DNA-Dependent ATPase

Transcription

1 JOURNAL OF VIROLOGY, Jan. 1975, P Copyright American Society for Microbiology Vol. 15, No. 1 Printed in U.S.A. Characterization of a Bacteriophage T4 Mutant Lacking DNA-Dependent ATPase MARGARET T. BEHME* AND K. EBISUZAKI Cancer Research Laboratory, University of Western Ontario, London, Ontario, Canada Received for publication 15 August 1974 A DNA-dependent ATPase has previously been purified from bacteriophage T4-infected Escherichia coli. A mutant phage strain lacking this enzyme has been isolated and characterized. Although the mutant strain produced no detectable DNA-dependent ATPase, growth properties were not affected. Burst sizes were similar for the mutant phage and T4D in polal, recb, recc, uvra, uvrb, uvrc, and various DNA-negative E. coli. UV sensitivity and genetic recombination were normal in a variety of E. coli hosts. Mapping data indicate that the genetic locus controlling the mutant occurs near gene 56. The nonessential nature of this gene is discussed. In a search for new enzymes involved in DNA metabolism, a DNA-dependent ATPase was discovered in extracts of bacteriophage T4- infected Escherichia coli cells (4). This enzyme is similar to two other DNA-dependent ATPases found in E. coli (7). The purified bacteriophage enzyme is not detectable in uninfected E. coli. In the presence of DNA, large amounts of ATP are hydrolyzed to ADP and inorganic phosphate without concomitant scission or elongation of the DNA. Other properties of the bacteriophage-induced enzyme demonstrate the unusual nature of the reaction. The ATPase activity is completely dependent on the presence of DNA and is stimulated by calf thymus DNA, E. coli DNA, and heat-denatured T4 and T7 DNAs.' However, native T4 and T7 DNAs do not activate the enzyme. Since the properties of the bacteriophageinduced ATPase provide few clues as to the function of the enzyme in DNA metabolism, mutants lacking this enzyme were sought. Appropriate known T4 mutants screened for the presence of the DNA-dependent ATPase showed normal levels of enzyme activity (4). We report here the isolation and characterization of a bacteriophage T4 mutant which fails to induce detectable amounts of the enzyme. MATERIALS AND METHODS Bacterial strains. E. coli strains used included B, BK(X), B3, ER22(endoI), and CR63(su+). The dna mutants listed in Table 1 were obtained from J. A. Wechsler. The genotypes and properties of the dna mutants are described by Wechsler and Gross (18). The uvra, -B, and -C strains AB1886, 1885, and and the rec strains AB2470 and AB3058, as well as the lex strain AB2494, were from the Coli Genetic Stock Center (Yale University School of Medicine); W3110 and P3478 (polal) were obtained from J. Cairns (5). The uvrd, -E, and -F strains were obtained from I. Mahler. Strain 152 (reca) and its parent 28 were obtained from M. Meselson. Bacteriophage strains. In this work T4D (originally obtained from R. S. Edgar) was used as wild type. Other bacteriophage strains from the laboratory collection were rii strains 1367 and 723 and ame605 (ligase-). Both 1367 and 723 lie in the riia cistron and are unmapped mutations derived by aminopurine mutagenesis. Deletion mutants 1 to 10 of the del(39-56) series lying between genes 39 and 56 were obtained from T. Homyk (Theodore Homyk, Jr., and Jon Weil, Virology, in press). The newly isolated mutant bacteriophages lacking DNA-dependent ATPase are designated dda and are represented by strain L148. Bacteriophage assays. PFU were determined using the EHA-bottom layer agar and EHA-top layer agar method of Steinberg and Edgar (17). UV irradiation. Bacteriophage (1010 PFU/ml in 0.05 M phosphate buffer, ph 6.8) were irradiated with a General Electric germidical lamp (G8T5) placed at a distance of 58.5 cm. The flux density was 7 ergs/mm2 per s. Multiplicity reactivation. E. coli were grown in Fraser and Jerrel medium (8) with appropriate growth requirements to 10' cells/ml and collected and resuspended to 4 x 10' cells/ml in fresh medium containing M KCN. UV irradiation of bacteriophage was performed as described above for various time periods up to 3 min. After incubating the bacteria at 37 C for 5 min, the bacteriophage were added at a multiplicity of 5. Unadsorbed bacteriophage were removed by the addition of anti-t4 serum 5 min after infection. After incubation for a further 5-min interval, the appropriately diluted reaction mixtures were assayed for the number of infective centers on E. coli B.

