Tumor Biol. DOI 10.1007/s13277-015-3078-y
RESEARCH ARTICLE
Association between ADH1B and ADH1C polymorphisms and the risk of head and neck squamous cell carcinoma Yong Bae Ji & Seung Hwan Lee & Kyung Rae Kim & Chul Won Park & Chang Myeon Song & Byung Lae Park & Hyoung Doo Shin & Kyung Tae
Received: 20 October 2014 / Accepted: 8 January 2015 # International Society of Oncology and BioMarkers (ISOBM) 2015
Abstract Alcohol consumption is one of the major risk factors for head and neck squamous cell carcinoma (HNSCC), and the alcohol dehydrogenase (ADH) family proteins are key enzymes in ethanol metabolism. We examined the associations between single nucleotide polymorphisms (SNPs) of ADH1B and ADH1C and the risk of HNSC C. We analyzed six SNPS of ADH1B, namely −992C>G, −957C > A, +3170A>G, +3377G>T, +3491G>A, and + 13543A>G, and five SNPs of ADH1C, namely −1064C>T, −325G>C, +5702A>G, +7462T>C, and +13044A>G, in 260 Korean HNSCC patients and 330 controls, using single base extension and the TaqMan assay. The odds ratios (ORs) and 95 % confidence intervals (95 % CIs) of the CG and GG genotypes of ADH1B −992C>G, the AA genotype of −957C>A, the GG genotype of +3170A>G, the GA genotype of +3491G>A, and +13543A>G were 0.51 (0.32–0.82), 0.63 (0.42–0.94), 1.84 (1.13–2.99), 1.77 (1.15–2.73), 2.34 (1.44– 3.79), and 2.21 (1.23–3.95), respectively. The ORs of ADH1C +13044A>G were 1.94 (1.01–3.71) and 1.97 (1.05–3.71) in the dominant and co-dominant models, respectively. The ORs of the GC genotype of ADH1C −325G>C and the AG genotype of +5702A>G were 2.52 (1.51–4.21) and 2.43 (1.36– 4.32), respectively. ADH1B +3170A>G and ADH1C + 13044A>G were in strong linkage disequilibrium with the other SNPs of ADH1B and ADH1C, respectively. There were gene-environment interactions between ADH1B +3170A>G Y. B. Ji : S. H. Lee : K. R. Kim : C. W. Park : C. M. Song : K. Tae (*) Department of Otolaryngology-Head and Neck Surgery, College of Medicine, Hanyang University, 222 Wangsimni-ro, Seongdong-gu Seoul 133-792, Korea e-mail: [email protected] B. L. Park : H. D. Shin Department of Genetic Epidemiology, SNP Genetics, Inc., Seoul, Korea
and ADH1C +13044A>G and alcohol consumption and smoking. ADH1B +3170A>G and ADH1C +13044A>G SNPs are associated with an increased risk of HNSCC, and they could be used as biomarkers for the high-risk group of HNSCC in Koreans. Keywords Polymorphism . SNP . ADH1B . ADH1C . Head and neck cancer
Introduction The development of head and neck squamous cell carcinoma (HNSCC) is associated with environmental factors such as chemical carcinogens, ultraviolet exposure, irradiation and viral infection, and host-related factors such as genetic susceptibility [1–3]. Many epidemiologic studies have shown that alcohol consumption is one of causative factors in the development of HNSCC [4]. According to the International Agency for Research on Cancer (IARC), there is adequate evidence that alcohol is a causative factor of malignancies such as oral cancer, pharyngeal cancer, laryngeal cancer, esophageal cancer, hepatoma, colorectal cancer, and breast cancer [5]. Experimental studies have shown that acetaldehyde, a metabolite of ethanol, inhibits DNA synthesis and repair [6]; it binds covalently to DNA, forming a DNA adduct and so causing DNA damage. It has been reported that this adduct is present at higher levels in alcohol drinkers than in nondrinkers [7]. Furthermore, the acetaldehyde concentration is 10–20 times higher in the saliva than in the serum after alcohol consumption. These findings suggest that acetaldehyde is associated with the development of HNSCC in alcohol consumers [8].
Tumor Biol.
The enzyme alcohol dehydrogenase (ADH) is responsible for the conversion of ethanol to acetaldehyde, and there are several ADH isoenzymes that differ in amino acid sequence. The isoenzymes are divided into five categories: class I (ADH1A, ADH1B, and ADH1C), class II (ADH4), class III (ADH5), class IV (ADH7), and class V (ADH6). Three of the class I isoenzymes (ADH1A, ADH1B, and ADH1C) are located close to the 4q21-23 locus [9]. ADH1B and ADH1C are predominantly expressed in the liver and are the main enzymes responsible for ethanol oxidation. Although alcohol consumption is associated with the development of HNSCC, this does not mean that the increased risk of developing HNSCC is strictly proportional to the amount of alcohol consumed because the susceptibility of individuals differs, and genetic polymorphisms of ADH lead to variability in the activity of ADH [10]. Studies of ADH polymorphism have focused mainly on the single nucleotide polymorphisms (SNPs) ADH1B +3170A>G His48Arg (rs1229984) and ADH1C +13044A>G Ile350Val (rs698). According to some studies, these SNPs are associated with an increased risk of developing HNSCC, while other studies have found no association [11–26]. However, there have been only a few studies of ADH1B and ADH1C SNPs other than ADH1B +3170A>G and ADH1C +13044A>G. Therefore, in this study, we analyzed six SNPs of ADH1B, including ADH1B +3170A>G, and five SNPs of ADH1C, including ADH1C +13044A>G, which are known to exist in Koreans, and we evaluated the relationship between these SNPs of and the risk of developing HNSCC.
Materials and methods Patients We conducted a hospital-based case-control study in 590 subjects. The case group consisted of 260 patients who were histopathologically diagnosed with HNSCC between January 2004 and December 2009 in a tertiary hospital. The 330 controls consisted of healthy volunteers who visited the hospital for routine health examinations, and patients admitted to the same hospital for treatment for non-cancerous diseases such as chronic tonsillitis, chronic sinusitis, or chronic otitis media during the same period. The control individuals had no history of previous malignancies or genetic disorders. All the subjects of this study were Korean, and we obtained written informed consent from all the subjects, and the study was approved by the institutional review board of Hanyang University. The demographic and clinical characteristics of the study subjects are presented in Table 1. The mean age was 62.1± 15.7 years (range, 28–90 years) in the HNSCC group and 43.7±10.9 years (range, 21–76 years) in the control group
(PG, and five SNPs of ADH1C, namely ADH1C −1064C>T, −325G>C, +5702A>G, +7462T>C, and +13044A>G. The TaqMan assay was used to genotype ADH1B + 3170A>G. Forward and reverse primer were 5′-TCTTTTCT GAATCTGAACAGCTTCTCTTT-3′ and 5′-GGTCACCA GGTTGCCACTAAC-3′, respectively. The probe for the wild-type A allele was 5′-CTGTAGGAATCTGTCGCACA3′, with the 5′ end and 3′ end tagged with the VIC™ fluorescent dye and the FAM™ quencher dye, respectively. The probe for the mutant-type G allele was 5′-CTGTAGGAAT CTGTCACACA-3′, tagged as for the wild-type probe. Polymerase chain reactions (PCR) were performed in 5 μL reaction mixtures containing TaqMan universal PCR master mix (Applied Biosystems, Foster City, CA, USA), UNG, primer (900 μM), probe (200 nM), and genomic DNA (20 ng). For genotyping the other ten SNPs, we used the single base extension (SBE) technique. PCR reaction mixtures of 5 μL contained primer (1.25 pM), genomic DNA (5 ng), dNTPs (250 M), and 0.15 U Taq DNA polymerase (Applied Biosystems). The reaction conditions were as follows for 40 amplification cycles: 50 °C for 2 min, 95 °C for 10 min followed by 15 s denaturing, then final annealing and extension at 60 °C for 1 min. The TaqMan assay plate was transferred to a Prism 9700HT instrument (Applied Biosystems), and fluorescence levels in each well were measured with SDS software (ver. 2.3., Applied Biosystems). The positive (12
Tumor Biol. Table 1 Clinical characteristics of head and neck squamous cell carcinoma patients and controls (n=590)
Variables
Case
p Value
Control
N
%
N
%
23 86 151
(8.9) (33.1) (58.0)
202 97 31
(61.2) (29.4) (9.4)
Age (year)
a
Less than 20 pack-years
b
20 pack-years or more
c
Consumption less than 120 g of alcohol per week
d
Consumption 120 g of alcohol or more per week
≤45 45–60 ≥60 Smoking status None Former Lighta Heavyb Age started smoking (year) Period of smoking (year) Alcohol consumption None Lightc Heavyd Age started drinking (year) Period of drinking (year) Gender Male Female
samples) and negative (12 blanks) controls were used instead of quantitative RT-PCR. The PCR products and Genescan 120Liz size-standard solution (Applied Biosystems) were placed in highly-deionized (Hi-Di) formamide, and subjected to electrophoresis using an ABI Prism 3100 Genetic Analyzer (Applied Biosystems). Genotyping was done with an ABI Prism GeneScan and Genotyper.
