In the present paper, we examined the incidence of polymorphic genes involved with the detoxification of exogenous chemicals, including carcinogens, namely GSTT1 (glutathione transferase θ1), GSTM1 (glutathione transferase μ1) and NQO1 (NAD(P)H:quinone oxidoreductase 1) in 60 Filipino children with ALL (acute lymphoblastic leukaemia). We found a significantly high incidence of the GSTM1 null genotype in ALL children (71.7%) compared with 51.7% in the control group of children (P<0.05). The GSTT1 null genotype was observed in 35.0% and 33.3% of the ALL cases and the control subjects respectively, with no significant difference. Screening for NQO1 (609C>T) mutant alleles showed a high incidence of the NQO1 C/C genotype (NQO1 homozygous wild-type allele genotype) in 60.0% of ALL cases and was significantly higher than in the control group (23.3%) (P<0.01). These GSTM1 null and NQO1 wild-type genotypes are independently associated with the risk of ALL in Filipino patients. When these two genotypes, GSTM1 null and NQO1 C/C, were combined, the hazard rate for childhood leukaemia was significantly increased (P<0.001). We also noticed that the incidences of GSTM1 null mutations and the NQO1 C/C genotype were significantly higher among Filipinos. These findings suggest a possible role of the GSTM1 null and NQO1 C/C genotypes in the susceptibility of paediatric ALL cases in the Philippines.

INTRODUCTION

Leukaemia, in particular ALL (acute lymphoblastic leukaemia), is the most common form of childhood malignancy [1]. The peak age incidence of ALL is 3–4 years [2] and it is most prevalent among Hispanics, followed by Caucasians and Asians, and lowest among Afro-Americans [1,3]. Despite the rates of successful induction of remission or cure by treatment improving significantly in ALL [4,5], the incidence of ALL cases has been increasing [1,6].

The aetiology of leukaemia is largely unknown, although it is considered to be multifactorial [7]. The involvement of certain environmental exposures has been linked to leukaemia, such as contact with industrial chemicals like benzene [8], industrial and household chemicals [9,10] and chemotherapeutic drugs, such as topoisomerase inhibitors [11] and ionizing radiation [12]. In addition, leukaemogenesis is considered to be affected by interactions between genes and environments [13]. In particular, it was reported that potential carcinogens from the environment or their metabolites, such as dipyrone [9] and permethrin [10], cause DNA damage that induces irrepairable chromosomal breaks and thus leads to initiation and progression of leukaemia [13,14].

Individuals, however, display varying degrees of susceptibility as well as severity of illness [15] towards potential carcinogens, indicating differences in genetic background among individuals, such as variations in enzyme expression involved in either detoxification or DNA repair. GSTT1 (glutathione transferase θ1), GSTM1 (glutathione transferase μ1) and NQO1 [NAD(P)H:quinone oxidoreductase 1] belong to the phase II group of common detoxification enzymes which facilitate the conversion of toxic metabolites (initially metabolized by phase I enzymes) into more water-soluble products that can easily be eliminated [16]. Genes coding for these enzymes are reported to be polymorphic, thereby the presence of both alleles renders an individual with the capacity to eliminate potential carcinogens more efficiently than those with a deletion or mutation of the gene.

GSTT1 and GSTM1 are dimeric soluble proteins of approx. 25 kDa [17]. A wide variety of electrophilic compounds, including well-known environmental carcinogens such as benzopyrene, epoxides and PAH (polyaromatic hydrocarbons), and other industrial carcinogens such as organophosphates, alkylating agents, dihalomethanes, buta-1,3-diene and ethylene oxides [18,19] are detoxified by these enzymes.

Previous studies have clarified the correlation of deletion of alleles of either GSTT1 and GSTM1 or a null genotype with the risk of developing ALL [2024], other leukaemia subtypes [2527] and other malignancies such as cervical, lung, prostate, esophageal, colon and breast cancers [2832]. Genotype frequencies for both GSTT1 and GSTM1 also vary with ethnic background, with the highest percentage of the null genotype observed in 80% of Asians compared with 20% of Caucasians [18]. Among Asians, the GSTT1 null genotype is more frequent among Japanese, Korean and Chinese populations [33,34].

NQO1 is a dimeric flavoprotein which specifically catalyses the reduction of numerous quinines [35]. It is an important chemoprotective enzyme against quinines or quinones derived from the oxidation of phenolic metabolites of benzene [36]. It also functions in generating antioxidant forms of vitamin E, quinine and ubiquinone, thereby protecting cells from oxidative stress [37], scavenging toxic superoxides within the cell [38] and stabilizing p53 [39]. By contrast, apart from detoxification, NQO1 is known to promote carcinogenesis by activating carcinogens such as nitrosamines and heterocyclic amines present in tobacco smoke and food [36]. Currently, 22 single-nucleotide polymorphisms in the NQO1 gene have been reported [40], including a cytosine to thymine point mutation in exon 6 (609C>T), which leads to a proline to serine amino-acid residue substitution (P187S) and results in low enzymatic activity [41]. Genotypes can either be NQO1 C/C (NQO1 homozygous wild-type allele genotype), NQO1 C/T (NQO1 heterozygous wild-type and 609C>T allele genotype) or NQO1 T/T (NQO1 homozygous 609C>T mutant allele genotype). Association of the NQO1 609C>T polymorphism with the risk of developing ALL and various cancer types have been reported previously [4244]. Similarly to GSTT1 and GSTM1, the frequency of the NQO1 609C>T genotype varies with race, with a high frequency of the mutant allele observed among Hispanics and Asians [45].

Considering that leukaemia is the leading malignancy among Filipino children [46] and that a high prevalence of gene deletion or mutation among Asians has been reported, we investigated whether this high incidence would be present among leukaemic children in the local region. The present study, therefore, was conducted to investigate the prevalence of these genes (GSTT1, GSTM1 and NQO1) among Filipino leukaemic children residing within the NCR (National Capital Region) in the northern part of the Philippines, including Manila, Quezon City, Pasig and Paranaque. This study also attempted to determine the independent and combined relative risks of GSTT1, GSTM1 and NQO1 genotypes attributable to the occurrence of ALL among Filipino children. In the present paper, we provide evidence of a positive association between GSTM1 null and NQO1 C/C genotypes and ALL susceptibility in the Philippines.

MATERIALS AND METHODS

Sample collection and clinical profiles

Genotype screening for GSTT1, GSTM1 and NQO1 was conducted in 60 unrelated ALL paediatric cases and 60 randomly-selected normal children, as controls, with ages less than 18 years old. Cases were newly diagnosed or receiving treatment at the University of Santo Tomas Hospital in Manila and the Children's Medical Center in Quezon City during the period of 1 June to 30 October 2005. Controls were chosen from children receiving routine check-ups at the University of Santo Tomas Hospital, with a complete blood count falling within normal range, reported to be in good health and were not suspected to have any type of malignancy. All subjects represented natural- born Filipino children, who are not products of marriages of a Filipino with another race (eg. Caucasians or other Asians) and were residing within the NCR. Peripheral-blood samples for both case and control subjects were collected. Informed consent of the subjects was given, and approval of the local bioethics committees was obtained.

Clinical information about these ALL cases, including age at diagnosis, gender, WBC (white blood cell) count at diagnosis and ALL subtypes (pre-B-cell ALL, T-cell ALL and undetermined lineage) is summarized in Tables 1 and 2. Among the 60 ALL cases, 70.0% were undergoing chemotherapy, 11.7% had clinical relapse of ALL at the time of this study and 13.3% were in remission.

Table 1
Comparison of general profiles of ALL and control cases

The number of individuals was 60 in both the ALL group and the control group. The values stated relate to the number of indivduals in each category, and the values in parentheses are the number of individuals in each category expressed as a percentage of the group. The statistical significance was estimated using the χ2 test.

General descriptionALL groupControl groupSignificance (P value)
Age (at diagnosis)    
 <1–5 years old. 42 (70.0) 22 (36.7) 0.0001 
 6–10 years old 13 (21.7) 14 (23.3) – 
 11–18 years old 5 (8.33) 24 (40.0) – 
Gender    
 Male 38 (63.3) 30 (50.0) 0.1406 
 Female 22 (36.7) 30 (50.0) – 
General descriptionALL groupControl groupSignificance (P value)
Age (at diagnosis)    
 <1–5 years old. 42 (70.0) 22 (36.7) 0.0001 
 6–10 years old 13 (21.7) 14 (23.3) – 
 11–18 years old 5 (8.33) 24 (40.0) – 
Gender    
 Male 38 (63.3) 30 (50.0) 0.1406 
 Female 22 (36.7) 30 (50.0) – 
Table 2
Summary of ALL clinical characteristics of 60 Filipino patients with paediatric leukaemia

The values are the number of indivduals in each category, and the values in parentheses are the number of individuals in each category expressed as a percentage of the group.

Clinical featuresNumber of individuals
WBC count (×109 cells per litre)  
 <10 29 (48.3) 
 11–30 16 (26.7) 
 31–50 8 (13.3) 
 >50 7 (11.7) 
Immunophenotype/lineage  
 Pre-B-cell 41 (68.3) 
 T-cell 6 (10.0) 
 Biphenotypic 5 (8.33) 
 Early Pre-B-cell 1 (1.67) 
 Mixed lineage 1 (1.67) 
 Undefined 5 (8.33) 
Chemotherapy status  
 Under chemotherapy 42 (70.0) 
 Relapsed cases 7 (11.67) 
 Remission 8 (13.3) 
 Off therapy 4 (6.67) 
Clinical featuresNumber of individuals
WBC count (×109 cells per litre)  
 <10 29 (48.3) 
 11–30 16 (26.7) 
 31–50 8 (13.3) 
 >50 7 (11.7) 
Immunophenotype/lineage  
 Pre-B-cell 41 (68.3) 
 T-cell 6 (10.0) 
 Biphenotypic 5 (8.33) 
 Early Pre-B-cell 1 (1.67) 
 Mixed lineage 1 (1.67) 
 Undefined 5 (8.33) 
Chemotherapy status  
 Under chemotherapy 42 (70.0) 
 Relapsed cases 7 (11.67) 
 Remission 8 (13.3) 
 Off therapy 4 (6.67) 

Genotyping

Genomic DNA was obtained from peripheral-blood samples of leukaemia and control cases and was extracted using the Wizard® genomic DNA purification kit (Promega). All samples were genotyped in triplicate with consistent results confirming that the genotypes have no mixture of germline and somatic mutations.

GSTT1 and GSTM1 genes were amplified using multiplex PCR as described previously [22,29,47,48], with co-amplification of the albumin gene as a positive internal control. A PCR mixture containing genomic DNA without the GSTT1 and GSTM1 primers was used as a negative control. A total reaction volume of 20 μl was prepared as described previously [47] containing 10 mM Tris/HCl (pH 8.3), 50 mM KCl, 3.5 mM MgCl2, 3.0 μg/ml of each GSTM1 primer and 1.0 μg/ml of each GSTT1 primer, 0.6 μg/ml of each albumin primer, 200 μM dNTPs, 0.5 unit of Taq DNA polymerase. Genomic DNA (100 ng/μl) from each individual was used as a template with the following primer sequences: GSTT1: 5′-TTCCTTACTGGTCCTCACATCTC-3′ and 5′-TCACCGGATCATGGCCAGCA-3′ [47,48]; GSTM1: 5′-GAACTCCCTGAAAAGCTAAAGC-3′ and 5′-GTTGGGCTCAAATATACGGTGG-3′ [47,48] and albumin: 5′-GCCCTCTGCTAACAAGTCCTAC-3′and 5′-GCCCTAAAAAGAAAATCGCCAATC-3′ as reported previously [47]. All samples were amplified in duplicate using the following PCR conditions: 95°C (1 min), 66°C (1 min), 72°C (1 min) for 40 cycles, with a final extension step of 10 min at 72°C in a PTC-200 Peltier thermal cycler (MJ Research).

Genotyping for NQO1 was conducted using primers constructed previously for the PCR–CTPP (PCR confronting two-pair primer) method [28,49]. A schematic diagram of the PCR–CTPP method for simultaneous detection of the NQO1 wild-type and mutant alleles as reported previously [28,49] is shown in Figure 2. The reaction mixture (25 μl) contained 1× PCR Buffer (Promega), 200 nM of each primer, 200 μM dNTPs, 3.5 mM MgCl2, 0.5 unit of Taq DNA polymerase and 100 ng/μl template DNA. A PCR mixture containing genomic DNA without the NQO1 C allele and T allele primers was used as a negative control. The primers used for NQO1 609C>T detection [28,49] were as follows: C allele (wild-type): 5′-TATCAGAGTGTCTTACTGAGA-3′ and 5′-AATGCTATATGTCAGTTGAGG-3′; T allele (mutant): 5′-GTGGCTTCCAAGTCTTAGAAT-3′ and 5′-TTTCTAGCTTTGATCTGGTTG-3′ with the following PCR conditions: 95°C (1 min), 60°C (1 min) and 72°C (1 min) for 40 cycles, with a final extension step of 10 min at 72°C using a PTC-200 Peltier thermal cycler.

