- Email: [email protected]
Mutations in the Exo III Motif of the Herpes Simplex Virus DNA Polymerase Gene Can Confer Altered Drug Sensitivities
VIROLOGY
246, 298±305 (1998) VY989201
ARTICLE NO.
Mutations in the Exo III Motif of the Herpes Simplex Virus DNA Polymerase Gene Can Confer Altered Drug Sensitivities Ying T. Hwang, Jennifer F. Smith, Li Gao, and Charles B. C. Hwang1 Department of Microbiology, and Immunology, College of Medicine, Health Science Center, State University of New York, Syracuse, New York 13210 Received March 2, 1998; returned to author for revision April 2, 1998; accepted April 20, 1998 Two herpes simplex virus mutants containing mutated residues within the conserved Exo III motif of the polymerase gene were previously shown to be defective in 39±59 exonuclease activity and exhibited extremely high mutation frequencies. In this study, we have shown that these mutants also exhibited higher resistance to phosphonoacetic acid and sensitivity to aphidicolin and all nucleoside analogs tested, including acyclovir and gancicolvir, compared to wild-type virus. Marker transfer experiments and sequencing analyses demonstrated that these altered phenotypes were the result of mutations within the Exo III motif. The data indicate that, aside from leading to exonuclease deficiency, mutations in the Exo III motif may also affect interaction of nucleoside triphosphates with the catalytic sites of polymerase activity. © 1998 Academic Press
ing of substrates and pyrophosphate (reviewed in Coen, 1996). The 39±59 exonuclease activity of HSV Pol is located in the N-terminal half of the protein, which contains three highly conserved motifs (Exo I, II, and III) (Fig. 1A) based on protein sequence alignments among several Pols containing exonuclease activity. Examination of mutant Pols containing mutations within these Exo motifs of Escherichia coli Pol I (Derbyshire et al., 1988, 1991; Freemont et al., 1988), Pols of bacteriophages (Frey et al., 1993; Reha-Krantz and Nonay, 1993; Reha-Krantz et al., 1991; Soengaas et al., 1992), and yeast Pol d (Morrison et al., 1991; Simon et al., 1991) confirmed that these conserved motifs are critical for 39±59 exonuclease activity, and some mutants, if not lethal, have mutator phenotypes (Frey et al., 1993). Similarly, HSV Pol mutants containing mutations within these motifs are severely impaired in exonuclease activity (Hall et al., 1995; Hwang et al., 1997; KuÈhn and Knopf, 1996). For example, two Exo III mutants, Y7 and YD12, have been shown previously to be able to replicate in Vero cells, yet they exhibit 300- to 800-fold increases in the mutation frequencies of the thymidine kinase (tk) gene, compared to wild-type strain KOS (Hwang et al., 1997). Their extremely high mutation frequencies are presumably due to impaired exonuclease activities; only less than 2%, if any, of wild-type activity is displayed by these two mutants. This finding suggests that the proofreading function of 39±59 exonuclease activity intrinsic to HSV Pol may not be absolutely essential for the function of HSV pol and DNA replication. Previous reports (Gibbs et al., 1988; Hall et al., 1989; Wang et al., 1992) have demonstrated that certain HSV-1 mutants with altered drug sensitivities contain mutations
INTRODUCTION The catalytic subunit of herpes simplex virus (HSV)2 DNA polymerase (pol) possesses polymerase and 39±59 exonuclease activity (Knopf, 1979; Knopf and Weisshart, 1990; Marcy et al., 1990). Protein sequence alignment of HSV Pol and other DNA polymerases reveals several homologous regions, designated regions I through VII (Gibbs et al., 1988; Hwang et al., 1992). HSV Pol also contains a conserved d region C shared among other eukaryotic viral DNA polymerases (Coen, 1996; Hwang et al., 1997), and DNA polymerase d (Zhang et al., 1991) and e (Kesti et al., 1993). Importantly, most HSV mutants conferring resistance to certain antiviral drugs have mutations within the conserved regions, including regions I, II, III, V, and VII and d region C (Gibbs et al., 1988; Hwang et al., 1992; Larder et al., 1987; Wang et al., 1992). Since these antiviral drugs mimic and/or compete with the natural substrates for incorporation by the viral DNA Pol, resistance to these antiviral drugs can result from altered specificity during insertion of these analogs. Furthermore, HSV pol mutants exhibiting resistance to nucleoside analogs are usually accompanied by altered sensitivities to phosphonoacetic acid (PAA), which is thought to be a pyrophosphate analog, and aphidicolin. Therefore, it had been suggested that these conserved regions are directly or indirectly involved in recognition and bind-
1 To whom reprint requests should be addressed. Fax: (315) 4647680. E-mail: [email protected]. 2 Abbreviations used: HSV, herpes simplex virus; PAA, phosphonoacetic acid; ACV, acyclovir; GCV, ganciclovir; AraA, arabinosyladenine; AraT, arabinosylthymidine; ICdR, iododeoxycytidine; BVdU, bromovinyldeoxyuridine.
0042-6822/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
298
HSV Exo III MUTANTS
299
FIG. 1. (A) HSV Pol open reading frame with regions of sequence similarity. Conserved amino acid regions I to VII, and d region C, which overlap the Exo III motif, among diverse DNA polymerases are shown as empty boxes. The dark boxes indicate the conserved Exo I, II, and III motifs critical for 39±59 exonuclease activity. The numbers above the line refer to amino acid residues of HSV Pol. Sequences shown below the line are the Exo III residues of HSV Pol; the altered amino acids in mutant Y7 and YD12 are also indicated. (B). Maps of SmaI and KpnI restriction sites of the HSV pol gene. The box below the map shows the DNA fragment which is able to rescue the altered drug sensitive phenotype of the Y7 mutant.
in the N-terminal half of the pol gene, including sequences nearby and upstream of the Exo III motif. However, their exonuclease capabilities have not been characterized. In this study, we have characterized the phenotypes of two Exo III mutants, Y7 and YD12. In addition to the loss of exonuclease activity, both mutants exhibit altered drug sensitivities. These results suggest that the conserved Exo III motif of HSV Pol may play an important role in maintaining the proper structure of the catalytic site of polymerase activity, in addition to its role in exonuclease activity. RESULTS
resistant to PAA, had an ED50 at 6 mg/ml, while the Klenow fragment of E. coli Pol I, which was known to be resistant to PAA, had an ED50 of .100 mg/ml. As expected, mutant Pol Y7 and YD12 were resistant to the inhibitory effects of PAA, with an ED50 of .100 and 28 mg/ml, respectively (Fig. 2). Therefore, both Y7 and YD12 mutant Pols were highly resistant to PAA. The Exo III mutants of HSV-1 pol exhibit altered drug sensitivities With recombinant viruses available, we also examined their sensitivities to various antiviral drugs, in-
Recombinant Pols containing the Exo III mutations are resistant to PAA Earlier studies (Gibbs et al., 1988; Hall et al., 1989; Hwang et al., 1997; Larder et al., 1987; Wang et al., 1992) demonstrate that HSV pol mutants with mutations located upstream and downstream of the Exo III motif can confer altered drug sensitivities. Because the Exo III motif lies at the center of the conserved d region C, we therefore hypothesized that mutations of the Exo III motif might also exhibit altered drug sensitivities. To test this hypothesis, we began to examine the effects of PAA on the polymerase activities of the wild-type, Y7, and YD12 recombinant HSV-1 Pols which were expressed and purified from recombinant baculovirus-infected insect Sf9 cells (Hwang et al., 1997). By the standard gap-filling reactions, the polymerase activity of wild-type Pol was significantly inhibited by PAA, with an effective dose inhibiting 50% of activity (ED50) at 6 mg/ml, while the Klenow fragment of E. coli Pol I, which was known to be
FIG. 2. Effects of PAA on polymerase activities of purified wild-type (X), Y7 (‚), and YD12 (E) HSV Pols and the E. coli Pol I Klenow fragment (h). Polymerase activity assays were performed as described under Materials and Methods.
300
HWANG ET AL.
