JOURNAL OF VIROLOGY, Jan. 1976, p. 247-253 Copyright 0 1976 American Society for Microbiology
Vol. 17, No. 1 Printed in U.SA.
Transformation of Hamster Kidney Cells by BK Papovavirus DNA KENNETH K. TAKEMOTO* AND MALCOLM A. MARTIN National Institute of Allergy and Infectious Diseases, Laboratory of Viral Diseases and Laboratory of Biology of Viruses, Bethesda, Maryland 20014 Received for publication 25 July 1975
Supercoiled BK papovavirus DNA was shown to transform hamster kidney cells using the calcium phosphate co-precipitation technique. The transformed cells contained intranuclear T-antigen(s) and rescuable virus and produced progressively growing tumors when inoculated into hamsters. A novel finding was the production in tumor-bearing animals of antinuclear antibody, which reacted against normal, untransformed cells; in addition, tumor serum contained antibody against virus-specific T-antigen(s). Since the discovery of BK papovavirus (BKV) in 1971 by Gardner et al. (2), a number of studies on the biological and biochemical properties of the virus have been reported. Biologically, BKV appears to be only weakly oncogenic for hamsters; Shah et al. (16) reported tumor formation in only 1 out of 52 hamsters that had been inoculated with the virus at birth. In this laboratory, no tumors have developed in newborn hamsters inoculated by various routes with BKV or a closely related virus isolated from the urine of a patient with Wiskott-Aldrich syndrome (18). Transformation of a continuous hamster cell line (BHK-21) by BKV has been observed by Major and DiMayorca (8); this study did not report virus-specific T-antigens in the transformed cells, nor was virus rescue attempted. In a more recent study by Portolani et al. (13), evidence for transformation of hamster cells by BKV was presented; i.e., the transformed cells contained T-antigens and produced tumors when inoculated into hamsters, and BKV was rescued from the transformed cells. In the study by Portolani et al. (13), however, transformation appeared to be inefficient, as judged by the number of cell passages (eight) which were required before transformed colonies appeared. As reported here, transformation with BKV DNA, on the other hand, appears to be more efficient and occurs more rapidly than with infectious virus. Cells transformed by BKV DNA possessed properties of transformed cells; i.e., they synthesized T-antigens, contained rescuable virus, and produced progressively growing tumors when injected into hamsters. An unusual finding was the presence of antinuclear antibody (ANA) to normal cells in tumor
serum; antibody to T-antigens was also de-
tected in the tumor serum. MATERIALS AND METHODS Virus growth and purification. BKV was grown in WI-38 fibroblast cells which were cultured in roller bottles. The method for virus purification was the same as previously reported (9). In brief, infected cells were collected, suspended in 2.5% deoxycholate and 0.25% trypsin, incubated for 30 min at 37 C, layered over a CsCl cushion (1.34 g/cm3), and then spun in an SW25. 1 rotor at 23,000 rpm for 90 min. The virus band that formed below the interface was collected and banded in CsCl by isopycnic density centrifugation at 35,000 rpm for 16 h. Cells. WI-38 fibroblast cells were used between 20 and 30 passages. Primary cultures of hamster kidney (HK) cells were prepared by trypsinization of kidneys from 10-day-old hamsters. Human fetal brain cultures were prepared as previously described (16). CV-1 cells, a continuous line of African green monkey cells, were originally obtained from the American Type Culture Collection. The medium for all cells was Eagle medium supplemented with 10% fetal bovine serum. Antiserum. The following sera were used in various tests. (i) Simian virus 40 (SV40) T-antibody was from hamsters with a transplantable SV40 tumor maintained in this laboratory. (ii) JC T-antibody was from hamsters with a transplantable JC virus-induced fibrosarcoma derived in this laboratory; this tumor developed in a hamster that had been inoculated at birth with JC virus, a papovavirus isolated by Padgett et al. (11) from brain tissue of a case of progressive multifocal leukoencephalopathy. (iii) SV40 anti-U serum was a generous gift of A. Lewis. The method for preparing this serum has been described (7). FA tests. The indirect procedure for fluorescent antibody (FA) tests was used, using goat anti-hamster globulin. Preparation of viral DNA. BKV, which had been 247
248
TAKEMOTO AND MARTIN
purified by isopycnic centrifugation in CsCl, was suspended in 0.1 M NaCl, 0.05 M Tris-hydrochloride, pH 8.0, 0.025 M EDTA, and 0.4% sodium dodecyl sulfate and treated with self-digested Pronase (100 Ag/ml) for 16 h at 37 C. The preparation was then extracted with an equal volume of phenol, dialyzed against 0.01 M Tris-hydrochloride, pH 7.5, and banded in CsCl containing ethidium bromide (300 ,g/ml) (15). Supercoiled BKV DNA was collected from the gradient, treated with isopropanol to remove the ethidium bromide, and dialyzed against 0.01 M Tris-hydrochloride, pH 7.5. Transformation with BKV DNA. BKV DNA (2 ,gg/ml) was suspended in Tris-buffered saline, 0.125 M CaCl2, and salmon sperm DNA (8 ,g/ml) as described by Graham and van der Eb (4). One-milliliter aliquots were added to subconfluent primary HK cells for approximately 8 h. The cultures were then washed once, and fresh medium was added. When infected and control cultures grew to confluence 5 days later, they were trypsinized and reseeded. Controls for these experiments consisted of cultures exposed to salmon sperm DNA only and untreated HK cultures.
