediated by the nuclear amine oxidase homolog LSD1 Cell 2004 119 941 953 15620353
Ahmad K Henikoff S The histone variant H3.3 marks active chromatin by replication-independent nucleosome assembly Mol Cell 2002 9 1191 1200 12086617
Ng HH Robert F Young RA Struhl K Targeted recruitment of Set1 histone methylase by elongating Pol II provides a localized mark and memory of recent transcriptional activity Mol Cell 2003 11 709 719 12667453
Dion MF Altschuler SJ Wu LF Rando OJ Genomic characterization reveals a simple histone H4 acetylation code Proc Natl Acad Sci U S A 2005 5501 5506 15795371
Schubeler D MacAlpine DM Scalzo D Wirbelauer C Kooperberg C The histone modification pattern of active genes revealed through genome-wide chromatin analysis of a higher eukaryote Genes Dev 2004 18 1263 1271 15175259
Bernstein BE Kamal M Lindblad-Toh K Bekiranov S Bailey DK Genomic maps and comparative analysis of histone modifications in human and mouse Cell 2005 120 169 181 15680324
Bernstein BE Liu CL Humphrey EL Perlstein EO Schreiber SL Global nucleosome occupancy in yeast Genome Biol 2004 5 R62 15345046
Lee CK Shibata Y Rao B Strahl BD Lieb JD Evidence for nucleosome depletion at active regulatory regions genome-wide Nat Genet 2004 36 900 905 15247917
Yuan GC Liu YJ Dion MF Slack MD Wu LF Genome-scale identification of nucleosome positions in S. cerevisiae
Science 2005 626 630 15961632
Bernstein BE Humphrey EL Erlich RL Schneider R Bouman P Methylation of histone H3 Lys 4 in coding regions of active genes Proc Natl Acad Sci U S A 2002 99 8695 8700 12060701
Robyr D Suka Y Xenarios I Kurdistani SK Wang A Microarray deacetylation maps determine genome-wide functions for yeast histone deacetylases Cell 2002 109 437 446 12086601
Liu CL Schreiber SL Bernstein BE Development and validation of a T7 based linear amplification for genomic DNA BMC Genomics 2003 4 19 12740028
Boeger H Griesenbeck J Strattan JS Kornberg RD Nucleosomes unfold completely at a transcriptionally active promoter Mol Cell 2003 11 1587 1598 12820971
Schwabish MA Struhl K Evidence for eviction and rapid deposition of histones upon transcriptional elongation by RNA polymerase II Mol Cell Biol 2004 24 10111 10117 15542822
Kim M Krogan NJ Vasiljeva L Rando OJ Nedea E The yeast Rat1 exonuclease promotes transcription termination by RNA polymerase II Nature 2004 432 517 522 15565157
Krogan NJ Dover J Wood A Schneider J Heidt J The Paf1 complex is required for histone H3 methylation by COMPASS and Dot1p: Linking transcriptional elongation to histone methylation Mol Cell 2003 11 721 729 12667454
Suka N Suka Y Carmen AA Wu J Grunstein M Highly specific antibodies determine histone acetylation site usage in yeast heterochromatin and euchromatin Mol Cell 2001 8 473 479 11545749
Roh TY Ngau WC Cui K Landsman D Zhao K High-resolution genome-wide mapping of histone modifications Nat Biotechnol 2004 22 1013 1016 15235610
Ng HH Robert F Young RA Struhl K Genome-wide location and regulated recruitment of the RSC nucleosome-remodeling complex Genes Dev 2002 16 806 819 11937489
Robert F Pokholok DK Hannett NM Rinaldi NJ Chandy M Global position and recruitment of HATs and HDACs in the yeast genome Mol Cell 2004 16 199 209 15494307
Lee TI Rinaldi NJ Robert F Odom DT Bar-Joseph Z Transcriptional regulatory networks in Saccharomyces cerevisiae
Science 2002 298 799 804 12399584
Harbison CT Gordon DB Lee TI Rinaldi NJ Macisaac KD Transcriptional regulatory code of a eukaryotic genome Nature 2004 431 99 104 15343339
Kurdistani SK Robyr D Tavazoie S Grunstein M Genome-wide binding map of the histone deacetylase Rpd3 in yeast Nat Genet 2002 31 248 254 12089521
Kadosh D Struhl K Targeted recruitment of the Sin3-Rpd3 histone deacetylase complex generates a highly localized domain of repressed chromatin in vivo Mol Cell Biol 1998 18 5121 5127 9710596
Kasten M Szerlong H Erdjument-Bromage H Tempst P Werner M Tandem bromodomains in the chromatin remodeler RSC recognize acetylated histone H3 Lys14 Embo J 2004 23 1348 1359 15014446
Hurowitz EH Brown PO Genome-wide analysis of mRNA lengths in Saccharomyces cerevisiae
Genome Biol 2003 5 R2 14709174
Sterner DE Berger SL Acetylation of histones and transcription-related factors Microbiol Mol Biol Rev 2000 64 435 459 10839822
Pray-Grant MG Daniel JA Schieltz D Yates JR Grant PA Chd1 chromodomain links histone H3 methylation with SAGA- and SLIK-dependent acetylation Nature 2005 433 434 438 15647753
Lo WS Trievel RC Rojas JR Duggan L Hsu JY Phosphorylation of serine 10 in histone H3 is functionally linked in vitro and in vivo to Gcn5-mediated acetylation at lysine 14 Mol Cell 2000 5 917 926 10911986
Cheung P Tanner KG Cheung WL Sassone-Corsi P Denu JM Synergistic coupling of histone H3 phosphorylation and acetylation in response to epidermal growth factor stimulation Mol Cell 2000 5 905 915 10911985
Wittschieben BO Otero G de Bizemont T Fellows J Erdjument-Bromage H A novel histone acetyltransferase is an integral subunit of elongating RNA polymerase II holoenzyme Mol Cell 1999 4 123 128 10445034
Agalioti T Chen G Thanos D Deciphering the transcriptional histone acetylation code for a human gene Cell 2002 111 381 392 12419248
Ferrell JE Tripping the switch fantastic: How a protein kinase cascade can convert graded inputs into switch-like outputs Trends Biochem Sci 1996 21 460 466 9009826
Ferrell JE How responses get more switch-like as you move down a protein kinase cascade Trends Biochem Sci 1997 22 288 289
Krogan NJ Kim M Tong A Golshani A Cagney G Methylation of histone H3 by Set2 in Saccharomyces cerevisiae is linked to transcriptional elongation by RNA polymerase II Mol Cell Biol 2003 23 4207 4218 12773564
Bernstein BE Tong JK Schreiber SL Genomewide studies of histone deacetylase function in yeast Proc Natl Acad Sci U S A 2000 97 13708 13713 11095743
Holstege FC Jennings EG Wyrick JJ Lee TI Hengartner CJ Dissecting the regulatory circuitry of a eukaryotic genome Cell 1998 95 717 728 9845373
Protacio RU Li G Lowary PT Widom J Effects of histone tail domains on the rate of transcriptional elongation through a nucleosome Mol Cell Biol 2000 20 8866 8878 11073987
Kristjuhan A Walker J Suka N Grunstein M Roberts D Transcriptional inhibition of genes with severe histone h3 hypoacetylation in the coding region Mol Cell Biol 2002 10 925 933
Santos-Rosa H Schneider R Bernstein BE Karabetsou N Morillon A Methylation of histone H3 K4 mediates association of the Isw1p ATPase with chromatin Mol Cell 2003 12 1325 1332 14636589
Friedman N PCluster: Probabilistic agglomerative clustering of gene expression profile. Jerusalem: Hebrew University. 6 p 2003 Available: http://ai.stanford.edu/~erans/module_nets/figures/pcluster.pdf . Accessed 28 July 2005
Benjamini Y Hochberg Y Controlling the false discovery rate: A practical and powerful approach to multiple testing J Royal Stat Soc B 1995 57 289 300
Duda RO Hart PE Pattern classification and scene analysis 1973 New York Wiley 482
Ben-Dor A Friedman N Yakhini Z Overabundance Analysis and Class Discovery in Gene Expression Data. Jerusalem: Hebrew University. 26 p 2002 Available: http://www.cs.huji.ac.il/∼nirf/Papers/BFY2Full.pdf . Accessed 28 July 2005
|
16122352
|
PMC1195719
|
CC BY
|
2021-01-05 08:21:26
|
no
|
PLoS Biol. 2005 Oct 30; 3(10):e328
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030328
|
oa_comm
|
==== Front
BMC GenomicsBMC Genomics1471-2164BioMed Central London 1471-2164-6-1031604281110.1186/1471-2164-6-103Research ArticleConjugative plasmid pAW63 brings new insights into the genesis of the Bacillus anthracis virulence plasmid pXO2 and of the Bacillus thuringiensis plasmid pBT9727 Van der Auwera Géraldine A [email protected] Lars [email protected] Jacques [email protected] Laboratory of Food and Environmental Microbiology, Université catholique de Louvain, Croix du Sud 2/12, B-1348 Louvain-la-Neuve, Belgium2 National Institute of Occupational Health, Lersø Parkallé 105, DK-2100 Copenhagen, Denmark2005 26 7 2005 6 103 103 9 5 2005 26 7 2005 Copyright © 2005 Van der Auwera et al; licensee BioMed Central Ltd.2005Van der Auwera et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Bacillus cereus, Bacillus anthracis and Bacillus thuringiensis belong to the genetically close-knit Bacillus cereus sensu lato group, a family of rod-shaped Gram-positive bacteria. pAW63 is the first conjugative plasmid from the B. cereus group to be completely sequenced.
Results
The 71,777 bp nucleotide sequence of pAW63 reveals a modular structure, including a 42 kb tra region encoding homologs of the Type IV secretion systems components VirB11, VirB4 and VirD4, as well as homologs of Gram-positive conjugation genes from Enterococcus, Lactococcus, Listeria, Streptococcus and Staphylococcus species. It also firmly establishes the existence of a common backbone between pAW63, pXO2 from Bacillus anthracis and pBT9727 from the pathogenic Bacillus thuringiensis serovar konkukian strain 97-27. The alignment of these three plasmids highlights the presence of well conserved segments, in contrast to distinct regions of high sequence plasticity. The study of their specific differences has provided a three-point reference framework that can be exploited to formulate solid hypotheses concerning the functionalities and the molecular evolution of these three closely related plasmids. This has provided insight into the chronology of their divergence, and led to the discovery of two Type II introns on pAW63, matching copies of the mobile element IS231L in different loci of pXO2 and pBT9727, and the identification on pXO2 of a 37 kb pathogenicity island (PAI) containing the anthrax capsule genes.
Conclusion
The complete sequence determination of pAW63 has led to a functional map of the plasmid yielding insights into its conjugative apparatus, which includes T4SS-like components, as well as its resemblance to other large plasmids of Gram-positive bacteria. Of particular interest is the extensive homology shared between pAW63 and pXO2, the second virulence plasmid of B. anthracis, as well as pBT9727 from the pathogenic strain B. thuringiensis serovar konkukian strain 97-27.
==== Body
Background
The Bacillus cereus sensu lato family of rod-shaped Gram-positive bacteria contains six subspecies that are genetically very close [1] but nonetheless have highly specialized lifestyles, especially as concerns their respective virulence spectra. Most notable are B. cereus sensu stricto, an opportunistic pathogen which has been implicated in food poisoning [2] and endophtalmitis [3], B. anthracis, the etiological agent of anthrax [4], and B. thuringiensis which produces δ-endotoxin crystals that are toxic to insect larvae [5]. These subspecies are thought to have emerged from a common ancestor following a series of genetic rearrangements mediated inter alia by mobile DNA elements (transposons, insertion sequences and phages), in synergy with various mechanisms of horizontal gene transfer (conjugation, transduction or transformation), leading to the acquisition of virulence genes. This is exemplified by the presence of large virulence plasmids in B. anthracis (pXO1 and pXO2) [6], emetic strains of B. cereus [7] and B. thuringiensis that carry the genes responsible for the main phenotypic properties by which these bacteria can be distinguished.
While the inter- and intra-molecular movements of mobile elements can obviously have major consequences for the organization and composition of the host genome, it is probably the mechanism of conjugation that best enables the dispersion of these elements throughout the gene pool. Several conjugation systems have been described in Gram-negative bacteria, all of them involving the formation of a sex pilus to bring the participants in close contact, followed by the actual transfer of genetic material via a type IV secretion system (T4SS) [8]. There is much less data available concerning conjugation among Gram-positive bacteria, but the present consensus distinguishes four main transfer strategies, the most common of which seems to be that of the so-called broad host range plasmids. At present, the best characterized of these are the multiresistance plasmids pSK41 [9] and pGO1 [10] from Staphylococcus, pRE25 [11] from Enterococcus, pMRC01 [12] from Lactococcus and pIP501 [13] from Streptococcus as recently reviewed by Grohmann and coworkers [14].
The broad-host-range conjugative plasmid pAW63 was identified in B. thuringiensis serovar kurstaki, where it displays an efficient ability to conjugate in liquid medium, both for its own transmission (around 10-3 transconjugants per donor in broth mating, up to a frequency of 1:1 between kurstaki strains) as well as that of small mobilizable plasmids [15]. Moreover, heterologous conjugation experiments have shown that it is also capable of transfer to its cousins B. thuringiensis serovar israelensis and B. cereus, as well as to Bacillus sphaericus, Bacillus licheniformis [15] and less closely related species such as Listeria innocua and Enterococcus faecalis (A. Wilcks, pers. comm.; G. Van der Auwera and J. Mahillon, unpublished).
In the present study, the complete sequence determination of pAW63 has led to a functional map of the plasmid yielding insights into its conjugative apparatus, which includes T4SS-like components, as well as its resemblance to other large plasmids of Gram-positive bacteria. Of particular interest is the extensive homology shared between pAW63 and pXO2, the second virulence plasmid of B. anthracis [16], as well as pBT9727 from the pathogenic strain B. thuringiensis serovar konkukian strain 97-27 [17].
Results
pAW63 is a 71,777 bp circular molecule
The complete nucleotide sequence of pAW63 was determined to be 71,777 bp long with a G+C content of 33.8%. Analysis of the coding content and organization revealed few intergenic regions, and a total of 76 coding sequences (CDSs) were identified (Fig. 1), of which all but 8 were found to be in the same orientation (defined as counter-clockwise). Interestingly, these 8 CDSs were all located in close proximity to each other, and one of these (CDS 47) corresponded to a component of the pAW63 replicon, which was isolated and sequenced in a previous study [18] while several others were later assigned putative functions as mobile genetic elements as detailed further in the text. The sequence was found to have an 81% coding ratio with an average CDS length of 768 bp. The largest CDS identified was CDS 26, weighing in at 4,248 bp. Several CDSs were associated to distinct spikes in G+C composition and will be discussed individually. The BLAST similarity searches showed that most of the 76 predicted CDSs encoded proteins with similarity to proteins from other organisms, although 36 of these were hypothetical proteins and only 26 could be attributed biological functions, while 14 of the predicted genes did not have any known homologs at all. The 26 genes encoding proteins with discernible functions were assigned to functional categories according to a classification scheme adapted from Riley [19]. The physical details and relevant BLAST hit results (scored by percentage of amino acid identity) of all 76 CDSs are summarized in Table 1.
Figure 1 Circular map of pAW63. Coding Sequences (CDS) are represented by block arrows on the outer circle. Predicted functions/homologies are indicated by the color key featured below; numbers in brackets refer to main CDSs (see Table 1 for details). The first circle from the center delineates the functional modules identified on the plasmid; T indicates the transfer (tra) region, C indicates the control region (replication and regulation), and M indicates the two mobile DNA-associated flanking regions. The second circle from the center is a circular bar graph of the G+C composition percentage of the plasmid sequence, with the overall mean value (33.8%) as baseline; values above the line are G+C rich (max value 51% G+C) and values below the line are A+T rich (min value 16% G+C). The third circle from the center is a graduated size scale with small tick marks every 1 kb and large tick marks every 10 kb.
Table 1 CDSs of pAW63: homologies and comparison with pXO2 and pBT9727
pAW63 s coordinates size (aa) most relevant homology pXO2 pBT9727
CDS* start stop function / microorganism / (amino acid identity percentage 'id %') CDS* id % CDS* id %
#001 - 1151 542 203 - #001 77% #001 67%
#002 - 1450 1162 96 hypothetical protein Efae03001129 Enterococcus faecium (29%) #002 55% #002 52%
#003 - 2585 1727 286 chromosome segregation ATPase E. faecium (25%) #004 80% #003 67%
#004 - 2857 2581 92 - #005 92% #004 62%
#005 - 3554 2945 203 - #006 93% #006 60%
#006 - 4333 3577 252 Tn916 ORF 14, similar to NLP/p60 family lipoprotein E. faecalis (37%) #007 88% #007 78%
#007 - 6254 4448 602 Type II intron reverse transcriptase IepA Bacillus megaterium (44%) - - - -
#008 - 7314 6927 129 Tn916 ORF 14, similar to NLP/p60 family lipoprotein E. faecalis (37%) #007 77% #007 61%
#009 - 9250 7315 645 VirB4, Type IV secretory pathway E. faecium (37%) #008 95% #008 80%
#010 - 9935 9266 223 ATPase involved in DNA repair E. faecium (30%) #009 91% #009 75%
#011 - 10315 10000 105 - #010 95% #010 79%
#012 - 11142 10377 255 - - - - -
#013 - 14900 11528 1124 adhesin AidA, Type V secretory pathway E. faecium (25%) #013 74% #013 48%
#014 - 16905 14925 660 VirD4, Type IV secretory pathway E. faecium (42%) #014 79% #014 85%
#015 - 18508 17839 223 VirD4, Type IV secretory pathway E. faecium (38%) #015 94% #014 83%
#016 - 20463 18534 643 putative membrane protein (no bacterial homologs) #016 65% #015 50%
#017 - 20847 20547 100 - #017 72% #016 43%
#018 - 21116 20864 84 - #018 89% #017 63%
#019 - 21662 21299 121 hypothetical protein Listeria monocytogenes (37%) #020 83% #018 58%
#020 - 22452 21702 250 conserved hypothetical proteinL. monocytogenes (34%) #021 99% #019 70%
#021 - 23282 22448 278 conserved hypothetical proteinL. monocytogenes (32%) #022 69% #020 61%
#022 - 24648 23316 444 VirB11, Type II/IV secretion system L. monocytogenes (35%) #023 87% #021 70%
#023 - 25024 24664 120 conserved hypothetical protein L. monocytogenes (37%) #024 94% #022 70%
#024 - 25273 25123 50 - #025 57% - -
#025 - 25735 25360 125 - #026 77% #023 68%
#026 - 30002 25754 1416 cell surface protein, similar to Rhs family Bacillus cereus (20%) #027 78% #024 84%
#027 - 30709 30556 51 - - - - -
#028 - 31117 30724 131 - - - - -
#029 - 31693 31204 163 - - - - -
#030 - 32268 31794 158 - - - - -
#031 - 32492 32300 64 - - - - -
#032 - 33216 32880 112 - - - - -
#033 - 33820 33694 42 - - - - -
#034 - 33693 33234 153 - - - #026 58%
#035 - 34082 33827 85 - - - - -
#036 - 34439 34100 113 - - - - -
#037 - 34874 34622 84 - - - #028 66%
#038 - 35824 34972 284 - - - - -
#039 - 36199 35968 77 - - - #033 50%
#040 - 36488 36371 39 - - - - -
#041 - 37060 36571 163 - #031 88% #034 39%
#042 - 37861 37336 175 - #032 84% #036 50%
#043 - 38649 38346 101 - #033 90% #037 60%
#044 - 39688 39532 52 - #034 66% - -
#045 - 40902 39684 406 prophage helix-turn-helix protein Bacillus cereus G9241 (25%) #036 89% #038 59%
#046 - 41848 41089 253 - #038 88% #045 70%
#047 - 43592 42050 514 replication protein RepE E. faecalis (40%) #039 96% #046 83%
#048 + 44458 45385 309 replication-associated protein RepB E. faecalis - - - -
#049 + 45347 45671 108 putative replication-associated (no homologies) (31%) - - - -
#050 - 46557 45735 274 - - - - -
#051 - 48301 46663 546 pheromone binding protein B. cereus G9241 (73%) #089 90% #055 80%
#052 - 49424 48788 212 DNA-binding protein B. cereus G9241 (53%) #093 61% #059 77%
#053 + 49620 49968 116 conserved hypothetical protein Staphylococcus aureus (31%) #095 75% #062 71%
#054 + 50020 50407 129 group specific protein B. cereus ZK (75%) #096 77% #063 76%
#055 - 51260 50471 263 hypothetical protein lpl1076 Legionella pneumophila (49%) - - - -
#056 - 51589 51343 82 hypothetical protein B. cereus ATCC 14579 (60%) - 86% - -
#057 + 52288 53362 358 RapD response regulator B. thuringiensis (94%) - - - -
#058 - 54506 53636 290 transposase, IS5 family Aneurinibacillus thermoaerophilus (52%) - - - -
#059 + 54988 55699 237 - - - - -
#060 + 56252 56870 206 site-specific recombinase, resolvase Clostridium thermocellum (54%) - - #065 69%
#061 + 57303 57834 177 phage site-specific recombinase B. cereus ATCC 14579 (64%) #101 - #066 73%
#062 - 58650 58203 149 CAAX N-terminal protease B. cereus ATCC 14579 (35%) #103 - #071 68%
#063 - 60821 58679 714 DNA topoisomerase, TrsI/TraI Lactococcus lactis (45%) #104 - #072 74%
#064 - 61323 60942 127 - - - - -
#065 - 62731 61429 434 conserved hypothetical protein B. cereus G9241 (35%) - - - -
#066 - 63383 62798 195 AbiQ abortive infection mechanism L. lactis (34%) - - #073 28%
#067 - 64659 63633 342 signal transduction histidine kinase E. faecium (38%) #107 90% #074 83%
#068 - 65290 64738 184 putative membrane-bound hydrolase Bacillus licheniformis (32%) #109 - #075 92%
#069 - 66240 65313 309 - #110 83% #076 60%
#070 - 67475 66302 391 ATPase involved in DNA repair E. faecium (42%) #111 77% #077 94%
#071 - 67681 67471 70 - #112 62% #078 75%
#072 - 68132 67703 143 - #113 80% #079 69%
#073 - 68628 68151 159 - - - - -
#074 - 69124 68695 143 - - - - -
#075 - 70679 69197 494 - - 49% #079 50%
#076 - 71773 70744 343 ATPase involved in DNA repair E. faecium (36%) - 96% #080 83%
* CDS numbers are abbreviated forms originating from their formal designations: pAW63-###, GBAA_pXO2_0###, pBT9727_0###
pAW63 generic relationships
Most of the significant similarity results originated from Gram-positive bacteria species such as Enterococcus spp., Streptococcus spp., Staphylococcus spp., Lactococcus spp. and Listeria spp. in addition to Bacillus halodurans, Bacillus subtilis, Bacillus licheniformis and the members of the B. cereus sensu lato group. It is of particular interest to note that out of the 76 CDSs predicted on pAW63, 50 CDSs showed strong similarity (between 48.8 and 98.7% amino acid identity, with an average of 81.1%) to CDSs found on pXO2 from B. anthracis [GenBank:NC_007323] and 49 CDSs showed strong similarity (between 43.8 and 97.4% amino acid identity, with an average of 70.5%) to pBT9727 from B. thuringiensis serovar konkukian [GenBank: CP_000047], with 42/76 (55.3%) of these CDSs being shared by all three plasmids. This was consistent with previous observations of sequence similarities between these plasmids [18,20,21].
Replication and regulatory functions are grouped in an 8 kb 'control center'
All the elements that were putatively identified as being directly involved in plasmid replication, copy control or other regulatory processes were found to reside within an 8 kb region of the plasmid containing 12 CDSs (CDS 46 to 57), of which 6 could not be assigned to a functional category. This region was delineated by mobile genetic elements or remnants thereof, and contained five of the eight CDSs that were found in clockwise orientation, the other three being located in its gene mobility-associated flanking sequences.
The 4.1 kb replicon of pAW63 was characterized in a previous study [18] and classified as belonging to the pAMβ1 family of theta-replicating conjugative plasmids, with which it shares a similar cis-functioning origin of replication (ori). The largest of the four CDSs contained in this region, Rep63A (CDS 47), displayed strong similarity to the replication proteins of several plasmids in this family, as well as 96% amino acid identity with the RepS protein of the pXO2 replicon [18]. Rep63B, the second largest CDS of the replicon (CDS 48), displayed strong similarity with copy control proteins RepB and PrgP from the Enterococcus faecalis conjugative plasmids pAD1 [22] and pCF10 [23], respectively [18]. Although the two smaller CDSs (49 and 50) did not display any significant homologies, the authors noted that one of these, CDS 49, showed properties (location, size, orientation and hydrophilicity) that likened it to repC and prgO, which are putative genes encoding stability functions on pAD1 and pCF10, respectively.
Upon further sequencing of the plasmid, three additional putative regulatory elements were identified in close proximity to the replicon. These elements (CDS 51, 52 and 57) were homologs of genes that are highly conserved throughout the B. cereus group as well as in some less closely related species. CDS 51 and 52 displayed significant homologies to an oligopeptide ABC transporter functioning as a pheromone binding protein, and to a DNA-binding protein, respectively. The clockwise-oriented CDS 57 showed a distinctly low overall G+C content and was found to encode a protein with 94% identity to RapD, a response regulator aspartate phosphatase acting as a transcriptional activator involved in the regulation of sporulation [24].
The conjugative functions of pAW63 are grouped in a 42 kb tra region
All 15 CDSs assigned putative conjugative functions were found to be located within a 42 kb region of the plasmid (CDS 1 to 26 and 62 to 76; hereafter referred to as the tra region). Every one of the 41 CDSs predicted within this region was in the same counter-clockwise orientation, suggesting an operon-style transcription scheme may be in place, which would be consistent with the genetic organization of most known conjugative systems [25].
pAW63 encodes three homologs of Gram-negative T4SS components
Three CDSs were found to encode proteins with significant similarity to components of the Vir secretion system, originally characterized in Agrobacterium tumefaciens, which is the archetypal model for the Type IV Secretion System (T4SS) that is believed to carry out the conjugative process in Gram-negative bacteria, as recently reviewed [8,26]. Briefly, the T4SS can be summarized as a two-step mechanism [27] which involves the replication of the plasmid DNA by effector proteins followed by the translocation of a single-stranded molecule across the cell envelope into the recipient cell. The transfer is powered by the action of a molecular pump and the DNA is thought to travel through a channel formed by a core protein complex, while the necessary close cell-to-cell contact is mediated by surface structures.
Analysis of the pAW63 sequence (Fig. 1) showed that CDSs 14 and 15 were in fact the two halves of an interrupted CDS encoding a homolog of the VirD4 component, which has been shown to function as a coupling protein between the DNA strand replication machinery and the transmembrane transfer complex and powerhouse. Interestingly, the 935 bp space between them was found to be completely devoid of coding sequences and showed a distinctly higher G+C content than was considered average for the rest of the plasmid. CDS 9 displayed 37% identity with VirB4, a membrane associated N-triphosphatase which is thought to play a role in providing the energy needed to power the actual DNA transfer. CDS 22 displayed 35% identity to VirB11, a subunit implicated in both Type II and Type IV secretion systems that was shown to be necessary for conjugation, although its exact role remains unclear. Structural analyses have shown it to possess a transmembrane domain as well as an ATPase domain, and recent research suggests that the subunits are arranged as a dynamic hexameric assembly associated with the inner membrane and functioning as a gating component [28].
pAW63 encodes homologs to components of Gram-positive conjugation systems from diverse species
Another 10 CDSs were classified as conjugative on the basis of their homologies, this time to genes from Gram-positive conjugation systems. Most notably, CDS 63 showed a distinctly high G+C content and was found to encode a topoisomerase displaying 45% identity to the TrsI/TraI protein from the Lactococcus lactis conjugative plasmid pMRC01 [12]. This component of the lactococcal conjugative system is thought to function as a relaxase, a role equivalent to that of VirD2 in the T4SS machinery.
CDSs 67 and 70 both displayed significant homology to LtrC-like (from the Lactococcal transfer ORF C) putative conjugative elements from Streptococcus, Staphylococcus, Lactococcus and Listeria species (with an average 30% identity), although they had different 'most relevant BLAST hits', as shown in Table 1. The function of the LtrC element itself, which was identified in the tra region of the lactococcal conjugative plasmid pRS01 [29], has not yet been determined.
CDSs 6 and 8 were predicted to be the two halves of an interrupted protein with significant resemblance to ORF 14 from the enterococcal conjugative transposon Tn916, a self-transmissible molecule encoding tetracycline resistance [30]. The function of Tn916 CDS 14 has not yet been elucidated, but further results of the similarity search displaying comparable identity levels included a soluble lytic murein transglycosylase from E. faecium as well as the invasion-associated extracellular protein Iap from L. monocytogenes.
Furthermore, three CDSs (CDS 13, 16 and 26) were predicted to encode proteins that may be involved in the establishment of intimate cell-to-cell contact, either as aggregation substances secreted into the medium, or as cell surface determinants. CDS 13 showed a distinctly high G+C content in its second half, which contained two tracts of distinct repeated units that were predicted to correspond to a series of helix-turn-helix motifs followed by a series of pleated sheets in the protein product. It also displayed 25% identity with the adhesin AidA from the Type V secretory pathway of E. faecium, which relies on an autotransporter system for cytolysin secretion. The adhesin itself is primarily implicated in the pheromone-mediated enterococcal mating process as an aggregation substance, but it has also been shown to act as an adhesion factor in the course of infection of eukaryotic cells by pathogenic E. faecalis strains and as such it may be considered a virulence determinant [31]. CDS 26 was the largest CDS found on the plasmid (4.2 kb) and showed little significant resemblance to any known sequences except for a positive match (20% identity) with a 3.3 kb long cell-surface protein from the B. cereus reference strain ATCC14579 and another (23% identity) with a 2.37 kb long Rhs family protein (for Recombination hot spot) from Bifidobacterium longum. Rhs products are thought to possess properties typical of cell surface proteins and often present internal rearrangements that lead to antigenic variation [32]. CDS 16 was unlike any known bacterial sequences (apart from those of plasmids pBT9727 and pXO2) but the putative 643 residue protein was predicted to have a secondary structure consisting essentially of hydrophilic helices, suggesting it may function as a surface-associated or free-acting aggregation substance.
Finally, the remaining CDSs correspond to ATPases predicted to be involved in segregation and DNA repair (CDSs 3, 10, 62 and 76), as well as a predicted membrane-bound hydrolase (CDS 68).
pAW63 and mobile genetic elements
Five CDSs were readily identified as mobile genetic elements on the basis of their homologies. Standing apart from the other four by both character and location, CDS 7 was found positioned right between the two fragments of the interrupted Tn916 ORF 14-like element and displaying 44% identity to IepA, a group II intron-encoded protein from Bacillus megaterium which further BLAST results revealed to be a reverse transcriptase and maturase. This immediately led to the conclusion that CDS 7 was part of a group II intron that had inserted itself in the uninterrupted ancestor of CDSs 6 and 8, a finding supported by the skewed G+C content profile corresponding to the site of the proposed insertion. The element was tentatively named B.th.I1 (for B. thuringiensis Intron #1). Furthermore, a closer examination of the 935 bp fragment of seemingly non-coding DNA with high G+C content found interrupting the VirD4 homolog (CDSs 14 and 15) revealed that the extremities of this insert were nearly identical to those of the B.th.I1 intron mentioned above, indicating the presence in this locus of a second retroelement presumably derived from the first. This second element was tentatively named B.th.I2 (for B. thuringiensis Intron #2).
Three other of these CDSs (CDS 58, 60 and 61) were found in close proximity to each other by the regulation side of the proposed 'control center' region, as mentioned previously, and were predicted to encode various types of DNA recombinases.
Finally, CDS 45 was found flanking the replication side of the replication control center and showing similarity to a prophage helix-turn-helix protein from the B. cereus strains G9241 and ATCC 14579. The 10 kb region downstream of CDS 45 was found to encode 18 rather short CDSs (≈300 bp on average) with no significant homologies in a configuration reminiscent of typical bacteriophage gene structure, which may correspond to a prophage integrated within the plasmid sequence. Intriguingly, while the first 4 kb of this region showed a G+C content in accordance with the rest of the plasmid, the following 6 kb segment showed a distinctly low G+C content along most of its length.
As a counterpoint to the considerable amount of phage-derived genetic material tentatively identified on the plasmid, CDS 66 displayed significant similarity to AbiQ, a single-protein abortive infection mechanism from L. lactis which limits phage dissemination by shutting down the lytic cycle and leading the infected host cells to their death [33].
pAW63 shares a common backbone with pXO2 from B. anthracis and pBT9727 from B. thuringiensis konkukian
Similarities between pAW63 and the virulence plasmid pXO2 from B. anthracis had been observed previously [18,20] and further investigation (C. Kuske, pers. comm.) had yielded thirty short sequences (400 bp long on average) corresponding to regions of pAW63 which had been shown to hybridize to pXO2. This data was used in this study as the basis for the sequencing of pAW63, in a strategy that involved the outlining of a backbone sequence from which primers pairs were designed to amplify and individually sequence the intervening regions.
These premises had guaranteed at least several hits to the virulence plasmid from B. anthracis, pXO2, and the final sequence of pAW63 certainly did not disappoint. Indeed, a preliminary alignment of their sequences showed that aside from a 37-kb region found only on pXO2 corresponding to the region containing the anthrax capsule genes and associated regulatory elements, they possessed a highly similar genetic makeup. Furthermore, the BLAST results had also revealed the existence of a third closely related plasmid, pBT9727, which was recently published along with the genome sequence of its host strain B. thuringiensis serovar konkukian strain 97-27 [GenBank:NC_005957] and shows a comparable overall gene organization and coding content to that of the other two. These findings were consolidated by performing a BLAST Score Ratio Analysis of the predicted proteomes of the three plasmids, the results of which indicated a high level of synteny, with the pAW63 and pBT9727 pair showing the highest ratio of shared CDSs although the associated identity percentages were the poorest. Regarding the CDSs shared by each of these plasmids with pXO2, pBT9727 possessed the highest number of these but for the most part pAW63 showed higher identity percentages (Tables 1 and 2). A global phylogenetic analysis was performed on the three proteomes by systematically aligning the 42 homologous predicted proteins shared by all three plasmids together along with their most relevant common BLAST hit result and building a tree from the four sequences (data not shown). Results were consistent for all quartets examined and suggested that pBT9727 had been the first to branch off from the common ancestor, while pAW63 and pXO2 diverged later. The detailed alignment of the three sequences shown in Figure 2 clearly illustrates the extent of the synteny observed between these plasmids.
Figure 2 Linear alignment of pXO2, pAW63 and pBT9727. CDSs are represented by block arrows. Several CDS numbers (see Table 1) are indicated for reference on each plasmid, just above or below their representation. Predicted functions/homologies are indicated by the color key featured below. Well conserved segments of the plasmids are paired by shaded regions (>40% amino acid identity); percentages for specific CDS pairs can be found in Tables 1 and 2. The proposed PAI of pXO2 is raised above the rest of the sequence for clarity. Scale is indicated by the bar in the lower right-hand corner.
Table 2 Comparison of pXO2 and pBT9727 CDSs with no homologs on pAW63
B. anthracis CDS id % B.t. konkukian CDS Putative gene functions
GBAA_pXO2_0011 65% pBT9727_0011 -
GBAA_pXO2_0012 60% pBT9727_0012 -
GBAA_pXO2_0028 67% pBT9727_0024 -
GBAA_pXO2_0035 64% pBT9727_0038 -
GBAA_pXO2_0040 91% pBT9727_0047 replication-associated protein
GBAA_pXO2_0041 79% pBT9727_0048 replication-associated protein
GBAA_pXO2_0052 89% pBT9727_0039 -
GBAA_pXO2_0053 72% pBT9727_0040 CAAX N-terminal protease
GBAA_pXO2_0054 92% pBT9727_0041 -
GBAA_pXO2_0055 97% pBT9727_0042 transcriptional regulator TetR
GBAA_pXO2_0056 96% pBT9727_0043 IS231 transposase
GBAA_pXO2_0057 66% genomic? bacitracin transport permease
GBAA_pXO2_0061 39% genomic? -
GBAA_pXO2_0075 36% genomic? sensor histidine kinase
GBAA_pXO2_0086 81% pBT9727_0043 IS231 transposase
GBAA_pXO2_0094 93% pBT9727_0061 DNA-damage repair protein
The tra region of pAW63 finds its equivalent on both pXO2 and pBT9727, with high levels of identity and contrasting discrete variations
Both pXO2 and pBT9727 were found to possess an approximately 42 kb long region that was almost identical to the tra region of pAW63 with respect to gene structure and organization, as illustrated in Fig. 2. All of the CDSs identified on pAW63 as possible conjugative genes were found to have highly similar homologs on both of these plasmids (see percentages in Tables 1 and 2), although several displayed key differences as detailed below.
While the copy of the Tn916 CDS 14-like element present on pAW63 had been found to be encoded by two CDSs (CDSs 6 and 8) due to the insertion of the group II intron B.th.I1 (see above), the corresponding homologs of this element were uninterrupted on pBT9727 and pXO2. The sequence comparison of these 'native' versions of the gene with the one containing the intron allowed for the precise site of insertion to be determined. This region of the Tn916 CDS 14-like element was then aligned with its closest homologs as identified by BLAST search, revealing that the insertion had taken place in a relatively well-conserved domain of the gene. This is consistent with the observation that group II introns target specific genes or gene domains [34] despite the lack of experimental data available concerning the insertion of this particular element.
Similarly, the three plasmids possessed separate versions of the VirD4 homolog. The CDS encoding this gene was found to be interrupted in a distinct manner on both pXO2 and pAW63, while it appeared to be intact on pBT9727. As indicated previously, the pAW63-borne homolog (CDS 14 and 15) was interrupted by the putative ORF-less group II intron B.th.I2, while the cause of the disruption of the pXO2-borne homolog was a frameshift due to the deletion of two nucleotides in a C-rich stretch located at two-thirds of the original protein.
Likewise, homologs for the putative cell-surface protein encoded by CDS 26 of pAW63 were found on both pXO2 and pBT972, albeit in two different forms. While pBT9727 was found to possess a presumably intact version of the gene, the pXO2-borne version appeared to be disrupted by a single nucleotide deletion causing a frameshift two-thirds into the original protein. This was confirmed to be a genuine feature and not a sequencing artifact by comparing the three available sequences of pXO2 [GenBank:NC_002146, GenBank:NC_007323, GenBank:NC_003981] which had identical frameshifts in the corresponding locus.
Finally, the three homologs of the putative adhesin (CDS 13 on pAW63) were found to differ in a very interesting way. While the first half of the protein was well conserved in all three versions, the second half containing the repeated motifs was almost identical between pXO2 and pAW63, although pXO2 lacked several repetitions of the basic units, and pBT9727 possessed completely different basic units which nevertheless gave rise to highly similar secondary structure predictions.
Sequence variability is highest in the replication control center area and is associated with the presence of multiple mobile genetic elements
The mobile DNA-based borders of the replication control center of each plasmid were found to be comparable as follows. The putative prophage structure bordering the replicon (CDS 27 to 45 on pAW63) was found on all three plasmids, although the second part of this structure was poorly conserved, to the point of being absent from pXO2. The other side of this region was delineated in all three cases by variants of the site-specific recombinase found on pAW63 (CDS 61). The area between the replicon and this recombinase was highly variable between the three plasmids, with pAW63 featuring the most CDSs therein (CDS 50 to 60).
Regarding the regulatory elements present in this region, it was observed that while both the putative pheromone receptor (CDS 51) and the DNA-binding protein (CDS 52) identified on pAW63 were also present as close homologs on the two other plasmids, the putative transcriptional activator RapD (CDS 57) was not found in any form on either of them. On the other hand, both pBT9727 and pXO2 were found to possess almost identical copies of two other genes that were not present on pAW63: a homolog of the DNA-damage repair gene uvr from B. subtilis, and the gene encoding the transcriptional activator TetR.
A recent recombination event caused pAW63 to exchange part of its replicon
It is interesting to note that while the origin of replication and main replication protein Rep63A (CDS 47) were found to be highly conserved across all three plasmids, no homologs were found for the two other replication-associated CDSs of pAW63 (CDS 48 and 49), but two CDSs of similar size and orientation were found in equivalent locations relative to the common backbone on each of the other two plasmids, suggesting they share a functional homology with the two pAW63 CDSs, despite their lack of sequence similarity. Furthermore, these pBT9727- and pXO2-borne pairs of CDSs were found to share a high level of sequence identity with each other, suggesting that the original corresponding part of the pAW63 replicon had been exchanged for its present form at some point following the divergence of pXO2 and pAW63. In support of this idea, a similar pattern of resemblance/difference was observed for a series of partially palindromic iterons present in the vicinity of the replication genes which were previously believed to be implicated in the replication process. While the iteron sequences found on pBT9727 and pXO2 were by no means identical, they did share similar repeated motifs that were completely different from those identified on pAW63. The discovery of further repetitions of these various motifs as well as several additional palindromes in key locations of the sequence have led to the elucidation of a complex structure of palindromic and/or iterative elements serving as node points for recombination events that may have been responsible for major divergences in synteny between the three plasmids in the replicon area, as illustrated in Figure 3.
Figure 3 Relational diagram of the replicon region of the three plasmids. Comparison of the replicon region of each plasmid reveals a complex structure of palindromic and/or iterative elements serving as node points for recombination events. Sequence segments are represented by thick horizontal lines joined by solid diagonal lines. The background striping highlights 'shared' versus 'unique' sequence segments as indicated by the text legend on the left hand side. Horizontal dashes are spacers to indicate a shorter length or lack of corresponding segment. Node points putatively involved in recombination events are represented by green circles; iterative and/or palindromic sequence units by half arrows; CDSs by rectangles, above the sequence line to indicate clockwise orientation, or below to indicate counter- clockwise orientation.
Comparative evidence supports the existence of a 37 kb pathogenicity island (PAI) on pXO2
The 37 kb region of DNA located after the replicon on pXO2 was found to exhibit most of the characteristics of a bacterial pathogenicity island (PAI) as defined by Hacker and Kaper [35]. All of the genes carried by pXO2 that are known to play a capital role in the course of anthrax infection, such as the capsule genes, were clustered in this region, along with the regulatory genes acpA and acpB, which are directly associated with their expression during the development of the pathology [36]. Furthermore, this region had distinctive genetic properties from the rest of the plasmid, such as a lower G+C content (30.9% against 34.3%), and more variation in the orientation and phase of coding sequences. It was also found to be rich in mobile genetic elements (some cryptic or defective), such as multiple inactivated copies of the IS231 transposase [16,37] and a vestigial class II transposon.
The comparison of the sequences flanking the proposed PAI on pXO2 and their equivalents on pAW63 and pBT9727, as described above (Fig. 3), revealed the existence of specific nodes consisting of iterative and/or palindromic sequences that could be held responsible for the acquisition of the PAI region through a past recombination event, though not excluding the possibility of later rearrangements having taken place within this region.
pXO2 and pBT9727 possess copies of Insertion Sequence IS231L in different loci, causing a visible interruption in synteny
The results of the three-way BLAST Score Ratio Analysis of this complex plasmid trio highlighted a group of five CDSs that were not present on the pAW63 sequence itself but were shared with a high degree of similarity between pBT9727 and pXO2. Their location was different on either plasmid with respect to the common backbone, thus representing a significant interruption in the synteny of the plasmids. It appeared that this small group had been previously described on pXO2 as an IS231-derived [38] Insertion Sequence (IS) and designated IS231L [37], although it was assumed to be incapable of autonomous transposition due to a frameshift in the CDS encoding the transposase. However, the copy of the IS231L element harboured by pBT9727 was found to have an uninterrupted and therefore supposedly functional transposase, as well as nearly perfect Inverted Repeats (IR), the canonical 20 bp boundaries of this type of mobile element matching those of the pXO2-borne IS231L element.
Interestingly, the copy of IS231L found on pBT9727 was found to be nested within the Tn1546-related [39] left IR of a novel Class II transposon belonging to the Tn3 family, encoding a single cryptic CDS and tentatively named TnBt9727. Likewise, the pXO2-borne copy of IS231L was nested within the TnARS1-related [GenBank AY780525] right IR of a vestigial class II transposon. No corresponding left IR was found for this element, but an IS3 family transposase was found in close proximity to the putative right IR, suggesting a possible structural association.
Furthermore, the presence of two other CDSs with homology to the IS231 transposase in the PAI of pXO2 prompted a search for other instances of IRs matching those of IS231L in the vicinity of these CDSs. A well-conserved copy of the right IR was indeed discovered near the end of the PAI beside a second CDS encoding an IS231 transposase-like gene (Fig. 2).
Discussion
A large proportion of the pAW63 sequence was observed to possess significant similarity to plasmids originating from several different bacterial species harbouring fundamentally different conjugative and replication systems. The resulting hybrid combination of genes probably opens up the mating scope of the plasmid. The physical arrangement of the plasmid was identified as a composite structure consisting of a 42 kb tra region encoding conjugation functions, organised in typical operon fashion, and an 8 kb control center encoding the replication and regulation functions, joined by mobile genetic elements or remains thereof (Fig. 1).
Analysis of the replication control center of pAW63 revealed that it possessed copy control proteins similar to those found on the enterococcal conjugative pheromone plasmids pAD1 and pCF10. However, pAW63 conjugative transfer among B. thuringiensis strains has been shown to be independent from recipient-encoded pheromones [15]. It has been suggested that in addition to their role in the transfer process, host-produced pheromones may play a role in the replication of the pAD1 and pCF10 plasmids [23]. Given a similar copy control mechanism, the apparent independence of pAW63 from pheromone signalling may account for the greater host range of pAW63 compared to pAD1 and pCF10, by removing the requirement for a compatible pheromone system in the recipient mating partner. It was therefore very interesting to find near the pAW63 replicon two CDSs respectively encoding an ABC transporter-type pheromone receptor, homologous to those found in Enterococcus and Listeria species, and a close homolog of RapD, a transcriptional activator from B. thuringiensis serovar morrisoni. At this point it is unclear whether the function of these elements is restricted to either the plasmid copy control or the regulation of conjugation, or whether these processes are directly linked by common regulatory pathway components. Nonetheless, the presence of a putative pheromone-sensing mechanism on pAW63 suggests that the conjugative transfer process of the plasmid may follow alternate or optionally additive regulatory pathways depending on the species of the mating cells.
Regarding basic conjugative function, the presence on the plasmid of genes related to the T4SS system corroborates similar findings [40] concerning the transfer system of the streptococcal plasmid pIP501, which bears a tra region containing VirD4, VirB4 and Vir B11 homologs. Interestingly, all three of the pAW63 and pIP501 T4SS-like proteins correspond to components of the system that operate within the cell or in association with the inner membrane to prepare the plasmid DNA for transfer and provide motor function, and no homologs have been found on the plasmid to the 'outer' components. This could be explained by the idea that the inner part of the Gram-positive cell wall structure is sufficiently similar to that of the Gram-negative hosts for some of the 'inner' components, designated as effector and transporter molecules, of both transfer systems to share their main characteristics and possibly a common evolutionary history derived from replication mechanisms. Conversely, it seems reasonable to assume that major differences in the cell wall structure and composition, such as the presence of a thick peptidoglycan layer and the absence of an outer membrane, would be mirrored by the presence of specific protein components responsible for the 'external' steps of the conjugative process. This includes sensing of recipients, cell to cell attachment and the bridging of cell walls to effect DNA transfer, and requires the assembly of a core complex as well as cell surface structures. With respect to the latter, several putative cell surface- or membrane-associated proteins, such as the adhesin encoded by CDS 13 and the very large protein encoded by CDS 26, have already been identified in the tra region of pAW63, although further studies will be necessary to work out their specific roles in the conjugative process.
The presence of introns B.th.I1 and B.th.I2 within two genes of which at least one (the VirD4 homolog) may be considered essential to the conjugative process, poses the question of the functionality of the interrupted genes, as well as that of possible collateral effects caused by these elements. Regarding maintenance of functionality of the host genes, current models of group II intron splicing as well as experimental research show that intron excision during transcription leads to the production of a complete and functional protein [41]. In addition, this type of mobile element is thought to insert preferentially into certain conjugative genes such as relaxases, and it has been suggested that such an insertion actually enhances the efficiency of the conjugative system [34,42]. Furthermore, the relationship between B.th.I1, which encodes its own reverse transcriptase, and B.th.I2, which does not, deserves further investigation. While such a configuration has rarely been observed among eubacteria, ORF-less introns that were closely related to an intron encoding a reverse transcriptase located in the same genome have been identified in cyanobacteria and archeae [43]. It has been proposed that these CDS-less elements were derivatives of their 'complete' RT-encoding relatives [44]. Significantly, several of these 'degenerate' elements were shown to be mobile, raising the possibility that mobility may be supplied in trans [41] as is the case for the Mobile Insertion Cassette MIC231 trans-activated by IS231 transposase [37,45].
The comparison of the tra regions of pAW63, pBT9727 and pXO2 has provided several valuable indications on the relative importance of specific genes in the conjugative process. While pAW63 is known to be fully functional and quite efficient as a conjugative plasmid, pXO2 is believed to be incapable of autonomous transfer although it has been shown to be mobilizable by the conjugative plasmid pXO14 from B. thuringiensis toumanoffi [46], suggesting that pXO2 carries a set of conjugative genes that is incomplete or disrupted in one or several key components. Both pXO2 and the newly sequenced plasmid pBT9727 were now shown to possess a set of potential conjugation genes closely related to those of pAW63. The transfer capabilities of pBT9727 have not yet been assayed, but its sequence features point towards partial conjugative function at the very least, and possibly full self-transmissibility. For these reasons, further studies will focus on loci exhibiting discrete variations in order to expose their specific contribution to the transfer process. For instance, the unique interruptions found in the pXO2 homolog of the VirD4 element and in that of the putative cell surface protein encoded by CDS 26 on pAW63 may prove to be particularly significant, especially if pBT9727 is shown to be fully conjugative.
From a broader point of view, the combination of the phylogenetic analysis of the three proteomes with the elucidation of the structural features delineating the major sequence differences in the replication control center has made it possible to roughly reconstruct the molecular evolution of the plasmid trio. The revelation that pBT9727 was the first to diverge from the common stem, leaving pXO2 and pAW63 to branch off from an intermediate form some time later, carries several important implications for the ecology of the B. cereus s.l. family. At a very basic level, it highlights the mixed lineage of these plasmids, as attested by their distinct genetic properties (conjugation versus virulence) contrasting with the nearly perfect conservation on all three of the minireplicon of pXO2 [47]; plasmids which were isolated from strains that are treated quite differently, namely as a biopesticide versus a dangerous pathogen. This is particularly relevant in the light of recent incidents linking B. cereus sensu stricto strains to cases of anthrax-like pathologies [21,48,49], as well as the discovery that the cereulide genetic determinants of emetic B. cereus are located on a plasmid [7,50]. Most significantly, it makes a strong case for the pXO2 capsule genes region to be officially designated a pathogenicity island, a hypothesis further validated by the structural features typical of mature PAIs that were identified in this study. The recognition of pXO2 as a PAI-equipped virulence plasmid would also be consistent with the observation that most of the major known pathogens possess such PAIs, many of these carried by virulence plasmids, and it would put pXO2 on par with pXO1, which is known to carry its own anthrax toxin-bearing PAI. In addition, it should be noted that the configuration of IS231-related IRs observed on pXO2 (Fig. 2), which is quite likely due to the insertion of the IS231L element into the PAI after its acquisition by pXO2, entails that a region representing almost the entire PAI, containing all the capsule genes and their associated regulatory elements, can technically be considered an IS231-derived Mobile Insertion Cassette (MIC). As such, this cassette could undergo a transposition event provided a functional IS231 transposase is supplied in trans. Incidentally, this opportunity for virulence genes to be directly mobilized by a mobile genetic element from within a larger PAI is paralleled in pXO1 by the presence in its PAI of a Class II transposon, TnXO1, which carries the anthrax germination operon gerX as its passenger genes [51].
Methods
Bacterial isolates and PCR amplification
B. thuringiensis strain HD73 containing pAW63 was grown in liquid LB medium at 30°C. 1 ml cultures grown overnight were centrifuged into a pellet, washed twice in ddH20 and resuspended in 250 μl ddH20. Freshly prepared aliquots of 5 μl of resuspended cell solution were used as template in each subsequent 50 μl PCR reaction. PCR amplification was performed using 'Hi-Fi polymerase mix' from Fermentas according to the manufacturer's specifications. For target fragments of over 3 kb it was necessary to adjust the MgCl2 concentration of the reaction mix as well as adding dimethylsulfoxide (DMSO) to optimize yield and specificity. Further details can be obtained upon request. Cloning procedures made use of One Shot® TOP10 Electrocomp™ E. coli from Invitrogen, and transformation was performed according to the manufacturer's specifications.
Sequencing strategy
The sequencing strategy took advantage of the previously observed similarity between pAW63 and the virulence plasmid pXO2 from B. anthracis [18,20]. Further investigation (C. Kuske, pers. comm.) yielded thirty short sequences (400 bp long on average) corresponding to regions of pAW63 which had been shown to hybridize to pXO2. This data was used to construct a backbone sequence on the basis of which primer pairs could be designed to amplify the intervening regions. The resulting amplicons were purified on gel using Qiagen QiaQuick Gel Purification Kit according to the manufacturer's specifications. Amplicons smaller than 5 kb were directly sequenced from the purified PCR sample, those between 5 and 10 kb were cloned into sequencing vectors, and those larger than 10 kb were each subcloned into separate libraries. All cloning procedures were performed using the Topo XL PCR cloning kit of Invitrogen according to the manufacturer's specifications. Amplicons were individually sequenced at the Flanders Interuniversity Institute for Biotechnology (VIB) Genetic Service Facility by double strand primer walking following standard procedures with a final accuracy of 99.98% per amplicon. The primary assembly of individual amplicon sequences was performed using the Seqman software package. The overall assembly quality was improved by sequencing regions of minimum 500 bp in length overlapping the extremities of adjoining amplicons using the same protocol as described above.
Sequence analysis and annotation
The full assembly of the amplicons, overlapping fragments and the previously published pAW63 replicon was done manually using the Accelrys DS Gene sequence editor. The coordinate start of the sequence was assigned with respect to that of pXO2 in order to facilitate future comparative analyses between the two plasmids, for which the B. anthracis 'Ames Ancestor' strain sequence of pXO2 was used [GenBank:NC_007323]. Potential coding regions were predicted with the prokaryotic gene finder EasyGene [52] using a hidden Markov model (HMM) pre-trained on the B. anthracis genome. Sequence similarity searches, alignments and further annotation were performed using the BioPerl open software package with standard default BLAST X and ClustalW parameters (details available on request). The BLAST score ratio analysis (BSRA) was performed as described by Rasko et al. [53]. Protein analyses were done using Accelrys DS Gene. Proteome phylogeny analyses were performed using the Tree Building Method (Neighbor Joining, uncorrected) in Accelrys DS Gene. All figures included in this work were generated using the BioPython, GenomeDiagram and ReportLab open software packages. The complete sequence of pAW63 has been deposited in the GenBank database under the accession number DQ025752 [GenBank: DQ025752].
Authors' contributions
All authors contributed substantially to the study, and all have read and approved the final manuscript.
Acknowledgements
We are grateful to Lasse Smidt for his many contributions to this project, and to Daniel De Palmenaer for a fruitful collaboration on the analysis of the IS231L elements. We would like to thank Dr. Cheryl Kuske for sharing primer sequences and preliminary pAW63 sequence data prior to publication. We also acknowledge the contribution of the VIB Genetic Service Facility for the sequencing of the plasmid amplicons. This work was supported by the FRIA (Fonds pour la Formation à la Recherche dans l'Industrie et dans l'Agriculture) (grant to G. Van der Auwera), the Carlsberg Foundation (grant to L. Andrup), the National Fund for Scientific Research, and the Université catholique de Louvain.
==== Refs
Helgason E økstad OA Caugant DA Johansen HA Fouet A Mock M Hegna I Kolstø AB Bacillus anthracis, Bacillus cereus, and Bacillus thuringiensis--one species on the basis of genetic evidence Appl Environ Microbiol 2000 66 2627 2630 10831447 10.1128/AEM.66.6.2627-2630.2000
Schoeni JL Wong AC Bacillus cereus food poisoning and its toxins J Food Prot 2005 68 636 648 15771198
David DB Kirkby GR Noble BA Bacillus cereus endophthalmitis Br J Ophthalmol 1994 78 577 580 7918273
Mock M Fouet A Anthrax Annu Rev Microbiol 2001 55 647 671 11544370 10.1146/annurev.micro.55.1.647
Gill SS Cowles EA Pietrantonio PV The mode of action of Bacillus thuringiensis endotoxins Annu Rev Entomol 1992 37 615 636 1311541 10.1146/annurev.en.37.010192.003151
Fouet A Mock M Differential influence of the two Bacillus anthracis plasmids on regulation of virulence gene expression Infect Immun 1996 64 4928 4932 8945528
Hoton FM Andrup L Swiecicka I Mahillon J The cereulide genetic determinants of emetic Bacillus cereus are plasmid-borne Microbiology
Christie PJ Type IV secretion: the Agrobacterium VirB/D4 and related conjugation systems Biochim Biophys Acta 2004 1694 219 234 15546668 10.1016/j.bbamcr.2004.02.013
Berg T Firth N Apisiridej S Hettiaratchi A Leelaporn A Skurray RA Complete nucleotide sequence of pSK41: evolution of staphylococcal conjugative multiresistance plasmids J Bacteriol 1998 180 4350 4359 9721269
Morton TM Eaton DM Johnston JL Archer GL DNA sequence and units of transcription of the conjugative transfer gene complex (trs) of Staphylococcus aureus plasmid pGO1 J Bacteriol 1993 175 4436 4447 7687249
Schwarz FV Perreten V Teuber M Sequence of the 50-kb conjugative multiresistance plasmid pRE25 from Enterococcus faecalis RE25 Plasmid 2001 46 170 187 11735367 10.1006/plas.2001.1544
Dougherty BA Hill C Weidman JF Richardson DR Venter JC Ross RP Sequence and analysis of the 60 kb conjugative, bacteriocin-producing plasmid pMRC01 from Lactococcus lactis DPC3147 Mol Microbiol 1998 29 1029 1038 9767571 10.1046/j.1365-2958.1998.00988.x
Thompson JK Collins MA Completed sequence of plasmid pIP501 and origin of spontaneous deletion derivatives Plasmid 2003 50 28 35 12826055 10.1016/S0147-619X(03)00042-8
Grohmann E Muth G Espinosa M Conjugative plasmid transfer in gram-positive bacteria Microbiol Mol Biol Rev 2003 67 277 301, table of contents 12794193 10.1128/MMBR.67.2.277-301.2003
Wilcks A Jayaswal N Lereclus D Andrup L Characterization of plasmid pAW63, a second self-transmissible plasmid in Bacillus thuringiensis subsp. kurstaki HD73 Microbiology 1998 144 ( Pt 5) 1263 1270 9611801
Okinaka R Cloud K Hampton O Hoffmaster A Hill K Keim P Koehler T Lamke G Kumano S Manter D Martinez Y Ricke D Svensson R Jackson P Sequence, assembly and analysis of pX01 and pX02 J Appl Microbiol 1999 87 261 262 10475962 10.1046/j.1365-2672.1999.00883.x
Hernandez E Ramisse F Ducoureau JP Cruel T Cavallo JD Bacillus thuringiensis subsp. konkukian (serotype H34) superinfection: case report and experimental evidence of pathogenicity in immunosuppressed mice J Clin Microbiol 1998 36 2138 2139 9650985
Wilcks A Smidt L økstad OA Kolstø AB Mahillon J Andrup L Replication mechanism and sequence analysis of the replicon of pAW63, a conjugative plasmid from Bacillus thuringiensis J Bacteriol 1999 181 3193 3200 10322022
Riley M Systems for categorizing functions of gene products Curr Opin Struct Biol 1998 8 388 392 9666336 10.1016/S0959-440X(98)80074-2
Pannucci J Okinaka RT Williams E Sabin R Ticknor LO Kuske CR DNA sequence conservation between the Bacillus anthracis pXO2 plasmid and genomic sequence from closely related bacteria BMC Genomics 2002 3 34 12473162 10.1186/1471-2164-3-34
Rasko DA Altherr MR Han CS Ravel J Genomics of the Bacillus cereus group of organisms FEMS Microbiol Rev 2005 29 303 329 15808746 10.1016/j.femsre.2004.12.005
Francia MV Fujimoto S Tille P Weaver KE Clewell DB Replication of Enterococcus faecalis pheromone-responding plasmid pAD1: location of the minimal replicon and oriV site and RepA involvement in initiation of replication J Bacteriol 2004 186 5003 5016 15262938 10.1128/JB.186.15.5003-5016.2004
Hedberg PJ Leonard BA Ruhfel RE Dunny GM Identification and characterization of the genes of Enterococcus faecalis plasmid pCF10 involved in replication and in negative control of pheromone-inducible conjugation Plasmid 1996 35 46 57 8693026 10.1006/plas.1996.0005
Koetje EJ Hajdo-Milasinovic A Kiewiet R Bron S Tjalsma H A plasmid-borne Rap-Phr system of Bacillus subtilis can mediate cell-density controlled production of extracellular proteases Microbiology 2003 149 19 28 12576576 10.1099/mic.0.25737-0
Koraimann G Miller RV, Day MJ Bacterial Conjugation: cell-to-cell contact dependent horizontal gene spread Microbial Evolution: Gene establishment, survival and exchange 2004 Washington, D.C. , ASM Press 111 124
Yeo HJ Waksman G Unveiling molecular scaffolds of the type IV secretion system J Bacteriol 2004 186 1919 1926 15028675 10.1128/JB.186.7.1919-1926.2004
Llosa M Gomis-Ruth FX Coll M de la Cruz Fd F Bacterial conjugation: a two-step mechanism for DNA transport Mol Microbiol 2002 45 1 8 12100543 10.1046/j.1365-2958.2002.03014.x
Savvides SN Yeo HJ Beck MR Blaesing F Lurz R Lanka E Buhrdorf R Fischer W Haas R Waksman G VirB11 ATPases are dynamic hexameric assemblies: new insights into bacterial type IV secretion Embo J 2003 22 1969 1980 12727865 10.1093/emboj/cdg223
Mills DA Choi CK Dunny GM McKay LL Characterization of the conjugation system associated with the Lactococcus lactis ssp. lactis plasmid pRS01 Dev Biol Stand 1995 85 543 548 8586229
Clewell DB Jaworski DD Flannagan SE Zitzow LA Su YA The conjugative transposon Tn916 of Enterococcus faecalis: structural analysis and some key factors involved in movement Dev Biol Stand 1995 85 11 17 8586160
Kreft B Marre R Schramm U Wirth R Aggregation substance of Enterococcus faecalis mediates adhesion to cultured renal tubular cells Infect Immun 1992 60 25 30 1729187
Hill CW Sandt CH Vlazny DA Rhs elements of Escherichia coli: a family of genetic composites each encoding a large mosaic protein Mol Microbiol 1994 12 865 871 7934896
Emond E Dion E Walker SA Vedamuthu ER Kondo JK Moineau S AbiQ, an abortive infection mechanism from Lactococcus lactis Appl Environ Microbiol 1998 64 4748 4756 9835558
Belhocine K Plante I Cousineau B Conjugation mediates transfer of the Ll.LtrB group II intron between different bacterial species Mol Microbiol 2004 51 1459 1469 14982638 10.1111/j.1365-2958.2004.03923.x
Hacker J Kaper JB Pathogenicity islands and the evolution of microbes Annu Rev Microbiol 2000 54 641 679 11018140 10.1146/annurev.micro.54.1.641
Koehler TM Bacillus anthracis genetics and virulence gene regulation Curr Top Microbiol Immunol 2002 271 143 164 12224521
De Palmenaer D Vermeiren C Mahillon J IS231-MIC231 elements from Bacillus cereus sensu lato are modular Mol Microbiol 2004 53 457 467 15228527 10.1111/j.1365-2958.2004.04146.x
Mahillon J Rezsohazy R Hallet B Delcour J IS231 and other Bacillus thuringiensis transposable elements: a review Genetica 1994 93 13 26 7813910 10.1007/BF01435236
Arthur M Molinas C Depardieu F Courvalin P Characterization of Tn1546, a Tn3-related transposon conferring glycopeptide resistance by synthesis of depsipeptide peptidoglycan precursors in Enterococcus faecium BM4147 J Bacteriol 1993 175 117 127 8380148
Kurenbach B Bohn C Prabhu J Abudukerim M Szewzyk U Grohmann E Intergeneric transfer of the Enterococcus faecalis plasmid pIP501 to Escherichia coli and Streptomyces lividans and sequence analysis of its tra region Plasmid 2003 50 86 93 12826062 10.1016/S0147-619X(03)00044-1
Lambowitz AM Zimmerly S Mobile group II introns Annu Rev Genet 2004 38 1 35 15568970 10.1146/annurev.genet.38.072902.091600
Belhocine K Yam KK Cousineau B Conjugative transfer of the Lactococcus lactis chromosomal sex factor promotes dissemination of the Ll.LtrB group II intron J Bacteriol 2005 187 930 939 15659671 10.1128/JB.187.3.930-939.2005
Dai L Zimmerly S ORF-less and reverse-transcriptase-encoding group II introns in archaebacteria, with a pattern of homing into related group II intron ORFs Rna 2003 9 14 19 12554871 10.1261/rna.2126203
Toor N Hausner G Zimmerly S Coevolution of group II intron RNA structures with their intron-encoded reverse transcriptases Rna 2001 7 1142 1152 11497432 10.1017/S1355838201010251
Chen Y Braathen P Leonard C Mahillon J MIC231, a naturally occurring mobile insertion cassette from Bacillus cereus Mol Microbiol 1999 32 657 668 10320586 10.1046/j.1365-2958.1999.01388.x
Reddy A Battisti L Thorne CB Identification of self-transmissible plasmids in four Bacillus thuringiensis subspecies J Bacteriol 1987 169 5263 5270 3117773
Tinsley E Naqvi A Bourgogne A Koehler TM Khan SA Isolation of a minireplicon of the virulence plasmid pXO2 of Bacillus anthracis and characterization of the plasmid-encoded RepS replication protein J Bacteriol 2004 186 2717 2723 15090513 10.1128/JB.186.9.2717-2723.2004
Rasko DA Ravel J økstad OA Helgason E Cer RZ Jiang L Shores KA Fouts DE Tourasse NJ Angiuoli SV Kolonay J Nelson WC Kolstø AB Fraser CM Read TD The genome sequence of Bacillus cereus ATCC 10987 reveals metabolic adaptations and a large plasmid related to Bacillus anthracis pXO1 Nucleic Acids Res 2004 32 977 988 14960714 10.1093/nar/gkh258
Hoffmaster AR Ravel J Rasko DA Chapman GD Chute MD Marston CK De BK Sacchi CT Fitzgerald C Mayer LW Maiden MC Priest FG Barker M Jiang L Cer RZ Rilstone J Peterson SN Weyant RS Galloway DR Read TD Popovic T Fraser CM Identification of anthrax toxin genes in a Bacillus cereus associated with an illness resembling inhalation anthrax Proc Natl Acad Sci U S A 2004 101 8449 8454 15155910 10.1073/pnas.0402414101
Ehling-Schulz M Vukov N Schulz A Shaheen R Andersson M Martlbauer E Scherer S Identification and partial characterization of the nonribosomal peptide synthetase gene responsible for cereulide production in emetic Bacillus cereus Appl Environ Microbiol 2005 71 105 113 15640177 10.1128/AEM.71.1.105-113.2005
Van der Auwera G Mahillon J TnXO1, a germination-associated class II transposon from Bacillus anthracis Plasmid 2005 53 251 257 15848228 10.1016/j.plasmid.2004.08.004
Larsen TS Krogh A EasyGene--a prokaryotic gene finder that ranks ORFs by statistical significance BMC Bioinformatics 2003 4 21 12783628 10.1186/1471-2105-4-21
Rasko DA Myers GS Ravel J Visualization of comparative genomic analyses by BLAST score ratio BMC Bioinformatics 2005 6 2 15634352 10.1186/1471-2105-6-2
|
16042811
|
PMC1196294
|
CC BY
|
2021-01-04 16:32:48
|
no
|
BMC Genomics. 2005 Jul 26; 6:103
|
utf-8
|
BMC Genomics
| 2,005 |
10.1186/1471-2164-6-103
|
oa_comm
|
==== Front
Emerg Themes EpidemiolEmerging Themes in Epidemiology1742-7622BioMed Central London 1742-7622-2-71604276610.1186/1742-7622-2-7CommentaryAssessing influenza-related mortality: Comment on Zucs et al. Dushoff Jonathan [email protected] Princeton U. Dept. of Ecology and Evolutionary Biology, Princeton NJ, 08544, USA2 National Institutes of Health, Fogarty International Center, Bethesda MD, 20892, USA2005 21 7 2005 2 7 7 23 6 2005 21 7 2005 Copyright © 2005 Dushoff; licensee BioMed Central Ltd.2005Dushoff; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
==== Body
Influenza is an important source of mortality and morbidity, and an important public health priority. Measuring the health burden imposed by influenza viruses is an important, and still controversial, question. Some authors argue that influenza is directly or indirectly responsible for the majority of seasonal excess deaths in temperate countries [1], while others argue that they trigger only a small minority [2]. Retrospective cohort studies have shown a surprisingly large protective effect of influenza vaccination against deaths from any cause [3-5], and one author has provocatively suggested that increased influenza vaccination of the elderly could halve the total mortality rate [6]. The population-level interpretation of these cohort studies is not clear, however [7], and studies in the Netherlands [8] and Britain [9] have found substantially lower protection. The present contribution [10] is a welcome addition to the data base that can be used to address this important question. But much more remains to be done to standardize and improve methods, and to reconcile the results obtained from different approaches.
The impact of influenza is difficult to measure, because there is a great deal of influenza-like illness (ILI) in the world, caused by a large number of viruses, and only a very small percentage of cases is confirmed virologically [11]. Deaths triggered by influenza may be attributed to a number of final causes, including pneumonia, heart disease and stroke, and may occur weeks after initial infection [1,12,13].
The work of [10] is based on a long tradition of estimating influenza deaths by inference from seasonal patterns in death series. This method was pioneered by Serfling [14], and further developed by workers including Simonsen and colleagues [7,15]. While this work is valuable, and has produced apparently robust results, it is largely disconnected from quantitative virologic data about influenza. Thus, attribution of health effects to influenza is based strongly on assumptions about underlying death trends.
Recently, Thompson and colleagues have used virologic surveillance data to estimate influenza mortality [13] and hospitalizations [16], using a weekly, seasonal regression model. These models are the first to link quantitative virologic information to measures of influenza burden at the population level. The estimates produced are consistent with those of Serfling-like estimates [7,17] and are robust to the addition of estimates of respiratory syncytial virus (RSV) prevalence to the regressions. Questions remain about these estimates, however. The work of Thompson and colleagues removes a sinusoidal trend (fit at the same time as influenza prevalence), but does not take into account issues of autocorrelations; or the possibility of seasonal confounding between influenza prevalence, morbidity and mortality, and such factors as day length, temperature or school terms; or the likelihood that deaths caused by influenza infection in a given week may not occur until several weeks later.
Keatinge, Donaldson and colleagues [2,18], also used simple regression methods that ignored autocorrelations and seasonal confounding to study the causes of winter mortality in Europe. Unlike Thompson and colleagues, they lacked virological data and instead used proxies for influenza, but included temperature data, and found that temperature rather than influenza explained most of the excess deaths in their models.
Approaching a consensus on the health and mortality burden of influenza, and on the cause of winter excess mortality in general, is an important scientific and public-policy goal. For this to happen, further progess is needed in several areas.
• Employing virological data. When possible, analyses of influenza burden should be tied to estimates of laboratory-confirmed influenza cases. In some cases, such measures can be combined with ILI surveillance to improve estimates. Efforts should be made to increase the amount of viral surveillance information available in the public domain, with spatial and temporal break downs.
• More sophisticated statistical analyses. Time series methods that address issues of seasonal confounding and autocorrelation are available [19,20], but have been little used in analyses of seasonal mortality. Helfenstein [21] analyzed paired pre-whitened mortality series and inferred a "hidden relation" underlying heart disease in women and men. Much more needs to be done to investigate the relationship between mortality (or morbidity) and co-factors including weather, air pollution and epidemics of influenza and other viruses, while accounting for seasonal confounding and autocorrelations. Methods should evaluate multiple risk factors and consider the possibility of interactions between them.
• Discuss and define time scales. An important, and usually unasked, question in comparing results from different estimation approaches is the time scale on which influenza deaths are being measured. Everybody dies, so what is being measured as the mortality burden of influenza (or of weather) is deaths that are hastened by the cause in question. The question is whether these deaths are being hastened by weeks, months or years. Regressions that use a weekly time frame are expected to count deaths hastened by even a few weeks, while traditional methods of summing excess deaths over a season of 3 months or longer will be measuring at a different time scale.
• Quantitative spatial comparisons. As regional surveillance data, and data from different countries, become more available, analyses that explicitly incorporate risk factors and health outcome variables from various localities have the potential to greatly increase statistical power and shed light on unravelling the contribution of influenza and other risk factors to mortality and morbidity.
The contribution of influenza to morbidity and mortality – and, more broadly, cataloging the causes of daily and seasonal excess deaths and hospitalizations – remain as unresolved questions with important scientific and public-health implications. There is a pressing need for more communication between researchers studying different causes, places and time scales, and for application of appropriate, powerful statistical methods to these questions.
Competing interests
The author(s) declare that they have no competing interests.
==== Refs
Reichert TA Simonsen L Sharma A Pardo SA Fedson DS Miller MA Influenza and the winter increase in mortality in the United States, 1959–1999 Am J Epidemiol 2004 160 492 502 15321847 10.1093/aje/kwh227
Donaldson GC Keatinge WR Excess winter mortality: influenza or cold stress? Observational study B M J 2002 324 89 90 10.1016/S0022-2836(02)01040-9
Nichol KL Goodman M The health and economic benefits of influenza vaccination for healthy and at-risk persons aged 65 to 74 years Pharmacoeconomics 1999 16 63 71 10623378
Hak E Nordin J Wei FF Mullooly J Poblete S Strikas R Nichol KL Influence of high-risk medical conditions on the effectiveness of influenza vaccination among elderly members of 3 large managed-care organizations Clin Infect Dis 2002 35 370 377 12145718 10.1086/341403
Nichol KL Nordin J Mullooly J Lask R Fillbrandt K Iwane M Influenza vaccination and reduction in hospitalizations for cardiac disease and stroke among the elderly N Engl J Med 2003 348 1322 1332 12672859 10.1056/NEJMoa025028
Poland GA If you could halve the mortality rate, would you do it? Clin Infect Dis 2002 35 378 380 12145719 10.1086/341404
Simonsen L Reichert TA Viboud C Blackwelder WC Taylor RJ Miller MA Impact of Influenza Vaccination on Seasonal Mortality in the US Elderly Population Arch Intern Med 2005 165 265 72 15710788 10.1001/archinte.165.3.265
Voordouw ACG Sturkenboom MCJM Dieleman JP Stijnen T Smith DJ van der Lei J Stricker BHC Annual revaccination against influenza and mortality risk in community-dwelling elderly parsons JAMA 2004 292 2089 2095 15523069 10.1001/jama.292.17.2089
Mangtani P Cumberland P Hodgson CR Roberts JA Cutts FT Hall AJ A cohort study of the effectiveness of influenza vaccine in older people, performed using the United Kingdom General Practice Research Database J Infect Dis 2004 190 1 10 15195237 10.1086/421274
Phillip Zucs Udo Buchholz Walter Haas Helmut Uphoff Influenza associated excess mortality in Germany, 1985-2001 Emerg Themes Epidoemiol 2005 2 6 10.1186/1742-7622-2-6
Cox NJ Subbarao K Influenza Lancet 1999 354 1277 1282 10520648 10.1016/S0140-6736(99)01241-6
Glezen WP Payne AA Snyder DN Downs TD Mortality and influenza J Infect Dis 1982 146 313 321 7108280
Thompson WW Shay DK Weintraub E Brammer L Cox N Anderson LJ Fukuda K Mortality associated with influenza and respiratory syncytial virus in the United States JAMA 2003 289 179 186 12517228 10.1001/jama.289.2.179
Serfling RE Methods for current statistical analysis of excess pneumonia influenza deaths Public Health Rep 1963 78 494 506
Simonsen L Clarke MJ Williamson GD Stroup DF Arden NH Schonberger LB The impact of influenza epidemics on mortality: Introducing a severity index Am J Public Health 1997 87 1944 1950 9431281
Thompson WW Shay DK Weintraub E Brammer I Bridges CB Cox NJ Fukuda K Influenza-associated hospitalizations in the United States JAMA 2004 292 1333 1340 15367555 10.1001/jama.292.11.1333
Simonsen L Blackwelder WC Reichert TA Miller MA Letter: Estimating deaths due to influenza and respiratory syncytial virus JAMA 2003 289 2499 2500 12759317 10.1001/jama.289.19.2499-b
Keatinge WR Donaldson GC Bucher K Jendritsky G Cordioli E Martinelli M Dardanoni L Katsouyanni K Kunst AE Mackenbach JP McDonald C Nayha S Vuori I Cold exposure and winter mortality from ischaemic heart disease, cerebrovascular disease, respiratory disease, and all causes in warm and cold regions of Europe Lancet 1997 349 1341 1346 9149695 10.1016/S0140-6736(96)12338-2
Haugh LD Box GEP Identification of dynamic regression (distributed lag) models connecting 2 time series J Am Stat Assoc 1977 72 121 130
Diggle PJ Time Series: A Biostatistical Introduction 1990 Oxford University Press, Oxford
Helfenstein U Detecting hidden relations between time series of mortality rates Methods Inf Med 1990 29 57 60 2308527
|
16042766
|
PMC1196295
|
CC BY
|
2021-01-04 16:38:27
|
no
|
Emerg Themes Epidemiol. 2005 Jul 21; 2:7
|
utf-8
|
Emerg Themes Epidemiol
| 2,005 |
10.1186/1742-7622-2-7
|
oa_comm
|
==== Front
Harm Reduct JHarm Reduction Journal1477-7517BioMed Central London 1477-7517-2-111604279510.1186/1477-7517-2-11Case ReportA case report: Pavlovian conditioning as a risk factor of heroin 'overdose' death Gerevich József [email protected]ácskai Erika [email protected] Lajos [email protected] Zoltán [email protected] Addiction Research Institute, Budapest2 ELTE University, Faculty of Orthopedagogics, Budapest, Hungary3 National Institute of Psychiatry, Budapest, Hungary2005 25 7 2005 2 11 11 4 1 2005 25 7 2005 Copyright © 2005 Gerevich et al; licensee BioMed Central Ltd.2005Gerevich 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 authors present a case illustrating a mechanism leading directly to death which is not rare but has received little attention.
Case presentation
The case was evaluated by autopsy, investigation of morphine concentration in the blood, and clinical data. The heroin dose causing the 'overdose' death of a young man who had previously been treated a number of times for heroin addiction did not differ from his dose of the previous day taken in the accustomed circumstances. The accustomed dose taken in a strange environment caused fatal complications because the conditioned tolerance failed to operate. The concentration of morphine in the blood did not exceed the level measured during earlier treatment.
Conclusion
These results are in line with the data in the literature indicating that morphine concentrations measured in cases of drug-related death do not differ substantially from those measured in cases where the outcome is not fatal. A knowledge of the conditioning mechanism can contribute to prevention of fatal cases of a similar type. The harm reduction approach places great stress on preventive intervention based on data related to drug-related death.
==== Body
Background
A number of mechanisms leading directly to drug-related death are known. One of the most widely known variants is where the active substance content of a drug bought on the black market differs from the accustomed level [1]. Lethal development related to drug overdose occurs most frequently when the patient accustomed to the drug gives up its use then after a while attempts to continue addictive behaviour with the same dose used immediately before withdrawal [2]. The use of drugs in combination also increases the danger of a fatal overdose [3].
However, there is also another explanatory model of cases of drug-related death. Siegel et al. showed that situation-specific tolerance is capable of preventing the fatal consequence of a fatal-sized opiate overdose. When rats are given a large dose of morphine following morphine dosing in an environment substantially differing from the one in which they experienced the effects related to morphine, signs of overdose rapidly appear and in a few cases lead to the death of the rat. In contrast, in the case of rats where the morphine is dosed in the same circumstances the same size dose has a substantially smaller effect since the substance was given in the accustomed environment and so they were "expecting" its effect [4].
Siegel interviewed 10 heroin overdose survivors in an attempt to ascertain whether the overdoses occurred following novel pre-drug cues. For seven of the overdoses, the drug was administered in an environment not previously associated with drug use [5].
O'Brien showed the conditioned tolerance phenomenon in detoxicated heroin addicts in a double blind situation, on four different occasions. On one occasion the subjects were given a moderate dose (4 mg) of hydromorphon in an infusion without knowing what they were being given and when. On the second occasion they injected the same dose themselves. On the second two occasions the same process was repeated with salt. When they were given the opiate without prior indication, the subjects showed a significantly greater physiological reaction following the full effect of the drug than when they knew what they were receiving (since they injected it themselves). The anticipation and preparation for taking the drug triggers responses contrary to the drug effect in persons already showing drug tolerance. The anticipation preceding the administration of opiate, acting as a conditioned stimulus, reduced the action of the drug and so contributed to the development of a mechanism corresponding to tolerance [6].
Gutiérrez-Cebollada et al. interviewed 76 heroin addicts admitted to the emergency room of a university hospital in Barcelona. Fifty-four patients were admitted because of heroin overdose, and 22 were seeking urgent medical care for unrelated conditions, but their interview revealed intravenous heroin self-administration 1 hr or less before admission. All of the patients who had recently used heroin, but had not suffered an overdose, injected the drug in their usual drug-administration environment. In contrast, 52% of the overdose victims administered "in an unusual setting" [7].
The case described here is the first in the literature of addiction medicine where death can be quite clearly attributed to Pavlovian conditioning.
Case presentation
K.J., a 26-year-old male, first presented at the Drug Prevention and Treatment Centre with his wife in November 1997. They both asked to be treated for heroin addiction. Before admission they had been treated once as out-patients without success. He first used heroin a year later, in 1995, intravenously from the start, beginning with half a gram once a week; six months later his dose had increased to a gram a day. By then he was shooting up daily. He had never had any physical illness. Once he was hospitalized because of overdose, although opiate antagonist medication was not necessary. The concentration in the blood of morphine, the catabolite of heroin, was 0.05 mg/l. At the time of admission no internal medicine or neurological disorder could be found, while dysthymia and emotional lability were observed in the psychiatric state without psychotic symptoms or disorientation. Laboratory tests showed no abnormality. Detoxification with clonidine was followed by rapid relapse. He was never abstinent for longer than a week.
His wife recounted that on January 8, 1999, the day before his death, they had decided to begin withdrawal the following day. Next day, January 9, the wife remained at home and K.J. set out for work. What happened after that can be reconstructed from the forensic medical report and from information given by drug-using friends. On the way to work K.J. changed his mind and, breaking his promise to his wife, went to the dealer and bought a dose of heroin. He met other drug-using friends there who had bought heroin from the same dealer that day and later told the author that the heroin purchased then did not differ in quality from the usual. K.J. did not return home with the heroin purchased as he did on other occasions but went to the public toilet in the pedestrian underpass at the Népliget Metro station where he injected the same quantity (0.5 gram) that he had taken the previous day in the accustomed place, at home with his wife. The authorities called out were unable to help and pronounced him dead. A syringe half filled with a yellowish-brown fluid and a sooty spoon were found beside the body. The fluid in the syringe was heroin, while the metabolite of heroin, 6-0-acetylmorphine, and morphine-3-0-glucuronid were found in the blood and urine.
The autopsy found numerous traces of punctures by injection needles of various age on both upper limbs, the left side of the neck and the lower limbs. Traces of an infected but healing needle puncture were found inside the right elbow. Examination of the internal organs showed signs of general, very acute circulatory failure: acute congestive plethora of the organs, cerebral oedema, heightened brain pressure, cerebellar inclusion, acutely inflated lungs. The concentration in the blood of morphine, the catabolite of heroin, was 0.05 mg/l. The dose did not differ from the accustomed, daily dose. Other substances (alcohol, benzodiazepines, barbiturates) were not found. Heroin 'overdose' was given as the cause of death.
Conclusion
The fatal consequence of the heroin injection may have been caused by the failure in the action of conditioned tolerance. As the figure shows, when a conditioned place preference arises, the user has to take a bigger dose each time to achieve the same effect as the user who does not have the opportunity for secondary conditioning with environmental stimuli since he or she constantly changes the place where the drug is taken [6]. When the drug is taken in a strange environment the conditioned tolerance does not operate since the organism is not "expecting" the drug. The end result is that the otherwise accustomed dose leads to an overdose and thereby to death. This is why the term "overdose" is misleading since the quantity taken was not greater than other doses taken without fatal complications [8].
In this case it could be determined that the heroin used by the patients did not differ in composition from what they had been using earlier. A number of people bought the substance from the same dealer at the same time and subsequently reported that it had not caused them any problem. The concentration of morphine found in the blood was below the morphine values given in the literature in fatal cases; median level: 0.35 mg/l (range: 0.08–3.2 mg/l) [9,10]. This corresponds to the lower limit of morphine levels measured in current heroin users [9]. Probably the user died because he did not take the drug in the accustomed place and circumstances. In the strange, unaccustomed environment the conditioned tolerance described above reducing the effect of the drug action did not operate and a relative overdose resulted (Figure 1). The chance of possible contamination of the heroin powder by actual poisoning substances or infective agents is minimal, since none of those who bought heroin together with the patient had any toxic complications.
Figure 1 Heroin concentration levels in a case A after conditioning in an accustomed place (A1) and in a new place (A2), and in a case B without conditioning.
In his in-depth study of 99 fatal cases Ingold lists among the risk situations injection of drugs in public places where there was no way of testing the drugs beforehand [11]. This is confirmed by other research [7]. Australian authors have reached the same conclusion: deaths attributed to overdose are likely to have morphine levels no higher than those who survive, or heroin users who die from other causes [8].
The phenomenon of conditioned overdose death is of great significance for harm reduction. Users familiar with the concept of conditioned place preference could have greater chances of survival than those who are not aware of it. This is why there is a need for educational programmes as part of the treatment, making users receiving treatment aware of the nature and risks of conditioning. The more users are aware of the role played by conditioned cues in drug action and in relapse, the greater the chance that they will avoid fatal complications.
We doctors have a great responsibility in alerting the patients we treat to the dangers of conditioning.
Acknowledgements
This article is written in a framework of the Pygmalion Project (NKFP-ø5/052/2004).
==== Refs
Brecher EM Licit and illicit drugs The International Journal of the Addictions 1980 15 359 7380597
Gardner R Deaths in United Kingdom opioid users 1965–69 The Lancet 1970 ii 650 653 10.1016/S0140-6736(70)91414-5
Gerevich J Fatal combination of MDMA and heroin Psychosomatics 2005 46 189 15774960 10.1176/appi.psy.46.2.189
Siegel S Hinson RE Krank MS McCully J Heroin "overdose" death: Contribution of drug-associated environmental cues Science 1982 216 436 437 7200260
Siegel S Pavlovian conditioning and heroin overdose: Reports from overdose victims Bulletin of the Psychonomic Society 1984 22 428 430
O'Brien CP Childress AR McLellan AT Ehrman R O'Brien CP, Jaffe JH A learning model of addiction Addictive States 1992 Raven Press, New York 157 177
Gutiérrez-Cebollada J de la Torre R Ortuño J Garcés JM Camí J Psychotropic drug consumption and other factors associated with heroin overdose Drug and Alcohol Dependence 1994 35 169 174 7914483 10.1016/0376-8716(94)90124-4
Darke S Zador D Fatal heroin 'overdose': a review Addiction 1996 91 1765 1772 8997759 10.1046/j.1360-0443.1996.911217652.x
Darke S Sunjic S Zador D Prolov T A comparison of blood toxicology of heroin-related deaths and current heroin users in Sydney, Australia Drug and Alcohol Dependence 1997 47 45 53 9279497 10.1016/S0376-8716(97)00070-7
Darke S Ross J Fatal heroin overdoses resulting from non-injecting routes of administration, NSW, Australia Addiction 2000 95 569 573 10829332 10.1046/j.1360-0443.2000.9545698.x
Ingold FR Study of deaths related to drug abuse in France and Europe Bull Narc 1986 38 81 89 3779181
Gutierrez-Cebollada J de la Torre R Ortuno J Garces JM Cami J Psychotropic drug consumption and other factors associated with heroin overdose Drug Alcohol Depend 1994 35 169 174 7914483 10.1016/0376-8716(94)90124-4
|
16042795
|
PMC1196296
|
CC BY
|
2021-01-04 16:36:49
|
no
|
Harm Reduct J. 2005 Jul 25; 2:11
|
utf-8
|
Harm Reduct J
| 2,005 |
10.1186/1477-7517-2-11
|
oa_comm
|
==== Front
Health Qual Life OutcomesHealth Qual Life OutcomesHealth and Quality of Life Outcomes1477-7525BioMed Central 1477-7525-3-441604277510.1186/1477-7525-3-44ResearchMethodology and measurement properties of health-related quality of life instruments: A prospective study of patients undergoing breast reduction surgery Thoma Achilleas [email protected] Sheila [email protected] Karen [email protected] Eric [email protected] William [email protected] Department of Surgery, Division of Plastic and Reconstructive Surgery, St. Joseph's Healthcare, Hamilton, Ontario, Canada2 Surgical Outcomes Research Centre (SOURCE), McMaster University, Hamilton, Ontario, Canada3 Department of Psychiatry and Behavioural Neurosciences, McMaster University, Hamilton, Ontario, Canada4 Centre for Health Economics and Policy Analysis, Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada5 Health Utilities Inc, Dundas, Ontario, Canada2005 22 7 2005 3 44 44 24 6 2005 22 7 2005 Copyright ©2005 Thoma et al; licensee BioMed Central Ltd.2005Thoma et al; licensee BioMed Central Ltd.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.Background
Breast hypertrophy is associated with clinically important morbidity. A prospective study was conducted to assess the change in health-related quality of life (HRQL) following breast reduction mammoplasty. This paper describes the measurement properties of each of the HRQL questionnaires used.
Methods
The reliability, responsiveness, and the construct validity of each HRQL instrument (the Health Utilities Index Mark 2 (HUI2) and Mark 3 (HUI3) and the Breast Reduction Assessment Value and Outcomes (BRAVO) instruments) were assessed. The BRAVO instruments are a set of separate instruments including the Short Form-36 (SF-36), the Multidimensional Body Self Relations Questionnaire Appearance Assessment (MBSRQ-AS), and the Breast Related Symptoms Questionnaire (BRSQ).
Results
The HUI2, the HUI3, the MBSRQ-AS, and the breast severity symptom (BSS) score from the BRSQ all demonstrated good test-retest reliability. The SF-36 physical component summary, the MBSRQ-AS, and the BSS score demonstrated high responsiveness. The SF-36 mental component summary and the HUI3 had a moderate effect size and the HUI2 had a small effect size. All of the changes in scales are correlated in the same direction except for the SF-36 physical component summary and the SF-36 mental component summary.
Conclusion
All four instruments were found to be reliable and responsive. These instruments can be used in similar clinical settings to evaluate the change in patients' HRQL.
Breast Reduction SurgeryHealth-Related Quality of LifeReliabilityResponsivenessValidity
==== Body
Background
Within the last decade the plastic surgical community has been encouraged to use health-related quality of life (HRQL) assessment instruments to report on the efficacy of surgical interventions [1-5]. There is also an increased awareness of the impact of health and healthcare on the quality of human life such as a patient's ability to perform daily activities. Positive themes of happiness, social well-being, and emotional well-being need to be measured as these variables are particularly relevant to plastic surgery. Various HRQL instruments, generic and disease or condition specific, have been applied to plastic surgery research, especially in the area of breast hypertrophy and reduction mammoplasty [6-19]. Evidence from other clinical settings has shown that the generic instruments may be as efficient as the disease-specific ones [20-22]. A recommendation was made by Guyatt et al to include both a generic and a disease (condition) specific instrument in the evaluation of medical interventions [23].
Breast hypertrophy has been reported by patients to be associated with important burdens in pain and discomfort as well as emotion [7]. Earlier breast studies used a variety of study designs, instruments, and outcome measures [6-19]. These studies found that breast hypertrophy was associated with significant morbidity and reduced HRQL. They also found that after breast reduction mammoplasty patients had a substantial improvement in HRQL. Kerrigan et al found that patients with breast hypertrophy had lower health utility scores compared to controls without breast hypertrophy [6]. In a second report, Kerrigan et al found that patients with breast hypertrophy scored lower on the EuroQol; McGill Pain Questionnaire, Multidimensional Body Self Relations Questionnaire (MBSRQ), Short Form 36 (SF-36), and breast-related symptoms questionnaire (BRSQ) than the controls [7]. A recent prospective study found that pre-operatively mammoplasty patients scored lower on the SF-36 compared to normative data and there was an improvement in SF-36 scores from pre-operative to post-operative and these improvements were maintained to 12 months [13]. The improvements noted after the reduction mammoplasty remained stable at three years post-surgery [14]. In a cohort study, Collins et al found that pre-surgery patients scored significantly lower on the SF-36 than normative data and that following reduction mammoplasty patients improved from pre-surgery in all eight domains of the SF-36 [8]. Collins et al also found that post-surgery pain was lower and that the benefits from breast reduction were not associated with body weight, bra cup size, or weight of tissue resection [8].
In a recent Canadian prospective study of patients with a body mass index (BMI) below 27, pre-surgery mammoplasty patients scored lower on the SF-36 compared to normative data and post-surgery these patients achieved scores similar to normative data [18]. Although several publications have addressed HRQL in patients with breast hypertrophy, reduction mammoplasty remains a controversial surgery because of the denial of insurance coverage based on BMI in certain jurisdictions [18,19].
A number of different instruments have been used in previous studies to measure HRQL in patients with breast hypertrophy. In terms of the hierarchy of evidence in surgical studies, the studies which provide the higher strength of evidence are prospective cohort studies which address important patient outcomes. These studies have shown an improvement from pre-operative to post-operative, which have been statistically significant. Our study is similar to the design of some of the earlier prospective cohort studies measuring HRQL in patients with breast hypertrophy [8,10,13-15,18]. A recent study and discussion by Kerrigan et al stresses the importance of measuring HRQL and incorporating patient-reported health status into everyday practice [18,24]. The current study is the first to use the Health Utilities Index (HUI) as an outcome assessment [25-28]. This study is also the first prospective study to simultaneously assess the measurement properties of four HRQL instruments in breast reduction patients.
The primary objective of this study is to look at the measurement properties, including the reliability and responsiveness, of each of the four HRQL instruments used. The secondary objective was to assess the concurrent validity of each of the four HRQL instruments.
Methods
Patient eligibility and study design
Consecutive patients seen by the senior author (AT) over a period of one-year, with the diagnosis of breast hypertrophy and who obtained government approval for reduction mammoplasty were invited to participate in this prospective study. After signing an informed consent form, patients were asked to complete several questionnaires at each assessment time: (one week (time one) and one day before surgery (time two) and at one month (time three), six months (time four), and 12 months after surgery (time five)). The questionnaires were the HUI [25-28], and the Breast Reduction Assessment Value and Outcomes (BRAVO) instruments which consist of a set of separate instruments including the SF-36 [29], the MBSRQ-AS [30], and the BRSQ [7,24]. The one-week recall period was used for the HUI, the MBSRQ-AS, and the BRSQ and a four-week recall period was used for the SF-36.
The patients were provided with the questionnaires at their clinic visits and they either completed them while at the clinic or they completed them at home and returned them to the clinic by mail. The patients completed the questionnaires at one week before surgery and at one day before surgery to assess the test-retest reliability of each instrument. The questionnaires were completed at three post-operative time-points to measure change and to assess the stability of change over one-year of follow-up. The Research Ethics Board of McMaster University and St. Joseph's Hospital approved this study.
Clinical and demographic measures
In addition to completing the quality of life instruments (described in detail below), each patient underwent a physical examination and the baseline information was recorded. Demographic information including age, height, and weight was obtained which permitted the calculation of BMI (kg/m2). Other baseline information collected included self-reported bra cup size, diabetes, history of depression, smoking history, shoulder grooving, shoulder pain, back pain, neck pain, breast pain, intertrigo, and history of headaches.
Generic utility instruments: HUI
The HUI is a well-known health status and quality of life assessment instrument developed as an indirect method of measuring utilities (preferences) in clinical trials and other studies [25-28]. The HUI is a comprehensive, reliable, responsive, and valid multi-attribute utility instrument [25-28]. Responses to the questionnaire are converted using standard algorithms to levels of the Health Utilities Index Mark 2 (HUI2) and Mark 3 (HUI3) multi-attribute health status classification systems. The attribute levels are combined with published scoring functions to calculate utility scores of overall HRQL.
The HUI2 and HUI3 health status classification systems are complementary. Together they provide descriptive measures of ability or disability for health-state attributes, and descriptions of comprehensive health status [28]. The HUI2 is composed of seven attributes or dimensions which are sensation, mobility, emotion, cognition, self-care, pain, and fertility [25-28]. The HUI3 is composed of eight attributes or dimensions: vision, hearing, speech, ambulation, dexterity, emotion, cognition, and pain with five to six levels per attribute [25-28]. A seven-element vector describes the HUI2 comprehensive health state of a patient. Standard HUI questionnaires do not assess HUI2 fertility and, for the purposes of calculating overall HRQL, patients in this study were assumed to have no problems with their fertility. An eight-element vector, one level for each attribute (domain or dimension) of health, describes the HUI3 comprehensive health state for a patient or group of patients. The levels range from highly impaired to normal. For overall health status, the HUI2 and HUI3 utility scales of HRQL are defined such that dead = 0.00 and perfect health = 1.00. The HUI2 describes 24,000 unique health states and the HUI3 describes 972,000 unique health states that are obtained from factorials of the number of levels in each attribute.
Utilities derived from responses to HUI questionnaires may be used to calculate quality adjusted life years (QALYs). QALYs are the measure of effectiveness in cost-utility analysis, a special type of cost-effectiveness analysis for comparing alternative surgical interventions [25-28,31].
Generic health profile: SF-36
The SF-36 is a multi-purpose, short-form health survey with 36 questions [29]. It is a generic measure, as opposed to one that targets a specific age, disease, or treatment group. Accordingly, the SF-36 has proven useful in surveys of general and specific populations, comparing the relative burden of diseases, and in differentiating the health benefits produced by a wide range of different treatments [29]. The experience to date with the SF-36 has been documented in nearly 4,000 publications; citations for those published in 1988 through 2000 are documented in a bibliography covering the SF-36 and other instruments in the "SF" family of tools [29].
The SF-36 contains multi-function item scales to measure eight domains: physical function (10 items); role physical (4 items); bodily pain (2 items); general health (5 items); vitality (4 items); social functioning (2 items); role emotional (4 items); and mental health (5 items) [29]. The two summary measures of the SF-36 are the physical component summary and the mental component summary [29]. The scores for the multi-function item scales and the summary measures of the SF-36 vary from zero to 100, with 100 being the best possible score and zero being the lowest possible score [29].
Disease (condition) specific quality of life instruments: MBSRQ-AS and BRSQ
The MBSRQ is a well-validated self-report inventory for the assessment of body image [30]. Body image is conceived as one's attitudinal dispositions toward the physical self. As attitudes, these dispositions include evaluative, cognitive, and behavioral components. The physical self encompasses not only one's physical appearance but also the body's competence or fitness and its biological integrity or health/illness. The MBSRQ is a 69-item self-report inventory for the assessment of self-attitudinal aspects of the body-image construct [30]. The MBSRQ is intended for use with adults and adolescents over the age of 15 years [30]. Two forms of the MBSRQ are available, the full version and the MBSRQ-Appearance Scales (MBSRQ-AS). The full, 69-item version consists of seven factor subscales: 1) appearance evaluation, 2) appearance orientation, 3) fitness evaluation, 4) fitness orientation, 5) health evaluation, 6) health orientation, and 7) illness orientation [30]. There are also three multi-item subscales: 1) the body areas satisfaction scale (BASS), 2) the overweight pre-occupation scales, and 3) the self-classified weight scale [30].
In this study, the shorter version of the MBSRQ-AS was used and only the appearance evaluation subscale was used, because we were concerned with measuring body image. Scores vary from one to five. A high score indicates emphasis on one's looks, attention to one's appearance, and engaging in extensive grooming behaviours. A low score indicates apathy about one's appearance, one's looks are not especially important, and not expending much effort to "look good". High scorers feel mostly positive and satisfied with their appearance; low scorers have a general unhappiness with their physical appearance [30].
The BRSQ lists 13 breast related symptoms and the respondent indicates how much of the time she has the symptoms [7,24]. From this questionnaire, two scores are derived. The first score is the breast symptom summary score (BSS score), which is calculated by taking the mean scores of all 13 items. The BSS score varies from zero to 100, with a high score corresponding to fewer and less severe breast symptoms. For the second score, seven items of the 13-item scale are used to provide the physical symptom count. However, we did not tabulate the physical symptom count for this prospective study, as we were only interested in the overall BRSQ summary score (BSS score). The BRSQ has been validated and has demonstrated good test-retest reliability [7,8,24].
Scoring of the questionnaires
Scores for the HUI2, the HUI3, and the SF-36 were generated according to algorithms from the developers [32] and the SF-36® Health Survey Manual & Interpretation Guide, [33] respectively. The MBSRQ-AS and the BRSQ were scored according to the algorithm provided by Cash et al and Kerrigan et al, respectively [7,24,30].
Reliability and validity testing of the HRQL questionnaires
A measure is reliable if it is sound and dependable. Reliability is assessed by tests of repeatability or reproducibility. Reliability is often assessed in terms of agreement between intra-subject test-retest measurements and inter-assessor measurements [34]. There are various ways of assessing reliability of a measure [35]. These can be classified as inter-observer reliability (degree of agreement between different observers) and intra-observer or test-retest reliability (agreement between observations made by the same observer). An intraclass correlation coefficient (ICC) is used in this paper as a statistical measure of agreement for assessing test-retest reliability.
To estimate test-retest reliability, the same HRQL instrument is completed by the same patient on two different occasions. The assumption is that there would be no change in the scorers if there is no substantial change in health status of the patient being measured between the two occasions. The test-retest reliability of patients' responses is extremely important as we were most interested in determining that the difference in scores, between pre- and post-operative times reflected a real change in the patient's health is a result of the surgical intervention. If patient reporting is not reliable then one cannot truly capture the change in health status in patients using HRQL questionnaires.
The reliability of a test is indicated by the reliability coefficient. Reliability is expressed as a number ranging between zero and one; as it approaches zero there is lower reliability and a reliability coefficient close to one indicates higher reliability. In other words, the larger a reliability coefficient is, the more repeatable or reliable the test scores. General guidelines exist for interpreting reliability coefficients. A reliability coefficient value of 0.90 and greater is said to be excellent; a reliability coefficient value of 0.80 to 0.89 is good; a reliability coefficient value of 0.70 to 0.79 is adequate; and a reliability coefficient value below 0.70 may have limited applicability [36].
The validity and reliability of the HUI2, HUI3, and the SF-36 instruments have been demonstrated in various populations [25-29]. The MBSRQ has been validated and some reliability testing has been completed [30]. The BRSQ has been tested for face validity and has undergone test-retest reliability [7,24].
In this study we assessed the test-retest reliability of the HUI2, the HUI3, the MBSRQ-AS, and the BRSQ in patients diagnosed with breast hypertrophy prior to undergoing breast reduction mammoplasty. We did not assess the test-retest reliability of the SF-36 because we had used the four-week recall period for the SF-36. This study also provides some evidence about the concurrent validity of the BRSQ.
Responsiveness of the HRQL questionnaires
We used two generic and two disease (condition) specific instruments in this prospective study. Generic health status measures seek a broad perspective that is not specifically related to the restricted score of the HRQL of a specific disease or condition. Using a generic instrument has the advantage of allowing comparisons of health status to be made across different diseases and health states [37]. Disease (condition) specific measures focus on the disease or condition being studied, allowing greater sensitivity to intervention-related change compared to generic measures [37]. When deciding to use a generic instrument or a disease (condition) specific instrument to measure HRQL, it is important to consider the responsiveness of a HRQL instrument [37]. There are two major aspects of responsiveness, internal responsiveness and external responsiveness [38]. Internal responsiveness characterizes the ability of a measure to change over a pre-specified timeframe, whereas external responsiveness reflects the extent to which change in a measure relates to a corresponding change in a reference measure of clinical or health status [38]. This study focuses on internal responsiveness.
The effect size index is a statistical measure that can be used as an indicator of internal responsiveness. The mathematical formula for the effect size is the difference (Δ) of mean follow-up assessment scores minus mean baseline assessment score divided by the standard deviation of the baseline scores [39]. Our baseline was one-day before surgery and follow-up was six months after surgery. According to the well-known thresholds set by Cohen, an effect size of less than 0.20 can be considered trivial, an effect size between 0.20 and 0.50 can be considered small, an effect size between 0.50 and 0.80 can be considered moderate, and an effect size greater than 0.80 is considered large [40]. The standardized response mean (SRM) is the mean change scores divided by the standard deviation of the change scores [40].
Minimum Important Differences (MID)
The minimum important difference is a measure of clinically important or relevant change in health [37]. In other words, the minimum clinically important difference is the minimum level of change of an outcome measure that is considered to be clinically relevant. Drummond reported that differences of 0.03 or greater in mean utility scores were definitely clinically important [41]. This is supported by Grootendorst et al and Horsman et al, who reported that a difference in mean overall HUI scores of 0.03 or more should be considered as clinically important, and by Samsa et al who indicate minimal clinically important differences of HUI overall scores are between 0.02 to 0.04 [28,42,43]. Differences in mean HUI single-attribute utility scores of 0.05 or greater are considered clinically important [28].
There is no rule for determining what constitutes the minimum clinically important difference on the SF-36 subscales [14]. A 10-point change in scores has been suggested as a rule of thumb to apply on 100-point quality of life scales [44]. Minimum important differences have not been reported for the MBSRQ-AS and the BSS score.
Correlation analyses for assessing redundancy among instruments and concurrent validity of BSS score
Correlation analysis will provide information about the degree of redundancy from measurements using various instruments and evidence about the concurrent validity of the BSS score. Concurrent validity is a form of construct validity [35]. With concurrent validity, a new scale is correlated with another measure thought to be measuring the same construct and both are administered at the same time points [35].
In the current study, the change score of each questionnaire was correlated with the change score of the other questionnaires to assess the degree of redundancy among measures and to assess the concurrent validity of the BSS score. We expected all of the change scores to be positively correlated with each other because they are all scored in a positive direction, measuring improvement.
Statistical analyses
The patient characteristics were described using frequency distributions and means. The ICC of test-retest reliability was computed using data from one week prior to surgery (time one) and one day prior to surgery (time two) for each HRQL instrument named above. To measure responsiveness, effect size, and standardized response means [39] were calculated for each of the HRQL instruments (HUI2, HUI3, SF-36, MBSRQ-AS, and BRSQ) from one-day before surgery (time two) to six-months after surgery (time four). The Pearson correlation coefficient was calculated using the change score from baseline (one-day before surgery, time two) to six-months after surgery (time four) to assess concurrent validity among the HRQL instruments used in this study. The six-month follow-up was used in the above analyses because there was a higher completion rate than the 12-month follow-up. All statistical analyses were performed using the SPSS statistical software (version 13.01).
Results
Completion rates
Fifty-two consecutive patients initially consented to participate in the study. The first patient was enrolled in April 2001 and the last patient was enrolled in May 2002. Of the 52 patients who had initially agreed to participate, 49 patients completed the baseline assessment. Patients did not complete the study for various reasons. One patient could not sufficiently understand English to complete the questionnaires, another patient cancelled her surgery after it had been booked, and one patient decided not to participate. Although 49 patients completed the baseline assessment, some patients did not return their HRQL questionnaires at all time-points despite several telephone calls and mailings (Table 1).
Table 1 Number of patients who completed each HRQL instrument at each time point
Time Point HUI2 and HUI3 SF-36 MBSRQ-AS BSS Score
1 Week Pre-Op (Time 1) 48 48 47 49
1 Day Pre-Op (Time 2) 47 46 48 49
1 Month Post-Op (Time 3) 42 42 43 44
6 Months Post-Op Time 4 43 40 41 43
1 Year Post-Op Time 5 32 30 30 33
HUI2 = Health Utilities Index Mark 2; HUI3 = Health Utilities Index Mark 3; SF-36 = Short-Form 36; MBSRQ-AS = Multidimensional Body Self Relations Questionnaire Appearance Assessment; BSS Score = Breast Symptom Summary Score
Clinical and demographic information
The mean age of the patients was 38 years (minimum 20 years; maximum 68 years). The mean BMI was 30.9 kg/m2 (minimum 21.8 kg/m2; maximum 49.5 kg/m2). Self-reported bra cup sizes ranged from D to H, with 65 percent of the patients having a cup size of DD. Eighteen percent of patients had a history of depression, eight percent experienced frequent headaches, and 12 percent were smokers. Prior to surgery, all of the patients experienced neck pain, 94 percent experienced back pain, 53 percent experienced shoulder grooving, 45 percent experienced shoulder pain, 14 percent had breast pain, and 39 percent had intertrigo. The mean tissue resection weight for the left breast was 757.8 grams and the mean tissue resection weight for the right breast was 822.6 grams.
Test-Rest reliability
The computed ICC for the HUI2 was 0.86, the HUI3 was 0.84, the MBSRQ-AS was 0.85, and BSS score was 0.87. The HUI2, the HUI3, the BMSRQ-AS, and the BSS score all demonstrated good test-retest reliability.
Responsiveness
The responsiveness of each instrument is shown in Table 2. The SF-36 physical summary score, the MBSRQ-AS, and the BSS score had a large effect size, therefore, demonstrating high responsiveness. The SF-36 mental component summary and the HUI3 had a moderate effect size and the HUI2 had a small effect size. The SF-36 mental component summary, the HUI2, and the HUI3 had somewhat of a lower responsiveness than the other HRQL instruments used in this study. The standard response means for the measures are of the same magnitude as the effect size.
Table 2 Responsiveness of the HRQL instruments used in this study
Measure Difference SD at Baseline (1-day pre-op) SD ES1 SRM
HUI2 (n = 41) 0.06 0.14 0.14 0.45 0.46
HUI3 (n = 41) 0.12 0.19 0.17 0.63 0.67
SF-36 (Physical) (n = 37) 10.16 8.43 7.45 1.21 1.36
SF-36 (Mental) (n = 37) 7.46 11.75 12.63 0.63 0.59
MBSRQ-AS (n = 40) 0.86 0.65 0.70 1.32 1.23
BSS Score (n = 41) 45.05 13.15 13.74 3.43 3.28
HUI2 = Health Utilities Index Mark 2; HUI3 = Health Utilities Index Mark 3; SF-36 = Short-Form 36; MBSRQ-AS = Multidimensional Body Self Relations Questionnaire Appearance Assessment; BSS Score = Breast Symptom Summary Score; SD= Standard Deviation; ES1 = Effect Size (based on Cohen, 1988); SRM = Standardized Response Mean; SRM = Δ/SD(Δ)
ES1 = Δ / SD at baseline; Difference (Δ) = mean score at 6 month assessment minus mean score at baseline.
* n's reflect the number of patients who have completed the measure at both time-points (baseline and six months) and hence the difference in numbers from Table 1.
Minimally Important Differences (MID)
In the current study, the difference identified between the baseline (the day before surgery) and at six-months after surgery was 0.06 for the HUI2 which is twice the minimal important difference identified by Horseman et al [28] (Table 2). For the HUI3, the observed difference was four times the minimal important difference identified above (Table 2). We observed a 10 point increase in the SF-36 physical component summary, which is considered to be of clinical importance (Table 2) [14,44]. However, we did not observe a clinically important increase in the SF-36 mental component summary. The difference observed for the score of the MBSRQ-AS and the BSS score from baseline to six months after surgery was 0.86 and 45.05, respectively (Table 2). Since an effect size of two or more is considered statistically significant (based on the standardized response mean), we believe that this change is clinically important and should be further investigated in other populations.
Assessing redundancy among measures and concurrent validity of the BSS score
The Pearson's correlations between changes in pairs of HRQL scores are presented in Table 3. Five of the 15 correlations are statistically significant. The HUI2 and HUI3 scores are significantly positively correlated with each other as expected, but scores from HUI2 and HUI3 are not significantly correlated with scores from any other measures. The BSS scores are positively correlated with both SF-36 physical component summary and MBSRQ-AS scores. The MBSRQ-AS scores are positively correlated with the SF-36 mental component summary. The SF-36 physical component summary and the SF-36 mental component summary are negatively correlated. Moderate or better associations were observed for HUI2 emotion with SF-36 mental component summary (r = 0.489, p = 0.003) and MBSRQ-AS (r = 0.618, p < 0.001), for the HUI3 emotion with SF-36 mental component summary (r = 0.501, p = 0.002), and for HUI3 pain with SF-36 physical component summary (r = 0.412, p = 0.013).
Table 3 Correlations between changes in the HRQL scores
HUI3 SF-36 (Physical) SF-36 (Mental) MBSRQ-AS BSS Score
HUI2
Pearson Correlation 0.625** 0.135 0.295 0.317 0.221
p-value (2-tailed) <0.001 0.431 0.081 0.053 0.170
n 41 36 36 38 40
HUI3
Pearson Correlation 0.128 0.273 0.127 0.198
p-value (2-tailed) 0.458 0.107 0.446 0.222
n 36 36 38 40
SF-36 (Physical)
Pearson Correlation -0.515** -0.187 0.359*
p-value (2-tailed) 0.001 0.289 0.029
n 37 34 37
SF-36 (Mental)
Pearson Correlation 0.484** 0.147
p-value (2-tailed) 0.004 0.386
n 34 37
MBSRQ-AS
Pearson Correlation 0.481**
p-value (2-tailed) 0.002
N 40
n's reflect the number of patients who have completed the measure at both time-points (baseline and six months) and hence the difference in numbers from Table 1.
** Correlation is significant at the 0.01 level (2 tailed)
* Correlation is significant at the 0.05 level (2 tailed)
HUI2 = Health Utilities Index Mark 2; HUI3 = Health Utilities Index Mark 3; SF-36 = Short-Form 36; MBSRQ-AS = Multidimensional Body Self Relations Questionnaire Appearance Assessment; BSS Score = Breast Symptom Summary Score
Discussion
This study included patients with the diagnosis of breast hypertrophy who had obtained government approval for reduction mammoplasty. In our geographical area (Ontario, Canada), in contrast to other jurisdictions, for example, Nova Scotia, Canada [18] and the United States [19], the approval for provincial coverage for reduction mammoplasty is almost always granted if the patient has a bra cup size of D or larger and is experiencing physical symptoms.
A number of previous studies have reported that women who suffer from breast hypertrophy frequently present with heightened body image dissatisfaction [45-48]. In Canada, when plastic surgeons are faced with lawsuits, it is most commonly from breast surgery and when they are sued by patients following a breast reduction surgery it is usually due to the appearance of the breast or scarring [49]. Body image is conceived as one's attitudinal dispositions toward the physical self. As attitudes, these dispositions include evaluative, cognitive, and behavioral components. A study of the preoperative body image concerns of breast reduction patients found increased dissatisfaction with both their overall body image and breast size [46]. In response to their excessive breast size, patients reported extreme embarrassment in public areas and social settings and significant avoidance of physical activity [46]. Several previous studies on patients with breast hypertrophy have used the MBSRQ-AS to measure body image and have found that women with breast hypertrophy had low scores on the MBSRQ-AS suggesting dissatisfaction with their overall body image [7,8,24,46].
Patients completed the HUI2, the HUI3, and the BRAVO instruments (the SF-36, the MBSRQ-AS, and the BRSQ) at one week and one day before surgery to measure the test-retest reliability of each instrument and at one, six, and 12 months after surgery to measure change in HRQL following breast reduction mammoplasty. The methodology used in this prospective study may interest those who wish to sponsor, design, or implement future HRQL studies in breast reduction surgery or other areas of plastic surgery.
Of the 52 patients who had initially agreed to participate, 49 patients completed the baseline assessment. Despite multiple reminders, 30 patients completed all of the HRQL questionnaires at the 12-month follow up. This equates to a compliance rate of 57.7 percent. The response rate in this study is comparable to response rates obtained in previous studies on HRQL in patients with breast hypertrophy. For instance, several authors have reported response rates ranging from 32.5 percent to 80 percent [5,11,12,14,16]. For future studies, it may be helpful to understand why patients may not complete all of the requirements of a research study. The burden of completing multiple questionnaires may have limited our rate of compliance at one year. Patients who withdrew consent from one multi-centre trial reported interference with work, lack of time, complicated and cumbersome record keeping requirements, difficult study medicine regimens, and difficulty scheduling appointments due to a lack of flexibility on the part of the study personnel [50]. In the above study, the matched patients who completed all of their follow up reported that remuneration, commitment to finish, and the belief that the study was important motivated them to fully complete the study [50]. Based on existing guidelines for self-administered questionnaires, the questionnaires used in the present study exceeded the 12-page upper limit recommendation [51].
To measure the test-retest reliability of each instrument, scores were obtained for each instrument using the recommended algorithms and the ICC was computed from these scores. We found that all HRQL instruments demonstrated good reliability, which reinforces previous reliability testing of the HUI2, HUI3, SF-36, MBSRQ-AS, and BSS score.
It is extremely important that there is low within-patient variability in stable patients, relative to the magnitude of change that is predicted following the intervention, while answering the various questions on quality of life questionnaires in surgical outcome studies. Absence of reliable reporting will reduce the ability of measures to assess the effectiveness of surgery. For the present study, the one-week interval (time one and time two) was chosen to assess patient reporting as it was not long enough for other adverse events to intervene and change the health status but appropriate to avoid recall bias.
Marx et al noted that if multiple questionnaires were administered, each consisting of numerous items, the effect of memory may be minimized and the effect of memory may be greater if only a single questionnaire was used [52]. In the present study, four HRQL questionnaires were administered, each with multiple questions so the effect of a patient's memory is likely to be limited, therefore not biasing the responses.
The SF-36 physical component summary, the MBSRQ-AS, and the BSS score showed high responsiveness. The SF-36 mental component summary, the HUI2, and the HUI3 had a lower responsiveness summary statistics than the other HRQL instruments used in this study but all three instruments were able to detect clinically important changes in overall HRQL scores. The HUI3 showed a moderate effect size and detected a clinically important reduction in pain scores. All of the statistically significant correlations are positive except for the SF-36 physical component summary with the SF-36 mental component summary. The negative correlation may be a function of the problem with the algorithms for calculating SF-36 physical and mental component summary scores described in the published literature including reports by Simon et al [53] and Cunningham et al [54]. This study confirms evidence of concurrent validity for the BSS score as the change in BSS score is highly correlated with the SF-36 and other HRQL measures [19]. The HUI scores appear to provide unique information, as they were not correlated with the other measures. There were moderate or stronger correlations of HUI single-attribute utility scores, for emotion and pain, with the SF-36 and MBSRQ-AS.
This study demonstrates that patient reporting using the HUI2, the HUI3, the MBSRQ-AS, and the BSS score are reliable in a sample of patients diagnosed with breast hypertrophy who had breast reduction mammoplasty. All instruments were equally reliable. The HUI is the only preference-based instrument and it was shown to be responsive. The two disease (condition) specific instruments were the most responsive of all the HRQL instruments used
Having established the reliability and responsiveness of two generic (HUI and SF-36) instruments and two disease (condition) specific (MBSRQ-AS and BSS score) instruments, and the concurrent validity of BSS score, the focus moves onto the clinical and policy implications of the prospective study by addressing the following question: Can the improvement in HRQL derived from breast reduction surgery be measured quantitatively? Research is underway to address four specific issues: 1) identifying health attributes affected most frequently in breast hypertrophy patients and describing the extent of the observed morbidity; 2) assessing the health status and HRQL of patients in short, intermediate, and long time periods after reduction mammoplasty (i.e. one, six, and 12 months); 3) determining if there is a relationship between tissue resection weight and changes in health status and HRQL; and 4) determining if there is a relationship BMI and changes in health status and HRQL to address the ongoing BMI discrimination by third party payers.
List of abbreviations
BRAVO Breast Reduction Assessment Value and Outcomes
BRSQ Breast Related Symptoms Questionnaire
BSS Score Breast Symptom Summary Score
HRQL Health-Related Quality of Life
HUI Health Utilities Index
HUI2 Health Utilities Index Mark 2
HUI3 Health Utilities Index Mark 3
ICC Intraclass Correlation Coefficient
MBSRQ-AS Multidimensional Body Self Relations Questionnaire Appearance Assessment
MID Minimally Important Differences
SF-36 Short Form-36
SRM Standardized Response Mean
Authors' contributions
AT: Conception of the study, design of the study, acquisition of data, drafting of the manuscript. SS: Drafting of the manuscript. KV: Study coordination, acquisition of data, critical review of the manuscript. ED: Design of the study, statistical analysis, critical review of the manuscript. BF: Study design, critical review of the manuscript. All authors have read and approved the final manuscript.
Acknowledgements
This study could not have been done without support of the patients who completed the questionnaires. We would like to acknowledge Dr. Kevin O'Grady for assistance with data entry for this project. W. Furlong has a proprietary interest in Health Utilities Inc. which distributes copyright Health Utilities Index (HUI®) instrumentation and provides methodological advice on the use of HUI.
==== Refs
Keller RB Amadio PC Boland AL Bourne RB Heck DL Rudicel SA Swiontkowski MF Fundamentals of outcome research 1994 Committee on Outcome Studies. American Academy of Orthopaedic Surgeons (Department of Research)
Klassen A Jenkinson C Fitzpatrick R Goodacre T Measuring quality of life in cosmetic surgery patients with a condition-specific instrument: the Derriford Scale Br J Plast Surg 1998 51 380 384 10.1054/bjps.1997.0250 9771365
Cole RP Shakespeare V Shakespeare P Hobby JAE Measuring outcome in low priority plastic surgery patients using quality of life indices Br J Plast Surg 1994 47 117 121 10.1016/0007-1226(94)90170-8 8149054
Rankin M Borah GL Perry AW Wey PD Quality-of-life outcomes after cosmetic surgery Plast Reconstr Surg 1998 10 2139 2145 9811016
Klassen A Jenkinson C Fitzpatrick R Goodacre T Patients' health related quality of life before and after aesthetic surgery Br J Plast Surg 1996 49 433 438 10.1016/S0007-1226(96)90025-9 8983542
Kerrigan CL Collins ED Kneeland TS Voigtlaender D Moncur MM Matheney TH Grove MR Tosteson AN Measuring health state preferences in women with breast hypertrophy Plast Reconstr Surg 2000 106 280 288 10946925
Kerrigan CL Collins ED Striplin D Kim HM Wilkins E Cunningham B Lowery J The health burden of breast hypertrophy Plast Reconstr Surg 2001 108 1591 1599 10.1097/00006534-200111000-00024 11711933
Collins ED Kerrigan CL Lowery JS Striplin DT Cunningham B Wilkins EG The effectiveness of surgical and nonsurgical interventions in relieving the symptoms of macromastia Plast Reconstr Surg 2002 109 1556 1566 10.1097/00006534-200204150-00011 11932597
Chang WT Collins Ed Kerrigan CL An internet-based utility assessment of breast hypertrophy Plast Reconstr Surg 2001 108 370 377 10.1097/00006534-200108000-00014 11496177
Klassen A Fitzpatrick R Jenkinson C Goodacre T Should breast reduction surgery be rationed? A comparison of the health status of patients before and after treatment: postal questionnaire survey BMJ 1996 313 454 457 8776311
Shakespeare V Postle K A qualitative study of patients' views on the effects of breast-reduction surgery: a 2-year follow-up survey Br J Plast Surg 1999 52 198 204 10.1054/bjps.1999.3106 10474471
Tykka E Asko-Seljavaara S Hietanen H Patients' satisfaction with breast reconstruction and reduction mammoplasty Scand J Plast Reconstr Surg Hand Surg 2002 35 399 405 11878176
Blomqvist L Eriksson A Brandberg Y Reduction mammoplasty provides long-term improvement in health status and quality of life Plast Reconstr Surg 2000 106 991 997 10.1097/00006534-200010000-00005 11039369
Blomqvist L Brandberg Y Three-year follow-up on clinical symptoms and health-related quality of life after reduction mammoplasty Plast Reconstr Surg 2004 114 49 54 10.1097/01.PRS.0000127794.77267.14 15220567
Chao JD Memmel HC Redding JF Egan L Odom LC Casas LA Reduction mammoplasty is a functional operation improving quality of life in symptomatic women: a prospective single-centre breast reduction outcome study Plast Reconstr Surg 2002 110 1644 1652 10.1097/00006534-200212000-00003 12447043
Horch RE Jaeger K Stark GB Quality of life after breast reduction-mammoplasty Handchir Mikrochir Plast Chir 1999 31 137 142 10.1055/s-1999-13510 10337560
Jones SA Bain JR Review of data describing outcomes that are used to assess changes in quality of life after reduction mammoplasty Plast Reconstr Surg 2001 108 62 67 10.1097/00006534-200107000-00010 11420505
Miller BJ Morris SF Sigurdsom LL Bendor-Samuel RL Brennan M Davis G Paletz Prospective study of outcomes after reduction mammoplasty Plast Reconstr Surg 2005 115 1025 1031 10.1097/01.PRS.0000154212.10845.54 15793440
Kerrigan CL Discussion: Prospective study of outcomes after reduction mammoplasty Plast Reconstr Surg 2005 115 1032 1033 10.1097/01.PRS.0000156148.95908.28 15793441
Bombardier C Ware J Russell IJ Larson M Chalmers A Read JL Auranofin therapy and quality of life in patients with rheumatoid arthritis: Results of a multicentre trial Am J Med 1986 81 565 578 10.1016/0002-9343(86)90539-5 3532786
Liang MH Larson MG Cullen KE Schwartz JA Comparative measurement efficiency and sensitivity of five health status instruments for arthritis research Arthrit Rheumat 1985 28 542 547
Parkerson GR Connis RT Broadhead WE Patrick DL Taylor TR Tse CK Disease-specific versus generic measurement of health-related quality of life in insulin-dependent diabetic patients Med Care 1993 31 629 639 8326776
Guyatt GH Naylor CD Juniper E Heyland DK Jaeschke R Cook DJ Users' guides to the medical literature: XII. How to use articles about health-related quality of life JAMA 1997 277 1232 1237 10.1001/jama.277.15.1232 9103349
Kerrigan CL Collins D Kim HM Schnur PL Wilkins E Cunningham B Lowery J Reduction mammoplasty: Defining medical necessity Med Decis Making 2002 22 208 217 10.1177/02789X02022003003 12058778
Torrance GW Feeny DH Furlong WJ Barr RD Zhang Y Wang Q Multiattribute utility function for a comprehensive health status classification system: Health Utilities Index Mark 2 Med Care 1996 34 702 722 10.1097/00005650-199607000-00004 8676608
Furlong WJ Feeny DH Torrance GW Barr RD The Health Utilities Index (HUI) system for assessing health-related quality of life in clinical studies Ann Med 2001 33 375 384 11491197
Feeny D Furlong W Torrance GW Goldsmith CH Zhu Z DePauw S Denton M Boyle M Multiattribute and single-attribute utility functions for the Health Utilities Index Mark 3 system Med Care 2002 40 113 128 10.1097/00005650-200202000-00006 11802084
Horsman J Furlong B Feeny D Torrance G The Health Utilities Index (HUI): concepts, measurement properties, and applications Health Qual Life Outcomes 2003 1 54 10.1186/1477-7525-1-54 14613568
Ware JE JrSpilker B The SF-36 health survey Quality of Life and Pharmacoeconomics in Clinical Trials 1996 Second Philadelphia PA: Lippincott-Raven Press 337 345
Cash TF Pruzinsky T Body images: Development, deviance, and change 1990 New York: The Guilford Press
Thoma A Veltri K Haines T Goldsmith C O'Brien B Quartly C A methodological guide to performing a cost-utility study comparing surgical techniques Can J Plast Surg 2004 12 179 187
Furlong W Feeny D Torrance GW HUI Health Utilities Index Procedures Manual for Self-Administered Questionnaires Dundas ON 2001 Canada; Health Utilities Inc
Ware J Snow KK Kosinski M Gandek B SF-36 Health Survey: Manual and Interpretation Guide 1993 Boston, Mass., The Health Institute, New England Medical Center
Furlong W Barr R Feeny D Yandow S Patient-focused measures of functional health status and health-related quality of life in pediatric orthopedics: A case study in measurement selection Health Qual Life Outcomes 2005 3 3 10.1186/1477-7525-3-3 15647118
Streiner DL Norman RG Health Measurement Scales: A practical guide to their development and use 1995 2 Oxford Medical Publications, Oxford: New York, Oxford University Press (reprinted 1998).
U.S. Department of Labour Employment and Training Administration Understanding Test Quality-Concepts of Reliability and Validity, 1999 http://www.hr-guide.com/data/G362.htm
Middel B Statistical significant change versus relevant or important change in (quasi) experimental design: Some conceptual and methodological problems in estimating magnitude of intervention-related change in health service research Int J Integrated Care 2002 2 1 22
Husted JA Cook RJ Farewell VT Gladman DD Methods for assessing responsiveness: a critical review and recommendations J Clin Epidemiol 2000 53 459 468 10.1016/S0895-4356(99)00206-1 10812317
Cohen J Statistical power analysis for the behavioral sciences 1988 2 Hillsdale, NJ: Lawrence Earlbaum Associates
Cohen J Statistical power analysis for the behavioural sciences 1977 revised New York: Academic Press
Drummond M Introducing economic and quality of life measurements into clinical studies Ann Med 2001 33 344 349 11491193
Grootendorst P Feeny D Furlong W Health Utilities Index Mark 3 evidence of construct validity for stroke and arthritis in a population health survey Med Care 2000 38 290 299 10.1097/00005650-200003000-00006 10718354
Samsa G Edelman D Rothman M Williams GR Lipscomb J Matchar D Determining clinically important differences in health status measures: A general approach with illustrations to the Health Utilities Index Mark 2 Pharmacoeconomics 1999 15 141 155 10351188
King M The interpretation of scores form the EORCT quality of life questionnaire QCQ-C30 Qual Life Res 1996 5 555 567 10.1007/BF00439229 8993101
Glatt BS Sarwer DB O'Hara DE Hamori C Buckley LP LaRossa D A retrospective study of changes in physical symptoms and body image after reduction mammoplasty Plast Reconstr Surg 1999 103 76 82 9915166
Sarwer DB Bartlett SP Bucky LP LaRossa D Low DW Pertschuk MJ Wadden TA Whitaker LA Bigger is not always better: Body image dissatisfaction in breast reduction and breast augmentation patients Plast Reconstr Surg 1998 101 1956 1961 9623843
Faria FS Guthrie E Bradbury E Brain AN Psychological outcome and patient satisfaction following breast reduction surgery Br J Plast Surg 1999 52 448 458 10.1054/bjps.1999.3167 10673920
Guthrie E Bradbury E Davenport P Faria FS Psychological status of women requesting breast reduction surgery as compared to a control group of large-breast women J Psychomsom Res 1998 45 331 339 10.1016/S0022-3999(98)00002-6
Callaghan WB Professional liability of plastic surgeons in Canada Can J Plast Surg 1997 5 203 232
Janson SL Alioto ME Boushey HA Attrition and retention of ethnically diverse subjects in a multicenter randomized controlled research trial Control Clin Trials 2001 22 236S 243S 10.1016/S0197-2456(01)00171-4 11728627
Dillman DA Mail and Telephone Surveys: The Total Design Method New York Wiley and Sons 1978 10.1186/1477-7525-3-3
Marx RG Menezes A Horovitz L Jones EC Warren RF A comparison of two time intervals for test-retest reliability of health status instruments J Clin Epidemiol 2003 56 730 735 10.1016/S0895-4356(03)00084-2 12954464
Simon G Revicki D Grothaus L Vonkorff M SF-36 summary scores: Are physical and mental health truly distinct Med Care 1998 36 567 572 10.1097/00005650-199804000-00012 9544596
Cunningham WE Nakazono TT Tsai KL Hays RD Do differences in methods for constructing SF-36 physical and mental health summary measures change their associations with chronic medical conditions and utilization? Qual Life Res 2003 12 1029 1035 10.1023/A:1026191016380 14651420
|
16042775
|
PMC1196297
|
CC BY
|
2021-01-04 22:25:28
|
no
|
Health Qual Life Outcomes. 2005 Jul 22; 3:44
|
utf-8
|
Health Qual Life Outcomes
| 2,005 |
10.1186/1477-7525-3-44
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612000710.1371/journal.pmed.0020210Neglected DiseasesInfectious DiseasesPharmacology/Drug DiscoveryInfectious DiseasesMedicine in Developing CountriesDrugs and adverse drug reactionsInternational healthHealth PolicyDesigning Drugs for Parasitic Diseases of the Developing World Neglected DiseasesMcKerrow James H James H. McKerrow is the Robert E. Smith Professor of Experimental Pathology, and Director of the Sandler Center for Basic Research in Parasitic Diseases, University of California, San Francisco, United States of America. E-mail: [email protected]
Competing Interests: JHM is Director of the Sandler Center for Research on Parasitic Diseases.
8 2005 30 8 2005 2 8 e210Copyright: © 2005 James H. McKerrow.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.McKerrow outlines three new strategies, all originating within academic centers, that provide a new drug pipeline for treating parasitic diseases.
==== Body
The term “parasite” derives from a classical Greek word that was used to refer to “a guest who comes to dinner and doesn't leave” or “a class of priests who had meals at public expense” (Clitodemus, in Athenaeus Grammaticus, ca. 378 B.C.). In modern usage, it refers to eukaryotic organisms that range from single-cell protozoa to complex multicellular worms. The diseases caused by these organisms represent some of the world's greatest health problems.
Malaria, for example, currently affects about 500 million people and causes about 3,000 deaths a day—mostly in sub-Saharan Africa [1]. Schistosomiasis, caused by Schistosoma spp. (blood flukes), affects over 250 million people in the tropical world [2], and a recent meta-analysis showed that the disease is significantly associated with anemia, chronic pain, diarrhea, exercise intolerance, and undernutrition [3]. Almost a billion people are infected with the nematode parasites: Ascaris (roundworm), Ancylostoma (hookworm), and Trichuris (whipworm) [2]. Why then have so few effective drugs been produced against these diseases?
The answer lies primarily in the fact that these are diseases affecting people who are poor and living in poor regions of the world. They represent little or no viable market for the pharmaceutical industry, especially given the market requirements of the recently merged pharmaceutical giants. Given this reality, how can imaginative and effective strategies be developed to meet this challenge?
New Strategies for Antiparasitic Drug Design
At last year's meeting of the American Society of Tropical Medicine and Hygiene, the organization that hosts one of the largest international gatherings of basic scientists and tropical medicine specialists, several symposia highlighted efforts in antiparasitic drug design. What was five years ago a fairly dark vision of the future, now appears brighter.
Several nonprofit organizations are now operating that are dedicated to this unmet medical need, some collaborative interest exists in industry, and philanthropies have backed new initiatives. There are also a number of new academic consortia with novel strategies to address issues of target discovery and preclinical development.
Strategy one: Developing a drug that has both a commercial market in the west and an application against a neglected parasitic disease. This strategy is exemplified by the international consortium of researchers organized by Dr. Richard Tidwell of the University of North Carolina. With funding from the Bill and Melinda Gates Foundation, more than a dozen faculty and scientists from six research institutions (see Box 1) are working in collaboration with Immtech International to manage the “cross-over” development of antifungals produced by Immtech as potential drugs for African trypanosomiasis, also known as African sleeping sickness. This approach takes advantage of drug development of a class of compounds targeting a viable commercial market for Immtech (fungal diseases of the developed world). At the same time, academic- and nonprofit-institute scientists are targeting these compounds in parallel toward an “unprofitable” major health problem—African sleeping sickness. Specifically, an analog of the antifungal pentamidine has been optimized as a trypanocidal agent and is in clinical trials in Africa.
Box 1. International Consortium Working with Immtech International to Develop Drugs for African Sleeping Sickness
University of North Carolina–Chapel Hill
Georgia State University
London School of Hygiene and Tropical Medicine
Ohio State University
Swiss Tropical Institute
Kenya Trypanosomiasis Research Institute
Strategy two: An academic consortium gets federal support for pharmacokinetic and toxicology studies. This second strategy is exemplified by work under the leadership of Dr. Donald Krogstad and colleagues at Tulane University. Here, without an industrial partner, Krogstad and colleagues are using in-house chemistry and computational assistance to modify chloroquine analogs in an effort to overcome the problem of chloroquine resistance in malaria treatment. Their main partner in this effort is the National Institute of Allergy and Infectious Diseases, which has provided support for pharmacokinetic and toxicology studies by virtue of pre-existing contracts with SRI International (an independent nonprofit research-and-development organization) in Menlo Park, California, United States. At least one compound from this series is entering clinical trials.
Strategy three: An academic center uses a philanthropic gift to woo expertise from industry and build infrastructure for preclinical drug development. This third strategy is represented by the center that I direct, the Sandler Center for Basic Research in Parasitic Diseases at the University of California, San Francisco (UCSF). Here, philanthropic support was used to build infrastructure for a consortium of laboratories that mimics what might be found in a small- to medium-sized pharmaceutical company organized to carry out preclinical drug discovery and development. These core laboratories include computational support for drug discovery and chemical library selection, X-ray crystallography, drug metabolism and toxicology facilities, animal models of parasitic diseases, high-throughput screening, and synthetic chemistry.
Initial work at this center focused on a drug lead for Chagas disease and was supported by a Tropical Disease Research Unit Program project grant from the National Institute of Allergy and Infectious Diseases. This compound, now a drug candidate, was originally synthesized by Dr. Jim Palmer of Khepri Pharmaceuticals (now Celera) as part of an industry search for anti-inflammatory and anticancer drugs focusing on cysteine proteases. Validation of this compound as an antiparasitic was carried out by the UCSF team through rodent models of disease and, next, supported, as in the case of the Tulane consortium, by Dr. Chuck Litterst, Director of the National Institute of Allergy and Infectious Diseases Drug Development and Surveillance Group. The UCSF group initially handed over further development of a drug candidate to the Institute for OneWorld Health (iOWH), one of the nonprofit organizations that has sprung up to meet “downstream” drug development needs for neglected diseases (see Sidebar). iOWH facilitated tests of drug safety, minimum effective dose, and a formulation for large-scale Good Manufacturing Practice manufacture. The Drugs for Neglected Diseases Initiative (www.dndi.org) has now become the main partner for further development [4]. The Sandler Center consortium itself is now focusing on completing preclinical development of similar drug candidates targeting homologous enzymes in several other major global pathogens.
Institute for OneWorld Health
iOWH (www.oneworldhealth.org) is a nonprofit pharmaceutical company founded in 2000 in San Francisco whose mission is to develop safe, effective, and affordable new medicines for people with infectious diseases in the developing world. The institute carries basic research forward through drug development—it identifies drug leads, secures resources, and completes their development. For example, the organization recently received a grant of nearly US$10 million from the Bill and Melinda Gates Foundation to continue advancing its promising drug for visceral leishmaniasis, paromomycin, through the approval and post-approval process.
There is another way in which this third strategy represents a radical departure from the usual research paradigm of academic centers. A High-Throughput Screening Center has been established by the Sandler Center consortium, building upon the expertise of Drs. Kip Guy (now director of the center) and Janice Williams (now manager of the center), who brought an industrial perspective to the university. A recently completed screen used an assay that allowed robotic screening of a library of thousands of FDA-approved drugs and natural products for activity against Trypanosoma brucei, the parasite that causes African sleeping sickness. The “hit list” from this screen is being made available on a Web site (http://itsa.ucsf.edu/~schisto/fruit.html) for those organizations (such as the Drugs for Neglected Diseases Initiative [4], iOWH, and the World Health Organization) whose missions include licensing, manufacture, and distribution. Similar screens are being set up for leishmaniasis, Chagas disease, schistosomiasis, and malaria.
A New Drug Pipeline
The success of these three unusual strategies, all originating within academic centers, provides a new drug pipeline for parasitic diseases with no market value for the pharmaceutical industry (Figure 1). The future is likely to see a number of academic or academic–industrial collaborations supporting preclinical development in consortia like those described above.
Figure 1 The Pipeline of Drug Development
The success of the three strategies discussed in this article, all originating within academic centers, provides a new drug pipeline for parasitic diseases with no market value for the pharmaceutical industry.
DNDi, Drugs for Neglected Diseases Initiative; IMMTECH, Immtech International; MMV, Medicines for Malaria Venture; NIAID, National Institute of Allergy and Infectious Diseases; UNC, University of North Carolina–Chapel Hill.
This prediction was already borne out at the April 2005 meeting, “Drug Development for Diseases of Protozoa,” sponsored by the Keystone Symposia. In the short time since the American Society of Tropical Medicine and Hygiene meeting, it has become clear that more academic laboratories are encouraged to pursue drug discovery and development avenues beyond research traditionally thought of as academic. New consortia modeled after the three described above are now functional or in late planning stages at the University of Washington, Johns Hopkins University, Harvard University, and the University of Dundee. In addition, significant help from some large pharmaceutical companies, notably the Trés Cantos laboratory of GlaxoSmithKline, has fueled focused academic- and nonprofit-organization drug development efforts. Three nonprofit organizations (the Drugs for Neglected Diseases Initiative, iOWH, and the Medicines for Malaria Venture) are actively working to support various stages of the drug-development pipeline. Their efforts are particularly key to the “downstream” elements of drug manufacture and clinical trials.
Finally, philanthropic organizations such as the Bill and Melinda Gates Foundation, the Sandler Family Supporting Foundation, and the Ellison Foundation have contributed to these endeavors. Their continued support, and the recruitment of other interested parties, will be crucial to maintaining the momentum engendered by the three projects outlined here.
Editorial Note: PloS has received financial support in the form of a grant from the Sandler Family Supporting Foundation.
Citation: McKerrow JH (2005) Designing drugs for parasitic diseases of the developing world. PLoS Med 2(8): e210.
Abbreviations
iOWHInstitute for OneWorld Health
UCSFUniversity of California at San Francisco
==== Refs
References
World Health Organization, United Nations Children's Fund Africa malaria report 2003 2003 Available: http://www.rbm.who.int/amd2003/amr2003/amr_toc.htm . Accessed 11 May 2005
Colley DG LoVerde PT Savioli L Infectious disease. Medical helminthologyin the 21st century Science 2001 293 1437 1438 11520969
King CH Dickman K Tisch DJ Reassessment of the cost of chronic helminthic infection: A meta-analysis of disability-related outcomes in endemic schistosomiasis Lancet 2005 365 1561 1569 15866310
Pécoul B New drugs for neglected diseases: From pipeline to patients PLoS Med 2004 1 e6 10.1371/journal.pmed.0010006 15526054
|
16120007
|
PMC1196468
|
CC BY
|
2021-01-05 10:40:26
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e210
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020210
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612000810.1371/journal.pmed.0020229Learning ForumDiabetes/Endocrinology/MetabolismEndocrinologyA 69-Year-Old Female with Tiredness and a Persistent Tan Learning ForumPerros Petros Petros Perros is in the Endocrine Unit at Freeman Hospital, Newcastle upon Tyne, United Kingdom. E-mail: [email protected]
Competing Interests: The author declares that he has no competing interests.
8 2005 30 8 2005 2 8 e229Copyright: © 2005 Petros Perros.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.In this case-based learning article, Perros discusses the work- up a woman who presented with palpitations, tiredness, shortness of breath, and a persistent tan.
==== Body
DESCRIPTION OF CASE
A 69-year-old female presented with palpitations and a history of tiredness and shortness of breath for several weeks. She had a previous history of Raynaud syndrome. She was an ex-smoker. She commented that she had not lost her tan since the previous summer. Her only medication was nifedipine for her Raynaud syndrome.
On examination, she was slim and tanned. Pulse rate was 86 beats per minute and regular. Her blood pressure (BP) was 105/74 mm Hg. Her chest was hyperinflated. The rest of her examination was recorded as normal. A chest X ray showed no evidence of cardiac failure. Electrocardiogram monitoring showed episodes of atrial fibrillation. Her routine biochemistry was as follows: serum sodium, 132 mmol/l (normal range, 135–145 mmol/l); potassium, 5.1 mmol/l (3.4–5 mmmol/l); urea, 8.6 mmol/l (2.5–6.4 mmol/l); and creatinine, 110 mmol/l (65–120 μmol/l).
She was commenced on digoxin and warfarin. Her breathlessness gradually improved, and she remained in sinus rhythm.
How Would You Explain the Hyponatraemia, and What Additional Investigations Are Required?
Hyponatraemia is a common electrolyte abnormality in hospitalised patients [1]. The cause is often obvious (e.g., a clear history of fluid and electrolyte loss through vomiting or diarrhoea, or through use of thiazides or loop diuretics). A patient with hyponatraemia should be assessed by first taking a thorough history, focusing on gastrointestinal symptoms, fluid intake, thirst, postural dizziness, and medication. Evidence of intravascular volume depletion should be sought by examining skin turgour, the tongue, jugular venous pressure, and pulse rate, and most importantly, by measuring the BP in the supine and erect positions. A drop of BP by more than 20 mm Hg is indicative of intravascular volume depletion. Measurement of urinary sodium is the single most useful test. A urinary sodium concentration less than 20 mmol/l is indicative of volume depletion due to extrarenal causes (except in certain oedematous states). A urinary sodium concentration greater than 40 mmol/l suggests syndrome of inappropriate ADH secretion (SIADH), or a salt-losing nephropathy (diuretics, primary renal tubular diseases, or adrenal failure). The diagnosis of SIADH requires demonstration of normal thyroid and adrenocortical function, in the absence of intravascular volume depletion [2].
Progress
The patient was admitted into hospital two weeks after the initial presentation. Her main complaints were increasing lethargy and tiredness, reduced appetite, an episode of fainting, and weight loss. On examination, she was pigmented and thin (Figures 1 and 2). Her pulse rate was 76 beats per minute in sinus rhythm. BP was 77/59 mm Hg. The rest of the examination was normal.
Figure 1 The Patient's Facial Appearance at Presentation versus after Treatment
Facial appearance at presentation (left), showing pigmentation, and after treatment (right).
Figure 2 Photo of the Patient's Hand at Presentation, Showing Pigmentation in the Creases
What Is the Differential Diagnosis?
A history of malaise, anorexia, and weight loss in a 69-year-old patient is worrisome, and malignancy requires exclusion. She is an ex-smoker; therefore, lung cancer should be considered. SIADH can be a complication of lung cancer, and she has a history of hyponatraemia. Pigmentation may also occur in some lung cancers and other malignancies associated with ectopic adrenocorticotrophic hormone (ACTH) production. However, ectopic ACTH is usually associated with hypokalaemia and hypertension, which have not been features in this patient.
Lung, breast, gastrointestinal, and other malignancies can cause hypercalcaemia due to parathyroid hormone–related peptide secretion, and common symptoms of hypercalcaemia include anorexia, lethargy, and dehydration.
The patient has recently been started on digoxin, and toxicity from this drug can manifest as anorexia and malaise. The recent introduction of warfarin may also have led to occult blood loss, which could account for her malaise and hypotension.
Generalised pigmentation can be a manifestation of other systemic disorders. The patient is known to have Raynaud syndrome, and systemic sclerosis is associated with both Raynaud and pigmentation but is rare, and she has no other features, such as arthropathy or vasculitic lesions. Haemochromatosis can cause pigmentation and present with malaise due to liver failure, diabetes, or other rarer endocrinopathies, and predisposes a patient to hepatocellular carcinoma. Thyrotoxicosis and type 1 diabetes sometimes present in an atypical manner in patients of this age group, and need to be excluded. Primary adrenal failure could explain the patient's anorexia, malaise, pigmentation, low blood pressure, fainting, and serum electrolytes.
If you suspected primary adrenal failure, the specific symptoms that you would ask about are shown in Box 1.
Box 1. Common Symptoms of Primary Adrenal Failure
Tiredness
Anorexia
Nausea and vomiting
Abdominal pain
Weight loss
Postural dizziness
Pigmentation
Craving for salt
Weakness
Hypoglycaemic episodes
Investigations following the Second Hospital Admission
Following the second hospital admission, the chest X ray was unchanged from previous, with no masses. Other investigations had the following results: electrocardiogram, sinus rhythm, no digoxin effect; creatinine, 87 mmol/l (65–120 μmol/l); plasma glucose, 3.4 mmol/l; potassium, 5.4 mmol/l (3.4–5 mmol/l); sodium, 113 mmol/l (135–145 mmol/l); urea, 5.2 mmol/l (2.5–6.4 mmol/l); urine sodium, 45 mmol/l; serum thyroid-stimulating hormone, 7.87 mU/l (0.3–4.1 mU/l); serum free thyroxine, 18 pmol/l (11–23 pmol/l). Serum calcium, haemoglobin, and liver function tests were normal. Ultrasound scan of the abdomen was normal. Autoantibodies and extractable nuclear antigen Scl70 were negative; extractable-nuclear-antigen ribonucleic proteins were positive; thyroid microsomal antibodies were positive with a titre greater than 1/800; and antinuclear antibodies were negative.
How Would You Manage the Patient's Hyponatraemia?
This patient now has developed severe hyponatraemia and is at risk of developing neurological problems (deterioration in level of consciousness, and fits); therefore, treatment is required as a matter of urgency. Correct management depends on the cause. If the diagnosis is SIADH, the patient needs to be fluid restricted; if it is volume depletion, she will require intravenous fluid and electrolytes.
The low BP and high potassium are against a diagnosis of SIADH. It can be argued that salt and water depletion that is sufficient to cause severe hyponatraemia would be expected to be associated with a higher level of urea. Serum urea, however, is also dependent on the patient's muscle mass, which in this case was low. SIADH usually is associated with a urea level of less than 4 mmol/l. The high level of urine sodium is consistent with SIADH, but also with salt loss from the kidneys.
The diagnosis, therefore, is volume depletion due to renal losses, and normal saline should be administered intravenously.
The rapidity by which hyponatraemia is corrected is crucial. If the serum sodium is corrected too quickly and the hyponatraemia is chronic, there is a risk of central pontine myelinolysis [3]. Frequent monitoring of serum sodium, aiming for a rise of serum sodium of no more than 10 mmol/l/d, should be undertaken when the duration of hyponatraemia is chronic or unknown [2]. In patients where the hyponatraemia is known to have developed within the previous 2–3 days, it can be corrected fast, if there is an indication for doing so (coma or fits).
How Would You Explain the Other Laboratory Abnormalities?
Antibodies to extractable-nuclear-antigen ribonucleic protein are associated with mixed connective tissue disorder, but the specificity of the test is only 60%–75% [4]. The diagnosis of mixed connective tissue disorder requires additional clinical features that this patient did not have, although the positive antibodies to extractable-nuclear-antigen ribonucleic protein indicate that this patient has an autoimmune predisposition.
The patient's thyroid blood tests show a raised serum thyroid-stimulating hormone and normal free thyroxine, a condition often referred to as “subclinical” hypothyroidism. Her thyroid microsomal antibodies are also markedly positive and suggest an underlying autoimmune thyroiditis.
What Tests Are Required to Confirm Primary Adrenal Failure (Addison Disease)?
The biochemical diagnosis of Addison disease is made by measuring the serum cortisol concentration. Cortisol is a stress hormone with a prominent diurnal variation, consisting of an early morning peak and low levels in the evening and early part of the night. A single, random cortisol measurement, therefore, may not be diagnostic, unless the patient is physiologically stressed at the time (for example, patients presenting with an adrenal crisis can be diagnosed by a random cortisol measurement taken at the time of their illness and before steroid therapy is initiated). Hypotensive patients with an intact adrenal axis are expected to have a serum cortisol greater than 550 mmol/l; therefore, a cortisol concentration significantly lower than that, in such circumstances, is inappropriate. A blood sample should also be taken for plasma ACTH, which will indicate whether the hypoadrenalism is due to primary adrenal disease (high ACTH) or hypothalamic/pituitary disease (low ACTH). Measurement of the plasma renin activity and aldosterone concentrations is also helpful: in Addison disease plasma renin activity is high (because of the low intravascular volume) and plasma aldosterone is low (because of inability of the adrenal cortex to produce it).
In more stable or ambulant patients, a short synacthen test is usually diagnostic. The criterion for a normal synacthen test is a 30-minute cortisol value of more than 550 mmol/l [5]. The protocol for a short synacthen test is shown in Box 2.
Box 2. Protocol for Short Synacthen Test
Test is best performed in the morning, e.g., at 9:00 am.
Insert intravenous cannula.
Take basal cortisol, ACTH, renin, aldosterone, urea and electrolytes, and adrenal antibodies.
Short synacthen, 250 μg, is given as intravenous bolus at time zero.
Take samples at 0, 30, and 60 minutes for cortisol.
This patient had a short synacthen test, and the results were as follows: cortisol at baseline, 135 mmol/l; cortisol at 30 minutes, 144 mmol/l; baseline ACTH, 434 pg/l (0–47 pg/l); plasma renin activity, 8.9 ng/ml/h (1.1–4.1 ng/ml/h); and aldosterone, <50 pmol/l (220–430 pmol/l).
What Steroid Regimen Should the Patient Have at This Stage?
Patients in adrenal crisis or those who are hypotensive or cannot take medication orally should have parenteral glucocorticoid, in addition to saline and dextrose. In such a scenario, hydrocortisone is administered intravenously or intramuscularly by multiple bolus injections or by continuous infusion. A common regimen is 100–150 mg of hydrocortisone daily, until the patient's condition improves [6]. With such supraphysiological doses of glucocorticoids, no mineralocorticoid is required. If the patient is not acutely ill and is able to take steroids orally, hydrocortisone may be given initially at a dose of 40 mg in the morning and 20 mg in the evening. Once the patient's condition is improved (usually 3-4 days), the dose of hydrocortisone is reduced to 20–30 mg daily in two or three divided doses, with one-half to two-thirds of the dose taken on rising in the morning and the rest at noon and in the early evening. Mineralocorticoid is usually also introduced at this stage (fluodrocortisone 50–100 mg daily, as a single dose).
What Additional Investigations Are Required?
The cause of Addison disease needs to be identified. Autoimmune adrenalitis is the most common cause in the developed world, but infectious diseases like tuberculosis and fungal infection are common in some areas [7]. In an older age group, metastases to the adrenal glands are a possibility. Haemorrhage into the adrenals or infarction can occur in the context of meningococcal septicaemia, overanticoagulation, and antiphospholid syndrome.
In this case, adrenal antibodies were strongly positive and an ultrasound of the abdomen showed no abnormalities; therefore, more investigations were not undertaken. If imaging of the adrenals is specifically indicated, computerised tomography scanning is the best modality.
Should the Hypothyroidism Be Corrected by Adding Thyroxine?
Autoimmune hypoadrenalism and thyroid disease often coexist. It is imperative that the adrenal failure is treated first, as thyroxine replacement in an undiagnosed patient with Addison disease can precipitate a crisis [8]. Often, mild abnormalities in thyroid function (subclinical hypothyroidism) resolve after initiation of steroid therapy without thyroid replacement, as was the case in this patient.
DISCUSSION
Addison disease usually presents with nonspecific symptoms and should be considered in patients with unexplained malaise. Pigmentation, when present, is an important clue to the diagnosis. Hyponatraemia is almost invariable, though it may be mild and easily missed, as it is a common finding in hospitalised patients. Hyponatraemia merits investigation to reveal the underlying cause, as the treatments may be very different—fluid restriction if due to SIADH and fluid administration if due to fluid and electrolyte depletion. If the hyponatraemia is chronic, correction should be slow and closely supervised. Subclinical hypothyroidism may coexist with Addison disease. As thyroid disease is much more common than Addison disease and thyroid function tests are frequently requested, the patient's symptoms may be attributed to the thyroid dysfunction. Initiation of thyroxine therapy in such cases can precipitate an adrenal crisis, and physicians should think about Addison disease in patients whose symptoms deteriorate after thyroxine.
Key Learning Points
Think of Addison disease in patients complaining of tiredness and who look genuinely ill.
Check supine and erect BP in hyponatraemic cases.
The most useful biochemical test in assessing hyponatraemia is the urine sodium.
The diagnosis of SIADH requires exclusion of adrenal and thyroid failure.
An inappropriately low serum cortisol in a critically ill patient may be diagnostic of hypoadrenalism.
The definitive diagnostic test for Addison disease is the short synacthen test.
Once primary adrenal failure is diagnosed, look for the underlying cause.
Citation: Perros P (2005) A 69-year-old female with tiredness and a persistent tan. PLoS Med 2(8): e229.
The Learning Forum section editors are Susan Lightman and William Lynn.
Abbreviations
ACTHadrenocorticotrophic hormone
BPblood pressure
SIADHsyndrome of inappropriate ADH secretion
==== Refs
References
Smith DM McKenna K Thompson CJ Hyponatraemia Clin Endocrinol (Oxf) 2000 52 667 678 10848869
Baylis PH The syndrome of inappropriate antidiuretic hormone secretion Int J Biochem Cell Biol 2003 35 1495 1499 12824060
Lin SH Hsu YJ Chiu JS Chu SJ Davids MR Osmotic demyelination syndrome: A potentially avoidable disaster Q J Med 2003 96 935 947
Phan TG Wong RCW Adelstein S Autoantibodies to extractable nuclear antigens: Making detection and interpretation more meaningful Clin Diagn Lab Immunol 2002 9 1 7 11777822
Trainer PJ Besser GM Grossman A Primary adrenal failure Clinical endocrinology 1998 Oxford Blackwell Science 474 483
Lamberts SW Bruining HA de Jong FH Corticosteroid therapy in severe illness N Engl J Med 1997 337 1285 1292 9345079
Ten S New M Maclaren N Addison's disease J Clin Endocrinol Metab 2001 86 2909 2922 11443143
Murray JS Jayarajasingh R Perros P Lesson of the week: Deterioration of symptoms after start of thyroid hormone replacement BMJ 2001 323 332 333 11498494
|
16120008
|
PMC1196471
|
CC BY
|
2021-01-05 10:40:23
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e229
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020229
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612000910.1371/journal.pmed.0020230PerspectivesGenetics/Genomics/Gene TherapyGastroenterology/HepatologyGeneticsInflammatory Bowel DiseaseExpressing the Differences between Crohn Disease and Ulcerative Colitis PerspectivesWijmenga Cisca Cisca Wijmenga is Professor of Human Genetics, Complex Genetics Section, Department of Biomedical Genetics, University Medical Center Utrecht, Utrecht, the Netherlands. E-mail: [email protected]
Competing Interests: The author declares that no competing interests exist.
8 2005 30 8 2005 2 8 e230Copyright: © 2005 Cisca Wijmenga.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.
Dissection of the Inflammatory Bowel Disease Transcriptome Using Genome-Wide cDNA
Test Your Knowledge: Ten Questions on Inflammatory Bowel Disease
Dissecting Out Differences in the Transcriptomes of Inflammatory Bowel Disease
Wijmenga discusses the implications of a study that used a comprehensive gene expression approach to find new genes and pathways relevant to the pathophysiology of inflammatory bowel disease.
==== Body
Inflammatory bowel disease (IBD) is mainly seen in developed, urbanized countries, and its incidence increased steeply at the beginning of the 20th century, concurrently with improved hygiene. Similar trends have been observed for allergic and autoimmune disorders, suggesting that a reduction in microbial burden contributes to disease pathogenesis. Apart from environmental factors playing a strong role, genetic factors are also involved in IBD, so it is regarded as a complex disorder.
IBD can be classified as Crohn disease (CD) or ulcerative colitis (UC). Although these two forms of IBD share similar clinical and pathological features, the disease is heterogeneous, with marked differences in clinical presentation, underlying genetic factors, and response to treatment. The differentiation between CD and UC has been debated for a long time. Elucidation of the underlying disease pathway(s) may help to resolve this debate and also provide important leads for novel therapeutic targets. Recent genetic studies have been fairly successful in identifying disease genes for CD, in particular, pointing toward an impaired integrity of the epithelial barrier (reviewed in [1]). However, a broad view of the underlying pathogenic mechanism in IBD is expected to come from complementary approaches, including gene expression profiling using microarray studies.
Molecular Profiling to Identify the Molecular Mechanisms Underlying IBD
In a study published in PLoS Medicine, Schreiber and coworkers have used a comprehensive gene expression approach to find new genes and pathways relevant to the pathophysiology of IBD [2]. They took a close look at the genes differentially expressed in mucosal biopsies taken from the sigmoidal colons of ten individuals with UC, ten individuals with CD, and 11 control individuals (Figure 1). The genes were studied by hybridizing RNA from these biopsies to membranes spotted with some 23,000 unique transcripts from the human genome. One of the strengths of this study is the access to tissue involved in the disease process, although the biopsies represent a mixed population of cells.
Figure 1 Colonoscopy Images
Above are colonoscopy images from a healthy control patient with a noninflamed colon (left) and from patients with highly inflamed CD (middle) and UC (right). Below are resulting heat maps of differentially expressed genes identified for control patients versus CD patients (left) and control patients versus UC patients (right).
(Image: Costello et al. [2])
The authors initially examined active disease tissues (i.e., from untreated patients), which comes with the risk of mainly seeing the consequence of the disease rather than the cause. Their experimental design is extremely robust for obtaining small but significant differences in gene expression: ten measurements per gene were made for each sample, as the genes were spotted in duplicate and the experiments were repeated five times. The authors were able to identify genes with at least a 1.2-fold change between groups, differences much smaller than commonly reported for these types of studies. Hence, the number of genes differentially expressed was rather large: 378 genes unique to CD, 150 genes unique to UC, and 122 genes differentially expressed in both groups, compared to normal control tissue. Although both UC and CD share a general inflammation profile, these results strengthen earlier suggestions that CD and UC are, at the molecular level, two related yet different forms of chronic intestinal inflammation. Follow-up studies indicated that most of these differentially expressed genes may not be IBD-specific, but rather a consequence of colon inflammation.
Three other microarray studies on IBD have been published, but there is little overlap between the individual genes identified [3–5]. Although the experimental designs of these four studies differ significantly, it is interesting that the studies all point to the involvement of similar biological processes: immune-related processes, oncogenesis/cell proliferation/growth, and structure/permeability-related processes.
What Can Be Learned from These Studies?
Gene-expression profiling studies appear to hold much promise. In cancer studies, this promise has been met with the identification of many profiles that can be used to classify different tumor stages or to predict response to therapy. The gene-expression changes in tumor tissues are, in general, much more pronounced than those seen in other diseased tissues. A number of potentially interesting, new leads for therapeutic targets or disease diagnosis have come from Costello et al.'s study. Upregulation of cancer-related genes, such as TFF1 and the gene that encodes the Wnt signaling molecule CSNK1D, is very specific to the UC profile, and may point toward these genes playing a role in the increased risk of malignant transformation for patients with a long history of UC. Uthoff et al. also suggested a potential role for the Wnt signaling pathway in UC carcinogenesis in an earlier but much smaller study [6], but this needs to be further investigated. It is encouraging to learn that many of the genes now being identified confirm the ideas on disease pathogenesis that have developed recently as a result of genetic studies, namely, that there is impaired integrity of the epithelial barrier. Hence, Costello et al.'s study may further our understanding of the underlying disease mechanism.
Another interesting avenue for gene expression data is its use in selecting novel disease gene candidates. Costello et al. have identified 59 differentially expressed genes that map to IBD linkage regions. Genetic variants contributing to a complex disease like IBD are expected to be regulatory variants resulting in differential gene expression, rather than structural variants (amino acid substitutions). Although the majority of these 59 genes are more the consequence of the disease than the cause, current technology and recent accessibility to large numbers of single nucleotide polymorphisms make it feasible to test them for genetic association.
Looking Toward the Future
As gene expression studies mainly generate hypotheses, they provide a basis for further detailed gene function studies. However, the majority of genes included in these types of studies have an unknown function or a limited functional annotation at best, making it difficult to identify functional relationships between genes. This situation is expected to change with the ongoing large-scale functional genomics studies now being conducted. Nevertheless, genes with an unknown function can still be excellent candidate genes for testing by genetic association. Since expression studies mainly highlight the perturbed pathways involved, and genetic studies point to critical players in disease pathways, it is obvious that much can be gained from integrating the two. As Schreiber's research group was instrumental in identifying DLG5 as one of the IBD genes, this group is in an excellent position to perform both types of studies [7].
A future application might be to use disease-specific signatures as predictive diagnostic tools for distinguishing CD from UC. Since mucosal biopsies are invasive and burdensome for the patients, it would be interesting to determine whether similar patterns of expression can also be obtained from peripheral blood.
Citation: Wijmenga C (2005) Expressing the differences between Crohn disease and ulcerative colitis. PLoS Med 2(8): e230.
Abbreviations
CDCrohn disease
IBDinflammatory bowel disease
UCulcerative colitis
==== Refs
References
Schreiber S Rosenstiel P Albrecht M Hampe J Krawczak M Genetics of Crohn disease, an archetypal inflammatory barrier disease Nat Rev Genet 2005 6 376 388 15861209
Costello CM Mah N Hasler R Rosenstiel P Waetzig GH Dissection of the inflammatory bowel disease transcriptome using genome-wide cDNA microarrays identifies novel candidate disease genes PLoS Med 2005 2 e199 10.1371/journal.pmed.0020199 16107186
Dieckgraefe BK Stenson WF Korzenik JR Swanson PE Harrington CA Analysis of mucosal gene expression in inflammatory bowel disease by parallel oligonucleotide arrays Physiol Genomics 2000 4 1 11 11074008
Lawrance IC Fiocchi C Chakravarti S Ulcerative colitis and Crohn's disease: Distinctive gene expression profiles and novel susceptibility candidate genes Hum Mol Genet 2001 10 445 456 11181568
Dooley TP Curto EV Reddy SP Davis RL Lambert GW Regulation of gene expression in inflammatory bowel disease and correlation with IBD drugs: Screening by DNA microarrays Inflamm Bowel Dis 2004 10 1 14 15058520
Uthoff SM Eichenberger MR Lewis RK Fox MP Hamilton CJ Identification of candidate genes in ulcerative colitis and Crohn's disease using cDNA array technology Int J Oncol 2001 19 803 810 11562759
Stoll M Corneliussen B Costello CM Waetzig GH Mellgard B Genetic variation in DLG5 is associated with inflammatory bowel disease Nat Genet 2004 36 476 480 15107852
|
16120009
|
PMC1196472
|
CC BY
|
2021-01-05 10:40:26
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e230
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020230
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612001010.1371/journal.pmed.0020231PerspectivesOtherGeriatricsNutritionNutrition and MetabolismGeriatric MedicineCalorie Restriction Extends Life Span— But Which Calories? PerspectivesHeilbronn Leonie K Ravussin Eric *Eric Ravussin is director of the Health and Performance Enhancement Division at Pennington Biomedical Research Centre in Baton Rouge, Louisiana, United States of America, and Leonie K. Heilbronn is a NHMRC, Peter Doherty Australian Biomedical Fellow at the Garvan Institute in Sydney, Australia.
Competing Interests: The authors are conducting the first randomized clinical trial of caloric restriction in non-obese humans, funded by the National Institute on Aging (Bethesda, Maryland, United States).
*To whom correspondence should be addressed. E-mail: [email protected] 2005 30 8 2005 2 8 e231Copyright: © 2005 Heilbronn and Ravussin.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.Heilbronn and Ravussin discuss the findings and implications of a study in PLoS Biology of caloric restriction in fruit flies. Are there implications for humans?
==== Body
Evidence that calorie restriction (CR) retards aging and extends median and maximal life span was first described in the 1930s by McCay et al. [1]. Since then, similar observations have been made in a variety of species including rodents, fish, fruit flies, worms, and yeast [2]—and although they are not yet definitive, results from ongoing longevity studies in monkeys suggest that CR will also extend life span in longer-lived species [3,4].
There are many theories explaining the mechanisms by which CR extends life span. An early hypothesis was that delayed sexual maturation might be a mechanism. However, it has since been shown that CR initiated in older animals also increases life span [5]. Reduced metabolic rate—with consequent reduction in free radical production—was another early, leading hypothesis to explain the anti-aging effects of CR. But there are many other metabolic effects that have been associated with CR, including altered insulin sensitivity and signaling, stress resistance, altered neuroendocrine function, and changes in nutrient signaling. Any, or a combination, of these biological changes may retard aging. However, recent studies seem to favor a highly conserved stress response that evolved early in most species to increase an organism's chance of surviving adversity (such as CR) by triggering concerted physiological responses [6].
Nutrient Composition of Calorie Restricted Diets
Previous studies in rodents have shown that the effects of CR on extending life span are dose dependent, with a 20% reduction in calorie intake producing a smaller increment in life span as compared to a 40% reduction in food intake [7]. From this work and others, the reduction in calories was recognized to be of paramount importance in the longevity response, and alterations in the nutrient content of diets were considered irrelevant. However, a recent study in PLoS Biology by Mair et al. challenges this long-held concept of CR [8].
From Fruit Flies to Rodents
In the study by Mair et al., the authors examined life span in fruit flies (Drosophila melanogaster) fed one of four different diets: (1) a combination of yeast and sugar (control), (2) restricted in yeast only, (3) restricted in sugar only, and (4) restricted in yeast and sugar. The authors observed that in the restricted sugar group, as compared to the controls, maximal life span was unchanged and median life span was increased by only 12%. On the other hand, both maximal and median life spans were increased substantially in the restricted yeast group and in the restricted yeast and sugar group (Figure 1). Importantly, the authors claim that total calorie contents of the restricted sugar and restricted yeast diets were similar. Thus, from this study it can be implied that restricting carbohydrate is less advantageous than restricting protein/lipid for mediating the effects of dietary restriction (DR) on life span.
Figure 1 Plot of Median Life Span of Female Drosophila against the Estimated Caloric Content of the Food Medium
(A) and (B) represent independent repeats. Red arrows link pairs of food types where differences in caloric content are due to different yeast concentrations. Blue arrows link pairs of food types where differences in caloric content are due to different sugar concentrations. Green arrow links food types where differences in caloric content are due to both different sugar and yeast concentrations. Life span is extended to a greater extent per calorie by reducing yeast concentration from control to DR levels than by reducing sugar. This is in contrast to what would be predicted if calorie intake were the key mediator of life-span extension by DR in fruit flies.
(Figure from [8])
It is of concern that the authors did not directly measure the flies' total food intake but only estimated intake by examining their feeding behavior. This method may not take into account possible differences in the rate of food uptake of restricted flies, which could affect the results.
Investigators have undertaken studies of prolonged calorie restriction in humans
Only a handful of studies have investigated the role of altering nutrient composition on longevity in rodents. The results from these studies have been contradictory, with some studies showing no life-span extension following restriction of fat only [9] and others showing increased life span following replacement of casein-protein for soy-protein [10]. However, the concept that it is not just reduced calorie intake that drives the life-span extension effect is not new, and the timing of food intake has also been proposed to be of importance. For example, when rodents are fed every other day, improvements in biomarkers of aging and increased life span are observed, even though measured calorie intake and body weight were not statistically different from ad-libitum or pair-fed animals [11,12]. These responses were dependent on genotype and the age at which the protocol was implemented. Furthermore, animals fed every other day had a better response to neurotoxic stressors as compared to animals maintained on prolonged CR.
The mechanisms behind the differences in restricting carbohydrate vs. protein/lipid on life-span extension were not examined in the Mair et al. study [8]. Such mechanisms obviously imply the existence of molecular systems in cells that sense macronutrients—systems that may respond not only to nutrient availability but also to the hormonal response elicited by these dietary nutrients [13]. For example, restricting dietary carbohydrates increases the plasma concentration of B-hydroxybutyrate (that is, ketogenesis), a shift that may counteract life-span extension in mammals. However, rat models of Alzheimer and Parkinson diseases fed a ketogenic diet exhibit increased resistance to seizures and have increased protection of neurons [14]. Ketogenic diets are also prescribed to patients with epilepsy, and although there have been no randomized controlled trials, large observational studies (some prospective) suggest that this diet does have a beneficial effect on seizures [15].
However, it is likely that carbohydrate and protein may differentially alter nutrient-sensing pathways such as Sir2 and mammalian target of rapomyocin (mTOR), which are gaining acceptance as mediators of the life-span extension effects of CR [16]. Sir2 (the mammalian homolog is SIRT1) is a nicotinamide-adenine-dinucleotide-dependent histone deacetylase that interacts with numerous transcription factors to silence gene transcription. Sir2 is upregulated by CR and is required for life-span extension effects of CR in Caenorhabditis elegans (reviewed in [6]). mTOR is a serine/threonine kinase that is activated by insulin, nutrients, and growth factors and is a central regulator of ribosome biogenesis, protein synthesis, and cell growth. Inhibition of mTOR increases life span in Drosophila and C. elegans (reviewed in [16]).
Are There Implications for Human Life Span?
Obviously, invertebrate organisms cannot serve as reliable models for human longevity, and the results by Mair et al. [8] should not be extrapolated to mammals in general. But if this result could be replicated in humans, then the prospect of DR to increase life span would be eminently more attractive than overall CR. This would mean that a change in food patterns could have a similar effect to the dramatic reduction of total food intake. However, the life-span extension effects of CR have not been proven in humans, and the jury is still out on whether nutrient composition will even affect life span in non-human primates.
In close collaboration with the National Institute on Aging, investigators in Baton Rouge, Boston, and Saint Louis (all in the United States) have undertaken studies of prolonged CR in humans. These studies aim to test the feasibility and safety of different types of calorie restricted diets in non-obese people and to determine the effects of CR on risk factors for age-related diseases, psychological factors, immune function, oxidative stress, and molecular pathways identified in lower species [17]. These kinds of studies will further help identify the mechanisms underpinning the effect of CR or DR on longevity.
Citation: Heilbronn LK, Ravussin E (2005) Calorie restriction extends life span—But which calories? PLoS Med 2(8): e231.
Abbreviations
CRcalorie restriction
DRdietary restriction
mTORmammalian target of rapomyocin
==== Refs
References
McCay CM Crowel MF Maynard LA The effect of retarded growth upon the length of the life span and upon the ultimate body size J Nutr 1935 10 63 79
Weindruch R Walford RL The retardation of aging and disease by dietary restriction 1988 Springfield (Illinois) Charles C. Thomas 436
Bodkin NL Alexander TM Ortmeyer HK Johnson E Hansen BC Mortality and morbidity in laboratory-maintained Rhesus monkeys and effects of long-term dietary restriction J Gerontol A Biol Sci Med Sci 2003 58 212 219 12634286
Lane MA Black A Ingram DK Roth GS Calorie restriction in non-human primates: implications for age-related disease risk J Anti-Aging Med 1998 1 315 326
Weindruch R Walford RL Dietary restriction in mice beginning at 1 year of age: Effect on life-span and spontaneous cancer incidence Science 1982 215 1415 1418 7063854
Sinclair DA Toward a unified theory of caloric restriction and longevity regulation Mech Ageing Dev 2005 E-pub ahead of print
Weindruch R Walford RL Fligiel S Guthrie D The retardation of aging in mice by dietary restriction: Longevity, cancer, immunity and lifetime energy intake J Nutr 1986 116 641 654 3958810
Mair W Piper MDW Partridge L Calories do not explain extension of life span by dietary restriction in Drosophila
PLoS Biol 2005 3 e223 10.1371/journal.pbio.0030223 16000018
Iwasaki K Gleiser CA Masoro EJ McMahan CA Seo EJ Influence of the restriction of individual dietary components on longevity and age-related disease of Fischer rats: The fat component and the mineral component J Gerontol 1988 43 B13 B21 3335742
Iwasaki K Gleiser CA Masoro EJ McMahan CA Seo EJ The influence of dietary protein source on longevity and age-related disease processes of Fischer rats J Gerontol 1988 43 B5 B12 3335746
Anson RM Guo Z de Cabo R Iyun T Rios M Intermittent fasting dissociates beneficial effects of dietary restriction on glucose metabolism and neuronal resistance to injury from calorie intake Proc Natl Acad Sci U S A 2003 100 6216 6220 12724520
Goodrick CL Ingram DK Reynolds MA Freeman JR Cider N Effects of intermittent feeding upon body weight and life span in inbred mice: Interaction of genotype and age Mech Ageing Dev 1990 55 69 87 2402168
Rodgers JT Lerin C Haas W Gygi SP Spiegelman BM Nutrient control of glucose homeostasis through a complex of PGC-1[alpha] and SIRT1 Nature 2005 434 113 118 15744310
Kashiwaya Y Takeshima T Mori N Nakashima K Clarke K D-beta-Hydroxybutyrate protects neurons in models of Alzheimer's and Parkinson's disease Proc Natl Acad Sci U S A 2000 97 5440 5444 10805800
Levy R Cooper P Ketogenic diet for epilepsy Cochrane Database Syst Rev 2003 3 CD001903
Beckman M More without TOR. Inhibiting nutrient sensor extends life span in fruit flies Sci Aging Knowledge Environ 2004 2004 36
Heilbronn LK Ravussin E Calorie restriction and aging: review of the literature and implications for studies in humans Am J Clin Nutr 2003 78 361 369 12936916
|
16120010
|
PMC1196473
|
CC BY
|
2021-01-05 11:13:39
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e231
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020231
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612001110.1371/journal.pmed.0020256Student ForumOtherClinical PharmacologyMedical EducationMedical EducationDrugs and adverse drug reactionsMedicine in Developing CountriesLearning How Drug Companies Promote Medicines in Nepal Student ForumGiri Bishnu Rath Shankar P. Ravi *Bishnu Rath Giri is a sixth-semester medical student, and P. Ravi Shankar is Assistant Professor of Pharmacology, at Manipal College of Medical Sciences, Pokhara, Nepal.
Competing Interests: The authors declare that they have no competing interests.
*To whom correspondence should be addressed: E-mail: [email protected] 2005 30 8 2005 2 8 e256Copyright: © 2005 Giri and Shankar.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.An educational initiative looks critically at the drug industry's promotional tactics.
An educational initiative looks critically at the drug industry's promotional tactics
==== Body
Armed with the power to prescribe, doctors are in a position to flood their patients' bodies with potent medicines at a stroke of their pen—and they have been consistently wooed by the pharmaceutical industry.
Doctors in Nepal have started to become targets for the pharmaceutical industry's promotional activities. The mountainous, landlocked, developing country of Nepal, which has a population of about 23 million, has an annual per capita income, in terms of purchasing power parity, of about US$2,000 in urban areas and US$1,000 in rural areas. In Nepal, Western allopathic medicine coexists with traditional medical practices like Ayurveda, Tibetan medicine, and faith healing. Many rural areas lack access to modern health care, but Kathmandu and other cities are becoming booming markets for pharmaceuticals.
Pharmaceutical Promotion and Its Impact in Nepal
The promotional activities of medical representatives (MRs, also known as drug company representatives), and other promotional activities by pharmaceutical companies, have a major influence on prescribing and on drug use by the general public all over the world [1]. The World Health Organisation's Department of Essential Drugs and Medicines Policy maintains a database of such promotional activities at http://www.drugpromo.info. In Nepal, common prescribing problems are polypharmacy, and overprescription of antibiotics and injections [2]. We are studying prescribing patterns in our hospital and have observed excessive use of expensive antibiotics, vitamins, and digestive and enzyme preparations. The preparations prescribed are those that are actively promoted by MRs. I (PRS) have sometimes found consultants recommending to the hospital Drug and Therapeutics Committee that drugs of doubtful efficacy be included in the hospital pharmacy, following visits by MRs.
Most hospitals in Nepal allow free access of MRs to doctors, and academic detailing is absent. MRs bring with them many creative ideas for drug promotion. I (BRG) observed this for the first time when I visited one of my professors. A signboard next to the door read “Doctor is IN-DIGENE” (Digene is a brand of antacid). After going inside, I noticed that there was at least one big poster promoting a pharmaceutical company on every wall. On the table was a beautifully handcrafted nameboard with the professor's name in golden letters. The side facing the professor had the brand name of a drug in equally stylish lettering.
Medical conferences in Nepal are strongly dominated by the pharmaceutical industry. Often, companies organize parties for doctors in which a continuing medical education topic is followed by a lavish cocktail dinner—but often the educational part is absent. One such party was recently organized by a manufacturer of suture material and was attended by almost all the consultants and medical officers of our hospital.
After doctors accept personal benefits from the industry, they might arguably feel an obligation to prescribe the promoted drug. There is certainly good evidence in the literature that such gifts from the industry are associated with changes in prescribing behavior (the evidence is collected at http://nofreelunch.org/requiredinfluence.htm). Pharmaceutical companies also sponsor the activities of medical students (such sponsoring can take the form of sports matches, publications, and parties). The industry expects a substantial return for every rupee spent on doctors and medical students.
Often, though, the cost of promotion is directly recovered from the patient. A lady I (BRG) knew once said to me, “I am taking only half the medicine. If it works well then I will get more. It is too expensive.” The tendency to buy a lesser amount of the drug or no drug at all is a common phenomenon in Nepal and leads to therapeutic failure.
An Educational Initiative on Promotion
As a student at Manipal College of Medical Sciences (Figure 1), I (BRG) have had the opportunity to experience an educational initiative on drug companies' promotional activities. This initiative involves small groups of students in interactive learning sessions. I have learned about common methods of drug promotion; unethical promotion practices; ways in which promotional materials give false impressions of the absolute and relative risks and the risk-benefit ratio; and the interpretation of graphs used in advertisements and optimizing time spent with MRs.
Figure 1 Manipal College of Medical Sciences, Pokhara, Nepal
(Photo: Ravi Shankar)
Class activities include critical analysis of promotional material and drug advertisements against the World Health Organization's Ethical Criteria for Medicinal Drug Promotion (http://www.who.int/medicines/library/dap/ethical-criteria/ethicalen.shtml). These criteria, prepared by an international group of experts, “constitute a frame of reference for judging proper behavior in drug promotion, whether involving the contents of advertisements and package inserts or the sponsorship of scientific symposia, and the use of representatives.” The criteria “give manufacturers, distributors, the promotion industry, prescribers and consumer groups a framework to ensure that promotional practices are in keeping with acceptable ethical standards.” Recently, role-play has been introduced into the class activities to present the interaction between a MR and a doctor. The interactions are critically analyzed by the groups. Students are assessed in critical analysis of promotional material during the pharmacology practical examinations.
Prescribers often do not realize the influence of promotional materials and activities [3]. In our hospital, through the Drug Information Center, drug bulletin, and other measures, we are sensitizing prescribers to critically evaluate industry sources of drug information. Recognizing that it is easier to inculcate good prescribing habits in future doctors, teaching and learning about rational use of medicines and medicinal drug promotion are emphasized in the medical students' pharmacology course.
Impact of the Initiative
There has been a noticeable impact on the attitudes of medical students towards drug promotion. I (BRG) and my fellow students have been sensitized to the negative effects of aggressive promotion. In our hospital, aggressive promotion is evident but criticism is often difficult to make. When I discussed this essay with an intern, he responded, “No MR will approach you. You will need them from your internship onwards.”
I (PRS) am analyzing feedback obtained from a questionnaire about the educational initiatives. Preliminary assessment showed that students were favorably disposed toward and enjoyed the sessions. We are investing in the future, and believe that around 10%–15% of our students have been convinced to take a hard, critical look at drug promotion. “Enlightened” students will be an important asset to their communities and countries. Long-term studies of the impact will be carried out in the future.
Conclusion
The initiatives have helped me (BRG) critically analyze sources of information from the pharmaceutical industry. In the future, I am confident that I will be prepared for the sales pitches and gimmicks of the industry.
When prescribers are made aware of the methods used by drug promoters, it is more difficult for unhealthy practices to take root; the ultimate beneficiaries are the patient and the community. We (PRS and colleagues) plan to bring in actual MRs to demonstrate the promotional techniques and to show videotaped encounters between doctors and MRs. We plan to strengthen the role-plays in the future. We are trying to enlist the support of the hospital's Drug and Therapeutics Committee to conduct sessions on drug promotion for the interns and medical officers.
Writing for Student Forum
If you would like to write an essay for the Student Forum, please send an enquiry to E-mail: [email protected], stating in 100 words what the essay will be about and why it will be of interest to our readers. Further instructions on writing for the Student Forum, together with archived Student Forum articles, can be found at http://studentforum.plosmedicine.org. An international team of medical student advisers helps to choose articles for this section.
Citation: Giri BR, Shankar PR (2005) Learning how drug companies promote medicines in Nepal. PLoS Med 2(8): e256.
Abbreviation
MRmedical representative
==== Refs
References
Laing RO Hogerzeil HV Ross-Degnan D Ten recommendations to improve use of medicines in developing countries Health Policy Plan 2001 16 13 20
International Network for Rational Use of Drugs Training manual of the 16th national training course on rational use of drugs, 13–18 October 2003 2003 Kathmandu (Nepal) International Network for Rational Use of Drugs 66
Holloway K Green T Drug and therapeutic committees—A practical guide 2004 Available: http://www.who.int/medicines/library/par/who-edm-par-2004_1/WHO_EDM_PAR_2004_1_Drugs_and_therapeutics_committees.pdf . Accessed 28 June 2005
|
16120011
|
PMC1196479
|
CC BY
|
2021-01-05 10:40:30
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e256
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020256
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612001210.1371/journal.pmed.0020270Correspondence and Other CommunicationsPharmacology/Drug DiscoveryEpidemiology/Public HealthHealth PolicyHIV/AIDSMedical EthicsEthicsHealth PolicyHIV Infection/AIDSMedicine in Developing CountriesResource allocation and rationingEstimating the Number of Antiretroviral Treatment Facilities Based on the Wilson–Blower Method CorrespondenceMalangu Ntambwe
1
1University of LimpopoPretoriaSouth AfricaE-mail: [email protected]
Competing Interests: The author has declared that no competing interests exist.
8 2005 30 8 2005 2 8 e270Copyright: © 2005 Ntambwe Malangu.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.
Designing Equitable Antiretroviral Allocation Strategies in Resource- Constrained Countries
==== Body
The implementation of the comprehensive plan for the care, management, and treatment of HIV and AIDS in South Africa [1] needs to be supported by all. It is encouraging to note that Wilson and Blower [2] used South Africa to develop a novel method to determine the optimal strategy for allocating antiretroviral treatment (ART) sites among health-care facilities (HCFs) in KwaZulu–Natal.
An equitable allocation of HCFs is necessary to ensure that each individual with HIV will have an equal chance of receiving antiretroviral drugs (ARVs). We have applied their method to determine the number of ART HCFs per district in KwaZulu–Natal.
We first set out to assemble basic details about the KwaZulu–Natal health districts, namely, population and number of hospitals and fixed and mobile clinics. Secondly, by using population data as reported in the KwaZulu–Natal Department of Health 2004 annual report [3], and based on a 10% HIV prevalence, we determined the HIV population per district and also as a percentage of the total HIV population in the province. Finally, we calculated the number of ART HCFs, based on the premise that 54 ART HCFs will serve 100% of the HIV population in the province. By contrasting the estimated number of ART HCFs with the current ART HCFs, we calculated the number of ART HCFs that still need to be established.
The national target for access to HCFs is 10,000 habitants per one fixed primary health-care (PHC) facility [2]. A PHC facility could be a clinic, a community health center, or a hospital. At present, there is one fixed PHC facility for 17,215 inhabitants in KwaZulu–Natal [3].
With regard to ART, the 54 HCFs proposed by Wilson and Blower translate to 18,076 people with HIV per facility (Table 1). This number is close to the actual figure of 17,215 inhabitants per facility in the province.
Table 1 Number of ART HCF Calculated for Each District
Source: Kwazulu–Natal Department of Health [2].
In terms of equity, it could be argued, for instance, that the two facilities in the eThekwini district cannot be expected to provide ARVs to the estimated 319,994 individuals with HIV. In comparison, the Umziyathi, Amajuba, Umkhanyakude, Uthungulu, and Ugu districts currently have the same number of ART HCFs as eThekwini, but serve smaller populations (Table 1). This reflects the fact that the choice of the current facilities was guided more by practical considerations, such as availability of staff and infrastructure, than by the principle of equity as suggested by the World Health Organization [4].
From our calculations, it seems that in order to achieve treatment equity for individuals with HIV in KwaZulu–Natal, more ARV HCFs should be established as follows: 15 in eThekwini, four each in Umgungundlovu and Zululand, three each in Uthukela and Uthungulu, two each in Ugu and iLembe, and one each in Amajuba, Sisonke, Umzinyathi, and Umkhanyakude.
As a recommendation, future rollout of ART should take into consideration the principle of equity. This will ensure that all people with HIV have equal access to ARVs from their nearest HCF. We show that by applying the Wilson–Blower method, it is possible to determine the number of health-care facilities where ARVs would be equitably provided.
This correspondence letter was peer reviewed.
Citation: Malangu N (2005) Estimating the number of antiretroviral treatment facilities based on the Wilson–Blower method. PLoS Med 2(8): e270.
==== Refs
References
Tshabalala-Msimang M Statement of Cabinet on a plan for comprehensive treatment and care for HIV and AIDS in South Africa 2004 February Capetown (South Africa) Cape Gateway Available: http://www.capegateway.gov.za/eng/pubs/news/2004/feb/29782 . Accessed 18 July 2005
Wilson DP Blower SM Designing equitable antiretroviral allocation strategies in resource-constrained countries PLoS Med 2005 2 e50 10.1371/journal.pmed.0020050 15737005
KwaZulu–Natal Department of Health Annual report 2003–2004 2004 Pietermaritzburg KwaZulu–Natal Department of Health Available: http://www.kznhealth.gov.za/0304report.htm . Accessed 18 July 2005
World Health Organization, Joint United Nations Programme on HIV/AIDS Treating 3 million by 2005: Making it happen 2003 Geneva World Health Organization Available: http://www.who.int/3by5/publications/documents/isbn9241591129/en/ . Accessed 18 July 2005
|
16120012
|
PMC1196485
|
CC BY
|
2021-01-05 10:40:31
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e270
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020270
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612001310.1371/journal.pmed.0020272EditorialScience PolicyMedical journalsEditorial policies (including conflicts of interest)Communication in Health CareMinimizing Mistakes and Embracing Uncertainty EditorialThe PLoS Medicine Editors E-mail: [email protected]
8 2005 30 8 2005 2 8 e272Copyright: © 2005 Public Library of Science.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.
Why Most Published Research Findings Are False
Scientific truth is a moving target. Journals must apply rigorous standards, but publication of data and conclusions before they are proven true is part of the research enterprise.
==== Body
“Truth in science can be defined as the working hypothesis best suited to open the way to the next better one.”—Konrad Lorenz, Austria
Scientific truth is a moving target. In the process of peer review, authors, reviewers, and editors work together to minimize the reporting of false results. However, even if one assumes no bias, wrongdoing, or ignorance on the part of any of the individuals involved—which is unrealistic, no doubt—chances are that some findings will turn out to be false. But is it inevitable, as John Ioannidis argues in an Essay in this issue of PLoS Medicine (DOI: 10.1371/journal.pmed.0020124), that the majority of findings are actually false?
Although his calculations are based on assumptions about complex scenarios that we do not fully understand—as is true for most research projects—Ioannidis argues convincingly that many published findings will turn out to be false.
Ioannidis is not the first to raise some of these concerns. Indeed, there are already initiatives under way that seek to address them. Increasingly, researchers design individual studies, systematic reviews, and meta-analyses using Bayesian statistics, in which the issue of pre-study odds is taken into account. And issues such as reducing sources of bias when assessing evidence are addressed in the methodology used by the Cochrane Collaboration in the production of its systematic reviews.
Ioannidis doesn't define “findings,” but it seems important to attempt to separate data (“in this study 5% of people examined who lived in San Francisco from 1965–1970 developed lung cancer compared with 20% of people studied who lived in Anchorage”) from conclusions (“lung cancer rates are higher in Anchorage than San Francisco”) and hypotheses (“cold weather exacerbates the consequences of smoking”).
Hypotheses will inevitably change, as they depend not only on the study but also on the context of other relevant research and knowledge. Conclusions are also often based on current knowledge and assumptions, and, thus, subject to change. The data should be more robust; for instance, other researchers applying the same methods to study the same group of patients at the same time should be able to generate the same data. However, research progress depends on conclusions being tested elsewhere. The major issue about the truth of research findings would therefore seem to concern the conclusions, and Ioannidis's claim that most conclusions are false is probably correct. Is that a problem? Can it be avoided?
The possibility that most conclusions are false might be an inevitable part of the research endeavor. That said, researchers and those involved in publication of research must do what they can to reduce false conclusions.
One way to do this is to delay publication until such a time when the chances that a conclusion is true are sufficiently high. If many published conclusions are false, we (editors and reviewers) need to ask ourselves whether we are setting the bar too low. But what is the consequence of setting it higher?
Research progress depends on dissemination of results, and journal articles are the most effective tool we currently have to share them. The answer, therefore, cannot be that we wait until conclusions are proven beyond a doubt before we publish them. Publication of preliminary findings, negative studies, confirmations, and refutations is an essential part of the process of getting closer to the truth. Everyone involved in the generation and publication of research results needs to be open-minded, rigorous, and honest in designing experiments, analyzing results, reporting findings, peer-reviewing manuscripts, providing comments, and accepting that uncertainty exists in research.
Ioannidis suggests how studies could be designed from the outset to increase their chances of producing true results. He also gives some corollaries that allow readers to get a sense of the extent of uncertainty for a particular study. He stresses that reliable evidence generally comes from several studies and from several teams of researchers, and that what matters is the totality of the evidence.
What can editors do? At high-impact journals such as PLoS Medicine, we see it as our job to select important articles. This means the conclusions reported should be more rather than less likely to be true. But better measures of importance are that a study should address a substantial clinical or public- health question, in as rigorous a way as possible, and the findings should be likely to have an effect on how other researchers think about the question. In reporting studies, we ask that data are clearly delineated from conclusions, and conclusions from hypotheses. In addition to individual studies, editors should (and at PLoS Medicine we do) ensure there is a place for articles that synthesize evidence from different sources.
Too often editors and reviewers reward only the cleanest results and the most straightforward conclusions. At PLoS Medicine, we seek to create a publication environment that is comfortable with uncertainty. We encourage authors to discuss biases, study limitations, and potential confounding factors. We acknowledge that most studies published should be viewed as hypothesis-generating, rather than conclusive. And we publish high-quality negative and confirmatory studies.
We also accept some responsibility for educating consumers of research about the research process. Consumers also need to become comfortable with uncertainty, and understand the strengths and weaknesses intrinsic to every study conducted and published. Besides selecting papers and influencing how results are reported, we use the synopses and patient summaries to highlight uncertainties in research papers. We also encourage contributions such as the essay by Ioannidis to our magazine section that will help research producers and consumers to understand research findings in context.
Citation:
PLoS Medicine Editors (2005) Minimizing mistakes and embracing uncertainty. PLoS Med 2(8): e272.
|
16120013
|
PMC1196486
|
CC0
|
2021-01-05 10:40:31
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e272
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020272
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612001410.1371/journal.pmed.0020277Correspondence and Other CommunicationsOtherEpidemiology/Public HealthEvidence Based PracticeSystematic reviews and meta-analysesEpidemiologyMedicine in Developing CountriesResearch MethodsEvidence-Based Medicine in Iberoamerica: Problems and Possible Solutions CorrespondenceOrtiz Zulma
1
Perel Pablo
1
Pardo Jordi
1
1Argentine Collaborating Center of the Iberoamerican Cochrane NetworkBuenos AiresArgentinaE-mail: [email protected]
Competing Interests: The authors have declared that no competing interests exist.
8 2005 30 8 2005 2 8 e277Copyright: © 2005 Ortiz 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.
Is Evidence-Based Medicine Relevant to the Developing World?
==== Body
We want to congratulate Chinnock and colleagues, who summarize very well the main problems that evidence-based medicine faces in developing countries [1]. As members of the Iberoamerican Cochrane Network, we would like to share some lessons learned and highlight possible solutions to the problems identified by Chinnock et al. [1].
We have learned from the experience of working in and with Latin American countries that one of the first barriers to overcome is inequity in accessing evidence. The second barrier is the English language. Efforts have been made by our network to overcome both barriers by providing free access to Biblioteca Cochrane Plus (BCP) (http://www.bibliotecacochrane.net). In addition to systematic reviews and protocols, this database contains evidence-based information not indexed in other sources.
However, ensuring access does not necessarily mean that reviews will be used in decision making. Many local problems do not appear in the BCP material, but those that are most prevalent and those with high impact on public health and clinical practice of these countries have been reviewed. Nevertheless, few health professionals apply the results of such reviews. One possible solution is that Cochrane centers, groups, or fields, most of which are based in developed countries, could invest resources in mass dissemination and promote their activities through organizations such as the Pan American Health Organization. This would encourage not only the use of systematic reviews, but also promote an interest in the Cochrane Collaboration from health authorities in the Americas.
Another aspect revealed in Chinnock et al.'s article is the need to get more people from developing countries involved in writing and peer-reviewing systematic reviews. The nature of the Cochrane Collaboration facilitates this, and we have had excellent results when working with several of its groups and fields. However, developing countries have a limited number of people qualified to participate in the writing and peer-reviewing of systematic reviews. Most of those who have the necessary skills also have an enormous load of teaching and clinical care, their salaries are insufficient to support a white-collar lifestyle, and, thus, private practice is the most common means of augmenting earnings. These economic issues are by far the major factor underlying the relative lack of research in developing countries [2]. Cochrane groups and centers based in developed countries should identify potential reviewers in developing countries and invest resources that provide them with spare time to devote to the promotion, production, and evaluation of systematic reviews. This idea is in line with the Millennium Development Goals [3], specifically number eight, which addresses the need to develop a global partnership for development. Nevertheless, the concerns exposed by Chinnock and colleagues in connection with the search for reviews performed in developing countries would decrease if the use of databases specific to these regions, such as LILACS (Literatura Latinoamericana en Ciencias de la Salud) in Latin America, was encouraged and if the use and development of these databases were supported.
Finally, we consider that advocacy on the importance of research and evidence-based public health should be strengthened in developing countries. This has been highlighted by Bernardo Houssay, the first Latin American honored with the Nobel Prize, who said, “Science is only science when it involves constant progress and improvement arising from research. Thus, there are only two possible standpoints: that of tuggers and that of others being tugged. In other words, you may either create knowledge at the same time others do, or accept a subordinate position and depend on what others produce.” When the response to his views was different from what he expected, he added, “It would not be ethical to base a research strategy on the unfair exploitation of sacrifices made by those with exceptional and determined minds. Wise countries do not live waiting for saints or miracles to occur” (quoted in [4]).
Citation: Ortiz Z, Perel P, Pardo J (2005) Evidence-based medicine in Iberoamerica: problems and possible solutions. PLoS Med 2(8): e277.
==== Refs
References
Chinnock P Siegfried N Clarke M Is evidence-based medicine relevant to the developing world? PLoS Med 2005 2 e107 10.1371/journal.pmed.0020107 15916456
Bhutta Z Practising just medicine in an unjust world BMJ 2003 327 1000 1001 14593005
United Nations UN Millennium development goals 2000 New York United Nations Available: http://www.un.org/millenniumgoals/ . Accessed 18 July 2005
Charreau EH Incorporación a la Academia Nacional de Medicina 2004 May 13 Buenos Aires Consejo Nacional de Investigaciones Científicas y Técnicas Available: http://www.conicet.gov.ar/NOTICIAS/2004/Abril/nota06.php . Accessed 21 July 2005
|
16120014
|
PMC1196490
|
CC BY
|
2021-01-05 10:40:31
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e277
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020277
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612001510.1371/journal.pmed.002027805-PLME-C-0341Correspondence and Other CommunicationsOtherMedical JournalsEditorial Policies (Including Conflicts of Interest)PLoS Takes a Step Backward CorrespondencePippin John J
1
1Dallas, TexasUnited States of AmericaE-mail: [email protected]
Competing Interests: The author is a medical research consultant for the Physicians Committee for Responsible Medicine.
8 2005 30 8 2005 2 8 e278Copyright: © 2005 John J. Pippin.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.
Some Tolerance for Fur-Animal Studies in PLoS Medicine
==== Body
The only people who don't know in 2005 that animal research is irrelevant for human disease are those who don't understand it or those who benefit from it. As a physician, clinical researcher, and former animal researcher, I know that though they are our closest genetic relatives, primates have failed as research models virtually whenever they have been used.
As a partial list of failures, allow me to submit the notorious forced smoking experiments, which allowed cigarettes to be promoted widely for decades; the abject failure of a quarter-century of primate research on AIDS to provide any useful insights; the false leads and dangerous vaccines produced during polio research (verified by Albert Sabin, himself); the failure of primate studies to improve risks for birth defects and premature births; and the failure of monkey studies to identify nonsteroidal anti-inflammatory drug cardiovascular risk [1].
The PLoS Medicine editors state in hopeful language that the Lassa fever vaccine was successful in four monkeys, and, thus, is a suitable agent for human study [2]. Recall that VaxGen's AIDS vaccine (AIDSVAX) showed great success in primate studies, but was an abject failure in two human clinical trials, including a trial of over 2,500 injection drug users in Thailand [3] and a multinational trial of over 5,000 high-risk individuals [4].
Consider the fruitless decades-long effort to produce an AIDS vaccine in primates, the failure to produce even a single case of human AIDS in any primate studied, or the failure to identify even one useful AIDS drug from primate studies. Genetic and physiological imperatives dictate that no animal model, even higher primates, gives information applicable to humans. The Human Genome Project [5] tells us that there is sufficient genetic diversity among humans that pharmacogenetic and pharmacogenomic techniques will have an increasing role in overcoming problems related to polymorphisms and other variations. We can't even apply scientific findings uniformly to humans, and PLoS Medicine is now promoting monkey research?
I am very disappointed that PLoS Medicine has regressed to reporting animal research. It is discouraging that in this era of rapid biomedical advancement, and appropriate relegation of animal research to the historical dustbin, PLoS has chosen to re-introduce an anachronistic, medically discredited, and unethical research tool to its reporting.
Citation: Pippin JJ (2005) PLoS takes a step backward. PLoS Med 2(8): e278.
==== Refs
References
Merck Research Laboratories Electrical injury model of arterial and venous thrombus formation in the anesthetized African green monkey: Measurement of prothrombotic and antithrombotic effects 2000
Barbour V Cohen B Yamey G Some tolerance for fur—Animal studies in PLoS Medicine
PLoS Med 2005 2 e203 10.1371/journal.pmed.0020203 15971961
McCarthy M AIDS vaccine fails in Thai trial Lancet 2003 362 1728
McCarthy M HIV vaccine fails in phase 3 trial Lancet 2003 361 755 756 12620743
Human Genome Project Information Pharmacogenomics 2005 Washington (DC) US Department of Energy Office of Science Available: http://www.ornl.gov/sci/techresources/Human_Genome/medicine/pharma.shtml . Accessed 21 July 2005
|
16120015
|
PMC1196491
|
CC BY
|
2021-01-05 11:13:40
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e278
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020278
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612001610.1371/journal.pmed.002028005-PLME-C-0343Correspondence and Other CommunicationsBiotechnologyOtherMedicine in Developing CountriesNanotechnology for the Poor? CorrespondenceFoladori Guillermo
1
Invernizzi Noela
1
1Universidad Autonoma de ZacatecasZacatecas, MexicoE-mail: [email protected]
Competing Interests: The authors have declared that no competing interests exist.
8 2005 30 8 2005 2 8 e280Copyright: © 2005 Foladori and Invernizzi.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.
Nanotechnology and the Developing World
==== Body
After interviewing 63 experts, Salamanca-Buentello et al. [1] identified the ten main nanotechnologies that could provide a solution to problems involving water, agriculture, and health. Overflowing with good intentions, the proposal reflects the idea that if a problem can be identified, all that has to be done is apply a suitable technology and it will be solved. Most of the examples do not take into account that the relationship between science and society is much more complex.
The authors suggest that quantum dots could detect HIV molecules in the early stages, facilitating the treatment and reducing the number of new cases. The authors seem to forget the story of recent years, which has been one of open war between multinational pharmaceutical corporations and countries seeking to manufacture antiretrovirals. Nanotechnology products are already being patented. A patent in the US costs US$30 000 in legal bureaucracy, and a worldwide patent may be as much as US$250 000 [2]. The moral of the story: the efficiency and implications of the application of technology depend on the social context.
The article identifies nanotechnology as the solution to five of the eight UN Millennium Development Goals [3]. Among these solutions are nanosensors to improve the dosage of water and fertilization of plants, and hopefully reduce poverty and hunger. Not so long ago, genetically modified organisms were hailed as the solution that would put an end to hunger. However, they ended up being used mainly in developed countries. There has been no improvement for developing countries; quite the contrary, transgenics turned up where they were not wanted, as was the case in Oaxaca [4]. The moral of the story: the choice of technology is not a neutral process. It is not necessarily true that the technology that is best and meets our needs will be the one to survive.
In a previous article [5] three of the same authors maintained that the position adopted by Prince Charles—arguing that nanotechnology will widen the gap between rich and poor countries—and the position of the Action Group on Erosion, Technology and Concentration—requesting a moratorium on manufacture and commercialization of synthetic nanoparticles—both ignore the voices of people in developing countries. With their research the authors intended to fill this gap. But the opinion of scientists involved in nanotechnology does not necessarily fall in with the most appropriate pathways for satisfying the needs of the poor. We may concur that infectious diseases are one of the main problems that the developing world is facing, but we may differ radically on how a solution to this problem should be attained. Prevention is not the same thing as cure. Nanotechnology is not necessary to reduce malaria radically, as is suggested by the authors. In Henan Province, China, malaria was reduced by 99% between 1965 and 1990 as a result of social mobilization, backed up by fumigation, the use of mosquito nets, and traditional medicine based on artemisinin [6]. Viet Nam reduced the number of malaria-related deaths by 97% between 1992 and 1997 with similar methods [7]. The moral of the story: there are many means to an end, and technology is not always the solution. Organizing people can be just as important.
Citation: Foladori G, Invernizzi N (2005) Nanotechnology for the poor? PLoS Med 2(8): e280.
==== Refs
References
Salamanca-Buentello F Persad DL Court EB Martin DK Daar AS Nanotechnology and the Developing World PLoS Med 2005 2 e97 10.1371/journal.pmed.0020097 15807631
Regalado A Nanotechnology patents surge Wall Street Journal 2004 June 18 1 Sect A
United Nations UN millennium development goals 2000 Available: http://www.un.org/millenniumgoals/ . Accessed 18 July 2005
Schapiro M Lightman A Sarewitz D Desser C Blowback in genetic engineering Living with the genie: Essays on technology and the quest for human mastery 2003 Island Press Washington (DC) 261 272
Court E Daar AS Martin E Acharya T Singer PA Will Prince Charles et al diminish the opportunities of developing countries in nanotechnology? 2004 January 28 Bristol (United Kingdom) Nanotechweb.org. Available: http://www.nanotechweb.org/articles/society/3/1/1/1 . Accessed 18 July 2005
Jackson S Sleigh AC Liu XL Economics of malaria control in China: Cost performance and effectiveness of Henan's consolidation programme 2002 Geneva World Health Organization Available: http://www.who.int/tdr/publications/publications/pdf/sebrep1.pdf . Accessed 20 July 2005
World Health Organization Viet Nam reduces malaria death toll by 97% within five years 2002 Available: http://www.who.int/inf-new/mala1.htm . Accessed 18 July 2005
|
16120016
|
PMC1196492
|
CC BY
|
2021-01-05 10:40:31
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e280
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020280
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612001710.1371/journal.pmed.0020281Correspondence and Other CommunicationsOtherPharmacology/Drug DiscoveryClinical PharmacologyEpidemiology/Public HealthHealth PolicyGeneral MedicineGuidelinesHealth PolicyMedical journalsPublic HealthRegulationWhy We Whistleblowers Are Passionate in Our Convictions CorrespondenceKruszewski Stefan P
1
1Harrisburg, PennsylvaniaUnited States of AmericaE-mail: [email protected]
Competing Interests: The author has declared that no competing interests exist.
8 2005 30 8 2005 2 8 e281Copyright: © 2005 Stefan P. Kruszewski.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.
Why PLoS Sponsored a Roundtable of Medical Whistleblowers
==== Body
Whistleblowers serve no function if they cannot tell their stories. The present story of whistleblowing—as discussed, in part, in PLoS Medicine—that involves the pharmaceutical industry, pharmaceutical benefit management corporations, the managed care industry, and the political and lobbying forces that zealously guard their secrets could not have been told without the help of courageous men and women [1, 2] For that reason, those of us who congregated in Washington, D.C., on May 15th, 2005, at the invitation and support of the Public Library of Science and the Government Accountability Project feel particularly humbled and grateful to these two sponsors. Our convictions could not have been aired were it not for the essential First Amendment work of responsible journalists, who exemplify the best in investigatory research.
For me, whistleblowing is not a theoretical exercise. It has a human face and tangible features. It is the face of children and adults who have been injured or killed by misrepresented pharmaceuticals; clinical research trial results that have been sequestered from the scientific community and whose incomplete findings cause injury; and pharmaceuticals that are detailed to physicians, not to save lives or necessarily improve the health or welfare of the recipients, but to make money.
In the lonely and, at times, discouraging world of whistleblowing, we whistleblowers are passionate, and often successful, because our efforts have a different goal than the corporations and political interests whose operations we occasionally challenge. Our goal is to tell the truth. That honest effort is the source of any ethical difference we can or might make. Truth is the basis for the power of a whistleblower, one that can withstand the assault of unprecedented odds against being heard put forth by that sum of political power, expediency, and money.
A whistleblower's success depends upon competent and articulate media. The debate to improve the status quo—be it in pharmaceutical marketing or managed-care decision making—cannot proceed or flourish without it.
Ralph Waldo Emerson, American essayist and philosopher (1803–1882), commented about success (I have adapted his comments for all of us who gathered in Washington in mid-May 2005): “To leave the world a bit better, whether by a healthy child, a garden patch or a redeemed social condition; to know even one life breathed easier because you have lived; this is to have succeeded [as a whistleblower].”
Citation: Kruszewski SP (2005) Why we whistleblowers are passionate in our convictions. PLoS Med 2(8): e281.
==== Refs
References
Barbour V Cohen B Yamey G Why PLoS sponsored a roundtable of medical whistleblowers PLoS Med 2005 2 e208 10.1371/journal.pmed.0020208 15913415
Lenzer J What can we learn from medical whistleblowers? PLoS Med 2005 9 e209 10.1371/journal.pmed.0020209
|
16120017
|
PMC1196493
|
CC BY
|
2021-01-05 10:40:30
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e281
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020281
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612001710.1371/journal.pmed.0020281Correspondence and Other CommunicationsOtherPharmacology/Drug DiscoveryClinical PharmacologyEpidemiology/Public HealthHealth PolicyGeneral MedicineGuidelinesHealth PolicyMedical journalsPublic HealthRegulationWhy We Whistleblowers Are Passionate in Our Convictions CorrespondenceKruszewski Stefan P
1
1Harrisburg, PennsylvaniaUnited States of AmericaE-mail: [email protected]
Competing Interests: The author has declared that no competing interests exist.
8 2005 30 8 2005 2 8 e281Copyright: © 2005 Stefan P. Kruszewski.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.
Why PLoS Sponsored a Roundtable of Medical Whistleblowers
==== Body
Whistleblowers serve no function if they cannot tell their stories. The present story of whistleblowing—as discussed, in part, in PLoS Medicine—that involves the pharmaceutical industry, pharmaceutical benefit management corporations, the managed care industry, and the political and lobbying forces that zealously guard their secrets could not have been told without the help of courageous men and women [1, 2] For that reason, those of us who congregated in Washington, D.C., on May 15th, 2005, at the invitation and support of the Public Library of Science and the Government Accountability Project feel particularly humbled and grateful to these two sponsors. Our convictions could not have been aired were it not for the essential First Amendment work of responsible journalists, who exemplify the best in investigatory research.
For me, whistleblowing is not a theoretical exercise. It has a human face and tangible features. It is the face of children and adults who have been injured or killed by misrepresented pharmaceuticals; clinical research trial results that have been sequestered from the scientific community and whose incomplete findings cause injury; and pharmaceuticals that are detailed to physicians, not to save lives or necessarily improve the health or welfare of the recipients, but to make money.
In the lonely and, at times, discouraging world of whistleblowing, we whistleblowers are passionate, and often successful, because our efforts have a different goal than the corporations and political interests whose operations we occasionally challenge. Our goal is to tell the truth. That honest effort is the source of any ethical difference we can or might make. Truth is the basis for the power of a whistleblower, one that can withstand the assault of unprecedented odds against being heard put forth by that sum of political power, expediency, and money.
A whistleblower's success depends upon competent and articulate media. The debate to improve the status quo—be it in pharmaceutical marketing or managed-care decision making—cannot proceed or flourish without it.
Ralph Waldo Emerson, American essayist and philosopher (1803–1882), commented about success (I have adapted his comments for all of us who gathered in Washington in mid-May 2005): “To leave the world a bit better, whether by a healthy child, a garden patch or a redeemed social condition; to know even one life breathed easier because you have lived; this is to have succeeded [as a whistleblower].”
Citation: Kruszewski SP (2005) Why we whistleblowers are passionate in our convictions. PLoS Med 2(8): e281.
==== Refs
References
Barbour V Cohen B Yamey G Why PLoS sponsored a roundtable of medical whistleblowers PLoS Med 2005 2 e208 10.1371/journal.pmed.0020208 15913415
Lenzer J What can we learn from medical whistleblowers? PLoS Med 2005 9 e209 10.1371/journal.pmed.0020209
|
16120018
|
PMC1196494
|
CC BY
|
2021-01-05 10:40:31
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e282
|
latin-1
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020282
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612001910.1371/journal.pmed.0020285Correspondence and Other CommunicationsOtherMedical EducationAcademic MedicineMedical EducationPatientsThe Need for a New Specialist Professional Research System of “Pure” Medical Science CorrespondenceCharlton Bruce G
1
Andras Peter
1
1University of Newcastle upon TyneNewcastle upon TyneUnited KingdomE-mail: [email protected]
Competing Interests: The authors have declared that no competing interests exist.
8 2005 30 8 2005 2 8 e285Copyright: © 2005 Charlton and Andras.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.
Five Futures for Academic Medicine
Future Scenarios for Academic Medicine Should Include Other Health Disciplines
==== Body
Awasthi et al.'s discussion of the future of academic medicine [1] is stimulating, but the primary focus of policy should be enhancing scientific progress in medicine. Science policy should address the decline in major, clinically relevant “breakthroughs” over recent decades [2].
Medical research has become mostly an “applied” science, which implicitly aims at steady progress by an accumulation of small improvements, each increment having a high probability of validity. Applied medical science is therefore a social system of communications for generating pre-publication peer-reviewed knowledge ready for implementation [3]. However, the need for predictability dictated by peer reviewing of research funding and the need for a high probability of validity in published research makes modern medical science risk-averse. This has led to a decline in major therapeutic breakthroughs where new treatments for new diseases are required [2].
There is a need for the evolution of a specialized professional research system of pure medical science, where the major evaluation of validity occurs (in the manner of classic sciences) post-publication and by peer usage, rather than peer review [3,4]. The role of pure medical science would be to generate and critically evaluate radically novel and potentially important theories, techniques, therapies, and technologies.
Pure science ideas typically have a lower probability of being valid, but have the possibility of much greater benefit if they turn out to be true [5]. The domination of medical research by “applied” criteria means that even good ideas from pure medical science are typically ignored or rejected as being too speculative. It is possible to publish radical and potentially important ideas in medical science, but at present there is no formal mechanism by which pure science publications may be received, critiqued, evaluated, and extended to become suitable for “application”.
Pure medical science needs to evolve to constitute a typical specialized scientific system of formal communications among a professional community with close research groupings, journals, meetings, and electronic and Web communications—like any other science. However, the pure medical science system would have its own separate aims, procedures for scientific evaluation, institutional organization, funding, and support arrangements, and it would have a separate higher professional career path with distinctive selection criteria. For instance, successful leaders of pure medical science institutions would need different qualities from many of the current leaders of medical science, and would need to be selected on the basis of their specialized cognitive aptitudes and their record of having generated science-transforming ideas.
The main “market” for pure medical science would be “applied” medical scientists who need radical strategies to solve important clinical problems that are not yielding to established methods. Pure medical science units might then arise as an elite grouping linked to existing world-class applied medical research institutions. The direct financial stimulus to create elite pure medical science institutions might come from the leadership of academic “entrepreneurs” and imaginative patrons in the major funding foundations.
Citation: Charlton BG, Andras P (2005) The need for a new specialist professional research system of “pure” medical science. PLoS Med 2(8): e285.
==== Refs
References
Awasthi S Beardmore J Clark J Hadridge P Madani H Five futures for academic medicine PLoS Med 2005 2 e207 10.1371/journal.pmed.0020207 16000024
Charlton BG Andras P Medical research funding may have over-expanded and be due for collapse QJM 2005 98 53 55 15625354
Charlton BG Conflicts of interest in medical science: Peer usage, peer review and ‘CoI consultancy’ Med Hypotheses 2004 63 181 186 15236772
Charlton BG Andras P The future of ‘pure’ medical science: The need for a new specialist professional research system Med Hypotheses 2005 65 419 425 15985341
Charlton BG Inaugural editorial Med Hypotheses 2004 62 1 2
|
16120019
|
PMC1196495
|
CC BY
|
2021-01-05 10:40:31
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e285
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020285
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612001910.1371/journal.pmed.0020285Correspondence and Other CommunicationsOtherMedical EducationAcademic MedicineMedical EducationPatientsThe Need for a New Specialist Professional Research System of “Pure” Medical Science CorrespondenceCharlton Bruce G
1
Andras Peter
1
1University of Newcastle upon TyneNewcastle upon TyneUnited KingdomE-mail: [email protected]
Competing Interests: The authors have declared that no competing interests exist.
8 2005 30 8 2005 2 8 e285Copyright: © 2005 Charlton and Andras.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.
Five Futures for Academic Medicine
Future Scenarios for Academic Medicine Should Include Other Health Disciplines
==== Body
Awasthi et al.'s discussion of the future of academic medicine [1] is stimulating, but the primary focus of policy should be enhancing scientific progress in medicine. Science policy should address the decline in major, clinically relevant “breakthroughs” over recent decades [2].
Medical research has become mostly an “applied” science, which implicitly aims at steady progress by an accumulation of small improvements, each increment having a high probability of validity. Applied medical science is therefore a social system of communications for generating pre-publication peer-reviewed knowledge ready for implementation [3]. However, the need for predictability dictated by peer reviewing of research funding and the need for a high probability of validity in published research makes modern medical science risk-averse. This has led to a decline in major therapeutic breakthroughs where new treatments for new diseases are required [2].
There is a need for the evolution of a specialized professional research system of pure medical science, where the major evaluation of validity occurs (in the manner of classic sciences) post-publication and by peer usage, rather than peer review [3,4]. The role of pure medical science would be to generate and critically evaluate radically novel and potentially important theories, techniques, therapies, and technologies.
Pure science ideas typically have a lower probability of being valid, but have the possibility of much greater benefit if they turn out to be true [5]. The domination of medical research by “applied” criteria means that even good ideas from pure medical science are typically ignored or rejected as being too speculative. It is possible to publish radical and potentially important ideas in medical science, but at present there is no formal mechanism by which pure science publications may be received, critiqued, evaluated, and extended to become suitable for “application”.
Pure medical science needs to evolve to constitute a typical specialized scientific system of formal communications among a professional community with close research groupings, journals, meetings, and electronic and Web communications—like any other science. However, the pure medical science system would have its own separate aims, procedures for scientific evaluation, institutional organization, funding, and support arrangements, and it would have a separate higher professional career path with distinctive selection criteria. For instance, successful leaders of pure medical science institutions would need different qualities from many of the current leaders of medical science, and would need to be selected on the basis of their specialized cognitive aptitudes and their record of having generated science-transforming ideas.
The main “market” for pure medical science would be “applied” medical scientists who need radical strategies to solve important clinical problems that are not yielding to established methods. Pure medical science units might then arise as an elite grouping linked to existing world-class applied medical research institutions. The direct financial stimulus to create elite pure medical science institutions might come from the leadership of academic “entrepreneurs” and imaginative patrons in the major funding foundations.
Citation: Charlton BG, Andras P (2005) The need for a new specialist professional research system of “pure” medical science. PLoS Med 2(8): e285.
==== Refs
References
Awasthi S Beardmore J Clark J Hadridge P Madani H Five futures for academic medicine PLoS Med 2005 2 e207 10.1371/journal.pmed.0020207 16000024
Charlton BG Andras P Medical research funding may have over-expanded and be due for collapse QJM 2005 98 53 55 15625354
Charlton BG Conflicts of interest in medical science: Peer usage, peer review and ‘CoI consultancy’ Med Hypotheses 2004 63 181 186 15236772
Charlton BG Andras P The future of ‘pure’ medical science: The need for a new specialist professional research system Med Hypotheses 2005 65 419 425 15985341
Charlton BG Inaugural editorial Med Hypotheses 2004 62 1 2
|
0
|
PMC1196496
|
CC0
|
2021-01-05 10:40:31
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e299
|
latin-1
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020299
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612001910.1371/journal.pmed.0020285Correspondence and Other CommunicationsOtherMedical EducationAcademic MedicineMedical EducationPatientsThe Need for a New Specialist Professional Research System of “Pure” Medical Science CorrespondenceCharlton Bruce G
1
Andras Peter
1
1University of Newcastle upon TyneNewcastle upon TyneUnited KingdomE-mail: [email protected]
Competing Interests: The authors have declared that no competing interests exist.
8 2005 30 8 2005 2 8 e285Copyright: © 2005 Charlton and Andras.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.
Five Futures for Academic Medicine
Future Scenarios for Academic Medicine Should Include Other Health Disciplines
==== Body
Awasthi et al.'s discussion of the future of academic medicine [1] is stimulating, but the primary focus of policy should be enhancing scientific progress in medicine. Science policy should address the decline in major, clinically relevant “breakthroughs” over recent decades [2].
Medical research has become mostly an “applied” science, which implicitly aims at steady progress by an accumulation of small improvements, each increment having a high probability of validity. Applied medical science is therefore a social system of communications for generating pre-publication peer-reviewed knowledge ready for implementation [3]. However, the need for predictability dictated by peer reviewing of research funding and the need for a high probability of validity in published research makes modern medical science risk-averse. This has led to a decline in major therapeutic breakthroughs where new treatments for new diseases are required [2].
There is a need for the evolution of a specialized professional research system of pure medical science, where the major evaluation of validity occurs (in the manner of classic sciences) post-publication and by peer usage, rather than peer review [3,4]. The role of pure medical science would be to generate and critically evaluate radically novel and potentially important theories, techniques, therapies, and technologies.
Pure science ideas typically have a lower probability of being valid, but have the possibility of much greater benefit if they turn out to be true [5]. The domination of medical research by “applied” criteria means that even good ideas from pure medical science are typically ignored or rejected as being too speculative. It is possible to publish radical and potentially important ideas in medical science, but at present there is no formal mechanism by which pure science publications may be received, critiqued, evaluated, and extended to become suitable for “application”.
Pure medical science needs to evolve to constitute a typical specialized scientific system of formal communications among a professional community with close research groupings, journals, meetings, and electronic and Web communications—like any other science. However, the pure medical science system would have its own separate aims, procedures for scientific evaluation, institutional organization, funding, and support arrangements, and it would have a separate higher professional career path with distinctive selection criteria. For instance, successful leaders of pure medical science institutions would need different qualities from many of the current leaders of medical science, and would need to be selected on the basis of their specialized cognitive aptitudes and their record of having generated science-transforming ideas.
The main “market” for pure medical science would be “applied” medical scientists who need radical strategies to solve important clinical problems that are not yielding to established methods. Pure medical science units might then arise as an elite grouping linked to existing world-class applied medical research institutions. The direct financial stimulus to create elite pure medical science institutions might come from the leadership of academic “entrepreneurs” and imaginative patrons in the major funding foundations.
Citation: Charlton BG, Andras P (2005) The need for a new specialist professional research system of “pure” medical science. PLoS Med 2(8): e285.
==== Refs
References
Awasthi S Beardmore J Clark J Hadridge P Madani H Five futures for academic medicine PLoS Med 2005 2 e207 10.1371/journal.pmed.0020207 16000024
Charlton BG Andras P Medical research funding may have over-expanded and be due for collapse QJM 2005 98 53 55 15625354
Charlton BG Conflicts of interest in medical science: Peer usage, peer review and ‘CoI consultancy’ Med Hypotheses 2004 63 181 186 15236772
Charlton BG Andras P The future of ‘pure’ medical science: The need for a new specialist professional research system Med Hypotheses 2005 65 419 425 15985341
Charlton BG Inaugural editorial Med Hypotheses 2004 62 1 2
|
0
|
PMC1196497
|
CC0
|
2021-01-05 10:40:30
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e320
|
latin-1
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020320
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 10.1371/journal.pmed.002032105-PLME-CN-0397CorrectionMedical JournalsHealth Education (Including Prevention and Promotion)Correction: What Are the Roles and Responsibilities of the Media in Disseminating Health Information? CorrectionSchwitzer Gary Mudur Ganapati Henry David Wilson Amanda Goozner Merrill Simbra Maria Sweet Melissa Baverstock Katherine A 8 2005 30 8 2005 2 8 e321Copyright: © 2005 Public Library of Science.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.
What Are the Roles and Responsibilities of the Media in Disseminating Health Information?
==== Body
In PLoS Medicine, vol 2, issue 7, DOI: 10.1371/journal.pmed.0020215
The following sentence contains an incorrect number: “Only by massaging the numbers could one figure out that physicians would need to put 700 women on statins to eliminate one cancer case (in medical parlance, this is called number needed to treat).”
The corrected sentence is as follows: “Only by massaging the numbers could one figure out that physicians would need to put 140 women on statins to eliminate one cancer case (in medical parlance, this is called number needed to treat).”
This correction note may be found online at DOI: 10.1371/journal.pmed.0020321. Published August 30, 2005
Citation: (2005) Correction: What are the roles and responsibilities of the media in disseminating health information? PLoS Med 2(8): e321.
|
0
|
PMC1196498
|
CC0
|
2021-01-05 11:13:53
|
no
|
PLoS Med. 2005 Aug 30; 2(8):e321
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020321
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 1612862210.1371/journal.pbio.0030314Research ArticleBioinformatics/Computational BiologyEvolutionGenetics/Genomics/Gene TherapyVertebratesTwo Rounds of Whole Genome Duplication in the Ancestral Vertebrate Vertebrate Genome DuplicationDehal Paramvir
1
Boore Jeffrey L
1
2
1Evolutionary Genomics Department, Department of Energy Joint Genome Institute and Lawrence Berkeley National Laboratory, Walnut Creek, California, United States of America,2Department of Integrative Biology, University of California, Berkeley, California, United States of AmericaHolland Peter Academic EditorUniversity of OxfordUnited Kingdom10 2005 6 9 2005 6 9 2005 3 10 e3147 9 2004 8 7 2005 Copyright: © 2005 Dehal and Boore.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.
Clear Evidence for Two Rounds of Vertebrate Genome Duplication
The hypothesis that the relatively large and complex vertebrate genome was created by two ancient, whole genome duplications has been hotly debated, but remains unresolved. We reconstructed the evolutionary relationships of all gene families from the complete gene sets of a tunicate, fish, mouse, and human, and then determined when each gene duplicated relative to the evolutionary tree of the organisms. We confirmed the results of earlier studies that there remains little signal of these events in numbers of duplicated genes, gene tree topology, or the number of genes per multigene family. However, when we plotted the genomic map positions of only the subset of paralogous genes that were duplicated prior to the fish–tetrapod split, their global physical organization provides unmistakable evidence of two distinct genome duplication events early in vertebrate evolution indicated by clear patterns of four-way paralogous regions covering a large part of the human genome. Our results highlight the potential for these large-scale genomic events to have driven the evolutionary success of the vertebrate lineage.
The authors compared genomic map positions of paralogous genes from four vertebrate genomes and found evidence for two distinct genome duplication events in the vertebrate lineage.
==== Body
Introduction
It has long been hypothesized that the increased complexity and genome size of vertebrates has resulted from two rounds (2R) of whole genome duplication (WGD) occurring in early vertebrate evolution, thus providing the requisite raw materials [1]. This seemed to be supported by the long-standing speculation that humans have about 100,000 genes, roughly four times the number expected for invertebrates' genomes, but this is now known to be incorrect, with the actual human gene count being closer to 30,000 [2,3]. Conflicting analyses have now made this very controversial, with some studies supporting 2R (e.g., [4–8]), others seeing only a single round of WGD (e.g., [9–11]), and still others refuting WGD altogether by concluding that nothing greater than limited segmental duplications have occurred (e.g., [12,13]).
The 2R hypothesis had been bolstered by observations that a few gene families, e.g., Hox clusters [14], follow a “4:1 rule” in the numbers of vertebrate to invertebrate genes. However, comparison of the complete genome sequences of human [2,3] and Drosophila [15] revealed that less than 5% of homologous gene families follow the 4:1 rule [12]. Further, although two sequential duplications are expected to generate the evolutionary topology (AB)(CD) for the descendent genes, rather than (A)(BCD), in fact, the relationships of vertebrate multigene families do not generally show this pattern, as indicated by early studies using only a few genes [16] and confirmed as complete genome sequences became available [2,13]. (However, for a different view, see [17].) Several studies have incorporated data from sparse sampling of genes from taxa thought to have branched near to these purported duplications, including lamprey [18], hagfish, amphioxus [17,19–22], and Ciona [23]; although these results are useful for timing duplications, the conclusions could never be viewed as definitively resolving this issue because these products could have alternatively been generated by duplications of individual genes or short gene segments rather than by WGDs. Even duplicating all of the genes in a genome individually is quite different from a whole genome duplicating simultaneously.
There are several reasons why this has been a difficult issue to resolve. After duplication, only the minority of gene pairs will adopt a new function (“neofunctionalization”) or partition old functions (“subfunctionalization”) quickly enough to escape disabling mutations that would lead to their eradication [24]; therefore, rampant gene loss rapidly erases this signal of genome duplication. Further, four-member gene families, even those with the (AB)(CD) topology, can be generated by two rounds of duplications of individual genes or of segments much smaller than the entire genome, generating a condition that cannot be differentiated on this basis from 2R followed by many gene losses. This alternative scenario seems especially plausible because recent analyses have shown that gene duplications occur much more frequently than had been thought, with the typical rate being sufficient to duplicate an entire genome equivalent every 100 million years (MY) [25,26]. Until recently, no complete genome sequence has been available from an outgroup that is closely related to vertebrates, and all methods of phylogenetic reconstructions are less accurate with more distant relatives such as Drosophila and yeast [20]. Lastly, there has not been to date a method to accurately and comprehensively cluster genes into homologous families because methods that rely on sequence similarity alone are highly subject to artifactual association of slowly evolving paralogs and to erroneous exclusion of the more rapidly evolving genes.
Fortunately, as has been shown convincingly for the yeast genome and for Arabidopsis [27–30], evidence of an ancient genome duplication can be seen in the large-scale pattern of the physical locations of homologous genes, even when the great majority of the duplicated genes have been lost. Studies have shown that the human genome also has multiple regions of colinear paralogous gene copies [4,21,22,31–37], but considered the arrangements of too small a number of genes and genomic regions to be comprehensive. This approach is now available for a large-scale evaluation of the vertebrate 2R hypothesis, because complete (i.e., at least draft quality) genome sequences are available for the tunicate Ciona intestinalis [38] (a basal chordate outgroup) and the vertebrates Takifugu rubripes [39] (a pufferfish “fugu”), Mus musculus [40] (mouse), and human [2,3]. Figure 1 illustrates how the signal of two rounds of genome duplication could be retained by the large-scale pattern in location of duplicated genes, in which many tracks of paralogous duplicates (which may not contain identical subsets of genes) each occur at exactly four positions in the genome, i.e., “tetra-paralogons.” No similar signal would be generated by repeated duplications of genes or even large gene segments; only WGDs would result in such global organization of paralogous genes.
Figure 1 Pattern Predicted for the Relative Locations of Paralogous Genes from Two Genome Duplications
(A) Representation of a hypothetical genome that has 22 genes shown as colored squares.
(B) A genome duplication generates a complete set of paralogs in identical order.
(C) Many paralogous genes suffer disabling mutations, become pseudogenes, and are then lost. One could imagine this condition being evidence of a single round of genome duplication followed by significant gene losses.
(D) A second genome duplication recreates another set of paralogs in identical order, with multigene families that retained two copies now present in four, and those that had lost a member now present in two copies.
(E) Again, many paralogous genes suffer disabling mutations, become pseudogenes, and are then lost. Of course, unrelated gene duplications and transpositions can occur. Even though this leaves only a few four-member gene families, the patterns of 2- and 3-fold gene families unite, in various combinations, all four genomic segments, revealing that the sequential duplications had been of very large regions, in this case all or nearly all of this hypothetical genome.
Results
Gene Clustering and Duplication Timing
A graph-based method was used with the complete gene sets of the four chordates (98,517 total genes; see Table 1 for details of each step in the analysis) to generate clusters such that each contains all, and only, those genes that descended from a single gene in their common ancestor (Figure 2). A multiple sequence alignment and a maximum likelihood evolutionary tree were constructed for each cluster, then a Web browser interface was built so that each can be viewed individually. (For more details and updates that include more taxa, see the “PhIGs” [Phylogenetically Inferred Groups] Web site at http://phigs.org/.) We could then easily determine when each gene duplicated relative to lineage splitting by comparing these gene trees with the known evolutionary relationships of the animals. For example, a gene duplication that is specific to only one animal's lineage is seen as two genes from the same genome clustered together. A gene that duplicated once in the unique common ancestor of mouse and human would generate a tree that groups gene copy 1 of human and mouse and, separately, gene copy 2 of human and mouse. Put more generally, gene duplications that are shared by more that one species are seen as a replication of the phylogeny of the descendant organisms for each gene copy. Of course, gene losses and various combinations of these processes are seen as well. Figure 3 shows all possible gene topologies along with how each would be interpreted.
Figure 2 Overview of the Building of a Gene Cluster and Phylogenetic Tree Shown by a Hypothetical Example
(A) Each circle represents a gene, labeled with the source genome according to the first letter of each taxon—C, M, H, and F for Ciona, mouse, human, and fugu, respectively—and further differentiated by numeral. BLASTP was first used to search all vertebrate genes for the one most similar to Ciona's C1 gene, in this case the mouse gene M1. Then other genes are recruited to the cluster if they have a higher similarity score to M1 that that between C1 and M1, indicated here by the red lines. The six genes shown on the right side of the diagram have some sequence similarity to those in the cluster, but less than that between C1 and M1, so are not included. Because the vertebrates are more closely related to each other than any is to Ciona, each cluster will include those genes descended from a single gene in the common chordate ancestor, having arisen by either lineage splitting or gene duplication specific to one or more vertebrates. (See Materials and Methods for more details.)
(B) Evolutionary tree of the genes in this cluster show separate duplications for fugu and for human. Because the maximum likelihood method does not rely solely on sequence similarity, there is no significance to the mouse gene being most similar to C1. The mouse genome simply contained the most slowly evolving vertebrate gene in this multigene family; this can be from any vertebrate taxon with approximately equal likelihood.
Figure 3 Hypothetical Phylogenetic Tree Showing All Possible Types of Gene Relationships and How They Are Most Parsimoniously Interpreted
Interior nodes are designated in lower case for those that simply result from lineage splitting and in upper case for gene duplications within a lineage. Although not shown, nodes are still scored if there is gene loss. Phylogenetic trees for each gene family can be viewed at http://phigs.org/, also providing a valuable tool for improving the inference of gene function. DBFTS, duplication before fish–tetrapod split; DBPRS, duplication before primate–rodent split; FD, fugu duplication; fts, fish–tetrapod split; HD, human duplication; MD, mouse duplication; prs, primate–rodent split.
Table 1 Overview of the Process for Analyzing the Complete Gene Sets with Number of Genes Included at Each Step
This reveals that 46.6% of the ancestral chordate genes appear in duplicate in one or more of the vertebrate lineages, with 34.5% having at least one duplication before the divergence of fish from tetrapods and 23.5% having at least one duplication afterward. (Some of these are counted twice, having had duplications both before and after the fish–tetrapod split.) This means that there are 3,753 gene duplications placed at the base of Vertebrata, which is remarkable because the ancestral genome would be reasonably estimated to have had fewer than 20,000 genes, which is the case for the tunicate as well as other invertebrate outgroups. However, as can be seen in Figure 4, gene duplications are in large numbers on every branch of the tree, making it unclear whether this, in itself, indicates a significant acceleration in duplication rate. Additionally, of the gene clusters with duplications basal to the fish–tetrapod split, 20.6% have had one duplication event, 10.8% have had two, and 5.1% have had more than two, counter to the expectation from 2R, and casting further doubt on the significance of this for the 2R hypothesis.
Figure 4 Phylogenetic Analysis of the Four Chordates with Drosophila as an Outgroup
This phylogenetic tree is based on 766 concatenated single copy protein sequences totaling 313,797 amino acid positions with branches proportional to the amount of change. Numerals in bold above the branches indicate the number of gene duplications occurring in each lineage; numerals below indicate branch lengths.
Gene Family Membership
An early observation in support of 2R was that several gene families have expanded from a single member in invertebrates to having four members for some vertebrates. Previous studies, confirmed in this analysis, have shown that this is not generally true for vertebrate multigene families [12]. As can be seen in Figure S1, there is no peak at four for gene family membership for any vertebrate. In fact, even gene duplications do not predominate; for each vertebrate species considered individually, one member per cluster is the largest category, accounting for 55%, 57%, and 59% of the fugu, mouse, and human genes, respectively, with 53.4% of the gene clusters having no duplication events whatsoever. Thus, there is no signal of 2R remaining in gene family membership, despite early anecdotal observations to the contrary.
Determination of Concordantly Duplicated Regions
To test the extent to which the 3,753 early duplication events that are timed to the base of Vertebrata were generated as part of larger scale, multigene duplications, we examined the relative positions of these resulting paralogs in the human genome (which is currently the best assembled and annotated vertebrate genome). These results are shown in Figure 5 (and more comprehensively in Figure S2) in which the linear array of genes for each chromosome is used to query for paralogs generated by any duplication event prior to the fish–tetrapod split. It is apparent from these figures that there is a large-scale pattern of genome segments that are concordant in having similar arrangements of paralogous genes. We quantified this by identifying all cases in which two or more different early-duplicating genes are within a 100-gene window, then, for each, querying all other places in the genome, using a sliding window to count the number of cases in which their respective paralogs are within both 50 genes upstream as well as 50 genes downstream from that point. There is a distinct pattern of having multiple chromosomes matching with long linear stretches of paralogous genes. This indicates that these duplications occurred in very large segments, consistent with the hypothesis of WGD(s). Having matches to three other chromosomal segments is the dominant category, as can be seen by the darker coloring in Figures 5 and S2 and in the histogram of Figure 6. These patterns, with each genomic region corresponding in gene arrangement to sets of paralogs in three other genomic segments, are strong support for the 2R hypothesis.
Figure 5 Plot of the Genomic Positions of Paralogous Pairs of Human Genes that Arose from Duplications Predating the Fish–Tetrapod Split
The queries shown here use Chromosomes 2, 4, 5, and 10, as indicated for the four panels. (The complete set can be seen in Figure S2.) On the x-axis is each chromosome arranged from p to q telomeres. On the y-axis is each of the 22 human autosomes plus the X and Y chromosomes. For each query gene on the x-axis, a “hit” is scored if the subject chromosome contains a paralog generated by a gene duplication prior to the fish–tetrapod split. The lower portion of each panel plots the n-fold redundancy along the query chromosome as defined by pairs of paralogs detected in a sliding window analysis. See the Material and Methods section for details, but briefly, for every human query gene, a window was considered of 50 genes to the left and 50 genes to the right, with a “hit” obtained for the subject chromosome if it includes the early-duplicated paralogs of genes on each side of the query. Four-fold (i.e., including the query) matching, as expected by the 2R hypothesis, is highlighted in a darker shade of blue.
Figure 6 Histogram Showing the Lower Bound Estimate of N-fold Redundancy Using the Analysis Reported in Figure 5
This histogram is generated by counting the depth of paralogon redundancy across all human chromosomes as shown in the lower part of Figure S2 (and subsampled for Figure 5). The peak at 4-fold coverage is consistent with the 2R hypothesis, and constitutes a lower bound estimate, because the sliding window examines only a small span of flanking genes and would be highly subject to effects of local gene rearrangements.
Although the 4-fold (i.e., including the query segment) category is the most prevalent, it accounts for only 25% of the genome. Nonetheless, it is striking that this remains the largest category despite approximately 450 MY of evolution. This constitutes a strong signal of 2R, and could not reasonably have been generated by a series of smaller duplication events. For the latter to have generated this pattern, multiple duplications of the same region (or its resulting duplicates) would have to have occurred three times, and have done so for many regions throughout the genome. We would expect, rather, that independent, random duplications would follow a Poisson distribution; this contrasting situation is seen when the same analysis is done with all human gene paralogs generated by duplication after the split of fish and tetrapods (not shown). Even if we were to consider the alternative of a single WGD followed by subsequent independent duplications of large segments, it would be difficult to explain why these would have been predominantly 2-fold for previously duplicated regions. The most parsimonious explanation for the observed pattern can only be 2R.
Tetra-Paralogon Detection
To further establish 2R, we evaluated these sets of paralogs for whether this 4-fold matching indicates that they fall into tetra-paralogons, as illustrated in Figure 1. We formalized this by first identifying paralogons (paralogous genomic segments) containing the same set of at least two duplicated gene pairs, while allowing for a maximum of 100 unduplicated genes in between (similar to the approach in [10]). (The allowance of 100 genes is arbitrary, but the results are not critically dependent on this number, which is only used to find the blocks of paralogous genes.) We infer that duplicated genes in paralogons are likely to have arisen from a single duplication involving all contained, duplicated genes, and that the unique, intervening genes have resulted from differential gene deletions and subsequent genome rearrangements.
We identified 2,953 paralogous human gene pairs that are inferred to have resulted from 1,912 genes that duplicated prior to the divergence of the fish and tetrapod lineages (with some gene losses also). Of these paralogous genes, 32.4% are still in 386 detectable paralogons comprising 772 individual genomic segments, containing from two to 42 gene pairs (Table S1). Of these 772 genomic segments, 454 comprise tetra-paralogons (Figures 7A and S3, Table 2) as shown hypothetically in Figure 1, in which overlapping sets of paralogs fall into 4-fold groups. (Unfortunately, it was not possible for us to evaluate the hypothesis of an additional genome duplication unique to ray-finned fish [41,42] because of the generally poor contiguity of the fugu draft assembly.)
Figure 7 An Arbitrarily Selected Subset of the Human Genome Showing the Physical Relationships Among Paralogous Genes
(A) This is an example of the tetra-paralogous relationships of a subset of human genes that are all inferred, by gene trees, to have duplicated prior to the split of fish from tetrapods, but after the split of Ciona from vertebrates. These genes are on four chromosomes with their identities indicated outside of the circle. The complete set of tetra-paralogons can be viewed in Figure S3.
(B) In contrast, paralogous human genes generated by duplications after the split of fish and tetrapods, as shown for this sample of the same four human chromosomes, do not form such tetra-paralogons. Their pattern appears to result from smaller-scale tandem duplications of individual genes or segments, followed by slow rearrangements. In addition to these apparently functional gene pairs shown in the figure for this portion of the human genome, we have identified eight pseudogene pairs that occur on different chromosomes; it is not clear whether these pseudogenes are the result of random retrotransposition (or other rearrangement mechanisms) rather than gene conversion events between older duplicates, which would make it appear as though these had duplicated later than they actually did, as has been observed in yeast [29].
Table 2 Distribution of the Human Genome's Tetra-Paralogons by Chromosome under the Most Permissive Model for Signal Detection
In contrast, when looking at the gene pairs that arose from a duplication event after the divergence of the fish and mammal lineages (see Figure 4), we find only 11% are detected in paralogons in the human genome, indicating that these duplications have less commonly included large segments of the genome (Figure 7B). This is especially interesting in that their relative recency would make it more likely that any large duplications would remain detectable, reinforcing the contrast with the large-scale structure of those earlier duplications. By looking specifically for tandemly duplicated genes by defining them as paralogs on the same chromosome that are separated by fewer than 10 intervening genes, we can recognize that 50% of these human gene pairs arose from tandem duplication, compared with 6% for the human gene pairs that arose before the divergence of the fish and tetrapod lineages.
Discussion
No detectible signal of WGD exists in the analysis of gene family membership. There is no peak at four genes per family for any of the vertebrates (Figure S1) as might result from 2R. Presumably this results from a great number of subsequent gene losses that have erased this signal. Likewise, the phylogenetic timing of the duplication events is also inconclusive, because duplications are common on every branch (see Figure 4). Although there is a somewhat greater number assigned to the base of vertebrates, there is no reliable way to evaluate the significance of this. In fact, even if this larger number could be found to be statistically significant, it may simply indicate that this was a period with an accelerated duplication of individual genes or multigene segments or a reduction in the rate of gene loss, rather than indicating WGD.
Conclusive evidence for 2R is seen only when data from gene families, phylogenetic trees, and genomic map position are all taken together, as has been advocated by others [21,32,43]. When examining the genomic map position of only those genes in the human genome that trace their ancestry back to a duplication event at the base of vertebrates, a clear pattern of tetra-paralogons emerges, indicating that 2R occurred at the base of vertebrates. This signal remains most clearly in 25% of the human genome that forms the largest category in the analysis shown in Figures 5 and 6, but we also find that 72% of all human genes are included in the total extent of all of the paralogons that overlap with these regions, providing the least constrained estimate of the portion of the human genome still retaining structure from the 2R. This is the outside estimate, because some portion could have as well been the result of segmental duplications of regions earlier established by WGD. This is in contrast to the pattern seen for the many other gene duplications, which generated paralogs that are predominantly arranged in tandem.
This is particularly compelling considering that this signal has survived more than 450 MY of genome rearrangements and the loss of many genes. We can imagine the effect that duplications, translocations, inversions, and deletions (and combinations thereof) would have had on this analysis: (1) Duplications would cause an increase beyond the 4-fold category; (2) translocations would decrease the 4-fold category if they are pervasive enough to clear large regions of paralogs; (3) inversions can either cause a decrease in the number of chromosomes hit by moving paralogous genes beyond the detection of the sliding window analysis or cause an increase by spreading some paralogous genes across the boundaries into adjacent segments; both of which can be exacerbated by gene translocations that blur the edges of the corresponding regions; and (4) deletions would generally increase the 3-fold chromosome category at the expense of the 4-fold category, and a deletion that occurred between the two WGDs would increase the 2-fold chromosome category. Additionally, in some cases, a few individual gene deletions or translocations may have eliminated the links between pairs of duplicated genes. Through these, and combinations of these events, the original 4-fold co-linearity established by 2R (or something less than the perfect 4-fold pattern, if these duplications were long separated) has been eroded.
These tetra-paralogons are spread across nearly all human chromosomes (Table 2). Notably, chromosome Y does not have any tetra-paralogons, perhaps due to its relatively recent origin and small number of genes, or perhaps this indicates a more rapid rate of gene movement. Chromosome 21 also has no tetra-paralogons, and Chromosome 18 has only one that is small. These chromosomes, and other regions without tetra-paralogons, could be of recent origin or they could have undergone multiple rearrangement events that would have destroyed the signal.
Although our study does not specifically address the effect that 2R has had on vertebrate evolution, we note two interesting observations. First, the vast majority of duplicated genes were subsequently deleted, indicating that relatively few genes may have been responsible for the increased complexity seen in vertebrates. Second, it is possible that many genes were loosed from constraint after the genome duplications and experienced an accelerated rate of sequence change before returning to single copy, and it is possible that this has played some role in the evolution of vertebrate complexity [44].
The mechanism of these genome duplication events, whether two separate rounds of either auto- or allo-tetraploidy or a single octoploidy, remains uncertain. We speculate that the most likely scenario is two rounds of closely spaced auto-tetraploidization events, based on the following observations. For most sets of tetra-paralogs, some pairs within the set extend over a longer region than others, indicating two distinct duplication events. If, alternatively, there had been a single octoploidy, then we would have to hypothesize multiple occasions in which two of the four descendant genomic segments lost the same sets of genes independently, which seems unlikely. The phylogenetic trees for the gene families are not consistently nested, as would be expected in the case of allo-tetraploidy or two widely spaced auto-tetraploidy events. Finally, tree topologies of genes within paralogy blocks are not always congruent, indicating that the process of gene loss and rediploidization spanned the duplication events [17].
It remains unclear to what extent such large-scale genomic events have driven macroevolutionary change versus the regular accumulation of small mutations, as is the central tenet of the classical model of evolution. We imagine that rapid and extensive evolutionary change could possibly be an emergent property of having all genes duplicated at the same time, allowing this expanded gene repertoire to evolve together, and so reach a greater level of interaction and complexity than could evolve from cumulative single gene duplications. WGDs have occurred in many lineages, including frogs [45,46], fish [41,42,47], yeast [27–30], Arabidopsis [27–30], and corn and several other crop species [48], all of which are being studied by modern genomics techniques. We view the broad and pervasive distribution of these tetra-paralogons in the human genome, despite the remarkably small number of genes remaining in duplicate, as robust evidence that 2R occurred at the base of Vertebrata, and anticipate that future studies will soon illuminate the roles this has played in the evolutionary success of the vertebrate lineage.
Materials and Methods
Obtaining chordate sequences
Sequences and gene annotations of the tunicate Ciona intestinalis and the pufferfish Takifugu rubripes were obtained from the Department of Energy Joint Genome Institute Web site at http://www.jgi.doe.gov. Sequences for Homo sapiens (version 19.34b.2) and Mus musculus (version 19.32.2) were obtained from the Ensembl project website at http://www.ensembl.org. For genes with multiple transcripts, only the longest protein sequence was taken, resulting in 15,852 Ciona, 37,241 fugu, 22,444 mouse, and 22,980 human genes. Table 1 shows an overview of the methods along with the numbers of genes and clusters included after each step.
Clustering
The objective of the gene clustering is to reconstruct groups of genes such that each includes all (and only) the descendents of a single gene in the ancestral chordate. The underlying assumption is that all of the vertebrate genes in such a cluster will have a higher degree of similarity to each other than they will to their ortholog in Ciona, because they have arisen after the Ciona–vertebrate divergence by either gene duplication or lineage splitting. We conceptually translated the protein sequences for all genes, then for each Ciona protein sequence, the best match to any vertebrate protein was found using BLASTP [49]. Likewise, for each vertebrate protein, the best Ciona match was found. This list of best Ciona–vertebrate hits was then ordered by raw score. A graph was constructed such that each protein sequence appears at a node and the raw BLASTP scores between each pair form the weight of each edge. These sequences were then grouped by using the pairs of best hits as seeds for a single linkage clustering of the graph with the minimum edge score of the seed. This recruits to each cluster any sequence with greater similarity to the individual seed sequences than they have to each other, ensuring that genes with similarity due to a duplication before the split of Ciona and Vertebrata are properly apportioned into separate clusters. Any cluster that attempts to use a protein that has already been assigned is eliminated to reduce ambiguity and any cluster with greater than 100 members in a single species is eliminated. Figure 2 illustrates this clustering process.
Phylogenetic analysis
A multiple sequence alignment for each cluster was created using ClustalW 1.81 [50]. This alignment was then trimmed by eliminating all positions with gap characters. If the remaining length of the multiple sequence alignment was less than 100 amino acids, the entire cluster was eliminated. Phylogenetic trees were constructed by using the quartet puzzling maximum likelihood method as implemented in TREE-PUZZLE 5.1 [51] using the JTT model of amino acid substitution and a gamma distribution of rates over eight rate categories with 10,000 puzzling steps used to assess reliability. Any tree whose nodes were not strictly bifurcating was eliminated. Even with strict requirements for membership in the clusters, for reliable sequence alignment, and for confidence of evolutionary analysis, we generated 6,641 gene family clusters that include 39,136 (39.7%) of 98,517 total chordate genes (see Table 1), of which 3,096 had duplicated vertebrate genes and 1,621 produced trees that are strictly bifurcating (i.e., having no polytomies).
Identification of node types
Each node of each tree was classified in comparison to the known evolutionary relationships of the animals. For example, if the gene cluster tree contains exactly four members, and one from each animal, then the parsimonious inference is that no gene duplication occurred. In the case of a similar cluster, but where one member is missing, this is a gene loss in a single group. Gene cluster trees can show duplications specific to individual lineages by having two genes clustered together for the same animal. A gene duplication that occurred before the split of fish from tetrapods is seen as a duplicated tree of the animal relationships after their splitting from Ciona and, similarly, a gene duplication that occurred in tetrapods but before the split of the mouse and human lineages is seen as a duplication of the mouse–human group. Combinations of these, such as a duplication for tetrapods followed by a loss in one of the tetrapod lineages, are also seen and scored (see Figure 3).
The sorting of orthologous and paralogous relationships for each gene cluster provides an effective tool for improving the inferences of gene function by allowing annotations from well studied genomes to be transferred to the orthologous genes of other species. Inferring function from orthology is expected to be more accurate than using sequence similarity alone, since the latter tends to incorrectly associate slowly evolving paralogs. We provide a web based resource for this sorting at http://phigs.org/.
Detecting concordantly duplicated genomic regions
An overview of the genomic distribution and patterns of paralogous gene arrangement was created by plotting the chromosomal location for all genes having a duplication before the fish–tetrapod divergence for each chromosome. We performed a sliding window analysis, looking upstream and downstream of each gene in turn, for 50 genes on each side, and asking whether paralogs generated prior to the fish–tetrapod split occur within 100 genes of each other at another genomic location. This is the most conservative approach for detecting this signal. The results are shown in Figures 5 and S2.
Detecting paralogons and tetra-paralogons
A paralogon was defined as two chromosomal locations in the same genome that have the same set of gene pairs, allowing for a maximum of 100 unduplicated genes between. This significantly expands the set of regions that can be detected by the sliding window analysis. These were detected for the entire human genome considering separately only those paralogs generated by duplications prior to the split between fish and tetrapods versus those arising from duplications after. Each paralogon was tested for its membership in additional paralogous-region pairs. When a segment pairs with three different paralogons, we considered all six possible pairings of the four regions. If, and only if, all six possible combinations are confirmed as paralogons, the group was defined as a tetra-paralogon (see Figure 1). A minimum reconstruction of the signal of 2R remaining in the human genome is found in the extent of complete 4-fold overlap and the maximum by extending these regions to include the complete extent of all contributing paralogons. Genome and chromosome coverage were calculated by summing the number of genes that are encompassed by this more expansive reconstruction and dividing by the total number of genes.
Inclusion of Drosophila genes and phylogenetic analysis
For each cluster that contained only a single gene from each of the four chordate species, the highest scoring BLASTP match to the Drosophila melanogaster gene set [15] was added. We then performed a multiple sequence alignment for each of these 766 sets of genes, followed by phylogenetic analysis of this concatenated data set as above. The resulting tree was rooted at the midpoint with branch lengths proportional to the amount of amino acid substitution as estimated by TREE-PUZZLE 5.1 [51].
Supporting Information
Figure S1 Histogram of Gene Cluster Membership
The numbers of genes per cluster are shown for each of the three vertebrates individually as well as for all three grouped together. There is no peak at four for any species, or at 12 as the total for all (or 16 for all, considering that there may have been a further genome duplication for fish), indicating that gene losses have been common and have eradicated this signal of genome duplications.
(30 KB JPG).
Click here for additional data file.
Figure S2 Plot of the Genomic Positions of Paralogous Pairs of Human Genes that Arose from Duplications Pre-dating the Fish–Tetrapod Split
On the x-axis is each chromosome arranged from p to q telomeres. On the y-axis is each of the 22 human autosomes plus the X and Y chromosomes. For each query gene on the x-axis, a “hit” is scored if the subject chromosome contains a paralog generated by a gene duplication prior to the fish–tetrapod split. The lower portion of each panel plots the n-fold redundancy along the query chromosome as defined by pairs of paralogs detected in a sliding window analysis. See the Material and Methods section for details, but briefly, for every human query gene, a window was considered of 50 genes to the left and 50 genes to the right, with a “hit” obtained for the subject chromosome if it includes the early-duplicated paralogs of genes on each side of the query. The expected value of four for the 2R hypothesis is highlighted in a darker shade of blue.
(5.2 MB DOC).
Click here for additional data file.
Figure S3 Illustration of Whole Genome 4-Fold Paralogy
The lines connect paralogous genes in the human genome that originated in duplications that occurred after the tunicate–vertebrate split but before the fish–tetrapod split. Numerals around the outside of the figure refer to human chromosome numbers.
(103 KB JPG).
Click here for additional data file.
Table S1 Paralogons in the Human Genome
Paralogons in the human genome are defined as having two or more pairs of paralogous genes separated by no more than 100 intervening genes (see Materials and Methods). A and B in the header refer to the first and second chromosome in considered pairs. The columns labeled “Start” and “End” define the extent of each paralogon by numbered genes. The number of paralogous gene pairs defining the paralogon and the total number of genes encompassed by the region are indicated.
(344 KB DOC).
Click here for additional data file.
We thank R. Baker, M. P. Francino, M. Medina, J. Schwarz, and Y. Valles for helpful comments on the manuscript, and S. Rash and W. Huang for technical assistance. This work was performed under the auspices of the U.S. Department of Energy, Office of Biological and Environmental Research, by the University of California, Lawrence Berkeley National Laboratory, under contract No. DE-AC03-76SF00098.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. PD and JB conceived and designed the experiments. PD performed the experiments. PD and JB wrote the paper.
Citation: Dehal P, Boore JL (2005) Two rounds of whole genome duplication in the ancestral vertebrate. PLoS Biol 3(10): e314.
Abbreviations
2Rtwo rounds of whole genome duplication
MYmillion years
WGDwhole genome duplication
==== Refs
References
Ohno S Evolution by gene duplication 1970 Berlin Springer-Verlag 160
Lander ES Linton LM Birren B Nusbaum C Zody MC Initial sequencing and analysis of the human genome Nature 2001 409 860 921 11237011
Venter JC Adams MD Myers EW Li PW Mural RJ The sequence of the human genome Science 2001 291 1304 1351 11181995
Lundin LG Evolution of the vertebrate genome as reflected in paralogous chromosomal regions in man and the house mouse Genomics 1993 16 1 19 8486346
Meyer A Schartl M Gene and genome duplications in vertebrates: the one-to-four (-to-eight in fish) rule and the evolution of novel gene functions Curr Opin Cell Biol 1999 11 699 704 10600714
Spring J Vertebrate evolution by interspecific hybridization—are we polyploid? FEBS Lett 1997 400 2 8 9000502
Wang Y Gu X Evolutionary patterns of gene families generated in the early stage of vertebrates J Mol Evol 2000 51 88 96 10903375
Larhammar D Lundin LG Hallbook F The human Hox-bearing chromosome regions did arise by block or chromosome (or even genome) duplications Genome Res 2002 12 1910 1920 12466295
Guigo R Muchnik I Smith TF Reconstruction of ancient molecular phylogeny Mol Phylogenet Evol 1996 6 189 213 8899723
McLysaght A Hokamp K Wolfe KH Extensive genomic duplication during early chordate evolution Nat Genet 2002 31 200 204 12032567
Gu X Wang Y Gu J Age distribution of human gene families shows significant roles of both large- and small-scale duplications in vertebrate evolution Nat Genet 2002 31 205 209 12032571
Friedman R Hughes AL The temporal distribution of gene duplication events in a set of highly conserved human gene families Mol Biol Evol 2003 20 154 161 12519918
Friedman R Hughes AL Pattern and timing of gene duplication in animal genomes Genome Res 2001 11 1842 1847 11691848
Popovici C Leveugle M Birnbaum D Coulier F Homeobox gene clusters and the human paralogy map FEBS Lett 2001 491 237 242 11240134
Adams MD Celniker SE Holt RA Evans CA Gocayne JD The genome sequence of Drosophila melanogaster
Science 2000 287 2185 2195 10731132
Hughes AL Phylogenies of developmentally important proteins do not support the hypothesis of two rounds of genome duplication early in vertebrate history J Mol Evol 1999 48 565 576 10198122
Furlong RF Holland PW Were vertebrates octoploid? Philos Trans R Soc Lond B Biol Sci 2002 357 531 544 12028790
Escriva H Manzon L Youson J Laudet V Analysis of lamprey and hagfish genes reveals a complex history of gene duplications during early vertebrate evolution Mol Biol Evol 2002 19 1440 1450 12200472
Holland PW Garcia-Fernandez J Williams NA Sidow A Gene duplications and the origins of vertebrate development Dev Suppl 1994 125 133 7579513
Holland PW More genes in vertebrates? J Struct Funct Genomics 2003 3 75 84 12836687
Abi-Rached L Gilles A Shiina T Pontarotti P Inoko H Evidence of en bloc duplication in vertebrate genomes Nat Genet 2002 31 100 105 11967531
Panopoulou G Hennig S Groth D Krause A Poustka AJ New evidence for genome-wide duplications at the origin of vertebrates using an amphioxus gene set and completed animal genomes Genome Res 2003 13 1056 1066 12799346
Leveugle M Prat K Popovici C Birnbaum D Coulier F Phylogenetic analysis of Ciona intestinalis gene superfamilies supports the hypothesis of successive gene expansions J Mol Evol 2004 58 168 181 15042337
Wolfe KH Yesterday's polyploids and the mystery of diploidization Nat Rev Genet 2001 2 333 341 11331899
Lynch M O'Hely M Walsh B Force A The probability of preservation of a newly arisen gene duplicate Genetics 2001 159 1789 1804 11779815
Lynch M Conery JS The evolutionary fate and consequences of duplicate genes Science 2000 290 1151 1155 11073452
Wong S Butler G Wolfe KH Gene order evolution and paleopolyploidy in hemiascomycete yeasts Proc Natl Acad Sci U S A 2002 99 9272 9277 12093907
Vision TJ Brown DG Tanksley SD The origins of genomic duplications in Arabidopsis Science 2000 290 2114 2117 11118139
Kellis M Birren BW Lander ES Proof and evolutionary analysis of ancient genome duplication in the yeast Saccharomyces cerevisiae
Nature 2004 428 617 624 15004568
Dietrich FS Voegeli S Brachat S Lerch A Gates K The Ashbya gossypii genome as a tool for mapping the ancient Saccharomyces cerevisiae genome Science 2004 304 304 307 15001715
Katsanis N Fitzgibbon J Fisher EM Paralogy mapping: identification of a region in the human MHC triplicated onto human chromosomes 1 and 9 allows the prediction and isolation of novel PBX and NOTCH loci Genomics 1996 35 101 108 8661110
Pebusque MJ Coulier F Birnbaum D Pontarotti P Ancient large-scale genome duplications: phylogenetic and linkage analyses shed light on chordate genome evolution Mol Biol Evol 1998 15 1145 1159 9729879
Gibson TJ Spring J Evidence in favour of ancient octaploidy in the vertebrate genome Biochem Soc Trans 2000 28 259 264 10816139
Vienne A Rasmussen J Abi-Rached L Pontarotti P Gilles A Systematic phylogenomic evidence of en bloc duplication of the ancestral 8p11.21–8p21.3-like region Mol Biol Evol 2003 20 1290 1298 12777526
Luke GN Castro LF McLay K Bird C Coulson A Dispersal of NK homeobox gene clusters in amphioxus and humans Proc Natl Acad Sci USA 2003 100 5292 5295 12704239
Castro LF Furlong RF Holland PW An antecedent of the MHC-linked genomic region in amphioxus Immunogenetics 2004 55 782 784 14749904
Castro LF Holland PW Chromosomal mapping of ANTP class homeobox genes in amphioxus: Piecing together ancestral genomes Evol Dev 2003 5 459 465 12950625
Dehal P Satou Y Campbell RK Chapman J Degnan B The draft genome of Ciona intestinalis Insights into chordate and vertebrate origins Science 2002 298 2157 2167 12481130
Aparicio S Chapman J Stupka E Putnam N Chia JM Whole-genome shotgun assembly and analysis of the genome of Fugu rubripes
Science 2002 297 1301 1310 12142439
Waterston RH Lindblad-Toh K Birney E Rogers J Abril JF Initial sequencing and comparative analysis of the mouse genome Nature 2002 420 520 562 12466850
Van de Peer Y Taylor JS Meyer A Are all fishes ancient polyploids? J Struct Funct Genomics 2003 3 65 73 12836686
Jaillon O Aury JM Brunet F Petit JL Stange-Thomann N Genome duplication in the teleost fish Tetraodon nigroviridis reveals the early vertebrate proto-karyotype Nature 2004 431 946 957 15496914
Horton AC Mahadevan NR Ruvinsky I Gibson-Brown JJ Phylogenetic analyses alone are insufficient to determine whether genome duplication(s) occurred during early vertebrate evolution J Exp Zool B Mol Dev Evol 2003 299 41 53 14508816
Seoighe C Johnston CR Shields DC Significantly different patterns of amino acid replacement after gene duplication as compared to after speciation Mol Biol Evol 2003 20 484 490 12654935
Tymowska J Fischberg M Tinsley RC The karyotype of the tetraploid species Xenopus vestitus Laurent (Anura: pipidae) Cytogenet Cell Genet 1977 19 344 354 611004
Jeffreys AJ Wilson V Wood D Simons JP Kay RM Linkage of adult alpha- and beta-globin genes in X. laevis and gene duplication by tetraploidization Cell 1980 21 555 564 7407927
Taylor JS Van de Peer Y Braasch I Meyer A Comparative genomics provides evidence for an ancient genome duplication event in fish Philos Trans R Soc Lond B Biol Sci 2001 356 1661 1679 11604130
Blanc G Wolfe KH Widespread paleopolyploidy in model plant species inferred from age distributions of duplicate genes Plant Cell 2004 16 1667 1678 15208399
Altschul SF Gish W Miller W Myers EW Lipman DJ Basic local alignment search tool J Mol Biol 1990 215 403 410 2231712
Thompson JD Higgins DG Gibson TJ CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice Nucleic Acids Res 1994 22 4673 4680 7984417
Schmidt HA Strimmer K Vingron M von Haeseler A TREE-PUZZLE: Maximum likelihood phylogenetic analysis using quartets and parallel computing Bioinformatics 2002 18 502 504 11934758
|
16128622
|
PMC1197285
|
CC BY
|
2021-01-05 08:21:27
|
no
|
PLoS Biol. 2005 Oct 6; 3(10):e314
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030314
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 1613122710.1371/journal.pbio.0030323Research ArticleNeuroscienceIn VitroDynamics of SNARE Assembly and Disassembly during Sperm Acrosomal Exocytosis SNARE Assembly Drives ExocytosisBlas Gerardo A. De
1
Roggero Carlos M
1
Tomes Claudia N [email protected]
1
Mayorga Luis S [email protected]
1
1Laboratorio de Biología Celular y Molecular, Instituto de Histología y Embriología, Consejo Nacional de Investigaciones Científicas y Técnicas, Facultad de Ciencias Médicas, Universidad Nacional de Cuyo, Mendoza, ArgentinaHughson Fred Academic EditorPrinceton UniversityUnited States of America10 2005 6 9 2005 6 9 2005 3 10 e32329 3 2005 14 7 2005 Copyright: © 2005 De Blas 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.
EnSNAREd by the Sperm Acrosome
The dynamics of SNARE assembly and disassembly during membrane recognition and fusion is a central issue in intracellular trafficking and regulated secretion. Exocytosis of sperm's single vesicle—the acrosome—is a synchronized, all-or-nothing process that happens only once in the life of the cell and depends on activation of both the GTP-binding protein Rab3 and of neurotoxin-sensitive SNAREs. These characteristics make acrosomal exocytosis a unique mammalian model for the study of the different phases of the membrane fusion cascade. By using a functional assay and immunofluorescence techniques in combination with neurotoxins and a photosensitive Ca2+ chelator we show that, in unactivated sperm, SNAREs are locked in heterotrimeric cis complexes. Upon Ca2+ entry into the cytoplasm, Rab3 is activated and triggers NSF/α-SNAP-dependent disassembly of cis SNARE complexes. Monomeric SNAREs in the plasma membrane and the outer acrosomal membrane are then free to reassemble in loose trans complexes that are resistant to NSF/α-SNAP and differentially sensitive to cleavage by two vesicle-associated membrane protein (VAMP)–specific neurotoxins. Ca2+ must be released from inside the acrosome to trigger the final steps of membrane fusion that require fully assembled trans SNARE complexes and synaptotagmin. Our results indicate that the unidirectional and sequential disassembly and assembly of SNARE complexes drive acrosomal exocytosis.
Unidirectional and sequential disassembly and assembly of SNARE complexes drive sperm acrosomal exocytosis.
==== Body
Introduction
Regulated exocytosis is a sophisticated process that requires the specific attachment of secretory granules to the plasma membrane and the opening of fusion pores connecting the interior of the granule to the extracellular medium [1]. Several of the proteins involved have been identified and characterized by genetic approaches, reconstitution assays, and biochemical means. The current consensus paradigm for membrane fusion is based on results obtained from diverse cellular systems, ranging from yeast to neurons. For instance, the roles assigned to small GTPases of the Rab family derive from studies carried out in endocytosis models [2], whereas those of SNAREs come from neuroendocrine cell exocytosis [1]. At the core of this paradigm, Rabs promote the tethering—loose and reversible attachment—of the compartments that will fuse [3]. Subsequently, the assembly of heterotrimeric trans SNARE complexes brings about the docking—tight and irreversible attachment—of the fusing membranes [4]. Docking is followed by the opening and expansion of the fusion pore. In regulated exocytosis, this final stage requires an increase in cytoplasmic Ca2+ and the action of Ca2+ sensor proteins such as synaptotagmin [1]. After membrane fusion, SNAREs remain engaged in heterotrimeric cis complexes. Disassembly of the latter is achieved by the concerted action of α-SNAP and NSF, and is required to prepare SNAREs for subsequent rounds of fusion.
SNAREs are classified as R or Q based on the identity of a highly conserved residue [5]. Q-SNAREs and R-SNAREs contribute three and one helixes, respectively, to ternary complexes. When Q- and R-SNAREs reside on the same membrane, complexes are in a cis, fusion-incompetent configuration. In contrast, when Q- and R-SNAREs reside on opposite membranes, complexes are in a trans, fusion-competent configuration. In neurosecretory cells, exocytotic SNARE complexes are composed of syntaxin1A and a synaptosome-associated protein of 25 kD (SNAP25), which are two plasma membrane Q-SNAREs, and vesicle-associated membrane protein (VAMP) 2, which is a R-SNAREs found in secretory vesicles. These proteins are the target of botulinum and tetanus toxins, a set of highly specific zinc-dependent endoproteases [6]. In fact, the role of SNAREs in regulated exocytosis was unequivocally established thanks to the striking inhibitory effect of these neurotoxins on secretion [7]. Only when not assembled in tight complexes are SNAREs susceptible to cleavage [8], making these toxins excellent tools for the diagnosis of SNARE assembly status.
The acrosome is a large membrane-limited granule that overlies the nucleus of the mature spermatozoon [9]. Upon stimulation, sperm undergo exocytosis of this granule in a synchronized wave, with no recycling of components. Acrosomal exocytosis (AE) is an all-or-nothing event that comprises the opening of hundreds of fusion pores between the outer acrosomal membrane and the plasmalemma. AE depends on Rab3A, NSF/α-SNAP, and toxin-sensitive members of SNARE families [10–13]. It also requires an efflux of Ca2+ from inside the acrosome even in the presence of high cytosolic concentrations [14]. Concurrence of Rab- and toxin-sensitive, SNARE-dependent pathways is a hallmark of AE that makes it a unique mammalian model to study the different phases of the membrane fusion cascade. This feature is not found in other systems. Most exocytotic models either do not have a well defined role for, or are negatively regulated by, Rabs [1]. Likewise, systems in which Rabs are necessary for fusion typically contain toxin-insensitive SNARE isoforms.
By using a combination of neurotoxins and a photosensitive Ca2+ chelator, we show here that AE proceeds through a sequential set of events initiated when Rab3 is activated and triggers NSF/α-SNAP-dependent disassembly of cis SNARE complexes. SNAREs then reassociate in loose trans complexes until an efflux of intra-acrosomal Ca2+ promotes synaptotagmin- and SNARE-dependent membrane fusion.
Results
Rab3A Is Required before, and SNAREs and Synaptotagmin VI after, Intra-Acrosomal Ca2+ Efflux
O-nitrophenyl EGTA–acetoxymethyl ester (NP-EGTA-AM ) is a photolabile Ca2+ chelator that prevents inducer-triggered AE in sperm permeabilized by streptolysin-O (SLO); NP-EGTA-AM does this by entering the cytosol, diffusing through the outer acrosomal membrane, and accumulating inside the acrosome [14]. UV photolysis of NP-EGTA-AM rapidly replenishes the acrosomal Ca2+ pool, resuming exocytosis (Figure 1). In combination with AE inhibitors, NP-EGTA-AM helps to determine whether fusion-related factors are required before or after the intra-acrosomal Ca2+-sensitive step. Briefly, NP-EGTA-AM allows an AE inducer to prepare the fusion machinery up to the point when intra-acrosomal Ca2+ is required. Inhibitors are then added and the tubes illuminated. Resistance to inhibitors—reflected in unaffected exocytosis—implies that the targets of the fusion-related factors are required upstream of intra-acrosomal Ca2+ efflux. Sensitivity to inhibitors—revealed by blocked exocytosis—means their targets are located after the intra-acrosomal Ca2+-sensitive step (see Figure 1). AE is always prevented when the inhibitors are added prior to the inducer and maintained throughout the experiment. An anti-Rab3A antibody inhibited exocytosis when added before challenging with Ca2+ but not afterward. In contrast, antibodies against syntaxin1A, SNAP25, VAMP2, and synaptotagmin VI were able to abrogate exocytosis even when added after the inducer (Figure 1). These results indicate that Rab3A is necessary early in the fusion cascade, before Ca2+ is released from the acrosome, whereas SNAREs and synaptotagmin VI are required later, during or after the intra-acrosomal Ca2+ efflux. Similar experiments support an early role for NSF/α-SNAP [13]. These observations are summarized in Table 1.
Figure 1 Rab3A Is Required before, and SNAREs and Synaptotagmin VI after, Intra-Acrosomal Ca2+ Efflux
Permeabilized spermatozoa were loaded with 10 μM NP-EGTA-AM (NP) for 15 min at 37 °C to chelate intra-acrosomal Ca2+. AE was then initiated by adding 0.5 mM CaCl2 (10 μM free Ca2+)(Ca2+). After further 15 min incubation at 37 °C to allow exocytosis to proceed up to the intra-acrosomal Ca2+-sensitive step, sperm were treated for 15 min at 37 °C with antibodies that recognize Rab3A (20 μg/ml, anti-Rab3A), SNAP25 (20 μg/ml, anti-SNAP25), syntaxin1A (1/25 dilution, anti-Stx1A), VAMP2 (20 μg/ml, anti-VAMP2), or synaptotagmin VI (30 μg/ml, anti-StgVI). All these procedures were carried out in the dark. UV flash photolysis of the chelator was induced at the end of the incubation period (hν), and the samples were incubated for 5 min to promote exocytosis (NP→Ca2+→antibody→hν, black bars; a diagram of the experiment is shown at the top of the figure). Sperm were then fixed and AE was measured as described in Materials and Methods. Several controls were included (grey bars): background AE in the absence of any stimulation (control); AE stimulated by 10 μM free Ca2+ (Ca2+), inhibitory effect of NP-EGTA-AM in the dark (NP→Ca2+→dark), and the recovery upon illumination (NP→Ca2+→hν); and inhibitory effect of the antibodies when present throughout the experiment (NP→antibody→Ca2+→hν). The data were normalized as described in Materials and Methods (mean ± SEM). Statistical analysis is provided in Table S2.
Table 1 Factors Acting on Exocytosis before or after the Release of Ca2+ from the Acrosome
SNAREs Are Assembled in Neurotoxin-Resistant Complexes in Resting Spermatozoa
NSF and α-SNAP are proteins that catalyze the priming of the fusion machinery, disrupting the cis SNARE complex and activating individual SNAREs, and are required for AE [13]. This suggests that sperm SNAREs are initially assembled in inactive cis complexes. Such complexes are resistant to cleavage by neurotoxins [8].
To assess whether SNARE proteins are accessible to neurotoxins in unstimulated spermatozoa, permeabilized sperm were incubated with the light chains of botulinum neurotoxin (BoNT) E or B—BoNT/E cleaves SNAP25 and BoNT/B cleaves VAMP—or tetanus toxin (TeTx), which cleaves VAMP. Toxins were then inactivated by the specific zinc chelator N,N,N′,N′-tetrakis (2-pyridymethyl) ethylenediamine (TPEN) [15]. AE was subsequently stimulated with Ca2+ and secretion assessed. Exocytosis was not abrogated under these conditions (Figure 2A), suggesting that SNAP25 and VAMP2 are in a toxin-resistant configuration in resting spermatozoa.
Figure 2 SNAREs Are Assembled in Neurotoxin-Resistant Complexes in Resting Spermatozoa
(A) Permeabilized spermatozoa were treated at 37 °C for 15 min with 357 nM BoNT/E, 100 nM BoNT/B, or 100 nM TeTx. Next, 2.5 μM TPEN (see [B]) was added and AE was activated by adding 0.5 mM CaCl2 (10 μM free Ca2+) and the incubation continued for an additional 15 min (black bars). Sperm were then fixed and AE was measured as described in Materials and Methods. Several controls were included (grey bars): background AE in the absence of any stimulation (control); AE stimulated by 10 μM free Ca2+ (Ca2+); TPEN effect on exocytosis (TPEN→Ca2+); inhibitory effect of the neurotoxins on exocytosis (neurotoxin→Ca2+); and block of neurotoxin activity by TPEN (TPEN→neurotoxin→Ca2+).
(B) Recombinant SNAP25 (0.7 μg) was incubated for 15 min at 37 °C in the presence of 0.6 μg of BoNT/E and increasing concentrations of TPEN. Samples were then resolved by SDS-PAGE and stained with Coomassie blue. Molecular weight standards are indicated on the left (in kilodaltons). Densitometry and quantitation of the stained bands show 100%, 7%, 92%, 98%, 100%, and 100% of intact SNAP25 in lanes 1–6 (from left to right), respectively.
(C) Treatment with TeTx was performed as described in (A), in the presence of 310 nM NSF and 500 nM α-SNAP (NSF/αS) to promote SNARE complex dissociation (black bar). Incubation with NSF/α-SNAP in the presence of TPEN-inactivated toxin did not affect exocytosis (grey bar).
The data in (A) and (C) were normalized as described in Materials and Methods (mean ± SEM). Statistical analysis is provided in Table S3.
To rule out the possibility that permeabilization would be artifactually responsible for cis SNARE complex formation in human sperm, we conducted experiments similar to those depicted in Figure 2A, but in intact human sperm. Because sperm were not permeabilized, holotoxins were used instead of the isolated light chains. These clostridial neurotoxins consist of a heavy subunit, responsible for binding to plasma membrane receptors (in particular gangliosides GT1b and GQ1b), and a light subunit that carries the proteolytic activity [6]. Heavy-chain-mediated binding is required for internalization of the light chain, which acts in the cytosol [6]. Because both GT1b and GQ1b gangliosides have been isolated from sperm preparations [16] we expected non-permeabilized sperm to be at least partially sensitive to holotoxins. Indeed, when intact (i.e., non-permeabilized) human sperm were exposed to BoNT/A (SNAP25-specific) and BoNT/F (VAMP-specific) before triggering exocytosis with the Ca2+ ionophore A23187, an approximately 50% inhibition in AE was observed (Figure S1). Toxin concentrations were 5-fold higher than those required to inhibit exocytosis in permeabilized sperm [12] to allow for the poor internalization of the catalytic subunits. Once again, we found that the SNARE proteins were not accessible to neurotoxins in unstimulated cells, since exocytosis was not attenuated when toxins were inactivated by TPEN before challenging with A23187 (Figure S1), indicating that SNAP25 and VAMP were in a toxin-resistant configuration prior to initiating AE.
To show directly that TPEN blocks the proteolytic activity of neurotoxins, recombinant SNAP25 was incubated with BoNT/E in the absence or presence of increasing concentrations of the zinc chelator. As shown in Figure 2B, BoNT/E cleaved SNAP25. TPEN completely inhibited its activity at concentrations of 1 μM and higher, in total agreement with the functional data.
These results show that, in the absence of sperm stimulation, Q- and R-SNAREs are protected from neurotoxin cleavage, not adopting a toxin-sensitive state during the 15-min incubation protocol. These results were similar in permeabilized and intact sperm. Thus, we conclude that SNAREs are engaged in a toxin-resistant configuration in resting sperm. An attractive possibility is that they are locked in ternary cis complexes rather than undergoing cyclic assembly and disassembly like SNAREs in other systems. If this is the case, addition of an excess of recombinant NSF/α-SNAP should force disassembly, rendering the SNAREs toxin-sensitive. We therefore incubated sperm with NSF/α-SNAP in the presence of TeTx. The toxin was subsequently inactivated by addition of TPEN, and AE was stimulated. Ca2+ failed to elicit exocytosis, indicating that NSF/α-SNAP rendered VAMP sensitive to toxin cleavage (Figure 2C). In conclusion, our data indicate that, in resting sperm, SNAREs are in toxin-resistant cis complexes that can be disassembled by NSF/α-SNAP.
Rab3A Triggers the Disassembly of SNARE Complexes before Intra-Acrosomal Ca2+ Efflux
In our permeabilized sperm model, exocytosis is achieved when an increase in the cytosolic Ca2+ concentration promotes the activation of Rab3A [10]. Alternatively, AE can be initiated by adding recombinant Rab3A preloaded with GTP-γ-S [11]. Rab3A-triggered exocytosis is sensitive to neurotoxins [12]. Therefore, at some point after Rab3A activation, SNAREs must pass from a toxin-resistant to a toxin-sensitive state. This transition may occur before or after the release of Ca2+ from the acrosome. To distinguish between these two possibilities, intra-acrosomal Ca2+ was chelated with NP-EGTA-AM and AE was initiated with Rab3A in the presence of BoNT/E. The toxin was then inactivated by TPEN and intra-acrosomal Ca2+ replenished by NP-EGTA-AM photolysis. As shown in Figure 3, exocytosis was efficiently inhibited, indicating that SNAREs were disassembled—because SNAP25 was cleaved—after Rab3A activation and before the intra-acrosomal Ca2+ efflux. In brief, Rab3A—and not the efflux of intra-acrosomal Ca2+—elicited SNARE complex disassembly. It is worth noting that disassembly appears to be Ca2+-independent, since it occurred at a low Ca2+ concentration in the medium (approximately 100 nM [14]).
Figure 3 Activation with Rab3A Triggers the Disassembly of SNARE Complexes before Intra-Acrosomal Ca2+ Efflux
Permeabilized spermatozoa were loaded with 10 μM NP-EGTA-AM (NP) for 15 min at 37 °C to chelate intra-acrosomal Ca2+. Then 357 nM BoNT/E was added to the system, and AE was initiated by adding 300 nM GTP-γ-S-loaded Rab3A. After 15 min at 37 °C, the toxin was inactivated with 2.5 μM TPEN, and photolysis of NP was induced by UV illumination (hν). The samples were incubated for a further 5 min to promote exocytosis (NP→BoNT/E→Rab3A→TPEN→hν, black bar). At the end of the incubation, sperm were fixed and AE was measured as described in Materials and Methods. Several controls are included (grey bars): background AE in the absence of any stimulation (control); AE stimulated by 10 μM free Ca2+ (Ca2+) or Rab3A (Rab3A); blockage of Rab3A-triggered exocytosis by BoNT/E (BoNT/E→Rab3A); inhibitory effect of NP-EGTA-AM in the dark (NP→Rab3A→dark) and the recovery upon illumination (NP→Rab3A→hν); and inactivation of the neurotoxin by TPEN (NP→TPEN→BoNT/E→Rab3A→hν). The data were normalized as described in Materials and Methods (mean ± SEM). Statistical analysis is provided in Table S4.
Immunofluorescence Experiments Indicate That SNAREs Are Disassembled by Sperm Activation
Neurotoxins are proteases; hence, their effect on sperm SNAREs should be reflected by immunostaining, provided that anti-SNARE antibodies recognizing the removed portions are used as probes. BoNT/C was selected for these experiments for the following reasons: (i) it cuts syntaxin1A near the transmembrane domain, eliminating most of the cytoplasmic region [6]; (ii) a mutant without catalytic activity is available (BoNT/C–E230A [EA], T. Binz, personal communication); and (iii) immunolabeling with a polyclonal antibody that recognizes the region of the protein cleaved by the toxin is highly reproducible. When BoNT/C was introduced into our assay, substantial inhibition of exocytosis was observed at low concentrations of the toxin (Figure 4). As expected, the protease-null mutant did not affect AE.
Figure 4 Syntaxin1A Is Assembled in Toxin-Resistant Complexes That Are Disassembled by NSF/α-SNAP or by Sperm Activation
(A) Permeabilized spermatozoa were incubated for 15 min at 37 °C with increasing concentrations of BoNT/C (black circles, wild type; grey circles, EA, a protease-inactive mutant) and then stimulated with 10 μM Ca2+ for 15 min at 37 °C. Afterwards, sperm were fixed and AE measured as described in Materials and Methods.
(B) To assess the assembly state of syntaxin1A, sperm were incubated with 100 nM BoNT/C (15 min at 37 °C), and the cells were then fixed and immunostained with an anti-syntaxin1A antibody recognizing an epitope that is cleaved by the toxin. To prevent AE, which would release syntaxin into the medium by vesiculation of the acrosome, intra-acrosomal Ca2+ was chelated with 10 μM BAPTA-AM (15 min at 37 °C, B-AM). The toxin treatment in resting sperm (BoNT/C) or B-AM-loaded sperm (B-AM→BoNT/C) had no effect on the syntaxin labeling compared to untreated sperm (control). However, when 310 nM NSF and 500 nM α-SNAP were added to the system to promote the disassembly of SNARE complexes, the toxin significantly decreased the syntaxin labeling (B-AM→NSF/αS→BoNT/C). The BoNT/C treatment also affected syntaxin labeling when sperm were stimulated for 15 min at 37 °C with 10 μM free Ca2+ (B-AM→ BoNT/C→Ca2+) or 300 nM Rab3A (B-AM→BoNT/C→Rab3A). The protease-inactive mutant did not affect labeling under these conditions (B-AM→BoNT/C-EA→Ca2+). Fluorescence was normalized as described in Materials and Methods. The data represent the mean ± SEM. Statistical analysis is provided in Table S5.
For immunofluorescence experiments, sperm were incubated under different conditions with or without BoNT/C. Cells were fixed and double-labeled with fluorescein isothiocyanate (FITC)–Pisum sativum agglutinin (PSA)—to distinguish between reacted and intact spermatozoa—and with an anti-syntaxin1A antibody (Figure 5). Under control conditions (incubation without BoNT/C), clear immunolabeling was observed in the acrosomal region of most cells (Figure 5A). This pattern is expected for a protein that participates in sperm exocytosis. Notice that spontaneously reacted sperm (not stained by FITC-PSA) did not exhibit syntaxin staining (Figure 5D and 5E). Addition of BoNT/C did not decrease the immunofluorescence labeling with the anti-syntaxin1A antibody (Figures 4B and 5D), indicating that this SNARE was protected from toxin cleavage in resting spermatozoa. Addition of NSF/α-SNAP in the presence of the toxin caused a significant decrease in the percentage of syntaxin-labeled cells (Figure 4B). Because these proteins disassemble SNARE complexes, their ability to confer BoNT/C vulnerability—marked by diminished immunolabeling—means that syntaxin is engaged in resistant complexes in untreated sperm. To verify that sperm activation triggers SNARE disassembly in a step previous to the Ca2+ efflux from the intra-acrosomal pool, this store was depleted with BAPTA-AM, and sperm exocytosis stimulated with Ca2+ or Rab3A in the presence of BoNT/C. Under these conditions, the percentage of sperm with acrosomal syntaxin1A labeling decreased significantly, indicating that the protein became toxin-sensitive after sperm activation and before the efflux of intra-acrosomal Ca2+ (Figures 4B and 5G). In contrast, stimulation in the presence of catalytically inactive BoNT/C caused no decrease in the percentage of labeled spermatozoa (Figures 4B and 5J), ruling out the possibility of an effect unrelated to the proteolytic activity of the toxin. FITC-PSA staining was not affected by the different conditions, indicating that the decrease in syntaxin1A immunolabeling was not due to AE that would have released the outer acrosomal membrane together with the apposed plasma membrane (Figure 5).
Figure 5 Effect of BoNT/C on Syntaxin1A Immunofluorescence
Sperm were incubated with 100 nM BoNT/C (15 min at 37 °C) as explained in Figure 4. The cells were then fixed and triple-stained with an anti-syntaxin1A antibody that recognizes an epitope trimmed by the toxin (red; [A, D, G, and J]), FITC-PSA to differentiate between reacted and intact sperm (green; [B, E, H, and K]), and Hoechst 33258 to visualize all cells in the field (blue; [C, F, I, and L]). Notice that spontaneously reacted sperm were negative for syntaxin1A staining (arrowheads in [D] and [E]). BoNT/C had no effect on resting sperm (compare [A–C] with [D–F]). However, labeling in sperm stimulated with 10 μM Ca2+ in the presence of BAPTA-AM to prevent exocytosis (observe that PSA staining is not affected) was significantly reduced by the toxin (asterisks, [G]). In contrast, the same experimental condition in the presence of the protease-inactive toxin (BoNT/C-EA) had no effect (J–L). Bars = 5 μm.
These results reaffirm the notion that SNAREs are toxin-protected in resting spermatozoa and that addition of NSF/α-SNAP unprotects them. Furthermore, activation of spermatozoa (with either Ca2+ or Rab3A) under conditions that deplete intra-acrosomal Ca2+ promotes their cleavage by BoNTs, indicating that cis SNARE complexes are disassembled at a step prior to the intra-acrosomal Ca2+ efflux.
SNAREs Are Not Engaged in Tight Complexes before the Efflux of Intra-Acrosomal Ca2+
Upon sperm stimulation, cis SNARE complexes disassemble and pass through a toxin-sensitive configuration. The next question we answered was whether they reassemble into toxin-resistant complexes before Ca2+ efflux from the acrosome triggers the final steps of exocytosis. For these experiments, sperm were stimulated with Ca2+ or Rab3A in the presence of NP-EGTA-AM. BoNT/E (or BoNT/C) was then added to the assay to cleave unprotected SNAP25 (or syntaxin). Finally, intra-acrosomal Ca2+ was replenished by photolysis of the chelator. BoNT/E inhibited both Ca2+- and Rab3A-triggered exocytosis even when added after AE had progressed to the intra-acrosomal Ca2+-sensitive step (Figure 6). A similar result was observed with BoNT/C (Figure 6). We conclude that SNAREs were not protected before the efflux of intra-acrosomal Ca2+; hence, they may remain as monomers after they are disassembled following sperm activation. Alternatively, they may reassemble partially in loose trans complexes that also are sensitive to neurotoxins. These complexes have been postulated in secretory vesicles and granules already attached to the plasma membrane. In the loose complexes Q- and R-SNAREs are contributed by different membranes but are not tightly packed. Hence, they can be cleaved by most neurotoxins, including BoNT/E and BoNT/C [17,18].
Figure 6 SNAREs Do Not Reassemble in Tight Complexes before the Efflux of Intra-Acrosomal Ca2+
Permeabilized spermatozoa were loaded with 10 μM NP-EGTA-AM (NP) for 15 min at 37 °C to chelate intra-acrosomal Ca2+. AE was then initiated by adding 10 μM free Ca2+ (Ca2+) or 300 nM Rab3A (Rab3A). After 15 min incubation at 37 °C to allow exocytosis to proceed to the intra-acrosomal Ca2+-sensitive step, neurotoxins recognizing SNAP25 (BoNT/E) and syntaxin (BoNT/C) were added to the tubes to assess whether the SNAREs had reassembled into toxin-resistant complexes. Intra-acrosomal Ca2+ was replenished by photolysis of NP (hν), and samples were incubated for 5 min to promote exocytosis (NP→Ca2+/Rab3A→neurotoxins→hν, black bars). Sperm were then fixed and AE measured as described in Materials and Methods. Several controls are included (grey bars): background AE in the absence of any stimulation (control); AE stimulated by 10 μM free Ca2+ (Ca2+) or Rab3A (Rab3A); inhibitory effect of NP-EGTA-AM in the dark (NP→Ca2+/Rab3A→dark) and recovery upon illumination (NP→Ca2+/Rab3A→hν); and AE inhibition when neurotoxins were present throughout the incubation (NP→neurotoxins→Ca2+/Rab3A→hν). The data were normalized as described in Materials and Methods (mean ± SEM). Statistical analysis is provided in Table S6.
SNAREs Are Forming Loose trans Complexes before the Efflux of Intra-Acrosomal Ca2+
Following cis complex disassembly during sperm activation, SNAREs remain toxin-sensitive until Ca2+ efflux from the acrosome triggers the last steps of AE. Both monomeric and heterotrimeric loose trans configurations are susceptible to toxin cleavage. To distinguish between the two, we used recombinant soluble (unpalmitoylated) SNAP25. This protein efficiently inhibits AE, most likely because it competes with endogenous, membrane-associated SNAP25, for the formation of productive SNARE complexes [12]. We reasoned that recombinant SNAP25 would be able to compete with the endogenous protein when monomeric, but not when engaged in complexes. Sperm were stimulated with Ca2+ in the presence of NP-EGTA-AM, recombinant SNAP25 was added, and intra-acrosomal Ca2+ replenished by UV photolysis. Under these conditions, recombinant SNAP25 was not able to inhibit exocytosis (Figure 7A). These observations suggest that after stimulation and before the Ca2+ efflux from the acrosome, endogenous SNAREs are engaged in SNARE complexes.
Figure 7 SNAREs Reassemble in Loose Complexes That Are Resistant to NSF/α-SNAP before the Efflux of Intra-Acrosomal Ca2+
(A) Permeabilized spermatozoa were loaded with 10 μM NP-EGTA-AM (NP) for 15 min at 37 °C to chelate intra-acrosomal Ca2+. AE was then initiated by adding 10 μM free Ca2+ (Ca2+). After 15 min incubation at 37 °C to allow exocytosis to proceed to the intra-acrosomal Ca2+-sensitive step, 800 nM recombinant SNAP25 (SNAP25) was added to compete with endogenous SNAP25. Intra-acrosomal Ca2+ was replenished by photolysis of NP-EGTA-AM (hν), and the samples were incubated for 5 min to promote exocytosis (NP→Ca2+→SNAP25→hν, black bar). Sperm were then fixed and AE was measured as described in Materials and Methods.
(B) Permeabilized spermatozoa were loaded with 10 μM NP-EGTA-AM (NP) for 15 min at 37 °C. AE was then initiated by adding 10 μM free Ca2+ (Ca2+) or 300 nM Rab3A (Rab3A). After 15 min incubation at 37 °C, 100 nM neurotoxin recognizing VAMP (BoNT/B and TeTx) was added to the tubes to assess whether the SNAREs had reassembled in loose trans complexes sensitive to BoNT/B but not to TeTx. After 15 min incubation at 37 °C, intra-acrosomal Ca2+ was replenished by photolysis of NP-EGTA-AM (hν), and the samples were incubated for 5 min to promote exocytosis (NP→Ca2+/Rab3A→neurotoxin→hν, black bars). Sperm were then fixed and AE measured as described in Materials and Methods.
(C) To assess whether NSF/α-SNAP can disassemble loose trans SNARE complexes, permeabilized sperm treated as in (B) were incubated with TeTx in the presence of 310 nM NSF and 500 nM α-SNAP (NP→Ca2+/Rab3A→NSF/αS+TeTx→hν, black bars).
Several controls were included in (A), (B), and (C) (grey bars): background AE in the absence of any stimulation (control); AE stimulated by 10 μM free Ca2+ (Ca2+) or 300 nM Rab3A (Rab3A); inhibitory effect of NP-EGTA-AM in the dark (NP→Ca2+/Rab3A→dark) and the recovery upon illumination (NP→Ca2+/Rab3A→hν); inhibitory effect when SNAP25 was present throughout the incubations (NP→SNAP25→Ca2+→hν); inhibitory effect when the neurotoxins were present throughout the incubations (NP→neurotoxin→Ca2+/Rab3A→hν); and the effect of NSF/α-SNAP on SNARE complexes in unstimulated sperm (NSF/αS+TeTx→TPEN→Rab3A→hν). The data were normalized as described in Materials and Methods (mean ± SEM). Statistical analysis is provided in Table S7.
Differential sensitivity to BoNT/B and TeTx constitutes an independent approach to distinguish between monomeric VAMP and that engaged in loose trans complexes. These toxins cleave the same peptide bond, exposed in both configurations [6]. Interestingly, TeTx binds to the N-terminus of the VAMP coil domain whereas BoNT/B binds the C-terminus. Since SNARE complex assembly begins at the N-terminus, the TeTx-recognition site is hidden in loose SNARE complexes while the BoNT/B recognition site is exposed. In other words, TeTx can only cleave monomeric VAMP while BoNT/B also cuts VAMP loosely assembled in SNARE complexes [19]. As shown in Figure 7B, BoNT/B—but not TeTx—was capable of inhibiting AE when the system was stimulated with Ca2+ and allowed to reach the intra-acrosomal Ca2+-sensitive step. To assess whether loose trans complex assembly takes place in the absence of added Ca2+, Rab3A was used as an inducer in the presence of EGTA. Again, both toxins blocked exocytosis when present from the beginning of the experiment, but only BoNT/B inhibited when added after stimulation (Figure 7B). These data suggest that assembly of trans SNARE complexes is Ca2+-independent.
The fact that exocytosis was accomplished in the constant presence of exogenous SNAP25 and TeTx (Figure 7A and 7B) indicates that trans SNARE complexes are not disassembled after intra-acrosomal Ca2+ efflux. We hypothesized that after sperm stimulation, SNAREs remain irreversibly engaged in loose trans complexes as long as intra-acrosomal Ca2+ is kept unavailable. As shown in Figure 2C, recombinant NSF/α-SNAP can disassemble cis SNARE complexes in unstimulated spermatozoa. We asked whether they could also disengage loose trans complexes. To this end, NP-EGTA-AM-loaded sperm were stimulated with Ca2+ or Rab3A and then treated with TeTx in the presence or absence of NSF/α-SNAP. The results, presented in Figure 7C, show that these chaperones failed to confer TeTx sensitivity when added after the exocytic process had progressed to the intra-acrosomal Ca2+-sensitive step. These observations indicate that trans SNARE complexes cannot be disassembled by NSF/α-SNAP. The notion that trans SNARE complexes are resistant to NSF/α-SNAP opposes the original SNARE hypothesis [20] and, although suggested in the past [21], has to our knowledge never before been shown in actual cells.
Immunofluorescence Experiments Show That VAMP2 Is Engaged in Loose trans SNARE Complexes after Stimulation
As shown in Figures 4B and 5, immunofluorescence is a powerful technique to monitor SNARE assembly status as reflected by SNARE sensitivity to cleavage by neurotoxins. We thus made use of the differential sensitivity of VAMP2 to TeTx and BoNT/B when engaged in loose complexes to show by VAMP2 immunostaining that these complexes form after sperm stimulation. BAPTA-AM was used in all cases to chelate intra-acrosomal Ca2+ and prevent VAMP2-containing membrane loss inherent to the acrosome reaction. As expected, acrosomal labeling for VAMP2 was not affected by incubating unstimulated sperm with TeTx or BoNT/B (Figures 8 and 9), consistent with the notion that VAMP2 is protected in cis SNARE complexes. A significant decrease in the percentage of cells exhibiting acrosomal staining was observed when toxin-loaded sperm were challenged with Ca2+ (Figures 8, 9G, and 9P). Thus, VAMP2 sensitization to cleavage was probably a consequence of cis SNARE disassembly by the stimulant. In contrast, when toxins were added after Ca2+ (i.e., after AE had progressed to the intra-acrosomal Ca2+-sensitive step), VAMP2 labeling was attenuated by BoNT/B (Figures 8 and 9J) but not by TeTx (Figures 8 and 9S). Loss of VAMP2 immunostaining following BoNT/B treatment is due to lack of immunostaining of the cleavage product (aa 1–76) by the anti-VAMP2 antibody [22]. This pattern of VAMP2 sensitivity to BoNT/B coupled to resistance to TeTx implies its engagement in loose trans SNARE complexes. In conclusion, immunofluorescence data further support the idea that, following sperm stimulation, SNAREs are engaged in loose trans complexes awaiting the release of intra-acrosomal Ca2+ that will trigger the final steps of AE.
Figure 8 VAMP2 Is Engaged in Loose SNARE Complexes before the Efflux of Intra-Acrosomal Ca2+
Permeabilized spermatozoa were loaded with 10 μM BAPTA-AM (B-AM) for 15 min at 37 °C to chelate intra-acrosomal Ca2+. AE was then initiated by adding 10 μM free Ca2+ (Ca2+). After 15 min incubation at 37 °C to allow exocytosis to proceed to the intra-acrosomal Ca2+-sensitive step, 100 nM neurotoxins recognizing VAMP (BoNT/B or TeTx) were added to the tubes and the samples were incubated for 15 min at 37 °C (B-AM→Ca2+→neurotoxin, black bars). Samples were then immunolabeled with an anti-VAMP2 antibody as described in Materials and Methods. Notice that at this stage VAMP2 immunolabeling was sensitive to BoNT/B but not to TeTx. Several other conditions are included (grey bars). The toxins did not affect VAMP2 staining in resting sperm (compare control versus B-AM→neurotoxin). However, the toxins decreased the VAMP2 labeling when present during stimulation (B-AM→neurotoxin→Ca2+). Fluorescence was normalized as described in Materials and Methods (mean ± SEM). Statistical analysis is provided in Table S8.
Figure 9 Effect of BoNT/B and TeTx on VAMP2 Immunofluorescence
Sperm were incubated with 100 nM BoNT/B or TeTx (15 min at 37 °C) as described in Figure 8. The cells were then triple-stained with an anti-VAMP2 antibody that recognizes an epitope that is cleaved by the toxin (red; [A, D, G, J, M, P, and S]), FITC-PSA to differentiate between reacted and intact sperm (green; [B, E, H, K, N, Q, and T]), and Hoechst 33258 to visualize all cells in the field (blue; [C, F, I, L, O, R, and U]). BoNT/B and TeTx had no effect on resting sperm (compare [D–F] and [M–O] with [A–C]). However, labeling in sperm stimulated with 10 μM Ca2+ in the presence of BAPTA-AM to prevent exocytosis (observe that the PSA staining is not affected) was significantly reduced by the toxins (asterisks, [G] and [P]). In contrast, when cells were first allowed to arrive at the intra-acrosomal Ca2+-sensitive step and then treated with toxins, BoNT/B caused a significant decrease of the VAMP2 label (asterisks, [J]), whereas TeTx had no effect (S). Bars = 5 μm.
Discussion
Exocytosis of the acrosome is a synchronized, all-or-nothing process that happens only once in the life of the spermatozoon and depends on both Rab3 activation and neurotoxin-sensitive SNAREs. Because of these special features, it constitutes a particularly attractive system to examine molecular aspects of regulated exocytosis that are not amenable to experimental manipulation in other mammalian models. These include the coupling between Rab and SNARE functions, the properties of loose trans SNARE complexes, the role of synaptotagmin in the dynamics of SNARE assembly, and the role of intravesicular Ca2+ in membrane fusion. Furthermore, AE is a central process in sperm physiology, and understanding the molecular mechanisms underlying it will be of outstanding importance for our ability to regulate fertilization. Sperm contact with glycoproteins in the zona pellucida of the oocyte leads to the opening of store-operated Ca2+ channels in the sperm plasmalemma followed by a massive entry of Ca2+ from the medium into the cytoplasm, leading to exocytosis [23]. Our SLO-permeabilized sperm model, in which Ca2+ can freely permeate into the cells, resembles the state of open store-operated Ca2+ channels in intact cells and is therefore suitable to study stages of exocytosis occurring downstream of store-operated Ca2+ channel opening.
SNAREs are required in multiple fusion events mandatory for cell survival even under resting conditions. The ratio of monomeric to assembled SNAREs depends on the type and physiological condition of the cell. Thus, while some studies indicate that most SNAREs are free in the plasma membrane [24], others suggest that they are engaged in complexes [25,26]. SNAP25 and syntaxins can form partial all-Q complexes (three helix bundles from one SNAP25 and one syntaxin; [25]) or complexes composed of four helix bundles from two syntaxins and one SNAP25 [27]. Pre-association of Q-SNAREs in a three-bundle complex creates the docking site for the cognate R-SNARE [28]. Unlike their ternary counterparts, binary complexes are unstable and sensitive to neurotoxins [25]. Whatever the steady-state configuration of SNAREs in neuroendocrine cells might be, exocytosis is blocked by neurotoxins, suggesting that SNAREs go through toxin-sensitive stages [18,29]. In resting human sperm, both R- and Q-SNAREs are protected from toxin cleavage. Susceptibility to toxins is conferred by both endogenous or exogenously added NSF/α-SNAP. Thus, we conclude that sperm SNAREs are locked in ternary complexes in a cis configuration, in contrast to cells with active vesicular recycling. Because sperm have only one chance to fertilize the oocyte, tight spatial and temporal regulation of the acrosome reaction is a prerequisite for their success. In this scenario, it is not surprising that SNAREs are bound in an inactive state until exocytosis is triggered.
The connection between the rise in cytosolic Ca2+ occurring upon sperm stimulation and Rab3A activation is not clear. A calmodulin-mediated effect is possible, since Ca2+/calmodulin binds Rab3A [30] and promotes the exchange of GDP for GTP in the Rab3A–GDP dissociation inhibitor complex [31]. Although we have shown that calmodulin has an inhibitory effect on AE in permeabilized sperm and that this effect is not mediated by binding to Rab3A [32], our results do not exclude a role for calmodulin in Rab3A activation. Once Rab3A is activated, cis SNAREs are disassembled by NSF/α-SNAP and become accessible to neurotoxins. In other systems, active Rabs recruit a variety of effectors that tether the membranes that will fuse. NSF has been found associated with tethering complexes [33]. Moreover, NSF binds several Rabs, including Rab3A [34]. Perhaps similar interactions favor the dissociation of cis SNAREs by sperm NSF/α-SNAP following Rab3A activation. The proximity of the tethered membranes plus the availability of free Q- and R-SNAREs would allow their association in loose trans complexes. The latter can form at the very low Ca2+ concentrations achieved by EGTA and NP-EGTA-AM chelating Ca2+ in the medium/cytosol and in the acrosome, respectively. While not required for the assembly of SNARE complexes from pure proteins [27], Ca2+ appears to be necessary for trans SNARE pairing during exocytosis in PC12 cells [35]. Perhaps this incongruity is due to the presence of specific regulatory proteins in these cells. We have shown that trans complexes are resistant to NSF/α-SNAP, in agreement with results obtained in a proteoliposome fusion assay [21]. In contrast, trans SNAREs can be dissociated by the NSF/α-SNAP homologs Sec18p/Sec17p in a yeast vacuole fusion assay [36]. The reasons for this discrepancy are unknown, but it is worth mentioning that both the models (yeast vacuole fusion versus mammalian regulated exocytosis) and the experimental readout (coimmunoprecipitation of SNAREs versus toxin sensitivity) are quite different. Resistance to NSF/α-SNAP, SNAP25, and TeTx indicates that, following sperm stimulation, SNAREs are engaged in trans complexes that do not spontaneously revert to the monomeric configuration from which such complexes arose. Temporally, the arrest of exocytosis until Ca2+ is released from the acrosome correlates with SNAREs being assembled in loose trans complexes. Resistance to NSF/α-SNAP, SNAP25, and TeTx might also be explained if SNAREs were dispensable for the final fusion steps, as has been suggested for yeast vacuole fusion and exocytosis of cortical granules of sea urchin oocytes [36,37]. Our data do not support this view, however, since BoNTs and antibodies to SNAREs continue to prevent exocytosis downstream of intra-acrosomal Ca2+ release. This indicates that SNAREs are required late in the exocytotic cascade.
A direct role for intravesicular Ca2+ in membrane fusion has been proposed in several transport events, including the exocytosis of secretory granules [38,39]. Why would Ca2+ efflux from the acrosome be necessary for exocytosis? We favor the idea that tight apposition of the membranes near the forming fusion pore prevents free accessibility to cytosolic Ca2+. Thus, a local release from the acrosome is necessary to activate the final steps of membrane fusion. Synaptotagmins are likely involved in this late Ca2+-sensitive step. Synaptotagmin VI, at least, is present in the acrosomal membrane of human sperm, and this protein is required at a step downstream of the intra-acrosomal Ca2+ efflux (Figure 1 and [40]). Therefore, synaptotagmin VI/Ca2+ may favor the full zippering of SNARE complexes and thus enable membrane fusion [41].
Our data fit the working model depicted in Figure 10, in which, for simplicity, only SNAREs are shown. Initially, SNAREs are locked in inactive cis complexes on plasma and outer acrosomal membranes. Rab3A is activated upon Ca2+ entrance into the cytoplasm, triggering the tethering of the acrosome to the plasma membrane. Next, NSF/α-SNAP disassemble cis SNARE complexes on both membranes. Monomeric SNAREs are free to assemble in loose trans complexes, causing the irreversible docking of the acrosome to the plasma membrane. At this point, Ca2+ is released from inside the acrosome through inositol 1,4,5-trisphosphate–sensitive Ca2+ channels to trigger the final steps of membrane fusion, which require SNAREs (presumably in tight trans complexes) and synaptotagmin.
Figure 10 Working Model for the Dynamics of SNARE Assembly and Disassembly during AE
The resistance to neurotoxin proteolysis is indicated as determined experimentally here. The block by intra-acrosomal Ca2+ chelators is marked in red. OAM, outer acrosomal membrane; PM, plasma membrane. See text for more details (SNARE drawings were modified from [4]).
Materials and Methods
Reagents
SLO was obtained from Corgenix (Peterborough, United Kingdom). Gamete preparation medium (Serono, Aubonne, Switzerland) was used to culture spermatozoa. TPEN, NP-EGTA-AM, Hoechst 33258, and BAPTA-AM were purchased from Molecular Probes (Eugene, Oregon, United States). Anti-VAMP2 (mouse monoclonal, clone 69.1, purified IgG), and anti-syntaxin1A (rabbit polyclonal, whole serum) were from Synaptic Systems (Göttingen, Germany). Anti-Rab3A (rabbit polyclonal, purified IgG) was from Santa Cruz Biotechnology (Santa Cruz, California, United States). Anti-SNAP25 (mouse monoclonal, clone SP12, purified IgG) was from Stress Gen (Victoria, British Columbia, Canada). The anti–synaptotagmin VI (rabbit polyclonal, affinity purified) has been previously described [42]. TRITC-conjugated goat anti-mouse and anti-rabbit IgG were from Kirkegaard and Perry Laboratories (Gaithersburg, Maryland, United States). All other chemicals were analytical-grade and were purchased from Sigma Chemical (St. Louis, Missouri, United States) or ICN Biochemicals (Aurora, Ohio, United States).
Recombinant proteins
Recombinant SNAP25-His6 was generously provided by U. Matti (Physiologisches Institut, Universität des Saarlandes, Homburg, Germany). Plasmids encoding His6-α-SNAP and His6-NSF in pQE9 (Qiagen, Valencia, California, United States) were a kind gift from S. Whiteheart (University of Kentucky, Lexington, Kentucky, United States). Plasmids encoding the light chain of BoNT/C, BoNT/C-EA, BoNT/B, and TeTx in pQE3 (Qiagen), and BoNT/E in pQE9 were generously provided by T. Binz (Medizinische Hochschule Hannover, Hannover, Germany). DNA encoding His6-α-SNAP was transformed into Escherichia coli XL1-Blue (Stratagene, La Jolla, California, United States) and induced overnight at 20 °C with 0.2 mM IPTG. The same protocol was used for the expression of BoNT/E, BoNT/C, BoNT/C-EA, and TeTx. Plasmid constructs encoding NSF and BoNT/B were transformed into E. coli M15pRep4 (Qiagen) and induced 4 h at 30 °C with 1 mM IPTG. Purification of recombinant proteins was accomplished according to [43], except that 0.5 mM ATP, 5 mM MgCl2, and 2 mM DTT were added to all buffers involved in the purification of His6-NSF. The expression plasmid pGEX2T containing the cDNA-encoding human Rab3A was generously provided by M. Colombo and P. Stahl (Washington University, St. Louis, Missouri, United States). GST-Rab3A was expressed in E. coli XL1-Blue, purified, prenylated, and loaded with guanosine 5′-O-3-thiotriphosphate following standard procedures [11].
AE in permeabilized sperm
Human semen samples were obtained from normal healthy donors, after at least 2 d of abstinence. Highly motile sperm were recovered following a swim-up separation for 1 h in gamete preparation medium at 37 °C in an atmosphere of 5% CO2/95% air. Concentration was adjusted to 5–10 × 106/ml, and incubation proceeded for at least 2 h under conditions that supported capacitation (gamete preparation medium, 37 °C, 5% CO2/95% air). Permeabilization was accomplished as described [11]. Briefly, washed spermatozoa were resuspended in cold PBS containing 0.4 U/ml SLO for 15 min at 4 °C. Cells were washed once with PBS, resuspended in ice-cold sucrose buffer (250 mM sucrose, 0.5 mM EGTA, 20 mM Hepes-K [pH 7]) containing 2 mM DTT. Sperm were incubated and spotted on eight-well slides, air-dried, and fixed/permeabilized in ice-cold methanol for 30 s. Acrosomal status was evaluated by staining with FITC-coupled PSA according to [44]. At least 200 cells were scored using a Nikon (Tokyo, Japan) microscope equipped with epifluorescence optics. Negative (no stimulation) and positive (10 μM free Ca2+) controls were included in all experiments. For each experiment, the data were normalized by subtracting the number of reacted spermatozoa in the negative control (range 18%–30%) from all values, and expressing the resulting values as a percentage of the acrosome reaction observed in the positive control (range 30%–50%).
Indirect immunofluorescence in SLO-permeabilized sperm
Sperm were permeabilized with SLO and treated as indicated in the figure legends. For syntaxin1A immunofluorescence, sperm were spotted on round cover slips and fixed/permeabilized in 2% paraformaldehyde–0.1% Triton X-100 in PBS for 10 min at room temperature. For VAMP2 immunofluorescence, 20 μg/ml anti-VAMP2 antibody was added to the sperm suspension and incubated for 10 min at 37 °C before spotting. After fixation, sperm were incubated in 50 mM glycine-PBS for 30 min at 20 °C and 1 h in 5% BSA-PBS containing 0.4% polyvinylpyrrolidone (PVP) (40,000 average MW). Cells were labeled with an anti-syntaxin antibody (overnight at 4 °C, 1/50 whole serum in 3% BSA-PBS/PVP), followed by a TRITC-labeled anti-rabbit IgG (1 h at 20 °C, 10 μg/ml in 0.5% BSA-PBS/PVP). TRITC-labeled anti mouse IgG was used when anti-VAMP2 was the primary antibody. Cover slips were washed (3×) with PBS/PVP between incubations. Finally, cells were incubated 1 min in cold methanol and stained with Hoechst 33258 (20 min at 20 °C, 1 μg/ml in PBS) followed by FITC-PSA (30 min at 20 °C, 50 μg/ml in PBS), and washed with distilled water 20 min at 4 °C. Cover slips were mounted in Gelvatol, and examined with an Eclipse TE3000 Nikon microscope equipped with a Plan Apo 60×/1.40 oil objective and a Hamamatsu (Bridgewater, New Jersey, United States) Orca 100 camera operated with MetaMorph 6.1 software (Universal Imaging, Downingtown, Pennsylvania, United States). Background was subtracted and brightness/contrast were adjusted to render an all-or-nothing labeling pattern using Jasc Paint Shop Pro 6.02 (Jasc Software, http://www.corel.com). The presence of immunostaining in the acrosomal region was evaluated in at least 200 cells in three independent experiments. Data were normalized with respect to the percentage of positive cells observed in untreated samples (range 30%–70%).
Statistical analysis
Data were evaluated using one-way ANOVA. The Tukey-Kramer post hoc test was used for pairwise comparisons. The results are listed in Tables S1–S8, which correspond to data depicted in Figures S1, 1, 2, 3, 4, 6, 7, and 8, respectively. Only significant differences (p < 0.05) are discussed in the text.
Supporting Information
Figure S1 SNAREs Are Assembled in Neurotoxin-Resistant Complexes in Non-Permeabilized Resting Spermatozoa
Spermatozoa maintained in gamete preparation medium were treated at 37 °C for 15 min with 150 nM BoNT/A (holotoxin) or 25 nM BoNT/F (holotoxin). Next, 25 μM TPEN (a zinc chelator that inactivates neurotoxins) was added, AE was activated by adding 10 μM A23187, and the incubation continued for an additional 15 min (black bars). Sperm were then fixed and AE was measured as described in Materials and Methods. Several controls are included (grey bars): background AE in the absence of any stimulation (control); AE stimulated by 10 μM A23187 (Iono); lack of TPEN effect on exocytosis (TPEN→Iono); inhibitory effect of the neurotoxins on exocytosis (neurotoxin→Iono); and block of neurotoxin activity by TPEN (TPEN→neurotoxin→ Iono). The data were normalized as described in Materials and Methods (mean ± standard error of the mean [SEM]). Statistical analysis is provided in Table S1.
(409 KB JPG).
Click here for additional data file.
Table S1 Accessory Data and Statistical Analysis for Figure S1
(38 KB DOC).
Click here for additional data file.
Table S2 Accessory Data and Statistical Analysis for Figure 1
(50 KB DOC).
Click here for additional data file.
Table S3 Accessory Data and Statistical Analysis for Figure 2A and 2C
(62 KB DOC).
Click here for additional data file.
Table S4 Accessory Data and Statistical Analysis for Figure 3
(39 KB DOC).
Click here for additional data file.
Table S5 Accessory Data and Statistical Analysis for Figure 4B
(37 KB DOC).
Click here for additional data file.
Table S6 Accessory Data and Statistical Analysis for Figure 6
(51 KB DOC).
Click here for additional data file.
Table S7 Accessory Data and Statistical Analysis for Figure 7
(72 KB DOC).
Click here for additional data file.
Table S8 Accessory Data and Statistical Analysis for Figure 8
(36 KB DOC).
Click here for additional data file.
The authors thank M. Furlán for technical assistance, Drs. Patterson, Stevens, and Matti for critical reading of the manuscript, and Drs. Matti, Stahl, Colombo, Binz, Whiteheart, and Fukuda for plasmids and proteins. Thanks are also given to L. de Jong and Dr. Fernandez for BoNT/A and BoNT/F (holotoxins), and to Dr. Álvarez for statistical advice. This work was supported partly by an International Research Scholar Award from the Howard Hughes Medical Institute to LSM and by grants from Consejo Nacional de Investigaciones Científicas y Técnicas (Argentina) and Agencia Nacional de Promoción Científica y Tecnológica (Argentina).
Competing interests. The authors have declared that no competing interests exist.
Author contributions. CNT and LSM conceived and designed the experiments. GAD, CMR, and CNT performed the experiments. GAD, CMR, CNT, and LSM analyzed the data. CNT and LSM wrote the paper.
Citation: De Blas GA, Roggero CM, Tomes CN, Mayorga LS (2005) Dynamics of SNARE assembly and disassembly during sperm acrosomal exocytosis. PLoS Biol 3(10): e323.
Abbreviations
AEacrosomal exocytosis
BoNTbotulinum neurotoxin
EAE230A
NP-EGTA-AM
O-nitrophenyl EGTA–acetoxymethyl ester
FITCfluorescein isothiocyanate
PSA
Pisum sativum agglutinin
PVPpolyvinylpyrrolidone
SEMstandard error of the mean
SLOstreptolysin-O
SNAP25synaptosome-associated protein of 25 kD
TeTxtetanus toxin
TPENN,N,N′,N′-tetrakis (2-pyridymethyl) ethylenediamine
VAMPvesicle-associated membrane protein
==== Refs
References
Burgoyne RD Morgan A Secretory granule exocytosis Physiol Rev 2003 83 581 632 12663867
Deneka M Neeft M van der Sluijs P Regulation of membrane transport by Rab GTPases Crit Rev Biochem Mol Biol 2003 38 121 142 12749696
Pfeffer SR Transport-vesicle targeting: Tethers before SNAREs Nat Cell Biol 1999 1 E17 E22 10559876
Sollner TH Regulated exocytosis and SNARE function Mol Membr Biol 2003 20 209 220 12893529
Fasshauer D Sutton RB Brunger AT Jahn R Conserved structural features of the synaptic fusion complex: SNARE proteins reclassified as Q- and R-SNAREs Proc Natl Acad Sci U S A 1998 95 15781 15786 9861047
Schiavo G Matteoli M Montecucco C Neurotoxins affecting neuroexocytosis Physiol Rev 2000 80 717 766 10747206
Schiavo G Benfenati F Poulain B Rossetto O Polverino dL Tetanus and botulinum-B neurotoxins block neurotransmitter release by proteolytic cleavage of synaptobrevin Nature 1992 359 832 835 1331807
Hayashi T Mcmahon H Yamasaki S Binz T Hata Y Synaptic vesicle membrane fusion complex: Action of clostridial neurotoxins on assembly EMBO J 1994 13 5051 5061 7957071
Yanagimachi R Knobil E Neill JD Mammalian fertilization The physiology of reproduction 1994 New York Raven Press 189 281
Michaut M Tomes CN De Blas G Yunes R Mayorga LS Calcium-triggered acrosomal exocytosis in human spermatozoa requires the coordinated activation of Rab3A and N-ethylmaleimide-sensitive factor Proc Natl Acad Sci U S A 2000 97 9996 10001 10954749
Yunes R Michaut M Tomes C Mayorga LS Rab3A triggers the acrosome reaction in permeabilized human spermatozoa Biol Reprod 2000 62 1084 1089 10727281
Tomes CN Michaut M De Blas G Visconti P SNARE complex assembly is required for human sperm acrosome reaction Dev Biol 2002 243 326 338 11884041
Tomes CN De Blas GA Michaut MA Farre EV Cherhitin O α-SNAP and NSF are required in a priming step during the human sperm acrosome reaction Mol Hum Reprod 2005 11 43 51 15542541
De Blas G Michaut M Trevino CL Tomes CN Yunes R The intraacrosomal calcium pool plays a direct role in acrosomal exocytosis J Biol Chem 2002 51 49326 49331
Aballay A Sarrouf MN Colombo MI Stahl PD Mayorga LS Zinc depletion blocks endosome fusion Biochem J 1995 312 919 923 8554539
Gore PJ Singh SP Brooks DE Composition of gangliosides from ovine testis and spermatozoa Biochim Biophys Acta 1986 876 36 47 3947668
Chen YA Scales SJ Scheller RH Sequential SNARE assembly underlies priming and triggering of exocytosis Neuron 2001 30 161 170 11343652
Xu T Binz T Niemann H Neher E Multiple kinetic components of exocytosis distinguished by neurotoxin sensitivity Nat Neurosci 1998 1 192 200 10195143
Hua SY Charlton MP Activity-dependent changes in partial VAMP complexes during neurotransmitter release Nat Neurosci 1999 2 1078 1083 10570484
Söllner T Whiteheart SW Brunner M Erdjument-Bromage H Geromanos S SNAP receptors implicated in vesicle targeting and fusion Nature 1993 362 318 324 8455717
Weber T Parlati F McNew JA Johnston RJ Westermann B SNAREpins are functionally resistant to disruption by NSF and α-SNAP J Cell Biol 2000 149 1063 1072 10831610
Synaptic Systems Synaptobrevin-1 (VAMP-1), synaptobrevin-2 (VAMP-2), cellubrevin (VAMP-3): Major vesicle proteins involved in fusion—Fact sheet 2005 Göttingen (Germany) Synaptic Systems Available: http://www.sysy.com/synaptobrevin/sbrevin_fs.html#vamp2 . Accessed 29 July 2005
Darszon A Beltran C Felix R Nishigaki T Trevino CL Ion transport in sperm signaling Dev Biol 2001 240 1 14 11784043
Lang T Margittai M Holzler H Jahn R SNAREs in native plasma membranes are active and readily form core complexes with endogenous and exogenous SNAREs J Cell Biol 2002 158 751 760 12177041
Rickman C Meunier FA Binz T Davletov B High affinity interaction of syntaxin and SNAP-25 on the plasma membrane is abolished by botulinum toxin E J Biol Chem 2004 279 644 651 14551199
Xiao J Xia Z Pradhan A Zhou Q Liu Y An immunohistochemical method that distinguishes free from complexed SNAP-25 J Neurosci Res 2004 75 143 151 14689457
Fasshauer D Otto H Eliason WK Jahn R Brunger AT Structural changes are associated with soluble N-ethylmaleimide-sensitive fusion protein attachment protein receptor complex formation J Biol Chem 1997 272 28036 28041 9346956
Fiebig KM Rice LM Pollock E Brunger AT Folding intermediates of SNARE complex assembly Nat Struct Biol 1999 6 117 123 10048921
Xu T Rammner B Margittai M Artalejo AR Neher E Inhibition of SNARE complex assembly differentially affects kinetic components of exocytosis Cell 1999 99 713 722 10619425
Coppola T Perret-Menoud V Luthi S Farnsworth CC Glomset JA Disruption of Rab3-calmodulin interaction, but not other effector interactions, prevents Rab3 inhibition of exocytosis EMBO J 1999 18 5885 5891 10545100
Park JB Kim JS Lee JY Kim J Seo JY GTP binds to Rab3A in a complex with Ca2+ /calmodulin Biochem J 2002 362 651 657 11879192
Yunes R Tomes C Michaut M De Blas G Rodriguez F Rab3A and calmodulin regulate acrosomal exocytosis by mechanisms that do not require a direct interaction FEBS Lett 2002 525 126 130 12163174
McBride HM Rybin V Murphy C Giner A Teasdale R Oligomeric complexes link Rab5 effectors with NSF and drive membrane fusion via interactions between EEA1 and syntaxin 13 Cell 1999 98 377 386 10458612
Han SY Park DY Park SD Hong SH Identification of Rab6 as an N-ethylmaleimide-sensitive fusion protein-binding protein Biochem J 2000 352 165 173 11062069
Chen YA Scales SJ Patel SM Doung YC Scheller RH SNARE complex formation is triggered by Ca2+ and drives membrane fusion Cell 1999 97 165 174 10219238
Ungermann C Sato K Wickner W Defining the functions of trans-SNARE pairs Nature 1998 396 543 548 9859990
Tahara M Coorssen JR Timmers K Blank PS Whalley T Calcium can disrupt the SNARE protein complex on sea urchin egg secretory vesicles without irreversibly blocking fusion J Biol Chem 1998 273 33667 33673 9837952
Scheenen WJ Wollheim CB Pozzan T Fasolato C Ca2+ depletion from granules inhibits exocytosis. A study with insulin-secreting cells J Biol Chem 1998 273 19002 19008 9668080
Mayer A Intracellular membrane fusion: SNAREs only? Curr Opin Cell Biol 1999 11 447 452 10449339
Michaut M De Blas G Tomes CN Yunes R Fukuda M Synaptotagmin VI participates in the acrosome reaction of human spermatozoa Dev Biol 2001 235 521 529 11437455
Tucker WC Weber T Chapman ER Reconstitution of Ca2+ -regulated membrane fusion by synaptotagmin and SNAREs Science 2004 304 435 438 15044754
Fukuda M Mikoshiba K A novel alternatively spliced variant of synaptotagmin VI lacking a transmembrane domain. Implications for distinct functions of the two isoforms J Biol Chem 1999 274 31428 31434 10531344
Qiagen (2003 June) The QIAexpressionist: A handbook for high-level expression and purification of 6xHis-tagged proteins Valencia (California) Qiagen Available: http://www1.qiagen.com/literature/handbooks/PDF/Protein/Expression/QXP_QIAexpressionist/1024473_QXPHB_0603.pdf . Accessed 29 July 2005
Mendoza C Carreras A Moos J Tesarik J Distinction between true acrosome reaction and degenerative acrosome loss by a one-step staining method using Pisum sativum agglutinin J Reprod Fertil 1992 95 755 763 1383539
|
16131227
|
PMC1197286
|
CC BY
|
2021-01-05 08:21:28
|
no
|
PLoS Biol. 2005 Oct 6; 3(10):e323
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030323
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 1612862310.1371/journal.pbio.0030324Research ArticleInfectious DiseasesMolecular Biology/Structural BiologyVirologyVirusesIn VitroDesign of Wide-Spectrum Inhibitors Targeting Coronavirus Main Proteases Inhibitors Targeting Coronavirus MprosYang Haitao
1
2
Xie Weiqing
3
Xue Xiaoyu
1
2
Yang Kailin
1
2
Ma Jing
1
2
Liang Wenxue
4
Zhao Qi
1
2
Zhou Zhe
1
2
Pei Duanqing
5
Ziebuhr John
6
Hilgenfeld Rolf
7
Yuen Kwok Yung
8
Wong Luet
9
Gao Guangxia
1
2
Chen Saijuan
4
Chen Zhu
4
Ma Dawei [email protected]
3
Bartlam Mark
1
2
Rao Zihe [email protected]
1
2
1Tsinghua-IBP Joint Research Group for Structural Biology, Tsinghua University, Beijing, China,2National Laboratory of Biomacromolecules, Institute of Biophysics, Chinese Academy of Sciences, Beijing, China,3State Key Laboratory of Bioorganic and Natural Products Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, Shanghai, China,4Shanghai Institute of Hematology, Rui-Jin Hospital affiliated to Shanghai Second Medical University, Shanghai, China,5Guangzhou Institute of Biomedicine and Health, Chinese Academy of Sciences, Guangzhou, China,6Institute of Virology and Immunology, University of Würzburg, Würzburg, Germany,7Institute for Biochemistry, University of Lübeck, Lübeck, Germany,8Department of Microbiology, University of Hong Kong, Hong Kong, China,9Department of Chemistry, Inorganic Chemistry Laboratory, University of Oxford, Oxford, United KingdomBjorkman Pamela Academic EditorHoward Hughes Medical Institute/California Institute of TechnologyUnited States of America10 2005 6 9 2005 6 9 2005 3 10 e32431 5 2005 13 7 2005 Copyright: © 2005 Yang 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.
Casting a Wide Net to Fight Coronaviruses
The genus Coronavirus contains about 25 species of coronaviruses (CoVs), which are important pathogens causing highly prevalent diseases and often severe or fatal in humans and animals. No licensed specific drugs are available to prevent their infection. Different host receptors for cellular entry, poorly conserved structural proteins (antigens), and the high mutation and recombination rates of CoVs pose a significant problem in the development of wide-spectrum anti-CoV drugs and vaccines. CoV main proteases (Mpros), which are key enzymes in viral gene expression and replication, were revealed to share a highly conservative substrate-recognition pocket by comparison of four crystal structures and a homology model representing all three genetic clusters of the genus Coronavirus. This conclusion was further supported by enzyme activity assays. Mechanism-based irreversible inhibitors were designed, based on this conserved structural region, and a uniform inhibition mechanism was elucidated from the structures of Mpro-inhibitor complexes from severe acute respiratory syndrome-CoV and porcine transmissible gastroenteritis virus. A structure-assisted optimization program has yielded compounds with fast in vitro inactivation of multiple CoV Mpros, potent antiviral activity, and extremely low cellular toxicity in cell-based assays. Further modification could rapidly lead to the discovery of a single agent with clinical potential against existing and possible future emerging CoV-related diseases.
Structure-assisted optimization of compounds capable of inactivating CoV Mpros may rapidly lead to an antiviral agent against Coronavirus-associated diseases.
==== Body
Introduction
The genus Coronavirus belongs to the plus-strand RNA virus family of the Coronaviridae and currently contains about 25 species that are classified into three groups according to their genetic and serological relationships [1–4]. Coronaviruses (CoVs) infect humans and multiple species of animals, causing a variety of highly prevalent and severe diseases [1,5]. For example, human coronavirus (HCoV) strains 229E (HCoV-229E), NL63 (HCoV-NL63), OC43 (HCoV-OC43), and HKU1 (HCoV-HKU1) cause a significant portion of upper and lower respiratory tract infections in humans, including common colds, bronchiolitis, and pneumonia. They have also been implicated in otitis media, exacerbations of asthma, diarrhea, myocarditis, and neurological disease [2,3,6–9]. A previously unknown HCoV, severe acute respiratory syndrome coronavirus (SARS-CoV), which is most closely related to the group II CoVs [10], proved to be the etiological agent of a global outbreak of a life-threatening form of pneumonia called severe acute respiratory syndrome (SARS), which, in 2003, was the cause of more than 800 fatalities worldwide [11–14]. Animal CoVs are mainly associated with enteric and respiratory diseases in livestock and domestic animals. Most of the viruses are highly contagious with significant mortality in young animals, resulting in considerable economic losses worldwide [5,9].
Although vaccines have been developed against avian infectious bronchitis virus (IBV), canine CoV, and porcine transmissible gastroenteritis virus (TGEV) to help prevent serious diseases, several potential problems remain. Vaccination against IBV is only partially successful due to the continual emergence of new serotypes and recombination events between field and vaccine strains. The development of vaccines against feline infectious peritonitis virus (FIPV) has been frustrated by the phenomenon of antibody-dependent enhancement. No licensed vaccines or specific drugs are available to prevent HCoV infection [6,9]. Following the SARS outbreak, a series of inhibitors was reported against the helicase and main protease (Mpro) of SARS-CoV to prevent viral replication [15–20]. However, previous research has only placed emphasis on SARS-CoV, and no structural data are available to confirm the direct interaction between these inhibitors and their targets, or for the further modification of these compounds.
In common with other RNA viruses employing RNA-dependent RNA polymerases for genome replication, CoVs are generally thought to mutate at a high frequency [21], although this phenomenon remains to be studied in detail. During the SARS epidemic in China, the emergence of SARS-CoV suggested an animal–human interspecies transmission [22,23]. The virus continued evolving to adapt to the human host during the course of the outbreak [22] with about one-third the mutation rate of human immunodeficiency virus [24]. The high degree of similarity between genome sequences of bovine CoV and the recently sequenced HCoV-OC43 suggested an earlier animal-to-human interspecies transmission than SARS-CoV [25]. Moreover, a high frequency of RNA recombination is a common feature of CoV genetics and has been demonstrated for representative viruses from all CoV groups, including murine hepatitis virus (MHV), TGEV, and IBV [9,26]. For instance, the outbreaks caused by variant strains of IBV that arose from recombination of vaccine and wild-type virulent strains in chicken flocks limit the usage of vaccines against IBV [27,28]. Consequently, it is of concern whether current vaccines or drugs in development will be effective against the next wave of attacks by altered SARS-CoV [22].
In view of the issues posed above, the development of wide-spectrum drugs against the existing pathogenic CoVs is a more reasonable and attractive prospect than individual strategies for drug design, and thereby could provide an effective first line of defense against future emerging CoV-related diseases such as SARS. However, some of the key factors controlling the host spectrum and viral pathogenicity are highly variable among CoVs. For instance, CoVs use different host receptors for cellular entry, have poorly conserved structural proteins (antigens), and encode diverse accessory genes in their 3'-terminal genome regions that probably contribute to the pathogenicity of CoVs in specific hosts [1–3,29–34]. Clearly, this structural and functional diversity presents a significant obstacle for designing a versatile compound against all CoVs unless a highly conserved target that is comparatively stable during evolution is identified within the genus Coronavirus. Here we report the discovery of a highly conserved region based on four crystal structures and one homology model of Mpro representing all three genetic clusters of the genus Coronavirus, and a uniform inhibition mechanism revealed from the structures of Mpro-inhibitor complexes from SARS-CoV and TGEV. A structure-assisted optimization program has yielded compounds with fast in vitro inactivation of multiple CoV Mpros, potent antiviral activity, and extremely low cellular toxicity in cell-based assays. Further modification could rapidly lead to the discovery of a single agent with clinical potential against existing and possible future emerging CoV-associated diseases.
Results/Discussion
Target Identification
Development of wide-spectrum inhibitors is an attractive strategy against CoV-associated diseases; however, it entirely depends on the availability of a conserved target within the whole genus Coronavirus. During the first round of target screening, all structural proteins (including S, E, M, HE, and N proteins) were excluded due to the considerable variations among different CoVs [1–3,33,34]. Subsequently, the RNA-dependent RNA polymerase, RNA helicase, and Mpro constitute attractive potential nonstructural protein targets for consideration. However, no structural data were available for the former two proteins, increasing the difficulties for rational drug design and downstream modification of possible drug leads.
The pivotal roles played by Mpros in controlling viral replication and transcription through extensive processing of replicase polyproteins, together with the absence of closely related cellular homologues, identify the Mpro as a potentially important target for antiviral drug design [35]. However, pairwise BLAST of the primary sequences among CoV Mpros showed identities of only 38% in some cases. Since it is acknowledged that three-dimensional structures are more closely conserved than primary sequences, we decided to investigate the conservation among the CoV Mpro structures. As the Mpros showed comparatively high sequence similarity within each CoV group, representatives from every group were chosen for comparison. The structures of Mpros from TGEV (group I), HCoV-229E (group I), and SARS-CoV are available [36–38]. Although the crystal structure of IBV (group III) Mpro is currently under refinement by our group, it can nevertheless be used as an experiment-based model. As the structure of MHV Mpro (group II) was unavailable, and previous studies have shown that SARS-CoV is related to group II [10], we constructed a homology model for MHV Mpro based on the structure of SARS-CoV Mpro. Superposition of the crystal structures and homology model showed approximately 2 Å root mean square deviation for all 300 Cα, but the most variable regions were the helical domain III and surface loops. The substrate-binding pockets located in a cleft between domains I and II, and especially the S4, S2, and S1 are highly conserved among CoV Mpros suggesting the possibility for wide-spectrum inhibitor design targeting this region in the Mpros of all CoVs. This hypothesis was further supported by enzyme activity assays (see Table 1). Based on the assumption that the substrate-binding sites are highly conserved among CoV Mpros, a fluorescence-labeled substrate MCA-AVLQ↓SGFR-Lys(Dnp)-Lys-NH2 was synthesized to determine the kinetic parameters of TGEV, HCoV-229E, FIPV, MHV, IBV, and SARS-CoV Mpros. The substrate sequence was derived from residues P4–P5′ of the SARS-CoV Mpro N-terminal autoprocessing site, which has the sequence AVLQSGFRK. IBV Mpro demonstrated an almost identical Km to that of SARS-CoV Mpro. An interesting observation was that four other CoV proteases showed marginally stronger binding affinity to the substrate than SARS-CoV Mpro itself. These results further support the preliminary biochemical studies on conservation of substrates of CoV Mpros [39].
Table 1 Enzyme Activity and Enzyme-Inhibition Data for Representatives of All Genetic Clusters of Genus Coronavirus
First Round of Inhibitor Design: Michael Acceptor Inhibitors
The structures of TGEV and SARS-CoV Mpros have previously been determined in complex with a substrate-analog chloromethyl ketone (CMK) inhibitor, Cbz-VNSTLQ-CMK. The sequence of this substrate-analog was derived from residues P6–P1 of the N-terminal autoprocessing site of TGEV Mpro [36,38]. However, the two protomers of SARS-CoV Mpro each exhibited an unexpected binding mode, possibly resulting from the comparatively weak binding of peptidyl elements derived from the substrate of TGEV Mpro and from the highly reactive electrophile CMK. This would suggest that nucleophilic attack might have occurred before a stable noncovalently bound enzyme-inhibitor complex was formed. Accordingly, the single binding mode in the TGEV Mpro complex was taken into account when designing possible broad-spectrum inhibitors on the basis of these structures and models. Although the CMK inhibitor is nonselective because of its high chemical reactivity and is susceptible to cleavage by gastric and enteric proteinases, it could provide structural insight into the substrate-binding pocket. Superposition of the structures and model revealed that all these proteases have a His–Cys catalytic dyad with relatively conserved orientations, in which His acts as a proton acceptor and Cys undergoes nucleophilic attack on the carbonyl carbon of the substrate. It is widely accepted that increased inhibitor potency can be achieved provided that a covalent bond is formed between the active Cys residue and the designed compound, resembling the intermediate during substrate cleavage. The Michael acceptors, a class of conjugated carbonyl compounds, were successfully introduced to devise irreversible Cys protease inhibitors, including the antirhinovirus compound ruprintrivir (formerly designated AG7088) [40–42], and so the highly reactive electrophile CMK was replaced by a less reactive trans-α, β-unsaturated ethyl ester, which was expected to readily extend into the bulky S1′ subsite of CoV Mpros.
During our initial round of inhibitor design, we focused on the S1, S2, and S4 subsites crucial for substrate recognition and utilized a strategy for mimicking the substrate side chains of residues P4–P1 to accommodate the corresponding subsites. Since backbones of CoV Mpros constituting this area superimposed particularly well, except for a small segment located on the outer wall of S2, we concentrated on the variation of side chains forming these pockets. In the TGEV Mpro complex structure [36], the side chains of 165-Glu, 162-His, 171-His, and 139-Phe (also conserved in other Mpros) are incorporated with other backbone elements to constitute the S1 site, which has an absolute requirement for Gln at the P1 position via two hydrogen bonds. Modeling showed that a lactam with (S) stereochemistry at the α-carbon might preserve the hydrogen bonds essential for S1 recognition; moreover, a comparatively bulky lactam ring would create additional van der Waals interactions. The side chains of 164-Leu, 51-Ile, 41-His, and 53-Tyr, as well as the alkyl portion of side chains of 186-Asp and 47-Thr, are involved in forming a deep hydrophobic S2 subsite that can accommodate the relatively large side chain of Leu in TGEV Mpro. This feature can also be observed in the HCoV-229E Mpro. Several conservative substitutions occur in other CoV Mpros (164-Leu → 165-Met in SARS-CoV and MHV Mpros; 53-Tyr → 50-Trp in IBV Mpro). Another minor difference was observed in SARS-CoV and MHV Mpros, where the outer wall segment is composed of a short 310-helix from residues 45–50, compared with a less regular structure in HCoV and TGEV Mpro. With respect to the structure of IBV Mpro undergoing refinement, no clear electron density was observed in the corresponding stretch of residues 44–47. We reasoned that variations in the segment located on the outer wall of S2 should not significantly affect the hydrophobicity of this deep subsite. This is supported by evidence wherein Leu is found at position P2 of substrates for all CoV Mpros. As P2 Phe is present in the C-terminal autocleavage site of SARS-CoV, phenyl was used as a smaller substituent to explore this subsite. The side chain of Thr at P3 is solvent-exposed, so this site was expected to tolerate a wide range of functionality. The side chains of 164-Leu, 166-Leu, 184-Tyr, and 191-Gln that form the S4 hydrophobic subsite of TGEV are conserved in other CoV Mpros, excluding the following conservative substitutions: 184-Tyr → 184-Phe in HCoV Mpro; 164-Leu → 165-Met, 184-Tyr → 185-Phe in SARS-CoV. A tertiary butyloxycarbonyl was introduced at the P4 position as an N-terminal protecting group to enter into the S4 site. Thus, by combining the modifications above, a novel compound designated as I2 (see Figure 1A) was designed and a series of analogs was synthesized for the inhibition assay (see Protocol S1).
Figure 1 Structures of Inhibitors and Their Interactions with SARS-CoV Mpro
(A) The structures of compounds I2, N1, and N3.
(B) A stereo view showing I2 bound into the substrate-binding pocket of the SARS-CoV Mpro at 2.7 Å. The I2 inhibitor is shown in gold and covered by an omit map contoured at 1.0 σ. Residues forming the substrate-binding pocket are shown in silver.
(C) A stereo view showing N1 bound into the substrate-binding pocket of the SARS-CoV Mpro at 2.0 Å. The N1 inhibitor is shown in gold and covered by an omit map contoured at 1.0 σ. Residues forming the substrate-binding pocket are shown in silver. Two water molecules (in red) form hydrogen bonds with N1.
(D) Detailed view of the interactions between the N1 and SARS-CoV Mpro. The N1 inhibitor is shown in green. Hydrogen bonds are shown as dashed lines, and interaction distances are given. The covalent bond is labeled in red.
Kinetic Mechanism of Michael Acceptor Inhibitors
Covalent irreversible inactivation of CoV Mpros by Michael acceptor inhibitors proceeds according to the kinetic mechanism presented in the scheme below:
The inhibitor initially forms a reversible complex with the protease, which then undergoes a chemical step (nucleophilic attack by Cys) leading to the formation of a stable covalent bond [42]. The evaluation of this series of time-dependent inhibitors requires both the equilibrium-binding constant Ki (designated as k
2/k
1) and the inactivation rate constant for covalent bond formation k
3 [43]. We avoided measurement of IC50 after preincubation to assess the effect of these time-dependent inhibitors, since there is a general trend for this value to decrease to zero with prolonged preincubation time, which would lead to an inappropriate evaluation.
The Structure of SARS-CoV Mpro in Complex with an Inhibitor I2
The compounds designed in the first round did not exhibit obvious inhibition on CoV Mpros without preincubation, suggesting a very poor Ki. We were able to solve a 2.7-Å resolution crystal structure of SARS-CoV Mpro complexed with I2 (see Figure 1B; Table S1) despite the weak noncovalent binding, in order to enhance the inhibitory effect of these compounds. Briefly, compound I2 binds to the shallow cleft formed by a portion of the strand eII and a segment of the loop linking domains II and III. The Cβ atom of the Michael acceptor forms a covalent bond with Sγ of 145-Cys as expected. The lactam P1 inserts favorably into S1 and the side chain of Val at P3 is solvent-exposed. However, the failure of P2 and P4 to be properly accommodated by their corresponding subsites attracted our attention, and might account for the poor inhibitory effect of this series of molecules. First, although phenyl at P2 could enter the S2 site, its rigidity prevents it from reorienting to insert further into this site. Second, the N-terminal protecting group tertiary butyloxycarbonyl did not insert into the S4 subsite, possibly as a result of the planar property of the butyloxyamide group. The other compounds designed in this round are listed in Table S2.
Second Round of Inhibitor Design: Optimization of Michael Acceptor Inhibitors
Based on the complex structure of I2, we entered into a second round of optimization focusing on the P2 and P4 recognition sites. For the P2 subsite, the phenyl group was substituted by a more flexible Leu side chain. In order to enhance the binding affinity, a series of residues were utilized as substituents at P4, followed by a heterocycle that should increase the Van der Waals contacts with residues flanking at either side. From this round of modification, two inhibitors designated as N1 and N9, and a more efficacious derivative named N3, were identified with fast in vitro inactivation of all CoV Mpros tested, including those of TGEV, HCoV-229E, FIPV, HCoV-NL63 (representatives from group I); MHV, HCoV-HKU1 (representatives from group II); SARS-CoV (related to group II); and IBV (representative from group III) in preliminary inhibition assays (see Figure S2). These inhibitors are not sensitive under 1 mM concentration of dithiothreitol (DTT), which is consistent with a previous report of this type of compound [42]. Subsequently, strict inhibition kinetic parameters were determined and are listed in Table 1 (determination of kinetic parameters of Mpros of HCoV-HKU1 and HCoV-NL63 is underway). These inhibitors showed more powerful inhibition of FIPV Mpro than other proteases with high inactivation rates (kobs/[I] ≥ 23,000 M−1•s−1), such that measurement of Ki and k
3 proved difficult. In this case, kobs/[I] was utilized to evaluate their inhibition as an approximation of the pseudo second-order rate constant (k
3/Ki) if very rapid inactivation occurs. The Ki of N1 ranges from approximately 1.11–10.7 μM and k
3 ranges from approximately 4.1–50 × 10−3s−1; the Ki of N9 ranges from approximately 0.9–6.7 μM, and k
3 ranges from approximately 2.6–19.5 × 10−3s−1. Compared with N1 and N9, N3 demonstrated more potent inhibition on TGEV, FIPV, MHV, and IBV Mpros with kobs/[I] ranging from approximately 4,700–47,000 M−1•s−1. We therefore solved the crystal structure of SARS-CoV and TGEV Mpros individually complexed with N1, which revealed a common mechanism of inhibition among CoV Mpros.
The Structure of SARS-CoV Mpro in Complex with the Inhibitor N1
N1 binds to protomers A and B of SARS-CoV in an identical and normal manner. On binding N1, the S1 subsite in protomer B adopts an active conformation compared with the partially collapsed S1 pocket of protomer B in the native structure [38], which can be ascribed to inhibitor-induced conformational changes. As a result, discussion will be focused entirely on protomer A (see Figures 1C, 1D, 2A, and 2B). From the omit map (contoured at 1.2 σ), clear electron density showed that N1 binds in an extended conformation with the inhibitor backbone atoms forming an antiparallel sheet with residues 164–168 of the long strand eII on one side, and with residues 189–191 of the loop linking domains II and III. Here we dissect the inhibitor into different parts for further discussion.
Figure 2 Surface Representation of Native SARS-CoV Mpro and Inhibitor Complexes
(A) Surface representation of conserved substrate-binding pockets of five CoV Mpros. Background is SARS-CoV Mpro. Red: identical residues among the five CoV Mpros; magenta: substitution in one CoV Mpro; orange: substitution in two CoV Mpros. The S1, S2, S4, and S1′ subsites and residues forming the substrate-binding pocket are labeled.
(B) Surface representation of SARS-CoV Mpro (blue) complexed with N1 (green). Water molecules are shown as red spheres. The P1–P5 and P1′ groups and residues forming the substrate-binding pocket are labeled.
(C) Surface representation of SARS-CoV Mpro (blue) in complex with N3 (green). Labels are the same as in Figure 2B.
Gate-regulated switch
Comparison between the molecular surfaces of SARS-CoV Mpro complexed with N1 and the native enzyme show that certain residues constituting the S1 and S2 subsites undergo large conformational changes on inhibitor binding (see Figure 2A and 2B). The side chain of 142-Asn flips over with a 6-Å shift to superpose onto the lactam like a lid when P1 inserts into the subsite. This might account for the movement of main chains of residues 141–143 toward the S1 site; 142-Asn, together with the main chains of neighboring residues, covers the P1 site like one half of a gate. On the opposite side, 49-Met protrudes by around 5Å from the hydrophobic S2 site and is situated parallel to the side chain of Leu at P2. The side chain of another residue, 189-Gln, moves upward to form a 3.0-Å hydrogen bond with the backbone NH of P2. These two residues constitute the other half of a gate. Together, these two halves should serve as a gate-regulated switch with an essential role in substrate or inhibitor recognition and binding.
Trans-α, β-unsaturated ethyl ester
Clear electron density showed that the Sγ atom of 145-Cys forms a standard 1.8-Å C–S covalent bond with the Cβ of vinyl group, which suggests a Michael addition reaction. The Sγ atom moved slightly (approximately 0.6 Å) toward the interior of the protein compared with the native enzyme. The Michael acceptor remains in a plane following the Michael addition since it is stabilized by a water molecule. This ordered water molecule donates a long 3.3-Å hydrogen bond to the carboxylate oxygen of the ester and then accepts a 2.8-Å hydrogen bond from the backbone NH of 143-Gly and a 3.0-Å hydrogen bond from the carboxamide nitrogen of 142-Asn. The position of Sγ in 145-Cys implies that it undergoes nucleophilic attack on Cβ by approaching the π-electron cloud from above. The carbonyl oxygen occupies the oxyanion hole and is close to backbone NHs of 143-Cys and 145-Cys, mimicking the tetrahedral oxyanion intermediate formed during Ser protease cleavage. However, the standard hydrogen bonds are not formed. The ethyl ester portion extends into the S1′ site, with sufficient size, and in an extended conformation, to interact with the alkyl portions of 25-Thr and 27-Leu by van der Waals interaction.
P1, P2, and P4 sites
The lactam at P1 inserts favorably into the S1 subsite and forms two stable 2.6-Å hydrogen bonds: one between the lactam oxygen and NE2 of 163-His, and another between the lactam NH and a water molecule at the bottom of this subsite. The Cα of Leu at the P2 site in N1 moves into the S2 subsite by approximately 1 Å relative to the corresponding carbon atom in I2, and Cβ–Cγ of Leu forms an angle of approximately 40° to the phenyl at P2 in the I2 complex, inserting deeply into the S2 subsite. Another notable difference between N1 and I2 is the insertion of an Ala between P3 and P4 in I2, the latter of which was replaced by an isoxazole to block the N-terminal. As expected, the side chain of Ala at the current P4 position readily enters into the S4 subsite. Simultaneously, the backbone NH of Ala donates a hydrogen bond to the carbonyl oxygen of 190-Thr. The isoxazole at P5 makes Van der Waals contacts with 168-Pro and the backbone of residues 190–191.
Further modifications of N1
A variety of substitutions were investigated for P4, P5, and P1′ (see Table S3). The 1.85-Å crystal structure of SARS-CoV Mpro complexed with N9 (see Figure S1) showed that Val could serve as a substituent at P4, slightly increasing the hydrophobic interactions. Another derivative N3 with benzyl ester exhibited improved inhibition, which could be seen from inhibition assays of FIPV and MHV Mpros (see Table 1). Its co-crystal structure with SARS-CoV Mpro indicated that the bulky benzyl group extends into the S1′ site, possibly enhancing the Van der Waals interaction with 25-Thr and 27-Leu (see Figure 2C).
The Structure of TGEV Mpro in Complex with an Inhibitor N1
There are two molecules per asymmetric unit in the co-crystal structure of TGEV Mpro with N1, compared with as many as six molecules per asymmetric unit in the native enzyme structure [37]. N1 binds to TGEV Mpro in a similar mode to SARS-CoV Mpro with some subtle differences (see Figure 3). First, after the nucleophilic addition reaction, the Michael acceptor does not remain in a plane as in the SARS-CoV Mpro complex structure, but instead flips over by about 90° to interact with the backbone atoms of residues 141–142. Unlike the SARS-CoV Mpro complexed with N1, the TGEV Mpro lacks a water molecule connecting the ethyl ester with the side chain of residue 142 (Asn → Ala in TGEV Mpro). The rate of chemical inactivation presumably depends on how the reactive vinyl group is oriented and on the extent to which the transition-state intermediate can be stabilized by proteases [42]. We suspect that in SARS-CoV Mpro, the water molecule prevents the Michael acceptor from reorienting to accept a proton from the imidazole of 41-His in the transition state. Although the intermediate remains to be unveiled, this could partially explain why N1 has a higher inactivation rate constant (k
3) against TGEV Mpro than SARS-CoV Mpros. Second, another water molecule in the TGEV Mpro complex occupies an equivalent position to the 189-Gln side chain, which interacts with the backbone NH of Leu at P2 in SARS-CoV Mpro–N1 complex. This water molecule, however, donates a 2.6-Å hydrogen bond to 47-Thr and accepts a 2.7-Å hydrogen bond from the NH backbone of Leu at P2. Third, the isoxazole sways to interact with the backbone atoms of residues 188–189. It is worth mentioning that these slight variations do not notably affect the Ki, as the binding modes of P1, P2, and P4 remain the same as in SARS-CoV Mpro.
Figure 3 The Structure of TGEV Mpro in Complex with N1
A stereo view showing N1 bound into the substrate-binding pocket of the TGEV Mpro at 2.7 Å. The N1 inhibitor is shown in gold and covered by an omit map contoured at 1.0 σ. Residues forming the substrate-binding pocket are shown in silver. The red sphere represents a water molecule that is hydrogen bonded to N1.
HCoV-229E, FIPV, and MHV Inhibition Assays
Despite the high multiplicity and single-cycle infection conditions, N3 displayed potent inhibition against HCoV-229E, FIPV, and MHV-A59 with individual IC50 of 4.0 μM, 8.8 μM, and 2.7 μM, respectively (see Figure 4A–4C). The dose response curves all show that N3 was able to penetrate cells derived from different species and tissues to access its targets. Consequently, the results strongly imply that N3 was a wide-spectrum anti-CoV lead compound. However, we noticed some small discrepancies in the data between enzyme-inhibition assays and cell-based assays. This can be explained by the different cells for the inhibitor to enter and by potential incongruities in the dependence of Mpro for different CoVs. Furthermore, we cannot exclude the potential existence of differences among the bacterially expressed proteases in enzyme-inhibition assays and subtle differences in activity that were not fully revealed by the SARS-CoV-derived substrate used in our in vitro assays.
Figure 4 Cell-Based Assays of N3 against HCoV-229E, FIPV, and MHV-A59
Inhibition of replication of three CoVs under high-multiplicity single-cycle growth conditions (MOI = 3) and protection of DBT cells from MHV infection under low-multiplicity growth conditions (MOI = 0.01). (A) Reduction of HCoV-229E titer in MRC-5 cell culture by N3. (B) Reduction of FIPV titer in FCWF cell culture by N3. (C) Reduction of MHV-A59 titer in DBT cell culture by N3. In (A–C), infections were done at an MOI of 3 TCID50 per cell, and titers were determined at 14 h postinfection. (D) Plaque-reduction assay of MHV-A59.
FCWF, F. catus whole fetus; MOI, multiplicity of infection.
MHV Plaque-Reduction Assay
To further substantiate the data and, in particular, to evaluate the ability of this type of compound to prevent cells from being infected by CoVs and their cellular cytotoxicity, a murine delay brain tumor (DBT) cell-based MHV plaque-reduction assay was performed for the following reasons: (1) three important human pathogens HCoV-HKU1, HCoV-OC43, and SARS-CoV belong to or relate to group II CoVs; (2) aged mice have been successfully used as a model for increased severity of SARS in elderly humans [44]. The EC50 of the MHV plaque-reduction assay was 3.4 μM (see Figure 4D), which was consistent with the IC50 determined in the MHV inhibition assay. It was observed that when the concentration of N3 increased to 8 μM, the DBT cells could be sufficiently protected. Moreover, 500 μM N3 only displayed 28.3% inhibition of cell growth, suggesting extremely low cellular toxicity (see Figure S3). These results demonstrate that N3 is a particularly promising lead compound for further development.
Future Prospects
Evidence suggests that CoVs may have completed at least two animal-to-human interspecies transmissions to date [22,24,25]. An alternative hypothesis has been proposed whereby the 1889–1890 pandemic characterized by malaise, fever, pronounced central nervous system symptoms, with a significant increase in case fatality with increasing age, was the result of interspecies transmission of bovine CoV to humans rather than an influenza virus [25]. Although this hypothesis needs more evidence to support, it is widely acknowledged that SARS resulted from animal-to-human transmission of a previously unknown CoV. CoVs, especially those that can infect hosts such as domestic animals and pets, which humans have frequent contact with, remain a potential threat to human health assuming they cross the interspecies barrier again. Hence, the development of wide-spectrum drugs will lead to increased protection of human health, a reduction of the considerable economic costs associated with CoVs, defense against endangered wild animals susceptible to infection, and valuable model animals such as transgenic mice with high mortality rates for CoVs. Identification of the CoV Mpro as a conserved target among all CoVs will provide an opportunity for the development of broad-spectrum inhibitors against all CoV-related diseases. Ruprintrivir, whose backbone was also a trans-α, β-unsaturated ester incorporated with the peptidyl portion, has entered clinical trials against rhinovirus infection [42], although it did not show inhibition of CoVs [20]. This is a compound with poor aqueous solubility and low oral bioavailability in animals. In preclinical animal studies, hydrolysis of this compound produced alcohol and carboxylic acid metabolite, which was 400-fold less active than ruprintrivir and was the predominant biotransformation pathway. Ruprintrivir is formulated as a suspension for intranasal delivery. Phase II studies reported ruprintrivir prophylaxis reduced the proportion of subjects with positive viral cultures and viral titers. Ruprintrivir is well tolerated, and the most common adverse effects of this compound are blood-tinged mucus and nasal passage irritation [45,46]. This highlights that structure-assisted optimization of N3 could possibly lead to the discovery of a single agent to enter clinical trials against all CoV-associated diseases, although ultimate clinical potential requires more sufficient investigation. Our latest results show that N3 could also strongly inhibit the replication of SARS-CoV and TGEV in cell-based assays (data to be published elsewhere). Furthermore, since this compound was designed against a highly conserved region within the genus Coronavirus, it should have efficient resistance to the high mutation and recombination rates of CoVs. It is noteworthy that N3 also exhibited potent inhibition on the Mpros of HCoV-NL63 and HCoV-HKU1, two recently identified HCoVs associated with bronchiolitis, conjunctivitis, and pneumonia [2,3], in preliminary inhibition assays (see Table S2). This strongly supports our hypothesis that a single agent developed from N3 could provide an effective first line of defense against future emerging CoV-related diseases. Moreover, it also suggests that incorporation of Michael acceptor with the peptidyl portion specific for proteases would be a good starting point for the development of inhibitors against viral Cys or Ser proteases. A comprehensive and systematic program of optimization of this class of inhibitors based on CoV Mpro-inhibitor complexes is underway. We have so far crystallized MHV Mpro, and the crystallization of Mpros of recently identified HCoV-NL63 and HCoV-HKU1 are in progress.
Materials and Methods
Protein cloning, expression, and purification
The preparation of SARS-CoV Mpro for structural analysis has been described previously [38]. The method of preparation of SARS-CoV Mpro for activity assay is almost identical except that the coding sequence was inserted into BamHI and XhoI sites of the expression vector pGEX-4T-1 (Pharmacia, New York, United States). The cDNA encoding IBV Mpro (M41 strain) was a gift from Professor Ming Liao (South China Agricultural University, China); the cDNA encoding Mpro of MHV (A59 strain) was a gift from Professor Guangxia Gao (Institute of Biophysics Chinese Academy of Sciences, China); the cDNA encoding Mpro of HCoV-HKU1 was kindly provided by Professor Kwok-yung Yuen (University of Hong Kong, China); coding sequences of TGEV, IBV, HCoV-HKU1, and HCoV-NL63 Mpros were inserted into BamHI and XhoI sites of the pGEX-4T-1 plasmid, and the subsequent methods for expression and purification were carried out as for SARS-CoV Mpro. After change of a BamHI cleavage site at 429–434 in the sequence coding MHV Mpro to
GGCTCC, this coding sequence was inserted into BamHI and XhoI sites of pGEX-4T-1 plasmid for expression. FIPV Mpro (15 mg/ml) and HCoV-229E Mpro (15 mg/ml; two amino acids deleted at C-terminal) were expressed and purified as described previously [39,47].
Crystallization and data collection
SARS-CoV Mpro was crystallized as previously reported [38]. The SARS-CoV Mpro inhibitor complexes were prepared as follows. First, the inhibitors were dissolved in 7.5% PEG 6000, 6% DMSO, and 0.1 M Mes (pH 6.0) with a concentration of 10 mM (supersaturation). Then, a 3-μl aliquot of such solution was added to the drop, and the crystals were soaked for approximately 2–6 days. A single crystal was prepared for low-temperature data collection by transfer to a cryoprotectant solution containing 30% PEG 400 and 0.1 M Mes (pH 6.0) and then flash frozen in a stream of N2 gas at 100 K. The set of SARS-CoV Mpro-I2 complex data was collected to 2.7 Å resolution using a Mar345 image plate (Marresearch, Norderstedt, Germany) mounted on a Rigaku RU2000 X-ray generator (Sevenoaks, United Kingdom) operated at 48 kV and 98 mA (λ = 1.5418 Å). Data for SARS-CoV Mpro individually complexed with N1 and N3 were collected at 100 K in-house on a Rigaku CuKα rotating-anode X-ray generator (MM007) at 40 kV and 20 mA (λ = 1.5418 Å) with a Rigaku image-plate detector. Data for SARS-CoV Mpro-N9 complex were collected at 100 K in-house on a Rigaku CuKα rotating-anode X-ray generator (FR-E) at 45 kV and 45 mA (λ = 1.5418 Å) with a Rigaku image-plate detector.
In respect to TGEV Mpro co-crystal preparation, TGEV Mpro was incubated with a 3-fold molar excess of N1 for 24 h at 4 °C. Crystallization trials were performed by the method published previously [37]. Briefly, the condition for crystal growth is 0.1 M HEPES (pH 8.5), 1.8 M (NH4)2SO4, 6% MPD, 5 mM DTT, and 5% dioxane. The set of TGEV Mpro-N1 complex data was collected according to the method for SARS-CoV Mpro-N9 complex All intensity data were indexed, integrated, and scaled with the HKL2000 programs DENZO and SCALEPACK [48]. Data collection statistics are summarized in Table S1. Since the refinement of the IBV Mpro structure is ongoing, the methods of crystallization and structure determination will be published elsewhere.
Structure elucidation, model building, and refinement
The methods for structure determination, model building, and refinement were published previously [38]. Briefly, the SARS-CoV Mpro-I2 complex structure was determined by molecular replacement from our native structure of SARS-CoV Mpro (pH 7.6) (PDB ID: 1UK3). The structures of SARS-CoV Mpro in complex with N1, N3, or N9 were determined from the isomorphous SARS-CoV Mpro-I2 complex structure. The TGEV Mpro-N1 structure was determined by molecular replacement using a single monomer of the native TGEV Mpro structure (PDB ID: 1P9U). All cross-rotation and translation searches for molecular replacement were performed with CNS [49]. Adjustments to the models were made in O [50]. Positional refinement, individual B-factor refinement, and water picking were performed with CNS [49]. Validation of the final models was performed with PROCHECK [51]. Detailed refinement statistics are summarized in Table S1.
Enzyme activity assay
The activity of Mpros was measured by continuous kinetic assays, using an identical fluorogenic substrate MCA-AVLQSGFR-Lys(Dnp)-Lys-NH2 (over 95% purity, GL Biochem Shanghai Ltd, Shanghai, China). The fluorescence intensity was monitored with a Fluoroskan Ascent instrument (ThermoLabsystems, Helsinki, Finland) using wavelengths of 320 and 405 nm for excitation and emission, respectively. The experiments were performed with a buffer consisting of 50 mM Tris-HCl (pH 7.3), 1 mM EDTA, with or without DTT. Kinetic parameters, Km and kcat, were determined by initial rate measurements at 30 °C. With respect to SARS-CoV Mpro, the reaction was initiated by adding protease (final concentration of 1 μM) to a solution containing different final concentrations of the substrate (3.2–40 μM). The concentrations of other Mpros and individual substrate range for activity assay are as follows: IBV Mpro: 0.8 μM, substrate range: 6.4–80 μM; HCoV-229E Mpro: 0.1 μM, substrate range: 1.6–20 μM; TGEV Mpro: 0.1 μM, substrate range: 6.4–80 μM; FIPV Mpro: 0.1 μM, substrate range: 1.6–20 μM; MHV Mpro: 1 μM, substrate range: 6.4–80 μM. Fluorescence was monitored at 1 point per 2 s. Initial rates were calculated by fitting the linear portion of the curves (the first 3 min of the progress curves) to a straight line using the program Origin 7.0 (OriginLab Corporation, Natick, Massachusetts, United States). The initial velocities were converted to enzyme activity (micromole substrate cleaved)/second. Kinetic constants were obtained from a double-reciprocal plot.
Mpro inhibition assays
As compounds with potent inhibition identified in preliminary inhibition assay, the strict kinetic parameters were determined. Time-dependent inhibitor progress curves were fit to a first-order exponential (equation 2) [43,52] to yield an observed first-order inhibition rate constant (kobs). P is the product fluorescence; v
0 is the initial velocity; t is time; D is a displacement term to account for the fact that the emission is nonzero at the start of data collection. The values of Ki and k
3 were calculated from plots of 1/kobs obtained from equation 2 versus 1/[I] according to equation 3. [I] is inhibitor concentration; [S] is substrate concentration; Km is the Michaelis-Menten constant for the substrate; k
3 is the rate constant of inactivation, and Ki is the equilibrium constant.pt?>
In the experiment, the Ki and k
3 values for the irreversible inhibitors were obtained from reactions initiated by addition of individual Mpro, the concentration of which was similar as that for the enzymatic activity assay, containing 10 or 20 μM substrate, which depends on the enzymatic activity. The inhibitors vary from 5–8 different concentrations (10-fold molar excess of the enzyme in most cases). Data from the continuous assays were analyzed with the nonlinear regression analysis program Origin. When fast inactivation occurs, the measurement of Ki and k
3 proved difficult. In this case, kobs/[I] was used as an approximation of the pseudo second-order rate constant to evaluate the inhibitors and was measured at approximately 2–4 different inhibitor concentrations. The error associated with this determination (kobs/[I]) is less than 20% of a given value.
MHV-A59 plaque-reduction assay
Murine DBT cells (generously provided by Dr. Lishan Su of University of North Carolina) were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (FBS) and antibiotics at 37 °C in 5% CO2.
DBT cells were suspended in growth medium in triplicate wells in 6-well plates and preincubated with appropriate concentrations of the inhibitor. The next day, the medium was aspirated, and MHV-A59 was added to each well at a titer of 100 PFU/well. After incubation for 1 h, the virus inoculum was aspirated, and 2 ml of a media-agar overlay with appropriate concentrations of inhibitor was added to each well. The plates were further incubated for 24 h and stained with neutral red to visualize plaques.
Cytotoxicity assay
DBT cells were suspended in growth medium in 96-well plates. The next day, appropriate concentrations of the inhibitor were added to the medium. Two days later, the relative numbers of surviving cells were measured by MTT (Sigma, St. Louis, Missouri, United States) assay in accordance with the manufacturer's instructions.
HCoV-229E, FIPV, and MHV-A59 infection assays
Human embryonic lung fibroblast cells (MRC-5;
ATCC [Manassas, Virginia, United States]: CCL 171), Felis catus whole fetus (macrophage) cells (FCWF,
ATCC: CRL 2787), and DBT cells were cultured in minimal essential medium (MEM) supplemented with 25 mM HEPES, Glutamax I, nonessential amino acids, 10% FBS, and antibiotics at 37 °C in 5% CO2. Nearly confluent monolayers of MRC-5 (incubated at 33 °C following infection), FCWF, and DBT cells, which were grown in 6-well plates, were infected with HCoV-229E, FIPV (strain 79–1146), and MHV-A59, respectively, at a multiplicity of infection of 3 TCID50 per cell. After 60 min of virus adsorption, the virus inoculum was replaced with cell culture medium containing varying concentrations of N3 or in the absence of inhibitor. At 14 h postinfection, the virus titers in the cell culture supernatants were determined using standard procedures. All experiments were performed in triplicate and mean values were determined.
Supporting Information
Figure S1 A Stereo View Showing N9 Bound into the Substrate-Binding Pocket of the SARS-CoV Mpro at 1.85 Å
The N9 inhibitor is shown in gold and covered by an omit map contoured at 1.0 σ. Residues forming the substrate-binding pocket are shown in silver. Two water molecules (in red) form hydrogen bonds with N9.
(425 KB PDF).
Click here for additional data file.
Figure S2 N3 Has Wide-Spectrum Inhibition on CoV Mpros
Activity profile curves were displayed at two different inhibitor concentrations for (A–F). (A) 0.1 μM HCoV 229E Mpro solution with 10 μM substrate.
(B) 0.1 μM TGEV Mpro solution with 20 μM substrate.
(C) 0.05 μM FIPV Mpro solution with 10 μM substrate.
(D) 0.6 μM MHV Mpro solution with 20 μM substrate.
(E) 0.8 μM IBV Mpro solution with 20 μM substrate.
(F) 1 μM SARS-CoV Mpro solution with 20 μM substrate.
(G) The preliminary inhibitory assay of N3 on Mpro of a newly identified CoV (HCoV-HKU1). Curve A represents the activity curve of 1 μM Mpro of HCoV-HKU1 in cleaving 20 μM substrate with time; curves B and C individually represent the decrease in enzyme activity when N3 was added with 2-fold and 4-fold molar of protease.
(H) The preliminary inhibitory assay of N3 on Mpro of a recently identified CoV (HCoV-NL63). Curve A represents the activity curve of 0.5 μM Mpro of HCoV-NL63 in cleaving 10 μM substrate with time; curves B and C individually represent the decrease in enzyme activity when N3 was added with 2-fold and 4-fold molar of protease.
(1.2 MB PDF).
Click here for additional data file.
Figure S3 The Cytotoxicity of N3 on Murine DBT Cells
(124 KB PDF).
Click here for additional data file.
Table S1 Data Collection and Refinement Statistics
(106 KB PDF).
Click here for additional data file.
Table S2 Representative Inhibitors Designed in the First Round (I2 not shown here)
(128 KB PDF).
Click here for additional data file.
Table S3 Representative Inhibitors Designed in the Second Round (N1 and N3 not shown here)
(121 KB PDF).
Click here for additional data file.
Protocol S1 (137 KB PDF).
Click here for additional data file.
Accession Numbers
The Protien Data Bank (http://www.rcsb.org/pdb/) accession numbers for the structures of SARS Mpro individually complexed with I2, N1, N3, and, N9, and TEGV Mpro in complex with N1 are 1WNQ, 1WOF, 2AMQ, 2AMD, and 2AMP, respectively. The GenBank (http://www.ncbi.nih.gov/Genbank/) accession number for IBV Mpro is DQ157446.
We thank Xuemei Li, Sheng Ye, Yi Han, Xiaoyun Ji, Chuan Qin, Andrew R. Chang, and Shengjian Li for technical assistance; Ming Liao and George F. Gao for supplying cDNA of IBV Mpro; Huanming Yang, Jan Wang, and Jun Yu for providing cDNA of SARS-CoV Mpro; Chih-chen Wang and Jun Gu for supplying fluorometers; Hualiang Jiang, Luhua Lai, Song Li, and Gang Liu for supplying substrates and advice; Hua Fu for discussion and advice. This work was supported by Projects 973 and 863 of the Ministry of Science and Technology of China (grant numbers 200BA711A12 and G199075600), the National Natural Science Foundation of China (grant numbers 30221003, 20342002, and 20321202), the Sino-German Center (grant number GZ236[202/9]), and the Sino-European Project on SARS Diagnostics and Antivirals of the European Commission (grant number 003831). JZ and RH were supported by the Deutsche Forschungsgemeinschaft.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. HY, DM, and ZR conceived and designed the experiments. HY, WX, XX, KY, JM, WL, QZ, ZZ, JZ, KYY, GG, DM, and MB performed the experiments. HY, XX, KY, QZ, ZZ, LW, DM, and MB analyzed the data. WX, DP, JZ, RH, KYY, GG, SC, ZC, and DM contributed reagents/materials/analysis tools. HY, MB, and ZR wrote the paper.
Citation: Yang H, Xie W, Xue X, Yang K, Ma J, et al. (2005) Design of wide-spectrum inhibitors targeting coronavirus main proteases. PLoS Biol 3(10): e324.
Note Added in Proof
The version of this paper that was first made available on 6 September 2005 has been replaced by this, the definitive, version: there was a typesetting error in equation 1 that has now been corrected.
Correction Note
There was a typesetting error remaining in equation 1, despite the Note Added in Proof indicating that the equation had been corrected. The second reaction step should not have had a reverse arrow, which has now been removed. Corrected 10/24/05.
Abbreviations
CMKchloromethyl ketone
CoVcoronavirus
DBTdelay brain tumor
DTTdithiothreitol
FIPVfeline infectious peritonitis virus
HCoVhuman coronavirus
HCoV-HKU1HCoV strain HKU1
HCoV-NL63HCoV strain NL63
HCoV-OC43HCoV strain OC43
HCoV-229EHCoV strain 229E
IBVavian infectious bronchitis virus
MHVmurine hepatitis virus
Mpromain protease
SARSsevere acute respiratory syndrome
SARS-CoVsevere acute respiratory syndrome coronavirus
TGEVporcine transmissible gastroenteritis virus
==== Refs
References
Lai MMC Holmes KV Knipe DM Howley PM
Coronaviridae The viruses and their replication Fields virology, 4th ed 2001 Philadelphia Lippincott Williams and Wilkins 1163 1179
Woo PC Lau SK Chu CM Chan KH Tsoi HW Characterization and complete genome sequence of a novel coronavirus, coronavirus HKU1, from patients with pneumonia J Virol 2005 79 884 895 15613317
van der Hoek L Pyrc K Jebbink MF Vermeulen-Oost W Berkhout RJ Identification of a new human coronavirus Nat Med 2004 10 368 373 15034574
Spaan WJM Cavanagh D
Coronaviridae
Virus taxonomy: Eighth report of the International Committee on Taxonomy of Viruses 2004 London Elsevier-Academic Press 945 962
Pereira HG Porterfield JS
Coronaviridae
Andrewes' viruses of vertebrates, 5th ed 1989 London Baillière Tindall 42 57
Siddell SG Ziebuhr J Snijder EJ Mahy BWJ ter Meulen V Coronaviruses, toroviruses, and arteriviruses Topley and Wilson's microbiology and microbia infections, 10th ed 2005 London Hodder Arnold 823 856
Myint SH Siddell SG Human coronavirus infections The Coronaviridae 1995 New York Plenum 389 401
Gagneur A Sizun J Vallet S Legr MC Picard B Coronavirus-related nosocomial viral respiratory infections in a neonatal and paediatric intensive care unit: A prospective study J Hosp Infect 2002 51 59 64 12009822
Holmes KV Knipe PM Howley PM Coronaviruses Fields virology, 4th ed 2001 Philadelphia Lippincott Williams and Wilkins 1187 1197
Snijder EJ Bredenbeek PJ Dobbe JC Thiel V Ziebuhr J Unique and conserved features of genome and proteome of SARS-coronavirus, an early split-off from the coronavirus group 2 lineage J Mol Biol 2003 331 991 1004 12927536
Drosten C Gunther S Preiser W van der Werf S Brodt HR Identification of a novel coronavirus in patients with severe acute respiratory syndrome N Engl J Med 2003 348 1967 1976 12690091
Ksiazek TG Erdman D Goldsmith CS Zaki SR Peret T A novel coronavirus associated with severe acute respiratory syndrome N Engl J Med 2003 348 1953 1966 12690092
Kuiken T Fouchier RA Schutten M Rimmelzwaan GF van Amerongen G Newly discovered coronavirus as the primary cause of severe acute respiratory syndrome Lancet 2003 362 263 270 12892955
Peiris JS Lai ST Poon LL Guan Y Yam LY Coronavirus as a possible cause of severe acute respiratory syndrome Lancet 2003 361 1319 1325 12711465
Bacha U Barrila J Velazquez-Campoy A Leavitt SA Freire E Identification of novel inhibitors of the SARS coronavirus main protease 3CLpro Biochemistry 2004 43 4906 4912 15109248
Blanchard JE Elowe NH Huitema C Fortin PD Cechetto JD High-throughput screening identifies inhibitors of the SARS coronavirus main proteinase Chem Biol 2004 11 1445 1453 15489171
Jain RP Pettersson HI Zhang J Aull KD Fortin PD Synthesis and evaluation of keto-glutamine analogues as potent inhibitors of severe acute respiratory syndrome 3CLpro J Med Chem 2004 47 6113 6116 15566280
Kao RY Tsui WH Lee TS Tanner JA Watt RM Identification of novel small-molecule inhibitors of severe acute respiratory syndrome-associated coronavirus by chemical genetics Chem Biol 2004 11 1293 1299 15380189
Tanner JA Zheng BJ Zhou J Watt RM Jiang JQ The adamantine-derived bananins are potent inhibitors of the helicase activities and replication of SARS coronavirus Chem Biol 2005 12 303 311 15797214
Wu CY Jan JT Ma SH Kuo CJ Juan HF Small molecules targeting severe acute respiratory syndrome human coronavirus Proc Natl Acad Sci U S A 2004 101 10012 10017 15226499
Steinhauer DA Holland JJ Direct method for quantitation of extreme polymerase error frequencies at selected single base sites in viral RNA J Virol 1986 57 219 228 3001347
Peiris JS Guan Y Yuen KY Severe acute respiratory syndrome Nat Med 2004 10 S88 S97 15577937
Guan Y Zheng BJ He YQ Liu XL Zhuang ZX Isolation and characterization of viruses related to the SARS coronavirus from animals in southern China Science 2003 302 276 278 12958366
Chinese SARS Molecular Epidemiology Consortium Molecular evolution of the SARS coronavirus during the course of the SARS epidemic in China Science 2004 303 1666 1669 14752165
Vijgen L Keyaerts E Moes E Thoelen I Wollants E Complete genomic sequence of human coronavirus OC43: Molecular clock analysis suggests a relatively recent zoonotic coronavirus transmission event J Virol 2005 79 1595 1604 15650185
van der Most RG Spaan WJ Siddell SG Coronavirus replication, transcription and RNA recombination The Coronaviridae 1995 New York Plenum 11 31
Kusters JG Jager EJ Niesters HG van der Zeijst BA Sequence evidence for RNA recombination in field isolates of avian coronavirus infectious bronchitis virus Vaccine 1990 8 605 608 1708184
Wang L Junker D Collisson EW Evidence of natural recombination within the S1 gene of infectious bronchitis virus Virology 1993 192 710 716 8380672
de Haan CA Masters PS Shen X Weiss S Rottier PJ The group-specific murine coronavirus genes are not essential, but their deletion, by reverse genetics, is attenuating in the natural host Virology 2002 296 177 189 12036329
Jeffers SA Tusell SM Gillim-Ross L Hemmila EM Achenbach JE CD209L (L-SIGN) is a receptor for severe acute respiratory syndrome coronavirus Proc Natl Acad Sci U S A 2004 101 15748 15753 15496474
Li W Moore MJ Vasilieva N Sui J Wong SK Angiotensin-converting enzyme 2 is a functional receptor for the SARS coronavirus Nature 2003 426 450 454 14647384
Hofmann H Pyrc K van der Hoek L Geier M Berkhout B Human coronavirus NL63 employs the severe acute respiratory syndrome coronavirus receptor for cellular entry Proc Natl Acad Sci U S A 2005 102 7988 7993 15897467
Marra MA Jones SJ Astell CR Holt RA Brooks-Wilson A The genome sequence of the SARS-associated coronavirus Science 2003 300 1399 1404 12730501
Rota PA Oberste MS Monroe SS Nix WA Campagnoli R Characterization of a novel coronavirus associated with severe acute respiratory syndrome Science 2003 300 1394 1399 12730500
Ziebuhr J Snijder EJ Gorbalenya AE Virus-encoded proteinases and proteolytic processing in the Nidovirales J Gen Virol 2000 81 853 879 10725411
Anand K Ziebuhr J Wadhwani P Mesters JR Hilgenfeld R Coronavirus main proteinase (3CLpro) structure: Basis for design of anti-SARS drugs Science 2003 300 1763 1767 12746549
Anand K Palm GJ Mesters JR Siddell SG Ziebuhr J Structure of coronavirus main proteinase reveals combination of a chymotrypsin fold with an extra alpha-helical domain EMBO J 2002 21 3213 3224 12093723
Yang H Yang M Ding Y Liu Y Lou Z The crystal structures of severe acute respiratory syndrome virus main protease and its complex with an inhibitor Proc Natl Acad Sci U S A 2003 100 13190 13195 14585926
Hegyi A Ziebuhr J Conservation of substrate specificities among coronavirus main proteases J Gen Virol 2002 83 595 599 11842254
Liu S Hanzlik RP Structure-activity relationships for inhibition of papain by peptide Michael acceptors J Med Chem 1992 35 1067 1075 1552501
Hanzlik RP Thompson SA Vinylogous amino acid esters: A new class of inactivators for thiol proteases J Med Chem 1984 27 711 712 6547487
Matthews DA Dragovich PS Webber SE Fuhrman SA Patick AK Structure-assisted design of mechanism-based irreversible inhibitors of human rhinovirus 3C protease with potent antiviral activity against multiple rhinovirus serotypes Proc Natl Acad Sci U S A 1999 96 11000 11007 10500114
Tian WX Tsou CL Determination of the rate constant of enzyme modification by measuring the substrate reaction in the presence of the modifier Biochemistry 1982 21 1028 1032 7074045
Roberts A Paddock C Vogel L Butler E Zaki S Aged BALB/c mice as a model for increased severity of severe acute respiratory syndrome in elderly humans J Virol 2005 79 5833 5838 15827197
Hayden FG Turner RB Gwaltney JM Chi-Burris K Gersten M Phase II, randomized, double-blind, placebo-controlled studies of ruprintrivir nasal spray 2-percent suspension for prevention and treatment of experimentally induced rhinovirus colds in healthy volunteers Antimicrob Agents Chemother 2003 47 3907 3916 14638501
Hsyu PH Pithavala YK Gersten M Penning CA Kerr BM Pharmacokinetics and safety of an antirhinoviral agent, ruprintrivir, in healthy volunteers Antimicrob Agents Chemother 2002 46 392 397 11796347
Ziebuhr J Heusipp G Siddell SG Biosynthesis, purification, and characterization of the human coronavirus 229E 3C-like proteinase J Virol 1997 71 3992 3997 9094676
Otwinowski Z Minor W Carter CW Jr Sweet RM Processing of X-ray diffraction data collected in oscillation mode Macromolecular crystallography, part A 1997 New York Academic Press 307 326
Brunger AT Adams PD Clore GM DeLano WL Gros P Crystallography & NMR system: A new software suite for macromolecular structure determination Acta Crystallogr D Biol Crystallogr 1998 54 905 921 9757107
Jones TA Zou JY Cowan SW Kjeldgaard Improved methods for binding protein models in electron density maps and the location of errors in these models Acta Crystallogr A 1991 47 110 119 2025413
Laskowski RA MacArthur MW Moss DS Thornton JM PROCHECK: A program to check the stereochemical quality of protein structures J Appl Cryst 1993 26 283 291
Meara JP Rich DH Measurement of individual rate constants of irreversible inhibition of a Cys proteinase by an epoxysuccinyl inhibitor Bioorg Med Chem Lett 1995 5 2277 2282
|
16128623
|
PMC1197287
|
CC BY
|
2021-01-05 08:21:27
|
no
|
PLoS Biol. 2005 Oct 6; 3(10):e324
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030324
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 1620707510.1371/journal.pbio.0030337Research ArticleDevelopmentDrosophilaInsectsArthropodsAnimalsEukaryotes
Drosophila Heartless Acts with Heartbroken/Dof in Muscle Founder Differentiation Muscle Fibre Formation in Adult DrosophilaDutta Devkanya
1
Shaw Sanjeev
1
Maqbool Tariq
1
Pandya Hetal
1
VijayRaghavan K [email protected]
1
1National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bangalore, IndiaLeptin Maria Academic EditorUniversity of CologneGermany10 2005 6 9 2005 6 9 2005 3 10 e33717 9 2004 29 7 2005 Copyright: © 2005 Dutta et al.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
A Signaling Pathway at the Heart of Muscle Development
The formation of a multi-nucleate myofibre is directed, in Drosophila, by a founder cell. In the embryo, founders are selected by Notch-mediated lateral inhibition, while during adult myogenesis this mechanism of selection does not appear to operate. We show, in the muscles of the adult abdomen, that the Fibroblast growth factor pathway mediates founder cell choice in a novel manner. We suggest that the developmental patterns of Heartbroken/Dof and Sprouty result in defining the domain and timing of activation of the Fibroblast growth factor receptor Heartless in specific myoblasts, thereby converting them into founder cells. Our results point to a way in which muscle differentiation could be initiated and define a critical developmental function for Heartbroken/Dof in myogenesis.
In the fly embryo, the founder cells that direct myofibre formation are selected through Notch-mediated signaling. The authors show that in adult animals, founder cells are specified by signaling through the FGF pathway.
==== Body
Introduction
Each multi-nucleate muscle fibre in an animal is uniquely positioned and performs a specific function. The development of these features is a consequence of the specification of the identity of the fibre, and its differentiation in the context of its innervation and attachment to tendon cells. In Drosophila, the identity of a muscle fibre is specified by the expression of a combination of transcription factors unique to each muscle and by its location [1–6]. In addition to characteristics that specify the identity of each muscle, all syncytial muscles in flies share a common mechanism of fibre formation. This common mechanism uses a special cell, the founder cell, that organises the fusion process and provides it directionality [1,4,7–9]. A founder cell attracts its neighbouring myoblasts—the “fusion-competent” myoblasts—that fuse with the founder, to form a multi-nucleate myotube. Specific molecules expressed in the founder and fusion-competent cells direct the fusion process [7,10]. One such molecule, expressed on the surface of the founder cells and encoded by the gene dumbfounded (duf, also called kin of irre or kirre), is an Ig domain containing membrane protein that interacts with other Ig domain proteins expressed on the surface of fusion-competent cells [11,12]. This interaction initiates the process of cell fusion. The selection of founder cells and the expression of duf in founder cells are thus important first steps in muscle differentiation.
In the embryo, a founder cell is selected from a cluster of equivalent myoblasts that are specified by activation of the Ras/mitogen-activated protein kinase (MAPK) pathway. From this “equivalent group”, a single cell—the precursor cell—is chosen by Notch-mediated lateral inhibition. The precursor cell divides to give rise to two embryonic founder cells or an embryonic founder cell and an adult myoblast progenitor [1,2,4,13]. Founder cells, thus specified, express transcription factors unique to each founder, but all founders express the duf-lacZ reporter gene [14]. The remaining cells of the equivalent group constitute the fusion-competent myoblasts.
The muscles of the adult fly are also made using the founder mechanism [15–17]. The formation of each adult muscle fibre appears to be nucleated by a duf-lacZ-expressing cell, and these cells exhibit features characteristic of founders [15]. In adult myogenesis, duf-lacZ expression in a founder pattern is currently the earliest available molecular marker for the formation of muscle founder. The monoclonal antibody 22C10 has also been shown to be a marker of muscle founder cells in the adult abdomen [15]. The antibody 22C10 recognises the microtubule-binding protein Futsch, which has been found to regulate microtubule architecture in neurons during development [18,19]. Futsch, perhaps, serves a similar function in the developing myofibres and its expression in founder cells could indicate this. The pattern of expression of duf-lacZ during myofibre formation and the function of duf-lacZ-expressing cells suggest that the founder cell mechanism is conserved through Drosophila development and could, perhaps, be even more generally applicable [15].
Significantly, though, the mechanism of founder cell selection during adult myogenesis is different from that used in the embryo: Notch-mediated lateral inhibition does not appear to be used in adult founder cell choice [15]. The embryonic origin of adult myoblasts [4,20,21]—they are clonally derived from siblings of embryonic founder cells—make this understandable, in hindsight. Indeed, in the third larval instar, adult myoblasts in the abdomen, seen associated with nerves, all express duf-lacZ at low levels [15]. During pupal development most myoblasts completely shut down duf-lacZ expression while some begin to express this reporter in a distinct founder cell pattern [15]. This is shown schematically in Figure 1. The question, then, in adult myogenesis is not how a founder cell is chosen from a group of apparently naïve myoblasts, but how founder properties are up-regulated in a defined set of myoblasts at the sites of fibre formation, while properties characteristic of fusion-competent cells are up-regulated in other cells.
Figure 1 Schematic Representation of the Appearance of Founder Cells in the Abdomen
(A) Late third instar larva. Precursor myoblasts for the adult abdominal muscles are associated with the segmental and inter-segmental nerves. These myoblasts express duf-lacZ in low levels.
(B) 13 h APF. The adult myoblasts, while continuing to express low levels of duf-lacZ, proliferate and migrate out along the nerve.
(C) 26 h APF. By this stage, low duf-lacZ expression in all myoblasts is replaced by higher levels of duf-lacZ in selected cells, the founder cells. The remaining cells, termed the fusion-competent cells, eventually fuse with the founder cells to form the adult myotubes.
(D) 28 h APF. The nascent myotubes, founded by the founder cells, can be visualised by staining with the antibody 22C10.
We show here that the maturation of adult founder cells in the abdomen of the fly is mediated by signalling through the Fibroblast growth factor (FGF) receptor Heartless (Htl). In particular, we demonstrate a novel interplay of components of FGF signalling that results in a precise pattern of founder cells for each multi-fibre array of muscles. This interplay involves the regulation of Sprouty (Sty) (a negative regulator of FGF signalling) and Heartbroken (Hbr) (a positive regulator required for Htl signalling).
Results
Our previous study had shown that Notch-mediated lateral inhibition does not function in the choice of adult founders [15]. We next investigated whether one of the receptor tyrosine kinase (RTK) family of proteins is involved in selection of adult founders. The family of FGF receptors (FGFRs) constitutes an important member of the RTK super-family, and we examined its role in adult founder cell selection. Drosophila has two FGFR-coding genes, htl [22] and breathless (btl) [23]. Htl plays a significant part in the patterning of a variety of mesodermal tissues in the embryo. It is required for cell migration during mesodermal invagination, specification of a subset of somatic muscle precursors, and differentiation of heart and midline glial cells [22,24,25]. In the pupa, htl mRNA is found to be expressed in the adult thoracic myoblasts [26]. In this study, we examined the involvement of Htl in adult founder selection. Our results illustrate the participation of the Htl pathway in selection of founder cells in the abdomen
Htl is Expressed in Adult Abdominal Myoblasts
In the third instar larva, we found that Htl was expressed in the Twist (Twi)–expressing adult abdominal myoblasts that are associated with nerves (Figure 2A–2C). Htl expression in these myoblasts was also observed during pupal development. Figure 2D–2I shows Htl expression at 18 h after puparium formation (APF) in the dorsal (Figure 2D–2F) and lateral (Figure 2G–2I) myoblasts. Double staining with anti-Htl and anti-β-galactosidase antibodies in duf-lacZ pupae at 28 h APF showed that Htl continued to be expressed in all myoblasts—in both the duf-lacZ-expressing founder cells and in fusion-competent myoblasts (Figure 2J–2L). We also examined the expression of the second Drosophila FGFR gene btl in the adult mesoderm using a btl-GAL4 driver. btl expression was not observed in abdominal myoblasts or in muscle founder cells (data not shown).
Figure 2 Expression of Htl in Abdominal Mesoderm
(A–I) twi-lacZ larval or pupal preparations double-labelled with anti-β-galactosidase (green) and anti-Htl (red). (A) twi-lacZ-expressing myoblasts in a third instar larva. (B) Same preparation as in (A), showing expression of Htl. (C) Merged image of (A) and (B) showing co-localisation of Htl with twi-lacZ. (D–F) A pupal preparation 18 h APF showing a dorsal cluster of twi-expressing myoblasts (D) also expressing Htl (E). (F) Merged image of (D) and (E). (G–I) Pupa 18 h APF showing a cluster of lateral myoblasts (G), all expressing Htl (H). (I) Merged image of (G) and (H). Htl expression is also observed in some epidermal cells. A few such cells are outlined in white in (H).
(J–L) A duf-lacZ pupal preparation 28 h APF double-labelled with anti-β-galactosidase (green) and anti-Htl (red). (J) A set of lateral founders expressing duf-lacZ. (K) Htl expression in the same preparation. (L) Merged image of (J) and (K). Htl is present in the founder cells (expressing duf-lacZ) and also in the remaining population of fusion-competent myoblasts.
Anterior is at top; dorsal midline is at right.
Scale Bar = 20 μm.
Modification of Htl Signalling Affects Founder Cell Number in the Abdomen
All available mutants of Htl are embryonic lethal, making scoring of Htl mutants for pupal or adult phenotypes difficult. We mis-expressed a dominant-negative construct of Htl (dnHtl) to examine the Htl loss-of-function effect in the adult myoblasts (Figure 3). Over-expression of UAS-dnhtl using 1151-GAL4, a GAL4 driver expressed in all adult myoblasts [27,28], resulted in a distinct decrease in the number of founders per hemi-segment. An average of four dorsal founders per hemi-segment, as opposed to wild-type numbers between 17 and 22, were observed in 49% of the hemi-segments, while dorsal founders were completely absent in 17% of these hemi-segments (n = 30). Lack of a complete suppression of founder cells in experimental pupae was most likely due to the incomplete penetrance of the dnHtl construct (also reported in [29]). Figure 3B and 3D show dorsal and lateral hemi-segments, respectively, of dnHtl mis-expressed pupae where the founder cells are completely lost. Htl function was additionally compromised by mis-expressing UAS-htl-RNAi using the 1151-GAL4 driver. In this case too, a decrease in founder number, similar to that in dnhtl mis-expression, was observed (data not shown). Yan, a downstream target of MAPK, functions as an RTK-pathway antagonist. It is an Ets-domain-containing transcription factor that keeps the RTK-responsive genes repressed. Phosphorylation of Yan by MAPK inactivates its function as a repressor [30–33]. We over-expressed an activated form of Yan (Yanact), which cannot undergo phosphorylation by MAPK [30], in adult myoblasts using the 1151-GAL4 driver. Yanact mis-expression resulted in a severe phenotype, with no founders being selected (Figure 3F) and no muscle fibres being formed (Figure 3H) in any hemi-segment. Figure 3E and 3G show the respective wild-type patterns.
Figure 3 Founder-Pattern upon Decrease in Htl Signaling
(A–D) Pupal preparations (grown at 29 °C) stained with 22C10 to visualise the founders. (A) A Canton-S pupa 30 h APF showing a subset of dorsal founders. (B) A similarly aged pupa of the genotype 1151/+; UAS-dnhtl/+; UAS-dnhtl/+ showing absence of dorsal founders. The white arrow points to a neuronal branching labelled by 22C10. (C) A preparation 28 h APF of Canton-S pupa showing lateral founders. (D) A 1151/+; UAS-dnhtl/+; UAS-dnhtl/+ pupa, similarly aged as in (C), showing no founders in the lateral hemi-segment.
(E and F) A preparation 28 h APF (grown at 29 °C) of a duf-lacZ pupa (E) and a 1151/duf-lacZ; UAS-Yanact/+ pupa (F) double-labelled with anti-β-galactosidase (red) and 22C10 (green). (E) shows wild-type pattern of a subset of dorsal founders. (F) shows complete absence of founder cells. Only the nerve, stained by 22C10, is visible.
(G and H) A preparation 42 h APF (grown at 29 °C) of a duf-lacZ pupa (G) and a 1151/duf-lacZ; UAS-Yanact/+ pupa (H) double-labelled with anti-β-galactosidase (red) and anti-MHC (green). In contrast to the wild-type pattern of dorsal muscles (G), no muscles are formed in the UAS-Yanact mis-expression pupa (H). The absence of muscles is not an experimental artefact because in the same preparation the unhistolysed larval muscles [66] can be viewed (white asterisks in [G] and [H]).
Anterior is at top; dorsal midline is at left.
Scale bar = 20 μm.
In a converse set of experiments we amplified Htl-mediated signal in myoblasts and obtained a diametrically opposite phenotype, i.e., an increase in number of founders. Over-expression of an activated form of Htl (UAS-λhtl), using the same 1151-GAL4 driver, led to an increased number of founders and, consequentially, an increased number of muscle fibres in the abdomen (Figure 4). The effect was most prominent for founders for the lateral muscles, where a 2-fold increase in number of founder cells per hemi-segment was observed. Figure 4B shows the excess lateral founders (26 in number) at 28 h APF compared to the control in Figure 4A (13 founders). The excess founder cells were longer and more stretched-out than the normal founders (see Discussion). Figure 4D and 4F show excessive lateral myofibres at 41 h APF (Figure 4C and 4E show the corresponding control preparations). In the dorsal muscles, the excess-founder phenotype was less prominent. Figure 4H shows a region of a dorsal hemi-segment in an activated-Htl pupa. The extra duf-lacZ-expressing dorsal founders (white arrows) are observed below the rest of the founders.
Figure 4 Increase in Founder and Fibre Number upon Activation of Htl Signalling in Myoblasts
(A and B) Pupae 28 h APF (grown at 29 °C) double-labelled with anti-β-galactosidase (red) and 22C10 (green). (A) duf-lacZ pupa showing wild-type pattern of founders in a section of lateral hemi-segment. The number of founders seen in this section is 13. (B) 1151/duf-lacZ; UAS-λhtl/+ pupa showing excess number of lateral founders (26) in an equivalent region.
(C and D) Pupae 42 h APF (at 29 °C) double-labelled with anti-β-galactosidase (red) and anti-MHC (green). (C) duf-lacZ pupa showing the wild-type lateral myofibres, each having one high duf-lacZ-expressing nucleus (white arrowheads). (D) 1151/duf-lacZ; UAS-λhtl/+ pupa showing extra number of lateral fibres. White arrowheads indicate high duf-lacZ-expressing nuclei within these fibres. For many fibres, in both (C) and (D), the high duf-lacZ-expressing nucleus is not within the field of view.
(E and F) DIC images of Canton-S (E) and 1151/+; UAS-λhtl/+ pupae labelled with anti-MHC at 42 h APF. In contrast to the neat array of wild-type muscle fibres (E), myofibres in UAS-λhtl mis-expressed pupa are heaped in bundles (F).
(G) Normal pattern of dorsal founders in duf-lacZ pupa at 28 hAPF (at 29 °C).
(H) Founder pattern in the dorsal hemi-segment of a 1151/duf-lacZ; UAS-λhtl/+ pupa similarly aged to that in (G). White arrows indicate the extra founder cells that have appeared below the normal set of fibres.
Anterior is at top; dorsal midline is at left.
Scale bar = 20 μm (A–D, G, and H), 13 μm (E and F).
Pointed (Pnt), an Ets family protein and a target of MAPK, functions as a positive regulator of the RTK pathway. Upon phosphorylation by MAPK, Pnt outcompetes Yan and turns on genes formerly repressed by Yan [32–34]. Over-expression of an active form of Pnt, therefore, should have the same effect as that of activated Htl. We found that mis-expression of PntP1, an isoform of Pnt [35,36], in adult myoblasts led to an increased number of founder cells compared to the wild-type number in the abdomen (data not shown).
In addition to the FGFRs, another member of the RTK family that is repeatedly used during development to direct cell fate choices is the Drosophila Epidermal growth factor receptor (DER) (reviewed in [37]). During embryonic myogenesis, DER provides inductive signals to specify equivalence group for muscle founder cell selection. In the absence of the function of DER or Spitz (a ligand for DER), a large subset of myofibres and their progenitors fail to form in the embryo [13,38] Since both DER and Htl function via the Ras pathway, we wanted to ascertain that the phenotypes observed for the adult founders were specific for the Htl pathway. We decreased DER signalling by mis-expressing a dominant-negative construct of DER (UAS-dnDER) using 1151. Separately, we also over-expressed the DER-inactivating ligand Argos (Aos) [39] in all myoblasts by driving UAS-aos using 1151-GAL4. In both cases, mis-expressed progeny had no abnormality in founder cell number and pattern when compared to wild-type (data not shown).
The above results ruled out the role of DER and confirmed the involvement of the Htl receptor in muscle founder cell choice. However, RTK pathways are also known to function in cell proliferation [40–43]. What, if any, is the contribution of such a role of Htl signalling to the phenotypes we see? The expectation if Htl is not involved in myoblast proliferation is that decreasing Htl signalling in myoblasts should not affect the number of fusion-competent myoblasts. However, since founders are selected from the Twi-expressing myoblast pool, absence of founder cell selection should correspondingly correlate with an increase in myoblast number. We examined the number of Twi-expressing myoblasts at 24 h APF in wild-type pupae and in pupae in which the Htl pathway was down-regulated (Figure S1). In these pupae, the number of myoblasts per hemi-segment did not show a significant change (lateral myoblasts [n = 5], wild-type, 58 ± 5, dnHtl, 60 ± 14; dorsal myoblasts [n = 4], wild-type, 96 ± 8, dnHtl, 85 ± 11). This ruled out the involvement of the Htl pathway in myoblast proliferation. However, because of the variation in number of myoblasts in control and in experimental pupae, it was not feasible to examine the conversion of specific founders into fusion-competent myoblasts. Activating the pathway using an activated-Htl construct caused a very slight increase in myoblast number (Figure S1) (lateral myoblasts [n = 5], wild-type, 58 ± 5, activated Htl, 68 ± 9; dorsal myoblasts [n = 4], wild-type, 96 ± 8, activated Htl, 122 ± 11). This effect is, however, still consistent with the Htl pathway not being involved in myoblast proliferation as activated Htl can activate downstream effectors common to an RTK pathway that controls proliferation. In order to examine the effect of GAL4/UAS-induced Htl activation specifically at the time of founder selection, we used a temperature-sensitive GAL80 (TARGET) system [44]. In this system, the activator function of GAL4 protein can be temporally regulated by a ubiquitously expressed temperature-sensitive GAL80 protein (GAL80ts) (functional at 18 °C, non-functional at 29 °C). When animals carrying the 1151-GAL4 driver, a UAS construct for the Htl pathway (either UAS-dnhtl, or UAS-λhtl), and a tubulin (tub)–GAL80ts gene were grown at 18 °C, we saw no effect on number of founders or fusion-competent myoblasts (Figure S2). This illustrated that the tub-GAL80ts was effective in preventing the function of GAL4 in the pupal mesoderm. The use of the GAL80ts conditional system thus allowed the Htl pathway to be modulated in a controlled background starting with a similar number of myoblasts. 1151-GAL4; tub-GAL80ts; UAS-dnhtl pupae, raised at 18 °C for 30 h followed by a heat shock at 29 °C for 11 h, were scored for founder and myoblast number. Down-regulation of Htl resulted in no founders being selected (Figure S2). However, the number of fusion-competent myoblasts in each hemi-segment did not show any significant increase (Figure S2) over that in the control (lateral myoblasts, control 54 ± 7, GAL4-GAL80ts-dnhtl 62 ± 8; n = 4). 1151-GAL4; tub-GAL80ts; UAS-λhtl pupae were grown similarly as above. The expectation was that the activation of Htl would result in an excess of founder cells at the expense of Twi-expressing cells. Indeed, when GAL80 was rendered inactive and the GAL4 used to activate Htl, an excess number of founders were seen (Figure S2). However, myoblast number was comparable to the control (lateral myoblasts in a hemi-segment, control 54 ± 7, GAL4-GAL80ts–activated Htl 62 ± 5; n = 4), though the pattern of the spatial location of myoblasts was found to be altered (Figure S2). Taken together, we can safely say that the effect of Htl signalling on founder number is a consequence of a direct involvement of the pathway in founder selection and not due to an indirect effect on myoblast number. However, given the large variation in control myoblast numbers (such that conversion of founder cells into fusion-competent myoblasts is not effectively reflected as an increase in myoblasts in dnHtl pupa), and given that not all myoblasts are converted to founders upon activating Htl, it is not feasible to deduce a precise and predictable relationship between numbers of myoblasts and founders.
Hbr/Dof, a Signalling Protein Specific for Drosophila FGFR-Mediated Pathway, Is Expressed in Founder Cells
The continued expression of Htl in adult myoblasts throughout pupal development suggested that restricted presence of receptor protein is not the mechanism behind selection of founder cells from the myoblast pool. We next examined the expression of the cytoplasmic protein Dof [45], also known as Hbr [46] or Stumps [47]. Unlike other intracellular signalling components common to the different RTK-mediated pathways, Hbr/Dof functions exclusively in the FGFR pathway and is essential for the activation of MAPK by the FGFRs [45,46].
From the late third instar larva till 24 h APF, Hbr/Dof was found to be expressed in all nerve-associated myoblasts. Figure 5A–5D show expression of Hbr/Dof in myoblasts at 18 h APF; starting 24 h APF, Hbr/Dof expression began to disappear from most myoblasts. By 28 h APF, Hbr/Dof expression was observed in only an array of cells, the founder cells. Figure 5E–5J show Hbr/Dof expression in a restricted number of twi-lacZ-expressing cells at 28 h APF while Figure 5K and 5L show Hbr/Dof expression in duf-lacZ-expressing founder cells at the same stage (i.e., 28 h APF).
Figure 5 Dof Expression during Abdominal Myogenesis
(A–D) twi-lacZ pupae 18 h APF double-labelled with anti-β-galactosidase (red) and anti-Dof (green). (A) A cluster of lateral myoblasts. (B) Dof expression in the same lateral cluster, showing expression in all myoblasts. (C) A cluster of dorsal myoblasts. (D) Dof expression in the dorsal cluster. Like in (B), Dof in (D) is expressed in all dorsal myoblasts at this stage.
(E–J) twi-lacZ pupae 28 h APF double-labelled with anti-β-galactosidase (red) and anti-Dof (green). (E) A cluster of twi-lacZ-expressing lateral myoblasts. (F) Dof expression in the same region. (G) Merged image of (E) and (F), showing Dof expression in only a subset of cells. (H) A cluster of twi-lacZ-expressing dorsal myoblasts. (I) Dof expression in the same region. (J) Merged image of (H) and (I), showing Dof expression only in selected cells within the cluster. In (G) and (J), the white arrows indicate a few myoblasts that do not express Dof.
(K and L) duf-lacZ pupae 28 h APF double-labelled with anti-β-galactosidase (red) and anti-Dof (green). The duf-lacZ-expressing lateral (K) and dorsal (L) founders express Dof. Thus, the cells that express Dof at 28 h APF are the founder cells.
Anterior is at top; dorsal midline is at right.
Scale bar = 30 μm.
Thus, the pattern of appearance of Hbr/Dof-expressing cells was synchronous with that of the appearance of founder cells. This led us to examine whether Hbr/Dof has a role in imparting founder identity to specific cells.
Mis-Expression of Hbr/Dof in All Myoblasts Leads to Increased Number of Founders
We maintained expression of Hbr/Dof in all myoblasts by crossing UAS-dof with 1151-GAL4. Hbr/Dof mis-expression led to a dramatic increase in the number of founder cells and muscle fibres in the abdomen compared to the wild-type (Figure 6). Figure 6B is a lateral hemi-segment of an 1151/duf-lacZ; UAS-dof/+ pupa, showing a 2-fold increase in number of founder cells (42) compared to that in a wild-type duf-lacZ pupa shown in Figure 6A (19). The excess founders were more elongated in shape than the wild-type founders, similar to the shape observed in activated-Htl pupae (see Figure 4B). Figure 6D shows a UAS-dof mis-expressed pupa at 42 h APF having an excess number of lateral myofibres compared to the number in an equivalent region in a similarly aged wild-type pupa (Figure 6C). There are approximately 23 high duf-lacZ-expressing nuclei within the field of view in Figure 6D, in contrast to the nine in the wild-type control in Figure 6C. These mis-expression results suggested that restricted expression of Hbr/Dof in pupal abdomen, beginning 24 h APF, is important for specification of adult founders.
Figure 6 Over-Expression of Hbr/Dof and Htl
(A–D) Increase in founder and fibre number upon over-expression of Hbr/Dof in myoblasts. (A and B) Pupae 28 h APF (grown at 29 °C) double-labelled with anti-β-galactosidase (red) and 22C10 (green). (A) duf-lacZ pupa showing wild-type pattern of founders in a lateral hemi-segment. There are 19 founders present in the field of view. (B) 1151/duf-lacZ; UAS-dof/+ pupa. Compared to the wild-type, an excess number of founder cells (∼42) are present in the mis-expression pupa. (C and D) Pupae 42 h APF (grown at 29 °C) double-labelled with anti-β-galactosidase (red) and anti-MHC (green). (C) duf-lacZ pupa showing lateral muscles in a hemi-segment. (D) 1151/duf-lacZ; UAS-dof/+ pupa showing excessive fibre formation.
(E–J) Expression of Dof levels upon activation of Htl. Preparations of duf-lacZ (E–G) and 1151/duf-lacZ; UAS-λhtl/+ (H–J) pupae double-labelled with anti-β-galactosidase (red) and anti-Dof (green) at 28 h APF (grown at 29 °C). For both pupae, images have been acquired at identical conditions. (E) Lateral founders in one hemi-segment of a duf-lacZ pupa. (F) Dof expression in the lateral founders. (G) Merged image of (E) and (F). (H) Lateral founders in an 1151/duf-lacZ; UAS-λhtl/+ pupa, showing an increase in founder number. (I) Dof expression in the same pupa. Dof is expressed in the extra founder cells. Also, level of Dof protein in the founders is higher than in the wild-type founders (in F). (J) Merged image of (H) and (I).
Anterior is at top; dorsal midline is at right.
Scale = 30 μm (A, B, and E–J), 20 μm (C and D).
Hbr/Dof Expression Persists in UAS-λhtl Pupae
We next asked whether the restricted expression of Hbr/Dof is generated by a feedback relationship involving Htl signalling. Such a feedback mechanism has been observed in the embryo where the Ras pathway up-regulates the expression of Hbr/Dof in the muscle progenitor cells [48]. We over-expressed the activated-Htl construct in adult myoblasts (using the 1151-GAL4 driver) and examined the pattern and levels of expression of the Hbr/Dof protein. Over-expression of UAS-λhtl resulted in a continued expression of Dof in cells, at a level higher than normal. Figure 6E–6G show a 28 h APF duf-lacZ pupa expressing Hbr/Dof in founder cells. In a similarly staged 1151/duf-lacZ; UAS-λhtl/+ pupa, Dof expression was observed in the extra founder cells (Figure 6H–6J).
sty, Initially Present in All Myoblasts, Disappears at the Founder Cell Stage
Both Htl and Hbr/Dof are expressed in undifferentiated myoblasts prior to the founder cell selection stage. However these myoblasts do not up-regulate duf-lacZ expression—the characteristic property of adult founders—showing that Htl, if active, does not specify founders at this stage. This could be because the ligand is not available or because of the presence of negative regulators. Dof-mediated mis-expression resulted in an increased founder and fibre number phenotype. Since Hbr/Dof function is ligand-dependent, this implies that the ligand, like the receptor, is generally available. Thus, there must be another factor that prevents premature receptor activation during normal development. We tested whether this factor might be encoded by sty. Sty is a negative regulator of FGF signalling [49–51]. We examined sty-lacZ expression in adult myoblasts. In the third larval instar, sty-lacZ expression was seen in all nerve-associated adult myoblasts (Figure 7). This expression declined gradually in the pupal stages. Figure 7B shows that sty-lacZ expression occurs only in a subset of cells at 18 h APF. By 28 h APF, when founders are chosen in the abdomen, sty-lacZ expression was not detected in adult myoblasts (Figure 7C and 7D). We do not know how exactly the expression of sty-lacZ correlates with that of Sty. However, these results are suggestive of a negative regulatory role of Sty that prevents premature receptor activation.
Figure 7 sty and Founder Cells
(A and B) sty-lacZ third instar larva (A) and pupa 20 h APF (B) double-labelled with anti-Twi (red) and anti-β-galactosidase (green). (A) A dorsal cluster of Twi-expressing myoblasts, all expressing sty-lacZ. (B) A dorsal cluster (yellow arrow) and a lateral cluster (yellow arrowhead) of myoblasts in a 20 h APF pupa. Not all myoblasts express sty-lacZ at this stage.
(C and D) sty-lacZ pupa 28 h APF double-labelled with 22C10 (red) and anti-β-galactosidase (green). sty-lacZ is not expressed in the dorsal (C) or lateral (D) set of founders.
(E–H) Pupae 28 h APF (grown in 29 °C) double-labelled with 22C10 (green) and anti-β-galactosidase (red). (E and G) duf-lacZ pupae showing sets of dorsal (E) and lateral (G) founders. (F and H) 1151/duf-lacZ; UAS-sty/+ pupae showing decreased number of founders in the dorsal (F) and lateral (H) hemi-segments. White arrows in (E) and (F) indicate 22C10-labelled neuronal branching.
Anterior is at top; dorsal midline is at left.
Scale = 30 μm.
Over-Expression of sty in Myoblasts Leads to Absence of Founder Cells
We maintained expression of sty in myoblasts by expressing UAS-sty under the 1151-GAL4 driver. sty mis-expression led to a discernible decrease in the number of founder cells. In contrast to wild-type numbers of about 17–22 dorsal founders and about 20 lateral founders per hemi-segment, 1151/duf-lacZ; UAS-sty/+ pupae had an average of seven dorsal founders and eight lateral founders per hemi-segment (n = 15). Figure 7F and 7H show, respectively, the decreased number of dorsal and lateral founders in 1151/UAS-sty pupae in comparison to the controls (Figure 7E and 7G). This suggested that down-regulation of sty is a prerequisite for cells to adopt a founder fate. We examined the status of the Twi-expressing myoblasts in 1151/+; UAS-sty/+ pupae at 28 h APF (Figure S1D). Sty over-expression produced no comparable change in the number of fusion-competent myoblasts when compared to the control (dorsal myoblasts, wild-type, 96 ± 8, UAS-sty, 84 ± 5, n = 4; qualitatively similar results obtained for lateral myoblasts). This result, along with the results of sty-lacZ expression and sty mis-expression, support a direct role of sty in founder cell selection.
Discussion
During somatic myogenesis in the Drosophila embryo, combinatorial functions of the Wingless, Decapentaplegic, and Ras pathways determine domains of mesodermal cells in each segment, from which a single precursor cell is chosen by Notch-mediated lateral inhibition [1,2,21]. The daughters of the precursor cell form two embryonic muscle founder cells—each with a characteristic pattern of expression of markers that specify its identity—or they form an embryonic muscle founder cell and an adult myoblast progenitor [1]. This latter cell type proliferates during larval life and its progeny, the adult myoblasts, are associated with imaginal discs and larval nerves. While embryonic founder cells shut down the expression of Twi, a marker of myoblast identity, the adult myoblasts retain Twi expression during their proliferative phase during larval life [52]. At the onset of metamorphosis, Twi levels decline in a group of cells, the adult founders, that express duf-lacZ at high levels and are located at the sites of myofibre formation [15]. Twi expression is also shut off in other myoblasts as they fuse with the founder to form multi-nucleate cells [53].
Interestingly, adult myoblasts, like the embryonic founders from whose siblings they are derived, express duf-lacZ (albeit at low levels) throughout larval life [15]. As adult muscle differentiation begins, this low-level expression changes dramatically to a pattern in which one founder cell—expressing duf-lacZ at high levels—is chosen to seed each muscle fibre [15]. How is this founder cell chosen? We have shown earlier that removal of Notch signalling in adult myoblasts does not result in an increase in the number of founders [15]. This suggested that lateral inhibition mediated by Notch, the process that operates in the embryo, is not the mechanism by which adult founders are chosen. Indeed, the requirements are quite different, for adult myoblasts all express duf-lacZ at low levels, suggesting—consistent with their origins as siblings of embryonic founders—that they all already have some properties similar to founder cells. In choosing adult founder cells, therefore, duf-lacZ is to be up-regulated in cells that will become founders and down-regulated in others that will become fusion-competent cells. Our results show that the Htl pathway plays a key role in choosing adult founders. We also suggest that Htl does this using an unusual mechanism in which an intracellular positive regulator plays an important role.
Adult myoblasts in the third larval instar express Twi, Hbr/Dof, Htl, and sty-lacZ. At the onset of adult abdominal myogenesis, Twi expression declines. With this, the expression of Hbr and Sty declines in myoblasts. We suggest that, in the third instar larva, the presence of Sty prevents the activation of the Htl receptor, even if the ligand and Hbr/Dof are available. However, since both Hbr/Dof and sty-lacZ expression decline with Twi, at the onset of myogenesis, the Htl receptor will still be unable to function, because Hbr/Dof is necessary for the function of the Htl receptor. We would like to suggest that, as Sty and Hbr/Dof expression decline (as Twi expression shuts down at the onset of myogenesis), the Htl receptor is active in some myoblasts. Htl signalling maintains Hbr/Dof expression in these cells by a positive feedback mechanism. Maintenance of Hbr/Dof expression reinforces the Htl signal, which in turn up-regulates the expression of founder-specific genes such as duf in these cells, thereby imparting them with founder properties. Consistent with this hypothesis, activating the Htl receptor results in the maintenance of Hbr/Dof in adult myoblasts. This prolonged activation of Hbr/Dof, and therefore of duf, could be the cause of morphological changes associated with the excess founder cells, as observed in Figures 4B and 6B.
How could this localised activation of the receptor occur? One way is via the localised availability of the Htl ligand. Proximity of some of the cells to the source of the ligand could cause higher levels of Htl signalling in those cells than others, thus biasing their fate towards that of a founder. Examining the expression pattern of the recently identified ligands of Htl [54,55] should be able to resolve whether this indeed is the case. A second, and more likely, mechanism for localised activation of receptor is via a process that does not involve the localised presence of the ligand. We suggest this possibility because the continued mis-expression of Hbr/Dof in all adult myoblasts results in an increased number of founders and muscle fibres. Since Hbr/Dof function is dependent on ligand activation of the receptor, the ligand must be available to Htl on all myoblasts. Local activation of the receptor could occur by Hbr/Dof being maintained briefly in a founder cell pattern in some myoblasts even as all of the others down-regulate Sty and Hbr/Dof at the onset of myogenesis (with the decline of Twi expression). This continued expression of Hbr/Dof in some myoblasts, and the absence of Sty, could allow local activation of the receptor and the consequent maintenance of Hbr/Dof in a founder pattern.
The problem then shifts to deciphering the mechanism by which the (hypothetical) localised activation of Hbr/Dof takes place. Since abdominal myoblasts are associated with nerves, one possibility is that the signal could come from the nerves. This “solution” has two problems, however. First, it is not clear how a precise periodicity of signal, expressed along the nerve and seen by associated myoblasts, would be generated to organise the correct spacing of founder cells. More pertinent perhaps is the observation that surgical removal of the nerve does not affect the number of muscle fibres [56]. Thus, nerves are unlikely to be the source for the signal that organises myoblasts in a founder pattern. Another possible source for a signal that maintains and elevates Hbr/Dof expression in a founder pattern could be the epidermis. The abdominal epidermis develops from ectodermal cells, the histoblasts [57]. As the epidermis differentiates during metamorphosis, muscle tendon precursor cells—specified by and expressing the stripe locus—can be identified [58]. The tendon precursor cells, given that they are in proximity to the differentiating myoblasts, could possibly be a source of organising signal that modulates Hbr/Dof expression to a founder pattern. Thus, the precise segmental and regional patterning of the epidermis could organise the pattern of founder cells in the developing abdominal musculature. In favour of this hypothesis is the finding that reduction of stripe-expressing cells in the dorsal thoracic disc results in the reduction of duf-lacZ expression in the larval templates that give rise to the thoracic dorsal longitudinal muscles [53], and increasing stripe expression in the ectoderm results in the increase of duf-lacZ expression in the developing dorsal longitudinal muscles (Arjumand Ghazi and K. VijayRaghavan, unpublished data). We do not know yet if these results apply to the abdomen.
A third possible mechanism of localised activation of Htl, not exclusive of either of the ones mentioned earlier, is that a dynamic interaction between ligands, other activators, and repressors results in the activation of Htl in a specific pattern. Such a process has been described in the embryo, e.g., in the anterior patterning of follicle cells in the Drosophila egg [59].
In conclusion, while many mechanistic details still remain elusive, the implication of the FGF pathway as a key player in adult founder cell choice provides the molecular tools to identify missing elements in the pathway. Integrated within the broad question of founder cell specification are more specific questions pertinent to the different muscle groups. Activation of Htl signalling produces a less prominent effect on the dorsal muscles than on the lateral muscles. Also, the extra founders of the dorsal muscles are located in a characteristic fashion (altered in orientation; see Figure 4H) that is different from that observed for the excess lateral founders. These observations raise questions about whether the dorsal and lateral groups of founders have different levels of sensitivity to the FGF pathway and whether they employ the pathway in different ways.
Our results allow the testing of whether this pathway operates in a similar manner during myogenesis in other contexts in Drosophila and in other animals, in particular the higher vertebrates. Vertebrate muscles are composed of multiple fibres, which make them similar to Drosophila adult muscles [15,60]. Vertebrate myogenesis shares several features with Drosophila myogenesis, at the level of genetic and molecular regulatory mechanisms [28,61]. The FGF pathway in vertebrates, mediated by multiple isoforms of the receptor and the ligand, has been found to play an instructive role in induction and commitment of myogenic cells. In Xenopus, for instance, an FGF-mediated pathway controls specification and differentiation of myotomal progenitors [62]. Also, signalling via FGFR4 positively regulates myogenic differentiation during avian limb muscle development [63]. The present study, showing the role of Htl in muscle differentiation, highlights yet another similarity. Our study also provides directions for probing how the number and location of fibres are regulated in vertebrates, questions that remain to be resolved in the field of vertebrate myogenesis.
Materials and Methods
Fly strains
To follow duf expression, the enhancer-trap line rp298 [14], which has a P element nuclear-localising lacZ inserted within the promoter region of duf, was used [11]. The GAL4-UAS system [64] was used for directed expression of genes during adult myogenesis. The 1151-GAL4 enhancer-trap strain, obtained from L. S. Shashidhara (Centre for Cellular and Molecular Biology, Hyderabad, India), is expressed in all adult myoblasts associated with the imaginal discs and nerves in the larvae [27,65]. UAS-htl-RNAi was obtained from Arno Müller (Institute für Genetik, University of Dusseldorf, Germany). UAS-yanac
t, UAS-sty, and sty-lacZ were gifts from Ben-Zion Shilo (Weizmann Institute of Science, Rehovot, Israel). UAS-dof was a gift from Maria Leptin (Institute für Genetik, University of Cologne, Germany). UAS-GFPN-lacZ/Cyo; btl-GAL4/TM3, Sb Ser was a gift from Shigeo Hayashi (Riken Center for Developmental Biology, Kobe, Japan). The following stocks were obtained from the Bloomington Drosphila Stock Center (http://flystocks.bio.indiana.edu/: UAS-dnDER, UAS-aos, UAS-dnhtl, UAS–activated htl (or UAS-λhtl), UAS-pntP1, and tub-GAL80ts. The various UAS strains (except the strains UAS-dnhtl, UAS-dnDER, and UAS-aos, which have double copy of the UAS insert) were crossed into the background of duf-lacZ using standard genetic techniques. For characterising the expression profile of genes and proteins, fly stocks and crosses were grown at 25 °C. For the GAL4-UAS mis-expression studies, progeny of the crosses and their controls were grown at 25 °C until early second instar stages, and then shifted to 29 °C until the required pupal stages.
Tissue preparation
Wandering third instar larvae were collected for larval dissections. White pre-pupae (0 h APF) were collected and grown at appropriate temperatures for desired times prior to dissection. The pupal and larval tissues were prepared for immunohistochemistry as described previously [53]. The larval or pupal preparations were mounted in 70% glycerol for DAB (di-amino benzidine)–stained preparations, or in Vectashield mounting medium (Vector Labs, Burlingame, California, United States) for fluorescent-labelled preparations.
Immunohistochemistry
Mouse anti-β-galactosidase antibody and 22C10 (both from Developmental Studies Hybridoma Bank; http://www.uiowa.edu/~dshbwww/) were used at a dilution of 1:50. Rabbit anti-β-galactosidase antibody (obtained from Molecular Probes, Eugene, Oregon, United States) was used at a dilution of 1:5,000. Rabbit anti-Twi antibody, a gift from Siegfried Roth (University of Cologne), and Rabbit anti–Myosin heavy chain (MHC) antibody, a gift from Dan Kiehart (Duke University, Durham, North Carolina, United States), were used at a dilution of 1:500. Rabbit anti-Dof antibody, a gift from Maria Leptin (Institute für Genetik, University of Cologne), was used at a dilution of 1:200. Rabbit anti-Htl antibody, a gift from Alan Michelson (Brigham and Women's Hospital, Boston, Massachusetts, United States), was pre-adsorbed at a dilution of 1:2,000 and used. For the anti-Htl antibody, the signal was amplified by Tyramide amplification kit (TSA, NEN Life Science Products, Boston, Massachusetts, United States). Tyramide was diluted in the amplification solution at a dilution of 1:50, and the streptavidin-conjugated Alexa 568 was used at a dilution of 1:200. Secondary antibodies conjugated to Alexa Fluor dyes (Molecular Probes) were used at a dilution of 1:200—Alexa 488 for green and Alexa 568 for red labelling.
Microscopy and cell count
Fluorescent preparations were scanned using a confocal microscope (MRC-1024, Bio-Rad Laboratories, Hercules, California, United States), and images were analysed using the software Metamorph (version 4.5) (Molecular Devices, Sunnyvale, California, United States). The DAB-stained preparations were examined using a Nikon (Tokyo, Japan) Eclipse E1000 microscope. For myoblast counts in wild-type and experimental pupae, abdominal segments A2 to A6 were considered.
Supporting Information
Figure S1 Status of Twi-Expressing Myoblasts upon Modification in Levels of Htl Signalling
Fluorescent preparations of pupae, grown for 28 h at 29 °C, stained with anti-Twi antibody to label the myoblasts. A cluster of dorsal myoblasts is in view in each of the images. See text for supporting quantitative data for each experiment.
(A) Wild-type pupa.
(B) 1151/+; UAS-dnhtl/+; UAS-dnhtl/+ pupa. The number of myoblasts remains unaffected.
(C) 1151/+; UAS-λhtl/+ pupa. A moderate increase in myoblast numbers is observed in this case.
(D) 1151/+; UAS-Sty/+ pupa. The number of myoblasts remains largely unaffected.
(3.9 MB TIF).
Click here for additional data file.
Figure S2 Founder Number and Myoblast Number upon Modification of Htl Levels Using the tub-GAL80ts System
Fluorescent images of pupae stained with antibodies 22C10 (red), to mark the founders, and anti-Twi (green), to mark the myoblasts. tub-GAL80ts [44] strain was put in the background of 1151-GAL4 using standard genetic techniques. The temperature-sensitive GAL80 protein is functional at 18 °C and represses the activating function of GAL4 protein. At a higher temperature (29 °C), the GAL80ts becomes non-functional and the GAL4 can activate the genes downstream of UAS sequence [44]. 1151-GAL4; tub-GAL80ts stock was crossed, separately, to UAS-λhtl and UAS-dnhtl stocks. The progeny of the above crosses were grown for 30 h at 18 °C, followed by 11 h at 29 °C. This timing corresponds to a stage prior to founder selection in wild-type. Pupae of 1151-GAL4; tub-GAL80ts were also treated similarly, to serve as control. To check whether the GAL80ts protein was functional in the pupal mesoderm in particular, 1151-GAL4, UAS-λhtl, tub-GAL80ts pupae were grown for 50 h at 18 °C followed by 4 h at 29 °C.
(A) 1151-GAL4/+; tub-GAL80ts/+ pupa grown for 30 h at 18 °C followed by 11 h at 29 °C (control). A set of lateral founders (one of them indicated by white arrow) is in view. The Twi-expressing myoblasts are seen aligned over the founders.
(B) 1151-GAL4; tub-GAL80ts; UAS-λhtl pupa grown for 50 h at 18 °C followed by 4 h at 29 °C. At 18 °C, the GAL80 protein is functional and represses GAL4 activation of UAS-λhtl. The founders and Twi-expressing fusion-competent myoblasts are present in wild-type pattern.
(C) 1151-GAL4; tub-GAL80ts; UAS-λhtl pupa grown for 30 h at 18 °C, followed by 11 h at 29 °C (i.e., similarly treated as in [A]). Founders are present in clusters (indicated by white arrows). The fusion-competent myoblasts are not aligned in a pattern similar to that observed in (A) or (B), but their number does not change significantly (see text).
(D) Dorsal region of an 1151-GAL4; tub-GAL80ts; UAS-dnhtl pupa similarly treated as in (A). Twi-expressing cells are present (white arrowhead) but founders are not observed.
For (A–C), anterior is at left; dorsal midline is at top. For (D), anterior is at top; dorsal midline is at left.
(2.5 MB TIF).
Click here for additional data file.
We would like to thank Mary Baylies, Maria Leptin, Alan Michelson, Benny Shilo, Siegfried Roth, Daniel Kiehart, Arno Muller, Helen Skaer, and Veronica Rodrigues for their generous help with reagents, suggestions, and advice. We also thank Giriraj Sharma for his help in confocal microscopy and Sapan Gandhi for his assistance in a few experiments. This work was supported by grants from the Department of Biotechnology and National Centre for Biological Sciences to KVR and a Kanwal Rekhi Fellowship to DD.
Competing interests. The authors have declared that no competing interests exist.
Author contributions. DD, SS, and KV conceived and designed the experiments. DD, SS, TW, and HP performed the experiments. DD and KV analyzed the data. DD and KV wrote the paper.
Citation: Dutta D, Shaw S, Maqbool T, Pandya H, VijayRaghavan K (2005) Drosophila Heartless acts with Heartbroken/Dof in muscle founder differentiation. PLoS Biol 3(10): e337.
Abbreviations
AosArgos
APFafter puparium formation
btl
breathless
duf
dumbfounded
DER
Drosophila Epidermal growth factor receptor
dnHtldominant-negative construct of Heartless
DofDownstream of fibroblast growth factor receptor
FGFFibroblast growth factor
FGFRFibroblast growth factor receptor
GAL80tstemperature-sensitive GAL80
HbrHeartbroken
HtlHeartless
MAPKmitogen-activated protein kinase
MHCMyosin heavy chain
PntPointed
RTKreceptor tyrosine kinase
tub
tubulin
StySprout
TwiTwist
UAS-dnDERdominant-negative construct of Drosophila Epidermal growth factor receptor
Yanactactivated Yan
==== Refs
References
Baylies MK Michelson AM Invertebrate myogenesis: Looking back to the future of muscle development Curr Opin Genet Dev 2001 11 431 439 11448630
Carmena A Baylies M Sink H Development of the larval somatic musculature Muscle development in Drosophila 2005 Georgetown (Texas) Eurekah
Frasch M Controls in patterning and diversification of somatic muscles during Drosophila embryogenesis Curr Opin Genet Dev 1999 9 522 529 10508697
Baylies MK Bate M Ruiz Gomez M Myogenesis: A view from Drosophila
Cell 1998 93 921 927 9635422
Ruiz Gomez M Bate M Segregation of myogenic lineages in Drosophila requires numb Development 1997 124 4857 4866 9428422
Bate M Rushton E Frasch M A dual requirement for neurogenic genes in Drosophila myogenesis 1993 149 161 Dev Suppl
Taylor MV Muscle differentiation: Signalling cell fusion Curr Biol 2003 13 R964 R966 14680655
Taylor MV
Drosophila development: Novel signal elicits visceral response Curr Biol 2002 12 R102 R104 11839291
Dworak HA Sink H Myoblast fusion in Drosophila
Bioessays 2002 24 591 601 12111720
Dworak HA Charles MA Pellerano LB Sink H Characterization of Drosophila hibris, a gene related to human nephrin Development 2001 128 4265 4276 11684662
Ruiz-Gomez M Coutts N Price A Taylor MV Bate M
Drosophila dumbfounded: A myoblast attractant essential for fusion Cell 2000 102 189 198 10943839
Strunkelnberg M Bonengel B Moda LM Hertenstein A de Couet HG rst and its paralogue kirre act redundantly during embryonic muscle development in Drosophila
Development 2001 128 4229 4239 11684659
Carmena A Gisselbrecht S Harrison J Jimenez F Michelson AM Combinatorial signaling codes for the progressive determination of cell fates in the Drosophila embryonic mesoderm Genes Dev 1998 12 3910 3922 9869644
Nose A Isshiki T Takeichi M Regional specification of muscle progenitors in Drosophila The role of the msh homeobox gene Development 1998 125 215 223 9486795
Dutta D Anant S Ruiz-Gomez M Bate M VijayRaghavan K Founder myoblasts and fibre number during adult myogenesis in Drosophila
Development 2004 131 3761 3772 15262890
Kozopas KM Nusse R Direct flight muscles in Drosophila develop from cells with characteristics of founders and depend on DWnt-2 for their correct patterning Dev Biol 2002 243 312 325 11884040
Rivlin PK Schneiderman AM Booker R Imaginal pioneers prefigure the formation of adult thoracic muscles in Drosophila melanogaster
Dev Biol 2000 222 450 459 10837132
Hummel T Krukkert K Roos J Davis G Klambt C
Drosophila Futsch/22C10 is a MAP1B-like protein required for dendritic and axonal development Neuron 2000 26 357 370 10839355
Roos J Hummel T Ng N Klambt C Davis GW
Drosophila Futsch regulates synaptic microtubule organization and is necessary for synaptic growth Neuron 2000 26 371 382 10839356
Carmena A Murugasu-Oei B Menon D Jimenez F Chia W Inscuteable and numb mediate asymmetric muscle progenitor cell divisions during Drosophila myogenesis Genes Dev 1998 12 304 315 9450926
Ruiz-Gomez M Romani S Hartmann C Jackle H Bate M Specific muscle identities are regulated by Kruppel during Drosophila embryogenesis Development 1997 124 3407 3414 9310335
Beiman M Shilo BZ Volk T Heartless, a Drosophila FGF receptor homolog, is essential for cell migration and establishment of several mesodermal lineages Genes Dev 1996 10 2993 3002 8957000
Klambt C Glazer L Shilo BZ breathless, a Drosophila FGF receptor homolog, is essential for migration of tracheal and specific midline glial cells Genes Dev 1992 6 1668 1678 1325393
Wilson R Leptin M Fibroblast growth factor receptor-dependent morphogenesis of the Drosophila mesoderm Philos Trans R Soc Lond B Biol Sci 2000 355 891 895 11128983
Gisselbrecht S Skeath JB Doe CQ Michelson AM heartless encodes a fibroblast growth factor receptor (DFR1/DFGF-R2) involved in the directional migration of early mesodermal cells in the Drosophila embryo Genes Dev 1996 10 3003 3017 8957001
Emori Y Saigo K Distinct expression of two Drosophila homologs of fibroblast growth factor receptors in imaginal discs FEBS Lett 1993 332 111 114 8405423
Anant S Roy S VijayRaghavan K Twist and Notch negatively regulate adult muscle differentiation in Drosophila
Development 1998 125 1361 1369 9502718
Roy S VijayRaghavan K Muscle pattern diversification in Drosophila The story of imaginal myogenesis Bioessays 1999 21 486 498 10402955
Michelson AM Gisselbrecht S Zhou Y Baek KH Buff EM Dual functions of the heartless fibroblast growth factor receptor in development of the Drosophila embryonic mesoderm Dev Genet 1998 22 212 229 9621429
Rebay I Rubin GM Yan functions as a general inhibitor of differentiation and is negatively regulated by activation of the Ras1/MAPK pathway Cell 1995 81 857 866 7781063
Hsu T Schulz RA Sequence and functional properties of Ets genes in the model organism Drosophila
Oncogene 2000 19 6409 6416 11175357
Brunner D Ducker K Oellers N Hafen E Scholz H The ETS domain protein pointed-P2 is a target of MAP kinase in the sevenless signal transduction pathway Nature 1994 370 386 389 8047146
Lai ZC Rubin GM Negative control of photoreceptor development in Drosophila by the product of the yan gene, an ETS domain protein Cell 1992 70 609 620 1505027
O'Neill EM Rebay I Tjian R Rubin GM The activities of two Ets-related transcription factors required for Drosophila eye development are modulated by the Ras/MAPK pathway Cell 1994 78 137 147 8033205
Klambt C The Drosophila gene pointed encodes two ETS-like proteins which are involved in the development of the midline glial cells Development 1993 117 163 176 8223245
Scholz H Deatrick J Klaes A Klambt C Genetic dissection of pointed, a Drosophila gene encoding two ETS-related proteins Genetics 1993 135 455 468 8244007
Shilo BZ Signaling by the Drosophila epidermal growth factor receptor pathway during development Exp Cell Res 2003 284 140 149 12648473
Buff E Carmena A Gisselbrecht S Jimenez F Michelson AM Signalling by the Drosophila epidermal growth factor receptor is required for the specification and diversification of embryonic muscle progenitors Development 1998 125 2075 2086 9570772
Schweitzer R Howes R Smith R Shilo BZ Freeman M Inhibition of Drosophila EGF receptor activation by the secreted protein Argos Nature 1995 376 699 702 7651519
Wiedlocha A Sorensen V Signaling, internalization, and intracellular activity of fibroblast growth factor Curr Top Microbiol Immunol 2004 286 45 79 15645710
Schlessinger J Cell signaling by receptor tyrosine kinases Cell 2000 103 211 225 11057895
Klint P Claesson-Welsh L Signal transduction by fibroblast growth factor receptors Front Biosci 1999 4 D165 D177 9989949
Jorissen RN Walker F Pouliot N Garrett TP Ward CW Epidermal growth factor receptor: Mechanisms of activation and signalling Exp Cell Res 2003 284 31 53 12648464
McGuire SE Mao Z Davis RL Spatiotemporal gene expression targeting with the TARGET and gene-switch systems in Drosophila
Sci STKE 2004 2004 pl6 14970377
Vincent S Wilson R Coelho C Affolter M Leptin M The Drosophila protein Dof is specifically required for FGF signaling Mol Cell 1998 2 515 525 9809073
Michelson AM Gisselbrecht S Buff E Skeath JB Heartbroken is a specific downstream mediator of FGF receptor signalling in Drosophila
Development 1998 125 4379 4389 9778498
Imam F Sutherland D Huang W Krasnow MA stumps, a Drosophila gene required for fibroblast growth factor (FGF)-directed migrations of tracheal and mesodermal cells Genetics 1999 152 307 318 10224263
Carmena A Buff E Halfon MS Gisselbrecht S Jimenez F Reciprocal regulatory interactions between the Notch and Ras signaling pathways in the Drosophila embryonic mesoderm Dev Biol 2002 244 226 242 11944933
Kramer S Okabe M Hacohen N Krasnow MA Hiromi Y Sprouty: A common antagonist of FGF and EGF signaling pathways in Drosophila
Development 1999 126 2515 2525 10226010
Hacohen N Kramer S Sutherland D Hiromi Y Krasnow MA sprouty encodes a novel antagonist of FGF signaling that patterns apical branching of the Drosophila airways Cell 1998 92 253 263 9458049
Casci T Vinos J Freeman M Sprouty, an intracellular inhibitor of Ras signaling Cell 1999 96 655 665 10089881
Bate M Rushton E Currie DA Cells with persistent twist expression are the embryonic precursors of adult muscles in Drosophila
Development 1991 113 79 89 1765010
Fernandes J Bate M VijayRaghavan K Development of the indirect flight muscles of Drosophila
Development 1991 113 67 77 1765009
Gryzik T Muller HA FGF8-like1 and FGF8-like2 encode putative ligands of the FGF receptor Htl and are required for mesoderm migration in the Drosophila gastrula Curr Biol 2004 14 659 667 15084280
Stathopoulos A Tam B Ronshaugen M Frasch M Levine M pyramus and thisbe: FGF genes that pattern the mesoderm of Drosophila embryos Genes Dev 2004 18 687 699 15075295
Currie DA Bate M Innervation is essential for the development and differentiation of a sex-specific adult muscle in Drosophila melanogaster
Development 1995 121 2549 2557 7671818
Fristrom JW Doctor J Fristrom DK Logan WR Silvert DJ The formation of the pupal cuticle by Drosophila imaginal discs in vitro Dev Biol 1982 91 337 350 6807731
Fernandes JJ Celniker SE VijayRaghavan K Development of the indirect flight muscle attachment sites in Drosophila : Role of the PS integrins and the stripe gene Dev Biol 1996 176 166 184 8660859
Wasserman JD Freeman M An autoregulatory cascade of EGF receptor signaling patterns the Drosophila egg Cell 1998 95 355 364 9814706
Soler C Daczewska M Da Ponte JP Dastugue B Jagla K Coordinated development of muscles and tendons of the Drosophila leg Development 2004 131 6041 6051 15537687
Taylor MV Sink H Comparison of muscle development in Drosophila and vertebrates Muscle development in Drosophila 2005 Georgetown (Texas) Eurekah
Pownall ME Gustafsson MK Emerson CP Myogenic regulatory factors and the specification of muscle progenitors in vertebrate embryos Annu Rev Cell Dev Biol 2002 18 747 783 12142270
Marics I Padilla F Guillemot JF Scaal M Marcelle C FGFR4 signaling is a necessary step in limb muscle differentiation Development 2002 129 4559 4569 12223412
Brand AH Perrimon N Targeted gene expression as a means of altering cell fates and generating dominant phenotypes Development 1993 118 401 415 8223268
Roy S VijayRaghavan K Homeotic genes and the regulation of myoblast migration, fusion, and fibre-specific gene expression during adult myogenesis in Drosophila
Development 1997 124 3333 3341 9310328
Bate M Rushton E Myogenesis and muscle patterning in Drosophila
C R Acad Sci III 1993 316 1047 1061 8076205
|
16207075
|
PMC1197288
|
CC BY
|
2021-01-05 08:21:28
|
no
|
PLoS Biol. 2005 Oct 6; 3(10):e337
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030337
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 10.1371/journal.pbio.0030344SynopsisBioinformatics/Computational BiologyEvolutionGenetics/Genomics/Gene TherapyVertebratesClear Evidence for Two Rounds of Vertebrate Genome Duplication Synopsis10 2005 6 9 2005 6 9 2005 3 10 e344Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Two Rounds of Whole Genome Duplication in the Ancestral Vertebrate
==== Body
As one of the most important sources of novel gene functions, gene duplications play a major role in evolutionary change. Though a gene copy will generally become inactive after duplication, it can be saved—either by acquiring a new function or dividing aspects of the original gene's function—on its way to becoming ubiquitous, or “fixed,” within the population.
The notion of “evolution by gene duplication” was proposed in 1970 by Susumu Ohno, who argued that gene and whole genome duplication provided the raw material for evolutionary innovations such as subcellular compartments, fins, and jaws. Having “extra” copies of genes provides the opportunity for duplicate genes to escape the constraints of purifying selection, and allows the genes to diverge and acquire novel functions. Ohno also proposed that two rounds of whole genome duplication occurred at some point in early vertebrate evolution—a possibility that could explain the relatively large size and complexity of the vertebrate genome.
The evolutionary patterns of gene duplications were reconstructed by comparing the complete gene sets of a tunicate (sea squirt), fish, mouse, and human. The 4-fold pattern in their global physical organization provides unmistakable evidence of two distinct, ancient whole genome duplications
Investigators equipped with far more powerful genome-mining tools than were available to Ohno have long sought evidence of this hypothesis (known as the 2R hypothesis, for “two rounds” of whole genome duplication), but with conflicting results. The observation that some gene families have four members in vertebrates but just one in invertebrates (the 4:1 rule) appeared to support the 2R hypothesis, until it was discovered that less than 5% of homologous gene families (similar genes with shared ancestry) followed the rule. But even when gene families do follow the rule, their configuration could just as likely arise from two rounds of single gene duplication as from whole genome duplication. And because duplicate genes are far more likely to degrade than to assume new or shared functions, the signal of whole genome duplication disappears.
Recent studies have shown that the global pattern of the physical location of homologous genes provides evidence of ancient whole genome duplications in yeast and plants, even when most of the duplicates have degraded. Now Paramvir Dehal and Jeffrey Boore have taken this approach to test the 2R hypothesis, by comparing the recently completed genome sequence of the invertebrate sea squirt with the genomes of three vertebrates—pufferfish, mouse, and human. (Because the sea squirt is a close relative of vertebrates, its genome can help reconstruct a more accurate tree of the organisms' evolutionary relationships than a more distant relative like the fruitfly could.)
After generating gene clusters that each contained “all, and only, those genes that descended from a singe gene in their common ancestor,” the authors used a method to infer the evolutionary relationships of the genes in each cluster. They could then compare these gene trees to the known evolutionary relationships of the organisms to determine when each gene duplicated in relation to when the lineages diverged. From this analysis, Dehal and Boore identified over 3,500 gene duplications present in multiple vertebrates, indicating they had occurred at the base of the vertebrate tree, dating back some 450 million years. But did these early duplication events arise from some large-scale duplication event, or were they simply the result of a great number of smaller scale duplications?
To explore this question, the authors analyzed the relative positions of the resulting paralogs in the vertebrate genome with the highest-quality data—the human genome. When considering only this subset of 3,500-plus early vertebrate duplications, they found a global pattern of human genome segments with similar arrangements of paralogous genes and multiple chromosomes with long linear stretches of interdigitated sets of paralogous genes—evidence that the duplications occurred in large segments. Even stronger support for the 2R hypothesis comes from the observation that the colinear arrangement of these genes is predominantly in a 4-fold pattern; this repetitive pattern is seen across almost all the human chromosomes. It's unlikely, the authors argue, that any combination of smaller, independent duplication events could have generated the same pattern.
Now that strong evidence for Ohno's hypothesis exists, researchers can investigate both the mechanism of genome duplication events and their possible effects on vertebrate evolution. It seems likely that a whole genome duplication would provide combinatorial possibilities that could permit a greater leap in evolution than could single gene duplications, even if the single gene duplications affected the complete set of genes. Studies that examine the function of these paralogous genes can explore whether these large-scale genomic events helped drive organismal complexity and diversification within the vertebrate lineage. —Liza Gross
|
0
|
PMC1197289
|
CC BY
|
2021-01-05 08:21:26
|
no
|
PLoS Biol. 2005 Oct 6; 3(10):e344
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030344
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 10.1371/journal.pbio.0030352SynopsisNeuroscienceIn VitroEnSNAREd by the Sperm Acrosome Synopsis10 2005 6 9 2005 6 9 2005 3 10 e352Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Dynamics of SNARE Assembly and Disassembly during Sperm Acrosomal Exocytosis
==== Body
Organelles in cells communicate with each other via small membrane-bound vesicles that carry molecular cargo from one part of the cell to another. Just as a cargo ship must know where to dock in order to empty goods at the correct port, vesicles in a cell must dock and empty their contents in the appropriate part of the cell. Studying the regulation of vesicle transport and membrane fusion is a major area of research in cell biology. Though the rules governing vesicle transport and fusion in the sea of cellular organelles have been deciphered in bits and pieces from various cell systems, all the components required for vesicle fusion have not been characterized for any single cell type. In a new study, Gerardo De Blas, Carlos Roggero, Claudia Tomes, and Luis Mayorga have elucidated the molecular mechanics of membrane fusion by studying the single vesicle of the sperm, the acrosome.
Enzymes released from the acrosome facilitate contact between the sperm and egg membranes during fertilization by dissolving the sheath surrounding an egg. For this, the acrosome membrane must fuse with the outer sperm membrane—a process called the acrosome reaction. This reaction happens only once in the lifetime of a sperm. In other systems, proteins involved in membrane fusion must be recycled so they can be reused in the fusion of a subsequent vesicle. But the acrosome reaction is unidirectional and requires no recycling, making it easier to decipher the steps involved.
Nonstimulated human sperm (left) and sperm undergoing acrosomal exocytosis (right). Multiple fusion events between the cell membrane and the membrane of the acrosome (the sperm's vesicle) trigger release of acrosomal content and formation of membranebound vesicles, thanks to the action of SNARE complexes
In previous work, Mayorga's group had shown that an increase in the concentration of intracellular calcium activates a molecule called Rab3A, thus initiating the acrosome reaction. The reaction proceeds with the help of various proteins, including NSF, SNAREs, and synaptotagmin VI, as well as calcium release from within the acrosome. Synaptotagmin VI is a calcium-sensitive protein. SNAREs are highly specialized proteins that exist in complexes of three molecules wound together in a helix and are present on both the acrosome and plasma membranes. NSF unwinds these helices so that molecules on opposite membranes can interact. To determine whether the SNARE proteins are in a single or triplet configuration at any given time, the authors used bacterial neurotoxins that can degrade single SNAREs but have no effect against the triplets.
In this study, De Blas and colleagues combine fluorescent techniques, a light-sensitive calcium chelator (which depletes all the calcium in the acrosome), and chemicals that inhibit specific steps in the cascade, to decipher whether each reaction occurs before or after the release of calcium. The researchers show that Rab3A activates NSF, which goes on to untwine helical, neurotoxin-resistant SNARE complexes on the acrosomal and sperm membranes, allowing opposing SNAREs to interact. Once SNAREs on opposite membranes form neurotoxin-sensitive loose complexes, calcium is released from within the acrosome. One protein that is required after the release of calcium is synaptotagmin VI. This protein, the authors suggest, could be responsible for the tight zippering of SNAREs on opposite membranes, converting them into toxin-resistant complexes. As the helix between molecules on opposite membranes becomes tighter, the membranes get pulled closer to each other, enabling membrane fusion.
How calcium activates Rab3A or synaptotagmin VI and how these proteins carry out their roles at the molecular level remain to be elucidated. However, this study elegantly demonstrates the cascade of players involved in the acrosomal membrane fusion reaction, from start to finish, in a single cell—something that had not been shown before. —Supriya Kumar
|
0
|
PMC1197290
|
CC BY
|
2021-01-05 08:21:27
|
no
|
PLoS Biol. 2005 Oct 6; 3(10):e352
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030352
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 10.1371/journal.pbio.0030353SynopsisInfectious DiseasesMolecular Biology/Structural BiologyVirologyVirusesIn VitroCasting a Wide Net to Fight Coronaviruses Synopsis10 2005 6 9 2005 6 9 2005 3 10 e353Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Design of Wide-Spectrum Inhibitors Targeting Coronavirus Main Proteases
==== Body
Viewed under a microscope, the coronavirus appears almost beautiful, thanks to the halo-like crown formed by its surface proteins. (“Corona” means “crown” in Latin.) Aesthetics aside, this genus of viruses is responsible for a wide range of animal and human diseases, from the common cold to the deadly severe acute respiratory syndrome, familiarly known as SARS. Research efforts to design antiviral agents to combat coronaviruses intensified after SARS killed at least 800 people in 2003 and have focused mostly on just this virus. But Haitao Yang, Dawei Ma, Zihe Rao, and colleagues reasoned that it might prove more efficient to develop wide-spectrum drugs and vaccines that could work against all coronaviruses—significantly reducing the health and economic burden associated with the 25 species of coronavirus.
Scientists fear that vaccines may prove ineffectual against coronaviruses because the viruses, like HIV, change their protein sequences and structures so often that a vaccine targeting one strain would likely be ineffective against another. The success of such a vaccine strategy depends on finding a protein target that is present, or well conserved, among all the different coronaviruses. By combining structural and biochemical analyses, Yang et al. not only identified such a target in a conserved region of a viral enzyme but also designed compounds with antiviral activity against multiple coronaviruses.
A broad-spectrum inhibitor that can recognize the active site of a coronavirus enzyme called the main protease could lead to the discovery of a single agent against coronaviruses. The protease structure is shown here in ribbon-and-surface representation; inhibitor molecules are the yellow, blue, and red balls
Because coronavirus species show great diversity among their structural proteins—which include the glyocoproteins that form the halo—the authors turned to three enzymes as potential targets. But since structural data were available for only one of the enzymes, called the main protease (Mpro), the authors focused on Mpro. Having structural data in hand greatly accelerates drug development, and since humans and other animals have no proteins similar to Mpro, the likelihood of deleterious side effects is low.
Initial computer analysis showed that the Mpro primary protein sequences (the linear amino acid sequences) have only 38% sequence identity between coronavirus species in some cases. But because three-dimensional structures tend to be more conserved than amino-acid sequences, the authors chose representative viruses from each group of coronavirus to study and compare the structure of their Mpro. This protease normally binds to its target protein (called the substrate) via a specific region, called the substrate-binding site. Structural analysis determined that this site is well conserved among coronaviruses, and biochemical tests confirmed that it would make a promising target for antiviral agents.
To test this hypothesis, the researchers created a synthetic version of the substrate that normally binds to the protease's substrate-binding site—reasoning that if they could inhibit the substrate's access to the binding site by the mimic (known as suicide inhibitors), they should be able to block the protease's activity and maybe halt viral replication. By studying the structure of the protease–substrate/inhibitor complex, Yang et al. continually improved their synthetic inhibitor until it bound strongly to the protease. Using this initial inhibitor as a base, the authors designed a panel of inhibitors and identified compounds that rapidly blocked proteases from multiple coronaviruses and kept the coronaviruses from reproducing. The compounds caused no obvious damage in human cells in the experiments.
The substrate-binding site identified by the researchers is an especially attractive target for drug development because evolutionarily conserved regions do not undergo high mutation rates like the rest of the viral genome, allowing antiviral drugs to maintain their effectiveness. Support for this hypothesis comes from the finding that a compound developed in this study also inhibits Mpro from new coronavirus strains that cause conjunctivitis, bronchiolitis, and pneumonia. By identifying promising candidates for drugs capable of targeting the entire Coronavirus genus, Yang et al. have laid the foundation for containing everything from the common cold to the deadly SARS virus. Preclinical and clinical trials will show whether these compounds live up to their promise. —Supriya Kumar
|
0
|
PMC1197291
|
CC BY
|
2021-01-05 08:28:16
|
no
|
PLoS Biol. 2005 Oct 6; 3(10):e353
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030353
|
oa_comm
|
==== Front
PLoS BiolPLoS BiolpbioplosbiolPLoS Biology1544-91731545-7885Public Library of Science San Francisco, USA 10.1371/journal.pbio.0030358SynopsisDevelopmentDrosophilaInsectsArthropodsAnimalsA Signaling Pathway at the Heart of Muscle Development Synopsis10 2005 6 9 2005 6 9 2005 3 10 e358Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
Drosophila Heartless Acts with Heartbroken/Dof in Muscle Founder Differentiation
==== Body
Unlike virtually every other type of cell, muscle cells contain dozens or even hundreds of nuclei. These multinucleate cells, called myofibers, form by fusion of precursor cells, called myoblasts, with “founder cells.” In the fruit fly, embryonic founder cells are formed by a well-known signaling pathway, but the same mechanism is not used to form adult founder cells. In this issue, K. VijayRaghavan and colleagues identify several key molecules involved in adult founder cell formation, and show that the process occurs through a novel mechanism.
In the fly embryo, founder cells differentiate from myoblasts through the actions of a membrane-bound receptor called Notch, an important player in several “signaling cascades” that use environmental signals to trigger changes in gene expression. However, in previous work, the authors have shown that Notch does not play a role in establishing adult founder cells. Instead, several clues pointed to the Fibroblast growth factor (FGF) family of receptors, one of which, in the fly, is called Heartless. Among other locations, Heartless is found on the surface of adult myoblasts in the abdomen. Reducing its expression in these cells, the authors showed, reduced the number of founder cells, while elevating it increased them.
But since Heartless is found in all adult myoblasts, it could not be responsible by itself for converting a myoblast into a founder cell. Another protein, called Heartbroken, seemed like a good candidate, since it functions exclusively within the FGF pathway. The authors showed that while the gene for Heartbroken is initially expressed in all myoblasts, over time its expression becomes restricted to those cells that develop into founder cells. Furthermore, by artificially maintaining Heartbroken expression, the authors dramatically elevated the number of founders in developing muscle, strengthening the case that Heartbroken is a key promoter of founder cell development.
But since both Heartless and Heartbroken are initially present in early myoblasts, what prevents wholesale Heartless signaling and premature, widespread founder cell formation? The authors show that a third factor, called Sprouty, declines in expression as founders are specified, and is not detected after founders are established. Sprouty is known to be a negative regulator of FGF signaling. VijayRaghavan and colleagues suggest that Sprouty interferes with Heartless signaling in early myoblasts, preventing founder cell formation even in the presence of Heartbroken. The gradual decline in the level of Sprouty may then “release the brakes” on Heartless signaling. Not every myoblast becomes a founder cell at that point, though, because the level of Heartbroken has also declined. Exactly which cells will maintain sufficient Heartbroken to become founders, and how those cells are specified, remains to be worked out.
Drosophila adult muscle founder cells are selected by the Heartless signaling pathway. Overexpression of components of the Heartless pathway increases the number of founder cells (right) compared to normal expression (left)
Still more remains to be discovered about the development of muscle in the adult fly. The significance of this work lies in identifying the FGF pathway as a critical component in muscle cell development, which provides leads that can be used to fill in the missing members of the pathway. While the details of vertebrate muscle development differ, the FGF pathway is known to be involved there too, and this work may shed light on aspects of that process as well. —Richard Robinson
|
0
|
PMC1197292
|
CC BY
|
2021-01-05 08:21:28
|
no
|
PLoS Biol. 2005 Oct 6; 3(10):e358
|
utf-8
|
PLoS Biol
| 2,005 |
10.1371/journal.pbio.0030358
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612862010.1371/journal.pmed.0020251Research ArticleMolecular Biology/Structural BiologyRespiratory MedicineRespiratory MedicineInterstitial lung diseaseUp-Regulation and Profibrotic Role of Osteopontin in Human Idiopathic Pulmonary Fibrosis Osteopontin in IPFPardo Annie
1
Gibson Kevin
2
Cisneros José
3
Richards Thomas J
2
Yang Yinke
2
Becerril Carina
3
Yousem Samueal
4
Herrera Iliana
1
Ruiz Victor
3
Selman Moisés
3
Kaminski Naftali
2
*1Facultad de Ciencias, Universidad Nacional Autónoma de México, Mexico City, Mexico,2The Dorothy P. and Richard P. Simmons Center for Interstitial Lung Diseases, Pulmonary Allergy and Critical Care Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of America,3Instituto Nacional de Enfermedades Respiratorias, Mexico City, Mexico,4Department of Pathology, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of AmericaBarnes Peter J. Academic EditorNational Heart and Lung InstituteUnited Kingdom*To whom correspondence should be addressed. E-mail: [email protected]¤ Current address: Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, United States of America
Competing Interests: The authors have declared that no competing interests exist.
Author Contributions: AP, KG, SY, VR, MS, and NK designed the study. KG, JC, TJR, YY, CB, IH, VR, and NK performed experiments. AP, KG, JC, YY, SY, IH, VR, MS, and NK analyzed the data. AP, KG, MS, and NK enrolled patients. AP, KG, MS, and NK contributed to writing the paper.
9 2005 6 9 2005 2 9 e25124 1 2005 10 6 2005 Copyright: © 2005 Pardo 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.
Role of Osteopontin in Idiopathic Pulmonary Fibrosis
Background
Idiopathic pulmonary fibrosis (IPF) is a progressive and lethal disorder characterized by fibroproliferation and excessive accumulation of extracellular matrix in the lung.
Methods and Findings
Using oligonucleotide arrays, we identified osteopontin as one of the genes that significantly distinguishes IPF from normal lungs. Osteopontin was localized to alveolar epithelial cells in IPF lungs and was also significantly elevated in bronchoalveolar lavage from IPF patients. To study the fibrosis-relevant effects of osteopontin we stimulated primary human lung fibroblasts and alveolar epithelial cells (A549) with recombinant osteopontin. Osteopontin induced a significant increase of migration and proliferation in both fibroblasts and epithelial cells. Epithelial growth was inhibited by the pentapeptide Gly-Arg-Gly-Asp-Ser (GRGDS) and antibody to CD44, while fibroproliferation was inhibited by GRGDS and antibody to αvβ3 integrin. Fibroblast and epithelial cell migration were inhibited by GRGDS, anti-CD44, and anti-αvβ3. In fibroblasts, osteopontin up-regulated tissue inhibitor of metalloprotease-1 and type I collagen, and down-regulated matrix metalloprotease-1 (MMP-1) expression, while in A549 cells it caused up-regulation of MMP-7. In human IPF lungs, osteopontin colocalized with MMP-7 in alveolar epithelial cells, and application of weakest link statistical models to microarray data suggested a significant interaction between osteopontin and MMP-7.
Conclusions
Our results provide a potential mechanism by which osteopontin secreted from the alveolar epithelium may exert a profibrotic effect in IPF lungs and highlight osteopontin as a potential target for therapeutic intervention in this incurable disease.
Osteopontin may have a critical role in the pathogenesis of idiopathic pulmonary fibrosis, and be a target for therapeutic intervention in this disease.
==== Body
Introduction
Idiopathic pulmonary fibrosis (IPF) is a chronic fibrosing interstitial pneumonia of unknown etiology characterized by alveolar epithelial cell injury/activation, fibroblast proliferation, and exaggerated accumulation of extracellular matrix in the lung parenchyma [1,2]. The disease is usually progressive and largely unresponsive to corticosteroid and immunosuppressive therapy [2,3].
A distinctive morphological feature of IPF is the development of fibroblastic/myofibroblastic foci, represented by widely scattered, small aggregates of subepithelial mesenchymal cells immersed within a myxoid-appearing extracellular matrix [4]. These fibroblastic/myofibroblastic foci represent areas of active fibrogenesis and play a crucial role in the progressive fibrotic response [1,5]. These sites are characterized by inappropriate re-epithelialization and impaired extracellular matrix remodeling. Although significant advances have been made in the characterization of the clinical and morphological features of this disease, the molecular mechanisms that underlie IPF pathogenesis in humans are still largely unknown [1,4–7].
Osteopontin (also termed secreted phosphoprotein 1) is a phosphorylated acidic glycoprotein that contains an Arg-Gly-Asp motif that binds to the integrin family of adhesion molecules [8]. It functions as a cell adhesion and migration molecule that can bind to several ligands, including αvβ3 integrin, some CD44 isoforms, and fibronectin [9,10]. Osteopontin has been implicated in a number of physiological and pathological processes including bone resorption, malignant transformation, and metastasis [11,12]. It is also considered a key cytokine regulating inflammation, cellular immune response, and tissue repair, with a unique effect on T cell function [11,12].
Using oligonucleotide microarrays we previously demonstrated that osteopontin is highly up-regulated in bleomycin-induced lung fibrosis in mice, and we reported similar results in a preliminary report involving five IPF lungs and four control samples [13,14]. Interestingly, osteopontin seems to have a profibrotic effect in the development of bleomycin-induced pulmonary fibrosis [15]. After bleomycin instillation, osteopontin-null mice developed reduced lung fibrosis characterized by dilated distal air spaces with decreased active transforming growth factor-beta 1 (TGF-β1) and reduced type I collagen expression compared with wild-type controls [16]. The mechanisms by which osteopontin, a cytokine with primarily TH1 effects (i.e., antifibrotic) on T lymphocytes [11,17,18], may cause profibrotic effects are not fully understood.
In this study, we applied microarrays to analyze gene expression patterns in a larger cohort of IPF lungs (13 IPF samples and 11 controls), and we analyzed the direct effects of osteopontin on human lung fibroblasts, alveolar epithelial cell migration and proliferation, and matrix metalloprotease (MMP) gene expression in vitro.
Methods
Study Population
Patients from the National Institute of Respiratory Diseases, Mexico City, México (Protocol S1), and the University of Pittsburgh, Pittsburgh, Pennsylvania, United States (Protocol S2), were included in this study. The protocol was approved by both Institutions, and written informed consent was obtained where required. Samples for oligonucleotide microarray were obtained from the tissue bank of the Department of Pathology at the University of Pittsburgh. The use of archived tissue has been approved by the local Institutional review board. Diagnosis of IPF was supported by history, physical examination, pulmonary function studies, chest HRCT, and bronchoalveolar lavage (BAL) findings, and was corroborated by open lung biopsy. The morphologic diagnosis of IPF was based on typical microscopic findings consistent with usual interstitial pneumonia [4]. The patients fulfilled the criteria of the American Thoracic Society and European Respiratory Society [7]. BAL samples were obtained at first consult as part of the initial diagnostic work-up. None of the patients had been treated with corticosteroids or immunosuppressive drugs at the time of BAL. Demographic data, pulmonary function data, and BAL differential cell counts are provide in Table 1.
Table 1 Patient Characteristics
Levels of osteopontin in BAL fluids were evaluated in ten healthy individuals (two current smokers, two former smokers, and six that had never smoked). All had normal chest X-rays and spirometries. Likewise, histologically normal lung tissues obtained at necropsy from six nonsmoking adult individuals who had died of causes unrelated to lung diseases were utilized for immunohistochemistry. For oligonucleotide microarrays, control samples included normal histology lung samples resected from patients with lung cancer obtained from the Pittsburgh Tissue Bank (Pittsburgh, Pennsylvania, United States).
Oligonucleotide Microarrays
Surgical remnants of biopsies or lungs explanted from patients with IPF that underwent pulmonary transplant were the sources of 13 IPF samples. Lung samples resected from patients with lung cancer, obtained from the tissue bank of the Department of Pathology at the University of Pittsburgh, were the sources of 11 normal samples. None of these samples had been included in our previous study. Total RNA was extracted and used as a template to generate double-stranded cDNA and biotin-labeled cRNA, as recommended by the manufacturer of the arrays and previously described [11]. Fragmented cRNA was hybridized to Codelink Uniset I slides. After hybridization, arrays were washed and stained with streptavidin-AlexaFluor 647. The arrays were scanned using a Genepix 4000B microarray scanner. Images were analyzed using Codelink expression II analysis suite. They were visually inspected for defects and quality control parameters as recommended by the manufacturer. Data files were imported into a microarray database and linked with updated gene annotations using SOURCE (http://genome-www5.stanford.edu/cgi-bin/SMD/source/sourceSearch) and then median scaled. Based on our previous experience, all expression levels below 0.01 were brought to 0.01. Statistical analysis was performed using Scoregene gene expression package (http://www.cs.huji.ac.il/labs/compbio/scoregenes, and data visualization was performed using Genexpress (http://genexpress.stanford.edu) and Spotfire Decision Site 8.0 (Spotfire, Göteborg, Sweden). The complete set of gene array data has been deposited in the Gene Expression Omnibus database with GEO serial accession number GSE2052 (http://www.ncbi.nlm.nih.gov/geo) according to MIAME guidelines. The general approach to analysis was previously described by us [19]. Weakest link models [20,21] were fitted using the Weaklink package for the R statistical software system (http://www.r-project.org). The p-value was obtained from fitting a logistic regression model with a single independent variable that is the minimum of the percentiles for the expression levels for the two genes. A Bonferroni correction was applied for multiple testing.
Bronchoalveolar Lavage
BAL was performed through flexible fiberoptic bronchoscopy under local anesthesia. Briefly, 300 ml of normal saline was instilled in 50-ml aliquots, with an average recovery of 60%–70%. The recovered BAL fluid was centrifuged at 250 g for 10 min at 4 °C. The cell pellet was resuspended in 1 ml of PBS and an aliquot was used to evaluate the total number of cells. Other aliquots were fixed in carbowax, stained with hematoxylin and eosin, and used for differential cell count. Supernatants were kept at −70 °C until use.
ELISA
Quantification of osteopontin was performed in BAL fluid samples from 18 IPF patients and 10 healthy individual controls, by using a commercial sensitive and specific ELISA following the instructions of the manufacturer (Calbiochem, La Jolla, California, United States).
Immunohistochemistry
Tissue sections were deparaffinized, rehydrated, and then blocked with 3% H2O2 in methanol for 30 min, then antigen was retrieved with citrate buffer (10 mM, pH 6.0) for 5 min in a microwave. Rabbit polyclonal antibody to human osteopontin (2 ng/ml; Calbiochem) was applied and samples were incubated at 4 °C overnight. A secondary biotinylated anti-immunoglobulin followed by horseradish peroxidase-conjugated streptavidin (BioGenex, San Ramon, California, United States) was used according to manufacturer's instructions. AEC (BioGenex) in acetate buffer containing 0.05% H2O2 was used as substrate [15,19]. The sections were counterstained with hematoxylin. The primary antibody was replaced by non-immune serum for negative control slides.
Two-Color Immunofluorescence Analysis
A standard two-stage double-immunofluorescence labeling technique was used. Briefly, frozen sections were washed in PBS (0.01 M [pH 7.4]) for 5 min and then fixed in cold acetone for 10 min, twice. Tissues were incubated in blocking buffer (1% BSA, 5% normal serum, 0.05% NP-40, in PBS) for 30 min. The slides were then incubated with a rabbit antibody to osteopontin (1:100; Abcam, Cambridge, Massachusetts, United States) for 1 h at room temperature and washed in PBST (0.05% Tween-20 in 0.01 M PBS [pH 7.4]) for 10 min, three times. A mouse anti-MMP-7 monoclonal antibody (1:1,000; Chemicon International, Temecula, California, United States) was added and the slides were incubated for an additional 1 h at room temperature. After three 10-min washes in PBST, slides were incubated with secondary antibodies (sheep anti-rabbit IgG-Cy3, 1:1,000; and goat anti-mouse IgG-FITC, 1:1,000; Sigma-Aldrich, St. Louis, Missouri, United States) for 30 min. Slides were then washed in the same buffer and mounted with antifade medium (containing DAPI to stain cell nuclei).
Cell Culture
Primary human normal lung fibroblasts were obtained as previously described [22], and the A549 cell line was obtained from
ATCC (Rockville, Maryland, United States).
Growth Rate Assay
Lung fibroblasts or A549 cells were seeded in 96-well culture plates at a cell density of 7.5 × 103 and 5 × 103 cells/well respectively, and incubated in Ham's F-12 and DMEM media, respectively (GIBCO-BRL, Grand Island, New York, United States), supplemented with 10% FBS at 37 °C in 5% CO2 and 95% air. After 12 h, the medium was replaced by medium with 0.1% FBS alone or 0.1% FBS plus increasing concentrations of osteopontin (0.4, 1, and 2 μg/ml) and the cells were maintained in culture for another 48 h. Cell growth was determined using the cell proliferation reagent WST-1 (Boehringer Mannheim, Mannheim, Germany) as previously described [19]. All assays were performed in triplicate. In parallel experiments, fibroblasts and A549 cells were pretreated for 25 min with antibody to human αvβ3 integrin (10 μg/ml; Chemicon), the pentapeptide Gly-Arg-Gly-Asp-Ser (GRGDS; Calbiochem), or antibody to human CD44 (NeoMarkers, Fremont, California, United States), and then osteopontin was added (2 μg/ml). Additionally, A549 cells were pretreated with antibody to human epidermal growth factor receptor (EGFR; 10 μg/ml; Chemicon).
Cell Migration Assay
Migration of fibroblasts and A549 cells was assayed using commercially available 24-well collagen-coated Boyden chambers (Chemicon) with an 8-μm pore size. Briefly, a semi confluent (∼80%) monolayer of lung fibroblasts or A549 cells was harvested with trypsin-EDTA, centrifuged, and resuspended in Ham's F-12 medium containing 5% BSA. The cell suspensions (3 × 105 cells/well) were added to the upper chamber. The lower chamber contained 0.3 ml of medium with 5% BSA alone or with 10 μg/ml of recombinant human osteopontin (Calbiochem). PDGF (8 ng/ml) and epidermal growth factor (EGF; 50 ng/ml) were used as positive controls for fibroblasts and for A549 cells, respectively. Additional BSA-coated chambers were used as blanks for each sample. After incubation for 8 h at 37 °C in a humidified incubator with 5% CO2 and 95% air, the nonmigrating cells on the top of Boyden chamber were scraped and washed. The migrating cells were quantitated according to manufacturer's instructions. Briefly, the cells were stained and the color eluted with 300 μl of extraction buffer, and 150-μl aliquots were measured in an ELISA plate reader at 545 nm. All assays were performed in duplicate. In parallel experiments, A549 cells and fibroblasts were pretreated as described above for growth rate.
RNA Isolation and Northern Blot Analysis
Total RNA was extracted from lung fibroblasts and A549 epithelial cells using the RNeasy Mini Kit (Qiagen GmbH, Germany). Cells were lysed and homogenized in the presence of a highly denaturing guanidine isothiocyanate-containing buffer. The samples were then applied to an RNeasy minicolumn, and RNA was eluted in 30 μl of water. Total RNA (20 μg/lane) was fractionated on a 1% agarose gel containing 0.66 M formaldehyde [21]. Ribosomal RNA was visualized with ethidium bromide, and the fractionated RNA was transferred onto a Nytran transfer membrane (NEN Life Science Products, Boston, Massachusetts, United States) by capillary blotting overnight. RNA was immobilized by baking at 80 °C for 2 h, and then prehybridized at 42 °C for 18 h in 5× SSC, 50% formamide, 5× Denhardt's solution, and 0.5% SDS, containing 100 μg/ml of denatured salmon sperm DNA. Hybridization was carried out at 42 °C for 18 h in hybridization buffer composed of formamide, 0.5% SDS, and heat-denatured 32P-labeled cDNA probes. After washing, membranes were dried and exposed to Kodak BIOMAX MS film at −70 °C with an intensifying screen. The cDNA clones for human MMP-1, TIMP-1, α1 type I collagen, and RNA ribosomal 18S were obtained from
ATCC. Human MMP-7 cDNA was a kind gift of Lynn Matrisian, Vanderbilt University, Nashville, Tennessee, United States. The cDNA probes were radiolabeled with 32P-dCTP to a specific activity of 200 × 106 dpm/μg using a Random Primer Labeling Kit (Stratagene, La Jolla, California, United States). All experiments were repeated twice.
Quantitative Real-Time RT-PCR
1 μg of RNA was treated with 1 unit of DNAase (Life Technologies, Grand Island, New York, United States). First-strand cDNA was synthesized by reverse transcription with random primers and Moloney-murine leukemia virus reverse transcriptase according to manufacturer's protocol (Advantage RT-for-PCR Kit; Clontech, Palo Alto, California, United States).
Real-time PCR amplification was performed using i-Cycler iQ Detection System (BioRad, Hercules, California, United States), using TAQMAN probes (PE Applied Biosystems, Wellesley, California, United States) labeled with FAM and TET. PCR was performed with the cDNA working mixture in a 25-μl reaction volume containing 3 μl of cDNA, 20 mM Tris-HCl (pH 8.3), 50 mM KCl, 2 mM MgCl2, 200 μM dNTPs, 0.2 μM specific 5′ and 3′ primers for 18S rRNA, 0.6 μM specific 5′ and 3′ primers for gene target, 0.2 μM of each probe TAQMAN (18S rRNA and gene target), and 1.25 units of AmpliTaq GOLD DNA polymerase (PE Applied Biosystems). A dynamical range was built with each product of PCR on copy number serial dilutions from 1 × 108 to 1 × 101; all PCRs were performed in triplicate. Standard curves were calculated referring the threshold cycle (Ct) to the log of each cDNA dilution step. Results were expressed as the number of copies of the target gene normalized to 18S rRNA. Some primers used in PCR reactions were designed using Beacon Designer software 2.1 (BioRad) and checked for homology in BLAST. The cycling conditions for PCR amplification were performed using the following protocol: Initial activation of AmpliTaq Gold DNA polymerase at 95 °C for 7 min; and 40 cycles of denaturation at 95 °C for 30s, annealing at 60 °C for 30 s, and extension at 72 °C for 30 s. The sequences of the PCR primer pairs and probes for each gene are shown in Table 2. All PCR experiments were done in duplicate.
Table 2 Primers and Probes for Real-Time PCR
PAGE Zymography
For MMP-7 analysis, conditioned media was electrophoresed in 12.5% SDS gels containing as substrate bovine CM-transferrin (0.3 mg/ml) and heparin [22]. After electrophoresis, gels were washed in a solution of 2.5% Triton X-100, and incubated overnight at 37 °C in 100 mM glycine (pH 7.6), plus 10 mM CaCl2 and 50 nM ZnCl2. Identical gels were incubated but in the presence of 20 mM EDTA. Gels were stained with Coomassie Brilliant Blue R250 and destained in a solution of 7.5% acetic acid and 5% methanol. Recombinant human MMP-7 (Chemicon) was used as positive control.
Western Blot Analysis
Serum-free conditioned media was centrifuged at 300 g at 4 °C for 30 min to remove cell debris, then concentrated by lyophilization. Samples were solubilized in water, and aliquots containing 5 μg of protein were mixed with Laemmli sample buffer and electrophoresed on 10% or 12.5% SDS-polyacrylamide gels. Proteins were transferred to nitrocellulose filters at 15 V for 20 min using semi-dry transfer cell (BioRad). Nonspecific sites were blocked overnight with 4% (w/v) nonfat dried milk in PBS, and membranes were incubated with rabbit antibody to human interstitial collagenase IgG (1:250 in PBS with 1% BSA; Santa Cruz Biotechnology, California, United States), antibody to human TIMP-1 (Chemicon), antibody to human MMP-7 (Oncogene, San Diego, California, United States), or monoclonal antibody to human α-smooth muscle actin (2 μg/ml; Cymbus Biotechnology, Chandlers Ford, Hants, United Kingdom) for 2 h at room temperature. Filters were incubated with secondary antibody conjugated to peroxidase for 1 h at room temperature. Finally, the filters were developed with the Enhanced Chemiluminescence detection system (Amersham Pharmacia Biotech, Piscataway, New Jersey, United States) using radiograph film (Hyperfilm, Amersham Pharmacia Biotech) according to the instructions of the manufacturer. Western blots were repeated twice.
Results
Osteopontin Gene Expression Is Highly Up-Regulated in IPF Lungs
The complete microarray dataset is available at the Gene Expression Omnibus database with GEO serial accession number GSE2052 (http://www.ncbi.nlm.nih.gov/geo). Gene expression patterns clearly distinguished IPF lungs from normal lungs. Osteopontin was the most up-regulated gene among the genes that distinguished IPF lungs (Figure 1A). Osteopontin levels were increased more than 20-fold (mean expression 0.62 in control lungs, 16.8 in IPF lungs; Figure 1B). This increase was statistically significant whether we applied threshold number of misclassifications (TNoM) score (p = 0.000221), or Student's t-test (p = 0.0000035). We corrected for multiple testing by controlling the false discovery rate at the 5% level [23]. Osteopontin expression levels were higher in most IPF lungs than in any control lung (Figure 1B).
Figure 1 Osteopontin Expression Levels by Microarray Analysis from Controls and IPF Lungs
Total RNA was used to generate double-stranded cDNA and biotin-labeled cRNA. Fragmented cRNA was hybridized to Codelink Uniset I slides and stained and scanned as described in Materials and Methods.
(A) A log scale scatter plot of the average of intensity of all the genes on the arrays in controls (x-axis) and IPF (y-axis). Colored points indicate 178 genes that were significantly changed (p < 0.01 in TNoM and Student's t-test). Points are colored by their fold ratios; progressive shades of blue indicate increase, and progressive shades of red indicate decrease. Points colored in gray did not reach significance. Oblique solid line indicates the line of equality. Two green dashed lines are lines that depict 10-fold change.
(B) Osteopontin levels in individual samples are shown. The y-axis is expression level in arbitrary fluorescence levels (log scale). The blue quadrangles are osteopontin levels in individual samples. Heat map shows sample osteopontin levels normalized to the geometric mean of osteopontin in controls and log base 2-transformed.
Osteopontin Is Increased in BAL Fluid from IPF Patients
Osteopontin protein was quantified in BAL fluids from 18 IPF patients and 10 healthy controls. As shown in Figure 2, ELISA measurement revealed a significant increase in the immunoreactive protein in the fluids derived from IPF lungs (148.8 ± 83 ng/ml in IPF lungs versus 3.8 ± 2.0 ng/ml in healthy controls, p < 0.01).
Figure 2 Osteopontin Levels in BAL Fluid
Quantification of osteopontin by ELISA was performed in BAL fluid samples from 18 IPF patients and 10 healthy individual controls. An increased concentration of soluble osteopontin was found in the BAL fluid obtained from IPF patients compared with healthy controls. The data represent the mean ± standard deviation (SD). * p < 0.01.
Immunoreactive Osteopontin Is Localized Primarily in Epithelial Cells
To examine the cellular source of osteopontin we analyzed IPF and control lungs by immunohistochemistry. As illustrated in Figure 3A–3C, osteopontin was localized primarily in alveolar epithelial cells that exhibited an intense cytoplasmic staining in IPF lungs. Occasionally, clusters of alveolar macrophages were also positive (Figure 3C). Immunohistochemical staining for osteopontin was negative in normal lungs as well as in lung tissue samples incubated with nonimmune sera (Figure 3D).
Figure 3 Localization of Osteopontin in IPF Lungs
Immunoreactive protein was revealed with AEC, and samples were counterstained with hematoxylin. Two representative IPF lung samples exhibited strong epithelial staining (original magnification, 40×) (A,B). Control lung showed no staining (C). Negative control section from IPF lung in which the primary antibody was replaced with nonimmune serum also showed no staining (40×) (D).
Osteopontin Induces Fibroblast and Epithelial Cell Growth Rate
To determine the effect of osteopontin on the growth rates of fibroblasts and epithelial cells, cells were stimulated with increasing concentrations of osteopontin, and the cell number was determined after 48 h using the cell proliferation reagent WST-1. Significant dose-dependent increases in cell proliferation were observed with 1 μg/ml and 2 μg/ml. Two different fibroblast lines reached 220% and 380% over controls (p < 0.01), while in two different experiments A549 cell lines exhibited 60% and 80% increases of growth rate over control (p < 0.01).
Osteopontin-induced fibroblast proliferation (2 μg/ml) was significantly suppressed by GRGDS-pentapeptide, which interrupts binding of RGD-containing proteins to cell surface integrins (p < 0.01) and by antibody to αvβ3 integrin (p < 0.05), suggesting that the effect of osteopontin on growth rate was mediated by the interaction of the GRGDS domain of osteopontin with αvβ3 integrin (Figure 4A). In contrast, epithelial cell growth was partially inhibited by antibody to CD44 (p < 0.05) (Figure 4B). Epithelial cell growth was also significantly suppressed by GRGDS but not by antibodies to αvβ3 (Figure 4B).
Figure 4 Effect of Osteopontin on Fibroblasts and Epithelial Cell Proliferation
Human normal lung fibroblasts (A) and A549 epithelial cells (B) were grown in Ham's F-12 medium with 0.1% FBS and stimulated with 2 μg/ml osteopontin. In parallel, osteopontin-stimulated cells were treated with anti-αvβ3, anti-CD44, and GRGDS. Each bar represents the mean ± SD of three experiments performed in triplicate; *p < 0.05; **p < 0.01.
OPN, osteopontin
Osteopontin Induces Fibroblast and Epithelial Cell Migration
To examine the effect of osteopontin on cell migration, we used collagen-coated Boyden chambers, a well-established in vitro assay system. The number of cells that migrated in absence of osteopontin was used as control (0% migration). As revealed in Figure 5A, fibroblasts significantly moved toward osteopontin compared with cells exposed to medium plus 5% BSA alone in the lower chamber. Osteopontin (10 μg) enhanced fibroblast migration by 120% ± 11% (p < 0.01), an enhancement similar to that obtained with the potent fibroblast mitogen platelet-derived growth factor, used as a positive control (157% ± 12%, unpublished data). To analyze possible mechanisms by which osteopontin stimulates migration, different blockers were used. Fibroblast migration was significantly reduced by GRGDS and antibody to αvβ3 integrin (p < 0.01), and by antibody to CD44 (p < 0.05). The inhibition of cell migration was specific, since incubation with IgG had no effect (unpublished data).
Figure 5 Effect of Osteopontin on Fibroblasts and Epithelial Cell Migration
Fibroblasts (A) and A549 epithelial cells (B) were placed in the upper compartment of a Boyden-type chamber, and Ham's F-12 medium containing 5% BSA alone or with 10 μg/ml of osteopontin was added to the lower compartment. After 8 h of incubation, the migrating cells were stained, and the absorbance of the stained solution was measured by ELISA. In parallel experiments, osteopontin-stimulated cells were treated with anti-αvβ3, anti-CD44, and GRGDS. Each bar represents the mean ± SD of three experiments; *p < 0.01; **p < 0.05.
OPN, osteopontin
A549 lung cells also showed a significant increase in cell migration in response to osteopontin (Figure 5B). After 8 h of incubation, A549 cells passing through the membrane increased by 114% ± 25% compared to control cells (p < 0.01). Epidermal growth factor, used as a positive control, induced a 168% ± 14% increase in cell migration (unpublished data). Osteopontin-induced migration was abolished when the epithelial cells were pretreated individually with GRGDS, anti-αvβ3 integrin, or anti-CD44 (p < 0.01; Figure 5B).
Osteopontin Induces an Environment That Favors Extracellular Matrix Deposition in Fibroblasts
MMP-1 expression by fibroblasts
Under basal conditions, some primary human lung fibroblast cell lines may express MMP-1, while others express the enzyme only when stimulated, for example, by aminophenylmercuric acetate (APMA). In this context, the effect of osteopontin on MMP-1 expression was examined by Northern blot analysis in a cell line producing MMP-1 under basal conditions, and in a cell line that did not produce MMP-1 but was stimulated by APMA (Figure 6A and 6B). Osteopontin down-regulated both the basal and the APMA-induced MMP-1 transcript level. When the signal of MMP-1 mRNA was normalized to the level of 18S RNA and quantified by densitometry, a reduction of ∼50% was noticed. This result was confirmed by real-time PCR that showed ∼40% inhibition (p < 0.05; Figure 6C). Inhibition of MMP-1 expression was partially blocked by antibody to CD44, while it was not affected by antibody to αvβ3 integrin. The inhibitory effect of osteopontin on MMP-1 was also observed at the protein level by Western blot analysis. As shown in Figure 6D, the level of immunoreactive MMP-1 present in the conditioned medium was decreased in the fibroblasts stimulated with osteopontin as compared with control cells.
Figure 6 Effects of Osteopontin on MMP-1 Gene and Protein Expression in Two Human Lung Fibroblast Cell Lines
(A) Representative Northern blot of 20 μg total cellular RNA per lane extracted from control cells and fibroblasts stimulated with 0.4 μg/ml and 1 μg/ml osteopontin. Both concentrations of osteopontin induced a down-regulation in the expression of MMP-1.
(B) Osteopontin also reduced overexpression of MMP-1 in APMA-stimulated cells.
(C) The expression level of MMP-1 by real-time PCR was determined as described in Materials and Methods and normalized to the level of 18S ribosomal RNA. In parallel experiments, osteopontin-stimulated cells were treated with anti-αvβ3 and anti-CD44. Bars represent mean ± SD (*p < 0.05).
(D) Representative Western blot demonstrating a decrease of immunoreactive MMP-1 in conditioned media from fibroblasts stimulated with osteopontin. Fibroblasts treated with APMA and FGF-1 plus heparin used as positive controls strongly induced MMP-1 expression.
C, control; FGF1, FGF-1 plus heparin; OPN, osteopontin; PMA, APMA-stimulated.
Osteopontin increases TIMP-1 expression by fibroblasts
The effect of osteopontin on TIMP-1 gene expression in fibroblasts is illustrated in Figure 7. Northern blot analysis (Figure 7A) revealed an increase in TIMP-1 gene expression at 0.4 μg/ml and 1 μg/ml osteopontin when compared to control. This result was confirmed by real-time PCR (Figure 7B). Osteopontin increased the TIMP-1/18S ribosomal RNA from 55.7 ± 14.2 copies to 121.2 ± 13.1 copies (p < 0.01). Both anti-CD44 and anti-αvβ3 integrin abolished this increase. Osteopontin also increased TIMP-1 protein expression in conditioned medium as illustrated in Figure 7C.
Figure 7 Effect of Osteopontin on TIMP-1 Gene and Protein Expression by Fibroblasts
(A) Northern blot of 20 μg total cellular RNA per lane extracted from control and fibroblasts stimulated with 0.4 μg/ml and 1 μg/ml osteopontin. Both concentrations of osteopontin induced an increase of TIMP-1 expression.
(B) The expression level of TIMP-1 by real-time PCR normalized to the level of 18S ribosomal RNA corroborates TIMP-1 up-regulation by osteopontin (*p < 0.01). In parallel experiments, osteopontin-stimulated cells were treated with anti-αvβ3 and anti-CD44.
(C) Western blot demonstrating an increase of immunoreactive TIMP-1 in conditioned media from fibroblasts stimulated with osteopontin.
C, control; OPN, osteopontin.
Osteopontin induces collagen gene expression in fibroblasts
The effect of osteopontin on collagen gene expression is depicted in Figure 8. Northern blot analysis (Figure 8A) revealed a 2-fold increase in α1 type I collagen gene expression. Expression of α-smooth muscle actin in fibroblasts was induced by TGF-β1 but not by osteopontin (Figure 8B).
Figure 8 Effects of Osteopontin on Collagen Gene Expression and Smooth Muscle Alpha Actin Protein in Human Lung Fibroblasts
(A) Northern blot of 20 μg total cellular RNA per lane extracted from control and fibroblasts stimulated with 0.4 μg/ml and 1 μg/ml osteopontin. Both concentrations of osteopontin induced an up-regulation in the expression of α1 type I collagen.
(B) Western blot showing no effect of osteopontin on immunoreactive α smooth muscle actin. Recombinant TGF-β1 was used as a positive control.
C, control; OPN, osteopontin.
Osteopontin Increases MMP-7 Expression by Epithelial Cells
Stimulation of A549 cells with osteopontin (0.4 μg/ml and 2.0 μg/ml) induced an up-regulation of MMP-7 gene expression as illustrated by Northern blot in Figure 9A. Real-time PCR revealed a 6-fold increase at 6 h after osteopontin stimulation, which was abolished by treatment with anti-αvβ3, anti-CD44, anti-EGFR, or GRGDS as shown in Figure 9B. Pro-MMP-7 overexpression was confirmed at the protein level by Western blot analysis of the epithelial cells conditioned medium (Figure 9C). As recombinant MMP-7 used as control showed a lower molecular weight band to that of conditioned medium, this one was treated with APMA to activate pro-MMP-7. Activated proenzyme showed a similar molecular weight band to the recombinant protein. Zymography using CM-transferrin as substrate showed that treatment of A549 epithelial cells with 0.4 μg/ml and 1 μg/ml osteopontin induced an increase of both the pro-MMP-7 and MMP-7 activity bands (Figure 9D).
Figure 9 Effect of Osteopontin on MMP-7 Gene and Protein Expression and Activity in A549 Epithelial Cells
(A) Northern blot of 20 μg total cellular RNA per lane extracted from control and A549 cells stimulated with 0.4, 1, and 2 μg/ml osteopontin.
(B) Densitometry of Northern blot and normalization of MMP-7 to 18S demonstrates a 3- to 4-fold increase in MMP-7 over control.
(C) Real-time PCR showing up-regulation of MMP-7 expression by osteopontin (*p < 0.01) and inhibition by anti-αvβ3, anti-CD44, anti-EGFR, and GRGDS.
(D) Western blot demonstrating an increase of immunoreactive MMP-7 in conditioned medium from A549 epithelial cells stimulated with osteopontin. Activation of pro-MMP-7 by APMA is shown in leftmost lane.
(E) Zymography of conditioned media in 12.5% SDS gels containing bovine CM-transferrin (0.3 mg/ml) and heparin as substrate.
C, control; OPN, osteopontin.
Osteopontin Colocalizes with MMP-7 in Alveolar Epithelial Cells from IPF Lungs
Since we have previously demonstrated that IPF lungs strongly express MMP-7 in alveolar epithelial cells [14], we evaluated whether the staining pattern of this enzyme is associated with that of osteopontin. Figure 10 illustrates confocal microscopy images showing that MMP-7 (Figure 10A) is partially overlapped with that of osteopontin (Figure 10B), giving substantial double labeling (Figure 10C and 10D). Normal lungs did not exhibit any significant staining for MMP-7 or osteopontin (unpublished data).
Figure 10 Osteopontin and MMP-7 Colocalization in IPF Lungs
(A−C) MMP-7 staining is shown in green (A), osteopontin is shown as red staining (B), and overlap of staining is shown in yellow (C), suggest colocalization of MMP-7 and osteopontin in alveolar epithelial cells in IPF lungs (60×).
(D) A lower-magnification image (20×) of the same region (A–C) with the same color coding. The white rectangle depicts area shown in (A−C).
Weakest Link Models Identify a Statistically Significant Interaction between Osteopontin and MMP-7
To determine whether MMP-7 and osteopontin expression levels jointly distinguish IPF and control samples, we applied weakest link statistical models [20]. Weakest link models assume that two genes jointly impact the probability that a randomly selected sample belongs to a certain phenotype (IPF in this case) only if expression levels of the two genes lie on a low-dimensional curve whose form is specified by the model and estimated using the observed quantiles of the data. Our analysis detected a statistically significant joint effect of MMP-7 and osteopontin on the IPF phenotype (p < 0.001). This joint relationship is illustrated in Figure 11 for both the gene expression values (Figure 11A) and the sample percentiles (Figure 11B). To assess the relevance of this association, we chose both substantively meaningful genes and a more general set of genes to evaluate the contribution of the weakest link model relating osteopontin and MMP-7 (Figure 11). Among 14 matrix metalloproteases and their inhibitors represented on the Human Uniset I chip, osteopontin and MMP-7 exhibited the most significant biological interaction. Osteopontin also had significant interactions (weakest link models with p < 0.05) with MMP-1, MMP-2, and MMP-11 but none was as statistically significant as MMP-7. We also ran weakest link models combining osteopontin with each of 400 genes that passed the false discovery rate of less than 0.05 for differential expression between IPF and control samples. The interaction of osteopontin with MMP-7 in a weakest link model was more significant than 371 of these 400 genes.
Figure 11 Weakest Link Models of Osteopontin and MMP-7
IPF samples are depicted in solid dots and controls in open dots; the y-axis is MMP-7 expression and x-axis is osteopontin expression. Black solid line is the curve of optimal use showing that the expression levels for MMP-7 and osteopontin jointly interact to determine the IPF phenotype. The probability contour plot is shown in terms of the observed expression data (scale is log base 2 for gene expression data) (A); and probability contour plot is shown in terms of percentiles of the data (scale is percentiles) (B).
Discussion
In the present study we focused on the profibrotic effects of osteopontin, a multifunctional cytokine that mediates diverse biological functions, including cell adhesion, chemotaxis, and signaling, as well as tissue reparative processes [8,11,12]. We demonstrated that osteopontin was the most up-regulated gene in lungs of IPF patients, and that it was mainly expressed by alveolar epithelial cells. To better understand the potential local profibrotic effects of osteopontin, we studied its effects on lung fibroblasts and epithelial cells. Functionally, osteopontin induced fibroblast and epithelial cell proliferation and migration. The effect on fibroblast migration and proliferation was dependent mainly on integrins, while in epithelial cells proliferation was mainly dependent on CD44 and migration was dependent on CD44 and integrin signaling. Osteopontin exhibited profibrotic effects on molecules involved in extracellular matrix remodeling. Thus, in fibroblasts, osteopontin increased TIMP-1 and type I collagen and inhibited MMP-1 expression, while in alveolar epithelial cells it induced MMP-7. The effects on TIMP-1 and MMP-1 expression appeared to be mostly dependent on CD44, while the effect on MMP-7 expression was dependent on CD44 and integrin signaling. Interestingly, osteopontin was colocalized with MMP-7 in alveolar epithelial cells of IPF lungs, and application of the weakest link models to microarray data suggested that the genes of both interacted to affect the IPF phenotype. Our results provide a potential mechanism by which osteopontin secreted from epithelial cells exerts its profibrotic effects through direct signaling on fibroblasts and epithelial cells.
Several studies in experimental tissue fibrosis have suggested a possible profibrotic role of osteopontin. In kidney fibrosis, osteopontin enhances macrophage recruitment and stimulates the development of renal scarring after an acute ischemic insult [24]. Also, osteopontin expression is increased in the myocardium after myocardial infarction, and the lack of this mediator is associated with decreased collagen accumulation [25]. Similarly, osteopontin appears to be an important mediator of the cardiac profibrotic effects of angiotensin II by promoting collagen synthesis and remodeling in the interstitial myocardium [26]. In experimental lung fibrosis it has been suggested that osteopontin produced by alveolar macrophages functions as a fibrogenic cytokine [15,16]. In granulomatous lung diseases, osteopontin is also up-regulated and its main sources are macrophages and T lymphocytes [27,28]. We observed that in human IPF lungs, hyperplastic alveolar epithelial cells seemed to be a source of osteopontin, which is consistent with the findings of Berman et al. [15,16]. These observations support the key role of alveolar epithelial cells as regulators of the lung profibrotic microenvironment in IPF, thus emphasizing the critical difference between the mechanisms of fibrosis in IPF compared with animal models or with lung fibrosis associated with inflammatory disorders.
Positive feedback mechanisms have been previously proposed for osteopontin and MMP-2 [29], where osteopontin induces MMP-2 expression [30] and is cleaved and activated by MMP-2 [29]. The same mechanism has been proposed for osteopontin and MMP-3, where osteopontin binds and activates MMP-3, which in turn can cleave and activate it [31]. Similarly, osteopontin is cleaved and activated by MMP-7 [31], and we observe that MMP-7 is induced by osteopontin in epithelial cells, suggesting that this positive feedback mechanism is also applicable to osteopontin and MMP-7. This is also supported by the colocalization of MMP-7 and osteopontin in IPF epithelial cells, and by the computational relationship of the expression levels of osteopontin and MMP-7. Interestingly, MMP-7 and osteopontin are β-catenin target genes [32,33]. Recently, Chilosi et al. reported impressive activation of WNT/β-catenin in IPF lungs [34]. They have mainly observed β-catenin nuclear localization in proliferative bronchiolar lesions, where it colocalized with MMP-7 [34]. Collectively, these results could indicate a mechanism by which osteopontin and MMP-7 are induced by aberrant activation of the WNT/β-catenin pathway, and each affects the function and expression of the other gene, thus representing a local positive feedback mechanism that facilitates a chronic, relentless lung disease.
Although intriguing, this proposed positive, self-perpetuating loop in itself cannot explain increased collagen deposition, the critical hallmark of fibrosis. In this context, our results suggest that osteopontin affects the critical balance between MMPs and their inhibitors through its cell-specific effects. In agreement with the inhibition of interleukin 1β-stimulated increases in MMP-2 and MMP-9 observed by Xie et al. [35], we observed that in human lung fibroblasts, osteopontin caused a significant reduction of baseline as well as APMA-stimulated MMP-1, an MMP responsible for fibrillar collagen degradation. Additionally, we observed a concomitant increase in TIMP-1, the main inhibitor of MMP-1 (as well as MMP-2 and −9), and in type I collagen gene expression, suggesting that osteopontin may facilitate a profibrotic environment not only by its effect on epithelial cells but also by inducing a nondegradative microenvironment similar to the one observed in IPF and experimental lung fibrosis [36,37]. Interestingly, osteopontin did not induce α-smooth muscle actin in lung fibroblasts in vitro, suggesting that although it had a role in facilitating the profibrotic environment in IPF, it had weaker role in the formation of myofibroblast foci. However, further experiments will be needed to determine this point.
Migration and proliferation of fibroblasts are essential for the expansion of their populations and the formation of the fibroblastic foci that seem to represent the “leading edge” of the progressive fibrotic process [5]. The observation that osteopontin induced both migration and proliferation of primary human lung fibroblasts suggests that osteopontin may be involved in this process and is supported by previous observations in murine and human fibroblast cell lines [15]. Osteopontin promotes fibroblast collagen gel contraction and rat cardiac fibroblast proliferation, and it has been suggested that it is an important mediator of angiotensin II regulation of fibroblast behavior in the cardiac remodeling process [38]. It also induced proliferation of vessel smooth muscle cells, suggesting that it is involved in vascular remodeling during the development of atherosclerosis [39,40]. These results highlight the sufficiency of a cytokine secreted from alveolar epithelial cell to induce many of the phenotypic changes associated with lung fibrosis.
The mechanisms by which osteopontin influences epithelial and fibroblast cells are not fully understood. In general it has been proposed that osteopontin affects cells by binding to CD44 isoforms, certain integrins, and EGFR [41,42]. It is unknown whether the expression of these receptors is changed in IPF; CD44 is expressed on lung fibroblasts and epithelial cells [43], is induced in radiation- and bleomycin-induced pulmonary fibrosis [44] and in acute alveolar fibrosis [45], and is critical for resolution of noninfectious lung inflammation [46]. The integrin receptors for osteopontin are widely expressed in lung epithelial cells and fibroblasts [47], but a change in their repertoire has not been reported yet with IPF. In this paper we did not seek to fully dissect these mechanisms; however, we explored some of the possible receptors that may be important to osteopontin effect on fibroblasts and epithelial cells. Our results support some of the previous observations regarding the importance of integrin-mediated signaling in fibroblast migration. Interestingly, we observed a differential effect in epithelial and fibroblasts cells. Inhibition of CD44 significantly reduced the effects of osteopontin on cell proliferation in epithelial cells, while αvβ3 inhibition seemed to affect mostly fibroblasts. This integrin is a receptor for a wide variety of extracellular matrix ligands with an exposed RGD sequence, including vitronectin, fibronectin, fibrinogen, thrombospondin, proteolyzed collagen, von Willebrand factor, as well as osteopontin, and appears to play a critical role in cell migration [48]. The hyaluronic acid receptor CD44 is also a receptor for osteopontin, and it has been implicated in chemotaxis mediated by this mediator [10,49,50]. In our experiments, the pan-integrin antagonist GRGDS-pentapeptide added to the culture medium abolished the effects in fibroblasts and epithelial cells, suggesting that different integrins may be involved in both cell types, and that for some effects, both integrins and CD44 are required. Detailed studies of the effects of variant osteopontin isoforms as well as cells that express different receptors will be needed to elucidate these mechanisms.
Our study focused on human tissues and human cell lines because of the unique features of IPF that are not readily mimicked by any animal models. We insisted on using primary human lung fibroblasts in our experiments; therefore, we are confident that these results represent a mechanism that may actually occur in the human lung. Unfortunately, it is nearly impossible to work with primary alveolar epithelial cells, and we had to resort to the epithelial cell line A549. However, we present evidence obtained from human lungs that suggest that the mechanisms that we proposed in vitro do exist in the human IPF lung.
In summary, in this study we highlight the role of osteopontin in human IPF. Although previous studies have suggested that osteopontin has a potential profibrotic effect in animal models of lung fibrosis, its role in human IPF was unclear. We demonstrated that osteopontin is highly expressed in IPF lungs, and that it is primarily expressed by hyperplastic alveolar epithelial cells. We demonstrated that osteopontin affected fibroblast and epithelial cell proliferation and migration, and that it had fibrosis-relevant effects on MMP and TIMP expressions. Our results suggest a mechanism explaining most of the profibrotic effects of osteopontin by its direct effects on fibroblasts and epithelial cells in the lungs. Furthermore, our results suggest that in IPF the interaction between MMP-7 and osteopontin may be involved in the relentlessly progressive nature of the disease, and highlight osteopontin as a potential target for therapeutic intervention in this incurable disease.
Supporting Information
Protocol S1 Sample Mexico
(746 KB ZIP).
Click here for additional data file.
Protocol S2 Microarrays
(2.1 MB ZIP).
Click here for additional data file.
Accession Numbers
The GenBank (http://www.ncbi.nlm.nih.gov/) accession numbers of the genes discussed in this paper are R18S (X03205), MMP-1 (NM_002421), MMP-7 (XM_017384), and TIMP-1 (X03124).
Patient summary
Background
Idiopathic pulmonary fibrosis is a chronic progressive disease of the lung that leads to increasing amounts of scar tissue with subsequent destruction of the lung. There is no specific cure at present. Patients may be treated with corticosteroids or drugs to suppress their immune system, although these drugs are usually not effective. Some patients receive lung transplants.
Why Was This Study Done?
Previous work has suggested that a protein called osteopontin is increased in mice that have lung fibrosis and that mice that do not have the gene for osteopontin are protected from lung fibrosis. The researchers wanted to investigate if osteopontin was also involved in the human disease.
What Did the Researchers Do and Find?
They looked at samples taken from the lungs of people with idiopathic pulmonary fibrosis and other diseases and measured many genes that are expressed there. They found that osteopontin was increased in the lungs of people with idiopathic pulmonary fibrosis. They then looked at cultures of lung cells and found that osteopontin caused an increase in the number and movement of cells that are involved in lung fibrosis. Its presence also affected other proteins that seem to be involved in fibrosis.
What Do These Findings Mean?
Osteopontin may have a key role in the pathway that causes fibrosis to occur in the lungs of people with idiopathic pulmonary fibrosis. Further work will need to be done to confirm these results, but in the future drugs directed against osteopontin or one of the related proteins might be a possible treatment for the disease. Currently there are no such drugs. Additionally osteopontin may be useful in the diagnosis and early detection of the disease, but further studies are required.
Where Can I Get More Information Online?
Medline Plus has links to many pages with information on the disease:
http://www.nlm.nih.gov/medlineplus/pulmonaryfibrosis.html
The Coalition for Pulmonary Fibrosis is a nonprofit organization that has information for patients as well as physicians:
http://www.coalitionforpf.org/
The Dorothy P. and Richard P. Simmons Center for Interstitial Lung Diseases contains information about IPF for patients and their families:
http://simmonscenterild.upmc.com/
The authors wish to acknowledge Inna Loutaev and Lara Chensny for their technical help and Mary Williams for her help in administering the collaborative efforts that underlie this manuscript. The authors also thank Dr. A. Choi, Dr. J. Dauber, and Dr. N. Friedman for their critical review and insightful comments. This work was supported National Institutes of Health grant 1R01 HL073745–01 and by a generous donation from the Simmons family. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Citation: Pardo A, Gibson K, Cisneros J, Richards TJ, Yang Y, et al. (2005) Up-Regulation and profibrotic role of osteopontin in human idiopathic pulmonary fibrosis. PLoS Med 2(9): e251.
Abbreviations
APMAaminophenylmercuric acetate
BALbronchoalveolar lavage
EGFRepidermal growth factor (receptor)
IPFidiopathic pulmonary fibrosis
MMPmatrix metalloprotease
SDstandard deviation
TGF-β1transforming growth factor-beta 1
TIMPtissue inhibitor of metalloprotease
==== Refs
References
Selman M King TE Pardo A Idiopathic pulmonary fibrosis: Prevailing and evolving hypotheses about its pathogenesis and implications for therapy Ann Intern Med 2001 134 136 151 11177318
Gross TJ Hunninghake GW Idiopathic pulmonary fibrosis N Engl J Med 2001 345 517 525 11519507
Collard HR King TE Treatment of idiopathic pulmonary fibrosis: The rise and fall of corticosteroids Am J Med 2001 110 326 328 11239857
Katzenstein AL Myers JL Idiopathic pulmonary fibrosis: Clinical relevance of pathologic classification Am J Respir Crit Care Med 1998 157 1301 1315 9563754
Crystal RG Bitterman PB Mossman B Schwarz MI Sheppard D Future research directions in idiopathic pulmonary fibrosis: Summary of a National Heart, Lung, and Blood Institute working group Am J Respir Crit Care Med 2002 166 236 246 12119236
American Thoracic Society American Thoracic Society. Idiopathic pulmonary fibrosis: Diagnosis and treatment. International consensus statement. American Thoracic Society (ATS), and the European Respiratory Society (ERS) Am J Respir Crit Care Med 2000 161 646 664 10673212
ATS/ERS International Multidisciplinary Consensus Classification of the Idiopathic Interstitial Pneumonias ATS/ERS International Multidisciplinary Consensus Classification of the Idiopathic Interstitial Pneumonias. This joint statement of the ATS and the ERS was adopted by the ATS board of directors, June 2001 and by the ERS Executive Committee, June 2001 Am J Respir Crit Care Med 2002 165 277 304 11790668
O'Regan AW Nau GJ Chupp GL Berman JS Osteopontin (Eta-1) in cell-mediated immunity: Teaching an old dog new tricks Immunol Today 2000 21 475 478 11071524
Reinholt FP Hultenby K Oldberg A Heinegard D Osteopontin—A possible anchor of osteoclasts to bone Proc Natl Acad Sci U S A 1990 87 4473 4475 1693772
Weber GF Ashkar S Glimcher MJ Cantor H Receptor-ligand interaction between CD44 and osteopontin (Eta-1) Science 1996 271 509 512 8560266
Denhardt DT Noda M O'Regan AW Pavlin D Berman JS Osteopontin as a means to cope with environmental insults: Regulation of inflammation, tissue remodeling, and cell survival J Clin Invest 2001 107 1055 1061 11342566
O'Regan A Berman JS Osteopontin: A key cytokine in cell-mediated and granulomatous inflammation Int J Exp Pathol 2000 81 373 390 11298186
Kaminski N Allard JD Pittet JF Zuo F Griffiths MJ Global analysis of gene expression in pulmonary fibrosis reveals distinct programs regulating lung inflammation and fibrosis Proc Natl Acad Sci U S A 2000 97 1778 1783 10677534
Zuo F Kaminski N Eugui E Allard J Yakhini Z Gene expression analysis reveals matrilysin as a key regulator of pulmonary fibrosis in mice and humans Proc Natl Acad Sci U S A 2002 99 6292 6297 11983918
Takahashi F Takahashi K Okazaki T Maeda K Ienaga H Role of osteopontin in the pathogenesis of bleomycin-induced pulmonary fibrosis Am J Respir Cell Mol Biol 2001 24 264 271 11245625
Berman JS Serlin D Li X Whitley G Hayes J Altered bleomycin-induced lung fibrosis in osteopontin-deficient mice Am J Physiol Lung Cell Mol Physiol 2004 286 L1311 1318 14977630
Koguchi Y Kawakami K Uezu K Fukushima K Kon S High plasma osteopontin level and its relationship with interleukin-12-mediated type 1 T helper cell response in tuberculosis Am J Respir Crit Care Med 2003 1355 1359 12574077
Ashkar S Weber GF Panoutsakopoulou V Sanchirico ME Jansson M Eta-1 (osteopontin): An early component of type-1 (cell-mediated) immunity Science 2000 287 860 864 10657301
Kaminski N Friedman N Practical approaches to analyzing results of microarray experiments Am J Respir Cell Mol Biol 2002 27 125 132 12151303
Reichert TE Scheuer C Day R Wagner W Whiteside TL The number of intratumoral dendritic cells and zeta-chain expression in T cells as prognostic and survival biomarkers in patients with oral carcinoma Cancer 2001 91 2136 2147 11391595
Richards TJ Chensny LJ Loutaev I Fernando HC Gibson KF Weakest link models for microarray data in idiopathic pulmonary fibrosis Am J Resp Crit Care Med 2004 169 A776
Ramos C Montano M Garcia-Alvarez J Ruiz V Uhal BD Fibroblasts from idiopathic pulmonary fibrosis and normal lungs differ in growth rate, apoptosis, and tissue inhibitor of metalloproteinases expression Am J Respir Cell Mol Biol 2001 24 591 598 11350829
Benjamini Y Hochberg Y Controlling the false discovery rate: A practical and powerful approach to multiple testing J R Stat Soc 1995 289 300
Persy VP Verhulst A Ysebaert DK De Greef KE De Broe ME Reduced postischemic macrophage infiltration and interstitial fibrosis in osteopontin knockout mice Kidney Int 2003 63 543 553 12631119
Trueblood NA Xie Z Communal C Sam F Ngoy S Exaggerated left ventricular dilation and reduced collagen deposition after myocardial infarction in mice lacking osteopontin Circ Res 2001 88 1080 1087 11375279
Matsui Y Jia N Okamoto H Kon S Onozuka H Role of osteopontin in cardiac fibrosis and remodeling in angiotensin II-induced cardiac hypertrophy Hypertension 2004 43 1195 1201 15123578
O'Regan AW Chupp GL Lowry JA Goetschkes M Mulligan N Osteopontin is associated with T cells in sarcoid granulomas and has T cell adhesive and cytokine-like properties in vitro J Immunol 1999 162 1024 1031 9916729
Carlson I Tognazzi K Manseau EJ Dvorak HF Brown LF Osteopontin is strongly expressed by histiocytes in granulomas of diverse etiology Lab Invest 1997 77 103 108 9251683
Gao YA Agnihotri R Vary CP Liaw L Expression and characterization of recombinant osteopontin peptides representing matrix metalloproteinase proteolytic fragments Matrix Biol 2004 23 457 466 15579312
Philip S Kundu GC Osteopontin induces nuclear factor kappa B-mediated promatrix metalloproteinase-2 activation through I kappa B alpha /IKK signaling pathways, and curcumin (diferulolylmethane) down-regulates these pathways J Biol Chem 2003 278 14487 14497 12473670
Agnihotri R Crawford HC Haro H Matrisian LM Havrda MC Osteopontin, a novel substrate for matrix metalloproteinase-3 (stromelysin-1) and matrix metalloproteinase-7 (matrilysin) J Biol Chem 2001 276 28261 28267 11375993
Brabletz T Jung A Dag S Hlubek F Kirchner T Beta-catenin regulates the expression of the matrix metalloproteinase-7 in human colorectal cancer Am J Pathol 1999 155 1033 1038 10514384
El-Tanani M Platt-Higgins A Rudland PS Campbell FC Ets gene PEA3 cooperates with beta-catenin-Lef-1 and c-Jun in regulation of osteopontin transcription J Biol Chem 2004 279 20794 20806 14990565
Chilosi M Poletti V Zamo A Lestani M Montagna L Aberrant Wnt/beta-catenin pathway activation in idiopathic pulmonary fibrosis Am J Pathol 2003 162 1495 1502 12707032
Xie Z Singh M Siwik DA Joyner WL Singh K Osteopontin inhibits interleukin-1beta-stimulated increases in matrix metalloproteinase activity in adult rat cardiac fibroblasts: Role of protein kinase C-zeta J Biol Chem 2003 278 48546 48552 14500723
Selman M Ruiz V Cabrera S Segura L Ramirez R TIMP-1, −2, −3, and −4 in idiopathic pulmonary fibrosis. A prevailing nondegradative lung microenvironment? Am J Physiol Lung Cell Mol Physiol 2000 279 L562 574 10956632
Kolb M Bonniaud P Galt T Sime PJ Kelly MM Differences in the fibrogenic response after transfer of active transforming growth factor-beta1 gene to lungs of “fibrosis-prone” and “fibrosis-resistant” mouse strains Am J Respir Cell Mol Biol 2002 27 141 150 12151305
Ashizawa N Graf K Do YS Nunohiro T Giachelli CM Osteopontin is produced by rat cardiac fibroblasts and mediates A(II)-induced DNA synthesis and collagen gel contraction J Clin Invest 1996 98 2218 2227 8941637
Gadeau AP Campan M Millet D Candresse T Desgranges C Osteopontin overexpression is associated with arterial smooth muscle cell proliferation in vitro Arterioscler Thromb 1993 13 120 125 8422334
Panda D Kundu GC Lee BI Peri A Fohl D Potential roles of osteopontin and alphaVbeta3 integrin in the development of coronary artery restenosis after angioplasty Proc Natl Acad Sci U S A 1997 94 9308 9313 9256478
O'Regan A The role of osteopontin in lung disease Cytokine Growth Factor Rev 2003 14 479 488 14563350
Sodek J Zhu B Huynh MH Brown TJ Ringuette M Novel functions of the matricellular proteins osteopontin and osteonectin/SPARC Connect Tissue Res 2002 43 308 319 12489175
Ponta H Sherman L Herrlich PA CD44: From adhesion molecules to signalling regulators Nat Rev Mol Cell Biol 2003 4 33 45 12511867
Kasper M Bierhaus A Whyte A Binns RM Schuh D Expression of CD44 isoforms during bleomycin-or radiation-induced pulmonary fibrosis in rats and mini-pigs Histochem Cell Biol 1996 105 221 230 8681040
Svee K White J Vaillant P Jessurun J Roongta U Acute lung injury fibroblast migration and invasion of a fibrin matrix is mediated by CD44 J Clin Invest 1996 98 1713 1727 8878421
Teder P Vandivier RW Jiang D Liang J Cohn L Resolution of lung inflammation by CD44 Science 2002 296 155 158 11935029
Sheppard D Functions of pulmonary epithelial integrins: From development to disease Physiol Rev 2003 83 673 686 12843406
Eliceiri BP Cheresh DA The role of alphav integrins during angiogenesis: Insights into potential mechanisms of action and clinical development J Clin Invest 1999 103 1227 1230 10225964
Isacke CM Yarwood H The hyaluronan receptor, CD44 Int J Biochem Cell Biol 2002 34 718 721 11950588
Zhu B Suzuki K Goldberg HA Rittling SR Denhardt DT Osteopontin modulates CD44-dependent chemotaxis of peritoneal macrophages through G-protein-coupled receptors: Evidence of a role for an intracellular form of osteopontin J Cell Physiol 2004 198 155 167 14584055
|
16128620
|
PMC1198037
|
CC BY
|
2021-01-05 11:13:39
|
no
|
PLoS Med. 2005 Sep 6; 2(9):e251
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020251
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612862110.1371/journal.pmed.0020255Research ArticleMedical InformaticsGeneral MedicineDrugs and Adverse Drug ReactionsThe Effect of Automated Alerts on Provider Ordering Behavior in an Outpatient Setting Automated Alerts in a Clinic SettingSteele Andrew W
1
*Eisert Sheri
2
Witter Joel
3
Lyons Pat
4
Jones Michael A
5
¤Gabow Patricia
6
Ortiz Eduardo
7
1Information Services, Denver Health, Denver, Colorado, United States of America,2Health Services Research, Denver Health, Denver, Colorado, United States of America,3General Internal Medicine, Denver Health, Denver, Colorado, United States of America,4Software Division, Siemens Medical Solutions, Malvern, Pennsylvania, United States of America,5Thomson Micromedex, Greenwood Village, Colorado, United States of America,6Medicine, Denver Health, Denver, Colorado, United States of America,7Medicine, Washington D.C. VA Medical Center, Washington, District of Columbia, United States of AmericaDavis Robert Academic EditorCenters for Disease Control and PreventionUnited States of America*To whom correspondence should be addressed. E-mail: [email protected]¤Current address: University of Colorado Hospital, Denver, Colorado, United States of America
Competing Interests: AWS, SE, JW, PG, and EO have no competing interests. PL is an employee of Siemens Medical Solutions. MAJ was employed by Thomson Micromedix during the time of the study.
Author Contributions: AWS, SE, JW, PL, MAJ, PG, and EO designed the study. AWS analyzed the data. AWS, JW, SE, PL, MAJ, PG, and EO contributed to writing the paper.
9 2005 6 9 2005 2 9 e25514 1 2005 21 6 2005 Copyright: © 2005 Steele 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.
Is it Possible to Change Prescribing Habits?
Background
Computerized order entry systems have the potential to prevent medication errors and decrease adverse drug events with the use of clinical-decision support systems presenting alerts to providers. Despite the large volume of medications prescribed in the outpatient setting, few studies have assessed the impact of automated alerts on medication errors related to drug–laboratory interactions in an outpatient primary-care setting.
Methods and Findings
A primary-care clinic in an integrated safety net institution was the setting for the study. In collaboration with commercial information technology vendors, rules were developed to address a set of drug–laboratory interactions. All patients seen in the clinic during the study period were eligible for the intervention. As providers ordered medications on a computer, an alert was displayed if a relevant drug–laboratory interaction existed. Comparisons were made between baseline and postintervention time periods. Provider ordering behavior was monitored focusing on the number of medication orders not completed and the number of rule-associated laboratory test orders initiated after alert display. Adverse drug events were assessed by doing a random sample of chart reviews using the Naranjo scoring scale.
The rule processed 16,291 times during the study period on all possible medication orders: 7,017 during the pre-intervention period and 9,274 during the postintervention period. During the postintervention period, an alert was displayed for 11.8% (1,093 out of 9,274) of the times the rule processed, with 5.6% for only “missing laboratory values,” 6.0% for only “abnormal laboratory values,” and 0.2% for both types of alerts. Focusing on 18 high-volume and high-risk medications revealed a significant increase in the percentage of time the provider stopped the ordering process and did not complete the medication order when an alert for an abnormal rule-associated laboratory result was displayed (5.6% vs. 10.9%, p = 0.03, Generalized Estimating Equations test). The provider also increased ordering of the rule-associated laboratory test when an alert was displayed (39% at baseline vs. 51% during post intervention, p < 0.001). There was a non-statistically significant difference towards less “definite” or “probable” adverse drug events defined by Naranjo scoring (10.3% at baseline vs. 4.3% during postintervention, p = 0.23).
Conclusion
Providers will adhere to alerts and will use this information to improve patient care. Specifically, in response to drug–laboratory interaction alerts, providers will significantly increase the ordering of appropriate laboratory tests. There may be a concomitant change in adverse drug events that would require a larger study to confirm. Implementation of rules technology to prevent medication errors could be an effective tool for reducing medication errors in an outpatient setting.
A computerized order entry system that alerted providers to potential problems was shown to be able to influence prescribing practice
==== Body
Introduction
Medical error reduction has become a major force for health care in the United States of America since the landmark publication of the 1999 Institute of Medicine report, To Err Is Human: Building a Safer Health System and the subsequent report, Crossing the Quality Chasm: A New Health System for the 21st Century [1,2]. It is estimated that 700,000 people die or are injured in hospitals each year from adverse drug events (ADEs) [2]. Among the preventable ADEs, more than half were associated with the ordering of the medication [3]. As many as 28% of the ADEs are thought to be associated with medication errors [4]. Computerized Physician Order Entry (CPOE) systems have the capability to address medication errors by utilizing dosing recommendations, reminders for selecting appropriate medications, and clinical checking for drug–allergy, and drug–drug interactions [5–7]. However, despite the potential, less than 10% of hospitals have implemented CPOE [8].
In the outpatient setting, studies have shown that between 18% and 25% of patients may have an ADE [9,10]. Many medications have specific renal, hematologic, and hepatic effects. Monitoring guidelines have been recommended for many of these medications. Health-care providers, however, may find it difficult to acquire and maintain the information to appropriately prescribe these medications. Utilization of a computerized clinical-decision support system linked to patient specific laboratory data within CPOE applications may assist in handling this information overload.
Types of ordering errors include selecting the wrong medication for the patient's illness, choosing an incorrect dose given specific patient characteristics such as age and renal function, and ordering medication when the patient is known to be allergic to the medication. Some errors occur when the provider is not aware of clinical information that is readily available within computer applications [11]. It has been shown that as many as 45% of medication errors may be related to drug–laboratory issues [12]. Further, appropriate monitoring of clinically relevant laboratory values is often inadequate. For example, despite four Federal Drug Administration warnings recommending monthly liver function testing for patients on an oral hypoglycemic, Troglitazone, follow-up studies found less than 5% compliance with these recommendations [13]. Lastly, it is impossible for providers to remember and stay current with all of the new information being provided. For example, the top forty prescribed medications each has on average seven drug–laboratory interactions [14].
This study evaluated the implementation of computerized provider order entry (CPOE) and an integrated computer-based clinical-decision support system (CDSS) in an outpatient clinic at Denver Health as tools for medication error reduction. The study was done in collaboration with Thomson Micromedex and Siemens Medical Solutions. The overarching purpose of the study was to determine the impact of using computerized alerts to improve the prescribing of medications in the outpatient setting. The study focused on drug–laboratory interactions related to medication use that can lead to hyper- and hypokalemia, nephrotoxicity, thrombocytopenia, and hepatic inflammation.
Methods
Patients and Setting
This study was conducted at Denver Health outpatient primary-care clinics. Denver Health is the principle safety net institution for Colorado. A single electronic record links primary-care clinics, specialty clinics, and the acute care hospital. Medication and laboratory orders are placed using CPOE. Table 1 provides the demographic characteristics of the patients seen during the study period. There were 19,076 patients seen during this 9-mo time period. Sixty-four percent of the patients were female, 82% were Hispanic, 42% had Medicaid coverage, 41% were uninsured, and 17% had Medicare, private, or other types of insurance.
Table 1 Demographics—All Patients, Westside Clinic (08/01/2002–04/30/2003)
All provider staff were allowed to enter medication orders including physicians, allied health providers (nurse practitioners, physician assistants), and residents. All registered patients were eligible for the intervention.
The Colorado Multiple Institutional Review Board (COMIRB) approved the study. COMIRB is a statewide board established to review biomedical and behavioral research involving humans to ensure appropriate ethical issues are addressed (see Protocol S1).
Data and Time Frame
Baseline results of rules application were collected from 08/01/2002 to 11/29/2002 (17 wk). The intervention was implemented on 12/01/2002. Postintervention data were collected between 12/01/2002 and 04/30/2003 (21 wk). The timeframe was chosen to provide a sample size of at least 475 which would provide a power (1 − β) of 90% with an α level of 0.05 to detect a difference of 5%, going from 5% to 10% between the order cessation rate comparing preintervention to postintervention phases.
Interventions
Denver Health has utilized rules technology for over two years. As part of the collaboration with Thomson Micromedex, this study used commercially available rules developed in Arden Syntax language as Medical Logic Modules (MLMs) and modified them to meet the local needs. Although other types of rules were considered, only rules that were commercially available from Thomson Micromedex and applicable to the outpatient setting were used for this study. All rules from Thomson Micromedex were reviewed, and it was determined that the most appropriate available rules to address patient safety in the outpatient setting covered the following five areas: (1) potential drug-induced hypokalemia; (2) potential drug-induced hyperkalemia;(3) potential drug-induced nephrotoxicity; (4) potential drug-induced thrombocytopenia; and (5) potential drug-induced hepatotoxicity.
This study focused mainly on medications that were associated with the above conditions and more commonly used in the outpatient clinic, compared to other medications that were largely used in an inpatient setting. These medications and the associated rules are shown in Table 2.
Table 2 Example of Medications Used in Rules for Associated Lab Abnormalities
For each drug–laboratory interaction, rules were written identifying medications, routes of administration, and abnormal laboratory threshold levels for inclusion in the rule. In addition, a determination was made for each medication as to whether an alert should be provided for an abnormal laboratory value only or either an abnormal laboratory value or a missing laboratory value, or, despite an association with the laboratory abnormality, no alert would be displayed to the provider.
In response to the alerts, providers could decide to keep, revise, or delete the medication order. They could also order any rule-associated laboratory tests. Changes to orders could not be collected given the technological capabilities of the CPOE system, but lack of a medication order after the rule was triggered was evidence that the provider decided to stop the ordering process and not order the medication during that session. Likewise, comparing the rates of ordering of rule-associated laboratory tests before and after the intervention provided a measure of the efficacy of the intervention.
Thomson Micromedex provided the knowledge content for each of the above rules. One physician (AWS) and a pharmacist (MAJ) utilized a reference THOMSON Micromedex Healthcare Series Integrated Databases and then agreed upon changes to the rules criteria. This was reviewed by a second physician (JW) and some minor changes were subsequently made. The final criteria list was provided to the rules builder staff to incorporate into the rules. The Arden Syntax rules contained within the Medical Logic Modules were integrated into the rules engine application. Due to differences in terminology and unique characteristics of the Denver Health laboratory, order entry, and pharmacy databases, portions of the rules had to be rewritten to conform to these local needs. However, the logic of the rule remained intact. The content and layout of the alert screens were designed by the Information Services staff with input from clinic providers. (Figure 1) The laboratory cutoff values for triggering an alert were the same as the Denver Health abnormal laboratory reference ranges. A timeframe of 6 mo was chosen for using historical laboratory data. Various rules output-message display formats were presented to a group of providers, and consensus was reached on a final display.
Figure 1 Rules Output Screen for “Recommended Laboratory Value Not Done in Last Six Months”
During 2002, CPOE was implemented at one of Denver Health's larger outpatient clinics, the Sam Sandos Family Health Clinic. Approximately 120 users were entering over 6,000 orders per week, 40% of which were for medications. As providers selected medications in the ordering process, a rule processed information on the five drug–laboratory interactions listed above. If rule criteria were met for a drug–laboratory interaction, an alert screen was presented to the provider with a message containing patient name, type of rule alert, name of medication that triggered the alert, and a message with laboratory results, if available, and suggestions to consider deleting or changing the medication or to consider ordering a rule-associated laboratory test. Providers did not need to respond to the alert, but needed to select “Continue” to proceed with the ordering session. They were free to make any changes in their ordering process.
No specific provider education was given to the staff concerning the importance and types of drug–laboratory interactions contained within this study.
Analytical Approach
A nonrandomized pre- and postcomparison of the intervention clinic was accomplished by turning rules on in the background without displaying any message to providers during the preintervention time period. Because the rules were processing in the background, the provider did not receive any alerts recommending changes in their orders. This baseline ordering behavior was then compared to ordering behavior after alerts were presented to the provider.
Medications prescribed and laboratory ordering volume and type were measured at intervention clinics from automated computerized order entry log files. Rules performance was measured by looking at total rules triggered, and further subdividing into rules triggered with no alert provided, rules triggered with alert for missing laboratory tests, and rules triggered with alert for abnormal laboratory values. Provider order behavior was monitored focusing on the number of medication orders not completed after alert display, and the number of rule-associated laboratory test orders initiated after alert display. ADEs were assessed by doing a random sample of chart reviews using the Naranjo scoring scale [15]. The chart audit was limited to a sample of charts for which the most recent rule-associated laboratory value was abnormal within the last 6 mo. The Naranjo criteria has four categories for ADEs; “definite,” “probable,” “possible,” and “doubtful.” This analysis used the combination of “definite” and “probable” as being a potential ADE. Statistical comparisons were made using Fisher's exact test or Generalized Estimating Equations, as appropriate, using a SAS statistical package (SAS Institute, Cary, North Carolina, United States).
Results
The rule processed 16,291 times during the study period: 7,017 during the preintervention period and 9,274 during the postintervention period (see Table 3) During the time span of the study there were 54,206 patient visits; medications were ordered on 17,444 (32%) of the visits. The rule processed on 49% of all medication orders. During the postintervention period, an alert was displayed to the care provider for 11.8% (1,093 out of 9,274) of the times the rule processed. Among these alerts, 5.6% were for only “missing laboratory values,” 6.0% were for only “abnormal laboratory values,” and 0.2% were for both types of alerts (see Figure 2). The rule did not have an appreciable negative effect on system performance. On average, the rule delayed processing of the screens in the CPOE application by less than 2 sec. There were no complaints from providers about slow system performance related to the rules processing during this study.
Figure 2 Medication Orders That Triggered Alerts
Table 3 Drug–Laboratory Interaction Rules Performance Summary
Comparing the pre- and postintervention periods for when any type of alert was presented, there was not a statistical difference in the rate of the provider not completing the medication order (5.4% vs. 8.3%, p = 0.17). However, when the alert was for an abnormal laboratory value, the percentage of times the medication order was not completed increased from 5.6% at baseline to 10.9% during the intervention (p = 0.03).
Comparing the pre- and postintervention periods for medication orders for which no alert was displayed shows no significant change in the percentage of time the provider ordered the rule-associated laboratory test (17.0% during preintervention period vs. 16.2% during the postintervention period, p = 0.38). This indicates that there was no trend, in general, to increased laboratory test ordering during the study period. Focusing on medication orders for which an alert was presented shows an increase in the percentage of time the provider ordered the rule-associated laboratory test (38.5% vs. 51.1%, p < 0.001). The largest effect was noticed when the alert was triggered for a missing laboratory test: the percentage of times the provider ordered the rule-associated laboratory test increased from 43.0% at baseline to 62.0% (p < 0.001).
One investigator (JW) reviewed a random sample of charts. The study focused on medication orders for which an alert was displayed indicating that the patient had a rule-associated abnormal laboratory value. A total of 163 charts were reviewed: 116 from the preintervention period and 47 from the postintervention period (Table 4). Overall, by combining “definite” and “probable” categories within the Naranjo scoring criteria, 12 (10.3%) of charts in the preintervention group had a potential ADE, and 2 (4.3%) of charts in the postintervention group had a potential ADE (p = 0.35 by Fisher's exact test).
Table 4 Adverse Drug Events Identified through Chart Review (Naranjo Scoring)
Discussion
Health-care organizations are struggling with methods to improve the quality of care provided in a cost-efficient manner. Patient safety issues are primary concerns for health-care institutions and providers. Numerous examples have been published assessing the role of technology in assisting in these efforts [16–19]. The vast majority of data on technological interventions is focused on the inpatient hospital setting, often at tertiary care institutions, usually with house staff (physicians in training) programs, and rarely looks at commercially available applications [17,20–22]. Among geriatric patients, studies have shown a rate of 13.8 preventable ADEs per 1,000 person-years [23]. At Brigham and Women's Hospital, Boston, Massachusetts, implementation of inpatient CPOE led to an 81% decline in non-missed-dose medication error rates overall, and an 86% reduction in the intensive care units [24]. In an emergency department setting, computer-assisted prescriptions were more than three times less likely to contain errors than handwritten prescriptions [25].
In contrast, this study at Denver Health looked at a very specific type of clinical-decision support system: the use of a rules technology to prevent drug–laboratory adverse drug events. The clinical-decision support application and rule knowledge were both obtained from commercial vendors, and the CPOE application was a commercially available application. The setting was unique in that it was a primary-care outpatient setting. Furthermore, faculty physicians, as compared to training physicians, entered the majority of orders.
The clinical outcome portions of the study focused on assessing the effect of clinical-decision support systems on changing ordering behavior and, ultimately, in reducing ADEs among the patients. The study was not designed to have an adequate sample size to detect statistical difference in ADEs, although there were non-statistically fewer ADEs during the intervention phase. The rules did demonstrate a significant ability to change the ordering behavior of the provider. The effect was modest in halting the ordering of the medication and appeared to be limited to occasions in which the alert presented an abnormal laboratory value, with almost a doubling in order cessation. Still, the provider continued with ordering the medicine despite a warning message in the vast majority (91.7%) of the orders. This may be due to the providers deciding that the benefits of the medication far outweighed potential adverse effects on the associated laboratory abnormalities. In contrast, across all medication orders, and all categories of rules, ordering of the appropriate rule-associated laboratory test increased significantly (33% increase) with the presentation of an alert. The strongest effect was when providers where alerted to “missing” laboratory results (42% increase). Similar results have been found by Galanter et al. [26] when looking at automated safety alerts interactions between digoxin and potassium. In their study, checking for unknown potassium values increase from 9% to 57% after implementation of alerts. Likewise, in a community-based intervention by Hoch et al. [27], computerized alerts for missing potassium values sent the day after physicians had ordered a diuretic led to a 9.8% increase in potassium testing. Our study differed from these studies in that we looked at numerous medications across different therapeutic categories.
There was less of an effect on ordering behavior when the alert informed the provider of the existence of an abnormal laboratory value (23% increase in ordering of the test). This may imply that the cutoff values for the “abnormal” trigger were set too low, and that providers felt that repeating the laboratory test was not warranted given the degree of the abnormality. Further analysis, looking at the severity of the laboratory abnormality and correlating that to ordering behavior, may provide more insight to this issue.
