Analysis of genetically driven alternative splicing identifies FBXO38 as a novel COPD susceptibility gene

Abstract

While many disease-associated single nucleotide polymorphisms (SNPs) are associated with gene expression (expression quantitative trait loci, eQTLs), a large proportion of complex disease genome-wide association study (GWAS) variants are of unknown function. Some of these SNPs may contribute to disease by regulating gene splicing. Here, we investigate whether SNPs that are associated with alternative splicing (splice QTL or sQTL) can identify novel functions for existing GWAS variants or suggest new associated variants in chronic obstructive pulmonary disease (COPD).RNA sequencing was performed on whole blood from 376 subjects from the COPDGene Study. Using linear models, we identified 561,060 unique sQTL SNPs associated with 30,333 splice sites corresponding to 6,419 unique genes. Similarly, 708,928 unique eQTL SNPs involving 15,913 genes were detected at 10% FDR. While there is overlap between sQTLs and eQTLs, 60% of sQTLs are not eQTLs. Co-localization analysis revealed that 7 out of 21 loci associated with COPD (p<1×10−6) in a published GWAS have at least one shared causal variant between the GWAS and sQTL studies. Among the genes identified to have splice sites associated with top GWAS SNPs was FBXO38, in which a novel exon was discovered to be protective against COPD. Importantly, the sQTL in this locus was validated by qPCR in both blood and lung tissue, demonstrating that splice variants relevant to lung tissue can be identified in blood. Other identified genes included CDK11A and SULT1A2.Overall, these data indicate that analysis of alternative splicing can provide novel insights into disease mechanisms. In particular, we demonstrated that SNPs in a known COPD GWAS locus on chromosome 5q32 influence alternative splicing in the gene FBXO38.Author SummaryWhile it is known that chronic obstructive pulmonary disease (COPD) is caused in part by genetic factors, few studies have identified specific causative genes. Genetic variants that alter the expression levels of genes have explained part of the genetic component of COPD, however, there are additional genetic variants with unknown function. In some genes the protein coding sequence can be altered by a mechanism known as RNA splicing. We hypothesized that some genetic variants that are associated with risk of COPD contribute to the disease by altering RNA splicing. In this study, we identified genetic variants that are associated both with COPD risk and RNA splicing. In particular, we found that a COPD associated variant of previously unknown function may contribute to the inclusion of a new exon in the FBXO38 gene. These finding are significant because they indicate that analysis of RNA splicing can help identify genes that contribute to disease.

Blood-Based Transcriptomic and Proteomic Biomarkers of Emphysema

“…However, most studies to date have not evaluated emphysema-associated alternative splicing mechanisms. Alternative splicing, the regulatory process in which multi-exon human genes are expressed in multiple transcript isoforms, has been implicated in the pathophysiology of several lung diseases such as asthma, pulmonary fibrosis, pulmonary arterial hypertension, and COPD (15)(16)(17)(18)(19)(20)(21). Protein levels have also been studied for potential emphysema biomarker identification and it was found that sRAGE, ICAM1, CCL20, and adiponectin levels in blood and eotaxin levels in bronchoalveolar lavage fluid are associated with emphysema severity (22)(23)(24)(25), though the protein panels used for these studies included fewer proteins than more recently developed panels (26,27).…”

Subcutaneous adipose tissue splice quantitative trait loci reveal differences in isoform usage associated with cardiometabolic traits

“…11 Previous studies have examined sQTLs in multiple tissues and populations. [13][14][15][16][17][18][19][20] While variants within canonical splice sites are more likely to be sQTL variants than eQTL variants, similar proportions of sQTLs and eQTLs have been observed at splice donor and acceptor sites. 13 In addition, sQTL variants have been identified in introns, exons, and intergenic regions, 19 suggesting that variants in diverse genomic contexts may influence splicing by affecting splice silencers and enhancers or through other mechanisms.…”

“…[13][14][15][16][17][18][19][20] While variants within canonical splice sites are more likely to be sQTL variants than eQTL variants, similar proportions of sQTLs and eQTLs have been observed at splice donor and acceptor sites. 13 In addition, sQTL variants have been identified in introns, exons, and intergenic regions, 19 suggesting that variants in diverse genomic contexts may influence splicing by affecting splice silencers and enhancers or through other mechanisms. 21 Prior studies have identified sQTLs colocalized with GWAS signals for traits including chronic obstructive pulmonary disease (COPD), Alzheimer disease, age-related macular degeneration (AMD), and Parkinson disease.…”

“…21 Prior studies have identified sQTLs colocalized with GWAS signals for traits including chronic obstructive pulmonary disease (COPD), Alzheimer disease, age-related macular degeneration (AMD), and Parkinson disease. [13][14][15][16][17]22 In several studies, sQTLs led to discovery of plausible mechanisms for known candidate genes. 14,15,20 Here, we present an analysis of subcutaneous adipose tissue splice junction expression levels from 426 participants in the Metabolic Syndrome in Men (METSIM) study.…”

Characterization of a COPD-Associated NPNT Functional Splicing Genetic Variant in Human Lung Tissue via Long-Read Sequencing

“…However, eQTL studies do not capture all of the potentially relevant functional mechanisms through which causal variants may alter gene expression. In particular, the alteration of gene splicing and isoform ratios is an important disease-causing gene regulatory mechanism that is not well captured by gene-level eQTL analyses 6,7 .…”