VIROLOGY 75, 319-334 (1976) Biological Activities of Deletion Mutants of Simian Virus 40! WALTER A. SCOTT,? WILLIAM W. BROCKMAN,? anp DANIEL NATHANS?* Department of Microbiology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205 Accepted August 1, 1976 Mutants of Simian virus 40 (SV40) with large deletions in the early or late regions of the genome were tested for biological activity. Deletion mutants lacking portions of both late genes (B/C and D), but with an intact early genomic segment, were able to induce T antigen in infected cells, replicate their DNA in the absence of helper virus, stimulate thymidine incorporation into cellular DNA, and transform mouse and hamster cells. Cells transformed by late deletion mutants were shown to contain the mutant genome by a fusion-complementation rescue procedure. Deletion mutants lacking substantial por- tions of the early genomic region, including those segments where fsA mutants map, lacked all of the above activities. However, both early and late deletion mutants interfered with SV40 DNA replication. INTRODUCTION Cloned deletion mutants of Simian virus 40 (SV40) have been isolated from two sources: from virus passaged serially at high multiplicity of infection (Brockman and Nathans, 1974; Mertz et al., 1974), and from enzymatically cleaved SV40 DNA (Lai and Nathans, 1974a; Mertz et al., 1974; Carbon et al., 1975). Some SV40 dele- tion mutants are viable, whereas others are sufficiently defective to require a helper virus for replication. We have been using cloned defective deletion mutants, whose genomes have been physically mapped, to localize SV40 genes (Lai and Nathans, 1974b; 1976) and to assign the various biological activities of the virus to specific regions of the genome (Brockman and Scott, 1975). In this communication we detail our experiments on the activities of 1 This is Publication Number 18 in a series on the Genome of Simian Virus 40. Publication 17 is Lai and Nathans (1975b). ? Present address: Department of Biochemistry, University of Miami, School of Medicine, Miami, Florida 33152. 3 Present address: Department of Microbiology, University of Michigan, School of Medicine, Ann Arbor, Michigan 48104. + Author to whom reprint requests should be ad- dressed. 319 Copyright © 1976 by Academic Press, Inc. All rights of reproduction in any form reserved. mutants containing large deletions of either the early or late region of the ge- nome. From the results we conclude that the early region plus immediately contig- uous segments of DNA are sufficient for viral DNA replication, stimulation of thy- midine incorporation into cellular DNA, T antigen formation, and cell transforma- tion. Moreover, cells transformed by a deletion mutant have been shown to yield the mutant genome after cell fusion and complementation. In contrast to the above activities, interference with SV40 DNA replication by a mutant virus does not ap- pear to be dependent on the function of a specific gene product. MATERIALS AND METHODS Virus and cells. Wild-type (wt) SV40 was derived from plaque-purified small plaque virus (No. 776) grown in the BSC-1 line of African green monkey cells, as de- tailed previously (Danna and Nathans, 1971). Evolutionary variants ev-1117, -1119, and -1114 are defective viruses cloned from serially passaged virus stocks (Brock- man et al., 1975); dl-1007 is a cloned, con- structed deletion mutant (Lai and Na- thans, 1974a). Each mutant was recloned at least once prior to preparation of virus stocks, as detailed earlier (Brockman et 320 al., 1975). Temperature-sensitive mutants of SV40 (tsA28, tsB4, and tsC219) were generously supplied by Peter Tegtmeyer (Tegtmeyer, 1972) or Robert Martin (Chou and Martin, 1974). BSC-40 cells (Brock- man and Nathans, 1974) were used for the growth of virus at elevated temperature. BALB/c-3T3 cells were obtained from G. Todaro and were grown in minimal Ea- gle’s medium (MEM) with 20% fetal calf serum. Chinese hamster lung (CHL) cells were secondary cultures obtained from Janice Chou and Robert Martin (Martin and Chou, 1975), and were grown in MEM with 10% fetal calf serum. Purification of deletion mutants. Stocks of SV40 deletion mutants contain comple- menting ¢s helper virus. To purify the mu- tant, clarified cell lysates were layered over a CsCl cushion (p = 1.34 g/ml), and virus particles were centrifuged into the CsCl layer. This entire layer was then added to fresh CsCl solution (final density, 1.34 g/ml), and after equilibrium centrifu- gation the light virus band was recovered and recentrifuged to equilibrium twice more to exclude most of the denser helper virus. The purified virus was then dialyzed against phosphate-buffered saline, treated with a trace of chloroform, and diluted into minimal medium with 2% fetal calf serum and stored frozen at —30°. Titration of deletion mutants. The con- centration of infectious mutant virus was assayed by infectious center complementa- tion plaque titration in the presence of an excess of a complementing ¢s mutant (Brockman and Nathans, 1974), as noted in the text. The titer is expressed as complementation plaque-forming units (cPFU). Mutant DNA was labeled with ”P and then purified by Hirt extraction (Hirt, 1967) and electrophoresis in 1.4% agarose gel slabs, as described elsewhere (Brock- man and Nathans, 1974). SV40 T antigen was detected in cells infected with mutant or wt DNA (Mc- Cutchan and Pagano, 1968) by the indirect immunofluorescence procedure using anti- T serum from Flow Laboratories and fluo- rescein-tagged anti-hamster serum from Nutritional Biochemicals Corp. For this SCOTT, BROCKMAN AND NATHANS purpose about 25 ng of electrophoretically purified DNA was used to infect approxi- mately 104 cells growing on a glass slide and about 1000 cells were scored for nu- clear fluorescence. Replication of mutant DNA. BSC-1 cell monolayers in 6-mm microwells were in- fected with purified mutant virus over a range of input multiplicities in the pres- ence or absence of wt SV40 at a multiplic- ity of 4 PFU per cell. After 24 and 48 hr, 5 uC of [*P Jorthophosphate was added, and at 72 hr, viral DNA was extracted by Hirt’s procedure. In every case a constant, known amount of [H]SV40 DNA was added with the 0.1 ml of lysing solution to monitor recovery of SV40 DNA. After cen- trifugation of the lysate, RNase treat- ment, and phenol extraction, 10 yg of tRNA was added, the DNA was precipi- tated with ethanol, and the precipitate taken up in a small volume of 15 mM NaCl-1.5 mM Na citrate, pH 7.0. An ali- quot of each sample was then subjected to electrophoresis in 1.4% agarose gel slabs to separate mutant from helper virus DNA, and all forms of unit length DNA weje pooled and counted as were the forms of deleted DNA present. At the same time the recovery of added [7HJDNA was as- sessed, and the [P]DNA values normal- ized to the original amount of [HJDNA present. Measurement of virus-stimulated incor- poration of thymidine into cellular DNA. BALB/c-3T3 cells were uniformly labeled by allowing them to grow in 10-cm dishes for three or four generations in 1 uCi of [4C]thymidine (50 Ci/mol) and _ then seeded in 6-mm microwells. One day after the culture had reached confluence, the medium was changed to 0.1 ml of MEM supplemented with 2% fetal calf serum which had been heated at 70° for 30 min. Forty-eight hours after this medium change, the first infection was performed by adding 25 yl of diluted virus stock with- out removing the medium or disturbing the monolayer. Virus stocks were diluted in MEM to give the appropriate virus mul- tiplicity and a final concentration of un- heated serum that was not more than 0.2%. Other cultures were infected in the SV40 DELETION MUTANTS same way at later times for other time points in the experiment. For the first time point, DNA was pulse-labeled by remov- ing the medium and by adding 50 wl of MEM containing 2% heated fetal calf se- rum and 40 yuCi/ml of [H]thymidine (20 Ci/mmol). After 1 hr the cultures were rinsed with cold Tris-buffered saline and harvested by the Hirt method. Culture ly- sates were transferred into small test tubes using a Pasteur pipet, and the super- natant from the Hirt fractionation proce- dure, which contained few counts, was dis- carded. The precipitate was dissolved in a few drops of 3 N NH,OH, and transferred to Whatman 3 MM filter disks. The disks were dried and counted. *H counts from each well were normalized to a constant amount of ['*]DNA to correct for variation in recovery of cell DNA during the lysis and fractionation procedures. Cell transformation. Subconfluent BALB/c-3T3 cells (1.5 x 10° cells) or CHL cells (1 x 105 cells) growing in 1.5-cm wells at 37° were infected at a series of multiplic- ities with SV40 or a purified deletion mu- tant. Twenty-four hours later the cells of each well were trypsinized and transferred to a 10-cm petri dish in MEM with 5% fetal calf serum. Three weeks later (3T3) or 4 weeks later (CHL), piled-up colonies were scored as transformants. Several trans- formed colonies were isolated and cloned twice by plating a dilute cell suspension and selecting colonies originating from single cells. Rescue of virus from transformed BALB/c-3T3 cells by fusion-complementa- tion. This procedure is based on the com- plementation-plaquing procedure used for cloning deletion mutants (Brockman and Nathans, 1974). A defective genome in transformed cells is induced by cell fusion with permissive cells and complemented by a tsA helper virus, thus leading to the interdependent growth of the defective ge- nome and the helper. In practice, cultures of transformed 3T3 cells at 80-90% conflu- ence were trypsinized and 10+ cells were seeded in 1 ml of MEM (containing 10% fetal calf serum) in 1.5-cm culture wells. After these cells had attached (6 hr, 37°), 3 x 10° BSC-40 cells were added to each 321 well. After an additional 12 hr at 37°, the cultures were infected at a multiplicity of 4 with a tsA mutant for a period of 2 hr, the medium was removed, and the monolayer was subjected to fusion with uv-inacti- vated Sendai virus (Microbiological Asso- ciates) as described by Davidson (1967). Briefly, the monolayer was washed twice with cold MEM without serum and then exposed to uv-inactivated Sendai virus (200 hemagglutinating units in 0.1 ml of MEM without serum) and incubated on ice for 20-25 min. At the end of the incuba- tion, the Sendai virus was removed and the monolayer again was rinsed twice with cold MEM without serum. MEM (0.2 ml), which had been preequilibrated at 37°, was added and the culture was further incu- bated at 37° for 30 min, following which 1 mi of MEM with 3% fetal calf serum was added and the incubation continued at 37° for 12-18 hr. At this time the culture was trypsinized, transferred to 6-cm_ petri dishes, assayed for infective centers at 40°, and the plaques then were surveyed for the presence of deletion mutants by the electrophoretic assay described earlier (Brockman and Nathans, 1974). Endo R analysis of DNA. The electro- phoretically fractionated DNA species pre- pared by infecting cells with virus from the complementation plaques described above were recovered from agarose gels, dialyzed extensively against 1.5 mM NaCl in 0.15 mM Na citrate (pH 7), concentrated 10-fold by evaporation, and subjected to digestion with endonuclease R- Hind (Danna et al., 1973). Hind fragments were separated by electrophoresis in 4% polyacrylamide gel slabs and fragments visualized by autora- diography. Virus induction from transformed CHL cells by mitomycin was based on the proce- dure of Rothschild and Black (1970). Transformed CHL cells were seeded in 1.5- cm wells (3 x 10* cells per culture) and allowed to attach at 37°. Mitomycin-C (Sigma) was dissolved in MEM and added in a darkened room to a final concentra- tion of 2 yg/ml. The cultures were wrapped in aluminum foil and incubated at 37° for 12 hr. Then the medium was removed and the cultures were rinsed once 322 with MEM containing 2% fetal calf serum and infected at a multiplicity of 5 with temperature-sensitive helper virus tsA28 or tsB4, or mock-infected. At the end of 2 hr, medium was added and the cultures were incubated for 2 days at 37°, at which time 10° permissive BSC-40 cells which had been infected with the same helper virus were added to each culture. When the permissive cells had attached, the cul- tures were shifted to 40° and incubated for 16 days. After freezing and thawing, the lysates were surveyed for the presence of deletion mutants as described above for virus rescued from transformed 3T3 cells. In one set of experiments defective virus was rescued from transformed CHL cells by the fusion-complementation procedure described above. RESULTS Structure of mutant genomes. The gen- omic structures of mutants used in the studies to be reported are shown in Fig. 1 in relation to the SV40 map (Brockman et al., 1975; Lai and Nathans, 1974a). In sum, ev-1114 and dl-1007 have deletions span- ning the two known late genes of SV40 (B/ C and D), ev-1114 retaining only a small segment of late DNA (about 450 nucleotide pairs) adjacent to each end of the early region of the genome. Both of these mu- tants have the entire early region (A gene) intact. The two other mutants (ev-1117 and ev-1119) have deletions which elimi- nate about 70 and 60% of the early region, respectively, and have the entire late re- gion uninterrupted. All three evolutionary variants, which were derived from serially passaged virus stocks, have duplications of DNA including the origin of DNA