Statistical analysis Student’s t test was used for continuous variables, and the chi-square test and Fisher’s exact test for categorical variables. Linkage disequilibrium (LD) was analyzed using Haploview v4.2 software (http://www.broadinstitute. org/mpg/haploview), examining Lewontin’s D′ (|D′|) and the LD coefficient r 2 between all pairs of bi-allelic loci. Haplotypes were estimated using PHASE v2.0 software from the University of Washington. Odds ratios (OR) and 95 % confidence intervals (CI) was calculated with a logistic regression model using SAS, version 9.2 (SAS Inc., Cary, NC, USA). Analyses of the referent (to the homozygotes of major alleles) and three alternative models (co-dominant, dominant, and recessive for the minor alleles) were used for OR and 95 % CI values. PC, +5702A>G, +7462T>C, and +13044A>G, had frequencies of 6.0, 15.1, 11.7, 2.0, and 6.0 %, respectively. Except for ADH1C +5702A>G, none of the gene polymorphisms deviated from Hardy-Weinberg equilibrium. Correlations between ADH1B and ADH1C SNPs and HNSCC The SNPs of ADH1B and their ORs for the development of HNSCC are summarized in Table 3. For ADH1B −992C>G, the CG and GG genotypes were associated with a decreased risk of HNSCC compared to the CC genotype. On the other hand, the AC genotype of ADH1B −957C>A, the GG genotype of +3170A>G, the AG genotype of +3491G>A, and the AG genotype of +13543A>G were associated with an increased risk of HNSCC compared to the corresponding
Tumor Biol. Table 2 Single nucleotide polymorphisms and allele frequencies of ADH1B and ADH1C in head and neck squamous cell carcinoma patients and controls
Gene
Locus
ADH1B
−992C>G
Amino acid change
−957C>A +3170A>G
His48Arg
+3377G>T +3491G>A +13543A>G ADH1C
−1064C>T −325G>C +5702A>G +7462T>C +13044A>G
Ile350Val
HWE P value for deviation from Hardy-Weinberg Equilibrium
homozygotes of the major alleles. ADH1B +3377G>T was not associated with the risk of HNSCC. The SNPs of ADH1C and their ORs are summarized in Table 4. The CG genotype of −325G>C and the AG genotype of +5702A>G were associated with an increased risk of HNSCC compared with the corresponding major allele homozygotes. The ORs of +13044A>G were 1.94 (1.01–3.71) and 1.97 (1.05–3.71), in the dominant and co-dominant models, respectively, while, the ORs of −1064C>T were 1.97 (1.05– 3.71) and 1.94 (1.01–3.71), in the same two models. ADH1C +7462T>C was not associated with the risk of HNSCC. Linkage disequilibrium and a haplotype analysis of the SNPs of ADH1B and ADH1C The SNPs of ADH1B were in strong linkage disequilibrium except for the linkage between ADH1B +3377G>T and +13543A>G. The SNPs of ADH1C also were in strong linkage disequilibrium except for the linkages between ADH1C −1064C>T and +7462T>C, and between +7462T>C and +13044A>G (Fig. 1). In terms of haplotype analysis, there were 11 haplotypes among ADH1B SNPs 992C>G, −957C>A, + 3170A>G, +3377G>T, +3491G>A, and +13543A>G, and 8 haplotypes among ADH1C −1064C>T, −325G>C,
Genotype
C 295 C 414 A 317 G 502 G 387
CG 238 AC 158 AG 232 GT 86 AG 177
G 57 A 18 G 41 T 2 A 26
N 590 N 590 N 590 N 590 N 590
A 489 C 520 G 430 A 472 T 566 A 520
AG 97 CT 69 CG 142 AG 98 CT 24 AG 69
G 4 T 1 C 18 G 20 C 0 G 1
N 590 N 590 N 590 N 590 N 590 N 590
Minor allele frequency
HWE
0.298
0.376
0.164
0.538
0.266
0.870
0.076
0.403
0.194
0.320
0.089
0.733
0.060
0.408
0.151
0.141
0.117
0.000
0.020
0.614
0.060
0.408
+5702A>G, +7462T>C, and +13044A>G. We analyzed the five haplotypes of ADH1B and three of ADH1C with frequencies exceeding 5 %. None were related to the risk of HNSCC. The risk of HNSCC and the SNPs of ADH1B and ADH1C excluding the influence of ADH1B +3170A>G and ADH1C +13044A>G In this study, we observed strong linkage disequilibrium between the SNPs of ADH1B and ADH1C. Therefore, we analyzed the five SNPs of ADH1, namely −992C>G, −957C>A, +3377G>T, +3491G>A, and +13543A>G, in 127 HNSCC patients and 190 normal controls who had the AA genotype of ADH1B +3170A>G to eliminate the effect of the +3170A>G His48Arg SNP of ADH1B that had been reported to be associated with an increased risk of HNSCC. The results revealed no significant risk of HNSCC associated with any of the five SNPs. We also analyzed the four SNPs of ADH1C, namely −1064C>T, −325G>C, +5702A>G, and +7462T>C, in 220 patients and 300 normal controls with the AA genotype of ADH1C +13044A>G to avoid the effects of ADH1C +13044A>G Ile350Val. There was again no significant risk of HNSCC associated with any of the SNPs of ADH1C.
His48Arg
2 (0.8) 153 (58.9) 92 (35.4) 15 (5.8) 206 (79.2) 51 (19.6) 3 (1.2)
TT GG AG AA AA AG GG
0 (0.0) 234 (70.9) 85 (25.8) 11 (3.3) 283 (85.8) 46 (13.9) 1 (0.3)
146 (44.2) 145 (43.9) 39 (11.8) 243 (73.6) 79 (23.9) 8 (2.4) 190 (57.6) 125 (37.9) 15 (4.6) 282 (85.5) 48 (14.6)
OR Adjusted odds ratio for gender and age, 95% CI 95 % confidence interval
+13543A>G
+3491G>A
+3377G>T
+3170A>G
−957C>A
149 (57.3) 93 (35.8) 18 (6.9) 171 (65.8) 79 (30.4) 10 (3.9) 127 (48.9) 107 (41.2) 26 (10.0) 220 (84.6) 38 (14.6)
CC CG GG CC AC AA AA AG GG GG GT . 1 2.34 (1.44–3.79) 1.53 (0.89–2.61) 1 2.21 (1.23–3.95) 1.31 (0.36–4.81)
1 0.51 (0.32–0.82) 0.63 (0.42–0.94) 1 1.84 (1.13–2.99) 1.46 (0.76–2.79) 1 1.49 (0.94–2.35) 1.77 (1.15–2.73) 1 1.07 (0.58–1.95)
0.008 0.68
0.0006 0.12
.
0.83
0.09 0.01
0.01 0.26
0.005 0.03
p
−992C>G
Control (%) OR (95 % CI)a
2.05 (1.18–3.54)
1.92 (1.30–2.81)
1.24 (0.71–2.17)
1.65 (1.16–2.33)
1.68 (1.12–2.53)
0.57 (0.41–0.80)
OR (95 % CI)
0.01
0.0009
0.46
0.005
0.01
0.001
p
Co-dominant analysis
ADH1B
Case (%)
Referent analysis
Loci
Gene
Amino acid change Geno type Distribution
Logistic analysis of ADH1B single nucleotide polymorphisms in head and neck squamous cell carcinoma patients and controls
Table 3
2.17 (1.23–3.85)
2.28 (1.44–3.60)
1.16 (0.64–2.09)
1.66 (1.08–2.55)
1.85 (1.16–2.95)
0.49 (0.31–0.75)
OR (95 % CI)
Dominant analysis
0.008
0.0004
0.63
0.02
0.01
0.001
p
1.53 (0.11–20.98)
1.75 (0.61–5.02)
.
2.85 (1.22–6.67)
1.74 (0.49-6.19)
0.52 (0.24–1.12)
OR (95 % CI)
Recessive analysis
0.75
0.30
.
0.02
0.39
0.09
p
Tumor Biol.
Ile350Val
16 (6.2) 0 (0.0) 220 (84.6) 39 (15.0) 1 (0.4)
CT CC AA AG GG
8 (2.4) 0 (0.0) 300 (90.9) 30 (9.1) 0 (0.0)
300 (90.9) 30 (9.1) 0 (0.0) 257 (77.9) 66 (20.0) 7 (2.1) 277 (83.9) 44 (13.3) 9 (2.7) 322 (97.6)
OR Adjusted odds ratio for gender and age, 95% CI 95 % confidence interval
+13044A>G
+7462T>C
+5702A>G
−325G>C
220 (84.6) 39 (15.0) 1 (0.4) 173 (66.5) 76 (29.2) 11 (4.2) 195 (75.0) 54 (20.8) 11 (4.2) 244 (93.9)
CC CT TT GG CG CC AA AG GG TT 2.11 (0.72–6.19) . 1 1.86 (0.97–3.59) .
1 1.86 (0.97–3.59) . 1 2.52 (1.51–4.21) 1.39 (0.78–2.49) 1 2.43 (1.36–4.32) 1.28 (0.73–2.22) 1
0.06 .
0.17 .
0.003 0.39
0.0004 0.27
0.06 .
p
1.97 (1.05–3.71)
2.11 (0.72–6.19)
1.74 (1.14–2.65)
1.95 (1.29–2.93)
1.97 (1.05–3.71)
OR (95 % CI)
0.04
0.17
0.01
0.001
0.04
p
Co-dominant analysis
−1064C>T
OR (95 % CI)a
Case (%)
Control (%)
Referent analysis
Distribution
ADH1C
Genotype
Loci
Gene
Amino acid change
Logistic analysis of ADH1C single nucleotide polymorphisms in head and neck squamous cell carcinoma patients and controls
Table 4
1.94 (1.01–3.71)
2.11 (0.72–6.19)
2.23 (1.32–3.78)
2.38 (1.47–3.86)
1.94 (1.01–3.71)
OR (95 % CI)
Dominant analysis
0.05
0.17
0.003
0.0004
0.05
p
.
.
1.37 (0.45–4.15)
1.51 (0.48–4.78)
.
OR (95 % CI)
Recessive analysis
.
.
.
0.58
0.48
p
Tumor Biol.
Tumor Biol. Fig. 1 Gene structure and linkage disequilibrium of the SNPs of ADH1B and ADH1C. Linkage disequilibrium is displayed, using the Haploview. D′ values are given in the cell at the intersection of each pair of SNPs. A blank cell indicates D′=1.0. The darker the cell, the greater the linkage disequilibrium between the SNPs. Haplotype blocks are outlined
Evaluation of gene-environment interactions according to alcohol consumption and smoking The ORs of ADH1B +3170A>G and ADH1C +13044A>G according to the amounts of alcohol consumption and smoking are presented in Table 5. In the ADH1B +3170A>G, the OR of the GG genotype to the AA genotype was 4.11 (95 % CI 1.26–13.51) in light drinkers and 6.54 (95 % CI 1.80–23.81) in heavy drinkers, with no significant OR in the non-drinkers. In ADH1C +13044A>G, the OR of the AG genotype relative to the AA genotype was 2.68 (95 % CI 1.18–6.06) in light drinkers and 2.44 (95 % CI 0.91–6.58) in heavy drinkers. The OR of the AG genotype in the non-drinker group was not significant. In the evaluation of interaction between the SNPs and smoking, we assessed it only in the light and heavy smoker groups because there was only one smoker in the control group. The ORs of the GG genotype to the AA genotype of ADH1B +3170A>G and the AG genotype relative to the AA genotype of ADH1C +13044A>G were 4.0 (95 % CI 1.73– 9.26) and 2.33 (95 % CI 1.24–4.37) in heavy smokers, respectively, while ORs was not significant in the light smoker group.
Discussion Of the SNPs of ADH1B, +3170A>G His48Arg in exon 3 is the main player and has been studied frequently. The adenine (A) and guanine (G) alleles of that SNP produce histidine and arginine residues, respectively. The ADH1B*1 allele is
defined by arginine at codons 48 and 370 and the ADH1B*2 allele by histidine at codon 48 and arginine at codon 370. Hence the ADH1B*1 and ADH1B*2 alleles can be used to represent the G and A alleles of ADH1B +3170A>G His48Arg, respectively [27]. The ADH1B*2 allele has an elevated metabolic activity. The ADH1B*2/*2 genotype has approximately 70–80-fold higher metabolic activity than the ADH1B*1/*1 genotype and approximately 40-fold higher than the ADH1B*1/*2 genotype [10]. The frequencies of the alleles and genotypes of the SNPs of ADH1B vary between ethnic groups. ADH1B*1 is highly prevalent in Caucasians, and ADH1B*2 among Asians [28]. Among Koreans, the frequency of ADH1B*2 has been reported to be 74–81 % [11, 29, 30]. We found a frequency of ADH1B*2 (the A allele of ADH1B +3170A>G) of 73.4 %, in agreement with previous reports. There are conflicting results for the association between ADH1B +3170A>G and the risk of HNSCC. We previously reported that the ADH1B*1/*1 genotype was associated with an increased risk of HNSCC [11]. Several studies including meta-analysis also reported that ADH1B*1 allele was found to be associated with the risk of HNSCC, especially in higher alcohol consumers [12–17]. On the other hand, other groups failed to detect any significant correlation between ADH1B*1 and *2 and the risk of HNSCC or esophageal cancer [18, 19]. In the present study, the OR of the ADH1B +3170A>G GG genotype compared to the AA genotype was 1.77 (95 % CI 1.15–2.73), pointing to a significant correlation between ADH1B +3170A>G and the risk of HNSCC. Of the other five SNPs of ADH1B found in Koreans, ADH1B −992C>G was associated with a decreased risk of HNSCC, while ADH1B −957C>A, +3491G>A, and +13543A>G were associated with an increased risk.
Tumor Biol. Table 5 Logistic analysis of ADH1B +3170A>G and ADH1C +13044A>G polymorphisms in head and neck squamous cell carcinoma patients and controls according to alcohol consumption and smoking
Gene ADH1B +3170A>G
Non-drinker
Light drinker
Heavy drinker
ADH1C +13044A>G
Non-drinker
Light drinker
Heavy drinker
ADH1B +3170A>G
Light smoker
Heavy Smoker
ADH1C +13044A>G
Light smoker
Heavy smoker OR Adjusted odds ratio for gender and age, 95% CI 95 % confidence interval
There was strong linkage disequilibrium between ADH1B +3170A>G and the other ADH1B SNPs. Therefore, to avoid the effects of +3170A>G, we analyzed the relative risks of the other five SNPs among individuals with the AA genotype of +3170A>G. The results showed that there was no significant correlation between the other five SNPs and the risk of HNSCC. This suggests that the increased risks associated with −957C>A, +3491G>A, and +13543A>G are due to their linkage disequilibrium with ADH1B +3170A>G. Of the SNPs of ADH1C, Ile350Val (ADH1C +13044A>G) and Arg272Gln are the key players. The A and G alleles of +13044 A>G produce isoleucine and valine, respectively. The ADH1C*1 allele is defined by isoleucine and arginine in codons 350 and 272, respectively, and the ADH1C*2 allele by valine and glutamine in codons 350 and 272, respectively [31]. Because Ile350Val and Arg272Gln are in perfect linkage disequilibrium, the Ile350 (A) and Val350 (G) alleles of ADH1C +13044A>G Ile350Val can represent
Genotype
Case (%)
Control (%)
AA AG GG AA AG
55 (9.3) 40 (6.8) 3 (0.5) 23 (3.9) 26 (4.4)
49 (8.3) 33 (5.6) 6 (1.0) 81 (13.7) 61 (10.3)
GG AA AG GG AA AG GG AA AG GG AA AG GG AA AG GG AA AG
7 (1.2) 49 (8.3) 41 (6.9) 16 (2.7) 87 (14.7) 11 (1.9) 0 (0) 43 (7.3) 13 (2.2) 0 (0) 90 (15.3) 15 (2.5) 1 (0.2) 24 (9.2) 19 (7.3) 6 (2.3) 76 (29.2) 66 (25.4)
6 (1.0) 60 (10.2) 31 (5.3) 3 (0.5) 79 (13.4) 9 (1.5) 0 (0) 133 (22.5) 15 (2.5) 0 (0) 88 (14.9) 6 (1.0) 0 (0) 45 (13.6) 31 (9.4) 6 (1.8) 144 (43.6) 94 (28.5)
GG AA AG GG AA AG GG
19 (7.3) 40 (15.4) 9 (3.5) 0 (0) 134 (51.5) 26 (10.0) 1 (0.4)
9 (2.7) 71 (21.5) 11 (3.3) 0 (0) 228 (69.1) 19 (5.8) 0 (0)
OR
95 % CI
p
1.08 0.45
0.59–1.97 0.11–1.88
0.879 0.313
1.50
0.78–2.88
0.247
4.11
1.26–13.51
0.037
1.62 6.54
0.89–2.95 1.80–23.81
0.13 0.002
1.11
0.44–2.82
1.0
2.68
1.18–6.06
0.022
2.44
0.91–6.58
0.104
1.15 1.88
0.54–2.44 0.55–6.45
0.847 0.345
1.33
0.87–2.02
0.198
4.0
1.73–9.26
0.001
1.45
0.55–3.80
0.461
2.33
1.24–4.37
0.009
the ADH1C*1 and *2 alleles, respectively [32]. The ADH1C*1 allele has approximately 2.5-fold higher oxidative function than the ADH1C*2 allele [10]. The frequencies of the alleles of the ADH1C SNPs also differ between ethnic groups. The frequencies of ADH1C*1 and *2 are similar in Caucasian people so that 45–70 % of Caucasians are ADH1C heterozygotes. However, the ADH1C*1 allele is present in 75–90 % of Africans and 85– 100 % of Asians [28]. In other studies conducted among Koreans, the frequency of the ADH1C*1 allele was 94– 95 % [29, 30]. In this study, its frequency was 94 %, in agreement with previous reports. With regard to the association between ADH1C SNPs and the risk of head and neck cancer, conflicting results have been reported. Some studies reported the ADH1C*1/*1 genotype was associated with an increased risk of HNSCC [20, 21]. However, on the contrary to this, other studies reported that ADH1C*1 allele was associated with a reduced risk of
Tumor Biol.
