Hepatic stellate cells (Ito-cells) but not collagen IV may partly be responsible for lower portal pressure after reversing secondary biliary cirrhosis in the rat

Abstract

Background/Aims: Chronic bile duct obstruction in the rat leads to biliary cirrhosis and portal hypertension. Biliary decompression with Roux-en-Y choledocho-jejunostomy (RY) reverses most but not all of these changes. The aim of the present study was to determine whether hepatic stellate cells, as a main source of extracellular matrix proteins, participate in this process. Methods: Sprague-Dawley rats were allocated to one of three groups: Bile duct ligation (BDL) for 3 weeks, BDL followed by RY and sham-operated animals as controls (SHAM). At the end of the experimental period, portal pressure was measured, livers subjected to random sampling and hepatocytes, bile ducts/ductules, hepatic stellate cells and collagen IV determined stereologically. Hepatic stellate cells and collagen IV were characterized immunohistochemically with an antibody against desmin and collagen IV, respectively. Results: Volume fraction of hepatocytes decreased from 65.6±5.3 in sham-operated animals to 27.9±8.8% in bile duct ligated animals ( p<0.05). In contrast, volume fraction of bile ducts/ductules increased from 0.4±0.2 in sham-operated animals to 25.3±8.6% in bile duct ligated ones; similarly, hepatic stellate cells increased from 0.4±0.2 in sham-operated animals to 2.6±0.9% in bile duct ligated ones ( p<0.01) and collagen IV from 10.0±2.3 in sham-operated animals to 24.5±8.0% ( p<0.01) in bile duct ligated animals. These changes were partially reversed by Roux-en-Y choledocho-jejunostomy; hepatocytes, bile ducts/ductules, hepatic stellate cells and collagen IV averaging 54.8±13.1, 6.1±6.8, 1.6±0.6 and 14.5±3.6%, respectively ( p<0.05 RY vs. BDL). Portal pressure in sham-operated animals, bile duct ligated animals and those with a Roux-en-Y choledocho-jejunostomy averaged 13.4±0.7, 20.1±2.7 and 16.9±1.6 cm H 2O, respectively, and correlated significantly with the volume fraction of hepatic stellate cells (r S=0.96; p<0.001) and less with collagen IV (r s=0.61; p<0.007). However, by stepwise regression, collagen IV did not significantly add to the ability of the equation to predict portal pressure. Conclusions: These results lend further support to the notion that hepatic stellate cells are prominently involved in fibrogenesis and in the reversibility of these changes, but hepatic stellate cells do not completely revert to normal even 4 weeks after successful decompression. Furthermore, our data suggest that hepatic stellate cells may be related to maintenance of portal hypertension.