ApoE Receptor-2 R952Q Variant in Macrophages Elevates Soluble LRP1 to Potentiate Hyperlipidemia and Accelerate Atherosclerosis in Mice

Abstract

Background: apoER2 (apolipoprotein E receptor-2) is a transmembrane receptor in the low-density lipoprotein receptor (LDLR) family with unique tissue expression. A single-nucleotide polymorphism that encodes the R952Q sequence variant has been associated with elevated plasma cholesterol levels and increased myocardial infarction risk in humans. The objective of this study was to delineate the mechanism underlying the association between the apoER2 variant with arginine-to-glutamine substitution at residue 952 (R952Q) and increased atherosclerosis risk.

Methods: An apoER2 R952Q mouse model was generated and intercrossed with LDLR knockout mice, followed by feeding a Western-type high-fat high-cholesterol diet for 16 weeks. Atherosclerosis was investigated by immunohistology. Plasma lipids and lipid distributions among the various lipoprotein classes were analyzed by colorimetric assay. Tissue-specific effects of the R952Q sequence variant on atherosclerosis were analyzed by bone marrow transplant studies. sLRP1 (soluble low-density lipoprotein receptor-related protein 1) was measured in plasma and conditioned media from bone marrow-derived macrophages by ELISA and GST-RAP (glutathione S-transferase-receptor-associated protein) pull-down, respectively. P38 MAPK (mitogen-activated protein kinase) phosphorylation in VLDL (very-low-density lipoprotein)-treated macrophages was determined by Western blot analysis.

Results: Consistent with observations in humans with this sequence variant, the apoER2 R952Q mutation exacerbated diet-induced hypercholesterolemia, via impediment of plasma triglyceride-rich lipoprotein clearance, to accelerate atherosclerosis in Western diet-fed LDLR knockout mice. Reciprocal bone marrow transplant experiments revealed that the apoER2 R952Q mutation in bone marrow-derived cells instead of non-bone marrow-derived cells was responsible for the increase in hypercholesterolemia and atherosclerosis. Additional data showed that the apoER2 R952Q mutation in macrophages promotes VLDL-induced LRP1 (low-density lipoprotein receptor-related protein 1) shedding in a p38 MAPK-dependent manner.

Conclusions: The apoER2 R952Q mouse model recapitulates characteristics observed in human disease. The underlying mechanism is that the apoER2 R952Q mutation in macrophages exacerbates VLDL-stimulated sLRP1 production in a p38 MAPK-dependent manner, resulting in its competition with cell surface LRP1 to impede triglyceride-rich lipoprotein clearance, thereby resulting in increased hypercholesterolemia and accelerated atherosclerosis.

Keywords: LDL receptor-related proteins; atherosclerosis; coronary artery disease; hypercholesterolemia.

Figure 1.. ApoER2 R952Q mutation accelerates atherosclerosis in Ldlr^−/− mice.

(A) Sequences for gRNA and donor oligo with changes indicated by lower case letters and HindIII restriction site (underlined); DNA electrophoresis of PCR products after HindIII restriction digest (left panel) and genome sequencing (right panel). (B) Representative images and quantification of Oil Red O-stained atherosclerotic lesion area from Lrp8^RRLdlr ^−/− (n = 12) and Lrp8^QQLdlr ^−/− (n = 9) in the whole aorta. (C) Representative photographs and quantification of lesion size from Lrp8^RRLdlr ^−/− (n = 25) and Lrp8^QQLdlr ^−/− (n = 18) in the aortic root. The scale bars represent 500 μm. Morphometric data are presented as mean ± SD. Student’s t-test was used to evaluate statistical significance.

Figure 2:. ApoER2 R952Q mutation promotes hypercholesterolemia.

(A) Body weights from Lrp8^RRLdlr^−/− (n = 18) and Lrp8^QQLdlr^−/− (n = 13) mice after 16 weeks of Western diet. (B) Fasting blood glucose levels of Lrp8^RRLdlr ^−/− (n = 24) and Lrp8^QQLdlr ^−/− (n = 21) mice. (C) Fasting plasma insulin levels of Lrp8^RRLdlr ^−/− (n = 6) and Lrp8^QQLdlr ^−/− (n = 8) mice. (D) Growth rate of smooth muscle cells isolated from Lrp8^RRLdlr ^−/− (n = 5) and Lrp8^QQLdlr ^−/− (n = 4) mice, represented as fold change. Fasting total plasma triglyceride levels (E), FPLC profile of triglyceride distribution in VLDL (F), total plasma cholesterol levels (G) and cholesterol distribution among various lipoprotein fractions (H) in Lrp8^RRLdlr ^−/− (n = 25) and Lrp8^QQLdlr ^−/− (n = 28) mice. The FPLC profiles were analyzed from 2 cohorts per genotype with pooled plasma from 9–12 mice per cohort, and the data represented as percent of total lipids in plasma of control Lrp8^RRLdlr^−/− mice. Statistical significance between different genotypes were evaluated using Student’s t-test (B,C) or Welch t-test with corrections for unequaled variance (A,E) and the data are presented as mean ± SD. Data that did not meet normality criteria were evaluated by Mann-Whitney test and the data are presented as median ± SD (G). Triglyceride (TG), cholesterol (CH), and phospholipid (PL) concentrations and their percent composition in VLDL (fractions 2–8) and IDL/LDL (fractions 9–30) were determined from pooled FPLC samples obtained from 2 separate cohorts of Lrp8^RRLdlr^−/− (I,K) and Lrp8^QQLdlr^−/− (J,L) mice and averaged.