FIG. 3. Effects of aphidicolin (A), PAA (B), ACV (C), GCV (D), AraA (E), AraT (F), ICdR (G), and BVdU (H) on plaque formation by wild-type KOS (X) and the mutants Y7 (‚) and YD12 (E) and the Kpn I 1.1-kb rescued Y7 virus (h). Plaque reduction assays were performed as described under Materials and Methods.
cluding PAA, aphidicolin, ACV, GCV, AraA, AraT, ICdR, and BVdU. Like most HSV pol mutants containing mutations within conserved regions (Gibbs et al., 1988; Hwang et al., 1992), both mutants were hypersensitive to aphidicolin (Fig. 3) and resistant to PAA (Fig. 3B), compared to wild-type virus. In contrast to most drugresistant mutants (Gibbs et al., 1988; Hwang et al., 1992; Wang et al., 1992), which confer resistance to both PAA and nucleoside analogs and map within the polymerase domain, both mutants exhibited different degrees of hypersensitivity to ACV (Fig. 3C), GCV (Fig. 3D), AraA (Fig. 3E), AraT (Fig. 3F), ICdR (Fig. 3G), and BVdU (Fig. 3H). The ED50 results of each drug exhibited by the wild-type, Y7, and YD12 viruses are summarized in Table 1. Thus, both Exo III mutant viruses are not only severely impaired in exonuclease activity (Hwang et al., 1997), but also exhibit altered drug sensitivities.
Y577H is the only change at the amino acid level in the pol open reading frame of the Y7 mutant Although both Exo III mutants were constructed to contain the desired mutations, these mutants might introduce an additional mutation(s) within the pol gene due to their high mutation rates (Hwang et al., 1997), and such unexpected change(s) might be attributed to the altered drug sensitivities. To rule out this possibility, the entire open reading of the pol gene of the Y7 mutant was sequenced as described under Materials and Methods. As expected, sequencing results revealed a T to C change in codon 577 (change of TAC to CAC), which changes tyrosine to histidine. In addition, a G to A change corresponding to nucleotide 459 of the pol gene was also found. However, this G to A alteration does not alter the leucine codon at residue 153. Therefore, the pol gene of the Y7 mutant contains
HSV Exo III MUTANTS
301
FIG. 3ÐContinued
only the expected tyrosine to histidine mutation at the amino acid level.
region of the Y7 pol gene conferring the altered phenotype. Although mapping can be accomplished by transferring the PAA resistant (PAAr) marker of Y7 into wildtype KOS virus, the progeny of transfection receiving the PAAr marker of Y7 may generate an additional mutation(s) if the marker is indeed the altered Exo III sequence. To avoid this potential problem, which may complicate the interpretation of the data, we choose to res-
A DNA fragment containing the Exo III motif can rescue the altered drug sensitivities Sequencing data indicated that the Y577H mutation was responsible for the altered drug sensitivities of mutant Y7. To verify this further, we attempted to map the
TABLE 1 ED50 of Various Drugs on Wild-Type, Y7, and YD12 Viruses
KOS Y7 YD12
PAA (mg/ml)
Aphidicolin (mg/ml)
ACV (mM)
GCV (mM)
AraA (mM)
AraT (mg/ml)
ICdR (mg/ml)
BVdU (mg/ml)
7 25 42
0.36 0.16 0.25
1.05 0.48 0.75
0.58 0.28 0.3
160 40 36
6 1.2 1.1
11 1.4 1.4
0.14 0.01 0.02
Note. The ED50, the concentration of each drug to reduce plaque formation by 50%, was determined directly from the plaque reduction curves shown in Fig. 3.
302
HWANG ET AL. TABLE 2 Marker Rescue of Aphhs Marker in Y7 Mutant Virus Progeny of Y7 DNA plus the following fragment:
Experiment No.
None
KpnI 1.0
KpnI 1.1
KpnI 1.8
SmaI C
SmaI D
I II
,0.04* ,0.03
,0.4 ND
3.6 2.8
,0.09 ND
ND ND
ND ,0.5
Note. Marker transfer experiments were performed by cotransfection of Y7 DNA prepared from mutant virus together with different DNA fragments (Fig. 1B) derived from plasmid containing wild-type pol sequences. Progeny of transfectants were titered on Vero cells in the absence or the presence of 0.5 mg/ml of aphidicolin. The efficiency of each DNA fragment to rescue the Aphhs of Y7 was expressed as the percentage of progeny (*) to form plaques on Vero cells in the presence of 0.5 mg/ml of aphidicolin. Values underlined are considered positive for marker rescue. ND, not determined.
cue the hypersensitivity phenotype of Y7 to aphidicolin (Aphhs) by using different DNA fragments derived from the wild-type pol gene. The reason for using Y7 for conducting these experiments is that the altered sensitivity of Y7 to aphidicolin was sufficient to permit marker rescue mapping of the Aphhs mutation. Experiments were performed by cotransfection of the Y7 infectious DNA with different DNA fragments derived from the wildtype pol gene. The progeny of transfections using infectious Y7 DNA alone formed no plaques in the presence of 0.5 mg/ml of aphidicolin. In two different experiments (Table 2), the 1.1-kb KpnI fragment of the wild-type pol gene that includes the Exo III motif was able to rescue the growth of Y7 effectively in the presence of 0.5 mg/ml of aphidicolin. The 1.0- and 1.8-kb KpnI fragments and the SmaI D fragment, which do not include the Exo III motif, were not able to rescue the Aphhs marker. Therefore, the marker conferring increased sensitivity to aphidicolin in Y7 mapped to the 1022 bp within the 1.1-kb KpnI fragment which includes the Exo III sequences, but excludes the sequences within the SmaI D fragment (Fig. 1B). Recombinant viruses resulting from the marker rescue experiments were plaque purified and their sensitivities to various drugs were analyzed. These viruses exhibited sensitivities to aphidicolin, PAA, and AraA indistinguishable from those of the wild-type KOS virus (Figs. 3A, 3B, and 3E). However, they were highly resistant to ACV, GCV, AraT, ICDR, and BVdU (Figs. 3C, 3D, 3F, 3G, and 3H). Since the HSV tk gene is required for phosphorylation of these nucleoside analogs (Elion et al., 1977; Cheng et al., 1983; Gentry and Aswell, 1975; Summers and Summers 1977; De Clerq et al., 1979), with the exception of AraA (North and Cohen, 1979), their resistances to these drugs suggested that they might contain tk mutations. It was possible that these rescued recombinants were derived from preexisting tk mutations in infectious Y7 viral DNA used for marker rescue experiments. Sequences flanking the Exo III motif of these marker rescued recombinant viruses were also analyzed as described above. These recombinant viruses no longer contained the mutation
at residue 577. We conclude that, similar to other mutations upstream and downstream of Exo III in d region C (Gibbs et al., 1988; Larder et al., 1987; Wang et al., 1992), the Y7 mutation altering the Exo III sequence can confer altered drug sensitivities in vitro and in infected cells. Alterations of the Exo III secondary structure induced by mutations within the conserved d region C The secondary structures of pol mutants containing mutations within the d region C were analyzed by the method of Garnier et al. (1978) via the PEPTIDESTRUCTURE and PLOTSTRUCTURE programs in GCG (Genetics Computer Group; Wisconsin Package; Madison, Wisconsin). While Exo III residues of wild-type Pol formed a continuous b sheet structure (residues 570 to 587) (Fig. 4), this method predicted that this b sheet structure was altered to form a turn structure in both Y7 and YD12 mutants. Interestingly, mutants Su1 (Wang et al., 1992), RSC-26 (Larder et al., 1987), tsD9 (Gibbs et al., 1988), and PFAr2 (Gibbs et al., 1988), which each confer altered drug sensitivities, also altered the structure of residues in the d region C (Fig. 4). It is possible that alterations of the local structure within Exo III and the d region C might also affect the proper structure of the catalytic sites of polymerase activity, which would result in altered sensitivities of these mutant pols to these drugs. It is also notable that the structural alteration of the d region C alone may not affect exonuclease activity, since PFAr2 mutant Pol had been shown to display exonuclease activity indistinguishable from that of the wild-type Pol (Derse et al., 1982). Exonuclease activity requires the Exo III motif, especially the conserved tyrosine and aspartic acid residues, which have the catalytic activity per se and bind the second divalent metal ion, respectively. Therefore, the Exo III motif may have a role in maintaining the proper structure of the catalytic sites of polymerase activity, in addition to its role in the 39±59 exonuclease activity.
HSV Exo III MUTANTS
303
FIG. 4. Predicted alterations of the secondary structures of the Exo III and d region C. Secondary structure predictions were analyzed by the method described in the text. The amino acid residues of the d region C of wild-type HSV-1 Pol are shown at the top; the numbers above it refer to the amino acid residues. The predicted secondary structure (a, a-helical; b, b sheet; c, coil; and t, turn) of wild-type sequences is shown below the sequences. The mutated residues are underlined. Only those with altered structure resulting from each mutation are shown. The following mutations were analyzed: Y577H (mutant Y7), Y577H/D581A (mutant YD12); D531N (mutant Su1); E597D (mutant RSC-26); E597K (mutant tsD9), and A605V (mutant PFAr2).
DISCUSSION The structural and functional independence of the polymerase and exonuclease domains of HSV Pol has been proposed (Haffey et al., 1990) according to the model of E. coli Pol I (reviewed in Joyce and Steitz, 1994). However, a previous study (Weisshart et al., 1994) indicates that the catalytic sites for both polymerase and exonuclease activities may reside on a single domain. Alternatively, the catalytic site for polymerase activity may reside on an independent structural domain but requires functional interactions with other domains (Weisshart et al., 1994). This result is based on the biochemical analysis of different fragments of Pol after limited proteolytic digestion. Other studies which demonstrate that several mutant pols containing mutations within or upstream of the conserved 39±59 exonuclease activity site exhibit altered polymerase activity (Gibbs et al., 1991; Hall et al., 1989; Wang et al., 1992) also implicate the interdependence between exonuclease and polymerase activities of HSV pol. It was possible that these mutants with altered polymerase phenotypes might be also associated with altered exonuclease activity. However, the effects of these mutations on 39±59 exonuclease activity had not been demonstrated. In this study, we demonstrate that two 39±59 exonuclease-deficient pols can lead to the altered polymerase activity with regard to the interactions with both pyrophosphate and nucleoside analogs. These altered phenotypes are due to mutations within the Exo III motif based on sequencing data and marker transfer experiments. Our results further support the model of interdependence between the exonuclease and polymerase activities of HSV pol. The altered drug sensitivities of both Y7 and YD12
mutants may also suggest that interactions with the sugar, base, and phosphate moieties of dNTPs are altered. Their high resistance to PAA reflected that these mutations might have substantial alterations in the structure of the pyrophosphate exchange-release site which could affect the interaction of catalytic site of polymerase activity with the triphosphate moiety of nucleoside triphosphates. Furthermore, the increased sensitivities of both mutant Pols to all nucleoside analogs tested in this study were substrate dependent. For example, both mutants exhibited two- to fivefold increased sensitivity to nucleoside analogs with altered sugar moieties, i.e., ACV, GCV, AraA, and AraT. Both ACV and GCV are guanosine analogs lacking the 29 and 39 or the 29 moieties, respectively. AraA and AraT are adenosine and thymidine analogs, respectively, with the arabinosyl conformation at the 29 position. These results suggested that both mutations could have intermediate effects on interactions with the sugar moiety of nucleoside triphosphates. The substantially increased sensitivities of both mutants to both ICdR and BVdU, which are nucleoside analogs with altered structures of pyrimidine bases, also suggested that these mutations could have dramatic effects on their interactions with the base moiety of nucleoside triphosphates. Thus, the mutated pol with altered residues 577 and 581 could alter interactions with the sugar, base, and triphosphate moieties of nucleoside triphosphates, in addition to its loss of exonuclease activity (Hwang et al., 1997). Although mutations in the Exo III motif and d region C could alter drug sensitivity, it is currently unclear whether residues within the Exo III motif and d region C could have direct or indirect interactions with nucleoside triphosphates. It is possible that they may function to
304
HWANG ET AL.
maintain a proper structure in the catalytic sites of polymerase activity. Therefore, the effects of these mutated residues on the structure of Pol demand further studies, such as X-ray crystallographic analysis. Alternatively, it will be of value if suppressor mutants can be isolated from exonuclease-deficient pol to demonstrate that the exonuclease-deficient phenotype can be reversed by mutations within the catalytic sites of polymerase activity. Such information would further support the model of structural and functional interdependence between these two activities of HSV pol. MATERIALS AND METHODS Cells and viruses Vero cells were grown and maintained as described (Hwang et al., 1997). HSV wild-type virus strain KOS and its derivative Y7, which contains a single mutation (a change of amino acid 577 from tyrosine to histidine; Y577H), and YD12, which contains double changes (the same change in Y577H plus a change of aspartic acid at residue 581 to alanine; Y577H/D581A), were propagated as described (Hwang et al., 1997). Chemicals All chemicals and antiviral drugs, except ganciclovir (GCV; Syntex, Palo Alto, CA), used in this study were purchased from Sigma (St. Louis, MO). Antiviral drug PAA is a pyrophosphate analog and binds at the pyrophosphate exchange-release site of the Pol. Acyclovir (ACV) and GCV are guanosine analogs lacking the 29 and 39 or the 29 of sugar moieties, respectively. Arabinosyladenine (AraA) and arabinosylthymidine (AraT) are adenine and thymidine analogs, respectively, with arabinosyl conformation at the 29 position. Iododeoxycytidine (ICdR) is a cytidine analog containing iodine at the 59 position. Bromovinyldeoxyuridine (BVdU) is a thymidine analog with a replacement of the 5-methyl group of the thymine base with a bromovinyl group. Polymerase activity assays The standard gap-filling reactions (Marcy et al., 1990; Hwang et al., 1997) were performed to determine the sensitivity of wild-type and mutant Pol proteins to PAA. Sensitivity of each Pol to PAA was determined by the amount of [a-32P]dCTP incorporated into activated calf thymus DNA under different concentrations of PAA and measured in a liquid scintillation counter. The Klenow fragment of E. coli Pol I was included as a control. Drug assays Plaque reduction assays were applied to determine relative sensitivity to antiviral drugs as described previously (Hwang et al., 1992).
Sequencing of the pol genes To confirm that Y7 virus contained only the Exo III mutation, three DNA fragments were amplified from total DNA prepared from Y7 infected Vero cells by polymerase chain reaction (PCR) using three pairs of primers from the pol gene, corresponding to nucleotides 218 to 21 (Pol-59) and 1593 to 1578 (2158L), 1189 to 1209 (1200R) and 2180 to 2790 (3389L), and 2745 to 2765 (3325R) and 3734 to 3717 (Pol-39), where the A nucleotide of the first ATG codon of the pol gene is defined as nucleotide 1. The PCR products, which covered the entire open reading frame of the pol gene, were then sequenced directly using primers corresponding to the pol gene nucleotides 133 to 149 (712R), 210 to 190 (270L), 339 to 354 (920R), 431 to 350 (430R), 2921 to 2938 (2921R), 3073 to 3094 (3073R), and 3432 to 3454 (4018R) and those described previously (Hwang et al., 1997). Marker rescue experiments and mutation mapping Marker rescue was performed as described previously (Hwang et al., 1992) by using Y7 infectious DNA and the KpnI or SmaI fragments of the wild-type pol gene isolated from plasmid pSVK-pol (Hwang et al., 1997). To confirm that the marker-rescued recombinant viruses contained the wild-type sequence, the corresponding 1.1-kb KpnI fragment was amplified by PCR from crude virion DNAs. The resulting DNAs were sequenced using the corresponding primers. ACKNOWLEDGMENT This work is supported by NIH Grant DE10051.
