Cardiac oxidative stress in a mouse model of neutral lipid storage disease

Abstract

Cardiac oxidative stress has been implicated in the pathogenesis of hypertrophy, cardiomyopathy and heart failure. Systemic deletion of the gene encoding adipose triglyceride lipase (ATGL), the enzyme that catalyzes the rate-limiting step of triglyceride lipolysis, results in a phenotype characterized by severe steatotic cardiac dysfunction. The objective of the present study was to investigate a potential role of oxidative stress in cardiac ATGL deficiency. Hearts of mice with global ATGL knockout were compared to those of mice with cardiomyocyte-restricted overexpression of ATGL and to those of wildtype littermates. Our results demonstrate that oxidative stress, measured as lucigenin chemiluminescence, was increased ~6-fold in ATGL-deficient hearts. In parallel, cytosolic NADPH oxidase subunits p67phox and p47phox were upregulated 4-5-fold at the protein level. Moreover, a prominent upregulation of different inflammatory markers (tumor necrosis factor α, monocyte chemotactant protein-1, interleukin 6, and galectin-3) was observed in those hearts. Both the oxidative and inflammatory responses were abolished upon cardiomyocyte-restricted overexpression of ATGL. Investigating the effect of oxidative and inflammatory stress on nitric oxide/cGMP signal transduction we observed a ~2.5-fold upregulation of soluble guanylate cyclase activity and a ~2-fold increase in cardiac tetrahydrobiopterin levels. Systemic treatment of ATGL-deficient mice with the superoxide dismutase mimetic Mn(III)tetrakis (4-benzoic acid) porphyrin did not ameliorate but rather aggravated cardiac oxidative stress. Our data suggest that oxidative and inflammatory stress seems involved in lipotoxic heart disease. Upregulation of soluble guanylate cyclase and cardiac tetrahydrobiopterin might be regarded as counterregulatory mechanisms in cardiac ATGL deficiency.

Keywords: (s)GC; (soluble) guanylate cyclase; 2,2-diethyl-1-nitroso-oxyhydrazine; ATGL; ATGL(−/−); Adipose triglyceride lipase; BH(2); BH(4); Cardiac hypertrophy; DAG; DEA/NO; FFA; GAPDH; IL-6; Inflammation; MCP-1; Mac-2; Mn(III)tetrakis (4-benzoic acid) porphyrin chloride; MnTBAP; NADPH; NADPH oxidase; NO; NOX; ONOO(−); Oxidative stress; PBS; PKC; PPARα; SOD; TG; TNFα; VASP; adipose triglyceride lipase; adipose triglyceride lipase knockout; diacylglycerol; dihydrobiopterin, [2-amino-6-(1,2-dihydroxypropyl)-7,8-dihydro-1H-pteridin-4-one]; eNOS; endothelial nitric oxide synthase; free fatty acid; galectin-3; glyceraldehyde-3-phosphate dehydrogenase; iNOS; inducible nitric oxide synthase; interleukin 6; monocyte chemotactic protein-1; nNOS; neuronal nitric oxide synthase; nicotinamide adenine dinucleotide phosphate; nitric oxide; pVASP; peroxisome proliferator receptor α; peroxynitrite; phosphate-buffered saline; phosphorylated vasodilator-stimulated phosphoprotein; protein kinase C; superoxide dismutase; tetrahydrobiopterin, [(6R)-2-amino-6-[(1R,2S)-1,2-dihydroxypropyl]-5,6,7,8-tetrahydropteridin-4(1H)-one]; triacylglycerol; tumor necrosis factor α; vasodilator-stimulated phosphoprotein.

Graphical abstract

Fig. 1

NOX2-dependent oxidative stress in cardiac ATGL deficiency. (A) gp91ds-tat-sensitive lucigenin chemiluminescence was measured in cardiac homogenates prepared from WT (open bars), ATGL(−/−) (solid bars), WT/MHC-A35 (striped bars), and ATGL(−/−)/MHC-A35 mice (gray bars) as described in Materials and methods. Data represent mean values ± S.E.M. of 4 individual experiments. Inset: Inhibition of NADPH-induced chemiluminescence by gp91ds-tat (50 μM), Cu, Zn SOD (500 U/ml) or MnTBAP (10 μM) in hearts of WT (open bars) and ATGL(−/−) mice (solid bars). (B) Cardiac NOX2 mRNA and protein expression were measured by qPCR and Western blot, respectively. Data are expressed as folds of WT controls (WT = 1) and represent mean values ± S.E.M. of 6 experiments. qPCR measurements were performed in triplicates. (C, D) Protein expression of p47^phox and p67^phox subunits, respectively. Data are expressed as folds of WT controls (WT = 1) and represent mean values ± S.E.M. of 6 individual experiments. (E) Representative Western blots of NOX2, p47^phox, p67^phox, and GAPDH; *p < 0.05 ATGL(−/−) vs WT.

