The anti-atherogenic effect of HDL has been suggested to be partly due to the action of HDL-associated paraoxonase (PON). Three distinct enzymes have been identified, encoded by PON1, PON2 and PON3, clustered on chromosome 7q21-q22. Two cSNPs in PON1 (L55M and Q192R) and one in PON2 (S311C) have been implicated as independent risk factors for coronary heart disease (CHD) in some, but not all, studies. A PON3 SNP (A99A) was identified and the effect of these four PON SNPs on HDL levels and CHD risk was examined in the prospective Northwick Park Heart Study II (NPHSII). Genotype frequencies did not differ between cases and controls but the CHD risk associated with smoking was significantly modified by PON1 L55M genotype. Compared to LL non-smokers, LL smokers had a hazard ratio (HR) of 1.30 (95% CI 0.81-2.06) while M-allele carriers had a HR of 1.76 (1.17-2.67). When genotypes were analysed in combination, men with the genotype PON1 55 LM/MM+PON2 311 CC, had HR of 3.54 (1.81-6.93) compared to PON1 LL+PON2 SS/SC men (interaction P=0.004). These effects were independent of classical risk factors. These data demonstrate the importance of stratifying by environmental factors and the use of multiple SNPs for genetic analysis.