1

Effect of the rs997509 Polymorphism on the Association Between ENPP1, Metabolic

Syndrome and Impaired Glucose Tolerance in Childhood Obesity.

Nicola Santoro, Grazia Cirillo, Maria Grazia Lepore, Alfonsina Palma, Alessandra Amato, Piera

Savarese, Pierluigi Marzuillo, Anna Grandone, Laura Perrone and Emanuele Miraglia del Giudice.

Department of Pediatrics “F. Fede” Seconda Università degli Studi di Napoli, Via Luigi De

Crecchio 2, 80138, Napoli.

Running title: ENPP1 gene and metabolic syndrome in children

Corresponding Author: Dr. Emanuele Miraglia del Giudice, Dipartimento di Pediatria, Seconda

Università di Napoli, Via Luigi De Crecchio N°2, 80138, Napoli, Italy.

Phone: ++39-081-5665420

Fax: ++39-081-5665427

E-mail: emanuele.miraglia@unina2.it

Word Count: 238 (abstract) 3,080 (text); Number of tables and figures: 3 Tables, 2 Figures

Disclosure Statement: The authors have nothing to disclose

Key words: ENPP1, children, metabolic syndrome, IGT, polymorphism.

J Clin Endocrin Metab. First published ahead of print October 21, 2008 as doi:10.1210/jc.2008-1659

Copyright (C) 2008 by The Endocrine Society

2

Abstract

Context: Variants on the nucleotide pyrophosphatase/phosphodiesterase-1 (ENPP-1) gene have

been associated with obesity and insulin resistance. Since insulin resistance is a pivotal factor in the

development of metabolic syndrome (MS) and impaired glucose tolerance (IGT), we aimed to test

the association between the K121Q and rs997509 ENPP-1 variants with obesity, MS and IGT in

obese children and adolescents.

Methods: We screened 809 children, 409 obese and 400 lean controls. Obese subjects underwent a

standard oral glucose tolerance test, whole body insulin sensitivity index (WBISI) and homeostasis

model assessment (HOMA) were calculated.

Results: No difference in prevalence for K121Q and rs997509 polymorphisms between obese and

controls (p>0.05) were observed. Obese children carrying the rs997509 rare allele showed higher

insulin (p=0.001), HOMA-IR (p<.001) and lower WBISI values (0.04) compared to common allele

homozygous. A similar observation was done for K121Q variant, with 121Q allele carriers showing

higher insulin (p=0.03) and HOMA (p=0.04) values than 121K homozygotes.

Moreover, subjects carrying the rs997509 rare allele had higher risk of MS (OR 2.4, 95% CI: 1.3-

4.3) and IGT (OR 4.7, 95% CI: 1.9 -11.4) than common allele homozygotes. Evaluating combined

effects of both polymorphisms, which are in strong linkage disequilibrium, we showed that the

effect on insulin sensitivity was due to the rs997509 T variant.

Conclusion: We conclude that the ENPP1 rs997509T allele can predispose obese children to MS

and IGT and that this variant might drive the association between the ENPP1 121Q allele and

insulin resistance.

3

Introduction

Children are becoming increasingly vulnerable to overweight and obesity around the world. At

least 155 million school-age children worldwide are overweight or obese, according to the latest

estimates from the International Obesity Task Force (1). One of the most frequent complications of

obesity is the early occurrence of metabolic syndrome, defined as the co-occurrence of

dyslipidaemia, hypertension and impaired glucose homeostasis (2). Insulin resistance represents the

link between obesity and these metabolic alterations (3). A crucial step in the pathogenesis of

insulin resistance is represented by the free fatty acids (FFA) accumulation in the liver, fat cells and,

particularly, skeletal muscle of obese patients, that, interfering with the normal insulin signalling

cascade, appears as the primary determinant of insulin resistance (4). As consequence of insulin

resistance, pancreas needs to increase its insulin production to maintain normal values of glycaemia,

but the progressive fat accumulation in pancreatic beta cells leads to beta cells failure causing a

defective insulin secretion, inadequate to maintain glycaemia in the normal range. This condition

appears as clinically evident when impaired glucose tolerance (IGT) or type 2 diabetes (T2D)

occurs. In fact, along with metabolic syndrome, IGT and T2D are nowadays much more common

that in the past years among obese children and adolescents, with a range of incidence between

0.5% and 4% for T2D and 5% and 25% for IGT (5, 6). The plasma membrane enzyme termed

ENPP1 (ectonucleotide pyrophosphatase phosphodiesterase 1), has been shown to inhibit insulin

receptor function by affecting its tyrosine kinase activity in peripheral tissues, including liver,

muscle and fat (7). It is a class II transmembrane glycoprotein located both on plasma membrane

and in the endoplasmatic reticulum (8). Although its physiological function is not completely

understood, there is evidence that ENPP1 plays a major role in insulin resistance development (9).

