Herpes Simplex-1 Helicase-Primase IDENTIFICATION OF TWO NUCLEOSIDE TRIPHOSPHATASE SITES THAT PROMOTE DNA HELICASE ACTION

Herpes simplex virus type 1 (HSV-1) encodes a heterotrimeric helicase-primase comprised of the products of the UL5, UL8, and UL52 genes (Crute, J. J., and Lehman, I. R. (1991) J. Biol. Chern. 266, 44844488). A steady state kinetic analysis of the enzyme isolated from HSV-1-infected CV1 cells or insect cells expressing the enzyme after infection with recombinant baculoviruses has shown it to possess two sites capable of hydrolyzing nucleoside triphosphates in a DNA-dependent manner. One site (Site I) hydrolyzes both ATP and GTP; the second (Site 11) hydrolyzes only ATP. These two sites are contained within a subassembly of the helicase-primase formed by coexpression of the UL5 and UL52 genes in insect cells. Sites I and I1 are activated by separate DNA effector sites, both of which support DNA helicase action. These findings are likely to be of importance in understanding how helicases in general catalyze the unwinding of duplex DNA and, in particular, how the helicase-primase functions at the HSV1 replication fork.

In this report, we present evidence that the heterotrimeric HSV-1 helicase-primase contains two separate nucleoside triphosphatase sites; one of these (Site I) catalyzes the hydrolysis * This research was supported in part by Grant A126538 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
$ Supported by a fellowship from the California Division of the American Cancer Society. Present address: Dept. of Molecular Biology, Boehringer Ingelheim Pharmaceuticals, Inc., P. 0. Box 368, 90 East Ridge, Ridgefield, CT 06877.
$ Supported by National Institutes of Health Fellowship 1F032 GM12092.
ll Supported by Damon Runyon-Walker Winchell Cancer Fund Fellowship DRG-961.

Materials
All reagents were obtained from sources previously described unless otherwise noted (15,16). ADP and GDP were from the Sigma. [3H] ATP (37 Ci/mmol) and [3H]GTP (7.0 Ci/mmol) were from Amersham Corp. Polyethyleneimine-cellulose F TLC plates were from EM Science (Gibbstown, NJ). Reagents for native and SDS-polyacrylamide gel electrophoresis were from Bio-Rad.

Nucleic Acids
The helicase substrate was prepared exactly as described previously (15). (dA),,o was obtained from Midland Certified Reagent Co. (Midland, TX). Activated calf thymus DNA was prepared as described (19).
Purifkation of HSV-1 Helicase-Primase and Subassembly The HSV-1 helicase-primase from HSV-1-infected CV-1 cells (Fraction IV) was purified as previously described (15,16). Recombinant baculovirus-expressed helicase-primase was isolated as described and purified through the gel filtration step (17). The U W / UL52 subassembly of the HSV-1 helicase-primase that contains both helicase and primase activities was isolated from doubly infected SF9 cells. The purification procedure was the same as that described for the recombinant three-subunit enzyme; only the gel filtration step was omitted (21). It was approximately 80% pure, as assessed by SDS-polyacrylamide gel electrophoresis followed by staining with Coomassie Brilliant Blue (22).
Enzyme Assays DNA-dependent NTPase-Reaction mixtures were assembled in buffer A (25 pl) with additions made as indicated. Either [3H]ATP (3.0 Ci/mmol) or [3H]GTP (3.0 Ci/mmol) served as the NTP substrate. MgCL was added to maintain the level of free MgZf at 3.5 mM, assuming 1 mol of MgZf complexed/mol of NTP, and activated calf thymus DNA (220 p~) was used as DNA cofactor unless otherwise indicated. Reactions were initiated by addition of enzyme and were stopped after 20 min at 34 "C by the addition of 5.0 pl of 0.5 M EDTA. To separate products from reactants, three 1.0-pl aliquots of the reaction mixture were spotted onto polyethyleneimine-cellulose F TLC plates and allowed to dry. After development of the plates with 0.5 M formic acid, 0.5 M LiCl for assay of DNA-dependent ATPase, and 0.5 M formic acid, 1.0 M LiCl for assay of DNA-dependent GTPase, the products, ADP or GDP, were visualized with ultraviolet light using standards run in a parallel lane. Regions of the plates corresponding to the NDP were excised and dissolved in a liquid scintillation fluor, and their radioactivity was determined.
Helicase-DNA helicase activity was measured essentially as described (15) using the 3'-tailed substrate. Assay mixtures (10 pl) were assembled in buffer A on ice and contained the helicase substrate (15 p~ in nucleotide) and the indicated concentration of either ATP or GTP. MgCI, was added to maintain the concentration of free MgZ' at 3.5 mM. Reactions were initiated by the addition of 60 ng of the recombinant baculovirus-expressed HSV-1 helicase-primase. To ensure that the helicase assay was within the linear range, reactions were terminated after 10 min at 34 "C by the addition of one-third volume of a freshly prepared solution consisting of 200 mM EDTA, 1% SDS, 50% glycerol. Under these conditions, less than 10% of the labeled oligonucleotide was displaced from the single-stranded DNA. Electrophoresis was performed as described (15); the gel was backed with DE-52 paper prior to drying, and the relative amounts of product in each reaction were determined by scanning with an AMBIS radioanalytic imaging system (AMBIS Systems, Inc., San Diego, CA).

