Defective AMPK regulation of cholesterol metabolism accelerates atherosclerosis by promoting HSPC mobilization and myelopoiesis

Abstract

Objectives: Dysregulation of cholesterol metabolism in the liver and hematopoietic stem and progenitor cells (HSPCs) promotes atherosclerosis development. Previously, it has been shown that HMG-CoA-Reductase (HMGCR), the rate-limiting enzyme in the mevalonate pathway, can be phosphorylated and inactivated by the metabolic stress sensor AMP-activated protein kinase (AMPK). However, the physiological significance of AMPK regulation of HMGCR to atherogenesis has yet to be elucidated. The aim of this study was to determine the role of AMPK/HMGCR axis in the development of atherosclerosis.

Methods: We have generated a novel atherosclerotic-prone mouse model with defects in the AMPK regulation of HMGCR (Apoe^-/-/Hmgcr KI mice). Atherosclerotic lesion size, plaque composition, immune cell and lipid profiles were assessed in Apoe^-/- and Apoe^-/-/Hmgcr KI mice.

Results: In this study, we showed that both male and female atherosclerotic-prone mice with a disruption of HMGCR regulation by AMPK (Apoe^-/-/Hmgcr KI mice) display increased aortic lesion size concomitant with an increase in plaque-associated macrophages and lipid accumulation. Consistent with this, Apoe^-/-/Hmgcr KI mice exhibited an increase in total circulating cholesterol and atherogenic monocytes, Ly6-C^hi subset. Mechanistically, increased circulating atherogenic monocytes in Apoe^-/-/Hmgcr KI mice was associated with enhanced egress of bone marrow HSPCs and extramedullary myelopoiesis, driven by a combination of elevated circulating 27-hydroxycholesterol and intracellular cholesterol in HSPCs.

Conclusions: Our results uncovered a novel signalling pathway involving AMPK-HMGCR axis in the regulation of cholesterol homeostasis in HSPCs, and that inhibition of this regulatory mechanism accelerates the development and progression of atherosclerosis. These findings provide a molecular basis to support the use of AMPK activators that currently undergoing Phase II clinical trial such as O-3O4 and PXL 770 for reducing atherosclerotic cardiovascular disease risks.

Keywords: AMPK; Atherosclerosis; Cholesterol; HMG-CoA reductase; HSPCs.

Figure 1

Inhibition of AMPK-HMGCR signalling increases atherogenesis. (A-E) Apoe^−/− and Apoe^−/−/Hmgcr KI mice were aged to 20 weeks old on a standard chow diet and aortas were collected to assess for plaque size and composition. In the aortic sinus, (A, B) lesion size were assessed by hematoxylin and eosin (H&E), (A, C) lipid content was assessed by Oil Red O (ORO) staining, (A, D) macrophages were assessed by staining for CD68, (A, E) collagen content was assessed by staining for picrosirius red. Results shown are a representative image. Scale bars = 50 μM. Data are means ± SEM and analysed using a Student unpaired t test, n = 8–16 per group (∗P < 0.05, ∗∗∗P < 0.001, when comparing Apoe^−/− versus Apoe^−/−/Hmgcr KI).

Figure 2

Elevated circulating cholesterol and its metabolite 27-hydroxoysterol levels in Apoe^−/−/Hmgcr KI mice. (A-B) Cholesterol synthesis in Apoe^−/− and Apoe^−/−/Hmgcr KI hepatocytes in response to AMPK activators metformin and O–3O4. n = 3 independent experiments; each experiment contains at least three replicates. (C-D) Liver and serum samples were collected from Apoe^−/− and Apoe^−/−/Hmgcr KI mice. Cholesterol levels, including total cholesterol, free cholesterol, and cholesteryl ester, were measured. (E) Serum LDL/VLDL cholesterol from Apoe^−/− and Apoe^−/−/Hmgcr KI mice. (F) Heat map of Z-scored lipid concentrations of species from the cholesteryl ester (CE), free cholesterol (COH) and dehydrocholesteryl ester (DE) classes from Apoe^−/− and Apoe^−/−/Hmgcr KI mice. (G) Overlayed extraction ion chromatogram of the 27-hydroxylated cholesterol measurement for two representative samples. Black trace, Apoe^−/−, Red trace, Apoe^−/−/Hmgcr KI. (H) Serum levels of the cholesterol metabolite 27-hydroxycholesterol from Apoe^−/− and Apoe^−/−/Hmgcr KI mice. Results are means ± SEM and analysed using a two-way ANOVA (in A-D) or a Student t test (in E-H), n = 7–10 per group (∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, when comparing Apoe^−/− versus Apoe^−/−/Hmgcr KI).

Figure 3

Increased atherosclerosis in Apoe^−/−/Hmgcr KI is associated with increased Ly6-C^hi monocytosis. (A-E) Apoe^−/− and Apoe^−/−/Hmgcr KI mice were aged to 20 weeks old on a standard chow diet and blood were collected at the end of the study. Circulating leukocytes were quantified by flow cytometry. (C) Ly6-C^hi monocytes, (D) Ly6-C^lo monocytes and (E) neutrophil populations were identified and overlayed on the same FACS plot for visualization purpose. Results are means ± SEM and analysed using a Student unpaired t test, n = 8 per group (∗P < 0.05 when comparing Apoe^−/− versus Apoe^−/−/Hmgcr KI).

Figure 4

Inhibition of AMPK-HMGCR signalling promotes HSPC mobilization and extramedullary myelopoiesis. (A-F) Apoe^−/− and Apoe^−/−/Hmgcr KI mice were aged to 20 weeks old on a standard chow diet. HSPCs in the (A) BM, (B) blood and (C) spleen were measured using flow cytometry. In the spleen, (D) CMPs, (E) GMPs and (F) Ly6-C^hi monocytes were measured via flow cytometry. Results are means ± SEM and analysed using a Student unpaired t test, n = 7–11 per group (∗P < 0.05, ∗∗P < 0.01 when comparing Apoe^−/− versus Apoe^−/−/Hmgcr KI).

Figure 5

Increased cholesterol accumulation in Apoe^−/−/HMGCR KI HSPCs contributes to HSPC mobilization and Ly6-C^hi monocytosis. (A-I) Apoe^−/− and Apoe^−/−/Hmgcr KI mice were aged to 20 weeks old on a standard chow diet. (A) BM HSPCs were sorted using FACS before ex vivo cholesterol synthesis was measured via ¹⁴C-acetate labelling and TLC fractionation. n = 4 independent experiments. BODIPY-Cholesterol levels were measured in (B) BM HSPCs and (C) blood HSPCs. BM HSPC (D) proliferation and (E) CBS were measured via flow cytometry. BODIPY-Cholesterol levels were measured in (F) splenic HSPCs, and (G) splenic HSPC proliferation and (H) CBS were measured via flow cytometry. BODIPY-Cholesterol levels were measured in (I) blood Ly6-C^hi monocytes. (J) Experimental overview. (K-L) Representative flow plots, and ratio of CD45.2 to CD45.1 cells in (K) blood Ly6-C^hi monocytes and (L) BM HSPCs. (M) Graphic abstract. Results are means ± SEM and analysed using a Student unpaired t test, n = 4–9 per group (∗P < 0.05, ∗∗P < 0.01 when comparing Apoe^−/− versus Apoe^−/−/Hmgcr KI or Wt/Wt versus Hmgcr KI/Wt).