J. Mol. Biol. (1992)

226,

989-996

dnaR Function of the pm Gene of Escherichia coli in Initiation of Chromosome Replication Yoshimasa Sakakibara Department of Chemistry, National Institute of Health 2-10-35, Kamiosaki, Shinqawa, Tokyo 141, Japan (Received

30 September 1991; accepted 8 April

1992)

Escherichia coli mutant named dnaR, which was temperature sensitive in initiation of DNA replication, has been characterized through identification of the mutant gene. A I.65 x lo3 base-pair chromosomal DNA fragment isolated from wild-type cells, but not the exhibited an activity that reversed corresponding fragment from the dnaR mutant, temperature-sensitive growth of the mutant. The DNA fragment was found to include the entire prs-coding sequence and specify a 34,000 M, protein with phosphoribosylpyrophosphate synthetase activity. The dnuR mutation resided within the prs-coding segment and made the synthetase thermolabile. The coding segment for the dnaR product was determined, by introduction of various mutations into the cloned fragment, to be the same as that for the synthetase. The dnuR function of the prs gene product in DNA replication is discussed on the basis of an observation that thermal treatment of the dn.aR mutant caused a delay in initiation of chromosome replication after the downshift, despite the presence of the synthetase activity at the preheat level.

A new

Keywords: replication

initiation;

1. Introduction

Jensen et al., 1986). This enzyme catalyses a pyrophosphoryl transfer from ATP to ribose-5-phosphate (Kornberg et al., 1955). PRPP is a precursor in the initial step of a highly branched network for biosynthesis of various nucleotides (for a review, see Neuhard & Nygaard, 1987). The result presented here discloses that the prs gene participating in nucleotide precursor synthesis is involved in initiation of chromosome replication. The function of the prs gene in initiation of chromosome replication has been studied by the analysis of DNA synthesis in the dnaR mutant after thermal treatment. Based on the results obtained, the possibility that the prs gene product exerts the dnaR function as one of the proteins involved directly in initiation of chromosome replication will be discussed.

Replication of the Escherichia eoli chromosome is initiated from a fixed site, oriC. The chromosomal DNA fragment including or& is capable of replicating autonomously as an extrachromosomal factor (Yasuda & Hirota, 1977). The replication of or&’ plasmid depends on various gene functions involved in bacterial DNA replication (for a review, see von Meyenburg & Hansen, 1987). In the accompanying paper (Sakakibara, 1992), I have reported that a newly isolated mutant dnaR, which is thermosensitive in initiation of chromosome replication, permits replication of oriC plasmid at a high temperature restrictive for its own chromosome replication. The plasmid replication in dnaRdeficient cells depends on the dnaA function essential for the initiation of replication from oriC. This lack of a requirement for the dnaR function in oriC plasmid replication distinguishes the gene encoding the dnaR product from previously identified genes that participate in replication of bacterial chromosome. Here, I report that the dnaR function is encoded by the prs gene, the gene for phosphoribosylpyrophosphate (PRPPT) synthetase (EC 2.7.6.1) (HoveTAbbreviations pyrophosphate;

kb,

dnaR; prs; PRPP synthetase

2. Materials (a)

Bacterial

and Methods and growth media

strains

strains used were HC5101 (dnaR+), HC5103 HC5105 (dnaRl30 rnhl7 :: Tn3) (Sakakibara, 1992) and minicell-producing TH1219 (Harayama & Hazelbauer, 1982). L broth contained l.Oo/, (w/v) Tryptone (Difco Laboratories, Detroit, MT), 0.5’7’ yeast extract (Difco) and 0.1 M-NaCI. Thymine was added at 4 pg/ml, except when labeling cells with radioactive Bacterial

(dnaRl30),

used: PRPP, 5-phosphoribosyl-lIO3 base-pairs.

