Isolation and Characterization of Mutants of the Bacillus subtilis ...

JOURNAL OF BACTERIOLOGY, Nov. 2003, p. 6425–6433 0021-9193/03/$08.00⫹0 DOI: 10.1128/JB.185.21.6425–6433.2003 Copyright © 2003, American Society for Microbiology. All Rights Reserved.

Vol. 185, No. 21

Isolation and Characterization of Mutants of the Bacillus subtilis Oligopeptide Permease with Altered Specificity of Oligopeptide Transport Jonathan Solomon,† Laura Su,‡ Stanley Shyn,§ and Alan D. Grossman* Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 Received 2 July 2003/Accepted 8 August 2003

Bacterial oligopeptide permeases are members of the large family of ATP binding cassette transporters and typically import peptides of 3 to 5 amino acids, apparently independently of sequence. Oligopeptide permeases are needed for bacteria to utilize peptides as nutrient sources and are sometimes involved in signal transduction pathways. The Bacillus subtilis oligopeptide permease stimulates competence development and the initiation of sporulation, at least in part, by importing specific signaling peptides. We isolated rare, partly functional mutations in B. subtilis opp. The mutants were resistant to a toxic tripeptide but still retained the ability to sporulate and/or become competent. The mutations, mostly in the oligopeptide binding protein located on the cell surface, affected residues whose alteration appears to change the specificity of oligopeptide transport. The oligopeptide permease of Bacillus subtilis is an ATP binding cassette (ABC) transporter that imports small peptides (30, 35). ABC transporters link ATP hydrolysis to the import or export of various compounds (3, 13, 43). It is estimated that the B. subtilis genome encodes approximately 60 different ABC transporters (34). Bacterial oligopeptide permeases typically import peptides of 3 to 5 amino acids, apparently independently of peptide sequence, although the oligopeptide permease of Lactococcus lactis has a preference for peptides containing 4 to 18 residues (4). Oligopeptide permeases are needed for bacteria to utilize peptides as nutrient sources. In addition, oligopeptide permeases are involved in signal transduction pathways regulating bacterial development, virulence, and conjugal plasmid transfer (2, 8, 15, 16, 22, 26). The B. subtilis opp operon encodes five proteins—OppA, OppB, OppC, OppD, and OppF. OppB and OppC are membrane-spanning proteins that create the transmembrane pore through which oligopeptides are imported (30, 35). OppD and OppF are ATP binding proteins that link transport to the hydrolysis of ATP. OppA is tethered to the extracellular face of the cell membrane via a lipid anchor. OppA binds to the peptide substrate and facilitates its uptake by interacting with and delivering the peptide to the OppBCDF complex. Mutations in oppA, oppB, oppC, and oppD completely block the ability to take up oligopeptides (30), whereas mutations in oppF cause a partial defect in transport (19). This partial defect causes resistance to low levels of a toxic tripeptide and a delay in competence gene expression (19, 35). Oligopeptide permeases can transport a wide variety of peptide substrates, in contrast to many other transporters, which

have a limited number of substrates. The structure of substrate binding protein OppA of Salmonella enterica serovar Typhimurium (36, 39, 40) and mutational analyses of OppA of L. lactis, based on the structure of OppA of Salmonella serovar Typhimurium (32), have provided insight into this unusual feature. When an oligopeptide binds in the cleft between two lobes (domains I and III) of OppA, the lobes close up, capturing the peptide. There are important charge interactions with the N and C termini of the bound peptide and weak bond interactions with the peptide backbone. The spacious cavity within the binding pocket of OppA can accommodate a wide variety of amino acid side chains, resulting in very little sequence specificity for binding peptides. In B. subtilis, the oligopeptide permease also functions to regulate two developmental processes, sporulation and genetic competence. In fact, B. subtilis opp was first identified as a sporulation locus, named spo0K, because null mutations caused decreased sporulation efficiency (30, 35). The B. subtilis oligopeptide permease contributes to the regulation of sporulation and competence, at least in part, by importing specific signaling peptides derived from phr gene products (16, 17, 27, 33). Seven phr genes have been recognized in the B. subtilis genome. Each encodes the precursor of a putative signaling peptide that is secreted and imported via the oligopeptide permease. Once inside the cell, a specific Phr peptide antagonizes the activity of its cognate regulatory protein(s) of the Rap family. The competence and sporulation-stimulating factor (CSF) is a pentapeptide derived from the phrC gene product (37). It is imported via the oligopeptide permease and, once inside the cell, stimulates the expression of genes (e.g., srfA) activated by transcription factor ComA. CSF most likely stimulates the activity of ComA by inhibiting RapC, a negative regulator of ComA (18, 37). At least three Rap proteins, RapA, RapB, and RapE, contribute to the regulation of sporulation (14, 28, 29). All three are response regulator aspartyl phosphate phosphatases and act to dephosphorylate Spo0F, thereby inhibiting sporulation. Pentapeptides derived from phrA, phrC, and phrE are imported via the oligopeptide permease and inhibit the activity of

* Corresponding author. Mailing address: Department of Biology, Building 68-530, Massachusetts Institute of Technology, Cambridge, MA 02139. Phone: (617) 253-1515. Fax: (617) 253-2643. E-mail: adg @mit.edu. † Present address: Elixir Pharmaceuticals, Cambridge, MA 02139. ‡ Present address: Department of Internal Medicine, Stanford Hospital and Clinic, Stanford, CA 94305. § Present address: Division of Biological Sciences, University of California at San Diego, La Jolla, CA 92093. 6425

6426

J. BACTERIOL.

SOLOMON ET AL. TABLE 1. B. subtilis strains used

Strain

Genotype

JRL291 JRL293 JMS198 JMS213 JMS289 JMS291 JMS319 Not5 Not10 Not16 Not17 Not37 Not39 Not47 Not50 Not103 Not104 Not105 Not109 Not110 Not201 Not202 Not203 Not204 Not205 Not206 Not207 Not208 Not209 Not210 Not211 Not212 Not213 Not371 Not372 Not373 Not374 Not375 Not376 Not377 Not378

⌬oppA amyE::(srfA-lacZ cat) amyE::(srfA-lacZ cat) ⌬oppA metC::Tn917 (mls) amyE::(comG-lacZ neo) ⌬oppA metC::Tn917 (mls) amyE::(srfA-lacZ cat) thrC::(comG-lacZ mls) yjbB::spc thrC::(comG-lacZ mls) ⌬oppA thrC::(comG-lacZ mls) oppA-G337R yjbB::spc thrC::(comG-lacZ mls) oppD-E239K yjbB::spc thrC::(comG-lacZ mls) oppA-E259K yjbB::spc thrC::(comG-lacZ mls) oppA-E259K yjbB::spc thrC::(comG-lacZ mls) oppA-R443H yjbB::spc thrC::(comG-lacZ mls) oppA-G337R yjbB::spc thrC::(comG-lacZ mls) oppA-G466E yjbB::spc thrC::(comG-lacZ mls) oppC-A266V yjbB::spc thrC::(comG-lacZ mls) oppA-G337R yjbB::spc thrC::(comG-lacZ mls) oppA-D449N yjbB::spc thrC::(comG-lacZ mls) oppA-E259K yjbB::spc thrC::(comG-lacZ mls) oppA-E259K yjbB::spc thrC::(comG-lacZ mls) oppA-G299E yjbB::spc thrC::(comG-lacZ mls) oppA-G337R yjbB::spc thrC::(comG-lacZ mls) oppA-E259K yjbB::spc thrC::(comG-lacZ mls) oppA-E259K yjbB::spc thrC::(comG-lacZ mls) oppA-R443H yjbB::spc thrC::(comG-lacZ mls) oppA-G337R yjbB::spc thrC::(comG-lacZ mls) oppA-G466E yjbB::spc thrC::(comG-lacZ mls) oppC-A266V yjbB::spc thrC::(comG-lacZ mls) oppD-E239K yjbB::spc thrC::(comG-lacZ mls) oppA-G337R yjbB::spc thrC::(comG-lacZ mls) oppA-D449N yjbB::spc thrC::(comG-lacZ mls) oppA-E259K yjbB::spc thrC::(comG-lacZ mls) oppA-E259K yjbB::spc thrC::(comG-lacZ mls) oppA-G299E yjbB::spc thrC::(comG-lacZ mls) oppA-G337R yjbB::spc amyE::(srfA-lacZ cat) oppA-E259K yjbB::spc amyE::(srfA-lacZ cat) oppA-R443H yjbB::spc amyE::(srfA-lacZ cat) oppA-G466E yjbB::spc amyE::(srfA-lacZ cat) oppA-D449N yjbB::spc amyE::(srfA-lacZ cat) oppA-G299E yjbB::spc amyE::(srfA-lacZ cat) oppC-A266V yjbB::spc amyE::(srfA-lacZ cat) oppD-E239K yjbB::spc amyE::(srfA-lacZ cat)

RapA, RapB, and RapE, respectively (18, 25), thereby stimulating sporulation. In this article we describe the isolation of rare, partly functional mutations in B. subtilis opp. The identification of these mutations was relatively simple because of the role of B. subtilis opp in regulating competence and sporulation. The mutants were deficient in the transport of a specific toxic peptide but still retained the ability to sporulate and/or become competent. The mutations, mostly in oppA, affected residues whose alteration appears to change the specificity of oligopeptide transport. MATERIALS AND METHODS Strains and plasmids. Plasmids were cloned and maintained in Escherichia coli strain AG1111, which is MC1061 [araD139 ⌬(ara-leu)7697 ⌬lacX74 galU galK rpsL hsdR] containing F⬘ proAB⫹ lacIq lacZM15 Tn10 (F⬘ in strain XL-1 Blue; Stratagene). The B. subtilis strains used are listed in Table 1. All are derived from strain JH642 and contain trpC2 and pheA1 mutations (31). The metC85::Tn917 allele (41) is from Bacillus Genetic Stock Center strain 1A607. ⌬oppA is a nonpolar, in-frame deletion of oppA (⌬oppA131) that removes the DNA from codon 18 (at the EspI site) to codon 477 (at the EcoRI site) out of a total of 545 codons (19). The yjbB::spc mutation disrupts the gene immediately downstream of oppF by

Reference, source, or comment

20 (also known as JMS747)

Strain that was mutagenized with EMS Original isolate Original isolate Original isolate Original isolate Original isolate Original isolate Original isolate Original isolate Original isolate Original isolate Original isolate Original isolate Original isolate Backcross of Not5 into JMS289 Backcross of Not16 into JMS289 Backcross of Not17 into JMS289 Backcross of Not37 into JMS289 Backcross of Not39 into JMS289 Backcross of Not47 into JMS289 Backcross of Not50 into JMS289 Backcross of Not10 into JMS289 Backcross of Not103 into JMS289 Backcross of Not104 into JMS289 Backcross of Not105 into JMS289 Backcross of Not109 into JMS289 Backcross of Not110 into JMS289 srfA-lacZ into thr⫹ derivative of Not201 srfA-lacZ into thr⫹ derivative of Not202 srfA-lacZ into thr⫹ derivative of Not204 srfA-lacZ into thr⫹ derivative of Not206 srfA-lacZ into thr⫹ derivative of Not210 srfA-lacZ into thr⫹ derivative of Not213 srfA-lacZ into thr⫹ derivative of Not207 srfA-lacZ into thr⫹ derivative of Not208

replacing the DNA between the BclI and SplI sites of yjbB with a spectinomycin resistance cassette. It was used to transfer unmarked alleles of opp by cotransformation, with selection for resistance to spectinomycin and screening for phenotypes associated with opp. yjbB::spc had no detectable effect on any of the phenotypes tested, including sporulation, competence gene expression, bialaphos resistance, and peptide transport (data not shown). The srfA-lacZ fusion (srfA-lacZ⍀1974) is a translational fusion located in a single copy at the amyE locus (10). The comG-lacZ fusion is a transcriptional fusion located in a single copy at the thrC locus (1, 20). Media. Routine growth and transformation of B. subtilis were done by using standard techniques and media (12). Liquid minimal medium was defined S7 medium (42), except that morpholinepropanesulfonic acid (MOPS) buffer was used at 50 mM instead of 100 mM. Minimal medium in agar plates contained Spizizen’s minimal salts (12). Minimal medium was supplemented with 1% glucose (0.1% glucose for minimal sporulation plates), 0.1% glutamate, and essential amino acids. Nutrient sporulation medium was DS medium (12). Competence indicator plates contained SpII medium (12). Antibiotics were used at the following concentrations: ampicillin, 100 ␮g/ml; chloramphenicol, 5 ␮g/ml; spectinomycin, 100 ␮g/ml; neomycin, 5 ␮g/ml; and erythromycin, 0.5 ␮g/ml, and lincomycin, 12.5 ␮g/ml, together to select for the mls gene. 5-Bromo-4-chloro3-indolyl-␤-D-galactopyranoside (X-Gal) was added to indicator plates from 30 to 120 ␮g/ml. Bialaphos (phosphinothricyl-alanyl-alanine) was a generous gift from Meiji Seika Kaisha, Ltd., Pharmaceutical Research Center, Yokohama, Japan, and was routinely used at 5 ␮g/ml unless otherwise indicated.

B. SUBTILIS Opp ALTERED-SPECIFICITY MUTANTS

VOL. 185, 2003 Mutagenesis and mutant isolation. Cells were grown in DS medium to midexponential phase, pelleted, and resuspended in Spizizen salts–1 mM MgSO4. Cells were then shaken at 37° for 30 min in the presence of ethyl methanesulfonate (EMS) (1.5% final concentration). EMS-treated cells were pelleted, washed twice in DS medium to remove the EMS, separated into 10 aliquots, and grown to confluence in Luria-Bertani medium. Mutants resistant to 5 ␮g of bialaphos/ml were selected either on competence plates (SpII medium) containing X-Gal or on minimal sporulation plates. Bialaphos-resistant colonies that appeared darker blue (comG-lacZ; Com⫹) or more opaque (Spo⫹) than opp-null mutants were purified, and chromosomal DNA was used for backcrossing. After backcrossing, with selection for a marker (yjbB::spc) linked to opp, 13 candidates retained bialaphos resistance and developmental phenotypes and were characterized further. Sequencing of mutations. We determined in which gene of the opp operon each mutation was located by complementation analysis. Wild-type copies of oppA and oppBCDF were introduced at the amyE locus, and the partial diploids were tested for mutant phenotypes (19). All but two mutations were complemented by oppA. The other two mutations were further tested by complementation with oppBC, oppD, and oppF at amyE (19). Each mutant allele was sequenced by using an fmol DNA sequencing kit (Promega). The DNAs for the sequencing reactions were PCR products derived from the chromosomal DNAs of the backcrossed mutants. PCR products were purified by using a Qiagen PCR purification kit. Bialaphos resistance assays. Cells were grown for at least three doublings in liquid minimal medium. The cultures were diluted 1/10 into minimal medium supplemented with various concentrations of bialaphos. The diluted cultures were grown for 3 h, after which the optical density at 600 nm (OD600) was determined. ␤-Galactosidase assays. ␤-Galactosidase specific activity was measured essentially as described previously (21) and is presented as 1,000 times the change in the A420 per minute per milliliter of culture per OD600 unit. Response to CSF. CSF is the pentapeptide ERGMT (37) and was synthesized by and purchased from Genemed Synthesis, Inc. (S. San Francisco, Calif.). CSF activity was measured essentially as described previously (20, 37, 38). Briefly, 0.25 ml of cells (containing an srfA-lacZ fusion) at an OD600 of 0.05 to 0.1 was added to 0.25 ml of medium containing various concentrations of CSF and bovine serum albumin (50 ␮g/ml) to prevent a nonspecific loss of activity. The mixture was incubated at 37°C for 70 min and then assayed for ␤-galactosidase specific activity. Sporulation assays. Cells were grown in DS medium at 37°C, and sporulation was assayed 18 to 24 h after the end of exponential growth. Percent sporulation is 100 times the number of spores per milliliter (heat-resistant CFU after treatment at 80°C for 20 min) divided by the number of viable cells per milliliter. OppA structural modeling. OppA (SwissProt accession number P24141) from B. subtilis was modeled by using the First Approach Mode at the Swiss-Model protein structure homology modeling server (http://us.expasy.org). B. subtilis OppA was modeled by using 33 structural templates of Salmonella serovar Typhimurium OppA and 5 structural models of E. coli dipeptide permease binding protein DppA (7). Amino acids 34 to 526 (out of 545) of B. subtilis OppA were included in the model. The graphics in Fig. 1 were produced by using Swiss-Pbd Viewer (9, 23, 24). The model is very similar to the solved crystal structure of Salmonella serovar Typhimurium OppA (36, 39, 40).

RESULTS Mutant isolation. We set out to identify mutations in the oligopeptide permease that cause a partial loss of function and perhaps alter the specificity of peptide transport. We selected mutations that allowed cells to grow in the presence of the toxic tripeptide bialaphos. Null mutations in opp are known to confer resistance to bialaphos (19, 30). To identify some of the rare opp mutants that still retained some function, the bialaphos-resistant mutants were screened for levels of competence gene expression or sporulation that were greater than those of opp-null mutants. Competence gene expression (comG-lacZ) was visualized on X-Gal indicator plates. The ability to sporulate was judged by colony appearance on minimal sporulation plates. opp-null mutants (and other early sporulation mutants) appear as translucent colonies after several days. Spo⫹ colonies are more opaque. Mutants resistant to 5 ␮g of bialaphos/ml were isolated at a

6427

FIG. 1. Model of B. subtilis OppA. OppA from B. subtilis was modeled as described in Materials and Methods. The three domains are shown in different colors: domain I (blue ribbon), domain II (red ribbon), and domain III (green ribbon). The interface between domains I and III comprises the peptide binding pocket. Residues altered by the oppA missense mutations are indicated as space-filling balls: D449N and R443H (aqua balls) probably affect charge interactions with the N and C termini of some oligopeptides. G337R and G466E (red balls) are in the peptide binding pocket. G299E (black balls) is in or next to the crossover region between domains III and I and is partly exposed in the peptide binding pocket. E259K (yellow balls) is in domain I.

frequency of approximately 10⫺3 following mutagenesis with EMS (see Materials and Methods). Of these, slightly more that 1 in 104 appeared to have levels of comG-lacZ expression or sporulation that were higher than those of opp-null mutants. We isolated 21 bialaphos-resistant mutants that appeared to retain the expression of comG-lacZ and 10 mutants that still appeared to sporulate. We isolated chromosomal DNA from these mutants and transferred (by transformation) the opp region to nonmutagenized cells by selecting for a marker, yjbB::spc, that is closely linked to opp. After backcrossing, 13 candidates retained their phenotypes. All further characterization was done with these strains or their derivatives. Mapping and sequencing of mutations. We determined in which gene of the opp operon each mutation was located by complementation analysis and DNA sequencing (see Materials and Methods). Most of the mutations (Table 2) were in oppA, encoding the substrate binding protein. Six different missense mutations in oppA were identified; one allele (oppA-E259K) was isolated four times (three independent isolates), and another (oppA-G337R) was isolated independently three times. One mutation was in oppC, encoding one of the two integral membrane proteins that make the pore. One mutation was in oppD, encoding one of the two ATP binding subunits. The oppD mutation is not in the conserved ABC region. That only two of the alleles were isolated multiple times indicates that the screen was probably not saturated. OppA structure and locations of missense mutations. We modeled the likely structure of B. subtilis OppA by using the

6428

SOLOMON ET AL.

J. BACTERIOL.

TABLE 2. Phenotypes of various opp mutants Allelea

Concn (␮g/ml) of bialaphos causing inhibitionb

% Sporulationc

srfAlacZd

Response to CSF (fold induction)e

Wild type ⌬oppA oppA-E259K (yellow) oppA-G299E (black) oppA-G337R (red) oppA-R443H (aqua) oppA-D449N (aqua) oppA-G466E (red) oppC-A266V oppD-E239K

⬍1.25 ⬎320 ⬃160 160–320 10–20 160–320 20–40 20–40 1.25–2.5 ⬎320

49 3 43 12 71 6 21 15 8 4

100 5 86 82 100 21 18 83 91 81

3.3 1.3 2.2 1.5 3.1 1.1 1.1 1.6 2.9 2.7

a Strains containing the indicated allele were tested for the various phenotypes. The colors correspond to the color of the amino acid in the model of OppA in Fig. 1. b Concentration of bialaphos at which growth was ⬃30% that in the absence of bialaphos, based on the experiments shown in Fig. 2. c Cells containing the indicated allele were grown in nutrient sporulation medium, and sporulation efficiency was measured ⬃18 h after the end of exponential growth. Similar results were obtained in multiple experiments. Strains used were the same as those used in the experiments shown in Fig. 2. d ␤-Galactosidase specific activity in late exponential phase (OD600, ⬃1.5), reported as a percentage of that in wild-type cells tested in parallel (from Fig. 3). e Maximum fold stimulation of srfA-lacZ in response to added CSF (from Fig. 4).

First Approach Mode at the Swiss-Model protein structure homology modeling server (http://us.expasy.org) and visualized the positions of the mutations in this model (Fig. 1). B. subtilis OppA is approximately 32% identical and 69% similar to Salmonella serovar Typhimurium OppA. The structures of Salmonella serovar Typhimurium OppA (39, 40) and E. coli dipeptide binding protein DppA are similar (7), and both were used to model B. subtilis OppA. OppA is composed of three domains. Domains I (Fig. 1, blue) and III (Fig. 1, green) are analogous to domains in many substrate binding proteins and form the surface interfaces that interact with the oligopeptide ligand (40). The function of domain II (Fig. 1, red) is not known. Five of the partly functional mutations in oppA are in domain III and likely affect peptide binding directly, and one mutation is in domain I. Two mutations, D449N and R443H, affect amino acids (Fig. 1, aqua balls) in domain III that are predicted to form important charge interactions with the N and C termini of at least some tri- and tetrapeptides (36, 39, 40). Two mutations, G337R and G466E, alter residues (Fig. 1, red balls) in domain III that are predicted to line the peptide binding pocket. These changes are likely to interfere directly with the binding of some oligopeptides. The G299E mutation also may affect peptide binding directly. This mutation alters an amino acid (Fig. 1, black balls) that is at the crossover region between domains III and I and is predicted to be partially exposed in the peptide binding pocket. The E259K mutation affects a residue (Fig. 1, yellow balls) in domain I that is predicted to be outside the peptide binding pocket. We suspect that E259K affects peptide binding indirectly. Of the amino acids altered in the mutants, only two, D449 and R443, are conserved between B. subtilis OppA and Salmonella serovar Typhimurium OppA. In Salmonella serovar Typhimurium OppA, these two amino acids form charge interactions with the N and C termini of at least some tri- and tetrapeptides. That we isolated mutations affecting these residues in B. subtilis OppA is consistent with the notion that they

affect peptide binding directly. These residues do not appear to be conserved in OppA of L. lactis, which binds larger oligopeptides. The precise role of the various amino acids of B. subtilis OppA in peptide binding ultimately will be elucidated when the structure is determined. Bialaphos transport. The eight opp mutations impaired but did not completely abolish bialaphos transport, as determined by resistance to various concentrations of the toxic tripeptide. We tested for resistance to bialaphos by measuring growth in liquid medium. Cells in exponential growth were diluted into various concentrations of bialaphos, and the change in the optical density of each culture was determined after 3 h and compared to that of a parallel culture without bialaphos (Fig. 2). The concentration of bialaphos at which growth was ⱕ30% that with no bialaphos was defined as the MIC (Table 2). oppA-null mutants were completely resistant and wild-type cells were completely sensitive to all concentrations of bialaphos tested (Fig. 2). The opp mutants were originally selected for resistance to 5 ␮g of bialaphos/ml, and all of the mutants showed at least some growth at that concentration (Fig. 2 and Table 2). The oppC-A266V mutation provided the lowest level and the oppD-E239K mutation conferred the highest level of resistance to bialaphos (Table 2 and Fig. 2D). All of the other mutations fell somewhere in between. At increased bialaphos concentrations, the growth of all of the opp missense mutants was impaired, indicating that bialaphos was being transported into the cells. This conclusion is consistent with the notion that opp alleles encode partly functional proteins. Sporulation. The partly functional opp mutations had a range of effects on sporulation. We measured sporulation frequencies for each of the missense mutants in DS medium and compared the frequencies to those of the wild type and an opp-null mutant (Table 2). The oppC-A266V and oppD-E239K mutants had large defects in sporulation, comparable to the defect in the oppA-null mutant. The oppD-E239K mutant also had a high level of bialaphos resistance, but the oppC-A266V mutant had a very low level of bialaphos resistance (Table 2 and Fig. 2). Thus, there is no correlation between a defect in bialaphos resistance (transport) and sporulation. The oppA alleles caused a range of sporulation defects and, again, the severity of the defect did not correlate with the level of resistance to bialaphos (Table 2 and Fig. 2). oppA-G337R (in the peptide binding pocket) and oppA-E259K (in domain I, apparently located away from the binding pocket) resulted in approximately wild-type levels of sporulation. In contrast, oppA-R443H (predicted to affect an interaction with the C terminus of some peptides), resulted in a low level of sporulation. The other three oppA mutants had levels of sporulation intermediate between those of the wild type and the oppA-null mutant. Competence gene expression. The partly functional opp mutants also had a range of effects on competence gene expression, from causing only slight delays to causing severely reduced expression (Fig. 3 and Table 2). We measured the expression of srfA-lacZ and comG-lacZ in each of the mutants grown in defined minimal medium. srfA is expressed in all cells and is controlled by the response regulator transcription factor ComA (6, 17; see also references therein). ComA acquires phosphate from the membrane-bound receptor kinase ComP, and the activity of ComP is stimulated by ComX pheromone.

VOL. 185, 2003


6429

FIG. 2. Inhibition of growth in the presence of different concentrations of toxic tripeptide bialaphos. Strains containing the indicated opp allele were grown in defined minimal medium to mid-exponential phase and diluted into fresh medium with the indicated concentrations of bialaphos. Cultures were grown at 37°C for 3 h, and the optical density was determined. Relative growth is the optical density achieved in the indicated concentration of bialaphos divided by the optical density of a parallel culture grown in the absence of bialaphos. Similar results were obtained in multiple experiments. Data for the wild type (wt) (JMS289) and the oppA-null mutant (JMS319) are repeated in each panel for comparison. Strains containing the opp missense mutants were Not201 (oppA-G337R), Not202 (oppA-E259K), Not204 (oppA-R443H), Not206 (oppA-G466E), Not210 (oppA-D449N), Not213 (oppA-G299E), Not207 (oppC-266V), and Not208 (oppD-E239K).

The activity of ComA is also modulated by CSF, the pentapeptide derived from the phrC gene product. CSF (and perhaps peptides derived from other phr genes) is transported into the cell via the oligopeptide permease, where it apparently inhibits the activity of RapC, a negative regulator of ComA. The expression of comG depends on the expression of comS, a small open reading frame in the srfA operon (5, 11), and comG expression is a good indicator of the level of competence development (6; see also references therein). As expected, there was a general correlation among the effects on the expression of srfA and comG (Fig. 3), consistent with the known regulatory pathway for competence development. oppA-G337R (Fig. 3A and B), oppA-E259K (Fig. 3A and B), oppA-G466E (Fig. 3C and D), and oppA-G299E (Fig. 3E and F) caused only a slight delay in competence gene expression. The two oppA mutations that affect putative charge interactions with the N and C termini of some tri- and tetrapeptides, oppA-R443H (Fig. 3C and D) and oppA-D449N (Fig. 3E and F), had the most severe effects on competence gene expression. It is not clear whether these residues are involved in pentapeptide binding, but these results indicate that they might be. The oppC-A266V mutation caused little if any decrease in competence gene expression (Fig. 3G and H). This mutation

also had the mildest effect on bialaphos resistance (Fig. 2 and Table 2). The oppD-E239K mutation caused only a slight to moderate delay in competence gene expression (Fig. 3G and H), even though it had a marked effect on bialaphos resistance (Fig. 2 and Table 2). These results indicate that there is no obvious correlation among defects in bialaphos transport, sporulation, and competence gene expression in the various partly functional opp mutants. Response to CSF (ERGMT). We tested the ability of each of the partly functional opp mutants to respond to CSF, a pentapeptide with the sequence ERGMT (18, 37). Cells were assayed for the amounts of srfA-lacZ expression induced by various concentrations of CSF (Fig. 4). As described previously (18, 37), in wild-type cells, the accumulation of ␤-galactosidase specific activity from an srfA-lacZ fusion increases approximately threefold 70 min after the addition of 5 to 10 nM CSF (Fig. 4A). Higher concentrations of CSF cause less stimulation until expression is inhibited (Fig. 4A). This inhibitory effect is not understood but is dependent on the import of CSF and is independent of RapC, the target of the stimulatory effect (18, 37). oppA-null mutants do not show any stimulation (or inhibition) of srfA expression in response to CSF (Fig. 4A). Several of the partly functional opp mutants were defective

6430

SOLOMON ET AL.

J. BACTERIOL.

FIG. 3. Expression of srfA-lacZ and comG-lacZ. Cells containing the indicated opp alleles and the indicated fusions were grown in defined minimal medium for at least three generations before the start of the experiment. Samples were taken at the indicated cell densities, and ␤-galactosidase specific activity was determined. Similar results were obtained in at least three independent experiments. Strains containing the srfA-lacZ fusion were JRL293 (opp⫹), JRL291 (⌬oppA), and Not371 to Not378 (each containing an opp missense mutation). Strains containing the comG-lacZ fusion were as in Fig. 2. Data for the wild type and the oppA-null mutants in panels A, B, E, and F are repeated in panels C, D, G, and H, respectively.

VOL. 185, 2003


6431

FIG. 4. Response to CSF (PhrC). Cells containing an srfA-lacZ fusion and the indicated opp allele were grown in defined minimal medium to an OD600 of 0.05 to 0.1. Cells were added to fresh medium containing the indicated concentration of CSF, incubated at 37°C for 70 min, and then assayed for ␤-galactosidase specific activity. Similar results were obtained in at least three independent experiments. Data for the wild type (wt) (JRL293) and the oppA-null mutant (JRL291) are repeated in each panel for comparison. ⌬oppA (multiplication signs) is labeled only in panel A. The opp missense mutant strains are the same as those containing the srfA-lacZ fusion in Fig. 3.

in responding to CSF (Fig. 4). However, the defects did not correlate strictly with the alterations in the expression of srfA or comG (Table 2 and Fig. 3 and 4). The oppA-D449N and oppA-R443H mutants (whose mutations were predicted to affect charge interactions with the N and C termini of some triand tetrapeptides) showed little or no response to CSF, similar to that of an oppA-null mutant (Fig. 4B). This finding is consistent with the notion that the affected residues are also involved in pentapeptide binding. Whereas these two mutants were unable to respond to CSF, they were only partly impaired in the expression of srfA-lacZ (Table 2 and Fig. 3). This finding is consistent with the proposal that peptides in addition to CSF contribute to the activation of the expression of srfA. oppA-E259K, oppA-G299E, and oppA-G466E also impaired the ability to respond to CSF (Fig. 4D) but had little if any effect on the expression of srfA (Table 2 and Fig. 3). These mutations did not alter the concentration of CSF at which a response is maximal but did alter the amplitude of the response. Finally, oppA-G337R and the mutations in the poreforming protein, oppC-A226V, and the ATP binding protein, oppD-E239K, had little or no effect on the response to CSF (Fig. 4C) as well as little or no effect on the expression of srfA (Table 2 and Fig. 3).

DISCUSSION We isolated and characterized rare partly functional mutations in opp of B. subtilis. The mutations all confer some level of resistance to the toxic tripeptide bialaphos, and the mutants retain at least partial function for sporulation and/or genetic competence. However, although they retain partial function, most of the mutants are also partly defective in sporulation or competence gene expression. The effects on sporulation and competence most likely are due to defects in the ability to transport signaling peptides derived from the phr family of genes. Each of the opp mutations probably affects the uptake of different signaling peptides. Some of the mutations probably alter the specificity of transport, reducing or preventing the uptake of some peptides while having relatively little effect on the uptake of others. At least three peptides, PhrA, PhrE, and CSF (PhrC), are known to affect sporulation (15, 26, 27). Several of the partly functional opp mutants (oppC-A266V, oppD-E239K, oppAG299E, and oppA-R443H) are more severely affected for sporulation than for competence gene expression. These mutants likely are defective in the uptake of peptides important for sporulation but still able to import peptides that stimulate competence gene expression.

6432

SOLOMON ET AL.

The only Phr-derived peptide known to affect competence gene expression is CSF (PhrC). However, the defect in competence caused by a phrC-null mutation is much less severe than that caused by an opp-null mutation (18, 37). This difference implies that signaling peptides in addition to CSF might be imported by the oligopeptide permease to stimulate ComA activity and competence gene expression. Consistent with this notion, we found no direct correlation between the response to CSF and the defect in the expression of srfA in the opp mutants (Table 2 and Fig. 3 and 4). We postulate that opp mutants (e.g., oppA-G299E and oppA-G466E) that are defective in the response to CSF but that have relatively normal expression of srfA (and comG) are able to import other signaling peptides that stimulate the activity of ComA. The identities of these peptides are presently not known. Two of the eight mutations that we characterized affect subunits other than the oligopeptide binding protein OppA. The membrane-spanning proteins of ABC transporters must interact with the transported molecule, and the mutation in oppC (oppC-A266V) could alter such an interaction. Alternatively, we suspect that the oppC-A266V and oppD-E239K mutations cause a more general defect in transport, perhaps affecting the ATPase cycle or communication between the substrate binding protein and the transport complex. If there is a more general defect in these mutants, then that defect has a much more severe effect on sporulation than on competence gene expression. The four mutations in oppA (oppA-G337R, oppA-R443H, oppA-D449N, and oppA-G466E) that change residues in the oligopeptide binding pocket probably directly alter the specificity of oligopeptide binding. The various effects of these mutations on bialaphos resistance, sporulation, and competence gene expression most likely reflect this altered specificity. The identification of additional signaling peptides affecting competence gene expression will help to clarify the effects of the various opp mutations. Furthermore, structural analysis of these mutant OppA proteins should be very useful for understanding the nature and specificity of peptide binding. ACKNOWLEDGMENTS We thank Meiji Seika Kaisha, Ltd., for the generous gift of bialaphos. We thank Beth Lazazzera for useful discussions, Cathy Lee for help with Fig. 1, and Cathy Lee and Jenny Auchtung for comments on the manuscript. J.S. was supported in part by a Howard Hughes predoctoral fellowship. This work was supported in part by Public Health Service grant GM50895 to A.D.G. from the NIH. REFERENCES 1. Albano, M., R. Breitling, and D. Dubnau. 1989. Nucleotide sequence and genetic organization of the Bacillus subtilis comG operon. J. Bacteriol. 171: 5386–5404. 2. Claverys, J. P., B. Grossiord, and G. Alloing. 2000. Is the Ami-AliA/B oligopeptide permease of Streptococcus pneumoniae involved in sensing environmental conditions? Res. Microbiol. 151:457–463. 3. Dassa, E., and P. Bouige. 2001. The ABC of ABCs: a phylogenetic and functional classification of ABC systems in living organisms. Res. Microbiol. 152:211–229. 4. Detmers, F. J. M., E. R. S. Kunji, F. C. Lanfermeijer, B. Poolman, and W. N. Konings. 1998. Kinetics and specificity of peptide uptake by the oligopeptide transport system of Lactococcus lactis. Biochemistry 37:16671–16679. 5. D’Souza, C., M. M. Nakano, and P. Zuber. 1994. Identification of comS, a gene of the srfA operon that regulates the establishment of genetic competence in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 91:9397–9401. 6. Dubnau, D., and C. M. J. Lovett. 2002. Transformation and recombination,

J. BACTERIOL.

7. 8. 9. 10. 11.

12. 13. 14. 15.

16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

30. 31. 32. 33. 34. 35.

p. 453–471. In A. L. Sonenshein, J. A. Hoch, and R. Losick (ed.), Bacillus subtilis and its closest relatives: from genes to cells. ASM Press, Washington, D.C. Dunten, P., and S. L. Mowbray. 1995. Crystal structure of the dipeptide binding protein from Escherichia coli involved in active transport and chemotaxis. Protein Sci. 4:2327–2334. Gominet, M., L. Slamti, N. Gilois, M. Rose, and D. Lereclus. 2001. Oligopeptide permease is required for expression of the Bacillus thuringiensis plcR regulon and for virulence. Mol. Microbiol. 40:963–975. Guex, N., and M. C. Peitsch. 1997. SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modelling. Electrophoresis 18:2714–2723. Hahn, J., L. Kong, and D. Dubnau. 1994. The regulation of competence transcription factor synthesis constitutes a critical control point in the regulation of competence in Bacillus subtilis. J. Bacteriol. 176:5753–5761. Hamoen, L. W., H. Eshuis, J. Jongbloed, G. Venema, and D. van Sinderen. 1995. A small gene, designated comS, located within the coding region of the fourth amino acid-activation domain of srfA, is required for competence development in Bacillus subtilis. Mol. Microbiol. 15:55–63. Harwood, C. R., and S. M. Cutting. 1990. Molecular biological methods for Bacillus. John Wiley & Sons, Inc., Chichester, England. Higgins, C. F. 2001. ABC transporters: physiology, structure and mechanism—an overview. Res. Microbiol. 152:205–210. Jiang, M., R. Grau, and M. Perego. 2000. Differential processing of propeptide inhibitors of Rap phosphatases in Bacillus subtilis. J. Bacteriol. 182:303– 310. Lazazzera, B., T. Palmer, J. Quisel, and A. D. Grossman. 1999. Cell density control of gene expression and development in Bacillus subtilis, p. 27–46. In G. M. Dunny and S. C. Winans (ed.), Cell-cell signaling in bacteria. ASM Press, Washington, D.C. Lazazzera, B. A. 2001. The intracellular function of extracellular signaling peptides. Peptides 22:1519–1527. Lazazzera, B. A., and A. D. Grossman. 1998. The ins and outs of peptide signaling. Trends Microbiol. 6:288–294. Lazazzera, B. A., J. M. Solomon, and A. D. Grossman. 1997. An exported peptide functions intracellularly to contribute to cell density signaling in B. subtilis. Cell 89:917–925. LeDeaux, J. R., J. M. Solomon, and A. D. Grossman. 1997. Analysis of non-polar deletion mutations in the genes of the spo0K (opp) operon of Bacillus subtilis. FEMS Microbiol. Lett. 153:63–69. Magnuson, R., J. Solomon, and A. D. Grossman. 1994. Biochemical and genetic characterization of a competence pheromone from B. subtilis. Cell 77:207–216. Miller, J. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Nodwell, J. R., K. McGovern, and R. Losick. 1996. An oligopeptide permease responsible for the import of an extracellular signal governing aerial mycelium formation in Streptomyces coelicolor. Mol. Microbiol. 22:881–893. Peitsch, M. C. 1995. Protein modeling by e-mail. Bio/Technology 13:658– 660. Peitsch, M. C. 1996. ProMod and Swiss-Model: Internet-based tools for automated comparative protein modelling. Biochem. Soc. Trans. 24:274– 279. Perego, M. 1997. A peptide export-import control circuit modulating bacterial development regulates protein phosphatases of the phosphorelay. Proc. Natl. Acad. Sci. USA 94:8612–8617. Perego, M. 1999. Self-signaling by Phr peptides modulates Bacillus subtilis development, p. 243–258. In G. M. Dunny and S. C. Winans (ed.), Cell-cell signaling in bacteria. ASM Press, Washington, D.C. Perego, M., and J. A. Brannigan. 2001. Pentapeptide regulation of aspartylphosphate phosphatases. Peptides 22:1541–1547. Perego, M., P. Glaser, and J. A. Hoch. 1996. Aspartyl-phosphate phosphatases deactivate the response regulator components of the sporulation signal transduction system in Bacillus subtilis. Mol. Microbiol. 19:1151–1157. Perego, M., C. Hanstein, K. M. Welsh, T. Djavakhishvili, P. Glasser, and J. A. Hoch. 1994. Multiple protein-aspartate phosphatases provide a mechanism for the integration of diverse signals in the control of development in B. subtilis. Cell 79:1047–1055. Perego, M., C. F. Higgins, S. R. Pearce, M. P. Gallagher, and J. A. Hoch. 1991. The oligopeptide transport system of Bacillus subtilis plays a role in the initiation of sporulation. Mol. Microbiol. 5:173–185. Perego, M., G. B. Spiegelman, and J. A. Hoch. 1988. Structure of the gene for the transition state regulator, abrB: regulator synthesis is controlled by the spo0A sporulation gene in Bacillus subtilis. Mol. Microbiol. 2:689–699. Picon, A., E. R. S. Kunji, F. C. Lanfermeijer, W. N. Konings, and B. Poolman. 2000. Specificity mutants of the binding protein of the oligopeptide transport system of Lactococcus lactis. J. Bacteriol. 182:1600–1608. Pottathil, M., and B. A. Lazazzera. 2003. The extracellular Phr peptide-Rap phosphatase signaling circuit of Bacillus subtilis. Frontiers Biosci. 8:32–45. Quentin, Y., G. Fichant, and F. Denizot. 1999. Inventory, assembly and analysis of Bacillus subtilis ABC transport systems. J. Mol. Biol. 287:467–484. Rudner, D. Z., J. R. LeDeaux, K. Ireton, and A. D. Grossman. 1991. The

VOL. 185, 2003


spo0K locus of Bacillus subtilis is homologous to the oligopeptide permease locus and is required for sporulation and competence. J. Bacteriol. 173:1388– 1398. 36. Sleigh, S., J. Tame, E. Dodson, and A. Wilkinson. 1997. Peptide binding in OppA, the crystal structures of the periplasmic oligopeptide binding protein in the unliganded form and in complex with lysyllysine. Biochemistry 36: 9747–9758. 37. Solomon, J. M., B. A. Lazazzera, and A. D. Grossman. 1996. Purification and characterization of an extracellular peptide factor that affects two different developmental pathways in Bacillus subtilis. Genes Dev. 10:2014–2024. 38. Solomon, J. M., R. Magnuson, A. Srivastava, and A. D. Grossman. 1995. Convergent sensing pathways mediate response to two extracellular competence factors in Bacillus subtilis. Genes Dev. 9:547–558.

39. Tame, J., E. Dodson, G. Murshudov, C. Higgins, and A. Wilkinson. 1995. The crystal structures of the oligopeptide-binding protein OppA complexed with tripeptide and tetrapeptide ligands. Structure 3:1395–1406. 40. Tame, J., G. Murshudov, E. Dodson, T. Neil, G. Dodson, C. Higgins, and A. Wilkinson. 1994. The structural basis of sequence-independent peptide binding by OppA protein. Science 264:1578–1581. 41. Vandeyar, M. A., and S. A. Zahler. 1986. Chromosomal insertions of Tn917 in Bacillus subtilis. J. Bacteriol. 167:530–534. 42. Vasantha, N., and E. Freese. 1980. Enzyme changes during Bacillus subtilis sporulation caused by deprivation of guanine nucleotides. J. Bacteriol. 144: 1119–1125. 43. Young, J., and I. B. Holland. 1999. ABC transporters: bacterial exportersrevisited five years on. Biochim. Biophys. Acta 1461:177–200.

6433