VIROLOGY
191,498-501
(1992)
African Swine Fever Virus-Induced
DNA Polymerase Is Resistant to Aphidicolin
MARIA ISABELMARQUESAND Jo&o V. COSTA’ Gulbenkian Institute of Science, Apartado
14, P-278 1 Oeiras Codex, Portugal
Received June 29, 1992; accepted August 3, 1992
African swine fever virus (ASFV) induces the synthesis of a virus-specific DNA polymerase, which is inhibited by phosphonoacetic acid and cytosine arabinoside. In contrast to all other a-like DNA polymerases of DNA viruses, ASFV-specific DNA polymerase is resistant to aphidicolin. Concentrations of the drug as high as 160 PM had no effect on virus production or plaquing efficiency. The resistance of ASFV DNA polymerase to aphidicolin was confirmed by analyzing the effect af the drug on viral DNA synthesis. A moderate inhibition of viral DNA synthesis was observed when aphidicolin was added immediately after virus adsorption but normal synthesis occurred, with a peak at 10 hr p.i., when the drug was added at 2 or 4 hr p.i. This suggests that a very early phase of ASFV DNA replication is sensitive to aphidicolin and is probably catalyzed by a different enzyme. An in vitro assay of DNA polymerase activity was used to assay the sensitivity of the virus-specific DNA polymerase to inhibitors. In correspondence to the results observed in vivo, phosphonoacetic acid strongly inhibited the enzyme activity, whereas aphidicolin had no effect. Resistance to aphidicolin was independent of the concentration of dCTP used in the assay. Three independent ASFV mutants resistant to phosphonoacetic acid showed the same resistance to aphidicolin as wild type virus. o 1992 Academic Press. Inc.
Aphidicolin, a tetracyclic diterpenoid obtained from Cephalosporium aphidicob, inhibits the activity of eukaryotic a-like DNA polymerases (I) and of DNA polymerases induced by herpesviruses (2, 3), poxviruses (2), and baculoviruses (4. The DNA polymerases of adenoviruses and bacteriophage 429, which use a protein primer, are also inhibited by aphidicolin, though at higher concentrations (5, 6). Another group of animal DNA viruses, hepadnaviruses, are resistant to aphidicolin (7), but their DkA is replicated by a polymerase with retrotranscriptase activity, not by a conventional DNA polymerase. All those aphidicolin-sensitive DNA polymerases belong to the same enzyme family. They are also inhibited by cytosine arabinoside and by phosphonoacetic acid and they share several conserved regions in amino acid sequence (8). The sensitivities to aphidicolin and to phosphonoacetic acid seem to be related and both inhibitors are presumed to interact with the enzyme site for dNTP binding. Mutations conferring resistance to aphidicolin or to phosphonoacetic acid map within the same conserved region (Q-72) and mutants resistant to one of the inhibitors frequently show altered sensitivity to the other inhibitor (10, 12, 73). African swine fever virus (ASFV) is a large DNA virus that replicates in the cytoplasm of infected cells (re-
’
viewed in (14)). The morphology of the virus particle is typical of iridoviruses but the genomic structure and molecular biology strategy of the virus is much closer to that of poxviruses. The viral genome encodes for all the enzymes required for DNA replication and transcription, including a DNA polymerase. As poxvirus DNA polymerase and the other a-like animal virus DNA polymerases, ASFV-induced DNA polymerase is inhibited by phosphonoacetic acid (15). We report in this paper that ASFV-induced DNA polymerase is fully resistant to aphidicolin, which constitutes an exception to known DNA polymerases induced by animal viruses. Figure 1 shows the effect of aphidicolin, phosphonoacetic acid, and cytosine arabinoside on the production of infectious virus and on the plaquing efficiency of the virus. Plaquing efficiency is defined as the ratio between the number of plaques formed in the presence and in the absence of the drug. One million Vero cells were infected with ASFV at 0.1 PFU per cell. After virus adsorption, aphidicolin, phosphonoacetic acid or cytosine arabinoside were added at different concentrations. Control cells to which aphidicolin was not added were infected in the presence of dimethylsulfoxide in the same amount as that introduced with aphidicolin. When cytoplasmic effect was complete in control-infected cells, the medium of all infected cells was harvested and titrated in the absence of any inhibitor. For determining plaquing efficiency, the same virus stock was plaque titrated on Vero cells in the absence of
To whom reprint requests should be addressed
0042-6822192 CopyrIght
498
$5.00
0 1992 by Academic
All rights of reproducton
Press, Inc.
in any form reserved.
SHORT COMMUNICATIONS
2 08
8
loo >r z
6
1@2
.$ s
@ v1. x=J
4
$2
*
104 10s
0) .E i% (II h
E 0 0
50
100 150
Aphidicolin,
PM
0
0.4
0.8
PAA,
mM
0
50
100 150
araC, PM
FIG. 1. Effect of aphidicolin, phosphonoacetic acid (PAA), and cytosine arabinoside (araC) on ASFV replication (solid circles) and plaquing efficiency (open circles). After virus adsorption, aphidicolin, phosphonoacetic acid, or cytosine arabinoside were added at the indicated concentrations. When cytoplasmic effect was complete in control-infected cells, the medium of all infected cells was harvested and titrated in the absence of inhibitors. For determining plaquing efficiency, ASFV was titrated in the absence of inhibitors or in the presence of different concentrations of aphidicolin, phosphonoacetic acid, or cytosine arabinoside.
inhibitors or in the presence of different concentrations of aphidicolin, phosphonoacetic acid, or cytosine arabinoside. As shown in Fig. 1, both phosphonoacetic acid and cytosine arabinoside at concentrations equal to or higher than 0.4 mM or 20 PM, respectively, reduced virus production by three or four orders of magnitude. Their effect on plaquing efficiency was even more pronounced. No plaques were formed in the presence of 0.8 mM phosphonoacetic acid or 20 PM cytosine arabinoside. In contrast, aphidicolin had no effect on virus production or plaquing efficiency when added to infected cells at concentrations as high as 160 PM. Identical results were obtained when cells were infected at 0.3 or 1 PFU per cell. As a control, we analyzed the effect of aphidicolin on plaquing efficiency of vaccinia virus and observed that 40 pM aphidicolin reduced to 1% the number of plaques and that no plaque was formed in the presence of 80 pM aphidicolin. Resistance to aphidicolin was confirmed by onestep growth experiments. Cells were infected at 10 PFU per cell in the presence of 80 p/1/1aphidicolin and the medium was collected at 4-hr intervals until 24 hr after infection and titrated. Infected control cells and cells infected in the presence of aphidicolin produced a maximum of 2 x 10’ PFU per million cells at 24 hr p.i., whereas the yield in the presence of either 0.8 mM phosphonoacetic acid or 20 pM cytosine arabinoside did not exceed the inoculum. The concentrations of aphidicolin to which ASFV was resistant are much higher than all those reported as inhibiting the growth of other viruses. Aphidicolin concentrations resulting in 50% reduction of virus
499
growth or more are 1.5 pM for the baculovirus Bombix mori nuclear polyhedrosis virus (4) 0.6 pM for herpes simplex virus type 1 and 1 1 pM for vaccinia virus (2). Viral DNA replication was assessed by determining the incorporation of tritiated thymidine into acid-insoluble material in the cytoplasm of infected cells. Cells were infected at 20 PFU per cell. After adsorption, the medium was replaced with medium containing 2% dialyzed newborn calf serum. At different times after infection, cells were pulse-labeled for 15 min with 10 &i/ml methyl-[3H]thymidine (47 Ci/mmol). The cells were washed and scraped into ice-cold PBS and resuspended in 20 mM Tris-HCI, pH 7.5, 150 mM NaCI, 5 mM MgCI,. After addition of 0.5% NP-40, the cells were incubated in ice for 15 min and the nuclei were removed by centrifugation. The cytoplasmic fraction was precipitated with TCA and counted. In some cases, TCA-precipitable radioactivity from washed nuclei was also counted. Differently from uninfected cells, where cytoplasmic incorporation of nucleotide was negligible, infected cells showed a marked increase in thymidine incorporation, starting about 4 hr p.i. and reaching a maximum at 10 hr p.i. (Fig. 2). Consistently with the effect on virus yield, both phosphonoacetic acid and cytosine arabinoside inhibited almost completely viral DNA replication (Fig. 2B). We tested two concentrations of aphidicolin, 60 pM and 120 p/1/1.Addition of aphidicolin at these concentrations reduced thymidine incorporation in the
30 25 m b ? x E 8
2015lo50 -. 0 2
4
6
6 10 12
I
2
4
6
6 10 12
hr p.i. FIG. 2. Time course of viral DNA synthesis in the cytoplasm of infected cells, assayed by measuring thymidine incorporation into TCA-insoluble material. Mock-infected cells (open circles) and infected cells (solid circles) were pulse-labeled for 15 min with 10 pCil ml methyl-[3H]thymidine. The cells were extracted and the cytoplasmic fraction was precipitated with TCA and counted. Inhibitors were added since the beginning of infection at the following concentrations: 60 pM (A, solid squares) or 120 PM (A, solid diamonds) aphidicolin. 0.8 mM phosphonoacetic acid (6, solid squares), and 20 @M cytosine arabinoside (6, solid diamonds). Addition of aphidicolin (A) and phosphonoacetic acid (B) was also done at 2 hr p.i. (open squares) or 4 hr p.i. (open diamonds).
SHORT COMMUNICATIONS
500
0
50
100
150 0 200 400 600 800
Aphidicolin, PM
PM
PM
FIG. 3. Effect of aphidicolin and phosphonoacetic acid (PAA) on DNA polymerase activity of ASFV. The activity of cytoplasmic extracts from infected (solid lines) or mock-infected cells (broken lines) was assayed by using activated calf thymus DNA in reaction mixtures containing the indicated concentrations of aphidicolin or phosphonoacetic acid. In a different experiment, the effect of aphidicolin was determined in reaction mixtures containing 2.5 FM (triangles), 5 gM (diamonds), or 100 pM (squares) dCTP. DNA polymerase activity is indicated as pmol incorporated dATP per min per milligram protein in the extract.
nucleus of infected or uninfected cells to 0.05% in average, but had no noticeable effect on virus growth, as described above. Surprisingly, we observed that both doses of aphidicolin caused a reduction of about 50% in thymidine incorporation into cytoplasmic viral DNA when the drug was added at 0 hr p.i. (Fig. 2A). This contradiction with the lack of effect on virus growth might be explained if we presume that ASFV is replicated in large excess and so a 50% reduction in viral DNA replication is not limitative to virus production. To determine whether the reduction in DNA synthesis was actually due to partial inhibition of ASFV DNA polymerase activity, we tested the effect of aphidicolin on the enzyme activity in vitro. Cells were infected at 20 PFU per cell and cytoplasmic extracts were prepared at 10 hr p.i. as described above. Reaction mixtures for DNA polymerase assays contained 10 ~1 cytoplasmic extract from 10” cells, 25 pg activated calf thymus DNA, and 2 &i [32P]dATP (800 Ci/mmol) in 100 ~1reaction buffer: 50 mMTris-HCI, pH 7.5,50 mM KCI, 10 mM MgCI,, 1 rnlLl dithiothreitol, 100 PM each dCTP, dGTP, and dTTP, and 20 PM dATP. Inhibitors were added at different concentrations. The mixtures were incubated for 45 min at 37” and insoluble radioactivity was counted. The results depicted on Fig. 3 confirm that ASFV DNA polymerase is indeed fully resistant to aphidicolin at concentrations up to 150 pm. In contrast, the enzyme activity in vitro was drastically inhibited by phosphonoacetic acid. In a first experiment (Fig. 3, solid circles) using dATP as the labeled nucleoside triphos-
phate, we used the same concentration of 100 PM for all other three dNTP. This could have masked an inhibitory effect of aphidicolin, because this drug inhibits DNA polymerase activity by competing with dCTP (16). To rule out this possible source of error, we also performed the experiment using much lower concentrations of dCTP, in the order of the in vivo concentration (about 3 @I). Again, the enzyme activity was completely resistant to aphidicolin when assayed in mixtures containing 2.5 PM or 5 &I dCTP. These results seem to exclude the hypothesis that we discussed above for explaining the effect of aphidicolin on thymidine incorporation into viral replicative DNA. It was recently reported that ASFV DNA replication occurs in two phases (17). Whereas most of viral DNA replication takes place in the cytoplasm of infected cells, with a peak at about 10 hr p.i., an earlier step in DNA replication occurs in the nucleus of all or some infected cells. If this phase of viral DNA replication is catalyzed by an aphidicolin-sensitive polymerase and provides templates for the cytoplasmic DNA replication, then the overall viral DNA replication would be affected, since we added the drug immediately after the end of virus adsorption. This explanation is supported by the fact that the degree of apparent inhibition of viral DNA replication in the cytoplasm is dependent on the time of aphidicolin addition. Instead of having aphidicolin present from the beginning of infection, we added the drug at 2 or 4 hr p.i., times when cytoplasmic DNA replication is still at very low levels. In our system, these are the times of start and end of a nuclear minor phase of viral DNA replication (data not shown). As a control, we performed similar tests using phosphonoacetic acid. In this case, addition of the inhibitor at 2 or 4 hr p.i. resulted in the same level of inhibition as the one observed when the drug was added at the beginning of infection (Fig. 2B). Differently, when we added 120 &I aphidicolin to infected cells at 2 hr p.i. the maximum incorporation of tritiated thymidine was intermediate between the peaks observed in the absence of the drug and in the presence of aphidicolin from the beginning of infection (Fig. 2A). The same concentration of aphidicolin added at 4 hr p.i. had practically no effect on the maximum of thymidine incorporation, as compared to the control in the absence of aphidicolin. We also analyzed the effect of aphidicolin on DNA polymerase activity of extracts from cells infected with three independent ASFV mutants resistant to phosphonoacetic acid. One of the mutants is a spontaneous mutant obtained after serial passages of the virus in the presence of phosphonoacetic acid and the other two mutants were isolated from a virus stock mutagenized with nitrosoguanidine. DNA polymerase activity of
SHORT COMMUNICATIONS
all the three cytoplasmic extracts, assayed as described above, was completely resistant to aphidicolin and was practically of the same level as the one of wild type virus. As far as we know, ASFV is the only animal DNA virus whose DNA polymerase is resistant to aphidicolin. Since the enzyme is sensitive to phosphonoacetic acid and cytosine arabinoside, this is an exception to the correlation between the action of the three inhibitors, which apparently act at the same site of a-like eukaryotic and viral DNA polymerases. Our results also suggest that the two phases of ASFV DNA replication may be catalyzed by two different DNA polymerases, the first nuclear phase being dependent on an aphidicolin-sensitive polymerase, possibly a cellular enzyme. Further experiments are in course to identify and characterize the enzyme involved in ASFV DNA replication at the nucleus of infected cells. REFERENCES 1. SUGINO. A., and NAKAYAMA, K., Proc. Nat/ Acad. Sci. USA 77, 7049-7053 (1980). 2. PEDRALI-NOY, G., and SPADARI, S., J. Viral. 36, 457-464 (1980).
501
3. Dc~occ~o, R. A., CHADHA, K., and SRIVASTAVA,B. I. S., Biochim. Biophys. Acfa 609, 224-231 (1980). 4. MIKHAILOV, V. S., MARLYEV, K. A.. ATAEVA, J. O., KULLYEV,P. K., and ATRAZHEV, A. M., Nucleic Acids Res. 14, 3841-3857 (1986). 5. FOSTER, D. A., HANTZOPOULOS, P., and ZUBAY, G., J. Viral. 43, 679-686 (1982). 6. BUNCO, L., and SALAS, M.. Virology 153, 179-187 (1986). 7. OFFENSPERGER,W.-B., WALTER, E., OFFENSPERGER,S., ZESCHNIGK, C., BLUM, H. E., and GEROK,W., Virology 164, 48-54 (1988). 8. ITO, J., and BRAITHWAITE,D. K., Nucleic Acids Res. 19, 40454057 (1991). 9. EARL, P. L., JONES, E. V., and Moss, B., Proc. Natl. Acad. Sci. USA 83,3659-3663 (1986). 10. TSURUMI, T., MAENO, K., and NISHIYAMA, Y., J. Viral. 61, 388-394 (1987). Il. DEFILIPPES,F. M., 1. l/ire/. 63, 4060-4063 (1989). 12. TADIE, J. A., and TRAKTMAN, P., J. L&o/. 65, 869-879 (1991). 13. MATSUMOTO, K., KIM, C. I., KOBAYASHI, H., KANEHIRO, H., and HIROKAWA,H., Virology 178, 337-339 (1990). 14. COSTA, J. V., In “Molecular Biology of Iridoviruses” (G. Darai, Ed.), pp. 247-270. Kluwer Academic, Boston, 1990. 15. MORENO, M. A., CARRASCOSA,A. L., ORTIN, J., and VIWUELA,E., J. Gen. Viral. 39, 253-258 (1978). 16. KROKAN, H., WIST. E., and KROKAN, R. H.. Nucleic Acids Res. 9, 4709-4719 (1981). 17. GARCIA-BEATO, R., SALAS, M. L., VIITIUELA,E.. and SALAS, J., Viralogy 188, 637-649 (1992).
