Vol. 137, No. 1
JOURNAL OF BACTERIOLOGY, Jan. 1979, p. 44-50 0021-9193/79/01-0044/07$02.00/0
Methylation-Dependent DNA Synthesis in Escherichia coli Mediated by DNA Polymerase I CYNTHIA LARK Department of Biology, University of Utah, Salt Lake City, Utah 84112
Received for publication 18 July 1978
An in vitro system was used to study DNA synthesis in lysates of Escherichia coli cells which had been grown in the presence of ethionine. Such lysates showed a reduced capacity to incorporate [3H]TTP into high-molecular-weight material. Activity could be restored by incubation with S-adenosyl methionine and ATP. S-adenosyl methionine-reactivated TTP incorporation required the presence of DNA polymerase I, ATP, and all four deoxyribonucleotide triphosphates. DNA polymerase Im was not required.
Methionine auxotrophs of Escherichia coli cease replicating DNA after one round of replication if grown in medium in which ethionine or norleucine are substituted for required methionine. Initiation of chromosome replication occurs in the presence of these analogs, and cells continue to make DNA until a region of the template is reached which was synthesized in the presence of the methionine analog (6). To explain these results, it was proposed that a methylated template was necessary for DNA
thesis can be restored by incubation of such lysates with S-adenosyl methionine (SAM) and ATP. However, the DNA synthesis which occurs after this treatment appears to be mediated by DNA polymerase I. MATERIALS AND METHODS Strains. H560 F- thyA arg poLAI endA tsx rpsL was obtained from F. Bonhoeffer. CR34 F- thyA thr leu thi lac tonA rpsL and E486 F- thyA thr leu met thi lac tonA rpsL dnaE486 (11) were obtained from J. Wechsler. BT4109 F- thyA poLA4109 was obtained from B Olivera. This strain is temperature sensitive for DNA polymerase I polymerizing activity but has normal 5' 3' exonuclease activity (8). WA3062 F- thyA arg t7p met, a derivative of E. coli 15T- (555-7) cured of P15 bacteriophage, was obtained from W Arber. A poL41 derivative of a methionine auxotroph of H560 was constructed, using P1 bacteriophage transduction and selecting for methyl methane sulfonate resistance. Methionine auxotrophs of H560, CR34, and BT4109 were isolated by using nitrosoguanidine mutagenesis followed by penicillin selection. The E486 strain is a spontaneous met mutant isolated by J. Wechsler. Growth of celis. M9 minimal medium (7) was used in all experiments. When required, nutrients were added to the following final concentrations (in micrograms per milliliter): thymine, 4; arginine, 100; methionine, 40; threonine, 40; leucine, 50; and thiamine, 0.1. Cells were counted in a Coulter counter, using a 30pam orifice. Reagents. Reagents for in vitro synthesis of DNA were cellophane, obtained from Kalle Einmach Cellophan, Kalle Alctiengesellschaft, Wiesbaden, Biebrich,
replication (5). Ethionine can be incorporated into proteins in place of methionine (10). It is possible, therefore, that proteins required for DNA synthesis might be unable to function when containing this amino acid analog. In this paper, it is shown that protein synthesis is needed to restore DNA synthesis in vivo after growth in the presence of ethionine. If ethionine is to be used to study the effects of preventing DNA methylation, it is essential to have a system in which these can be distinguished from those resulting from aberrant protein synthesis. This has been accomplished by using the in vitro method of Schaller et al. (9), in which protein synthesis cannot occur. In this system, cells are gently lysed on cellophane disks, and the lysate is incubated over droplets containing various reaction mixtures. This method does not subject the large-template DNA molecule to physical stress, which could result in breakage, and it appears to allow DNA synthesis to proceed in a manner characteristic of in vivo bacterial chromosome chain elongation. By using this system, it has been possible to show that lysates prepared from cells grown in the presence of ethionine show a marked decrease in ability to support DNA synthesis. Syn-
Germany; tris(hydroxymethyl)aminomethane (Tris), obtained from Sigma Chemical Co.; ATP, obtained from Sigma; dTTP, dCTP, dGTP, dATP, obtained from Schwarz/Mann; [3H]TTP, obtained from Amersham/Searle; and S-adenosyl methionine iodide, ob-
tained from Calbiochem. DNA polymerase I and anti-DNA polymerase I 44
VOL. 137, 1979
METHYLATION-DEPENDENT DNA SYNTHESIS
antiserum were gifts of A. Kornberg. DNA polymerase III was a gift of M. Gefter. Incorporation of radioactive isotope. Cultures growing in the presence of radioactive thymine were assayed by collecting samples (0.1 ml each) on Whatman 3 MM filter paper disks. After drying, the disks were treated with cold 10% trichloroacetic acid, washed with 5% trichloroacetic acid and ethanol, and dried. They were placed in organic scintillation fluid and counted in a Beckman scintillation counter. In vitro synthesis of DNA. Cells were grown in a rotary shaker bath to a titer of about 108 cells/ml at either 370C for H560 or CR34, or 300C for E486. [I4C]thymine (0.03 ,uCi/4 ,ug per ml) was added to prelabel the cells to calculate the cell concentration on each cellophane disk. When cells grown in ethionine medium were needed, the bacteria were collected on B-5 membrane filters (Schleicher & Schuell Co.), washed, and suspended in medium containing this analog. Such cells were grown in the presence of ethionine until DNA synthesis had ceased before harvesting. Lysates were prepared on cellophane disks on buffer agar by the method of Schaller et al. (9). A few minor changes were made. 0.5% Brij 58 was added to the lysozyme solution instead of to the bacterial suspension, and 0.2 M Tris buffer (pH 7.8) was substituted for morpholinopropane sulfonic acid (MOPS) in the agar plates. About 5 x 107 cells were spread on a cellophane disk (diameter, 16 mm). After lysis and drying, these disks were incubated on 50-ji drops of radioactive incorporation mixture at the desired temperature. The composition of the buffer for the incorporation mixture (TMK) was as follows: Tris buffer (pH 7.8), 20 mM; MgCl2, 5 mM, ethylenediaminetetraacetic acid, 0.1 mM; and KCl, 0.1 mM. When measuring DNA synthesis, this contained 1 mM ATP, 170 pmol of thymidine and 20 ,umol each of dATP, dCTP, dGTP and [3H]TTP (1 MuCi/ymol). All of the experiments are presented as counts per minute per 5 x 107 cells. This was calculated by prelabeling cultures with ['4C]thymine, during growth in methionine and ethionine, correlating the 14C counts with the cell number and correcting all of the counts to 5 x 107 cells. (The number of cells on a disk ranged from 3 x 107 to about 8 x 107. Usually, the density was about 5 x 107 disk.) The disks were boiled with 1 ml of sodium dodecyl sulfate (1%) and NaOH (0.01 M) and, after boiling, trichloroacetic acid was added. The precipitates were collected on B-2 membrane filters (Schleicher & Schuell), and the radioactivity was counted in a Beckman scintillation counter. For treatment of ethionine-grown cells with SAM, ATP (1 mM) and SAM (3.5 mM) were added to TMK buffer. The concentration of SAM was chosen as that capable of the greatest reactivation of DNA synthesis. This concentration also is that required to methylate DNA on cellophane disks (2). I have found that SAM cannot dialyze through all types of cellophane. The cellophane used in this study permits methylation. Lysates on cellophane disks were pretreated with SAM by incubation for 5 min on a drop (0.05 ml) of this SAM reactivation mixture. The disk was then
45
placed on agar at 4°C, and the SAM and ATP were removed by dialysis. Five minutes of such dialysis were sufficient to remove the ATP required for DNA synthesis or the SAM required to methylate DNA. Gomez-Eichelmann and Lark (2) have shown that DNA synthesized in vitro after removal of SAM in this manner can no longer function as a substrate for a deoxyribonuclease (DNase) of Diplococcus pneumoniae, described by Lacks and Greenberg (4), which is specific for methylated DNA. DNA polymerase I (0.2 U/disk) and DNA polymerase III (0.01 U/disk) were spread onto the cellophane disks as a mixture with the bacteria. Anti-DNA polymerase I antiserum (1:10 dilution of heated gamma-globulin) was either spread onto the cellophane disks before the bacteria-DNA polymerase I mixture was added, or it was added later, when the lysates were placed on agar plates to remove the SAM and ATP by dialysis.
RESULTS Synthesis of DNA after growth in ethionine-the inhibitory effect of chloramphenicol. When methionine auxotrophs are grown in ethionine medium, the DNA content of the culture doubles, after which DNA synthesis ceases. Addition of methionine reverses this inhibition. A culture of WA3062, similar to WA2365 used in previous experiments (6) but cured of P15 bacteriophage, was grown in medium containing [14C]thymine for three to four generations. The cells were transferred to ethionine medium containing ["4C]thymine ofthe same specific activity and were incubated until DNA synthesis ceased. Methionine and [3H]thymine were then added, the ["C]thymine remaining present. Incorporation of ["C]thymine (Fig. 1A) and ['H]thymine (Fig. 1B) was measured. A net increase in DNA was observed after a short lag. If chloramphenicol was added to inhibit protein synthesis, little increase in total DNA could be detected as incorporation of ["4C]thymine (Fig. 1A). Similarly, chloramphenicol markedly inhibits [3H]thymine
incorporation (Fig. 1B). These data indicate that a low level of DNA synthesis occurs in ethionine-treated cells which depends upon the presence of methionine but which does not require new protein synthesis. This finding prompted experiments to determine whether DNA synthesis requiring a methylated template can be demonstrated in vitro by using lysates of ethionine-grown cells. In vitro synthesis of DNA after incubation with ethionine. Lysates of E. coli methionine auxtrophs which had been grown in ethionine medium until DNA synthesis ceased were used for in vitro experiments. Figure 2 shows results obtained with lysates of CR34, using the cellophane disk system described by Schaller et
46
LARK
J. BACTERIOL.
which had been pretreated with SAM. If either dCTP or dGTP was omitted from the reaction mixture, incorporation was inhibited, despite 14 IL 0 SAM pretreatment and the presence of added DNA polymerase I. The [3H]TTP incorporation which follows treatment of lysates of ethionine-grown cells 5 with SAM is not dependent on DNA polymerase s o0 to ao x c 05 III. The same. amount of incorporation is obculture was transferred to [ medium, in 04C'thymi served in the presence or absence of N-ethyl FI.1. Inoprto frdoatv0hmn~~~~~2 fe tur of E.cl0At agow nmnmlmdu maleiamide (results not shown), and lysates of a strain lacking DNA polymerase III (dnaE486) 25 75 25 750 incorporate TTP after SAM pretreatment. FigMINUTES ure 4 shows the incorporation of [3H]TTP by FIG. 1. Incorporation of radioactive thymine after lysates of E486 (cf. Fig. 2). Lysates of this strain ethionine treatment: effect of chloramphenicolA cul- lack DNA polymerase III activity even at low syntesi esd was tti /a iemtinn t 20 WA3062 in minimal medium ture of E. coli grown to a density of 1.5 x 108 cellsmi. [SClthymine (1 temperatures, e.g., 250C (1); however; DNA poeaci4 pg per ml) was present during growth. The lymerase I activity is present in normal amounts. culture was transferred to[4Clthymine medium, in Addition of purified DNA polymerase III rewhich ethionine (20 pg/mi) was substituted for methi- stores some activity to lysates of methionineonine and then incubated for 1OtJ min until DNA grown cells. However, pretreatment of lysates of synthesis had ceased. At this time methionine (to 20 ethionine-grown E486 with SAM restores incorpg/in) and[BH]thnmine (to 10m tCiml) were added. poration without added DNA polymerase III, The culture was divided, and chloramphenicol was and addition of DNA polymerase III does not added to one portion (to 150 pg/mi). Samples (0.1 ml enhance the restored activity. each) were taken at intervals, and radioactivities ATP requirement and inactivation of were measured. (A) [hC]thymine; (B) [3H]thymine. Spymbols: 0, chioramphenicol added;, no chloram- DNA polymerase I activity by anti-DNA phenicol present. Note the differentt scales of the or- polymerase I antibody. The activity restored dinate for B. The samples from the culture without by SAM pretreatment requires ATP (Fig. 5). chioramphenicol are graphed on a scale which has After SAM pretreatment, SAM and ATP were been reduced threefold. A.
B.
0
al. (9). The in vitro incorporation of [3H]TTP into acid-precipitable material is reduced in lysates of cells which have been grown in the presence of ethionine when compared to lysates of methionine-grown cells. This is true for all strains of E. coli tested, including many not discussed in this paper. Incubation with SAM and ATP can restore DNA synthesis in lysates of cells grown in the presence of ethionine when these cells are poiAl (Fig. 2). The restored activity equals that observed with lysates of cells grown in methionine medium. Preincubation with SAM does not change (increase) the in vitro activity of lysates prepared from methionine-grown cells (results not shown, but see Fig. 3). The incorporated radioactivity is DNase digestible and can be banded in CsCl at a buoyant density of about 1.710 g/cm', and incorporation requires the presence of template DNA. Depenmdence of SAM-reactivated synthesis on DNA polymerase I but not DNA polymeragse MI. Attempts to use SAM to reactivate lysates of ethionine-grown poiA cells failed (Fig. 3). Addition of purified DNA polymerase I restored activity but only to thos'e lysates
20
40
MINUTES FIG. 2. In vitro DNA synthesis by lysates ofE. coli CR34 grown in ethionine. A culture of E. coli CR34 (10( cells/ml) was transferred to medium containing ethionine instead of methionine. After 160 min when DNA synthesis had ceased, lysates were prepared on cellophane disks and incubated on drops of reaction mix at 370C for various lengths of tine. One series (0) was preincubated for 5 min on TMK buffer plus ATP. Another (0) was preincubated for 5 min on TMK buffer plus ATP and SAM. DNA synthesis was also measured in a lysate that was prepared from a culture growing exponentially in medium plus methionine (0). This lysate was not pretreated with SAM.
METHYLATION-DEPENDENT DNA SYNTHESIS
VOL. 137, 1979
W3 -a
go
Sa
CD
47
-a -a U"
-a
x
Sn
NN
N%
a.
a10
20
1o
MINUTES
20
MINUTES
FIG. 4. In vitro DNA synthesis by lysates ofE. coli FIG. 3. In vitro DNA synthesis by lysates of E. coli E486 grown in ethionine medium. Three in vitro H560 grown in ethionine medium. Cultures of H560 synthesis experiments at 25°C (A) and 40°C (B and were transferred to ethionine medium and incubated C) are shown. Symbols: 0, culture grown in ethionine for 150 min until in vivo DNA synthesis had stopped. medium at 300 for 180 min-synthesis took place Lysates were prepared, and incorporation of I without preincubation on SAM; 0, the same lysate [3H]TTP was measured under different conditions was preincubated for 5 min on SAM; V, DNA polymat 37°C (A) or 25°C (B). Symbols: A, nopreincubation erase III was added to the lysates of the ethionineon SAM-this reaction is only shown for the 25°C grown cells but the lysates were not preincubated on experiment because the results for the 370C experi- SAM; V, DNA polymerase III was added to the lysate ment were identical; 0, 5 min ofpreincubation with of the ethionine-grown cells and the lysates were SAM; , no preincubation on SAM but DNA polym- preincubated for 5 min on SAM; 1, lysates of methierase I added to each disk; V, 5 min ofpreincubation onine-grown cells were used for incorporation; U, with SAM and DNA polymerase I added to each DNA polymerase III was added to lysates of methidisk; A, preincubation with SAM and DNA polym- onine-grown cells, and incorporation was measured. erase I added to each disk, but dCTP omitted from the reaction mixture; V, preincubation with SAM, and DNA polymerase I added to each disk, but dGTP omitted from the reaction mixture. Lysates also were prepared from cells growing exponentially in methi1200 onine medium. These samples also were preincubated with SAM for 5 min, and then incorporation was measured on disks with (U) or without (0) added DNA polymerase I.
removed by dialysis from the lysate (the ceilophane disks were transferred to agar plates and allowed to dialyze at 4°C). The activity of the lysates depended on the presence of ATP in the reaction mixture. In other experiments (not shown), a similar dependence could be demonstrated for poU lysates to which DNA polymerase I had been added. An attempt was made to demonstrate a requirement for DNA polymerase I activity in lysates of poLAU cells by adding anti-DNA polymerase I antibody. However, this did not inhibit the N-ethyl maleiamide-resistant activity which was observed after SAM pretreatment. Since the results with poU mutants indicated a requirement for DNA polymerase I, it appeared possible that the antibody had not reacted with the DNA polymerase I molecules because the enzyme was unavailable. This was tested by using antibody and purified DNA polymerase I (Fig. 6). When DNA polymerase I was added before the addition of antibody, the anti-DNA polymerase I serum had little effect. However, addition of antibody before adding enzyme in-
w
800
h0
_/ 400
10
20
MINUTES FIG. 5. Dependence of SAM-activated in vitro synthesis on ATP. A lysate was prepared from a culture of E. coli E486 after growth at 30°C in ethionine medium for 180 min. A, No preincubation on SAM. Incorporation measured on normal reaction mix (1 mMATP added). Other lysates were preincubated on 'SAM and ATP and then dialyzed to remove the ATP by placing the disks on chilled (4°C) agar plates (containing Tris magnesium buffer). One set (U) was incubated on reaction mixture containing the normal amount of ATP. Another (0) was incubated on a reaction mixture from which ATP had been omitted.
48
J. BACTERIOL.
LARK
B
A
CO) -A -a
CA -a -a
Cs. "Ca
1000
500
in.-
2
N..
SO
'I
C.)
an
10
20
10
20
MINUTES FIG. 6. The effect of anti-DNA polymerase I antiserum on SAM-activated DNA synthesis mediated by DNA polymerase I. In vitro incorporation at 37°C (A) and 25°C (B) utilized lysates of E. coli H560 previously grown in ethionine medium. AU lysates were preincubated on SAM for 5 min. Symbols: U, DNA polymerase I was not added to the lysate; *, DNA polymerase I was added to the lysate after preincubation with SAM; 0, anti-DNA polymerase I antiserum and DNA polymerase I were mixed with the lysate when the lysate was spread on the cellophane disk; O, DNA polymerase I was added when the lysate was spread on the cellophane disk. AntiDNA polymerase I antiserum was added after 5 min ofpreincubation on SAM.
hibited the reaction completely. It appears possible, therefore, that DNA polymerase I is active in a form in which it is unavailable to added antibody. Requirement for DNA polymerase I in vivo. The in vitro incorporation of TTP occurs in the absence of protein synthesis. In the experiment in Fig. 1, incorporation of [3H]thymine was observed after restoring methionine in the presence of chloramphenicol. This incorporation does not result in a net increase in DNA, and it is dependent on DNA polymerase I (Fig. 7). BT4109, a strain in which the polymerase activity of DNA polymerase I is temperature sensitive, was used. The culture was labeled with [14C]thymine and incubated at 300C in ethionine medium until DNA synthesis ceased. Methionine and [3H]thymine then were added as well as chloramphenicol and portions of the culture incubated at 30 and 420C. At 300C, the results were the same as those seen previously (Fig. 1). However, incorporation of [3H]thymine was inhibited at 420C by chloramphenicol. No such inhibition occurred at 420C in the wild-type strain (data not shown). Thus, the in vivo incorporation which can occur in the presence of chloramphenicol depends on DNA polymerase I. In the absence of chloramphenicol (Fig. 7), a net increase in DNA occurs at either 30 or 420C after a lag of almost 100 min. During this lag, incorporation of [3H]thymine occurs which appears to be somewhat temperature sensitive.
DISCUSSION When methionine auxotrophs of E. coli are grown in ethionine medium, DNA replication ceases after the DNA content of the cells has doubled. Lysates of such cells incorporate TTP poorly. The ability to incorporate [3H]TTP can be restored to lysates of ethionine-grown cells by preincubation with SAM if DNA polymerase I is present. DNA polymerase III does not appear to be required for this SAM-dependent DNA synthesis since lysates of a dnaE(Ts) mutant can incorporate [3H]TTP under these conditions, and the incorporation is resistant to Nethyl maleimide. In common with other in vitro systems for DNA synthesis in which undamaged bacterial or bacteriophage DNA serves as a template, DNA synthesis after SAM reactivation is dependent on the presence of ATP. Thus, this DNA polymerase I-mediated synthesis is different from that observed in purified systems using nicked DNA as a template which do not require ATP. The physical properties of the DNA synthesized by DNA polymerase I after SAM treatment are being investigated. Preliminary results show that this DNA is synthesized discontinuously as small polynucleotide chains (4 to 5S), which can be separated from larger DNA by denaturation and which can be chased into larger material. Therefore, this synthesis does not appear to be due to repair of gaps or nicks which might occur in large-template DNA. The role played by SAM in this in vitro system is not understood. It seems likely that the observed incorporation of TTP depends on methylation of DNA, although other effects of SAM cannot be ruled out. Several experiments show that DNA is methylated by the in vitro method used here. By using density label and [3H]methyl-labeled SAM, it is possible to show that SAM will methylate undermethylated DNA in lysates on cellophane disks (results not shown). More recently, Gomez-Eichelmann and Lark (2) have shown that unmethylated DNA synthesized in vitro can be methylated by SAM under the same conditions used here. Once methylated, this DNA can be degraded by the pneumococcal DNase, described by Lacks and Greenberg (4), specific for methylated DNA. They also showed that DNA made after the removal of SAM by dialysis is not a substrate for this enzyme and is, therefore, presumably not methylated. However, SAM can methylate other macromolecules and could be reactivating some other essential component of the system. Since preincubation with SAM is sufficient to activate the
METHYLATION-DEPENDENT DNA SYNTHESIS
VOL. 137, 1979
49
.,000 ..ooo 2,000
4,000
1,500
1,000 2,000
a IL
S00
U
U A.
U 00
300
MINUTES FIG. 7. Incorporation of radioactive thymine after ethionine treatment: effect of a polA(ts) mutation. A culture of E. coli BT4109 was grown at 30°C in ['4C]thymine (1 ,uCi/4 pg per ml) medium for three to four generations and then shifted to ethionine medium for 200 min (['4C]thymine was present). At this time, methionine and [3H]thymine were added, and the culture was divided into four portions. These were incubated at 30'C (0) or 42°C (0) with or without chloramphenicol (CM). [3HJthymine and [14CJthymine incorporation are shown in the upper and lower panels, respectively.
incorporation described here, it does not seem likely that SAM is required as a cofactor in the incorporation reaction, although a second requirement for low concentrations of SAM would be undetected. The experiments in Fig. 3 show that DNA polymerase I must be present in lysates of ethionine-grown cells for DNA synthesis to occur after treatment with SAM. This does not appear to be true in vivo, since BT4109, grown in ethionine medium, can resume replication at both 30 and 42°C after the addition of methionine (Fig. 7). Similarly, both H560 and its poUA+ derivative recover from growth in the presence of ethionine after addition of methionine to the medium (results not shown). This may be due to the presence of low levels of DNA polymerase I in polA cells sufficient for in vivo but not in vitro recovery. In support of this, it should be noted that recovery is more rapid when DNA polymerase I is active. Alternatively, some other component of replication may be able to substitute for DNA polymerase I in vivo but not in
ThepoUA-dependent turnover of DNA in the presence of chloramphenicol observed after addition of methionine to ethionine-grown cells most probably corresponds to the SAM-reactivated synthesis observed in vitro. DNA polymerase I is generally assumed to function in repair synthesis. This enzyme may be required for the repair of damage incurred either during growth in the presence of ethionine or as a result of methylation which is not synchronous with replication. It is possible that certain undermethylated regions of DNA are either not methylated or are methylated in such a way as to appear to be errors. These might then be recognized, excised, and replaced by DNA polymerase I. Such a mechanism could be responsible for the mismatch repair of X bacteriophage DNA described by Ryokowski et al. (M. Ryokowski, P. Pukkila, M. Radman, R. Wagner, and M. Meselson, Cold Spring Harbor Symp. Quant. Biol., in press). These authors observed that mismatch repair acts selectively on the replica strand rather than template DNA and depends upon undermethylation of the forvitro. The effect of chloramphenicol upon recovery mer. Altematively, the isolation of conditional lefrom growth in the presence of ethionine indicates that some protein required for replication thal mutants lacking the 5' -. 3' exonucleolytic is unavailable. As yet, it has not been possible to activity of DNA polymerase I may indicate that this enzyme is essential for replication (3, 8). I identify the protein (or proteins) lacking.
50
LARK
am currently investigating the effects which various conditional lethal mutations in genes essential for DNA replication may have on the DNA polymerase I activity described here. ACKNOWLEDGMENTS I thank K. G. Lark and J. Cairns for many helpful suggestions, and Edward Meenen for excellent technical assstance. This work was supported by Public Health Service grant 1R01-GM19703 from the National Institute of General Medical Sciences, and by grant NP 85-D from the American Cancer
Society. LITERATURE CITED 1. Gefter, M. L, Y. Hiroto, T. Kornberg, J. A. Wechsler, and C. Barnoux, 1971. Analysis of DNA polymerase II and m of Escherichia coli thermosensitive for DNA synthesis. Proc. Natl. Acad. Sci. U.S.A. 68:3150-3153. 2. Gomez-Eichelmann, M. C., and K. G. Lark. 1977. Endo R Dpn I Restriction of Escherichia coli DNA synthesized in vitro. Evidence that the ends of Okazaki pieces are determined by template deoxyribonucleic acid sequence. J. Mol. Biol. 117:621-635. 3. Konrad, E. B., and I. R. Lehman. 1974. A conditional lethal mutant of Escherichia coli K12 defective in the 5' --. 3' exonuclease associated with DNA polymerase I. Proc. Natl. Acad. Sci. U.S.A. 71:2048-2051.
J. BACTERIOL. 4. Lacks, S., and B. Greenberg. 1975. A deoxyribonuclease of Diplococcus pneumoniae specific for methylated DNA. J. Mol. Chem. 250:4060-4066. 5. Lark, C. 1968. The effect of methionine analogues ethionine and norleucine on DNA synthesis in Escherichia coli 15T-. J. Mol. Biol. 31:401414. 6. Lark, C., and W. Arber. 1970. Host specificity of DNA produced by Escherichia coli. Xm. Breakdown of cellular DNA upon growth in ethionine of strains with r15', rp,+ or rN3+ restriction phenotypes. J. Mol. Biol. 52: 337-348. 7. Lark, K. G., T. Repko, and E. Hoffman. 1963. The effect of amino acid deprivation on subsequent deoxyribonucleic acid replication. Biochim. Biophys. Acta 76: 9-24. 8. Olivera, B. M., and F. Bonhoeffer. 1974. Replication of Escherichia coli requires DNA polymerase I. Nature
(London) 250:513-514. 9. Schaller, H., 0. Bernd, V. Nusslein, J. Huf, R. Herrmann, and F. Bonhoeffer. 1972. DNA replication in vitro. J. Mol. Biol. 63:183-200. 10. Smith, R. C., and W. D. Salmon. 1965. Effect of ethionine on the ribonucleic acid, deoxyribonucleic acid, and protein content of Escherichia coli. J. Bacteriol. 89: 687-692. 11. Wechsler, J. A., and J. D. Gross. 1971. Escherichia coli mutants temperature-sensitive for DNA synthesis. Mol. Gen. Genet. 113:273-284.
JOURNAL OF BACTERIOLOGY, Jan. 1979, p. 44-50 0021-9193/79/01-0044/07$02.00/0
Methylation-Dependent DNA Synthesis in Escherichia coli Mediated by DNA Polymerase I CYNTHIA LARK Department of Biology, University of Utah, Salt Lake City, Utah 84112
Received for publication 18 July 1978
An in vitro system was used to study DNA synthesis in lysates of Escherichia coli cells which had been grown in the presence of ethionine. Such lysates showed a reduced capacity to incorporate [3H]TTP into high-molecular-weight material. Activity could be restored by incubation with S-adenosyl methionine and ATP. S-adenosyl methionine-reactivated TTP incorporation required the presence of DNA polymerase I, ATP, and all four deoxyribonucleotide triphosphates. DNA polymerase Im was not required.
Methionine auxotrophs of Escherichia coli cease replicating DNA after one round of replication if grown in medium in which ethionine or norleucine are substituted for required methionine. Initiation of chromosome replication occurs in the presence of these analogs, and cells continue to make DNA until a region of the template is reached which was synthesized in the presence of the methionine analog (6). To explain these results, it was proposed that a methylated template was necessary for DNA
thesis can be restored by incubation of such lysates with S-adenosyl methionine (SAM) and ATP. However, the DNA synthesis which occurs after this treatment appears to be mediated by DNA polymerase I. MATERIALS AND METHODS Strains. H560 F- thyA arg poLAI endA tsx rpsL was obtained from F. Bonhoeffer. CR34 F- thyA thr leu thi lac tonA rpsL and E486 F- thyA thr leu met thi lac tonA rpsL dnaE486 (11) were obtained from J. Wechsler. BT4109 F- thyA poLA4109 was obtained from B Olivera. This strain is temperature sensitive for DNA polymerase I polymerizing activity but has normal 5' 3' exonuclease activity (8). WA3062 F- thyA arg t7p met, a derivative of E. coli 15T- (555-7) cured of P15 bacteriophage, was obtained from W Arber. A poL41 derivative of a methionine auxotroph of H560 was constructed, using P1 bacteriophage transduction and selecting for methyl methane sulfonate resistance. Methionine auxotrophs of H560, CR34, and BT4109 were isolated by using nitrosoguanidine mutagenesis followed by penicillin selection. The E486 strain is a spontaneous met mutant isolated by J. Wechsler. Growth of celis. M9 minimal medium (7) was used in all experiments. When required, nutrients were added to the following final concentrations (in micrograms per milliliter): thymine, 4; arginine, 100; methionine, 40; threonine, 40; leucine, 50; and thiamine, 0.1. Cells were counted in a Coulter counter, using a 30pam orifice. Reagents. Reagents for in vitro synthesis of DNA were cellophane, obtained from Kalle Einmach Cellophan, Kalle Alctiengesellschaft, Wiesbaden, Biebrich,
replication (5). Ethionine can be incorporated into proteins in place of methionine (10). It is possible, therefore, that proteins required for DNA synthesis might be unable to function when containing this amino acid analog. In this paper, it is shown that protein synthesis is needed to restore DNA synthesis in vivo after growth in the presence of ethionine. If ethionine is to be used to study the effects of preventing DNA methylation, it is essential to have a system in which these can be distinguished from those resulting from aberrant protein synthesis. This has been accomplished by using the in vitro method of Schaller et al. (9), in which protein synthesis cannot occur. In this system, cells are gently lysed on cellophane disks, and the lysate is incubated over droplets containing various reaction mixtures. This method does not subject the large-template DNA molecule to physical stress, which could result in breakage, and it appears to allow DNA synthesis to proceed in a manner characteristic of in vivo bacterial chromosome chain elongation. By using this system, it has been possible to show that lysates prepared from cells grown in the presence of ethionine show a marked decrease in ability to support DNA synthesis. Syn-
Germany; tris(hydroxymethyl)aminomethane (Tris), obtained from Sigma Chemical Co.; ATP, obtained from Sigma; dTTP, dCTP, dGTP, dATP, obtained from Schwarz/Mann; [3H]TTP, obtained from Amersham/Searle; and S-adenosyl methionine iodide, ob-
tained from Calbiochem. DNA polymerase I and anti-DNA polymerase I 44
VOL. 137, 1979
METHYLATION-DEPENDENT DNA SYNTHESIS
antiserum were gifts of A. Kornberg. DNA polymerase III was a gift of M. Gefter. Incorporation of radioactive isotope. Cultures growing in the presence of radioactive thymine were assayed by collecting samples (0.1 ml each) on Whatman 3 MM filter paper disks. After drying, the disks were treated with cold 10% trichloroacetic acid, washed with 5% trichloroacetic acid and ethanol, and dried. They were placed in organic scintillation fluid and counted in a Beckman scintillation counter. In vitro synthesis of DNA. Cells were grown in a rotary shaker bath to a titer of about 108 cells/ml at either 370C for H560 or CR34, or 300C for E486. [I4C]thymine (0.03 ,uCi/4 ,ug per ml) was added to prelabel the cells to calculate the cell concentration on each cellophane disk. When cells grown in ethionine medium were needed, the bacteria were collected on B-5 membrane filters (Schleicher & Schuell Co.), washed, and suspended in medium containing this analog. Such cells were grown in the presence of ethionine until DNA synthesis had ceased before harvesting. Lysates were prepared on cellophane disks on buffer agar by the method of Schaller et al. (9). A few minor changes were made. 0.5% Brij 58 was added to the lysozyme solution instead of to the bacterial suspension, and 0.2 M Tris buffer (pH 7.8) was substituted for morpholinopropane sulfonic acid (MOPS) in the agar plates. About 5 x 107 cells were spread on a cellophane disk (diameter, 16 mm). After lysis and drying, these disks were incubated on 50-ji drops of radioactive incorporation mixture at the desired temperature. The composition of the buffer for the incorporation mixture (TMK) was as follows: Tris buffer (pH 7.8), 20 mM; MgCl2, 5 mM, ethylenediaminetetraacetic acid, 0.1 mM; and KCl, 0.1 mM. When measuring DNA synthesis, this contained 1 mM ATP, 170 pmol of thymidine and 20 ,umol each of dATP, dCTP, dGTP and [3H]TTP (1 MuCi/ymol). All of the experiments are presented as counts per minute per 5 x 107 cells. This was calculated by prelabeling cultures with ['4C]thymine, during growth in methionine and ethionine, correlating the 14C counts with the cell number and correcting all of the counts to 5 x 107 cells. (The number of cells on a disk ranged from 3 x 107 to about 8 x 107. Usually, the density was about 5 x 107 disk.) The disks were boiled with 1 ml of sodium dodecyl sulfate (1%) and NaOH (0.01 M) and, after boiling, trichloroacetic acid was added. The precipitates were collected on B-2 membrane filters (Schleicher & Schuell), and the radioactivity was counted in a Beckman scintillation counter. For treatment of ethionine-grown cells with SAM, ATP (1 mM) and SAM (3.5 mM) were added to TMK buffer. The concentration of SAM was chosen as that capable of the greatest reactivation of DNA synthesis. This concentration also is that required to methylate DNA on cellophane disks (2). I have found that SAM cannot dialyze through all types of cellophane. The cellophane used in this study permits methylation. Lysates on cellophane disks were pretreated with SAM by incubation for 5 min on a drop (0.05 ml) of this SAM reactivation mixture. The disk was then
45
placed on agar at 4°C, and the SAM and ATP were removed by dialysis. Five minutes of such dialysis were sufficient to remove the ATP required for DNA synthesis or the SAM required to methylate DNA. Gomez-Eichelmann and Lark (2) have shown that DNA synthesized in vitro after removal of SAM in this manner can no longer function as a substrate for a deoxyribonuclease (DNase) of Diplococcus pneumoniae, described by Lacks and Greenberg (4), which is specific for methylated DNA. DNA polymerase I (0.2 U/disk) and DNA polymerase III (0.01 U/disk) were spread onto the cellophane disks as a mixture with the bacteria. Anti-DNA polymerase I antiserum (1:10 dilution of heated gamma-globulin) was either spread onto the cellophane disks before the bacteria-DNA polymerase I mixture was added, or it was added later, when the lysates were placed on agar plates to remove the SAM and ATP by dialysis.
RESULTS Synthesis of DNA after growth in ethionine-the inhibitory effect of chloramphenicol. When methionine auxotrophs are grown in ethionine medium, the DNA content of the culture doubles, after which DNA synthesis ceases. Addition of methionine reverses this inhibition. A culture of WA3062, similar to WA2365 used in previous experiments (6) but cured of P15 bacteriophage, was grown in medium containing [14C]thymine for three to four generations. The cells were transferred to ethionine medium containing ["4C]thymine ofthe same specific activity and were incubated until DNA synthesis ceased. Methionine and [3H]thymine were then added, the ["C]thymine remaining present. Incorporation of ["C]thymine (Fig. 1A) and ['H]thymine (Fig. 1B) was measured. A net increase in DNA was observed after a short lag. If chloramphenicol was added to inhibit protein synthesis, little increase in total DNA could be detected as incorporation of ["4C]thymine (Fig. 1A). Similarly, chloramphenicol markedly inhibits [3H]thymine
incorporation (Fig. 1B). These data indicate that a low level of DNA synthesis occurs in ethionine-treated cells which depends upon the presence of methionine but which does not require new protein synthesis. This finding prompted experiments to determine whether DNA synthesis requiring a methylated template can be demonstrated in vitro by using lysates of ethionine-grown cells. In vitro synthesis of DNA after incubation with ethionine. Lysates of E. coli methionine auxtrophs which had been grown in ethionine medium until DNA synthesis ceased were used for in vitro experiments. Figure 2 shows results obtained with lysates of CR34, using the cellophane disk system described by Schaller et
46
LARK
J. BACTERIOL.
which had been pretreated with SAM. If either dCTP or dGTP was omitted from the reaction mixture, incorporation was inhibited, despite 14 IL 0 SAM pretreatment and the presence of added DNA polymerase I. The [3H]TTP incorporation which follows treatment of lysates of ethionine-grown cells 5 with SAM is not dependent on DNA polymerase s o0 to ao x c 05 III. The same. amount of incorporation is obculture was transferred to [ medium, in 04C'thymi served in the presence or absence of N-ethyl FI.1. Inoprto frdoatv0hmn~~~~~2 fe tur of E.cl0At agow nmnmlmdu maleiamide (results not shown), and lysates of a strain lacking DNA polymerase III (dnaE486) 25 75 25 750 incorporate TTP after SAM pretreatment. FigMINUTES ure 4 shows the incorporation of [3H]TTP by FIG. 1. Incorporation of radioactive thymine after lysates of E486 (cf. Fig. 2). Lysates of this strain ethionine treatment: effect of chloramphenicolA cul- lack DNA polymerase III activity even at low syntesi esd was tti /a iemtinn t 20 WA3062 in minimal medium ture of E. coli grown to a density of 1.5 x 108 cellsmi. [SClthymine (1 temperatures, e.g., 250C (1); however; DNA poeaci4 pg per ml) was present during growth. The lymerase I activity is present in normal amounts. culture was transferred to[4Clthymine medium, in Addition of purified DNA polymerase III rewhich ethionine (20 pg/mi) was substituted for methi- stores some activity to lysates of methionineonine and then incubated for 1OtJ min until DNA grown cells. However, pretreatment of lysates of synthesis had ceased. At this time methionine (to 20 ethionine-grown E486 with SAM restores incorpg/in) and[BH]thnmine (to 10m tCiml) were added. poration without added DNA polymerase III, The culture was divided, and chloramphenicol was and addition of DNA polymerase III does not added to one portion (to 150 pg/mi). Samples (0.1 ml enhance the restored activity. each) were taken at intervals, and radioactivities ATP requirement and inactivation of were measured. (A) [hC]thymine; (B) [3H]thymine. Spymbols: 0, chioramphenicol added;, no chloram- DNA polymerase I activity by anti-DNA phenicol present. Note the differentt scales of the or- polymerase I antibody. The activity restored dinate for B. The samples from the culture without by SAM pretreatment requires ATP (Fig. 5). chioramphenicol are graphed on a scale which has After SAM pretreatment, SAM and ATP were been reduced threefold. A.
B.
0
al. (9). The in vitro incorporation of [3H]TTP into acid-precipitable material is reduced in lysates of cells which have been grown in the presence of ethionine when compared to lysates of methionine-grown cells. This is true for all strains of E. coli tested, including many not discussed in this paper. Incubation with SAM and ATP can restore DNA synthesis in lysates of cells grown in the presence of ethionine when these cells are poiAl (Fig. 2). The restored activity equals that observed with lysates of cells grown in methionine medium. Preincubation with SAM does not change (increase) the in vitro activity of lysates prepared from methionine-grown cells (results not shown, but see Fig. 3). The incorporated radioactivity is DNase digestible and can be banded in CsCl at a buoyant density of about 1.710 g/cm', and incorporation requires the presence of template DNA. Depenmdence of SAM-reactivated synthesis on DNA polymerase I but not DNA polymeragse MI. Attempts to use SAM to reactivate lysates of ethionine-grown poiA cells failed (Fig. 3). Addition of purified DNA polymerase I restored activity but only to thos'e lysates
20
40
MINUTES FIG. 2. In vitro DNA synthesis by lysates ofE. coli CR34 grown in ethionine. A culture of E. coli CR34 (10( cells/ml) was transferred to medium containing ethionine instead of methionine. After 160 min when DNA synthesis had ceased, lysates were prepared on cellophane disks and incubated on drops of reaction mix at 370C for various lengths of tine. One series (0) was preincubated for 5 min on TMK buffer plus ATP. Another (0) was preincubated for 5 min on TMK buffer plus ATP and SAM. DNA synthesis was also measured in a lysate that was prepared from a culture growing exponentially in medium plus methionine (0). This lysate was not pretreated with SAM.
METHYLATION-DEPENDENT DNA SYNTHESIS
VOL. 137, 1979
W3 -a
go
Sa
CD
47
-a -a U"
-a
x
Sn
NN
N%
a.
a10
20
1o
MINUTES
20
MINUTES
FIG. 4. In vitro DNA synthesis by lysates ofE. coli FIG. 3. In vitro DNA synthesis by lysates of E. coli E486 grown in ethionine medium. Three in vitro H560 grown in ethionine medium. Cultures of H560 synthesis experiments at 25°C (A) and 40°C (B and were transferred to ethionine medium and incubated C) are shown. Symbols: 0, culture grown in ethionine for 150 min until in vivo DNA synthesis had stopped. medium at 300 for 180 min-synthesis took place Lysates were prepared, and incorporation of I without preincubation on SAM; 0, the same lysate [3H]TTP was measured under different conditions was preincubated for 5 min on SAM; V, DNA polymat 37°C (A) or 25°C (B). Symbols: A, nopreincubation erase III was added to the lysates of the ethionineon SAM-this reaction is only shown for the 25°C grown cells but the lysates were not preincubated on experiment because the results for the 370C experi- SAM; V, DNA polymerase III was added to the lysate ment were identical; 0, 5 min ofpreincubation with of the ethionine-grown cells and the lysates were SAM; , no preincubation on SAM but DNA polym- preincubated for 5 min on SAM; 1, lysates of methierase I added to each disk; V, 5 min ofpreincubation onine-grown cells were used for incorporation; U, with SAM and DNA polymerase I added to each DNA polymerase III was added to lysates of methidisk; A, preincubation with SAM and DNA polym- onine-grown cells, and incorporation was measured. erase I added to each disk, but dCTP omitted from the reaction mixture; V, preincubation with SAM, and DNA polymerase I added to each disk, but dGTP omitted from the reaction mixture. Lysates also were prepared from cells growing exponentially in methi1200 onine medium. These samples also were preincubated with SAM for 5 min, and then incorporation was measured on disks with (U) or without (0) added DNA polymerase I.
removed by dialysis from the lysate (the ceilophane disks were transferred to agar plates and allowed to dialyze at 4°C). The activity of the lysates depended on the presence of ATP in the reaction mixture. In other experiments (not shown), a similar dependence could be demonstrated for poU lysates to which DNA polymerase I had been added. An attempt was made to demonstrate a requirement for DNA polymerase I activity in lysates of poLAU cells by adding anti-DNA polymerase I antibody. However, this did not inhibit the N-ethyl maleiamide-resistant activity which was observed after SAM pretreatment. Since the results with poU mutants indicated a requirement for DNA polymerase I, it appeared possible that the antibody had not reacted with the DNA polymerase I molecules because the enzyme was unavailable. This was tested by using antibody and purified DNA polymerase I (Fig. 6). When DNA polymerase I was added before the addition of antibody, the anti-DNA polymerase I serum had little effect. However, addition of antibody before adding enzyme in-
w
800
h0
_/ 400
10
20
MINUTES FIG. 5. Dependence of SAM-activated in vitro synthesis on ATP. A lysate was prepared from a culture of E. coli E486 after growth at 30°C in ethionine medium for 180 min. A, No preincubation on SAM. Incorporation measured on normal reaction mix (1 mMATP added). Other lysates were preincubated on 'SAM and ATP and then dialyzed to remove the ATP by placing the disks on chilled (4°C) agar plates (containing Tris magnesium buffer). One set (U) was incubated on reaction mixture containing the normal amount of ATP. Another (0) was incubated on a reaction mixture from which ATP had been omitted.
48
J. BACTERIOL.
LARK
B
A
CO) -A -a
CA -a -a
Cs. "Ca
1000
500
in.-
2
N..
SO
'I
C.)
an
10
20
10
20
MINUTES FIG. 6. The effect of anti-DNA polymerase I antiserum on SAM-activated DNA synthesis mediated by DNA polymerase I. In vitro incorporation at 37°C (A) and 25°C (B) utilized lysates of E. coli H560 previously grown in ethionine medium. AU lysates were preincubated on SAM for 5 min. Symbols: U, DNA polymerase I was not added to the lysate; *, DNA polymerase I was added to the lysate after preincubation with SAM; 0, anti-DNA polymerase I antiserum and DNA polymerase I were mixed with the lysate when the lysate was spread on the cellophane disk; O, DNA polymerase I was added when the lysate was spread on the cellophane disk. AntiDNA polymerase I antiserum was added after 5 min ofpreincubation on SAM.
hibited the reaction completely. It appears possible, therefore, that DNA polymerase I is active in a form in which it is unavailable to added antibody. Requirement for DNA polymerase I in vivo. The in vitro incorporation of TTP occurs in the absence of protein synthesis. In the experiment in Fig. 1, incorporation of [3H]thymine was observed after restoring methionine in the presence of chloramphenicol. This incorporation does not result in a net increase in DNA, and it is dependent on DNA polymerase I (Fig. 7). BT4109, a strain in which the polymerase activity of DNA polymerase I is temperature sensitive, was used. The culture was labeled with [14C]thymine and incubated at 300C in ethionine medium until DNA synthesis ceased. Methionine and [3H]thymine then were added as well as chloramphenicol and portions of the culture incubated at 30 and 420C. At 300C, the results were the same as those seen previously (Fig. 1). However, incorporation of [3H]thymine was inhibited at 420C by chloramphenicol. No such inhibition occurred at 420C in the wild-type strain (data not shown). Thus, the in vivo incorporation which can occur in the presence of chloramphenicol depends on DNA polymerase I. In the absence of chloramphenicol (Fig. 7), a net increase in DNA occurs at either 30 or 420C after a lag of almost 100 min. During this lag, incorporation of [3H]thymine occurs which appears to be somewhat temperature sensitive.
DISCUSSION When methionine auxotrophs of E. coli are grown in ethionine medium, DNA replication ceases after the DNA content of the cells has doubled. Lysates of such cells incorporate TTP poorly. The ability to incorporate [3H]TTP can be restored to lysates of ethionine-grown cells by preincubation with SAM if DNA polymerase I is present. DNA polymerase III does not appear to be required for this SAM-dependent DNA synthesis since lysates of a dnaE(Ts) mutant can incorporate [3H]TTP under these conditions, and the incorporation is resistant to Nethyl maleimide. In common with other in vitro systems for DNA synthesis in which undamaged bacterial or bacteriophage DNA serves as a template, DNA synthesis after SAM reactivation is dependent on the presence of ATP. Thus, this DNA polymerase I-mediated synthesis is different from that observed in purified systems using nicked DNA as a template which do not require ATP. The physical properties of the DNA synthesized by DNA polymerase I after SAM treatment are being investigated. Preliminary results show that this DNA is synthesized discontinuously as small polynucleotide chains (4 to 5S), which can be separated from larger DNA by denaturation and which can be chased into larger material. Therefore, this synthesis does not appear to be due to repair of gaps or nicks which might occur in large-template DNA. The role played by SAM in this in vitro system is not understood. It seems likely that the observed incorporation of TTP depends on methylation of DNA, although other effects of SAM cannot be ruled out. Several experiments show that DNA is methylated by the in vitro method used here. By using density label and [3H]methyl-labeled SAM, it is possible to show that SAM will methylate undermethylated DNA in lysates on cellophane disks (results not shown). More recently, Gomez-Eichelmann and Lark (2) have shown that unmethylated DNA synthesized in vitro can be methylated by SAM under the same conditions used here. Once methylated, this DNA can be degraded by the pneumococcal DNase, described by Lacks and Greenberg (4), specific for methylated DNA. They also showed that DNA made after the removal of SAM by dialysis is not a substrate for this enzyme and is, therefore, presumably not methylated. However, SAM can methylate other macromolecules and could be reactivating some other essential component of the system. Since preincubation with SAM is sufficient to activate the
METHYLATION-DEPENDENT DNA SYNTHESIS
VOL. 137, 1979
49
.,000 ..ooo 2,000
4,000
1,500
1,000 2,000
a IL
S00
U
U A.
U 00
300
MINUTES FIG. 7. Incorporation of radioactive thymine after ethionine treatment: effect of a polA(ts) mutation. A culture of E. coli BT4109 was grown at 30°C in ['4C]thymine (1 ,uCi/4 pg per ml) medium for three to four generations and then shifted to ethionine medium for 200 min (['4C]thymine was present). At this time, methionine and [3H]thymine were added, and the culture was divided into four portions. These were incubated at 30'C (0) or 42°C (0) with or without chloramphenicol (CM). [3HJthymine and [14CJthymine incorporation are shown in the upper and lower panels, respectively.
incorporation described here, it does not seem likely that SAM is required as a cofactor in the incorporation reaction, although a second requirement for low concentrations of SAM would be undetected. The experiments in Fig. 3 show that DNA polymerase I must be present in lysates of ethionine-grown cells for DNA synthesis to occur after treatment with SAM. This does not appear to be true in vivo, since BT4109, grown in ethionine medium, can resume replication at both 30 and 42°C after the addition of methionine (Fig. 7). Similarly, both H560 and its poUA+ derivative recover from growth in the presence of ethionine after addition of methionine to the medium (results not shown). This may be due to the presence of low levels of DNA polymerase I in polA cells sufficient for in vivo but not in vitro recovery. In support of this, it should be noted that recovery is more rapid when DNA polymerase I is active. Alternatively, some other component of replication may be able to substitute for DNA polymerase I in vivo but not in
ThepoUA-dependent turnover of DNA in the presence of chloramphenicol observed after addition of methionine to ethionine-grown cells most probably corresponds to the SAM-reactivated synthesis observed in vitro. DNA polymerase I is generally assumed to function in repair synthesis. This enzyme may be required for the repair of damage incurred either during growth in the presence of ethionine or as a result of methylation which is not synchronous with replication. It is possible that certain undermethylated regions of DNA are either not methylated or are methylated in such a way as to appear to be errors. These might then be recognized, excised, and replaced by DNA polymerase I. Such a mechanism could be responsible for the mismatch repair of X bacteriophage DNA described by Ryokowski et al. (M. Ryokowski, P. Pukkila, M. Radman, R. Wagner, and M. Meselson, Cold Spring Harbor Symp. Quant. Biol., in press). These authors observed that mismatch repair acts selectively on the replica strand rather than template DNA and depends upon undermethylation of the forvitro. The effect of chloramphenicol upon recovery mer. Altematively, the isolation of conditional lefrom growth in the presence of ethionine indicates that some protein required for replication thal mutants lacking the 5' -. 3' exonucleolytic is unavailable. As yet, it has not been possible to activity of DNA polymerase I may indicate that this enzyme is essential for replication (3, 8). I identify the protein (or proteins) lacking.
50
LARK
am currently investigating the effects which various conditional lethal mutations in genes essential for DNA replication may have on the DNA polymerase I activity described here. ACKNOWLEDGMENTS I thank K. G. Lark and J. Cairns for many helpful suggestions, and Edward Meenen for excellent technical assstance. This work was supported by Public Health Service grant 1R01-GM19703 from the National Institute of General Medical Sciences, and by grant NP 85-D from the American Cancer
Society. LITERATURE CITED 1. Gefter, M. L, Y. Hiroto, T. Kornberg, J. A. Wechsler, and C. Barnoux, 1971. Analysis of DNA polymerase II and m of Escherichia coli thermosensitive for DNA synthesis. Proc. Natl. Acad. Sci. U.S.A. 68:3150-3153. 2. Gomez-Eichelmann, M. C., and K. G. Lark. 1977. Endo R Dpn I Restriction of Escherichia coli DNA synthesized in vitro. Evidence that the ends of Okazaki pieces are determined by template deoxyribonucleic acid sequence. J. Mol. Biol. 117:621-635. 3. Konrad, E. B., and I. R. Lehman. 1974. A conditional lethal mutant of Escherichia coli K12 defective in the 5' --. 3' exonuclease associated with DNA polymerase I. Proc. Natl. Acad. Sci. U.S.A. 71:2048-2051.
J. BACTERIOL. 4. Lacks, S., and B. Greenberg. 1975. A deoxyribonuclease of Diplococcus pneumoniae specific for methylated DNA. J. Mol. Chem. 250:4060-4066. 5. Lark, C. 1968. The effect of methionine analogues ethionine and norleucine on DNA synthesis in Escherichia coli 15T-. J. Mol. Biol. 31:401414. 6. Lark, C., and W. Arber. 1970. Host specificity of DNA produced by Escherichia coli. Xm. Breakdown of cellular DNA upon growth in ethionine of strains with r15', rp,+ or rN3+ restriction phenotypes. J. Mol. Biol. 52: 337-348. 7. Lark, K. G., T. Repko, and E. Hoffman. 1963. The effect of amino acid deprivation on subsequent deoxyribonucleic acid replication. Biochim. Biophys. Acta 76: 9-24. 8. Olivera, B. M., and F. Bonhoeffer. 1974. Replication of Escherichia coli requires DNA polymerase I. Nature
(London) 250:513-514. 9. Schaller, H., 0. Bernd, V. Nusslein, J. Huf, R. Herrmann, and F. Bonhoeffer. 1972. DNA replication in vitro. J. Mol. Biol. 63:183-200. 10. Smith, R. C., and W. D. Salmon. 1965. Effect of ethionine on the ribonucleic acid, deoxyribonucleic acid, and protein content of Escherichia coli. J. Bacteriol. 89: 687-692. 11. Wechsler, J. A., and J. D. Gross. 1971. Escherichia coli mutants temperature-sensitive for DNA synthesis. Mol. Gen. Genet. 113:273-284.
