JOURNAL OF BACTERIOLOGY, June 2009, p. 3981–3991 0021-9193/09/$08.00⫹0 doi:10.1128/JB.00278-09 Copyright © 2009, American Society for Microbiology. All Rights Reserved.
Vol. 191, No. 12
RemA (YlzA) and RemB (YaaB) Regulate Extracellular Matrix Operon Expression and Biofilm Formation in Bacillus subtilis䌤† Jared T. Winkelman,‡ Kris M. Blair, and Daniel B. Kearns* Indiana University, Department of Biology, 1001 East Third Street, Bloomington, Indiana 47405 Received 2 March 2009/Accepted 14 March 2009
Biofilms are multicellular aggregates stabilized by an extracellular matrix. In Bacillus subtilis, the biofilm matrix is composed of an extracellular polysaccharide and the secreted protein TasA. Expression of both of the matrix components is repressed by the DNA-binding master regulator, SinR. Here we identify two small protein regulators of the extracellular matrix: RemA (formerly YlzA) and RemB (formerly YaaB). Mutation of RemA or RemB impairs pellicle formation, complex colony architecture, and motility inhibition in a sinR mutant background. Both proteins are required for the activation of the matrix biosynthesis operons and appear to act in parallel to SinR and two other known biofilm regulators, AbrB and DegU. which is apparently dependent on two kinases, KinC and KinD (14, 28, 35). Spo0A-phosphate, in turn, represses the biofilm repressor AbrB and activates the expression of a small protein SinI in a subpopulation of cells (11, 13, 22). SinI binds directly to SinR to antagonize SinR DNA binding and relieve repression of the biofilm operons (4, 25). Two other proteins of unknown function, YlbF and YmcA, are required for relief of SinR repression, but their mechanism of action is unknown (9, 25). The transcription factor DegU also plays a role in the regulation of biofilms, but its relationship to SinR and the control of biofilm structural operons is poorly understood (27, 47, 48). These regulators are integrated to ensure that SinR is temporally and spatially controlled within the three-dimensional structure of the biofilm (49). Two regulators of biofilm formation, SlrR and EpsE, are downstream of SinR control (Fig. 1). SlrR is a helix-turn-helix containing protein homologous to SinR that is required for full expression of the yqxM operon (13, 29). Expression of the gene that encodes SlrR is repressed by SinR. EpsE is a glycosyltransferase homolog encoded within the eps operon that inhibits motility by acting like a flagellar clutch. EpsE interacts with flagellar motor components and disconnects the rotor from the power source, thereby preventing flagellar rotation (6). Motility inhibition by EpsE may stabilize biofilm structure. Thus, not only are cells mutated for sinR hyperaggregated by constitutive derepression of the operons that encode biofilm matrix components, but sinR cells are also nonmotile due to derepression of the EpsE clutch protein. We describe here the identification of two additional proteins, RemA (YlzA) and RemB (YaaB) (for regulators of the extracellular matrix) that are required for biofilm formation in B. subtilis. Mutations in either gene reduce transcription of the eps operon, reduce levels of the motility inhibitor EpsE, and restore motility to a sinR mutant. Furthermore, our findings indicate that RemA and RemB act independently of any previously identified biofilm pathway. Thus, we provide a further level of complexity in the regulation of B. subtilis biofilm formation.
Biofilms are multicellular aggregates of bacteria. Every bacterial species seems to differ in the structure and regulation of biofilm formation, but most biofilms share three properties in common (31). First, biofilms are stabilized by an extracellular matrix composed of various combinations of polysaccharides, proteins, and DNA. Second, cells lose flagellar motility (if present) upon entry to the biofilm state. Third, a complex series of regulatory proteins governs the transition to biofilm formation. Here, we explore the regulation of biofilm formation in the gram-positive bacterium, Bacillus subtilis. In the laboratory, B. subtilis biofilms manifest either as pellicles that float atop liquid or as colonies of complex architecture that grow atop solid surfaces. These biofilms are stabilized by an extracellular matrix assembled from two primary structural components: an extracellular polysaccharide (EPS) and the protein TasA (7, 8). The EPS is assembled by enzymes encoded within the 15 gene epsA-O operon (25). The TasA protein is encoded by the terminal member of the yqxM operon (also known as the yqxM-sipW-tasA operon). The yqxM operon also encodes the signal peptidase SipW required for TasA secretion and the YqxM protein required for localization of TasA to the extracellular matrix (7, 43, 44, 45). Cells unable to properly synthesize either the EPS or TasA produce fragile pellicles and colonies with a smooth texture (7, 8, 22). The master regulator protein SinR binds to the promoters and represses expression of the eps and yqxM operons (12, 15, 25). Derepression of biofilm formation is complex, and many regulators act in a pathway upstream of SinR (Fig. 1). The pathway begins when the alternative sigma factor SigH is activated upon entry into stationary phase, and SigH drives the expression of the master regulator of sporulation, Spo0A (8, 21, 46). Spo0A is activated to a low level by phosphorylation,
* Corresponding author. Mailing address: Department of Biology, Indiana University, 1001 East Third St., Bloomington, IN 47405. Phone: (812) 856-2523. Fax: (812) 855-6705. E-mail:
[email protected]. † Supplemental material for this article may be found at http://jb .asm.org/. ‡ Present address: Department of Bacteriology, University of Wisconsin-Madison, 1550 Linden Dr., Madison, WI 53706. 䌤 Published ahead of print on 10 April 2009.
MATERIALS AND METHODS Strains and growth conditions. B. subtilis strains were grown in Luria-Bertani (LB; 10 g of tryptone, 5 g of yeast extract, and 5 g of NaCl/liter) broth or on LB
3981
3982
WINKELMAN ET AL.
FIG. 1. Model of B. subtilis biofilm matrix operon regulation. A genetic model is depicted with known biofilm regulators and their effect on eps and yqxM operon expression. T-bars indicate repression. Arrows indicate activation. Solid lines indicate a confirmed direct interaction. Dashed lines indicate that the interaction is poorly understood and could be either direct or indirect.
