Genetic Variations in the Flanking Regions of miR-101-2 Are Associated with Increased Risk of Breast Cancer

Genetic variants in human microRNA (miRNA) genes may alter mature miRNA processing and/or target selection, and likely contribute to cancer susceptibility and disease progression. Previous studies have suggested that miR-101 may play important roles in the development of cancer by regulating key tumor-associated genes. However, the role of single nucleotide polymorphisms (SNPs) of miR-101 in breast cancer susceptibility remains unclear. In this study, we genotyped 11 SNPs of the miR-101 genes (including miR-101-1 and miR-101-2) in a case-control study of 1064 breast cancer cases and 1073 cancer-free controls. The results revealed that rs462480 and rs1053872 in the flank regions of pre-miR-101-2 were significantly associated with increased risk of breast cancer (rs462480 AC/CC vs AA: adjusted OR = 1.182, 95% CI: 1.030–1.357, P = 0.017; rs1053872 CG/GG vs CC: adjusted OR = 1.179, 95% CI: 1.040–1.337, P = 0.010). However, the remaining 9 SNPs were not significantly associated with risk of breast cancer. Additionally, combined analysis of the two high-risk SNPs revealed that subjects carrying the variant genotypes of rs462480 and rs1053872 had increased risk of breast cancer in a dose-response manner (P trend = 0.002). Compared with individuals with “0–1” risk allele, those carrying “2–4” risk alleles had 1.29-fold risk of breast cancer. In conclusion, these findings suggested that the SNPs rs462480 and rs1053872 residing in miR-101-2 gene may have a solid impact on genetic susceptibility to breast cancer, which may improve our understanding of the potential contribution of miRNA SNPs to cancer pathogenesis.


Introduction
Breast cancer is the leading cause of cancer-related deaths among women world-wide, with an estimated 1,383,500 new cases and 458,400 deaths in 2008. Although the incidence and mortality rates in developed countries have been decreasing during the past 25 years, both rates have been increasing in many developing countries [1][2]. In China, breast cancer is the most prevalent cancer and is ranked the sixth leading cause of death in Chinese women [3]. Except for ill-fitted environmental exposures, lifestyle and behavioral factors, many studies have also suggested that genetic factors are usually associated with the risk of breast cancer [4].
In recent years, the use of genome-wide association studies (GWAS) to screen disease-associated genetic variants, has led to the successful identification of numerous breast cancer susceptibility regions, supporting a polygenic model of breast cancer susceptibility [5][6][7][8]. However, such loci explain only a small percentage of the total risk, and important regions harboring genetic variants associated with breast cancer risk still remain to be identified.
MicroRNAs (miRNAs) are a class of small (,22 nucleotides), non-coding single-stranded RNAs, which regulate gene expression by targeting mRNA for deregulation or translational repression [9]. At present, the biogenesis of miRNAs has been clearly described in mammals. In general, miRNA genes are initially transcribed by RNA polymerase II to form large, primary miRNAs (pri-miRNAs). These pri-miRNAs are subsequently cleaved into pre-miRNAs by Drosha and processed into a miRNA duplex of 19-22 nt, by the endonuclease enzyme Dicer. In most cases, only one strand of the miRNA duplex from either the 59 or the 39 arm of the pre-miRNA is selected as the mature miRNA and incorporated into the RNA-induced silencing complex (RISC) that can target specific protein-coding messenger RNA (mRNA) [9][10]. Recently, increasing evidence suggests that miRNAs regulate the expression of almost one-third of the human genome. Deregulation of mature miRNA expression has been demonstrated in many human cancers including breast cancer, and specific miRNAs have been used as markers to define molecular subtypes of various cancers [11][12][13].
Although the exact mechanisms underlying miRNA deregulation in cancer are not clear, the presence of single nucleotide polymorphisms (SNPs) in miRNA genes, including pri-miRNAs, pre-miRNAs and mature miRNAs, has been shown to influence the processing and/or target selection of miRNAs, thus affecting the risk of cancer [14][15][16][17][18][19][20][21]. For example, Hu et al. first reported that the rs11614913 SNP in the miR-196a2 precursor was associated with survival of patients with non-small lung cancer (NSCLC) and risk of lung and breast cancer [15][16][17]. Further studies demonstrated that the rs11614913 SNP not only affected the mature processing of miR-196a2, but also influenced the interactions between miR-196a2 and its downstream targets [15]. The functional relevance of the rs11614913 SNP in the miR-196a2 precursor and breast cancer susceptibility was further confirmed by a research group at Yale University [18].
MiR-101 belongs to a family of miRNAs involved in various cellular activities, including cell proliferation, invasion and apoptosis [22]. Genomic loci for miR-101 have been identified on chromosome 1p31.3 (miR-101-1) and chromosome 9p24.1 (miR-101-2). MiR-101 is frequently expressed at low levels in multiple malignancies including breast cancer, hepatocellular carcinoma, glioblastoma, prostate and gastric cancers [23][24][25][26][27]. Over-expression of miR-101 has a tumor-suppressive effect in breast cancer, and miR-101 has been shown to negatively regulate oncogenes including EZH2 and STMN1 [22,27]. Furthermore, Sachdeva et al. reported that miR-101 may promote MCF-7 cell growth in an estrogen-independent manner by enhancing AKT activation, suggesting a link between miR-101 and estrogenindependent signaling in estrogen receptor (ER)-positive tumor cells [28]. However, to date, little is known about the role of miR-101-associated SNPs in breast cancer risk.
In this study, we hypothesized that polymorphisms of miR-101 are associated with the susceptibility of breast cancer in women. To test this notion, we investigated the association of 11 SNPs located in the miR-101 genes with breast cancer risk in a Chinese case-control study.

