Screening for Regulatory Variants in 460 kb Encompassing the CFTR Locus in Cystic Fibrosis Patients

Abstract

It is estimated that up to 5% of cystic fibrosis transmembrane conductance regulator (CFTR) pathogenic alleles are unidentified. Some of these errors may lie in noncoding regions of the locus and affect gene expression. To identify regulatory element variants in the CFTR locus, SureSelect targeted enrichment of 460 kb encompassing the gene was optimized to deep sequence genomic DNA from 80 CF patients with an unequivocal clinical diagnosis but only one or no CFTR-coding region pathogenic variants. Bioinformatics tools were used to identify sequence variants and predict their impact, which were then assayed in transient reporter gene luciferase assays. The effect of five variants in the CFTR promoter and four in an intestinal enhancer of the gene were assayed in relevant cell lines. The initial analysis of sequence data revealed previously known CF-causing variants, validating the robustness of the SureSelect design, and showed that 85 of 160 CF alleles were undefined. Of a total 1737 variants revealed across the extended 460-kb CFTR locus, 51 map to known CFTR cis-regulatory elements, and many of these are predicted to alter transcription factor occupancy. Four promoter variants and all those in the intestinal enhancer significantly repress reporter gene activity. These data suggest that CFTR regulatory elements may harbor novel CF disease-causing variants that warrant further investigation, both for genetic screening protocols and functional assays.

Figure 1

SureSelect targeted enrichment experimental design. A total of 12,053 SureSelect biotinylated probes span the 463-kb CFTR locus and encompass 20 previously identified CFTR cis-regulatory elements. Gaps in probe coverage correspond to highly repetitive regions that were excluded. Repeats shown correspond to RepeatMasker track from University of California, Santa Cruz, Genome Browser (hg19).

Figure 2

Promoter variants reduce activity of 2-kb CFTR promoter in a cell type–independent manner. A and B: Schematic of five variants identified within the 2-kb CFTR promoter (A) and four substitutions identified within the CFTR intron 11 cis-element (B). Stars represent individual patients with observed variant (blue, homozygous variant; yellow, heterozygous variant), and orange hexagons represent 10 patients heterozygous for the observed variant. C: Luciferase expression vectors containing the 2-kb CFTR promoter [wild type (WT) or with variants] were transiently cotransfected into 16HBE14o- or Caco2 cells. Data are shown relative to the CFTR 2-kb promoter parental vector. pGL3B, lacking a promoter, is shown for control. D: Luciferase expression vectors containing the 787-bp minimal CFTR promoter and the DHS11 long (WT or with variants) cis-element were transiently cotransfected into Caco2 cells. Data are shown relative to the CFTR minimal promoter parental vector (pGL3B.245). C and D: Luciferase expression levels were compared against pGL3B.2 kb (C) and pGL3B.245-DHS11(long) (D) using unpaired t-tests. Data are expressed as means ± SEM (C and D). n = 12 (C); n = 9 (D). ∗∗P < 0.01, ∗∗∗P < 0.001, and ∗∗∗∗P < 0.0001 versus pGL3B.2 kb (C; unpaired t-test) and versus pGL3B.245-DHS11(long) (D; unpaired t-test). DHS, DNaseI hypersensitive site.