HNSCC [22, 23]. Several studies including some metaanalyses were unable to conclude that there was a correlation between ADH1C +13044A>G and the risk of HNSCC because of the conflicting results n. In the current study, the OR of the AG genotype of ADH1C +13044A>G was not statistically significance in the referent model. However, it indicated an increased relative risk in the dominant (OR 1.94; 95 % CI 1.01–3.17) and co-dominant models (OR 1.97; 95 % CI 1.05–3.71). These results might suggest that the AG genotype of ADH1C +13044A>G is associated with an increased risk of HNSCC; however, its statistical power is not strong. The SNPs −1064C>T, −325G>C, and +5702A>G were associated with an increased risk of HNSCC, though ADH1C +7462T>C was not. However, the increased ORs of ADH1C −1064C>T, −325G>C, and +5702A>G SNPs might be effects of ADH1C +13044A>G because of the strong linkage disequilibrium between ADH1C +13044A>G and the other SNPs. We also analyzed ADH1B +3170A>G and ADH1C +13044A>G as a function of the level of alcohol consumption and smoking to evaluate gene-environment interactions. The relative risk of ADH1B +3170A>G and ADH1C +13044A>G on HNSCC was significantly increased in the drinkers, but not in the non-drinkers. Also, the relative risk of them was significantly increased in the heavy smokers than those in the light smokers. This reflects the gene-environment interaction between alcohol consumption and smoking, and the SNPs of ADH1B and ADH1C. In conclusion, ADH1B SNP +3170A>G His48Arg and ADH1C SNP +13044A>G Ile350Val are associated with increased risks of HNSCC, and could be useful molecular biologic markers for identifying patients at increased risk of developing HNSCC in Koreans. There is also a geneenvironment interaction between these SNPs and alcohol consumption and smoking.
Conflicts of interest None
References 1. Tae K, Lee HS, Park BJ, Park CW, Kim KR, Cho HY, et al. Association of DNA repair gene XRCC1 polymorphisms with head and neck cancer in Korean population. Int J Cancer. 2004;222:805–8. 2. Jeong SW, Tae K, Lee SH, Kim KR, Park CW, Park BL, et al. Cox-2 and IL-10 polymorphisms and association with squamous cell carcinoma of the head and neck in a Korean sample. J Korean Med Sci. 2010;25:1024–8. 3. Ji YB, Tae K, Lee YS, Lee SH, Kim KR, Park CW, et al. XPD polymorphisms and risk of squamous cell carcinoma of the head and neck in a Korean sample. Clin Exp Otorhinolaryngol. 2010;3: 42–7.
4. Blot WJ, McLaughlin JK, Winn DM, Austin DF, Greenberg RS, Preston-Martin S, et al. Smoking and drinking in relation to oral and phayngeal cancer. Cancer Res. 1988;48:3282–7. 5. IARC. Alcoholic consumption and ethyl carbamate. In: IARC monographs on the evaluation of carcinogenic risks to humans. Lyon, France: IARC; 2010, vol. 96. 6. Obe G, Jonas R, Schmidt S. Metabolism of ethanol in vitro produces a compound which induces sister-chromatid exchanges in human peripheral lymphocytes in vitro: acetaldehyde not ethanol is mutagenetic. Mutat Res. 1986;174:47–51. 7. Fang JL, Vaca CE. Detection of DNA adducts of acetaldehyde in peripheral white blood cells of alcohol abusers. Carcinogenesis. 1997;18:627–32. 8. Homann N, Jousimies-Somer H, Jokelainen K, Heine R. Salaspuro M High acetaldehyde levels in saliva after ethanol consumption: methodological aspects and pathogenetic implication. Carcinogenesis. 1997;18:1739–43. 9. Harada S, Misawa S, Agarwal DP, Goedde HW. Liver alcohol dehydrogenase and aldehyde dehydrogenase in the Japanese: isoenzyme variation and its possible role in alcohol intoxication. Am J Hum Genet. 1980;32:8–15. 10. Edenberg HJ. The genetics of alcohol metabolism-role of alcohol dehydrogenase and aldehyde dehydrogenase variants. Alcohol Res Health. 2007;30:5–13. 11. Ji YB, Tae K, Ahn TH, Lee SH, Kim KR, Park CW, et al. ADH1B and ALDH2 polymorphisms and their associations with increased risk of squamous cell carcinoma of the head and neck in the Korean population. Oral Oncol. 2011;47:583–7. 12. Yokoyama A, Muramatsu T, Omori T, Yokoyama T, Matsushita S, Higuchi S, et al. Alcohol and aldehyde dehydrogenase gene polymorphisms and oropharyngolaryngeal, esophageal and stomach cancers in Japanese alcoholics. Carcinogenesis. 2001;22:433–9. 13. Hiraki A, Matsuo K, Wakai K, Suzuki T, Hasegawa Y, Tajima K. Gene–gene and gene–environment interactions between alcohol drinking habit and polymorphisms in alcohol-metabolizing enzyme genes and the risk of head and neck cancer in Japan. Cancer Sci. 2007;98:1087–91. 14. Guo H, Zhang G, Mai R. Alcohol dehydrogenase-1B Arg47His polymorphism and upper aerodigestive tract cancer risk: a meta-analysis including 24,252 subjects. Alcohol Clin Exp Res. 2012;36:272–8. 15. Chang JS, Straif K, Guha N. The role of alcohol dehydrogenase genes in head and neck cancers: a systematic review and metaanalysis of ADH1B and ADH1C. Mutagenesis. 2012;27:275–86. 16. Hashibe M, Boffetta P, Zaridze D, Shangina O, SzeszeniaDabrowska N, Mates D, et al. Evidence for an important role of alcohol- and aldehyde-metabolizing genes in cancers of the upper aerodigestive tract. Cancer Epidemiol Biomarkers Prev. 2006;15: 696–703. 17. Zhang L, Jiang Y, Wu Q, Li Q, Chen D, Xu L, et al. Geneenvironment interaction on the risk of esophageal cancer among Asian poppulations with the G48A polymorphism in the alcohol dehydrogenase-2 gene: a meta-analysis. Tumor Biol. 2014;35: 4705–17. 18. Dura P, Berkers T, van Veen EM, Salomon J, te Morsche RH, Kristinsson JO, et al. Polymorphisms in alcohol-metabolizing enzymes and esophageal carcinoma susceptibility: a Dutch Caucasian case-control study. J Hum Genet. 2013;58(11):742–8. 19. Risch A, Ramroth H, Raedts V, Rajaee-Behbahani N, Schmezer P, Bartsch H, et al. Laryngeal cancer risk in Caucasians is associated with alcohol and tobacco consumption but not modified by genetic polymorphisms in class I alcohol dehydrogenases ADH1B and ADH1C, and glutathione-S-transferases GSTM1 and GSTT1. Pharmacogenetics. 2003;13:225–30. 20. Harty LC, Caporaso NE, Hayes RB, Winn DM, Bravo-Otero E, Blot WJ, et al. Alcohol dehydrogenase 3 genotype and risk of oral cavity and pharyngeal cancers. J Natl Cancer Inst. 1997;89:1698–705.
Tumor Biol. 21. Visapää JP, Götte K, Benesova M, Li J, Homann N, Conradt C, et al. Increased cancer risk in heavy drinkers with the alcohol dehydrogenase 1C*1 allele, possibly due to salivary acetaldehyde. Gut. 2004;53:871–6. 22. Muto M, Takahashi M, Ohtsu A, Ebihara S, Yoshida S, Esumi H. Risk of multiple squamous cell carcinomas both in the esophagus and the head and neck region. Carcinogenesis. 2005;26:1008–12. 23. Wang D, Ritchie JM, Smith EM, Zhang Z, Turek LP, Haugen TH. Alcohol dehydrogenase 3 and risk of squamous cell carcinomas of the head and neck. Cancer Epidemiol Biomarkers Prev. 2005;14:626–32. 24. Cadoni G, Boccia S, Petrelli L, Di Giannantonio P, Arzani D, Giorgio A, et al. A review of genetic epidemiology of head and neck cancer related to polymorphisms in metabolic genes, cell cycle control and alcohol metabolism. Acta Otorhinolaryngol Ital. 2012;32:1–11. 25. Hiyama T, Yoshihara M, Tanaka S, Chayama K. Genetic polymorphisms and head and neck cancer risk (Review). Int J Oncol. 2008;32:945–73. 26. Sturgis EM, Dahlstrom KR, Guan Y, Eicher SA, Strom SS, Spitz MR, et al. Alcohol dehydrogenase 3 genotype is not associated with risk of squamous cell carcinoma of the oral cavity and pharynx. Cancer Epidemiol Biomarkers Prev. 2001;10:273–5. 27. Osier MV, Pakstis AJ, Soodyall H, Comas D, Goldman D, Odunsi A, et al. A global perspective on genetic variation at the ADH genes
reveals unusual patterns of linkage disequilibrium and diversity. Am J Hum Genet. 2002;71:84–99. 28. Brennan P, Lewis S, Hashibe M, Bell DA, Boffetta P, Bouchardy C, et al. Pooled analysis of alcohol dehydrogenase genotypes and head and neck cancer: a HuGE review. Am J Epidemiol. 2004;159:1–16. 29. Lee HC, Lee HS, Jung SH, Yi SY, Jung HK, Yoon JH, et al. Association between polymorphisms of ethanol-metabolizing enzymes and susceptibility to alcoholic cirrhosis in a Korean male population. J Korean Med Sci. 2001;16:745–50. 30. Jung SH, Lee HC, Yoon JH, Lee HS, Kim CY. The association between genetic polymorphisms of ethanol-metabolizing enzymes and susceptibility to alcoholic liver cirrhosis. Korean J Hepatol. 1998;4:1–11. 31. Höög JO, Hedén LO, Larsson K, Jörnvall H, von Bahr-Lindström H. The gamma 1 and gamma 2 subunits of human liver alcohol dehydrogenase. cDNA structures, two amino acid replacements, and compatibility with changes in the enzymatic properties. Eur J Biochem. 1986;159:215–8. 32. Packer BR, Yeager M, Burdett L, Welch R, Beerman M, Qi L, et al. SNP500Cancer: a public resource for sequence validation, assay development, and frequency analysis for genetic variation in candidate genes. Nucleic Acids Res. 2006;1(34):D617–21.