There are various limitations to this study. The intervention only focused on a group of select drug–laboratory interactions and thus the results may not be generalizable to other types of interventions focusing on other patient-care issues. Further, the setting was in a single primary-care clinic outpatient setting within a large public-health integrated health-care delivery system and results may be different in other settings such as hospitals and private physician offices. The patient population served is primarily a lower income, minority-dominated (∼80% Hispanic), and medically underserved population. Different results may be obtained with a more affluent patient population. The study did not consider alert effectiveness based on the role of the provider. Further studies would be needed to determine if the provider role (i.e., staff physician, house staff, or nurse practitioner) may alter the effect of the alerts. Finally, as an evaluation of an intervention, the intervention was not randomized. Changes observed may have been occurring in the health-care environment irrespective of the intervention. The investigators are aware of no local or national initiatives to improve the care of these patients for the rule-associated conditions.
We conclude that with private–public entity collaboration, rules for drug–laboratory interactions can be encoded into computerized clinical applications in primary-care clinics within an integrated health-care delivery, safety-net institution. Further, with the use of clinical-decision support, providers will more often stop the ordering of medications when alerted to potential drug–laboratory interactions and will order more appropriate medication-associated laboratory tests. There may be an effect on ADEs. Future larger, more prolonged studies will help to determine the full relationship between automated alerts for drug–laboratory interactions and the related clinical outcomes of adverse drug events.
Supporting Information
Protocol S1 Human Participant Approval
(10 KB PDF).
Click here for additional data file.
Protocol S2 Initial Protocol
(61 KB DOC).
Click here for additional data file.
Patient Summary
Background
All drugs have unwanted side effects (also known as adverse drug events), and when drugs are combined the chances of side effects increase. It is almost impossible for individual physicians to keep up to date with all possible drug effects. Increasingly, prescription orders and patient records are transmitted and stored on computers rather than being handwritten. As well as improving their legibility, computer writing of prescriptions also makes it possible to design programs that look at a patient's record when the prescription is written and that check for any possible problems.
Why Was This Study Done?
The authors wanted to investigate whether such a program could be used in an outpatient setting to change the behavior of physicians and ultimately to reduce the number of adverse drug reactions.
What Did the Researchers Do and Find?
In a single outpatient facility in Denver, Colorado, they designed a program to alert prescribers when one of five possible adverse events was likely to occur, or when the patient required further tests to establish whether the drug was likely to be safe. They tested the effect of the program by looking at what physicians did when the alerting system was switched off, and then when it was switched on. They found that it was possible to alter the behavior of prescribers by alerting them to possible problems; prescribers were more likely to stop a prescription or to order more tests when they were alerted. However, the study was too small to show for sure whether there was any true effect on adverse drug reactions.
What Do These Findings Mean?
Programs such as this one might be useful in alerting prescribers to potential problems with the drugs they are intending to prescribe. However, further work will need to be done to see if these programs can reduce the adverse events that patients experience, and whether they will work in other hospitals and clinics.
Where Can I Get More Information Online?
The US Web site MedlinePlus has a page of links on patient issues such as adverse reactions:
http://www.nlm.nih.gov/medlineplus/patientissues.html
The US Agency for Healthcare Research and Quality (AHRQ) provides a continuously updated, annotated, and carefully selected collection of patient safety news, literature, tools, and resources, including the Patient Safety Network:
http://psnet.ahrq.gov/
In the UK, the National Patient Safety agency site has information on many aspects of patient safety:
http://www.npsa.nhs.uk/
We thank the staff and patients from the clinics who participated in this study and provided excellent input throughout the project. Richard Read of Read, Inc, Evergreen, Colorado, provided statistical support.
This research was performed by Denver Health through the support of the Agency for Healthcare Research and Quality (AHRQ) under Contract 290–00–0014, Task Order No. 3 (see Protocol S2). The authors of this article are solely responsible for its contents. No statements or views in this article should be construed as endorsements or positions of AHRQ, the U.S. Department of Health and Human Services, the Department of Veterans Affairs, Siemens Medical Solutions, Thomson Micromedex, or the Federal government.
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Citation: Steele AW, Eisert S, Witter J, Lyons P, Jones MA, et al. (2005) The effect of automated alerts on provider ordering behavior in an outpatient setting. PLoS Med 2(9): e255.
Abbreviations
ADEadverse drug event
CPOEComputerized Physician Order Entry
==== Refs
References
Kohn LT Corrigan JM Donaldson MS To err is human: Building a safe health system 1999 Washington (DC) National Academies Press 312
Institute of Medicine Crossing the quality chasm: A new health system for the 21st century 2001 Washington (DC) National Academies Press 364
Bates DW Cullen DJ Laird N Petersen LA Small SD Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE Prevention Study Group JAMA 1995 274 29 34 7791255
Bates DW Boyle DL Vander Vliet MB Schneider J Leape L Relationship between medication errors and adverse drug events J Gen Intern Med 1995 10 199 205 7790981
Kuperman GJ Teich JM Gandhi TK Bates DW Patient safety and computerized medication ordering at Brigham and Women's Hospital Jt Comm J Qual Improv 2001 27 509 521 11593885
Shojania KG Yokoe D Platt R Fiskio J Ma'luf N Reducing vancomycin use utilizing a computer guideline: Results of a randomized controlled trial J Am Med Inform Assoc 1998 5 554 562 9824802
Evans RS Pestotnik SL Classen DC Clemmer TP Weaver LK A computer-assisted management program for antibiotics and other antiinfective agents N Engl J Med 1998 238 232 238
Ash JS Gorman PN Seshadri V Hersh WR Computerized physician order eEntry in U.S. hospitals: Results of a 2002 survey J Am Med Inform Assoc 2004 11 95 99 14633935
Gandhi TK Burstin HR Cook EF Puopolo AL Haas JS Drug complications in outpatients J Gen Intern Med 2000 15 149 154 10718894
Gandhi TK Weingart SN Borus J Seger AC Peterson J Adverse drug events in ambulatory care N Engl J Med 2003 348 1556 1564 12700376
Leape LL Bates DW Cullen DJ Cooper J Demonaco HJ Systems analysis of adverse drug events. ADE Prevention Study Group JAMA 1995 274 35 43 7791256
Hulse RK Clark SJ Jackson JC Warner HR Gardner RM Computerized medication monitoring system Am J Hosp Pharm 1976 33 1061 1064 973633
Graham DJ Drinkard CR Shatin D Tsong Y Burgess MJ Liver enzyme monitoring in patients treated with troglitazone JAMA 2001 286 831 833 11497537
Schiff GD Klass D Peterson J Shah G Bates DW Linking laboratory and pharmacy: Opportunities for reducing errors and improving care Arch Intern Med 2003 163 893 900 12719197
Naranjo CA Busto U Sellers EM Sandor P Ruiz I A method for estimating the probability of adverse drug reactions Clinl Pharmacol Ther 1981 30 239 245
Mekhjian HS Kumar RR Kuehn L Bentley TD Teater P Immediate benefits realized following implementation of physician order entry at an academic medical center J Am Med Inform Assoc 2002 9 529 539 12223505
Bates DW Leape LL Cullen DJ Laird N Petersen LA Effect of computerized physician order entry and a team intervention on prevention of serious medication errors JAMA 1998 280 1311 1316 9794308
Raschke RA Gollihare B Wunderlich TA Guidry JR Leibowitz AI A computer alert system to prevent injury from adverse drug events: Development and evaluation in a community teaching hospital JAMA 1998 280 1317 1320 9794309
Pestotnik SL Classen DC Evans RS Burke JP Implementing antibiotic practice guidelines through computer-assisted decision support: Clinical and financial outcomes Ann Intern Med 1996 124 884 890 8610917
Teich JM Merchia PR Schmiz JL Kuperman GJ Spurr CD Effects of computerized physician order entry on prescribing practices Arch Intern Med 2000 160 2741 2747 11025783
Potts AL Barr FE Gregory DF Wright L Patel NR Computerized physician order entry and medication errors in a pediatric critical care unit Pediatrics 2004 113 59 63 14702449
King WJ Paice N Rangrej J Forestell GJ Swartz R The effect of computerized physician order entry on medication errors and adverse drug events in pediatric inpatients Pediatrics 2003 112 506 509 12949274
Gurwitz JH Field TS Harrold LR Rothschild J Debellis K Incidence and preventability of adverse drug events among older persons in the ambulatory setting JAMA 2003 289 1107 1116 12622580
Bates DW Teich JM Lee J Seger D Kuperman GJ The impact of computerized physician order entry on medication error prevention J Am Med Inform Assoc 1999 6 313 321 10428004
Bizovi KE Beckley BE McDade MC Adams AL Lowe RA The effect of computer-assisted prescription writing on emergency department prescription errors Acad Emerg Med 2002 9 1168 1175 12414466
Galanter WL Polikaitis A DiDomenico RJ A trial of automated safety alerts for inpatient digoxin use with computerized physician order entry J Am Med Inform Assoc 2004 11 270 277 15064288
Hoch I Heymann AD Kurman I Valinsky LJ Chodick G Countrywide computer alerts to community physicians improve potassium testing in patients receiving diuretics J Am Med Inform Assoc 2003 10 541 546 12925546
|
16128621
|
PMC1198038
|
CC BY
|
2021-01-05 11:13:39
|
no
|
PLoS Med. 2005 Sep 6; 2(9):e255
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020255
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1617383010.1371/journal.pmed.0020262Policy ForumMedical EthicsEthicsResearch MethodsResearch designClinical trialsDeception in Research on the Placebo Effect Policy ForumMiller Franklin G *Wendler David Swartzman Leora C Franklin G. Miller and David Wendler are in the Department of Clinical Bioethics, National Institutes of Health, Bethesda, Maryland, United States of America. Leora C. Swartzman is in the Department of Psychology, University of Western Ontario, London, Ontario, Canada. The opinions expressed are those of the authors and do not necessarily reflect the position or policy of the National Institutes of Health, the Public Health Service, or the Department of Health and Human Services.
*To whom correspondence should be addressed. E-mail: [email protected]
Competing Interests: The authors declare that no competing interests exist.
9 2005 6 9 2005 2 9 e262Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.A common feature of research investigating the placebo effect is deception of research participants about the nature of the research. Miller and colleagues examine the ethical issues surrounding such deception.
==== Body
The placebo effect is a fascinating yet puzzling phenomenon, which has challenged investigators over the past 50 years. Recently, it has been defined as the “positive physiological or psychological changes associated with the use of inert medications, sham procedures, or therapeutic symbols within a healthcare encounter” [1]. Increasing scientific inquiry has been aimed at elucidating the mechanisms responsible for placebo effects and determining how inert interventions can lead to positive changes in patients [1,2]. The majority of placebo mechanism research has been done within the context of experimental and clinical pain.
Patients' expectations for improvement, also referred to as “response expectancies,” are thought to be one of the central mechanisms responsible for placebo effects [3–5]. Brain imaging techniques are being used to explore both the neurophysiological correlates of these expectations and the mechanisms underlying placebo effects in a variety of contexts, including pain relief in healthy participants, relief of symptoms of depression, and motor functioning in patients with Parkinson disease [6–8]. Understanding these mechanisms is an important step in harnessing the placebo effect in patient care. In the words of a National Institutes of Health request for applications, “understanding how to enhance the therapeutic benefits of placebo effect in clinical practice has the potential to significantly improve healthcare” [9]. Toward that end, the National Institutes of Health invited submissions for systematic studies aimed at discerning the psychosocial factors (including expectancy) in the patient–clinician relationship and/or in the health-care environment that can potentiate healing.
A common feature of research investigating the placebo effect is deception of research participants about the nature of the research. This use of deception is considered necessary to understanding the placebo effect, but has received little systematic ethical attention. In this article, we examine ethical issues relating to deception in research on the placebo effect, with a particular focus on experiments involving patients in clinical settings. We propose a method of informing participants about the use of deception that can reconcile the scientific need for deceptive research designs with the ethical requirements for clinical research.
Altering Expectations to Examine Placebo Mechanisms
Response expectancy is seen to be a major driving force behind the placebo effect. Therefore, a common (and some would argue, necessary) feature of research aimed at elucidating placebo mechanisms is the use of deception in experimental manipulation of participants' expectations (e.g., about whether or not they will receive a “powerful pain killer” or a “sugar pill”), while holding constant the pharmacological (or other) properties of the administered intervention. This research has clearly shown (across a wide range of clinical conditions) that altering expectancies for improvement has an impact on therapeutic outcomes [8,10–13].
The tension between scientific methods for elucidating the placebo effect and ethical norms for conducting research involving human participants is illustrated most clearly by “the balanced placebo design,” an approach designed to disentangle the relative effects of pharmacology and response expectancy. Table 1 displays the balanced placebo design in a way that highlights the deception of participants that occurs in two of the four arms of the design.
Deception of research participants is considered necessary to understanding the placebo effect—but has received little ethical attention
(Illustration: Margaret Shear, Public Library of Science)
Table 1 The Balanced Placebo Design
In the balanced placebo design, investigators manipulate both expectancies (e.g., informing participants that they will receive a drug versus informing them that they will receive placebo) and the pharmacological agent (giving a drug versus giving a placebo). As reviewed by Swartzman and Burkell, researchers using this paradigm with healthy volunteers have shown that expectation plays a role in the subjective and behavioral effects of a range of psychoactive substances [14]. These substances include dexamfetamine, alcohol, caffeine, nicotine, and tetrahydrocannabinol [15–19].
The balanced placebo design offers a powerful and elegant approach to evaluate drug versus expectancy effects and their interactions. As Kirsch notes, this design yields information that cannot be obtained from conventional clinical trials [20]. It provides a direct assessment of the drug effect, independent of expectancy, and the nondeceptive arms are more ecologically valid than the double-blind administration in conventional randomized trials (i.e., they mimic what goes on in the real world of clinical practice). Thus, it is not surprising that Caspi recently suggested that the balanced placebo design “be used more often in clinical trials of drug efficacy” [21]. Despite the methodological virtues of the balanced placebo design, and its prior use in healthy volunteers, we are unaware of any trials that have employed this approach with patients. Clinical investigators likely have avoided use of the balanced placebo design out of concern for the ethical acceptability of deceiving patients.
An often cited article on the balanced placebo design characterized the deception in the following way: “Although deception is involved, it is no greater than is involved in any study using placebos” [22]. However, this defense of the balanced placebo design confuses the ethical issues it raises. Placebo-controlled trials aimed at evaluating the efficacy of treatments may be regarded as having an element of deception, since the placebo control is designed to appear indistinguishable from the active treatment under investigation. Nevertheless, when these studies are conducted under effective double-blind conditions, participants are told that they will receive either a drug or a placebo, and neither the investigators nor the research participants know which intervention is received by any of the participants. Accordingly, administering the study interventions, unlike the situation of the balanced placebo design, does not involve intentionally false communication; it requires investigators to withhold information, but not to lie to participants about the interventions they will receive.
When deception is used, a conflict between the means and ends of scientific investigation ensues.
Research designed to understand the placebo effect by deceptively manipulating the expectations of participants holds great promise for understanding the psychological and neurobiological dimensions of healing. However, to pursue this research while respecting participants, it is necessary to develop an approach that reconciles the outright deception involved in placebo research, including the balanced placebo design, with the ethical norms governing clinical research.
What Makes Deception in Scientific Investigation Ethically Problematic?
At the outset, it is useful to appreciate the conflict between the ethos of science and the use of deceptive techniques. Science aims to discover and communicate the truth about the natural world and human conduct. There are sound methodological reasons for using deception to probe for the truth about human attitudes and beliefs and their effects on behavior. It follows, however, that when deception is used, a conflict between the means and ends of scientific investigation ensues: the end of discovering the truth is pursued by the means of deliberate untruth.
It might be argued that deception in scientific investigation is no more problematic than the pervasive and accepted use of deception in daily life and in social contexts [23]. In a recent news article reporting advances in the design of computers to simulate human responsiveness, Clifford Nass, a professor of communication at Stanford University, endorses the deception involved in this project: “We spend enormous amounts of time teaching children to deceive—it's called being polite or social. The history of all advertising is about deceiving. In education, it's often important to deceive people—sometimes you say, ‘Boy you are really doing good,’ not because you meant it but because you thought it would be helpful” [24].
Deception in ordinary life typically is justified on the grounds that it is for the benefit of the individual who is being deceived. For instance, the polite and social deception that Nass cites is justified on the grounds that it is better to deceive someone slightly than to criticize the person or to hurt the person's feelings. Notice, however, that this condition is not relevant to placebo research, including the balanced placebo design. In placebo research, participants are not deceived for their own benefit. Rather, they are deceived for the benefit of science and society in general, through the development of generalizable knowledge.
Deception of research participants also clearly conflicts with the ethical norms governing clinical research [25,26]. First, it violates the principle of respect for persons by infringing on the right of prospective research participants to choose whether to participate in research based on full disclosure of relevant information. Second, it may manipulate individuals to volunteer when they otherwise would not have chosen to do so had they been informed accurately about the nature of the research, including its use of deception. For these reasons, deception, as it is currently practiced in the conduct of research on the placebo effect, is incompatible with informed consent.
Third, although scant systematic data have been collected on the effects of deception on clinical research participants, some available evidence indicates that when the deception is revealed, as in the debriefing process that often accompanies deceptive research, it causes distress to at least some participants [27]. The adverse impact of deception in psychological research, and whether it can be reversed adequately through a debriefing process, is a subject of debate [28–31]. Furthermore, deception in research involving patients in clinical settings may prove more upsetting. This is because participants in deceptive psychological research are, for the most part, psychology undergraduates who often are aware that deception is sometimes used in psychological research [32]. Patients, in contrast, legitimately expect to be able to trust in, and receive truthful communication from, clinicians and clinical investigators. This trust is violated by the use of deception. Especially problematic is the use of deception in experiments conducted by clinicians who have a prior clinician–patient relationship with the patients enrolled in the study. When patients learn about the use of deception in the process of debriefing, which is a common feature of deception research, they may feel that their trust has been violated. Consequently, deception of patients may have deleterious effects on the willingness of patients to volunteer for future clinical research. More importantly, by undermining patients' faith in the truthfulness of physicians, deception might interfere with the future medical care of those who have experienced deceptive research.
Deception may be harmful not only to those who are deceived but also to those who practice it.
Finally, deception in research raises ethical concern because it can be corrupting for the professionals who practice it, and for those who witness it. According to an ancient perspective in moral philosophy, moral character depends on habits of conduct [33]. The use of deception in research may interfere with the disposition not to lie or deceive persons. This problem is compounded when the study design requires deception at the initiation of the trial as well as repeated deception of participants while conducting the research. Those who witness deception, especially if performed or sanctioned by professionals in positions of authority, may develop skewed perceptions of the ethics of deception, which may have negative consequences for the development of moral character. In sum, deception in research is prima facie wrongful, and it may be harmful not only to those who are deceived but also to those who practice or witness it.
The American Psychological Association's guidelines [34] are perhaps the most prominent attempt to reconcile the use of deception with the ethical norms of human participant research. According to guideline 8.07 (Deception in Research), “(a) psychologists do not conduct a study involving deception unless they have determined that the use of deceptive techniques is justified by the study's significant prospective scientific, educational, or applied value and that effective nondeceptive alternative procedures are not feasible; (b) psychologists do not deceive prospective participants about research that is reasonably expected to cause physical pain or severe emotional distress; (c) psychologists explain any deception that is an integral feature of the design and conduct of an experiment to participants as early as is feasible, preferably at the conclusion of their participation, but no later than at the conclusion of the data collection, and permit participants to withdraw their data.”
We have argued elsewhere that these three conditions are not sufficient to address the ethical concerns raised by deceptive research [25,26]. In particular, these conditions fail to address the fact that concealing the use of deception itself may affect individuals' decision to participate in research and precludes individuals from deciding whether they want to participate in deceptive research. To be sure, the use of debriefing may mitigate the potential harmful consequences of deceitful communication by explaining the rationale for deception. However, just as compensation for damages caused by negligence or restitution for crime does not cancel an infringement of a person's rights, debriefing does not cancel the violation of the principle of respect for persons. To consider how these ethical concerns arise in actual practice, and what steps might be taken to address them, it will be helpful to consider specific examples of the use of deception in placebo research (Table S1).
Examples of Deception in Placebo Research
First, in an experiment investigating suggestion and expectation relating to placebo analgesia, 13 women with irritable bowel syndrome were recruited, and were subjected to visceral pain evoked by rectal distention, using a balloon attached to a rectal catheter. The experiment took place under five experimental conditions: (1) natural history (no intervention relating to or disclosure about the pain stimulus), (2) rectal placebo (a sterile surgical lubricant placed on the balloon, described as effective in relieving pain), (3) rectal lidocaine, (4) oral lidocaine, and (5) rectal nocebo (a placebo intervention accompanied by a disclosure that the intervention often causes increased pain) [13]. Notably, the research report stated that “the gastroenterologist who performed the study was the doctor the patients normally consulted in the clinic.”
The investigators described their disclosure to the participants as follows: “The patients were told that four drugs that reduced and increased pain in relation to IBS [irritable bowel syndrome], respectively, were being tested, and that they had been proven effective in preliminary studies” [13]. In reality, the participants were administered two different forms of only one drug, along with two placebos. Hence, the participants were deceived by being informed that they would receive drugs that in fact were placebo interventions.
Second, investigators recruited patients with asthma, from an academic medical center, to participate in an experiment examining changes in forced expiratory volume in one second following administration of inhaled saline described deceptively as either a bronchoconstrictor or a bronchodilator [12]. The purpose was to determine the impact of suggestion on a placebo intervention in patients identified as suggestible or suggestion-resistant, based on a validated rating scale. The disclosure to the research participants was described in the article reporting the experimental results as follows: “Patients were contacted via telephone and informed that the investigators were hoping to understand how medications produce beneficial effects in asthma, including whether telling subjects about the potential effects of various medications would alter response to these agents. Patients were not told that they would be exposed to placebo interventions.” The study thus used elaborate deception, which involved an inaccurate account of the nature of the research and false descriptions of a placebo intervention. It is therefore puzzling that the authors reported that “all patients gave informed consent to participate in the study,” especially since there was no indication that the participants were informed that deception would be employed. Instead, the participants were debriefed about the study at the end of the experiment.
Authorized Deception
Can deceptive research be made compatible with informed consent? Use of deception is not consistent with fully informed consent. If participants are told the true purpose of research and the nature of all procedures, there would be no deception. However, participants can be informed prior to deciding whether to volunteer for a study that the experimental procedures will not be described accurately or that some features of these procedures will or may be misleading or deceptive [25,26]. This approach, which we call “authorized deception,” permits research participants to decide whether they wish to participate in research involving deception and, if so, to knowingly authorize its use. Authorized deception is compatible with the spirit of informed consent. It fosters respect for persons, despite the use of deception, by alerting prospective participants to the fact that some or all participants will be deliberately deceived about the purpose of the research or the nature of research procedures.
For example, investigators using the balanced placebo design to study expectancy and pharmacological effects of dexamfetamine described the informed consent disclosure as follows: “For ethical reasons it was stated in the consent form that ‘…some information and/or instructions given [to the participant] may be inaccurate’” [15]. This statement recognizes the ethical force of authorized deception, but does not seem to go far enough. As illustrated above, the balanced placebo design involves lying to participants in two arms of the study: some participants are told that they are being administered a particular drug when in fact they receive placebo, and others that they are being administered placebo when in fact they receive the drug. Consequently, it is at best an understatement to describe the disclosure in this experiment as possibly involving “inaccurate” information. It would be more accurate to inform the prospective participants that some research participants will be misled or deceived.
Use of deception is not consistent with fully informed consent.
Variants of the authorized deception approach have been advocated, and sometimes evaluated experimentally, since the 1970s [23,35–37]; however, it has not become a routine feature of research using deception [38]. In order to solicit informed authorization for the use of deception, the informed consent document could be worded as follows: “You should be aware that the investigators have intentionally misdescribed certain aspects of the study. This use of deception is necessary to obtain valid results. However, an independent ethics committee has determined that this consent form accurately describes the major risks and benefits of the study. The investigator will explain the misdescribed aspects of the study to you at the end of your participation.”
When deception of study participants is necessary and justified by the scientific value of the study, the use of authorized deception makes the process of deceptive research transparent. Participants are informed that they will be misled or deceived, though obviously the exact nature of the deception cannot be disclosed. They are assured that the research has been reviewed and approved by an ethics oversight committee that has no vested interest in the research in question, and that no important risks, other than the risks of the deception itself, have been concealed. Finally, they are informed that debriefing will occur.
Methodological Objections and the Need for Future Study
One possible objection to the technique of authorized deception is that it is liable to defeat the purpose of using deception to obviate potentially biased responses of research participants to research interventions. Informing participants that deception will occur (particularly in a study that involves administration of a placebo) is apt to make them suspicious and wary, thus possibly contributing to biased data. This methodological risk is avoided in most deceptive research, which does not employ this technique, provided that prospective participants do not otherwise suspect that deception will be used. However, limited available research indicates that the anticipated biased results from disclosing the possibility of deception do not necessarily occur.
Holmes and Bennett assessed this methodological concern experimentally. Psychology students were exposed to a deceptive experiment in which they were falsely informed that two to eight “painful electric shocks” would be administered at random times after a red signal light appeared [35]. No shocks actually were administered. Measures of self-reported anxiety and physiological arousal (pulse and respiration rates) were obtained. Prior to the deceptive shock intervention, one experimental group was informed that deception is occasionally used in psychology experiments to assure unbiased responses. The other group exposed to the deceptive shock intervention did not receive any information about the possibility of deception. No outcome differences were observed for participants informed of the possibility of deception versus those not informed.
The information about deception in this experiment, however, falls short of the authorized deception approach that we recommend. It disclosed to prospective participants that deception is a possibility in “a few experiments,” rather than informing them that deception would actually be employed for all or some participants in the particular experiment in which they were invited to enroll. In contrast, Wiener and Erker directly tested the authorized deception approach, described as “prebriefing,” in an experiment evaluating attributions of responsibility for rape based on transcripts from an actual rape trial [37]. Participants (68 undergraduate psychology students) were either correctly informed or misinformed about the jury verdict regarding the defendant's guilt. Half of participants received an informed-consent document stating that “you may be purposefully misinformed.” The other half was not alerted to the possibility of deception. No differences on attribution of responsibility were observed depending on whether or not the participants were prebriefed about the use of deception.
The effects of the authorized deception approach on study outcomes merit investigation.
A second methodological objection to authorized deception is that it has the potential to reduce the comparability to previous research on placebo mechanisms that did not employ this technique. There is no way to avoid this problem. But to argue that consequently the authorized deception approach should not be adopted would suggest that past ethical lapses justify current ethically deficient practice. Finally, disclosure of the use of deception may lead to reduced participant enrollment, making it more difficult to complete valuable studies and possibly reducing their generalizability. At the extreme, it is possible that too few prospective participants will be willing to volunteer, especially for experiments recruiting patients. One clinical research study using the authorized deception approach (in this case, informing participants that details about the purpose of the research were withheld) found no substantial impact on enrollment [39]. This remains to be studied further. But if this approach reduces participant enrollment, it would indicate that eligible prospective participants do not wish to be deceived, casting doubts on the legitimacy of using deception without disclosing its use.
The results of psychology experiments that alerted participants to the possibility of deception and used prebriefing are encouraging, but may not be generalizable to the situation of patients in clinical research. The null findings obtained by Weiner and Erker and Holmes and Bennet need to be interpreted with caution [35,37]. Given that their study participants were psychology undergraduates, even those who were not prebriefed about the use of deception could have anticipated that they might be deceived [32].
Accordingly, the effects of the authorized deception approach on study outcomes merit investigation with respect to research on the placebo effect in a clinical setting. For example, a methodological experiment comparing the authorized deception approach to the traditional approach that does not reveal the use of deception might be attached to a study using the balanced placebo design to evaluate expectation effects relating to placebo analgesia among patients recovering from surgery. Patients would be randomized to the two methods of disclosure, which would be assessed in terms of their impact on reported pain relief among patients in the various arms of the underlying study. This would allow investigators to examine the extent to which the authorized deception approach biases the study outcomes. It might be desirable to conduct such a methodological experiment in connection with a diversity of underlying studies of the placebo effect and in various patient populations.
We suspect that the use of authorized deception will not bias studies of the placebo effect. Hence, the results of such experiments have the potential to pave the way for important research to proceed that uses the balanced placebo design in the clinical setting along with the authorized deception approach—research that otherwise might be rejected by ethics review committees, owing to concerns about using deception in clinical research. If authorized deception does produce some bias, decisions will have to be made by investigators and ethics review committees about the importance of this bias in compromising the validity of the research compared to the importance of respecting the autonomy of research participants. Conducting studies to estimate the extent of the bias will facilitate and inform these decisions.
If the use of authorized deception proved to produce seriously biased results, then it might be argued that it would be unethical to use the balanced placebo design in clinical research, owing to the extensive deception involved. Nevertheless, some aspects of the role of expectancy in therapeutic responses could still be evaluated in an ethical manner by using nondeceptive research paradigms in clinical settings [20,40,41], such as comparing an open versus closed [10,42] or an open versus double-blind administration of a therapeutic agent [11]. The problem with these experimental paradigms, however, is that because they do not fully manipulate expectancy and pharmacology in a factorial design (as does the balanced placebo design), they do not permit a rigorous evaluation of drug versus expectancy effects and their interaction.
Remaining Qualms about Deceptive Research
Deceptive research involving patients in the clinical setting might be considered unethical even when all pertinent safeguards are in place, including the use of authorized deception. This is because deception, even if authorized in advance, violates the ethical framework of the clinician–patient relationship, which is based on trust. It may be argued that clinician investigators who deceive patients in the course of research are acting fraudulently. Accordingly, professional ethics precludes participation in deceptive research.
This objection, however, confuses the ethics of clinical research with the ethics of medical care [43,44]. Clinical research aims at developing generalizable knowledge in order to improve medical care in the future. Promoting the medical best interests of particular patients is not part of the primary purpose of clinical research. Clinical research also departs from the ethics of medical care in the methods it uses, such as randomization, double-blind procedures, placebo controls, and the justification of risks. Nearly all clinical research, especially research that is not aimed at evaluating the efficacy or safety of treatment interventions, poses risks to participants that are not offset by potential medical benefits. Accordingly, the researcher is not functioning as a therapist in the context of clinical research. It follows that deceptive behavior that would be fraudulent in clinical practice is not necessarily unethical in clinical research. The informed-consent process should clarify that the research in question is different from and outside the purview of medical care. The use of authorized deception in this context makes research involving deception consistent with ethical guidelines appropriate to clinical research.
Experiments investigating the placebo effect evoke legitimate ethical concerns.
This objection cannot be so readily dismissed, however, if the investigator or members of the team of investigators include clinicians who have a prior therapeutic relationship with research participants, as in the experiment described earlier involving patients with irritable bowel syndrome [13]. When investigators simultaneously have both therapeutic and research roles, it is difficult, if not impossible, to avoid the violation of medical ethics constituted by deception, even if adequate safeguards are in place to make the deception justifiable in the context of research. In addition, the potential for negative consequences to patients from deception is likely to be greater in this situation. It is not clear why it would be necessary for a clinician having a prior therapeutic relationship with participants to conduct valuable research on the placebo effect. For example, in the case of Vase et al.'s irritable bowel syndrome experiment, an experienced clinician would be needed to safely administer the rectal distention procedure; however, someone other than the treating physician could be recruited to perform this function.
Conclusion
Research aimed at elucidating the placebo effect promises to produce valuable knowledge concerning the psychological and neurobiological dimensions of healing. Insights gleaned from this research may contribute to the development of clinical interventions that can enhance the therapeutic efficacy of existing treatments. Experiments investigating the placebo effect, however, evoke legitimate ethical concerns, owing to the use of deception.
Key safeguards to assure the ethical design and conduct of deceptive placebo research include (1) prior review and approval by an independent research ethics committee to determine that the use of deception is methodologically necessary and that the study protocol offers sufficient value to justify the risks it poses to participants, including the use of deception; (2) disclosure in the informed-consent document that the study involves the use of deception; and (3) debriefing of participants at the conclusion of research participation. To contribute to public accountability, articles reporting the results of research using deception should describe briefly adherence with these participant-protection guidelines [45,46]. As in all clinical research, an acceptable balance must be struck between promoting valuable knowledge and protecting the rights and well-being of participants.
Supporting Information
Table S1 Clinical Studies on the Placebo Effect Involving Deception
(69 KB DOC).
Click here for additional data file.
We thank Alan Chan, who conducted the literature search for the articles described in Table S1 and extracted the relevant information from the articles.
Citation: Miller FG, Wendler D, Swartzman LC (2005) Deception in research on the placebo effect. PLoS Med 2(9): e262.
==== Refs
References
Kleinman A Guess HA Wilentz JS Guess HA Kleinman A Kusek JW Engel LW An overview 2002 London BMJ 1 32 The science of the placebo: Towards an interdisciplinary research agenda
Harrington A The placebo effect: An interdisciplinary exploration 1997 Cambridge (Massachusetts) Harvard University Press 272
Kirsch I Response expectancy as a determinant of experience and behavior Am Psychol 1985 40 1189 1202
Pollo A Amanzio M Arslanian A Casadio C Maggi G Response expectancies in placebo analgesia and their clinical relevance Pain 2001 93 77 84 11406341
Stewart-Williams S Podd J The placebo effect: Dissolving the expectancy versus conditioning debate Psychol Bull 2004 130 324 340 14979775
Wager TD Rilling JK Smith EE Sokolik A Casey KL Placebo-induced changes in fMRI in the anticipation and experience of pain Science 2004 303 1162 1167 14976306
Mayberg HS Silva JA Brannan SK Tekell JL Mahurin RK The functional neuroanatomy of the placebo effect Am J Psychiatry 2002 159 728 737 11986125
de la Fuente-Fernandez R Schulzer M Stoessl AJ Placebo mechanisms and reward circuitry: Clues from Parkinson's disease Biol Psychiatry 2004 56 67 71 15231437
National Institutes of Health Elucidation of the underlying mechanisms of placebo effect 2001 Bethesda National Institute of Health Available: http://grants1.nih.gov/grants/guide/rfa-files/RFA-AT-02-002.html . Accessed 11 July 2005
Benedetti F Maggi G Lopiano L Lanotte M Rainero I Open versus hidden medical treatments: The patient's knowledge about a therapy affects the therapy outcome Prev Treat 2003 6 article 1. Available: http://journals.apa.org/prevention/volume6/toc-jun-03.html . Accessed 21 July 2005
Benedetti F Pollo A Lopiano L Lanotte M Vighetti S Conscious expectation and unconscious conditioning in analgesic, motor, and hormonal placebo/nocebo responses J Neurosci 2003 23 4315 4323 12764120
Leigh R MacQueen G Tougas G Hargreave FE Bienstock J Change in forced expiratory volume in 1 second after sham bronchoconstrictor in suggestible but not suggestion-resistant asthmatic subjects: A pilot study Psychosom Med 2003 65 791 795 14508022
Vase L Robinson ME Verne GN Price DD The contributions of suggestion, desire, and expectation to placebo effects in irritable bowel syndrome patients An empirical investigation. Pain 2003 105 17 25 14499416
Swartzman LC Burkell J Expectations and the placebo effect in clinical drug trials: Why we should not turn a blind eye to unblinding, and other cautionary notes Clin Pharmacol Ther 1998 64 1 7 9695713
Mitchell SH Laurent CL de Wit H Interaction of expectancy and the phenomenological effects of d-amphetamine: Subjective effects and self-administration Psychopharmacology (Berl) 1996 125 371 378 8826542
McKay D Schare ML The effects of alcohol and alcohol expectancies on subjective reports and physiological reactivity: A meta-analysis Addict Behav 1999 24 633 647 10574301
Kirsch I Weixel LJ Double-blind versus deceptive administration of placebo Behav Neurosci 1988 102 319 323 3365327
Perkins K Sayette M Conklin C Caggiula A Placebo effects of tobacco smoking and other nicotine intake Nicotine Tob Res 2003 5 695 709 14577986
Curran HV Brignell C Fletcher S Middleton P Henry J Cognitive and subjective dose-response effects of acute oral delt 9-tetrahydrocannabinol (THC) in infrequent cannabis users Psychopharmacology (Berl) 2002 164 61 70 12373420
Kirsch I Hidden administration as ethical alternatives to the balanced placebo design 2003 Washington (DC) American Psychological Association Available: http://journals.apa.org/prevention/volume6/pre0060005c.html . Accessed 11 July 2005
Caspi O When are placebo medication side effects due to the placebo phenomenon? JAMA 2002 287 2502
Rohsenow DJ Marlatt GA The balanced placebo design: Methodological considerations Addict Behav 1981 6 107 122 7023202
Milgram S Subject reaction: The neglected factor in the ethics of experimentation Hastings Cent Rep 1977 7 19 23
Vedantum S Human responses to technology scrutinized: Emotional interactions draw interest of psychologists and marketers Washington Post; Sect A: 14 2004 June 7
Wendler D Deception in medical and behavioral research: Is it ever acceptable? Milbank Q 1996 74 87 114 8596525
Wendler D Miller FG Deception in the pursuit of science Arch Intern Med 2004 164 597 600 15037487
Fleming M Bruno M Barry K Fost N Informed consent, deception, and the use of disguised alcohol questionnaires Am J Drug Alcohol Abuse 1989 15 309 319 2763986
Broder A Deception can be acceptable Am Psychol 1998 53 805 806
Kimmel AJ In defense of deception Am Psychol 1988 53 803 805
Ortman A Hertwig R Is deception acceptable? Am Psychol 1997 52 746 747
Ortman A Hertwig R The question remains: Is deception acceptable? Am Psychol 1998 53 806 807
Gallo PS Smith S Mumford S Effects of deceiving subjects on experimental results J Soc Psychol 1973 89 99 107
Aristotle Thomson JAK Tredennick H Barnes J Nichomachean ethics 2004 London Penguin Books 400
American Psychological Association Ethical principles of psychologists and code of conduct Am Psychol 2002 57 1060 1073 Available: http://www.apa.org/ethics/code2002.pdf . Accessed 21 July 2005 12613157
Holmes DS Bennett DH Experiments to answer questions raised by the use of deception in psychological research. I. Role playing as an alternative to deception. II. Effectiveness of debriefing after a deception. III. Effect of informed consent on deception J Pers Soc Psychol 1974 29 358 367 4814125
Bok S Lying: Moral choice in public and private life 1978 New York Random House 368
Wiener RL Erker PV The effects of prebriefing misinformed research participants on their attributions of responsibility J Psychol 1986 120 397 410
Sieber JE Iannuzzo R Rodriguez B Deception methods in psychology: Have they changed in 23 years? Ethics Behav 1995 5 67 85 11654171
Boter H van Delden JJM de Haan RJ Rinkel GJE Patients' evaluation of informed consent to postponed information: Cohort study BMJ 2004 329 86 87 15210571
Colloca L Benedetti F Price DD Bushnell MC The placebo in clinical studies and in medical practice 2004 Seattle IASP Press 187 205 Psychological methods of pain control: Basic science and clinical perspectives
Price D Assessing placebo effects without placebo groups: An untapped possibility? Pain 2001 90 201 203 11207391
Amanzio M Pollo A Maggi G Benedetti F Response variability to analgesics: A role for non-specific activation of endongenous opioids Pain 2001 90 205 215 11207392
Miller FG Rosenstein DL The therapeutic orientation to clinical trials N Engl J Med 2003 348 1383 1386 12672867
Miller FG Research ethics and misguided moral intuition J Law Med Ethics 2004 32 111 116 15152433
Pittinger DJ Deception in research: Distinctions and solutions from the perspective of utilitarianism Ethics Behav 2002 12 117 142 12956136
Miller FG Rosenstein DL Reporting of ethical issues in publications of medical research Lancet 2002 360 1326 1328 12414226
|
16173830
|
PMC1198039
|
CC BY
|
2021-01-05 10:40:31
|
no
|
PLoS Med. 2005 Sep 6; 2(9):e262
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020262
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1613878810.1371/journal.pmed.0020267Policy ForumStatisticsResearch MethodsData Cleaning: Detecting, Diagnosing, and Editing Data Abnormalities Policy ForumVan den Broeck Jan *Argeseanu Cunningham Solveig Eeckels Roger Herbst Kobus Jan Van den Broeck is an epidemiologist, and Kobus Herbst is a public-health physician at the Africa Centre for Health and Population Studies, Mtubatuba, South Africa. Solveig Argeseanu Cunningham is a demographer at the University of Pennsylvania, Philadelphia, Pennsylvania, United States of America. Roger Eeckels is Professor Emeritus of Pediatrics at the Catholic University of Leuven, Leuven, Belgium.
Competing Interests: The authors have declared that no competing interests exist.
*To whom correspondence should be addressed. E-mail: [email protected] 2005 6 9 2005 2 10 e267Copyright: © 2005 Van den Broeck 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.In this policy forum the authors argue that data cleaning is an essential part of the research process, and should be incorporated into study design.
==== Body
In clinical epidemiological research, errors occur in spite of careful study design, conduct, and implementation of error-prevention strategies. Data cleaning intends to identify and correct these errors or at least to minimize their impact on study results. Little guidance is currently available in the peer-reviewed literature on how to set up and carry out cleaning efforts in an efficient and ethical way. With the growing importance of Good Clinical Practice guidelines and regulations, data cleaning and other aspects of data handling will emerge from being mainly gray-literature subjects to being the focus of comparative methodological studies and process evaluations. We present a brief summary of the scattered information, integrated into a conceptual framework aimed at assisting investigators with planning and implementation. We recommend that scientific reports describe data-cleaning methods, error types and rates, error deletion and correction rates, and differences in outcome with and without remaining outliers.
The History of Data Cleaning
With Good Clinical Practice guidelines being adopted and regulated in more and more countries, some important shifts in clinical epidemiological research practice can be expected. One of the expected developments is an increased emphasis on standardization, documentation, and reporting of data handling and data quality. Indeed, in scientific tradition, especially in academia, study validity has been discussed predominantly with regard to study design, general protocol compliance, and the integrity and experience of the investigator. Data handling, although having an equal potential to affect the quality of study results, has received proportionally less attention. As a result, even though the importance of data-handling procedures is being underlined in good clinical practice and data management guidelines [1–3], there are important gaps in knowledge about optimal data-handling methodologies and standards of data quality. The Society for Clinical Data Management, in their guidelines for good clinical data management practices, states: “Regulations and guidelines do not address minimum acceptable data quality levels for clinical trial data. In fact, there is limited published research investigating the distribution or characteristics of clinical trial data errors. Even less published information exists on methods of quantifying data quality” [4].
Data cleaning is emblematic of the historical lower status of data quality issues and has long been viewed as a suspect activity, bordering on data manipulation. Armitage and Berry [5] almost apologized for inserting a short chapter on data editing in their standard textbook on statistics in medical research. Nowadays, whenever discussing data cleaning, it is still felt to be appropriate to start by saying that data cleaning can never be a cure for poor study design or study conduct. Concerns about where to draw the line between data manipulation and responsible data editing are legitimate. Yet all studies, no matter how well designed and implemented, have to deal with errors from various sources and their effects on study results. This problem occurs as much to experimental as to observational research and clinical trials [6,7]. Statistical societies recommend that description of data cleaning be a standard part of reporting statistical methods [8]. Exactly what to report and under what circumstances remains mostly unanswered. In practice, it is rare to find any statements about data-cleaning methods or error rates in medical publications.
Although certain aspects of data cleaning such as statistical outlier detection and handling of missing data have received separate attention [9–18], the data-cleaning process, as a whole, with all its conceptual, organizational, logistical, managerial, and statistical-epidemiological aspects, has not been described or studied comprehensively. In statistical textbooks and non-peer-reviewed literature, there is scattered information, which we summarize in this paper, using the concepts and definitions shown in Box 1.
Box 1. Terms Related to Data Cleaning
Data cleaning: Process of detecting, diagnosing, and editing faulty data.
Data editing: Changing the value of data shown to be incorrect.
Data flow: Passage of recorded information through successive information carriers.
Inlier: Data value falling within the expected range.
Outlier: Data value falling outside the expected range.
Robust estimation: Estimation of statistical parameters, using methods that are less sensitive to the effect of outliers than more conventional methods.
The complete process of quality assurance in research studies includes error prevention, data monitoring, data cleaning, and documentation. There are proposed models that describe total quality assurance as an integrated process [19]. However, we concentrate here on data cleaning and, as a second aim of the paper, separately describe a framework for this process. Our focus is primarily on medical research and on practical relevance for the medical investigator.
Data Cleaning as a Process
Data cleaning deals with data problems once they have occurred. Error-prevention strategies can reduce many problems but cannot eliminate them. We present data cleaning as a three-stage process, involving repeated cycles of screening, diagnosing, and editing of suspected data abnormalities. Figure 1 shows these three steps, which can be initiated at three different stages of a study. Many data errors are detected incidentally during study activities other than data cleaning. However, it is more efficient to detect errors by actively searching for them in a planned way. It is not always immediately clear whether a data point is erroneous. Many times, what is detected is a suspected data point or pattern that needs careful examination. Similarly, missing values require further examination. Missing values may be due to interruptions of the data flow or the unavailability of the target information. Hence, predefined rules for dealing with errors and true missing and extreme values are part of good practice. One can screen for suspect features in survey questionnaires, computer databases, or analysis datasets. In small studies, with the investigator closely involved at all stages, there may be little or no distinction between a database and an analysis dataset.
Figure 1 A Data-Cleaning Framework
(Illustration: Giovanni Maki)
The diagnostic and treatment phases of data cleaning require insight into the sources and types of errors at all stages of the study, during as well as after measurement. The concept of data flow is crucial in this respect. After measurement, research data undergo repeated steps of being entered into information carriers, extracted, transferred to other carriers, edited, selected, transformed, summarized, and presented. It is important to realize that errors can occur at any stage of the data flow, including during data cleaning itself. Table 1 illustrates some of the sources and types of errors possible in a large questionnaire survey. Most problems are due to human error.
Table 1 Issues to Be Considered during Data Collection, Management, and Analysis of a Questionnaire Study
Inaccuracy of a single measurement and data point may be acceptable, and related to the inherent technical error of the measurement instrument. Hence, data cleaning should focus on those errors that are beyond small technical variations and that constitute a major shift within or beyond the population distribution. In turn, data cleaning must be based on knowledge of technical errors and expected ranges of normal values.
Some errors deserve priority, but which ones are most important is highly study-specific. In most clinical epidemiological studies, errors that need to be cleaned, at all costs, include missing sex, sex misspecification, birth date or examination date errors, duplications or merging of records, and biologically impossible results. For example, in nutrition studies, date errors lead to age errors, which in turn lead to errors in weight-for-age scoring and, further, to misclassification of subjects as under- or overweight.
Errors of sex and date are particularly important because they contaminate derived variables. Prioritization is essential if the study is under time pressures or if resources for data cleaning are limited.
Screening Phase
When screening data, it is convenient to distinguish four basic types of oddities: lack or excess of data; outliers, including inconsistencies; strange patterns in (joint) distributions; and unexpected analysis results and other types of inferences and abstractions (Table 1). Screening methods need not only be statistical. Many outliers are detected by perceived nonconformity with prior expectations, based on the investigator's experience, pilot studies, evidence in the literature, or common sense. Detection may even happen during article review or after publication.
What can be done to make screening objective and systematic? To allow the researcher to understand the data better, it should be examined with simple descriptive tools. Standard statistical packages or even spreadsheets make this easy to do [20,21]. For identifying suspect data, one can first predefine expectations about normal ranges, distribution shapes, and strength of relationships [22]. Second, the application of these criteria can be planned beforehand, to be carried out during or shortly after data collection, during data entry, and regularly thereafter. Third, comparison of the data with the screening criteria can be partly automated and lead to flagging of dubious data, patterns, or results.
A special problem is that of erroneous inliers, i.e., data points generated by error but falling within the expected range. Erroneous inliers will often escape detection. Sometimes, inliers are discovered to be suspect if viewed in relation to other variables, using scatter plots, regression analysis, or consistency checks [23]. One can also identify some by examining the history of each data point or by remeasurement, but such examination is rarely feasible. Instead, one can examine and/or remeasure a sample of inliers to estimate an error rate [24]. Useful screening methods are listed in Box 2.
Box 2. Screening Methods
Checking of questionnaires using fixed algorithms.
Validated data entry and double data entry.
Browsing of data tables after sorting.
Printouts of variables not passing range checks and of records not passing consistency checks.
Graphical exploration of distributions: box plots, histograms, and scatter plots.
Plots of repeated measurements on the same individual, e.g., growth curves.
Frequency distributions and cross-tabulations.
Summary statistics.
Statistical outlier detection.
Diagnostic Phase
In this phase, the purpose is to clarify the true nature of the worrisome data points, patterns, and statistics. Possible diagnoses for each data point are as follows: erroneous, true extreme, true normal (i.e, the prior expectation was incorrect), or idiopathic (i.e., no explanation found, but still suspect). Some data points are clearly logically or biologically impossible. Hence, one may predefine not only screening cutoffs as described above (soft cutoffs), but also cutoffs for immediate diagnosis of error (hard cutoffs) [10]. Figure 2 illustrates this method. Sometimes, suspected errors will fall in between the soft and hard cutoffs, and diagnosis will be less straightforward. In these cases, it is necessary to apply a combination of diagnostic procedures.
Figure 2 Areas within the Range of a Continuous Variable Defined by Hard and Soft Cutoffs for Error Screening and Diagnosis, with Recommended Diagnostic Steps for Data Points Falling in Each Area
(Illustration: Giovanni Maki)
One procedure is to go to previous stages of the data flow to see whether a value is consistently the same. This requires access to well-archived and documented data with justifications for any changes made at any stage. A second procedure is to look for information that could confirm the true extreme status of an outlying data point. For example, a very low score for weight-for-age (e.g., −6 Z-scores) might be due to errors in the measurement of age or weight, or the subject may be extremely malnourished, in which case other nutritional variables should also have extremely low values. Individual patients' reports with accumulated information on related measurements are helpful for this purpose. This type of procedure requires insight into the coherence of variables in a biological or statistical sense. Again, such insight is usually available before the study and can be used to plan and program data cleaning. A third procedure is to collect additional information, e.g., question the interviewer/measurer about what may have happened and, if possible, repeat the measurement. Such procedures can only happen if data cleaning starts soon after data collection, and sometimes remeasuring is only valuable very shortly after the initial measurement. In longitudinal studies, variables are often measured at specific ages or follow-up times. With such designs, the possibility of remeasuring or obtaining measurements for missing data will often be limited to predefined allowable intervals around the target times. Such intervals can be set wider if the analysis foresees using age or follow-up time as a continuous variable.
Finding an acceptable value does not always depend on measuring or remeasuring. For some input errors, the correct value is immediately obvious, e.g., if values of infant length are noted under head circumference and vice versa. This example again illustrates the usefulness of the investigator's subject-matter knowledge in the diagnostic phase. Substitute code values for missing data should be corrected before analysis.
During the diagnostic phase, one may have to reconsider prior expectations and/or review quality assurance procedures. The diagnostic phase is labor intensive and the budgetary, logistical, and personnel requirements are typically underestimated or even neglected at the study design stage. How much effort must be spent? Cost-effectiveness studies are needed to answer this question. Costs may be lower if the data-cleaning process is planned and starts early in data collection. Automated query generation and automated comparison of successive datasets can be used to lower costs and speed up the necessary steps.
Treatment Phase
After identification of errors, missing values, and true (extreme or normal) values, the researcher must decide what to do with problematic observations. The options are limited to correcting, deleting, or leaving unchanged. There are some general rules for which option to choose. Impossible values are never left unchanged, but should be corrected if a correct value can be found, otherwise they should be deleted. For biological continuous variables, some within-subject variation and small measurement variation is present in every measurement. If a remeasurement is done very rapidly after the initial one and the two values are close enough to be explained by these small variations alone, accuracy may be enhanced by taking the average of both as the final value.
What should be done with true extreme values and with values that are still suspect after the diagnostic phase? The investigator may wish to further examine the influence of such data points, individually and as a group, on analysis results before deciding whether or not to leave the data unchanged. Statistical methods exist to help evaluate the influence of such data points on regression parameters. Some authors have recommended that true extreme values should always stay in the analysis [25]. In practice, many exceptions are made to that rule. The investigator may not want to consider the effect of true extreme values if they result from an unanticipated extraneous process. This becomes an a posteriori exclusion criterion and the data points should be reported as “excluded from analysis”. Alternatively, it may be that the protocol-prescribed exclusion criteria were inadvertently not applied in some cases [26].
Data cleaning often leads to insight into the nature and severity of error-generating processes. The researcher can then give methodological feedback to operational staff to improve study validity and precision of outcomes. It may be necessary to amend the study protocol, regarding design, timing, observer training, data collection, and quality control procedures. In extreme cases, it may be necessary to restart the study. Programming of data capture, data transformations, and data extractions may need revision, and the analysis strategy should be adapted to include robust estimation or to do separate analyses with and without remaining outliers and/or with and without imputation.
Data Cleaning as a Study- Specific Process
The sensitivity of the chosen statistical analysis method to outlying and missing values can have consequences in terms of the amount of effort the investigator wants to invest to detect and remeasure. It also influences decisions about what to do with remaining outliers (leave unchanged, eliminate, or weight during analysis) and with missing data (impute or not) [27–31]. Study objectives codetermine the required precision of the outcome measures, the error rate that is acceptable, and, therefore, the necessary investment in data cleaning.
Longitudinal studies necessitate checking the temporal consistency of data. Plots of serial individual data such as growth data or repeated measurements of categorical variables often show a recognizable pattern from which a discordant data point clearly stands out. In clinical trials, there may be concerns about investigator bias resulting from the close data inspections that occur during cleaning, so that examination by an independent expert may be needed.
In small studies, a single outlier will have a greater distorting effect on the results. Some screening methods such as examination of data tables will be more effective, whereas others, such as statistical outlier detection, may become less valid with smaller samples. The volume of data will be smaller; hence, the diagnostic phase can be cheaper and the whole procedure more complete. Smaller studies usually involve fewer people, and the steps in the data flow may be fewer and more straightforward, allowing fewer opportunities for errors.
In intervention studies with interim evaluations of safety or efficacy, it is of particular importance to have reliable data available before the evaluations take place. There is a need to initiate and maintain an effective data-cleaning process from the start of the study.
Documentation and Reporting
Good practice guidelines for data management require transparency and proper documentation of all procedures [1–4,30]. Data cleaning, as an essential aspect of quality assurance and a determinant of study validity, should not be an exception. We suggest including a data-cleaning plan in study protocols. This plan should include budget and personnel requirements, prior expectations used to screen suspect data, screening tools, diagnostic procedures used to discern errors from true values, and the decision rules that will be applied in the editing phase. Proper documentation should exist for each data point, including differential flagging of types of suspected features, diagnostic information, and information on type of editing, dates, and personnel involved.
In large studies, data-monitoring and safety committees should receive detailed reports on data cleaning, and procedural feedbacks on study design and conduct should be submitted to a study's steering and ethics committees. Guidelines on statistical reporting of errors and their effect on outcomes in large surveys have been published [31]. We recommend that medical scientific reports include data-cleaning methods. These methods should include error types and rates, at least for the primary outcome variables, with the associated deletion and correction rates, justification for imputations, and differences in outcome with and without remaining outliers [25].
This work was generously supported by the Wellcome Trust (grants 063009/B/00/Z and GR065377).
Citation: Van den Broeck J, Argeseanu Cunningham S, Eeckels R, Herbst K (2005) Data cleaning: Detecting, diagnosing, and editing data abnormalities. PLoS Med 2(10): e267.
==== Refs
References
International Conference on Harmonization Guideline for good clinical practice: ICH harmonized tripartite guideline 1997 Geneva International Conference on Harmonization Available: http://www.ich.org/MediaServer.jser?@_ID=482&@_MODE=GLB . Accessed 29 July 2005
Association for Clinical Data Management ACDM guidelines to facilitate production of a data handling protocol 2003 St. Albans (United Kingdom) Association for Clinical Data Management Available: http://www.acdm.org.uk/files/pubs/DHP%20Guidelines.doc . Accessed 28 July 2005
Food and Drug Administration Guidance for industry: Computerized systems used in clinical trials 1999 Washington (D. C.) Food and Drug Administration Available: http://www.fda.gov/ora/compliance_ref/bimo/ffinalcct.htm . Accessed 28 July 2005
Society for Clinical Data Management Good clinical data management practices, version 3.0 2003 Milwaukee (Wisconsin) Society for Clinical Data Management Available: http://www.scdm.org/GCDMP . Accessed 28 July 2005
Armitage P Berry G Statistical methods in medical research, 2nd ed 1987 Oxford Blackwell Scientific Publications 559
Ki FY Liu JP Wang W Chow SC The impact of outlying subjects on decision of bio-equivalence J Biopharm Stat 1995 5 71 94 7613561
Horn PS Feng L Li Y Pesce AJ Effect of outliers and non-healthy individuals on reference interval estimation Clin Chem 2001 47 2137 2145 11719478
American Statistical Association Ethical guidelines for statistical practice 1999 Alexandria (Virginia) American Statistical Association Available: http://www.amstat.org/profession/index.cfm?fuseaction=ethicalstatistics . Accessed 13 July 2005
Hadi AS Identifying multiple outliers in multivariate data J R Stat Soc Ser B 1992 54 761 771
Altman DG Practical statistics in medical research 1991 London Chapman and Hall 611
Snedecor GW Cochran WG Statistical methods, 7th ed 1980 Ames (Iowa) Iowa State University Press 507
Iglewicz B Hoaglin DC How to detect and handle outliers 1993 Milwaukee (Wisconsion) ASQC Quality Press 87
Hartigan JA Hartigan PM The dip test of unimodality Ann Stat 1985 13 70 84
Welsch RE Launer RL Siegel AF Influence functions and regression diagnostics 1982 New York Academic Press 149 169 Modern data analysis
Haykin S Neural networks: A comprehensive foundation 1994 New York Macmillan College Publishing 696
SAS Institute Enterprise miner, release 4.1 [computer program] 2002 Cary (North Carolina) SAS Institute
Myers RH Classical and modern regression with applications, 2nd ed 1990 Boston PWS-KENT 488
Wainer H Schachts S Gapping Psychometrika 1978 43 203 212
Wang RY A product perspective on total data quality management Commun ACM 1998 41 58 63
Centers for Disease Control and Prevention Epi Info, revision 1st ed. [computer program] 2002 Washington (D. C.) Centers for Disease Control and Prevention Available: http://www.cdc.gov/epiinfo . Accessed 14 July 2005
Lauritsen JM Bruus M Myatt MA EpiData, version 2 [computer program] 2001 Odense (Denmark) Epidata Association Available: http://www.epidata.dk . Accessed 14 July 2005
Bauer UE Johnson TM Editing data: What difference do consistency checks make? Am J Epidemiol 2000 151 921 926 10791565
Winkler WE Problems with inliers 1998 Washington (D. C.) Census Bureau Research Reports Series RR98/05. Available: http://www.census.gov/srd/papers/pdf/rr9805.pdf . Accessed 14 July 2005
West M Winkler RL Database error trapping and prediction J Am Stat Assoc 1991 86 987 996
Gardner MJ Altman DG Statistics with confidence 1994 London BMJ 140
Fergusson D Aaron SD Guyatt G Hebert P Post-randomization exclusions: The intention to treat principle and excluding patients from analysis BMJ 2002 325 652 654 12242181
Allison PD Missing data 2001 Thousand Oaks (California) Sage Publications 93
Twisk J de Vente W Attrition in longitudinal studies: How to deal with missing data J Clin Epidemiol 2002 55 329 337 11927199
Schafer JL Analysis of incomplete multivariate data 1997 London Chapman and Hall 448
South Africans Medical Research Council Guidelines for good practice in the conduct of clinical trials in human participants in South Africa 2000 Pretoria Department of Health 77
Gonzalez ME Ogus JL Shapiro G Tepping BJ Standards for discussion and presentation of errors in survey and census data J Am Stat Assoc 1975 70 6 23
|
16138788
|
PMC1198040
|
CC BY
|
2021-01-05 10:38:15
|
no
|
PLoS Med. 2005 Oct 6; 2(10):e267
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020267
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1612483410.1371/journal.pmed.0020284Research ArticleCancer BiologyImmunologyAllergy/ImmunologyOncologyPathologySurgeryWomen's HealthCancer: BreastImmunology and AllergyOncologyPathologyProfile of Immune Cells in Axillary Lymph Nodes Predicts Disease-Free Survival in Breast Cancer Lymph Node Immune Profile in Breast CancerKohrt Holbrook E
1
Nouri Navid
1
Nowels Kent
2
Johnson Denise
3
Holmes Susan
4
Lee Peter P [email protected] of Medicine, Division of Hematology, Stanford University, Stanford, California, United States of America,2Department of Pathology, Stanford University, Stanford, California, United States of America,3Department of Surgery, Division of Surgical Oncology, Stanford University, Stanford, California, United States of America,4Department of Statistics, Stanford University, Stanford, California, United States of AmericaHoughton Alan Academic EditorSloan-Kettering Cancer InstituteUnited States of America
Competing Interests: The authors have declared that no competing interests exist.
Author Contributions: HEK and PPL designed the study. HEK, NN, KN, DJ, SH, and PPL analyzed the data. HEK and PPL contributed to writing the paper.
9 2005 6 9 2005 2 9 e28418 10 2004 14 7 2005 Copyright: © 2005 Kohrt 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.
Getting More Mileage from Lymph Node Biopsies
Background
While lymph node metastasis is among the strongest predictors of disease-free and overall survival for patients with breast cancer, the immunological nature of tumor-draining lymph nodes is often ignored, and may provide additional prognostic information on clinical outcome.
Methods and Findings
We performed immunohistochemical analysis of 47 sentinel and 104 axillary (nonsentinel) nodes from 77 breast cancer patients with 5 y of follow-up to determine if alterations in CD4, CD8, and CD1a cell populations predict nodal metastasis or disease-free survival. Sentinel and axillary node CD4 and CD8 T cells were decreased in breast cancer patients compared to control nodes. CD1a dendritic cells were also diminished in sentinel and tumor-involved axillary nodes, but increased in tumor-free axillary nodes. Axillary node, but not sentinel node, CD4 T cell and dendritic cell populations were highly correlated with disease-free survival, independent of axillary metastasis. Immune profiling of ALN from a test set of 48 patients, applying CD4 T cell and CD1a dendritic cell population thresholds of CD4 ≥ 7.0% and CD1a ≥ 0.6%, determined from analysis of a learning set of 29 patients, provided significant risk stratification into favorable and unfavorable prognostic groups superior to clinicopathologic characteristics including tumor size, extent or size of nodal metastasis (CD4, p < 0.001 and CD1a, p < 0.001). Moreover, axillary node CD4 T cell and CD1a dendritic cell populations allowed more significant stratification of disease-free survival of patients with T1 (primary tumor size 2 cm or less) and T2 (5 cm or larger) tumors than all other patient characteristics. Finally, sentinel node immune profiles correlated primarily with the presence of infiltrating tumor cells, while axillary node immune profiles appeared largely independent of nodal metastases, raising the possibility that, within axillary lymph nodes, immune profile changes and nodal metastases represent independent processes.
Conclusion
These findings demonstrate that the immune profile of tumor-draining lymph nodes is of novel biologic and clinical importance for patients with early stage breast cancer.
==== Body
Introduction
Lymph node metastasis is well established among the strongest prognostic indicators of clinical outcome for patients with breast cancer [1–3]. The technique of sentinel lymph node (SLN) biopsy has been rapidly adopted over the past decade, as it accurately predicts axillary (nonsentinel) lymph node (ALN) metastasis and therefore identifies women who may be spared the morbidities of axillary dissection [4,5]. With the growing practice of SLN biopsy, new methods of lymph node analysis are being developed [3,6–8]. SLN evaluation by multiple hematoxylin and eosin stained sections (HES), immunohistochemistry (IHC), and most recently, RT-PCR for breast cancer-associated gene expression has increased metastasis detection by up to 42% [7,9]. Despite these technical advances, the prognostic significance of isolated tumor cells and RT-PCR–positive nodes remains inconclusive and highly debated.
Concurrent advances in pathological analysis of primary breast tumors have found infiltrating immune cells of prognostic significance [10,11]. Detailed histological analyses identified tumor-infiltrating T lymphocytes and dendritic cells, with diminished dendritic cell infiltration directly correlated with increased nodal metastasis and poor disease-free and overall survival [10,12–15]. Decreased circulating T lymphocyte populations have also been shown to correlate with poor overall survival [16]. Substantial evidence now exists showing impairment of the systemic and local immune response during breast cancer progression [10–17]. However, it is often overlooked that local tumor-draining nodes are the immunologically active sites where such immune responses, including tumor antigen presentation and lymphocyte activation, should develop. Impairment of the immune response is likely a critical step in lymph node invasion by tumor, and may precede microscopic metastasis detection. Indeed, a limited number of studies suggest that alterations in immune profile, including CD4 helper and CD8 cytotoxic T lymphocytes and CD1a dendritic cell populations, occur within the local nodes of breast cancer patients, although their clinical significance remains unknown [18–20]. Thus, we reasoned that immune profile analysis of tumor-draining nodes may be a more sensitive and earlier method of detecting metastasis, and may provide additional information on clinical outcome.
Materials and Methods
Study patients
Breast cancer patients aged 29–80 years treated at Stanford University Medical Center between February 1997 and January 1999 and found to have tumor-involved SLNs by multilevel HES or IHC were evaluated. Patients who subsequently underwent ALN dissection (ALND), as is standard clinical practice, with clinical outcome data available were selected. SLNs and ALNs were selected based on their designation as sentinel or axillary by the operative report; the majority of SLNs are ALNs based on their location within the breast at time of surgery. ALNs referred to in this study are all ALNs not designated as SLNs. For this reason, reference in this paper to ALNs and nonsentinel lymph nodes are synonymous. In surgical cases involving multiple SLNs and ALNs, one SLN (SLN series 1) and one ALN (ALN series 1) were arbitrarily selected by the Department of Pathology staff and represent the training set (n = 29). The Pathology staff member was blinded to the study design. As no randomization technique was employed, the training set selection process was by definition arbitrary rather than random. To test reliability and variance of immune profile, eight ALNs were selected from a single patient. For purposes of validating the training set, first, a second SLN and ALN were randomly selected (using the random selection function “sample” in R) for each individual within the training set; these represent training set SLN series 2 (n = 18; 11 of 29 patients had only a single SLN removed, which was included in SLN series 1), and training set ALN series 2 (n = 27; an additional ALN could not be retrieved for two of the original 29 patients). Second, a single ALN was randomly selected for all individuals within the test set (n = 48). SLNs and ALNs from patients within training set SLN series 2, training set ALN series 2, and the test set were randomly selected using the sample function in R [21]. As performed in prior studies to provide an average immune profile, ten control nodes—a single mesenteric node per control individual—were similarly examined from patients with benign disease without a history of malignancy or immune disorder [22–24]. All samples were collected from Stanford Department of Pathology Specimen Bank as coded specimens under a protocol approved by the Stanford University Medical Center Institutional Review Board.
All participants were untreated and without a history of cancer or immune disorder prior to breast cancer diagnosis and SLN biopsy. Following surgical management, patients received adjuvant therapy as determined by their medical and radiation oncologists. The duration of disease-free survival (DFS) was the time between initial diagnosis and first recurrence. All patients received SLN and ALN removal in conjunction with removal of primary tumor within 44 d of initial diagnosis. Initial diagnosis was performed by needle aspiration or core biopsy in the majority of cases. Final diagnosis was confirmed from the pathologic evaluation of the primary tumor from the lumpectomy specimen. The average difference between time of diagnosis and surgery was 12.3 d. We chose to use time of diagnosis rather than time of surgery to determine clinical outcome, as we were measuring the relationship between tumor and immune composition of local nodes versus the influence of surgery on outcome. All recurrences were based on documentation of local or systemic disease during a follow-up period of 5 y, after which data were censored. We recorded and verified patient, tumor, and lymph node characteristics [25].
Immunostaining
Tissue sections, 3 μm thick, were cut from formalin-fixed, paraffin-embedded nodes. HES and IHC were performed after antigen retrieval using Biogenex Genomx i1000 (San Ramon, California, United States). Antibodies included anti-CD4 (1/20, Novacastro; Vector Laboratories, Burlingame, California, United States), anti-CD8 (1/25; Dako, Glostrup, Denmark), anti-CD1a (1/100, Dako), anti-AE1/AE3 (1/25, Biogenex), and, as secondary antibody, EnVision dextran kit (1/5, Dako). Optimal concentrations were determined, and tested in sample node sections. Double staining using 3′3′ diaminobenzidene, VIP (Vector Laboratories), and a light counterstain with Mayer's hematoxylin (Innogenex) was performed for lymphocyte populations of interest, with colocalization of tumor cells. Isotype-matched antibodies were used as negative controls. All slides for the respective antibody were stained in the same run.
Presence of metastasis was verified by HES and IHC on four sections per node by two blinded investigators trained in breast cancer pathology. Area of node occupied by each immune cell type and by tumor was determined through computerized image acquisition and analysis software (BLISS; Bacus Laboratories, Lombard, Illinois, United States). Prior image analyses determine cell count and area from an average of five to 20 high-power fields [10,14,15,20,26]. Using BLISS we acquired 160–4,130 sequential images at 200× of the entire lymph node section, which were sequenced together by Metamorph Imaging System (Universal Imaging, Sunnyvale, California, United States). Objectives were calibrated to transform image pixels to microns. Control nodes were examined to standardize thresholds of each stain for cell of interest. Using an automated Metamorph script, standardized thresholds were applied with Metamorph log set to record areas occupied by cell of interest, tumor, and of entire node for all samples, thus minimizing any potential operator bias.
Statistical analysis
Univariate and multivariate analyses including logistic regression tested predictive capacity of patient characteristics. Immune profiles of patients with and without nodal metastasis, and with and without disease recurrence, were compared by Wilcoxon rank sum test. F-test for immune profile equality of variance analysis was used to determine variance between nodes from a single patient versus nodes from different patients with similar characteristics [21]. Variance was also calculated for pairs of ALNs with similar tumor status (either both tumor-free, or both tumor-involved) from the same patient, versus variance for pairs of ALNs with discordant tumor status (one tumor-free and one tumor-involved) from the same patient. ALN series 1 immune profile's sensitivity and specificity in predicting disease recurrence were determined from receiver-operating-characteristic (ROC) curves based on the ALN immune profile of patients with versus without disease recurrence from the training set. ALN series 1 immune profile thresholds were applied to SLN series 2, ALN series 2 and the test set with statistical comparison by X
2 test. We constructed Kaplan-Meier (KM) life-table curves for DFS, with permuted log-rank test comparisons, as the sample size was limited. The training set was stratified for KM curves by ALN series 1 and 2 immune profiles, established from ROC curves applied to ALN series 1, to test prediction of DFS. Nodal thresholds from the training set ALN series 1 were also applied to the test set in KM curves compared by permuted log-rank tests. For analyses involving ALNs from all participants, the only available ALN from the test set was selected. However, as the learning set had two possible ALNs (series 1 and series 2), the sample function in R was used to randomly select one of the two ALNs from the learning set by random number generation. Finally, immune profile and clinicopathologic characteristics significant by univariate analyses among all 77 patients, those with T1 tumors, and/or those with T2 tumors were entered into a Cox proportional hazards model. Two-sided p < 0.05 was considered a statistically significant difference. For analyses we used R statistical package [21,27,28].
Results
Patient, Primary Tumor, and Lymph Node Characteristics
Characteristics of the training set (29 patients) are shown in Table 1. Of 29 SLN metastases in SLN series 1, all were tumor-involved, five contained isolated tumor cells, 11 contained micrometastases, and 13 contained macrometastases. Of 18 SLNs in series 2, nine were tumor-involved, three contained micrometastases, and six contained macrometastases; 16 individuals had positive ALNDs. Of 29 arbitrarily selected series 1 ALNs, nine were found to be tumor-involved, with seven of the 20 tumor-free ALNs selected from patients with positive ALNDs (ALNs other than the one selected for series 1 were found to be tumor-involved) (Figure S1). Of 27 randomly selected series 2 ALNs, seven were tumor-involved (Figure S2). Recurrent disease developed in 11 of 29 patients with 5 y of follow-up; two of 11 recurrences (18%) occurred at a distant site, and ten of 11 developed locoregional relapse (91%), with one patient at time of relapse found to have both local and distant disease.
Table 1 Patient, Primary Tumor, and Lymph Node Characteristics
Table 1 Continued
Test set (48 patients) clinicopathologic characteristics are shown in Table 1. All patients had a tumor-involved SLN biopsy, with four containing isolated tumor cells, 24 containing micrometastases, and 20 containing SLN macrometastases. Recurrent disease developed in 22 (45.8%) of 48 patients during follow-up of 5 y; 14 of 22 occurred at distant sites, seven developed locoregional relapse, and one recurred both at a distant site and locally. ALNs selected from eight (36.3%) of 22 patients with disease recurrence were tumor-involved (Figure S3). Of the 26 ALNs selected from patients without recurrent disease, nine (34.6%) were tumor-involved.
Among all patients from both training set and test set (n = 77), only tumor size significantly correlated with disease recurrence (p = 0.015). Among patients with only T1 tumors (n = 41), percent tumor involvement in the SLN correlated with disease recurrence more closely than all other clinicopathologic characteristics, (p = 0.057). Likewise, among patients with only T2 tumors (n = 33), size of SLN metastasis correlated with disease recurrence more closely than all other clinicopathologic characteristics (p = 0.041).
Alterations in Immune Profile of Tumor-Draining Lymph Nodes
To determine whether tumor-draining lymph nodes from patients with breast cancer are different immunologically than lymph nodes from control individuals, we initially analyzed one SLN and one ALN from each of 29 breast cancer patients (training set, Table 1) by IHC for CD4 T cell, CD8 T cell, and CD1a dendritic cell populations (Figure 1). We found significant differences in CD4 and CD1a populations between SLN, ALN, and control nodes (Figure 1A). While control nodes contained the highest percentages of CD4 and CD8 T cells, ALNs contained the highest percentage of CD1a cells (Figure 1A). The magnitude of CD4 population decrease from control nodes to SLNs was over 10-fold greater than the CD8 decreases between these nodes. SLNs also displayed significant decreases in CD1a cells. Interestingly, CD1a cells were elevated in ALNs even above controls. To determine if tumor invasion is a prerequisite for alterations in immune profile, training set SLNs and ALNs were grouped together as tumor-free or tumor-involved, which revealed dramatic differences in CD4 and CD1a populations and CD4:CD8 ratio based on tumor status (Table 2). Furthermore, training set ALNs (Figure S1) were stratified as tumor-involved (n = 9), tumor-free from an individual with positive ALND (n = 7), or tumor-free from an individual with negative ALND (n = 13). CD4 and CD1a cells were significantly decreased in tumor-involved ALNs (Figure 1E). Intriguingly, CD4 populations were decreased even in tumor-free ALNs (Figure 1E), suggesting that these changes are not merely a reflection of tumor invasion. In contrast, tumor-free ALNs showed significant increases in CD1a cells, which is more dramatic in those from individuals with a positive ALND (Figure 1E). Analysis of percent of node involved by tumor and magnitude of CD4, CD8, or CD1a changes did not show a statistically significant relationship. These observations argue against a simple linear relationship between immune alterations and tumor invasion, but suggest that dynamic changes in the immune profile within tumor-draining lymph nodes may in fact precede tumor invasion.
Figure 1 Lymph Node Profile of Sentinel and Axillary Lymph Nodes
Mean and standard error of CD4 and CD8 T cell, CD1a dendritic cell populations as percent of lymph node, and CD4:CD8 cell ratio are shown for (A) SLN (n = 29), ALN (n = 29), and control lymph nodes (n = 10); (E) tumor-involved ALNs (n = 9), tumor-free ALNs (n = 7) from patients with a positive ALND, tumor-free ALNs from patients with a negative ALND (n = 13), and controls (n = 10); (I) SLNs and ALNs stratified by disease recurrence during 5 y of follow-up (11 of 29 with recurrent disease); (L) tumor-involved ALNs stratified by disease recurrence (n = 9); (M) tumor-free ALNs from patients with a positive ALND stratified by disease recurrence (n = 7); and (N) tumor-free ALNs from patients with a negative ALND stratified by disease recurrence (n = 13). Representative 200× images of lymphocyte population (brown staining) and infiltrating tumor (purple staining) by IHC, including CD8 T cells in (B) SLNs, (C) ALNs, and (D) controls; (F) CD4 T cells in tumor-involved ALNs, (G) tumor-free ALNs from patients with a positive ALND, (H) tumor-free ALNs from patients with a negative ALND; and (J) CD1a dendritic cells in ALNs from patients disease-free versus (K) patients who developed recurrence.
Table 2 Immune Profile and Nodal Status
Relationship between SLN Immune Profile and Axillary Metastasis or DFS
We investigated whether a relationship exists between SLN immune profile and ALN metastasis or DFS. While SLN CD4 populations and CD4:CD8 ratio demonstrated a trend toward an association with axillary metastasis (Table 3), CD8 and CD1a populations showed no such relationship. When SLN immune profile was analyzed for DFS, CD8 populations showed a trend; however, all other cell populations showed no statistically significant relationship with survival (Figure 1I; Table 3).
Table 3 SLN Immune Profile and Clinical Outcome
ALN Immune Profile and Disease-Free Survival
In contrast to SLNs, which exhibited similar immune profile changes in all 29 training set individuals, ALN CD4 and CD1a populations showed significant differences between patients with recurrence versus those disease-free at 5 y (p < 0.001) (Figure 1I and 1K; Table 4). Furthermore, associations between disease recurrence and changes in ALN CD4 and CD1a populations were independent of nodal metastasis or ALND status (Figure 1L–1N). Among patients with disease recurrence, degree of decrease in CD4 T cell and CD1a dendritic cell populations was similar (greater than 4-fold) among tumor-involved ALNs and tumor-free ALNs from either positive or negative ALNDs. These findings support a direct relationship between ALN immune profile and disease-free survival—even within these arbitrarily selected ALNs (series 1), regardless of nodal and locoregional metastasis status.
Table 4 ALN Immune Profile and DFS
To expand on the applicability of these findings, we randomly selected a second ALN from 27 of the 29 individuals in the training set (series 2, Figure S2). Immune profile thresholds determined from ROC curve analysis for maximal predictive accuracy among the training set ALN series 1 were applied to these additional 27 ALNs. Stratification of the training set into favorable and unfavorable prognostic groups for CD4 and CD1a populations was highly significant as displayed in KM curves of DFS (p = 0.005 and p = 0.007, respectively) (Figure 2A; Table 4).
Figure 2 Disease-free Survival Analysis of Women with Breast Cancer According to Immune profile Characteristics, Learning Set ALN Series 2, and Test Set
KM curves are shown for (A) median DFS applied to the learning set ALN series 2 (n = 27) and test set (n = 48) according to size of CD4 T cell and CD1a dendritic cell populations within learning set ALN series 2 (second, randomly selected ALN per individual); (B) DFS stratified by size of CD4 T cell and CD1a dendritic cell populations within test set ALNs; and (C) DFS applied to the learning set (n = 29) and test set (n = 48) according to size of ALN CD4 T cell and ALN CD1a dendritic cell populations. Thresholds for ALN CD4 T cell and ALN CD1a dendritic cell populations were determined by ROC curves as applied to the learning set (ALN series 1). Median duration of DFS are indicated; – indicates a median DFS greater than follow-up period, 5 y. Of 29 individuals in learning set ALN series 1, 11 had recurrent disease, and of 27 individuals in learning set ALN series 2, 11 had recurrent disease. Of 48 individuals in the test set of ALNs, 22 had recurrent disease. For ALN selection from the learning set (C), a single ALN was randomly selected from series 1 or series 2 per individual. Adjusted p-values were determined by the permuted log-rank statistic for comparison of DFS between groups.
Additional comparison of immune profile and patient characteristics within the training set demonstrated ALN CD4 T cell and CD1a dendritic cell populations had superior predictive capacity of DFS (p = 0.001 for both) compared to the degree of tumor involvement in SLNs and ALNs or to primary tumor size, by ROC curve analyses (p = 0.039, p = 0.102, and p = 0.072) (Figure S4). KM curves indicated significant stratification of DFS by percent of tumor involvement in SLN series 1, tumor stage, and ALN CD1a and CD4 populations (Figure 3) (p = 0.043, p = 0.096, p = 0.001, and p = 0.025). Patient stratification by both ALN CD4 T cell population and tumor stage predicted DFS equally as well as, if not better than, the most statistically significant clinicopathologic characteristics (tumor stage and percent of tumor involvement in the SLN) (Figure 3C).
Figure 3 DFS Analysis of Women with Breast Cancer According to Tumor and Immune profile Characteristics, Learning Set ALN Series 1
KM curves are shown for (A) median DFS applied to the learning set, n = 29, according to percent of SLN occupied by infiltrating tumor (determined by IHC), and stratified by tumor stage; (B) DFS according to size of CD4 T cell and CD1a dendritic cell populations within learning set ALN series 1 (first, arbitrarily selected ALN per individual); and (C) DFS stratified both by percent of SLN infiltrated by tumor and tumor stage, and by both axillary node CD4 T cell population and by tumor stage. A comparison of survival by all subgroups and a separate comparison of stratified T2 alone are included (* in [C]). Thresholds for percent tumor infiltration within SLN, ALN CD4 T cell, and ALN CD1a dendritic cell populations were determined by ROC curves as applied to the learning set (SLN and ALN series 1). Median duration of DFS are indicated; – indicates a median DFS greater than follow-up period, 5 y. Of 29 individuals, 11 had recurrent disease. Adjusted p-values were determined by the permuted log-rank statistic for comparison of disease-free survival between groups.
TI, tumor infiltration.
Intra-Individual Versus Inter-Individual Variance in Lymph Node Immune Profile
To more fully address the issue of internodal variance in immune profile from a single individual, we analyzed the immune profiles of eight randomly selected ALNs from a single patient. The variance of these nodes was compared to the variance of nodes from different individuals with similar patient characteristics, including similar recurrent disease state (n = 66). Equality of variance testing illustrated intra-individual homogeneity between nodes relative to inter-individual nodal variance for CD1a, CD4, and CD8 (F[65,7]-statistics of 24.65, 26.89, and 10.23; corresponding significances p < 0.001, p < 0.001, and p = 0.002, respectively).
Validation of the Predictive Capacity of ALN Immune Profile
To further validate the predictive capacity of ALN immune profile for DFS in breast cancer, we analyzed one randomly selected ALN from an additional 48 patients (test set, Table 1), 22 of which developed recurrent disease in 5 y. Thresholds determined by ROC curves from the training set series 1 were applied to the test set data, which demonstrated highly significant stratification of favorable and unfavorable risk of recurrent disease (KM curves of DFS and permuted log-rank tests significant with p < 0.001 for both CD4 and CD1a populations; Figure 2B). Final comparison of the predictive strength of ALN immune profile relative to the most predictive clinicopathologic characteristics was performed for all patients with recurrence status available (single ALN selected randomly from learning set series 1 or series 2, n = 27; and ALN test set, n = 48; total ALNs n = 77; Figure 2C). Of 77 patients analyzed, 33 developed recurrent disease during the follow-up period. Among all patients from both training set and test set, only tumor size significantly correlated with disease recurrence (p = 0.015). KM curves of DFS stratified by ALN CD4 population and ALN CD1a population demonstrate superior risk stratification for recurrence by immune profiling compared to tumor size (p < 0.001, p < 0.001, and p = 0.004, respectively; Figures 2C and 4A).
Figure 4 DFS Analysis of Women with Breast Cancer According to Tumor Stage, T1 and T2, and Immune Profile Characteristics, Learning and Test Sets
KM curves are shown for (A) median DFS applied to the learning set (n = 29) and test set (n = 48) according to tumor stage; (B) DFS stratified by size of ALN CD4 T cell and ALN CD1a dendritic cell populations among individuals with T1 tumors; and (C) DFS stratified by size of ALN CD4 T cell and ALN CD1a dendritic cell populations among individuals with T2 tumors. Thresholds for ALN CD4 T cell and ALN CD1a dendritic cell populations were determined by receiver-operating-characteristic curves as applied to the learning set (ALN series 1). Median duration of DFS are indicated; – indicates a median DFS greater than follow-up period, 5 y. Of 77 individuals, 33 had disease recurrence. Of 41 from individuals with T1 tumors, 15 had recurrent disease. Of 33 individuals with T2 tumors, 15 had disease recurrence. For ALN selection from the learning set, a single ALN was randomly selected from series 1 or series 2 per individual. Adjusted p-values were determined by the permuted log-rank statistic for comparison of DFS between groups.
Strength of ALN Immune Profile as Predictors of DFS in Early Stage Patients (T1 and T2 Tumors)
The predictive value of ALN immune profile was particularly striking in early stage breast cancer patients (with T1 and T2 tumors) (Figure 4). Among the learning set, patients with T2 tumors and ALN CD4 population less than 7.0% had a median duration to recurrence of 9 mo and five-year DFS rate of 0%, versus a median DFS greater than follow-up period of 5 y and DFS rate of 88% for those with T2 tumors and ALN CD4 population of 7.0% or above (p = 0.01) (Figure 4C). By immune profiling of the entire study population (n = 77), median DFS for the unfavorable CD4 and CD1a profiles among 33 patients with T2 tumors were both 24 mo with DFS rates of 13% and 0.0%, respectively. In contrast, favorable ALN CD4 and CD1a profiles portended DFS rate of 94% and 86%, respectively. DFS according to CD4 and CD1a immune profiles was superior to all other clinicopathologic characteristics, the most predictive characteristic being size of SLN metastasis (permuted log-rank test, ALN CD4, p < 0.001; ALN CD1a, p < 0.001; and size of SLN metastasis, p = 0.03). Furthermore, ALN immune profiles of CD4 or CD1a cells were significantly superior to prognostic capacity by amount of local metastatic tumor burden (number of tumor-involved ALNs, p > 0.05) among patients with T2 tumors.
For patients with T1 tumors, we similarly determined the best current clinicopathologic predictor of disease recurrence in 41 patients with T1 tumors among our study population. This characteristic, percent of tumor involvement within the SLN, was an inferior predictor to immune profiling by ALN CD4 and CD1a (permuted log-rank test, percent tumor involvement in SLN, p = 0.049; CD4, p < 0.001; and CD1a, p = 0.001; Figure 4B). By ALN immune profiling among patients with T1 tumors, median DFS for the unfavorable CD4 and CD1a profiles were both 36 mo with DFS rates of 20% and 29%, respectively. Favorable ALN immune profiles portended a significantly more favorable DFS rate of 88% and 81% for CD4 and CD1a among patients with T1 tumors. Thus, for patients with T1 tumors, DFS according to CD4 and CD1a immune profiles was also superior to current clinicopathologic characteristics, including the number of tumor-involved ALNs (p > 0.05).
Relationships between Immune Profile and Metastasis in SLN and ALN
To address potential mechanisms of immune changes in breast cancer-draining lymph nodes, we further explored the dependence of immune profile changes on nodal tumor metastasis in SLNs and ALNs. Immune profile thresholds determined from ROC curve analysis of training set series 1 lymph nodes (CD4 at 7%, CD1a at 0.6%) were applied to SLNs from training set series 2 and ALNs from training set series 2 and the test set. While all of the series 1 SLNs were tumor-involved, only 50% of the series 2 SLNs were involved, making such an analysis possible for both SLN and ALN. Lymph nodes were segregated based on immune profile changes and nodal metastasis (Table 5). Among the 18 SLNs, all nine (100%) tumor-involved SLNs showed decreased percentages of CD4 cells, and 77.8% showed decreased percentages of CD1a cells. Conversely, 81.8% and 77.8% of SLNs with relatively normal percentages of CD4 cells and CD1a cells, respectively, were tumor-free. X
2 testing for CD4 and CD1a, with p-values of less than 0.001 and 0.017, respectively, demonstrate the strength of relationship between tumor involvement and immune profile in SLNs.
Table 5 Sentinel and Axillary Lymph Node Immune Profile and Nodal Metastases
Importantly, ALN analysis of 75 nodes from training set series 2 and the test set, 24 of which were tumor-involved, did not demonstrate a similar effect of nodal tumor status on nodal immune profile (Table 5). Of the 24, 11 (46%) tumor-involved ALNs exhibited preserved CD4 percentages, and 14 (58%) exhibited preserved CD1a percentages. Furthermore, of 51 tumor-free ALNs, 21 (41%) and 23 (45%) exhibited decreased percentages of CD4 or CD1a cells, respectively. Hence, among these ALNs, no statistically significant association was found between decreased CD4 or CD1a populations and nodal tumor involvement (p-values 0.298 and 0.784, respectively). To address the dependence of ALN immune profile on nodal tumor status, we directly compared the immune profiles of series 1 and series 2 ALNs from the same patient. Of 27 paired ALNs, seven were discordant (one tumor-involved and one tumor-free), allowing us to address whether nodal metastasis is the dominant cause of ALN immune profile changes within individuals. Interestingly, the variance between discordant ALN pairs from the same patients was the same or even less than the variance between concordant ALN pairs (both tumor-involved or both tumor-free) (Table 6). This further supports the possibility that ALN immune profile change is driven by a separate process from nodal metastasis.
Table 6 Intra-Individual ALN Immune Profile Variance
Finally, the independent predictors of DFS are shown in Table 7. The most significant independent predictors were percent of CD1a and CD4 cells in the ALN (hazards ratios of 0.42 and 0.93, respectively). Tumor size displayed a trend with recurrence (although not significant at p < 0.05), with a hazards ratio of 1.18. Neither the percent of tumor within the analyzed ALN, nor the size of tumor metastasis within the SLN, were associated with DFS by Cox proportional hazards modeling. These findings point to the intriguing possibility that immune profile changes and nodal metastasis may be independent processes in ALN. This is in contrast to SLN, in which immune profile changes appear dependent on nodal metastasis. Importantly, our data show that ALN immune profile—not SLN immune profile (see Table 3) or ALN metastasis (Table 8)—predicts DFS in breast cancer.
Table 7 Cox Proportional Hazards Model for DFS
Table 8 ALN Immune Profile, Tumor Stage, and DFS
Discussion
It is now widely accepted that the status of tumor-draining lymph nodes significantly predicts clinical outcome in breast cancer. However, current clinical practice involves only histological examination of such nodes for the presence or absence of tumor, largely ignoring the immunological nature of lymph nodes in cancer. As the systemic immune response is clearly influenced by tumor progression, immune profile changes in early sites of immune system-cancer interactions, i.e., tumor-draining nodes, may represent a sensitive indicator of tumor metastasis [10,16,22]. More significantly, the nature of such immunological changes may provide additional biological and prognostic information. In this study, we analyzed the lymph node immune profiles in 77 breast cancer patients with tumor-involved SLNs, 42 of which had tumor-positive ALNDs. Importantly, in 5 y of follow-up, 33 patients had disease recurrence, allowing us to correlate nodal immune profile with clinical outcome. Four patients had SLNs containing isolated tumor cells (0.2 mm or smaller) detected by only IHC—these patients developed disease recurrence, supporting the clinical significance of IHC-only positive SLNs [6,7,29]. As in other studies, mesenteric nodes from patients with benign disease were used as comparisons, since axillary nodes are rarely excised for nonmalignant conditions [22–24]; immune profile of control nodes paralleled literature standards [23,24]. Importantly, new computer-based imaging techniques provided high-resolution image acquisition of the entire nodal surface. We acquired a total of 160–4,130 images (200× magnification) per nodal section, while prior studies based their results on only 5–20 images per section [10,14,15,20,26]. By such detailed, automated analysis of SLNs and ALNs, we identified unique patterns in the degree of CD4 helper T cell, CD8 cytotoxic T cell, and CD1a dendritic cell decreases relative to each other and controls.
An intriguing result from this study is that even tumor-free ALNs exhibited changes in immune profile, with suppression of CD4 and CD8 T cells relative to controls. In contrast, tumor-free ALNs exhibited higher dendritic cell populations than controls, and this elevation was more prominent in tumor-free ALNs from patients with positive ALNDs than from patients with negative ALNDs. This demonstrates that perturbations of the immune profile in tumor-free ALNs are dynamic and may occur before gross nodal metastasis. Our findings extend prior studies in melanoma, lung, head and neck, gastric, and breast cancer, which linked immune down-regulation only to tumor invasion, and also show that the relationship between increasing tumor invasion and changes in immune profile is not a simple linear one, as previously suggested [18,30–33].
While prognostic factors, including lymph node metastasis, tumor size, and histological grade, for breast cancer recurrence and overall survival are well established, few studies have thoroughly examined the influence of immune profile on clinical outcome [10,14,15]. To our knowledge, our findings represent the first demonstration of the clinical significance of T helper and dendritic profiles within tumor-draining nodes of breast cancer patients in predicting DFS. A recent study identified a direct relationship with SLN dendritic cell density and DFS in melanoma [34]. However, we found that the immune profile of SLNs does not display the predictive strength of ALN profiling, but rather reflects largely the metastatic status of the SLN (either tumor-involved or tumor-free). In contrast, the ALN immune profile appears much less influenced by the presence of intranodal metastatic tumor cells. We speculate that as the direct (tumor infiltration) and indirect (altered cytokine profile) effects of cancer progression alter the nodal environment, the predictive capacity of the SLN immune profile becomes diminished, and the influence of infiltrating tumor is augmented. This is analogous to observations in melanoma, in which proximity to primary tumor is the dominant determinant of immune profile [30,35,36]. By profiling ALNs, we observed a predictive accuracy of recurrence by dendritic and T cell populations that is superior even to the predictive accuracy of tumor involvement within the identical node. Furthermore, ALN immune profile predicted recurrence independent of presence or absence of metastasis on ALND. Therefore, a single axillary (nonsentinel) node, selected regardless of tumor involvement within the node or the overall status of all other nodes from the patient's ALND, contains a unique immune profile of potential prognostic value.
In summary, our findings suggest that changes in the immune profile of breast cancer-draining lymph nodes appear to accompany, and may precede, tumor invasion. Perturbation of the SLN immune profile, while highly correlated with the presence of infiltrating metastases, does not add further predictive value in patient prognosis. In contrast, our data show that ALN immune profile does predict DFS much better than it does ALN nodal metastasis. These findings raise the intriguing possibility that two independent processes may be responsible for the immune changes in sentinel versus axillary lymph nodes. The prognostic value of ALNs is highlighted by the capacity of immune profiling of a single, randomly selected ALN to stratify risk of recurrence among early stage breast cancer. Immune profiling of ALN CD4 T cells and CD1a dendritic cells among T1 and T2 tumors dramatically differentiates a population at high risk of recurrence significantly better than all available clinicopathologic patient characteristics. The additional prognostic significance of the immune profile among this subset of breast tumors is not possible by other patient, tumor, or lymph node characteristics. These observations warrant a larger, prospective confirmatory study. Our findings support that a subset of patients may be at higher risk of recurrence due to the extent of immune profile changes, and may therefore justify consideration of more aggressive therapy. Finally, our findings offer possible mechanisms underlying breast cancer's poor immunogenicity, due to either deficient co-stimulation secondary to low helper T cell populations, or inability to activate T cells as a result of down-regulation of antigen-presenting dendritic cells. Strategies to augment T cell and dendritic cell populations and function within tumor-draining nodes may increase the potential for an effective immune response and thus improve clinical outcome among breast cancer patients.
Supporting Information
Figure S1 ALN Status, Learning Set Series 1
ALND was positive in 16 of 29 individuals. Tumor involvement was determined for a single ALN per individual (learning set ALN series 1) (n = 29), and nine ALNs contained tumor infiltration. Of 20 tumor-free ALNs, seven were selected from patients with a positive ALND and 13 from patients with a negative ALND.
(64 KB TIF).
Click here for additional data file.
Figure S2 ALN Status, Learning Set Series 2
ALND was positive in 16 of 29 individuals. Tumor involvement was determined for a single ALN per individual (learning set ALN series 2) (n = 27), and seven ALNs contained tumor infiltration. Of 20 tumor-free ALNs, eight were selected from patients with a positive ALND and 12 from patients with a negative ALND.
(64 KB TIF).
Click here for additional data file.
Figure S3 ALN Status, Test Set
ALND was positive in 31 of 48 individuals. Tumor involvement was determined for a single ALN per individual (test set) (n = 48), and 17 ALNs contained tumor infiltration. Of 31 tumor-free ALNs, 14 were selected from patients with a positive ALND and 17 from patients with a negative ALND.
(64 KB TIF).
Click here for additional data file.
Figure S4 Predictive Strength of Patient and Immune Profile Characteristics, Learning Set
(A) ROC curve calculating the sensitivity and specificity of lymph node CD4 T cell, CD1a dendritic cell, and ratio of CD4:CD8 T cell populations in detecting nodal metastases from the learning set (SLN, n = 29; ALN series 1, n = 29).
(B) ROC curve calculating the sensitivity and specificity of, first, primary tumor size and percent of lymph node occupied by infiltrating tumor, and second, ALN series 1 CD4 T cell, CD1a dendritic cell, and ratio of CD4:CD8 T cell populations in predicting DFS. ALN series 1 represent the first, arbitrarily selected ALN per individual in the learning set. Greater area under the curve indicates greater predictive strength. Adjusted p-values were determined by ROC curve testing for comparison of variable's predictive capacity.
(59 KB TIF).
Click here for additional data file.
Patient Summary
Background
In its earliest stage, breast cancer is confined to the breast itself, but subsequently many cancers spread to other tissues. This often happens through the lymphatic system, a set of canals similar to blood vessels that transport lymph fluid. Lymph nodes are filters along the lymphatic system. The lymph fluid draining away from the breast area is mostly filtered in a set of lymph nodes in the armpit, the so-called axillary lymph nodes. To find out whether a breast cancer has started to spread, doctors routinely check the lymph nodes for breast cancer cells that have escaped from the tumor in the breast. This used to involve surgery to remove many or all of the approximately 30 axillary lymph nodes. Because the surgery can lead to side effects like chronic pain and swelling, doctors have started more recently to first remove the “sentinel” lymph node—the first filter through which the lymph from the tumor tissue drains. In most cases, additional nodes are removed only if this first one is found to contain cancer cells.
Why Was This Study Done?
We know that our immune system can recognize and fight cancer cells. Cancer develops only once the immune system has been compromised, and the actual state of the immune system might tell us something about how easy and quickly the cancer will grow and spread. Because the lymph fluid contains many immune system cells, the researchers thought that (besides looking for cancer cells) it might be worth checking the lymph nodes that are closest to the tumor for immune system activity.
What Did the Researchers Do and Find?
They counted the numbers of different immune system cells in lymph nodes from 77 breast cancer patients. All of the patients had tumor cells in their sentinel lymph nodes, and in 42 patients tumor cells were also found in other axillary lymph nodes. For all patients, the researchers knew whether their cancers came back within five years of removing the lymph nodes. They found that the pattern of immune cells in the sentinel lymph nodes correlated with the presence of cancer cells. In the axillary lymph nodes, however, the decrease in two types of immune cells was correlated with disease-free survival regardless of the presence or absence of tumor cells in these nodes.
What Does This Mean?
This suggests that immune cell characteristics in axillary nodes might provide information about how likely it is that a patient's cancer comes back. These are intriguing but early results that need to be confirmed by new and larger studies before it becomes clear whether regular examination of immune system cells in lymph nodes of breast cancer patients can tell us which cancers are likely to spread and thus should be treated more aggressively.
Where Can I Find More Information Online?
The following Web sites contain information on the role of lymph node dissection and examination in breast cancer.
Breastcancer.org (search for “lymph node removal” and “sentinel lymph node dissection”):
http://www.breastcancer.org
People Living with Cancer (search for “sentinel lymph node biopsy” or “axillary lymph node”):
http://www.plwc.org/
Medicineworld.org (search for “axillary lymph node dissection”):
http://medicineworld.org/
The authors thank the Stanford Histology Research Core for performing slide sectioning and immunohistochemistry, and JB Sneddon and the Brown lab for use of the BLISS imaging system. This work was supported by NIH R01 CA 090809 (PL), the Damon Runyon Cancer Research Foundation (Scholar Award to PL), and the American Cancer Society (Research Scholar Grant to PL). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Citation: Kohrt HE, Nouri N, Nowels K, Johnson D, Holmes S, et al. (2005) Profile of immune cells in axillary lymph nodes predicts disease-free survival in breast cancer. PLoS Med 2(9): e284.
Abbreviations
ALNaxillary (nonsentinel) lymph node
ALNDaxillary lymph node dissection
DFSdisease-free survival
HEShematoxylin and eosin stained sections
IHCimmunohistochemistry
KMKaplan-Meier
ROCreceiver-operating-characteristic
SLNsentinel lymph node
==== Refs
References
Badwe RA Thorat MA Parmar VV Sentinel-node biopsy in breast cancer N Engl J Med 2003 349 1968 1971
Page DL Jensen RA Simpson JF Routinely available indicators of prognosis in breast cancer Breast Cancer Res Treat 1998 51 195 208 10068079
Veronesi U Paganelli G Viale G Luini A Zurrida S A randomized comparison of sentinel-node biopsy with routine axillary dissection in breast cancer N Engl J Med 2003 349 546 553 12904519
Krag D Weaver D Ashikaga T Moffat F Klimberg VS The sentinel node in breast cancer—A multicenter validation study N Engl J Med 1998 339 941 946 9753708
Veronesi U Paganelli G Galimberti V Viale G Zurrida S Sentinel-node biopsy to avoid axillary dissection in breast cancer with clinically negative lymph-nodes Lancet 1997 349 1864 1867 9217757
Moore KH Thaler HT Tan LK Borgen PI Cody HS, 3rd Immunohistochemically detected tumor cells in the sentinel lymph nodes of patients with breast carcinoma: Biologic metastasis or procedural artifact? Cancer 2004 100 929 934 14983487
Dowlatshahi K Fan M Snider HC Habib FA Lymph node micrometastases from breast carcinoma: Reviewing the dilemma Cancer 1997 80 1188 1197 9317169
Pargaonkar AS Beissner RS Snyder S Speights VO Evaluation of immunohistochemistry and multiple-level sectioning in sentinel lymph nodes from patients with breast cancer Arch Pathol Lab Med 2003 127 701 705 12741893
Weigelt B Verduijn P Bosma AJ Rutgers EJ Peterse HL Detection of metastases in sentinel lymph nodes of breast cancer patients by multiple mRNA markers Br J Cancer 2004 90 1531 1537 15083181
Coventry BJ Morton J CD1a-positive infiltrating-dendritic cell density and 5-year survival from human breast cancer Br J Cancer 2003 89 533 538 12888826
Coventry BJ CD1a positive putative tumour infiltrating dendritic cells in human breast cancer Anticancer Res 1999 19 3183 3187 10652609
Georgiannos SN Renaut A Goode AW Sheaff M The immunophenotype and activation status of the lymphocytic infiltrate in human breast cancers, the role of the major histocompatibility complex in cell-mediated immune mechanisms, and their association with prognostic indicators Surgery 2003 134 827 834 14639362
Liyanage UK Moore TT Joo HG Tanaka Y Herrmann V Prevalence of regulatory T cells is increased in peripheral blood and tumor microenvironment of patients with pancreas or breast adenocarcinoma J Immunol 2002 169 2756 2761 12193750
Iwamoto M Shinohara H Miyamoto A Okuzawa M Mabuchi H Prognostic value of tumor-infiltrating dendritic cells expressing CD83 in human breast carcinomas Int J Cancer 2003 104 92 97 12532424
Lespagnard L Gancberg D Rouas G Leclercq G de Saint-Aubain Somerhausen N Tumor-infiltrating dendritic cells in adenocarcinomas of the breast: A study of 143 neoplasms with a correlation to usual prognostic factors and to clinical outcome Int J Cancer 1999 84 309 314 10371352
Blake-Mortimer JS Sephton SE Carlson RW Stites D Spiegel D Cytotoxic T lymphocyte count and survival time in women with metastatic breast cancer Breast J 2004 10 195 199 15125744
McDermott RS Beuvon F Pauly M Pallud C Vincent-Salomon A Tumor antigens and antigen-presenting capacity in breast cancer Pathobiology 2002 70 324 332 12865628
Laguens G Coronato S Laguens R Portiansky E Di Girolamo V Human regional lymph nodes draining cancer exhibit a profound dendritic cell depletion as comparing to those from patients without malignancies Immunol Lett 2002 84 159 162 12413731
Alam SM Clark JS George WD Campbell AM Altered lymphocyte populations in tumour invaded nodes of breast cancer patients Immunol Lett 1993 35 229 234 8390399
Poindexter NJ Sahin A Hunt KK Grimm EA Analysis of dendritic cells in tumor-free and tumor-containing sentinel lymph nodes from patients with breast cancer Breast Cancer Res 2004 6 R408 415 15217509
Dunn O Clark V Basic statistics: A primer for biomedical sciences, 3rd Ed 2000 Indianapolis John Wiley and Sons 231
Heidenreich W Jagla K Schussler J Borner P Dehnhard F Immunological characterization of mononuclear cells in peripheral blood and regional lymph nodes of breast cancer patients Cancer 1979 43 1308 1313 376089
Bryan CF Eastman PJ Conner JB Baier KA Durham JB Clinical utility of a lymph node normal range obtained by flow cytometry Ann N Y Acad Sci 1993 677 404 406 8494226
Vidal-Rubio B Sanchez-Carril M Oliver-Morales J Gonzalez-Femandez A Gambon-Deza F Changes in human lymphocyte subpopulations in tonsils and regional lymph nodes of human head and neck squamous carcinoma compared to control lymph nodes BMC Immunol 2001 2 2 11316463
Singletary SE Greene FL Sobin LH Classification of isolated tumor cells: Clarification of the 6th edition of the American Joint Committee on Cancer Staging Manual Cancer 2003 98 2740 2741 14669301
Zhang L Conejo-Garcia JR Katsaros D Gimotty PA Massobrio M Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer N Engl J Med 2003 348 203 213 12529460
Ihaka R Gentleman R R: A Language for Data Analysis and Graphics J Comp Graph Stat 1996 5 299 314
Maindonald J Braun J Data analysis and graphics using R 2003 Cambridge, United Kingdom Cambridge University Press 400
Dowlatshahi K Fan M Bloom KJ Spitz DJ Patel S Occult metastases in the sentinel lymph nodes of patients with early stage breast carcinoma: A preliminary study Cancer 1999 86 990 996 10491525
Cochran AJ Pihl E Wen DR Hoon DS Korn EL Zoned immune suppression of lymph nodes draining malignant melanoma: Histologic and immunohistologic studies J Natl Cancer Inst 1987 78 399 405 3469453
Battaglia A Ferrandina G Buzzonetti A Malinconico P Legge F Lymphocyte populations in human lymph nodes. Alterations in CD4+ CD25+ T regulatory cell phenotype and T-cell receptor Vbeta repertoire Immunology 2003 110 304 312 14632657
Shinkal H Kitayama J Kimura W Muto T Shibata Y Functional expression of CD11a on CD8+ cells is suppressed in regional lymph nodes with cancer involvement in patients with gastrointestinal carcinoma Cancer 1996 78 1677 1685 8859180
Takenoyama M Yasumoto K Harada M Matsuzaki G Ishida T Expression of activation-related molecules on regional lymph node lymphocytes in human lung cancer Immunobiology 1996 195 140 151 8877391
Cochran AJ Wen DR Huang RR Wang HJ Elashoff R Prediction of metastatic melanoma in nonsentinel nodes and clinical outcome based on the primary melanoma and the sentinel node Mod Pathol 2004 17 747 755 15098011
Cochran AJ Morton DL Stern S Lana AM Essner R Sentinel lymph nodes show profound downregulation of antigen-presenting cells of the paracortex: Implications for tumor biology and treatment Mod Pathol 2001 14 604 608 11406663
Essner R Kojima M Dendritic cell function in sentinel nodes Oncology (Huntingt) 2002 16 28 31
|
16124834
|
PMC1198041
|
CC BY
|
2021-01-05 11:13:39
|
no
|
PLoS Med. 2005 Sep 6; 2(9):e284
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020284
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1613878910.1371/journal.pmed.0020302Neglected DiseasesInfectious DiseasesMicrobiologyOtherPharmacology/Drug DiscoveryEpidemiology/Public HealthHealth PolicyInfectious DiseasesMedicine in Developing CountriesDrugs and adverse drug reactionsPublic HealthInternational healthHealth PolicyMicrobiologyA Breakthrough in R&D for Neglected Diseases: New Ways to Get the Drugs We Need Neglected DiseasesMoran Mary Mary Moran is Director of the Pharmaceutical R&D Policy Project, funded by the Wellcome Trust, at the London School of Economics and Political Science, London, United Kingdom. E-mail: [email protected]
Competing Interests: The author declares that she has no competing interests.
9 2005 8 9 2005 2 9 e302Copyright: © 2005 Mary Moran.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
A New Era of Hope for the World's Most Neglected Diseases
New research suggests that long-held beliefs on neglected disease drug development activity are no longer accurate.
==== Body
Research suggests that long-held beliefs on neglected-disease drug development activity are no longer accurate, and that these inaccurate beliefs have led—and are leading—to poorly designed and targeted government policies and incentives. On a more positive note, this research also highlights opportunities for better-targeted government policies that will more closely match the reality of neglected-disease drug development and the needs of public and industry groups.
Current Perceptions
Current government policies to stimulate development of new drugs for neglected diseases are based on a set of shared understandings.
One of these understandings is that only 13 new drugs have been developed for neglected tropical diseases since 1975, with the main problem being that these diseases are simply non-commercial for companies to invest in [1]. Another is that, although public-private partnerships (PPPs) for drug development have started, they are problematic. (In this article, PPPs are defined as public-health-driven not-for-profit organisations that drive neglected-disease drug development in conjunction with industry groups.) Governments are uncertain which of the plethora of PPPs they should support, particularly as most are thought to be too young to judge their success. At the time this article went to press, the Initiative on Public-Private Partnerships for Health (http://www.ippph.org/) listed 92 PPPs in its database (this number includes all PPP activity, including the small number of organisations that make drugs, vaccines, and microbicides; and one-off partnerings such as donations and cut-price deals).
The general view is that PPPs are inexperienced in drug development, and may eat up public cash without delivering the tools we need, while the real experience and capability in drug development lies with multinational pharmaceutical companies, which must be brought back into the neglected-disease field if we are to achieve success. For example, the Commission for Africa recently stated that we need to increase neglected-disease R&D by “giving large pharmaceutical firms incentives to investigate the diseases that affect Africa, instead of focusing on the diseases of rich countries” [2].
The logical outcome of these beliefs is to focus on new policies to commercialise neglected-disease markets for large companies, which seek peak sales of around $500 million per year to justify investment. This commercialisation approach is generally based on supplementing low developing country purchasing power with large—usually billion-dollar—market “pull” incentives, such as transferable intellectual property rights or advanced purchase commitments underwritten by Western governments.
The above all sounds fairly sensible. The only problem is that most of the above statements no longer hold true in the post-2000 world of neglected-disease drug development, and, as a consequence, government policies based on these beliefs are at risk of missing the mark.
The 2005 Reality
The landscape of neglected-disease drug development has changed dramatically during the past five years (I am not discussing vaccines, only drugs). There were 63 neglected-disease drug projects under way at the end of 2004, including two new drugs in registration stage and 18 new products in clinical trials, half of which were already at Phase III. Assuming there is sufficient funding, at standard attrition rates these projects would be expected to deliver eight to nine new neglected-disease drugs within the next five years, even if no further projects were commenced after this time (Figure 1). This expected yield was calculated using attrition rates from the Tufts Center for the Study of Drug Development [3] combined with PPP-specific rates when available. New projects have been commenced since the end of 2004, and activity is expected to further increase as newer PPPs and institutes get into stride.
Figure 1 Active Neglected-Disease Drug Projects by Institution* (Dec 2004)
DNDi, Drugs for Neglected Diseases Initiative; IOWH, Institute for OneWorld Health; MMV, Medicines for Malaria Venture; ND, neglected diseases.
This increase in activity is not a passing trend, but is a sign of deep-seated structural changes. In particular, it reflects the formation since 2000 of new pharmaceutical industry neglected-disease institutes, now employing over 200 scientists, and the creation of new drug development PPPs, which now conduct three-quarters of all identified neglected-disease drug development, sometimes in partnership with these industry institutes.
This renewed activity—at a level unheard of in the past two decades—commenced largely in the absence of significant new government incentives and generally without public intervention. Eighty percent of PPP drug development activity is funded through private philanthropy, while the industry institutes are largely self-funding, although sometimes with PPP project funding input. (One industry institute did benefit from generous support from the Government of Singapore.)
These findings led the Pharmaceutical R&D Policy Project at the London School of Economics, a project funded by the Wellcome Trust, to a closer examination: where is this new activity centred, what is motivating it, and what has suddenly made this “non-commercial” R&D possible? In line with its mandate, the project also sought to assess the performance of the different actors, in terms of both health outcomes and cost-efficiency. Our findings are set out below.
Neglected-Disease R&D Activity
Multinational drug companies now conduct half of new neglected-disease drug development activity (32 projects), either working with PPPs or working alone (usually with a view to subsequent partnering). In all cases, these companies are working on a non-commercial basis—that is, they are not motivated by commercial returns in neglected-disease markets—and they have agreed to provide the final products to poor patients in developing countries at not-for-profit prices. The bulk of this activity is accounted for by the four companies that have formal neglected-disease divisions—GlaxoSmithKline, Novartis, AstraZeneca, and Sanofi-Aventis—while four other companies have less formal neglected-disease activity, conducting perhaps one or two projects each, and generally on a more serendipitous basis.
A further half of the identified 60-plus neglected-disease drug projects are conducted by smaller-scale commercial firms working with PPPs, including small pharmaceutical companies, contract research organisations and developing country firms, and academic drug developers. The non-academic part of this activity is on a purely commercial basis, with small firms working on a different scale and being motivated by far smaller commercial returns than large multinational pharmaceutical companies. PPPs now spend as much on small company R&D as they do on multinational company projects. The R&D drug landscape for neglected diseases is shown in Figure 2.
Figure 2 The R&D Drug Landscape for Neglected Diseases (Dec 2004)
Three-quarters of all identified R&D (47 projects) was conducted by drug development PPPs, often working with the large and small pharmaceutical companies mentioned above. Nearly one-third of these projects are at the clinical trial stage, including seven drugs now in Phase III trials, and two further products are in the registration stage—rectal artesunate by the Special Programme for Research and Training in Tropical Diseases (TDR) (http://www.who.int/tdr), and paromomycin by the Institute for OneWorld Health (http://www.iowh.org) and TDR. Based on standard attrition rates, these PPP portfolios would be expected to yield six to seven new neglected-disease drugs within five years. This is a high proportion of the total eight to nine new drugs expected from all current avenues.
Once these data became clear, we were faced with the question of why pharmaceutical companies were investing in neglected-disease drug development. Current thinking—and our own initial beliefs—told us this was impossible without new commercial incentives, and no such incentives had been introduced.
Motivations and Facilitating Factors
An examination of the 60-plus neglected-disease projects now under way showed that our initial understanding of company motivations and PPPs' role was incorrect. In particular, commercial incentives were largely irrelevant to the decision by multinational companies to re-enter the neglected-disease field, while small companies involved in neglected-disease R&D were indeed responding to existing commercial drivers. In most cases, the involvement or planned involvement of PPPs was crucial to company activity, commercial or otherwise.
Multinational companies.
Big companies involved in neglected-disease R&D were not motivated by commercial returns in the neglected-disease market, but rather by longer-term business considerations, including: (1) minimising the risk to their reputation stemming from growing public pressure on companies over their failure to address developing country needs; (2) corporate social responsibility and ethical concerns; and (3) strategic considerations (for example, positioning themselves in emerging developing country markets, or building access to low-cost, high-skilled developing country researchers).
This renewed neglected-disease activity has been made possible by a major structural change in the way multinational companies approach neglected-disease R&D. Instead of conducting a limited number of more expensive late-stage drug development projects (the pre-2000 model), companies have moved upstream to the less expensive and more innovative drug discovery stages—allowing them to control costs and resource inputs to levels more acceptable to shareholders. The resulting drug leads can then be developed in conjunction with public partners. These public partners (usually PPPs) facilitate further development by subsidising clinical trial costs; by providing the necessary public health skills and developing country knowledge for clinical trials, registration, and implementation; and by sharing the risk of trials in important but high-liability patient groups, such as children and pregnant women.
This partnering model, sometimes called the “no profit–no loss model”, allows companies to participate in neglected-disease research (often providing substantial in-kind inputs) while still protecting shareholder value, and manufacturing and distributing final products to developing country patients at no mark-up. This has three clear advantages. First, it provides a source of high-quality innovative industry drug leads. Second, it uses the public health sector in its area of maximum comparative advantage (developing country clinical trials rather than drug design). Third, it provides final products to poor patients at not-for-profit prices.
Small companies
Small companies, on the other hand, have largely commercial motivations. Some see neglected-disease markets, particularly larger markets such as tuberculosis (TB) and malaria, as sufficiently attractive to warrant investment and will pursue these even without public support. For example, Zentaris (the small company that developed and registered the new anti-leishmaniasis drug, miltefosine), noted that: “While such a market would be negligible for a big pharmaceutical company, it has a good economic scale for us” (personal communication). A second—and potentially much larger—category is that of small firms that can use “add-on” neglected-disease R&D to promote their Western commercial goals. Examples of such goals are to expand information on their core commercial compounds, or help to establish proof-of-concept for a technology that can subsequently be transferred to commercial markets. These firms generally seek and need substantial PPP support, including full cost coverage and significant skills input—and are unlikely to continue neglected-disease R&D for developing countries if this support is not forthcoming. Finally, commercial contract research organisations increasingly see neglected-disease R&D as an interesting niche sector, and are now involved on a commercial basis in one-third of PPP projects.
While promising, small company neglected-disease activity remains largely under-exploited. Most small companies continue to be deterred by the substantial barriers to entry that large, disseminated, and unfamiliar developing country markets pose, while firms with a primary Western focus can have difficulty concluding financial agreements with cash-strapped PPPs, particularly if their intellectual property concerns are not adequately addressed.
PPPs
As noted above, the presence of PPPs is probably essential to multinational company participation, and plays a catalytic role in encouraging small Western-focussed companies to expand their remit to neglected-disease indications. However, PPPs also serve other useful functions, particularly from the public perspective, including: (1) integrating and coordinating multiple industry and academic partners and contractors along the drug development pipeline; (2) allocating public and philanthropic funds to the “right” kinds of R&D projects from a public health perspective (for example, since 2000, two-thirds of PPP R&D spending has gone directly to industry, almost equally divided between large and small companies, while a further one-third has been allocated to academics to translate basic research into new drug leads); and (3) managing neglected-disease drug portfolios, including selection and termination of projects based on their relative merits.
By virtue of these functions, PPPs are stimulating alternative approaches in addition to “classical” one-to-one partnerships with multinational pharmaceutical companies (these now represent only about one-third of all PPP projects). Their coordinating and integrating role allows PPPs to develop compounds from many different sources even if there is no interested industry partner. For example, the TB Alliance manages and integrates development of PA-824 (a Chiron compound that the company itself did not want to pursue through the full R&D process) on a purely subcontracted basis and without formal development partners.
Alternatively, by pairing up small Western companies or academics with developing country manufacturers, PPPs can—and do—foster a neglected-disease drug development pipeline that is far cheaper than the traditional commercial approach (see “Cost-efficiency” section). Nearly one-quarter of current PPP projects involve developing country pharmaceutical firms as the manufacturing (and sometimes development) partner for a range of compounds from academia, the public domain, or small firms. For example, the Medicine for Malaria Venture's synthetic peroxide project brings together academic discovery and early-development partners (Uni Nebraska, Uni Monash, Swiss Tropical Institute) with Ranbaxy (India) as the development and trial manufacturing partner.
Performance of the Different Approaches
Although increased neglected-disease R&D is always welcome, it is also important that this R&D is targeted to optimal health outcomes for developing country patients, and that it represents the most cost-efficient use of public and philanthropic funding (that is, that patients see maximum health returns for every dollar spent).
In order to assess performance of the various actors, we devised a range of metrics, including health value of the final products (safety, efficacy, suitability, and affordability for developing country patients), level of innovation, capacity, drug development times, cost-efficiency, and cost.
Measurement of the various drug development approaches against these metrics showed that both industry working alone and public groups working alone performed more poorly on most metrics than public–private collaborations. In other words—and perhaps unsurprisingly when we consider the matter more closely—drug development for neglected diseases is optimised by combining industry drug development expertise with public neglected-disease expertise. Below is a summary of outcomes against a sample of metrics.
Health outcomes
The PPP approach delivered the best health outcomes for developing country patients. Eight neglected-disease projects (Artemotil, Paluther, Coartem tablets paediatric label extension, Lapdap, Biltricide, Impavido, Ornidyl, and Mectizan) were conducted in public-industry collaborations (with TDR). Three of the resulting products had a major impact on developing country health—Mectizan (ivermectin), which halved the global burden of onchocerciasis between 1990 and 2000 [4]; praziquantel, which has helped to control schistosomiasis in Brazil, the Mahgreb, the Middle East, China, and the Philippines [5]; and the TDR-assisted label extension of Coartem tablets for paediatric use, which has delivered Africa its first safe, effective, suitable new anti-malarial for many years. We note that praziquantel's impact was greatly increased by the advent of a cheaper, simpler generic formulation by Shin Poong, which allowed broader rollout than the originator product, Biltricide.
By contrast, virtually all of the 13 drugs developed by industry working alone had a low overall health value for developing country patients (although, as noted above, industry has now largely moved to a partnering approach), with only one drug being widely accessible and useful in the developing world (Zentel/albendazole). These 13 drugs were Zentel, Lariam, Malarone, Mycobutin, Paser, AmBisome, Arsumax, Coartem original registration for adults and children above 10 kg in a four-dose (not six-dose) formulation, Halfan, Priftin, Rifampin, Rochagan, and Vansil. The chief obstacles to wider developing country use of industry-alone drugs were their high price—often due to expensive active pharmaceutical ingredients or high formulation cost—and their poor suitability to the target populations. Examples include the development of intravenous formulations suited to Western patients but difficult in poor-country settings, and exclusion of important patient groups, such as children with malaria or HIV-positive patients with TB.
Level of innovation and speed of development
Incremental innovation can offer marked benefits to patients. For instance, fixed-dose combinations of existing drugs can greatly improve ease of use and compliance; follow-on drugs in the same class may improve safety and efficacy; and paediatric formulations can make childhood treatments simpler and more reliable. However, if we are to effectively manage health outcomes in the long-term then we must also overcome drug resistance, which is a growing problem for many neglected diseases, including malaria, TB, leishmaniasis, and sleeping sickness. To do so, we need to focus R&D on “breakthrough” innovation—that is, novel compounds with a novel mechanism of action against parasites and microbes—as well as on activities that simplify or improve existing therapies from the patient perspective. Measurement of the level of “breakthrough” innovation under each approach shows that PPPs and industry partnering approaches perform best. Nearly half of all PPP projects (49%) and more than half of industry partnering projects (63%) are in the “breakthrough” category, compared to only 8% of drugs developed by industry working alone under the pre-2000 model.
The 8% number should not, however, be compared directly with the post-2000 number since the former is based on registered drugs while the latter is for a portfolio of ongoing projects. Given that R&D of “breakthrough” drugs is associated with higher attrition rates, the profile of finished drugs coming out of the post-2000 portfolio is likely to include fewer innovative products. Attrition rates alone, however, cannot account for the much higher share of breakthrough innovation post-2000. The key explanation for this difference is the recent major shift in industry neglected-disease R&D strategy, as noted above, where the serendipitous approach that characterised the past 25 years has given way to one that is specifically focussed toward “breakthrough” innovation. In the long-term, this approach will only deliver high-innovation products, irrespective of attrition rates.
Although the level of innovation is important, it is equally important that innovative R&D projects move quickly to bring new drugs to patients who need them. Time metrics show that PPP drug development trajectories matched or exceeded industry standards (based on data from [3] and [6]). In particular, they were significantly faster than public-alone drug development (see Figures 3 and 4); and they generally exceeded industry trajectories for neglected-disease new chemical entities (although the latter are too few in number to draw significant conclusions).
Figure 3 PPP Timelines
DNDi, Drugs for Neglected Diseases Initiative; MMV, Medicines for Malaria Venture.
Figure 4 Public Timelines
WRAIR, Walter Reed Army Institute of Research.
Cost-efficiency
The overall cost-efficiency of PPPs was superior to other approaches—partly, but not only, due to their ability to leverage substantial in-kind inputs from partners and by the exclusion of costs of capital from the PPP “push” model. The total cost of collective PPP drug development activity from 2000 to 2004 (excluding TDR, for which numbers were not available) was $112 million, including cost of failure for more than 40 projects (ten of these in clinical trials, including four at Phase III). Confidentiality agreements prevent us from disclosing project costs in many cases, in particular when company partnerships are involved. Full per-project costs were more readily disclosed on projects involving academics, developing country firms, and paid sub-contractors, and were remarkably low in most of the cases we examined. For example, Medicines for Malaria Venture's synthetic peroxide project has moved from exploratory, through lead identification, lead optimisation, and pre-clinical, and into Phase I trials at a total cost of $11.5 million. (Costs of completed projects will, of course, be higher.)
The industry cost of developing a new chemical entity for Western markets is substantially higher, estimated by the Tufts Institute at $802 million per drug including cost of failure and cost of capital, and at $403 million for out-of-pocket R&D costs per drug (including cost of failure) [3]. While indicative, these numbers do not hold fully for neglected-disease drug development, which some companies suggested at interview would be substantially cheaper due to lower developing-country trial costs (for example, around $50 million to take a new compound from discovery through to the start of clinical trials). Even using these lower estimates, however, figures to date suggest that PPPs can still be expected to perform significantly better on cost-efficiency and cost.
Overall performance
It is important to note that these outcomes are not evidence of the capacity of the individual players, but rather of the ability of different R&D approaches to deliver optimal outcomes. A company working in a partnership may be able to deliver a better outcome than the same company working alone, for a number of reasons. Companies working alone (as was generally the case under the pre-2000 model) tend to reduce the cost and risk of neglected-disease drug development by focussing on less-expensive, less-risky, “adaptive” R&D such as label extensions of veterinary drugs to humans, or new formulations of existing drugs, and/or by working slowly, as staff and funds are prioritised to more commercial programmes. Under the post-2000 partnering model, the same companies can still restrict costs and risks but in a far more productive way, focussing on discovery of breakthrough leads in the knowledge that others are available to help develop these and deliver them to patients.
What Does This Mean for Government Policies?
There is a clear disjunct between the reality of neglected-disease activity and current government thinking, which is focussed on “commercialising R&D to bring big companies back into the field”. This thinking is built firmly on the beliefs outlined at the start of this report and is now significantly out of kilter with the industry neglected-disease drug landscape.
Two policy issues stand out. Firstly, there is an urgent need to support the new model of neglected-disease drug development, in particular the PPP approach, which is already generating new drugs, is highly cost-effective, appears to offer the highest health value, and is a crucial factor in continuing cost-effective industry involvement in neglected-disease R&D. On this point, we note—and welcome—the recent G8 commitment to “increasing direct investment … through such mechanisms as Public Private Partnerships … to encourage the development of … drugs for AIDS, malaria, TB and other neglected diseases” [7]. We look forward to seeing the shape of new policies and mechanisms to make this commitment concrete, and encourage policy-makers to ensure that these are designed to incentivise optimal practices within the PPP approach, and to do so in the most cost-effective manner. Simply handing over cash may not be the best way.
Secondly, we suggest that policy-makers review their approach to “commercialising” R&D in light of the information above. If big companies tell us that public “commercial” markets are not a catalysing factor in their decision to engage in neglected-disease R&D, then we need to listen carefully to them. Policies to stimulate new multinational company activity are one thing; policies that shift existing industry activity from a not-for-profit approach to a for-profit approach are quite another, and may do so at a potential cost of many billions of dollars across all neglected-disease products. Policy-makers may also want to consider whether commercial incentives should be preferentially targeted toward smaller companies that have a closer fit with commercial neglected-disease markets, and whether these new incentives should be tailored to encourage industry-alone R&D, or to encourage partnered models, which metrics suggest may deliver a better health outcome. The latter is particularly a concern given the relative inexperience of most Western pharmaceutical companies (and in particular small companies) in later-stage clinical development and implementation of tropical disease or TB drugs for use in rural Africa or South Asia, as opposed to their undoubted experience in developing drugs for large-scale US and European disease markets.
Next Steps
The post-2000 renewal of neglected-disease R&D activity is good news for patients with neglected diseases, but it is only a beginning. We hope that this closer analysis will contribute to our store of information, and allow development of policies to encourage and improve these promising new trends in neglected-disease drug development.
A full report of the Pharmaceutical R&D Policy Group study is being published by the Wellcome Trust on its Web site to coincide with the publication of this article.
Citation: Moran M (2005) A breakthrough in R&D for neglected diseases: New ways to get the drugs we need. PLoS Med 2(9): e302.
Abbreviations
PPPpublic-private partnership
TBtuberculosis
TDRSpecial Programme for Research and Training in Tropical Diseases
==== Refs
References
Pecoul B Chirac P Trouiller P Pinel J Access to essential drugs in poor countries: A lost battle? JAMA 1999 281 361 367 9929090
Commission for Africa Final report: “Our common interest” 2005 March 2005. Available: http://www.commissionforafrica.org/english/report/thereport/english/11-03-05_cr_report.pdf . Accessed 29 July 2005
DiMasi J Hansen R Grabowski H The price of innovation: New estimates of drug development costs J Health Econ 2003 22 325 330 12606149
Shibuya K Bernard C Ezzati M Mathers C Global burden of onchocerciasis in the year 2000: Summary of methods and data sources [draft] 2005 Geneva Epidemiology and Burden for Disease (EBD) Global Programme on Evidence for Health Policy (GPE), World Health Organization
Chitsulo L Loverde P Engels D Schistosomiasis Nat Rev Microbiol 2004 2 12 13 15035004
PAREXEL International PAREXEL's Pharmaceutical R&D Statistical Sourcebook 2002/2003 2002 Waltham (Massachusetts) PAREXEL International 378
DFID Highlights of the G8 communiqué on Africa 2005 Available: http://www.number-10.gov.uk/output/Page7880.asp . Accessed 13 July 2005
|
16138789
|
PMC1198042
|
CC BY
|
2021-01-05 10:40:32
|
no
|
PLoS Med. 2005 Sep 8; 2(9):e302
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020302
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 10.1371/journal.pmed.0020314SynopsisMolecular Biology/Structural BiologyRespiratory MedicineInterstitial lung diseaseRespiratory MedicineRole of Osteopontin in Idiopathic Pulmonary Fibrosis Synopsis9 2005 6 9 2005 2 9 e314Copyright: © 2005 Public Library of Science.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.
Up-Regulation and Profibrotic Role of Osteopontin in Human Idiopathic Pulmonary Fibrosis
==== Body
Idiopathic pulmonary fibrosis is a chronic progressive scarring disease of the lung in which gradually the walls of the air sacs of the lungs become replaced by fibrotic tissue. When scarring forms, there is an irreversible loss of the tissue's ability to transfer oxygen into the bloodstream.
The disease affects more than 5 million people worldwide, of whom 40,000 die every year. Misdiagnosis is common because the origin and development of the disease is not completely understood. There are also no effective treatments; drugs to treat lung scarring are still in the experimental phase, and treatments to suppress inflammation have no beneficial effect in most patients.
Although significant advances have been made in the characterization of the clinical features of this disease, the molecular mechanism in humans is still largely unknown. In general, pulmonary fibrosis might be the result of an autoimmune disorder, the after effects of a viral infection, or a genetic condition. Pulmonary fibrosis also occurs after inhalation of some pollutants, in association with diseases such as scleroderma, rheumatoid arthritis, lupus, and sarcoidosis, and after certain medications and therapeutic radiation. However, the etiology of idiopathic pulmonary fibrosis is presently unknown.
Now Annie Pardo and colleagues examine the role of osteopontin, which has diverse functions as a cell-adhesion and migration molecule, in the pathogenesis of idiopathic pulmonary fibrosis. Osteopontin is a multifunctional cytokine that has been implicated in several physiological and pathological processes including bone resorption, malignant transformation, and metastasis. It is also considered a key molecule for regulating inflammation, cellular immune response, and tissue repair, with a unique effect on T cell function.
Osteopontin protein in lung from patient with IPF
Using oligonucleotide microarrays these researchers have previously demonstrated that osteopontin is highly upregulated in bleomycin-induced lung fibrosis in mice—an animal model of pulmonary fibrosis.
In the current study, they used microarrays to analyze gene expression patterns in lung samples (13 samples from people with idiopathic pulmonary fibrosis and 11 from control individuals). They found that osteopontin was the most upregulated gene in the lungs of patients with idiopathic pulmonary fibrosis, and that it was mainly expressed by alveolar epithelial cells.
To better understand the potential local profibrotic effects of osteopontin they then studied its effects on lung fibroblasts and alveolar epithelial cells and found that osteopontin induced a significant increase in migration and proliferation in both fibroblasts and epithelial cells. However, although the effect on fibroblast migration/proliferation was dependent mainly on integrins, in epithelial cells proliferation was mainly dependent on CD44 and migration was dependent on CD44 and integrin signaling.
Osteopontin also showed profibrotic-relevant effects on molecules involved in extracellular matrix remodeling. For example, in fibroblasts osteopontin increased TIMP-1 and type I collagen and inhibited MMP-1 expression, whereas in alveolar epithelial cells it induced MMP-7.
These findings concur with previous studies in experimental tissue fibrosis that have suggested a possible profibrotic role of osteopontin, said the authors. For example in kidney fibrosis, osteopontin enhances macrophage recruitment and stimulates the development of renal scarring after an acute ischemic insult; most importantly, mice that do not express the gene for osteopontin are protected from lung fibrosis induced by the drug bleomycin.
Altogether the results suggest a mechanism to explain most of the profibrotic effects of osteopontin by direct effects on fibroblasts and epithelial cells in the lungs. The findings also suggest that the interaction between MMP-7 and osteopontin might be involved in the progressive nature of the disease. Osteopontin is a potential target for therapeutic intervention in this relentless, incurable disease.
|
0
|
PMC1198043
|
CC0
|
2021-01-05 10:40:31
|
no
|
PLoS Med. 2005 Sep 6; 2(9):e314
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020314
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 1613879010.1371/journal.pmed.0020323EditorialInfectious DiseasesMicrobiologyOtherPharmacology/Drug DiscoveryHealth PolicyInfectious DiseasesMicrobiologyMedicine in Developing CountriesInternational healthHealth PolicyA New Era of Hope for the World's Most Neglected Diseases EditorialThe PLoS Medicine Editors E-mail: [email protected]
Competing Interests: Gavin Yamey, a senior editor at PLoS Medicine, was an initial signatory to the Drugs for Neglected Diseases Initiative appeal.
9 2005 8 9 2005 2 9 e323Copyright: © 2005 Public Library of Science.2005This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
A Breakthrough in R&D for Neglected Diseases: New Ways to Get the Drugs We Need
There are good reasons to be optimistic that tropical infectious diseases can be controlled, argue the PLoS Medicine editors.
==== Body
International health agencies have designated a group of 13 tropical infections as neglected diseases (Text S1). These diseases have much in common: they affect the world's poorest people (and are a cause of poverty); they are disabling, disfiguring, and stigmatizing; there is a shortage of safe and effective treatments; and affected patients have represented the lowest-priority markets for Western pharmaceutical manufacturers. Bernard Pécoul of the Drugs for Neglected Diseases Initiative (DNDi) has called these patients “the forgotten people.”
The conventional wisdom is that the outlook for such patients is hopeless. As one of us has argued, “[They] have no purchasing power, no vocal advocacy group is pleading for their needs, and no strategic interests—military or security—are driving concern about these conditions” (BMJ 325: 176–177). Hence, only 13 new drugs have been developed for neglected tropical diseases since 1975, and they have garnered little international attention compared with “the big three” global health threats: HIV/AIDS, tuberculosis, and malaria.
Fortunately, there is good news to report. The “2005 reality,” argues Mary Moran of the London School of Economics in an article in PLoS Medicine (DOI: 10.1371/journal.pmed.0020302), is that an explosion of public–private partnerships for health has dramatically altered the landscape of neglected-disease drug development. Her analysis shatters a common illusion—that industry is profoundly disinterested in developing drugs for neglected diseases. “Long-held beliefs on neglected-disease drug development activity are no longer accurate,” writes Moran.
According to her analysis, 63 neglected-disease drug projects were under way at the end of 2004. Three-quarters are being conducted by public–private partnerships that often involve multinational or small-scale commercial firms. Twenty new products are already in clinical trials, including half at the phase III or registration stage. Assuming there is sufficient funding, at standard attrition rates these projects would be expected to deliver eight or nine new neglected-disease drugs within the next five years. “The post-2000 renewal of neglected-disease R&D activity is good news for patients with neglected diseases,” writes Moran.
There are even more reasons for optimism. Firstly, many countries affected by neglected diseases, such as Brazil, Egypt, and India, now have the infrastructure to conduct their own neglected-disease research. Morel and colleagues have called these the “innovative developing countries” (Science 309: 401–404), and say they are now reaping the benefits of decades of investment in education, health research infrastructure, and manufacturing capacity. These countries can begin controlling their endemic tropical diseases themselves by developing their own treatments and vaccines with only modest technical or financial assistance from more developed countries.
Secondly, in addition to the expectation that new tools will be developed to control neglected diseases, there is a surge of interest in maximizing the effectiveness of existing tools. This interest focuses on the idea of taking the disparate vertical control programs, each targeting a specific neglected disease, and delivering them in one integrated package. For example, important work is under way to examine the impact of four drugs—albendazole, praziquantel, azithromycin, and ivermectin—in a single delivery mechanism in order to simultaneously target lymphatic filariasis, onchocerciasis, soil-transmitted helminthiases, schistosomiasis, and trachoma (Lancet 365: 1029–1030).
Thirdly, and perhaps most importantly, the neglected diseases have at long last caught the attention of the international development community. The penny has finally dropped: neglected-disease control will be crucially important to achieving many of the Millenium Development Goals, and investment into neglected diseases must become a priority for donors. On June 8, 2005, DNDi launched an appeal, signed by 17 Nobel laureates among others, calling on donors to create a new fund of $3 billion a year for neglected-disease R&D (http://www.researchappeal.org), while the Commission for Africa urged donors to “ensure that there is adequate funding for the treatment and prevention of parasitic diseases.”
It is therefore encouraging that the Gleneagles Communiqué, arising out of this year's G8 summit, specifically called for increased investment to encourage the development of tools for neglected-disease control. Another sign that donors are taking notice is that the US Senate recently passed a foreign operations appropriations bill that included $30 million for neglected diseases. This successful appropriation was in large part due to the advocacy of the Global Health Council, Eric Otteson at Emory University, and Peter Hotez at George Washington University.
There is one element that will be crucial to the success of all of these new initiatives—the ability to share research results and policy discussions freely across the globe, without access barriers. PLoS Biology and PLoS Medicine are committed to publishing the highest-quality basic and clinical research on neglected diseases, and PLoS is exploring the feasibility of launching a new journal devoted specifically to these diseases. The “2005 reality” is that there is an unprecedented window of opportunity for neglected-disease control, and PLoS will play its part.
Supporting Information
Text S1 The 13 Neglected Tropical Diseases
(25 KB DOC).
Click here for additional data file.
Citation:
PLoS Medicine Editors (2005) A new era of hope for the world's most neglected diseases. PLoS Med 2(9): e323.
|
16138790
|
PMC1198044
|
CC BY
|
2021-01-05 10:40:32
|
no
|
PLoS Med. 2005 Sep 8; 2(9):e323
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020323
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 10.1371/journal.pmed.0020324SynopsisCancer BiologyImmunologyAllergy/ImmunologyOncologyPathologyWomen's HealthOncologyCancer: breastImmunology and allergyPathologyWomen's HealthGetting More Mileage from Lymph Node Biopsies Synopsis9 2005 6 9 2005 2 9 e324Copyright: © 2005 Public Library of Science.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.
Profile of Immune Cells in Axillary Lymph Nodes Predicts Disease-Free Survival in Breast Cancer
==== Body
If breast cancer cells escape from the initial tumor site, they often travel to axillary lymph nodes (ALNs) in the armpit. Surgical removal and biopsy of axillary lymph nodes are routinely used to assess the stage of a patient's cancer, and prevent cancers that have spread into the lymph nodes from further metastasis. Because ALN dissections can cause uncomfortable chronic arm swelling, they are often preceded by a biopsy of the sentinel lymph node, the first node into which the tumor tissue drains. If sentinel lymph node biopsy reveals evidence of metastasis, surgeons then remove and dissect the ALNs. Lymph nodes are primarily examined for the presence of metastatic tumor cells, and, more recently, researchers are even searching for the presence of isolated tumor cells and breast cancer–associated gene expression patterns in local nodes.
In light of increasing evidence that the immune system is perturbed both locally at the tumor site and systemically as the cancer progresses, Peter Lee and colleagues set out to study the state of the immune system in draining lymph nodes. They determined profiles of CD4 and CD8 lymphocytes and dendritic cells from sentinel node biopsies and (nonsentinel) axillary lymph node dissections. Their results suggest that these immune profiles harbor independent information about the likelihood of tumor recurrence.
The researchers used automated high-resolution imaging to determine the numbers of CD4 T cells, CD8 T cells, and CD1a-positive dendritic cells in 47 sentinel and 104 axillary nodes from 77 patients with breast cancer. Five-year follow-up data were available for all patients, and 33 patients had disease recurrence within that time. The researchers found that sentinel and axillary nodes from cancer patients (whether the nodes contained tumor cells or were tumor free) contained fewer CD4 and CD8 T lymphocytes than nodes from cancer-free control patients. Dendritic cells were also reduced in tumor-positive nodes, but increased in tumor-free axillary nodes.
Sentinel lymph node section with immune cells and tumor cells
By dividing the patients into a “training set” of 29 individuals and a “test set” of 48 individuals, the researchers could test whether the correlations found in the test set could predict disease-free survival. They found that axillary node CD4 T cell and dendritic cell numbers, regardless of tumor status, were correlated with disease-free survival, but that this was not the case for immune parameters in the sentinel nodes (all of which were tumor-positive). Moreover, in their dataset, the predictive power of the immune parameters in the axillary nodes was better than that of any other characteristics of the patients, including pathological parameters such as tumor size and extent or size of nodal metastases.
These results suggest that, in patients with tumor-positive sentinel nodes, immune profile data from axillary nodes hold additional information on the probability of disease recurrence. As the authors suggest, these results warrant larger prospective studies to test these relationships and explore them in more detail. Another important open question is whether immune profile information from lymph nodes can predict risk of recurrence even in women whose cancers are caught at a stage where they have not yet spread to any lymph nodes.
|
0
|
PMC1198045
|
CC0
|
2021-01-05 10:40:32
|
no
|
PLoS Med. 2005 Sep 6; 2(9):e324
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020324
|
oa_comm
|
==== Front
PLoS MedPLoS MedpmedplosmedPLoS Medicine1549-12771549-1676Public Library of Science San Francisco, USA 10.1371/journal.pmed.0020328SynopsisMedical InformaticsGeneral MedicineDrugs and Adverse Drug ReactionsIs it Possible to Change Prescribing Habits? Synopsis9 2005 6 9 2005 2 9 e328Copyright: © 2005 Public Library of Science.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 Effect of Automated Alerts on Provider Ordering Behavior in an Outpatient Setting
Computerized Physician Order Entry Systems: The Coming of Age for Outpatient Medicine
==== Body
In the US more than 770,000 people are injured or die each year in hospitals from adverse drug events (ADEs), which can cost a hospital, depending on its size, about US$5.6 million every year, excluding ADE-associated costs for malpractice and litigation and the personal costs of injuries to patients. Nationally, hospital expenses to treat patients who have ADEs during hospital admission are enormous: between $1.6 billion and $5.6 billion annually. The cost to patients is also high and not just monetary: those who have an ADE spend on average 8–12 days longer in the hospital than patients who do not have an ADE, and their admission costs $16,000 to $24,000 more.
One way for hospitals to tackle the problem of medication errors is to install computerized monitoring systems, which can reduce ADEs by 28%–95%. Apart from the obvious benefits to patients, these systems can save hospitals as much as $500,000 annually in direct costs. However, despite the potential, fewer than 10% of hospitals have implemented such systems.
Less is known about the value of such systems in an outpatient setting. Now, Andrew Steele and colleagues from Denver have tested a computerized physician order entry (CPOE) system in a US hospital's outpatient clinic. The main purpose of the study was to determine the impact of using computerized alerts to improve the prescribing of medications in the outpatient setting. Studies have shown that 18%–25% of patients might have an ADE in the outpatient environment. This study evaluated a CPOE system alongside an integrated computer-based clinical-decision support system.
It focused on a very specific type of clinical-decision support system: the use of a rules technology to prevent drug–laboratory ADEs. The way the system worked was that providers ordered medications on a computer and an alert was displayed if a relevant drug–laboratory interaction existed.
Comparisons were made between baseline and post-intervention periods. Provider ordering behavior was monitored, focusing on the number of medication orders not completed and the number of rule-associated laboratory test orders initiated after alert display. The investigators found that the rule processed 16,291 times during the study period on all possible medication orders: 7,017 during the pre-intervention period (prescribing doctors did not receive alerts) and 9,274 during the post-intervention period (prescribing doctors received alerts). During the post-intervention period, an alert was displayed for 11.8% (1,093 out of 9,274) of the times the rule processed, with 5.6% of alerts being for “missing laboratory values,” 6.0% for “abnormal rule-associated laboratory” values, and 0.2% for both types of problems.
Providers did pay attention to the alerts; they increased ordering of the rule-associated laboratory test when an alert was displayed (39% at baseline versus 51% post-intervention, p < 0.001), thus showing that the rules had a significant ability to change the ordering behavior of the provider, said the authors. The strongest effect occurred when providers where alerted to “missing” laboratory results (42% increase), the investigators noted. There was less of an effect on ordering behavior when the alert informed the provider of the existence of an abnormal laboratory value (23% increase), which may imply that the cutoff values for the “abnormal” trigger were set too low, suggested the authors. However, there was only a modest effect on halting the ordering of medications, and this was limited to occasions in which the alert presented an abnormal laboratory value in which case there was almost a doubling in order cessation.
There are limitations to the study. For example, the intervention focused on a specific group of drug–laboratory interactions and thus the results may not be generalizable to other types of interventions. In addition, the setting was a single primary-care clinic outpatient setting within a large public-health-integrated health-care delivery system, and results may be different in other settings such as hospitals and private physician offices. However, changing prescriber practice at all is not easy to achieve and this approach thus warrants further research.
|
0
|
PMC1198046
|
CC0
|
2021-01-05 10:40:32
|
no
|
PLoS Med. 2005 Sep 6; 2(9):e328
|
utf-8
|
PLoS Med
| 2,005 |
10.1371/journal.pmed.0020328
|
oa_comm
|
==== Front
AIDS Res TherAIDS Research and Therapy1742-6405BioMed Central London 1742-6405-2-61610917810.1186/1742-6405-2-6ResearchCerebrospinal fluid signs of neuronal damage after antiretroviral treatment interruption in HIV-1 infection Gisslén Magnus [email protected] Lars [email protected] Lars [email protected] Steven G [email protected] Richard W [email protected] Department of Infectious Diseases, Göteborg University, Sahlgrenska University Hospital, Sweden2 Department of Neurology, Göteborg University, Sahlgrenska University Hospital, Sweden3 Department of Medicine, University of California San Francisco, San Francisco General Hospital, CA, USA4 Department of Neurology, University of California San Francisco, San Francisco General Hospital, CA, USA2005 18 8 2005 2 6 6 6 6 2005 18 8 2005 Copyright © 2005 Gisslén et al; licensee BioMed Central Ltd.2005Gisslén 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 neurofilament is a major structural component of myelinated axons. Increased cerebrospinal fluid (CSF) concentrations of the light chain of the neurofilament protein (NFL) can serve as a sensitive indicator of central nervous system (CNS) injury. To assess whether interrupting antiretroviral treatment of HIV infection might have a deleterious effect on the CNS, we measured NFL levels in HIV-infected subjects interrupting therapy.
We identified subjects who had CSF HIV RNA concentrations below 50 copies/mL at the time combination antiretroviral therapy was interrupted, and for whom CSF samples were available before and after the interruption.
Results
A total of 8 subjects were studied. The median (range) CSF NFL level at baseline was <125 (<125–220) ng/L (normal <250 ng/L). All 8 subjects exhibited an increase in CSF and plasma HIV RNA after stopping therapy, accompanied by intrathecal immunoactivation as evidenced by CSF lymphocytic pleocytosis (7/8 patients) and increased CSF neopterin concentration (5/6 patients). Three subjects showed a consistent increase in CSF NFL, rising from <125 ng/L to a maximum of 880 (at day 148), 1,010 (day 58) and 10,930 ng/L (day 101). None exhibited new neurological symptoms or signs, or experienced functional deterioration during the period off treatment; of 5 who underwent brief quantitative neurological testing, none showed worsening performance.
Conclusion
These findings suggest that resurgence of active HIV replication may result in measurable, albeit subclinical, CNS injury. Further studies are needed to define the frequency and pathobiological importance of the increase in CSF NFL.
==== Body
Background
The mortality and morbidity of HIV infection have substantially decreased in the developed world over the past decade, largely due to the introduction of combination antiretroviral therapy (ART) [1]. Widespread use of ART has reduced nearly all of the complications of advanced HIV infection and immunosuppression in regions where therapy is available, including CNS opportunistic infections and the AIDS dementia complex (ADC) [2].
Although effective in suppressing infection, current ART regimens do not eradicate HIV, and prolonged treatment may be complicated by development of drug resistance and an array of side effects [3]. A number of therapeutic strategies have been introduced to prolong the effectiveness of antiretroviral treatment while reducing drug exposure. Although structured (or strategic) treatment interruption (STI) was initially studied as a means of enhancing HIV-specific immunity via an autovaccination phenomena or reducing the levels of drug resistant HIV, its potential therapeutic utility is now being studied primarily as a way to maintain the immunologic and clinical benefit of therapy while reducing drug-toxicity and drug-costs [4]. Despite continued concerns over the safety of interrupting therapy, a large proportion of subjects in clinical practice interrupt therapy for reasons related to drug toxicity, treatment fatigue and/or drug costs.
We undertook the current study to determine whether interrupting therapy can cause deleterious CNS effects that are not appreciated by clinical history and examination. We used archived specimens from study subjects who had interrupted therapy and been followed by repeated lumbar puncture (LP) and CSF analysis before and at various times after remaining off treatment. As a measure of neurological injury, we used concentrations of the light chain of neurofilament protein (NFL). NFL is a major structural component of myelinated axons, and the CSF level can serve as a sensitive marker of axonal damage in a number of conditions [5,6]. Both focal and systemic ischemia cause neuronal damage detectable as a leakage of NFL into the CSF proportional to the severity of the injury [5,7,8]. In chronic disorders the release of NFL is less pronounced but still raised to several times normal in active degeneration of white matter or myelinated spinal tracts, for example in multiple sclerosis and in amyotrophic lateral sclerosis [5,9,10]. Dementia of vascular etiology with subcortical white matter pathology causes moderately increased CSF NFL levels [11,12]. CSF NFL determination has thus been shown to be a versatile tool to detect neuronal damage of any cause, prompting efforts to develop alternate methods of analysis [6,13,14].
In a previous study, increased CSF NFL concentrations were found in 12 of 18 AIDS patients without CNS opportunistic infections, but in only three out of 12 subjects with less advanced, asymptomatic HIV-1 infection [15].
Materials and methods
Study subjects
We retrospectively identified subjects from two centers who underwent serial lumbar punctures before and after interruption of a combination antiretroviral regimen, who had undetectable HIV RNA levels in CSF at the time of the interruption and who had longitudinal samples available for our studies. All subjects participated in separate local protocols evaluating the CSF responses to changes in antiretroviral treatment. Decisions to interrupt treatment were made by the subjects and their primary caregivers independent of the CSF studies. The protocols were approved by University of California San Francisco (UCSF) Committee on Human Research (CHR) and by the Research Ethics Committee of Göteborg University. All subjects provided informed consent.
CSF viral dynamics from some of these subjects have been previously described [16,17].
Laboratory methods
CSF concentrations of NFL were analysed using a previously described ELISA [5]. In brief, capturing antibody (hen anti-NFL IgG) was absorbed to microtest plates, and CSF samples or reference NFL were then incubated. Rabbit anti-NFL was used as secondary antibody. Bound secondary antibody was detected using peroxidase conjugated donkey anti-rabbit IgG. The sensitivity of the assay is 125 ng/L, and the upper normal reference value at the laboratory is 250 ng/L below the age of 60 years.
Cell-free CSF and plasma HIV-1 RNA were quantified with the Amplicor HIV-1 Monitor assay, version 1.5 (Roche Diagnostic System, Hoffman-La Roche, Basel, Switzerland) with a dynamic range down to 50 copies/mL and a detection limit of approximately 20 copies/mL.
Neopterin was measured by a commercially available radio-immunoassay (Henningtest Neopterin, BRAHMS, Berlin, Germany) with a normal reference value of ≤ 4.3 nmol/L in CSF [18].
Routine CSF assessments included CSF white blood cell (WBC) count and peripheral blood CD4+ T-lymphocyte (CD4) cell determination. Albumin was measured using standard Clinical Laboratory methods in each center, and the CSF:serum albumin ratios were calculated [CSF albumin (mg/L)/serum albumin (g/L)] and used as a measure of blood-brain barrier integrity [19]. The five San Francisco patients also underwent standardized neurological performance testing incorporating four tasks (timed gait, grooved pegboard with the dominant hand, finger tapping with the nondominant hand, and the Digit Symbol test of the WAIS-R), yielding an aggregate scaled z-score termed the QNPZ-4 score [20].
Statistics
Unless otherwise indicated, descriptive group statistics in this small study are presented as median values with the range of values, and comparisons between subjects with and without increases in CSF NFL used the Mann-Whitney U test.
Results
Baseline characteristics
Eight subjects, five from San Francisco, California, USA and three from Göteborg, Sweden were included in the study. Demographic and baseline characteristics of the subjects are listed in Table 1. All had CSF HIV RNA concentrations <50 copies/mL and were neurologically normal except for one identified as manifesting AIDS dementia complex (ADC) Stage 0.5 (equivocal/subclinical) disease on treatment. LPs were performed at various intervals before treatment was stopped and one to six (mean 2.9) times during the period off treatment. In two subjects (207 and 223) CSF results were also available from before the initiation of treatment.
Table 1 Baseline characteristics of patients.
Subject ID Age Sex Blood CD4+ CDC Stage ADC Stage HIV-1 RNA Antiretroviral therapies
baseline (nadir) CSF Plasma
(years) (cells/μL) (log10 copies/ml)
San Francisco
6004 56 M 275 (116) B3 0 <1.29 4,11 d4T,3TC,IDV
6008 39 M 834 (360) A2 0 1,61 <1.29 SQV, RTV
6011 47 M 375 (191) B3 0,5 <1.29 1,41 ZDV, 3TC, EFV
6012 54 M 464 (199) B3 0 <1.29 1,57 d4T, 3TC, EFV
6013 44 M 618 (275) C2 0 <1.29 3,69 d4T, 3TC, EFV
Göteborg
160 51 M 544 (195) C3 0 <1.29 <1.29 ZDV, 3TC, IDV
207 47 M 428 (210) A2 0 <1.29 <1.29 ZDV, 3TC, SQV, NFV
223 56 F 365 (110) B3 0 <1.29 <1.29 ZDV, 3TC, IDV/RTV
Changes in CD4-cell count, HIV RNA levels and pleocytosis
The time course of changes in subjects' salient laboratory findings are shown in Figure 1.
Figure 1 Time course of changes in salient laboratory findings in the eight subjects (A-H). Sets of three graphs for each subject arranged within eight panels (the top graph in each panel shows the changes in plasma and CSF HIV RNA and CSF WBCs, the center graph the NFL and QNPZ-4 for those with this assessment, and the bottom graph the blood and CSF neopterin and blood CD4+ T cell counts).
The peripheral blood CD4-cell count decreased from a median of 446 cells/μL (range, 275–834) while on treatment to 258 (208–738) cells/μL during the off-treatment follow-up.
All 8 subjects developed an increase in CSF and plasma HIV RNA concentrations. At baseline, the CSF HIV-1 RNA was <50 (<1.70 log10) copies/mL in all subjects, and the plasma HIV RNA levels was undetectable in six subjects. The median viral load increased to a maximum of 4.36 (3.38–4.87) log10 RNA copies/mL in CSF and to 5.23 (4.78–6.35) log10 copies/mL in plasma during the interruption period. All but one of the subjects developed an increase in CSF WBCs. CSF cell counts increased from a median of 0.5 (range, 0–4) cells/μL on treatment to a maximum of 14 (2–62) cells/μL after stopping therapy.
NFL
Three subjects developed elevations in NFL (panels A, B and C), while the remaining five subjects did not (panels D-H). As shown in the middle panels of each triplet set, all eight subjects had normal CSF NFL concentrations (<250 ng/L) while on treatment, with six of these below the detection limit (125 ng/L). For two subjects (223 and 207, panel series C and G), CSF samples were also available before initiation of antiretroviral treatment, 9–14 months prior to baseline, and CSF NFL concentrations were found to be <125 ng/L at these intervals as well. Treatment interruption resulted in CSF NFL increases in the three subjects (6012, 6011 and 223, panels A-C), rising from baselines at the detection limits to maximum values of 880 ng/L at day 148 (6011), 10,930 ng/L at day 101 (6012) and 1,010 ng/L at day 58 (223). These increases were delayed in relation to virological changes. In subject 223 the only interval testing after treatment stopped was at day 58, while in the two others who had more frequent measurements the earliest documented increase in NFL for subject 6012 was at day 59 and for subject 6011 the inflection point showing the initial increase was delayed to 86 and 99 days.
In the two subjects with increases in NFL and more frequent measurements, the rise in plasma and CSF HIV RNA preceded that of NFL, with the plasma HIV rising by the first sampling at 17 and 15 days and the CSF by days 38 and 36. By contrast, the first change in NFL was delayed by an additional three to seven weeks. The change from baseline to maximum post-STI HIV RNA in blood and CSF was not notably different in the three subjects with increases in NFL compared to those without any change in this neural marker. There was no clear association of NFL elevation with that of CSF WBCs; indeed, the one subject without CSF pleocytosis (6012, panel A) had the greatest increase in CSF NFL.
The subjects were all clinically stable during the period of observation (median 108, range 30–446 days, after treatment interruption). None developed either AIDS-related events or neurological symptoms during the off-treatment follow-up. Likewise, their neurological examinations were unchanged, and in the five tested for quantitative neurological performance (including two with NFL elevations – subjects 6012 and 6011), none showed deterioration in their QNPZ-4 scores (shown in middle graphs of subjects' triplet panels).
Changes in neopterin and its relationship to NFL
In keeping with the inflammatory response in CSF, all subjects in which CSF and serum neopterin were measured exhibited an increase in this marker while off therapy with the exception of subject 160 (panel H in the figure 1). CSF neopterin rose from a median of 6.65 (range, 1.80–16.4) to 23.0 (6.30–90.0) nmol/L in the six subjects analysed. Where the timing could be evaluated, the peak increase in CSF neopterin appeared to be delayed compared to that of CSF HIV, and in subjects 6012 and 6011 occurred near the time of NFL rise. However, based on these limited data, it was not clear that increases in CSF neopterin correlated with increases in NFL. For example, the increase in CSF neopterin in subject 207 (panel G) exceeded that of subjects 6012 and 6011 despite continued undetectable CSF NFL.
The CD4-cell decreases began earlier than the increases in NFL in subjects 6012 and 6011. Again, there was no clear association of development of NFL increase with baseline or nadir CD4 counts or the magnitude of their cell decline. Also, only one subject (223) showed evidence of blood-brain barrier disruption with an elevated CSF:serum albumin ratio. In all others the albumin ratios were stable or showed only minor change that did not parallel changes in CSF NFL (data not shown). As expected within this small dataset, no statistical significant difference could be found between subjects with or without CSF NFL increase, neither regarding the magnitude of increase in CSF neopterin, CSF WBC, CSF nor plasma viral load (data not shown).
Discussion
While the use of STI to stimulate and boost the immune system by re-exposure to viral antigens has not been proved effective and may be complicated by enduring T cell loss and enhanced resistance [4,21], there is still interest in judicious suspension of therapy in order to reduce cost and, more particularly, side effects of therapy. Additionally, patients may stop treatment for personal reasons or because of toxicities. Our observation of increased CSF NFL in three of eight subjects raises a previously unexamined potential concern when interrupting antiretroviral treatment – the possibility of nervous system injury – although further studies are needed to confirm the frequency of this finding and, more particularly, to understand its pathobiological and clinical implications.
The neurofilament is a major structural element of neurons, found most conspicuously in larger neurons and their myelinated axons. It is composed of a triplet protein of which the light subunit (NFL) is an essential component of the neurofilament core. Its main function is to maintain the axonal calibre and it thereby has a crucial role in the structural and functional integrity of axons and in their capacity to rapidly conduct nerve impulses [22]. Increased CSF NFL concentrations are thought to reflect principally injury of myelinated axons, and a clear association has been found between the presence of white matter changes and increased CSF NFL levels in patients with Alzheimer's disease and subcortical vascular dementia [12]. CSF NFL is also increased in several other neurodegenerative disorders linked to demyelination and/or axonal degeneration, including active multiple sclerosis and amyotrophic lateral sclerosis [5,10].
CSF NFL increased substantially in three of our eight subjects studied after stopping antiretroviral therapy, thereby indicating brain injury in this setting. The mechanisms responsible for this rise in NFL remain to be determined. Indeed, from a virological perspective, it is uncertain whether the increase in NFL is simply a complication of established viremia or is related more particularly to the abrupt surge in viral replication that follows treatment interruption. In other words, was this neurological damage due simply to higher levels of HIV replication or due to the "shock" associated with a rapid resurgence after interruption? Perhaps favouring the latter is the finding of elevations in NFL in these subjects without an AIDS diagnosis and CD4 counts above 200 cells/μL, contrasting somewhat with our earlier study that showing CSF NFL elevations chiefly in subjects with AIDS or with CNS opportunistic infections [15]. Normal CSF NFL concentration in one patient measured before treatment initiation gives further support to a rapid viral rebound as the triggering event.
As seen by others [23], plasma viral load increased rapidly and was already detected at the first follow-up (median 18 days) after treatment interruption. Subsequently, CSF viral load increased after a short interval in the subjects with sufficient observations to define these temporal relationships. In a larger experience, that included some of these same subjects, we have noted this delay and also found that an increase in CSF lymphocytosis developed in more than half of subjects [16,17].
The current study also shows that STI leads to increases in CSF and plasma neopterin levels and that these elevations peak later than plasma and CSF HIV RNA levels. Neopterin is an unspecific marker of immune activation that is largely derived from activated macrophages and microglia [24]. CSF neopterin is increased in patients with ADC and has also been found to be a predictive marker of ADC [25]. However, CSF neopterin is also frequently increased in HIV-infected patients without neurological complications, with higher levels found in severely immunocompromised patients than in those with CD4-cell counts above 200 cells/μL [26]. In a previous study, we found an association between increased NFL and CSF neopterin concentrations [15]. This led to our suggesting that immune activation was important in the neuronal injury and release of NFL in this setting. While elevations in CSF neopterin were found in all the subjects with increased NFL and there was the suggestion of temporal association, we also observed subjects with elevated CSF neopterin and normal, unchanging CSF NFL. Hence, further studies are needed to examine this association.
Whatever the underlying mechanism, the presumed axonal damage was subclinical, and no concomitant neurological deterioration was detected during the follow-up period either on the basis of symptoms or on clinical or more formal, though brief, quantitative testing. Based on this initial study, it remains unknown whether this axonal injury has clinical importance. The subject number was small and the follow-up period, for the most part, of limited duration. Hence, longer-term impact on neurological function could not be discerned, though one subject was followed for two years off treatment and did not exhibit neurological change over that time. With one exception (this same subject, 6011, with ADC 0.5 [27]), our subjects were neurologically normal and without prior evidence of ADC. Theoretically, for patients with ADC, resurgence of viremia and CNS infection might be more hazardous, and stopping treatment should be undertaken with particular caution. Whether treatment approaches with repeated cycles of treatment interruptions also can initiate repeated damage to the brain is an open question.
The present study also does not establish whether axonal injury is an acute short-term event after treatment interruption or if it is more chronic. In herpes simplex type 1 (HSV-1) encephalitis, CSF NFL levels increase to a maximum approximately 8–14 days after the onset of neurological symptoms, and only slowly decrease thereafter, with abnormal levels still detected as long as 3–10 months later [28]. Similar observations, with slow normalization of CSF NFL concentrations, have also been made after focal brain ischemia and after acute relapses of multiple sclerosis [10]. Wallerian degeneration, with an anterograde degeneration of axons and disruption of the axonal cytoskeleton, has been suggested as a principal cause of this delayed and long-lasting CSF NFL increase in these settings. The metabolic degradation and half-life of NFL in CSF is also not established.
The temporal delay between viral replication (as evidenced by plasma and more particularly CSF HIV RNA levels) and the onset of axonal disturbance (measured by NFL) may also imply that Wallerian degeneration is involved in STI. If the neuronal soma is the primary site of action, the axonal changes should be a later event. A direct effect on axons in myelinated tracts or spinal roots could also be hypothesised. Alternatively, there may be a lag between HIV replication and onset of neural injury, perhaps involving immunopathological pathways that lag behind the surge in viral replication. The magnitude of the CSF viral load increase, however, did not directly predict the development of elevated CSF NFL; as noted earlier, there were subjects with high CSF viral load without change in NFL levels.
Conclusion
In conclusion, our study raises the question of whether treatment interruption enhances the risk of brain injury. Conversely, it may also suggest that once started, continued treatment might prevent subclinical brain damage. NFL is a sensitive marker of axonal injury which in the present setting seems to disclose subclinical injury. Further studies are needed to define the frequency, duration, magnitude and pathobiological importance of this increase in CSF NFL in HIV infection.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
MG, LR, LH, and RWP contributed to the conception of the study, data interpretation, and writing of the paper. SGD contributed to the establishment of the clinical study and writing the paper.
Acknowledgements
This study was supported by grants from National Institutes of Health (R01 NS043103, R01 NS37660, R01 MH62701, and MO1-RR-00083-36), from the Medical Faculty of Göteborg University (ALFGBG-2874), and from the Research Foundation of Swedish Physicians against AIDS.
==== Refs
Mocroft A Ledergerber B Katlama C Kirk O Reiss P d'Arminio Monforte A Knysz B Dietrich M Phillips AN Lundgren JD Decline in the AIDS and death rates in the EuroSIDA study: an observational study Lancet 2003 362 22 29 12853195 10.1016/S0140-6736(03)13802-0
d'Arminio Monforte A Cinque P Mocroft A Goebel FD Antunes F Katlama C Justesen US Vella S Kirk O Lundgren J Changing incidence of central nervous system diseases in the EuroSIDA cohort Ann Neurol 2004 55 320 328 14991809 10.1002/ana.10827
Carr A Cooper DA Adverse effects of antiretroviral therapy Lancet 2000 356 1423 1430 11052597 10.1016/S0140-6736(00)02854-3
Lori F Lisziewicz J Structured treatment interruptions for the management of HIV infection Jama 2001 286 2981 2987 11743839 10.1001/jama.286.23.2981
Rosengren LE Karlsson JE Karlsson JO Persson LI Wikkelso C Patients with amyotrophic lateral sclerosis and other neurodegenerative diseases have increased levels of neurofilament protein in CSF J Neurochem 1996 67 2013 2018 8863508
Norgren N Rosengren L Stigbrand T Elevated neurofilament levels in neurological diseases Brain Res 2003 987 25 31 14499942 10.1016/S0006-8993(03)03219-0
Blennow M Savman K Ilves P Thoresen M Rosengren L Brain-specific proteins in the cerebrospinal fluid of severely asphyxiated newborn infants Acta Paediatr 2001 90 1171 1175 11697430 10.1080/080352501317061594
Rosen H Karlsson JE Rosengren L CSF levels of neurofilament is a valuable predictor of long-term outcome after cardiac arrest J Neurol Sci 2004 221 19 24 15178208 10.1016/j.jns.2004.03.003
Norgren N Sundstrom P Svenningsson A Rosengren L Stigbrand T Gunnarsson M Neurofilament and glial fibrillary acidic protein in multiple sclerosis Neurology 2004 63 1586 1590 15534240
Malmestrom C Haghighi S Rosengren L Andersen O Lycke J Neurofilament light protein and glial fibrillary acidic protein as biological markers in MS Neurology 2003 61 1720 1725 14694036
Rosengren LE Karlsson JE Sjogren M Blennow K Wallin A Neurofilament protein levels in CSF are increased in dementia Neurology 1999 52 1090 1093 10102440
Sjogren M Blomberg M Jonsson M Wahlund LO Edman A Lind K Rosengren L Blennow K Wallin A Neurofilament protein in cerebrospinal fluid: a marker of white matter changes J Neurosci Res 2001 66 510 516 11746370 10.1002/jnr.1242
Van Geel WJ Rosengren LE Verbeek MM An enzyme immunoassay to quantify neurofilament light chain in cerebrospinal fluid J Immunol Methods 2005 296 179 185 15680162 10.1016/j.jim.2004.11.015
Norgren N Karlsson JE Rosengren L Stigbrand T Monoclonal antibodies selective for low molecular weight neurofilaments Hybrid Hybridomics 2002 21 53 59 11991817 10.1089/15368590252917647
Hagberg L Fuchs D Rosengren L Gisslen M Intrathecal immune activation is associated with cerebrospinal fluid markers of neuronal destruction in AIDS patients J Neuroimmunol 2000 102 51 55 10626666 10.1016/S0165-5728(99)00150-2
Price RW Paxinos EE Grant RM Drews B Nilsson A Hoh R Hellmann NS Petropoulos CJ Deeks SG Cerebrospinal fluid response to structured treatment interruption after virological failure Aids 2001 15 1251 1259 11426069 10.1097/00002030-200107060-00006
Price RW Deeks SG Antiretroviral drug treatment interruption in human immunodeficiency virus-infected adults: Clinical and pathogenetic implications for the central nervous system J Neurovirol 2004 10 Suppl 1 44 51 14982739 10.1080/13550280490268223
Hagberg L Andersson LM Abdulle S Gisslen M Clinical application of cerebrospinal fluid neopterin concentrations in HIV infection Pteridines 2004 15 102 106
Tibbling G Link H Öhman S Principles of albumin and IgG analyses in neurological disorders. I. Establishment of reference values Scand J Clin Lab Invest 1977 37 385 390 337459
Price RW Yiannoutsos CT Clifford DB Zaborski L Tselis A Sidtis JJ Cohen B Hall CD Erice A Henry K Neurological outcomes in late HIV infection: adverse impact of neurological impairment on survival and protective effect of antiviral therapy. AIDS Clinical Trial Group and Neurological AIDS Research Consortium study team Aids 1999 13 1677 1685 10509569 10.1097/00002030-199909100-00011
Lawrence J Mayers DL Hullsiek KH Collins G Abrams DI Reisler RB Crane LR Schmetter BS Dionne TJ Saldanha JM Jones MC Baxter JD Structured treatment interruption in patients with multidrug-resistant human immunodeficiency virus N Engl J Med 2003 349 837 846 12944569 10.1056/NEJMoa035103
Hoffman PN Cleveland DW Griffin JW Landes PW Cowan NJ Price DL Neurofilament gene expression: a major determinant of axonal caliber Proc Natl Acad Sci U S A 1987 84 3472 3476 3472217
Fischer M Hafner R Schneider C Trkola A Joos B Joller H Hirschel B Weber R Gunthard HF HIV RNA in plasma rebounds within days during structured treatment interruptions Aids 2003 17 195 199 12545079 10.1097/00002030-200301240-00009
Huber C Batchelor R Fuchs D Hause A Lang A Niederwieser D Reibnegger G Swetly P Troppmair J Wachter H Immune response-associated production of neopterin release from macrophages primarily under control of interferon-gamma J Exp Med 1984 160 310 316 6429267 10.1084/jem.160.1.310
Brew BJ Dunbar N Pemberton L Kaldor J Predictive markers of AIDS dementia complex: CD4 cell count and cerebrospinal fluid concentrations of beta 2-microglobulin and neopterin J Infect Dis 1996 174 294 298 8699058
Gisslen M Fuchs D Svennerholm B Hagberg L Cerebrospinal fluid viral load, intrathecal immunoactivation, and cerebrospinal fluid monocytic cell count in HIV-1 infection J Acquir Immune Defic Syndr 1999 21 271 276 10428104
Price RW Brew BJ The AIDS dementia complex J Infect Dis 1988 158 1079 1083 3053922
Studahl M Rosengren L Gunther G Hagberg L Difference in pathogenesis between herpes simplex virus type 1 encephalitis and tick-borne encephalitis demonstrated by means of cerebrospinal fluid markers of glial and neuronal destruction J Neurol 2000 247 636 642 11041333 10.1007/s004150070134
|
16109178
|
PMC1198215
|
CC BY
|
2021-01-04 16:38:33
|
no
|
AIDS Res Ther. 2005 Aug 18; 2:6
|
utf-8
|
AIDS Res Ther
| 2,005 |
10.1186/1742-6405-2-6
|
oa_comm
|
==== Front
Ann Clin Microbiol AntimicrobAnnals of Clinical Microbiology and Antimicrobials1476-0711BioMed Central London 1476-0711-4-111609822910.1186/1476-0711-4-11Case ReportAcute cytomegalovirus infection complicated by venous thrombosis: a case report Rovery Clarisse [email protected] Brigitte [email protected] Philippe [email protected] Cédric [email protected] Philippe [email protected] Unité des Rickettsies, UMR6020, Faculté de Médecine, Université de la Méditerranée, Marseille, France2 Service de Maladies infectieuses et de Médecine tropicale, Hôpital Nord, Chemin des Bourrelys, 13915 Marseille Cedex 20, France3 Service de Médecine Interne, Hôpital Nord, Chemin des Bourrelys, 13915 Marseille Cedex 20, France2005 12 8 2005 4 11 11 31 5 2005 12 8 2005 Copyright © 2005 Rovery et al; licensee BioMed Central Ltd.2005Rovery 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
CMV-induced vasculopathy and thrombosis have been reported, but they are rare conditions usually encountered in immunocompromised patients. However more and more complications of CMV infections are recognized in immunocompetent patients.
Case presentation
We present a case report of a previously healthy adult with cytomegalovirus infection that was complicated by tibiopopliteal deep venous thrombosis and in whom Factor V Leiden heterozygous mutation was found.
Conclusion
This new case report emphasizes the involvement of cytomegalovirus in induction of vascular thrombosis in patients with predisposing risk factors for thrombosis. It is necessary to screen for CMV infection in patients with spontaneous thrombosis and an history of fever.
==== Body
Backgound
CMV-induced vasculopathy and thrombosis have been reported, but they are rare conditions. The few published reports on these conditions focus either on immunocompromised transplant recipients who are receiving high-dose immunosuppressive agents [1,2] or on HIV-infected patients [3-5]. We report one new case of acute CMV infection in a non-immunocompromised adult that presented as spontaneous venous thrombosis. There are few reports involving immunocompetent adults, and these reports are additional arguments for the implication of CMV in vasculopathy and thrombosis.
Case presentation
A 33-year-old white man with a 5-day history of fever, was hospitalized on May 2004. He also reported epigastralgia and pain in the right leg. He had an history of femoral deep venous thrombosis ten years before. His mother and his brother had an history of deep venous thrombosis and his father died because of pulmonary embolism. Physical examination revealed fever (temperature range 38°C–38°5C), asthenia, headache, anorexia and epigastralgia. His right leg was enlarged and painful. Tibiopopliteal deep venous thrombosis was confirmed by Doppler ultrasonography. At the time of hospital admission, his WBC was 6,000 cells/mm3, with 2,100 neutrophils/mm3 and 2,820 lymphocytes/mm3, with 9% of hyperbasophilic lymphocytes. His platelet count was 120,000/mm3 and his hemoglobin level was 141 g/l. The C-reactive protein level was 52 mg/l. Liver function tests revealed an ALT level of 65 UI/l, an AST level of 64 UI/l, an LDH level of 891 UI/l, a PAL level of 61 UI/l and a γ-GT level of 41 UI/l. Blood cultures were sterile. The results of serological tests for HIV ELISA, hepatitis A IgM, hepatitis B surface antigen, hepatitis C virus, Q fever, and toxoplasma were negative. VCA and EBNA IgG antibodies were positive suggesting past immunization. Serological test for CMV ELISA was strongly positive for IgM antibodies. The result of a CMV pp65 antigenemia assay, based on the direct detection of the CMV pp65 phosphoprotein was positive. A second serological test for CMV taken two weeks after the first one showed seroconversion with appearance of IgG. No pathological values for prothrombin time ratio, activated partial thromboplastin time, plasma antithrombin III, protein C and S activity were found. Results of tests for anticardiolipin antibodies, lupus anticoagulant and prothrombin 20210 were negative. Antinuclear antibodies were negative and complement was normal. Factor V Leiden heterozygous mutation was found. Anticoagulant treatment was introduced. Tthe patient became asymptomatic and was discharged from the hospital.
Discussion
Our observation, as well those of others [6,7], argues for implication of CMV in vasculopathy and thrombosis. In these cases, thrombosis occurred in immunocompetent patients during CMV infection, which was confirmed, in most cases, by pp65 antigen and/or CMV viremia/viruria. There are few reports involving immunocompetent adults since we retrieved only 13 reports through MEDLINE database [6-16]. Predisposing risk factors for thrombosis have sometimes been found in the reported cases such as protein C and S deficiency [15] and the presence of antiphospholipid antibodies [10,13,16,17]. In other reports, patients were women and had risk factors such as oral contraceptive use [8,9,14]. The factor V Leiden heterozygous mutation was only reported once previously [6]. However in some other reported cases, no hemostatic abnormalities can be found [6,7,17] and spontaneous resolutions of thrombosis were observed in some of these patients who did not receive anticoagulant treatment [9]. This suggests that CMV was the sole triggering factor. These observations suggest that acute CMV infection may be the cause of the thrombosis event in these patients. However, it is difficult to determine whether CMV is a direct cause of thrombosis or a precipitating factor in patients with underlying thrombogenic tendency. Considering that CMV could be a transient triggering factor for thrombosis in our patient, we decided not to treat lifelong as recommended for patients with factor V Leiden heterozygous mutation and 2 spontaneous thrombosis [18] but only for 6 months [19].
In conclusion, our case report reinforces the opinion that CMV is a rare but potentially signifiant cause or precipitating factor of arterial and venous thrombosis in immunocompetent hosts [20]. It seems to be necessary to screen for CMV infection in patients with spontaneous thrombosis and an history of fever. It is important to recognize CMV infection in these patient as it may be possible to consider the discontinuation of anticoagulant therapy earlier.
