A study of cyclic nucleotide metabolism and the histology of rat liver during 3'-methyl-4-dimethylamino-azobenzene carcinogenesis. II. Cyclic AMP metabolism.

We have studied cAMP metabolism in rat livers undergoing carcinogenesis induced by dietary 3'-methyl-4-dimethylaminoazobenzene. A correlation between the biochemical and the histological changes described in the companion paper has been made. In this study, we saw 100% incidence of cholangiocarcinoma by 10 weeks. During weeks 1--10, the biochemistry of tumor-free areas of the livers only was studied; during weeks 11-13, the increased size of the tumors made possible a biochemical study of the tumor tissue as well as the non-tumor tissue, and a comparison between the two was made. Alterations in all parameters of cAMP metabolism were seen from the earliest stages of treatemnt. Most striking were those of adenylate cyclase activity which preceded and accompanied tumor formation, and were seen in both non-tumor and tumor tissue. In the first few weeks of treatment, small acidophilic glycogen-deficient hepatocytes appeared in the periportal areas of the liver lobules. During this time, there was an increase in maximal isoproterenol stimulation of adenylate cyclase and to a lesser extent in the basal activity of the enzyme; increases in phosphodiesterase activity were seen, and were greatest in weeks 1, 2; cAMP levels were diminished in weeks 1, 2 and slightly but not significantly elevated at week 3. From week 4 onwards an even smaller glycogen-deficient cell population appeared in perilobular areas amongst the acidophilic hepatocytes, and tumors began to appear elsewhere in the livers; at this time, there were further marked increases in the basal activity and isoproterenol responsiveness of adenylate cyclase, and the appearance of increased Gpp(NH)p responsiveness of the enzyme; the increase in phosphodiesterase activities seen at week 3 (smaller than that seen in weeks 1, 2) was sustained but did not further increase; cAMP levels were now significantly elevated also, but they did not rise steadily as did the activity of adenylate cyclase. There was a marked difference between the adenylate cyclase activities in non-tumor tissue from tumor-bearing and non-tumor-bearing livers in weeks 4--10, but there was no difference between the phosphodiesterase activities or cAMP levels in these two groups. Adenylate cyclase activity was extremely high in both non-tumor tissue of tumor-bearing livers from weeks 4--10 and tumors from weeks 11--13. Although phosphodiesterase activities were most elevated in the tumors, there were extremely high cyclic AMP levels in these tissues. The difference between the cAMP levels of tumor and non-tumor tissue was striking. Our findings are discussed with respect to the two-state model of carcinogenesis...