Genotype frequencies and their evaluation

To determine the prevalence of the GSTT1, GSTM1 and NQO1 genotypes in the local population, allele and genotype frequencies were computed, including deviations from the Hardy–Weinberg Equilibrium. The Pearson χ2 test was used to compare the differences in genotype frequencies between leukaemia cases and controls. The relative risks of independent and combined genotypes for ALL leukaemogenesis among local children were determined by computing for the OR (crude odds ratio) with a 95% CI (confidence interval).

RESULTS

Genotype screening for GSTT1, GSTM1 and NQO1 by multiplex PCR was performed in 60 Filipino ALL paediatric patients and 60 randomly selected control children. Figure 1 shows the representative results of ten ALL patients for GSTT1 and GSTM1, and NQO1 genotype analyses showing most of the genotype varieties. The presence of a 459 bp PCR product for GSTT1 and 219 bp product for GSTM1 with reference to the 350 bp positive internal control (albumin) indicated a ‘positive’ genotype (Figure 1). Our study applied the conventional genotyping method described previously [22,29,47,48] and did not enable us to discriminate heterozygous from homozygous genotypes for the positive GSTT1 and GSTM1 results. With regards to NQO1, the presence of 161 bp and 283 bp PCR products corresponded to NQO1 wild-type and mutant alleles respectively (Figure 2A). Figure 2(A) illustrates the concept of the PCR–CTPP method for the multiplex detection of NQO1 wild-type and mutant alleles as reported previously [33].

Representative PCR results of DNA

Figure 1
Representative PCR results of DNA

Genotypes were screened for GSTT1 (459 bp) and GSTM1 (219 bp) by multiplex PCR, with albumin (350 bp) amplified as a positive internal control. The PCR results for ten patients is shown. The arrows indicate the positions of GSTT1 (459 bp), Albumin (350 bp) and GSTM1 (219 bp) genes that are amplified. The presence (+) or absence (−) of GSTT1 and GSTM1 is indicated below. We cannot distinguish between heterozygous and homozygous genotypes using the current PCR protocol. Mr, 100 bp DNA ladder marker is shown on the left-hand side (in bp); Pt, patient.

Figure 1
Representative PCR results of DNA

Genotypes were screened for GSTT1 (459 bp) and GSTM1 (219 bp) by multiplex PCR, with albumin (350 bp) amplified as a positive internal control. The PCR results for ten patients is shown. The arrows indicate the positions of GSTT1 (459 bp), Albumin (350 bp) and GSTM1 (219 bp) genes that are amplified. The presence (+) or absence (−) of GSTT1 and GSTM1 is indicated below. We cannot distinguish between heterozygous and homozygous genotypes using the current PCR protocol. Mr, 100 bp DNA ladder marker is shown on the left-hand side (in bp); Pt, patient.

Diagram of the PCR method used and representative results for NQO1 (609C>T) detection

Figure 2
Diagram of the PCR method used and representative results for NQO1 (609C>T) detection

(A) The PCR–CTPP method for multiplex detection of NQO1 wild-type and mutant alleles. The method has been described previously [33]. (B) Representative PCR results for ten patients (Pt) screened for the NQO1 wild-type allele (C; 161 bp) and the mutant allele (T; 283 bp) by multiplex PCR. Only patient 7 exhibited the NQO1 C/C genotype, whereas the other patients displayed the NQO1 C/T genotype. Mr, 100 bp DNA ladder marker is shown on the left-hand side (in bp).

Figure 2
Diagram of the PCR method used and representative results for NQO1 (609C>T) detection

(A) The PCR–CTPP method for multiplex detection of NQO1 wild-type and mutant alleles. The method has been described previously [33]. (B) Representative PCR results for ten patients (Pt) screened for the NQO1 wild-type allele (C; 161 bp) and the mutant allele (T; 283 bp) by multiplex PCR. Only patient 7 exhibited the NQO1 C/C genotype, whereas the other patients displayed the NQO1 C/T genotype. Mr, 100 bp DNA ladder marker is shown on the left-hand side (in bp).

Table 1 illustrates the general profile of the ALL patients and the control subjects. There are more children with ALL under 5 years old, presumably because the peak age of ALL was reported to be 3–4 years old [2]. Although there was an age difference between the ALL cases and the control cases, this profile is not concerned with the frequency distribution and case-control associations of the genotypes to leukaemia.

Table 2 shows the clinical features of 60 ALL patients enrolled in this study. The majority of the ALL patients have WBC counts of <1010 per litre (48.3%), exhibited the pre-B-cell immunophenotype (68.3%) and are undergoing chemotherapy (70.0%) (Table 2).

A comparison of each allele and the genotype frequencies for GSTT1, GSTM1 and NQO1 in between ALL cases and the control group is summarized in Table 3. Since allele frequencies may be used directly to state probabilities in observing the same genotypes in succeeding generations, we determined the allele frequencies for GSTT1, GSTM1 and NQO1. The GSTT1 null allele frequency is 0.59 and 0.58 in the ALL and normal groups respectively. Although there is no significant difference in the GSTT1 null allele frequencies, the frequency of GSTM1 null alleles in the ALL group (0.85) is significantly higher than the controls (0.72) (P<0.05). In addition, the NQO1 wild-type allele frequency is significantly higher among ALL patients (0.80) than the control group (0.61) (P<0.05).

Table 3
Comparison of GSTT1, GSTM1 and NQO1 allele and genotype frequencies in ALL and control Filipino children

The number of individuals was 60 in both the ALL group and the control group. For genotype frequency, the values are the number of individuals in each category, and the values in parentheses are the number of individuals in each category expressed as a percentage of the group.

