Anti-hepatocyte growth factor antibody inhibits hepatocyte proliferation during liver regeneration.

In vitro studies have shown hepatocyte growth factor (HGF) to be a potent mitogen for hepatocytes. Direct evidence of a mitogenic role in vivo was sought by inhibiting HGF activity, using continuous administration of neutralizing antibody to rats which had a stimulus for liver regeneration. Alzet osmotic mini-pumps, administering a constant supply of anti-HGF monoclonal antibody (clone D9), were inserted intraperitoneally into male Wistar rats; an irrelevant isotypical antibody was administered to controls. Forty-five animals received an intragastric bolus of 40 per cent carbon tetrachloride (CCl4) and groups of three test and control animals were killed at 24 h intervals for 7 days. Treatment with anti-HGF monoclonal antibody significantly inhibited the levels of immunodetectable HGF in the sera of rats following CCl4 administration. In comparison with controls, hepatocyte proliferation as assessed by bromodeoxyuridine labelling in anti-HGF-treated animals was significantly inhibited at 24 h (P < 0.001), 48 h (P < 0.001), and 96 h (P < 0.05) post-CCl4 administration. In contrast, sinusoidal cell proliferation was not significantly different from controls at any time point. Inhibition of the parenchymal proliferative response to acute CCl4-induced liver injury by the in vivo neutralization of HGF provides direct evidence that this growth factor plays an important role in liver regeneration following necrosis.

Macrophage and hepatic stellate cell responses during experimental hepatocarcinogenesis.

The aim of the study was to assess the monocyte/macrophage and hepatic stellate cell responses during experimental diethylnitrosamine (DEN)-induced hepatocarcinogenesis. Diethylnitrosamine (50mg/L) was administered to 39 rats for 10 weeks; liver tissue was obtained at weeks 10, 16 and 19. In this model, necroinflammatory damage occurs during the period of DEN administration but thereafter subsides; dysplastic nodules and carcinomas subsequently develop. Monocytes/ macrophages were detected immunohistochemically using ED1 and ED2 monoclonal antibodies; hepatic stellate cells (HSC) were detected using antibodies to alpha-smooth muscle actin (alpha-SMA) (activated HSC) and glial fibrillary acidic protein (GFAP). Parenchymal ED1- and ED2-positive monocytes/macrophages and alpha-SMA-positive HSC increased at week 10 when there was ongoing DEN-induced necroinflammatory activity. ED1- and ED2-positive cells were also prominent at weeks 16 and 19, particularly around the periphery of dysplastic and carcinomatous nodules, with occasional macrophages between dysplastic hepatocytes. alpha-SMA-positive HSC were present within sinusoids between dysplastic cells and were more abundant at weeks 16 and 19 than in control or week 10 animals. Activated HSC were prominent in fibrous septa around and within dysplastic and carcinomatous nodules at weeks 16 and 19. In contrast, GFAP-positive HSC did not accumulate in developing septa or within dysplastic and carcinomatous nodules. We have demonstrated changes in the monocyte/ macrophage and HSC populations during the development of hepatocellular dysplasia and carcinoma at time points when there is little necroinflammatory activity; this may therefore represent a host response to hepatocyte dysplasia. The HSC activation may be mediated, in part, by monocyte/ macrophage-derived factors, but we speculate that it may also result from direct stimulation by factors released from dysplastic hepatocytes.

Hepatocyte growth factor levels in liver and serum increase during chemical hepatocarcinogenesis.

Hepatocyte growth factor (HGF) is mitogenic for hepatocytes and some tumor cell lines. Elevations in plasma HGF levels have been detected in patients with hepatocellular carcinoma (HCC), and it is possible that HGF is involved in the promotion and/or progression of tumor growth. We measured serum and liver tissue HGF levels during chemically induced hepatocarcinogenesis. Wistar rats were given diethylnitrosamine (DEN) in drinking water for 10 weeks with controls receiving drinking water only. Animals were killed at 10, 16, and 19 weeks. Liver HGF levels were determined from immunoblotted protein by scanning densitometry, and serum HGF levels were measured by sandwich enzyme-linked immunosorbent assay (ELISA). HGF was also immunolocalized in fixed liver tissue sections. In DEN-treated animals, at 10 weeks, there was necroinflammation but no dysplasia. Serum HGF was elevated compared with controls (P < .001) but there was no increase in liver HGF. At 16 weeks, there was liver cell dysplasia with minimal necroinflammation; serum and tissue HGF levels were both significantly elevated above controls. At 19 weeks, hepatocellular carcinomas (HCC) were present in five of six DEN-treated animals; liver HGF (P < .05) and serum HGF (P < .001) were both elevated compared with controls. HGF was localized in basement membranes around bile ducts and vessels and some perisinusoidal cells. Increased HGF immunolabeling was observed at 16 and 19 weeks, but dysplastic hepatocytes and tumor cells were HGF-negative. HGF may serve as a growth promoter at early stages during liver tumor development acting through possibly endocrine and paracrine pathways. Recent observations have described HGF as being mitoinhibitory for HCC cell lines; it is possible therefore that the continued up-regulation of HGF in the latter stages of our DEN model may inhibit tumor cell growth, and thus represent a form of antitumor host response.

Transforming growth factor-alpha expression is altered during experimental hepatocarcinogenesis.

In order to characterize the role of transforming growth factor-alpha (TGF alpha) during hepatocarcinogenesis, liver tissue was examined at 10, 16, and 19 weeks following initial 10-week diethylnitrosamine (50 mg l-1 drinking water) exposure in female Wistar rats. Liver tissue protein extracts were electrophoresed and transferred to nitrocellulose filters. Levels of tissue-derived TGF alpha and epidermal growth factor receptor (EGFr) were assessed using an anti-TGF alpha monoclonal antibody (Ab-1) and an anti-EGFr polyclonal antibody (AB-4), coupled with scanning densitometric quantification. Immunolocalization of TGF alpha was performed in Bouin's-fixed, paraffin-embedded liver tissue sections. The distribution and intensity of TGF alpha immunoreactivity varied according to the degree of dysplasia, severely dysplastic cells being strongly immunoreactive. At week 10, mild hepatocyte dysplasia and perivenular inflammation were evident, together with a corresponding increase in perivenular TGF alpha immunoreactivity. By week 16, foci of moderate to severe dysplasia were observed; at this stage, there was a decrease in perivenular immunoreactivity but a further increase in overall liver tissue TGF alpha levels. Some 'altered foci' and dysplastic nodules showed intense immunoreactivity for TGF alpha. At these time points, immunodetectable liver EGFr was found to decrease significantly in comparison with normal control tissue. TGF alpha immunoreactivity was observed in fully developed carcinomas at week 19, although some tumours were negative by immunohistochemistry. The up-regulation of immunodetectable TGF alpha and the concomitant down-regulation of EGFr demonstrated positive (P < 0.01) and negative (P < 0.001) correlations, respectively, with hepatocyte proliferation indices. These findings suggest that the TGF alpha/EGFr ligand receptor system may be important during tumour promotion and in the stimulation of continued proliferation in hepatocellular carcinomas.

CD44 is expressed in hepatocellular carcinomas showing vascular invasion.

CD44 and its variant isoforms are a group of transmembrane glycoproteins which play important roles in immune recognition, in lymphocyte trafficking, and in cell-cell and cell-matrix interactions. Although CD44 is expressed by some normal human epithelial and mesenchymal cells, upregulation of CD44 expression has been related to the metastatic potential of some malignant tumours. In this study of 27 hepatocellular carcinomas (HCCs), an indirect immunohistochemical method was used to investigate the distribution of CD44 in normal liver and to determine whether expression of the standard form of CD44 (CD44s), or two of its variant isoforms (CD44-v3 and CD44-v6), correlated with tumour grade, proliferation indices, or histological evidence of vascular invasion. Fifteen of the tumours were Edmondson grade II, four were grade III, and eight were grade IV. Liver cell dysplasia was present in adjacent liver parenchyma in three cases and vascular invasion was observed in ten HCCs. Vascular invasion was found to be more frequent in high grade HCCs and a significant correlation was observed between tumour proliferation indices and vascular invasion. CD44s was not expressed by epithelial cells of normal liver but was expressed by tumour cells in six HCCs; vascular invasion was present in five of these HCCs. Three CD44s-positive cases also expressed CD44-v3 and two of these also expressed CD44-v6. CD44 was not expressed in areas of hepatocyte dysplasia. There was a significant correlation between CD44 expression and the presence of vascular invasion, but not between CD44 expression and tumour grade or tumour proliferation indices. It is concluded that upregulation of cell surface CD44 expression on malignant hepatocytes is related to their tendency to vascular invasion and may have implications relating to metastasis and prognosis in patients with HCCs.

Intrahepatic distribution of transforming growth factor-alpha (TGF alpha) during liver regeneration following carbon tetrachloride-induced necrosis.

The distribution of transforming growth factor-alpha (TGF alpha) in rat liver during regeneration was studied immunohistochemically using two antibodies, one a polyclonal (26T) raised against a synthetic peptide corresponding to the 17 C-terminal amino acids of the mature rat protein, and the other a monoclonal (Ab-2) raised against recombinant human protein. In normal liver, immunoreactive TGF alpha was detected in perivenular hepatocytes using both antibodies. No sinusoidal cells were found to contain the peptide. In response to carbon tetrachloride (CCI4)-induced necrosis, an initial increase in the intensity of immunoreactivity was noted at 24 h following exposure to the toxin. This coincided with the period immediately preceding the peak of hepatocyte proliferation; Ab-2 immunoreactive cells outnumbered 26T-positive cells. Thereafter there was a reduction in the number of TGF alpha-positive cells, but by day 4 the level of immunoreactivity had returned to that of normal liver. Using bromodeoxyuridine labelling, spatial and temporal relationships between TGF alpha expression and cell proliferation were identified, supporting the concept that this peptide plays an important role in the in vivo regenerative response to hepatic injury via an autocrine mechanism.