| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Psychiatry (K.A.B., K.C.L., K.M.G., E.E.B.) and Medicine (A.L.H.), University of North Carolina, Chapel Hill, North Carolina 27599-7175; and Department of Biobehavioral Health (S.G.W.), Pennsylvania State University, University Park, Pennsylvania 16802
Address all correspondence and requests for reprints to: Kim Brownley, Ph.D., Department of Psychiatry, CB #7175, University of North Carolina, Chapel Hill, North Carolina 27599-7175. E-mail: kim_brownley{at}med.unc.edu.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Obesity is a growing epidemic in the United States, and recent projections indicate obesity will soon surpass smoking as the number one preventable cause of cardiovascular-related death in our society (22). According to the Centers for Disease Control and Prevention (23) surveillance statistics, from 1991 to 2000, obesity increased 61% among U.S. adults, with the major burden of that increase evident in non-Hispanic blacks (12% increase), compared with non-Hispanic whites (8% increase). Overall, blacks are approximately 1.6 times more likely to be obese than whites. Obesity is related to multiple cardiovascular disease risk factors (24, 25, 26, 27), and black women, who suffer higher rates of obesity, hypertension, and diabetes, for example, than their white female counterparts (28), may be especially vulnerable.
To our knowledge, differences in ghrelin level as a function of race have not been investigated. Postprandial ghrelin is of special interest because those individuals who fail to maintain suppression of ghrelin release after a meal may be more susceptible to eating again soon. Furthermore, study designs that control not just a single meal but also all meals for an extended period before testing can provide a more naturalistic assessment of postprandial ghrelin dynamics. Therefore, the purpose of the present study was to examine ghrelin, and its relation to markers of cardiovascular and metabolic function, in a biracial sample of women who were tested after 4 d on the same controlled diet.
| Subjects and Methods |
|---|
|
|
|---|
This study is based on 43 women, 21 black and 22 white, aged 3663 yr, who participated in a study of dietary sodium and blood pressure, which was approved by the local Biomedical Ethics Committee. Six women were prehypertensive or had stage 1 hypertension (based on resting laboratory readings consistently exceeding 130/85 mm Hg), but none of the 43 women were taking any antihypertensive medications, and all were otherwise healthy. Seventeen women were premenopausal; 26 were postmenopausal (13 with and 13 without current hormone replacement).
Procedures
Dietary intervention.
Each participant initially completed a 4-d carefully controlled dietary intervention, during which she maintained her usual daily activities and visited the Clinical Research Center every 2448 h to eat breakfast and pick up food and beverages for outpatient consumption. Meals were selected and prepared to maintain correct daily electrolyte and nutrient levels by the Clinical Research Center nutritionists. All meals, snacks, and beverages were provided for two alternating daily menus of familiar whole foods to achieve the desired levels (
60% carbohydrate, 28% fat, 12% protein; 220 mEq Na, 40 mEq K). Total caloric load was adjusted for body size (i.e. higher calorie content meals were available to heavier individuals) to prevent larger persons from entering a state of negative energy balance, and pre- vs. postdiet weight comparisons verified that no weight loss occurred. Participants were limited to one cup of caffeinated coffee per day and were instructed to refrain from use of caffeine and any medications on the day of testing. Collections of 24-h urinary sodium and potassium excretion were measured on d 4 and compared with gender- and race-based norms for creatinine clearance to assess dietary compliance. Compliance rates were very high (>97%), in part, because subjects were offered a substantial monetary bonus for full compliance.
On d 4, participants completed a laboratory test session that began between 1630 and 1830 h and lasted 4.5 h. After arriving at the laboratory, they consumed a standardized dinner and underwent a 2-h period of rigidly controlled water drinking and urine collection to promote a steady state of hydration and sodium excretion. During this 2-h stabilization period, they were required to drink 1.5 liters of water and void four times on a fixed schedule. Instrumentation and standardization occurred during this time period. After stabilization, three experimental periods occurred: baseline, continuous mental stress, and recovery, each 40 min in length. Data reported in this paper were obtained at the midpoint of the baseline rest period (i.e. 2 h after ingestion of the standardized meal was complete).
