| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Endocrinology, Bone Diseases, Genetics, and Gynaecology (E.F., G.D., F.C.-A., C.M., J.-P.S., M.T.), Centre de Référence du Syndrome de Prader-Willi, Childrens Hospital, Centre Hospitalier Universitaire, and Laboratoire de Biochimie 3 (I.G.), Institut Fédératif de Biologie, Hôpital Purpan, 31059 Toulouse, France; and Institut National de la Santé et de la Recherche Médicale, U563 (F.C.-A., J.-P.S., M.T.) and U558 (C.A., M.T.), University Paul Sabatier, 31403 Toulouse, France
Address all correspondence and requests for reprints to: Professor M. Tauber, Unit of Endocrinology, Hôpital des Enfants, 330 Avenue de Grande Bretagne, TSA 70034, 31059 Toulouse Cedex 9, France. E-mail: tauber.mt{at}chu-toulouse.fr, molinas.c{at}chu-toulouse.fr.
| Abstract |
|---|
|
|
|---|
Objective: The objective of the study was to investigate total plasma ghrelin levels in children with PWS and controls from 2 months to 17 years.
Subjects and Methods: Forty children with PWS [24 boys, 16 girls, median age 3.6 yr, median body mass index (BMI) Z-score 0.3] were compared with 84 controls (57 boys, 27 girls, median age 4.2 yr median BMI Z-score 0.1). Children were then divided into two groups according to age and GH treatment.
Results: Median plasma ghrelin levels were significantly higher in children with PWS, compared with controls at any age (568 vs. 173, P < 0.0001) and decreased with age in both groups (P < 0.0001). In the whole group of PWS, we found an inverse relationship between ghrelin and BMI Z-score, insulin, homeostasis model assessment insulin resistance index, leptin, and lean mass. Plasma ghrelin levels were higher in children with PWS than controls, both in the youngest children below 3 yr who were not receiving GH (771 vs. 233, P < 0.0001) and in the children older than 3 yr, all of whom were treated with GH (428 vs. 159, P < 0.0001).
Conclusions: Plasma ghrelin levels in children with PWS are elevated at any age, including during the first years of life, thus preceding the development of obesity.
| Introduction |
|---|
|
|
|---|
Early diagnosis and multidisciplinary care involving GH treatment have been shown to improve the developmental outcome of these children and particularly to reduce the incidence of obesity (6).
Ghrelin is a 28-amino acid acylated peptide isolated in 1999 from stomach with a strong orexigenic effect and the most potent GH secretagogue through its hypothalamic actions (7). Ghrelin is now considered as a pleiotropic hormone with various secreting organs (duodenum, pancreas, pituitary, hypothalamus, testis, ovary, bone, and cartilage) and a wide range of target tissues varying with its acylation status (hypothalamus, stomach and gut, pancreas, adipose and cardiovascular tissues, testis, ovaries, and muscle) and acting through distinct receptors (8, 9, 10). Regarding its role at the level of arcuate nucleus, ghrelin demonstrated opposite effects to leptin, another hormone involved in the control of food intake and energy homeostasis, secreted by adipocytes, and acting on the same hypothalamic neurons (11). Leptin also plays a role in neuronal plasticity and particularly in the implementation of neuronal pathways in early life (12).
Circulating ghrelin levels have been shown to be low in obesity, suggesting that conversely to leptin, there is no ghrelin resistance in obesity (13, 14, 15). PWS individuals, unlike those with other known causes of obesity, had hyperghrelinemia (13, 14, 16). This hyperghrelinemia could explain at least two major endocrine dysfunctions observed in these patients such as obesity with reduced satiety and GH deficiency (described in 40–100% of the patients and explaining the short stature) (17). GH deficiency could be explained by ghrelin resistance, whereas obesity could result from a preserved action of ghrelin in controlling appetite possibly due to different receptors or different modulators. Ghrelin levels decreased with age in controls (16, 18, 19) and PWS (13, 14). Some but not all studies documented an inverse correlation between ghrelin levels and BMI in PWS individuals (14, 16). Nevertheless, ghrelin levels were reported negatively correlated with visceral adiposity, fasting insulin, and homeostasis model assessment insulin resistance index (HOMA-IR) (20, 21) in adults with PWS.