2 VOL. 15, 1974 DNA-DEPENDENT ATPase MUTANT OF PHAGE T4 Preparation of bacteriophage-infected cells. E. coli ER22 (10' cells/ml) was grown in Fraser and Jerrel medium (8), centrifuged, and resuspended in an equal volume of fresh medium at 42 C. These cells were infected at a multiplicity of 5 with the phage to be tested and incubated with aeration for 12 min. After the addition of 50,g of chloramphenicol per ml, the cells were cooled rapidly in ice, collected by centrifugation, and stored at -20 C. DNA-dependent ATPase assay. To screen for bacteriophage lacking enzyme, a short assay scheme was devised. Extracts of infected cells were prepared by a modification of the polyethylene glycol-dextran procedure described by Alberts (2). Bacteriophageinfected E. coli cells (0.4 g) were suspended in 3 ml of 5 M NaCl-0.2 M Tris (ph 8.0) 0.01 M MgCl,-2 mm glutathione and placed in a 10-ml beaker, 2.2 cm in diameter. The cell suspension was cooled in an ice bath and sonicated for two 15-s periods at 60 W using a W140 Branson sonifier equipped with the standard horn, 1.27 cm in diameter at the tip. After the addition of an equal volume of 120 mg of polyethylene glycol (Carbowax 6000) per ml-80 mg of dextran 500 per ml in suspending buffer, the preparations were agitated for 2 h at 4 C. After centrifugation, the upper phase was collected and dialyzed overnight against two changes of 0.01 M Tris (ph 7.6)-0.01 M mercaptoethanol. The dialyzed cell extracts were applied to DEAE-cellulose columns (1 by 2 cm) equilibrated with 0.01 M Tris (ph 7.6)-0.01 M mercaptoethanol. After washing the column with 4 ml of M NaCl-0.01 M Tris (ph 7.6)-0.01 M mercaptoethanol, the enzyme was eluted with 8 ml of the same buffer. From this fraction, 0.05 ml was assayed for ATPase activity using standard assay conditions (4) and heat-denatured calf thymus DNA. Hydroxylamine mutagenesis. Bacteriophage were mutagenized according to the procedure of Hall and Tessman (9). A sample of the phage treated with hydroxylamine was used to prepare a high-titer phage stock. Phages were grown with E. coli CR63 in Fraser and Jerrel medium (8) and purified by differential centrifugation. This phage stock was then mutagenized with hydroxylamine, and the procedure was repeated for a total of five successive hydroxylamine treatments. RESULTS Isolation of bacteriophage T4 strains deficient in DNA-dependent ATPase. After mutagenesis with hydroxylamine, as described in Materials and Methods, bacteriophage survivors were isolated and tested for induction of DNA-dependent ATPase. Those phage stocks lacking ATPase were crossed with wild type T4D. In this way four dda strains deficient in enzyme were obtained; one of these, designated L148, was subjected to detailed characterization. Backcrossing with T4D five times ensures that most other mutations have been eliminated, but a mutator activity discussed below remained. E. coli ER22 infected with bacteriophage L148 showed no DNA-dependent ATPase activity in the short assay described in Materials and Methods. Furthermore, these infected cells contained no demonstrable enzyme activity when tested more rigorously on the larger scale previously used (4). Figure 1 shows the amount of ADP formed from ATP on DEAEcellulose fractionation of extracts from cells infected with E605 (dda+ ligase-) and L148 (ddai). Growth properties of the DNA-dependent ATPase mutant. Bacteriophage L148 is viable on E. coli B at 42 C as well as at 37 C. Since the purified ATPase activity is specifically dependent on the presence of DNA, the enzyme presumably has a role in DNA metabolism. Thus, the growth properties of bacteriophage L148 were tested with DNA-negative strains of E. coli. Under nonpermissive growth conditions for the bacterial strains, all the dna mutants listed in Table 1 supported growth of bacteriophage L148 and T4D to a similar extent. In addition to having similar burst sizes, L148 and T4D showed almost identical patterns in one-step growth curves on the dna E. coli. Thus, the latent period for L148 is normal in dna mutants (data not shown). Sensitivity to UV irradiation. To investigate the possible role of the DNA-dependent ATPase in DNA repair reactions, bacteriophage L148 was tested for UV sensitivity. After UV irradiation the survival of strain L148 on various E. coli hosts was similar to that of T4D. The E. coli strains used included uvra, -B, -C, -D, -E, and -F, reca, -B, and -C, lex, and polal. Thus none of these functions appears to be mediated by the bacteriophage DNA-dependent ATPase E a- a-20,o ELUATE VOLUME ml FIG. 1. DEAE-cellulose chromatography of extracts of bacteriophage T4-infected cells. Conditions for extraction, chromatography, and assay of ATP hydrolysis were as reported (4). Symbols: Bacteriophage L148 (dda-)-infected cells, (0); bacteriophage E605 (ligase- dda+)-infected cells, (0). 51

3 52 BEHME AND EBISUZAKI J. VIROL. TABLE 1. Growth properties of bacteriophage L148 (dda) and T4D on DNA-negative E. colia Temp Burst size Bacteria Genotype (C) T4D it4z CR34 Parent to E strains E508 dnaa E107 dnab E279 dnab E194 dnab E486 dnae E101 dnaf DG75 Parent to PC strains PC1 dnac PC3 dnag PC7 dnad W3110 Parent to P P3478 polal a All dna bacteria were grown at 25 C to 5 x 108 cells/ml in nutrient broth containing 0.25% yeast extract, M MgCl2, 0.001% gelatin, M NaCl, and 4 lag of thiamine per ml. W3110 and P3478 were grown at 37 C. The cells were collected and resuspended in fresh medium at the indicated temperature and aerated for 60 min before infection with bacteriophage at a multiplicity of 0.1 to 0.3. Preincubation was omitted in the case of W3110 and P3478. Burst sizes were calculated from at least three independent measurements of infective centers and yields which were determined by standard methods (1). The efficiency of infection was normal in all cases since the number of infective centers agreed well with the number expected from phage input. In addition, characteristics of multiplicity reactivation after UV treatment were similar for T4D and L148 in polymerase I-deficient (polal) E. coli (data not shown). Capacity for genetic recombination. Lack of the DNA-dependent ATPase may affect the capacity for genetic recombination. To test this possibility, rii markers were crossed into the mutant L148 strain. The recombination frequencies were determined and compared with the corresponding control cross (riil x rii2) for bacteriophage strains having ATPase activity. No difference in recombination frequency was noted in these crosses. Moreover, tests of recombination capacity in E. coli strains deficient in recombination (reca, -B, -C) or deficient in polymerase I (polal) revealed a normal level of recombinants (Table 2). Both riidda strains contained the mutator activity discussed below. Position of DNA-dependent ATPase on T4 linkage map. The position of the dda mutation on the T4 linkage map was determined by the use of T4 deletion mutants. Cells infected with ten different deletion mutants of bacteriophage T4 were monitored for ATPase production using the short assay described in Materials and Methods. The ATPase activity was absent in all deletion strains except (39-56)1. This places the gene that determines or modifies DNA-dependent ATPase in a region near gene 56, which is approximately 11,000 base pairs from the reference loop r1589 (see Fig. 2). The dda locus in the deletion mutants appears identical to that in strain L148, since no complementation was observed between L148 and deletion strains (39-56)8 and (39-56)10. No DNA-dependent ATPase activity was detectable in E. coli coinfected with these deletion mutants and L148 (data not shown). Mutator activity found as an independent mutation in strain L148. Bacteriophage T4 mutant L148 has an enhanced spontaneous mutation rate (Table 3). Up to a fivefold greater frequency of typical r-type plaques was observed in fresh stocks of L148 than in T4D stocks. This mutator activity was corroborated by studying the spontaneous reversion of an rii marker crossed into strain L148. The average spontaneous reversion frequency in strain rii dda was up to 100-fold greater than in the rii strain (Table 4). However, this mutator activity did not correlate with the lack of enzyme activity in strain L148. This disparity became apparent from the nature of the progeny of the following cross: rii dda- 56+ x r+ dda Progeny of the type rii 56- were selected by their inability to grow on E. coli C600 (X) due to the rii marker and their restricted growth on E. coli B due to the amber gene 56 marker (strain E51, dctpase-). Approximately one-half of these progeny were mutator strains on the basis of the high reversion frequency of the rni marker. Presumably these mutator strains should lack DNA-dependent ATPase activity if the enzyme deficiency is the cause of the mutator activity; however, in several of these isolated strains, enzyme activity was present when measured by the short assay described in TABLE 2. Effect of dda on capacity for genetic recombination r+/rii + r+ x 100 E. coli rii, x rii2 rii,dda x rii,dda 28 (parent to 152) (reca) AB1157 (parent to AB strains) AB2470 (recb) AB3058 (recc) W3110 (parent to P3478) P3478 (pola I)

4 VOL.- 15, 1975 DNA-DEPENDENT ATPase MUTANT OF PHAGE T4 53 Position of Deletion (39-56)7 (39-56)9 (39-56)4 (39-56)6 i39-56)10 (39-56)I (39-56)2 (39-56)8 (39-56)3 ( I-i DNA-dependent ATPose act ivity ddo , DISTANCE FROM r 1589 IN BASE PAIRS FiG. 2. Position of dda on T4 linkage map. The physical location and extent of deletions in del(39-56) I to 10 are as given by Homyk and Weil (Virology, in press). Note that the interval indicated for dda represents an interval within which some part of the gene rather than the entire gene must lie. Materials and Methods. Furthermore, several isolated progeny having a low reversion frequency, thus lacking mutator activity, were found to lack DNA-dependent ATPase activity. Progeny of the type rii 56+ were isolated, and 44 out of 48 were classed as mutator strains based on their high reversion index. DNAdependent ATPase activity was present in six out of ten mutator strains tested by the short assay. Also, two of the four strains with a low reversion index were found to lack enzyme activity. These results show that the mutator activity associated with strain L148 is controlled by a gene which is distinct from that controlling DNA-dependent ATPase activity (dda). DISCUSSION The biological role of the DNA-dependent ATPase discovered in bacteriophage T4- infected E. coli remains elusive. The characteristics of a bacteriophage T4 mutant dda, which fails to induce detectable amounts of the enzyme, shed little light on the function of this ATPase. Under the conditions reported in this paper, the mutant T4 strain does not shiow any abnormalities in viability, UV sensitivity, genetic recombination, or mutator activity. Thus, the DNA-dependent ATPase appears to have a nonessential nature. The mapping data indicate that the locus for the DNA-dependent ATPase falls in a region of the genome near gene 56, where few markers are known (15). Since the deletion mutants used are all viable, although they lack sizable portions of the genome in the adjacent region, the functions controlled by these genes may be nonessential. However, this area of the genome may denote necessary functions under certain restrictive conditions. Indeed, a mutant E. coli strain AR-8 has been isolated which will not support growth of bacteriophage T4 having certain deletions in a segment of the (39-56) region differing from the dda segment (A. Rodriguez and J. Weil, personal communication). Therefore, the possibility that the role of this part of the genome is obscured by host functions cannot be excluded. Appropriate E. coli strains may not be available or have been tested to clarify the nature of the T4 mutant studied here. For example, several new recombination-deficient strains of E. coli isolated by Horii and Clark (10) demonstrate the existence of alternate pathways for recombination; such host pathways may nullify the deficiencies of the bacteriophage even in E. coli strains such as reca, which have low recombination capacities. In addition, some of the bacterial mutants tested were leaky. For example, polal has normal exonuclease activity and a 0.5 to 2% wild-type level of DNA polymerase activity (12). The mutator activity associated with the DNA-dependent ATPase mutant may be inci- TABLE 3. Increase in spontaneous mutation rate for T4 strains deficient in DNA-dependent ATPasea Mutation index Strain (frequency x 104) Exptl Expt2 L148 (dda) T4D a Spontaneous mutation rate of T4 strains to r type was determined by plating fresh phage stocks on CR63 at 500 PFU/plate. The plates were scored for typical r-type plaques, and the index of mutation was determined from the ratio of r- to r+ plaques. TABLE 4. Increase in spontaneous reversion frequency for T4 strains deficient in DNA-dependent ATPase activitya Strain No. of stocks Mutation index tested (frequency x 10') L723 (rildda) (rh) a Spontaneous reversion frequency of rii strains to r+ was determined by growing stocks from isolated plaques. Stocks were grown by adding 102 to 103 PFU to 2.5 x 108 E. coli B in 5 ml of Fraser and Jerrel medium (8) and aerating overnight at room temperature. After the addition of CHCI,, the progeny were plated on E. coli B and BK to determine the reversion frequency. The spontaneous reversion index is calculated as the fraction of r+ bacteriophage in the progeny times 10".

5 54 BEHME AND EBISUZAKI J. VIRtOL. dentally interesting although it was found not to be correlated with the lack of enzyme under study here. Similar effects on mutation rates have been observed with mutations in genes of bacteriophage T4 which have functions implicated in DNA replication and repair (6). A redundancy of phage genes for DNAdependent ATPase may also account for the nonlethal nature of dda mutants. However, duplication of the ATPase gene should result in observable levels of the enzyme. No enzyme activity is detected in strain L148, suggesting that the dda gene is not identically duplicated. Nevertheless, an alternate enzyme may fill the function of the DNA-dependent ATPase which has not been detected in this work. In this regard, the other ATPase activity seen eluting later from the DEAE-cellulose column in Fig. 1 is probably due to.a bacterial enzyme (4). The possibility that the ATPase in question is modified by dda so that it is no longer detectable by our assay procedures, but is still functional, also remains. Further attempts to elucidate the biological role of the DNA-dependent ATPase may depend on the isolation of a mutant E. coli strain which will not support the growth of dda bacteriophage. More extensive study of the characteristics of the purified enzyme may also aid in this search. These studies have much relevance, since dependence on ATP hydrolysis has been observed during in vitro DNA synthesis (11, 13, 14, 16). ACKNOWLEDGMENTS We thank Catherine L. Dewey, Susan McMahon, and Linda G. Cool for technical assistance. This work was supported by the Medical Research Council and National Cancer Institute of Canada. LITERATURE CITED 1. Adams, M. H Bacteriophages. Interscience Publishers, Inc., New York. 2. Alberts, B. M Fractionation of nucleic acids by dextran-polyethylene glycol two-phase systems, p In L. Grossman and K. Moldave (ed.), Methods in enzymology, Vol. 12, part A. Academic Press Inc., New York. 3. Chase, M., and A. H. Doermann High negative interference over short segments of the genetic structure of bacteriophage T4. Genetics 43: Debreceni, N., M. T. Behme, and K. Ebisuzaki A DNA-dependent ATPase from E. coli infected with bacteriophage T4. Biochem. Biophys. Res. Commun. 41: de Lucia, P., and J. Cairns Isolation of an E. coli strain with a mutation affecting DNA polymerase. Nature (London) 224: Drake, J. W The genetic control of spontaneous and induced mutation rates in bacteriophage T4. Genetics (Suppl.) 73: Ebisuzaki, K., M. T. Behme, C. Senior, D. Shannon, and D. Dunn An alternative approach to the study of new enzymatic reactions involving DNA. Proc. Nat. Acad. Sci. U.S.A. 69: Fraser, D., and E. A. Jerrel The amino acid composition of T3 bacteriophage. J. Biol. Chem. 205: Hall, D. H., and I. Tessman T4 mutants unable to induce deoxycytidylate deaminase activity. Virology 29: Horii, Z., and A. J. Clark Genetic analysis of the rec F pathways to genetic recombination in Escherichia coli K12: isolation and characterization of mutants. J. Mol. Biol. 80: Knippers, R DNA polymerase I. Nature (London) 228: Lehman, I. R., and J. R. Chien DNA polymerase I activity in polymerase I mutants of Escherichia coli, p In R. D. Wells and R. B. Inman (ed.), DNA synthesis in vitro. University Park Press, Baltimore. 13. Mordoh, J., Y. Hirota, and F. Jacob On the process of cellular division in Escherichia coli. V. Incorporation of deoxynucleoside triphosphates by DNA thermosensitive mutants of Escherichia coli also lacking DNA polymerase activity. Proc. Nat. Acad. Sci. U.S.A. 67: Moses, R. E., and C. C. Richardson Replication and repair of DNA in cells of Escherichia coli treated with toluene. Proc. Nat. Acad. Sci. U.S.A. 67: Mosig, G A map of distances along the DNA molecule of phage T4. Genetics 59: Smith, D. W., H. E. Schaller, and F. J. Bonhoeffer DNA synthesis in vitro. Nature (London) 226: Steinberg, C. M., and R. S. Edgar A critical test of a current theory of genetic recombination in bacteriophage. Genetics 47: Wechsler, J. A., and J. D. Gross Escherichia coli mutants temperature-sensitive for DNA synthesis. Mol. Gen. Genet. 113: