The benign c.344G > A: p.(Arg115His) variant in the LDLR gene interpreted from a pedigree-based genetic analysis of familial hypercholesterolemia

Background We previously identified the c.344G > A: p.(Arg115His) variant in the low-density lipoprotein receptor (LDLR) gene, which was interpreted as “conflicting interpretations of pathogenicity” in ClinVar, based on a genetic analysis of patients with familial hypercholesterolemia (FH). However, whether this variant affects the pathophysiology of FH remains unclear. Therefore, our aim was to annotate the c.344G > A: p.(Arg115His) variant in the LDLR gene in FH. We present 2 families harboring the c.344G > A: p.(Arg115His) variant in the LDLR gene. Methods Genetic analyses were performed for the coding regions and the exon-intron boundary sequence of the LDLR and proprotein convertase subtilisin/kexin type 9 (PCSK9) genes in 2 FH families. Next, the family without pathogenic variants in the LDLR and PCSK9 genes was screened by whole-exome sequencing. Detailed clinical and biochemical data were gathered from family members. Results In one family, the index case had biallelic c.1567G > A: p.(Val523Met) and c.344G > A: p.(Arg115His) variants in the LDLR gene, while the sibling had only the c.1567G > A: p.(Val523Met) variant in the LDLR gene. There was no difference in the FH phenotype between the siblings. In another family, the index case and the sibling had no pathogenic variants in the LDLR, PCSK9, and apolipoprotein B (APOB) genes, but the sibling’s wife with nonFH had the c.344G > A: p.(Arg115His) variant in the LDLR gene. The sibling and his wife had 4 children, including an unaffected child and an affected child who had the c.344G > A: p.(Arg115His) variant in the LDLR gene. In addition, the allele frequency of the c.344G > A: p.(Arg115His) variant (0.0023–0.0043) in Japanese and East Asian populations is relatively high compared with that of the other LDLR pathogenic variants (0.0001–0.0008). Conclusions The c.344G > A: p.(Arg115His) variant in the LDLR gene is interpreted as benign in individuals with FH.


Background
Familial hypercholesterolemia (FH) is a disease that leads to a high risk of coronary artery disease (CAD) because FH patients are exposed to high serum LDLcholesterol (LDL-C) levels from birth. The prevalence of FH is 1 per 200-500 individuals in the general population [1]. FH is caused by mutations in the low-density lipoprotein receptor (LDLR) and in the related genes apolipoprotein B (APOB) and proprotein convertase subtilisin/kexin type 9 (PCSK9). More than 4970 variants in the LDLR gene, 580 variants in the APOB gene, and 350 variants in the PCSK9 gene are shown in ClinVar [2]. In Japan, pathogenic variants in the APOB gene have not been reported [3]. Whether some variants of the LDLR and PCSK9 genes affect the pathophysiology of FH remains unclear.
In our FH cohort, patients were found to harbor the c.344G > A: p.(Arg115His) variant in the LDLR gene, which was interpreted as "conflicting interpretations of pathogenicity" in ClinVar. However, whether this variant affects the pathophysiology of FH remains to be elucidated. Herein, we present 2 families harboring the c.344G > A: p.(Arg115His) variant in the LDLR gene, including an index case with biallelic LDLR variants that comprised the c.344G > A: p.(Arg115His) and a pathogenic variant (Family 1) and nonFH patients with the c.344G > A: p.(Arg115His) variant (Family 2).

Methods
The protocol of this study was approved by the Ethics Review Committee of the National Cerebral and Cardiovascular Center (M17-56 or M24-80). Each patient provided written informed consent to participate in the study.

DNA analysis
Genomic DNA was extracted from whole blood from the sibling of Family 1 and the index case of Family 2 using an automated DNA extraction machine (QIAsymphony; QIAGEN, Valencia, CA) and from the other members of Family 2 using the Genome Extraction Kit (GENOMIX; Biologica, Nagoya, JAPAN) by SRL Inc. (Tokyo, JAPAN). The coding regions and exon-intron boundary sequences of the LDLR and PCSK9 genes were examined by Sanger sequencing as described previously [4]. For the index case of Family 2, multiplex ligationdependent probe amplification (MLPA) was performed to detect large rearrangements of the LDLR gene using the P062B LDLR MLPA Kit (MRC Holland, Amsterdam, the Netherlands). Next, whole-exome sequencing was performed for Family 2. Exome libraries were prepared using the SureSelect Human All Exon V7 Kit (Agilent Technologies, Santa Clara, CA). Sequencing was performed by NovaSeq 6000 (Illumina, San Diego, CA) with 150 bp paired-end reads at RIKEN GENESIS CO., LTD.
Sequence reads were aligned to the human reference genome (hg19) using BWA-MEM. Single nucleotide variants and small indels were called with HaplotypeCaller of the Genome Analysis Tool Kit. The presence/absence of the c.344G > A:p.(Arg115His) variant in the LDLR gene was confirmed by Sanger sequencing.

