Leptin Concentrations in Women in the San Antonio Heart Study: Effect of Menopausal Status and Postmenopausal Hormone Replacement Therapy

Transcription

1 American Journal of EpWemtokagy Copyright O 1997 by The Johns Hopkins University School of Hygiene and Public Health All rights reserved Vol. 146, No. 7 Printed In USA. A BRIEF ORIGINAL CONTRIBUTION Leptin Concentrations in Women in the San Antonio Heart Study: Effect of Menopausal Status and Postmenopausal Replacement Therapy Steven M. Haffner, Leena Mykkanen, and Michael P. Stem Leptin, the product of the human OB gene, is increased in obese individuals, suggesting resistance to its effect. However, there is variability in leptin levels at each level of body mass index, suggesting that genetic and environmental factors other than overall adiposity may regulate leptin concentrations. Leptin concentrations are higher in women relative to men, a difference that is only partially explained by the increased fat depots in women. The authors hypothesized that higher estrogen levels in women might be responsible for the sexual dimorphism in leptin concentrations. To test this hypothesis, they measured leptin concentrations in premenopausal women not on oral contraceptives (PRE) (n = 53), postmenopausal women on hormone replacement therapy (POSTY), and postmenopausal women not on hormone replacement therapy (POSTN) in the San Antonio Heart Study, a population-based study of diabetes and cardiovascular risk factors. Analyses were restricted to nondiabetic Mexican Americans. Subjects were well matched on obesity as assessed by body mass index (kg/m 2 ): PRE = 31.0, POSTY = 29.8, and POSTN = Leptin concentrations (ng/ml) were not significantly different among the three groups (PRE = 27.6, POSTY = 28.3, and POSTN = 27.8). The authors conclude that differences in estrogen levels are not likely to explain the sexual dimorphism in leptin concentrations. Am J Epidemiol 1997; 146: hormones; menopause; obesity; sex; sex hormones Leptin, a hormone regulating appetite and weight loss, has been the focus of major attention recently. Leptin is the product of the OB gene and is expressed in adipose tissue (1-4). Obese individuals have increased leptin levels (2-4) and thus may be resistant to the effects of leptin. After weight loss in humans, leptin levels decline (5, 6). Since there is residual variability in leptin levels at a given level of body mass index (5, 6), environmental and genetic factors other than obesity may regulate leptin concentrations. In all reports, leptin concentrations are higher in women than in men (5-10). However, there is disagreement about whether differences in adiposity explain the gender difference in leptin concentrations. In two reports (5, 6), adjustment for adiposity abolished the gender differences in leptin levels; however, in other studies (7-10), it did not. Studies of OB gene expression are more consistent in that higher OB gene expression is found in women than in men (2) and also Received for publication November 18, 1996, and accepted for publication June 5, From the Department of Medicine, University of Texas Health Science Center at San Antonio, San Antonio, TX. Reprint requests to Dr. Steven M. Haffner, Department of Medicine, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX in female mice compared with male mice at each level of body fat (11). Recently, Schwartz et al. (12) have found higher leptin levels in the cerebrospinal fluid of women than of men after controlling for age, adiposity, and serum leptin levels. Thus, the predominance of data suggests that there are gender differences in leptin concentrations that are unexplained by differences in adiposity. We hypothesized that high estrogen concentrations in women might explain the higher leptin levels in women. To explore this issue, leptin concentrations in premenopausal women (not on oral contraceptives) and postmenopausal women both on and off postmenopausal hormone replacement therapy were measured. These groups of women were closely matched on obesity and were selected from the San Antonio Heart Study, a population-based study of diabetes and cardiovascular risk factors. MATERIALS AND METHODS The San Antonio Heart Study is a population-based study of diabetes and cardiovascular disease in Mexican Americans and non-hispanic whites. From 1979 to 1982 (phase I) and from 1984 to 1988 (phase II), we randomly selected households in several San Antonio 581

2 582 Haffner et at. neighborhoods (13, 14). All men and nonpregnant women aged years who resided in the randomly selected households were eligible to participate. Only Mexican Americans were sampled in the barrio, but approximately equal numbers of each ethnic group were studied in the other types of neighborhoods. Mexican Americans were defined as individuals whose ancestry and cultural traditions derived from Mexican national origin (15). Detailed descriptions of the two study phases (I and II) have been published previously (13, 14). This study was approved by the Institutional Review Board of the University of Texas Health Science Center at San Antonio. All subjects gave informed consent. In October 1987, we began an 8-year follow-up of the phase I cohort to determine the incidence of noninsulin-dependent diabetes mellitus (type 2 diabetes) and cardiovascular disease (16). Beginning in October 1991, we started a similar 7-year follow-up of the phase II cohort (17). The results reported in this paper are based on cross-sectional analyses of a subset of subjects who participated in the 7-year follow-up of the phase II cohort. To simplify the analyses, subjects with diabetes by World Health Organization criteria (18) were excluded from this subset, although we have previously shown (19) that non-insulin-dependent diabetes is not associated with changes in leptin concentrations. In addition, to simplify the analyses, we restricted the analyses to Mexican Americans, although we have shown no ethnic difference in leptin levels between Mexican Americans and non-hispanic whites (20). Anthropometric measurements (height and weight, subscapular and triceps skinfolds, waist and hip circumferences) were made with the participant wearing an examining gown after having removed his/her shoes and upper garments (21). The triceps skinfold was measured directly posterior over the right triceps muscle, and the subscapular skinfold was measured just below the inferior angle of the right scapula. A Lange skinfold caliper (Cambridge, Maryland) was used, and the average of three readings, each made to the nearest 0.5 mm, was taken as the measurement for each skinfold. Waist circumference was measured at the level of the umbilicus. Hip circumference was measured at the level of the greater trochanters. Body mass index and the sum of the subscapular and triceps skinfolds were used as measures of overall adiposity and subcutaneous adiposity, respectively. Body mass index was calculated as weight (in kilograms) divided by height (in meters) squared. The waist-to-hip ratio was used as a measure of body fat distribution. At the follow-up of the phase II cohort, blood specimens were obtained after a 12- to 14-hour fast for determination of plasma glucose and serum insulin and proinsulin concentrations. Glucose was measured by a glucose oxidase method. A 75-g glucose equivalent load (Koladex or Orangedex, Custom Laboratories, Baltimore, Maryland) was then administered, and blood specimens were obtained 2 hours later for plasma glucose and serum insulin concentrations. We considered women to be postmenopausal if they had had a hysterectomy and bilateral oophorectomy and/or had had no menstrual periods in the year preceding their examination. We assessed whether postmenopausal women were taking postmenopausal estrogen by questionnaire. Women were considered to be postmenopausal if they had had fewer than two menstrual periods in the last year. Postmenopausal women taking hormones were initially matched to postmenopausal women not taking hormones by age (±2 years) and level of obesity (body mass index ± 2 kg/m 2. Seven hundred twenty-two women had contingency samples available. Women were considered to be premenopausal if they had had more than 10 periods in the year preceding their examination. We restricted analyses to premenopausal women who did not take oral contraceptives. Finally, premenopausal women were matched to the two groups of postmenopausal women on obesity (body mass index ±2 kg/ m 2 ). The premenopausal women, as expected, were younger than the postmenopausal women. The age range of the premenopausal women was years; the age range of the postmenopausal women was years. Serum samples later used for leptin determinations were stored at 70 C for an average duration of 2.9 years. These samples were not previously thawed until the leptin assay was performed. Leptin was measured by a commercial radioimmunoassay (Linco Research, Inc., St. Louis, Missouri) (7). The intraassay coefficient of variation was percent, and the interassay coefficient of variation was percent. Estradiol was measured by a solid phase radioimmunoassay (Diagnostic Products Corporation, Los Angeles, California). The lower limit of detectability was 8 pg/ml, and the intraassay and interassay coefficients of variation were 5.3 and 6.4 percent, respectively. Estrone was measured by a double antibody radioimmunoassay (Diagnostic Systems Laboratories, Webster, Texas). The lower limit of sensitivity was 1.2 pg/ml, and the intraassay and interassay coefficients of variation were 6.5 and 9.1 percent, respectively. The samples for sex hormones were not previously thawed until these assays were done; the average duration of storage was 3.8 years. Data analyses were performed using SAS statistical packages. The skewness and kurtosis of variables (lep-

3 Leptin Concentrations in Women: Menopausal Status 583 tin, body mass index, and so forth) were checked and found to be approximately normally distributed; therefore, parametric statistical tests were performed. RESULTS In table 1, the clinical and metabolic characteristics of subjects are shown by menopausal status. Subjects were matched on obesity as measured by body mass index. They also were similar with respect to overall adiposity as assessed by the sum of skinfolds and with respect to body fat distribution as determined by the waist-to-hip ratio or the ratio of subscapular-to-triceps skinfolds. Leptin concentrations (ng/ml) were not significantly different in premenopausal women (27.6 ng/ml), postmenopausal women on hormone replacement therapy (28.3 ng/ml), and postmenopausal women not on hormones (27.8 ng/ml). Estradiol and estrone levels were significantly different by menopausal status (p < 0.001). Estradiol and estrone levels were significantly higher in postmenopausal women on hormones than in premenopausal women (p < 0.001). Premenopausal women also had significantly higher estradiol and estrone levels than postmenopausal women not on hormones. The duration of postmenopausal hormone use was not significantly related to leptin levels (data not shown). The dose of Premarin therapy (used in 19 of 28 women) was also not significantly related to leptin concentrations (data not shown). Estradiol and estrone levels were not significantly related to leptin concentrations in the overall group or in any of the individual groups shown in table 1. In table 2, the clinical data are shown after adjustment for age and body mass index. Leptin concentrations were similar in each group. In table 3, we present a two-way analysis of variance with obesity (above and below the median for body mass index) and menopausal status as grouping variables. Obese subjects had significantly higher leptin concentrations than nonobese subjects (p < 0.001), but neither menopausal status nor hormone replacement therapy was related to leptin concentrations. DISCUSSION We have shown in this report that leptin concentrations are not related to menopausal status or postmenopausal hormone use. These findings argue against the hypothesis that increased estrogen concentrations in women are an explanation for their higher leptin concentrations. Further supporting the absence of a relation between hormone status and leptin levels are the absence of a dose response for premarin on leptin levels or a relation between the duration of estrogen use and leptin levels. These analyses, however, are limited by the small number of subjects. Hassink et al. (21) showed a rise in leptin levels (independently of adiposity) in girls with increased Tanner stage, suggesting a possible role for estrogen in modulation of leptin concentrations. Several reports have suggested higher leptin concentrations in women than in men (5-10, 21, 22). Two reports, which used bioimpedance to assess percentage of adiposity, reported no sex difference between men and women after adjustment for percentage of adipos- TABLE 1. Clinical and metabolic characteristics (mean ± SE*) of Mexican American women by menopausal status and hormone replacement (adjusted for age), San Antonio Heart Study follow-up, Age (yeare)f BMI* (kg/m*) WHR* Waist (cm) Fasting glucose (mg/dl) Leptin (ng/ml) Estrone (pg/ml) EstradBof (pg/ml) Skinfolds Subscapular (mm) Triceps (mm) STR* Sum of skinfolds (mm) PremenopausaJ (n = 53) 45.0 ± ± ± ± ± ± ± ± ± ± ± ± 2.8 Mean ±SE 51.5 ± ± ± ± ± ± ± ± ± ± ± ±3.6 PostmenopausaJ No hormone (n=28) Mean ± SE 52.2 ± ± ± ± ± ± ± ± ± ± ± ± 3.7 P valuet < < * SE, standard error, BMI, body mass index; WHR, waist-to-hip ratio; STR, ratio of subscapidar-to-triceps skinfolds. t Based on overall test of differences (F statistic from analysis of variance), i Not adjusted for age.

