Integration host factor is essential for the optimal expression of the styABCD operon in Pseudomonas...

Research in Microbiology 153 (2002) 527–536www.elsevier.com/locate/resmic

Integration host factor is essential for the optimal expressionof thestyABCD operon inPseudomonas fluorescens ST

Pedro Miguel Santos1, Livia Leoni, Ilaria Di Bartolo, Elisabetta Zennaro∗

Department of Biology, Third University of Rome, viale Marconi 446, 00146 Rome, Italy

Received 17 February 2002; accepted 24 April 2002

First published online 2 July 2002

Abstract

The StyS/StyR two-component regulatory system ofPseudomonas fluorescens ST controls the expression of thestyABCD operon codingfor the styrene degradation upper pathway. In a previous work we showed that the promoter of the catabolic operon (PstyA) is induced bystyrene and repressed to differing extents by organic acids or carbohydrates. In order to study the mechanisms controlling the expression ofthis operon, we performed a functional analysis on 5′ deletions ofPstyA by the use of a promoter–probe system. These studies demonstratedthat a palindromic region (sty box), located from nucleotides−52 to−37 with respect to the transcriptional start point is essential forPstyAactivity. Moreover, additional regulatory regions involved in the modulation ofPstyA activity were found along the promoter sequence. Inparticular, deletion of a putative StyR binding site, homologous to the 3′ half of thesty box and located upstream of this box, resulted in65% reduction of the induction level of the reporter gene. Additionally, we performed bandshift assays with a DNA probe corresponding toPstyA and protein crude extracts fromP. fluorescens ST, using specific DNA fragments as competitors. In these experiments we demonstratedthat IHF binds an AT-rich region located upstream of thesty box. On the basis of this finding, coupled with the results obtained withPstyAfunctional analysis, we suggest that the role of the IHF-mediated DNA bend is to bring closer, in an overlapping position, the upstream StyRputative binding site and the downstreamsty box, and that the formed complex enhances transcription. 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved.

Keywords: Integration host factor; Promoter; Two-component system; Bandshift; Styrene;Pseudomonas

1. Introduction

In Pseudomonas strains, styrene degradation has beenextensively studied both at the physiological and molecu-lar level. Interestingly, in all the studied strains, the styrenegene cluster is highly homologous and organized in a simi-lar way suggesting that all these strains received this genetictrait by horizontal transfer from a common ancestral bac-terium [3,31,32,37–41,50,58]. Up to now little attention hasbeen given to the regulatory mechanisms that govern the ex-pression of this cluster. The catabolic operonstyABCD codesfor the oxidation of the styrene lateral chain leading to theconversion of styrene into phenylacetic acid. The regulatory-associated operonstySR is located immediately upstream of

* Correspondence and reprints.E-mail address: [email protected] (E. Zennaro).

1 Present address: Centre for Biological and Chemical Engineering,Instituto Superior Técnico, 1049-001 Lisbon, Portugal.

the catabolic one and encodes a sensor kinase (StyS) and aregulatory DNA binding protein (StyR), which is essentialfor the expression of thestyABCD operon inE. coli [31,41,58]. Two-component regulatory systems for genes involvedin the degradation of other aromatic hydrocarbons haveonly been reported for toluene catabolism inPseudomonasputida F1 and DOT-T1 strains [30,35] and inThauera strainT1 [14] and for the degradation of biphenyls inRhodococcusM5 [29]. Interestingly, in the toluene and styrene-degradingstrains, the involved sensor kinase/response regulator pairsshare a high similarity both in gene organization and in pro-tein structures. In particular the sensor proteins are hybridkinases characterized by a leucine zipper dimerization motifat the N-terminus and a PAS domain, which is characteris-tic of proteins associated with oxygen sensing, light recep-tion, eukaryotic transcription regulation and clock proteins.Due to the different role of these proteins, a common fold insensors responsive to redox or related stimuli has been sug-gested [30,32,59].

0923-2508/02/$ – see front matter 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved.PII: S0923-2508(02 )01358-X

528 P.M. Santos et al. / Research in Microbiology 153 (2002) 527–536

A palindromic sequence (namedsty box), that could con-stitute the putative binding site of StyR, is present in thepromoters of all the studied styrene catabolic operons. Anidentical palindromic sequence was also found in the pro-moter of toluene catabolic operon inP. putida F1 (PtodX).In this last case, it was demonstrated that this box is the bind-ing site of TodT, a positive regulator that shares 49% iden-tity with StyR [30]. Interestingly,PtodX activity is inducedin response to a number of aromatic compounds includingstyrene [35]. Moreover, thesty box seems to be required forthe StyR-mediated activation of thestyABCD promoter inE. coli [41]. However, the molecular mechanism by whichthese promoters are regulated is not known.

Previous studies from our laboratory showed that inPseudomonas fluorescens ST, the promoter of thestyABCDoperon (PstyA) is induced by styrene and repressed todifferent extents by organic acids or carbohydrates [50].Similar results have been reported for the styrene-degradingstrainP. putida CA-3 [37,40]. The complexity of the hybrid

histidine kinase StyS and the different levels ofPstyAactivity under different growth conditions indicate that theexpression of styrene catabolic genes is finely regulated.

In this paper we report the structural and functionalanalysis ofPstyA. We also demonstrate that the integrationhost factor (IHF) protein is able to bind toPstyA and that itplays an important role in promoting transcription from thispromoter.

