Role of Four ABC Transporter Genes in Pharmacogenetic Susceptibility to Breast Cancer in Jordanian Patients

Breast cancer pharmacogenetics is increasingly being explored due to chemotherapy resistance among certain classes of patients. The ATP binding cassette (ABC) transporter genes have been previously implicated in breast cancer progression and drug response. In the present study, single nucleotide polymorphisms (SNPs) from the ABCC1, ABCC2, ABCB1, and ABCG2 genes were screened in breast cancer patients and healthy volunteers from the Jordanian-Arab population. Only the ABCB1 SNPs showed a significant association with BC in Jordanian-Arab patients, and the ABCB1 SNP rs2032582 exhibited a strong genotypic association with BC. With regard to the clinical characteristics of BC, the ABCC2 SNPs rs2273697 and rs717620 were found to be significantly associated with age at breast cancer diagnosis and breastfeeding status, while the ABCB1 SNP rs1045642 was significantly associated with age at breast cancer diagnosis. In terms of pathological characteristics, the ABCC1 SNP rs35628 and the ABCB1 SNP rs2032582 were significantly associated with tumor size, the ABCC2 SNP rs2273697 was significantly associated with estrogen receptor status, and the ABCG2 SNP rs2231142 was significantly associated with axillary lymph node status. In this current study, we assume that significant genetic variants within the ABC superfamily may increase the risk of breast cancer among Jordanian women. Furthermore, these variants might be responsible for worse BC prognosis.


Introduction
Breast cancer (BC) is the most common female malignancy in the majority of countries [1]. Arab populations suffer from lower but steadily rising BC incidence rates compared to their American and European counterparts, and the clinical characteristics of the disease also differ between the aforementioned populations [2]. Such population-level differences in BC predisposition have been attributed to genetics and have been widely investigated, with different mutations having different levels of association with BC [3]. Compounding this issue is the fact that Arab BC genetics are not well researched, and much less is known about the genes involved in BC progression and drug response in Arab patients [4].
The ATP binding cassette (ABC) transporters comprise seven subfamilies of membrane proteins that facilitate the transport and modulate the effects of a wide range of drugs and their metabolites [5,6]. Remarkably, an overexpression of certain ABC transporters in cancer cell lines resulted in multidrug resistance (MDR) and a potential failure of chemotherapy [7,8]. For example, the ABCC1 gene, also known as multidrug resistance-associated protein 1 (MRP1), is associated with worsened prognoses in a wide range of tumors, while the ABCC2 gene was found to contribute to drug resistance [9,10]. Likewise, the ABCB1 gene is highly 2 Journal of Oncology polymorphic and induces chemoresistance by preventing drug accumulation in cancer cells [7]. In addition, the ABCG2 gene, also known as the breast cancer resistance protein (BCRP), is responsible for the transport of many conventional chemotherapeutics and causes MDR in various cancer cells [11].
In the present study, four SNPs of ABC transporter genes, namely ABCC1, ABCC2, ABCB1, and ABCG2, were screened in Jordanian Arabs with and without breast cancer. Previous reports have indicated that these genes play a critical role in increasing tumor risk, especially in breast cancer [9,11]. The aim of this study is to determine whether the aforementioned genes play a significant role in Jordanian breast cancer patients.

Ethical Approval and Conduct.
The present study was given ethical approval by the Institutional Review Board (IRB) at Jordan University of Science and Technology. Written informed consent was obtained from all participants in this study before blood sample withdrawal.

Study Population and Design.
The study cohort consisted of 222 women diagnosed with breast cancer as well as 218 healthy matched volunteers. All participants were recruited from the Jordanian population and were of Arab descent. 5 ml of blood were withdrawn from each participant into EDTA tubes and refrigerated until DNA extraction.

Genomic Extraction and
Genotyping. Genomic DNA was extracted from a total of 440 blood samples using the Wizard5 Genomic DNA Purification Kit (Promega, USA). Extracted DNA was evaluated in terms of concentration (ng/ l) and purity (A260/280) quantity using the Nano-Drop ND-1000 UV-Vis Spectrophotometer (BioDrop, UK). DNA samples were then loaded onto an agarose gel to confirm product quality. Samples that met our requirements were diluted using nuclease-free water for a final concentration of 20 ng/ l and a final volume of 30 l. Genotyping was carried out by the Melbourne node of the Australian Genome Research Facility (AGRF) using the Sequenom MassAR-RAY5 system (iPLEX GOLD) (Sequenom, San Diego, CA, USA).

Denomination of Genotypic-Phenotypic Correlation.
In this study, several clinical and pathological features of BC were investigated in correlation with the studied variants. Clinical and pathological information for patients was collected from their medical records. P values were selected to estimate the association between SNPs and risk of BC. The analyses were done per genotype.

Statistical Analysis.
Case-control analyses were carried out using different statistical software. Allelic and genotypic frequencies were calculated using the Hardy-Weinberg equilibrium (HWE) equation (Court lab -HW calculator) (http://www.oege.org/software/hwe-mr-calc.html). The Statistical Package for the Social Sciences (SPSS), version 25.0 (SPSS, Inc., Chicago, IL) was used to calculate the p values that allowed discrimination between cases and controls in association with the genotype. It also facilitated the analysis of the different genotype models. On the other hand, genotypephenotype assessment was performed using the Chi-Square test and ANOVA tests [12]. P value denoted statistical significance if they were less than 0.05. Table 1 displays the SNPs of the ABCC1, ABCC2, ABCB1, and ABCG2 candidate genes. All of the polymorphic SNPs were tested for minor allele frequencies (MAF) and HWE p values in both the cases and controls (Table 1).

