| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Diabetes and Vascular Medicine, Peninsula Medical School (S.M.S.M., A.T.H., B.K., T.T., T.M.F.), Exeter, United Kingdom EX2 5AX; Department of Endocrinology and Metabolism, Peninsula Medical School (B.S.M., L.D.V., T.J.W.), Plymouth, United Kingdom PL6 80H; University of Wales College of Medicine (D.D.), Cardiff, United Kingdom; and University of Bristol School of Social Medicine (A.M., G.D.S., Y.B.-S.), Bristol, United Kingdom BS2 8DZ
Address all correspondence and requests for reprints to: Dr. Timothy M. Frayling, Department of Diabetes and Vascular Medicine, Peninsula Medical School, Barrack Road, Exeter, United Kingdom EX2 5AX. E-mail: t.m.frayling{at}exeter.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Reduced fetal growth is associated with an increased risk of type 2 diabetes and cardiovascular disorders in adulthood (5). The fetal insulin hypothesis postulates that this association could be due in part to fetal genes as well as to the intrauterine environment, as proposed by Barker and colleagues (6). In support of the fetal insulin hypothesis, subjects with a diabetic father are of lower birth weight than subjects with either a diabetic mother or nondiabetic parents (7, 8). Rare mutations in genes affecting insulin secretion and action also result in substantial effects on fetal growth (9, 10, 11, 12, 13, 14, 15, 16).
The genes influencing normal variation in fetal growth are not known. Recently, a genome-wide linkage analysis of 269 Pima Indian siblings found evidence for linkage of birth weight to chromosome 11 (map position 88cM) (17). In this region, birth weight was linked predominantly to paternally derived alleles. A number of studies have assessed the role of candidate variants in birth weight. These include the insulin gene variable number of tandem repeat polymorphism (INS-VNTR) (18, 19), the mt16189 variant (20), and the insulin receptor substrate 1 variant, G972R (21).
The INS-VNTR lies upstream of the insulin (INS) and IGF-II (IGF2) genes and is associated with variations in their transcription. The common INS-VNTR allele lengths fall into two broad classes: class I (26–63 repeats) and class III (141–209 repeats). INS-VNTR class I alleles increase INS transcription in the ß-cell and fetal pancreas in some studies (22, 23, 24). A correlation has also been found between increased insulin production at the physiological level in humans and class I alleles (25, 26), although two other studies have failed to find this association (27, 28). In contrast, class III alleles are associated with substantially higher levels of INS mRNA in the thymus than class I alleles (29, 30). In normal human term placenta, median IGF2 steady state mRNA, which is transcribed exclusively from the paternal chromosome, is significantly greater with class I alleles than with class III alleles (31).
INS and IGF2 are both imprinted genes, with preferential expression of the paternal allele in a number of tissues (32, 33, 34). Defects in imprinting mechanisms are often associated with alterations in growth (35, 36). This includes the VNTR-INS-IGF2 region, where loss of IGF2 imprinting resulting in the biallelic expression of IGF2 occurs in the majority of cases of Beckwith-Wiederman syndrome, which is characterized by prenatal and postnatal overgrowth (37, 38, 39).
The INS-VNTR class III allele is associated with increased head circumference and increased birth weight in a cohort of 758 term singletons from the Avon Longitudinal Study of Pregnancy and Childhood (ALSPAC). The association with birth weight was limited to the subset of children who did not show postnatal growth realignment (nonchangers); III/III children were 238 g heavier, on the average, than I/I children (P = 0.009) (18). In a study of 218 Hertfordshire men, this association was not found, although the small sample size provides low statistical power to replicate the ALSPAC result (19). Although the INS-VNTR-birth weight association has not been replicated in a normal cohort, a study of 141 girls with precocious puberty found the class I allele to be more frequent in the low birth weight group (birth weight SD score, less than -2) than the normal birth weight group (birth weight SD score, more than -2; P = 0.01) (40). Recently, a study of INS-VNTR genotype and birth weight in Pima Indians found the class III allele to be associated with lower, rather than higher, birth weight (-140 g/copy of the class III allele; P = 0.04) and also found a trend for lower birth weight with paternally transmitted class III alleles (paternally transmitted, -250 g; P = 0.05; maternally transmitted, -111 g; P = 0.43). However, this difference was not statistically significant (P = 0.50) (41).