2. Materials and methods

2.1. Bacterial strains, plasmids, media and chemicals

The bacterial strains and plasmids used in this study arelisted in Table 1.Pseudomonas and Escherichia coli cellswere routinely grown at 30 and 37◦C, respectively, in Luria-Bertani (LB) medium [34] or mineral salts medium [25] with0.1% glucose. In induction studies styrene was added via the

Table 1Bacterial strains and plasmids

Strains and plasmids Relevant genotype Reference or source

StrainsP. fluorescens ST Sty+ [1]P. putida KT2442 hsdMR, Rifr [22]P. putida A8759 KT2442ihfA::Kmr Tn4652 xylRS Pu-lacZ Rifr [8]P. putida RT31 A8759P. putida ihfA and ihfB under the control ofPtac promoter and

lacI q repressor; Rifr, Smr, Tcr[55]

E. coli DH5α endA1 hsdR17 supE44 thi-1 recA1 gyrA96 (Nalr) relA1 �(lacIZYA-argF)U169 deoR(�80dlacZ�M15)

Bethesda Res. Laboratories

PlasmidspBluescript II KS+ Cloning vector; Apr; 2.9 kb StratagenepRK2013 Helper plasmid; ColE1 replicon, Kmr Mob+ Tra+ [21]p907 pLAFR3 derivative with a 27-kb fragment fromP. fluorescent ST

chromosome carrying genes for styrene degradation[31]

pTE50 pTZ19R derivative containing a chromosomal fragment ofP. fluo-rescens ST carrying the genes coding forstyR, styA andstyB; Apr

[31]

pBSPa probe pBluescript II KS+ derivative in which a 201-bp PCR fragment(digested withEcoRI) containingPstyA was cloned into theEcoRI siteof the vector; Apr

This study

pJB137 Expression vector containing the entire promoter region (Pu) of thexylCMABN operon from the TOL plasmid, RK2 replicon; Apr

[5]

pPR9TT lacZ promoter probe vector; Apr; Cmr [51]pPR9Pa0 413-bp PCR-derived fragment, encompassing the entire intergenic

region betweenstySR andstyABCD operons and the first 240 nt ofstyAligated to theXhoI-BamHI sites of pPR9TT

This study

pPR9Pa1 Derivative of pPR9Pa0 carring a 5′ deletion from nucleotides−143 to−118 with respect to thestyA trascription start point

This study

pPR9Pa2 Derivative of pPR9Pa0 carrying a 5′ deletion from nucleotides−143 to−103 with respect to thestyA trascription start point

This study


This study


This study


This study


This study

P.M. Santos et al. / Research in Microbiology 153 (2002) 527–536 529

gas phase as previously described [31]. When necessary, cul-tures were supplemented with ampicillin (Ap, 100µg/ml),tetracycline (Tc, 20µg/ml), kanamycin (Km, 50µg/ml)or chloramphenicol (Cm, 30 to 200µg/ml). Isopropyl-β-D-thiogalactopyranoside (IPTG; 1 mM), 5′-bromo-4′-chloro-3′-indolyl-β-D-galactopyranoside (X-Gal, 1 mM),and 2-nitrophenyl-β-D-galactopyranoside (ONPG, 1 mM)were added to the media when appropriate.

2.2. DNA manipulation and genetic techniques

Plasmid DNA was prepared by the alkaline lysis proto-col [49] or with the QIAGEN Midi-isolation kit (Qiagen).DNA fragments were purified from agarose using theQiaquick gel extraction kit or the QIAEXII kit (Qiagen).Transformations ofE. coli, restrictions and ligations werecarried out by standard procedures [49]. All PCR amplifi-cations were performed by standard procedures usingPfupolymerase from Stratagene. Automated sequences wereperformed by MWG biotech sequence services.

For 5′ deletion analysis of thestyA promoter (PstyA),seven forward oligonucleotides were synthesized, the se-quences of which were chosen to be conducive to primingof the coding strand from positions−174, −148, −132,−110,−91,−83, and−72 relative to the ATG start codon ofthestyA gene (primers are named from Pa0/FW to Pa6/FW,Table 2). A reverse primer, named Pa/REV (Table 2) andannealing to the coding strand from nt+240 relative tothe styA ATG, was used for PCRs in combination witheach of the forward primers. The PCR products were firstblunt-cloned inHincII-digested pBluescriptII KS+ (Strata-gene) and checked by sequencing. Afterwards, the fragmentscloned in the right orientation were excised byXhoI-BamHIdigestion and ligated to compatible sites of the promoterprobe vector pPR9TT [51] giving rise to plasmids pPR9Pa0to pPR9Pa6 reported in Table 1. These plasmids were trans-ferred fromE. coli to P. fluorescens ST by triparental mat-ings with the helper plasmid pRK2013 [21].

To originate plasmid pBSPa probe, a 201-bp fragmentwas originated by PCR using primers PaprobeFW and Pa-probeREV (Table 2) and cloned byEcoRI into pBluescriptIIKS+.

2.3. β-Galactosidase and styrene monooxygenase assays

In order to measureβ-galactosidase activity,P. fluo-rescens ST cells harboring pPR9TT–derived plasmids, weregrown for 12 h at 30◦C in mineral salts medium contain-ing Cm 100µg/ml and supplemented with styrene or glu-cose as carbon sources; cells were then diluted 1:100 in thesame medium and subcultured until cultures reached a tur-bidity at 600 nm (OD600) of approximately 0.6. After col-lection, cells were lysed with toluene, andβ-galactosidaseactivity was measured as described by Miller, and expressedas Miller units [34]. The presented data correspond to the re-sults obtained from, at least, five independent experimentswith a standard deviation not higher than 10%.

To quantify styrene monooxygenase activity, produc-tion of indigo was assayed as previously described [38].Cells were grown in mineral medium containing styrene assole carbon source and harvested in the exponential phase(OD600= 0.6) by centrifugation, washed with 50 mM potas-sium phosphate buffer (pH 7), and resuspended in the samebuffer to an OD600 of 3.0. 100µl of concentrated cellswere added to 400µl of 50 mM phosphate buffer con-taining 0.25 mM indole. Samples were incubated at 30◦Cwith vigorous shaking for 30 min. After centrifugation, thecell pellets were resuspended in 1 ml of dimethylformamideand extracted by shaking for 15 min. The tubes were thencentrifuged to remove the cell debris. Accumulation of theblue dye indigo was followed by measuring the OD600 ofthe supernatants. Indigo concentration was calculated us-ing the molar extinction coefficient for indigo determinedby O’Connor et al. [38].

Table 2Oligonucleotides used in this study

Name Sequence (5′-3′) Starting Template Applicationpositiona

PaprobeFW CGGAATTCCGTTGACTGCTTCGGG −179 pTE50 PstyA probePaprobeREV CGGAATTCCAACAATACCGATACG +23 [31]Pa0/FW GCTCTAGACTGCTTCGGGCTCTGC −174 5′ PstyAPa1/FW GCTCTAGAGGTGTAGTAAATATAAGT −148 deletionsPa2/FW GCTCTAGAGTTTATGATTTTTAATATATA −132Pa3/FW GCTCTAGACTATTTTTAAGGATATTTTT −110Pa4/FW GCTCTAGATATACCGCATAAACCAC −91Pa5/FW GCTCTAGAATAAACCACGGTTTATTT −83Pa6/FW GCTCTAGATTTATTTCCTTTTTTGCT −72Pa/REV GGGGTACCTACGTAGTAGTAGTGG +240PuFW GAATTCATTTTAAGTCTCGTATAGCTAGC −22 pJB137 Pu coldPuREV GAATTCATCAAATCGACAGGTGGTTATGC −203 [5] competitor

aDistance in bp from thestyA or xylC ATG.


