Volume 8 Number 19 1980

Nucleic Acids Research

Initiation of polyoma virus DNA replication in vitro and its dependence on the viral gene A protein

Philippe Clertant and

Franjois Cuzin

Centre de Biochimie du C.N.R.S., Universite de Nice, Parc Valrose, 06034 Nice, France

Received 1 August 1980

AfTfACI

Initiation of polyana virus DNA replication is dependent on the activity of the early protein affected by the tsa mutations (large-T antigen). An in vitro DtA synthesizing system blocked at the initiation stage was designed by preparing nuclei fram cells shifted to high temperature after infection with a polyana tsa mutant. Addition to these nuclei of extracts fran wild type virus-infected cells resulted in a limited, but reproducible stimulation of decoynucleoside onnphosphate incorporation. At least for a significant part, this stimulation was shown to correspond to an increased synthesis of molecules identified as polyana replicative intermediates by their sedimentation coefficient and endonuclease Hpa II cleavage pattern. The ncn-random distribution of label ctserved among restriction fragments was that expected fran an initiation event occuring at the physiological origin. This activity was reduced to backgrond level in extracts fran tsa-infected cells shifted to high temperature and was specifically inhibited by addition of Fab fragments fran anti-polyana virus T antigen iuunoglcbulins. MUCTION Initiation of DNA synthesis is now being well &cumented in Bacteria (review in ref 1), but little is known of the initiation mechanism in Eukaryotes. Mach like the small DNA bactericphages, viruses like polyama and SV 40 may constitute useful models for such studies. As for other replicons (2), studies on the replication of polyana virus DNA in cells infected with temperature-sensitive ( ts ) mutants have demonstrated a requirement for expression of an early viral function in order to initiate new rounds of replication (3). The initiation ts mutants belong to a single canplementation group ( a gene ) (4-6). The tsa mutations map in the distal part of the early region (7,8), which codes for d polypeptide of approximate molecular weight 100,000, the large-T antigen recognized by antibodies fran tumour-bearing animals (912). This function appears to be the only one of viral genetic information which is required for the replication of viral DNA. As a first step towards a biochemical analysis of the initiation of polyoma DNA replication, a camplementation assay was designed, similar to iL

©) IRL Press Limited, 1 Falconberg Court, London W1V 5FG, U.K.

4377

Nucleic Acids Research those previously used for replication studies in Prokaryotes (13). The elongation, and to sone extent, the terrninaticn steps in polycaa virus DNA replication were shown by P. Reichard and his co-wrkers (14-16) to occur in a nuclear cell-free system. We isolated by the sane procedure nuclei fran tsa polyana-infected cells which had been shifted to the restrictive teriperature. Under these conditions, viral DNA synthesis in vivo was arrested, after conversion of the replicative intenediates into mature form I MA molecules (3). Extracts fran cells lytically infected with wild type virus were added to these nuclei, depleted of the a gene product initiating activity, but presuned to contain all other factors necessary for viral DN replication. It resulted in a reproducible stimulation of 32p incorporation fran [a-32p] deoxyribonucleoside triphosphates. In order to conclude that initiation events were actually occuring in vitro and that the early protein product of polyana a gene was required, the following points were studied: (i) production of labeled polyana replicative interediates (17), (ii) distribution of the radioactive label as a gradient originating at the physiological origin of replication (18), (iii) absence of this activity in similar extracts prepared fran nuck-infected cells, (iv) dependence on the a gene product, using extracts fran tsa-infected cells shifted to the restrictive terperature, (v) inhibition of the activity by anti-T antigen antibodies. NATERIAL ANED ETHC)S Cells and viruses. Nbuse 3T6 fibroblast cells (19) were routinely grown in Dulbeo nodified Eagle's mediurn (GIBOD) supplennted with 5 to 10% newborn calf serzn (GIBO0). Polyana wild type A2 strain (18) and tsa25 (6) viruses were grown at 330C in secondary nouse enrbryo cell cultures infected at low

mrultiplicity (0.05 PFU/cell). Irunological techniques. T antigen was detected by inmunofluorescence and cmplement fixation, as well as by inymmoprecipitation and subsequent gel electrophoresis, as previously described (20, 21). For the preparation of IgG inrnoglobulin Fab fragrents, hamster serun with a complerrent fixation titer of 50 to 100 (20) was pre-adsorbed on 3T6 cell layers fixed on tissue culture plates with nethanol:acetone (7:3) (1 ml serun107 cells, repeated 5 times). After anronium sulfate precipitaticn at 40% saturation, the insoluble glcbulin fraction was redissolved in 20 mM potassium phosphate buffer at pH 6.2 and passed on a DEAE-cellulose column equilibrated with the same buffer. The flowthrough fraction contained only IgG immunoglobulins, as checked on the basis of apparent mlecular weights estimated by gel filtration and sodium cbdecyl-

4378

Nucleic Acids Research sulfate (SDS) polyacrylamide gel electropioresis. It was concentrated by vacum dialysis against 0.1 M potassium phosphate buffer (pH 7.5), 3 0rM EiYTA, up to a final cxcentration of 10 /rg/m of protein. Papain (Sigma) was added (0.15 mg//ml), and dithioireitol (10 rt4. After 3 h incubation at 370C, the reaction was stoped by addition of 1/10 volume of 0.2 M iodoacetamide in 0.1 M phosphate buffer at pH 7.5. The soluticn was dialyzed against 0.1 M Tris-HCl pH 8, 1 M NaCl and centrifuged (10,000 x g, 5 min) . Fab and Fc fragnts were separated by gel filtration on Sephadex G 75 (Phannacia) and diluted to a final salt concentration of 0.2 M NaCl. Fc fragments were renoved by batch adsorption on protein A-Sepharose (Phamacia, 0.1 ml of a 1:1 suspension in water per rg protein). The supernatant was concentrated by vacuum dialysis against Isotonic Hepes Buffer (14), up to a conmcentration of 1 rig/ml protein, corresponding to the initial volume of serun. Isolation of nuclei frm cells infected with polycua virus. 3T6 cells were 42 gram at a density of ca. 5 x 10 cells/cma and infected with tsa25 polyom virus at a nultiplicity of 10 PFU/cell. After 48 h at 32°C, 90% or mre of the cells were staied by innunoflixrescence for T antigen, and 20 to 40%, for V antigen, whereas mximal rates of viral DM synthesis were observed, both in vivo and in isolated nuclei (Figure 1). Cells were labeled by addition to the inediun of [3H] thymidine (Comissariat a 1'Energie Atcrnique, France, final aoncentraticn: 2 wCi/r1). Thirty minutes later, cultures were transfered to 39.50C, and 2 h later, cells were washed and lyzed in Isotonic Hepes Buffer containing 0.5% Nonidet P40 as described in ref. 14. Cell lysis was mmritored by phase acntrast microscopy and nuclei harvested by centrifugation (800 x g, 10 min) and resuspended in an equal volure of Isotonic Hepes Buffer (final ENA concentration ca. 2 mg/ml). These preparations could be kept at -1960C for at least 6 mnths without loss of activity. In vitro incubation of the nuclei. Nuclei (10 wl of suspension) were preincubated for 15 min at 0°C in a total volume of 50 wl in the presence of 40 ruM NaCl, 6 iM MgCl2, 25 uM EXFA, 2 rM ATP, 0.1 mM each of dATP, dCIP and dIPE, 0.004 NrM [a-32P] dGTp (20-200 Ci/nmole, Amersham), 4 muM phosphoenolpyruvate, pyruvate kinase (20 wig/ml) in Isotonic Hepes Buffer with cell extracts added as indicated. The reaction was started by transfer to 260C, and arrested by addition of SD6 and EDLA for selective extraction acaording to Hirt (22). All neasurenents and further analysis were performed on the supernatants containing the viral ENA. As shown in Figure 1, nuclear preparatios fran tsa25infected cells shifted to the restrictive tenperature exhibited only backgroud levels of incorporation into this fraction, in parallel with the low 4379

Nucleic Acids Research rate of incorporation of [3H] thymidire in vivo. Purification of low molecular weight DE fran nuclei. The supernatants fran the Hirt selective extraction procedure (22) were treated with proteinase K (50 wg/ml, t4rck) for 4 h at 37°C, extracted twice with purified phenol, dialyzed against TE buffer (0.01 M Tris-HCl pH 7.5, 2 mrM EDIA), treated with pancreatic RNase (Nirck, 0.1 nig/ml) for 1 h at 37°C and extensively dialyzed against TE buffer cntaining 0.25 M NaCl. Chranatography on benzoylatednaphtcylated DEAE cellulose (BND cellulose) was used for the purification of polycana DNA replicative interre-diates as described in ref. 14. Neutral sucrose gradient centrifugation. Centrifugation through 5 ml preforred sucrose gradients (5-20% in TE buffer containing 1M NaCl) was for 165 mmcn at 48,000 rpn (200C) in the Spinco rotor SW 50-1. Fraction aliquots were counted on GF/C filters (Whatnan) after precipitation with 5% trichloroacetic acid. Hpa II restriction endonuclease digestion and agarose gel electrcoresis. [DN was precipitated with ethanol and redissolved in 10 IrM Tris-HCl buffer pH 7.5, 8 irM MgC22, 6 ntM KC1, 1 nM DIT, 50 wg/ml gelatin. 10 units of Hpa II restriction endunuclease (Boehringer) were added. After overnight incubation at 37°C, reaction was stopped by addition of EDTA to a final concentratiorr of 20 iuM. Proteins were renxved by extraction with chloroform and DN precipitated with ethanol, redissolved in 50 pl of E buffer (23) containing 20% sucrose. Electrophoresis was for 15 h at 2 V/cm on vertical slab gels of 1.4% agarose 15 an in length. Gels were stained after electrophoresis by inuersing them in E buffer containing 1 wg/ml ethidium bronide for 30 min, and photographed using a short wavelength ultraviolet lamp and a Polaroid camera. They were dried at 37°C under vacuum on several sheets of Thatmnan 3M paper and subaoitted to autoradiography at -700C on preflashed X-Onat Kodak filnm with intensifying screens (Dupont) (24) for 8 days. Fbr radioactivity counting, gel slices were boiled for 10 min in 1 ml H20, centrifuged, and the solution was counted in 10 ml of Picofluor solution (Packard). Independent measurements of band radloactivities were obtaired by scanning the autoradiograms using a Vernon densitometer. The later proved to be a mre reliable measurement of low levels of 32P radioactivity than scintillation counting. Prration of extracts. 3T6 cells were infected with either wild type or tsa25 polyona. After either 48 h at 320C, or a subsequent shift to 39.50C for 2 h, cells were lyzed in Hypotonic Hepes Buffer as described in ref. 14. The supernatant obtained after low speed centrifugation contained at least 75% of the T antigen camplement fixing activity. It was fractionated by amrmoium sulfate precipitation between 25 and 40% saturation. The final pellet was dissolved in Isotonic Hepes Buffer (1 ml per 108 cells), dialyzed against 4 4380

Nucleic Acids Research changes (100 vol) of the saTr buffer containing 20% glycerol and centrifuged for 10 min (40C, 10,000 x g). Protein concentration was 20 mg/ml and titers for T antigen varied between 10 and 30 ccrplarent fixing units (recovery between 50 and 100%). RESULTS Stinulation of lcmolecular weight DNA synthesis in initiation-deficient nuclei upon addition of extracts fran wild type polyara-infected cells. Nuclei were prepared fran tsa25-infected cells incubated at 32°C for

0I 0 b0n

I

I 0

a

z a

z

I

E

a

0. -

z

I

C00. m

0) cn

E

0)

\8*

E cL

I

Figure 1. Time-course of DNA synthesis in tsa25 infected cells and in nuclear preparations. A. In vivo thymidine incorporation: at the indicated times after infection at 250C, cultures were labeled for 1 h (5 iCi [3H] dT/ml) and the acid-insoluble radioactivity was measured on Hirt supernatants. T antigen ( 0 ) and V antigen-positive ( 0 ) cells were counted after immunofluorescence staining. B. In vitro dGMP incorporation in isolated nuclei: acid-insoluble radioactivity was measured after either 5 min ( 0 : Hirt supernatant; 0 : pellet) or 30 min ( C0: Hirt supernatant) incubation. C. Decay of viral DNA synthesis after a shift to 39.5°C of infected cell cultures (50 h after infection): [3H] dT ( 0 : Hirt in vivo incorporation during a 30 min pulse (5 ijCi/ml) supernatant; 0 : pellet) and [a-3 2p] dGMP in vitro incorporation during a 30 min incubation of the nuclei ( V : Hirt supernatant).

4381

Nucleic Acids Research 50 h, a timre corresponding to the maximal rate of both in vivo and in vitro viral DNA replication. Shifting these cells to 39.5°C resulted, in parallel with the in vivo situation, in a drastic decrease of the rate of dGMIP incorporation into low molecular weight DNA in the in vitro system (Figure 1). Incubation of extracts from wild type polycna-infected celIs under the same conditions, but in the absence of nuclei, resulted in a low incorporation into acid-insoluble material, which varied fran one extract to another. This endogenous activity was not due to incorporaticn in viral DNA (see Figure 6, lanes 1 and 5). It was always nmonitored in parallel incubations and its value was substracted fran the incorporation of the carplete system. As shown in Figure 2, addition to tsa25-infected nuclei of extracts from wild type polyoma-infected cells led to a reproucible stinulation of dGMP incorporation. No stimlation was observed after addition of equivalent amounts of protein prepared from mock-infected 3T6 cells.

Characterization of the DNA labeled in vitro with and without addition of extract. The nature of the DNA radioactively labeled in vitro was first analyzed by zonal ultracentrifugation on sucrose gradients. As shown in Figure 3, in

Figure 2. Stimulation of low molecuZar weight DNA

4

z

04

synthesis after addition of cell extracts to initiationdeficient nuclei. Concentration of total DNA in the assays was 0.4 mg/ml. Cell extracts were added at a ffinal concentration of 8

E 3

mg/ml:

o 0

EU
18 S

8S

7

= 0.6 E

6

0.5

5

c

o

o 0.4

*

0. o u

labeled DNA molecules with sedimentation coefficients corresponding to that of viral DNA ( ., 18 S) and in shorter molecules. Total radioactivity in fractions of zonal ultracentrifugation runs similar to that in Figure 3, after either 20 or 30 min incubation of the nuclear preparation either in the

~~~~~~~presence (U*) or in the absence

|

( I2! ) of extracts from virus-infected cells.

' 0.3

0.

0.1

I

0

E

20 20 40 40 incubation time ( min )

observed increase in dGMP incorporaticn. These rolecules were shown in inciependent experiments to be eluted from BND cellulose only in the presence of 2% caffein, as previously reported for the viral replicative interediates

(17) . These results indicate that synthesis dependent on the addition of the extract occurs primarily on templates with properties sinmilar to those of the viral replicative intermediates. Their positive identification as replicating viral DNA molecules was deduced from their restricticn cleavage map. Hpa II endonuclease cleavage of in vitro labeled DNA molecules. Fractions corresponding to sedimntation coefficients between 18 and 28 S (Figure 3) were pooled, and unlabeled polycara form I DNA was added as a carrier. After cleavage with Hpa II endonuclease, agarose gel electrophoresis and ethidium bronmide staining (Figure 6A), the known pattern of fragments (18, see Figure 5) was evidenced from the carrier polyana DNA. Fragments 5 and 6 were not resolved and 7 and 8, not detectable due to their small sizes. Autoradiographic exposure of the gels revealed the presence of 32p label in all detectable polyana fragments (bands 1 to 5 + 6 on Figure 6B) and also in additional discrete bands designated A to E. Sizes of the later fragments closely correspond to those of the Hpa II cleavage products fron mouse mito-

4384

Nucleic Acids Research Figure 5. Hpa II endonuclease

cleavage map of polyoma DNA. /B