DNA Repair Capacity as a Marker of Breast Cancer Susceptibility.

摘要

Introduction: The wide-ranging prognostic implications of a breast cancer diagnosis highlight the need to better enable women to make informed decisions regarding screening and treatment options. As several cancer susceptibility syndromes have been linked to germline mutations resulting in defective DNA repair, including the predisposition to breast cancer due to BRCA1 and BRCA2 mutations, more subtle defects in DNA repair capacity may contribute to the components driving differential susceptibility within the general population. Hence, understanding the role of DNA repair capacity in breast cancer onset may aid in the development of a more comprehensive risk profile, thereby furthering the effort to target relevant populations for early screening. In the studies undertaken for this dissertation, we employed various methodologies capturing endpoints across different repair pathways detectable in blood to both further elucidate the etiologic basis of breast cancer development and leverage the information into the potential development of a screening biomarker. Methods: For the phenotypic assessment of nucleotide excision repair (NER) capacity, we developed an ELISA-based method to determine benzo(a)pyrene diolepoxide (BPDE)-DNA adduct capacity in lymphoblastoid cell lines. Gene expression levels were assessed with pre-designed Taqman kits in RNA-derived cDNAs from mononuclear cells using a real-time PCR-based platform. Methylation analysis was conducted with in-house designed assays on bisulfite-converted DNA from mononuclear cells using a pyrosequencing platform. Finally, single nucleotide polymorphisms (SNP) genotyping was assessed in DNA derived from white blood cells with pre-designed Taqman SNP genotyping assays using a real-time PCR-based platform. All studies were conducted in sister-sets enrolled in the New York site within the Breast Cancer Family Registry and all statistical analysis was conducted using the R Foundation for Statistical Computing (2011). Results: We did not detect an association between the ELISA-based phenotypic assessment of NER capacity in the lymphoblastoid cells lines of the sister-sets (n=246, 114 sister-sets) and breast cancer risk (OR = 1.0, 95%CI=0.95, 1.04). Furthermore, we did not observe a correlation with previously determined NER capacity in the same population using an immunohistochemical-based method (r= -0.01, p=0.86). In our gene expression study (n=569, 218 sister-sets), women in the lowest tertile of ATM expression had a heightened risk of breast cancer compared to women in the highest tertile of expression, adjusted for age at blood draw and smoking status (OR=2.12, 95%CI=1.09, 4.12). This association was largely restricted to women with an extended family history of breast cancer (pinteraction = 0.06). Additionally, women in the lowest tertile of MSH2 expression also had a heightened risk of breast cancer compared to women in the highest tertile of expression, adjusted for age at blood draw and smoking status (OR=2.75, 95%CI=1.31, 5.79). The association observed between reductions in ATM expression level and breast cancer risk was lost upon incorporating previously determined end-joining capacity of EcoRI-generated sticky end substrates (OR=1.28, 95%CI=0.15, 11.2) and HincII-generated blunt end substrates (OR=1.55, 95%CI=0.15, 15.5) into the model, suggesting that the impact on risk due to reductions in ATM expression maybe partially driven by the reduction in double strand break repair capacity. In our study investigating breast cancer risk due to the impact of epigenetic modulation on DNA repair gene activity (n=569, 218 sister-sets), no association with risk was observed due to differential promoter methylation levels of BRCA1 (OR=1.09, 95%CI=0.98, 1.20), MLH1 (OR=1.19, 95%CI=0.91, 1.55) or MSH2 (OR=0.89, 95%CI=0.48, 1.64). Furthermore, no correlation between BRCA1 and expression (r=-0.05, p=0.39) or MSH2 methylation and expression (r=-0.04, p=0.39) was observed. Finally, our mismatch repair genotyping study (n=714, 313 sister-sets) indicated an association between the variant MutY_rs3219489 (OR=2.23, 95%CI=1.10, 4.52) and breast cancer risk, as well as a borderline association with risk due to the variant MSH2_rs2303428 (OR=1.71, 95%CI=0.99, 2.95). Furthermore, a protective effect was observed due to the variant MLH3_rs175080, restricted to women without an extended family history of breast cancer (pinteraction = 0.03). Conclusion: These studies suggest that the deregulation of targets spanning various DNA repair pathways contribute to the risk of familial breast cancer.