Association between ABC Transporter SNPs and Breast
Cancer (BC). The allelic and genotypic frequencies of the ABC transporter SNPs were determined for both cases and controls ( Table 2). All three ABCB1 SNPs were found to be significantly associated with BC in Jordanian patients, with rs1045642, rs1128503, and rs2032582 having p values of 0.01164587, 0.01610842, and 0.03565022, respectively. Figure 1 shows a representative scatter pattern for rs1045642 of ABCB1. In contrast, only the rs2032582 SNP of ABCB1 showed a strong genotypic association with BC (p value = 1e −8, OR =6.72, 95% CI = 4.27 to 10.57). rs2032582 is a triallelic polymorphism comprising the A, C, and T (minor) alleles (the homozygous TT variant was not estimated in the current study population). None of the other investigated SNPs showed any significant correlation with BC, as all the allelic and genotypic frequencies were greater than 0.05 (Table 2).
Further genetic analyses were carried out to test for the association of different genetic models with BC. Table 3 summarizes three different genetic models and the chisquared value for each. The ABCG2 gene was excluded from the analysis because it expressed only two genotypes. For the ABCC1 SNP rs4148351, Het (CT) versus Common Hz (CC) was found to be associated with BC in Jordanian Arabs ( 2 = 5.33; p value <0.05). Similarly, for the ABCB1 SNP rs1128503, the Rare Hz (AA) versus Common Hz (GG) model was related to BC in Jordanian Arabs ( 2 =4.52; p value <0.05). No such association was found for any of the ABCC2 SNPs (Table 3).

Association between ABC Transporter SNPs and Major Prognostic Factors of Breast Cancer (BC).
Certain clinical and pathological characteristics of BC serve as major prognostic factors for the disease that are exploited in the process of treatment selection. None of the ABCC1 SNPs showed any significant association with the clinical characteristics of BC, but the ABCC2 SNPs rs2273697 and rs717620 were found to be significantly associated with age at breast cancer diagnosis (p value = 0.042) and breastfeeding status (p value = 0.05), respectively (Table 4). Meanwhile, the ABCC1 SNP rs35628 was associated with the pathological characteristic of tumor size (p value = 0.014), while the ABCC2 SNP rs2273697 was significantly associated with estrogen receptor status (p value = 0.013) ( Table 4).
Likewise, rs1045642 was the only ABCB1 SNP to be significantly associated with a clinical characteristic of BC, namely, age at breast cancer diagnosis (p value = 0.029) ( Table 5). In contrast, rs2032582 was the only ABCB1 SNP to show significant association with a pathological characteristic of BC, namely, tumor size (p value = 0.03) ( Table 5). The ABCG2 SNP rs2231142 was found to be significantly associated with axillary lymph node status (p value = 0.001) but not with any clinical characteristic (Table 5).  Tables  4 and 5. Only the ABCC1 SNP rs35626 was found to be significantly correlated with the different IHC profiles (p value = 0.013).

Discussion
In the present study, four ABC transporter genes were screened in female BC patients and healthy volunteers from Jordan. Three SNPs from each of the ABCC1, ABCC2, and ABCB1 genes and one SNP from the ABCG2 gene were 4 Journal of Oncology  investigated for their association with BC in patients of Jordanian-Arab descent.
The ABCC1 (MRP1) gene has been previously reported as being a predictor of hematological toxicity in BC patients undergoing certain chemotherapy regimens [13]. It has also been found to be involved in MDR development in cases of neuroblastoma [14]. Moreover, ABCC1 expression was found to be increased in children with acute lymphoblastic leukemia, and ABCC1 gene induction resulted in worsened disease-free and overall survival rates [15,16]. Our results show that none of the three investigated ABCC1 SNPs showed any significant association with the clinical and pathological characteristics of BC. However, we found that the ABCC1 SNP rs35626 was significantly associated with different immunohistochemistry (IHC) profiles in Jordanian-Arab patients.
Similar to ABCC1, the ABCC2 gene is involved in decreased recurrence-free survival in BC patients receiving tamoxifen [17]. Nuclear expression of ABCC2 in BC cells was also found to be associated with worsened clinical outcome [18]. Our findings showed that the ABCC2 SNP rs2273697 was significantly associated with age at breast cancer diagnosis. Furthermore, rs2273697 was in correlation with estrogen receptor status for genotype association, patients were categorized according to the expression of estrogen receptor (positive versus negative) and tested with regard to their genotypes. However, in this study only gender was matched for the analysis. In addition, rs717620 was associated with breastfeeding status.
Three ABCB1 SNPs rs1045642, rs1128503, and rs2032582 have been suggested to play a role in altered doxorubicin pharmacokinetics in Asian BC patients [19]. In the present study, all three aforementioned ABCB1 SNPs were significantly associated with BC in Jordanian Arabs. Moreover, the ABCB1 SNPs rs1045642 and rs2032582 were significantly associated with age at breast cancer diagnosis and tumor size, respectively.
Overexpression of the ABCG2 gene was implicated in developing flavopiridol resistance in BC cells [20]. The homozygous genotype (CC) of the ABCG2 SNP rs2231142 of the ABCG2 gene resulted in significantly reduced intestinal transport activity compared to the wildtype (AA) [21]. In Kurdish BC patients, the A allele of the rs2231142 SNP may be a risk factor for BC progression, while the C allele was associated with poorer responses to anthracyclines and paclitaxel [22]. In contrast, the homozygous (CC) genotype of the ABCG2 SNP rs2231142 was significantly associated with longer progression-free survival in Han Chinese BC patients [23]. In the present study, the ABCG2 SNP rs2231142 was found to be significantly associated with axillary lymph node status in Jordanian BC patients.
Conclusively, screening certain ABC transporter genes in BC patients and healthy volunteers from the Jordanian-Arab