- Email: [email protected]
SV40 T-antigen binding to site II is functionally separated from binding to site I
VIROLOGY
164,
309-3
17
(1988)
SV40 T-Antigen
Binding to Site II Is Functionally
GijNTER Department
STElTER,
of Biochem\stry,
DOROTHEE University
Received
MijLLER.
of U/m, P. 0. Box 4066,
October
15, 1987; accepted
Separated AND
MATHIAS
from Binding
to Site I
MONTENARY’
O-7900
U/m, federal
January
14. 1988
Republic
of Germany
During lytic infection SV40 T antigen binds specifically to three different regions of the SV40 DNA to initiate viral DNA replication and to regulate early and late transcription. We have used the recently described plasmids pKB1, containing a 23-bp oligonucleotide coding for site I, pdl1085 containing sites II and Ill together with SV40 specific flanking sequences, and as a control pATC, a plasmid which contains all three binding sites (D. Mijller et al. (1987), Virology 161, 81-91) to analyze the differential binding of T antigen to these individual binding sites in the course of an SV40 infection. We found that shortly after infection the amount of bound DNA increased with the concentration of T antigen reaching a steady-state level at about 20 hr after infection. In comparison to binding at site I, binding to site II appeared with a delay of about 8-9 hr corresponding to the onset of viral DNA replication. The correlation between binding of T antigen to site II and the SV40 DNA replication could be further corroborated by using T antigen from the heat-sensitive mutant tsA58 which completely failed to bind to site II at nonpermissive temperature but exhibited a residual binding to site I. This reduced binding to site I proved insufficient for the proper functioning of autorepression. Our results support the hypothesis that distinctly different subclasses of T-antigen binding to site I or site II may exist. C 1989 Academic
Press,
Inc.
pairs (bp) that is both necessary and sufficient for Tantigen recognitron (Ryder et a/., 1985). To initiate viral DNA replication, direct interaction of T antigen with another distinct binding site within the SV40 control region is necessary (Deb et a/., 1987). This sequence, binding site I!, is a part of the core origin region reaching from nucleotides 5209 to 34. After binding DNA T antigen covers a ca. 50-bp-!ong sequence of the SV40 DNA. This binding site consists of a set of four pentanucleotides in palindromic order and a stretch of continuous A/T pairs which also seems to be crucial for binding (Mijller et a/., 1987). In addition to the origin binding activity, in viva and in vitro studies suggest that requirement for triggering viral DNA replication involve at least an ATPase and helicase activity of T antigen (Manos and Gluzman, 1984; Dean et al., 1987; Stahl er al., 1986). Moreover, Smale and Tjian (1986) found that T antigen coprecipitates with cellular DNA polymerase CY,an enzyme involved In DNA synthesis. We have recently described the c!oning of a 23.bp oligonucleotide representing binding site I alone and several plasmids containing the 23 bp of binding sate II either with or without netghboring SV40-specific flanking sequences (MOller et a/., 1987). We were ab!e to show that T antigen binds with high affinity to site I on its own and with a very much reduced affinity to site II but only in the presence of the AT-rich sequence between sites II and III. In the present paper we used two of these plasmids, pKB1 and pdllO85, representing either isolated site I or sites II and III with SV40-specific
INTRODUCTION The simian virus 40 (SV40) infectious cycle in cultured monkey cells is primarily regulated by one of the early gene products, large T (tumor) antigen (Tooze, 1981; Rigby and Lane, 1983). This multifunctional protein is required for initiation (Shortle et al., 1979; Margolskee and Nathans, 1984) and probably elongation (Wiekowskr et a/., 1987) of viral DNA replication, and for stimulation of late and repression of early viral gene expression (Keller and Alwine, 1985; Brady and Khoury, 1985; Myers et a/., 1981; Rio and Tjian, 1983). Several distinct biochemical activities are carried out by T antigen: specific binding to multiple sites at the SV40 origin (Shalloway et a/., 1980; Tegtmeyer et a/., l983), ATP hydrolysis (Clark et al., 1984; Giacherio and Hager 1979; Tjian and Robbins, 1979), unwinding of short double-helical DNA strands (Stahl eta/., 1986; Dean et a/., 1987), and association with a 53-kDa cellular protein (Lane and Crawford, 1979). Late in infection the expression of the early gene products, that is, small and large T antigen, is downregulated (Tegtmeyer et a/., 1975) by an interaction of T antigen with at least one sequence in the SV40 control region, binding site I (Rio et al., 1980; Hansen et al., 1981; Myers et a/., 1981). This site contains three pentanucleotides (-GAGGC-), two of which are separated by a cluster of six ArT pairs. ln vitro binding studies with site I defined a central core of 17 base ’ To whom
requests
for reprints
should
be addressed. 309
0042.6822/88 Copyright
$3.00
Q 1968 by Academic
Press. Inc
STE-iTER, MOLLER, AND MONTENARH
310
I SV 40 ori - region
II 4 origin
early
mRNA
9-1
I
II II
-
III
plate
mRNA II -
of replication
insert Hind site Hind
Ill C of SV40
---,
-
I of SV40
____ _-IL-
III/Sph
--md”“‘,.
l of cs 1085
I
_________
Ill
___________._
// -r---________ __________
-__-____.
___.______
_ ________-----
_-_
______-____vectorsequences -
SV40
sequences
with
binding
site
1. Plasmids used for binding studies. pATC contains the HindIll C fragment of SV40, pKB1 contains a synthetic oligonucleotide for site I, pdllO85 contains the origin fragment of ~~1085, a deletion mutant of SV40 lacking binding site I. FIG.
flanking sequences in quantitative in vitro binding assays to analyze a possible correlation between sitespecific binding of T antigen and its functions in the course of a SV40 virus infection. As a control we used T antigen from the temperature-sensitive SV40 virus mutant tsA58 deficient in the control of autoregulation and initiation of DNA synthesis, We found that shortly after SV40 infection of permissive monkey cells the increase in the T-antigen level correlated with an increase in binding to SV40 ori DNA. Although T-antigen concentrations increased further later on after infection, DNA binding reached a steady-state level. In the course of the infection T antigen first bound to site I and later on after infection to sites II and Ill also. Binding to sites II and III was detectable shortly before the onset of viral DNA synthesis. Furthermore, T antigen from the temperature-sensitive mutant tsA58 which is heat-sensitive for DNA replication and autoregulation is strictly heat-sensitive for binding to sites II and III but exhibited a residual binding to site I. MATERIALS
AND METHODS
Cells and viruses TC-7, an established line of African green monkey kidney cells, was used for all infections with SV40 wt and tsA mutants. The cells were grown in Dulbecco’s modified Eagle’s medium containing 5% fetal calf serum. Stocks of SV40 wt strain 776 and of tsA68 were prepared and tritrated as previously described (SchUrmann et a/., 1985). Template
DNAs
For specific DNA binding assays three different plasmid DNAs were used (Fig. 1): pATC represents the vector pATl53 with a HindIll C fragment of SV40 (nu-
coding
cleotides 1046-5172) which contains all three binding sites; pKB1 contains a synthetic oligonucleotide representing SV40 site I (nucleotides 5186-5209) cloned in the HindIIIIBamHI site of pBR322; pdl1085 contains the HindIll (5172) to Sphl (133) fragment of ~~1085, a deletion mutant of SV40 lacking binding site I of the SV40 ori sequence, which was cloned in pKP55 (Miller et al., 1987). Enzymes Restriction enzymes and Escherichia co/i DNA polymerase Klenow fragment were obtained from Pharmacia, Freiburg (FRG). DNA preparation
for binding
assays
Plasmid DNA was grown in E. co/i HBI 01, isolated, and purified according to Maniatis et a/. (1982). After digestion with HindIll (pATC) or Taql (pKB1, pdllO85) the restriction fragments were endlabeled with o~-~“Plabeled nucleotides. Alternatively for quantitative binding assays the restriction fragments were separated on 2% agarose gels and the band containing the fragment with the specific binding sequence was eluted, quantified, and also labeled by filling in the recessed ends with 32P-labeled nucleotides according to Maniatis et al. (1982). lmmunoprecipitation
and quantitation
of T antigen
TC-7 cells infected with either wt or tsA58 virus were extracted at different times after infection with lysis buffer as described in the text (50 mlLl Tris-HCI, pH 8.0, 120 mn/l NaCI, 5% Trasylol, and 0.5% (v/v) Nonidet P-40) for 30 min at 4”. Extracts were clarified by centrifugation at 105,000 g for 30 min at 4’, and to assess the exact amount of T antigen present, at each time point quantitation was carried out by Western blotting. For this purpose T antigen was immunopre-
SV40
T-ANTIGEN
BINDING
cipitated for 90 min with PAb108, a monoclonal antibody specific for T antigen (Gurney et a/., 1986) and fixed Staphylococcus aureus cells were prepared as described by Kessler (1975). PAb108 was chosen in particular because this monoclonal antibody is known to precipitate T antigen completely in sequential precipitation experiments (Gurney et a/., 1986). Immunocomplexes were washed three times with binding buffer (10 mM HEPES, pH 7.8, 1 mM dithiothreitol (D-IT), 0.7 mM EDTA, 80 mM KCI, 1 mg/ml bovine serum albumin) and three times with NET buffer (50 mM Tris-HCI, pH 7.5, 5 mM EDTA, 100 mM KCI, 0.05% Nonidet P-40), eluted with 50 mM NH,HCO,, 2% SDS, 29’0 mercaptoethanol, and run on 10% SDSpolyacrylamide gels as described (Montenarh and Henning, 1983). For Western blotting protein bands were transferred from the gel to nitrocellulose filters and T antigen was specifically recognized with PAb108 and labeled with ‘251-labeled protein A (specific radioactivity 20 &i/pg) as described previously (Schtirmann et a/., 1985). T-antigen-specific bands were excised from the filter and counted.
