Blackwell Science, LtdOxford, UKMMIMolecular Microbiology0950-382X© 2005 The Authors; Journal compilation © 2005 Blackwell Publishing Ltd? 2005??••••Original ArticlePseudomonas aeruginosa QscR activityJ.-H. Lee, Y. Lequette and E. P. Greenberg

Molecular Microbiology (2005)

doi:10.1111/j.1365-2958.2005.04960.x

Activity of purified QscR, a Pseudomonas aeruginosa orphan quorum-sensing transcription factor

Joon-Hee Lee, Yannick Lequette and E. Peter Greenberg* Department of Microbiology School of Medicine, University of Washington, Seattle, WA 98195-7242, USA. Summary The opportunistic pathogen Pseudomonas aeruginosa has two acyl-homoserine lactone (acyl-HSL) signalling systems, LasR-I and RhlR-I. LasI catalyses the synthesis of N-3-oxododecanoyl homoserine lactone (3OC12) and LasR is a transcription factor that requires 3OC12 as a ligand. RhlI catalyses the synthesis of N-butanoyl homoserine lactone (C4) and RhlR is a transcription factor that responds to C4. LasR and RhlR control the transcription of hundreds of P. aeruginosa genes. There is a third P. aeruginosa LasR-RhlR homologue encoded by qscR for which there is no cognate acyl-HSL synthase gene. To test the hypothesis that QscR functions by direct control of specific promoters in an acyl-HSL-dependent manner we purified QscR and characterized QscR activity in vitro. We also studied QscR activity in recombinant Escherichia coli. QscR binds to promoters that have elements similar in sequence to those found in LasRor RhlR-dependent promoters but QscR does not bind to the LasR- or RhlR-specific promoters we examined. QscR binding to DNA requires 3OC12, but QscR exhibits a relaxed acyl-HSL specificity compared with the 3OC12-cognate signal receptor LasR. Our results support the hypothesis that there is a specific QscRdependent regulon. We show that QscR controls genes in this regulon directly and that regulation is dependent on an acyl-HSL produced by LasI. Because of its relaxed signal specificity QscR may also respond to acyl-HSLs made by other bacteria in mixed bacterial communities. Introduction Quorum sensing controls expression of many virulence genes in the opportunistic human pathogen PseudomoAccepted 14 October, 2005. *For correspondence. E-mail [email protected]; Tel. (+1) 206 221 2797; Fax (+1) 206 616 2938.

© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd

nas aeruginosa (Fuqua et al., 2001; Smith and Iglewski, 2003). There are two acyl-homoserine lactone (acyl-HSL) quorum-sensing signals produced by P. aeruginosa, N-3oxododecanoyl homoserine lactone (3OC12) and Nbutanoyl homoserine lactone (C4). The 3OC12 signal is generated by an acyl-HSL synthase called LasI and the receptor is a transcriptional activator called LasR. The lasI and lasR genes are adjacent to each other. The C4 signal is produced by RhlI, and the transcriptional activator RhlR responds to this signal. The rhlI and rhlR genes are adjacent to each other (for reviews of quorum sensing in P. aeruginosa see Fuqua et al., 2001; Fuqua and Greenberg, 2002; Smith and Iglewski, 2003). The LasI-R and RhlI-R systems control expression of several hundred P. aeruginosa genes (Schuster et al., 2003; Wagner et al., 2003). The network of control for these genes is complicated (Wagner et al., 2003; Schuster et al., 2004a; Wu et al., 2004) and many may be controlled by LasR and RhlR indirectly. There is an additional P. aeruginosa gene that codes for a homologue of LasR and RhlR. This homologue is called QscR and it has no corresponding LasI or RhlI homologue. It is an orphan quorum-sensing signal receptor. QscR mutants are hypervirulent (Chugani et al., 2001) and a number of genes controlled by the other acyl-HSL signalling systems are repressed by QscR (Chugani et al., 2001). There are several possible mechanisms for QscR repression of LasR- or RhlR-activated genes. The QscR protein has been shown not only to form homomultimers but also heteromultimers with LasR and RhlR in vivo (Ledgham et al., 2003). Obviously, heteromultimer formation could interfere with the activity of LasR and RhlR. QscR might also bind 3OC12 or C4 and compete with LasR and RhlR for these signals. Although the idea has not received much attention, QscR might function by direct binding as a homomultimer to specific promoters. In this case it could function in an acyl-HSL-independent manner or it could borrow a signal produced by LasI or RhlI. Recently, the binding of purified LasR to a number of LasR-dependent promoters was reported (Schuster et al., 2004b). Several promoters served as good substrates for DNA binding. Others, even promoters with elements that appeared similar to known LasR binding sites, did not serve as binding sites for LasR. One might suspect that

