Structural Gene for Rnase H of Escherichia Coli by Cloning of The

Proc. NatL Acad. Sci. USA Vol. 78, No. 6, pp. 3770-3774, June 1981 Genetics

Identification of the dnaQ gene product and location of the structural gene for RNase H of Escherichia coli by cloning of the genes (mutator/maxicell/insertion mutant) TAKASHI HORIUCHI, HISAJI MAKI, MASAO MARUYAMA, AND MUTSUO SEKIGUCHI Department of Biology, Faculty of Science, Kyushu University 33, Fukuoka 812, Japan Communicated by Motoo Kimura, March 11, 1981

ABSTRACT By in vitro recombination we have constructed MATERIALS AND METHODS hybrid plasmids capable of complementing a conditional lethal mutator mutation, dnaQ49, in Escherichia coli K-12. The dnaQ+ Bacteria, Plasmids, and Phages. All bacterial strains used in plasmids consist of a full-length pBR322 DNA and a 1.5-kilobase these experiments are derivatives of E. coli K-12. Strains DNA fragment derived from the E. coli chromosome. Specific la- KH1171 (F-, recAl dnaQ49 strA) and KH1188 (Hfr, dnaQ49) beling of plasmid-encoded proteins by the maxicell method re- were isolated in this laboratory. Strain CSR603 (F-, uvrA6 vealed that the 1.5-kilobase insert codes for two proteins, one recAl phr-1) is a gift of D. Rupp and was used in maxicell ex- whose molecular weight is 25,000 [the 25-kilodalton (kDal) protein] periments (6). Strain KP370 (F+, recAl), provided by T. Miki, and the other whose molecular weight is 21,000 (the 21-kDal pro- was used in y8 insertion experiments (7). The vector plasmid tein). Because insertion of y8 sequence into the dnaQ gene of the pBR322 (8) and the vector phage Agt-AC (9) were furnished by plasmid resulted in disappearance of the 25-kDal protein, it was Y. Sakaki and Y. Nakamura, respectively. concluded that the 25-kDal protein is the dnaQ gene product. The Construction of Hybrid Plasmids. F+ strains harboring the 21-kDal protein was identified as RNase H on the basis of the fol- ColEl hybrid plasmids in the Clarke and Carbon colony bank lowing evidence. (i) Cells harboring the dnaQ+ plasmids, with or (10), provided by K. Ueda, were screened by conjugation. Cells without the y8insertion in the dnaQ gene, had a 5- to 7-fold higher harboring pLC28-22 and pLC34-20 were found to complement level of RNase H activity than cells harboring pBR322. (ii) After the dnaQ and dnaE defective mutants, respectively. The plas- induction of cells that are lysogenized with dnaQ+-transducing A mids and their derivatives were isolated (10) and digested with phages, RNase H activity increased considerably. A similar high EcoRI. The DNA fragments were inserted into the EcoRI site level of RNase H activity was observed with transducing phages of pBR322. Details of the construction and properties of the whose dnaQ function was inactivated by insertion ofa transposon, hybrid plasmids will be published elsewhere. Transformation Tn3, into the gene. (iii) The plasmid-encoded RNase H, labeled Mandel and with [3S]methionine, was purified in a manner essentially similar and transfection were performed as described by to that ofthe chromosome-encoded enzyme. These results suggest Higa (11). that the dnaQ gene and the structural gene for RNase H, termed Transposition of the y6 Sequence to Plasmids. The y8 se- gene rnh, are closely linked and located at 5 min on the linkage quence was inserted into plasmids according to the method of map. Guyer (7). Strain KP370 (F+, recAl) was transformed with the dnaQ+ plasmids, into which the yc was to be inserted, and the Mutators are a special class ofmutations that render many other transformants were subsequently conjugated with strain KH1171 genes unstable (1). In Escherichia coli, many types ofmutators (F-, recAl dnaQ49 strA). Because only the plasmids that re- have been isolated and characterized, mostly by genetic means ceived the y8 from F factor can be transferred to the recipient (24). Although speculative interpretations for mechanisms cells (7), the recombinants showing ampicillin- and tetracycline- leading to increment mutation frequency by the mutator mu- resistance characters were selected and their plasmids were tation have been proposed, little is known about the nature of analyzed. The existence ofthe y8 sequence (5.7 kb) in the plas- products encoded by mutator genes and their functions. mids was confirmed by electrophoresis ofthe DNA after EcoRI Recently, we discovered a strong mutator mutation that also digestion. Some ofthe plasmids with the y8 sequence exhibited caused temperature-sensitive growth and defective DNA syn- dnaQ- character, probably due to insertion ofthe y8 inside the thesis in cells carrying the mutation (5). It was found that the dnaQ gene. mutation was in a new gene, designated dnaQ, which was lo- Construction of Hybrid Phages. The dnaQ+-transducing A cated at 5 min on the E. coli genetic map. phage, designated AdnaQ+, was constructed by recloning the By using in vitro recombination techniques we have con- 1.5-kb DNA fragment derived from pMM5, which carries the structed plasmids carrying the dnaQ gene. The hybrid plasmids dnaQ gene (see Table 1), into a vector phage, Agt-AC (9). To contain a 1.5-kilobase (kb) insert of E. coli DNA that codes for transpose Tn3 to the AdnaQ+, the phage was propagated in cells two proteins whose molecular weights are 25,000 and 21,000. that harbor pKY2283 (ColEl: :Tn3). Progeny phages that had By biochemical and genetic analyses, the 25-kilodalton (kDal) acquired ampicillin resistance, conferred by Tn3 (12), were protein was identified as the dnaQ gene product and the 21- isolated and their DNAs were analyzed. kDal protein, as RNase H. It was suggested that the dnaQ gene Labeling of Plasmid-Encoded Proteins. The maxicell method and the structural gene for RNase H are adjacent on the E. coli of Sancar et al. (6) with a minor modification (13) was used. chromosome. Bacteria were grown at 37°C in M9 medium supplemented with 1% Casamino acids and 0.2% glucose. When OD at 660 nm x was irra- The publication costs ofthis article were defrayed in part by page charge reached about 0.4 (3 108 cells per ml), the culture payment. This article must therefore be hereby marked "advertise- nent" in accordance with 18 U. S. C. §1734 solely to indicate this fact. Abbreviations: kb, kilobase pair(s); kDal, kilodalton(s). 3770 Genetics: Horiuchi et al. Proc. Natl. Acad. Sci. USA 78 (1981) 3771 diated with ultraviolet light (60 J/m2). After the culture was Table 1. Properties of hybrid plasmids shaken for 1 hr, D-cycloserine was added to give a final concentration of 100 ,gg/ml and the culture was shaken overnight Complemen- Size of DNA, kb at 370C. One milliliter of the culture was centrifuged: the cells tation E. coli Y3 were washed and resuspended in 1 ml ofM9 supplemented with Plasmid dnaQ dnaE pBR322 chromosome insertion 25 ug each of 18 amino acids other than methionine and cys- pMM1 - + 4.4 9 - teine. [3S]Methionine (1000 Ci/mmol; 1 Ci = 3.7 X 1010 pMM4 + - 4.4 4.5 - becquerels) was added to give 82 ,uCi/ml, and the proteins syn- pMM5 + - 4.4 1.5 - thesized under these conditions were labeled for 1 hr at 370C. The labeled cells were centrifuged, washed, resuspended in 0.1 pMM5::y8(dnaQ+) + NT 4.4 1.5 5.7 ml of a buffer, and heated for 2 min. The sample (25 pl) was pMM5: :y(dnaQ-) - NT 4.4 1.5 5.7 applied to 12.5% NaDodSO4polyacrylamide gel. NT, not tested. Assay of RNase H Activity. The procedure for preparation ofcell extract was essentially the same as described by Wickner et al. (14). RNase H activity was assayed according to Berkower hand, pMML, which complements the dnaE but not the dnaQ et al. (15). The reaction mixture contained 3 ,uM poly(A) (483 defect of recipient cells, had a 9-kb EcoRI fragment. cpm/pumol: 40-140 nucleotide residues) hybridized with 10 ttM Identification of the dnaQ Gene Product. To identify the poly(dT), 40 mM Tris HCl (pH 7.7), 4 mM MgCl2, 1 mM di- dnaQ gene product, proteins encoded by the dnaQ+ plasmids thioerythritol, bovine serum albumin at 30 Ag/ml, 4% (vol/ were specifically labeled by the maxicell method developed by vol) glycerol, and an extract. The reaction ran for20 min at 30°C, Sancar etal. (6) and analyzed by NaDodSOjpolyacrylamide gel and then