INTRODUCTION

Staphylococcus aureus is a common human and animal pathogen (1, 2). The pathogenesis of S. aureus is primarily the result of secretion of a large number of extracellular and cell wall-associated proteins that facilitate the colonization, multiplication, and spread of the bacterium (3, 4). Three global elements, the accessory gene regulator (agr) (5, 6, 7, 8), the exoprotein regulator (xpr) (9, 10), and the Staphylococcal accessory regulator (sar) (11, 12, 13), have recently been implicated in the temporal and coordinate expression of these exoproteins. Genetic studies have revealed that the regulation of exoprotein gene expression by these global elements is at the level of transcription. To gain further insights into the complicated mechanism of transcriptional regulation, it is necessary to study transcription in vitro. In all bacterial systems studied, the RNA polymerase (RNAP)¹ core enzyme is composed of four subunits: two identical  subunits, , and . The association of  to the core allows the RNAP holoenzyme to recognize promoter elements and initiate transcription from specific sites (14, 15). The availability of RNA polymerase and different  factors from Escherichia coli (16, 17, 18) and Bacillus subtilis (19, 20, 21, 22, 23) have led to a very advanced level of understanding of gene regulation in these organisms. Biochemical studies on transcriptional regulation in S. aureus have not progressed to a comparable degree due to the lack of a defined in vitro transcription system. Only recently, RNAP from S. aureus has been purified and used in in vitro transcription studies (24, 25). The RNAP purified from exponentially growing S. aureus cells contains only small amounts of  subunit and is poorly active in transcription reactions. An overexpression system for obtaining large quantities of the  factor from S. aureus will be very helpful in elucidating the detailed mechanism of the regulation of gene expression in this organism.

Recently, a subunit of the purified S. aureus RNAP was identified as the putative  factor based on its ability to cross-react with the B. subtilis anti-^A antibody (24) and the E. coli anti-⁷⁰ antibody (25). The purified protein conferred on the E. coli core RNAP the ability to initiate authentic transcription from the sea promoter of S. aureus (25). No further characterization of this protein (`` factor'') has been reported as yet. It should be noted that the purification of this protein from exponentially growing S. aureus cells is labor-intensive, and the yield is very low. Interestingly, the chromosomally encoded plaC gene was predicted to encode the primary  factor in S. aureus based on the amino acid sequence identity of the plaC gene product with that of the B. subtilis vegetative  factor, ^A (26). In this paper, we report the overproduction and purification of the plaC-encoded gene product and demonstrate that the overproduced protein is the vegetative  factor.

MATERIALS AND METHODS

A list of plasmids used is given in Table I. The E. coli strain BL21(DE3)pLysE and the vector pET-24a(+) used for cloning and overproduction of the plaC gene were from Novagen, Inc.

Table I. Plasmids used and their characteristics Plasmid Promoter Reference pUC19spf ColE1 RNA-1 27 pMJB1168 sea 25 pMJB1166 hla 25 pSA2727 plaC 26 pRD101 Overexpression plasmid for SA This study pJET40 dnaK P1 28 pJET41 rpoH P1, P3, and P4 27 pMJB1164 agr P2 25 pMJB1167 sec 25

Lysostaphin was obtained from Bristol Myers Squibb Company. Restriction enzymes were from Life Technologies, Inc. and New England Biolabs. E. coli ⁷⁰ holoenzyme was purchased from Pharmacia Biotech Inc., and the E. coli RNA polymerase core enzyme was from Epicenter Technologies. Radioactive nucleotides were from either ICN Biomedicals or from Amersham Life Sciences.

