Differential effects of smoking on myocardial infarction risk according to the Gln/Arg 192 variants of the human paraoxonase gene.

Paraoxonase (PON1) seems to exert a major antioxidant effect by removing lipid peroxidation products. A common polymorphism of the PON1 gene modulates paraoxonase activity and has been related in some studies to coronary heart disease. PON1 genetic polymorphism includes PON1 Q, an isoform with a low activity toward paraoxon hydrolysis that has a glutamine at position 192, and PON1 R, the high-activity isoform with an arginine at position 192. In the present study, we investigated whether smoking, which is related to increased susceptibility to lipoprotein oxidation, has a differential effect by PON1-192 genotype on the risk of myocardial infarction (MI). One hundred fifty-six consecutive MI patients and 310 control subjects were studied. PON1 genotypes in the controls were distributed as follows: 154 (49.7%) QQ, 123 (39.7%) QR, and 33 (10.6%) RR. This distribution did not significantly differ from that of the MI patients: 84 (53.8%) QQ, 60 (38.5%) QR, and 12 (7.7%) RR. Subjects were classified into two groups, those who never smoked (n = 209) and those who were current smokers (n = 135) or ex-smokers (n = 122). In the latter, the variable "cigarette packs smoked per year" was defined as the number of packs smoked daily multiplied by the number of smoking years. As expected, smoking was significantly associated with an increased MI risk in the overall group. Subjects were then stratified by PON1 genotype. The packs smoked per year were significantly associated with an increased MI risk only in QQ homozygotes. This risk was higher among those in the higher tertile for cigarette packs smoked per year (odds ratio [OR] = 5.24, 95% confidence interval = 1.67 to 16.44, Pfor trend <.001). In contrast, the packs smoked per year were not significantly associated with MI risk in R-carrier subjects. We conclude that the risk of MI associated with smoking appears to be increased in subjects who are homozygous for the low-activity PON1 QQ genotype compared with R carriers, and this risk seems to be time- and dose-dependent.