Insulin resistant phenotype of polycystic ovary syndrome does not seem to be caused by variation in FTO.

Genetic variation in the FTO gene is associated with increased body weight and reduced insulin sensitivity. We investigated whether genetic variation in FTO is associated with polycystic ovary syndrome (PCOS), a condition also characterized by insulin resistance. Furthermore, we tested whether insulin resistance is specifically associated with genetic variation in FTO in women with PCOS. Sixty-two nondiabetic patients with PCOS defined by the Rotterdam criteria were compared to BMI and age-matched women. Each PCOS case was matched to 2 controls. All participants underwent an oral glucose tolerance test and were genotyped for the single nucleotide polymorphism rs8050136 in the FTO gene. There was no difference in the frequency of FTO genotypes between the PCOS and the non-PCOS groups. In non-PCOS participants, genetic variation in FTO is associated with insulin sensitivity (p=0.03). This association remained significant after adjustment for age and/or BMI (p<= 0.03). In subjects with PCOS, however, FTO did not associate with insulin sensitivity (p=0.67). Genetic variation in FTO does not have an impact on insulin sensitivity in women with PCOS and is therefore not involved in the pathogenesis of the insulin resistant phenotype seen in patients with PCOS.

Glucocorticoid receptor gene polymorphisms are associated with reduced first-phase glucose-stimulated insulin secretion and disposition index in women, but not in men.

AIM: Glucocorticoids are efficacious anti-inflammatory agents, but, in susceptible individuals, these drugs may induce glucose intolerance and diabetes by affecting ß-cell function and insulin sensitivity. We assessed whether polymorphisms in the glucocorticoid receptor gene NR3C1 associate with measures of ß-cell function and insulin sensitivity derived from hyperglycaemic clamps in subjects with normal or impaired glucose tolerance. METHODS: A cross-sectional cohort study was conducted in four academic medical centres in the Netherlands and Germany. Four hundred and forty-nine volunteers (188 men; 261 women) were recruited with normal glucose tolerance (n=261) and impaired glucose tolerance (n=188). From 2-h hyperglycaemic clamps, first- and second-phase glucose-stimulated insulin secretion, as well as insulin sensitivity index and disposition index, were calculated. All participants were genotyped for the functional NR3C1 polymorphisms N363S (rs6195), BclI (rs41423247), ER22/23EK (rs6189/6190), 9ß A/G (rs6198) and ThtIIII (rs10052957). Associations between these polymorphisms and ß-cell function parameters were assessed. RESULTS: In women, but not in men, the N363S polymorphism was associated with reduced disposition index (P=1.06 10(-4) ). Also only in women, the ER22/23EK polymorphism was associated with reduced first-phase glucose-stimulated insulin secretion (P=0.011) and disposition index (P=0.003). The other single-nucleotide polymorphisms were not associated with ß-cell function. Finally, none of the polymorphisms was related to insulin sensitivity. CONCLUSION: The N363S and ER22/23EK polymorphisms of the NR3C1 gene are negatively associated with parameters of ß-cell function in women, but not in men.

The secretion pattern of perivascular fat cells is different from that of subcutaneous and visceral fat cells.

AIMS/HYPOTHESIS: We have previously found that the mass of perivascular adipose tissue (PVAT) correlates negatively with insulin sensitivity and post-ischaemic increase in blood flow. To understand how PVAT communicates with vascular vessels, interactions between perivascular, subcutaneous and visceral fat cells with endothelial cells (ECs) were examined with regard to inflammatory, metabolic and angiogenic proteins. To test for possible in vivo relevance of these findings, circulating levels of the predominant secretion product, hepatocyte growth factor (HGF), was measured in individuals carefully phenotyped for fat distribution patterns. METHODS: Mono- and co-cultures of human primary fat cells with ECs were performed. mRNA expression and protein production were studied using Luminex, cytokine array, RealTime Ready and ELISA systems. Effects of HGF on vascular cells were determined by WST assays. In patients, HGF levels were measured by ELISA, and the mass of different fat compartments was determined by whole-body MRI. RESULTS: In contrast with other fat cell types, PVAT cells released higher amounts of angiogenic factors, e.g. HGF, acidic fibroblast growth factor, thrombospondin-1, serpin-E1, monocyte chemotactic protein-1 and insulin-like growth factor-binding protein -3. Cocultures showed different expression profiles from monocultures, and mature adipocytes differed from pre-adipocytes. HGF was preferentially released by PVAT cells and stimulated EC growth and smooth muscle cell cytokine release. Finally, in 95 patients, only PVAT, not visceral or subcutaneous mass, correlated independently with serum HGF levels (p = 0.03; r = 0.225). CONCLUSIONS: Perivascular (pre-)adipocytes differ substantially from other fat cells with regard to mRNA expression and protein production of angiogenic factors. This may contribute to fat tissue growth and atherosclerotic plaque complications. Higher levels of angiogenic factors, such as HGF, in patients with increased perivascular fat mass may have pathological relevance.

Visceral obesity modulates the impact of apolipoprotein C3 gene variants on liver fat content.

OBJECTIVE: It has not been solved whether subjects carrying the minor alleles of the -455T>C or -482C>T single nucleotide polymorphisms (SNPs) in the apolipoprotein-C3-gene (APOC3) have an increased risk for developing fatty liver and insulin resistance. We investigated the relationships of the SNPs with hepatic APOC3 expression and hypothesized that visceral obesity may modulate the effects of these SNPs on liver fat and insulin sensitivity (IS). METHODS: APOC3 mRNA expression and triglyceride content were determined in liver biopsies from 50 subjects. In a separate group (N=330) liver fat was measured by (1)H-magnetic resonance spectroscopy. IS was estimated during an oral glucose tolerance test (OGTT) and the euglycemic, hyperinsulinemic clamp (N=222). RESULTS: APOC3 mRNA correlated positively with triglyceride content in liver biopsies (r=0.29, P=0.036). Carriers of the minor alleles (-455C and -482T) tended to have higher hepatic APOC3 mRNA expression (1.80 (0.45-3.56) vs 0.77 (0.40-1.64), P=0.09), but not higher triglyceride content (P=0.76). In 330 subjects the genotype did not correlate with liver fat (P=0.97) or IS (OGTT: P=0.41; clamp: P=0.99). However, a significant interaction of the genotype with waist circumference in determining liver fat was detected (P=0.02) in which minor allele carriers had higher liver fat only in the lowest tertile of waist circumference (P=0.01). In agreement, during a 9-month lifestyle intervention the minor allele carriers of the SNP -482C>T in the lowest tertile also had less decrease in liver fat (P=0.04). CONCLUSIONS: APOC3 mRNA expression is increased in fatty liver and is regulated by SNPs in APOC3. The impact of the APOC3 SNPs on fatty liver is small and depends on visceral obesity.

Effects of a lifestyle intervention in metabolically benign and malign obesity.

AIMS/HYPOTHESIS: We and others recently characterised metabolically benign or healthy obesity (MHO). In the present study we investigated whether a lifestyle intervention is sufficient to place obese insulin-resistant (OIR) individuals in a position where the possible metabolic consequences are similar to those for MHO individuals. METHODS: A total of 262 non-diabetic individuals participated in a 9 month lifestyle intervention programme. Obese individuals (BMI ≥ 30.0 kg/m(2)) were stratified, based on their insulin sensitivity (IS) estimated from an OGTT, into MHO (IS in the upper quartile, n = 26) and OIR (IS in the lower three quartiles, n = 77). Total body and visceral fat were measured by magnetic resonance (MR) tomography and liver fat by (1)H-MR spectroscopy. RESULTS: During the intervention, visceral fat decreased significantly in both groups (both p ≤ 0.009), whereas total body and liver fat decreased only in the OIR group (p < 0.0001; MHO p = 0.12 for total body fat and p = 0.47 for liver fat). IS improved in the OIR group (p < 0.0001), but remained essentially unchanged in the MHO group (p = 0.30). However, despite the significant increase in the OIR group, IS at follow-up barely exceeded 50% of the IS of the MHO group (OIR 9.30 ± 0.53 arbitrary units [AU]; MHO 16.41 ± 1.05 AU; p < 0.0001). CONCLUSIONS/INTERPRETATION: IS improves during the lifestyle intervention in OIR individuals. However, it does not reach a level where adequate protection from type 2 diabetes and cardiovascular disease is expected. Thus, stratification of obese individuals based on their metabolic phenotype is important to identify those who are likely to need early pharmacological treatment in addition to the lifestyle intervention.