4 584 Haffner et al. TABLE 2. Clinical and metabolic characteristics of subjects (mean ± SE*) by menopausal status and hormone replacement (adjusted for age and BMI*), San Antonio Heart Study follow-up, Waist (cm) WHR* STR* Fasting glucose (mg/dl) Leptin (ng/ml) Estrone (pg/ml) EstracBol (pg/ml) Skinfolds (mm) Subscapular Triceps Sum of skinfolds Pre menopausal (n = 53) 95.1 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±3.6 Postman opau sal * SE, standard error; BMI, body mass index; WHR, waist-to-hip ratio. t Based on overall test of differences (Fstatistic from analysis of variance). No hormone 98.2 ± ± ± ± ± ± ± ± ± ±3.7 P valuet TABLE 3. Leptin levels (ng/ml) by hormonal status and level of obesity (adjusted for age), San Antonio Heart Study follow-up, No. Low BMI (<30.5 kg/m*) 26 High BMI (U30.5 kg/m2) 27 Premenopausal: no honnone Mean ± SEt 17.3 ± ± 2.4 No MeantSE 20.0 ± ± 3.3 Postmenopausal No hormone **><">* No. BMIt "" 7* ± ±3.1 * Computed by two-way analysis of variance with menopausal status and obesity as grouping variable, p value for interaction was fse, standard error; BMI, body mass index. ity (5, 6). In contrast, Saad et al. (9) have reported higher leptin levels in women than in men after adjustment for adiposity as assessed by the DEXA technique. However, Rosenbaum et al. (22) did not show a difference in leptin levels in adult men and women after adjustment for adiposity. Similarly, two other studies reported higher leptin levels in women than in men after adjustment for adiposity assessed by subcutaneous skinfolds (8, 10). In one of these studies (8), based on a population-based study (San Antonio Heart Study), the sex difference in leptin levels was 18.0 ng/ml. After adjustment for body mass index, the sex difference in leptin levels was essentially unchanged (17.6 ng/ml). After adjustment for subscapular/triceps skinfolds (which are increased in women relative to men), the sex difference in leptin was reduced to 9.0 ng/ml but was still highly significant (/? < 0.001). The current report has a number of limitations. Since the report is restricted to Mexican Americans, questions can be raised about its generalizability. In previous studies, no ethnic difference in leptin levels was observed between Mexican Americans and non- Hispanic whites (20) or between blacks and Caucasians (7). Furthermore, the relation between body mass index and leptin concentrations was similar in Mexican Americans and non-hispanic Whites (20). Although the validity of self-report may be limited, we have measured estradiol and estrone levels and shown results consistent with their self-reported status. The absence of a difference in leptin concentrations between pre- and postmenopausal women as well as the absence of an effect of postmenopausal hormone replacement therapy argues against an estrogenic effect regulating leptin levels. Moreover, the subjects were well matched for obesity using two different measures of adiposity; there were no differences in overall obesity or body fat distribution between the two groups. However, our study was cross-sectional. Definitive data would consist of observational studies going through natural menopause as well as direct studies of postmenopausal hormone replacement. It is also possible that differences in fat depots between men and women may help to explain gender differences in leptin levels. Men have lower levels of overall adiposity but greater visceral adiposity than women (23). Lonnqvist et al. (2) have shown that subcutaneous fat produces more leptin mrna than visceral fat (2), which could explain why women have higher leptin levels inasmuch as they have more subcutaneous fat than visceral fat. Our group, in fact, has p

5 Leptin Concentrations in Women: Menopausal Status 585 previously shown (8) that adjustment for subcutaneous fat reduced the sex difference in leptin concentrations much more than adjustment for waist circumference (which has been suggested as a good proxy for visceral fat) (24). Another possible explanation for the sexual dimorphism in leptin levels could be differences in the hypothalamus between men and women. There are binding sites for leptin in the hypothalamus, and leptin suppresses hypothalamic neuropeptide Y synthesis (25). The release of neuropeptide Y by the hypothalamus stimulates food intake (26). Urban et al. (27) have suggested that there are sex differences in the regional distribution of neuropeptide Y mrnacontaining cells in the hypothalamus of mice. However, it is not known whether this structural sex difference in the hypothalamus is associated with functional differences in leptin signaling and leptin levels. In conclusion, we have shown that leptin concentrations neither are significantly different in premenopausal and postmenopausal women nor are affected by postmenopausal hormone use. Our data suggest that sex hormones do not explain the sexual dimorphism in leptin levels, although direct studies of hormone replacement in both men and women would provide more definitive data. ACKNOWLEDGMENTS This work was supported by a grant from the National Heart, Lung, and Blood Institute (R01 HL24799). REFERENCES 1. Masuzaki H, Ogawa Y, Isse N, et al. Human obese gene expression: adipocyte-specific expression and regional differences in the adipose tissue. Diabetes 1995;44: Ltfnnqvist F, Amer P, Nordfors L, et al. Overexpression of the obese (OB) gene in adipose tissue of human obese subjects. Nature Med 1995;l: Hamilton BS, Paglia D, Kwan AYM, et al. Increased obese mrna expression in omental fat cells from massively obese humans. Nature Med 1995;l: Considine RV, Considine EL, Williams CJ, et al. Evidence against either a premature stop codon or the absence of obese gene mrna in human obesity. J Clin Invest 1995;95: Maffei M, Halaas J, Ravussin E, et al. Leptin levels in human and rodent: measurement of plasma leptin and OB RNA in obese and weight-reduced subjects. Nature Med 1995;1: Considine RV, Sinha MK, Heiman ML, et al. Serum immunoreactive leptin concentrations in normal weight and obese humans. N Engl J Med 1996;334: Ma ZA, Gingerich RL, Santiago JV, et al. Analyses of human plasma leptin by radioimmunoassay. Clin Chem 1996;42: Haffner SM, Gingerich RL, Miettinen H, et al. Leptin concentrations in relation to overall adiposity and regional body fat distribution in San Antonio. Int J Obes 1996;20: Saad MF, Damani S, Gingerich RL, et al. Sexual dimorphism in plasma leptin concentration. J Clin Endocrinal Metab 1997; 82: Ostlund RE Jr, Klein S, Yang JW. Relation between plasma leptin concentration, body composition and metabolic covariates. Diabetes 1996;45(suppl 2):41A. 11. Frederich RC, Hamann A, Anderson S, et al. Leptin levels reflect body lipid content in mice: evidence for diet-induced resistance to leptin action. Nature Med 1995;1: Schwartz MW, Peskind E, Raskind M, et al. Cerebrospinal fluid leptin levels: relationship to plasma levels and to adiposity in humans. Nature Med 1996,2: Stem MP, Rosenthal M, Haffner SM, et al. Sex difference in the effects of sociocultural status on diabetes and cardiovascular risk factors in Mexican Americans: the San Antonio Heart Study. Am J Epidemiol 1984; 120: Haffner SM, Stem MP, Hazuda HP, et al. Hyperinsulinemia in a population at high risk for non-insulin dependent diabetes meuitus. N Engl J Med 1986;315: Hazuda HP, Comeaux PJ, Stem MP, et al. A comparison of three indicators for identifying Mexican Americans in epidemiologic research: methodological findings from the San Antonio Heart Study. Am J Epidemiol 1986;123: Haffner SM, Hazuda HP, Mitchell BD, et al. Increased incidence of type II diabetes mellitus in Mexican Americans. Diabetes Care 1991;14: Haffner SM, Miettinen H, Gaskill SP, et al. Decreased insulin secretion and increased insulin resistance are independently related to the 7-year risk of NIDDM in Mexican Americans. Diabetes 1995;44: World Health Organization Study Group on Diabetes Mellitus. Diabetes mellitus: report of a WHO study group. Geneva: WHO, 1985:94-8. (Technical report no.727). 19. Haffner SM, Stem MP, Miettinen H, et al. Leptin concentrations in diabetic and non-diabetic Mexican Americans. Diabetes 1996,45: Wei M, Haffner SM, Stem MP. Serum leptin levels in Mexican Americans and non-hispanic whites: association with BMI and cigarette smoking. Ann Epidemiol 1997;7: Hassink SG, Sheslow DV, de Lancey E, et al. Serum leptin in children with obesity: relationship to gender and development Pediatrics 1997;98: Rosenbaum M, Nicolson M. Hirsch J, et al. Effects of gender, body composition, and menopause in plasma concentrations of leptin. J Clin Endocrinal Metab 1996;9: Haffner SM, Stem MP, Hazuda HP, et al. Upper body and centralized adiposity in Mexican Americans and non-hispanic whites: relationship to body mass index and other behavioral and demographic variables. Int J Obesity 1986; 10: Pouliot M, Despres JP, Lemieux S, et al. Waist circumference and abdominal sagittal diameter best simple anthropometric indexes of abdominal visceral adipose tissue accumulation and related cardiovascular risk in men and women. Am J Cardiol 1994;73: Stephens TW, Baslnski M, Bristown PK, et al. The role of nenropeptide Y in the antiobesity action of the obese gene product Nature 1995;377: Billington CJ, Briggs JE, Grace M, et al. Effects of intracerebroventricular injection of neuropeptide Y on energy metabolism. Am J Physiol 1991;26O:R Urban JH, Bauer-Dantoin AC, Levine JE. Neuropeptide Y gene expression in the arcuate nucleus: sexual dimorphism and modulation by testosterone. Endocrinology 1993;132: