( Originally Published Early 1960's )
The Role And Resurgence Of Viruses As cancer-info-Producing Agents.
To the two great classes of agents that can cause cancer, chemicals and radiant energy, we now add a group of self-reproducing giant molecules called viruses.
Toward the end of the Pasteur era of bacteriology, cancer, too, received attention. Many bacteria, fungi and other microorganisms were observed, isolated and claimed as causes of cancer. None of these reports was substantiated, and more careful work that avoided incidental contamination was so negative that by 1910 scientists became quite sure that cancer was not caused by microorganisms of any kind. From this was derived the dictum that cancer was not infectious. Even today, some stubborn workers continue to claim that fungi, usually with complex life cycles including filterable phases, produce cancer, but their demonstrations are unconvincing.
The period that closed with the disappointing bacteriological chapter in cancer research coincided with the earliest descriptions of diseases caused by viruses. The credit for this discovery belongs to a Russian, Demitri Ivanovsky, who in 1896 reported the transmission of tobacco mosaic disease by extracts that had been passed through filters too fine to allow the passage of any known bacteria; thus the term "filterable virus." Several years later, and probably independently, the Dutchman Van Berijinck made almost identical observations.
Peyton Rous, still a member of the Rockefeller Institute in New York, in 191 I succeeded in transmitting several sarcomas of chickens by means of cell-free, bacteria-free filtrates. The Rous sarcoma virus (RSV) is the first clear demonstration of the role of a virus in one type of malignant tumor. The discovery came at the wrong time, just when the noninfectious nature of cancers appeared to be established. Cancer research swung into a period of preoccupation with genetics, and tar-induced tumors. Only a few workers continued to explore the virus possibilities.
In the period of 1933 to 1937, two important cancer-info-virus discoveries were made. Richard Shope of the Rockefeller Institute succeeded in passing a wild rabbit papilloma or skin wart to domestic rabbits by means of cell-free filtrates, and in the new hosts the papillomas acquired the invasive characteristics of cancer. As observed with the Rous sarcoma of chickens, the filterable agent was seldom recoverable from the malignant tumor it had induced. In 1936, the workers at the Jackson Memorial Laboratory, primarily geneticists, discovered that the transmission of breast cancer in mice was extrachromosomal, in that it occurred in offspring of mice only if the mother came from the high breast cancer line. If' the father came from the high cancer line, the progeny did not develop breast cancers. The late John Bittner explored one of the simplest possibilities, that something was transmitted from the mother to the young after birth, and found that this something was indeed in the milk of the mothers. By the process of foster nursing high breast cancer strains by low breast cancer females, the occurrence of cancers was dramatically reduced, and by feeding or injecting young mice of low breast cancer strains with milk from mice of high cancer strains, the incidence of breast cancer could be strikingly increased.
During this period, too, cancer research was preoccupied with the polycyclic hydrocarbons, and the most promising approach appeared to be the biochemical one. Viruses were excluded in the research plans of many important institutions.
Credit for forcing the attention back to viruses is probably due to the discoveries of two workers. Ludwik Gross of New York showed that some leukemias of mice also were transmitted by cell-free filtrates. The filtrates were poor sources of the agents unless they were first "warmed up" by passage in tissue culture, and very young, preferably newborn animals had to be used for the transfer experiments. Sarah Stewart at the National Cancer Institute, following similar leads, succeeded in isolating from a mouse leukemia another agent that produced salivary gland cancers in mice, and on passage through tissue culture acquired even wider carcinogenic properties in being able to produce many different types of tumors, not only in mice but in several other rodents, such as rats and hamsters. This "Polyoma" virus broke through all previous reservations against virus research in cancer. Up to then, it was considered that the few cancer viruses such as the Rous virus in chickens or the Shope virus of rabbits were species and tumor specific.
Now both the tumor specificity and species specificity concepts were shattered, and many of the subsequent implications became rather obvious.
It should be remembered that during this period, 1945-1955, virus research was making great strides, especially in its stimulated attack on the problem of poliomyelitis. Tissue culture, the electron microscope, serological and immunological identification techniques were effective weapons that provided our victory over poliomyelitis, and at this appropriate time the same weapons could be trained against cancer.
In various animal species, there are now at least a dozen cancers that are established to be of viral etiology. In addition, there are several benign tumors and temporary masses of active cell proliferation stimulated by viruses to indicate a spectrum of cell reactions. There has been considerable elucidation of the structure and chemical composition of viruses. And the recent studies on the structure of animal cells are beginning to reveal how the virus-cell interaction takes place. These advances are creating so logical and orderly a conceptual picture that its general applicability, including to man, is no longer doubted, even though to date firm experimental evidence is not in hand. The question is no longer whether human cancers are due to viruses, but rather, what cancers are due to what viruses, and what are the mechanisms involved?
The type of cellular reaction produced by a virus apparently depends upon the degree to which the host can resist viral invasion. In many common virus diseases, the end-point is the destruction of virus-invaded cells. Limitation of viral reproduction and recovery of the host organism depend upon the development of an antibody response. The virus-host relationship in such virus entities as warts leads to the stimulation of cell growth by the virus, but the immune mechanisms eventually gain the upper hand and the lesion is localized or eliminated. In the benign tumors such as the papilloma of the wild rabbit, the fibroma of the squirrel, and the fibroma of deer, there is a more equal balance between the virus and the cells, with slow but controlled growth of the tissue. These tumors contain easily recoverable virus products. In the true invasive cancers, the causative virus is difficult to recover, since it may be in so perfect a balance with the cell thàt it actually becomes part of its structure.
The world of viruses is a large one, and includes entities that are quite different in size and in chemical structure. Basically, viruses are giant molecules of the same chemical structure as the components of animal and plant cells, with the characteristics of parasites in that they can replicate only within animal or plant cells. They are structurally either desoxyribonucleic acids (DNA) or ribonucleic acids (RNA), a series of nucleic acid components (purines and pyrimidines) attached along backbones of sugar-phosphoric acid building blocks. The nucleic acids during the extracellular phase of virus existence are contained in a protein coat, which is shed when the virus enters a cell.
There are examples of both DNA and RNA-type viruses among the carcinogenic virus classes. The DNA-types include the "Papova" group, an abbreviation-condensation word for the Shope Papilloma virus of rabbits, the Polyoma virus of mice, and the Vacuolating virus of the monkey. These viruses are initially seen in the nucleus of the cells they invade.
The RNA-types include the viruses of the lymphoma-leukemia complex, which produce the leukemias of chickens, the Rous sarcoma, and the leukemias of mice, of which there are a half-dozen or more different varieties. These viruses are initially seen in the cytoplasm of the cells. The Bittner virus, which is required for the occurrence of most breast cancers in mice, is another cytoplasmic RNA virus entity.
As more information regarding viruses and their behavior within the cell becomes available, it becomes increasingly clear that the so-called cancer viruses are not an unique class but represent examples of virus entities that fit them nicely into the whole world of viruses. The leukemia-lymphoma complex, for example, resembles the influenza group of viruses. The interrelationships have been emphasized by two recent discoveries. A common virus that is found in the monkey, the SV40 virus, when grown on tissue culture and injected into newborn hamsters, produces sarcomas at the site of injection. This finding led to considerable consternation among public health officials because SV40 contaminated many of the polio vaccine preparations that were used in immunizing children in 1955. Fortunately, careful observations fail to show any increase in cancers among children who received contaminated preparations, and now such contamination is avoided. Following this lead, John Trentin and his coworkers in Houston, Tex., found that a common virus that occurs in the nose and throat of man, Adenovirus 12, also evokes cancers in hamsters after being passed and activated by tissue culture. Another human virus, Adenovirus 18, also demonstrates this property. Thus, under special conditions, rather common viruses were shown to possess cancer producing activity.
Up to the time of this writing, no human cancer has been demonstrated to be due to a virus. Under the electron microscope, particles that resemble viruses have been photographed in cells from patients with acute leukemia, rectal polyps and gastric cancer. But such evidence is insufficient, because mere presence of viruses or bacteria does not establish a causal role; such particles could be incidental contaminants of no relation to the cancer process. Nevertheless, the experiences with the animal tumor viruses are now so extensive that to maintain that human beings do not develop reactions of a similar type becomes increasingly untenable.
The real question in regard to the virus-cancer studies is whether a virus will be found eventually as the basic, essential causative factor in all cancer reactions. No answer is possible at this time on this point. In my opinion, however, it is reasonable to postulate that permanent changes in the DNA or the RNA of a cell can be brought about by chemicals and by physical energy as well as by viruses, and that it is not necessary at this time to restrict the carcinogenic stimuli to any one class of agents, viruses or otherwise.
Considerable attention was focused on the experiments of Henry Kaplan and his group in California, who showed that leukemia in one strain of mice could be evoked by X-ray only if the animals were carriers of a leukemia virus. Thus, in effect, the role of X-rays in this instance was to exteriorize or activate a viral agent already present in the animal. It has not been demonstrated that a similar process applies to all leukemias of mice, much less whether this is a general explanation for the role of chemical carcinogens. Investigation of carcinogenic chemicals as activators of latent viruses is a fertile field, but extrapolations should be carefully controlled. What is actually dangerous is the false reasoning that now no heed needs to be paid to environmental carcinogens, such as tobacco smoke, because eventually the problem will be solved by the discovery of some viral agent. Even in the Kaplan experiments, it is evident that mice with leukemia virus would not have developed leukemia had they not been exposed to radiation, thus retaining radiation as an important factor in the causation. To intimate that the hazards of chemical carcinogens in our environment should be ignored because eventually they may be demonstrated to exert their effects through the activation of intracellular viruses is as absurd as to suggest that we should ignore fecal contamination of our drinking water because we have effective vaccines and antibiotics against typhoid.
A practical problem that has become of great importance with the resurgence of interest in viruses and cancer is whether cancer is infectious. The National Cancer Institute receives many letters of inquiry on this point, and the writers usually refer to the fear that contact with patients or household pets with cancer may be dangerous. We can no longer reply dogmatically, as we did 10 years ago, that cancer is not infectious.
It is more than probable that at least some human cancers will be found to be produced by viruses. Such discoveries may eventually allow us to prevent the occurrence of such cancers by the development of appropriate vaccines. But we still can be confident, on the basis of present knowledge, that no human cancer is communicable; that is, there is no evidence that a patient with cancer can transmit his disease to another person by contact or other direct means of infection. This is even more definite when transmission of cancer is feared between animals and man. How specific and limited we must be when we discuss cancer is immediately evident by the exceptions that are already known. In dogs, there does exist a cancer entity known as venereal sarcoma that is transmitted by direct contact. In chickens, the virus complex that causes several forms of leukemia and other cancers is transmitted from infected to young noninfected fowl.
Whether these and similar situations have their analogues in human populations remains to be established by future research.