Studies of association between the gene for calpain-10 and type 2 diabetes mellitus in the United Kingdom.

Variation in CAPN10, the gene encoding the ubiquitously expressed cysteine protease calpain-10, has been associated with type 2 diabetes in Mexican Americans and in two northern-European populations, from Finland and Germany. We have studied CAPN10 in white subjects of British/Irish ancestry, using both family-based and case-control studies. In 743 sib pairs, there was no evidence of linkage at the CAPN10 locus, which thereby excluded it as a diabetes-susceptibility gene, with an overall sib recurrence risk, lambda(S), of 1.25. We examined four single-nucleotide polymorphisms (SNP-44, -43, -19, and -63) previously either associated with type 2 diabetes or implicated in transcriptional regulation of calpain-10 expression. We did not find any association between SNP-43, -19, and -63, either individually or as part of the previously described risk haplotypes. We did, however, observe significantly increased (P=.033) transmission of the less common C allele at SNP-44, to affected offspring in parents-offspring trios (odds ratio 1.6). An independent U.K. case-control study and a small discordant-sib study did not show significant association individually. In a combined analysis of all U.K. studies (P=.015) and in combination with a Mexican American study (P=.004), the C allele at SNP-44 is associated with type 2 diabetes. Sequencing of the coding region of CAPN10 in a group of U.K. subjects revealed four coding polymorphisms-L34V, T504A, R555C, and V666I. The T504A polymorphism was in perfect linkage disequilibrium with the diabetes-associated C allele at SNP-44, suggesting that the synthesis of a mutant protein and/or altered transcriptional regulation could contribute to diabetes risk. In conclusion, we were not able to replicate the association of the specific calpain-10 alleles identified by Horikawa et al. but suggest that other alleles at this locus may increase type 2 diabetes risk in the U.K. population.

Analysis of parent-offspring trios provides evidence for linkage and association between the insulin gene and type 2 diabetes mediated exclusively through paternally transmitted class III variable number tandem repeat alleles.

Variation at the variable number tandem repeat (VNTR) minisatellite 5' of the insulin gene (INS) is associated with several phenotypes, including type 1 diabetes, polycystic ovary syndrome, and birth weight. Case-control studies have suggested that class III VNTR alleles are also associated with type 2 diabetes, but results have been inconsistent and may reflect population stratification. To explore further the role of the INS-VNTR in type 2 diabetes susceptibility, we used family-based association methods in 155 parent-offspring trios from the British Diabetic Association-Warren Trios repository, each ascertained via a Europid proband with type 2 diabetes. Overall, there was no significant association between diabetes and the INS-VNTR genotype, with 65 of 119 heterozygous parents (55%) transmitting class III and 54 class I (P = 0.16, one-sided). However, whereas maternal transmissions followed Mendelian expectation, there was a marked excess of class III transmission from the 49 heterozygous fathers (34 [69%] vs. 15, P = 0.003 vs. 50% expectation, P = 0.003 vs. maternal transmission). These results confirm that variation within the TH-INS-IGF2 locus, most plausibly at the VNTR itself, influences type 2 diabetes susceptibility. By demonstrating that this effect is mediated exclusively by the paternally derived allele, these findings implicate imprinted genes in the pathogenesis of type 2 diabetes.

Evidence that single nucleotide polymorphism in the uncoupling protein 3 (UCP3) gene influences fat distribution in women of European and Asian origin.

