This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wei, J.-R.
Right arrow Articles by Lai, H.-C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wei, J.-R.
Right arrow Articles by Lai, H.-C.

 Previous Article  |  Next Article 

Journal of Bacteriology, August 2005, p. 5683-5690, Vol. 187, No. 16
0021-9193/05/$08.00+0     doi:10.1128/JB.187.16.5683-5690.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Biochemical Characterization of RssA-RssB, a Two-Component Signal Transduction System Regulating Swarming Behavior in Serratia marcescens

Jun-Rong Wei,1 Yu-Huan Tsai,1 Po-Chi Soo,1 Yu-Tze Horng,1 Shang-Chen Hsieh,1 Shen-Wu Ho,1,2 and Hsin-Chih Lai1,2*

Department of Clinical Laboratory Sciences and Medical Biotechnology, National Taiwan University College of Medicine, Taipei, Taiwan, Republic of China,1 Department of Laboratory Medicine, National Taiwan University Hospital, Taipei, Taiwan, Republic of China2

Received 17 February 2005/ Accepted 16 May 2005


arrow
ABSTRACT
 
Our previous study had identified a pair of potential two-component signal transduction proteins, RssA-RssB, involved in the regulation of Serratia marcescens swarming. When mutated, both rssA and rssB mutants showed precocious swarming phenotypes on LB swarming agar, whereby swarming not only occurred at 37°C but also initiated on a surface of higher agar concentration and more rapidly than did the parent strain at 30°C. In this study, we further show that the predicted sensor kinase RssA and the response regulator RssB bear characteristics of components of the phosphorelay signaling system. In vitro phosphorylation and site-directed mutagenesis assays showed that phosphorylated RssA transfers the phosphate group to RssB and that histidine 248 and aspartate 51 are essential amino acid residues involved in the phosphotransfer reactions in RssA and RssB, respectively. Accordingly, while wild-type rssA could, the mutated rssA(H248A) in trans could not complement the precocious swarming phenotype of the rssA mutant. Although RssA-RssB regulates expressions of shlA and ygfF of S. marcescens (ygfFSm), in vitro DNA-binding assays showed that the phosphorylated RssB did not bind directly to the promoter regions of these two genes but bound to its own rssB promoter. Subsequent assays located the RssB binding site within a 63-bp rssB promoter DNA region and confirmed a direct negative autoregulation of the RssA-RssB signaling pathway. These results suggest that when activated, RssA-RssB acts as a negative regulator for controlling the initiation of S. marcescens swarming.


arrow
INTRODUCTION
 
To unravel the underlying regulatory mechanism of Serratia marcescens swarming, we had utilized a mini-Tn5 mutagenesis approach to discover a group of S. marcescens mutant strains that demonstrated precocious swarming at both 30 and 37°C (16). A pair of potential bacterial two-component signal transduction proteins (11, 27), RssA-RssB, had been identified as involved in the regulation of Serratia swarming. Further studies suggested that either saturated fatty acids or temperature shift sensed by RssA and RssB influence swarming behavior through changing the cellular fatty acid profile and altering the ratio of saturated fatty acids to unsaturated fatty acids (16). The negative regulatory effect of certain fatty acids on bacterial swarming was also observed in Proteus mirabilis (17), suggesting the existence of a common regulatory pathway in bacterial swarming.

Both rssA and rssB knockout mutants showed similar precocious swarming behaviors (16). To further analyze the biochemical property of this two-component system and understand the underlying mechanism by which this two-component system regulates swarming mobility, the phosphorelay between RssA and RssB during signal transduction was studied, and the interaction between RssB and the regulated target DNA fragments are characterized in this report. Our results (i) show that RssA and RssB are two-component signal transduction proteins involved in phosphorelay reactions and (ii) provide evidence of negative autoregulation.


arrow
MATERIALS AND METHODS
 
Bacterial strains, mutants, and culture conditions. S. marcescens CH-1 and the rssA mutant strain S. marcescens CH-1{Delta}A, in which rssA was inserted by a HindIII-digested {Omega}(Smr) gene cassette (21), were from a previous study (16). Escherichia coli DH5{alpha} (Invitrogen) was used as a host strain for the maintenance of recombinant DNA plasmids. E. coli BL21(DE3)pLysS (Novagen) was used to overexpress recombinant proteins. All bacteria used in this study were grown in Luria-Bertani (LB) medium at 37°C (23) supplemented with adequate antibiotics when necessary.

Enzymes, chemicals, and primers. DNA restriction and modification enzymes were purchased from Roche. Pfu polymerase and PCR-related products were from Stratagene and Perkin Elmer. Other laboratory-grade chemicals were purchased from Sigma, Merck, and BDH. The primers used in this study are summarized in Table 1.


View this table:
[in this window]
[in a new window]
  		TABLE 1. PCR primers used in this study

Analysis of DNA and protein sequences. The DNA and deduced protein sequences were compared with those in GenBank DNA and protein sequence databases using blastn, blastp, or tblastx via the NCBI Internet homepage (http://www.ncbi.nlm.nih.gov/). Protein sequence identities were analyzed by ExPASy proteomics tools (DAS, Tmpred, SOSUI, PredictProtein, and ProtScale) in the ExPASy Molecular Biology Server of the Swiss Institute of Bioinformatics (SIB) (http://tw.expasy.org).