Cardiovascular measures. Cardiac performance was measured using impedance cardiography, and blood pressure (BP) was determined every other minute during min 1119 of the baseline rest (29, 30, 31), as previously described. Briefly, stroke volume (SV) was derived using the Kubicek equation (32), heart rate was determined from the mean interbeat interval, and cardiac output and total peripheral resistance were calculated using standard formulas (33). To remove the influence of differences in body size on these measurements, cardiac output, SV, and total peripheral resistance were indexed for body size (33) and are hereinafter referred to as cardiac index, SV index, and vascular resistance index. Auscultatory BP was assessed using an automated monitor with R-wave gating, which provides accurate measurements within ± 2 mm Hg between 0 and 300 mm Hg (model 4240, SunTech Medical Instruments, Raleigh, NC).
Bioassays. Blood samples for plasma measures were drawn into prechilled, EDTA-containing tubes at 140 min from the start of the protocol, 120 min after the subjects finished eating the standardized meal and the initial controlled water drinking. These tubes were rapidly cold centrifuged and the resulting plasma pipetted and frozen at 80 C within minutes of collection.
Plasma levels of active acylated ghrelin were measured by a competitive enzyme immunoassay (Peninsula Labs, Belmont, CA). The sensitivity of this assay is 0.08 ng/ml with a standard range of 025 ng/ml. The intra- and interassay variation is 5 and 14%, respectively. Total (bound and free) leptin was measured in unextracted plasma by competitive enzyme immunoassay (Peninsula Labs) using a two-step enzyme color generation system to determine relative binding of biotinylated leptin conjugate (competitive ligand) vs. the sample peptide on leptin-specific antibody binding sites. Sample OD was plotted against the standard curve to determine leptin content. The sensitivity of the assay is 0.4 ng/ml with a standard range of 0500 ng/ml. Intra- and interassay variation is 7 and 10.2%, respectively. Urinary cortisol was determined using the Corte-Cote RIA (ICN Biomedicals, Inc., Aurora, OH). The sensitivity of the assay is 0.07 µg/dl. Intra- and interassay variation is 4.7 and 7.4%, respectively. Plasma norepinephrine was determined using HPLC with electrochemical detection (University of North Carolina General Clinical Research Center).
Statistical analysis
Average baseline values were calculated as the mean of min 15, 17, and 19. Ghrelin, leptin, cortisol, and norepinephrine data were log transformed before analysis. Group differences were determined using
2 and ANOVA, correlations were computed using Pearsons product moment, and linear regression was used to evaluate relationships between continuous variables. All analyses were carried out using SAS software (version 8.0, SAS Institute, Cary, NC). Results were expressed as the mean ± SD unless otherwise stated. P < 0.05 was considered statistically significant.
| Results |
|---|
|
|
|---|
|
|
2 >0.66). Race and obesity group differences in ghrelin
Effect of race.
Ghrelin levels differed as a function of racial group: mean (± SD) ghrelin was higher in black, compared with white, women (0.33 ± 0.16 vs. 0.24 ± 0.12 ng/ml). As seen in Fig. 1
(bottom), this racial group difference in ghrelin was in large part due to obese black women, who exhibited the highest average ghrelin level overall. However, the racial group difference in ghrelin remained significant after univariate adjustment for BMI, body weight, age, and leptin differences (P < 0.05), suggesting that race was not simply a proxy for BMI-related or caloric-intake-related differences in ghrelin.
Effect of obesity.
Ghrelin did not differ significantly between obese and nonobese subjects (0.31 ± 0.17 vs. 0.27 ± 0.12 ng/ml, respectively), and, unexpectedly, ghrelin was not inversely related to general obesity (as indexed by BMI) in this sample of women. Instead, central obesity, as measured by waist-hip ratio, did show the expected inverse correlation with ghrelin in white women (r = 0.43, P < 0.04), but this relationship was not present in black women (r = 0.09, P > 0.67). Similar to what was observed for ghrelin and leptin levels, this lack of the typical relationship between ghrelin and waist-hip ratio was largely due to obese black women (Fig. 2
).
|
As shown in Fig. 3
, ghrelin was strongly and significantly correlated with 24-h urinary cortisol in black women (r = 0.61, P < 0.004) but not in white women (r = 0.21, P > 0.34). In multivariate regression analysis, a significant race X ghrelin interaction was detected for cortisol (ß = 1.53, P < 0.007), and this effect remained intact after adjusting for age, BMI, waist-hip ratio, and leptin levels and removing the influence of one obese white womans cortisol value that was deemed an outlier (P < 0.03). Ghrelin was also associated with higher resting heart rate in white women (r = 0.44, P < 0.04) but not black women (r = 0.09, P > 0.68), yielding a marginally significant race X ghrelin interaction term for heart rate (P = 0.10).
|
|
Independently, ghrelin was elevated in black women and leptin levels were elevated in obese women, and the subset of obese, black women exhibited the highest overall ghrelin and leptin levels. In addition, in this same subset of obese black women, ghrelin levels appeared to be independent of central obesity, whereas in white women, the expected inverse relationship between central obesity and ghrelin was obtained.
The relationships between ghrelin and cortisol and between ghrelin and heart rate likewise differed as a function of race. Higher ghrelin was associated with higher cortisol only in black women and higher resting heart rate only in white women. Higher ghrelin was linked to higher diastolic BP and lower norepinephrine in both black and white obese women.
| Discussion |
|---|
|
|
|---|
Fasting ghrelin level is correlated with insulin resistance and inversely related to insulin level (9, 15, 34); controlling for insulin level or resistance negated the relationship between ghrelin and BMI (15). Higher insulin resistance has been observed in black, compared with white, women (35), and BMI and ghrelin were unrelated in the present study, particularly among obese black women, suggesting the possibility of a higher level of insulin resistance in that subgroup. Previous studies (36, 37) demonstrated a positive correlation between leptin and insulin resistance in women; thus, our finding of elevated leptin in black women, and particularly in obese black women, further supports the speculation of insulin resistance in that subgroup. Although not directly evaluated in the present study, insulin deficiency or insulin resistance may have accounted for the higher ghrelin level observed in black women. Future studies to elucidate racial differences in the connection between insulin resistance and ghrelin would be worthwhile.
Ghrelin is normally suppressed after meal intake, and this effect is at least in part dependent on insulin and obesity such that subnormal suppression of ghrelin might be anticipated in obese, insulin-resistant individuals (9, 13, 14, 38, 39) and type 2 diabetics (40). In addition, the return of ghrelin to premeal levels may be accelerated in obese individuals and diabetic patients (10). In the present study, ghrelin levels were measured 2 h after meal ingestion, a point at which ghrelin levels typically remain suppressed relative to premeal levels in healthy individuals but may already be rebounding in metabolically at-risk individuals (8, 10). Given the higher incidence of obesity and excess diabetes-related mortality in the larger population from which our sample was drawn (blacks living in the southeastern United States and in North Carolina in particular), it is not unreasonable to suspect that higher ghrelin levels observed in black women in this study were due, at least in part, to subnormal postprandial ghrelin suppression and/or accelerated ghrelin rebound.
It has been proposed that lesser postprandial ghrelin suppression and/or a more rapid postprandial ghrelin rebound may contribute to increased food intake in obesity (9) and hyperphagia that is common in insulin-deficient states (41, 42). Ghrelin promotes carbohydrate consumption and reduced fat depot use (43), and a high ghrelin level is associated with faster gastric emptying (8). According to a recent study (44), after a low- vs. high-caloric-load meal, ghrelin suppression is less dramatic, and the time to subsequent spontaneous request for more food is shorter. However, the findings did not support a definitive link between the extent of ghrelin suppression and time to spontaneous request for food. Given that all participants in the present study ate a controlled diet consisting of standard mixed-nutrient, fixed-calorie meals for 4 d and also 2 h before the measurement of ghrelin levels, it is unlikely that nonrandom differences in relative caloric load (or in macronutrient content) could explain the observations of higher ghrelin and higher leptin in black women. Because total caloric load was also adjusted for body size, it is also unlikely that ghrelin levels were higher in obese black women due to a systematic study-induced state of negative energy balance (11, 12).
Ghrelin administration induces the release of ACTH and cortisol (45). In the present study, ghrelin level correlated highly with 24-h urinary cortisol but only in black women. Elevated cortisol is a common finding in obesity (46, 47) and in patients with peripheral vascular disease (48) and diabetes mellitus, particularly those who are in poor control (49). Among women in particular, higher cortisol is independently associated with coronary stenosis (50), and adverse cardiovascular risk is greatest in those with the combination of obesity and failure to down-regulate cortisol levels (51). By self-report, all women in the present study had negative histories for coronary artery and peripheral vascular disease as well as diabetes, but it is not known whether black and white women differed in the incidence or severity of subclinical disease. Elevated cortisol in obesity, per se, is in large part due to hypothalamic dysregulation, particularly a disruption in the cortisol-ACTH feedback system (52). Peripheral ghrelin produced in the gastrointestinal tract stimulates GH release and regulates energy balance via its actions in the mediobasal and mediolateral hypothalamus (4, 5, 6). Ghrelin is also produced in hypothalamic neurons (5, 6). Thus, the elevations in ghrelin observed in black women and the strong ghrelin-cortisol relationship might reflect alterations in the control of ghrelin secretion (41, 42) and may be a precursor or marker of existing generalized hypothalamic dysregulation leading to subclinical disease in this group. Future angiographic studies could be instrumental in directly addressing this question.