Fig. 2

Cardiac NOX4 is upregulated in ATGL deficiency. (A) NOX4 mRNA was measured in hearts of WT (open bars), ATGL(−/−) (solid bars), WT/MHC-A35 (striped bars), and ATGL(−/−)/MHC-A35 mice (gray bars) by qPCR as described. Data represent mean values ± S.E.M. of 3–5 experiments performed in triplicate. (B) Protein expression of NOX4 was measured by Western blot analysis. Data are expressed as folds of WT controls (WT = 1) and represent mean values ± S.E.M. of 6 individual experiments. A representative Western blot is shown as inset; *p < 0.05 ATGL(−/−) vs WT.

Fig. 3

Upregulation of inflammatory markers in ATGL deficiency. (A) TNFα mRNA expression in cardiac homogenates of WT (open bars), ATGL(−/−) (solid bars), WT/MHC-A35 (striped bars), and ATGL(−/−)/MHC-A35 mice (gray bars) was measured by qPCR using TaqMan probes. (B, C) MCP-1 and IL-6 mRNA expression, respectively. Data represent mean values ± S.E.M. of 4–6 experiments performed in triplicate; *p < 0.05 ATGL(−/−) and ATGL(−/−)/MHC-A35 vs WT. (D) Protein expression of Mac-2 was measured by Western blot analysis. Data are expressed as folds of WT controls (WT = 1) and represent mean values ± S.E.M. of 9 individual experiments. (E) Activation of cardiac PKCs was measured by Western blot as phosphorylation of a crucial threonine. Data are expressed as folds of WT controls (WT = 1) and represent mean values ± S.E.M. of 9 individual experiments. *p < 0.05 ATGL(−/−) vs WT. (F) Representative Western blots of Mac-2, pPKC (pan), and GAPDH.

Fig. 4

Effects on NO/cGMP signaling. (A) 10,000 g supernatants of hearts from WT (open bars), ATGL(−/−) (solid bars), WT/MHC-A35 (striped bars), and ATGL(−/−)/MHC-A35 mice (gray bars) were prepared and assayed for Ca^2 +-dependent [³H]l-citrulline formation. Data are mean values ± S.E.M. of 4 experiments performed in duplicate. (B) Cardiac sGC activity was measured in cytosols stimulated with 0.1 mM DEA/NO with and without YC-1 (0.1 mM). The inset depicts basal enzyme activity assayed in the presence of MnCl₂. Data are mean values ± S.E.M. of 5 experiments performed in duplicate. *p < 0.05 ATGL(−/−) vs WT. (C) cGMP levels were determined by radioimmunoassay. Data represent mean values ± S.E.M. of 5–7 experiments performed in duplicate. (D) Phosphorylation of VASP at serine 239 was quantified by Western blot. Data are presented as the ratio of phosphorylated to total VASP protein and represent mean values ± S.E.M. of 6 individual experiments.

Fig. 5

Cardiac BH₄ is increased in ATGL deficiency. (A, B) BH₄ and BH₂ were quantified in hearts of WT (open bars), ATGL(−/−) (solid bars), WT/MHC-A35 (striped bars), and ATGL(−/−)/MHC-A35 mice (gray bars) by HPLC analysis. (C) BH₄/BH₂ ratio. Data represent mean values ± S.E.M. of 8 individual experiments. *p < 0.05 ATGL(−/−) vs WT.

Fig. 6

Systemic application of MnTBAP (5 mg × kg^− 1 × day^− 1) aggravates oxidative stress in ATGL deficiency. (A) gp91ds-tat-sensitive lucigenin chemiluminescence was measured in cardiac homogenates prepared from MnTBAP-treated WT (open bars) and ATGL(−/−) mice (solid bars) and compared to that of non-treated animals. Data represent mean values ± S.E.M. of 4 individual experiments. *p < 0.05 non-treated and MnTBAP-treated ATGL(−/−) vs non-treated WT; §p < 0.05 non-treated ATGL(−/−) vs MnTBAP-treated ATGL(−/−). (B, C) Cardiac BH₄ and BH₂ were determined by HPLC. Data represent mean values ± S.E.M. of 4 individual experiments. *p < 0.05 ATGL(−/−) vs WT.

Scheme 1

Signaling pathways that link cardiac ATGL deficiency to oxidative inflammatory stress and increased sGC activity. Impaired PPARα signaling in ATGL deficiency induces cardiac hypertrophy (path 1) that is compensated by increased sGC activity (path 6). Impaired PPARα signaling leads to increased superoxide production by NADPH oxidases (path 2) and augmented expression of inflammatory markers (path 3). Elevated TNFα levels induce expression of cardiac NADPH oxidases via activation of PKC (path 4). Lipid droplet surface-binding proteins directly initiate oxidative and inflammatory processes (path 5). NADPH oxidase-generated superoxide scavenges NO to form ONOO⁻ (path 7). Reduced NO availability induces activation of cardiac sGC. NADPH oxidase-derived superoxide production enhances cardiac hypertrophy and contractile dysfunction (path 8).