In the current study we found that the miR-146a rs2910164 and miR-155 rs767649 variants were significantly associated with the risk of NSCLC in an Iranian population. In particular, we demonstrate a significant increased risk of NSCLC in subjects with the CC genotype of miR-146a rs2910164 compared to subjects with the GG genotype. In contrast, the AT genotype of miR-155 rs767649 is protective against NSCLC. There was no link between miR-146a polymorphisms and smoking status although the AT genotype frequency of miR-155 was lower in smokers with NSCLC than in non-cancerous smoking controls. No polymorphisms were associated with NSCLC stage or type.
SNPs occur approximately once every three hundred base pairs in both coding and non-coding regions of the genome [12]. Approximately 93% of functional SNPs are located within non-coding regions which are the most common sites for miRNA genes [13]. SNPs located in mature miRNA regions may directly affect binding to target miRNAs and those SNPs occur in pre-miRNA, may interfere with miRNA maturation [13]. Importantly, miRNAs generally do not act individually but as part of a network in order to regulate the expression and/or function of numerous genes whilst variations in miRNA expression induces diverse functional outcomes [6].
miR-146a is known to negatively regulate severe inflammation [14] in an NF-kB-dependent manner and can bind to sequences within the 3′-UTRs of the TNF receptor-associated factor 6 (TRAF 6) and IL-1 receptor-associated kinase 1(IRAC1) genes [14]. NSCLC has been associated with enhanced inflammation [15] and changes in miR-146a expression may enhance lung cancer risk. In previous studies, Jia and colleagues showed that the CC genotype and C allele distribution in NSCLC patients was significantly higher (P = 0.03 and 0.03, respectively) and concluded that the rs2910164 polymorphism of miR-146a was associated with an increased risk of NSCLC in a Chinese population [8]. A meta-analysis study has also reported that the rs2910164 polymorphism is associated with lung cancer (NSCLC and SCLC) [16].
However, there is disparate data concerning miR-146a polymorphisms and the susceptibility to other cancers. Hashemi et al. showed no statistically significant association between miR-146a rs2910164 polymorphisms and prostate cancer in an Iranian population [17] despite a meta-analysis demonstrating that the rs2910164 CC genotype was associated with decreased prostate cancer risk in an Asian population [18]. In addition, whilst the miR-146a rs2910164 GC allele was significantly associated with breast cancer in a Pakistani population [9] this was not the case in Azeri subjects from Iran [19]. Furthermore, Gao and colleagues demonstrated a significant association of the rs2910164 CC genotype or C allele with susceptibility to colorectal cancer in Chinese males [20].
A recent meta-analysis showed that the CC genotype of miR-146a rs2910164 was associated with susceptibility to NSCLC but not to hepatocellular carcinoma and gastric cancer [12]. In stratified analysis by cancer type and ethnicity, is was reported that these polymorphisms were associated with a significant risk of lung cancer, breast cancer and colorectal cancer in Asian, but not Caucasian populations [13].
miR-155 expression is significantly up-regulated in lung cancer tissues, plasma and sputum and is associated with the risk of NSCLC [7]. possibly through a modulatory effect on NF-kB activity [7, 10]. The rs767649A>T polymorphism has been associated with an increased risk and poor prognosis of NSCLC. The T allele increased the transcriptional activity of miR-155 and patients with the rs767649-TT genotype had the highest risk ratio for NSCLC and had a reduced response to radiotherapy or chemotherapy [7].
In contrast, the rs767649 TT genotype is associated with a significantly reduced risk of cervical cancer and was linked with miR-155 overexpression in cancerous tissues compared with normal tissues suggesting an oncogenic role of miR-155 in cervical cancer. However, in vitro miR-155 promoter-luciferase assays indicated that the transition from an A to T allele might lead to miR-155 downregulation at the transcriptional level. This opposing effect on miR-155 expression indicates the importance of other factors in cervical cancer pathogenesis such as the presence or absence of human papillomavirus (HPV) infection. HPV is the main risk factor for cervical cancer and may impact on the direction and magnitude of the association between miR-155 rs767649 polymorphisms and the risk of cervical cancer [10].
miR-155 is overexpressed in hepatocellular carcinoma tissues compared with adjacent healthy tissues [11] . The rs767649 T allele is associated with higher miR-155 expression suggesting that variants of miR-155 contribute to the increased risk and poor prognosis of hepatocellular carcinoma [11]. A study has reported a population-dependent difference in the impact of the rs767649 TT genotype as a risk factor for multiple sclerosis. 93% of patients and 87% of controls had the T allele and none of the patients and controls had the AA genotype in one population but this was the opposite in subjects of Chinese ethnicity [21].
In conclusion, the current study indicates that the miR-146a rs2910164 and miR-155 rs767649 polymorphisms are associated with the risk of NSCLC with the AT genotype of miR-155 rs767649 being protective against NSCLC. Larger sample sizes with diverse ethnicities are needed to extend the findings.