the ul8 component of the herpes simplex virus helicase-primase

THE JOURNAL OF BIOIJXICAL CHEMISTRY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 269, No. 7, Issue of February 18, pp. 5030-5035, 1994 Printed in U.S.A.

The UL8 Component of the Herpes Simplex Virus Helicase-Primase Complex Stimulates Primer Synthesis by a Subassembly of the UL5 and UL52 Components*

(Received for publication, August 30, 1993, and in revised form, October 25, 1993)

Daniel J. TenneyS, Warren W. Hurlburt, Pamela A Micheletti, Marc Bifano, and Robert K. Hamatake From the Department of Virology, Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, New Jersey 08543-4000

The herpes simplex virus type 1 (HSV) ULS, UL8, and ULS2 proteins form a helicase-primase complex in infected cells. Several laboratories have demonstrated that helicase and nucleoside triphosphatase activities of the heterotrimer (ULS/8/62) are indistinguishable from that of a subassembly of ULS and ULS2 (ULS/S2). Al- though the ULS/S2 subassembly functions in coupled primase-polymerase assays on homopolymeric templates, its activity on natural DNA templates has been reported to require UL8. To determine the role of UL8 in primase assays, the activity of the U L S I S 2 subassembly was compared to that of the heterotrimer reconstituted by adding UL8 to UL5/52. We detected significant activity of the UL5/52 subassembly in coupled primase- polymerase and oligoribonucleotide primer synthesis assays on 4x174 and M13 virion DNAs. However the addition of UL8 to ULS/S2 stimulated this activity in a dose- dependent manner. We demonstrate that stimulation occurred at the level of primer synthesis. UL8 did not affect the amount or size of primers annealed to template, their utilization by DNA polymerase, or the use of specific initiation sites within the template. In kinetic studies, the rate of primer synthesis was increased by UL8 but the K, for 4x174 DNA template was unchanged. These results suggest that a function of the UL8 component of the HSV helicase-primase complex is to increase the efficiency of primer synthesis by UL5/52.

The replication of herpes simplex virus type 1 (HSV)’ DNA within host cells utilizes three origins of replication provided in cis and requires seven virally encoded proteins provided in trans (reviewed in Refs. 1 and 2). Studies of the activities of these proteins in vitro have revealed that their roles parallel those of other well defined DNAreplication proteins of prokaryotic and eukaryotic systems. The proteins include a processive DNA polymerase consisting of a catalytic subunit and an ac- cessory factor (Pol/uL42); a single-stranded DNA-binding protein (ICP8); a sequence-specific origin-binding protein (UL9); and a helicase-primase complex composed of three proteins (UL5, UL8, and UL52; reviewed in Refs. 1 and 2).

payment of page charges. This article must therefore be hereby marked * The costs of publication of this article were defrayed in part by the

“aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

f To whom corresDondence should be addressed. ”el.: 609-252-4368: FA: 609-252-6058.

The abbreviations used are: HSV, herpes simplex virus type 1; NTP, nucleoside 5’-triphosphate; NTPase, nucleoside triphosphatase; Pol, HSV DNA polymerase; TBE, Tris borate-EDTA, PAGE, polyacrylamide gel electrophoresis.

The subunits of the HSV helicase-primase complex were shown to be essential for HSV DNA replication using null mu- tant viruses (3-5) and transient assays of replication of HSV origin containing DNA (6-8). The complex was first purified from HSV-infected cells, identified as the UL5, UL8, and UL52 proteins, and shown to have helicase, DNA-dependent ATPase/ GTPase, and primase activities (9-11). Purification of the enzyme from recombinant baculovirus-infected insect cells en- abled further characterization of these activities (12-16). Several studies have found that a complex of the UL5/52 subunits (UL5/52 subassembly) was as active as the heterotrimer (UL5/8/52) in nucleoside triphosphatase (NTPase) and helicase activities, as well as in coupled primase-polymerase activity on poly(dT) DNA (13-16). Therefore the essential contribution of UL8 was not understood. Recently, UL8 was shown to be involved in the primase function of UL5/8/52 on natural DNA templates (16) although the mechanism remained unclear. UL8 might also play a role in nuclear localization of the complex as only the heterotrimer, not the UL5/52 subassembly or indi- vidual subunits, is capable of localizing to the nucleus of infected cells (17).

We sought to examine the role of UL8 in primase activity of the UL5/8/52 complex in vitro by comparing the activity of the UL5/52 subassembly with that of the reconstituted heterotrimer. We found that the UL5/52 subassembly alone was capable of enzyme concentration-dependent primase activity on the natural DNA template. However, comparison of the activity of the UL5/52 subassembly with that of the heterotrimer, reconstituted by the addition of UL8, revealed that the rate of primer synthesis is increased by UL8. UL8 did not affect several other parameters of primase activity: the size of primers synthesized, the annealing of primers to template DNA, the utilization of primers by DNA polymerase, the preferential sites of primer synthesis on the template, the affinity of the complex for template DNA, or the stability of the primase complex.