RESULTS
The HSV-1 Helicase-Primase Has One DNA-dependent GTPase and Two DNA-dependent ATPase Sites-Examination of the DNA-dependent NTPase activity of the HSV-1 helicase-primase by Lineweaver-Burk analysis showed that ATP and GTP were utilized differently ( Fig. la). When GTP was the substrate, a linear double-reciprocal plot was observed that yielded a K,(GTP) of 1.0 mM. In contrast, when ATP hydrolysis was measured, a curvilinear graph resulted that ATP should produce only partial inhibition. Under these conditions, the contribution of the high K, site to the total DNA-dependent ATPase activity should be inhibited by the added GTP, leaving the activity of the low K, DNA-dependent ATPase site unchanged.
As shown in Fig. 2, in reactions performed in the presence of sub-K, levels of [3H]GTP (250 p~) , addition of excess unlabeled ATP (up to 6.0 mM) resulted in complete inhibition of DNA-dependent GTPase activity. With [3H]ATP the addition of unlabeled GTP resulted in partial inhibition of the DNA-dependent ATPase activity until a plateau was reached. We conclude from these data that the HSV-1 helicase-primase has two DNA-dependent NTPase active sites. The higher K , site (i.e. &(ATP) = 1.3 mM and &(GTP) = 1.0 mM) can hydrolyze both ATP and GTP. This site will be referred to as Site I. The lower K, site, designated as Site I1 (ix. K, = 170 pM), is specific for ATP (Table I).
From measurements of DNA-dependent GTPase activity, we estimate the turnover number at Site I to be 35 s-l. From measurements of DNA-dependent ATPase activity, we esti- Initial reaction rates were determined by the analysis of aliquots (6.5 p l ) removed from the reaction mixtures and quenched with 0.5 M EDTA (4.0 pl) at 5, 10, and 20 min. Initial reaction rates were determined from the 5-min time point for the DNA-dependent ATPase and from the 5-and 10min time points for the DNA-dependent GTPase. In all cases, less than 10% of the input NTP was converted to NDP. 0, GTP; 0, ATP.
Thirty-two nanograms of the helicase-primase isolated from HSV-1infected cells were used in each reaction. a, Lineweaver-Burk plot; b, Eadie-Hofstee plot. mate the turnover number at Site I1 to be 20 s-l and the maximum turnover number for NTP hydrolysis to be 50 s" (Table I).
Sites I and 1 1 Are Present in the Helicase-Primase Subas-sernbly Containing the Products of the HSV-1 UL5 and UL.52 Genes-A baculovirus expressed subassembly of the helicaseprimase, consisting of the products of the UL5 and UL52 genes, contains the DNA-dependent NTPase activity associated with the three-subunit holoenzyme (21). We therefore wished to determine whether Sites I and I1 are both retained in the subassembly. Competition experiments of the type described above were performed utilizing both the baculovirus expressed helicase-primase holoenzyme and the two-subunit subassembly. As summarized in Table 11, the recombinant three-subunit helicase-primase and the two-subunit subassembly were indistinguishable. They were also indistinguishable from the enzyme isolated from HSV-1-infected cells (compare Table I1 and Fig. 2). With either recombinant enzyme, DNA-dependent GTPase activity was almost completely inhibited by the addition of excess unlabeled ATP. In contrast, approximately 30% of the DNA-dependent ATPase activity was retained in the presence of excess unlabeled GTP. These findings confirm the presence of both Sites I and I1 in the baculovirus-expressed HSV-1 helicase-primase and further localize these two sites to the two subunits encoded by the UL.5 and UL52 genes.
Sites Zand ZIAre Actiuated by Distinct DNA Effector Sites-Since Sites I and I1 represent distinct DNA-dependent NTPase active sites in the HSV-1 helicase-primase, we wished to determine whether these two sites were responsive to the same or different DNA effector sites. By measuring DNAdependent NTPase as described in the previous section, the &(DNA) at Sites I and I1 can be measured independently. The hydrolysis of [3H]GTP (4.0 mM) at subsaturating levels of ATP (250 p~) represents the NTPase associated with Site I; the hydrolysis of 13H]ATP (250 pM) in the presence of saturating levels of GTP (4.0 mM) represents the NTPase  activity contributed by Site 11. Under these conditions, the &(DNA) of the DNA effector for Site I activation was found to be 3.4 p~ (nucleotide) (Fig. 3a). An identical experiment performed to determine the K,,,(DNA) for Site I1 produced a nonlinar Lineweaver-Burk plot that yielded two K,(DNA) values (12 and 0.62 p~) (Fig. 3b). From these data, we conclude that activation of Sites I and I1 is mediated by distinct binding sites for the DNA effector that can be differentiated by their apparent affinity for DNA. Both Sites I and ZZ Participate in Helicase Activity-To determine the role of Site I and/or Site 11 in promoting helicase action, ATP and GTP were added individually at increasing concentrations to a helicase reaction. ATP at low concentrations would be expected to bind Site I1 preferentially, whereas at high concentrations it should be bound to both Sites I and 11. In contrast, GTP should bind Site I exclusively. As shown in Fig. 4, GTP supported displacement of the oligonucleotide annealed to M13 mp18 single-stranded DNA. Thus, NTP hydrolysis at Site I alone can promote helicase action. However, the reaction is less efficient than when both sites are occupied by a nucleoside triphosphate, i.e. in the presence of high ATP concentrations.
At low NTP levels, ATP was severalfold more effective in promoting helicase action than GTP. ATP at 0.60 and 1.0 mM produced a 3.6-and 3.5-fold greater rate of oligonucleotide displacement, respectively, than GTP (Fig. 4b). At these concentrations of ATP, Site I1 was about 90% saturated. The NTP saturation level at Site I was 25% for ATP and 40% for GTP.

