Molecular insights into genome-wide association studies of chronic kidney disease-defining traits

Abstract

Genome-wide association studies (GWAS) have identified >100 loci of chronic kidney disease-defining traits (CKD-dt). Molecular mechanisms underlying these associations remain elusive. Using 280 kidney transcriptomes and 9958 gene expression profiles from 44 non-renal tissues we uncover gene expression partners (eGenes) for 88.9% of CKD-dt GWAS loci. Through epigenomic chromatin segmentation analysis and variant effect prediction we annotate functional consequences to 74% of these loci. Our colocalisation analysis and Mendelian randomisation in >130,000 subjects demonstrate causal effects of three eGenes (NAT8B, CASP9 and MUC1) on estimated glomerular filtration rate. We identify a common alternative splice variant in MUC1 (a gene responsible for rare Mendelian form of kidney disease) and observe increased renal expression of a specific MUC1 mRNA isoform as a plausible molecular mechanism of the GWAS association signal. These data highlight the variants and genes underpinning the associations uncovered in GWAS of CKD-dt.

Fig. 1

Cis-expression quantitative trait locus (cis-eQTL) analysis in human kidney. a Study flow of cis-eQTL meta-analysis. SNPs—single nucleotide polymorphisms, eSNPs—genetic variants with at least one renal expression partner eGene. b Association between rs2927608 genotype and renal expression of ERAP2—the most significant result from cis-eQTL meta-analysis. RBINT—rank-based inverse normal transformation. The boxplot centre line shows the median, the two hinges show the upper and lower quartiles and the two whiskers show 1.5 times the interquartile range above and below the upper and lower quartiles respectively. c Flowchart showing the overlap between 117 variants associated with CKD-dt in genome-wide association studies (CKD-dt GWAS SNPs), their statistical proxies (linkage disequilibrium, r² >0.8) and kidney eSNPs (variants with at least one renal expression partner eGene). d Venn diagram—overlap between CKD-dt GWAS SNPs plus proxies (orange) and kidney eSNPs (green). e Venn diagram—overlap between CKD-dt GWAS SNPs plus proxies (orange) and kidney best eSNPs (purple). f Circular representation of findings from cis-eQTL analysis for variants identified CKD-dt GWAS. eGenes are ordered radially by genomic and chromosomal location, coloured by gene biotype (green—protein coding, purple—long non-coding, yellow—pseudogene) and labelled by their HUGO symbol. eGenes are connected to their eSNPs by lines whose colours are determined by the direction of gene expression change by GWAS CKD-dt risk allele (red—increase, blue—decrease). dbSNP reference cluster IDs are shown for each eSNP

Fig. 2

Tissue expression profiles of CKD-dt GWAS SNPs. a General overview of CKD-dt GWAS SNPs and their expression profiles in different tissues. b Detailed overview of CKD-dt GWAS SNPs in relation to presence (blue) or absence (grey) of eGenes across different tissues; kidney data are from the TRANSLATE study and TCGA, GTEx—Genotype-Tissue Expression project, eSNP—transcriptionally active single-nucleotide polymorphism; kidney specificity was defined as (i) exclusive presence of eGene in the kidney or (ii) exclusive presence of eGene–CKD-dt GWAS eSNP association in kidney tissue (yellow dbSNP rsID label) or (iii) difference in identity of eGene partner between the kidney and non-kidney tissues (red dbSNP rsID label) or (iv) difference in direction of association of CKD-dt GWAS SNP with eGene between kidney and non-kidney tissues (green dbSNP rsID label)

Fig. 3

Functional analysis of MUC1. a Functional annotations of MUC1 locus on chromosome 1. Top track—high-confidence CpG islands. Middle track—variants associated with the renal expression of MUC1 (MUC1 eSNPs) represented as triangles and coloured by their functional consequences; blue—non-coding exonic, pink—mapping onto CpG islands, green—splice variant, grey—no annotations, the sentinel variant outlined in red, other tested eSNPs are shown as circles. The height of each eSNP is determined by the negative log₁₀ P value for association with MUC1 expression in the kidney, so that eSNPs more significantly associated with MUC1 are higher than those less significantly associated. Bottom track—genes and a genomic scale bar for chromosome 1 in Mb. The coloured background reflects annotations from summarised chromatin states in adult human kidney tissue, red denotes transcription start site chromatin, green—transcribed chromatin, white—silent. b Renal expression of alternatively spliced MUC1 mRNA isoform in the TRANSLATE study and TCGA. Data are standardised expression (means and standard errors) stratified on rs4072037 genotype. The boxplot centre line shows the median, the two hinges show the upper and lower quartiles and the two whiskers show 1.5 times the interquartile range above and below the upper and lower quartiles respectively. c Renal expression of total MUC1 in the TRANSLATE study and TCGA. Data are standardised expression (means and standard errors) stratified on rs4072037 genotype. The boxplot centre line shows the median, the two hinges show the upper and lower quartiles and the two whiskers show 1.5 times the interquartile range above and below the upper and lower quartiles respectively. d N-terminus of pairwise alignment of alternatively spliced (bottom) and reference (top) MUC1 protein isoforms. Red arrows point to the predicted cleavage sites. e N-terminus of pairwise alignment of alternatively spliced (bottom) and reference (top) MUC1 protein isoforms after signal peptide removal. f Primary structure of the nine residues missing in the N-terminus of the alternatively spliced MUC1 isoform