989 0022-2836/92/16098948

$08.00/O

0

1992 Academic

Press

Limited

Y. Sakakibara

990

thymine. For solid media, Bacto agar (Difco) was added at 1.5% (w/v). D medium contained 7 g K,HPO,, 2 g KH,PO,, 0.5 g Na-citrate.2H,O, 1 g (NH&SO,, @I g MgSO,. 7H,O and 2 g glucose in 1000 ml H,O. DC medium contained casamino acids (Difco) at 95% (w/v) in D medium. (b) Drugs

and

radioactive

materials

Ampicillin and chloramphenicol were the products of Sigma Chemical Co., St Louis, MO and Sankyo Co. Ltd.. Tokyo, respectively. [2-14C]thymine (55 mCi/mmol), [methyZ-3H]thymidine (49 Ci/mmol) and [l-‘4C]leucine (50 mCi/mmol) were purchased from ICN Radiochemicals. Irvine, CA and [Y-~*P]ATP (IO Ci/mmol) from New England Nuclear, Boston. MA. (c) Measurement

of DNA

synthesis

Cells were grown in L broth containing [14C]thymine at 0.1 nCi/ml, and radioactivity incorporated into acid-insoluble material was measured as described (Sakakibara, 1992). To analyse DNA synthesis in the oriC and terC regions, cells were pulse-labeled with [3H]thymidine in a 1.0 ml culture as indicated in the text. DNA was extracted and dissolved in 650 ml of 95 M-Nacl after alkali denaturation, as described (Sakakibara, 1992). A 020 ml portion of the sample was incubated in duplicate at 42°C for 18 h with probe DNAs immobilized separately on nitrocellulose HA45 disks (Millipore Corp., Bedford, MA) in I.0 ml hybridization mixture (Sakakibara, 1988). The amount of probe DNA used was 92 lg for pSY317 (Yasuda & Takagi, 1983) and 64 pg for ISA5 (Kohara et al., 1987). A 40 ~1 portion of the sample was used to measure input radioactivity for hybridization. (d)

Cloning

of the gene with

dnaR

activity

HC5103 (dnaRl30) cells were transformed with pBR322 plasmids carrying BamHI-digested fragments of dnaR+ bacterial chromosome, using the method of Mandel & Higa (1970). One of the transformants obtained at 42°C carried pSRl30 plasmid, which had an insert of 28 kilobase-pairs (kb). A 2.3 kb EcoRI fragment of the insert was cloned into pBR322 plasmid at the EcoRI site to construct pSRl31. pSRl32 plasmid was constructed by ligation of PvuII-digested pSRl31 DNA (Fig. l(a)), which lacked the EcoRI-PvuII segment derived from pBR322. pSRl33 and pSRl34 were the subclones of pSRl30 plasmid, which lacked the 69 kb Hind111 region and I.9 kb AvaIII-Sal1 region, respectively (Fig. l(a)). To isolate the gene carrying the dnaR mutation, the BamHI DNA fragment of pSRl30 plasmid was cloned into 1D69 DNA (Mizusawa & Ward, 1982) and the resultant phage IdnaR+ was integrated into the chromosome of Aattl dnaRl30 cells by homologous recombination. Among phages induced from the lysogen were found those lacking the activity to reverse temperature-sensitive growth of HC5103 (dnaRZ30) cells lysogenic for iimm’l. One of the dnuR-defective phages, IdnaRl30, carried the BamHI fragment corresponding to that of IdnuR+ DNA. The BamHI insert of ldnaRl30 phage was cloned into pBR322 plasmid lacking the EcoRI-Hind111 region to construct pSR301 (Fig. l(b)). The BamHI fragment of pSRl30 (dnaR+) DNA was also cloned into the same vector to construct pSRl30b, which had single Hind111 and AvaIII sites in the insert and single EcoRV and Sal1 sites in the vector region. By replacing the 98 kb EcoRV-HindIII, 66 kb HindIII-AwaIII and 1.9 kb

AwaIII-Sal1 fragments of pSRl30b DNA with the corresponding fragments of pSR301 (dnaRl30) DNA, recombinant plasmids pSR302, pSR303 and pSR304 were constructed (Fig. l(b)). Enzymes used for DNA manipulation were purchased from Takara Shuzo Co. Ltd., Tyoto, Toyobo Co. Ltd.. Osaka. or New England Biolabs Inc., Beverly, MA. Nucleotide sequences were determined using the dideoxychain termination method (Sanger et al., 1977) according to the manual of Sequenase version 2.0 (US Biochemical Corp.. Cleveland, OH). (e) Construction

of frame-shift

and point

mutant

plasmids

Frame-shift mutants were constructed from pSR132 DNA by ligation of the SphI- or HindIII-digested DNA after blunt-end formation using T4 DNA polymerase or the Klenow fragment of DNA polymerase (see Fig. 4). Point mutants were constructed from pSRl32 DNA by replacing the SphI-Hind111 segment with the corresponding double-stranded synthetic oligonucleotide containing a substituted base-pair (see Fig. 4). The presence of the mutation in each resultant plasmid was confirmed by nucleotide sequencing of the region including the SphI and Hind111 sites. (f) Assay

of PRPP

synthetase

Cells were grown in 50 ml L broth at 30°C until A,,, reached I.0 ( -4 x lo* cells/ml). The cells were harvested, washed with 50 mM-potassium phosphate buffer (pH 7.5) at 4°C and suspended in 1 ml of this buffer. The cell suspension was sonicated 4 times for 15 s in an ice-water bath and centrifuged at 15,000 g for 10 min at 4°C. The supernatant was used for assay of PRPP synthetase activity. as described (Switzer & Gibson, 1978). The reaction mixture (50 /*l) consisted of 50 mM-triethanolamine50 mw-potassium phosphate buffer (pH 8.6), 5 mM-ribose5-phosphate, 10 miw-MgCl,, 25 mM-NaF, 637 mM-EDTA, 2 miw-[y-32P]ATP (1 mCi/mmol), 1 to 20 pg of cellular proteins. For dilution of the proteins, 50 mM-potassium phosphate buffer (pH 7.5) containing bovine serum albumin at 1 mg/ml was used. The reaction was conducted at 30°C for 5 min and stopped by addition of HCOOH to a final concentration of @I M. A reaction in the absence of ribose-5-phosphate was also run to estimate the activity specific for the synthetase. A portion of the reaction mixtures was applied to a polyethyleneiminecellulose thin-layer plate (Baker Inc., Phillipsburg, NJ) and developed in 685 M-potassium phosphate buffer (pH 3.4) (Jensen et al., 1979). The plate was exposed to an X-ray Rx film (Fuji Photo Film Co. Ltd., Tokyo) for about 15 h, and the PRPP region was cut from the chromatogram to measure radioactivity. The amount of proteins was determined as described (Bradford, 1976), using bovine serum albumin as standard. Most of chemicals used were the products of Sigma Chemical Co. (g) Expression

of pbsmid-specified

proteins

in minicells

TH1219 cells carrying plasmids were grown at 37°C in 100 ml DC medium containing ampicillin at 25 pg/ml to a cell density of about 5 x lOs/ml. The culture was centrifuged at 2900 g for 5 min at 4°C to remove most of the intact cells. Cells in the supernatant were harvested by centrifugation at 12,000 g for 10 min at 4°C and suspended in 1 ml DB buffer, which was D medium without glucose and MgSO,. The cell suspension was

dnaR Function of prs Gene applied to a 30 ml linear 5% to 20% sucrose gradient in DB buffer and centrifuged at 3000 g for 15 min at 4°C. Minicells in the middle fraction of the gradient were collected, concentrated by centrifugation and subjected to a 2nd sucrose gradient centrifugation. Minicells were harvested and suspended in 61 ml D medium at a concentration of A,,, = 2.5. After incubation at 37°C for 30 min, cells were labeled with 025 &i [14C]leucine for 60 min. The cells were harvested, washed with @15 M-N&I and suspended in 20 ~1 of the sample buffer described by Lsemmli (1970). After treatment in a boiling water bath for 5 min, a 10 ~1 portion of the sample was subjected to SDS-polyacrylamide gel electrophoresis as described (Laemmli, 1970). The acrylsmide concentrations in the stacking and separation gels were 35% and 125 y. (w/v), respectively. Electrophoresis was done at 18 mA for 90 min in a slab gel (10 cm x 10 cm x 01 cm). The gel was stained with a Coomassie brilliant blue R solution and treated with Amplify (Amersham, Buckinghamshire, England). The gel was dried and exposed to an Rx film (Fuji) for 15 h at -70°C. Proteins used as molecular weight markers were phosphorylase b, bovine serum albumin, ovo-albumin, carbonic anhydrase and lysozyme, which were purchased from Pharmacia Fine Chemicals, Uppsala, Sweden.