There were only two reports in the literature (22, 23) with plasma ghrelin measurements during the first years of life in PWS. We measured plasma ghrelin in 40 children with PWS aged from 2 months to 17 yr and in 84 controls to describe ghrelin changes over time and particularly before 3 yr. We also examined which factors influence ghrelin levels in PWS children and the effect of GH on ghrelin levels in nine children with repeated ghrelin sampling before and after GH treatment.
| Subjects and Methods |
|---|
|
|
|---|
Forty children with genetically confirmed diagnosis of PWS [24 boys, 16 girls; aged 0.2–17.2 yr, median 3.6 yr; body mass index (BMI) Z-score range –4.0 to 4.4, median 0.3] and 84 control children (57 boys, 27 girls; aged 0.3–17.1 yr, median 4.2 yr; BMI Z-score range –1.51 to 1.98, median 0.08) were enrolled in the present study. The methylation abnormalities at 15q11–13 were confirmed in all children with PWS. Thirty-two (80%) children had a paternal deletion, six (15%) had a uniparental maternal disomy, and one (2.5%) had an imprinting center mutation, and in one child (2.5%), the underlying genetic abnormality remained unknown.
The control group was represented by healthy nonobese children [BMI below the 97th percentile of French BMI curves (24)] hospitalized in the department of surgery for a scheduled minor intervention or an examination before anesthesia. The female to male ratio was comparable in the study groups, as was age and BMI distribution. Informed written consent was obtained from the parents of all subjects.
All the children with PWS were seen regularly in our center, and we divided them into two groups: one group involving 18 children younger than 3 yr without GH treatment (n = 18) and another group involving children aged older than 3 yr (n = 22) treated with GH (33 µg/kg·d sc). The group of children with PWS 0–3 yr were not on GH treatment because, until recently, GH treatment routinely started after the age of 3 yr. In the group of children older than 3 yr treated with GH, the median chronological age at start of GH treatment was 4.75 yr. The median duration of GH treatment was 1.9 yr, ranging from 0.1 to 7.4 yr. All the children with PWS (n = 40) underwent a GH stimulation test with a median peak of 8.4 µg/liter.
Nine children, all included in the group of children younger than 3 yr before receiving GH treatment, underwent two plasma ghrelin evaluations, before and after GH treatment. Median chronological age at start of GH treatment was 0.86 yr, ranging from 0.55 to 2.9 yr, and median duration of GH treatment was 1.0 yr, ranging from 0.6 to 2.8 yr.
Auxological and anthropological measures
Body weight and height were measured in all children at admission to the hospital. Height was measured with a Harpenden stadiometer and weight was measured using a Seca scale (Seca, Hamburg, Germany). BMIs were then calculated and expressed in Z-score according to the reference data for the French population (24). In children with PWS, gestational age at birth, birth weight, and birth length were also recorded, and BMI at birth was calculated. The values were transformed into SD score (SDS) using the standard tables by Usher and McLean (25); small for gestational age (SGA) was defined as birth weight and/or birth length less than –2 SDS.
Body composition
Dual-energy x-ray absorptiometry with a whole-body scanner (Lunar DPX-L Lunar Corp., Madison, WI) is included in the routine follow-up of children with PWS and was used to determine their body composition. Total body fat and lean body mass were analyzed using the Lunar 1.31 software. The values of total body fat and of lean body mass were expressed as percentage of body weight (26).
Biological measures
Total plasma ghrelin levels were measured using a commercial RIA (Phoenix Pharmaceutical, Belmont, CA.). The intra- and interassay coefficients of variation were 7.16 and 9.86%, respectively. All the measurements for ghrelin were performed using the same kit.
Fasting plasma ghrelin and insulin was measured in children (PWS and controls) with duration of fasting comparable. PWS children underwent fasting IGF-I, IGF binding protein (IGFBP)-3, glucose, and leptin measurements. All the biological parameters were analyzed in the same sample. IGF-I and IGFBP-3 were measured before GH treatment, if any, and before and after GH treatment in nine children.
