JOURNAL OF BACTERIOLOGY, Nov. 2002, p. 5862–5870 0021-9193/02/$04.00⫹0 DOI: 10.1128/JB.184.21.5862–5870.2002 Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Vol. 184, No. 21
Molecular Characterization of the Growth Phase-Dependent Expression of the lsrA Gene, Encoding Levansucrase of Rahnella aquatilis Jeong-Woo Seo,1† Ki-Hyo Jang,2 Soon Ah Kang,2 Ki-Bang Song,1 Eun Kyung Jang,3 Buem-Seek Park,1 Chul Ho Kim,1,3 and Sang-Ki Rhee1* Korea Research Institute of Bioscience and Biotechnology,1 and RealBioTech Co. Ltd.3 Korea Research Institute of Bioscience and Biotechnology, Yusong, Daejon 305-333, and Department of Medical Nutrition, Graduate School of East-West Medical Science, Kyung Hee University, Suwon 449-701,2 Korea Received 21 February 2002/Accepted 24 July 2002
Expression of the lsrA gene from Rahnella aquatilis, encoding levansucrase, is tightly regulated by the growth phase of the host cell; low-level expression was observed in the early phase of cell growth, but expression was significantly stimulated in the late phase. Northern blot analysis revealed that regulation occurred at the level of transcription. The promoter region was identified by primer extension analysis. Two opposite genetic elements that participate in the regulation of lsrA expression were identified upstream of the lsrA gene: the lsrS gene and the lsrR region. The lsrS gene encodes a protein consisting of 70 amino acid residues (Mr, 8,075), which positively activated lsrA expression approximately 20-fold in a growth phase-dependent fashion. The cis-acting lsrR region, which repressed lsrA expression about 10-fold, was further narrowed to two DNA regions by deletion analysis. The concerted action of two opposite regulatory functions resulted in the growth phasedependent activation of gene expression in Escherichia coli independent of the stationary sigma factor S. activates transcription of sacB directly or indirectly through SacX-SacY (9). Recently, we cloned and characterized a levansucrase gene, lsrA, from Rahnella aquatilis (27), a gram-negative enteric bacterium found in drinking water, river water, and plants and also in human clinical specimens (3). The lsrA gene was expressed well in Escherichia coli from its natural promoter upstream of the gene. The level of gene expression correlated with the degree of upstream DNA present in the cloned fragment. The lsrA gene was expressed at low levels in the early phase of cell growth but was stimulated significantly upon entry into the late part of the growth phase, a phenomenon that seemed to be caused by upstream DNA sequences. Based on these findings, we investigated the molecular basis for the growth phase-dependent expression of the gene and the role of the upstream region in regulation.
Bacterial levansucrase is an extracellular protein found in gram-negative and gram-positive bacteria. This protein catalyzes the synthesis of levan (-2,6-linked fructan) from sucrose (7). Bacterial levansucrases catalyze at least three different reactions: hydrolysis of sucrose, polymerization of fructose derived from sucrose, and hydrolysis of levan. Disruption of the gene encoding levansucrase of the fireblight pathogen Erwinia amylovora retards the development of necrotic symptoms in inoculated pear seedlings, suggesting that the enzyme participates in phytopathogenesis (14). This enzyme is also involved in the sucrose metabolism of Acetobacter diazotrophicus, in which disruption of the gene resulted in a mutant lacking the ability to utilize sucrose as a carbon source (1). In Bacillus subtilis, the sacB gene encodes levansucrase, and expression of this gene is induced by sucrose (29). Sucrose induction involves an antitermination mechanism employing an antiterminator (sacY or sacS), which prevents early termination at a rho-independent terminator located upstream of sacB (8). Sucrose modulates SacY activity through a regulatory cascade involving a phosphotransferase system and SacX, an inhibitor of SacY (31). Kunst and Rapoport reported that sacB expression was also stimulated by salt (20). The salt stimulus was transmitted via the DegS-DegU two-component system that has a pleiotropic regulatory role (20), in which DegU
MATERIALS AND METHODS Strains, plasmids, and media. R. aquatilis ATCC 33071 was grown in LuriaBertani (LB) broth at 30°C. The E. coli strains and plasmids used in this work are listed in Table 1. LB broth was used for growth of E. coli cells. An LB agar plate supplemented with 10% (wt/vol) sucrose was used for identification of the levansucrase phenotype. Antibiotics were used at the following concentrations: ampicillin,100 g/ml; chloramphenicol, 50 g/ml; and tetracycline, 20 g/ml. DNA manipulation and reagents. DNA isolation, restriction enzyme digestion, agarose gel electrophoresis, ligation, and transformation were all performed as described previously (16, 25). Unless otherwise specified, chemicals were purchased from Sigma. [␥-32P]ATP (⬎5,000 Ci/mmol) was obtained from Amersham. Restriction enzymes, calf intestinal alkaline phosphatase, the Klenow fragment, T4 polynucleotide kinase, T4 DNA ligase, Taq DNA polymerase, and a deoxynucleoside triphosphate mixture were obtained from Boehringer Mannheim or Takara. Oligonucleotide primers were synthesized by Takara. Construction of lsrA subclones with various lengths of upstream sequences. A number of deletion subclones of the upstream region of the lsrA gene, including subclones ⫺1146, ⫺798, ⫺419, ⫺274, and ⫺204, were derived from pRL1CPR
* Corresponding author. Mailing address: Korea Research Institute of Bioscience and Biotechnology (KRIBB), 52 Oun-dong, Yusong, Daejon 305-333, Korea. Phone: 82 42 860 4450. Fax: 82 42 860 4594. E-mail:
[email protected]. † Present address: Department of Biotechnology, Graduate School of Agriculture and Life Science, The University of Tokyo, Bunkyo-ku, Tokyo 113, Japan. 5862
REGULATION OF R. AQUATILIS lsrA
VOL. 184, 2002
5863
TABLE 1. Strains and plasmids used in this study Strain or plasmid
E. coli strains DH5␣ W3110 CP1005 Plasmidsa pBluescript II KS 2(⫹) pACYC184 pRLICPR pNd137 f118R pACNR(lsrS) a b c
Description
Source or reference
80dlacZ⌬M15 endA1 recA1 hsdR17 supE44 gyrA96 ⌬(lacZYA-agrF)U169 rpoS⫹ W3110 rpoS::Tn10
Bethesda Research Laboratories J. G. Pan J. G. Pan
Apr, cloning vector Cmr Tcr, cloning vector 3-kb ClaI lsrA⫹ fragment in pBluescript II KS (⫹) lsrA⫹ derivative of pRLICPR, 1.12 kb deleted from ClaI siteb lsrA derivative of pRLICPR, 1.88 kb deleted from ClaI sitec 0.6-kb NheI lsrS⫹ fragment in pACYC184
Stratagene New England Biolabs 27 This study This study This study
Plasmids used in deletion analyses of the upstream region are shown in Fig. 4 and 8. The ClaI site was located upstream of the lsrA structural gene. The ClaI site was located downstream of the lsrA structural gene.
by unidirectional deletion with an Exo mung bean deletion kit (Stratagene). pRL1CPR was digested with XbaI, treated with the Klenow fragment to fill in 5⬘ residues of the XbaI site with thiodeoxynucleoside triphosphates to block exonuclease III digestion, and digested with EcoRV. The double-digested DNA was treated with exonuclease III and mung bean nuclease to create nested deletions. The resulting plasmids were recircularized and transformed into E. coli DH5␣. Plasmids isolated from transformants were sequenced, and a collection of upstream deletion mutations was identified. Subsequent deletion of the lsrR region to prepare subclones ⫺175, ⫺146, ⫺119, and ⫺55 was performed by using the PCR method with sets of oligonucleotide primers. In all cases, the downstream primer, 3⬘-CGAAGCGTTACTG TCGAC-5⬘, which included an SalI site, was used. The following upstream primers, which included an XbaI site, were used: 5⬘-TCTAGAACCGGTAGAG GATA-3⬘ for ⫺175, 5⬘-TCTAGACTGACGATGAT-3⬘ for ⫺146, 5⬘-TCTAGA CTGACGATGAT-3⬘ for ⫺119, and 5⬘-TCTAGAACTGAGTGCATG-3⬘ for ⫺55. PCRs were carried out by using Taq DNA polymerase with pRL1CPR as the template. After an initial denaturation step of 10 min at 94°C, amplification was performed by using 30 cycles of denaturation at 94°C for 1 min, annealing at 50°C for 1 min, and extension at 72°C for 2 min. The products were gel purified, digested with XbaI and SalI, and ligated to pBluescript II KS⫹. A deletion derivative, ⫺(65-55), was prepared by ligation of f118R (see Fig. 5) and ⫺55 at the blunted XbaI site. Expression of lsrA. E. coli subclones carrying the lsrA gene were grown in 500-ml baffled flasks containing 100-ml portions of LB media supplemented with the appropriate antibiotics at 37°C with shaking at 100 rpm. Growth of the cells was monitored by determining the A600, and cells were harvested at the following stages of cell growth: early exponential phase (A600, 1), mid-exponential phase (A600, 2), late exponential phase (A600, 3.5), and early stationary phase (A600, 4.5). The harvested cells were used for further analysis as described below. RNA isolation and Northern blot