| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Departments of Internal Medicine (R.v.C., M.G.N., J.W.M.v.d.M.) and Gastroenterology (C.E.W.), University Medical Center Nijmegen, 6500 HB Nijmegen, The Netherlands; South-East Asian Ministries of Education Organization-Tropical Medicine, Regional Center for Community Nutrition (E.K.), and Working Group on Infectious Diseases, Faculty of Medicine (R.H.H.N.), University of Indonesia, Jakarta, Indonesia; and Division of Human Nutrition and Epidemiology (H.V., C.E.W.), Wageningen University, 6700 HB Wageningen, The Netherlands
Address all correspondence and requests for reprints to: Jos W. M. van der Meer, M.D., FRCP, Professor of Internal Medicine, University Medical Center Nijmegen, P. O Box 9101, 6500 HB Nijmegen, The Netherlands. E-mail: j.vandermeer{at}aig.azn.nl
Abstract
Tuberculosis patients often suffer from severe weight loss, which is considered to be immunosuppressive and a major determinant of severity and outcome of disease. Because leptin is involved in weight regulation and cellular immunity, its possible role in tuberculosis-associated wasting was investigated. In an urban clinic in Indonesia, plasma leptin concentrations, indicators of adipocyte mass, appetite, C-reactive protein (CRP), tuberculin reactivity, and cytokine response were measured in tuberculosis patients and healthy controls. Plasma leptin concentrations were lower in patients than in controls (615 vs. 2,550 ng/liter; P < 0.001). Multivariate regression analysis showed that body fat mass and inflammation were two independent factors determining plasma leptin concentrations; there was a positive correlation between fat and leptin, whereas, unexpectedly, leptin was inversely associated with CRP and tumor necrosis factor-
production. Concentrations of both CRP and leptin were independently associated with loss of appetite. Our results do not support the concept that weight loss in tuberculosis is caused by enhanced production of leptin. Rather, loss of body fat leads to low plasma leptin concentrations, and prolonged inflammation may further suppress leptin production. Because leptin is important for cell-mediated immunity, low leptin production during active tuberculosis may contribute to increased disease severity, especially in cachectic patients.
WASTING HAS LONG been recognized as a prominent feature of tuberculosis and is probably one of the determinants of the disease severity and outcome (1). However, uncertainty surrounds cause as well as effect of a poor nutritional status in tuberculosis patients. The cause, or pathogenesis of tuberculosis-associated wasting is incompletely understood, although it is likely that inflammatory mediators such as tumor necrosis factor- (TNF) do play a role (2). Similarly, a poor nutritional status is known to suppress cellular immunity, which is essential against Mycobacterium tuberculosis, but the precise mechanism remains uncertain (3, 4).
Leptin, the 16-kDa product of the ob-gene, may be involved in this cross-regulation between nutritional status and the immune response in tuberculosis. Leptin is produced by adipocytes and binds to specific receptors in the hypothalamus, from which it suppresses appetite (5). Concentrations of circulating leptin are proportional to fat mass (6), are reduced in starvation (7, 8), and are increased by inflammatory mediators (9). Administration of leptin to leptin-deficient ob/ob mice reduces food intake and increases energy expenditure (10). Experimental evidence has shown a number of other possible functions of leptin, including immune regulation. Leptin stimulates the proinflammatory response (11) and promotes proliferation, differentiation, and activation of hematopoietic cells (12). In mice, the reduction of leptin concentrations induced by starvation enhances sensitivity to endotoxic shock (13). Falling leptin concentrations also appear to be responsible for reduced T-cell function during starvation (14).
On the basis of the above, plasma leptin concentrations in tuberculosis may be the result of two antagonistic mechanisms. Whereas tuberculosis-associated loss of body fat mass may lead to reduced production of leptin (15), the host inflammatory response may increase leptin production (9). If, as an overall result, plasma leptin concentrations are increased in tuberculosis patients, then this might theoretically suppress appetite and food intake and be one of the mechanisms underlying weight loss. However, if plasma leptin concentrations are decreased in tuberculosis, then this might suppress cellular immunity and aggravate disease outcome. Therefore, the aim of this study was to measure plasma leptin concentrations in tuberculosis patients and to explore determinants of leptin such as nutritional status and the inflammatory response. The study was conducted in Indonesia, where malnutrition is highly prevalent among tuberculosis patients (3).
