| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Neuroendocrine Unit (M.M., K.K.M., V.S., E.H., K.K., A.K.), Pediatric Endocrine Unit (M.M.), and Eating Disorders Unit (D.B.H.), Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Anne Klibanski, M.D., BUL 457, Neuroendocrine Unit, Massachusetts General Hospital, 55 Fruit Street, Boston, Massachusetts 02114. E-mail: aklibanski{at}partners.org.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: We hypothesized that ghrelin may have independent effects on bone in AN adolescents.
Study Design, Subjects, and Outcome Measures: Frequent sampling was performed overnight every 30 min for 12 h in 23 adolescent AN girls aged 12–18 yr and 21 controls of comparable maturity. Ghrelin, leptin, cortisol, and GH secretion were examined using Cluster and deconvolution. We measured BMD and body composition (dual-energy x-ray absorptiometry) and carboxy-terminal peptide of type I procollagen and N-telopeptide levels.
Results: In healthy adolescents, ghrelin secretion strongly predicted BMD; secretory burst mass being the strongest predictor of lumbar spine (LS) bone mineral apparent density (BMAD) (r = 0.66, P = 0.003), LS BMAD z-scores (BMAD-z) (r = 0.59, P = 0.01), hip BMD (r = 0.55, P = 0.02), and hip BMD-z (r = 0.52, P = 0.03). When body composition measures (body mass index, lean and fat mass), and hormonal predictors (GH, IGF-I, cortisol, leptin, and estradiol) were entered into a regression model with ghrelin secretion to determine independent BMD predictors, ghrelin was the strongest predictor of LS BMAD, BMAD-z, hip BMD, and hip BMD-z, contributing to 43, 30, 26, and 19% of the variability, respectively, independent of GH or cortisol effects. Conversely, in AN, ghrelin secretion did not predict LS BMAD or hip-z and weakly predicted LS BMAD-z and hip BMD. Ghrelin did not predict carboxy-terminal peptide of type I procollagen or N-telopeptide/creatinine, which were predicted by GH and cortisol.
Conclusion: Ghrelin secretion predicts bone density independent of body composition, the GH-IGF-I axis, cortisol, or estradiol in healthy girls but not in those with AN.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Ghrelin is an orexigenic hormone (10, 11) that is also a GH secretagogue (12, 13, 14, 15, 16, 17, 18, 19). In addition, ghrelin increases secretion of ACTH (16, 17, 20, 21). We have demonstrated that ghrelin levels are elevated in girls with AN (22, 23), and ghrelin predicts GH and cortisol secretory burst frequency (23). Effects of ghrelin on tissues and organs other than the pituitary gland and hypothalamus have also been reported, including effects on adipogenesis (15). In addition, ghrelin mRNA is expressed in cartilage (24), and ghrelin administration increases osteoblast proliferation in vitro (25). The relationship between ghrelin parameters and bone metabolism in humans, however, has not been examined.
We hypothesized that ghrelin may have effects on bone metabolism that are independent of effects of known predictors and that high ghrelin levels in AN may contribute to low BMD observed in this condition, a unique model of undernutrition.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We examined 23 adolescent girls with AN (Diagnostic and Statistical Manual IV criteria) and 21 healthy adolescent girls. The two groups did not differ for chronological age (16.2 ± 1.6 vs. 15.4 ± 1.8 yr, probability value not significant) or for bone age (15.8 ± 1.5 vs. 15.7 ± 2.1 yr, probability value not significant). Mean duration of AN was 7.9 ± 10.5 months. Clinical and hormonal characteristics of our subjects have been reported earlier (8, 9, 22, 23, 26), but the relationship between ghrelin and bone metabolism has not been previously reported. Girls with AN were recruited through referrals from providers and Eating Disorder Centers in the New England area. Healthy adolescents were recruited through advertisements within the Partners HealthCare system, newspapers, and mailings to primary care providers. No healthy adolescent had a present or past history of an eating disorder. Informed assent and consent were obtained from all subjects and parents per Institutional Review Board recommendations. Use of medications and presence of disorders (other than AN) that may affect GH, cortisol, leptin or ghrelin levels were exclusion criteria for the study.