A polymorphism in the gene encoding AdipoR1 affects olfactory recognition.

Polymorphisms in the gene encoding adiponectin receptor 1 (AdipoR1) are associated with insulin resistance, fatty liver, increased risk for type 2 diabetes and cardiovascular disease. AdipoR1 is expressed in the central nervous system and in the olfactory mucosa of mice and humans. We therefore hypothesized that a common polymorphism in AdipoR1 might alter olfactory function. We investigated a group of 222 healthy subjects (male: n = 147, female: n = 75) for olfactory recognition, and genotyped them for the polymorphism rs6666089 in the human AdipoR1 gene. This polymorphism has been previously shown to be associated with insulin resistance. Olfactory recognition was tested using standardized sniffing sticks, and parameters of glucose metabolism and serum adiponectin levels were assessed. We found a significant olfactory impairment in carriers of the AdipoR1 polymorphism rs6666089 (olfactory recognition: GG: 89.4 ± 1.2%, GA: 86.9 ± 1.4%, AA: 77.2 ± 4.8%, additive model, P = 0.0004, adjusted for age). Adiponectin levels had no impact on olfactory recognition. Fasting plasma glucose, fasting plasma insulin, body mass index and HbA1c did not differ between the genotype groups. In conclusion, the presence of a genetic variation in AdipoR1 is associated with decreased olfactory recognition in healthy subjects. Adiponectin signalling may have an important role in olfactory function and regulation of appetite.

Genetic variants affecting incretin sensitivity and incretin secretion.

Recent genome-wide association studies identified several novel risk genes for type 2 diabetes. The majority of these type 2 diabetes risk variants confer impaired pancreatic beta cell function. Though the molecular mechanisms by which common genetic variation within these loci affects beta cell function are not completely understood, risk variants may alter glucose-stimulated insulin secretion, proinsulin conversion, and incretin signals. In humans, the incretin effect is mediated by the secretion and insulinotropic action of two peptide hormones, glucose-dependent insulinotropic polypeptide and glucagon-like peptide-1. This review article aims to give an overview of the type 2 diabetes risk loci that were found to associate with incretin secretion or incretin action, paying special attention to the potential underlying mechanisms.

The ENPP1 K121Q polymorphism determines individual susceptibility to the insulin-sensitising effect of lifestyle intervention.

AIMS/HYPOTHESIS: The K121Q (rs1044498) single nucleotide polymorphism (SNP) in the ENPP1 gene has shown association with insulin resistance and type 2 diabetes in various ethnic populations. We hypothesised that K121Q may predict the success of lifestyle intervention in terms of improvement of insulin sensitivity. METHODS: We genotyped 1,563 participants with an increased risk of type 2 diabetes for K121Q and performed correlational analyses with anthropometric data and variables of insulin sensitivity. For metabolic characterisation, all participants underwent an OGTT. A subgroup of 506 participants additionally underwent a euglycaemic-hyperinsulinaemic clamp. In 342 participants, metabolic traits and anthropometric data were re-evaluated after a 9 month lifestyle intervention. RESULTS: In the overall cohort, K121Q was not associated with measures of obesity, indices of glucose tolerance during OGTT and insulin sensitivity estimated from the OGTT or derived from a euglycaemic-hyperinsulinaemic clamp after appropriate adjustment. However, K121Q did significantly influence the change in insulin sensitivity during lifestyle intervention after appropriate adjustment (p (additive) = 0.0067, p (dominant) = 0.0027). Carriers of the minor allele had an impaired increase in OGTT-derived insulin sensitivity. A similar trend was obtained for clamp-derived insulin sensitivity, but did not reach significance. CONCLUSIONS/INTERPRETATION: In our population of European ancestry, the ENPP1 SNP K121Q influenced the change in insulin sensitivity during lifestyle intervention. Thus, this SNP may determine susceptibility to environmental changes and could predict the success of lifestyle intervention.

AHSG gene variation is not associated with regional body fat distribution--a magnetic resonance study.

Obesity-resistance in AHSG-knockout mice indicate an important role of alpha2-Heremans-Schmid glycoprotein/fetuin-A (AHSG) in the development of obesity. We studied whether genetic variation within AHSG affects whole-body adiposity and regional fat distribution in humans. We genotyped 321 subjects at increased risk for type 2 diabetes for five single nucleotide polymorphisms (SNP) rs2248690, rs4831, rs2070635, rs4917, and rs1071592. Body fat distribution and ectopic hepatic and intramyocellular lipids were assessed by magnetic resonance techniques. AHSG levels were determined by immunoturbidimetry. The five chosen SNPs covered 100% of common genetic variation (minor allele frequency >/=0.05) within AHSG (r (2)>/=0.8). All SNPs were significantly associated with AHSG levels (p<0.0001), except for rs4831 (p=0.9) after adjustment for gender, age, and body mass index (BMI). AHSG levels were associated with liver fat content (p=0.0160) and BMI (p=0.0247) after adjustment for gender and age. While rs2248690 was nominally associated with BMI in the dominant model (p=0.0432), none of the SNPs was associated with regional fat distribution. Common genetic variation within AHSG does not appear to influence regional body fat distribution, but may affect whole-body adiposity in humans.

Protein kinase C iota mediates lipid-induced apoptosis of human coronary artery endothelial cells.

Apoptosis is involved in the development and progression of atherosclerotic lesions. Protein kinase C (PKC) signalling is of importance in atherosclerosis as well as apoptosis. Therefore, we tested the involvement of PKC in lipid-induced apoptosis of human coronary artery endothelial cells (HCAEC). Protein expression of PKC isoforms alpha, beta I, delta, epsilon, and iota was detected, whereas no relevant protein amounts of PKC isoforms beta II, gamma, eta, theta, and zeta were found. Inhibition of classical and novel PKC isoforms by treatment with bisindolylmaleimide or PKC down-regulation by long-term treatment with 12-O-tetradecanoyl phorbol-13-acetate (TPA) could not prevent apoptosis induced by palmitate or stearate. In contrast, a specific myristoylated, cell-permeable PKC zeta/iota pseudosubstrate prevented lipid-induced apoptosis in HCAEC. Furthermore, saturated fatty acids activated PKC iota as evidenced by PKC iota down-regulation upon long-term treatment with stearate. Our data provide evidence that PKC iota is activated by saturated fatty acids and mediates lipid-induced apoptosis of HCAEC.

A common genetic variant in WFS1 determines impaired glucagon-like peptide-1-induced insulin secretion.

AIMS/HYPOTHESIS: WFS1 type 2 diabetes risk variants appear to be associated with impaired beta cell function, although it is unclear whether insulin secretion is affected directly or secondarily via alteration of insulin sensitivity. We aimed to investigate the effect of a common WFS1 single-nucleotide polymorphism on several aspects of insulin secretion. METHODS: A total of 1,578 non-diabetic individuals (534 men and 1,044 women, aged 40 +/- 13 years, BMI 28.9 +/- 8.2 kg/m(2) [mean +/- SD]) at increased risk of type 2 diabetes were genotyped for rs10010131 within the WFS1 gene. All participants underwent an OGTT (and a subset additionally an IVGTT [n = 319]) and a hyperglycaemic clamp combined with glucagon-like peptide-1 (GLP-1) and arginine stimuli (n = 102). RESULTS: rs10010131 was associated with reduced OGTT-derived insulin secretion (p = 0.03). In contrast, insulin secretion induced by an i.v. glucose challenge in the IVGTT and hyperglycaemic clamp was not different between the genotypes. GLP-1 infusion combined with a hyperglycaemic clamp showed a significant reduction of the insulin secretion rate during the first and second phases of GLP-1-induced insulin secretion in carriers of the risk allele (reduction of 36% and 26%, respectively; p = 0.007 and p = 0.04, respectively). CONCLUSIONS/INTERPRETATION: A common genetic variant in WFS1 specifically impairs GLP-1-induced insulin secretion independently of insulin sensitivity. This defect might explain the impaired insulin secretion in carriers of the risk allele and confer the increased risk of type 2 diabetes.

The risk allele load accelerates the age-dependent decline in beta cell function.

AIMS/HYPOTHESIS: Among the novel type 2 diabetes risk loci identified by genome-wide association studies, TCF7L2, HHEX, SLC30A8 and CDKAL1 appear to affect beta cell function. In the present study we examined the effect of these genes' risk alleles on the age-dependent decline in insulin secretion. METHODS: The SNPs rs7903146 (TCF7L2), rs7754840(CDKAL1), rs7923837 (HHEX) and rs13266634 (SLC30A8) were genotyped in 1,412 non-diabetic patients, who were subsequently grouped according to their number of risk alleles. All participants underwent an OGTT. Insulin secretion was assessed by validated indices and proinsulin conversion by calculating AUC(proinsulin)/AUC(insulin). RESULTS: The number of risk alleles revealed a Gaussian distribution, with most participants carrying four risk alleles. Stratification into groups with low (LAL, up to three alleles), median (MAL, four alleles) and high (HAL, five to eight alleles) allele load resulted in MAL and HAL participants displaying significantly lower insulin secretion and proinsulin conversion than LAL participants (p

Variation in the FTO gene influences food intake but not energy expenditure.

Polymorphisms in the FTO (fat mass- and obesity-associated) gene are associated with obesity. The mechanisms how genetic variation in this gene influences body weight are unknown. Body weight is determined by energy intake/storage and energy expenditure. In this study, we investigated whether genetic variation in FTO influences energy expenditure or food intake in carefully phenotyped subjects. In 380 German subjects, insulin sensitivity was measured by a hyperinsulinemic euglycemic clamp. Lean body mass and body fat were quantified using the bioimpedance method. Indirect calorimetry was used to estimate the metabolic rate. Food intake was assessed using food diaries (mean 11+/-1 d) in 151 subjects participating in a lifestyle intervention program to prevent diabetes. All subjects were genotyped for the FTO single nucleotide polymorphism (SNP) rs8050136. The risk allele of SNP rs8050136 was associated with higher body fat-related parameters (all p< or =0.04, additive inheritance model). Energy expenditure was not affected by the SNP. However, the risk allele of rs8050136 was significantly associated with higher energy intake (p=0.01, dominant inheritance model) during dietary restriction. Our data suggest that the increased body weight in carriers of the risk allele of FTO SNP rs8050136 is a consequence of increased food intake, but not of impaired energy expenditure.

Selective association of plasma non-esterified fatty acid species with circulating interleukin-8 concentrations in humans.

