THE MURINE LEUKEMIA-SARCOMA VIRUS COMPLEX* The

THE MURINE LEUKEMIA-SARCOMA VIRUS COMPLEX * BY ROBERT J. IIUE1NER

LABORATORY OF VIRAL DISEASES, NATIONAL INSTITUTE OF ALLMEItY AND INFECrIoUs DISE -SES, NATIONAL INSTITUTES OF HEALTH, BETHESDA, MARYLAND The development of in vitro (tissue culture) assay systems for murine leukemia (MuLV)l" 2 and sarcoma (MSV)3-7 viruses, the demonstration of the defectiveness of the MSV genome,3 and its rescue with various MuLV's7 have brought out certain remarkable similarities between these viruses and those of the avian leukosis- Rous sarcoma complex.3' 7, 8, "a Subsequent studies of the murine viruses, which I propose to outline briefly in this presentation, not only emphasize further the extent and significance of these similarities, but show also that the natural behavior, prevalence, and modes of spread of murine leukemia and sarcoma viruses follow very closely the patterns of the known avian tumor viruses. Avian Tumor Viruses.-During the many intervening years since Rous's discovery of the Rous sarcoma virus (RSV) in 1911, much information has been obtained concerning the nature and behavior of the RNA tumor viruses. However, as is true of all virus research, the modern era of RNA tumor virus research can be said to begin with the development of in vitro test systems. Manaker and Groupe reported altered foci in chick embryo cultures induced by RSV, following which Rubin developed a tissue culture test for the growth and assay of avian leukosis viruses (ALV).9' ga In reporting the resistance-inducing factor (RIF) test, Rubin showed that ALV interfered with the focus-forming activity of P SV in chick embryo fibroblast (CEF) cultures.10 Arined with two simplified and reproducible assay procedures, the focus-forming for RSV and the RIF for ALV, Rubin, Hana- fusa, Vogt, and others demonstrated the defectiveness of the Bryan strain of Rous sarcoma virus,iOa 11 described its rescue in vitro with various leukosis viruses,12 and defined to a considerable extent the natural history and immunological specificities of the latter in infected chicken populations."3 13a Subsequently my associates and I reported complement-fixing (CF) antigens in hamster tumors induced by the Schmidt-Ruppin strain of Rous sarcoma virus (SR-RSV). Using sera of tumored hamsters, we showed that these antigens were also present in chicken sarcomas and in chick embryo cells infected with avian leukosis viruses. The CF antigens were not sedimented with virus particles and appeared to be soluble internal antigenic components shared by all the known avian leukemia-sarcoma viruses.'4' 15 This latter finding provided the basis for the development of the COFAL test (complement fixation for avian leukosis),' a test now widelv used for detection and assay of ALV infections in chickens. These observations stimulated further studies on the nature of the internal antigens as structural components of the leukosis viruses;17 on their presence in "nonproducer" avian and hamster sarcoma cells;'8-20 and on the transfer of the noninfectious RSV genome from sarcoma cells to normal cells by direct contact.21 22 As a result of these studies it was soon established that ALV and RSV were similar to the influenza-like myxoviruses in that they contained anl internal group-specific antigen in addition to the outer envelope aiitigeim which had been shown earlier to be the determinant of serotype specificity. 12 This broad sharing of soluble virion antigens made it possible with the use of the COFAL test to detect, identify, and 835 Downloaded by guest on September 30, 2021 836 N. A. S. SYMPOSIUM PROC. N. A. S.

assay in tissue cultures all the established strains of ALV and many naturally occurring leukosis viruses as well. An in vitro Assay Systemfor Murine Leukemia Viruses.-The new in vitro methods for detection and assay of the avian tumor viruses led in our laboratory to attempts to develop similar procedures for murine leukemia viruses. An in vivo test developed earlier by Rowe for certain murine leukemias based on interference with the overt effects of Friend leukemia virus in mice,23 while successful, proved to be im- practical for most purposes. Since murine leukemia viruses were known to persist in cultures of leukemic tis- sue24 and to grow to a limited extent in covert fashion in normal mouse embryo fibroblast (MEF) cultures,25 a test modeled along the lines of the COFAL test seemed to offer the best solution. Eventually, Hartley, Rowe, and I developed such a test; we found that sera from rats carrying transplantable lymphosarcomas induced by the Rauscher and Gross leukemia viruses or from rats immunized by these and other leukemia viruses grown in MEF tissue culture would react in CF tests with specific antigens present in MEF cells infected with these same viruses. This led to standardized procedures for producing specific rat antisera, for assaying murine leukemia viruses in tissue culture, and for measuring CF and neutralizing serum antibodies to the latter. 2, 25 The initial test procedure, which might be termed the "complement fixation test for murine leukemia," or the COMuLV test, while useful for assaying the growth in MEF cultures of well-recognized strains of leukemia such as the Gross (G+) virus on the one hand and Friend, Moloney, and Rauscher (FMR) viruses on the other,26 was relatively insensitive for detecting naturally occurring leukemia 1 viruses. 2 While the various immune rat sera used to detect viral growth contained CF antibodies to sedimentable infectious virus particles, they did not react with soluble antigens that we suspected must also be present in the supernates. Thus we were unable in our early studies to confirm in the MuLV's the presence of a group-specific internal (CF) antigen similar to that found earlier in the ALV-RSV complex viruses. However, in 1966 Geering, Old, and Boyse26 reported a gel pre- cipitation test which revealed a group-reactive antigen released by ether treatment from murine leukemia viruses, which appeared to be shared by both theG+ and the FMR groups of murine leukemia viruses. We readily confirmed this observation in complement fixation tests using G + rat sera provided by Old and sera of rats in our own laboratory carrying tumors induced by the murine sarcoma virus (MSV).7 Prevalence of Murine Leukemia Viruses as Determined by the COMuLV Test.- Sera from inbred Fisher rats carrying fibrosarcomas produced by MSV and by pseudotypes of MSV developed CF and neutralizing antibodies to the corresponding strains of virus.3' I Following several serial transplantations of the sarcomas into newborn or weanling rats, neutralizing antibodies no longer appeared in sera and infectious virus was less evident in the tumors.27 Complement-fixing antibodies, however, were frequently present, and some were highly reactive with antigenic preparations of the well-known murine leukemia viruses and with preparations of many murine tissues and tissue cultures suspected to contain naturally occurring leukemia viruses.27 With the use of highly reactive rat sera we found that MSV and MuLV preparations usually contained large amounts of a group-specific antigen separable by Downloaded by guest on September 30, 2021 VOL. 58, 1967 ROBERT J. HUEBNER 837

TABLE 1 DEMONSTRATION OF SOLUBLE COMPLEMENT-FIXING" ANTIGENS IN MSV(RLV) WITH THE USE OF MSV RAT ANTISERUM

I Se I MSV RIRS (V)b (V + S) C Starting suspension" 16 16 Upper supernatant5 <2 16 1st Centrifugation Lower supernatant <2 16 Pelletf 16 16 ( Pellet 16 16 Upper supernatant <2 <2 2nd Centrifugation without ether Lower supernatant <2 <2 Pellet 16 16 (Pellet 16 16 Upper supernatant <2 16 2nd Centrifugation after ether RLQ Lower supernatant <2 16 Pellet 2 16 a The microcomplement fixation test was performed as described previously., b Sera of rats immunized with Rauscher tissue-culture-grown virus (antibodies to virus particle (V) only). C Sera of rats carrying MSV tumor (antibodies to viral (V) and soluble (S) antigens). d Ten per cent clarified suspension of MSV tumor tissue in EMEM. e Upper and lower halves of supernatant after centrifugation 15,000/rpm, 40 rotor/60 min, Spinco model L. f Pellet-resuspended in 1/4 volume. g Ether treatment according to method described by Geering et al." centrifugation from the viral particle and that identical "soluble" antigens were released from the latter by treatment with ether, whereas rat sera containing antibodies to the virus particle only did not detect the soluble antigens released by ether (see Table 1). Thus the leukemia viruses of mice also appeared to have a structure not unlike that of other myxoviruses which contain group-specific internal soluble (S) antigens and external or type-specific envelope (V) antigens.28 This finding proved to be particularly useful since the broadly reactive rat antisera made it possible for Hartley and her associates29 to detect and isolate within a relatively short period nearly 200 strains of murine viruses having the antigenic properties of the murine leukemia-sarcoma complex (Tables 2 and 3). Perhaps of greatest interest was the isolation of such viruses from tissues of normal mice, including those known to have very low spontaneous rates of clinical leukemia. The COMuLV test therefore makes it possible to design studies of the natural occurrence of murine leukemia viruses in their natural hosts, and our preliminary observations indicate that murine leukemia viruses, like many other viruses, are highly prevalent as well as covert infectious entities. Viewed in this context the TABLE 2 TABLE 3 In vitro ISOLATIONS OF LEUKEMIA In vitro ISOLATIONSa OF LEUKEMIA VIRUSES FROM MICE WITH LEUKEMIA VIRUS FROM TISSUEsb OF NORMAL MICE AND OTHER TYPES OF CANCER Wean- No. positive/ Embryo ling Adult no. tested High-incidence Virus-induced leukemia by 6 es- strainsc 13/14 13/13 40/47 tablished leukemia virus strains 5/6a Low-incidence Spontaneous leukemias 11/11 strainsd 3e/22 0/16 33/67 Irradiation-induced leukemia in a All isolations made in NIH Swiss MEF's. 2 strains of mice 2/2 b Spleen and thymic suspensions. Miscellaneous transplanted c AKR, C58, Call/I g. d BALB/c, CsH/HE, CFl, RF. tumors 7/9 e All three positives were from CsH/HE. a Kaplan virus (KLV) only strain not grown in MEF (derived from NIH Swiss, BALB/c, and C57B1 embryos). Most isolations from all sources inade in NIH Swiss MEF's. Downloaded by guest on September 30, 2021 838 N. A. S. SYMPOSIUM PROC. N. A. S.

development of clinical leukemia or lymphoma might then be regarded as the result of the additive effects of environmental factors such as physical and chemical carcinogens; thus such determinants would activate viral leukemias in the same way that various other inciting agents precipitate other types of viral activity. Additional determinants would be infection early in life with high dosage of virus and/or genetically determined host susceptibility. On the other hand, oncogenic mutants might also influence the onset and frequency of the neoplastic expressions of the leukemia-sarcoma viruses. Recent Observations on the Murine Sarcoma Viruses.-Following the development of tissue culture assay systems for murine leukemia and sarcoma viruses, we demonstrated that the genome of the latter persisted in a defective noninfectious state in transplantable sarcomas induced in newborn hamsters by MSV.7 We found also that the defective genome could be rescued and rendered fully infectious when MSV-induced hamster tumor cells (MSV-HT) were grown in mixed tissue cultures with CHIEF cells which were superinfected with the Moloney, Rauscher, and Friend variants of leukemia virus. The newly reconstituted pseudotype MSV's-MSV- (MLV), MSV(RLV), MSV(FLV)-were found to have the serological specificities of their helper leukemia viruses and to be as active