3. Results (a) Isolation of the gene with dnaR

function

A chromosomal DNA fragment encoding an activity that reversed temperature-sensitive growth Of the dnaR mutant was cloned from BamHI-digested E. coli DNA on pBR322 plasmid. The cleavage pattern of the 2.8 kb DNA fragment with a set of restriction enzymes (Fig. l(a)) corresponded to that of the chromosome region near the 1275 kb position on the map derived by Kohara et al. (1987). The position was very close to the genetic locus of the dnaR mutation at 26.3 minutes (Sakakibara, 1992). From the isolated’ clone pSRl30, deletion derivatives were constructed to delimit the region with the dnaR activity (Fig. l(a)). The activity was retained in pSR132 plasmid which carried the 1.65 kb EcoRI-PvuII segment, while it

had been lost in pSR133 plasmid with a deletion of the 0.5 kb BamHI-Hind111 with a deletion of the

segment. The HindIII-AvaIII

segment and pSR134 1.7 kb AvaIII-BumHI

results indicate that the 0.6 kb segment was required for the dnaR

activity. The dnaR activity

of the BamHI DNA fragment was also observed with ID69 phage vector (Mizusawa & Ward, 1982). When the LdnaR+ phage that was integrated into the dn.uR mutant chromosome by homologous recombination was induced, about one-fourth of the progeny had lost the dnaR activity. Among the dnuR-defective phages obtained was IdnuRl30, which carried the same insert as that of parental dnaR+ phage with respect to restriction enzyme cleavage sites (Fig. l(b)). This result shows that the dnaR mutation existed in the cloned fragment. When the BamHI insert of the dn.aR-defective

phage

DNA, the resultant temperature-sensitive

was

recloned

into

pBR322

plasmid pSR301 could reverse growth of the dnaR mutant.

991

(a 1 Eumn* himd’dnaR+ I himd~dMRI30 pSR3OldnnRI30 psR3owuau+ psR304dnaR+

HhdlI t

AVdI I

Ecvnlu I

I

I

(b)

prs-A

Eo”H1 I 1

n/Mm I

AWIII

PWII 1

I

tij _-------________-_-

PRPP synlheme

5,*

&

1785

. 1459 cc 1

Figure 1. A chromosomal

DNA segment including the prs gene. (a) pSR130 plasmid and its deletion derivatives. A restriction enzyme cleavage map of the 2.8 kb bacterial BamHI DNA fragment carried by pSR130 (dnuR+) plasmid is presented. The subclones indicated carry the bacterial DNA fragments indicated by the open boxes. (b) Location of the dnaR mutation. ldnaR+ phage carries the BamHI fragment of pSR130 DNA, and ldnaRl30 the corresponding fragment from the dnaR mutant chromosome (Materials and Methods). pSR301 plasmid carries the BumHI insert of MnaRl30 phage. Recombinant plasmids pSR302, pSR303 and pSR304 carry, respectively, the BarnHI-HindIII, HindIII-AwaIII and AvaIII-BamHI segments derived from pSR301 (Materials and Methods). The mutant fragments are indicated by the shaded boxes. (c) Primary structure of the prs gene. The coding segments for prs mRNA and PRPP synthetase are adapted from the results of Hove-Jensen et al. (1986). Positions of the initiation and termination presented are numbered from the first G residue of the BamHI site.

The complementation probably resulted from an elevated dose of the mutant gene by the multicopy plasmid. To delimit further the region including the dnaR mutation, the BamHI-HindIII, HindIII-AvaIII and AwaIII-BamHI segments of the dnaR+ plasmid were replaced with the corresponding mutant segments (Fig. 1(b)) and the BumHI insert of each plasmid was recloned into 12D69 DNA. The recombinant phage with the mutant HindIII-A411 segment had lost the dnaR activity, whereas the BamHI-Hind111 and AvaIII-BamHI recombinant phages retained it. Loss of the activity by replacement of the segment required for the dnaR activity with the corresponding mutant segment indicates that the cloned gene was dnaR, not a suppressor. (b) Presence of the prs gene within the dnaR region The nucleotide sequence of the EcoRI-PvuII region with the dnaR activity was determined (data

992

Y. Sakakibara

b

68-

46-

30-

14.6-

Figure 2. The activity of PRPP synthetase specified by pSR132 plasmid. HC5103 (dnaRI30) cells with and without pSR132 (dnaR+) plasmid were grown in L broth at 30°C to late logarithmic phase in the presence and absence of ampicillin (40 pg/ml), respectively. PRPP synthetase activity in extracts from the cells was assayed at 30°C by measuring pyrophosphoryl transfer from [y-3’P]ATP to ribose-5-phosphate, as described in Materials and Methods. Radioautograms of the reaction products separated from ATP by polyethyleneimine-cellulose thin-layer chromatography are presented. Assay conditions were with the extract from HC5103 cells with pSR132 (lanes 1, 2) or without the plasmid (lanes 3, 4) in t,he absence (lanes 1. 3) and presence of ribose-5-phosphate (lanes 2. 4).

not shown). Through a computer search for homology with known gene sequences, it was found that the dnaR region included the entire nucleotide sequence of the prs gene (Fig. l(c)), which encodes PRPP synthetase (Hove-Jensen et al., 1986). In fact, the activity of PRPP synthetase in a cell extract from pSR132 (dnaR+) plasmid-carrying dnaR mutant was very high compared with that from the plasmid-free mutant (Fig. 2). This plasmid specified a protein of 34,000 M, in minicells, which was absent from pBR322-specified proteins (Fig. 3 lanes a and b). These results were consistent with those reported by Hove-Jensen (1985). The 34,000 M, product was confirmed to be PRPP synthetase by purification to near homogeneity according to the procedure of Switzer & Gibson (1978) (data not shown). Thus, the dnuR-coding region contained the prs gene.

Figure 3. Proteins specified by pSR132 plasmid and its GTT and ATC mutants. The plasmid-carrying minicells were labeled with [‘4C]leucine at 37°C for 60 min. and rellular proteins were separated by SDS-polyacrylamide gel electrophoresis. as described in Materials and Met,hods. Radioautograms presented are for proteins of the cells carrying pBR322 (a), pSR132 (b) and pSR132 mutant with the GTT (c) or ATC conversion (d) shown in Fig. 4. Positions of molecular weight markers are indicated at the left.

(c) Alteration

of PRPP synthetase by the dnaR ,mutation Because the dnaR mutation resided within the region encoding PRPP synthetase (Fig. I), the effect of the mut,ation on the enzyme activity was examined. The extract prepared from dnaR mutant cells grown at 30°C exhibited about, half of the activity in that from dnaR+ cells (Table 1). When the mutant extract was incubated at 42°C for 5 min before assay at 3O”C, the synthetase activity became undetectable, although the dnaR+ extract retained about half of the activity after the treatment at 42°C (Table 1). Treatment of the mutant culture at 42°C for 5 min before preparation of the cell extract also caused a great decrease in the enzyme activity, although the treatment of dnaR+ cells had no significant effect on the activity (Table 1). Thus, PRPP synthetase in cell extracts from the dnaR mutant was thermosensitive. The synthetase activity in the extract, from pSR132 (dnaR+) plasmid-carrying dnaR+ cells was about 30-fold higher than that from the plasmidfree cells (Table 1). A similar high activity of the enzyme was yielded by introduction of pSR130 (dnaR+) plasmid into dnaR mutant or dnaR+ cells. In contrast, the synthetase specified by pSR303 (dnaRI30) plasmid exhibited a very low activity and very low thermal stability, compared to that by pSR132 (dnaR+) plasmid (Table I). A similar result

dnaR Function of pm Gene

993

Table 1 Thermosensitivity

of PRPP synthetase of the dnaR

Synthetase activity (pmol/mg protein per 5 min)

Preheat Bacteria dnaR+ dnaR+ dnaR+ dnaR130 dnaR1.30 dnaRl.30 dnaRld0 dnaRl.30 dnaR130 dnaRl.30 dnaRl30 dnaR1.30

Plasmid

mh mh

in viva

-

+

-

+

-

-

dmR+ dnaR+ dnaRl30 dnaR130

in 1itro

-

(d) Assignment of dnaR

to the prs gene

PRPP synthetase consists of 314 amino acid residues and the gene has own transcription promoter and t’erminator (Hove-Jensen et aZ., 1986) (Fig. l(c)). The nucleotide sequence for the enzyme has been determined by Hove-Jensen et al. (1986). They have suggested that the initiation codon sequence is most likely the GTG adjacent to CCT encoding the amino-terminal proline residue of the enzyme among several possible initiation codon sequences present’ upstream from the CCT in the

491

SPhI

0.18 0.10 0.16 0.07 < 0.01 0.01 0.08