2017•Zeitschrift für GastroenterologieRequires access
Molecular effects of the soluble guanylyl cyclase stimulator Riociguat in experimental cirrhosis and impact on transaminases in patients
Philipp Schwabl, Ksenia Brusilovskaya, Philipp Königshofer, Paul Supper, David Bauer, Florian Riedl, Nataliya Rohr‐Udilova, Hubert Hayden, Stefan Aschauer, Diana Bonderman, Georg Oberhuber, Bruno K. Podesser, Markus Peck‐Radosavljevic, Michael Trauner, Thomas Reiberger
Open publisher page 0 citations
Abstract
Background: Nitric oxide signaling including the downstream effector soluble guanylyl cyclase is impaired in cirrhosis. We previously demonstrated beneficial effects of the guanylyl cyclase stimulator riociguat (RIO) on liver fibrosis and portal hypertension. Now we aimed to assess (i) molecular mechanisms of RIO and (ii) effects on transaminases in humans. Methods: Cirrhosis was induced by bile duct ligation (BDL) in rats for 3 (W3, fibrosis) or 5 (W5, cirrhosis) weeks. Controls underwent sham operation. RIO (1 mg/kg/d) or vehicle was administered for the last 14 days. Liver specimens were stained for CK19 (ductular proliferation) and CD68 (macrophage infiltration). Intrahepatic markers of vascular dysfunction (eNOS, Myosin, Moesin), angiogenesis (CD31, VEGFR2) and inflammation (TNFa) were determined by western blotting (normalized to GAPDH; expressed as x-fold of control). Moreover, the course of transaminases was assessed in 31 patients with heart failure (HFpEF) after initiation of RIO therapy. Results: BDL-W3 rats showed significantly increased CK19 area, phosphorylated (p)- moesin, myosin and VEGFR2, while inflammatory marker remained unchanged. In the stage of fibrosis, RIO inhibited ductular proliferation (CK19: 6.14 ± 1.85 vs. 3.58 ± 0.81%; p = 0.005), decreased sinusoidal vasoconstriction (p-moesin: 22.8 ± 0.93 vs. 1.8 ± 0.45; p < 0.001; myosin: 1.78 ± 0.08 vs. 0.93 ± 0.51; p = 0.036) and reduced angiogenesis (VEGFR2: 2.41 ± 0.07 vs. 1.57 ± 0.50; p = 0.044). BDL-W5 cirrhotic rats presented with increased macrophage infiltration and TNFa levels, which were ameliorated by RIO treatment (CD68 area: 4.21 ± 1.33% vs. 2.87 ± 1.09%, p = 0.032; TNFa: 1.68 ± 0.16 vs.1.12 ± 0.09; p = 0.006). Additionally, RIO promoted sinusoidal vasodilation (p-eNOS: 1.14 ± 0.21 vs. 1.51 ± 0.02; p = 0.041) in cirrhotic BDL-W5 animals. In HFpEF patients, RIO significantly decreased AST (27.6 ± 1.6 vs. 23.3 ± 1.2U/L; p = 0.001) and ALT (23.6 ± 1.5 vs. 21.3 ± 1.4U/L; p = 0.046) levels after one month of treatment. After RIO treatment discontinuation, AST/ALT levels returned to baseline. Conclusions: RIO decreased ductular proliferation and improved vascular dysfunction in rats with fibrosis. In cirrhosis, RIO reduced macrophage infiltration and inflammation. In patients, RIO decreased levels of transmainases.