TO
SITES
I AND
II
311
immunopurified T antigen. The Cerenkov radiation of the input DNA fragment and of that bound to the immunocomplex after incubation for 90 min at 0” was counted and specific binding was calculated as the ratio of bound DNA to amount of T antigen. Viral DNA synthesis Subconfluent TC-7 cells in 60-mm petri dishes were infected with 0.3 ml of SV40 virus stock. The cultures were incubated for times indicated in the text and then radiolabeled with [3H]thymidine (7 &i/dish). Viral DNA was extracted by the SDS-l M sodium chloride method of Hirt (1967). Calf thymus DNA (100 pg) was added to the supernatant, and the DNA was precipitated with cold 10% trichloroacetic acid. The DNA was collected on Schleicher & Schijll No. 6 filter, washed three times with 10% trichloroacetic acid, and dried. 3H radioactivity was determined by liquid scintillation counting of the dried paper in 10 ml of Aquasol(New England Nuclear Corp., Dreieich, FRG). RESULTS
Specific
DNA binding
assay
Immunoprecipitates, which were normalized for T antigen, were used for the modified McKay assay (MOller et al., 1987). We used exactly the same amounts of T antigen for binding DNA fragments containing either site I or site II. lmmunoprecipitates were washed three times with binding buffer and resuspended in 500 ~1 binding buffer together with 8 pg sonicated calf thymus DNA. Two hundred nanograms of 32P-end-labeled restriction fragments (about 1 O-20 pmol) was added and the reaction mixture was incubated for 90 min at 0”. Bound fragments were separated from free DNA by washing the binding complexes three times with NET buffer and eluted with 200 ~1 of 10 mMTris-HCI, pH 9.0, 1 mM EDTA by incubation for 30 min at 37”. To ensure that DNA was eluted quantitatively the immunocomplexes were counted for Cerenkov radiation. Eluted DNA was lyophilized and electrophoresed on a horizontal 2% agarose gel. Gels were dried and exposed to Kodak X-AR films between intensifying screens. Proteins were eluted with 50 mM NHIHCOB, 2% SDS, 2% mercaptoethanol, and analyzed by Western blotting as described above. To calculate specific binding the bands containing labeled DNA or T antigen were excised and counted for Cerenkov radiation or 1251radioactivity, respectively. For quantitative binding experiments 500 ~1 of the reaction mixture contained 1 pg sonicated calf thymus DNA, and 200 fmol of the 32P-end-labeled specific restriction fragment was incubated with nearly equal amounts of
Specific origin-binding activity of T antigen increases in the course of infection reaching saturation It is a well-accepted fact that by ? G-l 2 hr after the uptake of SV40 virus into permissive celis T antigen is synthesized in detectable quantities whereas viral DNA synthesis begins several hours later (Tooze, 1981). This delay in replication in the presence of T antigen could be due to either cellular factors which are not available so early on after infection or the failure of the small amounts of T antigen present to bind DNA specifically. In a first step we therefore infected TC-7 cells with wild-type SV40, and at the time points indicated in Fig. 2 cells were extracted with 0.5% NP-40 at pH 8.0. T antigen was immunoprecipitated with PAb108, a monoclonal antibody specific for T antigen (Gurney et al., 1986), and quantified by Western blotting. As seen in Fig. 2A the amount of T antigen increased continuously from the time of infection, that is, from scarcely detectable amounts under the condrtions used to 20 times this amount after 24 hr. Parallel to these quantitation experiments we have also analyzed the DNA binding properties of T antigen at the same time points after infection. For these binding studies we began by using the plasmid pATC which contains the SV40 ori region with all three binding sites (Fig. 1). We applied the recently described very sensitive target bound DNA binding assay (Hinzpeter era/., 1986; Mtiller et al., 1987) which uses immunopurified T antigen and which allowed us to standardize the in
STETTER,
312
MULLER,
AND
MONTENARH
60
20
!..--12
I..
12 FIG. 2. Increase of and extracted at the blotting (A) and tested T antigen of all time
15
18
21
24
15
18
21
24
36
h.p.i.
36
T antigen during wt infection and specific binding to the SV40 origin fragment. TC7 cells were infected with SV40 wt virus time points indicated. T antigen was immunoprecipitated with monoclonal antibody PAb108 and analyzed by Western for specific DNA binding to a DNA fragment containing all three binding sites (pATC) (B). (C)The specific binding activity of points by plotting the amount of bound DNA versus the amount of T antigen.
vitro binding conditions. With this assay in hand we were able to determine the quantity and specificity of input protein and bound DNA at the same time. As shown in Fig. 2B with the increase in the level of T antigen more and more of the specific fragment of pATC was bound. The binding efficiency reached saturation almost 20 hr after infection although template DNA was present in excess as compared to T antigen (see Materials and Methods). These results indicate that early on after virus infection the subclass of T antigen which binds DNA increased to a steady-state level. On calculating the specific binding activity, i.e., the ratio between bound DNA and amount of T antigen, it becomes clear that early on after infection nearly all of the newly synthesized T antigen binds to DNA whereas at later stages the nonbinding molecules accumulate resulting in a continuous decrease of the specific binding activity of T antigen (Fig. 2C). Binding of T antigen to sites course of an SV40 infection
I and II in the
Having seen that DNA binding of T antigen increases with increasing concentrations of T antigen
very early on after infection, reaching a steady-state level at about 20 hr after infection, we now addressed the question whether there is a difference in the binding of T antigen to the individual sites I or II and Ill in the course of the SV40 infection. To test for binding to site I the plasmid pKB1 was used which contains a synthetic oligonucleotide representing the 23 bp of site I (Ballas et a/., in preparation; Miiller et al., 1987). For binding to site II we used pdl1085, a cloned fragment of ~~1085 with site I deleted, retaining sites II and III with the neighboring SV40-specific sequences (Fig. 1). Furthermore the same amount of T antigen was used for binding to site I or site Il. As seen in Fig. 3 binding to site I (pKB1) resembled that of pATC in that it appeared concurrently with T antigen and increased steadily in the course of infection reaching saturation after around 20 hr. However, binding to site II was detectable only 8 to 9 hr later than binding to site I and then increased steadily reaching saturation later than 36 hr after infection. To test whether there is a correlation between the beginning of binding of T antigen to site II and the onset of viral DNA replication we analyzed viral DNA synthesis in the course of the virus infection by measuring the incorporation of [3H]thymidine into viral
SV40
T-ANTIGEN
BINDING
DNA. At the same time points as before, viral DNA was extracted according to Hirt (1967) and [3H]thymidine incorporation was determined by liquid scintillation counting. Figure 3 shows that viral DNA synthesis began at about 2 1 hr after infection. This onset of viral DNA synthesis corresponded to the time at which binding to site II was initiated and binding to site I reached saturation. Thus, very early after infection T antigen was able only to bind to site I, whereas binding to site II was detectable about 8 hr later and corresponded to the onset of DNA replication. Different heat-sensitivities for binding sites I and II
of tsA58
T antigen
The results described so far indicated a correlation between binding of T antigen to binding site II and the onset of viral DNA synthesis in the course of an SV40 infection. To further test this correlation we analyzed the binding of T antigen from the temperature-sensitive SV40 mutant tsA58 to the individual binding sites I or II and III. As tsA58 is defective for replication and autoregulation (Tegtmeyer et al., 1975; Reed et a/., 1976) one might assume that T antigen from this mutant is heat-sensitive for binding to both sites I and Il. TC-7 cells were infected with tsA58 or, in control experiments, with wild-type SV40 and cultured either at 32 or at 39.5” for 72 or 48 hr, respectively. These time points were chosen particularly in order to ensure that (i) T antigen would be capable of binding to both binding sites I and II and (ii) that binding to both sites had reached the plateau of a steady-state level. Figure 4A shows by Western blotting the aliquot of wild-type T antigen which was tested for binding to pATC, pKB1, or pd11085. Although we observed a slightly lower level of T antigen from wild-type-infected cells at 39.5 than at 32” equal amounts of the SV40-specific fragment of either plasmid were bound to T antigen (Fig. 4C). By calculating the specific binding activities we
FIG. 3. Viral DNA replication and T-antrgen brndrng to site I or sites II and Ill. TC-7 cells were infected wrth SV40 wt virus and at the time points Indicated after Infectton labeled with [3H]thymrdine to measure viral DNA synthesis (M). Alternatively, T antigen was extracted and assayed for specrflc DNA binding, either to DNA containing only site I (0) or sates II and III (0).
TO SITES
I AND
Ii
313
found that wild-type T antigen binds slightly better to all three fragments at 39.5 than at 32” (Fig. 4E). In contrast, concentrations of T antigen from tsA58-infected cells were higher at 39.5 than at 32” (Fig. 4B). However, this T antigen from tsA58-infected cells cultured at 39.5” completely failed to bind to pdl1085, containing binding sites II and Ill. It bound very weakly to the specific fragments of pATC and pKB1 representing either all three binding sites or only site I (Fig. 4D). After calculating the specific binding activity it became clear that this binding affinity represents 20% of the activity observed at 32” (Fig. 4F). Consequently, we were able to show that the replication deficiency of this mutant T antigen can be directly correlated with its inability to bind to site Il. The residual binding to site I is either too ineffective for the maintenance of the autoregulation or binding affinities for both sites I and II are necessary for autorepression. DISCUSSION Since viral DNA and early RNA synthesis initiates in the same region of the SV40 genome where T antigen binds, this protein-DNA interaction plays a key role in the viral life cycle. The regulatory region on the SV40 DNA contains three distinct binding sites for T antigen termed I, Ilj and III. We have recentiy shown that T antigen can bind with a high affinity to an oligonucleotide representing just the binding site I of the SV40 DNA. For an efficient binding to site II an SV40-specific A/T-rich sequence between sites II and III is absolutely necessary, but even in the presence of this A/Trich region dissociation equilibrium constants (K, values) for site II are seven times lower than for site I. Binding to site III is considerably weaker, in the range of the nonspecific binding of T antigen to vector fragments (Muller et a/., 1987). Since there is only a negligible contribution of site III to the interaction of T antigen with site II we have chosen pdl1085 which contains the specific fragment of sites II and III to study the binding of T antigen to site Il. The use of immunopurified T antigen in a modified target bound DNA binding assay (Hinzpeter et a/., 1986; %lvlOller et a/., 1987) allows an easy quantification of T antigen as well as of bound DNA and thus facilitates the determination of the specific binding activity of T antigen for the individual binding sites. As shown by several laboratories [35S]methionrnelabeled T antigen appears 1O-1 2 hr after the uptake of SV40 virus (Tooze, 1981). In order to detect total amounts of T antigen we drd not metabolically label the protein but used instead the very sensitive Western blotting procedure (Burnette, 1981). Under our experimental conditions we found detectable amounts of
314
STElTER,
MULLER, AND MONTENARH
T-Ag
T-Ag
tsA
58
FIG. 4. Comparison of specific DNA binding of T antigen from wt SV40 and tsA58 virus. TC7 cells were infected with wt or tsA58 virus at 32 or 39.5” for 48 or 72 hr, respectively. T antigen was extracted, and one aliquot was assayed by Western blotting (A and B) and the other was tested for specific binding to a DNA fragment containing sites I, Ii, and Ill (pATC), only site I (pKBl), or sites II and Ill (pdl1085) (C and D). The specific binding activity was determined from A-D by the ratio of cpm of bound RNA to cpm of T antigen (E and F).
protein at 15 hr after virus infection. Due to its high stability T antigen accumulates in the course of the virus infection (Edwards et al., 1979). At early time points after infection binding of T antigen to the HindIll C fragment of the SV40 DNA increases with increasing concentrations of T antigen. Despite a further accumulation of T antigen later on in infection, binding to all three binding sites reached a plateau. Therefore, it is not surprising that specific binding showed a constant decrease. Since it has been shown that only newly synthesized T antigen binds with high affinity to SV40 DNA (Scheidtmann et a/., 1984; Fanning et a/., 1982) it seems reasonable to assume that this plateau in the binding of SV40 ori DNA is due to a steady-state level in newly synthesized T antigen. This subclass may be
the same as that described by Scheller et al. (1982) and which is specifically recognized by monoclonal antibady PAblOO. According to our present results this small subclass of DNA-binding T antigen could be further separated according to its differential binding for sites I or II and III in the course of the SV40 infection. The delay in binding of T antigen to site II as compared to site I might be due to a change in post-translational modification of T antigen. However, there was no drop in binding of T antigen to site I; on the contrary, binding to site I increased further and simultaneously to site II until both reached the plateau already observed for the HindIll C fragment. In order to exclude that differences in binding to sites I and II might
SV40
T-ANTIGEN
BINDING
-0
SITES
I AND
II
sv 40 E
pATC --CcLch 0
00 %l?,”
pKB1
,o o”0 N a-
315
tsA 58 pdllO85
pATC -
F
00 2“, :i%
E EE 52s “0” -y) aP z0
g 0 u .=E sm Pl-
:IG.
pKBl -
pd11085 --
100 80 60 40 20
l-
4.-Continued.
be due to different concentrations of T antigen equal amounts of T antigen were used for the binding assays or the specific binding activities were calculated as the ratio of bound DNA to the amount of T antigen. Thus, a subclass of T-antigen binding to site II appears probable in addition to the subclass which binds to site I. DNase footprinting experiments have shown that initially binding site I is protected and in the presence of elevated levels of T antigen also a protection of sites II and III is seen which supports the hypothesis of a cooperative binding of T antigen to SV40 DNA (Myers e2 al., 1981). However, by using SV40 mutant DNAs lacking site I, Jones et a/. (1984) were able to demonstrate that binding to site II in vitro is not dependant on binding of T antigen to site I. Since we used isolated binding sites I or II and III we can clearly exclude any cooperative effects which might be necessary, and therefore responsible, for the delay in binding to site II. By analyzing a large set of point mutants or deletion mutants several laboratories (Margolskee and Nathans, 1984; DiMaio and Nathans, 1980; Deb et a/., 1986, 1987) have shown a strict correlation between an intact sequence of binding site II and the ability for SV40 DNA replication. In the present study we could demonstrate a chronological order in binding of T antigen initially to site I and with a delay subsequently to site Il. Furthermore, site II binding correlates perfectly with the onset of viral DNA replication. Furthermore, T antigen from a SV40 temperature-sensitive mutant which is heat-sensitive for viral DNA replication is also heat-sensitive for binding to site II. Thus far, several laboratories have analyzed the binding behavior of
tsA58 T antigen using different methods, i.e., DNase footprinting (Wilson et al., 1982) or McKay assay (Schurmann et al., 1985; Hinzpeter eta/., 1987). These assays were performed with an in virro purified and heat-shocked T antigen (Wilson era/., 1982), by in vivo shift up of tsA58-infected cells from permissive to nonpermissive temperature, or after permanent cultivation of tsA58-infected cells at the nonpermissive temperature (Schurmann et al., 1985). All these various methods revealed that T antigen from tsA58-infected cells is heat-sensitive for binding to the SV40 on DNA sequence including sites I, II, and III. On the contrary, it has also been reported that tsA58 T antigen is not thermolabile for SV40 DNA binding in vitro (Myers et a/., 1981). With our very sensitive quantitative binding assay we could demonstrate that tsA58 T antigen is clearly heat-sensitive in its ability to bind to site II. Nevertheless, there is residual binding for site I which is about 20°h of the activity observed at permissive temperature. In order to avoid misleading conclusions it is obviously essential from the data shown in Fig. 2 to analyze the DNA binding properties at a point of time after infection when binding has already reached steady-state level. Therefore we have analyzed DNA binding 48 hr (wt T antigen) or 72 hr (tsA58 T antigen) after infection. Reed et a/. (1976) have shown that tsA58 T antigen is heat-sensitive for the repression of its own synthesis. Direct evidence for the implication of wild-type T-antigen binding to site I in autoregulation was provided by experiments in which T antigen suppresses early transcription in vitro and in viva by interaction with binding site I (Rio et al., 1980;
316
STETTER,
MULLER.
Hansen et a/., 1981; DiMaio and Nathans, 1982; Benoist and Chambon, 1981). Indeed, by analyzing a number of mutants with mutations in binding site II for their ability to regulate early gene transcription Rio and Tjian (1983) found that additional binding to site II might contribute to a proper regulation of early SV40 gene transcription. Our results with tsA58 T antigen at nonpermissive temperature showing a residual binding to site I but a complete defect in binding to site II seem to confirm these observations. Moreover, it is also possible that the level of T-antigen binding to site I is too low at nonpermissive temperature to regulate early gene transcription. Although the interaction of T antigen with binding sites I, II, and probably III is absolutely necessary for DNA replication and autoregulation, there is increasing evidence indicating that other distinct biochemical and structural properties of T antigen are also necessary (Manos and Gluzman, 1984; Dean et a/., 1987; Schijrmann et al., 1985; Wachter et a/., 1985; Runzler et a/., 1987; Montenarh and Miiller, 1987). In addition to a defect in DNA binding, T antigen from tsA58 exhibits several structural defects at nonpermissive temperature, i.e., its heat-sensitivity toward the aggregation into high-molecular-weight oligomers and for the formation of complexes with the cellular oncoprotein p53 (Fanning et al., 1981; Montenarh et a/., 1984). However, some other SV40 mutants are known whose T antigen can bind to SV40 ori DNA and exhibit ATPase activity but are nevertheless defective in their support of viral DNA replication (Manos and Gluzman, 1984). There is increasing evidence for an implication of the oligomerization of T antigen in the SV40 DNA replication (Schtirmann et a/., 1985; Wachter et a/., 1985; Montenarh et a/., 1986; Montenarh and Miiller, 1987; Runzler et al., 1987). Therefore, it is tempting to speculate that distinct biochemical subclasses of T antigen may differ in their binding affinities for site I or site II. Runzler et al. (1987) have recently shown that only the 5-7 S subclasses of T antigen have high binding affinities for sites I, II, and III. But, other structural subclasses have to be analyzed for their differential binding to sites I and II.
ACKNOWLEDGMENTS The authors thank J. Lyons for editorial help. This study was supported by a grant from the Deutsche Forschungsgemeinschaft to M.M. (SF6 322, Al).
REFERENCES BENOIST, C., and CHAMBON, P. (1981). In vivo sequence requirements of the SV40 early promoter region. Nature (London) 290, 304-310.
AND
MONTENARH
BRADY, J., and KHOURY, G. (1985). Trans.activation of the SV40 late transcription unit by T antigen. MO/. Cell. Biol. 5, 1391-l 399. BURNEITE, N. N. (1981). “Western blotting:” Electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal. Biochem. 112, 195-203. CLARK, R., TEVETHIA, M. J., and TJIAN, R. (1984). The ATPase activity of SV40 large T antigen. In “Cancer Cells” (G. F. Vande Woude, A. J. Levine, W. C. Topp, and J. D. Watson, Eds.), Vol. 2, pp. 363-368. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. DEAN, F. B., BULLOCK, P., MURAKAMI, Y., WOBBE, C. R., WEISSBACH, L., and HURWITZ, J. (1987). Simian virus 40 (SV40) DNA replication: SV40 large T antigen unwinds DNA containing the SV40 origin of replication. Proc. Nat/. Acad. Sci. USA 84, 16-20. DEB, S., DELUCIA, A. L., BAUR, C. P., KOFF, A., and TEGTMEYER, P. (1986). Domain structure of the SV40 core origin of replication. Mol. Cell. Biol. 6, 1663-1670. DEB, S., SUI. S., KOFF, A., DELUCIA, A. L., PARSONS, R., and TEGTMEYER, P. (1987). The T-antigen-binding domain of the simian virus 40 core origin of replication. J. Viral. 61, 2143-2149. DIMAIO, D., and NATHANS, D. (1980). Cold sensitive regulatory mutants of SV40.J. Mol. Biol. 140, 129-l 42. DIMAIO, D., and NATHANS, D. (1982). Regulatory mutants of simian virus 40. Effect of mutations at a T antigen binding site on DNA replication and expression of viral genes. J. Mol. Biol. 156, 531-548. EDWARDS, C. A. F., KHOURY, G., and MARTIN, R. G. (1979). Phosphorylation of T-antigen and control of T-antigen expression in cells transformed by wild-type and tsA-mutants of simian virus 40. J. Viral. 29, 753-762. FANNING, E., NOWAK, B., and BURGER, C. (1981). Detection and characterization of multiple forms of simian virus 40 large T antigen. J. Viral. 37, 92-102. FANNING, E., WESTPHAL, K.-H., BRAUER, D., and C~RLIN, D. (1982). Subclasses of simian virus 40 large T antigen differential binding of two subclasses of T antigen from productively infected cells to viral and cellular DNA. EMBO J. 1, 1023-l 028. GIACHERIO, D., and HAGER, L. P. (1979). A poly(dT) stimulated ATPase activity associated with SV40 large T antigen. J. Biol. Chem. 254, 8113-8116. GURNEY, E. G., TAMOWSKI, S., and DEPPERT, W. (1986). Antigenic binding sites of monoclonal antibodies specific for simian virus 40 large T antigen. J. Viral. 57, 1168-l 172. HANSEN, U., TENEN. D., LIVINGSTON, D. M., and SHARP, P. A. (1981). T antigen repression of SV40 early transcription from two promotars. Cell 27, 603-612. HINZPETER, M., and DEPPERT, W. (1987). Analysis of biological and biochemical parameters for chromatin and nuclear matrix association of SV40 large T antigen in transformed cells. Oncogene 1, 119-129. HINZPETER, M., FANNING, E., and DEPPERT, W. (1986). A new sensitive target-bound DNA binding assay for SV40 T antigen. Virology 148,159-167. HIRT, B. (1967). Selective extraction of polyoma DNA from infected mouse cell cultures. J. Mol. Biol. 26, 365-369. JONES, K. A., MYERS, R. M., and TJIAN, R. (1984). Mutational analysis of SV40 large T antigen DNA binding sites. EMBO J. 3, 3247-3255. KELLER, J. M., and ALWINE, J. C. (1985). Analysis of an activatable promoter: Sequence in the SV40 late promoter required for T antigen mediated transactivation. Mol. Cell. Biol. 5, 1859-l 869. KESSLER, S. W. (1975). Rapid isolation of antigens from cells with a staphylococcal protein A-antibody adsorbent: Parameters of the
SV40
T-ANTIGEN
BINDING
interaction of antibody-antigen complexes with protein A. J. Immunoi. 115, 1617-1624. LANE, D. P., and CRAWFORD, L. V. (1979). T antigen is bound to a host protein in SV40-transformed cells. Nature (London) 278, 261-263. MANIAIIS, T., FRITSCH, E. F., and SAMBROOK, J. (1982). “Molecular Cloning: A Laboratory Manual.” Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. MANOS, M., and GLUZMAN, Y. (1984). Simian virus 40 large T antigen point mutants that are defective in viral DNA replication but competent in oncogenic transformation. Mol. Cell. Biol. 4, 1 125-l 133. MARGOLSKEE, R., and NATHANS, D. (1984). SV40 mutant T antigens with relaxed specificity for the nucleotide sequence at the viral DNA origin of replication. 1. Viral. 49, 385-393. MONTENARH, M., and HENNING, R. (1983). Self assembly of SV40 large T antigen oligomers by divalent cations. J. Viral. 45, 531-538. MONTENARH, M., KOHLER, M., and HENNING, R. (1984). The oligomerization of simian virus 40 large T antigen is not necessarily effected by temperature sensitive A gene lesions. J. Viral. 49, 658-664. MONTENARH, M., and Mali-LER, D. (1987). The phosphotylation at Thr 124 of simian virus 40 large T antigen is crucial for its oligomerization. FEBS Letl. 221, 199-204. MONTENARH, M., VESCO, C., KEMMERLING, G., MOLLER, D., and HENNING, R. (1986). Regions of SV40 large T antigen necessary for oligomerization and complex formation with the cellular oncoprotein ~53. FEBS Lett. 204, 51-55. MOLLER, D., UGI, I., BALLAS, K.; REISER, P., HENNING, R., and MONTENARH, M. (1987). The AT-rich sequence of the SV40 control region influences the binding of SV40 T antigen to binding sites II and III. I/iro/ogy 161, 81-91. MYERS, R. M., RIO, D. C., ROBBINS, A. K., ~~~TJIAN, R. (1981). SV40 gene expression is modulated by the cooperative binding of T antigen to DNA. Cell 25, 373-384. REED, S. Jo, STARK, 6. R., and ALWINE, J. C. (1976). Autoregulation of simian virus 40 gene A by T antigen. Proc. Nat/. Acad. Sci. USA 73, 3083-308-Y. RIGEY, P., and ~.ANE, D. P. (1983). Structure and function of simian virus 40 large T antigen. in “Advances in Viral Oncology” (G. Klein, Ed.), Vol. 3., pp. 31-57. Raven Press, New York. RIO, D., ROBBINS, A., MYERS, R., and TJIAN, R. (1980). Regulation of SV40 early transcription in vitro by a purified tumor antigen. hoc. Nat/. Acad. Sci. USA 77, 5706-5710. RIO, D. C., and TJIAN, R. (1983). SV40 T antigen binding site mutations that affect autoregulation. Cell 32, 1227-l 240.
TO SITES
I AND
II
317
RUNZLER, R., THOMPSON, S., and FANNING, E. (1987). Oiigomerization and origin DNA-binding activity of simian virus 40 iarge T antigen. J. Viral 61, 2076-2083. RYDER, K., VAKALOPOULOU, E., MERTZ, R., MASTRANGELO, J., HOUGH, P., TEGTMEYER, P., and FANNING, E. (1985). Seventeen base pairs of region I encode a novel tripartite binding signal for SV40 T antigen. Cell 42, 539-548. SCHEIDTMANN, K.-H., HARDUNG, M., ECHLE, B., and WALTER, G. (1984). DNA binding activity of SV40 large T antigen correlates with a distinct phosphorylation state. J. Viral. 50, l-l 2. SCHELLER, A., COVEY, L., BARNET, B., and PRIVES, C. (1982). A small subclass of SV40 T antigen binds to the viral origin of replication. Cell 29, 375-383. SCH~~RMANN, C., MONTENARH, M., KOHLER, M., and HENNING, R. (1985). Oligomerization of simian virus 411 tumor antigen may be involved in viral DNA replication. Viroiogy 14& 1-l 1. SHALLOWAY, D., KLEINBERGER, T., and LIVINGSTON, D. M. (1980). Mapping of SV40 DNA replication origin region binding sites for the SV40 T antigen by protection against exonuciease III digestion. cell 20, 41 l-422. SHORTLE, D., MARGOLSKEE, R.; and NATCIANS, D. (1979). Mutational analysis of the SV40 replicon: Pseudorevertants of mutants with a defective replication origin. Proc. Nati. Acad. SC;. USA 76, 6128-6131. SMALE, S. T., and TJIAN, R. (1986). T-antigen-DNA polymerase CY complex implicated in simian virus 40 DNA replication. P/o/. Cell. Biol. 6, 4077-4087. STAHL, H., DR~GE, P., and KNIPPERS, R. (1986). DNA helicase activity of SV40 large tumor antigen. En/lBO J. 5, 1939-l 944. TEGTMEYER, P., LEW~ON, B. A., DELUCIA, A. L., WILSON, V. G., and RYDER, K. (1983). Topography of SV40 A protein-DNA compiexes: Arrangement of protein bound to the origin of replication. J. Virol. 46, 151-161. TEGTMEYER, P., SCHWARZ, M., COLLINS, J. K., and RUNDELL, K. R. (1975). Regulation of tumor antigen synthesis by simian virus 40 gene A. 1. Viral. 16, 168-178. TJIAN. R., and RO~BINS, A. (1979). Enzymatic activities associated with a purified simian virus 40 T-antigen-related protein. Proc. Nat/. Acad. Sci. USA 76, 61 O-61 5. TOOZE, J. (1981). “The Molecular Biology of Tumor Viruses,” 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. WACHTER, M., RIEDLE, G., and HENNING, R. (1985). Functional implications of oligomerization of simian virus 40 large T antigen during lytic virus infection. 1. Viral. 56, 520-526.
164,
309-3
17
(1988)
SV40 T-Antigen
Binding to Site II Is Functionally
GijNTER Department
STElTER,
of Biochem\stry,
DOROTHEE University
Received
MijLLER.
of U/m, P. 0. Box 4066,
October
15, 1987; accepted
Separated AND
MATHIAS
from Binding
to Site I
MONTENARY’
O-7900
U/m, federal
January
14. 1988
Republic
of Germany
During lytic infection SV40 T antigen binds specifically to three different regions of the SV40 DNA to initiate viral DNA replication and to regulate early and late transcription. We have used the recently described plasmids pKB1, containing a 23-bp oligonucleotide coding for site I, pdl1085 containing sites II and Ill together with SV40 specific flanking sequences, and as a control pATC, a plasmid which contains all three binding sites (D. Mijller et al. (1987), Virology 161, 81-91) to analyze the differential binding of T antigen to these individual binding sites in the course of an SV40 infection. We found that shortly after infection the amount of bound DNA increased with the concentration of T antigen reaching a steady-state level at about 20 hr after infection. In comparison to binding at site I, binding to site II appeared with a delay of about 8-9 hr corresponding to the onset of viral DNA replication. The correlation between binding of T antigen to site II and the SV40 DNA replication could be further corroborated by using T antigen from the heat-sensitive mutant tsA58 which completely failed to bind to site II at nonpermissive temperature but exhibited a residual binding to site I. This reduced binding to site I proved insufficient for the proper functioning of autorepression. Our results support the hypothesis that distinctly different subclasses of T-antigen binding to site I or site II may exist. C 1989 Academic
Press,
Inc.
pairs (bp) that is both necessary and sufficient for Tantigen recognitron (Ryder et a/., 1985). To initiate viral DNA replication, direct interaction of T antigen with another distinct binding site within the SV40 control region is necessary (Deb et a/., 1987). This sequence, binding site I!, is a part of the core origin region reaching from nucleotides 5209 to 34. After binding DNA T antigen covers a ca. 50-bp-!ong sequence of the SV40 DNA. This binding site consists of a set of four pentanucleotides in palindromic order and a stretch of continuous A/T pairs which also seems to be crucial for binding (Mijller et a/., 1987). In addition to the origin binding activity, in viva and in vitro studies suggest that requirement for triggering viral DNA replication involve at least an ATPase and helicase activity of T antigen (Manos and Gluzman, 1984; Dean et al., 1987; Stahl er al., 1986). Moreover, Smale and Tjian (1986) found that T antigen coprecipitates with cellular DNA polymerase CY,an enzyme involved In DNA synthesis. We have recently described the c!oning of a 23.bp oligonucleotide representing binding site I alone and several plasmids containing the 23 bp of binding sate II either with or without netghboring SV40-specific flanking sequences (MOller et a/., 1987). We were ab!e to show that T antigen binds with high affinity to site I on its own and with a very much reduced affinity to site II but only in the presence of the AT-rich sequence between sites II and III. In the present paper we used two of these plasmids, pKB1 and pdllO85, representing either isolated site I or sites II and III with SV40-specific
INTRODUCTION The simian virus 40 (SV40) infectious cycle in cultured monkey cells is primarily regulated by one of the early gene products, large T (tumor) antigen (Tooze, 1981; Rigby and Lane, 1983). This multifunctional protein is required for initiation (Shortle et al., 1979; Margolskee and Nathans, 1984) and probably elongation (Wiekowskr et a/., 1987) of viral DNA replication, and for stimulation of late and repression of early viral gene expression (Keller and Alwine, 1985; Brady and Khoury, 1985; Myers et a/., 1981; Rio and Tjian, 1983). Several distinct biochemical activities are carried out by T antigen: specific binding to multiple sites at the SV40 origin (Shalloway et a/., 1980; Tegtmeyer et a/., l983), ATP hydrolysis (Clark et al., 1984; Giacherio and Hager 1979; Tjian and Robbins, 1979), unwinding of short double-helical DNA strands (Stahl eta/., 1986; Dean et a/., 1987), and association with a 53-kDa cellular protein (Lane and Crawford, 1979). Late in infection the expression of the early gene products, that is, small and large T antigen, is downregulated (Tegtmeyer et a/., 1975) by an interaction of T antigen with at least one sequence in the SV40 control region, binding site I (Rio et al., 1980; Hansen et al., 1981; Myers et a/., 1981). This site contains three pentanucleotides (-GAGGC-), two of which are separated by a cluster of six ArT pairs. ln vitro binding studies with site I defined a central core of 17 base ’ To whom
requests
for reprints
should
be addressed. 309
0042.6822/88 Copyright
$3.00
Q 1968 by Academic
Press. Inc
STE-iTER, MOLLER, AND MONTENARH
310
I SV 40 ori - region
II 4 origin
early
mRNA
9-1
I
II II
-
III
plate
mRNA II -
of replication
insert Hind site Hind
Ill C of SV40
---,
-
I of SV40
____ _-IL-
III/Sph
--md”“‘,.
l of cs 1085
I
_________
Ill
___________._
// -r---________ __________
-__-____.
___.______
_ ________-----
_-_
______-____vectorsequences -
SV40
sequences
with
binding
site
1. Plasmids used for binding studies. pATC contains the HindIll C fragment of SV40, pKB1 contains a synthetic oligonucleotide for site I, pdllO85 contains the origin fragment of ~~1085, a deletion mutant of SV40 lacking binding site I. FIG.
flanking sequences in quantitative in vitro binding assays to analyze a possible correlation between sitespecific binding of T antigen and its functions in the course of a SV40 virus infection. As a control we used T antigen from the temperature-sensitive SV40 virus mutant tsA58 deficient in the control of autoregulation and initiation of DNA synthesis, We found that shortly after SV40 infection of permissive monkey cells the increase in the T-antigen level correlated with an increase in binding to SV40 ori DNA. Although T-antigen concentrations increased further later on after infection, DNA binding reached a steady-state level. In the course of the infection T antigen first bound to site I and later on after infection to sites II and Ill also. Binding to sites II and III was detectable shortly before the onset of viral DNA synthesis. Furthermore, T antigen from the temperature-sensitive mutant tsA58 which is heat-sensitive for DNA replication and autoregulation is strictly heat-sensitive for binding to sites II and III but exhibited a residual binding to site I. MATERIALS
AND METHODS
Cells and viruses TC-7, an established line of African green monkey kidney cells, was used for all infections with SV40 wt and tsA mutants. The cells were grown in Dulbecco’s modified Eagle’s medium containing 5% fetal calf serum. Stocks of SV40 wt strain 776 and of tsA68 were prepared and tritrated as previously described (SchUrmann et a/., 1985). Template
DNAs
For specific DNA binding assays three different plasmid DNAs were used (Fig. 1): pATC represents the vector pATl53 with a HindIll C fragment of SV40 (nu-
coding
cleotides 1046-5172) which contains all three binding sites; pKB1 contains a synthetic oligonucleotide representing SV40 site I (nucleotides 5186-5209) cloned in the HindIIIIBamHI site of pBR322; pdl1085 contains the HindIll (5172) to Sphl (133) fragment of ~~1085, a deletion mutant of SV40 lacking binding site I of the SV40 ori sequence, which was cloned in pKP55 (Miller et al., 1987). Enzymes Restriction enzymes and Escherichia co/i DNA polymerase Klenow fragment were obtained from Pharmacia, Freiburg (FRG). DNA preparation
for binding
assays
Plasmid DNA was grown in E. co/i HBI 01, isolated, and purified according to Maniatis et a/. (1982). After digestion with HindIll (pATC) or Taql (pKB1, pdllO85) the restriction fragments were endlabeled with o~-~“Plabeled nucleotides. Alternatively for quantitative binding assays the restriction fragments were separated on 2% agarose gels and the band containing the fragment with the specific binding sequence was eluted, quantified, and also labeled by filling in the recessed ends with 32P-labeled nucleotides according to Maniatis et al. (1982). lmmunoprecipitation
and quantitation
of T antigen
TC-7 cells infected with either wt or tsA58 virus were extracted at different times after infection with lysis buffer as described in the text (50 mlLl Tris-HCI, pH 8.0, 120 mn/l NaCI, 5% Trasylol, and 0.5% (v/v) Nonidet P-40) for 30 min at 4”. Extracts were clarified by centrifugation at 105,000 g for 30 min at 4’, and to assess the exact amount of T antigen present, at each time point quantitation was carried out by Western blotting. For this purpose T antigen was immunopre-
SV40
T-ANTIGEN
BINDING
cipitated for 90 min with PAb108, a monoclonal antibody specific for T antigen (Gurney et a/., 1986) and fixed Staphylococcus aureus cells were prepared as described by Kessler (1975). PAb108 was chosen in particular because this monoclonal antibody is known to precipitate T antigen completely in sequential precipitation experiments (Gurney et a/., 1986). Immunocomplexes were washed three times with binding buffer (10 mM HEPES, pH 7.8, 1 mM dithiothreitol (D-IT), 0.7 mM EDTA, 80 mM KCI, 1 mg/ml bovine serum albumin) and three times with NET buffer (50 mM Tris-HCI, pH 7.5, 5 mM EDTA, 100 mM KCI, 0.05% Nonidet P-40), eluted with 50 mM NH,HCO,, 2% SDS, 29’0 mercaptoethanol, and run on 10% SDSpolyacrylamide gels as described (Montenarh and Henning, 1983). For Western blotting protein bands were transferred from the gel to nitrocellulose filters and T antigen was specifically recognized with PAb108 and labeled with ‘251-labeled protein A (specific radioactivity 20 &i/pg) as described previously (Schtirmann et a/., 1985). T-antigen-specific bands were excised from the filter and counted.
TO
SITES
I AND
II
311
immunopurified T antigen. The Cerenkov radiation of the input DNA fragment and of that bound to the immunocomplex after incubation for 90 min at 0” was counted and specific binding was calculated as the ratio of bound DNA to amount of T antigen. Viral DNA synthesis Subconfluent TC-7 cells in 60-mm petri dishes were infected with 0.3 ml of SV40 virus stock. The cultures were incubated for times indicated in the text and then radiolabeled with [3H]thymidine (7 &i/dish). Viral DNA was extracted by the SDS-l M sodium chloride method of Hirt (1967). Calf thymus DNA (100 pg) was added to the supernatant, and the DNA was precipitated with cold 10% trichloroacetic acid. The DNA was collected on Schleicher & Schijll No. 6 filter, washed three times with 10% trichloroacetic acid, and dried. 3H radioactivity was determined by liquid scintillation counting of the dried paper in 10 ml of Aquasol(New England Nuclear Corp., Dreieich, FRG). RESULTS
Specific
DNA binding
assay
Immunoprecipitates, which were normalized for T antigen, were used for the modified McKay assay (MOller et al., 1987). We used exactly the same amounts of T antigen for binding DNA fragments containing either site I or site II. lmmunoprecipitates were washed three times with binding buffer and resuspended in 500 ~1 binding buffer together with 8 pg sonicated calf thymus DNA. Two hundred nanograms of 32P-end-labeled restriction fragments (about 1 O-20 pmol) was added and the reaction mixture was incubated for 90 min at 0”. Bound fragments were separated from free DNA by washing the binding complexes three times with NET buffer and eluted with 200 ~1 of 10 mMTris-HCI, pH 9.0, 1 mM EDTA by incubation for 30 min at 37”. To ensure that DNA was eluted quantitatively the immunocomplexes were counted for Cerenkov radiation. Eluted DNA was lyophilized and electrophoresed on a horizontal 2% agarose gel. Gels were dried and exposed to Kodak X-AR films between intensifying screens. Proteins were eluted with 50 mM NHIHCOB, 2% SDS, 2% mercaptoethanol, and analyzed by Western blotting as described above. To calculate specific binding the bands containing labeled DNA or T antigen were excised and counted for Cerenkov radiation or 1251radioactivity, respectively. For quantitative binding experiments 500 ~1 of the reaction mixture contained 1 pg sonicated calf thymus DNA, and 200 fmol of the 32P-end-labeled specific restriction fragment was incubated with nearly equal amounts of
Specific origin-binding activity of T antigen increases in the course of infection reaching saturation It is a well-accepted fact that by ? G-l 2 hr after the uptake of SV40 virus into permissive celis T antigen is synthesized in detectable quantities whereas viral DNA synthesis begins several hours later (Tooze, 1981). This delay in replication in the presence of T antigen could be due to either cellular factors which are not available so early on after infection or the failure of the small amounts of T antigen present to bind DNA specifically. In a first step we therefore infected TC-7 cells with wild-type SV40, and at the time points indicated in Fig. 2 cells were extracted with 0.5% NP-40 at pH 8.0. T antigen was immunoprecipitated with PAb108, a monoclonal antibody specific for T antigen (Gurney et al., 1986), and quantified by Western blotting. As seen in Fig. 2A the amount of T antigen increased continuously from the time of infection, that is, from scarcely detectable amounts under the condrtions used to 20 times this amount after 24 hr. Parallel to these quantitation experiments we have also analyzed the DNA binding properties of T antigen at the same time points after infection. For these binding studies we began by using the plasmid pATC which contains the SV40 ori region with all three binding sites (Fig. 1). We applied the recently described very sensitive target bound DNA binding assay (Hinzpeter era/., 1986; Mtiller et al., 1987) which uses immunopurified T antigen and which allowed us to standardize the in
STETTER,
312
MULLER,
AND
MONTENARH
60
20
!..--12
I..
12 FIG. 2. Increase of and extracted at the blotting (A) and tested T antigen of all time
15
18
21
24
15
18
21
24
36
h.p.i.
36
T antigen during wt infection and specific binding to the SV40 origin fragment. TC7 cells were infected with SV40 wt virus time points indicated. T antigen was immunoprecipitated with monoclonal antibody PAb108 and analyzed by Western for specific DNA binding to a DNA fragment containing all three binding sites (pATC) (B). (C)The specific binding activity of points by plotting the amount of bound DNA versus the amount of T antigen.
vitro binding conditions. With this assay in hand we were able to determine the quantity and specificity of input protein and bound DNA at the same time. As shown in Fig. 2B with the increase in the level of T antigen more and more of the specific fragment of pATC was bound. The binding efficiency reached saturation almost 20 hr after infection although template DNA was present in excess as compared to T antigen (see Materials and Methods). These results indicate that early on after virus infection the subclass of T antigen which binds DNA increased to a steady-state level. On calculating the specific binding activity, i.e., the ratio between bound DNA and amount of T antigen, it becomes clear that early on after infection nearly all of the newly synthesized T antigen binds to DNA whereas at later stages the nonbinding molecules accumulate resulting in a continuous decrease of the specific binding activity of T antigen (Fig. 2C). Binding of T antigen to sites course of an SV40 infection
I and II in the
Having seen that DNA binding of T antigen increases with increasing concentrations of T antigen
very early on after infection, reaching a steady-state level at about 20 hr after infection, we now addressed the question whether there is a difference in the binding of T antigen to the individual sites I or II and Ill in the course of the SV40 infection. To test for binding to site I the plasmid pKB1 was used which contains a synthetic oligonucleotide representing the 23 bp of site I (Ballas et a/., in preparation; Miiller et al., 1987). For binding to site II we used pdl1085, a cloned fragment of ~~1085 with site I deleted, retaining sites II and III with the neighboring SV40-specific sequences (Fig. 1). Furthermore the same amount of T antigen was used for binding to site I or site Il. As seen in Fig. 3 binding to site I (pKB1) resembled that of pATC in that it appeared concurrently with T antigen and increased steadily in the course of infection reaching saturation after around 20 hr. However, binding to site II was detectable only 8 to 9 hr later than binding to site I and then increased steadily reaching saturation later than 36 hr after infection. To test whether there is a correlation between the beginning of binding of T antigen to site II and the onset of viral DNA replication we analyzed viral DNA synthesis in the course of the virus infection by measuring the incorporation of [3H]thymidine into viral
SV40
T-ANTIGEN
BINDING
DNA. At the same time points as before, viral DNA was extracted according to Hirt (1967) and [3H]thymidine incorporation was determined by liquid scintillation counting. Figure 3 shows that viral DNA synthesis began at about 2 1 hr after infection. This onset of viral DNA synthesis corresponded to the time at which binding to site II was initiated and binding to site I reached saturation. Thus, very early after infection T antigen was able only to bind to site I, whereas binding to site II was detectable about 8 hr later and corresponded to the onset of DNA replication. Different heat-sensitivities for binding sites I and II
of tsA58
T antigen
The results described so far indicated a correlation between binding of T antigen to binding site II and the onset of viral DNA synthesis in the course of an SV40 infection. To further test this correlation we analyzed the binding of T antigen from the temperature-sensitive SV40 mutant tsA58 to the individual binding sites I or II and III. As tsA58 is defective for replication and autoregulation (Tegtmeyer et al., 1975; Reed et a/., 1976) one might assume that T antigen from this mutant is heat-sensitive for binding to both sites I and Il. TC-7 cells were infected with tsA58 or, in control experiments, with wild-type SV40 and cultured either at 32 or at 39.5” for 72 or 48 hr, respectively. These time points were chosen particularly in order to ensure that (i) T antigen would be capable of binding to both binding sites I and II and (ii) that binding to both sites had reached the plateau of a steady-state level. Figure 4A shows by Western blotting the aliquot of wild-type T antigen which was tested for binding to pATC, pKB1, or pd11085. Although we observed a slightly lower level of T antigen from wild-type-infected cells at 39.5 than at 32” equal amounts of the SV40-specific fragment of either plasmid were bound to T antigen (Fig. 4C). By calculating the specific binding activities we
FIG. 3. Viral DNA replication and T-antrgen brndrng to site I or sites II and Ill. TC-7 cells were infected wrth SV40 wt virus and at the time points Indicated after Infectton labeled with [3H]thymrdine to measure viral DNA synthesis (M). Alternatively, T antigen was extracted and assayed for specrflc DNA binding, either to DNA containing only site I (0) or sates II and III (0).
TO SITES
I AND
Ii
313
found that wild-type T antigen binds slightly better to all three fragments at 39.5 than at 32” (Fig. 4E). In contrast, concentrations of T antigen from tsA58-infected cells were higher at 39.5 than at 32” (Fig. 4B). However, this T antigen from tsA58-infected cells cultured at 39.5” completely failed to bind to pdl1085, containing binding sites II and Ill. It bound very weakly to the specific fragments of pATC and pKB1 representing either all three binding sites or only site I (Fig. 4D). After calculating the specific binding activity it became clear that this binding affinity represents 20% of the activity observed at 32” (Fig. 4F). Consequently, we were able to show that the replication deficiency of this mutant T antigen can be directly correlated with its inability to bind to site Il. The residual binding to site I is either too ineffective for the maintenance of the autoregulation or binding affinities for both sites I and II are necessary for autorepression. DISCUSSION Since viral DNA and early RNA synthesis initiates in the same region of the SV40 genome where T antigen binds, this protein-DNA interaction plays a key role in the viral life cycle. The regulatory region on the SV40 DNA contains three distinct binding sites for T antigen termed I, Ilj and III. We have recentiy shown that T antigen can bind with a high affinity to an oligonucleotide representing just the binding site I of the SV40 DNA. For an efficient binding to site II an SV40-specific A/T-rich sequence between sites II and III is absolutely necessary, but even in the presence of this A/Trich region dissociation equilibrium constants (K, values) for site II are seven times lower than for site I. Binding to site III is considerably weaker, in the range of the nonspecific binding of T antigen to vector fragments (Muller et a/., 1987). Since there is only a negligible contribution of site III to the interaction of T antigen with site II we have chosen pdl1085 which contains the specific fragment of sites II and III to study the binding of T antigen to site Il. The use of immunopurified T antigen in a modified target bound DNA binding assay (Hinzpeter et a/., 1986; %lvlOller et a/., 1987) allows an easy quantification of T antigen as well as of bound DNA and thus facilitates the determination of the specific binding activity of T antigen for the individual binding sites. As shown by several laboratories [35S]methionrnelabeled T antigen appears 1O-1 2 hr after the uptake of SV40 virus (Tooze, 1981). In order to detect total amounts of T antigen we drd not metabolically label the protein but used instead the very sensitive Western blotting procedure (Burnette, 1981). Under our experimental conditions we found detectable amounts of
314
STElTER,
MULLER, AND MONTENARH
T-Ag
T-Ag
tsA
58
FIG. 4. Comparison of specific DNA binding of T antigen from wt SV40 and tsA58 virus. TC7 cells were infected with wt or tsA58 virus at 32 or 39.5” for 48 or 72 hr, respectively. T antigen was extracted, and one aliquot was assayed by Western blotting (A and B) and the other was tested for specific binding to a DNA fragment containing sites I, Ii, and Ill (pATC), only site I (pKBl), or sites II and Ill (pdl1085) (C and D). The specific binding activity was determined from A-D by the ratio of cpm of bound RNA to cpm of T antigen (E and F).
protein at 15 hr after virus infection. Due to its high stability T antigen accumulates in the course of the virus infection (Edwards et al., 1979). At early time points after infection binding of T antigen to the HindIll C fragment of the SV40 DNA increases with increasing concentrations of T antigen. Despite a further accumulation of T antigen later on in infection, binding to all three binding sites reached a plateau. Therefore, it is not surprising that specific binding showed a constant decrease. Since it has been shown that only newly synthesized T antigen binds with high affinity to SV40 DNA (Scheidtmann et a/., 1984; Fanning et a/., 1982) it seems reasonable to assume that this plateau in the binding of SV40 ori DNA is due to a steady-state level in newly synthesized T antigen. This subclass may be
the same as that described by Scheller et al. (1982) and which is specifically recognized by monoclonal antibady PAblOO. According to our present results this small subclass of DNA-binding T antigen could be further separated according to its differential binding for sites I or II and III in the course of the SV40 infection. The delay in binding of T antigen to site II as compared to site I might be due to a change in post-translational modification of T antigen. However, there was no drop in binding of T antigen to site I; on the contrary, binding to site I increased further and simultaneously to site II until both reached the plateau already observed for the HindIll C fragment. In order to exclude that differences in binding to sites I and II might
SV40
T-ANTIGEN
BINDING
-0
SITES
I AND
II
sv 40 E
pATC --CcLch 0
00 %l?,”
pKB1
,o o”0 N a-
315
tsA 58 pdllO85
pATC -
F
00 2“, :i%
E EE 52s “0” -y) aP z0
g 0 u .=E sm Pl-
:IG.
pKBl -
pd11085 --
100 80 60 40 20
l-
4.-Continued.
be due to different concentrations of T antigen equal amounts of T antigen were used for the binding assays or the specific binding activities were calculated as the ratio of bound DNA to the amount of T antigen. Thus, a subclass of T-antigen binding to site II appears probable in addition to the subclass which binds to site I. DNase footprinting experiments have shown that initially binding site I is protected and in the presence of elevated levels of T antigen also a protection of sites II and III is seen which supports the hypothesis of a cooperative binding of T antigen to SV40 DNA (Myers e2 al., 1981). However, by using SV40 mutant DNAs lacking site I, Jones et a/. (1984) were able to demonstrate that binding to site II in vitro is not dependant on binding of T antigen to site I. Since we used isolated binding sites I or II and III we can clearly exclude any cooperative effects which might be necessary, and therefore responsible, for the delay in binding to site II. By analyzing a large set of point mutants or deletion mutants several laboratories (Margolskee and Nathans, 1984; DiMaio and Nathans, 1980; Deb et a/., 1986, 1987) have shown a strict correlation between an intact sequence of binding site II and the ability for SV40 DNA replication. In the present study we could demonstrate a chronological order in binding of T antigen initially to site I and with a delay subsequently to site Il. Furthermore, site II binding correlates perfectly with the onset of viral DNA replication. Furthermore, T antigen from a SV40 temperature-sensitive mutant which is heat-sensitive for viral DNA replication is also heat-sensitive for binding to site II. Thus far, several laboratories have analyzed the binding behavior of
tsA58 T antigen using different methods, i.e., DNase footprinting (Wilson et al., 1982) or McKay assay (Schurmann et al., 1985; Hinzpeter eta/., 1987). These assays were performed with an in virro purified and heat-shocked T antigen (Wilson era/., 1982), by in vivo shift up of tsA58-infected cells from permissive to nonpermissive temperature, or after permanent cultivation of tsA58-infected cells at the nonpermissive temperature (Schurmann et al., 1985). All these various methods revealed that T antigen from tsA58-infected cells is heat-sensitive for binding to the SV40 on DNA sequence including sites I, II, and III. On the contrary, it has also been reported that tsA58 T antigen is not thermolabile for SV40 DNA binding in vitro (Myers et a/., 1981). With our very sensitive quantitative binding assay we could demonstrate that tsA58 T antigen is clearly heat-sensitive in its ability to bind to site II. Nevertheless, there is residual binding for site I which is about 20°h of the activity observed at permissive temperature. In order to avoid misleading conclusions it is obviously essential from the data shown in Fig. 2 to analyze the DNA binding properties at a point of time after infection when binding has already reached steady-state level. Therefore we have analyzed DNA binding 48 hr (wt T antigen) or 72 hr (tsA58 T antigen) after infection. Reed et a/. (1976) have shown that tsA58 T antigen is heat-sensitive for the repression of its own synthesis. Direct evidence for the implication of wild-type T-antigen binding to site I in autoregulation was provided by experiments in which T antigen suppresses early transcription in vitro and in viva by interaction with binding site I (Rio et al., 1980;
316
STETTER,
MULLER.
Hansen et a/., 1981; DiMaio and Nathans, 1982; Benoist and Chambon, 1981). Indeed, by analyzing a number of mutants with mutations in binding site II for their ability to regulate early gene transcription Rio and Tjian (1983) found that additional binding to site II might contribute to a proper regulation of early SV40 gene transcription. Our results with tsA58 T antigen at nonpermissive temperature showing a residual binding to site I but a complete defect in binding to site II seem to confirm these observations. Moreover, it is also possible that the level of T-antigen binding to site I is too low at nonpermissive temperature to regulate early gene transcription. Although the interaction of T antigen with binding sites I, II, and probably III is absolutely necessary for DNA replication and autoregulation, there is increasing evidence indicating that other distinct biochemical and structural properties of T antigen are also necessary (Manos and Gluzman, 1984; Dean et a/., 1987; Schijrmann et al., 1985; Wachter et a/., 1985; Runzler et a/., 1987; Montenarh and Miiller, 1987). In addition to a defect in DNA binding, T antigen from tsA58 exhibits several structural defects at nonpermissive temperature, i.e., its heat-sensitivity toward the aggregation into high-molecular-weight oligomers and for the formation of complexes with the cellular oncoprotein p53 (Fanning et al., 1981; Montenarh et a/., 1984). However, some other SV40 mutants are known whose T antigen can bind to SV40 ori DNA and exhibit ATPase activity but are nevertheless defective in their support of viral DNA replication (Manos and Gluzman, 1984). There is increasing evidence for an implication of the oligomerization of T antigen in the SV40 DNA replication (Schtirmann et a/., 1985; Wachter et a/., 1985; Montenarh et a/., 1986; Montenarh and Miiller, 1987; Runzler et al., 1987). Therefore, it is tempting to speculate that distinct biochemical subclasses of T antigen may differ in their binding affinities for site I or site II. Runzler et al. (1987) have recently shown that only the 5-7 S subclasses of T antigen have high binding affinities for sites I, II, and III. But, other structural subclasses have to be analyzed for their differential binding to sites I and II.
ACKNOWLEDGMENTS The authors thank J. Lyons for editorial help. This study was supported by a grant from the Deutsche Forschungsgemeinschaft to M.M. (SF6 322, Al).
REFERENCES BENOIST, C., and CHAMBON, P. (1981). In vivo sequence requirements of the SV40 early promoter region. Nature (London) 290, 304-310.