Acknowledgements
We thank Ann Kasmar for reviewing the manuscript
==== Refs
Koskinen PK Nieminen MS Krogerus LA Lemstrom KB Mattila SP Hayry PJ Lautenschlager IT Cytomegalovirus infection and accelerated cardiac allograft vasculopathy in human cardiac allografts J Heart Lung Transplant 1993 12 724 29 8241209
Madalasso C de Souza NFJ Ilstrup DM Wiesner RH Rom RA Cytomegalovirus and its association with hepatic artery thrombosis after liver transplantation Transplantation 1998 66 294 97 9721795 10.1097/00007890-199808150-00003
Bayer DD Sorbello AF Condolucci DV Bilateral subclavian vein thrombosis in a patient with acquired immunodeficiency syndrome J Am Osteopath Assoc 1995 95 276 77 7744629
Jenkins PE Peters BS Pinching AJ Thromboembolic disease in AIDS is associated with cytomegalvirus disease AIDS 1991 5 1540 1542 1667576
Sullivan PS Dworkin MS Jones JL Hooper WC Epidemiology of thrombosis in HIV-infected individuals AIDS 2000 5 1540 1542
Abgueguen P Delbos V Chennebault JM Payan C Pichard E Vascular thrombosis and acute cytomegalovirus infection in immunocompetent patients: report of 2 cases and literature review Clin Infect Dis 2003 36 134 38 10.1086/374664
Ofotokun I Carlson C Gitlin SD Elta G Singleton TP Markovitz DM Acute cytomegalovirus infection complicated by vascular thrombosis: a case report Clin Infect Dis 2001 32 983 86 11247723 10.1086/319353
Inacio C Hillaire S Valla D Denninger MH Casadevall N Erlinger S Case report: cytomegalovirus infection as a cause of acute portal vein thrombosis J Gastroenterol Hepatol 1997 12 287 88 9195368
De Celis G Mir J Casal J Gomez D 31-year-old woman with an enlarged tender liver Lancet 1995 346 1270 7475722 10.1016/S0140-6736(95)91867-1
Uthman I Tabbarah Z Gharavi AE Hughes syndrome associated with cytomegalovirus infection Lupus 1999 8 775 77 10602454 10.1191/096120399678841034
Ailani RK Simms R Caracioni AA West WC Extensive mesenteric inflammatory veno-occlusive disease of unknown etiology after primary cytomegalovirus infection: first case Am J Gastroenterol 1997 92 1216 18 9219804
Lanari M Lazzarotto T Papa I Venturi V Bronzetti G Guerra B Faldella G Corvaglia L Picchio FM Landini MP Salvioli GP Neonatal aortic arch thrombosis as a result of congenital cytomegalovirus infection Pediatrics 2001 108 E114 11731641 10.1542/peds.108.6.e114
Labarca JA Rabaggliati RM Radrigan FJ Rojas PP Perez CM Ferres MV Acuna GG Bertin PA Antiphospholipid syndrome associated with cytomegalovirus infection: case report and review Clin Infect Dis 1997 24 197 200 9114147
Estival JL Debourdeau P Zammit C Teixeira L Guerard S Colle B Thrombose porte spontanée associée à une infection aiguë à cytomégalovirus chez une patiente immunocompétente Presse Med 2001 30 1876 1878 11791395
Arav-Boger R Reif S Bujanover Y Portal vein thrombosis caused by protein C and protein S deficiency associated with cytomegalovirus infection J Pediatr 1995 126 586 588 7699538
Youd P Main J Jackson E Cytomegalovirus infection and thrombosis: a causative association J Infect 2003 46 141 143 12634078 10.1053/jinf.2002.1035
Benoist S Laisné MJ Joly F Boudiaf M Panis Y Valleur P Cytomegalovirus infection as a cause of acute superior mesenteric vein thrombosis with jejunal infarction Surgery 2003 133 222 223 12605185 10.1067/msy.2003.8
Bauer KA The thrombophilias: well-defined risk factors with uncertain therapeutic implications Ann Intern Med 2001 135 367 373 11529700
Auerbach AD Sanders GD Hambleton J Cost-effectiveness of testing for hypercoagulability and effects on treatment strategies in patients with deep vein thrombosis Am J Med 2004 116 816 828 15178497 10.1016/j.amjmed.2004.01.017
Squizzato A Gerdes VEA Büller HR Effects of human cytomegalovirus infection on the coagulation system Thromb Haemost 2005 93 403 410 15735787
|
16098229
|
PMC1198216
|
CC BY
|
2021-01-04 16:38:21
|
no
|
Ann Clin Microbiol Antimicrob. 2005 Aug 12; 4:11
|
utf-8
|
Ann Clin Microbiol Antimicrob
| 2,005 |
10.1186/1476-0711-4-11
|
oa_comm
|
==== Front
Aust New Zealand Health PolicyAustralia and New Zealand Health Policy1743-8462BioMed Central London 1743-8462-2-171602951210.1186/1743-8462-2-17ResearchAn Australian childhood obesity summit: the role of data and evidence in 'public' policy making SA Nathan [email protected] Develin [email protected] Grove [email protected] Zwi [email protected] School of Public Health and Community Medicine, The University of New South Wales, Sydney NSW 2052 Australia2 Centre for Chronic Disease Prevention and Health Advancement, NSW Department of Health, Locked Mail Bag 961, North Sydney NSW 2059, Australia2005 20 7 2005 2 17 17 8 6 2005 20 7 2005 Copyright © 2005 SA et al; licensee BioMed Central Ltd.2005SA 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
Overweight and obesity in Australia has risen at an alarming rate over the last 20 years as in other industrialised countries around the world, yet the policy response, locally and globally, has been limited. Using a childhood obesity summit held in Australia in 2002 as a case study, this paper examines how evidence was used in setting the agenda, influencing the Summit debate and shaping the policy responses which emerged. The study used multiple methods of data collection including documentary analysis, key informant interviews, a focus group discussion and media analysis. The resulting data were content analysed to examine the types of evidence used in the Summit and how the state of the evidence base contributed to policy-making.
Results
Empirical research evidence concerning the magnitude of the problem was widely reported and largely uncontested in the media and in the Summit debates. In contrast, the evidence base for action was mostly opinion and ideas as empirical data was lacking. Opinions and ideas were generally found to be an acceptable basis for agreeing policy action coupled with thorough evaluation. However, the analysis revealed that the evidence was fiercely contested around food advertising to children and action agreed was therefore limited.
Conclusion
The Summit demonstrated that policy action will move forward in the absence of strong research evidence. Where powerful and competing groups contest possible policy options, however, the evidence base required for action needs to be substantial. As with tobacco control, obesity control efforts are likely to face ongoing challenges around the nature of the evidence and interventions proposed to tackle the problem. Overcoming the challenges in controlling obesity will be more likely if researchers and public health advocates enhance their understanding of the policy process, including the role different types of evidence can play in influencing public debate and policy decisions, the interests and tactics of the different stakeholders involved and the part that can be played by time-limited yet high profile events such as Summits.
==== Body
Background
Any policy-making process is complex – it deals with human and political dynamics, the use of resources, and power [1]. The development and implementation of policy in a democracy seeks to meet multiple objectives [2]: addressing major health and social policy problems, using public resources wisely, satisfying a range of stakeholders, avoiding conflict, and ensuring that political and economic objectives are met. Research is only one influence in the ongoing process of policy-making [3]. In setting the agenda, formulating policy, and implementing and evaluating it, various forms of evidence are sought and utilised. While conventionally such evidence is conceived as being derived from "scientific and objective" research, it is increasingly clear that a much wider range of sources and forms of evidence are influential [4]. There has been significant debate in Australia about the interface between evidence and policy-making [5], but little detailed analysis of the way evidence shapes the process of policy-making. This paper examines the role of data and evidence in public policy-making in response to childhood obesity in an Australian state, New South Wales (NSW).
Overweight and obesity (O&O) in Australia, as in many countries, has risen at an alarming rate over the last 20 years. Overweight is classified as a body mass index (BMI) of 25 and above, and obesity as 30 and above [6]. Obesity in men in Australia rose from 9.3% in 1980 to 17.1% in 2000 and for women from 8.0% to 18.9% [7,8]. O&O in children and young people has also increased markedly. From 1985 to 1995 the level of combined O&O in children more than doubled in all but the youngest age group of boys whilst the level of obesity tripled in all age groups and for both sexes [9].
Despite rising obesity, the policy response has been limited and hampered by a lack of evidence concerning effective interventions. The World Health Organisation (WHO) has highlighted "globesity" and released the Global Strategy on Diet, Physical Activity and Health [10]. Earlier, the United States (US) Surgeon General's Call to Action emphasised the need to create supportive environments which provide accessible and affordable healthy food choices and convenient opportunities for regular physical activity [11].
Australia was one of the first countries to produce an integrated national strategy for the prevention of O&O. The National Health and Medical Research Council (NHMRC) report 'Acting on Australia's weight: a strategic plan for the prevention of overweight and obesity' [12], was released in 1997, but its recommendations, which included strategies such as promoting physical activity, dietary monitoring, and encouraging the development of school canteen policies remained largely unaddressed.
Within this environment characterised by public policy inertia, the issue of childhood obesity, and the need for effective interventions, was brought to the forefront of one Australian state government's agenda through the NSW Childhood Obesity Summit (hereafter referred to as 'the Summit') in 2002. While obesity had already been identified as a problem, how to respond was unclear. With little evidence available to guide Government responses to the issue, the state health department's (hereafter referred to as NSW Health) articulated purpose of the Summit was to i) create better understanding in the community; ii) inform Members of Parliament; iii) hear and consider the views of families, parents and young people; iv) examine existing approaches and consider new ideas in a bipartisan forum; v) consider evidence; vi) identify ways to improve existing strategies and services; vii) build community consensus about future directions, and viii) recommend a future course of action so that the best available strategies, both long and short term, would be implemented to overcome the childhood overweight and obesity problem [13]. This paper examines the role of evidence and data in entrenching childhood obesity on the policy agenda, in shaping the Summit debate and informing the outcomes and the policies that were subsequently adopted.
Methods
Data collection
Data were collected from the transcripts of the Summit proceedings [14-16], media articles, the Summit Communiqué [17] which outlined the agreed resolutions, the Government Action Plan [18] published after the Summit and the announcement by the NSW Health Minister in December 2002 [19]. 'Factiva', a searchable archive of print media, was used to identify articles that referred to childhood obesity in the four main NSW statewide newspapers and one national newspaper (the Daily Telegraph, Sunday Telegraph, Sydney Morning Herald, The Sun Herald and the Australian) in the three months prior to the announcement of the Summit in July 2002 until the first public response from government in December 2002. There were 127 articles retrieved from this search.
Seven semi-structured key informant interviews [20] and one focus group discussion (FGD)[20] with three health staff involved in the Summit's organization were also conducted. The key informants included NSW health staff and experts in human nutrition, physical activity, and population health. The interviews and focus group discussion used a guide to elicit opinions on the stimulus for, and organization of, the Summit and its outcomes. The focus group discussion was transcribed for analysis and the interviews were used as background material.
Data analysis
The transcripts of the Summit proceedings [14-16], media articles and other key documents were reviewed and content analysed [20] to examine what type of evidence was used, by whom (eg. experts, industry, advocates) and for what purpose. Evidence that was valued or contested in the Summit debates and the media coverage received particular attention. The type of evidence used was categorised into three types based on a model adapted from Bowen & Zwi [4] who outlined five types of evidence. The categorisation used in the current study were empirical research (Type 1), such as randomised controlled trials, case control and cohort studies, time series analyses, observational studies, case reports and qualitative studies; ideas and opinions (Type 2) which incorporated the two categories of 'knowledge and information' and 'ideas and interests' outlined by Bowen & Zwi, and included evidence such as the results of consultation processes, opinions and views of "experts", interest groups and community members; and economic data (Type 3) which focused on economic evaluation, finance and resource implications.
Rigour
Rigour was addressed through triangulation, clear exposition of methods and reflexivity [21].
Triangulation is the use of different approaches, such as interviews and document analysis, to answer the same question which strengthens the rigour of a study and the interpretations made [22]. In the current study, interviews, analysis of transcripts from the summit debate and related documents and media coverage were used to answer questions posed in relation to the role of evidence in the NSW Childhood Obesity Summit.
It is important to consider the ways in which researchers and authors' past and present experiences may have shaped the way data was collected and interpreted – often referred to as reflexivity [22]. All the authors of this paper are involved, at some level, in public health advocacy and support a range of initiatives to address public health problems, including childhood obesity. The paper arose from a desire by the authors to better understand and reflect upon the role of evidence and its use by the different stakeholders in the Summit debate and how the debates around evidence were seen to influence the resolutions agreed. The involvement of all authors in the analysis and interpretation of the data presented in this paper, data triangulation, clear exposition of methods, conduct of a focus group with some of the key actors involved, and reflection on alternate ways of viewing the data were all important in enhancing the rigour of the study and the credibility of the interpretations made [22,23].
Results
Three phases were discernible in the process of policy making that occurred as part of the NSW Childhood Obesity Summit: 1) building and maintaining the momentum 2) summit debate and 3) outcomes and policy formulation.
1) Building and maintaining the momentum
Obesity had been recognised as a longstanding and increasingly important public health problem. Ebbeling et al (2002) pointed to publications decades earlier highlighting the issue and the need for a policy response[24]. Media interest in the issue of obesity in Australia was stimulated by available data highlighting "the doubling and tripling" of rates of obesity and concerns around the "second fattest kids in the world" (FGD). Obesity was seen as "the new tobacco" – the public health issue which was being recognised as demanding attention. Articles published in the peer review literature around this time [24,25] were triggers for media coverage and interviews with key informants and the focus group with NSW Health staff all emphasised the importance of media coverage in bringing the issue to public and policy attention:
"It [media coverage] was partly driven by data...the MJA [Medical Journal of Australia] also carried some data on childhood obesity and ... reinterpreting existing data sets. ...so that put it on the radar, that doesn't mean you've got [a] Summit happening yet... the data is essential – it is necessary, but not sufficient." (FGD)
"The doubling and tripling was the most used [news] grab everywhere, in every article, and it is still used." (FGD)
Why was NSW Health interested? The issue was shown to be important to the public. It provided the opportunity to divert attention away from other health issues which are considered solely the responsibility of government, for example, health care service provision. NSW Health also wanted to show leadership in an area where there was arguably Federal Government inaction. In New South Wales there was a clear perception that "prior to the Summit there was a national leadership vacuum" around childhood obesity (FGD). An earlier government summit on illicit drugs [26], had mobilised massive public attention and resources and it was hoped by NSW Health that a childhood obesity summit would draw in funds and resources to address this public health problem. A summit was seen as providing scope to debate interventions in an area where there was no scientific or political clarity at the time:
"there was interest, we were asked to do things, write things, pull things together... there were lots of false starts...we had things in train that were going to take another 5 or 10 years and they said they wanted a solution today... a summit was suggested as a way forward."(FGD)
Table 1 shows the number of media articles by month between April 2002 and December 2002. Within each month the percentage of articles that used Type 1, Type 2 or Type 3 evidence are identified. All the articles drew on more than one type of evidence. Peak months of coverage were July when the Summit was announced (n = 15), September when the Summit was held (n = 40), and December when the Health Minister announced the preliminary government response (n = 19). In the months prior to the announcement of the Summit, childhood obesity was covered 1–2 times per week in the newspapers studied.
Table 1 Number of media articles and evidence type used
Number of articles
Month Total Articles Type 1 Type 2 Type 3
April 5 5 5 0
May 7 6 3 0
June 9 9 5 2
July1 15 7 4 0
August 10 6 3 0
September2 40 15 16 2
October 11 4 5 0
November 11 8 8 2
December3 19 12 12 3
1Announcement of Summit
2 Month in which Summit was held
3 Month in which government initial policy response was announced
In the lead up to the Summit, most of the articles cited evidence of at least one type concentrating on Type 1 evidence focussed on the magnitude of the problem, backed up by expert opinion (Type 2). In the month before the first announcement by government in December economic data (Type 3), always referring to the cost of obesity to the health care system, were also reported.
Prior to the Summit and throughout the study period, Type 1 evidence was widely reported and largely uncontested, quoting authoritative sources such as the Lancet [24] and the Medical Journal of Australia [25] concerning the magnitude of the problem. Media representations drew on such data to present 'sound bites' to stimulate debate. The most commonly reported statistics were that either one in four, or one in 5 children in Australia was overweight or obese and that overweight and obesity had doubled between 1985 and 1995. These data from Magarey et al (2001) [9] were also contained in the background document prepared for the Summit [27] and included in the factual preamble to the Summit resolutions [16]. Once the Summit was underway, Type 2 data were more widely reported and ideas from experts, community members and key stakeholders concerning the way forward, were presented in the media. In putting forward their views, these stakeholders called on common sense understandings, research studies or pointed to a lack of conclusive evidence to support inaction.
Food advertising to children was a case in point. Prior to the Summit, debates about evidence in the media focused on taxing 'high fat foods' and banning food advertising to children. The soft drink industry spoke about the lack of good evidence for the effectiveness of such initiatives and the negative economic impact of a "fat tax". Physical activity and the role of parents as an influence on obesity were highlighted by the advertising and food industries as being the major influences on childhood obesity. Results from Sweden which were stated by the food and advertising industry as showing obesity rising despite an advertising ban were used to demonstrate that "there is no evidence that advertising makes children eat more fatty foods" (The Australian Newspaper, 1 July 2002)[28]. It became clear from the media coverage during the Summit that a ban on food advertising was the critical concern for industry who were calling a 'clear link' between harmful childhood behaviour and commercials, with editorials suggesting that instead "parents are the dominant influence on food choices" (Daily Telegraph, 12 September 2002) [29].
2) Summit debate
The Summit was held in September 2002 at Parliament House in NSW. An across-government organising committee oversaw delegate selection and sought to ensure balanced representation including: i) children and young people; ii) families, parents and community perspectives; iii) experts; iv) relevant peak bodies; v) special population perspectives, such as the socially disadvantaged, people from culturally and linguistically diverse communities, Aboriginal and Torres Strait Islanders, rural and remote communities, and people with disabilities.
The Summit provided an opportunity for delegates to present their case for action during plenary sessions. During the Summit, nine working groups (WGs) were convened: i) Early Childhood, ii) Family and Community, iii) School Education, iv) Health, v) Sport, Recreation & Fitness, vi) Local Government, vii) Commercial Food Industry, viii) Media, and ix) Transport and Planning.
The WGs were requested to put forward 10–15 resolutions for the Communiqué to be presented to government. The importance of evidence for the resolutions was made clear by the NSW Premier on the first day of the summit when he referred to the NSW Drug Summit [26], which had been held in May 1999.
"The Drug Summit emphasised looking at evidence, basing policies on evidence ... I would like that to be your guide too." (NSW Premier, Day 1 pg 33)
The case for action to tackle childhood obesity was uncontested from the outset. In opening the Summit the NSW Minister for Health referred to strong empirical evidence of the magnitude of childhood obesity:
"In Australia the level of combined overweight and obesity between 1985 and 1995 has more than doubled... Today in New South Wales one in five children aged between seven and 15 are classified as being either overweight or obese." (NSW Minister for Health, Day 1, pg 5)
Experts from government departments, academic institutions and the health service put forward similar statistics that highlighted the magnitude of obesity. Many outlined the consequences of overweight and obesity for type 2 diabetes in particular. The use of simple statistical concepts such as "doubling in rates" and "one in five of our children" were commonplace. Economic evidence highlighted the cost of the "obesity epidemic" to society – "it costs us a community $830 million a year" (Minister for Health, Day 1, pg 7) and individuals – "in one year the personal cost to individuals who are obese is $19 billion" (expert, Day 1 pg 16). Such data were uncritically and widely accepted during the Summit.
On the opening day, experts, parents, community groups and industry talked anecdotally about societal changes over decades and their impact on physical activity and food consumption. Statistics and studies were referred to in support of these observations, such as an increased reliance on carbonated sugared drinks, although no actual data were provided. Data from the US concerning changes in levels of physical activity were presented: "most people in my generation walked to school, today less than a third of children in the United States walk to school" (US expert, Day 1, pg 12) and Australian data on sedentary activity: "97% of our adolescents watch television ... between 60 and 80% play computer or video games." (Australian expert, Day 1, pg 16). Anecdotal observations about changing societal behaviours and environments were widely cited and seen as important factors to address despite the lack of reliable trend data and research evidence:
"we do not yet have evidence that any single one of these factors is driving the epidemic" (US expert, Day 1, pg 13)
"we know very little in any, firm solid way about the factors that influence young people to be active or sedentary – all we have to work with over the next three days are some recently informed guesses and some far less well-informed speculations" (Australian expert, Day 1, pg 12)
The views of young people, were an integral part of the Summit process and provided an emotive appeal to take action. Young people's stories were shown on video and they addressed the Summit. However, there appeared to be little attempt to draw these views together, articulate common threads or examine whether and how such views related to other empirical data and expert opinion. A young person opening the Summit stated that "It is genuinely important that our voice be heard"(young person, Day 1, pg 2). The FGD participants saw young people's stories as being powerful in stimulating action:
" we had to have lots of consultation processes that included the voice of the child... engaging the children... it was the most powerful thing." (FGD)
A young person at the Summit, however, expressed frustration about the focus on evidence:
"Unfortunately we have been bombarded with statistics. They have been repeated over and over again ... we are almost scared to put up a decent suggestion." (young person, Day 2, pg 30)
In contrast to the research evidence supporting the magnitude of the problem and the influencing factors, evidence supporting calls for action were mostly opinion and ideas with some reference to overseas efforts. Nonetheless, much was made of the need for evidence-based strategies, with a US expert claiming three strategies that were "defensible, but not conclusive" (US expert, Day 1, pg 13): breastfeeding, limiting television viewing and the promotion of physical activity. A Cochrane systematic review [30] covering 1985–2001 and encompassing 14,000 studies was reported (researcher, Day 1, pg 37). It found 11 studies of a high enough quality to examine the effectiveness of the intervention and it found only small or no effects with those interventions that were most effective focussed on reducing sedentary behaviour. A few delegates questioned the need for evidence from primary prevention trials, pointing to broader experiences that tell us "what works". They highlighted other successful public health campaigns such as in tobacco control as evidence for the success of a range of strategies, including advertising controls and taxes:
"we do need evidence, we do need to work at what has been shown to be the most effective, but that should not inhibit us from acting now. There have been a number of successful public health programs that have been introduced without definitive evidence." (expert, Day 1, pg 39)
The FGD and comments by Summit delegates highlighted the need for action coupled with thorough evaluation:
"There needs to be a recognition of the sense of urgency...that policy won't wait for the data." (FGD)
"This is about promising interventions, we have to just go with promising interventions, make sure they do no harm and just evaluate the heck out of them, and then maybe in ten years time, if they weren't the best things to do, well at least we did something" (FGD)
"we need periodic surveys to tell us how we are doing with respect to implementation of strategies ... we need causal models, that is, longitudinal studies which allow us to link risk factors like change in the food supply with changes in the prevalence of obesity." (US expert, Day 1, pg 13)
"we would like to see a regular – maybe five yearly – national nutrition, physical activity and health survey." (industry, Day 2, pg 2)
"Certainly, we need to take action, but at the same time we need to be doing research. We cannot continue to act in an evidence vacuum." (expert, Day 2, pg 20)
The most contentious issue centred on the role of food advertising to children (see Figure 1). The intensity of the debate between food industry representatives and the advocates of a ban on food advertising to children clearly illustrates the way different types of evidence are drawn upon to articulate a particular position or undermine that of opposing perspectives.
Figure 1 Contesting the evidence: food advertising and obesity.
3) Outcomes and policy formulation
The final Communiqué to government was to include a "factual foundation" and recommendations and resolutions for future action. The purpose of this component of the Summit was to: "Frame evidence-based solutions within a community-based 'reality check' perspective."(Day 3, pg1)
Evidence of the magnitude of overweight and obesity was included with little debate. Statements about the influencing factors were carefully worded to reflect agreement on importance and available evidence:
"Although physical activity trend data is lacking, it is apparent that children and adolescents are less physically active" (Day 2, pg 72)
"An increase in television viewing is associated with an increase in obesity in children. An increase in sedentary behaviour is associated with an increase in obesity in children. Experts have advised that television viewing needs to be one of the targets for obesity control efforts" (Day 2, pg 78)
Exposure to advertising messages was included in the factual preamble referring to the range of potential influences on food selection behaviours. The resolutions about food advertising to children generated the most debate concerning the evidence-base for such interventions (see Figure 1) and the relationship between food choice and television viewing. This debate illustrates the use of different types of evidence by industry representatives as one means of opposing calls for a ban on food advertising to children.
A resolution to ban food advertising to children was not agreed. In its place agreement was reached to have an independent review by the Federal government of the regulatory arrangements for food advertising "in recognition that food advertising is one of the contributing factors to the prevalence of eating habits that may promote obesity"(Day 3, pg 9) in addition to a review of a voluntary code to be undertaken by industry. Attempts by the Food Industry to have this statement deleted from the Communiqué were not successful. A systematic review of the impact of food advertising on diet, physical activity and childhood obesity was also recommended.
All other resolutions passed with minimal debate, including those addressing physical activity, school education, transport and planning. Most of the resolutions agreed at the Summit and taken up in the subsequent Government Action Plan [18] were focused on physical activity and nutrition education. Mandatory guidelines for school canteens also passed as a resolution despite some opposition from industry. Numerous resolutions in the Communiqué [17] referred to research and a detailed section on surveillance and monitoring proposed a funded collaborative centre of excellence in research, prevention and management.
Limited attention was devoted to the financial and logistical feasibility of the resolutions – this was apparent by the number of resolutions that required intervention at a federal rather than state level. However, the preliminary response from the government in December considered what was feasible in the current financial and political context:
"it wasn't really evidence-based, it was the feasibility of whatever strategy they had suggested..." (FGD)
In the final Summit address by the NSW Health Minister [16] the two resolutions specifically mentioned and strongly supported were the recommendations on school canteens and a collaborative centre for excellence for overweight and obesity research. These two initiatives were subsequently publicly announced in December as the key response to the Summit by the government [19]. The advisers to the Minister and NSW Health were concerned to ensure that the Summit resulted in some "announceable" interventions – and the two chosen seemed "doable", of value, and in some respects least contentious (FGD).
Discussion
This paper has sought to present key elements of the use of data and evidence in the NSW Childhood Obesity Summit. There are other dimensions of policy-making which deserve attention and other interpretations of the process possible. As indicated by Ham and Hill "It is rarely possible to agree on one version of events: the most that can be achieved is a plausible interpretation" [2] (p xi).
Empirical evidence of the magnitude of the obesity problem and the economic cost to the health system were critical to generating publicity and framing the case for action on childhood obesity in the lead up and during the Summit. This evidence was never contested and became a part of the factual foundation of the Summit Communiqué [17]. It is clear that the combination of Type I data, which was largely epidemiological in origin, and Type 3 data about the economic costs of the problem was persuasive.
The lack of empirical evidence for many of the influencing factors and related interventions, for example in the area of physical activity, did not hamper agreement of resolutions at the Summit and was instrumental in funding a research centre to collect better data and evidence for what works. Health officials who recognised the lack of an evidence base for interventions sought to promote those that seemed most logical and appropriate, along with a concern to ensure subsequent careful evaluation. The Summit demonstrated that policy action will move forward in the absence of strong research evidence if government sees the need to respond to public concerns.
However, lack of compelling evidence for interventions is likely to have been a factor in the failure of government to commit significant new funds and to agree to controversial recommendations around food advertising given strong industry opposition. The only contentious resolution taken up by government was for mandatory guidelines for school canteens: this appealed to many community groups and parents who attended the Summit and government is likely to have perceived strong public support for this intervention. Other commentators have questioned the soft policy options adopted in response to the obesity epidemic in Australia following the NSW Summit and raised questions about the way public health issues, such as obesity, are framed in public discourse [31].
The food and advertising industries who were represented at the Summit used the lack of well supported 'scientific' evidence to oppose controls on advertising. In contrast, the debates and resolutions around physical activity using anecdotal evidence, expert opinion and common-sense solutions garnered widespread support as there was no industry that stood to suffer financially from the action proposed. Where strong interests and powerful groups oppose policy direction, the evidence base required for government action, if it is to proceed, needs to be substantial. It is also possible that the more prominent role of the federal government in food advertising regulation and control worked against the agreement of concrete resolutions around food advertising to children.
Economically important industries have been seen by others as critical in the preparedness of governments to support controversial public health initiatives [32] and calls for more research have been presented as tactics to delay policy change [33]. However, creative and clear communication of the evidence has been instrumental in other areas, notably the successful efforts to ban tobacco advertising in Australia and in many other countries around the world despite powerful industry opposition [32, 33]. There is also more scope for interaction and collaboration with the food industry than with the tobacco industry as food as a product is not inherently harmful [33]. The food industry can have an important role in supporting a range of policy initiatives that promote healthy eating as was evident in the NSW Childhood Obesity Summit, but are likely to remain adversarial where industry profits are, or appear to be, at stake.
Conclusion
The NSW Childhood Obesity Summit played a role in promoting an agenda for action to address childhood obesity. It raised awareness in the public and political arena and provided a public forum for debating research evidence. The Summit demonstrated that while it is not necessary to have all the evidence in place to agree actions, that more radical policy change is much more difficult to achieve in the absence of established and detailed evidence, given the interests of important stakeholders, notably the private sector. The process and the outcomes of the Summit suggest that in the absence of strong Type 1 data, and where Type 2 evidence is contested, that policy-makers may opt for the path of least resistance: a call for more and better research and support for the systematic evaluation of interventions. While beneficial to researchers, direct and short term health gain may be limited.
The lack of an agreed evidence-base provides politicians with a freer hand in choosing actions which have wide appeal and are less controversial, rather than those which may produce greatest health benefit. The Summit's success in generating a set of resolutions should not be discounted even if large resource allocations were not forthcoming. Tobacco control initiatives have taken decades of concerted effort to realise [33] and obesity control efforts are likely to face the same challenges around evidence and action. The prospects of controlling obesity in the future will be amplified if researchers and public health advocates enhance their understanding of the policy process, the interests and tactics of the different stakeholders involved, and the role different types of evidence can play in influencing public debate and the decisions of policy-makers in time-limited yet high profile events such as Summits. Further research is needed to increase our understanding of the role of Summits in the broader politics and processes of policy-making.
Competing interests
Elizabeth Develin was involved in the original work described leading up to and including the NSW Child Obesity Summit and is employed by NSW Health. All the other authors are part of the School of Public health and Community Medicine, UNSW, which is in receipt of modest funds to collaboratively develop this reflection and paper by NSW Health, the body which organised the Summit described.
Authors' contributions
This paper was conceived jointly by all authors; all contributed to analysis and writing. SN prepared the first draft with significant contributions by ED and comments from AZ and NG. All authors contributed to writing successive drafts, feeding in literature, analysis and insights.
Acknowledgements
The authors wish to acknowledge insights and feedback from Bill Bellew (Director, Centre for Chronic Disease Prevention and Health Advancement, NSW Health) and Phillip Vita (former Manager, Nutrition & Physical Activity Branch, NSW Health, now Executive Officer, NSW Centre for Physical Activity & Health) and the work of Kate Hawkins and Sarah Yallop from NSW Health who conducted the key informant interviews. The authors would also like to thank the reviewers of the original submitted manuscript for their constructive and considered comments. The School of Public Health and Community Medicine, The University of New South Wales, and The Centre for Chronic Disease Prevention and Health Advancement, NSW Health, are collaborating on a project, funded by the Centre, to foster reflection, analysis and writing between academics, practitioners and policy-makers, and aimed at contributing to bridging the academic-policy divide.
==== Refs
Walt G Introduction to health policy. Process and power. 1994 London, Zed Books
Ham C Hill M The policy process in the modern capitalist state. (Second edition). 1993 Hertfordshire, Harvester Wheatsheaf
Rist RC Denzin NK and Lincoln YS Influencing the policy process with qualitative research. In (Editors). Handbook of Qualitative Research 1994 London, Sage Publications 545 557
Bowen S Zwi AB Pathways to "evidence informed" policy and practice: a framework for action. Public Library of Science Medicine 2005 2 e166 15913387
Lin V Gibson B Evidence-based Health Policy: Problems and Possibilities 2003 South Melbourne, Oxford University Press.
World Health Organisation Obesity: preventing and managing the global epidemic. Report of a WHO consultation. WHO Technical Report Series 2000
Bennett SA Magnus P Trends in cardiovascular risk factors in Australia: results from the National Heart Foundation's Risk Factor Prevalence Study, 1980-1989 Medical Journal of Australia 1994 161 519 527 7968750
Dunstan D Zimmet P Welborn T Diabetes & Associated Disorders in Australia: The final Report of the Australian Diabetes, Obesity and Lifestyle Study (AusDiab) 2000 Melbourne, International Diabetes Institute
Magarey A Daniels L Boulton TJC Prevalence of overweight and obesity in Australian Children and adolescents: reassessment of 1985 and 1995 data against new standard international definitions. The Medical Journal of Australia 2001 174 561 564 11453327
World Health Organisation Global strategy on diet, physical activity and health. 2004 World Health Organisation
US Department of Health and Human Services The Surgeon General's call to action to prevent and decrease overweight and obesity 2001 Washington, US Department of Health and Human Services.
National Health & Medical Research Council Acting on Australia's weight: a strategic plan for the prevention overweight and obesity. 1997 Canberra, Commonwealth of Australia
NSW Health Committee NSWCOSO NSW Childhood Obesity Summit Draft Program 2002 Sydney, NSW Department of Health
NSW Health NSW Childhood Obesity Summit: Report of proceedings of the first day
NSW Health NSW Childhood Obesity Summit: Report of proceedings of the second day
NSW Health NSW Childhood Obesity Summit: Report of proceedings of the third day
NSW Health NSW Childhood Obesity Summit Communique
NSW Health Prevention of Obesity in Children and Young People: NSW Government Action Plan 2003-2007 2003 Sydney, NSW Department of Health
NSW Minister for Health Youth Obesity - Five Year Plan Media Release 2002 , NSW Health
Patton MQ Qualitative Research and Evaluation Methods 2002 3rd edition California, Sage Publications Inc.
Mays N Pope C Assessing quality in qualitative research BMJ 2000 320 50 52 10617534 10.1136/bmj.320.7226.50
Denzin NK Lincoln Y Handbook of Qualitative Research: 2nd Edition 2000 California, Sage
Ebbeling CB Pawlak DB Ludwig DS Childhood obesity: public health crisis, common sense cure. Lancet 2002 360 473 482 12241736 10.1016/S0140-6736(02)09678-2
Goodman S Lewis PR Dixon AJ Travers CA Childhood obesity: of growing urgency. The Medical Journal of Australia 2002 176 400 401 12041641
NSW Government NSW Drug Summit
NSW Health NSW Childhood Obesity Summit Background Paper
The Australian Going off the scales The Australian 2002 1 July 2002 Sydney,
Daily Telegraph Dishing out blame Daily Telegraph 2002 Sydney,
Campbell K Waters E O'Meara S Kelly S Summerbell C Interventions for preventing obesity in children The Cochrane Database of Systematic Reviews 2002
Lin V Robinson P Australian public health policy in 2003-2004 Australian and New Zealand Health Policy 2005 2
Chapman S Lupton D The fight for public health: principles and practice of media advocacy 1994 London, BMJ Publishing Group
Yach D McKee M Lopez AD Novotny T Improving diet and physical activity: 12 lessons from controlling tobacco smoking BMJ 2005 330
|
16029512
|
PMC1198217
|
CC BY
|
2021-01-04 16:38:28
|
no
|
Aust New Zealand Health Policy. 2005 Jul 20; 2:17
|
utf-8
|
Aust New Zealand Health Policy
| 2,005 |
10.1186/1743-8462-2-17
|
oa_comm
|
==== Front
BMC BiolBMC Biology1741-7007BioMed Central London 1741-7007-3-161602949210.1186/1741-7007-3-16Research ArticleAlternative pre-mRNA processing regulates cell-type specific expression of the IL4l1 and NUP62 genes Wiemann Stefan [email protected] Anja [email protected] Annemarie [email protected] Molecular Genome Analysis, German Cancer Research Center, Im Neuenheimer Feld 580, Heidelberg, 69120, Germany2005 19 7 2005 3 16 16 8 3 2005 19 7 2005 Copyright © 2005 Wiemann et al; licensee BioMed Central Ltd.2005Wiemann 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
Given the complexity of higher organisms, the number of genes encoded by their genomes is surprisingly small. Tissue specific regulation of expression and splicing are major factors enhancing the number of the encoded products. Commonly these mechanisms are intragenic and affect only one gene.
Results
Here we provide evidence that the IL4I1 gene is specifically transcribed from the apparent promoter of the upstream NUP62 gene, and that the first two exons of NUP62 are also contained in the novel IL4I1_2 variant. While expression of IL4I1 driven from its previously described promoter is found mostly in B cells, the expression driven by the NUP62 promoter is restricted to cells in testis (Sertoli cells) and in the brain (e.g., Purkinje cells). Since NUP62 is itself ubiquitously expressed, the IL4I1_2 variant likely derives from cell type specific alternative pre-mRNA processing.
Conclusion
Comparative genomics suggest that the promoter upstream of the NUP62 gene originally belonged to the IL4I1 gene and was later acquired by NUP62 via insertion of a retroposon. Since both genes are apparently essential, the promoter had to serve two genes afterwards. Expression of the IL4I1 gene from the "NUP62" promoter and the tissue specific involvement of the pre-mRNA processing machinery to regulate expression of two unrelated proteins indicate a novel mechanism of gene regulation.
==== Body
Background
Many mechanisms for the alternative use of promoters, exons and polyadenylation signals within genes are known to significantly contribute to the complexity of the transcriptome [1-6]. These variations increase the number of products that can be generated from the currently recognized 20,000 – 30,000 protein-coding genes of the human genome [7]. For example, alternative promoters are used to confer specificity of mRNA expression in time and space [8,9] and of mRNA translation [10]. Often the N-terminal ends of proteins are altered to generate or remove signal sequences for protein localization [11]. Central exons may or may not be present thus changing the peptide sequence and properties [12]. The alternative use of polyA signals also has effects, for instance, on RNA stability [13,14].
The mechanisms described above all have in common the fact that the elements involved are associated only with the gene being transcribed and not with any other gene. The mechanism of trans-splicing, in which elements from more than one gene are involved in the generation of transcripts, is an open matter of discussion, although it appears to be rare and its function is still not well understood [15]. Overlapping genes and transcripts have been described in many species and occur in several varieties [16-18]. However, in vertebrates, few transcripts have been described which join two genes with different reading frames [19]. We have found evidence for sequence overlap of transcripts from two protein coding genes, NUP62 and IL4I1, where the latter is expressed in a tissue and cell-type specific manner. Both genes are transcribed from the same promoter and share the first two exons. A similar process has been described for Caenorhabditis elegans [20], in which mRNAs of two cholinergic proteins are transcribed from one promoter. Until now, this principle did not appear to be conserved in higher eukaryotes. The NUP62/IL4I1 genes are therefore the first proof that this mechanism is present in vertebrates. However, in contrast to what has been observed in C. elegans, the functions of the two proteins encoded by the one promoter are completely unrelated.
The protein encoded by NUP62 belongs to the class of nucleoporins (Nups) and is an essential part of the nuclear pore complex [21,22]. Its N terminus is believed to be involved in nucleocytoplasmic transport, while the C-terminal end contains a coiled-coil structure aiding in protein-protein interactions, and may function in anchorage of the protein in the pore complex (Annotation for P37198 in Swiss-Prot [23]). Nup62, like the other Nups, is conserved in the eukaryote kingdom [24,25]. The NUP62 gene consists of a single promoter with a CpG island and three transcribed exons. The protein is encoded exclusively by the terminal exon; the first two exons are non-coding. The second exon is prone to alternative splicing and is not contained in about half of the reported cDNAs derived from that gene (e.g., IMAGE:3050260 [26] and DKFZp547L134 [27]). NUP62 is ubiquitously expressed, an observation compatible with its essential role in transporting cargo across the nuclear envelope.
IL4I1 was initially identified to be exclusively expressed in B lymphoblasts as a gene that was induced by treatment with interleukin 4 (IL-4) [28,29]. Since then, the encoded protein has been identified as a leukocyte specific L-amino acid oxidase (LAAO; [30]) that specifically oxidizes aromatic amino acids. The protein contains an N-terminal signal peptide, which targets the protein to the endoplasmic reticulum and presumably to the lysosomes [30], where it is believed to be involved in antigen processing in B cells [30] and thus act in the immune response. The gene is reported to be transcribed from a single promoter, which appears to restrict expression to cells of the immune system, mostly in B lymphocytes [31]. It consists of eight exons, and the translation start is located in the second exon. The gene is conserved in eutherian mammals (NCBI HomoloGene:22567), but has not been identified in other eukaryotes and in prokaryotes.
We have identified several expressed sequence tags (ESTs) that indicate expression of IL4I1 in tissues other than B lymphocytes, namely human and mouse testis and brain. This expression of the IL4I1 gene was apparently driven by the same promoter as the upstream NUP62 gene. We have verified expression of the Il4i1_2 variant in mouse testis and brain, and thus show that the previously reported NUP62 promoter also drives expression of a second gene in a cell-type and tissue specific manner. The mRNA consists of sequence from both genes and two joining exons which are not part of either previously reported gene locus. Our findings indicate a new mechanism of gene regulation in which two genes that encode unrelated proteins share the same promoter but yet are still expressed in radically different cellular patterns. This suggests that the nature of the transcripts and proteins encoded by these two genes is controlled by tissue specific regulation of pre-mRNA processing.
Results
The exon structure of variant IL4I1_2 joins the described NUP62 and IL4I1 genes
Based on the available sequence information we predicted the gene structure for the human variant IL4I1_2 transcript represented by cDNA IMAGE: 5742307 in Fig. 1. To validate this structure we obtained several Mammalian Gene Collection clones that cover the splice variant and sequenced them to completion. One cDNA (IMAGE:4822638; Acc: BC026103) contained two mutations leading to premature in-frame stop codons. A second cDNA (IMAGE:5168029) contained exon 2 (35 nucleotides) of the previously reported IL4I1 gene [32], also disrupting the open reading frame (ORF). The remaining clones (IMAGE:5171014, IMAGE:5742307 and IMAGE:4838597) matched the predicted gene structure and thus supported the sequence of the variant. This gene structure includes the presumed first two exons of the NUP62 gene which are both part of the 5' untranslated region (UTR). Transcription of that variant is apparently controlled by the promoter that also controls expression of the NUP62 mRNA.
Figure 1 Structure of the human NUP62 and IL4I1 genes at chromosomal band 19q13.33. Genes are both shown from 5' (left) to 3' (right). Exons are represented by vertical bars and boxes, intronic sequence by horizontal lines. The NUP62 gene (three exons) is located upstream and tail-to-head to the previously reported IL4I1 gene (eight exons). NUP62 has two splice variants represented by cDNAs IMAGE:3050260 (includes exon 2), and DKFZp547L134 (skipped exon 2). The Nup62 protein is encoded by exon 3 (blue box). IL4I1 does not have reported splice variants and is represented by the sequence AF293462. Translation start of the protein is located in exon 2 (blue bars represent coding region). Several cDNAs were identified to link NUP62 and IL4I1 transcripts (representative: IMAGE:5742307). The splice form consists of the first two apparent exons of the NUP62 gene, two joining exons mapping in between the reported NUP62 and IL4I1 gene loci (red arrow heads), and exons 3–8 of the known IL4I1 gene. Translation start of the protein is in the second joining exon. The sequences of the joining exons are conserved in eutherian species, but not for instance in chicken or fish (taken from UCSC genome browser [54]). The reported promoter of the NUP62 gene contains a CpG island, while the promoter of previously known IL4I1 does not. The two genes cover 40 kb on the chromosome.
The terminal and coding exon from the NUP62 gene is not contained in the IL4I1_2 variant (Fig. 1). While the initiator ATG of the reported IL4I1 ORF is located in exon 2, the first two exons of the known IL4I1 gene are absent in the variant. Instead, the variant contains two additional exons (indicated with red arrowheads in Fig. 1) that are located in the region between the previously reported NUP62 and IL4I1 loci. The latter of the two exons contains the assumed translation initiator ATG.
The IL4I1_2 variant is conserved in eutherian mammals
The splice variant is conserved in other eutherian mammals where order and orientation of the NUP62 and IL4I1 genes are syntenic. Five ESTs from mouse verify the transcription and splicing of the Il4i1_2 variant. Like the human ESTs, the mouse ESTs were derived from cDNAs that had been generated from either testis or pooled tissues. One EST was sequenced from rat testis. All these cDNAs contain the first exon of the Nup62 gene, two intergenic exons and then exon 3 of the Il4i1 gene. There is apparently no homolog of human exon 2 of NUP62 in mouse and rat. Mouse Nup62 is thus the equivalent of the human splice variant represented by cDNA DKFZp547L134. The location and sequences of the joining exons that are specific for the IL4I1_2 variant are conserved between mouse, dog and human. Sequence conservation of the variant joining exons is higher than that of exons 1 and 2 of previously reported IL4I1 (Fig. 2). The probable translation initiation codon in exon 4 (exon 3 in mouse and rat) lies within a consensus Kozak sequence context (Fig. 2; [33]). An upstream ATG, which is in frame with the ATG we propose to initiate translation, does not match the Kozak consensus rules. It is present in human and chimpanzee, but not in mouse, rat or dog, and thus is not convincing; we suspect it could be prone to leaky scanning [33]. We conclude that translation either starts at the conserved ATG, or that use of the upstream ATG could possibly change some property of the encoded protein. While the N terminus of Il4I1_2 protein is predicted (SignalP [34]) to be a signal peptide (P = 0.969) when starting at the Kozak-ATG, the extended N terminus is predicted to be a signal anchor (P = 0.587) and not a signal peptide (P = 0.316). An extension at the N terminus might thus change localization of the protein.
Figure 2 Alignment of sequences upstream the IL4I1 and variant IL4I1_2 open reading frames. Alignments of IL4I1 (upper panel) and variant IL4I1_2 (lower panel) sequences are shown. The first two exons of IL4I1 are shown, and splice sites are underlined and in bold. Ten nucleotides of the next intron are displayed downstream of the first coding exon, which all match the consensus splice donor sequence. Reading frames are in all caps, and the deduced peptide sequences are given below. The human and chimpanzee sequences of IL4I1_2 have an upstream and in-frame start codon (bold), which is not conserved in mouse, rat or dog, and which does not match the Kozak consensus rules. All IL4I1_2 sequences have a short upstream ORF (all caps) which is conserved in location and sequence. The mouse, rat and human sequences were derived from cDNA sequences, while the chimpanzee and dog sequences were deduced from an alignment of human cDNA with the respective genome sequences [54].
All transcripts analyzed had a short six (dog: seven) residue upstream ORF, the localization and sequence of which was conserved. It remains to be determined whether this ORF is expressed in vivo as has been shown for other genes [35]. This ORF is too small and too close to the initiator ATG of the IL4I1-ORF to suggest an internal ribosome entry site (IRES) – type mechanism [36].
The IL4I1 gene has thus far only been found in eutherian mammals. This is supported by analysis of the genes downstream of the NUP62 orthologous genes in non-eutherian species. In Fugu rubripes, the next gene downstream of NUP62 is a homolog of human integrin alpha 6, and the two genes are oriented tail to tail. In Gallus gallus, the next gene downstream is the homolog of a human X-chromosomal gene (FLJ11016) with unknown function, and the genes are oriented head to tail. In Drosophila melanogaster, Nup62 is followed by a hypothetical WD-repeat protein (CG7989), which is in the opposite orientation (tail to tail) to Nup62. The situation in the opossum (Monodelphis domestica; thus far the only marsupial species sequenced) is unclear, as the sequence scaffold that covers NUP62 terminates 4 kb downstream and no gene is annotated there. However, the two genes that, according to annotation, flank opossum NUP62 do not map to the chromosomal region that harbors human NUP62 and IL4I1. In addition, no ortholog of the IL4I1 gene has yet been identified in the opossum genome. Thus, the evidence so far suggests that expression of variant IL4I1_2 (just as of original IL4I1) might be restricted to eutherian mammals. The sequencing and transcript analysis of more mammalian species will help to uncover the origin of the IL4I1 gene and its variant.
Mature ll4i1 protein and its variant are likely identical in sequence
Since the translation start in the previously reported IL4I1 transcript differs from that in the variant described here, the two protein products differ at their N termini. Fig. 3 shows a sequence alignment of the N-terminal ends of Il4I1 and the new variant. The N termini of Il4i1 and those of the variant Il4i1_2 are conserved in the species analyzed. The Il4i1 protein has been reported to be transported into the endoplasmic reticulum and the endosomes with help of an N-terminal signal peptide [30]. SignalP predicts such a signal peptide to be cleaved upstream of the glutamine residue at position 22 of the human variant protein. The homologous position is a leucine in the mouse protein and is there predicted to be the cleavage site. The same residue is the cleavage site also in the previously reported Il4i1. Consequently the processed proteins are probably identical in sequence, and only differ in the length of the respective signal sequences. We next analyzed whether the N terminus of the Il4i1_2 variant may serve as signal peptide in vitro and expressed the protein in fusion with green fluorescent protein (GFP) in mammalian cell culture [37,38]. The variant protein was indeed translocated into the endoplasmic reticulum [39] when overexpressed, and had the same localization as an overexpressed Il4I1-GFP fusion protein (not shown).
Figure 3 Sequence alignment of N-terminal ends of peptides encoded by the IL4I1 gene (Il4i1) and the novel variant (Il4i1_2). Sequences encoded by the first coding exon are different between the two variants, and the splice site is indicated with a red arrowhead. The mouse sequence of Il4I1_2 was derived from the following ESTs: BY100275, BY099330, BY087056, BY092834 and BY088421. The rat sequence was translated from the EST CV117152. Dog and chimpanzee sequences were deduced from aligned genomic sequences [54]. The cleavage site of the signal peptidase is predicted at the same position in all sequences and is indicated with a blue arrow head.
The IL4I1_2 variant is specifically expressed in testis and brain
EST evidence indicated that expression of the variant transcript might be tissue specific, as cDNAs exclusively from testis and brain had been sequenced. We analyzed the expression of the variant transcript in Northern blots of fetal and adult mouse (Fig. 4). A probe specific for the variant IL4I1_2 was employed, comprising the two joining exons downstream of the NUP62 coding exon. These exons are indicated with red triangles in Fig. 1. No expression of the Il4i1_2 variant was observed in fetal mice and in most adult tissues. A strong and specific band at 2.45 kb was only observed in the testis. The variant Il4i1_2 transcript is predicted to be 2.3 kb in size, not counting the polyA tail, when sequence of the ESTs (Methods) is extended with the known Il4i1 sequence towards its 3' end. The human variant IL4I1_2 is of similar size. Expression in the brain was expected because of cDNAs and ESTs available from that tissue, but not observed in Northern blot analysis. A smaller RNA of unknown sequence at 1.8 kb was visible in the fetal mouse, and in adult mouse liver, kidney and testis.
Figure 4 Northern hybridization with probe specific for the mouse Il4i1_2 splice variant. Blots contained poly A+ RNA from fetal mice 7 dpc (1), 11 dpc (2), 15 dpc (3), 17 dpc (4) and adult mouse tissues (heart (5), brain (6), spleen (7), lung (8), liver (9), skeletal muscle (10), kidney (11), and testis (12). Hybridization was with a probe comprising the joining exons 2 and 3 of the mouse Il4i1_2 variant, which are equivalent to human exons 3 and 4 (red arrowheads in Fig. 1). The blots were reprobed for beta actin mRNA as control for RNA content. The probe cross-hybridized with gamma actin mRNA where expressed.
Having identified expression of the Il4i1_2 variant in a tissue other than B lymphocytes, we next carried out RNA in situ hybridization to identify a possible cell-type specificity of this expression, and to find other tissues and cells where the variant is expressed. Expression of variant Il4i1_2 was found in testis to be predominantly in Sertoli cells at the periphery of the ducts (blue spots in Fig. 5, panels A1 and A2). In contrast to the Northern analysis, where brain did not have detectable expression of the Il4i1_2 variant, RNA in situ hybridization revealed expression of the variant transcript in the adult mouse brain (Fig. 6). Purkinje cells (cerebellum), cells of the hippocampus, and mitral cells in the olfactory bulb were specifically stained with the Il4i1_2 specific antisense probe (Fig. 6). Even though expression in some cell types within the brain was strong, overall expression of variant Il4i1_2 in the brain was weak, matching the results obtained with pooled brain tissue in the Northern analysis. No signals were detected in adult liver and kidney or in any of the embryonic stages by RNA in situ hybridization (not shown).
Figure 5 RNA in situ hybridization of variant Il4i1_2 in mouse testis. Probes were transcribed from the joining exons specific for variant Il4i1_2, which are located between the previously known Nup62 and Il4i1 gene loci. Signals (in blue) obtained with the antisense probe are on the left (A), and those obtained with sense probe are on the right (B). The scale bar is 100 μm in 1A and B, 50 μm in 2A and B.
Figure 6 RNA in situ hybridization of variant Il4i1_2 in different areas of adult mouse brain. Shown are sagittal sections from cerebellum (1), cortex (2), hippocampus (3) and olfactory bulb (4). Probes were transcribed from the joining exons specific for variant Il4i1_2, located between the previously known Nup62 and Il4i1 gene loci. Signals (in blue) obtained with the antisense probe are on the left (A), and those obtained with the sense probe are on the right (B). The scale bar is 500 μm in 2A and 3A, and 200 μm in the other images.
Discussion
We here report a novel transcript variant of the IL4I1 gene, which is a product of two exons from the previously described NUP62 gene, two apparently joining exons mapping between the reported NUP62 and IL4I1 gene loci, and six exons of the known IL4I1 gene. Expression of that variant is driven by the assumed NUP62 gene promoter with high tissue and cell type specificity. The protein encoded by the variant IL4I1_2 transcript is essentially the same as that of the originally described Il4i1 protein [32], since the primary structures of the encoded proteins are identical after probable cleavage of the predicted signal peptides. Although a different functionality of the variant signal peptides cannot be excluded [40], the expression of this otherwise B-cell specific gene in testis and brain already adds significantly to the previously known properties of that gene and the encoded enzyme. The tissue specificity of the reported IL4I1 promoter [31] appears to be essential for survival, and expression of that gene appears to be tightly controlled. Given the function of the encoded protein, a LAAO enzyme, such restriction of protein expression makes sense. Limiting IL4I1 expression to B cells would take reference to the specific function of that cell type (e.g. antigen processing). In contrast, the Il4i1 protein is likely not involved in the immune system/antigen processing when expressed in testis or the brain. While the function of that protein in these tissues thus remains to be established, a possible involvement in disease should be analyzed. The lysyl oxidase (LOX) has been found at elevated levels in amyotrophic lateral sclerosis (ALS) and in superoxide dismutase (mSOD1) knockout mice (which exhibit an ALS-like syndrome) and is believed to be involved in the progression of ALS [41]. The LAAO activity of Il4i1 makes this protein a new candidate not only for ALS, but also for other diseases associated with the death of Purkinje cells [42]. For example, the chromosomal location of the IL4I1 gene at 19q13.31 has been described as candidate region for spinocerebellar ataxia type SCA19. Elevated expression levels of IL4I1 have also been reported in primary mediastinal large B-cell lymphoma [43], thus associating this gene with cancer as well. Further experimentation will be necessary to establish a possible role of the variant IL4I1_2 in any of these or other diseases.
The previously described IL4I1 promoter appears to be strictly specific for B-cell expression. It does not contain a CpG island and is reported to be induced for instance by STAT6 [31]. In contrast, the IL4I1_2 variant in the human is likely to be expressed exclusively in testis and brain. The NUP62 gene has a CpG island and is ubiquitously expressed. In consequence, pre-mRNAs are spliced to produce the novel variant only in testis and brain. However Purkinje and Sertoli cells also require functional nuclear pore complexes to survive. Correct amounts of both mRNAs need to be generated within the cells. The amounts could be regulated most likely at the splicing and/or polyadenylation levels, or by specific mRNA degradation. In consequence, the variant IL4I1_2 transcript is indicative of a so far undetected mechanism of gene regulation. While the presence of alternative promoters is a common theme in many genes, the cell-type specific expression of two genes from one promoter is novel, especially when the transcripts contain exons from both genes.
Thus far, gene fusions had mostly been associated with disease [44]; for example, trans-splicing is associated with viral infection [45]. However, the process reported here occurs in normal individuals and could be essential in the expressing cell types. Apparent joining of genes as indicated by cDNA sequences takes place at a rather high rate, but in many cases these cDNAs are likely to have been the result of errors in the pre-mRNA processing machinery [46]. One example is AK074097, which points to a fusion between IL4I1 and the downstream gene encoding TBC1 domain family member 17. However, these genes are oriented tail to tail, and the sequence structure of AK074097 is not supported by any further cDNA data. AK074097 even extends into the next further downstream gene AKT1S1. The "splice variant" represented by this cDNA therefore most likely originated from the lack of transcriptional termination and mis-splicing of cryptic "exons". This cDNA could thus be regarded as biological noise [46]. While being probably not of functional relevance, this and many other similar cDNA sequences (also IMAGE:5168029) raise questions as to the fidelity of RNA production and processing in cells, and as to the requirement of biological systems to be able to tolerate such events. Since errors at the RNA level are not inherited per se, the observed phenomena presumably are indicative of the flexibility and stability of the cellular system, rather than that these RNAs themselves would contribute to the evolutionary principle directly. Our findings now suggest that promiscuity of the pre-mRNA processing machinery is a required mechanism on a higher than previously reported [5,6,47,48], i.e., a trans-gene level, and that it is regulated at tissue and cell-type levels.
Several questions remain unanswered. Why and how is the pre-mRNA spliced to specifically produce the variant IL4I1_2 mRNA? Is transcription of RNA polymerase past the 3'-terminal exon of NUP62, which is required to join exons from the apparent NUP62 and IL4I1 genes, restricted to the cell types and tissues where the variant is detected, or is the tissue specificity of that variant mRNA determined in the splicing/polyadenylation process? Why is IL4I1 expression driven by the NUP62 promoter at all? More globally, is this a unique mechanism or are there more genes that are driven by the promoters of upstream genes? Are there other cases where an apparently leaky splicing mechanism could be favourable over the risk of erroneous transcription from a more promiscuous promoter? And finally, how did this mechanism evolve? The evolution of this mechanism would have required at least three events to happen, probably in early eutherian development: 1) the installation of neighborhood and orientation of these two genes, 2) the continuation of transcription beyond the NUP62 translated exon and its transcription termination signals, and 3) the development of tissue specificity for NUP62 and IL4I1_2 pre-mRNA processing. Selective pressure appears to have favoured preservation of the status quo. Sequence conservation of those exons that are specific for the variant is even higher than that of exons 1 and 2 of the previously reported IL4I1 transcript. This could hint to an essential function of the variant and to the possibility that it was not IL4I1 that integrated downstream of NUP62, but that instead NUP62 integrated into the IL4I1 gene. The latter hypothesis is supported by the fact that the complete NUP62 ORF is located on one exon in mammals, while it is split into several exons in other eukaryotes. Thus the so-called NUP62 promoter might actually be an ancient IL4I1 promoter that triggered expression of two independent ORFs after integration of a NUP62 retroposon. An immediate question would follow: what happened to the original NUP62 gene in eutherians? The cDNA FLJ20130, which maps to human chromosome Xq22.3 encodes a protein that is homologous to part of Nup62, namely the most conserved nucleoporin Nsp1-like C-terminal domain (IPR007758). That domain is fundamental for interaction with Nup82, another protein of the nuclear pore complex. The exon/intron structure of at least three exons in the FLJ20130 locus is the same as in the chicken NUP62 gene (Fig. 7A). The conservation of FLJ20130 and NUP62 extends into the 3'UTR of FLJ20130, which is however part of the coding region of NUP62 (Fig. 7A). In human, dog, mouse and opossum, the FLJ20130 gene is flanked by CXorf 41 (upstream) and FLJ11016 (downstream) and their orthologs, respectively. The same homologous genes flank the NUP62 gene in chicken in identical order and orientation. FLJ20130 in human Xq22.3 might consequently be a remnant of the ancient NUP62 in mammals, having lost a number of exons and much of its coding region. Other examples of retrogenes have been reported [49]. In contrast to the ubiquitous expression of NUP62, EST data from mouse and human suggest that expression of FLJ20130 is mostly in early development. These findings indicate this probable ancient form of mammalian NUP62 may still be expressed but is likely to have acquired a novel function. Presence of the Nsp1_c like region, however, could implicate this protein to be involved in the nuclear pore complex as well.
Figure 7 Alignment of NUP62 and FLJ20130. (A) Protein sequences of human, opossum and chicken Nup62 were aligned with the peptide encoded by cDNA FLJ20130. While most of the mammalian NUP62 genes do not contain introns, the chicken NUP62 gene and the human gene encoding FLJ20130 do. Exon boundaries in the latter two sequences are indicated with pipes (|), and the lack of introns is indicated at the respective positions by dashes (-). The positions of exon boundaries are additionally highlighted in yellow. The N terminus of chicken Nup62 is probably incomplete. (B) Dot-plot of human nucleotide sequences FLJ20130 (presumed ancestor of NUP62) and AL162061 (NUP62). Windowsize was 55 bp, and stringency was set to 60% sequence identity. Coding regions in the two cDNAs are indicated with blue bars, and ORF boundaries are linked to the dot-plot alignment. The alignment extends past the protein coding sequences. The terminal and coding exon of AL162061 starts at position bp 129 and comprises the assumed retroposon.
The mechanism that determines the processing of NUP62/IL4I1_2 pre-mRNAs into its final form also remains to be identified. An attractive model would be the frequently observed use of alternative polyadenylation sites that is coupled to alternative splicing [50]. Then the terminal exon of NUP62 could be interpreted as "merely" an alternative 3'-end of the IL4I1 gene, or the downstream exons of IL4I1 were alternative ends of the NUP62 gene. A possible NUP62 retroposon would have contained a polyadenylation signal and a polyA-tail. Consensus polyadenylation signals (AAUAAA) are present in the NUP62 gene and transcripts while the polyA tail appears to have vanished since the time of integration. The downstream sequence element needed to make the polyadenylation signal functional [51] and to terminate transcription must have been present within the intron of the IL4I1 gene into which the retroposon inserted.
Conclusion
We have identified and verified a novel mechanism for regulation of gene expression that involves the transcription of two genes from the same promoter and the processing of two variant mRNAs from probably the same pre-mRNA. The encoded proteins are completely unrelated. Conservation of this mechanism in eutherians suggests both transcripts and the encoded proteins are essential for survival. Finally, our finding puts the current definition of the term "gene" in question, as the variants we have identified and analyzed are clearly the product of two genes. In addition to one promoter driving the expression of these genes, two of the formerly named NUP62 exons are also part of the IL4I1_2 variant. Should these exons be counted as belonging to the NUP62 or to the IL4I1 genes? One current definition of a gene is "a complete chromosomal segment responsible making a functional product" [52]. The chromosomal segment encoding the B-cell variant of IL4I1 appears completely separate from that of NUP62 and thus fulfils all criteria of the above definition. This is not true however for the newly detected IL4I1_2 variant. NUP62 and IL4I1_2 share noncoding regulatory DNA sequences, exons and introns within one chromosomal segment. The functional sequences of NUP62 and IL4I1_2, however, are unique and distinct, which is another criterion used to separate two genes. In consequence, the above definition of a "gene" should be put in question. Nature may have more surprises to reveal, and with increasing amounts of data on genomes, transcriptomes and proteomes being collected and analyzed, other paradigms may require revision.
Methods
Identification of splice variant
The cDNA IMAGE:4822638 (Acc:BC026103) was cloned and sequenced by the Mammalian Gene Collection [26]. More cDNAs were identified in the University of California, Santa Cruz (UCSC) genome browser [53,54] (assembly of May 2004), based on their EST sequences to cover part of the IL4I1_2 variant (IMAGE:5168029, IMAGE:5171014, IMAGE: 5742307, IMAGE:4838597). All these cDNAs were obtained from The German Resource Centre for Genome Research (RZPD; [55]) and completely sequenced with help of walking primers [56]. Sequences were assembled and aligned using the Staden package [57] to identify base substitutions and other alterations from the predicted consensus sequence.
Comparative genomic analysis
Comparative genomic analysis of the IL4I1_2 variant was done with help of the UCSC genome browser [53], which indicated variant cDNAs from mouse [58] (ESTs Acc:BY100275, BY099330, BY087056, BY092834, BY088421) and rat (Acc:CV117152). Alignment of protein sequences was done with Vector NTI software (Invitrogen). Synteny of genomic regions downstream of the NUP62 orthologs was analyzed in the genome assemblies and datasets of human (hg17), chimpanzee (panTro1), dog (canFam1), mouse (mm5), rat (rn3), opossum (monDom1), chicken (galGal2), Fugu (fr1), and Drosophila (dm1), all in the UCSC genome browser [54].
Northern hybridization
Multiple tissue Northern blots with poly-(A)+-RNA from mouse embryonic (Cat.# 636810) and mouse adult tissues (Cat.# 636808) were obtained from BD Biosciences Clontech. A probe specific for the mouse variant Il4i1_2 transcript was generated with the primers mmNupIlR1 (GAAGAACACAGGCAGATGCCCTG) and mmNupIlS1 (TGCATGGTGGTCTTTGTGGGGC), which were used to amplify the mouse joining exons 2 and 3 of the variant Il4i1_2 (equivalent to the human exons 3 and 4 indicated with red arrowheads in Fig. 1) from mouse testis RNA via RT-PCR. The 208 bp PCR product was cloned into the pCRII vector (Invitrogen), and sequence verified. Filters were hybridized with 32P-labelled purified PCR products from that clone. Hybridization was overnight in Church solution (1M Na2HPO4, 1M NaH2PO4·H2O, 10mM EDTA, pH8.0) at 65°C. Filters were washed once in 0.1% SDS/0.1xSSC for 10 min, once in 0.1% SDS/0.3xSSC for 10 min, and then exposed to Kodak Bio Max at -80°C.
RNA in situ hybridization
RNA in situ hybridization was performed on embryo sections at stages 10.5, 12.5, 14.5, 16.5 and different tissues of adult mice (testis, kidney, liver and brain). Embryos were isolated from pregnant NMRI mice. The day of plug detection was considered to be day 0.5 post conception (dpc). The tissues and embryonic stages were fixed over night in 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS) at 4°C. The tissues from adult NMRI mice were isolated after perfundation with 4% PFA in PBS. After embedding in paraffin, 6 μm sagittal sections were mounted on Superfrost+ slides. Cloned PCR products (see Northern hybridization) were sequence verified to identify orientation of the product within the vector. Antisense (T7) and sense (SP6) riboprobes labeled with digoxigenin-UTP (Enzo) were generated by in vitro transcription (Roche), after linearization of the constructs. Pre-treatment, hybridization and washing were carried out using a Ventana discovery system. Sense or antisense RNA probes were hybridized at 100ng RNA/ml in hybridization buffer in a volume of 100 μl/slide. Slides were analyzed using a Leica microscope.
Photographs were taken with a liquid crystal display (LCD) – camera (Power head, Sony) using AnalySIS software (Soft imaging System GmbH). The figures were assembled using Adobe Photoshop.
Authors' contributions
SW designed the study, carried out the sequence analysis and drafted the manuscript. AKK carried out the experimental research and helped to draft the manuscript. AP participated in study design and coordination. All authors read and approved the final manuscript.
Acknowledgements
We thank Jeremy Simpson for protein localization, and Ute Ernst and Hanna Bausbacher for excellent technical assistance. We thank Danielle and Jean Thierry-Mieg for interesting discussions and productive suggestions. This work was supported by the German Federal Ministry of Education and Research (BMBF) with grants 01GR0101 and 01GR0420.
==== Refs
Brett D Pospisil H Valcarcel J Reich J Bork P Alternative splicing and genome complexity Nat Genet 2002 30 29 30 11743582 10.1038/ng803
Modrek B Lee C A genomic view of alternative splicing Nat Genet 2002 30 13 19 11753382 10.1038/ng0102-13
Ast G How did alternative splicing evolve? Nat Rev Genet 2004 5 773 782 15510168 10.1038/nrg1451
Imanishi T Itoh T Suzuki Y O'Donovan C Fukuchi S Koyanagi KO Barrero RA Tamura T Yamaguchi-Kabata Y Tanino M Yura K Miyazaki S Ikeo K Homma K Kasprzyk A Nishikawa T Hirakawa M Thierry-Mieg J Thierry-Mieg D Ashurst J Jia L Nakao M Thomas MM Mulder N Karavidopoulou Y Jin L Kim S Yasuda T Lenhard B Eveno E Suzuki Y Yamasaki C Takeda J Gough C Amid C Bellgard M de Fatima Bonaldo M Bono H Bromberg SK Brookes A Bruford E Carninci P Chelala C Couillault C de Souza S Debily M Devignes M Dubchak I Endo T Estreicher A Eyras E Fukami-Kobayashi K Gopinathrao G Graudens E Hahn Y Han M Han Z Hanada K Hashimoto K Hinz U Hirai M Hishiki T Hopkinson I Imbeaud S Inoko H Kanapin A Kasukawa T Kelso J Kersey P Kikuno R Kimura K Korn B Kuryshev V Makalowska I Makalowski W Makino T Mano S Mariage-Samson R Mashima J Matsuda H Mewes HW Minoshima S Nagai K Nagasaki H Nigam R Ogasawara O Ohara O Ohtsubo M Okada N Okido T OOta S Ota M Ota T Otsuki T Piatier-Tonneau D Poustka A Ren S Saitou N Sakai K Sakamoto S Sakate R Schupp I Servant F Sherry S Shimizu N Shimoyama M Simpson AJ Soares B Steward C Suwa M Suzuki M Takahashi A Tamiya G Tanaka H Taylor T Terwilliger JD Unneberg P Watanabe S Wilming L Yasuda N Yoo H Veeramachaneni V Stodolsky M Go M Nakai K Takagi T Kanehisa M Sakaki Y Quackenbush J Okazaki Y Hayashizaki Y Hide W Chakraborty R Nishikawa K Sugawara H Tateno Y Chen Z Oishi M Tonellato P Apweiler R Okubo K Wagner L Wiemann S Strausberg RL Isogai T Auffray C Nomura N Gojobori T Sugano S Integrative Annotation of 21,037 Human Genes Validated by Full-Length cDNA Clones PLoS Biol 2004 2 856 875 15103394 10.1371/journal.pbio.0020162
Kornblihtt AR de la Mata M Fededa JP Munoz MJ Nogues G Multiple links between transcription and splicing RNA 2004 10 1489 1498 15383674 10.1261/rna.7100104
Kalnina Z Zayakin P Silina K Line A Alterations of pre-mRNA splicing in cancer Genes Chromosomes Cancer 2005 42 342 357 15648050 10.1002/gcc.20156
Wheeler DL Church DM Edgar R Federhen S Helmberg W Madden TL Pontius JU Schuler GD Schriml LM Sequeira E Suzek TO Tatusova TA Wagner L Database resources of the National Center for Biotechnology Information: update Nucleic Acids Res 2004 32 D35 40 14681353 10.1093/nar/gkh073
Anderson CL Zundel MA Werner R Variable promoter usage and alternative splicing in five mouse connexin genes Genomics 2005 85 238 244 15676282 10.1016/j.ygeno.2004.11.007
George CX Wagner MV Samuel CE Expression of interferon-inducible RNA deaminase ADAR1 during pathogen infection and mouse embryo development involves tissue selective promoter utilization and alternative splicing J Biol Chem 2005 280 15020 15028 15677478 10.1074/jbc.M500476200
Gebauer F Hentze MW Molecular mechanisms of translational control Nat Rev Mol Cell Biol 2004 5 827 835 15459663 10.1038/nrm1488
de Arriba Zerpa GA Saleh MC Fernandez PM Guillou F Espinosa de los Monteros A de Vellis J Zakin MM Baron B Alternative splicing prevents transferrin secretion during differentiation of a human oligodendrocyte cell line J Neurosci Res 2000 61 388 395 10931525 10.1002/1097-4547(20000815)61:4<388::AID-JNR5>3.0.CO;2-Q
Stamm S Ben-Ari S Rafalska I Tang Y Zhang Z Toiber D Thanaraj TA Soreq H Function of alternative splicing Gene 2005 344 1 20 15656968 10.1016/j.gene.2004.10.022
Stoecklin G Ming XF Looser R Moroni C Somatic mRNA turnover mutants implicate tristetraprolin in the interleukin-3 mRNA degradation pathway Mol Cell Biol 2000 20 3753 3763 10805719 10.1128/MCB.20.11.3753-3763.2000
Zhang H Hu J Recce M Tian B PolyA_DB: a database for mammalian mRNA polyadenylation Nucleic Acids Res 2005 33 D116 120 15608159 10.1093/nar/gki055
Hirano M Noda T Genomic organization of the mouse Msh4 gene producing bicistronic, chimeric and antisense mRNA Gene 2004 342 165 177 15527976 10.1016/j.gene.2004.08.016
Stallmeyer B Drugeon G Reiss J Haenni AL Mendel RR Human molybdopterin synthase gene: identification of a bicistronic transcript with overlapping reading frames Am J Hum Genet 1999 64 698 705 10053003 10.1086/302295
Veeramachaneni V Makalowski W Galdzicki M Sood R Makalowska I Mammalian overlapping genes: the comparative perspective Genome Res 2004 14 280 286 14762064 10.1101/gr.1590904
Makalowska I Lin CF Makalowski W Overlapping genes in vertebrate genomes Comput Biol Chem 2005 29 1 12 15680581 10.1016/j.compbiolchem.2004.12.006
Poulin F Brueschke A Sonenberg N Gene fusion and overlapping reading frames in the mammalian genes for 4E-BP3 and MASK J Biol Chem 2003 278 52290 52297 14557257 10.1074/jbc.M310761200
Alfonso A Grundahl K McManus JR Asbury JM Rand JB Alternative splicing leads to two cholinergic proteins in Caenorhabditis elegans J Mol Biol 1994 241 627 630 8057385 10.1006/jmbi.1994.1538
Gustin KE Sarnow P Effects of poliovirus infection on nucleo-cytoplasmic trafficking and nuclear pore complex composition EMBO J 2001 20 240 249 11226174 10.1093/emboj/20.1.240
Zhong H Takeda A Nazari R Shio H Blobel G Yaseen NR Carrier-independent nuclear import of the transcription factor PU.1 via RANGTP-stimulated binding to NUP153 J Biol Chem 2005 280 10675 10682 15632149 10.1074/jbc.M412878200
Swissprot Proteomics Server
Carmo-Fonseca M Kern H Hurt EC Human nucleoporin p62 and the essential yeast nuclear pore protein NSP1 show sequence homology and a similar domain organization Eur J Cell Biol 1991 55 17 30 1915414
Mans BJ Anantharaman V Aravind L Koonin EV Comparative Genomics, Evolution and Origins of the Nuclear Envelope and Nuclear Pore Complex Cell Cycle 2004 3 1612 1637 15611647
Strausberg RL Feingold EA Grouse LH Derge JG Klausner RD Collins FS Wagner L Shenmen CM Schuler GD Altschul SF Zeeberg B Buetow KH Schaefer CF Bhat NK Hopkins RF Jordan H Moore T Max SI Wang J Hsieh F Diatchenko L Marusina K Farmer AA Rubin GM Hong L Stapleton M Soares MB Bonaldo MF Casavant TL Scheetz TE Brownstein MJ Usdin TB Toshiyuki S Carninci P Prange C Raha SS Loquellano NA Peters GJ Abramson RD Mullahy SJ Bosak SA McEwan PJ McKernan KJ Malek JA Gunaratne PH Richards S Worley KC Hale S Garcia AM Gay LJ Hulyk SW Villalon DK Muzny DM Sodergren EJ Lu X Gibbs RA Fahey J Helton E Ketteman M Madan A Rodrigues S Sanchez A Whiting M Young AC Shevchenko Y Bouffard GG Blakesley RW Touchman JW Green ED Dickson MC Rodriguez AC Grimwood J Schmutz J Myers RM Butterfield YS Krzywinski MI Skalska U Smailus DE Schnerch A Schein JE Jones SJ Marra MA Generation and initial analysis of more than 15,000 full-length human and mouse cDNA sequences Proc Natl Acad Sci USA 2002 99 16899 16903 12477932 10.1073/pnas.242603899
Wiemann S Weil B Wellenreuther R Gassenhuber J Glassl S Ansorge W Bocher M Blocker H Bauersachs S Blum H Lauber J Dusterhoft A Beyer A Kohrer K Strack N Mewes HW Ottenwalder B Obermaier B Tampe J Heubner D Wambutt R Korn B Klein M Poustka A Toward a Catalog of Human Genes and Proteins: Sequencing and Analysis of 500 Novel Complete Protein Coding Human cDNAs Genome Res 2001 11 422 435 11230166 10.1101/gr.GR1547R
Chu CC Paul WE Fig1, an interleukin 4-induced mouse B cell gene isolated by cDNA representational difference analysis Proc Natl Acad Sci USA 1997 94 2507 2512 9122225 10.1073/pnas.94.6.2507
Chu CC Paul WE Expressed genes in interleukin-4 treated B cells identified by cDNA representational difference analysis Mol Immunol 1998 35 487 502 9798653 10.1016/S0161-5890(98)00031-5
Mason JM Naidu MD Barcia M Porti D Chavan SS Chu CC IL-4-induced gene-1 is a leukocyte L-amino acid oxidase with an unusual acidic pH preference and lysosomal localization J Immunol 2004 173 4561 4567 15383589
Schroder AJ Pavlidis P Arimura A Capece D Rothman PB Cutting edge: STAT6 serves as a positive and negative regulator of gene expression in IL-4-stimulated B lymphocytes J Immunol 2002 168 996 1000 11801631
Chavan SS Tian W Hsueh K Jawaheer D Gregersen PK Chu CC Characterization of the human homolog of the IL-4 induced gene-1 (Fig1) Biochim Biophys Acta 2002 1576 70 80 12031486
Kozak M Initiation of translation in prokaryotes and eukaryotes Gene 1999 234 187 208 10395892 10.1016/S0378-1119(99)00210-3
Bendtsen JD Nielsen H von Heijne G Brunak S Improved prediction of signal peptides: SignalP 3.0 J Mol Biol 2004 340 783 795 15223320 10.1016/j.jmb.2004.05.028
Oyama M Itagaki C Hata H Suzuki Y Izumi T Natsume T Isobe T Sugano S Analysis of small human proteins reveals the translation of upstream open reading frames of mRNAs Genome Res 2004 14 2048 2052 15489325 10.1101/gr.2384604
Fernandez J Yaman I Huang C Liu H Lopez AB Komar AA Caprara MG Merrick WC Snider MD Kaufman RJ Lamers WH Hatzoglou M Ribosome Stalling Regulates IRES-Mediated Translation in Eukaryotes, a Parallel to Prokaryotic Attenuation Mol Cell 2005 17 405 416 15694341 10.1016/j.molcel.2004.12.024
Simpson JC Wellenreuther R Poustka A Pepperkok R Wiemann S Systematic subcellular localization of novel proteins identified by large scale cDNA sequencing EMBO Rep 2000 1 287 292 11256614 10.1093/embo-reports/kvd058
Bannasch D Mehrle A Glatting K-H Pepperkok R Poustka A Wiemann S LIFEdb: A database for functional genomics experiments integrating information from external sources, and serving as a sample tracking system Nucleic Acids Res 2004 32 D505 508 14681468 10.1093/nar/gkh022
DKFZ Database for Localization, Interaction, Functional assays and Expression of human Proteins
Martoglio B Intramembrane proteolysis and post-targeting functions of signal peptides Biochem Soc Trans 2003 31 1243 1247 14641034
Li PA He Q Cao T Yong G Szauter KM Fong KS Karlsson J Keep MF Csiszar K Up-regulation and altered distribution of lysyl oxidase in the central nervous system of mutant SOD1 transgenic mouse model of amyotrophic lateral sclerosis Mol Brain Res 2004 120 115 122 14741400 10.1016/j.molbrainres.2003.10.013
Sarna JR Hawkes R Patterned Purkinje cell death in the cerebellum Prog Neurobiol 2003 70 473 507 14568361
Copie-Bergman C Boulland ML Dehoulle C Moller P Farcet JP Dyer MJ Haioun C Romeo PH Gaulard P Leroy K Interleukin 4-induced gene 1 is activated in primary mediastinal large B-cell lymphoma Blood 2003 101 2756 2761 12446450 10.1182/blood-2002-07-2215
Hahn Y Bera TK Gehlhaus K Kirsch IR Pastan IH Lee B Finding fusion genes resulting from chromosome rearrangement by analyzing the expressed sequence databases Proc Natl Acad Sci USA 2004 101 13257 13261 15326299 10.1073/pnas.0405490101
Caudevilla C Da Silva-Azevedo L Berg B Guhl E Graessmann M Graessmann A Heterologous HIV-nef mRNA trans-splicing: a new principle how mammalian cells generate hybrid mRNA and protein molecules FEBS Lett 2001 507 269 279 11696354 10.1016/S0014-5793(01)02957-X
Lareau LF Green RE Bhatnagar RS Brenner SE The evolving roles of alternative splicing Curr Opin Struct Biol 2004 14 273 282 15193306 10.1016/j.sbi.2004.05.002
Nogues G Kadener S Cramer P de la Mata M Fededa JP Blaustein M Srebrow A Kornblihtt AR Control of alternative pre-mRNA splicing by RNA Pol II elongation: faster is not always better IUBMB Life 2003 55 235 241 12880204
Shomron N Alberstein M Reznik M Ast G Stress alters the subcellular distribution of hSlu7 and thus modulates alternative splicing J Cell Sci 2005 118 1151 1159 15728250 10.1242/jcs.01720
Wang PJ X chromosomes, retrogenes and their role in male reproduction Trends Endocrinol Metab 2004 15 79 83 15036254 10.1016/j.tem.2004.01.007
Proudfoot NJ Furger A Dye MJ Integrating mRNA processing with transcription Cell 2002 108 501 512 11909521 10.1016/S0092-8674(02)00617-7
Proudfoot N New perspectives on connecting messenger RNA 3' end formation to transcription Curr Opin Cell Biol 2004 16 272 278 15145351 10.1016/j.ceb.2004.03.007
Snyder M Gerstein M Genomics. Defining genes in the genomics era Science 2003 300 258 260 12690176 10.1126/science.1084354
Kent WJ Sugnet CW Furey TS Roskin KM Pringle TH Zahler AM Haussler D The Human Genome Browser at UCSC Genome Res 2002 12 996 1006 12045153 10.1101/gr.229102. Article published online before print in May 2002
UCSC Genome Browser
Deutsches Ressourcenzentrum für Genomforschung
Haas S Vingron M Poustka A Wiemann S Primer design for large scale sequencing Nucleic Acids Res 1998 26 3006 3012 9611248 10.1093/nar/26.12.3006
Staden R The Staden sequence analysis package Mol Biotechnol 1996 5 233 241 8837029
Okazaki Y Furuno M Kasukawa T Adachi J Bono H Kondo S Nikaido I Osato N Saito R Suzuki H Yamanaka I Kiyosawa H Yagi K Tomaru Y Hasegawa Y Nogami A Schonbach C Gojobori T Baldarelli R Hill DP Bult C Hume DA Quackenbush J Schriml LM Kanapin A Matsuda H Batalov S Beisel KW Blake JA Bradt D Brusic V Chothia C Corbani LE Cousins S Dalla E Dragani TA Fletcher CF Forrest A Frazer KS Gaasterland T Gariboldi M Gissi C Godzik A Gough J Grimmond S Gustincich S Hirokawa N Jackson IJ Jarvis ED Kanai A Kawaji H Kawasawa Y Kedzierski RM King BL Konagaya A Kurochkin IV Lee Y Lenhard B Lyons PA Maglott DR Maltais L Marchionni L McKenzie L Miki H Nagashima T Numata K Okido T Pavan WJ Pertea G Pesole G Petrovsky N Pillai R Pontius JU Qi D Ramachandran S Ravasi T Reed JC Reed DJ Reid J Ring BZ Ringwald M Sandelin A Schneider C Semple CA Setou M Shimada K Sultana R Takenaka Y Taylor MS Teasdale RD Tomita M Verardo R Wagner L Wahlestedt C Wang Y Watanabe Y Wells C Wilming LG Wynshaw-Boris A Yanagisawa M Yang I Yang L Yuan Z Zavolan M Zhu Y Zimmer A Carninci P Hayatsu N Hirozane-Kishikawa T Konno H Nakamura M Sakazume N Sato K Shiraki T Waki K Kawai J Aizawa K Arakawa T Fukuda S Hara A Hashizume W Imotani K Ishii Y Itoh M Kagawa I Miyazaki A Sakai K Sasaki D Shibata K Shinagawa A Yasunishi A Yoshino M Waterston R Lander ES Rogers J Birney E Hayashizaki Y Analysis of the mouse transcriptome based on functional annotation of 60,770 full-length cDNAs Nature 2002 420 563 573 12466851 10.1038/nature01266
|
16029492
|
PMC1198218
|
CC BY
|
2021-01-04 16:39:15
|
no
|
BMC Biol. 2005 Jul 19; 3:16
|
utf-8
|
BMC Biol
| 2,005 |
10.1186/1741-7007-3-16
|
oa_comm
|
==== Front
BMC Blood DisordBMC Blood Disorders1471-2326BioMed Central London 1471-2326-5-41604280310.1186/1471-2326-5-4Research ArticleL-Glutamine therapy reduces endothelial adhesion of sickle red blood cells to human umbilical vein endothelial cells Niihara Yutaka [email protected] Neil M [email protected] Yamin M [email protected] Dean A [email protected] Cage S [email protected] M Alenor [email protected] John [email protected] Stephen H [email protected] Vijay K [email protected] Cho Seong [email protected] Kouichi R [email protected] Department of Medicine, Harbor-UCLA Medical Center, UCLA School of Medicine, Torrance, CA USA2 Department of Pediatrics, Harbor-UCLA Medical Center, UCLA School of Medicine, Torrance, CA USA3 Department of Medicine, San Francisco General Hospital, UCSF School of Medicine, San Francisco, CA USA4 Department of Biochemistry, USC Keck School of Medicine, Los Angeles CA USA5 Medicine/Hematology, USC Keck School of Medicine, Los Angeles CA USA2005 25 7 2005 5 4 4 2 11 2004 25 7 2005 Copyright © 2005 Niihara et al; licensee BioMed Central Ltd.2005Niihara 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 have previously demonstrated that therapy with orally administered L-glutamine improves nicotinamide adenosine dinucleotide (NAD) redox potential of sickle red blood cells (RBC). On further analysis of L-glutamine therapy for sickle cell anemia patients, the effect of L-glutamine on adhesion of sickle RBC to human umbilical vein endothelial cells (HUVEC) was examined.
Methods
The first part of the experiment was conducted with the blood samples of the 5 adult sickle cell anemia patients who had been on L-glutamine therapy for at least 4 weeks on a dosage of 30 grams per day compared to those of patient control group. In the second part of the experiment 6 patients with sickle cell anemia were studied longitudinally. Five of these patients were treated with oral L-glutamine 30 grams daily and one was observed without treatment as the control. t-test and paired t-test were used for determination of statistical significance in cross-sectional and longitudinal studies respectively.
Results
In the first study, the mean adhesion to endothelial cells with the autologous plasma incubated cells were 0.97 ± 0.45 for the treated group and 1.91 ± 0.53 for the nontreated group (p < 0.02). Similarly with lipopolysaccharide (LPS) incubated cells the mean adhesion to endothelial cells were 1.39 ± 0.33 for the treated group and 2.80 ± 0.47 for the untreated group (p < 0.001). With the longitudinal experiment, mean decrease in the adhesion to endothelial cells was 1.13 ± 0.21 (p < 0.001) for the 5 treated patients whereas the control patient had slight increase in the adhesion to endothelial cells.
Conclusion
In these studies, oral L-glutamine administration consistently resulted in improvement of sickle RBC adhesion to HUVEC. These data suggest positive physiological effects of L-glutamine in sickle cell disease.
==== Body
Background
Sickle cell disease is a devastating hereditary disorder with excruciating morbidities often resulting in the premature demise of the patients. This disease affects primarily those of African descent, although other ethnic groups such as the Hispanic population on American continents are also affected with relatively high incidence [1-3].
Incapacitating complications of sickle cell disease are usually due to severe anemia and frequent vasoocclusive processes which damage tissues [1-6]. The cause of these events is attributed largely to increased adherence of sickle red blood cells to vascular endothelium[4,7,8]. With alteration in adherence to endothelial cells, sickle RBC will have a longer transit time through capillaries which leads to vasoocclusion with accumulation of less deformable deoxygenated sickle RBC [7-9]. Thus, amelioration of pathologic RBC adherence is considered to be an important target in therapy of sickle cell disease.
In the current article, the effect of L-glutamine therapy on sickle RBC adherence to HUVEC cells is presented. L-glutamine is a precursor to NAD[10,11]. Previous studies have shown that oral L-glutamine therapy improves NAD redox potential of sickle RBC[10]. Also, among those patients treated with L-glutamine, there were subjective reports of improvement in clinical paramenters such as chronic pain and energy level[10].
The study presented here was conducted in an attempt to better understand mechanism by which L-glutamine may potentially exert a beneficial effect on sickle RBC. In the cohort that we have studied, oral L-glutamine therapy consistently led to reduced adhesion of sickle RBC to HUVEC in static assays with both cross-sectional and longitudinal studies.
Methods
Cross-sectional study
Reagents
LPS (lipopolysaccharide) preparation (Escherichia coli 0111:B4), Hanks' buffered saline solution, bovine serum albumin (BSA), HEPES buffer and all other chemicals were from Sigma Chemicals Co (St Louis, MO). Six-wells tissue culture plates and Falcon Primaria tissue culture flasks were from Fisher scientific (Pittsburgh, PA). Sodium-51Chromate (51Cr) and Aquasol scintillation fluid were obtained from NEN (Dupont NEN, Wilmington, DE).
Cell cultures
HUVEC were harvested from umbilical cord veins by collagenase digestion, as previously described[12]. Endothelial cells were identified by their cobblestone morphology, immunofluorescence staining with factor VIII-related antigen and uptake of diacetylated low-density lipoprotein (Biomedical Technologies, Stoughton, MA). Cells were passaged by trypsinization with 1% trypsin-EDTA (GIBCO BRL, Grand Island, NY) every 4–5 days. HUVEC were used from passages 2 to 6.
Patients, control and blood samples
Blood samples were drawn from homozygous sickle cell patients living in the Los Angeles area or from volunteer control subjects after obtaining informed consent as approved by the Harbor-UCLA Committee on Human Research. All subjects were in steady state condition without acute conditions including painful crisis. Nine adult sickle cell anemia patients 18 years and older participated. The treatment group consisted of 5 patients (mean age of 38.8 ± 7.8 with range of 30 to 47) who had been on L-glutamine therapy continuously for 8 weeks or longer at the dosage of 30 grams orally per day. The sickle cell control group consisted of 4 patients (mean age of 29.3 ± 9.0 with range of 20 to 37) who had not been on L-glutamine or any other anti-sickling therapy for at least a year. One of the patients in the non-treatment group participated at two separate time points as sickle cell control. Normal control samples were drawn from healthy volunteers.
51Cr labeling of RBC
Five to 7 ml of heparinized blood from normal donors and sickle cell subjects were centrifuged at 450 × g for 20 min and plasma was saved in a refrigerator and the buffy coat (leukocytes and platelets) was discarded. The pellet was washed three times in Hanks' buffered saline solution with centrifugation at 450 × g for 5 minutes and then RBC were suspended to 20% hematocrit in 5 ml HBSS (Hanks' buffered saline solution). RBC in HBSS were incubated with 10 μCi/ml 51Cr for 1 hr at 37°C with 5% CO2.
Static assay for the adherence of RBC to cultured HUVEC
51Cr-sodium chromate-labeled RBC were washed three times with 10 vol of HBSS and used at a hematocrit of 2% in HBSS. HUVEC grown to confluence in 6-well plates were washed twice with HBSS (3 ml) followed by the addition of 51Cr-labeled RBC (hct 2%) and HBSS to a final volume of 1 ml. Five percent autologous platelet-free plasma (3 wells) or 100 ng/ml of LPS (3 wells) was added and the contents were incubated for 45 minutes at 37°C with 5% CO2. Nonadherent RBC were removed by aspiration and monolayers were washed three times with 1 mL of HBSS containing 0.5% BSA.200 ul lysis buffer was added to each well and lysis solution was transferred to scintillation vial. 5 μl RBC was added to 200 μl lysis buffer and the solution was used as the baseline value. Two μl of scintillation solution were added to each vial and the radioactivity in the cell monolayer-containing adherent RBC was determined and the percentage of RBC adherent was calculated[8,13].
Statistical analysis
The ratio of percent adhesion of patient samples to that of normal control was obtained for each sample. The statistical significance of difference of the ratio between the treatment groups was determined using t-test.
Longitudinal study
Reagents
Bovine serum albumin, Hanks' buffered saline solution, calcium/magnesium-free phosphate-buffered saline, electron microscopic grade glutaraldehyde, and HEPES buffer were from Sigma Chemicals Co (St Louis, MO). Human umbilical vein endothelial cells, subculture reagent package (Trypsin EDTA, HEPES Buffered Saline Solution, Trypsin Neutralizing Solution) and endothelial cell growth medium (EGM) including all supplements and growth factors were from Clonetics Corp (San Diego, CA). Two-chambered culture slides, Falcon Primaria tissue culture flasks were from Fisher Scientific (Pittsburgh, PA).
Cell culture
Cells were grown at 37°C in 5% CO2/95% air in EGM at 85% confluence; the cells were subcultured to 2 chambered slides. The second to the fourth passages were used for adherence assays no later than 24 hours after HUVEC had reached confluence.
Patients, control and blood samples
Blood samples were drawn from homozygous sickle cell patients of the Harbor-UCLA Medical Center or from volunteer control subjects after obtaining informed consent as approved by the Harbor-UCLA Committee on Human Research. All subjects were in steady state condition, none had received blood transfusions within 4 months, and none had history of treatment with hydroxyurea. Six adult sickle cell anemia patients 18 years of age and older participated. Five patients (mean age 40.0 ± 16.4 with range of 19 to 57) were treated for 4 to 8 weeks with oral L-glutamine at dosage of 30 grams a day. One patient (49 years old) who served as sickle cell control had no treatment. The whole blood samples were drawn at baseline which was within 4 weeks of initiation of therapy and then after 4 to 8 weeks of therapy with L-glutamine for the 5 patients. The blood samples of untreated sickle cell control patient were drawn along with the treatment patients at baseline and then 8 weeks later. For each assay, normal control blood was also drawn from normal volunteers.
Gravity adherence assays
This method of assaying static adherence of RBC to HUVEC monolayers was adapted from a published method[14]. Heparinized blood from normal donors and sickle cell subjects was centrifuged at 5000 × g for 5 minutes and plasma was saved in the refrigerator and buffy coat (leukocyte and platelets) was discarded. RBC were washed three times in Hanks' buffered saline solution, centrifuged at 1000 × g for 5 minutes. RBC were then centrifuged at 5000 × g and suspended to 3.5% hematocrit in HAH buffer (Hanks' buffered saline solution, 1% BSA, 50 mmol/L HEPES pH 7.4) containing 17% autologous platelet-free plasma from the heparinized blood sample. Confluent HUVEC monolayers were washed with HAH buffer to remove traces of plasma. After washing with HAH buffer, each well was filled with 0.8 ml HAH and one well was treated with LPS (200 ng/ml) and incubated at 37 C, 5%CO2, 95% humidity for 1 hr. After treating cells with LPS for 1 hr, the monolayers were covered with sickle cell (SS) RBC suspensions and incubated at 37°C for 25 minutes. The wells were then filled completely with HAH, sealed with packing tape, and inverted at 37°C for 20 minutes. While still inverted, the well walls and gaskets of the slide chambers were removed. The slides were rinsed twice with HAH buffer to remove nonadherent RBC, fixed in 3% glutaraldehyde, stained, and mounted. RBC adherence was monitored visually by microscopy at 100× magnification and quantified by counting RBC adherent to HUVEC monolayers in 16 fields marked by a grid (same random fields for each sample). The mean adherence and standard error of mean (SEM) for each sample were calculated and results were used to determine the reported means from at least four experiments, only when the SEM of each sample was less than 20% of the means. Results were analyzed by the Student's t-test, with which p < 0.05 was considered significant.
Statistical analysis
The ratio of mean endothelial adhesion RBC of patient samples to that of normal control was obtained for each sample. The statistical significance of difference of the ratio between the baseline and after treatment was determined using paired t-test.
Results
Cross-sectional study
The mean ratios with assays using autologous plasma and LPS incubated cells were 0.97 ± 0.45 and 1.39 ± 0.33, respectively for the L-glutamine treated group (N = 5). For the untreated group the mean ratios were 1.91 ± 0.53 and 2.80 ± 0.47 for autologous plasma and LPS incubated cells respectively (N = 4 with one of the patients studied twice at two time points) The decreased adhesion to endothelial cells by the L-glutamine treated group were statistically significant in both assays with autologous plasma incubated cells (p < 0.02) and LPS incubated cells (p < 0.001) (Figures 1 and 2). The endothelial adhesion of RBC from non-treated sickle cell patients were as expected and several fold higher than that of normal control. On the other hand, the results of the treatment group were consistently similar to that of the normal control, thus resulting in ratio of close to 1 for both assays using LPS or autologous plasma only. Because of the relatively small size of cohorts, there was a trend for age difference in the ages of the two groups, mean age of 38.8 ± 7.8 with range of 30 to 47 for the treatment group and 29.3 ± 9.0 with range of 20 to 37 for the non-treatment group. However, there was no statistical significance with p > 0.1.
Figure 1 Ratios of HUVEC adhesion rates of sickle RBC to those of normal RBC for the treatment group (0.97 ± 0.45) and non-treatment group (1.91 ± 0.53) when the cells were incubated with autologous plasma alone. (p < 0.02)
Figure 2 Ratios of HUVEC adhesion rates of sickle RBC to those of normal RBC for the treatment group (1.39 ± 0.33) and non-treatment group (2.80 ± 0.47) when the cells were incubated with autologous plasma and LPS. (p < 0.001)
Longitudinal study
In contrast to the cross-sectional study, for this portion of experiment, each patient became his/her own reference and provided baseline value. The results were consistent with the data of cross-sectional study noted above. Compared to the baseline value, the endothelial adhesion of sickle RBC decreased in each patient (N = 5) within 4 to 8 weeks of oral L-glutamine therapy. The mean decrease was 1.13 ± 0.21 with p < 0.001. For the patient who participated as sickle cell control, the sickle RBC adhesion to endothelial cells increased slightly compared to the baseline (Figure 3).
Figure 3 Ratios of HUVEC adhesion rates of sickle RBC to those of normal RBC before and after treatment for the five sickle cell anemia patients with one sickle cell anemia control patient who was not treated. Mean decrease of the adhesion ratios was 1.13 ± 0.21 with p < 0.001 for the five treated patients, whereas the control patient actually had slight increase in the adhesion ratio over the same period.
Discussion
The vaso-occlusive process is mediated primarily by congestion and closure of the microvasculature with rigid sickle RBC[4,8,15]. It is well described that the oxygenated sickle RBC is relatively deformable but it starts to become less so upon unloading of oxygen molecules to tissue[16]. The duration of time required for sickle RBC to become less deformable is thought to be more than the usual capillary transit time for RBC[7,9]. Therefore, had it not been for the prolonged capillary transit time, the sickle RBC would have already exited the microvascular system before it becomes less deformable and trapped in the microvasculature. Thus, factors that affect the capillary transit time become crucial in initiation and propagation of vaso-occlusive event.
When Hebbel and colleagues first reported an increase in endothelial adhesion of sickle RBC two decades ago, one of the major factors involved in the pathophysiology of vasoocclusion started to unfold[8]. Since then, several endothelial adhesion molecules which are abnormally expressed in both RBC and endothelial cells of sickle cell disease patients have been described[7,14,17,18]. Also, involvement of leukocytes, platelets and inflammation in general with endothelial adhesion has been described[4,19,20].
In the study presented in this article, endothelial adhesion of sickle RBC was used as the primary end point to evaluate the efficacy of L-glutamine therapy. Although clinical outcome measurements, such as the incidence of sickle cell crises or mortality rate cannot be replaced for endothelial adhesion of sickle RBC, it may be a reasonable surrogate in predicting clinical efficacy of an experimental therapeutic modality.
A role for L-glutamine in sickle cell disease was conceived when Zerez and colleagues looked closely into the redox process in sickle RBC focusing on NAD which is a potent antioxidant[21]. L-glutamine is a precursor for NAD. In sickle RBC, NAD metabolism is altered favoring an increase in the synthesis of NAD but with decreased NAD redox potential[21] As NAD is a major antioxidant molecule in RBC, the phenomenon of increased synthesis for NAD with decreased NAD redox potential was interpreted as a compensatory mechanism of sickle RBC in the presence of increased oxidant stress[21]. The presence of increased oxidant stress and oxidant susceptibility of sickle cell has been described independently by numerous investigators[20,22-25]. Additional studies suggested that supplementation with the precursor of NAD, L-glutamine, may further enhance the NAD redox system[26].
With 30 grams of L-glutamine supplementation, there was improvement in NAD redox potential in 7 out of 7 patients. In addition, there were consistent reports of improved general clinical condition in such areas as energy level and chronic pain levels [10]. Furthermore, a short observation period of approximately 3 months suggested a decrease in the incidence of vaso-occlusive painful crises among those patients.
On the basis of these results, we proceeded to study the effect of L-glutamine therapy on endothelial adhesion of sickle RBC. As noted in the data presented here, in both cross-sectional and longitudinal studies, there is an indication that L-glutamine therapy improves the endothelial adhesion of sickle RBC.
In the cross-sectional study, patients who were already on the L-glutamine therapy were compared to those who have never been treated with L-glutamine. The adhesion in the untreated group was elevated as expected, which was consistent with previously published reports. On the other hand, the adhesion for the treatment group was very similar to those of normal control.
To confirm the cross-sectional data, the prospective study was conducted on 5 sickle cell anemia patients who were treated for 4 to 8 weeks. The results were consistent among each patient studied. With the therapy, each patient had significant improvement in endothelial adhesion while there was no improvement in the control patient who was not treated. The stimulants for endothelial cells were taken from the methods used in the previous studies by Kalra and collegues[13]. Their method is well established for static assay and based on their data, we felt their methods were applicable in our project. Adhesion assay for the cross-sectional study and longitudinal studies were conducted at two separate institutions but using same agents to stimulate the endothelial cells. The purpose of utilizing two expert sites for two parts of study was to improve objectivity by demonstrating that consistent data can be obtained even when assays were conducted by unbiased experts at multiple independent sites.
In the studies, we could not verify the increase in RBC survival duration as neither reticulocyte counts nor hemoglobin levels were not changed significantly. However, improvement in endothelial adhesion does not have to automatically result in increase in RBC survival. As these patients are asplenic, irreversibly sickled RBC are not rapidly removed and they will contribute to hemoglobin and hematocrit levels. Also, it is possible that with larger study with longer duration, there may be a change in these parameters with statistical significance. In general morphology of RBC appeared to have improved [see additional file 1 and 2].
Adhesion was assessed using a static endothelium-RBC adhesion system rather than a dynamic assay under conditions of flow over time. The decision to use the static assay was based on previous reports including that of the seminal article by Hebbel and colleagues[8,12,27]. In addition, as noted by Setty and others, the static assay may better simulate the interaction of sickle RBC and endothelium where microvessel flow may be slow and intermittent [27-29]. Obviously, there are also data suggesting that the dynamic endothelium-RBC adhesion system may better represent the in vivo activity. Such data favoring flow based assay may be found in the studies by Libowsky and colleagues, demonstrating with intravital microscopy that sickle RBC that became adherent during stasis created by a pressure cuff, dislodged when the cuff was removed and the flow resumed[30,31]. It is doubtful that there is an in vitro system that will perfectly represent the in vivo process. Rather, it is likely that both static and dynamic assays have roles in the investigation of sickle cell pathophysiology.
At present, the exact mechanism by which L-glutamine effectively decreases the endothelial adhesion of sickle RBC is not clear. However, there is evidence suggesting that one of the ways L-glutamine may benefit sickle RBC is by improvement of NAD redox potential[10]. This may prevent oxidant damage to RBC which may result in stimulation of inflammation and expression of adhesion molecules. In addition, L-glutamine may provide much needed energy and building material to maintain the integrity of sickle RBC. It has been reported that sickle cell children require increased amino acid consumption, especially glutamine[32].
Conclusion
L-glutamine is an inexpensive compound widely consumed as part of a normal diet or as a dietary supplement. There is essentially no major adverse effect when used as an oral agent both in a research setting and non-research setting [33-36]. The data gathered so far including the ones presented in this article suggest that there is a physiological basis for the potential clinical benefit of L-glutamine in management of sickle cell disease patients.
List of abbreviations
NAD = Nicotinamide Adenosine Dinucleotide
RBC = Red Blood Cells
HUVEC = Human Umbilical Vein Endothelial Cells
LPS = Lipopolysaccharide
BSA = Bovine Serum Albumin
HBSS = Hank's Buffered Saline Solution
HEPES = 4-(2-hydroxyethyl)-1 – piperazineethanesulfonic acid
EGM = Endothelial Cell Growth Medium
SEM = Standard Error of Mean
Competing interests
YN and KRT have proprietary interest in the product tested in this study.
Non-financial competing interests
None
Authors' contributions
YN is the primary investigator and senior author. YN conceived of the study, designed and coordinated the study. YN also drafted the manuscript. NMM and YMS performed the endothelial adhesion assays and coordinated with SHE and VKK in the analysis and interpretation of data. SHC, DAA, MAS and JM were involved in acquisition of patient data as well ad the performance of various assays. CSJ and KRT have been involved in the conception of the study as well as in supervising composition of manuscript. All authors read and approved the manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
This work was supported by NIH/NHLBI Grant: 1R29HL58640 and FDAOPD Grant: FD-R-002028. This work was also supported by Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center and NIH supported General Clinical Research Center at Harbor-UCLA Medical Center.
==== Refs
Hamdallah M Bhatia AJ Prevalence of sickle-cell trait in USA adolescents of Central American origin Lancet 1995 346 707 8 7658851 10.1016/S0140-6736(95)92321-7
Weissman AM Preventive health care and screening of Latin American immigrants in the United States J Am Board Fam Pract 1994 7 310 23 7942100
Diaz-Barrios V Newborn screening for sickle cell disease and other hemoglobinopathies. New York's experience Pediatrics 1989 83 872 5 2717317
Frenette PS Sickle cell vaso-occlusion: multistep and multicellular paradigm Curr Opin Hematol 2002 9 101 6 11844991 10.1097/00062752-200203000-00003
Perronne V Roberts-Harewood M Bachir D Roudot-Thoraval F Delord JM Thuret I Schaeffer A Davies SC Galacteros F Godeau B Patterns of mortality in sickle cell disease in adults in France and England Hematol J 2002 3 56 60 11960397 10.1038/sj.thj.6200147
Wierenga KJ Hambleton IR Lewis NA Survival estimates for patients with homozygous sickle-cell disease in Jamaica: a clinic-based population study Lancet 2001 357 680 3 11247552 10.1016/S0140-6736(00)04132-5
Barabino GA McIntire LV Eskin SG Sears DA Udden M Endothelial cell interactions with sickle cell, sickle trait, mechanically injured, and normal erythrocytes under controlled flow Blood 1987 70 152 7 3593962
Hebbel RP Yamada O Moldow CF Jacob HS White JG Eaton JW Abnormal adherence of sickle erythrocytes to cultured vascular endothelium: possible mechanism for microvascular occlusion in sickle cell disease J Clin Invest 1980 65 154 60 7350195
Vargas FF Blackshear GL Vascular resistance and transit time of sickle red blood cells Blood Cells 1982 8 139 45 7115971
Niihara Y Zerez CR Akiyama DS Tanaka KR Oral L-glutamine therapy for sickle cell anemia: I. Subjective clinical improvement and favorable change in red cell NAD redox potential Am J Hematol 1998 58 117 21 9625578 10.1002/(SICI)1096-8652(199806)58:2<117::AID-AJH5>3.0.CO;2-V
Zerez CR Lachant NA Lee SJ Tanaka KR Decreased erythrocyte nicotinamide adenine dinucleotide redox potential and abnormal pyridine nucleotide content in sickle cell disease Blood 1988 71 512 5 3337912
Kalra VK Ying Y Deemer K Natarajan R Nadler JL Coates TD Mechanism of cigarette smoke condensate induced adhesion of human monocytes to cultured endothelial cells J Cell Physiol 1994 160 154 62 7517402 10.1002/jcp.1041600118
Wali RK Jaffe S Kumar D Kalra VK Alterations in organization of phospholipids in erythrocytes as factor in adherence to endothelial cells in diabetes mellitus Diabetes 1988 37 104 11 3335275
Sugihara K Sugihara T Mohandas N Hebbel RP Thrombospondin mediates adherence of CD36+ sickle reticulocytes to endothelial cells Blood 1992 80 2634 42 1384794
Hebbel RP Schwartz RS Mohandas N The adhesive sickle erythrocyte: cause and consequence of abnormal interactions with endothelium, monocytes/macrophages and model membranes Clin Haematol 1985 14 141 61 3886233
Ohnishi ST Horiuchi KY Horiuchi K The mechanism of in vitro formation of irreversibly sickled cells and modes of action of its inhibitors Biochim Biophys Acta 1986 886 119 29 3955078 10.1016/0167-4889(86)90217-X
Hebbel RP Eaton JW Steinberg MH White JG Erythrocyte/endothelial interactions in the pathogenesis of sickle-cell disease: a "real logical" assessment Blood Cells 1982 8 163 73 7115974
Sultana C Shen Y Rattan V Johnson C Kalra VK Interaction of sickle erythrocytes with endothelial cells in the presence of endothelial cell conditioned medium induces oxidant stress leading to transendothelial migration of monocytes Blood 1998 92 3924 35 9808586
Inwald DP Kirkham FJ Peters MJ Lane R Wade A Evans JP Klein NJ Platelet and leucocyte activation in childhood sickle cell disease: association with nocturnal hypoxaemia Br J Haematol 2000 111 474 81 11122087 10.1046/j.1365-2141.2000.02353.x
Rattan V Sultana C Shen Y Kalra VK Oxidant stress-induced transendothelial migration of monocytes is linked to phosphorylation of PECAM-1 Am J Physiol 1997 273 E453 61 9316433
Zerez CR Lachant NA Lent KM Tanaka KR Decreased pyrimidine nucleoside monophosphate kinase activity in sickle erythrocytes Blood 1992 80 512 6 1320957
Klings ES Farber HW Role of free radicals in the pathogenesis of acute chest syndrome in sickle cell disease Respir Res 2001 2 280 5 11686897 10.1186/rr70
Anastasi J Hemoglobin S-mediated membrane oxidant injury: protection from malaria and pathology in sickle cell disease Med Hypotheses 1984 14 311 20 6472160 10.1016/0306-9877(87)90135-6
Wetterstroem N Brewer GJ Warth JA Mitchinson A Near K Relationship of glutathione levels and Heinz body formation to irreversibly sickled cells in sickle cell anemia J Lab Clin Med 1984 103 589 96 6699474
Beretta L Gerli GC Ferraresi R Agostoni A Gualandri V Orsini GB Antioxidant system in sickle red cells Acta Haematol 1983 70 194 7 6410646
Niihara Y Zerez CR Akiyama DS Tanaka KR Increased red cell glutamine availability in sickle cell anemia: demonstration of increased active transport, affinity, and increased glutamate level in intact red cells J Lab Clin Med 1997 130 83 90 9242370 10.1016/S0022-2143(97)90062-7
Setty BN Kulkarni S Stuart MJ Role of erythrocyte phosphatidylserine in sickle red cell-endothelial adhesion Blood 2002 99 1564 71 11861269 10.1182/blood.V99.5.1564
Rodgers GP Schechter AN Noguchi CT Klein HG Nienhuis AW Bonner RF Periodic microcirculatory flow in patients with sickle-cell disease N Engl J Med 1984 311 1534 8 6504081
Wun T Paglieroni T Field CL Welborn J Cheung A Walker NJ Tablin F Platelet-erythrocyte adhesion in sickle cell disease J Investig Med 1999 47 121 7 10198567
Lipowsky HH Sheikh NU Katz DM Intravital microscopy of capillary hemodynamics in sickle cell disease J Clin Invest 1987 80 117 27 3597770
Lipowsky HH Williams ME Shear rate dependency of red cell sequestration in skin capillaries in sickle cell disease and its variation with vasoocclusive crisis Microcirculation 1997 4 289 301 9219221
Salman EK Haymond MW Bayne E Sager BK Wiisanen A Pitel P Darmaun D Protein and energy metabolism in prepubertal children with sickle cell anemia Pediatr Res 1996 40 34 40 8798243
Garlick PJ Assessment of the safety of glutamine and other amino acids J Nutr 2001 131 2556S 61S 11533313
Ziegler TR Bazargan N Leader LM Martindale RG Glutamine and the gastrointestinal tract Curr Opin Clin Nutr Metab Care 2000 3 355 62 11151079 10.1097/00075197-200009000-00005
Hasebe M Suzuki H Mori E Furukawa J Kobayashi K Ueda Y Glutamate in enteral nutrition: can glutamate replace glutamine in supplementation to enteral nutrition in burned rats? JPEN J Parenter Enteral Nutr 1999 23 S78 82 10483902
Ziegler TR Benfell K Smith RJ Young LS Brown E Ferrari-Baliviera E Lowe DK Wilmore DW Safety and metabolic effects of L-glutamine administration in humans JPEN J Parenter Enteral Nutr 1990 14 137S 146S 2119459
|
16042803
|
PMC1198219
|
CC BY
|
2021-01-04 16:31:42
|
no
|
BMC Blood Disord. 2005 Jul 25; 5:4
|
utf-8
|
BMC Blood Disord
| 2,005 |
10.1186/1471-2326-5-4
|
oa_comm
|
==== Front
BMC Blood DisordBMC Blood Disorders1471-2326BioMed Central London 1471-2326-5-51604280910.1186/1471-2326-5-5Research ArticleTotal blood lymphocyte counts in hemochromatosis probands with HFE C282Y homozygosity: relationship to severity of iron overload and HLA-A and -B alleles and haplotypes Barton James C [email protected] Howard W [email protected] Ronald T [email protected] Rodney CP [email protected] Southern Iron Disorders Center, Birmingham, Alabama, USA2 Department of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, USA3 Department of Epidemiology and International Health, University of Alabama at Birmingham, Birmingham, Alabama, USA4 Immunogenetics Program, Department of Microbiology, University of Alabama at Birmingham, Birmingham, Alabama, USA2005 25 7 2005 5 5 5 10 3 2005 25 7 2005 Copyright © 2005 Barton et al; licensee BioMed Central Ltd.2005Barton 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
It has been reported that some persons with hemochromatosis have low total blood lymphocyte counts, but the reason for this is unknown.
Methods
We measured total blood lymphocyte counts using an automated blood cell counter in 146 hemochromatosis probands (88 men, 58 women) with HFE C282Y homozygosity who were diagnosed in medical care. Univariate and multivariate analyses of total blood lymphocyte counts were evaluated using these variables: sex; age, transferrin saturation, and serum ferritin concentration at diagnosis; units of blood removed by phlebotomy to achieve iron depletion; and human leukocyte antigen (HLA)-A and -B alleles and haplotypes.
Results
The mean age at diagnosis was 49 ± 14 years (range 18 – 80 years) in men and 50 ± 13 years (range 22 – 88 years) in women. The correlations of total blood lymphocyte counts with sex, age, transferrin saturation, and serum ferritin concentration at diagnosis, and units of blood removed by phlebotomy to achieve iron depletion were not significant at the 0.05 level. Univariate analyses revealed significant associations between total blood lymphocyte counts and presence of the HLA-A*01, -B*08, and -B*14 alleles, and the A*01-B*08 haplotype. Presence of the A*01 allele, B*08 allele, or A*01-B*08 haplotype were associated with a lower total blood lymphocyte count, whereas presence of the B*14 allele was associated with a greater total blood lymphocyte count. There was an inverse association of total blood lymphocyte count with units of phlebotomy to achieve iron depletion, serum ferritin concentration, and with presence of the A*01-B*08 haplotype.
Conclusion
We conclude that there is a significant inverse relationship of total blood lymphocyte counts and severity of iron overload in hemochromatosis probands with HFE C282Y homozygosity. The presence of the HLA-A*01 allele or the -B*08 allele was also associated with significantly lower total blood lymphocyte counts, whereas presence of the -B*14 allele was associated with significantly higher total blood lymphocyte counts. In univariate and multivariate analyses, total blood lymphocyte counts were significantly lower in probands with the HLA-A*01-B*08 haplotype than in probands without this haplotype.
==== Body
Background
Hemochromatosis occurs in 0.003 – 0.005 of persons of northwestern European descent, and is typically associated with homozygosity for the C282Y mutation of the HFE gene (exon 2, nt 845 G→A), located ~4 Mb telomeric to the human leukocyte antigen (HLA) region on Ch6p [1,2]. Some persons with hemochromatosis absorb increased quantities of iron and develop severe iron overload that is associated with hepatic cirrhosis, primary liver cancer, diabetes mellitus, other endocrinopathy, arthropathy, and cardiomyopathy, and with reduced longevity [1]. Total blood lymphocyte counts were lower in hemochromatosis index subjects with HFE C282Y homozygosity, higher iron stores, and hepatic cirrhosis than in those with lower iron burdens who did not have cirrhosis. In men, there was a significant negative correlation of total blood lymphocyte counts and body iron stores [3].
In the present study, we performed multivariate analyses of the variables sex; age; transferrin saturation and serum ferritin concentration at diagnosis; units of blood removed by phlebotomy to achieve iron depletion; and human leukocyte antigen (HLA)-A and -B alleles and haplotypes to determine their effects on total blood lymphocyte count at diagnosis in hemochromatosis probands with C282Y homozygosity. We discuss the implications of our observations in explaining quantities of total blood lymphocytes previously reported in persons with hemochromatosis associated with C282Y homozygosity.
Methods
General criteria for selection of study subjects
The performance of this work was approved by the Institutional Review Boards of Brookwood Medical Center and the University of Alabama at Birmingham. All subjects were adults (≥ 18 years of age) who identified themselves as Caucasians or whites; each resided in central Alabama. All hemochromatosis probands were diagnosed in a single community medical center; none was diagnosed by family or population screening. Probands with diagnoses of primary hematologic malignancies or those receiving anti-cancer chemotherapy were excluded. Persons of African ancestry were excluded for reasons described previously [4-7].
Selection of hemochromatosis probands
A presumptive diagnosis was established using an elevated transferrin saturation criterion; each proband was evaluated for iron overload and its complications [1,8,9]. We included probands who had: a) diagnosis in medical care during the interval 1997–2002; b) HFE C282Y homozygosity; c) available HLA-A and -B haplotypes; and d) therapeutic phlebotomy to induce iron depletion [9]. This cohort is the same as otherwise described and evaluated in a previous study [10].
Diagnosis of common variable immunodeficiency (CVID) and IgG subclass deficiency (IgGSD)
Diagnoses of CVID or IgGSD were based on demonstration of persistent, otherwise unexplained serum concentrations of Ig >2 SD below the corresponding mean levels [6,7,11]. Criteria for the diagnosis of CVID were: 1) decreased total serum IgG concentration; and 2) either decreased IgG subclass(es), decreased serum IgA concentration, or decreased serum IgM concentration [6,7,11]. Criteria for the diagnosis of IgGSD were: 1) normal total serum IgG concentration; and 2) abnormally low serum concentrations of one or more IgG subclasses; some patients with IgGSD also have decreased levels of IgA or IgM levels, although measurements of IgA or IgM are not diagnostic criteria for CVID or IgGSD [6,7,11].
Iron-associated measurements
Serum iron concentration, total serum iron-binding capacity, and serum ferritin concentration were measured using automated clinical methods and blood specimens obtained after an overnight fast. Transferrin saturation was expressed as the quotient of serum iron and iron-binding capacity × 100%. In some cases, percutaneous biopsy specimens of liver were obtained as an adjunct to hemochromatosis diagnosis and evaluation of hepatic pathology. Phlebotomy to induce iron depletion was performed as previously described; one unit of phlebotomy was defined as ~500 mL of blood [9]. We used presumptive criteria of iron overload as indications to perform therapeutic phlebotomy: serum ferritin ≥ 300 ng/mL (men) and ≥ 200 ng/mL (women) [9]. Iron overload was defined by demonstration of hepatic iron index ≥ 1.9 or removal of ≥ 2.0 g Fe by therapeutic phlebotomy [12]. Iron depletion was defined as complete when the serum ferritin level was 10 – 20 ng/mL, or when the hemoglobin concentration was <11.0 g/dL or the hematocrit was <33.0% for more than three weeks (in patients without chronic anemia) [9].
Immunoglobulin measurements
Serum concentrations of IgG, IgG subclasses, IgA, and IgM and were measured using standard automated methods before IgG replacement therapy was initiated, and as nadir values at the time of monthly IgG infusions in some patients. Reference ranges for serum Ig concentrations are: total IgG 700 – 1600 mg/dL; IgG1 422 – 1292 mg/dL; IgG2 117 – 747 mg/dL; IgG3 41 – 129 mg/dL; IgG4 1 – 291 mg/dL; total IgA 70 – 400 mg/dL; and IgM 40 – 230 mg/dL. The basis of these reference ranges has been reported elsewhere [13]. Deficiency of an Ig class or subclass was defined by a serum concentration at diagnosis that was less than the corresponding lower reference limit. Quantification of serum concentrations of total serum IgG and IgG subclasses was performed in all hemochromatosis probands. Quantification of IgA and IgM was performed in subjects with CVID and IgGSD, although these analytes were not measured in hemochromatosis probands whose total serum IgG and IgG subclass values were within the corresponding reference ranges. Measurement of IgA subclasses, IgD, or IgE in serum was not routinely performed in any of the present subjects.
Total blood lymphocyte counts
Blood specimens obtained by antecubital venipuncture from probands at the time of diagnosis of hemochromatosis were analyzed using a Cell-Dyne 1300 automated blood counter (Abbott Laboratories, Chicago, IL). Total blood lymphocyte counts were defined as the numbers of leukocytes of volume 40 – 100 fL detected by the counter in the respective specimens; counts are expressed as cells/mm3 × 10-3.
HFE and HLA analyses
HFE analyses were performed as described previously [14]. HLA-A and -B alleles were detected using low-resolution DNA-based typing (PCR/sequence-specific oligonucleotide probe) in hemochromatosis probands [14]. Control subjects were tested using the microdroplet lymphocytotoxicity test [15]; subjects were evaluated using antisera that detected allele assignments described in the 9th International Histocompatibility Workshop [16]. Because the levels of resolution of the DNA-based and serological typing methods we used are similar, alleles detected by these respective methods provide concordant allele assignments, with the exception of B*70 and B*71 that were not detected by serological methods. HLA typing of family members permitted assignment of Ch6p haplotypes defined by -A and -B alleles [5,7].
Statistical considerations
The data set included observations in 146 hemochromatosis probands. Numbers of men and women in various proband subgroups vary because some data were unavailable due to conditions of referral and prior management; we were unable to set phase for HLA haplotype determination in 20 probands. Analyses were performed with SAS [17], a computer spreadsheet (Excel 2000®, Microsoft Corp., Redmond, WA), and a statistical program (GB-Stat® v. 10.0, 2003, Dynamic Microsystems, Inc., Silver Spring, MD).
We determined that a loge (ln) transformation normalized the iron measures data and total blood lymphocyte counts (expressed as cells/mm3 × 10-3 ± 1 SD) [18], and thus permitted the use of statistical techniques that assume that values within a data set are normally distributed. Independent variables included a) sex; b) age at diagnosis; c) transferrin saturation at diagnosis; d) serum ferritin concentration at diagnosis; e) units of blood removed by phlebotomy to achieve iron depletion; and f) HLA-A and -B alleles and haplotypes. Descriptive data are displayed as enumerations, percentages, mean ± 1 S.D. or mean (95% confidence intervals (CI)). Frequency values were compared using chi-square analysis or Fisher exact test (one-tail), as appropriate. Mean values were compared using a student t-test (two-tail). Transformed measures are rounded to two decimal places. Blood lymphocyte counts were expressed to the nearest one decimal place. Frequencies and p values are expressed to four significant figures. We used an algorithm applicable to loci with multiple alleles [19] to estimate the significance level of Hardy-Weinberg proportions of HLA-A and -B allele frequencies in hemochromatosis probands. Two sets of analysis of variance (ANOVA) models were fit to the loge-transformed total blood lymphocyte count data. The first set used indicators of single HLA-A and -B alleles, and the second used HLA-A and -B haplotypes. The overall fits of the ANOVA models are indicated by R2 values; values of p < 0.05 were defined as significant.
Results
General characteristics of hemochromatosis probands with HFE C282Y homozygosity
There were 146 probands (88 men, 58 women). The mean age at diagnosis was 49 ± 14 years (range 18 – 80 years) in men and 50 ± 13 years (range 22 – 88 years) in women. Iron measures are displayed in Table 1. Eighty-six men and 42 women had iron overload. Fifteen men and six women had hepatic cirrhosis proven by liver biopsy (14.4%). Nine men and two women reported that they consumed ≥ 60 g of ethanol daily (7.5%). Three men had chronic hepatitis C; one man had porphyria cutanea tarda. None had undergone splenectomy, and none had lymphoproliferative disorders. Thirteen probands had either CVID (n = 3) or IgGSD (n = 10) (7 men, 6 women).
Table 1 Iron measures in hemochromatosis probands with HFE C282Y homozygosity1
Iron measure Men (95% CI) [n] Women (95% CI) [n] p value
Serum iron, μg/dL 209 (137, 317) [81] 193 (118, 315) [51] 0.0523
Transferrin saturation, % 85 (59, 121) [83] 77 (44, 133) [52] 0.0120
Serum ferritin, ng/mL 1097 (209, 5768) [81] 546 (84, 3535) [56] <0.0001
Phlebotomy to achieve iron depletion, units 29 (7, 122) [73] 17 (4, 76) [42] 0.0001
1Serum iron, transferrin saturation, and serum ferritin levels were measured at diagnosis of hemochromatosis. Values were transformed (loge) to achieve normal distributions; comparisons were made using a student t-test (two-tail). Data displayed in the table are expressed as mean (95% CI) after computing antilogse of the transformed data; p values < 0.05 were defined as significant.
Mean transferrin saturation, mean serum ferritin concentration, and mean units of phlebotomy to achieve iron depletion were greater in men than women with hemochromatosis (Table 1). The mean units of phlebotomy to achieve iron depletion was approximately twice as great in men as women (Table 1).
Hardy-Weinberg proportions of HLA-A and HLA-B alleles
Frequencies of HLA-A and -B alleles in hemochromatosis probands did not depart significantly from Hardy-Weinberg equilibrium.
Overall frequencies of HLA-A*03 allele and HLA-A and -B haplotypes
The frequencies of HLA-A*03 in male and female hemochromatosis probands were similar (0.8023 vs. 0.6727; p = 0.0823). Frequencies of HLA-A*03 in male and female control subjects were also similar (0.2662 vs. 0.2815; p = 0.7162) [10]. The overall frequency of A*03 in hemochromatosis probands was greater than that in control subjects (0.7518 vs. 0.2787; p < 0.0001).
Frequencies of the most common haplotypes detected in HFE C282Y homozygotes with a hemochromatosis phenotype from this geographic area [5,6,10] were compared with corresponding frequencies in control subjects. The overall frequency of A*01-B*08 was lower in hemochromatosis probands than in control subjects, but the difference was not significant (0.0603 vs. 0.0927, respectively; p = 0.0634). The overall frequency of A*02-B*44 was similar in hemochromatosis probands and in control subjects (0.0461 vs. 0.0620, respectively; p = 0.2846). The overall frequency of A*03-B*07 was greater in hemochromatosis probands than in control subjects (0.2447 vs. 0.0520, respectively; p < 0.0001, respectively). The overall frequency of A*03-B*14 was greater in hemochromatosis probands than in control subjects (0.0709 vs. 0.0113, respectively; p < 0.0001).
Comparison of loge total blood lymphocyte counts in men and women
Univariate analyses of mean total blood lymphocyte counts in men and women were expressed as cells/mm3 × 10-3 (95% CI). In men, the mean was 1.9/mm3 × 10-3 (1.1, 3.5). In women, the mean was 2.0/mm3 × 10-3 (1.1, 3.5). These values were not significantly different (p = 0.2473).
Correlation of loge total blood lymphocyte counts with clinical variables
The correlations of loge total blood lymphocyte counts with sex; age, loge transferrin saturation, loge serum ferritin concentration at diagnosis; and loge units of blood removed by phlebotomy to achieve iron depletion were not significant at the 0.05 level.
Univariate analyses of loge total blood lymphocyte counts and HLA-A and -B alleles
Univariate analyses revealed significant associations between mean loge total blood lymphocyte count and presence of the HLA-A*01, -B*08, and -B*14 alleles, and the A*01-B*08 haplotype (Table 2). Presence of the -A*01 allele, -B*08 allele, or A*01-B*08 haplotype was associated with lower total blood lymphocyte counts, whereas presence of the -B*14 allele was associated with greater total blood lymphocyte counts.
Table 2 Mean total blood lymphocyte counts in hemochromatosis probands with HFE C282Y homozygosity
HLA loge lymphocyte count (cells/mm3 × 10-3 (SD)) lymphocyte count (cells/mm3 × 10-3 (95% CI))1 p value (present vs. absent)2
A*01 present 1.012 (0.213) 2.8 (1.8, 4.2) 0.0362
A*01 absent 1.101 (0.184) 3.0 (2.1, 4.3)
B*08 present 0.975 (0.178) 2.7 (1.9, 3.8) 0.0048
B*08 absent 1.104 (0.189) 3.0 (2.1, 4.4)
B*14 present 1.160 (0.180) 3.2 (2.2, 4.5) 0.0317
B*14 absent 1.065 (0.192) 2.9 (2.0, 4.2)
A*01-B*08 present 0.943 (0.195) 2.6 (1.8, 3.8) 0.0026
A*01-B*08 absent 1.101 (0.185) 3.0 (2.1, 4.3)
1Lymphocyte counts were measured at diagnosis of hemochromatosis, and are expressed as mean (95% CI) after computing antilogse of the transformed data.
2Comparisons were made using loge lymphocyte counts and student t-test (two-tail); p values < 0.05 were defined as significant.
Multivariate analyses of loge total blood lymphocyte counts and clinical variables, HLA-A and -B alleles, and HLA-A and -B haplotypes
The residual variance formed by accounting for age and gender effects were used to explore multivariable associations. The HLA alleles that predicted loge total blood lymphocyte counts were B*08 (p = 0.0283) and B*14 (p = 0.0204). The presence of B*08 was associated with lower loge total blood lymphocyte counts, whereas the presence of B*14 was associated with higher loge total blood lymphocyte counts. When the residuals of the other clinical variables were added to the model, the effects of units of phlebotomy to induce iron depletion, serum ferritin concentration, and B*08 were significant (p = 0.0326, 0.0172, and 0.0127, respectively). Mean lymphocyte counts were lower with increasing serum ferritin concentration.
Similar models were fit using the HLA-A and -B haplotypes as variables. The A*01-B*08 haplotype was the only significant predictor of total blood lymphocyte count (p = 0.0021), and the presence of A*01-B*08 was associated with lower total blood lymphocyte counts after accounting for age and gender. When other variables were added to this same model, the units of phlebotomy to induce iron depletion (p = 0.0440), serum ferritin concentration (p = 0.0108) and the effect of A*01-B*08 (p = 0.0023) remained significant. A decrease in loge total blood lymphocyte count was associated with an increase in units of phlebotomy to induce iron depletion, loge serum ferritin concentration, and with presence of the A*01-B*08 haplotype.
Univariate analyses of loge total blood lymphocyte counts and hepatic cirrhosis
The mean total blood lymphocyte count in the 21 probands with hepatic cirrhosis proven by biopsy (1.9 cells/mm3 × 10-3 (95% CI: 1.1 × 10-3, 3.3 × 10-3)) was similar to that in the 125 probands without cirrhosis (1.9 cells/mm3 × 10-3 (95% CI: 1.1 × 10-3, 3.5 × 10-3); p = 0.8514).
Univariate analyses of loge total blood lymphocyte counts and CVID or IgGSD
The mean total blood lymphocyte count in the 13 probands with CVID or IgGSD (2.0 cells/mm3 × 10-3 (95% CI: 1.5 × 10-3, 2.6 × 10-3)) was similar to that in the 133 probands who did not have CVID or IgGSD (1.9 cells/mm3 × 10-3 (95% CI: 1.0 × 10-3, 3.5 × 10-3); p = 0.5814).
Discussion
The present study is comprised of the largest number of HFE C282Y homozygotes with hemochromatosis phenotypes who had available HLA-A and -B allele and haplotype data and were evaluated for the effects of clinical variables on total blood lymphocyte counts. Overall, the mean total blood lymphocyte counts in the present 88 male and 58 female hemochromatosis probands are very similar to those determined by automated methods in 100 male and 100 female healthy volunteer Caucasians [20]. Taken together, these results confirm observations of Cruz et al. that total blood lymphocyte counts of C282Y homozygotes with hemochromatosis phenotypes (37 men, 9 women) did not differ significantly from those of unrelated normal control subjects (116 men, 148 women) [21] and those of Porto et al. that there is an association between low CD8(+) numbers, HLA phenotype, and severity of iron overload [22].
In a multivariate analysis, we observed that total blood lymphocyte counts were lower in probands who required greater numbers of units of phlebotomy to achieve iron depletion or who had greater serum ferritin concentrations at diagnosis. This is consistent with a previous report of a significant inverse correlation of total blood lymphocyte counts with iron stores quantified by phlebotomy in a smaller hemochromatosis case series [3]. Although measurement of blood lymphocyte subsets was beyond the scope of the present study, it has been reported that proportions of the two major peripheral T-lymphocyte subsets expressed as CD4/CD8 ratio are stable before and after phlebotomy therapy for hemochromatosis, confirming the existence of a homeostatic mechanism that regulates the relative numbers of the two major blood T-lymphocyte populations [22,23]. Before the discovery of HFE, it was reported that inheritance of part or all of the hemochromatosis ancestral haplotype that includes HLA-A*03 and -B*07, particularly in a homozygous configuration, was associated with evidence of more severe iron overload in hemochromatosis patients in Australia, Alabama, and Italy [24-26]. Further, Porto et al. demonstrated that the severity of iron overload quantified by phlebotomy in patients with hemochromatosis was correlated with the proportions of CD4(+) and CD8(+) blood lymphocytes and the presence or absence of HLA-A*03 [22]. Thus, the latter report integrated then-existing knowledge of the relationships of severity of iron overload, HLA types, and blood lymphocyte subsets in persons with hemochromatosis.
Some observations support the hypothesis that lymphocyte numbers could influence iron absorption and therefore severity of iron overload in hemochromatosis. High CD4/CD8 ratios appear to precede the development of severe iron overload in persons with hemochromatosis [23,27]. Persons with hemochromatosis have significantly different CD(8)+ blood lymphocyte subsets than normal control subjects, based on analysis of CD(28) positivity or negativity [28]. In mice, blood lymphocyte numbers may influence iron overload severity in the absence of functional HFE protein [29]. There is a candidate mechanism that could account for a lymphocyte-mediated influence on iron absorption and severity of iron overload in hemochromatosis. Interleukin-6 (IL-6), a cytokine produced predominantly by lymphocytes and macrophages [30,31], induces expression of hepcidin, a potent inhibitor of iron absorption [32]. Further, hepcidin levels are significantly decreased in persons who have hemochromatosis associated with mutations of HFE (Ch6p21.3) [32]. Although most reports of lymphocyte numbers and subsets have been made in persons presumed or documented to have HLA- or HFE-associated hemochromatosis or iron overload, lymphopenia also occurred in an unusual case of early age-of-onset hemochromatosis and severe iron overload associated with homozygosity for a hepcidin promoter mutation on Ch19q13 [33].
Some reports indicate that lymphocyte numbers do not influence iron absorption either in patients with hemochromatosis or in those with iron overload due to other causes. In the present study, we observed that the mean blood lymphocyte counts were similar in men and women. In an earlier study of the same cohort, we observed that the severity of iron overload was significantly greater in men than women [10]. However, there was significant disparity in the frequency of certain HLA-A and -B types and haplotypes between men and women, but there was no significant association of these HLA markers with the severity of iron overload in a multivariate analysis that included sex as a independent variable [10]. Hepcidin levels are significantly decreased in hemochromatosis associated with TFR2, FPN1, and HJV mutations [34-36]. However, it is unknown whether there is an inverse association of blood lymphocyte numbers and the severity of iron overload or whether lymphocytes contribute to decreased hepcidin levels in these disorders. In patients with beta-thalassemia major, there was a highly significant linear increase in the percentages of blood OKT8(+) cells with an increasing number of units of erythrocytes transfused, irrespective of splenectomy [37]. The percentage of blood OKT4(+) cells varied inversely with increasing numbers of units of erythrocyte transfusion in patients who had not undergone splenectomy; in those who had undergone splenectomy, no significant correlation was observed [37]. Inverse relationships of CD8(+) blood lymphocytes and severity of transfusion iron overload were also observed in persons with beta-thalassemia, and deferoxamine therapy was associated with an increase in CD8(+) blood lymphocytes [38]. In sub-Saharan Africans with African iron overload, a disorder that is typically not linked to HLA or HFE C282Y, there was no significant association of serum ferritin concentrations and total blood lymphocyte counts [3,39]. In an experimental model of secondary iron overload in rats, the distribution of lymphocyte subsets in blood, thymus, spleen, mesenteric lymph nodes, Peyer patches, and bone marrow were similar in control and experimental groups [40]. Altogether, these results suggest that there is not a consistent relationship of severity of iron overload with CD4/CD8 ratios, blood T-lymphocyte subsets, or abnormal total blood lymphocyte counts in patients with hemochromatosis or in those with iron overload due to other causes [3,41].
A putative gene on Ch6p that modifies iron overload severity in hemochromatosis is presumed to be linked predominantly to A*03 or A*03-B*07 [14,22,24-26,42]. At present, there are two candidate genes. One is localized to the region of D6S105 [42]. The multivariate analysis of a large cohort of hemochromatosis probands with HFE C282Y homozygosity demonstrated that A*03-B*07 has no significant effect on units of phlebotomy to achieve iron depletion, but did not exclude a putative modifier gene in this region [10]. In another study, extended haplotypes of the Ch6p21.3 region in hemochromatosis patients and their "phenotypically unaffected" relatives with HFE C282Y homozygosity were similar [43]. Another candidate is tumor necrosis factor (TNF)-α promoter polymorphisms [44]. In an independent case series, however, a positive relationship of TNF-α promoter polymorphisms with iron overload severity or its complications was not confirmed [45]. Altogether, these later observations do not strongly support previous hypotheses that putative genes or alleles on Ch6p modify the severity of iron overload in C282Y homozygotes with hemochromatosis.
Our data set permitted an analysis of the relationship of total blood lymphocyte counts and HLA-A and -B alleles and haplotypes. In the present hemochromatosis probands, univariate analyses revealed that the presence of the HLA-A*01 allele or the -B*08 allele was associated with lower total blood lymphocyte counts, whereas presence of the -B*14 allele was associated with greater total blood lymphocyte counts. Bryan et al. first suggested a possible role for HLA in the interaction of iron, HLA, and lymphocytes by demonstrating that there was a differential response of peripheral blood mononuclear cells from HLA-A*02 and non-HLA-A*02 donors when the respective lymphocyte isolates were exposed to iron in a mixed lymphocyte culture reaction [46]. In hemochromatosis families and random population control subjects from Portugal, significantly higher blood CD8(+) lymphocyte counts were observed in subjects who had both the HFE H63D mutation and the HLA-A*29 allele [47]. In a control population from Portugal, there was a significant correlation of the HLA-A*01 with high numbers of CD8(+) blood lymphocytes, and an association of HLA-A*24 with low numbers of CD8(+) blood lymphocytes [21]. These observations indicate that total blood lymphocyte counts or blood T-lymphocyte subsets in persons who inherit common HFE missense mutations (with or without hemochromatosis) are associated with HLA-A and -B alleles.
In the present hemochromatosis probands, A*01-B*08 was a significant predictor of lower total blood lymphocyte counts in univariate and multivariate analyses. Further, A*01-B*08 is associated with greater serum ferritin concentrations in older hemochromatosis probands with C282Y homozygosity grouped by age than other haplotypes [10]. However, the association of A*01-B*08 and the severity of iron overload quantified by phlebotomy to achieve iron depletion was not significant [10]. In persons without hemochromatosis, total blood lymphocyte counts are lower in those with HLA-A*01-B*08, DR3 than in persons with other HLA haplotypes [48,49]. Persons with human immunodeficiency virus (HIV) infections and HLA-A*01-B*08 have lower total blood lymphocyte counts than persons with HIV infections who do not have HLA-A*01-B*08 [50,51]. Taken together, these observations suggest that there is a determinant of total blood lymphocyte counts or CD8(+) blood lymphocyte counts within the HLA-A*01-B*08 haplotype or in linkage disequilibrium with it. These observations also support previous reports that genetic factors in the region of the major histocompatibility complex on chromosome 6 have a major influence on the variation in blood lymphocyte numbers, especially those of T-lymphocyte subsets, in humans [21,52,53]. In a study of 15 CEPH families, quantitative trait loci that accounted for significant proportions of the phenotypic variance of blood lymphocyte counts and blood lymphocyte subpopulations were also detected on chromosomes 1, 2, 3, 4, 8, 9, 11, 12, and 18 [54].
In the present study, there was no significant difference in the total blood lymphocyte counts of hemochromatosis probands with or without hepatic cirrhosis in a univariate analysis. In contrast, it has been reported that total blood lymphocyte counts were lower in hemochromatosis index subjects with C282Y homozygosity with hepatic cirrhosis than in those with lower iron burdens who did not have cirrhosis [3]. However, the latter investigators indicated that their overall findings argue against the possibility that low blood lymphocyte counts in HFE hemochromatosis are a consequence of iron overload or represent an epiphenomenon of advanced cirrhosis [3,41]. The differences in the results of the present study and those of a previous report [3] may also be due to ethnic differences in the respective study populations, and the greater number of patients and lower prevalence of hepatic cirrhosis in the present report (n = 146; 14% had cirrhosis vs. previous report: n = 20; 65% had cirrhosis).
Total blood lymphocyte counts, including T-and B-lymphocyte subset counts, are subnormal in some persons with CVID [55,56]. However, we did not detect a significant difference in the total blood lymphocyte counts in hemochromatosis probands with or without CVID or IgGSD. This is consistent with the generally less severe blood lymphocyte subset deficits in IgGSD than in CVID [55,57], and with the greater proportion of the hemochromatosis probands in the present and another cohort who had IgGSD than CVID [6].
Conclusion
We conclude that there is a significant inverse relationship of total blood lymphocyte counts and severity of iron overload in hemochromatosis probands with HFE C282Y homozygosity. The presence of the HLA-A*01 allele or the -B*08 allele was also associated with significantly lower total blood lymphocyte counts, whereas presence of the -B*14 allele was associated with significantly higher total blood lymphocyte counts. In univariate and multivariate analyses, total blood lymphocyte counts were significantly lower in probands with the HLA-A*01-B*08 haplotype than in probands without this haplotype.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
JCB conceived and designed the study, diagnosed and treated the hemochromatosis probands and compiled their clinical data, performed some of the statistical analyses, and contributed to writing the manuscript. HWW performed statistical analyses and contributed to writing the manuscript. RTA compiled data on hemochromatosis probands, performed HLA and HFE typing of many of the probands, performed HLA typing of all control subjects, and contributed to writing the manuscript. RCP contributed to statistical analyses and writing the manuscript. All authors approved of the manuscript in its final form.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
This work was supported in part by Southern Iron Disorders Center, the Immunogenetics Program, grant MH-066181 from the National Institute of Mental Health, and grant NS-45934 from the National Institute of Neurological Diseases and Stroke.
==== Refs
Witte DL Crosby WH Edwards CQ Fairbanks VF Mitros FA Practice guideline development task force of the College of American Pathologists. Hereditary hemochromatosis Clin Chim Acta 1996 245 139 200 8867884 10.1016/0009-8981(95)06212-2
Feder JN Gnirke A Thomas W Tsuchihashi Z Ruddy DA Basava A Dormishian F Domingo RJ Ellis MC Fullan A Hinton LM Jones NL Kimmel BE Kronmal GS Lauer P Lee VK Loeb DB Mapa FA McClelland E Meyer NC Mintier GA Moeller N Moore T Morikang E Wolff RK . A novel MHC class I-like gene is mutated in patients with hereditary haemochromatosis Nat Genet 1996 13 399 408 8696333 10.1038/ng0896-399
Porto G Cardoso CS Gordeuk V Cruz E Fraga J Areias J Oliveira JC Bravo F Gangaidzo IT MacPhail AP Gomo ZA Moyo VM Melo G Silva C Justica B de Sousa M Clinical and genetic heterogeneity in hereditary haemochromatosis: association between lymphocyte counts and expression of iron overload Eur J Haematol 2001 67 110 118 11722599 10.1034/j.1600-0609.2001.t01-1-00481.x
Barton JC Acton RT Inheritance of two HFE mutations in African Americans: cases with hemochromatosis phenotypes and estimates of hemochromatosis phenotype frequency Genet Med 2001 3 294 300 11478530
Barton JC Acton RT HLA-A and -B alleles and haplotypes in hemochromatosis probands with HFE C282Y homozygosity in central Alabama BMC Med Genet 2002 3 9 12370085 10.1186/1471-2350-3-9
Barton JC Bertoli LF Acton RT Common variable immunodeficiency and IgG subclass deficiency in central Alabama hemochromatosis probands homozygous for HFE C282Y Blood Cells Mol Dis 2003 31 102 111 12850493 10.1016/S1079-9796(03)00116-5
Barton JC Bertoli LF Acton RT HLA-A and -B alleles and haplotypes in 240 index patients with common variable immunodeficiency and selective IgG subclass deficiency in central Alabama BMC Med Genet 2003 4 3 12803653 10.1186/1471-2350-4-3
Barton JC Barton NH Alford TJ Diagnosis of hemochromatosis probands in a community hospital Am J Med 1997 103 498 503 9428833 10.1016/S0002-9343(97)00276-3
Barton JC McDonnell SM Adams PC Brissot P Powell LW Edwards CQ Cook JD Kowdley KV Group HMW Management of hemochromatosis Ann Intern Med 1998 129 932 939 9867745
Barton JC Wiener HW Acton RT Go RC HLA haplotype A*03-B*07 in hemochromatosis probands with HFE C282Y homozygosity: frequency disparity in men and women and lack of association with severity of iron overload Blood Cells Mol Dis 2005 34 38 47 15607698 10.1016/j.bcmd.2004.08.022
Primary immunodeficiency diseases. Report of a WHO scientific group Immunodefic Rev 1992 3 195 236 1510837
McLaren CE Barton JC Adams PC Harris EL Acton RT Press N Reboussin DM McLaren GD Sholinsky P Walker AP Gordeuk VR Leiendecker-Foster C Dawkins FW Eckfeldt JH Mellen BG Speechley M Thomson E Hemochromatosis and Iron Overload Screening (HEIRS) study design for an evaluation of 100,000 primary care-based adults Am J Med Sci 2003 325 53 62 12589228 10.1097/00000441-200302000-00001
Dati F Schumann G Thomas L Aguzzi F Baudner S Bienvenu J Blaabjerg O Blirup-Jensen S Carlstrom A Petersen PH Johnson AM Milford-Ward A Ritchie RF Svendsen PJ Whicher J Consensus of a group of professional societies and diagnostic companies on guidelines for interim reference ranges for 14 proteins in serum based on the standardization against the IFCC/BCR/CAP Reference Material (CRM 470). International Federation of Clinical Chemistry. Community Bureau of Reference of the Commission of the European Communities. College of American Pathologists Eur J Clin Chem Clin Biochem 1996 34 517 520 8831057
Barton JC Shih WW Sawada-Hirai R Acton RT Harmon L Rivers C Rothenberg BE Genetic and clinical description of hemochromatosis probands and heterozygotes: evidence that multiple genes linked to the major histocompatibility complex are responsible for hemochromatosis Blood Cells Mol Dis 1997 23 135 145 9215758 10.1006/bcmd.1997.0129
Murphy CC Acton RT Barger BO Go RC Kirk KA Reitnauer PJ Roseman JM Population genetic analyses of insulin dependent diabetes mellitus using HLA allele frequencies Clin Genet 1983 23 405 414 6411398
Albert ED Baur MP Mayr WR Histocompatibility testing 1984 1984 Berlin, Springer-Verlag
SAS v. 9.0 2004 SAS Institute, Cary, NC
Box GEP Cox DR An analysis of transformations J Royal Stat Soc Series B (Methodological) 1964 26 211 252
Guo SW Thompson EA Performing the exact test of Hardy-Weinberg proportion for multiple alleles Biometrics 1992 48 361 372 1637966
Bain BJ Ethnic and sex differences in the total and differential white cell count and platelet count J Clin Pathol 1996 49 664 666 8881919
Cruz E Vieira J Goncalves R Alves H Almeida S Rodrigues P Lacerda R Porto G Involvement of the major histocompatibility complex region in the genetic regulation of circulating CD8 T-cell numbers in humans Tissue Antigens 2004 64 25 34 15191520 10.1111/j.1399-0039.2004.00245.x
Porto G Vicente C Teixeira MA Martins O Cabeda JM Lacerda R Goncalves C Fraga J Macedo G Silva BM Alves H Justica B de Sousa M Relative impact of HLA phenotype and CD4-CD8 ratios on the clinical expression of hemochromatosis Hepatology 1997 25 397 402 9021953
Reimao R Porto G de Sousa M Stability of CD4/CD8 ratios in man: new correlation between CD4/CD8 profiles and iron overload in idiopathic haemochromatosis patients C R Acad Sci III 1991 313 481 487 1760751
Barton JC Harmon L Rivers C Acton RT Hemochromatosis: association of severity of iron overload with genetic markers Blood Cells Mol Dis 1996 22 195 204 9075570 10.1006/bcmd.1996.0100
Crawford DH Powell LW Leggett BA Francis JS Fletcher LM Webb SI Halliday JW Jazwinska EC Evidence that the ancestral haplotype in Australian hemochromatosis patients may be associated with a common mutation in the gene Am J Hum Genet 1995 57 362 367 7668262
Piperno A Arosio C Fargion S Roetto A Nicoli C Girelli D Sbaiz L Gasparini P Boari G Sampietro M Camaschella C The ancestral hemochromatosis haplotype is associated with a severe phenotype expression in Italian patients Hepatology 1996 24 43 46 8707280
Porto G Reimao R Goncalves C Vicente C Justica B de Sousa M Haemochromatosis as a window into the study of the immunological system: a novel correlation between CD8+ lymphocytes and iron overload Eur J Haematol 1994 52 283 290 8020628
Arosa FA Oliveira L Porto G da Silva BM Kruijer W Veltman J de Sousa M Anomalies of the CD8+ T cell pool in haemochromatosis: HLA-A3-linked expansions of CD8+ Clin Exp Immunol 1997 107 548 554 9067531 10.1046/j.1365-2249.1997.d01-967.x
Miranda CJ Makui H Andrews NC Santos MM Contributions of beta2-microglobulin-dependent molecules and lymphocytes to iron regulation: insights from HfeRag1(-/-) and beta2mRag1(-/-) double knock-out mice Blood 2004 103 2847 2849 14656877 10.1182/blood-2003-09-3300
Hagiwara E Abbasi F Mor G Ishigatsubo Y Klinman DM Phenotype and frequency of cells secreting IL-2, IL-4, IL-6, IL-10, IFN and TNF-alpha in human peripheral blood Cytokine 1995 7 815 822 8664449 10.1006/cyto.1995.0098
Bruserud O Ehninger G Hamann W Pawelec G Secretion of IL-2, IL-3, IL-4, IL-6 and GM-CSF by CD4+ and CD8+ TCR alpha beta+ T-cell clones derived early after allogeneic bone marrow transplantation Scand J Immunol 1993 38 65 74 8327861
Nemeth E Valore EV Territo M Schiller G Lichtenstein A Ganz T Hepcidin, a putative mediator of anemia of inflammation, is a type II acute-phase protein Blood 2003 101 2461 2463 12433676 10.1182/blood-2002-10-3235
Porto G Cruz E Miranda HP Porto B Vasconcelos JC Lacerda R Roetto A Daraio F Bacelar C Growth hormone (GH)-induced reconstitution of CD8+ CD28+ T lymphocytes in a rare case of severe lymphopenia associated with Juvenile Haemochromatosis and Turner's syndrome Clin Endocrinol (Oxf) 2004 61 437 440 15473875 10.1111/j.1365-2265.2004.02069.x
Drakesmith H Schimanski LM Ormerod E Merryweather-Clarke AT Viprakasit V Edwards JP Sweetland E Bastin JM Cowley D Chinthammitr Y Robson KJ Townsend AR Resistance to hepcidin is conferred by hemochromatosis-associated mutations of ferroportin Blood 2005
Nemeth E Roetto A Garozzo G Ganz T Camaschella C Hepcidin is decreased in TFR2 hemochromatosis Blood 2005 105 1803 1806 15486069 10.1182/blood-2004-08-3042
Papanikolaou G Samuels ME Ludwig EH MacDonald ML Franchini PL Dube MP Andres L Macfarlane J Sakellaropoulos N Politou M Nemeth E Thompson J Risler JK Zaborowska C Babakaiff R Radomski CC Pape TD Davidas O Christakis J Brissot P Lockitch G Ganz T Hayden MR Goldberg YP Mutations in HFE2 cause iron overload in chromosome 1q-linked juvenile hemochromatosis Nat Genet 2004 36 77 82 14647275 10.1038/ng1274
Grady RW Akbar AN Giardina PJ Hilgartner MW de Sousa M Disproportionate lymphoid cell subsets in thalassaemia major: the relative contributions of transfusion and splenectomy Br J Haematol 1985 59 713 724 3872679
Cunningham-Rundles S Giardina PJ Grady RW Califano C McKenzie P de Sousa M Effect of transfusional iron overload on immune response J Infect Dis 2000 182 Suppl 1 S115 S121 10944493 10.1086/315919
Gordeuk V Mukiibi J Hasstedt SJ Samowitz W Edwards CQ West G Ndambire S Emmanual J Nkanza N Chapanduka Z . Iron overload in Africa. Interaction between a gene and dietary iron content N Engl J Med 1992 326 95 100 1727237
Melo RA Garcia AB Viana SR Falcao RP Lymphocyte subsets in experimental hemochromatosis Acta Haematol 1997 98 72 75 9286302
de Sousa M Porto G Arosa FA Cardoso C J.M. C Lacerda R Fraga J Barton JC and Edwards CQ T-lymphocyte expression and function in hemochromatosis Hemochromatosis Genetics, pathophysiology, diagnosis and treatment 2000 Cambridge, Cambridge University Press 396 407
Pratiwi R Fletcher LM Pyper WR Do KA Crawford DH Powell LW Jazwinska EC Linkage disequilibrium analysis in Australian haemochromatosis patients indicates bipartite association with clinical expression J Hepatol 1999 31 39 46 10424281 10.1016/S0168-8278(99)80161-5
Sachot S Moirand R Jouanolle AM Mosser J Fergelot P Deugnier Y Brissot P le Gall JY David V Low penetrant hemochromatosis phenotype in eight families: no evidence of modifiers in the MHC region Blood Cells Mol Dis 2001 27 518 529 11500063 10.1006/bcmd.2001.0394
Fargion S Valenti L Dongiovanni P Scaccabarozzi A Fracanzani AL Taioli E Mattioli M Sampietro M Fiorelli G Tumor necrosis factor alpha promoter polymorphisms influence the phenotypic expression of hereditary hemochromatosis Blood 2001 97 3707 3712 11389006 10.1182/blood.V97.12.3707
Beutler E Gelbart T Tumor necrosis factor alpha promoter polymorphisms and liver abnormalities of homozygotes for the 845G>A(C282Y) hereditary hemochromatosis mutation Blood 2002 100 2268 2269 12229882 10.1182/blood-2002-05-1520
Bryan CF Nishiya K Pollack MS Dupont B de Sousa M Differential inhibition of the MLR by iron: association with HLA phenotype Immunogenetics 1981 12 129 140 6451576 10.1007/BF01561656
Cardoso CS Alves H Mascarenhas M Goncalves R Oliveira P Rodrigues P Cruz E de Sousa M Porto G Co-selection of the H63D mutation and the HLA-A29 allele: a new paradigm of linkage disequilibrium? Immunogenetics 2002 53 1002 1008 11904676 10.1007/s00251-001-0414-8
Caruso C Bongiardina C Candore G Cigna D Romano GC Colucci AT Di Lorenzo G Gervasi F Manno M Potestio M Tantillo G HLA-B8,DR3 haplotype affects lymphocyte blood levels Immunol Invest 1997 26 333 340 9129986
Candore G Lio D GC R G C Pathogenesis of autoimmune diseases associated with 8.1 ancestral haplotype: effect of multiple gene interactions Autoimmun Rev 2002 1 29 35 12849055 10.1016/S1568-9972(01)00004-0
Kuntz BM Sobieraj B Bruster HT [T-helper (CD4+) cytopenia and HLA-A1, -B8, -C27, -DR3] Beitr Infusionsther Transfusionsmed 1994 32 310 313 9480115
McNeil AJ Yap PL Gore SM Brettle RP McColl M Wyld R Davidson S Weightman R Richardson AM Robertson JR Association of HLA types A1-B8-DR3 and B27 with rapid and slow progression of HIV disease QJM 1996 89 177 185 8731561
Hall MA Ahmadi KR Norman P Snieder H MacGregor AJ Vaughan RW Spector TD Lanchbury JS Genetic influence on peripheral blood T lymphocyte levels Genes Immun 2000 1 423 427 11196672 10.1038/sj.gene.6363702
Ahmadi KR Hall MA Norman P Vaughan RW Snieder H Spector TD Lanchbury JS Genetic determinism in the relationship between human CD4+ and CD8+ T lymphocyte populations? Genes Immun 2001 2 381 387 11704804 10.1038/sj.gene.6363796
Hall MA Norman PJ Thiel B Tiwari H Peiffer A Vaughan RW Prescott S Leppert M Schork NJ Lanchbury JS Quantitative trait loci on chromosomes 1, 2, 3, 4, 8, 9, 11, 12, and 18 control variation in levels of T and B lymphocyte subpopulations Am J Hum Genet 2002 70 1172 1182 11951176 10.1086/340090
Cunningham-Rundles C Bodian C Common variable immunodeficiency: clinical and immunological features of 248 patients Clin Immunol 1999 92 34 48 10413651 10.1006/clim.1999.4725
Di Renzo M Zhou Z George I Becker K Cunningham-Rundles C Enhanced apoptosis of T cells in common variable immunodeficiency (CVID): role of defective CD28 co-stimulation Clin Exp Immunol 2000 120 503 511 10844530 10.1046/j.1365-2249.2000.01239.x
Agematsu K Futatani T Hokibara S Kobayashi N Takamoto M Tsukada S Suzuki H Koyasu S Miyawaki T Sugane K Komiyama A Ochs HD Absence of memory B cells in patients with common variable immunodeficiency Clin Immunol 2002 103 34 42 11987983 10.1006/clim.2001.5197
|
16042809
|
PMC1198220
|
CC BY
|
2021-01-04 16:31:42
|
no
|
BMC Blood Disord. 2005 Jul 25; 5:5
|
utf-8
|
BMC Blood Disord
| 2,005 |
10.1186/1471-2326-5-5
|
oa_comm
|
==== Front
BMC CancerBMC Cancer1471-2407BioMed Central London 1471-2407-5-1101611783310.1186/1471-2407-5-110Research ArticleDistinctive serum protein profiles involving abundant proteins in lung cancer patients based upon antibody microarray analysis Gao Wei-Min [email protected] Rork [email protected] Randal P [email protected] David E [email protected] Ji [email protected] Alissa K [email protected] William N [email protected] Dean E [email protected] Gilbert S [email protected] Brian B [email protected] Samir M [email protected] Department of Internal Medicine, University of Michigan, Ann Arbor, MI 48109, USA2 Department of Critical Care Medicine, Safar Center for Resuscitation Research, University of Pittsburgh, Pittsburgh, PA 15260, USA3 Department of Pediatrics, University of Michigan, Ann Arbor, MI 48109, USA4 Van Andel Research Institute, Grand Rapids, MI 49503, USA5 Division of Pulmonary and Critical Care Medicine, NYU Cancer Institute, NYU School of Medicine NY, NY 10016, USA6 Fred Hutchinson Cancer Research Center, Seattle, WA 98109, USA2005 23 8 2005 5 110 110 15 6 2005 23 8 2005 Copyright © 2005 Gao 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
Cancer serum protein profiling by mass spectrometry has uncovered mass profiles that are potentially diagnostic for several common types of cancer. However, direct mass spectrometric profiling has a limited dynamic range and difficulties in providing the identification of the distinctive proteins. We hypothesized that distinctive profiles may result from the differential expression of relatively abundant serum proteins associated with the host response.
Methods
Eighty-four antibodies, targeting a wide range of serum proteins, were spotted onto nitrocellulose-coated microscope slides. The abundances of the corresponding proteins were measured in 80 serum samples, from 24 newly diagnosed subjects with lung cancer, 24 healthy controls, and 32 subjects with chronic obstructive pulmonary disease (COPD). Two-color rolling-circle amplification was used to measure protein abundance.
Results
Seven of the 84 antibodies gave a significant difference (p < 0.01) for the lung cancer patients as compared to healthy controls, as well as compared to COPD patients. Proteins that exhibited higher abundances in the lung cancer samples relative to the control samples included C-reactive protein (CRP; a 13.3 fold increase), serum amyloid A (SAA; a 2.0 fold increase), mucin 1 and α-1-antitrypsin (1.4 fold increases). The increased expression levels of CRP and SAA were validated by Western blot analysis. Leave-one-out cross-validation was used to construct Diagonal Linear Discriminant Analysis (DLDA) classifiers. At a cutoff where all 56 of the non-tumor samples were correctly classified, 15/24 lung tumor patient sera were correctly classified.
Conclusion
Our results suggest that a distinctive serum protein profile involving abundant proteins may be observed in lung cancer patients relative to healthy subjects or patients with chronic disease and may have utility as part of strategies for detecting lung cancer.
==== Body
Background
Lung cancer remains the leading cause of cancer mortality in the United States for both men and women [1,2]. Despite significant advances in understanding its biology and causes, the overall incidence of lung cancer is increasing, and improvements in outcome are not apparent [3]. As treatment is efficacious only for those patients who are diagnosed sufficiently early in the disease process, a significant reduction in patient mortality may result from earlier detection of lung cancer, including combinations of biomarkers with spiral CT imaging [2].
Identification of protein biomarkers in blood or serum may have utility for noninvasive disease detection and classification. Biomarker identification would be greatly enhanced by methodological improvements in protein detection. Direct serum protein profiling by matrix assisted laser desorption ionization (MALDI) mass spectrometry [4,5] has uncovered distinct mass profiles in several common types of cancer. However, the direct profiling of complex protein mixtures by MALDI has difficulties in providing the identification of the distinctive proteins. Further, given the limited dynamic range of MALDI, it is likely that distinctive features observed in serum with this approach represent relatively abundant proteins.
An alternative to mass spectrometry for protein profiling is the use of antibody microarrays. The field of protein microarrays currently encompasses applications that include profiling of serum and tissues from cancer patients [6,7], autoimmune diagnostics [8], protein interaction screening [9-12], as well as antibody-based detection of multiple antigens [13-17]. Recent increases in sensitivity and quantitative reproducibility has extended the utility of antibody microarrays [18,19]. In particular, direct multicolor labeling with rolling-circle amplification (RCA) has enabled enhanced sensitivity and reproducible measurements of low-abundance proteins, as compared to other direct or indirect labeling detection methods [20,21]. The strategy behind the two-color RCA detection is that two different protein samples can be labeled, respectively, with either biotin or digoxigenin, then both samples are co-hybridized to the antibody arrays. The bound proteins are then detected and individually quantitated, using RCA (and Cy3) to amplify fluorescence signals emanating from the bound biotin-labeled proteins, and RCA (and Cy5) to amplify signals from bound digoxigenin-labeled proteins. We have previously reported that, in comparison with either direct or indirect labeling detection, two-color RCA produced up to 30-fold higher fluorescence intensity measurements, enabling the reproducible measurements of lower abundance proteins in serum [20]. Importantly, we have been able to ascertain reproducible small expression differences between 2 different samples.
In the present study, we have utilized the RCA methodology and a panel of 84 antibodies to analyze the relative abundance of multiple proteins in sera from 24 newly diagnosed patients with lung cancer, 24 healthy controls, and 32 patients with chronic obstructive pulmonary disease (COPD). We have identified a distinctive serum protein profile for patients with lung cancers.
Methods
Serum samples
Serum samples were obtained following informed consent from 80 individuals, including 24 newly diagnosed lung cancer patients, 24 healthy subjects without a prior history of cancer, and 32 patients with COPD. All samples were collected under protocols approved by the local Institutional Review Board (IRB). The sera from lung cancer patients and healthy controls were collected through the Early Detection Research Network (EDRN) program at the University of Michigan. The sera from COPD patients were collected through the EDRN at New York University. All samples were stored frozen at -80°C prior to analysis.
Construction of antibody microarrays
The antibodies used in the preparation of the microarrays were purchased from various sources (a complete list and further information on each antibody is available online [22]). The 84 antibodies targeted 80 different proteins, present at a broad range of concentrations in serum, that could have levels associated with lung cancer, including acute phase reactants, proteases and protease inhibitors, immune system proteins, glycoproteins, extracellular matrix proteins, and cytokines. Microarray preparation was performed as described previously [20,23]. Briefly, samples (20 μl each) of 100–2000 μg/ml antibody solutions in PBS were prepared in polypropylene 384-well microtiter plates (MJ Research). Small amounts of each antibody solution were transferred to the surface of nitrocellulose-coated microscope slides (PATH slides, Gentel Biosurfaces) using a piezoelectric non-contact spotter (Biochip Arrayer, PerkinElmer Life Sciences). Twelve identical arrays were printed on each of seven slides; each array consisted of 96 antibodies or control proteins ("baits") printed in triplicate to form an 18 by 16 array of dots.
Serum labeling
An aliquot (1 μl) from each of 80 serum samples was labeled with N-hydroxysuccinimide (NHS)-Digoxigenin (Molecular Probes), and a second aliquot (1 μl) was labeled with NHS-biotin (Molecular Probes). Each 1 μl serum aliquot was diluted with 14 μl PBS containing 500 μM NHS-biotin or NHS-Digoxigenin. After the reactions had proceeded for 1 h on ice, 5 μl of 1 M Tris-HCl (pH 7.5) was added to each tube to quench the reactions, then the solutions were incubated on ice for an additional 20 min. Non-reactive dye molecules were removed by passing each solution through a size-exclusion chromatography spin column (Bio-Spin P6, Bio-Rad) with a molecular weight cutoff of 6 kDa. The digoxigenin-labeled samples were pooled, then distributed equally among the biotin-labeled samples. 4 μl of Tris-buffered saline (TBS) containing Super Block (Pierce), 1% Brij-35, and 1% Tween-20 was added to each sample, after which the total volume of each sample was adjusted to 40 μl with TBS.
Processing of antibody microarrays
Each of the 12 arrays on a slide was circumscribed with a wax border to segregate the arrays from each other. The slides were rinsed twice in PBS with 0.5% Tween-20 (PBST0.5) and then blocked 1 h at 4°C in PBS containing 0.1% Tween-20 (PBST0.1), 0.3% CHAPS, and 1% BSA. After the arrays were briefly rinsed twice with PBST0.5 and dried by centrifugation, 40 μl of each labeled serum sample mix was incubated on an array with gentle rocking at room temp for 1 h. The three groups of samples were arranged so as to balance the types of samples on each slide, as shown in a supplementary table. The arrays were rinsed in PBST0.1, briefly washed three times in PBST0.1, then dried by centrifugation. Mouse monoclonal anti-Biotin (Jackson ImmunoResearch) was covalently conjugated to a 20-base oligonucleotide (primer 1) as previously described [20]. Molecular Staging (New Haven, CT) kindly provided the other reagents necessary for RCA detection. These included a mouse monoclonal anti-Digoxigenin (Roche) antibody conjugated to a different 20-base oligonucleotide (primer 4.2), an 81-base circular DNA (circle 1) with a portion complementary to primer 1, and an 80-base circular DNA (circle 4.2) with a portion complementary to primer 4.2. The sequences of the primers, circles and decorators can be found in the supplementary information for Zhou et al. [20]. The microarrays were incubated for 1 h at room temp in PBST0.1 containing 1 mM EDTA, 5 mg/ml BSA, 75 nM circle 1, 75 nM circle 4.2, 1.0 μg/ml primer 1-conjugated anti-biotin, and 1.0 μg/ml primer 4.2-conjugated anti-Digoxigenin. The arrays were rinsed briefly in PBST0.1 then washed at room temp with gentle rocking three times for 3 min each in PBST0.1, after which they were incubated in 1X Tango buffer (Fermentas, Hanover, MD) containing Phi29 DNA polymerase (New England Biolabs), 0.1% Tween-20 and 0.4 mM dNTPs at 37°C for 30 min. Following a brief rinse in 2X SSC with 0.1% Tween-20 (SSCT0.1), the arrays were washed three times for 3 min each at room temperature with gentle rocking in 2X SSCT0.1, then dried by centrifugation. The arrays were incubated for 1 h (37°C) in 2X SSCT0.1 containing 0.5 mg/ml herring sperm DNA, Cy3-labeled 18-bp oligonucleotide complementary to the repeating DNA strand from primer 1 and a Cy5-labeled 22-bp oligonucleotide complementary to the repeating DNA strand from primer 4.2, each at 0.1 mM. The arrays were briefly rinsed in 2X SSCT0.1, washed three times for 3 min each at room temperature in 2X SSCT0.1, dried by centrifugation, then scanned (ScanArray, PerkinElmer Life Sciences).
Analysis
The Cy3 and Cy5 fluorescence was quantified using GenePix software (Axon Instruments). Of the total of 24192 dots, 206 were excluded as having defects by visually inspecting the images without reference to the quantitative data, with the most common cause of the defect being overlapping dots. The resultant ".gpr" files for each array were parsed to create a spreadsheet of the raw data, available as a supplement [22]. We took the negative of the base-2 logarithm of the "median of ratios" computed by the software, and averaged the triplicate measures for each bait, not including the excluded dots. This gave the average of the log-ratio of the sample (Cy3) to the standard pool (Cy5), hereafter referred to as the values.
We first performed a normalization in which the median value for each array was subtracted from all the values for that sample. Some antibodies displayed biases in favor of either the Cy3 or Cy5 channel, or showed large differences between groups. Consequently, we selected a subset of 48 antibodies that did not have large differences between groups, and had small within-group standard deviations in order to perform a normalization that would be less affected by antibodies with variable data or channel biases. We computed the average of the raw values for each antibody using the 80 arrays, and normalized the individual slides to this standard. For each slide, the median of the 48 differences for the array minus the corresponding values on the standard was subtracted from the array, subtraction being used rather than division because the values were already log-transformed. The averaged raw and normalized data are available as supplemental information [22].
Western blot analysis
We used Western blots to analyze the level of C-reactive protein (CRP) and serum-amyloid A (SAA) in sera of eight selected lung cancer patients and eight healthy controls. Subsequently, in order to validate our findings, we also analyzed the CRP and SAA levels in an independent set of 30 additional lung cancer patients and 30 additional healthy controls. Briefly, 5 μl of serum (from each patient) was resolved by 15% SDS-PAGE, and then transferred to a PVDF membrane. Following incubation in blocking buffer (PBST0.1 containing 2% nonfat dry milk (Bio-Rad)) for 2 h, the membrane was hybridized in blocking buffer containing either anti-CRP or anti-SAA mouse monoclonal antibodies at 0.5 μg/ml and 0.25 μg/ml for 1 h. The membrane was then washed and incubated with a horseradish peroxidase-conjugated sheep anti-mouse IgG (Amersham) at a 1:1000 dilution for 1 h. After washing, the membrane was briefly incubated in ECL (Enhanced Chemiluminescence, Amersham), then exposed to imaging film (Amersham). Integrated intensity measurements were made of the respective bands and the measurements were further analyzed statistically.
Results
Using microarrays containing 84 antibodies printed in triplicate on slides, we measured the amount of target protein bound from 80 individual sera, with each sample being compared to a pooled reference sample (consisting of a mixture of all of the sera) in a two-color assay. Figure 1 shows a representative image of antibody arrays from one slide. Eighty arrays with 24 sera from lung cancer patients, 24 normal sera, or 32 sera from patients with COPD were analyzed. The values determined were the normalized average of base-2 logarithms of the intensity arising from the individual sample divided by the intensity arising from the pooled sample, which was measured as Cy3 and Cy5 fluorescence, respectively. Values from triplicate antibody dots from the same array were quite reproducible, with average standard deviations of 0.14, corresponding to approximately 10% variation in the ratios.
Figure 2 depicts the first three principal components obtained using all 84 antibodies. While lung cancer patients were largely separated from the other two groups of patients, there was no clear separation between COPD and normal. This completely unsupervised view of the data indicates that the distinction between lung tumor patients' sera and the two other groups of sera was likely the largest source of variation in the data set (Figure 2A). The somewhat outlying samples were not associated with a particular microarray slide (Figure 2B) or brightness of the signals for either fluorescence. The first principal component was most highly correlated with C-reactive protein (CRP) and serum amyloid A (SAA).
In order to determine which antibodies distinguished sera of lung tumor patients from the other sera, we fit a 1-way analysis of variance model to the three groups of samples. Cancer patient sera gave significantly different mean values for 7/84 antibodies when compared to normal sera, and for 8/84 of the antibodies when compared to the COPD sera (both at p < 0.01). The 7 antibodies that yielded differences in the abundance of their corresponding proteins between tumor and normal sera were common to the group of 8 antibodies that yielded differences in the abundance of their corresponding proteins between tumor and COPD sera. The additional protein identified by the COPD comparison is troponin 1. We found increased levels of CRP, SAA, α-1-antitrypsin (AAT) by two distinct antibodies, and MUC1, and decreased levels of transferrin and gelsolin, in lung cancer sera (Table 1). Results obtained for the entire set of antibodies are available as supplemental data [22]. To assess the significance of these findings, we randomly permutated the sample labels 1000 times and performed the identical analysis on each resulting data set. On average this yielded only 0.1 antibodies for which the tumor samples were increased or decreased (at p < 0.1) compared to both other groups, with 1 or more significant antibody found in only 8.1% of the permuted data sets. Therefore, it is very unlikely that the occurrence of differences in levels of proteins for the 7 antibodies observed in the actual data is due to chance. The correlation within the group of lung cancer patients between the CRP, SAA, AAT, MUC1, transferrin and gelsolin data values are summarized in Table 2, and the two-dimensional log-scale plots for CRP and MUC1, and SAA and AAT are shown in Figure 3. The expression levels of CRP, SAA and AAT but not MUC1 were correlated with each other (r > 0.4, p < 0.05). The two AAT measurements, each derived from a different antibody, were significantly correlated (r = 0.72, p < 0.001).
We performed a leave-one-out validation of a Diagonal Linear Discriminant Analysis (DLDA) classifier that discriminates tumor vs. non-tumor samples [23]. We left out one sample at a time, then used the remaining 79 samples to select the 5 antibodies with values increased in tumor patient samples according to the p-values for 2-sample T-tests of tumor vs. non-tumor samples, and constructed the resulting discriminant function based on the 79 samples. When using all of the data CRP, SAA, MUC1, and 2 AAT antibodies would be selected as the top antibodies, in that order. The value of this function was then computed for the left out sample. Figure 4 shows the resulting Receiver Operating Characteristic (ROC) curve that was obtained. The calculations were also repeated using only the best 3 antibodies. Using 5 antibodies, the correct classification of all 56 of the non-tumor samples was associated with the correct classification of 15 of 24 cancer patient sera. We obtained the same result with a different classifier that used majority voting among the 5 closest neighboring samples, where the distances were computed after scaling each antibody's values by the pooled estimate of the standard deviation (in analogy to DLDA). Analogous results from cross-validating this simpler classifier using only the 3 best antibodies correctly classified 17 of 24 cancer patient sera, while misclassifiying 4 of 56 non-tumor samples, which also corresponds approximately to a point on the ROC curve for the DLDA classifier when it used 3 antibodies. This illustrates that the results obtained with DLDA classifiers were not particularly better than could be obtained with other simple methods.
CRP and SAA were selected for Western blot analysis in order to validate the specificity of antibody microarrays. Eight lung cancer sera and 8 normal sera were resolved by SDS-PAGE, then transferred to PVDF membranes. The membranes were probed with anti-CRP or anti-SAA antibodies. As shown in Figure 5, all of the sera from patients with lung cancer showed much higher levels of CRP and SAA compared to the sera from healthy controls. Subsequently, in order to validate our findings, we also analyzed the CRP and SAA levels in an independent set of 30 additional lung cancer patients and 30 additional healthy controls. Integrated intensity measurements were made of the respective bands and the measurements were further analyzed statistically. The distribution of integrated intensity measurement values obtained from the two groups of samples for both assays are shown in Figure 6. The number of tumor samples with values greater than the largest value for normal samples was 17/30 for CRP (p = 3.1 × 10-7) and 13/30 for SAA (p = 2.3 × 10-5).
Discussion
Four proteins were found to be more abundant in the lung cancer samples than those of the controls, namely CRP (13.3 fold), SAA (2.0 fold), AAT (1.4 fold) and MUC1 (1.4 fold). There were no significant protein expression differences observed in serum between the various lung cancer subtypes examined (adenocarcinoma, squamous and small cell carcinomas: data not shown). The significant increases in CRP and SAA protein levels found in the serum of lung cancer patients by protein microarray were confirmed by immunoassay. The increased levels of AAT in lung cancer patient sera (1.4 fold) were observed using two different antibodies, each obtained from a separate source.
The pattern of increased abundances of CRP, SAA, AAT and MUC1 in lung cancer patient sera that were observed in our microarray-based study is concordant with previous studies of individual proteins. An increased C-reactive protein level is part of the acute-phase response to most forms of inflammation, infection, tissue damage, and malignant neoplasia [25-27]. CRP [Uniprot PO2741] forms homopentamers (pentaxins); it promotes phagocytosis and complement fixation through calcium-dependent binding (two per 23 kDa subunit) to phosphorylcholine. CRP also interacts with DNA and histones to scavenge nuclear material from damaged circulating cells. The expression of CRP is induced by IL-1 and IL-6. While CRP itself is likely not useful as a single assay, it may have clinical utility as part of a panel of diagnostic biomarkers, especially in evaluating results from spiral CT imaging [2]. CRP is mainly expressed in hepatocytes; cytokines, especially interleukin-6, induce the expression and release of CRP [28,29]. CRP has been suggested as a useful prognostic indicator in esophageal carcinoma [30]. Studies also showed that CRP was an independent determinant of survival in non-small-cell lung cancer [31] and could be useful in the initial evaluation of patients with small cell lung cancer and in monitoring response to therapy [32].
