Genetic testing of Korean familial hypercholesterolemia using whole-exome sequencing

Abstract

Familial hypercholesterolemia (FH) is a genetic disorder with an increased risk of early-onset coronary artery disease. Although some clinically diagnosed FH cases are caused by mutations in LDLR, APOB, or PCSK9, mutation detection rates and profiles can vary across ethnic groups. In this study, we aimed to provide insight into the spectrum of FH-causing mutations in Koreans. Among 136 patients referred for FH, 69 who met Simon Broome criteria with definite family history were enrolled. By whole-exome sequencing (WES) analysis, we confirmed that the 3 known FH-related genes accounted for genetic causes in 23 patients (33.3%). A substantial portion of the mutations (19 of 23 patients, 82.6%) resulted from 17 mutations and 2 copy number deletions in LDLR gene. Two mutations each in the APOB and PCSK9 genes were verified. Of these anomalies, two frameshift deletions in LDLR and one mutation in PCSK9 were identified as novel causative mutations. In particular, one novel mutation and copy number deletion were validated by co-segregation in their relatives. This study confirmed the utility of genetic diagnosis of FH through WES.

Fig 1. Exome sequencing analysis of familial hypercholesterolemia (FH).

The steps for identifying FH-causing variants in three genes are shown, in addition to the subsequent genetic analyses of whole-exome sequencing data that led to the identification of pathogenicity.

Fig 2. Pedigree analysis of a patient with LDLR p.D834Rfs/- mutation.

(A) A simplified pedigree of the P05 family. The upper right arrow indicates the proband; squares indicate males, and circles indicate females. Open and filled symbols indicate unaffected and affected individuals, respectively. Asterisks indicate family members who underwent clinical examinations and molecular analyses. WT refers wild-type. (B) Clinical examination data and sequencing chromatograms. Vertical arrows indicate the mutation site. (C) Integrative Genomics Viewer screenshot of p.D834Rfs/-. Sequencing reads show that a single nucleotide substitution (G>C) and frameshift deletion (AT/-) occurred at the cis position.

Fig 3. Copy number variation (CNV) detection in LDLR.

SVD-ZRPKM values were used to detect CNVs by the CoNIFER algorithm and were calculated by transforming reads per kilobase per million values into standardized z-scores, based on the mean and standard deviation across all analyzed exomes. (A) The SVD-ZRPKM regional plot of the P25 patient with a large copy number deletion in LDLR. (B) The SVD-ZRPKM regional plot of the P17 patient and family member (P17-F01) with an inherited copy number deletion in LDLR. Green and blue indicate SVD-ZRPKM values of P17 and P17-F01, respectively. Values are plotted based on P17. (C) Pedigree of the P17 patient with CNV. The upper right arrow indicates the proband; squares indicate males, and circles indicate females. Open and filled symbols indicate unaffected and affected individuals, respectively. Asterisks indicate family members who underwent clinical examinations and CNV analyses. (D) TaqMan Copy Number Assay for P25, P17, and family members of P17. Red indicates the assay for P25 by probe #1 within intron 5; blue indicates the assay for P17 and other members by probe #2 (overlapped from intron 10 to exon 11). The assay was performed in duplicate and repeated. Results were plotted by CopyCaller software v.2.0.