Overexpression, Purification, and Characterization ... - Semantic Scholar

Val. 266, No. 28, Issue of October 5, pp. 18737-18744,1991 Printed in U.S. A.

OF BIOLOGICAL CHEMISTRY THEJOURNAL 0 1991 by The American Society for Biochemistry and Molecular Biology, Inc.

Overexpression, Purification,and Characterization of Escherichia coli Bacteriophage PRDl DNA Polymerase I N VITRO SYNTHESIS OF FULL-LENGTH PRDl DNA WITHPURIFIEDPROTEINS* (Received for publication, February 26, 1991)

Harri SavilahtiS, Javier CaldenteyS, Kenneth Lundstromg, Juhani E. Syvaojan, and Dennis H.BamfordSII From the $Department of Genetics, University of Helsinki, Arkadiankatu 7, SF-00100 Helsinki, Finland, §Orion Corporation Ltd.. Laboraton of Molecular Genetics. Valimotie 7. SF-00380 Helsinki, Finland, and the PDepartmentof Biochemistry, Uniuersity of Oulu, Linnunmaa, SF-90570 Oulu, Finland

genome is a linear molecule of about 15.0 kb’ (Bamford et al., Thebacteriophage P R D l DNA polymerasegene 1991). At both genome ends there are 110 bp long inverted (gene I) has been cloned into the expression vector pPLHlOl under the controlof the X & promoter. Tai- terminal repeats (Savilahti and Bamford, 1986) and 5”covaloring of an efficient ribosome bindingsite in frontof lently linked terminal proteins (Bamford et al., 1983). The the gene by polymerase chain reaction led to a high PRDl system is reviewed in more detail by Mindich and level heat-inducible expression of the corresponding Bamford (1988) and Caldentey et al. (1990). The left genome end codesfortwovery early proteins gene product (Pl) in Escherichia coli cells. Expression was confirmed in vivo by complementation of phage (Mindich et al., 1982;McGraw et al., 1983; Savilahtiand P R D l DNA polymerase gene mutants andin vitro by Bamford, 1987; Hsieh et al., 1987; Jung et al., 1987), the formation of the genome terminal protein PS-dGMP genome terminalprotein (gene VIZI, protein P8) and the replication initiation complex. Expressed P R D l DNA DNA polymerase (gene I , protein P l ) , in this order from the polymerase was purified to apparent homogeneity in genome terminus. The function of the terminal proteins is to serve as a primer in the initiation of DNA replication (Bama n active form. DNA polymerase, 3‘-5‘-exonuclease, ford and Mindich, 1984; Savilahti et al., 1989). In this event and PS-dGMP replication initiation complex formation activities cosedimented in glycerol gradient with a pro- the phage-encoded DNA polymerase catalyzes the linking of tein of 65 kDa, the size expected for P R D l DNA po- dGMP intothe Tyr-190 residue in the terminal protein(Hsieh lymerase. The DNA polymerase was active on DNase et al., 1990). DNA replication system with cell extracts has I-activated calf thymus DNA, poly(dA)*oligo(dT) and indicated that replication can start from either end of the molecule and thatterminal protein and DNA polymerase are poly(dA-dT) primer/templates as well as onnative phage P R D l genome. The 3’-5’-exonuclease activity the only phage-encoded proteins needed to synthesize the was specific forsingle-stranded DNA and released full-length phage genome in vitro (Yo0 and Ito, 1989). mononucleotides. No 5’-3’-exonuclease activity was Based on amino acid sequence comparison, PRDl DNA detected.Theinhibitor/activatorspectrum of the polymerase can be classified to thefamily of eukaryotic a-like P R D l DNA polymerase was also studied. An in vitro DNA polymerases (Savilahti and Bamford, 1987; Jung et al., replication system with purified components for bac- 1987; Bernad et al., 1987; Wonget al., 1988). A special feature teriophage PRDl was established. Formation of the of the enzyme is its capability of initiating the DNA replicaP8-dGMP replication initiation complex was a prereq- tion by aprotein priming mechanism (reviewed by Salas uisite for phage DNA replication,whichproceeded (1988a, 1988b)). This type of mechanism is also used by the from the initiation complex and yielded genomelength DNA polymerases from the extensively studied linear DNA replication products. replication systems of adenovirus (Horwitz, 1986; Tamanoi, 1986; Kellyet al., 1988) and bacteriophage (629 (Salas, 1988b). The time course of the expression of the early proteins in PRD1-infected cells suggests that theexpression of the phage genome terminal protein and the DNA polymerase are tranBacteriophage PRDl belongs to a close family of lipid- scriptionally coupled (Mindich et al., 1982).This is consistent containing dsDNA bacteriophages infecting Gram-negative with the finding that no obvious promoter sequence precedes bacteria harboring P, N, or W incompatibility group plasmids. the DNA polymerase gene (Savilahtiand Bamford, 1987). Among the hosts are Escherichia coli and Salmonella typhiThe abbreviations used are: kb, kilobase(s); Ap, ampicillin; bla, murium (Olsen et al., 1974; Bamford et al., 1981). The phage p-lactamase; bp, base pair(s);BSA, bovineserum albumin; BuAdATP, * This investigation was supported by a research grant (to D. H. B.) from the Academy of Finland. Additional funding was received from the Alfred Kordelin Foundation and Finnish Cultural Foundation (to H. S.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked“aduertkement”inaccordancewith 18 U.S.C. Section 1734 solely to indicate thisfact. 11 To whom correspondence shouldbe addressed Dept.of Genetics, University of Helsinki, Arkadiankatu 7, SF-00100 Helsinki, Finland. Tel.: 358-0-1917381; Fax: 358-0-1917376.

