IHF Is Required for the Transcriptional Regulation of the Desulfovibrio vulgaris Hildenborough orp Operons

Transcriptional activation of σ54-dependent promoters is usually tightly regulated in response to environmental cues. The high abundance of potential σ54-dependent promoters in the anaerobe bacteria, Desulfovibrio vulgaris Hildenborough, reflects the high versatility of this bacteria suggesting that σ54 factor is the nexus of a large regulatory network. Understanding the key players of σ54-regulation in this organism is therefore essential to gain insights into the adaptation to anaerobiosis. Recently, the D. vulgaris orp genes, specifically found in anaerobe bacteria, have been shown to be transcribed by the RNA polymerase coupled to the σ54 alternative sigma factor. In this study, using in vitro binding experiments and in vivo reporter fusion assays in the Escherichia coli heterologous host, we showed that the expression of the divergent orp promoters is strongly dependent on the integration host factor IHF. Bioinformatic and mutational analysis coupled to reporter fusion activities and mobility shift assays identified two functional IHF binding site sequences located between the orp1 and orp2 promoters. We further determined that the D. vulgaris DVU0396 (IHFα) and DVU1864 (IHFβ) subunits are required to control the expression of the orp operons suggesting that they form a functionally active IHF heterodimer. Interestingly results obtained from the in vivo inactivation of DVU0396, which is required for orp operons transcription, suggest that several functionally IHF active homodimer or heterodimer are present in D. vulgaris.


Introduction
Desulfovibrio vulgaris Hildenborough (DvH) is a well-studied sulfate-reducing bacteria (SRB).It is the first SRB for which the genome has been sequenced [1] and genetics and cellular tools have been recently developed [2].It is an anaerobic microorganism that exhibits a mode of growth based on the reduction of sulphate as a terminal acceptor and thus is directly linked to natural global sulphur and carbon cycles [3].These microorganisms have particularly important applications in biotechnology as they are involved in toxic heavy metal bioremediation, hydrogen sulfur decontamination in wastewater and biocorrosion [4].Although SRBs are ubiquitous in anoxic habitats, they are found in a wide variety of ecological niches as illustrated by their metabolic versatility.Such environmental adaptations require stimuli perception and the subsequent modulation of the expression of relevant genes to optimize metabolism and physiology.Indeed, an unusually large number of response regulators are encoded within the DvH chromosome.Interestingly, 70 potential s 54 promoters have been identified by computational prediction [5,6] suggesting that s 54 factor is the nexus of a large regulatory network in DvH.The alternative s 54 factor associates with the RNA polymerase holoenzyme to target specific genes.However, to proceed to transcription, the s 54 -RNA polymerase complex requires the action of enhancer binding proteins (EBPs).EBPs usually bind to regulatory sequences upstream the s 54 -dependent promoter [7,8,9,10].Efficient and functional interaction of s 54 -RNA polymerase with a cognate EBP requires DNA bending, which is often facilitated by the integration host factor (IHF) protein, especially for promoters that lack flexible intrinsic bends [11,12,13,14,15].In some cases, IHF has also been shown to recruit the s 54 -RNA polymerase to its promoter [14].IHF is involved in many cellular processes correlated with DNA functions such as replication, transcription, partitioning, transfer and packaging into phage particles [11,16,17].Aside its role in specific transcriptional activation of s 54 promoters, IHF contributes to a wide variety of macromolecular processes and is recognized, with CRP, FNR, FIS, ArcA, Lrp and H-NS, as one of the seven global regulators in Escherichia coli [18,19].HU, H-NS, FIS and IHF are basic proteins involved in the organization of the DNA nucleoid architecture and are therefore considered as histone-like proteins [20].In E. coli and related bacteria, IHF is a 20-kDa protein composed of two subunits, a and b.Although the ab heterodimer is the predominant active form, the aa and bb homodimers might be biologically active [21,22,23,24].
In addition to the 70 putative s 54 -dependent promoters, DvH genome analyses revealed the presence of 37 putative s 54associated EBPs and several genes encoding IHFa and IHFb subunits [25].To date, little is known about the specific roles of s 54 , the cognate EBPs and IHF in SRBs.Elucidation of their functions is thus essential to generate predictive model of the adaptation, stress responses of SRBs to environmental factors, and to develop effective SRB-based biotechnologies.In a recent study, we examined the function of the s 54 -dependent transcriptional regulator DVU2106 in the regulation of the divergent orp operons in DvH, which encode proteins putatively involved in positioning the septum in cell division [6].We showed that this EBP collaborates with the s 54 -RNA polymerase to orchestrate the simultaneous expression of the orp genes cluster [6].Here, we tested the contribution of IHF to this regulatory mechanism.In a reconstitution assay into the heterologous host E. coli, we found that IHF is required for efficient DVU2106-dependent activation of the s 54 orp promoters.Using electrophoretic mobility gel shift assays, the purified E. coli IHF protein was shown to bind to the orp promoters.We further localized IHF recognition sites between the s 54 -and DVU2106-binding sequences and determined which of the three potential IHF binding sites were of physiological relevance using site-directed mutagenesis, reporter fusion activity and in vitro gel shifts.We identified DVU0396 (IHFa) and DVU1864 (IHFb) subunits from DvH as the heterodimer functionally active to complement the transcription of orp operons in the IHF-deficient E. coli strain.Finally, we tested the role of the IHFa subunit on the transcription of the orp operons in DvH.

Bacterial Strains, Plasmids, Oligonucleotides and Growth Conditions
The bacterial strains and plasmids used in this study are listed in the Table S1.Custom oligonucleotides are listed in the Table S2.E. coli strains were routinely grown at 37uC in Luria-Bertani (LB) medium supplemented with the appropriate antibiotic when required (0.27 mM for ampicillin, 0.15 mM for chloramphenicol and 0.20 mM for kanamycin).Cultures of DvH were prepared in Postgate C medium [26] supplemented with 0.17 mM of kanamycin or 0.15 mM of thiamphenicol at 33uC in anaerobic conditions.

DNA Manipulations
Standard protocols were used for cloning and transformations.All restriction endonucleases and DNA modification enzymes were purchased from New England Biolabs.For electrophoretic mobility shift assay, PCRs were performed using the Expand High Fidelity (Roche Diagnostics).Site-directed mutagenesis was performed using the PfuTurboH or the PfuUltra TM High Fidelity DNA polymerases (Stratagene).Chromosomal DNA was purified using the Wizard Genomic DNA purification kit (Promega).Plasmid DNA was purified using the High Pure Isolation Plasmid Kit (Roche Diagnostics).PCR products and plasmid fragments were purified using MinElute kits (Qiagen).

Expression Plasmids
For plasmids producing DvH IHFa, IHFb and IHFa+b, the strong and constitutive lpp promoter was PCR amplified from E. coli K12 genomic DNA and cloned into the pOK12-2106 vector [6] downstream the DVU2106 gene using the unique EcoRI and NdeI restriction sites, yielding pOK12-2106-lpp.The IHFaencoding DVU0396 or IHFb-encoding DVU1864 genes were PCR-amplified from DvH genomic DNA and cloned downstream the lpp promoter in pOK12-2106-lpp using the unique NdeI and NotI restriction sites to yield pOK12-2106-lpp-0386 and pOK12-2106-lpp-1864 respectively.For construction of the plasmid carrying both genes, DVU0396 was PCR-amplified and cloned downstream DVU1864 using the NotI restriction site resulting in pOK12-2106-lpp-1864-0396.

Strain Construction
The W3110 DlacZ DihfA E. coli strain was constructed by P1 transduction from the BW25113 ihfA::kan R strain from the KEIO collection [27] and cassette excision using pCP20 as previously described [28] using W3110 DlacZ as recipient [29].The DvH Dihfa mutant strain was constructed using previously described protocols [30,31].Briefly, the 455-pb fragment upstream the DVU0396 gene was PCR-amplified using oligonucleotides DVU0395AscI_dir and CterDVU0395-SpeI and cloned into the AscI and SpeI sites of plasmid pDel.The 460-pb fragment downstream DVU0396 was PCR-amplified using oligonucleotides DVU0396MfeI_dir and DVU0396BglII_rev and cloned into MfeI and BglII sites of the plasmid previously constructed to yield pDelihfa.The accuracy of the cloned fragments was verified by restriction and DNA sequencing.The pDelihfa plasmid was transferred into DvH cells by electroporation as following: DvH cells grown in PC medium were harvested at early stationary phase (OD 600 ,0.8-1) and washed twice with cold and degassed sterile MilliQ-water.Approximately 500 ng of the pDelihfa plasmid were electroporated with a single 1.9 KV, 25 mF and 250 V pulse into DvH competent cells in BTX 1-mm gapped cuvette using a BTX Harvard Electro Cell ManipulatorH ECM630 apparatus.Cells were recovered in 30 ml of PC medium, and the culture was supplemented with 20 mg/ml of thiamphenicol after 6 hours of incubation at 33uC.After 7 days of culture in the same conditions, the culture was diluted in PC medium supplemented with thiamphenicol.After three rounds of culture and dilution, individual colonies were recovered by plating on PE medium supplemented with thiamphenicol.The deletion of the ihfa gene was verified by colony-PCR.

RNA Preparation
DvH was grown in three 80-ml independent cultures of DvH to an OD 595 ,0.4.The cells were harvested and resuspended in 200 ml of 10 mM Tris pH8.RNA was prepared by using the High Pure RNA kit (Roche Diagnostics) according to the manufacturer's instructions with an additional DNase I digestion step.
Real-time Quantitative PCR 10 mg of total extracted RNA were reverse transcribed by using random hexamers to yield cDNA. Real-time PCR was performed on a LightCyclerFastStart DNA Master PLUS SYBR Green I Kit (Roche Diagnostics).cDNA was mixed with 0.25 mM of each primer and 2 ml of master mix in a 10 ml final volume.The primers pairs used to quantify the expression levels of the selected genes are shown in Table S2.PCR amplifications were carried out with one cycle at 95uC for 8 min, followed by up to 45 cycles at 95uC for 12 s, 60uC for 6 s and 72uC for 20 s.The specificities of accumulated products were verified by melting-curve analysis.The fluorescence derived from the incorporation of SYBR Green I into the double-stranded PCR products was measured at the end of each cycle to determine the amplification kinetics of each product.The fit points method described by the manufacturer was then applied to the results.Briefly, a horizontal noise band was determined as well as a log line fitting the exponential portion of the amplification curve.The intersections of these log lines with the horizontal noise lines identified the crossing points.These crossing points were determined for each gene in three conditions.Relative Expression Software Tool (REST) was used to calculate the relative expression of each gene in each condition using the 16S RNA gene as reference for normalization.

b-galactosidase Assay
The activity of the b-galactosidase reporter was measured from mid-exponential growth phase cultures (OD 600 ,0.8) as previously described [32].Each enzymatic assay was performed in triplicate with three independent biological samples.The values are reported as the average of these nine measurements.

Purification of the E. coli IHF Heterodimer
E. coli MC4100 pTrc-IHF [33] was grown overnight in 10 ml of LB supplemented with kanamycin at 37uC with shaking and used to inoculate 1 liter of LB.This culture was grown until the OD 600 reached 0.6, and 1 mM of IPTG was then added.After 3H of induction, the E.coli IHF heterodimer was extracted as described [33].Briefly, cells were washed with 40 ml of Buffer E (25 mM Tris-HCl pH7.4, 1 mM EDTA, 3 mM b-mercaptoethanol, 100 mM NaCl, 1 mM PMSF and 10% glycerol).After centrifugation, cells were resuspended in 10 ml of Buffer E supplemented with DNase.Cells were broken by 3 passages in French-press and centrifugated at 45000 rpm during 1 H at 4uC.IHF heterodimer was then purified by two consecutive steps: immobilization onto a HiTrap Heparin HP column equilibrated in Buffer E and elution with a step gradient of KCl (0.2 M-2 M).The fractions containing IHF (0.8 to 1.2 M KCl) were pooled and diluted in Buffer A (25 mM Tris-HCl pH7.4, 1 mM EDTA, 3 mM b-mercaptoethanol and 10% of glycerol) to obtain a final KCl concentration of 100 mM, immobilized onto a HiTrap SP FF column and eluted with a step gradient of KCl.The fractions containing the IHF heterodimer (500-600 mM of KCl) were pooled and the buffer was exchanged against Buffer A. The final concentration of IHF was ,1.5 mg/ml.

IHF is Required for orp1 and orp2 Transcription
It was previously shown that the two divergent orp operons require the s 54 RNA polymerase and the cognate transcriptional EBP, DVU2106, to achieve transcription [6].It is well described that DNA bending allowing contacts between the cognate activator and s 54 -RNA polymerase to activate transcription is usually facilitated by the heterodimeric IHF protein.To test the contribution of IHF to the expression of these genes, the previously described orp1, orp2 and DVU2106 promoter-lacZ transcription fusions were introduced into the heterologous host E. coli.Expression of these promoter-lacZ fusions were examined in both IHF-proficient (wild-type) and IHF-deficient (DihfA) strains that produced the transcriptional regulator DVU2106 from a compatible multicopy plasmid (Figure 1A-D).The expression of DVU2106, controlled by the s 70 -RNA polymerase, was not influenced by the absence of IHF (Figure 1C).By contrast, the activities of the orp1 and orp2 promoters, under the control of the s 54 -RNA polymerase, were significantly affected by the absence of IHF as the orp1'::lacZ and orp2'::lacZ expression levels decreased 10-12 fold compared to the wild-type strain producing IHF (Figure 1B and 1D).These results show that the E. coli IHF heterodimer can activate orp operons transcription and that IHF is required for the expression of these s 54 -dependent operons.The E. coli IHF Protein Binds to the Promoter Regions of orp1 and orp2 To address whether IHF can bind to these promoters, we carried out electrophoretic mobility shift assays (EMSA). 32Plabeled orp1 and orp2 promoter fragments were incubated with increasing concentrations of purified E. coli IHF heterodimers.As shown in Figure 2, the two DNA fragments were specifically retarded in presence of $4 nM of IHF.These data demonstrate that the IHF protein interacts with the promoter regions of the orp1 and orp2 operons.