EXPERIMENTAL PROCEDURES

the HSV-1 UL5, UL8, and UL52 (12) and UL30 (Ref. 18; Pol) proteins Recombinant Baculouiruses-Recombinant baculoviruses expressing

were a generous gift of I. Robert Lehman. Construction of the recombinant baculovirus expressing HSV UL42 is described elsewhere (19,20). The recombinant baculovirus expressing HSV ICP8 was isolated after subcloning the ICP8 open reading frame contained within HSV nucleotides 56,77342,422 (21) into a baculovirus transfer vector (pBlue- bacIII) and standard co-transfection methods using a commercial kit (Invitrogen Corp., San Diego, CA). The HSV ICP8 open reading frame was subcloned from a vector used to express ICP8 in mammalian cells (pSV8.3) kindly provided by Min Gao (22).

Purification of Recombinant Proteins-Purification of the recombinant Pol.UL42 complex was as described (18). Apublished procedure for purifying the UL5/52 complex was followed up to the heparin-agarose column (12). The complex was then further purified by chromatography

5030

A Novel Function of the UL8 Subunit of the HSV Helicase-Primase 503 1

on single-stranded DNA cellulose and Q-Sepharose. UL8 was purified from cytosolic extracts, prepared as described (12), by chromatography on Q-Sepharose, phenyl-Sepharose, and Superose 12 columns. ICP8 was purified from nuclear extracts, prepared as described (18), by chromatography on S-Sepharose, ammonium sulfate precipitation with 30% saturated ammonium sulfate, and gel filtration chromatography on Superose 12. Protein concentrations were determined by absorbance measurements at 280 nm using calculated extinction coefficients (23) based on the predicted amino acid sequences of the proteins. Proteins were stored at -80 "C in buffer containing 20 m Tris-HC1 (pH 8.0), 10% (w/v) glycerol, 1 m EDTA, 10 m 2-mercaptoethano1, 1 l ~ l ~ phenyl- methylsulfonyl fluoride, and 50-300 m NaCl.

Coupled Primme-Polymerase Assays-Primase-dependent HSV Pol activity was assayed using single-stranded +X174 virion DNA (New England Biolabs). Reactions contained 50 II~M NaC1, 50 m Tris-HC1 (pH 8.0), 1 nw dithiothreitol, 200 pg/d acetylated bovine serum albumin, 10% (w/v) glycerol, 1 m ATP, 1 m GTP, 0.1 m CTP, 0.1 nw UTP, 50 w dATP, 50 pm dCTP, 50 w dGTP, 5 w dWP, 0.22 pCi of L3Hld'ITP (78 Ci/mmol, DuPont-New England Nuclear), 4 mrd MgClZ, 50 ng of +X174 virion DNA, Pol.UL42, ICP8, UL5/52, and, as indicated, UL8, in a total volume of 50 pl. Reactions were incubated at 30 "C for 2 h, and were terminated by the addition of trichloroacetic acid to 10%. and harvested for scintillation counting as described (19, 24). The back- ground level due to nuclease activity (in the absence of rNTPs) was routinely below 15% (data not shown).

Direct Oligoribonucleotide Primer Synthesis Assay--Radiolabeled oligoribonucleotide primers were synthesized in reactions containing 50 nw NaC1, 50 m~ Tris-HC1 (pH 8.0), 1 m dithiothreitol, 200 pg/ml acetylated bovine serum albumin, 10% (w/v) glycerol, 1 m ATP, 1 m GTP, 0.1 m CTP, 4 m MgCl,, 0.04 pCi/pl [&"PIUTP (DuPont-New England Nuclear; 3000 Ci/mmol; 10 pCi/pl), and +X174 or M13 virion DNA (30 pg/ml). Reactions were incubated for 60 min at 30 "C and terminated by adding formamide to 50% (for denaturing gels) or DNA loading buffer containing glycerol (25) with SDS added for a final concentration of 1% (for nondenaturing gels). Samples were assayed im- mediately (usually 5 pl of original reactions) or stored at -80 "C.

Direct examination of oligoribonucleotide primers used electrophoresis in denaturing 20% polyacrylamide, 7 M urea, Tris borate-EDTA (TBE) gels (25) (0.4 mm x 31 cm x 38.5 cm) at 55-65 watts for 75-90 min. Samples were heated to 70-80 "C for 5 min prior to loading. Size markers consisted of poly(A) (Pharmacia) radiolabeled using poly- nucleotide kinase and [-y-32PJATP (DuPont-New England Nuclear; 6000 Ci/mmol) and partially digested with nuclease U, (U. S. Biochemical Corp.). Following electrophoresis, gels were removed onto x-ray film backing and subjected to autoradiography at -80 "C.

