Inter-individual variations in DNA adduct levels assessed by analysis of multiple genetic polymorphisms in smokers

Hans B Ketelslegers¹, Ralph W.H. Gottschalk¹, Roger W.L. Godschalk¹,  Ad M. Knaapen¹, Frederik J. van Schooten¹, Robert F.M.H. Vlietinck^2,3, Jos C.S. Kleinjans¹ and Joost H.M. van Delft<sup>1*

1</sup>Department of Health Risk Analysis and Toxicology, Maastricht University, The Netherlands. ²Department of Population Genetics, Genomics and Bioinformatics, Maastricht University, The Netherlands. ³ Centre for Human Genetics, University of  Leuven, Belgium

Large inter-individual differences have been reported in susceptibility to health effects caused by exposure to (environmental) carcinogens. It was postulated that genetic polymorphisms may at least partly explain this variation. Because of the high number of single nucleotide polymorphisms (SNPs) that are involved in the process of carcinogenesis, high throughput genotyping methods are needed. To this end, we developed a genotyping method capable of genotyping a high number of SNPs in large populations.¹ So far, a database was developed containing 69 SNPs in 44 genes involved in biotransformation, DNA repair, oxidative stress, cell cycle control & apoptosis and inflammation.

A first application study was done in order to explain the inter-individual differences in levels of DNA damage in a population of 65 Caucasian smokers. Genetic polymorphisms in genes involved in pathways that lead to DNA damage and its repair may explain part of the large inter-individual variation in DNA adduct levels in smokers. To this end, we investigated 19 polymorphisms in 12 genes involved in carcinogen metabolism, DNA repair and oxidative stress defense on DNA adduct levels (determined by ³²P-postlabelling) in lymphocytes of 63 healthy Caucasian smokers. The total number of alleles that were categorized as putatively high risk alleles, appeared positively related with bulky DNA adduct levels (p=0.001). Subsequently, to investigate which polymorphisms may have the highest contribution to DNA adduct levels observed in these smokers, discriminant analysis was performed. In the investigated set of polymorphisms, GSTM1*0 (p<0.001), mEH*2 (p=0.001) and GPX1 (p<0.001) in combination with the level of exposure (p<0.001) were found to be the key effectors. This is the first demonstration of the involvement of GPX in DNA adduct formation. DNA-adduct levels in subjects with a relatively high number of risk alleles for these three genes were more than 2-fold higher than in individuals having no risk alleles (1.97±1.026 adducts per 10⁸ nucleotides versus 0.79±0.49). This study demonstrates that analysis of multiple genetic polymorphisms may predict the inter-individual variations in DNA adduct levels caused by exposure to a complex mixture of compounds such as cigarette smoke. Noteworthy, all three genes are involved in processes on deactivation of reactive carcinogenic metabolites. It is concluded that discriminant analysis is an important statistical tool studying the impact of multiple genotypes on molecular biomarkers, like DNA adduct levels.