Identification of the Functional IHF Binding Sites
In silico analyses of the intergenic region of the orp cluster using the Virtual Footprint program (http://prodoric.tu-bs.de/vfp/vfp_promoter.php)suggested the presence of two potential IHF binding sites within the orp1 promoter (called hereafter IHF1 and IHF2) located at positions 242 and 279, respectively, from the transcriptional start site of orp1 (Figures 3 and S1) and one potential binding site for IHF within the orp2 promoter, located at position 243 from the transcriptional start site of orp2 (Figures 3  and S1).The orp1 IHF1 site (AATCAGAATAAA) [conserved nucleotides are in bold]) has 78% similarity with the reported E. coli consensus sequence with the conserved nucleotides mainly located at the 59 extremity of the sequence whereas the orp1 IHF2 site (CATCACAAGCTCG) shares less similarity (67%) but better distributed throughout the sequence (Figures S1 and S3).The orp2 potential IHF site (AATCAAACATCTT) has 78% similarity with the reported E. coli consensus sequence.Regarding the relative position of these IHF binding sites compared to those of s 54 and of DVU2106, the orp1 IHF1 (242) and IHF2 (279) sites are located upstream the s 54 -binding sequence (position 213) and downstream the EBP binding site (position 2131) (Figure S1).The orp2 IHF binding site (243) is located upstream the s 54 -binding sequence (position 211) and downstream the EBP binding site (position 288) (Figure S1).Therefore, the locations of the three potential IHF binding sites are centred between s 54 -and DVU2106-binding sequences.
To determine the contribution of these IHF binding sites to the transcriptional activities of the orp1 and orp2 promoters, each of these consensus sites was altered by generating mutations at bases previously reported to be necessary for IHF binding [35] yielding orp1 IHF1-mut, orp1 IHF2-mut and orp2-IHF-mut and DVU2106-IHF-mut (DVU2106 promoter region in which orp2 IHF site was mutated) (Figure S2).These mutations were introduced into plasmids carrying promoters-lacZ transcription fusions.As done previously, expression of the wild-type and mutated promoter-lacZ fusions were compared in the IHF-proficient (wild-type) and IHFdeficient (DihfA) strains that produced the transcriptional regulator DVU2106 from a compatible multicopy plasmid.The accumulation of b-galactosidase was measured and reported in Figure 4.The expression of orp1 was not significantly affected by the mutation within the IHF1 site, as 98% of the b-galactosidase activity was retained compared to the IHF proficient strain with native orp1'::lacZ fusion (Figure 4A).By contrast, the b-galactosidase activity decreased to 8% of the activity of the wild-type promoter when the IHF2 site was altered (Figure 4B), a value similar to that of the wild-type reporter fusion in an IHF-deficient strain (compare Fig. 4B and Fig. 1B).When both sites were altered, the residual activity of the orp1 promoter was comparable to that observed with the single IHF2 site mutation (Figure 4C).Regarding the expression from the orp2 promoter, mutation of the potential IHF site led to a 10-fold decreased b-galactosidase activity compared to the wild-type reporter fusion (Figure 4E).Here again, this activity is comparable to that of the wild-type  fusion in IHF-deficient cells (compare Fig. 4E and Fig. 1D).The expression of DVU2106, controlled by s 70 -RNA polymerase, was not influenced by the absence of IHF protein and by the disruption of the orp2 IHF site (Figure 4D).These results demonstrate the requirement of the orp1 IHF2 and orp2 IHF sites for the efficient transcription of the s 54 -dependent orp1 and orp2 promoters, respectively.These results further suggest that theses sites could be targets for IHF binding.
To address whether IHF binds to these putative IHF binding sites, we carried out EMSA experiments with increasing concentrations of the purified E. coli IHF protein and 32 P-labeled orp1 and orp2 mutated promoters.In agreement with the b-galactosidase activities, figure 5 shows that disruption of the orp1 IHF1 site did not prevent formation of the IHF-DNA complex (Figure 5A) whereas mutation within the IHF2 site (or within both IHF1 and IHF2 sites) abolished IHF binding (Figure 5B and 5C).Similarly, alteration of the orp2 IHF site significantly prevented IHF binding to the orp2 intergenic region (Figure 5D).It should be noted that 8 nM of IHF are needed for observing the formation of the IHF/ orp1 IHF2 complex when the orp1 IHF1 site is altered (Figure 5A) whereas only 4 nM of IHF are necessary to form the same complex when orp1 IHF1 is not altered (Figure 2A) suggesting that orp1 IHF1 might promote binding of IHF on the IHF2 site.
Taken together, these results demonstrate that the orp1 IHF2 site is required for IHF binding and orp1 transcription and the orp2 IHF site recruits IHF to activate orp2 transcription.It is noteworthy that the purified IHF protein binds to sites centred between DVU2106 binding sites and s 54 promoters in both operons, a result compatible with IHF function in DNA bending to facilitate contacts between the distantly-located EBP and s 54 -RNA polymerase.

The DvH DVU0396-DVU1864 Heterodimer is Functionally Active to Stimulate the Transcription of the orp Operons
IHF is usually described as a heterodimeric protein complex composed of the IHFa and IHFb subunits encoded by the ihfA (himA) and ihfB (himB) genes.Genome analyses of various species known to encode active heterodimeric IHF revealed the presence of a single copy of genes encoding IHFa and IHFb, found at distinct locations on the chromosome and subject to independent regulatory influences [36].Surprisingly, two orthologs of himB, DVU1864 and DVU2973 that would encode IHFb, are annotated on the DvH genome.Amino acid sequences of DVU1864 (called here b1) and DVU2973 (called here b2) are 65% identical together and respectively 42% and 45% identical to the E. coli IHFb (Figure 6).By contrast, no gene encoding IHFa was annotated whereas six orthologs of hup genes encoding HU proteins were detected.Nevertheless, it has been previously proposed that two of the putative HU-encoding genes, DVU0396 and DVUA0004, correspond to himA that encodes the  IHFa protein [25].Primary sequence comparison of DVU0396 and DVUA0004 with IHFa from E. coli revealed only 31% and 21% identity, respectively (Figure 6).More recently, results obtained from pull-down experiments using DVU1864 as bait identified DVU0396 as the preferential prey [37].It should also be noted that DVUA0004 is located on the endogenous plasmid of DvH, which is lost in our growth conditions.
In order to identify the active IHF from DvH, orp1 and orp2 promoter-lacZ transcription fusions were introduced into the IHFa-deficient E. coli (DihfA) strain that produces the transcriptional regulator DVU2106 and the HU protein DVU0396 or both DVU0396 and DVU1864 from a compatible multicopy plasmid (Figure 7).Surprisingly, the construction producing DVU0396 was easily obtained in wild-type E. coli whereas the plasmid producing DVU0396 and DVU1864 was obtained after reducing the production of these two genes suggesting a toxic effect of the simultaneous production of both proteins.By contrast, no growth effect was observed when DVU0396 and DVU1864 are produced in the IHF-deficient E. coli strain (data not shown).Figure 7 shows that the b-galactosidase activities of the orp1 and orp2 reporter fusions were very low in the DihfA strain, independently of the presence of the DVU0396 protein.However, the concomitant production of DVU0396 and DVU1864 in the DihfA strain restored the activities of the orp1 and orp2 promoters to levels comparable to that observed in the wild-type strain.These results show that DVU0396 and DVU1864 are both required to activate orp1 and orp2 expression in vivo and further suggest that these two proteins may interact to assemble a functional heterodimer.

Functional Analysis of the DvH DVU0396 ihfa Mutant
As shown above, the DVU0396 and DVU1864 proteins may interact to stimulate the transcription of the orp operons in the in vivo reconstitution approach using the heterologous host E. coli.The amino-acid sequences of the two putative IHFb proteins encoded by DVU1864 and DVU2973 being 65% identical, one may suggest that both proteins may have redundant function.From these observations, and in order to determine whether IHF is physiologically required for the activation of the divergent orp operons in DvH, we deleted the DVU0396 gene on the DvH chromosome.Under anaerobic conditions and in lactate/sulfate medium, the deletion of DVU0396 did not induce significant effects on DvH growth (final biomass and growth rate).The transcription levels of orp1, orp2 and DVU2106 were compared by qRT-PCR in the wild-type and the DVU0396 mutant strains in both exponential and stationary phases.The expression ratio of orp1, orp2, and DVU2106 [log2(absolute gene regulation) = 20.406,20.404 and 20.805, respectively, using the wild-type strain as reference] showed that the absence of DVU0396 protein has no effect on orp1 and orp2 expression in DvH.Our further qRT-PCR analyses showed that the absence of DVU0396 is not compensated by an increasing expression of the genes encoding the IHFb1 (DVU1864) or IHFb2 (DVU2973) subunit [log2(absolute gene regulation) = 0.797 and 0.308 respectively, using the wild-type strain as reference].Taken together these results indicate that the absence of DVU0396 is not compensated by the upregulation of genes encoding IHFb subunits and does not affect the expression of the orp operons in DvH.

