Cardiomyocyte lipids impair β-adrenergic receptor function via PKC activation

Abstract

Normal hearts have increased contractility in response to catecholamines. Because several lipids activate PKCs, we hypothesized that excess cellular lipids would inhibit cardiomyocyte responsiveness to adrenergic stimuli. Cardiomyocytes treated with saturated free fatty acids, ceramide, and diacylglycerol had reduced cellular cAMP response to isoproterenol. This was associated with increased PKC activation and reduction of β-adrenergic receptor (β-AR) density. Pharmacological and genetic PKC inhibition prevented both palmitate-induced β-AR insensitivity and the accompanying reduction in cell surface β-ARs. Mice with excess lipid uptake due to either cardiac-specific overexpression of anchored lipoprotein lipase, PPARγ, or acyl-CoA synthetase-1 or high-fat diet showed reduced inotropic responsiveness to dobutamine. This was associated with activation of protein kinase C (PKC)α or PKCδ. Thus, several lipids that are increased in the setting of lipotoxicity can produce abnormalities in β-AR responsiveness. This can be attributed to PKC activation and reduced β-AR levels.

Fig. 1.

Treatment of cardiomyocytes with long-chain fatty acids (LCFA) alters β-adrenergic receptor (β-AR) function, heart failure marker gene expression, and PKC activity. A: Oil Red O staining of AC-16 cells treated with 0.4 mM palmitic acid (PA) or oleic acid (OA) for 14 h. B and C: triglyceride (TG; B) and free fatty acid (FFA; C) levels in OA- and PA-treated AC-16 cells. D: brain natriuretic peptide (BNP) mRNA levels determined by quantitative real-time (qRT)-PCR analysis; n = 7. *P < 0.01 compared with cells that were not treated with LCFA. E: cAMP levels in LCFA-treated cells stimulated with 100 nM isoproterenol for 15 min; n = 6. **P < 0.005 compared with cells that were not stimulated with isoproterenol. ∞P < 0.001 compared with cells that were treated with PA and stimulated with isoproterenol; 1-fold corresponds to 5.4 nM cAMP. F: β-AR density on plasma membrane preparations in baseline and stimulated conditions (100 nM isoproterenol, 15 min); n = 3, *P < 0.05 compared with cells that were not treated with LCFA; ‡P < 0.05 compared with cells that were not stimulated with isoproterenol. G: β-AR distribution in OA- and PA- ± PKC inhibitor-treated (Ro-31-8220, 5 μM, 14 h) GFP-β₂AR cells as monitored by confocal microscopy. H: Western blots of PKCα, PKCβ, PKCδ, PKCϵ, and PKCζ protein levels in membrane (M) and cytosolic (C) fractions obtained from PA-treated AC-16 cells. I and J: total (I) and isoform-specific (J) PKC activity in AC-16 cells treated with 0.4 mM PA for 14 h; n = 3, P < 0.05 compared with control cells. K: Western blots of PKCα and PKCδ protein in membrane (M) and cytosolic (C) fractions obtained from PA- or OA-treated AC-16 cells.

Fig. 2.

Role of ceramide in PA-mediated lipotoxicity in AC-16 cells. A: ceramide levels determined by diacylglycerol kinase assay in lipid extracts from control (CTRL) and 0.4 mM PA- and 0.4 mM OA-treated cells for 14 h. Control cells were treated with methanol. B: BNP mRNA levels determined by qRT-PCR analysis in cells treated with 10 μM of C6 ceramide for 14 h; n = 4. *P < 0.05. C: intracellular cAMP levels in AC-16 cells treated with 10 μM of C6 ceramide for 14 h and stimulated with 100 nM isoproterenol for 15 min; n = 6. **P < 0.01; 1-fold corresponds to 15.5 nM cAMP. D: intracellular cAMP levels in isoproterenol-stimulated AC-16 cells treated with 0.4 mM PA + 0.2 μM myriocin for 14 h and stimulated with 100 nM isoproterenol for 15 min; n = 6. *P < 0.05 compared with control cells that were not treated with LCFA and myriocin; 1-fold corresponds to 23.0 nM cAMP. E: Western blots of PKCα and PKCδ in membrane and cytosolic fractions of AC-16 cells treated with 10 μM C6 ceramide.

Fig. 3.

Role of diacylglycerol (DAG) in PA-mediated lipotoxicity in AC-16 cells. A: DAG levels determined by DAG kinase assay in lipid extracts from cells treated with 0.4 mM PA or OA for 14 h. Control cells were treated with methanol. B and C: BNP mRNA levels in AC-16 cells treated with 100 μM dioctanoylglycerol for 14 h; n = 4. P < 0.05 (B) or dipalmitoylglycerol; n = 4 (C). Control cells were treated with DMSO. D: intracellular cAMP levels in AC-16 cells treated with 100 μM dioctanoylglycerol for 14 h and stimulated with 100 nM isoproterenol for 15 min. Control cells were treated with DMSO; n = 6. P < 0.05 compared with cells that were not stimulated with isoproterenol; 1-fold corresponds to 14.0 M cAMP. E: Western blots of PKCα and PKCδ protein levels in membrane and cytosolic fractions obtained from AC-16 cells treated with 100 μM dioctanoylglycerol. Control cells were treated with DMSO. *P < 0.01.