Serum amyloid A [Uniprot PO2735] is an acute-phase protein that occurs in various isoforms in a molecular mass range of 11–14 kDa. SAA is produced by hepatocytes [33], secreted into serum and rapidly binds to high-density lipoprotein, with 90% occurring in the bound form [34]. SAA occurs at low levels in sera of healthy individuals [35]. Patients with neoplastic disease, including lung [36], renal [37], colorectal [38], prostate [39] and nasopharyngeal cancers [40] exhibit a dramatic elevation of serum SAA. However, SAA is not a cancer-specific marker per se. Its elevation in serum has been reported also in association with trauma, infection, inflammation, rheumatoid arthritis, and amyloidosis [41]. A study of 621 subjects with cancer found substantial increases of SAA levels in >95% (281 of 289) of patients with metastatic solid tumors, all myelocytic leukemia patients and all advanced lymphoma patients [42]. Interestingly, SAA was not elevated in the group of 32 COPD patients included in this study, suggesting a potential utility of SAA in distinguishing between the two conditions possibly due to a different cytokine profile between the two groups.
α-1-antitrypsin [A1AT/SERPINA1, Uniprot PO1009] is a secretory glycoprotein of molecular weight 44 kDa produced in the liver. It neutralizes the effects of proteases in several organ systems, mainly in the lung. The major physiological role of AAT in the lung is to bind and inhibit elastase released from leucocytes in the lower respiratory tract, thereby preventing the destruction of lung tissue [43,44]. The normal range of serum or plasma AAT concentrations is 1200–2000 mg/L, with large increases in inflammatory conditions, infections, cancer, liver disease, or pregnancy [43]. It was previously reported that the serum concentration of AAT increased with tumor growth and could be utilized following tumor resection as an indicator of relapse [45,46]. The prognostic significance of AAT expression in lung adenocarcinomas has been evaluated using immunohistochemistry [47]; strongly AAT-positive cases had a worse prognosis than weak-to-moderately AAT-positive or AAT-negative cases, suggesting that increased AAT expression in lung adenocarcinoma patients may be a prognostic indicator. The biological basis for the association of acute-phase proteins, including CRP, SAA, and AAT, with lung cancer remains largely unknown. The correlation between CRP, SAA, and AAT levels was significant (r > 0.4), likely reflecting a host response. Significantly higher levels occur in patients with metastatic disease compared to patients with limited disease [48].
We found serum MUC1 levels to be modestly elevated in lung cancer compared to controls. MUC1 [P15941] is a membrane-bound mucin of 122 kDa molecular weight with several interacting isozymes, polymorphic tandem repeats, and an extensively O-glycosylated core protein [49]. In vitro studies suggested that MUC1 reduces E-cadherin-mediated cell-cell adhesion by steric hindrance, which increases metastatic ability [50]. High MUC1 levels also reduces the integrin-mediated cell adhesion to the extracellular matrix [51]. The clinical importance of the MUC1 glycoprotein, however, is not clear. Previous studies have reported that MUC1 was developmentally regulated and aberrantly expressed by carcinomas, and a high level of MUC1 mRNA expression in adenocarcinoma has been associated with poor prognosis [52-58]. MUC1 was also found to be up-regulated in non-small-cell lung cancer [59-61]. MUC1 is shed into the blood stream and thus has a potential as a tumor marker, as demonstrated in breast cancer [62-64]. Consistent with this finding, we observed higher MUC1 expression levels in the sera of lung cancer patients than in either healthy subjects or patients with COPD. Additionally, MUC1 expression levels did not show significant correlation with CRP, SAA, or AAT, suggesting that the increased MUC1 levels might be due to a different biological process. Interestingly, MUC1 serum levels in breast cancer patients were not concordant with the levels observed in tumor tissues by immunohistochemistry [64,65], so the increased serum MUC1 expression may correspond to a specific isoform expressed by cancer cells. Thus, expression levels of the different MUC1 isoforms and their epitopes may need to be evaluated to fully explain the increased levels in serum of lung cancer patients.
Other acute-phase reactant serum proteins that have been reported as significantly elevated in certain cancers were not increased in this study of sera from lung cancer patients. Most notably, the alpha sub-unit of haptoglobin (MW 11.7 kDa) and isoforms of the haptoglobin-1 precursor (HAP1) have been reported to be increased in serum of patients with ovarian and other gynecologic cancers [66,67].
Conclusion
Our results suggest that a distinctive serum protein profile involving relatively abundant proteins may be observed in cancer patients relative to healthy subjects or patients with chronic disease. It is therefore likely that distinctive mass peak profiles observed by mass spectrometry in cancer sera relative to control and that may be predictive of outcome include a significant component related to host response to tumors and acute phase reactants. The extent to which such indicators of host response have clinical utility as a group, together with other tumor biomarkers remains to be determined. The use of antibody microarrays directed against a broad range of serum and lung tumor proteins would have utility for elucidating those proteins with the greatest diagnostic utility.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
W-MG carried out the antibody microarray studies, participated in the statistical treatment of the data and drafted the manuscript. RK performed the statistical analysis of the data and in drafting the manuscript. RPO participated in the antibody microarray studies. DEM participated in the design of the study, helped coordinate serum acquisition/usage and in drafting the manuscript. JQ carried out the independent validation of CRP and SAA quantitation. AKG and WNR participated in the acquisition of the COPD serum. DEB participated in the acquisition of both the normal and control serum. GSO participated in the design of the study and helped to draft the manuscript. BBH participated in the design of the study and its coordination, and helped to draft the manuscript. SMH conceived of the study, participated in its design, 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 grant MTTC GR356.
Figures and Tables
Figure 1 Scanned fluorescence image of an antibody microarray detected by two-color RCA. 96 baits including 84 antibodies were spotted onto microscope slides coated with nitrocellulose. 12 identical arrays were printed on each of seven slides. Each antibody was printed in triplicate on each array in order to form an 18 by 16 array of dots. A test sample labeled with biotin and a pooled reference sample labeled with digoxigenin were co-incubated on the microarray, and bound proteins from both samples were detected by RCA. The microarray was scanned for Cy3 fluorescence (from the test sample) and Cy5 fluorescence (from the reference sample).
Figure 2 The first 3 principal components from normalized log-base-2 ratios of sample to reference pool intensities, using all 84 antibodies. The full 3-dimensional figures that can be rotated are available in the supplementary materials. In A, normal, COPD and lung cancer patients are marked with yellow, blue and red, respectively. The first three principal components account for 43% of the variance. In B, seven slides are marked separately with blue, black, yellow, green, purple, brown and red.
Figure 3 Two-dimensional plots of normalized log-base-2 ratios of sample to reference pool intensities for CRP and MUC1, and SAA and AAT.
Figure 4 Receiver Operating Characteristic (ROC) curves from leave-one-out validation of a Diagonal Linear Discriminant Analysis classifier using the best 3 (or 5) antibodies. Both the antibodies selected and the discriminant function were based solely on the remaining 79 samples.
Figure 5 SDS-PAGE Western blot analysis of CRP and SAA. CRP and SAA levels in sera of eight lung cancer patients and eight healthy controls were analyzed. The sera chosen were those that gave extremely high or low values for the corresponding assay on the antibody microarrays.
Figure 6 A scatter plot of integrated intensity measurements derived from western blots of an independent set of sera from 30 additional lung cancer patients and 30 additional healthy controls, probed for SAA and CRP. Values are base two logarithms of the relative band intensities after adding 0.1 to each value (to force values to be greater than 0).
Table 1 Results for 7 antibodies showing significant differences between both lung tumor patients vs. normal controls and lung tumor patients vs. COPD patients.
Fold difference in means P-values from 1-way ANOVA P-value from 2-sample T-test
Antibody Tumor / Normal Tumor / COPD Tumor vs. Normal Tumor vs. COPD Tumor vs. Others
CRP 13.6 13.0 1.1 × 10-9 2.2 × 10-10 4.1 × 10-12
SAA 1.99 2.15 1.8 × 10-7 1.7 × 10-9 2.5 × 10-10
MUC1 1.30 1.42 3.3 × 10-3 4.3 × 10-5 5.3 × 10-5
AAT (1) 1.34 1.35 9.1 × 10-4 3.1 × 10-4 7.9 × 10-5
AAT (2) 1.42 1.33 1.1 × 10-3 3.7 × 10-3 5.2 × 10-4
Transferrin 0.73 0.71 2.7 × 10-4 2.3 × 10-5 7.2 × 10-6
Gelsolin 0.77 0.77 5.8 × 10-3 4.7 × 10-3 1.4 × 10-3
Table 2 Correlation between CRP, SAA, AAT, MUC1, and Transferrin protein expression in the serum of lung tumor patients.
SAA AAT (1) AAT (2) MUC1 Transferrin
r p value r p value r p value r p value r p value
CRP 0.60 0.0019 0.62 0.0012 0.61 0.0015 0.19 0.37 -0.175 0.41
SAA 0.48 0.019 0.41 0.049 -0.07 0.74 -0.121 0.57
AAT (1) 0.72 6.63 × 10-5 -0.02 0.94 0.012 0.96
AAT (2) 0.17 0.44 0.042 0.85
MUC1 0.226 0.29
Transferrin
==== Refs
Society, AC Cancer Facts and Figures 2004 Atlanta, GA 2004
Omenn GS Human lung cancer chemoprevention strategies: Parker B. Francis Lecture Chest 2004 125 123S 127S 15136459 10.1378/chest.125.5_suppl.123S
Carbone DP The biology of lung cancer Semin Oncol 1997 24 388 401 9280218
Xiao X Liu D Tang Y Guo F Xia L Liu J He D Development of proteomic patterns for detecting lung cancer Dis Markers 2003 19 33 39 14757945
Zhukov TA Johanson RA Cantor AB Clark RA Tockman MS Discovery of distinct protein profiles specific for lung tumors and pre-malignant lung lesions by SELDI mass spectrometry Lung Cancer 2003 40 267 279 12781425
Knezevic V Leethanakul C Bichsel VE Worth JM Prabhu VV Gutkind JS Liotta LA Munson PJ Petricoin EF 3rdKrizman DB Proteomic profiling of the cancer microenvironment by antibody arrays Proteomics 2001 1 1271 1278 11721638 10.1002/1615-9861(200110)1:10<1271::AID-PROT1271>3.3.CO;2-Y
Sreekumar A Nyati MK Varambally S Barrette TR Ghosh D Lawrence TS Chinnaiyan AM Profiling of cancer cells using protein microarrays: discovery of novel radiation-regulated proteins Cancer Res 2001 61 7585 7593 11606398
Joos TO Schrenk M Hopfl P Kroger K Chowdhury U Stoll D Schorner D Durr M Herick K Rupp S Sohn K Hammerle H A microarray enzyme-linked immunosorbent assay for autoimmune diagnostics Electrophoresis 2000 21 2641 2650 10949141 10.1002/1522-2683(20000701)21:13<2641::AID-ELPS2641>3.0.CO;2-5
MacBeath G Schreiber SL Printing proteins as microarrays for high-throughput function determination Science 2000 289 1760 1763 10976071
Ge H UPA, a universal protein array system for quantitative detection of protein-protein, protein-DNA, protein-RNA and protein-ligand interactions Nucleic Acids Res 2000 28 e3 10606671 10.1093/nar/28.2.e3
Lueking A Horn M Eickhoff H Bussow K Lehrach H Walter G Protein microarrays for gene expression and antibody screening Anal Biochem 1999 270 103 111 10328771 10.1006/abio.1999.4063
Zhu H Bilgin M Bangham R Hall D Casamayor A Bertone P Lan N Jansen R Bidlingmaier S Houfek T Mitchell T Miller P Dean RA Gerstein M Snyder M Global analysis of protein activities using proteome chips Science 2001 293 2101 2105 11474067 10.1126/science.1062191
Haab BB Dunham MJ Brown PO Protein microarrays for highly parallel detection and quantitation of specific proteins and antibodies in complex solutions Genome Biol 2001 2 RESEARCH0004 11182887 10.1186/gb-2001-2-2-research0004
Mendoza LG McQuary P Mongan A Gangadharan R Brignac S Eggers M High-throughput microarray-based enzyme-linked immunosorbent assay (ELISA) Biotechniques 1999 27 778 780 782–776,788. 10524321
de Wildt RM Mundy CR Gorick BD Tomlinson IM Antibody arrays for high-throughput screening of antibody-antigen interactions Nat Biotechnol 2000 18 989 994 10973222 10.1038/79494
Arenkov P Kukhtin A Gemmell A Voloshchuk S Chupeeva V Mirzabekov A Protein microchips: use for immunoassay and enzymatic reactions Anal Biochem 2000 278 123 131 10660453 10.1006/abio.1999.4363
Huang RP Huang R Fan Y Lin Y Simultaneous detection of multiple cytokines from conditioned media and patient's sera by an antibody-based protein array system Anal Biochem 2001 294 55 62 11412006 10.1006/abio.2001.5156
Haab BB Methods and applications of antibody microarrays in cancer research Proteomics 2003 3 2116 2122 14595810 10.1002/pmic.200300595
Haab BB Zhou H Multiplexed protein analysis using spotted antibody microarrays Methods Mol Biol 2004 264 33 45 15020778
Zhou H Bouwman K Schotanus M Verweij C Marrero JA Dillon D Costa J Lizardi P Haab BB Two-color, rolling-circle amplification on antibody microarrays for sensitive, multiplexed serum-protein measurements Genome Biol 2004 5 R28 15059261 10.1186/gb-2004-5-4-r28
Schweitzer B Wiltshire S Lambert J O'Malley S Kukanskis K Zhu Z Kingsmore SF Lizardi PM Ward DC Inaugural article: immunoassays with rolling circle DNA amplification: a versatile platform for ultrasensitive antigen detection Proc Natl Acad Sci U S A 2000 97 10113 10119 10954739 10.1073/pnas.170237197
Detailed antibody, patient serum and experimental information
Miller JC Zhou H Kwekel J Cavallo R Burke J Butler EB Teh BS Haab BB Antibody microarray profiling of human prostate cancer sera: antibody screening and identification of potential biomarkers Proteomics 2003 3 56 63 12548634 10.1002/pmic.200390009
Dudoit S Fridlyand J Speed TP Comparison of discrimination methods for the classification of tumors using gene expression data Journal of the American Statistical Association 2002 97 77 87 10.1198/016214502753479248
Pepys MB Hirschfield GM C-reactive protein: a critical update J Clin Invest 2003 111 1805 1812 12813013 10.1172/JCI200318921
Danesh J Wheeler JG Hirschfield GM Eda S Eiriksdottir G Rumley A Lowe GD Pepys MB Gudnason V C-reactive protein and other circulating markers of inflammation in the prediction of coronary heart disease N Engl J Med 2004 350 1387 1397 15070788 10.1056/NEJMoa032804
Tall AR C-reactive protein reassessed N Engl J Med 2004 350 1450 1452 15070796 10.1056/NEJMe048020
McKeown DJ Brown DJ Kelly A Wallace AM McMillan DC The relationship between circulating concentrations of C-reactive protein, inflammatory cytokines and cytokine receptors in patients with non-small-cell lung cancer Brit J Cancer 2004 91 1993 1995 15570310 10.1038/sj.bjc.6602248
Scott HR McMillan DC Crilly A McArdle CS Milroy R The relationship between weight loss and interleukin 6 in non-small-cell lung cancer Brit J Cancer 1996 73 1560 1562 8664130
Nozoe T Saeki H Sugimachi K Significance of preoperative elevation of serum C-reactive protein as an indicator of prognosis in esophageal carcinoma Am J Surg 2001 182 197 201 11574097 10.1016/S0002-9610(01)00684-5
Kato K Hitsuda Y Kawasaki Y Igishi T Yasuda K Mikami M Watanabe M Miyata M Sasaki T Shimizu E [The value of serum C-reactive protein as a survival determinant in patients with advanced non-small-cell lung cancer] Nihon Kokyuki Gakkai Zasshi 2000 38 575 580 11061080
Arpin D Trillet-Lenoir V Lasset C Souquet PJ Riou R Yoanidis I Guidollet J Cordier JF Brune J [Value of C-reactive protein determination in small cell lung cancer] Bull Cancer 1993 80 1063 1068 8081044
Raynes JG Eagling S McAdam KP Acute-phase protein synthesis in human hepatoma cells: differential regulation of serum amyloid A (SAA) and haptoglobin by interleukin-1 and interleukin-6 Clin Exp Immunol 1991 83 488 491 1706240
Bausserman LL Herbert PN Rodger R Nicolosi RJ Rapid clearance of serum amyloid A from high-density lipoproteins Biochim Biophys Acta 1984 792 186 191 6696928
d'Eril GM Anesi A Maggiore M Leoni V Biological variation of serum amyloid A in healthy subjects Clin Chem 2001 47 1498 1499 11468251
Biran H Friedman N Neumann L Pras M Shainkin-Kestenbaum R Serum amyloid A (SAA) variations in patients with cancer: correlation with disease activity, stage, primary site, and prognosis J Clin Pathol 1986 39 794 797 3734116
Kimura M Tomita Y Imai T Saito T Katagiri A Ohara-Mikami Y Matsudo T Takahashi K Significance of serum amyloid A on the prognosis in patients with renal cell carcinoma Cancer 2001 92 2072 2075 11596022 10.1002/1097-0142(20011015)92:8<2072::AID-CNCR1547>3.0.CO;2-P
Glojnaric I Casl MT Simic D Lukac J Serum amyloid A protein (SAA) in colorectal carcinoma Clin Chem Lab Med 2001 39 129 133 11341746 10.1515/CCLM.2001.022
Kaneti J Winikoff Y Zimlichman S Shainkin-Kestenbaum R Importance of serum amyloid A (SAA) level in monitoring disease activity and response to therapy in patients with prostate cancer Urol Res 1984 12 239 241 6516095 10.1007/BF00256147
Cho WC Yip TT Yip C Yip V Thulasiraman V Ngan RK Yip TT Lau WH Au JS Law SC Cheng WW Ma VW Lim CK Identification of serum amyloid a protein as a potentially useful biomarker to monitor relapse of nasopharyngeal cancer by serum proteomic profiling Clin Cancer Res 2004 10 43 52 14734450
Benson MD Cohen AS Serum amyloid A protein in amyloidosis, rheumatic, and enoplastic diseases Arthritis Rheum 1979 22 36 42 103558
Rosenthal CJ Sullivan LM Serum amyloid A to monitor cancer dissemination Ann Intern Med 1979 91 383 390 289303
Crystal RG The alpha 1-antitrypsin gene and its deficiency states Trends Genet 1989 5 411 417 2696185 10.1016/0168-9525(89)90200-X
Sun Z Yang P Role of imbalance between neutrophil elastase and alpha 1-antitrypsin in cancer development and progression Lancet Oncol 2004 5 182 190 15003202 10.1016/S1470-2045(04)01414-7
Homolka J Voslarova Z Malbohan I [Significance of alpha-1-antitrypsin and alpha-2-pregnancy-associated glycoprotein in the serum of patients with bronchial carcinoma] Z Erkr Atmungsorgane 1987 169 38 43 3118584
Di Martino G Iannucci F Bizzarro A Iacono G [Association of serum tumor markers in solid neoplasms (CEA, ferritin, alpha 1-antitrypsin, parathormone and calcitonin)] Boll Ist Sieroter Milan 1982 61 411 422 6985416
Higashiyama M Doi O Kodama K Yokouchi H Tateishi R An evaluation of the prognostic significance of alpha-1-antitrypsin expression in adenocarcinomas of the lung: an immunohistochemical analysis Br J Cancer 1992 65 300 302 1739634
Weinstein PS Skinner M Sipe JD Lokich JJ Zamcheck N Cohen AS Acute-phase proteins or tumour markers: the role of SAA, SAP, CRP and CEA as indicators of metastasis in a broad spectrum of neoplastic diseases Scand J Immunol 1984 19 193 198 6200925
Gendler SJ Spicer AP Epithelial mucin genes Annu Rev Physiol 1995 57 607 634 7778880 10.1146/annurev.ph.57.030195.003135
Wesseling J van der Valk SW Hilkens J A mechanism for inhibition of E-cadherin-mediated cell-cell adhesion by the membrane-associated mucin episialin/MUC1 Mol Biol Cell 1996 7 565 577 8730100
Wesseling J van der Valk SW Vos HL Sonnenberg A Hilkens J Episialin (MUC1) overexpression inhibits integrin-mediated cell adhesion to extracellular matrix components J Cell Biol 1995 129 255 265 7698991 10.1083/jcb.129.1.255
Wolman SR Pauley RJ Mohamed AN Dawson PJ Visscher DW Sarkar FH Genetic markers as prognostic indicators in breast cancer Cancer 1992 70 1765 1774 1355402
Sasaki M Nakanuma Y Shimizu K Izumi R Pathological and immunohistochemical findings in a case of mucinous cholangiocarcinoma Pathol Int 1995 45 781 786 8563941
Ajioka Y Allison LJ Jass JR Significance of MUC1 and MUC2 mucin expression in colorectal cancer J Clin Pathol 1996 49 560 564 8813954
McGuckin MA Walsh MD Hohn BG Ward BG Wright RG Prognostic significance of MUC1 epithelial mucin expression in breast cancer Hum Pathol 1995 26 432 439 7705823 10.1016/0046-8177(95)90146-9
Nakamori S Ota DM Cleary KR Shirotani K Irimura T MUC1 mucin expression as a marker of progression and metastasis of human colorectal carcinoma Gastroenterology 1994 106 353 361 7905449
Gendler SJ Lancaster CA Taylor-Papadimitriou J Duhig T Peat N Burchell J Pemberton L Lalani EN Wilson D Molecular cloning and expression of human tumor-associated polymorphic epithelial mucin J Biol Chem 1990 265 15286 15293 1697589
Ohgami A Tsuda T Osaki T Mitsudomi T Morimoto Y Higashi T Yasumoto K MUC1 mucin mRNA expression in stage I lung adenocarcinoma and its association with early recurrence Ann Thorac Surg 1999 67 810 814 10215233 10.1016/S0003-4975(99)00041-7
Nguyen PL Niehans GA Cherwitz DL Kim YS Ho SB Membrane-bound (MUC1) and secretory (MUC2, MUC3, and MUC4) mucin gene expression in human lung cancer Tumour Biol 1996 17 176 192 8638091
Guddo F Giatromanolaki A Koukourakis MI Reina C Vignola AM Chlouverakis G Hilkens J Gatter KC Harris AL Bonsignore G MUC1 (episialin) expression in non-small cell lung cancer is independent of EGFR and c-erbB-2 expression and correlates with poor survival in node positive patients J Clin Pathol 1998 51 667 671 9930070
Ho SB Niehans GA Lyftogt C Yan PS Cherwitz DL Gum ET Dahiya R Kim YS Heterogeneity of mucin gene expression in normal and neoplastic tissues Cancer Res 1993 53 641 651 7678777
Devine PL McGuckin MA Ramm LE Ward BG Pee D Long S Serum mucin antigens CASA and MSA in tumors of the breast, ovary, lung, pancreas, bladder, colon, and prostate. A blind trial with 420 patients Cancer 1993 72 2007 2015 7689923
Hayes DF Sekine H Ohno T Abe M Keefe K Kufe DW Use of a murine monoclonal antibody for detection of circulating plasma DF3 antigen levels in breast cancer patients J Clin Invest 1985 75 1671 1678 3889057
Croce MV Isla-Larrain MT Demichelis SO Gori JR Price MR Segal-Eiras A Tissue and serum MUC1 mucin detection in breast cancer patients Breast Cancer Res Treat 2003 81 195 207 14620915 10.1023/A:1026110417294
Cannon PM Ellis IO Blamey RW Bell J Elston CW Robertson JF Expression of tumour-associated antigens in breast cancer primary tissue compared with serum levels Eur J Surg Oncol 1993 19 523 527 8270037
Ahmed N Barker G Oliva KT Hoffmann P Riley C Reeve S Smith AI Kemp BE Quinn MA Rice GE Proteomic-based identification of haptoglobin-1 precursor as a novel circulating biomarker of ovarian cancer Brit J Cancer 2004 91 129 140 15199385 10.1038/sj.bjc.6601882
Ye B Cramer DW Skates SJ Gygi SP Pratomo V Fu L Horick NK Licklider LJ Schorge JO Berkowitz RS Mok SC Haptoglobin-alpha subunit as potential serum biomarker in ovarian cancer: identification and characterization using proteomic profiling and mass spectrometry Clin Cancer Res 2003 9 2904 2911 12912935
|
16117833
|
PMC1198221
|
CC BY
|
2021-01-04 16:03:06
|
no
|
BMC Cancer. 2005 Aug 23; 5:110
|
utf-8
|
BMC Cancer
| 2,005 |
10.1186/1471-2407-5-110
|
oa_comm
|
==== Front
BMC CancerBMC Cancer1471-2407BioMed Central London 1471-2407-5-931606096410.1186/1471-2407-5-93Research ArticleA molecular analysis by gene expression profiling reveals Bik/NBK overexpression in sporadic breast tumor samples of Mexican females García Normand [email protected] Fabio [email protected] la Vega Horacio [email protected] Everardo [email protected] Isabel [email protected]ñaloza Rosenda [email protected] Diego [email protected] Molecular Genetics Laboratory, Medical Research Unit (MRU), Pediatric Hospital, National Medical Center Century XXI, Mexican Social Security Institute, Mexico City, Mexico2 Molecular Oncology Laboratory, Oncology Hospital, National Medical Center Century XXI, Mexican Social Security Institute, Mexico City, Mexico3 Department of Anatomy-Pathology, Oncology Hospital, National Medical Center Century XXI, Mexican Social Security Institute, Mexico City, Mexico4 Genetics Engineering Laboratory, Biochemical Department, Biological Sciences National School, Polytechnic National Institute. Mexico City, Mexico2005 1 8 2005 5 93 93 11 1 2005 1 8 2005 Copyright © 2005 García 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
Breast cancer is one of the most frequent causes of death in Mexican women over 35 years of age. At molecular level, changes in many genetic networks have been reported as associated with this neoplasia. To analyze these changes, we determined gene expression profiles of tumors from Mexican women with breast cancer at different stages and compared these with those of normal breast tissue samples.
Methods
32P-radiolabeled cDNA was synthesized by reverse transcription of mRNA from fresh sporadic breast tumor biopsies, as well as normal breast tissue. cDNA probes were hybridized to microarrays and expression levels registered using a phosphorimager. Expression levels of some genes were validated by real time RT-PCR and immunohistochemical assays.
Results
We identified two subgroups of tumors according to their expression profiles, probably related with cancer progression. Ten genes, unexpressed in normal tissue, were turned on in some tumors. We found consistent high expression of Bik gene in 14/15 tumors with predominant cytoplasmic distribution.
Conclusion
Recently, the product of the Bik gene has been associated with tumoral reversion in different neoplasic cell lines, and was proposed as therapy to induce apoptosis in cancers, including breast tumors. Even though a relationship among genes, for example those from a particular pathway, can be observed through microarrays, this relationship might not be sufficient to assign a definitive role to Bik in development and progression of the neoplasia. The findings herein reported deserve further investigation.
==== Body
Background
Breast cancer is one of the most frequent causes in Mexican women over 35 years of age and mortality shows a tendency to increase over time [1]. The origin and course of sporadic breast cancer are not clear. At the molecular level, some alterations have been reported as associated with this neoplasia, such as changes in DNA quantity [2], cytogenetic alterations [3], amplification of some proto-oncogenes [4,5], loss of heterozygosity in some chromosomal regions [6,7], and mutations in at least four different susceptibility genes in hereditary forms [8-11].
The traditional way of classifying breast tumors is based on tumor size, degree of dissemination and Histopathology. Alterations in many genetic networks are involved in development of breast cancer; therefore, analysis of isolated genes is not sufficient to adequately understand this neoplasia. Transcriptional analysis of multiple genes expressed by breast tumors should provide a means to define a signature or molecular fingerprint of the disease and might progressively replace conventional diagnostic and prognostic parameters. Likewise, the capability to analyze simultaneous expression levels of thousands of genes offers better possibilities to understand and characterize the complete molecular mechanisms underlying cancer progression. Technologies such as DNA chips permit integral study of the advance of the disease, with the advantage of identifying marker genes for diagnosis, prognosis, and therapy.
In several studies found correlations among expression profiles and clinical characteristics, estrogen receptors, lymphatic nodes or treatment response [12-16]. However, this results can not be generalized to other populations because development of this heterogeneous disease have influence from multiple factors, including age, diet, genetics, environment, geographic location, no pregnancy and race [17]. To generate portraits for each population may help in the sub-classification of tumours, prognosis, and general understanding of breast cancer [18,19] and will allow us to identify characteristic genes from the Mexican population.
Tumor growth rates can be <5% of those predicted by proliferation measurements alone. Several types of human cancers such as colorectal [20], ovarian [21], endometrial [22], and cervical [23], also showed an increase in apoptotic index (by TUNEL assay) during tumorigenesis, which questions if cancer research fields related with inhibition of cell death might be a critical step in cancer development. Thus, in these cases a high rate of cellular proliferation must be responsible for tumor growth.
Human gene expression patterns derived from cDNA microarrays have been increasingly used to identify genes associated with human cancers [12,14,24]. On the basis of these studies, it appears that cDNA microarray based gene expression analysis of breast cancers tissues would reveal molecular characteristics associated with tumorigenesis.
Bcl-2 family proteins contain both, anti-apoptotic and pro-apoptotic, members and are essential to maintain organ systems. For many, but not for all apoptotic signals, balance between these two Bcl-2 subfamilies determine cell fate. The pro-apoptotic Bik protein is a member of one sub-class of the Bcl-2 family designated BH3-alone [25]. The human Bik gene is located on 22q13.3 and codifies a 160 amino acid protein. Its mRNA has ubiquitous distribution with elevated levels in heart and skeletal muscle [26].
Recently, the product of Bik gene has been associated with tumoral reversion in different cell lines and was proposed as therapy to induce apoptosis in cancer including breast tumors [27,28]. Using cDNA microarrays, we obtained gene expression profiles of 15 breast cancers and 5 normal breast tissues. Complete pairwise comparison of selected genes with real-time RT-PCR revealed consistent overexpression of Bik/NBK gene in tumor samples.
Methods
Breast tissue samples
The protocol of this work was approved by the Ethical Committee of our hospital with register number 98/718/43 and all patients provided informed consent in signed letter prior to the initiation of any procedure. Samples were taken from non-affected breast tissue and affected tissue from 15 non-related Mexican patients 40 years of age [mean of 52 years and range 40–68 years]. All tumors were sporadic, infiltrating, ductal adenocarcinomas from patients who had not received adjuvant chemotherapy. Tissues were obtained from the Oncology Hospital, at the Centro Medico Nacional Siglo XXI of the Instituto Mexicano del Seguro Social in Mexico City. Each sample was snap-frozen in liquid nitrogen and stored at -70°C until use. Specimens were selected for analysis by two criteria: (i) sufficient material for analysis, and (ii) histological evaluation by two pathologists demonstrating that samples contained at least 70% of tumor cells (in the case of cancer samples). Representative sections from each case were paraffin-embedded for staining with haematoxylin and eosin to assess histopathologic diagnosis using American Cancer Committee criteria [29] and for their use in Immunohistochemistry assays.
RNA extraction and mRNA purification
Total RNA of each sample was obtained with trizol and dissolved in RNase-free water for a final concentration of 1–2 μg/μl. Samples was stored at -70°C. Concentration and purity of each sample was assessed by gel electrophoresis and spectrophotometry (Ultrospec 2000, Pharmacia Biotech). Prior to obtaining poly A+ RNA, total RNA was treated with DNAse I according to Atlas pure total RNA labeling system (Clontech, Palo Alto, CA, USA). Poly A enrichment procedure was used to obtain mRNA (Atlas pure total RNA labeling system®, Clontech, Palo Alto, CA, USA).
Probe labeling and hybridization
cDNA label procedure was used to obtain 32P-labeled cDNA from 1 μg of poly A, with MoMLV reverse transcriptase (Atlas pure total RNA labeling system®, Clontech, Palo Alto, CA, USA). Labeled probes were purified with chroma SPIN-200 columns (Atlas cDNA Expression Arrays®, Clontech, Palo Alto, CA, USA). Incorporation of 32P into the probe was double-checked in a scintillation counter (Beckman, model LS 9000sc).
The hybridization procedure was performed with hybridization solution mixed with the entire pool of labeled cDNA probes with > 1.25 × 106 cpm on Atlas array membrane with 609 genes, double-spotted (Atlas human cancer cDNA expression arrays®; Clontech, Palo Alto, CA, USA). The membrane was hybridized at 68°C for 24 h, and then was washed with high-to-low astringency solutions. Membranes were exposed for 24 h in a phosphor screen (Kodak storage phosphor screen, Molecular Dynamics).
Image analysis and data collection
The digitalized image was obtained in a storm scanner (Storm 680, Molecular Dynamics, Inc.). Expression profiles were obtained with Atlas image 2.0 software (Clontech, Palo Alto, CA, USA) through comparison of normal vs. tumor tissues. We used a normalization coefficient in which average value of all genes was used to normalize the array. This coefficient was determined with the Sum method, which adds values of signal over background for all genes on arrays.
Data analysis
Hybridization profiles were analyzed with J-Express program developed and distributed by Molmine AS and collaborators [30], and Cluster and TreeView software's [31].
Real Time-RT-PCR analysis
The same source of total RNA used to define gene expression profiles was used in real time RT-PCR experiments, following all instructions outlined in the LightCycler-RNA amplification kit SYBR Green I manual (Roche Molecular Biochemicals, Mannheim, Germany). cDNA synthesis was carried out in LightCycler (Roche) in a capillary as follows: 20 μl mix reaction containing 500 ng of DNase I treated total RNA, 4 μl of LightCycler-RT-PCR reaction mix SYBR Green I (final concentration 1X), 5 mM MgCl2, 0.4 μl LightCycler-RT-PCR enzyme mix, and 5.0 pmol forward and reverse primers for both genes. Bik gene primer sequences were designed with OLIGO 4.1 and were as follows: forward 5' GAG ACA TCT TGA TGG AGA CC 3', reverse 5' TCT AAG AAC ATC CCT GAT GT 3'. HPRT gene primers were referred by Pieretti M et al; 1991 [32]. For reverse transcription, the reaction was incubated at 55°C for 30 min and at 95°C for 30 sec.
Amplification was carried out in the same capillary. LightCycler was programmed as follows: 50 three-segment cycles for amplification (10 sec at 95°C, 30 sec at 55°C, and acquisition (single-mode), 20 sec at 72°C) and three-segment cycle of product melting (0 sec at 95°C, 10 sec at 65°C, and 0 sec at 95°C at step-acquisition mode). Temperature transition rate for all segments of amplification cycles and melting curve cycle were set at 20°C/sec except for segment 3 that was set at 0.1°C/sec of product-melting curve analysis. Duplicate reactions were prepared for each sample along with a non-template negative control (H2O control). A standard curve was used with these assays and LightCycler3 data analysis software (Roche Molecular Biochemicals) was used in all processes.
Immunohistochemistry
For identification of Bik (NBK) protein expression, streptavidin-biotin peroxidase complex (DAKO LSAB Kit, Carpinteria, CA, USA) method with diaminobenzidine as chromogen was used. Epitopes were retrieved by autoclaving in 10 mM citric acid buffer, pH 6.1 for 2 min. As primary antibody, polyclonal anti-human NBK/Bik FL-160 (Santa Cruz Biotechnology, work dilution 1:100) in 1% bovine serum albumin-phosphate buffered saline (BSA-PBS) was used. Formalin-fixed, paraffin-embedded sections of normal skin epithelium served as positive controls for Bik (NBK). Negative control slides were processed in parallel, omitting primary antibody.
Results
Expression profiles of normal and malignant breast specimens
We analyzed gene expression patterns in dissected normal or malignant human breast tissue from 20 individuals, 15 infiltrating ductal carcinomas, nine stage II, six stage III, and five normal breast samples. Hierarchical cluster analysis was used to group genes on the basis of similarity in their patterns of expression. We used average cluster linkage, uncentered [31] based on Euclidean distance and removed all genes that were turned on in tumors but that were not expressed in normal tissue (NCK5AI, CDK5 activator, CDC25A, ERK4, K2P, COL11A1, OBCAM, AMPHIREGULIN, BCGF1 and BMP8).
Cluster analysis of tumors and controls is described in Figure 1. Patterns of gene expression among tumors showed great variation (Figure 1A); however, we identified two main groups. The left branch is conformed by four histologic type II and one type III, while the right branch contains five type II and five type III. Left branch shows distinctive underexpression and loss of expression areas with genes such as IFNGR2, interleukin 2, laminin B1, MSH2, MSH6 CASP8 and 10, RB1, PCNA, EGFR and PGS2, among others, and a few overexpressed genes such as MIF, CDK 2, 5, and 7, RPSA, JUP and ITGAE. On the other hand, right branch shows overexpression areas with genes such as CASP 10, CD30-L, and CD-40-L (TNF family), ras-like small, FGF 3 and 5, TDGF1, ERK5, MIF, IFNGR2, MIG, IFN-alpha, beta, gamma, IL-3, -4, -9 and -13, IP-10, HPRT, DCC, TEK, netrin-2, cadherins 4, 8, 12 and 13 [33], MMP3, 8, 10, 12 and 13 (metalloproteinase), COL11A2, CCNG1, CDC-6, ATM, N-MYC, and p53.
In Figure 1B, tumors and controls were associated in a dendrogram obtained with expression profile of each sample. With this analysis, tumors were separated in two main branches; left group contained five tumors defined predominantly by histologic type II, while right group was composed by stage II and III tumors, at the same percentage. In the lower part of the same Figure, we showed the gene expression pattern of eight genes related with apoptotic process obtained from each tumor, showing changes across all samples but exposing the possible relation between pro-apoptotic (BAK, BAX, BIK, BAD) and anti-apoptotic genes (BCL2, BCLW, MCL1, BCL2- A1). We showed that pro-apoptotic BCL2-interacting killer gene (BIK/NBK) overexpressed in nearly all samples (14/15), which makes it a possible candidate for further studies on its role in breast cancer. MCL1 was underexpressed or missing, at least 93.3% (14/15), relative to controls.
Due to consistent high expression of Bik gene in breast cancer, we validated its overexpression by real-time PCR and studied distribution of Bik protein in tumor breast samples by Immunohistochemistry.
Real time RT-PCR assay for Bik gene
According to the Methods section, we developed real-time RT-PCR using Bik and HPRT mRNAs as templates. In Figure 2, we showed amplification products of different samples of Bik and HPRT genes. With real-time RT-PCR, results of 9/13 samples (69%) agreed with those obtained through microarray analysis (considering NB and T14) and in 31% of tumors, we were unable to confirm the result. By using melting curves (data not shown), we confirmed that amplification products obtained with real-time RT-PCR were specific. As positive control, HPRT gene (housekeeping) was used (samples numbers 11 and 14) for tumors and normal breast (NB). We used tumor 14 as negative control (tumor that wich show expression for Bik gene in microarray assay).
Characterization of Bik/NBK expression by immunohistochemistry
After RT-PCR analysis, we checked the distribution and abundance of Bik protein on slides obtained from the same tissue used for histopathologic diagnosis. Only in poorly or well-differentiated tumor areas we found presence of BIK protein with predominant cytoplasmic location (Figure 3A). Figures 3C and 3D (magnification to 400×) show Bik protein in a sample of breast cancer. Presence of BIK protein was specific because, when primary antibody was omitted, we did not observe any signal in neoplasic tissue (Figure 3B).
Discussion
Advances in high-density DNA microarray technology have permitted to screen large numbers of genes and to correlate tumoral stages and gene-expression profiles in cancer research. The idea is that tumor behavior is ruled by expression of hundreds of genes; thus microarrays analysis allows those behaviors and clinical features to be predicted. In the present study, 10 genes were turned on in some tumors respect to the normal tissue.
With microarray analysis of samples, we identified two subgroups probably related with cancer progression. In the first group, we identified an underexpression area (lower part of dendrogram in Figure 1) Wich corresponded to four histological stage II tumors, (T2, T5, T12 and T14) and one tumor in histological stage III, (T3). The latter tumor probably corresponded to a stage II tumor at molecular level because its profile expression was similar to others in the same group. We think that this group belongs to an initial stage of development of neoplasia, because it presents fewer alterations than the second group, and because the behavior of some genes related with the beginning of tumor development such as genes related with DNA repairment, apoptosis, TNF, Fas-L route, checkpoint, remodeling, and maintenance of extracellular matrix, were underexpressed or lost their expression while other genes were overexpressed, such as human macrophage migration inhibitory factor and some cyclin-dependent kinases and cytoskeletal proteins.
The right branch (Figure 1) contains stage II or III tumors and probably correspond to more advanced stages of progression of neoplasia. In this group, we found some overexpressed genes; of these, COL11A2 and netrin-2 had not been previously associated with breast cancer. Other genes, such as DCC and cadherin 13, were found overexpressed in our samples but underexpressed in other reports [34]; these genes probably have mutations that differentially affect their proteins.
Apoptosis is a mechanism to control cell death which occurs during normal development, growth, and maintenance in multicellular organisms. It can be activated by stimulation of a cell surface death receptor or release of cytochrome c from mitochondria. In the cascade of events initiated by this signaling, cysteine aspartyl proteases, or caspases are cleaved from an inactive zymogen to an active heterodimer. These active caspases degrade several components critical for cell survival, such as DNA repair elements, structural proteins, and cell signaling peptides [35]. Caspase-3, the primary death effector, serves as a key target for monitoring apoptosis, for it is activated through both cell surface death receptor and mitochondrial release of cytochrome c [36]. Disruption in signaling pathways that regulate apoptosis can lead to a variety of pathologic conditions, making apoptosis an area of intense focus for research [37].
Eight genes related with apoptosis are found in the lower part of Figure 1B. Only two genes showed differential expression in nearly all tumors, Bik gene was overexpressed in 14/15 samples and MCL-1 gene expression was diminished or absent in the same number of tumors. The behavior of the remaining genes was not constant; in tumor 5, for example, nearly all selected genes were overexpressed, whereas in tumor 6 expression of nearly all selected genes was lost. The heterogeneous behavior of these genes makes it difficult to establish their role in apoptosis or survival of neoplasic cells. However, it was possible to establish which genes, as in the case of the Bik gene, are probably related with cancer progression or are involved in other genetic networks.
BH3-only proteins are structurally distant members of the Bcl-2 protein family which triggers apoptosis. These proteins share with each other and with the remainder of the Bcl-2 family only a nine amino acid BH3 (Bcl-2 Homology) region. This domain is required for its ability to bind to Bcl-2-like, pro-survival proteins and to initiate apoptosis. Mammals have at least 10 BH3-only genes that differ in expression pattern and model of activation (regulated by a diverse range of transcription factors). Certain BH3-only proteins, including Bad, Bik/Nbk, Bid, Bim/Bod, and Bmf, are restrained by post-translational modifications that cause their sequestration from pro-survival Bcl-2 family members [38]. Nbk/Bik (natural-born killer/Bcl-2-interacting killer) is a tissue-specific BH3-only protein which molecular function is largely unknown. Nbk fails to induce apoptosis in the absence of Bax. Nbk interacts with Bcl-xL and Bcl-2 but not with Bax. It was suggested that Nbk acts as an indirect killer which triggers Bax-dependent apoptosis, whereas Bak is not sufficient to confer sensitivity to Nbk [39].
BH3-only proteins require multidomain pro-apoptotic members Bax and Bak to release cytochrome c from mitochondria and kill cells. Short peptides representing alpha-helical BH3 domains of Bid or Bim are capable of inducing oligomerization of Bak and Bax to release cytochrome c. BH3 peptides from Bad and Bik cannot directly activate Bax or Bak, but instead bind anti-apoptotic members of BCL-2, resulting in displacement of Bid-like BH3 domains which initiate mitochondrial dysfunction. These data support two types of BH3 domains: Bid-like domains which activate Bax, Bak, and Bad-like domains which sensitize by occupying the pocket of anti-apoptotic members [40].
Bik, a BH3-alone protein (Bcl-2 homology domain), is a pro-apoptotic member of the BCL2 family. Missense Bik gene mutations and sequence alterations in intronic regions were observed in cell lymphomas; these data indicate that mutation of the Bik gene is relatively frequent [41,42]. The importance of phosphorylation and desphosphorylation reactions in intracellular signaling pathways has long been accepted. The importance of serine/threonine protein phosphatases in many processes including apoptosis is recognized. The phosphorylation state of anti-apoptotic (Bcl-2 and Bcl-xL) and pro-apoptotic (Bad, Bid and Bik) Bcl-2 proteins regulates their cellular activity and, therefore, cell survival and cell death. For instance, desphosphorylation of Bad allows it to interact with Bcl-xL and initiate cell death [43]. However, mutation of the Bik phosphorylation sites, in which the Thr and Ser residues were changed to alanine residues, reduced the apoptotic activity of Bik protein without significantly affecting its ability to heterodimerize with BCL-2 [44]
Of note, a significant fraction of either ectopic or endogenous BIK was found associated with the endoplasmic reticulum, suggesting that this organelle, in addition to mitochondria, may be a target of BIK function [45]. Collectively, the results identify BIK as an initiator of cytochrome c release from mitochondria operating from a location at the ER [46]. These results indicate that any function of Bik in programmed cell death and stress-induced apoptosis must overlap that of other BH3-only proteins [47].
Because the product of Bik gene is pro-apoptotic, this protein has been used to induce apoptosis in several cancer cell lines (PC-3, HT-29, MCF-7, MDA-MB-231, 435, 468 and A540) [27,28,48]. Fay M et al; in 2003 [49] reported underexpression of the CUL-5 gene in breast tumor tissue, whereas expression levels in several cancer cell lines (MCF7, MDA-MB-231) are essentially identical to those of normal breast. Probably breast tumor tissues (i.e., Bik gene) have a different behavior vs. breast cancer cell lines at a molecular level. Moreover we found some reports with different expression levels of Bik gene among multiple human tissues: Daniel T et al; 1999 [48] reported absence of Bik expression in heart, skeletal muscle, and brain, while Verma et al; 2000 [26] reported higher expression in heart and skeletal muscle. It is difficult to reconcile these reports.
Conclusion
It is not known whether Bik gene works as an apoptosis activator or only sensitizes the cell for death. In any event, it could be a prognostic factor and a possible therapy target in breast cancer. The findings reported in this paper deserve further investigation.
List of abbreviations
ATM Ataxia telangiectasia
BAD BCL2-antagonist of cell death
BAK BCL2-antagonist/killer 1
BAX BCL2-associated × protein
BCGF1 B-cell growth factor
BCL2- A1 BCL2-related protein A1
BCL2 B-cell CLL/lymphoma 2
Bcl-xL BCL2-like 1
CUL-5 Cullin 5
BCLW BCL2-like 2
Bik/Nbk BCL2-interacting killer/ natural-born killer
BMP8 Bone morphogenetic protein 8
CASP10 Apoptotic cysteine protease MCH4
CASP8 Apoptotic cysteine protease MCH5
CCNG1 Cyclin G1
CD30-L CD30 ligand
CD-40-L CD40 ligand
CDC25A Cell division cycle 25a
CDC-6 CDC6-related protein
CDK 2 Cyclin-dependent kinase 2
CDK 5 Cell division protein kinase 5
CDK 7 Cyclin-dependent kinase 7
CDK5 activator Cyclin-dependent kinase 5 activator p35
COL11A1 Collagen type xi alpha-1
COL11A2 Collagen type xi alpha-2
cpm Counts per million
DCC Tumor suppressor protein DCC
EGFR Epidermal growth factor receptor
ERK4 Extracellular signal-regulated kinase 4
ERK5 Extracellular signal-regulated kinase 5
FGF 3 Fibroblast growth factor-3
FGF 5 Fibroblast growth factor-5
HPRT Hypoxanthine-guanine phosphoribosyltransferase
IFN-alpha Interferon alpha-c leukocyte
IFN-beta Interferon beta-1
IFN-gamma Interferon gamma
IFNGR2 Interferon gamma accessory factor-1
IL-13 Interleukin-13
IL-3 Interleukin-3
IL-4 Interleukin-4
IL-9 Interleukin-9
IP-10 IFN-gamma-inducible chemokine IP-10
ITGAE Integrin alphae
JUP Junction plakoglobin
K2P Keratin, type ii cytoskeletal 2 oral
MCL1 Myeloid cell leukemia sequence 1
MIF Macrophage migration inhibitory factor
MIF Macrophage migration inhibitory factor
MIG Gamma interferon induced monokine
MMP10 Matrix metalloproteinase 10
MMP12 Matrix metalloproteinase 12
MMP13 Matrix metalloproteinase 13
MMP3 Matrix metalloproteinase 3
MMP8 Matrix metalloproteinase 8
MSH2 DNA mismatch repair protein MSH2
MSH6 DNA mismatch repair protein MSH6
NB Normal breast
NCK5AI Cyclin-dependent kinase 5 activator isoform p39i
N-MYC N-MYC proto-oncogene
OBCAM Opioid binding cell adhesion molecule
p53 cellular tumor antigen p53
PCNA Proliferating cell nuclear antigen
PGS2 Dermatan sulfate proteoglycan core protein
ras-like small Ras-like small GTPase TTF
RB1 Retinoblastoma susceptibility
RPSA Laminin 37kd receptor
TDGF1 Teratocarcinoma-derived growth factor
TEK TYROSINE-PROTEIN KINASE RECEPTOR
TNF Tumor necrosis factor
BH3 Bcl-2 Homology
Bid BH3 interacting domain death agonist
Bim/Bod Bcl-2 like11 (apoptosis facilitator)
Bmf Bcl-2 modifying factor
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
NG performed microarrays and real-time RT-PCR assays, contributed toward the design of the study and drafted the manuscript. FS participated in the conception and design of the study. HA performed real-time RT-PCR and immunohistochemistry assays. ECQ participated in the conception and design study. IA obtained and performed histopathologic diagnosis of the samples. RP participated in the conception of the study. DA drafted the manuscript, participated in the conception and design of study and its coordination. 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 grants No. 34451-M and Salud-2003-C01-074 from Mexican Council of Science and Technology (CONACYT) and by the Medical Research Council of Mexican Social Security Institute (IMSS) FOFOI FP-0038/1248; FP-0038/1247; FP-0038/763 México. The authors would like to thank PhD José Moreno (Specialties Hospital), PhD Guadalupe Rico (Pediatrics Hospital) for technical cooperation, M.A. Maggie Brunner (Managing Editor, Archives of Medical Research) for technical English assess, National Medical Center Century-XXI, IMSS, and PhD Juan Burgueño to assessment on statistical approach. E Curiel-Quesada is a COFAA fellow.
Figures and Tables
Figure 1 Dendrogram and picture of hierarchical cluster of 15 mammary tumors and normal tissue. Expression profiles from tumor and normal tissue (average from five normal tissues) were obtained through average-linkage, hierarchical, uncentered clustering analysis, as described in Methods. Genes unexpressed in normal tissue were not included in analysis. In dendrograms, each branch corresponds to similar expression profile among genes (A) or samples (B). Green squares in picture correspond to normal expression levels. Yellow squares correspond to absence of expression. Black squares correspond to underexpressed genes. Red squares correspond to overexpressed genes. Picture under dendrogram in panel details behavior of eight apoptosis-related genes along samples.
Figure 2 Real time RT-PCR validation of expression Bik and HPRT genes. Amplification curves of Bik (Bik T1 to T14), HPRT (T11 and T14) and non-affected breast (NB) mRNAs are shown. Graphic generated by LightCycler-based real time PCR thermocycler. The abscissa shows fluorescence relative units and ordinate cycle number.
Figure 3 Distribution of Bik protein in mammary ducts from adenocarcinomas. Anti-Bik/NBK polyclonal antibody) was reacted with HRP-secondary antibody and revealed with DAB. Haematoxylin was used as counter stain. A, C, and D are different tumor areas. B, negative control from the same patient.
==== Refs
SUIVE Sistema Unico de Informacion para la Vigilancia Epidemiologica /DGE/SSA. Incidencia de casos nuevos de enfermedades por grupo de edad Estados Unidos Mexicanos Registro Histopalógico de Neoplasias Malignas 2002
Cornelisse CJ Kuipers-Dijkshoorn N van Vliet M Hermans J Devilee P Fractional allelic imbalance in human breast cancer increases with tetraploidization and chromosome loss Int J Cancer 1992 50 544 548 1537620
Pandis N Jin Y Gorunova L Petersson C Bardi G Idvall I Johansson B Ingvar C Mandahl N Mitelman F Chromosome analysis of 97 primary breast carcinomas: identification of eight karyotypic subgroups Genes Chromosomes Cancer 1995 12 173 185 7536456
Borg A Baldetorp B Ferno M Killander D Olsson H Sigurdsson H ERBB2 amplification in breast cancer with a high rate of proliferation Oncogene 1991 6 137 143 1671531
Borg A Baldetorp B Ferno M Olsson H Sigurdsson H c-myc amplification is an independent prognostic factor in postmenopausal breast cancer Int J Cancer 1992 51 687 691 1612775
Bieche I Lidereau R Genetic alterations in breast cancer Genes Chromosomes Cancer 1995 14 227 251 8605112
Phelan CM Borg A Cuny M Crichton DN Baldersson T Andersen TI Caligo MA Lidereau R Lindblom A Seitz S Kelsell D Hamann U Rio P Thorlacius S Papp J Olah E Ponder B Bignon YJ Scherneck S Barkardottir R Borresen-Dale AL Eyfjord J Theillet C Thompson AM Larsson C Consortium study on 1280 breast carcinomas: allelic loss on chromosome 17 targets subregions associated with family history and clinical parameters Cancer Res 1998 58 1004 1012 9500463
Futreal PA Liu Q Shattuck-Eidens D Cochran C Harshman K Tavtigian S Bennett LM Haugen-Strano A Swensen J Miki Y BRCA1 mutations in primary breast and ovarian carcinomas Science 1994 266 120 122 7939630
Collins N McManus R Wooster R Mangion J Seal S Lakhani SR Ormiston W Daly PA Ford D Easton DF Consistent loss of the wild type allele in breast cancers from a family linked to the BRCA2 gene on chromosome 13q12-13 Oncogene 1995 10 1673 1675 7731724
Birch JM Hartley AL Tricker KJ Prosser J Condie A Kelsey AM Harris M Jones PH Binchy A Crowther D Prevalence and diversity of constitutional mutations in the p53 gene among 21 Li-Fraumeni families Cancer Res 1994 54 1298 1304 8118819
Vorechovsky I Luo L Lindblom A Negrini M Webster AD Croce CM Hammarstrom L ATM mutations in cancer families Cancer Res 1996 56 4130 4133 8797579
Perou CM Sorlie T Eisen MB van de RM Jeffrey SS Rees CA Pollack JR Ross DT Johnsen H Akslen LA Fluge O Pergamenschikov A Williams C Zhu SX Lonning PE Borresen-Dale AL Brown PO Botstein D Molecular portraits of human breast tumours Nature 2000 406 747 752 10963602 10.1038/35021093
Bertucci F Houlgatte R Benziane A Granjeaud S Adelaide J Tagett R Loriod B Jacquemier J Viens P Jordan B Birnbaum D Nguyen C Gene expression profiling of primary breast carcinomas using arrays of candidate genes Hum Mol Genet 2000 9 2981 2991 11115842 10.1093/hmg/9.20.2981
Sorlie T Perou CM Tibshirani R Aas T Geisler S Johnsen H Hastie T Eisen MB van de RM Jeffrey SS Thorsen T Quist H Matese JC Brown PO Botstein D Eystein LP Borresen-Dale AL Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications Proc Natl Acad Sci U S A 2001 98 10869 10874 11553815 10.1073/pnas.191367098
de Bono JS Tolcher AW Rowinsky EK The future of cytotoxic therapy: selective cytotoxicity based on biology is the key Breast Cancer Res 2003 5 154 159 12793897 10.1186/bcr597
Glanzer JG Eberwine JH Expression profiling of small cellular samples in cancer: less is more Br J Cancer 2004 90 1111 1114 15026786 10.1038/sj.bjc.6601668
Desai KV Kavanaugh CJ Calvo A Green JE Chipping away at breast cancer: insights from microarray studies of human and mouse mammary cancer Endocr Relat Cancer 2002 9 207 220 12542399 10.1677/erc.0.0090207
Forus A Sorlie T Borresen-Dale AL Myklebost O [Microarray technology--potential in cancer research] Tidsskr Nor Laegeforen 2001 121 2498 2503 11875927
Liu ET Classification of cancers by expression profiling Curr Opin Genet Dev 2003 13 97 103 12573442 10.1016/S0959-437X(03)00008-X
Watanabe I Toyoda M Okuda J Tenjo T Tanaka K Yamamoto T Kawasaki H Sugiyama T Kawarada Y Tanigawa N Detection of apoptotic cells in human colorectal cancer by two different in situ methods: antibody against single-stranded DNA and terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end-labeling (TUNEL) methods Jpn J Cancer Res 1999 90 188 193 10189889
Chan WY Cheung KK Schorge JO Huang LW Welch WR Bell DA Berkowitz RS Mok SC Bcl-2 and p53 protein expression, apoptosis, and p53 mutation in human epithelial ovarian cancers Am J Pathol 2000 156 409 417 10666369
Ioffe OB Papadimitriou JC Drachenberg CB Correlation of proliferation indices, apoptosis, and related oncogene expression (bcl-2 and c-erbB-2) and p53 in proliferative, hyperplastic, and malignant endometrium Hum Pathol 1998 29 1150 1159 9781657 10.1016/S0046-8177(98)90429-4
Astudillo H Lopez T Castillo S Gariglio P Benitez L p53, Bcl-2, PCNA expression, and apoptotic rates during cervical tumorigenesis Ann N Y Acad Sci 2003 1010 771 774 15033825 10.1196/annals.1299.138
Pollack JR Sorlie T Perou CM Rees CA Jeffrey SS Lonning PE Tibshirani R Botstein D Borresen-Dale AL Brown PO Microarray analysis reveals a major direct role of DNA copy number alteration in the transcriptional program of human breast tumors Proc Natl Acad Sci U S A 2002 99 12963 12968 12297621 10.1073/pnas.162471999
Adams JM Cory S The Bcl-2 protein family: arbiters of cell survival Science 1998 281 1322 1326 9735050 10.1126/science.281.5381.1322
Verma S Budarf ML Emanuel BS Chinnadurai G Structural analysis of the human pro-apoptotic gene Bik: chromosomal localization, genomic organization and localization of promoter sequences Gene 2000 254 157 162 10974546 10.1016/S0378-1119(00)00276-6
Tong Y Yang Q Vater C Venkatesh LK Custeau D Chittenden T Chinnadurai G Gourdeau H The pro-apoptotic protein, Bik, exhibits potent antitumor activity that is dependent on its BH3 domain Mol Cancer Ther 2001 1 95 102 12467227
Zou Y Peng H Zhou B Wen Y Wang SC Tsai EM Hung MC Systemic tumor suppression by the proapoptotic gene bik Cancer Res 2002 62 8 12 11782349
Cotran R Kumar V Robins S Schoen F Pathologic basis of disease 1994 Fifth edition Saunders
Bioinformatics software solutions. J-Express 2002
Statsoft TreeView & Cluster 2000
Pieretti M Zhang FP Fu YH Warren ST Oostra BA Caskey CT Nelson DL Absence of expression of the FMR-1 gene in fragile X syndrome Cell 1991 66 817 822 1878973 10.1016/0092-8674(91)90125-I
Lee SW H-cadherin, a novel cadherin with growth inhibitory functions and diminished expression in human breast cancer Nat Med 1996 2 776 782 8673923 10.1038/nm0796-776
O'Reilly PE Raab SS Niemann TH Rodgers JR Robinson RA p53, proliferating cell nuclear antigen, and Ki-67 expression in extrauterine leiomyosarcomas Mod Pathol 1997 10 91 97 9127313
Earnshaw WC Martins LM Kaufmann SH Mammalian caspases: structure, activation, substrates, and functions during apoptosis Annu Rev Biochem 1999 68 383 424 10872455 10.1146/annurev.biochem.68.1.383
Porter AG Janicke RU Emerging roles of caspase-3 in apoptosis Cell Death Differ 1999 6 99 104 10200555 10.1038/sj.cdd.4400476
Mattson MP Duan W Pedersen WA Culmsee C Neurodegenerative disorders and ischemic brain diseases Apoptosis 2001 6 69 81 11321043 10.1023/A:1009676112184
Puthalakath H Strasser A Keeping killers on a tight leash: transcriptional and post-translational control of the pro-apoptotic activity of BH3-only proteins Cell Death Differ 2002 9 505 512 11973609 10.1038/sj.cdd.4400998
Gillissen B Essmann F Graupner V Starck L Radetzki S Dorken B Schulze-Osthoff K Daniel PT Induction of cell death by the BH3-only Bcl-2 homolog Nbk/Bik is mediated by an entirely Bax-dependent mitochondrial pathway EMBO J 2003 22 3580 3590 12853473 10.1093/emboj/cdg343
Letai A Bassik MC Walensky LD Sorcinelli MD Weiler S Korsmeyer SJ Distinct BH3 domains either sensitize or activate mitochondrial apoptosis, serving as prototype cancer therapeutics Cancer Cell 2002 2 183 192 12242151 10.1016/S1535-6108(02)00127-7
Arena V Martini M Luongo M Capelli A Larocca LM Mutations of the BIK gene in human peripheral B-cell lymphomas Genes Chromosomes Cancer 2003 38 91 96 12874789 10.1002/gcc.10245
Klumpp S Krieglstein J Serine/threonine protein phosphatases in apoptosis Curr Opin Pharmacol 2002 2 458 462 12127881 10.1016/S1471-4892(02)00176-5
Verma S Zhao LJ Chinnadurai G Phosphorylation of the pro-apoptotic protein BIK: mapping of phosphorylation sites and effect on apoptosis J Biol Chem 2001 276 4671 4676 11084041 10.1074/jbc.M008983200
Mathai JP Germain M Marcellus RC Shore GC Induction and endoplasmic reticulum location of BIK/NBK in response to apoptotic signaling by E1A and p53 Oncogene 2002 21 2534 2544 11971188 10.1038/sj.onc.1205340
Germain M Mathai JP Shore GC BH-3-only BIK functions at the endoplasmic reticulum to stimulate cytochrome c release from mitochondria J Biol Chem 2002 277 18053 18060 11884414 10.1074/jbc.M201235200
Coultas L Bouillet P Stanley EG Brodnicki TC Adams JM Strasser A Proapoptotic BH3-only Bcl-2 family member Bik/Blk/Nbk is expressed in hemopoietic and endothelial cells but is redundant for their programmed death Mol Cell Biol 2004 24 1570 1581 14749373 10.1128/MCB.24.4.1570-1581.2004
Daniel PT Pun KT Ritschel S Sturm I Holler J Dorken B Brown R Expression of the death gene Bik/Nbk promotes sensitivity to drug-induced apoptosis in corticosteroid-resistant T-cell lymphoma and prevents tumor growth in severe combined immunodeficient mice Blood 1999 94 1100 1107 10419903
Fay MJ Longo KA Karathanasis GA Shope DM Mandernach CJ Leong JR Hicks A Pherson K Husain A Analysis of CUL-5 expression in breast epithelial cells, breast cancer cell lines, normal tissues and tumor tissues Mol Cancer 2003 2 40 14641918 10.1186/1476-4598-2-40
|
16060964
|
PMC1198222
|
CC BY
|
2021-01-04 16:03:05
|
no
|
BMC Cancer. 2005 Aug 1; 5:93
|
utf-8
|
BMC Cancer
| 2,005 |
10.1186/1471-2407-5-93
|
oa_comm
|
==== Front
BMC CancerBMC Cancer1471-2407BioMed Central London 1471-2407-5-951608079610.1186/1471-2407-5-95Research ArticleFolate system correlations in DNA microarray data Radivoyevitch Tomas [email protected] Department of Epidemiology and Biostatistics Case Western Reserve University, Cleveland, Ohio 44106 USA2005 4 8 2005 5 95 95 10 12 2004 4 8 2005 Copyright © 2005 Radivoyevitch; 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
Gene expression data is abundantly available from the Gene Expression Omnibus (GEO) and various websites. Pathway specific analyses of gene-gene correlations across these datasets remain relatively unexplored, though they could be informative.
Methods
Folate gene expression data is explored here in two ways: (1) directly, using gene-gene scatter plots and gene expression time course plots; and (2) indirectly, using de novo purine synthesis (DNPS) and de novo thymidylate synthesis (DNTS) flux predictions of a folate model perturbed by relative gene expression modulations of its Vmax parameters.
Results
Positive correlations within and between the DNPS and DNTS folate cycles are observed in the folate gene expression data. For steady state measurements across childhood leukemia patients, positive correlations between DNPS and DNTS are consistent with higher proliferative fractions requiring higher levels of both fluxes. For cells exposed to ionizing radiation, transient increases in both pathways are consistent with DNA damage driven dNTP demand, and a steadily decreasing backdrop is consistent with radiation induced cell cycle arrest. By and large, folate model based flux predictions paralleled these findings, the main differences being a gain of correlation information for the TEL-AML1 leukemia data, and the loss of one interesting inference, namely, that RNA repair driven DNPS precedes DNA repair driven DNTS after a 10 gray dose of ionizing radiation.
Conclusion
Pathway focused correlation analyses of DNA microarray data can be informative, with or without a mathematical model. Conceptual models are essential. Mathematical model based analyses should supplement, but should not replace, direct data analyses.
==== Body
Background
The folate system (Figure 1A) is central to de novo purine synthesis (DNPS) and de novo thymidylate synthesis (DNTS) and is a key target of several anti-cancer agents. For example, methotrexate (MTX), in its polyglutamated forms, inhibits dihydrofolate [DHF] reductase (DHFR), thymidylate synthetase (TS), glycinamide ribonucleotide formyltransferase (GART), and other folate system enzymes (see the MTX containing reaction equations [1] in Methods); the novel multi-targeted anti-folate ALIMTA has similar targets [2] though with a very different spectrum of inhibition constants Ki such that GART inhibition is dominant [3]; the anti-folate raltitrexed (Tomudex) mostly inhibits TS [4]; and 5-fluorouracil also inhibits TS (as FdUMP), though it also kills cells via incorporation into DNA and RNA [5].
Folates convert serine side chains into tetrahydrofolate (THF; Figure 1B) held reactive single-carbons with the following uses: 5-methyl-THF (CH3THF) is used to convert homocysteine into methionine; 5,10 methylene-THF (CH2THF) is used by TS of DNTS to convert dUMP (deoxyuridylate) into dTMP (thymidylate) for the sole purpose of DNA synthesis, be it scheduled (i.e. DNA replication driven) or unscheduled (i.e. DNA damage driven); and 10-formyl-THF (CHOTHF) is used to set up the purine ring closure reactions of DNPS for many purposes, including the synthesis of RNA, DNA, ATP, GTP and many other molecules, all of which are subject to purine base oxidation and thus replacement after irradiation. These three single-carbon consuming folate functions interact via competition for CH2THF. Using publicly available DNA microarray data, this study explores folate cycle interactions at the higher level of mRNA. To assess the added value to analyses of a published mathematical model of the folate system [1], inferences obtained directly from the DNA microarray data are compared here to those obtained indirectly using folate model predictions of DNPS and DNTS based on the same DNA microarray data.
Methods
Mathematical model of Morrison and Allegra
The folate cycle model of Morrison and Allegra [1] has the following mathematical form:
These equations restate the system configuration information of Figure 1A, i.e. they state that the rate at which a metabolite concentration increases equals the sum of the synthesis reaction fluxes (arrows into a node) minus the sum of the degradation reaction fluxes (arrows leaving a node). The ri in these equations are:
where DHFT-DHF (total DHF minus free DHF) is the concentration of DHF bound to DHFR, xi is the concentration of i-glutamated MTX, and all folates are assumed to be penta-glutamated. Not shown are 10 additional differential equations for up to penta-glutamation of MTX either free or bound to DHFR. These 10 equations are irrelevant when MTX = 0 as in the microarray data analyses below; they were used here only to validate the current implementation of the model against its previously published responses, see Figure 2[1].
Model limitations
Since the model has only one compartment, the cytosol, it cannot handle changes in the mitochondrial enzymes MTHFD2 and SHMT2, nor can it handle changes in the extra-cellular folate hydrolase gene FOLH1 (the gene that codes for prostate-specific membrane antigen, PSMA), so these genes were ignored in the analyses. Further, the folate genes GGH (polyglutamate hydrolase), FPGS (polyglutamate synthase), and RFC (reduced folate carrier), were not considered in the microarray analyses because these reactions are included in the model only for MTX and not folate, and because MTX = 0 for the microarray data.
Model modifications
Flux boundary conditions for dUMP and GAR synthesis in the original model [1] were replaced by downstream concentration boundary conditions set to their initial values. This was done because steady state flux differences across patients would otherwise be nullified; e.g. if the flux into GAR were fixed, the steady state flux through GART would also be fixed, and artificially then, there would be no variation in predictions of this flux across patients.
Model-data linkage
For steady state flux predictions of leukemia patient diagnostic samples, MAS5 microarray measurements of Ross et al [6] and Yeoh et al [7] were normalized by dividing by the mean of the leukemia subtype medians. Step functions from 1 (for t < 0) to the resulting metrics (for t ≥ 0) were then used as modulators of the baseline folate model Vmax values, i.e. microarray data normalization values were equated to the steady state of the model. Individualized patient steady states were then computed as simulation endpoints 40 hours after the Vmax perturbations; all time courses were inspected visually to assure settled steady states at 40 hours. For radiation response data [8], initial measurements were equated to the steady state of the model, so the data was normalized by its values at t = 0. Linear interpolations of the normalized data were then used as time-varying Vmax modulators. For both the steady state leukemia data and the time course radiation response data, proportionality between mRNA levels and protein levels was assumed. This assumption was motivated by simplicity and a lack of better alternatives. As proteomic-transcriptomic combined dynamic response data (e.g. [9]) accrues, it will likely be replaced by a set of gene specific lead-lag filter [10] assumptions. In the meantime, it can be viewed and used as a first order approximation, or as a temporary crutch.
Computational details
The computational environment R [11] was used with R packages of Bioconductor [12] to implement this study. Specifically, the package Biobase was used to manipulate microarray data as expression set (class eset) objects and SBMLR [13,14] was used to simulate a systems biology markup language (SBML) [15-17] representation of the folate metabolism model [1]. R scripts reproducing figures 2, 3, 4, 5, 6, 7, 8 are included with SBMLR as an illustrative example of package use. For convenience, array data used in this study have been repackaged as eset objects in R data packages available from the author's website [14]. Throughout, genes with multiple probe sets were represented by the set with the highest average value.
Results
Folate system correlations across childhood leukemias
Childhood acute lymphoblastic leukemia (ALL) microarray data of Ross et al [6] and Yeoh et al [7] is shown in Figures 3 and 4. Several points can be made regarding this steady state diagnostic bone marrow data. Firstly, since TYMS and DHFR (similarly MTHFD1, GART and ATIC) operate in series, it makes sense that the system would attempt to match these throughput capabilities as closely as possible to avoid the costs of maintaining unneeded excess "equipment." Thus, positive correlations within the DNPS and DNTS branches are expected. Secondly, growing cells require commensurate increases in both DNTSand DNPS, so positive correlations between these cycles are also expected. Finally, DNPS genes are higher in T cell leukemic cells than in B-cell leukemic cells, consistent with measured DNPS fluxes being three fold higher in T cells than in B cells [18].
To assess the added value of the folate model, since MTHFD1 and TYMS are the gatekeepers of DNPS and DNTS, respectively, correlation plots for these genes were compared to corresponding flux predictions in Figure 5. The plots show that, for the most part, model predicted fluxes are more correlated than measured MTHFD1 vs. TYMS mRNA. To estimate the amount of correlation attributable to steady state flux constraints alone, 1000 uncorrelated normally distributed (μ = 1; σ = 0.30) random numbers were applied to the model as Vmax modulators. The amount of correlation (r = 0.18, Figure 6) is more than that induced by the model (beyond gatekeeper correlations) for BCR-ABL and T cell leukemias (Figure 5), but less than induced for TEL-AML1 leukemias. Thus, for TEL-AML leukemias additional information must have been contributed by the other folate genes inputted into the model, suggesting more folate system coordination in this more curable leukemia subtype.
Folate system analysis of radiation time course data
Folate system correlations in radiation response time course data [8] were also investigated. The data (Fig. 7) shows that TS, DHFR, GARFT and MTHFD each have a dose-dependent transient increase after irradiation, consistent with radiation induced DNA damage causing a transient rise and fall in P53 activity [19] with subsequent induction of ribonucleotide reductase subunit P53R2 [20] causing a transient increase in de novo deoxynucleotide synthesis for DNA repair; a 17-fold increase in R2 protein 24 h after irradiation [21] is radioprotective [22], further supporting this conjecture. The data also shows a steady decline in many of the gene expression time courses, possibly due to radiation induced cell cycle arrest. At 10 gray, inspection of TS, DHFR, GARFT and MTHFD further suggests that RNA repair driven DNPS (peak at 2 hours) precedes DNA repair driven DNTS (peak at 6 hours).
In Figure 7, gene expression time courses are more likely to be signals if they differ between 3 and 10 gray only in terms of minor time shifts and dose ordered amplitude changes. Based on this, MTHFR and SHMT1 were dismissed as noise. The remaining six gene expression time courses were normalized by their values at t = 0 and applied to the folate model as linearly interpolated time-varying modulators of corresponding Vmax parameters. The resulting model-based predicted time courses are shown in Figure 8. These plots affirm the "dose-dependent spike resting on a decreasing backdrop" response of DNPS and DNTS that was qualitatively inferred from Figure 7. In contrast to the data itself, however, at 10 gray, RNA repair driven DNPS (peak at 12 hours) and DNA repair driven DNTS (peak at 10 hours) are reverse ordered and delayed. Since NTP production tends to be in high gear full time, compared to dNTP production, it likely has a greater ability to respond rapidly to oxidative damage. Further, a dNTP synthesis flux peak at 6 hours compared to 10 hours is more consistent with a P53 spike at 5 hours [19]. Thus, until measurements of DNPS and DNTS suggest otherwise, the data-based inference that RNA repair precedes DNA repair is tentatively more credible than its model-based counterpart.
Discussion
Pathway focused analyses are essential for gene-gene correlation studies because the number of possible correlations would otherwise be too large to investigate. For example, if a chip carries 10,000 genes, the number of 2D plots requiring correlation tests is then 10,000 choose 2, or ~50 million, i.e. a multiple testing problem not encountered in pathway focused studies.
Although TS and DHFR are predominantly controlled at the level of protein translation [23], Figures 3, 4 and 7 indicate that some control effort is also exerted at the mRNA level. Thus, protein level control may dominate a particular regulatory system, but mRNA signals may still be informative of what the overall system is trying to accomplish. That DNPS and DNTS are correlated in a consistent manner across disparate steady state-and transient datasets lends credence to the view that DNA microarray data is a valid source of biochemical system coordination and control information. Characterization of cancer differences in coordination and control could be relevant to future treatment designs and should thus be further explored.
Conclusion
The main conclusion of this paper is that interesting inferences can be gleaned from genome-wide microarray data (with or without mathematical models) if gene-gene correlations are analyzed in a pathway specific manner. The added value of analyzing microarray data using Morrison and Allegra's folate model, relative to simply eye-balling the gene expression data, was minimal. For example, in Figure 5, save the model's ability to identify TEL-AML1 leukemias as being additionally coordinated, gate-keeper focused gene expression scatter plots are almost as revealing as model-predicted DNPS vs. DNTS scatter plots. Similarly, for the radiation time course data in Figures 7 and 8, the spike increase in DNPS and DNTS and the baseline downward trend at larger times are apparent using either approach.
This study used conceptual models of folates and the biological effects of ionizing radiation to guide, focus, validate and discriminate microarray data inferences. Further, knowledge of system scope was used to reduce the analysis to a manageable dimension correlated subspace. This suggests that pathway focused analyses are more likely to be successful if they are applied to biochemical systems and experimental perturbations that are well understood.
To go beyond qualitative statements and to actually plot predictions (Figures 5 and 8), quantitative models are needed. Further, as biochemical system knowledge expands in scope and complexity, gedanken experiments underlying eye-ball data analyses will become increasingly difficult to carry out. Thus, model-based approaches must continue to be developed. At the same time, since the inference that RNA repair precedes DNA repair was lost in the model-based approach, such approaches should supplement, but should not replace, direct data analyses.
Abbreviations
DHF = dihydrofolate; THF = tetrahydrofolate; DNPS = de novo purine synthesis; DNTS = de novo thymidylate synthesis; DHFR = DHF reductase; TS = thymidylate synthetase (protein); TYMS = thymidylate synthetase (gene); MTHFR = methylene-THF reductase; MTR = methyl-THF transferase; FTS = formyl-THF synthetase; FDS = formyl-DHF synthetase; SHMT = serine hydroxy methyl transferase; GAR = glycinamide ribonucleotide; FGAR = formyl-GAR; GART = GAR formyltransferase; AICAR = aminoimidazole-4-carboxamide ribonucleotide; FAICAR = formyl-AICAR; dUMP = deoxyuridylate; dTMP = thymidylate; MTX = methotrexate; SBML = Systems Biology Markup Language.
Competing interests
The author declares that he has no competing interests.
Authors' contributions
The author is the sole contributor.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
This work was supported by the Comprehensive Cancer Center of Case Western Reserve University and University Hospitals of Cleveland (P30 CA43703), the American Cancer Society (IRG-91-022-09), the National Cancer Institute's Integrative Cancer Biology Program (P20 CA112963-01) and NIH grant 1K25 CA104791-01A1.
Figures and Tables
Figure 1 The folate cycle model of Morrison and Allegra [1] (A) and the molecular structure of tetrahydrofolate (B).
Figure 2 Morrison and Allegra's model [1] responding to 1 μM MTX applied continuously after t = 0. Concentrations are in μM (top 6 plots) and fluxes are in μM/hour (bottom 3 plots).
Figure 3 MAS5 U133a folate gene expression data of Ross et al [7]. Symbols are TEL-AML1 (B), BCR-ABL (b) and T-cell (T). In mirrored positions relative to the main diagonal, corresponding Pearson correlation coefficients r are given with their P values.
Figure 4 MAS5 U95av2 folate gene expression data of Yeoh et al [8]. Only the Yeoh et al [8] patients who are also in the Ross et al dataset [7] are considered. MAS5 summary measures were computed from cel files using Bioconductor's AFFY package. Symbols are as in Figure 3.
Figure 5 Comparisons of measured MTHFD1 vs. TYMS versus predicted DNPS vs. DNTS.
Figure 6 DNPS vs. DNTS predicted with random numbers replacing Figures 3 and 4.
Figure 7 Lymphocyte radiation response data of Jen and Cheung [8]. SHMT1 and MTHFR were not applied to the folate model, see text. Time is in hours.
Figure 8 The folate model's response to the radiation time course data. Fluxes are in μM/hr.
==== Refs
Morrison PF Allegra CJ Folate cycle kinetics in human breast cancer cells J Biol Chem 1989 264 10552 10566 2732237
Curtin NJ Hughes AN Pemetrexed disodium, a novel antifolate with multiple targets Lancet Oncol 2001 2 298 306 11905785 10.1016/S1470-2045(00)00325-9
Shih C Habeck LL Mendelsohn LG Chen VJ Schultz RM Multiple folate enzyme inhibition: mechanism of a novel pyrrolopyrimidine-based antifolate LY231514 (MTA) Adv Enzyme Regul 1998 38 135 152 9762351 10.1016/S0065-2571(97)00017-4
Yin MB Guimaraes MA Zhang ZG Arredondo MA Rustum YM Time dependence of DNA lesions and growth inhibition by ICI D1694, a new quinazoline antifolate thymidylate synthase inhibitor Cancer Res 1992 52 5900 5905 1394217
Spiegelman S Sawyer R Nayak R Ritzi E Stolfi R Martin D Improving the anti-tumor activity of 5-fluorouracil by increasing its incorporation into RNA via metabolic modulation Proc Natl Acad Sci USA 1980 77 4966 4970 6933541
Ross ME Zhou X Song G Shurtleff SA Girtman K Williams WK Liu HC Mahfouz R Raimondi SC Lenny N Patel A Downing JR Classification of pediatric acute lymphoblastic leukemia by gene expression profiling Blood 2003 102 2951 2959 12730115 10.1182/blood-2003-01-0338
Yeoh EJ Ross ME Shurtleff SA Williams WK Patel D Mahfouz R Behm FG Raimondi SC Relling MV Patel A Cheng C Campana D Wilkins D Zhou X Li J Liu H Pui CH Evans WE Naeve C Wong L Downing JR Classification, subtype discovery, and prediction of outcome in pediatric acute lymphoblastic leukemia by gene expression profiling Cancer Cell 2002 1 133 143 12086872 10.1016/S1535-6108(02)00032-6
Jen KY Cheung VG Transcriptional response of lymphoblastoid cells to ionizing radiation Genome Res 2003 13 2092 2100 12915489 10.1101/gr.1240103
Zheng PZ Wang KK Zhang QY Huang QH Du YZ Zhang QH Xiao DK Shen SH Imbeaud S Eveno E Zhao CJ Chen YL Fan HY Waxman S Auffray C Jin G Chen SJ Chen Z Zhang J Systems analysis of transcriptome and proteome in retinoic acid/arsenic trioxide-induced cell differentiation/apoptosis of promyelocytic leukemia Proc Natl Acad Sci U S A 2005
Oppenheim AV Willsky AS Young IT Signals and Systems 1983 Prentice Hall, Englewood Cliffs, NJ 763 763
Ihaka R Gentleman R R:a language for data analysis and graphics Journal of Computational and graphical statistics 1996 5 299 314
Gentleman RC Carey VJ Bates DM Bolstad B Dettling M Dudoit S Ellis B Gautier L Ge Y Gentry J Hornik K Hothorn T Huber W Iacus S Irizarry R Leisch F Li C Maechler M Rossini AJ Sawitzki G Smith C Smyth G Tierney L Yang JY Zhang J Bioconductor: open software development for computational biology and bioinformatics Genome Biol 2004 5 R80 15461798 10.1186/gb-2004-5-10-r80
Radivoyevitch T SBMLR
Radivoyevitch T Radivoyevitch Lab
Systems Biology Markup Language
Hucka M Finney A Sauro HM Bolouri H Doyle JC Kitano H Arkin AP Bornstein BJ Bray D Cornish-Bowden A Cuellar AA Dronov S Gilles ED Ginkel M Gor V Goryanin II Hedley WJ Hodgman TC Hofmeyr JH Hunter PJ Juty NS Kasberger JL Kremling A Kummer U Le Novere N Loew LM Lucio D Mendes P Minch E Mjolsness ED Nakayama Y Nelson MR Nielsen PF Sakurada T Schaff JC Shapiro BE Shimizu TS Spence HD Stelling J Takahashi K Tomita M Wagner J Wang J The systems biology markup language (SBML): a medium for representation and exchange of biochemical network models Bioinformatics 2003 19 524 531 12611808 10.1093/bioinformatics/btg015
Finney A Hucka M Systems biology markup language: Level 2 and beyond Biochem Soc Trans 2003 31 1472 1473 14641091
Dervieux T Brenner TL Hon YY Zhou Y Hancock ML Sandlund JT Rivera GK Ribeiro RC Boyett JM Pui CH Relling MV Evans WE De novo purine synthesis inhibition and antileukemic effects of mercaptopurine alone or in combination with methotrexate in vivo Blood 2002 100 1240 1247 12149204 10.1182/blood-2002-02-0495
Lahav G Rosenfeld N Sigal A Geva-Zatorsky N Levine AJ Elowitz MB Alon U Dynamics of the p53-Mdm2 feedback loop in individual cells Nat Genet 2004 36 147 150 14730303 10.1038/ng1293
Tanaka H Arakawa H Yamaguchi T Shiraishi K Fukuda S Matsui K Takei Y Nakamura Y A ribonucleotide reductase gene involved in a p53-dependent cell-cycle checkpoint for DNA damage Nature 2000 404 42 49 10716435 10.1038/35003506
Kuo ML Kinsella TJ Expression of ribonucleotide reductase after ionizing radiation in human cervical carcinoma cells Cancer Res 1998 58 2245 2252 9605773
Kuo ML Hwang HS Sosnay PR Kunugi KA Kinsella TJ Overexpression of the R2 subunit of ribonucleotide reductase in human nasopharyngeal cancer cells reduces radiosensitivity Cancer J 2003 9 277 285 12967138
Tai N Schmitz JC Liu J Lin X Bailly M Chen TM Chu E Translational autoregulation of thymidylate synthase and dihydrofolate reductase Front Biosci 2004 9 2521 2526 15353304
|
16080796
|
PMC1198223
|
CC BY
|
2021-01-04 16:03:06
|
no
|
BMC Cancer. 2005 Aug 4; 5:95
|
utf-8
|
BMC Cancer
| 2,005 |
10.1186/1471-2407-5-95
|
oa_comm
|
==== Front
BMC CancerBMC Cancer1471-2407BioMed Central London 1471-2407-5-961608079910.1186/1471-2407-5-96Research ArticleEffects of alpha fetoprotein on escape of Bel 7402 cells from attack of lymphocytes Li Mengsen [email protected] Xinhua [email protected] Sheng [email protected] Pingfeng [email protected] Gang [email protected] Department of Biochemistry and Molecular Biology, Peking University Health Science Center, Beijing 100083, China2 Department of Biochemistry, Hainan Medical College, Haikou 570102, China2005 5 8 2005 5 96 96 8 4 2005 5 8 2005 Copyright © 2005 Li 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
Involvement of AFP against apoptosis of tumor cell has been implicated in its evasion of immune surveillance. However, the molecular events of immune escape mechanisms are still unknown. The major observations reported here relate to a possible mechanism by which heptoloma Bel 7402 cells escape immune surveillance in vitro.
Methods
Western blotting and a well-characterized cofocal scanning image were performed to analyze the expression of Fas/FasL and caspase-3 in co-cultured Bel 7402 and Jurkat cells.
Results
After co-culture with Jurkat cells, up-regulated Fas and reduced FasL expression could be observed. Treatment with AFP could remarkably inhibit the elevated Fas and, whereas, induce the FasL expression in co-cultured Bel 7402 cells. Cells co-culture could induce the expression of caspase-3 in both cells line. The elevated caspase-3 in Bel 7402 cells was abolished following the treatment of AFP. The expression of caspase-3 was elevated in co-cultured Jurkat cells treated with AFP. No detectable change on the expression of survivin was examined in both cells line. Monoclonal antibody against AFP treatment alone did not obviously influence the growth of cells, as well as the expression of Fas/FasL and caspase-3. However, the effect of AFP could be blocked by antibody.
Conclusions
our results provide evidence that AFP could promote the escape of liver cancer cells from immune surveillance through blocking the caspase signal pathway of tumor cells and triggering the Fas/FasL interaction between tumor cells and lymphocytes.
==== Body
Background
Alpha fetoprotein (AFP) is one of several oncofetal proteins synthesized in large amounts by the fetus and drops in serum markedly shortly after birth. AFP as a tumor-associated fetal protein has demonstrated clinical utility as a tumor marker. Besides its role as a carrier or transporter for various serum ligands including fatty acids, retinoids, steroids, drugs, dyes and heavy metals, AFP has been reported to display growth regulatory properties. Previous studies have verified that AFP appears to functions as a growth regulator rather than only serum carrier. Multitude of cell types involving the growth and differentiation effects of AFP include placental [1], lymphoid [2], ovarian [3], uterine [4], gastric cancer [5], epidermal [6], breast cancer [7] and fetal fibroblasts [8]. Recently, some studies on the mechanisms of AFP suggested that AFP induced apoptosis in tumor cells independently of Fas/Fas ligand or TNFR/TNF signaling pathway, and AFP-mediated cell death involved activation of the effector caspase-3-like proteases, but was independent of upstream activation of the initiator caspase-1, caspase-8, and caspase-9-like proteases [9]. The intracellular mechanism of AFP involving to cAMP-PKA signaling pathway after its binding to different affinity receptors has been also reported [10].
Although the biological roles of the oncoembryonal protein AFP, including immunoregulatory functions in a variety of immune responses including the humoral and cell-mediated types, have been reviewed in detail, the evidences for the role of AFP in hepatoma cells escaping from host immune surveillance are still unknown [11,12]. In a recent study, AFP was used as an effective tumor rejection antigen to observe its effect in T-cell immune responses, implicating a gene therapy-based strategy for hepatoma cells [13]. However, the over-expression of AFP in human hepatoma cells is concurrent with aberrant growth manifestation. We presume that the altered serum AFP level is the cause of such changes rather than a coincident phenomenon and should be responsible for the malignant progression of liver cancer. Thus revealing the intracellular mechanisms underlying the evasion of tumor from host immune surveillance will provide further insights into the understanding for the biological role of AFP, particularly in the case of hepatocellular carcinomas.
Methods
Determination of cells proliferation
Jurkat T cells and Bel 7402 cells, the human hepatoma cell line, were adjusted to 3.0 × 104 per ml separately and maintained in PRMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum. The cells were seeded into 24-well plates and incubated at 37°C in a humidified atmosphere of 5% CO2. The supernatant was replaced to RPMI-1640 medium free serum for 24 h, then the various concentrations (5, 10, 20, 40, 80 or 100 mg/L) of AFP (Biodesign International Co. USA), human serum albumin (HSA, from Sigma, USA) and anti-AFP antibody (Santa Cruz. USA) were administrated into Jurkat T cells and Bel 7402 cells for 60 h respectively. The viability of cells was determined by Trypan blue exclusion assay.
Cell co-culture assay
To observe the effect of AFP on the escape of tumor cell from the attack of lymphocytes, 1.5 × 104 of Jurkat calls and equal Bel 7402 cells that grew under such conditions were mixed and co-cultured onto 24-well plate. Following the incubation in RPMI-1640 medium for 24 h, AFP (20 mg/L), HSA (20 gm/L), AFP (20 mg/L) plus anti-AFP antibody (40 mg/L) and anti-AFP antibody (40 mg/L) were added into culture for 60 h. The fraction containing Jurkat cells were removed from flask by resuspending the supernatant gently and transferring the supernatant to a centrifuge tube. Bel 7402 cells in the bottom were scraped and collected. The viability of each cell line was determined by Trypan blue exclusion.
Determination of Fas and FasL expression
Bel 7402 cells and Jurkat T cells were co-cultured as described above in Petri dish. AFP (20 mg/L), HSA (20 mg/L), AFP (20 mg/L) plus anti-AFP antibody (40 mg/L) and anti-AFP antibody (40 mg/L) were added into culture. At 48 hours treatment, Bel 7402 cells were washed 3 times with RPMI-1640 free serum to remove Jurkat cells. Cells were incubated with 0.5 ml of rabbit anti-Fas or anti-FasL solution (Santa Cruz Co, USA. 1:250 in RPMI-1640 medium) for 1 h. After washing with medium 3 times at room temperature, secondary goat anti-rabbit IgG antibodies conjugated with FITC (Santa Cruz, USA) were applied, and incubated for 1 hour at 37°C. After washing with PBS, the cells were observed under Confocal Laser Scanning Microscope (Leica TCS-NT SP2, Germany).
Western blot immunodetection
1.5 × 104 Jurkat T cells and 1.5 × 104 Bel 7402 cells per ml were co-cultured in 75-cm2 flasks and maintained in RPMI-1640 medium free serum for 24 h. To replace the supernatant with medium supplemented with 10% FCS, the supernatant which containing the Jurkat cells were removed, centrifuged and resuspended with fresh medium. The suspension was replaced into flasks to remix with Bel 7402 cells. Afterward, the co-cultured cells were treated with AFP (20 mg/L), HSA (20 gm/L), AFP (20 mg/L) plus anti-AFP antibody (40 mg/L) and anti-AFP antibody (40 mg/L) for 36 h respectively. Proteins were quantified before being loaded onto the gel. 40 μg of extracted proteins from each group was loaded onto 10% SDS polyacrylamide gels. Proteins were blotted onto a nitrocellulose membrane (Amersham, UK). Membranes were incubated for 1 h with anti-caspase-3 or anti-survivin monoclonal antibodies (Santa Cruz Co. USA) and then washed and revealed using anti-rabbit IgG horseradish peroxidase conjugate. Immunoreaction protein was detected by the chemiluminescence luminol reagent (Santa Cruz, USA).
Statistical analyses
All experiments were performed at least three times. Data were presented as mean ± SD. The significance of the difference between experimental and control groups was analyzed with Student's t test. A value of p < 0.05 was considered to be statistically significant.
Results
The effects of AFP on the proliferation of cells
AFP could enhance the proliferation of Bel 7402 cells in the dose-dependent manner in the concentration range of 10 to 80 mg/L (Fig 1A). The increment was up to 46.8% (80 mg/L) compared with control. However, the inhibited effect of AFP was observed in the growth of Jurkat cells. The maximum suppressive dose might be observed at the concentration of 80 mg/L (30.6% inhibition). As the controls, HSA and anti-AFP antibody did not display any detectable influences on the cell growth (Fig 1B and 1C).
The effects of AFP on the growth of co-cultured cells
To observe the effects of AFP on the escape of Bel 7402 cells from the attack of lymphocytes, AFP at a concentration of 20 mg/L, which has been reported and proved to be an optimal dose according to the dose response curves in our recent experiments [unpublished data], was added into co-cultured cells. Jurkat and Bel 7402 cells were separated as described in Materials and Methods after the treatment of 60 h and the viability of each cell was determined. The results showed that the growths of both cells in untreated group were inhibited in the cells co-culture (Fig 2). However, the administration of AFP could obviously inhibit the proliferation of Jurkat cells, but not Bel 7402 cells after cells co-culture. Anti-AFP antibody could block the effect of AFP on cell proliferation. Antibody alone and HSA did not display any significant influences on the cell growth.
Fas/FasL expression in Bel 7402
AFP was added into either separate-cultured or co-cultured cells to observe its effects on the expression of Fas and FasL in Bel 7402 under confocal fluorescent microscope. As shown in Fig 5, untreated cells cultured separately did not exhibit green fluorescence on the surface of Bel 7402 (Fig 3A). After the administration of AFP for 48 h, only a diffuse image could be observed (Fig 3B). Bel 7402 cells co-cultured with Jurkat cells for 90 min exhibited an ever-increasing green fluorescence throughout the time course that reflected the presence of expressed Fas, which was faded after the addition of AFP (Fig 3E and 3C). Anti-AFP antibody alone did not change the fluorescent intensity of image and however, could partly reverse the effect of AFP on the expression of Fas (Fig 3D and 3F). In the observation of FasL, separate-cultured Bel 7402 cells did not display the expression of FasL (Fig 4A). The slight green outline of cells was observed after the addition of AFP (Fig 4B). The cell co-culture could induce the expression of FasL in Bel 7402 cells (Fig 4C). However, Bel 7402, in contrast with that observed in the assay of Fas, co-cultured with Jurkat cells in the presence of AFP could show an enhanced green fluorescent. A representative image is shown in Fig 4D.
Immunodetection of apoptosis-related proteins
When different cells were separately cultured, AFP could enhance the intracellular expression of caspase-3 in Jurkat cells and suppress that in Bel 7402 cells (Fig 5). In the co-culture experiment, the contents of caspase-3 in both cell lines were increased even though no AFP treatment. As shown by the intensities of the immuno-positive bands, the synthesis of intracellular caspase-3 in Jurkat cells was potently up-regulated after the administration of AFP. In contrast, no detectable change in the content of caspase-3 could be observed in the co-cultured Bel 7402 cells, indicating the block of AFP. The effect of AFP on the expression of caspase-3 in both cells could be abolished by anti-AFP antibody. As shown in Fig 6, AFP did not display significant influence on the level of survivin although a slight increment in co-cultured Jurkat cells.
Discussion
During the last decade, it has been confirmed from a multitude of studies that AFP as a growth regulator modulates the ontogenic growth and tumor progression even though the overall findings remain controversial and their interpretations are still being debated. Previous studies also implicated that AFP expressed during pregnancy inhibits the maternal immune exclusion by suppressing immune responses of the humoral and cell-mediated types [14]. The higher level of AFP in the serum of liver cancer subject has been considered to be the reason of tumor development rather than merely the concomitant oncofetal protein. Hereby, the implication was emerged that AFP functions to constitute one of fundamental steps in the progression of hepatoma. It is in fact that one of mechanisms of progression and development of liver cancer is due to its escaping from immune surveillance. Experimental results obtained from the present study showed that AFP was capable of suppressing the growth of Jurkat cells with the concomitant proliferation of Bel 7402, indicating the diversity effects of AFP on the regulation of immune and tumor cells growth. This result was supported by a recent study, which indicated that AFP induced significant apoptosis of dendritic cells [15]. AFP-treated dendritic cells could produce low levels of IL-12 and TNF-α, a cytokine pattern that could hamper an efficient antitumor immune response. These results thereby offered a mechanism by which hepatocellular carcinoma escapes immunological control.
Co-culture experiment has been widely used for the determination of apoptosis, pathological response and tumor related protein synthesis [16-18]. The effects of Fas/FasL in the mechanism of tumor escaping from immune surveillance have been extensively documented [19-22]. It has been proposed that the expression of Fas/FasL in tumors may play a critical role in immune escape. In the present study, the expression of Fas/Fas L in target hepatoma Bel 7402 cells was examined after co-culture with the effector Jurkat cells in vitro. Our data showed that the expression of membrane Fas was enhanced in Bel 7402 cells co-cultured with Jurkat cells. The fact that the enhanced expression of Fas was abolished by AFP exposure indicated that the stimulation of Jurkat cells to Bel 7402 cells was able to be blocked by AFP. This is in accordance with our recent findings from immunodetection analysis (unpublished data). On the other hand, AFP enhanced the synthesis of FasL in Bel 7402 cells, which was consistent with previous findings [23-27]. AFP-induced over-expression of Fas on the surface of lymphocytes, together with simultaneous over-secreted FasL from tumor cells, could be one of reasons to accelerate the death of lymphocytes and facilitate the immune escape of liver cancer. This conclusion is supported by a recent study, which indicated that Fas-mediated apoptosis resulted by cancers expressing FasL and killing lymphocytes was irrespective of transforming growth factor-beta1 (TGF-beta1) expression [28].
Although the precise mechanism of AFP-mediated cell growth regulation and apoptosis induction remains obscure, there have been considerable investigations indicating that AFP can modulate apoptotic signals induced by various factors by either promoting or abrogating apoptosis. In the present study, the intracellular level of caspase-3 in Bel 7402 cells after co-culture with Jurkat cells was up-regulated, indicating the inducement of the apoptotic protein by lymphocytes. Furthermore, the pretreatment of AFP in vitro in co-cultured cells led to the full abolishment of caspase-3. Whereas, an increment at the level of caspase-3 protein was observed in co-cultured Jurkat cells after stimulated by AFP, which resulted in the depletion of lymphocytes in culture. The accelerated death of lymphocytes in current study might serve to certify the speculation that AFP-mediated apoptosis involved activation of the effector caspase-3-like proteases [9]. Therefore, it is reasonable to postulate that the escaping of tumor cells from immune surveillance is partly attributable to the apoptosis of lymphocytes induced by AFP. Taking into consideration the ability of AFP to alter the Fas/FasL expression in Bel 7402 and Jurkat cells, the signaling pathway involved might be initiated through the interaction of AFP with corresponding receptors and induced the activation of Fas/FasL and caspase-3 system. Our data elucidated that the death signal was triggered by activation of specific membrane AFP receptors involved in apoptosis signaling, leading to the expressional alteration of Fas/FasL and caspase-3. These findings are in accordance with the previous data showing that Fas/FasL was involved in AFP-induced immune escape [29-31].
Conclusions
In summary, the present study has at least partly explained demonstrated the potential effect of AFP in malignant growth and transformation of liver tumor cell. Take together all findings, we propose that the effect of AFP in the escape of hepatoma Bel 7402 cells from immune surveillance is achieved though decreasing Fas in its membrane and secreting FasL which in turn to trigger apoptosis of lymphocytes via caspase-3 cascade. Thus, the involvement of Fas- and caspase-related signal pathway in the occurrence of liver tumor was thereby indicated in this study. The precise elucidation of the biological effect of AFP will help to better understand the regulatory mechanism of immune surveillance in liver cancer.
Abbreviations
AFP: alpha fetoprotein; HSA: human serum albumin
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
Mengsen Li and Xinhua Liu: carried out the study and contributed equally to this study.
Sheng Zhou: participated in the statistical analysis.
Pingfeng Li: conceived of the study, and participated in its design and coordination and helped to draft the manuscript.
Gang Li: conceived and design of the study, corresponding author.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
This study is partly supported by the National Natural Science Foundation of China (No.30271174 and 30260117).
Figures and Tables
Figure 1 The effects of different concentrations of AFP on the growth of cells. The viability of Jurkat and Bel 7402 cells were observed after treated with different concentrations (5, 10, 20, 40, 80 or 100 mg/L) of either AFP (A), anti-AFP antibody (B) or HSA (C) for 60 h at 37°C in a humidified atmosphere of 5% CO2. *P < 0.05 and **P < 0.01 vs corresponding to normal control (0 mg/L) analyzed by t-test. Data were representative of an experiment that was repeated three times presented as mean ± SD of 12 samples.
Figure 2 The effects of AFP on the co-cultured cells. Bel 7402 and Jurkat cells were co-cultured for 24 h following treated with AFP (20 mg/L) (C), AFP (20 mg/L) plus anti-AFP antibody (40 mg/L) (D), anti-AFP antibody (40 mg/L) (E) and HSA (20 gm/L) (F) respectively. Graph bar A represents the viability of separate-cultured cells. Graph bar B represents co-cultured and no-treatment group. After 60 h, two cell lines were separated and the viability of each cell was observed respectively. *P < 0.05 and **P < 0.01 vs corresponding to normal control (0 mg/L) analyzed by t-test. Data were representative of an experiment that was repeated three times presented as mean ± SD of 12 samples.
Figure 3 Confocal assay for the expression of Fas on the membrane of co-cultured Bel 7402 cells. Bel 7402 cells and Jurkat T cells were co-cultured with AFP (20 mg/L), anti-AFP antibody (40 mg/L), and AFP (20 mg/L) plus anti-AFP antibody (40 mg/L) respectively for 48 h. After removing Jurkat cells, Bel 7402 cells were incubated with rabbit anti-Fas antibody and secondary goat anti-rabbit IgG antibodies conjugated with FITC. The cells were observed using a confocal laser scanning microscope. Image A: separate-cultured Bel 7402 cell; B: separate-cultured Bel 7402 cells treated with AFP; C: co-cultured Bel 7402 cells treated with AFP; D: co-cultured Bel 7402 cells treated with anti-AFP antibody. Right images from A to D were taken under common microscope to observe the state of cells. E: co-cultured Bel 7402 cells; F: co-cultured Bel 7402 cells treated with AFP and anti-AFP antibody. All images were representative of an experiment that was repeated three times.
Figure 4 Confocal assay for the expression of FasL on the membrane of co-cultured Bel 7402 cells. Bel 7402 cells and Jurkat T cells were co-cultured with AFP (20 mg/L) for 48 h. After removing Jurkat cells, Bel 7402 cells were incubated with rabbit anti-FasL antibody and secondary goat anti-rabbit IgG antibodies conjugated with FITC. The cells were observed using a confocal laser scanning microscope. Image A: separate-cultured Bel 7402 cell; B: separate-cultured Bel 7402 cells treated with AFP. Right images of A and B were taken under common microscope to observe the state of cells. C: co-cultured Bel 7402 cells without AFP treatment; D: co-cultured Bel 7402 cells treated with AFP. All images were representative of an experiment that was repeated three times.
Figure 5 The effects of AFP on the expression of caspase-3 protein. Jurkat T cells and Bel 7402 cells were co-cultured for 24 h and treated with AFP (20 mg/L), HSA (20 gm/L), AFP (20 mg/L) plus anti-AFP antibody (40 mg/L) and anti-AFP antibody (40 mg/L) for 36 h respectively. The expression of caspase-3 was detected with Western blotting. Lane 1: control; Lane 2: cells separate-cultured cells treated with AFP; Lane 3: co-cultured cells; Lane 4: co-cultured cells treated with AFP; Lane 5: co-cultured cells treated with AFP and anti-AFP antibody; Lane 6: co-cultured cells treated with anti-AFP antibody. All images were representative of an experiment that was repeated three times.
Figure 6 The effects of AFP on the expression of survivin protein. Jurkat T cells and Bel 7402 cells were co-cultured for 24 h and treated with AFP (20 mg/L), HSA (20 gm/L), AFP (20 mg/L) plus anti-AFP antibody (40 mg/L) and anti-AFP antibody (40 mg/L) for 36 h respectively. The expression of survivin was detected with Western blotting. Lane 1: control; Lane 2: cells separate-cultured cells treated with AFP; Lane 3: co-cultured cells; Lane 4: co-cultured cells treated with AFP; Lane 5: co-cultured cells treated with AFP and anti-AFP antibody; Lane 6: co-cultured cells treated with anti-AFP antibody. All images were representative of an experiment that was repeated three times.
==== Refs
Toder V Bland M Gold-Gefter L Nebel J The effect of alpha-fetoprotein on the growth of placental cells in vitro Placenta 1983 4 79 86 6190161
Hamel S Hoskin DW Hooper DC Murgita RA Mizejewski GJ, Jacobson HI Phenotype and function of bone marrow-derived T and non-T cells activated in vitro by alpha-fetoprotein Biological activities of alpha-fetoprotein 1987 CRC Press: Florida 167 177
Keel BA Eddy KB Cho S May JV Synergistic action of purified α-fetoprotein and growth factors on the proliferation of porcine granulose cells in monoplayer culture Endocrinology 1991 129 217 225 1711460
Mizejewski GJ Keenan JF Setty RP Separation of the estrogen-activated growth regulatory forms of alpha-fetoprotein in mouse amniotic fluid Biol Reprod 1990 42 887 898 1696509
Koide N Nishio A Igarashi J Kajikawa S Adachi W Amano J α-fetoprotein-producing gastric cancer: histochemical analysis of cell proliferation, apoptosis, and angiogenesis American J Gastroenterol 1999 94 1658 1663
Leal JA May JV Keel BA Human alpha-fetoprotein enhances epidermal growth factor proliferation activity upon porcine granulose cells in monolayer culture Endocrinology 1980 126 669 671
Bennett JA Semeniuk DJ Jacobson HI Murgita RA Similarity between natural and recombinant human alpha-fetoprotein as inhibitors of estrogen-dependent breast cancer growth Breast Cancer Res Treat 1997 45 169 179 9342442 10.1023/A:1005841032371
Dudich E Semenkova L Gorbatova E Dudich I Khromykh L Tatulov E Grechko G Sukhikh G Growth-regulative activity of human alpha-fetoprotein for different types of tumor and normal cells Tumour Biol 1998 19 30 40 9422080 10.1159/000029972
Dudich E Semenkova L Dudich I Gorbatova E Tochtamisheva N Tatulov E Nikolaeva M Sukhikh G alpha-fetoprotein causes apoptosis in tumor cells via a pathway independent of CD95, TNFR1 and TNFR2 through activation of caspase-3-like proteases Eur J Biochem 1999 266 750 61 10583368 10.1046/j.1432-1327.1999.00868.x
Li MS Li PF He SP Du GG Li G The intracellular mechanism of alpha-fetoprotein promoting the proliferation of NIH 3T3 cells Cell Res 2002 12 151 156 12118941
Mizejewski GJ Alpha-fetoprotein structure and function: relevance to isoforms, epitopes, and conformational variants Exp Biol Med (Maywood) 2001 226 377 408 11393167
Mizejewski GJ MacColl R α-fetoprotein growth inhibitory peptides: potential leads for cancer therapeutics Mole cancer Ther 2003 2 1243 1255
Vollmer CM JrEilber FC Butterfield LH Ribas A Dissette VB Koh A Montejo LD Lee MC Andrews KJ McBride WH Glaspy JA Economou JS Alpha-fetoprotein-specific genetic immunotherapy for hepatocellular carcinoma Cancer Res 1999 59 3064 7 10397245
Deutsch HF Chemistry and biology of alpha-fetoprotein Adv Cancer Res 1991 56 253 312 1709334
Um SH Mulhall C Alisa A Ives AR Karani J Williams R Bertoletti A Behboudi S α-Fetoprotein impairs APC function and induces their apoptosis J Immunol 2004 173 1772 1778 15265907
Lee TB Min YD Lim SC Kim KJ Jeon HJ Choi SM Choi CH Fas (Apo-1/CD95) and Fas ligand interaction between gastric cancer cells and immune cells J Gastroenterol Hepatol 2002 17 32 8 11895550 10.1046/j.1440-1746.2002.02657.x
Lim IJ Phan TT Bay BH Qi R Huynh H Tan WTL Lee ST Longaker MT Fibroblasts co-cultured with keloid keratinocytes: normal fibroblasts secrete collagen in a keloidlike manner Am J Physiol Cell Physiol 2002 283 C212 C222 12055090
Thornton MV Kudo D Rayman P Horton C Molto L Cathcart MK Ng C Paszkiewicz-Kozik E Bukowski R Derweesh I Tannenbaum CS Finke JH Degradation of NF-kappa B in T cells by gangliosides expressed on renal cell carcinomas The Journal of Immunology 2004 172 3480 3490 15004148
Uzzo RG Rayman P Kolenko V Clark PE Bloom T Ward AM Molto L Tannenbaum C Worford LJ Bukowski R Tubbs R Hsi ED Bander NH Novick AC Finke JH Mechanisms of Apoptosis in T Cells from Patients with Renal Cell Carcinoma Clinical cancer research 1999 5 1219 1229 10353760
Chen YL Chen SH Wang JY Yang BC Fas ligand on tumor cells mediates inactivation of neutrophils J Immunol 2003 171 1183 91 12874204
Elnemr A Ohta T Yachie A Kayahara M Kitagawa H Ninomiya I Fushida S Fujimura T Nishimura G Shimizu K Miwa K Human pancreatic cancer cells express non-functional Fas receptors and counterattack lymphocytes by expressing Fas ligand; a potential mechanism for immune escape Int-J-Oncol 2001 18 33 9 11115536
Perabo FG Kamp S Schmidt D Lindner H Steiner G Mattes RH Wirger A Pegelow K Albers P Kohn EC von-Ruecker A Mueller SC Bladder cancer cells acquire competent mechanisms to escape Fas-mediated apoptosis and immune surveillance in the course of malignant transformation Br J Cancer 2001 84 1330 8 11355943 10.1054/bjoc.2001.1808
Ito Y Monden M Takeda T Eguchi H Umeshita K Nagano H Nakamori S Dono K Sakon M Nakamura M Tsujimoto M Nakahara M Nakao K Yokosaki Y Matsuura N The status of Fas and Fas ligand expression can predict recurrence of hepatocellular carcinoma Br J Cancer 2000 82 1211 1217 10735508 10.1054/bjoc.1999.1065
Patel T Immune escape in hepatocellular cancer: is a good offense the best defense? Hepatology 1999 30 576 578 10421672 10.1002/hep.510300232
Lee SH Shin MS Lee HS Bae JH Lee HK Kim HS Kim SY Jang JJ Joo M Kang YK Park WS Park JY Oh RR Han SY Lee JH Kim SH Lee JY Yoo NJ Expression of Fas and Fas-related molecules in human hepatocellular carcinoma Hum Pathol 2001 32 250 256 11274632 10.1053/hupa.2001.22769
Fukuzawa K Takahashi K Furuta K Tagaya T Ishikawa T Wada K Omoto Y Koji T Kakumu S Expression of fas/fas ligand (fasL) and its involvement in infiltrating lymphocytes in hepatocellular carcinoma (HCC) J Gastroenterol 2001 36 681 688 11686478 10.1007/s005350170031
Nagao M Nakajima Y Hisanaga M Kayagaki N Kanehiro H Aomatsu Y Ko S Yagita H Yamada T Okumura K Nakano H The alteration of Fas receptor and ligand system in hepatocellular carcinomas: How do hepatoma cells escape from the host immune surveillance in vivo? Hepatology 1999 30 413 421 10421649 10.1002/hep.510300237
Houston A Bennett MW O'Sullivan GC Shanahan F O'Connell J Fas ligand mediates immune privilege and not inflammation in human colon cancer, irrespective of TGF-beta expression Br J Cancer 2003 89 1345 51 14520470 10.1038/sj.bjc.6601240
Semenkova L Dudich E Dudich I Tokhtamisheva N Tatulov E Okruzhnov Y Garcia-Foncillas J Palop-Cubillo JA Korpela T Alpha-fetoprotein positively regulates cytochrome c-mediated caspase activation and apoptosome complex formation Eur J Biochem 2003 270 4388 99 14622304 10.1046/j.1432-1033.2003.03836.x
Wu J Chen Y Li T Expression of Fas, p53 and AFP in development of human fetal germ cells in vitro Zygote 2002 10 333 40 12463529 10.1017/S0967199402004070
Li MS Li PF He SP Du GG Li G The promoting molecular mechanism of alpha-fetoprotein on the growth of human hepatoma Bel 7402 cell line World J Gastroenterol 2002 8 469 475 12046072
|
16080799
|
PMC1198224
|
CC BY
|
2021-01-04 16:03:05
|
no
|
BMC Cancer. 2005 Aug 5; 5:96
|
utf-8
|
BMC Cancer
| 2,005 |
10.1186/1471-2407-5-96
|
oa_comm
|
==== Front
BMC CancerBMC Cancer1471-2407BioMed Central London 1471-2407-5-971608349510.1186/1471-2407-5-97Research ArticleNovel curcumin- and emodin-related compounds identified by in silico 2D/3D conformer screening induce apoptosis in tumor cells Füllbeck Melanie [email protected] Xiaohua [email protected] Renate [email protected] Cornelius [email protected] Wolfgang [email protected] Robert [email protected] Institute of Biochemistry, Charité, Universitätsmedizin Berlin, Monbijoustr. 2, 10117 Berlin, Germany2 Division of Molecular Biology, Department of Surgery, Charité, Universitätsmedizin Berlin, Monbijoustr. 2, 10117 Berlin, Germany2005 5 8 2005 5 97 97 14 4 2005 5 8 2005 Copyright © 2005 Fuellbeck 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
Inhibition of the COP9 signalosome (CSN) associated kinases CK2 and PKD by curcumin causes stabilization of the tumor suppressor p53. It has been shown that curcumin induces tumor cell death and apoptosis. Curcumin and emodin block the CSN-directed c-Jun signaling pathway, which results in diminished c-Jun steady state levels in HeLa cells. The aim of this work was to search for new CSN kinase inhibitors analogue to curcumin and emodin by means of an in silico screening method.
Methods
Here we present a novel method to identify efficient inhibitors of CSN-associated kinases. Using curcumin and emodin as lead structures an in silico screening with our in-house database containing more than 106 structures was carried out. Thirty-five compounds were identified and further evaluated by the Lipinski's rule-of-five. Two groups of compounds can be clearly discriminated according to their structures: the curcumin-group and the emodin-group. The compounds were evaluated in in vitro kinase assays and in cell culture experiments.
Results
The data revealed 3 compounds of the curcumin-group (e.g. piceatannol) and 4 of the emodin-group (e.g. anthrachinone) as potent inhibitors of CSN-associated kinases. Identified agents increased p53 levels and induced apoptosis in tumor cells as determined by annexin V-FITC binding, DNA fragmentation and caspase activity assays.
Conclusion
Our data demonstrate that the new in silico screening method is highly efficient for identifying potential anti-tumor drugs.
==== Body
Background
The COP9 signalosome (CSN), a conserved multimeric protein complex, functions at the interface between signal transduction and ubiquitin (Ub)-dependent proteolysis [1]. Because of associated enzymes, the CSN possesses kinase acitivity. Two of the associated kinases are the protein kinase CK2 (CK2) and the protein kinase D (PKD) [2]. More than 200 proteins are known to be phosphorylated by the CK2, which is located nearly everywhere in the cell. The PKD is a serine/threonine kinase localized at either the plasma membrane or the cytosol of lymphocytes [3] and is associated with very diverse cellular functions, including Golgi organization, plasma membrane directed transport, metastasis, immune response, apoptosis and cell proliferation [4]. It is assumed that the CSN is a platform that brings together the kinases and appropriate substrates [5]. Transcriptional regulators such as p53 and c-Jun are phosphorylated by the CSN kinases [6,7]. The phosphorylation of p53 at Thr155 results in Ub-dependent degradation of the tumor suppressor [6]. In contrast, the CSN-directed phosphorylation of c-Jun leads to the stabilization of the transcription factor towards the Ub/26S proteasome system [8]. Cellular functions such as regulation of transcription, DNA repair, cell cycle regulation, senescence and apoptosis are modulated by p53 as well as c-Jun. Defects most frequently observed during tumorigenesis are mutations in the p53 gene [9]. It is well known that wild type p53 provides a critical brake in tumor development [10]. In contrast, as a component of the activator protein-1 the onco-protein c-Jun is mostly a positive regulator of cell proliferation and involved in oncogenic transformation (for review see [11]). Hence, the intracellular concentrations of p53 and c-Jun are decisive for tumor development. Therefore, in tumor therapy it is of great interest to control the stability of p53 and c-Jun in tumor cells. One strategy might be the inhibition of CSN-associated kinases, CK2 and PKD. It has been demonstrated before that blocking CSN-mediated phosphorylation causes an increase of p53 [6] and a decrease of c-Jun [12], very useful effects for anti-tumor drugs. Curcumin has been identified as an inhibitor of CSN-associated kinases [13], which is already in phase I clinical trials for evaluations concerning the prevention of colon, breast, lung and prostate cancer [14]. Former investigations showed that curcumin is a potent inhibitor of angiogenesis [15] and of the recombinant kinases CK2, PKD and the purified CSN complex from erythrocytes [2,13]. In addition, a natural product called emodin is also known as an inhibitor of the CK2 (PDB-Code: 1F0Q), PKD and the CSN complex [2].
In this study we developed an in silico screening to identify novel, more effective inhibitors of CSN-associated kinases by using our in-house database (more than 106 compounds). Curcumin and emodin served as lead structures in the screenings. Using a 3D superposition algorithm [16] the lead structures were compared with every compound of the database. For better coverage of the compounds and to assure their flexibility during usage of the algorithm a total of ~50 conformers were computed for every compound of the database. Compounds identified from the in silico screening were evaluated in kinase assays and cell culture experiments. With the new screening strategy potential new drugs for tumor therapy were identified, which stabilized endogenous p53 and induced apoptosis in tumor cells.
Methods
In silico screening
Three dimensional (3D) similarity search
Lead structures (curcumin and emodin) and compounds in the database were prepared for the 3D search, which is based on structural similarities. As a first step the centers of mass of each compound were determined and superimposed. The plane and the straight line of minimal quadratic distance to all atoms were computed to determine the least and largest (orthogonal) expansion. One structure was rotated such that the major directions coincide. In a further step the normalization of the atomic set was used to identify pairs of corresponding atoms. The root mean square distance (rmsd) was calculated for the related atomic pairs. An improvement of this value was obtained by removing or adding atoms to this superposition [16].
Two dimensional (2D) similarity search
Another possibility to find new inhibitors of CSN-associated kinases was the 2D search, which is based on chemical similarities. The presence or absence of common functional groups such as alcohols or ring systems such as pyrimidins was investigated. Each substructure element was assigned to one bit of a Boolean array. To calculate 2D similarities between two structures the Tanimoto-coefficient was estimated. Bits set in both structures (BitsAB) and bits, which were only set in one structure (BitsA and BitsB) were taken into consideration. The value varies between 0 (different molecules) and 1 (equal molecules): [17].
In further steps each compound was analyzed for its possible application as a drug. First we investigated the absorption and permeability using the Lipinski's rule-of-five, which implies that molecules should contain less than 10 H-bond-acceptors and less than 5 H-bond-donors. The calculated logP-value (describes the lipophilic properties) should be less than 5 and the molecular weight should be less than 500 g/mol [18]. Any compound violating more than one rule is not a promising candidate for a drug.
Based on the similar property principle one can predict toxic effects from the molecular structure. We used two quantitative structure toxicity relationship (QSTR) models to analyze the compounds for their toxicity. Using the software TOPKAT® DS MedChem Explorer (Discovery Studio, Accelrys Inc., [19]) from Accelrys Inc., which comprises the QSTR models, the toxicological data were obtained [20].
Kinase assay
Kinase activity was determined with [γ-32P]ATP (ICN) in a buffer containing 30 mM Tris, pH 7.6, 10 mM KCl, 0.5 mM DTT. Full-length c-Jun was used as substrate. Recombinant CK2 (2α2β) and recombinant PKD were obtained from Calbiochem and Upstate, respectively. The CSN was isolated from human erythrocytes as outlined before [7]. The protein kinase inhibitors emodin and resveratrol were purchased from Calbiochem, curcumin and piceatannol were from Sigma. BTB00363 (2-Pyridinecarboxylic acid, [(3,5-dichloro-2-hydroxyphenyl)methylene]hydrazide), SEW04213 (6-fluoro-3,4-dihydro-2H-pyrano [2,3-b]quinolin-5-amine), BTB14431 (9,10-dihydroxy-1,4-dihydroanthracene-1,4-dione), JFD02836 (3-methoxy-10-methyl-9,10-dihydro-9-acridinone) and JFD03665 (10-(hydroxymethylene)phenanthren-9(10H)-one) were obtained from Maybridge. Approximately 1 μg of full-length c-Jun was incubated with inhibitors at different concentrations (0, 20, 50, 100 and 200 μM) and 10 μCi [γ-32P]ATP for 60 min at 37°C. Reactions were stopped by adding 3 μl of 4 × SDS-sample buffer (Roth). The samples were separated by SDS-PAGE in a 12.5% gel, stained with Coomassie, dried and exposed to X-ray film. Subsequently the data were evaluated by densitometry. IC50 values were calculated assuming Michaelis-Menten kinetics.
Cell culture
HeLa cells were cultured in RPMI 1640 medium containing 10% (v/v) fetal calf serum, 2 mM glutamine (Life Technologies, Inc.), penicillin (100 U/ml) and streptomycin (100 μg/ml) in a humidified 5% CO2 atmosphere. For inhibitor treatment, 5–7 × 106 HeLa cells were incubated with curcumin, resveratrol, piceatannol, BTB00363, emodin, SEW04213, BTB14431, JFD02836 or JFD03665 in 0.25% dimethyl sulfoxide (DMSO) for 4 h at indicated concentrations. Control cells were treated with 0.25% DMSO. After inhibitor treatment, cells were harvested and lysed using ice-cold mono-detergent lysis buffer (50 mM Tris pH 8.0, 1% Triton-X-100, 1 mM EDTA, 150 mM NaCl, 1 μg/ml aprotinin, 1 μg/μl PMSF). HeLa cell lysate proteins were separated by SDS-PAGE in a 12.5% gel, blotted to nitrocellulose and probed with anti-c-Jun antibody (Calbiochem). For testing the intracellular concentration of p53 by Western blot analysis, B8 cells (murine fibroblasts) were cultured using Iscoves's MEM medium (Biochrom) with 125 μg/ml G418 (Gibco BRL). Cells (5 × 106 B8 cells/well) were incubated with resveratrol, piceatannol, emodin, BTB14431 or JFD02836 for 4 h at indicated concentrations. Control cells were treated with 0.25% DMSO. After inhibitor treatment, cell lysates were produced and analyzed as described for HeLa cells. Proteins were probed with an anti-p53 antibody (IC Chemikalien). All Western blots were developed using the ECL technique (Amersham).
Caspase-3/7 activity assay
To determine apoptosis, B8 cells (104 cells/well) were incubated with indicated inhibitors for 4 hours. After incubation, the caspase-3/7 reagent, containing a DEVD-peptide (Promega) was added as recommended by the manufacturer (Promega). The fluorescence of each well was measured at an excitation wavelength of 485 nm and an emission wavelength of 530 nm.
Cell viability assay
Cell viability was assessed by the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay (Sigma-Aldrich), which is based on the ability of a mitochondrial dehydrogenase from viable cells to oxidize the tetrazolium rings of the pale yellow MTT and to form a dark blue formazan crystals, which are largely impermeable to cell membranes and, therefore, accumulate within healthy cells. The number of surviving cells is directly proportional to the concentration of the formazan produced.
Cells were incubated with the indicated inhibitors at different concentrations (20 or 50 μM) for 24 h. Then a solution of MTT in phosphate-buffered saline (PBS) was added to each well to a final concentration of 0.5 μg/μl. After 4 h incubation dark blue formazan was solubilized with 100 μl DMSO. Absorbance was measured at 590 nm using an ELISA reader.
Annexin-V-FITC/propidium iodide (PI) double staining
Annexin V binds to phosphatidyl serine externalized to the outer leaflet of the plasma membrane bilayer during initial stages of apoptosis. To measure cell staining by annexin V the substance was labeled with FITC (fluorescein isothiocyanate). Simultaneously the cells were stained with PI. By the double staining the test discriminates between intact (FITC-/PI-), apoptotic (FITC+/PI-) and necrotic cells (FITC+/PI+) [21]. First, HeLa cells were incubated with the indicated inhibitors at different concentrations (20 or 50 μM) for 24 h. After harvesting and washing with PBS cells were resuspended in 100 μl annexin V binding buffer (containing 10 mM HEPES/NaOH, pH 7.4, 140 mM NaCl, 2.5 mM CaCl2). Subsequently cells were incubated with 5 μl of FITC-conjugated annexin V (ApoTarget) for 20 min at room temperature. After annexin-V-FITC staining, 400 μl of annexin V binding buffer containing PI (2.5 μg/ml) were added and the cells were analyzed by flow cytometry within 1 hour after staining.
DNA-fragmentation
Cells were seeded at a density of 105 cells/ml and treated with different concentrations (20 and 50 μM) of the indicated emodin- and curcumin-like compounds. After 24 h the cells were collected, washed with PBS at 4°C and fixed in PBS/2% (vol/vol) formaldehyde on ice for 30 min. After fixation cells were incubated with ethanol/PBS (2:1, vol/vol) for 15 min, pelleted and resuspended in PBS containing 40 μg/ml RNase A. After incubation for 30 min at 37°C cells were pelleted again and finally resuspended in PBS containing 50 μg/ml PI. The nuclear DNA fragmentation was then quantified by flow cytometric determination of hypodiploid DNA. Data were analyzed using the CELLQuest software. Data are given in percentage of hypoploidy (subG1), which reflects the number of apoptotic cells.
Results
In silico screening plays an important role in drug-design [21]. The method used here was developed to identify potential new inhibitors of the CSN-associated kinases, CK2 and PKD. Based on the structures of the known inhibitors curcumin and emodin [2] a search in our in-house database containing approximately 106 compounds was performed. A 3D superposition algorithm was developed to compare structures of the known inhibitors (lead structures) and the database compounds. Using the 3D superposition algorithm the identification of structures, which do not exactly fit into the pattern of the lead structure (scaffold hoppers, 2D similarity <85%), can be realized. To carry out specific searches, many features such as the size of a molecule (Å), limit of rmsd, number of assigned atoms and number of similar atoms were compared. With this approach 35 compounds similar to the lead structures curcumin and emodin were identified and further analyzed.
As a first step we tested the behavior of the compounds concerning the Lipinski's Rule-of-five using Accord for Excel from Accelrys Inc. Our investigation showed that no compound broke more than one rule (data not shown). Further the toxicological effects of the compounds were tested. Two QSTR models were employed: Rat Oral LD50 (Lethal Dose) and NTP (National Toxicology Program) Rodent Carcinogenicity. All identified compounds were predicted to be harmless.
Based on the different lead structures and the calculated Tanimoto-coefficients the identified compounds were divided into two groups (see Table 1). The compounds of the first group were found by searching the database with curcumin and the compounds of the second group are related to emodin. Fig. 1A shows the superposition of the structures curcumin (blue) and piceatannol (green). In addition, the chemical properties of the two compounds are compared (Fig. 1B). Both structures contain two aromatic rings and a number of H-bond acceptors, which seem to be important for the inhibitory effect.
All 35 compounds selected from the database by the screenings described above were tested in kinase assays for their ability to inhibit CSN-associated kinases. In these assays recombinant CK2 or PKD as well as purified CSN were incubated in the presence of [γ-32P]ATP and 50 or 200 μM of the potential inhibitors (data not shown). The data showed that only 7 out of 35 identified compounds inhibited the CSN-associated kinases significantly in the chosen concentration range. Therefore 7 reagents, 3 compounds of the curcumin-group and 4 of the emodin-group, were used for further analysis. Next IC50 values of these 7 compounds were determined with recombinant CK2, PKD or with the purified CSN. Kinase assays were performed in the presence of different inhibitor concentrations. After incubation assays were analyzed by SDS-PAGE and autoradiography. Kinase activity (%) was estimated by densitometry. Fig. 2 demonstrates the results for piceatannol (curcumin-group) and BTB14431 (emodin-group). Values obtained by densitometry were plotted against inhibitor concentrations. Obtained curves were used to calculate IC50 values, which are summarized in Table 1 for all tested compounds and compared with the values for curcumin and emodin determined before [2]. Data show that curcumin-derived compounds seem to have a higher affinity for PKD whereas inhibitors from the emodin-group were more efficient in suppressing the CK2. In most cases IC50 values obtained with the purified CSN were lower than those with recombinant kinases (Table 1).
Because cell treatment with curcumin or emodin led to Ub- and proteasome-dependent degradation of c-Jun [2,12], we tested whether HeLa cells incubated with different concentrations of the new inhibitors possess decreased c-Jun levels. It can be seen in the Western blot analysis (Fig. 3A) that resveratrol as well as piceatannol (curcumin-group) and BTB14431 as well as JFD02836 (emodin-group) induced a significant reduction of endogenous c-Jun in a dose-dependent manner.
It has been shown before that curcumin stabilizes endogenous p53 toward the Ub system in HeLa and MCF7 cells [6]. These cell lines possess wild type p53 [22-26]. We were interested to see whether our new inhibitors of CSN-associated kinases also increase intracellular steady state p53 concentration. Mouse B8 fibroblasts, also expressing wild type p53 [27] were treated with inhibitors. Subsequently cell lysates were tested by Western blot analysis using an anti-p53 antibody. Data are shown in Fig. 3B. All tested compounds induced significant stabilization of endogenous p53 in B8 cells in a dose-dependent manner. Interestingly, resveratrol at 200 μM was most effective in stabilizing the tumor suppressor (Fig. 3B).
It has been shown that emodin, curcumin and resveratrol induce apoptosis through p53-dependent pathways [28,29]. Therefore we asked the question whether piceatannol, BTB14431, SEW04213 and JFD02836 also induce apoptosis. Several studies on compounds similar to curcumin have been published. The ability to induce apoptosis was measured in approximately 60 tumor cell lines and effects were obtained in nearly all cells examined [30]. Most studies demonstrate the execution of apoptosis by the oxidation of tetrazolium [30], the measurement of DNA-fragmentation as well as caspase activity and annexin-V-FITC/propidium iodide (PI) double staining [31]; methods also applied in our investigations.
First the caspase activity was measured in B8 fibroblasts. Data shown in Fig. 4 demonstrate that all inhibitors tested caused an increase of caspase activity. The most pronounced increase was obtained with 50 μM of curcumin. Piceatannol (50 μM) and BTB14431 (50 μM) elevated caspase activity by approximately 5-times, whereas significant effects with SEW04213 and JFD02836 were only obtained at compound concentrations of 200 μM.
To corroborate these findings we investigated the DNA-fragmentation after treating HeLa cells with curcumin, emodin, resveratrol, piceatannol, BTB14431 and JFD02836 for 24 h. Except for JFD02836 in all experiments apoptosis was triggered (Fig. 5B) and the subG1 peak, which reflects the number of apoptotic cells increased in a dose-dependent manner (Fig. 5A).
In addition, by staining cells with annexin-V-FITC/PI cell death was quantified [32]. After 24 h of incubation a large number of apoptotic cells appeared in the experiments with the inhibitors curcumin, piceatannol as well as BTB14431 (Fig. 6B). Emodin and resveratrol produced less apoptotic cells. In experiments with high concentrations (50 μM) of curcumin and piceatannol (Fig. 6A) increased amounts of necrotic cells were observed.
The viability of the cells was checked by the MTT test, which detects the percentage of irreversibly damaged cells after treatment with the indicated inhibitors. By adding curcumin and BTB14431 at a concentration of 20 μM the percentage of viable cells decreased to ~25%. A significant reduction of viable cells with emodin, resveratrol or piceatannol was only detected at high inhibitor concentrations (50 μM) (Fig. 7).
Discussion and conclusion
Virtual screening versus high-throughput screening
Recently we have shown that curcumin and emodin, inhibitors of CSN-associated kinases, induce Ub- and proteasome-dependent degradation of c-Jun in tumor cells [2,12]. Moreover, curcumin treatment causes stabilization of the tumor suppressor p53 towards the Ub system [6]. It has been demonstrated that curcumin- and emodin-induced increase of p53 results in massive tumor cell death due to p53-dependent apoptosis [28,29]. Therefore, at least in tumor cells with wild type p53 elevation of cellular p53 levels could be of high therapeutic effect. Both events, the reduction of c-Jun and the increase of p53 are important for tumor therapy and can be accomplished by inhibition of CSN-associated kinases.
Therefore a new method was developed to identify compounds which are effective blockers of CSN-associated kinases and can be potentially used in tumor therapy. In our in silico screening we referred to curcumin and emodin as lead structures and compared them with approximately 106 compounds of our in-house database regarding their structural properties (similar property principal). In contrast to high-throughput screening (HTS) our method allows to exclude a large number of compounds before experimental testing. HTS is often used and appropriate assay systems can evaluate more than 125,000 compounds a day. However, HTS is not without problems [33,34]. HTS experiments become more and more expensive and the handling of the large amount of data is very time consuming. In addition, it is difficult to exclude false-positive hits [33]. By our virtual screening method we identified 35 compounds that seemed to be promising candidates for CSN kinase inhibition. Out of the 35 structures found by in silico screening 7 compounds had an inhibitory effect on recombinant CK2 and PKD and on the kinase activity of the purified CSN complex. Thus, the hit rate of our virtual screening was 20%. In contrast, the hit rate of HTS is usually approximately 2% [35]. We have used additional methods such as the Lipinski rule-of five and the toxicological investigations for ranking the found substances. These methods, however, did not serve to exclude compounds. Summing up, the data demonstrate that our in silico screening is a reliable and efficient method to find new active molecules.
The curcumin- and emodin-derived inhibitors of CSN-associated kinases
The present study includes only compounds which were identified using curcumin or emodin as lead structures. Our screening revealed 3 compounds of the curcumin-group and 4 compounds of the emodin-group, which showed inhibition of the kinases in vitro and in cell experiments. Interestingly, some of the identified compounds are more effective inhibitors than the lead structure. Therefore, in silico screening is a sensitive method for the identification of molecules with specific biological function. The two groups can be clearly divided by structural features. The different structures are most likely responsible for their different effects. Whereas members of the curcumin-group possess much better IC50 values with the PKD, the members of the emodin-group are better inhibitors of CK2. However, all compounds inhibit both CK2 and PKD relatively unspecifically. It has been shown before that emodin is a competitive CK2 inhibitor [36]. As demonstrated by crystal structure it binds into the ATP-binding pocket of the kinase [37]. Simulation on the basis of the emodin data revealed that curcumin also fits into the same ATP-binding pocket of the CK2 (data not shown). In addition, the effects of curcumin are reversible [12] as it would be expected of a competitive inhibitor. Therefore, we conclude that all identified compounds compete with ATP for the ATP-binding site of the kinases. Since the ATP-binding sites of CK2 and PKD are slightly different, members of the curcumin-group have another preference as compared to the emodin-group. On the other hand, ATP-binding sites are highly conserved among kinases. Therefore the identified kinase inhibitors are rather unspecific.
Inhibitors of CSN-associated kinases are potential drugs for tumor therapy
Interestingly, many identified inhibitors of CSN-associated kinases are compounds of natural products such as curcumin, resveratrol, piceatannol, emodin and honokiol, which have been shown to inhibit angiogenesis and development of malignancies [12,15,38,39]. Here we demonstrate possible anti-tumor mechanisms of known and new substances. The inhibitors of CSN-associated kinases exhibit two important effects. As shown here they reduce c-Jun levels in tumor cells. In addition, it has been demonstrated that inhibitors of CSN-associated kinases cause an increase of intracellular p53, which can be explained by the stabilization of the tumor suppressor towards the Ub/26S proteasome system [6]. However, because our data were obtained by estimating steady state levels of p53, altered expression might also contribute to the increased protein concentration in the cells. Nonetheless, elevated steady state levels of p53 are accompanied with massive cell death caused by apoptosis as demonstrated here with B8 fibroblasts and HeLa cells. In addition, the viability of the cells decreased after treatment with the selected substances.
Based on our data we cannot exclude that the tested compounds exert their pro-apoptotic effects independently of the CSN and p53. For example, in addition to CSN-associated kinases curcumin inhibits NF-κB activation associated with the induction of apoptosis [40]. Moreover, although the exact role of c-Jun in apoptosis is not known, low c-Jun steady state levels as measured in our experiments also might contribute to the induction of the apoptotic program (for rev. see [11]). In any case, the identified inhibitors exert their effects by stimulating apoptosis in tumor cells, which is most beneficial for tumor therapy. Based on its anti-tumor potential the CSN kinase inhibitor curcumin is already in phase I clinical trials [14] and perhaps inhibitors identified here will follow.
Competing interests
For compound BTB14431 a patent application is pending for assignee Charité.
Authors' contributions
MF designed and carried out the study, interpreted the data and drafted the manuscript; RP, WD, CF conceived, coordinated, and designed the study, interpreted the data and drafted the manuscript; XH participated in the cell culture experiments; RD participated in the in vitro assays. All authors have read and approved the final manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
This work is supported by the BMBF funded Berlin Center for Genome Based Bioinformatics (BCB). We thank the Director of the Department of Surgery, Dr. J. M. Müller, for continuous support.
Figures and Tables
Figure 1 3D superposition and 2D comparison of curcumin and piceatannol. Comparison of curcumin and piceatannol structures by three-dimensional (3D) and two-dimensional (2D) superposition. A, The lead structure curcumin (blue) was aligned in a 3D superposition with the structure of piceatannol (green) obtained in the database. B, 2D similarities of curcumin and piceatannol are demonstrated. Both structures contain two aromatic rings and a large number of H-bond-acceptors. Using this information the Tanimoto-coefficient shown in Table 1 was calculated.
Figure 2 In vitro kinase assays with piceatannol and BTB14431. Inhibition of CSN-associated kinases by curcumin- and emodin-derived compounds in vitro. A, Piceatannol, identified by searching the database with curcumin as a lead compound and B, BTB14431, a compound found by searching the database with emodin. Recombinant CK2 and PKD as well as purified CSN were used to phosphorylate c-Jun in vitro. The first lane (0) is without inhibitors. Piceatannol (A) and BTB14431 (B) were added in increasing concentrations (20, 50, 100 and 200 μM). The shown autoradiographs were evaluated by densitometry. The obtained data were plotted against increasing inhibitor concentrations and used to calculate IC50 values summarized in Table 2. The demonstrated results are representative for at least four independent experiments.
Figure 3 Effect of curcumin- and emodin-derived inhibitors on the stability of c-Jun and p53. Curcumin- and emodin-derived inhibitors of CSN-associated kinases affect the stability of c-Jun and p53 in HeLa or B8 cells. HeLa cells or B8 cells were incubated for 4 h with the inhibitors emodin, resveratrol, piceatannol, BTB14431 and JFD02836. The first lane is the control without inhibitor. Proteins of cell lysates were separated by SDS-PAGE and transferred to nitrocellulose. The Western blots analysis was probed with an anti-c-Jun or an anti-p53 antibody.
Figure 4 Measurement of caspase activity after cell treatment with curcumin- and emodin-derived inhibitors. Curcumin- and emodin-derived inhibitors of CSN-associated kinases induce apoptosis in B8 cells. Caspase-3/7 activities in mouse B8 fibroblasts were determined using a DEVD-peptide. Cells (104/well) were treated with the inhibitors curcumin, piceatannol, BTB14431, SEW04213 and JFD02836 for 4 h at 37°C. After incubating cells with caspase-3/7 reagent for 3 h at room temperature the fluorescence (RFLU: relative fluorescent light units) was measured at an excitation wavelength of 485 nm and an emission wavelength of 530 nm.
Figure 5 Detection of DNA-fragmentation in HeLa cells after 24 h incubation with kinase inhibitors. Dose-dependent DNA-fragmentation in HeLa cells after 24 h treatment with curcumin- and emodin-derived inhibitors. (A) Representative data for piceatannol (20 μM and 50 μM) and BTB14431 (20 μM and 50 μM) are shown. As a control the cells were incubated with DMSO (0.1%). Piceatannol and BTB14431 were added in two concentrations (20 μM and 50 μM). The marker indicates the percentage of subG1 cells (apoptotic cells). (B) The bar chart shows the percentage of hypodiploid cells (apoptotic cells) after treatment with the inhibitors curcumin, emodin, resveratrol, piceatannol, BTB14431 or JFD02836 at different concentrations (20 μM and 50 μM).
Figure 6 Apoptosis or necrosis: Annexin-V-FITC /PI double staining of HeLa cells after treatment with CSN kinase inhibitors. HeLa cells were cultured in medium containing 0.1% DMSO or the curcumin- and emodin-derived inhibitors at different concentrations (20 μM and 50 μM). After 24 h cell death was measured at the single-cell level by labeling cells with annexin-V-FITC and counterstaining with propidium iodide (PI). (A) Representative data for piceatannol and BTB14431 are shown. The numbers indicate the percentage of cells in each quadrant (lower left: FITC-/PI-, intact cells; lower right: FITC+/PI-, apoptotic cells; upper left: FITC-/PI+, necrotic cells; upper right: FITC+/PI+, late apoptotic or necrotic cells). (B) The bar chart describes the percentual distribution of necrotic, apoptotic and viable cells after treatment with curcumin, emodin, resveratrol, piceatannol or BTB14431.
Figure 7 MTT test for detecting the cell viability after treatment with kinase inhibitors. The cells were treated with curcumin, emodin, resveratrol, piceatannol or BTB14431 for 24 h at different concentrations (20 μM and 50 μM). As a control the cells were cultured with 0.1% DMSO. The bar chart displays the amount of viable cells after treatment.
Table 1 Creation of two groups of potential inhibitors of CSN-associated kinases. The structures found in the database by 3D and 2D screening were divided into two groups (curcumin-group and emodin-group) depending on the lead structures curcumin or emodin. The analysis of the 2D structures abet the division into the two groups. The double line marks the threshold, which normally limits the hits (2D similarity <85%).
Trade name 2D-structure 2D similarity (Tanimoto-coefficient)
Curcumin-group
Curcumin 1.0
Resveratrol 0.7
Piceatannol 0.7
BTB00363 0.7
Emodin-group
Emodin 1.0
BTB 14431 0.9
JFD02836 0.9
SEW 04213 0.8
JFD03665 0.8
Table 2 Determination of IC50 values in an in vitro kinase assay. IC50 values (μM) for the inhibition of recombinant CK2, PKD and the purified CSN complex by the inhibitors of the curcumin-group, piceatannol as well as BTB00363, and of the emodin-group, BTB14431, JFD02836, SEW04213 and JFD03665 were determined. The two groups of inhibitors were created based on the different structures of the lead compounds (see Table 1). Kinase assays were performed as outlined in materials and methods. For comparison IC50 values of curcumin, resveratrol and emodin from earlier measurements [2] are shown.
Inhibitor IC50 CK2 (μM) IC50 PKD (μM) IC50 CSN (μM)
Curcumin-group
Curcurmin 11.8 4.1 2.6 *
Resveratrol 51.0 17.6 32.1 *
Piceatannol 2.5 0.5 1.7
BTB00363 332.7 117.7 188.6
Emodin-group
Emodin 22.7 94.5 4.4 *
BTB14431 6.4 68.9 21.0
JFD02836 19.9 39.7 54.1
SEW04213 37.9 185.9 4.8
JFD03665 5.9 154.9 19.8
* [2]
==== Refs
Bech-Otschir D Kapelari B Dubiel W The COP9 signalosome: Its Possible Role in the Ubiquitin System Wiley-VCH Verlag GmbH & Co KGaA: Weinheim 2005
Uhle S Medalia O Waldron R Dumdey R Henklein P Bech-Otschir D Huang X Berse M Sperling J Schade R Dubiel W Protein kinase CK2 and protein kinase D are associated with the COP9 signalosome Embo J 2003 22 1302 1312 12628923 10.1093/emboj/cdg127
Marklund U Lightfoot K Cantrell D Intracellular location and cell context-dependent function of protein kinase D Immunity 2003 19 491 501 14563314 10.1016/S1074-7613(03)00260-7
Rykx A De Kimpe L Mikhalap S Vantus T Seufferlein T Vandenheede JR Van Lint J Protein kinase D: a family affair FEBS Lett 2003 546 81 86 12829240 10.1016/S0014-5793(03)00487-3
Bech-Otschir D Seeger M Dubiel W The COP9 signalosome: at the interface between signal transduction and ubiquitin-dependent proteolysis J Cell Sci 2002 115 467 473 11861754
Bech-Otschir D Kraft R Huang X Henklein P Kapelari B Pollmann C Dubiel W COP9 signalosome-specific phosphorylation targets p53 to degradation by the ubiquitin system Embo J 2001 20 1630 1639 11285227 10.1093/emboj/20.7.1630
Seeger M Kraft R Ferrell K Bech-Otschir D Dumdey R Schade R Gordon C Naumann M Dubiel W A novel protein complex involved in signal transduction possessing similarities to 26S proteasome subunits Faseb J 1998 12 469 478 9535219
Naumann M Bech-Otschir D Huang X Ferrell K Dubiel W COP9 signalosome-directed c-Jun activation/stabilization is independent of JNK J Biol Chem 1999 274 35297 35300 10585392 10.1074/jbc.274.50.35297
Slee EA O'Connor DJ Lu X To die or not to die: how does p53 decide? Oncogene 2004 23 2809 2818 15077144 10.1038/sj.onc.1207516
Vogelstein B Lane D Levine AJ Surfing the p53 network Nature 2000 408 307 310 11099028 10.1038/35042675
Shaulian E Karin M AP-1 as a regulator of cell life and death Nat Cell Biol 2002 4 E131 6 11988758 10.1038/ncb0502-e131
Pollmann C Huang X Mall J Bech-Otschir D Naumann M Dubiel W The constitutive photomorphogenesis 9 signalosome directs vascular endothelial growth factor production in tumor cells Cancer Res 2001 61 8416 8421 11731421
Henke W Ferrell K Bech-Otschir D Seeger M Schade R Jungblut P Naumann M Dubiel W Comparison of human COP9 signalsome and 26S proteasome lid' Mol Biol Rep 1999 26 29 34 10363643 10.1023/A:1006991419464
Moos PJ Edes K Mullally JE Fitzpatrick FA Curcumin impairs tumor suppressor p53 function in colon cancer cells Carcinogenesis 2004 25 1611 1617 15090465 10.1093/carcin/bgh163
Arbiser JL Klauber N Rohan R van Leeuwen R Huang MT Fisher C Flynn E Byers HR Curcumin is an in vivo inhibitor of angiogenesis Mol Med 1998 4 376 383 10780880
Thimm M Goede A Hougardy S Preissner R Comparison of 2D similarity and 3D superposition. Application to searching a conformational drug database J Chem Inf Comput Sci 2004 44 1816 1822 15446841 10.1021/ci049920h
Chen X Reynolds CH Performance of similarity measures in 2D fragment-based similarity searching: comparison of structural descriptors and similarity coefficients J Chem Inf Comput Sci 2002 42 1407 1414 12444738 10.1021/ci025531g
Lipinski CA Drug-like properties and the causes of poor solubility and poor permeability J Pharmacol Toxicol Methods 2000 44 235 249 11274893 10.1016/S1056-8719(00)00107-6
http://www.accelrys.com/dstudio Discovery Studio, Accelrys: http://www.accelrys.com/dstudio Discovery Studio Accelrys
Snyder RD Pearl GS Mandakas G Choy WN Goodsaid F Rosenblum IY Assessment of the sensitivity of the computational programs DEREK, TOPKAT, and MCASE in the prediction of the genotoxicity of pharmaceutical molecules Environ Mol Mutagen 2004 43 143 158 15065202 10.1002/em.20013
Liu R Hsieh CY Lam KS New approaches in identifying drugs to inactivate oncogene products Semin Cancer Biol 2004 14 13 21 14757532 10.1016/j.semcancer.2003.11.003
Hoppe-Seyler F Butz K Repression of endogenous p53 transactivation function in HeLa cervical carcinoma cells by human papillomavirus type 16 E6, human mdm-2, and mutant p53 J Virol 1993 67 3111 3117 8388491
Wesierska-Gadek J Schloffer D Kotala V Horky M Escape of p53 protein from E6-mediated degradation in HeLa cells after cisplatin therapy Int J Cancer 2002 101 128 136 12209989 10.1002/ijc.10580
Athanassiou M Hu Y Jing L Houle B Zarbl H Mikheev AM Stabilization and reactivation of the p53 tumor suppressor protein in nontumorigenic revertants of HeLa cervical cancer cells Cell Growth Differ 1999 10 729 737 10593649
Elledge RM Lock-Lim S Allred DC Hilsenbeck SG Cordner L p53 mutation and tamoxifen resistance in breast cancer Clin Cancer Res 1995 1 1203 1208 9815913
Choudhuri T Pal S Agwarwal ML Das T Sa G Curcumin induces apoptosis in human breast cancer cells through p53-dependent Bax induction FEBS Lett 2002 512 334 340 11852106 10.1016/S0014-5793(02)02292-5
Kast WM Offringa R Peters PJ Voordouw AC Meloen RH van der Eb AJ Melief CJ Eradication of adenovirus E1-induced tumors by E1A-specific cytotoxic T lymphocytes Cell 1989 59 603 614 2555063 10.1016/0092-8674(89)90006-8
Shieh DE Chen YY Yen MH Chiang LC Lin CC Emodin-induced apoptosis through p53-dependent pathway in human hepatoma cells Life Sci 2004 74 2279 2290 14987952 10.1016/j.lfs.2003.09.060
Liontas A Yeger H Curcumin and resveratrol induce apoptosis and nuclear translocation and activation of p53 in human neuroblastoma Anticancer Res 2004 24 987 998 15161054
Youssef KM El-Sherbeny MA Synthesis and antitumor activity of some curcumin analogs Arch Pharm (Weinheim) 2005 338 181 189 15864788 10.1002/ardp.200400939
Adams BK Cai J Armstrong J Herold M Lu YJ Sun A Snyder JP Liotta DC Jones DP Shoji M EF24, a novel synthetic curcumin analog, induces apoptosis in cancer cells via a redox-dependent mechanism Anticancer Drugs 2005 16 263 275 15711178 10.1097/00001813-200503000-00005
Vermes I Haanen C Steffens-Nakken H Reutelingsperger C A novel assay for apoptosis. Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled Annexin V J Immunol Methods 1995 184 39 51 7622868 10.1016/0022-1759(95)00072-I
Jenkins JL Kao RY Shapiro R Virtual screening to enrich hit lists from high-throughput screening: a case study on small-molecule inhibitors of angiogenin Proteins 2003 50 81 93 12471601 10.1002/prot.10270
Stahura FL Bajorath J Virtual screening methods that complement HTS Comb Chem High Throughput Screen 2004 7 259 269 15200375
Graham DL Bevan N Lowe PN Palmer M Rees S Application of beta-galactosidase enzyme complementation technology as a high throughput screening format for antagonists of the epidermal growth factor receptor J Biomol Screen 2001 6 401 411 11788058
Yim H Lee YH Lee CH Lee SK Emodin, an anthraquinone derivative isolated from the rhizomes of Rheum palmatum, selectively inhibits the activity of casein kinase II as a competitive inhibitor Planta Med 1999 65 9 13 10083837 10.1055/s-1999-13953
Sarno S Moro S Meggio F Zagotto G Dal Ben D Ghisellini P Battistutta R Zanotti G Pinna LA Toward the rational design of protein kinase casein kinase-2 inhibitors Pharmacol Ther 2002 93 159 168 12191608 10.1016/S0163-7258(02)00185-7
Huang MT Lou YR Ma W Newmark HL Reuhl KR Conney AH Inhibitory effects of dietary curcumin on forestomach, duodenal, and colon carcinogenesis in mice Cancer Res 1994 54 5841 5847 7954412
Bai X Cerimele F Ushio-Fukai M Waqas M Campbell PM Govindarajan B Der CJ Battle T Frank DA Ye K Murad E Dubiel W Soff G Arbiser JL Honokiol, a small molecular weight natural product, inhibits angiogenesis in vitro and tumor growth in vivo J Biol Chem 2003 278 35501 35507 12816951 10.1074/jbc.M302967200
Li L Aggarwal BB Shishodia S Abbruzzese J Kurzrock R Nuclear factor-kappaB and IkappaB kinase are constitutively active in human pancreatic cells, and their down-regulation by curcumin (diferuloylmethane) is associated with the suppression of proliferation and the induction of apoptosis Cancer 2004 101 2351 2362 15476283 10.1002/cncr.20605
|
16083495
|
PMC1198225
|
CC BY
|
2021-01-04 16:03:06
|
no
|
BMC Cancer. 2005 Aug 5; 5:97
|
utf-8
|
BMC Cancer
| 2,005 |
10.1186/1471-2407-5-97
|
oa_comm
|
==== Front
BMC Dev BiolBMC Developmental Biology1471-213XBioMed Central London 1471-213X-5-161610721310.1186/1471-213X-5-16Research ArticleBMP4 signaling is involved in the generation of inner ear sensory epithelia Li Huawei [email protected] Carleton E [email protected] Zhengmin [email protected] Yanling [email protected] Yucheng [email protected] Hong [email protected] Stefan [email protected] Department of Otolaryngology and Program in Neuroscience, Harvard Medical School and Eaton Peabody Laboratory, Massachusetts Eye and Ear Infirmary, Boston, MA 02114, USA2 Department of Otolaryngology, Central Laboratory of Eye, Ear, Throat and Nose Hospital, Shanghai Medical College of Fudan University, Shanghai, 200031, PR of China2005 17 8 2005 5 16 16 24 6 2005 17 8 2005 Copyright © 2005 Li et al; licensee BioMed Central Ltd.2005Li 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 robust expression of BMP4 in the incipient sensory organs of the inner ear suggests possible roles for this signaling protein during induction and development of auditory and vestibular sensory epithelia. Homozygous BMP4-/- animals die before the inner ear's sensory organs develop, which precludes determining the role of BMP4 in these organs with simple gene knockout experiments.
Results
Here we use a chicken otocyst culture system to perform quantitative studies on the development of inner ear cell types and show that hair cell and supporting cell generation is remarkably reduced when BMP signaling is blocked, either with its antagonist noggin or by using soluble BMP receptors. Conversely, we observed an increase in the number of hair cells when cultured otocysts were treated with exogenous BMP4. BMP4 treatment additionally prompted down-regulation of Pax-2 protein in proliferating sensory epithelial progenitors, leading to reduced progenitor cell proliferation.
Conclusion
Our results implicate BMP4 in two events during chicken inner ear sensory epithelium formation: first, in inducing the switch from proliferative sensory epithelium progenitors to differentiating epithelial cells and secondly, in promoting the differentiation of hair cells within the developing sensory epithelia.
==== Body
Background
The inner ear is derived from a thickened patch of ectodermal cells, the otic placode, which develops lateral to the developing hindbrain. In birds and mammals, the otic placode invaginates and pinches off to form the pear-shaped otic vesicle, or otocyst. The inner ear's sensory epithelia originate from the ventromedial region of the otocyst, an area that can be defined by the expression of several markers, such as BEN (bursal epithelium and neurons)[1], lunatic fringe and cSerrate-1 [2-4], islet-1 [5] and bone morphogentic proteins (BMPs) [4,6,7]. The distinctive temporal and spatial expression patterns of BMP4 in almost all developing chicken inner ear sensory epithelia has led to the suggestion that this protein may play an important role in the induction of inner ear sensory organs [4,6]. Because homozygous BMP4 knockout mice die between E6.5 and E9.5 [8], a period before the inner ear has formed, it is difficult to analyze the role of BMP4 in the genesis of inner ear sensory organs. As a consequence, other model systems, such as the chicken embryo, have been used to study possible roles of BMP4 in the inner ear. For example, noggin, an antagonist of BMPs, has been employed to interfere with BMP signaling during chicken inner ear development resulting in defects in semicircular canal formation and otic capsule malformation [9,10]. Although malformed or missing cristae were observed in these experiments, the hair cells developed normally. Thus these results did not clarify why BMP4 is robustly expressed in sensory epithelia primordia thereby neither confirming nor refuting the hypothesis that BMP signaling is involved in the genesis of inner ear sensory organs. A possible explanation for the previously observed lack of sensory epithelia defects after in ovo application of BMP4 antagonists is that the antagonists did not penetrate far enough to reach sufficiently high concentrations in the developing sensory epithelia to block BMP signaling effectively. To address this issue, we exploited a serum-free floating otocyst culture system, which allowed us to quantitatively analyze progenitor cell proliferation, apoptosis and cell differentiation in the developing otocyst with loss of function and gain of function experiments. Our experiments revealed that BMP4 signaling is involved in generation of sensory epithelia by negatively regulating inner ear progenitor cell proliferation through downregulation of the homeodomain transcription factor Pax-2. Subsequently, BMP4 signaling promotes hair cell differentiation.
Results
Exogenous BMP4 leads to increase in hair cell numbers and blockade of BMP signaling inhibits hair cell generation
Maintenance of the developing avian inner ear in vitro in an environment with largely reduced extrinsic influences can be facilitated by removal of periotic mesenchyme, which otherwise could serve as source for various signaling molecules [11,12]. As much as possible, we assured removal of the mesenchymal tissue surrounding the otocyst by mild enzymatic treatment with trypsin and careful dissection. We found that removal of the periotic mesenchyme did not affect the capability of stage 16 otic vesicles to generate major inner ear cell types after 7d in serum-free suspension culture (Fig. 1). Vesicles with obvious mesenchymal contributions that gave rise to cells with chondrocyte morphology (Fig. 1A, B) were excluded from this study. Only otic vesicles that did not display any obvious sign of mesenchymal contribution after culture were included in this study (Fig. 1C, D).
Figure 1 Stage 16 otocysts cultured for 7 days develop hair cells and neurons, independent of the presence of mesenchymal tissue. (A) Incomplete removal of periotic mesenchyme is clearly visible in cryosections of otocysts cultured for seven days. The asterisk labels such an area of mesenchyme-derived cells with chondrocyte morphology that label poorly with phalloidin (shown in B), indicating lower levels of F-actin in mesenchymal derivatives. (B) Hair cells, visualized by green myosin VIIA immunofluorescence and neurons (red β-III tubulin immunofluorescence) develop in the presence of mesenchyme. Filamentous actin is visualized with phalloidin-labeling (blue fluorescence) (C) Mesenchyme-free otocysts do not display mesenchyme-derived cell morphologies. (D) Hair cells (green) and neurons (red) develop in mesenchyme-free otocysts. Filamentous actin is visualized with phalloidin-labeling (blue fluorescence). (E) β-III tubulin-positive neurites (red) appear to contact myosin VIIA-positive hair cells shown in green (arrows). (F) Hair cells develop hair bundles that can be visualized with antibody to espin (green fluorescence), hair cell antigen (red fluorescence), and filamentous actin (blue fluorescence).
Although free-floating otocysts did not differentiate into morphologically well-defined and structured inner ears, they displayed remarkable constancy with regard to the number of total cells after 7 days in culture (approximately 40,000 cells per otocyst; Mutai and Heller, unpublished data) and of the number of hair cells formed during the culture period (648 ± 117 hair cells per otocyst, n = 8; about 1.6% of the total cell number). We identified hair cells with antibody to myosinVIIA [13] and by their hair bundles labeled with antibody to the hair bundle markers espin and hair cell antigen [14-16] (Fig. 1E, F). Furthermore, the hair cells that differentiated in free-floating otocysts appeared to be contacted by neurites, identified with antibody to neuron-specific β-III tubulin (Fig. 1E).
To assess the effects of BMP signaling on otocyst cells and hair cell generation, we analyzed complete sets of serial sections obtained from individual specimens of each experimental group. When exogenous BMP4 was added on culture day three, we observed a substantial increase in the number of hair cells, when compared with the untreated controls (Fig. 2A, B, F, G); hair cells were identified by double-immunostaining for myosinVIIA and hair cell antigen. Blocking of BMP signaling with either noggin or soluble BMP receptor 1a or 1b proteins dramatically reduced hair cell numbers after seven days in culture (Fig. 2C, E). The effect of noggin on hair cell generation was dosage-dependant and could be rescued by adding exogenous BMP4 (Fig. 2D, E). Exposing cultures to different concentrations of BMP4 revealed an optimal concentration of 3–5 ng/ml, which yielded the greatest number of hair cells (Fig, 2E).
Figure 2 Increased production of hair cells with exogenous BMP4 and decrease in hair cell numbers in response to blockade of BMP signaling. (A) Hair cell antigen (red) and myosin VIIA-positive hair cells (green) in a cryosectioned otocyst after seven days culture. Filamentous actin is visualized with phalloidin-labeling (blue fluorescence). We routinely observed both hair cells organized in epithelia (see also higher magnification in (F)) and scattered hair cells (arrowhead). (B) BMP4 at 5 ng/ml, applied on the third day in culture, leads to substantial increase in the number of hair cells in epithelial cells and also in the scattered population of isolated hair cells. (C) Noggin at 0.5 μg/ml diminishes the number of hair cells. (D) The effect of noggin-treatment (0.5 μg/ml)) can be rescued by addition of 5 ng/ml exogenous BMP4. (E) Dose-dependence of the effect of exogenously added BMP4 on the number of hair cells in otocysts after seven days in culture. BMP4 at 3 ng/ml and at 5 ng/ml significantly increases the number of hair cells when compared with control conditions (asterisks indicate p < 0.05, unpaired Student's t-test, n = 4–5). Noggin at various concentrations and soluble BMPR 1a and 1b significantly reduced the number of hair cells detected in otocysts after seven days in culture when compared to the untreated control (asterisks indicate p < 0.003, unpaired Student's t-test, n = 6–7); the effect of 0.5 μg/ml noggin can be fully rescued by addition of 5 ng/ml BMP4. Error bars represent standard deviations. (F,G) Higher magnification to show the morphology of hair cells observed in (A) and in (B).
BMP4 inhibits otocyst cell proliferation and induces apoptosis in prospective inner ear ganglion cells
To elucidate the effect of BMP4 on the increased generation of hair cells, we analyzed cell proliferation and apoptosis in cultures treated with exogenous BMP4 from day three in culture onward. We found that proliferation was significantly reduced in otocysts treated with BMP4 whereas apoptosis was increased (Fig. 3A, B). BMP4-induced apoptosis almost exclusively occurred in areas that we identified as neuronal using the strong islet-1 immunostaining of cochleovestibular neuron nuclei (Fig. 3C–D). Islet-1 is expressed in the developing inner ear in neurons as well as in the nascent sensory epithelia; neuronal islet-1 staining can easily be distinguished from islet-1-positive developing sensory epithelia by a more round shape of stained neuronal nuclei and higher intensity when compared to epithelial staining (see also [5]). The paired-box transcription factor Pax-2 is expressed in developing sensory epithelia and is down-regulated in cochleovestibular ganglion cells [17] and accordingly, neuronal domains in cultured otocysts lack Pax-2-expression (Fig. 3C–D). When BMP signaling was blocked with noggin, there was no significant increase in overall cell proliferation, but less apoptosis occurred in the neuronal area, identified by staining for β-III tubulin (Fig. 3E–G). Based on these results, we rule out that the increase in hair cell numbers in response to BMP4 is based on a higher rate of cell proliferation or lower rate of apoptosis.
Figure 3 Exogenous BMP4 leads to decreased overall proliferation and increased apoptosis in the neural domain of the cultured otocysts. (A) BMP4, when added at 10 ng/ml or 20 ng/ml to otocysts after three days in culture, significantly inhibits proliferation, tested by an 8 h BrdU pulse 24 h after adding BMP4. (Asterisks indicate p < 0.05 (10 ng/ml) and p < 0.01 (20 ng/ml), unpaired Student's t-test, n = 7; error bars represent standard deviations). (B) Addition of BMP4 after three days in culture leads to an increase in TUNEL-positive cells when analyzed on the seventh day (asterisks indicate p < 0.01 (10 ng/ml) and p < 0.001 (20 ng/ml), unpaired Student's t-test, n = 7; error bars represent standard deviations). (C) In sections of control otocysts analyzed on the seventh day in culture, apoptotic cells, identified by TUNEL staining (red), mostly occurred outside of the Pax-2-positive areas (shown in blue). (D) BMP4 at 10 ng/ml noticeably increased the number of TUNEL-positive cells in the potential neural domain, identified by green islet-1-positive cell nuclei. (E,F) Similar to (C,D), but the neural domain is labeled with antibody to beta-III tubulin (shown in green). Note the increase of TUNEL-positive cell numbers within the neural domain. (G) Noggin does not affect the number of apoptotic cells.
Otocyst cells fail to commit to a sensory epithelial fate in the absence of BMP signaling
The sensory epithelia of the inner ear, including hair cells and supporting cells, are derived from Pax-2-positive progenitor cells that also give rise to other inner ear cell types [17-20]. The reduction in the number of hair cells in response to blockade of BMP signaling with noggin was not accompanied by a detectable overall reduction of Pax-2-positive progenitor cells (Fig. 4A, B). As the sensory epithelium progenitor cells become committed to form sensory patches, Pax-2 expression decreases and islet-1 becomes detectable in both nascent hair cells and supporting cells before the onset of expression of hair cell markers [5,17]. We found that in noggin-treated otocysts, islet-1-positive epithelial patches containing hair cells appeared noticeably diminished, whereas the more strongly islet-expressing cochleovestibular neurons were apparently not affected (Fig. 4C, D). These data support our hypothesis that the generation of sensory patches from Pax-2-expressing progenitors depends on BMP signaling.
Figure 4 Sensory epithelium fails to develop in absence of BMP signaling. (A,B) Blockade of BMP signaling with 0.5 μg/ml noggin robustly decreased the number of hair cell antigen-expressing hair cells (HCA shown in red) that were detectable after seven days in culture. It appears that Pax-2 expression (shown in green) is not affected by noggin. (C,D) In noggin-treated otocysts, islet-1-positive sensory epithelia (visualized in green) were evidently diminished, but not completely absent (arrowheads), as shown in this representative set of images. The images shown are not representative for the neural domains where we observed that the overall number of islet-1-positive neurons (arrows in C and D) did not differ between noggin-treated and control otocysts (no significant difference, n = 5). The inset in (C) shows a higher magnification of the islet-1-positive sensory epithelia (see also Li et al., 2004b).
BMP4 inhibits cell proliferation through downregulation of Pax-2
The expression of Pax-2 in the early developing inner ear is usually correlated with increased cell proliferation and absence of apoptosis, which mostly occurs outside of the Pax-2-positive regions [17,21]. When BMP4 was added at the beginning of the 7d culture period, we noticed a substantial decrease in the number of Pax-2-positive cells, although without an apparent effect on the developing sensory epithelium (Fig. 5A–F). Western blot analysis corroborated that BMP4 down-regulates Pax-2 expression in a dose-dependent manner (Fig. 5G, H).
Figure 5 Downregulation of Pax-2 by BMP4. (A-C) Control group after seven days in culture. Similar to the in vivo situation [5], (B) islet-1-positive cells (green) appear in the incipient Pax-2-expressing sensory epithelia (red); (C) myosin VIIA-positive early hair cells (red) that express islet-1 (green) occur within islet-1-positive epithelial patches. (D-F) In otocysts treated with 10 ng/ml BMP4 from the beginning of the culture period, Pax-2 expression was clearly reduced, whereas islet-1-expression and hair cell generation were not affected. (G) In a comparative Western blot analysis, Pax-2-protein levels were clearly reduced in otocysts treated for 48 h with BMP4. The amount of protein loaded per lane was the total protein of 5 otocysts. (H) Similar experiment to the one shown in (G), but the duration of the BMP4-treatment was reduced to 24 h and the total protein load per lane was equalized. (I-K) In cryosections of stage 25 chicken inner ear, BMP4 mRNA is expressed in the basilar papilla sensory epithelium primordium (dark precipitate in (I)). (J)Pax-2 protein expression (depicted in red) becomes noticeably reduced in the BMP4-expressing area simultaneously when epithelial islet-1 expression (shown in green) occurs. Panel (K) is the merged image of (I) with (J).
In the normal inner ear, BMP4 mRNA is used as a marker to visualize presumptive sensory epithelium [4-7]. At the crucial time during development, when sensory epithelia become postmitotic, epithelial islet-1 expression overlaps with BMP4-expressing areas (Fig. 5I–K). Conversely, in the same presumptive sensory epithelia, Pax-2 expression is downregulated (Fig. 5I–K), suggesting that BMP4 acts as suppressor of Pax-2 expression during normal inner sensory epithelium formation.
Discussion
BMP signaling is required for inner ear sensory organ formation
The discovery of BMP4 mRNA expression in all sensory organ primordia of the developing inner ear has led to the hypothesis that this signaling protein plays a role in the induction or differentiation of inner ear sensory epithelia [4,6]. However, because homozygous BMP4 knockout mice do not live beyond the ninth day of embryonic development it is not feasible to analyze BMP4 function during murine otogenesis.
To circumvent this problem, we used a serum-free floating culture technique, which allowed us to quantitatively test the function of BMP4 on avian inner ear sensory epithelium formation in a controlled environment. This culture regimen enabled us to quantitatively analyze the generation of inner ear cell types by using cell-specific markers. We found that the number of hair cells that formed in otocysts cultured for 7 days is considerably lower than the number of hair cells that can be found in a chicken ear at the 10th day of embryonic development (E10; equivalent to E3 plus 7 days in vitro). Nevertheless, we argue that the chicken inner ear at E10 is also substantially larger than the 7-day cultured otocyst, which makes it difficult to assess whether a ratio of ~1.6% hair cells of the total cell number in a floating otocyst is comparable to the ratio of hair cells versus total inner ear cell number at E10. Nevertheless, we suggest that that the effects we observed in vitro are at least partially representative of the hair cell populations found in the various hair cell-bearing organs of the developing chicken ear. To the best of our ability, we removed periotic mesenchyme from the E3 otocysts. Specimens with obvious mesenchymal contribution were not scored in this study. Nevertheless, we did not observe a noticeable difference in the number of hair cells and neurons detectable in otocysts with clear mesenchymal contribution, which is an indication that the periotic mesenchyme appears to exert little influence on the otocyst after stage 16 (E3).
Loss of BMP signaling by application of noggin or a combination of dominant-negative BMP receptors markedly reduced the number of hair cells in the developing otocyst, which strongly supports the hypothesis that BMP signaling is required for hair cell generation. Nevertheless, even at the highest concentrations of noggin or dominant-negative BMP receptors, no complete blockade of hair cell generation was achieved. We suspect that the nature of BMP signaling in the developing sensory epithelia does not allow complete blockade with application of soluble inhibitors. BMP4 is a secreted protein that binds to cell surface receptors to activate the Smad signaling pathway (reviewed in [22,23]). Secreted BMP4 acts on neighboring cells in a paracrine manner as well as in autocrine fashion on the cells that produce the factor if these cells have appropriate BMP receptors. We hypothesize that extracellular BMP inhibitors cannot completely interfere with autocrine signaling, resulting in an incomplete blockade of BMP signaling in otocyst cells that express both BMP4 and BMP receptors.
Interference with BMP signaling in the inner ear in vivo by application of noggin or noggin-producing cells [9,10] led to inhibition or malformation of semicircular canals accompanied by occasional deformation of ampullae. No effects on sensory epithelium formation and hair cells were reported; however, these studies did not explicitly quantify the number of hair cells. Alternatively, we hypothesize that dorso-lateral transplantation of beads soaked with noggin or of noggin-producing cells into the periotic mesenchyme apparently did not suffice to reach sufficient concentrations of the inhibitor over the course of several days in the ventro-medial region of the otocyst, where the sensory epithelia originate. Using a serum-free floating culture environment allowed us to control the concentration of BMP4 or BMP signaling inhibitors more precisely than previous in ovo experiments.
BMP4 promotes hair cell differentiation
Three mechanisms could account for increase of the number of hair cells by exogenous BMP4: 1) promotion of sensory epithelial progenitor cell proliferation, 2) prevention of progenitor cell apoptosis, or 3) promotion of progenitor cell differentiation to hair cells. Our results suggest that BMP4 inhibits otocyst cell proliferation, but affects apoptosis only in the neuronal domain of cultured otocycts, implying that a plausible mechanism of BMP4's effect on increasing the number of hair cells is to promote sensory epithelium progenitor cells to differentiate.
Another line of evidence for BMP4's effect on sensory epithelium progenitor cells arises from our loss-of-function analysis with noggin. Here we found that noggin-treatment did not affect the number of Pax-2-positive cells, whereas the number of hair cells was considerably reduced. This result implies that Pax-2-positive progenitors fail to differentiate into sensory epithelium when BMP signaling is blocked.
Our observations that BMP4-treatment leads to increased hair cell generation and to reduction of otocysts cell proliferation, suggests a dual, concentration-dependant, action of BMP4: at concentrations of 3–5 ng/ml BMP4 augments cell differentiation. In fact, we did not find a significant inhibition of cell proliferation with BMP4 concentrations below 10 ng/ml (data not shown). At higher BMP4 concentrations (10–20 ng/ml), cell proliferation is sufficiently reduced to offset the BMP4 effect on hair cell differentiation. The result of the concentration-dependant dual effects of BMP4 is manifested in a bell-shaped dose-dependency of hair cell numbers (Fig. 2E).
The effects of BMP4 on otocyst cell proliferation and apoptosis that we report here differ from, but do not contradict, results obtained by in vivo administration of noggin to the developing chicken inner ear [9,10]. In these experiments, cell proliferation decreased and apoptosis increased after noggin-secreting beads or cells were transplanted into the mesenchyme adjacent to the otocysts. We argue that these previous analyses and our experimental protocol are not easily comparable as many factors could account for the observed differences. For example, interactions of noggin with potential signals in the periotic mesenchyme might produce different effects on inner ear tissue when compared with mesenchyme-free otocyst cultures. Another possible explanation is that our analysis included all otocyst-derived cells, whereas previous studies focused on the parts of the developing inner ear surrounding the noggin-secreting grafts.
The effect of BMP4 on promoting differentiation of sensory epithelium progenitor cells is consistent with its temporal and spatial expression in the inner ear sensory organ development [4,6,7]. BMP4 mRNA is expressed in all sensory organ primordia before the hair cells and supporting cells are developed. At this crucial period of sensory organ formation, the progenitor cells downregulate Pax-2, leave the cell cycle, and initiate differentiation into hair cells and supporting cells; Pax-2 expression subsequently persists in hair cells [5,17]. These events appear to be correlated with the presence of BMP4 in sensory patches, supporting a possible role for BMP4 in controlling proliferation and specifying differentiation in the inner ear's sensory epithelia.
Additional clues about the potential targets of BMP4 in the developing inner ear could arise from analysis of expression of BMP receptors and components of the BMP signaling pathways. Beside a comprehensive study in the developing zebrafish ear and lateral line [24], only one study has addressed in the developing chicken inner ear expression of BMP receptors and Smad proteins, the intracellular transducers of BMP signaling [25]. Although the latter study was focused on the role of BMP signaling during development of semicircular canals, it should be noted that the expression of BMP receptors in the developing chicken inner ear is probably not restricted to sensory patches or the Pax-2-positive domain. A more widespread expression of BMP receptors could potentially be the reason for the substantial downregulation of Pax-2 protein that we observed in free-floating otocycts in response to BMP4 treatment. At least in the developing zebrafish inner ear several BMP receptors are expressed widely, some of them ubiquitously, at all developmental stages [24].
BMP4 is a crucial control factor of proliferation, differentiation and apoptosis in the developing inner ear
At early stages of inner ear development, all cells of the otic placode and the otic pit express the transcriptional regulator Pax-2. Subsequently and noticeably evident at embryonic day three, Pax-2-expression gets more localized to the medio-ventral part of the otic vesicle, the region where the sensory epithelia originate [18-20]. At this time of development, cell proliferation mostly occurs in the Pax-2 expression domain and apoptosis happens outside of the Pax-2 positive domain [17]. We show that the numbers of Pax-2-positive cells as well as proliferating cells are markedly reduced when exogenous BMP4 is added to cultured otocysts. Coinciding with decreasing numbers of Pax-2-positive cells, we also detected substantial downregulation of the expression of Pax-2 protein. In the developing sensory patches, we were able to visualize that BMP4 mRNA is detectable in the incipient sensory patches just at the time when islet-1 expression becomes evident. The onset of islet-1 expression in sensory patches coincides with early differentiation of sensory epithelial cells [5]. In parallel, Pax-2 protein is markedly downregulated in these patches, implying that BMP4 also causes downregulation of Pax-2, cessation of proliferation, and upregulation of markers for early differentiating sensory epithelium in vivo.
Conclusion
In summary, our results indicate that BMP signaling is involved in the generation of inner ear sensory patches and hair cells. We also found evidence that BMP4 can affect cell proliferation and apoptosis in different cell populations of the developing inner ear. We therefore conclude that correct control of BMP signaling in all areas of the developing otocyst is central to obtaining an accurate balance of cell proliferation, apoptosis and cell differentiation needed for correct morphogenesis of the inner ear's sensory organs.
Methods
Chicken embryos
Fertilized chicken eggs of the white leghorn strain (Charles River SPAFAS) were stored at 14°C until embryonic development was initiated by placing them onto rocking platforms into a humidified incubator maintained at 38°C. Embryos were staged according to Hamburger and Hamilton [26].
Otocyst culture
Otocysts from stage 15–16 embryos were dissected in phosphate-buffered saline (PBS, pH7.2) and incubated for 30s in trypsin (0.125% in PBS) to aid in the removal of the periotic mesenchyme. The otocysts were rapidly transferred into 10 mL of serum-free culture medium in Petri dishes (non-tissue culture treated). Culture medium was mixed from equal parts of high glucose Dulbecco's modified Eagle's medium and F12 medium supplemented with N2 and B27 (Media and supplements were from Invitrogen/GIBCO/BRL, Carlsbad, CA). BMP4, noggin, and soluble BMPR1a and BMPR1b were obtained from R&D Systems (Minneapolis, MN). BMP signaling antagonists were continuously supplemented throughout the culture period. BMP4 was either applied from the beginning of the culture period or from the third day (72 h) in culture onward. The floating otocysts were maintained in a humidified incubator in a 5% CO2 atmosphere at 37°C. For quantitative analysis, cultured otocysts were harvested on the seventh day in culture, fixed overnight with paraformaldehyde (4% in PBS), cryoprotected for 48 h in sucrose (30% in PBS), embedded in TissueTek (EMS) and serially sectioned (16 μm thick) with a cryostat (CM3050, Leica).
