New Insights into Functional Implication of Genetic Variation in Association with Cancer

C l i n M e d International Library Citation: Zhang Y (2016) New Insights into Functional Implication of Genetic Variation in Association with Cancer. Int J Cancer Clin Res 3:054 Received: February 01, 2016: Accepted: April 14, 2016: Published: April 16, 2016 Copyright: © 2016 Zhang Y. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Zhang. Int J Cancer Clin Res 2016, 3:054


Introduction
Recent advances in genome-wide association studies (GWASs) and fine-mapping analyses have yielded a plethora of common and rare loci associated with diverse cancers and other complex diseases [1][2][3].These different penetrant risk loci together substantially unveil the heritable fraction of diseases [4].Due to the majority of cancer risk variants reside in intronic or intergenic regions of unknown function, how to effectively understand their molecular mechanisms has become a challenging but critical question in post-GWAS research [5].Several pioneering studies have revealed some promising molecular evidences at both transcriptional and regulatory levels for cancer genetic variants [6,7], and the contributions are continuously underway, especially in the area of gene expression and regulation.
Expression QTL: Expression QTL analysis has been widely used to detect the regulatory feature of genetic variants that influence the expression level of genes in cis and trans [15].Recent works did more deep exploration in the study of expression QTLs for cancerassociated genetic variants.One of the most powerful approaches to perform eQTL analyses in cancer genetic studies is the Li et al. developed method [33], which directly works through the TCGA data sets.The method did a residual linear regression test with tumor gene expression adjusted by somatic copy-number alternation and CpG methylation.This method shows considerably robust in functional annotation of cancer-associated variants.For example, Cai et al. applied this model to examine eQTLs for novel discovered breast cancer risk variants [34].Based on this model, Wong et al. also identified thousands of expression-associated somatic single nucleotide variants (eSNVs) in endometrial cancer [35].
It should be also noted that the non-coding RNAs, including long non-coding RNAs (lncRNAs), piRNAs, endogenous siRNAs, snoRNAs, have been reported to show aberrant expression in tumors [36][37][38].However, there are still less findings about whether cancerassociated genetic variants could functionally control the expression of the non-coding RNAs.Studies in other diseases or human traits have implicated the relationship between the expression of lncRNAs and genetic variants.In the study of Kumar et al. [39], they found that the majority of lncRNAs-eQTLs were specific to lncRNA alone and did not affect the expression of neighboring protein-coding genes in blood.Brown et al. proposed that the most promising molecular phenotype for genetic risk variants is cancer-relevant tissue or cell type specific eQTLs [40].Due to high tissue-restricted expression pattern for most lncRNAs, exploring the lncRNA-eQTL association for cancer risk loci could be conducted to understand the specificity of cancer-associated genetic variants.In Addition, relative to genetically steady eQTLs, some eQTLs are probably inducible [41], another missing layer involved in disease predisposition.
Transcriptional regulation by genetic variants is involved in allelic imbalance.The allelic specificity on gene expression is a phenomenon where two alleles for heterozygous SNPs show significantly biased expression [42].Recent studies have showed the allelic biased expression for cancer risk loci.For example, two breast cancer risk loci (rs2046210 at 6q25 and rs418269 at 8q24) are significantly associated with allelic specific expression of ESR1 and MYC, respectively [33].Due to the transcriptomic complexity and fine-tune regulation at both transcriptional and post-transcriptional levels, genetic variants associated with gene expression could be comprehensively analyzed, so as to maximize the understanding of cancer risk loci as regulators.
Protein-level QTL: At the translational level, many studies and reviews reported the regulatory effects of cancer genetic variants by transcription factors (TFs) focused on the difference of binding affinity on genetic variants in trans regulation [15,[43][44][45].Several in vitro and in vivo functional assays are developed to determine the binding affinity difference at two alleles of a locus of interest, including Electrophoretic Mobility Shift Assay (EMAS), gene reporter assays, chromatin immune-precipitation followed by quantitative PCR (ChIP-qPCR), and the rest [43,44,46,47].Here one newly discovery of the functional relationship of genetic variants with protein abundance, termed as protein quantitative trait loci (pQTLs), was first reported in the study of Wu et al. [48].Interestingly, some pQTLs could not be detected as eQTLs, providing a new layer for genetic variants in molecular phenotypic regulation.Similarly, another study identified that the SNP rs6834 was significantly correlated with DIDO1 protein levels relevant for cancer chemotherapy [49].Thus, combining genomic and proteomic quantification data could be considered for cancer risk variants.

Epigenetic regulation
Although several studies have manifested that cancers and other diseases associated causal non-coding variants function as enhancer or other cis-regulatory roles [50][51][52], how to establish the functional interaction between cis-regulatory elements and gene regulation becomes a critical issue.The latest studies have made some efforts on the chromatin architecture scale.
Chromosome conformation capture (3C) and its derived methods are the high-throughput molecular biology techniques used to analyze the topology of chromosomal regions in viable cells [53].With 3C technology, it is possible to identify the physically local or distal interaction between regulatory elements and genetic variants.For example, 3C followed by real-time PCR (3C-qPCR) has been successfully utilized in determining whether the two pre-defined genomic regions are physically interacted [6,33,43,[54][55][56].More recently, the development of targeted 3C and relevant technologies [57][58][59][60] further allow for a high-resolution survey of the whole genome for potential interactions with multiple regions of interest simultaneously.With these targeted chromosome architecture technologies, several studies have found both intra-chromosomal and inter-chromosomal physical interactions in high precision for cancer genetic variants.For example, Jager et al. found a regulatory network about the looping interactions between CCAT2, CCAT1 and MYC, at 8q24 risk locus associated with colorectal cancer [61].Another study also discovered the strongest long-range interaction was not at intra-chromosomal locus 8q24 but at inter-chromosomal locus 3q13 associated with prostate cancer [57].Therefore, for a given cancer risk locus, it may function as a regulatory hub by physical interactions with multiple genes important for carcinogenesis, implicating the multiple-directional regulation probably exists at specific risk loci.
There are still some issues for current epigenomic assays to link the chromosome-level DNA looping with cancer risk loci, because most chromosome-capturing assays are carried out in well-established cell lines.Due to the potential difference of genetic background between cell lines and patient-derived tissues or cells, it will be attractive to develop patient-derived cell/tissue system to characterize the personalized chromatin interaction landscape for cancer genetic studies.In addition, combining with single-cell based genomic and epigenomic sequencing technologies [62][63][64], experiments on the patient-derived ex vivo cells could provide a deeper knowledge about the genetic and etiologic susceptibility of cancers.

Environmental factors
Besides the genetic and epigenetic implications for cancer risk loci, a few studies reported other potential non-(epi)genetic factors in association with cancer susceptibility, including gene-trait interaction [65,66], gene-nutrition interaction [67,68], genetic-microbiome interaction [69], as well as electronic medical record (EMR)based genetic integrative analysis [70].Taking one as an example, Ramagopalan et al. [71] reported that the vitamin D receptor (VDR) binding sites were significantly enriched near cancer and autoimmune disease associated genes.Biochemical study demonstrated that the chromatin remodeler JMJD3 was regulated by vitamin D in colon cancer cells [72].Clinical surveys also demonstrated that a higher vitamin D level was significantly associated with a lower colorectal cancer risk [73,74], suggesting that cancer risk variants could be also implicated in the interaction with vitamins and other nutrients.

Conclusions and Prospects
In this review, the latest progression in the molecular phenotypes for cancer risk variants at multiple levels is summarized.We expect that these discussions will provide understanding of