RESEARCH ARTICLE
Association between ADH1B and ADH1C polymorphisms and the risk of head and neck squamous cell carcinoma Yong Bae Ji & Seung Hwan Lee & Kyung Rae Kim & Chul Won Park & Chang Myeon Song & Byung Lae Park & Hyoung Doo Shin & Kyung Tae
Received: 20 October 2014 / Accepted: 8 January 2015 # International Society of Oncology and BioMarkers (ISOBM) 2015
Abstract Alcohol consumption is one of the major risk factors for head and neck squamous cell carcinoma (HNSCC), and the alcohol dehydrogenase (ADH) family proteins are key enzymes in ethanol metabolism. We examined the associations between single nucleotide polymorphisms (SNPs) of ADH1B and ADH1C and the risk of HNSC C. We analyzed six SNPS of ADH1B, namely −992C>G, −957C > A, +3170A>G, +3377G>T, +3491G>A, and + 13543A>G, and five SNPs of ADH1C, namely −1064C>T, −325G>C, +5702A>G, +7462T>C, and +13044A>G, in 260 Korean HNSCC patients and 330 controls, using single base extension and the TaqMan assay. The odds ratios (ORs) and 95 % confidence intervals (95 % CIs) of the CG and GG genotypes of ADH1B −992C>G, the AA genotype of −957C>A, the GG genotype of +3170A>G, the GA genotype of +3491G>A, and +13543A>G were 0.51 (0.32–0.82), 0.63 (0.42–0.94), 1.84 (1.13–2.99), 1.77 (1.15–2.73), 2.34 (1.44– 3.79), and 2.21 (1.23–3.95), respectively. The ORs of ADH1C +13044A>G were 1.94 (1.01–3.71) and 1.97 (1.05–3.71) in the dominant and co-dominant models, respectively. The ORs of the GC genotype of ADH1C −325G>C and the AG genotype of +5702A>G were 2.52 (1.51–4.21) and 2.43 (1.36– 4.32), respectively. ADH1B +3170A>G and ADH1C + 13044A>G were in strong linkage disequilibrium with the other SNPs of ADH1B and ADH1C, respectively. There were gene-environment interactions between ADH1B +3170A>G Y. B. Ji : S. H. Lee : K. R. Kim : C. W. Park : C. M. Song : K. Tae (*) Department of Otolaryngology-Head and Neck Surgery, College of Medicine, Hanyang University, 222 Wangsimni-ro, Seongdong-gu Seoul 133-792, Korea e-mail: [email protected] B. L. Park : H. D. Shin Department of Genetic Epidemiology, SNP Genetics, Inc., Seoul, Korea
and ADH1C +13044A>G and alcohol consumption and smoking. ADH1B +3170A>G and ADH1C +13044A>G SNPs are associated with an increased risk of HNSCC, and they could be used as biomarkers for the high-risk group of HNSCC in Koreans. Keywords Polymorphism . SNP . ADH1B . ADH1C . Head and neck cancer
Introduction The development of head and neck squamous cell carcinoma (HNSCC) is associated with environmental factors such as chemical carcinogens, ultraviolet exposure, irradiation and viral infection, and host-related factors such as genetic susceptibility [1–3]. Many epidemiologic studies have shown that alcohol consumption is one of causative factors in the development of HNSCC [4]. According to the International Agency for Research on Cancer (IARC), there is adequate evidence that alcohol is a causative factor of malignancies such as oral cancer, pharyngeal cancer, laryngeal cancer, esophageal cancer, hepatoma, colorectal cancer, and breast cancer [5]. Experimental studies have shown that acetaldehyde, a metabolite of ethanol, inhibits DNA synthesis and repair [6]; it binds covalently to DNA, forming a DNA adduct and so causing DNA damage. It has been reported that this adduct is present at higher levels in alcohol drinkers than in nondrinkers [7]. Furthermore, the acetaldehyde concentration is 10–20 times higher in the saliva than in the serum after alcohol consumption. These findings suggest that acetaldehyde is associated with the development of HNSCC in alcohol consumers [8].
Tumor Biol.
The enzyme alcohol dehydrogenase (ADH) is responsible for the conversion of ethanol to acetaldehyde, and there are several ADH isoenzymes that differ in amino acid sequence. The isoenzymes are divided into five categories: class I (ADH1A, ADH1B, and ADH1C), class II (ADH4), class III (ADH5), class IV (ADH7), and class V (ADH6). Three of the class I isoenzymes (ADH1A, ADH1B, and ADH1C) are located close to the 4q21-23 locus [9]. ADH1B and ADH1C are predominantly expressed in the liver and are the main enzymes responsible for ethanol oxidation. Although alcohol consumption is associated with the development of HNSCC, this does not mean that the increased risk of developing HNSCC is strictly proportional to the amount of alcohol consumed because the susceptibility of individuals differs, and genetic polymorphisms of ADH lead to variability in the activity of ADH [10]. Studies of ADH polymorphism have focused mainly on the single nucleotide polymorphisms (SNPs) ADH1B +3170A>G His48Arg (rs1229984) and ADH1C +13044A>G Ile350Val (rs698). According to some studies, these SNPs are associated with an increased risk of developing HNSCC, while other studies have found no association [11–26]. However, there have been only a few studies of ADH1B and ADH1C SNPs other than ADH1B +3170A>G and ADH1C +13044A>G. Therefore, in this study, we analyzed six SNPs of ADH1B, including ADH1B +3170A>G, and five SNPs of ADH1C, including ADH1C +13044A>G, which are known to exist in Koreans, and we evaluated the relationship between these SNPs of and the risk of developing HNSCC.
Materials and methods Patients We conducted a hospital-based case-control study in 590 subjects. The case group consisted of 260 patients who were histopathologically diagnosed with HNSCC between January 2004 and December 2009 in a tertiary hospital. The 330 controls consisted of healthy volunteers who visited the hospital for routine health examinations, and patients admitted to the same hospital for treatment for non-cancerous diseases such as chronic tonsillitis, chronic sinusitis, or chronic otitis media during the same period. The control individuals had no history of previous malignancies or genetic disorders. All the subjects of this study were Korean, and we obtained written informed consent from all the subjects, and the study was approved by the institutional review board of Hanyang University. The demographic and clinical characteristics of the study subjects are presented in Table 1. The mean age was 62.1± 15.7 years (range, 28–90 years) in the HNSCC group and 43.7±10.9 years (range, 21–76 years) in the control group
(PG, and five SNPs of ADH1C, namely ADH1C −1064C>T, −325G>C, +5702A>G, +7462T>C, and +13044A>G. The TaqMan assay was used to genotype ADH1B + 3170A>G. Forward and reverse primer were 5′-TCTTTTCT GAATCTGAACAGCTTCTCTTT-3′ and 5′-GGTCACCA GGTTGCCACTAAC-3′, respectively. The probe for the wild-type A allele was 5′-CTGTAGGAATCTGTCGCACA3′, with the 5′ end and 3′ end tagged with the VIC™ fluorescent dye and the FAM™ quencher dye, respectively. The probe for the mutant-type G allele was 5′-CTGTAGGAAT CTGTCACACA-3′, tagged as for the wild-type probe. Polymerase chain reactions (PCR) were performed in 5 μL reaction mixtures containing TaqMan universal PCR master mix (Applied Biosystems, Foster City, CA, USA), UNG, primer (900 μM), probe (200 nM), and genomic DNA (20 ng). For genotyping the other ten SNPs, we used the single base extension (SBE) technique. PCR reaction mixtures of 5 μL contained primer (1.25 pM), genomic DNA (5 ng), dNTPs (250 M), and 0.15 U Taq DNA polymerase (Applied Biosystems). The reaction conditions were as follows for 40 amplification cycles: 50 °C for 2 min, 95 °C for 10 min followed by 15 s denaturing, then final annealing and extension at 60 °C for 1 min. The TaqMan assay plate was transferred to a Prism 9700HT instrument (Applied Biosystems), and fluorescence levels in each well were measured with SDS software (ver. 2.3., Applied Biosystems). The positive (12
Tumor Biol. Table 1 Clinical characteristics of head and neck squamous cell carcinoma patients and controls (n=590)
Variables
Case
p Value
Control
N
%
N
%
23 86 151
(8.9) (33.1) (58.0)
202 97 31
(61.2) (29.4) (9.4)
Age (year)
a
Less than 20 pack-years
b
20 pack-years or more
c
Consumption less than 120 g of alcohol per week
d
Consumption 120 g of alcohol or more per week
≤45 45–60 ≥60 Smoking status None Former Lighta Heavyb Age started smoking (year) Period of smoking (year) Alcohol consumption None Lightc Heavyd Age started drinking (year) Period of drinking (year) Gender Male Female
samples) and negative (12 blanks) controls were used instead of quantitative RT-PCR. The PCR products and Genescan 120Liz size-standard solution (Applied Biosystems) were placed in highly-deionized (Hi-Di) formamide, and subjected to electrophoresis using an ABI Prism 3100 Genetic Analyzer (Applied Biosystems). Genotyping was done with an ABI Prism GeneScan and Genotyper.