plates fortified with 1.5% Bacto agar at 37°C. When appropriate, antibiotics were included at the following concentrations: 10 g of tetracycline/ml, 100 g of spectinomycin/ml, 5 g of chloramphenicol/ml, 5 g of kanamycin/ml, and 1 g of erythromycin/ml plus 25 g of lincomycin/ml (mls). IPTG (isopropyl--Dthiogalactopyranoside; Sigma) was added to the medium at the indicated concentrations when appropriate. For pellicle formation experiments, 10 l of culture grown overnight at room temperature in LB medium was inoculated into 10 ml of minimal MSgg medium (5 mM potassium phosphate [pH 7], 100 mM morpholinepropanesulfonic acid [pH 7], 2 mM MgCl2, 700 M CaCl2, 50 M MnCl2, 50 M FeCl3, 1 M ZnCl2, 2 M thiamine, 0.5% glycerol, 0.5% glutamate, 50 g of tryptophan/ml, 50 g of phenylalanine/ml, and 50 g of threonine/ml) in six-well microtiter plates and incubated at 25°C (8). For colony architecture analysis, colonies were toothpick inoculated onto minimal MSgg medium fortified with 1.5% Bacto agar and incubated for 3 days at 25°C. For the motility assay, swarm agar plates containing 25 ml of LB medium fortified with 0.7% Bacto agar were prepared fresh and the following day were dried for 20 min in a laminar flow hood. Each plate was toothpick inoculated from an overnight colony and scored for motility after 18 h of incubation at 37°C (36). Plates were visualized by using a Bio-Rad Geldoc system and digitally captured by using Bio-Rad Quantity One software. Swarm expansion assay. Cells were grown to mid-log phase at 37°C in LB broth and resuspended to an optical density at 600 nm (OD600) of 10 in phosphate-buffered saline (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4; pH 8.0) containing 0.5% India ink (Higgins). Freshly prepared LB medium containing 0.7% Bacto agar (25 ml/plate) was dried for 20 min in a laminar flow hood, centrally inoculated with 10 l of the cell suspension, dried for another 10 min, and incubated at 37°C. The India ink demarks the origin of the colony, and the swarm radius was measured relative to the origin. For consistency, an axis was drawn on the back of the plate, and swarm radius measurements were taken along this transect. For experiments including IPTG, cells were propagated in broth in the presence of IPTG, and IPTG was included in the swarm agar plates. Western blotting. B. subtilis strains were grown in LB medium to an OD600 of ⬃1.0; 1-ml portions were harvested by centrifugation, resuspended to an OD600 of 10 in lysis buffer (20 mM Tris [pH 7.0], 10 mM EDTA, 1 mg of lysozyme/ml, 10 g of DNase I/ml, 100 g of RNase I/ml, 1 mM phenylmethylsulfonyl fluoride), and incubated 30 min at 37°C. Then, 10 l of lysate was mixed with 2 l of 6⫻ sodium dodecyl sulfate (SDS) loading dye. Samples were separated by SDS–12% polyacrylamide gel electrophoresis (PAGE). The proteins were elec-
J. BACTERIOL. troblotted onto nitrocellulose and developed with a 1:10,000 dilution of primary antibody (either anti-SigA, anti-EpsE, or anti-SinR) and a 1:10,000 dilution of secondary antibody (horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G). Immunoblots were developed by using an Immun-Star HRP developer kit (Bio-Rad). Strain construction. Unless otherwise indicated, all constructs were first introduced into the domesticated strain PY79 by natural competence and then transferred to the strain 3610 background by using SPP1-mediated generalized phage transduction (51). All strains used in the present study are listed in Table 1. All plasmids used in the present study are listed in Table S1 in the supplemental material. All primers used in the present study are listed in Table S2 in the supplemental material. (i) Complementation constructs. To generate the amyE::PremA-remA cat complementation construct pJW7, a PCR product containing the remA coding region plus ⬃500 bp of upstream sequence was amplified from B. subtilis 3610 chromosomal DNA by using the primer pair 1053/1054. The PCR product was digested with EcoRI and BamHI and cloned into the EcoRI and BamHI sites of pDG364 containing a polylinker and a chloramphenicol resistance cassette between two arms of the amyE gene (20). To generate the thrC::PremA-remA mls complementation construct pJW11, the same fragment was cloned into the EcoRI and BamHI sites of pDG1664, containing a polylinker and an mls resistance cassette between two arms of the thrC gene (20). To generate the amyE::PyloC-remA cat complementation construct pJW8, a PCR product containing the remA coding region was PCR amplified from B. subtilis 3610 chromosomal DNA by using the primer pair 1057/1054 and digested with XhoI and BamHI. A PCR product containing the putative promoter region upstream of yloC (PyloC) was PCR amplified from B. subtilis 3610 chromosomal DNA by using the primer pair 1055/1056 and digested with EcoRI and XhoI. The digested fragments containing PyloC and remA were then simultaneously ligated into the EcoRI and BamHI sites of pDG364. To generate the amyE::PremB-remB cat complementation construct pJW10, a PCR product containing the remB coding region plus ⬃700 bp of upstream sequence was amplified from B. subtilis 3610 chromosomal DNA by using the primer pair 1163/1164, digested with EcoRI and BamHI, and cloned into the EcoRI and BamHI sites of pDG364. To generate the amyE::PrecF-remB cat complementation construct pJW13, the PrecF promoter region was amplified by PCR from strain 3610 chromosomal DNA using the primer pair 1189/1190 and digested with EcoRI and XhoI. The remB gene was amplified from strain 3610 chromosomal DNA using the primer pair 1191/1164 and digested with XhoI and BamHI. The digested fragments containing PrecF and remB were then simultaneously ligated into the EcoRI/ BamHI sites of pDG364. To generate the thrC::PrecF-remB mls complementation construct pJW18, the two fragments were simultaneously ligated into the EcoRI/ BamHI sites of pDG1664. (ii) LacZ reporter constructs. To generate the -galactosidase (lacZ) reporter constructs pDP294, pDP296, pJW21, and pJW23, PCR products containing the following promoters were amplified from B. subtilis 3610 chromosomal DNA by using the primer pairs indicated in parentheses: PyloC (1055/1442), PremB (939/ 1443), PrecF (1300/1189), and PremA (1176/1053). Each PCR product was digested with EcoRI and BamHI and cloned independently into the EcoRI and BamHI sites of plasmid pDG268, which carries a chloramphenicol resistance marker and a polylinker upstream of the lacZ gene between two arms of the amyE gene (2). SPP1 phage transduction. To 0.2 ml of dense culture grown in TY broth (LB broth supplemented after autoclaving with 10 mM MgSO4 and 100 M MnSO4), serial dilutions of SPP1 phage stock were added, followed by static incubation for 15 min at 37°C. To each mixture, 3 ml of TYSA (molten TY supplemented with 0.5% agar) was added, poured