The RNAP enzyme was purified from the S. aureus strain RN4220 by a modification of published protocols (24, 25). The cells were grown in nutrient broth containing 2% casein enzymatic hydrolysate (Sigma) and 1% yeast extract (Difco). Cells were harvested at A₆₀₀ of 1.0 by centrifugation (8,000 rpm in a Sorvall GSA rotor) and washed, first with buffer A (10 mM Tris-HCl (pH 7.9), 10 mM MgCl₂, 1.0 M NaCl, 5 mM EDTA (pH 8.0), 0.2 mM DTT, and 5% glycerol) and then with the grinding buffer (29). Cells (25 g, wet weight) were resuspended in 50 ml of grinding buffer containing 5 mg of lysostaphin and incubated at room temperature for 30 min. The cells were then lysed by passing through a French pressure cell at 14,000-16,000 p.s.i. The cell debris was removed by centrifugation (14,000 rpm SS-34 rotor), and polyethyleneimine (Sigma) was added to 0.6% (v/v) to precipitate nucleic acids and bound proteins. Proteins weakly bound to nucleic acids were removed by washing with TGED (10 mM Tris-HCl (pH 7.9), 5% glycerol, 0.1 mM EDTA, and 0.2 mM DTT) plus 0.45 M NaCl. RNAP was eluted from nucleic acids with TGED, 1.0 M NaCl. After ammonium sulfate precipitation (35 g/100 ml) the proteins were resuspended in TGED, 0.1 M NaCl and applied to a heparin-agarose column (Sigma) preequilibrated with TGED, 0.1 M NaCl. The column was washed with TGED, 0.2 M NaCl, and the bound proteins were eluted off the column by applying a linear gradient from 0.2 to 1 M NaCl. The fractions containing RNAP subunits were pooled, concentrated by using Centriplus-10 concentrators (Amicon Inc.), and loaded onto a Sephacryl-300 column (Sigma), preequilibrated with TGED, 0.4 M NaCl. Fractions containing transcriptional activity were pooled, diluted to a conductivity of TGED-0.2 M NaCl, and passed through a Q-Sepharose Fast Flow anion exchange column (Sigma). The column was washed with TGED-0.2 M NaCl, and the bound proteins were eluted by applying a linear gradient from 0.2 M NaCl to 1.0 M NaCl.

Construction of the Plasmid pRD101 for Overproduction of ^SA

A DNA fragment containing the plaC gene was generated by PCR using the plasmid pSA2727 (26) as template. The primers used for amplification are plac1 (5-CCGGAATTCTAAAAGGAGCCGTTTCATGTCTGATAAC-3) and plac2 (5-CCCAAGCTTCAATGTACCAACCTCACTC-3). Amplification by using these primers generates EcoRI and HindIII sites at the termini of the amplified DNA. Amplification was carried out using recombinant Pfu DNA polymerase (Stratagene, CA). Reaction conditions were melting temperature, 96 °C (1 min); annealing temperature, 50 °C (1 min); elongation temperature, 72 °C (2 min). The amplified product was isolated from an agarose gel, digested with EcoRI and HindIII, and cloned into the EcoRI and HindIII sites of pET-24a(+), resulting in pRD101. The nucleotide sequence of the cloned plaC gene in pRD101 was confirmed by double-stranded DNA sequence analysis with Sequenase (U.S. Biochemical Corp.).

