Effect of Dysferlin Deficiency on Atherosclerosis and Plasma Lipoprotein Composition Under Normal and Hyperlipidemic Conditions

Abstract

Dysferlinopathies are a group of muscle disorders caused by mutations to dysferlin, a transmembrane protein involved in membrane patching events following physical damage to skeletal myofibers. We documented dysferlin expression in vascular tissues including non-muscle endothelial cells, suggesting that blood vessels may have an endogenous repair system that helps promote vascular homeostasis. To test this hypothesis, we generated dysferlin-null mice lacking apolipoprotein E (ApoE), a common model of atherosclerosis, dyslipidemia and endothelial injury when stressed with a high fat, and cholesterol-rich diet. Despite high dysferlin expression in mouse and human atheromatous plaques, loss of dysferlin did not affect atherosclerotic burden as measured in the aortic root, arch, thoracic, and abdominal aortic regions. Interestingly, we observed that dysferlin-null mice exhibit lower plasma high-density lipoprotein cholesterol (HDL-C) levels than their WT controls at all measured stages of the disease process. Western blotting revealed abundant dysferlin expression in protein extracts from mouse livers, the main regulator of plasma lipoprotein levels. Despite abnormal lipoprotein levels, Dysf/ApoE double knockout mice responded to cholesterol absorption blockade with lower total cholesterol and blunted atherosclerosis. Our study suggests that dysferlin does not protect against atherosclerosis or participate in cholesterol absorption blockade but regulates basal plasma lipoprotein composition. Dysferlinopathic patients may be dyslipidemic without greater atherosclerotic burden while remaining responsive to cholesterol absorption blockade.

Keywords: atherosclerosis; dysferlin; endothelial function; hyperlipidemia; lipids; plaque; vascular homeostasis.

FIGURE 1

High dysferlin expression in mouse and human atherosclerotic lesions. (A) Dysferlin Ab and control detection in representative ApoE-null mouse aortic sections (scale bar = 40 μm) and human coronary lesions (scale bar = 600 μm and scale bar = 50 μm) using goat anti-dysferlin antibodies or IgG with hematoxylin counterstain. (B) Tissue specific detection and knockout efficiency of Dysferlin in mouse lysates (skeletal muscle, liver, thoracic aorta, and small intestine) via western blot using the mouse anti-dysferlin; NCL-Hamlet 1 antibody in wild type (WT) and Dysf (KO) mice. Quantification of Dysf Ab is shown relative to ponceau staining, and expressed as a percentage of the WT:skeletal muscle signal. (C) Genotyping PCR using ear clip DNA biopsies of wild type (WT), heterozygous mutant (HET), and homozygous mutant Dysf (KO) mice, as well as a no template control (NTC).

FIGURE 2

Similar atherosclerotic plaque accumulation in aortic segments from ApoE and DKO mice after 5 and 9 months (mo) of HFD-feeding. (A) Representative images of Sudan IV stained plaque accumulation in distal, arch, thoracic, and abdominal aortic segments from WT, Dysf, ApoE, and DKO mice after 9 months of HFD-feeding, and quantification of plaque accumulation after both 5 and 9 months of HFD-feeding (B), 5 months: WT (N = 5), Dysf (N = 5), ApoE (N = 6), and DKO (N = 4), 9 months: WT (N = 6), Dysf (N = 6), ApoE (N = 5), and DKO (N = 7). Mean + SEM. WT and Dysf genotypes significantly different from ApoE and DKO at each time-point; P < 0.05 (*); P < 0.01 (**); P < 0.001 (***); P < 0.0001 (****); two-way ANOVA with Sidak’s post hoc tests. No significance changes to atherosclerosis was detected between 5 and 9 months-fed cohorts; two way ANOVA. Scale bar = 0.5 cm. Similar atherosclerotic plaque accumulation in the aortic sinus of ApoE and DKO mice after 5 and 9 months (mo) of HFD-feeding. (C) Representative images of Oil Red O stained plaque accumulation in the aortic root sinus of WT, Dysf, ApoE, and DKO mice after 9 months of HFD-feeding, and quantification of plaque accumulation after both 5 and 9 months of HFD-feeding (D), 5 months: WT (N = 7), Dysf (N = 4), ApoE (N = 6), and DKO (N = 4), 9 months: WT (N = 4), Dysf (N = 5), ApoE (N = 5), and DKO (N = 6). Mean + SEM. WT and Dysf genotypes significantly different from ApoE and DKO; P < 0.0001 (****). Two-way ANOVA with Sidak’s post hoc tests. No significance changes to atherosclerosis was detected between 5 and 9 months-fed cohorts; two way ANOVA. Scale bar = 0.5 cm.

FIGURE 3

Total plasma cholesterol, high density lipoprotein cholesterol and triglyceride levels in WT, Dysf, ApoE and DKO mice after 5 and 9 months (mo) of HFD-feeding and age-matched chow controls at 9 months. (A) Total cholesterol (CHOL) (B), high density lipoprotein (HDL-C), and (C), Triglycerides (TG). (D) Protein extracts from dysferlin-expressing BAEC (control cell lysate), and a its expression in a human hepatocyte derived cell line (HepG2). GAPDH is used as a loading control. 5 months HFD: WT (N = 10), Dysf (N = 5), ApoE (N = 7), DKO (N = 5), 9 months HFD: WT (N = 13), Dysf (N = 14), ApoE (N = 11), and DKO (N = 11). WT and Dysf genotypes significantly different from ApoE and DKO; P < 0.01 (**); P < 0.001 (***); P < 0.0001 (****); Significantly different from WT; P < 0.05 (^#); P < 0.01 (^##); P < 0.001 (^###); two way ANOVA with Sidak’s post hoc tests, 9 months Chow: WT (N = 8), Dysf (N = 11), ApoE (N = 4), AND DKO (N = 7). WT and Dysf genotypes significantly different from ApoE and DKO; P < 0.01 (**); P < 0.0001 (****); One way ANOVA Tukey’s post hoc tests. Mean + SEM.

FIGURE 4

Decreased atherosclerosis and plasma cholesterol levels in DKO mice treated with the cholesterol absorption blocker ezetimibe for 9 months. (A) Representative images and quantification of plaque accumulation from aortic arch, thoracic and abdominal segments, (B) total plasma cholesterol (CHOL), high density lipoprotein (HDL-C), and triglyceride (TG) levels from HFD-fed DKO mice, and (C) CHOL, HDL-C, and TG levels from HFD-fed Dysf-null mice treated with vehicle or ezetimibe for 9 months. DKO: Vehicle (N = 11), Ezetimibe (N = 7); Dysf: Vehicle (N = 14), Ezetimibe (N = 9); Significance at P < 0.01 (**); P < 0.001 (***); P < 0.0001 (****); Two way ANOVA Sidak’s post hoc tests. Mean + SEM. Scale bar = 0.5 cm.