REFERENCES Cheng, Y.-C., Huang, E.-S., Lin, J.-C., Mar, E.-C., Pagano, J. S., Dutschman, G. E., and Grill, S. P. (1983). Unique spectrum of activity of 9-[1,3-dihydroxy-2-propoxy)methyl]guanine against herpesvirus in vitro and its mode of action against herpes simplex virus type 1. Proc. Natl. Acad. Sci. USA 80, 2767±2770. Coen, D. M. (1996). Viral DNA polymerases. In ``DNA Replication in Eukaryotic Cells'' (M. L. Depamphilis, Ed.) Monograph 31, pp. 495± 523. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. De Clercq, E., Descamps, J., De Somer, P., Barr, P. J., Jones, A. S., and Walker, R. T. (1979). (E)-5-(2-Bromovinyl)-29-deoxyuridine: A potent and selective anti-herpes agent. Proc. Natl. Acad. Sci. USA 76, 2947±2951. Derbyshire, V., Grindley, N. D. F., and Joyce, C. M. (1991). The 39-59 exonuclease of DNA polymerase of Escherichia coli: Contribution of each amino acid at the active site to the reaction. EMBO J. 10, 17±24. Derbyshire, V., Freemont, P. S., Sanderson, M. R., Beese, L., Friedman, J. M., Joyce, C. M., and Steitz, T. A. (1988). Genetic and crystallographic identification of essential groups in the 39-59 exonucleolytic site of DNA polymerase. Science 240, 199±210. Derse, D., Bastow, K. F., and Cheng, Y.-C. (1982). Characterization of the DNA polymerases induced by a group of herpes simplex virus type I variants selected for growth in the presence of phosphonoformic acid. J. Biol. Chem. 257, 10251±10260. Freemont, P. S., Friedman, J. M., Beese, L. S., Sanderson, M. R., and Steitz, T. A. (1988). Cocrystal structure of an editing complex of
HSV Exo III MUTANTS Klenow fragment with DNA. Proc. Natl. Acad. Sci. USA 85, 8924± 8928. Frey, M. W., Nossal, N. G., Capson, T. L., and Benkovic, S. J. (1993). Construction and characterization of a bacteriophage T4 DNA polymerase deficient in 39-59 exonuclease activity. Proc. Natl. Acad. Sci. USA 90, 2579±2583. Elion, G. B., Furman, P. A., Fyfe, J. A., De Miranda, P., Beauchamp, L., and Schaeffer, H. J. (1977). Selectivity of action of an antiherpetic agent, 9-(2-hydroxyethoxymethyl)guanine. Proc. Natl. Acad. Sci. USA 74, 5716±5720. Garnier, J., Osguthorpe, D. J., and Robson, B. (1978). Analysis of the accuracy and implications of simple methods for predicting the secondary structure of globular proteins. J. Mol. Biol. 120, 97±120. Gentry, G. A., and Aswell, J. F. (1975). Inhibition of herpes simplex virus replication by AraT. Virology 65, 294±296. Gibbs, J. S., Chiou, H. C., Bastow, K. F., Cheng, Y.-C., and Coen, D. M. (1988). Identification of amino acids in herpes simplex virus DNA polymerase involved in substrate and drug recognition. Proc. Natl. Acad. Sci. USA 85, 6672±6676. Gibbs, J. S., Weisshart, K., Digard, P., DeBruynkops, A., Knipe, D. M., and Coen, D. M. (1991). Polymerization activity of an a-like DNA polymerase requires a conserved 39-59 exonuclease active site. Mol. Cell. Biol. 11, 4786±4795. Haffey, M. L., Novonty, J., Bruccoleri, R. E., Carroll, R. D., Stevens, J. T., and Matthews, J. T. (1990). Structure-function studies of the herpes simplex virus DNA polymerase. J. Virol. 64, 5008±5018. Hall, J. D., Orth, K. L., Sander, K. L., Swihart, B. M., and Senese, R. A. (1995). Mutations within conserved motifs in the 39-59 exonuclease domain of herpes simplex virus DNA polymerase. J. gen. Virol. 76, 2999±3008. Hall, J. D., Wang, Y., Pierpont, J., Berlin, M. S., Rundlett, S. E., and Woodward, S. (1989). Aphidicolin resistance in herpes simplex virus type 1 reveals features of the DNA polymerase dNTP binding site. Nucleic Acids Res. 17, 9231±9244. Hwang, C. B. C., Ruffner, K. L., and Coen, D. M. (1992). A point mutation within a distinct conserved region of the herpes simplex virus DNA polymerase gene confers drug resistance. J. Virol. 66, 1774±1776. Hwang, Y. T., Liu, B.-Y., Coen, D. M., and Hwang, C. B. C. (1997). Effects of mutations in the Exo III motif of the herpes simplex virus DNA polymerase gene on enzyme activities, viral replication, and replication fidelity. J. Virol. 71, 7791±7798. Joyce, C. M. and Steitz, T. A. (1994). Function and structure relationships in DNA polymerases. Annu. Rev. Biochem. 63, 777±822. Kesti, T., Frantti, H., and SyvaÈoja, J. E. (1993). Molecular cloning of the cDNA for the catalytic subunit of human DNA polymerase e. J. Biol. Chem. 268, 10238±10245. Knopf, C. W. (1979). Properties of herpes simplex virus DNA polymerase and characterization of its associated exonuclease activity. Eur. J. Biochem. 98, 231±244.
305
Knopf, C. W., and Weisshart, K. (1990). Comparison of exonucleolytic activities of herpes simplex virus type-1 DNA polymerase and DNase. Eur. J. Biochem. 191, 263±273. KuÈhn, F. J. P., and Knopf, C. W. (1996). Herpes simplex virus type 1 DNA polymerase: Mutational analysis of the 39-59 exonuclease domain. J. Biol. Chem. 271, 29245±29254. Larder, B. A., Kemp, S. D., and Darby, G. (1987). Related functional domains in virus DNA polymerases. EMBO J. 6, 169±175. Marcy, A. I., Olivo, P. D., Challberg, M. D., and Coen, D. M. (1990). Enzymatic activities of overexpressed herpes simplex virus DNA polymerase purified from recombinant baculovirus-infected insect cells. Nucleic Acids Res. 18, 1207±1215. Morrison, A., Bell, J. B., Kunkel, T. A., and Sugino, A. (1991). Eukaryotic DNA polymerase amino acid sequence required for 39-59 exonuclease activity. Proc. Natl. Acad. Sci. USA 88, 9473±9477. North, T. W., and Cohen, S. S. (1979). Aranucleosides and aranucleotides in viral chemotherapy. Pharmacol. Ther. 4, 81±108. Reha-Krantz, L. J., and Nonay, R. L. (1993). Genetic and biochemical studies of bacteriophage T4 DNA polymerase 39 3 59 exonuclease activity. J. Biol. Chem. 268, 27100±27108. Reha-Krantz, L. J., Stocki, S., Nonay, R. L., Dimayauga, E., Goodrich, L. D., Konigsberg, W. H., and Spicer, E. K. (1991). DNA polymerization in the absence of exonucleolytic proofreading: In vivo and in vitro studies. Proc. Natl. Acad. Sci. USA 88, 2417±2421. Simon, M., Giot, L., and Faye, G. (1991). The 39-59 exonuclease activity located in the DNA polymerase d subunit of Saccharomyces cerevisiae is required for accurate replication. EMBO J. 10, 2165±2170. Soengas, M. S., Esteban, J. A., Lazaro, J. M., Bernad, A., Blasco, M. A., Salas, M., and Blanco, L. (1992). Site-directed mutagenesis at the Exo III motif of f29 DNA polymerase: Overlapping structural domains for the 39-59 exonuclease and strand-displacement activities. EMBO J. 11, 4227±4237. Summers, W. C., and Summers, W. P. (1977). [125I]Deoxycytidine used in a rapid, sensitive, and specific assay for herpes simplex virus type 1 thymidine kinase. J. Virol. 24, 314±318. Wang, Y., Woodward, S., and Hall, J. D. (1992). Use of suppressor analysis to identify DNA polymerase mutations in herpes simplex virus which affect deoxynucleotide triphosphate substrate specificity. J. Virol. 66, 1814±1816. Weisshart, K., Kuo, A., Hwang, C. B. C., Kumura, K., and Coen, D. M. (1994). Structural and functional organization of herpes simplrx virus DNA polymerase investigated by limited proteolysis. J. Biol. Chem. 269, 22788±22796. Zhang, J., Chung, D. W., Tan, C.-K., Downey, K. M., Davie, E. W., and So, A. G. (1991). Primary structure of the catalytic subunit of calf thymus DNA polymerase d: Sequence similarities with other DNA polymerases. Biochemistry 30, 11742±11750.
