The present study examined the relationship between genetic combinations of the C1429T polymorphism in the G-protein beta-3 subunit (GNB3) gene and the A1330V polymorphism in the low-density lipoprotein receptor-related protein 5 (LRP5) gene and the risk of hypercholesterolaemia in Japanese workers. The present study included observations from 1997 to 2002 in 927 males and 662 females who were not hypercholesterolaemic on entry. The endpoint was the development of hypercholesterolaemia, defined as a total cholesterol level > or =240 mg/dl. The odds ratios for the combination of polymorphisms were calculated using pooled logistic regression analyses that incorporated other potential factors into the model. The odds ratios in males and females with GNB3/1429TT and LRP5/1330VV or AV genotypes were 4.17 compared to males with the 1429CT or TT and 1330AA genotypes and 3.53 compared to females with the 1429CC, CT or TT and 1330AA genotypes. Assuming these effects were a mere addition of two independent effects, the odds ratios for both GNB3/1429TT and LRP5/1330VV or AV were estimated to be 3.27 for males and 1.42 for females. Therefore, the synergic effects were shown to be 1.28 times in males (not significant) and 2.49 times in females (P<0.05 by bootstrap method). These results provide clear evidence that the genetic combination has a synergic effect. This study indicates that the combination of GNB3/C1429T and LRP5/A1330V is a very useful marker for predicting the development of hypercholesterolaemia in the general Japanese population.