J. VIROL.
RESULTS Establishment of transformed lines. Approximately 3 to 4 weeks after initial exposure of cells to viral DNA, colonies of cells epitheloid in morphology appeared in the culture. These colonies consisted of cells that grew more densely than surrounding cells and were readily visible without microscope examination. Uninfected, control cultures or cultures exposed to salmon sperm DNA did not contain such colonies. Upon subculture, rapidly growing, epithelial-like cells quickly became the predominant cell type, and the cultures thereafter could be transferred every week at a 1:5 split ratio. Cover slip cultures were prepared and stained by the FA method with SV40 and JC T-antibody; both antisera stained the cells equally well, with the antigen localized exclusively in the nucleus with nucleolar sparing (Fig. 1). One hundred percent of the cells contained T-anti-
FIG. 1. Immunofluorescent demonstration of intranuclear T-antigen(s) of BK-HK-transformed cells. Cells were reacted by the indirect procedure with SV40 T-antibody. x400.
VOL. 17, 1976
gen; viral antigen was not detected with antiBKV rabbit serum. The transformed HK cells will hereinafter be referred to as BK-HK cells. These cells were tested for pleuropneumonialike organisms and have remained negative up to the present time. FA tests for T- and U-antigens. Two intranuclear antigens (T and U) are induced by SV40 (7, 12). Since BKV is known to induce the synthesis of serologically indistinguishable Tantigens in lytic (17) as well as transformed cells (13, 16), it was of interest to determine whether U-antigen was also synthesized in cells infected or transformed by BKV. Cover slip cultures of human embryonic kidney (HEK; Flow Laboratories, Rockville, Md.) infected by BKV for 72 h were examined by the FA procedure for T- and U-antigens. As noted previously (17) and shown in Fig. 2 (top), BKV-infected cells synthesize intranuclear antigens, which react with SV40 T-antibody. When tested against anti-U antibody, BKV-infected cells were also positive (Fig. 2, bottom). BKV and SV40 thus induce the synthesis of serologically related T- and U-antigens. Although a strong reaction of U-antibody with lyrically infected cells was seen, BK-HKtransformed cells stained only weakly with anti-U serum. Similar weak reactivity and variability of staining was also noted among different lines of SV40-transformed cells by Lewis and Rowe (7). Cloning of BK-HK cells. Transformed BKHK cells were plated in 60-mm plastic petri dishes at sufficiently low cell numbers to derive isolated cell colonies. The plating efficiency ranged from 15 to 40%. Most of the colonies were similar in appearance and consisted of epithelial-like cells growing in multiple cell layers (Fig. 3). Cells from six different colonies were removed with trypsin by the method of Puck et al. (14) and were established as clonal lines. All were positive for T-antigen when examined by FA with SV40 T-antibody. In addition, BKV viral DNA has been detected in three clonal lines that were tested, and these results will be reported in a subsequent paper (P. Howley et al., manuscript in preparation). Rescue of BKV. BKV was rescued from BK-HK cells by fusion with permissive HEK cells using UV-irradiated Sendai virus according to a previously described procedure (19). After 3 weeks, the culture degenerated without specific viral cytopathic changes; supernatant fluids were negative for viral hemagglutinin with human type 0 erythrocytes. When clarified fluids were inoculated onto primary human
TRANSFORMATION WITH BKV DNA
249
fetal brain cell cultures, cytopathology characteristic of BKV (17) was noted by day 12, with extensive cytoplasmic vacuolization and rounding of the cells. When cellular destruction was complete, the supernatant fluid contained hemagglutinin at a titer of 1:160; this reaction was inhibited by BKV antiserum, but not by SV40 or JC antibody, proving that the rescued virus was indeed BKV. Tumor production by BK-HK cells. Twelve 5-day-old hamsters were inoculated subcutaneously with 106 uncloned BK-HK cells. Progressively growing tumors developed by 3 to 4 weeks in all of the animals. A tumor was excised and trypsinized, and tumor cells were grown in cell culture. The tumor cells again were positive by FA tests with SV40 and JC T-antibody. Production of T-antibody and unique ANA to normal cells by tumor-bearing animals. Six weeks after inoculation of transformed cells, the hamsters had developed tumors between 1 to 2 cm in diameter. Serum was obtained by intraorbital bleeding, and the pooled serum was tested against BK-HK cells as well as an SV40-transformed human cell line, WI-18 VA2 (obtained from the Wistar Institute). Both cell lines gave strongly positive intranuclear FA reactions. When tumor serum was reacted with uninfected HEK or CV-1 cells, an equally strong fluorescence was observed, and, like T-antigens, the staining was confined to the nucleus, with the nucleolus being spared (Fig. 4). Subsequent serum samples taken from the same animals over a 4-month period always were positive for ANA against normal HEK cells. A difference in antibody titers against transformed cells (BK-HK or WI-18 Va2) and HEK cells was noted (Table 1), indicating that the animals may have been responding to two different kinds of antigens. Subsequent experiments (in preparation) have shown that the ANA is directed against a nuclear component present in the nuclei of a variety of different kinds of cells from different species, including normal hamster cells. Because ANA was detected in pooled sera, it wag important to determine whether all of the tumored animals were producing ANA. The hamsters were therefore individually bled, and the serum was tested. Sera from 11 of the 12 hamsters were positive for ANA at titers of 1:20 to 1:80 using uninfected HEK cells. The single animal whose serum was negative for ANA had a titer of 1:40 against SV40-transformed cells; this result, together with the data above showing a relative difference in titers of ANA and T-antibody (Table 1), is perhaps indicative of
250
TAKEMOTO AND MARTIN
J. VIROL.
FIG. 2. HEK cells infected for 72 h and stained with SV40 T-antibody (top; x100) and SV40 U-antibody (bottom; x 400).
an immune response to two different kinds of antigens, one against T-antigens and the other against a normal cellular component.
DISCUSSION Numerous attempts in this laboratory to transform rodent cells (mouse, rat, or hamster) in culture with BKV have failed. FA studies on the induction of T-antigen in these cells have
also yielded negative results. Since T-antigen is apparently the only reliable marker for evidence of transformation by the papovavirus group, failure to detect T-antigens is indicative of the relative inefficiency of transformation by BKV. This finding is not a property of BKV only; three other recently isolated papovavirus strains from Wiskott-Aldrich patients (12), which are closely related to BKV, have also
TRANSFORMATION WITH BKV DNA
VOL. 17, 1976
251
0'*
FIG. 3. Colony of epithelial-like BK-HK cells, 14 days after plating. x 40.
FIG. 4. Immunofluorescent reaction of ANA in serum from hamsters with BK-HK tumors. Uninfected HEK cells were reacted with tumor serum by the FA procedure. Note the similarity of intense, intranuclear reaction with that usually seen in SV40-transformed cells. Cytoplasmic fluorescence is absent. x 200.
252
TAKEMOTO AND MARTIN
J. VIROL.
TABLE 1. Comparative T-antibody and ANA titers of pooled serum from hamsters with BK-HK tumors Time after initial inoculation (weeks)
T-antibody
ANA
celltieaier titer titer"
6
160
10
10 14
320 320 320
80 40 40
18C
aReciprocal of highest serum dilution giving 2+ fluorescence against the SV40-transformed human cell line WI-18 Va2. Reciprocal of highest serum dilution giving 2+ fluorescence against the uninfected HEK cell line or monkey cell line CV-1. c Serum from two surviving animals.
behaved in a similar fashion, and transformation by these viruses has not been successful. In vivo, BKV as well as the Wiskott-Aldrich isolates also appear to be only weakly oncogenic. We have injected high-titer, CsCl-banded BKV and related strains into hamsters and mice by various routes without the production of tumors. The experiments of Shah et al. (16) also support the conclusion that BKV is not a highly oncogenic virus. Out of 52 hamsters inoculated at birth, a single animal developed a fibrosarcoma after 8 months. This is in sharp contrast to the highly oncogenic SV40, which produces tumors in virtually 100% of hamsters (1). The JC papovavirus isolated from a case of progressive multifocal leukoencephalopathy (11) is also highly oncogenic and produces tumors in over 50% of hamsters inoculated intracerebrally, intraperitoneally, or subcutaneously (20; Walker, personal communication). In view of the fact that all three viruses, SV40, BKV, and JC, induce common T-antigens (13, 16, 17, 20) and therefore can carry out the early functions critical for transformation and oncogenicity, this difference in the lack of oncogenicity of BKV is puzzling. One possible explanation is that BKV virions do not penetrate or become uncoated very efficiently in rodent cells. The relative ease with which BKV viral DNA transformed hamster cells supports this hypothesis. The malignancy of the transformed cells was proven by the production of tumors in 100% of hamsters inoculated with the cells. An unusual and unique finding in tumor-bearing animals was the production of antibodies not only against the virus-specific T-antigens, but against intranuclear antigens in normal cells as well. To our knowledge, autoantibodies due to tumor development in animals have never been reported. In further studies (in preparation) we
have shown that BKV tumors transplanted into other hamsters also evoke ANA in recipients. The mechanism concerning ANA development, as well as the antigen involved in the reaction, is under further investigation. One possible explanation for the phenomenon is that T-antigens may not be entirely virus coded or specified, and that they may in large part be cellular proteins. We have recently reported that, although the T-antigens of BKV and SV40 are immunologically very similar, the regions of homology in the genomes of the two viruses are entirely in the endonuclease-cleaved fragments of SV40 that specify "late" viral gene products, i.e., capsid proteins (6). No homology was detected in the "early" regions of the SV40 DNA, which presumably are responsible for the production of various antigens (T, U, TSTA). This finding raises important questions concerning the origin and nature of T and U antigens. If they are mainly cellular proteins, a small segment must be virus specified to account for virus specificity; i.e., antigens induced by unrelated viruses (polyoma, adeno, and SV40) do not cross-react. Whether the finding of ANA in hamsters with tumors produced as a result of inoculation of BKV-transformed cells is a more general phenomenon and may be found in animals with tumors induced by other viruses remains to be determined. We have not observed ANA in serum from hamsters with SV40, polyoma, JC-papovavirus, or adeno-12 tumors, nor have we detected ANA in normal hamsters housed in the same environment during the course of these experiments. Recently, we have also tested serum from a BKV tumor induced in vivo by virus inoculation (kindly provided by K. Shah); pooled sera from hamsters with this tumor contain T-antibody but not ANA. Transformation of rat cells with sheared fragments of adeno-5 DNA, using the calcium phosphate precipitation method, has recently been reported by Graham et al. (5), enabling these workers to determine the size and location of the transforming activity within the viral genome. Graham et al. (3) have also been able to transform cells with fragments of SV40; approximately three-fourths of the viral genome was found to be capable of transformation. Similar experiments are underway in this laboratory with BKV DNA fragments obtained after restriction endonuclease cleavage; this procedure yields only four discrete fragments (5, 10). Such experiments should provide information regarding minimal size and precise location of the transforming activity of these viruses.
TRANSFORMATION WITH BKV DNA
VOL. 17, 1976 LITERATURE CITED 1. Eddy, B. E. 1964. Simian virus 40 (SV-40): an oncogenic virus. Prog. Exp. Tumor Res. 4:1-26. 2. Gardner, S. D.. A. M. Field, D. V. Coleman, and B. Hulme. 1971. New human papovavirus (B. K.) isolated from urine after renal transplantation. Lancet 1:1253- 1257. 3. Graham, F. L., P. J. Abrahams, C. Mulder, H. L. Heijneker, S. 0. Warnaar, F. A. J. deVries, W. Fiers, and A. J. van der Eb. 1974. Studies on in vitro transformation by DNA and DNA fragments of human adenoviruses and simian virus 40. Cold Spring Harbor Symp. Quant. Biol. 39:637-650. 4. Graham, F. L., and A. J. van der Eb. 1973. A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 52:456-467. 5. Graham, F. L., A. J. van der Eb, and H. L. Heijneker. 1974. Size and location of the transforming region in human adenovirus type 5 DNA. Nature (London) 251:687-691. 6. Khoury, G., P. M. Howley, C. Garon, M. F. Mullarkey, K. K. Takemoto, and M. A. Martin. 1975. An analysis of the homology and relationship between the genomes of papovaviruses BKV and SV40. Proc. Natl. Acad. Sci. U.S.A. 72:2563-2567. 7. Lewis, A. M., Jr., and W. P. Rowe. 1971. Studies on nondefective adenovirus-simian virus 40 hybrid viruses. I. A newly characterized simian virus 40 antigen induced by Ad2 + ND virus. J. Virol. 7:189-197. 8. Major, E. O., and G. DiMayorca. 1973. Malignant transformation of BHK-21 clone 13 cells by BK virus, a human papovavirus. Proc. Natl. Acad. Sci. U.S.A. 70:3210-3212. 9. Mullarkey, M. F., J. F. Hruska, and K. K. Takemoto. 1974. Comparison of two human papovaviruses with simian virus 40 by structural protein and antigenic analysis. J. Virol. 13:1014-1019. 10. Osborn, J. E., S. M. Robertson, B. L. Padgett, G. M. ZuRhein, D. L. Walker, and B. Weisblum. 1974. Comparison of JC and BK human papovaviruses with simian virus 40: restriction endonuclease digestion and
11.
12. 13. 14.
15.
16.
17.
18.
19.
20.
253
gel electrophoresis of resultant fragments. J. Virol. 13:614-622. Padgett, B. L., D. L. Walker, G. M. ZuRhein, R. J. Eckroade, and B. H. Delsel. 1971. Cultivation of papova-like virus from human brain with progressive multifocal leukoencephalopathy. Lancet 1:1257-1260. Pope, J. H., and W. P. Rowe. 1964. Detection of specific antigen in SV40-transformed cells by immunofluorescence. J. Exp. Med. 120:121-128. Portolani, M., G. Barbanti-Brodano, and M. LaPlaca. 1975. Malignant transformation of hamster kidney cells bv BK virus. J. Virol. 15:420-422. Puck, T. T., P. I. Marcus, and S. J. Cieciura. 1956. Clonal growth of mammalian cells in vitro. Growth characteristics of colonies from single HeLa cells with and without a "feeder" layer. J. Exp. Med. 103:273-284. Radloff, R., W. Bauer, and J. Vinograd. 1967. A dyebuoyant-density method for the detection and isolation of closed circular duplex DNA: the closed circular DNA in HeLa cells. Proc. Natl. Acad. Sci. U.S.A. 57:1514-1521. Shah, K. V., R. W. Daniel, and J. Strandberg. 1975. Sarcoma in a hamster inoculated with a human papovavirus, BK virus. J. Natl. Cancer Inst. 54:945-950. Takemoto, K. K., and M. F. Mullarkey. 1973. Human papovavirus, BK strain: biological studies including antigenic relationship to simian virus 40. J. Virol. 12:625-631. Takemoto, K. K., A. S. Rabson, M. F. Mullarkey, R. M. Blaese, C. F. Garon, and D. Nelson. 1974. Isolation of papovavirus from brain tumor and urine of a patient with Wiskott-Aldrich syndrome. J. Natl. Cancer Inst. 53:1205-1207. Takemoto, K. K., G. J. Todaro, and K. Habel. 1968. Recovery of SV40 with genetic markers of original inducing virus from SV40 transformed mouse cells. Virology 35:1-8. Walker, D. L., B. L. Padgett, G. M. ZuRhein, A. E. Albert, and R. F. Marsh. 1973. Human papovavirus (JC): induction of brain tumors in hamsters. Science 181:674-676.
