Genotype-phenotype correlation for thiopurine S-methyltransferase in healthy Italian subjects

Objective: The aim of the present study was to estimate the concordance rat e between erythrocyte thiopurine methyltransferase (TPMT) activity and geno type at the TPMT locus in an Italian population sample. Methods: The TPMT phenotype and genotype were determined in an unrelated po pulation of 103 Italian healthy blood donors. Erythrocyte TPMT activity was measured with a radiochemical assay using 12.5 muM S-adenosyl-L-(methyl-C- 14)-methionine and 4 mM 6-mercaptopurine. The genotyping assay was based on restriction fragment length polymorphism polymerase chain reaction (RFLP-P CR) and allele-specific oligonucleotide polymerase chain reaction (ASO-PCR) methods. Results: All subjects had detectable TPMT activity. The activity of TPMT va ried 2.8-fold between the 5th and 95th percentile. This variation was neith er age (P = 0.63) nor gender (P = 0.44) regulated and the frequency distrib ution of TPMT activity is compatible with a polymorphic distribution. The p resence of the four most common defective alleles, i.e. TPMT*2, TPMT*3A, TP MT*3B and TPMT*3C, was examined through the entire phenotyped population. N inety-two subjects did not carry any of the tested mutations, Eleven indivi duals were heterozygous for one of the mutant alleles and had a TPMT activi ty lower than 30 pmol/min/mg. Eight subjects were TPMT*1/TPMT*3A, two TPMT* 1/TPMT*3C and one was TPMT*1/TPMT*2. The TPMT*3B allele was not detected in the samples analysed. Conclusion: There was a concordance of 97% between genotype and phenotype. All the heterozygotes had an intermediate phenotype. However, the wide vari ation range in TPMT activity detected in the wild-type homozygotes indicate s that other genetic or epigenetic factors influence the TPMT phenotype.