We found significant differences in CYP1A2 enzyme activity between Swedes and Koreans that could not be explained by environmental factors or the CYP1A2 haplotypes examined, despite differences in allele frequencies. None of the investigated CYP1A2 haplotypes are critical in inducing variations in enzyme activity, with the exception of CYP1A2*1F.

CYP1A2 polymorphism -163C>A has an important increasing effect on CYP1A2 inducibility by heavy coffee consumption and may possibly be a contributing factor for interindividual variations in CYP1A2 enzyme activity.

Habitual heavy coffee consumption increases CYP1A2 activity. Polycyclic aromatic hydrocarbons formed during roasting of coffee beans might partly be responsible for this effect. The reason for the observed lower CYP1A2 activity in Serbs as compared to Swedes remains to be investigated.

The basis for interindividual variation in the CYP1A2 gene expression is not fully understood and the known genetic polymorphisms in the gene provide no explanation. We investigated whether the CYP1A2 gene expression is regulated by DNA methylation and displays allele-specific expression (ASE) using 65 human livers. Forty-eight percent of the livers displayed ASE not associated to the CYP1A2 mRNA levels. The extent of DNA methylation of a CpG island including 17 CpG sites, close to the translation start site, inversely correlated with hepatic CYP1A2 mRNA levels (P ¼ 0.018). The methylation of two separate core CpG sites was strongly associated with the CYP1A2 mRNA levels (P ¼ 0.005) and ASE phenotype (P ¼ 0.01), respectively. The CYP1A2 expression in hepatoma B16A2 cells was strongly induced by treatment with 5-aza-2 0 -deoxycytidine. In conclusion, the CYP1A2 gene expression is influenced by the extent of DNA methylation and displays ASE, mechanisms contributing to the large interindividual differences in CYP1A2 gene expression.

At 12 h after dose intake, the regression model predicted a 5.1-fold higher olanzapine plasma level in a non-smoking female patient who did not carry the UGT1A4 142T>G SNP compared to a smoking man treated with the same dose but heterozygous for UGT1A4 142T>G SNP. Whether these combined genetic and environmental factors influence the risk of therapeutic failure remains to be established.

In this study, we analyzed the influence of CYP1A2 genetic variation and enzyme activity on lung cancer risk in a high-incidence area. A total of 95 lung cancer patients and 196 controls were genotyped for the -3860G/A, -3113A/G, -2467T/delT, -739T/G, and -163C/A polymorphisms in the 5'-untranslated region of the gene. In addition, a subset of 70 patients and 115 controls were phenotyped by high-performance liquid chromatography determination of the caffeine metabolic ratio (CMR). The -2467T/delT polymorphism and the CYP1A2*1V haplotype (-163C>A, -2467T>delT) were inversely associated with lung cancer risk (odds ratio [OR] = 0.47 [0.2-0.9]; P = 0.02 and OR = 0.13 [0.02-1.0]; P = 0.04; respectively). In addition, the CYP*1A/*1V and *1F (-163C>A)/*1D (-163C>A, -2467T>delT) diplotypes were absent in the patients group, whereas accounting for 7.1% (P = 0.017) and 5.6% (P = 0.037) of controls, respectively. Mean CMR was significantly higher in patients than in controls (10.50 ± 17.31 vs. 6.52 ± 6.26, P = 0.01) but regression analyses did not yield significant ORs for the association with lung cancer risk. Similarly, no significant correlations were found between any genetic variant and enzyme activity. Several CYP1A2 haplotypes and diplotypes containing the -2467delT variant were associated with lower lung cancer risk; however, they did not correlate with significant changes in CYP1A2 metabolic activity toward caffeine.