| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Children’s Hospital of Pittsburgh, Weight Management and Wellness Center, and Division of Pediatric Endocrinology, Metabolism, and Diabetes Mellitus, Pittsburgh, Pennsylvania 15213
Address all correspondence and requests for reprints to: Silva A. Arslanian, M.D., Director, Weight Management and Wellness Center, Children’s Hospital of Pittsburgh, 3705 Fifth Avenue at DeSoto Street, Pittsburgh, Pennsylvania 15213. E-mail: silva.arslanian{at}chp.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Research Design and Methods: We investigated: 1) ghrelin suppression/PYY elevation in response to an oral glucose tolerance test (OGTT) in AA vs. AW, and 2) the relationship of ghrelin and PYY dynamics to insulin sensitivity. Thirty-three AA and 54 AW prepubertal children underwent an OGTT measuring ghrelin, PYY, glucose, and insulin. Fasting glucose to insulin ratio (GF/IF) was used to assess the relationship of insulin sensitivity to fasting and post-OGTT ghrelin and PYY levels.
Results: OGTT-induced suppression in ghrelin (
ghrelin) was lower in AA youth. ghrelin correlated with GF/IF (r = 0.47, P < 0.001) and insulin at 30 min (r = –0.47, P < 0.001). In multiple regression analysis, race (P = 0.013) and GF/IF (P = 0.004) contributed independently to the variance in ghrelin (R2 = 0.28, P < 0.001). Fasting and post-OGTT PYY levels were lower in AAs and were not related to insulin sensitivity.
Conclusion: The lower suppression of ghrelin in AA, but not the lower PYY levels, correlates with insulinemia and insulin resistance. Less ghrelin suppression and PYY elevation after a meal in black youth could be a potential mechanism of race-related differences in hunger/satiety predisposing to risk of obesity.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Race-related differences in ghrelin are suggested by studies revealing lower ghrelin levels in Pima Indians vs. Caucasians (4, 5). Also, a more recent study demonstrated elevated 2-h postmeal ghrelin levels in black vs. white women (6).
Glucose and insulin metabolism appear to be involved in the regulation of ghrelin levels. An inverse pattern of ghrelin and insulin levels has been detected during 24-h observation (2) and during hyperinsulinemic-euglycemic clamp studies (7). Moreover, an inverse relationship between fasting ghrelin and insulin resistance indices has been reported in adults and children (8, 9). We recently reported that post-oral glucose tolerance test (OGTT) ghrelin dynamics are impaired in overweight compared with normal-weight children and are modulated by insulin sensitivity (10).
PYY infusion to mimic postprandial levels inhibits food intake (3). Similar to ghrelin, PYY levels are reduced in obese subjects who also have a reduced postprandial rise in PYY (11). It is not clear, however, what regulates PYY levels acutely after food ingestion.
We hypothesized that race-related differences in ghrelin and PYY dynamics could be a potential mechanism responsible for the increased risk of obesity in blacks. More specifically, we theorized that: 1) in black children, there is less suppression in ghrelin levels and less elevation in PYY after food intake contributing to sustained hunger and lack of satiety; and 2) racial differences in ghrelin and PYY dynamics are modulated by insulin sensitivity. Therefore, we investigated ghrelin suppression and PYY elevation in response to an OGTT in AA vs. AW children and the relationship of ghrelin and PYY dynamics to insulin sensitivity.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study population consisted of 82 healthy prepubertal children, 7–12 yr of age. They included 33 AA (19 males and 14 females) and 54 AW (27 males and 27 females) children with a wide range of body mass index (BMI) percentiles for age and sex (fifth to 99th percent). AA or AW descent was verified by self-identification in three generations. Some of these children were reported in one of our previous publications (10). All studies were approved by the Institutional Review Board of The University of Pittsburgh. Children were recruited through newspaper advertisement and flyers posted in the health center. All research participants and their parents or guardians gave informed consent/assent. All subjects were documented to be in good health by a thorough medical interview and physical examination. Children at risk of overweight (BMI 85th to 95th percentile) and overweight (BMI > 95th percentile for age and sex) were free of any associated comorbidities or syndromes leading to obesity. Subjects were not receiving any medications. All subjects were assessed to be in Tanner stage 1 puberty by careful physical examination and confirmatory hormonal levels. Characteristics of study subjects are detailed in Table 1
.
|
After an overnight fast of 10–12 h, subjects were studied at the General Clinical Research Center of Children’s Hospital of Pittsburgh. Anthropometric measurements including height and weight were obtained without shoes and in light clothing. Children received an OGTT (1.75 g/kg, maximum 75 g). The average dose of glucose administered was 64.9 ± 12.9 g in AA and 64.3 ± 12.7 g in the AW group. Blood samples were drawn at 0, 15, 30, 60, 90, 120, and 180 min for determination of glucose, insulin, PYY, and ghrelin levels. Fasting IGF binding protein-1 (IGFBP-1) levels were determined. Fasting estradiol (in females) and testosterone (in males) were obtained to confirm pubertal staging. Body composition was assessed by dual-energy x-ray absorptiometry using an absorptiometer (Lunar, Madison, WI) (10).
Measurements
Plasma glucose was measured by the glucose oxidase method with the use of a YSI glucose analyzer (Yellow Springs Instruments, Yellow Springs, OH). Plasma insulin was measured by RIA (Linco Research Inc., St. Charles, MO) as reported previously (10). IGFBP-1 levels were measured by immunochemiluminescent assay (Esoterix, Inc., Calabasas Hills, CA), as reported by us previously (10). Estradiol level was measured by chemiluminescent assay. A level less than 21 pg/ml (77.1 pmol/liter) is consistent with prepubertal status. Testosterone panel including total and free testosterone was performed by HPLC tandem mass spectrometry (Esoterix). A level less than 18 ng/dl (624.1 pmol/liter) was considered to be consistent with prepubertal status. PYY levels were determined by RIA (Linco Research), which uses an antibody that recognizes both 1–36 and 3–36 forms of human PYY with less than 0.1% cross-reactivity with pancreatic polypeptide or neuropeptide Y. The intraassay and interassay coefficients of variation are 3.7–8.1% and 9.2–10.9%, respectively. Ghrelin levels were determined by RIA (Linco Research), which is specific for total ghrelin as reported by us before (10).
Calculations
Fasting glucose to insulin ratio (GF/IF) was used as a surrogate estimate of insulin sensitivity as described by us before (12). This estimate was used to assess the relationship of insulin sensitivity to fasting ghrelin and PYY levels and assess the relationship of insulin sensitivity to ghrelin suppression and PYY elevation during the OGTT. Ghrelin, PYY, insulin, and glucose area under the curve (AUC) during the OGTT was calculated by the trapezoidal rule.
Statistical analysis
The distribution of the different variables was examined and the appropriate parametric or nonparametric statistical test applied. Student’s t test or Mann-Whitney test was used for two group comparisons. Pearson or Spearman’s correlations were used to examine bivariate relationships. Multiple regression analysis was used to examine multivariate relationships. Repeated-measures ANOVA was used for comparison of ghrelin and PYY levels at different time points after OGTT. Ghrelin levels reached a statistical nadir at 60 min, and glucose and insulin peaked at 30 min during the OGTT. PYY levels reached a peak at 15 min after the glucose load. P < 0.05 was considered statistically significant. Results are reported as mean ± SD.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The two groups had similar age and pubertal stage. The AA and AW children had similar BMI and body composition (Table 1
). There were no significant differences between the two groups in serum testosterone [5.9 ± 3.7 ng/dl (204.7 ± 128.2 pmol/liter) in AA vs. 5.6 ± 2.9 ng/dl (194.3 ± 100.5 pmol/liter) in AW], free testosterone [0.6 ± 0.4 ng/ml (2.1 ± 1.4 pmol/liter) in AA vs. 0.7 ± 0.4 ng/ml (2.4 ± 1.4 pmol/liter) in AW] among the males, or estradiol among the females [13.6 ± 5.3 pg/ml (49.9 ± 19.5 pmol/liter) in AA vs. 13.6 ± 6.4 pg/ml (49.9 ± 23.5 pmol/liter) in AW].
Fasting data
Fasting ghrelin was similar in both groups (Table 1). Fasting PYY levels were significantly lower in the AA group (Table 1). There was no statistically significant difference in fasting glucose [84.1 ± 7.1 mg/dl (4.7 ± 0.4 mmol/liter) in AA vs. 83.6 ± 5.6 mg/dl (4.6 ± 0.3 mmol/liter) in AW], fasting insulin (18.9 ± 25.2 µU/ml in AA vs. 16.1 ± 18.0 µU/ml in AW), or GF/IF (6.5 ± 2.5 in AA vs. 8.0 ± 5.3 in AW) between the two groups. However, when the normal-weight (NW) children are analyzed separately, fasting insulin is significantly higher [10.7 ± 2.7 µU/ml (64.2 ± 16.2 pmol/liter) vs. 9.7 ± 5.6 µU/ml (58.2 ± 33.6 pmol/liter), P = 0.03], and GF/IF is lower (8.1 ± 1.9 vs. 10.7 ± 6.1, P = 0.04) in AA children. IGFBP-1 level was significantly lower in the AA children (21.6 ± 20.2 vs. 51.6 ± 47.0 ng/ml, P = 0.002).
There was no significant difference in fasting ghrelin or PYY levels between males and females in each racial group. There were no significant differences in fasting PYY levels in NW vs. obese children within each race (135.1 ± 29.3 and 144.8 ± 24.1 pg/ml in AA and 165.4 ± 55.6 and 173.1 ± 59.3 pg/ml in AW, NW vs. obese, respectively).
OGTT data
Ghrelin levels reached a nadir at 60 min after oral glucose load in AA and AW children. The mean absolute suppression in ghrelin was 76.4 ± 51.7 pmol/liter in AA vs. 112.6 ± 83.9 pmol/liter in AW children (P = 0.034). Percent suppression in ghrelin was also significantly lower in AA vs. AW children (18.3 ± 8.6 vs. 24.3 ± 10.7%, P = 0.007) (Fig. 1). Ghrelin AUC was not significantly different between AA and AW children. Insulin peaked at 30 min after the glucose load with insulin at 30 min of 110.3 ± 79.5 µU/ml (661.8 ± 477.0 pmol/liter) in AA and 75.3 ± 53.7 µU/ml (451.8 ± 322.2 pmol/liter) in AW (P = 0.025). Glucose levels were not significantly different between AA and AW youth at 0, 15, or 30 min during the OGTT. After the glucose peaked at 30 min, glucose levels were significantly lower in AA children. Subsequently glucose AUC was lower in AA children (18,810.5 ± 2,293.6 vs. 20,373.8 ± 3,067.1 mg/dl·min, P = 0.03). After the glucose load, PYY levels were significantly lower in the AA group at all time points measured except 120 min (P = 0.065) (Fig. 2). As a result, PYYAUC was significantly lower in the AA vs. the AW group (24,569.4 ± 5,340.9 vs. 30,425.0 ± 9,339.4 pg/ml·min, P = 0.002) (Fig. 2). After adjusting for lower glucose AUC in AA youth, PYYAUC remained significantly lower in AAs (P = 0.002).
|
|
Fasting ghrelin levels correlated inversely with BMI (r = –0.56, P < 0.001), BMI Z-score (r = –0.54, P < 0.001), percent body fat (r = –0.50, P < 0.001), and fat mass (r = –0.51, P < 0.001). Fasting ghrelin levels correlated positively with GF/IF (r = 0.55, P < 0.001) and inversely with fasting insulin levels (r = –0.53, P < 0.001). The correlation of fasting ghrelin with fasting insulin levels and GF/IF persisted after controlling for BMI (r = –0.25, P = 0.02 and r = 0.25, P = 0.02, respectively) or BMI Z-score (r = –0.29, P = 0.007 and r = 0.29, P = 0.006, respectively). Ghrelin AUC during the OGTT correlated inversely with insulin AUC (r = –0.57, P < 0.001) in the total group of subjects and in each racial group separately (r = –0.50, P = 0.003 in AA and r = –0.63, P < 0.001 in AW). Ghrelin suppression at 60 min correlated inversely with insulin increment at 30 min (r = –0.47, P < 0.001) and positively with GF/IF (r = 0.47, P < 0.001). When the same correlations were done using percent suppression in ghrelin at 60 min, it correlated inversely with insulin (r = –0.25, P = 0.021) but not GF/IF. IGFBP-1 as a marker of insulin sensitivity correlated with fasting ghrelin (r = 0.49, P < 0.001), absolute ghrelin suppression (r = 0.50, P < 0.001), and percent ghrelin suppression (r = 0.24, P = 0.025) at 60 min. The correlation of fasting ghrelin and ghrelin suppression to fasting IGFBP-1 levels persisted after controlling for BMI (r = 0.21, P = 0.055 and r = 0.23, P = 0.04, respectively) or BMI Z-score (r = 0.26, P = 0.018 and r = 0.23, P = 0.037, respectively).
Fasting PYY levels correlated with fasting ghrelin (r = –0.24, P = 0.03) and fasting glucose (r = 0.2, P = 0.07) but not GF/IF, insulin, BMI, BMI Z-score, or measures of body composition. PYY at its peak at 15 min correlated with glucose at 15 min (r = 0.29, P = 0.017). PYYAUC did not correlate with ghrelin, insulin, or glucose AUC.
Multiple regression analysis
In a multiple regression analysis, fasting insulin (beta = –0.31, P = 0.011) or GF/IF (beta = 0.31, P = 0.010) and percent body fat (beta = –0.33, P = 0.006) contributed to the variance in fasting ghrelin (R2 = 0.33, P < 0.001) independent of race. In a multiple regression analysis with race, GF/IF, and percent body fat as independent variables, GF/IF (beta = 0.36, P = 0.004) and race (beta = 0.21, P = 0.03), together and independently explained approximately 30% of the variance in absolute ghrelin suppression at 60 min (R2 = 0.28, P < 0.001). With percent ghrelin suppression as the dependent variable and race, GF/IF (or IGFBP-1), and percent body fat as independent variables, only race (beta = 0.29, P = 0.008) contributed significantly to the variance in percent suppression in ghrelin (P = 0.05). With PYYAUC as the dependent variable and percent body fat, GF/IF, and race as independent variables, only race (beta = 0.35, P = 0.001) contributed significantly to the variance in PYYAUC (R2 = 0.13, P = 0.009). With race and glucose AUC as independent variables, only race contributed significantly to the variance in PYYAUC (R2 = 0.11, P = 0.007).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
AA children have been shown to be hyperinsulinemic and insulin resistant, compared with their white peers of similar body composition (13). In this study, consistent with previous reports, AA children were hyperinsulinemic in response to the OGTT and had significantly lower IGFBP-1 levels, a measure of insulin resistance (14). This race-related difference in insulin sensitivity is evident in NW youth but not obese youth as reported by us previously (13, 15) and consistent with the present findings. It is our belief that the race-related difference in insulin sensitivity, which is of relatively small magnitude, is overshadowed by the much larger influence of obesity-related insulin resistance.
Ghrelin suppression was significantly lower in AA children and correlated inversely with insulin response to the OGTT and positively with insulin sensitivity after adjusting for body composition. Fasting ghrelin levels were not significantly different between the two racial groups and correlated with fasting insulin sensitivity measures. We previously demonstrated that fasting ghrelin and the dynamics of ghrelin response to glucose are modulated by the degree of insulin sensitivity. Thus, the more insulin resistant the child, the lower the fasting ghrelin and the lower the degree of suppression in ghrelin with OGTT (10). The current study extends our previous findings by demonstrating racial differences in absolute and percent suppression in ghrelin, which are independent of racial differences in insulin sensitivity and hyperinsulinemia. Our results are in agreement with many studies that reported an inverse relationship between fasting ghrelin levels and fasting insulin and insulin resistance indices in adults (4, 8) and children (9, 16). Also, ghrelin levels are lower in conditions associated with insulin resistance such as polycystic ovarian syndrome (17) and type 2 diabetes (18, 19). Moreover, ghrelin suppression during hyperinsulinemic clamp studies has been shown to be positively associated with insulin sensitivity (18).
Race-related differences in fasting ghrelin have been noted in a previous study reporting lower ghrelin levels in Pima Indians, compared with Caucasians of similar BMI (4). In that study, ghrelin levels were inversely related to fasting insulin. Salbe et al. (5) reported that fasting ghrelin levels are 50% lower in Pima Indians vs. Caucasians and remained lower after adjusting for age, gender, and body weight. In a study that measured a single postprandial ghrelin level 2 h after a standardized meal in AA and AW women, postprandial ghrelin levels were higher in AA women and more so in obese AA women. This difference remained significant after adjusting for BMI, age, and leptin levels (6). The authors speculated about an influence of insulin resistance on their findings. However, insulin levels were not determined.
Our findings confirm that there are racial differences in ghrelin regulation in children. In response to nutrients, there is less suppression in ghrelin levels in AA youth. This difference occurs despite significantly higher insulin excursion in AA children, suggesting resistance to insulin’s effect in lowering ghrelin in AA youth. However, in a multiple regression analysis, both race and insulin sensitivity contribute independently to the variance in ghrelin suppression. This could translate into impaired satiety in AA children. However, there are no epidemiological data regarding difference in quantities of food consumed between AA and AW children. In addition, we measured only total ghrelin levels and not acylated ghrelin. Even though total ghrelin is a good surrogate of active ghrelin, this may not be the case when addressing differences in two genetically different ethnic groups.
With respect to PYY, the "satiety" hormone, AA children have lower fasting and postprandial levels, which are independent of obesity and not related to measures of insulin sensitivity. Peptide YY is released from L-type cells in the distal gastrointestinal tract and increases after meal ingestion (20). It inhibits food intake in rodents (3, 21) and monkeys (22) and reduces weight gain in rodents (3), although these effects were not replicated in other studies in rodents (23). In humans, PYY infusion reduced food intake in 12 NW and 12 obese individuals (11). In that study, PYY levels were inversely related to BMI, and PYY infusion decreased ghrelin levels. In agreement with Batterham et al. (11), PYY levels increased in all our subjects after OGTT: in both NW and obese and both races. However, our results differ in that in our group of young subjects, there were no significant differences in PYY levels between NW and obese children and no significant relationship between PYY levels and BMI or percent body fat.
Similar to our findings, no correlation was found between PYY and BMI or insulin levels in 30 female adolescents with anorexia nervosa, NW, and obesity (24). In our study, the differences in fasting PYY and postprandial PYY levels were mainly related to the ethnicity of the subjects. The underlying mechanism for this racial difference is not explained by differences in insulin or postprandial ghrelin levels, but rather appears to be influenced by postprandial glucose levels. Similar to our findings, the above studies did not find a relationship between PYY levels and insulin (24), and there was no effect of PYY infusion on insulin levels (11). PYY levels correlated with fasting ghrelin levels in our subjects. However, unlike the study by Batterham et al. (11), there was no relationship between ghrelin and PYY postprandially. The difference in our results could be due to differences in the relationship after the physiological stimulant of glucose ingestion vs. a PYY infusion in their study.
In support of our findings, a study in mice revealed that the effect of PYY injections to reduce short-term food intake was not affected by endogenous ghrelin levels (21). In this study, we measured total PYY levels. Even though PYY 3–36 has been associated with satiety effects, a recent study revealed that PYY 1–36 also inhibits food intake in rodents (25). Our findings underscore the need for future investigations to clarify the mechanisms of regulation of PYY. A genetic basis for the observed racial differences is possible, given recent reports of PYY variants and polymorphisms in the Y2 receptor that were found to be related to severe obesity in Pima Indian men (26).
In summary, the results of this study reveal that AA children have lower absolute and percent suppression in ghrelin levels in response to feeding, compared with their AW peers. Ghrelin suppression in AA children appears to be resistant to the effect of insulin in lowering ghrelin. In addition, AA children have lower fasting and postprandial levels of PYY. This alteration in PYY levels and meal-induced ghrelin suppression could be responsible for differences in hunger and/or satiety between AA and AW children and could be a potential mechanism for the increased risk of obesity in blacks.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online May 23, 2006
Abbreviations: AA, African-American; AUC, area under the curve; AW, American white; BMI, body mass index; GF/IF, fasting glucose to insulin ratio; IGFBP, IGF binding protein; NW, normal weight; OGTT, oral glucose tolerance test; PYY, peptide YY.
Received November 9, 2005.
Accepted May 16, 2006.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. Z. Kasa-Vubu, A. Rosenthal, E. G. Murdock, and K. B. Welch Impact of Fatness, Fitness, and Ethnicity on the Relationship of Nocturnal Ghrelin to 24-Hour Luteinizing Hormone Concentrations in Adolescent Girls J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3246 - 3252. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Germain, B. Galusca, C. W Le Roux, C. Bossu, M. A Ghatei, F. Lang, S. R Bloom, and B. Estour Constitutional thinness and lean anorexia nervosa display opposite concentrations of peptide YY, glucagon-like peptide 1, ghrelin, and leptin Am. J. Clinical Nutrition, April 1, 2007; 85(4): 967 - 971. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-P. Chanoine Individual Differences in the Hormonal Control of Appetite: A Step toward a (More) Successful Treatment of Childhood Overweight? J. Clin. Endocrinol. Metab., August 1, 2006; 91(8): 2864 - 2866. [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 |
