Deletion of Osteopontin Enhances β₂-Adrenergic Receptor-Dependent Anti-Fibrotic Signaling in Cardiomyocytes

Abstract

Cardiac β₂-adrenergic receptors (ARs) are known to inhibit collagen production and fibrosis in cardiac fibroblasts and myocytes. The β₂AR is a Gs protein-coupled receptor (GPCR) and, upon its activation, stimulates the generation of cyclic 3',5'-adenosine monophosphate (cAMP). cAMP has two effectors: protein kinase A (PKA) and the exchange protein directly activated by cAMP (Epac). Epac1 has been shown to inhibit cardiac fibroblast activation and fibrosis. Osteopontin (OPN) is a ubiquitous pro-inflammatory cytokine, which also mediates fibrosis in several tissues, including the heart. OPN underlies several cardiovascular pathologies, including atherosclerosis and cardiac adverse remodeling. We found that the cardiotoxic hormone aldosterone transcriptionally upregulates OPN in H9c2 rat cardiac myoblasts-an effect prevented by endogenous β₂AR activation. Additionally, CRISPR-mediated OPN deletion enhanced cAMP generation in response to both β₁AR and β₂AR activation in H9c2 cardiomyocytes, leading to the upregulation of Epac1 protein levels. These effects rendered β₂AR stimulation capable of completely abrogating transforming growth factor (TGF)-β-dependent fibrosis in OPN-lacking H9c2 cardiomyocytes. Finally, OPN interacted constitutively with G_αs subunits in H9c2 cardiac cells. Thus, we uncovered a direct inhibitory role of OPN in cardiac β₂AR anti-fibrotic signaling via cAMP/Epac1. OPN blockade could be of value in the treatment and/or prevention of cardiac fibrosis.

Keywords: CRISPR; Epac1; cAMP; cardiac myocytes; fibrosis; osteopontin; signal transduction; β2-adrenergic receptor.

Figure 1

β₂-adrenergic receptor (AR) inhibits aldosterone-induced osteopontin (OPN) upregulation in H9c2 cardiomyocytes. H9c2 cells were treated with 10 nM aldosterone (Aldo), 10 µM salbutamol (Sal), or 10 nM aldosterone in the presence of 10 µM salbutamol (Sal+Aldo) for 2 h. At the end of this 2-h period, cells were harvested, total RNA isolated, and real-time PCR for OPN mRNA quantitation was performed. *, p < 0.05, vs. any other treatment; n = 4 independent experiments/condition.

Figure 2

Enhanced β₂AR-dependent cAMP accumulation in the absence of OPN in H9c2 cardiomyocytes. (A) Western blotting to confirm OPN CRISPR-mediated deletion in H9c2 cells infected with OPN-specific CRISPR lentivirus (knockout, KO) or mock CRISPR lentivirus (wild type, WT). Blotting for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) is also shown as loading control. Representative blots from three independent experiments per condition with similar results are shown. (B) cAMP accumulation in response to 10 µM isoproterenol (Iso) or 10 µM salbutamol (Salb) in control WT and in OPN-depleted (OPN KO) H9c2 cells, expressed as percent of the respective cAMP production induced by 10 µM forskolin (Fsk). *, p < 0.05, vs. WT; n = 3 independent experiments/condition/cell clone. (C) cAMP accumulation in response to 10 µM forskolin (Fsk) or vehicle (Veh) in control WT and in OPN-depleted (OPN KO) H9c2 cells. No significant differences were observed between WT-Fsk and OPN KO-Fsk at p = 0.05 (n = 3 independent experiments/condition/cell clone).

Figure 3

Enhanced βAR-dependent Epac1 protein levels in the absence of OPN in H9c2 cardiomyocytes (Epac: exchange protein directly activated by cAMP). H9c2 cells were treated with 10 µM isoproterenol (Iso) or vehicle for 24 h in the presence (WT) or absence (KO) of OPN, then cells were harvested and total protein extracts prepared for Epac1 immunoblotting. Representative blots are shown in (A), including GAPDH as loading control, and the densitometric quantitation of three independent experiments per condition performed in duplicate is shown in (B). *, p < 0.05, n = 3.

Figure 4

Absence of OPN potentiates the β₂AR-mediated inhibition of TGFβ-dependent profibrotic factor mRNA induction in H9c2 cells. mRNA levels of (A) type I collagen, (B) type III collagen, (C) type IV collagen, and (D) fibronectin, in WT or OPN-depleted (OPN KO) H9c2 cells treated with 10 ng/mL TGF-β₁ (TGF) with or without 10 µM salbutamol (Salb). *, p < 0.05, vs. TGF; ^#, p < 0.05, vs. WT-Salb; n = 3 independent experiments per condition (two-way ANOVA with Bonferroni test).

Figure 5

OPN opposes β₂AR signaling via physical interaction with G_αs in H9c2 cardiomyoblasts. Co-immunoprecipitation (co-IP) of OPN with G_αs in native WT H9c2 cells treated with vehicle (Veh) or 10 µM salbutamol (Salb). Representative blots are shown in (A), and the densitometric quantitation of three independent experiments is shown in (B). IB: immunoblotting; IP: immunoprecipitation; IgG: IP with a general IgG antibody (negative control for the OPN IP). No significant difference (at p = 0.05) in the amount of G_αs co-IP’d with OPN was observed between Veh and Salb (n = 3).

Figure 6

Schematic illustration of the proposed role of OPN in β₂AR ant-fibrotic signaling in H9c2 cardiomyocytes. CA: catecholamine; MR: mineralocorticoid receptor. See text for details and for all other molecular acronym descriptions.