analysis. Total RNA was isolated by using the hot phenol extraction procedure (25). E. coli cells harboring pRL1CPR or pNd137 prepared as described above were disrupted by vortexing with glass beads. After centrifugation to remove the cell debris, the supernatant was extracted three times with equal volumes of phenol-chloroform (5:1, vol/vol). The RNAs were ethanol precipitated, washed with 70% ethanol, dried, dissolved in diethyl pyrocarbonate-treated water, quantified on the basis of A260, and analyzed by Northern hybridization. The RNA samples (10 g) were separated by electrophoresis at 50 V on a 1% agarose gel poured in 1.1 M formaldehyde–10 mM NaPO4 (pH 7.4) and were transferred to a nylon membrane (Nytran; Schleicher & Schuell) by the capillary method (25). The membrane was hybridized to a 1.13-kb PmaCI-AflII internal fragment of lsrA labeled with digoxigenin (Boehringer Mannheim) at 50°C for 24 h. Hybridization was carried out in a solution containing 50% formamide, 1⫻ Denhardt’s solution, 100 g of carrier per ml, 0.1% sodium dodecyl sulfate (SDS), and 5⫻ SSC (1⫻ SSC is 0.15 M NaCl plus 0.015 M sodium citrate). After hybridization, the membrane was washed twice (10 min each) with 2⫻ SSC–0.1% SDS and then with 0.1⫻ SSC–0.1% SDS as described above. The hybridized RNA blots were developed as recommended by the manufacturer. Primer extension analysis. An oligonucleotide complementary to nucleotides 37 to 60 relative to the translation start codon of lsrA (5⬘-GTTGACCTTCAGC GAATCAGCACG-3⬘) was used as a primer to map the 5⬘ termini of the lsrA transcript. Twenty picomoles of ␥-32P-labeled primer was mixed with 5 g of
RNA. The mixture was heated at 58°C and cooled on ice. The annealed primers were extended with avian myeloblastosis virus reverse transcriptase (Promega Corp.) at 42°C for 30 min. The primer extension products were electrophoresed on a 6% sequencing gel along with lsrA sequencing reaction mixtures (Sequenase; United States Biochemical) by using pRL1CPR as the template and the same oligonucleotide as the primer. Enzyme activity assay. The activity of levansucrase was assayed by the method of O’Mullan et al. (24). A cell-free lysate was mixed with 50 mM sodium acetate (pH 6.0)–1% sucrose in a final volume of 1 ml, and the mixture was incubated at 37°C for 1 h. The concentration of glucose released by the sucrose hydrolysis activity of levansucrase was measured by the glucose oxidase method with a GOD-PAP kit (Sigma). All the results presented below were obtained from at least three independent experiments. Representative values are shown. One unit of enzyme activity was defined as the amount of enzyme that produced 1 mol of glucose per min, and specific activity was expressed in units per milligram of protein. The amount of protein was determined with a protein assay kit (BioRad) by using bovine serum albumin as the standard (Sigma). Nucleotide sequence analysis. Nucleotide sequencing was performed by the dideoxy chain termination method (26). The reaction was carried out with an ABI Prism dye terminator cycle sequencing Ready Reaction kit (Perkin-Elmer), and the mixture was analyzed with an automatic DNA sequencer (model 373A; Applied Biosystems). Nucleotide sequence accession number. The DNA sequence upstream of the lsrA gene, including the lsrS gene and the lsrR region, has been deposited in the GenBank nucleotide sequence database under accession number U91484.
RESULTS Growth phase-dependent expression of the lsrA gene in E. coli. To investigate the regulatory mechanism for the expression of the R. aquatilis lsrA gene in E. coli, Northern blot analysis was performed with total RNAs prepared at various stages of growth of E. coli DH5␣ harboring recombinant plasmid pRL1CPR, which contains the lsrA gene with 1.3 kb of upstream sequence. As shown in Fig. 1, 1.3-kb transcripts corresponding to the lsrA gene were detected by using a 1.13-kb PmaCI-AflII DNA fragment of the lsrA gene as the probe. Specific transcripts were barely detectable in the early phase of cell growth, but the level increased dramatically upon entrance into the late phase. However, when a similar Northern analysis was performed with RNA prepared from an E. coli strain harboring pNd137, which contains the lsrA gene with a 0.1-kb upstream sequence, the amount of lsrA transcripts increased in a linear fashion with cell growth (Fig. 1). These results imply that regulation of lsrA gene expression in E. coli occurs at the
5864
SEO ET AL.
J. BACTERIOL.
FIG. 1. Northern blot analysis of the lsrA transcript. The lanes contained 10 g of RNA from E. coli DH5␣ harboring pRL1CPR or pNd137 harvested at various growth stages. A digoxigenin-labeled internal PmaCI-AflII DNA fragment of the lsrA gene was used as the probe. The position of hybridizing RNAs (⬃1.3 kb) is indicated by an arrow.
transcriptional level in a growth phase-dependent manner. Moreover, the DNA sequence residing between 1.3 and 0.1 kb upstream from the lsrA gene seems to contain a regulatory component(s) responsible for the drastic increase in lsrA expression in the late growth phase. Most bacterial genes are transcribed by the RNA polymerase complex employing sigma factor 70 in the growing state. On the other hand, transcription induced under starvation conditions in the stationary phase of cell growth is carried out by the RNA polymerase complex employing another sigma factor, S (formerly RpoS) (23). In order to examine whether the stationary-phase-specific transcription of the lsrA gene was dependent upon S, expression of the lsrA gene from plasmids pRL1CPR and pNd137 in E. coli W3110 and isogenic rpoS mutant strain CP1005 (rpoS::Tn10) was analyzed by measuring levansucrase activity. The lsrA gene from pRL1CPR was expressed in a growth phase-dependent manner in both W3110 and CP1005 (Fig. 2), eliminating the possibility that RpoS mediates regulation of lsrA gene expression. Identification of transcription initiation site of the lsrA gene. To characterize the regulatory element(s) responsible for growth phase-dependent expression of the lsrA gene in E. coli, we first determined the transcription initiation site of the lsrA gene by primer extension analysis. RNA samples from latestage cultures of E. coli harboring pRL1CPR or pNd137 were analyzed by using a primer designed to span the nucleotide sequence of the lsrA coding region. The results obtained indicate that lsrA transcription initiates 51 bp upstream of the coding region of lsrA (Fig. 3). Moreover, transcription starts at the same site regardless of growth phase-dependent or -independent expression, since pRL1CPR and pNd137 produced transcripts starting at the same site. Close inspection of the nucleotide sequences upstream of the transcription initiation site revealed potential ⫺35 (CTGAGT) and ⫺10 sequences (TGTAAC) of prokaryotic vegetative promoters (see Fig. 5). Regulation of lsrA gene expression by two opposite regulatory functions from the upstream region. In order to search for a regulatory element(s) involved in the growth phase-dependent expression of lsrA, various subclones with shortened upstream sequences of the lsrA gene were constructed (Fig. 4A). The expression profiles of the lsrA gene from the subclones were examined (Fig. 4B) by measuring levansucrase activity. Growth phase-dependent regulation of gene expression was
maintained until the upstream extension was shortened to 798 bp upstream of the transcription initiation site. Further deletion of another 56 bp (subclone ⫺742) eliminated the stationary-phase-specific stimulation of lsrA expression. This strongly indicated that the nucleotide sequences in this 56-bp region were important for the growth phase-dependent increase in lsrA expression in the late phase of cell growth. In addition, the lower levels of both early- and late-phase expression of the ⫺742 construct compared to the levels expressed from pNd137 suggested that there is an additional negative regulatory function. The nucleotide sequences required for the negative regulation were located within 257 bp (subclone ⫺204) upstream of the lsrA gene (Fig. 4B), and this region was designated lsrR (Fig. 5). Identification of the lsrS gene encoding a protein homolo-
FIG. 2. Expression profiles of the lsrA gene in E. coli W3110 (A) and CP1005 (rpoS::Tn10) (B). The strains harboring pRL1CPR (squares) or pNd137 (circles) were cultured in LB medium supplemented with ampicillin (100 g/ml) and tetracycline (20 g/ml) at 37°C with shaking at 100 rpm. At intervals, the A600 (solid symbols) and the enzyme activity (open symbols) were measured.
VOL. 184, 2002
FIG. 3. Mapping of 5⬘ terminus of the lsrA transcript. Primer extension reactions were performed with 5 g of RNA extracted from a stationary-phase culture (A600, 4.5) of E. coli harboring pRL1CPR (lane 1) or pNd137 (lane 2), as described in Materials and Methods. Lanes G, A, T, and C show the results of the dideoxy sequencing reactions carried out with pRL1CPR as the template and the same oligonucleotide as the primer. The arrow indicates the position of the primer extension product. The sequence surrounding the transcription initiation site of the lsrA gene (⫹1, boldface G) is shown.
gous to Ogr, a late gene activator of bacteriophage P2. Sequence analysis revealed two open reading frames (ORFs) that overlapped the 56-bp region necessary for positive regulation (Fig. 5). ORF1 (⫺906 to ⫺727 bp) encodes a protein consisting of 59 amino acid residues, and ORF2 (⫺591 to ⫺803 bp) encodes a protein consisting of 70 amino acid residues. According to the deletion analysis, ORF2 is likely to be involved in the positive regulation of lsrA expression (Fig. 4). Neither the basal activity nor the induced activity was affected by the truncation at ⫺798, where sequences encoding the C-terminal 23 amino acid residues of the ORF1 product were deleted, whereas the late-phase activation was completely eliminated by truncation at ⫺742, where sequences encoding the N-terminal 20 amino acids of the ORF2 product were deleted. This finding was supported by multiple alignments of the deduced amino acid sequences of the ORF products obtained by using the BLAST program. The deduced amino acid sequence of the ORF1 product did not show homology to proteins reported previously. In contrast, the deduced amino acid sequence of the product of ORF2 (designated lsrS) showed a high level of similarity to the bacteriophage P2 Ogr family of transcriptional activators, such as the Ogr protein of phage P2 and E. coli K-12 cryptic phage (44%), the B protein of phage 186 (50%), the ␦ protein of phages P4 and R73 (53%), and the NucC and
REGULATION OF R. AQUATILIS lsrA
5865
RecC proteins of Serratia marcescens (48%) (Fig. 6) (5, 10, 17, 19, 22, 28, 30, 32). P2 Ogr is a zinc-binding protein, and four cysteine residues (C-X2-C-X22-C-X4-C) play a role in the binding process (21). As shown in Fig. 6, this feature is also well conserved in the deduced amino acid sequences of the lsrS gene product. It is also noteworthy that N-terminal residues are more closely conserved in these proteins. In fact, it has been reported that deletion of the C-terminal 21 amino acid residues did not alter the activity of the P2 Ogr protein, suggesting the importance of N-terminal residues for activity (11). Further functional analysis of the LsrS protein was carried out as follows. A 0.6-kb NheI fragment including the lsrS gene (Fig. 5) was blunted with the Klenow fragment and ligated into the EcoRV site of pACYC184, resulting in pACNR(lsrS). This plasmid was cotransformed into E. coli with one of various subclones containing the lsrA gene with different lengths of upstream sequence and was assayed for levansucrase activity. Cotransformation with pACNR(lsrS) restored the growth phase-dependent expression of lsrA from subclones ⫺742, ⫺419, ⫺274, and ⫺204 (Fig. 7), confirming that LsrS is responsible for the stationary-phase induction of lsrA. The activation effect of the LsrS protein was more striking in the ⫺86 subclone, which lacked sequences responsible for repression. lsrA expression was increased approximately 30-fold by introduction of pACNR(lsrS) in a growth phase-independent fashion (Fig. 7). Analysis of lsrR region. To further characterize the lsrR region, the DNA sequences around this region were deleted by using the PCR method with synthetic oligonucleotides to obtain constructs with end points (Fig. 8A) at ⫺175, ⫺146, ⫺119, and ⫺55, as well as a construct containing 204 bp with a deletion from ⫺67 to ⫺55 (Fig. 8B). (i) Potential binding sites of LsrS protein. Two potential LsrS binding sites were detected based on nucleotide sequence homology with the well-characterized P2 Ogr binding site sequence TGT-N12-ACA (33). The first binding site (binding site I), centered at ⫺55, exhibited an exact match with the sequence of the P2 Ogr binding site. Deletion of sequences upstream of ⫺55 resulted in loss of LsrS-dependent activation (Fig. 8B). However, lsrA expression could be still induced by LsrS even when binding site I was destroyed by deletion of sequences between ⫺67 and ⫺55 (Fig. 8B), suggesting that a second binding site exists. In fact, another putative LsrS binding site (binding site II), TGT-N10-ACA, was identified at ⫺119 (Fig. 8A). Deletion of binding site II also resulted in a decrease in activation by LsrS (Fig. 8B), although the extent of the decrease was less than that resulting from deletion of binding site I, implying that binding site II also participates in LsrS-dependent activation. (ii) Identification of two DNA regions necessary for repression. The absence of an ORF in the lsrR region suggested that these sequences function as a cis-acting regulatory element repressing lsrA expression during both the early and stationary phases of growth. When 29 bp between ⫺204 and ⫺175 was deleted, lsrA gene expression in the presence of LsrS was increased about 27-fold compared to that of the ⫺204 subclone (Fig. 8B). This region was designated R1. When nucleotide sequences spanning ⫺67 to ⫺55 were deleted from subclone ⫺204, the lsrA expression in the presence of LsrS was approximately fivefold higher than that of the ⫺204 subclone, sug-
5866
J. BACTERIOL.
SEO ET AL.
FIG. 4. Effects of upstream deletions on lsrA expression. (A) Deletion subclones of the upstream region. The designations of the deletion subclones are based on the number of base pairs present upstream of the transcription initiation site (⫹1). The open boxes and the arrow indicate the lsrA gene and its site of transcription initiation, respectively. (B) Expression profiles of the lsrA gene in the deletion subclones. Cell-free lysates prepared from cultures in two growth phases, the early phase (A600, 2) and the late phase (A600, 4.5), were assayed for the enzyme activity.
gesting that this 13-bp region, designated R2, has a regulatory role. It is interesting that there is an inverted repeat centered at ⫺60, although its role has not been determined. DISCUSSION In this study, we identified regulatory elements which coordinate growth phase-dependent regulation of the recombinant lsrA gene of R. aquatilis in E. coli (Fig. 9). The LsrS protein activated lsrA expression up to 20-fold during the late phase of cell growth. In addition, a cis-acting lsrR region was identified that represses expression of the lsrA gene 10-fold. Enzyme activity assays of subclones with various lengths of upstream sequences resulted in identification of at least two DNA regions (R1 and R2) responsible for repression (Fig. 8).
The LsrS protein exhibits amino acid sequence homology with and shares properties with members of the phage P2 Ogr family of transcription activators. The phage proteins are known to activate transcription of phage late genes (13, 15). Jin et al. (17) and Ferrer et al. (10) cloned and characterized cryptic prophage genes, including the P2 ogr homologs (nucC and recC) from S. marcescens. The gene products NucC and RecC are transcriptional activators of extracellular nuclease and bacteriocin 28b, respectively. Interestingly, this regulation is also subject to the cell growth phase (10, 17). Julien and Calendar identified the binding sequence of the P4 ␦ protein, a member of the P2 Ogr family (18). Based on nucleotide sequence homology with the well-characterized P2 Ogr binding site and deletion analysis (Fig. 8), we identified two potential LsrS binding sites upstream of the lsrA promoter. The se-
VOL. 184, 2002
REGULATION OF R. AQUATILIS lsrA
5867
FIG. 5. Nucleotide sequence of the upstream region of lsrA. All numbers shown in the nucleotide sequence are based on the transcription initiation site of lsrA (⫹1). The arrows indicate 5⬘ ends upstream of the lsrA gene in the deletion subclones (Fig. 4). The deduced amino acid sequences of the products of two overlapping ORFs are shown below (ORF1; lowercase letters; ⫺906 to ⫺727 bp) and above (ORF2 or LsrS; uppercase letters; ⫺591 to ⫺803 bp) the nucleotide sequence. The NheI sites used in the construction of pACNR(lsrS) are indicated. The region enclosed in a box is the lsrR region. The potential promoter sequence of the lsrA gene is underlined. The potential binding site for LsrS is double underlined. For the nucleotide sequence of the lsrR and lsrA regions, the complementary strand is shown in lowercase letters.
quence of binding site I, centered at ⫺55, is an exact match with the consensus binding site sequence of the P2 Ogr family proteins. The P2 Ogr protein activates the transcription of the target gene through direct interaction with the ␣ subunit of the
RNA polymerase complex (2, 12). The proximity of binding site I to the promoter corresponds to the predicted model for action of the activators. Binding site II was found to be centered at ⫺119, considerably upstream from the lsrA promoter,
5868
SEO ET AL.
J. BACTERIOL.
FIG. 6. Homology alignment of the deduced amino acid sequence of the lsrS gene product. Asterisks indicate conserved residues. Four cysteine residues of the potential zinc-binding site are indicated by boldface type. NucC and RecC, transcriptional activators of extracellular nuclease (nucA) and bacteriocin b28 (bss) of S. marcescens, respectively; P2 Ogr, 186 B, and R73 ␦, late gene transcriptional activators of bacteriophages P2, 186, and R73, respectively; P4 ␦ N-region and P4 ␦ C-region, amino and carboxyl regions of late gene activator of P4 phage, respectively; and K-12 Ogr, P2 ogr homologous gene of cryptic phage on E. coli K-12 chromosome.
and its sequence, TGT-N10-ACA, was not an exact match with the consensus sequence. We have demonstrated that this site is functional through deletion analysis. Moreover, we have observed the presence of binding site II binding protein in crude
extracts from late-phase cells of R. aquatilis by using an electrophoretic mobility shift assay (unpublished data). To our knowledge, this is the first example of a transcriptional activator of a member of the P2 Ogr family interacting with the
FIG. 7. Effect of the lsrR region on the stationary-phase-specific expression of lsrA by the LsrS protein. Two deletion subclones, ⫺204 (circles) and ⫺86 (pNd137) (triangles), were transformed with pACNR(lsrS) (solid symbols) and pACYC184 (open symbols), grown in LB media supplemented with ampicillin (100 g/ml) and chloramphenicol (50 g/ml), and assayed for lsrA expression at intervals during the cultures.
VOL. 184, 2002
REGULATION OF R. AQUATILIS lsrA
5869
FIG. 8. (A) Sequence analysis of the lsrR region. All numbers shown on the nucleotide sequence are based on the transcription initiation site of lsrA (⫹1). Two putative binding sequences (binding sites I and II) of LsrS are indicated by boldface type. An inverted repeat sequence covering the R2 region (B) is indicated. The arrows indicate 5⬘ ends of deletion subclones prepared as described in Materials and Methods. (B) Deletion analysis of the lsrR region. Boxes I and II indicate putative binding sites for LsrS. R1 and R2 indicate nucleotide regions involved in repression of lsrA expression. The level of activation was determined by dividing the LsrA activity of E. coli containing plasmid pACNR(lsrS) by the LsrA activity of E. coli containing plasmid pACYC184.
second binding site remote from the target promoter. Although the mechanism of action of the LsrS activator at the remote binding site is unknown, we suggest that looping or bending must occur to allow interaction between the activator and the RNA polymerase complex (6). The late-phase-specific activation of lsrA expression by LsrS may be explained by a property of the P2 Ogr family activities (e.g., autoactivation) (4, 10, 17). We have found a putative LsrS binding sequence upstream of the lsrS gene (TGTATCAGACAGTAAGTACA) (Fig. 5). The presence of the lsrR region, in which two DNA regions (R1 at ⫺204 to ⬃⫺175 bp and R2 at ⫺67 to⬃⫺55 bp) have been defined, repressed expression of the lsrA gene in all
phases of cell growth. Consistent with this result, a protein that interacted with R1 sequences was detected during all growth phases of R. aquatilis by an electrophoretic mobility shift assay (unpublished data), indicating that the R1 region might be an operator sequence for a repressor involved in the regulation of lsrA expression. Interestingly, an R1 binding protein was also detected with E. coli crude extracts. This probably explains why lsrA expression could be repressed by the presence of the R1 sequence in E. coli. Like the lsrA gene, the nucC and recC genes of S. marcescens are also negatively regulated via the SOS system (10, 17). Finally, we considered whether the lsrA gene was also subject to growth phase-dependent regulation in R. aquatilis. A
5870
SEO ET AL.
FIG. 9. Summary of regulatory functions of the lsrA gene identified in this work.
modest increase in lsrA expression was detected upon entrance into the late phase of cell growth in R. aquatilis (data not shown). The increase was not as large as that observed in E. coli because of the high basal level of gene expression in wildtype R. aquatilis. However, the data obtained in this study of the regulation of R. aquatilis lsrA in E. coli could still provide some information concerning the gene regulation mechanism in R. aquatilis. Finally, we eliminated the possibility of involvement of the stationary-phase-specific sigma factor S, a sigma factor which regulates many cell growth phase-dependent bacterial genes (23), in the regulation of lsrA (Fig. 2). Thus, analysis of the regulation of the lsrA gene by coordination of two opposite regulatory functions could provide insights into a new growth phase-dependent gene regulation mechanism. ACKNOWLEDGMENTS We thank S. Shin, S. G. Song, D. S. Lee, J. G. Pan, and C. Park for providing the E. coli strain. We are grateful to O. S. Kwon for his critical reading of the manuscript. This work was supported by the Ministry of Science and Technology, Republic of Korea. REFERENCES 1. Arrieta, J., L. Hernandez, A. Coego, V. Suarez, E. Balmori, C. Menendez, M. F. Petit-Glatron, R. Chambert, and G. Selman-Housein. 1996. Molecular characterization of the levansucrase gene from the endophytic sugarcane bacterium Acetobacter diazotrophicus SRT4. Microbiology 142:1077–1085. 2. Ayers, D. J., M. G. Sunshine, E. W. Six, and G. E. Christie. 1994. Mutations affecting two adjacent amino acid residues in the alpha subunit of RNA polymerase block transcriptional activation by the bacteriophage P2 Ogr protein. J. Bacteriol. 176:7430–7438. 3. Berge, O., T. Heulin, W. Achouak, C. Richard, R. Bally, and J. Balandreau. 1991. Rahnella aquatilis, a nitrogen-fixing enteric bacterium associated with the rhizosphere of wheat and maize. Can. J. Microbiol. 37:195–203. 4. Birkeland, N. K., B. H. Lindqvist, and G. E. Christie. 1991. Control of bacteriophage P2 gene expression: analysis of transcription of the ogr gene. J. Bacteriol. 173:6927–6934. 5. Birkeland, N. K., and B. H. Lindqvist. 1986. Coliphage P2 late control gene ogr, DNA sequence and product identification. J. Mol. Biol. 188:487–490. 6. Busby, S., and R. H. Ebright. 1994. Promoter structure, promoter recognition, and transcription activation in prokaryotes. Cell 79:743–746. 7. Chambert, R., G. Treboul, and R. Dedonder. 1974. Kinetic studies of levansucrase of Bacillus subtilis. Eur. J. Biochem. 41:285–300. 8. Crutz, A.-M., M. Steinmetz, S. Aymerich, R. Richter, and D. Le Coq. 1990. Induction of levansucrase in Bacillus subtilis: an antitermination mechanism negatively controlled by the phosphotransferase system. J. Bacteriol. 172: 1043–1050.
J. BACTERIOL. 9. Crutz, A.-M., and M. Steinmetz. 1992. Transcription of Bacillus subtilis sacX and sacY genes, encoding regulators of sucrose metabolism, is both inducible by sucrose and controlled by the DegS-DegU signaling system. J. Bacteriol. 174:6087–6095. 10. Ferrer, S., M. B. Biejo, J. F. Guasch, J. Enfedaque, and M. Regue. 1996. Genetic evidence for an activator required for induction of colicin-like bacteriocin 28b production in Serratia marcescens by DNA-damaging agents. J. Bacteriol. 178:951–960. 11. Gebhardt, K., R. A. King, G. E. Christie, and B. H. Lindqvist. 1993. Mutational analysis of the bacteriophage P2 Ogr protein: truncation of the carboxy terminus. J. Bacteriol. 175:7724–7726. 12. Gralla, J. D. 1990. Promoter recognition and mRNA initiation by Escherichia coli E-sigma-70. Methods Enzymol. 185:37–54. 13. Grambow, N. J., N. K. Birkeland, D. L. Anders, and G. E. Christie. 1990. Deletion analysis of a bacteriophage P2 late promoter. Gene 95:9–15. 14. Gross, M., G. Geier, K. Rudolph, and K. Geider. 1992. Levan and levansucrase synthesized by the fireblight pathogen Erwinia amylovora. Physiol. Mol. Plant Pathol. 40:371–381. 15. Halling, C., and R. Calendar. 1990. Bacteriophage P2 ogr and P4 ␦ genes act independently and are essential for P4 multiplication. J. Bacteriol. 172:3549– 3558. 16. Hanahan, D. 1983. Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166:557–580. 17. Jin, S., Y.-C. Chen, G. E. Christie, and M. J. Benedik. 1996. Regulation of the Serratia marcescens extracellular nuclease: positive control by a homolog of P2 Ogr encoded by a cryptic prophage. J. Mol. Biol. 256:264–278. 18. Julien, B., and R. Calendar. 1995. Purification and characterization of the bacteriophage P4 ␦ protein. J. Bacteriol. 177:3743–3751. 19. Kalionis, B., I. B. Dodd, and J. B. Egan. 1986. Control of gene expression in the P2-related template coliphages. J. Mol. Biol. 191:199–209. 20. Kunst, F., and G. Rapoport. 1995. Salt stress is an environmental signal affecting degradative enzyme synthesis in Bacillus subtilis. J. Bacteriol. 177: 2403–2407. 21. Lee, T.-C., and G. E. Christie. 1990. Purification and properties of the bacteriophage P2 ogr gene product. J. Biol. Chem. 265:7472–7477. 22. Lin, C.-S. 1984. Nucleotide sequence of the essential region of bacteriophage P4. Nucleic Acids Res. 12:8667–8684. 23. Loewen, P. C., and R. Hengge-Aronis. 1994. The role of the sigma factor S (KatF) in bacterial global regulation. Annu. Rev. Microbiol. 48:53–80. 24. O’Mullan, P. J., M. Szakucs-Dobozi, and D. E. Eveleigh. 1991. Identification of saccharolytic enzymes of Zymomonas mobilis CP4. Biotechnol. Lett. 13: 137–142. 25. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 26. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463–5467. 27. Seo, J.-W., K.-H. Jang, K.-B. Song, C.-H. Kim, B.-H. Chung, and S.-K. Rhee. 2000. Molecular cloning of a gene encoding the thermoactive levansucrase from Rahnella aquatilis and its growth phase-dependent expression in Escherichia coli. J. Biotechnol. 81:63–72. 28. Slettan, A., K. Gebhardt, E. Kristiansen, N. K. Birkeland, and B. H. Lindqvist. 1992. Escherichia coli K-12 and B contain functional bacteriophage P2 ogr genes. J. Bacteriol. 174:4094–4100. 29. Steinmetz, M., D. Le Coq, S. Aymerich, G. Geonzy-Treoul, and P. Gay. 1985. The DNA sequence of the gene for the secreted Bacillus subtilis enzyme levansucrase and its genetic control sites. Mol. Gen. Genet. 200:220–228. 30. Sun, J., M. Inouye, and S. Inouye. 1991. Association of a retroelement with a P4-like cryptic prophage (retronphage R73) integrated into the selenocystyl tRNA gene of Escherichia coli. J. Bacteriol. 173:4171–4181. 31. Tortosa, R., and D. Le Coq. 1995. A ribonucleic antiterminator sequence (RAT) and a distant palindrome are both involved in sucrose induction of the Bacillus subtilis sacXY regulatory operon. Microbiology 141:2921–2927. 32. Van Bokkelen, G. B., E. C. Dale, C. Halling, and R. Calendar. 1991. Mutational analysis of a bacteriophage P4 late promoter. J. Bacteriol. 173:37–45. 33. Winslow, R. H., B. Julin, R. Calendar, and G. E. Christie. 1998. An upstream sequence element required for NucC-dependent expression of the Serratia marcescens extracellular nuclease. J. Bacteriol. 180:6064–6067.