Ethics Statement
This case-control study was approved by the institutional review board of Nanjing Medical University. The design and performance of current study involving human subjects were clearly described in a research protocol. All participants were voluntary and would complete the informed consent in written before taking part in this research.

Study Population
A total of 1064 breast cancer cases and 1073 cancer-free controls were included in this study, which has been described previously [29]. Patients with breast cancer were consecutively recruited between January 2004-April 2010 from the First Affiliated Hospital of Nanjing Medical University, Gulou Hospital and the Cancer Hospital of Jiangsu Province (Nanjing, China). Cancer-free controls were randomly selected from a cohort of more than 30,000 participants in a community-based screening program for non-infectious diseases conducted in the Jiangsu Province during the same period as breast cancer patients were recruited. Control subjects had no self-reported cancer history and were frequency matched to the breast cancer patients by age (65 years) and residential areas (urban and rural). Information related to demographic data, menstrual and reproduction history and environment exposure history was obtained from each patient during a standardized interview, and 5 ml of venous blood was subsequently collected from each participant for genotyping assays. The estrogen receptor (ER) and progesterone receptor (PR) status of each patient was obtained from the hospital medical records.

Statistical analyses
Differences in the distributions of demographic characteristics, selected variables and genotypes frequencies between breast cancer cases and controls were analyzed by x 2 test and student t test. Associations between the genotypes and breast cancer risk were estimated by computing the odds ratios (ORs) and 95% confidence intervals (CIs) from logistic regression analyses. Adjustment factors for the associations included age, age at menarche and menopausal status. The Hardy-Weinberg equilibrium was tested by a goodness-of-fit x2 test to compare the observed genotype frequencies to expected frequencies among the control subjects. All statistical analyses were performed with Statistical Analysis System software (version 9.1.3; SAS Institute, Cary, NC, USA).

Results
Demographic information and variables for breast cancer patients (n = 1064) and controls (n = 1073) are presented in Table  S1. The age of patients and controls was comparable after frequency-matching (P.0.05). Compared with control subjects, patients with breast cancer experienced an earlier menarche, later first live birth and a lower proportion of natural menopausal status(P,0.05). Of the 1064 patients with breast cancer, 490 (46.05%) were ER positive, while 506 (47.56%) were PR positive.
The chromosomal position and disease association of 11 SNPs are described in Table 1. The SNP rs1011210 deviated from Hardy-Weinberg equilibrium among controls (P,0.05) and was excluded from subsequent analyses. In multivariate logistic regression models, the rs462480 and rs1053872 residing in the 61 bp and 10 kb downstream of pre-miR-101-2 were significantly associated with an increased risk of breast cancer, but not the remaining nine SNPs (Table 1).
Furthermore, we conducted a joint analysis of these two SNPs rs462480 and rs1053872 (Table 2). There was a significant trend miR-101-2 with the Breast Cancer Risk for the increased risk of breast cancer with the increasing number of variant genotypes (P trend = 0.002). In the combined dataset, compared with subjects with ''0-1'' risk allele of the two SNPs, subjects carrying ''2-4'' risk alleles resulted in 1.29-fold (95% CI, 1.08-1.54; P = 0.005) increased risk of breast cancer.
We also performed stratification analyses for the combined effect of rs462480 and rs1053872 based on age, age at menarche, first live birth, menopausal status, ER and PR status. As shown in Table 3, the increased breast cancer risk associated with ''2-4'' risk alleles of rs462480 and rs1053872 was significant among women with older age, earlier menarche, later first live birth, negative ER and PR and postmenopausal status, compared with subjects with the ''0-1'' risk allele. However, we did not observe any significant differences of these two SNPs in different strata (P.0.05 for heterogeneity tests).