In addition to the associations with altered fetal growth, the INS-VNTR polymorphism is associated with susceptibility to disease. The strongest association is with type 1 diabetes, where class I alleles predispose to diabetes in a recessive manner, whereas class III alleles are dominantly protective (42). The INS-VNTR III/III genotype has also been associated with anovulatory polycystic ovary syndrome (43), paternally derived class III alleles with type 2 diabetes (44), and paternally derived class I alleles with childhood obesity (45). There have been few studies of the role of the INS-VNTR in type 2 diabetes-related intermediate traits. One study of 218 Hertfordshire men showed an association between the class III allele and impaired glucose tolerance/type 2 diabetes in those subjects who showed no postnatal growth realignment (nonchangers) (19).
Although imprinting occurs at the INS-VNTR locus, only one study has been performed to assess the potential role of the parent of origin of the INS-VNTR allele and fetal growth (41). The association between the class III allele and increased fetal growth in the normal population has also not been replicated. The replication of associations between gene variants and polygenic traits is vital before conclusions can be drawn about the roles of these variants. We therefore investigated the role of the INS-VNTR in fetal growth in 3 large Caucasian cohorts consisting of a total of 1184 children and 1099 parents. Parents were available from 2 of these cohorts, which enabled the assessment of potential parent of origin effects in 467 children. We also investigated the role of variation at the INS-VNTR in postnatal growth, including body mass index (BMI) and intermediate traits associated with type 2 diabetes, such as insulin resistance and ß-cell function.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Barry-Caerphilly growth cohort (BCG)
Details of the BCG cohort are available elsewhere (46). Briefly, it consists of 660 subjects born between 1971–1974 with detailed anthropometric measurements available at birth through 8 yr of age and oral glucose tolerance test data taken at a mean age of 25 yr. Insulin was measured using an ELISA (K6219, Dako, Carpenteria, CA). The insulinogenic index was calculated as (insulin 30 min - insulin 0 min)/(glucose 30 min - glucose 0 min). Birth weight was obtained from hospital records, and subsequently weight was measured on a portable beam balance. Head circumference at birth data were available from 614 subjects.
Exeter Family Study (EFS)
The EFS is a consecutive cohort of subjects born in Exeter, Devon, UK, from 1999 on. A fasting blood sample was taken from both parents, coinciding with approximately 28 wk gestation. Insulin was measured using an immunoenzymometric assay (Insulin EASIA, BioSource, Brussels, Belgium).
For this study we genotyped the 762 parents and 360 newborns delivered at 36–44 wk gestation from whom DNA was available. Cord blood was used to extract DNA from newborns. Head circumference at birth data were available from 359 babies. Fasting glucose and fasting insulin were available from 446 parents.
Plymouth EarlyBird Study
The EarlyBird Study is a prospective cohort of subjects born 1995–1996 that started school in Plymouth, Devon, UK. A fasting blood sample was taken from both parents coinciding with the child’s first baseline visit (mean, 4.9 yr after birth of study child). Insulin was measured using an insulin-specific monoclonal antibody (Immulite Diagnostic Products, Los Angeles, CA). We genotyped the 164 children and 337 parents from whom DNA was available. Head circumference at birth data are not available from this study.
Materials and Methods
Genotyping. We amplified, using PCR, the INS -23A/T polymorphism, which is in very tight linkage disequilibrium with INS-VNTR class (18). A 360-bp fragment containing the INS -23 polymorphism was PCR-amplified with the following primers: forward, 5'-AGCAGGTCTGTTCCAAGG-3'; and reverse, 5'-CTTGGGTGTGTAGAAGAAGC-3'. PCR was performed in a 25-µl volume containing in addition to the standard reagents: 1 M betaine, 5% dimethylsulfoxide, and 0.5 U AmpliTaq Gold (PE Applied Biosystems, Warrington, UK). PCR cycling conditions were denaturation at 96 C for 12 min, followed by 35 cycles of denaturation at 94 C for 1 min, annealing at 54 C for 1 min, and extension at 72 C for 45 sec, with a final 10-min extension at 72 C. We digested the PCR product with the restriction enzyme Hph1 (New England Biolabs) according to the manufacturer’s instructions, which cuts at the INS -23 site.
Genotyping quality control. For all 3 cohorts, 1 column of 8 wells from each 96-well plate of DNA was randomly selected and blindly regenotyped. The genotyping error rate was less than 0.6%. Families from the EFS inconsistent for alleles of the INS-VNTR or 7 other previously genotyped single nucleotide polymorphisms were genotyped for 6 microsatellite markers. Allele inconsistencies were confirmed in 6.8% of all families in the EFS. One family (0.6%) from the Plymouth EarlyBird cohort was inconsistent for INS-VNTR alleles, but no other inconsistent families were identified in this cohort from 3 single nucleotide polymorphisms previously genotyped. All members of a family inconsistent for a polymorphism were excluded from further analyses.
Statistical analyses
Hardy-Weinberg equilibrium was assessed by 
2 comparisons of observed genotype frequencies with expected genotype frequencies inferred from observed allele frequencies. Comparisons of data by genotype were made on log-transformed data when the data distribution was skewed [e.g. fasting insulin, homeostasis model assessment score (HOMAS), and insulinogenic index], and the data presented have been back transformed.
In all cohorts, extremes of gestation age (<36 and >44 wk) were excluded. Birth weight was analyzed using a general linear model, with sex as a factor and gestational age as a covariate. The criteria defined by Dunger and colleagues (18) were used to stratify by postnatal growth realignment in the BCG cohort. Subjects were assigned as changers if their z-score differed by 0.67 SD of weight or more between birth and 2 yr of age; otherwise they were assigned as nonchangers. In the EarlyBird study the same criteria were used, except weight percentiles at entry to the study (4.9 yr) were used rather than 2 yr percentiles.
Parent of origin effects were examined in the EFS and the EarlyBird study in the 467 families when DNA from both parents was available. An additional 28 families from the EFS and 9 families from the EarlyBird study were excluded because all 3 family members were heterozygous for the I/III polymorphism, meaning we could not determine the parent of origin of the class III allele inherited by the child.
Combined effect sizes and P values were determined using means and SDs from individual cohorts after checking for homogeneity between studies (Studentized range, Q) and assuming a random effects model.
For postnatal anthropometry, data were derived after adjusting for age of the subject at the time of measurement. As the insulin assays used in the three cohorts are not identical, we have only combined the fasting insulin values and the HOMAS derived from it using z-scores. Insulin resistance was calculated using HOMA from fasting insulin and glucose (47). Subjects were excluded if they had diabetes or another known factor that alters glucose tolerance or insulin sensitivity, e.g. pregnancy (unless stated) or taking hormonal contraceptives. Oral glucose tolerance tests were performed in the BCG cohort; therefore, 2 h glucose values and the insulinogenic index [(30 min insulin - fasting insulin)/(30 min glucose - fasting glucose)] were used to examine ß-cell function (46).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Tables 1–3
show the details of the analysis of INS-VNTR genotype in relation to birth weight. There was no effect of genotype on birth weight in the total of 1130 subjects either across all 3 genotypes or when comparing III/III subjects with I/I subjects (Table 1).
|
|
|
). In addition to birth weight, measures of neonatal head circumference were available from the BCG cohort and EFS cohort. However, no effects of genotype (at P < 0.05) were seen with head circumference at birth, when adjusting for the baby’s sex and gestational age: EFS: I/I, 35.3 cm [95% confidence interval (CI), 35.1–35.5; n = 177]; I/III, 35.3 cm (95% CI, 35.1–35.5; n = 150); III/III, 35.3 cm (95% CI, 34.8–35.7; n = 32); I/I vs. III/III, P = 1.00; BCG: I/I, 35.9 cm (95% CI, 35.8–36.0; n = 293); I/III, 36.0 cm (95% CI, 35.8–36.1; n = 258); III/III, 35.7 cm (95% CI, 35.4–36.0; n = 63); I/I vs. III/III, P = 0.16; EFS and BCG combined: I/I vs. III/III, P = 0.34; Q, P = 0.49. No effect of INS-VNTR genotype on head circumference at birth was seen in nonprimiparous EFS babies: I/I, 35.4 cm (95% CI, 35.2–35.6; n = 108); I/III, 35.2 cm (95% CI, 34.9–35.4; n = 87); III/III, 35.3 (95% CI, 34.8–35.8; n = 18); I/I vs. III/III, P = 0.67. Maternal weight at 28 wk pregnancy and maternal fasting glucose, which is known to be associated with fetal growth levels, were available for 279 subjects from the EFS. Again, there was no correlation between fetal growth and INS-VNTR genotype when correcting for these variables (data not shown).
The parent of origin of the INS-VNTR allele was ascertained for 467 EFS and EarlyBird subjects. However, there was no difference in birth weight across genotype groups (at P < 0.05) when stratified by parent of origin of the class III allele (paternal allele first: III/xvs. I/x, -15 g (-107 to +77; P = 0.74); size of effect vs. I/I subjects: I/III, -37 g (-146 to +73; P = 0.51); and III/I, -59 g (-172 to +55; P = 0.31; Table 2).
We explored the effect of maternal genotype on birth weight, adjusting for maternal weight at 20 wk gestation and the baby’s gestational age and sex. We also explored the effect of paternal genotype on birth weight, adjusting for the baby’s gestational age and sex. We found no support for an effect of parental genotype on the baby’s birth weight (maternal genotype: I/I vs. III/III, P = 0.91; paternal genotype: I/I vs. III/III, P = 0.72).
INS-VNTR and type 2 diabetes-related intermediate traits
Measures of fasting and 2-h insulin and fasting and 2-h glucose were available from 660 subjects in the BCG cohort. Measures of glycemia (fasting and 2 h), insulin sensitivity, and ß-cell function were not associated with INS-VNTR genotype (Table 4a).
|
]. When the studies were studied separately, this association was significant in the EarlyBirds, but not in the BCG or EFS populations. In each cohort, class III homozygotes had greater BMI than class I homozygotes (except EFS males, in whom the mean BMI of both homozygote genotypes was 26.5). This trend is statistically significant when the 3 cohorts are combined [Table 4d; excluding EFS pregnant females; III/III vs. I/I: effect size, +0.98 kg/m2 (95% CI, +0.15–1.82); P = 0.02]. Class III homozygous pregnant women from the EFS are heavier than class I homozygous pregnant women at 20 wk gestation, but this was not significant (P = 0.08). |
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Parent of origin effects have been reported between INS-VNTR genotype and type 1 diabetes (42), type 2 diabetes (44), and juvenile obesity (45). This could be through imprinting at the VNTR-INS-IGF2 locus or maternal genotype effects. We hypothesized that an effect of the INS-VNTR genotype on birth weight could be mediated through a parent of origin effect. However, we found no support for parent of origin effects of the INS-VNTR genotype on fetal growth.
A number of reasons may explain our failure to replicate the results of the ALSPAC study. Firstly, the initial study may have overestimated the size of the effect of the INS-VNTR on fetal growth, which may be small or even nonexistent. This may have reduced our power to detect an effect of genotype on birth weight. Secondly, the heterogeneity of our samples may have reduced our power. However, our biggest single studies consisted of 660 newborns, compared with 758 newborns in the ALSPAC study, and we found no significant heterogeneity between the studies when they were combined. Recent meta-analyses of genetic association studies have highlighted the problems of establishing an accurate assessment of whether a phenotype is associated with common DNA variants. It has been advocated that evidence for the role of genetic variation in polygenic traits is built up over many studies (48). Therefore, we may not know the true size effect of INS-VNTR on fetal growth until further studies are performed.
An additional reason for the discrepancy may be the different intrauterine exposures between studies. However, all of our studies and the initial positive study (18) took place in the United Kingdom in more than 98% white Caucasian subjects. Our largest cohort (BCG) was born between 1971 and 1974, and birth weight has increased with time, possibly reflecting the changing intrauterine environment. The EFS and EarlyBird cohorts were born at a similar time period (within 10 yr) and within 150 miles of the initial study, and it is difficult to consider that this would represent a significantly different environmental exposure. There was no evidence of heterogeneity between our cohorts for measures of size at birth, although we cannot exclude the possibility of different effects operating within them. There is good evidence for association of the INS-VNTR polymorphism with type 2 diabetes (19, 44) and altered insulin levels in juvenile obesity, with parent of origin effects (45, 49). It is not yet possible to test these associations in our young cohorts. We found some evidence in our population cohorts for altered insulin resistance and BMI, but not ß-cell dysfunction or raised fasting glucose. However, our population cohorts are underpowered, as other studies used extreme phenotypes to observe these associations, and phenotypes related to subsequent type 2 diabetes may not be seen in young adults. Heterogeneity of samples may have underestimated an effect of INS-VNTR on BMI. We recommend that large studies be undertaken to directly examine the hypothesis generated by our observations that the INS-VNTR genotype could influence BMI and insulin sensitivity.
Parity could be a confounding factor on size at birth (50). However, we did not see any trend in the EFS between birth weight and INS-VNTR genotype when correcting for parity (n = 313; I/I vs. III/III, P = 0.88) or when dichotomizing babies into first pregnancies (n = 97; I/I vs. III/III, P = 0.94) and nonfirst pregnancies (n = 206; I/I vs. III/III, P = 0.85).
The -23Hph polymorphism that we used to determine INS-VNTR genotype is in tight linkage disequilibrium with INS-VNTR allele class and allows comparison with previous population and functional studies. However, the -23 polymorphism cannot be used to discriminate the many subcategories of class I alleles. The primary sequence of the repeat elements varies and could have potential functional consequences. For example, class I alleles overall show a very strong association with type 1 diabetes, yet the most common class I allele subcategory (allele 814) was not found to be overtransmitted to type 1 diabetic offspring. Furthermore, class I allele 698 was found to be undertransmitted to diabetic offspring, and in vivo studies suggest that insulin RNA expression associated with allele 618 differs from that of other class I alleles (24). Further in vitro studies are required to determine the association between class I subcategory and insulin transcription.
We found no evidence for heterogeneity among the 3 adult populations (excluding EFS pregnant women) for any anthropometric measurement, and we therefore combined the data using means and SDs from individual populations. We observed a trend between INS-VNTR genotype and adult BMI, with III/III subjects being heavier than I/I, in the BCG and Plymouth EarlyBird populations, but not in men from the EFS. This trend was statistically significant when the populations were combined (P = 0.02). Pregnant women from the EFS were not included in this analysis. However, we noted that there was a weak trend between INS-VNTR genotype and mother’s 20 wk gestation weight (P = 0.08), and these observations warrant further investigation in other populations. Previous studies have not found a similar trend in a normal population (51), and we have performed many statistical tests for this study. We observed a nominally significant association between INS-VNTR genotype and insulin sensitivity (HOMAS) when the cohorts were combined (I/I vs. I/III, P = 0.04). Association between the class III allele and insulin resistance has been reported in studies of INS-VNTR and disease-enriched populations (27, 40, 43, 49). This association was apparent in the EarlyBirds, but we did not find it in the BCG or EFS populations. Given the number of statistical tests we have undertaken during this study, replication of this observation in other healthy populations is required. The trends among INS-VNTR, BMI, and insulin sensitivity are small, which could be a result of the nature of the phenotypes measured and the heterogeneity among the cohorts.
Our lack of support for an effect of INS-VNTR genotype on measures of fetal growth and nominal evidence for some intermediate traits are not incompatible with the findings that this locus is associated with adult diseases such as polycystic ovarian syndrome and type 2 diabetes. Apart from the possibilities of false positive and false negative results, it is possible that some type 2 diabetes genes influence the rate of decline in ß-cell function or insulin sensitivity, rather than alter a trait throughout life.
In conclusion, we did not replicate previous reports of altered fetal growth associated with INS-VNTR genotype in our three large cohorts despite allowing for parent of origin effects and postnatal growth realignment.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
A.T.H. is a Wellcome Trust Career Leave Research fellow.
T.M.F. is a career scientist of the South and West National Health Service Research Directorate.
Abbreviations: ALSPAC, Avon Longitudinal Study of Pregnancy and Childhood; BCG, Barry-Caerphilly growth cohort; BMI, body mass index; CI, confidence interval; EFS, Exeter Family Study; HOMAS, homeostasis model assessment score; INS-VNTR, insulin gene variable number of tandem repeats minisatellite.
Received April 11, 2003.
Accepted October 3, 2003.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. O Mook-Kanamori, J J Miranda Geelhoed, E. A P Steegers, J. C M Witteman, A. Hofman, H. A Moll, C. M van Duijn, A. C S Hokken-Koelega, and V. W V Jaddoe Insulin gene variable number of tandem repeats is not associated with weight from fetal life until infancy: the Generation R Study Eur. J. Endocrinol., December 1, 2007; 157(6): 741 - 748. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. B. Dunger, C. J. Petry, and K. K. Ong Genetics of Size at Birth Diabetes Care, July 1, 2007; 30(Supplement_2): S150 - S155. [Full Text] [PDF]
|
|
|
|
 |
|
 E. Landmann, F. Geller, J. Schilling, S. Rudloff, E. Foeller-Gaudier, and L. Gortner Absence of the Wild-type Allele (192 Base Pairs) of a Polymorphism in the Promoter Region of the IGF-I Gene but Not a Polymorphism in the Insulin Gene Variable Number of Tandem Repeat Locus Is Associated With Accelerated Weight Gain in Infancy Pediatrics, December 1, 2006; 118(6): 2374 - 2379. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Bougneres Genotypic and phenotypic complexity at the insulin variable number of tandem repeats locus. J. Clin. Endocrinol. Metab., November 1, 2006; 91(11): 4246 - 4249. [Full Text] [PDF]
|
|
|
|
|
|
B. Heude, C. J. Petry, the Avon Longitudinal Study of Parents Children (A, M. Pembrey, D. B. Dunger, and K. K. Ong The Insulin Gene Variable Number of Tandem Repeat: Associations and Interactions with Childhood Body Fat Mass and Insulin Secretion in Normal Children J. Clin. Endocrinol. Metab., July 1, 2006; 91(7): 2770 - 2775. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T.-A. Vu-Hong, E. Durand, S. Deghmoun, P. Boutin, D. Meyre, D. Chevenne, P. Czernichow, P. Froguel, and C. Levy-Marchal The INS VNTR Locus Does Not Associate with Smallness for Gestational Age (SGA) but Interacts with SGA to Increase Insulin Resistance in Young Adults J. Clin. Endocrinol. Metab., June 1, 2006; 91(6): 2437 - 2440. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Sandhu, B. Heude, E. H. Young, R. Luben, J. Luan, K.-T. Khaw, J. Todd, and N. J. Wareham INS VNTR Class Genotype and Indexes of Body Size and Obesity: Population-Based Studies of 7,999 Middle-Aged Men and Women Diabetes, September 1, 2005; 54(9): 2812 - 2815. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Meyre, P. Boutin, A. Tounian, M. Deweirder, M. Aout, B. Jouret, B. Heude, J. Weill, M. Tauber, P. Tounian, et al. Is Glutamate Decarboxylase 2 (GAD2) a Genetic Link between Low Birth Weight and Subsequent Development of Obesity in Children? J. Clin. Endocrinol. Metab., April 1, 2005; 90(4): 2384 - 2390. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. B. Meigs, J. Dupuis, A. G. Herbert, C. Liu, P. W. F. Wilson, and L. A. Cupples The Insulin Gene Variable Number Tandem Repeat and Risk of Type 2 Diabetes in a Population-Based Sample of Families and Unrelated Men and Women J. Clin. Endocrinol. Metab., February 1, 2005; 90(2): 1137 - 1143. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Bennett, U. Sovio, A. Ruokonen, H. Martikainen, A. Pouta, S. Taponen, A.-L. Hartikainen, V. J. King, P. Elliott, M.-R. Jarvelin, et al. Variation at the Insulin Gene VNTR (Variable Number Tandem Repeat) Polymorphism and Early Growth: Studies in a Large Finnish Birth Cohort Diabetes, August 1, 2004; 53(8): 2126 - 2131. [Abstract] [Full Text] [PDF]
|
|
Read all eLetters
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
