Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
[CANCERRESEARCH34, 2835—284 1, November 1974] On the Origin of the Genes for Neoplasia: G. H. A. Clowes Memorial Lecture' Howard M. Temin McArdle Laboratory, University of Wisconsin,Madison, Wisconsin53706 NORMAL CELL 1@ 4..-@CELLDNA \ @\ c::@ CANCERGENES) CANCER CELL Chart 1. The origin of the genes for cancer. A cancer cell has in its DNA (zig-zag line) genes for cancer (oval). virions contain RNA and a DNA polymerase. I shall particularly deal with the two groups of avian ribodeoxy viruses, the avian leukosis-sarcoma viruses and the reticulo endotheliosis viruses. I shall then present our current ideas about the origin of these ribodeoxyviruses and indicate why I believe that this origin provides a model for the formation in normal V for cancer. Ribodeoxyviruses , \:a.*@. t cells of the genes Ribodeoxyvirus is the name of a large group of animal viruses (20). Virions of ribodeoxyviruses have a diameter of about 100 nm and consist of an envelope surrounding an internal core with no clearly observable symmetry. The envelope consists of a lipid bilayer and external glycoproteins. 4 The core contains a ribonucleoprotein particle consisting of 60 to 70 S RNA, basic proteins, and a DNA polymerase. The ribodeoxyvirus group includes viruses such as RSV,2 mouse mammary tumor virus, mouse leukemia virus, and visna virus. As will be discussed below for avian ribodeoxy viruses, the various ribodeoxyviruses differ very much in their effects on infected cells and organisms. Introduction RSV The question I am going to discuss is, “How is the genome RSV is the prototype ribodeoxyvirus. It, as well as all of the of a normal cell changed into the genome of a cancer cell?― other avian ribodeoxyviruses (7), have C-type morphology; This question is distinct from the questions of how the that is, the central core is not eccentrically located. RSV is a biochemistry of a cancer cell differs from that of a normal cell member of the avian leukosis-sarcoma group of viruses as or how the genes for cancer are expressed in a cancer cell. The shown by the antigenicity of its core proteins (the group question of the origin of cancer genes can be formalized as in specific basic proteins and the group-specific DNA polymer ase) and by the nucleotide sequence of its RNA. It causes rapid Chart 1. Before considering this question, I shall discuss some characteristics 1 Presented of ribodeoxyviruses, at the 1974 meeting of that the is, viruses whose American Association for Cancer Research in Houston, Texas. The work in my laboratory was supported by USPHS Research Grant CA-07175 from the National Cancer Institute and Research Grant VC-7 from the American Cancer appearance of sarcomas after injection into sensitive birds. In cell culture, it causes rapid cell transformation with alterations in cell morphology and cell multiplication. These alterations occur within 2 or 3 days and can involve all of the cells in a culture. The cell properties converted by infection with RSV include the serum requirement for cell multiplication, the rate Society. I formerly held Research Career Development Award CA-8182 from the National Cancer Institute and now am an American Cancer Society Research Professor. 2The abbreviations used are: Rous-associated virus-O. RSV, Rous sarcoma virus; NOVEMBER 1974 Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research. RAV-O, 2835 @1@ H.M. Temin of hyaluronic acid synthesis and the amount of the enzyme hyaluronic acid synthetase, the control of glycolysis, the rate of glucose transport, the concentration of cyclic 3', 5'-AMP, the production of plasminogen activator, and some surface properties. Variants of RSV exist that are altered in the type of neoplastic transformation that they cause, the morphology of the transformed cells, the temperature stability of the neoplastic transformation (mutants temperature sensitive for transformation), and the presence or absence of neoplastic transformation (td or NT