Hans Sägesser

Pharmacological Inhibition of Integrin αvβ3 Aggravates Experimental Liver Fibrosis and Suppresses Hepatic Angiogenesis

The vitronectin receptor integrin αvβ3 promotes angiogenesis by mediating migration and proliferation of endothelial cells, but also drives fibrogenic activation of hepatic stellate cells (HSCs) in vitro. Expecting antifibrotic synergism, we studied the effect of αvβ3 inhibition in two in vivo models of liver fibrogenesis. Liver fibrosis was induced in rats by way of bile duct ligation (BDL) for 6 weeks or thioacetamide (TAA) injections for 12 weeks. A specific αvβ3 (αvβ5) inhibitor (Cilengitide) was given intraperitoneally twice daily at 15 mg/kg during BDL or after TAA administration. Liver collagen was determined as hydroxyproline, and gene expression was quantified by way of quantitative polymerase chain reaction. Liver angiogenesis, macrophage infiltration, and hypoxia were assessed by way of CD31, CD68 and hypoxia‐inducible factor‐1α immunostaining. Cilengitide decreased overall vessel formation. This was significant in portal areas of BDL and septal areas of TAA fibrotic rats and was associated with a significant increase of liver collagen by 31% (BDL) and 27% (TAA), and up‐regulation of profibrogenic genes and matrix metalloproteinase‐13. Treatment increased gamma glutamyl transpeptidase in both models, while other serum markers remained unchanged. αvβ3 inhibition resulted in mild liver hypoxia, as evidenced by up‐regulation of hypoxia‐inducible genes. Liver infiltration by macrophages/Kupffer cells was not affected, although increases in tumor necrosis factor α, interleukin‐18, and cyclooxygenase‐2 messenger RNA indicated modest macrophage activation. Conclusion: Specific inhibition of integrin αvβ3 (αvβ5) in vivo decreased angiogenesis but worsened biliary (BDL) and septal (TAA) fibrosis, despite its antifibrogenic effect on HSCs in vitro. Angiogenesis inhibitors should be used with caution in patients with hepatic fibrosis. (HEPATOLOGY 2009.)

Reversibility of secondary biliary fibrosis by biliodigestive anastomosis in the rat

Biliary cirrhosis with portal hypertension and hepatocellular failure is a well-known complication of extrahepatic obstruction. It is unclear to what extent these changes are reversible by biliodigestive anastomosis. Therefore a rat model of relief of biliary obstruction was developed by performing Roux-en-Y choledochojejunostomy in rats after bile duct obstruction. Patency of the biliodigestive anastomosis was documented by biliary scintigraphy. Microsomal function was assessed in vivo by the aminopyrine breath test and portal hypertension by spleen pulp pressure. Microsomal function was markedly impaired in obstructed animals but recovered after biliodigestive anastomosis. Microsomal cytochrome P450 content paralleled these changes. Similarly, portal hypertension was reversed after successful relief of obstruction. Stereologic analysis showed that biliodigestive anastomosis partially reversed bile ductular proliferation and fibrosis. Studying the time course of recovery showed that restoration of microsomal function was achieved after 2 weeks whereas recovery from portal hypertension required 4 weeks of biliary drainage. Recovery of microsomal function was paralleled by normalization of microsomal lipid composition while resolution of portal hypertension occurred parallel to resolution of the histologic abnormalities.

The effect of endothelin and its antagonist Bosentan on hemodynamics and microvascular exchange in cirrhotic rat liver

BACKGROUND/AIMS To characterize the effects of endothelin-1 and of Bosentan, a mixed endothelin antagonist, on hepatic hemodynamics in cirrhotic animals in vivo and on hepatic microvascular exchange in the perfused rat liver. METHODS Biliary cirrhosis was induced by bile duct ligation, and micronodular cirrhosis by chronic exposure to phenobarbital/CCl4 in male rats. Hepatic hemodynamics were studied under basal conditions and after administration of Bosentan (3-30 mg/kg) by the microsphere technique. Microvascular exchange was assessed in the in situ perfused rat liver by the multiple indicator dilution technique. RESULTS Bosentan lowered portal pressure in a dose-dependent fashion; at the highest dose tested, this decrease averaged -29+/-11 and -26+/-8% in biliary and micronodular cirrhosis, respectively (p<0.01). This was achieved mainly via a marked decrease in hepatic arterial flow. In the perfused liver, endothelin-1 induced a dose-dependent vasoconstriction; up to 10(-9) mol/l; this was not associated with any effect on viability. At this dose, endothelin-1 markedly decreased extravascular albumin space in both controls and micronodular cirrhosis; this could be antagonized by Bosentan 10(-5) mol/l. CONCLUSIONS Endothelin-1 affects hepatic microvascular exchange, presumably by a direct effect on hepatic sinusoidal endothelial cells. A mixed endothelin antagonist lowers portal pressure in vivo, presumably by acting on hepatic stellate cells, and counteracts the microvascular effects of endothelin-1 in vitro. These properties could prove useful in treatment of portal hypertension.

Overexpression of endothelin-1 in bile duct ligated rats: correlation with activation of hepatic stellate cells and portal pressure

BACKGROUND/AIMS Hepatic stellate cells (HSC) are involved in the pathogenesis of liver fibrosis; although ET-1 is increased in cirrhosis, its pathophysiological role in fibrogenesis and portal hypertension remains controversial. The aim of this study was to investigate splanchnic hemodynamics and to correlate them with changes in ET-1 expression and HSC activation in bile duct ligated (BDL) rats. METHODS/RESULTS Expression of the ET-1 gene was increased early as measured by quantitative reverse transcriptase-polymerase chain reaction (6-fold 3 days after BDL) whereas ET-1 peptide measured by RIA increased significantly only in the late phase (30-fold at 28 days). There was a linear correlation between portal pressure and the amount of ET-1 in the portal vein (r = 0.66; P = 0.003), as well as between ET-1 and the volume fraction of myofibroblasts (r = 0.80, P < 10(-7)) as assessed by morphometry and immunohistochemical staining using alpha-smooth muscle actin. CONCLUSIONS During chronic liver injury activation of HSCs and of preproET-1 mRNA is accentuated in the early phase after BDL. The late increase in ET-1 peptide may indicate that this peptide is only secondarily involved in HSC activation. The correlation between ET-1 in portal vein and portal pressure suggests that ET-1 may play an important role in the development of portal hypertension.

Toxicity of Valproic Acid in Mice with Decreased Plasma and Tissue Carnitine Stores

The aim of this study was to investigate whether a decrease in carnitine body stores is a risk factor for valproic acid (VPA)-associated hepatotoxicity and to explore the effects of VPA on carnitine homeostasis in mice with decreased carnitine body stores. Therefore, heterozygous juvenile visceral steatosis (jvs)+/– mice, an animal model with decreased carnitine stores caused by impaired renal reabsorption of carnitine, and the corresponding wild-type mice were treated with subtoxic oral doses of VPA (0.1 g/g b.wt./day) for 2 weeks. In jvs+/– mice, but not in wild-type mice, treatment with VPA was associated with the increased plasma activity of aspartate aminotransferase and alkaline phosphatase. Furthermore, jvs+/– mice revealed reduced palmitate metabolism assessed in vivo and microvesicular steatosis of the liver. The creatine kinase activity was not affected by treatment with VPA. In liver mitochondria isolated from mice that were treated with VPA, oxidative metabolism of l-glutamate, succinate, and palmitate, as well as β-oxidation of palmitate, were decreased compared to vehicle-treated wild-type mice or jvs+/– mice. In comparison to vehicle-treated wild-type mice, vehicle-treated jvs+/– mice had decreased carnitine plasma and tissue levels. Treatment with VPA was associated with an additional decrease in carnitine plasma (wild-type mice and jvs+/– mice) and tissue levels (jvs+/– mice) and a shift of the carnitine pools toward short-chain acylcarnitines. We conclude that jvs+/– mice reveal a more accentuated hepatic toxicity by VPA than the corresponding wild-type mice. Therefore, decreased carnitine body stores can be regarded as a risk factor for hepatotoxicity associated with VPA.

Characterisation of portal hypertension models by microspheres in anaesthetised rats: a comparison of liver flow.

Several portal hypertensive animal models are available and frequently used for haemodynamic studies. The portal venous inflows, measured with microspheres in pentobarbital anaesthetised rats, are compared here. The partial portal vein ligation model is characterised by a high portal venous inflow, together with extensive portal systemic shunting, at the cost of portal sinusoidal flow. In carbon tetrachloride-induced micronodular cirrhosis, portal sinusoidal flow, which reaches liver parenchyma, is high, and this is more pronounced in the presence of ascites. In bile duct ligation and excision-induced cirrhosis, an increase in liver weight was not equally followed by an increase in portal sinusoidal flow, pointing to a relatively underperfused liver.

Nitric oxide synthase 1 is partly compensating for nitric oxide synthase 3 deficiency in nitric oxide synthase 3 knock-out mice and is elevated in murine and human cirrhosis.

Abstract: 
Background: The role of endothelial nitric oxide synthase 3 (NOS‐3) in the hyperdynamic circulation associated with cirrhosis is established but not that of the neuronal (NOS‐1) isoform. We therefore investigated aortic NOS‐1 levels in NOS‐3 knock‐out (KO) and wildtype (WT) mice and in hepatic arteries of patients.

Type A, but not type B, endothelin receptor antagonists significantly decrease portal pressure in portal hypertensive rats

BACKGROUND/AIM Endothelin-1 plays an important role in the regulation of portal hypertension; endothelin antagonists have been extensively studied in portal hypertensive animals. We aimed to evaluate the efficacy of highly selective endothelin antagonists in partial portal vein ligated (PPVL) rats. METHODS Four groups of 7 male Sprague-Dawley rats were administered orally ABT-627 (ET(A)-selective), A-192621 (ET(B)-selective), or A-182086 (non-selective), with the fourth group serving as control. On the 3rd day after beginning treatment animals underwent PPVL. On the 11th day hemodynamics were studied and portal vein ET-1 was measured. RESULTS In the control group portal pressure was 13.4+/-SD 0.2 mmHg; this increased to 14.9+/-1.8 (p<0.05) in the ET(B) blocked group. In contrast, ET(A) blockade improved portal hypertension (11.7+/-1.1, p<0.05), while the treatment with the non-selective antagonist had no effect (12.3+/-0.7 n.s.). Mean arterial pressure was not significantly affected by any treatment. Portal vein ET-1 was increased in all groups compared to controls; this increase was limited to the pre-stenotic area (79+/-43 vs 194+/-76 in the pre- and post-stenotic portal vein; p<0.0025). CONCLUSIONS Oral administration of an ET(A) antagonist ameliorated portal hypertension; we suggest that long-term therapy of portal hypertension with selective ET(A) antagonists may be more beneficial than mixed antagonists.

Intraportal administration of glyceryl trinitrate or nitroprusside exerts more systemic than intrahepatic effects in anaesthetised cirrhotic rats

BACKGROUND/AIMS Increased intrahepatic vascular tone can be pharmacologically manipulated in isolated cirrhotic livers. Intrahepatic endothelial dysfunction may lead to a decreased production of the potent endogenous vasodilator nitric oxide in cirrhotic livers. The aims of the study were to determine whether portal pressure can be lowered in vivo by injecting nitric oxide donors glyceryl trinitrate or nitroprusside directly in the portal vein and whether this is related to a decrease in intrahepatic resistance. METHODS In anaesthetised CCl4 cirrhotic rats, intraportal doses of glyceryl trinitrate 0.5, 1 or 5 microg/kg/ min or nitroprusside 1, 5 or 10 microg/kg/min did not decrease portal pressure but only arterial pressure. Systemic and splanchnic haemodynamics were measured before and during 15 min intraportal infusion of glyceryl trinitrate 10 microg/kg/min or nitroprusside 20 microg/kg/min. RESULTS Glyceryl trinitrate decreased portal pressure from 14.0+/-1.1 to 11.8+/-1.4 mm Hg, splanchnic perfusion pressure from 102+/-10 to 74+/-5 mm Hg and portal sinusoidal flow from 2.11+/-0.38 to 1.70+/-0.35 ml/min/g liver (all p<0.05). Nitroprusside did not decrease portal pressure significantly but led to a reduction of the splanchnic perfusion pressure (104+/-9 to 66+/-7 mm Hg) and the portal sinusoidal flow (2.39+/-0.50 to 1.77+/-0.31 ml/min/g liver; all p<0.05). Portal sinusoidal resistance was not altered by either drug. CONCLUSIONS Intraportal infusion of nitric oxide donors decreased arterial pressure more than portal pressure. Portal sinusoidal resistance remained unaffected, but the liver parenchyma became less perfused with high doses. The systemic effects of nitric oxide donating drugs prevailed.

Attenuated portal hypertension in germ‐free mice: Function of bacterial flora on the development of mesenteric lymphatic and blood vessels

Intestinal bacterial flora may induce splanchnic hemodynamic and histological alterations that are associated with portal hypertension (PH). We hypothesized that experimental PH would be attenuated in the complete absence of intestinal bacteria. We induced prehepatic PH by partial portal vein ligation (PPVL) in germ‐free (GF) or mice colonized with altered Schaedlers flora (ASF). After 2 or 7 days, we performed hemodynamic measurements, including portal pressure (PP) and portosystemic shunts (PSS), and collected tissues for histomorphology, microbiology, and gene expression studies. Mice colonized with intestinal microbiota presented significantly higher PP levels after PPVL, compared to GF, mice. Presence of bacterial flora was also associated with significantly increased PSS and spleen weight. However, there were no hemodynamic differences between sham‐operated mice in the presence or absence of intestinal flora. Bacterial translocation to the spleen was demonstrated 2 days, but not 7 days, after PPVL. Intestinal lymphatic and blood vessels were more abundant in colonized and in portal hypertensive mice, as compared to GF and sham‐operated mice. Expression of the intestinal antimicrobial peptide, angiogenin‐4, was suppressed in GF mice, but increased significantly after PPVL, whereas other angiogenic factors remained unchanged. Moreover, colonization of GF mice with ASF 2 days after PPVL led to a significant increase in intestinal blood vessels, compared to controls. The relative increase in PP after PPVL in ASF and specific pathogen‐free mice was not significantly different. Conclusion: In the complete absence of gut microbial flora PP is normal, but experimental PH is significantly attenuated. Intestinal mucosal lymphatic and blood vessels induced by bacterial colonization may contribute to development of PH. (Hepatology 2015;61:1685‐1695)