Kinetics of primer synthesis were analyzed in the presence of various concentrations of +X174 virion DNAand 1.88 pmol of UL5/52 without or with 5.6 pmol of UL8. One-half of reaction products was directly analyzed as described above. Production of the predominant oligoribonucleotide primer was compared for quantitative purposes: the synthesis of additional smaller products was consistently proportional. Autoradiographs were scanned using an LKB scanning laser densitom- eter and analyzed using Gelscan XL soRware (Pharmacia LKB Biotech- nology Inc.). The analysis of kinetic data to determine V,, and K , was performed using the KinetAsyst I1 software (IntelliKinetics, Inc.) which fit the amount of primers synthesized ( u ) to the equation u = V,, * SKK, + S).

Visualization of radiolabeled primers annealed to single-stranded +X174 DNA was accomplished by electrophoresis of products in nondenaturing agarose TBE gels containing 1 pg/ml ethidium bromide. Gels were rinsed briefly in HZO and dried under vacuum, followed by autoradiography. To determine the size of primers annealed to single- stranded DNA template, annealed primer-template complexes were excised from gels and eluted from the agarose, followed by ethanol precipitation, resuspension in 50% formamide, and electrophoresis in denaturing 20% PAGE gels, as described above.

DNA Polymerase Extension of Radiolabeled RNA Primers- Following primer synthesis, dNTPs were added to 50 p each along with additional MgClz to 200 w for Pol.UL42 or to approximately 10 m final for modified T7 DNA polymerase (Sequenase; U. S. Biochemi- cal Corp.). ICP8 was added to reactions containing Pol.UL42 to 1 moleculdl7 nucleotides of DNA, approximately the concentration that optimally stimulates polymerase activity (18). Incubation was carried out at 37 "C for 60 min. Reactions were terminated as above for primer synthesis reactions or directly analyzed by restriction enzyme digestion using standard protocols. After restriction enzyme digestion, products were analyzed by electrophoresis in nondenaturing 4% PAGE TBE gels, followed by autoradiography.

0.0 1.0 2.0 3.0

pmoles mu52

20 -

15 .

10 -

51

0 - 0 5 10 15

molar ratio of ULB to UL5/52

FIG. 1. Activity of the ULW52 subassembly and the effect of uL8 in the coupled phase-polymerase assay. Coupled primase- polymerase assays were performed using 50 ng of +X174 virion DNA (28.6 fmol of circular DNA molecules), 40 fmol of Pol.UL42,900 fmol of ICP8, and [3Hldl'TP, as described under "Experimental Procedures."A, coupled primase-polymerase activity with increasing levels of UL5/52. B , titration of UL8 into coupled primase-polymerase assays containing 190 fmol of UL5/52.

The Stability of the Helicase-Primase under Reaction Conditions- 170 fmol of UL5/52 without or with 600 fmol of UL8 was combined with 1 m ATP and 50 ng of +X174 DNA or with no ATP or DNA substrates in buffer containing 50 IILM Tris-HC1 (pH 8), 10% glycerol, 500 pg/ml bovine serum albumin, and 1 m dithiothreitol. The mixtures were heated at 30 "C for the various times and then transferred to an ice water bath. The amount of UL5/52 primase activity remaining after the heat treatment was determined by the coupled primase-polymerase assay after addition of the reaction mixture and the protein components necessary.

RESULTS

UL8 Reconstitution of UL5f 52 Primase Subassembly Results in Increased Activity-Previous reports suggested that the UL5/52 subassembly of the helicase-primase complex retained the full NTPase and helicase activities of the UL5/8/52 heterotrimer (13-16). Additionally, coupled primase-polymerase activity using homopolymer DNA template was unaffected by UL8 reconstitution of UL5/52 (14, 16) but activity on M13 bac- teriophage DNA required the inclusion of UL8 (16). We tested the requirement for UL8 in the coupled primase-polymerase assay using single-stranded 4x174 virion DNA as template. UL5/52 exhibited significant levels of activity in a concentration-dependent manner in the absence of UL8 (Fig. lA). This activity was stimulated by reconstitution of the UL5/8/52 heterotrimer upon addition of increasing amounts of UL8 (Fig. 1B). The activity of UL5/52 and its stimulation by UL8 was seen whether HSV Pol (Fig. 1) or modified T7 DNA polymerase (Sequenase; U. S. Biochemical Corp.) was used to extend primers made by HSV helicase-primase (data not shown). Agarose gel electrophoresis of reaction products confirmed that the amounts of d lTP incorporated reflected the amount of single- stranded DNA converted to double-stranded forms (data not shown). The activity of the UL5/52 subassembly in coupled