in producing sarcomas in mice and transformed foci in MEF as the original MSV virus (Tables 4 and 5). More recently, other pseudotypes were produced by using tissue-culture-grown Gross leukemia virus and a strain of leukemia virus derived from normal cells of AKR newborn mice.29 Both pseudotypes-MSV(GLV) and MSV(AKRLY')- were derived by superinfection of mixed hamster tumor (MSV-HT) and MEF cells with tissue-culture-grown preparations of the respective leukemia viruses. Neu- tralization tests in MEF tissue cultures using type-specific rat antisera described previously" 2 revealed that despite certain differences MSV(GLV) and MSV- (AKRLV) were antigenically closely related to each other, but were quite distinct from the pseudotype MSV's induced by Friend, Moloney, and Rauscher viruses (Table 5). Sedimentation and treatment with ether showed that these pseudotypes also contained the group-specific soluble antigen described above. These and other investigations confirmed results obtained earlier showing that the MSV pseudotypes, like the RSV pseudotypes, carried type-specific outer envelopes identi- TABLE 4 INFECTIOUS VIRUS AND CF ANTIGENS IN EXTRACTS OF MOUSE SARCOMAS INDUCED IN NEWBORN NIH MICE BY PSEUDOTYPE MOLONEY SARCOMA VIRUSES Focus- CF Mouse Type of forming antigen Pseudotypea Derivation passageb extracts titer titer Tumor Inductiond MSV(MLV) Mixed TC + MLVe P0 Crude 103.1 4 NT NT 4i4i P1 Conc. 105-0 16 13/13 8 days it HT-1 + MLV (in vivo) P0 Crude 10'.7 <2 4/12 12 days MSV(RLV) Mixed TC + RLV P0 Crude 103.5 4 NT NT ithit Pi Conc. 104.7 8-16 21/21 8 days it HT-1 + RLV (in vivo) P0 Conc. 10'-4 <2 20/20 8 days MISV(FLV) Mixed TC + FLV P0 Crude 101.3 <2 17/17 9 days "t "t P1 Conc. 1060- 2 22/22 8 days itAdt P2 Conc. 106*7 4 16/16 7 days a Viruses rescued with various helpers. MLV = Moloney leukemia virus. RLV = Rauscher leukemia virus. FLV - Friend leukemia virus. NT = not tested. b Po = Primary inoculation. Pi - First passage (extract or transplant), etc. c Crude = 10% extract. Conc. = Concentrated virus prepared by Moloney procedure (see ref. 7, footnote 8). d No. with tumor/no. mice inoculated with undiluted crude extract or 10-1 dilution of concentrate. Days indicate time of first tumor. Mixed TC = HT-1 hamster tumor cells plus MEF. Downloaded by guest on September 30, 2021 VOL. 58, 1967 ROBERT J. HUEBNER 839

TABLE 5 ANTIGENIC CHARACTER OF MSV PSEUDOTYPES DETERMINED BY SERUM NEUTRALIZATION TESTS Neutralizing Antibody Titera vs: Rat antiserum MSV (RL) MSV (ML) MSV (FL) MSV(GL) MSV(AKR) Controlb Od 0 0 0 0 Rauscherb 40 0 20 0 0 Moloneyb 0 40 0 0 0 Friendb 0 0 40 0 0 Grossc 0 0 0 40 80 AKRc 0 0 0 0 160 C58c 20 20 20 320 320 a Reciprocal of serum dilution giving 67% or greater reduction in number of MSV foci. b Sera of rats immunized with control tissue culture fluid or murine leukemia viruses grown in tissue culture. c Sera of rats bearing transplanted tumors from leukemias induced by indicated viruses. The serum versus the C58 (G +) virus was kindly supplied by Dr. Lloyd Old. d 0 = <20. cal with those of the respective leukemia viruses used for rescuing the sarcoma genome. The newly established pseudotypes also contained an internal antigenic component found in all the murine leukemia viruses (vide supra). Rescue of the defective MSV genome was achieved in several different ways. It can be seen in Table 6 that, as reported previously,7 the MSV genome was rescued from the MSV-HT cells only when the latter were exposed in tissue culture to both murine leukemia virus (MuLV) and normal mouse cells; rescue was achieved in vito as well when MSV-HT cells mixed with MuLV were injected subcutaneously and/or intramuscularly into newborn NIH Swiss or BALB/c mice. However, we found that the genome was also rescued by injecting MSV-HT cells into newborn AKR mice without the addition of murine leukemia virus.29 Since tissues from normal newborn AKR mice regularly yielded demonstrable leukemia virus, it seemed clear that the vertically transmitted leukemia virus present in AKR mice provided an indigenous source of envelope material for the