br Results br Association analysis of PARP SNPs in BC
3.1. Association analysis of PARP1 SNPs in BC patients and healthy individuals
In this study, to perform the association of PARP1 studied SNPs with the prevalence of BC, 186 BCE patients and 200 healthy donors were genotyped. The distributions of allele and genotype frequencies of studied SNPs in both study groups are shown in Table 1. All the gen-otypic frequencies were in Hardy–Weinberg equilibrium (p > 0.05).
Genotype frequencies of PARP1 gene polymorphisms in breast cancer cases and controls.
rs907187 was related to increased BC risk in the CC and GG genotypes,
3.2. Linkage disequilibrium results
Analyses indicated that all three SNPs had a minor allele frequency higher than 5%. The linkage disequilibrium plot of the studied SNPs was represented in Fig. 1. The pairwise linkage disequilibrium is given for each pair of SNPs showing pairwise D′ values (Fig. 2). The observed pairwise D′ values showed that rs4653734 and rs907187 were in strong linkage disequilibrium. Rs4653734 and rs907187 indicated 100% linkage disequilibrium. Also, rs1136410 and rs4653734 were at 66% of linkage disequilibrium.
3.3. Comparison of haplotype frequencies in BC patients and healthy controls
3.4. Genotypes and clinical data
Among 186 patients with BC, the clinical characteristics of 86 in-dividuals were completely available. Out of 86 patients, 44 individuals had invasive ductal carcinoma (IDC), 42 patients had invasive lobular carcinoma (ILC), 12 individuals had Lobular Carcinoma in Situ (LCIS) and 8 patients had ductal carcinoma in situ (DCIS). Among 44 patients with IDC, all were stage 3, and the cancer had spread to lung (16 pa-tients), bone (10 patients), liver (4patients), spleen (2 patients), and in 12 patients BC metastasized to more than a location.
Among 42 patients with ILC, 4 patients were at stage 4 and 38 patients were at stage 3. The location of metastases in these 42 patients was lung (12 patients) and bone (12 patients). Also in 18 patients BC spread to two or more tissues. The characteristics of the 86 patients (ILC and IDC) presented in Table 3.
3.5. Bioinformatics results
TFBIND, Genomatix and Regulome DB softwares were applied to predict TF Belinostat (PXD101) and the analyses revealed that the single-nu-cleotide alterations in rs4653734 (C/G) and rs907187 (C/G) regions possibly modifies the TF binding site. To determine the relationships between the identified SNPs and regulatory sequences, the list of reg-ulatory elements containing binding sites were obtained from Regulome DB. Online servers predicted that E2F and E2F-4 TFs bind to C allele of rs907187; however, these binding sites were not predicted in the presence of the mutant allele (G). Previous studies reported that E2F-1 and E2F-4 are involved in the expression of PARP1. In addition, it has been predicted that SHARP1 bound to wild allele (C) of rs4653734 whereas, mutant allele (G) didn't bound to this TF. These basic helix-loop-helix TFs have oncogenic or tumor suppressor role in development of various cancers. Another bioinformatics sub-analysis was performed for rs4653734 site as a putative CpG-SNP through MethPrimer tool. Results indicated that rs4653734 and rs907187 were in a CpG island of PARP1 promoter region; based on this, substituting the wild-type alleles of rs4653734 and rs907187 with mutant alleles (C allele with G allele for both of them) revealed that rs4653734 in mutant form can make a
Fig. 2. Schematic visualization of CpG islands in the PARP1 promoter including rs907187 and rs4653734 SNPs using Methprimer tool. Schematic map of the CpG islands indicated two SNPs in wild type (a) and mutant (b) forms. The map extended form 100 bp upstream to 100 bp downstream of two polymorphisms.
Haplotype distribution between patients and control subjects.
Overall Controls Patients p-Value P p-Value
P p-value refers to Permutation p-value.
Clinical characteristics of the 82 patients (ILC and IDC) with breast cancer.
Location of primary tumor IDC ILC
N refers to Number. IDC and ILC means Invasive Ductal Carcinoma and Invasive Lobular Carcinoma, respectively.
new CpG site in the island (Fig. 2). The recessive carriers of rs4653734 (G allele) may have a different methylation pattern compared with the dominant ones (C allele).
PARP1 regulates many biological processes including DNA repair, maintenance of genomic integrity, regulation of telomerase activity, metabolism, signaling, transcription regulation, chromatin archi-tecture, inflammation and cell death (Schreiber et al., 2006; Kim et al., 2005; Gibson and Kraus, 2012). Breast cancer is the major cause of female cancer-related death (DeSantis et al., 2015). Ongoing statistical models for evaluating BC risk have restricted specificity and sensitivity (Amir et al., 2010). SNPs play critical roles in different types of cancers such as breast cancer and genotyping of key SNPs could be a more accurate tool for cancer diagnosis and management (Nahon and Zucman-Rossi, 2012; Multani and Saranath, 2016). The current study investigated the association of the rs907187, rs4653734 and rs1136410 SNPs in PARP1 gene with susceptibility to BC. To the best of our knowledge, this is the first study to examine the distribution of rs4653734 and its possible relation to BC development. Clinical data indicate that lung metastasis was one of the most frequent breast me-tastasis of IDC. Moreover, lung and bone were the most common site of extra mammary metastasis of ILC.