Study protocol
After a screening visit to confirm eligibility for the study, subjects were admitted overnight to the General Clinical Research Center (GCRC) for frequent blood sampling every 30 min from 2000–0800 h for ghrelin, leptin, GH, and cortisol. GH, cortisol, and leptin data were available for all subjects, and ghrelin data for 21 girls with AN and 18 controls. All subjects had supper before 1900 h and were fasting thereafter. Girls with AN and controls were allowed to choose the dinner menu to mimic their natural intake rather than induce an artificial state of intake for the day. Weight was measured in a hospital gown on a single electronic scale in the fasting state the following morning at the GCRC. Height was measured in triplicate on a single stadiometer and the average of three readings taken. BMI was calculated as (weight in kilograms)/(height in meters)2. Fasting blood samples were obtained at 0800 h for IGF-I, estradiol, and levels of carboxy-terminal peptide of type I procollagen (PICP) (bone formation marker). A second morning urine sample 2 h after the first void was obtained for N-telopeptde (NTX) (bone resorption marker) and creatinine (cr). A bone age was obtained to determine skeletal maturity. Bone age was read by a single observer, a pediatric endocrinologist, using the methods of Greulich and Pyle (27). Bone density testing at the lumbar spine and hip was performed by dual energy x-ray absorptiometry (DXA) (Hologic 4500; Hologic, Waltham, MA). Lumbar bone mineral apparent density (BMAD), a surrogate for volumetric bone density, was calculated using the formula described by Katzman et al. (28). z-Scores were obtained using the bone density applet of Bachrach, Hastie, and Narasimhan (http://www-stat-class.stanford.edu/pediatric-bones). DXA was also used to determine fat mass and lean body mass.
We examined ghrelin, leptin, GH, and cortisol data from frequent sampling by both Cluster (1 x 2) and deconvolutional analysis using methods previously described to determine concentration and secretory characteristics (8, 9, 29, 30). Concentration characteristics reported for ghrelin include area under the curve (AUC), nadir, and valley mean levels. Secretory characteristics reported include secretory burst mass and pulsatile and total nocturnal secretion. Total secretion refers to the sum of basal and pulsatile nocturnal ghrelin secretion. Data for GH and cortisol have been previously reported (8, 9).
Biochemical assessment
Ghrelin was measured using a RIA (Phoenix Pharmaceuticals, Belmont, CA; sensitivity, 2 pg/ml; coefficient of variation, 10%). We used an immunoradiometric assay to measure GH (Nichols Institute Diagnostics, San Juan Capristano, CA; detection limit, 0.05 ng/ml; coefficient of variation, 2.4–9.4%) and IGF-I (Nichols Institute Diagnostics; detection limit, 30 µg/liter; coefficient of variation, 3.1–4.6%). RIA was used to measure serum cortisol (Diagnostic Products Corp., Los Angeles, CA; limit of sensitivity, 1.0 µg/dl; coefficient of variation, 2.5–4.1%), leptin (Linco Diagnostics, St. Louis, MO; sensitivity, 0.5 ng/ml; coefficient of variation, 3.4–8.3%), estradiol (Diagnostic Systems Laboratories, Inc., Webster, TX; limit of detection, 2.2 pg/ml; coefficient of variation, 6.5–8.9%), and PICP (Diasorin Inc., Stillwater, MN; limit of detection, 25 ng/ml; coefficient of variation, 1.3–3.8%). We measured urine NTX in a 2-h, second morning urine sample normalized for cr excretion. NTX was measured by ELISA (Ostex International Inc., Seattle, WA; detection limit, 20 nmol bone collagen equivalent (BCE); intraassay coefficient of variation, 5–19%). Urine cr was measured by the hospital laboratory using published methods. Samples were stored at –80 C until analysis. All samples were run in duplicate.
