Heme oxygenase gene targeting to adipocytes attenuates adiposity and vascular dysfunction in mice fed a high-fat diet

Abstract

We examined the hypothesis that adipocyte dysfunction in mice fed a high-fat (HF) diet can be prevented by lentiviral-mediated and adipocyte specific-targeting delivery of the human heme oxygenase-1 (aP2-HO-1). A bolus intracardial injection of aP2-HO-1 resulted in expression of human HO-1 for up to 9.5 months. Transduction of aP2-HO-1 increased human HO-1 expression in fat tissues without affecting murine HO-1. In mice fed a HF diet, aP2-HO-1 transduction attenuated the increases in body weight, blood glucose, blood pressure, and inflammatory cytokines, as well as the content of both visceral and subcutaneous fat. Transduction of aP2-HO-1 increased the numbers of adipocytes of small cell size (P<0.05), insulin sensitivity (P<0.05), adiponectin levels, as well as vascular relaxation to acetylcholine compared with HF mice administered the aP2-green fluorescent protein. Adipocytes of mice fed a HF diet expressed high levels of peroxisome proliferator activator receptor, aP2, C/EBP, and Wnt5b proteins and displayed marked increases in Peg1/Mesoderm specific transcript (P<0.03). Transduction of aP2-HO-1 lowered the elevated levels of these proteins and increased Sonic hedgehog, Wnt10b, and β-catenin (P<0.05). Inhibition of HO activity by administration of tin mesoporphyrin to HF-fed mice transduced with the aP2-HO-1 reversed the decrease in Peg1/Mesoderm-specific transcript, TNFα, and MCP-1 levels. Collectively, this novel study demonstrates that adipocyte-specific overexpression of HO-1 attenuates HF-mediated adiposity and vascular dysfunction; increases insulin sensitivity; and improves adipocyte function by increasing adiponectin, Shh, and WNT10b, and by decreasing inflammatory cytokines. These effects are reversed by the HO activity inhibitor, stannous mesoporphyrin.

Figure 1

A) body weight; B) visceral fat and C) subcutaneous fat; D) serum TNFα; and E) serum MCP-1 in mice fed a normal diet (control) and mice fed a HF diet and transduced with the aP2-GFP (aP2-GFP) or aP2-HO-1 untreated (aP2-HO-1) and treated with SnMP (+SnMP) (*p<0.05 vs. control; **p<0.05 vs. HF+aP2-GFP; ^#p<0.05 vs. HF+aP2-HO-1).

Figure 2

A)Hematoxylin-eosin staining of visceral fat and quantitative analysis of adipocyte size in visceral fat surrounding the aorta in mice fed a normal diet (control) and mice fed a HF diet and transduced with the aP2-GFP (aP2-GFP) or aP2-HO-1 (aP2-HO-1). Results are means±S.E.; *P<0.05 vs. control; ^#p<0.05 vs. HF+aP2-GFP, +p<0.05 vs. aP2-HO-1. B)Plasma adiponectin levels C)Blood glucose levels and D)intraperitoneal insulin tolerance test (IPITT)in mice fed a normal diet (control) and mice fed a HF diet and transduced with the aP2-GFP (aP2-GFP) or aP2-HO-1 untreated (aP2-HO-1) and treated with SnMP (+SnMP). Results are mean±SE; n=4; *p<0.05 vs. control; **p<0.05 vs. aP2-GFP; #p<0.05 vs. aP2-HO-1.

Figure 3

A) Blood pressure, which was measured by tail cuff method. Results are mean±SE (n=10/group). Systolic blood pressure; *p<0.05 vs. control; **p<0.05 vs. aP2-GFP; ^#p<0.05 vs. aP2-HO-1. Diastolic blood pressure; ⁺p<0.05 vs. control; ⁺⁺p<0.05 vs. aP2-GFP; ^##p<0.05 vs. aP2-HO-1. B) Relaxation-response curves to acetylcholine in aorta. Results are means±SE, n=5; *p<0.05 vs. control; **p<0.05 vs. aP2-GFP; ^#p<0.05 vs. aP2-HO-1.

Figure 4

Western blots and densitometry analyses of (A) PPARγ, C/EBPα, (B) adiponectin and Shh proteins in adipocytes from mice fed a normal diet (control) and mice fed a HF diet and transduced with the aP2-GFP (aP2-GFP) or aP2-HO-1 untreated (aP2-HO-1) and treated with SnMP (+SnMP). Results are mean±SE; n=4; *p<0.05 vs. control; **p<0.05 vs. HF+aP2-GFP; ^#p<0.05 vs. HF+aP2-HO-1.

Figure 5

Western blots and densitometry analyses of (A) aP2 and (B)Mest/peg1 proteins in adipocytes from mice fed a normal diet (control) and mice fed a HF diet and transduced with the aP2-GFP (aP2-GFP) or aP2-HO-1 untreated (aP2-HO-1) and treated with SnMP (+SnMP). Results are mean±SE; n=4; *p<0.05 vs. control; **p<0.05 vs. HF+aP2-GFP; ^#p<0.05 vs. HF+aP2-HO-1. Western blots and densitometry analyses of (C) Wnt10b, (D) Wnt5b and (E) β-catenin proteins in adipocytes from mice fed a normal diet (control) and mice fed a HF diet and transduced with the control vector aP2-GFP (aP2-GFP) or with the aP2-HO-1 (aP2-HO-1) untreated and treated with SnMP (+SnMP). Results are mean±SE; n=4; *p<0.05 vs. control; **p<0.05 vs. HF+aP2-GFP; ^#p<0.05 vs. HF+aP2-HO-1.

Figure 6

Proposed mechanisms underlying the adiposity lowering effect of adipocyte-specific upregulation of HO-1. Targeted overexpression of HO-1 to the adipocyte is sufficient to counteract the detrimental effects of high fat diet on body weight, adiposity and insulin sensitivity. Within the adipocyte, HO-1 suppresses adipocyte differentiation by decreasing expression of key regulators including PPARγ, E/CBPα, Peg1/Mest and aP2 leading to decreases in lipid accumulation and increases of pre-adipocytes and “healthy adipocytes” that produce cytoprotectiveadipokines such as adiponectin.