| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Pediatric Endocrinology (C.L.S., P.B.), Pôle d’Endocrinologie Enfants-Adultes Cochin-St. Vincent de Paul, APHP, Hôpital Saint Vincent de Paul, Paris V University, and Institut National de la Santé et de la Recherche Médicale U561 (C.L.S., P.B.), Hôpital Saint Vincent de Paul, 75014 Paris, France; Service de Biostatistique et d’Information Médicale (A.D.), Hôpital Necker, 75015 Paris, France; Department of Pediatrics (E.M.D.G.), Second University of Naples, 80138 Naples, Italy; and Centre National de la Recherche Scientifique UMR8090 (P.F.), Pasteur Institute, 59021 Lille, France
Address all correspondence and requests for reprints to: Pierre Bougnères, Pediatric Endocrinology, Hôpital Saint Vincent de Paul, 82 Avenue Denfert Rochereau, 75014 Paris, France. E-mail: pierre.bougneres{at}wanadoo.fr.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Methods: We conducted an association study in 580 severely obese European children (body mass index > 99.6th centile) and 606 nonobese control children, in whom glucose and insulin were measured in the fasting state. The homeostasis model assessment insulin resistance index was used to estimate IR.
Results: We found that a single-nucleotide polymorphism (rs361072) located in the promoter of the p110β gene was associated with fasting glucose (P = 0.0002), insulin (P = 2.6 10–8), and homeostasis model assessment insulin resistance index (P =1 10–9) in the severely obese children. The effect of rs361072 was marginal or not significant in nonobese children.
Conclusions: The C allele of rs361072 attenuates IR in superobese children.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
IR is influenced by not only the degree of obesity but also ethnic and genetic factors. To search for these genetic factors, we focused on the initial steps of insulin signaling. Class IA phosphatidyl-inositol-3 kinase (PI3K) is a crucial effector of insulin action (5), and signaling through the PI3K pathway depends on a critical balance between catalytic p110 and regulatory p85 subunits (6). Individual variations in the degree of IR could thus be associated with variations in the content or function of p110
or p110β, the catalytic subunits expressed in insulin target tissues (7). We studied the association of a common p110β single-nucleotide polymorphism (SNP) (8) with IR, reflected by the homeostasis model assessment insulin resistance index (HOMA-IR) index, in a cohort of white superobese children of European ancestry. Our objective was to test genetic factors that can be predictors of IR in adolescents with severe obesity. To avoid the effects of treatments and hypocaloric diets on insulin-glucose homeostasis, we selected patients who have never received drugs, and whose obesity course was unabated since its onset in early ages.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
We enrolled only white Caucasian patients with all grandparents born in Europe and/or Mediterranean countries and having no known African or Asian ancestor. To match the natural history of IR in severe juvenile obesity, our study was performed during the dynamic phase of obesity, before patients were submitted to weight reduction attempts. Inclusion criteria were severe obesity defined as a body mass index (BMI) greater than the 99.6th centile, European ancestry, a BMI greater than the 85th centile before age 6 yr. Studies were approved by institutional ethics committees, and written informed consent was obtained from the parents and oral consent from the children. No subject had diabetes, all were considered healthy at the time of study, and none was taking any medication. Two hundred ninety-nine adolescents with severe obesity came from the SUPEROBGEN cohort, a hospital-based cohort recruited between 1996 and 2003 (9). The LILLE obese children cohort included 104 severely obese children from northern France (10). The NAPOLI obese children cohort included 177 superobese children recruited between 1999 and 2006 (11). We were careful that none of the studied obese or nonobese subjects were from the same siblingship, the same family, or even related families. Children belonging to isolated communities having a high level of consanguinity like French gypsies or Basques were excluded from the study as well as seven severely obese children with mutations of the MC4R gene (12).
The LEANGEN cohort of 606 nonobese European children was recruited from 1986 to 2007 among the investigators’, colleagues’, nurses’, and friends’ families as well as among patients hospitalized for reasons not interfering with insulin and glucose homeostasis (benign surgery, evaluation of familial mild short stature, or isolated cryptorchidism) (13), totaling 666 nonobese healthy children in whom fasting glucose and insulin measurements were available as well as DNA samples. Body weight had to be within 90–110% of ideal body weight for age. Among these 666, 51 children had missing or obviously erroneous phenotypic data or were too young and were excluded from analysis and nine samples could not be genotyped; thus, 606 children were genotyped and included in the study. In all of them, sexual maturation was consistent with chronological age, and physical examination was normal except for the aforementioned abnormalities. Subjects with a familial history of diabetes or abnormally high birth weights were excluded. Sexual maturation at puberty was staged according to Tanner (14).
Experimental procedures for clinical studies
All children had been gaining weight the 6 months preceding the study to ensure that sampled insulin values truly reflected the natural history of IR (9). These data were not available; thus, continuous weight gain was assumed by interviews in the NAPOLI and LILLE cohorts. During the 3 d preceding insulin measurements, children were asked to keep their usual feeding habits. Parents or nurses checked the diet, and children with insufficient intake were excluded. Children fasted 12 h overnight, from the end of dinner (2000 h) to time 0 of the test (0800 h).
Insulin and glucose measurements
Plasma glucose was determined with a glucose analyzer (Beckman glucose analyzer, Beckman Coulter, Fullerton, CA) and serum insulin by time-resolved fluoroimmunoassay using Wallac Delfia reagents (PerkinElmer SAS, Courtaboeuf, France). For insulin, the intra- and interassay coefficients of variation at the level of 10 µU/ml were 4.1 and 5.0%, respectively, at the level of 27.5 µU/ml 3.8 and 5.1%, and at the level of 65 µU/ml 3.7 and 4.9%. The detection limit was 0.5 µU/ml. The cross-reactivity of C-peptide was 0.01%, that of proinsulin 0.1%. Duplicate insulin measurements on following days in 100 children showed a 3–7% intraindividual coefficient of variation. Insulin measurements were performed in the same laboratory for the SUPEROBGEN and LEANGEN cohorts (9) and standardized by exchanging samples blindly with the LILLE and NAPOLI centers.