Elevated plasma concentrations of non-esterified fatty acids (NEFA), due to high fat intake and/or adipose tissue lipolysis, are a hallmark of the metabolic syndrome. We assessed whether certain plasma NEFA species contribute to the chronic low-grade inflammatory state seen in the metabolic syndrome. We determined the fasting plasma NEFA patterns of 75 overweight non-diabetic subjects and analysed their relationship with plasma inflammatory parameters. After adjustment for gender, age, body fat, and waist-hip ratio, no strong correlations of single NEFA species with leukocyte number, C-reactive protein, interleukin-6, tumour necrosis factor-alpha, or monocyte chemoattractant protein-1 were detected. However, oleate was negatively (r=-0.36, p=0.0015) and myristate (r=0.41, p=0.0003) as well as the omega3-polyunsaturated NEFA alpha-linolenate (r=0.37, p=0.0011), eicosapentaenoate (r=0.40, p=0.0003), and docosahexaenoate (r=0.40, p=0.0004) were positively associated with interleukin-8 concentrations. The other NEFA species as well as the total plasma NEFA concentration did not correlate with interleukin-8. The correlations of myristate, oleate, and the sum of all omega3-polyunsaturated NEFA with interleukin-8 were independent of plasma tumour necrosis factor-alpha and overall adiposity. Our data demonstrate close and selective associations of oleate, myristate, and omega3-polyunsaturated NEFA with plasma concentrations of the pro-inflammatory chemokine interleukin-8. Thus, these NEFA species may represent specific determinants of plasma interleukin-8.

Genetic variability of procolipase associates with altered insulin secretion in non-diabetic Caucasians.

AIMS: Procolipase (CLPS) is secreted from the exocrine pancreas into the gastrointestinal tract and becomes proteolytically cleaved into colipase and the pentapeptide enterostatin. While colipase is an indispensable cofactor for pancreatic lipase, enterostatin acts as a hormone that inhibits insulin secretion and confers satiety signals to the brain, thereby restricting further food intake in animal models. As both high fat diet and obesity contribute to insulin resistance, we investigated whether genetic variability of CLPS associates with metabolic traits in non-diabetic humans at diabetes risk. METHODS: Tagging single nucleotide polymorphisms (SNPs) in the human CLPS locus on chr6p21.1 were selected using HapMap data. 498 humans, phenotyped for different glucose and lipid metabolic traits, were genotyped by bidirectional sequencing and multivariate linear regression analyses were undertaken. RESULTS: 2 tagging SNPs (rs3748050 in the Kozak sequence: A/G and rs3748051 in intron 1: A/G), covering 100% of CLPS variability including 8 kb of its promoter, were genotyped for association analyses. The minor alleles of both tagging SNPs associated significantly with a reduced insulin secretion (-20.2%, both SNPs) in various estimation models derived from the oral glucose tolerance test (OGTT; rs3748050/51: 30 min C-peptide levels: p=0.001/0.01, insulinogenic index: p=0.02/0.02, AUC C-peptide/AUC glucose: p=0.01/0.01) after adjustment for relevant covariates. No significant associations with fasting total cholesterol (c), HDL-c, LDL-c, triglycerides and free fatty acids were found (all p > 0.11). CONCLUSIONS: CLPS genetic variability associates with insulin secretory function in non-diabetic humans and may represent a novel candidate gene for development of type 2 diabetes.

Variants of CDKAL1 and IGF2BP2 affect first-phase insulin secretion during hyperglycaemic clamps.

AIMS/HYPOTHESIS: Genome-wide association studies have recently identified novel type 2 diabetes susceptibility gene regions. We assessed the effects of six of these regions on insulin secretion as determined by a hyperglycaemic clamp. METHODS: Variants of the HHEX/IDE, CDKAL1, SLC30A8, IGF2BP2 and CDKN2A/CDKN2B genes were genotyped in a cohort of 146 participants with NGT and 126 with IGT from the Netherlands and Germany, who all underwent a hyperglycaemic clamp at 10 mmol/l glucose. RESULTS: Variants of CDKAL1 and IGF2BP2 were associated with reductions in first-phase insulin secretion (34% and 28%, respectively). The disposition index was also significantly reduced. For gene regions near HHEX/IDE, SLC30A8 and CDKN2A/CDKN2B we did not find significant associations with first-phase insulin secretion (7-18% difference between genotypes; all p > 0.3). None of the variants showed a significant effect on second-phase insulin secretion in our cohorts (2-8% difference between genotypes, all p > 0.3). Furthermore, the gene variants were not associated with the insulin sensitivity index. CONCLUSIONS: Variants of CDKAL1 and IGF2BP2 attenuate the first phase of glucose-stimulated insulin secretion but show no effect on the second phase of insulin secretion. Our results, based on hyperglycaemic clamps, provide further insight into the pathogenic mechanism behind the association of these gene variants with type 2 diabetes.

Polymorphisms in the TCF7L2, CDKAL1 and SLC30A8 genes are associated with impaired proinsulin conversion.

AIMS/HYPOTHESIS: Variation within six novel genetic loci has been reported to confer risk of type 2 diabetes and may be associated with beta cell dysfunction. We investigated whether these polymorphisms are also associated with impaired proinsulin to insulin conversion. METHODS: We genotyped 1,065 German participants for single nucleotide polymorphisms rs7903146 in TCF7L2, rs7754840 in CDKAL1, rs7923837 and rs1111875 in HHEX, rs13266634 in SLC30A8, rs10811661 in CDKN2A/B and rs4402960 in IGF2BP2. All participants underwent an OGTT. Insulin, proinsulin and C-peptide concentrations were measured at 0, 30, 60, 90 and 120 min during the OGTT. Insulin secretion was estimated from C-peptide or insulin levels during the OGTT using validated indices. We used the ratio proinsulin/insulin during the OGTT as indicator of proinsulin conversion. RESULTS: In our cohort, we confirmed the significant association of variants in TCF7L2, CDKAL1 and HHEX with reduced insulin secretion during the OGTT (p<0.05 for all). Variation in SLC30A8, CDKN2A/B and IGF2BP2 was not associated with insulin secretion. The risk alleles of the variants in TCF7L2, CDKAL1 and SLC30A8 reduced proinsulin to insulin conversion (p<0.05 for all), whereas the risk alleles in HHEX, CDKN2A/B and IGF2BP2 were not associated with reduced proinsulin to insulin conversion (p>0.6). CONCLUSIONS/INTERPRETATION: Diabetes-associated variants in TCF7L2 and CDKAL1 impair insulin secretion and conversion of proinsulin to insulin. However, both aspects of beta cell function are not necessarily linked, as impaired insulin secretion is specifically present in variants of HHEX and impaired proinsulin conversion is specifically present in a variant of SLC30A8.

Endothelial NO-synthase intron 4 polymorphism is associated with disturbed in vivo nitric oxide production in individuals prone to type 2 diabetes.

Insulin resistance, as well as vascular disease, both share a relevant genetic background taking the influence of a positive family history of these disorders. On the other hand, insulin resistance is associated with a proatherosclerotic disturbance in nitric oxide dependent vasodilation, probably contributing to the link between these two disorders. We examined the association between nitric oxide dependent vasodilation (measured with high resolution ultrasound at 13 MHz) and three relevant NO-synthase (eNOS)-polymorphisms in 200 insulin resistant subjects participating in the Tuebinger Lifestyle Intervention Program (TULIP). This study revealed that carriers of the eNOS intron 4 polymorphism (aa 2.16%; ab 24.2%; bb 73.2%) show significantly worse endothelial, and thereby eNOS dependent vasodilation (p=0.03, multivariate ANOVA), as compared to wildtype carriers. The 5' UTR T-786C and the G894 T polymorphism did not show any influence on eNOS-activity. In subjects at increased risk to develop type 2 diabetes, the eNOS intron 4 polymorphism is independently associated with endothelial function as indicated by disturbed endothelial NO production. Due to the high prevalence and the relatively strong effect, this polymorphism might help to identify subjects at increased risk for atherosclerosis associated with overweight and insulin resistance.