Myocardial oxidative stress contributes to transgenic β₂-adrenoceptor activation-induced cardiomyopathy and heart failure

Abstract

Background and purpose: While maintaining cardiac performance, chronic β-adrenoceptor activation eventually exacerbates the progression of cardiac remodelling and failure. We examined the adverse signalling pathways mediated by nicotinamide adenine dinucleotide phosphate (NADPH) oxidase and reactive oxygen species (ROS) after chronic β₂-adrenoceptor activation.

Experimental approach: Mice with transgenic β₂-adrenoceptor overexpression (β₂-TG) and non-transgenic littermates were either untreated or treated with an antioxidant (N-acetylcysteine, NAC) or NADPH oxidase inhibitors (apocynin, diphenyliodonium). Levels of ROS, phosphorylated p38 mitogen-activated protein kinase (MAPK), pro-inflammatory cytokines and collagen content in the left ventricle (LV) and LV function were measured and compared.

Key results: β₂-TG mice showed increased ROS production, phosphorylation of p38 MAPK and heat shock protein 27 (HSP27), expression of pro-inflammatory cytokines and collagen, and progressive ventricular dysfunction. β₂-adrenoceptor stimulation similarly increased ROS production and phosphorylation of p38 MAPK and HSP27 in cultured cardiomyocytes. Treatment with apocynin, diphenyliodonium or NAC reduced phosphorylation of p38 MAPK and HSP27 in both cultured cardiomyocytes and the LV of β₂-TG mice. NAC treatment (500 mg·kg⁻¹ ·day⁻¹) for 2 weeks eliminated the up-regulated expression of pro-inflammatory cytokines and collagen in the LV of β₂-TG mice. Chronic NAC treatment to β₂-TG mice from 7 to 10 months of age largely prevented progression of ventricular dilatation, preserved contractile function (fractional shortening 37 ± 5% vs. 25 ± 3%, ejection fraction 52 ± 5% vs. 32 ± 4%, both P < 0.05), reduced cardiac fibrosis and suppressed matrix metalloproteinase activity.

Conclusion and implications: β₂-adrenoceptor stimulation provoked NADPH oxidase-derived ROS production in the heart. Elevated ROS activated p38 MAPK and contributed significantly to cardiac inflammation, remodelling and failure.

Figure 1

Increased ROS production in the LV of β₂-TG mice. A, representative fluorescent probe DHE staining for oxidative fluorescent signal of LV sections from NTG and β₂-TG mice (5-month-old) and densitometric analysis of DHE fluorescence. (Bar = 50 µm). B, ROS production was determined in freshly harvested LV (5-month-old) by electron spin resonance assay using superoxide probe CMH, or C, lucigenin-enhanced chemiluminescence assay for NADPH oxidase activity. D, quantitative real-time PCR for mRNA expression of NOX2 and NOX4 isoforms in the LV of NTG and β₂-TG mice at both 5 and 15 months of age. Data are presented as relative changes to age-matched NTG mice (n= 5–9/group). *P < 0.05 versus NTG mice.

Figure 2

p38 MAPK activation by NADPH oxidase-derived ROS in the LV of β₂-TG. A, β₂-TG mice (5- to 7-month-old) were treated with N-acetylcysteine (NAC, 250 mg·kg⁻¹, i.p.), or NADPH inhibitors, apocynin (Apo, 2 mg·kg⁻¹, i.p.) or diphenyliodonium chloride (DPI, 1 mg·kg⁻¹, i.p.). The LV was collected 1 h after the treatment. Western blot analyses were performed using antibodies against phospho-p38 MAPK, p38 MAPK, phospho-HSP27, HSP27 and tubulin respectively. B, Levels of phosphorylated and total p38 MAPK or HSP27 and tubulin, quantified by densitometry and presented as changes relative to age-matched NTG mice. *P < 0.05 versus NTG mice; ^†P < 0.05 versus vehicle-treated β₂-TG group, n= 6/group.

Figure 3

p38 MAPK activation by NADPH oxidase-derived ROS in cultured cardiomyocytes following β₂-adrenoceptor stimulation. A, Serum-starved cells were loaded with 10 µmol·L⁻¹ H₂DCFDA for 30 min at 37°C. Cells were then preincubated with CGP20712A (CGP, 1 µmol·L⁻¹, selective β₁-adrenoceptor antagonist), together with ICI118551 (ICI, 1 µmol·L⁻¹, selective β₂-adrenoceptor antagonist), NAC (1 mmol·L⁻¹), DPI (10 µmol·L⁻¹) or apocynin (100 µmol·L⁻¹), respectively, for 30 min, and then stimulated with isoprenaline (ISO, 1 µmol·L⁻¹) for 10 min. Cellular fluorescence intensity was measured using a fluorescence reader and expressed as the -fold increase of unstimulated cells. B, Cells were preincubated with CGP, together with apocynin, DPI, NAC, PEG-SOD (25 U·mL⁻¹) or ICI, respectively, for 30 min, and then stimulated with ISO for 10 min. Western blot analyses were performed using antibodies against phospho-p38 MAPK, p38 MAPK, phospho-HSP27 and HSP27 respectively. The blot shown is a representative of three similar experiments. Pooled data are presented as relative changes to ISO + CGP stimulated cells. *P < 0.05 versus ISO + CGP.