Statistical analysis Student’s t test was used for continuous variables, and the chi-square test and Fisher’s exact test for categorical variables. Linkage disequilibrium (LD) was analyzed using Haploview v4.2 software (http://www.broadinstitute. org/mpg/haploview), examining Lewontin’s D′ (|D′|) and the LD coefficient r 2 between all pairs of bi-allelic loci. Haplotypes were estimated using PHASE v2.0 software from the University of Washington. Odds ratios (OR) and 95 % confidence intervals (CI) was calculated with a logistic regression model using SAS, version 9.2 (SAS Inc., Cary, NC, USA). Analyses of the referent (to the homozygotes of major alleles) and three alternative models (co-dominant, dominant, and recessive for the minor alleles) were used for OR and 95 % CI values. PC, +5702A>G, +7462T>C, and +13044A>G, had frequencies of 6.0, 15.1, 11.7, 2.0, and 6.0 %, respectively. Except for ADH1C +5702A>G, none of the gene polymorphisms deviated from Hardy-Weinberg equilibrium. Correlations between ADH1B and ADH1C SNPs and HNSCC The SNPs of ADH1B and their ORs for the development of HNSCC are summarized in Table 3. For ADH1B −992C>G, the CG and GG genotypes were associated with a decreased risk of HNSCC compared to the CC genotype. On the other hand, the AC genotype of ADH1B −957C>A, the GG genotype of +3170A>G, the AG genotype of +3491G>A, and the AG genotype of +13543A>G were associated with an increased risk of HNSCC compared to the corresponding
Tumor Biol. Table 2 Single nucleotide polymorphisms and allele frequencies of ADH1B and ADH1C in head and neck squamous cell carcinoma patients and controls
Gene
Locus
ADH1B
−992C>G
Amino acid change
−957C>A +3170A>G
His48Arg
+3377G>T +3491G>A +13543A>G ADH1C
−1064C>T −325G>C +5702A>G +7462T>C +13044A>G
Ile350Val
HWE P value for deviation from Hardy-Weinberg Equilibrium
homozygotes of the major alleles. ADH1B +3377G>T was not associated with the risk of HNSCC. The SNPs of ADH1C and their ORs are summarized in Table 4. The CG genotype of −325G>C and the AG genotype of +5702A>G were associated with an increased risk of HNSCC compared with the corresponding major allele homozygotes. The ORs of +13044A>G were 1.94 (1.01–3.71) and 1.97 (1.05–3.71), in the dominant and co-dominant models, respectively, while, the ORs of −1064C>T were 1.97 (1.05– 3.71) and 1.94 (1.01–3.71), in the same two models. ADH1C +7462T>C was not associated with the risk of HNSCC. Linkage disequilibrium and a haplotype analysis of the SNPs of ADH1B and ADH1C The SNPs of ADH1B were in strong linkage disequilibrium except for the linkage between ADH1B +3377G>T and +13543A>G. The SNPs of ADH1C also were in strong linkage disequilibrium except for the linkages between ADH1C −1064C>T and +7462T>C, and between +7462T>C and +13044A>G (Fig. 1). In terms of haplotype analysis, there were 11 haplotypes among ADH1B SNPs 992C>G, −957C>A, + 3170A>G, +3377G>T, +3491G>A, and +13543A>G, and 8 haplotypes among ADH1C −1064C>T, −325G>C,
Genotype
C 295 C 414 A 317 G 502 G 387
CG 238 AC 158 AG 232 GT 86 AG 177
G 57 A 18 G 41 T 2 A 26
N 590 N 590 N 590 N 590 N 590
A 489 C 520 G 430 A 472 T 566 A 520
AG 97 CT 69 CG 142 AG 98 CT 24 AG 69
G 4 T 1 C 18 G 20 C 0 G 1
N 590 N 590 N 590 N 590 N 590 N 590
Minor allele frequency
HWE
0.298
0.376
0.164
0.538
0.266
0.870
0.076
0.403
0.194
0.320
0.089
0.733
0.060
0.408
0.151
0.141
0.117
0.000
0.020
0.614
0.060
0.408
+5702A>G, +7462T>C, and +13044A>G. We analyzed the five haplotypes of ADH1B and three of ADH1C with frequencies exceeding 5 %. None were related to the risk of HNSCC. The risk of HNSCC and the SNPs of ADH1B and ADH1C excluding the influence of ADH1B +3170A>G and ADH1C +13044A>G In this study, we observed strong linkage disequilibrium between the SNPs of ADH1B and ADH1C. Therefore, we analyzed the five SNPs of ADH1, namely −992C>G, −957C>A, +3377G>T, +3491G>A, and +13543A>G, in 127 HNSCC patients and 190 normal controls who had the AA genotype of ADH1B +3170A>G to eliminate the effect of the +3170A>G His48Arg SNP of ADH1B that had been reported to be associated with an increased risk of HNSCC. The results revealed no significant risk of HNSCC associated with any of the five SNPs. We also analyzed the four SNPs of ADH1C, namely −1064C>T, −325G>C, +5702A>G, and +7462T>C, in 220 patients and 300 normal controls with the AA genotype of ADH1C +13044A>G to avoid the effects of ADH1C +13044A>G Ile350Val. There was again no significant risk of HNSCC associated with any of the SNPs of ADH1C.
His48Arg
2 (0.8) 153 (58.9) 92 (35.4) 15 (5.8) 206 (79.2) 51 (19.6) 3 (1.2)
TT GG AG AA AA AG GG
0 (0.0) 234 (70.9) 85 (25.8) 11 (3.3) 283 (85.8) 46 (13.9) 1 (0.3)
146 (44.2) 145 (43.9) 39 (11.8) 243 (73.6) 79 (23.9) 8 (2.4) 190 (57.6) 125 (37.9) 15 (4.6) 282 (85.5) 48 (14.6)
OR Adjusted odds ratio for gender and age, 95% CI 95 % confidence interval
+13543A>G
+3491G>A
+3377G>T
+3170A>G
−957C>A
149 (57.3) 93 (35.8) 18 (6.9) 171 (65.8) 79 (30.4) 10 (3.9) 127 (48.9) 107 (41.2) 26 (10.0) 220 (84.6) 38 (14.6)
CC CG GG CC AC AA AA AG GG GG GT . 1 2.34 (1.44–3.79) 1.53 (0.89–2.61) 1 2.21 (1.23–3.95) 1.31 (0.36–4.81)
1 0.51 (0.32–0.82) 0.63 (0.42–0.94) 1 1.84 (1.13–2.99) 1.46 (0.76–2.79) 1 1.49 (0.94–2.35) 1.77 (1.15–2.73) 1 1.07 (0.58–1.95)
0.008 0.68
0.0006 0.12
.
0.83
0.09 0.01
0.01 0.26
0.005 0.03
p
−992C>G
Control (%) OR (95 % CI)a
2.05 (1.18–3.54)
1.92 (1.30–2.81)
1.24 (0.71–2.17)
1.65 (1.16–2.33)
1.68 (1.12–2.53)
0.57 (0.41–0.80)
OR (95 % CI)
0.01
0.0009
0.46
0.005
0.01
0.001
p
Co-dominant analysis
ADH1B
Case (%)
Referent analysis
Loci
Gene
Amino acid change Geno type Distribution
Logistic analysis of ADH1B single nucleotide polymorphisms in head and neck squamous cell carcinoma patients and controls
Table 3
2.17 (1.23–3.85)
2.28 (1.44–3.60)
1.16 (0.64–2.09)
1.66 (1.08–2.55)
1.85 (1.16–2.95)
0.49 (0.31–0.75)
OR (95 % CI)
Dominant analysis
0.008
0.0004
0.63
0.02
0.01
0.001
p
1.53 (0.11–20.98)
1.75 (0.61–5.02)
.
2.85 (1.22–6.67)
1.74 (0.49-6.19)
0.52 (0.24–1.12)
OR (95 % CI)
Recessive analysis
0.75
0.30
.
0.02
0.39
0.09
p
Tumor Biol.
Ile350Val
16 (6.2) 0 (0.0) 220 (84.6) 39 (15.0) 1 (0.4)
CT CC AA AG GG
8 (2.4) 0 (0.0) 300 (90.9) 30 (9.1) 0 (0.0)
300 (90.9) 30 (9.1) 0 (0.0) 257 (77.9) 66 (20.0) 7 (2.1) 277 (83.9) 44 (13.3) 9 (2.7) 322 (97.6)
OR Adjusted odds ratio for gender and age, 95% CI 95 % confidence interval
+13044A>G
+7462T>C
+5702A>G
−325G>C
220 (84.6) 39 (15.0) 1 (0.4) 173 (66.5) 76 (29.2) 11 (4.2) 195 (75.0) 54 (20.8) 11 (4.2) 244 (93.9)
CC CT TT GG CG CC AA AG GG TT 2.11 (0.72–6.19) . 1 1.86 (0.97–3.59) .
1 1.86 (0.97–3.59) . 1 2.52 (1.51–4.21) 1.39 (0.78–2.49) 1 2.43 (1.36–4.32) 1.28 (0.73–2.22) 1
0.06 .
0.17 .
0.003 0.39
0.0004 0.27
0.06 .
p
1.97 (1.05–3.71)
2.11 (0.72–6.19)
1.74 (1.14–2.65)
1.95 (1.29–2.93)
1.97 (1.05–3.71)
OR (95 % CI)
0.04
0.17
0.01
0.001
0.04
p
Co-dominant analysis
−1064C>T
OR (95 % CI)a
Case (%)
Control (%)
Referent analysis
Distribution
ADH1C
Genotype
Loci
Gene
Amino acid change
Logistic analysis of ADH1C single nucleotide polymorphisms in head and neck squamous cell carcinoma patients and controls
Table 4
1.94 (1.01–3.71)
2.11 (0.72–6.19)
2.23 (1.32–3.78)
2.38 (1.47–3.86)
1.94 (1.01–3.71)
OR (95 % CI)
Dominant analysis
0.05
0.17
0.003
0.0004
0.05
p
.
