The Val985Met insulin-receptor variant in the Danish Caucasian population: lack of associations with non-insulin-dependent diabetes mellitus or insulin resistance.

To the Editor: A recent study in a Dutch population sample has shown a strong association between a Val985Met missense mutation in the insulin-receptor gene and the common form of non-insulin-dependent diabetes mellitus (NIDDM) (5.6%, vs. 1.3% in the control population). The authors also conclude that this Val985Met variant in the insulin receptor associates with hyperglycemia ('t Hart et al. 1996). The Val985Met missense mutation in exon 17 of the insulin-receptor gene first was found in sporadic form in cohorts ofNIDDM patients and later in control subjects (O'Rahilly et al. 1991, 1992), although without consequences for the insulin-receptor phosphorylation, and, investigated in vivo by overexpression in Chinese hamster-ovary cells, this Val985Met insulin-receptor variant had normal phosphorylation in response to insulin (Flier et al. 1993). When studied in NIDDM pedigrees, this Val985Met insulin receptor variant did not segregate with NIDDM, although carriers of the polymorphism tended to have higher levels of postglucose load levels of plasma glucose (Elbein et al. 1993). In an attempt to replicate the Dutch findings in the Danish Caucasian population, we have determined the distribution of the three known polymorphisms in exon 17 of the insulin-receptor gene in 254 unrelated Danish Caucasian NIDDM patients recruited from the outpatient clinic at Steno Diabetes Center (Copenhagen) and in 243 age-matched, unrelated, and glucose-tolerant Danish Caucasian control subjects traced in the Danish Central Population Register and living in the same area of Copenhagen as the NIDDM patients. Furthermore, we have estimated the potential physiological impact of the Val985Met receptor variant on the whole-body insulin-sensitivity index and on both glucose-stimulated serum insulin and C-peptide release in 380 study participants randomly recruited from a population of young individuals 18-32 years of age, who, in 1979-80 and again in 1984-85, as children, had participated in blood-pressure surveys in a representative and specified part of Copenhagen city and who are characterized by an intravenous glucose-tolerance test in combination with tolbutamide injection and who were analyzed in accordance with the Bergman minimal model (Clausen et al. 1996). In this study material we previously have reported changes in the whole-body insulin-sensitivity index and glucose-stimulated insulin secretion in individuals carrying genetic variants in key proteins in the insulin-signaling cascade (Clausen et al. 1995; Hansen et al. 1995). Genotyping was performed by PCR amplification of the entire exon 17 of the human insulin-receptor gene, from 0.1 ig of genomic DNA, by use of 5 pmol each of intronic sense primer, 5'-tgggtggaaggtggcgtcaga-3', and antisense primer, 5'-tcaggaaagccagcccatgtc-3', yielding a 353-bp DNA segment (Seino et al. 1990). BsaAI and RsaI restriction analyses of 10 jul of PCR product were performed in separate 20-gi reactions. All the reactions were analyzed on a 2% agarose gel. Exon 17 of the insulin-receptor gene contains two sites for BsaAI (ACGT) and only a single site for RsaI (GTAC). One of the BsaAI sites is disrupted by the Val985Met (GTG/ ATG), as well as by the TAC984TAT polymorphism, and the other is disrupted by the CAC1058CAT polymorphism, whereas the RsaI site is disrupted only by the TAC984TAT polymorphism. Therefore, depending on the genotype, BsaAI restriction digestion of exon 17 (353 bp) gives a combination of four major visible