Recent findings suggest that ghrelin exerts cardiovascular and hemodynamic effects, in part, via its actions on renal sympathetic nerve activity and vasopressin regulation (19, 20, 21). In healthy male subjects, acute ghrelin administration reduced mean arterial pressure and increased SV and cardiac index (18); and fasting plasma ghrelin was inversely associated with several indexes of right ventricular function (17). Our finding of higher heart rate associated with ghrelin in white women appears consistent with these findings, yet our data suggesting that higher ghrelin is associated with higher diastolic BP and lower plasma norepinephrine in obese women is not. Individual differences in response to the dietary sodium load during the 4 d before testing could contribute to these latter findings. Salt loading is associated with decreased urinary norepinephrine (53) and is linked to diastolic BP increases in persons who are salt sensitive, a trait that may be more common in obese individuals (54). The effects of salt loading on ghrelin have not been examined, but there are limited data suggesting a role of sodium-dependent channels in ghrelins stimulatory effect on intracellular calcium (55). Although the present study did not evaluate salt sensitivity, per se, it is reasonable to speculate a renal/sodium-mediated ghrelin-cardiovascular connection underlying the race and obesity differences in our data. Taken together, this collection of studies suggests a need for additional investigations of sodium load (along with potassium, calcium, and magnesium loads) on ghrelin and associated cardiovascular effects, and such studies should take into account the interactive effects of race, feeding state, and gender.
The strengths of the present study include the novel examination of race as a potential modifier of ghrelin, and the timing of metabolic and cardiovascular measurements after a rigidly timed postmeal ingestion protocol. In addition, whereas previous studies have focused on changes in fasting ghrelin after a single meal or ghrelin administration, the present study reflects a more naturalistic challenge that ensued after a 4-d controlled diet representative of local and regional dietary intake (56). Limitations of this study include the absence of insulin and insulin resistance measures and the absence of premeal and more frequent postmeal ghrelin assessments to better characterize interim ghrelin dynamics. Collection of such data in future studies will play an integral role in clarifying whether the current observations reflect ethnic differences in chronic ghrelin levels in immediate premeal levels, postprandial suppression and rebound, or some combination thereof.
In summary, in black women, independent of BMI, ghrelin was elevated and higher ghrelin was associated with higher cortisol. These findings were especially noteworthy in the subset of obese black women, who also exhibited significantly elevated leptin. Previously our research group demonstrated differences between blacks and whites in physiological responses that may contribute to increased cardiovascular risk. Prior observations include increased vascular resistance during stress, increased left ventricular wall thickness for the same level of blood pressure, decreased ß-receptor responsivity, and slowed sodium excretion during stress in blacks (57, 58, 59, 60). Blacks living in the southeastern United States, particularly in North Carolina in which this study was conducted, experience higher rates of overweight/obesity and suffer total mortality burdens from heart disease, diabetes, and stroke in excess of whites. The burden of premature death (before age 65 yr) due to heart disease is particularly high among black women (21.7%), compared with their white counterparts (7.7%). The link between ghrelin and cortisol and subsequently between cortisol and these disease states suggests that further investigation of the role of ghrelin in cardiovascular morbidity among black women is warranted.
| Footnotes |
|---|
Abbreviations: BMI, Body mass index; BP, blood pressure; SV, stroke volume.
Received March 30, 2004.
Accepted June 8, 2004.
| References |
|---|
|
|
|---|
- and ß-adrenergic receptor agonists: effects of race in borderline hypertensive compared with normotensive men. Am J Hypertens 6:630635[Medline]This article has been cited by other articles:
![]() |
F. Bacha and S. A. Arslanian Ghrelin and Peptide YY in Youth: Are There Race-Related Differences? J. Clin. Endocrinol. Metab., August 1, 2006; 91(8): 3117 - 3122. [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 |