Comparison of the promoting effects of various agents in induction of preneoplastic lesions in rat liver.

The promoting activities of 29 compounds on induction of hyperplastic (neoplastic) liver nodules (HN) and the dose-dependent effects of tumor promoting agents were compared by using a short-term test system developed in this laboratory. In tests on promoting activity, F344 rats were given a single dose (200 mg/kg) of N-nitrosodiethylamine (DEN), and from two weeks later, were treated with various test compounds for 6 or 10 weeks. They were subjected to partial hepatectomy 3 or 4 weeks after DEN treatment. The results showed that strong hepatocarcinogens, such as aflatoxin B1, DEN, N-nitrosodimethylamine (DMN), 2-acetylaminofluorene (2-AAF), 3'-methyl-4-dimethylaminoazobenzene (3'-Me-DAB) and ethionine, induced many hyperplastic liver nodules, whereas dieldrin, 1,1,1-trichloro-2,2-bis(4-chlorophenyl)ethane (DDT), polychlorinated biphenyls (PCB) and alpha-hexachlorocyclohexane (alpha-HCH) induced few lesions. Nonhepatocarcinogens, such as N-nitrosoethylurea (ENU) and 3-methylcholanthrene (3-MC), only slightly induced hyperplastic nodules. Of the miscellaneous compounds tested, phenobarbital, deoxycholic acid and ethynyl estradiol also induced gamma-glutamyl transpeptidase (gamma-GT) positive foci. In tests on the dose-dependent effects of promoting agents, DMN was given at different concentrations for 6 weeks from 2 weeks after DEN treatment. Results were quantitated by histochemical measurement of the number or area of gamma-GT positive lesions induced. A long-term experiment on the effect of feeding DMN for 96 weeks was also done. Clear dose-dependent effects of DMN were seen in induction of gamma-GT positive foci in the short-term experiment and neoplastic lesions in the long-term one.


Introduction
Since the proposal of a two-stage process of chemical carcinogenesis in mouse skin, extensive investigations have been carried out to understand the mechanism of the initiation-promotion process (1). Initiation is thought to introduce neoplastic information into the cells, possibly by some alteration of cellular DNA, whereas treatment with a promoter given continuously over a relatively long period is thought to allow phenotypic expression of the altered genome eventually leading to neoplastic growth. This two-stage process of carcinogenesis was also demonstrated more recently in liver carcinogenesis with 2-acetylaminofluorene (2-AAF) as an initiator and phenobarbital as a promoter (2,3). A similar two-stage mechanism has been proposed for carcinogenesis in urinary bladder (4)(5)(6), mammary gland (7), thyroid gland (8) and intestines (9).
The actual risk to humans of chemicals in the environment is usually the result of multiple exposures to a wide variety of agents, including carcinogens and promoters, and there is increasing evidence that these compound infact contribute to the induction of human cancer. However, it is difficult to test all possible factors in long-term in vivo experiments, and no suitable in vitro short-term test for promoting agents has yet been reported. Thus a short-term in vivo assay for promoting agents is urgently needed. Previous studies on this problem in this laboratory indicated that when rats were pretreated with N-nitrosodiethylamine (DEN) and then exposed to various promoting agents for 6 to 10 weeks, quantitatively detectable numbers of hyper-plastic (neoplastic) nodules (HN) were formed in their liver (10). The HN measured in this experiment included hyperplastic foci (areas of cellular alterations) and hyperplastic nodules (neoplastic nodules) of more than 0.2 mm diameter, as previously described (10). The aims of this study were to oW serve the sequence in which hepatocellular carcinomas develop from early preneoplastic lesions and to define HN; to study the dose-dependent effects in long-term experiments for carcinogenicity and in short-term tests for promoting activity on the incidence of neoplastic and preneoplastic lesions, and to compare the activities of various compounds as promoters of preneoplastic liver lesions in the shortterm model system. The numbers and areas of HN per unit area in microscopic sections of liver are shown in Figures 1  and 2, respectively. In the liver of rats treated with 2-AAF, the number and'area of HN were maximal in week 10, and then gradually decreased to week 50 (p<0.001). However, the area remained almost constant from week 30 to 50. In the group given a-HCH, the number and area of HN gradually increased throughout the experiment. The decrease of HN in 2-AAF treated rats after week 10 was apparently due to their degeneration or disappearance. Therefore, the numbers and areas of degenerated or necrotic nodules which had a spongy or cystic appearance were also measured. The histological features of the lesions, which were diagnosed as degenerated hyperplastic nodules (DHN), were similar to those seen in spongiosis hepatis reported by Bannasch et al. (12). The number and area (not shown) of these degenerated hyperplastic nodules (DHN) increased with time until week 30 and then decreased or remained almost constant in rats given 2-AAF or a-HCH (Fig. 3). The maximal number of hepatocellular carcinoma was reached at week 40 (Fig. 4).   This experiment shows that detectable preneoplastic lesions in the liver, measured as HN, were induced in DEN-treated rats by exposure to promoting agents for 6 weeks in combination with partial hepatectomy. Their number and area started to decrease 2 weeks after the end of exposure to promoting agents (10 weeks after DEN) due to degenerative change.

Dose-Dependent Effects of DEN in Induction of Preneoplastic Lesions of Rat Liver
N-Nitrosodimethylamine (DMN) was given at concentrations of 10.0, 1.0 and 0.1 ppm in the diet for 6 weeks after initiation with DEN to test its dose-dependent effect in promotion (Fig. 5). In this experiment, the number and area of preneoplastic liver lesions giving a positive reaction for y-glutamyltranspeptidase (y-GT) were measured (13). A long-term experiment in which DMN was given continuously in the diet for 96 weeks at the same concentrations as in the short-term test (10.0, 1.0 and 0.1 ppm) was also carried out to examine whether the incidence of y-GT positive foci in the short-term experiment was comparable to that of HN in the long-term experiment. A significant dose-dependent response in the induction of y-GT positive foci in the short-term test and also HN or malignant neoplasias in the long-term test were shown among the higher range of doses (14). In these experiments the results of the short-term experiment clearly reflected those of the long-term one (Tables 1 and 2). Similarly, a clear dose-dependent effect on the enhancement of HN induction was seen with different doses of phenobarbital (15,16). Since most y-GT positive cell foci develop into HN (17), this system for detecting preneoplastic changes seems suitable for analytical studies on promoters or factors that enhance chemical carcinogenesis.   (7) 0 -

Effects of Exogenous and Endogenous Promoting Agents on Induction of Preneoplastic Lesions
In tests on the promoting activities of various compounds, rats were injected intraperitoneally with DEN or saline (control), given test compounds by an appropriate route for 6 weeks and then killed (Fig. 5). One week after the start of treatment with test compounds, a two-thirds partial hepatectomy was performed to stimulate cell proliferation (18). The doses and routes of administration of test compounds are shown in Tables 3-6. In tests on hormones and bile acids, a modified experimental schedule was used: test compounds were given for 10 weeks and partial hepatectomy was performed in week 4 (Fig. 5). Doses of chemicals were chosen on the basis of their LD. values (19).
Among the hormones and bile acids tested by the modified schedule, ethynyl estradiol, dexamethasone, testosterone and deoxycholic acid greatly increased both the number and area of y-GT positive foci (Table 6). In this experiment, y-GT positive foci were measured quantitatively becasue they appear earlier than HN and therefore their measurement is more sensitive than that of HN.
Hepatocarcinogens and chemicals with promoting activity in liver carcinogenesis gave positive results in this system. Liver carcinogens had strong promoting effects, and some nonliver carcinogens had weak promoting effects. Thus our system could be used for screening for hepatocarcinogens or promoters of hepatocarcinogenesis as epigenetic factors and also for estimating their potency.

Conclusions
The present work shows that detectable preneoplastic lesions are induced by short-term exposure pendent in both shortand long-term tests. There was a good correlation between quantitative values for early preneoplastic lesions (y-GT positive foci) and the incidences of the more advanced lesions, such as of hyperplastic (neoplastic) nodules and hepatocellular carcinomas.
Strong liver carcinogens were shown to have strong promoting effects. Thus various tumor promoting agents can be detected and their potencies can be measured with the system developed in this study. Figure 6 shows diagrammatically possible relationships between different neoplastic stages and their modifying factors. When the initiation is strong and proliferation of initiated cells is rapid, the patient will die of cancer before completing lifespan. However, when the proliferation is slow, the lifespan is completed before cancer develops; in other words, the patient does not die of cancer. The effect of promoter is thought to shorten the period of early or preneoplastic stage by increasing the angle of aslant lines in the diagram. If we could detect some compound with an effect to decrease this angle, the compound would be used as an antipromoter.