3. Results
Both the infected and control groups of hamsters were healthy throughout the experiment and no hamsters died. Liver fluke eggs were first observed in the faeces one month after infection [11].
3.1. Histological changes
The biliary system consists of two main parts: the intrahepatic and extrahepatic bile ducts including the gall bladder. In this study, the intrahepatic bile ducts of the hamster were classified as the first and second order bile ducts. The second order bile ducts were large bile ducts and their main branches, starting from the hepatic hilus, are characterised by the presence of a fibrous ductal wall and lined by low columnar to columnar epithelium. They corresponded to the left or right hepatic ducts, segmental ducts and area ducts in human [16] and [17]. The first order bile ducts were small bile ducts corresponded to the septal and interlobular bile ducts, and bile ductules. Normally, the adult liver flukes inhabited the second order bile ducts and were not found in the first order bile ducts.
No obvious pathological changes were observed in any specimens in the control group. In infected hamsters, the liver changes were similar to those previously described [4]. A distinct inflammatory response was observed around the juvenile flukes in the intrahepatic second order bile duct at the hepatic hilus and extrahepatic bile duct by day 3 p.i. Mild inflammation without flukes was also seen in the periportal areas and first order bile ducts. The cell types mainly comprised mononuclear cells and eosinophils. The severity of the inflammation gradually increased from day 7 p.i. onwards. Heavy mononuclear cell and eosinophil infiltration around the intrahepatic bile ducts was noted in the areas with epithelial desquamation. Focal liver cell necrosis was also observed early in the infection. Maximal inflammation was established by 1 month p.i.
From the second month onwards, active inflammatory changes around the bile ducts (both the first and second order bile ducts) regressed slightly and lymphoid follicles, as well as plasma cell infiltration, were predominant. Active germinal centers were seen in most large lymphoid follicles of the liver, extrahepatic bile duct and gall bladder. The most obvious histological changes in the liver in chronic infection were periductal fibrosis, ductal dilatation and mononuclear cell infiltration with lymphoid aggregation. The pathological changes were more severe in the liver and extrahepatic bile duct than that in the gall bladder. Granulomatous inflammation around fluke eggs was frequently seen in the intrahepatic periductal areas and protruding from the mucosal infolding within the bile duct lumen.
3.2. Immunohistochemical localisation of Opisthorchis antigens
A positive reaction to the immunoperoxidase staining for Opisthorchis antigens is recognised by reddish-brown deposits. To compare, immunofluorescent staining produced a bright fluorescence. In both methods, the more intense the reaction, the greater the amount of antigens. Both antiserum preparations (i.e. anti-ES and anti-somatic antibodies) produced strongly positive staining in the tegument, muscle, digestive tract and reproductive organs including eggs of the liver fluke. The anti-ES antibody exhibited less activity for the parasite eggs than that of the anti-somatic. The specificity of the reaction was confirmed by comparing the staining of the adjacent sections treated with either the rabbit anti-Opisthorchis antibody alone, the antibody after absorption with O. viverrini antigens or normal rabbit serum. The purified antibody to ES antigens cross-reacted with the host vascular smooth muscle. The sensitivity of both staining techniques for localisation of Opisthorchis antigens were similar. Immunoperoxidase staining was therefore used in most of the specimens because of its convenience, permanence and excellent morphology.
Opisthorchis antigens were not detected in any of the uninfected control specimens. Positive staining was seen in all stages of the parasite in the infected tissue sections. Extraparasitic staining was found in the mucus coat and on the apical surface of the biliary epithelium lining in close contact with the flukes as early as day 3 p.i. Obvious staining of parasites and biliary epithelium was observed by day 7 p.i.. On day 14 p.i., when the parasites were growing, parasite antigens were observed diffusely distributed in the flukes, biliary epithelium and surrounding stromal tissue near the flukes.
The presence of parasite antigen-positive biliary epithelium was associated with heavy mononuclear cell infiltration. The staining of biliary epithelium was either homogeneous or heterogeneous. Similar patterns of parasite antigen staining, which were slightly increased in intensity, were seen from day 30 p.i. until the end of the experiment. Interestingly, intense staining was observed in the epithelium and periductal fibrotic areas in close contact with the flukes of chronically infected specimens. The gall bladder epithelium contained smaller amounts of antigens on the mucosal coat than on the epithelia of the intrahepatic and extrahepatic bile ducts. Representative photomicrographs of the sequential immunohistochemical localisation of Opisthorchis antigens in the infected biliary system are shown in Fig. 1A–F.