Friedreich's ataxia is the most common hereditary ataxia and is frequently associated with disturbances of glucose metabolism. This autosomal recessive disease is caused by the decreased expression of a mitochondrial protein, frataxin, encoded by the X25 gene. Homozygous expansion of a GAA repeat in the first intron of X25 inhibits frataxin expression and is associated with clinical disease. We evaluated whether heterozygous expansions of the triplet repeat in the frataxin gene X25 may be associated with NIDDM in two genetically distinct populations--one in Germany (n = 358) and the other in the U.S. (n = 292)--using a polymerase chain reaction-based assay. Intermediate expansions (10-36 repeats), which are longer than normal but not sufficient for the appearance of the ataxia phenotype, were found in 24.7 and 27.3% of these two NIDDM cohorts compared with 7.6 and 6.3% of the matched control subjects (both P < 0.001). The odds ratios were 3.36 (95% CI 1.72-6.55) for the German group and 4.01 (2.08-7.74) for the U.S. group. Therefore, we conclude that the X25/frataxin GAA repeat polymorphism is associated with NIDDM in a frequency higher than any other mutation heretofore described. Further studies are needed to elucidate the possible role of frataxin in the pathogenesis of NIDDM.