atop fresh TY plates, followed by incubation at 30°C overnight. The top agar from the plate containing near confluent plaques was harvested by scraping into a 50-ml conical tube, vortexed, and centrifuged at 5,000 ⫻ g for 10 min. The supernatant was treated with 25 g of DNase/ml (final concentration) before being passed through a 0.45-m-pore-size syringe filter and stored at 4°C. Recipient cells were grown to stationary phase in 2 ml of TY broth at 37°C. Then, 0.9-ml portions of cells were mixed with 10 l of SPP1 donor phage stock. Afterward, 9 ml of TY broth was added to the mixture, which was allowed to stand at 37°C for 30 min. The transduction mixture was then centrifuged at 5,000 ⫻ g for 10 min, the supernatant was discarded, and the pellet was resuspended in the remaining volume. A total of 100 l of cell suspension was then plated onto TY fortified with 1.5% agar, the appropriate antibiotic, and 10 mM sodium citrate. -Galactosidase assay. Cells were harvested from cultures growing at 37°C in LB or MSgg broth. Cells were collected in 1-ml aliquots and suspended in equal volumes of Z buffer (40 mM NaH2PO4, 60 mM Na2HPO4, 1 mM MgSO4, 10 mM
B. SUBTILIS BIOFILM ACTIVATORS
VOL. 191, 2009
3983
TABLE 1. Strains examined in this studya Strain
Genotype
Reference
3610 DS92 DS177 DS1837 DS1839 DS1840 DS1855 DS1882 DS1928 DS1929 DS1930 DS2025 DS2315 DS2319 DS2320 DS2323 DS2324 DS2326 DS2566 DS2567 DS2594 DS2595 DS2617 DS2609 DS2641 DS2646 DS2647 DS2651 DS2679 DS2911 DS2912 DS2913 DS3330 DS3351 DS3352 DS3375 DS3374 DS3390 DS3391 DS3407 DS3629 DS3630 DS3631 DS3718 DS3741 DS3744 DS3745 DS3783 DS3908 DS4051 DS4052 DS4070 DS4141 DS4142 DS4143 DS4156 DS4160 DS4168 DS4173 DS4174 DS4175 DS4176 DS4260 DS4261 DS4618 DS4621
Undomesticated wild type sinR::spec abrB::mls epsH::tet sinR::kan sor2 关remB(L69Stop)兴 epsH::tet sinR::kan sor4 关remA(R18W)兴 epsH::tet sinR::kan sor5 amyE::Physpank-epsE epsH::tet amyE::Peps-lacZ cat epsH::tet sinR::kan sor2 关remB(L69Stop)兴 amyE::Physpank-epsE spec epsH::tet sinR::kan sor4 关remA(R18W)兴 amyE::Physpank-epsE spec epsH::tet sinR::kan sor5 amyE::Physpank-epsE spec degU::Tn10 spec epsH::tet amyE::Peps-lacZ cat epsH::tet sinR::spec sor20 关remB(⍀Aframeshift@codon73)兴 epsH::tet sinR::spec sor24 关remB(⍀Aframeshift@codon37)兴 epsH::tet sinR::kan sor25 epsH::tet sinR::spec sor28 关remB(⌬Aframeshift@codon72)兴 epsH::tet sinR::kan sor29 epsH::tet sinR::spec sor31 关remA(P29S)兴 sinR::spec epsH::tet amyE::Peps-epsE cat sinR::spec epsH::tet amyE::Peps-epsE cat pMarA kan mls sinR::spec epsH::tet amyE::Peps-epsE cat remB::TnYLB kan (TATATTCTG) sinR::spec epsH::tet amyE::Peps-epsE cat remB::TnYLB kan (TATTCTGAA) sinR::spec epsH::tet amyE::Peps-lacZ cat remB::TnYLB kan (TATATTCTG) sinR::spec epsH::tet amyE::Peps-lacZ cat (Kearns and Losick, 2005) remB::TnYLB kan (TATATTCTG) sinR::spec epsH::tet amyE::Peps-epsE cat remA::TnYLB kan (TACGTTCCC) sinR::spec epsH::tet amyE::Peps-epsE cat remA::TnYLB kan (TAAGACCGT) sinR::spec epsH::tet amyE::Peps-epsE cat remB::TnYLB kan (TATATATTC) remA::TnYLB kan (TACGTTCCC) sinR::spec epsH::tet amyE::Peps-lacZ cat remA::TnYLB kan (TACGTTCCC) remB::TnYLB kan epsH::tet amyE::Peps-lacZ cat remA::TnYLB kan epsH::tet amyE::Peps-lacZ cat remA::TnYLB kan (TACGTTCCC) amyE::PremA-remA cat epsH::tet remA::TnYLB kan (TACGTTCCC) amyE::PyqxM-lacZ cat epsH::tet amyE::PyqxM-lacZ cat remA::TnYLB kan (TACGTTCCC) amyE::PyloC-remA cat sinR::spec epsH::tet amyE::PyqxM-lacZ cat remA::TnYLB kan (TACGTTCCC) sinR::spec epsH::tet amyE::PyqxM-lacZ cat remB::TnYLB (TATATTCTG) sinR::spec epsH::tet amyE::PyqxM-lacZ cat epsH::tet remB::TnYLB kan (TATATTCTG) amyE::PyqxM-lacZ cat sinR::spec remA::TnYLB kan (TACGTTCCC) sinR::spec remA::TnYLB kan (TACGTTCCC) amyE::PremA-remA cat sinR::spec remA::TnYLB kan (TACGTTCCC) amyE::PyloC-remA cat sinR::spec remB::TnYLB kan (TATATTCTG) remB::TnYLB kan (TATATTCTG) amyE::PremB-remB cat sinR::spec epsH::tet amyE::Peps-lacZ cat remA::TnYLB kan (TACGTTCCC) thrC::PremA-remA mls sinR::spec epsH::tet amyE::PyqxM-lacZ cat remA::TnYLB kan (TACGTTCCC) thrC::PremA-remA mls sinR::spec remB::TnYLB kan (TATATTCTG) amyE::PremB-remB cat sinR::spec remB::TnYLB kan (TATATTCTG) amyE::PrecF-remB cat sinR::spec epsH::tet amyE::Peps-lacZ cat remB::TnYLB kan (TATATTCTG) thrC::PrecF-remB mls sinR::spec epsH::tet amyE::PyqxM-lacZ cat remB::TnYLB (TATATTCTG) thrC::PrecF-remB mls remB::TnYLB kan (TATATTCTG) amyE::PrecF-remB cat epsH::tet amyE::PrecF-lacZ cat sinR::spec epsH::tet amyE::PrecF-lacZ cat abrB::mls epsH::tet amyE::PrecF-lacZ cat degU::Tn10 spec epsH::tet amyE::PrecF-lacZ cat sinI::spec epsH::tet amyE::PrecF-lacZ cat epsH::tet amyE::PremA-lacZ cat sinI::spec epsH::tet amyE::PremA-lacZ cat sinR::spec epsH::tet amyE::PremA-lacZ cat degU::Tn10 spec epsH::tet amyE::PremA-lacZ cat abrB::mls epsH::tet amyE::PremA-lacZ cat sinR::spec epsH::tet thrC::Peps-lacZ mls remA::TnYLB kan (TACGTTCCC) amyE::PyloC-remA cat sinR::spec epsH::tet thrC::PyqxM-lacZ mls remA::TnYLB kan (TACGTTCCC) amyE::PyloC-remA cat sinI::spec epsH::tet amyE::Peps-lacZ cat epsH::tet amyE::PremB-lacZ cat
26 6 6 6 6 26
6 6 6 6 6 6
12
12
Continued on following page
3984
WINKELMAN ET AL.
J. BACTERIOL. TABLE 1—Continued
Strain
DS4622 DS4623 DS4624 DS4625 DS4627 DS4634 DS4635 DS4636 DS4637 DS4638 DS4639 DS4640 DS4641 DS4642 DS4643 DS4945 DS4946 DS4947 DS4948 DS4949 DS4950 DS4951 DS4952 DS4953 DS4954 DS4955 DS4956 DS4957 DS4958 DS4962 a
Genotype
Reference
remA::TnYLB kan epsH::tet amyE::PremA-lacZ cat remA::TnYLB kan epsH::tet amyE::PrecF-lacZ cat remB::TnYLB kan epsH::tet amyE::PremA-lacZ cat remB::TnYLB kan epsH::tet amyE::PrecF-lacZ cat epsH::tet amyE::PyloC-lacZ cat remA::TnYLB kan epsH::tet amyE::PremB-lacZ cat remB::TnYLB kan epsH::tet amyE::PremB-lacZ cat sinR::spec epsH::tet amyE::PremB-lacZ cat sinI::spec epsH::tet amyE::PremB-lacZ cat degU::Tn10 spec epsH::tet amyE::PremB-lacZ cat remA::TnYLB kan epsH::tet amyE::PyloC-lacZ cat remB::TnYLB kan epsH::tet amyE::PyloC-lacZ cat degU::Tn10 spec epsH::tet amyE::PyloC-lacZ cat sinI::spec epsH::tet amyE::PyloC-lacZ cat sinR::spec epsH::tet amyE::PyloC-lacZ cat degU::Tn10 spec remB::TnYLB kan (TATATTCTG) degU::Tn10 spec remA::TnYLB kan (TACGTTCCC) degU::Tn10 spec abrB::mls remB::TnYLB kan (TATATTCTG) abrB::mls remA::TnYLB kan (TACGTTCCC) abrB::mls epsH::tet amyE::Peps-lacZ cat abrB::mls epsH::tet amyE::PyloC-lacZ cat abrB::mls epsH::tet amyE::PyaaB-lacZ cat epsH::tet amyE::PslrR-lacZ cat sinR::spec epsH::tet amyE::PslrR-lacZ cat sinI::spec epsH::tet amyE::PslrR-lacZ cat degU::Tn10 spec epsH::tet amyE::PslrR-lacZ cat remA::TnYLB kan (TACGTTCCC) epsH::tet amyE::PslrR-lacZ cat abrB::mls epsH::tet amyE::PslrR-lacZ cat remB::TnYLB kan (TATATTCTG) epsH::tet amyE::PslrR-lacZ cat
All strains are in the undomesticated strain 3610 parental background. Sequences in parentheses indicate the sites of transposon insertion.
