Susceptibility to oral squamous cell carcinoma: correlation with variants of CYP1A1-MspI, GSTT1, GSTM1, ALDH2, EC-SOD and Lifestyle factors

This prospective study included 750 patients who were admitted to our hospital from June 2011 to May 2015. Another 750 participants who received physical examinations during the same time period were selected as healthy controls; the physical examination showed no cancer or hereditary diseases. There was no statistically significant difference in age, gender, place of origin or nationality, and the subjects were unrelated. Participant demographic data, smoking history, alcohol drinking history, occupational history and family tumor history were collected. The smoking status was evaluated using the smoking index (SI) that was the product of the daily number of cigarettes multiplied by number of years of smoking. Based on SI values, the participants were divided into the following categories: non smokers, individuals with SI ≤400, and individuals with SI 400. Alcohol consumption was evaluated with the drinking index (DI) that was the product of the daily amount of drinking (in grams) multiplied by the number of years of drinking. Based on DI values, the participants were classified as non drinkers, those with DI ≤3000, and those with DI 3000. The study was approved by the General Hospital of Daqing Oil Field, Daqing, Heilongjiang Province, People’s Republic of China (PRC). Informed written consent was obtained from all participants.

From each participant, 3 mL blood was collected in vacutainers with EDTA as anticoagulant. The QIAmpDNA extraction kit (Qiagen GmbH, Hilden, Germany) was used to extract DNA from white blood cells. Extracted DNA was stored at –30 °C. Polymerase chain reaction (PCR) was used to amplify the DNA to the levels required for restriction fragment length polymorphism (RFLP) analysis. For all reactions, a total volume of 25 μL comprised 2.5 μL 10 × buffer, 2.5 μL dNTP, 20 pmol upstream primers, 20 pmol downstream primers (see below), 0.75 μL Taq DNA polymerase (all PCR reagents from Promega, Madison, WI, USA), and 100 ng template DNA. Reactions were performed on PE480 thermocycler (Perkin Elmer?, Norwalk, CT, USA), as follows: initial denaturation at 94 °C for 4 min., and 35 cycles of denaturation at 94 °C for 40 seconds, annealing at 55 °C for 40 seconds, and extension at 72 °C for 50 seconds, and final extension at 72 °C for 5 min. The PCR products were digested with restriction endonucleases as appropriate (described below). A total reaction volume of 20 μL comprised 1 ng PCR product, 2 μL 10 × NEB (New England Biolabs, Ipswich, MA, USA) reaction buffer, and 10 U endonuclease; reactions were performed at 37 °C for 3 hours. Digestion products were separated by 100V electrophoresis on a 3.0% agarose gel, for 1 hour. After 30 min. in ethidium bromide, bands were detected by ultraviolet light.

The CYP1A1-MspI polymorphism was detected using the following primer sequences: upstream primer (5’-CAG TGA AGA GGT GTA GCC GCT-3’), downstream primer (5’-TAG GAG TCT TGT CTC ATG CCT-3’ (synthesized by TaKaRa, Dalian, China). Restriction digestion with MspI produced three genotypes: wild-type (m1/m1) with a band at 340 bp, heterozygous (m1/m2) with bands at 340, 200 and 140 bp, and homozygous mutant (m2/m2) with bands at 200 and 140 bp.

The EC-SOD polymorphism was detected using primer sequences as reported in a previous study [7]: upstream primer (5’-GCA ACC AGG CCA GCG TGG AGA ACG GGA A-3’), and downstream primer 5’-CCA GAG GAG AAG CTC AAA GGC AGA-3’). The PCR product was digested with restriction endonuclease PauI. Two genotypes were produced: homozygous C/C with bands at 111 and 109 bp; and heterozygous C/G with bands at 220, 111 and 109 bp.

The GSTT1 polymorphism was detected using primer sequences according to Wilson et al. [8]: upstream primer (5’-TCT CCT TAC TGG TCC TCA CAT CTC-3’), and downstream primer (5’-TCA CCG GAT CAT GGC CAG CA-3’). GSTT1 positive [+] was indicated by the presence of a 480 bp fragment after PCR amplification; GSTT1 negative [–] meant lack of PCR product. The control β-globin gene was detected with upstream primer (5’-CAA GAG CCA ACC ACA GGT AC-3’) and downstream primer (5’-GAA GAG CCA AGG ACA GGT AC-3’).

The GSTM1 polymorphism was detected using primer sequences selected according to Nguyen et al. [9]: upstream primer (5’-GAA CTC CCT GAA AAG CTA AAG C-3’), and downstream primer (5’-GTT GGG CTC AAA TAT ACG GTG G-3’). GSTM1 positive [+] indicated the presence of bands at 230 and 219 bp; GSTM1 negative [–] indicated the presence of only one 219 bp band. The control β-globin gene was detected with upstream primer (5’-CAA GAG CCA ACC ACA GGT AC-3’), and downstream primer (5’-GAA GAG CCA AGG ACA GGT AC-3’).

The ALDH2 polymorphism was detected using primer sequences selected according to Ishibashi et al. [10]: upstream primer (5’-CCC TTT GGT GGC TAG AAG ATG-3’), and downstream primer (5’-CCA CAC TCA CAG TTT TCT CTT-3’). The PCR products were digested with the restriction endonuclease MboI. The amplified fragment was 91 bp and three genotypes could be seen after digestion: homozygous G/G with a band at 55 bp, heterozygous G/L with bands at 65 and 55 bp, and homozygous L/L with a band at 65 bp. The ALDH2 (G/L) and ALDH2 (L/L) were combined and marked as ALDH2 (non G/G) for co-analysis.

The difference in genotype distributions between the patient and control groups was determined using the χ² test. A non conditional logistic regression model was used to analyze the adjusted odds ratio (OR) for risk of oral squamous cell carcinoma with different genotypes, as well as the combined effects between smoking, drinking, and genotype. The 95% confidence interval (CI) is also reported. The Statistical Package for the Social Sciences (SPSS Inc., Chicago, IL, USA) version 11.0 was used for statistical analysis. Statistical significance was accepted at a p value of <0.05.

General demographic and clinical information for participants in both groups are provided in Table 1. There was no statistically significant difference in gender or age between case and control groups (p >0.05). However, smoking and drinking were significantly more common in individuals with OSCC than in controls (p <0.01).

Table 1

General characteristics of healthy controls and patients with oral squamous cell carcinoma.

For each of the enzyme encoding genes we investigated, the distribution of genotypes, significantly differed between cases and controls (all p <0.01). CYP1A1-MspI (m2/m2), EC-SOD (C/G), GSTT1 [–], GSTM1 [–] and ALDH2 (non G/G) genotypes were significantly more common in individuals with OSCC than in control individuals (p <0.01) (Table 2).

Table 2

Distribution of CYP1A1-MspI, EC-SOD, GSTT1, GSTM1, and ALHD2 genotypes in the control and patient groups.

Because there was a significant shift of the genotype frequencies observed in individuals with OSCC compared to controls, we also examined whether OSCC patients were more likely to carry multiple variant genotypes than healthy subjects. The combined distribution of CYP1A1-MspI (m2/ m2), EC-SOD (C/G), GSTT1 [–], GSTM1 [–] and ALHD2 (non G/G) genotypes in both groups are provided in Table 3. Within the patient group, the majority of individuals (90.4%) had at least two of the variants. Namely, 1.9% of individuals were positive for five of the variant genotypes, 20.8% for four genotypes, 37.1% for three genotypes, 30.7% for two genotypes, 8.5% for only one variant genotype, and only 1.1% did not have any of the variant genotypes. In contrast, in the control group, the majority of individuals had one or no variants (51.6%). Thus, the presence of two or more variants in the OSCC population was significantly more common than in controls (p <0.01).

Table 3

The combined distribution of CYP1A1-MspI (m2/m2), EC-SOD (C/G), GSTT1[–], GSTM1 [–] and ALHD2 (non G/G) variant genotypes in the control and patient groups.

Table 4 depicts the results of the analysis of CYP1A1-MspI (m2/ m2), EC-SOD (C/G), GSTT1 [–], GSTM1 [–] and ALHD2 (non G/G) variant genotype presence and smoking status. In individuals with OSCC, multiple variants and smoking were significantly more common compared to the control group; smoking in combination with multiple variant genotypes had a synergistic effect on cancer likelihood. Smokers with five variant genotypes were 32-times more likely to have OSCC.

Table 4

Combined analysis of smoking status, and variant genotype distribution in relation to oral squamous cell carcinoma susceptibility.

Furthermore, within the sub-population of smokers, SI was synergistic with multiple variant genotypes (Table 5). In the patient group, the presence of multiple variant genotypes and an SI>400 was significantly more common compared to the control group (p <0.01).

Table 5

Combined analysis of smoking index, and variant genotype distribution in relation to oral squamous cell carcinoma susceptibility.

A similar analysis was applied to CYP1A1-MspI (m2/m2), EC-SOD (C/G), GSTT1 [–], GSTM1 [–] and ALHD2 (non G/G) variant genotypes and alcohol consumption (Table 6). The presence of multiple variants and alcohol consumption was significantly more common in patients compared to the control group; furthermore, as with smoking, the drinking behavior and combined variant genotypes had a synergistic effect on the likelihood of cancer.

Table 6

Combined analysis of drinking status, and variant genotype distribution in relation to oral squamous cell carcinoma susceptibility.

The synergistic effect of the DI and combined expression of variant genotypes is depicted in Table 7. In the patient group, the presence of multiple variants and a DI >3000 was significantly more common compared to the control group (p <0.01).

Table 7

Combined analysis of drinking index, and variant genotype distribution in relation to oral squamous cell carcinoma susceptibility.

Table 8 displays the results of the analysis of the presence of variant genotypes combined with smoking and drinking status. In the patient group, there were significantly more patients who had multiple variants and were positive for both smoking and drinking history as compared to the control group.

Table 8

Combined analysis of smoking and drinking status, and variant genotype distribution in relation to oral squamous cell carcinoma susceptibility.