The priB Gene Encoding the Primosomal Replication n Protein of Escherichia coli*

The gene encoding protein n of the Escherichia coli primosome has been discovered in the rpsF-rpsR-rpll ribosomal protein operon and designatedpriB. The low copy number of PriB protein and the distinctive codon usage of its gene argue against its being a ribosomal protein. A strain which overproduces PriB was constructed and has been used to purify the protein to homogeneity. The overproduced protein behaves like that purified from wild-type cells. The replication of 4x174 single-stranded viral DNA to generate the duplex replicative can be reconstituted in vitro using Escherichia roles of replication biochemically. DnaB, DnaC, DnaG DnaT, SSB, and DNA polymerase I11 holoenzyme. of host proteins: PriA, PriB, PriC n’, n, and n”, functions proteins in phage replication been of their involvement in isolation gene-encoding and gene, cZ857, and the T7 RNA polymerase gene under control of the X PL promoter. Shifting the growth temperature from 30 to 42 “C derepresses the h PL pro- moter allowing expression of the T7 RNA polymerase that then transcribes the priB gene. Other Methods-SDS-PAGE and Coomassie Brilliant Blue stain-ing were performed as described (10). The protein concentration was determined by the Bradford method (11) with bovine serum albumin as a standard.

The replication of 4x174 single-stranded viral DNA to generate the duplex replicative form can be reconstituted in vitro using only Escherichia coli proteins (1). The roles of many of these proteins in both phage and host replication have been studied genetically as well as biochemically. These proteins include: DnaB, DnaC, DnaG (primase), DnaT, SSB, and DNA polymerase I11 holoenzyme. The replication of 9x174 requires three additional host proteins: PriA, PriB, PriC (previously known as n', n, and n", respectively). The functions of these remaining proteins in phage replication have been explored (1). However, the limited amount of material and the lack of identification of their genes have prevented their further characterization and studies of their involvement in replication of the host chromosome. The recent isolation of the priA gene-encoding protein n' (2, 3) has enabled the construction of a null priA mutant and provided insights into its role in replication (4).
In the 9 X SS' to RF reaction, PriA acts at an early stage of primosome assembly by binding a specific hairpin on E. coli single-stranded DNA binding protein-coated 9X ssDNA. This structure is then recognized and bound by proteins PriB and PriC. Formation of the primosome proceeds with the subsequent actions of DnaB, DnaC, DnaT, and primase (5). As a mobile complex, the primosome lays down the RNA primers that DNA polymerase I11 can elongate.
We report here the discovery of the priB gene that encodes the primosomal protein PriB. Identification of the gene has enabled us to the overproduce and purify PriB to homogeneity.
* This work was supported by National Institutes of Health Grant GM07851 and National Science Foundation Grant DMB87-1007945. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. j: Supported by a predoctoral fellowship from the Howard Hughes Medical Institute.
Replication Assay for PriB Protein-The reconstituted $X ss to RF replicative assay was essentially as previously described (7) with the addition of potassium glutamate to 20 mM. One unit of replication at 30 "C. activity promotes the incorporation of 1 pmol of nucleotide in 1 min Identification of the priB Gene-The PriB protein was purified to near homogeneity through the DNA-cellulose step by Dr. Philippe Tacchini (University of Geneva) as described (8). The N-terminal amino acid sequence determined by Allan Smith (Stanford University) was NH,XNRLVLSGTVSRAPLRKVVPXGI. This protein sequence was used to search the Swiss Protein data bank using the programs of IntelliGenetics (9).
Construction of an Overexpression Vector-A plasmid, pRA47, expressing the priB gene from the T7 610 promoter and under translational control of the $10 Shine-Dalgarno was constructed analogously to pRA45, as described (2). Briefly, the primary construction consisted of the 571-base pair Sau3Al fragment from the rpsF-priB-rpsR-rpll operon (nucleotides 657-1228 containing the priB ORF) subcloned into the BamHI site of pT7-6. Site-directed mutagenesis was used to create an NdeI site at the priB start codon into which the $10 Shine-Dalgarno cassette was inserted. The plasmid, pGP1-2, contains the temperature-sensitive h repressor gene, cZ857, and the T7 RNA polymerase gene under control of the X PL promoter. Shifting the growth temperature from 30 to 42 "C derepresses the h PL promoter allowing expression of the T7 RNA polymerase that then transcribes the priB gene.
Other Methods-SDS-PAGE and Coomassie Brilliant Blue staining were performed as described (10). The protein concentration was determined by the Bradford method (11) with bovine serum albumin as a standard.

RESULTS
Identification of thepriB Gene-The N-terminal amino acid sequence derived from PriB protein was used to search the Swiss Protein data bank which contains potential proteins derived from ORFs as well as known protein sequences. This search identified the second ORF in the ribosomal operon containing rpsF (S6), rpsR (SlS), and rplI (L9) ( (Table I).
Although located in a ribosomal protein operon, PriB is 10Tbp FIG. 1. The operon containing the priB gene. The putative promoter ( P ) , the ORFs (indicated by boxes), and the putative transcription terminator ( t ) were derived from the nucleotide sequence as described (12). The previously unidentified open reading frame has been designated priB. bp, base pairs.

TABLE I
Overproduction of PriB activity Crude Fraction I1 was prepared and assayed in @X ss to RF reconstitution reaction. The host strain was K38 harboring pGP1-2 and either pT7-6 or its derivative (pRA47). Cells were grown a t 30 "C as described (6) and induced by adding an equal volume of L broth Dreheated to 54 "C and then were erown further a t 37 "C for 2 h.

