As noted in Figure 7.1, the application of a single appropriate low dose of a carcinogenic poly- cyclic hydrocarbon does not of itself result in the development of any tumors. However, the po- tential for preneoplastic development may be realized by addition of the promoting agent, strongly arguing for the presence of cells altered or “initiated” in the carcinogen-treated skin.
Until recently, characterization and quantification of the stage of initiation could be ac- complished well only after the process of carcinogenesis had begun. Thus, the presence of pre- neoplastic and/or neoplastic lesions implied that initiation had occurred. Many preneoplastic and neoplastic lesions appear to be derived from single cells (cf. Tanooka, 1988). It is on this basis, then, that some of the characteristics of the stage of initiation have been delineated. To date, it has not been possible to identify in an unequivocal manner single initiated cells, although single cells having some of the characteristics of initiated cells have been described in several experi- mental systems (Moore et al., 1987; Nakano et al., 1994; Solt et al., 1985). Therefore most of the discussion on the cellular and molecular biology of initiation has been deduced from experi- ments in which the implication is that the characteristics of the stage of initiation are being investigated.
Morphologic and Biological Characteristics of the Stage of Initiation
One of the striking characteristics of the stage of initiation, deduced from the experiments out- lined in Figure 7.1, was the irreversibility of its effects. An extensive delay of administration of the promoting agent after initiation still led to the development of tumors. Other biological char- acteristics of initiation could also be concluded from the experiments of Figure 7.1. Application of the promoting agent must follow initiation, although in some investigations in the skin, Fürstenberger et al. (1985) suggested that a partial inversion of this sequence could induce mouse skin carcinogenesis. Other biological characteristics of the stage of initiation are given in Table 7.3. The effectiveness of initiation by chemicals is sensitive to alterations in xenobiotic metabolism (Talalay et al., 1988) and in trophic hormones (Liao et al., 1993). As discussed in Chapter 3, the ability of a carcinogenic agent to initiate cells may depend on the ability of the cell to metabolize the agent to its ultimate carcinogenic form. The process of initiation appears to be most effective when the agent is administered during the DNA synthesis phase of the cell cycle, as noted both in vivo (Rabes et al., 1986) and in cell culture (Grisham et al., 1980). Sev- eral studies (Borek and Sachs, 1968; Columbano et al., 1981; Kakunaga, 1975) have indicated the critical importance of one or several rounds of cell division that must occur in the presence of the initiating agent for a cell to become initiated.
An important characteristic of the stage of initiation is the fact that spontaneous (endogenous or fortuitous) initiation occurs in living organisms, probably rather frequently. The existence of spontaneously initiated cells can be deduced from the occurrence of sponta- neous neoplasms in both the human and in experimental animals. Single putatively initiated cells have been described in the rat liver (Moore et al., 1987), and their spontaneous clonal progeny have been identified by several investigators (Schulte-Hermann et al., 1983; Xu et al., 1990). Other spontaneous preneoplastic lesions have been described in experimental sys- tems (Maekawa and Mitsumori, 1990; Pretlow, 1994) as well as in the human (Dunham,
1972; Pretlow, 1994). Therefore it would appear that the spontaneous or fortuitous initiation of cells in a variety of tissues is a very common occurrence. It follows, then, that the develop- ment of neoplasia may be a function solely of the action of agents at the stages of promotion and/or progression.
The absence of either a readily measurable threshold or no-effect dose level and of a max- imal response of the formation of initiated cells short of lethal cellular toxicity to the tissue and/or animal are major characteristics of the process of initiation. Such a dose-response curve is seen in Figure 7.2. As noted in the figure, extrapolation of the data points of those animals administered the carcinogen will extend the curve to the ordinate where some tumors are seen to
Table 7.3 Morphological and Biological Characteristics of the Stage of Initiation
Irreversible in viable cells
Initiated “stem cell” usually not morphologically identified
Efficiency sensitive to metabolic and other cellular factors and the cell cycle
Spontaneous (endogenous) occurrence of initiated cells
Requires cell division for “fixation”
Dose-response does not exhibit a readily measurable threshold
Relative potency of initiators depends on quantitation of the numbers of preneo- plastic lesions after defined period of promotion
Figure 7.2 Liver tumor incidence expressed as percentage response in mice given various daily doses of
2-acetylaminofluorene for 24 months. These data were taken from the review by Fishbein (1980). A small percentage of animals not receiving any of the carcinogen developed liver neoplasms within the 24-month period, as evidenced by the symbol slightly above the origin on the y axis. However, extrapolation of the data points through the origin of animals given 2-acetylaminofluorene can be accomplished.
occur in the absence of any administered carcinogen. As the curve is drawn, however, even at the lowest theoretical doses, one must presume an ever-so-slight increase in tumor incidence over that seen in the absence of the carcinogen. While the potential pitfalls in the extrapolation of data points have been considered by a number of individuals (Preussmann, 1980; Aldridge,
1986), at present there is insufficient evidence for the existence of measurable thresholds of the effects of initiation. Further support for this concept may be seen in Figure 7.3, which depicts a dose-response curve of the appearance of single putative initiated hepatocytes as a function of the amount of a single dose of the chemical carcinogen diethylnitrosamine. Although this curve
Figure 7.3 The effect of varying the dose of diethylnitrosamine administered to rats 24 hours after a 70% partial hepatectomy on the number of single hepatocytes expressing the placental form of glutathione S-transferase (PGST+) per liver 14 days after carcinogen administration.
appears to exhibit a threshold, the data points may readily be extrapolated through the origin. Other studies in which DNA adducts of a specific carcinogenic agent are measured as a func- tion of dose similarly result in a linear relationship extrapolatable through the origin (cf. Zeise et al., 1987). While the determination of the initiating potency of chemical agents has gener- ally not been done in rat liver, a measure of the relative potency of several initiating agents has been made on the basis of the quantitation of numbers of preneoplastic focal lesions (Pitot et al., 1987).