2 J.-H. Lee, Y. Lequette and E. P. Greenberg LasR regulates these other promoters indirectly. One such LasR-dependent promoter that does not serve as a binding substrate for purified LasR is the PA1897 promoter (Whiteley and Greenberg, 2001; Schuster et al., 2004b). PA1897 is adjacent to and divergently transcribed from qscR (Stover et al., 2000). Thus, PA1897 is a candidate for a gene controlled by QscR directly. We thus purified QscR and tested the ability of the purified protein to bind the PA1897 promoter region. We show that it binds this promoter and promoters with related sequences and the binding depends on the 3OC12 signal produced by LasI. Furthermore, QscR did not bind to LasR-dependent promoters we tested.

ever, we used a P. aeruginosa strain capable of acyl-HSL production. To determine whether folding of QscR into a soluble protein requires the presence of an acyl-HSL, we constructed a PAO-T7 derivative with lasI and rhlI mutations and studied pJLQhis-directed expression of Histagged QscR in this strain. In the absence of added acylHSL or in the presence of 20 µM C4 we detected no soluble QscR in cell lysates. In the presence of 20 µM 3OC12 about half of the total QscR was soluble (data not shown). Thus, like other LuxR homologues studied to date, QscR requires the presence of an appropriate acylHSL during cell growth for folding into a soluble, active form.

Results

Specific 3OC12-dependent QscR promoter binding

Purification of QscR

The qscR gene is adjacent to and divergently transcribed from PA1897, a gene of unknown function. The intergenic region is 761 bp. A previous transcriptome analysis showed that PA1897 was activated by 3OC12 (Schuster et al., 2003). A subsequent study showed that purified LasR did not bind to the PA1897 promoter region in vitro (Schuster et al., 2004b). There is an inverted repeat centred about 45 bp from the transcript start site of PA1897 that shows sequence similarity to binding sites for a variety of other acyl-HSL regulator binding sites (Whiteley and Greenberg, 2001). These data suggest that QscR might bind to the PA1897 promoter directly. Thus, we wanted to test the ability of QscR to bind the PA1897 promoter region. We found that QscR binds specifically to a probe that extends 300 bp from the PA1897 translation start codon and includes the inverted repeat element (Fig. 1B). It does not bind to a 216 bp DNA fragment that extends up to but does not include the inverted repeat (data not shown). QscR had an apparent affinity for the PA1897 promoter of about 0.3 nM and Hill coefficient of 1.8, indicating cooperative binding (see below). To determine more precisely where QscR binds to the 339 bp DNA fragment we performed a DNase I footprint analysis (Fig. 2). As predicted, QscR protects the 20 bp inverted repeat sequence in the presence of 3OC12. In fact the protected area extends into the −35 region (from −54 to −29). The gel shift and footprinting data support the hypothesis that QscR regulates PA1897 directly. Is QscR binding to the PA1897 promoter region sufficient to activate transcription? To address this question we measured β-galactosidase activity in Escherichia coli containing a PA1897–lacZ reporter. There is a strong activation of lacZ expression in E. coli containing the qscR expression plasmid pJN105Q and the PA1897–lacZ reporter pJL101 in response to 10 µM 3OC12, 4000 units compared with <100 units without 3OC12. Thus, we conclude QscR binding to the PA1897 promoter region can trigger transcription.

We used P. aeruginosa PAO-T7 containing the expression vector pJLQhis as a source of His-tagged QscR (see Experimental procedures). After nickel-affinity chromatography of the soluble polypeptides we obtained a fraction that was judged by SDS-PAGE analysis to be >95% QscR (Fig. 1A). Other reports have shown the production of soluble LasR (Schuster et al., 2004b), Vibrio fischeri LuxR, and Agrobacterium tumefaciens TraR (Zhu and Winans, 1999) required the presence of the cognate acylHSL in the growth medium used for bacterial expression. We did not add acyl-HSLs to our culture medium; howA

B 1

2

3

4

83 62 47.5 32.5 25

B F

16.5

N

Fig. 1. The 3OC12-dependent binding of His-tagged QscR to the PA1897 promoter region. A. SDS-PAGE of His-tagged QscR after nickel column affinity chromatography. Left lane shows molecular mass standards and right lane shows column purified polypeptides (pooled fractions from 100 mM to 400 mM in the imidazole gradient) stained with Coomassie brilliant blue. The masses of the standards are shown to the left of the gel (kDa). B. Binding of purified QscR to a PA1897 promoter fragment as measured by gel shift assay. Lane 1, control with no QscR in the reaction mixture. Lane 2, 1 nM QscR and no added 3OC12. Lane 3, 0.35 nM QscR plus 20 µM 3OC12. Lane 4, 1 nM QscR plus 20 µM 3OC12. The non-specific DNA (N), the QscR-free target DNA (F) and the QscR-bound DNA (B) are indicated by the arrowheads. The promoter fragment is 339 bp and it includes 300 bp upstream of PA1897.

© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology

Pseudomonas aeruginosa QscR activity 3 A

B –54 –29 TCTCTCCGCAGATACCTGCCCGGAAGGGCAGGTTGTCCC –35 TGCCGGGCTGTGACAATTTAATTCGACCAGGCATTTCAT –10 TGTCCGTGCCGATTTTCACGAAGCGCATTCTGAGGCAAT TAAAAAGAGCGCTCCATTCGACCATGGACAAGCTATCCA NcoI CGCCTGACCGAGATCGCCTTCCGAATATAGCGAAGCGAT AACCGCAGCCTGCCGAGAAGTGCTTCAGACAATAAACAG GACGCTGGCCTTTCGTATCGATGAAAGTTCCGCATGGCG TCCGCCCCTAAGGAAGAGGAGATAAATATGATTTATTAC TTGATCGGAGTGGCGCTATTCATCTTC

Fig. 2. Identification of the QscR binding site in the PA1897 promoter region by DNase I footprinting. A. DNase I footprinting gel with sequencing ladder on the left, followed by DNase I-digested target DNA without QscR, with 3 nM and 9 nM His-tagged QscR from left to right. The target DNA is a 141 bp fragment of the qscR-PA1897 intergenic region extending from the NcoI site towards qscR. Reaction mixtures contained 5 µM 3OC12. B. Sequence of the qscR-PA1897 intergenic region. The NcoI site is underlined. The PA1897 translational start codon is indicated in bold. The transcription start site (Whiteley and Greenberg, 2001) is indicated by the arrow. The predicted −10 and −35 regions are underlined and labelled. The QscR binding site protected from DNase I digestion is boxed and dyad symmetry is indicated in bold.

hypothesis that positions 8 and 13 are critical, we showed the probe with the artificial sequence serves as a binding substrate for QscR (Fig. 3). Furthermore, a probe containing the PA1897 QscR binding site with bp 8 and 13 changed to a LasR-binding 8A-13T motif does not serve as a QscR target (Fig. 3). PA1897 is on one flank of qscR. The phenazine synthesis phzA2 operon is on the other flank. Like the hcn genes the phenazine synthesis genes are activated by quorum sensing (Schuster et al., 2003) and transiently repressed by QscR (Chugani et al., 2001). Thus, we wanted to determine whether QscR binds to the phzA2 promoter region. We could not detect any binding of QscR to the region between qscR and phzA2 (data now shown). Furthermore, we could not detect any binding to the promoter region of the unlinked phzA1 operon. This suggests that QscR control of phenazine biosynthesis and hydrogen cyanide synthesis may not be through a direct interaction of QscR with promoter elements of phz and hcn genes. Do any other genes regulated by QscR in vivo contain QscR-binding elements? We searched the P. aeruginosa genome for promoter elements with sequences similar to

A

PA1897mut

GCTACCTGCCAGTTCTGGCAGGTTTG

lasB mut 0

T

ACCTGCCCGGAAGGGCAGGTTGTCCC

PA1897 lasB

C .35

1

3.5

10

PA1897wt

G 0

.35

1

3.5

10

1

3.5

10

PA1897mut

DNA-binding specificity of QscR We tested the ability of purified QscR to bind to hcnA and lasB probes shown previously to contain elements sufficient for LasR binding (Schuster et al., 2004b). QscR has been shown to repress hcnA transcription in P. aeruginosa (Chugani et al., 2001). Gel shift experiments showed no evidence of binding to the hcnA (data not shown) and lasB (Fig. 3) probes. The LasR binding site in the lasB probe is identical to the QscR binding site in PA1897 in 15 out of the 20 bp of the inverted repeat. Nevertheless, just as QscR does not bind to the lasB promoter, LasR does not bind to the PA1897 promoter (Fig. 3 and Schuster et al., 2004b). Previous work has suggested that positions 8 and 13 in P. aeruginosa quorum sensing-dependent promoters are important specificity determinants (Whiteley and Greenberg, 2001). Positions 8 and 13 in the LasR binding site of the lasB promoter represent two of the five bases that diverge from the PA1897 binding site. The QscR box in the PA1897 has an 8C-13G motif. We created a probe with 8C-13G in place of 8A-13T in the lasB promoter element to test the hypothesis that these bases were critical determinants for QscR binding. In support of the

B F N 0 lasB

.35

1

3.5

10

0

.35

lasBmut

B F

Fig. 3. The 8C-13G motif in the QscR binding site is critical for specific QscR binding. Top: an alignment of the PA1897 promoter region sequence protected from DNase I digestion by QscR and the sequence of the lasB promoter protected by LasR (Schuster et al., 2004b). The changes made to create the mutated PA1897 (PA1897mut) and mutated lasB (lasBmut) DNA are indicated by the arrows. Bottom: gel shift assays of QscR binding to wild-type and mutant qscR and lasR boxes. All reaction mixtures contained 2 µM 3OC12 and His-tagged QscR at the concentrations indicated. The PA1897 probe DNA is as described in Fig. 1. The lasB DNA fragment is as described in Schuster et al. (2004b).

© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology

4 J.-H. Lee, Y. Lequette and E. P. Greenberg 0

PA1897 25 35 90 180 260 350 875 1800 2600

PA 5351 100 350 1000 3500 10000

0

B F N

●

0

● -2 ● 1.5

●

●

0

● -2

2

2.5

3

log[QscR(pM)]

●

●

●

● 2

2.5

3 3.5 log[QscR(pM)]

the QscR binding site in PA1897. One gene that is not influenced by LasR or RhlR, PA5351, contains an element with identity to the PA1897 palindrome in 14 out of the 20 bp. This element is an imperfect inverted repeat but it has the critical 8C-13G residues. QscR shows 3OC12dependent binding to DNA containing the imperfect inverted repeat of PA5351, but it has a lower apparent binding affinity than the affinity to the PA1897 binding site, 4 nM compared with 0.3 nM for PA1897, and QscR does not bind to PA5351 cooperatively (Fig. 4). We imagine that QscR binds to promoters of P. aeruginosa genes other than PA1897 and PA5351 but this remains to be determined. Purified QscR exists as a monomer in solution Purified QscR shows a predicted molecular mass of about 29 K (Fig. 5). This is consistent with the monomeric molecular weight of the His-tagged polypeptide (29.3 K). The molecular mass is not altered by 3OC12 (data not shown). This is in contrast to several other QscR homologues that have been purified including LasR (Schuster et al., 2004b), TraR (Zhu and Winans, 2001) and CarR (Welch et al., 2000), which exist as homodimers in solution. LuxR has also been purified but the oligomeric state of this protein in solution is not known (Urbanowski et al., 2004). Our footprint experiments (Fig. 2) indicate that QscR is bound to regulatory DNA as a dimer. The region of DNA protected from DNase I is similar in length to regions protected by other QscR homologues known to function as dimers. We showed that QscR binding to the PA1897 promoter is cooperative (Fig. 4). This indicates that binding of one monomer to the PA1897 promoter region stimulates binding of the second and thus supports

4

the conclusion that our purified QscR is in a monomeric state. We believe it is possible that QscR may in fact weakly dimerize in solution and that because of weak dimerization under the specific conditions of our analysis it is in a monomeric state. Acyl-HSL-binding requirements We have shown that 3OC12 can facilitate QscR binding to the PA1897 promoter region (Figs 1 and 4), but is this the only acyl-HSL that can function with QscR? Gel shift experiments indicate that QscR binds DNA in the presence of acyl-HSLs (at 2 µM) with acyl groups of 10–14 carbons but not with acyl groups of 4, 6 or 8 carbons (data not shown). To better assess the acyl-HSL specificities we compared the activity of QscR and LasR in recombinant

BD

AO C R

End

0.006 UV absorbance

●

log(bound/free)

2

●

log(bound/free)

2

Fig. 4. Concentration-dependent binding of His-tagged QscR to a PA1897 promoter fragment and a PA5351 promoter fragment. Left: gel shift (top) and Hill plot (bottom) for PA1897. The DNA probe was the 239 bp QscR binding site containing fragment extending from nucleotide 2068790 in the P. aeruginosa genome. Right: gel shift (top) and Hill plot (bottom) for PA5351. The PA5351 DNA probe is a 177 bp fragment extending from position 6019329– 6019505 in the P. aeruginosa genome (Stover et al., 2000). The left lane in each gel is a control with no His-tagged QscR or 3OC12. All the lanes represent reaction mixtures with 2 µM 3OC12 and the indicated concentration of Histagged QscR in pM.

0.004

0.002 0

1000

2000 Time (seconds)

3000

Fig. 5. Molecular mass of His-tagged QscR as estimated by gel filtration column chromatography. Elution profile of 40 µg purified Histagged QscR. Absorbance at 280 nm. Arrows indicate the following mass standards: A, albumin [molecular weight (MW) 67 K]; O, ovalbumin (MW 45 K); C, chymotrypsinogen A (MW 25 K); R, ribonuclease A (MW 14 K). BD indicates the void volume as indicated by elution of blue dextran (MW 2000 K), and End marks the total bed volume of the column.

© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology

Pseudomonas aeruginosa QscR activity 5 8000 A

6000

b-Galactosidase activity

4000

2000

0 0

50

100

150

200

8000 B

6000

C6 3OC6 C8 C10 3OC10 C12 3OC12 C14 C16

4000

2000

0 0

20

40

60

80

100

Acyl-HSL(nM)

Fig. 6. Activity of QscR and LasR in recombinant E. coli in response to different acyl-HSLs. A. E. coli containing the QscR expression vector pJN105Q and the PA1897–lacZ vector pJL101. B. E. coli containing the LasR expression vector pJN105L and lasI– lacZ vector pSC11. The acyl chains of each acyl-HSL are indicated on the right.

E. coli (Fig. 6). QscR can activate a PA1897–lacZ reporter at nM concentrations of C10, 3OC10, C12 or 3OC12. There is also a low level of activation with C8 and C14. In fact QscR is slightly more responsive to low concentrations of 3OC10 than it is to 3OC12 (Fig. 6A). LasR responds best to 3OC12, responds reasonably well to 3OC10 and shows some response to C14, but it shows little or no response to the other acyl-HSLs we tested (Fig. 6B). Discussion Previous work on QscR showed that it could repress the activation of selected LasR- and RhlR-dependent quorum-sensing responsive genes (Chugani et al., 2001). This could be the result of competition for signal, competition for binding sites on the regulatory DNA, or heterodimer formation (Chugani et al., 2001; Ledgham et al., 2003). Here we have tested the hypothesis that QscR functions alone to control a gene or genes in a way that does not depend on LasR or RhlR. We show that the promoter regions of two genes (PA1897 and PA5351) contain QscR binding sites (Figs 1, 2 and 4) and we show that QscR activates PA1897 in a 3OC12-dependent man-

ner (Fig. 5). We believe that the direct activity of QscR represents its main function and that other effects that might stem from signal binding competition with LasR or heterodimer formation are minor effects. We analysed the site to which QscR binds in the promoter of PA1897. QscR protects a region that includes a 20 bp inverted repeat (Fig. 2) that was previously thought to serve as a LasR binding site (Whiteley and Greenberg, 2001). Our recent in vitro studies of LasR suggested that it might not control PA1897 directly (Schuster et al., 2004b) and our current investigation supports this view. It now seems evident that QscR is the activator for PA1897. Subtle differences in the sequence of the DNA binding region can determine specificity for QscR and LasR (Fig. 3). The LasR binding site in the lasB promoter region shows identity with the QscR binding site of PA1897 in 15 of the 20 positions in the inverted repeat (Fig. 3), but QscR does not bind to the LasR binding site in lasB and LasR does not bind to the QscR binding site in PA1897. We believe PA1897 was originally misclassified as a LasRcontrolled gene because it is dependent on 3OC12. We now show that QscR also depends on 3OC12 for activity. We suspect there may be other genes in this category, genes that require 3OC12 for activation but that require QscR rather than LasR. The qscR gene codes for an orphan acyl-HSL transcription factor. Unlike lasR and rhlR, which are linked to lasI and rhlI, genes that code for production of acyl-HSL signals, there is no I gene linked to qscR. Thus, to what signal QscR responds or whether it even responds to a signal directly has not been clear. We have established that the DNA-binding activity of QscR is dependent on the presence of a long-chain acyl-HSL (Figs 1, 4 and 6). The 3OC12 signal produced by LasI and to which LasR responds is an effective ligand for QscR. For overexpression of active LasR, 3OC12 must be present during growth of the LasR-expressing bacteria, and the purified LasR is bound tightly to 3OC12 (Schuster et al., 2004b). Activity of the purified protein does not require added 3OC12. Like LasR, QscR requires 3OC12 to fold into an active conformation but unlike LasR signal binding to QscR is not irreversible. In contrast to LasR, binding of purified QscR to DNA is dependent on added acyl-HSL. That two 3OC12-responsive transcription factors differ fundamentally in their ability to exist in the absence of the signal may have important implications. This would allow for a very rapid response of the QscR regulon to sudden decreases in environmental levels of 3OC12 where the LasR regulon may respond more slowly. Like other transcriptional activators in the LuxR family that have been studied, QscR requires the presence of an acyl-HSL in the culture growth medium for folding in an active state. However, unlike LasR, which also responds to 3OC12, purified QscR requires exogenous addition of

© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology

6 J.-H. Lee, Y. Lequette and E. P. Greenberg 3OC12 for binding to target DNA. LasR binds the signal so avidly during purification that it is not necessary to add signal for DNA-binding activity (Schuster et al., 2004b). The other example of a LuxR homologue that shows in vitro activity that depends on added signal is LuxR itself. However, LuxR is unstable in the absence of signal and LuxR purification required the presence of 3OC6 in all of the solutions used during purification (Urbanowski et al., 2004). This was not true of QscR. We did not have to include any acyl-HSLs in the purification solutions. Thus, QscR seems like a good candidate for attempts to determine the structure of a LuxR homologue in the ligand-free state. There are other notable differences between LasR and QscR. One we find particularly intriguing is the fact that QscR has a broader signal specificity than does LasR, and QscR may even respond to 3OC10, C10 and C12 better than it does to 3OC12 (Fig. 6). This suggests that QscR might respond to signals produced by other bacteria that coexist with P. aeruginosa. We believe that this also points to the possibility that qscR and the genes surrounding it may be relatively recent acquisitions in the P. aeruginosa genome. The qscR gene is divergently transcribed from PA1897 and several other genes that may be involved in lipid metabolism. The phzA2–G2 cluster is on the opposite flank of qscR. The phzA2–G2 cluster is a second set of phenazine synthesis genes in the P. aeruginosa genome (Mavrodi et al., 2001). QscR is a transcription factor involved in the acyl-HSL quorum-sensing signalling pathways of P. aeruginosa. We show that in response to signals produced by the LasI protein it controls genes unique from those controlled by LasR directly. It also represses, at least transiently, some genes controlled by LasR and RhlR (Chugani et al., 2001) in what appears to be an indirect fashion. Our evidence suggests that the indirect transient repression of genes such as the hcn genes is not due to the ability of QscR to bind to LasR- or RhlR-dependent promoters but we cannot address the issue of what mechanisms might be involved in indirect repression. Rather we can conclude that QscR does control some genes via direct interaction

with promoters and it does so in a manner dependent on 3OC12. For some reason this possibility of direct action for QscR has not received serious consideration in the past. Experimental procedures Bacterial strains, plasmids and culture conditions The plasmids we used are described in Table 1. Growth of P. aeruginosa PAO-T7 (Hoang et al., 2000) and E. coli DH5α (Sambrook et al., 1989) were in Luria–Bertani (LB) broth containing 50 mM 3-(N-morpholino) propanesulphonic acid (MOPS), pH 7.0 at 37°C with vigorous shaking. Growth was monitored as optical density at 600 nm. As appropriate the following antibiotics were used (per ml): 100 µg ampicillin, 200 µg carbenicillin and 15 µg gentamicin. Synthetic acylHSLs and isopropyl-β-D-thiogalactoside (0.4 mM) were added where indicated. For construction of the His-tagged QscR expression vector pJLQhis we amplified qscR from P. aeruginosa PAO1 genomic DNA by polymerase chain reaction (PCR). The primers were 5′-AAGCTCATATGCATGATGAGAG-3′ with the underlined sequence indicating a NdeI restriction site, and 5′-AACGGGATCCGGCCATTCGG-3 with the underlined sequence indicating a BamHI restriction site. The PCR product was digested with NdeI and BamHI. The digested fragment was ligated with NdeI–BamHI-digested pET16b to form the His-tagged qscR vector pET16b-qscR. The Ori1600 and pMB1 replication origins were obtained as a NdeI–ScaIdigested fragment from pQF50. The ends were filled and this origin of replication was introduced into NruI–ScaI-digested pET16b-qscR to form pJLQhis, which expresses the Histagged QscR. This cloning procedure replaced the ColE1 replication origin with Ori1600 (broad host range) and pMB1 (E. coli ) replication origins. Plasmids were transferred to P. aeruginosa PAO-T7 by transformation with carbenicillin selection. To create the PA1897–lacZ reporter plasmid, pJL101, a 339 bp PCR fragment from −300 to +39 relative to the PA1897 translation start codon was ligated with SmaIdigested pQF50. To construct the QscR and LasR expression plasmids, pJN105Q and pJN105L, we used the broad host range expression vector, pJN105, which carries an Larabinose-inducible promoter. The qscR open reading frame (ORF) was amplified by using primers containing a XbaI site,

Table 1. Plasmids used in this study. Plasmids

Genotype

Reference

pQF50 pJL101 pSC11 pET16b pET16b-qscR pJLQhis pJN105 pJN105Q pJN105L

Broad-host-range lacZ fusion vector, Apr PA1897–lacZ reporter in pQF50, Ori1600, pMB1-containing NdeI–ScaI fragment, Apr lasI–lacZ reporter in pQF50, Apr Overexpression plasmid for His-tagged proteins, ApR QscR overexpression vector His-tagged QscR overexpression vector araC-pBAD cloned in pBBR1MCS-5, GmR qscR in pJN105, GmR lasR in pJN105, GmR

Farinha and Kropinski (1990) This work Chugani et al. (2001) Novagen This work This work Newman and Fuqua (1999) This work This work

© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology

Pseudomonas aeruginosa QscR activity 7 and the ORF was cloned into the XbaI site of pJN105. Similarly, the lasR ORF was amplified with the primers containing EcoRI and XbaI sites and cloned into EcoRI–XbaIdigested pJN105. The size, orientation and integrity of all constructs were confirmed by restriction patterns and DNA sequencing. DNA fragments containing site-directed mutations in the PA1897 and lasB OP1 promoter regions were created in two PCR steps by using the same flanking primers and internal complementary primers containing the point mutations of interest. The first reaction used an end primer and an internal primer containing the two indicated base changes (Fig. 3). The two PCR products were size-fractionated on the agarose gels and purified using QIAquick gel extraction kits (Qiagen). These PCR products were then annealed and amplified with flanking primers in a second PCR. The final PCR products were size-fractionated and purified by agarose gel electrophoresis. The mutations were confirmed by DNA sequencing.

Expression and purification of His-tagged QscR Pseudomonas aeruginosa PAO-T7 containing pJLQhis was grown in 900 ml of LB broth with carbenicillin at 37°C to optical density at 600 nm of 0.5. The culture was shifted to 16°C and IPTG was added. After another 16 h at 16°C with shaking, cells were harvested by centrifugation at 8000 g for 10 min at 4°C. The harvested cells were stored at −80°C overnight. Subsequent QscR purification procedures were performed at 0–4°C. The cells were suspended in 16 ml of binding buffer (20 mM Tris-HCl, 500 mM NaCl and 5 mM imidazole, pH 7.9), and were lysed by sonication. The lysate was centrifuged at 20 000 g for 30 min and the resulting supernatant fluid was centrifuged at 20 000 g for 30 min. The cleared supernatant fluid was fractionated by Ni-NTA agarose column chromatography (Qiagen). The bound protein was washed with a buffer containing 20 mM Tris-HCl, 500 mM NaCl and 68 mM imidazole, pH 7.9. Proteins were eluted by increasing concentrations of imidazole. Fractions containing His-tagged QscR were pooled, dialysed in 100 mM KCl, 50 mM NaCl, 2 mM EDTA, 0.5% Tween 20, 20% glycerol and 50 mM Tris-HCl (pH 7.0), and stored at −80°C. We note that none of the solutions we used for QscR purification contained acyl-HSLs.

Gel shift assays The gel shift assay was based on a protocol described by Urbanowski et al. (2004). Each gel shift reaction contained both specific and non-specific DNA probes. Specific DNA probes were prepared by PCR amplification of the region upstream of PA1897 [from nucleotides (nt) 2068690– 2069028 in the P. aeruginosa chromosome], PA5351 (nt 6019329–6019505), qscR (nt 2069190–2069490), hcnA (nt 2412246–2412545), phzA1 (nt 4713465–4713788), phzA2 (nt 2070385–2070685), PA4209 (nt 4713159–4713498), lasB (nt 4170610–4170705), and the mutant lasB and PA1897 promoter regions (regions identical to those for the wild-type fragments except for the introduced mutations). The nonspecific probe was generated by PCR amplification of the mini-CTX multiple cloning site (Becher and Schweizer, 2000).

The PCR products were end labelled using [γ-32P]-ATP and T4 nucleotide kinase. Binding reactions contained 10–30 pM of specific and non-specific DNA in 20 µl of DNA-binding buffer (50 mM KCl, 1 mM EDTA, 1 mM DTT, 0.1 mg ml−1 BSA, and 5% glycerol and 20 mM Tris, pH 7.5). Purified QscR and acyl-HSLs were added as indicated, and the reaction mixtures were incubated at room temperature for 20 min. The reaction mixtures were then separated by electrophoresis on native 5% Tris-glycine-EDTA polyacrylamide gels (1:30 bi-acrylamide to acrylamide ratio) at 10 V cm−1 at room temperature. The probes were detected and relative abundances determined by using a Typhoon model 8600 phosphorimager with ImageQuant software (Molecular Dynamics, Sunnyvale, CA) with dried gels.

DNase I protection assays DNase I footprinting was based on previously described methods (Schuster et al., 2004b; Urbanowski et al., 2004). The 339 bp PA1897 promoter fragment used for gel shift experiments has NcoI site in asymmetric position. After PCR amplification, the 339 bp fragment was labelled at both end using [γ-32P]-ATP and T4 nucleotide kinase. To generate a probe labelled at one end, the DNA fragment was digested with NcoI and the 141 bp fragment was purified by standard procedures (Sambrook et al., 1989). For DNA–protein binding, we incubated the 32P-labelled probe, 3–9 nM purified Histagged QscR and 5 µM 3OC12 in 120 µl of DNA binding buffer for 20 min at room temperature. The reaction mixtures were then treated with 6 µl of RQ1 DNase (0.24 units, Promega) for 5 min, after which 25 µl of footprint stop solution (3 M ammonium acetate, 0.25 M EDTA and 15 µg ml−1 sheared calf thymus DNA) was added. The DNA was precipitated with 370 µl of ethanol. DNase I digestion products were separated on 7% denaturing polyacrylamide gels and were visualized with X-ray film.

Gel filtration analysis of purified QscR Purified QscR (40 µg in 0.2 ml) was eluted on a Superose 12 10/300 GL column (Amersham Pharmacia) at 4°C using a Bio-Rad DuoFlow fast performance liquid chromatography apparatus. Gel filtration was in 20 mM Tris-HCl, pH 7.5, 50 mM KCl, 1 mM EDTA, 1 mM DTT and 5% glycerol with or without 5 µM 3OC12 at a flow rate of 0.4 ml min−1. The column was calibrated with protein standards from a lowmolecular-weight gel filtration kit (Sigma).

Measurement of lacZ transcription in recombinant E. coli To monitor lacZ transcription we measured β-galactosidase activity by using a Galacto-Light PlusTM kit (Tropix) as described elsewhere (Whiteley et al., 1999). Results are given in units of β-galactosidase activity per optical density at 600 nm.

Acknowledgements We thank Herbert P. Schweizer for generously providing P. aeruginosa PAO-T7 and Amy L. Schaefer for helpful sug-

© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology

8 J.-H. Lee, Y. Lequette and E. P. Greenberg gestions. This study was supported by USPHS Grant GM59026 and by a grant from the W. M. Keck Foundation.

References Becher, A., and Schweizer, H.P. (2000) Integration-proficient Pseudomonas aeruginosa vectors for isolation of singlecopy chromosomal lacZ and lux gene fusions. Biotechniques 29: 948–952. Chugani, S.A., Whiteley, M., Lee, K.M., D’Argenio, D., Manoil, C., and Greenberg, E.P. (2001) QscR, a modulator of quorum-sensing signal synthesis and virulence in Pseudomonas aeruginosa. Proc Natl Acad Sci USA 98: 2752–2757. Farinha, M.A., and Kropinski, A.M. (1990) Construction of broad-host-range plasmid vectors for easy visible selection and analysis of promoters. J Bacteriol 172: 3496–3499. Fuqua, C., and Greenberg, E.P. (2002) Listening in on bacteria: acyl-homoserine lactone signalling. Nat Rev Mol Cell Biol 3: 685–695. Fuqua, C., Parsek, M.R., and Greenberg, E.P. (2001) Regulation of gene expression by cell-to-cell communication: acyl-homoserine lactone quorum sensing. Annu Rev Genet 35: 439–468. Hoang, T.T., Kutchma, A.J., Becher, A., and Schweizer, H.P. (2000) Integration-proficient plasmids for Pseudomonas aeruginosa: site-specific integration and use for engineering of reporter and expression strains. Plasmid 43: 59–72. Ledgham, F., Ventre, I., Soscia, C., Foglino, M., Sturgis, J.N., and Lazdunski, A. (2003) Interactions of the quorum sensing regulator QscR: interaction with itself and the other regulators of Pseudomonas aeruginosa LasR and RhlR. Mol Microbiol 48: 199–210. Mavrodi, D.V., Bonsall, R.F., Delaney, S.M., Soule, M.J., Phillips, G., and Thomashow, L.S. (2001) Functional analysis of genes for biosynthesis of pyocyanin and phenazine1-carboxamide from Pseudomonas aeruginosa PAO1. J Bacteriol 183: 6454–6465. Newman, J.R., and Fuqua, C. (1999) Broad-host-range expression vectors that carry the 1-arabinose-inducible Escherichia coli araBAD promoter and the araC regulator. Gene 227: 197–203. Sambrook, J.E., Fritsch, F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Schuster, M., Lostroh, C.P., Ogi, T., and Greenberg, E.P. (2003) Identification, timing, and signal specificity of

Pseudomonas aeruginosa quorum-controlled genes: a transcriptome analysis. J Bacteriol 185: 2066–2079. Schuster, M., Hawkins, A.C., Harwood, C.S., and Greenberg, E.P. (2004a) The Pseudomonas aeruginosa RpoS regulon and its relationship to quorum sensing. Mol Microbiol 51: 973–985. Schuster, M., Urbanowski, M.L., and Greenberg, E.P. (2004b) Promoter specificity in Pseudomonas aeruginosa quorum sensing revealed by DNA binding of purified LasR. Proc Natl Acad Sci USA 101: 15833–15839. Smith, R.S., and Iglewski, B.H. (2003) Pseudomonas aeruginosa quorum sensing as a potential antimicrobial target. J Clin Invest 112: 1460–1465. Stover, C.K., Pham, X.Q., Erwin, A.L., Mizoguchi, S.D., Warrener, P., Hickey, M.J., et al. (2000) Complete genome sequence of Pseudomonas aeruginosa PA01, an opportunistic pathogen. Nature 406: 959–964. Urbanowski, M.L., Lostroh, C.P., and Greenberg, E.P. (2004) Reversible acyl-homoserine lactone binding to purified Vibrio fischeri LuxR protein. J Bacteriol 186: 631–637. Wagner, V.E., Bushnell, D., Passador, L., Brooks, A.I., and Iglewski, B.H. (2003) Microarray analysis of Pseudomonas aeruginosa quorum-sensing regulons: effects of growth phase and environment. J Bacteriol 185: 2080–2095. Welch, M., Todd, D.E., Whitehead, N.A., McGowan, S.J., Bycroft, B.W., and Salmond, G.P. (2000) N-acyl homoserine lactone binding to the CarR receptor determines quorum-sensing specificity in Erwinia. EMBO J 19: 631–641. Whiteley, M., and Greenberg, E.P. (2001) Promoter specificity elements in Pseudomonas aeruginosa quorumsensing-controlled genes. J Bacteriol 183: 5529–5534. Whiteley, M., Lee, K.M., and Greenberg, E.P. (1999) Identification of genes controlled by quorum sensing in Pseudomonas aeruginosa. Proc Natl Acad Sci USA 96: 13904–13909. Wu, H., Song, Z., Hentzer, M., Andersen, J.B., Molin, S., Givskov, M., and Hoiby, N. (2004) Synthetic furanones inhibit quorum-sensing and enhance bacterial clearance in Pseudomonas aeruginosa lung infection in mice. J Antimicrob Chemother 53: 1054–1061. Zhu, J., and Winans, S.C. (1999) Autoinducer binding by the quorum-sensing regulator TraR increases affinity for target promoters in vitro and decreases TraR turnover rates in whole cells. Proc Natl Acad Sci USA 96: 4832–4837. Zhu, J., and Winans, S.C. (2001) The quorum-sensing transcriptional regulator TraR requires its cognate signaling ligand for protein folding, protease resistance, and dimerization. Proc Natl Acad Sci USA 98: 1507–1512.

© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology