Surprisingly, given our relatively detailed knowledge of the genomes of several murine and in- sect species, fewer studies aimed at identifying genes whose expression modifies the develop- ment of neoplasia have been carried out in these lower species. Studies in the mouse are perhaps the best documented. In Table 5.16 may be seen a listing of most of the known genes that have been found to modify the development of neoplasia in animals treated with chemical carcino-
Table 5.16 Modifying Genes in Germline Genetic (GG) and Chemical (C) Carcinogenesis in Animals
gens. In at least one instance, that of the Min mouse, a germline genetic mutation in the APC gene (see above) is the basis for neoplastic development. In this latter case, the modifying gene termed Mom1 has been reported to be identical to the gene for secretory type II phospholipase A2 (Pla2s), as reported by MacPhee et al. (1995). These latter authors suggest that the modify- ing gene acts to alter the cellular microenvironment within the intestinal crypts, thereby modify- ing the number of polyps that develop.
A number of examples of genes, all of which have been only partially or very little charac- terized, have been shown to alter the tumor response to the administration of one or a few doses of a chemical carcinogen. The Hcs locus first described by Drinkwater modifies the develop- ment of hepatomas in the liver of the mouse after ethylnitrosourea administered during neonatal life or hepatomas that develop spontaneously. The action of the gene is to alter the rate of repli- cation of preneoplastic hepatocytes. Manenti et al. (1994) have also reported at least five other Hcs loci in the mouse, but the functions of these are as yet unknown. Malkinson and his associ- ates have reported the existence of multiple pulmonary adenoma susceptibility (pas) genes that affect the spontaneous or chemically induced numbers of pulmonary adenomas in appropriate mouse strains (Malkinson, 1991). In studies directed towards identifying such genes, they have presented evidence that different forms of Ki-ras proto-oncogene in which a 37-bp sequence within the first intron of the gene is present once in the DNA of sensitive strains and twice in that of resistant mice (Malkinson and You, 1994). The authors have considered several possible mechanisms for the modifying action of this mutation. Fijneman et al. (1994) have also shown that other loci within the major histocompatibility complex are likely to modify lung tumor sus- ceptibility in the mouse as well. Susceptibility to estrogen-induced pituitary tumors in rats segre- gates with the prolactin (Prl) gene or in some closely linked gene (Shepel and Gorski, 1990).
There have also been reported instances of enhanced spontaneous and chemically induced neoplasia in both rats and mice exhibiting specific genotypes. One of the most extensively stud- ied is the viable yellow gene (Avy), the expression of which increases the incidence of spontane- ous and induced tumors in mouse strains carrying the nude (nu/nu) mutation (Chapter 15; Stutman, 1979). In addition, mice of the C3H strain carrying the Avy gene show an extremely high spontaneous incidence of mammary and liver neoplasms (Heston and Vlahakis, 1968). In rats, two separate mutations not obviously related to cancer development have been shown to act
as modifiers in the development of specific neoplasms. Rats of the LEC strain bearing a deletion in the copper transporting ATPase gene homologous to that causing Wilson’s disease (Wu et al.,1994) in the human exhibit a high spontaneous rate of both renal and liver cancer (Izumi et al., 1994; Ono et al., 1991). Whether this is related to the abnormality in copper metabolism must await further investigation. Another rat strain bearing a small deletion mutation in the albumin gene, in which homozygotes exhibit almost no measurable serum albumin, exhibit a higher inci- dence of neoplasms of the kidney (Nagase et al., 1983) after dimethylnitrosamine administration and of the brain after a single transplacental administration of ethylnitrosourea to homozygous rats (Usuki et al., 1992).
Pharmacogenetics as a Multifactorial Variable in Animal Cancer
The study of multiple forms of phase I and phase II genes in animals has, for some species, been almost as extensive as that in humans. However, the initial studies relating the metabolic activa- tion of chemical carcinogens to specific genes was carried out in the mouse more than two de- cades ago, when the relationship between the induction of a phase I activity, aryl hydrocarbon hydroxylase (AHH) by the carcinogen methylcholanthrene was related to the “carcinogenic in- dex” or susceptibility to chemical carcinogenesis of the skin in a series of inbred mouse strains. These data are depicted in Figure 5.17. These studies stimulated investigations mentioned above
Figure 5.17 Relationship between the carcinogenic index (Iball index) for the induction of neoplasms in mice by subcutaneous administration of 3-methylcholanthrene as a function of the induction of aryl hydro- carbon hydroxylase (AHH) activity by 3-methylcholanthrene in the liver for each of 14 genetically distinct inbred strains of mice, each indicated by a separate solid circle. The Iball index is defined as:
The carcinogenic potency (I) of a chemical is thus related to both the number (as %) of neoplasms induced and the time required (latent period) for neoplasms to appear (Chapter 13). By this reference potent carcin- ogens usually induce many neoplasms in a short time, whereas weak carcinogens induce fewer neoplasms only after a prolonged period. (After Thorgeirsson and Nebert et al., 1977, with permission of the authors and publisher.)
in relation to aryl hydrocarbon hydroxylase induction in human cells in patients with lung can- cer. Interestingly, induction of sarcomas by the subcutaneous implantation of the same carcino- gen gave various results in relation to the aryl hydrocarbon hydroxylase activity, depending on the dose of the carcinogen implanted (Prehn and Lawler, 1979). The genetic locus that is critical for the environmentally induced response in these phase I genes as well as a number of phase II genes has been termed the Ah locus. This locus codes for one member of a family of proteins that are ligand-activated transcription factors involved in the regulation of the expression of a number of genes. The molecular mechanisms of this regulatory process are being elucidated but are beyond the scope of this text. The interested reader is referred to recent reviews (Swanson and Bradfield, 1993; Okey et al., 1994).
A mutation in the cytochrome P450–metabolizing debrisoquine has been reported in the rat (Matsunaga et al., 1989) owing to an absence of expression of the gene. No carcinogenesis investigations have yet been carried out on this strain, but in another study (Aitio et al., 1991)— with outbred rats administered the carcinogen diethylnitrosamine for 20 weeks—a correlation of individual susceptibility to the development of liver cancer was related to the pattern of a num-ber of cytochrome P450 gene products. This relationship was most accentuated at low dose lev- els and thus was felt to be a closer reflection of human exposure.
Several phase II genes have also been investigated in rodents, but most extensively investi- gated and compared with the human are the N-acetyltransferases. Polymorphic forms of slow and rapid acetylator phenotypes have been described in the hamster (Ferguson et al., 1994), mouse (Chung et al., 1993), rabbit (Sasaki et al., 1991), and rat (Hein et al., 1991). Mutations in the hamster are somewhat analogous to those seen in the human, but those in the rabbit appear to be quite different. McQueen et al. (1982) demonstrated that, in hepatocytes from rapid and slow acetylator rabbits maintained in culture and treated with hydralazine, DNA repair was present in cells from the slow acetylators but not in those from the rapid acetylators. This suggested a po- tential difference in the susceptibility to amine carcinogens as well. A similar result was seen in hepatocytes of rapid and slow acetylator strains of mice treated with aminofluorene (cf. Weber and Hein, 1985).
GENETIC CARCINOGENESIS: THE KEY TO UNLOCKING THE SECRETS OF CANCER?
From this overview, the student will hopefully appreciate that the explosion in our knowledge of genetic mechanisms is reflected in a dramatic increase in our understanding of neoplasia and its genesis. In the remainder of this text, there are numerous other examples verifying this state- ment. However, as indicated earlier in this chapter, germline genetic alterations do not constitute the major cause of neoplasia in the human, although especially the multifactorial, polygenic area may be involved in the causation of a great percentage of all human cancer. The next chapter views the genetics of neoplasia not from the germline but from the inheritance of somatic cells. This process, as we have noted from our definition of neoplasia (Chapter 2), is ubiquitous, and much of our knowledge of somatic cell genetics has now evolved from the genetic revolution of the latter part of this century.