Association between H19 SNP rs217727 and lung cancer risk in a Chinese population: a case control study

Lung cancer (LC) has a high incidence and will continue to be the most common cause of cancer-related death around the world [1]. In China, this malignancy has the highest mortality and accounts for an estimated 25% of cancer-related deaths [2]. LC is a complex pathological process. The major risk factors by far are cigarette smoking and air pollution. Because a proportion of individuals exposed to carcinogens may have genetic factors associated with the development of cancer, predisposing genic elements should be weighed as risk factors for LC.

Long non-coding RNAs (lncRNAs) are longer than 200 nucleotides and are defined as non-protein-coding transcripts that are universally transcribed in the genome [3]. LncRNAs are transcribed as sense, antisense, bidirectional, intronic, or intergenic [4]. They can work by binding to chromatin-modifying complexes to specifically silence genomic loci both in cis and trans [5]. Increasingly, more studies have revealed that lncRNAs play a major role in many aspects of tumorigenesis at the epigenetic, transcriptional, and posttranscriptional levels, including cell growth, apoptosis, invasion, and metastasis. Based on the latest studies, there is evidence that lncRNAs can control gene expression through multiple mechanisms, such as transcription, translation, imprinting, genome rearrangement, and chromatin modification [6]. H19 is a maternally expressed imprinted gene on chromosome 11p15.5 that encodes for a capped and spliced RNA and has been implicated in cancer [7]. It was the first lncRNA discovered in the human genome and plays a crucial role in mammalian development [8, 9].

Single nucleotide polymorphisms (SNPs) have been widely used in plant, livestock, and animal genetic analyses. SNPs may affect gene expression and function. In addition, SNPs can be associated with the susceptibility to cancer. To date, there have been rare reports of genetic mutations in lncRNAs and their possible correlations to LC susceptibility. Thus far, the association between H19_rs217727 polymorphisms and LC has not been studied in the Chinese population.

In this hospital-based case-control study, we hypothesized a possible association between variant genotypes of the human H19 gene (rs217727) and LC. To test our hypothesis, SNPs within the H19 gene were genotyped from blood DNA samples of 555 LC patients and 618 age- and gender-matched general population controls.

LncRNAs, which characterize a functionally varied class of transcripts, have been found in many different species, such as humans, animals, plants, yeast, and viruses [14‐17]. Many researchers suggest that lncRNAs play a key role in tumorigenesis and during cellular development, differentiation, and many other biological processes. Furthermore, several studies have reported that lncRNAs are misregulated in various types of cancers [18‐20]. Significant overexpression of lncRNAs-CCAT2 was found in lung adenocarcinoma [21]. Nie et al. [22] reported that the lncRNA ANRIL was overexpressed in NSCLC patient tissues and associated with advanced “tumor node metastasis (TNM)” subsets, tumor size, and prognosis. Therefore, abnormalities of the expression of lncRNAs may be involved in the tumorigenesis of LC. Genetic variants in lncRNAs could be a biomarker for the prediction of cancer susceptibility in humans. Liu et al. [23] found that lncRNAs-MALAT1_rs619586 was associated with decreased hepatocellular carcinoma risk. LncRNAs-HOTAIR_rs12826786 in strong linkage disequilibrium with rs1899663 (r² = 1) was associated with the risk of gastric cardia adenocarcinoma [24]. However, their definitive roles in cancer development and progression remain largely unclear.

The H19 lncRNA gene does not encode a protein, but an oncofetal RNA [25, 26]. Deregulation of oncofetal RNA plays a critical role in tumorigenesis [26]. Accumulating evidence suggests that loss of imprinting and deregulation of the H19 gene are associated with human cancer, and its overexpression is a frequent event in lung cancer development [27, 28]. H19 is abnormally expressed in many types of cancers, including gastric [29], liver [30], colorectal [31], bladder [32], and pancreatic cancer [33], and increases the tumorigenic properties of tumor cells [34‐37]. In addition, studies have shown that H19 enhances invasion and migration of pancreatic ductal adenocarcinoma cells by decreasing let-7 and subsequently increasing the HMGA2-mediated epithelial-mesenchymal transition (EMT) [33]. Barsyte-Lovejoy et al. [34] found that the knockdown of H19 inhibited colony formation and anchorage-independent growth in lung cancer cells. Other studies have reported that H19 could be induced under hypoxic stress through the p53/HIF1-α pathway. Moreover, the knockdown of H19 could significantly suppress hypoxia-induced cancer cell proliferation in vivo [36]. Furthermore, high expression of H19 was positively associated with advanced TNM stage and was a predictor of overall survival (OS) in gastric cancer patients [38, 39]. Studies have shown that the H19_rs2107425 SNP was related to the susceptibility of bladder cancer, and showed a significant correlation with LC susceptibility (P = 0.02, age under 50 years) [40, 41]. However, Riaz et al. [42] found that H19_rs2107425 did not alter H19 mRNA expression in breast cancer. Yang et al. [43] reported that the variant H19 genotypes (CT + TT rs217727, CT + TT rs2839698) were correlated with an increased risk of gastric cancer (P = 0.040, P = 0.033), and the CT and TT genotypes in rs2839698 were also related to higher H19 mRNA levels in serum. In contrast, the rs217727 polymorphism did not affect the H19 mRNA level. To the best of our knowledge, the role of the H19_rs217727 polymorphism in LC susceptibility is still unknown in the Chinese population. Accordingly, we investigated whether this polymorphism was associated with the risk of LC in the Chinese population.

In this study, the A/A genotype of H19_rs217727 was significantly higher in the LC patients than in the controls (P = 0.006). In particular, there was a significantly increased risk of squamous cells carcinoma (P = 0.004) and adenocarcinoma (P = 0.045). However, when the combined A/G + A/A genotypes were compared with the wild-type G/G genotype, there was no significant difference. Therefore, the G allele may be a protective factor and people who carry this allele may be less likely to develop lung cancer. However, the present research was limited with respect to geographical variation, nation, and sample size. These factors may greatly affect the accuracy of this experiment. Additional studies that encompass more geographical regions, additional ethnic groups, and larger sample size should be performed. Although all subjects were enrolled from only two hospitals and selection bias could not be avoided, the genotype distribution of the controls in our study did accord with the HWE. Additional studies are also necessary to understand the mechanism by which the rs217727 SNP affects H19 mRNA expression, alters the translational efficiency, or leads to alterations in the H19 structure in LC.

In the current study, we found that the H19_rs217727 polymorphism plays a crucial role in the risk of LC in a Chinese population. Larger population-based studies are required to confirm the relationship between H19 expression levels and the susceptibility to LC. H19_rs217727 SNPs may be potential clinical markers for predicting the risk of LC.