Serum insulin was assessed by immunoelectrochemoluminescence (Bayer Corp., Tarrytown, NY) and serum leptin by RIA (Linco Research, Inc., St. Charles, MO). IGF-I and IGFBP-3 were measured by an immunoradiometric assay (Immunotech, Beckmann, Fullerton, CA). The interassay and intrassay coefficients of variation for intermediate values were, respectively, 6.8 and 6.3% for IGF-I and 9.5 and 6% for IGFBP-3. Plasma glucose was measured with enzymatic reagents on an automated analyzer (Olympus, Bayer Diagnostics, Puteaux, France). The estimate of insulin resistance by homeostasis model assessment score was calculated in all subjects with the following formula: fasting serum insulin (milliinternational units per liter) x fasting plasma glucose (millimoles per liter)/22.5, as described by Matthews et al. (27).
Statistical analysis
The statistical analysis was performed using the StatView software (version 5.0, SAS, Cary, NC). Comparisons between and within groups were carried out using Mann-Whitney U test for unpaired data and Wilcoxon test for paired data. The effect of nominal variables on a continuous variable was determined by the Kruskall-Wallis test. The relationships between variables within the separate groups were evaluated using linear regression analyses. We analyzed in detail the evolution of plasma ghrelin in PWS children and particularly those in children younger than 3 yr. All data are presented as medians and ranges or in percent. P < 0.05 values were considered statistically significant.
| Results |
|---|
|
|
|---|
|
|
Plasma ghrelin levels were significantly higher in the total group of PWS children, compared with controls [568 (144–1363) vs. 173 (55–516) pg/ml, P < 0.0001].
Figure 1
shows the evolution of ghrelin levels from 0.3 to 17 yr in all patients. Plasma ghrelin levels showed wide interindividual variations, especially in children with PWS during the first year of life, with values ranging from 320 to 1363 pg/ml. We did not observe a significant difference in plasma ghrelin levels between children with paternal deletion and those with a uniparental maternal disomy.
|
In children younger than 3 yr, median plasma ghrelin levels were significantly higher in PWS, compared with controls [771 (241–1363) vs. 233 (110–516) pg/ml, P < 0.0001], whereas BMI and age were not significantly different. Although there was no significant difference in insulin between PWS and controls under 3 yr old (P = 0.27), values were only available for a limited number of the group.
In children older than 3 yr, median plasma ghrelin levels were also significantly higher in PWS, compared with controls [428 (144–823) vs. 159 (55–410) pg/ml, P < 0.0001]. In this group, BMI expressed in Z-score and fasting insulin were significantly higher in PWS patients (P = 0.02 and P < 0.0001, respectively).
Factors influencing ghrelin levels
Whole group We observed a significant inverse relationship between ghrelin levels and age in the whole PWS group (n = 40) (R2 = 0.359, P < 0.0001) and the whole control group (n = 84) (R2 = 0.203, P < 0.0001), whereas no statistical differences between gender were found in both groups. In children with PWS, we observed a significant inverse relationship between ghrelin levels and BMI expressed in Z-score (R2 = 0.207, P = 0.0032), insulin (P < 0.0001), HOMA-IR (P = 0.0002), and leptin (P = 0.0027). We did not find a relationship between ghrelin and fat mass, but there was a positive relationship between ghrelin and lean mass (R2 = 0.180, P = 0.04) in the whole group of children with PWS. There was no significant association between ghrelin and GH peak at stimulation test (R2 = 0.004, P = 0.75).
Children younger than 3 yr In both children with PWS (n = 18, all without GH treatment) and controls (n = 27), the only parameter influencing ghrelin levels was age (P = 0.05), and there was a trend for an inverse relationship with BMI expressed in Z-score (P = 0.098). In children with PWS, there was no relationship between ghrelin and any of the parameters measured: insulin, HOMA-IR, and leptin.
Children older than 3 yr We also observed an inverse relationship with age in both PWS children, all treated with GH (n = 22), and controls (n = 57) (P < 0.0001). An inverse relationship between ghrelin and fasting insulin (P = 0.001), HOMA-IR (P = 0.002), and leptin (P = 0.028) was observed only in PWS children. All these children with PWS received GH treatment and had significantly higher insulin levels, compared with controls (P < 0.0001) and with a group of GH-untreated PWS children (P < 0.05); this group included six children with PWS (three males, three females, median age 7.2 yr) regularly followed up in our department but not included in our present study. Data regarding these six GH-untreated children older than 3 yr were: median BMI Z-score 1.89, median ghrelin level 612 pg/ml, and median insulin level 3.5 mU/liter).
