Relationship of tobacco smoking, CYP1A1, GSTM1 gene polymorphism and esophageal cancer in Xi’an

Abstract

AIM: To analyze the association of tobacco smoking, polymorphism of CYP1A1 (7^th exon) and GSTM1 genotype and esophageal cancer(EC) in Xi’an.

METHODS: A hospital based case-control study, with molecular epidemiological method, was carried out. Polymorphism of CYP1A1 and GSTM1 of samples from 127 EC cases and 101 controls were detected by PCR method.

RESULTS: There were no significant difference of age and gender between cases and controls. Tobacco smoking was the main risk factor(OR = 1.97; 95%CI = 1.12-3.48) for EC in Xi’an. The proportions of CYP1A1 Ile/Ile, Ile/Val and Val/Val gene types in cases and controls was 19.7%, 45.7%, 34.6% and 30.7%, 47.5%, 21.8% respectively (P = 0.049).Individuals with CYP1A1 Val/Val genotype compared to those with CYP1A1 Ile/Ile genotype had higher risk for EC increased (OR = 2.48, 95%CI = 1.12-5.54). The proportions of GSTM1 deletion genotype in cases and controls were 58.3% and 43.6% (OR = 1.81, 95%CI = 1.03-3.18, P = 0.028). Analysis of gene-environment interaction showed that tobacco smoking and CYP1A1 Val/Val genotype; tobacco smoking and GSTM1 deletion genotype had synergism interaction respectively. Analysis of gene-gene interaction did not find synergistic interaction between these two genes. But in GSTM1 deletion group,there was significant difference of distribution of CYP1A1 genotype between cases and controls (P = 0.011).

CONCLUSION: CYP1A1 Val/Val and GSTM1 deletion genotypes are genetic susceptibility biomarkers for EC. The risk increases, when person with CYP1A1 Val/Val and/or GSTM1 deletion genotype. And these two-metabolic enzymes seem to have interactions with tobacco smoking, in which the mechanism still needs further study.

Key Words: $[Keywords]

INTRODUCTION

Esophageal cancer (EC) is one of the most common malignant tumors of human being. The incidence of EC varies in different countries. China is the country with highest incidence and mortality rate of EC. Research showed that risks for EC in different countries or different places were different[1-6]. In western countries alcohol intake and tobacco smoking were studied deeply[7-12]. It was thought that besides tobacco smoking and alcohol drinking, nutrition factors, life style, viruses infection, heredity or exposure to nitrosamines, fungi or AFB1 maybe involved in the process of EC[1,3,13-19]. In China, researches showed risks for EC were different in areas with different incidence[1-5,16,18,20,21]. The mortality rate of EC of Xi’an city in Shaanxi province is about 24 per 100000, which ranks first in all cancer mortalities. Previous studies showed that both of tobacco smoking and family history of EC were main risk factors for EC in Xi’an city[2,22,23].

EC is a multi-etiology disease; environmental risks exposures and genetic susceptibility may take the role part[2,22-24]. Almost all of the environmental carcinogens (procarcinogens) are activated to be ultimate carcinogens before initiate the process of carcinogenesis. Some metabolic enzymes are closely related to the activation and detoxification of procarcinogens. Alterations of the key oncogene or tumor suppress gene can disturb the cycle of cell proliferation, which can also initiate the process of carcinogenesis[23,25,26]. Susceptibility of cancer is associated with the genetics polymorphism of related metabolic enzymes. Both certain susceptibility related biomarkers and certain environmental carcinogens perhaps are indispensable factors for EC[20,23,27,28]. To explore the bio-basis of genetic susceptibility of EC in Xi’an, we carried out a hospital based case- control study to analyze the associations of tobacco smoking, CYP1A1, GSTM1 gene polymorphism and EC.

MATERIALS AND METNODS

Seletion of patients and controls

All cases with esophageal cancer (confirmed by pathological diagnosis) came from inpatients of Tandu Hospital during half a year period (December, 1999 to April, 2000). All controls were stratified randomly selected from non-cancer inpatients from different department of the same hospital during the same period. Both cases and controls were confined to residents with long-term living in Xi’an (with similar proportion of gender and age).

Collected data

Trained interviewers using a structured questionnaire interviewed cases and controls in the hospital. The questionnaire obtained detailed information on residence, occupation, tobacco smoking habit and so on. Here tobacco smoking was defined as smoking at least one cigarette per day and persisting for more than one year. 127 cases (male 97, female 30) and 101 (male 78, female 23) controls were included. Blood samples were also collected for extraction of DNA genome. All blood samples had been stored at -70 °C before started DNA extraction.

PCR methods to detect polymorphism of CYP1A1 and GSTM1

Digested by Proteinase K, DNA genomes were extracted from blood clot of cases and controls with hydroxybenzene, chloroform method in a uninterrupted period. CYP1A1 and GSTM1 polymorphisms were identified by polymerase chain reaction (PCR) before which DNA samples were stored at 4 °C. Primers for GSTM1(P1: 5’-GTACCCTACTTGATTGATGGG-3’; P2: 5’-CTGGATTGTAGCAGATCATGC-3’) and for CYP1A1 (P3: 5’-CGGAAGTGTATCGGTGAGACCA-3’ P4: 5’-CGGAAGTGTATCGGTGAGACCG-3’; P5: 5’-GTAGACAGAGTCTAGGCCTCA-3’) were synthesized by Shenggong bio-technical company of Shanghai. PCR condition for GSTM1 as follows, 50 µL solution including 10 × buffer 5 µL, Mg²⁺ 2 µL, P1, P2 1 µL respectivly, template DNA 1.5 µL, dDNTPs 1 µL and Taq DNA polymerase 3^U. After denaturation at 94 °C for 10 min, Taq DNA polymerase was added,followed by 30 cycles with 94 °C 1min, 60 °C 1 min, and 72 °C 1 min. 20 g·L^-1 agar was used to electrophoresis PCR production, then observed under the violate light. GSTM1 exist genotype was characterized as had a 273 bp fragment; while GSTM1 deletion genotype had no fragment (Figure 1).

Figure 1 Identify the GSTM1 genotype. M:100 bp DNA ladder, 3,4 were GSTM1 exist 2,5 were GSTM1 deletion; 1 positive control, 6 negative control, 7 was blank control (without DNA template)

We used two pairs of primers to detect the polymorphism of CYP1A1 (7^th exon). For each DNA sample two sets of PCR were carried out using P3, P5 (marked as tube A) and P4, P5 (marked as tube B) respectively. PCR conditions were the same:50 µL solution including 10 × buffer 5 µL, Mg²⁺ 2 µL, P3,P5 (or P4, P5) 1 µL, template DNA 1.5 µL, dDNTPs 1 µL and Taq DNA polymerase 3^U, then 94 °C 10 min followed by 94 °C 1 min, 55 °C 1 min, 72 °C 1 min, 35 cycles, 72 °C extending 10 min. PCR products were observed. The PCR was conducted to detect the mutation of A-G in CYP1A1 7^th exon, the mutation can leads to change of one amino acid (Ile to Val). If there was only tube A had the specifically fragment (200 bp), the DNA was regarded as CYP1A1 Ile/Ile genotype (pure wild genotype); if only tube B had the positive fragment, CYP1A1 Val/Val genotype (pure mutation) was considered; and CYP1A1 Ile/Val genotype was identified with both tube A and tube B had the fragment (Figure 2).

Figure 2 Identified the genotypes of CYP1A1. M: 100 bp DNA ladder; 1(A),2(B) represent Ile/Ile genotype; 3(A), 4(B) represent Ile/Val genotype; 5(A), 6(B) represent Val/Val genotype; 7(A),8(B) as blank control( without DNA template).

Quality control

DNA extraction and PCR were conducted in different period and places. The genotypes of DNA samples were identified blindly. Every PCR had were set controls as blanket control (without DNA template), positive control and negative control, and when any one of these controls was failure, PCR wasre-conducted.

Statistical analysis

Data were input into computer, then the values of χ², odds ratio (OR) and 95%CI (confidence intermediate) were calculated. And ORs of gene-environment and gene-gene interaction were also estimated.

RESULTS

Comparability between cases and controls

The age and gender in cases and controls were comparable(Table 1).

Table 1 Comparability of age and gender in cases and controls.

Factor	Case	Control	χ2	P
Age(year)
< 50	28	16
50-	38	44
≥ 60	61	41	4.73	0.094
Gender
Male	97	78
Female	30	23	0.02	0.80

Risk factors for EC

The proportions of tobacco smoking, GSTM1 deletion genotype and CYP1A1 genotype (Val/Val) in cases and controls were significantly different (P < 0.05) (Table 2).

Table 2 Distributions of smoking, GSTM1 deletion and CYP1A1 genotypes.