Clinical and laboratory data
Serum total cholesterol (TC), triglycerides (TG), and high-density lipoprotein-cholesterol (HDL-C) levels were measured using enzymatic methods. LDL-C levels were calculated by the Friedewald formula or were measured using enzymatic methods. Achilles tendon thickness (ATT) was measured by X-ray. CAD was evaluated by the presence of myocardial infarction, angina pectoris, or coronary arteries with ≥ 75% stenosis by coronary angiography or electrocardiogram.

Case presentation Family 1
The index case was a 19-year-old woman who had biallelic LDLR c.1567G > A: p.(Val523Met) and c.344G > A: p.(Arg115His) variants ( Fig. 1, Table 1). She was referred to our lipid clinic with her brother at the age of 19 years for dyslipidemia, and her untreated LDL-C level was 208 mg/dL. She did not have ATT. Her medications included 2.5 mg of rosuvastatin, and her LDL-C level was 100 mg/dL on this medication regimen. Her sibling (II-1) was a 21-year-old male who had the LDLR c.1567G > A: p.(Val523Met) variant. His untreated LDL-C level was 204 mg/dL at 21 years old. His ATT values were 6.7 and 9.1 mm. His medication included 2.5 mg of rosuvastatin, and his LDL-C level was 99 mg/dL under this medication regimen. Their father was diagnosed with FH, but his lipid profile was unknown.

Family 2
The index case was a 52-year-old man who had no pathogenic variants in the LDLR and PCSK9 genes (Fig. 1). He was referred to the lipid clinic at the age of 52 years for dyslipidemia (II-1). His untreated LDL-C level was 237 mg/dL, and his ATT values were 7.1 mm and 9.2 mm. He was diagnosed with hypertension at the age of 45 years. His medication regimen included 5 mg of rosuvastatin and 10 mg of ezetimibe, and his LDL-C level was 111 mg/ dL under this regimen. The lipid profiles of his parents were unknown. His father died of liver cancer at 65 years old, and his mother died at 64 years old due to unknown causes. The sibling (II-2) of the index case was a 49-yearold man who was diagnosed with FH. His medication regimen included 2 mg of rosuvastatin, 10 mg of ezetimibe, and 2 g of omega-3 fatty acid ethyl, and his LDL-C level was 106 mg/dL under this regimen. The index case and   http://www.genome.med.kyoto-u.ac.jp/SnpDB) [5], the Tohoku Medical Megabank Organization database (n = 3381; tommo_3.5KJPN) [6], and ExAC (n = 66,000, [7]), respectively. The allele frequency of the c.344G > A: p.(Arg115His) variant in the LDLR gene was relatively high compared with that of the other LDLR pathogenic variants (0.0001-0.0008) detected in these databases.  [9]. In addition, the Arg (pI = 10.76) to His (pI = 7.59) substitution does not change the positive charge of this sequence. For the c.344G > A: p.(Arg115His) variant, only one study exists; functional assay results revealed that receptor activities were 64% of normal in COS7 cells transfected with LDLR cDNA containing the c.344G > A: p.(Arg115His) variant [10]. However, the genotype-phenotype correlation of the c.344G > A:p.(Arg115His) in the LDLR gene in pedigrees has not been demonstrated. Thus, based on 2 pedigree-based genetic analyses, the c.344G > A: p.(Arg115His) variant in the LDLR gene was classified as benign according to the guidelines issued by the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Thus, examining the genotype-phenotype correlation in a pedigree is important for providing a genetic diagnosis with high accuracy.

Discussion
In our cohort, we showed that unrelated patients harboring no pathogenic variants in the LDLR and PCSK9 genes comprised approximately 40% of FH patients [4]. FH may be partly explained by the accumulation of common SNPs [11]. A number of factors, including somatic genetic changes, environmental factors, and other genetic factors, might contribute to the pathogenesis and phenotypic variations observed in FH. We are currently searching for new candidate FH genes that may be responsible for FH in Family 2. Finally, in vitro functional analyses are needed to quantify the actual effect of the c.344G > A: p.(Arg115His) variant in the LDLR gene.

Conclusions
The c.344G > A: p.(Arg115His) variant in the LDLR gene was not shared by the sibling with FH in Family 1, and patients with nonFH also had the variant in Family 2. In conclusion, the c.344G > A: p.(Arg115His) variant in the LDLR gene is interpreted as benign based on pedigreebased genetic analysis.