Research Highlights

A Polymorphism in the SOD2 Gene and its Association with Esophageal Cancer

Evaluation of: Sun L, König IR, Homann N: Manganese superoxide dismutase (MnSOD) polymorphism, alcohol, cigarette smoking and risk of oesophageal cancer. Alcohol Alcoholism 44(4), 353–357 (2009).

As alcohol consumption, smoking and Barrett‘s syndrome can result in gastro-esophageal reflux, they are important risk factors for esophageal cancer. In addition, Zhang et al. have shown that a genetic component may be implicated in the pathogenesis of esophageal squamus cell carcinoma and adenocarcinoma [1]. With regard to the molecular mechanism involved in oesophegeal carcinogenesis, oxidative stress is likely to play a role and is mediated by the production of reactive oxidative species (ROS). The manganese superoxide dismutase (MnSOD) enzyme, which is found in mitochondria, is part of the antioxidant defense system that counteracts the possible damaging effects of ROS. Polymorphisms in the SOD2 gene encoding MnSOD, which cause an altered level of gene expression, may subsequently influence the cell‘s ability to counteract ROS-induced damage. A polymorphism in SOD2, consisting of a T to C substitution and resulting in an amino acid change from valine (Val) to alanine (Ala) has been associated with several types of cancer [2], but studies on the link with esophageal cancer have not been reported.

Sun, König and Homann examined three Caucasian cohorts for the MnSOD allele frequencies; an esophageal cancer cohort (adenocarcinoma: n = 61; squamous cell carcinoma: n = 109), a population-based healthy cohort (n = 400) and a cohort of heavy drinkers (n = 160) [3]. Although the Ala/Ala genotype was more frequent in cancer patients (27.7%) than in healthy blood donors (12.5%) and heavy drinkers (23.1%), these differences were not statistically significant when the researchers controlled for age, alcohol and tobacco use. Similarly, nonsignificant differences in genotype frequencies were obtained when the cancer cohort was subdivided according to cancer type and compared with the healthy controls or heavy drinkers. Interestingly, in an additional, exploratory analysis, it was found that SOD2 Ala/Ala genotype carriers among the three cohorts did consume significantly more alcohol (uncorrected p < 0.0001) and cigarettes (uncorrected p = 0.041) than those carrying Val/Ala or Val/Val genotypes.

The Sun et al. study was the first to examine the association between a polymorphism in the SOD2 gene and esophageal cancer [3]. However, whether the SOD2 gene and environmental risk factors interact in increasing the risk for esophageal carcinoma remains a topic for future research. Esophageal cancer is a complex disease that is most likely affected by multiple genes [4], environmental factors and by the interaction between these genes and environment. Hence, the relationship between SOD2 genotype and esophageal cancer might be dependent on levels of alcohol use.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

Relationships between the ADH1C Gene, Alcohol Use and Risk Factors for Cardiovascular Disease

Evaluation of: Latella MC, Di Castelnuovo A, De Lorgeril M et al.: Genetic variation of alcohol dehydrogenase type 1C (ADH1C), alcohol consumption, and metabolic cardiovascular risk factors: Results from the IMMIDIET study. Atherosclerosis (2009) (Epub ahead of print).

Epidemiological studies showed that light- to-moderate alcohol consumption protects against cardiovascular diseases (e.g., [1]). The mechanism involved in the cardioprotection through alcohol might involve its enhancing effect on high-density lipoprotein (HDL) cholesterol levels. In addition, interindividual variability in alcohol metabolism may help understand the cardioprotective effects of alcohol [2]. In the human body, alcohol is converted by alcohol dehydrogenase (ADH) into acetaldehyde, which is subsequently oxidized into acetate by acetaldehyde dehydrogenase. The γ2 low-activity allele of a polymorphism in the ADH1C gene (rs689) has been associated with higher levels of HDL cholesterol and a lower risk at myocardial infarction, in individuals consuming at least one alcoholic drink per day [3]. These results, however, have not been replicated in subsequent studies.

