Escherichia coli Deoxyribonucleic Acid Synthesis Mutants - Journal of

Vol. 124, No. 1 Printed in U.S.A.

JOURNAL OF BACTERIOLOGY, Oct. 1975, p. 167-175 Copyright 0 1975 American Society for Microbiology

Escherichia coli Deoxyribonucleic Acid Synthesis Mutants: Their Effect upon Bacteriophage P2 and Satellite Bacteriophage P4 Deoxyribonucleic Acid Synthesis DONALD W. BOWDEN,* REBECCA S. TWERSKY, AND RICHARD CALENDAR Department of Molecular Biology, University of California, Berkeley, California 94720

Received for publication 4 June 1975

Escherichia coli C strains containing different deoxyribonucleic acid (DNA) synthesis mutations have been tested for their support of the DNA synthesis of bacteriophage P2 and its satellite phage P4. Bacteriophage P2 requires functional dnaB, dnaE, and dnaG E. coli gene products for DNA synthesis, whereas it does not require the products of the dnaA, dnaC, or dnaH genes. In contrast, the satellite virus P4 requires functional dnaE and dnaH gene products for DNA synthesis and does not need the products of the dnaA, dnaB, dnaC, and dnaG genes. Thus the P2 and P4 genomes are replicated differently, even though they are packaged in heads made of the same protein. A large number of temperature-sensitive de- studying P2 and P4 DNA synthesis. Ideally, we oxyribonucleic acid (DNA) synthesis mutants wanted to have mutant dna genes in a backhave now been isolated in Escherichia coli (6, ground in which [8H]thymidine labeling was 14, 17, 27, 33). These mutations define genes optimal (thy-) and in which host DNA synthebelieved to be directly involved in replicating sis could be selectively eliminated by treatment the E. coli genome. with ultraviolet light or mitomycin C (uvrA-) Even though the proteins coded by the dna (23) genes have in many cases been purified (for Our work was helped by Dumas et al. (12, summary, see reference 29), the exact role 13, 18) and Sakai et al. (27), who, in order which many of them play in vivo or in vitro is to study 4X174 DNA synthesis, isolated temnot known. The notable exceptions are that perature-sensitive dna mutants from E. coli DNA polymerase III has been identified as the C parents that were thy - and uvrA -. As a product of the dnaE gene (15), and there is consequence, we were required to construct evidence suggesting that dnaG codes for a only a few E. coli C strains carrying mutant ribonucleic acid-synthesizing activity (J. dna alleles. To obtain conclusive results we used two Bouche, K. Zechel, and A. Kornberg, J. Biol. types of experiments to study P2 and P4 reChem., in press). A considerable amount of information about quirements for the E. coli dna genes. First, we the mechanism of DNA synthesis has been measured the burst from P2- or P4-infected cells obtained by studying the requirements of small, at 30 and 42 C. Next, we measured directly the single-stranded DNA viruses for the E. coli dna amount of DNA synthesis in a P2- or P4genes (29). We thought that by studying the infected host after mitomycin C treatment by requirements of P2 bacteriophage (5) and its the incorporation of labeled thymidine into satellite phage P4 (31) for the E. coli dna genes acid-insoluble material. we might obtain a clearer picture of P2 and P4 MATERIALS AND METHODS DNA replication. Bacteria and bacteriophage. Bacterial strains are Until recently the in vivo study of P2 and P4 phage requirements for DNA replication has listed in Table 1. P2 vir22 is insensitive to P2 been limited, due to the lack of appropriate immunity (10). P2 virl aml2 is defective in cell lysis strains. Temperature-sensitive dna mutants (20). P2 virl is defective in establishment of immubut is sensitive to P2 prophage repression (3). P2 have almost always been isolated from E. coli nity Ig cc (2) was used to make lysogenic strains of H502, K-12 strains (17), whereas P2 adsorbs and grows HF4704, and their derivatives. P4 virl is a clear considerably better on E. coli C strains (5). In plaque mutant (24). P2 and P4 stocks were prepared addition, the genetic backgrounds in which as described previously (19 and 16, respectively). most dna mutants have been isolated, while C-1055 and C-1197 were used as indicator strains for good for bacterial genetics, are inappropriate for P2 and P4, respectively. The terms P2 and P4 are used 167

168

BOWDEN, TWERSKY, AND CALENDAR

J. BACTERIOL.

TABLE 1. Bacterial strainsa Strain

E. coli C strains C-1055 C-1197 C-1521 H502 and derivatives H502 LD301 LD312 LD331 HF4704 and derivatives HF4704 C-1900 C-2128 C-2301 C-2307 C-2309 HF4704S E. coli K strains PC314 S. typhimurium strains RC903

