JOURNAL OF BACTERIOLOGY, Aug. 2005, p. 5166–5178 Vol. 187, No. 150021-9193/05/$08.00�0 doi:10.1128/JB.187.15.5166–5178.2005Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Bacillus subtilis Phosphorylated PhoP: Direct Activation of the E�A-and Repression of the E�E-Responsive phoB-PS�V Promoters

during Pho ResponseWael R. Abdel-Fattah, Yinghua Chen,† Amr Eldakak, and F. Marion Hulett*

Laboratory for Molecular Biology, Department of Biological Sciences, University of Illinois at Chicago,Chicago, Illinois 60607

Received 6 January 2005/Accepted 25 April 2005

The phoB gene of Bacillus subtilis encodes an alkaline phosphatase (PhoB, formerly alkaline phosphatase III)that is expressed from separate promoters during phosphate deprivation in a PhoP-PhoR-dependent mannerand at stage two of sporulation under phosphate-sufficient conditions independent of PhoP-PhoR. Isogenicstrains containing either the complete phoB promoter or individual phoB promoter fusions were used to assessexpression from each promoter under both induction conditions. The phoB promoter responsible for expres-sion during sporulation, phoB-PS, was expressed in a wild-type strain during phosphate deprivation, butinduction occurred >3 h later than induction of Pho regulon genes and the levels were approximately 50-foldlower than that observed for the PhoPR-dependent promoter, phoB-PV. E�E was necessary and sufficient forPS expression in vitro. PS expression in a phoPR mutant strain was delayed 2 to 3 h compared to the expressionin a wild-type strain, suggesting that expression or activation of �E is delayed in a phoPR mutant underphosphate-deficient conditions, an observation consistent with a role for PhoPR in spore development underthese conditions. Phosphorylated PhoP (PhoP�P) repressed PS in vitro via direct binding to the promoter, thefirst example of an E�E-responsive promoter that is repressed by PhoP�P. Whereas either PhoP or PhoP�Pin the presence of E�A was sufficient to stimulate transcription from the phoB-PV promoter in vitro, roughly10- and 17-fold-higher concentrations of PhoP than of PhoP�P were required for PV promoter activation andmaximal promoter activity, respectively. The promoter for a second gene in the Pho regulon, ykoL, was alsoactivated by elevated concentrations of unphosphorylated PhoP in vitro. However, because no Pho regulon geneexpression was observed in vivo during Pi-replete growth and PhoP concentrations increased only threefold invivo during phoPR autoinduction, a role for unphosphorylated PhoP in Pho regulon activation in vivo is notlikely.

Bacillus subtilis is a gram-positive bacterium that normallyresides in soil in which inorganic phosphate is present at verylow concentrations (approximately 2 to 3 orders of magnitudelower than the concentrations of other required ions) (34). B.subtilis has a two-gene operon encoding the two-componentPhoP-PhoR signal transduction system (42, 43). The histidinekinase, PhoR, is an autokinase that serves as a phosphodonorto the response regulator, PhoP. During phosphate starvation,when external inorganic phosphate (Pi) concentrations de-crease to less than 0.1 mM (40), PhoP is required to activate orrepress transcription of a set of genes, including the genesinvolved in alkaline phosphatase (APase) production, cell wallbiosynthesis, inorganic phosphate transport, and cytochromebiogenesis (21).

Different patterns of PhoP binding are observed at PhoP-activated and -repressed promoters (for details see references3 and 21). Activated promoters contain a core binding regionpositioned on the coding strand roughly between positions �20and �60 relative to the transcription start site that contains

four repeated 6-bp consensus PhoP-binding sequences thatform two-dimer binding sites where PhoP dimers bind coop-eratively (13). Activated promoters often have secondary PhoPdimer binding sites either 3� or 5� of the core binding regionthat are required for full promoter activity (13, 31). At re-pressed promoters phosphorylated PhoP (PhoP�P) binds toregions that overlap the transcription initiation site, and insome cases PhoP�P oligomerizes along the DNA as far as 168bp into the coding region. Repressed promoters with only onePhoP dimer binding site, not two PhoP dimer binding sites,require phosphorylation for PhoP binding (4, 28).

Alkaline phosphatases in B. subtilis are encoded by membersof a multiple-gene family that is expressed differentially underphosphate starvation conditions and sporulation conditions(14, 22). Phosphate starvation-inducible (PSI) alkaline phos-phatases require the phoPR gene products, whereas sporula-tion-inducible alkaline phosphatases are expressed indepen-dent of the Pi concentration or phoP and/or phoR geneproducts but rather require the products of at least three stageII sporulation genes, including spoIIAC (encoding the fore-spore-specific sigma factor, �F, which regulates the �E activa-tion), spoIIE (a PP2C phosphatase required for the activationof �F), and spoIIGB (encoding the mother cell specific sigmafactor, �E) (5, 9). Two major alkaline phosphatases, PhoB(formerly APase III) and PhoA (formerly APase IV) (19),account for approximately 98% of total alkaline phosphatase

* Corresponding author. Mailing address: Laboratory for MolecularBiology, Department of Biological Sciences, University of Illinois atChicago, 900 S. Ashland Avenue (M/C 567), Chicago, IL 60607.Phone: (312) 996-5460. Fax: (312) 413-2691. E-mail: Hulett@uic.edu.

† Present address: Department of Medicine, Section of Hematology/Oncology, University of Chicago, Chicago, Ill.

5166

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

specific activity in response to phosphate starvation (6, 24).Other alkaline phosphatases account for less than 2% of thetotal alkaline phosphatase specific activity induced duringphosphate starvation, including that of PhoD phosphodiester-ase (13, 14).

phoB is among the members of both the �E-controlled genesand the Pho regulon genes that have recently been identifiedusing DNA microarray analysis (15, 32), reporter gene tech-nology screening (37), and transcriptional profiling (16). Thesegenes include phoB-ydhF (2, 15, 32), phoPR (15, 32), yhaX (15,37), yhbH (15, 37), yycP (15, 32), and glnQ (15, 32), all of whichwere reported to be PSI in a phoPR-dependent manner and arepotentially transcribed from �E-dependent promoters, at leastduring sporulation induction. Inclusion of the phoPR promoteramong these genes and operons is explained by evidence indi-cating that three of the six phoPR operon promoters (35; A.Puri and F. M. Hulett, unpublished data) require PhoP�P forenhanced transcription. One of these promoters is an E�E-responsive promoter, and two are E�A promoters (35).

phoB promoter deletion analysis was previously performedusing a number of nonisogenic B. subtilis strains that containedmutations in important regulatory genes, including abrB,spo0A, spo0B, sigF, spoIIE, sigE, phoR, phoP, and phoS (9, 20).These studies established that phoB gene transcription is ini-tiated from two transcription start sites at two different pro-moters. A strong vegetative cell promoter (PV) was phosphatestarvation inducible only in a phoP- and/or phoR-dependentmanner and accounted for approximately 40% of the totalalkaline phosphatase activity during phosphate starvation. A50-fold-weaker sporulation promoter (PS) was expressed atstage II during phosphate-replete growth on modified Schaf-fer’s sporulation medium; its expression accounted for approx-imately 45% of the total alkaline phosphatase specific activityduring sporulation induction (6, 22, 24).

Sequence analysis of the phoB promoter region (see Fig. 4A)revealed the presence of a consensus �10 region (TATAAT)for �A-dependent promoters upstream of the PV transcriptionstart site, but no consensus �35 region (TTGACA) for �A-dependent promoters was observed (9). A moderately con-served consensus �10 region (AATAACCA) and a �35 region(TTCTAAA) for �E-dependent promoters (15) were observedupstream of the PS promoter transcription start site. The re-sponse regulator PhoP was reported to bind to the phoB pro-moter exclusively between the PV and PS transcriptional initi-ation sites (29), a region which contains four direct repeats(TT[C/A/T]A[C/T]A) of the conserved sequence for PhoPbinding (13, 30).

In this study, we used isogenic B. subtilis strains, descendantsof parental strain JH642, to study the effect of a phoPR or sigEmutation on the temporal and differential transcription fromeither the PV or PS promoter during phosphate starvation orsporulation induction. Our studies indicated that the PS pro-moter was expressed during phosphate starvation but was ex-pressed later than Pho regulon genes. PS promoter expressionwas further delayed in a phoPR mutant strain. We used in vitrotranscription to explore the relative contributions of PhoP,PhoP�P, �A, and/or �E in transcriptional activation from thephoB promoters.

MATERIALS AND METHODS

Bacterial strains and growth conditions. All strains, plasmids, and primersused in this study are listed in Table 1. Bacterial stocks maintained at �80°Cwere used to inoculate LB agar plates for Escherichia coli strains or tryptoseblood agar base containing 0.5% glucose for B. subtilis strains. The media weresupplemented with selective antibiotics at the following concentrations: tetracy-cline, 10 mg/liter; erythromycin, 0.25 mg/liter; chloramphenicol, 5 mg/liter; andampicillin, 200 mg/liter. To obtain isogenic wild-type and mutant B. subtilisstrains harboring the required phoB promoter-lacZ reporter fusion, threepDH32 derivatives, pRC695, pRC696, and pPS2, containing the phoB promoterregions (Fig. 1) (relative to the translation start codon at position 1) at positions�178 to 34 (PS�V), �118 to 34 (PV), and �432 to �87 (PS), respectively, wereconstructed (9) and used for transformation of wild-type (JH642), phoPR(MH5913), and sigE (EU8701) strains. Before transformation, plasmids pRC695and pRC696 were linearized by ScaI digestion, and plasmid pPS2 was linearizedby NruI digestion. Transformants were selected for chloramphenicol resistanceand amyE phenotypes. The resultant nine strains were grown in low-phosphatedefined medium (LPDM) for Pi starvation conditions or in modified Schaffersporulation medium (43 mM Pi) amended with 0.1% glucose (SSG medium) forsporulation induction in Pi-replete conditions.

To construct a plasmid for overexpression of the mature active form lackingthe transcription-inhibiting prosequence (25) of the �E protein (*�E), the sigEgene lacking the region encoding the prosequence was PCR amplified usingJH642 chromosomal DNA with primers FMH490 and FMH491 containingcleavage sites for NdeI and BamHI, respectively. The PCR-amplified fragmentswere cloned into plasmid pCR2.1, resulting in plasmid pCH201. The mutant sigEgene was then released from pCH201 by digestion with NdeI and BamHI andcloned into plasmid pET16b at the same sites to generate plasmid pCH202. Theaccuracy of all plasmid constructs was confirmed by DNA sequencing. PlasmidpCH202 was used for transformation of E. coli BL21(DE3)/pLysS, and repre-sentative transformants were selected for the chloramphenicol resistance phe-notype.

Enzyme assays. For determination of total alkaline phosphatase specific ac-tivity, 250 �l of an LPDM culture was added directly to 1 ml of the substrate, 0.1M p-nitrophenyl phosphate in 1 M CHES (N-cyclohexyl-2-aminoethanesulfonicacid; pH 9.5), and the reaction rates were measured at an optical density at 420nm (OD420). One unit of APase activity was equivalent to 1 �mol of p-nitrophe-nol released per min at 37°C. The APase specific activity was expressed in unitsper unit of OD540.

The �-galactosidase (�-Gal) specific activity was measured by the method ofFerrari et al. (17) using o-nitrophenyl-�-D-galactopyranoside as the substrate.One unit of �-Gal activity was equivalent to 0.33 nmol of o-nitrophenol releasedper min at 37°C. The �-Gal specific activity was expressed in units per milligramof total cellular protein. The amount of B. subtilis JH642 total cellular proteinwas calculated as previously described by Eder et al. (14).

Overexpression and purification of proteins. For overexpression of His10PhoP,His10*�E, GST*PhoR, and His6�

A proteins, E. coli BL21(DE3)/pLysS was used asa host for plasmids pCH128, pCH201, pLS21, and pLC2, respectively (Table 1).Overexpression and purification of His-tagged and glutathione S-transferase(GST)-tagged proteins were performed as described previously (29). *PhoR isthe soluble cytoplasmic portion of the PhoR protein. The His tags were cleavedby factor Xa treatment, whereas the GST tag was removed by thrombin cleavage(29). For preparation of the B. subtilis His6-tagged RNA polymerase (His6RNAP)core enzyme (E), B. subtilis strain MH5636 (38) was grown in 12 1-liter portionsof LB medium. Cells were collected at the end of the exponential phase ofgrowth. Cell lysis and RNAP core enzyme purification were performed as pre-viously described (38).

Gel mobility shift assay. For preparation of probes, the phoB-PS�V promoterfragment was amplified by PCR using B. subtilis JH642 genomic DNA as thetemplate and primers FMH745 and FMH746, whereas the phoB-PS promoterfragment was PCR amplified using plasmid pPS2 as the template and primersFMH124 and FMH128 (FMH124 and FMH128 are homologous to DNA inpDH32 adjacent to the EcoRI and BamHI cloning sites, respectively.). 32Plabeling was performed during PCR by incorporating [�-32P]dATP and/or[�-32P]dGTP (0.2 �Ci/�l) into the newly synthesized DNA strand in the presenceof limited dATP and/or dGTP concentrations (20 �M, usually one-tenth theconcentration of dCTP and dTTP). The 32P-labeled probes were ethanol pre-cipitated and extracted from a 1.2% agarose gel using a gel extraction kit(QIAGEN), and the final concentration was adjusted to 0.1 �M in nuclease-freetranscription buffer (NFTB) (see below). The binding reactions (8 �l) wereperformed in NFTB containing, in order of addition, 12 nM 32P-labeled DNA(�15,000 cpm), 180 �M ATP, 0.18 �M *PhoR, and PhoP at concentrations

VOL. 187, 2005 TRANSCRIPTIONAL ANALYSIS OF B. SUBTILIS phoB PROMOTER 5167

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

ranging from 0.35 to 2.8 �M. The reaction mixtures were incubated at 37°C for30 min and were subsequently loaded onto 5% nondenaturing polyacrylamidegels (acrylamide/bisacrylamide, 29:1), which had been prerun in Tris-borate-EDTA buffer at 100 V/10°C for 30 min, and electrophoresed for 2 h at 100V/10°C. The free probe and shifted PhoP�P-DNA complex on dried gels werevisualized using a PhosphorImager (Molecular Dynamics).

In vitro transcription analysis. All components used for in vitro transcriptionreactions were adjusted to the required concentrations by dilution in NFTB(Tris-HCl [pH 7.8], 10 mM; KCl, 50 mM; MgCl2, 5 mM; CaCl2, 1 mM; EDTA-Na2, 0.1 mM) containing 5% glycerol, unless indicated otherwise. To obtain thetemplate DNA for in vitro transcription analysis, the full-length phoB-PS�V

promoter region (positions �222 to 106 relative to the translation start codon atposition 1 [see Fig. 4A]) was PCR amplified using JH642 chromosomal DNA andprimers FMH745 and FMH746 (Table 1), and the ykoL promoter region (posi-tions �106 to 151 relative to the transcription start site at position 1 [see Fig.6A]) was PCR amplified using primers FMH769 and FMH768. The phoB-PS

promoter region was obtained by PCR using plasmid pPS2 as the template andprimers FMH124 and FMH128. Promoter DNA was extracted from a 1.2%agarose gel using a gel extraction kit (QIAGEN), ethanol precipitated, andresuspended in NFTB to a final concentration of 100 nM. To prepare RNAPholoenzyme, core RNAP was incubated with each � factor on ice for 30 min. Foreach transcription reaction (final volume, 15 �l), template DNA (33 nM) was

mixed with ATP (180 �M), *PhoR protein (0.18 �M), and PhoP protein atdifferent concentrations (0.02 to 10 �M in twofold increments) and incubated at37°C for 20 min. *PhoR and PhoP dilutions were prepared in phosphorylationbuffer (HEPES [pH 8.0], 50 mM; KCl, 50 mM; MgCl2, 50 mM; 7% glycerol)prior to addition to the reaction mixtures. To produce various concentrations ofPhoP�P in the in vitro transcription reaction mixtures, PhoP phosphorylationwas performed during the initial DNA binding step using *PhoR at a fixedconcentration, 0.18 �M. After the binding reaction, the reconstituted RNAPholoenzyme (core enzyme concentration, 0.09 �M; � factor concentration, 0.7�M) was added to the reaction mixture; this was followed by addition of anucleoside triphosphate mixture (180 �M each of ATP, GTP, and CTP, 18 �MUTP, 0.1 �M [�-32P]UTP [3,000 Ci/mmol; 10 �Ci/�l], and 1 U/�l RNasin RNaseinhibitor in NFTB) and incubation at 37°C for another 30 min. In vitro tran-scription reactions were stopped by addition of 7.5 �l stop buffer (7 M urea, 100mM EDTA-Na2, 0.05% xylene cyanol, 0.05% bromophenol blue, 5% glycerol),and the mixtures were heated at 75°C for 5 min. Ten microliters of each reactionmixture was loaded onto a sequencing gel (6% polyacrylamide [19:1], 8 M urea)and electrophoresed at 350 V for 1 h. The RNA transcripts on dried gels werevisualized with a PhosphorImager (Molecular Dynamics) and subjected to quan-titative analysis using the volume report function of the ImageQuant software,version 5.1, with corrections for background by subtraction of intensities below

TABLE 1. Bacterial strains, plasmids, and primers

Strain, plasmid, orprimer Genotype or characteristics Source or

reference

E. coli strainsDH5� InvitrogenBL21(DE3)/pLysS Novagen

B. subtilis strainsJH642 pheA1 trpC2 J. A. HochMH5636 pheA1 trpC2 rpoC pYQ52 (Cmr) 38MH5913 pheA1 trpC2 phoPR::tet (Tetr) 8EU8701 pheA1 trpC2 sigE::erm (Ermr) C. P. MoranMH6141 pheA1 trpC2 amyE::pRC695 (Cmr) This studyMH6142 pheA1 trpC2 amyE::pRC696 (Cmr) This studyMH6143 pheA1 trpC2 amyE::pPS2 (Cmr) This studyMH6144 pheA1 trpC2 phoPR::tet amyE::pRC695 (Tetr Cmr) This studyMH6145 pheA1 trpC2 phoPR::tet amyE::pRC696 (Tetr Cmr) This studyMH6146 pheA1 trpC2 phoPR::tet amyE::pPS2 (Tetr Cmr) This studyMH6147 pheA1 trpC2 sigE::erm amyE::pRC695 (Ermr Cmr) This studyMH6148 pheA1 trpC2 sigE::erm amyE::pRC696 (Ermr Cmr) This studyMH6149 pheA1 trpC2 sigE::erm amyE::pPS2 (Ermr Cmr) This study

PlasmidspRC695 pDH32::�178 to 34 of phoB promoter region (PS�V) 9pRC696 pDH32::�119 to 34 of phoB promoter region (PV) 9pPS2 pDH32::�432 to �78 of phoB promoter region (PS) 5pCH128 pSKB3::phoP (Cmr) 8pCR2.1 oriEc (Ampr) InvitrogenpCH201 pCR2.1::*sigE (Ampr) This studypET16b Vector for N-terminal fusion to factor Xa-cleavable His tag sequence; transcription

is controlled from T7 lac promoter (Cmr)Novagen

pCH202 pET16b::*sigE (Cmr) This studypLC2 E. coli overexpression plasmid for B. subtilis �A (Ampr) J. D. HelmannpLS21 pEGEX-2T::*phoR (Ampr) 45

Primersa

FMH745 5�-CAGTCCGCCTATCCACGGCCAGTATCT-3� This studyFMH746 5�-CTTTTCCTGGGCGCTGGCTTCAGGCACAC-3� This studyFMH490 5�-TGCATATGGGCGGGAGTGAAGCCCTGCCGCCTCCAT-3� This studyFMH491 5�-CTGGATCCTTACACCATTTTGTTGAACTC-3� This studyFMH769 5�-AAGAATTCGATCGGTTCAAAACGGACAATAC-3� This studyFMH768 5�-AAGGATCCGCTCAGAAATCGTCATTTTGTTCATG-3� This studyFMH124 5�-GTTCACCACCTTTTCCC-3� R. ChesnutFMH128 5�-GGTAACTATTGCCGATG-3� R. Chesnut

a The underlined restriction endonuclease sequences were added to the 5� ends of the primers for subsequent cloning applications. These sequences are NdeI,BamHI, EcoRI, and BamHI sites in primers FMH490, FMH491, FMH769, and FMH768, respectively.

5168 ABDEL-FATTAH ET AL. J. BACTERIOL.

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

and above each specified band. The radiolabeled RNA marker was preparedaccording to the manufacturer’s instructions (Novagen).

Estimation of cellular PhoP concentrations by Western immunoblotting. Pro-tein concentrations were determined using the Bio-Rad protein assay (Bio-RadLaboratories) with bovine serum albumin as the standard. For preparation of aPhoP standard curve, dilutions of purified standard PhoP were prepared inphoPR mutant cell lysates and subjected to immunoblot detection usingPhoPCTD-specific antibody (a polyclonal antibody developed against the PhoPC-terminal DNA binding domain) as previously described (8). To determine thecellular PhoP concentrations, wild-type strain MH6143 was grown in LPDM, andsampling began when the OD540 reached 0.5 � 0.1. Culture samples (15 ml) werecollected hourly from T�2 to T5 during phosphate starvation (where T0 repre-sents sporulation stage 0). Microscopic cell counting of culture samples wasperformed using a Petroff-Hausser counter (Hausser Scientific Partnership) anda phase-contrast microscope at a magnification of �400. Whole-cell extractswere prepared by resuspension of cell pellets in appropriate volumes of lysisbuffer (50 mM Tris-HCl [pH 7.0], 10 mM EDTA, 15 mg/ml lysozyme, 10 �g/mlDNase I, 100 �g/ml RNase A, 1 mM phenylmethylsulfonyl fluoride) to bring thefinal cell concentration to 5 � 107 cells/�l and were incubated at 37°C for 15 min;this was followed by addition of sodium dodecyl sulfate (SDS) sample buffer (0.5M Tris-HCl [pH 7.0], 12% SDS, 0.1% bromophenol blue, 20% 2-mercaptoetha-nol, 30% glycerol). Samples were incubated in a boiling water bath for 10 minprior to loading. Efficient cell lysis was monitored microscopically. Samples werediluted in phoPR cell lysates to obtain a final concentration equivalent to 2 � 107

cells/�l. From calibrated samples, 4 �l (equivalent to lysate from 8 � 107 cells)or 10 �l (equivalent to lysate from 2 � 108 cells) was electrophoresed on 12%SDS-polyacrylamide gel electrophoresis gels and subjected to immunoblot de-tection as previously described by Chen et al. (7). Developed blot membraneswere scanned using a computer scanner (MICROTEK ScanMaker 5), and bandswere quantified using the volume report function of the ImageQuant software.Band intensities were normalized by using standard PhoP protein applied toseparate lanes in the same gel. For estimation of cellular PhoP concentrations,we assumed that the cell volume was 10 �m3 (44).

Determination of inorganic phosphate concentration. The extracellular inor-ganic phosphate concentrations were determined for LPDM cultures using themethod of Ames and Dubin (1). For analysis, 300 �l of properly diluted culturesupernatant (usually 10% in deionized H2O for an LPDM culture) was mixedwith 700 �l of ascorbic acid reagent (10% ascorbic acid in H2O and 0.4 g ofammonium molybdate liter in 1 N H2SO4, mixed at a ratio of 1:6 [vol/vol]), andthe reaction mixtures were incubated at 45°C for 20 min. The inorganic phos-phate concentration (�M) was equivalent to the OD820 of the reaction mixturedivided by the slope obtained from a Pi standard curve. Under our experimentalconditions the slope of the Pi standard curve was �0.008 OD820 unit/�M.

RESULTS

For in vivo transcriptional analysis of the phoB promoter,three DNA fragments representing the full-length phoB pro-moter (PS�V), the vegetative promoter (PV), or the sporula-tion promoter (PS) were inserted upstream of the promoterlesslacZ reporter gene in the pDH32 plasmid, yielding plasmidspRC695, pRC696, and pPS2, respectively (Fig. 1). A singlecopy of each promoter-lacZ fusion was inserted into the amyElocus of a set of isogenic strains, including the parental B.subtilis strain (JH642) and phoPR (MH5913) and sigE(EU8701) derivatives of it. The �-galactosidase specific activitywas monitored for all nine strains during growth in LPDM forPi starvation induction or during growth in SSG medium forsporulation induction (Fig. 2). The growth kinetics of all pro-moter derivative mutants were similar to those of the parentalwild-type, phoPR, and sigE mutant strains.

Both PV and PS are expressed during phosphate starvation,and a phoPR mutation or a sigE mutation affects both promot-ers, although differently. During growth in LPDM, the paren-tal wild-type strains containing each of the three phoB pro-moter fusions induced expression from both the PV and PS

promoters (Fig. 2A). Expression from the PV or PS�V pro-moter fusions was induced at T0, when the Pi concentrationwas limiting (�100 �M), and peaked at T5, when maximumexpression levels of �6,000 �-Gal units were detected. Expres-sion from the PS promoter fusion was induced at T4, and thelevels were 50-fold lower than the levels from the PV pro-moter; however, they failed to reach the maximum levels in thefirst 12 h of growth. No expression from the PV promoterfusion was observed in the phoPR mutant grown in LPDM(Fig. 2B), indicating that the PSI expression from the strong PV

promoter is phoPR dependent. In addition, phoPR deletionaffected the timing of induction from the weak PS promoter.Expression from the PS promoter in phoPR mutants, fromeither the PS or PS�V promoter fusions, exhibited a further 2-to 3-h lag compared to the wild-type strain. During growth ofthe three sigE mutant strains in LPDM (Fig. 2C), no expressionwas observed from the PS promoter fusion, indicating that thePSI expression from the PS promoter is sigE dependent. Likeexpression in the wild-type strains, expression from the PV

promoter in the sigE mutants, from either the PV or PS�V

promoter fusions, exhibited the strong induction pattern thatwas induced at T0 and peaked at T5 during phosphate starva-tion. This expression was �40% higher than that in the paren-tal strains.

Sporulation-inducible phoB expression is solely from thesigE-dependent PS promoter. During growth of wild-type,phoPR, or sigE mutant strains in SSG medium (Pi-replete con-ditions), no expression was observed from the strong PV pro-moter fusion (Fig. 2D through F), indicating that the sporula-

FIG. 1. Chromosomal organization of the phoB-ydhF operon andits promoter deletions. Genes are represented by arrows that indicatethe direction of transcription. The original DNA clone containing theydhG-phoB intergenic region was fused with a promoterless lacZ genein pDH32, resulting in plasmid pCB619. The physical locations of thephoB-PV and phoB-PS transcription start sites (arrows), the putative �A

and �E consensus �10 and �35 recognition sequences (solid boxes),and the PhoP core binding region (open boxes) are indicated. The lineseparation was used to reduce the physical distance between bases�178 and �361 (the putative location of the ydhG promoter) relativeto the phoB translational start site at position 1. For promoter deletionanalysis various DNA fragments containing the PS and/or PV transcrip-tion start sites were fused with a promoterless lacZ gene in pDH32,resulting in plasmids pRC695, pRC696, and pPS2, as indicated.

VOL. 187, 2005 TRANSCRIPTIONAL ANALYSIS OF B. SUBTILIS phoB PROMOTER 5169

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

tion-inducible expression of phoB is from the weak PS

promoter only. Both the wild-type and phoPR mutant strains(Fig. 2D and E, respectively) induced expression from the PS

promoter, either from the PS�V or the PS promoter fusions, atT4 and peaked at T6, when the maximum expression levels(�80 �-Gal units) were detected. No expression from the PS

promoter fusion was observed during growth of the sigE mu-tant strains in either SSG medium (Fig. 2F) or LPDM (Fig.2C).

phoPR mutation affects the level and timing of expressionfrom the PS promoter during phosphate limitation. To deter-mine the extent to which a mutation in phoPR affected thetiming and inducing strength of the PS promoter under phos-phate deficiency stress conditions, we compared the parentaland phoPR mutant strains for growth and expression from thePS promoter fusion for 24 h in LPDM (Fig. 3). In the parentalstrain, expression from the PS promoter fusion was induced atT4 and peaked at T8 with a maximum expression of �110�-Gal units. Expression from the PS promoter fusion was de-layed 2 to 3 h in the phoPR mutant strain, and the maximumexpression level was reduced about 25%. In addition, the max-imum expression levels detected from the PS promoter in

FIG. 2. Expression of phoB promoter-lacZ reporters from different phoB promoter deletions in wild-type, phoPR, and sigE mutant B. subtilisstrains during 12 h of growth in LPDM and SSG medium. Plasmids pRC695, pRC696, and pPS2 containing the lacZ reporter fused to the PS�V(circles), PV (squares), and PS (triangles) promoters, respectively, were linearized and used to transform wild-type (WT) (JH642), phoPR(MH5913), and sigE (EU8701) parental strains. The resulting isogenic strains (Table 1) were grown under phosphate-limiting conditions in LPDM(A, B, and C) or SSG medium (D, E, and F). Growth (solid symbols) and �-galactosidase specific activity (open symbols) were determined at thetimes indicated. Time zero was the transition from the exponential to the stationary phase of growth. The outer right ordinate in panel Acorresponds to the high expression level from the PS�V (circles) and PV (squares) promoter derivatives, while the inner right ordinate correspondsto expression from the PS promoter derivatives (triangles).

FIG. 3. Expression of phoB-PS lacZ in wild-type and phoPR mutantB. subtilis strains during 24 h of growth in LPDM. Expression ofPS-lacZ was determined in wild-type strain MH6143 (squares) andphoPR MH6146 (triangles) grown under phosphate-limiting conditionsin LPDM. Solid symbols, growth; open symbols, �-galactosidase spe-cific activity. Time zero was the transition from the exponential to thestationary phase of growth.

5170 ABDEL-FATTAH ET AL. J. BACTERIOL.

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

LPDM (Fig. 3) were approximately twofold greater than thelevel detected in SSG medium (Fig. 2D), suggesting that me-dium composition also influences the strength of expressionfrom the sigE-dependent PS promoter.

phoB-PS promoter is an E�E-responsive promoter whosetranscription in vitro is inhibited by high PhoP�P concentra-tions. To determine if the role of PhoP-PhoR and the mothercell specific sigma factor (�E) in transcription of the phoB-ydhFoperon was direct or indirect, an in vitro transcription analysiswas performed using a 325-bp linear DNA fragment containingthe full-length phoB promoter (102 bp upstream of PS [position1] and 142 bp downstream of PV [position 1]) (Fig. 4A). Theexpected transcript initiating from the PS transcription startsite was 223 nucleotides long. As shown in Fig. 4B, when PhoPwas not present in the in vitro transcription reaction mixtures(lanes 1 through 8), �E-containing RNA polymerase (E�E) was

sufficient for transcription initiation from the phoB-PS pro-moter. No detectable PS transcripts were observed when eithercore RNA polymerase or �E was not present in the in vitrotranscription reaction mixtures (data not shown). Eight reac-tion mixtures (lanes 1 to 8) containing decreasing concentra-tions of core RNAP (0.18 to 0.01 �M) and increasing concen-trations of �E (0.09 to 0.18 �M) yielded similar concentrationsof transcript (lanes 1 to 4). E�E (molar ratio, 0.09:0.7) was usedin the following studies. When PhoP was present in the in vitrotranscription reaction mixtures at concentrations ranging from0.02 �M to 2.8 �M (Fig. 4B, lanes 9 through 16), no significanteffect on transcription from the �E-dependent PS promoterwas observed at PhoP concentrations of �0.35 �M (lanes 9through 13), but concentrations of 0.7 �M resulted in areduction in transcription. In the presence of *PhoR, PhoPconcentrations from 0.02 �M to 2.8 �M (lanes 17 through 24)

FIG. 4. In vitro transcription analysis of the phoB promoter. (A) Nucleotide sequence of the phoB promoter coding strand. Regulatory elementsare indicated as follows: arrows, transcription start sites for PS and PV; subscripts, �10 sequences for both promoters and �35 sequence for thePS promoter; superscripts, consensus PhoP tandem binding sequences; SD, Shine-Dalgarno sequence; Met, methionine (translational start codon).The lowercase letters indicate mismatches with the consensus sequence. The thick line and the dashed line indicate the DNase I-protected regionsfor PhoP(�P) on the coding and noncoding strands, respectively, and the arrowheads indicate the previously identified hypersensitive sites in thefootprints. Primers FMH746 and FMH745, used to amplify the phoB promoter, are indicated by arrows (5� to 3� direction) for the coding andnoncoding strands, respectively. The numbers below the sequences indicate positions relative to the phoB translational start site (position 1). Thenumbers above the sequences indicate positions relative to the transcription start site of the PS or PV promoter. (B) In vitro transcription analysisof phoB promoter using various concentrations of the core (E) and sigma factor (�) (lanes 1 through 8) in the reconstituted RNAP. (Upper panel)E�E; (lower panel) E�A. Linearly increasing concentrations of unphosphorylated PhoP (in twofold increments) were used in both the E�E- andE�A-driven transcription reactions in the absence of *PhoR (lanes 9 through 16) or in the presence of *PhoR at a final concentration of 0.18 �M(lanes 17 through 24). The protein concentrations (�M) are indicated above the lanes. Lane M contained a radiolabeled RNA marker. nts,nucleotides.

VOL. 187, 2005 TRANSCRIPTIONAL ANALYSIS OF B. SUBTILIS phoB PROMOTER 5171

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

resulted in greatly reduced transcription from the PS promoter.Thus, PhoP was not required for transcription initiation fromthe PS promoter, but the presence of low concentrations ofPhoP�P or high concentrations of PhoP repressed transcrip-tion from the �E-dependent PS promoter.

PhoP�P and E�A requirement for transcription from thephoB-PV promoter in vitro. For in vitro transcription analysisof the PV promoter, we used the reaction conditions describedabove, except that we initiated transcription with �A-saturatedRNAP (E�A). The expected transcript initiating from the PV

transcription start site (position 1) was 142 nucleotides long.As shown in Fig. 4B (lower panel), in the absence of PhoP,�A-containing RNA polymerase (E�A) failed to stimulatetranscription from the phoB promoter (lanes 1 through 8) atany concentration used. Addition of PhoP at concentrationsranging from 0.02 to 2.8 �M (lanes 9 through 16) stimulated adose-dependent increase in transcription from the PV pro-moter only when the unphosphorylated PhoP concentrationwas �0.18 �M (lanes 12 through 16). When *PhoR waspresent in the in vitro transcription reaction mixtures (lanes 17through 24), PhoP concentrations ranging from 0.02 to 0.18�M (lanes 17 through 20) stimulated a dose-dependent in-crease in transcription from the PV promoter, and the maximalyield was obtained at a PhoP concentration of 0.18 �M (lane21). Further increases in PhoP concentrations (lanes 21through 24) resulted in reduced transcription. The use of PhoPat concentrations of �1.4 �� significantly inhibited transcrip-tion activation from the PV promoter (lanes 23 and 24).

PhoP inhibition of PS promoter transcription is specific andinvolves PhoP binding to the PS promoter in addition to the PV

promoter. A phoB promoter fragment containing only the PS

promoter (from pPS2 [Fig. 1]) was used as a template for invitro transcription to determine if reduction of the PS tran-script from the phoB-PS�V promoter in the presence of phos-phorylated PhoP was exclusively the result of PhoP binding tothe PV promoter core binding region, which blocked RNAPpassage (Fig. 5A, top panel). Lane 1 contained the transcriptfrom the E�E-dependent PS promoter without PhoP or*PhoR. Lanes 2 through 6 contained increasing concentrationsof PhoP (0.09 to 1.4 �M) plus *PhoR, which were shown toincreasingly inhibit PS transcription from the phoB-PS�V pro-moter in the presence of *PhoR (0.18 �M) and ATP (Fig. 4B,upper panel). Under these conditions higher concentrations ofPhoP increasingly inhibited the PS promoter, although not asstrongly as identical concentrations did when the phoB-PS�V

promoter was used as the template. A dominant role forPhoP�P in PS inhibition was suggested by the strong transcriptin the reaction with 2.8 �M PhoP in the absence of *PhoR(lane7). Quantitation of the PS transcript with increasing con-centrations of PhoP plus *PhoR from the PS template alone iscompared in Fig. 5A (bottom panel) to quantitation of thetranscripts from the phoB-PS�V promoter fragment (Fig. 4B)with the same concentrations of PhoP or PhoP and *PhoR orwith E�E alone. Note the difference in PS transcription in thereaction mixture containing 1.4 �M PhoP with the full-lengthPS template in the presence of *PhoR.

To determine if inhibition of PS transcription by high con-centrations of PhoP and *PhoR was specific, we compared PS

transcription to transcription with another E�E-dependentpromoter, spoIIID, under conditions described above (Fig. 5A,

FIG. 5. Effect of deleting the downstream direct repeats on PhoP�P-mediated repression of transcription from the �E-dependent phoB-PSpromoter in vitro. (A) The upper panel shows in vitro transcription of the phoB-PS promoter lacking the downstream direct repeats, using E�E

alone (lane 1) and various concentrations of PhoP�P (lanes 2 through 6) or unphosphorylated PhoP (2.8 �M) (lane 7). The histogram in the lowerpanel shows the relative amounts of phoB-PS transcripts in vitro (in PhosphorImager output units) obtained using the PS�V promoter withunphosphorylated PhoP (open bars) or PhoP�P (solid bars) and using the PS promoter with PhoP�P (shaded bars). For comparison, themaximum transcript output obtained from the PS�V or PS promoter with E�E alone was normalized to 100%. (B) Specificity of the PhoP�P-mediated repression of transcription from the phoB-PS promoter. The phoB-PS promoter and the spoIIID promoter (a PhoP�P-insensitive,�E-dependent promoter) were used as templates for in vitro transcription reactions using E�E alone (lane 1), E�E with unphosphorylated PhoP(lane 2), and E�E with PhoP in the presence of a low *PhoR concentration (0.18 �M) (lane 3) or a high *PhoR concentration (0.35 �M) (lane4). (C) PhoP�P binds to the phoB-PS promoter independent of the downstream direct repeats. Gel mobility shift assays were performed using a325-bp 32P-labeled phoB-PS�V or 500-bp 32P-labeled phoB-PS promoter fragment in the absence of PhoP�P (lane 1) or in presence of variousPhoP�P concentrations (lanes 2 through 8). The positions of the free DNA probe (Fp) and shifted PhoP�P-DNA complex (CPhoP�P) areindicated on the right. The protein concentrations (�M) are indicated above the lanes. nts, nucleotides.

5172 ABDEL-FATTAH ET AL. J. BACTERIOL.

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

bottom panel) that inhibited the phoB-PS�V promoter. Lane 1in Fig. 5B shows the level of transcript from the phoB-PS (toppanel) or spoIIID (lower panel) template with E�E alone. E�E

with PhoP (2.8 �M) resulted in some decrease in the PS pro-moter transcript, but in the presence of *PhoR (0.18 �M) andPhoP the transcript was completely inhibited, similar to theresults obtained for the phoB-PS�V promoter (Fig. 4B, toppanel, lane 24). Under the same conditions and when the*PhoR concentration was increased from 0.18 to 0.35 �M, thelevel of the spoIIID transcript (lanes 3 and 4, respectively) wassimilar to the level observed with E�E without PhoP and*PhoR (lane 1), indicating that the role of PhoP and *PhoR ininhibition of the phoB promoter is specific.

Because transcription from the PS promoter was reduced inthe presence of PhoP�P, we asked if PhoP directly bound tothe PS promoter fragment in a gel mobility shift assay and howthe binding compared to PhoP binding to the full-length phoBpromoter (Fig. 5C). Lane 1 in Fig. 5C contained the full-lengthphoB-PS�V free probe in the top panel and the PS free probein the bottom panel. Lanes 2 to 8 in each panel containedincreasing concentrations of PhoP in the presence of ATP and0.18 �M *PhoR. Both probes revealed a partial shift at thelowest concentration of PhoP (lane 2), although the percent-age of the phoB-PS�V probe shifted was significantly greaterthan the percentage of the PS probe shifted. Roughly twofold-higher concentrations of PhoP were required for a completeshift of the phoB-PS probe (1.4 �M PhoP) (Fig. 5C, bottompanel, lane 6) than for a complete shift of the phoB-PS�V

probe (0.7 �M PhoP) (Fig. 5C, top panel, lane 4). These dataindicate that PhoP binds directly to the PS promoter indepen-dent of the core binding region in the PV promoter previouslyreported. In light of the current data, we reexamined the PhoPfootprinting protection on the phoB-PS�V promoter (29).While nearly complete protection of the large core bindingregion of the PV promoter was observed even at the lowestPhoP�P concentration (55 nM), there was also a previouslyunnoticed, smaller protected region upstream of the Ps tran-scription start site on the noncoding strand. Partial protectionwas observed starting at 110 nM PhoP�P, and there was in-creasing protection at PhoP�P concentrations of 220 and 440nM. The protected region in the promoter sequence is indi-cated in Fig. 4A. Interestingly, the protected region on thenoncoding strand contains two putative consensus repeats fora PhoP dimer-binding sequence on either side of the �10element of the E�E-responsive PS promoter (�2TAAATAATGGTTATTCTTT�19), with a conservation match of 5/6 and4/6, respectively. PhoP protection at this PS promoter site re-quired PhoR and ATP for PhoP phosphorylation, just as re-tardation of the PS promoter probe in the studies describedabove did (data not shown).

PhoP activates in vitro transcription from the phoB-PV pro-moter in a phosphorylation- and/or concentration-dependentmanner. The concentration range of PhoP used was expandedup to �10 �M to compare the maximum transcription of PV

with PhoP compared to that with PhoP�P and to determine ifhigh concentrations of PhoP inhibited transcription from thePV promoter (Fig. 6A). Because these experiments were de-signed to investigate the effects of various unphosphorylatedPhoP and PhoP�P concentrations on transcription from thePV promoter, the only experimental variable was the PhoP

concentration, and all other reaction components were un-changed. This strategy resulted in three changes in the ratio ofPhoP to PhoR. In the first case, the concentration of PhoP(0.02 to 0.09 �M) was less than that of PhoR (0.18 �M).Consistent with the data shown in Fig. 4B, this concentrationrange of PhoP stimulated transcription from the PV promoteronly when PhoR was present (Fig. 6A). In the second case,PhoP and PhoR were present at equimolar concentrations ornearly equimolar concentrations. The concentration ofPhoP�P (0.18 �M) was the optimum concentration for max-imum transcriptional yield from the PV promoter. In the lastcase, PhoP concentrations between 0.35 and 10 �M werehigher than the PhoR concentration. In contrast to the othercases, PhoP at concentrations of 0.18 �M in the presence ofPhoR resulted in linearly decreasing transcriptional yields, andcomplete repression was detected at concentrations of �1.4�M. Interestingly, the same range of PhoP concentrations wasable to stimulate transcription from the PV promoter in theabsence of PhoR. The optimum unphosphorylated PhoP con-centration that stimulated the maximum transcriptional yieldfrom the PV promoter was 2.8 �M or near 2.8 �M, a repressingconcentration in the case of PhoP�P. Decreased transcriptionfrom the PV promoter was also observed at higher concentra-

FIG. 6. Effects of various concentrations of unphosphorylatedPhoP, PhoP�P, and �A on transcription activation from the phoB-PVpromoter in vitro. (A) The upper panels show the results of an in vitrotranscription analysis of the phoB promoter using reconstituted E�A inthe presence of various concentrations of unphosphorylated PhoP(Un) or phosphorylated PhoP (�P). The graph in the lower panelshows the amounts of the phoB-PV transcripts (in PhosphorImageroutput units) plotted as a function of the unphosphorylated PhoP(open circles) or PhoP�P (solid circles) concentration. The PhoP-to-PhoR molar ratios (plus signs) are plotted as a function of the variousPhoP concentrations used in the in vitro transcription reactions.(B) Dependence of unphosphorylated PhoP-driven transcription reac-tions on �A concentration: in vitro transcription of the phoB promoterusing various concentrations of �A in reconstituted RNAP in thepresence of the optimum concentration of either unphosphorylatedPhoP (Un*) (2.8 �M) or PhoP�P (�P*) (0.18 �M). The proteinconcentrations (�M) are indicated above the lanes.

VOL. 187, 2005 TRANSCRIPTIONAL ANALYSIS OF B. SUBTILIS phoB PROMOTER 5173

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

tions of unphosphorylated PhoP (concentrations above 5.25�M). Thus, both unphosphorylated PhoP and PhoP�P stim-ulated transcription from the PV promoter in a concentration-dependent manner for a maximum transcriptional yield, andaddition of higher concentrations of either form decreasedtranscription from the PV promoter. The maximum transcrip-tional yield with the unphosphorylated PhoP was �85% thatobtained with PhoP�P and required �16-fold-higher PhoPconcentrations.

To determine if phosphorylation of PhoP might influence itsinteraction with the E�A form of RNAP, we varied the con-centration of �A in the reconstituted RNAP (Fig. 6B) whilePhoP and PhoP�P were used at their optimal concentrations,determined as described above. At the optimal PhoP�P con-centration (0.18 �M), limiting concentrations of �A (0.02 �M)stimulated transcription, although the transcriptional yield wasreduced compared to that obtained using higher �A concen-trations. Increasing the �A concentration so that it was up toeightfold greater than the PhoP�P concentration had no sig-nificant effect on the transcriptional yield. In contrast, the �A

concentration significantly influenced the ability of the unphos-phorylated PhoP to stimulate transcription from the PV pro-moter in vitro. Concentrations of �A of �0.35 �M were notsufficient to allow the optimal unphosphorylated PhoP concen-tration to stimulate transcription. �A concentrations of �0.35�M allowed maximum transcription activation at the optimalunphosphorylated PhoP concentration. Interestingly, �A wasrequired at a �17-fold-higher concentration to allow transcrip-tion activation at the optimum PhoP concentration comparedto the concentration required at the optimum PhoP�P con-centration.

In general, this in vitro transcription analysis illustrated twodifferent conditions for PhoP-dependent transcription activa-tion from the E�A-responsive phoB-PV promoter. The firstcondition is phosphorylation dependent, which allows a rela-tively low PhoP�P concentration compared to the PhoP con-centration to stimulate transcription even at limiting �A con-centrations. The second condition is concentration dependent,requiring considerably higher concentrations of both PhoP and�A in order to stimulate transcription in vitro.

Unphosphorylated PhoP or PhoP�P is essential for tran-scriptional activation of a second PhoP-regulated promoter,ykoL. Because the ability of unphosphorylated PhoP to activatetranscription in vitro was observed for the first time during thecourse of this study, we explored the use of a second PhoP-regulated promoter. The ykoL promoter exhibits significantsimilarity to the phoB promoter (Fig. 7A). First, it is tran-scribed from a potential �A-dependent promoter which alsolacks a consensus �35 sequence essential for E�A recognition.Second, it contains the four TT(C/A/T)A(C/T)A repeats char-acteristic of PhoP-activated promoters, which are quite similarto the phoB promoter in terms of conservation, relative dis-tances between tandem sequences, and position upstream ofthe transcription start site (41). Recently, DNase I footprintanalysis of the ykoL promoter region with PhoP indicated thatthe levels of protection afforded by the phosphorylated andunphosphorylated forms of PhoP were similar (36), a situationthat we observed for the phoB-PV promoter. The in vitro tran-scription analysis of the ykoL promoter (Fig. 7B) showed thatdespite the ability of E�A alone to direct transcription from a

nonspecific start site (PX), which produced an RNA transcriptthat was �100 nucleotides long (Fig. 7B, lanes 2 through 8), noPV transcript was observed. Addition of PhoP in the absence of*PhoR (lanes 9 through 16) or in the presence of *PhoR (lanes17 through 23) switched transcription initiation to a site thatyielded a transcript that was �150 nucleotides long, consistentwith the PV transcription start site that was identified previ-ously by primer extension analysis of the ykoL mRNA duringgrowth of a wild-type strain in phosphate starvation medium(41) (Fig. 7A). Similar to the maximum transcriptional yieldwith the phoB promoter, the maximum transcriptional yieldobserved with the ykoL-PV promoter in the presence of un-phosphorylated PhoP was �75% that observed in the presenceof PhoP�P. Higher concentrations of unphosphorylated PhoPdecreased transcription activation from the ykoL-PV promoterstart site, and there was a 10% reduction in the transcriptionyield from the ykoL-PV start site at a PhoP concentration of 10�M (lane 16). PhoP�P (0.35 �M) showed maximal transcrip-tion activation. Figure 7C shows quantitation of the ykoL-PV

transcript with PhoP alone (lanes 9 to 16) and with PhoP plus*PhoR (lanes 17 to 23).

Cellular concentrations of PhoP increase threefold duringphosphate-limited growth. To determine the fold increase inthe cellular PhoP concentration induced upon Pi depletion, wecarried out a quantitative Western blot analysis of PhoP syn-thesis during growth in LPDM. Figure 8A shows the detectableconcentrations of PhoP in the linear range under the condi-tions used. PhoP exhibited steady-state levels of �1 �M (Fig.8B) during the pre-Pho induction period (from T�2 to T0).When the Pi concentration in the medium decreased below 100�M, alkaline phosphatases were induced and PhoP concentra-tions began to increase. Total APase specific activity increasedrapidly until T2, while the PhoP concentrations continued toincrease until T4. At T4, the PhoP concentration was approx-imately threefold higher than the concentration detected at T0.These data are consistent with our understanding of the twoAPase promoters and the complex phoPR promoter. It is likelythat production of PhoP continues into the stationary phasebecause the phoPR operon has equally strong, PhoP�P-en-hanced E�A and E�E promoters. In contrast, the APase geneseach have a strong, PhoP-dependent E�A promoter (phoA andphoB-PV), but the E�E promoter of phoB-PS is very weak(roughly 2% the strength of the E�A promoter) and it appearsto be inhibited by PhoP�P at high concentrations. Thus, thecontribution of the weak E�E phoB-PS promoter to totalAPase activity should be negligible, like its contribution tophoB expression from the phoB-PS�V promoters during phos-phate starvation (Fig. 2A).

DISCUSSION

PhoPR plays a dual role in regulation of the E�E-responsivephoB-PS promoter. In vitro transcription data reported hereestablished that E�E was necessary and sufficient for phoB-PS

promoter function. A direct role for E�E in phoB-PS promoterexpression was corroborated by phoB-PS promoter fusion dataobtained with a sigE mutant strain during phosphate-limitedgrowth in LPDM and during sporulation in SSG medium (Fig.2). Previous phoB-lacZ fusion data (9) failed to identify PS

promoter expression during phosphate-limited growth because

5174 ABDEL-FATTAH ET AL. J. BACTERIOL.

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

the promoter fusion contained the 50-fold-stronger PV pro-moter in addition to the PS promoter, which masked the con-tribution of the PS promoter. PS promoter expression duringphosphate-limited growth was delayed �2 h in the phoPRmutant (Fig. 3), suggesting that either synthesis or maturationof �E is delayed under the Pi deficiency stress conditions in theabsence of PhoPR. This implies that PhoPR or another Phoregulon gene is needed for efficient sporulation, which may beespecially important in nature as inorganic phosphate is oftenthe most limiting nutrient in the soil (34). PS promoter expres-sion from the PS�V promoter fusion was not observed previ-ously in phoP or phoR mutant strains grown in LPDM because

experiments were terminated before delayed PS promoter ac-tivation at T5 to T6 (9).

PhoP and PhoR have two roles in transcription from thephoB-PS promoter. As stated above, the positive role forphoPR affects the timely expression from the E�E-responsivepromoter, PS, during phosphate starvation. At the same time,our in vitro data suggest that PhoP�P represses transcriptionfrom the PS promoter by binding directly to the PS promoter.These results suggest that PhoPR has dual roles in develop-ment affecting the timing of at least one E�E promoter bypositively affecting when E�E is available during phosphatestarvation and also preventing early expression via PhoP�P

FIG. 7. In vitro transcription analysis of a second PhoP-regulated promoter, ykoL. (A) Nucleotide sequence of the ykoL promoter codingstrand. Regulatory elements are indicated as follows: arrows, transcription start site for PV; subscripts, �10 sequences for PV promoter;superscripts, consensus PhoP tandem binding sequences; SD, Shine-Dalgarno sequence; Met, methionine (translational start codon). Thelowercase letters indicate mismatches with the consensus sequence. Primers FMH768 and FMH769 used to prepare the ykoL promoter areindicated by arrows under the coding and noncoding strands, respectively. The numbers below the sequences indicate the positions relative to theykoL translational start site (position 1). The numbers above the sequences indicate the positions relative the transcription start sites of PV.(B) Runoff in vitro transcription analysis of ykoL promoter using various concentrations of purified B. subtilis sigma factor (�A) and core RNAP(E) (lanes 1 through 8). Linearly increasing concentrations of unphosphorylated PhoP (in twofold increments) were used in the E�A-driventranscription reactions in the absence of *PhoR (lanes 9 through 16) or in the presence of *PhoR at a final concentration of 0.18 �M (lanes 17to 23). The protein concentrations (�M) are indicated above the lanes. Lane M contained a radiolabeled RNA marker. (C) Amount of ykoL-PVtranscript (in PhosphorImager output units) plotted as a function of the unphosphorylated PhoP concentration (open circles) or the PhoP�Pconcentration (solid circles). nts, nucleotides.

VOL. 187, 2005 TRANSCRIPTIONAL ANALYSIS OF B. SUBTILIS phoB PROMOTER 5175

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

before Pho regulation is turned off. The fact that the latter rolewas not detected during previous induction studies of PS inphosphate-replete sporulation media when the PV promoterwas silent is understandable because there would have been nosignal for PhoP phosphorylation and thus no repression of PS

(9).Although the binding site in the phoB-PS promoter, up-

stream of the PV promoter, is positioned like other secondarybinding sites for PhoP-activated promoters, it is clear that ithas no role in PV activation because expression from the phoB-PS�V promoter fusion and expression from the PV promoterare the same (Fig. 2A and C). A requirement for phosphory-lation of PhoP for binding at PS was observed for two otherPhoP repressed promoters, resD and tagA (4, 28), that containa single consensus PhoP dimer binding site.

Different mechanisms control genes and operons that havebeen classified as both Pho and E�E regulon members. Forgenes classified as both Pho and E�E regulon genes, the mech-anism responsible for inclusion of the phoPR promoter has

been explained by PhoP�P amplification of �A- and �E-re-sponsive phoPR operon promoters (35). Here we showed thatthe PhoP-regulated phoB promoter, PV, is a �A promoter andthat transcription from the second promoter, PS, requires onlyE�E, although the PS transcription was delayed in a phoPRmutant (Fig. 2B and 3) and was negatively regulated byPhoP�P. yhaX and yhbH are similar to the phoB-PS promoterin three ways. Both these genes were expressed during Pi-limiting growth in a wild-type phoPR strain at a time later thanthe time that other Pho-regulated genes were expressed, nei-ther gene was transcribed in a sigE mutant strain, and like thePS promoter, neither yhaX nor yhbH has consensus tandemrepeats for PhoP binding on the coding strand upstream of theputative E�E promoter (37). Because the Pho regulon assign-ment for yhaX and yhbH was based solely on the absence ofpromoter expression in a phoR mutant strain, it seems that thepossible delay of E�E-dependent gene expression in thatstrain, not the lack of direct PhoPR promoter regulation, mayhave caused the negative results. Consistent with this idea,PhoP does not bind to the yhaX promoter (36). No furtherinformation is available on the mechanism of Pho regulationand �E-dependent regulation of yycP or glnQ. The putativeyycOP and glnQ promoters are interesting in that both containtwo putative consensus repeats for a PhoP dimer binding siteon the coding strand on either side of the �35 element of theputative �E promoter. It should be interesting to determine ifE�E-responsive PhoP-activated genes exist (in addition to au-toregulation) and if their requirement for PhoP binding isdifferent than the requirement for all PhoP-activated �A pro-moters studied (13) or for the PhoP-enhanced autoregulationE�E promoter (35).

PhoP or PhoP�P plus E�A is sufficient for phoB-PV or ykoLpromoter transcription in vitro, and the PhoP concentration isthe key. The failure of E�A to activate in vitro transcription ofphoB-PV (Fig. 4) or ykoL (Fig. 7) is likely due to the absenceof a �35 �A consensus binding region in either promoter, as istrue for other E�A PhoP-activated promoters which absolutelyrequire PhoP�P for activation, including the phoA, pstS, andtuaA promoters (30, 40). Previous studies indicated that un-phosphorylated PhoP was unable to activate transcription invitro. The maximum concentrations of unphosphorylated PhoPand �A used in the previous studies were �0.25 �M and 0.02�M, respectively. Consistent with the data presented here (Fig.5B), these concentrations of unphosphorylated PhoP plus E�A

were not sufficient to activate transcription in vitro. Here, weshowed that unphosphorylated PhoP stimulated transcriptionfrom the phoB-PV promoter in vitro. However, the concentra-tion of unphosphorylated PhoP required for maximum tran-scription was much higher than that required for PhoP�P witheither promoter (17-fold higher for phoB and 10-fold higherfor ykoL), indicating that phosphorylation has an importantrole.

While it is possible that PhoP isolated from E. coli is phos-phorylated, the activity of equimolar concentrations of PhoPcompared to the activity of PhoP in the presence of PhoR plusATP suggests that the percentages of PhoP phosphorylated inthe two cases differ considerably. Furthermore, a PhoPN struc-tural analysis (3) for crystals formed in 1 day at 4°C showed nophosphorylation. However, a number of E. coli-expressed un-phosphorylated response regulators have been shown to acti-

FIG. 8. Western immunoblot detection of cellular PhoP levelsduring the phosphate starvation response. (A) Dilutions of purifiedstandard PhoP were prepared in phoPR cell lysates and used forWestern immunoblot detection using PhoP-specific antibody (leftpanel). The standard curve in the graph on the right was preparedas described in Materials and Methods. (B) For Western immuno-blot detection of intracellular PhoP concentrations, the wild-typestrain (MH6143) was allowed to grow in LPDM, and samples werecollected at the times indicated. Enzymatic cell lysis was performedas described in Materials and Methods. Microscopic cell countingwas performed, and samples were calibrated in phoPR cell lysates toa final concentration equivalent to 2 � 107 cells/�l. From the cali-brated samples, 4 �l (equivalent to lysate from 8 � 107 cells) and 10�l (equivalent to lysate from 2 � 108 cells) were electrophoresed on12% SDS-polyacrylamide gel electrophoresis gels and subjected toimmunoblot detection using PhoPCTD-specific antibody. Growth(OD540), microscopic cell counts, APase specific activities, and ex-tracellular inorganic phosphate concentrations were determined atthe times indicated. Time zero was the transition from the expo-nential to the stationary phase of growth.

5176 ABDEL-FATTAH ET AL. J. BACTERIOL.

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

vate transcription in vitro (PrrA [10], Spo0A [27], and DesR[11]), only to lose this ability in the variant with a mutation atthe phosphorylatable aspartate residue and in strains contain-ing this variant (PrrA [10] and DesR [11]). Spo0A (26) andDesR (11) were proven to be phosphorylated when they wereisolated. UhpA or the UhpA D54N variant is constitutivelyactive for expression of the sugar phosphate transporter UhpTin E. coli when it is overexpressed, but the variant is totallyinactive as a single copy, suggesting that phosphorylation ofUhpA is required for a single copy but not when UhpA isoverexpressed (12, 33, 47). In the case of the PhoP/PhoQtwo-component system of Salmonella, overexpression of theresponse regulator PhoP or the D52A variant activated regu-lon genes in vivo, albeit in a signal-independent manner, re-portedly due to concentration-dependent activation of PhoPvia dimerization. Further studies indicated that signal-depen-dent (low-Mg2�) PhoP DNA binding is phosphorylation de-pendent in vivo (46). In contrast to the examples describedabove, ResD D57A was shown to induce transcription of theResDE regulon in response to oxygen limitation, suggestingthat ResD senses oxygen-limiting conditions via an unknownmechanism in addition to being activated by phosphorylationby ResE (18).

Because of the high concentrations of unphosphorylatedPhoP required for Pho regulon promoter activation in vitroand the modest threefold autoinduction of PhoP in vivo, webelieve that a physiological role for unphosphorylated PhoP ina wild-type strain is unlikely. However, activation of Pho regu-lon genes has been reported in a phoR deletion strain. Theability to induce low-level expression from both tuaA and pstSpromoters was retained in a phoR mutant strain during phos-phate-deficient growth (30, 40). In the latter study (40), theauthors proposed that PhoP was likely phosphorylated in aPhoR-independent manner (i.e., by unknown phosphodonors).In a second case, Ogura and colleagues (32) reported thatoverproduction of PhoP in vivo in a phoR deletion strain acti-vated expression of phoB and a number of Pho regulon genesin a signal-independent manner; these genes are not normallyactivated in a phoR mutant during phosphate starvation. Al-though a number of possibilities may explain this observation,during PhoP overproduction under nonphysiological condi-tions concentration-dependent activation of PhoP is a poten-tial factor.

Phosphorylation of PhoP minimizes the E�A levels requiredfor maximum transcription activation in vitro, consistent within vivo conditions during Pho induction. Because both un-phosphorylated PhoP and PhoP�P are dimers in solution (29)and both forms of PhoP bind to phoB (29) or ykoL (36) pro-moters with similar affinity, we asked if phosphorylation ofPhoP had an effect on its interaction with the transcriptionalmachinery (7). Consistent with this notion, when PhoP orPhoP�P was used at the concentration determined to be op-timal for phoB-PV transcription (Fig. 5A), concentrations of�A (�0.09 �M) required for PhoP�P-stimulated transcriptionactivity were significantly lower than the concentrations re-quired by unphosphorylated PhoP (Fig. 5B). Analysis of the invivo concentration of PhoP showed that autoinduction of PhoPsynthesis and APase expression initiated in parallel, whenPhoP levels were the lowest, at T0. During the following tran-sition period (T1 and T2), PhoP concentrations increased mod-

estly, roughly onefold, while the abundance of E�A is known todecrease due to sigma factor displacement (23). Taken to-gether, these data suggest that the phosphorylated form ofPhoP is probably responsible for in vivo activation of Phoregulon genes, including genes encoding APases.

Elevated concentrations of PhoP�P inhibited transcriptionin vitro, providing a possible explanation for previous in vivodata. PhoP�P is known to function not only as a transcriptionactivator but also as a transcription repressor of at least threeE�A-responsive promoters, including tagA/D (30, 39) and resD(4). One difference observed in vitro between repressed andactivated promoters is the customary requirement for PhoPphosphorylation for promoter binding at PhoP-repressed pro-moters (4, 30) compared to activated promoters, where eitherPhoP or PhoP�P binds. Also, it appears that similar levels ofPhoP�P are required for activation and repression in vivobecause repression of one set of genes and activation of an-other set by PhoP�P occur simultaneously. For example, uponphosphate starvation the genes responsible for synthesis ofteichoic acid (the tagA/D operons) are repressed (30), whilethe tuaA operon encoding proteins for the synthesis of teichu-ronic acid is activated (30), both in a phoPR-dependent man-ner. Thus, it seems unlikely that elevated levels of PhoP�Phave a physiologically relevant role in Pho gene repression in awild-type strain. However, when overexpression studies areconducted, PhoP�P may play a role. Data consistent with invivo repression of Pho regulon genes by elevated PhoP�Pconcentrations come from two in vivo studies, both involvingnonphysiological overexpression of PhoP. It was observed thatexpression of PhoP from its own promoter on a multicopyplasmid in a phoPR wild-type background resulted in a reducedgrowth rate and repression of Pho regulon genes even underPi-deficient conditions (W. Liu and F. M. Hulett, unpublished).In an independent study (32), overproduction of any one ofthree response regulators (PhoP, DegU, and ComA) in a wild-type background failed to result in expression of target genes,and in each case deletion of a cognate histidine kinase genewas required for expression. Our in vitro data are consistentwith the hypothesis that repression by elevated concentrationsof PhoP�P is one possible explanation for the previously ob-served in vivo repression of Pho regulon genes.

In summary, data presented here show that phoB-PS is anE�E-responsive promoter that is expressed under phosphate-deficient conditions and that the timing of this expression isdelayed in a phoPR mutant strain. These observations suggestthat workers should be cautious when analyzing promotersbelieved to depend on both E�E and phoPR. While a role forPhoP�P in activation of an E�E promoter has been reportedpreviously (35), repression of the PS promoter by PhoP�P isthe first example of a repressor role for PhoP�P at E�E-responsive promoters. Two E�A promoters that require PhoPfor expression in vivo and in vitro, phoB-PV and ykoL, wereshown to be activated at a high concentration of PhoP and tobe repressed by high concentrations of PhoP�P in vitro,thereby providing a possible explanation for previous in vivoobservations (32, 39). Phosphorylation of PhoP resulted indecreased concentration requirements for both PhoP�P andE�A for stimulation of Pho regulon promoters in vitro.

VOL. 187, 2005 TRANSCRIPTIONAL ANALYSIS OF B. SUBTILIS phoB PROMOTER 5177

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

ACKNOWLEDGMENTS

We thank Salbi Paul for providing the purified spoIIID promoterpreparation.

This work was supported by National Institutes of Health grantGM-33471 to F.M.H.

REFERENCES

1. Ames, B. N., and D. T. Dubin. 1960. The role of polyamines in the neutral-ization of bacteriophage deoxyribonucleic acid. J. Biol. Chem. 235:769–775.

2. Antelmann, H., C. Scharf, and M. Hecker. 2000. Phosphate starvation-inducible proteins of Bacillus subtilis: proteomics approach and transcrip-tional analysis. J. Bacteriol. 182:4478–4490.

3. Birck, C., Y. Chen, F. M. Hulett, and J.-P. Samama. 2003. The crystalstructure of the phosphorylation domain in PhoP reveals a functional tan-dem association mediated by an asymmetric interface. J. Bacteriol. 185:254–261.

4. Birkey, S. M., W. Liu, X. Zhang, M. F. Duggan, and F. M. Hulett. 1998. Phosignal transduction network reveals direct transcriptional regulation of onetwo-component system by another two-component regulator: Bacillus subtilisPhoP directly regulates production of ResD. Mol. Microbiol. 30:943–953.

5. Birkey, S. M., G. Sun, P. J. Piggot, and F. M. Hulett. 1994. A pho regulonpromoter induced under sporulation conditions. Gene 147:95–100.

6. Bookstein, C., C. W. Edwards, N. V. Kapp, and F. M. Hulett. 1990. TheBacillus subtilis 168 alkaline phosphatase III gene: impact of a phoAIIImutation on total alkaline phosphatase synthesis. J. Bacteriol. 172:3730–3737.

7. Chen, Y., W. R. Abdel-Fattah, and F. M. Hulett. 2004. Residues required forBacillus subtilis PhoP DNA binding or RNA polymerase interaction: alaninescanning of PhoP effector domain transactivation loop and � helix 3. J.Bacteriol. 186:1493–1502.

8. Chen, Y., C. Birck, J.-P. Samama, and F. M. Hulett. 2003. Residue R113 isessential for PhoP dimerization and function: a residue buried in the asym-metric PhoP dimer interface determined in the PhoPN three-dimensionalcrystal structure. J. Bacteriol. 185:262–273.

9. Chesnut, R. S., C. Bookstein, and F. M. Hulett. 1991. Separate promotersdirect expression of phoAIII, a member of the Bacillus subtilis alkaline phos-phatase multigene family, during phosphate starvation and sporulation. Mol.Microbiol. 5:2181–2190.

10. Comolli, J. C., A. J. Carl, C. Hall, and T. Donohue. 2002. Transcriptionalactivation of the Rhodobacter sphaeroides cytochrome c2 gene P2 promoterby the response regulator PrrA. J. Bacteriol. 184:390–399.

11. Cybulski, L. E., G. del Solar, P. O. Craig, M. Espinosa, and D. de Mendoza.2004. Bacillus subtilis DesR functions as a phosphorylation-activated switchto control membrane lipid fluidity. J. Biol. Chem. 279:39340–39347.

12. Dahl, J. L., B. Y. Wei, and R. J. Kadner. 1997. Protein phosphorylationaffects binding of the Escherichia coli transcription activator UhpA to theuhpT promoter. J. Biol. Chem. 272:1910–1919.

13. Eder, S., W. Liu, and F. M. Hulett. 1999. Mutational analysis of the phoDpromoter in Bacillus subtilis: implications for PhoP binding and promoteractivation of Pho regulon promoters. J. Bacteriol. 181:2017–2025.

14. Eder, S., L. Shi, K. Jensen, K. Yamane, and F. Hulett. 1996. A Bacillussubtilis secreted phosphodiesterase/alkaline phosphatase is the product of aPho regulon gene, phoD. Microbiology 142:2041–2047.

15. Eichenberger, P., S. T. Jensen, E. M. Conlon, C. van Ooij, J. Silvaggi, J. E.Gonzalez-Pastor, M. Fujita, S. Ben-Yehuda, P. Stragier, J. S. Liu, and R.Losick. 2003. The �E regulon and the identification of additional sporulationgenes in Bacillus subtilis. J. Mol. Biol. 327:945–972.

16. Eymann, C., H. Mach, C. R. Harwood, and M. Hecker. 1996. Phosphate-starvation-inducible proteins in Bacillus subtilis: a two-dimensional gel elec-trophoresis study. Microbiology 142:3163–3170.

17. Ferrari, E., D. J. Henner, M. Perego, and J. A. Hoch. 1988. Transcription ofBacillus subtilis subtilisin and expression of subtilisin in sporulation mutants.J. Bacteriol. 170:289–295.

18. Geng, H., S. Nakano, and M. M. Nakano. 2004. Transcriptional activation byBacillus subtilis ResD: tandem binding to target elements and phosphoryla-tion-dependent and -independent transcriptional activation. J. Bacteriol.186:2028–2037.

19. Hulett, F., E. Kim, C. Bookstein, N. Kapp, C. Edwards, and H. Wyckoff.1991. Bacillus subtilis alkaline phosphatases III and IV. Cloning, sequencing,and comparisons of deduced amino acid sequence with Escherichia colialkaline phosphatase three-dimensional structure. J. Biol. Chem. 266:1077–1084.

20. Hulett, F., J. Lee, L. Shi, G. Sun, R. Chesnut, E. Sharkova, M. Duggan, andN. Kapp. 1994. Sequential action of two-component genetic switches regu-lates the PHO regulon in Bacillus subtilis. J. Bacteriol. 176:1348–1358.

21. Hulett, F. M. 2002. The Pho regulon, p. 193–203. In A. L. Sonenshein, J. A.Hoch, and R. Losick (ed.), Bacillus subtilis and its closest relatives: fromgenes to cells. ASM Press, Washington, D.C.

22. Hulett, F. M., C. Bookstein, and K. Jensen. 1990. Evidence for two structuralgenes for alkaline phosphatase in Bacillus subtilis. J. Bacteriol. 172:735–740.

23. Ju, J., T. Mitchell, H. Peters III, and W. G. Haldenwang. 1999. Sigma factordisplacement from RNA polymerase during Bacillus subtilis sporulation. J.Bacteriol. 181:4969–4977.

24. Kapp, N. V., C. W. Edwards, R. S. Chesnut, and F. M. Hulett. 1990. TheBacillus subtilis phoAIV gene: effects of in vitro inactivation on total alkalinephosphatase production. Gene 96:95–100.

25. LaBell, T. L., J. E. Trempy, and W. G. Haldenwang. 1987. Sporulation-specific �-factor �29 of Bacillus subtilis is synthesized from a precursor pro-tein, P31. Proc. Natl. Acad. Sci. USA 84:1784–1788.

26. Ladds, J. C., K. Muchova, D. Blaskovic, R. J. Lewis, J. A. Brannigan, A. J.Wilkinson, and I. Barak. 2003. The response regulator Spo0A from Bacillussubtilis is efficiently phosphorylated in Escherichia coli. FEMS Microbiol.Lett. 223:153–157.

27. Lewis, R. J., D. J. Scott, J. A. Brannigan, J. C. Ladds, M. A. Cervin, G. B.Spiegelman, J. G. Hoggett, I. Barak, and A. J. Wilkinson. 2002. Dimerformation and transcription activation in the sporulation response regulatorSpo0A. J. Mol. Biol. 316:235–245.

28. Liu, W., S. Eder, and F. M. Hulett. 1998. Analysis of Bacillus subtilis tagABand tagDEF expression during phosphate starvation identifies a repressorrole for PhoP�P. J. Bacteriol. 180:753–758.

29. Liu, W., and F. Hulett. 1997. Bacillus subtilis PhoP binds to the phoB tandempromoter exclusively within the phosphate starvation-inducible promoter. J.Bacteriol. 179:6302–6310.

30. Liu, W., and F. Hulett. 1998. Comparison of PhoP binding to the tuaApromoter with PhoP binding to other Pho-regulon promoters establishes aBacillus subtilis Pho core binding site. Microbiology 144:1443–1450.

31. Liu, W., Y. Qi, and F. M. Hulett. 1998. Sites internal to the coding regions ofphoA and pstS bind PhoP and are required for full promoter activity. Mol.Microbiol. 28:119–130.

32. Ogura, M., H. Yamaguchi, K.-I. Yoshida, Y. Fujita, and T. Tanaka. 2001.DNA microarray analysis of Bacillus subtilis DegU, ComA and PhoP regu-lons: an approach to comprehensive analysis of B. subtilis two-componentregulatory systems. Nucleic Acids Res. 29:3804–3813.

33. Olekhnovich, I. N., J. L. Dahl, and R. J. Kadner. 1999. Separate contribu-tions of UhpA and CAP to activation of transcription of the uhpT promoterof Escherichia coli. J. Mol. Biol. 292:973–986.

34. Ozanne, P. G. 1980. Phosphate nutrition of plants—a general treatise, p.559–585. In E. Khasswenh (ed.), The role of phosphorus in agriculture.American Society of Agronomy, Madison, Wis.

35. Paul, S., S. Birkey, W. Liu, and F. M. Hulett. 2004. Autoinduction of Bacillussubtilis phoPR operon transcription results from enhanced transcription fromE�A- and E�E-responsive promoters by phosphorylated PhoP. J. Bacteriol.186:4262–4275.

36. Pragai, Z., N. E. Allenby, N. O’Connor, S. Dubrac, G. Rapoport, T. Msadek,and C. R. Harwood. 2004. Transcriptional regulation of the phoPR operon inBacillus subtilis. J. Bacteriol. 186:1182–1190.

37. Pragai, Z., and C. R. Harwood. 2002. Regulatory interactions between thePho and �B-dependent general stress regulons of Bacillus subtilis. Microbi-ology 148:1593–1602.

38. Qi, Y., and F. M. Hulett. 1998. PhoP-P and RNA polymerase �A holoenzymeare sufficient for transcription of Pho regulon promoters in Bacillus subtilis:PhoP-P activator sites within the coding region stimulate transcription invitro. Mol. Microbiol. 28:1187–1197.

39. Qi, Y., and F. M. Hulett. 1998. Role of PhoP�P in transcriptional regulationof genes involved in cell wall anionic polymer biosynthesis in Bacillus subtilis.J. Bacteriol. 180:4007–4010.

40. Qi, Y., Y. Kobayashi, and F. Hulett. 1997. The pst operon of Bacillus subtilishas a phosphate-regulated promoter and is involved in phosphate transportbut not in regulation of the pho regulon. J. Bacteriol. 179:2534–2539.

41. Robichon, D., M. Arnaud, R. Gardan, Z. Pragai, O. R. M., G. Rapoport, andM. Debarbouille. 2000. Expression of a new operon from Bacillus subtilis,ykzB-ykoL, under the control of the TnrA and PhoP-PhoR global regulators.J. Bacteriol. 182:1226–1231.

42. Seki, T., H. Yoshikawa, H. Takahashi, and H. Saito. 1987. Cloning andnucleotide sequence of phoP, the regulatory gene for alkaline phosphataseand phosphodiesterase in Bacillus subtilis. J. Bacteriol. 169:2913–2916.

43. Seki, T., H. Yoshikawa, H. Takahashi, and H. Saito. 1988. Nucleotide se-quence of the Bacillus subtilis phoR gene. J. Bacteriol. 170:5935–5938.

44. Sharpe, M. E., P. M. Hauser, R. G. Sharpe, and J. Errington. 1998. Bacillussubtilis cell cycle as studied by fluorescence microscopy: constancy of celllength at initiation of DNA replication and evidence for active nucleoidpartitioning. J. Bacteriol. 180:547–555.

45. Shi, L., and F. M. Hulett. 1999. The cytoplasmic kinase domain of PhoR issufficient for the low phosphate-inducible expression of Pho regulon genes inBacillus subtilis. Mol. Microbiol. 31:211–222.

46. Shin, D., and E. A. Groisman. 2005. Signal-dependent binding of the re-sponse regulators PhoP and PmrA to their target promoters in vivo. J. Biol.Chem. 280:4089–4094.

47. Webber, C. A., and R. J. Kadner. 1997. Involvement of the amino-terminalphosphorylation module of UhpA in activation of uhpT transcription inEscherichia coli. Mol. Microbiol. 24:1039–1048.

5178 ABDEL-FATTAH ET AL. J. BACTERIOL.

on March 11, 2020 by guest

http://jb.asm.org/

Dow

nloaded from