Fig. 4.

Role of PKCs in the lipid-driven impairment of β-AR function in AC-16 cells A: total PKC activity in AC-16 cells treated with 0.4 mM PA for 14 h and 5 μM Ro-31-8220 [PKC inhibitor (PKCinh)]; n = 6. P < 0.05 compared with control cells that were treated neither with PA nor with PKCinh. B and C: cAMP levels (B) and membrane β-AR density (C) in AC-16 cells treated with 0.4 mM PA and 5 μM PKCinh for 14 h and then stimulated with 100 nM isoproterenol for 15 min; n = 6. P < 0.05 compared with cells that were not treated with PA; n = 3. **P < 0.05 compared with cells that were not stimulated with isoproterenol and ‡P < 0.05 compared with cells that were not treated with LCFA; 1-fold corresponds to 23.0 nM cAMP. D and E: Western blots of PKCα, PKCδ, and β-actin total protein levels obtained from AC-16 cells treated with 50 nM siRNA oligos that target PKCα (D) or PKCδ (E). Nontransfected cells and cells treated with scrambled siRNA oligos were used as controls. F: intracellular cAMP levels in AC-16 cells transfected with 50 nM siRNA oligos that target PKCα or PKCδ, treated with 0.4 mM PA for 14 h and then stimulated with 100 nM isoproterenol for 15 min. Control cells were treated with methanol and scrambled siRNA; n = 6. *P < 0.05 compared with control cells; **P < 0.05 compared with cells that were not stimulated with isoproterenol; ‡P < 0.05 compared with cells that were not treated with LCFA; ∼P < 0.05 compared with cells treated with scrambled siRNA and PA; 1-fold corresponds to 4.8 nM cAMP.

Fig. 5.

Impaired β-AR responsiveness in isolated adult rat primary cardiomyocytes. A: rat primary cardiomyocytes were treated for 14 h with 0.4 mM PA or OA and then stained with Oil Red O. Control cells were treated with methanol. B: intracellular cAMP levels in primary cardiomyocytes treated with LCFA, 10 mM l-glutathione, or 50 μM salubrinal and stimulated with 100 nM isoproterenol for 15 min; n = 6. *P < 0.05 compared with cells that were not treated with LCFA; 1-fold corresponds to 1.5 nM cAMP.

Fig. 6.

Impaired β-AR responsiveness to dobutamine in lipotoxic hearts of mice expressing specifically in cardiomyocytes a GPI-anchored lipoprotein lipase [α-myosin heavy chain (MHC)-LpL^GPI]. A: LVdP/dt (1st-order derivative of the left ventricular pressure waveform) recording in response to increasing doses of dobutamine in 4-mo-old wild type (WT) and α-MHC-LpL^GPI mice. Dobutamine was injected via the femoral vein every 1–2 min, and the LVP waveform was recorded via a pressure transducer/catheter in the left ventricle; n = 4/genotype. B: LVdP/dt max as an index of cardiac contractility to increasing doses of dobutamine in 4-mo-old WT and α-MHC-LpL^GPI mice. Data computation was achieved using the PowerLab Software; n = 4/genotype. C: LVdP/dt min as an index of myocardial relaxation to increasing doses of dobutamine in 4-mo-old WT control and α-MHC-LpL^GPI mice; n = 4/genotype. D: membrane β-AR density in 4-mo-old WT and α-MHC-LpL^GPI hearts measured by saturation-ligand binding assay; n = 4–5/genotype. E: mouse plasma norepinephrine (NE) and epinephrine (Epi) levels measured by ELISA after a 4-h fast; n = 6–7. F: Western blots of PKCα and PKCδ in the membrane and cytosolic fractions of hearts of 4-mo-old WT and α-MHC-LpL^GPI mice; n = 3/genotype.

Fig. 7.

Impaired β-AR responsiveness in in hearts of mice expressing specifically in cardiomyocytes a GPI-anchored peroxisome proliferator-activated receptor-γ (α-MHC-PPARγ) or acyl-CoA synthetase (α-MHC-ACS) and high-fat-fed C57BL/6 mice. A–C: LVdP/dt max as an index of cardiac contractility to increasing doses of dobutamine, 1–30 μg/kg, α-MHC-PPARγ and control C57BL/6 littermate mice (A), α-MHC-ACS and control FVB littermate mice (B), or high-fat- and chow-fed C57BL/6 mice (chow: 32.5 ± 0.6 g, HFD: 53.7 ± 1.1 g; C). PowerLab Software was used for data computation; n = 4/genotype. *P < 0.05 compared with WT, P < 0.05 compared with WT basal. D–F: Western blots of PKCα and PKCδ protein levels in the membrane and cytosolic fractions obtained from cardiac tissue of WT and α-MHC-PPARγ (D), α-MHC-ACS (E), or high-fat-fed mice (F).