Herpes simplex virus type 1 mutants for the origin-binding protein induce DNA amplification in the absence of viral replication

VIROLOGY

179,478-481

(1990)

Herpes Simplex Virus Type 1 Mutants for the Origin-Binding Protein DNA Amplification in the Absence of Viral Replication

Induce

REGINE HEILBRONN,’ SANDRA K. WELLER,* AND HARALDZUR HAUSEN Deutsches Krebsforschungszentrum, and *Department of Microbiology,

Im Neuenheimer Feld 280, D-6900 Heidelberg, Federal Republic of Germany; University of Connecticut Health Center, Farmington. Connecticut 06032

Received Herpes of seven gene we negative relevant

January

22, 1990; accepted

simplex virus (HSV) induces DNA amplification within the host cell genome, which is mediated by a set of six HSV replication genes. The origin-binding protein (UL9) is dispensible. By the use of HSV mutants for the UL9 show here that HSV can induce DNA amplification in the absence of lytic viral growth in contrast to replicationmutants for either the UL8 or UL52 gene used as control. The amplification-inducing potential of HSV may be for the pathogenicity of the virus. 0 1990Academic PWSS. IIIC.

Herpes simplex virus (HSV) infection can induce a variety of effects on the host cell genome including various types of chromosomal aberrations, DNA repair, mutations, and DNA amplification (7-10). The relevance of these HSV-induced genotoxic effects for the pathogenicity of the virus in viva, however, is still unclear for several reasons: Until now it has not been known whether HSV-induced genotoxic effects can occur in the absence of lytic viral growth, which of course would be a prerequisite for the persistence of any HSVinduced genetic alteration. With the recent identification of the HSV genes mediating DNA amplification, this question can now be tackled. A subset of HSV genes, namely the UL5, -8, -42, -52, major DNA-binding protein (UL29), and DNA polymerase (UL30) genes, are necessary and sufficient to induce DNA amplification of integrated SV40 DNA sequences upon cotransfection into SV40-transformed hamster cells (10). All these genes code for replication functions; however, an additional replication gene coding for the originbinding protein (UL9) (12- 14) is dispensible for DNA amplification (10). These data suggested that a set of genes not sufficient for lytic viral growth can induce DNA amplification. To provide more direct evidence that these data from transfection experiments reflect the biological behavior of the virus, we made use of HSV mutants with a disrupted UL9 gene which express the amplification-inducing HSV genes in the context of authentic early viral gene expression. Due to the defect in the origin-binding protein (UL9) the mutants cannot replicate; therefore, DNA amplification can be studied in the absence of viral replication.

’ To whom 0042-6822190

July 3, 1990

requests

for reprints

$3.00

CopyrIght 0 1990 by Academic Press. Inc. All rights of reproduction I” any form reserved.

should

Two well-characterized HSV host range mutants with a defective UL9 gene (hr27 and hr48) were tested to amplify chromosomally integrated SV40 DNA sequences in SV40-transformed hamster cells. Since hr27 and hr48 (Fig. 1) cannot grow in Vero cells, a helper cell line constitutively expressing the defective gene in vans is needed to propagate the viruses (74). The two mutants and wild-type HSV (strain KOS) were used to infect the SV40-transformed hamster cell line Elonal 1, which has previously been shown to amplify the integrated SV40 DNA sequences over lOO-fold upon infection with different HSV wild-type strains (8, 9). Cells were harvested for Southern blot analysis at 40 hr p.i. Genomic DNAs were digested with Sacl (noncut enzyme for SV40 DNA) and hybridized to 32P-labeled SV40 DNA (Fig. 2A). Integrated single-copy SV40 DNA sequences correspond to the 15-kb band present in all lanes. It is evident that the ULS-defective mutants hr27 and hr48 can induce DNA amplification equally well as wild-type virus (KOS) (Fig. 2A). To confirm that hr27 and hr48 are defective in viral DNA replication under the conditions used, replication of a transfected or&-containing plasmid was tested in parallel using the Dpnl assay. Dpnl will digest the transfected plasmid DNA which has been methylated during procaryotic replication, whereas transfected plasmids replicated by eucaryotic cells will not be methylated and thus become Dpnl resistent. Cells were transfected with 1 pg of pH 10 (carrying or&), and 6 hr later infected with the respective virus (MOI 1). Total genomic DNA was extracted at 40 hr p.i. for Southern blot analysis. EcoRIl HindllIIDpnl-digested DNA was probed with a 32P-labeled oris subfragment. As can be seen in Fig. 2B there is high level oris replication after infection with wild-type virus (KOS) but no (hr27), or low level (hr48) replication

be addressed 478

SHORT

COMMUNICATIONS

HSV-1

UL

GENES

us

c* dbp

PLASMIDS

479

HI’ pcmUL5

pal

k-lH pm-ULB

tA

pewDBP

*

*

UL42

UL52

H pm-POL

pCM-UL42

H pCM-UL52

I

DCM-UL9

MUTANTS

FIG. 1. Localization of the HSV replication/amplification genes on the HSV-1 genome. Schematrc representatron of the 153.kb HSV-1 genome with the unrque long (UJ and unique short (U,) regions and flanking inverted repeats (black boxes for the long region; lrght boxes for the short region). The replication/amplification inducing genes are indicated below: 1115/-8/-g/-42/-52; dbp, DNA binding protein; pol, DNA polymerase. The arrows indicate the direction of transcription. Black arrows are genes necessary for both HSV replication and SV40 DNA amplification (UL5, UL8, pot, dbp, UL42, UL52); the stippled arrow indicates UL9, which is the only additional gene required for HSV DNA replication. The relative positions of cloned subfragments covering individual open reading frames under the control of the human cytomegalovirus (HCMV) promotor are given below. The construction of the pCM-UL series has been described (10). The approximate map positions of the HSV host range mutants hr80 (UL8), hr27/hr48/hr94 (UL9), and hrl 14 (UL52) are given below. Their isolation has been described previously (74- 16; S. K. Weller manuscript in preparation). hr27 and 48 are spontaneous host range mutants for the UL9 gene, whereas hr80, hrl 14, and hr94 have been constructed by insertion of the laci’gene under the control of the ICP6 promotor into the respective open reading frame.

of oris after infection with the UL9 mutants. High levels of oris replication, however, can be seen after cotransfection of the defective gene (UL9) expressed underthe control of the human cytomegalovirus (HCMV) immediate-early promotor (pCM-ULS). The low level of oris replication in hr48 and hr27 (visible only on very long exposures) in the absence of transfected pCM-UL9 was unexpected, since these mutants had been plaque-purified several times and been shown before to be deficient in viral DNA synthesis (14). We therefore assume that these mutants tend to revert at a relatively high frequency. Recently, the precise defects of another spontaneous UL9 mutant have been defined by nucleic acid sequencing. The mutant hr156 isolated at the same time as hr48 and hr27 (14) contains two point mutations. It is presumed that one or both are responsible for the growth phenotype of the virus. The high reversion rates of the UL9 mutants used here, hr48 and hr27, suggest that the defects may be due to single point mutations (E. P. Carmichael and S. K. Weller, unpublished data).

To circumvent problems with revertants of the UL9 mutants we decided to use HSV null-mutants generated by insertional mutagenesis using an ICP6::lacZ gene cassette. A mutant for the UL9 gene has been isolated which exhibits a clear replication-negative phenotype (S. K. Weller, manuscript in preparation). Two other similarly constructed replication-negative null-mutants for the UL8 and UL52 gene, respectively (15, 16). were used as contols. Elonal 1 cells were infected with these three replication-negative HSV mutants (MO1 1) 6 hr after transfection of 1 pg oris plasmid (pHl0) with or without the HCMV-IE promoter-driven expression construct (2 pg) to complement for the defective viral gene as indicated in Fig. 3. Genomic DNA was extracted for Southern blot analysis at 40 hr p.i. &cl-digested DNA was probed with 32P-labeled SV40 DNA to assay for DNA amplification. In parallel, EcoRIl HindllllDpnl-digested DNA was probed with 32P-labeled oris subfragment to assay for oris replication as described above. The negative control, mock-infected cells transfected with pH10 and vector DNA (Blue-

