Interrelation between Viral and Cellular DNA Synthesis in ... - CiteSeerX

J. gen. Virol. (1983), 64, 1991-1998. Printed in Great Britain

1991

Key words: MVM/parvovirus/DNA synthesis/aphidicolin

Interrelation between Viral and Cellular DNA Synthesis in Mouse Cells Infected with the Parvovirus Minute Virus of Mice By N. H A R D T , 1 C. D I N S A R T , z S. S P A D A R I , 2 G. P E D R A L I - N O Y 2 AND J. R O M M E L A E R E ~* 1Laboratoire de Biophysique et Radiobiologie, Universit~ libre de Bruxelles, rue des Chevaux, 67, B-1640 Rhode-St Genkse, Belgium and Zlstituto di Genetica Biochimica ed Evoluzionistica, Consiglio Nazionale delle Ricerche, Via S. Epifanio, 14, 1-27100 Pavia, Italy (Accepted 25 May 1983) SUMMARY

Mouse fibroblasts arrested in Go by isoleucine deprivation were inoculated with the autonomous parvovirus minute virus of mice (MVM). Infected cells were released from the Go block by transfer to complete medium and their progression to and through the S phase was monitored. The onset of viral and cellular DNA synthesis coincided, suggesting that cellular factor(s) required for MVM DNA replication became available as soon as cells entered the S phase. Cellular DNA synthesis was reduced to about 6 0 ~ by MVM infection. However, this inhibition did not decrease significantly the overall rate of DNA replication in infected cells because it was compensated by concomitant viral DNA synthesis. MVM infection delayed the movement of the cells out of S phase by at least 5 h. At any time post-infection, more than 9 5 ~ of both viral and cellular DNA synthesis was sensitive to inhibition by aphidicolin. Since this drug is highly specific for cellular DNA polymerase ~, the data are consistent with a major role of this enzyme in the in vivo DNA replication of autonomous parvovirus. The assembly of 95 ~ of virus progeny particles was concomitant with a late phase of viral DNA replication which accounted for 30 ~ of the total viral DNA synthesized. The inhibition of this residual viral DNA replication by aphidicolin reduced dramatically the size of the burst of infectious particles; this observation concurs with other evidence to suggest that encapsidation is driven by a late replication event sensitive to this drug. INTRODUCTION Parvoviruses are among the smallest known animal viruses and contain a predominantly single-stranded, linear DNA genome (Tattersall & Ward, 1978). The replication of parvovirus DNA takes place in the cell nucleus and involves three steps: (i) the conversion of singlestranded DNA to double-stranded replicative forms, (ii) the replication of double-stranded DNA and (iii) the asymmetric production of progeny single-stranded DNA from replicative forms (Berns & Hauswirth, 1978). Parvoviruses of the non-defective subgroup are able to replicate autonomously provided that S phase-specific host cell function(s) is (are) available (Tattersall & Ward, 1978). The S phase-dependent step of the parvovirus life cycle is likely to be viral DNA replication, in particular the conversion of the input single-stranded genome to a duplex form (Wolter et al., 1980). The cellular 'S factors' essential for parvovirus replication have not been identified so far and the time of their appearance during the S phase is disputed (Parris & Bates, 1976; Wolter et al., 1980). Productive infection by the non-defective parvoviruses does not require a virion-associated DNA polymerase (Pritchard et al., 1978a) and is likely to rely on cellular DNA polymerases. Although both DNA polymerases ~ and ~ have been shown to catalyse parvovirus DNA replication in vitro (Handa & Carter, 1979; Pritchard et al., 1981; Kollek & Goulian, 1981; Kollek et al., 1982; Faust & Rankin, 1982), the contribution of these enzymes to viral DNA synthesis in vivo is unclear. The involvement of DNA polymerase ~ is supported by indirect evidence showing a temporal correlation between this enzymic activity 0022-1317/83/0000-5661 $02.00 © 1983 SGM

1992

N.

HARDT

AND

OTHERS

and parvovirus D N A synthesis in infected cells (Pritchard et al., 1978 b). Studies of the effect of parvovirus replication on the progression of cellular D N A replication have given contradictory results. Some authors have reported that parvovirus infection causes an early reduction in the rate of D N A synthesis (Salzman et al., 1972) whereas others have found no alteration in the progression of infected cells through S phase (Hampton, 1970; Parris & Bates, 1976). These uncertainties concerning the interrelation between cellular and parvovirus D N A replication prompted us to investigate the time course of these processes in synchronized mouse cells following infection with the autonomous parvovirus minute virus of mice (MVM) (Ward & Tattersall, 1982). The data presented suggest that MVM D N A replication (i) depends on cellular function(s) expressed early in S phase, (ii) is accompanied by an inhibition of cellular D N A synthesis, and (iii) is inhibited by aphidicolin, a specific inhibitor of cellular D N A polymerase (Ikegami et al., 1978; Pedrali-Noy & Spadari, 1979), to a similar extent as cellular D N A replication. Moreover, a late, aphidicolin-sensitive event in MVM D N A replication seems to drive the assembly of infectious progeny particles.

METHODS Cell culture and viral infection. Mouse A9 cells (Littlefield, 1964) were accumulated in G O by isoleucine deprivation as described previously (Rommelaere et al., 1981). The arrested monolayer culture (4 × 105 cells) was infected with the plaque-purified strain T of MVM (Tattersall, 1972) at 4 p.f,u./cell. Immediately after infection, cells were transferred to Eagle's minimal essential medium supplemented with 5 ~ foetal calf serum. Autoradiography. At intervals following MVM or mock infection, cells attached to glass slides were pulselabelled with [3H]thymidine (TdR) (5 ~tCi/ml, 5 Ci/mmol) for 45 min and further incubated for 10 min in phosphate-buffered saline (PBS) containing 1 mg/ml of unlabelled TdR. Cultures were then rinsed, fixed and prepared for autoradiography, as described previously (Hardt et al., 1980). Virus titration. At intervals post-infection, cells were collected and the virus was extracted by freeze-thawing (Tattersall et al., 1976). Extracts were assayed for infectivity by direct plaque assay (Tattersall, 1972). In some experiments, infected cells were exposed to 15 ktm-aphidicolin from diffferent times until 36 h post-infection, at which time the infectivity of cell-associated viral particles was measured. DNA analysis Radioactive labelling. At intervals after MVM or mock infection, cells were pulse-labelled with [3H]TdR (20 pCi/ml, 25 Ci/mmol) for 45 min. Cultures were then incubated for 7 min with PBS containing l mg/ml of unlabelled thymidine. After several rinses, cells were collected and suspended in PBS (2 x 105 cells/ml). When present, aphidicolin was used at 15 ~tM during both the pulse and chase periods. Measurement o f the overall rate o f DNA synthesis. A 0.1 ml amount of the cell suspension was collected on a Whatman GF/C filter. The filter was washed twice in 5~o trichloroacetic acid (TCA) and once in ethanol. When dry, the filter was dipped in Econofluor scintillator (New England Nuclear) and the radioactivity was measured in a liquid scintillation spectrometer. It was verified that the rate of incorporation of[3H]TdR in both uninfected and MVM-infected cells was not changed by increasing the concentration of thymidine in the external medium from 0.8 ktM (our standard radioactive labelling conditions) to 5 pM. Therefore, the incorporation of [3H]TdR is not likely to be affected by endogenous deoxyribonucleotide pools (Cleaver & Holford, 1965) and should provide a reliable measurement of DNA synthesis. Alkaline sucrose sedimentation analysis. The cell suspension was exposed to 20 Gy of y-radiation delivered from a Gamma cell 200 (6°Co). A 4 ml gradient of 5 to 20~ sucrose in 0.1 M-NaC1, 0.1 M-NaOH was formed on top of a 1 ml cushion of 50~ sucrose. Cell suspensions were made 0.2 Min NaOH, incubated for 15 rain at room temperature and layered (0-2 ml) on top of the gradients. These were centrifuged immediately at 4 °C, in the SW50.1 rotor in a Beckman model L ultracentrifuge at 48000 rev/min for 4 h. After centrifugation, 24 to 26 fractions were collected and their radioactivity was measured by liquid scintillation spectrometry using Aquasol-2 fluid (New England Nuclear). DNA purified from MVM virions and labelled with [3H]TdR (Bourguignon et al., 1976) was used as a marker and centrifuged in parallel gradients. Determination ofL,iralDNA by D N A - D N A reassociation. Samples of labelled cells were suspended in 0.02 M-Tris HC1, 0-15 M-NaCI, 0.01 M-EDTA, pH 7-5 (2 × l07 cells/ml) and lysed with SDS (0.6%) and self-digested Pronase (2 mg/ml) for 1 h at 37 °C. Extracted DNA was sheared to a length of 500 to 600 nucleotides by heating the lysates at 100 °C in the presence of 0.3 M-NaOH for 15 min~ After cooling to 0 °C, samples were neutralized with K H , P O 4 to a final pH 7.3 to 7.5 and dialysed overnight against 5 mM-Tris-HC1, 0.5 M-NaCI, 1 mM-EDTA, pH 7.7. The resultant DNA solutions were used for reassociation experiments as described by Pedrali-Noy & Weissbach (1977) except that an appropriate excess of both strands of non-radioactive viral DNA was added to each sample to drive

