The identification of allelic variants of human genes is of great importance when assessing genetic susceptibility. The emerging role of genetic polymorphisms in association studies has created the need for high throughput genotyping methodologies that allow a more rapid identification of relevant polymorphisms related to individual cancer risk enabling the extension to large-scale association studies. DNA pooling methodology may be of great importance considering the cost, time and labor that are involved in large-scale genotyping analysis carried out on individual samples. Alternatively, when using pooled samples which are made up of DNA from many individuals treated as a single sample, these factors are decrease drastically. In this way, the use of DNA pooling methodology, as a pre-selection tool, allows the identification of the most relevant polymorphisms to be studied, facilitating the estimation of the allelic frequency of each SNP in different populations. The present study initially aimed to validate the DNA pooling approach for the identification of genetic polymorphisms potentially associated with individual cancer risk generating pools with known allelic frequencies and using studies ongoing in the laboratory. Finally, our main aim was to test the accuracy of the pooled DNA analysis comparing the results of the allelic frequencies determined using pooled samples with the allelic frequency previously estimated by individual genotyping and previously published. In order to analyze the possibility of establishing differences between populations, we created DNA pools using a Portuguese control population, a breast cancer population and a Xavante Indian population characterized by a total absence of breast cancer cases. The pools were firstly created with known allelic frequencies, previously determined by individual genotyping, and, latter, randomly incremented in sample size to 200 patients and controls. Our results showed that the DNA pooling approach was a useful tool for the analysis of allelic distribution in the different populations studied. In conclusion, our results showed that this methodology can be applied as an effective approach to identify SNPs of importance in genetic susceptibility to disease which may be used in association studies conducted subsequently by individual genotyping.

Background: MMR is responsible for the repair of base-base mismatches and insertion/deletion loops. Besides this, MMR is also associated with an anti-recombination function, suppressing homologous recombination. Losses of heterozygosity and/or microsatellite instability have been detected in a large number of skin samples from breast cancer patients, suggesting a potential role of MMR in breast cancer susceptibility. Methods: We carried out a hospital-based case-control study in a Caucasian Portuguese population ( 287 cases and 547 controls) to estimate the susceptibility to non-familial breast cancer associated with some polymorphisms in mismatch repair genes (MSH3, MSH4, MSH6, MLH1, MLH3, PMS1 and MUTYH). Results: Using unconditional logistic regression we found that MLH3 (L844P, G>A) polymorphism GA (Leu/Pro) and AA (Pro/Pro) genotypes were associated with a decreased risk: OR = 0.65 (0.45-0.95) ( p = 0.03) and OR = 0.62 (0.41-0.94) ( p = 0.03), respectively. Analysis of two-way SNP interaction effects on breast cancer revealed two potential associations to breast cancer susceptibility: MSH3 Ala1045Thr/MSH6 Gly39Glu - AA/TC [ OR = 0.43 (0.21-0.83), p = 0.01] associated with a decreased risk; and MSH4 Ala97Thr/MLH3 Leu844Pro - AG/AA [ OR = 2.35 (1.23-4.49), p = 0.01], GG/AA [ OR = 2.11 (1.12-3,98), p = 0.02], and GG/AG [ adjusted OR = 1.88 (1.12-3.15), p = 0.02] all associated with an increased risk for breast cancer. Conclusion: It is possible that some of these common variants in MMR genes contribute significantly to breast cancer susceptibility. However, further studies with a large sample size will be needed to support our results.

Glutathione-S-transferases (GSTs) are a superfamily of phase II metabolizing enzymes that catalyse the detoxification of a large range of endogenous and exogenous toxic compounds, playing an important role in protecting cells against damage, through glutathione conjugation with electrophilic substances. Polymorphic variation in these enzymes that affect its activity seems to be related to individual susceptibility to various human diseases, including cancer. Of the GST super-family, the alpha class GSTs have commonly been described as one of the most versatile class, since it is responsible for detoxification of compounds such as bilirubin, bile acids and penicillin, thyroid and steroid hormones, allowing its solubilization and storage in the liver. Among the alpha class, GSTA1 and GSTA2 isoforms are the most widely expressed in human tissues. Additionally, these enzymes can catalyse conjugation of the nitrogen mustard group of alkylating anticancer drugs, some heterocyclic amines and alpha,beta-unsaturated aldehydes. Since some risk factors for increased breast cancer risk could be related to high production of reactive oxygen species during the metabolism of estrogens by catechol estrogens, or to the exposure to genotoxic compounds, and some of these toxic compounds are usually metabolized by GSTA2, we carried out a hospital based case-control study in a Caucasian Portuguese population (291 breast cancer patients without familiar history of breast cancer and 547 controls matched for age, sex and ethnicity) in order to evaluate the potential modifying role of three non-synonymous polymorphisms in the GSTA2 gene (P110S Ex 5+56C>T;, rs2234951; S112T Ex5+63G>C, rs2180314 and E210A Ex7+83A>C, rs6577) on the individual susceptibility to breast cancer. Our data show that the Studied polymorphisms are in strong linkage disequilibrium, but no association was observed between individual GSTA2 polymorphisms and haplotypes and individual susceptibility to breast cancer.