5032 A Novel Function of the UL8 Subunit of the HSV Helicase-Primase

A B C UL5/52 - 0.6 1.7 5 15 0.61.7 5 15 5 5 5 5 5 5

U L 8 - - - _ _ 1.7 5 15 45 - 0.5 1.5 5 15 50

d 20 -

15 -

-a

10-

1 2 3 4 5 6 7 8 Y 10 11 12 13 14 15

FIG. 2. The effect of UL8 on primer synthesis by the UL5/52 subassembly. Products of primer synthesis assay resolved on 20% denaturing PAGE, as described under “Experimental Procedures.” Re- actions contained increasing amounts (in picomoles) of UL5/52 without (A) or with ( B ) a 3-fold molar excess of UL8. In C, UL8 was titrated into reactions containing a constant amount (5 pmol) of UL5/52. 5 pl of the reaction products (50 pl total) were analyzed. The migration of the predominant oligoribonucleotide primer is indicated with an arrow at the right. The migration of size markers consisting of y-”P-labeled poly(A), partially digested with U2 nuclease, is indicated by their sizes in nucleotides (nth

primase-polymerase assays and its stimulation by UL8 were not limited to 4x174 DNA, M13 virion DNA also served as a suitable template (data not shown). Therefore, on natural DNA templates, the UL5/52 subassembly demonstrated significant primase activity that can be augmented by the addition of UL8.

UL8 Stimulates Primer Synthesis by the UL5/52 Subassem- bly-To determine the level at which coupled primase-polymerase activity was stimulated by UL8, we tested the effect of UL8 on the synthesis of oligoribonucleotide primers by UL5/52. The UL5/52 subassembly displayed an enzyme concentration-dependent increase in primer synthesis on the 4x174 template (Fig. 2A) . Using [cY-~~PIUTP to label oligoribonucleotide primers, we found that the most abundant primers synthesized by the HSV primase on 4x174 virion DNA migrated between the 12- and 13-nucleotide bands of the poly(A)/U2-digested size markers. Low amounts of smaller products migrating from the 6 to 11-nucleotide length markers were also produced. These sizes are in agreement with those reported previously (11) using a combination of [CY-~~PIUTP and [cY-~~PICTP for labeling.

The levels of primer synthesis by the UL5/52 subassembly were increased further upon reconstituting the heterotrimer by the addition of UL8 (Fig. 2 B , 3-fold molar excess UL8 to UL5/ 52). UL8 affected the amount but not the size or distribution of primers synthesized. At 3-fold molar excess of UL8 to UL5/52, synthesis of the predominant oligoribonucleotide primer was stimulated approximately 3-fold. Titration of UL8 into the UL5/52 subassembly (Fig. 2C) demonstrated that UL8 en- hancement of primer synthesis was detectable when UL8 concentrations were as low as one-third that of UL5/52. Stimula- tion leveled off after addition of a 3-fold molar excess of UL8 to UL5/52: addition of 50-fold excess UL8 to UL5/52 resulted in a 3.5-fold increase in activity (data not shown). UL8 also stimulated primer synthesis when M13 virion DNA template was used (data not shown). We conclude that the UL8 protein

stimulates the synthesis of oligoribonucleotide primers by the UL5/52 subassembly.

Near maximal levels of primers were detected after addition of a %fold molar excess of UL8 to UL5/52. Fractionation using a Superose-12 gel filtration column revealed that a combination of UL5/52 mixed with a 3-fold molar excess of UL8 resulted in virtually complete reconstitution of the heterotrimer (data not shown). This correlation of functional and physical data is consistent with the hypothesis that stimulation of primer synthesis results from reconstitution of the UL5/8/52 heterotrimer by the association of UL8 with the UL5/52 subassembly. A 3-fold molar excess of UL8 to UL5/52 was used in subsequent experiments to achieve near maximal reconstitution.

Effect of UL8 on Primer Annealing and Extension by DNA Polymerase-Although primer synthesis is stimulated approximately %fold by UL8, greater levels of stimulation are observed upon titration of UL8 into the coupled primase-polymerase assay (Fig. 1B). Another potential mechanism for UL8 stimulation in the coupled primase-polymerase assay is that primers made by the UL5/8/52 heterotrimer were more efficiently utilized by DNA polymerase than those made by the UL5/52 subassembly. This difference could arise by changes in the level of primer annealing. We tested this hypothesis by comparing the amount of primers synthesized to the amount of primers annealed to template DNA and to the amount of primers extended by DNA polymerase.

Radiolabeled primers that remain annealed to template DNA after their synthesis migrate with the template in nondenaturing agarose gels. Extension of these primers by the addition of DNA polymerase and unlabeled dNTPs resulted in the formation of replicative form I1 and I11 DNA. We found that the amount of primers that remained annealed to template DNA (Fig. 3 A ) , and the amount extended by DNA polymerase (Fig. 3B) correlated directly with the amounts synthesized (Fig. 3C), regardless of whether the primers were made by UL5/52 or the reconstituted heterotrimer. Extension of the primers with HSV Pol.UL42 or with Sequenase yielded similar results: the results with Sequenase are presented in Fig. 3 because a single extension product is made while Pol.UL42 yielded several shorter extension products because of decreased processivity through regions of secondary Therefore, UL8 did not affect the annealing of primers to template DNA or their extension by DNA polymerase.

