Background: A simple method of genotyping and phenotyping cytochrome P450 2A6
(CYP2A6) was previously reported using individual blood samples and urinary caffeine metabolite
ratios of 1,7-dimethyluric acid (17U) to 1-methylxanthine (1X).
Objective: Blood spotted onto storage cards and salivary caffeine metabolites were analyzed in 27
healthy non-smoking Japanese volunteers with no prior abstention from dietary caffeine intake.
Methods: 1,7-Dimethylxanthine (17X), 17U, 1X, and caffeine levels in spot saliva samples were
determined in Japanese non-smokers by high-performance liquid chromatography under normal dietary
Results: 17U/17X ratios in saliva were almost constant over time, but those of 17U/1X were variable in
two subjects tested before and 1-2.5 h after caffeine treatment (a cup of black tea). In seven subjects,
17U/17X ratios in saliva were highly correlated with those in plasma (r = 0.98, p < 0.01) and well correlated
with those in urine samples (r = 0.78, p < 0.05). The average 17U/17X ratios, but not 17U/1X
ratios, in saliva under dietary caffeine consumption obtained from subjects with CYP2A6*1/*4 (n=11)
and CYP2A6*4/*4 (whole-gene deletion, n=2) genotypes were significantly lower than those from subjects
with wild-type CYP2A6*1/*1 (n=14). Genotyping was done by a multiplex real-time polymerase
chain reaction method using blood spotted onto storage cards.
Conclusion: The present results suggest that the decreased CYP2A6 function associated with the
whole-gene deletion genotype (determined using blood samples) could be detected using 17U/17X
ratios, but not 17U/1X ratios, in spot saliva samples under normal dietary caffeine consumption in Japanese
non-smokers, just as it could be detected using urinary 17U/1X ratios.