Background: Inosine triphosphate (ITP) pyrophosphohydrolase (ITPA) catalyzes the pyrophosphohydrolysis of ITP/dITP and xanthosine triphosphate to prevent incorporation of unusual nucleotides into RNA and DNA. Important mutations leading to enzyme deficiency are 94C>A and IVS2 + 21A>C. An association between ITPA 94C>A and adverse reactions during azathioprine treatment has been shown. To investigate the ITPA phenotype, an HPLC procedure was developed and phenotype-genotype correlations were assessed.
Methods: The enzymatic conversion of ITP to inosine monophosphate (IMP) was terminated by perchloric acid and saturated dipotassium hydrogen phosphate. We quantified the IMP at 262 nm after separation on an Aqua perfect C18 column using 20 mmol/L phosphate buffer, pH 2.5. We also genotyped samples for ITPA 94C>A and IVS2 + 21A>C by real-time fluorescence PCR.
Results: The assay was linear to 3 mmol/L IMP [approximately 500 micromol/(g Hb x h)] with a lower limit of quantification of 4 micromol/L [approximately 0.5 micromol/(g Hb x h)]. With IMP-enriched samples, within- and between-day imprecision was < or = 3.6% and < or = 4.9%, respectively, and the inaccuracy was < or = 5.2%. With pooled erythrocytes, within- and between-day imprecision was 3.8% and 7.5%, respectively. ITPA activity in 130 healthy controls was between < 0.5 and 408 micromol IMP/(g Hb x h). Mutant allele frequencies were 0.062 (94C>A) and 0.131 (IVS2 + 21A>C). When we used a cutoff of 125 micromol IMP/(g Hb x h), phenotyping detected all 94C>A mutant cases, all 94C>A and IVS2 + 21A>C compound heterozygotes, all IVS2 + 21A>C homozygotes, and 6 of 24 IVS2 + 21A>C heterozygote-only cases. A novel IVS2 + 68T>C mutation was also found.
Conclusions: The HPLC procedure provides an excellent ITPA phenotype-genotype correlation and led to the discovery of a novel IVS2 + 68T>C mutation. The method could facilitate investigation of the role of ITPA activity for drug toxicity during thiopurine therapy.