Swarming motility assay. The swarming assay was performed on LB medium solidified with 0.8% Eiken agar (Eiken) by inoculating 3-µl portions of an overnight LB broth culture onto the centers of agar plate surfaces and incubating at 37°C.

Construction of a recombinant plasmid, pBAD18RssA, containing complete rssA. Full-length rssA (GenBank accession number AF465237) was amplified by PCR using the primer pair RssAF and RssAR with the S. marcescens CH-1 chromosome as the template. The PCR product was cloned into the EcoRI-XbaI site of plasmid pBAD18-Kan (9). The resultant plasmid was designated pBAD18RssA.

Construction of recombinant plasmids pET28cRssA and pET28RssB. A truncated version of rssA, encoding residues 190 to 469 of RssA (cytoplasmic RssA [cRssA]), lacked the two presumptive N-terminal transmembrane domains amplified by PCR with primer pair cRssAF and cRssAR by use of S. marcescens CH-1 chromosomal DNA as the template. The PCR product was cloned into the NdeI-XhoI site of plasmid pET-28c(+) (Novagen) to create pET28cRssA. Plasmid pET28RssB was created by amplifying the complete rssB gene from S. marcescens CH-1 chromosomal DNA with primers RssBF and RssBR and then cloned into pET-28c via the NdeI-XhoI site. Histidine (His) tags were added at the N termini of both cRssA and RssB in these constructs.

Site-directed mutagenesis of cRssA, RssA, and RssB. Primer-mediated PCR mutagenesis (12) was used to mutate the presumptive phosphorylation site of RssA from H248 to A248 to create a cRssA mutant protein designated cRssA(H248A). The mutagenic oligonucleotide primer pair consisting of rssAHtAwR and rssAHtAwF (codon 248; CAC to GCC) was used for PCR with pET28cRssA as the template. The amplified DNA product was then cloned into pET28c(+) to form pET28cRssA(H248A). DNA sequencing analysis confirmed that the sequence is identical to the corresponding fragment of pET28cRssA, except for codon 248, which was changed from CAC to GCC. Similar procedures were used for pBAD18RssA(H248A) construction. PCR-amplified products, with pBAD18RssA as the template, were cloned into pBAD18-Kan to generate pBAD18RssA(H248A). To construct pET28RssB(D51E), in which the D51 is mutated in RssB to become E51 (GAT to GAA), the primer pair consisting of rssBDtEwR and rssBDtEwF was used.

Purification of His-tagged recombinant proteins. To oversynthesize cRssA, RssB, cRssA(H248A), and RssB-D51E, E. coli strain BL21(DE3)pLysS constructs containing pET28cRssA, pET28cRssA(H248A), pET28RssB, and pET28RssB(D51E), respectively, were grown in 500-ml volumes of LB broth supplemented with 50 µg/ml of kanamycin at 37°C. When the optical densities at 600 nm (OD600) reached 0.5 to 0.6, IPTG (isopropyl-ß-D-thiogalactopyranoside) was added at final concentrations of 0.2 mM, and incubation was continued for overnight at 12°C. The His-tagged proteins were purified as follows. E. coli cells were suspended in 20 mM sodium phosphate (pH 7.4) containing 5 mM imidazole, 400 mM NaCl, and 0.1% NP-40. After incubation for 30 min on ice, bacteria were broken by freezing and thawing followed by sonication (Misonix). After centrifugation at 12,000 x g for 15 min at 4°C, the spent supernatant was applied to a Ni2+-nitrilotriacetic acid column (Pharmacia) before washing three times with 20 mM sodium phosphate (pH 7.4) containing 400 mM NaCl and 50 mM imidazole. The protein was eluted with 20 mM sodium phosphate (pH 7.4) containing 400 mM NaCl and 500 mM imidazole. All proteins were dialyzed against TKMD buffer (50 mM Tris-HCl, pH 8.0, 5 mM MgCl2, 50 mM KCl, 0.2 mM dithiothreitol, and 10% glycerol). The purity of eluted protein was judged to be above 90 to 95%, based on a Coomassie blue staining assessment. Protein concentrations were determined by measuring absorbance at 595 nm using the Bradford protein assay (Bio-Rad).

In vitro phosphorylation and phosphotransfer assay. For single-time-point phosphorylation reactions, each protein at a final concentration of 10 µM was used for a protein phosphorylation assay. cRssA was incubated for 1 h without or with RssB or RssB(D51E) proteins in TKMD buffer with 20 µM ATP containing 10 µCi of [{gamma}-32P]ATP before the addition of RssB or RssB(D51E). RssB or RssB(D51E) was added for another 1-h incubation, and the reaction was stopped by adding an amount of stop buffer (4x sodium dodecyl sulfate [SDS] buffer containing 100 mM EDTA) equivalent to one-third of the original volume.

For the dynamic phosphorylation assay, 40 µM cRssA was autophosphorylated in a total volume of 100 µl TKMD buffer with 20 µM ATP and with 50 µCi of [{gamma}-32P]ATP added. At the given times, aliquots of 10 µl were withdrawn, and the reaction was stopped by adding an amount of stop buffer equivalent to one-third of the original volume. After 45 min of incubation, an equal amount of RssB (20 µM) was added into the reaction mixture for the phosphotransfer assay. All phosphotransfer reactions were carried out at room temperature. Reaction mixtures were separated by 15% SDS-polyacrylamide gel electrophoresis (10), dried, and exposed to X-ray film. Autoradiographs were scanned using a scanning densitometer (PDI).

