Hyperglycemia impairs atherosclerosis regression in mice

Abstract

Diabetic patients are known to be more susceptible to atherosclerosis and its associated cardiovascular complications. However, the effects of hyperglycemia on atherosclerosis regression remain unclear. We hypothesized that hyperglycemia impairs atherosclerosis regression by modulating the biological function of lesional macrophages. HypoE (Apoe(h/h)Mx1-Cre) mice express low levels of apolipoprotein E (apoE) and develop atherosclerosis when fed a high-fat diet. Atherosclerosis regression occurs in these mice upon plasma lipid lowering induced by a change in diet and the restoration of apoE expression. We examined the morphological characteristics of regressed lesions and assessed the biological function of lesional macrophages isolated with laser-capture microdissection in euglycemic and hyperglycemic HypoE mice. Hyperglycemia induced by streptozotocin treatment impaired lesion size reduction (36% versus 14%) and lipid loss (38% versus 26%) after the reversal of hyperlipidemia. However, decreases in lesional macrophage content and remodeling in both groups of mice were similar. Gene expression analysis revealed that hyperglycemia impaired cholesterol transport by modulating ATP-binding cassette A1, ATP-binding cassette G1, scavenger receptor class B family member (CD36), scavenger receptor class B1, and wound healing pathways in lesional macrophages during atherosclerosis regression. Hyperglycemia impairs both reduction in size and loss of lipids from atherosclerotic lesions upon plasma lipid lowering without significantly affecting the remodeling of the vascular wall.

Figure 1

Experimental design and metabolic parameters. A: Hypomorphic apoE mice carrying the Mx1-cre transgene (Apoe^h/h MX1-Cre) were fed a high-fat diet (HFD) for 18 weeks. All mice received either a series of five consecutive i.p. injections of saline (baseline and saline groups) or streptozotocin (STZ group) during the 17^th week of HFD. A group of mice was sacrificed after 18 weeks of HFD and used as the baseline group. The two other groups of mice were given an i.p. injection of pIpC and switched to a low-fat diet (LFD) after 18 weeks of HFD. These two groups of mice were sacrificed 4 weeks after the pIpC injection and consumption of a chow diet. Blood glucose levels over time (B) and percentage of HbA_1c at sacrifice (C) for both the saline and STZ groups. Plasma cholesterol (D) and triglyceride (E) levels before (average of weeks 17 and 18) and 4 weeks after the induction of diabetes. Two-way analysis of variance, followed by Bonferroni posttests: ^∗∗∗P < 0.001, ^∗∗∗∗P < 0.0001.

Figure 2

Plasma lipoprotein profile. A: Pooled plasma samples from four animals per group were fractionated with fast protein liquid chromatography (FPLC), and cholesterol was measured in each individual fraction. B: FPLC fractions were resolved by gel electrophoresis and probed for apoB100, apoB48, apoA1, and apoE. C: Total plasma was separated by gel electrophoresis and probed for apoB100, apoB48, apoA1, and apoE and the corresponding quantification. One-way analysis of variance, followed by Bonferroni posttests: ^∗P < 0.05, ^∗∗P < 0.01, and ^∗∗∗P < 0.001.

Figure 3

Hyperglycemia impairs atherosclerosis regression. A: Lesion areas were measured from aortic root cross sections of baseline and both saline and streptozotocin (STZ) regression groups. B and C: Quantification of the ORO lesion content (B) and representative images of aortic root cross sections stained with oil red O (ORO) (C). Adjacent histological cross sections of aortic roots stained with apoA1 (green), apoB (red), and Hoechst (a nuclear dye, blue; D) or apoE (red) and Hoechst (E) and the corresponding quantifications (immunofluorescence intensity (IFI); F). Quantification (G) and representative (H) images of aortic root cross sections labeled with mac-2 antibody (a macrophage marker, green) and Hoechst (blue). One-way analysis of variance, followed by Bonferroni posttests: ^∗P < 0.05, ^∗∗P < 0.01, and ^∗∗∗P < 0.001 between baseline and regression groups; ^†P < 0.05 between saline and STZ groups.

Figure 4

Flow cytometry analysis of circulating leukocytes in saline control and streptozotocin (STZ)-treated mice at baseline and after 4 weeks of regression. Leukocyte (CD45⁺ cell) counts (A) and percentage of leukocytes identified as monocytes (CD115⁺ cells), granulocytes (CD115⁻, Ly6G⁺ cells), B cells (B220R⁺ cells), and T cells (CD3⁺ cells) (B). Monocyte subtype Ly6C^high (C) and Ly6C^low (D) counts and the expression level of CD62L (L-selectin) on Ly6C^high monocytes (E). Mean fluorescence intensity (MFI): ^∗P < 0.05, ^∗∗P < 0.01 by one-way analysis of variance (A) and two-way analysis of variance (B), followed by Bonferroni posttests.

Figure 5

Hyperglycemia does not impair lesion remodeling during atherosclerosis regression. Quantification of total collagen (A) and representative images of cross sections of aortic roots stained with Sirius red (bright-field and polarized images) (B). C: Quantification of collagen type I (bright orange-yellow polarized fibers) and collagen type III (green polarized fibers) is shown. Quantification (D) of smooth muscle cell (SMC) content of the lesion and representatives images of aortic root cross sections labeled with anti–SMC-α-actin antibody (red) and Hoechst (blue; E, top). Quantification (F) of fibronectin content of the lesion and representatives images of aortic root cross sections labeled with anti-fibronectin antibody (green) and Hoechst (blue; E, bottom). One- and two-way analysis of variance, followed by Bonferroni posttests: ^∗P < 0.05, ^∗∗P < 0.01 between baseline and regression groups.