HOMA-IR index
IR was quantified in the studied children using the HOMA-IR index (15) calculated as the product of fasting plasma insulin (in microunits per milliliter) and fasting plasma glucose (in millimoles per liter), divided by 22.5. Higher HOMA values indicate higher IR. This index does not have the quality of the gold standard euglycemic clamp procedure to evaluate insulin sensitivity. It is, however, the only one that we could apply to the hundreds of studied children, including the nonobese controls, for both ethical and practicality reasons. As many authors (1, 2, 3, 16, 17, 18), we used the HOMA index to estimate IR because it is the only validated index than can be used in large cohorts of children.
Genotyping
Genomic DNA samples were prepared from peripheral blood using DNA extraction kit PureGene (Gentra, Minneapolis, MN) or by using a standard phenol/chloroform method. In the 374-kb region neighboring rs361072, we selected 11 SNPs with minor allele frequencies greater than 0.10 and tested them for association with IR. Two other SNPS, rs361080 in intron 10 of p110β and rs600590 in intron 5 of FAIM, were in near complete linkage disequilibrium with rs361072 and thus associated with HOMA-IR (P = 1.3 10–6 and 5 10–5). We were not able to assign any potential function to these SNPS, which together with rs361072 formed a haplotype block associated with HOMA-IR to the same extent as rs361072. Following these observations, our study focused on rs361072. The PCR for rs361072 was carried out in a volume of 50 µl containing 200 ng genomic DNA, 1 µM of each primer (forward, 5'-CCTGTCAAGTGCTGGTTAACTA-3' and reverse, 5'-CAATCCATACCACCAACTAAAG-3'), 0.2 mmol/liter of each deoxynucleotide triphosphate, 1.5 mM MgCl2, and 1.25 U Taq polymerase Fast Start (Roche, Stockholm, Sweden). Allele positivity (presence of a T at position –359) and allele negativity (a C in the same position) were identified through the AATATT sequence recognized by Ssp1. Eight internal controls were systematically introduced into each 96-well plate. These internal controls were made of samples that were previously genotyped. We genotyped a total of 1303 DNA samples including 96 internal controls and 21 samples that had to be rerun for uncertain results. The average error rate estimated by internal duplicates was less than 0.6% (one error among 168 duplicates). Allele frequency was in Hardy-Weinberg equilibrium in the superobese and nonobese cohorts.
Statistical analyses
All values are expressed as means ± SE. Variables that were not normally distributed (BMI, fasting glucose, and insulin, HOMA-IR index) were log transformed for ANOVA between the genotypic groups. However, for clarity of interpretation, results are expressed in the table as untransformed values. All P values were two sided, and P < 0.05 was considered significant. Data analyses involved use of R statistical software.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
In the superobese children, we found that the –359 C/T SNP (rs361072) was strongly associated (1.10–6) with HOMA-IR (Table 1
). HOMA-IR was 40% lower in the C/C, compared with T/T superobese children. The intermediate HOMA values in C/T obese children supported an additive effect of the C allele. Significant differences were observed in fasting glucose and insulin, which were both lower in C/C patients (Table 1).
|
). In the studied cohorts, age, puberty, and gender showed no significant effect on HOMA-IR variation in superobese (P > 0.50) or nonobese children (P = 0.16 for age, P = 0.14 for puberty).
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Intramyocellular and intraabdominal lipid accumulation are associated with the development of IR in juvenile obesity (4). Our hypothesis is that p110β being a major effector of insulin action on adipocytes (5, 6, 7, 19), the p110β promoter variant could possibly affect adipose tissue development and partition between visceral and sc fat depots (4, 18) and hence modify IR in the liver and muscle as reflected by the HOMA-IR (21).
We found that the effect of rs361072 was present and of significant magnitude in severely obese adolescents. Although studied C/C children were slightly older and with a more advanced pubertal stage, they showed a clear tendency to have lower fasting glucose and insulin concentration and lower HOMA-IR. Obese adolescents have a narrow range of fasting glycemia, and thus, the magnitude of the effect of rs361072, although statistically detectable, was very small. Effects were minimal or absent in nonobese children, as previously reported in nonobese adults (8), which we interpreted as a strong dependence of the rs361072 association with IR on fat accumulation and mechanisms related to adipose tissue development and metabolism (4).
The found effects of rs361072 support that genetic markers could help identify superobese adolescents who are at greatest risk of developing IR and for whom more aggressive weight control is warranted (22, 23). Before being considered valid, however, the present association needs replication in further cohorts and different populations (24). It will also be interesting to test the effects of rs361072 on IR in superobese adults and evaluate its influence on IR complications in this rapidly expanding part of the population.
|
Footnotes |
|---|
Disclosure Statement: The authors have nothing to declare.
First Published Online October 30, 2007
Abbreviations: BMI, Body mass index; HOMA-IR, homeostasis model assessment insulin resistance index; IR, insulin resistance; PI3K, phosphatidyl-inositol-3 kinase; SNP, single-nucleotide polymorphism.
Received August 14, 2007.
Accepted October 22, 2007.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
) and p110(β) have differential roles in PDGF- and insulin-mediated signaling. J Cell Sci 113(Pt 2):207–214
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
