The research of liver metastasis is a developing field. The ability of tumor cells to invade the liver depends on the complicated interactions between metastatic cells and local subpopulations in the liver (including Kupffer cells, hepatic stellate cells, liver sinusoidal endothelial cells, and immune-related cells). These interactions are mainly mediated by intercellular adhesion and the release of cytokines. Cell populations in the liver microenvironment can play a dual role in the progression of liver metastasis through different mechanisms. At the same time, we can see the participation of liver parenchymal cells and nonparenchymal cells in the process of liver metastasis of different tumors. Therefore, the purpose of this article is to summarize the relationship between cellular components of liver microenvironment and metastasis and emphasize the importance of different cells in the occurrence or potential regression of liver metastasis.

This review on the unique patterns of metastases by common and rare types of cancer addresses regional lymphatic metastases but also demonstrates general principles by consideration of vital organ metastases. These general features of successfully treated metastases are relationships to basic biological behavior as illustrated by disease-free interval, organ-specific behavior, oligo-metastatic presentation, genetic control of the metastatic pattern, careful selection of patients for surgical resection, and the necessity of complete resection of the few patients eligible for long-term survival after resection of vital organ metastasis. Lymph node metastases, while illustrating these general features, are not related to overall survival because lymph node metastases themselves do not destroy a vital organ function, and therefore have no causal relationship to overall survival. When a cancer cell spreads to a regional lymph node, does it also simultaneously spread to the systemic site or sites? Alternatively, does the cancer spread to the regional lymph node first and then it subsequently spreads to the distant site(s) after an incubation period of growth in the lymph node? Of course, if the cancer is in its incubation stage in the lymph node, then removal of the lymph node in the majority of cases with cancer cells may be curative. The data from the sentinel lymph node era, particularly in melanoma and breast cancer, is consistent with the spectrum theory of cancer progression to the sentinel lymph node in the majority of cases prior to distant metastasis. Perhaps, different subsets of cancer may be better defined with relevant biomarkers so that mechanisms of metastasis can be more accurately defined on a molecular and genomic level.

Osteopontin (OPN) is a glycophosphoprotein with multiple intracellular and extracellular functions. In vitro, OPN enhances migration of mouse neutrophils and macrophages. In cancer, extracellular OPN facilitates migration of cancer cells via its RGD sequence. The present study was designed to investigate whether osteopontin is responsible for neutrophil and macrophage infiltration in human cancer and in particular in glioblastoma. We found that in vitro mouse neutrophil migration was RGD-dependent. In silico, we found that the OPN gene was one of the 5% most highly expressed genes in 20 out of 35 cancer microarray data sets in comparison with normal tissue in at least 30% of cancer patients. In some types of cancer, such as ovarian cancer, lung cancer and melanoma, the OPN gene was one of those with the highest expression levels in at least 90% of cancer patients. In glioblastoma, the most invasive type of brain tumours/glioma, but not in lower grades of glioma it was one of the 5% highest expressed genes in 90% of patients. In situ, we found increased protein levels of OPN in human glioblastoma versus normal human brain confirming in silico results. OPN protein expression was co-localized with neutrophils and macrophages. In conclusion, OPN in tumours not only induces migration of cancer cells but also of leucocytes.

The innate immune mechanisms of the liver represent an important first line of defense against bacterial products, toxins, and food antigens coming from the intestine. Natural Killer (NK) and Natural Killer T cells (NKT) are components of the innate immune system with increased presence in the liver compared to other organs and have been reported to participate in the inflammatory processes during hepatic diseases. However significant confusion has been noted in this field mainly due to changes in the characterization of these cells as new knowledge accumulates and due to differences in the approaches used for their study. Both cell types can mediate hepatic injury in several models but studies in human liver diseases have not managed to fully explain their functions. However accumulating evidence supports an antifibrotic role of NK cells mainly via an inhibitory effect on hepatic stellate cells by inducing apoptosis and via production of interferon-gamma. Therefore, downregulation of NK cells during most types of liver injury may facilitate liver fibrosis. Data about the role of NKT cells in liver fibrosis are limited. This review will summarize the studies about the role of NK and NKT cells in liver diseases with a special interest in hepatic injury and liver fibrosis.

To investigate chemopreventive effect of liposomal beta-sitosterol on tumor metastasis, we prepared liposomal beta-sitosterol composed of egg yolk phosphatidylcholine for oral delivery. Although orally administered beta-sitosterol (4 micromol as beta-sitosterol/mouse) was not absorbed into plasma, the amount of immune response cytokines such as IL-12 and IL-18 was increased in the small intestine after the liposome intake. Moreover, after daily oral administration of the liposome for 7 d, natural killer (NK) cell activity in the mice was increased, suggesting that the immune surveillance activity of mice was enhanced by the liposomal beta-sitosterol intake. Thus, we examined metastatic potential of B16BL6 melanoma cells, which were intravenously injected into mice after sequential administration of liposomal beta-sitosterol for 7 d. The number of metastatic colonies in the lungs was significantly less than that of control group two weeks after the injections of the cells. These results suggest that daily liposomal beta-sitosterol intake prevents tumor metastasis may be due to enhancement of gut immune surveillance systems.

The activity of a set of peptidases (proteases) involved in cancer progression is collectively known as the cancer 'degradome'. Invasion and metastasis were initially considered as late events in cancer development and the processes in which proteases were involved. However, recent studies indicate that invasion and metastasis are not late events, but can occur during early stages as well. Moreover, other processes occurring in various stages of cancer progression are also protease-dependent, such as (upregulation of) cell proliferation, (downregulation of) apoptosis, involvement of white blood cells, angiogenesis and induction of multi-drug resistance. Proteolytic activity in tumours is regulated in a complex manner, as both genetically unstable cancer cells and stable stromal cells, such as fibroblasts, endothelial cells and inflammatory cells, are involved. In vitro studies and studies using animal models have clearly shown protease dependency of many processes in carcinogenesis. However, clinical trials using protease inhibitors have thus far been unsuccessful except for a few applications of matrix metalloprotease (MMP) inhibitors when used in combination with cytostatic anticancer agents and/or in the early stages of cancer. Antithrombotics, such as low-molecular-weight heparin and warfarin, were also successful in clinical trials, probably by interfering with proteases of the coagulation cascade. The two-way association between cancer and thrombosis has long been recognised in the clinic. The poor outcome of other clinical trials of protease inhibitors is probably due to the late stages of cancer of the patient populations included, and the limited understanding of the complex regulation and effects of the activity of the various proteases in tumours depending on, among others, tumour type and stage, interactions between the cancer cells, other cells and the extracellular matrix in tumours. Therefore, a better fundamental understanding of the proteolytic complexity in tumours is essential before clinical trials can be rationally designed. At present, antithrombotics, the urokinase-type plasminogen activator system, the membrane-bound membrane-type 1-MMP, cathepsin L and the proteasome seem the most promising candidates as targets for anticancer strategies in early stages of cancer in combination with cytotoxic drugs. Moreover, metronomic therapy is an attractive approach using low doses of inhibitors for prolonged periods of time without interruption to specifically target endothelial cells that are involved in angiogenesis.

Liver sinusoids harbor populations of 2 important types of immunocompetent cells, Kupffer cells (KCs) and natural killer (NK) cells, which are thought to play an important role in controlling hepatic metastasis in the first 24 hr upon arrival of the tumor cells in the liver. We studied the early interaction of KCs, NK and CC531s colon carcinoma cells in a syngeneic rat model by confocal laser scanning microscopy. Results showed a minority of KCs (19% periportal and 7% pericentral) involved in the interaction with 94% of tumor cells and effecting the phagocytosis of 92% of them. NK cell depletion decreased the phagocytosis of tumor cells by KCs by 33% over a period of 24 hr, leaving 35% of the cancer cells free, as compared to 6% in NK-positive rats. Surviving cancer cells were primarily located close to the Glisson capsule, suggesting that metastasis would initiate from this region.

Glucose-6-phosphate dehydrogenase (G6PD; EC 1.1.1.49) is the key regulatory enzyme of the pentose phosphate pathway and produces NADPH and riboses. In this study, the kinetic properties of G6PD activity were determined in situ in chemically induced hepatocellular carcinomas, and extralesional and control parenchyma in rat livers and were directly compared with those of the second NADPH-producing enzyme of the pentose phosphate pathway, phosphogluconate dehydrogenase (PGD). Distribution patterns of G6PD activity, protein, and mRNA levels were also compared to establish the regulation mechanisms of G6PD activity. In (pre)neoplastic lesions, the V(max) of G6PD was 150-fold higher and the K(m) for G6P was 10-fold higher than in control liver parenchyma, whereas in extralesional parenchyma, the V(max) was similar to that in normal parenchyma but the K(m) was fivefold lower. This means that virtual fluxes at physiological substrate concentrations are 20-fold higher in lesions and twofold higher in extralesional parenchyma than in normal parenchyma. The V(max) of PGD was fivefold higher in lesions than in normal and extralesional liver parenchyma, whereas the K(m) was not affected. Amounts of G6PD protein and mRNA were similar in lesions and in extralesional liver parenchyma. These results demonstrate that G6PD is strongly activated post-translationally in (pre)neoplastic lesions to produce NADPH.

In the present study, B16 melanoma cells were found to produce inhibitory and cytotoxic substances with a molecular weight lower than 3000 Da against macrophages in a conditioned medium. The B16 melanoma-conditioned medium suppressed nitric oxide (NO) production only by mouse peritoneal macrophages and the mouse macrophage-like cell line, RAW264.7 cells, but not by rat peritoneal macrophages. In addition, it showed cytotoxicity against mouse peritoneal macrophages and mouse macrophage-like cell lines, RAW264.7 and J774A.1 cells, but not against rat cells (peritoneal macrophages, 3Y1, hepatocytes), human cells (HeLa, KB, MCF-7), or mouse 3T3-L1 cells. The inhibitory activity of NO production was not affected by trypsin treatment or arginine supplementation, but it was abolished by heat treatment at 95 degrees C for 3 min. On the other hand, the cytotoxicity was not influenced by these treatments. Inducible NO synthase induction following lipopolysaccharide stimulation was reduced by treatment of mouse peritoneal macrophages with B16 melanoma-conditioned medium. These results suggest that metastatic B16 melanoma cells produce two distinct substances: to suppress NO production by macrophages and to kill macrophages and macrophage-like cell lines. We propose that these activities may help metastatic B16 melanoma cells to escape a host immunosurveillance system and to metastasize to target organs.

The expression of the following cell adhesion molecules, their beta1 and beta2 integrin ligands and the cytokine tumour necrosis factor-alpha (TNF-alpha) was investigated by light and electron microscope immunohistochemistry in the liver tissue in 20 patients with colorectal and gastric cancer also presenting with liver metastases: intercellular adhesion molecule-1 (ICAM-1), vascular endothelial adhesion molecule-1 (VCAM-1), E-selectin, leucocyte function-associated antigen-1 (LFA-1), macrophage antigen-1 (Mac-1), and very late antigen-4 (VLA-4). We have found a parallel enhancement of the adhesion molecules and of TNF-alpha in liver sinusoids surrounding metastases. The expression of ICAM-1 was enhanced on sinusoidal cells in all zones of the acinus. VCAM-1 immune reactivity was diffuse but less intensive in the lobule. E-selectin expression was observed in sinusoidal cells attached to metastases. In tumour metastases the expression of ICAM-1, VCAM-1, and E-selectin was visible on the tumour vascular endothelium. Tumour infiltrating host cells sowing positive immunoreactivity for ICAM-1, VCAM-1, LFA-1, Mac-1, and VLA-4 were located mainly at the boundary between liver parenchyma and the metastasis. At the ultrastructural level, ICAM-1-positive immune deposits were observed on the cellular membrane and in some transport vesicles of gastric metastatic cells. Further, the expression of all adhesion molecules was confirmed to sinusoidal endothelial cells and tumour vessels. It is concluded that the enhanced expression of adhesion molecules in liver sinusoids could be a marker for the assessment of the ability of sinusoidal endothelial cells to control the recruitment of leukocytes and monocytes to the metastatic site. They could also direct the adhesion of new circulating tumour cells to sinusoidal endothelium.

Recently, it was demonstrated that dietary omega-3 polyunsaturated fatty acids (PUFAs) induce 10-fold more metastases in number and 1000-fold in volume in an animal model of colon cancer metastasis in rat liver. It was observed that tumors of rats on a fish oil diet lacked peritumoral stroma unlike tumors in livers of rats on a low fat diet or a diet containing omega-6 PUFAs. In the present study, only one-third of the tumors in livers of rats on omega-3 PUFA diet contained peritumoral stroma, whereas peritumoral stroma was present in 87% of the tumors in livers of rats on low fat diet. To explain these findings, we tested the hypothesis that fish oil exerts a direct inhibiting effect on the formation of extracellular matrix in tumor stroma as a consequence of blocking transformation of fat storing cells into myofibroblasts. It was found with immunohistochemical analysis of desmin as marker for fat storing cells and alpha-smooth muscle actin as marker for myofibroblasts that numbers of myofibroblasts were higher in tumors containing intratumoral stroma only than in tumors containing both peritumoral and intratumoral stroma. As most of the tumors in fish oil-treated rats contained intratumoral stroma only, this suggests that transformation of fat storing cells into myofibroblasts was highest in tumor stroma of fish oil-treated rats. Therefore, it is unlikely that the lack of stroma around tumors in fish oil-treated rats is due to inhibition of transformation of fat storing cells into myofibroblasts, but lack of peritumoral stroma is rather a consequence of rapid development of tumors in livers of fish oil-treated rats.

Metastasis of colorectal carcinomas rarely occurs in cirrhotic livers. Our study investigated the influence of activated Kupffer cells from cirrhotic rat livers on hepatic colonization and FasR-mediated apoptosis of colon cancer cells. A rat colon cancer cell line, RCN-9, was used to inoculate rat livers. Treatment with conditioned media of Kupffer cells isolated from CCl(4)-induced cirrhotic rat livers (cirrhotic KCM) significantly reduced the incidence of hepatic colonization of RCN-9 cells. In vitro cytotoxicity of Kupffer cells and tumour infiltrating lymphocytes (TILs) on RCN-9 cells was evaluated using [(3)H]-release assay. RCN-9 cells were resistant to cytotoxicity mediated by cirrhotic Kupffer cells, but were sensitized to TIL-mediated killing after treatment with cirrhotic KCM. The specific killing induced by TILs was FasR-mediated, as it was inhibited by ZB4, an antagonistic anti-FasR antibody. In agreement, cirrhotic KCM increased recombinant Fas ligand-induced apoptosis of RCN-9 cells, and up-regulated FasR expression on RCN-9 cells as evaluated by RT-PCR and flow cytometry. These findings suggest that Kupffer cells in cirrhotic livers sensitize metastatic colon cancer cells to FasR-mediated apoptosis by up-regulating the receptors, which thus prepare them to be eliminated by infiltrating lymphocytes.

Immune surveillance may play a role in protecting against the establishment of metastasis. The authors previously observed that the injection of as few as 10(4) lung metastatic B16BL6 melanoma cells/0.2 mL resulted in no metastasis and in a reduced rate of cell accumulation in the lung, the target organ. In the current study, the authors examined the correlation between metastatic potential and tumor cell trafficking by using a liver-metastatic model.

The liver-metastatic potential of RAW117-H10 cells was examined by varying the number of cells injected into mice through the portal vein. To investigate the trafficking of the cells, the authors performed positron emission tomography (PET) analysis, because advances in this technology now enable the use of PET to investigate the real-time trafficking of as few as 10(4) cells/0.2 mL. Furthermore, to clarify the role of the immune defense system, metastatic potential and cell trafficking also were examined by using macrophage-depleted mice.

When 10(6) or 10(5) RAW117-H10 cells/0.2 mL were injected into mice, both quantities of cells caused liver metastasis, cells accumulated in the liver at a similar rate, and there was an approximately 10-fold difference in the number of accumulated cells between the two doses. However, the injection of 10(4) cells/0.2 mL did not produce metastasis, and the accumulation rate in the liver was less than one-tenth of that after the injection of 10(5) cells/0.2 mL. The treatment of mice with 2-chloroadenosine for depleting macrophages prior to the injection of 10(4) cells/0.2 mL resulted in the suppression of the fast elimination of the cells from the liver. Corresponding to this change in PET images, the injection of 10(4) cells/0.2 mL into 2-chloroadenosine-pretreated mice resulted in metastasis.

The current study suggests that immune surveillance suppresses accumulation of tumor cells to the target and suppresses metastasis, and this effect is obvious when small numbers of tumor cells are used for the challenge. Furthermore, the immune defense system plays a role in the early stage of the metastatic process.

The alteration in sinusoidal collagen type IV occurrence, and myofibroblastic (alpha-SMA-positive) Ito cellular transformation are described in the liver of patients with malignant gastric and colorectal tumors, using electron microscopy as well as light microscopical and ultrastructural immunohistochemistry. The ultrastructural finding revealed transformation of Ito cells mostly into transitional cells in highly differentiated primary tumors and into transitional and myofibroblast-like cells with expressed changes in the other sinusoidal cells in poorly differentiated tumors. Ito cell numbers increased significantly in the livers of cancer patients. A highly significant statistical association was obtained between Ito cell numbers on the one hand and collagen type IV and alpha-SMA immunoreactivity on the other hand in the pericentral zone of the liver lobule. Ultrastructural immunohistochemistry showed increased collagen IV immune deposits in the space of Disse, assembled for the most part around and inside transitional cells. Alpha-SMA immunoreactivity was detected in activated Ito cells diffuse in the lobule, with stronger expression in the intermediate and pericentral zones. It is suggested that stimuli which can influence Ito cell transformation are produced by tumor cells from the primary tumor (TGF-beta1, TNF-alpha, PDGF-beta etc.) and from the metastasizing gastric or colorectal tumor cells--matrix metalloproteinase-2 (MMP-2). It is suggested that sinusoidal extracellular matrix deterioration creates a barrier for cancer invasion on the one hand, or possibly facilitates metastasizing by ensurance of matrix for adhesion on the other hand.

The relationship among the real-time trafficking of lung metastatic B16BL6 cells, metastatic potential, and the injected number of the cells was examined, since the smaller the number of tumor cells injected, the more clearly the immune defense may be observed. When 1x10(6) or 1x10(5) B16BL6 cells were injected into mice via the tail vein, both numbers of cells accumulated in the lung at a similar rate: there was an approximately 10-fold difference in the number of accumulated cells between the two doses. Elimination from the lung was not dependent on the cell number but on the proportion of accumulated cells. However, the injection of 1x10(4) cells resulted in lung accumulation less than one-tenth of that obtained with 1x10(5) cell injection. Metastasis was observed when 1x10(5) or 1x10(6) B16BL6 cells were injected, but not after injection of 1x10(4) cells. To clarify the roles of the immune defense system at the initial phase of metastasis, we challenged macrophage-depleted mice with 1x10(4) tumor cells. Treatment of mice with 2-chloroadenosine prior to the tumor cell challenge cancelled the suppression of not only metastasis but also the lung accumulation. Furthermore, the administration of 2-chloroadenosine following the tumor cell challenge had little effect on the metastatic potential. These results suggest that the immune surveillance whose action was obvious at the low dose of challenged tumor cells functions strongly at the initial phase but not at the advanced stages of the metastatic process, and that macrophages play an important role in the suppression of metastasis.

In preclinical studies, tumor cells genetically altered to secrete granulocyte-macrophage colony-stimulating factor (GM-CSF) can generate systemic antitumor immunity. Clinically relevant immunotherapeutic approaches for the treatment of colorectal cancer should address efficacy within the liver, a common site of metastatic disease. We investigated the effect of irradiated colon cancer cells engineered to produce GM-CSF on protecting from and treating established liver metastases.

Using a model of liver metastasis by intrahepatic injection of CT-26 murine colon carcinoma cells in syngeneic BALB/c mice, GM-CSF-producing irradiated cells were given as an intradermal vaccine either 14 days prior to hepatic challenge or in animals with early established tumor (days 5 and 10). The presence of tumor, tumor volume, and survival were endpoint determinants.

Animals receiving GM-CSF-producing vaccination demonstrated significant protection from subsequent hepatic challenge of viable tumor cells, even at the highest challenge doses. In animals with early established tumors, a significant response was seen with prolongation in survival.

We conclude that GM-CSF autologous tumor vaccination was effective for the treatment of hepatic colorectal metastases in this murine model. These findings provide support for immunotherapeutic approaches for metastatic liver cancer.

Kupffer cells (KC), the liver macrophages, are able to produce PGE(2), which is involved in immune suppression and in the aggravation of cancer cachexia due to interference with lipid metabolism in the liver. Since tumour-bearing (TB) rats present high plasma epinephrine levels, and this hormone is able to affect macrophage metabolism and function, we have assessed the effect of epinephrine (5 nM) upon Kupffer cell PGE(2) production. Epinephrine induced increased production of PGE(2) both by control (3.5-fold) and TB rats (27 per cent) KC, an effect blocked by propranolol. Enhancement of cAMP content in the cells by addition of isoproterenol (0.1 microM) to the incubations, however, failed to induce the same response in the cells. Nevertheless, when phenylephrine (1 microM) was added to the incubation, a similar pattern of PGE(2) production to that observed for epinephrine was found for control and TB rat KC. We propose that the effect of epinephrine upon KC PGE(2) production is mediated by alpha-adrenergic receptors and that Ca(2+) is involved in the response, since increasing concentrations of the ion added to the incubation medium (0.25, 0.5 and 1.0 mM) enhanced the eicosanoid production, while EDTA abolished the response.

Disseminated colon carcinoma metastases in the liver are associated with low cure rates and constitute a serious therapeutic problem. Appropriate experimental models which mimic metastases development and outgrowth can provide insight into the mechanism of this lethal process and facilitate the finding of new approaches for its control. We established an orthotopic liver metastases model based on CC531 rat colon adenocarcinoma cells which were transfected with a beta-galactosidase gene as marker to facilitate their detection. Intraportal injection of CC531-lac-Z cells resulted in a rapid and locally aggressive growth within the liver and was characterised by a tumour volume doubling time of 20 h and abundant angiogenesis. A commercially available chemi-luminescence assay allowed rapid, quantitative and sensitive detection of the diffusely growing tumour cells. Immunogenicity of CC531-lac-Z cells induced by the marker gene was significantly reduced by co-administering the tumour cells with matrigel. Within an observation period of three weeks following tumour cell injection only 6% of the animals showed lung involvement, thus indicating a specific homing of CC531-lac-Z cells to the liver. This period appears long enough to allow therapeutic manipulations at various stages of tumour growth in the liver. It is envisaged that the model will have applications for various therapeutic strategies.

Techniques for creation of colon carcinoma epithelial cells lines in long-term culture have been available for years, but these techniques have involved mechanical or enzymatic methods to separate epithelial cells from surrounding tissues. While this practice has been intermittently successful, the effect of these traumatic methods on long-term cellular behavior is unknown. Samples of colon carcinoma from patient volunteers were subjected to serial nonenzymatic disruptions of carcinoma cells from surrounding fibrous tissues. Cells were collected, allowed to proliferate, and then tested for their epithelial characteristics (mucin, vimentin, cytokeratin, colon-specific antigen, carcinoembryonic antigen) by immunohistochemistry and flow cytometry. Growth characteristics were determined by phase-contrast microscopy, multiple passage, and freeze/thaw effects. Tumorigenicity was proven in nude mice. Of 11 initial attempts, three resulted in stable long-term culture lines of cells which are demonstrated to behave similarly to the original tumors from which they were derived. This technique adds another reliable in vitro tool for the study of colon carcinoma.

To elucidate the early events of blood-borne metastasis under actual blood flow, real-time trafficking of RAW117 large cell lymphoma cells, namely parental RAW117-P and liver-metastatic RAW117-H10 cells, was investigated using positron emission tomography (PET). Both types of cells accumulated in the liver immediately after injection via the portal vein, and were eliminated from the liver time-dependently. The elimination rate of RAW117-H10 cells, however, was slower than that of RAW117-P cells, suggesting that RAW117-H10 cells interact more strongly with hepatic sinusoidal endothelium than the parental cells. This result correlated with the metastatic potential of these cells: RAW117-H10 cells metastasized in the liver to a greater extent than RAW117-P cells after injection via this route. To investigate the role of sialylglycoconjugates in the interaction of RAW117-H10 cells with the hepatic endothelium after injection via the portal vein, the trafficking of RAW117-H10 cells was examined after the cells had been treated with sialidase. The elimination rate of RAW117-H10 cells from liver was observed to be greatly accelerated by sialidase treatment. To elucidate what kind of sialylglycoconjugates is related to this phenomenon, we analyzed the distribution of sialyl Lewis A and sialyl Lewis X antigens of both sublines of RAW117 by using flow cytometry. RAW117-H10 cells were found to express a much higher level of sialyl Lewis A than RAW117-P cells, whereas the amount of sialyl Lewis X did not differ significantly. These findings suggest that some sialylglycoconjugates, perhaps sialyl Lewis A in particular, play an important role in the initial interaction of RAW117-H10 cells with the hepatic endothelium, leading to metastasis.

ICAM-1 mediates the recruitment of neutrophils through the endothelium to the site of inflammation by the ICAM-1/Mac-1 and ICAM-1/LFA-1 adhesion pathways. In extrahepatic cholestasis, recruitment of neutrophils is a main feature of the inflammatory infiltrate in areas of parenchymal damage. The aim of the present study was to describe the light and electron microscopical localization of ICAM-1 expression in the liver of cholestatic patients. The peroxidase-antiperoxidase technique was used. Increased ICAM-1 expression was detected on sinusoidal endothelial and Kupffer cells. A de novo ICAM-1 expression was described on some Ito cells and the sinusoidal hepatocyte membrane in areas of parenchymal injury. In the portal areas of livers of cholestatic patients, ICAM-1 was observed on the endothelial surface of portal veins and on hepatic arteries. Occasionally, ICAM-1 was found on the surface of bile duct epithelia. It is suggested that ICAM-1 expression is up-regulated by cytokines like TNF-alpha, IL-1 and interferons released from activated Kupffer cells. The mechanisms of ICAM-1 upregulation and neutrophil recruitment in the liver during extrahepatic cholestasis are discussed.

Metastatic rat colon cancer cells but not normal rat hepatocytes showed activity of cathepsin B on their plasma membranes. Activity was visualized in living cells with a new fluorogenic substrate, [Z-Arg]2-cresyl violet, and confocal microscopy. When these cancer cells were injected into the portal vein of rats, the animals developed tumors in the liver in a heterogeneous fashion. Three- to four-fold more tumors were found in the small caudate lobe than in the other three large lobes of the liver. Oral treatment with a selective water-soluble inhibitor of extracellular cathepsin B, Mu-Phe-homoPhe-fluoromethylketone, resulted in 60% reduction of the number of tumors and 80% reduction of the volume of tumors in the three large lobes whereas tumor development was not affected in the small caudate lobe. This study supports the conclusions that (a) extracellular cathepsin B plays a crucial but complex role in liver colonisation by rat colon carcinoma cells in vivo, (b) its selective inhibition suppresses tumor growth heterogeneously in the liver and (c) the caudate lobe of the liver is a relatively large risk factor for tumor development.

The deterioration of extracellular matrix turnover is a key event in tumor progression. It has been assumed that Ito cell transformation is stimulated by tumor-derived factors. In the present study changes in the occurrence of collagen type III and IV and Ito cell transformation are described in the sinusoids of patients with malignant gastrointestinal tumors without liver metastases, and around metastatic liver tumors using routine histology, electron microscopy as well as light microscopical and ultrastructural immunohistochemistry. Dilated sinusoids filled with lymphoid cells and variable perisinusoidal fibrosis were detected light microscopically. Collagen type III and IV immune deposits were increased perisinusoidally. Ultrastructural immunohistochemistry showed increased staining in the space of Disse and around Ito and transitional cells for both types of collagen. Ito cells were transformed into transitional cells. Pit cells appeared in the inflammatory infiltrate in sinusoids. Ito cells were significantly increased in number pericentrally and periportally. It is suggested that stimuli, which can influence Ito cellular behaviour are produced by inflammatory cells in sinusoids, resident sinusoidal cells, tumor cells or by tumor stroma. It is concluded that transformed Ito cells and increased amounts of collagen type III and IV in sinusoids of patients with malignant tumors without liver metastases or around metastatic tumors may predict tumor-related alterations of liver parenchyma, which may serve as a barrier for further outgrowths of the cancer cells.