Search

Continuous feeding of alpha-naphthylisothiocyanate to young male Sprague-Dawley rats was shown to produce a concentration-dependent increase in the number of hepatic ductular cells and a concentration- and time-dependent elevation of serum and liver gamma-glutamyl transpeptidase and alpha-fetoprotein. In liver, the increased gamma-glutamyltranspeptidase and alpha-fetoprotein were predominantly confined to the proliferative ductular cell population. It is concluded that early stages of intoxication by the noncarcinogen alpha-naphthylisothiocyanate resemble early stages in induction of liver neoplasia by carcinogens that evoke ductular proliferation. Elevation of gamma-glutamyltranspeptidase and alpha-fetoprotein expression by an expanding ductular cell population characterizes both processes. However, the increase is rapidly reversed after alpha-naphthylisothiocyanate is discontinued, in contrast to the persistence that has been reported when acetylaminofluorene was administered.

The objective of the present study was to define early biochemical changes occuring in livers of rats that were fed various chemical carcinogens. Rats were subjected to partial hepatectomy and subsequently given multiple injections of radioactive thymidine to prelabel DNA in their liver. Following a 4-week recovery period the rats were placed on either basal diets or diets containing either 0.05% 3'-methyl-4-dimethylaminoazobenzene (3'-MeDAB), 0.028% 2-acetylaminofluorene, or 0.05% 2-methyl-4-dimethylaminoazobenzene for various periods. After 5 weeks 3'-MeDAB had caused a dose-dependent loss of prelabeled DNA demonstrating the cytotoxicity of this carcinogen. The comparatively noncarcinogenic 2-methyl-4-dimethylaminoazobenzene caused only a small loss of prelabeled DNA. In contrast, the hepatocarcinogen 2-acetylaminofluorene did not cause a loss of prelabeled DNA, demonstrating its low cytotoxicity. Autoradiography and histology revealed that the loss of prelabeled DNA in livers of rats fed 3'-MeDAB was largely due to parenchymal cell death. Experiments designed to separate liver regenerative hyperplasia from neoplastic hyperplasia revealed the presence of both an early and a delayed elevation of thymidine incorporation into liver DNA in rats fed 0.05% 3'-MeDAB. An "early" elevation of incorporation occurred during and shortly after 3'-MeDAB feeding, and a "delayed" elevation of incorporation occurred some weeks after the dye was discontinued. Autoradiography revealed that parenchymal cells were largely responsible for the increased incorporation. Feeding of 2-methyl-4-dimethylaminoazobenzene depressed thymidine incorproation. A direct comparison of the effects of isomolar levels of 3'-MeDAB and 2-acetylaminogluorene on hepatic hyperplasia indicated that both carcinogens caused comparable increases in thymidine incorporation, which returned to control levels upon feeding of carcinogen-free diet. The differences and similarities between the responses to the three compounds are discussed and considered with regard to initiation and promotion of hepatoma formation.

The mathematical science of quantitative stereology has established relationships for the quantitation of elements in three-dimensional space from observations on two-dimensional planes. This report describes the utilization and importance of such mathematical relationships for the quantitative analysis of focal hepatic lesions in terms relative to the volume of the liver. Three examples are utilized to demonstrate the utility of such calculations in the three-dimensional quantitation of hepatic focal lesions. The first is that of a computer-simulated experiment based on defined hypothetical situations. The simulations demonstrate the applicability of the computations described in this report to the evaluation of two-dimensional data from typical animal experiments. The other two examples are taken from actual experiments and involve the transplantation of hepatic cell populations into the liver suitably prepared hosts and the quantitation of altered foci produced by initiation with diethylnitrosamine-partial hepatectomy followed by promotion with phenobarbital. The quantitation of altered foci by means of a two-dimensional analysis (simple enumeration of focal intersections/area of tissue section) is proportional to the quantitation of foci per volume of liver provided that the mean diameter of the foci for each treatment is sufficiently uniform, as exemplified in the text by the transplantation experiment. When such mean diameters are unequal as in the diethylnitrosamine-phenobarbital experiment described herein, quantitation from three-dimensional analysis gives significantly different results as compared with enumeration of focal intersections on two-dimensional areas. These studies clearly demonstrate that the frequency and size of foci intersections viewed on two-dimensional tissue sections do not necessarily reflect the number of size of foci in the three-dimensional tissue. Only by quantitating the number and size of the foci in relation to the three-dimensional volume of the tissue can one determine the validity of the proportionality of data from two-dimensional measurements to the total number of foci per volume of tissue. Such a conclusion has important implications for quantitative studies on hepatocarcinogenesis as well as for the enumeration of premalignant lesions which occur during the natural history of carcinogenesis in any solid tissue.

Characteristic and patterns of gamma-glutamyltranspeptidase (GGT) expression were studied in four rat and three human hepatoma cell lines. Phenotypic diversity of GGT expression was demonstrated by the following findings. (a) GGT specific activity increased rapidly in three of four rat lines during the first 72 hr after subculture. (b) GGT activity was detected in the fourth rat cell line only from 96 to 120 hr after subculture. (c) In late log or stationary cultures, each of the four rat lines assumed a unique and characteristic level of GGT specific activity. (d) The intracellular GGT distribution pattern was markedly varied in rat and human cell lines. (e) GGT activity was confined to isolated cell clusters in one human line in vitro and one rat line both in vivo and in vitro. And (f) there was poor correlation between GGT specific activity and several liver-associated and hepatoma-associated properties. In contrast to evidence of diversity in GGT expression, GGT was shown to be a nonsecreted protein in all four rat cell lines. The constitutive or autogenous nature of the GGT phenotype in rat hepatoma cells was demonstrated by the retention of the GGT-positive and GGT-negative phenotypes of two strains grown in mixed culture; the lack of change in GGT activity when cells were cultured on different substrata, in different media, or in media containing hormones (insulin, dexamethasone, triiodothyronine, or glucagon); and the assumption of nearly constant levels of GGT specific activity in late log or stationary cultures. The results suggest that GGT activity is expressed in hepatomas as a result of disturbed differentiation and that this expression is not necessarily linked to cell proliferation.