mutants) (see review in Ref. 20). All of these genetically different viruses appear to have very similar virions. Therefore, the existence of these different virus variants indicates that the RSV genome consists of two parts: one part contains the genes specifying the virion proteins and, therefore, the virion structure; the other part contains the genes for neoplastic transformation (Chart 2). These genes are RNA in RSV virions. When RSV causes cancer, these genes are transcribed into DNA and integrated with the cell DNA forming new DNA genes for neoplastic transformation, as diagramed in Chart 3. This assertion, called the DNA provirus hypothesis, is supported by many lines of evidence. The most conclusive is the isolation from cells infected by RSV of infectious DNA containing the informa tion for production of RSV and for neoplastic transformation. The infectious RSV DNA was first reported by Hill and Hillova (8) in 197 1. We have confirmed and extended their results (Table 1). The data in this table indicate that DNA purified from either chicken or rat cells infected with RSV can cause the production of new virus and morphological transformation in recipient chicken cells. The amount of DNA required to cause transformation in one-half of the treated cultures was 0. 1 pg. The infectious material was shown to be DNA by its sensitivity to treatment with DNase and its resistance to treatment with alkali, RNase, and Pronase. This description makes it clear that RSV is extremely efficient at causing tumors and forming cancer genes. There is no progression in the sense used in cancer research in RSV-induced transformation of chicken cells. All of the cancer genes appear to be present soon after infection. However, RSV is not a natural virus. It does not persist in nonlaboratory populations. RSV is a laboratory creature, passaged and preserved by virologists. RAV-O RAV-O is another avian leukosis virus. Its virions contain the same group-specific antigens and DNA polymerase as do RSV virions, and its RNA has nucleotide sequences homolo gous to most of the nucleotide sequences in Rous sarcoma and other avian leukosis virus RNA (13, 14). However, RAV-O causes no disease in infected chickens. Crittenden found no increase in the incidence of leukemia in chickens either naturally or artificially infected with RAV-O (L. B. Critten den, personal communication). RAV.O is a natural virus, in contrast to RSV which is a laboratory virus. However, RAV-O is maintained in chicken populations, not as a virus but as a provirus or cellular genes. The existence of RAV-O as cellular genes in chickens is most clearly shown by the experiments of Crittenden et a!. (4) 2836 RSV RNA GENOME VIRION GENES, CANCER GENES ED Chart 2. The genome of RSV. This genome is RNA in the vision and DNA in infected cells. NORMAL CELL + RSV (@Ws@ PROVIRUS) CANCER CELL Chart 3. The mechanism of neoplastic transformation by RSV. The provirus is indicated by a heavy zig-zag line. Other symbols are as in Chart 1. Table 1 Infectious DNA for RSV DNA was extracted by a modified Marmur procedure from Ch(RSV)a and from rat cells infected with B77 avian sarcoma virus, but producing no virus. Serial 2-fold dilutions of DNA were added to quadruplicate cultures of chicken embryo fibroblasts previously treated with DEAE-dextran. Foci were scored after 8 to 14 days, and 50% infectious doses were calculated. Other samples of DNA were treated as indicated before infection. Data from paper of Cooper and Temin (3). infectious dose Donor DNA50% (ag)Ch(RSV)0.1Ch(RSV)-DNase>10Ch(RSV)-alkali1.0Ch(RSV)-RNase, salt0.1Ch(RSV)-Pronase0.1R(B77V)0.1 low a Ch(RSV), chicken cells infected with and producing RSV. diagramed in Chart 4. Crittenden et a!. showed that a cross of two chickens that did not produce RAV-O yielded progeny that could produce RAV-O. The two parental chickens did not produce RAV-O for different reasons. One did not carry the gene for RAV-O production (V), while the other was resist ant to the spread of RAV-O (tvb'). While most of the nucleotide sequences of RAV-O RNA are the same as those of RSV RNA, some are different. Some nucleotide sequences in RSV RNA are not present in RAV-O RNA. This is most clearly established by experiments of Neiman et a!. (14), who showed by competition hybridization that RSV RNA contained some nucleotide sequences that were not competed by RAV-O RNA. The genes for RAV-O are present in some chicken cells as DNA. This can be shown by nucleic acid hybridization (10, 13) or more definitively by infectious DNA assay (Table 2). These experiments demonstrate that the RAV-O genome is present in cells as DNA and that the RAV-O genome does not contain genes for transformation, only the genes for virus production. Therefore, the RAV-O genome (Chart 5) is simpler than the RSV genome (Chart 2). The RAV-O genome contains genes for virion proteins, but it does not contain genes for transformation. CANCER RESEARCH Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research. VOL. 34 Origin of the Genes for Neoplasia @ @\@@@tvb8 @@@tvbr No RAV—O NoRAV—O RAV—O Crittenden, Smith,Weiss,Sarma ‘974 Chart 4. The genetic transmission in chickens of the potential to produce RAV-O. V, genes for RAV-O; tvl/, gene for sensitivity to Subgroup B avian leukosis-sarcoma viruses; tvb', alielic gene for resistance to Subgroup B avianleukosis-sarcomaviruses. young chickens and selecting virus that caused rapid produc tion of myeloblastosis (2). Rous-associated virus-A is by the same criteria another avian leukosis virus. In animals, it causes some leukosis (chicken leukemia) with a low frequency. These leukemias appear only after a high rate of virus production for a long time, that is, after a long latent period. In cell culture, Rous-associated virus-A causes no transformation of fibroblasts or of reticulo endothelial cells. It is, therefore, a weakly transforming virus. Rous-associated virus-A is a natural virus. It is passed from infected hens to their eggs by infection by virions (5). The infected hens do not develop leukemia in all cases, and when they do it is after they have laid many infected eggs. Thus, Rous-associated virus-A persists in nonlaboratory populations of chickens. Reticuloendotheliosis Table 2 Infectious DNA for RA V.0 DNA was extracted and assayed as described in the legend of Table 1 from Ch(RSV)a, Ch(RAV-O) or from Cli. One day after treatment with Ot and Ch(RAV-O) DNA'S, pheasant cells were added. After 1 week, the supernatants were assayed for sedimentable DNA polymerase or uridine-3 H-labeled particles. RSV production was assayed on chicken cells. Data from papers of Cooper and Temin (Ref. 3 ; G. M. Cooper and H. M. Temin. Infectious DNA from Cells Infected with Rous Sarcoma Virus, Reticuloendotheliosis Virus, or Rous-associated Virus—O.Cold Spring Harbor Symp. Quant. Biol., in press). doseVirusTransformationCh(RSV)0.10.1Ch(RAV-O)0.2>1001>10>10 infectious Donor DNA50% Viruses Reticuloendotheliosis viruses form another group of avian ribodeoxyviruses. They share with the avian leukosis-sarcoma viruses the general characteristics of ribodeoxyviruses and even C-type morphology. They differ from the avian leukosis sarcoma viruses in the nature of their envelope glycoproteins, their core basic proteins, and their DNA polymerase and in the nucleotide sequence of their 60 to 70S RNA. After injection into young fowl, reticuloendotheliosis viruses may cause rapid death as a result of an acute infection with necrosis of the spleen and liver, or reticuloendotheliosis, which is a proliferation of primitive mesenchymal elements (17). The acute disease is mimicked in cell culture of fibroblasts where these viruses cause rapid appearance of a strong cytopathic effect. A chronic infection without cell death or transformation is then established (23). Like the avian leukosis viruses, the reticuloendotheliosis a Ch(RSV), chicken cells producing RSV; Ch(RAV-O), chicken cells producing RAV-O; Cli, chicken cells not producing virus. RAV—OGENOME VIRION GENES chart 5. The genome of RAV-O. The genome consists of RNA in visionsand of DNA in infected cells. Other Avian Leukosis Viruses Avian myeloblastosis virus is another avian leukosis virus. Its virions contain the same core group-specific antigens and DNA polymerase as do RSV and RAV-O virions, and its RNA has nucleotide sequences similar to those of the RNA's of other avian leukosis viruses. On injection into susceptible chickens, avian myeloblastosis virus causes rapid appearance of myelo blastosis, a myeloid leukemia of chickens. In cell culture, it can infect fibroblasts with virus production, but there is no transformation. However, it can cause rapid transformation of embryonic reticuloendothelial and muscle cells (12). There fore, avian myeloblastosis virus is a cell-dependent, strongly transforming virus. Again, as in the case of the strongly viruses replicate through a DNA intermediate. As shown by the data in Table 3, DNA. isolated from cells infected with reticuloendotheliosis viruses can cause formation of plaque forming virus in sensitive cells. Nucleic acid hybridization is another way to demonstrate the DNA intermediate of reticuloendotheliosis viruses. Sixty S Table 3 Infectious DNA for reticuloendotheliosis viruses DNA was extracted and assayed as described in the legend of Table 1 from Ch(TDSNVY' j@either the acute (3 DAI) or the chronic (14 DAI) phase of infection and from Ch spleen(REV-T). Virus production was measured by assay of supernatants for cytopathic effect on chicken and duck fibroblasts. Data from paper of Cooper and Temin (3). infectious dose Donor DNA50% (pg)Ch(TDSNV) DAI0.1Ch(TDSNV) 3 DAI0.1Ch 14 spleen(REV-T)0.2Ch(TDSNV)-DNase>10Ch(TDSNV)-RNase, salt0.1Ch(TDSNV)-alkali1.0Oi(TDSNV)-l.71 low g CsCl/cucm0.05 transforming RSV, avian myeloblastosis virus is a laboratory a Ch(TDSNV), chicken fibroblasts producing Trager duck spleen necrosis virus; DA!, days after infection; Ch spleen(REV-T), chicken virus. It was selected by passing virus from myeloblastosis in spleen cells producing reticuloendotheiosis virus (strain T). NOVEMBER 1974 Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research. 2837 H. M. Temin @ RNA was isolated from purified reticuloendotheliosis virus virions, labeled with ‘ 25 and annealed to DNA from infected cells (Chart 6). Most of the RNA became RNase resistant, indicating that most of the nucleotide sequences of the RNA were present as DNA in infected cells. RNA Viruses and Cancer Our discussion of these avian RNA viruses with different degrees of pathogenesis allows us to formulate criteria for whether an RNA virus can modify a cell genome to cause neoplasia. As can be seen in Table 4, the viruses that cause neoplasia contain cancer genes. The more efficient the viruses are at causing neoplastic transformation, the more cancer genes they contain. The viruses that do not cause transforma tion, like RAV-O and Trager duck spleen necrosis virus, do not contain genes for cancer. Rous-associated virus-A, which has a low efficiency of causing cancer, apparently has only some of the genes for cancer (1 1, 14). Ui F@4 0 m >- a0 z However, this formulation provides an answer only to a special case of the question posed in Chart 1. When the strongly transforming RNA viruses cause cancer, they do so as described for RSV in Chart 3, that is, by synthesizing in cells the genes for cancer coded in the viral genome. However, this formulation does not tell us how the genes for cancer are formed when there is no overt evidence of a strongly transforming RNA tumor virus. Possible Mechanisms of Carcinogenesis A number of hypotheses can be proposed to explain the creation of the genes for cancer in normal cells (Chart 7). [The infection mechanism discussed above (Chart 3) is the fourth one.] In all the hypotheses the cancer cells contain the genes for neoplastic transformation. The hypotheses differ as to whether or not these genes are present in normal cells and, therefore, the nature of the genetic changes in carcinogenesis. In a mutation hypothesis, a normal cell does not contain genes for cancer. During carcinogenesis, these appear by mutation. In a differentiation hypothesis, normal cells contain the genes for cancer in an inactive form. During carcinogenesis, these genes are activated so that their phenotypic expression leads to the formation of a cancer cell. The oncogene hypothesis is a special case of a differentiation hypothesis (6, 9). In the oncogene hypothesis, as in all other differentiation hypotheses, the genes for cancer are present in normal cells in an inactive form and become active during carcinogenesis. The i0 N NORMAL CELL @ ) CANCER CELL MUTATION ,ww\,'-f\,\ > DIFFERENTiATION @ cot (molesec/ Chart 6. Hybridization liter) of Trager duck spleen necrosis virus (TDSNV) @ I 2 S I-labeled @ virus-infected and uninfected cells. TDSNV ‘ 25I-labeled RNA (3000 cpm) was hybridized to DNA from uninfected chickens (Ch DNA), an uninfected Pekin duck (P. Du DNA), and TDSNV-infected chicken RNA to DNA's from Trager duck spleen ONCOGENE necrosis INFECTION cells ICh(TDSNV) DNA I . After different times of hybridization, the extent of hybridization was determined by RNase A and RNase T, digestion. The RNase-resistant cpm (200 cpm) were subtracted before the calculation of percentage of hybridization. Data from paper Kang and Temin (10). of ,@w%,%A@,#\ > ,W'1s.@,\ > PROTOVIRUS ,w@v',w' Chart 7. Hypotheses > of possible genetic mechanisms of carcinogene 515.Symbols as in Charts 1 and 3. Table 4 Formation of neoplasiaby RNA viruses This table summarizes the data discussed in the preceding part of the text. to cause neoplasiaPresence VirusAbility virusRSV++4*Avian myeloblastosis virus virus-A1* ±bRAV-O00Trager Rous-associated of cancer genes in ±4*a duck spleen necrosis virus00 a Hypothesized. b Lai et a!. (1 1); Neiman 2838 et al.(14). CANCER RESEARCH VOL. 34 Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research. Origin of the Genes for Neoplasia oncogene hypothesis differs from other differentiation hy potheses in that the oncogene hypothesis proposes that the genes for cancer are those of a strongly transforming RNA tumor virus. Therefore, in the oncogene hypothesis these genes for cancer are associated with the genes for an RNA tumor virus virion. The protovirus hypothesis is a special case of a mutation hypothesis (20). In the protovirus hypothesis, normal cells do not have the genes for neoplastic transformation. These genes appear by variational processes during carcinogenesis. The protovirus hypothesis differs from other mutation hypotheses by postulating that the genetic variations leading to the formation of the genes for cancer occur in a part of the normal cell genome involved in DNA to RNA to DNA information transfer. Some data can be secured relating to these hypotheses by asking what genes for cancer are present in normal cells. When DNA was extracted from normal cells, as it was extracted for the experiments described in Tables 1, 2, and 3, and added to other normal chicken cells, we found no transformation or virus production (Table 2). This experiment indicates that normal, nonvirus-producing chicken cells do not contain genes for neoplastic transformation. In addition, it shows that normal, nonvirus-producing chicken cells do not contain genes for a complete provirus. Therefore, the genes for RAV-O in normal chickens discussed above (Chart 4) are for a defective or incomplete provirus of RAV-O. Some genetic change must occur for the cells to start producing RAV-O. This genetic change is reflected in the ability of the DNA from RAV-O-producing chicken cells to transfer the ability to produce RAV-O and the lack of ability of DNA from chicken cells not producing virus to transfer the ability to produce RAV-O. These experiments are not consistent with simple differenti ation hypotheses, including the oncogene hypothesis. These results, however, do not rule out more complex differentiation hypotheses in which there are several genes for cancer and these genes are present in normal cells in an unlinked form. The Protovirus Hypothesis Now, I shall consider the protovirus hypothesis for cancer in more detail. This hypothesis states that genes for neoplastic transformation arise in an organism as a result of misevolution of a normal system of DNA to RNA to DNA information transfer. At present, there is no evidence directly testing this hypothesis. There is, however, evidence consistent with a protovirus hypothesis for the origin of ribodeoxyviruses. This hypothesis states that ribodeoxyviruses arise as a result of misevolution of normal cellular genes. The evidence comes from experiments involving nucleic acid hybridization and study of DNA polymerase relationships. When nucleic acid hybridization experiments, like those described in Chart 6, were carried out between RNA's from a reticuloendotheliosis virus or from RAV-O and DNA from a variety of fowl, different extents of hybridization were found (Table 5). The distribution of sequences of RAV-O RNA in DNA paralleled the closeness of the relationship of the fowl to chickens. There was most hybridization to chicken DNA; less to pheasant, quail, and turkey DNA; and none to duck DNA. Endogenous sequences of reticuloendotheliosis virus RNA in fowl DNA were found to the extent of about 10% in all of the gallinaceous birds, but none were found in duck DNA. In addition to these nucleotide sequence homologies between virus RNA and cell DNA, there are serological relationships between the DNA polymerases of both reticulo endotheliosis and avian leukosis-sarcoma viruses and the DNA polymerases of normal cells. An example of these relationships is seen in the data in Chart 8, which indicate that reticuloendotheliosis virus DNA polymerases specifically bind antibody made to the chicken large DNA polymerase. In addition to these relationships, there are a large number of other homologies between the genes of avian ribodeoxy viruses and chicken cells and the genes of mammalian ribodeoxyviruses and mammalian cells. (See Refs. 21 and 22 for a more complete discussion of these relationships.) These data indicate that the avian ribodeoxyviruses are related to the class Ayes, especially to the order Galliformes. The avian leukosis-sarcoma viruses apparently are an offshoot from the genus Gallus. However, it cannot yet be determined from which fowl the reticuloendotheliosis viruses evolved (2 1, 22). To explain these relationships between ribodeoxyviruses and cells, I have postulated an evolution, described in Chart 9, in which successive DNA to RNA to DNA information Table 5 Endogenousvirus-specificnucleotide sequencesin DNA of different fowl Hybridization was performed as described in the legend to Chart 6 with ‘ 25I-labeled 60S RNA's from Trager duck spleen necrosis virus (2200 cpm) or RAV-0 (3000 cpm) and DNA from livers of the indicated fowl or calf thymus DNA. Background counts of 135 cpm were subtracted before calculation of percentages of hydridization. Data from paper of Kang and Temin (10). DNAChickenPheasantQuailTurkeyDuckCalf of RNA-' 2 5 thymusTrager @Source spleenba101010<2<2necrosis duck virusRAV-O55201510<1<1 @ a RNA-' 2 5 hybridized at a C0 t of 8 X 10@ moles-sec/liter. Only the saturation value is given. NOVEMBER 1974 Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research. 2839 @1@ H. M. Temin transfer system is involved in differentiation (19, 21). IgG-L and in evolution >- I— Summary and Conclusions > I— C-) There are two groups of avian ribodeoxyviruses, the avian leukosis-sarcoma viruses and the reticuloendotheliosis viruses. Virions of these viruses contain RNA, and the viruses replicate through a DNA intermediate. There is wide variation from none to great among these viruses in their ability to cause z C-) 0 4 + 20 + + 5 10 \j@vWW\ + 15 DNA POLYMERASE(p g) Chart 8. Antibody-blocking activity \JW@fW\ of DNA polymerases for antibody against chicken large DNA polymerase (IgG-L). Antibody against chicken large DNA polymerase (4 @zg)was incubated in 25 @zl 4, VV'VWvt\ with the indicated amounts of reticuloendotheliosis virus (strain T) (REV-fl, chicken embryo large (CH-L), Trager duck spleen necrosis virus (TDSNV), chicken embryo small (CH-S), rat liver large (RL-L), RSV-RAV-O, and Escherichia co/i I DNA polymerases. After 20 min, the mixtures were heated to inactivate the DNA polymerases. (The E. co/i DNA polymerase had been previously inactivated.) Partially purified chicken embryo large DNA polymerase was added to the + viruses. Zig-zag line, DNA; straight line, RNA. The increasing and curvature indicate the evolution of viral genes. against chicken embryo large DNA polymerase and was called 0% blocking. The percentage of blocking was calculated as the percentage of increase in activity after preincubation . VVV@@W\ with the indicated ‘1' DNA polymerases. Data from work of Mizutani and Temin (15). The apparent better blocking by reticuloendotheliosis virus (strain T) DNA polymerase than by the homologous chicken embryo large DNA polymerase is the result of the impurity of the chicken embryo large DNA polymerase preparation. transfers to thickness \vAvt\A/W\ incubation mixtures. After 30 mm of further incubation, the DNA polymerase activity remaining was assayed with activated calf thymus DNA as a template•primer.The DNA polymerase activity remaining in the sample incubated without antibody was set at 100%. Eighty % of the chicken embryo large DNA polymerase was neutralized by the 4 @g of antibody VIRUS Chart 9. The protovirus hypothesis for the origin of ribodeoxy 4 VvWvV\ 4, lead changes in the cell genome. These changes result in the accumulation and alteration of nucleic acid sequences until the genome of an infectious virus is produced. The virus then exists apart from the cell and can infect other organisms. I postulate a similar process for the formation of the genes for cancer. As a result of misevolution or variation in successive DNA to RNA to DNA information transfers, the genes for cancer are produced (Chart 10). Strongly transforming viruses, like RSV, arise as a result of recombination between the new genetic elements described in Charts 9 and 10 (1 , 18). This recombination creates a virus genome containing genes for virions and for neoplastic transformation. The cancer genes exist before the cancer virus. The formation of Kirsten murine sarcoma virus may be an example of this process (16). The cellular system of DNA to RNA to DNA information transfer does not exist to cause formation of ribodeoxyviruses or the genes for cancer. I have postulated that this information 2840 4, \/WWvA\ 4, 4, @vWWV\ 4, \A@A/\ Chart 10. The protovirus hypothesis for the origin of the genes for cancer. Zig-zag line, DNA; straight line, RNA. The increasing thickness and curvature indicate the evolution of cancer genes. Oval, cancer genes. CANCER RESEARCH VOL. 34 Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research. Origin of the Genes for Neoplasia cancer. The viruses that cause cancer seem to have genes for cancer. Carcinogenesis by these strongly transforming viruses involves the formation in infected cells of genes for cancer copied from the viral genome. Normal chicken cells do not contain infectious DNA for transformation or for virus production. Therefore, there is not an inactive provirus of a strongly transforming virus in normal chicken cells. The protovirus hypothesis for cancer states that the genes for neoplastic transformation arise in an organism as a result of misevolution of a normal system of DNA to RNA to DNA information transfer. The protovirus hypothesis for the origin of ribodeoxyviruses states that ribodeoxyviruses arise as a result of misevolution of normal cellular genes. There is evidence from experiments involving nucleic acid hybridiza tion and study of DNA polymerase relationships for the protovirus hypothesis for the origin of ribodeoxyviruses. REFERENCES 1. Altaner, C., and Temin, H. M. Carcinogenesis by RNA Sarcoma Viruses. XII. A Quantitative Study of Infection of Rat Cells in Vitro by Avian Sarcoma Viruses. Virology, 40: 118—134, 1970. 2. Beard, J. W. Avian Virus Growths and Their Etiologic Agents. Advan. Cancer Res., 7: 1—127,1963. 3. Cooper, G. M., and Temin, H. M. Infectious Rous Sarcoma Virus and Reticuloendotheliosis Virus DNA. J. Virol., in press. 4. Crittenden, L. B., Smith, A. J., Weiss, R. A., and Sarma, P. S. Host Gene Control of Endogenous Avian Leukosis Virus Production. Virology,57: 128—138,1974. 5. Di Stefano, H. S., and Dougherty, R. M. Mechanisms for Congenital Transmissions of Avian Leukosis Virus. J. Nati. Cancer Inst., 37: 869—883,1966. 6. Gross, L. Facts and Theories on Viruses Causing Cancer and Leukemia. Proc. Natl. Acad. Sci. U. S., 71: 2013—2017, 1974. 7. Hanafusa, H. Avian RNA Tumor Viruses. In: Cancer—ACompre hensive Treatise, in press. 8. Hifi, M., and Hillova, J. RNA and DNA Forms of the Genetic Material of C-Type Viruses and the Integrated State of the DNA Form in the Cellular Chromosome Biochim. Biophys. Acta, 355. 7—48,1974. 9. Huebner, R. J., and Todaro, G. J. Oncogenes of RNA Tumor NOVEMBER Viruses as Determinants of Cancer. Proc. Natl. Acad. Sci. U. S., 64: 1087—1094, 1969. 10. Kang, C.-Y., and Temin, H. M. Reticuloendotheliosis Virus Nucleic Acid Sequencesin Cellular DNA. J. Virol., in press. 11. Lai, M. M. C., Duesberg, P. H., Horst, J., and Vogt, P. K. Avian Tumor Virus RNA: A Comparison of Three Sarcoma Viruses and Their Transformation-Defective Derivatives by Oligo-nucleotide Fingerprinting and DNA-RNA Hybridization. Proc. Nat!. Aced. Sci. U. S., 70: 2266—2270, 1973. 12. Moscovici, C., Moscovici, M. G., and Zanetti, M. Transformation of Chick Fibroblast Cultures with Avian Myeloblastosis Virus. J. Cellular Physiol., 73: 105—108,1969. 13. Nieman, P. E. Measurement of Endogenous Leukosis Virus Nucleotide Sequences in the DNA of Normal Avian Embryos by RNA-DNA Hybridization. Virology, 53: 196—204,1973. 14. Nieman, P. E., Wright, S. E., McMillin,C., and MacDonnell, D. Nucleotide Sequence Relationships of Avian RNA Tumor Viruses: Measurement of the Deletion in a Transformation-Defective Mutant of Rous Sarcoma Virus. J. Virol., 13: 837—846,1974. 15. Mizutani, S., and Temin, H. M. Specific Serological Relationships Among Partially Purified DNA Polymerases of Avian Leukosis Sarcoma Viruses, Reticuloendotheliosis Viruses, and Avian Cells. J. Virol., 13:1020—1029, 1974. 16. Scolnick, E. M., Rands, E., Williams,D., and Parks, W. P. Studies on the Nucleic Acid Sequences of Kirsten Sarcoma Virus: A Model for Formation of a Mammalian RNA-Containing Sarcoma Virus. J. Virol., 12: 458—463, 1973. 17. Taylor, H. W., and Olson, L. D. Chronological Study of the T-Virus in Chicks. I. Development of Lesions. Avian Diseases, I 7: 782—794,1973. 18. Temin, H. M. Malignant Transformation of Cells by Viruses. Perspectives Biol. Med., 14: 11—26,1970. 19. Temin, H. M. The Protovirus Hypothesis. J. Nail. Cancer Inst., 46: hI—Vu!, 1971. 20. Temin, H. M. Cellular and Molecular Biology of RNA Tumor Viruses, Especially Avian Leukosis-Sarcoma Viruses and Their Relatives. Advan. Cancer Res., 19: 47—104,1974. 21. Temin, H. M. On The Origin of RNA Tumor Viruses. Ann. Rev. Genet., inpress. 22. Temin, H. M. On the Origin of RNA Tumor Viruses. Harvey Lectures, in press. 23. Temin, H. M.,and Kassner,V. K. Replication of Reticuloendothei osis Viruses in Cell Culture: 291—297,1974. Acute Infection. J. Virol., 13: 1974 Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research. 2841 On the Origin of the Genes for Neoplasia: G. H. A. Clowes Memorial Lecture Howard M. Temin Cancer Res 1974;34:2835-2841. Updated version E-mail alerts Reprints and Subscriptions Permissions Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/34/11/2835 Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected]. To request permission to re-use all or part of this article, contact the AACR Publications Department at [email protected]. Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research.