In vitro detection of differential and cell-specific hepatobiliary toxicity induced by geldanamycin and 17-allylaminogeldanamycin in rats.

The differential toxicity of two anticancer agents is described using the in vitro rat liver slice culture model. Liver slices from F-344 rats were cultured for 5 days in Waymouth's-based medium with exposure to a range of geldanamycin (GEL) or 17-allylaminogeldanamycin (17-AAG) concentrations. GEL induced concentration-dependent reduction of alkaline phosphatase and of gamma-glutamyl transferase levels, which are indicators of biliary epithelial cell(s) (BEC) viability, and exhibited hepatocellular toxicity at higher concentrations. Histologically, BEC cell injury was evident at the lowest GEL concentration (0.1 microM) and progressed to overt bile duct necrosis at 5 microM, a level at which hepatocellular damage was also more prominent. Slices exposed to the same concentrations were more sensitive to toxic effects of GEL than of 17-AAG. 17-AAG at the lowest concentration had more slice biomarker retention than GEL, and histological analysis revealed minimal toxic effect on BEC. With increasing concentration, BEC were progressively lost, and BEC proliferation was completely inhibited at 5 microM 17-AAG. Hepatocellular injury was evident only at high dose exposures. This is believed to be the first use of an in vitro liver tissue model to accurately predict the differential and concentration-dependent toxicities of these compounds.

Regulation of differentiated phenotype of rat hepatic lipocytes by retinoids in primary culture.

Activation of hepatic lipocytes to myofibroblastlike cells observed in cell culture and during liver fibrogenesis is characterized by an increase in collagen formation and cell proliferation. These changes appear to be associated with the loss of intracellular retinoid in lipocytes, the principal storage site for vitamin A in the body. To evaluate whether retinoids have the capability to suppress lipocyte activation, we exposed cultured lipocytes in both native and myofibroblastlike states to retinoids and determined their effects on collagen production, intracellular retinoid level, and cell proliferation. Retinol (1 microM) and retinoic acid (1 microM) supplementation of primary rat lipocyte cultures inhibited the spontaneous increase in collagen synthesis associated with lipocyte activation; lower concentrations of retinol (10 or 100 nM) were also effective. Simultaneously, retinol addition prevented a precipitous decline in intracellular retinoid content in the absence of added retinoid. These retinoid effects were reversed by a change to unsupplemented control medium. When cells in the myofibroblastlike state were exposed to retinol (> or = 1 microM), a significant increase in intracellular retinoid levels and reduction in collagen synthesis occurred. Lipocytes in both native and myofibroblastlike states secreted four to five times higher amounts of type I collagen than type III collagen, but retinol and retinoic acid particularly inhibited production of type I collagen. Cell proliferation measured by [3H]thymidine incorporation was also inhibited by retinol. These results demonstrate that extracellular retinoids suppress lipocyte-activated collagen synthesis and cell proliferation and support the interpretation that retinoids themselves are regulatory factors in maintenance of the lipocyte in its native, differentiated state.

Cyanide-induced cytotoxicity to isolated hepatocytes.

The responses of various cell functional parameters to potassium cyanide (KCN) were investigated in isolated rat hepatocytes to examine their quantitative and temporal relationship to inhibition of mitochondrial respiration. Both cultures and suspensions of hepatocytes, maintained in hormone-supplemented culture medium at 37 degrees C under an air: CO(2) (95:5) atmosphere, were used in these studies. The earliest and most sensitive change detected was inhibition of O(2) consumption (EC (50) = 78 mu m ); other early changes included depression in ATP and ATP/ADP, inhibition of urea synthesis, and elevation in lactate/pyruvate. The effect on ATP depression at 10 min was reversed by replacement with fresh medium containing no cyanide (t (1 2 )=9 min ). Increased release of lactate dehydrogenase and acid phosphatase occurred much later during the incubation (120 or 240 min) and at much higher KCN concentrations (>/=0.5 mm) than the other changes. These observations are consistent with inhibition of mitochondrial respiration being the initiating or major contributing factor in cyanide-induced cytotoxicity in hepatocytes, but because of the difference in EC(50) values for depression of O(2) consumption and energy-dependent parameters and for intracellular enzyme release, other mechanisms may also contribute to cell death. Reduced glutathione content and lipid peroxidation in the cells, assessed by measuring thiobarbituric acid reactants, were not significantly changed, and the mechanism leading up to cell death remains uncharacterized. The cyanide concentrations in vitro that produced inhibition of O(2) consumption were in the same range as those in blood and liver that inhibited cytochrome oxidase in vivo at lethality. The results in the hepatocyte model, therefore, can explain why the liver is not a target organ in massive, acute cyanide poisoning, in that overt signs develop at higher concentrations and after longer times than are applicable for the death of the animal.

Subchronic inhalation toxicity of 1,3,5-trichlorobenzene.

Male and female rats were exposed to 0, 10, 100 or 1000 mg/m3 of 1,3,5-trichlorobenzene vapors for 6 hours daily, 5 days a week, for up to 13 weeks. After 4 and 13 weeks of exposure, animals were sacrificed and examined for changes in blood, clinical chemistry, internal organs, and tissues resulting from the 1,3,5-trichlorobenzene treatment. No treatment-related effects on the blood and clinical chemistry were evident. The only effects that were considered treatment-related were a squamous metaplasia and hyperplasia in the respiratory epithelium in the nasal passages of high-dose rats and the increased incidence of dried red material on the faces of these rats during exposures compared with other groups.

Evaluation of hepatocytes isolated by a nonperfusion technique in a prescreen for cytotoxicity.

Twenty-three chemicals, differing widely in cytotoxic (hepatotoxic) potency in vivo, were examined to determine their ability to release glutamic-oxaloacetic transaminase (GOT) from hepatocytes isolated by a nonperfusion method from rat liver. The test chemicals were carbon tetrachloride, chloroform, 1,1,2- and 1,1,1-trichloroethane, six bromobenzene analogs, tri-n-butyl tin, chlorpromazine, tetracycline, halothane, phenobarbital, L-ethionine, acetaminophen, thioacetamide, allyl alcohol, ethanol, ascorbic acid, dimethyl sulfoxide, and acetone. In all but two cases--thioacetamide and allyl alcohol--there was a good correspondence between chemicals active in the assay as now performed and those that elevate serum transaminase and cause liver injury on short-term exposure in vivo. These results indicate that with further effort it may be possible to develop an effective, inexpensive, and rapid prescreen to identify drugs and environmental chemicals that are potentially cytotoxic to animals and humans.