.
1.37 (0.45–4.15)
1.51 (0.48–4.78)
.
OR (95 % CI)
Recessive analysis
.
.
.
0.58
0.48
p
Tumor Biol.
Tumor Biol. Fig. 1 Gene structure and linkage disequilibrium of the SNPs of ADH1B and ADH1C. Linkage disequilibrium is displayed, using the Haploview. D′ values are given in the cell at the intersection of each pair of SNPs. A blank cell indicates D′=1.0. The darker the cell, the greater the linkage disequilibrium between the SNPs. Haplotype blocks are outlined
Evaluation of gene-environment interactions according to alcohol consumption and smoking The ORs of ADH1B +3170A>G and ADH1C +13044A>G according to the amounts of alcohol consumption and smoking are presented in Table 5. In the ADH1B +3170A>G, the OR of the GG genotype to the AA genotype was 4.11 (95 % CI 1.26–13.51) in light drinkers and 6.54 (95 % CI 1.80–23.81) in heavy drinkers, with no significant OR in the non-drinkers. In ADH1C +13044A>G, the OR of the AG genotype relative to the AA genotype was 2.68 (95 % CI 1.18–6.06) in light drinkers and 2.44 (95 % CI 0.91–6.58) in heavy drinkers. The OR of the AG genotype in the non-drinker group was not significant. In the evaluation of interaction between the SNPs and smoking, we assessed it only in the light and heavy smoker groups because there was only one smoker in the control group. The ORs of the GG genotype to the AA genotype of ADH1B +3170A>G and the AG genotype relative to the AA genotype of ADH1C +13044A>G were 4.0 (95 % CI 1.73– 9.26) and 2.33 (95 % CI 1.24–4.37) in heavy smokers, respectively, while ORs was not significant in the light smoker group.
Discussion Of the SNPs of ADH1B, +3170A>G His48Arg in exon 3 is the main player and has been studied frequently. The adenine (A) and guanine (G) alleles of that SNP produce histidine and arginine residues, respectively. The ADH1B*1 allele is
defined by arginine at codons 48 and 370 and the ADH1B*2 allele by histidine at codon 48 and arginine at codon 370. Hence the ADH1B*1 and ADH1B*2 alleles can be used to represent the G and A alleles of ADH1B +3170A>G His48Arg, respectively [27]. The ADH1B*2 allele has an elevated metabolic activity. The ADH1B*2/*2 genotype has approximately 70–80-fold higher metabolic activity than the ADH1B*1/*1 genotype and approximately 40-fold higher than the ADH1B*1/*2 genotype [10]. The frequencies of the alleles and genotypes of the SNPs of ADH1B vary between ethnic groups. ADH1B*1 is highly prevalent in Caucasians, and ADH1B*2 among Asians [28]. Among Koreans, the frequency of ADH1B*2 has been reported to be 74–81 % [11, 29, 30]. We found a frequency of ADH1B*2 (the A allele of ADH1B +3170A>G) of 73.4 %, in agreement with previous reports. There are conflicting results for the association between ADH1B +3170A>G and the risk of HNSCC. We previously reported that the ADH1B*1/*1 genotype was associated with an increased risk of HNSCC [11]. Several studies including meta-analysis also reported that ADH1B*1 allele was found to be associated with the risk of HNSCC, especially in higher alcohol consumers [12–17]. On the other hand, other groups failed to detect any significant correlation between ADH1B*1 and *2 and the risk of HNSCC or esophageal cancer [18, 19]. In the present study, the OR of the ADH1B +3170A>G GG genotype compared to the AA genotype was 1.77 (95 % CI 1.15–2.73), pointing to a significant correlation between ADH1B +3170A>G and the risk of HNSCC. Of the other five SNPs of ADH1B found in Koreans, ADH1B −992C>G was associated with a decreased risk of HNSCC, while ADH1B −957C>A, +3491G>A, and +13543A>G were associated with an increased risk.
Tumor Biol. Table 5 Logistic analysis of ADH1B +3170A>G and ADH1C +13044A>G polymorphisms in head and neck squamous cell carcinoma patients and controls according to alcohol consumption and smoking
Gene ADH1B +3170A>G
Non-drinker
Light drinker
Heavy drinker
ADH1C +13044A>G
Non-drinker
Light drinker
Heavy drinker
ADH1B +3170A>G
Light smoker
Heavy Smoker
ADH1C +13044A>G
Light smoker
Heavy smoker OR Adjusted odds ratio for gender and age, 95% CI 95 % confidence interval
There was strong linkage disequilibrium between ADH1B +3170A>G and the other ADH1B SNPs. Therefore, to avoid the effects of +3170A>G, we analyzed the relative risks of the other five SNPs among individuals with the AA genotype of +3170A>G. The results showed that there was no significant correlation between the other five SNPs and the risk of HNSCC. This suggests that the increased risks associated with −957C>A, +3491G>A, and +13543A>G are due to their linkage disequilibrium with ADH1B +3170A>G. Of the SNPs of ADH1C, Ile350Val (ADH1C +13044A>G) and Arg272Gln are the key players. The A and G alleles of +13044 A>G produce isoleucine and valine, respectively. The ADH1C*1 allele is defined by isoleucine and arginine in codons 350 and 272, respectively, and the ADH1C*2 allele by valine and glutamine in codons 350 and 272, respectively [31]. Because Ile350Val and Arg272Gln are in perfect linkage disequilibrium, the Ile350 (A) and Val350 (G) alleles of ADH1C +13044A>G Ile350Val can represent
Genotype
Case (%)
Control (%)
AA AG GG AA AG
55 (9.3) 40 (6.8) 3 (0.5) 23 (3.9) 26 (4.4)
49 (8.3) 33 (5.6) 6 (1.0) 81 (13.7) 61 (10.3)
GG AA AG GG AA AG GG AA AG GG AA AG GG AA AG GG AA AG
7 (1.2) 49 (8.3) 41 (6.9) 16 (2.7) 87 (14.7) 11 (1.9) 0 (0) 43 (7.3) 13 (2.2) 0 (0) 90 (15.3) 15 (2.5) 1 (0.2) 24 (9.2) 19 (7.3) 6 (2.3) 76 (29.2) 66 (25.4)
6 (1.0) 60 (10.2) 31 (5.3) 3 (0.5) 79 (13.4) 9 (1.5) 0 (0) 133 (22.5) 15 (2.5) 0 (0) 88 (14.9) 6 (1.0) 0 (0) 45 (13.6) 31 (9.4) 6 (1.8) 144 (43.6) 94 (28.5)
GG AA AG GG AA AG GG
19 (7.3) 40 (15.4) 9 (3.5) 0 (0) 134 (51.5) 26 (10.0) 1 (0.4)
9 (2.7) 71 (21.5) 11 (3.3) 0 (0) 228 (69.1) 19 (5.8) 0 (0)
OR
95 % CI
p
1.08 0.45
0.59–1.97 0.11–1.88
0.879 0.313
1.50
0.78–2.88
0.247
4.11
1.26–13.51
0.037
1.62 6.54
0.89–2.95 1.80–23.81
0.13 0.002
1.11
0.44–2.82
1.0
2.68
1.18–6.06
0.022
2.44
0.91–6.58
0.104
1.15 1.88
0.54–2.44 0.55–6.45
0.847 0.345
1.33
0.87–2.02
0.198
4.0
1.73–9.26
0.001
1.45
0.55–3.80
0.461
2.33
1.24–4.37
0.009
the ADH1C*1 and *2 alleles, respectively [32]. The ADH1C*1 allele has approximately 2.5-fold higher oxidative function than the ADH1C*2 allele [10]. The frequencies of the alleles of the ADH1C SNPs also differ between ethnic groups. The frequencies of ADH1C*1 and *2 are similar in Caucasian people so that 45–70 % of Caucasians are ADH1C heterozygotes. However, the ADH1C*1 allele is present in 75–90 % of Africans and 85– 100 % of Asians [28]. In other studies conducted among Koreans, the frequency of the ADH1C*1 allele was 94– 95 % [29, 30]. In this study, its frequency was 94 %, in agreement with previous reports. With regard to the association between ADH1C SNPs and the risk of head and neck cancer, conflicting results have been reported. Some studies reported the ADH1C*1/*1 genotype was associated with an increased risk of HNSCC [20, 21]. However, on the contrary to this, other studies reported that ADH1C*1 allele was associated with a reduced risk of
Tumor Biol.
HNSCC [22, 23]. Several studies including some metaanalyses were unable to conclude that there was a correlation between ADH1C +13044A>G and the risk of HNSCC because of the conflicting results n. In the current study, the OR of the AG genotype of ADH1C +13044A>G was not statistically significance in the referent model. However, it indicated an increased relative risk in the dominant (OR 1.94; 95 % CI 1.01–3.17) and co-dominant models (OR 1.97; 95 % CI 1.05–3.71). These results might suggest that the AG genotype of ADH1C +13044A>G is associated with an increased risk of HNSCC; however, its statistical power is not strong. The SNPs −1064C>T, −325G>C, and +5702A>G were associated with an increased risk of HNSCC, though ADH1C +7462T>C was not. However, the increased ORs of ADH1C −1064C>T, −325G>C, and +5702A>G SNPs might be effects of ADH1C +13044A>G because of the strong linkage disequilibrium between ADH1C +13044A>G and the other SNPs. We also analyzed ADH1B +3170A>G and ADH1C +13044A>G as a function of the level of alcohol consumption and smoking to evaluate gene-environment interactions. The relative risk of ADH1B +3170A>G and ADH1C +13044A>G on HNSCC was significantly increased in the drinkers, but not in the non-drinkers. Also, the relative risk of them was significantly increased in the heavy smokers than those in the light smokers. This reflects the gene-environment interaction between alcohol consumption and smoking, and the SNPs of ADH1B and ADH1C. In conclusion, ADH1B SNP +3170A>G His48Arg and ADH1C SNP +13044A>G Ile350Val are associated with increased risks of HNSCC, and could be useful molecular biologic markers for identifying patients at increased risk of developing HNSCC in Koreans. There is also a geneenvironment interaction between these SNPs and alcohol consumption and smoking.
Conflicts of interest None
References 1. Tae K, Lee HS, Park BJ, Park CW, Kim KR, Cho HY, et al. Association of DNA repair gene XRCC1 polymorphisms with head and neck cancer in Korean population. Int J Cancer. 2004;222:805–8. 2. Jeong SW, Tae K, Lee SH, Kim KR, Park CW, Park BL, et al. Cox-2 and IL-10 polymorphisms and association with squamous cell carcinoma of the head and neck in a Korean sample. J Korean Med Sci. 2010;25:1024–8. 3. Ji YB, Tae K, Lee YS, Lee SH, Kim KR, Park CW, et al. XPD polymorphisms and risk of squamous cell carcinoma of the head and neck in a Korean sample. Clin Exp Otorhinolaryngol. 2010;3: 42–7.
4. Blot WJ, McLaughlin JK, Winn DM, Austin DF, Greenberg RS, Preston-Martin S, et al. Smoking and drinking in relation to oral and phayngeal cancer. Cancer Res. 1988;48:3282–7. 5. IARC. Alcoholic consumption and ethyl carbamate. In: IARC monographs on the evaluation of carcinogenic risks to humans. Lyon, France: IARC; 2010, vol. 96. 6. Obe G, Jonas R, Schmidt S. Metabolism of ethanol in vitro produces a compound which induces sister-chromatid exchanges in human peripheral lymphocytes in vitro: acetaldehyde not ethanol is mutagenetic. Mutat Res. 1986;174:47–51. 7. Fang JL, Vaca CE. Detection of DNA adducts of acetaldehyde in peripheral white blood cells of alcohol abusers. Carcinogenesis. 1997;18:627–32. 8. Homann N, Jousimies-Somer H, Jokelainen K, Heine R. Salaspuro M High acetaldehyde levels in saliva after ethanol consumption: methodological aspects and pathogenetic implication. Carcinogenesis. 1997;18:1739–43. 9. Harada S, Misawa S, Agarwal DP, Goedde HW. Liver alcohol dehydrogenase and aldehyde dehydrogenase in the Japanese: isoenzyme variation and its possible role in alcohol intoxication. Am J Hum Genet. 1980;32:8–15. 10. Edenberg HJ. The genetics of alcohol metabolism-role of alcohol dehydrogenase and aldehyde dehydrogenase variants. Alcohol Res Health. 2007;30:5–13. 11. Ji YB, Tae K, Ahn TH, Lee SH, Kim KR, Park CW, et al. ADH1B and ALDH2 polymorphisms and their associations with increased risk of squamous cell carcinoma of the head and neck in the Korean population. Oral Oncol. 2011;47:583–7. 12. Yokoyama A, Muramatsu T, Omori T, Yokoyama T, Matsushita S, Higuchi S, et al. Alcohol and aldehyde dehydrogenase gene polymorphisms and oropharyngolaryngeal, esophageal and stomach cancers in Japanese alcoholics. Carcinogenesis. 2001;22:433–9. 13. Hiraki A, Matsuo K, Wakai K, Suzuki T, Hasegawa Y, Tajima K. Gene–gene and gene–environment interactions between alcohol drinking habit and polymorphisms in alcohol-metabolizing enzyme genes and the risk of head and neck cancer in Japan. Cancer Sci. 2007;98:1087–91. 14. Guo H, Zhang G, Mai R. Alcohol dehydrogenase-1B Arg47His polymorphism and upper aerodigestive tract cancer risk: a meta-analysis including 24,252 subjects. Alcohol Clin Exp Res. 2012;36:272–8. 15. Chang JS, Straif K, Guha N. The role of alcohol dehydrogenase genes in head and neck cancers: a systematic review and metaanalysis of ADH1B and ADH1C. Mutagenesis. 2012;27:275–86. 16. Hashibe M, Boffetta P, Zaridze D, Shangina O, SzeszeniaDabrowska N, Mates D, et al. Evidence for an important role of alcohol- and aldehyde-metabolizing genes in cancers of the upper aerodigestive tract. Cancer Epidemiol Biomarkers Prev. 2006;15: 696–703. 17. Zhang L, Jiang Y, Wu Q, Li Q, Chen D, Xu L, et al. Geneenvironment interaction on the risk of esophageal cancer among Asian poppulations with the G48A polymorphism in the alcohol dehydrogenase-2 gene: a meta-analysis. Tumor Biol. 2014;35: 4705–17. 18. Dura P, Berkers T, van Veen EM, Salomon J, te Morsche RH, Kristinsson JO, et al. Polymorphisms in alcohol-metabolizing enzymes and esophageal carcinoma susceptibility: a Dutch Caucasian case-control study. J Hum Genet. 2013;58(11):742–8. 19. Risch A, Ramroth H, Raedts V, Rajaee-Behbahani N, Schmezer P, Bartsch H, et al. Laryngeal cancer risk in Caucasians is associated with alcohol and tobacco consumption but not modified by genetic polymorphisms in class I alcohol dehydrogenases ADH1B and ADH1C, and glutathione-S-transferases GSTM1 and GSTT1. Pharmacogenetics. 2003;13:225–30. 20. Harty LC, Caporaso NE, Hayes RB, Winn DM, Bravo-Otero E, Blot WJ, et al. Alcohol dehydrogenase 3 genotype and risk of oral cavity and pharyngeal cancers. J Natl Cancer Inst. 1997;89:1698–705.
Tumor Biol. 21. Visapää JP, Götte K, Benesova M, Li J, Homann N, Conradt C, et al. Increased cancer risk in heavy drinkers with the alcohol dehydrogenase 1C*1 allele, possibly due to salivary acetaldehyde. Gut. 2004;53:871–6. 22. Muto M, Takahashi M, Ohtsu A, Ebihara S, Yoshida S, Esumi H. Risk of multiple squamous cell carcinomas both in the esophagus and the head and neck region. Carcinogenesis. 2005;26:1008–12. 23. Wang D, Ritchie JM, Smith EM, Zhang Z, Turek LP, Haugen TH. Alcohol dehydrogenase 3 and risk of squamous cell carcinomas of the head and neck. Cancer Epidemiol Biomarkers Prev. 2005;14:626–32. 24. Cadoni G, Boccia S, Petrelli L, Di Giannantonio P, Arzani D, Giorgio A, et al. A review of genetic epidemiology of head and neck cancer related to polymorphisms in metabolic genes, cell cycle control and alcohol metabolism. Acta Otorhinolaryngol Ital. 2012;32:1–11. 25. Hiyama T, Yoshihara M, Tanaka S, Chayama K. Genetic polymorphisms and head and neck cancer risk (Review). Int J Oncol. 2008;32:945–73. 26. Sturgis EM, Dahlstrom KR, Guan Y, Eicher SA, Strom SS, Spitz MR, et al. Alcohol dehydrogenase 3 genotype is not associated with risk of squamous cell carcinoma of the oral cavity and pharynx. Cancer Epidemiol Biomarkers Prev. 2001;10:273–5. 27. Osier MV, Pakstis AJ, Soodyall H, Comas D, Goldman D, Odunsi A, et al. A global perspective on genetic variation at the ADH genes
reveals unusual patterns of linkage disequilibrium and diversity. Am J Hum Genet. 2002;71:84–99. 28. Brennan P, Lewis S, Hashibe M, Bell DA, Boffetta P, Bouchardy C, et al. Pooled analysis of alcohol dehydrogenase genotypes and head and neck cancer: a HuGE review. Am J Epidemiol. 2004;159:1–16. 29. Lee HC, Lee HS, Jung SH, Yi SY, Jung HK, Yoon JH, et al. Association between polymorphisms of ethanol-metabolizing enzymes and susceptibility to alcoholic cirrhosis in a Korean male population. J Korean Med Sci. 2001;16:745–50. 30. Jung SH, Lee HC, Yoon JH, Lee HS, Kim CY. The association between genetic polymorphisms of ethanol-metabolizing enzymes and susceptibility to alcoholic liver cirrhosis. Korean J Hepatol. 1998;4:1–11. 31. Höög JO, Hedén LO, Larsson K, Jörnvall H, von Bahr-Lindström H. The gamma 1 and gamma 2 subunits of human liver alcohol dehydrogenase. cDNA structures, two amino acid replacements, and compatibility with changes in the enzymatic properties. Eur J Biochem. 1986;159:215–8. 32. Packer BR, Yeager M, Burdett L, Welch R, Beerman M, Qi L, et al. SNP500Cancer: a public resource for sequence validation, assay development, and frequency analysis for genetic variation in candidate genes. Nucleic Acids Res. 2006;1(34):D617–21.