AIMS/HYPOTHESIS: Uncoupling proteins are mitochondrial transmembrane carriers implicated in the regulation of energy balance. Dysfunction of UCP3 (the predominant uncoupling protein in skeletal muscle) might therefore be expected to reduce thermogenic capacity, alter energy homeostasis and influence predisposition to obesity and Type II (non-insulin-dependent) diabetes mellitus. A variant in the putative promoter region of UCP3 (-55 c-->t) has recently been identified, and an association with obesity reported in French subjects. Our aim was to study the pathophysiological role of this variant in diabetes-related and obesity-related traits using two distinct ethnic populations. METHODS: The -55 c-->t variant was genotyped in 85 South Indian and 150 European parent-offspring trios ascertained through Type II diabetic probands and in 455 South Indian subjects initially recruited to an urban survey into the prevalence of diabetes. RESULTS: In South Indian and European parent-offspring trios there was no preferential transmission of either allele at the -55 c-->t polymorphism to diabetic offspring (South Indians, p = 0.60; Europeans, p = 0.15). When family members were analysed for intermediate traits, the t-allele was associated with increased waist-to-hip ratio but only in females (South Indian mothers p = 0.036, daughters p = 0.032: European mothers p = 0.037, daughters p = 0.14). These findings were replicated in South Indian females from the population-based survey (p = 0.039). CONCLUSION/INTERPRETATION: The consistent association between the t-allele at this locus and increased waist-to-hip ratio in women from three separate data sets indicates that variation at this polymorphism (or another locus with which it is in linkage disequilibrium) influences fat distribution but that this effect is restricted to females.

An uncoupling protein 2 gene variant is associated with a raised body mass index but not Type II diabetes.

AIMS/HYPOTHESIS: Linkage between markers close to the uncoupling protein 2 and 3 genes (11q13) and resting metabolic rate and a pre-diabetic phenotype have been found. The syntenic region in mouse has been found to be linked to quantitative traits associated with obesity and diabetes. UCP2 and UCP3 could therefore have an important role in body weight regulation and susceptibility to diabetes. We investigated a recently identified variant of the UCP2 gene in exon 8 as a marker for glucose and weight homeostasis. METHODS: Length variation of the UCP2 exon 8 variant was studied by the polymerase chain reaction and agarose gel electrophoresis. Sequence variants of the UCP3 gene were sought by semi-automated DNA sequencing. RESULTS: In 453 South Indian subjects, we found an association in women between the UCP2 exon variant and body mass index (p = 0.018). These findings were replicated in a separate group of South Indian subjects (n = 143, p < 0.001) irrespective of sex. Although no association was found between the UCP2 exon 8 variant and overt obesity in British subjects, the UCP2 genotype of obese women (n = 83) correlated with fasting serum leptin concentration (p = 0.006) in the presence of extreme obesity. These observations could not be explained by tight linkage disequilibrium with a coding region variant in the region of the UCP3 gene of biological significance. Lastly, no association was found between UCP2 and Type II (non-insulin-dependent) diabetes using either a family based design (85 families) or case control study (normal glucose tolerance n = 335, impaired glucose tolerance n = 42, Type II diabetes n = 76). CONCLUSION/INTERPRETATION: We have described a UCP2 gene exon 8 variant that may affect susceptibility to weight gain by influencing regulation of leptin.

Mutation Screening Using PCR-SSCP : Silver Staining and Isotopic Protocols.

Screening for mutations prior to sequencing can reduce the time and costs of identifying mutations. When the DNA sequence is known, the technique of detecting mutations as single-stranded conformational polymorphisms (SSCP) is a convenient method of screening for possible mutations. SSCP was originally developed by Orita et al. (1). It has the ability of detecting a single base change, and has been applied to a number of genes, including the insulin receptor (2), GLUT 4 (3), glucokinase (4), and the mitochondrial genome (5).

UKPDS 21: low prevalence of the mitochondrial transfer RNA gene (tRNA(Leu(UUR))) mutation at position 3243bp in UK Caucasian type 2 diabetic patients.