ALL groupControl group
GeneAllele frequencyGenotype frequencyAllele frequencyGenotype frequencyOR (95% CI)Significance (P value)
GSTT1       
 Positive 0.41 39 (65.0) 0.42 40 (66.7) 1.0* 0.85 
 Null 0.59 21 (35.0) 0.58 20 (33.3) 1.08 (0.51–2.29) – 
GSTM1       
 Positive 0.15 17 (28.3) 0.28 29 (48.3) 1.0* 0.02 
 Null 0.85 43 (71.7) 0.72 31 (51.7) 2.37 (1.11–5.04) – 
NQO1       
 C/C 0.80 (C allele) 36 (60.0) 0.61 (C allele) 14 (23.3) 4.82 (2.18–10.6) 0.0002 
 C/T 0.20 (T allele) 24 (40.0) 0.39 (T allele) 45 (75.0) 1.0* – 
 T/T 0.00 0 (0) 0.00 1 (1.67) – – 
ALL groupControl group
GeneAllele frequencyGenotype frequencyAllele frequencyGenotype frequencyOR (95% CI)Significance (P value)
GSTT1       
 Positive 0.41 39 (65.0) 0.42 40 (66.7) 1.0* 0.85 
 Null 0.59 21 (35.0) 0.58 20 (33.3) 1.08 (0.51–2.29) – 
GSTM1       
 Positive 0.15 17 (28.3) 0.28 29 (48.3) 1.0* 0.02 
 Null 0.85 43 (71.7) 0.72 31 (51.7) 2.37 (1.11–5.04) – 
NQO1       
 C/C 0.80 (C allele) 36 (60.0) 0.61 (C allele) 14 (23.3) 4.82 (2.18–10.6) 0.0002 
 C/T 0.20 (T allele) 24 (40.0) 0.39 (T allele) 45 (75.0) 1.0* – 
 T/T 0.00 0 (0) 0.00 1 (1.67) – – 
*

Value used as a reference.

In terms of the genotype incidence of GSTT1, GSTM1 and NQO1 from both the ALL cases and the control group, we found a high incidence of the GSTM1 null (71.7%) and NQO1 C/C (60.0%) genotypes among ALL patients (Table 3). This high GSTM1 null genotype incidence among ALL patients is significantly higher (P<0.05) than the incidence in the control group (51.7%). However, the incidence of the NQO1 C/C genotype is significantly higher in ALL children (P<0.05) than in the control group (23.3%). In addition, the incidence of the NQO1 C/T genotype is 40% in ALL children, compared with 75.0% in the control group. The GSTT1 null genotype in individuals with leukaemia and normal individuals, on the other hand, accounts for 35.0% and 33.3% respectively, and is consistent with the Hardy–Weinberg equilibrium. However, GSTM1 and NQO1 gene frequencies were not completely consistent, suggesting the possibility of slight sampling bias or population stratification, although in every genotyping procedure, population stratification may be present, even in well-designed studies [50].

With an observed high GSTM1 null incidence among ALL children, the risk of developing ALL was further determined and we found a 2.37-fold increase in the risk (OR=2.37; 95% CI=1.11–5.04) for children possessing the GSTM1 null genotype. By further determining the relative risk for NQO1, we found a significant 4.82-fold increase in the risk of developing ALL (OR=4.82; 95% CI=2.19–10.6) due to a NQO1 C/C genotype. Since both GSTM1 null and NQO1 C/C genotypes are associated with an increased risk of ALL, we further determined the combined effect of these two genotypes. Table 4 showed that the relative risk of developing ALL was further increased by 11.9-fold (OR=11.9; 95% CI=3.5–41.1) when both GSTM1 null and NQO1 C/C genotypes were combined.

Table 4
Comparison of relative risks of ALL among the ALL case and control groups with combinations of GSTM1 and NQO1 genotypes

The number of individuals was 60 in both the ALL group and the control group. The values are the number of indivduals in each category, and the values in parentheses are the number of individuals in each category expressed as a percentage of the group.

Genotype combinationsALL groupControl groupOR (95% CI)
GSTM1 positive+NQO1 C/T 6 (10.0) 20 (33.3) 1.0* 
GSTM1 positive+NQO1 C/C 11 (18.3) 7 (11.7) 5.24 (1.41–19.52) 
GSTM1 null+NQO1 C/T 18 (30.0) 26 (43.3) 2.31 (0.77–6.88) 
GSTM1 null+NQO1 C/C 25 (41.7) 7 (11.7) 11.9 (3.45–41.09) 
Genotype combinationsALL groupControl groupOR (95% CI)
GSTM1 positive+NQO1 C/T 6 (10.0) 20 (33.3) 1.0* 
GSTM1 positive+NQO1 C/C 11 (18.3) 7 (11.7) 5.24 (1.41–19.52) 
GSTM1 null+NQO1 C/T 18 (30.0) 26 (43.3) 2.31 (0.77–6.88) 
GSTM1 null+NQO1 C/C 25 (41.7) 7 (11.7) 11.9 (3.45–41.09) 
*

Value used as a reference.

DISCUSSION

In the present paper, we report the genotype incidence of GSTT1, GSTM1 and NQO1 genes coding for phase II detoxification enzymes in children with ALL. We found a high incidence of deletion of both GSTT1 and GSTM1 genes and the presence of the NQO1 C/C genotype among Filipino ALL paediatric patients. A gross deletion of GSTT1 and GSTM1 or a single-base substitution in the NQO1 gene leads to limited or no expression and thus a diminished function of these enzymes [18,40]. To our knowledge, this is the first report of the incidence of GSTT1, GSTM1 and NQO1 among children diagnosed with ALL in the Philippines.

The positive association we observed in the GSTM1 null genotype with the risk of ALL coincides with three independent leukaemia studies of Canadian [21], Caucasian [26] and Thai populations [23]. Also, the high GSTM1 null prevalence (71.7%) we observed in this study coincides with previous reports based on Canadian (64.9%) [21] and Black patients (41.2%) [22]; however, the GSTM1 null incidence among Filipinos is slightly elevated compared with these two ethnicities. Furthermore, the GSTM1 null genotype frequency from other ethnicities with various malignancies (40.0%–64.0%) [19,25,26,29] is similarly prevalent.

Although a 2.37-fold risk of ALL is associated with the GSTM1 null genotype, we found that the GSTT1 null genotype was not associated with ALL among Filipinos. Our finding is similar with association studies reported previously with ALL in the U.S.A. [20,21,22] and Thailand [23], and with other cancer types such as AML (acute myeloid leukaemia) [25,26], cervical [28] and lung cancers [19].

This association of the GSTM1 null genotype with ALL risk and not the GSTT1 null genotype can probably be explained by differences in catalytic properties among GST subtypes. Although GSTT1 and GSTM1 have overlapping substrate specificity, these enzymes exhibit different catalytic activities [18]. For instance, deletion of the GSTM1 gene results in an increase in the DNA-damaging effect of TSO (trans-stilbene oxide) in cultured human lymphocytes, but deletion of GSTT1 has no effect [51]. Other potential carcinogens which are specific substrates for GSTM1 and not GSTT1 may be present in the environment thus increasing the risk of leukaemogenesis in individuals with the GSTM1 null genotype.

Similar to our observation on the association of the GSTM1 null genotype with ALL risk, we found a positive association of the NQO1 C/C genotype with ALL. However, two previous reports on ALL observed the opposite results [43,52]. Krajinovic et al. [43] and Smith et al. [52] reported a positive association of the NQO1 C/T genotype with ALL. In contrast, other reports support our results with NQO1 genotype analysis. For example, a positive association of the NQO1 C/C genotype with CML (chronic myelogenous leukaemia) [27] and lung cancer [29,53] was reported. In addition, another study [25] showed a negative association of the NQO1 C/T genotype with de novo AML risk. Although we do not currently know the reason, it is possible that these apparent discrepancies might reflect differences in the chemical carcinogens involved in leukaemogenesis in different countries. Furthermore, the association of the NQO1 C/C genotype with the risk of developing ALL may be attributable to the function of the enzyme in activating carcinogens aside from detoxification. A review by Ross et al. [36] postulated that some endogenous metabolites can be transformed by NQO1 to yield more active products that can produce reactive oxygen or alkylating species, thus attacking nucleophilic sites within essential biomolecules such as DNA. An example of this is the reduction and activation of nitro compounds found in cooked foods such as 4NQO (4-nitroquinoline-1-oxide) [36,54].

We also determined the possible enhanced risk of ALL with a combination of the GSTM1 null and NQO1 C/C genotypes, and observed that ALL susceptibility was enhanced by a combination of these genotypes. Although the extent of increased risk was expected, our findings support the idea that GSTM1 and NQO1 genotypes both contribute to ALL independently. The positive association of the combined GSTM1 null and NQO1 C/C genotypes with cervical cancer in Japan was reported previously [28]. However, no such previous genetic analysis of ALL has been reported.

From our findings, we therefore consider that the GSTM1 null and NQO1 C/C genotype combination may be likely to play a role in ALL leukaemogenesis among Filipino children. In addition, considering that NQO1 has the potential to activate probable substrates into carcinogens [36], including GSTT1, which is capable of transforming some substrates such as ethylene dibromide and halogenated methanes into mutagenic electrophiles [19], it is possible that the putative and so far unidentified carcinogen(s) involved may be activated by NQO1 and detoxified by GSTM1.

Further studies in combination with other genes for detoxification or DNA repair is recommended, for instance, CYP1A1, a phase 1 enzyme which, in combination with the GSTM1 null genotype, is known to be associated with an increased risk of ALL [26]. Similarly, CYP2E1 (cytochrome P450 2E1) and MPO (myeloperoxidase) in combination with NQO1 further increase the risk of malignancies [48]. It would be thus interesting to investigate combinations of phase 1 and phase 2 groups of enzymes, since the former are generally the first enzymatic defence against foreign compounds [55] and can generate more reactive molecules that are further metabolized by GSTs and NQO1.

In conclusion, we clarified the genetic contribution of GSTM1 null and NQO1 C/C genotypes among children living in the Philippines to the risk of increased ALL leukaemogenesis (11.9-fold). Although we still have not identified the carcinogenic substance involved, we assume that metabolic activation of potential carcinogen(s) might play a role. From this baseline information, we hope to stimulate further studies on ALL risk factors, particularly to investigate the influence of environmental exposure that can contribute to the awareness, preventive and management measures of ALL in the Philippines.

Abbreviations

     
  • ALL

    acute lymphoblastic leukaemia

  •  
  • AML

    acute myeloid leukaemia

  •  
  • CI

    confidence interval

  •  
  • GSTM1

    glutathione transferase μ1

  •  
  • GSTT1

    glutathione transferase θ1

  •  
  • NQO1

    NAD(P)H:quinone oxidoreductase 1

  •  
  • NQO1 C/C genotype

    NQO1 homozygous wild-type allele genotype

  •  
  • NQO1 C/T genotype

    NQO1 heterozygous wild-type and mutant 609C>T allele genotype

  •  
  • NCR

    National Capital Region

  •  
  • OR

    crude odds ratio

  •  
  • PCR–CTPP

    PCR confronting two-pair primer

  •  
  • WBC

    white blood cell

We thank Dr A. Cavilles, Dr L. Naranjo, Dr L. Lepongo, Dr A. Bautista, Dr J. Dadura, Dr Hernandez and the staff of the Hematology Unit of the Children's Medical Center and the University of Santo Tomas Hospital for recruiting the ALL patients. We also thank Maria Teresa Sur, R. Sio, Dr M. Orillaza and Dr Raymundo for sample collection. We also thank the Philippine Council for Health Research and Development for financial support. We thank Dr Ann Florence B. Victoriano and Marni E. Cueno for language editing and Dr John Donnie A. Ramos for his comments and suggestions.

References

References
McNeil
 
D. E.
Cote
 
T. R.
Clegg
 
L.
Mauer
 
A.
 
SEER update of incidence and trends in pediatric malignancies: acute lymphoblastic leukemia
Med. Pediatr. Oncol.
2003
, vol. 
41
 (pg. 
413
-
414
)
Sather
 
H. N.
 
Age at diagnosis in childhood acute lymphoblastic leukemia
Med. Pediatr. Oncol.
1986
, vol. 
14
 (pg. 
166
-
172
)
Matasar
 
M. J.
Ritchie
 
E. K.
Consedine
 
N.
Magai
 
C.
Neugut
 
A. I.
 
Incidence rates of the major leukemia subtypes among US hispanics, blacks, and non-hispanic whites
Leuk. Lymphoma
2006
, vol. 
47
 (pg. 
2365
-
2370
)
Pui
 
C. H.
Sandlund
 
J. T.
Pei
 
D.
Rivera
 
G. K.
Howard
 
S. C.
Ribeiro
 
R. C.
Rubnitz
 
J. E.
Razzouk
 
B. I.
Hudson
 
M. M.
Cheng
 
C.
, et al 
Results of therapy for acute lymphoblastic leukemia in black and white children
JAMA, J. Am. Med. Assoc.
2003
, vol. 
290
 (pg. 
2001
-
2007
)
Silverman
 
L. B.
Gelber
 
R. D.
Dalton
 
V. K.
Asselin
 
B. L.
Barr
 
R. D.
Clavell
 
L. A.
Hurwitz
 
C. A.
Moghrabi
 
A.
Samson
 
Y.
Schorin
 
M. A.
, et al 
Improved outcome for children with acute lymphoblastic leukemia: results of Dana-Farber Consortium Protocol 91–01
Blood
2001
, vol. 
97
 (pg. 
1211
-
1218
)
Kamsa-ard
 
S.
Wiangon
 
S.
Kamsa-ard
 
S.
Suwanrungruang
 
K.
Jetsrisuparb
 
A.
Horsith
 
S.
 
Trends in incidence of childhood leukemia, Khon Kaen, Thailand, 1985–2002
Asian Pac. J. Cancer Prev.
2006
, vol. 
7
 (pg. 
75
-
78
)
Okamoto
 
T.
Ohno
 
Y.
Tsugane
 
S.
Watanabe
 
S.
Shimoyama
 
M.
Tajima
 
K.
Miwa
 
M.
Shimotohno
 
K.
 
Multi-step carcinogenesis model for adult T-cell leukemia
Jpn. J. Cancer Res.
1989
, vol. 
80
 (pg. 
191
-
195
)
Steffen
 
C.
Auclerc
 
M. F.
Auvrignon
 
A.
Baruchel
 
A.
Kebaili
 
K.
Lambilliotte
 
A.
Leverger
 
G.
Sommelet
 
D.
Vilmer
 
E.
Hémon
 
D.
Clavel
 
J.
 
Acute childhood leukemia and environmental exposure to potential sources of benzene and other hydrocarbons; a case control study
Occup. Environ. Med.
2004
, vol. 
61
 (pg. 
773
-
778
)
Alexander
 
F. E.
Patheal
 
S. L.
Biondi
 
A.
Brandalise
 
S.
Cabrera
 
M. E.
Chan
 
L. C.
Chen
 
Z.
Cimino
 
G.
Cordoba
 
J. C.
Gu
 
L. J.
, et al 
Transplacental chemical exposure and risk of infant leukemia with MLL gene fusion
Cancer Res.
2001
, vol. 
15
 (pg. 
2542
-
2546
)
Borkhardt
 
A.
Wilda
 
M.
Fuchs
 
U.
Gortner
 
L.
Reiss
 
I.
 
Congenital leukaemia after heavy abuse of permethrin during pregnancy
Arch. Dis. Child Fetal Neonatal Ed.
2003
, vol. 
88
 (pg. 
436
-
437
)
Adamson
 
R.
Seiber
 
S.
 
Chemically induced leukemia in humans
Environ. Health Perspect.
1981
, vol. 
39
 (pg. 
93
-
103
)
Infante-Rivard
 
C.
 
Diagnostic X-rays, DNA repair genes and childhood acute lymphoblastic leukemia
Health Phys.
2003
, vol. 
85
 (pg. 
60
-
64
)
Lightfoot
 
T.
Roman
 
E.
 
Causes of childhood leukemia and lymphoma
Toxicol. Appl. Pharmacol.
2004
, vol. 
199
 (pg. 
104
-
117
)
Pui
 
C.
Relling
 
M.
Pharm
 
D.
Downing
 
J.
 
Mechanisms of disease: acute lymphoblastic leukemia
N. Engl. J. Med.
2004
, vol. 
15
 (pg. 
1535
-
1548
)
Perera
 
F. P.
 
Molecular epidemiology: insights into cancer susceptibility, risk, assessment and prevention
J. Natl. Cancer Inst.
1996
, vol. 
88
 (pg. 
496
-
509
)
Krajinovic
 
M.
Labuda
 
D.
Sinnett
 
D.
 
Childhood acute Lymphoblastic leukemia: genetic determinants of susceptibility and disease outcome
Rev. Environ. Health
2001
, vol. 
16
 (pg. 
263
-
279
)
Coggan
 
M.
Whitbread
 
L.
Whittington
 
A.
Board
 
P.
 
Structure and organization of the human θ-class glutathione S-transferase and D-dopachrome tautomerase gene complex
Biochem. J.
1998
, vol. 
334
 (pg. 
617
-
623
)
Landi
 
S.
 
Mammalian class θ GST and differential susceptibility to carcinogens: a review
Mutat. Res.
2000
, vol. 
463
 (pg. 
247
-
283
)
Schneider
 
J.
Bernges
 
U.
Philipp
 
M.
Woitowitz
 
H. J.
 
GSTM1, GSTT1, and GSTP1 polymorphism and lung cancer risk in relation to tobacco smoking
Cancer Lett.
2004
, vol. 
10
 (pg. 
65
-
74
)
Davies
 
S.
Bhatia
 
S.
Ross
 
J. A.
Kiffmeyer
 
W. R.
Gaynon
 
P. S.
Radloff
 
G. A.
Robison
 
L. L.
Perentesis
 
J. P.
 
Glutathione S-transferase genotypes, genetic susceptibility, and outcome of therapy in childhood acute lymphoblastic leukemia
Blood
2002
, vol. 
100
 (pg. 
67
-
71
)
Krajinovic
 
M.
Labuda
 
D.
Richer
 
C.
Karimi
 
S.
Sinnett
 
D.
 
Susceptibility to childhood acute lymphoblastic leukemia: influence of CYP1A1, CYP2D6, GSTM1, and GSTT1 genetic polymorphisms
Blood
1999
, vol. 
93
 (pg. 
1496
-
1501
)
Chen
 
C.
Liu
 
Q.
Pui
 
C.
Rivera
 
G.
Sandlund
 
J.
Ribeiro
 
R.
Evans
 
W.
Relling
 
M.
 
Higher frequency of glutathione S-transferase deletions in black children with acute lymphoblastic leukemia
Blood
1997
, vol. 
89
 (pg. 
1701
-
1707
)
Pakakasama
 
S.
Mukda
 
E.
Sasanakul
 
W.
Kadegasem
 
P.
Udomsubpayakul
 
U.
Thithapandha
 
A.
Hongeng
 
S.
 
Polymorphisms of drug metabolizing enzymes and risk of childhood acute lymphoblastic leukemia
Am. J. Hematol.
2005
, vol. 
79
 (pg. 
202
-
205
)
Canalle
 
R.
Burim
 
R. V.
Tone
 
L. G.
Takahashi
 
C. S.
 
Genetic polymorphisms and susceptibility to childhood acute lymphoblastic leukemia
Environ Mol. Mutagen.
2004
, vol. 
43
 (pg. 
100
-
109
)
Naoe
 
T.
Takeyama
 
K.
Yokozawa
 
T.
Kiyoi
 
H.
Seto
 
M.
Uike
 
N.
Ino
 
T.
Utsunomiya
 
A.
Maruta
 
A.
Jin-nai
 
I.
, et al 
Analysis of genetic polymorphism in NQO1, GST-M1, GST-T1, and CYP3A4 in 469 Japanese patients with therapy-related leukemia/myelodysplastic syndrome and de novo acute myeloid leukemia
Clin. Cancer Res.
2000
, vol. 
6
 (pg. 
4091
-
4095
)
Davies
 
S.
Robinson
 
L.
Buckley
 
J.
Radloff
 
G.
Ross
 
J.
Parentesis
 
J.
 
Glutathione S-transferase polymorphisms in children with myeloid leukemia: a children's cancer group study
Cancer Epidemiol. Biomarkers Prev.
2000
, vol. 
9
 (pg. 
563
-
566
)
Hishida
 
A.
Terakura
 
S.
Emi
 
N.
Yamamoto
 
K.
Murata
 
M.
Nishio
 
K.
Sekido
 
Y.
Niwa
 
T.
Hamajima
 
N.
Naoe
 
T.
 
GSTT1 and GSTM1 deletions, NQO1 C609T polymorphism and risk of chronic myelogenous leukemia in Japanese
Asian Pac. J. Cancer Prev.
2005
, vol. 
6
 (pg. 
251
-
255
)
Niwa
 
Y.
Hirose
 
K.
Nakanishi
 
T.
Nawa
 
A.
Kuzuya
 
K.
Kazuo
 
T.
Hamajima
 
N.
 
Association of the NAD(P)H:quinone oxidoreductase C609T polymorphism and the risk of cervical cancer in Japanese subjects
Gynecol. Oncol.
2005
, vol. 
96
 (pg. 
423
-
429
)
Sunaga
 
N.
Kohno
 
T.
Yanagitani
 
N.
Sugimura
 
H.
Kunitoh
 
H.
Tamura
 
T.
Takei
 
Y.
Tsuchiya
 
S.
Saito
 
R.
Yokota
 
J.
 
Contribution of the NQO1 and GSTT1 polymorphisms to lung adenocarcinoma susceptibility
Cancer Epidemiol. Biomarkers Prev.
2002
, vol. 
11
 (pg. 
730
-
738
)
Nakazato
 
H.
Suzuki
 
K.
Matsui
 
H.
Koike
 
H.
Okugi
 
H.
Ohtake
 
N.
Takei
 
T.
Nakata
 
S.
Hasumi
 
M.
Ito
 
K.
, et al 
Association of genetic polymorphisms of glutathione S-transferase genes (GSTM1, GSTT1 and GSTP1) with familial prostate cancer risk in a Japanese population
Anticancer Res.
2003
, vol. 
23
 (pg. 
2897
-
2902
)
Habdous
 
M.
Siest
 
G.
Herbeth
 
B.
Vincent-Viry
 
M.
Visvikis
 
S.
 
Glutathione S-transferases genetic polymorphisms and human diseases: overview of epidemiological studies
Ann. Biol. Clin. (Paris)
2004
, vol. 
62
 (pg. 
15
-
24
)
Mitrunen
 
K.
Hirvonen
 
A.
 
Molecular epidemiology of sporadic breast cancer: the role of polymorphic genes involved in oestrogen biosynthesis and metabolism
Mutat. Res.
2003
, vol. 
544
 (pg. 
9
-
41
)
Hamajima
 
N.
 
PCR-CTPP: A new genotyping technique in the era of genetic epidemiology
Expert Rev. Mol. Diagn.
2001
, vol. 
1
 (pg. 
119
-
123
)
Nelson
 
H. H.
Wiencke
 
J. K.
Christiani
 
D. C.
Cheng
 
T. J.
Zuo
 
Z. F.
Schwartz
 
B. S.
Lee
 
B. K.
Spitz
 
M. R.
Wang
 
M.
Xu
 
X.
Kelsey
 
K. T.
 
Ethnic differences in the prevalence of the homozygous deleted genotype of glutathione S-transferase θ
Carcinogenesis
1995
, vol. 
16
 (pg. 
1243
-
1245
)
Belinsky
 
M.
Jaiswal
 
A. K.
 
NAD(P)H:quinone oxidoreductase1 (DT-diaphorase) expression in normal and tumor tissues
Cancer Metastasis Rev.
1993
, vol. 
12
 (pg. 
103
-
117
)
Ross
 
D.
Kepa
 
J. K.
Winski
 
S. L.
Beall
 
H. D.
Anwar
 
A.
Siegel
 
D.
 
NAD(P)H:quinine oxidoreductase 1 (NQO1): chemoprotection, bioactivation, gene regulation and genetic polymorphisms
Chem. Biol. Interact.
2000
, vol. 
129
 (pg. 
77
-
97
)
Ross
 
D.
 
Quinone reductases multitasking in the metabolic world
Drug Metab Rev.
2004
, vol. 
36
 (pg. 
639
-
654
)
Siegel
 
D.
Gustafson
 
D.
Dehn
 
D.
Han
 
J.
Boonchoong
 
P.
Berliner
 
L.
Ross
 
D.
 
NAD(P)H:quinone oxidoreductase 1: role as a superoxide scavenger
Mol. Pharmacol.
2004
, vol. 
65
 (pg. 
1238
-
1247
)
Asher
 
G.
Lotem
 
J.
Sachs
 
L.
Kahana
 
C.
Shanl
 
Y.
 
Mdm-2 and ubiquitin-independent p53 proteosomal degradation regulated by NQO1
Proc. Natl. Acad. Sci. U.S.A.
2002
, vol. 
99
 (pg. 
13125
-
13130
)
Nebert
 
D. W.
Roe
 
A. L.
Vandale
 
S. E.
Bingham
 
E.
Oakley
 
G. G.
 
NAD(P)H:quinone oxidoreductase (NQO1) polymorphism, exposure to benzene, and predisposition to disease: a HuGE review
Genet. Med.
2002
, vol. 
4
 (pg. 
62
-
70
)
Traver
 
R. D.
Siegel
 
D.
Beall
 
H. D.
Phillips
 
R. M.
Gibson
 
N. W.
Franklin
 
W. A.
Ross
 
D.
 
Characterization of a polymorphism in NAD(P)H:quinone oxidoreductase (DT-diaphorase)
Br. J. Cancer
1997
, vol. 
75
 (pg. 
69
-
75
)
Lanciotti
 
M.
Dufour
 
C.
Corral
 
L.
Di Michele
 
P.
Pigullo
 
S.
De Rossi
 
G.
Basso
 
G.
Leszl
 
A.
Luciani
 
M.
Lo Nigro
 
L.
, et al 
Genetic polymorphism of NAD(P)H:quinone oxidoreductase is associated with an increased risk of infant acute lymphoblastic leukemia without MLL gene rearrangements
Leukemia
2005
, vol. 
19
 (pg. 
214
-
216
)
Krajinovic
 
M.
Sinnett
 
H.
Richer
 
C.
Labuda
 
D.
Sinnett
 
D.
 
Role of NQO1, MPO and CYP2E1 genetic polymorphisms in the susceptibility to childhood acute lymphoblastic leukemia
Int. J. Cancer
2002
, vol. 
97
 (pg. 
230
-
236
)
Larson
 
R.
Wang
 
Y.
Banerjee
 
M.
Wiemels
 
J.
Hartford
 
C.
Le Beau
 
M. M.
Smith
 
M. T.
 
Prevalence of the inactivating 609 C-T polymorphism in the NAD(P)H:quinone oxidoreductase (NQO1) gene in patients with primary and therapy-related myeloid leukemia
Blood
1999
, vol. 
2
 (pg. 
803
-
807
)
Kelsey
 
K. T.
Ross
 
D.
Traver
 
R. D.
Christiani
 
D. C.
Zuo
 
Z-F.
Spitz
 
M. R.
Wang
 
M.
Xu
 
X.
Lee
 
B. K.
Schwartz
 
B. S.
Wiencke
 
J. K.
 
Ethnic variation in the prevalence of a common NAD(P)H:quinone oxidoreductase polymorphism and its implications for anti-cancer chemotherapy
Br. J. Cancer
1997
, vol. 
76
 (pg. 
852
-
854
)
Laudico
 
A.
Esteban
 
D.
Reyes
 
L.
Liquido
 
J.
 
Philippine Cancer Facts and Estimates
1998
Manila
Phillipine Cancer Society
Arand
 
M.
Mühlbauer
 
R.
Hengstler
 
J.
Jäger
 
E.
Fuchs
 
J.
Winkler
 
L.
Oesch
 
F.
 
A multiplex polymerase chain reaction protocol for the simultaneous analysis of the glutathione S-transferase GSTM1 and GSTT1 polymorphisms
Anal. Biochem.
1996
, vol. 
236
 (pg. 
184
-
186
)
Wan
 
J.
Shi
 
J.
Hui
 
L.
Wu
 
D.
Jin
 
X.
Zhao
 
N.
Huang
 
W.
Xia
 
Z.
Hu
 
G.
 
Association of genetic polymorphisms in CYP2E1, MPO, NQO1, GSTM1, and GSTT1 genes with benzene poisoning
Environ. Health Perspect.
2002
, vol. 
110
 (pg. 
1213
-
1218
)
Kawase
 
H.
Hamajima
 
N.
Tamakoshi
 
A.
Wakai
 
K.
Saito
 
T.
Tajima
 
K.
 
Triplex polymerase chain reactions with confronting two-pair primers (PCR-CTPP) for NQO1 C609T, GSTM1 and GSTT1 polymorphisms: a convenient genotyping method
Asian Pac. J. Cancer Prev.
2003
, vol. 
4
 (pg. 
67
-
70
)
Freedman
 
M. L.
Reich
 
D.
Penney
 
K. L.
McDonald
 
G. J.
Mignault
 
A. A.
Patterson
 
N.
Gabriel
 
S. B.
Topol
 
E. J.
Smoller
 
J. W.
Pato
 
C. N.
, et al 
Assessing the impact of population stratification on genetic association studies
Nat. Genet.
2004
, vol. 
36
 (pg. 
388
-
393
)
Bernardini
 
S.
Hirvonen
 
A.
Ventaus
 
H.
Norppa
 
H.
 
Trans-stilbene oxide-induced sister chromatid exchange in cultured human lymphocytes: influence of GSTM1 and GSTT1 genotypes
Mutagenesis
2001
, vol. 
16
 (pg. 
277
-
281
)
Smith
 
M. T.
Wang
 
Y.
Kane
 
E.
Rollinson
 
S.
Wiemels
 
J. L.
Roman
 
E.
Roddam
 
P.
Cartwright
 
R.
Morgan
 
G.
 
Low NAD(P)H:quinone oxidoreductase 1 activity is associated with increased risk of acute leukemia in adults
Blood
2001
, vol. 
97
 (pg. 
1422
-
1426
)
Chen
 
H.
Lum
 
A.
Seifried
 
A.
Wilkens
 
L.
Marchand
 
L.
 
Association of the NAD(P)H: quinone oxidoreductase 609 C3T polymorphism with a decreased lung cancer risk
Cancer Res.
1999
, vol. 
159
 (pg. 
3045
-
3048
)
Sugimura
 
T.
Terada
 
M.
 
Experimental chemical carcinogenesis in the stomach and colon
Jpn. J. Clin. Oncol.
1998
, vol. 
28
 (pg. 
163
-
167
)
Liska
 
D.
 
The detoxification enzyme systems
Altern. Med. Rev.
1998
, vol. 
3
 (pg. 
187
-
196
)

Author notes

1

This paper is dedicated to the memory of Dr Gloria Bernas, in appreciation of her contributions to science education and research in the Philippines.