The discussion so far has been concerned with the matter of where the polyoma agent acts, in terms of species, strain, and tissue specificity. It is also important to know when it can act, in terms of stage of morphogenesis or differentiation of the tissues, and how the response may differ depending on differences in these influences. Here the advantages of the polyoma virus-tumor system become more apparent, for at this point there exists a crossroads of three major disciplines (virology, oncology, and experimental embryology) largely surveyable by variations of a single methodology, tissue culture.

For nearly a decade, studies of virus tumors in mammals have been increasingly influenced by the findings made possible by wider use of young and, particularly, newborn animals. The ostensible reasons for the greater sensitivity of young animals to tumor-inducing viruses are (1) as for other viruses, the undifferentiated cells of young or embryonic animals are "inherently" more susceptible to virus infection, as are cells in tissue culture, and (2) the poorly developed reactivity of immune response in young animals permits viruses to disseminate, infect cells, and multiply to still higher concentrations before the defenses come fully into play.

Tissue-culture methods make it possible to control the second influence, thereby simplifying (not without pitfalls) evaluation of the first. Evidence will be presented here to show that for the action of polyoma virus on salivary-gland tissue, the first tenet deserves critical re-examination, at least with regard to the response elicited by virus acting on cells in different stages of pre- and post-natal development.

The observation that the oncogenic response of salivary glands in the intact mouse is more marked in the newborn period and declines progressively with increasing age remains unchallenged. The absolute limitation of the responsive period to the first postnatal day (75) found refutation, however, when highly active tissue-culture preparations of virus made it possible to show that salivary tumor could be induced in 50 percent of mice inoculated as late as the 14th postnatal day by the usual subcutaneous route (13). Parotid tumors also have appeared after long latent periods (8 months or more) in mice inoculated intravenously, when 30 to 40 days old, with massive doses of the agent (13). Furthermore, it became apparent that ability to respond could be restored to some degree in young adult mice by total-body irradiation prior to exposure to virus (13). This observation has been repeated with mice up to 5 months of age (76), though it has not been possible by X irradiation to restore the same degree of response seen in newborn mice, either in terms of proportion of animals developing tumors, multiplicity of tumor types appearing, or shortness of the latent period. The important single fact remains, however, that salivary glands of fully mature mice are capable of tumor response to polyoma virus.

A possible cotumorigenic action of radiation, as against an immune-response suppressing effect, has not been disproved from the foregoing observations. Use has been made, therefore, of the tissue-culture system previously reported (39), in which it was observed that fragments of salivary gland exposed to the virus in tissue culture responded in a dual manner, and exhibited both proliferative and necrotic response, with the former dominating during the periods studied.

The results of this study will be reported briefly here and are tabulated in table 2. It is recognized that two objections can be raised to this approach: (1) It has not yet been established that the morphologic changes observed in this tissue-culture system are biologically equivalent to tumor genesis in the intact animal. It appears probable, however, that there is a close similarity between the phenomenon in vitro and in vivo. We have recently observed the appearance of a tumor, morphologically identical with salivary-gland tumors induced in vivo, at the site of transplantation of a 30-day organ culture of submandibular gland infected with the polyoma virus (figs. 11, 12, and 13). More observations on such transplanted cultures will be required for careful evaluation of the nature of the changes seen in vitro. (2) Although the tissue-culture system provides conditions that sustain cellular survival and proliferation, these conditions do not duplicate conditions in vivo, either in regard to constancy of the milieu or to qualitative and quantitative nutritional factors. As with all tissue-culture studies, the value of the observations depends on how well they accord with other information, and on what leads they provide toward the design of other investigations.

Gelatin sponge matrix cultures of submandibular (submaxillary) gland from a 5-month-old C3H/Bi male mouse were the first to be studied besides the glands of the newborn. The technique was the same as that previously used, except that it was found necessary to replace the gelatin sponges repeatedly during the first 10 days of culture because the adult male submandibular gland secretes a protease that rapidly digests the gelatin (77). Activity of this enzyme was not apparent after 10 days in culture.

Table 2. Response To Polyoma Virus By Submandibular Glands Of Mice At Various Ages

Age of donor animal

Hemagglutination inhibition titer of donor or donor's mother

esponse at indicator 1 interval after isolation and exposure to viirus

1 Week

2 Weeks

3 Weeks

4 Weeks

6 to 8 Weeks

13-Day embryos (complete rudiment of submandibular gland)

Not determined, mothers from polyoma-free breeding colony

± Cytolysis, mesenchyme only

+ Nuclear inclusions

+ Cytolysis of epithelium and mesenchyme

+ Nuclear inclusions

+ Cytolysis of epithelium and mesenchyme

+ Nuclear inclusions

-

-

14-Day embryos (complete rudiment of submandibular gland)

Not determined, mothers from polyoma-free Weeding colony

-

+ Cytolysis of epithelium and mesenchyme

+ Nuclear inclusions

+ Cytolysis of epithelium and mesenchyme

+ Nuclear inclusions

+ Cytolysis of epithelium and mesenchyme

+ Nuclear inclusions

+ Epithelial proliferation (minimal)

-

Newborn, 12 hours

Negative

? No definite reaction

+ Cytolysis of epithelium and mesenchyme

+ Nuclear inclusions

+ Epithelial oroliferation

+ Cytolysis of epithelium and mesenchyme

+ Nuclear inclusions

+ Epithelial Droliferation

+ Cytolysis of epithelium and mesenchyme

+ Nuclear inclusions

+ Epithelial Droliferation

+ Cytolysis of epithelium and mesenchyme

+ Nuclear inclusions

+ Epithelial Droliferation

Young adults, 6 and 8 weeks

Negative

-

-

-

+ Cytolysis of epithelium and mesenchyme

+ Nuclear inclusions

+ Epithelial proliferation

-

Adult, 5 months Adult, 8 months

Negative + 1:800

-

-

-

-

-

-

Same as above Same as above + Epithelial proliferation (maximal)

-

The culture medium used throughout the work reported here was composed of human adult serum (30%) and Morgan, Morton, and Parker's medium 199 (70%). It also contained 100 units per ml. of penicillin G and phenol red at a concentration of 0.001 percent.

After 30 days, the cultures were fixed, serially sectioned, and examined. Not only had the adult tissue survived well, but the response to the virus preparation was similar to that observed previously in cultures of salivary glands from newborn mice. In fact, the proliferative aspect of the response was perhaps more marked. Figure 6 shows the heaping up of large, altered cells around the periphery of an explant, and extension of groups of cells into the compartments of the adjacent gelatin sponge. The nuclear changes and cell necrosis seen in the experiments with glands from newborns were also present.

Five additional culture series were then set up, using as donor mice 4 C3H/Bi males at 6 and 8 weeks of age with negative hemagglutination inhibition (HI) serologic reactions to polyoma virus, and one C3H/Bi female at 8 months of age, with a positive HI reaction to a dilution of 1:800.6 This last animal had never received polyoma virus by injection and bore no tumors, but was a member of a colony in which the incidence of natural infection was high. The first 4 animals had been reared and kept in a room free of polyoma-virus infection.

The general results in all these culture series were similar, though certain differences should be noted. Regardless of the age, sex, or serologic status of the donor, cytologic alterations including nuclear inclusions and clear areas, nucleolar enlargement, cell enlargement, proliferation and extension of cell groups into the sponge, as well as necrosis, were found. All these changes were similar to the response exhibited by glands of newborns. Figures 7 and 8 illustrate explants from test and control cultures, respectively, from an 8-week-old mouse. It will be noted that the control explant contains relatively well-preserved, though atrophied, remains of acinar and duct structures, and that these epithelial structures are completely quiescent and remain within the confines of the original explant. In contrast, there is a rather massive outgrowth of cells from the virus-exposed explant into the sponge. Figures 9 and 10 show the cytologic differences between the surviving cells in a control explant and the outgrowing cells from an infected culture. It is evident in figure 10 that, as in the study with glands from newborns, cell necrosis and mitosis can exist side by side among the altered cells. Another feature that has been consistently present in infected cultures of newborn as well as of adult glands is not so readily apparent in figures 7 and 8, but was quite evident in the sections. This consisted in the virtually complete necrosis of the connective-tissue elements, including macrophages, within the explants that showed the most widespread proliferative response. Van Gieson connective-tissue stains showed less collagen in the infected cultures than in the controls, and the collagen in the infected cultures was distributed mainly about portions of ducts, where it probably represented the persistence of collagen that was already present in the original explants. Surviving fibroblasts were rarely seen within this collagen. Thus within a single organ, mesenchymal elements responded solely by necrosis, while the epithelial elements responded with both necrosis and proliferation.