Figure 4

Myocardial fibrosis and extracellular matrix remodelling in the LV of β₂-TG mice. A, Gelatine zymography revealed increased levels of activated and latent forms of MMP-2 (A-MMP-2 and L-MMP-2) in β₂-TG than NTG mice (5- and 15-month-old). Data are presented as changes relative to age-matched NTG mice (n= 5–9/group). B, Collagen content in the LV of β₂-TG and NTG mice (5 and 15 months of age, n= 5–9/group) was quantified by hydroxyproline assay. *P < 0.05 versus respective NTG controls; ^†P < 0.05 versus untreated β₂-TG group.

Figure 5

β₂-TG mice developed progressive LV dysfunction with aging. Representative M-mode echocardiographic tracings from short-axis LV 2-D images (A) and pressure-volume loops (B) from NTG and β₂-TG mice (5- and 15-month-old), β₂-TG mice at 5 months of age manifested similar chamber size and enhanced contractility, while they displayed impaired LV compliance versus their NTG counterparts. Older β₂-TG mice (15-month-old) showed stiffened LV, impaired systolic and diastolic function and dilated LV.

Figure 6

The antioxidant NAC significantly reduced oxidative stress and restored gene expression in the LV of β₂-TG mice. β₂-TG mice (7-month-old) were treated with NAC (500 mg·kg⁻¹·day⁻¹ in drinking water) for 2 weeks. ROS production was determined by DHE fluorescence, Bar = 50 µm (A), electron spin resonance assay using superoxide probe, CMH, or lucigenin-enhanced chemiluminescence assay (B). C, quantitative real-time PCR for mRNA expression of NOX2 and NOX4 with data presented as changes relative to NTG (5–10/group). D, mRNA expression of β₂-adrenoceptor (β₂-AR) transgene and endogenous α-MHC was assayed by quantitative real-time PCR, and expressed as -fold changes relative to untreated β₂-TG mice. Changes in the expression level of human β₂-adrenoceptor were plotted against that of α-MHC. β₂-adrenoceptor density was assayed by binding assay (n= 6/group). *P < 0.05 versus NTG mice; ^†P < 0.05 versus untreated β₂-TG group.

Figure 7

Chronic treatment with NAC prevented the progression of cardiomyopathy. β₂-TG and NTG mice (7-month-old) were treated with NAC (500 mg·kg⁻¹·day⁻¹ in drinking water) for 3 months. LV function was assessed by echocardiography (A, B) and pressure-volume catheterization (C). LVDd, LV dimension at end-diastole; LVDs, LV dimension at end-systole; FS, fractional shortening; EDPVR, end-diastolic pressure-volume relationship; ESPVR, end-systolic pressure-volume relationship. Data (n= 6–17/group) were analysed by two-way repeated measures anova followed by post hoc tests. *P < 0.05 versus NTG mice; ^†P < 0.05 versus untreated β₂-TG mice; ^‡P < 0.05 versus same mice at 7-month-old.

Figure 8

Chronic treatment with NAC halted extracellular matrix remodelling and cardiomyocyte apoptosis. A, gelatine zymography was performed to determine the level of active and latent form of MMP-2 (A-MMP-2 and L-MMP-2) in the LV of β₂-TG and NTG mice after 3 months of NAC treatment. Data are presented as relative changes to NTG mice (n= 6–12/group). B, LV sections of NTG, untreated and NAC-treated β₂-TG mice were stained with Masson's trichrome or Picrosirius red (bar = 100 µm). Cardiomyocyte diameter and LV collagen content were measured (n= 3/group). LV collagen content was quantified by hydroxyproline assay (n= 6–12/group). C, TUNEL-positive LV cardiomyocyte nuclei (arrows) were counted under fluorescent or light microscopy (left panels) counterstained with DAPI (top right) or haematoxylin (bottom right). Results are expressed as a percentage of total cardiomyocytes (4000–6000/heart). *P < 0.05 versus NTG mice; ^†P < 0.05 versus untreated β₂-TG mice. n= 4/group.