Figure 3:. R952Q mutation in apoER2 impedes triglyceride-rich lipoprotein clearance.

(A) Triglyceride levels of plasma from Lrp8^RRLdlr ^−/− (n = 4) and Lrp8^QQLdlr ^−/− (n = 3) mice measured at hourly intervals after lipolysis inhibition. (B) Clearance rates of TRL emulsion particles containing [³H]cholesteryl hexadecyl ether in Lrp8^RRLdlr ^−/− (n = 7) and Lrp8^QQLdlr ^−/− (n = 7) mice. Data in B were analyzed by nonlinear regression and fitted to a 2-phase decay curve to determine the half-life of initial fast phase of TRL clearance (C) and the half-life of the second slow phase of TRL clearance (D). The half-life data are reported as mean ± SD using Student’s t-test (C) or median ± SD using Mann-Whitney test (D) to evaluate statistical significance.

Figure 4:. ApoER2 R952Q mutation in bone marrow cells exacerbates diet-induced hypercholesterolemia and atherosclerosis.

Total plasma cholesterol levels (A) and distribution among various lipoproteins (B), as well as plasma triglyceride levels (C) and distribution among various lipoproteins (D) in mice with Lrp8 RR (n = 16) or Lrp8 QQ (n = 16) bone marrow transplanted into Lrp8^RRLdlr^−/− recipient mice. Atherosclerosis in Lrp8^RRLdlr^−/− mice transplanted with Lrp8 RR or QQ bone marrows (n = 7 in each group) was analyzed by Oil Red O-staining of the whole aorta (E), or serial sections of the aortic roots to determine lesion size (F) and quantification of Oil Red O-stained lesion area within the plaque in the aortic root (G). Fasting levels of plasma cholesterol (H) and its lipoprotein distribution (I) as well as triglycerides (J) levels and lipoprotein distribution (K) of pooled plasma from Lrp8^RRLdlr^−/− (n = 7) and Lrp8^QQLdlr^−/− (n = 8) mice after transplant with bone marrow from Lrp8^RRLdlr^−/− mice. Atherosclerosis in Lrp8 RR (n = 7) or Lrp8 QQ (n = 6–9) mice transplanted with Lrp8^RRLdlr^−/− bone marrow was quantified by Oil Red O-staining of the whole aorta (L) and the aortic roots to determine lesion size (M). Data are reported as mean ± SD using Student’s t-test (A,E,F,G,H,J,L), or Welch’s t-test for unequal variants correction (M), or as median ± SD using Mann-Whitney test (C) to evaluate statistical significance between groups.

Figure 5:. ApoER2 R952Q mutation promotes sLRP1 and TNFα release.

(A) Levels of sLRP1 in plasma of Lrp8^RRLdlr^−/− recipient mice transplanted with bone marrow from Lrp8^RRLdlr^−/− (n = 13) and Lrp8^QQLdlr^−/− (n = 12) mice. (B) sLRP1 levels in plasma of Lrp8^RRLdlr^−/− (n = 7) and Lrp8^QQLdlr^−/− (n = 8) recipient mice transplanted with Lrp8^RRLdlr^−/− bone marrow. (C) Levels of TNFα in plasma of Lrp8^RRLdlr ^−/− (n = 8) and Lrp8^QQLdlr ^−/− (n = 7) mice. Data are presented as mean ± SD using Student’s t-test (A, B) or as median ± SD using Mann-Whitney test (C) to evaluate statistical significance.

Figure 6:. ApoER2 R952Q mutation in macrophages increases sLRP1 levels in a p38 MAPK-dependent manner.

(A), Representative Western blot and quantification of sLRP1 levels in conditioned media from BMDM from Lrp8^RRLdlr ^−/− (n = 4) and Lrp8^QQLdlr ^−/− (n = 4) mice after treatment with VLDL. A two-way ANOVA with Student-Newman-Keuls post-analysis was used to determine significance. (B) Representative Western blot and quantification of phosphorylated and total p38 MAPK and quantification of phosphorylated p38 MAPK in Lrp8^RRLdlr^−/− and Lrp8^QQLdlr^−/− BMDM (n = 5) treated with VLDL for the indicated times. Student’s t-test was used to determine genotype-specific differences. (C) Representative Western blot and quantification of sLRP1 levels in conditioned media from BMDM isolated from Lrp8^QQLdlr ^−/− (n = 4) mice after stimulation with VLDL with or without a 15 min pre-treatment with 25 μM p38 MAPK inhibitor SB202190. A one-way ANOVA with Student-Newman-Keuls post-analysis was used to determine significance.