Overproduction and Purification of ^SA

E. coli BL21(DE3)pLysE cells containing the plasmid pRD101 were grown in 2XTY (16 g of Bacto tryptone, 10 g of yeast extract, and 5 g/liter NaCl plus 0.4% glucose) in the presence of 30 µg/ml kanamycin and 20 µg/ml chloramphenicol at 37 °C. The cells were grown to an A₆₀₀ of 0.6 and induced by isopropyl--D-thiogalactopyranoside for 2 h. The cells were harvested by centrifugation and used immediately for purification of ^SA. The method of purification was modified from that used to purify B. subtilis ^A factor (30). Cells from 2 liters of culture were suspended in 20 ml of cell disruption buffer (50 mM Tris-HCl (pH 7.9), 2 mM EDTA (pH 8.0), 0.2 mM DTT, and 0.05% sodium deoxycholate), incubated on ice for 20 min, and lysed by passing through a French pressure cell twice at 14,000-16,000 p.s.i. The lysate was centrifuged at 14,000 rpm in a Sorvall SS34 rotor at 4 °C. After washing with TGED buffer containing 0.25 M NaCl, the cell pellet was resuspended in 20 ml of TGED buffer containing 6 M guanidine hydrochloride, incubated on ice for 10 min, and centrifuged at 15,000 rpm for 30 min at 4 °C. The supernatant was diluted gradually to 1,200 ml with cold TGED buffer. The protein suspension was centrifuged at 7,500 rpm in a GSA rotor at 4 °C. The supernatant was further diluted to 2,400 ml with cold TGED buffer. DEAE-Sephacel resin (25 ml) (Pharmacia Biotech Inc.) preequilibrated with TGED buffer containing 0.1 M NaCl was added to the protein suspension, and the suspension was stirred slowly for 1 h at 4 °C. The suspension was packed in a column. The column was washed with TGED buffer containing 0.1 M NaCl, and the proteins were eluted by applying a step gradient from 0.2 to 0.5 M NaCl in increments of 0.05 M. The fractions containing the overproduced protein were pooled and precipitated by adding ammonium sulfate (42 g/100 ml). The protein pellet was resuspended in 0.5 ml of TGED buffer containing 0.25 M NaCl and chromatographed on a Sephadex G-75 column (Sigma) equilibrated with the same buffer. The fractions containing the overproduced protein were pooled, concentrated by using Centriplus-10 concentrators, dialyzed against storage buffer (20 mM Tris-HCl (pH 7.9), 10 mM MgCl₂, 1 mM EDTA, 0.1 mM DTT, 1 mM phenylmethylsulfonyl fluoride, 50% glycerol), and stored in small aliquots at 20 °C. The amount of protein was quantitated by a Bio-Rad protein determination kit using bovine serum albumin as standard. Protein samples were electrophoresed on 12% SDS-polyacrylamide gels and visualized by staining with Coomassie Blue R-250.

Purification of the S. aureus  Factor from SDS-Polyacrylamide Gels

The ^SA protein was eluted from a 12% SDS-polyacrylamide gel as described by Hager and Burgess (31) except that we electroeluted the protein instead of eluting it by diffusion. The subsequent steps for SDS removal and renaturation of the ^SA protein were the same as described (31).

The proteins separated on SDS-polyacrylamide gels were transferred to a nitrocellulose membrane (Bio-Rad). The ECL Western blotting analysis system (Amersham Corp.) was used for detection of proteins on the membrane. The primary antibody (B. subtilis anti-^A polyclonal antibody) was used at a 1:3,000 dilution.

The purified RNAP enzyme from S. aureus or the E. coli core enzyme with or without ^SA was incubated in a microcentrifuge tube for 20 min on ice. To this mixture, 40 ng of plasmid DNA was added and incubated at 35 °C for 10 min. A mixture containing 40 mM Tris acetate (pH 7.9), 100 mM NaCl, 5 mM MgCl₂, 0.2 mM DTT, 100 µg/ml bovine serum albumin, 0.25 mM each of ATP, CTP, and GTP, 0.015 mM UTP, 10 µCi of [-³²P]UTP, 50 µg/ml heparin, and 0.5 units of Prime RNase Inhibitor (5 Prime  3 Prime, Inc., Boulder, CO) was used in each reaction and incubated at 35 °C for 10 min. The reactions were terminated by the addition of 100 µl of a stop solution (0.4 M ammonium acetate, 20 mM EDTA, 0.3% (w/v) SDS, and 4 µg of tRNA) and precipitated with 300 µl of 100% ethanol for 1 h at 20 °C. After centrifugation, the pellet was washed with 70% ethanol, resuspended in sequencing gel loading buffer containing 98% deionized formamide, 10 mM EDTA, 0.025% xylene cyanol FF, and 0.025% bromphenol blue. The samples were incubated in boiling water for 5 min and electrophoresed in a 6% polyacrylamide gel containing 7 M urea.

RNA used for mapping the 5-end was prepared as described above, but the reactants were scaled up to 10-fold. Heparin was excluded from the reaction mixture. The concentration of each of the nucleoside triphosphates used was 0.5 mM, and none was radiolabeled. The transcription reactions were terminated by adding 5 units of RNase-free DNase I (Boehringer Mannheim) followed by incubation at 35 °C for 10 min. Then the samples were heated at 90 °C for 5 min to inactivate DNase I and extracted twice with phenol-chloroform and once with chloroform. RNA was ethanol-precipitated, washed with 70% ethanol, and resuspended in nuclease-free water.

For primer extension reactions, synthetic oligonucleotide primers (25) were end-labeled with [-³²P]ATP using T4 polynucleotide kinase (Promega Corp.). Reactions were carried out at 42 °C for 45 min with 200 units of Moloney murine leukemia virus reverse transcriptase (Promega) following the manufacturer's protocol. DNA sequencing was performed using the same primers except that none was radiolabeled. The primer extension products were run on a 7.5% polyacrylamide-urea sequencing gel in parallel with sequencing reactions.

The NH₂-terminal amino acid sequencing of the overproduced protein was performed by the method of Matsudaira (32). The proteins were separated on a 12% SDS-polyacrylamide gel and blotted to a Westran polyvinyldiene difluoride membrane (Schleicher & Schuell). The membrane was briefly stained with Coomassie Blue and destained. The band corresponding to the overproduced protein was excised, and the NH₂ terminus of the protein was sequenced at the Genetic Research Facility of the University of Illinois at Urbana-Champaign using a 477A NH₂-terminal protein sequencer (Applied Biosystems).

RESULTS

Overexpression and Purification of the  Factor Encoded by plaC

The predicted protein product of the S. aureus chromosomal gene plaC has 78.8% amino acid sequence identity with the B. subtilis SigA, the vegetative  factor (26). To characterize whether the plaC gene product functions as a  factor in S. aureus, we attempted to clone plaC under the control of the T7 promoter, as was done in the case of the B. subtilis sigA gene (30). Unlike the cloning of B. subtilis sigA, when the S. aureus plaC DNA was ligated with the appropriate vector for expression from the T7 promoter and was transformed into the E. coli strain BL21(DE3), we were unable to obtain any stable transformant. We have successfully used the E. coli strain BL21(DE3)pLysE (BL21(DE3)pLysS was not suitable) for overexpression of plaC (Fig. 1). Substantial levels of T7 lysozyme accumulate in this strain. Any T7 RNAP produced from the ``leaky'' lacUV5 promoter is readily inactivated by the lysozyme in the uninduced state, thus preventing the expression of the cloned gene product that might be toxic for E. coli growth. In our case apparently, the plaC gene product is toxic for E. coli growth. The presence of resident lysozyme also makes it much easier to lyse the cells for the preparation of the cell extracts. The overproduced plaC gene product obtained from isopropyl--D-thiogalactopyranoside-induced cell lysate was mostly present in inclusion bodies and migrated on a polyacrylamide gel at a mobility corresponding to 55 kDa (Fig. 1, lane 2). The proteins present in the inclusion bodies were solubilized with 6 M guanidine hydrochloride and chromatographed successively on DEAE Sephacel and G-75 Sephadex columns (Fig. 1, lanes 3 and 4). At this stage of purification, the protein preparation was about 85% pure (Fig. 1, lane 4). Approximately 3 mg of protein was obtained from 2 liters of cell culture. However, some contaminants were present after chromatography on Sephadex G-75 gel filtration. The overproduced protein was purified to near homogeneity from a preparative SDS-polyacrylamide gel (Fig. 1, lane 5). The experimentally determined N-terminal sequence of the overproduced protein (XSDNTVKIKKQ) matched with the predicted amino acid sequence of the plaC gene product, except that it lacked the first amino acid, methionine.

Western blot analyses revealed that the overproduced protein cross-reacted with a polyclonal antibody raised against B. subtilis ^A (Fig. 2, lane 2). The cross- reactive band co-migrated with ^A of B. subtilis (Fig. 2, compare lanes 2 and 3). We previously identified a 55-kDa subunit of the purified S. aureus RNAP as a candidate for the vegetative  factor based on its cross-reactivity with anti-^A antibody (24). Western blot analysis showed that the overproduced protein and the 55-kDa subunit of the purified S. aureus RNAP had the same antigenicity, and they co-migrated (Fig. 2, lanes 1 and 2). It is also apparent from the Western blot analyses that the purified S. aureus RNAP contains negligible amounts of the 55-kDa subunit (Fig. 2, lanes 2 and 3). Although the amount of purified S. aureus RNAP used in lane 2 is 25-fold higher than the amount of B. subtilis RNAP used in lane 3, the intensity of the 55-kDa band in lane 2 is almost 5-fold less than the co-migrating ^A protein band in lane 3.

To investigate the ability of the plaC gene product to act as a  factor, we reconstituted RNAP holoenzyme by adding the purified protein to the core RNAP from E. coli (henceforth referred to as heterologous RNAP) and assayed for transcriptional activity from different promoters present on supercoiled plasmids. The results of in vitro transcription are shown in Fig. 3. Three separate experiments gave similar results. Note that all of the transcription reactions were performed in the presence of heparin, unless indicated otherwise, to prevent reinitiation of transcription. The purified plaC gene product confers on the core RNAP the ability to initiate specific transcription from several S. aureus promoters. The transcripts produced from the three S. aureus promoters sea, sec, and hla were of the same size as previously reported using the  factor purified from exponentially growing S. aureus cells (25). Neither the core RNAP alone nor the purified protein was able to initiate any specific transcript from any of the promoters tested. The sizes of the transcripts synthesized by the E. coli E⁷⁰ from the sea and hla promoters are identical to those produced by the heterologous RNAP (Fig. 3, compare lane 8 with lane 9, and lane 14 with lane 15). Note that unlike the heterologous RNAP, the E. coli E⁷⁰ failed to produce a specific transcript from the sec promoter (Fig. 3, compare lanes 11 and 12). Based on the above results, the plaC- encoded gene product is designated as ^SA (S. aureus-derived  factor).

with or without

E

We also tested the ability of the purified ^SA protein to initiate specific transcription from several E. coli promoters. The plasmid pUC19spf, used to clone the different promoters and their flanking DNA elements, contains the ⁷⁰-dependent ColE1 RNA-1 promoter (27). Both the E. coli E⁷⁰ and the heterologous RNAP produced the expected 102- and 108-nucleotide transcripts from the RNA-1 promoter (Fig. 3) (27). Transcription was also detected from the ⁷⁰-dependent rpoH P1 (428-nucleotide) and rpoH P4 (290-nucleotide) promoters (33) present on the plasmid pJET41 (Fig. 3, lanes 2 and 3). Note that the E. coli E⁷⁰ RNAP transcribed from the rpoH P1 promoter more efficiently than the heterologous RNAP, but the transcription from the rpoH P4 promoter was equally efficient for the E⁷⁰ RNAP or the heterologous RNAP. Neither the heterologous RNAP nor the E. coli E⁷⁰ produced any transcript from the E. coli ^E-dependent rpoH P3 promoter (295 nucleotides) (27, 33), present on the pJET41 plasmid (Fig. 3, lanes 2 and 3). We also did not detect any transcript from the E. coli ³²-dependent dnaK P1 promoter (28) (data not shown).

Concentration-dependent Stimulation of Transcriptional Activity of the S. aureus RNA Polymerase by ^SA

While the specific initiation of transcription from several S. aureus promoters by the overproduced protein establishes that it is a  factor, the RNAP core enzyme used in the above transcriptional assays was from E. coli, a Gram-negative bacterium. For biologically relevant studies, it is desirable to evaluate the role of the  factor in the homologous RNAP holoenzyme (all of the RNAP subunits derived from S. aureus). Consistent with a report from another laboratory (25), the RNAP that we prepared from S. aureus contains trace amounts of the  subunit and is poorly active in in vitro transcription reactions. We chose two S. aureus promoters, sea and agr P2, to study the effect of ^SA on the purified RNAP from S. aureus. As shown in Fig. 4, a high concentration of the purified RNAP was required for detection of transcripts from the sea promoter. However, the addition of ^SA stimulated transcription from the sea promoter by about 72-fold (Fig. 4, compare lanes 5 and 10) and that from the agr P2 promoter by about 30-fold (data not shown).

RNAP and

Having the above results, we titrated the concentration of ^SA required to stimulate optimal transcriptional activity of our purified RNAP preparation from S. aureus. The concentrations of purified RNAP used for these assays did not result in any detectable transcriptional activity from both the sea and agr P2 promoters in the absence of any added ^SA (Figs. 5A, lane 1, and data not shown), and the trace amounts of ^SA present in the purified RNAP are ignored for the analyses of data presented below. There is a linear relationship between the increase in transcriptional activity and the increasing concentrations of ^SA added. The peak transcriptional activity was reached at a ^SA:core molar ratio of approximately 7 ± 1, for both the sea and agr P2 promoters (Figs. 5B and data not shown).

RNAP by

The 5-ends of the ^SA-initiated mRNAs synthesized in vitro from the sea and agr P2 promoters were mapped to compare in vitro and in vivo transcription start sites. There are three putative promoter elements, P2a, P2b and P2c, within the cloned agr P2 promoter region (Fig. 6A). Five major primer extension products were obtained from the mRNA synthesized in vitro using the plasmid DNA containing the agr P2 region as the DNA template. Three of these products might have resulted from the mRNA synthesized from the P2a promoter. We do not know whether this is due to ``stuttering'' of RNAP in vivo or a similar effect of the reverse transcriptase in vitro. Reported results of 5-end mapping of in vivo synthesized RNA are somewhat ambiguous. S1 nuclease data suggested one transcript from the agr P2 promoter corresponding to the topmost band indicated in the P2a region (Fig. 6A) (5, 34). Previously reported primer extension data suggested one transcript corresponding to the smallest RNA product in the P2a region (Fig. 6A) and another transcript corresponding to that indicated by P2c (25). The primer extension product P2b (Fig. 6A), obtained using in vitro synthesized RNA as the template, is in agreement with that reported previously (25). Since the P2b product is not detected in vivo, it might be an artifact of the in vitro experiment, or transcription from this promoter is normally down-regulated or repressed in vivo.

Primer extension with in vitro produced mRNA from the sea promoter revealed that the transcription start site corresponded to a guanosine nucleotide (Fig. 6B), which is one nucleotide upstream of the start site observed with the in vivo synthesized mRNA (35). The above results confirm that the ^SA protein directs the S. aureus RNAP to initiate specific transcription from the cognate promoters.

DISCUSSION

We have cloned the gene encoding the putative vegetative  factor of S. aureus and have demonstrated that the cloned gene product can functionally replace the  factor isolated from exponentially growing S. aureus cells. This is the first  factor from S. aureus that has been overexpressed and successfully used for in vitro transcription studies. Note that the S. aureus RNAP isolated from exponentially growing cells is virtually a core enzyme, nearly devoid of the  factor and poorly active in transcription assays (Figs. 2 and 4A). The purification of the  factor from E. coli cells, with the cloned gene, is very efficient and far less labor intensive as compared with that from S. aureus cells (Fig. 1) (25).

We confirmed that the purified protein is encoded by plaC by matching its amino-terminal 10-amino acid sequence with that of the predicted protein. The predicted plaC gene product has a molecular weight of 42,177, which is close to the calculated molecular weight of 42,957 of the B. subtilis sigA gene product. Note that  factors migrate abnormally on SDS-polyacrylamide gels because of highly positive and negative charge clusters (36, 37). Both the plaC gene product ^SA and the B. subtilis ^A migrate in gels as 55-kDa proteins (Fig. 2) (30). The common antigenic nature of the purified protein and the ^A subunit of the purified B. subtilis RNAP and their co-migration in a polyacrylamide gel support the authenticity of the purified plaC gene product.

Several different criteria establish that the overproduced plaC gene product acts as a  factor, designated as ^SA. The overproduced ^SA enables the core RNAP from E. coli to initiate promoter-specific transcription (Fig. 3). Note that although the transcription specificities of E⁷⁰ and E^SA RNAPs were similar for several S. aureus and E. coli promoters, there were both qualitative and quantitative differences in activities between the two forms of the RNAPs. The quantitative difference is evident from the 428-nucleotide rpoH P1 promoter transcript (Fig. 3). The qualitative difference is obvious from the fact that E⁷⁰ did not transcribe from the sec promoter and that E^SA gave a transcript corresponding to the in vivo synthesized mRNA (Fig. 3). The activity of the purified S. aureus RNAP enzyme was enhanced by almost 70-fold (sea promoter, Fig. 4) or 30-fold (agr P2 promoter, data not shown) by the addition of ^SA, and the activity increased in a concentration-dependent fashion with the addition of ^SA (Fig. 5B). The primer extension results with the RNA synthesized in vitro, using S. aureus RNAP supplemented with ^SA, are in general agreement with previously reported data on 5-end mapping of in vivo synthesized RNA (see ``Analyses of the Transcriptional Start Sites of the sea and agr P2 Promoters'' under ``Results''). All of these results taken together convincingly demonstrate that the ^SA protein acts as a  factor in S. aureus.

What is known about the functional role of plaC in cells? A point mutation in plaC, plaC1 in the S. aureus chromosome, resulted in an increase of the copy number of the plasmid pT181 by partially derepressing the synthesis of its counter transcript RNA (38). Note that the plaC1 mutation results in a change of proline to serine (209th amino acid residue from the NH₂ terminus) in the conserved 10 recognition element of ⁷⁰ class of proteins (26). The plaC1 mutation also resulted in the decrease in agr P2 promoter activity. However, several other promoter activities tested under plaC1 background were unaffected. The plaC1 mutation did not affect the growth of cells (26). Thus, the plaC1 mutation most likely impairs transcription from a limited number of promoters that direct the expression of genes, whose products are not essential for cell growth.

A GenBank^TM data base search using the BLASTP program (39) revealed that ^SA has significant homology with putative primary  factors of several Gram-positive organisms (Table II). In Gram-negative bacteria, the gene encoding the primary  factor is a part of a three-gene cluster in an operon consisting of rpsU (encoding ribosomal protein S21), dnaG (encoding DNA primase), and rpoD (encoding the principal  factor) (49). In the Gram-positive bacterium B. subtilis this operon apparently lacks rpsU but maintains dnaG (49). In contrast, the open reading frames preceding the S. aureus plaC gene do not have any similarity with dnaG or rpsU (26). The principal  factor genes of some other Gram-positive bacteria have recently been found to have a similar organization as the plaC gene of S. aureus. The hrdB genes of S. coelicolor and S. griseus and the sigA gene of B. lactofermentum are not associated with either dnaG or rpsU (45, 47, 50).

Table II. Homology of SA with primary  factors of other Gram-positive bacteria Organism Protein No. of amino acids Identitya Reference % B. subtilis SigA 371 79 36 Listeria monocytogenes LmRpoD 374 78 40 Enterococcus faecalis SigA 368 70 41 Clostridium acetobutylicum SigA 378 67 42 Lactococcus lactis L1RpoD 340 66 43 Streptomyces aureofaciens HrDB 525 53 44 Streptomyces coelicolor HrDB 442 53 45 Streptomyces griseus HrdB 445 52 46 Brevibacterium lactofermentum SigA 497 37 47 a  The sequence alignments were performed by using the CLUSTAL W program (48) with default settings.

Of all Gram-positive organisms, the vegetative  factor from B. subtilis has been well characterized. Besides the strong similarity of the ^SA protein with the vegetative  factor of B. subtilis, several lines of evidence suggest that ^SA is the vegetative  factor of S. aureus. To identify genes for the principal  factors of various eubacteria, Tanaka et al. (51) performed DNA hybridization analysis with a synthetic oligonucleotide probe corresponding to the highly conserved amino acid sequence present within the ``RpoD box'' of primary  factors (15). They were able to identify only one signal corresponding to a single gene for S. aureus. Consistent with the above results, Basheer and Iordanescu (26) detected only one homology region corresponding to the plaC sequence in the S. aureus chromosome by Southern analysis. A construct expressing B. subtilis sigA was found to complement in S. aureus the plaC1 mutation, proving functional relatedness of the two  factors.² The Staphylococcal promoters recognized by ^SA, e.g. sea, sec, and hla, have high degree of sequence similarity with that of the consensus sequence recognized by E. coli ⁷⁰ and B. subtilis ^A. ^SA also recognizes the E. coli ⁷⁰-dependent rpoH P1, rpoH P4 and ColE1 RNA-1 promoters but not the E. coli ³² and ^E-dependent promoters. Finally, the disruption of the S. aureus chromosomal plaC gene by insertion of a tetracycline-resistant cassette (52) was obtained only in cells also carrying plaC cloned on a plasmid vector. In such cells the maintenance of the cloned plaC was found to be required for cell viability.² These results convincingly prove that the plaC gene product is the vegetative  factor.