Leukemia is a disease which occurs spontaneously among mice of most -- perhaps all -- laboratory strains, with an incidence considerably in excess of that seen in humans. Aside from the curious and exceedingly useful strains (e.g., AKR and C58) which show a near 100% incidence of the disease by one year of age, mice of most other strains, if observed throughout their life spans, show an incidence of at least a few percent, occurring usually well after one year of age. The bulk of these leukemias are lymphatic, frequently T-cell leukemias or lymphomas, although a few strains show a significant incidence of other types, such as Hodgkin's-like disease of SJL mice.
In addition to spontaneously occurring leukemia, the disease may be readily induced in mice by leukemogenic viruses of the C-type, RNA oncovirus family and by treatment with x-rays and chemical carcinogens of several different types. Susceptibility to the induction of the disease varies widely among mice of different inbred strains, however, and high susceptibility to one inducing agent is not necessarily accompanied by a similar level of susceptibility to another agent.
A considerable body of evidence points to the involvement of endogenous murine leukemia virus (MuLVs) in the leukemias occurring in the high-incidence strains. It is less certain, however, that the same process is necessarily involved in all cases of spontaneous or induced leukemia occurring in low-incidence strains. As with cancer in general, mouse leukemia may well represent a cellular phenotype which may be attained by several different pathways, some associated with viruses and some not. In any case, viral leukemias in mice are of special interest because the pathways followed in their genesis seem likely to be the first or among the first such pathways to be elucidated. Nevertheless, our present understanding of the details of viral leukemogenesis is fragmentary at best.
Genetic analyses of strain differences in the occurrence of spontaneous or induced leukemias in mice have indicated some of the major subdivisions of the pathways in viral leukemogenesis ( 1). A number of single genes have been identified which are polymorphic in the laboratory mouse and which govern events and/or conditions that appear to be favorable to the course of the disease. Specifically, three broad categories of genes have been revealed: (a) chromosomally located viral genomes, either complete or defective, representing at least two different varieties of murine leukemia viruses (MuLV) which are transmitted vertically as Mendelian traits; (b) genes which govern the expression of MuLV genomes by, for example, influencing the horizontal spread between nearby cells of the viral infection; and (c) genes which influence the capacity of the host to respond immunologically to virus-infected and/or transformed cells.
Inherited Viral Genomes. Different classes of nondefective MuLVs share a common mechanism of replication ( 2). They depend on their endogenous RNA-dependent DNA polymerases (reverse transcriptases) for the formation of a DNA copy of the viral RNA immediately following penetration of the host cell and uncoating; this DNA copy is inserted into a chromosomal location and serves as the source of viral information for all succeeding events in the replication cycle. If this chromosomal insertion occurs in a somatic cell, then the viral genome will be transmitted to all its daughter cells, but other cells of the individual can become similarly infected only by means of a new infectious event mediated by a mature virus particle produced by the original infected cell. If, however, the chromosomal insertion occurs in a germline cell, then it may be transmitted to an individual of the next generation. In this case the viral genome will be present in all cells of the individual and will be transmitted by it to future generations as a Mendelian character.
Mice infected with MuLV of exogenous origin do not routinely transmit their infection in this manner, suggesting that germline cells are protected by special mechanisms from infection with these viruses. But Mendelian transmission by mice infected in vitro as early embryos has been demonstrated ( 3).
Mice of the inbred AKR strain show high levels of ecotropic MuLV, infectious for murine cells but noninfectious in cells of most other species from soon after birth ( 4). Studies by Rowe and his collaborators ( 5) have indicated that this virus derives from MuLV genomes situated at two different chromosomal sites -- one of which is very precisely mapped with respect to closely linked markers. That these genes, designated Akv-1 and Akv-2, do in fact consist of viral genomes, rather than regulators of viral genomes located elsewhere, has been proven by experiments showing that nucleic acid sequences specific to this particular MuLV map uniquely at these two chromosomal sites in AKR mice ( 6).
Ecotropic MuLV genomes are present chromosomally in mice of strains other than AKR as well, both high- and low-leukemic. From what is presently known of their chromosomal locations, it appears that the sites of insertion of these viral genomes are rarely, if ever, the same in unrelated mouse strains. This observation clearly implies that possible insertion sites may be very numerous in the host DNA, although they may not all be of equally ready access for the proviral DNA. In a parallel manner, it is clear that xenotropic MuLVs, closely related to ecotropic MuLVs but showing a markedly different host range pattern, are also present in the genomes of most if not all mouse strains ( 7), again showing different strain-specific integration sites into the host chromosomal DNA.
Control of Virus Expression. Studies in crosses of high-leukemic AKR mice with mice of low-leukemic strains indicate that the introduction of genes capable of suppressing the expression of endogenous MuLV genomes results in suppression of the high-leukemic phenotype. Apparently it is not sufficient that the viral genome be present in a mouse for leukemia to occur; it must also be expressed in the form of infectious viral particles.
The Fv-1 gene is the best studied example of a gene which suppresses MuLV expression and also leukemogenesis. The Fv-1b allele, present in BALB/c and other inbred strains, possesses the dominant property of suppressing the replication of certain MuLVs, including the ecotropic MuLV of AKR mice, in the tissues of young mice ( 8), and as a consequence the pattern of segregation of this gene in AKR x BALB/c crosses coincides closely with the pattern of leukemia suppression ( 9). The molecular events associated with this virus suppression are not yet elucidated, but it is clear that restriction of cell-to-cell spread of virus from rare spontaneously producing cells is the basis of the low-MuLV phenotype associated with the Fv-1b allele in these crosses ( 10).
Recent studies by Allen Mayer in my laboratory have indicated that another allele at the Fv-1 locus, that present in the RF strain, also suppresses leukemia in crosses with AKR mice, but it does so without showing the severe suppressive effect of Fv-1b on ecotropic MuLV expression. Rather, this allele exerts its suppressive effect on the xenotropic MuLV which is detected in AKR mice at around six months of age. Thus suppression of either ecotropic or xenotropic MuLV expression appears sufficient to suppress leukemogenesis, a finding that lends support to the current idea that the occurrence of the disease results from a collaborative effort on the part of both types of MuLV ( 11).
Control of Tumor Cell Outgrowth. Following virus replication and spread, one or more cells of the thymic lymphoid series may be transformed, acquiring malignant growth potential. Failure of the host to suppress the outgrowth of such cells then results in clinically recognizable disease and death. Suppression of tumor growth appears to result from an immunologic response to virus-induced, tumor-specific antigens expressed on the surfaces of leukemia cells. The occurrence of this immune response is strongly influenced by the host's genotype at the complex H-2 region ( 12) of chromosome 17.
The I subregion of the H-2 complex includes the determinants of immune response genes, which markedly influence the capacity of mice to respond to specific antigenic determinants ( 13). Thus it is readily conceivable that certain tumor antigens might induce strong responses leading to tumor rejection in some hosts but only weak responses with little effect on tumor growth in others. Such a system is exemplified by the X.1 antigen of x-ray-induced leukemias of BALB/c mice ( 14). Hosts bearing the I region of the H-2b haplotype responds strongly to X.1, whereas hosts possessing only the H-2d or H-2k haplotypes are nonresponders with little capacity to reject X.1-positive tumor cells.
Recent evidence suggests another different mechanism whereby the H-2 type of the host may influence rejection of virus-induced tumor cells. Cytolytic T lymphocytes specific for viral antigens show a marked preference in vitro for target cells of the same H-2 type as that of the tumor cell which induced the response ( 15). It appears that this H-2 restriction phenomenon is attributable to the fact that the antigen recognized consists of a viral molecule complexed on the cell surface with a molecule governed by the K or D loci of the H-2 region ( 16). Not all products of the polymorphic K and D loci show the capacity to form an appropriate complex with viral molecules, and mice possessing only non-complexing K and D molecules do not exhibit a cytotoxic T cell response to their leukemia cells because the cells lack the proper antigenicity.
The genetic evidence summarized above makes a strong case for the involvement of C-type viruses in the leukemias occurring spontaneously in AKR mice, adding thereby further weight to the observations of Gross ( 17) that the disease is transferable under certain conditions with cell-free extracts. Comparable viruses have also been isolated from leukemias induced with x-irradiation or with hydrocarbon carcinogens, suggesting that induction of endogenous virus expression was also the critical factor in these leukemias ( 18, 19).
It has long been known that skin painting of mice of some strains with the carcinogen, 3-methylcholanthrene (MCA), results in a high incidence of lymphatic leukemia. Recent findings indicate that this response is specific for mice homozygous for the recessive Ahd allele ( 20). Mice possessing a dominant Ahb allele, and who therefore respond to the treatment by augmented production of the inducible enzyme, aryl hydrocarbon hydroxylase, show a strong skin tumor response rather than leukemia. In this case it is clear that the Ahd genotype of the mice does not cause the leukemia but merely represents a prerequisite for its occurrence in response to MCA. It is quite possible that the detailed mechanism of this response involves endogenous MuLV as an intermediate, and if so the marked influence of the Ah locus on this disease amply suggests the complexity of the interactions between the host, the virus, and the environment in the genesis of leukemia. The study of genetic factors in mouse leukemia has contributed enormously to our understanding of the disease, but it is clear that we have merely scratched the surface of this complex problem.
1. Steeves, R.A., and Lilly, F. (1977). Ann. Rev. Genet. 11: 277.
See also
MGI.
2. Tooze, J. (1973). "The Molecular Biology of Tumor Viruses." Cold Spring Harbor Laboratory, Cold Spring Harbor.
3. Jaenisch, R. (1976). Proc. Natl. Acad. Sci. USA 73: 1260.
See also
PubMed.
4. Rowe, W.P., and Pincus, T. (1972). J. Exp. Med. 135: 429.
See also
PubMed.
5. Rowe, W.P. (1972). J. Exp. Med. 136: 1272.
See also
PubMed.
6. Chattopadhyay, S.K., Rowe, W.P., Teich, N.M., and Lowry, D.R. (1975). Proc. Natl. Acad. Sci. USA 72: 906.
See also
MGI.
7. Kozak, C., and Rowe, W.P. (1978). Science 199: 1448.
See also
PubMed.
8. Rowe, W.P., and Hartley, J.W. (1972). J. Exp. Med. 136: 1286.
See also
MGI.
9. Lilly, F., Duran-Reynals, M.L., and Rowe, W.P. (1975). J. Exp. Med. 141: 882.
See also
PubMed.
10. Rowe, W.P. (1973). Cancer Res. 33: 3061.
See also
MGI.
11. Hartley, J.W., Wolford, N.K., Old, L.J., and Rowe, W.P. (1977). Proc. Natl. Acad. Sci. USA 74: 789.
See also
MGI.
12. McDevitt, H.O., and Benacerraf, B. (1969). Adv. Immunol. 11: 31.
See also
PubMed.
13. Klein, J. (1975). "Biology of the Mouse Histocompatibility-2 Complex." Springer-Verlag, New York.
See also
MGI.
14. Sato, H., Boyse, E.A., Aoki, T., Iritani, C., and Old, L.J. (1973). J. Exp. Med. 138: 593.
See also
PubMed.
15. Doherty, P.C., and Zinkernagel, R.M. (1975). J. Exp. Med. 141: 502.
See also
PubMed.
16. Blank, K.J., and Lilly, F. (1977). Nature (London) 269: 808.
See also
PubMed.
17. Gross, L. (1978). Cancer Res. 38: 485.
See also
PubMed.
18. Lieberman, M., and Kaplan, H.S. (1959). Science 130: 387.
See also
PubMed.
19. Ball, J.K., and McCarter, J.A. (1971). J. Natl. Cancer Inst. 46: 751.
See also
PubMed.
20. Duran-Reynals, M.L., Lilly, F., Bosch, A., and Blank, K.J. (1978). J. Exp. Med. 147: 459.
See also
PubMed.