Subjects and Methods
Subjects
In an outpatient tuberculosis clinic in Jakarta, Indonesia, 60 consecutively selected patients with pulmonary tuberculosis were evaluated before and after 2 months of standard antituberculous treatment with isoniazide, rifampicin, pyrazinamide, and ethambutol. Diagnosis was based on clinical presentation and radiology and confirmed by sputum microscopy and culture for M. tuberculosis. In a subgroup of 20 (untreated) patients, tuberculin reactivity was measured. Thirty healthy individuals resident in the same neighborhood as the patients were selected for comparison. These controls had no history or signs of active pulmonary tuberculosis and had no abnormalities on chest x-ray examination. Informed consent was obtained from all patients and control subjects, and the study was approved by the ethical committee of the Faculty of Medicine, University of Indonesia, Jakarta.
Anthropometric measurements
Patients and control subjects were weighed barefoot with minimum clothing using an electronic weighing scale (SECA-770). Body weight was recorded to the nearest 0.1 kg. Height was measured to the nearest 0.1 cm using a microtoise. Total body fat was estimated from the average of two duplicate measurements of skinfold thickness at four sites (biceps, triceps, subscapular, and suprailiac regions) (16). Food intake was estimated from two 24-h recalls using World Food version 2.0 (University of California, Berkeley, CA).
Laboratory methods
Plasma leptin concentrations were measured by capture ELISA according to guidelines of the manufacturer (Quantikine DLP00, R&D Systems, Minneapolis, MN). Plasma C-reactive protein (CRP) concentrations were measured by standard turbidimetry. In a subgroup of 20 patients, ex vivo production of cytokines was assessed in whole blood as previously described (17). Briefly, whole blood was incubated in closed vacutainer tubes at 37 C without stimulation or with lipopolysaccharide (LPS; final concentration, 10 µg/liter), phytohemoagglutinin (10 mg/liter), or purified protein derivative (PPD; 10 mg/liter). Supernatants were harvested after incubation for 6 h (LPS and phytohemoagglutinin) or 24 h (PPD). Concentrations of TNF, IL-1ß, and IL-1 receptor antagonist (IL-1ra) were measured by specific RIA (18), and concentrations of interferon-
(IFN-) and IL-6 were measured by ELISA (Pelikine, CLB, Amsterdam, The Netherlands). All relevant comparisons were made within single assays. The intra-assay coefficient of variation was 3% for leptin, 6% for IFN-, and less than 10% for TNF, IL-1ß, and IL-1ra. Day to day variation of whole blood ex vivo cytokine production in humans is 12% for TNF, 23% for IL-1ß, 5% for IL-1ra, and 47% for IFN- (17).
Statistical analyses
Data were analyzed using SPSS version 7.5.2 for Windows (SPSS, Inc., Chicago, IL). Patients and controls were compared regarding their plasma leptin concentrations, body weight, body mass index (BMI) (calculated as weight/height2, kg/m2), body fat mass, sex, and age, using t test or Mann-Whitney U test as appropriate. The relationship between plasma leptin concentrations and body fat mass, and between plasma concentrations of leptin and CRP was analyzed by univariate regression. Multivariate regression models were used to assess whether data were consistent with our hypothesis that a tuberculosis-associated change in plasma leptin concentration is mediated through changes in body fat mass and CRP. Hence, log (plasma leptin concentration) was modeled with body fat mass and CRP as main terms. A term for tuberculosis was retained to account for possible mechanisms through which it might influence plasma leptin concentration independently from body fat and CRP. The geometric mean changes in plasma leptin concentration, food intake, and the mean change in body weight, body fat mass, and CRP after 2 months of antituberculous treatment were evaluated by one-sample t tests. Multivariate logistic regression analysis was used to evaluate plasma leptin concentrations and inflammation as determinants of reported appetite (normal or below normal) in tuberculosis patients at baseline. CRP was used as a marker of the inflammatory response, which was considered as a possible confounder in the relation of plasma leptin concentration with appetite.