noninfectious sarcoma genome. We have not succeeded in transferring the cell-transforming principle from MSV- HT cells to normal mouse cells to acquire "nonproducer" mouse sarcoma cells such as Sarma et al.22 reported for the avian sarcoma system. Perhaps of special interest is the pseudotype MSV produced by my associate Dr. Igel, who used as a helper the leukemia virus isolated several years ago from irradiated C57B1 mice by Kaplan and his associatesY3' 31 Despite years of effort, this virus (KLV) has not been adapted for growth in tissue culture, and bioassays of the virus which must be done in mice frequently require six months or longer. The VISV(KLV) pseudotype made it possible to study the envelope properties of the KLV in tissue culture as well as in newborn mice. MSV(KLV) appears to have the antigenic properties of MSV(GLV) but is peculiar in that it is poorly adapted for growth in any MEF cells except those derived from the C57B1 strain, and its tumorigenic activity appears to be restricted largely to C57B1 newborn mice.30 The MSV viruses therefore resemble the RSV viruses in several important ways (Table 7). Like RSV, MSV's contain virus-specific envelope antigens that reveal easily detectable antigenic differences between various antigenically different helper leukemia viruses. Thus the Gross virus group was shown to be distinct from the Friend, Moloney, and- Rauscher virus group. The tissue culture plaque reduc- Downloaded by guest on September 30, 2021 840 N. A. S. SYMPOSIUM PROC. N. A. S.

TABLE 6 In vitro AND In vivo ATTEMPTS TO RESCUE MSV GENOME FROM HAMSTER TUMOR (HT) CELLS HT MSV (-) lr:.Tissue culture experiments In vivo experiments (NIH Swiss mice) HT cells + MuLV-yield MSV (-) HT cells only into NIH mice-yield MSV (-) HT cells + MEF cells-yield MSV (-) HT cells + MuLV into NIH mice-yield MSV (+) HT cells + MEF cells + MuLV-yield HT cells into AKR mice-yield MSV (+) MSV (+) MSV (8th passage, BALB/c); newborn hamsters; 9th-30th tissue culture passage of HT cells. MSV = Moloney sarcoma virus; (-) = virus-free; (+) = positive for MSV. MEF = cells cultured from special NIH strain Swiss mice. MuLV = murine leukemia virus (several strains). The NIH Swiss mice were MuLV-free. The AKR mice were positive for MuLV. TABLE 7 SIMILARITIES BETWEEN AvIAN AND MURINE-SARCOMA VIRUSES Avian Murine Focus formation + + D)efective sarcoma genome Rescue with leukemia virus + + In vitro In vivo + + Antigens + + Type-specific envelope + + Group-specific internal + + Replication by budding at plasma membrane + + Host range (determined by helper virus envelope) + + Pathogenesis determined by sarcoma genome, not by helper virus + + Interference by leukemia viruses with sarcoma virus + + Transfer of genome from nonproducer sarcoma cells to normal cells without helper virus

tion neutralization test described earlier1' I provides perhaps the simplest procedure available for distinguishing the antigenically different murine leukemia viruses. Since the AMLV-MSV internal (S) antigens are generally produced in higher titer than the envelope (V) antigens, they are therefore easily detected in the COMuLV test, an observation that, as described above, has greatly facilitated the detection and assay of prevalent leukemia viruses. The murine leukemia viruses have been found by Sarma et al.32 to interfere with MSV focus formation in tissue cultures, thus behaving very much like the ALV- RSV complex viruses. This observation was utilized to develop a standard assay procedure for the known leukemia viruses. This new test appears to have a sen- sitivity nearly equal to that of the COMuLV test. Recently we found that host and host cell susceptibilities to infection with various MSV's are genetically determined. Cell penetration apparently is determined by the envelope furnished by the helper MuLV. Thus the host range as well as the development of neutralizing antibodies in the infected host are helper-dependent properties of the MSV's. The pathogenic behavior of the sarcoma genome, on the other hand, appears to be helper virus-independent in that the tumors produced by all the pseudotypes have the same cytopathic appearance. 0 General Considerations.-One of the more obvious yet frequently overlooked facts in considering causes of disease is that very few etiological agents behave exclusively in highly specific fashion. Studies of the whole "icebergs" including the hidden (subclinical) activities of most infectious agents show that the clinical manifestations regarded as most characteristic nearly always represent relatively unusual Downloaded by guest on September 30, 2021 VOL. .;8, 1967 ROBERT J. HUEBNER 841

