Perspectivesin CancerResearch Retroviruses as Carcinogens and Pathogens: Expectations and Reality1

Retroviruses (without transforming genes) are thought to cause leukemias and other cancers in animals and humans because they were originally isolated from those diseases and because experimental infec tions of newborns may induce leukemias with probabilities of 0 to 90%. According to this hypothesis viral cancers should be contagious, polyclonal, and preventable by immunization. However, retroviruses are rather widespread in healthy animals and humans where they typically cause latent infections and antiviral immunity. The leukemia risk of such infections is less than 0.1% and thus about as low as that of virus-free controls. Indeed retroviruses are not sufficient to initiate transformation (a) because of the low percentage of symptomatic virus carriers and the complete lack of transforming function in vitro-,(b) because of the striking discrepancies between the long latent periods of 0.5 to 10 years for carcinogenesis and the short eclipse of days to weeks for virus replication and direct pathogenic and immunogenic effects; (c) because there is no gene with a late transforming function, since all genes are essential for replication; (d) because host genes, which do not inhibit virus, inhibit tumorigenesis up to 100% if intact and determine the nature of the tumor if defective; and above all (e) because of the monoclonal origin of viral leukemias, defined by viral integration sites that are different in each tumor. On these bases the probability that a virus-infected cell will become transformed is estimated to be about 10 ", The viruses are also not necessary to maintain transformation, since many animal and all bovine and human tumors do not express viral antigens or RNA or contain only incomplete proviruses. Thus as carcinogens retrovinises do not necessarily fulfill Koch's first postulate and do not or only very rarely ( 10 " ) fulfill the third. Therefore it has been proposed that retroviruses transform inefficiently by activating latent cellular oncogenes by for example provirus integration. This predicts diploid tumors with great diversity, because integration sites are different in each tumor. However, the uniformity of different viral and even nonviral tumors of the same lineage, their common susceptibility to the same tumor resistance genes, and transformation-specific chromosome abnormalities shared with nonviral tumors each argue for cellular transforming genes. Indeed clonal chromosome abnormalities are the only known transformation-specific determinants of viral tumors. Since tumors originate with these abnor malities, these or associated events, rather than preexisting viruses, must initiate transformation. Therefore it is proposed that transformation is a virus-independent event and that clonal viral integration sites are conse quences of clonal proliferation of transformed cells. The role of the virus in carcinogenesis is limited to the induction of hyperplasia which is necessary but not sufficient for carcinogenesis. Hyperplasia depends on chronic viremia or high virus expression which are very rare in animals outside the laboratory and have never been observed in humans. Since latent viruses, which are typical of nearly all natural infections, are neither direct nor indirect carcinogens, they are not targets for cancer prevention. Viruses are also not targets for cancer therapy, since tumors are not maintained and not directly initiated by viral genes and occur naturally despite active antiviral immunity. Lymphotropic retrovirus has been proposed to cause AIDS because 90% of the patients have antibody to the virus. Therefore antibody to the virus is used to diagnose AIDS and those at risk for AIDS. The virus has also been suggested as a cause of diseases of the lung and the nervous system. Promiscuous male homosexuals and recipients of frequent trans fusions are at a high risk for infection and also at a relatively high annual Received 6/2/86; revised 10/14/86; accepted 11/11/86. 'Supported by (OIG) National Cancer Institute Grant CA-39915A-01 and Council for Tobacco Research Grant 1547 and by a scholarship in residence of the Fogarty International Center, NIH, Bethesda, MD. risk for AIDS, which averages 03% and may reach 5%. Others are at a low risk for infection and if infected are at no risk for AIDS. AIDS viruses are thought to kill 1-cells, although these viruses depend on mitosis for replication and do not lyse cells in asymptomatic infections. Indeed the virus is not sufficient to cause AIDS (a) because the percentage of symptomatic carriers is low and varies between 0 and 5% with the risk group of the carrier, suggesting a cofactor or another cause; (b) because the latent period for AIDS is 5 years compared to an eclipse of only days to weeks for replication and direct pathogenic and immunogenic effects; and (c) because there is no gene with a late AIDS function, since all viral genes are essential for replication. Moreover the extremely low levels of virus expression and infiltration cast doubt on whether the virus is even necessary to cause AIDS or any of the other diseases with which it is associated. Typically, provira I DNA is detectable in only 15% of AIDS patients and then only in 1 of Id2 to III' lymphocytes and is expressed in only 1 of IO4to IO5 lymphocytes. Thus the virus is inactive or latent in carriers with and without AIDS. It is for this reason that it is not transmitted as a cell-free agent. By contrast, all other viruses are ex pressed at high titers when they function as pathogens. Therefore AIDS virus could be just the most common occupational infection of those at risk for AIDS because retroviruses are not cytocidal and unlike most viruses persist as latent, nonpathogenic infections. As such the virus is an indicator of sera that may cause AIDS. Vaccination is not likely to benefit virus carriers, because nearly all have active antiviral immunity. How often have I said to you, that when you have eliminated the impossible, whatever re mains however improbable must be the truth. —Sherlock Holmes The irreversible and predictable courses of most cancers indicate that cancer has a genetic basis. In 1914 Boveri (1) proposed that cancer is caused by chromosomal mutations. This hypothesis has since received ample support (2-4), al though a cellular cancer gene has yet to be identified (5). In the light of the spectacular discovery of RSV2 in 1911, which proved to be a direct, infectious carcinogen, the hypothesis emerged that viruses may be a significant source of exogenous cancer genes (6). The virus-cancer hypothesis has since steadily gained support because retroviruses and DNA viruses were frequently isolated from animal leukemias and other tumors, and occa sionally from human leukemias, in efforts to identify causative agents (7-16). However, once discovered in tumors and named tumor viruses, most of these viruses were subsequently found to be widespread in healthy animals and humans (8, 12-18). Thus these viruses are compatible with the first but apparently not necessarily with the third of Koch's postulates3 as viral carcinogens. Only a few of the many tumor viruses are indeed directly oncogenic, such as RSV and about 20 other types of 'The abbreviations used are: RSV, Rous sarcoma virus: AIDS, acquired immunodeficiency syndrome; HTLV-1, human T-cell leukemia virus; MMTV, mouse mammary tumor virus; ATLV, adult T-cell leukemia virus; STLV-IH, simian T-cell leukemia virus; ATL, adult T-cell leukemia; MCF, mink cell focusforming; HIV, human immunodeficiency virus, ARV, AIDS-associated retrovirus. 3Koch's postulates define the steps required to establish a microorganism as the cause of a disease: (a) it must be found in all cases of the disease; (b) it must be isolated from the host and grown in pure culture; (c) it must reproduce the original disease when introduced into a susceptible host; and (</)it must be found present in the experimental host so infected. 1199 on April 21, 2017. © 1987 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from RETROVIRUSES AS CARCINOGENS AND PATHOGENS retroviruses (5, 13, 19, 20), and hence compatible with Koch's third postulate. Therefore, if we want to assess the role of viruses in cancer, there must be a clear separation between those viruses which are directly oncogenic and those which are not. The directly oncogenic retroviruses owe their transforming function to a particular class of genes which are termed one genes (20). These are as yet the only known autonomous cancer genes that can transform diploid cells in vitro as well as in animals susceptible to the particular virus (5). Since susceptible cells are inevitably transformed as soon as they are infected, the resulting tumors are polyclonal (13, 16). Nevertheless, directly oncogenic retroviruses have never caused epidemics of cancer. The probable reason is that one genes are not essential for survival of the virus and hence are readily lost by sponta neous deletion or mutation (5). Indeed, one genes were origi nally discovered by the analysis of spontaneous one deletion mutants of RSV (21). Moreover, because one genes typically replace essential genes (except in some strains of RSV) these viruses cannot replicate unless aided by a nondefective helper virus (5, 13). The vast majority of the tumor viruses are retroviruses and DNA viruses that do not contain one genes. The RNA genomes of all retroviruses without one genes measure only 8 to 9 kilobases (13, 22). They all encode three major essential genes which virtually exhaust their coding capacity. These are in the 5' to 3' map order gag which encodes the viral core protein, pol which encodes the reverse transcriptase, and env which encodes the envelope glycoprotein (23, 24). Although these viruses lack one genes they are considered tumor viruses, be cause they were originally isolated from tumors and because experimental infections may induce tumors under certain con ditions. However, in contrast to tumors caused by viruses with one genes, such tumors are always monoclonal and induced reproducibly only in genetically selected animals inoculated as newborns after latent periods of over 6 months (see below). Because of the long latent periods, these retroviruses are said to be "slow" viruses (13, 16), although their mechanism of replication is exactly the same as that of their fast and efficient relatives with one genes that transform cells as soon as they infect them (5, 19) (Table 1). The retroviruses are also consid ered to be plausible natural carcinogens because they are not cytocidal and hence compatible with neoplastic growth and other slow diseases. Indeed, retroviruses are the only viruses that depend on mitosis for replication (13, 25). However, the retroviruses without one genes are also the most common and benign passenger viruses of healthy animals and humans probably because of their unique noncytocidal mechanism of replication and their characteristic ability to coexist with their hosts without causing any pathogenic symp toms either as latent infections, which make no biochemical demands, or even as productive infections. Based on the per missiveness of a host for expression and reproduction, they have been divided into exogenous viruses which are typically expressed and hence potentially pathogenic and endogenous viruses which are typically latent and hence nonpathogenic (1618). Because they are so readily suppressed in response to as yet undefined cellular suppressors (8, 11, 12, 16-18), endoge nous viruses are integrated as proviruses into the germ line of most if not all vertebrates (8, 13, 16-18). Nevertheless, the endogenous and exogenous retroviruses are entirely isogenic and there is no absolute biochemical or functional distinction between them except for their response to suppressors of a particular host (13, 16-18) (Part I, Section A). Therefore the association of these viruses with a given disease is not sufficient even to suggest a causative role in it. Indeed there is as yet no direct evidence that retroviruses play a role as natural carcino gens of wild animals and humans. Thus the critical expectations of the virus-cancer hypothesis, namely that RNA or DNA tumor viruses would be direct carcinogens, that viral tumors would be polyclonal because each virus-infected cell would be transformed, and above all that viral carcinogenesis would be preventable by immunization, remain largely unconfirmed. Recently retroviruses without one genes have been isolated from patients with AIDS and those at risk for AIDS and have since been considered the cause of AIDS (26). In contrast to other retroviruses, the AIDS viruses are thought to act as direct, cytocidal pathogens that kill susceptible T-cells (13, 27). Here we discuss how the retroviruses without one genes fit the role of viral carcinogens or AIDS pathogens and whether these viruses are indeed the vessels of evil they have been labeled to be. Above all we hope to identify transformation-specific or AIDS-specific viral and cellular determinants and functions. Since the genetic repertoire of all retroviruses without one genes, including that of the AIDS viruses (28), is exhausted by genes that are essential for virus replication (13, 24), a hypo thetical oncogenic or AIDS function would have to be indirect or it would have to be encoded by one of the essential genes. In the second case the virus would be oncogenic or cause AIDS whenever it replicates. A survey of the best studied animal and human retroviruses demonstrates that these viruses are not sufficient to cause tumors and not necessary to maintain them. Most likely these viruses play a role in inducing tumors indi rectly. Indeed transformation appears to be a virus-independent, cellular event for which chromosome abnormalities are the only specific markers. Likewise the AIDS viruses are shown not to be sufficient to cause AIDS, and the evidence that they are necessary to cause it is debated. I. Retroviruses and Cancer A. Retroviruses Are Not Sufficient for Transformation Because Less Than 0.1% of Infected Animals or Humans Develop Tumors Avian lymphomatosis virus was originally isolated from leukemic chickens (29). However, subsequent studies proved that latent infection by avian lymphomatosis viruses occurs in all chicken flocks and that by sexual maturity most birds are infected (30-32). Statistics report an annual incidence of 2 to 3% lymphomatoses in some inbred flocks. Yet these statistics include the more common lymphomas caused by Marek's virus (a herpes virus) (33, 34). The apparent paradox that the same virus is present in most normal and healthy animals (30) but may be leukemogenic in certain conditions was resolved at least in descriptive terms by experimental and congenital contact infections. Typically experimental or contact infection of new born animals that are not protected by maternal antibody would induce chronic (31, 32) or temporal (35, 36) viremia. The probability of such animals for subsequent lymphomatosis ranges from 0 to 90% depending on tumor resistance genes (Section C). However, infection of immunocompetent adults or of newborn animals protected by maternal antibody and later by active immunity would induce latent, persistent infections with a very low risk of less than 1% for lymphomatosis (32).4 Thus only viremic animals are likely to develop leukemia at a predictable risk. Viremia has a fast proliferative effect on hemopoietic cells and generates lymphoblast hyperplasia (Fig. 1) (32, 36, 37). ' H. Rubin, personal communication.