Several studies have reported that the locus where ENPP1 gene maps (6q22-q23) is linked to both

insulin resistance and T2D, highlighting the potential role of this gene in modulating susceptibility

to these metabolic issues (9). The most widely investigated ENPP1 polymorphism is the

substitution of a lysine with a glutamine at codon 121 (K121Q), with the 121Q variant associated

4

with type 2 diabetes (9). This aminoacid change is located in the second somatomedin-B-like

domain of ENPP1 and may interfere with protein-protein interactions (10). Human studies

demonstrated that obese adults carrying the Q variant show lower insulin sensitivity than

noncarriers (11). Moreover, association studies have shown that ENPP1 121Q variant may

influence the susceptibility to develop obesity and IGT during childhood (12, 13). Although the

majority of the studies dealing with the ENPP1 gene have focused their attention on the K121Q

variant obtaining discordant results (9), a recent report by Bochenski et al., examining the

associations of tagging single nucleotide polymorphisms (SNPs) and haplotypes of the linkage

disequilibrium block containing K121Q with T2D, showed that another variant, the rs997509, lying

in the intron 1 and consisting in a substitution of a cytosine (C) with a timine (T), had a strong

correlation with T2D (14). The authors suggested that the T allele of the new identified SNP, is

sufficed to distinguish the 121Q carrying haplotype that was significantly more associated with

T2D.

Given these evidences we decided to screen the K121Q and rs997509 in a population of obese

children and adolescents to try to replicate previous results and to verify which relationship

intercourse between the two polymorphisms and the predisposition to obesity and insulin resistance

in our population.

Materials and Methods

Cohort description and clinical evaluation

Four hundred and nine Caucasian obese children and adolescents, referred to our ward (childhood

obesity service) since the 1999 have been enrolled. Subjects between 2 and 16 years of age and with

a BMI exceeding the 97th

percentile for their age and sex were eligible. Subjects using medications

that alter blood pressure, glucose or lipid metabolism were excluded. Moreover, all subjects missing

the complete data set needed for the analysis were excluded. The ethical committee of the Second

5

University of Study of Naples approved the study. Informed consent was obtained by parents and,

where appropriate, by children.

Of the 409 subjects enrolled, 208 were girls. This sample was representative of the 2482 children

referred to our ward from 1999 to 2005; in fact, no differences in mean age, sex distribution and

pubertal stage were observed between the study sample and the sample of excluded subjects.

Weight and height were measured and BMI was calculated. Standard deviations scores for BMI

were calculated by using the LMS method (15). The population mean age was 11.5 ± 3.0 years; the

mean z-score BMI was 3.6 ± 1.0. Systolic blood pressure (SBP) and diastolic blood pressure (DBP)

were measured three times while the subjects were seated and the two last measurements were

averaged for the analysis (16). Pubertal stage was assessed using Tanner criteria (17). Thirty one

percent of children included in the study were pubertal (71 girls). All measurements were taken

twice by the same operator.

To test the ENPP1 allelic distribution a group of controls, composed by 400 (200 girls) lean age and

sex matched children (27% pubertal, 63 girls), was recruited as previously described (18). Briefly,

unrelated lean children who were age and sex matched and belonged to the same geographic area

were recruited as controls (mean age, 10.5 ± 2.3 yr; mean BMI z-score, 0.4 ±

0.4). They consulted

the Department of Pediatrics of the Second University of Naples for presumed diseases and were

found to be normal. Informed consent from parents and assent from children

were obtained before

participation in the study. The dosage of fasting insulin, serum glucose, plasma lipids and blood

pressure measurement has been performed also in 200 controls.

Metabolic evaluation

After informed consent, a blood sample was drawn from each patient after an overnight fast. The

serum was frozen at -20°C until analysed. Triglycerides and total cholesterol levels were

determined by an enzymatic colorimetric test with lipid clearing factor and HDL-cholesterol (HDL)

6

was measured by precipitation. To underwent an oral glucose tolerance test (OGTT) by assuming

1.75 g of glucose per kilogram of body weight, subjects were evaluated at 8 a.m. after an overnight

fast; they consumed a diet containing at least 250 g of carbohydrates per day for three days before

the study and refrained from vigorous physical activity; insulin and glucose levels were measured

during the OGTT at baseline and later every 30 minutes for 120 minutes.

Given the changes occurring in body composition during growth, to diagnose the metabolic

syndrome the criteria of the National Cholesterol Education Program's Adult Treatment Panel (19)

and the World Health Organization (20) were modified according to Weiss et al (2). Consequently,

children and adolescents in our study were classified as having the metabolic syndrome if they met

three of the following criteria needed to be present: BMI exceeding the 95th

percentile, systolic

and/or diastolic blood pressure exceeding the 95th

percentile, triglycerides levels higher than 110

mg/dl, HDL cholesterol lower than 40 mg/dl for males and 50 mg/dl for females, impaired glucose

tolerance (glucose level greater than 140 mg/dl, but less than 200 mg/dl after two hours from the

beginning of the OGTT).

Insulin resistance was assessed using the homeostasis model assessment (HOMA) as follows:

fasting insulin (pmol/liter) x fasting glucose (mmol/liter)/135. The degree of insulin sensitivity was

assessed by using the whole body insulin sensitivity index (WBISI). The composite WBISI is based

on values of insulin and glucose obtained from the OGTT and the corresponding fasting values, as

originally described (21). It represents good estimates for clamp-derived insulin sensitivity and it

has been demonstrated to be correlated with intramyocellular lipid content (22). It has been

obtained according to the following formula: 10.000/square root of [(fasting insulin X fasting

glycaemia) X (mean insulin concentration during OGTT) X (mean glycaemia during OGTT) (22).

Genotyping

All the patients and controls were genotyped for the K121Q and rs997509 variants. For the

rs997509 C/T substitution the following primers were used, F: 5’-ATTTTTTCCTTCAGTGTATA-

7

3’ and R: 5’-ACACCCTACAACCCCTAAGA-3’. The AciI restriction enzyme was used to identify

the variant, since the T allele eliminates an AciI restriction site. To detect the K121Q variant the

couple of primers used were: F: 5’-GCAATTCTGTGTTCACTTTGG-3’ and R: 5’-

GAGCACCTGTTGACACA-3’. AvaII restriction enzyme was used to detect the variant.

Statistical analysis

Power calculations were performed using the genetic power calculator.

http://pngu.mgh.harvard.edu/~purcell/gpc/. Since previous studies showed an association between

childhood obesity and K121Q variant (12, 13), prior to analyse the association of Q variant with

obesity we calculated that the statistical power in the proband – control data set was 98% at p= 0.05

to detect an association using the control frequencies and Odds Ratios (OR) shown in the German

population (13) (allele frequency 0.11. OR=1.82 for T allele). To reach a statistical power of 98% at

p=0.05 we screened 400 age and sex matched lean controls.

Chi Square test was used to verify whether the genotypes were in Hardy-Weinberg equilibrium and

to compare allele frequencies between obese and non-obese subjects and metabolic syndrome and

IGT prevalence between the different genotypes. Linkage disequilibrium between markers was

assessed as described (23, 24). A general linear model (GLM) was used to evaluate the differences

between groups of genotype and the effect of genotypes interaction on phenotype. Because of the

low prevalence of rare allele homozygotes for the K121Q and rs997509 variants, to test

genotype/phenotype associations rare allele homozygotes and heterozygotes were merged. When

necessary the variables were adjusted for age, sex, BMI and pubertal stage. A logistic regression

was generated to calculate the odds of developing the metabolic syndrome or IGT for a given

genotype. Although raw values are shown, non-normally distributed variables were log-transformed

before performing the analysis. Data are expressed as means and standard deviations. P values <

0.05 were considered statistically significant. The SAS Statistical Software Package version 8.2

(SAS institute, Clary, NC) was used to analyze data.

8

Results

One hundred and forty three (35%) patients (66 girls) showed the criteria to make diagnosis of

metabolic syndrome, three patients (0.7%) showed impaired fasting glucose (IFG) and 23 subjects

(5.6%) had impaired glucose tolerance (IGT); all IFG patients had also IGT. None showed type 2

diabetes. Subjects with metabolic syndrome had higher insulin levels, HOMA, triglycerides, z-score

systolic blood pressure (SBP), z-score diastolic blood pressure (DBP) and lower WBISI and HDL

levels than subjects who did not meet the criteria for the diagnosis of metabolic syndrome (table 1).

Subjects with IGT showed higher serum insulin levels (40.6±32 vs 27.4±18; p<.001), higher

HOMA (9.40±6.2 vs 5.6±3.8; p<.001) and lower WBISI (1.4±0.7 vs 2.5±2.0; p=0.007) than those

with a normal glucose tolerance. None of the lean controls showed dyslipidemia, hypertension or

IGT and the serum insulin levels (8.8±3.3) and the HOMA values (0.9±0.6) were significantly

different from those observed in obese (both p<.001).

The genotype distribution of both the studied variants was in Hardy-Weinberg equilibrium both in

obese and control cohorts (both p>0.05). The K121Q and rs997509 showed a strong linkage

disequilibrium (D’ = 0.9). The allelic distribution of the K121Q polymorphism was similar to that

shown in the Italian population (25) with a prevalence of 121Q variant of about 15%. Allelic

prevalence of the K121Q and rs997509 did not differ between obese and controls. The Q allele was

present in 16% of obese and in the 13% of controls (p> 0.05), the T allele of the rs997509 was

present in 7% of obese and in 6% of controls (p>0.05). Also genotype frequencies were similar in

obese and controls both for K121Q (obese KK 0.72, KQ 0.24, QQ 0.04 - controls KK 0.75, KQ

0.23, QQ 0.02 ; p= 0.3) and rs997509 (obese CC 0.87, CT 0.12, TT 0.01 - controls CC 0.89, CT

0.1 , TT 0.01; p= 0.8 ). The T allele of the rs997509 was present in about the 7% of population,

according to the study by Bochensky et al., performed in a Polish population showing a prevalence

of about 6% (15).

When compared for the clinical features, subjects carrying the 121Q allele showed significantly

higher serum insulin levels and HOMA values than subjects homozygotes for the common allele

9

and subjects carrying the rs997509T allele showed higher serum insulin levels, higher HOMA and

lower WBISI than children homozygotes for the common allele (table 2).

A significant higher prevalence of the metabolic syndrome and IGT in the group of patients

carrying the rs997509 T allele was observed. In fact, the metabolic syndrome was present in 52% of

obese children with this allele (OR: 2.4, 95% CI: 1.3-4.3; p= 0.005) and IGT was present in 13% of

obese children carrying the rs997509 T allele (OR 4.7, 95% CI: 1.9 -11.4; p<.001). Although not

statistically significant, subjects carrying the 121Q allele showed the same trend, with a higher

prevalence of IGT and Metabolic syndrome than those carrying the wild allele (table 3). Moreover,

subjects carrying the rs997509 T allele had significantly higher serum insulin levels during the

curve, showing a trend for higher serum glucose levels (figure 1). Although a similar trend was

evident, no significant differences for serum insulin and glucose levels during the OGTT were

observed for Q121 (all p-values higher than 0.05, except baseline insulin values) (figure 2).

Differences in fasting serum insulin levels and HOMA values according to the rs997509 (serum

insulin = 9.2±4.8 vs. 12.3±5.7, p = 0.6; HOMA = 1.1±0.7 vs. 1.4±1.2 p=0.9) and K121Q (serum

insulin = 8.3±2.6 vs. 8.9±3.5, p = 0.3; HOMA = 0.8±0.7 vs. 1.0±0.6 p=0.5) genotypes were not

observed in 200 subjects of the control group.

When the combined effects of the genotypes was tested, we used a model where, at every turn,

HOMA, insulin or WBISI was the dependent variable and sex, age, pubertal stage, K121Q and

rs997509 were the covariates. The results showed that the major effect on these three variables

(insulin, HOMA and WBISI) was ascribed to the rs997509 variant that appeared to drive also the

association between K121Q variant and insulin resistance (table 3).

Discussion

In vitro and in vivo studies clearly indicate that membrane glycoprotein ENPP1 contributes to the

decreased function of the insulin receptor observed in the insulin resistance. ENPP1 is, in fact,

10

elevated in muscle, fat and other tissues of subjects with insulin resistance (26, 27) and transgenic

mice over-expressing ENPP1 in different tissues are insulin resistant and show type 2 diabetes (9).