2.4. Gel mobility shift assay

Cells grown under different conditions were harvested bycentrifugation (7000g for 20 min. at 4◦C), resuspended insonication buffer (50 mM Tris-HCl pH 7.5, 100 mM KCl,1 mM EDTA, 10% glycerol), and disrupted by sonication(6 cycles of 15 s burst with 15 s interval) in ice. Aftercentrifugation (17000g at 0◦C) protein crude extracts werequantified and normalized (using sonication buffer) to a totalprotein concentration of 10 mg/ml (reconfirmed by SDS-PAGE), aliquoted, and stored at−80◦C until their use. Inorder to generate the DNA probe, a 201-bp DNA fragmentcontainingPstyA was obtained byEcoRI digestion of thepBSPaprobe. This fragment was labeled with [α-32P]dATPby fill-in with Klenow enzyme. The radiolabeled DNAfragment was purified by gel filtration with a sepharose G50spun column followed by phenol:chloroform extraction andethanol precipitation. ThePstyA-derived DNA fragmentsused as cold competitors in mobility shift experiments(100 ng per sample) corresponded to the same PCR productsused to generate plasmids pPR9Pa0, pPR9Pa2, pPR9Pa3,pPR9Pa4 and pPR9Pa6. A 181-bp DNA fragment containingthe integration host factor (IHF) binding site upstream of thePu promoter ofxyl genes in the TOL plasmid was originatedby PCR using primersPuFW andPuREV (Table 2). ThisPCR product was used as IHF-specific cold competitor.

The binding reactions were carried out in a volume of20 µl. DNA probes (0.5 ng, 2000 cpm) were incubated at4◦C for 30 min with 20µg of protein crude extracts in bind-ing buffer (10 mM Tris-HCl pH 8.0, 50 mM NaCl, 2.5 mMMgCl2, 0.025% NP-40, 125µg/ml sonicated salmon spermDNA, 5% glycerol). When non-labeled DNA competitorswere used, protein crude extracts were added last to the bind-ing mixture. After incubation, the reaction mixtures wereloaded (10µl) onto a 30-min pre-run 6% continuous non-denaturing polyacrylamide gel. The electophoresis was car-ried out at 4◦C in 0.5× Tris-borate-EDTA buffer (pH 8.0) at10 V/cm for 2.5 h. Gels were then dried and exposed.

2.5. Curvature predictions

Computer predictions of DNA curvature were obtainedwith the aid of a refined version of the WEDGE pro-gram [13].

3. Results

3.1. Deletion analysis of the styA promoter

The sequence reported in Fig. 1B corresponds to thestyR–styA intergenic region and begins just downstream ofthe stop codon ofstyR and ends at the start codon ofstyA,the first gene of thestyABCD catabolic operon (see Fig. 1A).

In Fig. 1C the promoter region of thetodX gene fromP. putida F1 is shown. This promoter is under the control

of a two-component system, whose members, TodS–TodT,share a high similarity with the StyS–StyR pair [30]. TheTodT-binding site (tod box) has been confirmed by foot-printing analysis and is identical, except for one nucleotide,to thesty box that is located from nucleotides−52 to −37of the sequence in Fig. 1B. Interestingly, sequences similarto each half-palindrome are located upstream (gray-shadednucleotides−102 to −97) and downstream (black-shadednucleotides+25 to +30) of the sty box. The same situa-tion is also present in the corresponding promoters of theother studied styrene-degrading strains (accession numbers,AJ000330 and AF031161). Additional motifs that are sim-ilar to each half of thetod box are also present in thetodXpromoter. It is tempting to speculate that these motifs canbe alternative binding sites for the corresponding regulators.Other important features ofPstyA are the presence of a directrepeat (5′-CTGCTTC-3′) at the 5′ end and of an AT rich re-gion spanning approximately from nucleotides−114 to−57(Fig. 1B). This region contains three 5′-ATTTTTA-3 ′ repeats(Fig. 1, dotted underlines) which are present in several bac-terial promoters which are bound by IHF [11,23,26,27,36,53,56,57]. A putative IHF binding site is also present down-stream of thetod box in todX promoter (Fig. 1C). Moreover,a canonical−35 was not found, but only an “extended”−10,as often happens when positive regulators attract RNA poly-merase by binding near or overlapping the−35 σ70 factor-binding site [2,12]. The common structural features of TodTand StyR-dependent promoters might play a functional rolein transcription regulation.

With the aim of assigning a functional role to the putativeregulatory elements found inPstyA, a 5′ deletion analysiswas performed. The 5′ deletedPstyA fragments were clonedinto the pPR9TT promoter probe plasmid (Table 1). Theobtained constructs, indicated in Fig. 1B as pPR9Pa0-Pa6,were transferred intoP. fluorescens ST andβ-galactosidaseactivities were measured in cells grown on styrene or glucoseas sole carbon sources. Results are showed in Fig. 2.

The pPR9Pa0 plasmid, containing the entirestyR–styAintergenic region, showed a strong styrene-induced pro-moter activity. However, deletion of the direct repeat5′-CTGCTTC-3′ located at the 5′ end of PstyA (constructpPR9Pa1) led to an enhancement of styrene-dependent pro-moter activity of about 1.5-fold in comparison to pPR9Pa0.Thus, this motif might have a negative role in modulatingPstyA. Plasmids pPR9Pa3, pPR9Pa4 and pPR9Pa5 showedcomparable levels ofβ-galactosidase activity, revealing a2-fold reduction of styrene-dependent promoter activity anda 3-fold enhancement of the basal activity in glucose with re-spect to pPR9Pa1. Thereforecis-acting regulative elementsinvolved in modulation ofPstyA are located in the DNAregion spanning from nucleotides−102 to−80 of the se-quence reported in Fig. 1B. This region ofPstyA is character-ized by the presence of the 5′-GTTTAT-3′ motif, a putativeStyR binding site, and of one of the three 5′-ATTTTTA-3 ′motifs.


Fig. 1. (A) Regulatory and catabolic operons for the styrene degradation system inP. fluorescens ST. styS, sensor;styR, regulator; styAB, styrenemonooxygenase;styC, epoxystyrene isomerase;styD, phenylacetaldehyde dehydrogenase,paak, phenylacetyl-coenzimeA ligase.PstySR, promoter ofstySR;PstyA, promoter ofstyABCD; I1, putative styrene-sensing domain; I2, putative redox potential-sensing domain; HK1 and HK2, histidine kinase domains; R1and R2, receiver domains; B DNA binding domain [50,58]. (B) Sequence analysis of thestyA promoter region. The reported sequence starts just downstream ofthestyR stop codon. Numbering is relative to thestyABCD transcription start point (in boldface and marked by an asterisk [58]). ThestyA ATG start codon is inboldface and underlined. The ribosome binding site (RBS) and the extended−10 region are underlined (double and single underlines, respectively). The StyRputative binding site, namedsty box, is signed by white arrows. Each esameric sequence of this inverted repeat is boxed in black and gray, respectively. Thedirect repeat (5′-CTGCTTC-3′) at the early 5′ of the promoter region is boxed, and the direct repeats (5′-ATTTTTA-3′) located just upstream of thesty box areunderlined with dotted lines. pPR9Pa0 to pPR9Pa6 correspond to 5′ deletions ofPstyA used to investigate the role of each of the putative regulatory elements.(C) todX promoter region [30]. Numbering is relative to thetodX transcription start point (in boldface and marked by an asterisk). The ATG start codon isin boldface and underlined. The ribosome-binding site (RBS) and the extended−10 region are underlined (double and single underlines, respectively). TheTodT binding site (tod box ) is indicated by white arrows, and each half of this palindrome is marked with black and gray boxes, respectively. A sequenceresembling the consensus for IHF (WATCAANNNNTTR, [24]) is underlined with dotted lines.

Promoter activity of plasmid pPR9Pa6, carrying a dele-tion of a further 11 bp with respect to pPR9Pa5, was totallysilenced. This deletion eliminates part of thesty box, demon-strating that its integrity is essential forPstyA activity.

3.2. Interactions of P. fluorescensST crude extractswith PstyA

The structural and functional complexity ofPstyA sug-gests that multiple regulatory factors might be involved intranscriptional control of this promoter. Indeed, bandshift

experiments using a probe corresponding toPstyA incu-bated with protein crude extract from styrene-inducedP. flu-orescens ST cells showed a complex pattern of retardationwith three main bandsa, b, c (Fig. 3, lane 2). In order toassess which of the DNA motifs localized onPstyA were in-volved in formation of the different complexes, competitionexperiments with DNA fragments corresponding to 5′ dele-tions ofPstyA were performed. The DNA fragments used asspecific cold DNA competitors were the same PCR productsused to generate plasmids pPR9Pa0, pPR9Pa2, pPR9Pa3,pPR9Pa4 and pPR9Pa6 and here are referred to as Pa0, Pa2,


Fig. 2. Effects of 5′ deletions onstyA promoter activity.β-galactosidase activities (reported in Miller units [34]) were measured inP. fluorescens ST strainharboring the indicatedPstyA::lacZ fusions grown to OD600 ∼= 0.6 on styrene (A) or glucose (B).P. fluorescens ST carrying the control plasmid pPR9TTdisplayedβ-galactosidase activity<4 Miller units in all tested conditions. The reported data represent the average of at least five independent experiments.Vertical lines indicate standard deviations.

Fig. 3. Gel retardation assays of in vitro binding of thestyA promoterwith protein crude extracts fromP. fluorescens ST cells grown on styrene.The three complexes formed (a, b, c) and the probe are indicated byarrows. 0.5 ng of [α-32P]dATP-labeledPstyA probe were used for eachreaction. Lane 1,PstyA probe; lane 2,PstyA probe incubated with 20µgof P. fluorescens ST crude extracts; lanes 3 to 7,PstyA probe incubatedwith 20 µg of P. fluorescens ST crude extracts and with the indicatedDNA-competitor fragments (100 ng per sample) corresponding to different5′ deletions ofPstyA (see Fig. 1). Reaction mixtures were prepared asdescribed in Materials and methods.

Pa3, Pa4 and Pa6, respectively. If a DNA binding factor in-teracts within a DNA region encompassed by the specificcompetitor, this factor is sequestered by the competitor andis no longer available for binding to the full-length promoter.

As expected, competition with Pa0 DNA fragment, whichcorresponds to the entire promoter region, resulted in theabsence of a retardation pattern. The same result wasobtained in the competition with Pa2 (Fig. 3, lanes 3 and 4)even if a very faint band, corresponding to complexa

could be present. Two bands corresponding to complexesa and c appeared when Pa3 was the competitor (Fig. 3,

lane 5). As stated above, the DNA region between Pa2 andPa3 contains a putative StyR binding site and one of thethree 5′-ATTTTTA-3 ′ motifs and is required for fullPstyAactivity.

Competition with Pa4 resulted in the localization of theDNA region where the DNA binding factor responsiblefor the formation of complexb interacts (Fig. 3, lane 6).Finally, competition with Pa6 resulted in a pattern verysimilar to that obtained with Pa4 (Fig. 3, lane 7). Inthese two last patterns, all the bands present when nocompetitor was added, are represented, though with a loweramount of complexesa andc. This result was unexpectedsince the competitor Pa4 contains thesty box and onedownstream putative StyR binding site, while in Pa6 thesty box is in large part deleted. The band-shift patternobtained with Pa4 and Pa6 competitors can be explainedassuming that our experimental conditions did not permita stable binding of StyR to the promoter or that in theused cell extracts StyR was in an inactive form. In thiscase the retarded bands may be due totrans-acting elementsother than StyR. Alternatively, it is possible that StyRbinding to the different sites is cooperative and requiresspecific promoter architecture, only achieved when the entirepromoter sequence is present. In this case, competition withprogressively shorter 5′-deleted promoter fragments wouldbe less efficient or prevented. The fact that the intensity ofthe banda increases with the decreasing of the length of thecompetitor is in agreement with this hypothesis.

3.3. Interaction of IHF with PstyA

The DNA region between Pa3 and Pa4 is the target sitefor a factor that triggers the formation of complexb. Thisregion contains two 5′-ATTTTTA-3 ′ motifs which might beinvolved in binding of IHF. Moreover, computer predictions,based on published estimates of dinucleotide wedge angles,suggest the presence of an intrinsic DNA curvature locatedin this region ofPstyA (between nucleotides−73 and−63


Fig. 4. Gel retardation assays demonstrating specific binding of IHF tothe styA promoter. The arrows indicate the probe and the three complexesformed (a, b, c) with P. fluorescens ST crude extract proteins. 0.5 ng of[α-32P]dATP-labeledPstyA probe were incubated with 20µg of crudeextract proteins from the indicatedPseudomonas strains. Additions to thebinding buffer for each specific binding reaction are reported above eachlane. Where indicated, 100 ng of DNA competitor were added: IHF-specificDNA competitor corresponds to a PCR-originated fragment encompassingthe Pu promoter region (lanes 3 and 5); Pa3 and Pa4 competitor DNAfragments correspond toPstyA 5′-deletions reported in Fig. 1 (lanes 9and 10, respectively). ST was grown on styrene while the other strains weregrown on LB medium added with 1mM IPTG where indicated (lanes 8to 10).1. KT2442ihfA− derivative.2. A8759 derivative withihfA andihfB under the control ofPtac promoter.

of the sequence of Fig. 1). Binding of IHF to intrinsicallycurved DNAs also not matching the consensus has beenreported [52]. Hence, using a specific DNA competitorcontaining a demonstrated IHF binding region within thePu promoter region [18] as well as protein crude extractsfrom aP. putida ihfA mutant [8,55], the possible IHF–PstyAinteraction was investigated.

In Fig. 4 the results of these experiments are reported.Specific competition with thePu DNA fragment resulted inthe absence of complexb in bothP. fluorescens ST (lane 3)andP. putida KT2442 (lane 5). Moreover, when the reactionmixture contained protein crude extracts from theP. putidaihfA mutant A8759 (lane 6), no retardation band correspond-ing to complexb was observed. Accordingly, the IPTG-controlled expression of IHF inP. putida RT31 (lanes 7and 8) revealed that IHF expression led to the formationof a retardation band correspondent to complexb. These re-sults confirm the interaction of IHF with thePstyA promoterregion. Interestingly, the KT2442 strain showed a retarda-tion pattern similar to that of the ST strain. It is possiblethat nucleoid proteins or heterologous regulators can bind toPstyA promoter sequences in the absence of specific regu-lators properly bound at specific sites within the promoter,

giving rise to bands of comparable mobility due to the sim-ilar molecular weight of many different regulators [43]. In-deed, longer gel electrophoresis showed that complexesa

andc obtained with crude extracts from KT2442 run fasterthan that from the ST strain. Overexpression of IHF in RT31cell extract (Fig. 4, lane 8) probably prevents the accessionof heterologous DNA binding factors to thePstyA promoter,since only the complex corresponding to IHF binding couldbe observed with these extracts.

In order to investigate the in vivo involvement of IHF inPstyA activation, plasmid p907 was introduced inP. putidaKT2442 and in itsihfA− derivative A8759. Both strainsacquired the ability to grow on styrene as sole carbonsource since plasmid p907 contains a 27-kb fragment fromP. fluorescent ST chromosome encompassing thestySR andstyABCD operons [31]. The first two genes of thestyABCDoperon (styAB) encode the styrene monooxygenase (SMO)enzyme which converts styrene to styrene oxide and alsoindole to indoxyl, which spontaneusly dimerizes to indigo.Therefore SMO activity can be determined by monitoringthe accumulation of the blue dye indigo from indole [38].SMO activities were measured in cells grown to OD600 of0.6 in mineral medium containing styrene as sole carbonsource. KT2442(p907) cells produced 3.2 nmoles indigomin−1 mg dry wt−1, approximately two-fold more thanA8759(p907) which produced 1.5 nmoles indigo min−1 mgdry wt−1, demonstrating that IHF is functionally involved inPstyA regulation.

Putting together the results obtained withPstyA func-tional analysis and band-shift experiments, some consider-ations can be made on thecis- and trans-acting elementsidentified. There is approximately a 40% reduction in thepromoter activity, corresponding to about 65% reduction inthe induction level with Pa3 or Pa4 deletions. To reduction inthe induction level also contributes a higher level of basal ex-pression, indicating that a less stringent control of promotersilencing in non-induced conditions occurred. The promoterregion between Pa2 and Pa3 contains a putative StyR bind-ing site and that between Pa3 and Pa4 constitutes the targetsite for IHF binding, as demonstrated by band-shift experi-ments. Since the effect of IHF-mediated bent is to bring dis-tal regions closer [42], it is not surprising that similar levelsof β-galactosidase expression were found with Pa3 and Pa4deletions. In fact, IHF binds to a DNA region close to thebeginning of Pa3 deletion and its positive effect can only beobserved when upstream sequences are present, as in Pa2 fu-sion. Whether the positive effect exerted by the DNA regionbetween Pa2 and Pa3 is due to the presence of a putativeStyR binding site or whether it simply has a structural func-tion remains to be established. However, we have observedthat, upon IHF binding, the upstream StyR putative bindingsite and thesty box would be brought closer in an overlap-ping position.


4. Discussion

This paper represents the first structural and functionalanalysis of the promoter of thestyABCD operon coding forstyrene degradation inP. fluorescens ST. Our data indicatethat this promoter is finely regulated and we have demon-strated that IHF plays an important role in its modulation.The promoter region contains multiple putative StyR bind-ing sites. The fact that the same situation is maintained in thecorresponding promoters of other styrene-degrading strainsand in thetodX promoter, where the TodT–todX binding hasbeen demonstrated, indicates that these sequences are im-portant for the modulation ofPstyA activity. Often, regula-tors bind multiple sites, and the different binding positionscan affect promoter activity [12,15,47]. Regulativecis ele-ments located downstream of the transcription origin havebeen found in many promoters and it was demonstrated thatthey act as repressive elements [48].

Experiments performed with several 5′ PstyA deletionsfused with thelacZ reporter gene clearly showed that thestybox is essential for promoter activity. Most probably, also theupstream StyR binding site has an important role inPstyAactivation. Indeed, deletion of a fragment containing thissequence led to 45% reduction inβ-galactosidase activity,corresponding to a 65% reduction in the inducible level. Thesame effect was also observed when the downstream IHFbinding site was deleted. This suggests that the role of IHF isto bring closer the StyR putative binding site and thesty boxlocated upstream and downstream of the IHF-mediated DNAbend, and that the formed complex enhances transcription.IHF is thought to affect transcription mainly by acting as anarchitectural factor by several different mechanisms, oftenin concert with other global or gene-specific transcriptionalregulators [15,24,33,44]. The best studied examples ofactivation are atσ54-dependent promoters, where the mainrole of IHF is to position a NtrC family activator such thatit can directly contact RNA polymerase [4,7,9,27,54]. Apartfrom promoters controlled by NtrC-response regulators, fewpromoters regulated by a two-component system have beendescribed, which show an IHF-mediated activity [6,46,53].

PstyA expression might also be subjected to a negativecontrol. In fact deletion of the 5′ end direct repeat led to anincrease in the expression level ofβ-galactosidase in styreneand to a 2-fold increase in the induction level. A databasesearch did not allow the identification of possible regulatorsknown for binding sequences homologous to the identifieddirect repeat.

The contemporary presence inPstyA promoter of thepalindromicsty box sequence and of sequences correspond-ing to each half palindrome suggest that StyR could bindboth as a monomer and as a dimer. StyR is a responsetranscriptional activator that belongs to the FixJ family. Ithas been demonstrated that, in the non-phosphorylated FixJ,the receiver domain acts as an inhibitory module of theC-terminal DNA binding domain and that phosphorylationabolishes this inhibition through a conformational change

in the receiver domain [16,28]. More recently, Da Re et al.have demonstrated that FixJ phosphorylation also triggersits dimerization and that this process enhances ten times theaffinity to the promoter DNA [17]. Therefore in the case ofFixJ, the two main mechanisms of the activation of responseregulators are not mutually exclusive. It is possible that thesame mechanisms are present in other response regulators ofthe FixJ family. The amino acid sequence of the dimeriza-tion interface of FixJ [17] is 68% identical and 88% similarto the corresponding region of the StyR protein, where theV91 and K95 residues, which have been demonstrated to beessential for FixJ dimerization, are conserved.

In the bandshift competition experiments it was notpossible to identify the DNA complex(es) correspondingto StyR binding. This could be due to a low amount ofthe StyR active form in our cell extracts. This impliesthat complexesa and c might be generated by additionaltrans-acting elements. Still, we observed that complexesa and c were also formed in the presence of competitorDNAs which retained most of thecis elements essential forpromoter activity. This observation led us to suggest that5′ promoter deleted fragments were poor competitors andthat, whatever the nature of thetrans-acting elements, theformation of stable complexes probably requires a specificpromoter structure, only achieved when the entire promotersequence is present.

A vast body of literature is available on the regulatorymechanisms which control catabolism of aromatic com-pounds [10,20,45]. The picture emerging from these datais that in many cases both catabolic promoters and regula-tors have a certain degree of non-specificity with respect tothe signal they respond to, suggesting that in these systemsevolution towards optimal regulation is not completed [19].Nevertheless, all the catabolic systems for aromatic com-pounds studied up to now are subjected to catabolite repres-sion by preferred carbon sources with a plethora of mech-anisms most of which are not yet understood (reviewed inRef. [10]). This indicates that coupling of a catabolic path-way with the energetic status of the cell is a priority inthe establishment of new catabolic functions. In the two-component regulatory system here described, the responseto these two main signals, i.e., the presence of the specifichydrocarbon and the redox status of the cell, is most likelyintegrated at the level of the hybrid sensor kinase leadingto fine modulation of the amount of the response regulator-active form, which in turn determines the promoter activity.

Acknowledgements

We would like to thank Maia Kivisaar for the strain gifts,Gioacchino Micheli for computer prediction of DNA curva-ture and Valeria Di Stefano, and Sara Merlo for technical as-sistance. This work was supported by a grant from the targetproject on biotechnology (grant no. 9701252.49), ConsiglioNazionale delle Ricerche, Rome, Italy. P.M. Santos is the re-


cipient of a Ph.D. fellowship from the FCT (grant: PRAXISXXI/BD/15899/98), Portugal.

References

[1] G. Baggi, M.M. Boga, E. Catelani, E. Galli, V. Treccani, Styrenecatabolism by a strain ofPseudomonas fluorescens, Syst. Appl.Microbiol. 4 (1983) 141–147.

[2] K.A. Barne, J.A. Bown, S.J. Busby, S. Minchin, Region 2.5 of theEscherichia coli RNA polymerase sigma70 subunit is responsible forthe recognition of the ‘extended−10’ motif at promoters, EMBO J. 16(1997) 4034–4040.

[3] F. Beltrametti, A.M. Marconi, G. Bestetti, C. Colombo, E. Galli,M. Ruzzi, E. Zennaro, Sequencing and functional analysis of styrenecatabolism genes fromPseudomonas fluorescens ST, Appl. Environ.Microbiol. 63 (1997) 2232–2239.

[4] G. Bertoni, N. Fujita, A. Ishihama, V. de Lorenzo, Active recruitmentof sigma54-RNA polymerase to thePu promoter ofPseudomonasputida: Role of IHF and alphaCTD, EMBO J. 17 (1998) 5120–5128.

[5] J.M. Blatny, T. Brautaset, H.C. Winther-Larsen, K. Haugan, S. Valla,Construction and use of a versatile set of broad-host-range cloningand expression vectors based on the RK2 replicon, Appl. Environ.Microbiol. 63 (1997) 370–379.

[6] D.F. Browning, J.A. Cole, S.J. Busby, Suppression of FNR-dependenttranscription activation at theEscherichia coli nir promoter by Fis,IHF and H-NS: Modulation of transcription initiation by a complexnucleo-protein assembly, Mol. Microbiol. 37 (2000) 1258–1269.