AND
MONTENARH
BRADY, J., and KHOURY, G. (1985). Trans.activation of the SV40 late transcription unit by T antigen. MO/. Cell. Biol. 5, 1391-l 399. BURNEITE, N. N. (1981). “Western blotting:” Electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal. Biochem. 112, 195-203. CLARK, R., TEVETHIA, M. J., and TJIAN, R. (1984). The ATPase activity of SV40 large T antigen. In “Cancer Cells” (G. F. Vande Woude, A. J. Levine, W. C. Topp, and J. D. Watson, Eds.), Vol. 2, pp. 363-368. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. DEAN, F. B., BULLOCK, P., MURAKAMI, Y., WOBBE, C. R., WEISSBACH, L., and HURWITZ, J. (1987). Simian virus 40 (SV40) DNA replication: SV40 large T antigen unwinds DNA containing the SV40 origin of replication. Proc. Nat/. Acad. Sci. USA 84, 16-20. DEB, S., DELUCIA, A. L., BAUR, C. P., KOFF, A., and TEGTMEYER, P. (1986). Domain structure of the SV40 core origin of replication. Mol. Cell. Biol. 6, 1663-1670. DEB, S., SUI. S., KOFF, A., DELUCIA, A. L., PARSONS, R., and TEGTMEYER, P. (1987). The T-antigen-binding domain of the simian virus 40 core origin of replication. J. Viral. 61, 2143-2149. DIMAIO, D., and NATHANS, D. (1980). Cold sensitive regulatory mutants of SV40.J. Mol. Biol. 140, 129-l 42. DIMAIO, D., and NATHANS, D. (1982). Regulatory mutants of simian virus 40. Effect of mutations at a T antigen binding site on DNA replication and expression of viral genes. J. Mol. Biol. 156, 531-548. EDWARDS, C. A. F., KHOURY, G., and MARTIN, R. G. (1979). Phosphorylation of T-antigen and control of T-antigen expression in cells transformed by wild-type and tsA-mutants of simian virus 40. J. Viral. 29, 753-762. FANNING, E., NOWAK, B., and BURGER, C. (1981). Detection and characterization of multiple forms of simian virus 40 large T antigen. J. Viral. 37, 92-102. FANNING, E., WESTPHAL, K.-H., BRAUER, D., and C~RLIN, D. (1982). Subclasses of simian virus 40 large T antigen differential binding of two subclasses of T antigen from productively infected cells to viral and cellular DNA. EMBO J. 1, 1023-l 028. GIACHERIO, D., and HAGER, L. P. (1979). A poly(dT) stimulated ATPase activity associated with SV40 large T antigen. J. Biol. Chem. 254, 8113-8116. GURNEY, E. G., TAMOWSKI, S., and DEPPERT, W. (1986). Antigenic binding sites of monoclonal antibodies specific for simian virus 40 large T antigen. J. Viral. 57, 1168-l 172. HANSEN, U., TENEN. D., LIVINGSTON, D. M., and SHARP, P. A. (1981). T antigen repression of SV40 early transcription from two promotars. Cell 27, 603-612. HINZPETER, M., and DEPPERT, W. (1987). Analysis of biological and biochemical parameters for chromatin and nuclear matrix association of SV40 large T antigen in transformed cells. Oncogene 1, 119-129. HINZPETER, M., FANNING, E., and DEPPERT, W. (1986). A new sensitive target-bound DNA binding assay for SV40 T antigen. Virology 148,159-167. HIRT, B. (1967). Selective extraction of polyoma DNA from infected mouse cell cultures. J. Mol. Biol. 26, 365-369. JONES, K. A., MYERS, R. M., and TJIAN, R. (1984). Mutational analysis of SV40 large T antigen DNA binding sites. EMBO J. 3, 3247-3255. KELLER, J. M., and ALWINE, J. C. (1985). Analysis of an activatable promoter: Sequence in the SV40 late promoter required for T antigen mediated transactivation. Mol. Cell. Biol. 5, 1859-l 869. KESSLER, S. W. (1975). Rapid isolation of antigens from cells with a staphylococcal protein A-antibody adsorbent: Parameters of the
SV40
T-ANTIGEN
BINDING
interaction of antibody-antigen complexes with protein A. J. Immunoi. 115, 1617-1624. LANE, D. P., and CRAWFORD, L. V. (1979). T antigen is bound to a host protein in SV40-transformed cells. Nature (London) 278, 261-263. MANIAIIS, T., FRITSCH, E. F., and SAMBROOK, J. (1982). “Molecular Cloning: A Laboratory Manual.” Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. MANOS, M., and GLUZMAN, Y. (1984). Simian virus 40 large T antigen point mutants that are defective in viral DNA replication but competent in oncogenic transformation. Mol. Cell. Biol. 4, 1 125-l 133. MARGOLSKEE, R., and NATHANS, D. (1984). SV40 mutant T antigens with relaxed specificity for the nucleotide sequence at the viral DNA origin of replication. 1. Viral. 49, 385-393. MONTENARH, M., and HENNING, R. (1983). Self assembly of SV40 large T antigen oligomers by divalent cations. J. Viral. 45, 531-538. MONTENARH, M., KOHLER, M., and HENNING, R. (1984). The oligomerization of simian virus 40 large T antigen is not necessarily effected by temperature sensitive A gene lesions. J. Viral. 49, 658-664. MONTENARH, M., and Mali-LER, D. (1987). The phosphotylation at Thr 124 of simian virus 40 large T antigen is crucial for its oligomerization. FEBS Letl. 221, 199-204. MONTENARH, M., VESCO, C., KEMMERLING, G., MOLLER, D., and HENNING, R. (1986). Regions of SV40 large T antigen necessary for oligomerization and complex formation with the cellular oncoprotein ~53. FEBS Lett. 204, 51-55. MOLLER, D., UGI, I., BALLAS, K.; REISER, P., HENNING, R., and MONTENARH, M. (1987). The AT-rich sequence of the SV40 control region influences the binding of SV40 T antigen to binding sites II and III. I/iro/ogy 161, 81-91. MYERS, R. M., RIO, D. C., ROBBINS, A. K., ~~~TJIAN, R. (1981). SV40 gene expression is modulated by the cooperative binding of T antigen to DNA. Cell 25, 373-384. REED, S. Jo, STARK, 6. R., and ALWINE, J. C. (1976). Autoregulation of simian virus 40 gene A by T antigen. Proc. Nat/. Acad. Sci. USA 73, 3083-308-Y. RIGEY, P., and ~.ANE, D. P. (1983). Structure and function of simian virus 40 large T antigen. in “Advances in Viral Oncology” (G. Klein, Ed.), Vol. 3., pp. 31-57. Raven Press, New York. RIO, D., ROBBINS, A., MYERS, R., and TJIAN, R. (1980). Regulation of SV40 early transcription in vitro by a purified tumor antigen. hoc. Nat/. Acad. Sci. USA 77, 5706-5710. RIO, D. C., and TJIAN, R. (1983). SV40 T antigen binding site mutations that affect autoregulation. Cell 32, 1227-l 240.
TO SITES
I AND
II
317
RUNZLER, R., THOMPSON, S., and FANNING, E. (1987). Oiigomerization and origin DNA-binding activity of simian virus 40 iarge T antigen. J. Viral 61, 2076-2083. RYDER, K., VAKALOPOULOU, E., MERTZ, R., MASTRANGELO, J., HOUGH, P., TEGTMEYER, P., and FANNING, E. (1985). Seventeen base pairs of region I encode a novel tripartite binding signal for SV40 T antigen. Cell 42, 539-548. SCHEIDTMANN, K.-H., HARDUNG, M., ECHLE, B., and WALTER, G. (1984). DNA binding activity of SV40 large T antigen correlates with a distinct phosphorylation state. J. Viral. 50, l-l 2. SCHELLER, A., COVEY, L., BARNET, B., and PRIVES, C. (1982). A small subclass of SV40 T antigen binds to the viral origin of replication. Cell 29, 375-383. SCH~~RMANN, C., MONTENARH, M., KOHLER, M., and HENNING, R. (1985). Oligomerization of simian virus 411 tumor antigen may be involved in viral DNA replication. Viroiogy 14& 1-l 1. SHALLOWAY, D., KLEINBERGER, T., and LIVINGSTON, D. M. (1980). Mapping of SV40 DNA replication origin region binding sites for the SV40 T antigen by protection against exonuciease III digestion. cell 20, 41 l-422. SHORTLE, D., MARGOLSKEE, R.; and NATCIANS, D. (1979). Mutational analysis of the SV40 replicon: Pseudorevertants of mutants with a defective replication origin. Proc. Nati. Acad. SC;. USA 76, 6128-6131. SMALE, S. T., and TJIAN, R. (1986). T-antigen-DNA polymerase CY complex implicated in simian virus 40 DNA replication. P/o/. Cell. Biol. 6, 4077-4087. STAHL, H., DR~GE, P., and KNIPPERS, R. (1986). DNA helicase activity of SV40 large tumor antigen. En/lBO J. 5, 1939-l 944. TEGTMEYER, P., LEW~ON, B. A., DELUCIA, A. L., WILSON, V. G., and RYDER, K. (1983). Topography of SV40 A protein-DNA compiexes: Arrangement of protein bound to the origin of replication. J. Virol. 46, 151-161. TEGTMEYER, P., SCHWARZ, M., COLLINS, J. K., and RUNDELL, K. R. (1975). Regulation of tumor antigen synthesis by simian virus 40 gene A. 1. Viral. 16, 168-178. TJIAN. R., and RO~BINS, A. (1979). Enzymatic activities associated with a purified simian virus 40 T-antigen-related protein. Proc. Nat/. Acad. Sci. USA 76, 61 O-61 5. TOOZE, J. (1981). “The Molecular Biology of Tumor Viruses,” 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. WACHTER, M., RIEDLE, G., and HENNING, R. (1985). Functional implications of oligomerization of simian virus 40 large T antigen during lytic virus infection. 1. Viral. 56, 520-526.