Discussion
In this study, we evaluated the association of 11 tagging SNPs located in the miR-101 gene and predisposition to breast cancer in a case-control study. We found that rs462480 and rs1053872 in the flank region of pre-miR-101-2 were significantly associated with the increased risk of breast cancer in the Chinese population. To our knowledge, this is the first study to evaluate the association of miR-101-related polymorphisms and breast cancer susceptibility.
MiR-101, a miRNA commonly down-regulated in cancer, has been implicated in several key cancer-related processes including cell growth, migration, invasion and apoptosis. Recently, several studies supporting the considerable role of miR-101 in the development of breast cancer have been reported. Sachdeva et al. demonstrated that miR-101 stimulated estrogen-independent growth via upregulation of phosphorylated AKT [28]. Frankel et al. revealed that miR-101 could act as a key regulator of autophagy, which may sensitize breast cancer cells to 4-hydroxytamoxifen (4-OHT)-mediated cell death [30]. Wang et al. revealed that miR-101 was down-regulated in different subtypes of breast cancer, and subsequently showed that miR-101 could inhibit tumor growth and stimulate breast cancer cells to apoptosis by targeting STMN1 [31]. To date, only one published association study investigated the effect of miR-101 polymorphisms on risk of hepatitis B-related liver disease [32]. This study revealed that the rs7536540 polymorphism located in the primary region of miR-101-1 was significantly decreased the risk of liver cirrhosis and hepatocellular carcinoma (OR = 0.63, 95% CI 0.42-0.93 and OR = 0.63, 95% CI 0.46-0.85 under the dominant model). Furthermore, rs12375841 in miR-101-2 was significantly associated with clearance of hepatitis B viral infection (OR = 1.24, 95% CI 1.03-1.48 under the co-dominant model). In this study, we did not observe the significant association between the 5 tagging SNPs (rs555146, rs578481, rs705509, rs7536540, rs1011210) in the vicinity of miR-101-1 gene and the risk of breast cancer. We found that rs462480, a SNP in high LD with rs12375841 (r 2 = 1) in miR-101-2, was significantly associated with an increased risk of breast cancer (OR = 1.182, 95% CI 1.030-1.357 under the additive  model). Meanwhile, the SNP rs1053872 in the flanking region of pre-miR-101-2 was also associated with the susceptibility of breast cancer. In addition, we observed a clear and significant trend toward increased breast cancer risk as the number of variant genotypes of the two SNPs rs462480 and rs1053872. The pri-miRNA, which is hundreds to thousands of nucleotides in length, is cleaved into pre-miRNA by Drosha in the nucleus, and is subsequently cleaved by Dicer in the cytoplasm to generate the final miRNA duplex [9]. For many pri-miRNAs, RNA folding algorithms have predicted that the sequences flanking either side of the pre-miRNA hairpin, may anneal to form a long, imperfect stem. A modest stem extension adjacent to the pre-miRNA is essential for excision of the pre-miRNA intermediate from a pri-miRNA substrate [33]. Although the function of these extensions or how they regulate the Drosha enzyme remains unclear, the extra flanking sequences may be required initially to tether or recruit the Drosha-DGCR8 complex to RNA [34]. Previous studies have demonstrated that genetic variants in the extensions may affect the Drosha recognition and cleavage [35][36]. Therefore, we speculate that the rs462480 at 61 bp downstream of pre-miR-101-2 may influence the processing of mature miRNA by affecting cleavage of Drosha. Further studies are warranted to investigate the underlying biologic mechanisms for the association of SNPs of pre-miR-101 and susceptibility to breast cancer.
In conclusion, our results indicated that genetic variants in the vicinity of pre-miR-101-2 were associated with breast cancer risk in the Chinese population. The rs462480 and rs1053872 SNPs may be considered as candidate genetic markers for the susceptibility to breast cancer in Chinese women. Further studies incorporating diverse populations and functional assays are required to validate and extend these findings.