INTRODUCTION

Top Abstract Introduction Materials and methods Results Discussion References

Studies of the genetics of the mouse in relation to infections by the oncogenic murine polyomavirus promise to reveal multiple factors with diverse modes of action. Variations in susceptibility to tumor induction by polyomavirus among inbred mouse strains have been noted (8, 11, 14; data not shown). Much of this variation reflects differences in the abilities of different strains of mice to generate virus-specific anti-tumor immune responses, resulting in variable overall tumor frequencies. In addition, some strains exhibit resistance to tumor development in specific target tissues while being susceptible at other sites, suggesting the existence of nonimmunological mechanism(s). The host genetic background can influence events as diverse as the early events of virus replication and spread as a necessary prelude to the induction of a broad array of tumors and late events affecting the likelihood of metastases of specific tumor types. The identity and mechanisms of action of most of these genes remain unknown. Here we describe two contrasting forms of host resistance, one that prevents tumor induction by targeting the virus and preventing its spread and one that acts downstream of virus spread, targeting incipient tumors for immunological rejection.

A survey of over 30 inbred mouse strains has uncovered several strains which show a total or near-total absence of tumors or development of a restricted spectrum of tumor types following inoculation of newborn animals with highly tumorigenic strains of virus. These resistant strains can be divided into two groups based on the responses of adult mice to ionizing radiation followed by virus inoculation. The use of radiation is necessary because newborn mice of even the most susceptible strains are known to acquire resistance as adults (16). Maturation of immune mechanisms is generally regarded as the basis for this acquisition of resistance, although other mechanisms may also be involved.

Most resistant strains show a radiation-sensitive form of resistance, designated R^r-s, indicated by the development of tumors in a high proportion of treated animals. Earlier studies have established the R^r-s phenotype as the norm of host resistance, mediated by cellular immune responses to polyomavirus-induced tumors (1, 15, 16, 24). MHC (major histocompatibility complex) type is a major determinant of this kind of resistance (11). The present study reveals a second form of host resistance, based on the observation that mice of certain strains fail to develop tumors following radiation and virus challenge. This radiation-resistant form of resistance, designated R^r-r, operates through a nonimmunological mechanism that blocks virus dissemination in the intact host.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and methods Results Discussion References

C3H/BiDa mice were obtained initially from Clarence Reeder (National Cancer Institute, Fredrick Cancer Research and Development Center, Fredrick, Md.). C57BR/cdJ, MA/MyJ, CE/J, and C57BL/6J mice were obtained from the Jackson Laboratory, Bar Harbor, Maine. All mice were maintained in our virus-free breeding colony. F₁ crosses were performed reciprocally.

Mice younger than 24 h of age were inoculated intraperitoneally with 0.05 ml of A2, A2+ (10), or LID (2, 3) polyomavirus containing approximately 2 × 10⁵ to 1 × 10⁶ PFU. Four- to six-week-old mice received between 500 and 900 rads of  radiation in a Gammacell 40 (Atomic Energy of Canada, Ltd., Ottawa) and 24 h later were inoculated intraperitoneally with virus. Mice were inspected twice weekly and necropsied when moribund or at 6 to 12 months of age. Tumors and tissues were fixed in Bouins' solution and embedded in paraffin, and sections were stained with Harris' hematoxylin and eosin as previously described (5, 13). All gross tumors other than those of bone were confirmed by microscopic examination.

Whole mouse section hybridization was performed as previously described (6, 7). Briefly, mice inoculated at birth were sacrificed at 10 days and frozen in Tissue-Tek embedding medium (OCT 4583). Parasagittal sections were transferred to nitrocellulose membranes and hybridized to ³⁵S-labelled nick-translated polyomavirus DNA, and the hybridization signal was visualized by autoradiography.

Tests for innate immunity (NK cells and antiviral interferons) were carried out as described previously (25; see text).

	RESULTS

Top Abstract Introduction Materials and methods Results Discussion References

Patterns of susceptibility and resistance to tumor induction by polyomavirus. Three patterns of host responses are apparent based on comparisons of tumor development in newborn versus irradiated adult mice (Table 1) as follows:

(i) Susceptible strains (S). C3H/BiDa is one of several inbred strains known to be highly susceptible to the oncogenic effects of polyomavirus. When inoculated within 24 h of birth with an appropriate strain of virus, these mice develop multiple tumors in frequencies approaching 100% within a period of 3 to 4 months (5, 13). Susceptibility in these strains is based in large part on a common MHC background (H2^k) coupled with an endogenous superantigen encoded by a particular mouse mammary tumor provirus (Mtv-7 sag). In the H-2^k background, the Mtv-7 superantigen effectively deletes precursors of polyomavirus-specific cytotoxic T lymphocytes, rendering these mice unable to reject polyomavirus-induced tumors (17, 18).

(ii) Radiation-sensitive resistant strains (R^r-s). Many strains of mice show inoculated as newborns partial or complete resistance to tumor induction. Earlier studies have shown that normally resistant strains such as C57BL/6J can be rendered susceptible following immune suppression by a variety of mechanisms, such as radiation or neonatal thymectomy or administration of antilymphocyte serum (1, 15, 16, 24).

(iii) Radiation-resistant resistant strains (R^r-r). Not all neonatally-resistant mice develop a radiation-sensitive form of resistance as adults as is clearly illustrated with mice of the MA/MyJ strain. Adult mice of this strain maintain complete resistance to tumor induction by polyomavirus, even when they are given immunosuppressive doses (up to 900 rads) of radiation.

MA/MyJ mice are resistant to the virulent LID virus strain. LID is a virulent strain of polyomavirus which spreads rapidly in newborn mice, inducing widespread cytolytic damage, particularly in the kidney and brain (2, 3). Newborn mice were inoculated intraperitoneally with 10⁶ PFU of the LID strain, as previously described (2). Following are data on the fractions of animals surviving up to 30 days. All 20 newborn C3H/BiDa mice inoculated with this virus died within 2 weeks. All of the mice of the R^r-s strains C57BR/cdJ (n = 20) and C57BL/6J (n = 21) were also killed by LID. Some of these mice survived for up to 3 weeks but uniformly succumbed, with the same pattern of virus spread and cytolytic damage as that seen in C3H/BiDa mice. The slightly extended average survival time of R^r-s compared to that of susceptible mice may reflect a degree of viral clearance by polyomavirus-specific cytotoxic T lymphocytes present in the former but absent in the latter strains (17, 18).

Newborn MA/MyJ mice are highly resistant to virus spread. To test these predictions, newborn mice of each prototype strain were inoculated with either the high tumor (A2) or virulent (LID) strain of virus, and the degree of virus spread was determined by whole mouse section hybridization at 10 days (Fig. 1). The results show that MA/MyJ (R^r-r) mice are highly resistant to the spread of both virus strains (Fig. 1A and D). C57BR/cdJ (R^r-s) mice, on the other hand, are susceptible to disseminated infection by both viruses (Fig. 1B and E), showing degrees of spread only slightly less than those seen in C3H/BiDa (S) mice (Fig. 1C and F).

View larger version (32K): [in this window] [in a new window]   FIG. 1.   Whole mouse section hybridization. MA/MyJ (A and D), C57BR/cdJ (B and E), and C3H/BiDa (C and F) mice were inoculated at <1 day old with A2 (A to C) or LID (D to F) virus and sacrificed at 10 days. See text.

Cell cultures derived from MA/MyJ mice are susceptible to lytic virus infection. The resistance of MA/MyJ mice could be accounted for by a lack of or a reduced level of expression of functional virus receptors. This possibility was examined by a comparison of the results of infection of primary kidney cell cultures prepared from baby MA/MyJ and C3H/BiDa mice. Cultures were infected at a low multiplicity (<0.1 PFU/cell), so that multiple cycles of infection would be required to induce full cytopathic effects. As shown in Fig. 2, cultures from both mice developed extensive cytopathic changes typical of polyomavirus infection after 1 week. Thus, while kidneys of MA/MyJ mice are largely spared in vivo (Fig. 1), kidney cells in culture from these mice are readily killed by the virus. This result demonstrates that at the cellular level, MA/MyJ mice are capable of expressing functional virus receptors as well as other factors essential for productive viral infection.

View larger version (154K): [in this window] [in a new window]   FIG. 2.   Uninfected and infected kidney cultures from baby C3H/BiDa (A and B, respectively) and MA/MyJ (C and D, respectively) mice. Infected cultures are shown 7 days postinfection. See text.

Tests for innate immunity in S, R^r-s, and R^r-r mice. Several aspects of innate immunity are known to affect host resistance to viral infection. Among these, the production of NK cells is of possible relevance to the R^r-r phenotype, first because of the known relative resistance of NK cells to radiation, and second because NK cells are regulated in part by type 1 (antiviral) interferons. To test the possibility that resistance of MA/MyJ mice may be due to these aspects of innate immunity, levels of NK activity and antiviral interferons were measured and compared in young adult mice of each prototype strain. NK activity was measured in a standard cytotoxic assay with spleen-derived leukocytes and chromium-labeled YAC-1 target cells (25). No significant levels of NK activity were seen, and no differences among the three mouse strains were noted (Fig. 3).

View larger version (15K): [in this window] [in a new window]   FIG. 3.   Assay for NK cytotoxic activity. E/T, effector/target ratios with splenocyte suspensions from normal mouse spleens as effectors and chromium-51-labelled YAC-1 target cells. Incubations were overnight at 37°C. Percent cytotoxicity was calculated as described previously (25).

Differences in genetic transmission of resistance by R^r-s and R^r-r mice. The resistance of R^r-s mice was previously found to be recessive in a cross between C57BR/R^r-r cdJ (R^r-s) and C3H/BiDa (S) (17, 19). In contrast, resistance conferred by the R^r-r trait is inherited in a dominant or codominant fashion. The dominance of R^r-r is evident with variable penetrance, depending on the assay. Under the conditions of our tests, the R^r-r trait shows incomplete penetrance in F₁ animals, based on protection from tumor development, but complete penetrance in protecting the host from the acute lethality of the virulent LID strain.

	DISCUSSION

Top Abstract Introduction Materials and methods Results Discussion References

The resistance of MA/MyJ mice to polyomavirus differs from that of C57BR/cdJ and other resistant strains which mount effective anti-polyomavirus tumor immune responses. Indeed, MA/MyJ mice carry both the MHC type (H-2^k) and an endogenous mouse mammary tumor virus superantigen (Mtv-43 sag, with V specificity identical to that of Mtv-7 sag [21]) predictive of a highly susceptible phenotype based on an inability to mount such antitumor responses (17). The fact that their resistance cannot be overcome by immunosuppressive doses of radiation further indicates a nonimmunological basis.

The R^r-r form of resistance exhibited by MA/MyJ mice contrasts with the R^r-s form of C57BR/cdJ in several additional ways. The resistance of MA/MyJ extends in dramatic fashion to newborns inoculated with the virulent LID strain of virus, which kills C57BR/cdJ and other R^r-s strains but not MA/MyJ. Virus replication and spread are sharply curtailed in MA/MyJ compared with those in C57BR/cdJ mice. In crosses with the highly susceptible C3H/BiDa strain, MA/MyJ transmits its resistance in a dominant or codominant manner, while that of C57BR/cdJ is completely recessive. MA/MyJ is unique among the strains we have examined so far in that it exhibits all of the above-mentioned properties. However, the CE/J mouse resembles MA/MyJ in its adult onset R^r-r phenotype, showing resistance to the induction of tumors following radiation as adults but limited susceptibility as newborns.

The R^r-r phenotype is based on a nonimmunological mechanism that blocks dissemination of the virus in vivo. Neither a high level of NK cells nor a high level of antiviral interferons accounts for this form of resistance. The failure of the virus to replicate and spread in vivo does not result from an inability of the host to express virus receptors, since kidney cultures from baby MA/MyJ mice are fully susceptible to infection. A simple deficiency of virus receptors in vivo would also be incompatible with the observed dominance of the R^r-r phenotype. A limited array of tumors arise in roughly half of MA/MyJ mice inoculated with LID as newborns, indicating that at least some cell types in these mice bear receptors for the virus and can be transformed in vivo. These results are consistent with a strong down-regulation of virus spread which spares the animal from the acute lethal effects of the virus while allowing enough dissemination for the induction of some tumors.

Expression in the intact host of some factor(s) that binds and diverts the virus from infecting target cells or acts intracellularly to prevent virus release could, in principle, account for the R^r-r phenotype. An efficient apoptotic response of MA/MyJ mice to viral infection, i.e., one that acts in some way to override the polyomavirus's antiapoptotic mechanism (4, 20), would also be consistent with the present findings. Experimental approaches to identify the physiological and molecular bases of the R^r-r form of polyomavirus resistance are limited at present due to inability to model the resistance in culture. However, with the relatively rapid assay LID survival test and the clear dominance of the R^r-r trait, a genetic approach to identify the gene(s) involved should be possible.