5-bromo-2-deoxyuridine (BrdU) labeling
BrdU (Sigma) was added (3 μg/ml) at specific time points for an exposure period of 8 h, after which the otocycts were harvested and processed for cryosectioning. BrdU incorporation was detected immunohistochemically (see below). Quantification was done by determining the fraction of BrdU-positive cells of the total cell number in every other section obtained from each otocyst.
Apoptosis labeling
Apoptotic cells were identified with the TUNEL labeling technique [27] (TUNEL enzymatic labeling kit, Roche). The fixed sections were washed three times with PBS, 5 min each, treated for 2 min with ice-cold 0.1% sodium tartrate, and incubated for 1 h with fluorescein-dNTP and terminal deoxynucleotideyl transferase at 37°C. After triple washing with PBS, the sections were immunostained for additional cell markers (see below).
Immunohistochemistry
The cryosections were blocked for 1 h with 1% BSA, 5% heat-inactivated goat serum and 0.1% Triton-100 in PBS (PBT1). For BrdU labeling, the sections were exposed to 2N HCl for 30 min before adding the primary antibody. The slides were then incubated overnight at 4°C in PBT1 with diluted antibodies: 1:5000 for mouse anti-HCA (a gift from G. Richardson [28]), 1:3000 for rabbit anti-myosin VIIA (a gift from A. El-Amraoui and C. Petit), 1:100 for monoclonal anti-islet-1 (clone 40.3A4, cell culture supernatant, Developmental Studies Hybridoma Bank, University of Iowa, Iowa City, IA), 1:100 for rabbit anti-Pax2 (Covance, Princeton, NJ), 1:500 for monoclonal anti-neuron-specific β-III tubulin (TuJ) (Chemicon), and 1:1000 for mouse anti-BrdU (Sigma). Unbound antibodies were removed with three PBT1 washes and one PBT2 (same as PBT1 but without serum) wash for 15 min each at room temperature. FITC-conjugated, TRITC-conjugated, and cy5-conjugated goat anti-rabbit and anti-mouse secondary antibodies (Jackson ImmunoResearch) were used at a dilution of 1:400 in PBS. A 45–60 min incubation period in the secondary-antibody mixture preceded three washes for 15 min each in PBS. Counterstaining with long-wavelength nuclear staining agent TOTO-3 (Molecular Probes) was done to visualize cell nuclei and TRITC-conjugated phalloidin was used to visualize filamentous actin. The coverslipped slides were analyzed by confocal microscopy (TCS SP2, Leica). For quantitative studies, we used the data measured in all serial sections of individual specimen and we calculated mean values from individual specimens derived from at least three independent experiments. For counting of labeled cell nuclei, we analyzed each section with ImageJ (version 1.31 v for Mac OSX available at ). A section thickness of 16 μm cannot completely exclude double counting of hair cells but assessment of pilot experiments with complete series' of sections counterstained with TOTO-3 did not reveal any evidence that our analysis overestimated hair cell numbers. Incomplete series were not used for quantitative assessments of cell numbers.
In situ hybridization/fluorescent immunostaining
Digoxigenin-labeled sense and antisense probes for chicken BMP4 were synthesized (DIG RNA Labeling Kit, Roche) from 1 μg of linearized plasmid DNA and resuspended in 100 μl water. In situ hybridization was initiated by rehydrating the sections in 100 μl diluted probe (1:100) in 50% (v/v) formamide, 10% (w/v) dextran sulfate, 1 mg·ml-1 yeast RNA, 1 × Denhardt's solution, 185 mM NaCl, 5.6 mM NaH2PO4, 5 mM Na2HPO4, 5 mM EDTA, and 15 mM Tris at pH7.5 and preheated to 70°C. After coverslipping and overnight incubation at 65°C in a chamber humidified with 50% (v/v) formamide in 150 mM NaCl and 15 mM trisodium citrate at pH7 (1 × SSC), the coverslips were removed in 5 × SSC and the slides were washed twice for 30 min each in 50% (v/v) formamide and 0.1% (v/v) polyoxyethylene sorbitan monolaurate (Tween-20) in 1 × SSC at 65°C. Thereafter, the slides were washed for 15 min in 0.2 × SSC and for 15 min in PBS at room temperature.
For the first antibody detection, the sections were blocked for 1 h in PBT1. The slides were then incubated for 2 h at room temperature with alkaline phosphatase-conjugated anti-digoxigenin Fab fragments in PBT1 (1:1000; Boehringer Mannheim). Unbound Fab fragments were removed by washing twice for 30 min each in PBT2. For detection, the sections were covered with 100 μl of nitro blue tetrazolium chloride and 5-bromo-4-chloro-3-indolyl phosphate substrate (1-STEP NBT/BCIP, Pierce), coverslipped, and incubated overnight at room temperature in a humidified chamber. For the second immunostaining with fluorescent antibodies, coverlips were removed in PBS and the slides were incubated overnight at 4°C in PBT1 with primary antibodies to Pax-2 and islet-1. Fluorescent secondary antibody detection and image acquisition by confocal microscopy was conducted as described above.
Western blot analysis
Proteins were separated on 7.5% SDS-polyacrylamide gels and transferred to nitrocellulose membranes using a semi-dry transfer system (Trans-Blot SD, BioRad). Relative amounts of total protein and concentration differences among samples were determined with a protein assay kit (BCA kit, PIERCE). Western blot membranes were incubated for 1 h at room temperature in 2.5% (vol/vol) Liquid Block (Amersham Pharmacia Biotech) and 0.1% (vol/vol) Tween-20 in PBS. Pax-2 antiserum was diluted 1:250 in 2.5% (vol/vol) Liquid Block and 0.1% (vol/vol) Tween-20 in PBS and blots were incubated overnight at 4°C in diluted antiserum. Unbound primary antibodies were removed by four washes for 15 min each at room temperature in 0.1% Tween-20 in PBS. Bound primary antibodies were detected with horseradish peroxidase-conjugated antibody to rabbit IgG (Amersham Pharmacia Biotech) at a dilution of 1:7500 in 0.1% Tween-20 in PBS. Unbound secondary antibodies were removed by two washes of 15 min each in 0.1% Tween-20 in PBS and two washes for 15 min each in PBS. Detection was performed with chemiluminescence substrate (ECL plus, Amersham Pharmacia Biotech) and exposure to Hyperfilm ECL (Amersham Pharmacia Biotech).
Authors' contributions
HuL carried out most otic vesicle culture experiments, immunohistochemistry, in situ hybridization, microscopical imaging, and statistical analysis. EC participated in immunohistochemical and quantitative analyses and performed Western blot experiments. ZW, YZ, and YW provided corroborating evidence for the effects observed and contributed additional quantitative data. HoL carried out cryosectioning and organotypic culture experiments. HuL and SH conceived of the study, HuL drafted the manuscript and SH finalized the manuscript, which has been read and approved by all authors.
Acknowledgements
The authors thank Dr. D.K. Wu for the chicken BMP4 cDNA and Sabine Mann for assistance with Western blot analysis. This work was supported by grants DC04563 and DC006167 (S.H.) and by Core grant P30 DC05209 from the National Institutes of Health and by the National Nature Science Foundation of China, grant 30271397 as well as by the Shanghai Science and Technology Commission grant 04JC14094 (both to H.L.).
==== Refs
Goodyear RJ Kwan T Oh SH Raphael Y Richardson GP The cell adhesion molecule BEN defines a prosensory patch in the developing avian otocyst J Comp Neurol 2001 434 275 288 11331529 10.1002/cne.1177
Myat A Henrique D Ish-Horowicz D Lewis J A chick homologue of Serrate and its relationship with Notch and Delta homologues during central neurogenesis Dev Biol 1996 174 233 247 8631496 10.1006/dbio.1996.0069
Adam J Myat A Le Roux I Eddison M Henrique D Ish-Horowicz D Lewis J Cell fate choices and the expression of Notch, Delta and Serrate homologues in the chick inner ear: parallels with Drosophila sense-organ development Development 1998 125 4645 4654 9806914
Cole LK Le Roux I Nunes F Laufer E Lewis J Wu DK Sensory organ generation in the chicken inner ear: contributions of bone morphogenetic protein 4, serrate1, and lunatic fringe J Comp Neurol 2000 424 509 520 10906716 10.1002/1096-9861(20000828)424:3<509::AID-CNE8>3.0.CO;2-Q
Li H Liu H Sage C Huang M Chen ZY Heller S Islet-1 expression in the developing chicken inner ear J Comp Neurol 2004 477 1 10 15281076 10.1002/cne.20190
Oh SH Johnson R Wu DK Differential expression of bone morphogenetic proteins in the developing vestibular and auditory sensory organs J Neurosci 1996 16 6463 6475 8815925
Takemura T Sakagami M Takebayashi K Umemoto M Nakase T Takaoka K Kubo T Kitamura Y Nomura S Localization of bone morphogenetic protein-4 messenger RNA in developing mouse cochlea Hear Res 1996 95 26 32 8793505 10.1016/0378-5955(95)00233-2
Winnier G Blessing M Labosky PA Hogan BL Bone morphogenetic protein-4 is required for mesoderm formation and patterning in the mouse Genes Dev 1995 9 2105 2116 7657163
Gerlach LM Hutson MR Germiller JA Nguyen-Luu D Victor JC Barald KF Addition of the BMP4 antagonist, noggin, disrupts avian inner ear development Development 2000 127 45 54 10654599
Chang W Nunes FD De Jesus-Escobar JM Harland R Wu DK Ectopic noggin blocks sensory and nonsensory organ morphogenesis in the chicken inner ear Dev Biol 1999 216 369 381 10588886 10.1006/dbio.1999.9457
Montcouquiol M Kelley MW Planar and vertical signals control cellular differentiation and patterning in the mammalian cochlea J Neurosci 2003 23 9469 9478 14561877
Ladher RK Anakwe KU Gurney AL Schoenwolf GC Francis-West PH Identification of synergistic signals initiating inner ear development Science 2000 290 1965 1967 11110663 10.1126/science.290.5498.1965
Hasson T Heintzelman MB Santos-Sacchi J Corey DP Mooseker MS Expression in cochlea and retina of myosin VIIa, the gene product defective in Usher syndrome type 1B Proc Natl Acad Sci U S A 1995 92 9815 9819 7568224
Zheng L Sekerkova G Vranich K Tilney LG Mugnaini E Bartles JR The deaf jerker mouse has a mutation in the gene encoding the espin actin-bundling proteins of hair cell stereocilia and lacks espins Cell 2000 102 377 385 10975527 10.1016/S0092-8674(00)00042-8
Li H Liu H Balt S Mann S Corrales CE Heller S Correlation of expression of the actin filament-bundling protein espin with stereociliary bundle formation in the developing inner ear J Comp Neurol 2004 468 125 134 14648695 10.1002/cne.10944
Bartolami S Goodyear R Richardson G Appearance and distribution of the 275 kD hair-cell antigen during development of the avian inner ear J Comp Neurol 1991 314 777 788 1816275 10.1002/cne.903140410
Li H Liu H Corrales CE Mutai H Heller S Correlation of Pax-2 expression with cell proliferation in the developing chicken inner ear J Neurobiol 2004 60 61 70 15188273 10.1002/neu.20013
Lawoko-Kerali G Rivolta MN Holley M Expression of the transcription factors GATA3 and Pax2 during development of the mammalian inner ear J Comp Neurol 2002 442 378 391 11793341 10.1002/cne.10088
Hutson MR Lewis JE Nguyen-Luu D Lindberg KH Barald KF Expression of Pax2 and patterning of the chick inner ear J Neurocytol 1999 28 795 807 10900085 10.1023/A:1007057719025
Sanchez-Calderon H Martin-Partido G Hidalgo-Sanchez M Differential expression of Otx2, Gbx2, Pax2, and Fgf8 in the developing vestibular and auditory sensory organs Brain Res Bull 2002 57 321 323 11922981 10.1016/S0361-9230(01)00725-0
Lang H Bever MM Fekete DM Cell proliferation and cell death in the developing chick inner ear: spatial and temporal patterns J Comp Neurol 2000 417 205 220 10660898 10.1002/(SICI)1096-9861(20000207)417:2<205::AID-CNE6>3.0.CO;2-Y
Mehler MF Mabie PC Zhang D Kessler JA Bone morphogenetic proteins in the nervous system Trends Neurosci 1997 20 309 317 9223224 10.1016/S0166-2236(96)01046-6
Itoh S Itoh F Goumans MJ Ten Dijke P Signaling of transforming growth factor-beta family members through Smad proteins Eur J Biochem 2000 267 6954 6967 11106403 10.1046/j.1432-1327.2000.01828.x
Mowbray C Hammerschmidt M Whitfield TT Expression of BMP signalling pathway members in the developing zebrafish inner ear and lateral line Mech Dev 2001 108 179 184 11578872 10.1016/S0925-4773(01)00479-8
Chang W ten Dijke P Wu DK BMP pathways are involved in otic capsule formation and epithelial-mesenchymal signaling in the developing chicken inner ear Dev Biol 2002 251 380 394 12435365 10.1006/dbio.2002.0822
Hamburger V Hamilton HL A series of normal stages in the development of the chick embryo. 1951 Dev Dyn 1992 195 231 272 1304821
Gorczyca W Traganos F Jesionowska H Darzynkiewicz Z Presence of DNA strand breaks and increased sensitivity of DNA in situ to denaturation in abnormal human sperm cells: analogy to apoptosis of somatic cells Exp Cell Res 1993 207 202 205 8391465 10.1006/excr.1993.1182
Richardson GP Bartolami S Russell IJ Identification of a 275-kD protein associated with the apical surfaces of sensory hair cells in the avian inner ear J Cell Biol 1990 110 1055 1066 2182645 10.1083/jcb.110.4.1055
|
16107213
|
PMC1198226
|
CC BY
|
2021-01-04 16:30:32
|
no
|
BMC Dev Biol. 2005 Aug 17; 5:16
|
utf-8
|
BMC Dev Biol
| 2,005 |
10.1186/1471-213X-5-16
|
oa_comm
|
==== Front
BMC GenomicsBMC Genomics1471-2164BioMed Central London 1471-2164-6-1011604279410.1186/1471-2164-6-101Methodology ArticleApplication of four dyes in gene expression analyses by microarrays Staal Yvonne CM [email protected] Herwijnen Marcel HM [email protected] Schooten Frederik J [email protected] Delft Joost HM [email protected] Department of Health Risk Analysis and Toxicology, Maastricht University, P.O. box 616, 6200 MD Maastricht, The Netherlands2005 25 7 2005 6 101 101 25 2 2005 25 7 2005 Copyright © 2005 Staal et al; licensee BioMed Central Ltd.2005Staal 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
DNA microarrays are widely used in gene expression analyses. To increase throughput and minimize costs without reducing gene expression data obtained, we investigated whether four mRNA samples can be analyzed simultaneously by applying four different fluorescent dyes.
Results
Following tests for cross-talk of fluorescence signals, Alexa 488, Alexa 594, Cyanine 3 and Cyanine 5 were selected for hybridizations. For self-hybridizations, a single RNA sample was labelled with all dyes and hybridized on commercial cDNA arrays or on in-house spotted oligonucleotide arrays. Correlation coefficients for all combinations of dyes were above 0.9 on the cDNA array. On the oligonucleotide array they were above 0.8, except combinations with Alexa 488, which were approximately 0.5. Standard deviation of expression differences for replicate spots were similar on the cDNA array for all dye combinations, but on the oligonucleotide array combinations with Alexa 488 showed a higher variation.
Conclusion
In conclusion, the four dyes can be used simultaneously for gene expression experiments on the tested cDNA array, but only three dyes can be used on the tested oligonucleotide array. This was confirmed by hybridizations of control with test samples, as all combinations returned similar numbers of differentially expressed genes with comparable effects on gene expression.
==== Body
Background
DNA microarray technology is widely used for gene expression analysis studies [1-5], as it is a high throughput technique by which the expression of all genes in a whole genome can be studied in a single assay. For many microarrays, the probe consists of cDNA or oligonucleotides spotted on a glass slide, and the target is fluorescent labelled cDNA (or cRNA). Both direct as well as indirect labelling protocols are applied: either, one target cDNA or cRNA is labelled with a single dye and hybridized on a microarray slide, or two targets are labelled with two different dyes, one for the reference and one for the test sample, and co-hybridized on a microarray slide. In dual label experiments, most often Cyanine 3 (Cy3) and Cyanine 5 (Cy5) are used as fluorescent dyes, although other dyes have been suggested [6]. In this way differential expression for thousands of genes between two different RNA samples can be measured simultaneously. Usually these experiments are time consuming, and, because microarray slides and fluorescent labels are expensive, the experiments are also high in costs. Moreover, several replicates need to be performed to increase statistical significance and to detect small differences in gene expression [7,8].
The application of four different dyes to label targets would be a major advantage as fewer microarrays will be required, leading to a reduction of costs and time without compromising gene expression data. A larger number of samples can be compared directly on a single microarray by labelling with more dyes, suggesting that fewer arrays will be required and that the hybridization design can be further optimized [9,10]. For instance, in the case that four samples need to be compared in all combinations, a dual-label common reference design requires four arrays for a single analysis of each sample, whereas a four-label design would require no common reference because all samples can be hybridized on a single array and only one array for a single analysis of each sample is needed. This will reduce variation, since variation between signal intensities for two dyes on a single spot is much smaller than variation between spots on different arrays [11]. Furthermore, day to day variation is reduced since it is possible to achieve more hybridizations on the same day [12]. In toxicogenomics assessments, as well as in other research areas, the approach to use multiple dyes can be of high value as it allows comparing several exposure conditions or time series simultaneously.
Forster et al [13] were the first to study the feasibility of using a third dye (Alexa 594) for labelling in microarray based gene expression analyses. Although they found that Alexa 594 gave a small signal in the Cy3 channel during scanning and Cy3 gave a small signal in the Alexa 594 channel, they concluded that Alexa 594 could be used besides Cy3 and Cy5 for direct comparison of two experimental samples and measuring these samples in relation to a reference sample.
The goal of our study was to investigate whether more than three different fluorescent dyes can be applied in gene expression studies using DNA microarrays. This was studied using microarrays with cDNA and oligonucleotide probes by hybridizing with a single sample labelled with four dyes (a quadruple self-hybridization or further stated as self-hybridization). Self-hybridization experiments are useful for measuring microarray data variability since any deviation from the expected value of 0 (for log transformed data) is caused by systemic or technical variation [13,14]. We also studied the application of more than two dyes for gene expression changes caused by exposure of cells to benzo[a]pyrene, to verify that the new dyes can be applied simultaneously in microarray studies. In the present study, we demonstrate that on our cDNA arrays four dyes can be applied, but that hybridization on the oligonucleotide arrays should be restricted to three dyes.
Results
Selection of fluorescent dyes
Four different dyes were tested for signal cross-talk at the emission / excitation settings of the ScanArrayExpress, namely Alexa 488, Alexa 594, Cyanine 3 and Cyanine 5. Therefore, the fluorescence of each dye at scanner settings of all tested dyes was measured. Results are summarized in Table 1. Since none of these dyes gives hardly any signal at settings for any other dye, it can be concluded that all dyes can be used simultaneously and were therefore considered suitable for use in microarray experiments. These dyes were further examined on two different microarray platforms.
Table 1 Cross-fluorescence of tested dyes. Fluorescence of dyes at scanner settings of all dyes expressed as a percentage of fluorescence at its own settings (the latter was set to 100%).
dye Alexa 488 Alexa 594 Cy3 Cy5
settings
Alexa 488 100.0 0.0 1.1 0.0
Alexa 594 0.9 100.0 1.3 13.0
Cy3 0.0 2.2 100.0 0.0
Cy5 0.1 0.1 0.2 100.0
Optimizing laser power and PMT gain settings
The cDNA microarray from PHASE-I Molecular Toxicology was hybridized with a single cDNA target labelled with four different dyes (Cy3, Cy5, Alexa 488 and Alexa 594). Initial laser power settings for Alexa 488, Alexa 594, Cyanine 3 and Cyanine 5 were respectively 93, 91, 89 and 80%, and initial PMT gain settings were respectively 72, 71, 61, 60%. In order to obtain the optimal scan settings for each dye, the array is scanned at different laser power and PMT gain setting. Figure 1 shows, as an example, the data for varying laser power and PMT gain settings for Alexa 594.
Figure 1 (A) Effect of varying laser power settings on Alexa 594 fluorescence signals. Results for the PHASE-I cDNA microarray scanned with constant PMT gain and varying laser power. Average 10log transformed fluorescence data for each gene of the dyes at varying setting (y-axis) was plotted against the initial fluorescence data (x-axis). (B) Effect of varying PMT gain settings on Alexa 594 fluorescence signals. Results for the PHASE-I cDNA microarray scanned with constant laser power and varying PMT gain. Average 10log transformed fluorescence data for each gene of the dyes at varying setting (y-axis) was plotted against the initial fluorescence data (x-axis)
In the scatter plots of data of one scan versus another, in general the data points indicate parallel lines when the settings are varied between the scans, implying that the fluorescent signals are consistent for all levels of gene expression when targets are labelled with these dyes. The larger distribution of the data points at low signals is a normal effect, which is due to reduced accuracy to measure signals from low expressed genes. Compared to Alexa 594, varying laser settings gave similar results for Alexa 488. For Cy3 and Cy5, the data points in the scatter plots run parallel for each setting. Varying laser settings gave similar results for all tested dyes. The Alexa 488 and Alexa 594 graphs, however, show a minor disturbance in the lines of the data points when the laser is varied (shown for Alexa 594 in Figure 1a). This suggests that for these two dyes, a fixed laser power should always be applied, whereas the other dyes allow some variation. Furthermore, these data indicate that laser power and PMT gain can be varied to some extend without affecting relative gene expression levels, as long as there is no saturation of signal intensities.
We also tested photo bleaching of the 4 dyes by scanning the microarray slide up to 5 times with the same scanner settings for all 4 dyes, and plotted the mean signal intensities as percentage of the signal intensity at the first scan (Figure 2). As is evident, photo bleaching occurs for all dyes as for all the signals decreases. The reduction was highest for Alexa 488 and least for Alexa 594, but was always small (<11% between the first and second round of scanning). Furthermore, the signal-to-noise ratio did not change for either of the dyes after repetitive scanning (data not shown). Therefore, we conclude that the photo bleaching is not expected to hamper gene expression analyses on microarrays.
Figure 2 Photo bleaching of Alexa 488, Alexa 594, Cy3 and Cy5 after repetitive scanning of the microarray. Mean signal intensity of Alexa 488, Alexa 594, Cy3 and Cy5 is presented after repetitive scanning, relative to the signal at the first scan.
Correlation coefficients between dyes at different laser power settings
The influence of laser power and PMT gain settings on the correlations between the combinations of dyes to a trend line was examined. A cDNA microarray was scanned at the initial settings (mentioned above), and with a laser power of 70% or 100% and with adjusted PMT gain until none of the spots gave saturated signals. Results are shown in the Table 2. These correlation coefficients show that for all possible combinations of dyes, increasing the laser power, and thereby reducing the PMT gain, results in a higher correlation coefficient. This suggests that these cDNA microarrays with targets labelled with Alexa 488, Alexa 594, Cy3 and Cy5, and scanned with the ScanArrayExpress, could best be scanned at 100% laser power setting and adjusted PMT gain settings, in order to obtain the smallest variation in gene expression values. Although the correlations are high and differences are marginal, the poorest correlation for the first array was found for Alexa 488 combined with Cy5 (0.935), and the highest correlation for Alexa 594 with Cy3 (0.988).
Table 2 Correlation coefficients between gene expression measured for different dyes at various laser settings for the cDNA array. Correlation coefficients for all genes on the PHASE-I cDNA array between combinations of dyes at different scanner settings for 1 array, and at laser power 100 settings for 5 arrays.
array1 laser power = 100 (n = 5)
Settings laser = 70 original settings laser = 100 mean ± stdev
Dye combination
Alexa 488 vs Alexa 594 0.953 0.967 0.965 0.953 ± 0.020
Alexa 488 vs Alexa Cy3 0.916 0.941 0.955 0.923 ± 0.044
Alexa 488 vs Cy5 0.890 0.935 0.935 0.942 ± 0.014
Alexa 594 vs Cy3 0.958 0.983 0.988 0.938 ± 0.053
Alexa 594 vs Cy5 0.946 0.978 0.979 0.986* ± 0.005
Cy3 vs Cy5 0.975 0.981 0.987 0.926* ± 0.062
Mean 0.940 0.964 0.968 0.945 ± 0.025
Original scanner settings for Alexa 488, Alexa 594, Cy3 and Cy5 are: laser power 93, 91, 89 and 80% respectively, and PMT Gain 72, 71, 91, 60% respectively.
* The correlation coefficient between Alexa 594 – Cy5 is significantly higher than the correlation coefficient for any dye combination with Alexa 488, and the combination Alexa 594-Cy3 has a significantly higher correlation coefficient than Cy3-Cy5 (t-test, p < 0.05).
The reproducibility was tested by several other self-hybridizations of different RNA samples. Table 2 shows the results for the correlation coefficients calculated for all combinations of dyes. Numerical data from the table indicate that correlation coefficients for the repeated experiments are similar with mean correlation coefficients varying between 0.923 (Alexa 488 and Cy3) and 0.986 (Alexa 594 and Cy5).
For the rat oligonucleotide microarray, also self-hybridizations with targets labelled with Cy3, Cy5, Alexa 488 and Alexa 594 were also conducted and the laser power was set to 70 or 100% with adjustment of the PMT gain until no saturation of fluorescence occurred. Table 3 represents the correlation coefficients for these settings, and similar on this array, the correlations for all combinations of dyes are higher at laser power settings of 100% compared to 70%. However in all cases the correlation coefficients were smaller (varying between 0.486 and 0.887) compared to the cDNA array. Furthermore, Table 3 shows that correlations between Alexa 488 and any other dye are much lower than the correlation for any of the other combinations. This is probably due to the high background fluorescence for Alexa 488 on these arrays compared to the spot signals. The ratio of mean spot signal to mean background variation (signal-to-noise ratio) was clearly lower for Alexa 488 then for the other dyes (namely, 1.25, 1.65, 2.88 and 1.88 for Alexa 488, Alexa 594, Cy3 and Cy5, respectively). The high background signal in the Alexa 488 channel can not be due to auto-fluorescence of the Corning slides alone as it was not observed when scanning an unhybridized microarray.
Table 3 Correlation coefficients between gene expression measured for different dyes at various laser settings for the oligonucleotide array. Correlation coefficients for all genes on the oligonucleotide array between combinations of dyes at different scanner on 1 array, and at laser power 100 settings for 7 arrays. Alexa 488 labelled samples were only hybridized on the first array.
array1 laser power = 100 (n = 7)
Settings laser = 70 laser = 100 mean ± stdev
Dye combination
Alexa 488 vs Alexa 594 0.118 0.561
Alexa 488 vs Alexa Cy3 0.080 0.512
Alexa 488 vs Cy5 0.127 0.486
Alexa 594 vs Cy3 0.334 0.857 0.855 ± 0.032
Alexa 594 vs Cy5 0.279 0.808 0.853 ± 0.027
Cy3 vs Cy5 0.313 0.887 0.890* ± 0.052
Mean 0.208 0.685 0.843 ± 0.072
* The correlation coefficient between Cy3 – Cy5 is significantly higher than the correlation coefficient of Alexa 594 – Cy3 (t-test, p < 0.05).
To reduce the background binding on the oligonucleotide arrays, we applied several different hybridisation and washing protocols. We varied BSA concentration in the hybridization buffer, added tRNA, Cot1 or PolyA and used a commercial hybridization buffer (DIG Easy Hyb granules, Roche, Germany). We also varied the concentrations SSC and SDS in the washing buffers. The best results for all dyes were obtained by using the hybridization protocol as described in "Microarray hybridizations" of the Methods section. The data from this most optimal protocol are presented here.
With the exclusion of Alexa 488, the other dyes were tested in several more self-hybridizations with for each array a different RNA sample in order to confirm the reproducibility. Table 3 shows the correlation coefficients for all combinations of the 3 dyes. The correlation coefficients are similar for all repetitive experiments with mean values varying between 0.854 and 0.891.
Standard deviation in relation to spot intensity for all combinations of dyes
The standard deviation for the 10log transformed expression ratios of the 3 or 4 replicate spots per gene on the arrays was calculated and plotted against the mean signal intensity of the corresponding dyes (Figure 4). For both arrays, the standard deviation decreased with increasing gene expression level. For the cDNA array, the standard deviation was equal for all combinations of dyes at a 10log signal intensity of 3 and higher. At lower signal intensities, however, the standard deviation for combinations of any dye with Alexa 488 were higher than for Cy3-Cy5 combinations, and standard deviations for combinations with Alexa 594 are intermediate. For the oligonucleotide array, the standard deviation for all combinations of dyes with Alexa 488 is higher at any signal intensity than for any other combination of dyes.
Figure 3 (A) Standard deviation of the expression ratio to the relative expression level for the PHASE-I cDNA array. Standard deviation of 10log transformed expression ratios for the 4 replicate spots of each gene (y-axis) plotted against the mean 10log transformed signal intensities (x-axis) for the corresponding dyes for all combinations of dyes and for the cDNA array. Regression lines are based on a power model. (B) Standard deviation of the expression ratio to the relative expression level for the Oligonucleotide array. Standard deviation of 10log transformed expression ratios for the 3 replicate spots of each gene (y-axis) plotted against the mean 10log transformed signal intensities (x-axis) for the corresponding dyes for all combinations of dyes and for the oligonucleotide array. Regression lines are based on a power model.
Identification of modulated genes for various dye combinations
As microarrays are intended to identify genes that are differentially expressed between different RNA samples, we tested the applicability of four dyes by analyzing RNA samples from cells exposed to 3 concentrations of B[a]P versus a vehicle control. Table 4 shows the labelling and hybridization schedule for the B[a]P exposed samples on the arrays (per array, four RNA samples were simultaneously hybridized), which was conducted to the two independent treatments (see Materials and Methods). Every dye was used for every RNA sample, but not each dye combination was applied for each combination of control and test sample. For every B[a]P concentration a confidence analysis was performed to select modulated genes for each dye combination separately. Also, for all dye combinations combined (paired data), a confidence analysis was conducted. For the cDNA array 20, 31 and 45 genes were found modulated for paired data of respectively 3, 10 and 30 μM. For the oligonucleotide array 121, 97 and 195 genes were found modulated for paired data of respectively 3, 10 and 30 μM. Modulated genes for each dye combination were compared to modulated genes found all dye combinations paired. Table 5 and 6 summarize the results for respectively the cDNA arrays and the oligonucleotide arrays; they present numbers of modulated genes for specific dye combinations as a percentage of numbers of modulated genes by all dye combinations combined (in bold). On average, this percentage is approximately 45%, although in some cases it is clearly lower or higher. This deviation, however, is not consistent for a dye or a combination of dyes, so it can be concluded that all dyes perform equally well in identifying differentially expressed genes. Also in these Tables, the different dye-combinations are compared to each other, all as a percentage of modulated genes by all dye combinations (in italics). Once again differences are observed, which are not sufficient consistent to conclude that one combination of dyes performs worse or better than another to identify modulated genes.
Table 4 Labelling schedule for B[a]P exposed samples. Each HepG2 sample is labelled with each fluorescent dye. Rat liver samples were labelled as shown by array number 1–3, without the application of Alexa 488.
fluorescent label
array no. Cy3 Cy5 Alexa 594 Alexa 488
1 0 μM 3 μM 10 μM 30 μM
2 10 μM 0 μM 30 μM 3 μM
3 3 μM 30 μM 0 μM 10 μM
4 30 μM 10 μM 3 μM 0 μM
Table 5a Performance of a dye combination in revealing modulated genes in B[a]P (3 μM) treated HepG2 cells using a cDNA array.
Cy5-Cy3 Cy3-A594 A488-Cy5 A594-A488
Cy5-Cy3 50*
Cy3-A594 40 50
A488-Cy5 25 35 50
A594-A488 40 40 25 40
* In bold the intersection of two gene lists indicating the modulated genes for a dye combination as percentage of all modulated genes (20 genes) found by analysis of all dye combinations combined. In italics the intersection of three gene lists indicating the modulated genes for two dye combinations as percentage of all modulated genes found for all dye combinations combined. The first dye was used for B[a]P treatment, the second for the control.
Table 5b Performance of a dye combination in revealing modulated genes in B[a]P (10 μM) treated HepG2 cells using a cDNA array.
Cy3-Cy5 A594-Cy3 Cy5-A488 A488-A594
Cy3-Cy5 29*
A594-Cy3 23 58
Cy5-A488 6 10 16
A488-A594 26 39 13 74
* In bold the intersection of two gene lists indicating the modulated genes for a dye combination as percentage of all modulated genes (31 genes) found by analysis of all dye combinations combined. In italics the intersection of three gene lists indicating the modulated genes for two dye combinations as percentage of all modulated genes found for all dye combinations combined. The first dye was used for B[a]P treatment, the second for the control.
Table 5c Performance of a dye combination in revealing modulated genes in B[a]P (30 μM) treated HepG2 cells using a cDNA array.
Cy3-A488 A488-Cy3 Cy5-A594 A594-Cy5
Cy3-A488 35*
A488-Cy3 24 53
Cy5-A594 22 33 38
A594-Cy5 16 20 16 31
* In bold the intersection of two gene lists indicating the modulated genes for a dye combination as percentage of all modulated genes (45 genes) found by analysis of all dye combinations combined. In italics the intersection of three gene lists indicating the modulated genes for two dye combinations as percentage of all modulated genes found for all dye combinations combined. The first dye was used for B[a]P treatment, the second for the control.
Table 6 Performance of a dye combination in revealing modulated genes in B[a]P treated liver slices using an oligonucleotide array.
B[a]P concentration
dye combination 3 μM 10 μM 30 μM
Cy5 – Cy3 36* 11
Cy3 – A594 64
Cy3 – Cy5 47 13
A594 – Cy3 48
Cy5 – A594 51 8
A594 – Cy5 38
* In bold the intersection of two gene lists indicating the modulated genes for a dye combination as percentage of all modulated genes (for 3, 10 and 30 μM: 121, 97 and 195 genes respectively) found by analysis of all dye combinations combined. In italics the intersection of three gene lists indicating the modulated genes for two dye combinations as percentage of all modulated genes found for all dye combinations combined. The first dye was used for B[a]P treatment, the second for the control.
Additionally, the performance of the dye combinations was evaluated by comparing the gene expression difference. Figure 4, which represents the results for the experiment with HepG2 cells on DNA microarrays with the application of four dyes simultaneously, can be used as an example. For each dye combination a similar effect on gene expression is observed and it can be summarized that all dye combinations result in similar gene expression changes. For the rat liver slices similar results were found.
Figure 4 (A) Gene expression difference for several genes after exposure of HepG2 cells to 3 μM B[a]P during 6 h. Gene expression difference for several genes, in varying relative gene expression level of high (FASN and HIST1H2AL), middle (CYP1A2, HMGCS1, VMP1 and IGFBP1) and low (CYP1A1, SLC22A3), as measured by different dye combinations by using four dyes simultaneously on cDNA microarrays in RNA samples from HepG2 cells exposed to 3 μM B[a]P during 6 hours. Error bars indicate the standard deviation for the replicate spots. (B) Gene expression difference for several genes after exposure of HepG2 cells to 10 μM B[a]P during 6 h. Gene expression difference for several genes, in varying relative gene expression level of high (FASN and HIST1H2AL), middle (CYP1A2, HMGCS1, VMP1 and IGFBP1) and low (CYP1A1, SLC22A3), as measured by different dye combinations by using four dyes simultaneously on cDNA microarrays in RNA samples from HepG2 cells exposed to 10 μM B[a]P during 6 hours. Error bars indicate the standard deviation for the replicate spots. (C) Gene expression difference for several genes after exposure of HepG2 cells to 30 μM B[a]P during 6 h. Gene expression difference for several genes, in varying relative gene expression level of high (FASN and HIST1H2AL), middle (CYP1A2, HMGCS1, VMP1 and IGFBP1) and low (CYP1A1, SLC22A3), as measured by different dye combinations by using four dyes simultaneously on cDNA microarrays in RNA samples from HepG2 cells exposed to 30 μM B[a]P during 6 hours. Error bars indicate the standard deviation for the replicate spots.
Discussion
We have investigated the applicability of four fluorescent dyes in gene-expression analysis by microarrays. By using more than two dyes in microarray experiments, without lessening the data obtained, costs and time can be decreased as fewer microarrays are needed.
Initially, several dyes were tested for cross-talk on the ScanArrayExpress reader, and ultimately 4 dyes were tested for parallel use in microarray experiments. Today, Cy3 and Cy5 are the most widely used dyes in microarray experiments and much research has been done on these dyes [4,8,11,15,16], although Alexa 555 and Alexa 647 have been suggested by Cox et al [6]. It was our intention to select dyes that could complement Cy3 and Cy5 and we show that Alexa 488 and Alexa 594 are suited for this and can be used for parallel hybridization in microarray experiments. All dyes were applicable on the tested cDNA arrays. On the tested oligonucleotide arrays, however, only three dyes, namely Alexa 594, Cy3 and Cy5, could be used.
Selection of fluorescent dyes
Based on cross-talk signals, four dyes – Alexa 488, Alexa 594, Cy3 and Cy5 – were found suitable for hybridization on microarrays and some cross-talk did occur for this combination. The highest fluorescence for a dye at settings of another dye was observed for Cy5, namely 13% cross-talk at the settings for Alexa 594. This cross-talk may influence differential gene expression analyses, especially if the signals for Cy5 and Alexa 594 differ drastically within a spot. Therefore, in order to minimize artificial gene expression differences, scan settings should be optimized such that emission intensities are gross similar (e.g. by assuring that the brightest spots are on the edge of saturation). Furthermore, dye swap design on replicate arrays will reduce the bias resulting from cross-talk, and algorithms can be developed to eliminate this bias.
Dye bias
Dye bias is the difference in labelling efficiency between different dyes as one dye can be better incorporated than another; this can affect the gene expression data [17-19]. When using more than one dye, dye bias may occur and most likely, it is enhances with increasing number of dyes. Dye bias can be reduced by using the indirect amino-allyl labelling instead of direct labelling, but it is not clear whether dye bias is fully eliminated [11]. However, dye bias can be eliminated by LOWESS normalization of the data, combined with a labelling and hybridization design in which each target is labelled with each different dye [20]. Liang et al [7] showed that the correlation between predicted and observed gene expression ratios increased by adding a second microarray with dye switching. This confirms that accuracy can be improved by adding dye swap replicates and applying a balanced labelling design. A balanced labelling design with four dyes may increase the number of required arrays, but still saves the total number of arrays. For example, when 3 treatments and a control are to be compared using 4 data points per comparison, 16 microarrays are needed for a common reference design, 12 arrays for a block design (treatment vs. control on an array), 8 when using a loop design, but only 4 with 4 dyes and the design shown in Table 4.
Applicability of selected dyes
The applicability of the dyes was analyzed in four different ways. First by calculating the correlation coefficients between dyes in self-hybridizations, second by calculating the standard deviation of their log ratio per gene for replicate spots in the self-hybridizations, third by comparing numbers of modulated genes for all dye combinations in samples exposed to B[a]P and finally by comparing gene expression modulation for several genes from samples exposed to B[a]P.
When applied on the cDNA array, all combinations of dyes gave high correlation coefficients (>0.9) and thus seem suitable for parallel hybridization in microarray experiments. On the oligonucleotide array, the correlation coefficients were high for all combinations (>0.8), except for combinations with Alexa 488 (≌0.5). The correlation coefficients for all combinations of dyes on both arrays are constant in multiple repeated hybridizations. These results are supported by the plots for the standard deviation of the replicate spots. For the cDNA array, the standard deviation is equal for all combinations of dyes at high gene expression level. However, for the oligonucleotide array the standard deviation of the signal intensity of high expressed genes for all combinations with Alexa 488 is higher than the standard deviation for all other combinations of dyes. Since the correlation coefficient of Alexa 488 with other dyes is low and the standard deviation for Alexa 488 is high, it is not advisable to use Alexa 488 for labelling and hybridization on the oligonucleotide array.
The correlation coefficients observed for all combinations of Alexa 488 with any other dye on the oligonucleotide array are lower than any of the other correlation coefficients. This was due to a high background signal and a lower signal-to-noise ratio in the Alexa 488 channel, which can not be attributed to auto fluorescence. This background signal was much less pronounced on the cDNA array, which may be explained by a different coating of the microarray slides. Alexa dyes have a net negative charge, which may cause non-specific electrostatic interaction with positively charged molecules [21]. This may be a reason for why the dye adhered differently to the two different microarray slides. However, this does not explain why the background binding for Alexa 594 is much less in comparison to that of Alexa 488.
For all dyes tested on the oligonucleotide array, many genes showed a low gene expression level compared to the cDNA array. In general, weak signals are detected with lower accuracy than strong ones [22]. This is reflected by the higher standard deviations for lower signals in the plots for the cDNA and oligonucleotide array (Figure 4). Lyng et al [22] showed that reliable data for mean signal intensities were only achieved within a range of 200 to 50,000 (no background correction performed). This clarifies the lower correlation coefficients found for the oligonucleotide array compared to the cDNA array.
For all dye combinations, percentages of modulated genes relative to modulated genes for all dye combinations combined are generally equal (Tables 5 and 6). This indicates that any dye combination has approximately the same sensitivity to identify differentially expressed genes, and that the traditional combination of Cy3-Cy5 is not necessarily preferable above the others. Therefore, we consider all dyes suitable for usage in gene expression studies by microarrays. This was further substantiated by the observation for several differentially expressed genes that the level of modulation is in the same range for all dye combinations.
Although Forster et al [13] used a different approach to test the use of Alexa 594 besides Cy3 and Cy5 in microarray analysis, their conclusions are in agreement with that of this study. Forster et al [13] tested the use of different combinations of two dyes in hybridization, and found some cross-talk between Cy3 and Alexa 594 and between Cy5 and Alexa 594. Although, some cross-talk was observed between Cy5 and Alexa 594 (13%), only small cross-talk was noticed (<3%) for Cy3 and Alexa 594 in this study. Forster et al [13] also found a more linear relation between Cy3 and Alexa 594 than for Cy3 and Cy5. However, we noticed only a small difference in correlation coefficient for Cy3 / Cy5 and for Alexa 594 / Cy3 (Table 2 and 3). These differences could be due to the different testing methods and different arrays used.
Conclusion
All our experiments demonstrate that for gene expression analyses on microarrays Alexa 594 is best suited as a third dye in addition to Cy3 and Cy5, and that Alexa 488 can be applied as a fourth dye on some microarray platforms, but unfortunately not on all array platforms. The general applicability of four dyes on other microarray systems is therefore uncertain, and needs to be investigated on a case-by-case basis.
Methods
Cross-talk analysis of fluorescent dyes
Two ARES™ Alexa fluor® dyes (Alexa 488 and 594) (Molecular Probes, Leiden, The Netherlands) and conventionally used Cyanine3 (Cy3) and Cyanine5 (Cy5) (Amersham Biosciences, Uppsala, Sweden) were tested for cross-talk of excitation / emission signals. All dyes were dissolved according to the producer's manual and applied on a glass slide. The slide was scanned with a ScanArrayExpress microarray scanner (Packard BioChip Technologies, Perkin Elmer life sciences, Boston, USA) with laser wavelengths for Alexa 488, Cy3, Alexa 594 and of 488, 543.8, 594 and 632.8 nm respectively, and emission filter of 522, 570, 614 and 670 nm respectively. The images were analyzed with ImaGene (BioDiscovery, USA). Fluorescence of each dye at the scanning settings of all tested dyes was measured and four dyes were selected for further use (see Results).
Source of RNA samples
RNA was isolated from cultured HepG2 cells or from rat liver or precision-cut liver slices and used for the microarray hybridizations. HepG2 cells were cultured in Minimal Essential Medium (MEM) supplemented with 1% non-essential amino acids, 1% sodium-pyruvate, 2% penicillin/streptomycin and 10% Foetal Bovine Serum (all from Gibco/BRL, Breda, The Netherlands) in T25 culture flasks at 37°C and 5% CO2. HepG2 cells were exposed to 3, 10 or 30 μM of Benzo[a]pyrene (B[a]P, from Sigma-Aldrich, the Netherlands) and a vehicle control (DMSO) during 6 hours in two independent experiments. DMSO concentration in de cell culture media was 0.1%. After exposure, media was removed and 1 ml Trizol (Gibco/BRL, Breda, The Netherlands) was immediately added to the cells.
A male Wistar albino rat (200 g) was killed by cervical dislocation, and the liver after removal, was snap frozen in liquid nitrogen and stored at -80°C. Liver tissue (8.6 g) was crushed using a mortar and pester. An amount of 0.05 g crushed liver tissue was dissolved in 1 ml Trizol reagent. Additionally, precision-cut liver slices were obtained by using a Krumdieck tissue slicer [23]. Cylindrical liver cores with a diameter of 8 mm were sliced into 250 μm thick slices. In the two independent experiments, slices were exposed to 3, 10 or 30 μM B[a]P or a solvent control (DMSO 0.067%) during 24 hours. After exposure, slices were snap frozen in liquid nitrogen and RNA was isolated in a manner similar to that of the whole liver tissue.
RNA isolation and cDNA synthesis
RNA was isolated from the Trizol solutions according to the producer's manual and purified with the RNeasy mini kit (Qiagen Westburg bv., Leusden, The Netherlands). RNA quantity was measured on a spectrophotometer and the quality was determined on a BioAnalyzer (Agilent Technologies, Breda, The Netherlands). Only RNA samples which showed clear 18S and 28S peaks were used for labelling and hybridization. In order to generate sufficient large uniform samples for the multiple self hybridizations, RNA samples from several isolations were pooled.
RNA was reverse transcribed into cDNA in quadruplicate with amino allyl labelled dUTP (Sigma-Aldrich, St Louis, USA) and subsequently labelled with one of the dyes (based on Van Delft et al [24]). For each sample, a mixture of 10 μg of RNA and 6 μg of random hexamer primers were incubated in 18.5 μl at 70°C for 10 minutes and snap frozen on dry ice / ethanol for 30 seconds. Thereafter DTT (final concentration 10 mM), 0.5 mM dATP, dCTP and dGTP, 0.3 mM dTTP, 0.2 mM 5-(3-aminoallyl)-2'deoxyuridine-5'-triphosphate (aa-dUTP), and 400 U Superscript II reverse transcriptase (Invitrogen, Life Technologies, Breda, The Netherlands) were added to a final volume of 30.1 μl and incubated overnight at 42°C. RNA was hydrolyzed by adding 10 μl of 1 M NaOH and 10 μl of 0.5 M EDTA followed by an incubation of 15 minutes at 65°C. To neutralize, 10 μl of 1 M HCl was added. cDNA samples were purified to remove unincorporated amino allyl dUTP and buffers using a QIAquick PCR Purification Kit (Qiagen Westburg bv., Leusden, The Netherlands) according to the producer's manual. However, in order to eliminate interference of amines during labelling, buffers were substituted by phosphate buffers (wash buffer: 5 mM KPO4 pH 8.0, 80% ethanol; elution buffer: 4 mM KPO4 pH 8.5). The sample was eluted in duplicate using 30 μl elution buffer and dried in vacuo. Following amino-allyl labelling, cDNA targets were resolved in 4.5 μl of 0.1 M Na2CO3 pH 9.0 and 4.5 μl of a 2.25 μM of Cy™5 or Cy™3 Monofunctional Reactive Dye esters (Amersham Biosciences, Uppsala, Sweden) was added. Samples were incubated in the dark at room temperature for 1 hour. Targets to be labelled with Alexa dyes were resolved in 5 μl of MilliQ, 3 μl of labelling buffer (Sodium bicarbonate, prepared according to the producers' manual) and 2 μl of a 6.3 μM of ARES™ Alexa Fluor® (Molecular Probes, Leiden, The Netherlands) (dissolved in DMSO according to the producers' manual) was added. The sample was incubated for 1 hour in the dark at room temperature. After incubation 35 μl of 100 mM NaAc pH 5.2 was added to the Cy-labelled targets and 90 μl MilliQ was added to the Alexa labelled targets. The samples were purified using a QIAquick PCR Purification Kit (Qiagen Westburg bv, Leusden, The Netherlands) to remove unincorporated dyes. To the rat liver targets, additional 4 μl of 100 U/ml Poly-dA (Amersham Biosciences, Uppsala, Sweden) and 3 μl of 1 mg/ml mouse Cot1-DNA (Invitrogen, Breda, The Netherlands) were added to block an unspecific binding of the targets to the array. The targets were dried in vacuo.
Microarray hybridizations
HepG2 targets were hybridized to the PHASE-1 Microarray Human-600 (PHASE-1 Molecular Toxicology, Santa Fe, USA), containing 597 sequence verified cDNA clones from human genes, representing a number of toxicologically relevant, as well as control, genes, each printed in quadruplicate. Hybridization and washing was done according to the producer's manual as previously described [15]. The labelled cDNA target was dissolved in 30 μl hybridization buffer (50% formamide, 5× SSC, 0.1% SDS, 0.1 mg/ml Salmon Sperm DNA) and incubated for 15 minutes in the dark at room temperature. The target was denatured by heating for 5 minutes at 95°C, centrifuged for 3 minutes at maximum speed, and placed in a heat block at 70°C until further use. The target (28 μl) was applied on the cover slip (24–32 mm) and the microarray was placed on top of the cover slip. The slide was hybridized overnight in a humidified hybridization chamber (Corning, Life Sciences, The Netherlands) in a water bath at 42°C. After incubation, the slide was placed in wash buffer (2× SSC, 30–34°C) to remove the cover slip, and washed 5 minutes in 2× SSC / 0.1% SDS, 5 minutes in 0.1× SSC / 0.1% SDS, 2 times 5 minutes in 0.1× SSC at 32°C, and 1 minute in MilliQ all at room temperature. The slide was centrifuged to dry.
Rat liver targets were hybridized on an Operon rat oligonucleotide array containing 5700 oligonucleotides (Operon, Qiagen, Venlo, The Netherlands) printed in triplicate on Corning UltraGAPS Coated Slides (Corning Life Sciences, New York, USA) by the Genome Centre Maastricht (Maastricht University, Maastricht, The Netherlands). Hybridization and washing was done according to Corning's protocol for oligonucleotide arrays. The labelled cDNA target was dissolved in 65 μl hybridization buffer (30% formamide; 5× SSC; 0.1% SDS) and incubated for 15 minutes in the dark at room temperature. The target was denatured by heating for 5 minutes at 95°C, centrifuged for 2 minutes at maximum speed, and kept at room temperature until further use. The microarray slide and cover slip (24 × 60 mm) were prehybridized for 45 minutes in preheated prehybridization buffer (5× SSC; 0.1% SDS; 1% BSA) at 42°C. Slides and cover slips were washed several times in MilliQ followed by dipping in isopropanol and centrifugation to dry. The target (60 μl) was applied on the cover slip and the microarray was placed on top of the cover slip. The slide was hybridized overnight in a humidified hybridization chamber (Corning, Life Sciences, The Netherlands) in a water bath at 42°C. After incubation, the slide was placed in wash buffer (2× SSC / 0.1% SDS) at 42°C to remove the cover slip. The slide was washed for 2 times 5 minutes in 2× SSC / 0,1% SDS at 42°C, 2 times 10 minutes in 0.1× SSC / 0.1% SDS at room temperature and 4 times 1 minute in 0.1× SSC at room temperature. The slide was centrifuged to dryness.
Microarray data analysis
The microarray slides were scanned on a ScanArrayExpress (Packard Biochip Technologies, Perkin Elmer life sciences, Boston, USA). All four channels were scanned at several different settings for laser power and / or photo multiplier tube (PMT Gain). Settings were optimized such that the signal of the highest fluorescent spots is just below the maximum measurable level. Laser power settings were set at 100% and PMT Gain was adjusted, unless otherwise stated. The images (10 micron resolution; 16 bit tiff) were processed with ImaGene 5.0 software (BioDiscovery Inc., Los Angeles, USA) to quantify spot signals. Irregular spots were manually or automatically flagged and not included in the data analysis.
For the self-hybridizations, data from ImaGene were exported to Microsoft Excel (Microsoft, USA) for transformations and analysis. For each spot, mean local background signal was subtracted from the mean spot signal, negative signals were excluded, and the resulting net spot signal data were log transformed. These log transformed background corrected expression signals for all combinations of dyes at all scanner settings were plotted and analyzed by linear regression and correlation coefficients (R2) were calculated. Furthermore, standard deviations of 10log transformed expression ratios for each gene (for 3 or 4 replicate spots, depending on the array used), were plotted against the mean 10log transformed expression signals and analyzed by regression analysis.
For the B[a]P exposed samples, data from ImaGene were transported to GeneSight software version 4.1.5 (BioDiscovery Inc, Los Angeles, USA) for transformations and analyses. For each spot, background was subtracted; flagged spots and spots with a net expression level below 5 were omitted. Data were log base 2 transformed and expression difference between exposed and control were calculated. Data normalization was done by LOWESS and centring expression differences by subtracting mean values (the latter only for the oligonucleotide arrays). Data of replicate spots were combined while omitting outliers (>2 standard deviations). In order to estimate the number of differentially expressed genes following a treatment, the confidence analysis tool from GeneSight was used. For confidence analyses, for each B[a]P concentration, data of the two replicate arrays with the same dye combination were combined. Up-regulated and down-regulated genes were identified at 99% confidence intervals with up-regulation or down regulation levels set at 0.2 (2log-scale) for the cDNA arrays and respectively 99.5% and 0.5 for the oligonucleotide arrays.
Authors' contributions
YS carried out the cell culturing and preparation of the liver slices including exposure and RNA isolation. YS and MvH carried out the labeling and hybridization of the samples and the image quantification. YS carried out the data analysis and drafted the manuscript. FvS and JvD participated in design of the study and preparation of the manuscript. All authors have read and approved the manuscript.
Acknowledgements
The authors thank Dr. C Ioannides and his colleagues for his help in making and the exposure of the rat liver slices and Ms D. Pushparajah for her help in correcting the English language. The research was carried out in the context of the AMBIPAH project (mechanism-based approaches to improved cancer risk assessment of ambient air polycyclic aromatic hydrocarbons), funded by the European Union (No. QLRT-2001-02402).
==== Refs
Duggan DJ Bittner M Chen Y Meltzer P Trent JM Expression profiling using cDNA microarrays Nat Genet 1999 21 10 14 9915494 10.1038/4434
Li X Gu W Mohan S Baylink DJ DNA microarrays: their use and misuse Microcirculation 2002 9 13 22 11896556 10.1038/sj.mn.7800118
Nuwaysir EF Bittner M Trent J Barrett JC Afshari CA Microarrays and toxicology: the advent of toxicogenomics Mol Carcinog 1999 24 153 159 10204799 10.1002/(SICI)1098-2744(199903)24:3<153::AID-MC1>3.0.CO;2-P
van Hal NLW Vorst O van Houwelingen AM Kok EJ Peijnenburg A Aharoni A van Tunen AJ Keijer J The application of DNA microarrays in gene expression analysis J Biotechnol 2000 78 271 280 10751688 10.1016/S0168-1656(00)00204-2
Quackenbush J Computational analysis of microarray data Nat Rev Genet 2001 2 418 427 11389458 10.1038/35076576
Cox WG Beaudet MP Agnew JY Ruth JL Possible sources of dye-related signal correlation bias in two-color DNA microarray assays Anal Biochem 2004 331 243 254 15265729 10.1016/j.ab.2004.05.010
Liang M Briggs AG Rute E Greene AS Cowley AWJ Quantitative assessment of the importance of dye switching and biological replication in cDNA microarray studies Physiol Genomics 2003 14 199 207 12799473
Rosenzweig BA Pine PS Domon OE Morris SM Chen JJ Sistare FD Dye bias correction in dual-labeled cDNA microarray gene expression measurements Environ Health Perspect 2004 112 480 487 15033598
Lee ML Kuo FC Whitmore GA Sklar J Importance of replication in microarray gene expression studies: statistical methods and evidence from repetitive cDNA hybridizations Proc Natl Acad Sci U S A 2000 97 9834 9839 10963655 10.1073/pnas.97.18.9834
Rodi CP Bunch RT Curtiss SW Kier LD Cabonce MA Davila JC Mitchell MD Alden CL Morris DL Revolution through genomics in investigative and discovery toxicology Toxicol Pathol 1999 27 107 110 10367683
Dobbin K Shih JH Simon R Questions and answers on design of dual-label microarrays for identifying differentially expressed genes J Natl Cancer Inst 2003 95 1362 1369 13130111
Chen JJ Delongchamp RR Tsai CA Hsueh HM Sistare F Thompson KL Desai VG Fuscoe JC Analysis of variance components in gene expression data Bioinformatics 2004 20 1436 1446 14962916 10.1093/bioinformatics/bth118
Forster T Costa Y Roy D Cooke HJ Maratou K Triple-target microarray experiments: a novel experimental strategy BMC Genomics 2004 5 13 15018645 10.1186/1471-2164-5-13
Hessner MJ Wang X Khan S Meyer L Schlicht M Tackes J Datta MW Jacob HJ Ghosh S Use of a three-color cDNA microarray platform to measure and control support-bound probe for improved data quality and reproducibility Nucleic Acids Res 2003 31 e60 12771224 10.1093/nar/gng059
Dobbin K Shih JH Simon R Statistical design of reverse dye microarrays Bioinformatics 2003 19 803 810 12724289 10.1093/bioinformatics/btg076
Wang Y Wang X Guo SW Ghosh S Conditions to ensure competitive hybridization in two-color microarray: a theoretical and experimental analysis Biotechniques 2002 32 1342 1346 12074165
Dombkowski AA Thibodeau BJ Starcevic SL Novak RF Gene-specific dye bias in microarray reference designs FEBS Lett 2004 560 120 124 14988009 10.1016/S0014-5793(04)00083-3
Tseng GC Oh MK Rohlin L Liao JC Wong WH Issues in cDNA microarray analysis: quality filtering, channel normalization, models of variations and assessment of gene effects Nucleic Acids Res 2001 29 2549 2557 11410663 10.1093/nar/29.12.2549
Yang YH Dudoit S Luu P Lin DM Peng V Ngai J Speed TP Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation Nucleic Acids Res 2002 30 e15 11842121 10.1093/nar/30.4.e15
Churchill GA Fundamentals of experimental design for cDNA microarrays Nat Genet 2002 32 Suppl 490 495 12454643 10.1038/ng1031
Panchuk-Voloshina N Haugland RP Bishop-Stewart J Bhalgat MK Millard PJ Mao F Leung WY Alexa dyes, a series of new fluorescent dyes that yield exceptionally bright, photostable conjugates J Histochem Cytochem 1999 47 1179 1188 10449539
Lyng H Badiee A Svendsrud DH Hovig E Myklebost O Stokke T Profound influence of microarray scanner characteristics on gene expression ratios: analysis and procedure for correction BMC Genomics 2004 5 10 15018648 10.1186/1471-2164-5-10
Hashemi E Dobrota M Till C Ioannides C Structural and functional integrity of precision-cut liver slices in xenobiotic metabolism: a comparison of the dynamic organ and multiwell plate culture procedures Xenobiotica 1999 29 11 25 10078837 10.1080/004982599238786
van Delft JH van Agen E van Breda SG Herwijnen MH Staal YC Kleinjans JC Discrimination of genotoxic from non-genotoxic carcinogens by gene expression profiling Carcinogenesis 2004 25 1265 1276 14963013 10.1093/carcin/bgh108
|
16042794
|
PMC1198227
|
CC BY
|
2021-01-04 16:32:48
|
no
|
BMC Genomics. 2005 Jul 25; 6:101
|
utf-8
|
BMC Genomics
| 2,005 |
10.1186/1471-2164-6-101
|
oa_comm
|
==== Front
BMC GenomicsBMC Genomics1471-2164BioMed Central London 1471-2164-6-1081609823010.1186/1471-2164-6-108Research ArticleComparative analysis of expression of histone H2a genes in mouse Nishida Hiromi [email protected] Takahiro [email protected] Hiroki [email protected] Yasuhiro [email protected] Yoshihide [email protected] Laboratory for Genome Exploration Research Group, RIKEN Genomic Sciences Center (GSC), RIKEN Yokohama Institute, 1-7-22 Suehiro-cho, Tsurumi-ku, Yokohama, Kanagawa 230-0045, Japan2005 13 8 2005 6 108 108 23 2 2005 13 8 2005 Copyright © 2005 Nishida et al; licensee BioMed Central Ltd.2005Nishida 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
At least 18 replication-dependent histone H2a genes are distributed in 3 Hist gene clusters on different chromosomes of the mouse genome. In this analysis we designed specific PCR primers for each histone H2a transcript and studied the expression levels and patterns using quantitative RT-PCR (qRT-PCR). In addition, we compared histone H3 K9 acetylation levels in the promoter regions of H2a genes by ChIP (chromatin immunoprecipitation) – quantitative PCR (qPCR) analysis.
Results
RT-PCR analysis indicated that all 20 histone H2a genes assessed in this study are expressed. The replication-dependent histone H2a genes have different expression levels but similar expression patterns. Among the 20 histone H2a genes, the expression-level of H2afz, a replication-independent gene, was highest, and that of Hist1h2aa, a replication-dependent gene, was lowest. Among 18 replication-dependent H2a genes, the expression level of Hist3h2a was highest. The ChIP-qPCR analysis showed that histone H3 K9 acetylation levels in promoter regions of both H2afz and Hist3h2a are clearly higher than that in the promoter region of Hist1h2aa. The H3 K9 acetylation level in the promoter of Hist1h2aa is similar to that in the γ-satellite region.
Conclusion
These results strongly suggest that histone H3 K9 acetylation plays a role in the expression of histone genes.
==== Body
Background
Eukaryotic genomic DNA is packaged with chromosomal proteins, forming chromatin. The most fundamental repeating unit of chromatin is the nucleosome. The nucleosome core consists of 146 bp of DNA wrapped around an octamer of histone proteins made up of 2 copies each of histones H2A, H2B, H3, and H4, in 1.65 turns [1]. Replication of the eukaryotic chromosomes requires the synthesis of histones to package the newly replicated DNA into chromatin. Control of the level of histone mRNA accounts for much of the control of histone protein synthesis [2]. It is still an open question as to how the expression of individual histone genes is controlled.
The variants and modifications of the histone proteins are related to chromatin structure [3-6]. Specific amino acids within histone tails are targets for a number of post-transcriptional modifications, i.e., acetylation, methylation, phosphorylation, and ubiquitination [3]. In particular, the modification of histone H3 K9 affects chromatin structure. H3 K9 methylation is enriched in transcriptionally silent genes and heterochromatin. On the other hand, H3 K9 acetylation is enriched in transcriptionally active genes [7]. Is this modification related to histone gene expression?
Eighteen replication-dependent histone H2a genes were identified in the mouse genome sequence [8]. Among these 18 genes, 13 are located in the Hist1 cluster on chromosome 13, 4 in the Hist2 cluster on chromosome 3, and 1 in the Hist3 cluster on chromosome 11 [8]. Thus, replication-dependent histone H2a genes are distributed in at least 3 Hist clusters. In addition, the mouse has 2 replication-independent histone H2a genes, H2afx on chromosome 9 and H2afz on chromosome 3. Recently we reported a novel replication-independent histone H2a gene (H2afj) on chromosome 6 [9]. H2afz and H2afj are typical replication-independent genes [9,10]. The H2afz protein is enriched in euchromatic regions and acts synergistically with a boundary element to prevent the spread of heterochromatin [6]. On the other hand, H2afx mRNA has both a polyadenylated tail and a stem-loop structure [11], elements typical of, respectively, replication-independent and replication-dependent histone genes.
As cells progress from G1 to S phase, the rate of histone gene transcription increases 3- to 5-fold, and the efficiency of histone pre-mRNA processing increases 8- to 10-fold, resulting in a 35-fold increase in histone protein levels [2,12]. Most promoters of histone genes have CCAAT and TATA boxes [9,13]. Some promoters have an E2F binding motif between the CCAAT and TATA boxes. This E2F binding motif is recognized, and then the E2F transcription factor activates an H2a gene in early S-phase of the cell cycle [14]. However, it is not known how transcription-related proteins cooperate to coordinately regulate histone gene transcription during the cell cycle.
The amino acid sequences of histone H2a proteins are very similar, except for that of H2afz protein [9]. For example, Hist1h2ab, 2ac, 2ad, 2ae, 2ag, 2ai, 2an, and 2ao encode the same structural protein. Among these 8 genes, Hist1h2ad and 2ao have the same nucleotide sequence; however, the others have different nucleotide sequences. Quantitative RT-PCR analysis can be used to show the expression levels of different genes (for example [15]). Thus, in this study we designed the specific PCR primers for each histone H2a gene and studied the expression levels and patterns by qRT-PCR.
Results and discussion
Each product of the qRT-PCR gave a single band on the agarose gel, located in the expected position (Fig. 1). This result indicates that all histone H2a genes are expressed in Hepa 1–6 cells. The expression levels of 18 replication-dependent histone genes and H2afx increased along with cell cycle progression from the beginning (0 h) of S-phase to the middle (2–4 h) of S-phase, and then decreased from the middle to the end (6 h) of S-phase (Fig. 2). On the other hand, the expression level of the replication-independent gene H2afz lacked such a single peak during S-phase (Fig. 2).
Figure 1 RT-PCR products. Lanes 1 and 19, DNA ladder marker; 2, Hist1h2aa transcript; 3, Hist1h2ab transcript; 4, Hist1h2ac transcript; 5, Hist1h2ad/1h2ao transcripts; 6, Hist1h2ae transcript; 7, Hist1h2af transcript; 8, Hist1h2ag transcript; 9, Hist1h2ah transcript; 10, Hist1h2ai/1h2aj transcripts; 11, Hist1h2ak transcript; 12, Hist1h2an transcript; 13, Hist2h2aa1/2h2aa2 transcripts; 14, Hist2h2ab/2h2ac transcripts; 15, Hist3h2a transcript; 16, H2afj transcript [9]; 17, H2afx transcript; 18, H2afz transcript.
Figure 2 Expression patterns and levels: results of a) first and b) second qRT-PCR analyses. X-axis, time (hours); Y-axis, expression level relative to H2afz expression level adjusted to 1.0 at 0 h in each experiment.
H2afz is regulated in a replication-independent manner, but H2afx is regulated in a replication-dependent manner. This pattern is consistent with the results of a previous report that indicated that H2afx gives rise to a cell-cycle-regulated mRNA ending in the stem-loop during S-phase, and a polyadenylated mRNA during G1-phase [10]. Therefore, H2afx is regulated in a replication-dependent manner (Fig. 2). On the other hand, H2afz lacks regulation of a polyadenylated mRNA. Interestingly, expression levels of H2afz decreased at the end (6 h) of S-phase, similar to those of replication-dependent genes (Fig. 2). This result suggests that the decrease at the end of S-phase is independent of the histone H2a mRNA structure.
We compared the sum of expression levels at 0, 1, 2, 3, 4, 5, and 6 h (S-phase) from each histone H2a gene (Fig. 3). Amino acid sequences from the proteins encoded by Hist1h2ab, 2ac, 2ad, 2ae, 2ag, 2ai, 2an, and 2ao were identical. However, among these 8 genes, the expression level of Hist1h2ae was 10 to 30 times that of Hist1h2ag (Fig. 3). Thus, the expression levels of the genes encoding the same structural protein were different.
Figure 3 Sum of expression levels at 0, 1, 2, 3, 4, 5, and 6 h (S-phase) in Fig. 2. Blue and red indicate a) and b) in Fig. 2, respectively. Y-axis, sum of expression levels.
Among the 13 genes in the Hist1 cluster, the expression level of Hist1h2ae was approximately 100 times that of Hist1h2aa (Fig. 3). In addition, the 4 genes in the Hist2 cluster had different expression levels. Thus, the expression level of Hist2h2aa1/2aa2 was approximately 10 times that of Hist2h2ab/2ac (Fig. 3). Therefore, the expression levels of genes belonging to the same gene cluster were different.
One possibility is that such different expression levels are caused by different promoters and different binding proteins bound to the promoters. For example, the promoters of Hist1h2ad, Hist1h2af, Hist1h2ag, and Hist1h2ah have the E2F binding motif (5'-TTTTCGCGCCC-3') between the CCAAT and TATA boxes [9]. Among these 4 replication-dependent genes, the expression level of Hist1h2ah was approximately 10 to 20 times that of Hist1h2ag (Fig. 3). In addition, compared among all 20 genes assessed in this paper, the expression levels of H2afz, Hist3h2a, Hist2h2aa1/2aa2, Hist1h2ae, and Hist1h2ai/aj were higher than that of Hist1h2ah, and those of Hist1h2ak and Hist1h2aa were lower than that of Hist1h2ag (Fig. 3). Thus, the relation between the E2F binding motif and the expression level is not clear. Unfortunately, we cannot determine here which structure of the promoters causes such different expression levels.
Next, we compared the histone H3 K9 acetylation levels in the promoter regions of H2afz (highest expression), Hist3h2a (highest expression among replication-dependent H2a genes), and Hist1h2aa (lowest expression). The ChIP-qPCR analysis showed that histone H3 K9 acetylation levels in the promoter regions of both H2afz and Hist3h2a were clearly higher than that in the promoter region of Hist1h2aa. The H3 K9 acetylation level in the promoter of Hist1h2aa was similar to that in the γ-satellite heterochromatin region (Table 1). This result indicates that the expression of histone H2a genes is related to the acetylation of histone H3 K9 in the promoter region.
Table 1 CT values of quantitative PCR for pull-down DNA fragments in ChIP analysis.
Hist1h2aa promoter Hist3h2a promoter H2afz promoter γ-satellite
1st 2nd 1st 2nd 1st 2nd 1st 2nd
A: No antibody 26.9 26.95 27.46 27.15 29.45 30.58 8.2 8.17
B: Antibody of H3 K9 acetylated 27.61 27.06 24.14 23.59 26.47 26.68 8.81 8.57
A – B -0.71 -0.11 3.32 3.56 2.98 3.9 -0.61 -0.4
Conclusion
This study strongly suggests that histone H3 K9 acetylation plays a role in the expression of histone genes.
Methods
Cell cycle synchronization
The cell cycle of mouse Hepa 1–6 cells was synchronized at the end of G1-phase by the addition of thymidine-hydroxyurea. The cell cycle arrest was released by washing out the thymidine-hydroxyurea, then the cells were harvested at intervals of 1 h from 0 to 13 h.
RNA extraction
Total RNA was extracted by using the RNeasy mini kit (Qiagen) according to the instructions in the manual for the cell line. After that, each sample was treated with DNase I.
cDNA synthesis
RNA (approximately 0.5 μg) and random hexamer primers were heated to 70°C for 10 min, followed by cooling on ice for 5 min. The cDNA was synthesized in Superscript III First Strand buffer (Invitrogen) according to the manual. The reverse transcriptase was inactivated by a 15-min incubation at 70°C.
Quantitative PCR
The primers used in this analysis are shown in Table 2. Quantification of GAPDH (glyceraldehyde-3-phosphate dehydrogenase) mRNA (primers 5'-TGTGTCCGTCGTGGATCTGA-3' and 5'-CCTGCTTCACCACCTTCTTGA-3' ; product size 76 bp) was used as a control for data normalization. PCR amplification was performed on an ABI PRISM 7700 Sequence Detection System (Applied Biosystems). The PCR conditions were an initial step of 30 s at 95°C, followed by 40 cycles of 5 s at 95°C and 30 s at 60°C. The SYBR premix Ex Taq (Takara) was used according to the manual. Each amplification curve was checked [16]. Expression was assessed by evaluating threshold cycle (CT) values. The relative amount of expressed RNA was calculated by using Livak and Schmittgen's method [17]. The qRT-PCR analyses were performed twice. In each analysis, we adjusted the H2afz expression level to 1 at 0 h.
Table 2 Primers used in this analysis.
Transcript Sequences (5' to 3'), forward and reverse Product size (bp)
Hist1h2aa cggcagtgctagaatacttgaca, gcaggtggcgaggagta 96
Hist1h2ab gcctgcagttccccgta, atctcggccgtcaggtactc 121
Hist1h2ac ggctgctccgcaagggt, cttgttgagctcctcgtcgtt 191
Hist1h2ad/2ao tggacgcggcaagcagggt, agcacggccgccaggtag 162
Hist1h2ae accggctgctccgcaaa, tgatgcgcgtcttcttgttgt 144
Hist1h2af cgaggagctcaacaagctgt, ttgggcttatggtggctct 111
Hist1h2ag tggacgcggcaaacagggc, cagcacggccgccaggtaga 162
Hist1h2ah atatgtctggacgcggt, acgcgctccgagtagttg 133
Hist1h2ai/2aj tcgcgccaaggccaagact, cccacgcgctccgagtagtt 102
Hist1h2ak tacctggcagccgtgcta, cagcttgttgagctcctcgtc 141
Hist1h2an gaggagctcaacaagctgct, ggtggctctcggtcttcttc 100
Hist2h2aa1/2aa2 aactgtagcccggcccg, ttcgtctgtttgcgcttt 100
Hist2h2ab/2ac ggcaaagtgacgatcgca, gtggctctcggtcttcttgg 78
Hist3h2a agcagggcggcaaagctcgt, ttacccttacggagaaggcg 100
H2afx aaggccaagtcgcgctctt, tcggcgtagtggcctttc 86
H2afz actccggaaaggccaagaca, gttgtcctagatttcaggtg 100
Chromatin immunoprecipitation
A total of 2 × 107 cells were cross-linked with 1% formaldehyde for 10 min at room temperature. First, genomic DNA was cut by micrococcal nuclease. Then it was cut by sonication. The precleared extract was divided into 2 equal portions. One was used for control lacking antibody, and the other was incubated with acetylated histone H3 K9 antibody (Upstate Biotechnology). Following immunoprecipitation, beads were washed in low salt, then high salt, then LiCl, then TE buffers. The qPCR analyses were performed two times. Primers used in quantitative PCR were the Hist1h2aa promoter (5'-TTATAGGCGTGGACATT-3' and 5'-CACAGCTTGAATTCCCC-3'), the Hist3h2a promoter (5'-CCGCGTTCTTTTCTGAT-3' and 5'-AATTCGTAAGCGCCAGC-3'), and the H2afz promoter (5'-GCGCCAATCATCGCTCG-3' and 5'-TCGGGACGCGTCCTTGA-3'). We used γ-satellite as a constitutive heterochromatin. The γ-satellite PCR primers have been reported [18].
Authors' contributions
HN designed this study and carried out the molecular biological studies. TS and HO carried out the ChIP experiment and qPCR. YT carried out synchronization of cells. YH helped design the study.
Acknowledgements
This work was supported in part by a Research Grant for the RIKEN Genome Exploration Research Project, a Research Grant for Advanced and Innovational Research Programs in Life Sciences, and a Research Grant for the Genome Network Project from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of the Japanese Government; in part by a Research Grant for CREST of Japan Science and Technology Corporation to Y.H; and in part by grant 15770055 from the MEXT to H.N.
==== Refs
Luger K Mäder AW Richmond RK Sargent DF Richmond TJ Crystal structure of the nucleosome core particle at 2.8 Å resolution Nature 1997 389 251 260 9305837 10.1038/38444
Marzluff WF Duronio RJ Histone mRNA expression: multiple levels of cell cycle regulation and important developmental consequences Curr Opin Cell Biol 2002 14 692 699 12473341 10.1016/S0955-0674(02)00387-3
Rice JC Allis CD Histone methylation versus histone acetylation: new insights into epigenetic regulation Curr Opin Cell Biol 2001 13 263 273 11343896 10.1016/S0955-0674(00)00208-8
Sims RJ IIIMandal SS Reinberg D Recent highlights of RNA-polymerase-II-mediated transcription Curr Opin Cell Biol 2004 16 263 271 15145350 10.1016/j.ceb.2004.04.004
Ehrenhofer-Murray AE Chromatin dynamics at DNA replication, transcription and repair Eur J Biochem 2004 271 2335 2349 15182349 10.1111/j.1432-1033.2004.04162.x
Meneghini MD Wu M Madhani HD Conserved histone variant H2A.Z protects euchromatin from the ectopic spread of silent heterochromatin Cell 2003 112 725 736 12628191 10.1016/S0092-8674(03)00123-5
Roh T-Y Cuddapah S Zhao K Active chromatin domains are defined by acetylation islands revealed by genome-wide mapping Genes Dev 2005 19 542 552 15706033 10.1101/gad.1272505
Marzluff WF Gongidi P Woods KR Jin J Maltais LJ The human and mouse replication-dependent histone genes Genomics 2002 80 487 498 12408966 10.1016/S0888-7543(02)96850-3
Nishida H Suzuki T Tomaru Y Hayashizaki Y A novel replication-independent histone H2a gene in mouse BMC Genetics 2005 6 10 15720718 10.1186/1471-2156-6-10
Hatch CL Bonner WM The human histone H2A.Z gene. Sequence and regulation J Biol Chem 1990 265 15211 15218 1697587
Mannironi C Bonner WM Hatch CL H2A.X. a histone isoprotein with a conserved C-terminal sequence, is encoded by a novel mRNA with both DNA replication type and polyA 3' processing signals Nucleic Acids Res 1989 17 9113 9126 2587254
Harris ME Böhni R Schneiderman MH Ramamurthy L Schümperli D Marzluff WF Regulation of histone mRNA in the unperturbed cell cycle: evidence suggesting control at two posttranscriptional steps Mol Cell Biol 1991 11 2416 2424 2017161
Osley MA The regulation of histone synthesis in the cell cycle Annu Rev Biochem 1991 60 827 861 1883210 10.1146/annurev.bi.60.070191.004143
Oswald F Dobner T Lipp M The E2F transcription factor activates a replication-dependent human H2A gene in early S phase of the cell cycle Mol Cell Biol 1996 16 1889 1895 8628255
Crechowski T Bari RP Stitt M Scheible WR Udvardi MK Real-time RT-PCR profiling of over 1400 Arabidopsis transcription factors: unprecedented sensitivity reveals novel root- and shoot-specific genes Plant J 2004 38 366 379 15078338 10.1111/j.1365-313X.2004.02051.x
Nishida H Tomaru Y Oho Y Hayashizaki Y Naturally occurring antisense RNA of histone H2a in mouse cultured cell lines BMC Genetics 2005 6 23 15892893 10.1186/1471-2156-6-23
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
Shestakova EA Mansuroglu Z Mokrani H Ghinea N Bonnefoy E Transcription factor YY1 associates with pericentromeric γ-satellite DNA in cycling but not in quiescent (G0) cells Nucleic Acids Res 2004 32 4390 4399 15316102 10.1093/nar/gkh737
|
16098230
|
PMC1198228
|
CC BY
|
2021-01-04 16:32:48
|
no
|
BMC Genomics. 2005 Aug 13; 6:108
|
utf-8
|
BMC Genomics
| 2,005 |
10.1186/1471-2164-6-108
|
oa_comm
|
==== Front
BMC MedBMC Medicine1741-7015BioMed Central London 1741-7015-3-141611531110.1186/1741-7015-3-14Research ArticleSuicide attempts in clinical trials with paroxetine randomised against placebo Aursnes Ivar [email protected] Ingunn Fride [email protected] Jorund [email protected] Bent [email protected] Department of Pharmacotherapeutics, University of Oslo, Oslo, Norway2 Department of Mathematics, University of Oslo, Oslo, Norway2005 22 8 2005 3 14 14 8 3 2005 22 8 2005 Copyright © 2005 Aursnes et al; licensee BioMed Central Ltd.2005Aursnes 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
Inclusion of unpublished data on the effects of antidepressants on children has suggested unfavourable risk-benefit profiles for some of the drugs. Recent meta-analyses of studies on adults have indicated similar effects. We obtained unpublished data for paroxetine that have so far not been included in these analyses.
Methods
The documentation for drug registration contained 16 studies in which paroxetine had been randomised against placebo. We registered the number of suicides, suicide attempts and ideation. We corrected for duration of medication and placebo treatment and used a standard Bayesian statistical approach with varying priors.
Results
There were 7 suicide attempts in patients on the drug and 1 in a patient on placebo. We found that the probability of increased intensity of suicide attempts per year in adults taking paroxetine was 0.90 with a "pessimistic" prior, and somewhat less with two more neutral priors.
Conclusion
Our findings support the results of recent meta-analyses. Patients and doctors should be warned that the increased suicidal activity observed in children and adolescents taking certain antidepressant drugs may also be present in adults.
==== Body
Background
The debate about whether the use of antidepressant drugs increases suicidal activity has recently been sharpened after more than 10 years of turmoil [1]. Conclusions concerning children and adolescents have been drawn [2]. Inclusion of unpublished data suggested unfavourable risk-benefit profiles for some of the drugs. For adults, industry points to the absence of data refuting the null hypothesis (no such increase).
In a February 19th BMJ editorial [3] accompanying two meta-analyses of suicidal activities in adult patients on SSRIs [4,5], the authors failed to convey the unanimous conclusion in the reviewed studies of an increased risk of suicidal attempts. Admittedly, one of the analyses only touched statistical significance, but that might have been due to the withholding of data by the manufacturer of one of the drugs. We have had access to some of those missing data. Recently we were given the opportunity to review the clinical data on paroxetine as presented to the world's drug regulatory agencies in 1989. One of the published meta-analyses [4] contained summaries of the documentation provided by the marketing authorization holders to the MHRA, which did not distinguish between suicidal attempt and suicidal ideation for paroxetine. We studied the primary data and even the individual case descriptions when available. Another meta-analysis [5] reported published data only, whereas most of our data were from unpublished studies. Moreover, as opposed to the BMJ authors, we have based our statistical analysis on comparing intensities of suicide attempts per year in drug and placebo groups, taking the exposure time of the patients properly into account. We now present our findings and estimate the degree of support for the idea of an increased intensity per year of suicide attempts in adults.
Methods
We included only double blind, parallel design studies with patients (all adults) randomised to either paroxetine or placebo. Altogether 16 studies met these criteria (references 79 to 93 and 95 in the Expert Report), containing respectively 916 and 550 paroxetine and placebo treated patients. The study period was in most instances 6 weeks. One important exception was a study (reference 91) with a preponderance of paroxetine use over placebo and lasting for 17 weeks. Patients were excluded from the studies after a suicide-related event. Taking this censoring into account, paroxetine treatment made up 190.7 patient years altogether and placebo 73.3 patient years. Suicide-related events could be found in tables in the Expert Report, in the adverse reactions section in the individual study reports, and in the individual patient descriptions.
We let θp be the intensity per year of a suicide attempt in the placebo group and θd the intensity per year in the drug group, for a random patient in the 16 studies; correspondingly, Xp and Xd represent the total numbers of suicide attempts. We can have at most one suicide attempt for each patient. Taking this censoring into account, we denoted the corresponding patient years in the 16 studies combined by mp and md. In addition, patients in both the placebo and drug groups are supposed to behave in a similar manner. It then follows that the likelihood of the experiment corresponds to Xp and Xd having Poisson distributions respectively with parameters (mpθp) and (md θd). In addition, we assume that the two variables were conditionally independent given the parameters. The corresponding observed data are (xp, mp) and (xd, md), and the prior information is denoted by (xop, mop) and (xod, mod).
The Bayesian approach is based on the construction of probability distributions for θp and θd . This does not mean that these parameters are to be interpreted as random variables, but our knowledge of the parameters is uncertain and we describe this uncertainty with the help of probability distributions. Probability distributions describing our initial uncertainty are called prior distributions (that is, before real data are collected). When the real data are taken into account, the prior distributions are updated by Bayes' formula to posterior distributions. An excellent introduction to Bayesian methods in medicine is given by Spiegelhalter et al. [6].
We assume that the prior distribution for θp is gamma, with parameters xop and mop, while correspondingly θd has the parameters xod and mod and is assumed to be independent of the prior distribution for θp. Hence, standard Bayesian theory gives the posterior distribution of θp as gamma, with parameters xop + xp and mop + mp, while θd will have the parameters xod + xd and mod + md. We performed simulations by making 80000 random draws of θd and θp from their independent gamma posterior distributions, computed the logarithms of the ratios θd/θp, and constructed diagrams by applying a standard density estimation technique to these logarithms. (The logarithm was introduced to avoid an unwelcome feature of the density estimation method.) Note that the logarithm of the ratio θd/θp is greater than zero whenever θd is greater than θp. Hence, we calculated the probabilities that medication with paroxetine is associated with an increased intensity of a suicide attempt per year as the proportions of logarithmic ratios greater than zero in the samples. This corresponds to areas below the densities to the right of zero in the diagrams.
The grounds for a pessimistic prior have been given by Healy and Whitaker [7] who, relating the occurrence of suicidal activities to the use of antidepressant drugs, estimated an odds ratio of 2.4 from evidence given in clinical trials, epidemiological observations and case histories. The clinical trial data they used included, but were not restricted to, studies with the active drugs randomised against placebo. Mathematically, we chose to express this view as equivalent to observing two (xod) events with paroxetine during 50 (mod) patient years and one (xop) with placebo during 50 (mop) patient years, adding up to 3 attempts per 100 patient years, which is similar to our observed average value for paroxetine and placebo taken together. We based the calculations on a total of only 100 (mod + mop) patient years in the prior, compared to 264 (md + mp) patient years in the real data, in order to increase the importance of the real data over the prior information. The slightly optimistic and slightly pessimistic priors represent respectively a paper by Lapierre [8] (appearing in tandem with Healy and Whitaker) and the article that reported suicidal ideation in children medicated with paroxetine [2]. The former author took the attitude that, if anything, there were slight signs of reduced suicidal activity connected with antidepressants, whereas the latter authors left the reader with the assumption that the observed increased suicidal ideation in children must somehow be reflected in adults. We assigned the numbers of suicidal patients on paroxetine and placebo per 50 patient years to be respectively 1.35 and 1.65 and vice versa.
Results
There were no suicides in the 16 studies with paroxetine randomised against placebo. Suicidal activities are listed in table 1. Summarising the suicide attempts, there are seven among the patients on paroxetine and one among the patients on placebo. (One event tabulated in the Expert Report as occurring with placebo did in fact happen during the run-in period before randomisation.)
Table 1 Suicide attempts and ideation in randomised clinical trials with paroxetine against placebo. Extracted from "APPLICATION FOR MARKETING AUTHORIZATION: SEROXAT" 1989. * Referring to list in Part I, Volume 3
Patient identification number Study reference number* Suicidal attempt Suicidal ideation Medication Tabulated Individually described
02 01 009 79 X Placebo Yes Yes
02 04 089 82 X Paroxetine Yes Yes
03 002 034 84 X Placebo No Yes
04 02 056 84 X Paroxetine Yes Yes
1 09 021 90 X Washout Yes Yes
09 01A 005 91 X Paroxetine Yes Yes
09-01A-006 91 X Paroxetine No Yes
09 01E 260 91 X Paroxetine Yes No
09 01J 573 91 X Paroxetine Yes Yes
09-OU-620 91 X Paroxetine No Yes
09-01G-405 91 X Paroxetine No Yes
037 93 X Paroxetine No Yes
07 01A 001 95 X Paroxetine Yes Yes
The three prior distributions of the logarithm of the ratio θd/θp are shown in figure 1, and the corresponding posterior distributions are shown in figure 2. The probability that medication with paroxetine is associated with an increased intensity per year of a suicide attempt is 0.90 with the pessimistic prior (Healy and Whitaker [7]), and 0.79 (Lapierre [8]) and 0.85 (Whittington et al. [2]) with the two other priors. The corresponding prior probabilities were respectively 0.75, 0.42 and 0.58.
Figure 1 Prior intensity of suicidal attempt. Distributions of three different prior (see text) logarithmic intensity ratios ln(θd/θp) (logarithmic intensity of a suicide attempt on drug minus logarithmic intensity of a suicide attempt on placebo).
Figure 2 Posterior intensity of suicidal attempt. Distributions of posterior logarithmic intensity ratios ln(θd/θp) with three different priors.
Discussion
We believe that the chosen studies are similar enough to be pooled for analysis. This view is supported by the similarities of the protocols for the various studies, although the populations that were studied differed considerably. We also believe it is best to count patient years rather than patients, although claims have been made for the contrary [7]. At least, counting patient years is a more conservative approach. Furthermore, to treat the patient as a unit and to assume binomial distributions would be inappropriate since patients have different follow-up times and hence different probabilities of suicide attempts within both the placebo and the drug groups.
Another statistical approach would have been to express the prior distributions in terms of independent priors for θp and the ratio θd/θp, with a common, relatively weak prior for θp in all three formulations. This was our approach in a previous publication using basically the same method [9]. To make things simpler and perhaps more transparent in this short paper, we have not used this approach here.
Conclusion
Although we report a small data set, by taking various priors into account the data strongly suggest that the use of SSRIs is connected with an increased intensity of suicide attempts per year. The two meta-analyses and our contribution taken together make a strong case for the conclusion, at least with a short time perspective, that adults taking antidepressants have an increased risk of suicide attempts. We also conclude that the recommendation of restrictions on the use of paroxetine for children and adolescents recently conveyed by regulatory agencies [10] should be extended to include usage by adults.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
IA collected the data, presented the problem and drafted the manuscript along with BN, who suggested the statistical solution based on earlier work by the present authors [9]; IFT did the computations and took part in the planning along with JG. All authors read and approved the final manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
Journalist Ane Hoyem, the Norwegian Broadcasting Corporation (NRK), encouraged this investigation while researching for the medical information program "Puls".
==== Refs
Breggin PR Suicidality, violence and mania caused by selective serotonin reuptake inhibitors (SSRI): A review and analysis International Journal of Risk and Safety in Medicine 2004 16 31 49
Whittington CJ Kendall T Fonagy O Cottrell D Cotgrove A Boddington E Selective serotonin reuptake inhibitors in childhood depression: systematic review of published versus unpublished data Lancet 2004 363 1341 1345 15110490 10.1016/S0140-6736(04)16043-1
Cipriani A Barbui C Geddes JR Suicide, depression, and antidepressants. Patients and clinicians need to balance benefits and harms BMJ 2005 330 373 374 15718515 10.1136/bmj.330.7488.373
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 330 385 388 15718537 10.1136/bmj.330.7488.385
Fergusson D Doucette S Glass KC Shapiro S Healy D Hebert P Hutton B Association between suicide attempts and selective serotonin reuptake inhibitors: systematic review of randomised controlled trials BMJ 2005 330 396 399 15718539 10.1136/bmj.330.7488.396
Spiegelhalter DJ Myles JP Jones DR Abrams KR An introduction to Bayesian methods in health technology assessment BMJ 1999 319 508 512 10454409
Healy D Whitaker C Antidepressants and suicide: risk-benefit conundrums J Psychiatry Neurosc 2003 28 331 337 14517576
Lapierre YD Suicidality with selective serotonin reuptake inhibitors: Valid claim? J Psychiatry Neurosci 2003 28 340 347 14517577
Aursnes I Tvete IF Gaasemyr J Natvig B Clinical efficacies of antihypertensive drugs Scand Cardiovasc J 2003 37 72 79 12775305 10.1080/14017430310001186
FDA issues public health advisory on cautions for use of antidepressants in adults and children
|
16115311
|
PMC1198229
|
CC BY
|
2021-01-04 16:27:56
|
no
|
BMC Med. 2005 Aug 22; 3:14
|
utf-8
|
BMC Med
| 2,005 |
10.1186/1741-7015-3-14
|
oa_comm
|
==== Front
BMC MicrobiolBMC Microbiology1471-2180BioMed Central London 1471-2180-5-481610916410.1186/1471-2180-5-48Research ArticleHerpes simplex virus interferes with amyloid precursor protein processing Shipley Suzanne J [email protected] Edward T [email protected] Ruth F [email protected] Curtis B [email protected] Faculty of Life Sciences, Moffat Building, University of Manchester, Manchester, M60 1QD, UK2 School of Biochemistry and Molecular Biology, University of Leeds, Leeds, W. Yorks, LS2 9JT, UK2005 18 8 2005 5 48 48 9 5 2005 18 8 2005 Copyright © 2005 Shipley 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 early events underlying Alzheimer's disease (AD) remain uncertain, although environmental factors may be involved. Work in this laboratory has shown that the combination of herpes simplex virus type 1 (HSV1) in brain and carriage of the APOE-ε4 allele of the APOE gene strongly increases the risk of developing AD. The development of AD is thought to involve abnormal aggregation or deposition of a 39–43 amino acid protein – β amyloid (Aβ) – within the brain. This is cleaved from the much larger transmembranal protein 'amyloid precursor protein' (APP). Any agent able to interfere directly with Aβ or APP metabolism may therefore have the capacity to contribute towards AD. One recent report showed that certain HSV1 glycoprotein peptides may aggregate like Aβ; a second study described a role for APP in transport of virus in squid axons. However to date the effects of acute herpesvirus infection on metabolism of APP in human neuronal-type cells have not been investigated. In order to find if HSV1 directly affects APP and its degradation, we have examined this protein from human neuroblastoma cells (normal and transfected with APP 695) infected with the virus, using Western blotting.
Results
We have found that acute HSV1 (and also HSV2) infection rapidly reduces full length APP levels – as might be expected – yet surprisingly markedly increases levels of a novel C-terminal fragment of APP of about 55 kDa. This band was not increased in cells treated with the protein synthesis inhibitor cycloheximide
Conclusion
Herpes virus infection leads to rapid loss of full length APP from cells, yet also causes increased levels of a novel 55 kDa C-terminal APP fragment. These data suggest that infection can directly alter the processing of a transmembranal protein intimately linked to the aetiology of AD.
==== Body
Background
The key events which initiate Alzheimer's disease (AD) remain unclear, though environmental factors have been shown to be involved [1]. The two main neuropathological features of AD – senile plaques (SP) and neurofibrillary tangles – occur also in the normal elderly. AD may be triggered when the numbers of these features increase to abnormal levels. Although there is much evidence supporting the involvement of one or both of these structures in the disease process, it is unclear whether they are involved directly, what processes underlie their formation, and why their numbers rise during the development of AD. However, any environmental stimulus capable of leading to production of the abnormal proteins which make up these structures might thereby contribute to the occurrence of AD.
β-amyloid protein (Aβ) is the major proteinaceous component of senile plaques, and the abnormal deposition of aggregates of this protein is thought to give rise to SP formation. Aβ comprises a sequence of 39–43 amino acids and is formed by cleavage from the much longer amyloid precursor protein (APP), first by β-secretase then by γ-secretase. Generation of Aβ may lead eventually to the development of neurotoxic SPs, which themselves cause further tissue damage and SP generation, or alternatively to the production of small neurotoxic Aβ assemblies, which may be the most damaging form of Aβ [2]. Whatever the mechanism leading to neuronal damage after Aβ generation, any agent which interferes with amyloid systems may thereby contribute to the development of AD.
Studies of the possible role of pathogens in AD were made possible in the early 1990s by the development of PCR, allowing detection of low levels of viral or bacterial DNA within human brain tissues. However very few investigators have undertaken such studies, and most work has focused on herpes simplex virus type 1 (HSV1). A causal role for this virus in triggering AD was suggested by Itzhaki et al., who found that the risk of developing AD associated with carriage of an APOE-ε4 allele depends on the presence of latent HSV1 in the brain [3,4]. Coupled with the finding here of a higher APOE-ε4 allele frequency amongst individuals who suffer damage after HSV1 reactivation in the periphery – seen as cold sores – this indirectly supports a role for HSV1, perhaps acting with other pathogens [5], as a key environmental contributor to AD [3,4].
Recent epidemiological studies support the involvement of a pathogen in AD: cognitive function in AD patients declines for at least 2 months after a systemic infection [6]; cognitive decline in elderly cardiovascular patients correlates with viral burden [7]. These could be explained by systemic infection causing brain inflammation which, in turn, leads to reactivation of latent HSV1 – and consequent further damage in the CNS. Consistently, in middle-aged controls, very few of whom would harbour HSV1 in brain [8], no cognitive decline occurs after systemic infection [9]. Enhancement of damage in brain, due to the presence of an infectious agent, is supported by the finding that inflammation in brain caused by lipopolysaccharide is augmented in pre-clinical prion-infected mice [10].
One mechanism by which HSV1 might contribute to AD was suggested by the detection of a sequence homology between Aβ and the HSV1 glycoprotein B (gB), a viral coat protein which is involved in attachment of the virus to cells, and by the finding that synthetic peptides derived from gB can give rise to Aβ-like aggregates. The authors suggested that such viral proteins might act as a seed for amyloid plaque formation [11]. Another way in which infection might affect amyloid biochemistry is suggested by studies demonstrating that Aβ or APP are upregulated in response to a wide range of injurious stimuli, including head injury [12], stroke [13] or HIV infection [14]. Increased expression of APP – though not of the related protein amyloid precursor-like protein 2 (APLP2) – has been reported in cutaneous wound repair [15]. APP foci colocalise with sites of opportunistic infection in HIV dementia patients, including sites of herpes virus (cytomegalovirus) infection [16]. Another type of APP-HSV1 interaction has been demonstrated in axonal transport: APP was found to be present in HSV1 particles and it was suggested that this could lead to alterations in location and processing of APP at the nerve terminal [17] causing synaptic and neuronal dysfunction.
No studies have been made on HSV1 effects on APP or Aβ in cells in culture. Here we describe the first to investigate whether levels of APP and its metabolites are affected by HSV1 infection, using human neuronal-type cells in culture.
Results and discussion
We carried out initial experiments on SHSY5Y cells to determine the level of HSV1 needed to ensure that on inoculation most cells were infected. We then examined levels of APP in lysates prepared from the cells 6 h after infection with this dose of HSV1. Western blots stained with anti C-terminal APP antibody (Fig. 1A) revealed, as expected, various full-length APP bands in uninfected cells corresponding to the three main isoforms in multiple glycosylation states. Deglycosylation experiments have shown that the lower single band (100 kDa) is immature APP695, that the doublet above this (approx molecular weight 115 kDa) is mature APP695 and immature APP751/APP770 and that the highest molecular weight group (approx. 125 to 135 kDa) comprises a mixture of mature APP751 and mature APP770 (Parkin et al., unpublished observations).
The intensity of these bands decreased appreciably in cells harvested only 3 h after inoculation with HSV1, as did those from cells treated with the protein synthesis inhibitor cycloheximide. This decline after infection may reflect loss of ability of the cells to generate new APP due simply to virus-induced shut-down of protein synthesis [18].
Of particular interest was the relative intensity of a 55 kDa band, again apparent after C-terminal APP antibody staining. This band was similarly intense in control-treated and cycloheximide-treated cells, but had far greater intensity after HSV1 infection. This increase in HSV1-infected cells alone was highly reproducible, occurring in many experiments, each carried out independently, from the infection to the blotting stage. Band intensity for cells treated for 6 hours was quantified using Syngene Genetools software and averaged over five such experiments is shown in Figure 1B; value for HSV1-infected cells was significantly different from control, being elevated by 124% (95%CI 101 – 148%; p < 0.002), whereas that for cycloheximide-treated cells was similar to the control value.
To exclude the possibility that the band reflected synthesis of a viral protein which cross-reacts non-specifically with the antibody, we immunostained blots produced from gels of viral proteins from our HSV1 preparations. No evidence for any cross-reaction of the Sigma C-terminal APP antibody with these proteins could be found, suggesting that the strengthened band was cell-derived (results not shown). This result does not preclude any cross-reactivity with an ICP, as the latter would not be present within virions, but the presence of the 55 kDa protein in uninfected cells, and also, at a higher level, in APP-transfected cells (see below), strongly supports a cellular, non-viral origin of the protein. Interestingly, APP has been detected in HSV1 virions by Western blotting [17], but this probably reflects the authors' usage of a much more concentrated and purified virus preparation than that of the inoculate we use for infecting cells.
We next examined the influence of duration of infection on these viral-induced changes (Fig. 1C). As anticipated, with increasing duration before harvesting, the depletion of APP became more pronounced (this was found also with cycloheximide treatment); in contrast, the intensity of the C-terminal 55 kDa fragment seen in the infected cells was not higher than in uninfected cells at 3 hr, but increased at 6 hr and 9 hr. However there was minimal further change as incubation increased to 24 hr (not shown). This result also was found to be reproducible on repeating the experiment several times from infection to blotting. It suggests that infection prevents synthesis of APP, leading to its gradual depletion, but causes also the abnormal accumulation of the 55 kDa fragment due, presumably, to a decrease in degradation of the latter. However, for longer incubation times, we cannot exclude the possibility of sequestration of some APP within newly synthesised viral particles, as reported elsewhere [17], which might reduce the amount of APP in the virus-free cell lysates used for the preparation of cell proteins.
We repeated these experiments with herpes simplex virus type 2 (HSV2), to see whether these effects were specific to HSV1. Fig. 2 shows that very similar findings were obtained, with 55 kDa band intensity for HSV2-infected cells 9 hr after infection being 101% greater than that for control cells (95%CI 94 – 107%; p < 0.005) suggesting that this phenomenon is not uniquely associated with HSV1 infection.
On the assumption that a corresponding N-terminal fragment of approximately similar size to that of the 55 kDa C-terminal fragment would be produced as a result of this abnormal processing (the size of APP being 110 kDa to 135 kDa), we probed our blots with the Sigma N-terminal antibody (Fig 3A). However we found no clear evidence for an increased amount of an N-terminal fragment of around this size after infection. Possibly, such a fragment would be either degraded or released into the cell culture medium, thus precluding its detection here.
To confirm further that the C-terminal fragment is derived from APP (and not for example from an APLP, i.e. of a type of protein related to APP, though with a less clear involvement in AD) we obtained human SYSY5Y cells that had been transfected with a human APP gene (APP695), and which over-express APP. We stained blots of proteins from non-transfected, APP transfected, and mock-transfected (i.e. with vector alone) cells with the C-terminal antibody, and found the 55 kDa band in all three cases, though its intensity was greater in the APP transfected cells alone (see Fig. 3B). This indicates that the 55 kDa fragment is indeed an APP product.
Conclusion
Infection of neuronal cells by HSV1 is known to cause rapid shutoff of protein synthesis prior to viral replication [19]. This is mediated by the virion host shut-off (vhs) protein, which destabilises cellular mRNAs causing them to be degraded. Consequently, and unsurprisingly, transcription of the gene(s) for APP would cease, as would its synthesis and that of other proteins. Our finding that levels of full length APP decline rapidly in infected cells and in cells treated with cycloheximide, suggest that in both cases turnover of full length APP is rapid.
The surprising increase in amount of the 55 kDa C-terminal APP fragment in infected cells shows that processing of APP as well as its synthesis is affected by HSV1, that this occurs within 6 h of infection, and that the level increases with time after infection up to at least 9 h. The size of this fragment and its staining with C-terminal antibody indicate that it includes the region of APP which contains Aβ.
The possibility that the 55 kDa band derives from an APLP or merely reflects a non-specific reaction of the APP antibody with a viral protein or ICP is unlikely. Our viral preparations did not cross-react with this antibody, and at the time when this band begins to increase in intensity (6 h) the majority of ICPs would have already appeared. Furthermore the 55 kDa band is present (at lower levels) in cells not infected with virus. Also, cells transfected with APP (though not mock-transfected cells) have a higher level of this 55 kDa fragment than do untransfected uninfected cells.
The fact that the mock-transfected and cycloheximide-treated cells do not show any intensification of the 55 kDa band suggests that this phenomenon is not due merely to a non-specific stress response, but occurs in response to infection, although the increase in the fragment after HSV2 infection shows the effect is not specific for HSV1 infection alone. Interestingly, infection of human brain microvascular endothelial cells with Chlamydia pneumonia (Cpn) may increase the intensity of a similar 55 kDa C-terminal APP band (personal communication, Professor Brian Balin, PA, USA). Thus, increased production of this fragment may be a general response to infection.
Currently HSV1 is the only virus that has been shown to be present in brain of most elderly humans [20] – and may be present as a whole functional genome [21]. Active infection of brain cells with other agents might lead to the same APP effects, but until or unless they are shown to be present, they can not be proposed as possible factors in AD. The situation regarding presence or absence of the bacterium Cpn in brain appears unresolved, although a direct role for Cpn in amyloid systems is supported by recent studies in mice showing amyloid deposition in animals intracerebrally infected with Cpn [22]. Confirmation of Cpn presence would support the possibility that, as with HSV1, the increased level of the 55 kDa fragment that it causes might contribute to AD.
The low levels of Aβ secreted by SHSY5Y cells preclude our examining at present the effects of HSV1 infection on levels of Aβ in cell culture models. Eventual identification of those components of Aβ protein systems which are altered by active viral infection will clarify whether pathogens such as HSV1 can contribute directly to the development of AD neuropathology.
We report here that levels of full length APP rapidly decline in human neuronal type cells acutely infected with either HSV1 or HSV2. Also, the amount of a C-terminal 55 kDa APP fragment which contains the Aβ sequence appears to increase rapidly in infected cells. The fragment level is greater in (non-virally infected) SHSY5Y cells transfected with APP695, suggesting that it is almost certainly derived from APP rather than from APLP. The fragment may increase as part of a host defence mechanism, and/or it might lead to increased generation of Aβ. We are now investigating whether the latter possibility is correct.
Methods
Cell culture
Human neuroblastoma (SHSY5Y) cells were maintained in Eagle's minimum essential medium (EMEM) supplemented with 10% (v/v) foetal bovine serum (FBS), 2% (v/v) glutamine and 1% (v/v) penicillin/streptomycin (10% medium), hereafter referred to as growth medium. Cells were incubated at 37°C in a humidified atmosphere of 5% carbon dioxide.
SHSY5Y neuroblastoma cells over-expressing APP695 were prepared by double blunt end ligation of the human APP695 sequence into the BstXI site of pIREShyg (BD Biosciences Clontech, California, USA). For stable transfections 30 μg of DNA was introduced to cells by electroporation and selection was performed in normal growth medium containing 100 μg ml-1 hygromycin B selection antibiotic (Gibco BRL, Paisley, UK). 'Mock-transfected' cells were stably transfected with the empty pIREShyg vector.
HSV1 (strain SC16) stocks were prepared from a primary stock kindly provided by Prof. Roy Jennings (Department of Virology, University of Sheffield), using Vero cells as host. Herpes simplex virus type 2 stocks were prepared from a clinical isolate kindly provided by Prof Anthony Hart (University of Liverpool) as for HSV1, using HEp2a cells. In both cases, confluent cell monolayers were inoculated at high multiplicities of infection (>10 plaque-forming units (pfu) per cell) with virus suspended in growth medium containing only 1% FBS. Once cytopathic end point was reached (after about two days) virus was harvested from medium and from cells disrupted by low power sonication. Cellular debris was removed by low speed centrifugation (1000 g, 10 min), and virus was isolated by high speed centrifugation of the supernatant (10000 g, 2 h, 4°C) using a Sorvall SS34 rotor. Virus-containing pellets were suspended in PBS, and stored in aliquots at -85°C; their infectivity was assessed by plaque assay of serial dilutions. The virus preparations were checked for bacterial sterility by inoculation into beef heart agar plates and confirming the absence of bacterial growth after incubation for several days at 37°C. For both viruses three sequential passages were prepared, and only passage 3 stocks used here.
SHSY5Y cells were seeded at a concentration of 8 million cells per flask (T175) and incubated overnight. Prior to infection, growth medium was discarded and cells were then washed briefly 10 ml of PBS at 37°C. HSV1 (or in some experiments HSV2) was introduced in 10 ml of growth medium (containing only 0.5% serum), at 3 pfu/cell. For controls, either 0.5% serum containing growth medium alone or the latter containing cycloheximide at 10 μg/ml was used. After 1 hr incubation, the inoculating medium (or control treatment) was removed, and 10 ml of fresh 0.5% serum containing growth medium was added, followed by further incubation for various times.
Protein extraction and Western blotting
Cells were harvested by removing medium, washing twice with 10 ml PBS, and incubated in 1 mM EDTA (pH 7.4) (in PBS) at room temperature for 10 min. The cell suspension was centrifuged (500 g, 5 min, 4°C), and the cell pellet was resuspended in 400 μl of homogenisation buffer (0.5% Triton X-100 in PBS; 2 mM phenylmethylsulphlfluoride (PMSF); 100 μg/ml of aprotinin and 100 μg/ml leupeptin). Cell lysis was completed by sonication (MSE sonicator, 4°C, 10 μm amplitude, 6 × 10 s).
After measuring the protein concentration of each lysate (BCA protein assay; Pierce), samples were prepared for polyacrylamide gel electrophoresis (PAGE) by mixing 60 μg of protein with 0.25 volume of ×5 Laemmli sample buffer containing 25% β-mercaptoethanol, and boiling for 5 min. Samples were subjected to electrophoresis on 10% SDS-PAGE gels and the proteins transferred to PVDF membranes (Immobilin-P, Millipore) by Western blotting.
Membranes were blocked for 1 h using 8% skimmed-powdered milk in 0.5% Tween in TBS (TBST). The membrane was then washed (this and subsequent washes involved 5 separate 5 min washes in TBS) and then incubated with primary antibody, diluted 1:4000 with TBS (for 1 1/2 h). Primary antibodies used were anti-C-terminal APP (Sigma; A8717), and the anti-N-terminal APP antibody 22C11. After a further wash, membranes were incubated with secondary antibody conjugated with peroxidase (Pierce) which had been diluted 1:1250 with 8% milk in TBS for 1 h. After a final wash the membrane was incubated with Supersignal West Pico Chemiluminescent Substrate Kit (Pierce) for 10 min, and proteins were then visualised by chemiluminescence using a gel documentation system (Syngene).
Authors' contributions
SJS carried out virology experiments and Western blotting studies, and helped to draft the manuscript. ETP prepared the APP-transfected cells, and participated in the study design. RFI participated in the study design and helped to draft the manuscript. CBD conceived of the study, coordinated it, and drafted the manuscript. All authors read and approved the final manuscript.
Acknowledgements
We are grateful to the Humane Research Trust, the Wellcome Trust, the Nuffield Foundation, and the Manchester Alzheimer's Research Trust Network for financial support. We thank also Edward Tsao who carried out some preliminary experiments and Prof Nigel Hooper (University of Leeds) for helpful discussion.
Figures and Tables
Figure 1 Effect of HSV1 infection on APP processing in SHSY5Y cells. (A) SHSY5Y human neuroblastoma cells were either untreated (cont), infected with herpes simplex virus type 1 (HSV1), or treated with the protein synthesis inhibitor cycloheximide (CH), then incubated for a further 6 h. Cell lysates were subjected to Western blot analysis, using an anti-C-terminal APP antibody (Sigma Aldrich A8717) at 1:4000 dilution. Each lane contains cell lysate prepared from a single flask (two flasks were used per treatment). Several full-length APP bands are clearly visible. Arrows indicate the band intensified by HSV1 infection. (B) Quantification of 55 kDa band in control (cont), cycloheximide-treated (CH), or HSV1 infected cells, assessed using Syngene GeneTools software. Values show average band height from five independent experiments for cells treated for 6 h, each involving two or four separately processed flasks. Bars show standard deviation. Treatments resulting in values significantly different from control are marked * (indicates p < 0.002; ANOVA). (C) Effect of time period after inoculation with virus on height of 55 kDa band. SHSY5Y human neuroblastoma cells were either untreated (cont), infected with HSV1, or treated with cycloheximide (CH), then incubated for a further 3, 6 or 9 h. Cell lysates were analysed by Western blotting as in Fig 1A. Arrows indicate the band intensified by HSV1 infection.
Figure 2 Effect of HSV2 infection on APP processing in SHSY5Y cells. SHSY5Y human neuroblastoma cells were either untreated (cont), infected with HSV2, or treated with cycloheximide (CH), then incubated for 9 h. Cell lysates were analysed by Western blotting as in Fig 1A. Arrows indicate the band intensified by HSV2 infection.
Figure 3 Absence of N-terminal fragment and confirmation that 55 kDa fragment is derived from APP. (A) Staining with antibody to N-terminal APP. SHSY5Y human neuroblastoma cells were either untreated (cont) or infected with HSV1, then incubated for a further 6 h. Cell lysates were analysed by Western blotting using an anti-N-terminal APP antibody 22C11. No evidence for increase in an N-terminal fragment of the expected M Wt. was observed, suggesting the latter may be metabolised further or secreted. (B) Strengthening of 55 kDa band in uninfected APP-transfected SHSY5Y cells. Normal SHSY5Y human neuroblastoma cells, mock-transfected SHSY5Y cells (mTf) or APP695 transfected cells (APP Tf) were uninfected prior to harvest. In addition normal SHSY5Y cells which had been infected with HSV1 (lane labelled HSV1) for 6 h were also prepared, to allow position of the HSV1-senstive 55 kDa band to be clearly identified. Cell lysates were subjected to Western blot analysis, and probed with an anti-C-terminal APP antibody (Sigma Aldrich A8717) at 1:4000 dilution. Arrows indicate the position of the band intensified after HSV1 infection. The intensity of full-length APP bands was appreciably greater than those in the non-transfected cells. In addition the intensity of several smaller bands was also greater than the intensity of the same bands in the mock-transfected (mTf) or normal SHSY5Y (cont), including a band at the same position as the band intensified on HSV1 infection of normal SHSY5Y cells (marked by the arrows), suggested this band is genuinely derived from APP.
==== Refs
Raiha I Kaprio J Koskenvuo M Rajala T Sourander L Alzheimer's disease in Finnish twins Lancet 1996 347 573 578 8596319 10.1016/S0140-6736(96)91272-6
Selkoe DJ Deciphering the genesis and fate of amyloid beta-protein yields novel therapies for Alzheimer disease J Clin Invest 2002 110 1375 1381 12438432 10.1172/JCI200216783
Itzhaki RF Lin WR Shang D Wilcock GK Faragher B Jamieson GA Herpes simplex virus type 1 in brain and risk of Alzheimer's disease Lancet 1997 349 241 244 9014911 10.1016/S0140-6736(96)10149-5
Lin WR Graham J MacGowan SM Wilcock GK Itzhaki RF Alzheimer's disease, herpes virus in brain, apolipoprotein E4 and herpes labialis Alzheimer's Reports 1998 1 173 178
Dobson CB Wozniak MA Itzhaki RF Do infectious agents play a role in dementia? Trends Microbiol 2003 11 312 317 12875814 10.1016/S0966-842X(03)00146-X
Holmes C El-Okl M Williams AL Cunningham C Wilcockson D Perry VH Systemic infection, interleukin 1beta, and cognitive decline in Alzheimer's disease J Neurol Neurosurg Psychiatry 2003 74 788 789 12754353 10.1136/jnnp.74.6.788
Strandberg TE Pitkala KH Linnavuori KH Tilvis RS Impact of viral and bacterial burden on cognitive impairment in elderly persons with cardiovascular diseases Stroke 2003 34 2126 2131 12920256 10.1161/01.STR.0000086754.32238.DA
Jamieson GA Maitland NJ Wilcock GK Yates CM Itzhaki RF Herpes simplex virus type 1 DNA is present in specific regions of brain from aged people with and without senile dementia of the Alzheimer type J Pathol 1992 167 365 368 1328575 10.1002/path.1711670403
Dickerson FB Boronow JJ Stallings C Origoni AE Cole S Krivogorsky B Yolken RH Infection with herpes simplex virus type 1 is associated with cognitive deficits in bipolar disorder Biol Psychiatry 2004 55 588 593 15013827 10.1016/j.biopsych.2003.10.008
Combrinck MI Perry VH Cunningham C Peripheral infection evokes exaggerated sickness behaviour in pre-clinical murine prion disease Neuroscience 2002 112 7 11 12044467 10.1016/S0306-4522(02)00030-1
Cribbs DH Azizeh BY Cotman CW LaFerla FM Fibril formation and neurotoxicity by a herpes simplex virus glycoprotein B fragment with homology to the Alzheimer's A beta peptide Biochemistry 2000 39 5988 5994 10821670 10.1021/bi000029f
Roberts GW Gentleman SM Lynch A Graham DI beta A4 amyloid protein deposition in brain after head trauma Lancet 1991 338 1422 1423 1683421 10.1016/0140-6736(91)92724-G
Popa-Wagner A Schroder E Walker LC Kessler C beta-Amyloid precursor protein and ss-amyloid peptide immunoreactivity in the rat brain after middle cerebral artery occlusion: effect of age Stroke 1998 29 2196 2202 9756603
Esiri MM Biddolph SC Morris CS Prevalence of Alzheimer plaques in AIDS J Neurol Neurosurg Psychiatry 1998 65 29 33 9667557
Kummer C Wehner S Quast T Werner S Herzog V Expression and potential function of beta-amyloid precursor proteins during cutaneous wound repair Exp Cell Res 2002 280 222 232 12413888 10.1006/excr.2002.5631
Wiley CA Achim CL Hammond R Love S Masliah E Radhakrishnan L Sanders V Wang G Damage and repair of DNA in HIV encephalitis J Neuropathol Exp Neurol 2000 59 955 965 11089573
Satpute-Krishnan P DeGiorgis JA Bearer EL Fast anterograde transport of herpes simplex virus: role for the amyloid precursor protein of alzheimer's disease Aging Cell 2003 2 305 318 14677633 10.1046/j.1474-9728.2003.00069.x
Matis J Kudelova M Early shutoff of host protein synthesis in cells infected with herpes simplex viruses Acta Virol 2001 45 269 277 12083325
Roizman B Sears A Fields B, Knipe D and Howley P Herpes simplex viruses and their replication Fundamental virology 1995 Philadelphia, Lippincott-Raven 1043 1107
Lin WR Wozniak MA Cooper RJ Wilcock GK Itzhaki RF Herpesviruses in brain and Alzheimer's disease J Pathol 2002 197 395 402 12115887 10.1002/path.1127
Wozniak MA Shipley SJ Combrinck M Wilcock GK Itzhaki RF Productive herpes simplex virus in brain of elderly normal subjects and Alzheimer's disease patients J Med Virol 2005 75 300 306 15602731 10.1002/jmv.20271
Little CS Hammond CJ MacIntyre A Balin BJ Appelt DM Chlamydia pneumoniae induces Alzheimer-like amyloid plaques in brains of BALB/c mice Neurobiol Aging 2004 25 419 429 15013562 10.1016/S0197-4580(03)00127-1
|
16109164
|
PMC1198230
|
CC BY
|
2021-01-04 16:03:40
|
no
|
BMC Microbiol. 2005 Aug 18; 5:48
|
utf-8
|
BMC Microbiol
| 2,005 |
10.1186/1471-2180-5-48
|
oa_comm
|
==== Front
BMC Med Res MethodolBMC Medical Research Methodology1471-2288BioMed Central London 1471-2288-5-221602661210.1186/1471-2288-5-22Research ArticleOblique decision trees for spatial pattern detection: optimal algorithm and application to malaria risk Gaudart Jean [email protected] Belco [email protected] Stéphane [email protected] Ogobara [email protected] Medical Statistics and Informatics Research Team, LIF -UMR 6166 – CNRS/ Aix-Marseille University, Faculty of Medicine, 27 Bd Jean Moulin 13385 Marseille Cedex 05, France2 Immunology and Genetic of Parasitic Diseases, UMR 399 – INSERM/ Aix-Marseille University, Faculty of Medicine, 27 Bd Jean Moulin 13385 Marseille Cedex 05, France3 Malaria Research and Training Centre, Faculty of Medicine, Pharmacy and Odonto-Stomatology, University of Mali, BP 1805, Bamako, Mali2005 18 7 2005 5 22 22 2 3 2005 18 7 2005 Copyright © 2005 Gaudart et al; licensee BioMed Central Ltd.2005Gaudart 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 order to detect potential disease clusters where a putative source cannot be specified, classical procedures scan the geographical area with circular windows through a specified grid imposed to the map. However, the choice of the windows' shapes, sizes and centers is critical and different choices may not provide exactly the same results.
The aim of our work was to use an Oblique Decision Tree model (ODT) which provides potential clusters without pre-specifying shapes, sizes or centers. For this purpose, we have developed an ODT-algorithm to find an oblique partition of the space defined by the geographic coordinates.
Methods
ODT is based on the classification and regression tree (CART). As CART finds out rectangular partitions of the covariate space, ODT provides oblique partitions maximizing the interclass variance of the independent variable. Since it is a NP-Hard problem in RN, classical ODT-algorithms use evolutionary procedures or heuristics. We have developed an optimal ODT-algorithm in R2, based on the directions defined by each couple of point locations. This partition provided potential clusters which can be tested with Monte-Carlo inference.
We applied the ODT-model to a dataset in order to identify potential high risk clusters of malaria in a village in Western Africa during the dry season. The ODT results were compared with those of the Kulldorff' s SaTScan™.
Results
The ODT procedure provided four classes of risk of infection. In the first high risk class 60%, 95% confidence interval (CI95%) [52.22–67.55], of the children was infected. Monte-Carlo inference showed that the spatial pattern issued from the ODT-model was significant (p < 0.0001).
Satscan results yielded one significant cluster where the risk of disease was high with an infectious rate of 54.21%, CI95% [47.51–60.75]. Obviously, his center was located within the first high risk ODT class. Both procedures provided similar results identifying a high risk cluster in the western part of the village where a mosquito breeding point was located.
Conclusion
ODT-models improve the classical scanning procedures by detecting potential disease clusters independently of any specification of the shapes, sizes or centers of the clusters.
==== Body
Background
Since the development of warning systems and environmental hazards awareness, a wide range of statistical methods has been provided to identify disease clusters and spatial patterns. These methods have been classified into three groups [1-3]:
- Tests for focused clustering where the putative source is prespecified [4,5,2];
- Tests for global clustering with statistics using the distance between cases [6-9];
- General tests for localized clusters where the putative source or potential clusters cannot be prespecified [6,10,11].
This paper focuses on the latter tests i.e. on general procedures for the determination of spatial patterns. These patterns allow us to localize disease clusters where the disease rate is particularly high. Since the Openshaw's Geographical Analysis Machine (GAM), numerous works have proposed extensions or modifications of this method. The GAM lays out a regular grid of points covering the region under study. Then it generates overlapping circular windows centered at each grid point with constant radii depending on the grid spacing. The procedure is repeated at different predetermined values of the radius and thus defines potential clusters. Alternative procedures use circular windows centered at the observed point locations [10] and scan the area through this irregular grid. The use of squared shaped windows has also been proposed [6]. A general review of spatial methods is provided by Waller and Gotway [12] as well as in several publications [13-15].
The Kulldorff's scan statistic is one of the most interesting and used methods for cluster analysis [16,1,17]. The scan statistic is a likelihood ratio based method, which Kulldorff [11] defined without any assumptions about the shape, size or collection of locations for the scanning windows. However, various algorithms are necessary to calculate the test statistics for different defined types of scanning windows. Softwares (e.g. SaTScan™ [18]) have been implemented for some of these particular windows/algorithms. The SaTScan™ imposes on the map circular windows positioned on regular (such as GAM) or irregular grid (defined by the observed point locations). For each center point, the radius varies continuously from zero to a pre-specified upper bound. Each of the circular windows, moving through the different centers and with different radii, is a possible candidate for containing a cluster of cases.
It is noteworthy that the detection of potential clusters is enforced on circular shaped (or squared shaped) windows. The various algorithms applied the scan statistic method to windows centered at either grid or observed point locations. These two procedures define different sets of potential clusters and therefore may not provide exactly the same results. Furthermore, changing the windows' shape may also provide different clusters. Gangnon and Clayton introduced a bayesian approach [19] for clustering which does not require cluster's locations or shapes to be specified but which requires some prior specifications of the distribution of various cluster size and shapes (hierarchical priors). However, given the large number of potential models, the posterior distribution cannot be directly provided. Therefore, Gangnon and Clayton limit the number of models under consideration by using a randomized method to build models with high posterior distributions. They approximate the posterior distribution over the limited number of cluster models incorporating hierarchical prior. Patil and Taillie [20] proposed an adaptation of the scan statistics to detect clusters without restricted shape. It reduces the size of the potential cluster set by determining levels of the rates of cases. The potential cluster set consists on all the connected components that have rates higher than a fixed level. Each level determines a potential cluster set. But the determination of levels is data-dependent. Furthermore in a practical point of view, not all of the observed rates can be used as levels in order to avoid providing a computationally impracticable number of potential cluster sets. Other procedures use stochastic optimization algorithm to reduce the number of examined potential clusters [21]. But again these methods used for the determination of potential clusters are not optimal from a classification viewpoint.
The aim of the present work is to provide an optimal partitioning procedure using Oblique Decision Trees in order to detect spatial patterns and to optimize the potential clusters determination without prior specifications. Rather than using a likelihood ratio test, this new approach, which is not a scan statistic, is based on the calculus of the interclass variance during each of many splits of the space before providing the final pattern.
Methods
CART and ODT-models
Tree-based models such as CART (Classification And Regression Trees) [22] are non-linear and non-parametric alternatives to linear models for regression and classification problems (such as linear regression, logistic regression, linear discriminant analysis, linear proportional hazard models). CART models are fitted by binary recursive partitioning of a multidimensional covariate space, in which the dataset is successively split into increasingly homogeneous subsets until a specified criterion is satisfied. For the first partition, CART searches the best possible place to split a continuous variable into two classes and defines two subspaces which maximize overall class separation (i.e. interclass variance of the dependent variable). Each of these subspaces subsequently serves as the basis for further partitioning independently of the others and so on. At each step the variable used for each split is selected from all the explicative variables so as to provide an optimal partition given the previous actions. Partitions' sequence is summarized by a binary tree. The root node of tree corresponds to the entire data space. Partitions of the space are associated with descendants of the root node. The leaves of the tree, or terminal nodes, correspond to subspaces which are not further partitioned. The stability of the procedure can be improved using Data resampling.
While CART-models are widely used as exploratory techniques they are less-commonly used for prediction. Trees generally rely on fewer assumptions than classical methods and handle a wide variety of data structures. Furthermore they are easy to use and to interpretate, and thus provide a wide range of application fields. The use of CART procedure has been considered by others in a variety of medical problems [22,23] such as, for example, survival analysis [24-26], longitudinal analysis, diagnostic and prognostic studies or clinical trials [27-30].
One particular application is signal processing [31], in which the problem concerns the detection of multiple change points in the mean. The CART procedure can be used to estimate simultaneously the change-points and the means by recovering an underlying piecewise constant function f(t).
If mk are the means for each piecewise k, then tk are the change-points:
If we extend this point of view to the covariate space defined by geographic coordinates, CART estimates the "change-lines" (instead of change-points) of a piecewise constant function on R2. In other words, tree-based procedure can easily determine spatial patterns.
However, one limitation is that CART provides axis-parallel splits i.e. rectangular spatial patterns. Oblique decision trees (ODT) deal with this problem. Those algorithms produce oblique (and then polygonal) partitioning of the covariate space. However, oblique trees are less popular than axis-parallel trees because the splits are less straightforward to interpret and oblique procedures require greater computational complexity than axis-parallel algorithms. Finding the best oblique tree in the covariate space is a NP-Hard problem [32]. Therefore, existing ODT algorithms use deterministic heuristics or evolutionary algorithms (like the OC1 system [33]) to find appropriate hyperplanes for partitioning the covariate space [22,34,32,33]. Comparisons of the different procedures are provided, for example, by Murthy [33] Cantu-Paz [34] and Bradley [35].
Despite this difficulty in RN, it is easier to find an oblique partition in the particular case of a space determined by the geographic coordinates, i.e. in R2. Evolutionnary or heuristic algorithms are not robust. They provide occasionally local minima [33] and therefore are not optimal procedures in R2. The ODT-algorithm we have developped is an optimal procedure to reach an optimal solution without using heuristics or evolutionary procedures.
ODT algorithm
The general purpose of the entire procedure consists on finding several partitions of the plane. We present the first step which allows finding the best oblique split of the plane. Going recursively, this algorithm will split the plane into several partitions, until reaching a specific criterion.
This subsection is organized as follow:
i. First, we will introduce how the plane is splitted into two adjacent partitions according to the interclass variance.
ii. Second, we will present how the finite set of oblique lines is determined, still within the first step of the entire procedure.
iii. Third, we will propose an optimization of this algorithm.
i. The splitting method proceeds as follows.
Consider, in the geographical space represented by the plane with an orthogonal basis {x, y} and a fixed origin O, n points Mi with coordinates {xi, yi}. These coordinates can represente the geographic coordinates of a point location provided by GPS. To each point Mi a numeric random variable Zi (called explained or dependant variable) is associated with the observation zi Whereas the CART procedure partitions the plane according to a line parallel to the axis maximizing the interclass variance of zi, our procedure partitions the plane according to an oblique line maximizing in the same way the interclass variance of zi.
To find this oblique line according to the direction we have to define the perpendicular direction u and the angle .
From a general viewpoint, for a fixed direction the procedure has to:
- Orthogonally projects the points Mi on the (O, u) direction, defining the coordinate ui;
- Considers all the ui as potential threshold in the way to split the plane with the direction perpendicular to the direction u and going through ui;
- Finds the optimal split between two adjacent classes, maximizing the interclass variance of zi according to theses projections.
ii. The splitting method provides a finite set of cluster proceeding as follows.
Before detailing the algorithm, we have to study the different splitting directions i.e. to specify wich angles θ have to be analyzed. For a global solution the algorithm can scan all the oblique directions (i.e. all the θ) between zero and π. In a heuristic way one can also discretize this interval providing a finite number of angles θ. But these two methods are not optimal.
The optimal algorithm for an optimal solution is easy to implement. Obviously, two points Mi (xi, yi) and Mj (xj, yj) have the same projected coordinates on the (O, u) direction if and only if Mi Mj is perpendicular to (O, u) (Figure 1). Then the number of critical directions, defined by the θij angles, exists and is finite.
Figure 1 Construction of the critical angle θij of the direction u. - the geographical space is represented by the plane with an orthogonal basis {x, y} and a fixed origin O; - u is a direction perpendicular to the splitting direction ; - Mi and Mj are two point locations in the geographical space.
For each direction passing through two points Mi (xi, yi) and Mj (xj, yj), we define φij the angle between the line Mi Mj and the x-axis.
Then:
As previously defined, θ is the angle between the x-axis and the direction (O, u) perpendicular to Mi Mj.
Then for each couple (Mi, Mj), we have
Each critical angle θij defines an angular sector. Within each sector, the order of the coordinates projected on the (O, u) direction does not depend on this direction. For points Mi and Mj the difference (uj - ui) of their coordinates projected on (O, u) verifies:
(uj - ui) cos(φij) = (xj - xi) sin(θ - θij) (1)
with (uj - ui) = (yj - yi) sin(θ) for xi = xj ⇔ φij =
Thus (uj - ui) depends continuously on θ. The sign of this difference cannot change within the angular sector since (uj - ui) = 0 only if θ = θij.
It follows that the interclass variances (and then the ODT procedure) is not modified within each sector. As a direct consequence of (1) the transition from a sector to another via the critical angle θij (Figure 2) induces the same order except the permutation of the two adjacent elements (ui, uj).
Figure 2 Passage through the critical direction u, from sector 1 to sector 2. - u is a direction perpendicular to the splitting direction ; Mi and Mj are two point locations in the geographical space; - Change in the order of the projected coordinates on the u' and u" directions; - u' and u" are directions with intermediate angles, belonging respectively to sector 1 and sector 2; - u'i, u'j, u"i, and u"j are the projected coordinates of points Mi and Mj: u'i > u'j and u"i <u"j.
Note that for aligned points Mi, Mj and Mk the algorithm has to permute the adjacent element group (ui, uj, uk). Similarly for parallel directions Mi Mj and Mk Ml, the algorithm has to permute at the same time the couples of adjacent elements (ui, uj) and (uk, ul).
Note again that all these angular sectors define as much covariates. Thus the procedure comes to the usual CART procedure. But the number of different critical angles is and using CART this way over-consumes time and space. For example, in our application the number of point locations is n = 164, hence the number of different angular sectors is N = 13270.
iii. We present now an optimization of our algorithm. A less time consuming and more efficient algorithm is a stepwise analysis of the angular sector, ordered according to the observed θij. At each step the algorithm uses the previous calculus.
Because only two elements between two adjacent sectors are permuted only one interclass variance has to be reloaded, related to the single different split (or some interclass variances for the group of permuted couples, related to a few different splits). The procedure inherits the calculus of the other interclass variances from the previous sector with the exception of the interclass variance related to the single permutation.
Thus, the algorithm complexity is (n2 log n) in time and (n) in space for one split. Finally, our algorithm splits the plane into two adjacent partitions as follows:
• Arrange the xi;
• Calculate and arrange the θij via the aij;
• Calculate ;
• For each potential split of the first angular sector (corresponding to the x-axis), i.e. for each value of xi:
- Calculate the ∑ zi for each class (on both sides of the threshold xi) and then the interclass variance, using the previous results;
- If the calculated interclass variance is greater than the previous one, store the results;
• For the next angular sector
- Permute the corresponding xi xj (or the group of elements);
- Calculate the ∑ zi only for the two classes generated by the split between xj and xi (or some splits for the group of permuted elements);
- If the new interclass variance is greater than the previous optimum, store the results;
• Until all sectors are scanned.
This algorithm goes on recursively until a specific criterion is reached and the Oblique Decision Tree is completed.
For simplicity we will not herein discuss special procedures of CART such as stopping rules, pruning algorithms or resampling methods; these are examined elsewhere [22,36].
Dataset
Malaria is the major parasitic disease in the world affecting approximately 300–500 million individuals annually. About two percents of the individuals infected with Plasmodium falciparum die. Most of the deaths occur in children. In the last decade, the incidence of malaria has been increasing at an alarming rate in Africa representing over 90% of the reported cases in the world [37].
The study area was the whole village of Bancoumana located in the administrative circle of Kati (Mali, Western Africa). This village is located in the high Niger's valley, a Sudanese savannah area, about 60 km south-west from the capital city Bamako. The main activities are rice cultures and truck farming along the Niger river. This village is 2.5 km2 wide, with 8 000 inhabitants (MRTC census, 1998) and about 1 600 children under 9 years. The transmission of malaria is high during the rain season (usually from June to October, with temperatures varying between 25°C and 40°C). It decreases then, reaching a low level of transmission one or two months thereafter.
The project investigated at a village-level approach (using a 1–3 m resolution scale) the risk of malaria infection. The presence of P. falciparum, the main infectious agent of malaria in this area, in blood smears was investigated in 1 461 children living in 164 households during the dry season in March 2000. Among them, 474 children had a positive blood smear (32.44%, CI95% [30.09–34.89]). Localization was performed through GPS receivers. Thus, all children were geocoded at a point location (corresponding to their house). Geo-database and cartographic displays were provided with the ArcGIS 8.3 software (ESRI, Redlands, CA).
Human subjects' research conducted in these studies was approved by the Institutional Committee on Ethics of the Mali Faculty of Medicine and Pharmacy, University of Mali. To obtain informed consents a stepwise consent process was applied as described by Doumbo [38]. First, the community informed consent was obtained before the beginning of the study. Second, the informed consent of the parents or guardians of the children were orally obtained before each clinical or biological investigation.
Data analysis
The ODT-algorithm was implemented with the Matlab Software 7.0.1 (The Mathworks Inc. 2004). We applied the ODT procedure to the dataset using the GPS coordinates of each location as independent covariates and the parasitic positivity rate (rate of positive blood smears per houses) as dependant variable. Thus ODT provided an optimal partition of the geographical area, i.e. a spatial pattern of the disease risk. We chose to use two classical stopping rules [22]. First, the ODT stopped if a class was made up of less than 15 locations. Second, we prunned a node if, after partition, one of the two resulting classes was made up of less than 3 locations.
For inference we considered the constant risk hypothesis as a model of "no pattern". Under this null hypothesis each child is at the same disease risk within the observation period regardless of his location. Thus the classes issued from the ODT displayed similar disease risk. However in keeping with many spatial health applications [12], we can not rely on asymptotic arguments to derive theoretically the associated distributions under the null hypothesis. Monte Carlo (MC) simulations were flexible tools for such assessment.
Similarly to many statistical models, we used for inference the explained variability rate Rv, defined as the ratio of the interclass sum of squared errors (SCE) (outcome of the ODT model) and the total SCE. We considered a Monte Carlo inference conditional on the set of all locations and on the local number of subjects. The total number of cases varied from simulation to simulation with an expected value (the total number of cases on the observed dataset). In this way, the simulations assessed spatial variations in the local proportion of cases conditional on the set of all locations. Monte Carlo simulations reflected a constant risk hypothesis similarly to the Rushton and Lolonis [39] approach. We ran 999 simulations under the constant risk hypothesis i.e. homogeneous Poisson distribution. Under this null hypothesis we applied the ODT-algorithm for each of the random dataset and calculated the empirical distribution of Rv. Thus the MC inference provided p-values for testing whether or not the observed explained variability rate is a realization of the theoretical (simulated) distribution under the constant risk hypothesis. In other words, MC inference tested the ODT-model and provided the significance of the spatial pattern issued from the oblique decision tree.
We compared the ODT-model outputs with those of the scan statistic method. For the latter, we used the software program SaTScan™ [18] in order to test for the presence of spatial clusters of malaria infection and to estimate their locations. The identification of high risk clusters with the SaTScan™ was performed under the Poisson probability model assumption using a maximal cluster size of 50% of the total population. For statistical inference, 999 Monte Carlo replications were performed. The null hypothesis of no clustering was rejected when the simulated p-value was lower than or equal to 0.05.
During the data analysis we calculated all confidence intervals of rates according to the Wilson method [40].
Results
Oblique Decision Tree
The ODT (Figure 3) partitioned the village into four risk classes. The explained variability rate is high, i.e. Rv = 83.96% of the variability is explained by the ODT-model. The global risk of disease (Table 1) was 32.44%, CI95% [30.09–34.89]. The ODT provided two classes of high infection risk. In the first high risk class (P2), located in the western part of the village (Figure 5), the risk was 60%, CI95% [52.22–67.55]. In the second high risk class (P3), located in the southern part of the village, the risk was about 50% with a large confidence interval. Note that during the rain season about 80% of the children had a positive blood smear in the whole village. Investigations at this site pointed to a small pond located within the western high risk class, and to ricefields located in the southern part of the village, both having been identified as Anopheles (the vector of malaria) breeding places.
Figure 3 Oblique Decision Tree for spatial partitioning. The geographical area is splited into 6 partitions. Nloc: number of locations belonging to each partition; n: total number of children of each partition; R: infectious rate; θ: critical angle for each split; Vic: interclasses variance for each split.
Table 1 Spatial pattern resulting from the ODT-model. The first line refers to the areas without any partition.
Centroid's Coordinatesa Pop.b Risk of infection [CI95%] Number of Locationsc
No pattern X = -8.266497256
Y = 12.20520982 1 461 32.44% [30.09–34.89] 164
P1 X = -8.270634
Y = 12.202594 30 26.67% [14.18–44.45] 5
P2 X = -8.27019
Y = 12.20438615 153 60.13% [52.22–67.55] 13
P3 X = -8.26849
Y = 12.1999733 26 50.0% [32.06–67.94] 3
P4 X = -8,2659751
Y = 12.205486 1 252 28.83% [26.39–31.4] 143
a- The coordinates are for the centroid of each partition.
b- Pop. refers to the total number of children included in each partition.
c- The number of locations refers to the total number of households within each partition.
Figure 5 The village of Bancoumana. - The circle S1 refers to the significative cluster provided by the Kulldorff's SaTScan. - The strait lines are the 3 splits resulting from the ODT-model, providing 4 partitions P1, P2, P3 and P4. - The bold grey line represents the Niger river. - Each location is represented by its own risk value. The scale of risks is discretized in 6 equal sized intervals.
Monte Carlo inference provided a global test, testing the null hypothesis of a homogeneous Poisson distribution of the malaria infection cases within the study area. Under this null hypothesis we provided (999 simulated sets and one observed set) the empirical distribution of the explicated variability rate Rv (Figure 4). In this application the Rv provided by the ODT-model significantly differed (p < 0.0001) from the one provided under the homogeneous Poisson distribution, i.e. the spatial pattern was significant.
Figure 4 Empirical distribution of the explained variability rate Rv. The distribution was provided by Monte Carlo procedure (999 simulated sets and one observed set).
Satscan approach
The Satscan results yielded one significant clusters (Table 2). In the first cluster (S1) the risk of disease was high with an infectious rate of 54.21% (CI95% [47.51–60.75]). Obvioulsy, his center was located within the high risk ODT partition (P2) and the risk of disease were similar in S1 and P2 (Figure 5). The second and third high clusters were not significant, totalizing only one point location each.
Table 2 High risk of malaria spatial clusters in Bancoumana, Mali, march 2000.
Cluster
Coordinatesa Radius Km Popb Risk of infection [CI95%] Cases Obs/exp Locc pd
S1 X = -8.27047
Y = 12.205325 0.27 214 54.21% [47.51–60.75] Obs:116
Exp:69.43 22 0.001
S2 X = -8.26701
Y = 12.205729 0.00 7 85.71% [48.69–97.43] Obs:6
Exp:2.27 1 0.998
S3 X = -8.26469
Y = 12.207877 0.00 20 60.00% [38.66–78.12] Obs:12
Exp:6.49 1 0.999
a- The coordinates are for the center of each circle.
b- Pop. refers to the total number of children into each cluster.
c- Loc. refers to the number of locations belonging to each cluster.
d- p-values refer to the Monte Carlo inference, after 999 replicates.
Discussion
For spatial cluster detection, the specification of the shape and size of the clusters is required rather than using political or administrative definitions of zones. For this purpose scanning methods provide sets of potential clusters but the problem of the choice of the shape still remains. Different scanning grid and different windows' shapes or sizes may provide different sets of potential clusters. To reduce this difficulty we introduced ODT-models with the aim to detect spatial pattern without pre-specifying windows' shape. In contrast to classical scanning procedures, neither the shape, size, nor centroid location have to be specified by the users. Thus, ODT are optimal procedures from the classification viewpoint.
Furthermore the spatial pattern obtained by the ODT-model defines a potential clusters set which can then be tested using the classical Monte Carlo inference. Similarly to Satscan, inference analysis has to avoid multiple testing inherent to such a procedure. The Kulldorffs procedure provides first a potential cluster set. Second, this procedure performs a significance test based on the local likelihood ratio statistic for each cluster in a way that compensates for the multiple testing. In our work we provide a global inference, testing the significativity of the spatial pattern obtained by ODT. Note that, similarly to Kulldorff's inference, likelihood ratio tests can be used to test the spatial pattern.
Recently, Tango proposed a flexibly spatial scan statistic to detect noncircular clusters [41]. But the Tango's method is not practically feasible for large clusters (more than 30 point locations).
Our findings indicate that the ODT-method is consistent with the classical Kulldorf's scan statistic. ODT procedure is thus a classification tool widely usable for spatial pattern detection. When compared to ODT, the scan statistic did not detect the second high risk cluster (P3). This is probably due to the lack of points fitting in this cluster (3 point locations and 26 childrens). The 95% confidence interval of the disease rate in this cluster is large (32.06%-67.94%). Nevertheless, after investigations in the village, a putative source of disease risk has been detected at this location. The two non-significant high risk clusters (S2 and S3) enclose only one point location each. This might explain both the lack of significativity with the satscan method and the lack of detection by the ODT.
After detection of a significant spatial pattern, the next logical step is to test whether this pattern can be explained by known or suspected risk factors. For example in the context of malaria, environmental factors such as mosquitoes breeding sites or thatched habitations might be identified and appropriate measures can be proposed to enhance the disease's control policy.
ODT-models allow for a flexible relationship between the variables. The relationships between covariates do not need to be linear or additive and the interactions do not need to be prespecified or to be of a particular multiplicative form. The literature about tree-based models is increasing particularly for studies focusing on formal inference procedures [31,36]. In contrast to classical ODT-procedures the algorithm herein described is optimal since it uses neither evolutionary algorithms [32,34] nor heuristics [22]. While the problem is NP-Hard in RN, the algorithm remains polynomial in R2.
The stability of tree-models can be improved by resampling methods. It is noteworthy that scanning methods such as satscan can also benefit from resampling. Resampling methods may improve the determination of the potential cluster set when the SaTScan™ procedure uses windows centered at each point location. Among different stopping rules, criteria have to be chosen according to the trade-off between variance and bias of prediction. The usually chosen rules are known as flexible and robust methods [31]. But as our application results indicate, less restrictive rule can be used for specific epidemiological dataset in order to improve the interpretation of the ODT-models' output. For rare diseases it might be necessary to use less stringent stopping rules than for diseases characterized by an epidemic evolution. This is related to the definition of "cluster of cases" which depends on the epidemiological profile of the disease.
The risk of infection has a high geographic variability [42,43] and the knowledge of this variability is essential to enhence malaria control programs' efficiency [44]. Moreover, the detection of high-risk locations is one recommendation of the 20th WHO technical report [45]. In this context, the development of GIS displays data on local malaria cases and then stratification of malaria risk providing the opportunity for more focal (and then efficient) malaria control programs [42].
Conclusion
In conclusion, Oblique Decision Tree is a new approach for spatial pattern detection and has the following features:
- ODT improve the classical scanning procedures by providing polygonal potential clusters;
- ODT are not bound by fixed centroid locations, sizes or shapes. Thus, first they have an enhanced flexibility. Second, the results are independent from the shape size and center pre-specification;
- ODT provide an optimal partition in the classification viewpoint.
Thus, ODT-models favorably compare with other cluster detection methods for spatial epidemiology.
Abbreviations
CART: Classification And Regression Tree
CI95%: 95% Confidence Interval
GIS: Geographic Information System
GPS: Global Positioning System
MC: Monte Carlo
MRTC: Malaria Research and Training Center, Bamako, Mali.
ODT: Oblique Decision Tree
SCE: sum of squared errors
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
J Gaudart provided the ODT procedure, implemented the algorithm, performed statistical and geographical analysis, and drafted the manuscript.
B Poudiougou, S Ranque and O Doumbo carried out the epidemiological georeferenced data and provided the parasitological analysis.
O Doumbo initiated and supervised the epidemiological study.
All authors wrote 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 French Research Ministry (PAL+ 2001 Program) and the NIH Grant Mali-Tulane TMRC N° AI 95-002-P50
We thank Dr Bernard Fichet for many valuable discussions, Dr Bernard Giusiano for program improvements, and Dr Joanny Gouvernet for his helpful comments.
==== Refs
Kulldorff M Feuer EJ Miller BA Freeman LS Breast cancer in northeastern United States: a geographical analysis Am J Epidemiol 1997 146 161 170 9230778
Bithell JF The choice of test for detecting raised disease risk near a point source Stat Med 1995 14 2309 2322 8711271
Cuzick J Edwards R Spatial clustering for inhomogeneous populations J R Stat Soc [Ser B] 1990 52 73 104
Tango T A class of tests for detecting 'general' and 'focused' clustering of rare diseases Stat Med 1995 14 2323 2334 8711272 8711272
Diggle PJ Morris S Elliott P Shaddick G Regression modelling of disease risk in relation to point sources J R Stat Soc [Ser A] 1997 160 491 505 10.1111/1467-985X.00076
Anderson NH Titterington DM Some methods for investigating spatial clustering, with epidemiological applications J R Stat Soc [Ser A] 1997 160 87 105 10.1111/1467-985X.00047
Tango T Score tests for detecting excess risks around putative sources Stat Med 2002 21 497 514 11836732 10.1002/sim.1003
Diggle PJ Chetwynd AG Second-order analysis of spatial clustering for inhomogeneous populations Biometrics 1991 47 1155 1163 1742435
Gomez-Rubio V ferrandiz J Lopez A Kurt Hornik, Friedrich Leisch, Achim Zeileis Detecting clusters of diseases with R Proceedings of the 3rd International Workshop on Distributed Statistical Computing: March 20–22 2003; Vienna Austria 2003
Turnbull BW Iwano EJ Burnett WS Howe HL Clark LC Monitoring for clusters of disease: application to leukemia incidence in upstate New York Am J Epidemiol 1990 132 S136 143 2356825
Kulldorff M A spatial scan statistic Commun Stat Theor M 1997 26 1481 1496
Waller LA Gotway CA Applied spatial statistics for public health data 2004 Wiley: Hoboken New Jersey
Wakefield J Elliott P Issues in the statistical analysis of small area health data Stat Med 1999 18 2377 2399 10474147 10.1002/(SICI)1097-0258(19990915/30)18:17/18<2377::AID-SIM263>3.3.CO;2-7
Kulldorff M Nargawalla N Spatial disease clusters: detection and inference Stat Med 1995 14 799 810 7644860
Thomas AJ Carlin BP Late detection of breast and colorectal cancer in Minnesota counties: an application of spatial smoothing and clustering Stat Med 2003 22 113 127 12486754 10.1002/sim.1215
Sheehan TJ De Chello LM Kulldorff M Gregorio DI Gershman S Mroszczyk M The geographic distribution of breast cancer incidence in Massachusetts 1988 to 1997, adjusted for covariates Int J Health Geogr 2004 3 17 15291960 10.1186/1476-072X-3-17
Hjalmars U Kulldorff M Gustafsson G Nagarwall N Childhood leukemia in Sweden: using GIS and spatial scan statistic for cluster detection Stat Med 1996 15 707 715 9132898 10.1002/(SICI)1097-0258(19960415)15:7/9<707::AID-SIM242>3.3.CO;2-W
Kulldorff M SaTScanTM v5.l-Software for the spatial and space-time scan statistics 2004 Information Management Services Inc., Silver Spring, Maryland
Gangnon RE Clayton MK Bayesian detection and modeling of spatial disease clustering Biometrics 2000 56 922 935 10985238 10.1111/j.0006-341X.2000.00922.x
Patil GP Taillie C Upper level set scan statistic for detecting arbitrarily shaped hotspots Environ Ecol Stat 2004 11 183 197 10.1023/B:EEST.0000027208.48919.7e
Duczmal L Assunciao RM A simulated annealing strategy for the detection of arbitrarily shaped spatial clusters Comput Statist Data Anal 2004 45 269 286 10.1016/S0167-9473(02)00302-X
Breiman L Friedman JH Olshen RA Stone CJ Classification and regression trees 1993 Chapman & Hall: New York
Segal MR Tager IB Trees and tracking Stat Med 1993 12 2153 2168 8310186
Xu R Adak S Survival analysis with time-varying regression effects using a tree-based approach Biometrics 2002 58 305 315 12071403 10.1111/j.0006-341X.2002.00305.x
Leblanc M Crowley J Relative Risk trees for censored survival data Biometrics 1992 48 411 425 1637970
Schmoor C Ulm K Schumacher M Comparison of the Cox model and the regression tree procedure in analyzing a randomized clinical trial Stat Med 1993 12 2351 2366 8134738
Zhang H Holford T Bracken MB A tree-based method of analysis for prospective studies Stat Med 1996 15 37 49 8614744 10.1002/(SICI)1097-0258(19960115)15:1<37::AID-SIM144>3.3.CO;2-S
Crichton NJ Hinde JP Marchini J Models for diagnosing chest pain: is cart helpful? Stat Med 1997 16 717 727 9131760 10.1002/(SICI)1097-0258(19970415)16:7<717::AID-SIM504>3.0.CO;2-E
Fu CY Combining loglinear model with classification and regression tree (CART): an application to birth data Comput Statist Data Anal 2004 45 865 874 10.1016/S0167-9473(03)00092-6
McBride WJH Mullner H Muller R Labrooy J Wronski I Determinants of dengue 2 infection among residents of charters towers, Queensland, Australia Am J Epidemiol 1998 148 1111 1116 9850134
Gey S Bornes de risque, détection de ruptures boosting: trois thèmes statistiques autour de CART en régression PhD thesis 2002 University of Paris XI
Heath D Kasif M Salzberg S Ruzena Bajcsy Induction of oblique decision trees Proceedings of the 13th International Joint Conference on Artificial Intelligence: August 28-September 3 1993; Chambery, France 1993 Morgan Kaufmann 1002 1007
Murthy SK Kasif M Salzberg S A system for induction of oblique decision trees J Artif Intell Res 1994 2 1 32
Cantu-Paz E Kamath C Inducing oblique decision trees with evolutionary algorithms IEEE Trans Evol Comput 2003 7 54 68 10.1109/TEVC.2002.806857
Brodley CE Utgoff PE Multivatiate decision trees COINS technical reports 92-82 1992 University of Massachusetts
Ghattas B Agrégation d'arbres de décision binaires: application à la prévision de l'ozone dans les Bouches du Rhône PhD thesis 2001 University of Aix-Marseille II
Tanser F Le Sueur D The application of geographical information systems to important public health problems in Africa Int J Health Geogr 2002 1 1 12437788
Doumbo OK It takes a village: medical research and ethics in Mali Science 2005 307 679 681 15692036 10.1126/science.1109773
Rushton G Lolonis P Exploratory spatial analysis of birth defect rates in an urban population Stat Med 1996 15 717 726 9132899 10.1002/(SICI)1097-0258(19960415)15:7/9<717::AID-SIM243>3.0.CO;2-0
Newcombe RG Two-sided confidence intervals for the single proportion: comparison of seven methods Stat Med 1998 17 857 872 9595616 10.1002/(SICI)1097-0258(19980430)17:8<857::AID-SIM777>3.0.CO;2-E
Tango T Takahashi K A flexibly shaped spatial scan statistic for detecting clusters Int J Health Geogr 2005 4 11 15904524 10.1186/1476-072X-4-11
Booman M Durrheim DN La Grange K Martin C Mabuza AM Zitha A Mbokazi FM Fraser C Sharp BL Using a geographical information system to plan a malaria control programme in South Africa Bull World Health Organ 2000 78 1438 1444 11196490
Baird JK Agyei SO Utz GC Koram K Barcus MJ Jones TR Fryauff DJ Binka FN Hoffman SL Nkrumah FN Seasonal malaria attack rates in infants and young children in Northern ghana Am J Trop Med Hyg 2002 66 280 286 12139221
Killeen GF Seyoum A Knols BGJ Rationalizing historical successes of malaria control in africa in terms of mosquito resource availability management Am J Trop Med Hyg 2004 71 87 93 15331823
World Health Organization Expert Committee on Malaria 20th report World Health Organ Tech Rep 2000 735
|
16026612
|
PMC1198231
|
CC BY
|
2021-01-04 16:32:51
|
no
|
BMC Med Res Methodol. 2005 Jul 18; 5:22
|
utf-8
|
BMC Med Res Methodol
| 2,005 |
10.1186/1471-2288-5-22
|
oa_comm
|
==== Front
BMC NephrolBMC Nephrology1471-2369BioMed Central London 1471-2369-6-91608683410.1186/1471-2369-6-9Study ProtocolCase-control study on analgesics and nephropathy (SAN): protocol Heinemann Lothar AJ [email protected] Edeltraut [email protected] Michael [email protected] der Woude Fokko [email protected] Helmut [email protected] Centre for Epidemiology & Health Research Berlin, Invalidenstr. 115, 10115 Berlin, Germany2 Institute for Clinical Pharmacology, Charité-University Medicine Berlin, Schumannstr. 20–21, 10117 Berlin, Germany3 EPES Epidemiology, Pharmacoepidemiology and Systems Research GmbH, Wulfstr. 8, 12165 Berlin, Germany4 Nephrology, 5. Med. Klinik, Klinikum Heidelberg-Mannheim, Theodor-Kutzer-Ufer 1–3, 68167 Mannheim, Germany5 Neprologie, Krankenanstalt der Stadt Wien, Rudolfstifting, 3. Med. Abteilung, Juchgasse 25, 1030 Wien, Austria2005 8 8 2005 6 9 9 19 4 2005 8 8 2005 Copyright © 2005 Heinemann 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 association between intake of non-phenacetin-containing analgesics and the occurrence of chronic renal failure is still controversially discussed. A new epidemiologic study was planned and conducted in Germany and Austria.
Methods/design
The objective of the international, multicenter case-control study was to evaluate the association between end-stage renal disease (ESRD) and use of non-phenacetin-containing analgesics with particular emphasis on combined formulations. A targeted sample of 1000 new (incident) dialysis patients, aged less than 50 years, was planned to recruit between January 1, 2001 and December 31, 2004. The age limit was chosen to avoid contamination of the study population with phenacetin-containing analgesics to the extent possible. Four control subjects per ESRD case, matched by age, sex, and region were selected from the population living in the region the case came from.
Lifetime exposure to analgesics and potential renal risk factors were recorded in a single face-to-face interview. A set of aids was introduced to reinforce the memory of study participants.
A standardized, pre-tested interview questionnaire (participants), a medical documentation sheet (physicians in dialysis centres), a logbook for all activities (dialysis centres) were used to collect the necessary data.
Quality management consisted of the standardized procedures, (re-) training and supervision of interviewers, regular checks of all incoming data for completeness and plausibility.
The study is scientifically independent and governed by a international Scientific Advisory Committee that bridged the gap between the sponsoring companies and the investigators. Also other advisory groups assisted the managing committee of the study. All relevant German and Austrian nephrological associations supported the study, and the study design was carefully reviewed and approved by the Kidney Foundation of Germany.
Discussion
The study is expected to answer the main research question by end 2005. There is however a high potential for various biases that we tried to address with adequate measure. One limitation however cannot be overcome: The methodologically needed age-limitation of the study will make it not easy to generalize the results to age groups over 50 years. It might be suggested to repeat the study for persons over 50 years in 10 years when contamination with phenacetin use early in life is likely to be outgrown.
==== Body
Background
It was long time ago that Dubach et al. published their paper on phenacetin nephropathy [1]. At that time, phenacetin-containing analgesics were freely available in bulk to all workers in the Swiss watch industry [1], and sachets of phenacetin-containing analgesic powders were even given as a tip for small services to apprentices in the Huskvarna factory in Sweden [2]. Several epidemiological studies were undertaken to clarify the association between phenacetin overuse and development of end stage renal disease (ESRD)1 (see [3,4] for review). As a result, phenacetin was banned at different times in almost all states of the world. However, although "phenacetin nephropathy" – thought to develop only after decades of chronic phenacetin overuse – could not be expected to disappear soon, nephrologists suspected that non-phenacetin analgesic combinations, especially those containing paracetamol and aspirin (plus caffeine) might cause ESRD as well and introduced the hypothetical term "analgesic nephropathy". Since empirical data were scarce, this hypothesis engendered some controversy. Safety concerns led the regulatory authorities of Germany, Austria and Switzerland to initiate a scientific re-evaluation. A peer review committee of scientists was jointly selected by the regulatory authorities and the pharmaceutical industry and was asked to critically review data on the relationship between non-phenacetin combined analgesics and nephropathy. The committee's two main conclusions were that (1) there is insufficient evidence to associate non-phenacetin combined analgesics with nephropathy and that (2) new studies should be done to provide appropriate data to resolve the question [5]. Meanwhile the US Food and Drug Administration (FDA) had approved an analgesic combination containing paracetamol, aspirin, and caffeine as safe and effective for the use in uncomplicated migraine [6].
The Scientific Advisory Committee recommended the new study to be conducted within a relatively short time period to meet regulatory needs. The Drug Authorities (Germany, Austria, Switzerland) agreed that the year 2006 is as an acceptable deadline for obtaining final results. The discussions emphasized that the study should be designed and conducted in a way that ruled out any possibility of contamination by prior phenacetin use. In addition, it was agreed to conduct the study in those countries that had expressed concern about the problem, since the relevant compound analgesics are available on the markets of these countries and have a sufficient market share.
A further question to be addressed was whether caffeine co-formulated with analgesics could lead to a dependence on analgesics, contribute to heavy long-term use, and thereby result in a higher risk of ESRD. Although an additional Expert Meeting on Caffeine found no evidence for such an effect [7], and although a further special review did not reveal any substantial data to support a pivotal role of caffeine [8], the new study was intended to provide information on the role of caffeine in risk development.
Since ESRD is a rare outcome, a case-control study seemed to be the appropriate design (initially called "Study on Analgesic Nephropathy", SAN). Neither a study outside Europe (where combined analgesics are used to a lesser extent) nor long-term prospective cohort studies were regarded as acceptable alternatives, since the latter would not provide timely results. It was not possible to conduct a historical cohort study, because the required data were not available.
The protocol of the case control study as agreed upon by the Scientific Advisory Committee was approved by the regulatory authorities of Germany, Austria and Switzerland and was also reviewed by 'The Lancet' without any recommendation for change [personal communication; Prof. Fox, 2004] The initial study protocol can be read in the website ).
The objective of this publication is to present the study protocol with some minor revisions as approved by the Scientific Advisory Committee prior to the final analyses and subsequent publication in 2006.
Methods/design
General study design and rationale
A case-control study design was chosen because the study addresses the potential association between a rare medical condition (ESRD) and a relatively common exposure (analgesics) and because it needed to be conducted within a short period of time at affordable costs. Cases would be individuals initiating dialysis (incident ESRD) at one of the participating centers, and controls would be taken from the general population.
Two major concerns influenced the planning of the study:
- To exclude the possibility of inadvertent contamination with phenacetin among study subjects, study participation was restricted to individuals under 50 years of age. While this measure greatly impedes case accrual (only 25% of incident ESRD is within this age range), it assures minimal phenacetin contamination, because phenacetin use declined in the 1970ies and the drug was removed from the market in the countries under study in mid to late 1980. This age limit assures good recall of drug histories among these young study subjects and captures the group most likely to be affected if the metabolite of phenacetin, paracetamol, were associated with ESRD.
- The second issue concerns the time from onset of the disease until the diagnosis of ESRD which leads to dialysis, which is estimated to be between 5 to 10 years. Because the applicability of this pattern in association with a putative cause such as analgesic use is unclear, the protocol permits analyses of lag time and of appropriate subgroups.
Objectives of the study
The study was designed to address as many questions as possible about the association between ESRD and analgesic use in the range of recommended dosage with sufficient statistical power and in a timely manner.
The primary study objective is to determine whether a significant association between the occurrence of ESRD and the use of phenacetin-free analgesics in the average population exists or not.
A. The following three groups are of primary interest in the context of the main research objectives (basis for sample size estimations):
- Use of any phenacetin-free analgesics
- Use of any phenacetin-free analgesics in fixed combinations ("combis")
- Use of any phenacetin-free single substance – analgesics ("monos")
B. Drug-specific analyses will be done for groups with sufficient numbers, focusing on paracetamol with and without other analgesics, and with and without caffeine in the formulation.
C. In addition to the comparison between use and non-use (see "exposure"), the magnitude of use, such as dose, duration, cumulative lifetime dose per duration of use and other measures for the extent of exposure will be assessed. The desirable sub-group analyses (cf. exposure groups) are listed in advance for transparency, although it is recognized that some analyses will lack statistical power.
If a significant association of analgesic use and risk of ESRD becomes apparent, the secondary objective of the study is to analyze the patterns of use among individuals with excessive analgesics use (over 30% in excess of the recommended dose) and to address the role of caffeine co-formulated with analgesics and in conjunction with caffeine use in beverages, medical conditions, and indicators of general health behavior. This, however, is not part of the main research question focused on the risk of ESRD in the average population, i.e. not in specific sub-groups.
Study design
The study is conducted as a multicenter case-control study in Germany and Austria. New (incident) cases of ESRD will be collected from dialysis centers. On average, four population controls will be individually matched per case for age, gender and region the case came from.
Cases are defined as persons with end-stage renal disease (ESRD), newly admitted to a chronic dialysis program. Since there may be a considerable lag time between the underlying condition and the diagnosis of ESRD and because the underlying renal disease itself may have led to the use of analgesics, four index dates preceding the actual diagnosis of ESRD were defined. All potentially associated exposures (analgesic use, exposure to other chemicals at the work site), and conditions or complaints occurring after a given index date will not be considered in the statistical analyses. The four index dates were defined as follows:
• Index date 1: Time of first entry into the dialysis program for cases; time of personal interview for controls.
• Index date 2: Time the case was first informed about incipient renal failure from the treating physician; the index date for the control is the index date of the respective case.
• Index date 3 represents an arbitrarily defined lag-time of 5 years prior to the start of dialysis; the index date for the control is the index date of the respective case.
• Index date 4 represents an arbitrarily defined lag-time of 10 years prior to the start of dialysis; the index date for the control is the index date of the respective case.
Since index date 2 might be subject to information- and recall bias, index date 3 is considered to be the most reliable index date for risk estimates. Index 2 might be used if the number of analgesic users is too small at index date 3. The analgesic exposure of cases and controls at index date 4 (10 years before terminal renal insufficiency) was predicted to become too small for stable analyses, although this date might be biologically sound. The figure shows the relationship between index dates and putative analgesic use. Note that Index date 2 may have a wide range.
Outcome measure
The outcome measure is the relative risk of ESRD associated with use of analgesics either as single or combined formulation: heavy use (see later) vs. no use; lifetime dose, or duration of use.
Study participants
Recruitment of centers
The study aimed to collaborate with as many dialysis centers as possible in Germany and Austria to guarantee a sufficiently high and complete recruitment of incident ESRD patients in this young age group. A compensation for the time spent was offered to controls – if requested- and on demand to cases (up to 20 €) as well as to dialysis centers (up to 40€ for completing the medical history and diagnostic documentation).
The Clinical Principal Investigators, Prof. van der Woude for Germany and Prof. Graf for Austria, were responsible for organizing the recruitment of dialysis centers via the relevant professional organizations. In Germany, a joint effort of three nephrological associations under the auspices of the German Kidney Foundation facilitated contacts to hundreds of dialysis centers and thereby a good study performance. The German Kidney Foundation independently approved the study protocol, established an organizational network for the smooth collaboration among centers and with the Data Management and Coordination Center (DMCC – ZEG Berlin), and officially invites the dialysis units on behalf of the German Kidney Foundation. In Austria, Prof. Graf organized the study after approval from the Nephrological Association of Austria, which officially invited the dialysis centers' participation. In both countries, an information and training program was introduced for all dialysis centers participating in the study.
By the end of 2004, 259 dialysis centers had agreed to participate in the study and 170 contributed at least one incident case of ESRD. Details about centers and geographical distribution can be seen at the website of the SAN study .
Recruitment of cases and controls
All study participants received a comprehensive letter of invitation which explained the study objective of investigating various health outcomes with respect to life-time exposure to chemicals including drug treatment. No reference to analgesics is made in the invitation letter. The letter in addition provides some sample questions about exposure to chemicals and health outcomes similar to those that might be asked in the study interview. One intention of the letter is to stimulate controls, prior to the interview, to carefully think about their lifetime history of exposure to any chemicals, including drugs, in relation to medical and other conditions. This approach was chosen to minimize recall bias that could arise because cases in the past were likely asked more often for potential exposures than controls. Cases may therefore be more fully aware of previous exposures than controls. The letter was therefore designed to create a more similar interview situation for cases and controls by stimulating controls to reflect about prior exposures and health outcomes before the study interview.
Cases
Cases are persons with terminal renal failure with a first admission to renal replacement treatment because of severe renal insufficiency – end stage renal disease ("ESRD cases"). Individuals with conditions leading to an acute dialysis program were not eligible as cases. The renal function of subjects included as cases was re-evaluated after 3 months. Patients who were found to be acute dialysis patients or patients for re-dialysis after transplant rejection were then excluded.
Logbooks were kept in all participating centers for the duration of their participation in the study to monitor case accrual and completeness of ESRD patient capture. It was anticipated that not all centers would be able to participate for the entire study period, some entering late, and others discontinuing for various reasons – e.g. due to excessive workload.
Cases are eligible if they are new (incident chronic dialysis, not acute or recurrent) and under 50 years of age, if they are in adequate physical and mental condition to be interviewed, and if they are willing to participate in the study as confirmed by signed informed consent. Some of the new cases reported by the dialysis centers (documented in the logbook) are expected to meet not the eligibility criteria. It was expected that more than 70% of reported cases are eligible (eligibility rate). A response rate of about 80% of all eligible incident cases admitted in the respective dialysis unit (as documented in their logbook) was assumed for the sample size calculation.
Cases are invited to participate in the study by the treating physician at the dialysis center, who explains the objectives of the study as being "research into associations between chemicals and health". Care is taken that no reference is made to a suspected association between renal failure and analgesics. The physician provides the patient with the invitation letter, which is identical for cases and controls and which contains details on the interview process as well as examples of the questions asked in the interview. If the patient agrees to participate, his or her contact details are sent to the Data Management and Coordinating Center (DMCC: ZEG Berlin) that informs the interviewer staff to make an appointment. The interviewer then visits the patient at home after having made an appointment. No interview is done without a signed informed consent form.
All cases and controls were normally interviewed in person at their homes on the basis of a detailed and structured questionnaire. The same interviewer conducts all interviews for each case-control cluster (one case and up to four controls).
The interview provides information about exposure, relevant co-variables, and times of exposure (diary or calendar method), permitting the full use of the time-related variables for the statistical evaluation of time dependent data or pre-planned sub-groups (cf. analyses plan).
Controls
The study is designed to use population controls as reference group. Eligible controls are persons without ESRD aged less than 50 years matched individually for age (same 5-year age group as the case), sex and region of the respective case at entry in the dialysis program. Controls are eligible if they are willing to participate in the study, provide informed consent, and are in adequate physical and mental condition to be interviewed. Controls may have any medical conditions except ESRD to be eligible, because the controls should represent the population with regard to medical conditions. In equivocal circumstances (e.g. controls with kidney stones or other kidney diseases), controls will be considered separately in the analysis but will not be stopped to complete the interview.
Controls are recruited after the case interview is completed and the case's matching region is known. Four matched controls were tried to accrue per case. If one potential control refuses participation the next eligible person replaces him or her. The method of control identification varies depending on location. Whereas for larger cities listings of residents form the basis for recruitment, smaller villages are more suited for random route sampling and neighborhood or telephone contacts.
A response rate of 60% was assumed (after correction for non-eligible controls resulting from death, incorrect address, and other reasons that prevent to get into contact with the potential control person). Based on the experience from recent population-based case-control studies in Germany, this response rate is a realistic estimate.
Non-participation, and non-response for cases and controls
There are various reasons for non-participation in the study or exclusion from the analysis for individuals who were considered initially eligible to participate in the study.
The dialysis centre for various reasons might inadvertently miss eligible cases or report cases that are not eligible. These are documented in the logbook and expressed in the "eligibility rate" on the basis of all new patients reported from each of the collaborating centers.
An identified case might become ineligible due to poor physical or mental condition, insurmountable language problems, and death after identification but before interview, dialysis outside the study period, or being outside the age limit. Some of these reasons also apply to ineligibility of population controls. The proportion ineligible cases for these and other reasons forms the "eligibility rate" (Number of eligible cases divided by the number of initially identified cases
Eligible cases or controls may refuse to sign the informed consent and decline participation, or refuse to complete the interview. This is called "non-response" in the context of this study (Number of included/interviewed cases divided by the number of eligible cases).
Finally, cases or controls will be excluded from the analysis due to the use of any phenacetin-containing analgesic in any dose during their lifetime.
Exposure and co-variables
Analgesics
The main exposure variable is analgesic use. A lifetime history of analgesic use is obtained during the personal interview. Memory aids include a list of brand names of analgesics and an atlas with pictures of packages of all analgesics, both past and present. Participants are asked for the brand names, number of daily doses, start and stop of use, and the reasons/indications for use. The brand names were used to identify all analgesics whether prescription or over-the-counter (OTC), single-substance and combination analgesics with and without caffeine, phenacetin, or additional substances such as codeine and barbiturates. Information is also collected on switching analgesics and the reasons for switching, as well as on subjective symptoms of "dependence" and psychological or behavioral patterns.
The exposure definitions are similar to those of two recent studies in Germany (MURC study [9] and Pommer et al [10]), to facilitate comparisons with these studies that are the only available on analgesic use in Germany or Austria, particularly on OTC analgesics.
The cumulative lifetime use of all analgesics and of specific analgesic subgroups will be based on grams of the respective analgesic group (see below). The cumulative lifetime dose of analgesics will be stratified into tertiles to obtain the distribution in population controls, being called low, moderate, and heavy use. The Scientific Advisory Committee (SAC, see later) recommended the use of the top tertile of lifetime analgesic dose (grams) as the measure for heavy analgesic exposure in preference to the definition of "substantial use" as it was defined in the study protocol before revision. This decision was made after data from the population controls in the planned review (according to the first protocol) showed that the utilization of analgesics had changed considerably in the decade since "substantial use" [10] was defined:
- "Heavy users" are defined as persons whose use of analgesics is in the top tertile of the accumulated lifetime dose in grams. Alternatively, heavy use can also be defined in a similar way for shorter periods, i.e. not for lifetime but for any of 12-months periods or other units. In contrast to lifetime use this was defined as "peak use". However, the main analyses will be based on cumulated lifetime dose that will have priority for answering the main research question.
- "Non-users" are defined as never-users or short-term/irregular users of any analgesics (<1 dose of analgesic per month over all possible previous 12-month periods). Participants matching this definition are members of a unique, general reference group which will be identical across all comparisons.
Persons with recalled use of phenacetin-containing analgesics in the past will be excluded from the main analyses of the study – even if phenacetin use cannot be confirmed because of vague recall.
The coding of lifetime use of analgesics follows the discussion and agreement of the Advisory Committee with a classification into 13 categories (see Table 1).
Due to the low prevalence for some analgesic groups found in the analyses of the drug use in controls after year one of the study, two broader categories (sub-categories) of phenacetin-free analgesics were formed: MONOS (all monos; paracetamol; ASA; rest of monos) and COMBIS (all combis; combis with paracetamol; combis without paracetamol; combis with caffeine; combis without caffeine).
These categories of lifetime use of analgesics are not mutually exclusive. Pure groups of users of one type of analgesics are very rare and only available for very commonly used analgesics such as ASA. This distinguishes an observational study from a clinical trial and requires an analysis adapted to the prevalence in users as found in the database. The magnitude of usage (lifetime dose, peak dose, duration, dose per duration of use) will be examined by categorizing these variables into tertiles or other groupings depending on the number of users in each group.
Other variables
The following co-variables, evaluated as independent risk factors for either initiation or promotion of ESRD, will be considered as potential confounders or effect modifiers of a potential association between ESRD and analgesic use:
Age, sex, education, region, selected conditions (renal diseases, urinary tract infections, family history of renal diseases, repeated abdominal X-ray, psychological conditions), co-medication (rheumatic drugs, immunosuppressants, anti-cancer drugs, antibiotics, self reported dependence on any drugs), complaints (gastrointestinal, heart, vomiting, depression, anxiety, sleeplessness, fatigue, irritability, stress, eating problems, and various pain [joints, back pain, migraine, other headache, menstruation), and exposure to potentially nephrotoxic agents at the work site (such as heavy or other metals, special or general silicates, solvents, soldering/welding fumes).
Sample size estimation
The sample size needed to address the three main research questions was estimated with the following assumptions: Alpha 5%, beta 20% (power 80%), case-control ratio 1:4, minimal detectable risk (odds ratio) 2.0, and revised prevalence of heavy analgesic use (lifetime dose in grams; top tertile of use in each of the analgesic groups; adapted to the more accurate estimates from the study population; old estimates (see ) based on empirical data of population controls of the SAN study from end 2003 (see table 2). The sample size of cases and controls in table 2 is based on prevalence estimates of heavy use in different analgesic groups and thereby provides an impression what comparisons are likely or unlikely to be finally analyzed.
We conclude from this table that our initial estimate of the required sample size (cf. 1st study protocol in ) is still roughly valid. About 1000 cases of ESRD need to be identified in the co-operating dialysis centers to include 800 cases and 3200 controls in the final analysis which would result in sufficient power for the analysis of the main study questions and would additionally provide results for many but not all of the subgroups.
Quality control & assurance measures
Quality control and assurance measures are implemented to ensure that the procedures and data are reasonably valid, and compatible within and among centers and between cases and controls.
The quality assurance measures were directed at the interview technique, the preparation of the field work, the conduct of the study, and finally at the plausibility of the database. The interview was standardized and the questionnaire was tested in a pilot phase for the clarity of its items. The interviewers were trained and re-trained and appropriate explanations were integrated into the interviewer training plan and guideline. A set of aids was introduced to reinforce the memory of study participants such as picture displays of analgesic packages, tables with lists of drugs for the study subject to read etc. to avoid recall bias. Training and retraining of all personnel involved in the study was done: of investigators, abstractors of medical records, interviewers, following a standardized training plan.
A detailed logbook was kept in each of the participating dialysis centers. This provides information on the eligibility or ineligibility of cases and controls, participation rates and the reasons for non-responses or non-participation, the status of recruitment and other details. Additionally, site visits were done, if required.
Incoming data were checked in the Data Management and Coordinating Centre Berlin for quality and comprehensiveness by a sophisticated quality assurance system and queries were made to the study center if there was doubt about the validity of the data or if there were missing data.
Data management and analysis plan
Data management
The data were collected locally and thereafter transferred to the central Data Management and Coordination Centre (DMCC) Berlin according to clear time lines and defined responsibilities. The DMCC consists of ZEG Berlin (data management) and EPES Berlin (data analysis).
The data were entered into the database and thoroughly checked for errors and plausibility. The database will then be transferred into the "clean data set" that will contain the following groups of data:
1. Case/control variable ESRD (diagnostic categories, time variables as far as available)
2. Main exposure variables – use of analgesics (up to the time of respective index dates)
3. Co-variables:
- Co-morbidity: renal, circulatory, metabolic, psychiatric, and other conditions; headache and other painful conditions;
- Treatment history: types of treatment with potential relation to outcome and risk factors;
- Exposure to caffeine: coffee, tea, and caffeine-containing beverages;
- Job exposure: occupation & industrial branch; exposure in selected occupational categories; exposure to groups of certain chemicals/minerals;
- Other personal data: Age, sex, centre, health care contacts, socio-demographic markers
4. Factors potentially related to overuse of analgesics: psychological / vegetative / psychiatric conditions; behavioral factors (such as smoking, alcohol, physical activity).
Several databases will have to be established for different analyses which relate to the different index dates, different classifications of user groups of analgesics etc.
Analysis plan
In the final analysis, appropriate multivariate analyses will be done to assure that the risk estimates for ESRD and use of phenacetin-free analgesics are appropriately adjusted for confounding and effect modification. A detailed analysis plan was discussed at the meeting of the SAC in April 2004, and approved after revisions. Briefly, the "core analyses" will include:
- Information on selection such as included cases by regions/country, non-participation rate by case/control status, non-response rate by case/control status, reasons for non-participation and non-response by case/control status.
- Frequency distributions of exposure with analgesics by case/control status and by index dates such as ever vs. never use, lifetime dose (grams) in tertiles vs. non-use, duration of use (years) in tertiles vs. non-use, density of use (dose per duration of use).
- Logistic regression analyses of ESRD and analgesic exposure by index dates: Ever vs. never use, lifetime dose (grams) in tertiles vs. non-use, duration of use (years) in tertiles vs. non-use, density of use (dose / duration).
Several methodological issues will be addressed in the analyses:
One such issue relates to the precise definition and status of the index dates. Index date 1 is used only for description of the group of cases and controls, but not for the interpretation of the risk of nephropathy. The preferable index time for risk estimates is index 3. If the exposure data at index date 3 are insufficient, then SAC endorsed the use of index date 2.
The analgesic groups analyzed in the "core analyses" are not mutually exclusive. This means that the analysis of heavy lifetime use in one specific analgesic group is usually contaminated by use in other analgesic subgroups. It is unlikely to find sufficiently large "mutually exclusive groups of heavy analgesic use" will exist in the database. Moreover, it is unlikely that a satisfactory statistical approach for this problem will be found. The final decision on which method will be employed depends on the quantity of data available.
The inclusion of dosage of a specific analgesic subgroup (grams) as a continuous variable was discussed to provide a more comprehensive analysis of the dose-response effects in the "normal" user population. Such a variable might preferably be defined in broader categories of grams, showing not the increase of risk per one gram but per 10 or 50 or 100 or even 500 grams of lifetime dose.
An analysis of peak use might be appropriate if the data indicate a consistently increased risk in some of the analgesic subgroups. Peak use is defined as high use within a brief period of time (12 month) with only little or average use before and after this episode. This approach differs from the analyses of the cumulated lifetime dose, and it is not the focus to answer the main research question.
These and other data-driven analyses are suitable to test or confirm the biological plausibility of results. However, these analyses are hypothesis generating and will not be conducted to answer the planned main research questions.
The final analysis will face the problem of small numbers in many analgesic subgroups. The SAC strongly recommended reducing the number of adjustment variables to an absolute minimum. The appropriate statistical methods will be applied to select the most important confounders to get more stable risk estimates for small numbers. Further, the committee recommended not calculating or report adjusted odds ratios if any cell contains less than 10 subjects.
The findings must be interpreted with the consideration in mind that numerous analyses will be done to scrutinize the many available databases. Therefore it is statistically expected that some significant results will be found. This does not necessarily reflect a causal association because it may be a statistical artifact. The option of controlling for multiple testing in an observational study was discussed but not finally decided. It might be sufficient to interpret the results with great caution without formally applying further statistical procedures.
Study management
The management of the study with regard to clinical/nephrological aspects, including recruitment of co-operating dialysis centers, is the responsibility of Prof. van der Woude (V. Medizinische Klinik- Nephrologie, Klinikum Mannheim, Medizinische Fakultät der Universität Heidelberg) for Germany, and of Prof. Graf (Abteilung für Nephrologie des Krankenhauses Rudolfstiftung in Wien) for Austria.
The management of the whole study with regard to epidemiological expertise, data management and analyses is the responsibility of Prof. Heinemann (Centre for Epidemiology & Health Research Berlin).
The three Principal Investigators mentioned above form the Managing Committee of the SAN study in Germany and Austria.
The German Kidney Foundation plays a key role in supporting the study, particularly with regard to co-operation with the three German Nephrological Associations (see acknowledgments) and thereby with the collaborating dialysis centers. The same role plays the Austrian Nephrological Association for this country.
Agreements on terms of reference of various partners in the study, on ownership of data, accessibility of the database, and publication policy are available on the website of the SAN study . This website also contains the current status of the fieldwork and other accessible data.
Advisory committees
The scientific advisory committee (SAC) was jointly nominated by the Drug Authorities and the Industry. It consists of internationally acknowledged experts for different fields of expertise. The terms of reference of this committee and the names of its members are detailed in annex 1 and on the SAN website . The SAC also discussed and approved revisions of the protocol. It will finally approve the main analyses and the main publication prior to submission.
Delegates of the Nephrological Associations (three German and one Austrian Association) formed a Steering Committee to support the study. Two members of each Association of the two countries where the study is conducted have a seat in this advisory committee. The steering committee received monthly reports from the study. It can request additional information, and supports the study by identifying and resolving problems that occur within the study period. Details about Terms of Reference and names of members can be found on the SAN website .
Communication
Periodic reports of study progress are made to the SAC, and the Steering Committee of the study.
Additionally, annual interim analyses for selected questions were submitted to the SAC for discussion and to get advice. Summaries of the annual reports were submitted to interest Drug Regulators only after approval by the SAC.
A publication policy was agreed with the collaborating study centers (see for details).
The study will be analyzed and published in 2005/06.
Discussion
Because confounding and bias can affect every observational study, potential forms of bias received special consideration in the design of the initial study protocol in 2000/2001. Several sources of bias, such as recall bias, selection bias, reverse causality (protopathic) bias, interviewer bias, and (residual) confounding have been addressed in critical reviews of previous case-control studies on the association between analgesics and kidney disease [5,6] and were of concern in the design of this case-control study.
Recall bias assumes a differential recollection of exposures between cases and controls. In order to minimize recall bias, cases and controls are interviewed in their homes to provide for a comparable interview situation. Interviewers are trained to conduct standardized interviews using the same material in cases and controls. Visual and other aids are used to facilitate memory. These include lists of the trade names of analgesics and other drugs which were or are on the market, pictures of packages of different analgesic brands etcetera. Drug use is ascertained according to the calendar method so that a lifetime history of analgesic use can be constructed for each study subject. Interviewers are trained to ask for beacon points in the patients' lives in order to facilitate memory of the time period of drug use.
Although differential recall often tends to lead to inflated risk estimates, it is also of concern in this study that patients with analgesic nephropathy may deny their previous analgesic intake that would bias the risk estimate for analgesics towards the null. Therefore, in the personal interview, questions about medical conditions for which analgesics may have been taken precede questions about analgesic drug use. If the patients report medical conditions for which analgesics may have been used but do not report exposure to analgesics the interviewers have been instructed to ask the patients again about drug use for the respective conditions.
Selection bias: Since it was assumed that cases of ESRD are equally identified among analgesic users and nonusers due to the severity of their medical condition, exposed subjects should not have a higher likelihood of being selected into the study. This could be a problem if cases were defined as patients with early renal disease and analgesic users had more serum creatinine measurements done and therefore a higher likelihood of being detected as a case. For the endpoint ESRD no selection bias due to differential selection of exposed cases is anticipated. Selecting cases with advanced disease has been criticized in previous case-control studies, since the etiologically relevant exposures are those that antedate the onset of disease and not those that antedate end-stage renal disease. To address this concern, several index dates are considered (see reverse causality bias).
The study uses a largely population-based approach (primary study base) by attempting to recruit all regional dialysis units located in these regions (usually federal states), enrolling all incident cases in these units into the study and by selecting population controls from the same regions where the case came from. The representativeness of the control group can be investigated by comparing characteristics in controls to those in the population using official statistics as a source.
A system of motivation and reminder is in place to assure adequate response rates in both cases and controls. In order to avoid non-participation of cases with analgesic nephropathy, the information on the type of question in the study information is subtle and avoids mentioning of analgesics.
Reverse causality (protopathic) bias is also of concern in this study. Chronic renal disease may lead to symptoms or physical changes (for example backache, headache, malaise, intercurrent infections, fever etc), which in turn may lead the patient to use analgesics. Exposure to analgesics in cases of ESRD may therefore reflect exposure which truly started before the chronic renal disease process or which was a consequence of the disease process.
It is very difficult to address this type of bias in a case-control study. To minimize reverse causality bias, several index dates are considered in the statistical analyses. One index date to be considered is the date the renal disease was first suspected/diagnosed (taken from the interview where the patient is asked when the doctor first mentioned an increased serum creatinine level). However, this date may already represent an advanced stage of the disease, since chronic renal disease commonly has an insidious onset and therefore is often not diagnosed in the early stages. Therefore, two other index dates are considered which are based on fixed lag-times of 5 and 10 years. This approach is commonly used in cancer epidemiology and avoids bias by (differential) detection of the disease. Discrepancies in results between index date 5 and 10 years could be indicative of protopathic bias (cf. discussion). Protopathic bias cannot be ruled out if the risk estimate for ESRD in analgesic users is increased with use of the index date with a lag time of 5 but not with a lag time of 10 years., According to recommendations of the SAC, the index date with a lag time of 5 or 10 years is given priority in the analysis to avoid protopathic bias (see above).
Interviewer bias: Although it will not be possible to blind interviewers to the case or control status, interviewers are blinded to the main study hypothesis. The interview does not only ask about kidney disease, but also about a broad range of health outcomes. Furthermore, interviewers acquire information not only about the use of analgesics, but also about exposure to many other drugs as well as exposure to many chemicals at the work place.
Confounding: Previous studies were criticized for confounding by chronic phenacetin use. In this study, chronic phenacetin use will be identified by means of a database that lists the names of all analgesics that contained phenacetin in the past and the respective time periods in which these analgesics contained phenacetin (since the composition of analgesics was often changing in Germany, but the trade name of the drug remained the same). With this database and the calendar method in the interview it will be possible to identify patients with significant phenacetin use and exclude them from testing the main study hypotheses. This study should provide a better opportunity to study the association between phenacetin-free analgesics and chronic renal disease than previous case-control studies, since only patients below age 50 are included. Also, many of the earlier studies were conducted more than 10 years ago when phenacetin use had not been banned long enough to enroll a large enough number of patients without significant phenacetin use. Now, more than 10 years later, this will be more likely possible.
Due to methodological limitations of previous case-control and cohort studies the definition of significant phenacetin use is not straightforward. The cumulative amount of phenacetin that carries an increased risk of chronic kidney disease (and therefore which patients with phenacetin use in the past are to be excluded) is not clear. If the threshold value for significant phenacetin use defined is too high, residual confounding by phenacetin use may lead to an inflated risk estimate for non-phenacetin containing analgesics. Previous studies have shown high risk estimates for ≥ 1 kg cumulative dose of phenacetin: McCredie (1988)[11]: OR = 19 (10–37); Pommer (1989)[10]: OR = 9 (2–39) and also for regular use of phenacetin: Morlans (1990)[12]: OR = 19 (2–157); Sandler (1989)[13]: OR = 5 (1.7–14.9). Some studies have also suggested an increased risk for smaller doses of phenacetin. McCredie observed an odds ratio of 15 (8–28) for ≥ 0.1 kg of phenacetin (compared with < 0.1 kg of phenacetin) and in the study by Pommer [10], a significant increase in risk was observed for cumulative doses of phenacetin of 100–499 g of phenacetin (OR = 1.99 (1.14–3.48)) and 500–999 g (OR = 2.58 (1.19–5.59). No increased risk was observed in this study for cumulative doses of phenacetin of less than 100 g. Sandler[13] also observed an increased odds ratio for lower doses of phenacetin (OR = 1.92 (1.06–3.49) for weekly use of phenacetin defined as use at least once a week for at least one year). To avoid confounding by residual phenacetin use, we will exclude all cases and controls with any detectable phenacetin use from the test of the main study hypotheses.
It is also very important to ascertain information on medical conditions which can independently affect kidney function and lead to an increased use of analgesics such as rheumatic disorders, diabetes, hypertension, kidney stones etc. This study carefully collects information on these conditions and their date of occurrence in the medical documentation form and in the interview in order to be able to control for these conditions in the statistical analyses.
Competing interests
Investigators, drug regulators and the pharmaceutical companies in a common approach initiated this study. The investigators and the SAC exclusively designed the study with some critical remarks from drug regulators and industry. The SAN study is scientifically independent and governed by an independent Advisory Committee. A group of pharmaceutical companies (Boehringer-Ingelheim Pharma GmbH&Co.KG; Dr. Mann Pharma; Whitehall-Much GmbH; Berlin-Chemie AG; Dr. R. Pfleger GmbH; Roche Consumer Health) provided an unconditional grant to cover the costs of the preparatory meetings, the conduction of the study, and the meetings of the advisory committees. The investigators are accountable to the SAC in all scientific matters. None of the investigators have any financial relationship with the group of manufacturers of analgesics.
Authors' contributions
LAJH: Responsible for the epidemiological study design and involved in writing of the paper. EG: Contributed to the study design and to writing & revising the paper. ML: Contributed to the study design particularly data management, analysis plan, and revision of the paper. FvdW: Responsible for the nephrological issues of the study protocol, the collaboration of German nephrological societies and dialysis centres and contributed to writing/revising of the paper. HG: Responsible for the nephrological issues, the collaboration of Austrian nephrological society and dialysis centres, and contributed to writing/revising of the paper.
Appendix 1
The Scientific Advisory Committee (SAC)
Prof. A. R. Feinstein, Dept. Epidemiology, Yale University, Medical School, New Haven, USA: (Chairman until end 2001, died)
Prof. K.-M. Koch, Abt. Nephrologie, Medizinische Hochschule Hannover, Hannover, Deutschland: (Chairman from 2002)
Prof. S. Shapiro, Div. Epidemiology, School of Public Health, Columbia University, New York, USA (from 2002)
Prof. P. Bauer, Institute for Medical Statistics, University of Vienna, Vienna, Austria
Prof. E. Delzell, Occupational Epidemiology, Scholl of Medicine, University Birmingham, USA
Dr. G. Curhan, Channing Laboratory, Harvard University, Boston, USA (until 2004)
Prof. J.M. Fox, University of Saarland, Germany
Prof. P. Michielsen, University of Leuven, Belgium
Prof. S. Suissa, Pharmacoepidemiology Research Unit, Div. Clinical Epidemiology, McGill University, Montreal, Canada
Dr. B. Kasiske, Internal Medicine & Nephrology, Hennepin County Medical Center, University Minneapolis, Minneapolis, USA (from 2002)
Prof. M. Mihatsch, Institute for Pathology, University Basel, Switzerland
The Steering Committee (SC)
Prof. Dr. H. Holzer, Nephrologie, Universitätsklinik, Graz, Austria
Prof. Dr. W. H. Hörl, Medizinische Klinik III am Allgemeinen Krankenhaus der Stadt Wien, Austria
Prim. Doz. Dr. H.K. Stummvoll, Krankenhaus der Elisabethinen, Linz, Austria
Prof. Dr. Florian Lang, Generalsekretär der Gesellschaft für Nephrologie, Tübingen, Germany
Prof. Dr. D. Schlöndorff, Medizinische Poliklinik Klinikum Innenstadt, München, Germany
Prof. Dr. T. Risler, Leiter der Sektion Nephrologie, Medizinische Klinik III, Tübingen, Germany
Prof. Dr. W. Boesken, II. Medizinische Abt., Krankenhaus der Barmherzigen Brüder, Trier, Germany
Prof. Dr. G. Schultze, Dialyse Institut Villingen-Schwenningen, Germany
Study coordination
Germany: Dr. A. Busauschina, Dr. P. Schnülle (Klinikum Heidelberg-Mannheim, 5. Med. Klinik – Nephrologie, Mannheim, Germany
Austria: K. Stadler (2001/2002), G. Schneidewind (from 2002)
International coordinating and data management center
Organization of field work, quality management, and data management: A. Assmann, Dr. S. Möhner, Dr. T. DoMinh (ZEG Berlin),
Data analyses: D. Kuehl-Habich (EPES Berlin)
Advisor for Clinical Pharmacology of analgesics: Dr. K. Farker, Institute for Clinical Pharmacology, Jena, Germany
Statistical advisory group
Prof. Dr. S. Suissa, McGill University Montreal, Canada
Prof. Dr. P. Bauer, University of Vienna, Austria
Prof. Dr. J. Röhmel, German Drug Authority, Bundesinstitut für Arzneimittel und Medizinprodukte, BfArM, Bonn, Germany
Collaborators in dialysis centers
The collaborators in the 170 German and Austrian dialysis centers are listed in the study website .
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
We thank the Nephrological Associations from Germany and Austria for their support to make the study feasible: Oesterreichische Gesellschaft für Nephrologie, Deutsche Gesellschaft für Nephrologie, Deutsche Dialysegesellschaft niedergelassener Ärzte e.V., Deutsche Arbeitsgemeinschaft fuer Klinische Nephrologie.
We would also like to thank the representatives of the German Drug Authority BfArM, Dr. A. Thiele and Dr. K. Menges, for their continuous support and particular their critical review in all phases of the study. We are also grateful for the support coming from the representatives of the Steering Committee of the sponsoring companies, Dr. B. Aicher and Dr. T. Schettler.
We are indebted to the members of the Scientific Advisory Committee scientifically governing the study from the beginning, the members of the Steering Committee, and the coworkers in the study coordinating institutions in Mannheim-Heidelberg, Wien, and Berlin (listed in annex 1).
Figures and Tables
Figure 1 Index dates and relationship with putative analgesic use in the SAN case-control study.
Table 1 Categories of lifetime use of analgesics
Category of analgesics Description
1. Phenacetin-free analgesics All analgesics not containing phenacetin
2. Phenacetin-containing analgesics Analgesics containing phenacetin
3. Mono analgesics (MONOS); phenacetin-free
One analgesic compound, no non-analgesic active agents (except vitamins)
3.1 Paracetamol (acetaminophen)
3.2 ASA
3.3 Ibuprofen
3.4 Metamizol, Diclofenac, Propyphenazon, Indometacin, Aminophenazon, Phenylbutazon)
4. Combination analgesics (COMBIS); phenacetin-free
Combination of more than one active analgesic agent – or – one or more analgesic active agent with one or more non-analgesic agents
4.1 Paracetamol + caffeine (Paracetamol and caffeine, no other analgesic or non-analgesic agent)
4.2 Paracetamol + ASA (Combination of paracetamol and ASA, no other analgesic or non-analgesic agent, no caffeine, no codeine, no barbiturates
4.3 Paracetamol + ASA + Caffeine (Combination of paracetamol and ASA and caffeine, no other analgesic or non-analgesic agent)
4.4 Other monos + Caffeine (One other analgesic agent – except phenacetin or paracetamol with caffeine, no additional non-analgesic agents
4.5 Other combis without caffeine (Two or more analgesic agents – except combination paracetamol + ASA with or without non-analgesic agents, or one analgesic agent with one or more non-analgesic agents, but no caffeine, no codeine, no barbiturates.
4.6 Other combis + caffeine (ASA or paracetamol and/or other analgesic agents and caffeine, no phenacetin, no codeine, no barbiturates)
4.7 Combis with codeine and/or barbiturates (One or more analgesic agents and codeine and/or barbiturates, independent of other agents, i.e. even with additional caffeine, but no phenacetin.
Table 2 Number of cases and controls needed to detect a 2-fold risk (alpha 5%, 1-beta = 80%). Comparison of heavy analgesic use (top tertile in grams) vs. non use in the respective categories of analgesics. Important: the subcategories are not mutually exclusive and cannot be added up to 100%.
Prevalence in controls (%) Nb of cases Nb of controls
All analgesics (phenacetin-free) 22.8 104 416
Phenacetin-containing 2.5 616 2464
MONOS
Paracetamol 5.3 308 1232
ASA 15.7 129 516
Ibuprofen 2.1 728 2912
Other monos 5.0 324 1296
COMBIS
Paracetamol+caffeine 0.1 14 670 58 680
Paracetamol+ASA 0.9 1657 6628
Paracet+ASA+caffeine 4.1 388 1552
Others+caffeine 0.6 2470 9880
Other combis without caffeine 3.5 449 1800
Other combis+caffeine 3.3 475 1900
Combis with codeine or barbiturates 1.9 654 2616
==== Refs
Dubach UC Levy PS Minder F Epidemiological study of analgesic intake and its relationship to urinary tract disorders in Switzerland Helv Med Acta 1968 34 297 312 5704615
Grimlund K ktryckeriet Bo Phenacetin and renal damage at a Swedish factory. Kungl 1963 PA. Norstedt & Söner, Stockholm
Delzell E Shapiro S A review of epidemiologic studies of nonnarcotic analgesics and chronic renal disease Medicine 1998 77 102 21 9556702 10.1097/00005792-199803000-00003
McLaughlin JK Lipworth L Wong-Ho C Blot WJ Analgesic use and chronic renal failure: A critical review of the epidemiologic literature Kidney International 1998 54 679 686 9734593 10.1046/j.1523-1755.1998.00043.x
Feinstein AR Heinemann LAJ Curhan GC Delzell E DeSchepper PJ Fox JM Graf H Luft FC Michielsen P Mihatsch MJ Suissa S Van der Woude F Willich S (ad-hoc review committee) The relationship between non-phenacetin combined analegesics and nephropathy: A review Kidney Internat 2000 58 2259 2264 10.1046/j.1523-1755.2000.00410.x
Hersh EV Moore PA Ross GL Over-the-counter analgesics and antipyretics: a critical assessment Clin Ther 2000 22 500 548 10868553 10.1016/S0149-2918(00)80043-0
Feinstein AR Heinemann LAJ Dalessio D Fox JM Goldstein J Haag G Ladewig D O'Brien CP Do Caffeine-containing analgesics promote dependence? A review and evaluation Clin Pharmacol Ther 2000 68 457 467 11103748 10.1067/mcp.2000.110974
Fox JM Siebers U Caffeine as a promotor of analgesic-associated nephropathy – where is the evidence? Fundamental & Clinical Pharmacology 2003 17 377 92 12803578 10.1046/j.1472-8206.2003.00174.x
Greiser E Molzahn M eds Multizentrische Nieren- und Urothel-Carcinom-Studie (Abschlußbericht). Schriftenreihe der Bundesanstalt für Arbeitsschutz und Arbeitsmedizin Fb 780 1998 Wirtschaftsverlag NW, Bremerhaven
Pommer W Bronder E Greiser E Helmert U Jesdinsky HJ Regular analgesic intake and the risk of end-stage renal failure Am J Nephrol 1989 9 403 12 2801788
McCredie M Stewart JH Mahony JF Is phenacetin responsible for analgesic nephropathy in New South Wales Clin Nephrol 1982 17 134 40 7067175
Morlans M LaPorte J Vidal X Cabeza D Stolley PD End-stage renal disease and non-narcotic analgesics: a case-control study Br J Clin Pharmacol 1990 30 717 723 2271370
Sandler DP Smith JC Weinberg CR Buckalew VM Dennis VW Analgesic use and chronic renal disease N Engl J Med 1989 320 1238 43 2651928
|
16086834
|
PMC1198232
|
CC BY
|
2021-01-04 16:03:43
|
no
|
BMC Nephrol. 2005 Aug 8; 6:9
|
utf-8
|
BMC Nephrol
| 2,005 |
10.1186/1471-2369-6-9
|
oa_comm
|
==== Front
BMC Plant BiolBMC Plant Biology1471-2229BioMed Central London 1471-2229-5-171611783510.1186/1471-2229-5-17Methodology ArticleDesign and fabrication of adjustable red-green-blue LED light arrays for plant research Folta Kevin M [email protected] Lawrence L [email protected] Ryan [email protected] Hyeon-Hye [email protected] J Dustin [email protected] Raymond [email protected] John C [email protected] Horticultural Sciences Department and the Plant Molecular and Cellular Biology Program, University of Florida, Gainesville, USA2 Biological Sciences Office, Space Life Sciences Laboratory, Kennedy Space Center FL, USA3 Dynamac Corporation, Space Life Sciences Laboratory, Kennedy Space Center, FL, USA2005 23 8 2005 5 17 17 31 3 2005 23 8 2005 Copyright © 2005 Folta 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 specific light attributes, such as color and fluence rate, influence plant growth and development, researchers generally cannot control the fine spectral conditions of artificial plant-growth environments. Plant growth chambers are typically outfitted with fluorescent and/or incandescent fixtures that provide a general spectrum that is accommodating to the human eye and not necessarily supportive to plant development. Many studies over the last several decades, primarily in Arabidopsis thaliana, have clearly shown that variation in light quantity, quality and photoperiod can be manipulated to affect growth and control developmental transitions. Light emitting diodes (LEDs) has been used for decades to test plant responses to narrow-bandwidth light. LEDs are particularly well suited for plant growth chambers, as they have an extraordinary life (about 100,000 hours), require little maintenance, and use negligible energy. These factors render LED-based light strategies particularly appropriate for space-biology as well as terrestrial applications. However, there is a need for a versatile and inexpensive LED array platform where individual wavebands can be specifically tuned to produce a series of light combinations consisting of various quantities and qualities of individual wavelengths. Two plans are presented in this report.
Results
In this technical report we describe the practical construction of tunable red-green-blue LED arrays to support research in plant growth and development. Two light fixture designs and corresponding circuitry are presented. The first is well suited for a laboratory environment for use in a finite area with small plants, such as Arabidopsis. The second is expandable and appropriate for growth chambers. The application of these arrays to early plant developmental studies has been validated with assays of hypocotyl growth inhibition/promotion and phototropic curvature in Arabidopsis seedlings.
Conclusion
The presentation of these proven plans for LED array construction allows the teacher, researcher or electronics aficionado a means to inexpensively build efficient, adjustable lighting modules for plant research. These simple and effective designs permit the construction of useful tools by programs short on electronics expertise. These arrays represent a means to modulate precise quality and quantity in experimental settings to test the effect of specific light combinations in regulating plant growth, development and plant-product yield.
==== Body
Background
Light drives the processes of photosynthesis and plant development, and ultimately affects crop yield. The culmination of over a century of plant photobiology research shows that plants possess complicated photosensory networks that monitor and respond to a wide spectrum of ambient light energies. The spectral sensitivity of the plant light-sensors greatly exceeds the range of human vision, as light effects on physiology have been observed from energies arising from the UV-B wavebands [1] into the near infra-red [2]. This broad range of environmental information is processed by integrated signaling networks that tailor growth and development to best fit ambient light conditions.
Light sensing pathways have been dissected through use of narrow-bandwidth light sources. Since individual photoreceptors are generally tuned to sense specific regions of the spectrum, narrow-bandwidth irradiation allows isolation of effects associated with a particular waveband. For instance, the phytochromes mediate responses to red and far-red light, with partial activity extended through the green, blue and near-UV wavebands. Cryptochromes are required for maximal response to blue and UV-A [3], while the phototropins exhibit autophosphorylation when stimulated with light qualities from the blue-green interface (500 nm) to UV-C [4]. Other sensors share sensory overlap with the phototropins, and at least several other receptors remain to be characterized [5].
Narrow-bandwidth, research-quality light is typically generated using a broad spectrum source (such as an incandescent lamp) filtered through an infra-red heat sink and several layers of acetate theatre filters or colored plastic. Fluorescent lamps also are used, as they emit three principle light qualities that can be readily filtered to obtain narrow-bandwidth light. LED light has been used principally in studies of phytochrome reversibility, as switchable red and far-red LED arrays are commercially available. Many reports have demonstrated the utility of red/far-red LED sources in modulating phytochrome responses during de-etiolation [6-8], modulation of root growth [9], root greening [10] and senescence [7]. LED technology has been incorporated into lighting regimes to modulate plant growth and development for decades as acute supplementation of sunlight (such as [11-13]) or as the basis of plant growth in commercial chambers (as in [14]). Additionally, LED-generated light is well-suited for small growth chambers and other applications where a significant fluence rate is required, but little physical space is available for conventional lamps. The practical aspects of LED lighting make them particularly well-suited for space applications where light treatments need to be precise and reliable with little heat radiation and low weight. LEDs may also be especially useful in retrofitting incandescent or fluorescent growth chambers for terrestrial applications.
There is a need for a light source that may facilitate plant growth in a chamber environment, yet still allow for dynamic variation in light quality and quantity for experimental studies. Such devices may reveal interesting interactions between light sensing systems. Recent findings demonstrate that even "benign" wavebands (like green light) have significant influence in plant physiology in concert with red and blue light [15-17]. In this report we detail the design and construction of a compact LED light array for use in plant research. These light sources utilize Norlux hex-LED arrays and are capable of delivering 0 – 150 μmol m-2 s-1 of combined red (R), green (G), and/or blue (B) light. The design allows precise fluence-rate control of individual wavebands, allowing growth of plants under different combinations of light energies. These designs are the same as those used to generate plant-growth data in a number of recent studies of red, blue and green light interaction [17,18]. Two designs are presented in Figures 1, 2, 3, 4. The first is a plan that may be suitable for use in the laboratory or classroom for fewer than one thousand dollars. The second depicts an expandable system that may be appropriate for large growth chambers. Together the two plans presented represent tested and proven designs to introduce efficient lighting to chambers where light quality, quantity, duration and mixture can be readily controlled.
Results and discussion
The advent of new semiconductor technologies has inspired a marked decrease in the price of LED-based devices. An increased number of consumer-grade products have become available to the researcher, and now these new tools may be integrated into various light-research applications. The goal of this work is to provide an interface between research needs and new technology. With this, the best-available research tools may be implemented by researchers without a significant investment in development. The plans presented herein offer two options for LED light source construction, based on the need of an experimental illumination tool or requirement for large-area irradiation.
LED-based lighting regimes are being adopted by municipalities and medical facilities for their consistent, low-power, low-maintenance output. However, one of the most important practical applications of this technology is in the design for lighting regimes to support plant growth. It is of great interest to not only to foster plant growth, but to control plant growth. Basic plant research has demonstrated that specific light wavebands may affect discrete aspects of plant physiology, such as germination [19], stem growth [20], biomass [15,21,22] and the transition to flowering [23]. The supplementation of specific wavebands or skewing of overall spectrum may help modulate the progression of these developmental events. The possibility that combinatorial light regimes may help to optimize growth and control developmental transitions makes the implementation of LED technology particularly attractive to the design of controlled environments targeted to plant production for aesthetic applications, or applications relevant to human nutrition. If spectral quality alone can delay or hasten the floral transition it may have profound effects on modulating the delivery of nursery goods or perhaps affect the availability of consumable produce in a finite, controlled environment. This attribute alone makes LED lighting a compelling platform for specific plant growth routines, such as those proposed for long-term inhabitation of space. Since humans rely specifically on vegetative parts (like stems, leaves and tubers) or reproductive parts (berries and seeds) of plants for nutrition, it is critical to develop systems which impart control of the progression of plant development to affect plant output toward the particular needs of humans.
The implementation of narrow-wavelength LED technology may benefit plant growth schema through supplementation or complete retrofitting of existing chambers. Its compact design may replace existing infrastructure with long-life and consistent output. Here, antiquated lamp systems, replete with toxic, inefficient fluors, may be refitted with efficient light sources that require little to no maintenance with comparable light output. Although previously unattainable without substantial engineering, the geometry of the systems provided in this report brings LED technology to the average plant biology laboratory.
Despite their vast advantages over conventional lighting systems, the LED arrays described in this report offer opportunities for improvement and expansion. Larger installations (e.g. 100 HEX arrays) require close attention to array density, as the fluence rate of RGB HEX LED lights decays significantly toward the edges of the irradiation area. Careful arrangement modified to the application lessens the frequency of "hotspots" or other gradients of light intensities under the light fixture. It is impossible to eliminate all variability under the arrays under all fluence rates and light combinations. The spacing of HEX units in individual systems needs to be carefully tailored for the specific application.
Another potential improvement would be to integrate PAR sensors into the system to provide irradiance feedback and adjust light intensity through a computer-aided regulatory circuit. This would allow the user to enter a specific irradiance value for the desired wavelength and would compensate for changes in LED output that occur over time and with temperature changes in the ambient environment.
This system has been developed using LEDs emitting three principal wavebands. The clear extrapolation is to add additional LED types to generate additional spectrum coverage. LEDs currently manufactured include UV, far-red and infra-red light. From the plans presented within this report it may be possible to develop lighting systems that roughly approximate solar output by compounding the effects of multiple LEDs. Such a system may prove especially valuable in optimizing plant physiology and may have applications to human physiology as well.
The arrays described were tested for support of normal plant developmental responses. These are most conspicuous in early seedling development, as initial responses to light are rapid, robust, and have been well characterized [5,24-26]. Three responses to light, namely inhibition of stem elongation (red or blue activated), stem growth promotion (green activated) and phototropism (blue activated), have been extremely well characterized and may be used to test and verify the utility of these LED arrays on early plant developmental responses. Three assays were conducted. First, end-point stem growth was measured in plants grown under three fluence rates of red or blue light. Red and blue light strongly inhibit early stem elongation through the phytochrome and cryptochrome systems, respectively [20,26]. The results of two independent fluence-rate/response experiments using over 60 seedlings are shown in Figure 5A and 5B. Figure 5A shows the height of seedlings grown for 72 h under constant blue light and Figure 5B shows the effect of the same treatment with red light. The cry1 and phyB mutants are presented as controls. The data show that constant blue or red light inhibit seedling elongation in a manner roughly proportional to fluence rate. Inhibition is detectable even at low fluence rate (<100 μmol m-2 s-1), leading to a strong inhibition of stem growth elongation. The results of these trials mirror the previously published results [27,28], suggesting that the LED arrays described herein act in a manner similar to those used previously.
While phytochromes and cryptochrome effects are salient as stem growth inhibition after days of growth in constant light, other rapid responses involve other light sensors and can be measured on the order of minutes rather than days. Contrary to the effects of red and blue, a short single pulse of green light stimulates rapid elongation of the hypocotyl in the dark-grown seedling [17]. The response persists in all photomorphogenic mutants, it occurs when plants are grown in constant dim red light, and growth promotion is the opposite of what occurs when seedlings are irradiated with red or blue light. This evidence renders it difficult to conveniently ascribe this response to any of the known light sensors, and it is likely being mediated by a separate green-sensitive transduction pathway. Figure 5C, shows the results of 20 independent seedlings treated with green light from the Norlux HEX arrays, compared to previously published data [17]. Seedlings irradiated with a short, single pulse of green light begin to grow rapidly within 15 min, attaining 140% of their dark rate before growth rate declines to dark levels after an hour. The results are highly similar to published findings, again indicating that Norlux HEX arrays are a suitable alternative to other LED or fluorescent light sources.
Phototropism is the rapid curvature of the hypocotyl toward a unilateral light source. In Arabidopsis phototropism is blue light induced and is mediated by the phototropins, autophosphorylating serine-threonine kinases associated with the plasma membrane [29]. The response is exquisitely sensitive to blue light. Here wild-type seedlings were irradiated from the side while being imaged in 5-minute intervals with an infra-red CCD camera (as in [30]). The degree of curvature was monitored over two hours and compared to previous results (Figure 5D). The results indicate that the rapid response of phototropism is very similar between the arrays designed in this study and those produced commercially.
Conclusion
As the availability of new technologies increases, research programs are challenged with the need to retool their capabilities to exploit their potential. Since many of these new technologies are electronic and/or computer related, a certain degree of technical prowess is required to move ideas from the drawing board to application. The development costs of these technologies may be significant. This report offers two clearly applicable models of LED light source development that may be implemented in the study of plant growth and output, helping to narrow the long-term challenges of illumination to support plant growth. These designs allow the fine control of specific wavebands shown to influence plant growth and development. Such designs represent the first step in defining conditions that will optimize, or perhaps even control, plant growth and development. The light sources used in this report sustain normal plant early developmental responses, suggesting that they are an appropriate substitute for, or complement to, other plant illumination solutions.
Methods
The NorLux HEX LEDs (Norlux Corp., Carol Stream, IL) were chosen for these applications because of their compact size and power input to output efficiency. The device is a 90-die, densely-clustered set of LED chips, 30 red (628 nm), blue (470 nm) and green (530 nm) in one device. The HEX platform implements chip-on-board technology which allows high fluence rates to be generated from a relatively small ~2 cm-diameter source. The solid-state design is affixed to a ceramic and steel support to facilitate efficient heat transfer to the mounting substrate. These qualities make the Norlux HEX array useful in applications where high fluence rates need to be generated in minimal space, such as a growth chamber or experimental setting. Most importantly, the RGB components may be independently controlled to finely adjust the quality and quantity of irradiance. Two designs of array controller and array construction are presented. The first is the design of the compact 5-HEX arrays, modular and tunable RGB banks at the University of Florida. These were designed for small spaces and use in studies of stem growth and plant development in the model system, Arabidopsis thaliana. The second design details the construction of the larger light banks of 36 HEX arrays built for studies at Kennedy Space Center. Here, larger light banks are required to assess the effects of light quantity and quality on the growth of specific crop species likely to be grown in space. Complete parts lists for both array designs are presented in Table 1. All materials are common electronics components and may be obtained through local (eg. Radio Shack) or online vendors [31,32].
Design 1 – The tunable RGB banks at the University of Florida
Research activities by this research group at the University of Florida center on developmental transitions, namely the progression of seedling growth between dark and light environments as well as the vegetative/floral transition. These processes are all controlled by a well-defined set of photosensors, responsive to the various portions of the spectrum. The ability to independently control specific wavebands makes it possible to assess how independent light sensing systems integrate environmental information to tailor the developmental response in question. Experiments are performed on small seedlings or throughout the life cycle of Arabidopsis thaliana plants, so an optimal light bank would be self contained, relatively small, and easily moveable. With these considerations in mind, the tunable RGB banks were developed. Each tunable RGB bank consists of 5 HEX arrays (Figure 1), 150 total dies, where the red, blue and green dies are independently adjustable. 20 HEX arrays allow for 4 independent light banks that may be run simultaneously with equal spectral output to irradiate a large area (~1.0 × 0.25 m). Alternatively, each may be independently controlled to illuminate up to four separate spectral quality/quantity conditions at the one time. Two separate photoperiods may be controlled using two independent power supplies and timers.
Control unit
The Control Unit (CU) was designed to support twenty HEX units, four sets of controls to regulate RGB independently on five HEX units. By specification, the Norlux HEX array has a maximum input voltage of 11.7 V and 1.5 A of input current (200 mA per HEX). A circuit was developed where a single potentiometer controls the output from each of the red, blue and green LED arrays in 5 separate HEX units. A single circuit that controls one color in five arrays is shown in Figure 2A. This circuit consists of a standard 25 A power supply delivering 13.8 VDC to a common bus that feeds an LM317 voltage regulator. Voltage to the array is modulated by a potentiometer as well as an in-line switch. A 1 K ohm power resistor is placed in the circuit as a current limiter. The voltage regulator and the LEDs require a stable DC input. Input and output capacitors were added to minimize ripple for improved transient response. A more stable voltage waveform assures consistent output. This simple configuration is repeated for each color. There are twelve individual circuits in the control unit, each controlling R, G, or B in each of four light arrays.
In Design 1, the LM317 voltage regulator is used to ensure an output of 1.5 A, which is important to achieve the maximum light output, yet each generates substantial heat. The voltage regulator has a threshold temperature of 125°C, and the twelve regulators used in this design must be equipped with individual heat sinks. The CU was outfitted with two 80 mm 12 V fans, one facing into, and one facing out of the CU. Each of the individual voltage regulators was outfitted with an individual heat sink to assure adequate cooling.
The RGB bank
Each Tunable RGB Bank consists of five RGB HEX arrays affixed to a 12.5 mm solid aluminum support (Figure 2B and 2C). To allow for modular removal, replacement or service of individual HEX units, the four HEX array connection wires (1 each, RGB anodes and ground) were attached to a standard RJ11 telephone connector. The output of the CU terminates in a 5-input RJ11 coupler where all five HEX arrays could be easily, yet reversibly interfaced.
Thermal issues must be considered in construction of the light source. The LEDs have a maximum temperature rating of 110°C. To maintain full operation without heat damage to components, the NorLux HEX arrays required mounting to a substrate capable of efficient heat transfer and dissipation. In this design, HEX arrays were affixed to solid 18 mm aluminum blocks with heat-sink compound and metal screws (Figure 2B). A black-anodized alloy heat sink was attached to the aluminum block and an 80 mm fan was installed over each unit (Figure 2C). With this configuration the aluminum block temperature did not exceed 45°C. A low operating temperature not only assures consistent LED output, but also is important as radiant heat may perturb plant growth and/or development.
The modular design of the four Tunable LED Banks permits the capacity to control the quantity and quality of actinic light in plant experiments (Figure 3A). In this arrangement up to nine 2.5" (6.35 cm) pots containing rosette plants such as Arabidopsis may be grown under a fluence rate of 75 μmoles m-2 s-1 (+/- 5 μmoles m-2 s-1) at 15 cm vertically below from the light fixture (Figure 3B). Alternatively, seedling hypocotyl growth experiments on Petri dishes may be performed at much closer proximity (5 to 7 cm) where a uniform fluence rate of over 100 μmoles m-2 s-1 is readily obtained.
Design 2 – The large RGB light banks at Kennedy Space Center
Unlike the small 5-array banks described previously, large 36 HEX LED light fixtures were developed to accommodate the uniform irradiation of large flats of plants used in the study of crop production for advanced life support in space. The fixtures were made from polycarbonate and CPVC plastic sheets, aluminum and stainless steel. Like the previous design, the large banks are comprised of three separate units: a power supply/control unit, a timer, and a light fixture.
Control unit, power requirements and capability
The physical layout of the 36-Norlux HEX array control unit is shown in Figure 3C. Each LED component was independently controlled by using a pulse wave modulator (PMW; number 7 in Figure 3C) made by NorLux to control the electronic circuits for each RGB component to the desired irradiance. The controller consisted of a NEMA electrical box housing the power supply, the circuit boards and the PWM with control knobs for adjusting each circuit of HEX LED units. Each set of 30 R, G, or B dies in each HEX array was powered by a single circuit consisting of a TIP 107 PNP transistor, a 22 ohm, 1 watt resistor and a 20 ohm, 1 watt resistor assembled as in Figure 4A. This unit is repeated multiple times, one for each color on each array (Figure 4B). For each color, all of these circuits are placed in parallel and controlled by a single potentiometer. An electronic timer was used to set the photoperiod for the lights with the capability of 4 on/off programs for the complete system. This timer was purchased locally that could satisfy the voltage and ampere load requirement. The electrical power requirements were 110 volts AC and 18.72 amps maximum for all the HEX LEDs. When the HEX LEDs are operating at 100% power they are using approximately 432 watts. The entire system uses approximately 1980 watts at full operating power. At 25°C each RGB HEX LED requires 520 mA total, red 120 mA, green 200 mA, and blue 200 mA.
The RGB 36-array bank
The aluminum, CPVC, and polycarbonate sheets were measured, cut, and assembled to create a light fixture to mount the Norlux HEX LEDs and designed to fit into custom racks within the Advance Life Sciences growth chambers at Kennedy Space Center. The custom light fixture with attached aluminum plate has dimensions of 20" × 20" × 3" (50.8 cm × 50.8 cm × 7.62 cm). The HEX LEDs were mounted directly to 0.125" (3.175 mm) aluminum sheet 20" × 20" (50.8 cm × 50.8 cm) with thermal paste to allow sufficient heat dissipation. The HEX LEDs were evenly distributed on the aluminum sheet (Figures 3D and 3E). Although the aluminum sheet was the primary heat sink for cooling the HEX arrays, secondary aluminum heat sinks were added in a vertical arrangement inside the light fixture for added cooling (Figure 3F). Three cables with 37 conductors (22 AWG) were connected from the light fixture to the circuit boards, one cable for controlling the red components (i.e., 1080 red dies) from 36 HEX LEDs, one for the green components (i.e., 1080 green dies) from 36 HEX LEDs, and one for the blue components (i.e., 1080 blue dies) from 36 HEX LEDs. The light output is approximately 150 μmoles m-2 s-1 at the center and approximately 85 μmoles m-2 s-1 at the edges when measured with a radiometer 30.5 cm vertically below from the light fixture. Fluence rates decreased at the edges, suggesting great care to be exercised in positioning of LED arrays relative to the specific application.
Validation of LED HEX arrays using early photomorphogenic responses
The end-point hypocotyl elongation assays were performed by arranging seed in a line on a square Petri dish placed vertically under the light source. Individual seeds were placed on media containing 1 mM KCl and 1 mM CaCl2, they were stratified at 4°C for 48 h and then were placed into the respective conditions for 72 h. Seedlings were imaged at high resolution and hypocotyl length was measured from digital images using Image Tool (Windows Version 3.0) software. Hypocotyl kinetics in response to green light and phototropic curvature induced by blue light were performed as previously described [17,30].
Authors' contributions
KF designed and fabricated facets of the RGB HEX array electronics and structural elements at the University of Florida (UF), performed all physiological assays and drafted the manuscript. LK designed and fabricated the HEX array platform at Kennedy Space Center (KSC). RM and JDK designed and assembled the circuit board in the UF control unit and engaged in troubleshooting the arrays. H-HK, RW and JS provided conceptual designs and oversaw the practical implementation of the technology at KSC.
Acknowledgements
The Norlux HEX arrays at the University of Florida were purchased with funding from NASA grant NAG10-316 (KMF). The control unit and student technical support were funded by the University of Florida University Scholars Program (RM), with additional support from the National Research Council Resident Research Associate Program (H-HK). This work was supported by the Florida Agricultural Experiment Station and was approved for publication as Journal Series Number R-10919.
Figures and Tables
Figure 1 A simplified diagram of Design 1. Here 20 Norlux HEX arrays are configured into four individual banks of five arrays each. R, G and B are individually controllable on each bank, allowing the use of the controller to regulate light conditions over four independent light quality and quantity conditions. The four banks are regulated by two independent power supplies, actuated by two separate timers. In this configuration two separate photoperiods may be tested for any set of experimental parameters.
Figure 2 A) The schematic of the Tunable LED controller circuit used in Design 1. The controller regulates the output of four sets of five LEDs. The schematic represents the circuit used to power one color (R, G, or B) in five separate HEX chips, and therefore is repeated twelve times in the controller unit in this arrangement. The 5 HEX LED bank is shown from a bottom (B) and top (C) view.
Figure 3 A) The components and organization of the RGB control unit used in Design 1; 1. 12 VDC power supply input bus, 2. voltage regulator array, one regulator to supply one color on five HEX arrays, 3. power resistor array, one for each color on 5 hex arrays, 4. output fan, 5. output bus, 6. potentiometer bus. B) The Design 1 arrays in operation, tuned to an environment without red (left) and without green (right) C) The components of the 36-element control unit; 1. power supply, 2. 12 VDC bus, 3,4,5, red, blue and green circuit boards, 6. on-off switch/breaker, 7. pulse-wave modulator, 8. cooling fan. D) A diagrammatical representation of the 36 HEX LED layout. The position of HEX arrays (green hexagons), wire pass-through holes (red circles), ventilation holes (blue circles), and the vertical heat sinks (black bars) are presented. The complete 36-element array (top; Panel E) and bottom (Panel F).
Figure 4 The circuitry powering the large LED banks. A) Two repeat units, driving two independent sets of 30 dies in Norlux HEX arrays. Q = TIP 170 PnP transistor; R1 = 22 ohm, 1 watt, metal film resistor (+/- 5%); R2 = 20 ohm, 1 watt, metal film resistor (+/- 5%); Rb = 4.7 k ohm, ½ watt, metal film resistor (+/- 5%). B) The assembly of the individual units into a controller for red, blue and green circuits in a 16 HEX array.
Figure 5 Norlux arrays induce typical light responses during early seedling development. A) Blue light fluence-rate response experiments demonstrate the effect of increased fluence rates of blue light on stem growth after 72 h (filled circles; n = 60). The cry1 mutant was tested for comparison (n = 62). B) Fluence-rate/response experiments were performed using red light and demonstrate growth inhibition in wild-type seedlings (filled circles; n = 70) compared to phyB mutants (open circles; n = 60). C) The stem-stimulatory effect of a short single pulse of green light is shown (open circles) and compared to previously published results ([17]; filled circles), and dark growth kinetics (dotted line). D) Phototropic curvature in response to blue light was measured in etiolated seedlings (filled circles; n = 21) and compared to results from previous reports ([30]; open circles). These findings demonstrate how the HEX arrays and the designs presented herein have utility in study of integration of light information from multiple photosensory systems.
Table 1 The complete parts list for both array designs developed.
Design 1 – 4 × 5 RGB Arrays Design 2 – 1 × 36 RGB Arrays
20 – NorLux HEX LED 90 Die, RGB Array, Part Number N9100-0005
4 – 14" × 18" × 3/4" aluminum blocks
4 – alloy heat sinks, approximately 5" × 3" base
2 – 12 VDC, 25 A Regulated Switching Power supply
6 – 3" (80 mm) 12 VDC fan
12 – 1 K ohm potentiometer
12 – LM317 voltage regulator with matching heat sink
12 – SPST switch
4 – 5-input RJ11 coupler on single telephone cable
12 – RJ11 connector
1 – large plastic project box
2 – terminal bus, 1 ea per power supply
2 – 110 VAC standard 24 h timer
12 – 1.0 μF capacitor
12 – 0.1 μF capacitor
12 – 1 Ω power resistor
12 – 1 K Ω resistor
1 – blank circuit board 36 – NorLux HEX LED 90 Die RGB Array, Part Number N9100-0005
1 – 20" × 20" × 1/8" sheet of aluminum
1 – 24" × 24" × 3/8" sheet of gray CPVC plastic
1 – 48" × 48" × 1/4" sheet of clear polycarbonate plastic
1 – 8" × 16" × 6" NEMA waterproof Fiberglas composite electrical box
1 – 15 VDC, 22 A Regulated Switching Power Supply
1 – 25 A 110 VAC breaker
2 – 3" (80 mm) 110 VAC fan
1 – 5" (120 mm) 15 VDC fan
36 – 22 Ω, 2 W composite resistor
72 – 20 Ω, 1 W composite resistor
108 – TIP107 NPN Regulated Transistor
108 – 4.7 K Ω 1 W composite resistor
1 – Pulse Wave Modulator (manufactured by NorLux)
3 – Blank prototype circuit board
Misc. – Connectors, wire and cables, stainless steel nuts, bolts, and screws.
Telephone cable, 2-pair, 22 AWG
Misc. – wire and cable, stainless steel nuts, bolts and screws.
==== Refs
Shinkle JR Atkins AK Humphrey EE Rodgers CW Wheeler SL Barnes PW Growth and morphological responses to different UV wavebands in cucumber (Cucumis sativum) and other dicotyledonous seedlings Physiol Plant 2004 120 240 248 15032858 10.1111/j.0031-9317.2004.0237.x
Johnson CF Brown CS Wheeler RM Sager JC Chapman DK Deitzer GF Infrared light-emitting diode radiation causes gravitropic and morphological effects in dark-grown oat seedlings Photochem Photobiol 1996 63 238 242 11536734
Ahmad M Grancher N Heil M Black RC Giovani B Galland P Lardemer D Action spectrum for cryptochrome-dependent hypocotyl growth inhibition in Arabidopsis Plant Physiol 2002 129 774 785 12068118 10.1104/pp.010969
Knieb E Salomon M Rudiger W Autophosphorylation, Electrophoretic Mobility and Immuno-reaction of Oat Phototropin 1 under UV and Blue Light Photochem Photobiol 2004
Spalding EP Folta KM Illuminating topics in plant photobiology Plant, Cell and Environment 2005 28 39 53 10.1111/j.1365-3040.2004.01282.x
Yadav V Kundu S Chattopadhyay D Negi P Wei N Deng XW Chattopadhyay S Light regulated modulation of Z-box containing promoters by photoreceptors and downstream regulatory components, COP1 and HY5, in Arabidopsis Plant J 2002 31 741 753 12220265 10.1046/j.1365-313X.2002.01395.x
Rousseaux MC Ballare CL Jordan ET Vierstra RD Directed overexpression of PHYA locally suppresses stem elongation and leaf senescence responses to far-red radiation Plant Cell and Environment 1997 20 1551 1558 10.1046/j.1365-3040.1997.d01-51.x
Parks BM Spalding EP Sequential and coordinated action of phytochromes A and B during Arabidopsis stem growth revealed by kinetic analysis Proc Natl Acad Sci U S A 1999 96 14142 14146 10570212 10.1073/pnas.96.24.14142
Kiss JZ Mullen JL Correll MJ Hangarter RP Phytochromes A and B mediate red-light-induced positive phototropism in roots Plant Physiol 2003 131 1411 1417 12644690 10.1104/pp.013847
Usami T Mochizuki N Kondo M Nishimura M Nagatani A Cryptochromes and phytochromes synergistically regulate Arabidopsis root greening under blue light Plant Cell Physiol 2004 45 1798 1808 15653798 10.1093/pcp/pch205
Casal JJ Sanchez RA Deregibus VA The Effect of Plant-Density on Tillering - the Involvement of R/Fr Ratio and the Proportion of Radiation Intercepted Per Plant Environ Exp Bot Environ Exp Bot 1986 26 365 371
Heo JW Lee CW Murthy HN Paek KY Influence of light quality and photoperiod on flowering of Cyclamen persicum Mill. cv. 'Dixie White' Plant Growth Regulation 2003 40 7 10 10.1023/A:1023096909497
Heo J Lee C Chakrabarty D Paek K Growth responses of marigold and salvia bedding plants as affected by monochromic or mixture radiation provided by a Light-Emitting Diode (LED) Plant Growth Regulation 2002 38 225 230 10.1023/A:1021523832488
Markelz NH Costich DE Brutnell TP Photomorphogenic responses in maize seedling development Plant Physiol 2003 133 1578 1591 14645729 10.1104/pp.103.029694
Kim HH Goins GD Wheeler RM Sager JC Green light supplementation for enhanced lettuce growth under red and blue light-emitting diodes Hortscience 2004 39 1617 1622 15770792
Kim HH Goins GD Wheeler RM Sager JC Stomatal conductance of lettuce grown under or exposed to different light quality Annals of Botany 2004 94 91 97 10.1093/aob/mch192
Folta KM Green light stimulates early stem elongation, antagonizing light-mediated growth inhibition Plant Physiol 2004 135 1407 1416 15247396 10.1104/pp.104.038893
Kim HH Wheeler R Sager JC Norkiane J Photosynthesis of Lettuce Exposed to Different Short Term Light Qualities Environmental Control in Biology 2005 43 113 119
Shinomura T Nagatani A Hanzawa H Kubota M Watanabe M Furuya M Action spectra for phytochrome A- and B-specific photoinduction of seed germination in Arabidopsis thaliana Proc Natl Acad Sci U S A 1996 93 8129 8133 8755615 10.1073/pnas.93.15.8129
Parks BM Folta KM Spalding EP Photocontrol of stem growth Curr Opin Plant Biol 2001 4 436 440 11597502 10.1016/S1369-5266(00)00197-7
Went FW The Experimental Control of Plant Growth 1957 Waltham, MA, Chronica Botanica 343
Klein RM Edsall PC Gentile AC Effects of near ultraviolet and green radiations on plant growth Plant Physiol 1965 40 903 906
Valverde F Mouradov A Soppe W Ravenscroft D Samach A Coupland G Photoreceptor regulation of CONSTANS protein in photoperiodic flowering Science 2004 303 1003 1006 14963328 10.1126/science.1091761
Briggs WR Huala E Blue-light photoreceptors in higher plants Annu Rev Cell Dev Biol 1999 15 33 62 10611956 10.1146/annurev.cellbio.15.1.33
Lin C Blue Light Receptors and Signal Transduction Plant Cell 2002 Supplement 2002 S207 S225
Chen M Chory J Fankhauser C Light signal transduction in higher plants Annu Rev Genet 2004 38 87 117 15568973 10.1146/annurev.genet.38.072902.092259
Huq E Tepperman JM Quail PH GIGANTEA is a nuclear protein involved in phytochrome signaling in Arabidopsis Proc Natl Acad Sci U S A 2000 97 9789 9794 10920210 10.1073/pnas.170283997
Lin C Yang H Guo H Mockler T Chen J Cashmore AR Enhancement of blue-light sensitivity of Arabidopsis seedlings by a blue light receptor cryptochrome 2 Proc Natl Acad Sci U S A 1998 95 2686 2690 9482948 10.1073/pnas.95.5.2686
Briggs WR Christie JM Phototropins 1 and 2: versatile plant blue-light receptors Trends Plant Sci 2002 7 204 210 11992825 10.1016/S1360-1385(02)02245-8
Folta KM Leig EJ Durham T Spalding EP Primary inhibition of hypocotyl growth and phototropism depend differently on phototropin-mediated increases in cytoplasmic calcium induced by blue light Plant Physiol 2003
Digikey Inc. www.digikey.com
Mouser Electronics www.mouser.com
|
16117835
|
PMC1198233
|
CC BY
|
2021-01-04 16:03:52
|
no
|
BMC Plant Biol. 2005 Aug 23; 5:17
|
utf-8
|
BMC Plant Biol
| 2,005 |
10.1186/1471-2229-5-17
|
oa_comm
|
==== Front
BMC Womens HealthBMC Women's Health1472-6874BioMed Central London 1472-6874-5-101607899710.1186/1472-6874-5-10Research ArticleCytologic features of nipple aspirate fluid using an automated non-invasive collection device: a prospective observational study Proctor Kerry AS [email protected] Leslie R [email protected] Joel S [email protected] Department of Pathology, University of Utah, Salt Lake City, Utah, USA2 Institute for Clinical and Experimental Pathology, Associated Regional and University Pathologists (ARUP) Laboratories Inc., Salt Lake City, Utah, USA2005 3 8 2005 5 10 10 28 2 2005 3 8 2005 Copyright © 2005 Proctor et al; licensee BioMed Central Ltd.2005Proctor 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
Detection of cytologic atypia in nipple aspirate fluid (NAF) has been shown to be a predictor of risk for development of breast carcinoma. Manual collection of NAF for cytologic evaluation varies widely in terms of efficacy, ease of use, and patient acceptance. We investigated a new automated device for the non-invasive collection of NAF in the office setting.
Methods
A multi-center prospective observational clinical trial involving asymptomatic women designed to assess fluid production, adequacy, safety and patient acceptance of the HALO NAF Collection System (NeoMatrix, Irvine, CA). Cytologic evaluation of all NAF samples was performed using previously described classification categories.
Results
500 healthy women were successfully enrolled. Thirty-eight percent (190/500) produced fluid and 187 were available for cytologic analysis. Cytologic classification of fluid producers showed 50% (93/187) Category 0 (insufficient cellular material), 38% (71/187) Category I (benign non-hyperplastic ductal epithelial cells), 10% (18/187) Category II (benign hyperplastic ductal epithelial cells), 3% (5/187) Category III (atypical ductal epithelial cells) and none were Category IV (unequivocal malignancy). Overall, 19% of the subjects produced NAF with adequate cellularity and 1% were found to have cytologic atypia.
Conclusion
The HALO system is a simple, safe, rapid, automated method for standardized collection of NAF which is acceptable to patients. Cytologic assessment of HALO-collected NAF showed the ability to detect benign and pre-neoplastic ductal epithelial cells from asymptomatic volunteers.
==== Body
Background
The majority of breast cancers originate in the epithelium lining the milk ducts. It is believed that most breast cancers are slow growing and progress from precancerous cells, which have cellular and nuclear changes that can be identified microscopically. Finding microscopic evidence of ductal epithelial atypia/atypical ductal hyperplasia (ADH) has been shown in previous epidemiologic studies to be a predictor of future breast cancer development in an individual woman. [1-10] This increased risk has been identified using random peri-areolar fine needle aspiration (FNA), tissue biopsy or nipple secretion samples for assessment of cytologic atypia.
Nipple fluid or secretions, usually aspirated from the breast ducts, is a protein rich material termed nipple aspirate fluid (NAF) which can be microscopically examined for the presence of atypical ductal epithelial cells. Nipple fluid can be obtained from many women, with reports of NAF production ranging from 25% [11] to more than 95% [12] of women. There are a variety of factors associated with the ability to produce nipple fluid, particularly intrinsic breast characteristics [13]. Nipple fluid acquisition methods are various, including manual breast compression, either followed by manual breast pump or syringe-type device with suction, sometimes repeated up to 10 minutes on each breast.
Breast cancer risk assessment using breast fluid cytology has been suggested to have a role in risk stratification and clinical decision making for women who are at high risk for breast cancer development. Ductal lavage is for clinical use in high-risk women and involves identification and cannulation of one or more fluid-yielding duct(s) then rinsing each with saline, collecting and analyzing the lavaged fluid. The finding of atypical cells could potentially influence a woman's decision for more aggressive surveillance or chemoprevention. Women with atypical ductal hyperplasia in the Breast Cancer Prevention Trial showed an 86% risk reduction with tamoxifen chemoprevention [14].
We analyzed cytologic features of samples obtained during a pilot study using a new suction-based automated mechanical device for the non-invasive collection of NAF in the office setting. This article reports the results from a multi-center prospective observational clinical trial involving asymptomatic women designed to evaluate fluid production, adequacy, safety, patient acceptance and ability to detect atypical breast epithelial cells.
Methods
Study sponsor and study design
The study sponsor was Neomatrix, LLC (Irvine, Ca). The study design and execution was the responsibility of the study sponsor. The author's institution (ARUP Laboratories) provided contracted pathology services for the study sponsor, on a usual and customary fee-for-service basis. Study administration expenses at the parent institution (University of Utah) were paid by the study sponsor. No direct compensation was made to the manuscript authors.
The study was conducted over a one year period and no preparatory phase was incorporated. The participant enrollment sites were obstetric and gynecology clinic practices located in Avon, CT., Farmington, CT., and Baton Rouge, LA. There was a single set-up instructional visit that took less than one hour and was performed prior to initiation of the investigation. The only variation in collection rates was seen between centers. One site had a lower NAF collection rate initially. The HALO System at this site was evaluated and the equipment was found to be operating outside of its specified performance parameters. The equipment was replaced and the variability in NAF collection rates between centers was no longer observed. During the study, enrollment was stopped for approximately 6 months in order to make some minor design modifications to enhance equipment performance.
Patient enrollment
The study population included only asymptomatic, non-pregnant, non-lactating women with no history of breast cancer, breast surgery (e.g. breast augmentation or breast reduction), or nipple piercing who were asked to volunteer as part of a prospective multi-center observational study.
Participants were required to be at least 18 years old and there was no upper age limit. Written informed consent was obtained from all subjects before enrollment in the study. The study protocol was approved for all participating collection centers by Biomedical Research Institute of America (San Diego, CA), an independent Institutional Review Board (Protocol PP-01) and the University of Utah (IRB #11588) Institutional Review Board. All rules and regulations concerning biomedical studies with human subjects were followed. A standardized questionnaire was completed for all participants that included medical history, current medications, family history, obstetric and gynecologic history, and breast health history including any previous biopsies and mammogram results. Gail model 5-year risk profiles were calculated for each woman over the age of 35 .
Nipple aspiration procedure
After obtaining written informed consent and completing a study questionnaire, a clinical breast exam was performed. The subject was wearing a front-opening examination gown and seated in a comfortable position. Each breast was cleaned using an alcohol wipe. Abrasive cleanser was not used. The HALO NAF Collection System (NeoMatrix, Irvine, CA) (Fig 1) has adjustable breast cups with disposable sample collection cups which were placed simultaneously on each breast and manually adjusted to fit snugly around the nipple and areola. The application of topical anesthetic was not required. Occasionally, some breast tissue proximal to the areola was covered by the petals depending on the breast size. After both breast cups are secure, the START button is depressed to initiate the automatic NAF acquisition cycle (Fig. 2). The HALO console initiates the vacuum, providing a gentle suction (similar to that of a breast pump) on both breasts. Simultaneously with suction, heat is applied to the covered areas via circulating warm fluid within the Breast Cup petals. Towards the end of the cycle, the HALO system initiates mild compression of the Breast Cup petals to retrieve any fluid from the ducts. The entire cycle is 5 minutes in duration. The Console indicates when the NAF acquisition cycle has completed. Suction is gently automatically released from the breast cups.
Figure 1 The HALO NAF collection system (photos courtesy of NeoMatrix, Irvine, CA). A. Control Console B. Adjustable Breast Cups with Fluid Reservoir Cassette C. Disposable Sample Collection Cups.
Figure 2 The HALO device applied to a breast. (Photos courtesy of NeoMatrix, Irvine, CA.) An alcohol wipe is used to cleanse the nipples (step 1); cups are placed on the woman's breasts and adjusted to fit (step 2); heat and suction are simultaneously applied via the breast cup petals (step 3); mild compression is initiated by the console (step 4); any sample of NAF obtained is transferred to a vial of cytology fixative and transported to the cytology lab (step 5).
Any collected NAF was transferred from the nipple or sample collection cup(s) to a vial of fixative (CytoLyt, Cytyc Corporation, Boxborough, MA) using a pipette if necessary. If fluid was obtained from either one or both breasts, all samples were combined into a single sample preservative vial. Only one attempt was made to obtain NAF in the five minute session, and if no NAF was produced by either breast, the participant was considered to be a non-producer.
Sample processing and cytologic examination
All samples were shipped to a single reference laboratory (ARUP Laboratories, Inc., Salt Lake City, UT). Microscopic slides were prepared from the entire NAF sample using a Millipore filter technique (Millipore Corp., Billerica, MA), which was chosen due to the low cellularity of the specimens. The filter preps were stained with the modified Papanicolaou stain technique.
Each slide was reviewed by one of a group of three cytopathologists with experience examining breast cytologic specimens, including ductal lavage, and who were trained in the NAF cytologic categories prior to beginning the study. All difficult or borderline cases were resolved by reviewing cases at a multi-headed microscope. The slides were classified according to the most severe abnormality detected using one of five categories based upon increasing nuclear abnormality. The classification system used categories developed by King et al (Table 1). Category 0 was designated for unsatisfactory specimens, with insufficient material for complete evaluation, defined microscopically as having less than 10 ductal epithelial cells. Category I samples contained benign non-hyperplastic ductal epithelial cells. Category II was defined as ductal epithelial hyperplasia with cellular arrangements of cohesive clusters (greater than 10 – 50 cells) and minimal cellular changes including mild nuclear and cellular enlargement and occasional nucleoli but finely granular and evenly distributed chromatin. Category III, atypical hyperplasia, included cells with more distinct nuclear enlargement, increased nuclear to cytoplasmic ratios, coarsely granular chromatin with prominent chromocenters and irregular nuclear membranes with nuclear variation. Increased numbers of atypical single cells were also included in this group. Category IV was defined as unequivocally malignant cells.
Table 1 Nipple aspirate fluid cytology classification*
Classification Characteristics Interpretation
Unsatisfactory (Category 0) <10 ductal epithelial cells. Unsatisfactory specimen.
Benign (Category I) Duct epithelial cells within normal limits. Foam cells. Apocrine metaplastic cells. No malignant cells identified. Benign (non-hyperplastic) ductal epithelial cells present.
Hyperplasia (Category II) Minimal changes including slight cell and nuclear enlargement. Chromatin remains finely granular and evenly distributed. Small and regular nucleoli sometimes present. Cell distribution predominately in groups and cohesive with >10–50 cells (papillary and apocrine subcategories). No malignant cells identified. Benign hyperplastic ductal epithelial cells present.
Atypical Hyperplasia (Category III) Moderate to severe abnormalities with distinct nuclear enlargement, increasing nuclear to cytoplasmic ratio, irregular nuclear borders, and nuclear variation. Coarsely granular chromatin. Prominent chromocenters. Cell distribution in groups with some papillary formations. Increased numbers of single atypical cells (apocrine type subcategory). Atypical hyperplastic ductal epithelial cells present. Malignancy cannot be completely excluded.
Malignancy (Category IV) Single cells and groups of cells with unequivocal nuclear features of cancer. Malignant cells present derived from adenocarcinoma.
* King et al [4]
Data collection and monitoring
Enrollment site investigators and study coordinators filled out case report forms recording relevant information about the subjects' medical history, eligibility, study procedures, adverse events, and any available follow-up care. All clinical sites were monitored and all study data, including final cytology results, where collected by the study sponsor, NeoMatrix.
Patient acceptance and post-procedure survey
Adverse events were noted immediately after the procedure as well as at the four to eight week post-procedure survey. The participant was asked to rate her comfort level immediately following the procedure using a visual analog scale of 1 to 10, with one being most comfortable and 10 being least comfortable. Participants were also contacted four to eight weeks after the procedure to assess satisfaction with the procedure. Women with cytologic atypia or worse (Class III or IV) were referred to their regular physicians who determined the appropriate follow-up care. A standard protocol for following patients with atypia was not included as part of this investigation. This follow-up care may have included further assessment of the patient via imaging, biopsy in some cases, risk counseling, and increased surveillance.
Statistical analysis
The result of NAF producers and non-producers were expressed as raw numbers for each demographic category. Comparison between two groups was performed using the χ2 test of association. The difference between values was considered significant at P < 0.05.
Results
Enrollment demographics
Five hundred (n = 500) participants were successfully enrolled. Overall characteristics of the study participants are summarized in Table 2. The average age was 41.1 years (range 18–65). There was no significant difference between age and fluid production (p <= 1.0). One-hundred and ninety (38%) of the women were fluid producers. Thirty-eight percent (162/426) of women less than 55 years old were fluid producers, while 38% (28/74) of women aged 55 or older produced fluid. Eighty-nine percent (445/500) were Caucasian, 9% (47/500) were African American, 1% (7/500) were Hispanic and one participant was Asian. Forty-eight percent of nulliparous women were fluid producers whereas 36% of parous women produced NAF, which is not statistically significant for the group of pre-menopausal women (p <= 1.0), but is significant if all women are included (p < 0.05). Thirty-nine percent of Caucasians produced NAF while NAF was obtained from 31% of non-white subjects (p <= 1.0). Thirty-six percent of subjects with no 1st degree family history produced NAF, 46% of subjects with one 1st degree relative with breast cancer, and 75% of subjects who had more than one 1st degree relative with breast cancer (p <= 0.10). Forty-two percent of women with a lactation history produced NAF while 34% of women who never lactated were fluid producers (p < 0.10). Overall, 14% had at least one first degree relative with cancer and 11% had a history of a previous breast biopsy. In summary, none of the differences between fluid producers and non-producers with regards to any of the listed demographics was statistically significant.
Table 2 Participant characteristics, demographics and fluid production status
Overall, No. (%) NAF Producers, No. (% of subgroup) p-value
Total No. of Women Enrolled 500 190 (38.0)
Age groups, y, No. (%) p <= 1.0
18–24 63 (12.6) 16 (25.4)
25–34 93 (18.6) 36 (38.7)
35–44 115 (23.0) 45 (39.1)
45–54 155 (31.0) 65 (41.9)
55–64 71 (14.2) 26 (36.6)
65+ 3 (0.6) 2 (66.7)
Parity, No. (%) p <= 0.05 ***
Nulliparous 83 (16.6) 40 (48.2)
Parous 417 (83.4) 150 (36.0)
Age at Menarche, years, No. (%) p <= 1.0
<=12 241 (48.2) 90 (37.3)
13–14 193 (38.6) 77 (39.9)
>=15 61 (12.2) 21 (34.4)
Missing 5 (1.0) 2 (40)
Ethnicity, No. (%) p <= 1.0
Caucasian 445 (89.0) 173 (38.9)
Non-Caucasian* 55 (11.0) 17 (30.9)
1st Degree Relatives with breast cancer, No. (%) p <= 0.1
No 429 (85.8) 156 (36.4)
Yes, 1 67 (13.4) 31 (46.3)
Yes, >=2 4 (0.8) 3 (75.0)
History of breast biopsy, No. (%) p <= 1.0
Yes** 56 (11.2) 22 (39.3)
Menstrual status, No. (%) p <= 1.0
Pre-Menopausal 358 (71.6) 137 (38.3)
Menopausal 142 (28.4) 53 (37.3)
Lactation history, No. (%) p <= 0.1
Never lactated 268 (53.6) 92 (48.4)
History of lactation 232 (46.4) 98 (51.6)
* 47 total African American, 1 Asian, and 7 Hispanic. **type of biopsy not reported *** If only pre-menopausal women are included in the analysis of NAF production vs. parity there is no significant difference (p <= 1.0)
Nipple aspirate fluid analysis
Three of the 190 specimens collected from the fluid producers had a container leak during specimen transport and therefore could not be analyzed, with the remaining 187 available for evaluation. The final cytology results are summarized in Table 3. Fifty percent (93/187) of the NAF samples were classified as Category 0, 38% (71/187) Category I (Figure 3), 10% (18/187) Category II (Figure 4), and 3% (5/187) Category III (Figures 5 and 6). No Category IV (unequivocal malignancy) samples were identified. Statistical analysis whereby all patients 55+ are combined into one group in order to strengthen the raw numbers showed there was no significant difference between age groups or cytologic categories (p = 0.27).
Table 3 Nipple aspirate fluid (NAF) cytologic findings
Cytologic diagnosis No. of women/Total No. fluid producers (%) 18–24 yr, No. (%) 25–34 yr, No. (%) 35–44 yr, No. (%) 45–54 yr, No. (%) 55–64, No. (%) 65 + yrs, No. (%)
Unsatisfactory (Category 0) 93/187 (49.7) 9/93 (9.7) 14/93 (15.1) 22/93 (23.7) 32/93 (34.4) 15/93 (16.1) 1/93 (1.1)
Benign (Category I) 71/187 (38.0) 5/71 (7.0) 15/71 (21.1) 21/71 (29.6) 25/71 (35.2) 5/71 (7.0) 0
Hyperplasia (Category II) 18/187 (9.6) 1/18 (5.6) 6/18 (33.3) 1/18 (5.6) 5/18 (27.8) 5/18 (27.8) 0
Atypical Hyperplasia (Category III) 5/187 (2.8) 1/5 (20.0) 0 1/5 (20.0) 2/5 (40.0) 0 1/5 (20.0)
Malignancy (Category IV) 0/187 (0.0) 0 0 0 0 0 0
Sample Leak 3/187 (1.6) 0 1 0 1 1 0
Figure 3 A-B. Category I. Benign (non-hyperplastic) ductal epithelial cells. The breast ductal epithelial cells are single, small, and uniform (arrow-A). Foam cells are a frequent finding (arrow-B). Apocrine metaplastic cells are sometimes identified. Pap 100X.
Figure 4 A-B. Category II. Benign hyperplasia. The cells are distributed mainly in cohesive groups of 10–50 cells. Minimal cytologic changes are seen including slight cell and nuclear enlargement. The nuclear chromatin is finely granular and evenly distributed and small regular nucleoli are sometimes present. Pap 100X.
Figure 5 A-D. Category III. Atypical hyperplasia. NAF from a 65-year-old Caucasian woman with a family history of breast cancer and an elevated Gail index of 3.0%. NAF analysis reveals moderate to severe cytologic abnormalities including distinct nuclear enlargement, increasing nuclear to cytoplasmic ratio, irregular nuclear borders, and nuclear variation. The chromatin is coarsely granular and there are prominent chromocenters. While the cells are distributed mainly in groups with occasional papillary formations, there are also increased numbers of single atypical cells (arrow-A). After the Category III findings, a ductal biopsy was performed that was found to be benign. A breast biopsy six months later showed DCIS. Pap 100X.
Figure 6 A-D. Category III. Atypical hyperplasia. NAF from a 45-year-old Caucasian woman with a Gail index of 1.1% and no significant medical history. NAF analysis reveals moderate to severe cytologic abnormalities including distinct nuclear enlargement, increasing nuclear to cytoplasmic ratio, irregular nuclear borders, and nuclear variation. The chromatin is coarsely granular and there are prominent chromocenters. Follow-up bilateral biopsies showed LCIS in the right breast, and hyperplastic changes in the left. The changes depicted in these micrographs are not specific for either of those entities and may have originated from other areas of atypia (e.g., DCIS) that was not sampled by the biopsies. Pap 100X.
Procedure acceptance and adverse events
A total of 419/500 (84%) women were surveyed for procedure acceptance four to eight weeks after their procedure. The average comfort assessment rating immediate post-procedure was 5.0 on a scale of 1–10 (one being most comfortable) and 4.2 at the four to eight week telephone/mail post-procedure survey (Table 4). The nipple, areola, and breast areas were visually assessed by the study nurse immediately following the procedure. Twenty-six percent of the participants had no observed skin redness after the procedure, 59% mild redness, 14% moderate redness and less than one per cent had severe redness reported. No major adverse events were reported. Two participants chose to discontinue the procedure mid-cycle due to discomfort and there were five reported minor events including bleeding or small surface lacerations. These were treated with topical ointment, observation, Keflex for one suspected mild mastitis, and Mycolog for candiasis noted in one participant. All resolved without further intervention. Eighty-three percent of the participants reported that they would have the HALO procedure again and 88% said they would recommend the procedure to others.
Table 4 Patient acceptance and adverse events
Initial (n = 500) Post-procedure survey (n = 419)
HALO comfort assessment (scale 0–10) 5.0 4.2
Redness post HALO No. (%) No. (%)
None 130 (26.0) 0
Mild 297 (59.4) 0
Moderate 71 (14.2) 0
Severe 2 (0.4) 0
Would recommend HALO to others N/A 367 (87.6)
Would choose HALO again N/A 349 (83.3)
Nipple aspirate fluid and Gail score
Gail 5 year risk profiles were obtained for the participants over the age of 35. Overall, no statistical difference was seen with regards to fluid production and calculated Gail profile result (p = 0.2). Comparison of Gail risk (>1.7% vs. <1.7%) and cytology category results, for the 190 women assessed, showed no significant difference (p = 0.68).
Post-procedure monitoring
The five women found to have Category III changes were referred for further breast care by their regular physicians. One of these women initially had a benign breast biopsy but was subsequently noted to have ductal carcinoma in situ (DCIS) on a follow-up repeat breast biopsy six months later done as part of an increased surveillance plan as determined by her physician. A second woman had negative bilateral biopsies and the third woman with surgical follow-up had a biopsy that showed lobular carcinoma in situ (LCIS). One of the five Category III women had negative follow-up ultrasound imaging and is being closely followed with mammography every six months. The last of the Category III women, a 24-year-old Caucasian woman with a strong family history of breast cancer in her mother, maternal grandmother and aunt (all diagnosed premenopausally), had negative initial imaging but was subsequently found to have a lump by clinical breast exam. Follow-up ultrasound imaging was again negative. It was recommended that this woman undergo genetic counseling. Follow-up information was not obtained for the Category 0, I or II women and further follow-up of the Category III subjects was not part of this protocol. The Category III women are being followed as high-risk individuals per their physicians' standard protocol outside of this investigation. Only one of these Category III women was identified as being high-risk prior to the NAF collection, using the Gail risk profile calculated at enrollment.
Discussion
Many studies have shown that finding atypical hyperplasia of the breast ductal epithelium is associated with an increased risk of subsequent development of breast cancer [1-7]. Wrensch and colleagues have observed in a prospective trial that NAF production and NAF atypia in a screening population are associated with an increased risk of breast cancer. Further, cytologic assessment of NAF may modestly improve the discriminatory accuracy of the Gail risk model in a screening population [10]. However, NAF collection requires time and experienced trained personnel. We report prospective cytologic examination of NAF collected from otherwise asymptomatic healthy women obtained in a pilot study using the automated HALO System. We found that it is technically feasible to detect normal and atypical breast ductal epithelial cells using routine cytologic preparation methods and a modified classification system, adapted from King et al [4]. Thirty-eight percent of participants produced a NAF sample and of the samples obtained, 50% had adequate ductal epithelial cells for cytology analysis and five asymptomatic women (5/500, 1%) had Category III changes (atypical hyperplasia). Compared to other studies reporting non-invasive NAF collection (Table 5), the percentage of participants who produced fluid using the HALO collection system falls into the range of these previous manual methods (18%–74%). We observed that 19% of the participants had adequate cellularity (defined in this study as greater than 10 ductal epithelial cells present) which is similar to the few studies that recorded cytology results, although there is wide variability (18%–71%).
Table 5 Comparison of non-invasive NAF collection by series
Series Subject Population Fluid Acquisition Method % Fluid Obtained % samples with >10 ductal epithelial cells present for cytologic assessment
Papanicolaou et al 1958 [2] n = 917 asymptomatic women; 19–75 yrs Manual compression followed by manual breast pump 18% Specimen cellularity not specifically reported; ~50% of samples "sparsely cellular with no evidence of atypia"
Petrakis et al 1975 [15] n = 606 healthy volunteers; >18 yrs; Caucasian, Filipina, African American, Mexican, Asian Manual compression & suction 48% Findings reported as secretor or non-secretor; cellularity not reported
Sartorius 1977 [3] n = 203 without breast disease; n = 1503 patients with positive or suspect breast disease Sartorius syringe – device; manual compression & suction 65% (age 31–50); 30–40% (age <20/>60) 48% of 203 without disease; 54% of women with known or suspected breast disease
Buehring 1979 [16] n = 1744 self-selected; mostly asymptomatic volunteers; >18 yrs; Caucasian Sartorius method 49% 36% NAF samples; 18% overall "satisfactory"
Petrakis et al 1981 [17] n = 3929 volunteers from health fairs; >=18 yrs; Caucasian Sartorius method 56% Findings reported as secretor or non-secretor; cellularity not reported
Wynder et al 1981 [18] n = 244 Finnish volunteers; age 20–69 Sartorius method; repeated up to 10 min 38% Cytology not assessed
Wynder et al 1985 [19] n = 990 volunteers; age 30–70 (289 "healthy"'; 548 with benign breast disease; 153 with untreated breast cancer) Sartorius method; repeated up to 10 min 38–57% Cytology not assessed
Wrensch et al 1990 [13] n = 1428 with no history of breast cancer; age 20–74 Sartorius method 37% Findings reported as secretor or non-secretor; cellularity not reported
Wrensch et al 1992 [8] n = 2701 Caucasian volunteers; free from breast cancer Sartorius method 74% 87% NAF samples; 71% overall fluid with satisfactory cytology
Sauter et al 1997 [12] n = 177 non-Asian subjects including women with history of breast cancer, precancerous mastopathy and invasive cancer Modified breast pump consisting of syringe attached to endotracheal tube and respiratory humidification adapter 94–99% 96% NAF samples; 53% sufficiently cellular for DNA analysis
HALO Series 2004 n = 500 healthy volunteers; ages 18–65 yrs; asymptomatic, no breast cancer history Automated five minute cycle (heat, suction, compression) 38% 50% NAF samples; 19% overall produced samples with >10 ductal epithelial cells
Overall, the HALO NAF collection procedure was found to be acceptable by the women studied, rapid, and posed little physical risk. This device has an advantage over other methods of NAF collection in that it is automatic and easy to use, thereby removing most clinician variability. It is also less invasive than other methods of sampling breast epithelium in asymptomatic women, such as ductal lavage and fine needle aspiration.
Criticisms of using NAF to detect cellular abnormalities include the observation that fewer adequately cellular specimens are obtained than with ductal lavage. Dooley et al compared NAF and DL specimens [20] and found that on average a larger number of breast ductal epithelial cells were obtained with DL (13,500) versus nipple aspiration (120) leading to a greater percentage of adequately cellular specimens obtained with DL versus nipple fluid aspiration (78% versus 27% in their study, respectively). In addition, they found abnormal cytologic findings in a greater percentage of DL specimens (24%) than in NAF specimens (9%). In the present study, 38% of the participants produced fluid with the HALO NAF collection system, and of these 50% had specimens that were adequately cellular giving a 19% overall adequacy rate. Even though not all women produce NAF, Wrensch et al [9] showed that women who produce nipple fluid had a slight (1.5 times) increase in the relative risk of breast cancer development and non-producers had a decreased relative risk compared to all fluid producers regardless of final diagnosis. One of Papanicolaou's early studies [1] reported obtaining nipple secretions via breast palpation, massage or a hand-held breast pump, from approximately 50% of the patient population studied; however, this percentage also included women with spontaneous nipple discharge. Secretions were obtained from approximately 19% of asymptomatic women, with pre-menopausal women more likely to produce fluid than post-menopausal women in the Papanicolaou study. Sartorius et al [3] developed a hand held aspiration device, composed of a syringe attached to a small plastic cup placed over the nipple. Their study obtained fluid from approximately 50% of the symptomatic women studied. Wrensch et al [8,9] were able to obtain fluid using a similar device as that described by Sartorius from 40–80% of the women studied, with greater percentages obtained in pre-menopausal women. Krishnamurthy et al [21] obtained fluid from 81% of their study participants; however, these patients had known cancer diagnoses and were under general anesthesia. One pilot study has suggested that administering nasal oxytocin prior to collection can improve the yield of NAF [22].
Despite these limitations, the ease and convenience of this method of obtaining breast ductal epithelial cells might make it a more acceptable option for women who are undergoing NAF collection. As shown in this study, participant acceptance of the procedure was adequate with an average initial comfort assessment rating of 5.0 on a scale of 1–10. Dooley et al [20] reported a lower pain rating for nipple aspiration (8 mm on a 100 mm scale where 0 mm represented "no pain") yet all of the patients in that study underwent either local or general anesthesia. Eighty-three percent of the participants in the current study reported they would have the HALO procedure again and 88% said they would recommend the procedure to others. Since it can be quickly performed in an office setting, patients do not need to be referred to a specialist or scheduled in advance, and can potentially have the test performed at the time of their annual gynecologic exam. The procedure is automated to reduce operator variability and can be performed by non-physician staff.
No prospective studies have been done to date that assess the negative predictive value of ductal cytology in asymptomatic or high risk women. Nipple aspirate fluid assessment is not a diagnostic test for breast cancer. NAF production and cytologic assessment may be used in conjunction with the Gail model for risk prediction and the automated HALO system may facilitate NAF collection in the office setting.
While clinical follow-up care of patients with abnormal breast fluid cytology is not standardized, Shaughnessy et al [23] published an algorithm for management of high-risk women who undergo either nipple aspiration or ductal lavage. Interim management guidelines from the Breast Cancer Risk Assessment Group (BCRAG) [24] also emphasized the need for patients with cytologic abnormalities to undergo a diagnostic work-up to discover potential occult lesion.
NAF specimens have important applications in addition to the cytologic assessments that are described in this paper. A study by Chagpar et al [25] showed that NAF specimens could be tested for the presence of the Thomsen-Friedenreich (TF) antigen, which has been found to be elevated in breast carcinoma. They found a significant difference between healthy breasts and breasts with early-stage carcinoma, even when measured on NAF samples as small as 15 μL. Fackler et al [26] recently reported a new method (quantitative multiplex methylation-specific PCR) that assesses for promoter hypermethylation of DNA in breast ductal epithelial cells, even in samples with as few as 50 cells. This method showed high sensitivity (84%) and specificity (89%) for four genes seen in breast carcinomas. Prostate specific antigen levels in NAF have been shown to decrease with advanced disease stage, larger tumor size, and nodal involvement in women with breast cancer [27]. As additional proteomic, biochemical and genomic biomarkers are identified, testing of NAF samples may become more commonplace.
Conclusion
The HALO procedure has been shown to be a feasible way to obtain NAF samples for cytologic assessment and it appears to be a safe, rapid, fairly well-tolerated, and non-invasive procedure. While fluid production and adequacy rates may not be as high as reported in manual NAF collection series, there may be advantages to using the HALO System over manual collection techniques. The collection cycle is automated, thus removing any clinician variability and allowing women to be consistently, objectively screened routinely to assess any NAF changes. The system is user-friendly, requires minimal training and can be performed by clinical staff. The system is designed for bilateral, simultaneous collection using heat, compression, and suction combined in a single five minute cycle. Further prospective studies with long-term clinical follow-up are necessary to determine the clinical significance of non-producers vs. producers, insufficient samples and other cytologic categories found on NAF samples collected with the HALO System.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
All authors contributing equally to this work. Dr. Proctor is a pathology resident. Dr. Proctor tabulated the final study data and wrote the draft of the manuscript. Ms. Rowe was the cytology data coordinator and research and development scientist. Dr. Bentz was Director of Cytopathology, and responsible for review of study samples. All authors read and approved this manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
Acknowledgements
The authors wish to thank Dr's. C. Jay Marshall and Evelyn V. Gopez for providing expert interpretation of nipple aspirate fluid samples. Thanks also to Sandy Anthony, CT (ASCP) for technical expertise in the processing of study samples, without whom the study would not have gone so smoothly. Thanks go to all the volunteers for this study.
==== Refs
Papanicolaou GN Brader GM Holmquist DG Falls EA Cytologic evaluation of breast secretions Ann N Y Acad Sci 63 1409 21 1956 Mar 30 13314483
Papanicolaou GN Holmquist DG Brader GM Falk EA Exfoliative cytology of the human mammary gland and its value in the diagnosis of cancer and other diseases of the breast Cancer 1958 11 377 409 13511360
Sartorius OW Smith HS Morris P Benedict D Friesen L Cytologic evaluation of breast fluid in the detection of breast disease J Natl Cancer Inst 1977 59 1073 80 903989
King EB Chew KL Petrakis NL Ernster VL Nipple aspirate cytology for the study of breast cancer precursors J Natl Cancer Inst 1983 71 1115 21 6581355
Dupont WD Page DL Risk factors for breast cancer in women with proliferative breast disease N Engl J Med 1985 312 146 51 3965932
Fabian CJ Kimler BF Zalles CM Klemp JR Kamel S Zeiger S Mayo MS Short-term breast cancer prediction by random periareolar fine needle aspiration cytology and the Gail risk model J Natl Cancer Inst 2000 92 1217 21 10922407 10.1093/jnci/92.15.1217
Marshall CJ Schumann GB Ward JH Riding JM Cannon-Albright L Skolnick M Cytologic identification of clinically occult proliferative breast disease in women with a family history of breast cancer Am J Clin Pathol 1991 95 157 65 1992606
Wrensch MR Petrakis NL King EB Miike R Mason L Chew KL Lee MM Ernster VL Hilton JF Schweitzer R Goodson WH IIIHunt TK Breast cancer incidence in women with abnormal cytology in nipple aspirates of breast fluid Am J Epidemiol 1992 135 130 41 1536131
Wrensch MR Petrakis NL Miike R King EB Chew K Neuhaus J Lee MM Rhys M Breast cancer risk in women with abnormal cytology in nipple aspirates of breast fluid J Natl Cancer Inst 2001 93 1791 8 11734595 10.1093/jnci/93.23.1791
Tice JA Miike R Adduci K Petrakis NL King E Wrensch MR Nipple aspirate fluid cytology and the Gail model for breast cancer risk assessment in a screening population Cancer Epidemiol Biomarkers Prev 2005 14 324 8 15734953
Rose DP Lahti H Laakso K Kettunen K Wynder EL Serum and breast duct fluid prolactin and estrogen levels in healthy Finnish and American women and patients with fibrocystic disease Cancer 1986 57 1550 4 2418944
Sauter ER Ross E Daly M Klein-Szanto A Engstrom PF Sorling A Malick J Ehya H Nipple aspirate fluid: a promising non-invasive method to identify cellular markers of breast cancer risk Br J Cancer 1997 76 494 501 9275027
Wrensch MR Petrakis NL Gruenke LD Ernster VL Miike R King EB Hauck WW Factors associated with obtaining nipple aspirate fluid: analysis of 1428 women and literature review Breast Cancer Res Treat 1990 15 39 51 2183892 10.1007/BF01811888
Fisher B Costantino JP Wickerham DL Redmond CK Kavanah M Cronin WM Vogel V Robidoux A Dimitrov N Atkins J Daly M Wieand S Tan-Chiu E Ford L Wolmark N Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 study J Natl Cancer Inst 1998 90 1371 88 9747868 10.1093/jnci/90.18.1371
King EB Barrett D King MC Petrakis NL Cellular composition of the nipple aspirate specimen of breast fluid. I. The benign cells Am J Clin Pathol 1975 64 728 38 1202937
Buehring GC Screening for breast atypias using exfoliative cytology Cancer 1979 43 1788 99 445368
Petrakis NL Ernster VL Sacks ST King EB Schweitzer RJ Hunt TK King MC Epidemiology of breast fluid secretion: association with breast cancer risk factors and cerumen type J Natl Cancer Inst 1981 67 277 84 6943366
Wynder EL Hill P Laakso K Littner R Kettunen K Breast secretion in Finnish women: a metabolic epidemiologic study Cancer 1981 47 1444 50 7226070
Wynder EL Lahti H Laakso K Cheng SL DeBevoise S Rose DP Nipple aspirates of breast fluid and the epidemiology of breast disease Cancer 1985 56 1473 8 4027883
Dooley WC Ljung BM Veronesi U Cazzaniga M Elledge RM O'Shaughnessy JA Kuerer HM Hung DT Khan SA Phillips RF Ganz PA Euhus DM Esserman LJ Haffty BG King BL Kelley MC Anderson MM Schmit PJ Clark RR Kass FC Anderson BO Troyan SL Arias RD Quiring JN Love SM Page DL King EB Ductal lavage for detection of cellular atypia in women at high risk for breast cancer J Natl Cancer Inst 2001 93 1624 32 11698566
Krishnamurthy S Sneige N Thompson PA Marcy SM Singletary SE Cristofanilli M Hunt KK Kuerer HM Nipple aspirate fluid cytology in breast carcinoma Cancer 2003 99 97 104 12704689 10.1002/cncr.10958
Zhang L Shao ZM Beatty P Sartippour M Wang HJ Elashoff R Chang H Brooks MN The use of oxytocin in nipple aspiration fluid Breast J 2003 9 266 268 12846857 10.1046/j.1524-4741.2003.09402.x
O'Shaughnessy JA Ljung BM Dooley WC Chang J Kuerer HM Hung DT Grant MD Khan SA Phillips RF Duvall K Euhus DM King BL Anderson BO Troyan SL Kim J Veronesi U Cazzaniga M Ductal lavage and the clinical management of women at high risk for breast carcinoma: a commentary Cancer 2002 94 292 8 11900214 10.1002/cncr.10238
Hollingsworth AB Singletary SE Morrow M Francescatti DS O'Shaughnessy JA Hartman AR Haddad B Schnabel FR Vogel VG Current comprehensive assessment and management of women at increased risk for breast cancer Am J Surg 2004 187 349 62 15006563 10.1016/j.amjsurg.2003.12.025
Chagpar A Evelegh M Fritsche HA JrKrishnamurthy S Hunt KK Kuerer HM Prospective evaluation of a novel approach for the use of a quantitative galactose oxidase-Schiff reaction in ductal fluid samples from women with breast carcinoma Cancer 2004 100 2549 54 15197795 10.1002/cncr.20311
Fackler MJ McVeigh M Mehrotra J Blum MA Lange J Lapides A Garrett E Argani P Sukumar S Quantitative multiplex methylation-specific PCR assay for the detection of promoter hypermethylation in multiple genes in breast cancer Cancer Res 2004 64 4442 52 15231653
Sauter ER Kein G Wagner-Mann C Diamandis EP Prostate-specific antigen expression in nipple aspirate fluid is associated with advanced breast cancer Cancer Detect Prev 2004 28 27 31 15041074 10.1016/j.cdp.2003.11.003
|
16078997
|
PMC1198234
|
CC BY
|
2021-01-04 16:30:35
|
no
|
BMC Womens Health. 2005 Aug 3; 5:10
|
utf-8
|
BMC Womens Health
| 2,005 |
10.1186/1472-6874-5-10
|
oa_comm
|
==== Front
Cancer Cell IntCancer Cell International1475-2867BioMed Central London 1475-2867-5-281612021910.1186/1475-2867-5-28Primary ResearchResponses of genes involved in cell cycle control to diverse DNA damaging chemicals in human lung adenocarcinoma A549 cells Zhu Huijun [email protected] Catherine [email protected] Charles [email protected] Nigel J [email protected] Molecular Toxicology (Biological Chemistry), Division of Biomedical Sciences, Faculty of Life Sciences, Imperial College London, Sir Alexander Fleming Building London, SW7 2AZ, UK2005 24 8 2005 5 28 28 21 9 2004 24 8 2005 Copyright © 2005 Zhu et al; licensee BioMed Central Ltd.2005Zhu 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 anticancer agents and carcinogens are DNA damaging chemicals and exposure to such chemicals results in the deregulation of cell cycle progression. The molecular mechanisms of DNA damage-induced cell cycle alteration are not well understood. We have studied the effects of etoposide (an anticancer agent), cryptolepine (CLP, a cytotoxic alkaloid), benzo [a]pyrene (BaP, a carcinogenic polycyclic aromatic hydrocarbon) and 2-amino-1-methyl-6-phenylimidazo [4,5-b]pyridine (PhIP, a cooked-meat derived carcinogen) on the expression of cell cycle regulatory genes to understand the molecular mechanisms of the cell cycle disturbance.
Results
A549 cells were treated with DMSO or chemicals for up to 72 h and periodically sampled for cell cycle analysis, mRNA and protein expression. DMSO treated cells showed a dominant G1 peak in cell cycle at all times examined. Etoposide and CLP both induced G2/M phase arrest yet the former altered the expression of genes functioning at multiple phases, whilst the latter was more effective in inhibiting the expression of genes in G2-M transition. Both etoposide and CLP induced an accumulation of p53 protein and upregulation of p53 transcriptional target genes. Neither BaP nor PhIP had substantial phase-specific cell cycle effect, however, they induced distinctive changes in gene expression. BaP upregulated the expression of CYP1B1 at 6–24 h and downregulated many cell cycle regulatory genes at 48–72 h. By contrast, PhIP increased the expression of many cell cycle regulatory genes. Changes in the expression of key mRNAs were confirmed at protein level.
Conclusion
Our experiments show that DNA damaging agents with different mechanisms of action induced distinctive changes in the expression pattern of a panel of cell cycle regulatory genes. We suggest that examining the genomic response to chemical exposure provides an exceptional opportunity to understand the molecular mechanism involved in cellular response to toxicants.
==== Body
Background
Many chemical carcinogens and therapeutic agents interact with cells, leading to temporary/permanent cell growth arrest, genetic modification or cell death. The ultimate effect of a chemical on cells is largely determined by the chemical's ability to elicit genomic response. The recent launch of the National Institutes of Health NCI Chemical Genomics Initiative [1] heralds a new era of chemical-genome research. In the current study, we have used chemical-genomics and phenotypic expression to understand the molecular mechanisms involved in cellular response to chemical exposure.
We have examined four chemicals. Etoposide, a topoisomerase II inhibitor, induces DNA double strand breaks by promoting the formation of cleavable DNA-protein complexes and causes cell cycle arrest at S phase or G2/M phase dependent on the cell type [2,3]. Cryptolepine (CLP), an alkaloid extracted from the West African climbing shrub Cryptolepis sanguinolenta, interferes with topoisomerase II and inhibits DNA synthesis and is potently cytotoxic to tumor cells [4-6]. Benzo(a)pyrene (BaP), one of the polycyclic aromatic hydrocarbons (PAHs) derived from incomplete combustion of organic matter, is an archetypal procarcinogen. Epidemiological studies indicate a positive link between exposure to BaP and the occurrence of human cancers [7-9]. BaP exerts its genotoxicity via cytochrome P450-mediated metabolism, namely CYP1A1 and CYP1B1, to form electrophiles that covalently bind to DNA [10,11]. The expression of CYP1A1 and CYP1B1 can be induced by the activation of aryl hydrocarbon receptor (AhR), which is a ligand-activated transcription factor [12]. Upon binding to its ligands, such as dioxin and PAHs, AhR translocates to the nucleuswhere it complexes with ARNT to stimulate the transcription of genes via transactivation through enhancer domains known as AHR-, dioxin-, or xenobiotic-response elements [13,14]. The cytochrome P450 Cyp1 family, including CYP1B1 and CYP1A1, as well as several phase II detoxification genes are among those regulated by AhR [15,16]. Studies also provide evidence that AHR participates in the modulation of the transcriptional program at least in part by associating with additional transcription factors [17,18]. Such associations may be responsible for the effects of the ligand-activated AHR in the regulation of proliferation [19]. 2-Amino-1-methyl-6-phenylimidazo [4,5-b]pyridine (PhIP), a diet-derived heterocyclic amine formed during the cooking of meat [20], is a rodent carcinogen and suspected human carcinogen. It is known that PhIP and other heterocyclic amines are metabolized chiefly by CYP1A2 but also CYP1A1 and CYP1B1 to form electrophiles that bind to DNA to form DNA adducts [21-23].
In the current study we have used A549 human lung adenocarcinoma cells to examine the cellular and genomic responses to the chosen DNA reactive chemicals. The alveolar epithelial type II cell-derived A549 cells have been extensively used to test the cytotoxicity of therapeutic agents and environment toxicants [24]. These cells express aryl hydrocarbon receptor (AhR) [25], providing a useful model for studying AhR mediated gene regulation. Cell cycle analysis combined with cDNA Microarray assay allowed us to distinguish the molecular mechanisms of the cell cycle disturbance induced by the different chemicals and to more precisely predict the fate of cells after chemical exposure.
Results
Effects of treatments on cell cycle
Cell growth status was examined by microscopy and flow cytometry. Cells treated with DMSO, BaP and PhIP, but not etoposide and CLP, reached confluence after 24 h, as examined under microscope (data not shown). Figure 1 shows that in all DMSO treated samples, a predominant number of cells distributed in G1 phase of the cell cycle. In comparison, treatment with etoposide induced a time dependent decrease in the number of cells in G1 phase and an accumulation of cells in G2/M phase of the cell cycle. Cells treated with CLP displayed a cell cycle profile with elevated G2/M phase peak after 24 h. In comparison, BaP and PhIP had no persistent cell cycle specific effects. At 72 h, the sample treated with etoposide exhibited a nearly 2–3 fold increase in subG1 phase cells, compared with samples treated differently. Consistent with the occurrence of subG1 cells, we also observed more floating cells in the etoposide treated sample compared to the other treatments at 72 h, suggesting that etoposide induced cytoxicity at this late timepoint.
Figure 1 Effects of chemical treatment on cell cycle. A549 cells were treated with DMSO (< 0.1% v/v), etoposide (10 μM), BaP (25 μM), CLP (2.5 μM) or PhIP (50 μM) for the times indicated. Cell cycle distribution was assessed by flow cytometry. The gating represents % cells in each phase of the cell cycle. M1: sub-G1; M2:G0/G1 phase; M3: S phase; M4: G2/M phase.
Changes in the expression of P53 and its target genes in response to chemical treatments
p53 is well-established as a primary responder to cellular and genetic stress [26]. Many of the p53 transcriptional target genes are involved in the cell cycle checkpoints. We have examined the levels of p53 protein together with some of its transcriptional target genes in response to different chemical treatments (table 1, group 1). Figure 2 shows a low level of p53 protein in DMSO treated cells. The expression levels of WAF1 and TGF-β mRNA, the protein products of which are involved in the G1/S transition cell cycle checkpoint [27,28], increased with time in DMSO treated cells (Fig. 3A). This pattern of change in the expression of WAF1 and TGF-β indicates that prolonged culture caused cell cycle arrest at G1 phase, which was supported by the observations that the number of cells in G1 phase was increased after 24 h compared with that at 6 h (Fig, 1). Indeed, after 24 h, cells treated with DMSO formed a uniform confluent monolayer. Exposure to etoposide and CLP induced an accumulation of p53 at 6 h and 24 h (Fig. 2). Consistent with this, WAF1, TGF-β, BAX, MDM2 all showed a positive response to both agents within 24 h at mRNA level (Fig. 3A, 3B). Although the ability for etoposide to induce the accumulation of p53 was largely diminished after 48 h (Fig. 2), exposure to the chemical was still effective in up-regulating WAF1 and MDM2 (Fig. 3A, 3B) Western blotting analysis also confirmed that exposure to etoposide and CLP upregulated the expression of proteins WAF1, MDM2 family members and BAX as early as 6 h (Fig. 2). Exposure to BaP and PhIP, on the other hand, increased the expression of BAX at 24 h (Figs. 2, 3A), but had little effects on any of the other p53 gene targets during the 72 h of treatment compared with the DMSO control (Figs. 2, 3A, 3B), consistent with the inability of these two chemicals to induce the accumulation of p53 protein (Fig. 2). These results suggest that exposure to both etoposide and CLP induced p53 activation, whilst etoposide, BaP and PhIP may also activate p53-independent pathways to regulate the expression of the so-called p53 target genes in A549 cells.
Table 1 Genes and their accession number in Genbank
Genes Access. No Genes Access. No
Group 1 Group 4
Waf U03106 CDC25C M34065
TGF-beta AB000584 CENPf U30872
Bax-delt U19599 NEK2 Z2906
Bax-alpha L22473 BUB1 AF053305
MDM2-D U33202 BUB1B AF053306
MMD2-A U33199 TTK M86699
MDM2-E U33203
Group 2 Group 5
Cyclin D1 M64349 CDC34 L22005
CDK4 U37022 CDC45L AJ223728
Cyclin E M74093 MCM2 D21063
E2F M96577 MCM6 D84557
DP1 L23959 MCM7 D55716
Rb M15400 MCM8 D55083
Group 3 Group 6
BIRC3 U75285 CYP11A M14565
CCNB1 M25753 CYP1A1 X02612
CCNA2 X51688 CYP1B1 U03688
CDKN3 L25876 CYP24 L13286
HSCDC6 U77949 CYP2A7 M33317
MYT1 U56816 CYP2B6 M29874
CYP4B1 J02871
CYP51 U23942
Group 1: P53 target genes; group 2: genes involved in G1/S transition; group 3: genes involved in G2/M phase transition; group 4: genes involved in mitosis; group 5: genes involved in DNA replication initiation; group 6: CYP genes.
Figure 2 Western blots of p53, p53-transcriptional target gene proteins and proteins involved in cell cycle execution. Lysates from A549 cells treated as described in Fig. 1. were subjected to SDS-PAGE, electroblotted onto nitrocellulose and probed for specific immunoreactive proteins.
Figure 3 Effects of chemical treatment on mRNA levels of p53 target genes and cyclin D1 in A549 cells that were treated as described in Fig. 1 were measured using cDNA microarray hybridisation. A. p53 transcriptional targets (I). B. p53 transcriptional targets (II). C. Cyclin D1. GenBank accession number for each gene is displayed in table 1.
Effects on genes involved in G1/S transition
Apart from the p53 target genes WAF1 and TGF-β, we have also examined the expression of other genes involved in the G1-S phase transition (genes in group 2 in table 1). These genes showed no appreciable response to the treatments (data not shown) with the exception of CYCLIN D1, which appeared to be upregulated by exposure to etoposide, CLP and BaP (Fig. 3C). The effects of all chemical treatments on the expression of G1/S transition genes seemed to be insufficient to cause cell cycle arrest in G1 phase.
Effects on genes involved in G2/M transition
In DMSO treated cells, the expression levels of genes involved in G2/M phase transition (Table 1, group 3) were lowest at 72 h (Fig. 4A), suggesting a low growth potential in these cells at this stage. Etoposide persistently inhibited the expression of all these genes. The effects on CYCLIN A (CCNA2) and CYCLIN B1 (CCNB1) were observed as early as 6 h. CLP inhibited the expression of many of these genes within 24 h and all of them by 72 h. Interestingly, exposure to BaP also inhibited the expression of the majority of these genes, although the effects were not observed before 48 h. Exposure to PhIP, on the other hand, upregulated some of these genes at the later timepoints, though the effects were not substantial. These results suggest that etoposide is the most effective agent in inhibiting cell cycle progression through G2/M phase transition. The marginal effect elicited by PhIP indicated a small growth stimulus competing against confluence-related growth inhibition.
Figure 4 Effects of chemical treatment on mRNA levels of genes involved in G2/M transition and mitosis. mRNA levels in A549 cells that were treated as described in Fig. 1 were measured using cDNA microarray hybridisation. A. G2/M genes. B. Mitosis genes. GenBank accession number for each gene is displayed in table 1.
Effects on genes involved in mitosis
The expression of genes involved in mitosis (Table 1, group 4) in cells treated with DMSO was lowest at 72 h, suggesting that the rate of cell division was lowest at this time point (Fig. 4B). Exposure to etoposide repressed the expression of all these genes within 6 h of treatment. In comparison, CLP showed no such inhibition up to 24 h of treatment, although inhibition was evident at 72 h. Exposure to BaP, again, had no effects on the expression of these genes before 24 h, but inhibited the expression of all these genes after 48 h of treatment. In complete contrast, cells treated with PhIP expressed higher levels of these mitosis-related genes at the 72 h time point. This suggests that PhIP either had the potential to delay the growth inhibition initiated by confluence or exerted a mild growth stimulation effect.
Effects on genes involved in DNA replication initiation
In cells treated with DMSO, the expression of genes involved in DNA replication initiation (Table 1, group 5) was highest at 6 h (Fig. 5A). Although the expression of most of these genes had dramatically decreased at 24 h, there was no time-dependent further decrease thereafter. This pattern of change suggests that the process of DNA synthesis initiation was still active when cell growth had reached confluence. Exposure to etoposide, CLP and BaP all caused inhibitory effects on the expression of many of these genes, although the effects of BaP were only apparent after 48 h. Again, in contrast to the other treatments, PhIP appeared to upregulate the expression of many genes in this group at the 48 h and 72 h time points. These results suggest that cells treated with PhIP were more active in initiating DNA synthesis than cells treated with DMSO and with the other three agents, again supporting a growth stimulating effect.
Figure 5 Effects of chemical treatment on expression of genes involved in DNA replication initiation and CYP1B1. mRNA levels in A549 cells that were treated as described in Fig. 1 were measured using cDNA microarray hybridization and protein levels by Western blotting. A. mRNA levels of DNA replication initiation genes. B. mRNA levels of CYP1B1 C. Expression of CYP1B1 protein. GenBank accession number for each gene is displayed in table 1.
Western blotting was also used to examine the expression level of a number of gene products (Fig. 2). Consistent with the data seen at the mRNA level, the protein levels of cyclin A and cyclin B were much lower at 72 h than any other time point in DMSO treated cells (Fig. 2). Exposure to etoposide and CLP decreased the levels of cyclin A and cyclin B, consistent with their effects on the mRNA levels of these proteins. Exposure to BaP, which reduced the mRNA levels of cyclin A and cyclin B at late time points, also reduced the levels of these proteins. Finally and again in contrast to the other treatments, at 72 h PhIP treatment induced the expression of cyclin A, cyclin B and MCM7 proteins (Fig. 2) consistent with a growth stimulatory effect. These results clearly show that changes at mRNA level resulted in changes at protein level. The temporal aspect of altered expression of mRNA to altered expression of protein depends upon the stability of the respective message and protein.
Effect on CYP1B1
The expression level of CYP1B1 in DMSO treated cells was relatively low at 6 h and 24 h but was about 10-fold higher after 48 h (Fig. 5B). Compared with DMSO, exposure to BaP induced a time-dependent upregulation of CYP1B1 from 6 h to 24 h (Fig. 5B). The induction of CYP1B1 protein in A549 cells treated with BaP for 24 h was confirmed by Western blot (Fig. 5C). The ability of BaP to increase the expression of CYP1B1 is consistent with the general perception that this chemical can bind to the AhR, which is the transcriptional factor of CYP1B1. We have confirmed the presence of the AhR protein in this cell line although its expression does not appear to have been affected by the drug treatments (data not shown). We also examined the expression of other CYP genes (group 6 in table 1) and found no appreciable response to the treatments at any of the time points (data not shown).
Discussion
To gain insights into the mechanisms through which chemicals interfere with cell cycle progression, we monitored cell cycle changes by flow cytometry analysis combined with cDNA microarray assay using cells from the same experiments. Cells treated with DMSO for 6 h showed well-defined G1 and G2/M phase peaks and a good proportion of cells distributed in S phase, indicative of proliferation status. The G2/M phase peak was smaller at 24 h and not readily discernable at 48 h and 72 h, indicating that cell growth had reached stationary phase at this late stage of the culture. An increase of cytoxicity was detected in cells treated with etoposide for 72 h. We examined the nature of etoposide-induced cytotoxicity and found no degradation of PARP and caspase-3, well documented features of apoptotic cell death (data not shown).
Although exposure to etoposide appeared to induce cell cycle arrest predominantly at G2/M phase, global gene expression analysis suggested that at the dose used (10 μM) the chemical has the ability to inhibit cell cycle progression at multiple stages. It effectively caused an accumulation of p53 protein, which leads to the up-regulation of its transcriptional gene targets. Included in these gene targets were Waf1 and TGF-β, both protein products are involved in the G1 checkpoints with WAF1 also being involved in G2/M checkpoint. Additionally, etoposide inhibited the expression of many other genes that function in the execution of cell cycle progression through S phase, G2/M transition and mitosis.
Although exposure to CLP induced a cell cycle profile similar to that induced by etoposide at the 24 h, there was a preferential inhibition of the expression of genes involved in G2/M transition. This contrasts with etoposide, which was more effective in inhibiting the expression of genes controlling both G2/M phase and mitosis.
BaP and PhIP treated cells showed no obvious phase specific effect on cell cycle progression over the period of study (72 h). The lack of effect of polycyclic aromatic hydrocarbon carcinogens on cell cycle in cells, including A549 cells, has been reported by others and described as stealth carcinogenesis [29]. Exposure to BaP and PhIP had little, if any, effect on the expression of cell cycle regulatory genes at the 6 h and 24 h time points. However, after prolonged treatment (48 – 72 h), BaP and PhIP induced alterations in the expression of many cell cycle regulatory genes. The effects exerted by BaP and PhIP on gene expression were not associated with dramatic changes in cell cycle profile, which may be due to a delay in the ability of the chemicals to influence transcription by which time the cells had achieved confluence.
p53 protein accumulation often occurs as an indicator of DNA damage [30,31]. The products of p53 transcriptional target genes are known to play important roles in multiple cellular biological processes, including cell cycle checkpoint control [32,33], DNA synthesis [34], DNA damage repair [26] and apoptosis [35]. P53 selectively regulates the expression of its targets in response to certain treatments [36]. Moreover, it has been found that some p53 target genes are also regulated in a p53-independent manner [37,38]. Our experiments in A549 cells showed that etoposide and CLP were almost equally effective in inducing the accumulation of p53 protein at 6 h of treatment, whilst BaP and PhIP failed to do so up to 72 h. We selected a number of p53 target genes to determine whether the dynamics of their expression was altered by the various chemical treatments applied. It was observed that both etoposide and CLP showed similar patterns of effects on the expression of some p53 target genes, including Waf1, MDM2 and TGF-β and Bax, although the etoposide effects were more prominent. The ability of etoposide and CLP to affect the expression of the p53 target genes was not reflected in changes in the expression of these protein products. This was exemplified by the observation that CLP was more effective than etoposide in upregulating the expression of MDM2 protein despite being less effective in upregulating the expression of MDM2 at mRNA level. These results suggest that there must be differences in the post-transcriptional regulation of the expression of MDM2 protein in response to the treatments of etoposide and CLP.
MDM2 acts as an E3 ubiquitin ligase which mediates autoubiquitination and ubiquitination of other proteins including p53 [39]. The balance between auto- and substrate-ubiquitination of MDM2 is modulated physiologically by posttranslational modifications, including sumoylation and phosphorylation. If SUMO conjugates to MDM2, its E3 ligase activity is shifted toward p53, while self-ubiquitination is minimized [40]. The tumour suppressor P19ARF associates with MDM2 to inhibit the ubiquitination, export and subsequent degradation of p53 [41,42]. Given that MDM2 sumoylation is also stimulated significantly by ARF [43], SUMO and ubiquitin modifications appear to be mutually antagonistic. The switch in modification status is stress-responsive, because UV irradiation leads to loss of MDM2 sumoylation [44]. Further study will be needed to investigate whether CLP and etoposide induce different modification of MDM2.
More interestingly, it was observed that p53 protein was much reduced in cells treated with etoposide for 48 h and 72 h, compared with those treated with CLP at the same time points. This suggests that etoposide and CLP influence the expression of p53 by different mechanisms. The fact that etoposide remained effective in altering the expression of the so-called p53 target genes in the absence of p53 protein accumulation suggests that exposure to etoposide may also result in regulation of genes via p53-independent mechanisms. BaP and PhIP, neither of which effected p53 protein, selectively increased the expression of Bax and its protein product at 24 h but had little effect on other p53 regulated genes. Again it seems that exposure to BaP and PhIP can result in regulation of the expression of BAX via a p53-independent mechanism. The p53-independent regulation of BAX has been previously reported [38].
Among the many CYPs, CYP1A1 and CYP1B1 are major enzymes involved in the metabolism of procarcinogen PAHs to their DNA reactive species [45,46]. It has been shown that A549 cells express AhR and both CYP1A1 and CYP1B1 [25] and are able to activate BaP to form DNA adducts [[47] and our unpublished data]. In the present study we found that A549 cells treated with DMSO constitutively express more CYP1B1 mRNA than CYP1A1 mRNA (data not shown) The expression of CYP1B1 increased dramatically after treatment with DMSO for 48 h and 72 h compared with those treated for shorter times yet DMSO is not thought to induce this enzyme. It has been reported that CYP1B1 is a senescence related gene in human cells and in mouse cells [48,49], thus the late increase in the expression of CYP1B1 in DMSO treated cells may be related to confluence-mediated cell growth inhibition. BaP drastically increased the expression of CYP1B1 mRNA at early times with little effect on CYP1A1 mRNA (data not shown), suggesting that in A549 cells CYP1B1 may be the key enzyme in metabolizing BaP. The early upregulation of CYP1B1 in response to exposure to BaP was followed by downregulation of many cell cycle regulatory genes, supporting the proposal that CYP1B1 protein may have an important role in the regulation of cell cycle. Alternatively, the downregulation of key genes for cell cycle progression induced by BaP may be mediated by the interaction of AhR with other transcription factors.
By 72 h, the expression of key cell cycle regulatory genes in DMSO treated cells was lowest (measured at mRNA and protein level), indicating the status of growth inhibition. However many genes, particularly those involved in regulating the cell cycle in mitosis phase and in initiating DNA synthesis, were expressed higher in PhIP treated cells at this late time point, implying that cells treated with PhIP retain the potential to proliferate at confluence phase. At present it is unclear whether the effects of PhIP on cell cycle progression are due to DNA reactivity or to an epigenetic mechanism. However, our results are coincident with the report that PhIP activates the MAP kinase pathways in MCF-10A cells [50] and MCF7 cells [51]. The ability of PhIP to enhance cell growth signals under constraint culture conditions may be important to its carcinogenic properties.
Conclusion
Studying mRNA expression in concert with cellular dynamics provides an effective way of understanding molecular mechanisms of action of chemicals that interfer with cell cycle progression. We have shown that exposure to the DNA intercalating chemicals etoposide and CLP can rapidly induce cell cycle disturbance by inhibiting the expression of multiple cell cycle regulatory genes, whereas the chemical carcinogens BaP and PhIP are insidious and pronounced cellular change requires more prolonged treatment. We conclude that the present work has identified a panel of genes that are responsive to the chosen chemicals and provides evidence supporting the proposal that genomic manipulation with chemicals will influence cellular outcome and that the nature and temporal aspects of the genomic response is predictive of prognosis.
Materials and methods
All reagents were purchased from Sigma-Aldrich (Poole, UK) unless otherwise indicated.
Cell culture and treatments
The human lung adenocarcinoma cell line A549 (European Collection of Cell Cultures, Wiltshire, UK) was cultured in Ham's F12 medium supplemented with l-glutamine (2 mM), 10% fetal bovine serum and 10 μg/ml gentamicin (all from Invitrogen, Paisley, UK) in a humidified incubator at 37°C with 5% CO2. When 70 % confluence was reached, cells were treated with DMSO (< 0.1% v/v), etoposide (10 μM), benzo(a)pyrene (25 μM), cryptolepine (2.5 μM) (gift from Dr Addae-Kyeremeh, University of Science and Technology, Kumasi, Ghana) and PhIP (50 μM) (purchased from Toronto Research Chemicals, Canada), for up to 72 h. All cells were harvested using 0.1% trypsin solution in EDTA.
Flow cytometry
Cellular DNA content was determined by propidium iodide staining flow cytometry as described previously [52].
Western blotting
Cell lysates (10 μg protein) were resolved by SDS-polyacrylamide gel electrophoresis, electroblotted onto nitrocellulose (0.45 μm), and blocked by incubation in 5% nonfat dry milk in phosphate buffered saline for 1 h at room temperature. The nitrocellulose was incubated with primary antibody overnight at 4°C, followed by secondary antibody conjugated to horseradish peroxidase for 1 h at room temperature. Detection was achieved using ECL kit (Amersham Life Science, UK). The antibodies against p53 (sc-6243), Bax (sc-493), cyclin A (sc-751), cyclin B (sc-752) and MCM7 (sc-9966) were purchased from Santa cruz, (California, USA). Antibody against p21/waf1 (CP74) was purchased from Neomarkers (Fremont, USA). Antibody against MDM2 (OP114) was purchased from CN Biosciences (Nottingham, UK).
DNA Microarray Assay
Total RNAs were prepared with the RNeasy Mini Kit (Qiagen), according to manufacture's instruction. Ten μg of total RNA was reverse transcribed to cDNA by using T7-(dT)24 primer and Super Script Double-stranded cDNA Synthesis Kit (Invitrogen). Biotin-labeled cRNA was synthesized from cDNA by using ENZO bioArray High Yield RNA Transcript Labeling Kit (Enzo Diognostics). The cRNA samples were hybridized to human GeneChip arrays containing approximately 12,000 human genes (Human Genome U95A, Affymetrix). All analyses were performed at the Affymetrix core facility (Microarray Centre, Hammersmith Hospital, Imperial College London), in accordance with the MAIME protocol [53]. The average intensity of all genes on each chip was adjusted to 1,500 to allow for comparison and subsequent analysis [54].
Gene grouping
All genes were categorized according to the MAPP files from the GenMAPP database [55]. Due to a failure in sample processing, data at 48 h was not available for agent CLP.
Real-Time quantitative RT-PCR
Total RNA samples from preparations used for the GeneChip hybridisation were also used as templates for Real-time quantitive RT-PCR to confirm the microarray expression data. Reverse transcription was performed using SuperScriptII (Invitrogen). For each RNA sample, three cDNAs and one negative control (no transcriptase) were synthesized by using 2 μg of RNA sample and 0.2 μg/μl random hexamer primer, in total volume of 20 μl. RT-PCR was performed using Applied Biosystems 7900 sequencer. Primers and probes were purchased from Applied Biosystems and designed within an exon for each gene using the Primer Express program, version 1.5 (sequences are available on request). Each reaction contained 200 nM forward and reverse primers, 200 nM probe, 5 μl cDNA template, 12.5 μl Taqman mastermix 2× (Applied Biosystems) and ddH2O q.f to 25 μl. The cycling parameters were an initial 50°C for 2 mins and 95°C 10 mins, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. We used primers and a probe corresponding to the housekeeper gene 18 s rRNA to which we normalized the expression of genes of interest. Data was retrieved from SDS2.0 software (Applied Biosystems) and analysed using the SAS system for Windows: Release 8.01 TS level 01M0.
Validation of the DNA microarray data
We employed real-time PCR to quantify 9 DNA-repair gene targets (Table 2) and used the data to validate the DNA microarray assay. Results of the two assays were compared and the correlation coefficients are presented in table 2. The two assays showed good consistency with an overall correlation of > 0.73.
Table 2 Correlation between DNA repair gene expression determined by cDNA microarray assay and RT-PCR.
Treatment Correlation Coefficient
6 h 24 h
DMSO 0.75 0.70
ETOP 0.76 0.80
BaP 0.69 0.79
CLP 0.62
PHIP 0.71 0.77
The expression of 9 genes, including APEX (M80261), ATM (U26455), Pol-beta (D29013), c-fos (V01512), Gadd45 (M60974), RAP1 (M63488), TPA (M15518), XRCC1 (NM 00639) and XRCC2 (Y08837) were analyzed using real-time quantitative RT-PCR, and the results were compared with that derived from DNA microarray assay.
Abbreviations
BaP, benzo(a)pyrene; DMSO, dimethyl sulphoxide; PAH, polycyclic aromatic hydrocarbon; PhIP, 2-amino-1-methyl-6-phenylimidazo [4,5-b]pyridine; CLP, Cryptolepine; ETOP, etoposide.
Acknowledgements
This work was supported by the Food Standards Agency, UK (T09002).
==== Refs
Garber K NIH launches chemical genetics initiative J Natl Can Inst 2003 95 1740 1741
Knudsen KE Booth D Naderi S Sever-Chroneos Z Fribourg AF Hunton IC Feramisco JR Wang JY Knudsen ES RB-dependent S-phase response to DNA damage Mol Cell Biol 2000 20 7751 63 11003670 10.1128/MCB.20.20.7751-7763.2000
Kao GD McKenna WG Yen TJ Detection of repair activity during the DNA damage-induced G2 delay in human cancer cells Oncogene 2001 20 3486 96 11429695 10.1038/sj.onc.1204445
Bonjean K De Pauw-Gillet MC Defresne MP Colson P Houssier C Dassonneville L Bailly C Greimers R Wright C Quetin-Leclercq J Tits M Angenot L The DNA intercalating alkaloid cryptolepine interferes with topoisomerase II and inhibits primarily DNA synthesis in B16 melanoma cells Biochemistry 1998 37 5136 46 9548744 10.1021/bi972927q
Yang SW Abdel-Kader M Malone S Werkhoven MC Wisse JH Bursuker I Neddermann K Fairchild C Raventos-Suarez C Menendez AT Lane K Kingston DG Synthesis and biological evaluation of analogues of cryptolepine, an alkaloid isolated from the Suriname rainforest J Nat Prod 1999 62 976 83 10425120 10.1021/np990035g
Ansah C Gooderham NJ The popular herbal antimalarial, extract of Cryptolepis sanguinolenta, is potently cytotoxic Toxicol Sci 2002 70 245 51 12441369 10.1093/toxsci/70.2.245
Smith LE Denissenko MF Bennett WP Li H Amin S Tang M Pfeifer GP Targeting of Lung Cancer Mutational Hotspots by Polycyclic Aromatic Hydrocarbons J Natl Cancer Inst 2000 92 803 811 10814675 10.1093/jnci/92.10.803
Kadlubar FF Badawi AF Genetic susceptibility and carcinogen-DNA adduct formation in human urinary bladder carcinogenesis Toxicol Lett 1995 82–83 627 32 10.1016/0378-4274(95)03507-9
Li D Walcott FL Chang P Zhang W Zhu J Petrulis E Singletary SE Sahin AA Bondy ML Genetic and environmental determinants on tissue response to in vitro carcinogen exposure and risk of breast cancer Cancer Res 2002 62 4566 70 12183407
Harvey RG Polycyclic aromatic hydrocarbons: chemistry and carcinogenicity (Cambridge U.K.) 1991 Cambridge University Press
Shimada T Gillam EM Oda Y Tsumura F Sutter TR Guengerich FP Inoue K Metabolism of benzo[a]pyrene to trans-7, 8-dihydroxy-7, 8-dihydrobenzo[a]pyrene by recombinant human cytochrome P450 1B1 and purified liver epoxide hydrolase Chem Res Toxicol 1999 12 623 9 10409402 10.1021/tx990028s
Nebert DW Roe AL Dieter MZ Solis WA Yang Y Dalton TP Role of the aromatic hydrocarbon receptor and [Ah] gene battery in the oxidative stress response, cell cycle control, and apoptosis Biochem Pharmacol 2000 59 65 85 10605936 10.1016/S0006-2952(99)00310-X
Murray GI Melvin WT Greenlee WF Burke MD Regulation, function, and tissue-specific expression of cytochrome P450 CYP1B1 Annu Rev Pharmacol Toxicol 2001 41 297 316 11264459 10.1146/annurev.pharmtox.41.1.297
Hoffman EC Reyes H Chu FF Sander F Conley LH Brooks BA Hankinson O Cloning of a factor required for activity of the Ah (dioxin) receptor Science 1991 252 954 958 1852076
Whitlock JPJ Mechanistic aspects of dioxin action Chem Res Toxicol 1993 6 754 763 8117913 10.1021/tx00036a003
Hankinson O The aryl hydrocarbon receptor complex Annu Rev Pharmacol Toxicol 1995 35 307 340 7598497 10.1146/annurev.pa.35.040195.001515
Sutter TR Greenlee WF Classification of members of the Ah gene battery Chemosphere 1992 25 223 226 10.1016/0045-6535(92)90519-W
Wang F Wang W Safe S Regulation of constitutive gene expression through interactions of Sp1 protein with the nuclear aryl hydrocarbon receptor complex Biochemistry 1999 38 11490 11500 10471301 10.1021/bi982578f
Marlowe JL Knudsen ES Schwemberger S Puga A The Aryl Hydrocarbon Receptor Displaces p300 from E2F-dependent Promoters and Represses S Phase-specific Gene Expression J Biol Chem 2004 279 29013 22 15123621 10.1074/jbc.M404315200
Sugimura T Sato S Mutagens-carcinogens in foods Cancer Res 1983 43 2415s 2421 6682010
Zhao K Murray S Davies DS Boobis AR Gooderham NJ Metabolism of the food derived mutagen and carcinogen 2-amino-1-methyl- 6-phenylimidazo(4,5-b)pyridine (PhIP) by human liver microsomes Carcinogenesis 1994 15 1285 1288 8020169
Crofts FG Sutter TR Strickland PT Metabolism of 2-amino-1- methyl-6-phenylimidazo[4,5-b]pyridine by human cytochrome P4501A1, P4501A2 and P4501B1 Carcinogenesis 1998 19 1969 1973 9855011 10.1093/carcin/19.11.1969
Turesky RJ Lang NP Butler MA Teitel CH Kadlubar FF Metabolic activation of carcinogenic heterocyclic aromatic amines by human liver and colon Carcinogenesis 1991 12 1839 1845 1934265
Hukkanen J Lassila A Päivärinta K Valanne S Sarpo S Hakkola J Pelkonen O Raunio H Induction and regulation of xenobiotic-metabolizing cytochrome P450s in the human A549 Lung adenocarcinoma cell line Am J Respir Cell Mol Biol 2000 22 360 366 10696073
Dohr O Sinning R Vogel C Munzel P Abel J Effect of transforming growth factor-beta1 on expression of aryl hydrocarbon receptor and genes of Ah gene battery: clues for independent down-regulation in A549 cells Mol Pharmacol 1997 51 703 10 9145908
Kastan MB Zhan Q el-Deiry WS Carrier F Jacks T Walsh WV Plunkett BS Vogelstein B Fornace AJ Jr A mammalian cell cycle checkpoint pathway utilizing p53 and GADD45 Is defective In ataxia-telangiectasia Cell 1992 71 587 97 1423616 10.1016/0092-8674(92)90593-2
Massagué J Chen Y Controlling TGF-β signalling Genes & Dev 2000 14 627 644 10733523
Vidal A Koff A Cell-cycle inhibitors: three families united by a common cause Gene 2000 247 1 15 10773440 10.1016/S0378-1119(00)00092-5
Khan QA Anderson LM Hydrocarbon carcinogens evade cellular defense mechanism of G1 arrest in nontransformed and malignant lung cell lines Toxicol Appl Pharmacol 2001 173 105 13 11384212 10.1006/taap.2001.9172
Kastan MB Onyekwere O Sidransky D Vogelstein B Craig RW Participation of p53 protein in the cellular response to DNA damage Cancer Res 1991 51 6304 6311 1933891
Nelson WG Kastan MB DNA strand breaks: the DNA template alterations that trigger p53-dependent DNA damage response pathways Mol Cell Biol 1994 14 1815 1823 8114714
El-Deiry WS Tokino T Velculescu VE Levy DB Parsons R Trent JM Lin D Mercer WE Kinzler KW Vogelstein B WAF1, a potential mediator of p53 tumor suppression Cell 1993 75 817 25 8242752 10.1016/0092-8674(93)90500-P
El-Deiry WS Harper JW O'Connor PM Velculescu VE Canman CE Jackman J Pietenpol JA Burrell M Hill DE Wang Y WAF1/CIP1 is induced in p53-mediated G1 arrest and apoptosis Cancer Res 1994 54 1169 1174 8118801
Shivakumar CV Brown DR Deb S Deb SP Wild-type human p53 transactivates the human proliferating cell nuclear antigen promoter Mol Cell Biol 1995 15 6785 6793 8524244
Miyashita T Reed JC Tumor suppressor p53 is a direct transcriptional activator of the human bax gene Cell 1995 80 293 9 7834749 10.1016/0092-8674(95)90513-8
MacLachlan TK Takimoto R El-Deiry WS BRCA1 directs a selective p53-dependent transcriptional response towards growth arrest and DNA repair targets Mol Cell Biol 2002 22 4280 4292 12024039 10.1128/MCB.22.12.4280-4292.2002
Seoane J Le H Massague J Myc suppression of the p21Cip1 Cdk inhibitor influences the outcome of the p53 response to DNA damage Nature 2002 419 729 734 12384701 10.1038/nature01119
Butt AJ Firth SM King MA Baxter RC Insulin-like growth factor-binding protein-3 modulates expression of Bax and Bcl-2 and potentiates p53-independent radiation-induced apoptosis in human breast cancer cells J Biol Chem 2000 275 39174 39181 10998426 10.1074/jbc.M908888199
Fang S Jensen JP Ludwig RL Vousden KH Weissman AM Mdm2 is a RING finger-dependent ubiquitin protein ligase for itself and p53 J Biol Chem 2000 275 8945 8951 10722742 10.1074/jbc.275.12.8945
Honda R Yasuda H Activity of MDM2, a ubiquitin ligase, toward p53 or itself is dependent on the RING finger domain of the ligase Oncogene 2000 19 1473 1476 10723139 10.1038/sj.onc.1203464
Honda R Yasuda H Association of p19(ARF) with Mdm2 inhibits ubiquitin ligase activity of Mdm2 for tumor suppressor p53 EMBO J 1999 18 22 7 9878046 10.1093/emboj/18.1.22
Stott FJ Bates S James MC McConnell BB Starborg M Brookes S Palmero I Ryan K Hara E Vousden KH Peters G The alternative product from the human CDKN2A locus, p14(ARF), participates in a regulatory feedback loop with p53 and MDM2 EMBO J 1998 17 5001 14 9724636 10.1093/emboj/17.17.5001
Xirodimas DP Chisholm J Desterro JMS Lane DP Hay RT p14ARF promotes accumulation of SUMO-1 conjugated (H)Mdm2 FEBS Lett 2002 528 207 211 12297306 10.1016/S0014-5793(02)03310-0
Buschmann T Lerner D Lee CG Ronai Z The Mdm-2 amino terminus is required for Mdm2 binding and SUMO-1 conjugation by the E2 SUMO-1 conjugating enzyme Ubc9 J Biol Chem 2001 276 40389 95 11384992 10.1074/jbc.M103786200
Shimada T Martin MV Pruess-Schwartz D Marnett LJ Guengerich FP Roles of individual human cytochrome P-450 enzymes in the bioactivation of benzo(a)pyrene, 7,8-dihydroxy-7,8- dihydrobenzo(a)pyrene, and other dihydrodiol derivatives of polycyclic aromatic hydrocarbons Cancer Res 1989 49 6304 6312 2509067
Shimada T Hayes CL Yamazaki H Amin S Hecht SS Guengerich FP Sutter TR Activation of chemically diverse procarcinogens by human cytochrome P-450 1B1 Cancer Res 1996 56 2979 2984 8674051
Feldman G Remsen J Shinohara K Cerutti P Excisability and persistence of benzo(a)pyrene DNA adducts in epithelioid human lung cells Nature 1978 274 796 798 683316 10.1038/274796a0
Kang MK Kameta A Shin KH Baluda MA Kim HR Park NH Senescence-associated genes in normal human oral keratinocytes Exp Cell Res 2003 287 272 278 12837283 10.1016/S0014-4827(03)00061-2
Alexander DL Ganem LG Fernandez-Salguero P Gonzalez F Jefcoate CR Aryl-hydrocarbon receptor is an inhibitory regulator of lipid synthesis and of commitment to adipogenesis J Cell Sci 1998 111 3311 3322 9788873
Venugopal M Agarwal R Callaway A Schut HA Snyderwine EG Inhibition of cell death in human mammary epithelial cells by the cooked meat-derived carcinogen 2-amino-1-methyl-6- phenylimidazo [4,5-b]pyridine Biochem Biophys Res Commun 1999 266 203 7 10581190 10.1006/bbrc.1999.1801
Lauber SN Ali S Gooderham NJ The cooked food derived carcinogen 2-amino-1-methyl-6-phenylimidazo[4,5-b] pyridine is a potent oestrogen: a mechanistic basis for its tissue-specific carcinogenicity Carcinogenesis 2004 25 2509 17 15319301 10.1093/carcin/bgh268
Zhu H Gooderham NJ Neoplastic transformation of human lung fibroblast MRC-5 SV2 cells induced by benzo[a]pyrene and confluence culture Cancer Res 2002 62 4605 4609 12183415
Brazma A Hingamp P Quackenbush J Sherlock G Spellman P Stoeckert C Aach J Ansorge W Ball CA Causton HC Gaasterland T Glenisson P Holstege FCP Kim FI Markowitz V Matese JC Parkinson H Robinson A Sarkans U Schulze-Kremer S Stewart J Taylor R Vilo J Vingron M Minimum information about a microarray experiment (MIAME) – toward standards for microarray data Nat Genetics 2001 29 365 371 11726920 10.1038/ng1201-365
Wodicka L Dong H Mittmann M Ho M Lockhart DJ Genome-wide expression monitoring in Saccharomyces cerevisiae Nat Biotechnol 1997 15 1359 1367 9415887 10.1038/nbt1297-1359
Dahlquist KD Salomonis N Vranizan K Lawlor SC Conklin BR GenMAPP, a new tool for viewing and analyzing microarray data on biological pathways Nat Genetics 2002 31 19 20 11984561 10.1038/ng0502-19
|
16120219
|
PMC1198235
|
CC BY
|
2021-01-04 16:24:16
|
no
|
Cancer Cell Int. 2005 Aug 24; 5:28
|
utf-8
|
Cancer Cell Int
| 2,005 |
10.1186/1475-2867-5-28
|
oa_comm
|
==== Front
Cell Commun SignalCell communication and signaling : CCS1478-811XBioMed Central London 1478-811X-3-101609114810.1186/1478-811X-3-10Researchβ-adrenergic receptor activation in immortalized human urothelial cells stimulates inflammatory responses by PKA-independent mechanisms Harmon Erin B [email protected] Jill M [email protected] James E [email protected] Department of Pharmacology, Physiology, and Therapeutics; University of North Dakota; School of Medicine & Health Sciences; Grand Forks, ND 58202-9037, USA2005 9 8 2005 3 10 10 23 3 2005 9 8 2005 Copyright © 2005 Harmon et al; licensee BioMed Central Ltd.2005Harmon 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
Interstitial cystitis (IC) is a debilitating disease characterized by chronic inflammation of the urinary bladder, yet specific cellular mechanisms of inflammation in IC are largely unknown. Multiple lines of evidence suggest that β-adrenergic receptor (AR) signaling is increased in the inflamed urothelium, however the precise effects of these urothelial cell signals have not been studied. In order to better elucidate the AR signaling mechanisms of inflammation associated with IC, we have examined the effects of β-AR stimulation in an immortalized human urothelial cell line (UROtsa). For these studies, UROtsa cells were treated with effective concentrations of the selective β-AR agonist isoproterenol, in the absence or presence of selective inhibitors of protein kinase A (PKA). Cell lysates were analyzed by radioimmunoassay for generation of cAMP or by Western blotting for induction of protein products associated with inflammatory responses.
Results
Radioligand binding demonstrated the presence of β-ARs on human urothelial UROtsa cell membranes. Stimulating UROtsa cells with isoproterenol led to concentration-dependent increases of cAMP production that could be inhibited by pretreatment with a blocking concentration of the selective β-AR antagonist propranolol. In addition, isoproterenol activation of these same cells led to significant increases in the amount of phosphorylated extracellular signal-regulated kinase (pERK), inducible nitric oxide synthase (iNOS) and the induced form of cyclooxygenase (COX-2) when compared to control. Moreover, preincubation of UROtsa cells with the selective PKA inhibitors H-89 or Rp-cAMPs did not diminish this isoproterenol mediated phosphorylation of ERK or production of iNOS and COX-2.
Conclusion
Functional β-ARs expressed on human urothelial UROtsa cell membranes increase the generation of cAMP and production of protein products associated with inflammation when activated by the selective β-AR agonist isoproterenol. However, the increased production of iNOS and COX-2 by isoproterenol is not blocked when UROtsa cells are preincubated with inhibitors of PKA. Therefore, UROtsa cell β-AR activation significantly increases the amount of iNOS and COX-2 produced by a PKA-independent mechanism. Consequently, this immortalized human urothelial cell line can be useful in characterizing potential AR signaling mechanisms associated with chronic inflammatory diseases of the bladder.
==== Body
Background
Interstitial cystitis (IC) is a debilitating disease characterized by chronic pain in the urinary bladder along with increased urinary frequency and urgency. IC is a complex disease with multiple etiologies, yet inflammatory pain is a common mechanism of all IC symptoms [1]. Prostanoids, arachidonic acid metabolites of the cyclooxygenase (COX) pathway, and nitric oxide (NO), whose formation is catalyzed by nitric oxide synthase (NOS), both play major roles in regulating the inflammatory response. Increased levels of prostaglandins generated by the inducible form of cyclooxygenase (COX-2) mediate the vasodilatation and vascular permeability observed during the early events of inflammation [2]. Moreover, animal models lacking the PGE2 prostaglandin receptor demonstrate a reduced algesic response indicating the importance of prostanoids in the signaling and perception of inflammatory pain [3]. Finally, increased COX-2 expression documented for an in vivo model of cystitis supports the idea that increased prostaglandin signaling sensitizes bladder afferents that control micturition and pain [4].
The expression of the inducible form of NOS, iNOS, has been characterized in numerous cell types as a consequence of the inflammatory processes that follow tissue damage [5]. Large amounts of NO generated by iNOS surpass homeostatic concentrations formed by endothelial eNOS or neuronal nNOS [6]. This difference in kinetics of NO formation by iNOS leads to multiple inflammatory responses that include neutrophil activation, DNA damage, protein nitration and induction of apoptosis [5]. Furthermore, animal models deficient in iNOS establish this enzyme's importance as a pathophysiological mediator of chronic inflammatory diseases [7]. Moreover, increased levels of luminal NO, recognized as a causative agent for bladder excitability and micturition, has been documented in patients with IC, which could represent a mechanism of hyperexcitability documented for this disease [8].
Multiple lines of evidence suggest that increased signaling through the G protein-coupled β-adrenergic receptor (AR) may be linked to inflammation associated with IC. Patients with IC have been found to have increased nerve fiber innervation of the urinary bladder. Further study has shown these fibers to be solely sympathetic nerves, which would correspond to an increase in AR signaling [9]. Moreover, elevated urinary levels of norepinephrine have been found in IC patients, which is also consistent with greater AR activity in the urinary bladder [10]. Finally, genomic profiling found increased transcription of the β2-AR gene in a mouse bladder inflammation model [11]. Together, these observations suggest that chronic β-AR stimulation may be linked to inflammatory bladder diseases like IC. Therefore, we hypothesize that urothelial β-AR activation mediates specific inflammatory responses that can be linked to bladder hyperexcitability and pain documented in chronic inflammatory bladder diseases like IC.
In order to test this hypothesis, we have studied the effects of β-AR activation in an immortalized cell line of human urothelium, UROtsa cells [12]. UROtsa cells exhibit numerous properties of basal bladder epithelial cells, including the potential to differentiate into the stratified cell types found in the mammalian bladder lining. Our results using these UROtsa cells as an in vitro model of bladder urothelium, reveals a correlation between β-AR activation and the production of specific pro-inflammatory proteins via a PKA-independent mechanism.
Results
Identification of specific β-AR binding sites
It was important to determine if specific AR binding sites are expressed on UROtsa cell membranes in order to study specific inflammatory responses that may be linked to β-AR activation. Therefore, radioligand binding analysis was performed to characterize the level of β-AR expression on UROtsa cell membranes. Increasing amounts of the iodinated non-selective β-AR antagonist (-)3-[125I]iodocyanopindolol (125I-CYP) specifically labeled a saturable, homogeneous, high affinity binding site on these UROtsa membranes (Figure 1). The number of binding sites was determined to be 951 ± 43 fmol/mg protein (n = 3), and the equilibrium dissociation constant of 125I-CYP for these binding sites was 32.4 ± 1.1 pM (n = 3). This 125I-CYP equilibrium dissociation constant is similar to the calculated affinity value of this radiochemical when it was used by others to identify β-AR binding sites on cell membranes [13].
Figure 1 Specific β-AR binding sites expressed on UROtsa cell membranes. UROtsa cell membranes were incubated with increasing amounts of 125I-CYP in the absence and presence of 10 μM propranolol to determine total and nonspecific binding, respectively. Specific 125I-CYP binding was plotted versus the amount of free radioligand concentration and fitted to a rectangular hyperbola using non-linear regression analysis. From this curve fit the number of specific 125I-CYP binding sites was determined to be 951 ± 43 fmol/mg protein. In addition, the equilibrium dissociation constant of 125I-CYP for these binding sites was calculated to be 32.4 ± 1.1 pM. Values are presented as the mean ± S.E. for n = 3 experiments performed in duplicate.
Isoproterenol Induces cAMP Accumulation
β-AR activation is classically linked to adenylate cyclase activation. Therefore, to test the functionality of β-ARs expresssed on UROtsa cells, we analyzed levels of cAMP generated after incubation with a selective β-AR agonist. Isoproterenol induced a concentration-dependent increase of cAMP in UROtsa cells (Figure 2). From this data a concentration-response curve was generated and used to calculate a half-maximal isoproterenol concentration (EC50) for generating cAMP in these cells. The calculated EC50 of 170 ± 66 nM (n = 4) for generating cAMP in these cells is similar to results by others when isoproterenol was used to stimulate β-AR production of cAMP [13]. In addition, this UROtsa cell response was abated to basal levels by preincubation with 1 μM of the selective β-AR antagonist propranolol (data not shown). These results together demonstrate that β-ARs are expressed on UROtsa cells and can be stimulated by selective receptor agonists to generate cAMP, which is characteristically linked to this AR type.
Figure 2 β-AR activation stimulates cAMP production in UROtsa cells. UROtsa cells incubated for 30 min with the selective β-AR agonist isoproterenol generated increasing amounts of cAMP in a concentration-dependent manner. From these data the concentration of isoproterenol that generated a half-maximal response (EC50) was calculated using non-linear regression analysis. The EC50 value (ϒ) of isoproterenol to increase cAMP in UROtsa cells was calculated to be 170 ± 66 nM and is presented as the mean ± S.E. for n = 4 experiments performed in duplicate.
Selective Production of Inflammatory Mediators by β-AR Stimulation
Increased sympathetic innervation and elevated urinary catecholamine levels in IC patients implies that chronic bladder inflammation may be linked to stimulation of β-ARs. Therefore, we stimulated UROtsa cells with an effective concentration (100 nM) of isoproterenol and probed the cell lysates for mediators of inflammatory responses using Western analysis. Semi-qualitative analysis demonstrated a significant increase in protein levels for COX-2 and iNOS 2 hr after addition of isoproterenol (Figure 3). Specifically there was a 1.8 ± 0.3 and 2.1 ± 0.3 fold increase in COX-2 (n = 8) and iNOS (n = 9) expression, respectively when compared to basal. Conversely, isoproterenol treatment did not raise levels of the inflammatory cytokines IL-1β, IL-10, and IL-8 (data not shown). In separate experiments, UROtsa cells were also treated with a nonspecific agent of inflammation in order to confirm the inflammatory nature of protein expression observed after isoproterenol addition. Stimulation of UROtsa cells with 300 nM lipopolysaccharide (LPS) induced the expression of both COX-2 and iNOS (n = 3; Figure 3). Maximum expression of COX-2 and iNOS was observed 2 hr after LPS addition, which is similar to the time course observed using isoproterenol. These results demonstrate a selective production of inflammatory mediators by UROtsa cells in response to β-AR activation.
Figure 3 UROtsa cell β-AR activation stimulates production of pro-inflammatory mediators. UROtsa cells were incubated with 100 nM isoproterenol or 300 nM lipopolysaccharide (LPS) in serum-free DMEM for the indicated times. Cells were lysed in modified RIPA and total cell lysates were processed as described under "Methods" for immunoblotting with anti-COX-2 or anti-iNOS antibody to determine β-AR mediated changes in protein expression. As reported by the antibody supplier, COX-2 is identified as a doublet band running at 72–74 kD, and iNOS is detected as a band at 130 kD. Peak expression of the pro-inflammatory mediators COX-2 and iNOS was observed 2 hrs after addition of isoproterenol. Semi-quantitative analysis revealed that there was a significant 1.8 ± 0.3 and 2.1 ± 0.3 fold increase in COX-2 and iNOS expression, respectively when compared to basal. Similarly, a 2 hr incubation with LPS led to increased expression of COX-2 and iNOS when compared to basal. Values are presented as the mean ± S.E. and the autoradiographs are representative immunoblots of n = 3–9 independent UROtsa cell treatments.
β-AR Mediated Activation of MAPK Pathway
Increased production of inflammatory mediators is associated with induction of the mitogen activated protein kinase (MAPK) signal transduction pathways [14]. Stimulation of β-ARs, although classically linked to cAMP accumulation, has also has been shown to activate pro-inflammatory MAPK pathways [15]. To test whether β-AR stimulation activates intercellular MAPK we assayed isoproterenol treated UROtsa cells for levels of pERK using Western analysis. Cells stimulated with an effective isoproterenol concentration (100 nM) showed significant increases in the amount of pERK 5 min after drug addition (Figure 4). Semi-qualitative analysis revealed a 2.4 ± 0.4 fold increase in pERK levels (n = 10) when compared to basal while levels of non-phosphorylated ERK2 remained constant. As a control, UROtsa cells treated with 300 nM LPS showed an increase in ERK phosphorylation after 5 min, which is similar to what was observed using isoproterenol (n = 3; Figure 4). These results demonstrate that a MAPK signal transduction pathway is activated by UROtsa cell β-AR stimulation and that ERK phosphorylation precedes transcriptional induction of COX-2 or iNOS.
Figure 4 β-AR mediated activation of the MAPK pathway. UROtsa cells were incubated with 100 nM isoproterenol or 300 nM LPS for the indicated times, and cell lysates were immunoblotted with anti-pERK to determine the receptor mediated phosphorylation state of ERK or anti-ERK2 to establish the total amount of ERK2 in the samples. As per manufacturer's literature, pERK antibody binds phorphorylated ERK1 and ERK2 in two bands found at 44 and 42 kD, respectively. The ERK2 antibody only binds ERK2 at 42 kD. Peak levels of pERK were observed 5 min after addition of isoproterenol with no significant changes in the total cell lysate levels of ERK2. Densitometric analysis demonstrated a significant 2.4 ± 0.4 fold increase in pERK levels when compared to basal. Similarly, LPS treatment also increased ERK phosphorylation over basal within 5 min without a change in ERK2 production. Values are presented as the mean ± S.E. and the autoradiographs are representative of n = 3–10 independent UROtsa cell treatments.
ERK Phosphorylation is Independent of cAMP Mediated Activation of PKA
Increased levels of cAMP are canonically associated with subsequent activation of protein kinase A (PKA). While our results demonstrated a β-AR mediated rise in cAMP production, it remained unclear whether phosphorylation of ERK was dependent upon this cAMP-dependent activation of PKA. Therefore, we studied the effects of UROtsa cell β-AR stimulation after treatment with selective concentrations of the PKA inhibitors H-89 or Rp-cAMPS. Cell pretreatment in serum-free DMEM containing 100 nM of H-89 or 10 μM of Rp-cAMPS did not significantly alter levels of pERK after stimulation with isoproterenol when compared to cells preincubated in the absence of PKA inhibitors (Figure 5). Increases in pERK when compared to basal were still detected following H-89 or Rp-cAMPS pretreatment with peak levels generated at 5 min. Semi-quantitave analysis of the immunoblots revealed a 1.9 ± 0.4 and 2.6 ± 0.7 fold increase in pERK levels over basal after H-89 (n = 4) or Rp-cAMPS (n = 3) pretreatment, respectively, which are not significantly different than the fold increase for pERK observed without inhibitor pretreatment (2.6 ± 0.9; n = 4). Results of these experiments demonstrate that β-AR stimulated MAPK activation in UROtsa cells is not dependent upon generation of cAMP production and subsequent activation of PKA.
Figure 5 β-AR stimulation activates the MAPK pathway after treatment with selective PKA inhibitors. After a 30 min preincubation with 100 nM of the PKA inhibitors H-89 (panel A) or 10 μM of Rp-cAMPS (panel B), UROtsa cells were stimulated with 100 nM isoproterenol for the indicated times and immunoblotted with anti-pERK or anti-ERK2. Peak levels of pERK were observed within 5 min after the addition of isoproterenol with no changes in the total cell lysate levels of ERK2. Semi-quantitave analysis of the immunoblots revealed a time-dependent increase in ERK phosphorylation (panel C). Five minutes after treatment, isoproterenol significantly induced a 1.9 ± 0.4 and 2.6 ± 0.7 fold increase in pERK levels over basal after H-89 or Rp-cAMPS pretreatment, respectively. These peak values are not significantly different than the fold increase over basal for pERK measured in the absence of PKA inhibitor (2.6 ± 0.9). Values are presented as the mean ± S.E. and the autoradiographs are representative immunoblots of n = 3–4 independent UROtsa cell treatments.
PKA-Independent Production of Inflammatory Mediators
Although previous results revealed that MAPK activation is not dependent upon PKA activation, we were interested in whether or not production of inflammatory mediators initiated by β-AR activation also occurred under PKA-independent mechanisms. Pretreatment of UROtsa cells with selective concentrations of H-89 (100 nM) or Rp-cAMPS (10 μM), again did not affect production of COX-2 or iNOS 2 hrs after addition of isoproterenol, when compared to cells pretreated in the absence of inhibitor (Figure 6). In these experiments, levels of COX-2 generated by β-AR activation in the presence of H-89 (n = 3) or Rp-cAMPS (n = 3) were increased 3.0 ± 0.4 and 2.5 ± 0.7 fold over basal, respectively. This level of COX-2 expression in response to isoproterenol was not significantly different from cells pretreated in the absence of inhibitor (1.9 ± 0.5 fold over basal; n = 5). Likewise, levels of β-AR mediated iNOS production after H-89 (1.8 ± 0.3 fold over basal; n = 5) or Rp-cAMPS pretreatment (2.2 ± 0.6 fold over basal; n = 3) were not significantly different from levels seen in the absence of PKA inhibitors (2.0 ± 0.7 fold over basal; n = 3). These results provide evidence for selective production of inflammatory mediators in UROtsa cells through activation of β-ARs that is independent of PKA.
Figure 6 β-AR stimulated expression of pro-inflammatory mediators occurs through PKA-independent mechanisms. After a 30 min pre-incubation with 100 nM of H-89 (panel A) or 10 μM of Rp-cAMPS (panel B), UROtsa cells were stimulated with 100 nM isoproterenol in serum-free DMEM for the indicated times and immunoblotted with anti-COX-2 or anti-iNOS antibody to determine β-AR mediated changes in protein expression. Peak UROtsa cell expression of the pro-inflammatory mediators COX-2 and iNOS was observed 2 hrs after addition of isoproterenol even after pre-incubation with selective PKA inhibitors (panels C and D). Levels of COX-2 generated in the presence of H-89 or Rp-cAMPS were significantly increased 3.0 ± 0.4 and 2.5 ± 0.7 fold over basal, respectively. However these isoproterenol induced levels of COX-2 were not significantly different from cells pretreated in the absence of inhibitor (1.9 ± 0.5 fold over basal). Likewise, levels (fold over basal) of iNOS production generated by isoproterenol after H-89 (1.8 ± 0.3) or Rp-cAMPS pretreatment (2.2 ± 0.6) were significantly greater than basal. However these responses were not significantly different from levels observed for isoproterenol induced iNOS production in the absence of PKA inhibitors (2.0 ± 0.7). Values are presented as the mean ± S.E. and the autoradiographs are representative immunoblots of n = 3–5 independent UROtsa cell treatments.
PKA-Dependent Phosphorylation of the Cyclic AMP-Responsive Element Binding Protein
Since our results reveal no changes in the amount of ERK phosphorylation or induction of COX-2 and iNOS after pretreatment with selective concentrations of H-89 or Rp-cAMPS, it was necessary to confirm that these PKA inhibitors were being used effectively in our system. Therefore, we examined the isoproterenol mediated phosphorylation state of the cAMP responsive element binding protein (CREB) in the absence and presence of selective concentrations of H-89 or Rp-cAMPS. CREB is a well characterized transcriptional factor, which is activated by cAMP-dependent PKA phosphorylation of specific serine residues [16]. In our experiments, the isoproterenol induced phosphorylation state of CREB was observed over a 2 hr interval (Figure 7). In the absence of PKA inhibitors, isoproterenol increased the phosphorylation state of CREB within 5 min returning to basal after 60 min (n = 3). This time course for CREB phosphorylation is similar to what has been described by others [17]. Pretreatment of UROtsa cells with 100 nM H-89 or 10 μM Rp-cAMPS significantly blocked the isoproterenol mediated CREB phosphorylation (n = 3). These results confirm that selective concentrations of H-89 and Rp-cAMPS used for this study can effectively block a cAMP-dependent PKA phosphorylation process. Furthermore, this finding supports our previous observations that the selective production of inflammatory mediators through induction of β-ARs is unrelated to the cAMP-dependent activation of PKA.
Figure 7 PKA inhibitors H-89 and Rp-cAMPS block phosphorylation of CREB. After a 30 min pre-incubation with serum-free DMEM alone or with 100 nM of H-89 or 10 μM of Rp-cAMPS, UROtsa cells were stimulated with 100 nM isoproterenol in serum-free DMEM for the indicated times and immunoblotted with an antibody specific to CREB phosphorylated at S129 and S133 (pCREB). Without inhibitor pretreatment, isoproterenol induced CREB phosphorylation within 5 min. Significant phosphorylation was still observed 15 min. after drug addition with banding intensity patterns similar to basal after 60 min. Preincubation with H-89 or Rp-cAMPS significantly decreased the phosphorylation of CREB at the 5 and 15 min time points in UROtsa cells, demonstrating the effectiveness of PKA inhibition for these compounds. As reported by the antibody suppliers, pCREB is identified as a band running at 43 kD. The second band seen on the autoradiographs represents a previously-reported alternative splice variant of CREB. Autoradiographs are representative immunoblots of n = 3 independent cell treatments.
Discussion
This study characterizes a novel role of β-AR signaling in urothelial cells that leads to selective induction of protein products associated with inflammatory responses. Our results demonstrate that a previously described human urothelial cell line expressing functional β-ARs increases production of cAMP, phosphorylated ERK and heightened translation of COX-2 and iNOS in response to agonist activation. β-AR stimulation classically precedes cAMP accumulation, which regulates the activity of PKA leading to phosphorylation of PKA-sensitive substrates. However, phosphorylation of ERK and selective production of inflammatory mediators in UROtsa cells occurs independently of PKA activation, as similar results were observed in the presence of two analogous inhibitors specific for this cAMP-dependent kinase. Effective use of these compounds was confirmed by documenting the inhibition of PKA dependent protein phosphorylation in our same model system. Therefore, functional β-ARs present on these human urothelial cells elicit pro-inflammatory responses by a PKA-independent mechanism.
Previous studies by others have demonstrated the link between activation of MAPK pathways and the induction of inflammatory mediators [14]. In these studies, receptor regulated expression of COX-2 and iNOS was dependent upon the intermediary phosphorylation of ERK. Moreover, β-AR activation, although classically linked to generation of cAMP, has been shown in other studies to influence MAPK activation in a PKA-independent manner [18]. These PKA-independent mechanisms associated with β-AR mediated phosphorylation of ERK have been shown to involve β-arrestin scaffolding complexes [18]. In our studies we show that β-AR mediated ERK phosphorylation in UROtsa cells is independent of active cAMP-dependent PKA. Whether other scaffolding complexes caused by β-AR stimulation in these cells are associated with ERK phosphorylation is currently under investigation by our laboratory.
Despite the fact that a specific etiology has yet to be identified, inflammatory pain is a common mechanism associated with the symptoms of IC [1]. With reference to our human urothelial cell model, we demonstrate an induction of mediators associated with inflammatory pain and bladder hyperexcitability in response to β-AR activation. Clinical correlations have recognized an increased sympathetic innervation as well as elevated catecholamine levels in IC patients when compared to controls [9,10]. Our studies suggest that chronic urothelial β-AR stimulation in these patients may induce COX-2 and iNOS leading to the increased progression of inflammatory pain and bladder hyperexcitability associated with this disease. Induction of COX-2 by bacterial lipopolysaccharide or endogenous cytokines has been shown to elevate prostanoid levels that are linked to the increased vasodilatation, vascular permeability and hyperalgesic responses of inflammation [2]. In other models, induction of iNOS by these same agonists to generate nitric oxide contributes to the nociceptive processing of inflammatory pain [19]. Therefore, we suggest that chronic urothelial β-AR stimulation leading to increased levels of prostaglandins and NO is one potential mechanism of inflammatory pain in IC. Moreover, higher levels of prostanoids and NO may also contribute to the symptomatic increases in urinary frequency and urgency diagnosed in patients with IC [4,8].
Support of this hypothesis has been reported using a mouse model of bladder inflammation in which genes encoding for iNOS and the β2-AR subtype were upregulated when compared to control [11]. Interestingly, a significant increase in genomic expression of the β2-AR subtype was only observed in a chronic and not an acute bladder inflammation model. Conversely, other investigators have shown using transient application of β-AR agonists that an increase in cAMP is sufficient to generate maintenance levels of NO in primary rat urothelial cells [20]. However, our studies using a human urothelial cell model, demonstrates that cAMP-dependent PKA activation is not necessary to induce inflammatory mechanisms for generating NO. Moreover, generation of homeostatic levels of NO in the rat model was sensitive to Ca2+ indicating that the responsible enzyme was eNOS, although transcriptional message (mRNA) for iNOS was well documented in this same report [20]. This suggests that chronic β-AR stimulation may induce expression of iNOS, which would generate higher levels of NO contributing to the production of inflammatory pain and increased micturition associated with IC.
In our human urothelial cell model we document a β-AR stimulated, PKA-independent signaling pathway that simultaneously increases the expression of two mediators of inflammation, COX-2 and iNOS. In addition, the pathophysiology linked to increased prostaglandin and NO production correlate well with the clinical manifestations associated with chronic inflammatory diseases like IC. Consequently, we believe that UROtsa cells serve as a readily accessible model for studying the β-AR-effector system associated with inflammation in IC. Nonsteroidal anti-inflammatory drugs (NSAIDs), which block the synthesis of prostaglandins by inhibiting COX, are commonly prescribed to relieve discomforts associated with IC. Moreover, NSIADs have been shown to decrease the amount of NO in vivo indicating the importance of COX-2 activity in regulating NO production during inflammation [21]. Furthermore, combined pharmacological inhibition of COX-2 and iNOS in a rat model of tonic pain, produces a synergistic antinociceptive effect [22]. This data suggests a common mechanism of action between these two drug classes, however, the associations between COX-2 and iNOS effector systems are currently unknown. Therefore, UROtsa cells represent a unique cell model whereby signal-transduction pathways common to the induction of both COX-2 and iNOS can be investigated. These studies not only may reveal novel targets of inflammatory pain that could be exploited therapeutically, but would increase our understanding of the etiology for general bladder inflammation and hyperexcitability in IC.
Conclusion
Stimulation of β-ARs expressed on cultured human urothelial cells leads to ERK phosphorylation and production of the pro-inflammatory enzymes. While cAMP levels rise in these cells after β-AR activation, production of COX-2 and iNOS are not dependent upon an increased cAMP regulated PKA activity. Continual initiation of AR function documented for patients diagnosed with IC would likely stimulate urothelial cell inflammatory responses thereby contributing to the etiology of this disease. Our results suggest that by focusing on common urothelial β-AR mediated inflammatory signaling pathways, reasonable pathophysiological mechanisms and potential therapeutic strategies could be developed for chronic inflammatory diseases like IC.
Methods
Cell Culture
The immortalized human urothelial (UROtsa) cell line was a gift from Donald Sens (University of North Dakota) and was propagated as previously reported [12]. Briefly, undifferentiated UROtsa cells were grown to confluence in serum-containing Dulbecco's Modified Eagle's Medium (DMEM) under standard cell culture conditions. Confluent UROtsa cells were washed in serum-free DMEM and pre-incubated with or without inhibitors protein kinase A (PKA) inhibitors 1 hr before addition of the selective β-AR agonist, isoproterenol. The PKA inhibitors H-89 (Sigma, St. Louis, MO) was used at a final concentration of 100 nM, while Rp-cAMPS (BioLog, Bremen, Germany) was used at 10 μM. Unless noted otherwise, isoproterenol (Sigma, St. Louis, MO) was added to cells at a final concentration of 100 nM. As a nonspecific initiator of inflammation control, cells were incubated with 300 nM lipopolysaccharide (CalBioChem, La Jolla, CA).
Membrane Preparation
A crude cell membrane preparation was prepared as previously described [23]. Briefly, UROtsa membranes were prepared by transferring suspended cells to a 50 mL conical tube and twice washing by centrifugation at 1000 × g using cold Hank's balance salt solution (HBSS). The intact cell pellet was resuspended in 10 mL of 0.25 M sucrose containing 10 μg/mL bacitracin, 10 μg/mL benzamidine, 10 μg/mL leupeptin, and 20 μg/mL phenylmethysulfonylfluoride. The cells were disrupted by freezing followed by Dounce homogenization of the thawed suspension using 20 strokes from a loose fitting (B) pestle. This mixture was then centrifuged at 1260 × g for 5 min at 4°C. Buffer A (20 mM HEPES, pH 7.5, 1.4 mM EGTA, 12.5 mM MgCl2) was added to the supernatant and centrifuged again at 30,000 × g for 15 min at 4°C. The resultant pellet was kept, resuspended in buffer A then centrifuged once more at 30,000 × g for 15 min at 4°C. The final crude membrane pellet was resuspended in buffer A containing 10% glycerol and stored in aliquots at -70°C until used for radioligand binding. Protein concentrations were measured using the method of Bradford [24].
Radioligand Binding
The radioligand binding protocol used for this study was performed as previously described [13]. Briefly, the density of expressed β-ARs on UROtsa cells was determined by saturation binding experiments using the nonselective β-AR antagonist 125I-CYP as the radiolabel (NEN Life Sciences, Boston, MA). Crude UROtsa cell membranes were allowed to equilibrate at 37°C with increasing concentrations of 125I-CYP (5–600 pM) in a 0.25 mL total volume of buffer A using 10-5 M propranolol to determine non-specific binding. Binding was stopped by filtering the membranes though Whatman GF/C glass fiber filters, followed by 5 – 5 mL washes with cold buffer A to remove any unbound drug. Amounts of total and non-specific radiolabel bound to cell membranes were calculated from radioactive counts remaining on the glass fiber filters. From the plotted saturation hyperbola, β-AR density (Bmax) and the equilibrium dissociation constant (Kd) of 125I-CYP for specific UROtsa cell binding sites were calculated using iterative non-linear regression analysis [25].
cAMP Assay
Confluent UROtsa cells used for the quantification of cAMP were treated in serum-free DMEM containing 1 mM 1-methyl-3-isobutylxanthine (IBMX) to inhibit phosphodiesterase. After 30 min of isoproterenol treatment, cells were lysed using 0.1 M HCl and collected for determination of cAMP production according to the Biotrak Assay System protocol (Amersham, Buckinghamshire, UK). Briefly, 3H-cAMP added to cell lysates was used to compete with endogenous cAMP for binding to a specific cAMP-binding protein. 3H-cAMP levels were then counted by liquid scintillation and related to endogenously generated cAMP by comparison with known standards. The concentration of isoproterenol that caused a half-maximal generation of cAMP (EC50) was calculated from non-linear regression analysis using Prism 4 (Graphpad Software, San Diego, CA).
Western Hybridization
After an appropriate period of time, treated cells were lysed using a modified RIPA buffer (150 mM NaCl, 10 mM Tris, pH 7.2, 0.1% sodium dodecylsulphate, 1.0% Triton X-100, 1.0% sodium deoxycholate, 5 mM EDTA, 1.0% protease inhibitor cocktail; Sigma, St. Louis, MO). Total cell lysate protein concentrations were estimated using Bradford protein assay reagent (Bio-Rad, Hercules, CA) before lysates were resolved by SDS-PAGE and transferred to nitrocellulose membranes. Protein expression was measured by 4°C overnight immunoblotting with diluted antibodies: extracellular signal-regulated kinase 2 (mouse monoclonal ERK2, 1:1000; Santa Cruz Biotechnology, Santa Cruz, CA), phosphorylated ERK1/2 (mouse monoclonal pERK, 1:500; Santa Cruz Biotechnology), COX-2 (goat polyclonal, 1:500; Santa Cruz Biotechnology) iNOS (rabbit polyclonal, 1:500; Santa Cruz Biotechnology) and phosphorylated CREB (rabbit polyclonal, 1:1000; AbCam, Cambridge, MA). After washing, membranes were incubated at 25°C for 90 min with diluted horseradish peroxidase-linked secondary antibody (1:1000–5000). Bound antibody was visualized by the Supersignal West Pico chemiluminescent system (Pierce, Rockford, IL) and exposed to radiographic film. Developed films were subsequently photographed and protein band intensity was estimated by semi-quantitative densitometric methods using LabWorks v.4.5 (UVP, Upland, CA). Protein levels are presented as the mean fold increase in pixel intensity over control, plus or minus the standard error for n experiments. Differences between control and drug treated groups was determined using a paired one-tailed Student's t test with a p < .05 level of probability accepted as significant. Equal protein loading was confirmed by Ponceau-S staining of nitrocellulose membranes.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
EBH performed all of the experiments and was responsible for acquisition, analysis and interpretation of the data. JMP provided assistance for the western blot analysis as part of a NSF funded summer fellowship. JEP conceived, monitored, and coordinated the experimental design. Both EBH and JEP contributed equally to the writing of this manuscript.
Acknowledgements
This work was supported in part by grants from the National Science Foundation 0235146 and the National Institutes of Health R21DK062865 (JEP).
==== Refs
Wesselmann U Interstitial cystitis: A chronic visceral pain syndrome Urology 2001 57 32 39 11378048 10.1016/S0090-4295(01)01123-2
Funk CD Prostaglandins and leukotrienes: Advances in eicosanoid biology Science 2001 294 1871 1878 11729303 10.1126/science.294.5548.1871
Stock JL Shinjo K Burkhardt J Roach M Taniguchi K Ishikawa T The prostaglandin E2 EP1 receptor mediates pain perception and regulates blood pressure J Clin Invest 2000 107 325 331 11160156
Hu VY Malley S Dattilio A Folsom JB Zvara P Vizzard MA COX-2 and prostanoid expression in micturition pathways after cyclophosphamide-induced cystitis in the rat Am J Physiol Regul Integr Comp Physiol 2003 284 R574 R585 12388444
Guzik TJ Korbut R Adamek-Guzik T Nitric oxide and superoxide in inflammation and immune regulation J Physiol Pharmacol 2003 54 469 487 14726604
Xie Q Nathan C The high-output nitric oxide pathway: role and regulation J Leukoc Biol 1994 56 576 582 7525816
Cuzzocrea S Chatterjee PK Mazzon E McDonald MC Dugo L Di Paola R Beneficial effects of GW274150, a novel, potent and selective inhibitor of iNOS activity, in a rodent model of collagen-induced arthritis Eur J Pharmacol 2002 453 119 129 12393067 10.1016/S0014-2999(02)02338-5
Ehren I Hosseini A Lundberg JO Wiklund NP Nitric oxide: a useful gas in the detection of lower urinary tract inflammation J Urol 1999 162 327 329 10411031 10.1097/00005392-199908000-00011
Hohenfellner M Nunes L Schmidt RA Lampel A Thuroff JW Tanagho EA Interstitial cystitis: increased sympathetic innervation and related neuropeptide synthesis J Urol 1992 147 587 591 1538434
Stein PC Torri A Parsons CL Elevated urinary norepinephrine in interstitial cystitis Urology 1999 53 1140 1143 10367842 10.1016/S0090-4295(98)00663-3
Saban MR Nguyen NB Hammond TG Saban R Gene expression profiling of mouse bladder inflammatory responses to LPS, substance P, and antigen-stimulation Am J Pathol 2002 160 2095 2110 12057914
Rossi MR Masters JR Park S Todd JH Garrett SH Sens MA The immortalized UROtsa cell line as a potential cell culture model of human urothelium Environ Health Perspect 2001 109 801 808 11564615
Zuscik M Porter JE Gaivin RJ Perez DM Identification of a conserved switch residue responsible for selective constitutive activation of the β2-adrenergic receptor J Biol Chem 1998 273 3401 3407 9452461 10.1074/jbc.273.6.3401
Jiang B Xu S Hou X Pimentel DR Brecher P Cohen RA Temporal control of NF-κB activation by ERK differentially regulates interleukin-1β-induced gene expression J Biol Chem 2004 279 1323 1329 14581482 10.1074/jbc.M307521200
Daaka Y Luttrell LM Lefkowitz RJ Switching of the coupling of the β2-adrenergic receptor to different G proteins by protein kinase A Nature 1997 390 88 91 9363896 10.1038/36362
Meyer TE Habener JF Cyclic adenosine 3',5'-monophosphate response element binding protein (CREB) and related transcription-activating deoxyribonucleic acid-binding proteins Endocr Rev 1993 14 269 290 8319595 10.1210/er.14.3.269
Yabe T Kanemitsu K Sanagi T Schwartz JP Yamada H Pigment epithelium-derived factor induces pro-survival genes through cyclic AMP-responsive element binding protein and nuclear factor kappa B activation in rat cultured cerebellar granule cells: Implication for its neuroprotective effect Neuroscience 2005 133 691 700 15893882 10.1016/j.neuroscience.2005.03.007
Luttrell LM Ferguson SSG Daaka Y Miller WE Maudsley S Della Rocca GJ β-Arrestin-dependent formation of β2 adrenergic receptor-Src protein kinase complexes Science 1999 283 655 661 9924018 10.1126/science.283.5402.655
Sung CS Wen ZH Chang WK Ho ST Tsai SK Chang YC Intrathecal interleukin-1β administration induces thermal hyperalgesia by activating inducible nitric oxide synthase expression in the rat spinal cord Brain Res 2004 1015 145 153 15223378 10.1016/j.brainres.2004.04.068
Birder LA Nealen ML Kiss S de Groat WC Caterina MJ Wang E β-Adrenoceptor agonists stimulate endothelial nitric oxide synthase in rat urinary bladder urothelial cells J Neurosci 2002 22 8063 8070 12223560
Uno K Iuchi Y Fujii J Sugata H Iijima K Kato K In vivo study on cross talk between inducible nitric-oxide synthase and cyclooxygenase in rat gastric mucosa: Effect of cyclooxygenase activity on nitric oxide production J Pharmacol Exp Ther 2004 309 995 1002 14988416 10.1124/jpet.103.061283
Dudhgaonkar SP Kumar D Naik A Devi AR Bawankule DU Tandan SK Interaction of inducible nitric oxide synthase and cyclooxygenase-2 inhibitors in formalin-induced nociception in mice Eur J Pharmacol 2004 492 117 122 15178354 10.1016/j.ejphar.2004.03.021
Porter JE Perez DM Influence of a lysine 331 counterion on the pKa of aspartic acid 125: Evidence for a salt-bridge interaction and role in Alpha-1b adrenergic receptor activation J Pharmacol Exp Ther 2000 292 440 448 10604981
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
Motulsky HJ Ransnas LA Fitting curves to data using nonlinear regression: A practical and nonmathmatical review FASEB J 1987 1 365 374 3315805
|
16091148
|
PMC1198236
|
CC BY
|
2021-01-04 16:39:15
|
no
|
Cell Commun Signal. 2005 Aug 9; 3:10
|
utf-8
|
Cell Commun Signal
| 2,005 |
10.1186/1478-811X-3-10
|
oa_comm
|
==== Front
Cerebrospinal Fluid ResCerebrospinal Fluid Research1743-8454BioMed Central London 1743-8454-2-41602950010.1186/1743-8454-2-4EditorialCerebrospinal Fluid Research: The first six months and the introduction of article processing charges Jones Hazel C [email protected] Editorial Office Gagle Brook House, Chesterton, Bicester, Oxon OX26 1UF, United Kingdom2005 19 7 2005 2 4 4 16 7 2005 19 7 2005 Copyright © 2005 Jones; licensee BioMed Central Ltd.2005Jones; 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.
Article processing charges (APCs) have recently been introduced for authors submitting papers to Cerebrospinal Fluid Research. This editorial is to inform readers about the need and use of APCs and about the advantages of free open access publishing.
==== Body
Introduction
It is now over six months since the launch of Cerebrospinal Fluid Research in December 2004. Cerebrospinal Fluid Research is published by BioMed Central, an independent publisher committed to ensuring peer-reviewed biomedical research is Open Access. To fund this, from July 1st 2005 authors of articles accepted for publication will be asked to pay an article-processing charge (APC) of £330.00.
Traditionally, readers pay to access articles, either through subscriptions or by paying a fee each time they download an article. Escalating journal subscriptions and shrinking library budgets have resulted in libraries subscribing to fewer journals [1], and the range of articles available to readers is becoming more limited. Although traditional journals publish authors' work for free (unless there are page or colour charges), paying to access articles limits the number of people that can read, use and cite them. The benefits of Open Access and of Cerebrospinal Fluid Research's Open Access policy were highlighted in a previous editorial [2].
Description of Payment
APCs will allow continued Open Access to all of Cerebrospinal Fluid Research's articles. Authors are asked to pay £330 if their article is accepted for publication. Waiver requests will be considered on a case-by-case basis, by the Editor-in-Chief. Authors can circumvent the charge if their institution becomes a 'member' of BioMed Central, in which case the annual membership fee covers the APCs for the authors of all BioMed Central journals at that institution for that year. Current members include NHS England, the World Health Organization, the US National Institutes of Health, Harvard, Princeton and Yale universities, and all UK universities [3]. No charge is made for articles that are rejected after peer review. Many funding agencies have also realized the importance of Open Access publishing and have specified that their grants may be used directly to pay APCs [4]. Additionally, the UK Research Councils recently announced that they are considering their position on improved access to research outputs [5].
What is your APC used for?
The APC pays for online submission and efficient peer review, for the article to be freely and universally accessible in various formats online, and for the processes required for inclusion in PubMed and archiving in PubMed Central, e-Depot, Potsdam and INIST. There is no remuneration of any kind provided to the Editor-in-Chief, to any members of the editorial board, or to peer reviewers; all of whose work is entirely voluntary. Although some authors may consider £330 expensive, it must be remembered that Cerebrospinal Fluid Research does not levy additional page or colour charges on top of this fee. With the article being online only, any number of colour figures and photographs can be included, at no extra cost. Another common expense with traditional journals is the purchase of reprints for distribution, and the cost of these reprints is frequently greater than our APCs. Cerebrospinal Fluid Research provides free, publication-quality pdf files for distribution, in lieu of reprints.
Advantages of Open Access
Although several journals now offer free online access to their articles, this is different from Open Access (as defined by the Bethesda Statement [6]). Journals often delay free access for 6–12 months, and even when the full text is available, readers are not allowed to reproduce and/or disseminate the work because of restrictions imposed by the copyright policy. That said, Cerebrospinal Fluid Research is not alone in the move to Open Access funded by APCs: many publishers now offer various "hybrid" forms where the authors are asked to contribute to the publishing costs. The Public Library of Science also has Open Access journals that are funded by grants and APCs of about US$1500 per article [7]. The high profile of these journals will raise awareness of Open Access and encourage researchers in all disciplines to understand and accept Open Access, with APCs as an acceptable method to fund it.
Conclusion
By providing a forum for Open Access, APCs will enable Cerebrospinal Fluid Research to continue to publish attractive and important papers on outstanding research that will be accessible to everyone worldwide. We believe this process will bring together people in a vigorous field of research, and we hope you will support this progress by submitting your next article to this Open Access journal.
==== Refs
Tamber PS Is scholarly publishing becoming a monopoly? BMC News and Views 2000 1 1
Jones HC Cerebrospinal Fluid Research: A new platform for dissemination of research, opinions and reviews with a common theme, Cerebrospinal Fluid Research 2004 1 1 15679934 10.1186/1743-8454-1-1
BioMed Central Institutional members
Which funding agencies explicitly allow direct use of their grants to cover article processing charges?
Research Councils UK: Access to Research Outputs
Bethesda Statement on Open Access Publishing
Public Library of Science
|
16029500
|
PMC1198237
|
CC BY
|
2021-01-04 16:37:38
|
no
|
Cerebrospinal Fluid Res. 2005 Jul 19; 2:4
|
utf-8
|
Cerebrospinal Fluid Res
| 2,005 |
10.1186/1743-8454-2-4
|
oa_comm
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.