Search

Recent studies have suggested that prostaglandin-endoperoxide synthase-2 (PTGS2), also known as cyclo-oxygenase 2, plays an etiological role in the development of type 2 diabetes mellitus (T2DM). PTGS2 generates prostaglandins, which negatively modulate glucose-stimulated insulin secretion, and functions as a mediator of the inflammatory response, which is associated with decreased insulin sensitivity. Moreover, the gene encoding this enzyme, PTGS2, is located on 1q25.2, a region that has been linked with early onset T2DM in Pima Indians. To determine the possible role played by PTGS2 in modulating susceptibility to T2DM, we screened approximately 7.0 kb of the gene, corresponding to the promoter, coding sequence, and flanking exon-intron boundaries, and identified five variants, including three single nucleotide polymorphisms (SNPs) in the promoter, one intronic SNP, and one in the 3' untranslated region. With the exception of one rare promoter SNP (minor allele frequency <0.03), all SNPs were typed in approximately 1000 Pima Indians. The range of frequencies for the more common alleles was 0.65-0.88, and we found substantial linkage disequilibrium between all PTGS2 SNP pairs (D'>/=0.95). Variant alleles at two markers, rs20417 and rs2066826, which are located in the promoter and intron 6, respectively, were in strong linkage disequilibrium with each other (D'=0.97) and were associated with a higher prevalence of T2DM. For marker rs20417, individuals with the variant CC genotype had a 30% higher T2DM prevalence compared with subjects with the GG genotype (odds ratio=1.6 per copy of C allele; P=0.01). The variant C allele of rs20417 has been associated with decreased PTGS2 promoter activity, thereby suggesting a possible biological consequence attributable to this polymorphism. These findings indicate that genetic variants in PTGS2 may play a role in mediating susceptibility to T2DM in Pima Indians and are consistent with the hypothesis that chronic inflammation may contribute to the development of T2DM in some individuals.

Several diseases including type 2 diabetes mellitus (T2DM) are associated with abnormal O-glycosylation of proteins. beta-O-linked N-acetylglucosaminidase (O-GlcNAcase) encoded by MGEA5 on 10g24.1-q24.3 removes N-acetylglucosamine (O-GlcNAc), and we investigated this locus in Pima Indians who have the world's highest prevalence of T2DM. We detected two variants but there was no association with parameters of insulin resistance or diabetes in approximately 1300 Pimas. We conclude that mutations in MGEA5 are unlikely to contribute to T2DM in this population.

Selected candidate genes have been analyzed in the Pima Indians of Arizona based on evidence that insulin resistance and type 2 diabetes have significant genetic determinants. An amino acid substitution at codon 905 of the glycogen-targeting subunit of type 1 protein phosphatase that regulates skeletal muscle glycogenesis was recently reported to be associated with changes in insulin action in Danish subjects. In addition to the variant at 905, we report here a novel substitution at codon 883 and common variant of an "ATTTA" element in the 3'-untranslated region (UTR) of the corresponding gene (PPP1R3). The 3'-UTR variant resembled the mRNA-destabilizing AT(AU)-rich elements (AREs) and resulted in a 10-fold difference in reporter mRNA half-life, was correlated with PPP1R3 transcript and protein concentrations in vivo, and was associated with insulin resistance and type 2 diabetes in the Pimas. The variant is more common in Pimas (0.56) than in Caucasians (0.40). Because of its apparent effect on expression of PPP1R3, it may, in part, contribute to the higher prevalence of type 2 diabetes in this Native American population.

Skeletal muscle glycogen synthase (encoded by GYS1 on chromosome 19q13.3) is the rate-limiting enzyme in insulin-mediated non-oxidative glucose disposal. Our previous studies have demonstrated an impairment of insulin-stimulated GYS1 activities in insulin-resistant Pima Indians, and associations of non-insulin-dependent diabetes mellitus (NIDDM) with the GYS1 locus were reported recently in Finnish and Japanese populations. We have performed linkage and association analyses of GYS1 and seven additional DNA markers on 19q with NIDDM, and with parameters of insulin action in the Pima Indians. We have found a significant association of NIDDM with GYS1 genotypes (p = 0.009), and with common GYS1 alleles (p = 0.022) in the Pima Indians. We have performed a detailed comparative analysis of the GYS1 gene, mRNA, and protein product in insulin-sensitive and insulin-resistant Pima Indians. No mutations in GYS1 coding sequences were detected; nor did we find alterations of GYS1 mRNA expression or of its basal enzymatic activity in insulin-resistant Pima Indians. These results contrasted with a 25% reduction of immunoreactive protein in insulin-resistant subjects as detected by Western blotting with an antibody specific for the C-terminal end of GYS1 (t-test p = 0.024; Wilcoxon's rank-sum test, p = 0.04). Because no mutations were detected in the DNA encoding this epitope, the difference in immunoreactivity may reflect post-translational modification(s) of the protein rather than a difference in the gene itself, or it could have occurred by chance. We conclude that our data do not indicate alterations in the GYS1 gene as the cause for the observed association, and that a different locus near GYS1 may be the contributing genetic element.

To investigate the relationship of insulin receptor kinase with insulin resistance in humans, we studied insulin-sensitive tyrosine kinase activity in muscle biopsies taken from 20 Pima Indians [14 nondiabetics, 6 with non-insulin-dependent mellitus (NIDDM)] during euglycemic clamps, at insulin concentrations of approximately 68 microU/ml (low dose) and approximately 1,170 microU/ml (high dose). In the nondiabetics, the low dose, insulin-induced kinase activation in vivo was 1.5-fold the activity in the fasting state (P less than 0.05), whereas in the diabetics, the kinase activity actually decreased by 40% relative to fasting (P less than 0.05). The difference in delta-kinase in vivo was significant (P less than 0.01) between the two groups. Similarly, the kinase activation in vitro in response to 1 nM insulin was lower in diabetic subjects compared with nondiabetics (P less than 0.01). These data indicate that, in NIDDM, both in vitro and in vivo insulin-stimulated tyrosine kinase activity is impaired. Among nondiabetics, the kinase sensitivity to insulin, calculated as the ratio of the kinase activity at 1 nM insulin in vitro to the kinase activity at 100 nM insulin, was positively correlated with plasma insulin concentrations 2 h after an oral glucose load (r = 0.69, P less than 0.01). Thus, in nondiabetic subjects with insulin resistance, insulin activation of the kinase is not reduced, but the kinase sensitivity to insulin increases with increasing plasma insulin levels. Therefore, the site of insulin resistance in nondiabetic subjects is distal to the insulin receptor kinase. Furthermore, it is possible that circulating insulin, by increasing the kinase sensitivity to insulin, is a determinant of the receptor kinase activity.