Levels of GH, IGF-I, and leptin can be converted to SI units (micrograms per liter) by multiplying by 1, and serum cortisol can be converted to SI units (nanomoles per liter) by dividing by 0.0363.
Statistical analysis
Data are reported as means ± SD. The Student’s t test was used for comparison of means. Correlational analyses were first performed to determine predictors of bone density and bone turnover markers, followed by mixed model regression analysis (P = 0.15 for entry into the model and P = 0.10 for leaving the model) to determine independent predictors of these measures.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Clinical characteristics of our subjects have been reported earlier (8, 23) and are summarized here. Girls with AN had markedly lower BMI (16.7 ± 1.2 vs. 21.7 ± 3.7 kg/m2, P < 0.0001) and fat mass (8.8 ± 2.7 vs. 17.4 ± 5.7 kg, P < 0.0001) than healthy adolescents of comparable maturity. Lean body mass did not differ between the groups (37.0 ± 3.9 vs. 38.8 ± 6.6 kg, P not significant). Bone density and bone turnover markers and ghrelin parameters are summarized in Table 1
. GH AUC was higher in AN than controls (3879 ± 2007 vs. 2596 ± 1112 ng/ml, P = 0.01), as was cortisol AUC (6112 ± 1467 vs. 4117 ± 8802 µg/dl, P < 0.0001). Total GH and cortisol secretions were similarly higher in girls with AN (238 ± 135 vs. 169 ± 71 ng/ml, P = 0.04; 89.6 ± 18.8 vs. 71.2 ± 17.6 µg/dl, P = 0.002 respectively). Leptin AUC was markedly lower in girls with AN (3,417 ± 2,574 vs. 11,439 ± 4752 ng/ml, P < 0.0001), as was total leptin secretion (130 ± 97 vs. 462 ± 169 ng/ml, P < 0.0001).
|
In healthy adolescents, we observed strong positive correlations among ghrelin parameters (burst mass and pulsatile and total secretion) and between lumbar spine BMAD and BMAD z-scores and hip BMD and BMD z-scores. Ghrelin burst mass correlated positively with lumbar spine BMAD (r = 0.66, P = 0.003) and BMAD z-scores (r = 0.59, P = 0.01) and hip BMD (r = 0.55, P = 0.02) and BMD z-scores (r = 0.52, P = 0.02). Similar correlations were observed between ghrelin pulsatile secretion and bone density measures. The relationship between total ghrelin secretion and lumbar spine BMAD and hip BMD is shown in Fig. 1. Even when a possible outlier was eliminated from the analysis, total ghrelin secretion correlated with lumbar spine BMAD (r = 0.54, P = 0.03) and hip BMD (r = 0.52, P = 0.04). Ghrelin AUC correlated positively with lumbar spine BMAD (r = 0.53, P = 0.02) and lumbar spine BMAD z-scores (r = 0.59, P = 0.01), and weakly with hip BMD (r = 0.45, P = 0.06) and hip BMD z-scores (r = 0.42, P = 0.08). However, ghrelin parameters did not correlate with bone turnover markers in this healthy group.
|
When girls with AN and healthy adolescents were divided into two groups based on median ghrelin levels, girls with ghrelin pulsatile secretion greater than the median value had lower levels of NTX/cr (105.0 ± 71.2 vs. 176.4 ± 143.0 nmol BCE/mmol cr, P = 0.05), as did girls with higher total ghrelin secretion (NTX/cr, 102.3 ± 70.9 vs. 183.7 ± 143.0 nmol BCE/mmol cr, P = 0.03.
Girls with greater ghrelin burst mass (by median values) had higher total GH secretion (240 ± 122 vs. 169 ± 101 ng/ml, P = 0.05) and greater cortisol burst frequency (7.0 ± 1.1 vs. 6.0 ± 1.4/12 h, P = 0.02) and cortisol AUC (5831 ± 1635 vs. 4691 ± 1235 µg/dl, P = 0.02). Greater pulsatile and ghrelin secretion (by median) was also associated with greater cortisol burst frequency (7.1 ± 1.1 vs. 5.9 ± 1.4 /12, P = 0.005) and cortisol AUC (5816 ± 1590 vs. 4705 ± 1306 µg/dl, P = 0.02), as was ghrelin total secretion (7.1 ± 1.2 vs. 5.7 ± 1.2/12 h, P = 0.0005 for cortisol burst frequency; 5920 ± 1620 vs. 4533 ± 1082 µg/dl, P = 0.003 for cortisol AUC).
Stepwise regression analysis to determine contribution of ghrelin parameters to bone density measures
Using mixed model regression analysis, we examined the contribution of BMI, fat mass, lean mass, GH and cortisol AUC, IGF-I, estradiol, and ghrelin and leptin total secretion to bone density and bone turnover in controls and in AN (significant predictors of bone density and bone turnover were used in this model). We report these results in Tables 2 and 3 for healthy and AN adolescents, respectively.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Girls with AN have lower BMD and BMD z-scores compared with healthy adolescents (1, 2, 8). Ghrelin levels are higher in AN than controls (22, 31), and a role for ghrelin in low BMD in AN has not been previously examined. Ghrelin is the endogenous ligand of the GH secretagogue (GHS) receptor (12, 13, 14, 15, 16, 17, 18, 19) and is also known to stimulate appetite (10, 11). Ghrelin also increases secretion of ACTH (16, 17, 20, 21), and we have demonstrated that ghrelin concentration predicts both GH and cortisol burst frequency (23). In addition to effects of ghrelin on regulation of pituitary secretion, ghrelin may have effects on tissues such as fat, cartilage, and bone. A role for ghrelin in adipogenesis has been reported (15), and a recent study demonstrated expression of ghrelin mRNA in cartilage (24). Ghrelin administration also increased osteoblast proliferation in an in vitro study (25).
We observed that ghrelin was a strong predictor of bone density measures both at the lumbar spine and at the hip in healthy adolescents. To determine whether ghrelin predicted bone density independent of known body composition predictors such as BMI, fat mass, and lean body mass, and known hormonal predictors such as GH, cortisol, and estradiol, we used stepwise regression analysis with ghrelin and these known predictors of bone density entered into the model (when the relationship was significant on linear regression). Ghrelin emerged as a significant and independent predictor of all BMD measures in healthy adolescents. Other significant predictors were body composition measures and leptin and cortisol levels. However, ghrelin did not predict bone turnover markers in healthy adolescents. In contrast, GH concentration predicted levels of both PICP and NTX/cr in this group, and cortisol concentration predicted levels of NTX/cr. As demonstrated earlier by our group (8), GH correlated positively with bone turnover markers. Given that ghrelin is an independent predictor of both GH and cortisol burst frequency, one may speculate that the effects of ghrelin on bone may be mediated via the combined effects of GH and cortisol concentrations. Conversely, ghrelin may have direct effects on bone through mechanisms not yet understood. One recent in vitro study reported that ghrelin stimulates osteoblast proliferation in cell cultures and increases levels of bone formation markers (25). In addition to the GHS type 1a receptor, expression of a second unknown receptor was implicated in osteoblasts in this study. Another study has also demonstrated that ghrelin effects on chondrocytes are mediated via a hitherto uncharacterized receptor different from the GHS-1a receptor (24).
In contrast to healthy adolescents, we observed weak associations between ghrelin and lumbar spine BMAD z-scores in girls with AN on univariate regression analysis, and ghrelin contributed to a much smaller fraction of the variability in bone density measures on stepwise regression. Body composition and estradiol levels were significant predictors of bone density on multiple regression analysis. Ghrelin concentration correlated inversely with bone turnover markers in AN, but was not an independent predictor of these markers on stepwise regression analysis. As reported earlier, GH did not predict bone turnover markers in AN (8), whereas cortisol was a significant predictor of bone turnover markers, with high cortisol levels being associated with lower levels of PICP and NTX/cr (9). High ghrelin levels in AN do not appear to have a direct effect on bone in AN; however, by altering cortisol secretion (23) may contribute to low bone density indirectly via effects of high cortisol levels on bone turnover. Of interest, a recent study reported that ghrelin effects on osteoblasts are dose dependent, and osteoblast proliferation observed with lower doses of ghrelin administered to osteoblast cultures was not observed with higher ghrelin doses (25). Stimulatory effects of ghrelin on GH secretion are mediated via the GHS-1a receptor, whereas effects of ghrelin on bone may be mediated via a different receptor. One may speculate that high levels of ghrelin have differing effects on the different receptors mediating ghrelin effects, and very high ghrelin levels may increase GH and ACTH levels but not stimulate osteoblast proliferation. Conversely, ghrelin by increasing ACTH and cortisol secretion may indirectly decrease osteoblast proliferation.
Our data suggest that ghrelin is an independent predictor of bone density measures in healthy adolescent girls. Given that ghrelin is an independent predictor of both GH and cortisol burst frequency (23), one may speculate that the effects of ghrelin on bone may be mediated via the combined effects of GH and cortisol, which predict bone turnover, or may be an independent effect mediated by a receptor other than the GHS-1a receptor. In girls with AN, however, high ghrelin levels are weak predictors of bone density, and an inability to respond to ghrelin may contribute to low bone mass. The correlational nature of these analyses cannot prove causality, and studies examining effects of ghrelin administration on markers of bone turnover, in particular in osteopenic women, are necessary to better understand the contribution of ghrelin to bone metabolism.
|
Footnotes |
|---|
This work was supported, in part, by National Institutes of Health Grants M01-RR-01066, DK 52625-05, DK 062249, and K23-RR-018851.
First Published Online July 5, 2005
Abbreviations: AN, Anorexia nervosa; AUC, area under the curve; BCE, bone collagen equivalent; BMAD, bone mineral apparent density; BMD, bone mineral density; BMI, body mass index; cr, creatinine; DXA, dual-energy x-ray absorptiometry; GHS, GH secretagogue; NTX, N-telopeptde; PICP, carboxy-terminal peptide of type I procollagen.
Received March 8, 2005.
Accepted June 28, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Misra, R. Prabhakaran, K. K. Miller, M. A. Goldstein, D. Mickley, L. Clauss, P. Lockhart, J. Cord, D. B. Herzog, D. K. Katzman, et al. Weight Gain and Restoration of Menses as Predictors of Bone Mineral Density Change in Adolescent Girls with Anorexia Nervosa-1 J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1231 - 1237. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Misra, R. Prabhakaran, K. K. Miller, M. A. Goldstein, D. Mickley, L. Clauss, P. Lockhart, J. Cord, D. B. Herzog, D. K. Katzman, et al. Prognostic Indicators of Changes in Bone Density Measures in Adolescent Girls with Anorexia Nervosa-II J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1292 - 1297. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Misra, K. K. Miller, J. Cord, R. Prabhakaran, D. B. Herzog, M. Goldstein, D. K. Katzman, and A. Klibanski Relationships between Serum Adipokines, Insulin Levels, and Bone Density in Girls with Anorexia Nervosa J. Clin. Endocrinol. Metab., June 1, 2007; 92(6): 2046 - 2052. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. A. Tritos and E. G. Kokkotou The Physiology and Potential Clinical Applications of Ghrelin, a Novel Peptide Hormone Mayo Clin. Proc., May 1, 2006; 81(5): 653 - 660. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Misra, K. K. Miller, P. Tsai, K. Gallagher, A. Lin, N. Lee, D. B. Herzog, and A. Klibanski Elevated Peptide YY Levels in Adolescent Girls with Anorexia Nervosa J. Clin. Endocrinol. Metab., March 1, 2006; 91(3): 1027 - 1033. [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 |
