Whole Exome/Genome Sequencing Joint Analysis of a Family with Oligogenic Familial Hypercholesterolemia

Abstract

Autosomal Dominant Hypercholesterolemia (ADH) is a genetic disorder caused by pathogenic variants in LDLR, APOB, PCSK9 and APOE genes. We sought to identify new candidate genes responsible for the ADH phenotype in patients without pathogenic variants in the known ADH-causing genes by focusing on a French family with affected and non-affected members who presented a high ADH polygenic risk score (wPRS). Linkage analysis, whole exome and whole genome sequencing resulted in the identification of variants p.(Pro398Ala) in CYP7A1, p.(Val1382Phe) in LRP6 and p.(Ser202His) in LDLRAP1. A total of 6 other variants were identified in 6 of 160 unrelated ADH probands: p.(Ala13Val) and p.(Aps347Asn) in CYP7A1; p.(Tyr972Cys), p.(Thr1479Ile) and p.(Ser1612Phe) in LRP6; and p.(Ser202LeufsTer19) in LDLRAP1. All six probands presented a moderate wPRS. Serum analyses of carriers of the p.(Pro398Ala) variant in CYP7A1 showed no differences in the synthesis of bile acids compared to the serums of non-carriers. Functional studies of the four LRP6 mutants in HEK293T cells resulted in contradictory results excluding a major effect of each variant alone. Within the family, none of the heterozygous for only the LDLRAP1 p.(Ser202His) variant presented ADH. Altogether, each variant individually does not result in elevated LDL-C; however, the oligogenic combination of two or three variants reveals the ADH phenotype.

Keywords: CYP7A1; LDL uptake; LDLRAP1; LRP6; autosomal dominant hypercholesterolemia; linkage analysis; next-generation sequencing; oligogenic hypercholesterolemia; polygenic risk score; protein structural models.

Figure 1

Pedigree of the family HC438 with the segregation of p.(Val1382Phe) variant in LRP6, p.(Pro398Ala) variant in CYP7A1 and p.(Ser202His) variant in LDLRAP1 and the weighted Polygenic Risk Score (wPRS). The proband II-7 is indicated by the black arrow. Squares and circles represent men and women, respectively. Affected family members are indicated by values highlighted in gray. More severely affected family members are indicated by bold values. * Age at lipid measurement in years. ** Myocardial infarction at 75 years old. *** Severe atheroma. # Under ciprofibrate. ## Under 5 mg rosuvastatin treatment. Patients for whom ♦ whole exome, □ whole genome sequencing was performed. Lipid values in mmol/L: TC for total cholesterol; LDL-C for LDL cholesterol; HDL-C for HDL cholesterol; TGs for triglycerides.

Figure 2

Structure of the LRP6 receptor and position of the identified variants. The LRP6 receptor contains the following structural motifs: signal peptide (SP), 4 β-propeller domains, 4 EGF-like domains (involved in the pH-dependent release of ligands in endosome), 3 LDLR type A repeats (responsible for the binding of ligands), a transmembrane anchor (binds the receptor to the cell membrane), and a cytoplasmic domain with PPPSP motifs (2 motifs at position 1487 and 1604 that allow the receptor to function in the Wnt/β-catenin pathway). Red arrows indicate the position of the variants identified in this study. Figure built from data from UniProt (www.uniprot.org (accessed on 12 October 2020)) and Ensembl (www.ensembl.org/index.html (accessed on 12 October 2020)) databases.

Figure 3

Crystal structure of wild-type and mutant LRP6-E3E4 with β-propeller domains (green) and epidermal growth factor (EGF)-like domains (gray). (A,B) LRP6-E3E4 Tyr972 residue (red) has polar contacts (yellow dotted lines) with Asp971 and Glu993 (blue). (C,D) LRP6-E3E4 mutant Cys972 residue (red) has a polar contact (yellow dotted line) only with Asp971 (blue).

Figure 4

Effect of inhibited, overexpressed or mutated LRP6 in HEK293T and HuH7 cells. (A) LRP6 mRNA expression in HuH7 after the silencing of LRP6. Reactions were run in triplicate for each cDNA. POLR2A was used as the reference housekeeping gene. The relative quantification of gene expression was performed using the ∆∆C_T method and non-transfected cells were used for calibration. (B) LDL-Bodipy uptake in HuH7 after silencing of LRP6. Median fluorescence intensity of 50,000 events was acquired for each sample, but only the median fluorescence intensity of living cells is presented. Data represent three independent assays performed in triplicate. (C) Expression of WT or mutated LRP6 at the cell surface of transfected HEK293T. The median fluorescence intensity of 100,000 events was acquired for each sample, but only the median fluorescence intensity of living cells is presented. Data represent four independently performed assays. (D,E) LRP6 expression in HEK293T cells after transfection with LRP6-WT or mutated plasmid (p.(Thr1479Ile) and p.(Tyr972Cys) variants). Proteins were extracted from transfected cells, separated by electrophoresis and then transferred onto PVDF membrane. The membrane was incubated with primary antibody (anti-LRP6), followed by incubation with secondary antibody before detection using the iBright^TM FL1500 imaging system. Protein was quantified by ImageJ software. Equal loading was confirmed using the ß-actin antibody. Data represent three independent assays. (F) LDL uptake in HEK293T after transfection with an empty vector, LRP6-WT or mutated plasmid. The median fluorescence intensity of 100,000 events was acquired for each sample, but only the median fluorescence intensity of living cells is presented. The fluorescence of each sample was normalized using the empty vector (PcM) as a reference. Data represent three independent assays, each performed in triplicate. In all experiments, the difference between conditions was determined by Bonferroni’s Multiple Comparison Test in one-way ANOVA and * p < 0.05, ** p < 0.01, *** p < 0.001 were considered as statistically significant. Results are shown as mean ± SD. Error bars represent ± SD.

Figure 5

LDL receptor expression, LDL binding and uptake, and LRP6 gene expression in patients EBV-transformed B-lymphocytes. (A) LDL receptor, (B) LDL-Bodipy binding and (C) LDL-Bodipy uptake quantification in EBV-transformed B-lymphocytes from normocholesterolemic subjects (N), LDLR mutation carriers (FH), hypercholesterolemic patients without an identified mutation (FH/M-), and two LRP6-p.(Val1382Phe) carriers from the HC438 family: II-4 and III-6 (see Figure 1). The median fluorescence of living cells is presented. Data represent five independently performed assays. (D) LRP6 gene expression. Relative Quantification (RQ) of LRP6 in EBV-transformed B-lymphocytes. Reactions were run in triplicate for each cDNA. HPRT and POL2RA were used as reference genes. The relative quantification was performed using the ∆C_T method. (A–D). Bonferroni’s Multiple Comparison Test in one-way ANOVA: * p < 0.05, ** p < 0.01, *** p < 0.001.