IMM-H007 promotes hepatic cholesterol and triglyceride metabolism by activating AMPK α to attenuate hypercholesterolemia

Abstract

Hypercholesterolemia is a significant risk factor for the development of atherosclerosis. 2',3',5'-Tri-O-acetyl-N ⁶-(3-hydroxyphenyl) adenosine (IMM-H007), a novel AMPK agonist, has shown protective effects in metabolic diseases. However, its impact on cholesterol and triglyceride metabolism in hypercholesterolemia remains unclear. In this study, we aimed to elucidate the effects and specific mechanisms by which IMM-H007 regulates cholesterol and triglyceride metabolism. To achieve this goal, we used Apoe ^-/- and Ldlr ^-/- mice to establish a hypercholesterolemia/atherosclerosis model. Additionally, hepatocyte-specific Ampka1/2 knockout mice were subjected to a 5-week high-cholesterol diet to establish hypercholesterolemia, while atherosclerosis was induced via AAV-PCSK9 injection combined with a 16-week high-cholesterol diet. Our results demonstrated that IMM-H007 improved cholesterol and triglyceride metabolism in mice with hypercholesterolemia. Mechanistically, IMM-H007 modulated the AMPKα1/2-LDLR signaling pathway, increasing cholesterol uptake in the liver. Furthermore, IMM-H007 activated the AMPKα1-FXR pathway, promoting the conversion of hepatic cholesterol to bile acids. Additionally, IMM-H007 prevented hepatic steatosis by activating the AMPKα1/2-ATGL pathway. In conclusion, our study suggests that IMM-H007 is a promising therapeutic agent for improving hypercholesterolemia and atherosclerosis through the activation of AMPKα.

Keywords: AMPK; Atherosclerosis; Cholesterol; FXR; Hypercholesterolemia; IMM-H007; LDLR; Triglyceride.

Graphical abstract

Figure 1

IMM-H007 improves HFHC diet-induced hypercholesterolemia and atherosclerosis in Apoe^−/− mice. (A) Male apolipoprotein E (Apoe)^−/− mice (∼8 weeks old) were randomly divided into two groups (15 mice/group). In the HFHC group, the mice were fed a high-fat and high-cholesterol diet (HFHC) containing 21% fat plus 0.5% cholesterol for 16 weeks; in the HFHC–H007 group, the mice were fed an HFHC for 10 weeks and then treated with IMM-H007 (200 mg/kg body weight/day) contained in an HFHC for 6 weeks. (B) Body weight from the 9th to the 16th week. (C) Determination of serum total cholesterol (TC). (D) Determination of serum non-high-density lipoprotein-cholesterol (non-HDL-C). (E) Determination of serum high-density lipoprotein-cholesterol (HDL-C) levels. (F) Determination of serum triglyceride (TG) levels. (G) En face Oil red O (ORO) staining. Scale bar = 5 mm. (H) Lesions in en face aortas were quantified via a computer-assisted image analysis protocol. The lesion areas are expressed as percentages of the en face aorta area. (I) Representative photomicrographs of aortic root sections stained with ORO. Scale bar = 500 μm. (J) Lesions in aortic root cross-sections were quantified via a computer-assisted image analysis protocol. The data are expressed as μm² in aortic root cross-sections. (K) Representative photomicrographs of aortic root sections stained with hematoxylin and eosin (H&E). Necrotic cores are marked by red dashed lines. Scale bar = 200 μm. (L) Representative photomicrographs of aortic root sections stained with Masson staining. Fibrous caps are marked by black dashed lines. Scale bar = 200 μm. (M, N) Quantitative analysis of the necrotic core area (M) and fibrous cap area (N). (O) Representative photomicrographs of aortic root sections stained with Verhoeff–van Gieson (VVG). Scale bar = 250 μm. (P) The collagen-positive areas are expressed as percentages of the aortic area. Statistical data are presented as mean ± SEM. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; NS, not significant versus the indicated group (n = 15).

Figure 2

IMM-H007 inhibits atherosclerosis in Ldlr^−/− mice. (A, B) Male low-density lipoprotein receptor (Ldlr)^−/− mice were fed high-fat, high-cholesterol (HFC) or HFC containing IMM-H007 (200 mg/kg body weight/day) for 20 weeks (A), and their body weights were measured in the 20th week (B). (C–F) Determination of serum TC (C), non-HDL-C (D), HDL-C (E), and TG (F) levels (n = 6). (G) En face lesions were determined by Oil Red O staining and quantified as a percentage (n = 8). Scale bar = 5 mm. (H, I) Aortic root sections were subjected to Oil Red O staining (H), followed by quantification of lesions in the aortic root (I) (n = 8). Scale bar = 500 μm. (J) Aortic root sections were subjected to H&E staining. Necrotic cores are marked by red dashed lines. Scale bar = 200 μm. (K) Representative photomicrographs of aortic root sections stained with Masson staining. The fibrous cap is marked by a black dashed line. Scale bar = 250 μm. (L, M) Quantification of lesions in the necrotic core area (L) and fibrous cap area (M) (n = 8). (N) The collagen-positive areas are expressed as percentages of the aortic area (n = 8). Statistical data are presented as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01; NS, not significant versus the indicated group.