Conflicting results exist concerning the role played in humans by ENPP1 variants and particularly

the K121Q polymorphism in predisposing to the quantitative traits related to insulin resistance:

obesity and type 2 diabetes (9). We did not show any association between obesity and K121Q. This

observation is in contrast with several studies ruled out in Europe and in USA (12, 13, 28, 29), but it

is consistent with three recent large multicentric studies in European and African-American

individuals (30-32). Although it is difficult to establish the reasons of these different results, it has

been suggested that differences in the genetic and/or environmental background and in recruitment

procedures may have played a role (9).

Another important clinical outcome we examined was the occurrence of insulin resistance, IGT and

MS according to K121Q and rs997509, the latter being a polymorphism which has been suggested

to drive the association between K121Q and diabetes. Particularly, Bochenski et al., by studying

five ENPP1 gene variants identified a risk haplotype for developing type 2 diabetes. The haplotype

contained both the rs997509 T allele and the 121Q allele. The authors observed that when the

carriers of the T allele of rs997509 were excluded from the analysis, the frequency of the remaining

121Q carriers was not different in the type 2 diabetic cases and controls (14).

We report that rs997509 variant has an effect on glucose metabolism and MS independently of the

presence of 121Q and that the highest degree of insulin resistance in our population was apparent

only when the rs997509 T allele was present (table 3). Our results suggest that rs997509 T allele is

strongly associated with insulin resistance and related abnormalities in obese children and

adolescents and that the T allele drives also the 121Q allele association with insulin resistance. In

fact, the association observed between 121Q and insulin resistance disappeared when we compared

only subjects not carrying the rs997509 T allele. This observation can be ascribed to the strong

linkage disequilibrium existing between the two variants and can explain why contrasting results

concerning the association between K121Q and type 2 diabetes have been reported. Moreover, the

11

observation that this association was not present in lean subjects means that obesity is essential for

revealing the phenotype, with variant’s effect being noticed only when, as usually happens in obese

children, an increasing insulin production in response to fat accumulation is required.

Remarkably, in obese subjects, the higher degree of insulin resistance seemed to influence also

serum glucose levels and consequently insulin secretion during OGTT. Subjects carrying the

rs997509 T allele, in fact, showed during the curve constantly higher glucose levels and

consequently higher insulin secretion than C homozygotes. The progressive inadequateness of

insulin response to hyperglicemia of subjects carrying rs997509 T allele is documented also by the

higher prevalence of IGT in this group of patients respect to obese subjects homozygotes for the C

allele. Children with the T allele, in fact, showed, not only a higher prevalence of IGT, but also of

metabolic syndrome, that is consequent to insulin resistance (3). Likely, in subjects carrying the

rs997509 T allele, hyperinsulinemia causes not only the worsening of the metabolic status, but also

a progressive impairment of beta cell function, via fat deposition in the pancreas.

Unfortunately, functional studies exploring the effect of the rs997509 variant on ENPP1 synthesis

and function are not available. The rs997509 is located in the 3’ end of intron 1 in a region

containing a regulatory element, as reported previously (14) and suggested by the Five Regulatory

track of the University of California Santa Cruz genome browser (14). Moreover, another

polymorphism, rs9493114, which is in complete linkage disequilibrium with the rs997509, has been

described in intron 8 and may be the functional responsible of this association, although other in

vivo and in vitro reports on this variant, at our knowledge, do not exist (14).

In conclusion, consistently with previous reports this study supports a genetic link between ENPP1

gene variants and metabolic abnormalities in obese children. Moreover, we suggest that the ENPP1

rs997509 variant is strongly associated with insulin resistance, metabolic syndrome and IGT in

obese children and adolescents and that it drives the known association between the 121Q variant

and insulin resistance.

12

References

1) http://www.iotf.org/childhoodobesity.asp

2) Weiss R, Dziura J, Burgert TS, Tamborlane WV, Taksali SE, Yeckel CW, Allen K,

Lopes M, Savoye M, Morrison J, Sherwin RS, Caprio S 2004 Obesity and the metabolic

syndrome in children and adolescents. N Engl J Med.; 350: 2362-74

3) Reaven GM Banting lecture 1988. Role of insulin resistance in human disease.

Diabetes. 1988 ; 37: 1595-607

4) McGarry JD 2002 Banting lecture 2001: dysregulation of fatty acid metabolism in the

etiology of type 2 diabetes. Diabetes. 51: 7-18

5) Sinha R, Fisch G, Teague B, Tamborlane WV, Banyas B, Allen K, Savoye M, Rieger V,

Taksali S, Barbetta G, Sherwin RS, Caprio S 2002 Prevalence of impaired glucose

tolerance among children and adolescents with marked obesity. N Engl J Med.;346: 802-10.