DISCUSSION
The HSV-1-encoded helicase-primase shows complex kinetics with respect to its NTP substrates. In the presence of a DNA cofactor, the enzyme displayed one K,,, for GTP (1.0 mM) and two for ATP (1.3 mM and 170 pM). Competition experiments showed the high K, (1.3 mM) DNA-dependent ATPase and GTPase sites to be the same (Site I), and distinct from the low K,,, DNA-dependent ATPase site with a K,,, of 170 p~ (Site 11). Both Sites I and I1 are present in the enzyme from HSV-1 infected CV-1 cells, as well as in the recombinant helicase-primase isolated from insect cells. The two-subunit subassembly composed of the products of the UL.5 and UL.52 genes also contained both sites, thereby localizing them to two of the polypeptides of the three-subunit holoenzyme.
In our initial studies of the helicase-primase from HSV-1-

TABLE I1
Utilization of nucleoside triphosphates by HSV-1 helicase-primase and UWIUL52 subassembly Assays were performed as described under "Experimental Procedures" using activated calf thymus DNA as cofactor. Additions of 3H-labeled and unlabeled NTP were as indicated. Assays were initiated with 60 ng of the baculovirus-expressed HSV-1 helicase-primase or 80 ng of the baculovirus-expressed U W / U W 2 subassembly.  infected Vero cells, we observed substantial proteolytic degradation in the course of purification (16). However, the major degradation product was separated from the intact enzyme at an early stage in the purification (15). The degradation product lacked DNA-dependent GTPase activity but retained DNA-dependent ATPase activity? Preliminary experiments showed this fragment to have a &(ATP) of 300 PM, close to the 170 PM that we have observed for Site 11. These findings suggest that Site I can be differentially removed from the helicase-primase by proteolysis to yield an enzyme that retains a protease-resistant Site 11. Inspection of the nucleotide sequence of the U W gene has shown it to belong to a superfamily of NTP-binding proteins that may function as DNA helicases (23). A conserved ATPbinding motif, encompassing residues 92-119, within the UL5 gene can account for one of the two DNA-dependent NTPase sites, but not both. Analysis of the nucleotide sequence of UW2 has not revealed a consensus NTP-binding domain.' The introduction of specific changes in the U W and UW2 genes by site-directed mutagenesis, followed by analysis of the mutant proteins, should permit the localization of Sites I and I1 within the helicase-primase.
Sites I and I1 are activated by different DNA effector sites.
In the presence of GTP, the KJDNA) for Site I is 3.4 PM  increase is a consequence of the ability of Site I and Site I1 to function independently in the unwinding of DNA or whether there is interaction between the two sites.
The prokaryotic helicases that have been examined appear to contain a single NTPase site. Thus, the finding of multiple NTPase sites in the HSV-1 helicase-primase is unusual and may be reflective of eukaryotic helicases in general.