replica- tion. In all experiments to be described below, doubly cloned mutant virions were first separated from ts helper virus by CsCl centrifugation or purified mutant DNA was prepared by agarose gel electro- phoresis, as described in Methods. Titer of purified mutant virions. To de- termine the number of biologically active deletion mutant particles in purified virus preparations and the amount of residual ts helper virus present, the number of plaque-forming units was determined at SCOTT, BROCKMAN AND NATHANS <_ Bra 6. ; \ Fic. 1. HindII/TIIl maps of mutant genomes in relation to the map of wt SV40 DNA (from Brock- man et al., 1975; Lai and Nathans, 1975, 1976). Du- plicated regions are indicated by the stippled areas, and deleted regions by the wedge-shaped extension from the circles. In the wt map are shown location of templates for early and late mRNA’s (Khoury et al., 1975), the three known genes of SV40 (Ozer and Tegtmeyer, 1971; Chou and Martin, 1974; Lai and Nathans, 1975), and the origin of DNA replica- tion(—). 40° (nonpermissive for ts mutants) by in- fectious center assay, in the presence of excess tsA or ¢tsB helper, as described in Methods. The titers, expressed as comple- mentation plaque-forming units or cPFU, are given in Table 1. (To determine the efficiency of complementation, titers of ts mutants measured by direct plaque assay at 32° were compared with result of com- plementation plaque assay; cPFU at 40° were about 30-70% of PFU at 32°.) As seen in Table 1, residual contamination by ts helper virus varied from about 2% (ev- 1119, preparation 1) to about 0.2% (dl- 1007). Formation of T antigen in infected cells. Mutant-infected cells were tested for the SV40 DELETION MUTANTS presence of T antigen, as described in Methods. To minimize the effect of con- taminating helper virus, mutant DNA purified by agarose gel electrophoresis was used. The cells were fixed at 28 hr postin- fection and stained for T antigen by the indirect immunofluorescence procedure. DNA from two late deletion mutants, ev- 1114 and di-1007, induced T antigen pro- duction in 1.2 and 4.4% of cells scored, respectively, compared to 1.7 % for cells infected by wt SV40 DNA. DNA from the two early mutants did not induce detecta- ble antigens (<0.1% of cells scored). These TABLE 1 Titers or SV40 Mutants Mutant Infectious centers (cPFU/ml) Mutant + tsA28 + tsB4 ts helper ev-1114 7x 10’ 5 x 105 140 ev-1117 2x 105 4x 10’ 200 ev-1119 #1 9 x 105 5 x 107 55 ev-1119 #2 1 x 106 1 x 10° 100 dl-1007 3 x 107 6 x 10+ 500 ev-Hll4 12 3 4 5 6 M7 8 9 10 tt fz : r +e de < a 7 *- 323 results are consistent with prior conclu- sions that the genetic determinant of T antigen is in the early region of the ge- nome (Lewis et al., (1974). Viral DNA synthesis. Next we tested the ability of variant genomes to replicate in the absence of helper virus. Since each variant virus preparation contained con- taminating helper virus (Table 1), we used a range of variant multiplicities and com- pared results with and without coinfecting SV40 in order to determine whether var- iant DNA replication was independent of helper virus. For this purpose two sets of monolayers of BSC-1 cells in 6-mm mi- crowells were infected in parallel with pur- ified deletion mutant virions at different multiplicities of infection, and one set of infected wells was coinfected with wt SV40 at a multiplicity of 4 PFU/cell. Viral DNA was labeled with **P, and variant and 5V40 DNA were isolated from the Hirt supernatant by electrophoresis in agarose, as detailed in Methods and illustrated in Fig. 2. The yield of each [*P]DNA species was then determined by counting dis- « we «= 2s * #» © * * 6@ #660 wT *@ « ev IL # #2 @ wt o. ev I Fic. 2. Replication of viral DNA in mutant-infected cells: autoradiograms of |**PIDNA extracted from infected cells followed by electrophoresis in agarose. The origin is at the top (arrow). On the left are the results with ev-1114; on the right are the results with ev-1117. Slot 1, no virus; slots 2-6, decreasing multiplicity of mutant; slot 7, wt SV40 only; slots 8-12, as in 2-6 plus wt SV40. M, reference SV40 DNA. For quantitation, the amount of ”P in short genomes (ev I and II) and in wi length genomes (wt I and II) was measured for each sample and normalized to the recovered [*HJDNA, as detailed in Methods. Results are plotted in Fig. 3. 324 207 ev-1l4 cts/m (thousands) o ae 0.005 = 0.05 0.8 $.0 50 epfu/ cell eb ev- III? Variant DNA 0.01 100, ot cptu/ cell Fic. 3. Replication of mutant DNA in the pres- ence (©) or absence (@) of wt SV40. On the ordinate is plotted normalized **P radioactivity in short ge- nomes and on the abscissa is input multiplicity of deletion mutant virus (cPFU/cell). When added, wt $SV40 was used at a multiplicity of 4. 0, Unit length DNA in absence of added wt SV40. solved gel segments after autoradi- ographic localization of DNA. As indicated in Methods, all values were normalized to recovered [9H}]SV40 DNA added to infected cells with the lysing solution. Quantitative results for this series of experiments with ev-1114 and ev-1117 are shown in Fig. 3. As seen in this figure, ev- 1114 DNA replication was detectable at a multiplicity of 0.05 cPFU/cell (about 0.0004 cPFU/cell of contaminating tsA28 helper virus) and was not stimulated by coinfection with SV40 over the entire range of variant multiplicities used. In contrast, ev-1117 DNA replication was first measurable at a multiplicity of 10 cPFU/cell, under which conditions the cells were coinfected with about 0.05 cePFU/cell of contaminating tsB4 helper vi- rus. Moreover, except at the highest mul- tiplicity of variant used, coinfection with SV40 resulted in stimulation of ev-1117 DNA replication. We conclude that ev-1114 DNA can replicate independently of helper virus, whereas ev-1117 DNA replication is dependent on helper. Less detailed experi- SCOTT, BROCKMAN AND NATHANS ments with ev-1119 and dl-1007 gave re- sults similar to those shown for ev-1117 and ev-1114, respectively. In this experiment we also scored cyto- pathic effect 72 hr postinfection at the var- ious multiplicities of infection. Cells in- fected with ev-1117 showed no visible cyto- pathic effect at input multiplicities of 10 cPFU/cell, whereas a cell layer infected with ev-1114 at 0.5 cPFU/cell showed ob- vious cytopathic effect, similar to that of cells infected with wt virus. Interference. In the course of the above studies of mutant DNA replication in the presence and absence of wt SV40, it be- came apparent that the mutants had in- hibited the replication of wt virus DNA (Fig. 2). To quantitate this inhibition we determined the yield of wt virus and wt virus DNA after coinfection of cells with a given mutant and wt SV40, as detailed in Methods. To avoid competitive absorption, mutant virus was added 1 hr after wt SV40. The results for ev-1114 and ev-1117 are presented in Table 2; ev-1119 gave es- sentially the same results as ev-1117. As seen in Table 2, both ev-1114 and ev-1117 inhibited the production of infectious SV40 and also inhibited the formation of SV40 DNA. Since ev-1114 lacks the major por- tion of the late region of the SV40 genome while ev-1117 is missing most of the early region, this interference phenomenon does not appear to be dependent on a specific gene product. One difference in the effect of the two variants is that in ev-1114-in- fected cells, total viral DNA synthesis is undiminished or somewhat greater than in cells infected with SV40, whereas in ev- 1117 cells, total viral DNA synthesis is below that seen in cells infected with SV40 alone. A second difference is seen in the ratios of variant to wt SV40 DNA (Table 2). These differences probably reflect the fact that replication of the ev-1114 genome does not depend on coinfecting wt SV40, whereas replication of the ev-1117 genome does (see Fig. 3). Stimulation of cellular DNA synthesis. One of the unusual activities of SV40 pos- sibly related to its oncogenic activity is the stimulation of cellular DNA synthesis in both permissive and nonpermissive cells SV40 DELETION MUTANTS TABLE 2 YIELD oF SV40 AND oF ViraL DNA 1n CELLS CoINFECTED WITH MUTANT AND WT VIRUS Mutant Multi- Yield as % of Total Variant plicity control viral DNA/wt (cPFU/ — DNA DNA cell) wt vi- wt (% of rus DNA con- trol) ev-1114 5 20 9.7 131 13 50 2 3.6 127 34 ev-1117 10 25 15 42 1.8 100 14 4.9 24 3.9 “ Each monolayer, consisting of 5 x 104 BSC-1 cells, was infected with 5 PFU of wt SV40 per cell, followed 1 hr later by the indicated mutant at the multiplicity shown. In the absence of mutant (con- trol), the yield of wt virus was 4.8 x 107 PFU at 96 hr postinfection and the radioactivity of unit length [?PJDNA isolated from cell lysate 72 hr postinfec- tion was 2 x 104 cpm. ® Total viral DNA equals unit length plus short genomes, i.e., wt + mutant, expressed as % of viral DNA from cells infected with wt SV40 alone. (Hatanaka and Dulbecco, 1966; Gershon et al., 1966). Since cells infected with tsA (early) mutants of SV40 show much greater temperature sensitivity of viral DNA replication than of stimulation of cell DNA synthesis (Tegtmeyer, 1972; Chou and Martin, 1975b), it was of interest to determine whether variants containing deletions in the early or late region of the SV40 genome could stimulate cell DNA synthesis. To test for this function, BALB/ c-3T3 cells were chosen because of the low background of DNA synthesis in contact- inhibited cells. Cell monolayers prelabeled with [“C}thymidine were infected with SV40 or with deletion mutant virions as detailed in Methods, and the incorporation of [H]thymidine into high molecular weight DNA was determined at various times after infection. The results are pre- sented in Fig. 4, each value plotted having been normalized to a constant amount of cellular DNA by means of the C counts in the same sample. As seen in the figure, ev- 1119 (and ev-1117 in a separate experi- ment), which is missing much of the early genome region, failed to stimulate thymi- dine incorporation into cellular DNA. In contrast, ev-1114 (and dl-1007 in a separate experiment), which has a deletion of most 325 of the late region, stimulates thymidine incorporation at least as well as SV40. We conclude that part or all of the early region is clearly necessary for this viral function. On the assumption that the late genome segment present in ev-1114 DNA does not code for some unknown functional protein, we can also conclude that late gene prod- ucts are not required. Whether ev-1114 is actually more active than wt SV40 in stim- ulating thymidine incorporation into cel- lular DNA, as suggested by the results shown in Fig. 4, is not clear, since in other experiments with different mutant prepa- rations, ev-1114- and dl-1007-stimulated incorporation was somewhat less than that due to wt SV40. Cell transformation. The ability of mu- tants to transform cells in culture was tested with nonpermissive BALB/c-3T3 cells and with semipermissive CHL. For this purpose subconfluent cells growing in multi-well dishes were infected at various multiplicities with wild-type or deletion mutant virions, and the number of mor- phologically transformed colonies was scored following transfer of the cells, as detailed in Methods. Only dense, multi- layered colonies were scored as trans- formed. The results of representative ex- periments are presented in Table 3. As 404 ev-1H4 304 wtSv40 204 H cts/min in DNA (thousonds) 3 ev-I1I9 (4) uninfected (®) ————¢-— & 24 48 72 Hrs. post Infection Fic. 4. Incorporation of [?H]thymidine into DNA of 3T3 cells after infection by mutant virus. On the ordinate is plotted the normalized incorporation into high molecular weight DNA vs time after infection. Twenty cPFU or PFU/cell were used in each case. See Methods for experimental details. The results for dl-1007 and ev-1117 were similar to those for ev- 1114 and ev-1119, respectively. 326 TABLE 3 TRANSFORMATION BY MUTANTS Virus PFU or cPFU/ Transformed colonies (% helper vi- cell per 10° cells? rus contami- nation)* BALB 3T3 CHL SV40 30 PFU 30 - 6 15 9 3 8 9 1.2 5 - 0.6 - 2 0.06 0 0 dl-1007 5 cPFU 12 _ (0.2%) 1 1 5 0.1 0 2 ev-1114 15 cPFU 12 ~ (1.0%) 3 1 7 0.3 0 2 ev-1117 30 cPFU 0 — (0.5%) 6 0 0 3 - 0 ev-1119 30 cPFU 0 0 (1.0%) 15 0 0 “The mutant viruses were assayed about 1 month prior to use. * Average of two original microwell monolayers. seen in the table, ev-1114 and dl-1007 pro- duced morphologically transformed colo- nies from both 3T3 and from CHL cells, whereas ev-1117 and ev-1119 had no trans- forming activity in either cell type at com- parable multiplicities of infection. Since similar amounts of contaminating ts helper virus were present in the serially diluted variants, it is unlikely that trans- formation by ev-1114 and dl-1007 was due to contaminating ¢sA mutant still present in the preparation. Moreover, SV40 at multiplicities similar to that of contami- nating ts helper virus present in the active. virus stocks failed to transform. When transformed colonies were selected from SV40-, ev-1114-, or dl-1007-infected 3T3 or CHL dishes and recloned twice, the result- ing clones all contained nuclear SV40 T antigen detectable by fluorescent antibody staining. We conclude that ev-1114 and di- 1007, but not ev-1117 nor ev-1119, can transform mouse and hamster cells, and therefore it is likely that the early region of the SV40 genome is not only necessary but also sufficient for transformation. Virus production by transformed cells. SCOTT, BROCKMAN AND NATHANS Since deletion mutant preparations used to transform CHL and 3T3 cells contained contaminating helper virus (tsA28), ex- periments were carried out to determine whether any of the cloned transformed cell lines contained rescuable virus. In the case of CHL cells, which are semipermissive for SV40 (Martin and Chou, 1975), attempts were made to induce virus production with mitomycin-C at 32°, as described in Meth- ods. In the case of transformed 3T3 clones, virus rescue at 32° by fusion with BSC-1 cells was attempted. For each cell type, two independent clones of wt SV40-trans- formed cells from the experiment shown in Table 3 were used as controls. As shown in Table 4, both wt SV40- transformed CHL cell lines (SV-CHL-1 and 2) yielded small amounts of virus in the absence of mitomycin-C. After drug treatment the yield increased more than 100-fold. However, none of the mutant- transformed CHL cells yielded detectable infectious virus at 32° with or without mi- tomycin-C treatment. In the case of 3T3 cells both wt SV40-transformed cell lines (SV-3T3-11 and 12) and one of the four ev- 1114-3T3 lines yielded virus after fusion with BSC-1 cells and prolonged incuba- tion, as judged either by cytopathic effect of the cell lysate on a fresh monolayer of BSC-1 cells (Table 4) or by plaque assay (results now shown), whereas the other mutant-transformed lines produced no de- tectable virus. Presumably, ev-1114-3T3-2 had been transformed by the tsA helper virus or a derivative thereof. However, the other mutant-transformed CHL or 3T3 cells showed no evidence of an intact SV40 genome by this test. Rescue of deletion mutants from 3T3- transformed cells by complementation. The rescue and induction experiments just described did not establish that mutant- transformed cells contain the genome of the SV40 mutants used to transform them. In an attempt to determine whether mu- tant genomes were actually present we ap- plied the complementation plaquing proce- dure described by Brockman and Nathans (1974) to the rescue of mutant genomes, as detailed in Methods and diagrammed in Fig. 5. In this procedure, cells infected SV40 DELETION MUTANTS 327 TABLE 4 Virus RESCUE FROM TRANSFORMED CLONES CHL? 3T3° Cells Virus production (PFU) Cells Virus produc- tion (cpe) Not induced Mitomycin induced CHL 0 0 3T3 - ev-1114-CHL-1 0 0 ev-1114-3T3-2 + ev-1114-CHL-2 0 0 ev-1114-3T3-3 - dl-1007-CHL-1 0 0 ev-1114-3T3-4 _ dl-1007-CHL-2 0 0 ev-1114-3T3-5 - SV-CHL-1 100 1.3 x 10% dl-1007-3T3-2 = SV-CHL-2 370 1.0 x 10° SV-3T3-11 + SV-3T3-12 + 7 A 1.5-em monolayer containing about 10° cells was treated with mitomycin-C (where indicated), incubated for 6 days at 32°, frozen and thawed, and PFU of lysate was measured on BSC-1 cells at 32°. ®’ Three x 10* BALB 3T3 cells were fused with 3 x 10° BSC-1 cells using uv-inactivated Sendai virus, and the mixture was plated in a 10-cm petri dish. Following incubation at 32° for 22 days, the entire undiluted frozen and thawed cell lysate was then used to infect BSC-1 cells at 32°, and cytopathic effect was scored after 10 days. Rescue of Defective SV40 by Fusion- Complementation \) Transformed 2)BSC-40 cells cells Microwell 5) Transfer to petri dish 3) Infect with tsA | 6) Add BSC-40cells 4) Fuse cells : Complementation plaques Fic. 5. Diagram of defective virus rescue by fu- sion-complementation. with two complementing mutants are plated, and excess indicator cells are then added to allow plaque formation. Plaque formation should be dependent on comple- mentation or recombination between a rescued genome and the added helper ts genome in a single heterokaryon; revert- ants in the ts virus stock would also pro- duce plaques. In this series of experiments we concentrated on analyzing clones ev- 1114-3T3-5 and dl-1007-3T3-2. The results of complementation plaqu- TABLE 5 Virus RESCUE FROM TRANSFORMED CELLS AT 40°¢ Cells Infectious centers after fusion No + tsA28 + helper tsC219 3T3 0, 0 2,1,1 0,0 ev-1114-3T3-5 0, 0 11, 13, 14 0,0 dl-1007-3T3-2 0, 0 4,0, 1 0, 0 SV-3T3-11 3,1 2, 2,4 1,0 “ Rach value is the yield from 10‘ transformed or nontransformed 3T3 cells. Multiple values refer to replicate dishes. ing are shown in Table 5. As controls, untransformed 3T3 and wt SV40-3T3-11- transformed clones were included, and plaque formation was also assessed with a late mutant (tsC219), instead of tsA28, as helper virus. As seen in the table, tsA28 with fused 3T3-BSC-40 cells resulted in a small number of plaques. These are proba- bly revertants in the tsA stock (see below), since 10° PFU were used in each experi- ment. In the case of SV-3T3 clones, fusion alone resulted in plaque formation with no apparent increase in plaques due to super- infection by ts mutants. In contrast, both ev-1114- and dl-1007-transformed clones produce plaques only in the presence of tsA helper virus; ésC could not substitute for tsA. These results suggest that comple- mentation and fusion resulted in rescue of 328 a resident mutant genome. As shown be- low, this inference was substantiated by analysis of individual plaques. Analysis of complementation plaques. To determine whether a putative comple- mentation plaque contained the input de- fective genome in addition to helper virus, plaques were screened by infecting BSC-40 cells grown in 1.5-cm wells with a few drops of plaque suspension at 40°. This temperature favors the replication of ge- nomes containing a normal early region over the tsA helper, since the multiplicity of infection is low. **P viral DNA was pre- pared from the infected cells and subjected to electrophoresis in agarose gels to sepa- rate short genomes from unit length helper virus DNA as described in Meth- ods. To determine whether short genomes were derived from the mutant used for Electrophoresis of Plaque No. i2345678a deb ge — Genome Length 100%— . 86%-— 65%-—— & a SCOTT, BROCKMAN AND NATHANS transformation or from the ¢sA helper, re- covered DNA was analyzed by digestion with endo R- HindII/dIIl and electropho- resis of digest products. In some cases plaque suspensions were first used to pre- pare virus stocks in order to obtain larger quantities of DNA. Plaques from tsA-infected untrans- formed 3T3/BSC_ heterokaryons. Three plaques formed by the tsA-infected un- transformed 3T3-BSC cells (see Table 5) all yielded full-length SV40 DNA; the HindII/ dlII digest of one of these was identical toa digest of tsA DNA (Lai and Nathans, 1974b). These plaques therefore appear to be derived from revertants of tsA. Plaques from tsA-infected ev-1114-3T3- 5/BSC heterokaryons. Eight plaques formed by tsA-infected ev-1114-3T3-5/BSC cells were analyzed (Fig. 6 and Table 6). Hin Digests = Plaque No. = = 314567 3 = eS wo * @ A in A: — Led “ “ - —Ar B-_ “ “ 3 —_B nn fat -—I b Fic. 6. (a) Agarose gel electrophoresis of viral DNA rescued from ev-1114-3T3-5. The numbers at the top refer to individual plaques, as described in the text and Table 6. The genome lengths recorded at the left are based on electrophoretic mobilities (Brockman et al., 1975). I, II refer to the positions of wt DNA forms I and IE, respectively. (b) Representative Hin digest patterns of dominant genomes shown in Fig. 6a: autoradi- ogram of [*PJDNA fragments following electrophoresis. Numbers at the top refer to individual plaques noted in Fig. 6a and Table 6. The three lanes on the right are from a different electrophoretic run than those on the left. Fragment A, is the CG fusion fragment (Fig. 1). SV40 DELETION MUTANTS TABLE 6 VIRAL GENOMES RESCUED FROM ev-1114-3T3-5 By FuSION-COMPLE MENTATION Plaque num- ber: 1 65, 78, 86, 90, 100 Inferred structure of predominant genome Lengths of genomes* (% of SV40 DNA) ev-1114 minus du- plication 2 100, 88 Variant of ev-1114 or recombinant 3 69, 90, 100 Variant of ts helper 4 86 ev-1114 5 86 ev-1114 6 78, 65, 86, 92, 100 Variant of ev-1114 7 65, 76, 86, 94 ev-1114 minus du- plication 8 82, 65, 76, 92, 100 Variant of ev-1114 * The predominant genome length is underlined. Six of these yielded more than one size class of DNA, but each had a predominant species. Only one plaque (plaque no. 2 in Table 6) produced unit length DNA as the major species; Hin digestion revealed a complex fragment pattern suggesting deri- vation from ev-1114 DNA. A shorter ge- nome (88% of full length) formed from the same plaque was a variant of tsA, as judged by its fragment pattern. Two plaques (Nos. 4 and 5 in Fig. 6a and Table 6) produced predominantly a DNA species of 86% of unit length, each of which gave an Hin digest pattern identical to that of ev-1114 DNA (Fig. 6b). Two plaques (Nos. 1 and 7) produced predominantly a DNA species of 65% of unit length, each of which gave a Hin digest pattern qualitatively identical to the ev-1114 DNA pattern, but with only one molar equivalent corre- sponding to the Hin-A fragment (Fig. 6b), indicating loss of the duplication present in ev-1114 (see Fig. 1). Of the remaining three plaques analyzed (nos. 3, 6, and 8), one produced a predominant DNA species 69% of unit length which was derived from tsA DNA. The other two (nos. 6 and 8) produced predominantly genomes of 78 and 82% of unit length, respectively, each of which showed Hin digest patterns re- sembling the 65% DNA present in plaques 1 and 7, except that each digest had one additional new fragment (Fig. 6b). It is of interest that each of the last two plaques 329 also produced a smaller amount of a DNA species 65% of full length (Fig. 6a). From the foregoing results we conclude that clone ev-1114-3T3-5 harbors the ev- 1114 genome, and after fusion-complemen- tation the original mutant genome or a derivative thereof replicates and becomes encapsidated (see Discussion). Plaques from tsA-infected dl-1007-3T3- 2/BSC heterokaryons. Five plaques formed by tsA-infected dl-1007-3T3-2/BSC cells were analyzed (Fig. 7 and Table 7). Four plaques (Nos. 2 to 5 in Table 7) yielded DNA which included molecules 76% of SV40 DNA length, i.e., the length of dl-1007 DNA (Fig. 7a). However, this DNA species was the major one in only one case (no. 2); Hin digestion of the 76% DNA from plaque no. 2 yielded an electropho- retic fragment pattern identical to di-1007 DNA (Fig. 7b). The unit length DNA spe- cies appeared to be derived from the tsA helper virus, as judged by their length or Hin digestion (Fig. 7b). On the basis of the results with rescued plaque no. 2, we con- clude that clone dl-1007-3T3-2 harbors the dl-1007 genome. Rescue of deletion mutants from trans- formed CHL cells. Similar analyses of vi- rus rescued from three transformed CHL clones were carried out, in this case in the presence of mitomycin-C and without clon- ing, as detailed in Methods, or by comple- mentation-fusion as with 3T3 cells. After superinfection with the tsA28 mutant (but not with ¢sB4), dl-1007 virus was rescued from dl-1007-transformed CHL cells (1007- CHL-1 and -2). Identification of the res- cued mutant was carried out by electro- phoretic isolation of its genome followed by analysis of Hin digests, as described above for virus rescued from transformed 3T3 cells. Similarly, from ev-1114-transformed CHL cells (ev-1114-CHL-2) superinfected with tsA28 and fused with BSC-40 cells, mutant genomes 65% of the length of SV40 DNA were detected. This species yielded the Hin digest pattern of ev-1114 DNA mi- nus its duplicated segment. Thus, mutant- transformed clones of CHL cells as well as mutant-transformed clones of 3T3 cells produced the original transforming virus or a derivative thereof. 330 Electrophoresis of DNA NS PlaqueNo. 8 [2345 3% — I — I e — dl-1007 SCOTT, BROCKMAN AND NATHANS Hin Digests Plaque No.2 dl-lIOO7 76% 100% ay B—w «= « —B _ L —cC j <= —D —E —F G—*# —G H— —H | — — | —J Fic. 7. (a) Agarose gel electrophoresis of viral DNA rescued from di-1007-3T3. Numbers at the top refer to individual plaques, as described in the text and Table 7. I, II refer to the positions of wt DNA forms I and II, respectively. (b) Hin digest patterns of genomes shown in Fig. 7, plaque no. 2: autoradiogram of [P]DNA fragments following electrophoresis. Notations at the top refer to the 76% genome and the unit length genome from plaque no. 2. DISCUSSION Three genes have been identified and mapped in the SV40 genome (for a recent review, see Kelly and Nathans, 1976): the A gene which codes for a protein (A pro- tein) of about 100,000 daltons that contains determinants for SV40 T antigen (Lewis et al., 1974; Tegtmeyer, 1974; Rundell et al., 1976); the B/C gene which codes for the major virion protein (VPI) (Prives et al., 1974; May et al., 1975; Khoury et al., 1976; Lai and Nathans, 1976); and the D gene which may code for a minor virion protein (Chou and Martin, 1975b). Based on the properties of tsA mutants, which map in the early region of the genome (Lai and Nathans, 1975), it has been inferred that the A protein is required for viral DNA replication (Tegtmeyer, 1972), stimulation of cellular DNA synthesis (Chou and Mar- SV40 DELETION MUTANTS TABLE 7 VirAL GENOMES RESCUED FROM dl-1007-3T3-2 BY Fusion-COMPLEMENTATION Plaque Length of ge- ‘Inferred structure number nomes* (% of SV40_ of predominant DNA) genome 1 100 ts helper 2 76, 100 dl-1007 3 100, 76 ts helper 4 100, 76 ts helper 5 100, 76 ts helper « The predominant genome length is underlined. tin, 1975), and cell transformation (Brugge and Butel, 1975; Martin and Chou, 1975; Osborn and Weber, 1975; Tegtmeyer, 1975; Kimura and Itagaki, 1975). The experi- ments reported here with mutants con- taining deletions within the early region confirm these earlier findings with tsA mutants. Of particular note is the com- plete lack of stimulation of