Discussion
In most environments, microorganisms are under tough competition for available resources and have to perceive, integrate and respond to multiple signals pertaining to a variety of stresses that include limited nutrient availability, physicochemical stresses and pollutant toxicity.The ability to appropriately adapt to prevailing conditions usually involves complex regulatory mechanisms such as two-component systems and s 54 -dependent promoters.In addition to the s 54 alternative factor, the s 54 pathway requires enhancer binding proteins and DNA bending.EBPs interact with an upstream activating sequence, and through DNA looping, which can be facilitated by IHF, contact the s 54 -RNA polymerase.Interestingly, a large number of potential s 54promoters and EBPs have been identified in DvH, an observation in accordance with the high metabolic versatility of this species [2].To date nothing is known on the role of IHF in this bacterium and more particularly on its contribution to the s 54 regulatory mechanism.We recently showed that two gene operons, orp1 and orp2, are regulated by s 54 and a cognate EBP, DVU2106 [6].In this study, we have refined the regulatory mechanism underlying orp gene expression by demonstrating that IHF contributes to the regulation of these genes.
Using an in vivo reconstitution approach in E. coli, we showed that the activities of the divergently transcribed orp1 and orp2 operon promoters require the integration host factor IHF.Although the predominant role of IHF is to facilitate contacts between the EBP and the s 54 -RNA polymerase, it was shown in P. putida that IHF is also important to recruit the s 54 -RNA polymerase to its promoter [14].The low expression of the orp operons measured in absence of IHF and the location of the IHF binding sites, located between the DVU2106 EBP-and the s 54binding elements suggest a direct role of IHF to assist in the formation of direct contacts between DVU2106 and the RNA polymerase and/or to recruit the s 54 -RNA polymerase to the appropriate site [38,39,40].The activities of the orp1 and orp2 promoters were reduced 10-12 fold in absence of IHF or when IHF-binding boxes were altered, probably reflecting the limitations imposed by the absence of IHF-induced DNA bending.In several instances, IHF has been shown to be dispensable for s 54dependent regulation; however, in these cases, DNA bending is induced by stretches of A/T-rich sequences between s 54 -and EBP-binding elements [35,41].No A/T-rich region that may induce natural curvature of DNA is found between the DVU2106and the s 54 -binding sites of orp1 and orp2 (Figure S1), an observation that may explain the important requirement for IHF for the regulation of the orp1 and orp2 operons.Such drastic effect was observed for the TodS-TodT two-component regulatory system from P. putida in which the absence of IHF causes a 8-fold decrease of the todX promoter expression compared to the wildtype strain [38].
It is noteworthy that these assays have been performed in the heterologous host E. coli.The fact that the absence of the E. coli IHF causes a decreased in expression of the orp1 and orp2 promoters demonstrates that despite the weak homology between the E. coli and DvH proteins, they are functionally similar.We therefore took advantage of this observation to (i) use the purified E. coli IHF protein for in vitro studies and (ii) to identify the DvH functional IHF heterodimer in reconstitution experiments in the E. coli DihfA strain.
The E. coli IHF protein was purified and electrophoretic mobility shift experiments using orp1 and orp2 promoter probes showed that, as expected from the IHF requirement for transcriptional activation, IHF binds to both promoters.We therefore went further by identifying the IHF binding sequences within the orp1 and orp2 promoters using a combination of computational and mutational analyses coupled to mobility shift and promoters-lacZ transcription reporter fusion assays.Bioinformatic analyses revealed the presence of two putative IHF boxes upstream the orp1 s 54 promoter and one upstream the orp2 s 54 promoter.We showed that IHF binds the orp1 IHF2 and the orp2 IHF binding sequences.The results from the transcriptional reporter fusions were consistent with these data as alteration within the orp1 IHF1 site has no effect on the expression level, whereas alterations within the orp1 IHF2 or orp2 IHF sites decrease the expression to levels comparable to that of the wild-type promoters in the DihfA strain.Therefore, the orp1 IHF2 and orp2 IHF binding site are required for IHF-mediated activation of the orp1 and orp2 promoters respectively.It is worth to note that orp1 expression is mediated by the IHF2 site and not IHF1.This result is, at first, surprising because comparison of these sequences with the E. coli IHF binding consensus motif shows that the orp1 IHF2 site is less conserved than IHF1 (67% and 78% respectively, Figure S3) and the 59 of the IHF1 sequence is 100% identical to the reported consensus (WATCAR) whereas the IHF2 sequence is less conserved in this region (Figure S3).
An in vivo reconstitution approach in E. coli allowed us to identify DVU0396 (IHFa) and DVU1864 (IHFb) as a functionally active heterodimer able to fully complement the absence of the E. coli IHFa subunit.By contrast, DVU0396 alone is unable to complement for the absence of the E. coli IHFa subunit and to functionally interact with the E. coli IHFb contrarily to what has been described for P. putida IHFa [42].This could be explained by the difference between the similarities of E. coli IHFa and DvH DVU0396 (32%) or P. putida IHFa (86%).The fact that DVU0396 and DVU1864 assemble a functionally active heterodimer is in agreement with recent results demonstrating that DVU1864 interacts preferentially with DVU0396 in pull-down experiments [36].While expression of the orp operons is on the same order of magnitude in the presence of E. coli IHF and DvH DVU0396/ DVU1864 heterodimer, it will be interesting in the future to test DVU0396/DVU1864 binding on the orp promoters by EMSA.Experiments are currently underway to purify the DvH DVU0396/DVU1864 complex.While DVU0396 is active in E. coli, the absence of DVU0396 does not significantly impact the expression of the orp operon genes in DvH as shown by quantitative RT-PCR.One hypothesis to explain this discrepancy is the large number of HU and IHF proteins in DvH that might have redundant functions.By contrast to the E. coli situation, DvH appears to lack the diversity of nucleoid-associated proteins such as Fis or HNS but seems to encode several copies of HU and IHF proteins.DvH genome encodes up to eight DNA-binding HU/ IHF proteins belonging to the COG776 family.Among these, two are annotated as encoding IHFb-like proteins (DVU1864 and DVU2973, which are 65% identical) and one, DVU1134, is annotated as a true DNA-binding HUb protein.Whereas DVU1864 and DVU2973 contains a central region extremely well conserved shown to be essential for IHF activity and specificity in E. coli [42,43], this central region in DVU1134 is less conserved (Figure 8).Three other HU-like proteins (DVU0764, DVU1795, DVU3187) exhibit 25-31% sequence identity with DVU0396.All three proteins possess the G62 residue required for IHF activity and possess one additional residue P65 known to be crucial determinants of IHF specificity in E. coli [43].Such redundancy in genes encoding HU/IHF proteins has not been observed in other organisms studied to date.Therefore, although the DVU0396-DVU1864 combination is functionally active in E. coli one may hypothesize that distinct homodimers or heterodimers involving different combinations will be used in DvH as previously described in other organisms [21,22,23,24], and therefore the DVU0396 mutation will not be sufficient to observe an effect on orp gene expression in DvH.This hypothesis is strengthened by pull-down experiments demonstrating that DVU1864, while interacting preferentially with DVU0396, is also able to interact with DVU2973, while this later also interacts with DVU0396 and DVU0764 [37].Further experiments are therefore needed to establish the role of each HU/IHF proteins and identify all functional IHF complexes coexisting in DvH.It would be also interesting to know whether this atypical gene redundancy is linked to the large number of putative EBPs observed in DvH [6].
We recently proposed a model on the regulation of the orp operons in which the s 54 -transcriptional regulator DVU2106 interacts with specific palindromic sequences and collaborates with s 54 to activate and synchronize the transcription of the two divergent orp operons [6].DVU2106 is under the control of the housekeeping s 70 factor but the overlap between the s 70 promoter and the DVU2106-binding sequence creates a negative regulatory loop [6].With the data presented here, we implemented this model by adding a new player in the game, IHF (Figure 9).We showed that IHF is required for full activity of the orp1 and orp2 promoter.As previously suggested, since IHF facilitates DNA bending but does not influence directly the formation of the open complex, its role in s 54 -dependent systems has to be considered as a co-regulator [44].In this context, we demonstrated that IHF binds to specific sequences, closely related to the E. coli IHFbinding consensus motif, located between the orp1 and orp2 s 54promoters and DVU2106 palindromic sites, to co-regulate and orchestrate the simultaneous expression of the divergent orp operons.In contrast no effect of IHF was observed on the s 70dependent DVU2106 expression.

Figure 1 .
Figure 1.IHF is involved in the transcriptional activation of both operons orp1 and orp2.(A) Schematic representation of the transcriptional elements of promoter regions of orp1, DVU2106 and orp2.The positions of the s 54 and s 70 promoters are indicated by bent arrows and the DVU2106-binding sites are indicated by solid line boxes.The lacZ reporter fusions with the promoter regions of orp1 (B), DVU2106 (C) and orp2 (D) are represented on the left.The transcript start sites are indicated by bent arrows.The positions of the 210 and 235 sequences of the s 70 promoter and the 212 and 224 sequences of the s 54 promoters are indicated by black rectangles.The activities measured in various backgrounds are shown on the right: E.coli wildtype strain (dark-grey) and E.coli DihfA strain (light-grey).The activity is the average of three independent measurements (the error bars show the standard deviations).doi:10.1371/journal.pone.0086507.g001

Figure 3 .
Figure 3. Sequence analyses of the orp1 and orp2 promoter regions.(A) The s 54 and s 70 promoters of the ORP system are indicated by bent arrows.The solid-line boxes indicate the palindromic DVU2106-binding sites.The dashed boxes represent the three putative IHF-binding sequences identified by comparison to the consensus sequence of the E.coli IHF-binding site (59-WATCARxxxxTTR-39).(B) DNA sequence alignment between the consensus sequence of E.coli IHFbinding site and each putative IHF-binding sequence of orp1 and orp2 promoter regions.The sequence identity between the consensus sequence of the E.coli IHF-binding site and the different orp putative IHF-binding site are indicated in bold.doi:10.1371/journal.pone.0086507.g003

Figure 4 .
Figure 4. orp1 IHF-2 and orp2 IHF sites are involved in transcriptional activation of orp1 and orp2, respectively.The lacZ reporter fusions with the promoter regions of (A) orp1 IHF1-mut, mutated for the putative IHF-binding site 1, (B) orp1 IHF2-mut, mutated for the putative IHF-binding site 2, (C) orp1 IHF-mut, mutated for both IHF-binding sites, (D) DVU2106 IHF-mut, mutated for putative IHFbinding site of orp2 promoter and (E) orp2 IHF-mut, mutated for the putative IHF-binding site are represented on the left.The nature of these mutations is described in materials and methods and in figure S4.Promoters s 54 and s 70 are indicated by bent arrows.The putative IHFbinding sites are indicated by boxes: solid-line boxes for wild-type sites and dashed boxes for mutated sites.The activities measured in various backgrounds are shown on the right: E.coli wild-type strain (dark-gray) and E.coli DihfA strain (light-gray).The measures shown in black and indicated by (+) correspond to the b-galactosidase activity observed in a wild-type E.coli strain carrying the fusion between wild-type promoters and lacZ.The activity is the average of three independent measurements (the error bars show the standard deviations).doi:10.1371/journal.pone.0086507.g004

Figure 6 .
Figure 6.Protein sequence alignment of IHFb and IHFa subunits of E.coli and DvH.(A) The figure shows the sequence of IHFb from E. coli aligned with the sequences of the two putative IHFb from DvH, DVU1864 and DVU2973.(B) The sequence of IHFa from E.coli was aligned with the sequences of the two putative IHFa, DVU0396 and DVUA004 from DvH using ClustalW (http://www.ebi.ac.uk).Highlighted residues correspond to crucial amino-acids for IHF function in E. coli.doi:10.1371/journal.pone.0086507.g006

Figure 7 .
Figure 7.The heterodimer DVU0396-DVU1864 from DvH is involved in the transcriptional activation of orp operons.The lacZ reporter fusions with the promoter regions of orp1 (A) and orp2 (B) are represented on the left.The positions of the 212 and 224 sequences of the s 54 promoters are indicated by black rectangles.The transcript start sites are indicated by bent arrows.The activities measured in various backgrounds are shown on the right: E.coli wildtype strain (black), E.coli DihfA strain (dark-grey), E.coli DihfA strain producing DVU0396 (middle-grey) and E.coli DihfA strain producing DVU0396-DVU1864 (light-grey).The activity is the average of three independent measurements (the error bars show the standard deviations).doi:10.1371/journal.pone.0086507.g007

Figure 8 .Figure 9 .
Figure 8.Protein sequence alignment of putative IHFb and IHFa subunits from DvH.The figure shows the sequence of DVU1864 aligned with the sequences of the two putative IHFb from DvH, DVU1864 and DVU2973 and the sequence of DVU0396 (IHFa) aligned with the sequences of three putative IHFa from DvH using ClustalW (http://www.ebi.ac.uk).Highlighted residues correspond to crucial amino-acids for IHF function in E. coli.doi:10.1371/journal.pone.0086507.g008

Figure
Figure S1 Sequences of orp1 (A) and orp2 (B) promoter regions.The indentified s 54 promoters of orp1 and orp2 are boxed in solid lines.The transcriptional start point of both operons is indicated by bent arrows.The stop codon and the start codon are in italics and underlined, and the shine-Dalgarno are indicated in bold.The two palindromic binding sites of the s 54 transcription activator, DVU2106, are boxed in dotted lines.The sequences of the putative IHF-binding sites are in bold and underlined: the orp1 promoter contains two putative IHF-binding sites and one putative IHF-binding sequence was found in the orp2 promoter.(PDF) Figure S2 Sequences of the different IHF sites and its variants.Mutated bases are indicated in bold for each mutant variant.These sequences are aligned with the IHF-binding consensus of E.coli.(PDF) Figure S3 Sequence alignment of each DvH IHF-binding site with IHF-binding site consensus sequences of E.coli (A) and P.putida (B).(PDF) Table S1 Bacterial strains and plasmids used in this study.