Vol. 17, No. 1 Printed in U.SA.
Transformation of Hamster Kidney Cells by BK Papovavirus DNA KENNETH K. TAKEMOTO* AND MALCOLM A. MARTIN National Institute of Allergy and Infectious Diseases, Laboratory of Viral Diseases and Laboratory of Biology of Viruses, Bethesda, Maryland 20014 Received for publication 25 July 1975
Supercoiled BK papovavirus DNA was shown to transform hamster kidney cells using the calcium phosphate co-precipitation technique. The transformed cells contained intranuclear T-antigen(s) and rescuable virus and produced progressively growing tumors when inoculated into hamsters. A novel finding was the production in tumor-bearing animals of antinuclear antibody, which reacted against normal, untransformed cells; in addition, tumor serum contained antibody against virus-specific T-antigen(s). Since the discovery of BK papovavirus (BKV) in 1971 by Gardner et al. (2), a number of studies on the biological and biochemical properties of the virus have been reported. Biologically, BKV appears to be only weakly oncogenic for hamsters; Shah et al. (16) reported tumor formation in only 1 out of 52 hamsters that had been inoculated with the virus at birth. In this laboratory, no tumors have developed in newborn hamsters inoculated by various routes with BKV or a closely related virus isolated from the urine of a patient with Wiskott-Aldrich syndrome (18). Transformation of a continuous hamster cell line (BHK-21) by BKV has been observed by Major and DiMayorca (8); this study did not report virus-specific T-antigens in the transformed cells, nor was virus rescue attempted. In a more recent study by Portolani et al. (13), evidence for transformation of hamster cells by BKV was presented; i.e., the transformed cells contained T-antigens and produced tumors when inoculated into hamsters, and BKV was rescued from the transformed cells. In the study by Portolani et al. (13), however, transformation appeared to be inefficient, as judged by the number of cell passages (eight) which were required before transformed colonies appeared. As reported here, transformation with BKV DNA, on the other hand, appears to be more efficient and occurs more rapidly than with infectious virus. Cells transformed by BKV DNA possessed properties of transformed cells; i.e., they synthesized T-antigens, contained rescuable virus, and produced progressively growing tumors when injected into hamsters. An unusual finding was the presence of antinuclear antibody (ANA) to normal cells in tumor
serum; antibody to T-antigens was also de-
tected in the tumor serum. MATERIALS AND METHODS Virus growth and purification. BKV was grown in WI-38 fibroblast cells which were cultured in roller bottles. The method for virus purification was the same as previously reported (9). In brief, infected cells were collected, suspended in 2.5% deoxycholate and 0.25% trypsin, incubated for 30 min at 37 C, layered over a CsCl cushion (1.34 g/cm3), and then spun in an SW25. 1 rotor at 23,000 rpm for 90 min. The virus band that formed below the interface was collected and banded in CsCl by isopycnic density centrifugation at 35,000 rpm for 16 h. Cells. WI-38 fibroblast cells were used between 20 and 30 passages. Primary cultures of hamster kidney (HK) cells were prepared by trypsinization of kidneys from 10-day-old hamsters. Human fetal brain cultures were prepared as previously described (16). CV-1 cells, a continuous line of African green monkey cells, were originally obtained from the American Type Culture Collection. The medium for all cells was Eagle medium supplemented with 10% fetal bovine serum. Antiserum. The following sera were used in various tests. (i) Simian virus 40 (SV40) T-antibody was from hamsters with a transplantable SV40 tumor maintained in this laboratory. (ii) JC T-antibody was from hamsters with a transplantable JC virus-induced fibrosarcoma derived in this laboratory; this tumor developed in a hamster that had been inoculated at birth with JC virus, a papovavirus isolated by Padgett et al. (11) from brain tissue of a case of progressive multifocal leukoencephalopathy. (iii) SV40 anti-U serum was a generous gift of A. Lewis. The method for preparing this serum has been described (7). FA tests. The indirect procedure for fluorescent antibody (FA) tests was used, using goat anti-hamster globulin. Preparation of viral DNA. BKV, which had been 247
248
TAKEMOTO AND MARTIN
purified by isopycnic centrifugation in CsCl, was suspended in 0.1 M NaCl, 0.05 M Tris-hydrochloride, pH 8.0, 0.025 M EDTA, and 0.4% sodium dodecyl sulfate and treated with self-digested Pronase (100 Ag/ml) for 16 h at 37 C. The preparation was then extracted with an equal volume of phenol, dialyzed against 0.01 M Tris-hydrochloride, pH 7.5, and banded in CsCl containing ethidium bromide (300 ,g/ml) (15). Supercoiled BKV DNA was collected from the gradient, treated with isopropanol to remove the ethidium bromide, and dialyzed against 0.01 M Tris-hydrochloride, pH 7.5. Transformation with BKV DNA. BKV DNA (2 ,gg/ml) was suspended in Tris-buffered saline, 0.125 M CaCl2, and salmon sperm DNA (8 ,g/ml) as described by Graham and van der Eb (4). One-milliliter aliquots were added to subconfluent primary HK cells for approximately 8 h. The cultures were then washed once, and fresh medium was added. When infected and control cultures grew to confluence 5 days later, they were trypsinized and reseeded. Controls for these experiments consisted of cultures exposed to salmon sperm DNA only and untreated HK cultures.
J. VIROL.
RESULTS Establishment of transformed lines. Approximately 3 to 4 weeks after initial exposure of cells to viral DNA, colonies of cells epitheloid in morphology appeared in the culture. These colonies consisted of cells that grew more densely than surrounding cells and were readily visible without microscope examination. Uninfected, control cultures or cultures exposed to salmon sperm DNA did not contain such colonies. Upon subculture, rapidly growing, epithelial-like cells quickly became the predominant cell type, and the cultures thereafter could be transferred every week at a 1:5 split ratio. Cover slip cultures were prepared and stained by the FA method with SV40 and JC T-antibody; both antisera stained the cells equally well, with the antigen localized exclusively in the nucleus with nucleolar sparing (Fig. 1). One hundred percent of the cells contained T-anti-
FIG. 1. Immunofluorescent demonstration of intranuclear T-antigen(s) of BK-HK-transformed cells. Cells were reacted by the indirect procedure with SV40 T-antibody. x400.
VOL. 17, 1976
gen; viral antigen was not detected with antiBKV rabbit serum. The transformed HK cells will hereinafter be referred to as BK-HK cells. These cells were tested for pleuropneumonialike organisms and have remained negative up to the present time. FA tests for T- and U-antigens. Two intranuclear antigens (T and U) are induced by SV40 (7, 12). Since BKV is known to induce the synthesis of serologically indistinguishable Tantigens in lytic (17) as well as transformed cells (13, 16), it was of interest to determine whether U-antigen was also synthesized in cells infected or transformed by BKV. Cover slip cultures of human embryonic kidney (HEK; Flow Laboratories, Rockville, Md.) infected by BKV for 72 h were examined by the FA procedure for T- and U-antigens. As noted previously (17) and shown in Fig. 2 (top), BKV-infected cells synthesize intranuclear antigens, which react with SV40 T-antibody. When tested against anti-U antibody, BKV-infected cells were also positive (Fig. 2, bottom). BKV and SV40 thus induce the synthesis of serologically related T- and U-antigens. Although a strong reaction of U-antibody with lyrically infected cells was seen, BK-HKtransformed cells stained only weakly with anti-U serum. Similar weak reactivity and variability of staining was also noted among different lines of SV40-transformed cells by Lewis and Rowe (7). Cloning of BK-HK cells. Transformed BKHK cells were plated in 60-mm plastic petri dishes at sufficiently low cell numbers to derive isolated cell colonies. The plating efficiency ranged from 15 to 40%. Most of the colonies were similar in appearance and consisted of epithelial-like cells growing in multiple cell layers (Fig. 3). Cells from six different colonies were removed with trypsin by the method of Puck et al. (14) and were established as clonal lines. All were positive for T-antigen when examined by FA with SV40 T-antibody. In addition, BKV viral DNA has been detected in three clonal lines that were tested, and these results will be reported in a subsequent paper (P. Howley et al., manuscript in preparation). Rescue of BKV. BKV was rescued from BK-HK cells by fusion with permissive HEK cells using UV-irradiated Sendai virus according to a previously described procedure (19). After 3 weeks, the culture degenerated without specific viral cytopathic changes; supernatant fluids were negative for viral hemagglutinin with human type 0 erythrocytes. When clarified fluids were inoculated onto primary human
TRANSFORMATION WITH BKV DNA
249
fetal brain cell cultures, cytopathology characteristic of BKV (17) was noted by day 12, with extensive cytoplasmic vacuolization and rounding of the cells. When cellular destruction was complete, the supernatant fluid contained hemagglutinin at a titer of 1:160; this reaction was inhibited by BKV antiserum, but not by SV40 or JC antibody, proving that the rescued virus was indeed BKV. Tumor production by BK-HK cells. Twelve 5-day-old hamsters were inoculated subcutaneously with 106 uncloned BK-HK cells. Progressively growing tumors developed by 3 to 4 weeks in all of the animals. A tumor was excised and trypsinized, and tumor cells were grown in cell culture. The tumor cells again were positive by FA tests with SV40 and JC T-antibody. Production of T-antibody and unique ANA to normal cells by tumor-bearing animals. Six weeks after inoculation of transformed cells, the hamsters had developed tumors between 1 to 2 cm in diameter. Serum was obtained by intraorbital bleeding, and the pooled serum was tested against BK-HK cells as well as an SV40-transformed human cell line, WI-18 VA2 (obtained from the Wistar Institute). Both cell lines gave strongly positive intranuclear FA reactions. When tumor serum was reacted with uninfected HEK or CV-1 cells, an equally strong fluorescence was observed, and, like T-antigens, the staining was confined to the nucleus, with the nucleolus being spared (Fig. 4). Subsequent serum samples taken from the same animals over a 4-month period always were positive for ANA against normal HEK cells. A difference in antibody titers against transformed cells (BK-HK or WI-18 Va2) and HEK cells was noted (Table 1), indicating that the animals may have been responding to two different kinds of antigens. Subsequent experiments (in preparation) have shown that the ANA is directed against a nuclear component present in the nuclei of a variety of different kinds of cells from different species, including normal hamster cells. Because ANA was detected in pooled sera, it wag important to determine whether all of the tumored animals were producing ANA. The hamsters were therefore individually bled, and the serum was tested. Sera from 11 of the 12 hamsters were positive for ANA at titers of 1:20 to 1:80 using uninfected HEK cells. The single animal whose serum was negative for ANA had a titer of 1:40 against SV40-transformed cells; this result, together with the data above showing a relative difference in titers of ANA and T-antibody (Table 1), is perhaps indicative of
250
TAKEMOTO AND MARTIN
J. VIROL.
FIG. 2. HEK cells infected for 72 h and stained with SV40 T-antibody (top; x100) and SV40 U-antibody (bottom; x 400).
an immune response to two different kinds of antigens, one against T-antigens and the other against a normal cellular component.
DISCUSSION Numerous attempts in this laboratory to transform rodent cells (mouse, rat, or hamster) in culture with BKV have failed. FA studies on the induction of T-antigen in these cells have
also yielded negative results. Since T-antigen is apparently the only reliable marker for evidence of transformation by the papovavirus group, failure to detect T-antigens is indicative of the relative inefficiency of transformation by BKV. This finding is not a property of BKV only; three other recently isolated papovavirus strains from Wiskott-Aldrich patients (12), which are closely related to BKV, have also
TRANSFORMATION WITH BKV DNA
VOL. 17, 1976
251
0'*
FIG. 3. Colony of epithelial-like BK-HK cells, 14 days after plating. x 40.
FIG. 4. Immunofluorescent reaction of ANA in serum from hamsters with BK-HK tumors. Uninfected HEK cells were reacted with tumor serum by the FA procedure. Note the similarity of intense, intranuclear reaction with that usually seen in SV40-transformed cells. Cytoplasmic fluorescence is absent. x 200.
252
TAKEMOTO AND MARTIN
J. VIROL.
TABLE 1. Comparative T-antibody and ANA titers of pooled serum from hamsters with BK-HK tumors Time after initial inoculation (weeks)
T-antibody
ANA
celltieaier titer titer"
6
160
10
10 14
320 320 320
80 40 40
18C
aReciprocal of highest serum dilution giving 2+ fluorescence against the SV40-transformed human cell line WI-18 Va2. Reciprocal of highest serum dilution giving 2+ fluorescence against the uninfected HEK cell line or monkey cell line CV-1. c Serum from two surviving animals.
behaved in a similar fashion, and transformation by these viruses has not been successful. In vivo, BKV as well as the Wiskott-Aldrich isolates also appear to be only weakly oncogenic. We have injected high-titer, CsCl-banded BKV and related strains into hamsters and mice by various routes without the production of tumors. The experiments of Shah et al. (16) also support the conclusion that BKV is not a highly oncogenic virus. Out of 52 hamsters inoculated at birth, a single animal developed a fibrosarcoma after 8 months. This is in sharp contrast to the highly oncogenic SV40, which produces tumors in virtually 100% of hamsters (1). The JC papovavirus isolated from a case of progressive multifocal leukoencephalopathy (11) is also highly oncogenic and produces tumors in over 50% of hamsters inoculated intracerebrally, intraperitoneally, or subcutaneously (20; Walker, personal communication). In view of the fact that all three viruses, SV40, BKV, and JC, induce common T-antigens (13, 16, 17, 20) and therefore can carry out the early functions critical for transformation and oncogenicity, this difference in the lack of oncogenicity of BKV is puzzling. One possible explanation is that BKV virions do not penetrate or become uncoated very efficiently in rodent cells. The relative ease with which BKV viral DNA transformed hamster cells supports this hypothesis. The malignancy of the transformed cells was proven by the production of tumors in 100% of hamsters inoculated with the cells. An unusual and unique finding in tumor-bearing animals was the production of antibodies not only against the virus-specific T-antigens, but against intranuclear antigens in normal cells as well. To our knowledge, autoantibodies due to tumor development in animals have never been reported. In further studies (in preparation) we
have shown that BKV tumors transplanted into other hamsters also evoke ANA in recipients. The mechanism concerning ANA development, as well as the antigen involved in the reaction, is under further investigation. One possible explanation for the phenomenon is that T-antigens may not be entirely virus coded or specified, and that they may in large part be cellular proteins. We have recently reported that, although the T-antigens of BKV and SV40 are immunologically very similar, the regions of homology in the genomes of the two viruses are entirely in the endonuclease-cleaved fragments of SV40 that specify "late" viral gene products, i.e., capsid proteins (6). No homology was detected in the "early" regions of the SV40 DNA, which presumably are responsible for the production of various antigens (T, U, TSTA). This finding raises important questions concerning the origin and nature of T and U antigens. If they are mainly cellular proteins, a small segment must be virus specified to account for virus specificity; i.e., antigens induced by unrelated viruses (polyoma, adeno, and SV40) do not cross-react. Whether the finding of ANA in hamsters with tumors produced as a result of inoculation of BKV-transformed cells is a more general phenomenon and may be found in animals with tumors induced by other viruses remains to be determined. We have not observed ANA in serum from hamsters with SV40, polyoma, JC-papovavirus, or adeno-12 tumors, nor have we detected ANA in normal hamsters housed in the same environment during the course of these experiments. Recently, we have also tested serum from a BKV tumor induced in vivo by virus inoculation (kindly provided by K. Shah); pooled sera from hamsters with this tumor contain T-antibody but not ANA. Transformation of rat cells with sheared fragments of adeno-5 DNA, using the calcium phosphate precipitation method, has recently been reported by Graham et al. (5), enabling these workers to determine the size and location of the transforming activity within the viral genome. Graham et al. (3) have also been able to transform cells with fragments of SV40; approximately three-fourths of the viral genome was found to be capable of transformation. Similar experiments are underway in this laboratory with BKV DNA fragments obtained after restriction endonuclease cleavage; this procedure yields only four discrete fragments (5, 10). Such experiments should provide information regarding minimal size and precise location of the transforming activity of these viruses.
