|
|
||||||||
CLINICAL STUDY |
1 Departments of Endocrinology and2 Molecular Genetics, Hospital Universitario Ramón y Cajal and Universidad de Alcalá, Carretera de Colmenar Km. 9.1, Madrid E-28034, Spain, and 3 Section of Diabetes, Endocrinology and Nutrition, Hospital Universitario Dr Josep Trueta, E-17007 Girona, Spain
(Correspondence should be addressed to H F Escobar-Morreale; Email: hescobarm.hrc{at}salud.madrid.org)
| Abstract |
|---|
|
|
|---|
Design: Casecontrol association study.
Methods: One hundred and forty-five type 2 diabetic patients and 217 non-diabetic controls were genotyped for the ACAA-insertion/deletion polymorphism at the 3' UTR of IGFIIR. Phenotyping included anthropometrics and a metabolic profile, including serum lipid levels and surrogate indexes of insulin resistance whenever possible.
Results: Diabetic patients were more frequently homozygous for the wild type 144 bp allele of IGFIIR compared with controls (diabetic patients 77.2%, controls 51.6%, P<0.001) suggesting a potential protective role against type 2 diabetes for the IGFIIR 140 bp variant. Carrying 140 bp alleles was associated with an odds ratio of having diabetes of 0.290 (95% confidence interval 0.1090.770), and controls homozygous for the wild type 144 bp allele presented with lower insulin and triglyceride levels, which are proxies for insulin resistance.
Conclusions: The ACAA-insertion/deletion polymorphism at the 3' UTR of IGFIIR is associated with type 2 diabetes and influences surrogate markers of insulin resistance in non-diabetic subjects.
| Introduction |
|---|
|
|
|---|
On the other hand, overexpression of IGF-II in pancreatic ß cells in transgenic mice models is followed by development of type 2 diabetes, fatty liver and obesity, despite an increase in ß cell mass (6). The role of IGF-II in the development of ß cells is further supported by the finding that impaired IGF-II production during fetal development may result into retardation of ß cell growth in a rat model of type 2 diabetes (7). Supporting a role of IGF-II in the development of insulin-resistant disorders in humans, IGF-II gene polymorphisms have been found in association with increased body mass index (BMI) in adult males (8), and with the polycystic ovary syndrome in Caucasian premenopausal women (9).
The IGF-II receptor is a multifunctional glycoprotein that has been proposed to influence tumor suppression and immunity (10, 11). The IGF-II receptor gene (IGFIIR) is located at chromosome 6q (12). Although in rodents, IGFIIR is imprinted and only maternally inherited alleles are expressed, in humans repressed expression of paternal IGFIIR alleles is only present in a small subset of subjects (13). Considering that the 6q region has shown evidence for linkage with insulin resistance and obesity-related metabolic phenotypes (14), we hypothesized that IGFIIR could be a positional candidate gene for type 2 diabetes.
This suspicion was further highlighted by our previous unpublished data suggesting an unexpected association of the ACAA-insertion/deletion polymorphism in the 3' UTR of IGFIIR with obesity and insulin resistance in hyperandrogenic women, although we were not able to demonstrate an association of this polymorphism with polycystic ovary syndrome (9), which was the actual aim of the study. Together with the association of the 6q region with insulin resistance and obesity-related metabolic phenotypes described above (14), our findings in hyperandrogenic women prompted us to evaluate the possible association of the ACAA-insertion/deletion polymorphism in the 3' UTR of IGFIIR with insulin resistance and type 2 diabetes in Caucasian individuals from Spain using a casecontrol approach.
| Subjects and methods |
|---|
|
|
|---|
Three hundred and sixty-two Caucasian subjects recruited for an ongoing study addressing non-classical cardiovascular risk factors, were included in the present study. Two hundred and seventeen non-diabetic controls were selected on the basis of fasting plasma glucose <6.1 mmol/l and BMI <40 kg/m2. Control subjects were neither receiving pharmacological treatment or had any evidence of metabolic disease other than overweight or non-morbid obesity. One hundred and forty-five patients with type 2 diabetes, defined according to the criteria of the American Diabetes Association (15), were prospectively recruited from diabetes outpatient clinics. Only patients who presented with a stable metabolic control within the previous 6 months, as defined by HbAlc values, were recruited. Pharmacological therapies for these patients included insulin (44.8%), oral hypoglycemic agents (72.9%), statins (38.0%), fibrates (10.6%), blood pressure lowering agents (61.5%), aspirin (42.7%) and allopurinol (4.2%).
For both diabetic patients and controls, exclusion criteria included: (i) clinically significant hepatic, neurological, endocrine or other major systemic disease, including malignancy, (ii) hemochromatosis, (iii) drug or alcohol abuse, defined as alcohol intake >80 g/day in men and >40 g/day in women, or serum trans-aminase activity more than twice the upper limit of normal, (iv) elevated serum creatinine concentration, (v) occurrence of an acute major cardiovascular event in the previous 6 months, (vi) acute illnesses and current evidence of acute or chronic inflammatory or infective diseases and (vii) mental illness rendering the subjects unable to understand the nature, scope and possible consequences of the study. All the subjects gave written informed consent and the Institutional Review Boards of the participating hospitals approved the protocol.
Methods
Anthropometric variables and clinical blood pressure were measured. A complete metabolic profile, including fasting glucose and insulin levels and serum lipid concentrations, was performed whenever possible as previously described (16). DNA was extracted from lymphocytes by the salting-out method. Genotyping of the ACAA-insertion/deletion polymorphism at the 3' UTR of IGFIIR (GenBank sequence NT-007422, NCBI SNPs Data Base: rs8191962), which results in alleles sized 140 or 144 bp, were performed by PCR using fluorescent dye-labeled forward primers, followed by use of an ABI310 Automated Sequencer (Applied Biosystems, Foster City, CA, USA). Primer sequences and allele sizes were described previously (17). The accuracy of genotyping was evaluated by retesting one of each of the eight samples, selected in a blinded manner, on different days. The identity of the PCR product was verified by direct sequencing of ten samples, representing the three genotypes, amplified with non-fluorescent primers.
Statistical analysis
A priori power analysis of the differences in allele frequencies between type 2 diabetic patients and controls was performed by PASS2000 software (Number Cruncher Statistical Systems, Kaysville, UT, USA). For a 0.80 power with 0.05
, the
2 test with 1 degree of freedom allowed the detection of an effect size for the difference between frequencies (w) of 0.15 by including a sample size of 350 individuals. To take into account the case to control ratio, power analysis using the
2 test for two group proportions was also performed (18). Our previous data indicated a 55% prevalence of IGFIIR 140 carriers among healthy control women (9). Taking the latter data and for a 0.80 power with 0.05
, in order to detect a 0.15 difference in frequencies between type 2 diabetic patients and controls, and for a patient/control ratio of 1/1.5, sample sizes should be at least of 125 and 185 individuals respectively. Of note, effect sizes between 0.10 and 0.30 are considered small by convention (19). Results are expressed as mean ± S.D. unless otherwise stated. The KolmogorovSmirnov statistic was applied to continuous variables. Logarithmic transformation was applied as needed to ensure normal distribution of the variables. Unpaired t-tests were used to compare the central tendencies of the different groups. Analysis of covariance using the univariate general linear model (GLM) tool was performed to correct for the effect of age between patients and controls. To evaluate the association between discontinuous variables, we used the
2 test and Fishers exact test as appropriate. The deviation from HardyWeinberg equilibrium was also evaluated by the
2 test.
The influence of the different genotypes on clinical and biochemical variables was analyzed by unpaired t-tests. Multivariate logistic regression was performed to evaluate the impact of the IGFRII genotype and other risk factors on the presence of type 2 diabetes in our series. A backwards stepwise introduction of independent variables with a probability of F to enter <0.05 and a probability of F to remove >0.1 was employed. Analyses were performed using SPSS 10 for the Macintosh (SPSS Inc, Chicago, IL, USA). P<0.05 was considered statistically significant.
| Results |
|---|
|
|
|---|
|
|
The model retained carrier status for 140 bp alleles of the IGFIIR ACAA-insertion/deletion polymorphism (odds ratio 0.290 (95% confidence interval (CI) 0.1090.770)), age (odds ratio 1.117 (95% CI 1.0741.161)), BMI (odds ratio 1.141 (95% CI 1.0371.256)) triglycerides levels (odds ratio 1.015 (95% CI 1.0091.021)) and male gender (odds ratio 5.167 (95% CI 1.67415.956)) as statistically significant predictors of type 2 diabetes in our model (
2=133.024, P<0.001, Nagelkerkes R2=0.637), further suggesting a potential protective role of the IGFIIR 140 bp variant against the development of type 2 diabetes. Finally, we studied the influence of IGFIIR ACAA-insertion/deletion genotypes on the phenotype of the subjects studied here. Yet considering that the phenotype of diabetic patients was interfered by diabetes itself and by the treatment received by these individuals, these analyses were restricted to the non-diabetic controls. Controls carrying 140 bp alleles of the IGFIIR ACAA-insertion/deletion polymorphism presented with lower fasting insulin and serum triglycerides, and a near-significant tendency towards lower total serum cholesterol levels, compared with non-diabetic controls homozygous for wild-type alleles (Table 3
). These results further suggest a functional protective role of the IGFIIR ACAA-insertion/deletion variant against insulin resistance and associated metabolic traits in the non-diabetic population.
|
| Discussion |
|---|
|
|
|---|
Diabetic patients showed a decreased frequency of carriers of 140 bp alleles compared with the controls, suggesting a potential protective role of this particular variant against the development of type 2 diabetes. This particular finding was strongly supported by the results of the logistic regression model, in which carrying 140 bp alleles of the ACAA-insertion/deletion polymorphism at the 3' UTR of IGFIIR was associated with a marked decrease in the risk of having type 2 diabetes, as indicated by the low odds ratio found. Furthermore, the R2 of this analysis was considerably high (0.637) indicating that the independent variables included in the model explained a substantial proportion of the risk of having type 2 diabetes in the population studied here.
Moreover, non-diabetic carriers of 140 bp alleles of the ACAA-insertion/deletion polymorphism in IGFIIR presented with decreased fasting insulin and triglyceride levels which are well-known proxies for insulin resistance further suggesting that this variant may play a protective role against insulin resistance-related traits in the general population.
A limitation of our study is that non-diabetic controls were selected on the basis of normal fasting glucose concentrations, in accordance with the American - Diabetes Association criteria (15). However, even assuming that 1.2% of our controls would have been considered diabetic had an OGTT been performed (20), we actually found an association of the ACAA-insertion/deletion polymorphism at the 3' UTR of IGFIIR with type 2 diabetes, that might have been stronger if these putative cases of undiagnosed diabetes have been removed from the control population. It seems improbable also that the decreased fasting insulin and triglyceride levels found in non-diabetic carriers of this 140 bp alleles of the ACAA-insertion/deletion polymorphism in IGFIIR could be attributed to a hypothetical 1.2% of undiagnosed diabetics among them. It should be highlighted that our present findings may also explain the report, in non-diabetic Mexican American individuals of the San Antonio Family Diabetes Study, of linkage of insulin-resistant phenotypes including fasting insulin with the marker D6S264, located at 6q25.2-q26 (14). This marker is very close to IGFIIR, located at 6q26 (12), and therefore, our present results provide a convincing explanation for this previously reported evidence of linkage.
The IGFIIR gene encodes a multifunctional receptor that possesses binding sites for mannose-6-phosphate and IGF-II (21, 22). The mannose-6-phosphate binding site of this receptor is involved in 25 cell-mediated immune functions (11). Tumor suppression (10) and embryonic development (23) are other important functions attributable to the IGFIIR. In conceptual agreement, IGF-II has been shown to affect intrauterine growth (24) as well as pancreatic ß cell proliferation (6, 7). A recent study has suggested that mutations at the 3' UTR of IGFRII may induce defects in the post-transcriptional process during the synthesis of the IGF-II receptor (25). We hypothesize that the ACAA-insertion/deletion polymorphism at the 3' UTR of the IGFIIR might be involved in the pathogenesis of type 2 diabetes by its regulatory functions on fetal growth and/or ß cell proliferation (26). As opposed to the association of the ACAA-insertion/deletion polymorphism at the 3' UTR of the IGFIIR with type 2 diabetes and insulin-resistant traits described here by us, a previous study (27) showed that an IGFIIR polymorphism located at exon 16, and not the ACAA-insertion/deletion polymorphism at the 3' UTR, resulted in statistically significant transmission distortion in parent offspring trios of children with type 1 diabetes. Possibly, this polymorphism influences some of the immune functions mediated by the IGF-II receptor, which has also been identified as a potential target for human autoimmune responses (28).
In conclusion, the ACAA-insertion/deletion polymorphism at the 3' UTR of IGFIIR is associated with type 2 diabetes in the Spanish population, and influences surrogate markers of insulin resistance in non-diabetic subjects, providing an explanation for the evidence for linkage between insulin-resistant traits and the D6S264 marker at 6q25.2-q26 previously reported in MexicanAmericans.
| Acknowledgements |
|---|
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |