Apolipoprotein A-IV (apo A-IV) is involved in the metabolism of both triglycerides and high-density lipoproteins (HDLs). Apo A-IV has been suggested as participating in several stages of reverse cholesterol transport. Uncertainty about the exact biochemical function of apo A-IV has made the use of genetic apo A-IV polymorphism (variants) attractive in evaluating its physiological role. To date, although some reports indicate that DNA polymorphisms at this locus play an important role in the metabolism of lipids and lipoproteins in western (Caucasian) populations, no similar comprehensive analysis has been performed in a distinct Japanese population. Using DNA sequencing and a restriction fragment length polymorphism (RFLP) study with polymerase chain reaction (PCR), the following allele frequencies were established: (a) codon -8 (G-->A, non-synonymous) allele 2 = 0 (n = 105); (b) codon 9 (A-->G, synonymous) allele 2 = 0.388 (n = 152); (c) codon 347 (A-->T, non-synonymous) allele 2 = 0 (n = 900); (d) codon 360 (T-->G, non-synonymous) allele 2 = 0 (n = 800); (e) VNTR exon 3 [(CTGT)3 and (CTGT)4] (CTGT)3 = 0.262 (n = 105); and (f) MspI (newly detected polymorphic site) polymorphism (C C/T GG) within intron 2, allele 2 = 0.096 (n = 193). The frequencies of these polymorphisms, except for that of the newly identified MspI site, are completely different from those reported in western populations. Among the 900 subjects examined, we found one ACT (Thr) to ACG (Thr) synonymous mutation at codon 347, which does not change the primary structure of apo A-IV. The apo A-IV allele frequency in patients (166 men and 56 women) with angiographically proven coronary heart disease (CHD) was also studied [codon 9 allele 2 = 0.329 (n = 217); VNTR exon 3 (CTGT)3 = 0.262 (n = 84); MspI within intron 2, allele 2 = 0.092 (n = 222)]. Furthermore, we evaluated serum lipid and lipoprotein levels quantitatively in control subjects and Japanese CHD patients. These polymorphisms did not show any consistent and significant association with lipid and lipoprotein parameters. In addition, no gender-specific effects of apo A-IV polymorphisms on lipid parameters adjusted for confounding factors were observed in either CHD patients or control subjects. Our results indicate that the apo A-IV gene is not a major determinant of the risk for CHD in Japanese.