TRANSFORMATION WITH BKV DNA
VOL. 17, 1976 LITERATURE CITED 1. Eddy, B. E. 1964. Simian virus 40 (SV-40): an oncogenic virus. Prog. Exp. Tumor Res. 4:1-26. 2. Gardner, S. D.. A. M. Field, D. V. Coleman, and B. Hulme. 1971. New human papovavirus (B. K.) isolated from urine after renal transplantation. Lancet 1:1253- 1257. 3. Graham, F. L., P. J. Abrahams, C. Mulder, H. L. Heijneker, S. 0. Warnaar, F. A. J. deVries, W. Fiers, and A. J. van der Eb. 1974. Studies on in vitro transformation by DNA and DNA fragments of human adenoviruses and simian virus 40. Cold Spring Harbor Symp. Quant. Biol. 39:637-650. 4. Graham, F. L., and A. J. van der Eb. 1973. A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 52:456-467. 5. Graham, F. L., A. J. van der Eb, and H. L. Heijneker. 1974. Size and location of the transforming region in human adenovirus type 5 DNA. Nature (London) 251:687-691. 6. Khoury, G., P. M. Howley, C. Garon, M. F. Mullarkey, K. K. Takemoto, and M. A. Martin. 1975. An analysis of the homology and relationship between the genomes of papovaviruses BKV and SV40. Proc. Natl. Acad. Sci. U.S.A. 72:2563-2567. 7. Lewis, A. M., Jr., and W. P. Rowe. 1971. Studies on nondefective adenovirus-simian virus 40 hybrid viruses. I. A newly characterized simian virus 40 antigen induced by Ad2 + ND virus. J. Virol. 7:189-197. 8. Major, E. O., and G. DiMayorca. 1973. Malignant transformation of BHK-21 clone 13 cells by BK virus, a human papovavirus. Proc. Natl. Acad. Sci. U.S.A. 70:3210-3212. 9. Mullarkey, M. F., J. F. Hruska, and K. K. Takemoto. 1974. Comparison of two human papovaviruses with simian virus 40 by structural protein and antigenic analysis. J. Virol. 13:1014-1019. 10. Osborn, J. E., S. M. Robertson, B. L. Padgett, G. M. ZuRhein, D. L. Walker, and B. Weisblum. 1974. Comparison of JC and BK human papovaviruses with simian virus 40: restriction endonuclease digestion and
11.
12. 13. 14.
15.
16.
17.
18.
19.
20.
253
gel electrophoresis of resultant fragments. J. Virol. 13:614-622. Padgett, B. L., D. L. Walker, G. M. ZuRhein, R. J. Eckroade, and B. H. Delsel. 1971. Cultivation of papova-like virus from human brain with progressive multifocal leukoencephalopathy. Lancet 1:1257-1260. Pope, J. H., and W. P. Rowe. 1964. Detection of specific antigen in SV40-transformed cells by immunofluorescence. J. Exp. Med. 120:121-128. Portolani, M., G. Barbanti-Brodano, and M. LaPlaca. 1975. Malignant transformation of hamster kidney cells bv BK virus. J. Virol. 15:420-422. Puck, T. T., P. I. Marcus, and S. J. Cieciura. 1956. Clonal growth of mammalian cells in vitro. Growth characteristics of colonies from single HeLa cells with and without a "feeder" layer. J. Exp. Med. 103:273-284. Radloff, R., W. Bauer, and J. Vinograd. 1967. A dyebuoyant-density method for the detection and isolation of closed circular duplex DNA: the closed circular DNA in HeLa cells. Proc. Natl. Acad. Sci. U.S.A. 57:1514-1521. Shah, K. V., R. W. Daniel, and J. Strandberg. 1975. Sarcoma in a hamster inoculated with a human papovavirus, BK virus. J. Natl. Cancer Inst. 54:945-950. Takemoto, K. K., and M. F. Mullarkey. 1973. Human papovavirus, BK strain: biological studies including antigenic relationship to simian virus 40. J. Virol. 12:625-631. Takemoto, K. K., A. S. Rabson, M. F. Mullarkey, R. M. Blaese, C. F. Garon, and D. Nelson. 1974. Isolation of papovavirus from brain tumor and urine of a patient with Wiskott-Aldrich syndrome. J. Natl. Cancer Inst. 53:1205-1207. Takemoto, K. K., G. J. Todaro, and K. Habel. 1968. Recovery of SV40 with genetic markers of original inducing virus from SV40 transformed mouse cells. Virology 35:1-8. Walker, D. L., B. L. Padgett, G. M. ZuRhein, A. E. Albert, and R. F. Marsh. 1973. Human papovavirus (JC): induction of brain tumors in hamsters. Science 181:674-676.
