A Mutant of Escherichia Coli Defective in DNA Polymerase II Activity

Proc. Nat. Acad. Sci. USA Vol. 69, No. 11, pp. 323&-3242, November 1972

A Mutant of Escherichia coli Defective in DNA Polymerase II Activity (genetic mapping/bacteriophage) YUKINORI HIROTA*, MALCOLM GEFTERt, AND LEONARD MINDICHt * Dbpartement de Biologie Mol6culaire de l'Institut Pasteur, Paris; t Department of Biological Sciences, Columbia University, New York, N.Y. 10027; and t The Public Health Research Institute of the City of New York, New York, 10006 Communicated by Jacques Monod, September 6, 1972

ABSTRACT A mutant of E. coli defective in DNA poly- H102616: F+ thr- leu- endI- polAl- strr polBl- thyA- merase II activity was isolated. Extracts had about 0.1% from strain H10265 by transfer of thr- leu- by a of the normal activity of the DNA polymerase. The effect of (derived the mutation on the ability of cells to replicate DNA of cross with Hfr Hayes carrying thr- and leu- markers, and by various sources was analyzed. Mutant bacteria grow nor- selection for strr and dnaE+ recombinants. mally, at 410 as well as at 300. All bacteriophages, F factors, and R factors examined so far grow normally in the Bacteriophages. Coli-phages, T1, T3, T4, T5, T6? T7, XV3, mutant. Sensitivity of the mutant to ultraviolet radiation Xbio, Xvh, 480, Pivir, P2vir, 4X174, fl, and f2, and temperate and alkylating reagents in growth media was the same as that of the wild type. phages isolated by Jacob and Wollman (10), 18, 80, 82, 186, The mutation, polBI, is recessive with respect to wild 299, 370, 381, 424, 434, 466, and W, were obtained from the type. The polBI mutation can coexist with recombination- bacteriophage collection of D6partment de Biologie Mole- defective mutations. Genetic mapping studies show the culaire de l'Institut Pasteur. Temperate phages (O D series) mutation to be located at about 2 min on the E. coli map. isolated by Bertani, (11) 5, 103, 145, 160, 193, 218, 219, 266, Isolation of an amber mutant of E. coli, polAl (1), which is 326, and 328, were the kind gift of Dr. G. Bertani. defective in the DNA polymerase described by Kornberg (2) Media. L-broth and L-agar (12) supplemented with 50 /Ag (DNA polymerase I), yet grows normally, led to the con- of thymine per ml and 2.5 mM CaCl2 were used to grow clusion that DNA polymerase I might not be a necessary com- bacterial cultures and for assays of phage and bacteria. ponent for replication of E. coli DNA. It has been suggested Minimal synthetic medium 63 (13) supplemented with 0.2% that this enzyme plays a role in DNA repair (1) and in repli- glucose and 5 /Ag of thymine per ml, was used. Synthetic cation of a colicinogenic factor El (3, 4). Availability of this medium with 0.2% glucose and 5 ,ug of thymine per ml was mutant stimulated enzymological investigations for new DNA used. Synthetic medium 63 supplemented with 50 1Ag of polymerases: DNA polymerase II (5-8) and DNA poly- novobiocin per ml, a gift of Laboratories Upjohn (Paris), and merase III (5) were discovered, purified, and characterized. McConkey-medium (Difco Laboratories) were used to score a We have analyzed these DNA polymerase activities in a mutation defective in lipopolysaccharide synthesis (14). series of thermosensitive mutants defective in DNA replication, dnaA through G. Normal levels of activity were found Bacterial Crosses. Bacteria, Hfr and F-, exponentially for DNA polymerase I and II; however, strains having growing at 300 in broth supplemented with thymine were thermosensitive mutations at the dnaE locus were defective in mixed (1 Hfr/10 F-) in the same medium. After 2 hr at 300, DNA polymerase III (9). We concluded, therefore, that DNA the cultures were washed, diluted, and plated on appropriate polymerase III is an enzyme required for DNA replication in media. The plates was incubated at 300. Recombinant colonies E. coli. were isolated from the plates, purified, and tested for their In this paper, we report the isolation and characterization character by replica plating on suitable media. of a mutant of E. coli defective in DNA polymerase II activity. Enzyme Assay. Procedures used for growth of cells, prepa- ration of cell-free extracts, and separation of polymerases II MATERIALS AND METHODS and III by phosphocellulose chromatography have been Bacterial Strains. The following strains were mainly used: described (9). For routine screening of recombinants, a H1026: F+ dnaEts endl polAl- strr thyA- (obtained from modification of the original method was used. The cell-free Dr. F. Bonhoeffer). extract from 2 to 3 g of cells was treated as before through the H10261: F+ dnaEts endI- polAl- strr polBl- thyA- lpc- DEAE-cellulose step. Gradient elution from phosphocellulose (derived from strain H1026 by selection for resistance to was replaced by a step elution. DNA polymerase III was phage 4X174). eluted with 0.12 M potassium phosphate buffer (pH 7.5), and H10265: F+ dnaEts endI- polAl - strr polBl - thyA- lpc+ DNA polymerase II was eluted with 0.25 M potassium phos- (derived from strain H10261 after treatment with N-methyl- phate buffer (pH 7.5). This procedure is not quantitative N'-nitronitrosoguanidine, with selection on McConkey agar but allows easy determination of the presence of polymerase medium (Difco Laboratories). II. 3238 Downloaded by guest on September 24, 2021 Proc. Nat. Acad. Sci. USA 69 (1972) Mutant of E. coli Defective in DNA Polymerase II 3239

RESULTS TABLE 2. Linkage of ipc with thyA and str in the cross Isolation of the mutant of E. coli defective in HfrPlO* x H10261t DNA polymerase II activity On the basis of our previous results (9), we thought that a Recombinants Recombinant types No of DNA polymerase II-negative mutant of E. coli might grow examined str lpc thy recombinants normally, but fail to replicate some other replicons such as thr + leu + thy+ «X174 that are known to some bacterial host func- require (152) 1 1 1 40 (26%) tions. For this purpose, OX174-resistant mustants of E. coli 0 1 1 45 t(30%) Were sought and DNA polymerases of the mutants were 1 0 1 4(2.6%) examined. 0 0 1 63 (41%) Our first isolate of a OX174-resistant mutant, H10261, Frequency of selected without prior mutagenesis on I-agar plates with occurence of pX174, was found to be defective in DNA polymerase II male markers 28% 56% 100% 152 (100%) activity. The results of analysis of DNA polymerase II and III from JG112 (9) (polAl - polB+ dnaE+), H1026, and H10261 Experimental procedures for mating of bacteria and scoring are shown in Fig. la, b, and c, respectively. As can be seen, of markers of recombinants are described in Methods. Alleles of strains H1026 and H10261 each contain 10% of the wild- markers derived from the Hfr parent are represented by 1, and from are type content of DNA polymerase III, as expected, since those derived the recipient parent represented by 0. * HfrPlO: Hfr thr leu polAl + strs lpc +thy+: 111. these strains harbor a dnaEts lesion. Unlike the parent H10261: F+ thr+ leu+ polAl strr thy: 000. which is in its t lpc strain, H1026, normal content of polymerase II, t 8 Recombinants of this class were examined for DNA poly- strain H10261 contains 0.1-0.2% of the wild-type content of merase II activities; all are defective for the enzyme activity. DNA polymerase II. The residual DNA polymerase II activity elutes slightly behind that of the wild type. If the crude extract of strain H10261 is supplemented with DNA polymerase II isolated from strain H1026 in an amount rates at 30 and 450 are compared. These data are summarized equivalent to that normally found in a polB+ strain, the in Table 1. As reported by other workers on multiple-phage enzyme can be recovered in 83% yield after phosphocellulose resistant mutants of E. coli, lpc or tfr (14, 15), H10261 did not chromatography (Fig. id). Thus, there does not appear to be support the growth of bacteriophages T4, T6, T7, X, P1, P2, an inhibitor of DNA polymerase II present in extracts of or OX174, but did support normal growth of T1, T5, and 080. strain H10261. Furthermore, the mutant is hypersensitive to bile salts or novo- The small amount of enzyme activity eluting from the biocine. These characteristics have been postulated to be due phosphocellulose column at the position of DNA polymerase to a defect in the lipopolysaccharide structure of the mutant. II shows a stimulation of the rate of reaction when ammonium This defect results in the loss of phage receptors and a higher sulfate is added to the reaction mixture. Stimulation by salt permeability of the cell to novobiocine (14). The two mu- has only been observed with DNA polymerase II; DNA tations thus found in H10261, polBi and lpc, were therefore polymerase I is unaffected by salt, and DNA polymerase III is analyzed further. inhibited by salt (5). Unlike the DNA polymerase II found PolBI and Ipc of H10261 are separate mutations in polB+ strains, the enzyme isolated from strain H10261 is temperature-sensitive. The reaction rate measured at 450 is From the mutant, H10261, we isolated a series of revertants 60% of that measured at 300. DNA polymerase II isolated that are able to grow normally on McConkey's agar medium, from the parent strain shows a 150% increase in activity when which contains bile salts. All the independent revertants isolated by the selection, 10 out of 10, become sensitive to all the phages; however, the enzyme activity of DNA polymerase TABLE 1. Effect of temperature on the rate of reaction II of the revertants remained defective. The amount of DNA polymerase II as well as its elution position and response to Reaction rate 450/300 salt are indistinguishable in the extracts of the original mutant H10261 and lpc+ revertants. DNA polymerase DNA polymerase The map location of genes controlling the activity of DNA Strain II III polymerase II, polBi, and lipopolysaccharide synthesis, JG112 1.9 1.5 lpc, on the E. coli chromosome were analyzed. For the mapping H1026 1.5 <0.1 of the lpc mutation, a cross [HfrP10 (16) x H10261] was per- H10261 0.6 <0. 1 formed with the selection of thymine-plus recombinants. The results summarized in Table 2 indicates that the mutant gene DNA polymerases II and III were obtained from the peak controlling defective synthesis of lipopolysaccharide of fractions of phosphocellulose chromatography (Fig. 1). Before H10261 is located close to the str and thy genes, the probable assay, DNA polymerase II activity was concentrated 20-fold by. order being str-lpc-thy. Thus, the mutation is located close to, adsorption to and elution from phosphocellulose as described and may be identical with, the gene, lpc, reported by Tamaki (5). Assays were performed by equilibrating the standard reaction mixture (5) to the appropriate temperature and then start- et al. (14). ing the reaction by the addition of enzyme. 5 min later, the reac- Recombinant colonies that received thy+ lpc+ genes from tion was terminated by the addition of cold 5% trichloroacetic the Hfr parent, thy+ lpc+ polBi +, were examined for enzyme acid. DNA pTymeraseII and III activities were measured as activity. All the 8 thy+ lpc+ recombinants examined were de- described (5). fective in DNA polymerase II activity, indicating that lpc Downloaded by guest on September 24, 2021 3240 Genetics: Hirota et al. Proc. Nat. Acad. Sci. USA 69 (1972)

were constructed in H10261 by the selection of thymine prototrophy, about 60% of the thy+ F-ductants became lpc-positive. Two thy+ lpc+ and 2 thy+ lpc- F-ductants were examined for DNA polymerase II activity, and all were found defective. I Additional multiple-phage resistant mutants were isolated 2 20i by the selection of 4X174 or T7 resistance: 16 from H1026, 0 0 8 from HfrlOM7, and 8 from PA3367. All mutants are E 6 C .06 X E ~~~~~~~E very similar to H10261 in the patterns of phage resistance and XL 4 - l~ .04 af hypersensitivity to novobiocin. All contained the normal 2 t 1r1 ~ .02 levels of DNA polymerase II activity. 6 -d 60 From the results listed above, we conclude that the genes

4 40 ipc and polB of H10261 are separate defects, and the mecha-

2 -20 nism that led us to isolate our polB1 - mutant, H10261, is therefore not clear. 25 35 45 55 FRACTION Mapping of polBI and dnaE FIG. 1. Isolation of DNA polymerases II and III. Various Mapping of the polB1 mutation is made difficult by the lack of strains were analyzed for the content of DNA polymerases II an easily recognizable phenotypic character, and recombi- and III. For each analysis, 10 g of cells were used and enzymes nants must be tested directly for the activity of DNA poly- were fractionated by phosphocellulose chromatography as de- merase II. Our approach was to select recombinants that have scribed (9). The elution profiles obtained from strains JG112, received small specific regions of the donor chromosome and H1026, and H10261 are shown in panels a, b, and c, respectively. The earlier eluting peak (fractions 25-35) represents polymerase to test small numbers of these for the presence or absence of III. The observed activity is recorded on the left ordinate. The DNA polymerase II activity. later eluting peak (fractions 45-55) represents polymerase II; Preliminary results indicated that polBi might be located its activity is recorded on the right ordinate. For measurements in the region of dnaE. The gene dnaE, which was shown by us of enzyme activity, column fractions were incubated routinely to code for the structure of DNA polymerase III (9), was for 5 min at 300. In the analysis of polymerase II activity from reported to be cotransducible with tonA (17), but the map strain H10261 (panel c), incubations were done for 50 minat 300; sequence of these genes was not determined. Three-point the data are plotted relative to a 5-min incubation period. Panel mapping of tonA, dnaE, and tfr (15) was made by a cross, d represents the enzyme activities obtained from strain H10261, Hfr 10M7 tfr- x E486, and the results shown in Table 3 which had been supplemented in the crude cell-free extract with demonstrate that the mapping order is thr-leu-tonA-dnaE-tfr. polymerase II (normally obtained from 10 g of cells). Strain H10265, polAl - polB1 - dnaEts thy- strT, was then mated with an Hfr strain, Hfr Hayes 1181, tonA, pro dnaE+, and polB are distinct genes and are not closely linked. Further- and recombinants were selected for streptomycin resistance more, an F' KLF22 or F122, isolated by B. Low, carrying the and thermoresistance. Recombinants of the three expected chromosome segment from argG through thyA, was found to classes with respect to proline requirement and phage resis- have the wild-type allele of lpc. Among 300 F-ductants that tance were assayed for DNA polymerase II activity. The re-

01 N0 TABLE 3. Map location of dnaE

\AAs Recombinant No. of Recombinants types recombinants examined thr-leu tonA dnaE tfr (%) strr thr+ leu+ 1 1 1 1 29 (11.6%) 1 1 1 0 60 (24.0%) 250 1 1 0 1 1 (0.4%) 1 0 1 1 4 (1.6%) 1 1 0 0 37 (14.8%) 1 0 0 1 0 (0%) 1 0 1 0 12 (4.8%) FIG. 2. Linkage map of E. coli K12 that shows the position 1 0 0 0 107 (42.8%) of reference markers used to locate mutations altering DNA Percentage polymerase activities, polAl (DNA polymerase I) (19), polB1 occurrence of (DNA polymerase II), and dnaE (DNA polymerase III) (9). male markers Relative map position of the reference markers and the symbols in recom-

are those listed by Taylor (18). Inner short lines represent F- binants (100) 50.8 42 13.6 250 (100%) primes, F15 (21), F122 (given by Dr. B. Low), and F144 (given by Dr. B. Low). Location and length of those lines indicate the HfrlOM7-1: Hfr stre thr + leu + tonA+ dnaE + tfr-: 1111. chromosome fragments carried on the F-primes. E486: F+ met thy strr thr leu tonA dnaEts tfr+: 0000. Downloaded by guest on September 24, 2021 Proc. Nat. Acad. Sci. USA 69 (1972) Mutant of E. coli Defective in DNA Polymerase II 3241

TABLE 4. Mapping of polBi

Recombinant of II Recombinants types No. of Assay polymerase examined ton dnaE pro recombinants No. examined No. positive strr dnaE+ 1 1 1 42 (58%) 3 3 1 1 0 22 (31%) 4 3 72 0 1 1 5 (7%) 4 0 0 1 0 3 (4%) Total 72 (100%) Frequency of occurrence of male markers 89% 100% 65%o

1181: HfrH tonA dnaE+ proA str8: 111. H 10265: F- polAl polBi tonA + dnaEts pro+ thy strr: 000.

sults shown in Table 4 demonstrate that the mapping order (22) in order to examine the ability of the polBi - strain to is: 0 (transfer origin of HfrH)-polB1-tonA-dnaE. hold R factors in an extra-chromosomal state. Strains con- The polBi mutation is cotransduced by P1 with markers taining R factor were as follows: R11-2(I), RPC3(N), leu and azi: 3 out of 4 leu-transductants, and 4 out of 4 leu- RS-a(W), RPu(P), R22(V), R40a(VI), R11-1(VIII), and- and azi-transductants also received polBi. It is concluded, R79(IX), where the letter or roman numeral in parentheses therefore, that the mapping location of polBl is about 2 min denotes the compatibility group. All 8 R factors, representing and that the probable order is: leu-polBl-tonA. different compatibility groups, were transferred into the An F', KLF44 or F144, isolated by B. Low, carrying the mutant at the same frequency as in the wild type. The chromosome segment from thr through purE was found to polBl- mutant maintained the acquired R factors as stably have the wild-type allele of the polBi. F144 is introduced into as does the wild type. Mutant strains carrying R factors or H102616, polBi- thr- leu-, by the selection of thr+ leu+ F15 are able to transfer these factors again to other strains at F-ductants. Two such F-ductants were examined for DNA similar frequencies to those of wild-type strains. polymerase II activity. Both F-ductants contained wild-type From H10261, a series of derived strains carrying polAl- levels of this enzyme activity. We conclude, therefore, that the polB1- dnaE+ or polA1 + polB1- dnaE+ were constructed by polBl mutation is recessive with respect to wild-type gene crosses with Hfr strains. Sensitivities of polBi - mutant function. strains to UV radiation and to methvlmethane sulfonate In Fig. 2, tentative map positions of genes controlling or to mitomycin C were measured. No difference was found DNA polymerase activities, polAl (19), polBl, and dnaE, are between strains carrying polBi - and polB1i+, as measured by shown on the E. coli chromosome map with the reference counting of surviving colonies. The gene for recA was intro- markers listed by Taylor (18). duced into polA + polB1- dnaE+ strains without difficulty. This indicates. that the recA and polB mutations can coexist DNA synthesis and DNA repair of polBI mutant in the same strain, in contrast to the situation found for The polBi - mutant multiplies at the same rate as the parental polA mutations (23). strain, in minimal and complete medium. Thermoresistant revertants that are dnaE+ but still polBi - grow normally DISCUSSION and form normal colonies at temperatures from 30 to 42°. There are two interesting aspects of the nature of the polBi The parental strain, H1026, and a novobiocin-resistant mutant. The first one concerns the role of DNA polymerase II revertant of H10261, H10265, which recovered adsorption of in DNA replication. Replication of bacterial, bacteriophage, bacteriophages yet remained defective in DNA polymerase II and plasmid DNAs was found to be normal in the mutant activity, were used to test the growth of phages. Both strains defective in DNA polymerase II activity. Wild-type E. coli are equally susceptible to infection with bacteriophages T1, contains, at most, about 90 molecules of DNA polymerase II T3, T5, T6, T7, OX174, X, P1, P2, f1, and f2. A mutant phage, per cell; the residual activity found in extracts of H10261 Xbioll, which is unable to grow on polAl (20), was found to could only represent about 0.1 molecule per cell, a number grow on the polBi mutant. Furthermore, all the temperate that would be hardly sufficient for normal DNA replication. phages isolated by Jacob and Wollman (10), such as 18, 80, We did not prove, however, that the mutant enzyme itself is 82, 186, 299, 370, 381, 424, 434, 466, and It, and by Bertani, active in vivo in spite of its not being detectable in the in vitro 5, 103,145,160, 193, 218, 219, 326 and 328, grow normally on assay. On the other hand, we showed that DNA polymerase both strains. The plaque morphologies of these phages are the III was indispensable for DNA replication in E. coli (9) and same on both strains. for temperate phages X, P1, and P2 (Hirota, unpublished Both F factor and R factors are able to replicate auton- results). It is likely that DNA polymerase III, and not omously on strains H1026, H10261, and H10265. An F- DNA polymerase I or II, is essential in the elongation of prime factor, F15 (21), can be transferred into the polBl- E. coli DNA and temperate-phage DNA. This does not negative mutants and it is maintained stably. A series of R exclude the possibility of an active role for DNA polymerase I factors were given to us by Dr. Chabbert and his associates or II in the replication of other DNA species. Recent work of Downloaded by guest on September 24, 2021 3242 Genetics: Hirota et al.. Proc. Nat. Acad. Sci. USA 69 (1972)

Kingsburg and Helinsky (3) and of Gbebel (4) have shown 3. Kingsburg, D. T. & Helinsky, D. R. (1971) Biochem. Bio- that colEl DNA replication requires DNA polymerase I but phys. Res. Commun. 41,1538-1544. 4. Goebel, W. (1972) Nature New Biol. 237, 67-70. not III. 5. Kornberg, T. & Gefter, M. L. (1971) Proc. Nat. Acad. Sci. The second aspect of the problem concerns the role of USA 68. 761-764. DNA polymerase II in DNA repair and the interactions be- 6. Knippers, R. (1970) Nature 228, 1050-1053. tween recombination-defective mutations and DNA poly- 7. Moses, R. & Richardson, C. C. (1970) Biochem. Biophys. Res. Commun. 41,1565-1571. merase-defective mutations. No difference was found between 8. Wickner, R. B., Ginsberg, B., Beckower, I. & Hurwitz, J. the mutant and the wild type in sensitivity to UV-light, (1972) J. Biol. Chem. 247, 489-497. methylmethane sulfonate, or mitomycin C. 9. Gefter, M. L., Hirota, Y., Kornberg, T., Wechsler, J. A. & Strains carrying potAl and potBi mutations will be very Barnoux, C. (1971) Proc. Nat. Acad. Sci. USA 68, 3150-3153. 10. Jacob, F. & Wollman, E. L. (1958) Ann. Inst. Pasteur 95, useful in the study of DNA synthesis since the major, if not 497-519. the only, DNA polymerase activity remaining is that of 11. Bertani, L. & Bertani, G. (1971) Advan. Genet. 16, 199-237. polymerase III. Assay of cell-free extracts of such mutants 12. Lennox, E. S. (1955) Virology 1, 190-206. should be considerably simplified with these double mutants, 13. Pardee, A. B., Jacob, F. & Monod, J. (1959) J. Mol. Biol. 1, 165-178. enabling studies of the regulation involved in the synthesis 14. Tamaki, S., Sato, T. & Matsuhashi, M. (1971) J. Bacteriol. of DNA polymerase III in E. coli and temperate coliphages. 105, 968-975. A similar E. coli mutant has also been isolated and char- 15. Curtiss, R., III, (1965) J. Bacteriol. 89, 28-40. acterized by others (24), and the results are generally in agree- 16. Jacob, F. & Wollman, E. L. (1961) Sexuality and Genetics of Bacteria (Academic Press Inc., New York). ment with ours. 17. Wechsler, J. A. & Gross, J. D. (1971) Mol. Gen. Genet. 113, We thank Miss C. Barnoux, Mrs. M.-C. Moreau, and Mrs. S. 273-284. Yang for excellent technical assistance. We also thank Dr. F. 18. Taylor, A. L. (1970) Bacteriol. Rev. 34, 155-175. Jacob for his interest and encouragements during the course of 19. Gross, J. D. & Gross M. (1969) Nature 224, 1166-1168. this study and Dr. E. C. Cox for performing P1-transduction 20. Zissler, J. & Shaefer, S. (1971) in "The Bacteriophage experiment. This work was aided by grants from the USPHS Lambda", Cold Spring Harbor Lab., Cold Spring Harbor, (GM CA 18943-01) and the Philippe Foundation Inc. to M. G., .N.Y. pp. 455-468. and grants from the Centre National de la Recherche Scien- 21. Ishibashi, M., Sugino, Y. & Hirota, Y. (1964) J. Bacteriol. tifique, the D6l6gation G6nerale A la Recherche Scientifique et 87, 554-561. Technique to Y.H. The work of L. M. was done at the Institut 22. Chabbert, Y. A., Scavizzi, M. R., Witchitz, J. L., Gerbaud, a leave of absence from the Public Health Re- G. & Bouanchaud, D. H. (1972) J. Bacteriol., in press. Pasteur during 23. Gross, J. D., Grunstein, J. & Witkin, E. M. (1971) J. Mol. search Institute of the City of New York. Biol. 58, 631-634. 1. De Lucia, P. & Cairns, J. (1969) Nature 224, 1164-1166. 24. Campbell, J. L., Soll, L. & Richardson, C. C. (1972) Proc. 2. Kornherg, A. (1969) Science 163, 1410-1418. Nat. Acad. Sci. USA 69, 2090-2094. Downloaded by guest on September 24, 2021