M V M D N A synthesis

1993

the reassociationreaction. This DNA was prepared by copyingwith Escherichia cob DNA polymerase I (Klenow fragment) self-primed single-stranded MVM DNA purified from viral particles, essentially as described by Bourguignon et al. (1976). Chemicals. [Me-3H]TdR was from Amersham International. Aphidicolin was kindly supplied by Dr A. Todd, Imperial Chemical Industries, Macclesfield, Cheshire, U.K. RESULTS Pattern of overall D N A synthesis in MVM-infected cells Mouse cells were accumulated in Go by incubation in isoleucine-depleted medium. Arrested cultures were transferred to complete medium and pulse-labelled with [3H]TdR at intervals. D N A synthesis was measured by determining the radioactivity incorporated into TCAinsoluble material and by cell autoradiography. Cultures released from the Go block resumed the cell cycle in a synchronous fashion. Fig. 1 (solid lines) illustrates the time course of cellular D N A synthesis. The traverse of S phase is indicated by a sharp and transient increase of both the fraction of labelled cells (Fig. 1a) and the incorporated radioactivity (Fig. 1 b). Entry into S phase and maximum labelling occurred about 10 and 18 h following transfer to complete medium, respectively. G0-arrested cultures were infected with MVM at a multiplicity of 4 infectious units/cell, immediately prior to transfer to complete medium. MVM infection did not disturb significantly the entry of the cells into S phase (Fig. 1 a). In contrast, the labelling index increased for 5 h more in MVM- than in mock-infected cultures, indicating that the movement of MVM-inoculated cells out of the phase of D N A synthesis was delayed (Fig. 1a). As illustrated by Fig. 1 (b), MVM infection caused little reduction in the maximum rate of D N A synthesis but slowed its decline, resulting in a 1.2-fold increase of the total amount of labelled D N A precursors incorporated within 35 h post-infection. Time course of viral and cellular D N A replication Incorporation of [3H]TdR into cellular and viral D N A was measured in cells pulse-labelled at intervals after Go release and MVM infection. The D N A isolated from infected cells was fractionated by sedimentation through alkaline sucrose gradients; it generated bimodal profiles (Fig. 2a). Molecules of high and low sedimentation rates co-migrated with D N A from uninfected cells and with purified viral DNA, respectively. The slow-sedimenting peak was used for the quantification of viral D N A (Fig. 2a). The fraction of virus-specific tritiated D N A was also determined by D N A - D N A reassociation in the presence of an excess of non-radioactive viral D N A (Table 1). Although Fig. 1 (b) shows that the overall rate of D N A synthesis in infected cells closely approximated that of uninfected cells, Fig. 2 (b) clearly shows that cellular D N A replication was markedly depressed after MVM infection. However, this inhibition, presumably due to some event(s) in MVM replication, was compensated for by a concomitant increase in viral D N A synthesis. The extent of the inhibition of cellular D N A synthesis occurring within 35 h after infection varied from 20 to 6 0 ~ in independent experiments. Correspondingly, viral D N A replication accounted for a fraction, which ranged from 25 to 65 %oof total incorporated radioactivity. The reason for this fluctuation in the level of viral D N A synthesis is not known. The time courses of viral and cellular D N A synthesis in infected cultures overlapped to a great extent (Fig. 2b; Table 1). The onset of replication of both types of D N A could not be dissociated and took place around 10 h after infection and release from Go. Infected cells achieved the highest rate of viral D N A replication about 5 h after residual cellular D N A synthesis started to decline. Effect of aphidicolin on viral D N A replication and encapsidation Viral replication was tested for its sensitivity to aphidicolin. The drug was used at a concentration, 15 gM, that inhibits over 95~o of cellular D N A replication (Pedrali-Noy & Spadari, 1980) with no effect on R N A and protein synthesis (Pedrali-Noy et al., 1982). Consistently, aphidicolin reduced [3H]TdR incorporation into cellular D N A of both mock- and MVM-

1994

N.

HARDT

AND

OTHERS I

(a)

I

I

i t

I

I

40-

I

9, tl

75

(a)

"~-"A,

!?% o~

",,

/

50

~ 20-

i

II III

!

/o._

e~

25

I

"._.

6

',,

I

i ~0,.

0

//

1

10 15 Fraction number

5

#

(b)

(b) 15

15

"~ l0

lO

?

/

I

20

I

I

X

d

5

-

, iII i

0

1t

•

I I

10 20 30 Time post-infection (h) Fig. 1

•

D .....

~

A,.

. ? , '

•

I0

..

•1

a

o

|1

•

20 30 Time post-infection (h) Fig. 2

Fig. 1. Effect of MVM infection on overall DNA synthesis in synchronized A9 ceils. Cultures were pulse-labelled with [3H]TdR at intervals following release from G Oblock and MVM infection. (a) Percentage of labelled cells determined by autoradiography. At least 3000 cells were examined for each point. Similar results were obtained with autoradiograms exposed from 6 h up to 20 days. (b) Radioactivity incorporated into TCA-insoluble material. Cells were pulse-labelled in the absence (O, A) or presence ( 0 , A) of aphidicolin. O, O, Uninfected cells; A, A , MVM-infected cells. Fig. 2. Time course and aphidicolin sensitivity of cellular and viral DNA synthesis in MVM-infected A9 cells. A9 cells were pulse-labelled with [3H]TdR at intervals after MVM or mock infection and release from Go block. (a) DNA fractionation by alkaline sucrose gradient sedimentation. Lysates of MVM-infected (O) and mock-infected (0) cells harvested 24 h after infection were run in parallel gradients. The arrow indicates the position of purified MVM virion DNA. For the purpose of quantification, DNA within the region of the gradient defined by the bar was considered to be of viral origin. (b) Rates of viral and cellular DNA synthesis. Cultures were pulse-labelled in the absence (O, A , IS]) or presence (O, A, II) of aphidicolin. The radioactivity incorporated into viral and cellular DNA fractions (see a) was measured at different times post-infection. O, 0 , Cellular DNA from mockinfected cells; A, A , cellular DNA from MVM-infected cells; I-q, II, viral DNA from MVM-infected cells. infected A9 cells to 3 or 5 ~ of its n o r m a l level at any t i m e after release f r o m Go (Fig. 1 b a n d 2 b). M o r e o v e r , viral D N A in M V M - i n f e c t e d cells was as sensitive to a p h i d i c o l i n as cellular D N A replication, irrespective o f t i m e post-infection (Fig. 2b). A p h i d i c o l i n did not alter the f r a c t i o n o f the culture i n v o l v e d in D N A synthesis but rather the level o f i n c o r p o r a t i o n p e r r e p l i c a t i n g cell (data not shown). T h e influence of the i n h i b i t i o n o f D N A synthesis on the f o r m a t i o n o f

M V M DNA synthesis

1995

T a b l e 1. Time course of viral and cellular DNA synthesis in MVM-infected A9 cells Rate of DNA synthesis (ct/min/104 cells)* Time post-infection (h) 5 19 26 32 45

Total DNA t 2050 27800 28679 6219 5720

Viral DNA~ ND[] 3367 6564 1293 795

Cellular DNA§ ND 24433 22115 4926 4925

* Cultures were pulse-labelled with [3H]TdR and lysed at intervals following release from G O block and infection. Radioactivity incorporated into TCA-insoluble material was measured. t Total DNA extracted' from infected cells. Virus-specific DNA, as determined by DNA DNA reassociation. The reassociation reaction was completed within 8 min in the presence of 22 Ixg/ml of non-radioactive viral DNA. Values were corrected for the extent of reassociation of i:ellular DNA under those conditions (approximately 5 ~). § Total DNA synthesis minus viral DNA synthesis. II ND, Not done.

I

I

I

-3

f x

< Z

2~ 0.5

. ~0

0

0

10 20 30 Time post-infection (h)

Fig. 3. Effect of aphidicolin on progeny virus formation in MVM-infected A9 cells, At intervals after MVM infection, A9 cells were either collected (closed symbols, continuous line) or further incubated in the presence of 15 ~tM-aphidicolinuntil harvest at 36 h post-infection (same open symbols, interrupted lines). Collected cells were disrupted and the infectivity of associated viral particles was measured. The dotted line is the cumulative curve of viral DNA synthesis calculated from Fig. 2(b) and considered as 1.0 at 36 h post-infection.

infectious progeny particles was studied in M V M - i n f e c t e d cultures exposed to a p h i d i c o l i n from different times to harvest time at 36 h after infection. It is a p p a r e n t from Fig. 3 that little virus p r o d u c t i o n occurred in the presence of aphidicolin. I n the a b s e n c e of the drug, most of the viral p r o g e n y was assembled d u r i n g the interval from 24 to 36 h post-infection. T h e p r o d u c t i o n of the burst was c o n c o m i t a n t with a late phase o f viral D N A replication a c c o u n t i n g for only 30 ~ o f the total viral D N A synthesized from the time of infection. T h e i n h i b i t i o n of this residual D N A synthesis by aphidicolin (Fig. 2b) was associated with a d r a m a t i c r e d u c t i o n of the burst size (Fig. 3). Thus, although a major fraction of total viral D N A had b e e n synthesized by 24 h, little p r o g e n y single-stranded D N A was available for e n c a p s i d a t i o n unless a n a p h i d i c o l i n - s e n s i t i v e event(s) occurred d u r i n g the s u b s e q u e n t 12 h.

1996

N. HARDT AND OTHERS

DISCUSSION At any time after infection, aphidicolin reduced viral DNA synthesis by more than 9 5 ~ (Fig. 2b). Parvovirus DNA replication involves the conversion of input single-strands into duplex parental replicative forms (RF). The multistep amplification of parental RF DNA gives rise to a large pool of RF molecules and eventually to progeny single-stranded DNA (Berns & Hauswirth, 1978). Parvovirus replication accounts for up to 75~ of uninfected cell DNA synthesis within a 35 h period which comprises the crossing of S phase (Fig. 2b). Most of the incorporation of [3H]TdR into viral DNA is associated with the viral DNA amplification phase; the conversion of the few 5000 nucleotides-long input gcnomes contributes minimally to total viral DNA synthesis. Consequently, the results presented indicate that aphidicolin prevents most of the amplification of parvovirus DNA, but they provide no information on the drug sensitivity of parental RF formation. Aphidicolin is known to inhibit cellular nuclear DNA replication (Geuskens et al., 1981). The inhibition of viral DNA synthesis by this drug is unlikely to result merely from the inhibition of cellular DNA replication since viral DNA synthesis does not require concomitant cellular D N A replication (Parris & Bates, 1976). A more likely interpretation is that an aphidicolin-sensitivc effector participates in a step of MVM DNA replication which is limiting for the amplification of viral DNA throughout the virus life cycle. The dependence of parvovirus DNA replication on host cell enzymes (Rhode, 1978) and the specificity of aphidicolin for cellular DNA polymerase czstrongly suggest that the latter enzyme is involved in viral DNA synthesis. It cannot be ruled out, however, that the drug affects other as yet undiscovered cellular or viral factors involved in parvovirus DNA synthesis. The role of DNA polymerase ct in DNA replication of autonomous parvovirus is further supported by the temporal correlation between this enzymic activity and viral DNA synthesis in vivo (Pritchard et al., 1978b) and by the ability of partially purified preparations of this enzyme to catalyse steps of viral DNA synthesis in vitro (Pritchard et al., 1981 ; Kollek et al., 1982; Faust & Rankin, 1982; Hfibscher et al., 1982; M. GiJnther, personal communication). The inhibition of MVM DNA synthesis was greater than 95 ~ and not significantly different from that of cellular DNA synthesis (Fig. 2 b). Such a sensitivity to aphidicolin would be consistent with the fact that amplification of the DNA of autonomous parvoviruses relies essentially or even exclusively on DNA polymerase ct, as does the replication of cellular chromosomal DNA. However, the DNA amplification of autonomous parvoviruses is a multistep process involving the replication of RF molecules and the asymmetric production of single-stranded DNA of viral polarity (Berns & Hauswirth, 1978). Thc temporal sequence and the enzymology of these steps are still unclear. The permanent need for DNA polymerase ct therefore does not rule out additional polymerase requirements for the completion of viral DNA synthesis. Such a possibility is raised by recent studies showing that DNA polymerase ), participates in parvovirus DNA replication in vitro (Kollek et al., 1982; M. Giinther, personal communication). Very little new virus assembly occurrcd in the presence of aphidicolin (Fig. 3). Although a major fraction of viral DNA had been synthesized by 24 h after infection, cells did not contain, at this stage, a large pool of progeny viral strands which could be cncapsidated in the absence of an aphidicolin-sensitive event. The specificity of this drug points to the dependence of MVM production on DNA synthesis. The inhibition by aphidicolin of residual cellular D N A replication is unlikely to account for the reduced formation of infectious particles since the assembly and processing of autonomous parvoviruses do not require the host cell to be in S phase (Parris & Bates, 1976; Richards et al., 1978). It follows that the burst of virus encapsidation appears to require concomitant synthesis of viral DNA. A similar conclusion was drawn from the effect of hydroxyurea on MVM production by rat cells (Richards et al., 1977, 1978). Little, if any, free single-stranded DNA was found in infected cells (Tattcrsall & Ward, 1976), suggesting that the production of progeny strands is tightly coupled to their packaging. The ultimate step in the formation of viral single strands is thought to be their displacement from duplex forms by asymmetric replication (Berns & Hauswirth, 1978). It was proposed that this strand displacemcnt synthesis may be driven by encapsidation (Tattersall & Ward, 1976; Richards et al., 1978) thereby accounting for the dependence of the assembly of new particles on continuing viral DNA synthesis.

M V M D N A synthesis

1997

The burst of MVM packaging was associated with a late phase of viral DNA replication accounting for only 30 ~ of the total viral DNA synthesized after infection (Fig. 3). Since virion morphogenesis lags behind the amplification of the pool of RF DNA (Ward & Tattersall, 1982), this late phase is likely to be mainly devoted to the asymmetric strand displacement synthesis of single-stranded progeny DNA. The aphidicolin sensitivity of both virion assembly and late viral DNA replication strongly suggests that DNA polymerase a participates in vivo in the displacement synthesis. This conclusion is consistent with the aphidicolin sensitivity of this reaction in parvovirus DNA replication in vitro (Pritchard et al., 1981 ; Kollek et al., 1982). A9 cells were accumulated in Go, inoculated with MVM and released from growth inhibition. MVM uptake was presumably completed when cells entered into S phase since this process is independent of the cell cycle and is carried out within a few hours after infection (Linser et al., 1979; J. Rommelaere, unpublished results). Therefore, initiation of viral DNA replication was mainly limited by the dependence of the conversion of the parental genome into duplex RF, on cellular factors expressed specifically during the S phase ('S factors') (Tattersall, 1972). The onset of viral DNA synthesis was found to coincide with the burst of cellular DNA replication (Fig. 2b). Consistently, the first viral double-stranded DNA was detected in early S phase (Wolter et al., 1980). In contrast, other authors reported that viral DNA synthesis was delayed until the end of the S phase (Parris & Bates, 1976). Taken together, these data suggest that cellular S factor(s) might be available throughout the S phase and that the relative timing of cellular and viral DNA synthesis may vary depending on the system and cell synchronization method used. Viral DNA synthesis was accompanied by an inhibition of cellular DNA replication detectable as soon as 10 h after infection (Fig. 2b). Viral DNA synthesis might compete with cellular DNA replication since it appears to make use, in particular, of the cellular DNA polymerase ct (see above) and it accounts for as much as 75 ~ of thymidine incorporation during a normal S phase (Fig. 2b; Parris & Bates, 1976). The decrease in RNA and protein syntheses (Parris & Bates, 1976) and the degeneration of nuclear elements (Richards et al., 1977) which occur at the time of viral DNA synthesis and virion assembly, respectively, might also participate in the reduction of cellular DNA replication. The inhibition of cellular DNA synthesis was compensated by concomitant viral DNA replication. Therefore, MVM infection caused only a slight decrease of the rate of overall thymidine incorporation, which was restricted to early S (Fig. 1 b). Similarly, cells inoculated with the bovine (Parris & Bates, 1976) and hamster osteolytic (Hampton, 1970) parvoviruses proceed through S phase without a significant reduction in the rate of total DNA synthesis. Taken together, the data presented and discussed point to the intimate relationship between DNA replication of autonomous parvoviruses and their host cells with regard to their nuclear location, cell cycle dependency and enzymology. In this respect, parvoviruses appear to provide an attractive model for studying the molecular mechanism of DNA replication in eukaryotic cells. We wish to thank Professor M. Errera for his support and Drs J. Tal and J. J. Cornelis for helpful discussions. This work was supported by an agreement between the Belgian Government and the Universit6 libre de Bruxelles 'Actions de Recherche Concert6e', the European Communities [contract BIO-359-B(G) and ENV/335/B], the Minist6re de la Sant6 Publique (contract G/826/1981), the Fonds National de la Recherche M6dicale (contract 3.4514.80) and partly by the 'Programma finalizzato Controllo delle Malattie da Infezione'. J. Rommelaere is Chercheur Qualifi6 du Fonds National Beige de la Recherche Scientifique. REFERENCES

BERNS, K. I. & HAUSWIRTH,W. W. (1978). Parvovirus D N A structure and replication. In Replication of Mammalian Parvoviruses, pp. 13-32. Edited by D. C. Ward & P. Tattersall. New York: Cold Spring Harbor Laboratory. BOURGUIGNON, G. J., TATrERSALL, P. & WARD, D. C. (1976). D N A of minute virus of mice: self-priming, nonpermuted, single-stranded genome with a 5'-terminal hairpin duplex. Journal of Virology 20, 290-306. CLEAVER, J. E. & nOLFORD, R. M. (1965). Investigations into the incorporation of [3H]thymidine into D N A in Lstrain cells and the formation of a pool of phosphorylated derivatives during pulse labelling. Biochimiea et biophysiea aeta 103, 654-671. FAUST,E. A. & RANK.IN,C. D. (1982). In vitro conversion of MVM parvovirus single-stranded D N A to the replicative form by D N A polymerase c~ from Ehrlich ascites tumour cells. Nucleic Acids Research 10, 4181~1201.

1998

N.

HARDT

AND

OTHERS

GEUSKENS, M., HARDT, N., PEDRALI-NOY, G. & SPADARI, S. (1981). An autoradiographic demonstration of nuclear D N A replication by D N A polymerase ct and of mitochondrial D N A synthesis by D N A polymerase y. Nucleic Acids Research 9, 1599 1613. HAMPTON, E. G. (1970). H-1 virus growth in synchronized rat embryo cells. Canadian Journal of Microbiology 16, 266~268. HANDA, H. & CARTER, B. J. (1979). Adeno-associated virus D N A replication complexes in herpes simplex virus or adenovirus-infected cells. Journal of Biological Chemistry 254, 6603-6610. HARDT, N., DE KEGEL, D., VAN HEULE, L., VILLANI, G. & SPADARI, S. (1980). D N A polymerase 7, cytochrome C oxidase and mitochondrial integrity in rabbit spleen lymphocytes stimulated with concanavalin A. Experimental Cell Research 127, 269-276. HUBSCHER, U., GERSCHWlLDER, P. & McMASTER, G. K. (1982). Evidence for a eukaryotic D N A polymerase ct holoenzyme. EMBO Journal 12, 1513-1519. IKEGAMI, S., TAGUCHI, T. & OKASHI, M. (1978). Aphidicolin prevents mitotic cell division by interfering with the activity of D N A polymerase ct. Nature, London 275, 458-460. KOLLEK, R. & GOULIAN, M. (1981). Synthesis of parvovirus H-1 replicative form from viral D N A by D N A polymerase y. Proceedings of the National Academy of Sciences, U.S.A. 78, 6206-6210. KOLLEK, R., TSENG, B. Y. & GOULIAN, M. (1982). D N A polymerase requirements for parvovirus H-1 D N A replication in vitro. Journal of Virology 41, 982-989. LINSER, P., BRUNING, H. & ARMENTROUT,R. W. (1979). U p t a k e of minute virus o f mice into cultured rodent cells. Journal of Virology 31, 537 545. LITTLEFIELD, J. W. (1964). Three degrees of guanylic acid inosinic acid pyrophosphorylase deficiency in mouse fibroblasts. Nature, London 203, 1142-1144. PARRIS, D. S. & BATES,R. C. (1976). Effect of bovine parvovirus replication on D N A , R N A and protein synthesis in S phase cells. Virology 73, 72-78. PEDRALI-NOY, G. & SPADARI, S. (1979). Effect of aphidicolin on viral and h u m a n D N A polymerases. Biochemical and Biophysical Research Communications 88, 1194-2002. PEDRALI-NOY, G. & SPADARI, S. (1980). M e c h a n i s m of inhibition of herpes simplex virus and vaccinia virus D N A polymerases by aphidicolin, a highly specific inhibitor of D N A replication in eukaryotes. Journal of Virology 36, 457-464. PEDRALI-NOY, G. & WEISSBACH,A. (1977). Evidence of a repetitive sequence in vaccinia virus D N A . Journal of Virology 24, 406-407. PEDRALI-NOY,G., BELVEDERE,M., CREPALD1,T., FOCHER, F. & SPADARI,S. (1982). Inhibition of D N A replication and growth of several h u m a n and murine neoplastic cells by aphidicolin without detectable effect upon synthesis of immunoglobulins and H L A antigens. Cancer Research 42, 3810-3813. PRtTC~L~RD,C., PATTON,J. T., BATES,R. C. & STOUT,E. R. (1978 a). Replication of non-defective parvoviruses: lack of a virion-associated D N A polymerase. Journal of Virology 28, 20-27. PRITCHARD, C., BATES,R. ¢. & STOUT, E. R. (1978b). Levels of cellular D N A polymerases in synchronized bovine parvovirus-infected cells. Journal of Virology 27, 258 261. PRITCHARD, C., STOUT,E. R. & BATES,R. C. (1981). Replication of parvoviral D N A . I. Characterization of a nuclear lysate system. Journal of Virology 37, 352-362. RHODE, S. L (1978). H-1 D N A synthesis. In Replication of Mammalian Parvoviruses, pp. 279-296. Edited by D. C. W a r d & P. Tattersall. New York: Cold Spring Harbor Laboratory. RICHARDS,R. LINSER, P. & ARMENTROUT,R. W. (1977). Kinetics of assembly of a parvovirus, minute virus of mice, in synchronized rat brain cells. Journal of Virology 22, 778 793. RICHARDS,R., LINSER, P. & ARME~rrROUT,R. W. (1978). Maturation of minute virus of mice particles in synchronized rat brain cells. In Replication of Mammalian Par~,oviruses,pp. 447-458. Edited by D. C. W a r d & P. Tattersall. New York: Cold Spring Harbor Laboratory. ROMMELAERE,J., VOS,J. M., CORNELIS,J. J. & WARD, D. C. (1981). UV-enhanced reactivation of minute virus of mice : stimulation of a late step in the viral life cycle. Photochemistry and Photobiology 33, 845-854. SALZMAN,L. A., WHITE, W. L. & McKERLIE, L. (1972). Growth characteristics of K i l h a m rat virus and its effect on cellular macromolecular synthesis. Journal of Virology 10, 573-577. TATTERSALL, P. (1972). Replication of the parvovirus MVM. Dependence of virus multiplication and plaque formation on cell growth. Journal of Virology 10, 586-590. TATTERSALL,P. & WARD, D. C. (1976). Rolling hairpin model for replication of parvovirus and linear chromosomal D N A . Nature, London 263, 106-109. TATTERSALL,P. & WARD, D. C. (1978). The parvoviruses: an introduction. In Replicationof Mammalian Parvoviruses, pp. 3-12. Edited by D. C. Ward & P. Tattersall. New York: Cold Spring Harbor Laboratory. TATTERSALL,P., CAWTE,P. J., SHATKIN,A. J. & WARD, D. C. (1976). Three structural polypeptides coded for by m i n u t e virus of mice, a parvovirus. Journal of Virology 20, 273-289. WARD, D. C. & TATTERSALL,P. (I982). Minute virus of mice. In The Mouse in Biomedical Research, vol. 2, pp. 313 334. New York: Academic Press. WOLTER, S., RICHARDS, R. & ARMENTROUT,R. W. (1980). Cell cycle-dependent replication of the D N A of minute virus of mice, a parvovirus. Biochimica et biophysica acta 607, 420M31.

(Received 7 March 1983)