Gel mobility shift assay. Promoter DNA fragments for gel mobility shift assays were amplified by PCR using S. marcescens CH-1 chromosomal DNA as the template. The primers used for PCR amplification are listed in Table 1. Amplified DNA fragments were then end labeled with digoxigenin (DIG)-11-ddUTP via terminal transferase supplied in a DIG gel shift kit (Roche). Reaction mixtures for the binding assay comprised 1 µM of RssB protein, which was with or without acetylphosphate (Ac-P) treatment, and DIG-labeled promoter DNA fragments of masses ranging from 1 to 10 ng. For the serial dilution experiment, phosphorylated or unphosphorylated RssB proteins were serially diluted three times in binding reaction buffer (20 mM Tris-HCl [pH 7.5], 10 mM MgCl2, 100 mM KCl, 10 mM EDTA). The binding reaction was performed in binding reaction buffer supplemented with 0.5 µg poly(dI-dC) and 5 µg bovine serum albumin. The reaction mixtures were incubated for 20 min at room temperature before loading into 6% nondenaturing polyacrylamide gels containing 0.5x Tris-borate-EDTA buffer. Electrophoresis was performed at 100 V for 1 to 1.5 h. The DNA-protein complexes were then electroblotted onto a positively charged Hybond-N nylon membrane (Amersham) and detected by anti-DIG-alkaline phosphatase. CSPD (Applied Biosystems) was added as the substrate as described by the manufacturer (Roche). Membranes were exposed to X-ray film at room temperature for 2 to 30 min.

RT-PCR and primer extension. The relative amounts of transcripts from the rssB gene were evaluated by the reverse transcription-PCR (RT-PCR) and primer extension methods (7). Total RNA was extracted by use of a Trizol kit (Invitrogen), treated with RNase-free DNase I (Promega), and then reverse transcribed into cDNA with Moloney murine leukemia virus reverse transcriptase (New England Biolabs) by use of primer RssBR (for RT-PCR) or PrssBR-3 (DIG-labeled primer ordered for primer extension). The products were amplified by PCR with the primer pair RssBR/RssBF and then analyzed by electrophoresis on 1% agarose gels for RT-PCR. For the primer extension, the products were further concentrated, dissolved in formamide loading buffer, and separated on a 10% Tris-borate-EDTA acrylamide gel with 7 M urea before being transferred to a nylon membrane to perform the DIG detection assay as described above.


arrow
RESULTS
 
A potential two-component system involved in regulation of S. marcescens CH-1 swarming. Through a mini-Tn5 transposon mutagenesis assay, a potential sensor kinase RssA and a response regulator RssB were identified as involved in the regulation of Serratia swarming (16). RssA contains all characteristics of sensor proteins, including blocks H (residues Met-242 to Val-257, with His-248 being the putative phosphorylation site), N (Leu-353 to Arg-373), D/F (Gly-385 to Gly-414), and G (Thr-415 to Lys-443) (Fig. 1A) (28). Its N-terminal region was also predicted to contain two hypothetical transmembrane domains (Lys5-Trp33, Gly162-Arg189) that may serve as anchors to the cytoplasmic membrane (data not shown). RssB shows a high level of amino acid sequence identity to members of the OmpR family (19, 20). Sequence alignments revealed that RssB bears all conserved residues identified as necessary for function in response regulator members, including Asp-8, Asp-9, Asp-51 (the putative phosphorylation site), and Lys-101, which are predicted to constitute the active sites in the receiver domain. The RssB C-terminal region contains a DNA-binding motif (residues 189 to 201), which was expected to bind and regulate the expression of downstream target genes in response to phosphorylation (Fig. 1B).



View larger version (76K):
[in this window]
[in a new window]
  		FIG. 1. Comparison of RssA blocks and RssB domains with two-component family proteins and purification of proteins used in this study. (A) Comparison of S. marcescens RssA with the conserved H, N, D/F, and G blocks of sensor proteins of two-component systems, including those of Bordetella pertussis BvgS, Neisseria gonorrhoeae BasS, and E. coli EvgS. (B) Comparison of S. marcescens RssB with the conserved phosphorylation and DNA-binding motifs of E. coli response regulators CpxR, OmpR, and PhoB. Asterisks indicate identical amino acids, and encircled asterisks indicate essential amino acid residues H248 in RssA and D51 in RssB for phosphorylation and phosphorelay reactions. (C) cRssA, RssB, cRssA(H248A), and RssB(D51E) were overexpressed and purified. Lanes 1, 8, and 10, protein molecular mass standards expressed in kDa; lanes 2 and 3, whole-cell lysates of E. coli BL21(DE3)/pET28cRssA before and after IPTG induction, respectively; lane 4, purified cRssA; lanes 5 and 6, whole-cell lysates of E. coli BL21(DE3)/pET28RssB before and after IPTG induction, respectively; lane 7, purified RssB; lanes 9 and 11, purified cRssA(H248A) and RssB(D51E), respectively. All proteins were purified from crude cell extracts after elution from Ni2+-nitrilotriacetic acid column affinity chromatography.

Autophosphorylation of cRssA and phosphotransfer to RssB. RssA was predicted to contain two transmembrane domains (4) anchored to the cytoplasmic membrane (data not shown). Previous in vitro studies on several sensor kinases showed that the removal of the transmembrane segments does not affect the process of autophosphorylation and the subsequent transphosphorylation to the cognate response regulator proteins (1, 3, 27). For facilitating the purification of RssA, cRssA and RssB were both His tagged in the N termini and purified from E. coli BL21(DE3)/pET28cRssA and E. coli BL21(DE3)/pET28RssB, respectively (Fig. 1C).