480

SHORT

COMMUNICATIONS

A

23.1 kb -

S,4kb

-

B virus pCM-ULS

oris

-

FIG. 2. SV40 DNA amplification and oris replication induced by the spontaneous UL9 mutants hr27 and hr49. (A) Elonal 1 cells (1 X 1 OS) were either mock-infected or infected with the respective viruses (MO1 1) as indicated. (B) Cells were transfected with a combination of 1 pg pH 10 (carrying on,) with or without 2 & pCM-UL9 and 6 hr later infected as indicated. Genomic DNA was extracted at 40 hr p.i. for Southern blot analysis. For the analysis of SV40 DNA amplification (A) &cl-digested DNA was probed with 32P-labeled SV40 DNA. For the analysis of oris replication (B) a Dpnl assay was performed with genomic DNA of Elonal 1 cells after transfection with an or&carrying plasmid (pH10) and subsequent virus infection. EcoRIIHindlll/Donl-digested DNA was probed with a 3zP-labeled oris subfragment as described (IO).

script, Stratagene) showed neither DNA amplification nor oris replication (Fig. 3, lane 1). The positive control, cells transfected with the entire set of HCMV-driven HSV replication genes (pCM-UL5, -UL8, -UL9, -dbp, -poi, -UL42, -UL52), induced DNA amplification as well as replication of the cotransfected ori, plasmid (Fig. 3, lane 2) as described before (10). The two replicationnegative HSV mutants hr80 (UL8) and hrl14 (UL52) did not induce either DNA amplification or oris replication, unless the defective gene was supplied in trans by cotransfection of the respective HCMV expression plasmid (Fig. 3, lanes 3-6). In contrast, the ICP6::facZ insertion mutant for the ULQ gene (hr94) induced high levels of DNA amplification comparable to those of wild-type virus (KOS) (Fig. 3, lanes 7-9, and 1 O-l 2 as short exposure). However, oris replication was found only after transfection of pCM-UL9 to complement for the defective ori-binding protein. Even on very long exposures, no trace of oris replication is seen for hr94 in the absence of pCM-ULQ (data not shown). Recently HSV-induced episomal SV40 DNA replication was de-

scribed in BHK cells transfected with cloned SV40 DNA (18). In this system the DNA-negative HSV ts-mutant tsS induced episomal SV40 DNA replication in the absence of oris replication. The lesion of tsS has been mapped to a region between nucleotide positions 21,650-25,149 (19) which spans the coding regions of the ULS-ULI 2 genes. Of these only UL9 codes for a HSV replication gene (70, 12, 14). Since we have shown that UL9 is dispensible for SV40 DNA amplification (70), and that tsS shares the same complementation group as hr27 (14), it is likely that the lesion of tsS actually lies within the UL9 gene. From our experiments we can conclude that UL9 is not required for the induction of DNA amplification, whereas UL8 and UL52 are indispensible. These data fully confirm our previous report using transfection of combinations of isolated genes for the identification of the six HSV amplification genes (10). With the experiments described here we can extend our earlier observations to showing that the amplification effect induced upon cotransfection of the six HSV amplification genes reflects the biological behavior of the virus: In the context of authentic early viral gene expression the amplification genes behave like the isolated cloned genes.

A

virus

-

-

--hrS0

pm-UL

-

ell

-

8

hr114

-- hrS4

-

-

52

KOS a

-

--tlts4 s

Kos -

B oris

123456

7

6

s

10

11

12

FIG. 3. SV40 DNA amplification and or& replication induced by the ICP6::tacZ insertion mutants for UL8 (hr80), UL52 (hrl14), and UL9 (hr94). The experiment was performed as described in Fig. 2. The HCMV-IE promoter-driven HSV amplificatio~replication genes pCMUL8, Pam-UL52, and pCM-ULS were transfected as indicated to complement for the defective mutant virus phenotype. “All” designates the transfection of the complete set of HCMV-driven HSV replication genes (pCM-UL5, -UL8, -UL9, -dpb, -pal. -UL42, UL52). Hr94 and KOS induce a very strong amplification effect. The short exposure (2 hr versus 5 days; lanes 1 O-l 2) reveals a comparable extent of SV40 DNA amplification induced by hr94 and KOS. The relatively weak amplification effect of hr80 and hrl14 upon complementation of the defective gene by transfection of pCM-ULB or pCM-UL52, respectively (lanes 4 and 6) is due to the lower efficiency of transfection versus infection.

SHORT

COMMUNICATIONS

Similar results were obtained with antiviral drugs and ts-mutants for the HSV DNA polymerase (8) and a nullmutant for the major DNA-binding protein (d21; described in Ref. 17; R. Heilbronn, unpublished data). We assume that null-mutants for UL42 and UL5 will behave similarly. The most important point of our experiments is the demonstration that HSV can provide gene functions sufficient for the induction of DNA amplificationand possibly other genotoxic effects-in the complete absence of HSV DNA replication and lytic viral growth. It will be interesting to test the UL9 mutants described here for other genotoxic effects in vitro and also in viva to study genotoxic effects during the lifelong virushost relationship. The possible separate expression of the amplification-inducing genes may eventually lead to the persistence of a cell with a modified genotype with possibly adverse consequences for the affected host.

ACKNOWLEDGMENTS We thank Dr. P. A. Schaffer for the HSV-mutant d21, Drs. M. Boshart and A. Bijrkle for stimulating discussion and critical reading of the manuscript, S. Stephan for excellent technical assistance, and M. Marquard for secretanal assistance during preparation of the manuscript. This work was supported by the Deutsche Forschungsgemeinschaft Grant He1 598/l -1.

REFERENCES 1. HAMPAR. (1961).

B., and ELLISON. S. A., Nature

(London)

192,

145-147

481

2. CHENET-MONTE, C., MOHAMMAD, F.. CELLUZZI, C. M., SCHAFFER, P. A., and FARBER, F. E., Virus Res. 6,245-260 (1986-l 987). 3. NISHIYAMA, Y., and RAPP, F.. Virology 110, 466-475 (1981). 4. SCHLEHOFER, J. R.. and ZUR HAUSEN, H., Virology 122, 471-475 (1982). 5. PILON. L., ROYAL, A., and LANGELIER, Y., J. Gen. Virol. 66, 259265 (1985). 6. PILON, L., LANGELIER, Y., and ROYAL, A., MO/. Cell. Biol. 6, 2977.. 2983 (1986). 7. SCHLEHOFER, J. R., GISSMANN, L., MATL, B., and ZUR HAUSEN, H., hr. 1. Cancer 32, 99-l 03 (1983). 8. MATZ, B.. SCHLEHOFER, J. R., and ZUR HAUSEN, H., Virology 134, 328-337 (1984). 9. MAR. B.,/. Viral. 61, 1427-1434 (1987). 10. HEILBRONN, R., and ZUR HAUSEN. H., 1. Viral. 63, 3683-3692 (1989). 11. CHALLBERG, M.D., froc. Nat/. Acad. Sci. USA 83, 9094-9098 (1986). 12. Wu, C. A., NELSON, N. J., MCGEOCH, D. J., and CHALLBERG, M. D., J. Viral. 62, 435-443 (1988). 13. Ouvo, P. D., NELSON, N. J., and CHALLBERG, M.D., Proc. Nat/. Acad. Sci. USA 85,5414-54 18 (1988). 14. CARMICHAEL, E. P.. KOSOVSKY. M. J., and WELLER, S. K., J. Viral. 62, 91-99 (1988). 15. GOLDSTEIN, D. J., and WELLER, S. K., J. Viral. 62, 2970-2977 (1988). 16. CARMICHAEL, E. P., and WELLER, S. K.. /. Viral. 63, 591-599 (1989). 17. ORBERG, P. K., and SCHAFFER P. A., /. Viral. 61, 1136-l 146 (1987). 18. MATZ, B., f. Gen. Viral. 70, 1347-1358 (1989). 19. MATZ, B., SUBAK-SHARPE, 1. H., and PRESTON, V., I Gen. Viral. 64, 2261-2270(1983),