Figure 3

The increase in LDLR protein levels induced by IMM-H007 depends on AMPKα1/2. (A) LDLR protein levels in Apoe^−/− mouse livers were determined via immunofluorescence staining, and the mean immunofluorescence intensity (MFI) of the images was quantified. Scale bar = 50 μm. n = 5. (B) HepG2 cells were treated with H007-M1 at the indicated concentrations for 24 h. LDLR protein expression was determined by Western blotting. (C) HepG2 cells were pretreated with H007-M1 (100 μmol/L) for 12 h and then incubated with CHX (100 μmol/L) for 0, 3, or 6 h. LDLR protein expression was determined by Western blot (top), and the half-life was calculated (bottom). (D, E) HepG2 cells were transfected with scrambled siRNA or AMP-activated protein kinase (Ampk) a1 siRNA and treated with H007-M1 (100 μmol/L) for 24 h. LDLR and AMPKα1 protein expression was subsequently determined via Western blotting (D), and the band density (E) was quantified. (F, G) HepG2 cells were transfected with scrambled siRNA or AMPKA2 siRNA and treated with H007-M1 (100 μmol/L) for 24 h, after which LDLR and AMPKα2 protein expression was determined by Western blotting (F), and the band density was quantified (G). (H, I) HepG2 cells were transfected with scrambled siRNA or AMPKG siRNA and treated with H007-M1 (100 μmol/L) for 24 h, after which LDLR and AMPKγ protein expression was determined by Western blotting (H), and the band density was quantified (I). (J, K) HepG2 cells were transfected with scrambled siRNA or AMPKB siRNA and treated with H007-M1 (100 μmol/L) for 24 h. LDLR and AMPKβ protein expression levels were subsequently determined via Western blotting (J), and the band density (K) was quantified. Statistical data are presented as mean ± SEM. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001 versus the indicated group (n = 3).

Figure 4

Hepatocyte Ampka1/2 deficiency blocked the effect of IMM-H007 on hypercholesterolemia. (A) Ampka^flox/flox and hAmpka KO mice were fed a high-fat (HF)/high-cholesterol (HC)/bile-salt (BS) diet with or without IMM-H007 (200 mg/kg body weight/day) for 5 weeks. (B) Determination of serum TC. (C) Determination of serum non-HDL-C. (D) Determination of serum HDL-C levels. (E) Determination of serum TG levels. Statistical data are presented as mean ± SEM. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001 versus the indicated group (n = 5). (F) RNA-seq assay with mouse liver total RNA. Enrichment analysis of the KEGG and REACTOME pathways (n = 3). (G, H) Protein levels of LDLR (G, scale bar = 100 μm) AMPKα1/2 (H, scale bar = 200 μm) in mouse liver was determined by immunofluorescent staining.

Figure 5

IMM-H007 upregulates LDLR expression via the AMPKα1/2–SENP1–IDOL signaling pathway. (A, B) HepG2 cells were posttranslationally transfected with scrambled siRNA, proprotein convertase subtilisin/kexin type 9 (PCSK9) siRNA (A), or inducible degrader of LDLR (IDOL) siRNA (B), after which LDLR, PCSK9, and IDOL protein expression was determined by Western blotting. (C) mRNA levels of PCSK9 and IDOL in HepG2 cells after H007-M1 treatment were measured via qRT-PCR (n = 3). (D) HEK-293T cells were transfected with Flag-IDOL, HA-small ubiquitin-like modifier 1 (SUMO1), or His-ubiquitin-conjugating enzyme (UBC9) for 24 h and then treated with H007-M1 (100 μmol/L) for 24 h. The SUMOylation of Flag-IDOL was determined by an IP assay using Flag beads. (E) HepG2 cells were transfected with scrambled siRNA or SUMO-specific protease 1 (SENP1) siRNA, after which the protein expression of IDOL, SENP1, and LDLR was determined by Western blotting. (F) Cas9-Ctrl and Cas9-AMPKA cells were treated with H007-M1 at the indicated concentrations for 24 h. IDOL, SENP1, and AMPKA protein expression was determined by Western blotting. (G) Protein expression in the livers of the mice in Fig. 4A was determined by Western blotting. (H) mRNA levels of SENP1 in HepG2 cells after H007-M1 treatment or transfection with AMPKA1/2 siRNA were measured via qRT-PCR (n = 3). (I) Putative cAMP response element (CRE) region in the hSENP1 promoter and the mutated sites. (J) HepG2 cells were transfected with the SENP1 promoter or SENP1-CRE-mut promoter and Renilla (as an internal control) overnight, followed by 100 μmol/L H007-M1 treatment or AMPKA1/2 overexpression for another 24 h. The activity of firefly and Renilla luciferases in the cellular lysate was determined via a dual-luciferase reporter assay (n = 8). Statistical data are presented as mean ± SEM. ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001; NS, not significant versus the indicated groups.

Figure 6

IMM-H007 promotes hepatic cholesterol metabolism by activating the AMPKα1–FXR pathway. (A) Heatmap of the differentially expressed genes related to bile acid metabolism in the liver identified via RNA-seq. (B) Cholesterol metabolism-related protein expression in the livers of the mice shown in Fig. 4A was determined by Western blotting. (C) 293T cells were transfected with the farnesoid X receptor (FXR) promoter and Renilla (as an internal control) overnight, followed by 100 μmol/L H007-M1 treatment for another 24 h. The activity of firefly and Renilla luciferases in the cellular lysate was determined via a dual-luciferase reporter assay (n = 5). (D, E) HepG2 cells were transfected with scrambled siRNA or AMPKA1 siRNA and treated with H007-M1 (100 μmol/L) for 24 h. Then, protein expression was determined by Western blotting (D), and the band density (E) was quantified. (F, G) HepG2 cells were transfected with scrambled siRNA or AMPKA2 siRNA and treated with H007-M1 (100 μmol/L) for 24 h. Then, protein expression was determined by Western blotting (F), and the band density was quantified (G) (n = 3). (H) Hepatic cholesterol quantitative analysis of the mice in Fig. 4A. (I) Hepatic total bile acid (TBA) quantitative analysis of the mice in Fig. 4A (n = 5). Statistical data are presented as mean ± SEM. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; NS, not significant versus the indicated group.

Figure 7

Hepatocyte AMPKα1/2 deficiency blocked the effect of IMM-H007 on lipid accumulation in the liver. (A) Liver photos of the mice in Fig. 4A. (B, C) H&E (B) and ORO (C) staining of liver sections from the mice in Fig. 4A. Scale bar = 100 μm. (D) TG quantitative analysis of total liver lipid extracts from the mice in Fig. 4A (n = 5). (E) Adipose triglyceride lipase (ATGL) mRNA expression in the livers of the mice shown in Fig. 4A was determined by qRT-PCR (n = 5). (F) ATGL, diacylglycerol O-acyltransferase 1 (DGAT1) protein expression in the livers of the mice in Fig. 4A was determined by Western blotting (n = 3). (G) Cas9-Ctrl and Cas9-AMPKA HepG2 cells were treated with H007-M1 at the indicated concentrations for 12 h. Protein expression was determined by Western blotting (n = 3). (H) Schematic diagram of the potential mechanism by which IMM-H007 decreases hepatic lipid accumulation. Statistical data are presented as mean ± SEM. ∗P < 0.05 versus the indicated groups.

Figure 8

Hepatocyte Ampka1/2 deficiency blocked the effect of IMM-H007 on atherosclerosis. (A) Ampka^flox/flox and hAmpka KO mice were administered AAV-PCSK9 through the tail vein. Subsequently, they were fed a HF/HC/BS diet for 16 weeks, with two groups receiving IMM-H007 (200 mg/kg body weight/day). (B) Determination of serum TC. (C) Determination of serum HDL-C. (D) Determination of serum non-HDL-C levels. (E) Determination of serum TG levels. (F) En face ORO staining and quantification via a computer-assisted image analysis protocol. The lesion areas are expressed as percentages of the en face aorta area. Scale bar = 5 mm. (G) Representative photomicrographs of aortic root sections stained with ORO. Scale bar = 500 μm. (H) Lesions in aortic root cross-sections were quantified via a computer-assisted image analysis protocol. The lesion areas are expressed as μm² in aortic root cross-sections. (I) Representative photomicrographs of aortic root sections stained with HE. Necrotic cores are marked by red dashed lines. Scale bar = 200 μm. (J) Representative photomicrographs of aortic root sections stained with Masson staining. Fibrous caps are marked by black dashed lines. Scale bar = 200 μm. (K, L) Masson staining followed by quantitative analysis of the necrotic core area (K) and fibrous cap area (L). (M) The collagen-positive areas are expressed as percentages of the aortic area (n = 8). Statistical data are presented as mean ± SEM. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; NS, not significant versus the indicated group.