Spectrum and Prevalence of Rare APOE Variants and Their Association with Familial Dysbetalipoproteinemia

Abstract

Familial dysbetalipoproteinemia (FD) is a highly atherogenic, prevalent genetically based lipid disorder. About 10% of FD patients have rare APOE variants associated with autosomal dominant FD. However, there are insufficient data on the relationship between rare APOE variants and FD. Genetic data from 4720 subjects were used to identify rare APOE variants and investigate their pathogenicity for autosomal dominant FD. We observed 24 variants in 86 unrelated probands. Most variants were unique (66.7%). Five identified APOE variants (p.Glu63ArgfsTer15, p.Gly145AlafsTer97, p.Lys164SerfsTer87, p.Arg154Cys, and p.Glu230Lys) are causal for autosomal dominant FD. One of them (p.Lys164SerfsTer87) was described for the first time. When we compared clinical data, it was found that carriers of pathogenic or likely pathogenic APOE variants had significantly higher triglyceride levels (median 5.01 mmol/L) than carriers of benign or likely benign variants (median 1.70 mmol/L, p = 0.034) and variants of uncertain significance (median 1.38 mmol/L, p = 0.036). For the first time, we estimated the expected prevalence of causal variants for autosomal dominant FD in the population sample: 0.27% (one in 619). Investigating the spectrum of APOE variants may advance our understanding of the genetic basis of FD and underscore the importance of APOE gene sequencing in patients with lipid metabolism disorders.

Keywords: APOE; apolipoprotein E; autosomal dominant; dyslipidemia; familial dysbetalipoproteinemia; hyperlipoproteinemia type III; pathogenicity; phenotype/genotype correlation; remnant lipoproteins; triglycerides.

Figure 1

The study design. FD—familial dysbetalipoproteinemia; FH—familial hypercholesterolemia.

Figure 2

Spectrum of rare APOE variants identified in the study. The N-terminal signal peptide and three major APOE regions are shown: the N-terminal region containing the receptor binding site; the C-terminal region, which contains the lipid binding site; and a hinge region connecting the N- and C-terminal regions. The number of index patients (n = 86) is indicated in the circle. The number of carriers of two rare APOE variants is shown in white circles. Black brackets indicate the APOE variants in these compound heterozygotes. The color of the circle indicates the clinical interpretation: red, orange, yellow, light green, and green for pathogenic, likely pathogenic, variant of uncertain significance, likely benign, and benign variants, respectively. Novel variants are highlighted in red, previously associated with autosomal dominant FD; in blue, with other hyperlipidemias; in orange, registered in dbSNP but not published; in black.

Figure 3

Prevalence of rare APOE variants and the expected prevalence of pathogenic or likely pathogenic APOE variants associated with autosomal dominant FD in the ESSE-Ivanovo sample. The color of the circle indicates the pathogenicity of the variant.: red, orange, yellow, light green, and green for pathogenic, likely pathogenic, variant of uncertain significance, likely benign, and benign variants, respectively. Gray indicates carrier of p.Gly145Asp variant with ε2ε3 genotype (information on cis- or trans-position of this variant was not available for evaluation). VUS—variant of uncertain significance.

Figure 4

TG levels of carriers of pathogenic or likely pathogenic variants compared to carriers of benign or likely benign variants and carriers of variants of uncertain significance. Central lines represent the median, box limits represent upper and lower quartiles, vertical lines represent 1.5 times the quartile range, and individual data points outside this range are shown as outliers. The colored points represent carriers of APOE variants, highlighting their specific TG levels within the overall distribution (green for carriers of benign or likely benign variants, red for carriers of pathogenic or likely pathogenic variants, and yellow for carriers of variants of uncertain significance). t-test was used for pairwise comparison of TG levels. p-values were adjusted by the Holm–Bonferroni method. Only significant differences are indicated among groups. B—benign; LB—likely benign; LP—likely pathogenic; Me—median; P—pathogenic; VUS—variant of uncertain significance.