Genotype

Source or reference

F+ grthr leu xan his str F + ,rthr leu xan his str (P2) his metE dnaA46 rha

(37) (25) (21)

uvrA thyA endoI uvrA thyA endoIdnaE uvrA thyA endoI dnaB uvrA thyA endoIdnaC

(12) (12) (13) (18)

uvrA thy uvrA thy ilv uvrA thy str uvrA thy str toiC uvrA thy dnaA46 uvrA thy dnaG3 str uvrA thy dnaH

(23) D. Denhardt M. Sunshine This paper This paper This paper (27)

Hfr C dnaG3

P. Carl

Col El

(36)

Lysogenic strains carrying P2 Ig cc were derived from HF4704, H502, and derivatives; lysogens were used where noted. Allele numbers, where assigned, are from Gross (17). a

to indicate P2 vir22 and P4 virl, respectively, unless noted otherwise. P1 1127 is a high-frequency transducing phage mutant (34) obtained from P. Harriman. It will be called P1 hereafter. P1 transducing stocks were grown as described by Wall and Harriman (34), except that LB broth was used rather than P1 broth, and stocks were grown at 33 C. Media. TPG basal medium is described by Lindqvist and Six (24). Super TPG (25) with the addition of thymidine, but without the addition of other nucleic acid precursors and spermidine, was used for radioactive labeling. LB broth (4) with NaCl reduced to 0.1 M was used for experiments which did not involve labeling. LC agar, LB agar (4) with 2.5 mM CaCl,, was used for phage assays. Davis minimal agar (28) was used where appropriate in transduction experiments. KAB adsorption buffer contains 10 mM tris-

(hydroxymethyl)aminomethane - hydrochloride, pH 7.4, 10 mM MgCl,, 100 mM NaCl, and 5 mM

CaCl,. Special chemicals. Mitomycin C and sodium deoxycholate were purchased from Sigma Chemical Co. [methyl-PH]thymidine was purchased from ICN Corp. Construction of strains. Transductions were carried out by modification of the procedure described by Wall and Harriman (34). For transductions involving temperature-sensitive mutations, the recipient cells were grown in LB broth at 30 C. P1 infection was for 30 min at 33 C. P1-infected cells were plated with P1 antiserum (K = 1) to prevent infection on the plates. C-2307 was constructed by transduction of C-1900 to ilv+ with P1 grown on C-1521. One of 25 ilv+ recombinants also received the dnaA46 mutation. C-2309 (dnaG3) was constructed by co-transduc-

tion of the dnaG3 mutation with tolC+. This was achieved by using P1 grown on PC314 (dnaG3 tolC+) to infect an E. coli C tolC- strain, C-2301. Some tolC+ recombinant colonies also carried dnaG3. This approach was suggested by P. Carl. C-2301 was isolated by spotting samples of a chloroform-treated overnight culture of RC903, a colicin El-carrying Salmonella typhimurium, on a lawn of C-2128 supplemented with streptomycin. Colonies which grew were tested for their sensitivity to deoxycholate (36). One of the colonies, C-2301, was shown to be tolC- by its insensitivity to colicin El produced by RC903 and by its sensitivity to concentrations of deoxycholate, which did not affect the growth of the parent strain C-2128. The transduction experiment was carried out in the following manner. P1 grown in PC314 (dnaG3) was used to infect C-2301. The infected cells were plated on LC agar plates and incubated at 30 C for 3 to 5 h, at which time the infected cells were covered with an overlay of LC agar containing 0.2 ml of 5% sodium deoxycholate. Deoxycholate-resistant recombinants were then tested for co-inheritance of temperature sensitivity. Twelve of 50 colonies were temperature sensitive. This is in agreement with results obtained by P. Carl (personal communication). Burst experiments. E. coli strains were grown in LB broth to a density of approximately 2 x 108 cells/ml. Ten milliliters of cells was centrifuged at 5,000 x g for 10 min and then suspended in 1.0 ml of KAB adsorption medium. The resuspended cells were divided in half; one half was placed at 42 C and the other at 30 C. Wild type and strains containing immediate stop dna mutations (dnaB-, dnaG-, and dnaE-) were incubated for 15 min, whereas slow stop mutants (dnaA-, dnaC-, and dnaH-) were incubated for 50 min. After this preincubation period, P2 vir22 or

P2 AND P4 DNA SYNTHESIS IN E. COLI dna MUTANTS

VOL. 124, 1975

P4 virl was added to each half of the bacteria at a multiplicity of 10 phage/bacterium. The infected cells were incubated 5 min more, and then anti-P2 antiserum was added to inactivate unadsorbed phage. The final K was equal to approximately 5. The cells were incubated an additional 5 min and were vortexed each minute. At the end of 5 min, a sample of the infected cells was diluted 10-' into 10 ml of prewarmed LB broth plus 20 Ml of 4% ethylenediaminetetraacetic acid. The P2 vir22-infected cells were then incubated at 42 or 30 C for 90 min, and P4 virl-infected cells were incubated for 120 min. The number of productively infected cells, called infective centers, was determined by plating a sample of the infected cells on the appropriate indicator at 30 C immediately after the 10-' dilution. The number of unadsorbed phage was determined by adding a sample of infected cells to chloroform immediately after the 10-4 dilution and then plating at 37 C on the appropriate indicator. The burst of P2 vir22 or P4 virl was measured by plating samples of the diluted infected cells at 37 C after the 90- or 120-min incubation. Nonlysogenic strains were used as host for P2 vir22 infections, whereas P2 Ig cc lysogens were the hosts for P4 virl infections. P4 infection of a P2 lysogen leads to a burst of phage, all of which are P4 (31). Measurement of DNA synthesis. Host DNA synthesis was stopped by treatment of E. coli uvrA cells with mitomycin C, as described by Lindqvist and Sinsheimer (23). Ten milliliters of the strain to be tested was grown to a density of approximately 2 x 108 cells/ml at 30 C in super TPG which contained 10 gg of thymidine per ml. At this time, 0.5 ml of 1 mg of mitomycin C per ml was added to the culture for HF4704 and its derivatives, or 0.65 ml for H502 and its derivatives. The cells were then incubated in the dark for 15 to 25 min at 33 C. The cells were then membrane filtered (Millipore Corp.), washed with 7.5 ml of TPG basal medium, and suspended in 1.0 ml of KAB buffer. A 0.1-ml amount of cells was then added to each of six tubes. Tubes numbered 1, 3, and 5 were placed at 42 C and tubes numbered 2, 4, and 6 were placed at 33 C. The tubes were preincubated for 15 or 50 min for different strains as described in the preceding section. At the end of preincubation P2

vir22 at a multiplicity of 10 phage/bacterium was added to tubes 1 and 2. P4 virl at a multiplicity of 10 was added to tubes 3 and 4. Tubes 5 and 6 were uninfected controls. Five minutes after infection 0.9 ml of prewarmed super TPG (2.5 Mg of thymidine per ml) containing 12 MCi of [methyl-'H]thymidine per ml was added to each tube. Samples (50 Ml) were taken from each tube, spotted on Whatman 3 MM filter paper, and placed in cold 5% trichloroacetic acid at intervals after infection. The filters were batch washed and counted by the procedure of Wolf (38). We used 33 C as the permissive temperature for measuring [IH]thymidine incorporation because P4 synthesizes DNA poorly at 30 C. HF4704 is thy- at 37 C, but will form colonies on minimal agar at 30 C (11; D. Bowden, unpublished data). H502 will not form colonies at 30 C on minimal agar. Incorporation of ['H ]thymidine into mitomycin C-treated HF4704 after phage infection was similar to incorporation into H502 under the same conditions (at 33 or 42 C). Upon the introduction of a DNA synthesis mutation into HF4704, incorporation was drastically reduced at both permissive and restrictive temperatures compared to HF4704. This problem was overcome by growing HF4704 and its derivatives with 0.5 mg of deoxyadenosine per ml in the medium and labeling with 1.0 mg of deoxyadenosine per ml in the labeling medium (7). With the addition of deoxyadenosine, which promotes uptake of exogenous thymidine (7), incorporation in the mutant strains was similar to the HF4704 parent under permissive conditions. RESULTS

Growth of P2 and P4 on H502 and its derivatives. Starting with an E. coli C parent strain, H502, L. B. Dumas isolated three strains which have temperature-sensitive mutations in the dnaB, dnaC, and dnaE genes (12, 13, 18). Our first step in investigating whether P2 or P4 required these genes for growth consisted of measuring the number of phage resulting from an infection at permissive (30 C) or restrictive temperature (42 C). Table 2 shows the results from burst experi-

TABLE 2. Growth of P2 and P4 in H502 and dna mutant derivatives at 30 and 42 C Infecting

St.a

phage P2 P4 P2 P4 P2 P4 P2 P4

Relevant

genotype

H502 H502(P2) LD312 LD312(P2) LD301 LD301(P2) LD331 LD331(P2)

169

Wild type Wild type dnaB dnaB dnaE dnaE dnaC dnaC

aLysogenic strains carry P2 Ig cc. b Each result is the average of at least two experiments.

Burst/infective center" 42 C

290 97 0.8 51 20 13 71 84

Ratio

30 C

(42 C/30 C)

202 129 242 123 150 210 120 87

1.4 0.75 0.003 0.41 0.13 0.06 0.60 0.97

170


J. BACTERIOL.

ments on strains containing dnaB, dnaE, and that treating uvr- cells with mitomycin C prior dnaC mutations in addition to their parent to virus infection depressed host DNA synthesis. strain H502. Infection of the wild-type strain Phage DNA synthesis could be clearly seen by H502 or H502 (P2 Ig cc) by P2 or P4 results in measuring [3H ]thymidine incorporation into a substantial burst at 30 or 42 C. In addition, a acid-insoluble material. We used this method to substantial burst of progeny phage comes from measure P2 and P4 DNA synthesis in H502 each infected cell at 30 C in the dna - mutants. and the temperature-sensitive DNA synthesis At restrictive temperature (42 C) this is not true. mutants derived from H502. The results of these P2 and P4 infection of strains carrying a tem- experiments are shown in Fig. 1. Fig. 1A shows the incorporation of [3H Ithyperature-sensitive mutation in dnaB (LD312) gave an interesting result (Table 2). The burst midine into continuously labeled, mitomycin Cof P2 in LD312 is drastically reduced at 42 C treated, virus-infected H502 (P2 Ig cc). This compared to 30 C. This result agrees with sim- control experiment on a wild-type host exhibits ilar experiments performed by M. Sunshine a pattern with which to compare similar experiet al. (32). This result is in contrast to the re- ments on mutant derivatives of H502. First, the sult obtained with P4 infection, which showed a amount of P2 DNA synthesized at 42 C is aplarge burst at 42 C. Whereas the ratio of the proximately one-half the amount synthesized at burst at 42 C to the burst at 30 C is reduced 33 C over the time course of the experiment. compared to the infection of the wild-type On the other hand, the amount of P4 DNA parent, the reduction is nothing like that seen synthesized at 42 C is three times the amount with P2 infecting LD312 at 42 C. These results synthesized at 33 C. The amount of label inled us to the conclusion that the dnaB gene is corporated into host DNA (uninfected cells) is necessary for P2 but is not necessary for the small at both temperatures. The same experiment using LD312 (P2 Ig cc) growth of P4. Similar results are reported by (dnaB-) as a host is shown in Fig. 1B. The E. W. Six (Virology, in press). Infection of the dnaE- strains LD301 and pattern of P2 DNA synthesis is clearly different LD301 (P2 Ig cc) by P2 and P4, respectively, compared to Fig. 1A. P2 makes a reduced but results in a considerable reduction of progeny significant amount of DNA at 33 C but shows phage at 42 C. Both the P2 and P4 bursts at 42 C no DNA synthesis at 42 C. This confirms the are reduced about 90% compared to the burst at result obtained with one-step growth experi30 C. Comparing these results with the results ments: P2 requires dnaB. In contrast to the P2 from infection of the wild-type H502 led us to result, the pattern of P4 DNA synthesis seems conclude that a functional dnaE-coded protein virtually the same in H502 (P2 Ig cc) and LD312 (P2 Ig cc). This, again, is in agreement is required for the growth of P2 and P4 phages. A different picture results from infection of with the result of one-step growth experiments LD331 (dnaC-). The bursts of P2 and P4 at examining P4 infection of LD312; P4 does not 30 and 42 C are both large. The ratio of the require dnaB. progeny at 42 versus 30 C is reduced in P2 The pattern of P2 and P4 DNA synthesis in infection compared to H502, but is still large. LD331 (P2 Ig cc) (dnaC-) (Fig. 1C) is very In the case of P4 infection the ratio is even similar to the pattern obtained from infecting higher than for H502. These results indicate H502 (P2 Ig cc). Twice as much P2 DNA is that a functional dnaC protein is not required synthesized at 33 as at 42 C, and one-third as for P2 or P4 growth. much P4 DNA is synthesized at 33 as at 42 C. [3H]thymidine labeling of P2 and P4 DNA This result again substantiates the results from after infection of H502 and mutant deriva- one-step growth experiments. P2 and P4 do not tives of H502. The burst of phage resulting require dnwC for DNA synthesis. from a bacteriophage infection is the result of a Measurement of P2 or P4 DNA synthesis in large number of metabolic steps, many of which the dnaE- strain LD301 (P2 Ig cc) (Fig. 1D) are only indirectly involved in DNA synthesis confirms the results obtained in one-step growth (8). It is possible that a mutation for DNA experiments. Both P2 and P4 incorporate label synthesis may exert pleiotropic effects which into DNA at 33 C. P2 DNA synthesis seems to affect the maturation of progeny phage but do be completely stopped at 42 C though. On the not directly inhibit phage DNA synthesis. To other hand, P4 DNA synthesis is still at a eliminate this possible artifact we also measured significant level above background. This P4 inthe DNA synthesis by P2 vir22 or P4 virl corporation is greatly.reduced compared to P4 in infected cells using [3H Ithymidine labeling. incorporation in infected wild-type cells. ConLindqvist and Sinsheimer (23) have shown sequently, we interpret this result as showing


VOL. 124, 1975

171

0 0 0 It

E 0 I'

~~jJ~~~un~~~nin eted

J

U.924jj..

150 120 90 60 30 0 150 M inutes ofter InfectI0n FIG. 1. DNA synthesis in mitomycin-treated H502 and mutant derivatives after infection with P2 vir22 or P4 (P2 virl. (A) Wild-type H502 (P2 Ig cc); (B) LD312 (P2 lg cc) (dnaB-); (C) LD331 (P2 Ig cc) (dnaC-); (D) LD301@E Ig cc) (dnaE-). Symbols: A, P2 at 33 C; A, P2 at 42 C; 0, P4 at 33 C; 0, P4 at 42 C; 0, uninfected at 33 *, uninfected at 42 C. 0

30

60 90

120

that P4 requires a functional dnaE product to synthesize its DNA. Growth of P2 and P4 on HF4704 and its derivatives. In addition to H502-derived strains, we used strains derived from HF4704 for one-step growth and [3H ]thymidine labeling experiments. The growth of P2 and P4 in HF4704, C-2307 (dnaA-), and HF4704S (dnaH-) is shown in Table 3. One can see from P2 infection of HF4704 and C-2307 that, although the bursts in C-2307 at 42 and 30 C are smaller than in HF4704, the ratios of progeny P2 at 42 C to progeny at 30 C are almost the same. P4 in-

fection of P2 Ig cc lysogens of C-2307 and HF4704 also yield very similar bursts and ratios. Consequently, we think P2 and P4 do not require a functional dnaA gene product. The results of P2 infection of HF4704S (dnaH-) proved to be more difficult to interpret. The ratio of bursts at 42 and 30 C are reduced sixfold from the value for a wild-type infection. This initially led us to believe that dnaH might be required for P2 growth. Measurement of DNA synthesis in this strain after P2 infection did not support this conclusion, however (see below). P4 infection of HF4704S (P2 Ig cc) resulted in


J. BACTERIOL.

100-fold-reduced burst at 42 C. From this result we concluded that P4 requires the dnaH

A number of conclusions can be drawn from this experiment. First, P4 can grow at 42 C in a host carrying a mutation in the dnaG gene. In addition, the mutant dnaG-carrying chromosome is affected at 42 C in such a way that P4 cannot activate prophage P2 late genes sufficiently to give a burst. This block can be overcome by use of a co-infecting P2. [3H]thymidine labeling of P2 and P4 DNA after infection of HF4704 and mutant derivatives of HF4704. To clarify the results obtained in burst experiments we measured P2 and P4 DNA synthesis in mitomycin C-treated HF4704 and its derivatives. The results are shown in Fig. 2. The kinetics of P2 vir22 or P4 virl DNA synthesis in HF4704 (P2 Ig cc) (Fig. 2A) do not differ appreciably from the pattern in H502 (P2 Ig cc) (Fig. 1A). P2 DNA synthesis at 42 C is less than P2 DNA synthesis at 33 C. P4 DNA synthesis at 42 C is still about three times the P4 DNA synthesis at 33 C. Fig. 2B shows the results from the same experiment performed on C-2307 (P2 Ig cc), which carries the mutation dnaA46. The patterns of P2 or P4 DNA synthesis are seen to be unchanged in C-2307 (P2 Ig cc) when compared to infection of HF4704 (P2 Ig cc). These results concur with the conclusion of the one-step growth experiment: P2 and P4 do not require dnaA.

172 a

protein for growth. Growth of P2 and P4 in C-2309 and C-2309 (P2 Ig cc). Infection of C-2309 (dnaG3) gave the results shown in Table 4. At 30 C the burst of P2 is substantial, whereas at 42 C the P2 burst is reduced 1,000-fold. This indicates that P2 requires a functional dnaG protein for growth. Initially we thought P4 also required dnaG for growth. P4 infection of C-2309 (P2 Ig cc) (Table 4) gave a very small burst at 42 compared to 30 C. Measurement of P4 DNA synthesis in C-2309 (P2 Ig cc) showed, however, that P4 does indeed synthesize DNA at 42 C in a dnaG3 host (see below). These contradictory results were further examined. P4 requires P2 late -genes to produce progeny phage (Six, Virology, in press). If the presence of the dnaG3 mutation leads to a functional breakdown of the E. coli DNA at 42 C, the lysogenic helper P2 Ig will also be affected. One consequence of this inactivation would be the inability of P4 to activate P2 late genes. Thus, even if P4 DNA replicated, no burst would result. To investigate this possibility we infected C-2309 (P2 ig cc) with both P4 and P2 virl aml2. The results presented in Table 4 show that the burst of P4 at 42 C is now the same as the burst of P4 at 30 C. cc

TABLE 3. Growth of P2 and P4 in HF4704 and dna mutant derivatives at 30 and 42 C Infecting phage

St*0

P2 P4 P2 P4 P2 P4

HF4704 HF4704(P2) C-2307 C-2307(P2) HF4704S

Burst/infective centerb 42 C 30 C

Relevant

train"

genotype

Wild type Wild type dnaA dnaA dnaH dnaH

HF4704S(P2)

Rtio (42 C/30 C)

363 130

426 69 249 77 37 0.2

1.2 0.53 1.0 0.41 0.20 0.003

242 186 184 57

Lysogenic strains carry P2 Ig cc. 'Each result is the average of at least two experiments.

a

TABLE 4. Growth of P2 and P4 in C-2309 and C-2309(P2)

Infecting phage P2 P4 P4 + P2 virl aml2

Strain'

C-2309

dnaG3

C-2309(P2) C-2309(P2)

dnaG3 dnaG3

C-2309(P2) is lysogenic for P2 Ig cc. Each result is the average of two experiments. c Burst of P4.

a

"

Relevant genotype

Burst/infective center' 42 C

30 C

0.2 2.6

177 45

38c

34c

(42 Rtio C/30 C) 0.001 0.06

1.1

VOL. 124, 1975


173

Minutes after infection FIG. 2. DNA synthesis in mitomycin-treated HF4704 and mutant derivatives after infection with P2 vir22 or P4 virl. (A) Wild-type HF4704 (P2 lg cc); (B) C-2307 (P2 lg cc) (dnaA-); (C) C-2309 (P2 lg cc) (dnaG-); (D) HF4704S (P2 lg cc) (dnaH-). Symbols: A, P2 at 33 C; A, P2 at 42 C; 0, P4 at 33 C; *, P4 at 42 C; 0, uninfected at 33 C; U, uninfected at 42 C.

In Fig. 2C the DNA synthesis in a dnaG3 host is shown. P2 DNA synthesis is reduced at 33 C, and at 42 C there appears to be no P2 DNA synthesis. The conclusion is that P2 requires a functional dnaG product for DNA synthesis. This is in agreement with the results of the onestep growth experiments. In the same figure one can also see that P4 synthesizes DNA quite well at 42 C. At 33 C P4 DNA synthesis was, as in the wild-type infection, about one-third the amount of DNA synthesis at 42 C. These results led us to conclude that P4 does not need the dnaG gene to synthesize DNA. P2 and P4 DNA synthesis in HF4704S (P2 Ig cc) is shown in Fig. 2D. In contrast to the

results of the one-step growth experiments P2 DNA synthesis appears to be unaffected at 42 C in the dnaH- host. We consider [8H Ithymidine incorporation a better indicator of DNA synthesis and consequently believe dnaH is not required by P2 for DNA synthesis. The pattern of P4 DNA synthesis in HF4704S (P2 Ig cc) is very similar to P4 DNA synthesis in the dnaE- strain LD301 (P2 Ig cc) (see Fig. 1D). P4 apparently synthesizes some DNA at 42 C, but this is still reduced from the amount synthesized upon infection of a wildtype strain. This result, in agreement with the one-step growth results, shows that dnaH is required for P4 DNA synthesis.

174


J. BACTERIOL.

DISCUSSION Prior to this work there were indications that P2 and its satellite phage P4 replicated by different modes. P2 was known to require the non-essential host gene rep for DNA replication (9), whereas P4 can replicate in a rep- mutant (24). In addition P2 DNA replication is known to be unidirectional (30), whereas P4 replicates bidirectionally (R. Inman, personal communication). The product of both replicating DNA molecules is a closed circular monomer which is packaged into heads (22, 24; G. Pruss, J. C. Wang, and R. Calendar, J. Mol. Biol., in press) composed of P2 proteins (31). This paper presents data which show that P2 and P4 require different host dna genes for replication. P2 requires the products of the elongation genes dnaB, dnaE, and dnaG to synthesize DNA (Fig 1 and 2; Tables 2 and 4). In contrast the only host gene which P4 requires is the dnaE gene, which codes for DNA polymerase III (15) (Fig. 1 and 2; Tables 2, 3, and 4). The E. coli dnaG protein appears to be involved in chain initiation (26). Recent work with the phage G4 suggests that the dnaG protein is a ribonucleic acid polymerase involved in synthesizing ribonucleic acid primers for subsequent DNA synthesis (Bouche et al., J. Biol. Chem., in press). P4 phage DNA syn. thesis is independent of dnaG and is also rifampin resistant (1). Thus, P4 DNA synthesis does not require the host enzymes known to initiate chains in DNA synthesis (29). This data would imply that P4 codes for a ribonucleic acid-polymerizing activity involved in DNA synthesis. There is indeed evidence for such a phage-coded protein (1). P4-infected cells contain a 90,000-dalton protein which synthesizes poly(rG) on poly(dG):(dC) (1). Amber mutants in the P4 A gene do not synthesize P4 DNA (16) and do not make the 90,000-dalton A protein (Barrett, Marsh, and Calendar, manuscript in preparation). The apparent requirement of P4 for dnaG (Table 4) when P2 is supplied as a prophage can be explained in terms of the late gene expression of prophage P2 rather than as being the result of a block in DNA synthesis. If the dnaG3 allele leads to breakdown of host DNA at 42 C, a P2 helper prophage will also be broken down. Temperature-sensitive dnaG mutants show degradation of their DNA after being shifted to 42 C (C. Sevastopoulos, personal communication). If P4 cannot transactivate P2 late proteins, it will not produce mature P2 phage particles (31). The host dnaB protein mig4t also

be replaced by the P4 gene A protein or another P4 protein. We do not favor the latter possibility, because P4 DNA has a molecular weight of only 7 x 106 (35), and no P4 genes have been implicated in the synthesis of DNA except the P4 gene A. We presently think a dnaB-like activity is not essential for P4 DNA synthesis. We were surprised that P2 and P4 showed different requirements for dnaH (Fig. 2; Table 3). A revertant of HF4704S (dnaH-), which forms colonies at 42 C, shows the same incorporation kinetics as wild-type HF4704 when infected with P2 or P4 at 33 or 42 C (data not shown). This would imply that the mutant dnaH gene is the cause for the suppression of P4 DNA synthesis at 42 C (Fig. 2). In addition, P4 does not give a burst when it co-infects with a P2 helper (data not shown). Reports of requirement or nonrequirement of host dna genes for particular phages have proven incorrect in the past due to leakiness, indirect effects, and unknown outside mutations. By using one-step growth experiments and measuring DNA synthesis directly we have tried to decrease the chances that our results are due to the effects mentioned above. In the future we hope to characterize P2 and P4 DNA replication in greater depth by studying P2 and P4 requirements for the other known dna genes (6, 14, 33). In addition, we are investigating the possibility that P2 or P4 may code for proteins whose function is similar to the products of non-essential host dna genes. ACKNOWLEDGMENTS This investigation was supported by Public Health Service research grant AI-08722 from the National Institute of Allergy and Infectious Diseases and predoctoral training grant GM 01389 from the National Institute of General Medical Sciences and by grant BMS 74-19607 from the National Science Foundation. R. S. T. was supported by the summer research associate program of the California Heart Association. We thank P. Carl, L. B. Dumas, P. Harriman, and T. Komano for generously providing bacterial and phage strains.

LITERATURE CITED 1. Barrett, K. J., W. Gibbs, and R. Calendar. 1972. A 2. 3. 4.

5.

transcribing activity induced by satellite phage P4. Proc. Natl. Acad. Sci. U.S.A. 69:2986-2990. Bertani, G., B. K. Choe, and G. Lindahl. 1969. Calcium sensitive and other mutants of bacteriophage P2. J. Gen. Virol. 5:97-104. Bertani, L. E. 1957. The effect of the inhibition of protein synthesis on the establishment of lysogeny. Virology 4:53-71. Bertani, L. E., and G. Bertani. 1970. Preparation and characterization of temperate, non-inducible bacteriophage P2 (host: Escherichia coli). J. Gen. Virol. 6:201-212. Bertani, L. E., and G. Bertani. 1971. Genetics of P2 and related phages. Adv. Genet. 16:199-237.

VOL. 124, 1975


6. Beyersmann, D., W. Messer, and M. Schlicht. 1974. Mutants of Escherichia coli B/r defective in deoxyribonucleic acid initiation: dnaI, a new gene for replication. J. Bacteriol. 118:783-789. 7. Boyce, R. P., and R. B. Setlow. 1962. A simple method of increasing the incorporation of thymidine into the deoxyribonucleic acid of Escherichia coli. Biochim. Biophys. Acta 61:618-620. 8. Calendar, R. 1970. The regulation of phage development. Annu. Rev. Microbiol. 24:241-296. 9. Calendar, R., B. Lindqvist, G. Sironi, and A. J. Clark. 1970. Characterization of REP- mutants and their interaction with P2 phage. Virology 40:72-83. 10. Chattoraj, D. K., and R. B. Inman. 1972. Position of two deletion mutations on the physical map of bacteriophage P2. J. Mol. Biol. 66:423-434. 11. Denhardt, D. T., M. Iwaya, and L. L. Larison. 1972. The rep mutation. II. Its effect on Escherichia coli and on the replication of bacteriophage OX174. Virology 49:486-496. 12. Dumas, L. B., and C. A. Miller. 1973. Replication of OX174 DNA in a temperature-sensitive dnaE mutant of Escherichia coli C. J. Virol. 11:848-855. 13. Dumas, L. B., and C. A. Miller. 1974. Inhibition of bacteriophage OX174 DNA replication in dnaB mutants of Escherichia coli C. J. Virol. 14:1369-1379. 14. Filip, C. C., J. S. Allen, R. A. Gustafson, R. G. Allen, and J. R. Walker. 1974. Bacterial cell division regulation: characterization of the dnaH locus of Escherichic coli. J. Bacteriol. 119:443-449. 15. Gefter, M. L., Y. Hirota, T. Kornberg, J. Wechsler, and C. Barnoux. 1971. Analysis of DNA polymerases II and III in mutants of Escherichia coli thermosensitive for DNA synthesis. Proc. Natl. Acad. Sci. U.S.A. 68:3150-3153. 16. Gibbs, W., R. N. Goldstein, R. Wiener, B. Lindqvist, and R. Calendar. 1973. Satellite bacteriophage P4: characterization of mutants in two essential genes. Virology 53:24-39. 17. Gross, J. D. 1972. DNA replication in bacteria. Curr. Top. Microbiol. Immunol. 57:39-74. 18. Kranias, E. G., and L. B. Dumas. 1974. Replication of bacteriophage OX174 DNA in a temperature-sensitive dnaC mutant of Escherichia coli C. J. Virol. 13: 146-154. 19. Lengyel, J. A., R. N. Goldstein, M. Marsh, M. G. Sunshine, and R. Calendar. 1973. Bacteriophage P2 head morphogenesis: cleavage of the major capsid protein. Virology 53:1-23. 20. Lindahl, G. 1971. On the control of transcription in bacteriophage P2. Virology 46:620-633. 21. Lindahl, G., Y. Hirota, and F. Jacob. 1971. On the process of cellular division in Escherichia coli: replication of the bacterial chromosome under control of prophage P2. Proc. Natl. Acad. Sci. U.S.A. 68:2407-2411.

175

22. Lindqvist, B. H. 1971. Vegetative DNA of temperate coliphage P2. Mol. Gen. Genet. 110:178-196. 23. Lindqvist, B. H., and R. L. Sinsheimer. 1967. Process of infection of bacteriophage OX174. XIV. Studies on macromolecular synthesis during infection with a lysis defective mutant. J. Mol. Biol. 28:87-94. 24. Lindqvist, B., and E. Six. 1971. Replication of bacteriophage P4 DNA in a nonlysogenic host. Virology 43:1-7. 25. Pruss, G., R. N. Goldstein, and R. Calendar. 1974. In vitro packaging of satellite phage P4 DNA. Proc. Natl. Acad. Sci. U.S.A. 71:2367-2371. 26. Olivera, B. M., K. G. Lark, R. Herrmann, and F. Bonhoeffer. 1973. Discontinuous DNA synthesis in vitro: a method for defining the role of factors in replication, p. 215-231. In R. D. Wells and R. B. Inman (ed.), DNA synthesis in vitro. University Park Press, Baltimore. 27. Sakai, H., S. Hashimoto, and T. Komano. 1974. Replication of deoxyribonucleic acid in Escherichia coli C mutants temperature sensitive in the initiation of chromosome replication. J. Bacteriol. 119:811-820. 28. Sasaki, I., and G. Bertani. 1965. Growth abnormalities in Hfr derivatives of Escherichia coli strain C. J. Gen. Microbiol. 40:365-376. 29. Schekman, R., A. Weiner, and A. Kornberg. 1974. Multienzyme systems of DNA replication. Science 186:987-993. 30. Schnos, M., and R. B. Inman. 1971. Starting point and direction of replication in P2 DNA. J. Mol. Biol. 55:31-38. 31. Six, E. W., and C. A. C. Klug. 1973. Bacteriophage P4: a satellite virus depending on a helper such as prophage P2. Virology 51:327-344. 32. Sunshine, M., D. Usher, and R. Calendar. 1975. Interaction of P2 bacteriophage with the dnaB gene of Escherichia coli. J. Virol. 16:284-289. 33. Wada, C., and T. Yura. 1974. Phenethyl alcohol resistance in Escherichia coli. HI. A temperature-sensitive mutation (dnaP) affecting DNA replication. Genetics 77:199-220. 34. Wall, J. D., and P. D. Harriman. 1974. Phage P1 mutants with altered transducing abilities for Escherichia coli. Virology 59:532-544. 35. Wang, J. C., K. V. Martin, and R. Calendar. 1973. On the sequence similarity of the cohesive ends of coliphage P4, P2, and 186 deoxyribonucleic acid. Biochemistry 121:2119-2123. 36. Whitney, E. 1971. The toIC locus in Escherichia coli K12. Genetics 67:39-53. 37. Wiman, M., G. Bertani, B. Kelley, and I. Sasaki. 1970. Genetic map of Escherichia coli strain C. Mol. Gen. Genet. 107:1-31. 38. Wolf, B. 1972. The characteristics and genetic map location of a temperature sensitive DNA mutant of E. coli K12. Genetics 72:569-593.