Factors	Case	Control	OR	OR95%CI	χ2	P
Smoking Yes	69	38	1.97	1.12-3.48	6.28	0.012
No	58	63
GSTM1
Deletion	74	44	1.81	1.03-3.18	4.85	0.028
Exist	53	57
CYP1A1
Ile/Ile	25	31	1.00		1.00
Ile/Val	58	48	1.50	0.74-3.03	1.48	0.22
Val/Val	44	22	2.48	1.12-5.54	5.93	0.015

Interaction of tobacco smoking and GSTM1 deletion genotype or CYP1A1 Val/Val genotypes

Analysis showed that there was synergistic interaction between tobacco smoking and GSTM1 deletion genotype (Table 3). Tobacco smoking and CYP1A1 Val/Val genotype also appeared synergistic interaction(Table 4). But CYP1A1 mutation genotypes(Val/Val, Ile/Val) and GSTM1deletion genotype did not show significant interaction (Table 5). OR of individuals with CYP1A1 mutation genotype and GSTM1deletion genotype was greater than those with of any other forms of the two genotypes. But there did not show any synergistic interaction between CYP1A1 mutation genotypes and GSTM1deletion genotype.

Table 3 Interaction of smoking and GSTM1 deletion genotype.

Smoking	GSTM1 deletion	Case	Control	OR	OR95%CI	χ2	P
No	No	25	37	1.00
Yes	No	28	20	2.07	0.90-4.80	3.48	0.062
No	Yes	33	26	1.88	0.86-4.13	2.93	0.087
Yes	Yes	41	18	3.37	1.49-7.69	10.29	0.0013

SIA = 3.37/(1.88+2.07-1.00) = 1.14

Table 4 Interaction of tobacco smoking and CYP1A1 Val/Val genotype.

Smoking	CYP1A1(Val/Val)	Case	Control	OR	OR95%CI	χ2	P
No	No	36	47	1.00
Yes	No	47	32	1.92	0.98-3.76	4.18	0.04
No	Yes	22	16	1.80	0.77-4.20	2.18	0.14
Yes	Yes	22	6	4.79	1.62-14.83	10.30	0.0013

SIM = 4.79 / (1.92 × 1.80) = 1.39

Table 5 Interaction of CYP1A1 mutation genotypes(Val/Val, Ile/Val) and GSTM1 deletion genotype.

GSTM1 deletion	CYP1A1 mutation	Case	Control	OR	OR95%CI	χ2	P
No	No	10	20	1.00
No	Yes	43	37	2.32	0.89-6.14	3.61	0.057
Yes	No	15	11	2.73	0.81-9.42	3.28	0.070
Yes	Yes	59	33	3.58	1.39-9.38	8.66	0.0033

Stratified with GSTM1 deletion genotype to analyze the distributions of CYP1A1 genotypes in cases and controls. Results showed that there were significant different in cases and controls (P = 0.011) in GSTM1 exist genotype CYP1A1 genotypes, whereas there were no significant difference (P = 0.83) between cases and controls in GSTM1 deletion genotype (Table 6).

Table 6 Analysis of CYP1A1 genotypes in cases and controls with Stratified GSTM1 deletion.

CYP1A1 genotype	GSTM1 deletion		existing GSTM1
	Case	Control	Case	Control
Ile/Ile	15	11	10	20
Ile/Val	35	19	23	29
Val/Val	24	14	20	8
χ2	0.39		9.04
P	0.83		0.011

DISCUSSION

Under similar environmental carcinogens exposure only a few of individuals get neoplasm, for there were individual difference to environmental exposure. The different liability to cancer was called genetic susceptibility of cancer. Genetic susceptibility can affect on every step of carcinogenesis, including modify the effect of environmental carcinogens[24,29-35].Oncogenes and tumor suppressor genes can also affect individual’s susceptibility to cancer.Cancer susceptibility genes includes typeI, typeII metabolism enzyme gene,DNA repair gene and those affect cell proliferation rate gene. In recent years evidence has accumulated to support the hypothesis that cancer susceptibility gene may be of importance in determining individual susceptibility to cancer[34,36-46].

EC is a multi-factor determined disease; including environmental risk factors and genetic factors. In recent years, more and more researches considered environmental and genetic susceptibility factors and their interactions in evaluating the risks of cancer[2,17,43,47-50]. Investigations showed the mortality rate of EC in Shaanxi province did not decreased during the late 20 years, and risks factors for EC in Xi’an city were discussed in several researches[2,22,23]. In this hospital based case- control study, the results showed that tobacco smoking was a risk factor; and tobacco smoking had interactions with GSTM1 deletion genotype and CYP1A1 Val/ Val genotype.

Most chemical carcinogens in environment are pro-carcinogens. And aromatic hydrocarbons (AHs) in tobacco smoking are pro-carcinogens, they need to be activated to reactive electrophilic forms by type ¢ñ metabolic enzymes (CYP450s), then initiate the carcinogenesis. On the other hand the reactive electrophilic forms of carcinogen can be detoxified and excreted by type I metabolic enzymes such as GSTM1. Although theoretically the increase of activity of type II metabolic enzymes and/or decrease of activity of type I metabolic enzymes can increase the risk for cancer, there were different results in different researches, some supported this hypothesis and others did not[27,34,35,37,40-42,51-56]. Our results showed that individuals with the GSTM1 deletion genotype or/and CYP1A1 Val/Val genotype had increased risks for EC.

P450 CYP1A1 gene located in chromosome 15q22 mainly metabolizes pro-carcinogens. There are three kinds of polymorphism of CYP1A1: MspI site, 7^th exon (Ile-Val) and AA polymorphism . MspI polymorphism include three genotypes: without MspI enzyme cleavage site allele gene m1 (m1/m1) as A genotype;having MspI cleavage site allele (m2/m2) as C genotype and m1/m2 as B genotype. In different populations the distribution of these three genotypes were different. CYP1A1 Ile-Val polymorphism caused by 7^th exon 4889^th base difference (A or G), transition of A to G results in 462^th amino turned from isoleucine to valine[13],then form three kinds of genotypes: homozygote wild genotype (Ile/Ile), mutation genotype(Val/Val) and heterozygote Ile/Val genotype. Polymorphism of 7^th exon correlated with polymorphism of MspI in Asia and Caucasian populations, and in Americans from Africa these two kind of CYP1A1 polymorphism were independent, CYP1A1 7^th exon polymorphism and Msp I site were incomplete linkage. Research showed CYP1A1 Val/Val genotype have higher ability to activate pro-carcinogen than CYP1A1 Ile/Ile genotype. PAH-DNA adducts in leukocyte were higher in heavy smoking population with CYP1A1 Val/Val genotype than those with CYP1A1 Ilel/Val or Ile/Ile genotype. AA polymorphism was new special MspI polymorphism, which still under discussion.

Although evidence showed that CYP1A1 mutation genotype (Val/Val) had the strongest ability to activate pro-carcinogens, the associations between CYP1A1 genotype and susceptibility to cancers were varied[30-33,37,57,58]. Data from Guandong province in China showed that MspI C correlated with no-smoking population’s lung cancer susceptibility[52]. Study in Shanghai and Haerbin no significant relation was discovered between CYP1A1 (Ile-Val) polymorphism and lung cancer susceptibility in non-smoking female patients[51].CYP1A1 Val/Val genotype only appear about 3.2%-5% in white population,while in Japanese it was about 19.8%, in Chinese it was 22.3%. Our study showed that distributions of CYP1A1 genotypes in cases and controls were different (P = 0.049), CYP1A1 Val/Val genotype was associated with EC (OR = 2.48, 95%CI = 1.12-5.54) and there was interaction of tobacco smoking and CYP1A1 Val/Val genotype.

GSTM1 can detoxify a number of reactive electrophilic compound substances, including the carcinogens PAHs. If individuals with GSTM1 deletion genotype, the ability of detoxify the carcinogens decreased. Individuals with GSTM1 deletion can have the increased risk of cancers[24,43,46]. In China there were similar research on GSTM1 deletion genotype and the risks of lung cancer(OR = 2.56)[53],and stomach cancer(OR = 1.90, 95%CI = 1.01-3.56)[54].Researches showed that in Henan province, high incidence of EC in China, GSTM1 deletion gene polymorphisms had not significant relation with EC susceptibility[20]. Results of our study indicated GSTM1 deletion genotype was significant different in cases and controls (P = 0.028) and the OR was 1.81(95%CI = 1.03-3.18). GSTM1 deletion genotype had synergistic interaction with tobacco smoking.

In summary, we found tobacco smoking, CYP1A1 Val/Val genotype; GSTM1 deletion genotype had associations with EC in Xi’an area. Gene-environment interaction analysis showed that tobacco smoking had synergistic interactions with CYP1A1 Val/Val genotype, and with GSTM1 deletion genotype. Gene-gene interaction analysis did not find synergistic interaction between CYP1A1 mutation genotypes and GSTM1 deletion genotype, though individuals with these two genotypes had increased risk for EC. The synergistic interactions and their mechanisms of tobacco smoking with these two metabolic enzymes gene polymorphisms still need further study with large (population-based) samples and modified designs.

ACKNOLEDGMENTS

We would like to thank Bing-Quan Gu for their help in collecting blood sample.