cellular DNA synthesis (assessed by thymidine incorpo- ration into high molecular weight DNA) after infection with ev-1117 or ev-1119, in contrast to the “leakiness” for this prop- erty of tsA mutants (Tegtmeyer, 1972; Chou and Martin, 1975). As shown in Fig. 1, both early deletion mutants tested are missing extensive segments of the early region of the genome, including that por- tion where almost all available tsA mu- tants map. Therefore the possibility of as- signing separate functions to specific seg- ments of the early region remains open. For example, it is possible that the A pro- tein is a primary gene product that is sub- sequently cleaved into polypeptides with distinct functions or that the A protein itself has multiple, distinct activities. In contrast to the early deletion mu- tants, the two mutants (d/-1007 and ev- 1114) with intact A gene but extensive deletions of the late genomic segment, in- cluding deletions of portions of both the B/ C and D genes (Fig. 1), can replicate their DNA without helper virus, stimulate thy- midine incorporation into high molecular weight DNA of nonpermissive mouse cells, and transform mouse and hamster cells with about the same efficiency as wild- type virus. Furthermore, we have shown by fusion-complementation that the mu- 331 tant-transformed cells harbor the mutant genome. Since the ev-1114 genome retains only a small segment of DNA outside the early genomic region, we conclude that the early region plus immediately contiguous DNA segments are sufficient for all of the above viral functions. Again, these results agree with earlier studies of tsB/C and tsD mutants (Tegtmeyer and Ozer, 1971; Mar- tin and Chou, 1975), and with the finding that a linear fragment of DNA derived from the genome segment between 0.15 and 0.74 map units (and thus the entire early region) can transform rat cells (Abrahams et al., 1975). Extension of these several studies by use of other deletion mutants or DNA fragments should estab- lish whether any DNA segments outside the early region are needed (e.g., tran- scription or replication signals), and also what parts of the early region are indis- pensable. Recent studies by Shenk e¢ al. (1976) with viable deletion mutants al- ready indicate that the genomic segment between 0.54 and 0.59 map units is not necessary for cell transformation. The finding that the early genome seg- ment of SV40 DNA is sufficient for viral DNA replication indicates that this seg- ment of DNA will be useful as a vector for constructing self-replicating plasmids con- taining inserted DNA segments. If the re- combinant is of appropriate size, it can be encapsidated into an SV40 virion in the presence of a helper mutant and cloned as described for constructed deletion mutants (Lai and Nathans, 1974a). Uchida et al. (1966) first noted that de- fective particles of SV40 interfere with the replication of wt virus. The coinfection of cells with wt virus (under conditions which minimize competitive absorption) and either ev-1114, ev-1117, or ev-1119, re- sulted in inhibition of wt DNA replication as well as inhibition of virus production. Since both the early and late deletion mu- tants produce this inhibition, interference apparently does not depend on the pres- ence of a specific gene product. On the other hand, since all three mutants con- tain a duplication of the replication initia- tion site, interference may be related to the number of initiation signals present; 332 l.e., interference may be due to competi- tion for a limiting initiation protein which interacts at this site. According to this notion, molecules with multiple initiation sites should initiate replication more fre- quently than molecules with a single site. Consistent with this expectation is the ob- servation that in cells coinfected at about equal multiplicities with wt SV40 and ev- 1114 (which is not dependent on helper virus for DNA replication), variant DNA accumulates (Table 2). Presumably this is the basis for the evolution of these var- iants (Brockman ef al., 1973). It will be of interest to determine in a more systematic way both rates of replication and interfer- ing activity of SV40 variants as a function of the number of initiation sites per mole- cule (Lee et al., 1975). The observation that a single clone of 3T3 cells transformed by ev-1114, an evolu- tionary variant with a tandem duplication of DNA, yielded a variety of genomes re- lated to ev-1114 after fusion-complementa- tion is of interest. Among these genomes were molecules lacking the tandem dupli- cation and others with what appear to be new duplications of DNA. In all of the latter instances, molecules the size of those lacking a duplicated segment were also present in the same virus plaque. One possible explanation for these observations is that excision of integrated ev-1114 DNA can occur in two ways (see Fig. 8): (1) by reversal of the integration step, i.e., re- combination at the site of integration, giv- ing rise to the original variant genome; or (2) by recombination within the duplicated segment, giving rise to a genome without the original duplication, as in the case of tandemly duplicated SV40 DNA in certain adeno-SV40 hybrid viruses (Kelly et al., 1974). In the second mechanism, genomes 65% of the length of SV40 DNA are gener- ated; since this length is probably below the level needed for efficient encapsida- tion, secondary duplications may be se- lected during subsequent rounds of virus replication. In fact, further propagation of the plaque suspension containing virus with predominant genomes 65% of unit length leads to the appearance of longer viral DNA molecules and disappearance of the 65% genome. On the other hand, rou- SCOTT, BROCKMAN AND NATHANS CELL ¢” cra HI B DNA ~ I B Fic. 8. Two possible modes of excision of inte- grated ev-1114 DNA. See the text for discussion of the model. Duplicated regions are indicated by the stippled segments. tine propagation of ev-1114 in BSC-1 cells does not generate the 65% variant, sug- gesting that this genome was not derived in the fusion-complementation experi- ments from free ev-1114 DNA. The in- ferred mechanism of excision of integrated viral genomes containing duplications de- serves more detailed and quantitative study. Finally, we point out that the procedure used to rescue defective SV40 genomes from transformed cells is similar in princi- ple to that described by Yoshiike et al. (1974), which was based on triple fusion of two transformed cells with a permissive cell. In the procedure described in the pres- ent communication, complementing func- tions are supplied by a superinfecting tsA mutant rather than a wt SV40 viral ge- nome present in transformed cells. There- fore virus growth at elevated temperature is dependent on rescuesof the endogenous genome, and infectious centers can be plated directly to isolate clones of rescued virus. This rescue procedure may have other applications, e.g., rescue of viruses from cells suspected of harboring specific defective genomes (e.g., from tumors or certain degenerative lesions), or rescue of misexcised viral DNA from transformed cells. ACKNOWLEDGMENTS We thank C.-J. Lai and Gary Ketner for helpful advice, and Douglas Welsh for able technical assist- SV40 DELETION MUTANTS ance. This research has been supported by a Na- tional Cancer Institute Grant (1 PO1 CA16519) and a grant from the Whitehall Foundation. W. W. B. and W. A. S. were supported by fellowships from the National Cancer Institute. REFERENCES ABRAHAMS, P. J., MULDER, C., VAN DE Voorpe, A., WaARNAAR, S. O., and VAN DER Eps, A. J. (1975). Transformation of primary rat kidney cells by fragments of simian virus 40 DNA. J. Virol. 16, 818-823. Brockman, W. W., and Natuans, D. (1974). The isolation of simian virus 40 variants with specifi- cally altered genomes. Proc. Nat. Acad. Sci. USA 71, 942-946. BrockMan, W. W., and Scott, W. A. (1975). Biolog- ical activities of cloned evolutionary variants of simian virus 40. Fed. Proc. 34, 526 (Abstract). BrockMaN, W. W., Gutal, M. W., and Natuans, D. (1975). Evolutionary variants of simian virus 40: Characterization of cloned complementing var- iants. Virology 66, 36-52. BruacE, J. S., and Bure, J. S. (1975). Role of simian virus 40 gene A function in maintenance of transformation. J. Virol. 15, 619-635. Carpon, J., SHENK, T. E., and Bere, P. (1975). Biochemical procedure for production of small deletions in simian virus 40 DNA. Proc. Nat. Acad. Sci. USA 72, 1392-1396. Cuou, J. Y., and Martin, R. G. (1974). Complemen- tation analysis of simian virus 40 mutants. J. Virol. 18, 1101-1109. Cuou, J. Y., and Martin, R. G. (1975a). Products of complementation between temperature sensitive mutants of simian virus 40. J. Virol. 15, 127-136. Cuou, J. Y., and Martin, R. G. (1975b). DNA infec- tivity and the induction of host DNA synthesis with temperature-sensitive mutants of simian vi- rus 40. J. Virol. 15, 145-150. Danna, K. J., and Natuans, D. (1971). Specific cleavage of simian virus 40 DNA by restriction endonuclease of Hemophilus influenzae. Proc. Nat. Acad. Sci. USA 68, 2913-2917. Danna, K. J., Sacx, G. H., Jr., and Nathans, D. (1973). Studies of SV40 DNA. VII. A cleavage map of the SV40 genome. J. Mol. Biol. 78, 363-376. Gersuon, D., Sacus, L., and Winocour, E. (1966). The induction of cellular DNA synthesis by sim- ian virus 40 in contact inhibited and X-irradiated cells. Proc. Nat. Acad. Sci. USA 56, 918-925. Hatanaka, M., and DutBecco, R. (1966). Induction of DNA synthesis by SV40. Proc. Nat. Acad. Sci. USA 56, 736-740. Hirt, B. (1967). Selective extraction of polyoma DNA from infected mouse cell cultures. J. Mol. Biol. 36, 365-369. Keuty, T. J., Jr., and Natuans, D. (1976). The genome of simian virus 40. Advances in Virus 333 Research, Vol. 21. Academic Press, New York, in press. KE ty, T. J., Jr., Lewis, A. M., Jr., LEvINgE, A. S., and SIEGEL, S. (1974). Structure of two Ad2-SV40 hybrids which contain the entire SV40 genome. JJ. Mol. Biol. 89, 113-126. Kuoury, G., Carter, B. J., GERDINAND, F.-J., How ey, P. M., Brown, M., and Martin, M. A. (1976). Genome localization of simian virus 40 RNA species. J. Virol. 17, 832-840. Kimura, G., and Iracakt, A. (1975). Initiation and maintenance of cell transformation by simian vi- rus 40: A viral genetic property. Proc. Nat. Acad. Sci. USA 72, 673-677. Lat, C.-J., and NatHans, D. (1974a). Deletion mu- tants of simian virus 40 generated by enzymatic excision of DNA segments from the viral genome. J. Mol. Biol. 89, 179-193. Lal, C.-J., and Natruans, D. (1974b). Mapping the genes of simian virus 40. Cold Spring Harbor Symp Quant. Biol. 39, 53-60. Lat, C.-J., and NatHans, D. (1975). A map of tem- perature-sensitive mutants of simian virus 40. Vi- rology 66, 70-81. Lar, C.-J., and Natuans, D. (1976). B/C gene of simian virus 40. Virology, in press. Lrg, T. N. H., Brockman, W. W., and Natuans, D. (1975). Evolutionary variants of simian virus 40: Cloned substituted variants containing multiple initiation sites for DNA replication. Virology 66, 53-69. Lewis, A. M., Jz., BREEDEN, J. H., WEwerkKA, Y. L., ScHNrPPER, L. E., and Levine, A. S. (1974). Studies of hamster cells transformed by adenovi- rus 2 and the nondefective Ad2-SV40 hybrids. Cold Spring Harbor Symp. Quant. Biol. 39, 651- 656. Martin, R. G., and Cuou, J. Y. (1975). Simian virus 40 functions required for the establishment and maintenance of malignant transformation. J. Vi- rol. 15, 599-612. May, E., Kopecka, H., and May, P. (1975). Mapping the transcription site of the SV40-specific late 16S mRNA. Nucleic Acids Res. 2, 1995-2005. McCurtcuan, J. H., and Pacano, J. S. (1968). En- hancement of the infectivity of SV40 DNA with DEAE-dextran. J. Nat. Cancer Inst. 4, 351-357. Mertz, J. E., Carson, J., HERZBERG, M., Davis, R. W., and Bera, P. (1974). Isolation and characteri- zation of individual clones of simian virus 40 mu- tants containing deletions, duplications and in- sertions in their DNA. Cold Spring Harbor Symp. Quant. Biol. 39, 69-84. Osporn, M., and WEBER, K. (1975). Simian virus 40 gene A function and maintenance of transforma- tion. J. Virol. 15, 636-644. Prives, C. L., Aviv, H., Gitpoa, E., Reve, M., and Winocour, BE. (1974). The cell-free translation of SV40 messenger RNA. Cold Spring Harbor Symp. 334 Quant. Biol. 39, 309-316. Rotuscuitp, H., and Buack, P. (1970). Analysis of SV40-induced transformation of hamster kidney tissue in vitro. Virology 42, 251-256. RuNDELL, K., Co.iins, J. K., TEGTMEYER, P., Lat, C.-J., and Naruans, D. (1976). Identification of simian virus 40 protein A. J. Virol., in press. Suenk, T. E., Carson, J., and Bera, P. (1976)..Con- struction and analysis of viable deletion mutants of simian virus 40. J. Virol. 18, 664-671. TectTmMEYER, P. (1972). SV40 DNA synthesis. The viral replicon. J. Virol. 10, 591-598. TeGTMEYER, P. (1974). Altered patterns of protein synthesis in infection by SV40 mutants. Cold SCOTT, BROCKMAN AND NATHANS Spring Harbor Symp. Quant. Biol. 39, 9-15. TEGTMEYER, P. (1975). Function of simian virus 40 gene A in transforming infection. J. Virol. 15, 613-618. TEGTMEYER, P., and Ozer, H. J. (1971). Tempera- ture sensitive mutants of simian virus 40: Infec- tion of permissive cells. J. Virol. 8, 516-524. Ucuipa, S., WATANABE, S., and Kato, M. (1966). Incomplete growth of simian virus 40 in African green monkey kidney cultures induced by serial undiluted passages. Virology 27, 135-141. Yosuke, K., Furuno, A., and Ucuipa, 8. (1974). Rescue of defective SV40 from a transformed mouse 3T3 cell line. Virology 60, 342-352.