2-(p-n-butylanilino)dATP; BuPdGTP, N2-(p-n-butylpheny1)dGTP; DMSO, dimethylsulfoxide; DTT,dithiothreitol;HEPES,N-2-hydroxyethylpiperazine-Nf-2-ethanesulfonic acid ITR, inverted terminal repeat; Km, kanamycin; NEM, N-ethylmaleimide; P I , bacteriophage P R D l DNA polymerase encoded by gene I; P8, bacteriophage P R D l genome terminal protein encoded by gene VIII; PCR, polymerase chain reaction; Klenow enzyme, large fragment of E. coli DNA polymerase I; RBS, ribosome binding site; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; Sr,streptomycin; Tc, tetracyclin; TCA, trichloroacetic acid; [ 1, designates the plasmidcarrier state.

18737

18738

BacteriophagePRDl DNA Polymerase

T h e DNA polymerase gene is expressed at a much lower level Cosedimentation of DNA polymerase,3’-5’-exonuclease, than the terminal protein gene. The low expression is sug- and protein-priming activities with P1 indicated that it was gested to result either from the lack of a RBS leading to a responsible for all these activities. The otherwell charactercoupled translation to gene VZIZ or from an inefficient RBS ized DNA polymerases from protein-primed DNA replication in front of gene Z (Savilahti and Bamford,1987). systems, adenovirus (Fieldet al., 1984) and bacteriophage 629 One line of researchinourlaboratory is aimedtoward of Bacillus subtilis (Blanc0 and Salas,1984, 1985a), also have understanding of the replication mechanism of P R D l genome, similar activities present in the single polypeptide.This might the only protein primed replication systemknown to operate be a general feature of DNA polymerases capable of protein in E. coli. As a first step, we are characterizing the products priming. Native P R D l genome served as a template for P1 of all of the phagegenes that areinvolved in thisprocess. We when primed with terminal protein P8. The enzyme could report here the overexpression, purification, and characteriza- also use a variety of DNA primer/templates such as poly(&). tion of the bacteriophage P R D l DNA polymerase. The en- oligo(dT),poly(dA-dT),andactivated DNA. Inthelatter zyme was shown to possess DNA polymerase activity and cases the PRDl DNA polymerase activity catalyzed the inprotein-priming activity, as well as 3’-5’-exonuclease activity. corporation of nucleotides from a DNA primer. These priming Replication of bacteriophage P R D l DNA was achieved with properties of P1, together with its 3’-5’-exonuclease activity purified proteins. Accordingly, this is the first time that a are in agreement with the generally accepted view of replicaprotein-primed DNA replication system originally operating tion of DNA molecules with covalently linked terminal proin E. coli. is available with purified components in uitro. teins (Salas, 1988b). Originally based on drug sensitivity, and later on amino acid sequence analysis, DNA polymerases have been divided EXPERIMENTAL PROCEDURES AND RESULTS~ into two groups: eukaryotic a-like DNA polymerases and E. coli polymerase I-like DNA polymerases (Bernad et al., 1987; DISCUSSION Jung et al., 1987; Wong et al., 1988). a-like DNApolymerases We constructed recombinant plasmids, where the bacteri- have been reported to be sensitive to the drug aphidicolin and ophage PRDl DNA polymerase gene (gene I ) was under the to certain nucleotide analogues such as BuAdATP and Bucontrol of the X p~ promoter. Because plasmid PUSH10 was PdGTP (Bernadet al., 1987). Here we have demonstrated the abletocomplement P R D l gene Z mutant phages and, in sensitivity of P R D l DNA polymerase to thesecompounds. In addition, extractsof strains harboring this plasmid were active ourhands, however, Klenowenzyme was equally or more in the protein-priming assay, we conclude that there is a sensitive to these drugs with poly(dA).(dT) primer/template. functional phage-derived RBS in front of gene Z. However, The reason for this remains unknown. The drug sensitivities the expression could not readily be detected by SDS-PAGE. of DNA polymerases may, however, depend also on reaction Downstream of gene Z, plasmid PUSH10 contained alsoa 5’- conditions. A more reliable basis for classification of DNA terminal portionof the PRDlgene XVpreceded by its original polymerases is certainly the aminoacid sequence data, which phage RBS. This truncatedgene was highly expressed allow- verifies that PRDl DNA polymerase belongs to a-like DNA ing the detectionof its gene product by SDS-PAGE. polymerases (SavilahtiandBamford, 1987; Bernad et al., To increase the gene I expression, we tailored a n efficient 1987; Jung et al., 1987). The higher sensitivity of P R D l DNA RBS in front of the gene. For this purpose, we applied ex- polymerase to NEM as well as its activation by dimethyl tended primer PCR methodology. Successful use of a similar sulfoxide clearly distinguished it from Klenow enzyme. The strategy in plasmid constructions and cloning has been de- latter observation might later be helpful in establishing a scribed(Vallette et al. 1989; MacFerrin et al. 1990). The simple andspecific assay for purificationof this protein. plasmid with changed RBS in front of gene I (pUSH100) P R D l DNA polymerase contained 3’-5’-exonuclease activdirected high level expression of P1 which could be visualized ity, which actedonsingle-strandedbutnoton doubleby SDS-PAGE (Fig. 2). Even though the P1 expression was stranded DNA and released mononucleotides. The same propincreased at least 100-fold compared with that from plasmid erties have been found to be associated with bacteriophage pUSH10, the expression of gene X V fragment was the same from both plasmids. This indicated that the changes made in front of gene Z were responsible for the increased expression of this gene. Unfortunately most of the overexpressed P1 and the N-terminal polypeptide coded by the truncated gene X V aggregated into an insoluble form. Nevertheless, the amount of the soluble form of P1 produced with plasmid pUSHlOO was considerablyhigher than that produced with plasmid pUSH10. In an attempt to purifyaggregated the material, the .polypeptides copurified showingthe heterologous composition of the aggregate (data not shown).T o overcome the aggrega- i o 3010 20 IO 20 30 IO 20 30 tion problem, gene X V from plasmid pUSHlOO was deleted, Incubation time (min) and in addition, the induction temperature was lowered. NeiFIG. 8. DNA polymerase activity of protein P1 with P R D l ther approach, however, helped to solve this problem. Therefore, purification of the small, soluble fraction of the protein DNA-PS as a template. Reactions were performed and samples processed as described under “Methods.” Reaction mixtures of 25 p1 was carried out. Plasmid pUSHllO was chosen to expressP1 contained 50 FM each dATP, dCTP, dGTP, and [3H]dTTP (0.5 pCi). for purification because it contained thewhole gene Z but only Purifiedprotein P1 (42 ng) was incubatedwith 500 ng of P R D l a small portion of gene XV. DNA-protein P8 complex (U),with 500 ng of pronase-treated P R D l

-

-c1

0 0

’ Portions

of this paper (including “Experimental Procedures,” “Results,” Figs. 1-7, and Tables I and 11) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are included in the microfilm edition of the Journal that available is from Waverly Press.

DNA (0),or without DNA (A) in the presence (panel A ) or absence (panel B ) of purified protein P8 (300 ng). Purified protein P8 in the absence of protein P1 was incubated with the same DNA templates as controls (panel C). After the indicated times a t 37 “C, the incorporation of [3H]dTMP intoa trichloroacetic acid-insoluble form was measured.

Bacteriophage DNA PRDl

Polymerase

18739

protein-priming mechanism to initiate its replication. In this event the phage DNA polymerase catalyzes the linking of dGMP to theTyr-190 residue in the terminal protein (Bamford and Mindich, 1984; Hsieh et al., 1990). Full-length PRDl genome replication in vitro has been described earlier using cell extract which contained both PRDl DNA polymerase and terminal protein (Yo0 and Ito, 1989). However, in this case the interpretation of the results does not rule out the participation of host factors in replication. We have shown here that two purified proteins, DNA polymerase and terminal protein, were sufficient to replicate the PRDlgenome in vitro when intact PRDl DNA-protein P8 template was used (Fig. 8).This in vitro replication yielded full-length products in 10 min (Fig. 9). The rate of elongation at 37 “C was approximately 25 nucleotides/s, which is close to thatobtained in the 429 system (Blanco and Salas, 1985b). Bacteriophage 429 DNA polymerase has been shown to replicate the phage DNAby a stranddisplacement mechanism supporting the symmetric model of DNA replication, where the replication can start from either terminus of the molecule (Blanco et al., 1989). In analogy to the 429 system, together FIG. 9. In vitro synthesis of P R D l DNA with purified com- with the fact that PRDlreplication can start from both ends ponents. The in oitro replication assay was carried out in a volume of the molecule (Yo0 and Ito, 1989), we assume that the of 200 pl as described under “Methods.” The reaction mixture con- mechanism of PRDl replication is similar. First protein primtained 50 p~ each dATP, dCTP, dTTP, and [a-”’PIdGTP (4 pCi), ing takes place, after which DNA polymerasecontinues elon2.4pgof purified terminal protein (PS), 336ng of PRDl DNA gation simultaneously displacing the parental strand of the polymerase (Pl), and4 pg ofPRDl DNA-protein P8. After incubation a t 37‘C, samples of25 p1 were withdrawn at the indicated times, same polarity. Ammonium ions stimulate the initiation of PRDl DNA processed, and analyzed by alkaline agarose gel electrophoresis as described under “Methods.” Lanes a-g, replication products (0, 1,2.5, replication in vitro (Savilahti et al., 1989). The elongation on 5, 10, 20, and 30 min, respectively); lane h, protease-treated PRDl poly(dA) oligo(dT) primer/template in this study, however, genome (1pg) stained with EtBr asa control. was not stimulated by ammonium indicating the specificity of stimulation to theinitiation step. Similar results have been 429 (Blanco and Salas, 1985a) and adenovirus (Field et al., achieved in the bacteriophage 429 system, where the stimu1984) DNA polymerases, and thus all so far studied DNA lation is specifically due to the stabilization of the complex polymerases with protein-priming activity share these exo- between the soluble terminal protein and DNA polymerase nuclease properties. Based on the amino acid sequence ho- prior to theinitiation of replication (Blanco et al., 1987). The mology analysis and site-directed mutagenesis, the active site specific mechanism in the PRDlsystem is stillunknown but of 3’-5’-exonuclease activity in DNA polymerases, including might well be similar to thatof 429. PRDl DNApolymerase, has beenlocalized in the three The availability of this in vitro DNA replication system and conserved regions (Bernad et al., 1989). Here we have con- a high frequencyelectroporation system (Lyra et al., 1991) for firmed at the enzymatic level that this activity is present in PRDl genome allowsthe introduction of the in vitro produced PRDl DNA polymerase. Accordingly, the active site in this molecules into the cell to test their biological activity. Furenzyme would contain the highly conservedexonuclease I, 11, thermore, since many DNAreplication mutants of E. coli and and I11 residues (amino acids 10-27,67-81, and 177-190, purified replication proteins are available, the PRDl system respectively) in the N-terminal domain of the molecule (Sav- offers a good opportunity to study the participation of possible ilahti and Bamford, 1987;Bernad et al., 1989). host factors in protein-primed DNA replication in general. The 3’-5’-exonuclease activity of DNA polymerases has been proposed to be responsible for proof-reading, which Acknowledgments-We want to thank Sisko Litmanen and Merja increases the replication fidelity (Brutlag and Kornberg, 1972; Nissinen for their skillful technical assistance. Tapio Kesti is acKunkel, 1988). It is logical to assume a similar function also knowledged for preparing different primer/templates and for helpful for P1 3’-5‘-exonuclease activity. However, the 3‘45‘- exo- discussions. Dr. G . E. Wright is appreciated for supplying the nucleonuclease activity of PRDl DNA polymerase mayalso have an tide analogues BuAdATP and BuPdGTP. additional function. It is known that both termini (ITR seREFERENCES quences) of linear DNA with terminal proteins are conserved within the molecule inspite of their physical separation Bamford, D. H., and Mindich, L. (1984) J. Virol. 50,309-315 (Savilahti and Bamford, 1986;Sakaguchi, 1990). Althoughthe Bamford, D. H., Rouhiainen, L., Takkinen, K., and Sderlund, H. (1981) J. Gen. Virol. 57,365-373 mechanism responsible for this conservation is unknown, it D., McGraw, T., MacKenzie, G., and Mindich, L. (1983)J. has been proposed that a panhandle formation between the Bamford, Virol. 47, 311-316 ITR sequences of the displaced strand might take place (Lech- Bamford, J. K. H., Hanninen, A.-L., Pakula, T., Ojala, P., Kalkkinen, ner and Kelly, 1977; Stow, 1982; Hay et al., 1984; Leegwater N., Frilander, M., and Bamford, D. H. (1991) Virology 183, 658et al., 1988; Lippe and Graham, 1989). The coevolution of 676 PRDl ITRs (Savilahti and Bamford, 1986) might be based Bernad, A., Zaballos, A., Salas, M., and Blanco, L. (1987) EMBO J. 6,4219-4225 on a similar mechanism, wherethe 3’-5’-exonuclease activity A., Blanco, L., Lazaro, J. M., Martin, G., and Salas, M. of P1 degrades the displaced single-stranded molecule from Bernad, (1989) Cell 59,219-228 the 3‘-termicus followed by DNA polymerase activity using Blanco. L.. and Salas., M. (1984) . . Proc. Natl. Acad. Sci. U.S. A. 81. the other ITR as a template. 532515329 It has been shown earlier that bacteriophage PRDl uses a Blanco, L., and Salas, M. (1985a) Nucleic Acids Res. 13,1239-1249

a b c d e

f g h

Bacteriophage PRDl DNA Polymerase

18740

Blanco, L., and Salas, M. (1985b) Proc. Natl. Acad. Sci. U. S. A. 82, 6404-6408 Blanco, L., Prieto, I., Gutierrez, J., Bernad, A,, Lazaro, J., Hermoso, J. M., and Salas, M. (1987) J. Virol. 61, 3983-3991 Blanco, L., Bernad, A., Lazaro, J . M., Martin, G., Garmendia, C., and Salas, M. (1989) J. Biol. Chem. 264, 8935-8940 Bradford, M. M. (1976) Anal. Biochem. 72, 248-254 Brutlag, D., and Kornberg, A. (1972) J. Bid. Chem. 247, 241-248 Caldentey, J., Bamford,J. K.H., and Bamford, D. H. (1990) J.Struct. B i d . 104. -,44-51 Field, J., Gronostajski, R. M., and Hurwitz, J. (1984) J . Biol. Chem. 259,9487-9495 Hay, R. T., Stow, N. D., and McDougall, I. M. (1984) J . Mol. Biol. 175,493-510 Horwitz, M. S. (1986) in Fundamental Virology (Fields, B. N., and Knipe, D. M., eds) pp. 563-606, Raven Press, New York Hsieh, J.-C., Jung, G., Leavitt, M. C., and Ito, J. (1987) Nucleic Acids Res. 15,8999-9009 Hsieh, J.-C., Yoo, S.-K., and Ito, J. (1990) Proc. Natl. Acad. Sci. U. 5’. A. 87,8665-8669 Jung, G., Leavitt, M. C., Hsieh, J.-C., and Ito, J. (1987) Proc. Natl. Acad. Sci. U. S. A . 84,8287-8291 Kelly, T. J., Wold, M. S., and Li, J . (1988) Adu. Virus Res. 34, 1-42 Kunkel, T. A. (1988) Cell 53,837-840 Laemmli, U. K. (1970) Nature 227, 680-685 Lechner, R. L., and Kelly, T. J., Jr. (1977) Cell 12, 1007-1020 Leegwater, P. A. J., Rombouts, R. F. A., and van der Vliet, P. C. (1988) Biochim. Biophys. Acta 951, 403-410 Liljestrom, P., Laamanen, I., and Palva, E. T. (1988) J. Mol.Biol. 201,663-673 Lippe, R., and Graham, F. L. (1989) J. Virol. 63, 5133-5141 Lyra, C., Savilahti, H., and Bamford, D. H. (1991) Mol. Gen. Genet., in press MacFerrin, K. D., Terranova, M. P., Schreiber, S. L., and Verdine, G. L. (1990) Proc. Natl. Acad. Sci. U. S. A. 87,1937-1941 Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY McGraw, T., Yang, H.-L., and Mindich, L. (1983) Mol. Gen. Genet. 190, 237-244 Mindich, L., and Bamford, D. H. (1988) in T h e Bacteriophages (Calendar, R., ed) Vol. 2, pp. 475-520, Plenum Press, New York Mindich, L., and McGraw, T. (1983) Mol. Gen. Genet. 190, 233-236 Mindich, L., Cohen, J., and Weisburd, M. (1976) J . Bacteriol. 126, 177-182 ~

~

~~

Mindich,L.,Bamford,D.H.,Goldthwaite, C., Laverty, M., and MacKenzie, G. (1982) J. Virol. 44, 1013-1020 Mullis, K. B., and Faloona, F. A. (1987) Methods Enzymol. 155,335350 Olkkonen, V. M., and Bamford, D. H. (1989) Virology 171, 229-238 Olsen, R. H., Siak, J.-S., and Gray, R. H. (1974) J. Virol. 14, 689699 Pakula, T. M., Savilahti, H., and Bamford, D. H. (1989) Gene (Amst.) 85,53-58 Panet, A., van de Sande, J. H., Loewen, P. C., Chorana, H. G., Raae, A. J.,Lillehang, J . R., and Kleppe, K. (1973) Biochemistry 12, 5045-5050 Remaut, E., Tsao, H., and Fiers, W. (1983) Gene (Amst.) 22, 103113 Sakaguchi, K. (1990) Microbiol. Reu. 54, 66-74 Salas, M. (1988a) Curr. Top. Microbiol. Immunol. 136, 71-88 Salas, M. (198815) in T h e Bacteriophages (Calendar R., ed) Vol. 1,pp. 169-191, Plenum Press, New York E. F., andManiatis, T. (1989) Molecular Sambrook,J.,Fritsch, Cloning: A Laboratory Manual, 2nd Ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY Savilahti, H., and Bamford, D. H. (1986) Gene (Amst.)49, 199-205 Savilahti, H., and Bamford, D. H. (1987) Gene (Amst.)57,121-130 Savilahti, H., Caldentey, J., and Bamford, D. H. (1989) Gene (Amst.) 85,45-51 Schlabach, A., Friedlender, B., Bolden, A., and Weissbach, A. (1971) Biochem. Biophys. Res. Commun. 44,879-885 Stow, N.D. (1982) Nucleic Acids Res. 10, 5105-5119 Syvaoja, J., Suomensaari, S., Nishida, C., Goldsmith, J. S., Chui, G. S. J., Jain, S., and Linn, S. (1990) Proc. Natl. Acad. Sci. U. S. A. 87,6664-6668 Tabor, S., and Richardson, C. C. (1987) Proc. Natl. Acad. Sci. U. S. A. 84,4767-4771 Tamanoi, F. (1986) in Adenovirus DNA (Doerfler, W., ed) pp. 97128, Martinus Nijhoff, Boston Vallette, F., Mege, E., Reiss, A,, and Adesnik, M. (1989) Nucleic Acids Res. 17, 723-733 Wong, S. W., Wahl, A. F.,Yuan, P.-M., Arai, N., Pearson, B. E., Arai, K., Korn, D., Hunkapiller, M. W., and Wang, T. S.-F. (1988) E M B O J. 7.37-47 Wray, W., Boulikas, T., Wray, V. P., and Hancock, R. (1981) Anal. Biochem. 118, 197-203 Yanisch-Perron, C., Vieira, J., and Messing, J. (1985) Gene (Amst.) 33, 103-119 Yoo, S.-K., and Ito, J.(1989) Virology 170,442-449

Supplementary mntrrial

10

Ovr~mpression.purification and characterization of Escherichia coli bacteriophage PRDl DNA polymerase: In uilro synthesis of full length V R D I DNA with purified proteins by liarri Savilahti. Jailer Caldentey. Kenneth LundstrBm. Juhani E. S y 6 o j a and Dennis 13. Bamford

Pnrnerftempiates .[or po~ymerasea s s a p Rnrlrnophage I X D i genome was Isolated by phenol extra~tionwilhout pronase treatment as drscribed [Lyra el ai. 1991). Calf thymus DNA was lrom Boehringer Mannhdm and it was aruvnted as described LXhlabach ei ai. 19711. Denatured calf thymus DNA was prepared by healing for 5 mi” at 90°C and rapidly cooling on ice. Poly(dAl.oligoldn was prepared by annealmg IdA),., (Slgmal and (dT),,,“ (Pharmacial in a nucleotide ratio of 1 0 1 as described (syv3aja et ai. 19901. Nternating poly IdA-dm was purchased from Sigma. Concentration of DNAs are given in nucleotides unless otherwise indicated.

l,.,

D a ~ i e l l l lsiruins. l bacletiophoges and plasmidsPlasmid p U C l 8 sequencing v ~ r f o rWanlsch-Perron el ai. 19851 and plasnild pr1857 (Remaut (11 01. 19831 rantaming lhe lambda rcpressor gene wcrc propagated on E. coli Strain D115a IGibco-BRLI. Plasmid pLM3 Illamford and Mindich 19841 ronlalning the l d l very early W l o n of phage P I W I DNA was propagated on E. coli HB94 [Mlndich and McCraw 19831. The expression V C C ~ OpPI.Hl01 ~ ILiijestram el (11. 19881 was propagalcd on E. coil strain PK4G6 iSavllahli el al. 19891 at 28°C. Recombinant MI3 phages were g r o w on E. coli Strain J M i 0 7 IYanlsch-Perron el 01. 1985). PRDl mutant phages were provldcd by Dr. Leonard Mindich. Thr Public Health Resrarrh institute. New York. Wild t y p e PRDl phage (Barnford PI nl. 19811 was propagated on S. l!,p!zimuriurn I,T21pLM21 and g e m I nwtanls Isus2 and sus38. Mindlch and McCraw 1983)on S . iyphimuriurn suppressor strdm PSAIpLMZI IMindich el al. 19821. Piasmld pLMZ IMindirh el a[.1976) encodes the receptor for PIWI

Bacteriophage PRDI D N A Polymerase

1874 1

18742

Bacteriophage PRDI DNA Polymerase 1

.II

I

IO.

I -II. IO

I I1

111 IV

\, VI

\,,.I*

I,

I

Bacteriophage PRDl DNA Polymerase

:. I

-

I

I

1

Bacteriophage PRDl DNA Polymerase 3'~s' emrt~icleuscacliuily~ To further charncterlze the nuclease activity detected in the cosedlmentation cxpenment. we rnrrled out the exonucl~ase dlrect~onallty analysis. Llnearizcd plasmld pUC 18 was labelled with "P elthcr at the 3 . or S-termlnl. Furlfled protein PI was incubated with these molecules and

rPmalning as well as releascd radloacuvlty was analyzed as described in "Melhods". Radloactivlty was released from the 9"temini of heat denatured DNA but not from the 3'~ tcrmini of natlve DNA [Fig. 7. panel AI, indlcatlng lhat the nuclea~eacllvlty was specific for SSDNA. ln control Cxperlment [PI absent) radlaacuvlly was not released from the 5"termlni of SSDNA.No release of radioactlvlty was detected from the 5"terminl of either natlve or heat denatured DNA [panel BI. The rclcasp of radioncuvlty from lhe 3~termlni of the heatdenatured DNA was also denionstrated by alkaline agarose gcI electrophoresls (Fig. 7 , panel Cl. T h e sbe of the radloactlve band remained the s a m even though the amount of radioactlvlty decreased. As enprcted. no r e i ~ a s cof radloactiwty from the 5'-termlni was detectcd (panel Dl. Furthermore. thcre was no detrClablc change in the siee of the labelled pLlCl8 DNA. This was expected. slnrc plasmid pUC18 is 2.7 kbp long and only a considerable nucleotide rclease from the 3'~ terminl could have been vlsualrzcd as a shonenlng of the radlaactwe S~lahelled molecules in this assay. The releaspd radioartmeproducts from heatdenatured 3~labdled p U C I 8 DNA were further analyzed by 20 0% PAGE (data not shown) as desrnbed In "Methods". For CompmSOn. the release enpcrlment was also conducted with Klenow ~ r u y m eAs . Internal standards we used inorganlc phosphate ['?I and Io4qdCTP. Slnce we labelled Srnd~cleavedplasmld pUC18 Wth

la~"PldCTP. theterminalnucleotlde released by Klenaw enzyme was dCMP. Radioacuve nucleotides released by prateln PI migrated as those released by Klenow ergrmc. The obvious ronclu~l~ IS. n that P I released mononucleotldes from the 3 - t e m i n l of the Ilnearlzed slnglestranded PUCIB DNA. Exonuclease expellments with I'HldT- showed that P1 can convert all radioactivity to acid soluble form (data not shownl. This verlned the S U C C ~ S S ~ Vnucleotlde ~ release by P I .

Repl!camn OJ bacteriophage P m l genome with p u r ! ! componenlr~ As descrlbed In "Methods"bactellophage PRO1 DNA repllcatlon with PI was studled uslng PRDl DNA~protein P8 (natlve PRDl genome) as a template. genome terminal prateln [Pal as a prlmer and all four nucleoudes as substrates (Fig. 8).When d l the components were present. accumulaUon of TCA-Insoluble radloactlvlly was detected (panel AI. If protease treated PRO1 genome was used as a template. or P I or P8 were omitted. no accumulation of radioactivity was detected lpanels A. B and C . respectlvelyl. lhese results lndlcated that only l h e Intact genome with termlnal proteins Could be repllcated and. In addltlon. free proteln P8 and protein P1 were both required for the repllcatlon. Thls specific repllcauon of PRDl DNA-protein P8 lndlcated that elongatlonproceeded from the P8-dCMP InlUatlon complex and that the nucleoude lncarparallon was not due to a repaldike actlvlty. As ana1y.d by alkallne agarose gel electrophoresis [Fig. 9). the length of l h e replicauonproducts Increased with tlme and genome length products were obtained already in IO mlnutes.