Involvement of Cyclic Guanosine Monophosphate-Dependent Protein Kinase I in Renal Antifibrotic Effects of Serelaxin

Abstract

Introduction: Kidney fibrosis has shown to be ameliorated through the involvement of cyclic guanosine monophosphate (cGMP) and its dependent protein kinase I (cGKI). Serelaxin, the recombinant form of human relaxin-II, increases cGMP levels and has shown beneficial effects on kidney function in acute heart failure patients. Antifibrotic properties of serelaxin are supposed to be mediated via relaxin family peptide receptor 1 and subsequently enhanced nitric oxide/cGMP to inhibit transforming growth factor-β (TGF-β) signaling. This study examines the involvement of cGKI in the antifibrotic signaling of serelaxin.

Methods and results: Kidney fibrosis was induced by unilateral ureteral obstruction in wildtype (WT) and cGKI knock-out (KO) mice. After 7 days, renal antifibrotic effects of serelaxin were assessed. Serelaxin treatment for 7 days significantly increased cGMP in the kidney of WT and cGKI-KO. In WT, renal fibrosis was reduced through decreased accumulation of collagen1A1, total collagen, and fibronectin. The profibrotic connective tissue growth factor as well as myofibroblast differentiation were reduced and matrix metalloproteinases-2 and -9 were positively modulated after treatment. Moreover, Smad2 as well as extracellular signal-regulated kinase 1 (ERK1) phosphorylation were decreased, whereas phosphodiesterase (PDE) 5a phosphorylation was increased. However, these effects were not observed in cGKI-KO.

Conclusion: Antifibrotic renal effects of serelaxin are mediated via cGMP/cGKI to inhibit Smad2- and ERK1-dependent TGF-β signaling and increased PDE5a phosphorylation.

Keywords: Relaxin; cGMP-dependent protein kinase; interstitial fibrosis; kidney; nitric oxide; serelaxin; signaling.

FIGURE 1

Determination of (A) cGMP levels and (B) protein kinase activity of cGKI, determined by VASP phosphorylation (Ser239) in healthy or fibrotic kidney tissue of WT mice 7 days after UUO; mice were untreated or treated with serelaxin for 7 days during UUO. As described in methods increase of renal cGMP levels was quantified with ELISA according to manufacturer’s instructions. VASP phosphorylation at Ser 239 was determined by western blotting (50 μg protein) after normalization to GAPDH; cGMP, cyclic guanosine monophosphate; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; P-VASP, phospho-vasodilator- stimulating phosphoprotein; RLX, serelaxin; UUO, unilateral ureteral obstruction; ^∗∗∗p < 0.001.

FIGURE 2

Effect of serelaxin on (A) mRNA and (B) protein expression of α-SMA in kidney tissue of WT and cGKI-KO mice. mRNA was determined with RT-qPCR and 18s rRNA as housekeeping gene. Results are shown as fold change of mRNA expression (2^ΔΔct) in the fibrotic kidney relative to contralateral healthy kidney, which was set as 1. The increase of α-SMA protein expression [%] in fibrotic renal tissue compared to contralateral kidney was quantified by immunohistochemistry; α-SMA, α-smooth muscle actin; cGKI, cGMP-dependent protein kinase I; KO, knockout; RLX, serelaxin; UUO, unilateral ureteral obstruction; WT, wildtype, ^∗p < 0.05.

FIGURE 3

(A) Immunohistochemical staining of fibronectin in kidney tissue of WT and cGKI-KO mice – healthy or 7 days after UUO treated or untreated with serelaxin; effect of serelaxin on (B) mRNA and (C) protein expression of fibronectin in kidney tissue of WT and cGKI-KO mice; mRNA was determined with RT-qPCR and 18s rRNA as housekeeping gene. Results are shown as fold change of mRNA expression (2^ΔΔct) in the fibrotic kidney relative to contralateral healthy kidney, which was set as 1. The increase of fibronectin protein expression [%] in fibrotic renal tissue compared to contralateral kidney was quantified by immunohistochemistry; cGKI, cGMP-dependent protein kinase I; KO, knock out; RLX, serelaxin; UUO, unilateral ureteral obstruction; WT, wildtype. ^∗∗p < 0.01.

FIGURE 4

Effect of serelaxin on (A) mRNA and (B) protein expression of Col1A1 as well as (C) protein expression of total collagen in kidney tissue of WT and cGKI-KO mice. mRNA was determined with RT-qPCR and 18s rRNA as housekeeping gene. Results are shown as fold change of mRNA expression (2^ΔΔct) in the fibrotic kidney relative to contralateral healthy kidney, which was set as 1. The increase of Col1A1 protein expression [%] in fibrotic renal tissue compared to contralateral kidney was quantified by immunohistochemistry; the increase of total collagen [%] was quantified with sirius red/fast green method; cGKI, cGMP-dependent protein kinase I; Col1A1, collagen1A1; KO, knock out; RLX, serelaxin; UUO, unilateral ureteral obstruction; WT, wildtype. ^∗p < 0.05, ^∗∗p < 0.01.

FIGURE 5

Effect of serelaxin on (C), (E) mRNA and (A), (B), (D), (F) protein expression of MMP-2, MMP-9 determined by zymography (each 70 μg protein) in kidneys from WT and cGKI-KO mice. (A) Representative blot from untreated WT, which illustrates active and latent MMP expression in healthy and fibrotic renal tissue; (B) representative blot from fibrotic WT kidneys untreated or treated with serelaxin; mRNA was determined with RT-qPCR and 18s rRNA as housekeeping gene. Results are shown as fold change of mRNA expression (2^ΔΔct) in the fibrotic kidney relative to contralateral kidney, which was set as 1 [(C) MMP2; (E) MMP9]; latent and active MMP expression determined by zymography (70 μg protein) and is shown only in fibrotic tissue. Each value of WT and cGKI-KO is related to untreated fibrotic WT, which was set as 1 [(D) MMP2; (F) MMP9]; cGKI, cGMP-dependent protein kinase I; KO, knock out; MMP, matrix metalloproteinase; RLX, serelaxin; UUO, unilateral ureteral obstruction gibt es nicht; WT, wildtype, ^∗p < 0.05, ^∗∗∗p < 0.001.

FIGURE 6

(A) Representative western blots (each 50 g protein) of signaling molecules in kidney tissue of WT mice with or without UUO untreated or treated with serelaxin; (B) effect of serelaxin on protein expression of signaling molecules (P-ERK2/ERK2, P-ERK1/ERK1, PDE5a, P-PDE5a, TGF-β, CTGF, P-SMAD2) in fibrotic kidney from WT and cGKI-KO mice; in Figure 5 (B) effects of serelaxin on protein expression of markers (normalized to GAPDH) was determined only in fibrotic tissue. Each value of WT and cGKI-KO is related to untreated fibrotic WT, which was set as 1; α-SMA, α-smooth muscle actin; cGKI, cGMP-dependent protein kinase I; cGMP, cyclic guanosine monophosphate; CTGF, connective tissue growth factor; ECM, extracellular matrix; eNOS, endothelial nitric oxide synthase; ERK, extracellular-signal regulated kinase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GMP, guanosine monophosphate; KO, knock out; MMPs, matrix metalloproteinases; PDE5a, phosphodiesterase 5a; RLX, serelaxin; RXFP1, relaxin receptor; Smad, small mothers against decapentaplegic protein; TGF-β, transforming growth factor-β; UUO, unilateral ureteral obstruction; WT, wildtype, ^∗p < 0.05, ^∗∗p < 0.01.

FIGURE 7

Serum creatinine levels in healthy mice or 7 days after UUO in WT and cGKI-KO mice. Serum creatinine levels were determined via high pressure liquid chromatography as described in methods. cGKI, cGMP-dependent protein kinase I; KO, knock out; RLX, serelaxin; WT, wildtype, ^∗p < 0.05.

FIGURE 8

Proposed hypothesis for in vivo antifibrotic signaling pathway of serelaxin in kidney. Serelaxin mediates its antifibrotic effect via RXFP1/cGMP/cGKI to inhibit TGF-β dependent Smad- and ERK-phosphorylation, which subsequently decreases ECM accumulation and suppresses profibrotic stimuli; α-SMA, α-smooth muscle actin; cGKI, cGMP-dependent protein kinase I; cGMP, cyclic guanosine monophosphate; CTGF, connective tissue growth factor; ECM, extracellular matrix; ERK, extracellular-signal regulated kinase; GMP, guanosine monophosphate; MMPs, matrix metalloproteinases; PDE5a, phosphodiesterase 5a; RXFP1, relaxin family peptide receptor 1; Smad, small mothers against decapentaplegic protein; TGF-β, transforming growth factor-β.