246, 298±305 (1998) VY989201
ARTICLE NO.
Mutations in the Exo III Motif of the Herpes Simplex Virus DNA Polymerase Gene Can Confer Altered Drug Sensitivities Ying T. Hwang, Jennifer F. Smith, Li Gao, and Charles B. C. Hwang1 Department of Microbiology, and Immunology, College of Medicine, Health Science Center, State University of New York, Syracuse, New York 13210 Received March 2, 1998; returned to author for revision April 2, 1998; accepted April 20, 1998 Two herpes simplex virus mutants containing mutated residues within the conserved Exo III motif of the polymerase gene were previously shown to be defective in 39±59 exonuclease activity and exhibited extremely high mutation frequencies. In this study, we have shown that these mutants also exhibited higher resistance to phosphonoacetic acid and sensitivity to aphidicolin and all nucleoside analogs tested, including acyclovir and gancicolvir, compared to wild-type virus. Marker transfer experiments and sequencing analyses demonstrated that these altered phenotypes were the result of mutations within the Exo III motif. The data indicate that, aside from leading to exonuclease deficiency, mutations in the Exo III motif may also affect interaction of nucleoside triphosphates with the catalytic sites of polymerase activity. © 1998 Academic Press
ing of substrates and pyrophosphate (reviewed in Coen, 1996). The 39±59 exonuclease activity of HSV Pol is located in the N-terminal half of the protein, which contains three highly conserved motifs (Exo I, II, and III) (Fig. 1A) based on protein sequence alignments among several Pols containing exonuclease activity. Examination of mutant Pols containing mutations within these Exo motifs of Escherichia coli Pol I (Derbyshire et al., 1988, 1991; Freemont et al., 1988), Pols of bacteriophages (Frey et al., 1993; Reha-Krantz and Nonay, 1993; Reha-Krantz et al., 1991; Soengaas et al., 1992), and yeast Pol d (Morrison et al., 1991; Simon et al., 1991) confirmed that these conserved motifs are critical for 39±59 exonuclease activity, and some mutants, if not lethal, have mutator phenotypes (Frey et al., 1993). Similarly, HSV Pol mutants containing mutations within these motifs are severely impaired in exonuclease activity (Hall et al., 1995; Hwang et al., 1997; KuÈhn and Knopf, 1996). For example, two Exo III mutants, Y7 and YD12, have been shown previously to be able to replicate in Vero cells, yet they exhibit 300- to 800-fold increases in the mutation frequencies of the thymidine kinase (tk) gene, compared to wild-type strain KOS (Hwang et al., 1997). Their extremely high mutation frequencies are presumably due to impaired exonuclease activities; only less than 2%, if any, of wild-type activity is displayed by these two mutants. This finding suggests that the proofreading function of 39±59 exonuclease activity intrinsic to HSV Pol may not be absolutely essential for the function of HSV pol and DNA replication. Previous reports (Gibbs et al., 1988; Hall et al., 1989; Wang et al., 1992) have demonstrated that certain HSV-1 mutants with altered drug sensitivities contain mutations
INTRODUCTION The catalytic subunit of herpes simplex virus (HSV)2 DNA polymerase (pol) possesses polymerase and 39±59 exonuclease activity (Knopf, 1979; Knopf and Weisshart, 1990; Marcy et al., 1990). Protein sequence alignment of HSV Pol and other DNA polymerases reveals several homologous regions, designated regions I through VII (Gibbs et al., 1988; Hwang et al., 1992). HSV Pol also contains a conserved d region C shared among other eukaryotic viral DNA polymerases (Coen, 1996; Hwang et al., 1997), and DNA polymerase d (Zhang et al., 1991) and e (Kesti et al., 1993). Importantly, most HSV mutants conferring resistance to certain antiviral drugs have mutations within the conserved regions, including regions I, II, III, V, and VII and d region C (Gibbs et al., 1988; Hwang et al., 1992; Larder et al., 1987; Wang et al., 1992). Since these antiviral drugs mimic and/or compete with the natural substrates for incorporation by the viral DNA Pol, resistance to these antiviral drugs can result from altered specificity during insertion of these analogs. Furthermore, HSV pol mutants exhibiting resistance to nucleoside analogs are usually accompanied by altered sensitivities to phosphonoacetic acid (PAA), which is thought to be a pyrophosphate analog, and aphidicolin. Therefore, it had been suggested that these conserved regions are directly or indirectly involved in recognition and bind-
1 To whom reprint requests should be addressed. Fax: (315) 4647680. E-mail: [email protected]. 2 Abbreviations used: HSV, herpes simplex virus; PAA, phosphonoacetic acid; ACV, acyclovir; GCV, ganciclovir; AraA, arabinosyladenine; AraT, arabinosylthymidine; ICdR, iododeoxycytidine; BVdU, bromovinyldeoxyuridine.
0042-6822/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
298
HSV Exo III MUTANTS
299
FIG. 1. (A) HSV Pol open reading frame with regions of sequence similarity. Conserved amino acid regions I to VII, and d region C, which overlap the Exo III motif, among diverse DNA polymerases are shown as empty boxes. The dark boxes indicate the conserved Exo I, II, and III motifs critical for 39±59 exonuclease activity. The numbers above the line refer to amino acid residues of HSV Pol. Sequences shown below the line are the Exo III residues of HSV Pol; the altered amino acids in mutant Y7 and YD12 are also indicated. (B). Maps of SmaI and KpnI restriction sites of the HSV pol gene. The box below the map shows the DNA fragment which is able to rescue the altered drug sensitive phenotype of the Y7 mutant.
in the N-terminal half of the pol gene, including sequences nearby and upstream of the Exo III motif. However, their exonuclease capabilities have not been characterized. In this study, we have characterized the phenotypes of two Exo III mutants, Y7 and YD12. In addition to the loss of exonuclease activity, both mutants exhibit altered drug sensitivities. These results suggest that the conserved Exo III motif of HSV Pol may play an important role in maintaining the proper structure of the catalytic site of polymerase activity, in addition to its role in exonuclease activity. RESULTS
resistant to PAA, had an ED50 at 6 mg/ml, while the Klenow fragment of E. coli Pol I, which was known to be resistant to PAA, had an ED50 of .100 mg/ml. As expected, mutant Pol Y7 and YD12 were resistant to the inhibitory effects of PAA, with an ED50 of .100 and 28 mg/ml, respectively (Fig. 2). Therefore, both Y7 and YD12 mutant Pols were highly resistant to PAA. The Exo III mutants of HSV-1 pol exhibit altered drug sensitivities With recombinant viruses available, we also examined their sensitivities to various antiviral drugs, in-
Recombinant Pols containing the Exo III mutations are resistant to PAA Earlier studies (Gibbs et al., 1988; Hall et al., 1989; Hwang et al., 1997; Larder et al., 1987; Wang et al., 1992) demonstrate that HSV pol mutants with mutations located upstream and downstream of the Exo III motif can confer altered drug sensitivities. Because the Exo III motif lies at the center of the conserved d region C, we therefore hypothesized that mutations of the Exo III motif might also exhibit altered drug sensitivities. To test this hypothesis, we began to examine the effects of PAA on the polymerase activities of the wild-type, Y7, and YD12 recombinant HSV-1 Pols which were expressed and purified from recombinant baculovirus-infected insect Sf9 cells (Hwang et al., 1997). By the standard gap-filling reactions, the polymerase activity of wild-type Pol was significantly inhibited by PAA, with an effective dose inhibiting 50% of activity (ED50) at 6 mg/ml, while the Klenow fragment of E. coli Pol I, which was known to be
FIG. 2. Effects of PAA on polymerase activities of purified wild-type (X), Y7 (‚), and YD12 (E) HSV Pols and the E. coli Pol I Klenow fragment (h). Polymerase activity assays were performed as described under Materials and Methods.
300
HWANG ET AL.
FIG. 3. Effects of aphidicolin (A), PAA (B), ACV (C), GCV (D), AraA (E), AraT (F), ICdR (G), and BVdU (H) on plaque formation by wild-type KOS (X) and the mutants Y7 (‚) and YD12 (E) and the Kpn I 1.1-kb rescued Y7 virus (h). Plaque reduction assays were performed as described under Materials and Methods.
cluding PAA, aphidicolin, ACV, GCV, AraA, AraT, ICdR, and BVdU. Like most HSV pol mutants containing mutations within conserved regions (Gibbs et al., 1988; Hwang et al., 1992), both mutants were hypersensitive to aphidicolin (Fig. 3) and resistant to PAA (Fig. 3B), compared to wild-type virus. In contrast to most drugresistant mutants (Gibbs et al., 1988; Hwang et al., 1992; Wang et al., 1992), which confer resistance to both PAA and nucleoside analogs and map within the polymerase domain, both mutants exhibited different degrees of hypersensitivity to ACV (Fig. 3C), GCV (Fig. 3D), AraA (Fig. 3E), AraT (Fig. 3F), ICdR (Fig. 3G), and BVdU (Fig. 3H). The ED50 results of each drug exhibited by the wild-type, Y7, and YD12 viruses are summarized in Table 1. Thus, both Exo III mutant viruses are not only severely impaired in exonuclease activity (Hwang et al., 1997), but also exhibit altered drug sensitivities.
Y577H is the only change at the amino acid level in the pol open reading frame of the Y7 mutant Although both Exo III mutants were constructed to contain the desired mutations, these mutants might introduce an additional mutation(s) within the pol gene due to their high mutation rates (Hwang et al., 1997), and such unexpected change(s) might be attributed to the altered drug sensitivities. To rule out this possibility, the entire open reading of the pol gene of the Y7 mutant was sequenced as described under Materials and Methods. As expected, sequencing results revealed a T to C change in codon 577 (change of TAC to CAC), which changes tyrosine to histidine. In addition, a G to A change corresponding to nucleotide 459 of the pol gene was also found. However, this G to A alteration does not alter the leucine codon at residue 153. Therefore, the pol gene of the Y7 mutant contains
HSV Exo III MUTANTS
301
FIG. 3ÐContinued
only the expected tyrosine to histidine mutation at the amino acid level.
region of the Y7 pol gene conferring the altered phenotype. Although mapping can be accomplished by transferring the PAA resistant (PAAr) marker of Y7 into wildtype KOS virus, the progeny of transfection receiving the PAAr marker of Y7 may generate an additional mutation(s) if the marker is indeed the altered Exo III sequence. To avoid this potential problem, which may complicate the interpretation of the data, we choose to res-
A DNA fragment containing the Exo III motif can rescue the altered drug sensitivities Sequencing data indicated that the Y577H mutation was responsible for the altered drug sensitivities of mutant Y7. To verify this further, we attempted to map the
TABLE 1 ED50 of Various Drugs on Wild-Type, Y7, and YD12 Viruses
KOS Y7 YD12
PAA (mg/ml)
Aphidicolin (mg/ml)
ACV (mM)
GCV (mM)
AraA (mM)
AraT (mg/ml)
ICdR (mg/ml)
BVdU (mg/ml)
7 25 42
0.36 0.16 0.25
1.05 0.48 0.75
0.58 0.28 0.3
160 40 36
6 1.2 1.1
11 1.4 1.4
0.14 0.01 0.02
Note. The ED50, the concentration of each drug to reduce plaque formation by 50%, was determined directly from the plaque reduction curves shown in Fig. 3.
302
HWANG ET AL. TABLE 2 Marker Rescue of Aphhs Marker in Y7 Mutant Virus Progeny of Y7 DNA plus the following fragment:
Experiment No.
None
KpnI 1.0
KpnI 1.1
KpnI 1.8
SmaI C
SmaI D
I II
,0.04* ,0.03
,0.4 ND
3.6 2.8
,0.09 ND
ND ND
ND ,0.5
Note. Marker transfer experiments were performed by cotransfection of Y7 DNA prepared from mutant virus together with different DNA fragments (Fig. 1B) derived from plasmid containing wild-type pol sequences. Progeny of transfectants were titered on Vero cells in the absence or the presence of 0.5 mg/ml of aphidicolin. The efficiency of each DNA fragment to rescue the Aphhs of Y7 was expressed as the percentage of progeny (*) to form plaques on Vero cells in the presence of 0.5 mg/ml of aphidicolin. Values underlined are considered positive for marker rescue. ND, not determined.
cue the hypersensitivity phenotype of Y7 to aphidicolin (Aphhs) by using different DNA fragments derived from the wild-type pol gene. The reason for using Y7 for conducting these experiments is that the altered sensitivity of Y7 to aphidicolin was sufficient to permit marker rescue mapping of the Aphhs mutation. Experiments were performed by cotransfection of the Y7 infectious DNA with different DNA fragments derived from the wildtype pol gene. The progeny of transfections using infectious Y7 DNA alone formed no plaques in the presence of 0.5 mg/ml of aphidicolin. In two different experiments (Table 2), the 1.1-kb KpnI fragment of the wild-type pol gene that includes the Exo III motif was able to rescue the growth of Y7 effectively in the presence of 0.5 mg/ml of aphidicolin. The 1.0- and 1.8-kb KpnI fragments and the SmaI D fragment, which do not include the Exo III motif, were not able to rescue the Aphhs marker. Therefore, the marker conferring increased sensitivity to aphidicolin in Y7 mapped to the 1022 bp within the 1.1-kb KpnI fragment which includes the Exo III sequences, but excludes the sequences within the SmaI D fragment (Fig. 1B). Recombinant viruses resulting from the marker rescue experiments were plaque purified and their sensitivities to various drugs were analyzed. These viruses exhibited sensitivities to aphidicolin, PAA, and AraA indistinguishable from those of the wild-type KOS virus (Figs. 3A, 3B, and 3E). However, they were highly resistant to ACV, GCV, AraT, ICDR, and BVdU (Figs. 3C, 3D, 3F, 3G, and 3H). Since the HSV tk gene is required for phosphorylation of these nucleoside analogs (Elion et al., 1977; Cheng et al., 1983; Gentry and Aswell, 1975; Summers and Summers 1977; De Clerq et al., 1979), with the exception of AraA (North and Cohen, 1979), their resistances to these drugs suggested that they might contain tk mutations. It was possible that these rescued recombinants were derived from preexisting tk mutations in infectious Y7 viral DNA used for marker rescue experiments. Sequences flanking the Exo III motif of these marker rescued recombinant viruses were also analyzed as described above. These recombinant viruses no longer contained the mutation
at residue 577. We conclude that, similar to other mutations upstream and downstream of Exo III in d region C (Gibbs et al., 1988; Larder et al., 1987; Wang et al., 1992), the Y7 mutation altering the Exo III sequence can confer altered drug sensitivities in vitro and in infected cells. Alterations of the Exo III secondary structure induced by mutations within the conserved d region C The secondary structures of pol mutants containing mutations within the d region C were analyzed by the method of Garnier et al. (1978) via the PEPTIDESTRUCTURE and PLOTSTRUCTURE programs in GCG (Genetics Computer Group; Wisconsin Package; Madison, Wisconsin). While Exo III residues of wild-type Pol formed a continuous b sheet structure (residues 570 to 587) (Fig. 4), this method predicted that this b sheet structure was altered to form a turn structure in both Y7 and YD12 mutants. Interestingly, mutants Su1 (Wang et al., 1992), RSC-26 (Larder et al., 1987), tsD9 (Gibbs et al., 1988), and PFAr2 (Gibbs et al., 1988), which each confer altered drug sensitivities, also altered the structure of residues in the d region C (Fig. 4). It is possible that alterations of the local structure within Exo III and the d region C might also affect the proper structure of the catalytic sites of polymerase activity, which would result in altered sensitivities of these mutant pols to these drugs. It is also notable that the structural alteration of the d region C alone may not affect exonuclease activity, since PFAr2 mutant Pol had been shown to display exonuclease activity indistinguishable from that of the wild-type Pol (Derse et al., 1982). Exonuclease activity requires the Exo III motif, especially the conserved tyrosine and aspartic acid residues, which have the catalytic activity per se and bind the second divalent metal ion, respectively. Therefore, the Exo III motif may have a role in maintaining the proper structure of the catalytic sites of polymerase activity, in addition to its role in the 39±59 exonuclease activity.
HSV Exo III MUTANTS
303
FIG. 4. Predicted alterations of the secondary structures of the Exo III and d region C. Secondary structure predictions were analyzed by the method described in the text. The amino acid residues of the d region C of wild-type HSV-1 Pol are shown at the top; the numbers above it refer to the amino acid residues. The predicted secondary structure (a, a-helical; b, b sheet; c, coil; and t, turn) of wild-type sequences is shown below the sequences. The mutated residues are underlined. Only those with altered structure resulting from each mutation are shown. The following mutations were analyzed: Y577H (mutant Y7), Y577H/D581A (mutant YD12); D531N (mutant Su1); E597D (mutant RSC-26); E597K (mutant tsD9), and A605V (mutant PFAr2).
DISCUSSION The structural and functional independence of the polymerase and exonuclease domains of HSV Pol has been proposed (Haffey et al., 1990) according to the model of E. coli Pol I (reviewed in Joyce and Steitz, 1994). However, a previous study (Weisshart et al., 1994) indicates that the catalytic sites for both polymerase and exonuclease activities may reside on a single domain. Alternatively, the catalytic site for polymerase activity may reside on an independent structural domain but requires functional interactions with other domains (Weisshart et al., 1994). This result is based on the biochemical analysis of different fragments of Pol after limited proteolytic digestion. Other studies which demonstrate that several mutant pols containing mutations within or upstream of the conserved 39±59 exonuclease activity site exhibit altered polymerase activity (Gibbs et al., 1991; Hall et al., 1989; Wang et al., 1992) also implicate the interdependence between exonuclease and polymerase activities of HSV pol. It was possible that these mutants with altered polymerase phenotypes might be also associated with altered exonuclease activity. However, the effects of these mutations on 39±59 exonuclease activity had not been demonstrated. In this study, we demonstrate that two 39±59 exonuclease-deficient pols can lead to the altered polymerase activity with regard to the interactions with both pyrophosphate and nucleoside analogs. These altered phenotypes are due to mutations within the Exo III motif based on sequencing data and marker transfer experiments. Our results further support the model of interdependence between the exonuclease and polymerase activities of HSV pol. The altered drug sensitivities of both Y7 and YD12
mutants may also suggest that interactions with the sugar, base, and phosphate moieties of dNTPs are altered. Their high resistance to PAA reflected that these mutations might have substantial alterations in the structure of the pyrophosphate exchange-release site which could affect the interaction of catalytic site of polymerase activity with the triphosphate moiety of nucleoside triphosphates. Furthermore, the increased sensitivities of both mutant Pols to all nucleoside analogs tested in this study were substrate dependent. For example, both mutants exhibited two- to fivefold increased sensitivity to nucleoside analogs with altered sugar moieties, i.e., ACV, GCV, AraA, and AraT. Both ACV and GCV are guanosine analogs lacking the 29 and 39 or the 29 moieties, respectively. AraA and AraT are adenosine and thymidine analogs, respectively, with the arabinosyl conformation at the 29 position. These results suggested that both mutations could have intermediate effects on interactions with the sugar moiety of nucleoside triphosphates. The substantially increased sensitivities of both mutants to both ICdR and BVdU, which are nucleoside analogs with altered structures of pyrimidine bases, also suggested that these mutations could have dramatic effects on their interactions with the base moiety of nucleoside triphosphates. Thus, the mutated pol with altered residues 577 and 581 could alter interactions with the sugar, base, and triphosphate moieties of nucleoside triphosphates, in addition to its loss of exonuclease activity (Hwang et al., 1997). Although mutations in the Exo III motif and d region C could alter drug sensitivity, it is currently unclear whether residues within the Exo III motif and d region C could have direct or indirect interactions with nucleoside triphosphates. It is possible that they may function to
304
HWANG ET AL.
maintain a proper structure in the catalytic sites of polymerase activity. Therefore, the effects of these mutated residues on the structure of Pol demand further studies, such as X-ray crystallographic analysis. Alternatively, it will be of value if suppressor mutants can be isolated from exonuclease-deficient pol to demonstrate that the exonuclease-deficient phenotype can be reversed by mutations within the catalytic sites of polymerase activity. Such information would further support the model of structural and functional interdependence between these two activities of HSV pol. MATERIALS AND METHODS Cells and viruses Vero cells were grown and maintained as described (Hwang et al., 1997). HSV wild-type virus strain KOS and its derivative Y7, which contains a single mutation (a change of amino acid 577 from tyrosine to histidine; Y577H), and YD12, which contains double changes (the same change in Y577H plus a change of aspartic acid at residue 581 to alanine; Y577H/D581A), were propagated as described (Hwang et al., 1997). Chemicals All chemicals and antiviral drugs, except ganciclovir (GCV; Syntex, Palo Alto, CA), used in this study were purchased from Sigma (St. Louis, MO). Antiviral drug PAA is a pyrophosphate analog and binds at the pyrophosphate exchange-release site of the Pol. Acyclovir (ACV) and GCV are guanosine analogs lacking the 29 and 39 or the 29 of sugar moieties, respectively. Arabinosyladenine (AraA) and arabinosylthymidine (AraT) are adenine and thymidine analogs, respectively, with arabinosyl conformation at the 29 position. Iododeoxycytidine (ICdR) is a cytidine analog containing iodine at the 59 position. Bromovinyldeoxyuridine (BVdU) is a thymidine analog with a replacement of the 5-methyl group of the thymine base with a bromovinyl group. Polymerase activity assays The standard gap-filling reactions (Marcy et al., 1990; Hwang et al., 1997) were performed to determine the sensitivity of wild-type and mutant Pol proteins to PAA. Sensitivity of each Pol to PAA was determined by the amount of [a-32P]dCTP incorporated into activated calf thymus DNA under different concentrations of PAA and measured in a liquid scintillation counter. The Klenow fragment of E. coli Pol I was included as a control. Drug assays Plaque reduction assays were applied to determine relative sensitivity to antiviral drugs as described previously (Hwang et al., 1992).
Sequencing of the pol genes To confirm that Y7 virus contained only the Exo III mutation, three DNA fragments were amplified from total DNA prepared from Y7 infected Vero cells by polymerase chain reaction (PCR) using three pairs of primers from the pol gene, corresponding to nucleotides 218 to 21 (Pol-59) and 1593 to 1578 (2158L), 1189 to 1209 (1200R) and 2180 to 2790 (3389L), and 2745 to 2765 (3325R) and 3734 to 3717 (Pol-39), where the A nucleotide of the first ATG codon of the pol gene is defined as nucleotide 1. The PCR products, which covered the entire open reading frame of the pol gene, were then sequenced directly using primers corresponding to the pol gene nucleotides 133 to 149 (712R), 210 to 190 (270L), 339 to 354 (920R), 431 to 350 (430R), 2921 to 2938 (2921R), 3073 to 3094 (3073R), and 3432 to 3454 (4018R) and those described previously (Hwang et al., 1997). Marker rescue experiments and mutation mapping Marker rescue was performed as described previously (Hwang et al., 1992) by using Y7 infectious DNA and the KpnI or SmaI fragments of the wild-type pol gene isolated from plasmid pSVK-pol (Hwang et al., 1997). To confirm that the marker-rescued recombinant viruses contained the wild-type sequence, the corresponding 1.1-kb KpnI fragment was amplified by PCR from crude virion DNAs. The resulting DNAs were sequenced using the corresponding primers. ACKNOWLEDGMENT This work is supported by NIH Grant DE10051.
REFERENCES Cheng, Y.-C., Huang, E.-S., Lin, J.-C., Mar, E.-C., Pagano, J. S., Dutschman, G. E., and Grill, S. P. (1983). Unique spectrum of activity of 9-[1,3-dihydroxy-2-propoxy)methyl]guanine against herpesvirus in vitro and its mode of action against herpes simplex virus type 1. Proc. Natl. Acad. Sci. USA 80, 2767±2770. Coen, D. M. (1996). Viral DNA polymerases. In ``DNA Replication in Eukaryotic Cells'' (M. L. Depamphilis, Ed.) Monograph 31, pp. 495± 523. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. De Clercq, E., Descamps, J., De Somer, P., Barr, P. J., Jones, A. S., and Walker, R. T. (1979). (E)-5-(2-Bromovinyl)-29-deoxyuridine: A potent and selective anti-herpes agent. Proc. Natl. Acad. Sci. USA 76, 2947±2951. Derbyshire, V., Grindley, N. D. F., and Joyce, C. M. (1991). The 39-59 exonuclease of DNA polymerase of Escherichia coli: Contribution of each amino acid at the active site to the reaction. EMBO J. 10, 17±24. Derbyshire, V., Freemont, P. S., Sanderson, M. R., Beese, L., Friedman, J. M., Joyce, C. M., and Steitz, T. A. (1988). Genetic and crystallographic identification of essential groups in the 39-59 exonucleolytic site of DNA polymerase. Science 240, 199±210. Derse, D., Bastow, K. F., and Cheng, Y.-C. (1982). Characterization of the DNA polymerases induced by a group of herpes simplex virus type I variants selected for growth in the presence of phosphonoformic acid. J. Biol. Chem. 257, 10251±10260. Freemont, P. S., Friedman, J. M., Beese, L. S., Sanderson, M. R., and Steitz, T. A. (1988). Cocrystal structure of an editing complex of
HSV Exo III MUTANTS Klenow fragment with DNA. Proc. Natl. Acad. Sci. USA 85, 8924± 8928. Frey, M. W., Nossal, N. G., Capson, T. L., and Benkovic, S. J. (1993). Construction and characterization of a bacteriophage T4 DNA polymerase deficient in 39-59 exonuclease activity. Proc. Natl. Acad. Sci. USA 90, 2579±2583. Elion, G. B., Furman, P. A., Fyfe, J. A., De Miranda, P., Beauchamp, L., and Schaeffer, H. J. (1977). Selectivity of action of an antiherpetic agent, 9-(2-hydroxyethoxymethyl)guanine. Proc. Natl. Acad. Sci. USA 74, 5716±5720. Garnier, J., Osguthorpe, D. J., and Robson, B. (1978). Analysis of the accuracy and implications of simple methods for predicting the secondary structure of globular proteins. J. Mol. Biol. 120, 97±120. Gentry, G. A., and Aswell, J. F. (1975). Inhibition of herpes simplex virus replication by AraT. Virology 65, 294±296. Gibbs, J. S., Chiou, H. C., Bastow, K. F., Cheng, Y.-C., and Coen, D. M. (1988). Identification of amino acids in herpes simplex virus DNA polymerase involved in substrate and drug recognition. Proc. Natl. Acad. Sci. USA 85, 6672±6676. Gibbs, J. S., Weisshart, K., Digard, P., DeBruynkops, A., Knipe, D. M., and Coen, D. M. (1991). Polymerization activity of an a-like DNA polymerase requires a conserved 39-59 exonuclease active site. Mol. Cell. Biol. 11, 4786±4795. Haffey, M. L., Novonty, J., Bruccoleri, R. E., Carroll, R. D., Stevens, J. T., and Matthews, J. T. (1990). Structure-function studies of the herpes simplex virus DNA polymerase. J. Virol. 64, 5008±5018. Hall, J. D., Orth, K. L., Sander, K. L., Swihart, B. M., and Senese, R. A. (1995). Mutations within conserved motifs in the 39-59 exonuclease domain of herpes simplex virus DNA polymerase. J. gen. Virol. 76, 2999±3008. Hall, J. D., Wang, Y., Pierpont, J., Berlin, M. S., Rundlett, S. E., and Woodward, S. (1989). Aphidicolin resistance in herpes simplex virus type 1 reveals features of the DNA polymerase dNTP binding site. Nucleic Acids Res. 17, 9231±9244. Hwang, C. B. C., Ruffner, K. L., and Coen, D. M. (1992). A point mutation within a distinct conserved region of the herpes simplex virus DNA polymerase gene confers drug resistance. J. Virol. 66, 1774±1776. Hwang, Y. T., Liu, B.-Y., Coen, D. M., and Hwang, C. B. C. (1997). Effects of mutations in the Exo III motif of the herpes simplex virus DNA polymerase gene on enzyme activities, viral replication, and replication fidelity. J. Virol. 71, 7791±7798. Joyce, C. M. and Steitz, T. A. (1994). Function and structure relationships in DNA polymerases. Annu. Rev. Biochem. 63, 777±822. Kesti, T., Frantti, H., and SyvaÈoja, J. E. (1993). Molecular cloning of the cDNA for the catalytic subunit of human DNA polymerase e. J. Biol. Chem. 268, 10238±10245. Knopf, C. W. (1979). Properties of herpes simplex virus DNA polymerase and characterization of its associated exonuclease activity. Eur. J. Biochem. 98, 231±244.
305
Knopf, C. W., and Weisshart, K. (1990). Comparison of exonucleolytic activities of herpes simplex virus type-1 DNA polymerase and DNase. Eur. J. Biochem. 191, 263±273. KuÈhn, F. J. P., and Knopf, C. W. (1996). Herpes simplex virus type 1 DNA polymerase: Mutational analysis of the 39-59 exonuclease domain. J. Biol. Chem. 271, 29245±29254. Larder, B. A., Kemp, S. D., and Darby, G. (1987). Related functional domains in virus DNA polymerases. EMBO J. 6, 169±175. Marcy, A. I., Olivo, P. D., Challberg, M. D., and Coen, D. M. (1990). Enzymatic activities of overexpressed herpes simplex virus DNA polymerase purified from recombinant baculovirus-infected insect cells. Nucleic Acids Res. 18, 1207±1215. Morrison, A., Bell, J. B., Kunkel, T. A., and Sugino, A. (1991). Eukaryotic DNA polymerase amino acid sequence required for 39-59 exonuclease activity. Proc. Natl. Acad. Sci. USA 88, 9473±9477. North, T. W., and Cohen, S. S. (1979). Aranucleosides and aranucleotides in viral chemotherapy. Pharmacol. Ther. 4, 81±108. Reha-Krantz, L. J., and Nonay, R. L. (1993). Genetic and biochemical studies of bacteriophage T4 DNA polymerase 39 3 59 exonuclease activity. J. Biol. Chem. 268, 27100±27108. Reha-Krantz, L. J., Stocki, S., Nonay, R. L., Dimayauga, E., Goodrich, L. D., Konigsberg, W. H., and Spicer, E. K. (1991). DNA polymerization in the absence of exonucleolytic proofreading: In vivo and in vitro studies. Proc. Natl. Acad. Sci. USA 88, 2417±2421. Simon, M., Giot, L., and Faye, G. (1991). The 39-59 exonuclease activity located in the DNA polymerase d subunit of Saccharomyces cerevisiae is required for accurate replication. EMBO J. 10, 2165±2170. Soengas, M. S., Esteban, J. A., Lazaro, J. M., Bernad, A., Blasco, M. A., Salas, M., and Blanco, L. (1992). Site-directed mutagenesis at the Exo III motif of f29 DNA polymerase: Overlapping structural domains for the 39-59 exonuclease and strand-displacement activities. EMBO J. 11, 4227±4237. Summers, W. C., and Summers, W. P. (1977). [125I]Deoxycytidine used in a rapid, sensitive, and specific assay for herpes simplex virus type 1 thymidine kinase. J. Virol. 24, 314±318. Wang, Y., Woodward, S., and Hall, J. D. (1992). Use of suppressor analysis to identify DNA polymerase mutations in herpes simplex virus which affect deoxynucleotide triphosphate substrate specificity. J. Virol. 66, 1814±1816. Weisshart, K., Kuo, A., Hwang, C. B. C., Kumura, K., and Coen, D. M. (1994). Structural and functional organization of herpes simplrx virus DNA polymerase investigated by limited proteolysis. J. Biol. Chem. 269, 22788±22796. Zhang, J., Chung, D. W., Tan, C.-K., Downey, K. M., Davie, E. W., and So, A. G. (1991). Primary structure of the catalytic subunit of calf thymus DNA polymerase d: Sequence similarities with other DNA polymerases. Biochemistry 30, 11742±11750.