Incubating with [{gamma}-32P]ATP, cRssA was rapidly phosphorylated within 5 min (Fig. 2A), which was in agreement with the results of previous studies on the autophosphorylation of two-component-system sensor kinases (11, 27). No radioactive labeling was observed when cRssA was incubated with [{alpha}-32P]ATP or [{gamma}-32P]GTP (data not shown), suggesting that cRssA is an ATP-dependent kinase. We then addressed the question of whether RssA is able to transphosphorylate RssB in vitro. At 45 min after cRssA autophosphorylation, an equal amount of purified RssB (20 µM) was added to the reaction mixture. A kinetic analysis revealed that RssB was rapidly labeled within 3 min (Fig. 2A), while RssB incubated with [{gamma}-32P]ATP alone was not (data not shown). cRssA retained labeled up to 45 min after the addition of RssB, possibly due to the remaining free [{gamma}-32P]ATP existing in the reaction mixture. A quantification of the dynamic phosphorelay process is shown (Fig. 2B).



View larger version (23K):
[in this window]
[in a new window]
  		FIG.2. Autophosphorylation of cRssA and cRssA(H248A) and transphosphorylation of RssB and RssB(D51E) by cRssA. (A) Purified cRssA was incubated with [{gamma}-32P]ATP, and samples were harvested at the time points indicated. At 45 min, an equal amount of RssB was added. Samples harvested were separated by 15% SDS-polyacrylamide gel electrophoresis before autoradiography. (B) The result of autoradiography was quantified by PhosphorImager analysis. (C) In vitro phosphorylation of the indicated proteins by [{gamma}-32P]ATP was carried out as described in Materials and Methods. Molecular mass markers (kDa) are shown, and proteins contained (+) in the reaction mixtures are shown in the list at the top. cRssA and cRssA(H248A) at concentrations of 10 µM were incubated with [{gamma}-32P]ATP for 1 hour, and an equal amount RssB or RssB(D51E) was added into the reaction mixtures for another hour. Reactions were stopped by the addition of 4x SDS sample buffer. Each reaction mixture (5 µl) was subjected to separation by 15% SDS-polyacrylamide gel electrophoresis, followed by visualization using Coomassie blue staining (upper panel) or autoradiography (lower panel). (D) Swarming phenotypes of S. marcescens strains CH-1 (I), CH-1{Delta}A (II), CH-1{Delta}A complemented with wild-type RssA (pBAD18RssA) (III), and CH-1{Delta}A complemented with mutant RssA [pBAD18RssA(H248A)] (IV) after 10 h of incubation on LB swarming plates at 37°C.

cRssA(H248A) and RssB(D51E) cannot be phosphorylated, and RssA(H248A) fails to complement S. marcescens CH-1{Delta}A. Amino acid sequence alignment analyses suggested that His248 of RssA and Asp51 of RssB are the phosphorylation sites for the phosphorelay reaction during the process of RssA-RssB signal transduction (Fig. 1A and B). To confirm this, His248 in RssA and Asp51 in RssB were replaced with alanine and glutamate, respectively, to form His-tagged cRssA(His248Ala) [hereafter referred to as cRssA(H248A)] and His-tagged RssB(Asp51Glu) [hereafter referred to as RssB(D51E)], respectively, by site-directed mutagenesis. cRssA(H248A) incubated in the same phosphorylation reaction buffer with [{gamma}-32P]ATP did not autophosphorylate itself (Fig. 2C). When autophosphorylated cRssA was mixed with RssB or RssB(D51E), 32P-labeled cRssA transferred its radioactivity to RssB but not to RssB(D51E) even after elongated reaction times of up to 1 hour (Fig. 2C). The amount of 32P-labeled cRssA was observed to increase slowly during the course of the experiment, possibly due to the excessive [{gamma}-32P]ATP existing in the reaction mixture.

To see whether the precocious swarming phenotype of S. marcescens CH-1{Delta}A (16) can be restored by rssA, the recombinant plasmid pBAD18RssA (16) was transformed to CH-1{Delta}A, which inhibited the precocious swarming phenotype of CH-1{Delta}A at 37°C (Fig. 2D, panel III). In comparison, no complementation was observed when the recombinant plasmid pBAD18RssA(H248A) was transformed to CH-1{Delta}A (Fig. 2D, panel IV). These data suggested that the precocious swarming phenotype observed in CH-1{Delta}A is indeed caused by a defect in rssA and that RssA(H248), which is shown to be involved in the phosphorelay reaction, is essential for RssA function.

RssB binds directly to its own promoter. Our previous studies showed that the expression of the hemolysin gene shlA and a functionally uncharacterized ygfF gene of S. marcescens (ygfFsm), which is located immediately upstream of rssB, is under the control of the RssA-RssB protein pair (13, 16). Furthermore, evidence showed that two-component response regulator proteins frequently autoregulate the expression of themselves (24, 26). These data suggested that the promoters of the shlBA operon, ygfFSm, or rssB might be directly bound and regulated by RssB. An in vitro electrophoretic mobility shift assay was performed to see the interaction between RssB and these promoter DNA fragments. Promoter regions of these three genetic determinants were amplified by PCR using primer pairs PshlBF/PshlBR, PygfFF/PygfFR, and PrssBF-1/PrssBR-1, followed by end labeling with DIG-ddUTP. The phosphorylated form of RssB (RssB-P) was prepared by incubation with Ac-P prior to being added to the reaction mixture (18). In the presence of an excess of nonspecific competitor DNA [0.5 mg of poly(dI-dC)] and bovine serum albumin, RssB-P bound to the rssB promoter and completely shifted the DNA fragment, forming a major band of protein/DNA complex (Fig. 3A, panel III, lane 3). Under the same assay conditions, no DNA shift phenomenon was observed when shlBA and the ygfFSm promoter were assayed (Fig. 3A, panels I and II). By comparison, the addition of the untreated RssB for the electrophoretic mobility shift assay led to only a partial shift of the rssB promoter DNA fragment (Fig. 3A, panel III, lane 2). To define the specific DNA region where RssB interacts with the rssB promoter, different primer pairs were designed and used in PCRs to amplify different DNA fragments within the rssB promoter region (Fig. 3B). While RssB could not bind fragments amplified by primer pairs PrssBF-2/PrssBR-1 and PrssBF-2/PrssBR-2 (Fig. 3A, panels IV and V), it bound to the 104-bp DNA fragment located at the region from –100 to –203 bp amplified by the primer pair PrssBF-3/PrssBR-3 (Fig. 3B and 4A and B). Cold competition by unlabeled PrssB fragments further confirmed the binding specificity and narrowed down the potential binding region to be within a 63-bp DNA fragment (Fig. 3B). While the unlabeled PrssBF-3/PrssBR-3 fragment could compete and reduce the shift due to RssB-P binding to the same fragment (Fig. 3A VII) in a manner similar to that of the PygfFF/PygfFR fragment, the annealed PrssBF-4/PrssBR-4 DNA fragment did not show competition evidence (Fig. 3A, panel VI). These data indicated that RssB specifically binds the region between PrssBR-4 and PrssBR-3.



View larger version (43K):
[in this window]
[in a new window]
  		FIG. 3. In vitro RssB-DNA gel mobility shift assay. (A) Purified RssB was used to test whether it interacted with DIG-labeled predicted promoter DNA regions, including shlB (I), ygfFSm (II), PrssBF-1/PrssBR-1 (III), PrssBF-2/PrssBR-2 (IV), and PrssBF-2/PrssBR-1 (V). Lane 1, DNA fragments only; lane 2, DNA fragments with purified RssB; lane 3, DNA fragments with purified RssB phosphorylated by Ac-P. (VI) Cold competition assay by unlabeled DNA fragment PrssBF-4/PrssBR-4 (lane 3) or ygfFSm (lane 4). Lane 1, labeled PrssBF-3/PrssBR-3 DNA fragment only; lane 2, PrssBF-3/PrssBR-3 with RssB-P only. (VII) Cold competition assay by unlabeled DNA fragment PrssBF-3/PrssBR-3. Lane 1, labeled PrssBF-3/PrssBR-3 DNA fragment only; for lanes 2 to 5, the competitive cold fragment concentrations were decreased from 50x, 10x, and 1x, respectively, to 0x. (B) rssB promoter region, where the locations of primers designed are shown with solid arrows. Arrows with filled circles indicate the predicted transcription and translation start sites. RssB-P binding DNA regions (63 bp) are shadowed.



View larger version (48K):
[in this window]
[in a new window]
  		FIG. 4. Effect of in vitro phosphorylation to RssB on DNA-binding capability. We used the DIG-labeled 104-bp rssB promoter region amplified by PrssBF-3/PrssBR-3 to test the effect of RssB phosphorylation on binding efficiency and DNA mobility shift by (A) RssB and (B) RssB-P. Protein concentrations used are indicated at the bottom. (C) DNA mobility shift by RssB-P (0.02 µM) (lane 2), RssB (1.5 µM) (lane 3), and RssB (D51E; 10 µM) (lane 4) with the same DNA fragment. Lane 1, target DNA only, serving as the mobility shift control. Positions of DNA-protein complexes CI and CII are indicated with arrows.

Phosphorylated RssB shows higher DNA-binding affinity. The DNA-binding activities of two-component-system response regulator proteins are usually dependent on their phosphorylation statuses (3, 27). To examine whether phosphorylation influences the efficiency of the binding of RssB to its own promoter DNA, RssB and RssB-P were used in a DNA mobility shift assay to verify the binding ability. To this end, the 104-bp DNA fragment amplified by PrssBF-3/PrssBR-3 was incubated with RssB or RssB-P at serially diluted concentrations. While DNA mobility shift phenomena were observed in the presence of both RssB (Fig. 4A) and RssB-P (Fig. 4B), a significantly lower RssB-P concentration was needed for the initiation of the DNA band shift. RssB-P at concentrations of as low as 0.002 µM started to show DNA-binding activity, and the fast-migration band complex I (CI) started to appear from these concentrations (Fig. 4B). Formation of the slow-migration band complex II (CII) occurred as the RssB-P concentration was increased from 0.02 µM (Fig. 4B). At concentrations of 0.05 µM or above, almost all DNA fragments were shifted by RssB-P under the detection conditions. By comparison, although RssB was also able to bind to PrssBF-3/PrssBR-3, a much higher protein concentration (0.006 µM) was needed to form DNA-protein complex CI, and little or no CII complex was formed (Fig. 4A). Moreover, RssB failed to shift all DNA fragments at concentrations of up to 1.5 µM (Fig. 4A). Similar results were also reported for many other response regulators, whereby the response regulator could bind to its target DNA without Ac-P treatment, and phosphorylation by Ac-P significantly enhances the binding activity (1, 2, 22).

As it was reported that the binding of an unphosphorylated response regulator to target DNA was due to partial phosphorylation during propagation in E. coli (5, 15), we then checked if RssB(D51E), which could not be phosphorylated, still bound to the 104-bp DNA fragment. As shown in Fig. 4C, under the same assay conditions, RssB(D51E) at the very high concentration of 10 µM still did not lead to the electrophoretic mobility shift of the DNA fragment tested. The data therefore suggested that the ability of RssB to bind to its target DNA is dependent on its phosphorylation status.

rssB promoter activity was activated in S. marcescens CH-1{Delta}A. The fact that phosphorylated RssB binds to its own promoter encouraged us to investigate whether the transcriptional activity of rssB was affected when it was not phosphorylated by RssA.

Results from the primer extension assay showed that rssB promoter activity is activated in S. marcescens CH-1{Delta}A, weakly in log phase and significantly in early stationary phase (Fig. 5A). The RT-PCR assay further confirmed that rssB RNA levels are increased in S. marcescens CH-1{Delta}A in both growth phases (Fig. 5B). Briefly, these data showed that in the absence of rssA, rssB promoter activity was activated, especially in early stationary phase, suggesting that RssA-RssB signaling negatively regulates rssB promoter expression.



View larger version (67K):
[in this window]
[in a new window]
  		FIG. 5. rssB transcription level is elevated in S. marcescens CH-1{Delta}A. Primer extension (A) and RT-PCR (B) assays were performed for the quantification of rssB transcripts. Total cellular RNA was extracted from S. marcescens CH-1 and CH-1{Delta}A grown to log phase (OD600 = 0.5) (lanes 1 and 3) and early stationary phase (OD600 = 1.2) (lanes 2 and 4). Lanes 1 and 2, S. marcescens CH-1; lanes 3 and 4, S. marcescens CH-1{Delta}A.


arrow
DISCUSSION
 
In a previous study, we demonstrated that a potential two-component system consisting of the protein pair RssA-RssB is involved in the regulation of swarming in S. marcescens CH-1 in response to saturated fatty acids and changes in the cellular fatty acid profile (16). In this study, we further examine the underlying biochemical mechanism by which the RssA-RssB system transfers signal, binds to its target gene, and therefore controls the swarming behavior.

Computer-assisted comparison analyses showed that RssA resembles histidine kinase (8) and RssB resembles a certain response regulator of a two-component regulatory system (27). On the basis of the following observations, we confirmed that RssA is a histidine kinase which, after phosphorylation, transfers the phosphate to RssB as interacting components of a two-component regulatory system in S. marcescens CH-1. First, cRssA autophosphorylates and serves as a phosphate donor to RssB rapidly and efficiently. Second, the replacement of His248 of cRssA by Ala prevents in vitro autophosphorylation, and replacement of Asp51 of RssB by Glu prevents its activity as a phosphate acceptor from cRssA (Fig. 2). Third, RssB binds to a 104-bp DNA fragment located within the predicted rssB promoter region in vitro (Fig. 3 and 4) and negatively autoregulates its own RNA transcription, especially in stationary phase (Fig. 5). While autoactivation phenomena are observed in many other two-component systems, such as HP166-HP165 (7) and CpxR-CpxA (6), our results suggested an autoinhibition phenomenon. A similar negative autoregulation phenomenon is also observed in PmrA-PmrB (14).

Most response regulators bind to their target DNAs while being phosphorylated (27). It is reported in some cases that a response regulator could bind DNA, albeit less efficiently, under unphosphorylated conditions. Unphosphorylated Spo0A binds detectably to the 0A box regions of a set of promoters, and the extent of binding is increased up to ~20-fold after phosphorylation (22). Ac-P treatment increases the binding affinity of DcuR to the dcuB promoter from an equilibrium dissociation constant of 6 µM to one of <1 µM (1). h-DesR could bind a specific des promoter region without further phosphorylation treatment (2). However, further examination shows that the binding of DesR to the des promoter is due to the partial phosphorylation of DesR during overexpression in E. coli (5). Here we demonstrate that although RssB purified from E. coli without in vitro phosphorylation treatment by Ac-P or phosphotransfer by phosphorylated cRssA shows binding ability to its target DNA fragments, Ac-P treatment of RssB significantly enhances its binding activity (Fig. 4). However, the fact that RssB(D51E) could not bind to target DNA even with Ac-P treatment suggested that RssB purified from E. coli may be phosphorylated to some degree. These data suggested that the efficiency of RssB binding to its own promoter is closely related to its phosphorylation status.

Two significant shifted bands, CI and CII, were observed after the binding of phosphorylated RssB to the 104-bp DNA fragment within the rssB promoter region amplified by PrssBF-3/PrssBR-3. CI was present both in the reaction mixture containing RssB and in the reaction mixture containing RssB-P. CII migrated even more slowly and existed mostly in the reaction mixture containing RssB-P. We preliminarily observed that RssB forms dimers or even multimers after being phosphorylated by Ac-P at 37°C (unpublished work). This suggested that further complicated conformational change and protein-protein interactions might happen while RssB-P binds to the target promoter.

