Search

The cytotoxic activity of cis-[((1R,2R)-1,2-cyclohexanediamine-N, N')bis(myristato)]platinum (II) (SM-11355) was evaluated in a cisplatin (CDDP)-resistant tumor cell line, and compared with that of CDDP. H4-II-E / CDDP with acquired resistance to CDDP was established by continuous exposure of a rat hepatic tumor line, H4-II-E, to increasing concentrations of CDDP over 12 months. Compared with the parental cell line, this cell line exhibited an 8.8-fold increase in resistance to CDDP and was not cross-resistant to 1,2-diaminocyclohexane platinum (II) dichloride (DPC). There were no differences in sensitivity to six non-platinum antitumor drugs between H4-II-E and H4-II-E / CDDP, which suggests that H4-II-E / CDDP is not multidrug-resistant. Intracellular platinum accumulation and the formation of a platinum-DNA adduct following CDDP exposure were significantly reduced in H4-II-E / CDDP compared to the parental cell line. The acquired CDDP resistance in H4-II-E / CDDP appeared to be predominantly due to reduced CDDP uptake. H4-II-E / CDDP was also resistant to CDDP suspended in Lipiodol (CDDP / Lipiodol), but was not cross-resistant to SM-11355 suspended in Lipiodol (SM-11355 / Lipiodol). Also, there were no differences in intracellular platinum accumulation or the formation of platinum-DNA adducts after SM-11355 / Lipiodol exposure between H4-II-E and H4-II-E / CDDP. These results suggest that acquired CDDP resistance in H4-II-E / CDDP does not influence the cytotoxic activity of SM-11355 / Lipiodol.

In man, liver cancer is on the increase, especially in males. Sex differences also exist in rodent models. To elucidate the mechanisms, chimeric mice were produced by amalgamation of early embryos from high and low hepatocarcinogen-susceptible strains, C3H and BALB / c. Tumor formation was initiated with 10 mg / kg of diethylnitrosamine at the ages of 7 and 14 days and mice were sacrificed at 30 and 45 weeks. The chimeras were classified into XY <--> XY, XY <--> XX, XX <--> XY, and XX <--> XX in terms of sex chromosomes by means of polymerase chain reaction-simple sequence length polymorphism analysis (SSLP) using Y chromosome-specific Sry primers in combination with the D3Mit21 marker. Liver lesions were analyzed histopathologically, by immunostaining using a C3H strain-specific antibody and by DNA in situ hybridization with the Y chromosome-specific digoxigenin-labeled Y353 / B probe. Sex and strain genotyping by SSLP analysis matched histological observations, confirming the reliability of our system. The strain differences in liver tumor numbers of each strain type in XY <--> XY and XX <--> XX subtypes of C3H <--> BALB / c chimeras were retained well (P < 0. 0001 and P < 0.001, respectively), indicating a minimum influence of the C3H or BALB / c surrounding milieu on development of individual lesions. On the other hand, significant promotion of XX cell tumors was evident in phenotypically male sexually chimeric XY <--> XX and XX <--> XY chimeras for both C3H (P < 0.02) and BALB / c (P < 0.01) lesions compared to the XX <--> XX case. The results suggest the presence of hormonal or micro-environmental factors specific for males, which are not caused cell-autonomously. Basic strain differences, however, are determined by intrinsic genetic factors rather than the strain-dependent micro-environment.

This study was designed to investigate the chemopreventive action of three natural products, coumaperine, aurapten and an extract from rosemary, against the initiation stage of rat hepato-carcinogenesis. Coumaperine has been isolated from white pepper as a naturally occurring antioxidative agent, but its potential modifying effects on carcinogenesis remain unclear. In experiment 1, a modification of the model developed by Tsuda et al. was applied, with assessment of numbers and areas of induced glutathione S-transferase placental form (GST-P)-positive hepatocellular foci in male F344 rats. Coumaperine, aurapten and the extract from rosemary were administered i.g. at 100 mg / kg / day once daily for 5 days with initiation by diethylnitrosamine (DEN) on day 4 (20 mg / kg, i.p.). Numbers and areas of GST-P-positive foci in each group given test chemicals tended to be decreased as compared to the vehicle control group values, significance being achieved for number with coumaperine. Experiment 2 was planned to investigate the mechanism of the inhibitory effects of coumaperine. Livers at 8 h after initiation by DEN were examined with coumaperine administered at 100 mg / kg / day once daily for 3 days. Proliferating cell nuclear antigen (PCNA)-positive cells tended to be decreased as compared to the vehicle control, but no effects on apoptosis or cytochrome P-450 (CYP) 2E1 expression were apparent. Our results suggest that coumaperine provides protection against initiation of hepatocarcinogenesis, and that this is related to inhibition of cell proliferation.

The in vitro sensitivities to cisplatin of AH66 and AH66F cells, a variant obtained from AH66 cells, were very similar, when assayed in a medium containing 5% fetal bovine serum (FBS), whereas in the in vivo experiments AH66F cells were sensitive and AH66 cells were highly resistant to cisplatin. In this study, we examined the mechanism of the in vivo cisplatin resistance of AH66 cells. The in vitro cisplatin sensitivity of AH66 cells was lowered by changing FBS to 5% ascites fluid (ASF) in the assay medium and the sensitivity in FBS by treatment with buthioninesulfoximine (BSO). The sensitivity of AH66F cells was not changed by these treatments. Moreover, after culture in 5% ASF for 48 h, the accumulation of cisplatin in AH66 cells was decreased and the efflux of cisplatin from the cells was accelerated. The accumulation of cisplatin in AH66 cells in ASF was increased by pretreatment with BSO, sodium azide or probenecid. Then, we examined the expression of the glutathione (GSH) conjugate efflux pump family. Among them, only the expression of canalicular multispecific organic anion transporter (cMOAT) in AH66 cells was decreased by culture in FBS and enhanced by ASF. These results suggest that some substances contained in ASF enhanced the expression of cMOAT in the plasma membrane of AH66 cells and this transporter actively extruded cisplatin-GSH conjugate from the cells. Consequently, AH66 cells afford a cisplatin-resistant tumor in the host.

The effects of sardine fish oil or corn oil on diethylnitrosamine (DEN)-induced hepatocarcinogenesis were investigated in male F344 rats. Starting at 5 weeks of age, animals were divided into 11 groups and fed 23.5% corn oil (HCO) (groups 1 and 7) or 5% corn oil (LCO) (groups 2 and 8), 22.5% sardine oil + 1% corn oil (FO) semipurified diet (groups 3 and 9) or basal diet (CE-2) (groups 4-6, 10 and 11). At 6 weeks of age, all animals except the vehicle-treated groups were given DEN (200 mg/kg body weight, i.p. once weekly for 3 weeks). One week after the final exposure to DEN, groups 1-3 were changed to the basal diet, and groups 4-6 were switched to the HCO, LCO or FO diet, respectively. Animals in groups 1-3 and 10 were given drinking water containing 0.05% phenobarbital (PB). Liver sections from the animals at the termination of the experiment (24 weeks) were doubly stained for glutathione S-transferase placental form (GST-P) and silver-stained nucleolar organizer regions (AgNORs). The multiplicity of hepatocellular neoplasms of group 1 was significantly larger than that of group 2 or 3. The number of GST-P-positive foci of group 2 or 3 was significantly smaller than that of group 1. Among the groups fed the experimental diets in the postinitiation phase (groups 4-6), no significant difference was found in the incidence of liver tumors. AgNORs values of the enzyme-altered foci in rats of the HCO diet groups were larger than those of the other diet groups. These results indicate that the enhancing effect of a high dose of corn oil in hepatocarcinogenesis is mainly present during the initiation phase but not during postinitiation phase, and fish oil rich in polyunsaturated omega-3 fatty acids could inhibit DEN-induced hepatocarcinogenesis in rats.

The modifying effects of quinacrine administration during the post-initiation phase of carcinogenesis were investigated in hamsters treated with N-nitrosobis(2-oxopropyl)amine (BOP). Female Syrian hamsters were given three weekly s.c. injections of BOP at a dose of 10 mg/kg and then 300 or 100 ppm quinacrine in their diet for 37 weeks. Additional groups of animals received the BOP injection alone, or only the 300 ppm quinacrine treatment as BOP-negative controls. At week 40 of the experiment, all surviving animals were killed and development of proliferative lesions was assessed histopathologically. The multiplicity of pancreatic adenocarcinomas and dysplastic lesions per hamster was significantly higher (P<0.01 and P<0.05) in the BOP/Q100 group (1.92 and 1.78) than in the BOP-alone group (1.07 and 0.79). The incidence of hepatocellular adenomas plus carcinomas was also significantly elevated (P<0.05) in the BOP/Q300 and BOP/Q100 groups. In contrast, the multiplicity of lung adenomas plus adenocarcinomas was significantly decreased (P<0.05) by the Q300 treatment. Neither the incidence nor the multiplicity of renal cell tumors (adenomas and carcinomas) or nephroblastomas significantly differed between the BOP-treated groups. Electron microscopic examination revealed an abundance of myeloid lamellar bodies filling the cytoplasm of hepatocytes and pancreatic ductular and acinar cells, and epithelial cells of the gallbladder in the quinacrine-treated animals, the degree being dose-dependent. Our results indicate that quinacrine enhances pancreatic and hepatic carcinogenesis in hamsters induced by BOP.

Our previous study showed that treatment of highly invasive rat ascites hepatoma (LC-AH) cells with alpha-difluoromethylornithine (DFMO), an inhibitor of ornithine decarboxylase, decreased both their intracellular level of putrescine and their in vitro invasion of a monolayer of calf pulmonary arterial endothelial (CPAE) cells, and that both these decreases were completely reversed by exogenous putrescine, but not spermidine or spermine. Here we show that all adhering control (DFMO-untreated) cells migrated beneath CPAE monolayer with morphological change from round to cauliflower-shaped cells (migratory cells). DFMO treatment increased the number of cells that remained round without migration (nonmigratory cells). Exogenous putrescine, but not spermidine or spermine, induced transformation of all nonmigratory cells to migratory cells with a concomitant increase in their intracellular Ca2+ level, [Ca2+]i. The putrescine-induced increase in their [Ca2+]i preceded their transformation and these effects of putrescine were not affected by antagonists of the voltage-gated Ca2+ channel, but were completely suppressed by ryanodine, which also suppressed the invasiveness of the control cells. The DFMO-induced decreases in both [Ca2+]i and the invasiveness of the cells were restored by thapsigargin, which elevated [Ca2+]i by inhibiting endoplasmic Ca2+-ATPase, indicating that thapsigargin mimics the effects of putrescine. These results support the idea that putrescine is a cofactor for Ca2+ release through the Ca2+ channel in the endoplasmic reticulum that is inhibited by ryanodine, this release being initiated by cell adhesion and being a prerequisite for tumor cell invasion.

Expression of prohibition, a growing-regulatory protein, was immunohistochemically investigated in normal rat tissues, regenerating livers, and chemically induced preneoplastic and neoplastic hepatic lesions. Specific cell types including hepatocytes, striated and smooth muscle cells, salivary gland duct epithelial cells, chondrocytes, immature spermatocytes and oocytes were found to be positive. In regenerating livers, prohibitin protein disappeared as early as 3 h after two-thirds hepatectomy and returned to near the original level by 24 h, while its mRNA level did not markedly vary. The timing of the disappearance was coincident with the expression of c-myc, suggesting a relation to quiescent hepatocytes entering the cell cycle. However, no pronounced decrease was evident in the most hyperplastic hepatic nodules and hepatocellular carcinomas investigated. Examination of 9 rat hepatocellular carcinoma cell lines, 6 hyperplastic hepatic nodules and 5 hepatocellular carcinomas revealed a single case of a base substitution in prohibition cDNA, identified as a synonymous sense change. The observed abundant expression of prohibitin in quiescent hepatocytes and its rapid loss under conditions of regeneration indicate a growth-regulatory function, but our results do not suggest any critical role in rat hepatocarcinogenesis.

The distribution of hepatocyte growth factor (HGF)-synthesizing cells in rat liver during development of glutathione S-transferase P form (GST-P)-positive nodules after diethylnitrosamine initiation followed by promotion with 2-acetylaminofluorene plus partial hepatectomy (PH) was investigated using in situ hybridization. HGF-producing cells were non-parenchymal in nature, and were suspected to be mainly of Kupffer type. They were mostly located outside GST-P-positive lesions, in the surrounding parenchyma. In the oval cell proliferation phase 1 week after PH, they increased and they were mainly localized around the portal triads. It is concluded that HGF is directly involved in an endogenous paracrine growth pathway controlling proliferation in oval cells and in normal, but not GST-P-positive, hepatocytes.

To examine the mechanism of inhibition by protein kinase C (PKC) inhibitors of the adhesion of highly malignant hepatoma AH66F cells to the mesentery-derived mesothelial cell (M-cell) layer through leukocyte function-associated antigen-1 (LFA-1)/intercellular adhesion molecule-1, the effects of a PKC inhibitor, NA-382, on the expression of LFA-1 molecules in AH66F cells were examined and compared with those in thymocytes from normal rats. NA-382 inhibited the adhesion of AH66F cells to the M-cell layer and the expression of LFA-1 on the membrane of the hepatoma cells after treatment for more than 24 h. It was confirmed that AH66F cells express similar mRNAs for LFA-1 subunits to those of thymocytes, and their levels were also decreased after treatment with NA-382. On the other hand, the LFA-1 mediated adhesion and the expression of both protein and mRNA for LFA-1 subunits in thymocytes were not changed by the PKC inhibitor. These results suggest that the expression of LFA-1 molecules in AH66F cells may be regulated by PKC via quite different mechanisms from those in normal lymphocytes.

Hepatocellular carcinoma (HCC) is supplied only with arterial blood and has capillaries instead of sinusoids, unlike the normal hepatic tissue. In order to reveal the sequential changes of sinusoidal structure and blood supply during hepatocarcinogenesis, we examined hyperplastic nodules (HPN), atypical hyperplastic nodules (AHPN) and HCC in F344 rats fed with 2-acetylaminofluorene for 25-35 weeks. The extent of arterial blood supply and the degree of sinusoidal capillarization were assessed by the staining of hepatic nodules with arterially infused ink and by immunohistochemical and ultrastructural analyses of the sinusoids, respectively. The proportion of arterialized nodules was 47% in HPN, 57% in AHPN and 85% in HCC, which indicates that arterialization begins at the stage of HPN and a few HCC still receive portal venous blood. The proportion of Factor-VIII-related antigen (F-VIII)-positive nodules was 25% in HPN, 40% in AHPN and 100% in HCC. Non-arterialized nodules did not express F-VIII, while some F-VIII-negative nodules were already arterialized. Electron microscopically, non-arterialized HPN exhibited normal features of sinusoids, while arterialized HPN showed disappearance of Kupffer cells and partial defenestration of sinusoidal endothelial cells. In non-arterialized and arterialized AHPN, basement membrane became continuous and lipid droplet-containing stellate cells were no longer seen. In HCC, endothelial fenestrae were diminished and basement membrane became thick. The present study has demonstrated that sinusoids are altered in the following order as hepatocarcinogenesis advances from HPN to HCC; onset of arterialization and decrease of endothelial fenestrae, expression of F-VIII by endothelial cells, disappearance of Kupffer cells, diminution of lipid droplets in stellate cells and thickening of basement membrane.

The effect of plant glycosides on tumor cell invasion was examined. Among the glycosides tested, ginsenoside Rg3 was found to be a potent inhibitor of invasion by rat ascites hepatoma cells (MM1), B16FE7 melanoma cells, human small cell lung carcinoma (OC10), and human pancreatic adenocarcinoma (PSN-1) cells, when examined in a cell monolayer invasion model. Structurally analogous ginsenosides, Rb2, 20(R)-ginsenoside Rg2 and 20(S)-ginsenoside Rg3 (a stereoisomer of Rg3), showed little inhibitory activity. Neither Rh1, Rh2, 20(R)-ginsenosides Rh1, Rb1, Rc nor Re had any effect. The effective ginsenoside, Rg3, tended to inhibit experimental pulmonary metastasis by highly metastatic mouse melanoma B16FE7 cells as well. Taking account of our previous finding that 1-oleoyl-lysophosphatidic add (LPA) induced invasion by MM1 cells in the monolayer invasion model, the effect of Rg3 on molecular events associated with the invasion induced by LPA was analyzed in order to understand the mechanism of the inhibition. Rg3, which suppressed the invasion induced by LPA, dose-dependently inhibited the LPA-triggered rise of intracellular Ca2+. Protein tyrosine phosphorylation triggered by LPA was not inhibited by Rg3.

A food-borne heterocyclic amine, 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx), induces hepatocellular carcinomas (HCCs) in F344 male rats at an incidence of 95%, when fed in the diet at 400 ppm for 61 weeks. In this study, the effect of a low dose of MeIQx was examined in Long-Evans with cinnamon-like coat color (LEC) rats, which have a mutation in Atp7b and suffer from hereditary hepatitis and HCCs, with high levels of copper accumulation in the liver. Rats of the LEC and Long-Evans with agouti coat color (LEA) sibling lines were given a diet containing 40 ppm MeIQx from the age of 23 weeks to 63 weeks, for a total administration period of 40 weeks. In LEC rats, HCCs were observed in 8/8 animals administered MeIQx, and 2/8 rats receiving a normal diet. The number of HCCs per rat (mean +/- SD) was 2.8 +/- 2.0 and 0.3 +/- 0.5, respectively. In the LEA rats, however, no tumors were induced by administration of MeIQx. These results indicate that damaged liver associated with compensatory cell proliferation is much more susceptible to chemical hepatocarcinogens, including MeIQx, than the normal liver.

Rat ascites hepatoma AH66F cells adhered better than AH130 cells to mesentery-derived mesothelial cells (M-cells), though both cells secreted Mr 92,000 matrix metalloproteinase on a gelatin zymogram with similar activity. AH66F cells expressed leukocyte function-associated antigen-1 (LFA-1) and intercellular adhesion molecule-1 (ICAM-1), a ligand of LFA-1, on the cell surface, while AH130 cells had ICAM-1 alone. The adhesion of M-cells of AH66F cells was inhibited to the adhesion level of AH130 cells by anti-rat LFA-1 alpha -and/or beta-chain monoclonal antibody (mAb) and also by anti-rat ICAM-1 mAb. This is the first report to show the LFA-1-dependent adhesion of cells other than leukocytes, because AH66F cells did not express CD45, T cell-alpha beta receptor or Cd11b/c (Mac-1/p150,95). These results indicate that a part of the adhesion of AH66F cells to M-cells is due to LFA-1/ICAM-1 interaction, and we suggest that this characteristic feature of AH66F cells may be related to the malignant properties.

The effect of cell cycle disturbance due to colchicine on the induction of enzyme-altered foci during liver regeneration in rats was studied. For initiation, diethylnitrosamine (DEN) at a dose of 10 mg/kg was injected intraperitoneally and partial hepatectomy (PH) was performed 4 h thereafter. Colchicine at doses of 0, 0.1, 0.25 and 0.5 mg/kg was injected intraperitoneally 1 and 3 days after the initiation, followed by application of selection pressure consisting of 2-acetylaminofluorene (AAF) and carbon tetrachloride (CCl4) administration. As end point lesions, gamma-glutamyltransferase (GGT)-positive enzyme-altered foci were assayed at week 5. There was no significant effect of colchicine on numbers of foci. However, a significant, dose-dependent increase in the area of GGT-positive lesions in the groups treated with colchicine was observed. Bromodeoxyuridine labeling indices were higher in foci induced in colchicine-treated rats than in the untreated rats. In a separate experiment, serum glutamic pyruvic transaminase was not increased significantly after DEN and colchicine treatment, and the mitotic index at 6 days after PH was increased in the liver of colchicine-treated rats. These results suggest that the cell cycle disturbance induced by colchicine causes more pronounced selective growth of cells initiated by DEN and colchicine, and this experimental model may be useful for analyzing the mechanisms underlying that growth advantage and the effects of cell cycle abnormalities in liver carcinogenesis.

Using anti-rat hepatocyte growth factor (HGF) antibody, we investigated the distribution of HGF-positive cells in the liver tissues of LEC rats at various phases of liver diseases. During the phase of fulminant hepatitis, HGF-positive cells increased remarkably, and many of them were localized at the portal triads; these cells were identified from their shape as non-epithelial cells. A reduced number of HGF-positive cells was observed during the phase of chronic hepatitis, while no HGF-positive cells were seen in the tissue of cholangiofibrosis. During the phase of carcinoma, staining revealed that both the hepatocellular carcinoma cells and the non-epithelial cells in cancerous liver tissue were HGF-positive. These results suggest that, in LEC rats, HGF may play an important role in the regeneration of hepatocytes as well as in the development of hepatocellular carcinoma.

Sodium D-glucaro-delta-lactam (ND2001) inhibited spontaneous pulmonary metastases of the highly metastatic B16 melanoma variant with a maximal inhibition rate of 99.5%, and 6 of 7 animals remained metastasis-free. Likewise, ND2001 inhibited the spontaneous pulmonary metastases of both Lewis lung carcinoma (3LL) with a rate of 98.0% (3 of 5 animals remaining metastasis-free) and rat KDH-8 liver carcinoma with a rate of 82.5% (3 of 7 animals remaining metastasis-free), although it was unable to inhibit the metastases of mouse BMT-11 fibrosarcoma and rat SST-2 breast carcinoma. Pretreatment with ND2001 in vitro inhibited the pulmonary metastases of the B16 variant and 3LL cells, which indicates direct action upon the cancer cells. When the invasive activity of cancer cells was measured by the Boyden chamber method, the number of invading B16 variant or 3LL cells was reduced with maximal inhibition rates of 93.0% or 89.9%, respectively, but pretreatment with ND2001 failed to reduce the invasive activity of BMT-11 or SST-2 cells. ND2001 showed neither cytocidal nor antitumor activity. These results suggest that ND2001 inhibited pulmonary metastases at the invasive step into the basement membrane by directly changing some property of the tumor cells.

Four organosulfur compounds from garlic and onions were examined for modifying effects on diethylnitrosamine (DEN)-induced neoplasia of the liver in male F344 rats using the medium-term bioassay system based on the two-step model of hepatocarcinogenesis. Carcinogenic potential was scored by comparing the numbers and areas per cm2 of induced glutathione S-transferase placental form-positive foci. Isothiocyanic acid isobutyl ester (IAIE), dipropyl trisulfide (DPT), and allyl mercapton (AM) exerted enhancing effects on their development, while dimethyl trisulfide also tended to increase them. To investigate possible mechanisms of the modifying influence, sequential changes in ornithine decarboxylase activity (ODC) over 24 h were measured in AM-treated liver tissue without prior DEN initiation. The activity started to increase by 4 h after AM-treatment, and reached maximum at 16 h, compared to controls. Spermidine/spermine N1-acetyltransferase activity was not significantly changed. An increase in proliferating cell nuclear antigen-positive cells followed the elevation of ODC activity. These results suggest that IAIE, DPT, and AM promote rat hepatocarcinogenesis and their promoting effect might be caused by increased cell proliferation with increased polyamine biosynthesis. In evaluating relationships between diet and cancer, it is thus appropriate to consider not only a possible protective role of garlic and onions, but also enhancing effects.

To compare the hepatotoxicity and hepatocarcinogenicity of N-nitroso-2-acetylaminofluorene (NO-AAF) and its parent compound, 2-acetylaminofluorene (AAF), male Sprague-Dawley rats were given intraperitoneal (i.p.) or subcutaneous (s.c.) injections of AAF or NO-AAF (60 mg/kg body weight/week) for ten months. In the AAF group, morphological changes were produced which involved gross distortions of the liver with multiple nodule formations. The rat livers in the NO-AAF group appeared to be smooth with a blunt-thick superior segment of the lateral lobe. The serum gamma-glutamyl transpeptidase activity in both the AAF group and the NO-AAF group was significantly elevated (P < 0.0005). The present study shows that i.p. and s.c. injections of NO-AAF resulted in a high incidence of well-differentiated hepatocellular carcinomas (HCC) (7/9 and 4/6, respectively), while poorly differentiated HCCs were induced by i.p. or s.c. administration of AAF (6/9 or 2/6, respectively). Subcutaneous lesions consisting of an inflammatory reaction and fibroadenoma formation were observed in the NO-AAF-treated rats, whereas no such skin lesions were detected in the AAF-treated animals. These results suggest that NO-AAF is a new direct-acting carcinogen which may be useful for investigating hepatocarcinogenesis.

The effect of cyclosporine A (CsA) on in vivo growth of hepatic metastasis was studied. Murine colon 38 tumor cells (1 x 10(5) were inoculated via the superior mesenteric vein. Mice were grouped depending on CsA dosage and time schedules: Group A: CsA 30 mg/kg body weight on the 7, 8 and 9th post tumor inoculation days by gavage; Group B: CsA 15 mg/kg body weight 30 min before tumor inoculation and 2 times more at 24 h intervals; Group C: CsA 30 mg/kg body weight at the same dose timing as Group B. Measurement of the diameter of the largest tumor serially by weekly laparotomy up to 4 weeks revealed that the tumor growth rates were significantly greater in Groups B and C than those in Group A or the control (without CsA). The mean tumor doubling times in the control, and Groups A, B and C were 2.2 +/- 1.3, 2.0 +/- 0.5, 1.5 +/- 0.4 and 1.3 +/- 0.8 days, respectively. The mean tumor numbers of hepatic metastasis were 13.2 +/- 8.3, 11.3 +/- 7.3, 19.4 +/- 8.7 and 19.6 +/- 6.8, respectively. Values of tumor proliferation index obtained by bromodeoxyuridine immunohistochemistry were 10.0 +/- 6.1%, 14.9 +/- 8.0%, 28.6 +/- 8.2% and 30.1 +/- 12.4%, respectively, with significant differences (Groups B and C vs. A or control, P < 0.05). In vitro MTT assay showed that cell viability rates were greater than 100% in the medium containing CsA concentrations of less than 10(-7) mol/liter. However, a cytostatic effect of CsA was apparent at higher concentrations. In contrast to the previous in vivo finding of a cytostatic effect of CsA on tumor cells, we found a cytoproliferative action when CsA was administered early in the course of metastatic tumor implantation in the liver. The mechanism of cytoproliferative effect of CsA is unknown but may involve multiple factors.

A sequential decrease in the number of hepatocyte tight junctions during the course of rat hepatocarcinogenesis was demonstrated by immunohistochemistry with a new 7H6 monoclonal antibody generated in our laboratory. Semiquantitative analysis by confocal laser scanning microscopy revealed that the expression of 7H6 antigen was reduced in hyperplastic foci, hyperplastic nodules and hepatocellular carcinomas (HCC) to 43%, 28% and 25%, respectively, compared to corresponding normal liver tissues. 7H6 antigen was scarce in HCC with a trabecular pattern, whereas it was expressed intensely at the apical and basolateral membrane of HCC with a glandular pattern. Immunoblot analysis of 7H6 expression in hepatocellular carcinomas showed a decrease roughly coincident with that shown by immunohistochemistry. These results indicated, for the first time, that tight junctions decrease progressively during carcinogenesis, leading to disruption of cellular polarity and cellular adhesiveness.

The possible mechanism(s) responsible for the different effects exerted by proliferative stimuli of different nature on the appearance of enzyme-altered hepatic foci, were investigated in male Wistar rats. Rats given an initiating dose of diethylnitrosamine (150 mg/kg body weight) were fed a diet containing 0.03% acetylaminofluorene for 2 weeks. Between the first and the second week, cell proliferation was induced by a proliferative stimulus of compensatory type (partial hepatectomy) or by a direct mitogenic stimulus (lead nitrate, 100 mumol/kg). The effect of the two different proliferative stimuli on the appearance of gamma-glutamyl transferase-positive foci was monitored by killing the rats for examination at 1, 2, 3, 5, and 6 days after the induction of cell proliferation. The results indicate that while enzyme-altered hepatocytes can be observed as early as 3 days after partial hepatectomy and are characterized by a rapid growth, direct hyperplasia did not exert any effect on the growth capacity of initiated cells. No effect of lead nitrate-induced hyperplasia was observed following three administrations of the mitogen. When platelet-poor plasma taken from animals exposed to the different proliferative stimuli was tested in primary cultures of hepatocytes, it was found that it induced a significant increase in the labeling index of normal hepatocytes. However, while serum taken 6 days after partial hepatectomy was still able to induce a significant increase in the labeling index, platelet-poor plasma from lead-treated rats had lost part of its effect at 5 days after treatment. The inability of direct hyperplasia to stimulate the development of enzyme-altered hepatic foci was not unique to lead nitrate since the same phenomenon was observed when three other hepatomitogens, nafenopin, cyproterone acetate, and ethylene dibromide, were used.

Development of preneoplastic lesions in the rat liver under the influence of various modifiers was investigated with particular attention to changes in simultaneous expression of altered enzyme phenotype within the lesions (conformity) and proliferation potential. Degree of conformity of marker enzymes such as glutathione S-transferase placental form (GST-P), glucose-6-phosphate dehydrogenase (G6PD), glucose-6-phosphatase, adenosine triphosphatase and gamma-glutamyltranspeptidase was compared with levels of 5-bromo-2-deoxyuridine labeling. After initiation with diethylnitrosamine, rats were administered the hepatopromoter sodium phenobarbital (PB, 0.05%), the antioxidant ethoxyquin (EQ, 0.5%), or a peroxisome proliferator, clofibrate (CF, 1.0%) or di(2-ethylhexyl)-phthalate (0.3%) and killed at week 16 or 32. The PB promoting regimen was clearly associated with increase in the numbers of high conformity class lesions simultaneously expressing three to five enzymes, and elevated proliferation potential. The inhibitor, EQ, in contrast, brought about a time-dependent decrease in conformity so that only 1 or 2 alterations were most commonly observed at week 32. Lesion populations in the peroxisome proliferator- and especially CF-treated cases were characterized by obvious dissociation between degree of conformity and proliferative status. Such treatment-dependent differences were not always correlated with the size of the lesion. The results thus suggested that the conformity and proliferation potential of preneoplastic lesions are dependent on modification treatment. Overall, GST-P was found to be the most reliable marker, although G6PD was less influenced in the peroxisome proliferator cases.

The inhibitory effects of a combination of Antineoplaston A-10 Injection with a small dose of cis-diamminedichloroplatinum (CDDP) on cell and tumor growth was tested in vitro and in vivo settings. A human hepatocellular carcinoma cell line (KIM-1) was used for the cell growth and transplanted tumor growth studies. In the cell growth study, one-hour exposure of KIM-1 cells to CDDP in the medium at concentrations of 0.5, 1.0, and 2.0 micrograms/ml inhibited cell growth dose-dependently. Continuous exposure of cultured cells to Antineoplaston A-10 Injection at concentrations of 4, 6, and 8 mg/ml also inhibited tumor growth dose-dependently. The combination of 0.5 microgram/ml CDDP and 6 mg/ml A-10 Injection inhibited cell growth more than did each agent individually. Electron microscopic study showed well-maintained organelle structures in Antineoplaston A-10 Injection-treated cells compared to CDDP-treated cells. alpha-Fetoprotein (AFP) production by 10(4) cells in 48 h increased in the A-10 Injection-treated and A-10 Injection+CDDP-treated groups as the concentration of these agents increased. In the tumor growth study, daily administration of Antineoplaston A-10 Injection 75 mg with once a week administration of 20 micrograms of CDDP for 5 weeks inhibited transplanted tumor growth in athymic mice after 33 days of treatment, while administration of 75 mg of A-10 Injection or 20 or 60 micrograms of CDDP alone showed no significant inhibition of tumor growth.

We have developed transgenic mice that inherit albumin promoter-regulated simian virus 40 (SV40) large T antigen gene, expressed specifically in hepatocytes. These mice all develop multifocal hepatocellular carcinomas at around 5 months and die of liver insufficiency by 7 months. Sequential morphological observation of hepatocarcinogenesis revealed 5 distinct stages: (I) newborn to 2 weeks of age, neither recognizable histological changes nor cellular replication in spite of T antigen expression; (II) between 3 and 7 weeks, diffuse cytomegalic change of hepatocytes with numerous abnormal mitoses, usually resulting in cell death; (III) from 7 weeks onwards, quasi-regenerative small hepatocyte foci with a decreased tendency for cytomegaly in spite of T antigen expression, rapidly replacing the hepatic tissue; (IV) 11 weeks of age and thereafter, neoplastic foci and nodules with enzymatic alteration; (V) 20 weeks of age and thereafter, gross hepatocellular carcinomas with occasional pulmonary metastases. Considerable variation existed both in morphological and enzymatic features and T antigen expression among neoplastic lesions, including carcinomas. Thus, these transgenic mice clearly show a multistep process in hepatocarcinogenesis with remarkable synchrony and provide a promising model for analyzing the essential events of carcinogenesis at different stages.

The inhibitory effect of sarcophytol A, a cembrane-type diterpene isolated from a marine soft coral, Sarcophyton glaucum, on development of spontaneous hepatomas was investigated in C3H/HeNCrj mice. A total of 80 mice were divided equally into two groups. The experimental and control groups were given basal diets with and without 0.01% sarcophytol A, respectively. At week 65 of the experiment, mice were examined for hepatomas. The percentages of hepatoma-bearing mice of the subgroup with three or more tumors and the tumor diameters of the group treated with sarcophytol A were smaller than those of the control group. Ridit analysis revealed that these differences were statistically significant. The body weight gain, and the food intake were not significantly different between these two groups. Analysis of blood serum revealed that feeding the diet containing 0.01% sarcophytol A for 65 weeks did not show any adverse effects. These results suggest that sarcophytol A inhibits the development of spontaneous hepatomas without toxicity, and should be considered as a possible cancer chemopreventive agent for hepatomas in humans.

Hydroquinone (HQ) was administered to F344 rats and B6C3F1 mice of both sexes at a level of 0.8% in the diet for two years. This treatment induced renal tubular hyperplasia as well as adenomas, predominantly in males of both species, and was associated with chronic nephropathy in rats. In addition, the occurrence of epithelial hyperplasia of the renal papilla was increased in male rats. Foci of cellular alteration of the liver were significantly reduced in number by HQ in rats, but in contrast, were increased in mice, where development of hepatocellular adenoma was also enhanced in males. The incidence of squamous cell hyperplasia of the forestomach epithelium was significantly higher in mice of both sexes given HQ than in the controls, but no corresponding increase in tumor development was observed. The present study strongly indicates potential renal carcinogenicity of HQ in male rats and hepatocarcinogenicity in male mice. Thus, it is possible that HQ, which is present in the human environment, may play a role in cancer development in man.

To elucidate the significance of angiotensin II (AII)-induced hypertension chemotherapy, changes of tissue blood flow both in normal subcutis and in tumors (AH109A, LY80) were measured with the hydrogen gas clearance method. A newly-developed anesthetic machine was used to keep the animals' condition constant. Tissue blood flow in normal subcutis and tumors always fluctuated with time under normotension. The nature and the rate of fluctuation in tumor blood flow were almost identical in two different types of tumors. However, the fluctuation of blood flow in tumor and that in normal subcutis were almost always inversely related when blood flows in these different tissues were measured simultaneously, i.e., when tissue blood flow in normal subcutis decreased, tumor blood flow increased, and vice versa. The findings supported the idea that the connection mode between the tumor vascular bed and normal vascular bed is a parallel circuit. Vascular resistance in the normal vascular bed under AII-induced hypertension seemed to be greater than that under normotension, because the AII-increased tumor blood flow always exceeded the maximum tumor blood flow under normotension. Due to the fluctuations of tumor blood flow, no-flow or low-flow areas, resistant to delivery of anti-cancer drugs, moved sporadically within the tumor under the normotensive condition. However, good conditions for drug delivery to tumor tissue were induced by AII-induced hypertension.

The relationship between tumor invasiveness in vitro and methylation of plasma membrane phospholipids was investigated. For this purpose, two hepatoma cell lines, C1-30 and LC-AH, were used which show specific penetration to below cultured monolayers of mesothelial cells from rat mesentery and endothelial cells from calf pulmonary artery, respectively. Methylthiodeoxyadenosine (MTA) and five of its analogs, difluoro-MTA, deoxyadenosine, sinefungin, phenylthiodeoxyadenosine and fluorophenylthiodeoxyadenosine, inhibited the invasion of the tumor cells without affecting their proliferation. This inhibition was associated with reduction in the incorporation of radioactivity of [methyl-3H]methionine into cellular phosphatidylethanolamine derivatives without changes in the labelings of RNA and DNA and carboxylmethylation of protein. These compounds also decreased the membrane fluidity of the tumor cells, measured by a steady-state fluorescence polarization method. Three other MTA analogs (fluorodideoxyadenosine, fluoroazidodideoxyadenosine and fluoroaminodideoxyuridine) did not affect the invasiveness of the tumor cells or alter their phospholipid methylation or membrane fluidity at concentrations that did not inhibit proliferation. These results suggest that the decrease in invasiveness of tumor cells by MTA and its analogs is due to alterations in the phospholipid composition and fluidity of the tumor cell membranes.

Modifying potentials of various chemicals on tumor development were investigated in a wide-spectrum organ carcinogenesis model using male F344/DuCrj rats. The animals were treated with N-nitrosodiethylamine (100 mg/kg body weight, ip, single injection at the commencement of the study), N-methyl-N-nitrosourea (20 mg/kg body weight, ip, 4 times during weeks 1 and 2) and N-bis(2-hydroxypropyl)nitrosamine (0.1% in drinking water, during weeks 3 and 4) for multi-organ initiation and then were given one of 14 test chemicals including 6 hepatocarcinogens, 7 non-hepatocarcinogens and 1 non-carcinogen, or basal diet for 16 weeks. All rats were killed at the end of week 20, and the major organs were carefully examined for preneoplastic and neoplastic lesions. Immunohistochemical demonstration of glutathione S-transferase-positive foci was also used for quantitative assessment of liver preneoplastic lesion development. Modifying effects were shown for 11 out of 14 test agents in the liver, forestomach, glandular stomach, lung, urinary bladder or thyroid, 7 of them targeting more than two organs. This was the first demonstration to our knowledge that clofibrate possesses enhancing potential for urinary bladder carcinogenesis and an inhibiting effect on thyroid carcinogenesis. Caprolactam showed no effect in any organ, in agreement with its established inactivity. The results indicated that the system could be reliably applied as a medium-term multiple organ bioassay for assessment of the modification potential of test agents in unknown target sites.

The effect of fetal calf serum (FCS) on in vitro invasion by rat ascites hepatoma cells (AH130) was studied by using the in vitro invasion assay. Although the coculture of the highly invasive clone (MM1) of AH130 cells and the mesothelial cell layer or endothelial cell layer in modified minimum essential medium supplemented with 10% FCS resulted in extensive penetration of the layer by the tumor cells, the omission of FCS resulted in an almost complete elimination of the in vitro invasion. The in vitro invasiveness by human small cell lung cancer cells (OC10) was also remarkably reduced by the omission of FCS from the assay medium, suggesting a requirement of serum for the in vitro tumor cell invasion. When 10% FCS was added to the medium 2 h after the tumor cell seeding in FCS-free invasion assay system, penetration by MM1 cells was observed within an hour. This rate of penetration was almost the same as that when 10% FCS was added at the time of tumor cell seeding. FCS was also required for the penetration of a mesothelial cell monolayer by MM1 cells in a defined growth medium (SFM-101), in which MM1 cells were well maintained. The invasion-inducing activity appears to be independent of the growth-stimulating activity in serum.

To investigate molecular responses to lipid peroxidative stimuli in neoplastic cells, lipid peroxidation was induced in liver of rats bearing 3'-methyl-4-dimethylaminoazobenzene-induced hepatocellular carcinoma by injecting a high dose of carbon tetrachloride (CCl4), a strong lipoperoxidative reagent. Normal rat livers with or without CCl4 treatment served as controls. CCl4 administration markedly provoked fatty metamorphosis, visualized by oil red O staining, in normal livers while minimal fatty changes were seen in hepatocellular carcinomas, where necrosis was often observed instead. After CCl4 treatment, the thiobarbituric acid values (representing levels of lipid peroxides in the tissue) were increased two-fold in the untreated normal liver, but were unchanged in the cancer tissue. Levels of vitamin C, an acutely reactive antioxidant, measured by high-performance liquid chromatography were not influenced by the CCl4 injection in the cancer tissue whereas a significant decrease was evident in normal livers. The total fatty acid content, measured by gas chromatography, was significantly lower in the cancer tissue than in the normal liver while the ratio of polyunsaturated fatty acids (PUFAs) in total fatty acids was little changed. Resistance of hepatocellular cancer cells to fatty metamorphosis and their susceptibility to necrosis induced by free radicals may be due to the paucity of the target PUFAs in their cell membrane fraction, resulting in low levels of lipid peroxides. Peroxidation of PUFAs might act as a "shock absorber" against free radical-induced toxic cell death in normal cells.

This study investigated the histological distribution of argyrophilic cells in experimental hepatic neoplasms, the number of these cells, and the proportion of neoplasms with such cells. Seventy 6-week-old male Donryu rats were given a 0.06% 3'-methyl-4-dimethyl-aminoazobenzene (3'-MeDAB) diet for 10 weeks, followed by an ordinary diet for an additional 10 weeks. Of the 70 rats, 50 were used for this investigation; 29 had hepatic tumors, 18 had cholangiofibrosis, and the other three had oval cell proliferation only. Hepatic tissues were stained with Grimelius and Fontana-Masson stains as well as routine hematoxylin-eosin stain. Argyrophilic cells were found in the hepatic neoplasms of 8 rats without argentaffin cells, while cholangiofibrosis was associated with argentaffin cells in almost all cases. Of the 8 rats with argyrophilic cells, three had an abundant population of these cells. The argyrophilic cells were found in areas of the neoplasms with a glandular, trabecular, tubular, or poorly differentiated pattern. Electron microscopy revealed that the neoplastic cells with a positive argyrophil reaction contained small round electron-dense endocrine granules. In addition, in the areas of cholangiofibrosis, two different types of gut endocrine cells were present (G and EC cells). These results suggest that 3'-MeDAB might induce hepatic carcinoid under certain conditions, though we have yet to confirm the development of a pure hepatic carcinoid due to this substance.

Effects of age on the induction of glutathione S-transferase placental form (GST-P)-positive hepatic foci in rats were examined using a medium-term liver bioassay system (for carcinogens). F344 male rats aged 6, 26 and 46 weeks were initially given a single intraperitoneal injection of diethylnitrosamine (DEN, 200 mg/kg) and, beginning 2 weeks later, received 0.02% 2-acetylaminofluorene (2-AAF), 0.05% phenobarbital (PB) or 1.3% acetaminophen (AAP) in the diet for 6 weeks. All animals were subjected to two-thirds hepatectomy 3 weeks after the DEN injection and were killed at week 8. Quantitative analysis of GST-P-positive foci revealed significantly (P less than 0.001) increased induction over control levels in terms of both numbers and areas for 2-AAF at all ages (6, 26 and 46 weeks), but especially in the 6-week-old case. In the PB- and AAP-treated groups, the respective enhancing and inhibitory influences were most pronounced in the animals aged 6 weeks, and were less marked in older rats. Thus, the response of F344 rats to the modifying effects of chemicals was age-dependent, the conclusion being drawn that young rats are more susceptible and therefore more appropriate for assessment of carcinogenic, promoting and inhibitory effects of chemicals.

Angioarchitectures of ascites hepatoma AH109A and Sato lung carcinoma (SLC) were quantitatively compared by measuring the following morphometric parameters: vascular density, vascular length, distance between tissues and their nearest blood vessel, and total length of microvascular network per unit area. When the vascular networks in these two types of tumors were compared in the initial stage, the morphological parameters were almost identical. Correlations between tumor size and the number of starting vessels and between enlargement of the tumor and the ensuing increase in pressure of the starting vessel were also evaluated with a microcomputer and an apparatus for measuring microvascular pressure. The total length of tumor vascular network to which one starting vessel supplied blood increased exponentially as the tumor increased in size exponentially. There was a positive correlation between tumor size and the number of starting vessels. The range of the blood supply from one starting vessel was evidently limited. The pressure of the starting vessel increased with enlargement of the tumor size. As soon as the pressure of the starting vessel reached a plateau, however, there was a rapid increase in low-flow or no-flow areas in regions within the tumor. From the results obtained, we consider that low-flow or no-flow areas, resistant to delivery of anticancer drugs, inevitably appear with the progression of tumor growth.

Captafol was administered at dietary levels of 0 (control), 750 and 1,500 parts per million (ppm) to groups of 50 male and 50 female F344/DuCrj rats for 104 weeks, and then all animals were maintained without captafol for a further 8 weeks, and killed in week 113. Renal cell carcinoma was found in eight of 50 male rats treated with 1,500 ppm and in one of 50 male rats treated with 750 ppm of captafol. The incidences of renal adenomas, including micro-adenomas, and basophilic altered cell tubules were significantly higher in both sexes treated with captafol than in controls, and the increases were apparently dose-dependent except that of adenomas in females. The incidences of neoplastic and preneoplastic lesions of the kidney in captafol-treated animals were higher in males than in females. Captafol also induced hepatocellular carcinomas in four of 50 female rats in the 1,500 ppm group. The incidences of hyperplastic (neoplastic) nodules and foci of cellular alterations in the liver were also significantly increased in both sexes treated with captafol, the increases being dose-dependent. In conclusion, captafol induced renal cell carcinomas in male rats and hepatocellular carcinomas in female rats.

A highly reproducible model of a solitary hepatic tumor in rats using ascites hepatoma AH13 has been developed using a two-step method which was suitable for quantitative chemotherapeutic studies. Diffuse hepatic metastases were induced first by inoculation of three different ascites hepatomas, AH13, AH130 and AH7974 into the portal vein in a dose-dependent fashion. Second, the induced hepatic tumor (3 x 10(7) cells) was minced into 1 x 1 x 4 mm fragments and implanted in the liver of normal rats. In this procedure, the AH13 strain proved best suited for the generation of a solitary hepatic tumor. The growth of the solitary liver tumor using AH13 was highly reproducible. To demonstrate the suitability of this solitary hepatic tumor model for the evaluation of chemotherapy, the tumor-burdened rats were treated with adriamycin (ADR) and mitomycin C (MMC). The reduction in tumor size was proportional to dosage, and the statistical significance of the differences between the treatment group and control group was proportional to dosage. A synergistic effect of ADR and MMC on the tumor also was demonstrated. This model should prove to be a useful tool for the testing of newly developed treatments of hepatic cancer.

The three-day-old larvae of self-fertilizing hermaphroditic fish Rivulus ocellatus marmoratus were fed a diet containing the antioxidant butylated hydroxyanisole (BHA) at levels of 0.01-0.8% for 12 days. Six months after BHA administration, hepatic tumors were found in all groups of BHA-treated fish. The BHA-induced tumor incidences were clearly dose-dependent. These results show that dietary BHA is hepatocarcinogenic in Rivulus even at 0.01% dose.

By using transparent chambers in rats, we have directly observed tumor-induced neovascularization in the early stage and the formation of intricate networks in Yoshida rat ascites hepatoma AH109A and Sato lung carcinoma at high magnification. We counted branching point numbers per unit area in the microvascular network with and without tumors in order to clarify the sites from which new vascular sprouts originate. Branching point number per unit area in normal tissue was 13.6 +/- 7.4/0.1 mm2 in the field near a terminal arteriole, and 12.9 +/- 7.3/0.1 mm2 in the field distant from a terminal arteriole. There was no significant difference between these two fields in the normal vascular network. On the other hand, in the tumor vascular network, the branching point number in the field near a terminal arteriole was 50.4 +/- 12.6/0.1 mm2, and 30.1 +/- 11.5/0.1 mm2 in the field distant from a terminal arteriole. The difference is highly significant (P less than 0.001). The frequency with which new capillaries originated from veins and venules was very low. We concluded from these results that the position from which tumor vessels originated was usually the terminal portion of a terminal arteriole.

The carcinogenic potential of 2-amino-3-methylimidazo[4,5-f]quinoline (IQ) was evaluated in cynomolgus monkeys. Monkeys received IQ, beginning at the age of one year, at doses of 10 or 20 mg/kg by gavage. Thus far, IQ has induced hepatocellular carcinoma in three monkeys with a latent period of 27 to 37 months. Metastases to the lung occurred in two of the three monkeys. Microscopically, the hepatocellular carcinoma in all three cases demonstrated a trabecular pattern. These data demonstrate that IQ is a potent carcinogen in nonhuman primates and support the idea that it is a potential carcinogen for humans.

A remarkably high incidence of hepatocellular carcinomas was observed in long-surviving LEC rats with hereditary hepatitis. Among the 60 LEC rats examined between 12 and 28 months of age from F29 and F30, 55 (92%) developed putative preneoplastic and neoplastic lesions such as hyperplastic foci and nodules, and hepatocellular carcinomas. Of these, hepatocellular carcinomas were observed with a high frequency (46/55; 84%). All rats of advanced age that survived more than 18 months developed hepatocellular carcinomas. These results suggest that the development of liver tumors in LEC rats is an age-associated phenomenon with serial hepatic alterations after the subsidence of acute hepatitis. The long-surviving rats had no normal tissue and showed chronic hepatitis in nontumorous tissues of the liver. Cholangiofibrosis was also found in most rats with hepatic lesions. Metastasis of hepatocellular carcinomas was found in four rats. Histologically, the hepatocellular carcinomas were of a well-differentiated type with a typical trabecular structure. Thus, LEC rats seem to be a promising animal model for studying the pathogenesis of hepatitis and hepatocellular carcinoma.

To elucidate the role of oncogene expression in hepatocarcinogenesis, we examined the expression of 4 cellular oncogenes (c-myc, c-fos, Ha-ras and c-erbA) in liver tissues induced by chemical agents. Four groups of male Sprague-Dawley rats were examined in the present study. Rats of the first and second groups were given a single intraperitoneal injection of diethylnitrosamine (DEN), 200 mg/kg body weight. Two weeks later, these rats were divided into two groups; the DEN-C group received no further medication, whereas the DEN-DES group was given diethylstilbestrol (DES), 0.5 mg/day, for 12 months. The DEN group was given DEN, 100 ppm, in drinking water for five months as the hepatocellular carcinoma (HCC) group. The DES group was given DES, 0.5 mg/day, from the start for 8 months. Rats of the DEN-DES and DEN groups developed grossly visible hepatic tumors. Significantly higher levels of c-myc gene expression were observed in tissues of HCC of the DEN group and in neoplastic nodules of the DEN-DES groups than in the DES and DEN-C group. The increase of c-myc mRNA seemed to begin after 1 month of treatment and became significant at 4 months in the DEN-DES group. On the other hand, no significant differences in mRNA levels of c-fos, Ha-ras and c-erbA were observed among these four groups. Although the significance of increased c-myc gene expression in neoplastic liver is still not known, it is conceivable that the persistent elevation of c-myc gene expression in the DEN and DEN-DES groups might contribute to the development of rat chemical hepatotumorigenesis.

The carcinogenicity of chrysazin (1,8-dihydroxy-9,10-anthracenedione) was examined by dietary administration to C3H/HeN mice. All of the effective mice (17) which were given 0.2% chrysazin diet and which survived more than 500 days developed adenomatous hyperplasia with cystic glands of the cecum. Similar lesions were also seen in the colon of mice in this group. These intestinal lesions were not obtained in any effective mouse (19) of the control group. The incidence of hepatocellular carcinoma of mice given chrysazin (4/17) was significantly higher than that of the controls (0/19). These results indicate that chrysazin is carcinogenic in mice as well as in rats. Some mechanistic aspects of the causation of these intestinal lesions and liver neoplasms are also discussed.

Three inhibitors of S-adenosylmethionine-mediated transmethylation, 5'-methylthioadenosine (MTA), 2'-deoxyadenosine and sinefungin, inhibited in vitro invasion by a highly invasive clone (Cl-30) of rat ascites hepatoma cells, AH 130 (AH cells). Difluoromethylthioadenosine (DFMTA), a non-metabolizable derivative of MTA, also caused strong inhibition of invasion at concentrations that did not suppress the growth of the tumor cells. Cl-30 cells precultured in methionine-depleted medium showed decreased invasiveness. DFMTA was also effective on the invasion by fibrosarcoma, B16 melanoma and human lung carcinoma cell lines.

Three hundred and ninety-five four-week-old SPF female ddY mice were each exposed to 5 or 6 cercariae of Schistosoma japonicum (Japanese strain) on their shaved abdomens and were maintained in a conditioned clean environment and fed on sterilized food and water. Fecal examinations at 8 to 10 weeks postinfection (PI) revealed 169 mice to be infected. More than half of them died within 30 weeks PI and 70 mice that survived to the 50th week PI were sacrificed. At autopsy, we could find no schistosome eggs in the liver or intestinal wall of 9 mice, and they were excluded. Out of 61 mice which showed S. japonicum eggs in their livers, 48 had single or multiple hepatoma, while no tumor was observed in the livers of the 60 control mice. The tumors were yellowish-white in color with distinct boundaries and the centers of the tumors were depressed in some cases. The size of the tumors varied from 1 to 20 mm in diameter. Most of the tumors retained the normal trabecular pattern, but in some cases the trabeculae were thickened, having wide vascular spaces. The tumor cells were PAS-negative and showed varieties of pleomorphism. The sizes of cells and nuclei varied greatly. These findings suggested some causal relationship between S. japonicum infection and the hepatoma formation in the host's liver. In the chronic course of schistosomiasis japonica in the endemic areas, S. japonicum infection probably plays a role in hepatoma formation of patients.

An antitumor principle (SAGP) has been purified from cell-free crude extract (CE) of group A Streptococcus hemolyticus, Su strain. The antitumor activity of each fraction was evaluated by measuring the in vitro growth inhibitory effect on transformed hamster embryonic lung cells (THEL). CE was subjected to thermal treatment, streptomycin precipitation, ammonium sulfate precipitation, and chromatography on octyl-Sepharose CL-4B, DEAE-cellulose, and Sephadex G-200 in that order. The active fraction from the last chromatography was dialyzed against distilled water. The resulting precipitate was removed and the supernatant was lyophilized (SAGP). SAGP is a homogeneous glycoprotein as judged by polyacrylamide gel electrophoresis, gel filtration, immunodiffusion, and PAS-staining. The molecular weight of SAGP was determined to be 140,000 to 150,000 by the gel filtration technique. SAGP is composed of subunits, each of which has a molecular weight of about 50,000. The isoelectric point of SAGP is 4.3. Amino acid analysis revealed an abundance of aspartic and glutamic acid residues in SAGP. The 50% growth inhibitory dose of SAGP on THEL was 0.062 micrograms/ml. Intraperitoneal administration of SAGP (20 mg/kg/day X 4) to ICR mice bearing Ehrlich ascites carcinoma cells increased their life span to 254% of the control.

The in vitro invasive capacity of poorly invasive cells (W1), which were cloned from rat ascites hepatoma cells (AH 130), was potentiated dose- and time-dependently by pretreating the cells with transforming growth factor-beta (TGF-beta). This potentiation of invasive capacity was completely abolished by anti-TGF-beta antibody. When the treated cells were ip inoculated into rats, the cells extensively invaded the peritoneum, and formed penetrating tumor nodules. The effect of TGF-beta was reversed by subculturing the treated cells without TGF-beta. The potentiation of invasive capacity by TGF-beta might participate in platelet-associated enhancement of tumor cell metastasis.

Male Sprague-Dawley rats were given a single intraperitoneal injection of diethylnitrosamine (DEN, 200 mg/kg body weight). Two weeks later the rats were divided into 4 groups; DEN-C group rats were given no further treatment; DEN-DES group rats were fed diethylstilbestrol (DES, 0.5 mg/day); DEN-TMX group rats were given tamoxifen (TMX, 1.0 mg/day) orally; DEN-DES TMX group rats were fed both DES and TMX for 8 months. Rats of the DEN-DES group developed grossly visible hepatic tumors. On the other hand, tumor development was significantly inhibited in rats of the DEN-DES TMX group. Total area of gamma-glutamyl transpeptidase-positive lesions and the mean area per lesion were significantly larger in rats of the DEN-DES group than those of the DEN-C, DEN-TMX or DEN-DES-TMX group. Estrogen receptor (ER) content of liver cytosol assayed by enzyme immunoassay (EIA) was significantly greater in rats of the DEN-DES group than in those of the DEN-C group and smaller in rats of the DEN-TMX and DEN-DES TMX group than in the DEN-C group. On the contrary, ER content of liver nuclei was significantly greater in rats of the DEN-TMX and DEN-DES TMX group than in those of the DEN-C or DEN-DES group. These results suggest that the promotive action of DES and the inhibitory action of TMX on DES-promoted hepatic tumorigenesis are, at least in part, mediated by ER in the rat.

The effect of dietary benzylselenocyanate (BSC), a novel organoselenium compound and its sulfur analog, benzylthiocyanate (BTC), on hepatocarcinogenesis induced by azoxymethane (AOM) was investigated in male F344 rats. Eighty-one weanling rats were divided into 3 groups and were raised on a semipurified diet (control diet). Starting from 5 weeks of age, groups of animals consuming the control diet were fed one of the experimental diets containing 25 ppm BSC or BTC. An additional group was continued on the control diet. At 7 weeks of age, animals were given weekly sc injections of AOM (15 mg/kg body weight once weekly for 2 weeks). One week after the second AOM injection, those groups receiving BSC and BTC diets were transferred to the control diet and continued on this diet until termination of the experiment at 34 weeks after the last AOM injection. For quantitative analysis of enzyme-altered liver cell foci, glutathione S-transferase placental form was stained by an immunohistochemical technique. The results indicate that the incidence and the density of the enzyme-altered foci were significantly lower in AOM-treated rats fed the diet containing 25 ppm BSC (foci incidence 56%, foci density 2.43/cm2) than in AOM-treated animals fed the control diet (foci incidence 92%, foci density 4.79/cm2). The incidence of small altered foci was significantly inhibited in rats fed the BTC diet (35%) as compared to those fed the control diet (68%), but the degree of inhibition was more pronounced in animals fed the BSC diet than in those fed the BTC diet.