Latella et al. examined 1604 subjects (from 974 male–female pairs participating in the IMMIDIET study) from Italy, Belgium and England for relationships between ADH1C genotype, alcohol consumption and various metabolic cardiovascular risk factors in a cross-sectional design [2]. In general, no direct associations between ADH1C genotype and cardiovascular risk factors or alcohol use were found. Independent of ADH1C genotype, alcohol intake was significantly and positively related to several cardiovascular risk factors, including HDL-cholesterol levels. An interaction effect in which ADH1C γ2 allele carriers who consumed over 48 g of pure ethanol per day showed higher levels of BMI and waist circumference than males who did not carry this allele, but this was not observed in females. This effect of ADH1C genotype did not seem to be mediated via the levels of either HDL cholesterol or other metabolic factors investigated. In fact, ADH1C γ2 allele carriers showed even lower HDL-cholesterol levels at high-alcohol consumption levels than subjects homozygous for the g1 allele.

The attempt to clarify the complex relationships between alcohol, specific genes and cardiovascular risk factors in the cross-sectional study by Latella and coworkers is admirable, as is their attempt to use multivariate analyses to approach this subject [2]. However, a prospective study, in which the participants are regularly examined for cardiovascular disease, would potentially provide a more ideal setting for such a study. Although the researchers used a multivariate design, ample different analyses were still carried out, which may have created a multiple testing problem. Since p-values were not corrected, and effect sizes were relatively low, the results should be replicated before firm conclusions can be drawn from this study.

Polymorphisms in the CYP2E1, GSTM1 and GABRG2 Genes and Risk for Alcoholic Liver Cirrhosis

Evaluation of: Khan AJ, Ruwali M, Choudhuri G et al.: Polymorphisms in cytochrome P450 2E1 and interaction with other genetic risk factors and susceptibility to alcoholic liver cirrhosis. Mutat. Res. 664, 55–63 (2009).

Although heavy alcohol consumption is a major risk factor for liver diseases, only 10–20% of heavy drinkers develop liver cirrhosis. Cirrhosis is marked by the replacement of normal liver tissue by collagen scar tissue, which disrupts the blood flow through the liver and results in loss of liver function [1]. Genetic factors, particularly those that influence alcohol metabolism or risk for alcoholism, may be involved in the risk for cirrhosis [1,2]. Since the cytochrome P450 2E1 (CYP2E1) enzyme is a key player in alcohol oxidization, differences in CYP2E1 activity caused by polymorphisms in the CYP2E1 gene may help explain individual variability in susceptibility for liver cirrhosis. Several polymorphisms have been identified in the CYP2E1 gene, of which the CYP2E1*5B (RsaI) SNP (C>T) leads to an increased transcription and enzyme activity. In previous studies, some of the CYP2E1 polymorphisms have been related to liver diseases, though with inconsistent results.

Kahn et al. examined associations between three polymorphisms in the CYP1E1 gene (CYP2E1*5B, CYP2E1*6 and MspI) and alcohol liver cirrhosis in an Indian sample, consisting of 160 alcoholic cirrhotic patients, 125 nonalcoholic cirrhotic patients, 250 nonalcohol dependent controls and 100 subjects who were alcohol dependent but showed no signs of liver disease [3]. In addition, two polymorphisms in genes involved in the detoxification of ROS and in alcohol consumption (glutathione S-transferases M1 [GSTM1] and γ-aminobutyric acid receptor γ2 [GABRG2], respectively) were also included in the analyses. Allele frequency of the rare CYP2E1*5B variant, but not of the more common CYP2E1*6 and MspI variants, was significantly higher in alcoholic cirrhotic patients compared with all other groups (0.002 ≤ p ≤ 0.034). The haplotype carrying the variant allele of CYP2E1*5B and the wild-types of the CYP2E1*6 and MspI was significantly more frequent in alcoholic cirrhotic patients than in nonalcoholic controls (odds ratio: 5.3; 95% CI: 1.46–19.60; p = 0.005). Also, a GSTM1 gene deletion polymorphism and a functional GABRG2 variant (intron 8, 84435G>A) separately showed a significant association with alcohol-related cirrhosis. Two combined gene effects were also found: carriers of both the CYP2E1*5B variant allele and the GSTM1 null allele showed a significantly higher risk for alcoholic liver cirrhosis (odds ratio: 7.7; 95% CI: 1.60–37.16; p = 0.003), as did those with both the CYP2E1*5B variant allele and the GABRAG2 A allele (odds ratio: 6.4; 95% CI: 1.92–21.50; p < 0.001).

That epistasis (i.e., the interaction between genes) is highly likely to affect complex human traits such as liver diseases is nicely shown by the results of the paper by Kahn and coworkers [3]. Testing for epistasis may be especially important since the effect of one locus can be masked by the effects of a second locus [4]. Epistasis effects may therefore explain considerably more phenotypic variance than two single variants, helping to explain the relative paucity of significant findings so far identified in genome-wide association studies on complex disorders [5]. There are presumably more than three loci involved in the development of alcoholic liver diseases and testing for multiway genetic interactions would further increase our knowledge of the etiology of liver diseases. However, a concern with the investigation presented in the current paper is the fact that the frequency of the risk alleles (especially the CYP2E1*5B allele) is low, reducing generalizability of the findings and making replications in larger samples necessary before firm conclusions can be drawn.

Interactions between the GPX1 Pro198Leu Polymorphism and Substance Use and Susceptibility to Colorectal Cancer

Evaluation of: Hansen RD, Krath BN, Frederiksen K et al.: GPX1 Pro198Leu polymorphism, erythrocyte GPX activity, interaction with alcohol consumption and smoking, and risk of colorectal cancer. Mutat. Res. 664, 13–19 (2009).

There is evidence that oxidative stress plays a role in the pathogenesis of colorectal cancer, since increased levels of ROS and lipid peroxidation have been found in colorectal cancer tissue. The antioxidant GPX1 and the DNA-repairing OGG1 enzymes are known to counteract oxidative DNA damage. Therefore, genetic polymorphisms in GPX1 and OGG1, such as the GPX1 rs1050450 (Pro¹⁹⁸Leu) and the OGG1 rs1052134 (Ser³²⁶Cys) variants, that result in lower enzyme expression levels, might play a role in colorectal carcinogenesis. In addition, fruit, vegetable and alcohol intake, tobacco smoking and selenium supplementation may influence GPX1 enzyme activity. Although the GPX1 and OGG1 variants have not consistently been associated with colorectal cancer in previous studies [1], Kim and colleagues did report interactions between the OGG1³²⁶Cys allele and meat consumption, and between the OGG1³²⁶Cys allele and smoking, which both increased the risk for colorectal cancer [2].

Hansen et al. examined whether 375 patients with colorectal cancer and 779 control subjects selected from a large, prospective cohort study differed in GPX1¹⁹⁸Leu or OGG1³²⁶Cys allele frequencies and/or various lifestyle factors [3]. GPX1 Pro¹⁹⁸Leu and OGG1 Ser³²⁶Cys genotypes were not directly related to colorectal cancer. Direct differences in lifestyle factors between patients with colorectal cancer and controls were also not observed. However, significant interaction effects were found; alcohol consumption and smoking augmented risk of colorectal cancer among homozygous GPX1¹⁹⁸Leu allele carriers (p = 0.02, uncorrected), while intake of vitamin C from diet reduced risk among homozygous carriers of the GPX1¹⁹⁸Pro allele (p = 0.05). In contrast to the findings by Kim et al., the OGG1 Ser³²⁶Cys polymorphism showed no significant interaction with lifestyle factors in the current study. Hansen et al. also measured erythrocyte GPX activity in prediagnostic blood samples. In univariate analyses GPX activity proved to be lower in women, GPX1¹⁹⁸Leu allele carriers, smokers and subjects who consumed less fruit and vegetables. In the multivariate analysis, however, these findings did not remain significant, except for fruit and vegetable intake, which explained approximately 5% of variance. Selenium supplementation and alcohol use were not related to erythrocyte GPX enzyme activity.

The setup of this study, a large (n = 57,053) prospective follow-up design, is ideally suited for the analyses of interest. Multivariate analyses like these, including both environmental and genetic factors, such as the one described by Hansen and coworkers [3], are likely to provide a better insight into the etiology of colorectal cancer and other complex disorders than the analysis of single factors or domains. Nevertheless, the number of cases that could be investigated is still limited, and findings from this study did not reach significance when a Bonferroni correction for multiple testing was applied. This study clearly needs replication, but is an important starting point that future studies on colorectal carcinogenesis may elaborate on.