Some patients with Type 2 (non-insulin-dependent) diabetes mellitus possess a mitochondrial mutation in the tRNA(Leu(UUR)) gene at position 3243 bp. These subjects show a maternal mode of inheritance and often have hearing defects. In French and Japanese populations, this mutation may be present in 1-3% of subjects with a family history of diabetes. We assessed the prevalence of this mutation in newly diagnosed diabetic subjects in the UK white Caucasian population. The 3243 bp mutation was not detected in 500 randomly selected Type 2 diabetic subjects, 50 gestational diabetic subjects, and members of a MODY pedigree. Two of 748 (0.27%) Type 2 diabetic subjects with a family history of diabetes were found to possess the mutation. These subjects had an early age of diagnosis (M 38 years; F 36 years) and were non-obese. The male patient showed evidence of markedly impaired beta-cell function and deafness, while the female was not deaf, had approximately 50% of normal pancreatic function and responded well to diet. The mutation in the tRNA(Leu(UUR)) gene probably occurs in only approximately 0.1-0.2% of white Caucasian Type 2 diabetic patients in the UK.

High prevalence of a missense mutation of the glucokinase gene in gestational diabetic patients due to a founder-effect in a local population.

A high proportion of the female patients who are members of maturity onset diabetes of the young (MODY) pedigrees, and whose diabetes mellitus is due to a glucokinase mutation, originally presented with gestational diabetes. To establish whether glucokinase mutations could be a common cause of gestational diabetes, we studied 50 subjects who presented with gestational diabetes and on follow-up had hyperglycaemia (5.5-10.0 mmol/l). Screening for glucokinase mutations using single-stranded conformational polymorphism (SSCP) analysis detected a missense mutation at position 299 (Gly299-->Arg) in three subjects. As two pedigrees in the Oxford area had the same glucokinase mutation, we suspected the role of a founder-effect, and carried out pedigree extension, haplotype construction (using microsatellite markers GCK1 and GCK2) and mutation screening of at-risk subjects from the same geographical area. One of the gestational diabetic subjects was found to be related to one of the previous pedigrees via her paternal grandmother. Subjects with the mutation were found to have the Z + 4/2 (GCK1/ GCK2) haplotype, suggesting that the observed high prevalence of the Gly299-->Arg glucokinase mutation in the Oxford region was due to a founder-effect. Since glucokinase mutations predominantly induce subclinical hyperglycaemia, it is likely that in the locality of other pedigrees there will be undiagnosed subjects with the same glucokinase mutation, which remains undetected unless pregnancy occurs.

Mitochondrial FAD-glycerophosphate dehydrogenase and G-protein-coupled inwardly rectifying K+ channel: No evidence for linkage in maturity-onset diabetes of the young or NIDDM.

Two genes that have potentially important regulatory roles in insulin secretion are both located on chromosome 2q24.1. G-protein-coupled muscarinic potassium channel (GIRK1) is an inwardly rectifying K+ channel that helps to maintain the resting potential and excitability of cells. Mitochondrial FAD-linked glycerophosphate dehydrogenase (m-GDH) catalyzes a rate-limiting step of the glycerol phosphate shuttle in pancreatic islets. Reduced m-GDH activity has been demonstrated in islets isolated from diabetic subjects compared with islets from nondiabetic control subjects and from the diabetic GK rat. To study the relationship between these candidate genes and NIDDM, we have examined a simple tandem-repeat polymorphism (STRP) close to both the KCN J3 (GIRK1) locus and the m-GDH locus. In a linkage study of three maturity-onset diabetes of the young (MODY) pedigrees, not linked to MODY1, MODY2, or MODY3, a cumulative score of - 9.6 at a recombination fraction of theta = 0 excluded linkage. In a population-association study, no linkage disequilibrium for the STRP was found between 190 unselected NIDDM patients and 60 geographically and age-matched white nondiabetic subjects (chi2 = 1.51 on 3 df, P = 0.68). Thus, mutations involving the genes for GIRK1 or FAD-glycerophosphate dehydrogenase are unlikely to cause MODY, and a common mutation in either gene is unlikely to contribute to NIDDM in whites. These data do not exclude mutations in some families or other ethnic groups.

Mutation of the glucagon receptor gene and diabetes mellitus in the UK: association or founder effect?

Recent evidence suggests that a mutation of the glucagon receptor (GCG-R) gene is involved in the development of type 2 diabetes in French patients. We have examined patients from three geographically distinct regions in the UK and found the GGT40 (Gly) to AGT40 (Ser) mutation to be present in 15/691 (2.2%) of patients with type 2 (non-insulin dependent) diabetes and 1/425 (0.2%) of geographically matched controls and have therefore replicated association of the GCG-R mutation with classical type 2 diabetes (Fisher's exact test p = 0.008). An increased frequency of the mutation of the GCG-R gene was also found in probands of type 1 (insulin dependent) diabetic multiplex (affected sib pair) families, (10/404, 2.5%). However, a lack of preferential transmission from parents heterozygous for the mutation, to affected type 1 diabetic sibs may suggest population stratification. This in turn cannot be excluded as an alternative explanation for the difference in frequency of the GCG-R gene mutation between subjects with type 2 diabetes and normal controls.

Candidate gene studies in pedigrees with maturity-onset diabetes of the young not linked with glucokinase.

Maturity-onset diabetes of the young (MODY) is a form of non-insulin-dependent diabetes mellitus characterised by an early age of onset and an autosomal dominant mode of inheritance. Only a proportion of cases are due to mutations in the glucokinase gene. We have studied five Caucasian MODY families, including the first MODY family to be described, with five candidate genes implicated in regulation of insulin secretion. The affected subjects showed more marked hyperglycaemia than that found in subjects with glucokinase mutations. We assessed polymorphic markers close to the genes for glucokinase, hexokinase II, adenosine deaminase, pituitary adenylate cyclase-activating polypeptide receptor, and glucagon-like peptide-1 receptor. Linkage analysis with diabetes gave cumulative log of the odds (LOD) scores of less than -3, implying that mutations in these genes are unlikely to provide a major genetic contribution to this form of MODY.

Non-linkage of the glucagon-like peptide 1 receptor gene with maturity onset diabetes of the young.

Glucagon-like peptide-1 (GLP-1) is a hormone derived from the preproglucagon molecule that is secreted by intestinal L cells and stimulates insulin secretion from beta cells. The GLP-1 receptor is a candidate gene for diabetes mellitus, as mutations may induce the impaired insulin response that is a characteristic feature of NIDDM. To study the relationship between the GLP-1 receptor gene and NIDDM, linkage of a microsatellite polymorphism flanking the GLP-1 receptor gene with diabetes was investigated in three Caucasian families with MODY and in the nuclear families of 12 NIDDM probands. A cumulative LOD score -8.50 excludes linkage in these MODY pedigrees. A LOD score of -1.24 in the NIDDM nuclear pedigrees makes linkage improbable. Mutations in or near the GLP-1 receptor gene are unlikely to be the major cause of the inherited predisposition to NIDDM in Caucasian pedigrees, but we cannot exclude a role for this locus in a polygenic model or a major role in some pedigrees.

Microsatellite polymorphisms at the glucokinase locus: a population association study in Caucasian type 2 diabetic subjects.

Glucokinase has a central role in glucose metabolism in pancreatic beta cells and hepatocytes and is an important candidate gene for Type 2 diabetes. Mutations of the glucokinase gene have been reported in Caucasian pedigrees with maturity-onset diabetes of the young and late-onset Type 2 diabetes. In population studies of American Blacks and Mauritian Creoles an association between alleles of a glucokinase polymorphism and Type 2 diabetes has been described. Two microsatellite polymorphisms (GCK 1 and GCK 2) flanking the glucokinase gene were investigated in Caucasian subjects. There was no significant linkage disequilibrium between the alleles of the two polymorphisms. The overall allelic frequencies for GCK 1 and the combined haplotyes did not significantly differ between 95 Type 2 diabetic and 76 normoglycaemic subjects. In an expanded cohort of 151 diabetic subjects the allelic frequencies at GCK 2 were also similar to controls. These results suggest that a single mutation of the glucokinase gene is not a common cause of Type 2 diabetes in English Caucasians.