We also tested whether UL8 affected the amount of smaller primers extended by DNA polymerase (Fig. 4A) or annealed to template DNA (Fig. 4B). Extension of radiolabeled primers prior to separation in 20% denaturing PAGE revealed that the longer primers, migrating with the 12-mer markers and some migrating at markers 8 through 11 nucleotides, were extended by Pol.UL42 (Fig. 4A). Extension of smaller sized products was not observed in the presence or absence of UL8. The size of primers annealed to template DNA was determined by excision of primer:templates from agarose gels, elution, and ethanol precipitation, followed by denaturing 20% PAGE. The sizes of primers annealed to template DNA were the same with UL5/52 and the reconstituted heterotrimer (Fig. 4B). These results ruled out the possibility that UL8 facilitated the extension of smaller primers by DNA polymerase or their annealing to template DNA. Taken together, the results presented in Figs. 1-4 conclusively demonstrate that the stimulatory effect of UL8 in the coupled primase-polymerase assay is at the level of primer synthesis and not at the level of primer annealing to template or extension by DNA polymerase. UL8 Does Not Increase the Stability of the UL5152

* R. K. Hamatake, M. Bifano, W. W. Hurlburt, and D. J. Tenney, manuscript in preparation.

U L5/52 U L8

0x174)

U L5/52 U L8

RFII>

RFIII>

A Novel Function of the UL8 Subunit of the HSV Helicase-Primase 5033

A

- 1.7 5 15 1.7 5 15 5 5

45 - - - 5 15 45 - 15

4 M 1 3

1 2 3 4 5 6 7 0 9

A UL5/52 4 4 4 4 4 4

UL8 - - 4 4 12 12

PoVUL42 - + - + - +

-0" 1

B 15 -

- 1.7 5 15 1.7 5 15 15 15

45"- 5 1 5 4 5 - 4 5

1 2 3 4 5 6 7 8 9

C UL5/52 - 1.7 5 15 1.7 5 15

UL8 45 - - - 5 15 45 nl

20 :

15 -

10 -

5 -

1 2 3 4 5 6 7

extension producls

B UL5/52 UL5/52 +UL8

T A T A !x 20

15 15

- 45

15

<

10

8 9

FIG. 3. The effect of ULS on primer annealing and the ability to be extended by DNA polymerase. Primers were synthesized and analyzed as described under "Experimental Procedures" using enzyme amounts (in picomoles) indicated in 25(A) or 50-pl ( B and C) reaction volumes. Annealed primers (A, 5 pl) were visualized by electrophoresis in 0.8% agarose-TBE and autoradiography. Extended primers (B , 10 pl) were analyzed similarly after incubation with 0.26 units of Sequenase and dNTPs for 60 min a t 37 "C. Total primers synthesized (C , 5 pl) were visualized in 20% denaturing PAGE. Lanes 1-7 used +X174 virion DNA as template. Lanes 8 and 9 used M13 virion DNA as template.

Subassembly-One possible explanation for the UL8 stimulation of UL5/52 primase activity is that UL8 increases the stability of the UL5/52 subassembly at the reaction temperature of 30 "C. To investigate this possibility, we compared the activity of UL5/52 preincubated for various times at 30 "C in the pres-

5

: 2 ; 4

FIG. 4. The effect of ULS on the length of primers extended by DNA polymerase and annealed to template DNA. A, primers were synthesized on 4 pg/ml +X174 virion DNA using 4 pmol of UL5/52 f various amounts of UL8 as indicated in a 25-pl total volume. 5 pl of reactions was run directly (-PoWL42; lanes 1, 3, and 5) or after extension with 50 p~ dNTPs, 27 fmol of PoIAJL42, and 2.8 pmol of ICP8 (+PoVUL42; lanes 2, 4, and 6), as described under "Experimental Pro- cedures." The predominant primer is indicated by an arrowhead. Slower migrating products resulting from extension of radiolabeled primers are indicated with a bracket. B, primers were synthesized using 5 pmol of UL5/52 (UL5/52) or 5 pmol of UL5/52 plus 15 pmol of UL8 (UL5/52 + UL8). Primers annealed to template DNA were isolated by electrophoresis in nondenaturing 0.8% agarose-TBE gels, excised, and electrophoresed in denaturing 20% PAGE in parallel with samples of total products prior to fractionation. T, total primers; A, annealed primers.

5034 A Novel Function of the UL8 Subunit of the HSV Helicase-Primase TABLE I

The effect of UL.8 on kinetics of primer synthesis at varied DNA concentrations

Assays were performed and kinetic parameters calculated as described under “Experimental Procedures.”

Enzyme K,” VWl.Xh

UL5/52 UL5/8/52

P M 14.3 2.8 15.0 8.2

The K,,, for &X174 virion DNA is expressed in micromolar concen-

* The V,,, is expressed in absorbance units from scanning laser tration of nucleotides.

densitometry of autoradiographs.

ence or absence of UL8, as described under “Experimental Pro- cedures.” We found that in the presence of substrate ATP and DNA, the primase was stable at 30 “C over the 60 min of incubation, regardless of the presence of UL8 (data not shown). In contrast, the primase was highly unstable when incubated in the absence of substrate ATP and DNA half the activity was lost after 5 min of incubation (data not shown). However, under these conditions the level of stability was again independent of UL8. Therefore the stimulatory effect of UL8 on primase activity of UL5/52 was not due to an effect on the stability of the enzyme during incubation at 30 “C.

Kinetics of Primer Synthesis-One possible mechanism for UL8 stimulation of primer synthesis by UL5/52 is an increase in the affinity of the enzyme for the template DNA. Alterna- tively, the UL5/8/52 heterotrimer may synthesize primers a t a greater rate than the UL5/52 subassembly. To distinguish these two possibilities, we compared primer synthesis over a range of 4x174 virion DNA concentrations as described under “Experi- mental Procedures.” The kinetic parameters determined from the experimental data are shown in Table I. While the K,,, for 6x174 virion DNA was unchanged by the addition of a 3-fold excess of UL8 to UL5/52, the V,,, was increased approximately g-fold. The degree of increase in V,,, agrees closely with the level of stimulation of primer synthesis by UL8. We conclude that UL8 increases UL5/52 primer synthesis by increasing the rate of synthesis and not by changing the affinity of the complex for DNA.

HSV Primase Recognition /Initiation Sites-Eukaryotic primases are thought to initiate in random locations on template DNA while those of prokaryotes have more specific recognition sequences and structures (reviewed in Ref. 26). In order to determine if the stimulation of primer synthesis by UL8 resulted from an increase in the number of DNA sites or sequences that are recognized for initiation, we determined the number and location of predominant initiation sites for UL5/52 and the UL5/8/52 heterotrimer. Radiolabeled primers were synthesized on 4x174 and M13 virion DNA and were then extended with DNA polymerase. The resulting replicative form DNAduplexes were then digested with restriction enzymes and electrophoresed in nondenaturing PAGE gels and subjected to autoradiography. Radiolabeling of specific restriction fragments indicated sites of primer initiation and synthesis. It can be seen from Fig. 5 that there was one predominant site of primase initiation within 4x174 DNA and three predominant sites in M13 virion DNA. All the restriction enzyme fragments were labeled to some extent, as seen in long autoradiographic exposures (data not shown), indicating that initiation occurred, albeit at low levels, in many sequences. There was no significant difference in the use of specific initiation sites in the presence or absence of UL8. Radiolabeling at all the predominant sites of initiation appeared to be increased by UL8. Therefore UL8 does not increase primer synthesis by increasing the number of DNA sequences used for initiation.

UL5/52 - 1.7 5 15 1.7 5 15 5 15

UL8 45 - - - 5 15 45 - - - .

> - ..

1 2 3 4 5 6 7 8 9 1011

FIG. 5. The effect of ULS on HSV primase initiation sites. Prim- ers were synthesized using the indicated amounts of UL5/52 and UL8 in 50-pl reactions. 30 pl were extended using 0.78 units of Sequenase and dNTPs. Products were digested with restriction enzymes and separated by electrophoresis in nondenaturing, 4% TBE-PAGE followed by autoradiography. The results presented are from 4 pl of the initial reaction. Lanes 1-7 contained &X174 virion DNA. Lanes 8-11 contained M13 virion DNA. The migration of restriction fragments of &X174 or M13 duplex DNAs are indicated by hatch marks on the left- and right-hand margins, respectively. Restriction fragments containing predominant initiation sites for HSV primase are indicated with arrowheads.

DISCUSSION

The three proteins of the HSV helicase-primase complex are essential for viral DNA replication in infected cells (3-5). The UL5, UL8, and UL52 heterotrimeric complex (UL5/8/52) can be isolated from HSV (10 , l l ) or recombinant baculovirus-infected cells (12, 13, 161, or reconstituted by addition of UL8 to a subassembly of the UL5 and UL52 proteins (16). The complex possesses helicase, NTPase, and primase activities in vitro.

Previously, the UL5/52 subassembly was shown to contain the helicase and NTPase activities of the UL5/8/52 heterotrimer (13-16). In addition, the UL5/52 subassembly was active in coupled primase-polymerase assays on homopolymeric poly(dT) template (14, 16) but not on natural DNA templates (16). Therefore, UL8 was implicated in primase activity on natural DNA template, perhaps in the annealing or extension of primers synthesized (16). In this work, we have conclusively shown that the UL8 protein stimulates the in vitro primase activity of the UL5/52 subassembly by stimulating the rate of primer synthesis. Moreover, after examining primer synthesis and coupled primase-polymerase activity on natural DNA templates we found that the UL5/52 subassembly alone contains the in vitro primase activity of the UL5/8/52 heterotrimer: the primers synthesized by UL5/52 were identical to the UL5/8/52 heterotrimer and only their rate of synthesis was affected by UL8.

The UL5/52 subassembly alone was capable of primase activity on the natural DNA templates and exhibited concentration dependent activity in the coupled primase-polymerase assay and in the direct primer synthesis assay. Previously, priming by UL5/52 could be detected on homopolymer DNA but not on natural DNA templates. Our detection of UL5/52 activity in coupled primase-polymerase assays on natural DNA templates could be due to more sensitive assays or more active enzyme preparations. We have consistently detected activity, however, with many different batches of the purified UL5/52 subassembly. In addition, the detection of UL8 stimulation of

A Novel Function of the UL8 Subunit of the HSV Helicase-Primase 5035

direct primer synthesis employed assay conditions resembling other published reports (11) for incorporation of [a-32PlUTP and [a-32P]CTP to label oligoribonucleotides. These labeled oligoribonucleotides served as primers for direct extension by DNApolymerase, both in the presence and absence of UL8 (Fig. 4A). We found that a critical parameter of this assay was the inclusion of high levels of ATP and GTP. Primer synthesis on 4x174 was optimum between 1 and 3 m~ ATP and GTP and was undetectable at 50.25 m~ (data not shown). Labeling of primers with radiolabeled ATP or GTP, with a corresponding decrease in the levels of unlabeled nucleotides, resulted in the detection of species that were variably responsive to enzyme concentration, were template independent, and were unable to be extended by DNA polymerases.

We found that the HSV primase initiates predominantly a t discrete sites on natural DNA templates, resembling the primases of prokaryotic rather than eukaryotic organisms (reviewed in Ref. 26). Perhaps a lack of HSV primase recognition sequences within homopolymer DNA accounts for low primase activity and the absence of a UL8 affect on these templates (14, 16h3 Preliminary results confirm that 4x174 and M13 initiation sequences best utilized by HSV primase are not encoded within homopolymer template^.^ It is interesting to note that homopolymer templates serve as excellent substrates for DNA- dependent NTPase activity but relatively poor substrates for primase activity ( 1 4 ~ ~ It would be of interest to determine the DNA sequences or structures that are utilized preferentially by the primase and whether they cause a transition of the protein from a DNA-dependent NTPase to a primase.

In addition to discrete initiation sites, a multiprotein complex encoding helicase and primase activities is reflective of prokaryotic primases (reviewed in Ref. 26). Both activities of the T4 phage helicase (gene 41 protein) and primase (gene 61 protein) are stimulated by formation of the heterodimer (27, 28). Primase activity of the 63-kDa component of the T7 helicase-primase is stimulated approximately 3-fold by addition of the 56-kDa component (29). This %fold stimulation in vitro may be responsible for an increased rate of T7 DNA synthesis observed in vivo (30, 31).

Unlike the 56-kDa protein of the T7 system, however, the UL8 protein is essential in vivo (3). Although UL8 stimulates primer synthesis by the UL5/52 subassembly, it is most likely not the only important function of the protein. In fact, we previously noted that the extent of UL8 stimulation of coupled primase-polymerase activity (Fig. 1) was higher than that observed in direct primer synthesis (Fig. 21, consistent with the possibility of additional levels of stimulation. In recent experiments we have in fact found that the HSV single-stranded DNA-binding protein (ICP8) forms a specific, functional inter- action with the UL5/8/52 heterotrimer but not the UL5/52 subassembly, indicating another potential function of ULtL2 Fur- thermore, experiments using temperature-sensitive mutants (17) suggested that none of the helicase-primase components localized to the nucleus of HSV-infected cells unless all three were co-expressed, consistent with the hypothesis that UL8 may be involved in nuclear localization of the complex. Detailed analysis of UL8 null mutants (3) should further elucidate this

D. J. Tenney, W. W. Hurlburt, P. A. Micheletti, M. Bifano, and R. K. Hamatake, unpublished observations.

possible function. The mechanism by which UL8 stimulates the rate of primer

synthesis by UL5/52 remains unknown. We have found that UL8 is also capable of stimulating the ATPase activity of the UL5/52 subassembly, although to a lesser degree than the stimulation of primer ~yn thes i s .~ The fact that UL8 is not required for helicase, NTPase, and primase activities of the UL5/52 subassembly, however, is inconsistent with a UL8-in- duced change in the global structure of the enzyme. More detailed analysis, including structure-function studies, are required to define the effect of UL8 binding to UL5/52 and the mechanism for the resulting increase in primer synthesis.

Acknowledgments-We are indebted to I. Robert Lehman for the generous gift of recombinant baculoviruses expressing HSV helicase- primase and Pol proteins and to Min Gao for the pSV8.3 vector containing the HSV ICP8 open reading frame. We appreciate insightful discussions during the course of these studies with Carolyn DiIanni, Mike Cordingley, and Rich Colonno and helpful critique of the manuscript from Steve Weinheimer, Min Gao, and Amy Sheaffer.

1. 2.

3. 4. 5. 6. 7.

8. 9.

10.

11. 12.

13. 14.

15.

16. 17. 18. 19.

20.

21.

22. 23. 24.

25.

26.

27.

28. 29.

30.

31.

REFERENCES Challberg, M. D., and Kelly, T. J. (1989)Annu. Rev. Biochem. 58,671-717 Weller, S . K. (1991) in Herpes Virus Panscription and Its Regulation (Wagner,

Carmichael, E. P., and Weller, S . K. (1989) J. Virol. 63, 591-599 Goldstein, D. J., and Weller, S . K. (1988) J. virol. 62, 297C~2977 Zhu, L., and Weller, S . K. (1992) J. virol. 66, 45-68 Challberg, M. D. (1986) Proc. Natl. Acad. Sei. U. S. A. 83,9094-9098 Wu, C. A,, Nelson, N. J., McGeoch, D. J., and Challberg, M. D. (1988) J. virol.

Stow, N. D. (1992) J. Gen. Virol. 73, 313-321 Cmte, J. J., Mocarski, E. S . , and Lehman, I. R. (1988) Nucleic Acids Res. 16,

65854596 Crute, J. J., Tsurumi, T., Zhu, L., Weller, S. K., Olivo, P. D., Challberg, M. D.,

Mocarski, E. S. , and Lehman, I. R. (1989) Proc. Natl. Acad. Sci. U. S . A. 8 6 , 218G2189

E. K., ed) pp. 105-135, CRC Press, Boca Raton, FL

62,435443

Crute, J. J., and Lehman, I. R. (1991) J. Biol. Chem. 266,44844488 Dodson, M. S . , Crute, J. J., Bruckner, R. C., and Lehman, I. R. (1989) J. Biol.

Calder, J. M., and Stow, N. D. (1990) Nucleic Acids Res. 18,3573-3578 Dodson, M. S . , and Lehman, I. R. (1991) Proc. Natl. Acad. Sci. U. S. A. 8 8 ,

Crute, J. J., Bruckner, R. C., Dodson, M. S . , and Lehman, I. R. (1991) J. Biol.

Calder, J. M., Stow, E. C., and Stow, N. D. (1992) J. Gen. virol. 73, 531538 Sherman, G., Gottlieb, J., and Challberg, M. D. (1992) J. Virol. 66,48844892

Hernandez, T. R., and Lehman, I . R. (1990) J. Biol. Chem. 265, 11227-11232 Hamatake, R. K., Bifano, M., Tenney, D. J., Hurlburt, W. W., and Cordingley,

Tenney, D. J., Hurlhurt, W. W., Bifano, M., Stevens, J. T., Micheletti, P. A,, M. G. (1993) J. Gen. virol. 74,2181-2189

McGeoch, D. J., Dalrymple, M. A,, Davidson, A. J., Dolan, A., Frame, M. C., Hamatake, R. K., and Cordingley, M. G. (1993) J. virol. 67, 1959-1966

McNab, D., Perry, L. J., Scott, J. E., and Taylor, P. (1988) J. Gen. Virol. 69, 1531-1574

Chem. 264,2083S20838

1105-1109

Chem. 266,2125221256

Gao, M., and Knipe, D. M. (1992) Mol. Cell. Biol. 12, 1330-1339 Gill, S. C., and von Hippel, P. H., (1989) Anal. Biochem. 182,319-326 Tenney, D. J., Micheletti, P. A,, Stevens, J. T., Hamatake, R. IC, Matthews, J.

T., Sanchez, A. R., Hurlburt, W. W., Bifano, M., and Cordingley, M. G. (1993) J. Virol. 67, 543-547

Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd Ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Komberg, A., and Baker, T. (1992) DNA Replication, 2nd Ed., pp. 275-306, W.

Venkatesan, M., Silver, L. L., and Nossal, N. G . (1982) J. Biol. Chem. 257,

Hinton, D. M., and Nossal, N. G. (1987) J. Biol. Chem. 262, 10873-10878 Patel, S. S. , Rosenberg, A. H., Studier, F. W., and Johnson, K. A. (1992) J. Biol.

Mendelman, L. V., Notarnicola, S. M., and Richardson, C. C. (1992) Proc. Natl.

Rosenberg, A. H., Patel, S. S. , Johnson, K. A,, and Studier, F. W. (1992) J. Biol.

H. Freeman and Co., New York

1242G12434

Chem. 267, 15013-15021

Acad. Sci. U. S . A. 89, 1063%10642

Chem. 267, 15005-15012

the ul8 component of the herpes simplex virus helicase-primase

Documents