the radioactivity converted into acid-soluble form was electrophoresis. Besides two known pBR322-encoded proteins determined. One unit of activity was defined as the amount of (28 and 37 kDal), pMM5 produced two specific proteins, a 25- enzyme producing 1 nmol ofacid-soluble material in 20 min at kDal protein that comigrated with 25-kDal chymotrypsinogen 300C. A and a 21-kDal protein that ran slightly faster than 21.5-kDal Analysis of RNase H. E. coli CSR603 harboring pMM5 or soybean trypsin inhibitor (Fig. 1). The corresponding bands pBR322 (60 ml each) was labeled with [35S]methionine (1000 Ci/ were found also in pMM4 (data not shown). mmol; 40 ,Ci/ml) by the maxicell method. The procedure and To determine which protein is specified by the dnaQ gene, the conditions were as described in a preceding section except we introduced the yO sequence into pMM5. Two types ofplas- that the dose ofultraviolet irradiation was 30 J/m2. The labeled mids with yv insertion were selected: pMM5:: y6(dnaQ+), cells were collected by centrifugation, washed, and frozen in which still complements the dnaQ mutation, and liquid nitrogen. The cells were thawed, resuspended in 2 ml pMM5:: y8(dnaQ7), which does not complement the mutation of2.5 M NaCVl mM EDTA/1 mM 2-mercaptoethanoV20 mM (Table 1). Then, proteins produced by these plasmids were ana- Tris*HCl, pH 7.4, and disrupted by sonication. To the crude lyzed by gel electrophoresis. pMM5:: y5(dnaQ+) produced the extract (1.5 ml), polyethylene glycol 6000 was added to a final two proteins originally found in pMM5 but pMM5:: y6(dnaQ7) concentration of 10%, and the mixture was centrifuged. The lacked the 25-kDal protein (Fig. 1, lanes D and E). It is likely, supernatant fluid was dialyzed against buffer C (16) and 1 ml of the dialysate was applied to a column ofphosphocellulose (2 ml). The column was washed with 4 ml of buffer C and eluted with A B C D E F a linear gradient of KC1 (0-0.8 M) in buffer C. Fractions (20 drops each) were collected and assayed for RNase H activity. Fractions at or around a peak ofthe enzyme activity were concentrated and subjected to NaDodSOJpolyacrylamide gel -68 electrophoresis followed by staining and fluorography. Other Methods. Polyacrylamide gel electrophoresis in the presence of NaDodSO4 was essentially the procedure of Lae- -39 mmli (17). Fluorography was done by soaking the stained gel for 1 hr in EN3HANCE (New England Nuclear) and for 1 hr in water, drying, and exposing to x-ray film (Kodak XR-1) at --25-. 5 -70°C. Protein was determined by the method of Lowry et al. (18). ii. ....; -21.5 RESULTS E 4 Construction of Plasmids Carrying the dnaQ gene. Hybrid .:...... }=Ni..:'-- plasmids capable of complementing dnaQ49 mutation were M..s made by in vitro recombination using a multicopy plasmid, pBR322, as vector. Both high mutability and conditional le- thality of the dnaQ mutant were almost completely suppressed by introduction ofthe dnaQ+ plasmids into the cells. Similarly, FIG. 1. Identification ofproteins produced by pMM5 and its inser- a plasmid carrying the dnaE gene was constructed, which tion mutants. The polypeptides codedforbyplasmids were labeled with served as a control. [35S~methionine by using a maxicell system and analyzed by electro- One of the dnaQ+ plasmids, pMM5, carried a 1.5-kb EcoRI phoresis on 12.5% NaDodSO/polyacrylamide gels followed by fluo- fragment derived from E. coli chromosome; the other, pMM4, rography. The positions of bovine serum albumin (68 kDal), E. coli a EcoRI RNA polymerase a subunit (39 kDal), chymotrypsinogen A (25 kDal), had 4.5-kb fragment (Table 1). The 4.5-kb fragment of and soybean trypsin inhibitor (21.5 kDal) are indicated. Each lane rep- pMM4 was further divided into a 1.5-kb fragment and a 3-kb resents CSR603 cells harboring one of the following plasmids; A, fragment by EcoRI treatment. Thus, the dnaQ gene seems to pBR322; B, pMM5; C, no plasmid; D, pMM5:: y5(dnaQ+); E, reside on the 1.5-kb insert ofthe hybrid plasmids. On the other pMM5:: yO(dnaQ-); F, pMM5. 3772 Genetics: Horiuchi et al. Proc. Natl. Acad. Sci. USA 78 (1981)

Table 2. RNase H activity in cells harboring various plasmids Table 3. RNase H activity in cells harboring plasmids RNase H with y8 insertion E. coli gene Specific activity, Ratio to dnaQ RNase H Plasmid* on plasmid units/mg protein pBR322 genotype Specific activity, Ratio to pBR322 - 17.9 1.0 Plasmid* of plasmid units/mg protein pBR322:: ('y8) pMM4 dnaQ 102.6 5.7 pBR322:: y8 None 21.2 1.0 pMM5 dnaQ 120.1 6.7 pMM4 dnaQ+ 134.9 6.4 pMMl dnaE 16.3 0.9 pMM4::y(dnaQ-) dnaQ: :y8 182.0 8.6 None - 4.4 0.25 pMM5:: y(dnaQ+) dnaQ+ 167.1 7.9 pMM5:: y(dnaQ-) dnaQ: : y8 154.0 7.3 * Strain CSR603 was used as host for these plasmids. * The host for the plasmids was strain CSR603. therefore, that the dnaQ gene codes for a protein whose size In the case ofcontrol A phage, no such increase was observed. is approximately 25 kDal. RNase H Activity in Insertion Mutants. To determine Increased Level of RNase H Activity in Cells Harboring whether the dnaQ gene itself controls the RNase H activity or dnaQ+ Plasmids. In a search for an enzyme activity relevant to not, we measured the enzyme activity in the dnaQ+ plasmids the dnaQ phenotype we noticed that an extract of cells that into which the y8 sequence was inserted. The same high level harbor the dnaQ+ plasmids contained a higher level of RNase of the enzyme activity was found in the two types of insertion H activity than did that ofnormal cells. RNase H activity ofcells mutants, one that produces the dnaQ gene product and the harboring pMM4 or pMM5 was 5- to 7-fold that of cells har- other unable to produce the gene product (Table 3). Thus, the boring pBR322 (Table 2). RNase H activity in cells harboring dnaQ gene does not seem to be related to RNase H. pMMl, which carries the dnaE but not the dnaQ gene, was This idea was supported by other experiments in which normal. dnaQ-transducing A phages were used. Transducing phages In these experiments, 3H-labeled poly(A) hybridized with whose dnaQ function was inactivated by transposition of Tn3 unlabeled poly(dT) was used as substrate to assay the enzyme were able to induce high levels of RNase H activity (Table 4). activity. When [3H]poly(A) alone was used no difference among Thus, there arises a possibility that, in the two proteins encoded the extracts was observed. Moreover, RNase H activity in cells by the 1.5-kb insert, the 21-kDal protein may be concerned with harboring dnaQ+ plasmids was sensitive to N-ethylmaleimide, RNase H. as was observed with the enzyme in normal cells (15). From Identification of the 21-kDal Protein as RNase H. A culture these results it was suggested that a gene that controls RNase of E. coli CSR603 harboring pMM5 was labeled with H activity is present in the dnaQ+ plasmids. [3S]methionine by the maxicell method and the extract was To confirm this, we recloned the 1.5-kb DNA into A phage subjected to phase separation and phosphocellulose chroma- vector. Fig. 2 shows one ofthe representative results. After in- tography. As a control, CSR603 cells harboring pBR322 were duction of cells that are lysogenized with dnaQ+-transducing RNase H activity increased almost linearly. processed similarly. phage (AdnaQ+), In both the activity was eluted as a single peak (Fig. The level at 3 hr ofincubation was about 8 times the initial level. samples 3) at the same position at which RNase H was reported to be eluted under these conditions (15, 19). Almost all the activity applied to the column was recovered in the peak fractions, and 10 r the ratio ofthe activities ofthe two samples in the partially purified fractions was essentially similar to that in the crude extracts. Thus, RNase H found in CSR603(pMM5) is chromato- graphically indistinguishable from that of normal cells. 04 The enzyme was further analyzed by NaDodSOJpolyacryl-

Table 4. RNase H activity in induced AdnaQ+ lysogens dnaQ RNase H U - genotype Specific activity, Ratio to Phage* on phage units/mg protein control Exp. I None (control) 10.6 1.0 AdnaQ+ dnaQ+ 47.0 4.4 0 1 2 3 AdnaQ(Tn3)-6 dnaQ+ 88.5 8.3 dnaQ: :Tn3 86.0 8.1 Time after induction, hr AdnaQ(Tn3)-3 FIG. 2. Increase of RNase H activity after induction of AdnaQ+ Exp. II phage lysogen. The lysogenic cells grown overnight were diluted into AdnaQ(Tn3)-1 dnaQ+ 79.2 7.5 fresh L broth and shaken at 30°C. When OD at 660 nm reached 0.2 AdnaQ(Tn3)-2 dnaQ+ 121.7 11.5 (2 x 108 cells per ml), the culture was exposed to 420C for 10 min and AdnaQ(Tn3)-3 dnaQ: :Tn3 88.0 8.3 then shaken at 3700. At appropriate intervals, aliquots ofthe culture AdnaQ(Tn3)-4 dnaQ: :Tn3 91.6 8.6 were withdrawn and centrifuged, and the precipitated cells were frozen AdnaQ(Tn3)-5 dnaQ: :Tn3 88.3 8.3 in liquid nitrogen. After thawing, the cells were resuspended in 0.1 vol of buffer C (16) without glycerol, disrupted by sonication, and centri- * KH1188(AC1857 Qam) was used as host for phages. All phages car- fuged; the supernatant was used for the assay of RNase H activity. ried CI857 and Qam mutations in addition to the dnaQ allele. Ex- *, KH1188(AdnaQ+ CI857 Qam, ACI857 Qam); 0, KH1188(ACI857 tracts were prepared from cells incubated for 3 hr after heat Qam). induction. Genetics: Horiuchi et al. Proc. Natl. Acad. Sci. USA 78 (1981) 3773

A amide gel electrophoresis. Fig. 4 A and B present patterns of BI I proteins present in each of the enzyme fractions from the two types of cells. In the peak fractions (fractions 15 and 16) of .RS 1( l.8 CSR603 (pMM5) there was a distinct band which corresponds II~~~~~~~ l.7 to the 21-kDal protein, whereas only a faint band was observed P. 0.6 at the corresponding position of the CSR603(pBR322) sample. 0.5 Clearer resolution of the protein was obtained with fluoro- 0 5 0.4 0.3 g graphy (Fig. 4 C and D). A band for the 21-kDal protein was 0.2 found in fractions 15 and 16 of CSR603(pMM5) but not in the w 0.1 corresponding fractions of CSR603(pBR322). These results in- 0 dicate that the 21-kDal protein encoded by the dnaQ+ plasmid 5 10 15 20 5 10 15 20 is indeed RNase H. Fraction FIG. 3. Resolution of RNase H by phosphocellulose chromatogra- DISCUSSION phy. Extracts of cells, whose plasmid-encoded proteins were labeled We have isolated and characterized a conditional lethal mutator with were applied to a phosphocellulose column, and [35S]methionine, mutant in E. coli K-12 All the the proteins were eluted by a linear gradient ofKCL. RNase H activity (dnaQ49) (5). dnaQ-defective was assayed and the fractions at or around a peak, indicated by a mutants thus far examined show the following pleiotropic phe- bracket, were used for further analyses, as shown in Fig. 4. (A) notype (refs. 5 and 20; unpublished results): (a) strong mutator CSR603(pBR322); (B) CSR603(pMM5). activity: (b) inability to produce colonies at 44.5°C in NaCl-free

C 12 14 16 18 20 M C 12 14 16 18 20 M

,.011i.. it. _II .iEWL Im I;

w 3-

A B

- - -

C 0

FIG. 4. Analysis ofRNase H fractions by NaDodSO4/polyacrylamide gel electrophoresis. Enzyme fractions indicated by brackets in Fig. 3 were concentrated by lyophilization and applied to 12.5% NaDodSO4/polyacrylamide gels. After electrophoresis, the gels were stained with dye and dried (A and B). The gels were then exposed to x-ray films for 16 days for fluorography (C and D). The numbers above the gels correspond to fraction numbers in Fig. 3; C, sample of whole cells of CSR603(pMM5); M, molecular weight markers; arrows, pMM5-specific 21-kDal protein. (A and C) Proteins of CSR603(pBR322); (B and D) proteins of CSR603(pMM5). 3774 Genetics: Horiuchi et al. Proc. Natl. Acad. Sci. USA 78 (1981) L-broth plates; (c) decreased level of DNA synthesis under the that the enzyme might be responsible for removing RNA that restrictive conditions; (d) increased sensitivity to DNA-inter- acts as a primer for DNA synthesis (25, 26). Recently, Ito and calating reagents or inhibitors of DNA gyrase, such as 5-ami- Tomizawa (27) reported that RNase H has a specific role in the noacridine and novobiocin; and (e) enhancement ofthermosen- process ofinitiation ofColEl DNA synthesis in vitro. However, sitivity of dnaEt" mutants, having mutations in the gene for definite evidence that RNase H is involved in cellular DNA DNA polymerase III. These properties of the mutants sug- replication has not been available. The plasmids carrying the gested that the dnaQ product might be a component ofthe cel- rnh gene should be useful in solving this problem. lular DNA synthesizing machinery or a factor stabilizing the chromosome structure and that its defectiveness may cause a We are grateful to S. Hirose for his helpful suggestion and discussion. decrease in the fidelity of DNA replication. We also thank D. Rupp, K. Ueda, K. Sakakibara, Y. Sakaki, K. Shimada, It seemed to be important to identify the dnaQ gene product T. Miki, and Y. Nakamura for bacterial strains, phages, plasmids, and and to elucidate its function. As a first step in this direction we helpful advice. This work was supported in part by grants from the constructed plasmids carrying the dnaQ gene by in vitro re- Ministry ofEducation, Science and Culture ofJapan and by a grant from combination. The hybrid plasmids contain a 1.5-kb insert ofE. the Sakkokai Foundation. coli DNA that codes for two proteins; one ofthe proteins, with 1. Cox, E. C. (1976) Annu. Rev. Genet. 10, 135-156. a molecular weight of 25,000, was identified as the dnaQ gene 2. Cox, E. C. & Yanofsky, C. (1969)J. Bacteriol. 100, 390-397. product. 3. Degnen, G. E. & Cox, E. C. (1974)J. Bacteriol. 117, 477-487. Among many proteins known to be involved in DNA repli- 4. Sevastopoulos, C. G. & Glaser, D. A. (1977) Proc. Nati. Acad. cation and related processes, there are some whose molecular Sci. USA 74, 3947-3950. weights are around 25,000. These include subunit E ofthe DNA 5. Horiuchi, T., Maki, H. & Sekiguchi, M. (1978) Mol.b Gen. Genet. 163, 277-283. polymerase III core enzyme (21) and discriminatory factor f3 6. Sancar, A., Hack, A. M. & Rupp, W. D. (1979)J. Bacteriol. 137, (22). There is a possibility that the dnaQ gene may encode one 692-693. of these proteins. Availability of the dnaQ+ plasmids makes it 7. Guyer, M. S. (1978)J. Mol. Biol. 126, 347-365. possible to examine such a possibility. 8. Bolivar, F., Rodriguez, R. L., Greene, P. J., Betlach, M. C., In the course of analysis of the dnaQ+ plasmids, we found Heyneker, H. L., Boyer, H. W., Crosa, J. H. & Falkow, S. that a protein with a molecular weight of 21,000 also was en- (1977) Gene 2, 95-113. 9. Leder, P., Tiemeier, D. & Enquist, L. (1977) Science 196, 175- coded by the 1.5-kb DNA insert. The protein was identified as 177. RNase H on the basis of the following criteria (15, 16, 19): (a) 10. Clarke, L. & Carbon, J. (1976) Cell 9, 91-99. the molecular weight is identical to that RNase H of E. coli; 11. Mandel, M. & Higa, A. (1970)J. Mol. Biol. 53, 159-162. (b) the enzyme specifically hydrolyzes RNA in a DNARNA 12. Kopecko, D. J. & Cohen, S. N. (1975) Proc. Natl. Acad. Sci. USA hybrid, and the activity is inhibited more than 90% by N-ethyl- 72, 1373-1377. and the enzyme is purified in a manner similar 13. Little, J. W., Edmiston, S. H., Pacelli, L. Z. & Mount, D. W. maleimide; (c) (1980) Proc. Natl. Acad. Sci. USA 77, 3225-3229. to that ofE. coli RNase H. Thus, it is evident that the gene for 14. Wickner, S., Wright, M., Berkower, I. & Hurwitz, J. (1974) in RNase H resides on the DNA fragment. Because the radioactive Methods in Molecular Biology: DNA Replication, ed. Wickner, 21-kDal protein was produced under conditions such that only R. B. (Dekker, New York), Vol. 7, pp. 195-219. plasmid-encoded proteins are labeled, itwas concluded that the 15. Berkower, I., Leis, J. & Hurwitz, J. (1973) J. Biol. Chem. 248, gene is the structural gene rather than the regulatory gene. 5914-5921. current for nomenclature ofgenes 16. Ito, T. & Tomizawa, J.-I. (1978) Cold Spring Harbor Symp. According to the convention Quant. Biol. 43, 409-417. (23), the gene is to be named rnh. 17. Laemmli, U. K. (1970) Nature (London) 227, 680-685. The sum of the molecular weights of the two proteins, the 18. Lowry, O. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. dnaQ gene product and RNase H, is approximately 46,000, (1951) 1. Biol. Chem. 193, 265-275. which isjust within the coding capacity ofthe cloned DNA frag- 19. Darlix, J.-L. (1975) Eur. J. Biochem. 51, 369-376. ment. From the space limitation it seems that there is no other 20. Horiuchi, T., Maki, H. & Sekiguchi, M. (1981) Mol. Gen. Genet. in intact form in the fragment, implying that dnaQ and rnh 181, 24-28. gene 21. McHenry, C. S. & Crow, W. (1979) J. Biol. Chem. 254, 1748- are adjacent. The dnaQ gene has been mapped at 5 min on the 1753. E. coli chromosome (5, 23). The rnh gene must lie within 0.1 22. Sumida-Yasumoto, C., Ikeda, J.-E., Benz, E., Marians, K. J., min from the dnaQ locus. Vicuna, R., Sugrue, S., Zipursky, S. L. & Hurwitz, J. (1978) Cold Recently, Carl et al. (24) isloated an E. coli mutant with a Spring Harbor Symp. Quant. Biol. 43, 311-329. moderately lowlevel ofRNase H activity (30% ofparental level). 23. Bachmann, B. J. & Low, K. B. (1980) Microbiol. Rev. 44, 1-56. on the E. coli 24. Carl, R. L., Bloom, L. & Crouch, R. J. (1980)J. Bacteriol. 144, The mutation (rnh) has been mapped at 5.1 min 28-35. chromosome. Because this is very close to the rnh locus deter- 25. Keller, W. (1972) Proc. Natl. Acad. Sci. USA 69, 1560-1564. mined in the present study, it seems that the mutation arose 26. Watson, J. D. (1972) Nature (London) New Biol. 239, 197-201. in the structural gene for RNase H. 27. Ito, T. & Tomizawa, J.-I. (1980) Proc. Natl. Acad. Sci. USA 77, From the mode of action of RNase H it has been supposed 2450-2454.