Results
Plasma leptin concentrations are lower in tuberculosis patients than in controls
The subjects included in this study were mostly young adults and more often male (Table 1
). Patients with tuberculosis presented with a 2- to 6-month history of respiratory symptoms (100%), fever (60%), night sweats (68%), fatigue (83%), and weight loss (80%). None of the investigated patients was HIV-positive. Reported weight loss ranged from 0–25 kg (median, 5 kg). Patients with untreated tuberculosis had substantially lower weight, BMI, and body fat mass than controls (Table 1). BMI in patients was reduced by 16% compared with controls, and body fat mass (percentage) by 45%. Patients had a geometric mean [95% confidence interval (CI)] plasma leptin concentration of 617 (range, 469–810) ng/liter, compared with 2,539 (range, 1,548–4,168) ng/liter in healthy control subjects (P < 0.001; Fig. 1A). This difference corresponded with a 76% reduction in patients. Patients were somewhat older than controls, but no association was found between age and plasma leptin concentration among either controls or patients. Women had substantially higher plasma leptin concentrations than men (geometric mean, 1,938 ng/liter compared with 575 ng/liter; ratio, 3.4; 95% CI, 2.0–5.6), but the sex ratio was similar among patients and controls (Table 1). In both male and female subjects, plasma leptin concentrations were significantly reduced in tuberculosis patients (Fig. 1B).
|
|
Univariate analysis showed that plasma leptin concentration increased proportionally with body fat mass (Fig. 2A). For every 10 U increment in fat percentage, plasma leptin concentration increased 3.1-fold (95% CI, 2.4–3.9) and 2.6-fold (95% CI, 1.4–4.7) in patients and control subjects, respectively. Median CRP was 52 mg/liter (interquartile range, 19–95 mg/liter) in tuberculosis patients and 2 mg/liter (1–3 mg/liter) in controls (P < 0.001). Among patients, plasma leptin concentrations were reduced by 33% (95% CI, 15–46%) for every 50 mg/liter increment in CRP. In control subjects, there was insufficient variation in CRP to assess its relation with plasma leptin concentration.
|
shows the results of a multivariate regression model to assess the role of body fat mass and possible other mechanisms whereby tuberculosis may affect plasma leptin concentration. This figure shows that patients with untreated tuberculosis have lower plasma leptin concentrations than expected for body fat mass. Table 2 gives a comparison of two regression models to assess the role of body fat mass, CRP, and other possible mechanisms whereby tuberculosis may affect plasma leptin concentrations. Multivariate regression analysis showed no evidence for effect modification or confounding by age and sex in these associations. Univariate regression (model 1) gave a crude assessment of the association between tuberculosis and plasma leptin concentration. Multivariate regression, with addition of CRP and body fat mass (model 2), reduced this association (from 76–35%) and shows that plasma leptin concentration was positively associated with body fat and inversely associated with CRP. The estimated proportion of variability in plasma leptin concentration explained (multiple R2) in model 2 was 0.64. Taken together, these findings are consistent with our hypothesis that tuberculosis-associated reductions in plasma leptin concentrations are mediated independently through a decreased percentage of body fat, inflammation, and possibly other, unidentified mechanisms.
|
The effect of antituberculous treatment on plasma leptin concentrations, appetite, nutritional status, and acute phase proteins was evaluated in 38 patients for whom blood test results were available after 2 months of treatment. These patients reported a clear improvement of symptoms within 1–3 wk after start of treatment. Plasma leptin concentrations were substantially higher after treatment (geometric mean difference, 409 ng/liter; 95% CI, 219–676 ng/liter; P < 0.001), corresponding to an increase of 64% relative to baseline (Fig. 2B). There was no evidence that selection bias caused by missing data substantially affected the estimated effect of treatment on the change in plasma leptin concentration (data not shown). Loss of appetite was reported by 27 patients (71%) before treatment and by none after treatment. After 2 months of treatment, patients had a higher energy intake (mean difference, 375 kJ), body weight (mean difference, 1.6 kg; 95% CI, 0.9–2.3 kg; P < 0.001), and body fat mass (mean difference, 1.6%, 95% CI, 0.6–2.5%; P = 0.003), and lower CRP (mean difference, 50 mg/liter; 95% CI, 26–74 mg/liter; P = 0.001).
Plasma leptin and cytokine response
The relationship between plasma leptin concentrations and the ex vivo production of proinflammatory cytokines was investigated in 19 tuberculosis patients for whom data were available. Linear regression analysis showed an inverse correlation between spontaneous ex vivo production of TNF and plasma leptin concentrations before treatment. Plasma leptin concentrations reduced by 42% (95% CI, 12–62%) for every 0.1 ng/ml increment in plasma TNF concentration. LPS-mediated production of TNF, as well as the production of IL-1ß, IL-1ra, and IL-6, was not significantly associated with plasma leptin concentrations before treatment. After treatment, there was a substantial decrease of plasma concentrations of IL-6 and ex vivo production of IL-1ß, IL-6, and TNF (data not shown). However, we could not directly demonstrate a significant association between changes in leptin concentrations and changes in cytokine production.
To investigate the relationship between plasma leptin concentrations and T-cell immunity, tuberculin skin tests and production of IFN- were evaluated in untreated patients. The size of skin reactions to PPD showed a positive but statistically nonsignificant correlation with plasma leptin concentrations (P = 0.30). A similar result was found for plasma leptin concentrations and PPD-induced ex vivo production of IFN- (P = 0.38). Ex vivo production of IFN- increased after treatment (mean difference, 188 pg/ml; 95% CI, -7 to 539 ng/liter). However, no significant relationship could be shown between change in plasma leptin concentrations and IFN- production (P = 0.39).
Determinants of appetite
Both inflammation and plasma leptin concentration were associated with loss of appetite in tuberculosis patients. Every 50-mg/liter increment in CRP was associated with a 1.4-fold increase (95% CI, 0.8–2.6) in the odds of reporting loss of appetite. When adjusted for CRP, every 1,000-ng/liter increment in plasma leptin concentration was associated with a 1.7-fold increase (95% CI, 0.7–4.3) in the odds of reporting loss of appetite.
Discussion
Tuberculosis often leads to severe weight loss (wasting), probably through the production of inflammatory mediators (2). Wasting, in turn, affects the inflammatory response, suppresses cellular immunity, and aggravates the outcome of tuberculosis (19). In these complex relations between tuberculosis, nutritional status and the host immune response, leptin is a possible mediator. In this study, plasma leptin concentrations were significantly suppressed in tuberculosis patients in Indonesia. Body fat mass was strongly correlated with plasma leptin concentrations, both in patients and controls. Unexpectedly, in tuberculosis patients, plasma CRP and in vitro production of TNF showed an inverse correlation with plasma leptin concentrations. Results of multivariate regression analysis support the hypothesis that tuberculosis-associated reductions of plasma leptin were mediated independently by weight loss and inflammation. Although previous data have shown that leptin stimulates cell-mediated immunity, we were unable to demonstrate a statistically significant correlation of plasma leptin concentrations with tuberculin reactivity or IFN- production.
To our knowledge, there is one previous study on plasma leptin concentrations in tuberculosis patients (21). In that report, leptin concentrations, as determined by RIA, were much higher than in ours. The (Turkish) patients in that report had a much higher BMI, but it seems surprising to us that after treatment they had 3-fold higher leptin concentrations than control subjects. Also, the control subjects had increased plasma TNF values, an unexpected finding in healthy individuals.
Loss of body fat mass could not entirely explain the observed low plasma leptin concentrations in tuberculosis patients in our study. Body fat mass is the most important determinant of plasma leptin concentrations, but starvation, hormones (including insulin and cortisol), as well as inflammatory mediators are able to modulate leptin production (22). Animal studies have shown that LPS, TNF, and IL-1ß raise leptin concentrations in serum and leptin mRNA in adipose tissue (7). Similarly, in cancer patients, recombinant TNF (22) and IL-1ß (23) increased plasma leptin. In sepsis patients, leptin levels were found to be elevated (23, 24, 25). To our surprise, in our study in tuberculosis, CRP and TNF production were inversely correlated with plasma leptin concentrations. Attenuation of the acute phase response and proinflammatory cytokine production during antituberculous treatment was accompanied by an impressive increase of plasma leptin concentrations. Of course, the acute inflammatory response in the animal and patient studies described above is different from the more chronic response in tuberculosis patients. The pattern of plasma leptin concentrations in weeks or months before diagnosis remains unknown, but one may hypothesize that the prolonged inflammatory response in tuberculosis down-regulates or exhausts leptin production.
In this study, multivariate analysis indicated that plasma leptin concentrations were associated with loss of appetite in tuberculosis. However, plasma leptin concentrations were substantially higher in control subjects (without anorexia), and patients regained appetite during treatment, despite a substantial increase in plasma leptin concentrations. Therefore, anorexia in tuberculosis seems to be determined to a much larger degree by inflammatory mediators (e.g. proinflammatory cytokines) than by leptin. Leptin signals the brain to decrease food intake, but so far no evidence has been found that anorexia in AIDS (26, 27) and other inflammatory disorders is caused by increased leptin levels (28, 29). In fact, it may be the other way around; in both laboratory animals (30) and human subjects (8), fasting induces falling leptin levels that evoke a number of adaptive responses, including suppression of metabolic rate (7). Similarly, in tuberculosis, decreased energy intake may reduce leptin production. We did not measure energy intake, but it is likely to be lower among tuberculosis patients than healthy controls.
Suppressed production of leptin may be detrimental for host defense against infections. In septic shock, mortality was found to be associated with decreased plasma leptin levels (23). In an animal model, the absence (13) or starvation-induced down-regulation of leptin increased susceptibility to endotoxic shock, and leptin partially reversed this effect (13). In addition, leptin reversed starvation-induced T-cell suppression (14). Host defense against tuberculosis depends on cell-mediated immunity, with a crucial role for Th1-type cytokines, primarily IFN- (31). Therefore, it may be hypothesized that decreased leptin production during active tuberculosis contributes to T-cell unresponsiveness. Indeed, in our patient group, both plasma leptin and ex vivo IFN- production were low and increased upon successful antituberculous treatment. We did not find a significant correlation between these two variables, which might be due either to the limited number of patients analyzed for cytokine production or to substantial intra- and interindividual variation of ex vivo cytokine production (17). We were also unable to show a statistical association between leptin and tuberculin reactivity, but skin testing, which was only done before treatment, is a rather crude measurement.
Based on our data and results from previous studies, we hypothesize that in untreated tuberculosis, loss of body fat, reduced energy intake, and the host immune response reduce leptin production (Fig. 3). Because leptin is important for cell-mediated immunity, suppressed leptin concentrations may contribute to a worse outcome of tuberculosis, especially in cachectic patients. In theory, administration of leptin might benefit tuberculosis patients, but this is not feasible in a country like Indonesia. Supplementation of micronutrients such as vitamin E (32) or zinc, which are known to increase leptin production (33), might be a cost-effective alternative. Of interest, zinc has the additional advantage of stimulating appetite (34).
|
Acknowledgments
We greatly appreciate the help of Wilma de Lenne and Yelilsan Veeraragu and the staff members of the outpatient clinic of the Perkumpulan Pemberantasan Tuberkulosis Indonesia, Jl Baladewa, Jakarta. Dr. Iskandar Zulkarnain, head of the Division of Tropical Medicine and Infectious Diseases, Department of Internal Medicine, Faculty of Medicine, University of Indonesia, provided staff to conduct this study. We thank Trees Verver, Liesbeth Jacobs, and Johanna van de Ven-Jongekrijg for their help with the cytokine assays.
Footnotes
R.v.C. is financially supported by the Dutch Organization for Scientific Research NWO (SGO Stipendium Infectious Diseases Grant SGO-INF 002). E.K. is financially supported by grants from Gesellschaft für Technische Zusammenarbeit, GmbH (Eschborn, Germany), the Neys-van Hoogstraten Foundation, Directorate General of Communicable Disease Control and Environmental Health, Ministry of Health, Republic of Indonesia, and the Integrated Excellent Research project from the Ministry of Research and Technology, Republic of Indonesia. H.V. is supported by a grant from The Netherlands Foundation for the Advancement of Tropical Research (NWO/WOTRO, Grant WV93-273). Quantikine leptin ELISAs (R&D Systems) were provided free of charge by Amgen, Inc. (Breda, The Netherlands).
Abbreviations: BMI, Body mass index; CI, confidence interval; CRP, C-reactive protein; IL-1ra, IL-1 receptor antagonist; IFN-, interferon-; LPS, lipopolysaccharide; PPD, purified protein derivative; TNF, tumor necrosis factor-.
Received June 14, 2001.
Accepted October 31, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  D. Ordway, M. Henao-Tamayo, E. Smith, C. Shanley, M. Harton, J. Troudt, X. Bai, R. J. Basaraba, I. M. Orme, and E. D. Chan Animal model of Mycobacterium abscessus lung infection J. Leukoc. Biol., June 1, 2008; 83(6): 1502 - 1511. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Trajman, M. Pai, K. Dheda, R. van Zyl Smit, A. A. Zwerling, R. Joshi, S. Kalantri, P. Daley, and D. Menzies Novel tests for diagnosing tuberculous pleural effusion: what works and what does not? Eur. Respir. J., May 1, 2008; 31(5): 1098 - 1106. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A I Martin, E Castillero, M Granado, M Lopez-Menduina, M A Villanua, and A Lopez-Calderon Adipose tissue loss in adjuvant arthritis is associated with a decrease in lipogenesis, but not with an increase in lipolysis J. Endocrinol., April 1, 2008; 197(1): 111 - 119. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Kumpers, R. Horn, G. Brabant, A. Woywodt, M. Schiffer, H. Haller, and M. Haubitz Serum leptin and ghrelin correlate with disease activity in ANCA-associated vasculitis Rheumatology, April 1, 2008; 47(4): 484 - 487. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. W. Wieland, S. Florquin, E. D. Chan, J. C. Leemans, S. Weijer, A. Verbon, G. Fantuzzi, and T. van der Poll Pulmonary Mycobacterium tuberculosis infection in leptin-deficient ob/ob mice Int. Immunol., November 1, 2005; 17(11): 1399 - 1408. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. van Lettow, J. W. M. van der Meer, C. E. West, R. van Crevel, and R. D. Semba Interleukin-6 and Human Immunodeficiency Virus Load, But Not Plasma Leptin Concentration, Predict Anorexia and Wasting in Adults with Pulmonary Tuberculosis in Malawi J. Clin. Endocrinol. Metab., August 1, 2005; 90(8): 4771 - 4776. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C Popa, M G Netea, T R D S Radstake, P L van Riel, P Barrera, and J W M van der Meer Markers of inflammation are negatively correlated with serum leptin in rheumatoid arthritis Ann Rheum Dis, August 1, 2005; 64(8): 1195 - 1198. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Franchimont, S. Roland, T. Gustot, E. Quertinmont, Y. Toubouti, M.-C. Gervy, J. Deviere, and A. Van Gossum Impact of Infliximab on Serum Leptin Levels in Patients with Crohn's Disease J. Clin. Endocrinol. Metab., June 1, 2005; 90(6): 3510 - 3516. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Gaetke, H. S. Oz, R. C. Frederich, and C. J. McClain Anti-TNF-{alpha} Antibody Normalizes Serum Leptin in IL-2 Deficient Mice J. Am. Coll. Nutr., October 1, 2003; 22(5): 415 - 420. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. West, A. Eilander, and M. van Lieshout Consequences of Revised Estimates of Carotenoid Bioefficacy for Dietary Control of Vitamin A Deficiency in Developing Countries J. Nutr., September 1, 2002; 132(9): 2920S - 2926. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