events in the natural histories of such agents. More often thait not, therefore, epidemiological studies based exclusively on highly specific clinical manifestations of widespread agents are likely to result in incorrect conclusions which hinder ra- tional control efforts. In the past, misconceptions about the prevalence of virus infections in relation to poliomyelitis or respiratory diseases, for instance, were anything but helpful. Until recently, similar misconceptions about the rarity, specificity, and uniqueness of animal leukemia viruses may have obstructed progress in developing more useful detection and assay systems for these viruses. Similarly, recent emphasis on the increasing incidence of lung cancer induced by chemical carcinogens found in to- bacco tars may have tended on the one hand to obscure the importance of such carcinogens in causing other cancers and on the other the importance of other carcinogens in lung cancer. In addition, recent investigations suggest that multiple carcinogenic agents (physical, chemical, and viral) may combine to produce certain tumors.33 Carcino- genesis resulting from the interactions of two or more chemicals, two viruses, viruses plus chemicals, irradiation and viruses, are known to occur. However, almost nothing has been done as yet to determine to what extent such synergistic phe- nomena occur naturally in general populations. The recent development of in vitro laboratory assay systems for viruses and environmental pollutants now should make such efforts possible. Conclusion.--The newer in vitro techniques for detecting and assaying the leukemia and sarcoma viruses of chickens and mice have produced radical changes in previous concepts concerning the natural behavior of these viruses. Like many other viruses, leukemia virus infection appears to be widespread, yet in most natural circumstances it rarely results in clinical disease during the normal lifetime of the infected animal. Indeed, one might conclude from studies of the prevalence of ALV's and MuLV's that they represent the most common virus infections of the mouse and the chicken. If this is so, as I pointed out previously for other pathogenic viruses, ecological studies of a limited or uncontrolled nature could be expected to uncover such infections as frequently in association with good as with ill health, despite the fact that under certain natural conditions they do cause clinically ob- servable disease.34 Finally, with the tools at hand, meaningful field studies can now be done to determine more specifically the roles of the avian and murine leukemia-sarcoma viruses in producing leukemia and other neoplastic illnesses in their natural hosts. The many similarities exhibited by these "natural" models in two different classes of animals suggest that they may represent expressions of a general biological pat- tern likely to be expressed also in man and his domestic animals. * This work was partially supported by the Etiology Area, National Cancer Institute, NIH. 1 Hartley, J. W., W. P. Rowe, W. I. Capps, and R. J. Huebner, these PROCEEDINGS, 53, 931 (196.5). 2 Rowe, W. P., J. W. Hartley, and W. I. Capps, NOCI Monograph No. 22 (1966), p. 15. I Hartley, J. W., and W. P. Rowe, these PROCEEDINGS, 55, 780 (1966). 4 Moloney, J. B., in Some Recent Developments in Comparative Medicine (London: Academic Press, 1966), 251. 5 Harvey, J. J., Nature, 204, 1104 (1964). 6Ting, R. C., Virology, 28, 783 (1966). Downloaded by guest on September 30, 2021 842 N. A. S. SYMPOSIUM PROC. N. A. S.