6) Invitti C, Gilardini L, Viberti G 2002 Impaired glucose tolerance in obese children and

adolescents. N Engl J Med. 347: 290-2

7) Maddux BA, Sbraccia P, Kumakura S, Sasson S, Youngren J, Fisher A, Spencer S,

Grupe A, Henzel W, Stewart TA 1995 Membrane glycoprotein PC-1 and insulin

resistance in non-insulin- dependent diabetes mellitus. Nature. 373: 448-51

8) Goding JW, Terkeltaub R, Maurice M, Deterre P, Sali A, Belli SI 1998 Ecto-

phosphodiesterase/pyrophosphatase of lymphocytes and non-lymphoid cells: structure and

function of the PC-1 family. Immunol Rev.161:11-26

9) Goldfine ID, Maddux BA, Youngren JF, Reaven G, Accili D, Trischitta V, Vigneri R,

Frittitta L 2008 The role of membrane glycoprotein plasma cell antigen 1/ectonucleotide

pyrophosphatase phosphodiesterase 1 in the pathogenesis of insulin resistance and related

abnormalities. Endocr Rev. 29: 62-75

13

10) Bollen M, Gijsbers R, Ceulemans H, Stalmans W, Stefan C 2000 Nucleotide

pyrophosphatases/phosphodiesterases on the move. Crit Rev Biochem Mol Biol. 35: 393-

432

11) Costanzo BV, Trischitta V, Di Paola R, Spampinato D, Pizzuti A, Vigneri R, Frittitta L

2001 The Q allele variant (GLN121) of membrane glycoprotein PC-1 interacts with the

insulin receptor and inhibits insulin signaling more effectively than the common K allele

variant (LYS121). Diabetes. 50: 831-6

12) Meyre D, Bouatia-Naji N, Tounian A, Samson C, Lecoeur C, Vatin V, Ghoussaini M,

Wachter C, Hercberg S, Charpentier G, Patsch W, Pattou F, Charles MA, Tounian P,

Clément K, Jouret B, Weill J, Maddux BA, Goldfine ID, Walley A, Boutin P, Dina C,

Froguel P 2005Variants of ENPP1 are associated with childhood and adult obesity and

increase the risk of glucose intolerance and type 2 diabetes. Nat Genet. 37: 863-7

13) Böttcher Y, Körner A, Reinehr T, Enigk B, Kiess W, Stumvoll M, Kovacs P 2006

ENPP1 variants and haplotypes predispose to early onset obesity and impaired glucose and

insulin metabolism in German obese children. J Clin Endocrinol Metab. 91: 4948-52

14) Bochenski J, Placha G, Wanic K, Malecki M, Sieradzki J, Warram JH, Krolewski AS

2006 New polymorphism of ENPP1 (PC-1) is associated with increased risk of type 2

diabetes among obese individuals. Diabetes. 55: 2626-30

15) Cole TJ 1990 The LMS method for constructing normalized growth standards. Eur J Clin

Nutr. 44: 45-60

16) Rosner B, Prineas RJ, Loggie JM, Daniels SR 1993 Blood pressure nomograms for

children and adolescents, by height, sex and age in the United States. J Pediatr. 123: 871-86

17) Tanner JM, Whitehouse RH 1976 Clinical longitudinal standards for height, weight,

height velocity, weight velocity, and stages of puberty. Arch Dis Child. 51: 170-9

18) Santoro N, Miraglia del Giudice E, Cirillo G, Raimondo P, Corsi I, Amato A,

Grandone A, Perrone L 2004 An Insertional polymorphism of the Proopiomelanocortin

14

(POMC) gene is associated with fasting insulin levels in childhood obesity. J Clin End

Metab 89: 4846-9

19) Third report of the National Cholesterol Education Program Expert Panel on

Detection, Evaluation and Treatment of High Blood Cholesterol in Adults (Adult

Treatment Panel III). Bethesda, Md.: National Heart, Lung, and Blood Institute, May 2001.

(NIH publication no. 01-3670.)

20) Alberti KG, Zimmet PZ. 1998 Definition, diagnosis and classification of diabetes mellitus

and its complications. 1. Diagnosis and classification of diabetes mellitus provisional report

of a WHO consultation. Diabet Med 15:539-553.