KCl, 38 mM 2-mercaptoethanol). Lysozyme was added to each sample to a final concentration of 0.2 mg ml⫺1, followed by incubation at 30°C for 15 min. Each sample was diluted in Z buffer to a final volume of 500 l, and the reaction was started with 100 l of 4 mg of 2-nitrophenyl -D-galactopyranoside ml⫺1 in Z buffer and stopped with 250 l of 1 M Na2CO3. The OD420 of the reaction mixture was measured, and the -galactosidase-specific activity was calculated as follows: [OD420/(time ⫻ OD600)] ⫻ dilution factor ⫻ 1,000. To generate bluewhite colonies, colonies were grown on MSgg medium supplemented with X-Gal (5-bromo-4-chloro-3-indolyl--D-galactopyranoside) for 48 h at 30°C and photographed with a Canon Powershot A620 digital camera. Transposon mutagenesis. The temperature-sensitive vector pMarA carrying TnYLB was introduced to strain DS2566 by SPP1-mediated generalized transduction to generate DS2567 (33). Next, a transposon library was generated by inoculating DS2567 into 3 ml of LB broth containing kanamycin and incubated at 22°C in a roller drum for 14 h. The culture was then serially diluted and spread on prewarmed LB plates fortified with 1.5% agar and kanamycin and incubated overnight at 42°C. The colonies were pooled and spotted in the center of a 0.7% swarm agar plate. Most of the cells remained nonmotile, but rare motile flares emerged from the central colony and were clonally isolated. Cells from the motile flares likely contained transposons linked to the mutation of interest. To confirm that the transposon was linked to the suppressor mutation, a lysate was generated on the suppressor mutant, and the transposon was transduced to the parent strain lacking the suppressor. Transposon insertion sites were identified by partially degenerate touchdown PCR using primer 766 and hybrid degenerate primer 749, 50 ng of purified chromosomal DNA, and Phusion polymerase (New England Biolabs) (34). Sequencing remA and remB. A PCR product containing the remA gene was amplified from B. subtilis chromosomal DNA (either from strain 3610 or the appropriate suppressor strain) by using the primer pair 906/907. The remA PCR product was then sequenced by using primers 906 and 907 individually. A PCR product containing the remB gene was amplified from B. subtilis chromosomal DNA (either from strain 3610 or the appropriate suppressor strain) by using the primer pair 884/885. The remB PCR product was then sequenced by using primers 884 and 885 individually.
RESULTS Suppressors that restore motility to sinR mutants reduce the levels of the motility inhibitor EpsE. SinR is a DNAbinding transcription factor that represses biofilm formation in B. subtilis (12, 15). Cells mutated for sinR are enhanced for biofilm formation and are also nonmotile (25). In previous work, 18 different suppressor of sinR (sor) mutations were isolated that restored motility in the absence of SinR. Nine of the sor strains had second-site mutations in the gene epsE, which encodes EpsE, a protein that disables flagella by uncoupling the FliG rotor from the MotA/MotB proton channel (6). The location of the nine other sor mutations, however, remained unknown and served as the starting point for this investigation. One way in which the unidentified sor mutations might restore motility is by reducing the levels of the motility inhibitor, EpsE. To determine EpsE levels in the suppressor strains, proteins from whole-cell lysates were resolved by SDS-PAGE, electroblotted, and probed with anti-EpsE polyclonal antibodies in Western blot analyses. EpsE protein was detectable in cells mutated for sinR, but EpsE protein was undetectable in a sinR mutant strain that also contained either an in-frame deletion of epsE or any of the remaining unidentified sor mutations (Fig. 2A). To determine whether the reduction of EpsE in the sor strains was sufficient to explain motility rescue, a copy of epsE was expressed from an artificial IPTG-inducible Physpank promoter and introduced at an ectopic locus (amyE::
VOL. 191, 2009
FIG. 2. Suppressor of sinR (sor) mutations restore motility by reducing levels of EpsE. (A) Western blot analysis of B. subtilis sor mutant cell lysates. Lysates were normalized and loaded according to cell number, resolved by SDS-PAGE, electroblotted, and separately probed with anti-SigA, anti-EpsE, and anti-SinR primary antibodies. SigA is the constitutively expressed housekeeping sigma factor and serves as a loading control. SlrR cross-reacts with anti-SinR antibodies. The following strains were used to generate this panel: DS1837, DS1839, DS1840, DS2315, DS2319, DS2320, DS2323, DS2324, and DS2326. (B) Swarm expansion assays of the wild type and sinR sor2, sinR sor4, and sinR sor5 mutants containing an IPTG-inducible epsE construct in the absence (open and gray symbols) or presence (solid symbols) of 1 mM IPTG. All points are the average of three replicas. The following strains were used to generate this panel: DS1855, DS1928, DS1929, and DS1930.
Physpank-epsE) in several of the sor mutant backgrounds. Each strain was motile in the absence of IPTG, but motility was inhibited when IPTG was included in the medium (Fig. 2B). We conclude that the sor mutations restored motility to the sinR parent by reducing the levels of the EpsE protein. We further conclude that because motility inhibition could be restored by expressing the epsE gene from a heterologous promoter, the genes disrupted by the unidentified sor mutations did not encode any essential component of motility inhibition that acts downstream of EpsE. The sor mutations disrupt previously unidentified activators of eps operon expression. One plausible explanation for the reduced levels of EpsE was that the sor mutations disrupted activators of the 15-gene eps operon, of which epsE is a member. To find activators of the eps operon, and thus candidate sites of the sor mutations, we designed a transposon-based
B. SUBTILIS BIOFILM ACTIVATORS
3985
genetic screen. A parental strain was generated in which two copies of epsE were transcribed redundantly from Peps promoters at two different locations (native and amyE), and both copies were derepressed by mutation of sinR. An epsH mutation was included to prevent EPS biosynthesis and cell aggregation that could confound mutant recovery. We hypothesized that this parental background might identify single transposon insertion mutations that abolish the expression of both copies of epsE and relieve motility inhibition by disrupting as-yetunidentified genetic trans activators of the Peps promoter. Colonies from a transposon insertion library generated in the parental background were pooled and spotted in the center of a soft agar petri plate. Most of the population remained nonmotile but, upon prolonged incubation, motile cells emerged from the central colony in the form of a flare. Five insertion mutations were independently isolated that had restored motility. For each suppressor, SPP1 phage-mediated generalized transduction backcross experiments demonstrated that the rescue-of-motility phenotype was inseparable from the transposon insertion. Two of the isolated strains had transposon insertions that disrupted the region immediately upstream of the gene remA (formerly ylzA), predicted to encode RemA, a broadly conserved 89-amino-acid protein of unknown function. The other three isolates had transposon insertions that disrupted the gene remB (formerly yaaB), predicted to encode RemB, a narrowly conserved 81-amino-acid protein of unknown function. Sequencing revealed that two of the nine unidentified sor strains had missense mutations in remA, while four other sor strains had nonsense or frameshift mutations in remB (Fig. 3 and see Fig. S1 in the supplemental material). The remaining three sor mutations were in neither remA nor remB and remain unidentified. We conclude that RemA and RemB are required for motility inhibition in a sinR mutant. We hypothesize, based on the rationale for the genetic screen, that these mutations restored motility by preventing sufficient levels of transcription from the Peps promoter. To directly measure the effects of remA or remB mutations on the level of transcription from the Peps promoter, a genetic reporter that placed expression of the gene encoding -galactosidase (lacZ) under the control of Peps was inserted at the ectopic amyE locus (amyE::Peps-lacZ). An epsH mutation was introduced to prevent cell aggregation that could confound accurate measurement of cell number and, consequently, gene expression. A sinR mutation increased expression of Peps-lacZ more than eightfold when grown in broth medium similar to that from which the sor mutants were originally isolated (LB medium) (Fig. 4A). Cells doubly mutated for sinR and either remA or remB exhibited 35- and 3-fold decreases in Peps-lacZ expression, respectively, relative to the sinR single mutant (Fig. 4A). We conclude that the RemA and RemB proteins encoded by the remA and remB genes, either directly or indirectly activate expression of the eps operon. We infer that RemA is the more potent of the two activators. RemA and RemB are global activators of biofilm formation. The eps operon encodes not only the motility inhibitor EpsE but also 14 other proteins required for EPS biosynthesis. Accordingly, reduction in eps operon expression has been correlated with defects in biofilm formation (25). To determine the consequence that mutations in remA and remB have on biofilm
3986
WINKELMAN ET AL.
J. BACTERIOL.
FIG. 3. remA and remB genetic regions. Open arrows indicate open reading frames. Bent arrows indicate the location of promoters. Circles are putative transcriptional terminators. Vertical arrows indicate the location of sor mutations (DS1837, DS1839, DS2315, DS2319, DS2323, and DS2326, see Fig. S1 in the supplemental material). Carets indicate the location of transposon insertions (DS2594, DS2595, DS2646, DS2647, and DS2651). The boundaries of the remA and remB complementation constructs are also indicated. A dashed line indicates a fusion between two genetic fragments.
phenotypes, cells were grown in biofilm-promoting MSgg medium in microtiter wells and on MSgg solid medium in petri plates. Cells mutated for sinR produced a robust and highly structured floating pellicle and hyper-rugose colonies due to the derepression of the eps operon (Fig. 5) (6, 25). A sinR remA double mutant produced no pellicle whatsoever and flat, featureless colonies (Fig. 5). A sinR remB double mutant produced a smooth featureless pellicle and less-rugose colonies
FIG. 4. Expression of biofilm matrix genes is reduced in remA and remB mutants. Open bars indicate the relative expression of the PepslacZ reporter construct. Gray bars indicate the relative expression of the PyqxM-lacZ reporter construct. All strains contain an epsH::tet allele to prevent cell clumping that might confound accurate measurement of cell number and enzyme activity. (A) Cells of the indicated genotypes were grown to an OD600 of 1.5 in LB medium, harvested, and assayed for -galactosidase expression levels. All values were normalized relative to the wild-type expression level of 1. Table S3 in the supplemental material contains all of the -galactosidase measurements used to generate this panel expressed as Miller units of activity. (B) Cells of the indicated genotypes were grown to an OD600 of 1.0 in MSgg medium, harvested, and assayed for -galactosidase expression levels. All values were normalized relative to the wild-type expression level of 1. Table S4 in the supplemental material contains all -galactosidase measurements used to generate this panel expressed as Miller units of activity.
(Fig. 5). Mutation of either remA or remB in otherwise wildtype cells resulted in defects in pellicle formation similar to the double mutants containing sinR (Fig. 5). We conclude that mutations in either remA or remB are epistatic to mutations in sinR. We further conclude that RemA is explicitly required for biofilm formation, whereas RemB appears to be a biofilm enhancer. Mutations in remA or remB reduced expression of the eps operon in LB medium that poorly supports biofilm formation (35). To more precisely correlate the biofilm phenotypes with gene expression defects, expression of the Peps-lacZ reporter was measured in biofilm-promoting MSgg media. Cells singly mutated for either remA or remB exhibited a 10-fold decrease in Peps-lacZ expression relative to wild type in liquid MSgg (Fig. 4B). To assess gene expression on a surface where biofilm formation occurs, wild-type colonies containing the Peps-lacZ reporter turned blue when grown on biofilm-promoting MSgg medium containing X-Gal (Fig. 6). The intensity of the blue color increased when the SinR repressor was mutated (Fig. 6). Conversely, blue color was not detectable due to constitutive repression by SinR when SinI, the antagonist of SinR, was mutated (Fig. 6). Consistent with the severe biofilm defects associated with remA, mutation of remA resulted in white colonies indicative of a dramatic decrease in Peps-lacZ expression (Fig. 6). Consistent with the more subtle biofilm defects associated with remB, mutation of remB resulted in a modest reduction in blue colony color (Fig. 6). We conclude that mutation of remA or remB resulted in a decrease in eps operon expression that was independent of medium composition (LB or MSgg) or state (liquid or solid). In addition to the EPS, the extracellular matrix of B. subtilis contains a protein, TasA (7). The gene encoding TasA is a member of the three-gene yqxM operon, the expression of which is also directly repressed by SinR (12, 44). To determine the effect of RemA and RemB on the transcription of the yqxM operon, we monitored the expression of a transcriptional reporter in which the lacZ gene was placed under the control of the PyqxM promoter and introduced at an ectopic locus (amyE:: PyqxM-lacZ). An epsH mutation was once again included to reduce cell aggregation and permit accurate cell density measurements. A sinR mutation increased expression from PyqxM
VOL. 191, 2009
B. SUBTILIS BIOFILM ACTIVATORS
3987
FIG. 6. Expression of remA and remB is largely independent of known biofilm regulators. Colonies contain the indicated lacZ reporter constructs (rows) and regulatory mutations (columns). “wt” indicates an otherwise wild-type strain that was mutated for epsH to abolish complex colony architecture. The epsH mutation was included in all strains in the grid to permit comparison of colonies and lacZ expression. The following strains were used to generate this figure: Peps series (DS2609, DS4618, DS2025, DS2913, DS2912, DS1882, and DS4950), PslrR series (DS4953, DS4954, DS4955, DS4956, DS4957, DS4958, and DS4962), PremA series (DS4174, DS4173, DS4175, DS4622, DS4624, DS4168, and DS4176), PyloC series (DS4643, DS4642, DS4641, DS4639, DS4640, DS4627, and DS4951), PremB series (DS4636, DS4637, DS4638, DS4634, DS4635, DS4621, and DS4952), and PrecF series (DS4142, DS4160, DS4156, DS4623, DS4625, DS4141, and DS4143).
FIG. 5. Cells mutated for remA and remB are defective in biofilm formation. The pellicle column depicts microtiter wells (six-well plate) in which cells have been grown in MSgg medium for 3 days at 25°C. The scale bar is 1 cm. The colony column depicts ⫻10 magnification images of individual colonies grown on MSgg medium for 3 days at 25°C. The scale bar is 1 mm. The motility column depicts petri plates containing LB medium and 0.7% agar, centrally inoculated and incubated at 37°C overnight. Swarm plates were filmed against a black background such that zones of bacterial colonization appear white and zones of uncolonized agar appears black. The scale bar is 2 cm. The indicated wild-type (strain 3610) and mutant strains are as follows: sinR (DS92), sinR remA (DS3629), sinR remA PremA-remA (DS3630), sinR remA PyloC-remA (DS3631), sinR remB (DS3718), sinR remB PremB-remB (DS3783), and sinR remB PrecF-remB (DS3908).
more than sixfold relative to the wild type (Fig. 4A). Relative to the sinR mutant, a sinR remA double mutant reduced expression more than 30-fold, and a sinR remB double mutant reduced expression more than 8-fold (Fig. 4A). As observed
with Peps, mutation of either remA or remB resulted in a dramatic reduction in PyqxM expression in otherwise wild-type cells (Fig. 4B). We conclude that RemA and RemB activate transcription of both operons that control synthesis of the biofilm extracellular matrix components. SlrR is a recently discovered protein that is homologous to SinR that directly or indirectly activates expression of the yqxM operon (13, 29). To determine whether RemA and RemB possibly affected yqxM operon expression by controlling SlrR protein levels, proteins from whole-cell lysates of the nine sor mutant strains were separated by using SDS-PAGE, electroblotted, and subjected to Western blot analysis with anti-SinR polyclonal antibodies. Antibodies raised against SinR have been shown to cross-react with SlrR due to the amino acid sequence similarity between the two proteins (13). SlrR was detectable in cells mutated for sinR and in cells doubly mutated for sinR and epsE (Fig. 2A). SlrR protein was reduced in all of the sor mutant strains except the as-yet-unidentified sor5. The reduction in SlrR protein levels in RemA or RemB mutants was correlated with a decrease in expression from a lacZ reporter fused to the slrR promoter when grown on solid MSgg medium containing X-Gal (Fig. 6). We conclude that, in addition to activating matrix biosynthesis operons, RemA and RemB also promote biofilm formation by activating expression and increasing the levels of the biofilm activator protein, SlrR. Thus, the effects of RemA and RemB on PyqxM expression are, at least in part, indirect through a reduction in SlrR. Since SlrR
3988
WINKELMAN ET AL.
does not affect the Peps promoter (13), RemA and RemB must activate the eps operon by another mechanism. Genetic complementation of remA and remB. Both the remA gene and the remB gene are in a genomic context such that they could be part of larger operons (Fig. 3). Thus, mutations in either gene, particularly the transposon insertions, could potentially cause polarity on downstream gene expression. To determine whether each gene was directly responsible for promoting motility inhibition and biofilm formation, genetic complementation analyses were conducted. To complement remA, the remA gene plus an ⬃500-bp region located immediately upstream (PremA-remA) was introduced at the ectopic amyE locus (Fig. 3). The PremA-remA complementation construct was partially functional: it restored motility inhibition to the sinR remA double mutant background but produced colonies and pellicles with less architectural complexity and failed to elevate Peps and PyqxM gene expression to levels found in a sinR mutant (Fig. 4A and 5). Colonies containing a -galactosidase reporter under the control of PremA (PremA-lacZ) were blue when grown on biofilm-promoting MSgg medium containing the chromogenic -galactosidase indicator X-Gal (Fig. 6). We conclude that remA is expressed from its own promoter but that this promoter is insufficient to fully complement all of the phenotypes of the remA mutant. To identify other potential promoters of remA, the remA gene was fused to a region from immediately upstream of yloC (PyloC-remA) and introduced at the ectopic amyE locus (Fig. 3). The amyE::PyloC-remA complementation construct fully rescued motility inhibition, pellicle formation, complex colony architecture, and Peps and PyqxM gene expression to the sinR remA mutant (Fig. 4A and 5). In addition, the amyE::PyloCremA complementation construct fully restored pellicle formation to a remA single mutant (see Fig. S2 in the supplemental material). Colonies containing a -galactosidase reporter under the control of PyloC (PyloC-lacZ) were blue when grown on biofilm-promoting MSgg medium containing X-Gal (Fig. 6). We conclude that there is a promoter immediately upstream of yloC and that remA may be cotranscribed with yloC under some circumstances. Furthermore, because we were able to achieve full complementation, we conclude that all remA mutant phenotypes were due to the disruption of remA and not due to polar effects on downstream genes. In an initial attempt to complement remB, the remB gene plus an ⬃700-bp region located immediately upstream (PremBremB) was introduced at the ectopic amyE locus (Fig. 3). The PremB-remB complementation construct did not rescue motility inhibition, pellicle formation, or complex colony architecture (Fig. 5). Furthermore, colonies containing a -galactosidase reporter under the control of PremB (PremB-lacZ) were white when grown on biofilm-promoting MSgg medium containing X-Gal (Fig. 6). We conclude that the region upstream of remB does not contain a functional promoter. In a second attempt to complement remB, the remB gene was fused to a region taken from immediately upstream of recF (PrecF-remB) and introduced at the ectopic amyE locus (Fig. 3). The PrecF-remB complementation construct rescued motility inhibition, pellicle formation, and complex colony architecture to a sinR remB double mutant (Fig. 5). In addition, the amyE:: PrecF-remB complementation construct fully restored pellicle formation to a remB single mutant (see Fig. S2 in the supple-
J. BACTERIOL.
mental material). Colonies containing a -galactosidase reporter under the control of PrecF (PrecF-lacZ) were blue when grown on biofilm-promoting MSgg medium containing X-Gal (Fig. 6). We conclude that the region immediately upstream of recF contains a promoter and we infer that remB is cotranscribed with recF. Finally, we note that, whereas the PrecF-remB construct appeared to completely complement the remB multicellular phenotypes, complementation of gene expression from the Peps and PyqxM promoters was incomplete (Fig. 4A). remA and remB expression is not controlled by SinR. One way in which remA and remB could be epistatic to sinR is if their expression was under SinR control. To determine whether the remA and remB genes were under the control of SinR, strains containing lacZ reporters relevant to either remA or remB expression were plated on MSgg biofilm-promoting media containing X-Gal. Colonies of strains containing lacZ genes expressed either from the remA-complementing promoter PremA or the remB-complementing promoter PrecF did not appear to be affected by mutation of sinR (Fig. 6). Colonies containing the lacZ gene expressed from the remA-complementing promoter PyloC (PyloC-lacZ) appeared to have a more intense blue color in the absence of SinR. To test whether PyloC (or any of the other promoters) was repressed by SinR, the reporter strain was mutated for SinI. Mutation of sinI appeared to have no effect on the expression of any of the reporters relevant to remA or remB, suggesting that none of the reporters were directly repressed by SinR (Fig. 6). We conclude that the expression of remA and remB is largely independent of SinR. Since remA and remB mutations are epistatic to mutations in sinR and neither gene is under SinR control, we conclude that RemA and RemB act independently and in parallel to the SinR pathway. Furthermore, disruption of neither RemA nor RemB appeared to have an effect on their own expression, suggesting that the two proteins are not autoregulatory (Fig. 6). Mutations in remA and remB are epistatic to mutations in abrB and degU. AbrB represses biofilm formation by repressing expression of the PyqxM promoter (13, 21). Mutation of abrB did not dramatically affect the expression from the PslrR, PremA, or PrecF promoters, but colonies containing Peps and PyloC reporters appeared to have a more intense blue color in the absence of AbrB (Fig. 6). Cells mutated for abrB were motile and formed rugose pellicles and colonies with rough architecture (Fig. 7). To determine the relationship between AbrB and the Rem proteins, double mutants were tested for multicellular phenotypes. Cells doubly mutated for abrB and remA displayed the remA phenotype with no pellicles and smooth colonies (Fig. 7). Cells doubly mutated for abrB and remB displayed the remB phenotype with smooth pellicles and flat colonies (Fig. 7). We conclude that mutations in remA and remB genes are epistatic to mutations in abrB. DegU activates and represses biofilm formation in a complex manner that depends on the DegU level and phosphorylation state (27, 47). Mutation of degU did not appear to have a dramatic effect on the expression of the Peps promoter, the PslrR promoter, or any promoter related to either remA or remB (Fig. 6). Cells mutated for degU formed smooth pellicles and colonies with rough architecture (Fig. 7). Consistent with published reports, cells mutated for DegU were nonmotile on surfaces for reasons unknown (Fig. 7) (27, 47). To determine
B. SUBTILIS BIOFILM ACTIVATORS
VOL. 191, 2009
FIG. 7. Mutations in remA and remB are epistatic to mutations in abrB and degU. The pellicle column depicts microtiter wells (six-well plate) in which cells have been grown in MSgg medium for 3 days at 25°C. The scale bar is 1 cm. The colony column depicts ⫻10 magnification images of individual colonies grown on MSgg medium for 3 days at 25°C. The scale bar is 1 mm. The motility column depicts petri plates containing LB medium and 0.7% agar, centrally inoculated and incubated at 37°C overnight. Swarm plates were filmed against a black background such that zones of bacterial colonization appear white and zones of uncolonized agar appear black. The scale bar is 2 cm. The indicated mutant strains are as follows: abrB (DS177), abrB remA (DS4949), abrB remB (DS4948), degU (DS4947), degU remA (DS4946), and degU remB (DS4945).
the relationship between DegU and the Rem proteins, double mutants were tested for multicellular phenotypes. Cells doubly mutated for degU and remA displayed the remA phenotype with no pellicles and flat colonies (Fig. 7). Cells doubly mutated for degU and remB displayed the remB phenotype with weaker pellicles and smooth colonies (Fig. 7). We conclude that mutations in remA and remB genes are epistatic to mutations in degU for biofilm formation. Thus, mutations of the genes encoding RemA and RemB are epistatic to mutations encoding all known major regulators of biofilm formation, including DegU, AbrB, and SinR. DISCUSSION Here we identify two proteins RemA and RemB that are required to activate the expression of matrix biosynthesis genes and biofilm formation in B. subtilis. Three lines of evidence support the conclusion that RemA and RemB act in parallel to the pathway that governs SinR derepression. First, mutations in remA or remB resulted in biofilm defects in otherwise wildtype cells, and thus the phenotypes did not depend on the
3989
absence of SinR (Fig. 5). Second, mutations in either remA or remB were epistatic to mutations in sinR (Fig. 5). Third, the expression of promoters required for complementation of remA and remB mutations did not appear to be under SinR control (Fig. 6). Thus, RemA and RemB appear to be the first proteins required for matrix operon expression in B. subtilis to act in parallel to SinR (Fig. 1). We conclude that RemA and RemB are fundamental to biofilm activation because they are also required in the absence of either AbrB or DegU (Fig. 7). The genes that encode RemA and RemB were identified as sites of bypass suppressors that restored motility to a SinR mutant. Artificial overexpression of the EpsE flagellar clutch protein restored motility inhibition to the sinR remA or sinR remB double mutants, suggesting that the two proteins acted upstream of EpsE and were not themselves part of the clutch mechanism. Instead, RemA and RemB controlled EpsE levels by activating transcription of the operon containing the epsE gene. Of 18 of the original suppressor of sinR (sor) mutations, 9 directly disrupted the motility inhibitor EpsE, and 6 disrupted either RemA or RemB that control EpsE protein levels. The locations of the remaining three sor mutations are still unknown, but these seem to control EpsE protein levels as well (Fig. 2A). Thus, EpsE appears to be the ultimate determinant of motility inhibition in sinR mutants. Mutation of remA resulted in a severe defect in biofilm formation due to a near complete loss of expression from the matrix synthesis operons. RemA is an 89-amino-acid protein that is highly conserved throughout the bacterial domain, more so than SinR, the eps operon, or TasA, suggesting that it may play a fundamental role in B. subtilis physiology. Interestingly, the genes in proximity to remA suggest a connection to the stringent response. Two genes are situated downstream of, and in the same direction as, remA: gmk encoding Gmk the essential enzyme guanylate kinase (18, 30) and rpoZ encoding the subunit of RNA polymerase (10, 16). The stringent response in B. subtilis is thought to be mediated through GTP levels, presumably controlled, at least in part, through nucleoside synthesis by Gmk (32, 36, 41). The subunit enhances the ability of E. coli RNA polymerase to respond to the modified GTP stringent response alarmone, ppGpp (17, 24, 50). Like the stringent response, biofilm formation in B. subtilis is thought to be triggered by nutrient depletion (8, 21, 46) and, like RemA, the subunit has been shown to be required for extracellular matrix production in a gram-positive bacterium, Mycobacterium smegmatis (37). The remA gene appears to be expressed from two promoters. Complementation of all remA mutant phenotypes required that remA be expressed from the PyloC promoter region, suggesting that remA may, under some circumstances, be cotranscribed with yloC, encoding YloC, a 291-amino-acid protein of unknown function. No obvious rho-independent terminator resides in the 78 bp that separate yloC and remA, supporting the idea that yloC and remA may be cotranscribed. Nonetheless, the remA gene is expressed independently from the auxillary promoter PremA, which supported motility inhibition by EpsE but not biofilm formation during complementation analysis. If the two promoters that express remA are controlled separately, then motility inhibition by EpsE may be uncoupled from biofilm formation under some conditions. RemB is a supplemental enhancer of biofilm gene expres-
3990
WINKELMAN ET AL.
sion. Mutation of remB impaired biofilm formation, reduced the levels of EpsE, and restored motility to a sinR mutant with only a minor reduction in the expression of the eps operon (⬃3-fold, Fig. 4A). Thus, it is possible that activation of the eps operon is not the primary function of RemB. We suspect that RemB may have other roles in the cell because of its proximity to the origin and genes involved in chromosome replication and maintenance. The remB gene is located less than 5 kb from the origin of replication and is likely to be cotranscribed with recF, encoding the RecF DNA repair protein (1, 3, 42). No obvious rho-independent terminator resides in the 19 bp that separate recF and remB, supporting the idea that recF and remB are cotranscribed. Immediately downstream of the remB gene is gyrB, encoding the essential GyrB DNA gyrase (19, 30, 38). RemB is an 81-amino-acid protein and, unlike its genetic neighbors, is conserved only in the low-G⫹C gram-positive bacteria. The mechanisms by which RemA and RemB activate gene expression are unknown. RemA and RemB could act as transcription factors that bind DNA directly, but neither protein is predicted to contain a helix-turn-helix DNA-binding motif. Instead, they may activate gene expression indirectly by interaction with another protein. For example, RemA or RemB might function in a fashion similar to the paradigm of SinI (57 amino acids) and AbbA (65 amino acids) that interact with, and antagonize DNA binding of, the repressor proteins SinR and AbrB, respectively (4, 5). Alternatively, some small regulatory proteins are known to interact with RNA polymerase directly. In E. coli, the stringent response is potentiated by the nonclassical transcription factor DksA (39). DksA binds in the secondary channel of RNA polymerase and can act to positively or negatively regulate transcription depending on the kinetic properties of the given promoter (23, 40). RemA and RemB may act by different mechanisms because, in all assays, the consequences of mutation of remA were more severe than mutation of remB. The proteins that regulate biofilm matrix gene expression in B. subtilis indirectly indicate the types of environmental stimuli that control biofilm formation. For example, stationary-phase proteins such as SigH and Spo0A are part of a pathway involved in derepression of SinR and indicate that starvation plays an important role in promoting biofilms. Here we identify two small proteins, RemA and RemB, that regulate the extracellular matrix in parallel to the SinR derepression pathway. The requirement of parallel regulators suggests that activation of the extracellular matrix integrates multiple stimuli. By elucidating the mechanism of RemA and RemB and by determining their regulation, we may deduce additional environmental or physiological inputs on B. subtilis biofilm formation. ACKNOWLEDGMENTS We thank Loralyn Cozy, Joyce Patrick, and Malcolm Winkler for experimental and technical support. We are especially grateful to Melissa Parrot for generating genetic constructs and conducting assays. We thank Rich Losick for critical reading of the manuscript and helpful discussion. This study was supported by NSF grant MCB-0721187 to D.B.K. REFERENCES 1. Alonso, J. C., and G. Lu ¨der. 1991. Characterization of recF suppressors in Bacillus subtilis. Biochimie 73:277–280.
J. BACTERIOL. 2. Antoniewski, C., B. Savelli, and P. Stragier. 1990. The spoIIJ gene, which regulates early developmental steps in Bacillus subtilis, belongs to a class of environmentally responsive genes. J. Bacteriol. 172:86–93. 3. Ayora, S., and J. C. Alonso. 1997. Purification and characterization of the RecF protein from Bacillus subtilis 168. Nucleic Acids Res. 25:2766–2772. 4. Bai, U., I. Mandic-Mulec, and I. Smith. 1993. SinI modulates the activity of SinR, a developmental switch protein of Bacillus subtilis, by protein-protein interaction. Genes Dev. 7:139–148. 5. Banse, A. V., A. Chastanet, L. Rahn-Lee, E. C. Hobbs, and R. Losick. 2008. Parallel pathways of repression and antirepression governing the transition to stationary phase in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 105:15547– 15552. 6. Blair, K. M., L. Turner, J. T. Winkelman, H. C. Berg, and D. B. Kearns. 2008. A molecular clutch disables flagella in the Bacillus subtilis biofilm. Science 320:1636–1638. 7. Branda, S. S., F. Chu, D. B. Kearns, R. Losick, and R. Kolter. 2006. A major component of the Bacillus subtilis biofilm matrix. Mol. Microbiol. 59:1229– 1238. 8. Branda, S. S., J. E. Gonza ´lez-Pastor, S. Ben-Yehuda, R. Losick, and R. Kolter. 2001. Fruiting Body Formation by Bacillus subtilis. Proc. Natl. Acad. Sci. USA 98:11621–11626. 9. Branda, S. S., J. E. Gonza ´lez-Pastor, E. Dervyn, S. D. Ehrlich, R. Losick, and R. Kolter. 2004. Genes involved in formation of structured multicellular communities by Bacillus subtilis. J. Bacteriol. 186:3970–3979. 10. Burgess, R. R. 1969. Separation and characterization of the subunits of RNA polymerase. J. Biol. Chem. 244:2168–2176. 11. Chai, Y., F. Chu, R. Kolter, and R. Losick. 2008. Bistability and biofilm formation in Bacillus subtilis. Mol. Microbiol. 67:254–263. 12. Chu, F., D. B. Kearns, S. S. Branda, R. Kolter, and R. Losick. 2006. Targets of the master regulator of biofilm formation in Bacillus subtilis. Mol. Microbiol. 59:1216–1228. 13. Chu, F., D. B. Kearns, A. McLoon, Y. Chai, R. Kolter, and R. Losick. 2008. A novel regulatory protein governing biofilm formation in Bacillus subtilis. Mol. Microbiol. 68:1117–1127. 14. Fujita, M., J. E. Gonza ´lez-Pastor, and R. Losick. 2005. High- and lowthreshold genes in the Spo0A regulation of Bacillus subtilis. J. Bacteriol. 187:1357–1368. 15. Gaur, N. K., J. Oppenheim, and I. Smith. 1991. The Bacillus subtilis sin gene, a regulator of alternate developmental processes, codes for a DNA-binding protein. J. Bacteriol. 173:678–686. 16. Gentry, D. R., and R. R. Burgess. 1989. rpoZ, encoding the omega subunit of Escherichia coli RNA polymerase, is in the same operon as spoT. J. Bacteriol. 171:1271–1277. 17. Gentry, D., H. Xiao, R. Burgess, and M. Cashel. 1991. The omega subunit of Escherichia coli K-12 RNA polymerase is not required for stringent RNA control in vivo. J. Bacteriol. 173:3901–3903. 18. Gentry, D., C. Bengra, K. Ikehara, and M. Cashel. 1993. Guanylate kinase of Escherichia coli K-12. J. Biol. Chem. 268:14316–14321. 19. Go ¨ttler, T., and D. Klostermeier. 2007. Dissection of the nucleotide cycle of B. subtilis DNA gyrase and its modulation by DNA. J. Mol. Biol. 367:1392– 1404. 20. Gue´rout-Fleury, A.-M., N. Frandsen, and P. Stragier. 1996. Plasmids for ectopic integration in Bacillus subtilis. Gene 180:57–61. 21. Hamon, M. A., and B. A. Lazazzera. 2001. The sporulation transcription factor Spo0A is required for biofilm development in Bacillus subtilis. Mol. Microbiol. 42:1199–1209. 22. Hamon, M. A., N. R. Stanley, R. A. Britton, A. D. Grossman, and B. A. Lazazzera. 2004. Identification of AbrB-regulated genes involved in biofilm formation by Bacillus subtilis. Mol. Microbiol. 52:847–860. 23. Haugen, S. P., W. Ross, M. Manrique, and R. L. Gourse. 2008. Fine structure of the promoter-sigma region 1.2 interaction. Proc. Natl. Acad. Sci. USA 105:3293–3297. 24. Igarashi, K., N. Fujita, and A. Ishihama. 1986. Promoter selectivity of Escherichia coli RNA polymerase: omega factor is responsible for the ppGpp sensitivity. Nucleic Acids Res. 17:8755–8765. 25. Kearns, D. B., F. Chu, S. S. Branda, R. Kolter, and R. Losick. 2005. A master regulator of biofilm formation by Bacillus subtilis. Mol. Microbiol. 55:739– 749. 26. Kearns, D. B., and R. Losick. 2003. Swarming motility in undomesticated Bacillus subtilis. Mol. Microbiol. 49:581–590. 27. Kobayashi, K. 2007. Gradual activation of the response regulator DegU controls serial expression of genes for flagellum formation and biofilm formation in Bacillus subtilis. Mol. Microbiol. 66:395–409. 28. Kobayashi, K., R. Kuwana, and H. Takamatsu. 2008. kinA mRNA is missing a stop codon in the undomesticated Bacillus subtilis strain ATCC 6051. Microbiol. 154:54–63. 29. Kobayashi, K. 2008. SlrR/SlrA controls the initiation of biofilm formation in Bacillus subtilis. Mol. Microbiol. 69:1399–1410. 30. Kobayashi, K., S. D. Ehrlich, A. Albertini, G. Amati, K. K. Andersen, M. Araund, K. Asai, et al. 2003. Essential Bacillus subtilis genes. Proc. Natl. Acad. Sci. USA 100:4678–4683.
VOL. 191, 2009 31. Kolter, R., and E. P. Greenberg. 2006. The superficial life of microbes. Nature 441:300–303. 32. Kra ´sny´, L., and R. L. Gourse. 2004. An alternative strategy for bacterial ribosome synthesis: Bacillus subtilis rRNA transcription regulation. EMBO J. 23:4473–4483. 33. Le Breton, Y., N. P. Mohapatra, and W. G. Haldenwang. 2006. In vivo random mutagenesis of Bacillus subtilis by use of TnYLB-1 a mariner-based transposon. Appl. Environ. Microbiol. 72:327–333. 34. Levano-Garcia, J., S. Verjovski-Almeida, and A. C. R. da Silva. 2005. Mapping transposon insertion sties by touchdown PCR and hybrid degenerate primers. BioTechniques 38:225–229. 35. Lo ´pez, D., M. A. Fischbach, F. Chu, R. Losick, and R. Kolter. 2009. Structurally diverse natural products that cause potassium leakage trigger multicellularity in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 106:280–285. 36. Lopez, J. M., C. L. Marks, and E. Freese. 1979. The decrease of guanine nucleotides initiates sporulation of Bacillus subtilis. Biochim. Biophys. Acta 587:238–252. 37. Mathew, R., R. Mukherjee, R. Balachandar, and D. Chatterji. 2006. Deletion of the rpoZ gene, encoding the subunit of RNA polymerase, results in pleiotropic surface-related phenotypes in Mycobacterium smegmatis. Microbiology 152:1741–1750. 38. Orr, E., and W. L. Staudenbauer. 1982. Bacillus subtilis DNA gyrase: purification of subunits and reconstitution of supercoiling activity. J. Bacteriol. 151:524–527. 39. Paul, B. J., M. M. Barker, W. Ross, D. A. Schneider, C. Webb, J. W. Foster, and R. L. Gourse. 2004. DksA: a critical component of the transcription initiation machinery that potentiates the regulation of rRNA promoters by ppGpp and initiating NTP. Cell 118:311–322. 40. Paul, B. J., M. B. Berkman, and R. L. Gourse. 2005. DksA potentiates direct activation of amino acid promoters by ppGpp. Proc. Natl. Acad. Sci. USA 102:7823–7828.
B. SUBTILIS BIOFILM ACTIVATORS
3991
41. Ratnayake-Lecamwasam, M., P. Serror, K.-W. Wong, and A. L. Sonenshein. 2001. Bacillus subtilis CodY represses early-stationary-phase genes by sensing GTP levels. Genes Dev. 15:1093–1103. 42. Sanchez, H., B. Carrasco, M. C. Cozar, and J. C. Alonso. 2007. Bacillus subtilis RecG branch migration translocase is required for DNA repair and chromosomal segregation. Mol. Microbiol. 65:920–935. 43. Sto ¨ver, A. G., and A. Driks. 1999. Secretion, localization, and antibacterial activity of TasA, a Bacillus subtilis spore-associated protein. J. Bacteriol. 181:1664–1672. 44. Sto ¨ver, A. G., and A. Driks. 1999. Regulation of synthesis of the Bacillus subtilis transition-phase, spore associated antibacterial protein TasA. J. Bacteriol. 181:5476–5481. 45. Sto ¨ver, A. G., and A. Driks. 1999. Control of synthesis and secretion of the Bacillus subtilis protein YqxM. J. Bacteriol. 181:7065–7069. 46. Veening, J. W., O. P. Kuipers, S. Brul, K. J. Hellingwerf, and R. Kort. 2006. Effects of phosphorelay perturbations on architecture, sporulation, and spore resistance in biofilms of Bacillus subtilis. J. Bacteriol. 188:3099–3109. 47. Verhamme, D. T., T. B. Kiley, and N. R. Stanley-Wall. 2007. DegU coordinates multicellular behavior exhibited by Bacillus subtilis. Mol. Microbiol. 65:554–568. 48. Verhamme, D. T., E. J. Murray, and N. R. Stanley-Wall. 2009. DegU and Spo0A jointly control transcription of two loci required for complex colony development by Bacillus subtilis. J. Bacteriol. 191:100–108. 49. Vlamakis, H., C. Aguilar, R. Losick, and R. Kolter. 2008. Control of cell fate by the formation of an architecturally complex bacterial community. Genes Dev. 22:945–953. 50. Vrentas, C. E., T. Gaal, W. Ross, R. H. Ebright, and R. L. Gourse. 2005. Response of RNA polymerase to ppGpp: requirement for the subunit and relief of this requirement by DksA. Genes Dev. 19:2378–2387. 51. Yasbin, R. E., and F. E. Young. 1974. Transduction in Bacillus subtilis by bacteriophage SPP1. J. Virol. 14:1343–1348.