Plasmid
Insert PriB activity ( X 10-4) Overproduction pT7-6 D R A~~ oriB with T 7 dl0 SD 2830 1180 unlike ribosomal proteins in codon usage and abundance. Ribosomal protein genes exhibit a distinct codon bias for certain amino acids. The ribosomal protein genes which flank priB on either side (Fig. 1) share this codon usage pattern (12) whereas that of priB is clearly different ( Table 11). The abundance of PriB protein at 80 copies/cell (8) is comparable with that of many of the replication proteins rather than that of ribosomal proteins which number 1,000 to 10,000 copies/ cell.
Purification and Characterization of PriB-PriB was purified from an overproducing strain using the $X ss to RF reconstituted replication assay. The purification from the soluble lysate was 11-fold with an overall yield of 71% (Table  111). The final product was homogeneous as judged by SDS-PAGE (Fig. 2).
The PriB protein isolated from the overproducing strain was characterized to verify that it behaves like PriB derived from wild-type cells (8). The molecular mass calculated from the gene sequence predicts an 11.4-kDa protein. The overproduced PriB has an apparent mass of 12.2 kDa on SDS-PAGE (Fig. 2) as reported for the previously purified protein (8).

Purification of PriB
All operations were carried out a t 0-4 "C unless otherwise noted. E. coli K38 (pGP1-2, pRA47) was grown in L broth containing 25 Fg/ml ampicillin and 30 pg/ml kanamycin at 30 "C to A,,,, of 0.8. The cells were then induced by shifting to 42 "C for 15 min, then grown further for 2 h a t 37 "C, harvested, resuspended in buffer A to A,," of 550, and frozen immediately in liquid nitrogen. A heat lysozyme lysis was carried out starting with 740 g of frozen cells. Dithiothreitol (1 M ) was added to 10 mM, 0.5 M EDTA (pH 8.0) to 10 mM, 0.8 M spermidine HC1 (pH 7.5) to 20 mM, 5 M NaCl to 100 mM, and ammonium sulfate (saturated a t 4 "C) to 5%. The remainder of the heat lysis was as described (7) yielding Fraction I (670 ml). Ammonium sulfate (0.225 g/ml fraction I) was added the precipitate was collected and resuspended in buffer R + 100 mM NaCl (Fraction 11, 68 ml). Fraction I1 was diluted to the conductivity of buffer R + 100 mM NaCl and applied to a 600-ml Bio-Rex 70 column (7.5 X 13.6 cm) equilibrated in buffer R + 100 mM NaCl at a rate of 1 column volume/h. The column was washed with 3 column volumes of buffer R + 100 mM NaCl and eluted with a 6-column volume linear gradient of buffer R + 100 mM to 1 M NaCI. Peak fractions were pooled (Fraction 111, 990 ml). For gel filtration chromatography, 0.08% of fraction 111 was clarified in a Beckman TLA1OO.l (15 min, 95,000 rpm) and applied to a 25-ml Superose-12 fast protein liquid chromatography column (HR 10/30) equilibrated in buffer R + 460 mM NaC1. The column was run at 0.2 ml/min, and 0.3-ml fractions were collected. Peak fractions were pooled ( The priB Gene Encoding Protein n requirement for approximately 1 dimer per ss @X circle replicated assuming that synthesis is processive once initiated (Fig. 4). Also as anticipated, the purified overproduced PriB is heat-stable (65% activity remains after 5 min at 100 "C) and is N-ethylmaleimide-sensitive (<3% activity remained after a 10-min treatment at 30 "C with 10 mM N-ethylmaleimide).

DISCUSSION
The priB gene is found in an operon flanked by ribosomal protein genes: upstream by rpsF and downstream by rpsR and rplI. Sequence analysis of this region identified a potential promoter and transcription terminator, and insertional mutagenesis indicated that these genes form an operon (12). The priB gene differs from its neighboring, ribosomal protein genes in two respects: 1) a 100-fold lower cellular abundance of its protein product and 2) distinctive codon preferences.

-
Control at the level of translational efficiency may be responsible for the low protein copy number. The clustering of macromolecular synthesis genes for replication, transcription, and translation into operons has been reported in three other instances in E. coli: the rpsU-dnaG-rpoD operon that encodes ribosomal protein S21, primase, and the a ' " subunit of RNA polymerase (13); the rplK-rplA-rpW-rplL-rpoB-rpoC operon encoding ribosomal proteins, L11, L1, L10, and L7/L12 and the / 3 and p' subunits of RNA polymerase (14); and the rpsM-rpsK-rpsD-rpoA-rplQ operon that encodes ribosomal proteins, S13, S11, S4, and L17, and the a subunit of RNA polymerase (15). The benefits of organizing these genes as operons are still unclear particularly in situations where the final levels of the proteins encoded in the operon vary greatly. The possible advantage of linking the biosynthesis of RNA, DNA, and proteins by proximity of key genes deserves further attention.
The location of the priB gene within the rpsF-rpsR-rpll ribosomal protein operon suggests that it plays a critical role for E. coli. The other gene products encoded by these four operons are centrally involved in RNA, DNA, and protein metabolism. The particular aspect of these processes that priB contributes to is still unclear. With the identification of the priA and priB genes, priC is the only gene among those  Table 111.
as described under "Materials and Methods." encoding primosomal proteins that remains at large. With these genes in hand, the overproduction of the gene products will provide access to biochemical studies of primosomal assembly, structure, and function. Mutation as well as regulated usage of each of the genes will afford the means to elucidate their individual and collective roles in replication of the host chromosome, plasmids, and phages.