Background: The NQO1 gene is located on chromosome 16q22.1. This gene encodes a cytosolic enzyme which is expressed in many tissues including the hematopoietic system. NQO1 has an important role in detoxifying quinones, which are widespread compounds that are present thoughout the body (vitamin K), the diet (proteins, fruits and vegetables) and the environment (solvents, cosmetics, cigarette smoke). A single nucleotide polymorphism (C to T) at position 609 (NQO1*2) produces a proline to serine substitution at codon 187 and results in very reduced enzyme activity (null allele). Wild type genotype individuals have normal NQO1 activity, heterozygotes have reduced activity and homozygotes for the mutation have nearly absent activity. The polymorphism is present at varying frequencies in different ethnic groups throughout the world. Several reports have shown an association of this polymorphism and susceptibility to nonhematological malignancies such as lung cancer in smokers. Other reports have found that it is associated with de novo adult AML, therapy-related AML, infant leukemia with MLL rearrangements and benzene-induced hematotoxicity/leukemia.

Aims: To determine the frequency of the NQO1*2 allele in Israeli ethnic groups and to determine if the allele predisposes to de novo adult AML in Israeli patients.

Methods: We used PCR to analyze for the presence of the NQO1*2 allele, utilizing the methodology of MT Smith (Blood 2001) with nested PCR and digestion with Hinf1. We studied adult AML patients of different ethnic origins (156 Jews and 106 Arabs) and normal controls of three major ethnic groups in Israel (270 Caucasian Jews, 250 Arabs and 168 Ethiopian Jews). Too few Ethiopian adult AML patients were identified to be included in the analysis.

Results: In the control Arab group, the distribution of the alleles was: homozygous normal (CC) 54%, heterozygotes (CT) 38.5%, and homozygous mutant (TT) 7.5%, which was virtually identical to the distribution in Arab AML patients (CC 52.8%, CT 41.5% and TT 5.7%). Similarly, the distribution in Caucasian Jewish controls was: CC 60.8%, CT 37.2%, and TT 2%, virtually identical to CC 61.1%, CT 35% and TT 3.9% in Caucasian AML patients. Ethiopian controls had a very different allelic distribution: CC 72.6%, CT 26.2% and TT 1.1 %. While the differences among the various ethnic groups was highly statistically significant, there were no statistically significant differences in allelic frequencies of patients as compared to ethnic matched controls.

Conclusions: In Israel there are significant differences in allelic distribution of NQO1 among various ethnic groups. However, the null allele was not present at a higher frequency in patients, suggesting that the null allele did not predispose to the development of de novo adult AML in Israel. The reason for the discrepancy between our data and other published studies may relate to underlying genetic differences in our population, for example, in CYP2E1, which participates in the same metabolic pathways as NQO1. Alternatively, it is possible that NQO1 protects against de novo AML only under specific environmental conditions, which may not exist in Israel. Our findings emphasize the importance of cross-cultural validation of pharmacogenetic studies, in order to assure the reliability of the findings.

Author notes

Corresponding author