191,498-501
(1992)
African Swine Fever Virus-Induced
DNA Polymerase Is Resistant to Aphidicolin
MARIA ISABELMARQUESAND Jo&o V. COSTA’ Gulbenkian Institute of Science, Apartado
14, P-278 1 Oeiras Codex, Portugal
Received June 29, 1992; accepted August 3, 1992
African swine fever virus (ASFV) induces the synthesis of a virus-specific DNA polymerase, which is inhibited by phosphonoacetic acid and cytosine arabinoside. In contrast to all other a-like DNA polymerases of DNA viruses, ASFV-specific DNA polymerase is resistant to aphidicolin. Concentrations of the drug as high as 160 PM had no effect on virus production or plaquing efficiency. The resistance of ASFV DNA polymerase to aphidicolin was confirmed by analyzing the effect af the drug on viral DNA synthesis. A moderate inhibition of viral DNA synthesis was observed when aphidicolin was added immediately after virus adsorption but normal synthesis occurred, with a peak at 10 hr p.i., when the drug was added at 2 or 4 hr p.i. This suggests that a very early phase of ASFV DNA replication is sensitive to aphidicolin and is probably catalyzed by a different enzyme. An in vitro assay of DNA polymerase activity was used to assay the sensitivity of the virus-specific DNA polymerase to inhibitors. In correspondence to the results observed in vivo, phosphonoacetic acid strongly inhibited the enzyme activity, whereas aphidicolin had no effect. Resistance to aphidicolin was independent of the concentration of dCTP used in the assay. Three independent ASFV mutants resistant to phosphonoacetic acid showed the same resistance to aphidicolin as wild type virus. o 1992 Academic Press. Inc.
Aphidicolin, a tetracyclic diterpenoid obtained from Cephalosporium aphidicob, inhibits the activity of eukaryotic a-like DNA polymerases (I) and of DNA polymerases induced by herpesviruses (2, 3), poxviruses (2), and baculoviruses (4. The DNA polymerases of adenoviruses and bacteriophage 429, which use a protein primer, are also inhibited by aphidicolin, though at higher concentrations (5, 6). Another group of animal DNA viruses, hepadnaviruses, are resistant to aphidicolin (7), but their DkA is replicated by a polymerase with retrotranscriptase activity, not by a conventional DNA polymerase. All those aphidicolin-sensitive DNA polymerases belong to the same enzyme family. They are also inhibited by cytosine arabinoside and by phosphonoacetic acid and they share several conserved regions in amino acid sequence (8). The sensitivities to aphidicolin and to phosphonoacetic acid seem to be related and both inhibitors are presumed to interact with the enzyme site for dNTP binding. Mutations conferring resistance to aphidicolin or to phosphonoacetic acid map within the same conserved region (Q-72) and mutants resistant to one of the inhibitors frequently show altered sensitivity to the other inhibitor (10, 12, 73). African swine fever virus (ASFV) is a large DNA virus that replicates in the cytoplasm of infected cells (re-
’
viewed in (14)). The morphology of the virus particle is typical of iridoviruses but the genomic structure and molecular biology strategy of the virus is much closer to that of poxviruses. The viral genome encodes for all the enzymes required for DNA replication and transcription, including a DNA polymerase. As poxvirus DNA polymerase and the other a-like animal virus DNA polymerases, ASFV-induced DNA polymerase is inhibited by phosphonoacetic acid (15). We report in this paper that ASFV-induced DNA polymerase is fully resistant to aphidicolin, which constitutes an exception to known DNA polymerases induced by animal viruses. Figure 1 shows the effect of aphidicolin, phosphonoacetic acid, and cytosine arabinoside on the production of infectious virus and on the plaquing efficiency of the virus. Plaquing efficiency is defined as the ratio between the number of plaques formed in the presence and in the absence of the drug. One million Vero cells were infected with ASFV at 0.1 PFU per cell. After virus adsorption, aphidicolin, phosphonoacetic acid or cytosine arabinoside were added at different concentrations. Control cells to which aphidicolin was not added were infected in the presence of dimethylsulfoxide in the same amount as that introduced with aphidicolin. When cytoplasmic effect was complete in control-infected cells, the medium of all infected cells was harvested and titrated in the absence of any inhibitor. For determining plaquing efficiency, the same virus stock was plaque titrated on Vero cells in the absence of
To whom reprint requests should be addressed
0042-6822192 CopyrIght
498
$5.00
0 1992 by Academic
All rights of reproducton
Press, Inc.
in any form reserved.
SHORT COMMUNICATIONS
2 08
8
loo >r z
6
1@2
.$ s
@ v1. x=J
4
$2
*
104 10s
0) .E i% (II h
E 0 0
50
100 150
Aphidicolin,
PM
0
0.4
0.8
PAA,
mM
0
50
100 150
araC, PM
FIG. 1. Effect of aphidicolin, phosphonoacetic acid (PAA), and cytosine arabinoside (araC) on ASFV replication (solid circles) and plaquing efficiency (open circles). After virus adsorption, aphidicolin, phosphonoacetic acid, or cytosine arabinoside were added at the indicated concentrations. When cytoplasmic effect was complete in control-infected cells, the medium of all infected cells was harvested and titrated in the absence of inhibitors. For determining plaquing efficiency, ASFV was titrated in the absence of inhibitors or in the presence of different concentrations of aphidicolin, phosphonoacetic acid, or cytosine arabinoside.
inhibitors or in the presence of different concentrations of aphidicolin, phosphonoacetic acid, or cytosine arabinoside. As shown in Fig. 1, both phosphonoacetic acid and cytosine arabinoside at concentrations equal to or higher than 0.4 mM or 20 PM, respectively, reduced virus production by three or four orders of magnitude. Their effect on plaquing efficiency was even more pronounced. No plaques were formed in the presence of 0.8 mM phosphonoacetic acid or 20 PM cytosine arabinoside. In contrast, aphidicolin had no effect on virus production or plaquing efficiency when added to infected cells at concentrations as high as 160 PM. Identical results were obtained when cells were infected at 0.3 or 1 PFU per cell. As a control, we analyzed the effect of aphidicolin on plaquing efficiency of vaccinia virus and observed that 40 pM aphidicolin reduced to 1% the number of plaques and that no plaque was formed in the presence of 80 pM aphidicolin. Resistance to aphidicolin was confirmed by onestep growth experiments. Cells were infected at 10 PFU per cell in the presence of 80 p/1/1aphidicolin and the medium was collected at 4-hr intervals until 24 hr after infection and titrated. Infected control cells and cells infected in the presence of aphidicolin produced a maximum of 2 x 10’ PFU per million cells at 24 hr p.i., whereas the yield in the presence of either 0.8 mM phosphonoacetic acid or 20 pM cytosine arabinoside did not exceed the inoculum. The concentrations of aphidicolin to which ASFV was resistant are much higher than all those reported as inhibiting the growth of other viruses. Aphidicolin concentrations resulting in 50% reduction of virus
499
growth or more are 1.5 pM for the baculovirus Bombix mori nuclear polyhedrosis virus (4) 0.6 pM for herpes simplex virus type 1 and 1 1 pM for vaccinia virus (2). Viral DNA replication was assessed by determining the incorporation of tritiated thymidine into acid-insoluble material in the cytoplasm of infected cells. Cells were infected at 20 PFU per cell. After adsorption, the medium was replaced with medium containing 2% dialyzed newborn calf serum. At different times after infection, cells were pulse-labeled for 15 min with 10 &i/ml methyl-[3H]thymidine (47 Ci/mmol). The cells were washed and scraped into ice-cold PBS and resuspended in 20 mM Tris-HCI, pH 7.5, 150 mM NaCI, 5 mM MgCI,. After addition of 0.5% NP-40, the cells were incubated in ice for 15 min and the nuclei were removed by centrifugation. The cytoplasmic fraction was precipitated with TCA and counted. In some cases, TCA-precipitable radioactivity from washed nuclei was also counted. Differently from uninfected cells, where cytoplasmic incorporation of nucleotide was negligible, infected cells showed a marked increase in thymidine incorporation, starting about 4 hr p.i. and reaching a maximum at 10 hr p.i. (Fig. 2). Consistently with the effect on virus yield, both phosphonoacetic acid and cytosine arabinoside inhibited almost completely viral DNA replication (Fig. 2B). We tested two concentrations of aphidicolin, 60 pM and 120 p/1/1.Addition of aphidicolin at these concentrations reduced thymidine incorporation in the
30 25 m b ? x E 8
2015lo50 -. 0 2
4
6
6 10 12
I
2
4
6
6 10 12
hr p.i. FIG. 2. Time course of viral DNA synthesis in the cytoplasm of infected cells, assayed by measuring thymidine incorporation into TCA-insoluble material. Mock-infected cells (open circles) and infected cells (solid circles) were pulse-labeled for 15 min with 10 pCil ml methyl-[3H]thymidine. The cells were extracted and the cytoplasmic fraction was precipitated with TCA and counted. Inhibitors were added since the beginning of infection at the following concentrations: 60 pM (A, solid squares) or 120 PM (A, solid diamonds) aphidicolin. 0.8 mM phosphonoacetic acid (6, solid squares), and 20 @M cytosine arabinoside (6, solid diamonds). Addition of aphidicolin (A) and phosphonoacetic acid (B) was also done at 2 hr p.i. (open squares) or 4 hr p.i. (open diamonds).
SHORT COMMUNICATIONS
500
0
50
100
150 0 200 400 600 800
Aphidicolin, PM
PM
PM
FIG. 3. Effect of aphidicolin and phosphonoacetic acid (PAA) on DNA polymerase activity of ASFV. The activity of cytoplasmic extracts from infected (solid lines) or mock-infected cells (broken lines) was assayed by using activated calf thymus DNA in reaction mixtures containing the indicated concentrations of aphidicolin or phosphonoacetic acid. In a different experiment, the effect of aphidicolin was determined in reaction mixtures containing 2.5 FM (triangles), 5 gM (diamonds), or 100 pM (squares) dCTP. DNA polymerase activity is indicated as pmol incorporated dATP per min per milligram protein in the extract.
nucleus of infected or uninfected cells to 0.05% in average, but had no noticeable effect on virus growth, as described above. Surprisingly, we observed that both doses of aphidicolin caused a reduction of about 50% in thymidine incorporation into cytoplasmic viral DNA when the drug was added at 0 hr p.i. (Fig. 2A). This contradiction with the lack of effect on virus growth might be explained if we presume that ASFV is replicated in large excess and so a 50% reduction in viral DNA replication is not limitative to virus production. To determine whether the reduction in DNA synthesis was actually due to partial inhibition of ASFV DNA polymerase activity, we tested the effect of aphidicolin on the enzyme activity in vitro. Cells were infected at 20 PFU per cell and cytoplasmic extracts were prepared at 10 hr p.i. as described above. Reaction mixtures for DNA polymerase assays contained 10 ~1 cytoplasmic extract from 10” cells, 25 pg activated calf thymus DNA, and 2 &i [32P]dATP (800 Ci/mmol) in 100 ~1reaction buffer: 50 mMTris-HCI, pH 7.5,50 mM KCI, 10 mM MgCI,, 1 rnlLl dithiothreitol, 100 PM each dCTP, dGTP, and dTTP, and 20 PM dATP. Inhibitors were added at different concentrations. The mixtures were incubated for 45 min at 37” and insoluble radioactivity was counted. The results depicted on Fig. 3 confirm that ASFV DNA polymerase is indeed fully resistant to aphidicolin at concentrations up to 150 pm. In contrast, the enzyme activity in vitro was drastically inhibited by phosphonoacetic acid. In a first experiment (Fig. 3, solid circles) using dATP as the labeled nucleoside triphos-
phate, we used the same concentration of 100 PM for all other three dNTP. This could have masked an inhibitory effect of aphidicolin, because this drug inhibits DNA polymerase activity by competing with dCTP (16). To rule out this possible source of error, we also performed the experiment using much lower concentrations of dCTP, in the order of the in vivo concentration (about 3 @I). Again, the enzyme activity was completely resistant to aphidicolin when assayed in mixtures containing 2.5 PM or 5 &I dCTP. These results seem to exclude the hypothesis that we discussed above for explaining the effect of aphidicolin on thymidine incorporation into viral replicative DNA. It was recently reported that ASFV DNA replication occurs in two phases (17). Whereas most of viral DNA replication takes place in the cytoplasm of infected cells, with a peak at about 10 hr p.i., an earlier step in DNA replication occurs in the nucleus of all or some infected cells. If this phase of viral DNA replication is catalyzed by an aphidicolin-sensitive polymerase and provides templates for the cytoplasmic DNA replication, then the overall viral DNA replication would be affected, since we added the drug immediately after the end of virus adsorption. This explanation is supported by the fact that the degree of apparent inhibition of viral DNA replication in the cytoplasm is dependent on the time of aphidicolin addition. Instead of having aphidicolin present from the beginning of infection, we added the drug at 2 or 4 hr p.i., times when cytoplasmic DNA replication is still at very low levels. In our system, these are the times of start and end of a nuclear minor phase of viral DNA replication (data not shown). As a control, we performed similar tests using phosphonoacetic acid. In this case, addition of the inhibitor at 2 or 4 hr p.i. resulted in the same level of inhibition as the one observed when the drug was added at the beginning of infection (Fig. 2B). Differently, when we added 120 &I aphidicolin to infected cells at 2 hr p.i. the maximum incorporation of tritiated thymidine was intermediate between the peaks observed in the absence of the drug and in the presence of aphidicolin from the beginning of infection (Fig. 2A). The same concentration of aphidicolin added at 4 hr p.i. had practically no effect on the maximum of thymidine incorporation, as compared to the control in the absence of aphidicolin. We also analyzed the effect of aphidicolin on DNA polymerase activity of extracts from cells infected with three independent ASFV mutants resistant to phosphonoacetic acid. One of the mutants is a spontaneous mutant obtained after serial passages of the virus in the presence of phosphonoacetic acid and the other two mutants were isolated from a virus stock mutagenized with nitrosoguanidine. DNA polymerase activity of
SHORT COMMUNICATIONS
all the three cytoplasmic extracts, assayed as described above, was completely resistant to aphidicolin and was practically of the same level as the one of wild type virus. As far as we know, ASFV is the only animal DNA virus whose DNA polymerase is resistant to aphidicolin. Since the enzyme is sensitive to phosphonoacetic acid and cytosine arabinoside, this is an exception to the correlation between the action of the three inhibitors, which apparently act at the same site of a-like eukaryotic and viral DNA polymerases. Our results also suggest that the two phases of ASFV DNA replication may be catalyzed by two different DNA polymerases, the first nuclear phase being dependent on an aphidicolin-sensitive polymerase, possibly a cellular enzyme. Further experiments are in course to identify and characterize the enzyme involved in ASFV DNA replication at the nucleus of infected cells. REFERENCES 1. SUGINO. A., and NAKAYAMA, K., Proc. Nat/ Acad. Sci. USA 77, 7049-7053 (1980). 2. PEDRALI-NOY, G., and SPADARI, S., J. Viral. 36, 457-464 (1980).
501
3. Dc~occ~o, R. A., CHADHA, K., and SRIVASTAVA,B. I. S., Biochim. Biophys. Acfa 609, 224-231 (1980). 4. MIKHAILOV, V. S., MARLYEV, K. A.. ATAEVA, J. O., KULLYEV,P. K., and ATRAZHEV, A. M., Nucleic Acids Res. 14, 3841-3857 (1986). 5. FOSTER, D. A., HANTZOPOULOS, P., and ZUBAY, G., J. Viral. 43, 679-686 (1982). 6. BUNCO, L., and SALAS, M.. Virology 153, 179-187 (1986). 7. OFFENSPERGER,W.-B., WALTER, E., OFFENSPERGER,S., ZESCHNIGK, C., BLUM, H. E., and GEROK,W., Virology 164, 48-54 (1988). 8. ITO, J., and BRAITHWAITE,D. K., Nucleic Acids Res. 19, 40454057 (1991). 9. EARL, P. L., JONES, E. V., and Moss, B., Proc. Natl. Acad. Sci. USA 83,3659-3663 (1986). 10. TSURUMI, T., MAENO, K., and NISHIYAMA, Y., J. Viral. 61, 388-394 (1987). Il. DEFILIPPES,F. M., 1. l/ire/. 63, 4060-4063 (1989). 12. TADIE, J. A., and TRAKTMAN, P., J. L&o/. 65, 869-879 (1991). 13. MATSUMOTO, K., KIM, C. I., KOBAYASHI, H., KANEHIRO, H., and HIROKAWA,H., Virology 178, 337-339 (1990). 14. COSTA, J. V., In “Molecular Biology of Iridoviruses” (G. Darai, Ed.), pp. 247-270. Kluwer Academic, Boston, 1990. 15. MORENO, M. A., CARRASCOSA,A. L., ORTIN, J., and VIWUELA,E., J. Gen. Viral. 39, 253-258 (1978). 16. KROKAN, H., WIST. E., and KROKAN, R. H.. Nucleic Acids Res. 9, 4709-4719 (1981). 17. GARCIA-BEATO, R., SALAS, M. L., VIITIUELA,E.. and SALAS, J., Viralogy 188, 637-649 (1992).