The facts that S. marcescens rssA mutant strain CH-1{Delta}A swarms vigorously at 37°C and that complementation of CH-1{Delta}A by wild-type but not mutant rssA [rssA(H248A)] returns the swarming phenotype suggested that RssA acts as a repressor regulating S. marcescens CH-1 swarming at 37°C. Although the underlying mechanism by which RssA-RssB regulates swarming is not confirmed yet, evidence suggested that RssA-RssB is involved in the regulation of the flhDCSm flagellar motility system (unpublished work). As swarming is an energy-consuming process, it is speculated that S. marcescens does not initiate swarming immediately but delays until the depletion of nutrients or the accumulation of metabolic wastes during the process of colonial cell growth occurs. RssB does not bind to the promoter regions of shlAB, ygfFSm (Fig. 3A I and II), or other genes (data not shown) which are expressed at significantly higher levels in CH-1{Delta}A than in the wild type. Besides, we have performed a solid-phase DNA binding assay as described previously (7) to search for the potential target gene of the RssA-RssB signaling system. However, neither experimental results nor a computational similarity search have yet given us any hint about what might be bound and regulated by RssB. Not every two-component system regulates its downstream genes by promoter binding. CheY, the sensor of a chemotaxis two-component system comprising only one phosphorylation domain, regulates the flagella system by interacting with the flagella motor switch complex (25). The existence of some posttranscriptional mechanism by which RssA-RssB regulates other genes is highlighted. We speculated that, at early growth stage, phosphorylated RssB inhibits S. marcescens swarming. As cells are growing into stationary phase, phosphorylated RssB binds and represses its own transcription and leads to the derepression of downstream target genes, and swarming initiates from the colonial edge. Further experiments aiming at understanding the in vivo RssA-RssB phosphorelay situation and identifying the RssA-RssA target genes involved in the regulation of swarming are being performed. These will help us unravel the molecular mechanisms underlying S. marcescens swarming.


arrow
ACKNOWLEDGMENTS
 
This work was supported by grants from National Science Council (NSC-92-2314-B-002-356 and NSC-93-2314-B-002-281) and Technology Development Program for Academia, Ministry of Economical Affairs (91-EC-17-A-10-S1-0013), which are greatly appreciated.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Department of Clinical Laboratory Sciences and Medical Biotechnology, National Taiwan University College of Medicine, No. 1. Chang-Der Street, Taipei 100, Taiwan, Republic of China. Phone: 886 2 2312-3456, ext. 6931. Fax: 886 2 2371-1574. E-mail: hclai{at}ha.mc.ntu.edu.tw. Back


arrow
REFERENCES
 
    1
  1. Abo-Amer, A. E., J. Munn, K. Jackson, M. Aktas, P. Golby, D. J. Kelly, and S. C. Andrews. 2004. DNA interaction and phosphotransfer of the C4-dicarboxylate-responsive DcuS-DcuR two-component regulatory system from Escherichia coli. J. Bacteriol. 186:1879-1889.[Abstract/Free Full Text]
    2. 2
  2. Aguilar, P. S., A. M. Hernandez-Arriaga, L. E. Cybulski, A. C. Erazo, and D. de Mendoza. 2001. Molecular basis of thermosensing: a two-component signal transduction thermometer in Bacillus subtilis. EMBO J. 20:1681-1691.[CrossRef][Medline]
    3. 3
  3. Bird, T. H., S. Du, and C. E. Bauer. 1999. Autophosphorylation, phosphotransfer, and DNA-binding properties of the RegB/RegA two-component regulatory system in Rhodobacter capsulatus. J. Biol. Chem. 274:16343-16348.[Abstract/Free Full Text]
    4. 4
  4. Cserzo, M., E. Wallin, I. Simon, G. von Heijne, and A. Elofsson. 1997. Prediction of transmembrane alpha-helices in prokaryotic membrane proteins: the dense alignment surface method. Protein Eng. 10:673-676.[Abstract/Free Full Text]
    5. 5
  5. Cybulski, L. E., G. del Solar, P. O. Craig, M. Espinosa, and D. de Mendoza. 2004. Bacillus subtilis DesR functions as a phosphorylation-activated switch to control membrane lipid fluidity. J. Biol. Chem. 279:39340-39347.[Abstract/Free Full Text]
    6. 6
  6. De Wulf, P., O. Kwon, and E. C. Lin. 1999. The CpxRA signal transduction system of Escherichia coli: growth-related autoactivation and control of unanticipated target operons. J. Bacteriol. 181:6772-6778.[Abstract/Free Full Text]
    7. 7
  7. Dietz, P., G. Gerlach, and D. Beier. 2002. Identification of target genes regulated by the two-component system HP166-HP165 of Helicobacter pylori. J. Bacteriol. 184:350-362.[Abstract/Free Full Text]
    8. 8
  8. Grebe, T. W., and J. B. Stock. 1999. The histidine protein kinase superfamily. Adv. Microb. Physiol. 41:139-227.[Medline]
    9. 9
  9. Guzman, L. M., D. Belin, M. J. Carson, and J. Beckwith. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177:4121-4130.[Abstract/Free Full Text]
    10. 10
  10. Harlow, E., and D. Lane. 1988. Antibodies: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
    11. 11
  11. Hoch, J. A., and T. J. Silhavy (ed.). 1995. Two-component signal transduction. ASM Press, Washington, D.C.
    12. 12
  12. Horton, R. M., Z. L. Cai, S. N. Ho, and L. R. Pease. 1990. Gene splicing by overlap extension: tailor-made genes using the polymerase chain reaction. BioTechniques 8:528-535.[Medline]
    13. 13
  13. Hsieh, S. C., Y. S. Fang, Y. S. Yang, H. C. Lai, and S. W. Ho. Activation of ygfFSm expression by long chain fatty acids and 3-oxo-C12 quorum sensing signal is antagonized by RssA, a two-component system sensor kinase involved in regulation of Serratia marcescens swarming. J. Microbiol. Immunol. Infect., in press.
    14. 14
  14. Hyytiainen, H., S. Sjoblom, T. Palomaki, A. Tuikkala, and P. E. Tapio. 2003. The PmrA-PmrB two-component system responding to acidic pH and iron controls virulence in the plant pathogen Erwinia carotovora ssp. carotovora. Mol. Microbiol. 50:795-807.[CrossRef][Medline]
    15. 15
  15. Ladds, J. C., K. Muchova, D. Blaskovic, R. J. Lewis, J. A. Brannigan, A. J. Wilkinson, and I. Barak. 2003. The response regulator Spo0A from Bacillus subtilis is efficiently phosphorylated in Escherichia coli. FEMS Microbiol. Lett. 223:153-157.[CrossRef][Medline]
    16. 16
  16. Lai, H. C., P. C. Soo, J. R. Wei, W. C. Yi, S. J. Liaw, Y. T. Horng, S. M. Lin, S. W. Ho, S. Swift, and P. Williams. 2005. The RssAB two-component signal transduction system in Serratia marcescens regulates swarming motility and cell envelope architecture in response to exogenous saturated fatty acids. J. Bacteriol. 187:3407-3414.[Abstract/Free Full Text]
    17. 17
  17. Liaw, S. J., H. C. Lai, and W. B. Wang. 2004. Modulation of swarming and virulence by fatty acids through the RsbA protein in Proteus mirabilis. Infect. Immun. 72:6836-6845.[Abstract/Free Full Text]
    18. 18
  18. McCleary, W. R., and J. B. Stock. 1994. Acetyl phosphate and the activation of two-component response regulators. J. Biol. Chem. 269:31567-31572.[Abstract/Free Full Text]
    19. 19
  19. Mizuno, T., and I. Tanaka. 1997. Structure of the DNA-binding domain of the OmpR family of response regulators. Mol. Microbiol. 24:665-667.[CrossRef][Medline]
    20. 20
  20. Pao, G. M. and M. H. Saier, Jr. 1995. Response regulators of bacterial signal transduction systems: selective domain shuffling during evolution. J. Mol. Evol. 40:136-154.[CrossRef][Medline]
    21. 21
  21. Prentki, P., and H. M. Krisch. 1984. In vitro insertional mutagenesis with a selectable DNA fragment. Gene 29:303-313.[CrossRef][Medline]
    22. 22
  22. Rowe-Magnus, D. A., and G. B. Spiegelman. 1998. Contributions of the domains of the Bacillus subtilis response regulator Spo0A to transcription stimulation of the spoIIG operon. J. Biol. Chem. 273:25818-25824.[Abstract/Free Full Text]
    23. 23
  23. Sambrook, J., and D. W. Russell. 2001. Molecular cloning, a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
    24. 24
  24. Scarlato, V., A. Prugnola, B. Arico, and R. Rappuoli. 1990. Positive transcriptional feedback at the bvg locus controls expression of virulence factors in Bordetella pertussis. Proc. Natl. Acad. Sci. USA 87:6753-6757.[Abstract/Free Full Text]
    25. 25
  25. Silversmith, R. E., and R. B. Bourret. 1999. Throwing the switch in bacterial chemotaxis. Trends Microbiol. 7:16-22.[CrossRef][Medline]
    26. 26
  26. Soncini, F. C., E. G. Vescovi, and E. A. Groisman. 1995. Transcriptional autoregulation of the Salmonella typhimurium phoPQ operon. J. Bacteriol. 177:4364-4371.[Abstract/Free Full Text]
    27. 27
  27. Stock, A. M., V. L. Robinson, and P. N. Goudreau. 2000. Two-component signal transduction. Annu. Rev. Biochem. 69:183-215.[CrossRef][Medline]
    28. 28
  28. Wolanin, P. M., P. A. Thomason, and J. B. Stock. 2002. Histidine protein kinases: key signal transducers outside the animal kingdom. Genome Biol. 3:reviews3013.

Journal of Bacteriology, August 2005, p. 5683-5690, Vol. 187, No. 16
0021-9193/05/$08.00+0     doi:10.1128/JB.187.16.5683-5690.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Soo, P.-C., Horng, Y.-T., Wei, J.-R., Shu, J.-C., Lu, C.-C., Lai, H.-C. (2008). Regulation of Swarming Motility and flhDCSm Expression by RssAB Signaling in Serratia marcescens. J. Bacteriol. 190: 2496-2504 [Abstract] [Full Text]  
  • Wei, J.-R., Tsai, Y.-H., Horng, Y.-T., Soo, P.-C., Hsieh, S.-C., Hsueh, P.-R., Horng, J.-T., Williams, P., Lai, H.-C. (2006). A Mobile Quorum-Sensing System in Serratia marcescens. J. Bacteriol. 188: 1518-1525 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wei, J.-R.
Right arrow Articles by Lai, H.-C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wei, J.-R.
Right arrow Articles by Lai, H.-C.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»