21) Matsuda M, DeFronzo RA 1999 Insulin sensitivity indices obtained from oral glucose

tolerance testing; comparison with euglycemic insulin clamp. Diabetes Care 22: 1462-70

22) Yeckel CW, Weiss R, Dziura J, Taksali SE, Dufour S, Burgert TS, Tamborlane WV,

Caprio S 2004 Validation of insulin sensitivity indices from oral glucose tolerance test

parameters in obese children and adolescents. J Clin Endocrinol Metab. 89: 1096-1101

23) Terwilliger J, Ott J: Handbook for human genetic linkage. Baltimore, MD, Johns Hopkins

University Press, 1994

24) Santoro N, Perrone L, Cirillo G, Raimondo P, Amato A, Brienza C, Miraglia del

Giudice E 2007 Effect of the melanocortin-3 receptor C17A and G241A variants on weight

loss in childhood obesity. Am J Clin Nutr. 85: 950-3

25) Pizzuti A, Frittitta L, Argiolas A, Baratta R, Goldfine ID, Bozzali M, Ercolino T,

Scarlato G, Iacoviello L, Vigneri R, Tassi V, Trischitta V 1999 A A polymorphism

(K121Q) of the human glycoprotein PC-1 gene coding region is strongly associated with

insulin resistance.Diabetes. 48: 1881-4

26) Goldfine ID, Maddux BA, Youngren JF, Frittitta L, Trischitta V, Dohm GL 1998

Membrane glycoprotein PC-1 and insulin resistance. Mol Cell Biochem. 182: 177-84

15

27) Goldfine ID, Maddux BA, Youngren JF, Trischitta V, Frittitta L 1999 Role of PC-1 in

the etiology of insulin resistance Ann N Y Acad Sci. 892:204-22

28) Barroso I, Luan J, Middelberg RP, Harding AH, Franks PW, Jakes RW, Clayton D,

Schafer AJ, O'Rahilly S, Wareham NJ 2003 Candidate gene association study in type 2

diabetes indicates a role for genes involved in beta-cell function as well as insulin action.

PLoS Biol. 1:E20.

29) Matsuoka N, Patki A, Tiwari HK, Allison DB, Johnson SB, Gregersen PK, Leibel RL,

Chung WK 2006 Association of K121Q polymorphism in ENPP1 (PC-1) with BMI in

Caucasian and African-American adults. Int J Obes . 30: 233-7.

30) Grarup N, Urhammer SA, Ek J, Albrechtsen A, Glümer C, Borch-Johnsen K,

Jørgensen T, Hansen T, Pedersen O 2006 Studies of the relationship between the ENPP1

K121Q polymorphism and type 2 diabetes, insulin resistance and obesity in 7,333 Danish

white subjects. Diabetologia. 49: 2097-104.

31) Lyon HN, Florez JC, Bersaglieri T, Saxena R, Winckler W, Almgren P, Lindblad U,

Tuomi T, Gaudet D, Zhu X, Cooper R, Ardlie KG, Daly MJ, Altshuler D, Groop L,

Hirschhorn JN 2006 Common variants in the ENPP1 gene are not reproducibly associated

with diabetes or obesity. Diabetes. 55: 3180-4.

32) Weedon MN, Shields B, Hitman G, Walker M, McCarthy MI, Hattersley AT, Frayling

TM 2006 No evidence of association of ENPP1 variants with type 2 diabetes or obesity in a

study of 8,089 U.K. Caucasians. Diabetes. 55: 3175-9.

16

Figures legends.

Figure 1: Serum insulin levels (A) and serum glucose levels (B) during OGTT according to the

rs997509 SNP genotype. The squares represent the rare allele carriers, the circle represent the

common allele homozygotes. Data are shown as means and deviation scores.

Figure 2: Serum insulin levels (A) and serum glucose levels (B) during OGTT according to the

K121Q genotype. The squares represent the rare allele carriers, the circle represent the common

allele homozygotes. Data are shown as means and deviation scores.

Table1 Clinical features of obese subjects according to the occurrence of Metabolic Syndrome.

Presence of Metabolic

Syndrome (n 143 )

Absence of Metabolic

Syndrome (n 266)

P

Age (years) 11.3±3.0 11.6±3.0 0.3

Pubertal subjects (%) 37 % 28 % 0.05

BMI (Kg/m2) 34 ±6 32±5 0.01

z-score BMI 3.9±0.9 3.4±1.3 0.01

Insulin (µµµµU/L)1 35.1±24.2 24.6± 16.2 <0.001

HOMA1 7.5±5.2 5.3±3.3 <0.001

WBISI1 1.9±1.7 2.7±2.0 0.001

HDL (mg/dl)1 37±5.9 50±8.9 <0.001

Triglycerides (mg/dl)1 114±41 86±72 <0.001

SBP (mmHg)2 121±13 116±14 <0.001

z-score SBP 1.2±1.2 0.8±1.0 0.007

DBP (mmHg)2 73±8 61±13 <0.001

z-score DBP 1.4±0.9 1.0±0.9 0.008

1 Adjusted for age, sex and pubertal stage. 2 Adjusted for age, sex and height.

K121Q rs997509

KK (297) KQ (97) + QQ (15) P CC (357) CT (48) + TT (4) P

Age (years) 11.2±3 11.5±2 0.5 11.7 ± 2.7 11.2 ± 2.9 0.2

Pubetal subjects (%) 30 33 0.4 30 35 0.5

BMI (Kg/m2) 32.6±5 33.0 ± 6 0.5 32.2 ± 5.4 33 ± 6.7 0.2

z-score BMI 3.1±0.9 2.9±0.9 0.4 3.0 ± 0.7 2.9 ± 0.4 0.8

Plasma glucose (mg/dl) 1 79.7±12 79.8±7 0.9 79.8 ± 8.9 81.5 ± 9.1 0.2

Insulin (µµµµU/l) 1 31.2±18.0 35.7±25.1 0.03 31.0±19.2 39.8±22.3 0.003

HOMA 1 6.2±3.7 7.2±5.4 0.04 6.2 ± 4.1 8.3 ± 5.2 <0.001

WBISI 1 2.5±2.0 2.3±1.5 0.09 2.5 ± 2.0 2.0 ± 1.5 0.04

HDL (mg/dl) 1

45±9.8 44±13 0.2 44±12 41±17 0.1

Triglycerides (mg/dl) 1

93±67 96±72 0.7 92±51 104±58 0.08

SBP (mmHg)2 118±18 120±31 0.4 117±19 120±33 0.2

z-score SBP 1.0±0.7 1.1±0.3 0.1 0.9±0.8 1.1±0.9 0.07

DBP (mmHg)2 67±14 69±21 0.2 64±17 69±35 0.09

z-score DBP 1.1±0.8 1.2±0.9 0.06 1.0±0.7 1.3±1.0 0.04

metabolic syndrome (%) 27 37 0.1 31 52 0.005

IGT (%) 3 7 0.06

4 13 <0.001

Before comparison the variable were log transformed, but geometric means are shown. When the general linear model was run

age, sex, BMI, pubertal stage (1) and height (

2) were used as covariates, the p-value shown are those obtained after adjustment.

Table 2 Clinical features of the obese subjects according to the K121Q and rs997509 genotypes

K121Q rs997509 subjects Insulin HOMA WBISI

KK CC 292 32.2±16.7 6.4±3.3 2.4-±1.3

KQ+QQ CC 65 31.9±19.3 6.4±4.1 2.5±1.6

KQ+QQ CT+TT 47 42.0±25.0 8.9±4.4 1.8±0.9

KK CT+TT 5 41.9±17.0 8.7±2.0 1.9±0.2

P-value 0.001 0.002 0.03

When the effect of both genotypes was tested with a general linear model, we used a model where, at every turn, HOMA,

insulin or WBISI was the dependent variable and sex, age, pubertal stage, K121Q and rs997509 were the covariates. The

results showed that the major effect on these three variables (insulin, HOMA and WBISI) was ascribed to the rs997509

variant that appeared to drive also the association between K121Q variant and insulin resistance. The data are presented as

mean +/- standard deviation.

Table 3: Association between parameters of glucose and insulin metabolism with a combined evaluation of the ENPP1

rs997509 and K121Q polymorphisms

0

50

100

150

200

250

0 30 60 90 120

AµU/L

time

0

20

40

60

80

100

120

140

160

180

0 30 60 90 120

mg/dl

time

B

Insulin

Plasma glucose

*

**

*

*

*

Figure 1

0

50

100

150

200

250

0 30 60 90 120

AµU/L

time

0

20

40

60

80

100

120

140

160

180

0 30 60 90 120

mg/dl

time

B

Insulin

Plasma glucose

*

Figure 2