Effect of GH treatment on plasma ghrelin levels in nine children with PWS
There was no significant difference between ghrelin levels before and after GH treatment (715 vs. 687 pg/ml, P = 0.37) and BMI during GH treatment did not significantly changed (–0.4 vs. +0.6 Z-score, P = 0.11).
| Discussion |
|---|
|
|
|---|
The study by Soriano-Guillen et al. (18) on a control population of 29 preterm newborns and 124 full-term newborns reported high total plasma ghrelin levels at birth and during the first 2 yr of life, followed by a progressive decrease until the end of puberty. However, in that study, for unexplained reasons, plasma ghrelin levels were higher than in our control children, the method of measurement being the same.
The acylated ghrelin was not evaluated in our study, but the ratio of acylated and total ghrelin has been reported to remain constant in animals and humans (28, 29), as well as in children with PWS (21).
Our data enabled us to describe the evolution of plasma ghrelin levels from 2 months to 3 yr of life in children with PWS showing that ghrelin levels were already elevated. Nevertheless, in the young children with PWS, the small numbers of patients in each year requires further studies.
High plasma ghrelin levels are present very early in children with PWS and could play a role in early-onset obesity. We could compare our results with the observation in a transgenic mouse model for the region equivalent to the human region of 15q11-q13, in which high plasma ghrelin levels were observed from the third day after birth, compared with control animals (30).
From a pathological point of view, we could hypothesize that high ghrelin in early years induces developmental neuroanatomical and synaptic plasticity changes akin to that seen with leptin in animals (12).
We found a significant decrease in total plasma ghrelin levels with age in children with PWS and controls, thus confirming the findings of some previous studies (18, 31, 32).
Our study confirms the existence of inverse relationship between fasting plasma ghrelin and fasting insulin, which was shown by previous studies in PWS as in controls (20, 21, 33, 34). In those papers that examined adults with PWS, ghrelin was more strongly negatively correlated with insulin than BMI or percent fat. Relative fasting hypoinsulinemia and increased insulin sensitivity were reported in children and adults with PWS, mainly in obese patients (35, 36, 37, 38). However, in our study, young children with PWS untreated with GH had fasting insulin levels not significantly different, compared with controls. In young children with PWS, we did not find any relationship between ghrelin and other clinical and biological parameters. Nevertheless, there was small numbers of young children with insulin measurements and further studies are needed. Conversely, in the older children with PWS treated with GH, fasting insulin was significantly higher, compared with controls, due to GH treatment, and there was a negative relationship between ghrelin and insulin and HOMA-IR and leptin.
The other correlations that we found are consistent with some previously reported data (18, 19, 29). We did not study visceral fat in the young children, but visceral fat had been reported lower in obese adult women with PWS, compared with non-PWS adult obese women (36). For ethical reasons we did not perform a body composition dual-energy x-ray absorptiometry in control children.
The analysis of ghrelin levels before and after GH treatment in nine children with PWS does not show a significant effect of GH on total plasma ghrelin levels, thus confirming our previously published results (16). The published data regarding the effect of GH on plasma ghrelin levels are variable in patients with PWS. One study (39) reported a nonsignificant tendency toward a decrease in plasma ghrelin levels in children after GH treatment; others (40) demonstrated a significant decrease in total plasma ghrelin, whereas a study on adult patients (41) did not show any effect.
Our study shows that ghrelin dysregulation in PWS occurs very early and precedes the onset of obesity. Plasma ghrelin levels in children with PWS are constantly high from birth to adulthood, regardless of age, but physiological decrease with age was preserved.
To our knowledge, this is the first study to describe plasma ghrelin levels in children with PWS during the first 3 yr of life. Further studies on a larger cohort are needed to describe more precisely the evolution of plasma ghrelin levels from birth to the first years of life.
| Acknowledgments |
|---|
| Footnotes |
|---|
First Published Online May 6, 2008
Abbreviations: BMI, Body mass index; HOMA-IR, homeostasis model assessment insulin resistance index, IGFBP, IGF binding protein; PWS, Prader-Willi syndrome; SDS, SD score; SGA, small for gestational age.
Received September 21, 2007.
Accepted April 29, 2008.
| References |
|---|
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |