On the other hand, we have a corresponding line of tumor in mice that has been consistently positive for virus through ten transplants. It is interesting here that, occasionally, with Dr. Rowe's indirect method of inoculating an emulsion of this tumor into adult mice and then bleeding them and testing for antibodies, certain passages of this consistently positive tumor will be negative by the MAP test. Yet, when we carry the tumor in tissue culture for 21 to 28 days, in every instance we get cytopathogenic effect (CPE) in the developing cells and the appearance of hemag-glutinating virus in the supernatant fluid of the culture, even though the original tumor emulsion might have been negative by the indirect test of inoculating mice. Thus, although it may be laborious and may introduce the possible hazard of tissue-culture contamination, here is another means, as Dr. Eddy mentioned, which, with material carried through several subcultures, brings out virus that may have been masked by the presence of antibodies in the original material.

Dr. Henle (University of Pennsylvania): When my colleagues, Dr. Gertrude Henle and Dr. Philip Garnard became interested in the polyoma virus, the hemagglutination phenomenon associated with the polyoma virus had not yet been recognized. To ordinary virologists, a wait of 2 weeks or longer for the development of CPE in mouse embryonic tissue in culture seemed rather arduous. Consequently, we thought that perhaps one could get evidence of viral multiplication earlier by the interference phenomenon. The data we have obtained in this way are similar to those which Dr. Rubin mentioned earlier.

If 10^5 tissue-culture doses of polyoma virus are injected onto mouse embryo cultures, complete resistance to the virus of vesicular stomatitis (VSV) will become apparent in 3 to 4 days. If a smaller amount such as 10^2 doses is inoculated, 10 or more days may be required until resistance to VSV is complete. Thus one can obtain a rough idea of the concentration of polyoma virus in a given preparation on the basis of the time necessary for the induction of resistance to VSV.

The challenge with VSV can be applied at a time when hemagglutination is not as as yet detectable, but protection is obtained. Nevertheless, by the time the results are available, 2 or 3 days later, there is usually some measurable hemagglutination, so that VSV interference is not of extreme practical importance. Another point of interest is that by the time resistance to VSV is measurable, relatively few cells can be shown to contain specific polyoma antigens when they are stained with fluorescent antibodies.

There has been a rumor that interferon has been demonstrated in polyoma-infected cultures. When I tried to learn the source of it I was told, "Wasn't it observed in your laboratory?" I can put this rumor at rest for the time being. If you believe that interference and interferon production are synonymous, we certainly have shown it, but all our efforts so far to demonstrate interferon in polyoma-infected cultures have failed. This might merely be on a quantitative basis, and further efforts are under way, by concentration procedures and other means, to demonstrate an interferon in these cultures.

Dr. Koprowski (Wistar Institute): There is no one here from Dr. Dulbecco's laboratory to comment briefly on his results in tissue culture. Would you care to do it, Dr. Rubin?

Dr. Rubin (University of California): I can only be brief, because all this work was done after I left. If one makes primary cultures from mouse embryos, one gets largely an outgrowth of fibroblasts. When Dulbecco and Vogt infected these primary outgrowths with polyoma virus, they observed a great deal of cellular destruction. This continues for some time with continuous release of fairly large amounts of virus, but ultimately the amount of virus decreases. At about the same time, a new type of cell appears in the culture which is specifically a much smaller cell with the characteristic that it will grow to about 3 or 4 times the cell concentration that the fibroblast will.

This cell inoculated subcutaneously in the mouse will cause a tumor, and, further, this cell is resistant to superinfection by polyoma virus. More or less the same thing has been done with hamster cells, I believe.

Dr. Sanford (National Institutes of Health): I want to comment on this work of Vogt and Dulbecco in relation to what we have done with the mouse parotid cell. They did not find CPE so much in the hamster cells, and it was one of the points in their article that this was not a selective action of the virus on the cell population to bring out a cell that might already have become neoplastic.

In our work with single-cell clones from the normal parotid we do get this CPE, and the recovered cultures have been implanted into adult mice along with control cultures at rather high concentrations of cells. In the transplantation of infected cells to prove whether they are malignant, autonomously growing cells, it is important to know that the implanted cells are the source of the tumor and that the virus released by the cells has not induced a tumor in the host. Controls can be run of the specific cell-free virus material, but one can never be sure whether filtered virus controls of this sort are going to be quite the same in their reaction on host tissues as cells that are implanted and are continuously releasing active virus into the host tissues.

We would like to suggest a system in which we have implanted C3H cells into the hybrids (C3H X BALB/c)F1. The tumors induced can then be subinoculated, and all our subinoculated tumors grow progressively in the hybrid and in the C3H mouse, but not in strain BALB/c.

Since the hybrid tissues, especially in this combination, are not ordinarily transplantable to the C3H mouse, it is thus possible to tag genetically the cells that are introduced and to prove that the tumors have actually arisen from the implanted cells.

Dr. Dawe: It is important, in interpreting the results of both Dr. Sanford's work and that of Dulbecco and Vogt to take careful account of the morphology of the tumors that are induced from the fibroblast cultures. I have seen the tumors that Dr. Sanford obtained by transferring infected cultures to mice, and the ones that we have obtained from transplanting infected organ cultures back into mice. Those from our organ cultures are identical with the induced tumors of the parotid gland, whereas those of Dr. Sanford's studies are definitely dissimilar and are, by her interpretation, with which I agree, more similar to the so-called tissue-culture tumors which she has obtained previously from isolated cell lines.

Dr. Dmochowski (M. D. Anderson Hospital): An associate of mine, Miss Elizabeth Bereczki, is engaged in the study of the relationship of polyoma virus to host cells. Among other results I would like to mention those obtained by acridine orange staining of mouse embryo cells infected with polyoma virus. The inclusions which are formed after infection are Feulgen negative. Then, eventually, they become lavender and usually remain so, though they sometimes turn pink. However, they take counterstain with azure blue.

I will not go into an interpretation of this, but it should be mentioned that the inclusions do not correspond to the virus particles seen in the electron microscope, as shown by Miss Bereczki. Thus these inclusions we often see may be an expression of reaction of the host and not the representation of aggregates of the virus itself.