[7] M. Buck, M.T. Gallegos, D.J. Studholme, Y. Guo, J.D. Gralla,The bacterial enhancer-dependent sigma(54) (sigma(N)) transcriptionfactor, J. Bacteriol. (2000) 4129–4136.

[8] R. Calb, A. Davidovitch, S. Koby, H. Giladi, D. Goldenberg,H. Magalit, A. Holtel, K. Timmis, J.M. Sanchez-Romero,V. de Lorenzo, A.B. Oppenheim, Structure and function of thePseudomonas putida integration host factor, J. Bacteriol. 178 (1996)6319–6326.

[9] M. Carmona, F. Claverie-Martin, B. Magasanik, DNA bending and theinitiation of transcription at sigma54-dependent bacterial promoters,Proc. Natl. Acad. Sci. USA 94 (1997) 9568–9572.

[10] I. Cases, V. de Lorenzo, The black cat/white cat principle of signalintegration in bacterial promoters, EMBO J. 20 (2001) 1–11.

[11] D. Charlier, M. Roovers, D. Gigot, N. Huysveld, A. Pirard, N.Glansdorff, Integration host factor (IHF) modulates the expression ofthe pyrimidine-specific promoter of thecarAB operons ofEscherichiacoli K12 and Salmonella typhimurium LT2, Mol. Gen. Genet. 237(1993) 273–286.

[12] J. Collado-Vides, B. Magasanik, J.G. Gralla, Control site locationand transcriptional regulation inEscherichia coli, Microbiol. Rev. 55(1991) 371–394.

[13] M.B. Colonna, G. Prosseda, G. Micheli, C.O. Gualerzi, Thermo-regulation of Shigella and Escherichia coli EIEC pathogenicity.A temperature-dependent structural transition of DNA modulatesaccessibility of virF promoter to transcriptional repressor H-NS,EMBO J. 17 (1998) 7033–7043.

[14] P.W. Coschigano, L.Y. Young, Identification and sequence analysisof two regulatory genes involved in anaerobic toluene metabolism bystrain T1, Appl. Environ. Microbiol. 63 (1997) 652–660.

[15] X. Dai, L.B. Rothman-Denes, DNA structure and transcription, Curr.Opin. Microbiol. 2 (1999) 126–130.

[16] S. Da Re, S. Bertagnoli, J. Fourment, J.M. Reyrat, D. Kahn, Intramole-cular signal transduction within the FixJ transcriptional activator: Invitro evidence for the inhibitory effect of the phosphorylatable regula-tory domain, Nucleic Acids Res. 22 (1994) 1555–1561.

[17] S. Da Re, J. Shumacher, P. Rousseau, J. Fourment, C. Ebel, D. Kahn,Phosphorylation-induced dimerization of the FixJ receiver domain,Mol. Microbiol. 34 (1999) 504–511.

[18] V. de Lorenzo, M. Herrero, M. Metzke, K.N. Timmis, An upstreamXylR- and IHF-induced nucleoprotein complex regulates the sigma54-dependentPu promoter of TOL plasmid, EMBO J. 10 (1991)1159–1167.

[19] V. de Lorenzo, J. Perez-Martin, Regulatory noise in prokaryoticpromoters: How bacteria learn to respond to novel environmentalsignals, Mol. Microbiol. 19 (1996) 1177–1184.

[20] E. Díaz, M.A. Prieto, Bacterial promoters triggering biodegradation ofaromatic pollutants, Curr. Opin. Biotechnol. 11 (2000) 467–475.

[21] D.H. Figursky, D.R. Helinski, Replication of an origin-containingderivative of plasmid RK2 dependent on a plasmid function providedin trans, Proc. Natl. Acad. Sci. USA 76 (1979) 1648–1652.

[22] F.C.H. Franklin, M. Bagdasarian, M.M. Bagdasarian, K.N. Timmis,Molecular and functional analysis of the TOL plasmid pWWO fromPseudomonas putida and cloning of the genes of the entire regulatedaromatic ringmeta-cleavage pathway, Proc. Natl. Acad. Sci. USA 78(1981) 7458–7462.

[23] J.A. Goodrich, M.L. Schartz, W.R. McClure, Searching for andpredicting the activity of sites for DNA binding proteins: Compilationand analysis of the binding sites fromE. coli integration host factor(IHF), Nucleic Acid Res. 18 (1990) 4993–5000.

[24] N. Goosen, P. van de Putte, The regulation of transcription initiationby integration host factor, Mol. Microbiol. 16 (1995) 1–7.

[25] S. Hartmans, J.P. Smits, M.J. Van der Werf, F. Volkering, J.A.M. deBont, Metabolism of styrene oxide and 2-phenylethanol in the styrene-degradingXanthobacter strain 124X, Appl. Environ. Microbiol. 55(1989) 2850–2855.

[26] F. Hayes, S. Austin, Topological scanning of the P1 plasmid partitionsite, J. Mol. Biol. 243 (1994) 190–198.

[27] T.R. Hoover, E. Santero, S. Porter, S. Kustu, The integration hostfactor stimulates interaction of RNA polymerase with NifA, thetrascriptional activator for nitrogen fixation operons, Cell 63 (1990)11–22.

[28] D. Kahn, G. Ditta, Modular structure of FixJ: Homology of thetranscriptional activator domain with the−35 binding domain ofsigma factors, Mol. Microbiol. 5 (1991) 987–997.

[29] D. Labbé, J. Garmon, P.C.K. Lau, Characterization of the genes en-coding a receptor-like histidine kinase and cognate response regulatorfrom biphenyl/polychlorobiphenyl-degrading bacterium,Rhodococ-cus sp. strain M5, J. Bacteriol. 179 (1997) 2772–2776.

[30] P.C.K. Lau, Y. Wang, A. Patel, D. Labbé, H. Bergeron, R. Brousseau,Y. Konisshi, M. Rawlings, A bacterial basic region leucine zipperhistidine kinase regulating toluene degradation, Proc. Natl. Acad. Sci.USA 94 (1997) 1453–1458.

[31] A.M. Marconi, F. Beltrametti, G. Bestetti, F. Solinas, M. Ruzzi,E. Galli, E. Zennaro, Cloning and characterization of styrenecatabolism genes fromPseudomonas fluorescens ST, Appl. Environ.Microbiol. 62 (1996) 121–127.

[32] A.M. Marconi, M. Ruzzi, S. Iurescia, E. Zennaro, Genetics of styrenein bacteria, Rec. Res. Dev. Microbiol. 2 (1998) 95–105.

[33] S.M. McLeod, R.C. Johnson, Control of transcription by nucleoidproteins, Curr. Opin. Microbiol. 4 (2001) 152–159.

[34] J.H. Miller, Experiments in Molecular Genetics, Cold Spring Harbor,New York, 1972.

[35] G. Mosqueda, M.I. Ramos-Gonzales, J.L. Ramos, Toluene metabolismby the solvent-tolerantPseudomonas putida DOT-T1 strain, and itsrole in solvent impermeabilization, Gene 232 (1999) 69–76.

[36] K. O’Brien, G. Deno, P. Ostrovsky de Spicer, J.F. Gardner, S.R.Maylor, Integration host factor facilitates repression of theput operonin Salmonella typhimurium, Gene 118 (1992) 13–19.

[37] K. O’Connor, C.M. Buckley, S. Hartmans, A.D.W. Dobson, Possibleregulatory role for nonaromatic carbon source in styrene degradationby Pseudomonas putida CA-3, Appl. Environ. Microbiol. 61 (1995)544–548.

[38] K.E. O’Connor, A.D.W. Dobson, S. Hartmans, Indigo formationby microorganisms expressing styrene monoxygenase activity, Appl.Environ. Microbiol. 63 (1997) 4287–4291.


[39] K. O’Connor, W. Duetz, B. Wind, A.D.W. Dobson, The effect ofnutrient limitation on styrene metabolism inPseudomonas putidaCA-3, Appl. Environ. Microbiol. 62 (1996) 3594–3599.

[40] N.D. O’Leary, K.E. O’Connor, W. Duetz, A.D.W. Dobson, Transcrip-tional regulation of styrene degradation inPseudomonas putida CA-3,Microbiology 147 (2001) 973–979.

[41] S. Panke, B. Witholt, A. Schmid, M. Wubbolts, Towards a biocatalystfor (s)-styrene oxide production: Characterization of styrene degra-dation pathway ofPseudomonas sp. Strain VLB120, Appl. Environ.Microbiol. 64 (1998) 2032–2043.

[42] L. Paul, R.M. Blumenthal, R.G. Matthews, Activation from a distance:Roles of Lrp and integration host factor in transcriptional activation ofgltBDF, J. Bacteriol. 183 (2001) 3910–3918.

[43] J. Perez-Martin, V. de Lorenzo, Integration host factor (IHF) sup-presses promiscuous activation of theσ54-dependent promoterPu ofPseudomonas putida, Proc. Natl. Acad. Sci. USA 92 (1995) 7277–7281.

[44] J. Pérez-Martín, K.N. Timmis, V. de Lorenzo, Co-regulation by bentDNA, J. Biol. Chem. 269 (1994) 22657–22662.

[45] J. Powlowski, V. Shingler, Genetics and biochemistry of phenoldegradation byPseudomonas sp. CF600, Biodegradation 5 (1994)219–236.

[46] N. Ramani, L. Huang, M. Freundlich, In vitro interactions of inte-gration host factor with theompF promoter-regulatory region ofEs-cherichia coli, Mol. Gen. Genet. 231 (1992) 248–255.

[47] V.A. Rhodius, S.J. Busby, Positive activation of gene expression, Curr.Opin. Microbiol. 1 (1998) 152–159.

[48] F. Rojo, Mechanisms of transcriptional repression, Curr. Opin. Micro-biol. 4 (2001) 145–151.

[49] J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular Cloning: A Labora-tory Manual, 2nd edn., Cold Spring Harbor Laboratory, Cold SpringHarbor, NY, 1989.

[50] P.M. Santos, J.M. Blatny, I. Di Bartolo, S. Valla, E. Zennaro,Physiological analysis of expression of styrene degradation gene

cluster inPseudomonas fluorescens ST, Appl. Environ. Microbiol. 66(2000) 1305–1310.

[51] P.M. Santos, I. Di Bartolo, J.M. Blatny, E. Zennaro, S. Valla, Newbroad-host-range promoter probe vectors based on the plasmid RK2replicon, FEMS Microbiol Lett. 195 (2001) 91–96.

[52] M. Shimizu, M. Miyake, F. Kanke, U. Matsumoto, H. Shindo,Characterization of the binding of HU and IHF, homologous histone-like protein of Escherichia coli, to curved and uncurved DNA,Biochim. Biophys. Acta 1264 (1995) 330–336.

[53] A. Sirko, E. Zehelein, M. Freundlich, G. Sawers, Integration hostfactor is required for anaerobic pyruvate induction ofpf1 operonexpression inE. coli, J. Bacteriol. 175 (1993) 5769–5777.

[54] C.C. Sze, A.D. Laurie, V. Shingler, In vivo and in vitro effects ofintegration host factor at the DmpR-regulated sigma(54)-dependentPopromoter, J. Bacteriol. 183 (2001) 2842–2851.

[55] R. Teras, R. Hõrak, M. Kivisaar, Transcription from fusion promot-ers generated during transposition of transposon Tn4652 is positivelyaffected by integration host factor inPseudomonas putida, J. Bacte-riol. 182 (2000) 589–598.

[56] R.J. Thompson, G. Mosig, Integration host factor (IHF) repressesa Chlamydomonas chloroplast promoter inE. coli, Nucleic AcidsRes. 16 (1988) 3313–3326.

[57] P. Tsui, M. Freundlich, Integration host factor binds specifically tosites in theilvGMEDA operon inEscherichia coli, J. Mol. Biol. 203(1988) 817–820.

[58] A. Velasco, S. Alonso, J.L. García, J. Perera, E. Díaz, Genetic andfunctional analysis of the styrene catabolic cluster ofPseudomonassp. strain Y2, J. Bacteriol. 180 (1998) 1063–1071.

[59] I.B. Zhulin, B.L. Taylor, R. Dixon, PAS domain S-boxes in Archea,Bacteria and sensors for oxygen and redox, Trends Biochem. Sci. 22(1997) 331–333.

Integration host factor is essential for the optimal expression of the styABCD operon in Pseudomonas...

Documents

Transcript of Integration host factor is essential for the optimal expression of the styABCD operon in Pseudomonas...