7Hluebner, R. J., J. W. Hartley, W. P. Rowe, W. T. Lane, and W. I. Capps, these PNRocE:- INGS, 56, 1164 (1966). 8 Huebner, R. J., in Perspectivee in Leukemia, Proceedings of the Leukemia Society, Inc. (New Orleans, La.: in press). 8a Huebner, R. J., in Carcinogenesis: A Broad Critique, Proceedings of the Twentieth Annual Symposium on Fundamental Cancer Research (Houston, Texas, M.D. Anderson Hospital and Tumor Institute, in press). 9 Manaker, R. A., and V. Groupe, Virology, 2, 838 (1956). 9a Rubin, H., and P. K. Vogt, Virology, 17, 184 (1962). 10Rubin, H., these PROCEEDINGS, 46, 1105 (1960). 1Oa Vogt, P. K., and H. Rubin, Virology, 13, 528 (1961). 11 Hanafusa, H., T. Hanafusa, and H. Rubin, these PROCEEDINGS, 49, 572 (1963). 12 Ibid., 51, 41 (1964). 13 Rubin, H., Bacteriol. Rev., 26, 1 (1962). 18a Vogt, P. K., and R. Ishizaki, in Viruses Inducing Cancer-Implications for Therapy (Salt Lake City: University of Utah Press, 1966), p. 71. 14 Huebner, R. J., D. Armstrong, M. Okuyan, P. S. Sarma, and H. C. Turner, these PROCEED- INGS, 51, 742 (1964). 15Armstrong, D., M. Okuyan, and R. J. Huebner, Science, 144, 1584 (1964). 16 Sarma, P. S., R. J. Huebner, and D. Armstrong, Proc. Soc. Exptl. Biol. Med., 115, 481 (1964). 17 Bauer, H., and W. Schafer, Z. Naturforsch., 20b, 815 (1965). 18 Vogt, P. K., P. S. Sarma, and R. J. Huebner, Virology, 27, 233 (1965). 19 Kelloff, G., and P. K. Vogt, Virology, 29, 377 (1966). 20 Payne, F. W., J. J. Solomon, and H. G. Purchase, these PROCEEDINGS, 55, 341 (1966). 21 Hanafusa, H., and T. Hanafusa, these PROCEEDINGS, 55, 532 (1966). 22 Sarma, P. S., W. Vass, and R. J. Huebner, these PROCEEDINGS, 55, 1435 (1966). 23 Rowe, W. P., Science, 141, 40 (1963). 24 Manaker, R. A., P. C. Strother, A. A. Miller, and C. V. Piczak, J. Natl. Cancer Inst., 25, 1411 (1960). 25 Hartley, J. W., W. I. Capps, W. P. Rowe, and R. J. Huebner, unpublished data. 26 Geering, G., L. J. Old, and E. A. Boyse, J. Exptl. Med., 124, 753 (1966). 27 Huebner, R. J., W. T. Lane, H. C. Turner, and J. W. Hartley, unpublished data. 28 Schaffer, F. L., and C. E. Schwerdt, in Viral and Rickettsial Infections of Man (Philadelphia: J. B. Lippincott Co., 1965), p. 119. 29 Hartley, J. W., W. P. Rowe, W. I. Capps, and R. J. Huebner, unpublished data. 30 Igel, H. R., and R. J. Huebner, unpublished data. 31 Lieberman, M., N. Haran-Ghera, and H. S. Kaplan, Nature, 203, 420 (1964). 32 Sarma, P. S., M. Cheong, J. W. Hartley, and R. J. Huebner, Virology, in press. 33 Symposium: Conference on Epidemiologic Approaches to Cancer Etiology, Cancer Research, 25, 1271 (1965). 34 Huebner, R. J., Ann. N. Y. Acad. Sci., 67, 430 (1957). Downloaded by guest on September 30, 2021