山本 和秀

Serum des-gamma-carboxy prothrombin (DCP) is commonly used to detect hepatocellular carcinoma (HCC). This review focuses on the clinical features of DCP-positive HCC and the molecular function of DCP in HCC. DCP-positive HCC demonstrates more aggressive clinicopathological features than DCP-negative HCC. Analysis of the biological effects of DCP revealed that DCP acts as a growth factor in both an autocrine and paracrine manner. DCP stimulates HCC cell proliferation through the Met-Janus kinase 1-signal transducer and activator of transcription 3 signaling pathway, whereas for vascular endothelial cells, it stimulates cell proliferation and migration through the kinase insert domain receptor-phospholipase C-gamma-mitogen-activated protein kinase signaling pathway.

Anti-idiotype antibodies (Ab2) play an important role in the homeostasis of immune responses and are related to the development and the disease activity of certain autoimmune diseases. The asialoglycoprotein receptor (ASGPR) is considered one of the target antigens in the pathogenesis of autoimmune chronic active hepatitis (AIH). We previously developed a mouse monoclonal antibody (clone 8D7) which recognizes rat and human ASGPR. In this study, to help investigate the anti-ASGPR antibody-anti-idiotype antibody network in patients with AIH, we developed a syngeneic mouse monoclonal Ab2 to the 8D7 anti-ASGPR antibody (Ab1). One clone, designated as 3C8, tested positive for specific reactivity to 8D7-Ab1 and did not bind to other irrelevant immunoglobulins. By competitive inhibition assays, the binding of 8D7-Ab1 to liver membrane extracts, i.e., the crude antigen preparation, was inhibited by 3C8-Ab2 in a dose-dependent manner, and the binding of 8D7-Ab1 to 3C8-Ab2 was inhibited by the liver membrane extracts. In the immunohistochemical analysis, 3C8-Ab2 blocked the specific staining of sinusoidal margins of rat hepatocytes by 8D7-Ab1. These results suggest that 3C8 anti-idiotype antibody recognizes the specific idiotypic determinants within the antigen-binding site of 8D7-Ab1.

In autoimmune chronic active hepatitis (AIH) and primary biliary cirrhosis (PBC), various autoantibodies including anti-asialoglycoprotein receptor (ASGPR) antibodies have been found in patients' sera. We have previously developed a mouse monoclonal antibody against rat and human ASGPR. In this study, we developed a capture enzyme-linked immunosorbent assay (ELISA) for detection of anti-ASGPR antibodies using this monoclonal antibody and investigated the occurrence of anti-ASGPR antibodies in the sera of patients with various liver diseases. Serum samples were obtained from 123 patients with various liver diseases, including 21 patients with AIH and 40 patients with PBC. In this capture ELISA, the target antigen in the crude rat liver membrane extracts was captured on the ELISA wells by the ASGPR-specific mouse monoclonal antibody. Thus, the cumbersome process of antigen purification was rendered unnecessary. Using this capture ELISA, we detected the anti-ASGPR antibody in 67% of the patients with AIH, in 100% of the patients with PBC, and in 57% of the patients with acute hepatitis type A. However, the anti-ASGPR antibody was rarely detected in patients with other liver diseases such as primary sclerosing cholangitis and obstructive jaundice. Our findings suggest that this capture ELISA would be useful for the detection of anti-ASGPR antibodies in autoimmune liver diseases.

We have previously developed an enzyme-linked immunosorbent assay (ELISA) to measure stool decay-accelerating factor (DAF) and found that stool DAF concentrations were significantly elevated in patients with colorectal cancer, suggesting that the measurement of stool DAF may be valuable for the detection of colorectal cancer. In order to refine the assay for the measurement of stool DAF, we investigated 1) effects of centrifugation of stool samples, 2) effects of detergents, and 3) adequate combination of various anti-DAF monoclonal antibodies for the ELISA system using only monoclonal antibodies. We found that high-speed centrifugation could be omitted and that only the removal of large undigested food residues by centrifugation of short duration in a low-speed benchtop microcentrifuge sufficed to adequately prepare the stool samples. Addition of 2 detergents, octyl beta-glucoside and sodium deoxycholate, known to solubilize glycosyl-phosphatidylinositol-anchored proteins such as DAF, did not influence stool DAF values. By using 2 mouse anti-DAF monoclonal antibodies (clone 4F11 and 1C6), we were able to achieve a stable ELISA for the measurement of stool DAF using a uniform source of antibodies. The results should allow us to consistently apply the DAF assay for routine use in the detection of colorectal cancer.

The preS2 region of the hepatitis B virus (HBV) has been reported to have human polymerized albumin receptor (PAR) activity, which correlates with viral replication. Here, we studied the genomic sequence of the preS region from rare patients lacking PAR activity, despite active viral replication. PAR and DNA polymerase activity was identified in 178 HBe antigen-positive HBV carriers, and a significant correlation between 2 markers was shown, except in 2 hepatitis patients lacking PAR activity. Nucleotide sequences of the preS region of HBV from both patients were examined by direct sequencing of PCR products. In one patient, a 45-base deletion was found to overlap half of the putative polymerized human albumin binding site in the preS2 region. In the other patient, a point mutation at the first nucleotide of the start codon of the preS2 region of HBV was found. There was no such genomic change in the 3 control HBV sequences. These results indicate that the preS2 region is necessary for binding of polymerized human albumin, and this is the first report of naturally existing mutant virus with no or low PAR activity.

We developed a monoclonal antibody (MoAb) (clone 5E8) against an antigen on the bile canalicular membrane of rat hepatocyte. By immunoblotting, MoAb 5E8 detected a band of 110 kD. In this study, we used the phage display technique to identify the target antigen recognized by MoAb 5E8. We screened a random phage display library expressing 12-mer peptide sequences and identified a peptide sequence, FHFNPYTGHPLT, as an epitope. We compared this peptide sequence with those of dipeptidyl peptidase IV (DPP IV, E.C.3.4.14.5) and Cell-CAM105, which proteins were located by a database search based on the information of tissue localization and approximate molecular weight of the MoAb 5E8 antigen, and sequence similarity with a region in DPP IV (amino acids 225-233) but not with Cell-CAM105 was found. In addition, we immunohistochemically stained various tissues (liver, small intestine, and kidney) of Japanese Fischer 344 rats, known to be deficient for DPP IV, with MoAb 5E8 and showed that the expression of MoAb 5E8 antigen was negligible or weak. In contrast, tissues sampled from the same organs of Sprague-Dawley rats, known to express DPP IV, were positively stained. These findings suggest that the antigen recognized by MoAb 5E8 is DDPIV and its major epitope is located in amino acids at positions 225-233.

The three-dimensional arrangement of ductular structures formed by oval cells in rats fed 2-acetylaminofluorene (2-AAF) was studied by scanning electron microscopy (SEM) of biliary tract casts and light microscopy of sections of liver injected with india ink via the biliary tract. Both resin and india ink were well injected up to bile ductules, and the findings of each method correlated with each other. By the second week after 2-AAF administration, a few oval cells appeared in the periportal areas forming ductular structures which connected with the portal bile ducts. At the 4th week, increased ductular structures occupied two thirds of the lobule and formed networks communicating with each other, and with the portal bile ducts. At the 8th week, such ductular structures were compressed around hyperplastic nodules and appeared like a basket in biliary casts examined by SEM. Although a histochemical study of gamma-glutamyl transpeptidase revealed activity both on the luminal side of the ductular structures and hepatocytes in hyperplastic nodules, no transition was observed between these two cell populations. These results suggest that oval cells have characteristics more similar to those of biliary epithelia than of hepatocytes, and have no relation to the development of hyperplastic nodules.

Phalloidin, a toxin from the plant Amanita phalloides, irreversibly polymerizes actin filaments and causes cholestasis. Three-dimensional structural changes induced by phalloidin in the bile canaliculi and the intra-acinar localization of these changes were studied in the rat liver by scanning and transmission electron microscopy. After 3 days of treatment, canalicular changes appeared mainly in zones 2 and 3 of Rappaport's acinus, but after 7 days of treatment changes occurred in bile canaliculi of the whole acinus. The changes in the bile canaliculi included tortuosity, saccular dilatation, loss of microvilli, bleb formation and elongation of canalicular side branches. Some side branches extended near to Disse's space, leaving only a thin cytoplasmic rim between the canalicular lumen and Disse's space. Kupffer cells were occasionally situated near such extended bile canaliculi and protruded their processes into the hepatic cord. These results suggest that bile canaliculi in zone 3 are more susceptible to phalloidin toxicity than those in zone 1 and that biliary constituents may leak from such altered bile canaliculi.

Autoimmune hepatitis (AIH) is a chronic and progressive disease characterized by histological interface hepatitis, hypergammaglobulinemia, and circulating autoantibodies. Multiple factors, including molecular mimicry, a genetic background including major histocompatibility complex class II, and defective function of regulatory T-cells, are involved in the pathogenesis. The diagnosis is made based on the scoring system of the International Autoimmune Hepatitis Group, the sensitivity and specificity of which are90%, respectively. AIH is classified into 3 sub-types based on the profiles of circulating autoantibodies: anti-nuclear antibody and/or smooth muscle antibody-positive (type 1), anti-liver-kidney microsomal antibody-positive (type 2), and anti-soluble liver antigen/liver-pancreas antigen antibody- positive (type 3). Recently, however, the number of atypical cases lacking the usual features has increased-for example, patients with acute-onset or fulminant-type AIH, autoantibody-negative patients, male patients, and patients with bile duct injury-and thus the clinical features of AIH have been diversified. AIH is responsive to immunosuppressive treatment in most cases; however, relapse occurs in more than 80% of patients within 1 year after immunosuppressive treatment withdrawal. The 10-year survival rate and the 10-year hepatocellular carcinoma-free rate are90%, respectively, indicating that some patients reach liver failure or develop hepatocellular carcinoma. To improve the prognosis of these patients, persistent normalization of transaminase is required.

To characterize primary sclerosing cholangitis (PSC) in Japanese patients and its association with inflammatory bowel disease (IBD), 155 reported cases of PSC, including 6 cases of our own, were reviewed. The prevalence of IBD was less in Japanese PSC patients than in Western patients (23% versus 62-100%). Japanese PSC patients with IBD were younger (mean age, 33.1 versus 51.8 years) and were more often women (51% versus 36%) than those without IBD. Seventy-four percent of PSC patients with IBD had extensive colonic lesions, and 89% of those developed IBD simultaneously, with or prior to PSC. There were 3 cases of neutrophilic cholangitis among the PSC patients with IBD but none in those without IBD. Based on these observations, we speculate that there may be subtypes of PSC which differ pathophysiologically.