Identification of the Actinobacillus pleuropneumoniae Leucine-Responsive Regulatory Protein and Its Involvement in the Regulation of In Vivo-Induced Genes

Abstract

Actinobacillus pleuropneumoniae is a gram-negative bacterial pathogen that causes a severe hemorrhagic pneumonia in swine. We have previously shown that the limitation of branched-chain amino acids (BCAAs) is a cue that induces the expression of a subset of A. pleuropneumoniae genes identified as specifically induced during infection of the natural host animal by using an in vivo expression technology screen. Leucine-responsive regulatory protein (Lrp) is a global regulator and has been shown in Escherichia coli to regulate many genes, including genes involved in BCAA biosynthesis. We hypothesized that A. pleuropneumoniae contains a regulator similar to Lrp and that this protein is involved in the regulation of a subset of genes important during infection and recently shown to have increased expression in the absence of BCAAs. We report the identification of an A. pleuropneumoniae serotype 1 gene encoding a protein with similarity to amino acid sequence and functional domains of other reported Lrp proteins. We further show that purified A. pleuropneumoniae His₆-Lrp binds in vitro to the A. pleuropneumoniae promoter regions for ilvI, antisense cps1AB, lrp, and nqr. A genetically defined A. pleuropneumoniae lrp mutant was constructed using an allelic replacement and sucrose counterselection method. Analysis of expression from the ilvI and antisense cps1AB promoters in wild-type, lrp mutant, and complemented lrp mutant strains indicated that Lrp is required for induction of expression of ilvI under BCAA limitation.

Actinobacillus pleuropneumoniae is a bacterial pathogen that causes both acute and chronic forms of necrotizing hemorrhagic pleuropneumonia in swine (16, 28, 42, 51). The severe economic effect of this disease on the swine industry has been ameliorated by improvements in detection and prevention of the disease and in management practices. However, the methods by which A. pleuropneumoniae infects and causes disease in swine are still not fully understood. While a variety of virulence factors have been reported to contribute to the pathogenesis of A. pleuropneumoniae (1, 3, 4, 9, 13, 39, 48, 56, 62, 63), little is known about what signals induce expression of these virulence factors during infection. Certain environmental cues, such as iron limitation, heat shock, oxidative stress, and osmotic stress, have been shown to play a part in the regulation of virulence genes in other organisms. We have recently shown that the limitation of branched-chain amino acids (BCAAs), which include leucine, isoleucine, and valine, is an additional cue that induces in vitro the expression of a subset of A. pleuropneumoniae promoters that were previously identified by in vivo expression technology as in vivo induced (ivi) (19, 59). However, the mechanism or mechanisms by which these ivi genes are regulated in response to BCAA limitation in A. pleuropneumoniae is unknown.

The study of gene regulators in A. pleuropneumoniae has been limited and has yielded the identification of only two to date. These include HlyX (33, 35), a homologue of the Escherichia coli global regulator FNR, and the ferric uptake regulator protein Fur (27). To elucidate how genes are regulated in response to BCAA limitation, a better understanding of potential regulators in A. pleuropneumoniae is needed.

One mechanism known to regulate genes in response to BCAA limitation is the leucine-responsive regulatory protein (Lrp). Lrp was first identified in E. coli as the positive regulator of ilvI (44, 46), a gene whose protein product is involved in BCAA biosynthesis. Other genes, both activated and repressed by Lrp, have been subsequently identified (reviewed in references 6, 7, 14, 40, and 41). A DNA microarray study by Tani et al. showed Lrp to be involved in the regulation of up to 10% of all E. coli genes either directly or indirectly (54). In general, Lrp positively regulates genes involved in biosynthesis of amino acids and negatively regulates genes involved in catabolism of amino acids in E. coli. However, Lrp has been shown to regulate, either directly or indirectly, genes associated with virulence, such as those for fimbriae in E. coli (21, 24, 57, 65, 66) and the hpmBA hemolysin operon of Proteus mirabilis (17). Recently, Lrp was shown to positively regulate the XhlA hemolysin of Xenorhabdus nematophila (10), which is required for virulence in insects.

Genes either directly or indirectly regulated by Lrp may respond to Lrp differently depending upon availability of BCAAs in the environment. Lrp can be a positive or negative regulator, with leucine antagonizing the effect of Lrp, potentiating the effect of Lrp, or having no effect on Lrp (34, 54). For example, Lrp positively regulates the E. coli ilvI gene in the absence of leucine, but the effect is antagonized in the presence of leucine (46). In contrast, the livJ gene, involved in BCAA transport, is repressed by both Lrp and leucine together, but repression is not achieved by either individually (34).

The presence of Lrp and its role in gene expression in A. pleuropneumoniae have not been investigated. We hypothesized that A. pleuropneumoniae contains an Lrp homologue and that this protein is involved in the regulation of a subset of genes expressed during infection and recently shown to have increased in vitro expression in the absence of BCAAs (59). In this study, we have identified an A. pleuropneumoniae serotype 1 gene with similarity to the lrp gene of E. coli. The A. pleuropneumoniae serotype 1 lrp gene was cloned, sequenced, and expressed in a protein expression vector, and hexahistidine (His₆)-tagged protein was purified. We report that A. pleuropneumoniae His₆-Lrp binds to two in vivo-induced promoters, the iviI promoter, which has been identified as the promoter for the ilvIH operon, and the iviG promoter, which expresses a transcript antisense to the A. pleuropneumoniae cps1AB capsule biosynthetic genes, as well as to the nqr promoter and to its own lrp promoter. Furthermore, we report the construction and confirmation of an A. pleuropneumoniae lrp mutant and show through complementation assays that A. pleuropneumoniae Lrp regulates expression of ilvI in A. pleuropneumoniae.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

The bacterial strains and plasmids used in this study are listed in Table 1. A. pleuropneumoniae strains were cultured in brain heart infusion (BHI) (Difco Laboratories, Detroit, MI) or chemically defined medium (CDM) (59) and incubated either at 35°C with 5% CO₂ for agar media or at 35°C and 150 rpm for broth media. For growth rate experiments, heart infusion (HI) broth (Difco) was also used. Media were supplemented with NAD (also designated V factor) (Sigma Chemical Company, St. Louis, MO) to a final concentration of 10 μg/ml and with riboflavin as needed (Sigma) to a final concentration of 200 μg/ml. Ampicillin and kanamycin, when required, were added to 50 μg/ml for plasmid selection in A. pleuropneumoniae. When investigating the response of A. pleuropneumoniae to the limitation of BCAAs, the amino acids isoleucine, leucine, and valine were excluded (CDM−ILV) from the complete CDM (CDM+ILV). For analysis of gene expression in CDM broth, bacterial strains grown for 18 h on BHI agar were inoculated into 5 ml of CDM broth to an optical density at 520 nm (OD₅₂₀) of ∼0.1.

TABLE 1.

Bacterial strains and plasmids used in this study

Strain or plasmid	Characteristic(s)	Reference or origin
Strains
A. pleuropneumoniae
AP100	ATCC 27088, serotype 1A, passaged through pigs	ATCCa
AP225	Spontaneous nalidixic acid-resistant mutant of AP100	20
APTW405	lrp single-crossover mutant of AP225	This work
AP359	lrp double-crossover mutant of AP225	This work
E. coli
CV975	F−ara thi Δ(lac-pro) ilvIH::Mu dI1734	44
XL1-Blue mRF′	Δ(mcrA)183 Δ(mcrCB-hsdSMR-mrr)173 endA1 supE44 thi-1 recA1 gyrA96 relA1 lac [F′ proAB lacIqZΔM15 Tn10 (Tetr)]	Stratagene
S17-1(λpir)	λpir recA thi pro hsd (rK−mK+) RP4-2-Tc::Mu Km::Tn7 Tmpr Smr	53
Plasmidsb
pUC19	Apr; high-copy-number cloning vector	58
pTF86	Apr; A. pleuropneumoniae IVET vector containing promoterless luxAB and ribBAH genes downstream of a unique BamHI cloning site	19
pQE30	Apr; protein expression vector	QIAGEN
pCV294	Apr; E. coli (His6-Lrp) protein expression vector	36
pER187	Apr Cmr; CAT cassette-containing vector	49
pUM24Cm	Cmr Kanr; sacR-sacB-nptI cassette-containing vector	47
pGP704	Apr; broad-host-range suicide vector	37
pGZRS39	Kanr; A. pleuropneumoniae shuttle vector	64
pKB11	Apr; A. pleuropneumoniae nqr promoter region cloned into pKB10, a derivative of pTF86 with the ribBAH genes deleted	K. Bachus
pTW296	Apr; E. coli ilvIH promoter (amplified by PCR and cloned into BamHI site of pUC19)	This work
pTW313	Apr; A. pleuropneumoniae lrp (amplified by PCR and cloned into pQE30)	This work
pTW328	Apr; E. coli ilvIH promoter (constructed by cloning the ∼300-bp BamHI fragment from pTW296 into the BamHI site of pTF86)	This work
pTW338	Apr; rnd′-lrp-ftsK′ (amplified by PCR and cloned into SphI/SalI sites of pUC19)	This work
pTW355	Apr; rnd′-Δlrp-ftsK′ (constructed by inverse PCR with pTW338 as template)	This work
pTW401	Apr Cmr; rnd′-Δlrp::CAT-ftsK′ (constructed by cloning a PCR-generated CAT cassette into the PstI site of pTW355)	This work
pTW402	Apr Cmr Kanr; rnd′-Δlrp::CAT-ftsK′-sacR-sacB-nptI (constructed by cloning the 3.4-kb BamHI sacR-sacB-nptI fragment into partially BamHI-digested pTW401)	This work
pTW404	Apr Cmr Kanr; rnd′-Δlrp::CAT-ftsK′-sacR-sacB-nptI (constructed by cloning the SphI/SacI insert of pTW402 into pGP704)	This work
pTW415	Kanr; rnd′-lrp-ftsK′ (constructed by cloning the SphI/SacI insertion of pTW338 into pGZRS39)	This work
pIviA	Apr; pTF86 containing a 333-bp insert that is homologous to the mrp gene	19
pIviG	Apr; pTF86 containing a 211-bp insert that is homologous to, but in an antisense orientation to, the cps1AB genes (involved in capsule biosynthesis)	19
pIviI	Apr; pTF86 containing a 623-bp insert that is homologous to the ilvI gene (required for branched chain amino acid biosynthesis)	19
pIviP	Apr; pTF86 containing a 175-bp insert that is homologous to the comJ promoter	30
pIviS	Apr; pTF86 containing a 352-bp insert that is homologous to the hfq promoter	30
pIviU	Apr; pTF86 containing a 604-bp insert that is homologous to mioC	30
pIviX	Apr; pTF86 containing a 490-bp insert that is homologous to the vapB vapC promoter	30
pIviY	Apr; pTF86 containing a 782-bp insert that is homologous to ftsY, which encodes the signal recognition particle receptor	30
pIvi17g	Apr; pTF86 containing a 290-bp insert that is homologous to, but in an antisense orientation to, the coaE gene	30

^aATCC, American Type Culture Collection.

^bpIvi plasmids contain in vivo-induced promoter fragments fused to promoterless luxAB genes of pTF86.

E. coli XL1-Blue mRF′ (Stratagene, La Jolla, CA) and E. coli S17-1(λpir) (53) were cultured in Luria-Bertani (LB) medium and used for cloning and mating, respectively. Ampicillin and kanamycin, when required, were added to 100 μg/ml and chloramphenicol was added to 10 μg/ml for plasmid selection in E. coli.

The pIvi plasmids listed in Table 1 were constructed by cloning Sau3A-digested A. pleuropneumoniae serotype 1 genomic DNA fragments into the promoter trap in vivo expression technology (IVET) plasmid pTF86 (19) and were identified as containing promoters that were induced during infection in a swine animal model (19, 30). pTF86 contains promoterless luciferase genes (luxAB), promoterless riboflavin genes (ribBAH), and a unique BamHI cloning site. When a functional promoter is cloned into pTF86 in the proper orientation, both the luciferase and riboflavin genes are expressed when the promoter is active. Expression of the riboflavin genes complements an attenuating mutation in the host A. pleuropneumoniae strain used for in vivo expression studies and restores virulence. Expression of luciferase activity from the promoter:Lux fusions can be used to measure the promoter activity.

Molecular manipulations.

Genomic DNA from A. pleuropneumoniae was isolated using a QIAGEN-tip 500 as described by QIAGEN (44a).Plasmid DNA was purified using QIAprep spin columns (QIAGEN). DNA-modifying enzymes were obtained from Roche (Roche Applied Science, Indianapolis, IN) and New England Biolabs (New England Biolabs, Inc., Beverly, MA) and used according to the respective manufacturer's specifications. Electrocompetent AP225 was prepared and electroporated as previously described (20). E. coli XL1-Blue mRF′ was electroporated using the same conditions as those for A. pleuropneumoniae.

Luciferase assays.

For quantitative measurement of luciferase activity, a Turner model 20e luminometer (Turner Designs, Sunnyvale, CA) was utilized as previously described (19). Briefly, 20 μl of broth culture was added to 20 μl of luciferase substrate and mixed for 10 s. The substrate was made by dissolving 20 mg/ml Essentially Fatty Acid Free bovine serum albumin (BSA) (Sigma) and 1 μl of N-decyl aldehyde in 1 ml of H₂O and sonicating the solution. The luminometer was set to a delay of 10 s, an integration of 30 s, and a sensitivity of 39.9%. The luminometer relative light unit (RLU) readings were normalized to the OD₅₂₀ units of the culture.

Induction, purification, and quantification of A. pleuropneumoniae and E. coli His₆-Lrp.

The A. pleuropneumoniae lrp gene was amplified by PCR using AP100 genomic DNA, Pfu Turbo DNA polymerase (Stratagene), and A. pleuropneumoniae lrp-specific primers MM379-SalI and MM430-BamHI and ligated in frame into pQE30 (QIAGEN) to generate pTW313.

E. coli XL1-Blue mRF′/pTW313 and XL1-Blue mRF′/pCV294, the E. coli His₆-Lrp protein expression vector, were grown in LB medium at 35°C and 150 rpm for 4.5 h. One millimolar isopropyl-β-d-thiogalactoside (IPTG) was added to each culture and incubated for an additional 3 h to an OD₆₀₀ of 0.4, at which time cell pellets were harvested and frozen. Frozen pellets were resuspended in 4 ml ice-cold binding buffer (5 mM imidazole, 0.5 M NaCl, 20 mM Tris-HCl [pH 7.9]). Samples were sonicated for 1 min until not viscous and then centrifuged at 14,000 × g for 20 min at 4°C to remove cellular debris. His-tagged proteins were purified using Novagen His-Bind Quick 900 cartridges (EMD Biosciences, Inc., Madison, WI) according to the manufacturer's instructions. The cartridges were washed with 20 ml of binding buffer and 10 ml of wash buffer (60 mM imidazole, 0.5 M NaCl, 20 mM Tris-HCl [pH 7.9]) and eluted with 4 ml of elution buffer (1 M imidazole, 0.5 M NaCl, 20 mM Tris-HCl [pH 7.9]). Eluted proteins were dialyzed into 20 mM Tris-HCl (pH 8.0)-0.2 mM EDTA-0.2 mM dithiothreitol-0.4 M NaCl by using a Centricon 10 (Millipore Co., Bedford, MA), and an equal volume of 100% ultrapure glycerol was added to each sample. Bio-Rad protein microassays were performed using BSA as a standard to determine final protein concentrations. The purified E. coli His₆-Lrp and A. pleuropneumoniae His₆-Lrp protein samples were stored at −80°C until use.

SDS-PAGE.

Protein samples were resuspended in an equal volume of 2× sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer (125 mM Tris-HCl [pH 6.8], 25% glycerol, 2.5% SDS, 2.5% β-mercaptoethanol, 1.25% bromophenol blue) and loaded on a 12% polyacrylamide gel, as described by Laemmli (31). Gels were stained with Coomassie blue (22).

EMSA.

A. pleuropneumoniae DNA fragments for electrophoretic gel mobility shift assays (EMSA) were isolated by PCR using AP100 genomic DNA and gene-specific primers, or vector-specific primers in the case of ivi fragments. The 623-bp genomic fragment in pIviI was cloned into pUC19 to generate the iviI PCR template, pTW286. Primers specific to the pUC19 vector, MM531 and MM532, were used in this case to generate the iviI fragment. For all other ivi fragments, primers specific to pTF86, MM478-lux and MM533-T4, were used.

For an internal control, an E. coli ilvIH promoter DNA fragment was PCR amplified from pTW328 by using primers MM478-lux and MM533-T4. To generate pTW328, the ilvIH promoter was PCR amplified from a colony of CV975 by using primers MM362-BamHI and MM363-BamHI. The PCR product was digested with BamHI and ligated to BamHI-digested pUC19 to generate pTW296. The ∼300-bp BamHI fragment from pTW296 was ligated to BamHI-digested pTF86 to generate pTW328.

PCR products were gel extracted and purified using the QIAEX II system (QIAGEN). For radiolabeling of DNA fragments, 1 pmol of purified DNA fragment was combined with 10 units T4 polynucleotide kinase (Roche), 50 μCi [γ-³²P]ATP (Amersham Biosciences, Piscataway, NJ), and 1× polynucleotide kinase buffer in a final volume of 20 μl and incubated at 37°C for 1 h. Completed reactions were inactivated by heating to 68°C for 10 min and cleaned by centrifuging the reaction volume through a Quick Spin column for radiolabeled DNA purification (Roche). Purified His₆-Lrp was diluted to 5 ng/μl in binding buffer [20 mM Tris-HCl (pH 8.0), 75 mM NaCl, 5 mM MgCl₂, 1 mM dithiothreitol, 12.5% glycerol, 0.1 mg/ml BSA, 25 μg/ml poly(dI-dC)]. Binding reaction mixtures containing 0, 5, 30, or 60 ng of His₆-Lrp were incubated with 0.05 pmol of radiolabeled DNA fragment and 0.5 μg poly(dI-dC) brought to a final volume of 20 μl with binding buffer (8). Binding reaction mixtures were incubated at room temperature for 20 min, and then reactions were stopped by adding 5 μl of STOP solution (USB Co., Cleveland, OH). The entire reaction volume was loaded onto a 5% nondenaturing polyacrylamide gel prepared in 1× Tris-borate-EDTA (pH 8.0) and electrophoresed at 200 V for 2 to 3 h. Gels were dried at 80°C for 40 min and exposed to Amersham Biosciences Hyperfilm MP film. For quantitative analysis, gels were scanned on a Storm 860 PhosphorImager (Molecular Dynamics, Sunnyvale, CA), and images were analyzed using ImageQuant-TL v2005 software from Amersham Biosciences (Piscataway, NJ).

Primer extension analysis.

Total RNA was isolated from the cell pellets of cultures of AP225/pTW338 and AP225/pKB11 grown in CDM−ILV to an OD₅₂₀ of 0.8, using TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's protocol. Primer extension analysis was performed as previously described (59) to identify the transcriptional start points for lrp and nqr, using gene-specific primers. Primer extension analysis of the iviI and iviG promoters was previously published (59).

Construction of an A. pleuropneumoniae lrp mutant.

Primers MM459-PstI and MM460-PstI, both designed with internal PstI restriction sites, were used with an inverse PCR technique to amplify around the pTW338 construct in opposite directions. The 4.5-kb linear product from the inverse PCR was digested with PstI and ligated to itself to generate the reconstituted plasmid, pTW355. The annealing position of the primers during the inverse PCR resulted in a ∼100-bp deletion from the center of lrp. An 0.9-kb PstI-digested chloramphenicol acetyltransferase (CAT) resistance cassette, from a PCR using pER187 as the template and MM489-PstI and MM511-PstI as primers, was ligated to the newly generated PstI site of pTW355 to generate pTW401. The 3.4-kb sacR-sacB-nptI BamHI fragment from pUM24Cm (47) was ligated to partially BamHI-digested pTW401 to generate pTW402. The sacR and sacB genes confer sucrose sensitivity in the presence of sucrose and allow for future selection of double-crossover events. The nptI gene confers resistance to kanamycin. The SphI/SacI insert from pTW402 was ligated to similarly digested pGP704 (37) to generate the knockout construct pTW404. The pTW404 construct was electroporated into E. coli S17-1(λpir), and this strain was filter mated with AP225, a nalidixic acid-resistant derivative of AP100, according to the protocol of Mulks and Buysse (38). Transconjugants were isolated on BHI agar supplemented with V factor (BHIV agar) and containing 2 μg/ml chloramphenicol and 50 μg/ml nalidixic acid after 48 h and screened by PCR for single- or double-crossover events at the lrp locus. A single-crossover transconjugant, designated APTW405, was selected and grown overnight at 35°C and 5% CO₂ on BHIV agar medium supplemented with 2 μg/ml chloramphenicol. The following day, the single-crossover mutant was inoculated into 1 ml of BHIV supplemented with 5 μg/ml chloramphenicol and grown at 37°C and 220 rpm for 2 h until slightly turbid. At this point, 1 ml of BHIV broth medium supplemented with 20% sucrose and 10 μg/ml chloramphenicol was added to the single-crossover mutant culture to achieve final concentrations of 10% sucrose and 7.5 μg/ml chloramphenicol. This culture was incubated at 37°C and 220 rpm for 5 h to select for chloramphenicol resistance and sucrose insensitivity. Dilutions of the chloramphenicol selection/sucrose counterselection culture were plated on BHIV agar supplemented with 5 μg/ml chloramphenicol and 10% sucrose and incubated overnight at 35°C and 5% CO₂. This chloramphenicol selection/sucrose counterselection on APTW405 resulted in chloramphenicol-resistant and sucrose-insensitive bacteria at a density of 6.7 × 10⁷ CFU/ml, suggesting that the single-crossover event in APTW405 was forced into a double-crossover event to generate an lrp mutant.

Southern blot analysis.

Chromosomal DNA and plasmid controls were digested with the restriction enzyme EcoRI, and the DNA fragments were separated on an 0.8% ultrapure agarose gel in Tris-acetate-EDTA buffer. Southern blotting was performed as described by Sambrook et al. (50). DNA probes were labeled with digoxigenin (DIG) by using either the PCR DIG probe synthesis or the DIG DNA labeling kit (Roche Applied Science, Indianapolis, IN). Probes included an 0.5-kb lrp PCR fragment, an 0.8-kb CAT cassette fragment, and a 3.7-kb pGP704 (37) fragment. The lrp fragment was generated by PCR using the MM430-BamHI and MM379-SalI primers with AP100 genomic DNA as a template. The CAT cassette fragment was generated by PCR using the MM508 and MM509 primers with pER187 (49) as a template. The pGP704 fragment was generated by digesting the plasmid with BglII. Hybridizations, washes, and developing were performed as described by Fuller et al. (20). Hybridizations were carried out at 42°C for 18 h in 50% formamide, 2% blocking solution (Roche), 5× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate), 0.1% Sarkosyl detergent, and 0.02% SDS. Blots were washed three times in 2× SSC-0.1% SDS for 15 min at room temperature and twice in 0.1× SSC-0.1% SDS for 60 min at 68°C. Blots were developed with an alkaline phosphatase-conjugated antidigoxigenin and CDP-Star substrate kit (Roche) according to the manufacturer's instructions.

Complementation induction assays.

The ∼1.9-kb SphI/SacI restriction digest fragment from pTW338 was ligated to similarly digested pGZRS39 to generate the lrp complementation plasmid, pTW415. Wild-type (AP225/pGZRS39), lrp mutant (AP359/pGZRS39), and complemented mutant (AP359/pTW415) strains containing pIviI or pIviG as a second plasmid were grown overnight on BHIV agar supplemented with 50 μg/ml kanamycin and 50 μg/ml ampicillin for maintenance of both plasmids. The pIviI and pIviG plasmids contain previously identified (19) in vivo-induced promoters fused to luciferase. A sterile cotton-tipped swab was used to resuspend each bacterial strain in 1.2 ml of CDM−ILV broth medium, and 100 μl was used to inoculate 5-ml cultures of CDM-−ILV and CDM+ILV broth medium for each strain. Cultures were grown at 37°C for 8 h at 220 rpm, with samples taken every 1 to 2 h and analyzed for luciferase expression by quantitative luciferase assays. Assays were performed in triplicate at each time point in each experiment, and experiments were repeated a minimum of three times.

Nucleotide sequence accession numbers.

A sequence of the A. pleuropneumoniae lrp gene has been deposited in the GenBank database under accession number DQ370064. Sequences of the inserts in the pIvi plasmids used in this study have been previously submitted to GenBank under the following accession numbers: clone iviA, DQ370062; iviG, DQ370063; iviI, DQ370055; iviP, DQ370061; iviS, DQ370056; iviU, DQ370060; iviX, DQ370059; iviY, DQ370058; ivi17b, DQ667682; and ivi17g, DQ370057. Sequence data for the nqr promoter have been submitted to GenBank as an amendment to accession number U24492, which contains the sequence for the A. pleuropneumoniae nqr (previously designated aopA) gene (11).

RESULTS

Cloning of A. pleuropneumoniae lrp.

We have previously identified the A. pleuropneumoniae ilvI gene as an in vivo-induced (ivi) gene (19) and have shown that expression of a luciferase reporter is induced from the ilvI promoter under BCAA-limiting conditions (59). ilvI has been extensively studied in E. coli and shown to be positively regulated by Lrp. To investigate the role of Lrp in the regulation of ilvI and other A. pleuropneumoniae ivi genes, we identified a gene with similarity to lrp in the unfinished A. pleuropneumoniae serotype 5 genome by homology to known lrp genes. Primers MM450-SphI and MM451-SalI were designed from the A. pleuropneumoniae serotype 5 sequence and used to clone a ∼1.9-kb region from A. pleuropneumoniae serotype 1 into pUC19 (58), which included the 3′ end of the upstream rnd gene, the complete lrp gene, and the 5′ end of the downstream ftsK gene, to generate pTW338. The insert from pTW338 was sequenced, and the translated lrp sequence from pTW338 was aligned with Lrp sequences from eight different bacterial species, including four members of the family Pasteurellaceae, to which A. pleuropneumoniae belongs (Fig. 1). The translated protein sequence of A. pleuropneumoniae is 71% identical to that of E. coli Lrp (Fig. 1). The alignment showed an overall amino acid sequence conservation, including within the domains for DNA binding, transcriptional activation, and leucine response identified in Lrp from E. coli (43) and Pyrococcus furiosis (32). The conservation of the domains within A. pleuropneumoniae Lrp suggests that the domains may have functions similar to those characterized for E. coli Lrp.

FIG. 1.

Lrp domain and amino acid alignment. Lrp amino acid sequences from nine bacterial species were aligned with ClustalX (55) and shaded with Boxshade v3.31. Residues that are identical in the majority of species are shown on a black background, while residues that are functionally conserved are shown on a gray background. The locations of the Lrp functional domains for DNA binding (dashed line), transcriptional activation (solid line), and leucine response (dotted line) are indicated above the sequence (32, 43). The position of the amino acid important in the activation of transcription by Lrp in E. coli but unconserved in A. pleuropneumoniae is indicated by an asterisk. Ec, Escherichia coli (GenBank accession number BAA35614); Ka, Klebsiella aerogenes (AAD12584); Aa, Actinobacillus actinomycetemcomitans; Pm, Pasteurella multocida (AAK02338); Hi, Haemophilus influenzae (AAC23241); Hd, Haemophilus ducreyi (AAP96279); Ap, Actinobacillus pleuropneumoniae serotype 1; Pa, Pseudomonas aeruginosa (AAG08693); Lp, Legionella pneumophila (AAU27568). The A. actinomycetemcomitans sequence was identified by similarity in the University of Oklahoma unfinished A. actinomycetemcomitans genome and translated.

Purification of A. pleuropneumoniae and E. coli His₆-Lrp.

To obtain a better understanding of the function of A. pleuropneumoniae Lrp, both E. coli and A. pleuropneumoniae His₆-Lrp were purified. Induction of both A. pleuropneumoniae (Fig. 2A) and E. coli (data not shown) His₆-Lrp resulted in a dominantly expressed ∼23-kDa protein, as shown by SDS-PAGE. This was ∼4.7 kDa larger than the expected size of 18.3 kDa as predicted from the translated A. pleuropneumoniae lrp gene and ∼4.2 kDa larger than what has been reported for E. coli Lrp. The addition of the His₆ tag could account for this difference. Purification of A. pleuropneumoniae (Fig. 2B) and E. coli His₆-Lrp (data not shown) was analyzed by SDS-PAGE.

FIG. 2.

Induction and purification of Actinobacillus pleuropneumoniae serotype 1 His₆-Lrp. (A) SDS-PAGE of A. pleuropneumoniae serotype 1 His₆-Lrp protein induction samples. Lanes: M, 10 μl of LOW SDS-PAGE standard; 1, 0.4 ml culture equivalent of cell extract before induction; 2, 0.1 ml culture equivalent cell extract after induction; 3, 0.05 ml culture equivalent induced soluble fraction; 4, 0.05 ml culture equivalent induced insoluble fraction. Lysozyme, used in the lysis procedure, appears at 14 kDa in lanes 3 and 4. (B) SDS-PAGE of A. pleuropneumoniae serotype 1 His₆-Lrp purification samples. Lanes: 1, 35 μl culture equivalent of cell extract; 2, 35 μl culture equivalent of cell extract flowthrough; 3, 17.5 μl culture equivalent of first-wash flowthrough; 4, 35 μl culture equivalent of second-wash flowthrough; 5, 35 μl culture equivalent of protein elution; 6, 140 μl culture equivalent of concentrated His₆-Lrp; 7, 280 μl culture equivalent of concentrated His₆-Lrp; M, 10 μl of LOW SDS-PAGE standard.

His₆-Lrp binding to A. pleuropneumoniae promoters.

To investigate whether the A. pleuropneumoniae Lrp directly regulates expression of ivi genes, we analyzed whether purified A. pleuropneumoniae His₆-Lrp would bind in vitro to the purified DNA inserts of A. pleuropneumoniae ivi promoter clones. To confirm that the EMSA were functioning as designed, the E. coli ilvIH promoter fragment from pTW328 was analyzed by EMSA. The mobility of the E. coli ilvIH promoter fragment was retarded in an EMSA when either A. pleuropneumoniae or E. coli His₆-Lrp was present (data not shown). Furthermore, E. coli His₆-Lrp bound to the A. pleuropneumoniae iviI insert (data not shown).

We then analyzed eight ivi promoter clones that had previously been shown to be induced under BCAA limitation; these were iviG, iviI, iviP, iviS, iviU, iviX, iviY, and ivi17g (Table 1) (59). iviA was used as a negative control because the iviA clone did not respond to limitation of BCAAs (59) and the pIviA insert DNA sequence did not have any similarity to published Lrp consensus binding sites (12, 60). Binding of A. pleuropneumoniae His₆-Lrp to the iviG and iviI inserts, but not to the iviA control, was demonstrated by EMSA (Fig. 3A). The presence of A. pleuropneumoniae His₆-Lrp retarded the migration of both the 623-bp iviI and the 211-bp iviG fragments in a dose-dependent manner. The presence of two separate retarded bands with the iviI insert suggests that iviI has multiple A. pleuropneumoniae His₆-Lrp binding sites. In contrast, A. pleuropneumoniae His₆-Lrp did not bind to the iviP, iviS, iviU, iviX, iviY, and ivi17g fragments under these assay conditions (data not shown).

FIG. 3.

Analysis of A. pleuropneumoniae His₆-Lrp binding to ivi clone DNA inserts by electrophoretic mobility shift assays. The iviI, iviG, iviA, lrp, and nqr labeled DNA fragments were mixed with 0, 5, 30, or 60 ng of His₆-Lrp in a binding reaction. Gels from at least two replicate experiments were scanned using a Storm PhosphorImager and the quantitative data used to calculate the percentage of labeled DNA bound to protein.

The DNA sequences of the inserts to pIviI and pIviG have been identified (59). The pIviI insert sequence has similarity to the 5′ end of ilvI, a gene involved in the biosynthesis of BCAAs. The pIviG insert sequence has sequence similarity to, but is in an antisense orientation to, the cps1AB genes identified to be involved in the biosynthesis of A. pleuropneumoniae serotype 1 capsule.

To obtain a better understanding of the Lrp regulon of A. pleuropneumoniae, we extended this analysis to include the promoter regions of other genes. The A. pleuropneumoniae putative promoter regions of the Apx toxin genes apxI, apxII, and apxIV; the apf type IV fimbria cluster; the flp-rcp-tad locus for type IV fimbriae, involved in biofilm formation; the ilvG gene, involved in BCAA biosynthesis; the lrp gene; the nqr operon, encoding the Na⁺-translocating NADH:ubiquinone oxidoreductase; and the serA gene, involved in serine biosynthesis, were isolated by PCR using gene-specific primers (Table 2) and analyzed by EMSA. The migrations of the 778-bp nqr and 751-bp lrp promoter fragments were retarded when A. pleuropneumoniae His₆-Lrp was added to the reaction mixture (Fig. 3B). In contrast, A. pleuropneumoniae His₆-Lrp was not observed to bind to the apxI, apxII, apxIV, apf, flp-rcp-tad, ilvG, or serA promoter fragments under these assay conditions (data not shown).

TABLE 2.

Primers used in this study

Primer	Sequence (5′ to 3′)	Description
MM362-BamHI	GGCCATCAGTGGATCCAAGAGC	E. coli ilvIH promoter forward primer
MM363-BamHI	CTCGGATCCAGACAACATCTCC	E. coli ilvIH promoter reverse primer
MM379-SalI	TAATCGTTCAATCGTCGACTAACTCC	lrp downstream primer
MM430-BamHI	CTGGGAAGACGGATCCATGGAACATAAAAAAC	lrp upstream primer
MM437-SphI	CCCATCGGATCCGTTACATAAGCATGCTCGGC	nqr promoter forward primer
MM438-BamHI	GTGTACCTGCGATAGGTGGATCCAAGCC	nqr promoter reverse primer
MM450-SphI	GCAGCGGCATGCGTATGGTATTTGTTA	rnd internal forward primer
MM451-SalI	AAGCGGTCGACTTTCGTCATTTTGTTGC	ftsK internal reverse primer
MM459-PstI	AACTGCAGGAATGTCATTTGGTTTCGGGTGATTTCG	lrp inverse PCR forward primer
MM460-PstI	ATCTGCAGGCTTCAAGCAATTCCGGATTCAGTAATG	lrp inverse PCR reverse primer
MM478-lux	GCTGCCTCCATCCATGGGGTTCCTC	pTF86 lux primer
MM480	TGGGATACCGTGCATTACTGAA	lrp forward PCR screen primer
MM481	CGGGTAATCGCAGTAAAGTCG	lrp reverse PCR screen primer
MM489-PstI	ATGACCCTGCAGACGAATTCGAGCTCGG	pER187::CAT forward primer
MM508	CTAATGAAGAAAGCAGACAAGTA	CAT forward PCR screen primer
MM509	GGGGCAGGTTAGTGACATT	CAT reverse PCR screen primer
MM511-PstI	GTCGACTCTAGACTGCAGCCTCCGTC	pER187::CAT reverse primer
MM525-BamHI	AAAGGGGATCCGCCAGACGAAT	ilvG promoter forward primer
MM526-XbaI	ACCGACATCTAGACCGACATAAAGTA	ilvG promoter reverse primer
MM527-BamHI	GAAATGGATCCGCCTGATTTAGCAC	lrp promoter forward primer
MM528-XbaI	CGTTCAATTCTAGATTAACTCCTTATTTTA	lrp promoter reverse primer
MM529-BamHI	GGAACGGATCCTCAATAACATCGCC	serA promoter forward primer
MM530-XbaI	CGCATTCTAGAAATTTTTGGTTGAGGC	serA promoter reverse primer
MM531	GTTTTCCCAGTCACGACGTTGT	pUC19 forward primer
MM532	CACAGGAAACAGCTATGACCATG	pUC19 reverse primer
MM533-T4	CTGGGGGATGAGTGGCACACC	pTF86 T4 primer
MM534-BamHI	GCAATATCGGATCCGGCTAATGTAATC	flp-rcp-tad promoter forward primer
MM535-XbaI	TGGCAATGAGTCTAGAAATAAAGAGTGTC	flp-rcp-tad promoter reverse primer
MM536-BamHI	AGCGGATCCTAAAACGAGGCACTTCTTG	apf promoter forward primer
MM537-XbaI	CGATGCCGCCTCTAGATCCGAAAGC	apf promoter reverse primer
MM538-BamHI	CCGTAATTGGATCCAAAATACCGTGAAGCAG	apxI promoter forward primer
MM539-XbaI	AAGTCTAGACATCCGCAACTAGCGAGGCAAC	apxI promoter reverse primer
MM540-BamHI	TACGGATCCTTGGTACAAAAAATTTTACAG	apxII promoter forward primer
MM541-XbaI	CCTTTCTAGAATTGATCATTTAAAGATATTATTGTAG	apxII promoter reverse primer
MM544-BamHI	AACCAGGATCCCACAAAGAACACAAGC	apxIV promoter forward primer
MM545-XbaI	CACAAAACGTCTAGACCCCACCATAAAT	apxIV promoter reverse primer

^aRestriction sites are underlined.

Accurate calculation of binding affinities of Lrp for each of these genes will require more extensive EMSA with a greater range of Lrp concentrations. However, using the band intensities seen with iviI (Fig. 3A), as measured using a Storm PhosphorImager, we can calculate an estimated binding constant for the binding of Lrp to the ilvIH promoter encoded in the iviI insert to be 125 nM.

Identification of transcriptional start sites and analysis of putative promoter regions.

Primer extension analysis was used to determine the transcriptional start sites of iviI, iviG, lrp, and nqr. Figure 4 shows the DNA sequences of the regions upstream from the transcriptional start sites for these four promoters. The initiating nucleotide in all four transcriptional start sites identified was predicted to be T. The regions immediately upstream of the transcriptional start sites displayed a conserved −10 region with the sequence TATA(A/T)T as well as a conserved −35 region with the consensus sequence TAGACA, with spacings of 17 to 20 bp between the putative −35 and −10 regions.

FIG. 4.

Alignment of the promoter regions from A. pleuropneumoniae iviI, iviG, lrp, and nqr. The transcriptional start points identified by primer extension are shown in boldface and underlined. The putative −10 and −35 regions are shown in boldface and shaded, and the consensus sequences are shown at the bottom. Dashes indicate gaps to align the promoter regions. Sequences with at least 80% homology to the consensus binding site for E. coli Lrp are boxed with a solid line; sequences with 73 to 80% homology are boxed with a dashed line. Where overlapping potential binding sites were identified, the site with the best homology is marked.

Using the E. coli consensus binding site for Lrp of YAGHAWATTWTDCTR, where Y is C or T; H is C, A, or T; W is A or T; D is G, T, or A; and R is G or A, we identified potential binding sites for Lrp in all four of these promoter regions (Fig. 4). All four contained a potential Lrp binding site overlapping the −10 region. There were also potential binding sites overlapping or close to the −35 region in the iviI, iviG, and lrp promoters, as well as an additional site further upstream in iviI and lrp.

Construction and confirmation of an A. pleuropneumoniae lrp mutant.

To further investigate the role of Lrp in the regulation of ivi genes responding to limitation of BCAAs, an lrp mutant was constructed (Fig. 5). To confirm that the chloramphenicol selection/sucrose counterselection was successful in producing an lrp mutant, 20 potential lrp mutants were screened by PCR using lrp-specific primers MM480 and MM481. Fifteen of the 20 colonies displayed a single 1-kb product that corresponded to a mutant lrp allele. Five colonies displayed both the wild-type 200-bp and mutant 1-kb bands predicted for a single-crossover event (data not shown).

FIG. 5.

Construction of the knockout construct pTW404. An inverse PCR technique was performed to amplify around the lrp-containing plasmid pTW338 in opposite directions and generate a ∼100-bp deletion from the center of lrp to make pTW355. A CAT resistance cassette DNA fragment was generated by PCR and cloned into the PstI site of pTW355 to form pTW401. A BamHI fragment containing the sacR, sacB, and nptI genes was excised from pUM24Cm and ligated into a BamHI site in pTW401 to generate pTW402. The insert from pTW402 was cloned into the conjugative suicide vector pGP704 to form pTW404.

Seven of the 15 lrp mutants were further characterized and were confirmed as gram-negative coccobacilli that were NAD dependent, chloramphenicol resistant, sucrose insensitive, kanamycin sensitive, and nalidixic acid resistant. These characteristics were as predicted for a double-crossover lrp mutant. A single representative mutant, AP359, was selected for more thorough validation experiments.

A colony PCR on AP359 using MM480 and MM481 resulted in a 1-kb product, as predicted for an lrp mutant (Fig. 6B). In comparison, a predicted 200-bp product was observed for wild-type AP225, and a 1-kb product for E. coli S17-1 (λpir)/pTW404 (Fig. 6B). The PCR screen supported AP359 as an lrp mutant.

FIG. 6.

Confirmation of the A. pleuropneumoniae lrp mutant. (A) Genetic maps of the lrp locus in wild-type, pTW404, and predicted double-crossover mutant strains. The predicted genomic DNA EcoRI fragment sizes of the wild-type and double-crossover mutant strains are shown along with predicted EcoRI fragments for the knockout construct pTW404. E, EcoRI; X_A, site of genetic recombination site A; X_B, site of genetic recombination site B; X_AX_B, resulting double-crossover event at sites X_A and X_B. (B) A 2% agarose gel with PCRs using A. pleuropneumoniae lrp-specific primers MM480 and MM481. PCRs using the following DNA templates were loaded into each lane: AP359 lrp mutant genomic DNA (lane 1), AP100 wild-type genomic DNA (lane 2), knockout construct pTW404 (lane 3), no DNA (lane −), Invitrogen 1-kb DNA ladder (lane M). (C to E) Southern blots probed with lrp (C), the CAT cassette (D), and the pGP704 BglII fragment (E). Lanes: 1, AP100 wild-type genomic DNA; 2, AP359 lrp mutant genomic DNA; 3, knockout construct pTW404.

Southern blot analyses of genomic DNAs prepared from wild-type AP100 and the lrp mutant, AP359, are shown in Fig. 6C, D, and E. In AP100, the lrp probe hybridized to a 5.2-kb EcoRI fragment (Fig. 6C, lane 1), but there was no reaction with either the CAT (Fig. 6D, lane 1) or the pGP704 (Fig. 6E, lane 1) probe. In the lrp mutant, the lrp probe hybridized to 3.4-kb and 2.6-kb EcoRI fragments (Fig. 6C, lane 2), while the CAT probe hybridized to a 3.4-kb EcoRI fragment (Fig. 6D, lane 2); there was no reaction with the pGP704 probe (Fig. 6E, lane 2). The hybridization pattern seen with AP359 genomic DNA is the pattern predicted for transconjugants where the wild-type lrp locus has been replaced by the mutated lrp::CAT locus by a double-crossover event (Fig. 6A). All three probes bound to EcoRI-digested pTW404 (Fig. 6C, D, and E, lanes 3). These data confirm AP359 as an A. pleuropneumoniae lrp deletion disruption mutant.

Growth of wild-type, lrp mutant, and complemented mutant strains in CDM+ILV and CDM−ILV.

Exponential growth rates of wild-type A. pleuropneumoniae, the lrp mutant, and a complemented mutant constructed by cloning pTW415 into the lrp mutant were compared in both complex and chemically defined growth media (Table 3). Specific growth rates for wild-type and lrp mutant A. pleuropneumoniae were similar in complex media and similar although much lower in complete chemically defined medium. However, while the growth rate for the wild type was further reduced in CDM−ILV, there was no detectable growth of the lrp mutant in this medium. The ability to grow in CDM−ILV was restored in the complemented mutant. In addition to an inability to grow in CDM−ILV, the lrp mutant displayed a slightly longer lag time than wild-type A. pleuropneumoniae in all growth media (data not shown).

TABLE 3.

Specific growth rates of wild-type, lrp mutant, and complemented mutant strains in various growth media

Medium	Specific growth rate (h−1)a
	AP225 (wild type)	AP359 (lrp mutant)	AP359/pTW415 (complemented mutant)
BHIV	0.83 ± 0.05	0.81 ± 0.08	NDb
HIV	0.83 ± 0.02	0.90 ± 0.05	ND
CDM+ILV	0.36 ± 0.05	0.39 ± 0.09	0.39 ± 0.12
CDM−ILV	0.26 ± 0.03	No growth	0.26 ± 0.02

^aData presented are the averages and standard deviations from at least three separate growth curves. Specific growth rate was calculated as ln2/T_d, where T_d is the doubling time during exponential growth.

^bND, not done.

Luciferase expression from iviI and iviG in wild-type, lrp mutant, and complemented lrp mutant strains.

To examine the effect of the loss of Lrp on expression from putative Lrp-regulated ivi promoters, luciferase activity expressed from the promoters in pIviI and pIviG, which contain promoter:luxAB fusions, in each strain was compared in CDM+ILV and CDM−ILV broth media (Fig. 7A and B). The luciferase activity expressed from the iviI promoter showed a basal level of expression of ∼500 RLU/OD unit in CDM+ILV in either a wild-type background (AP225/pGZRS39/pIviI), the lrp mutant background (AP359/pGZRS39/pIviI), or the complemented mutant background (AP359/pTW415/pIviI), with no significant increase over the course of the experiment. However, luciferase activity in the wild-type background increased on average over 7-fold in CDM−ILV compared to CDM+ILV in 1 hour and over 10-fold in 2 hours. In sharp contrast, when the iviI promoter was placed in an lrp mutant background, no increase in luciferase activity was observed in CDM−ILV. In the complemented mutant, induction of luciferase expression from the iviI promoter in CDM−ILV was restored. The apparent decrease in relative RLU/OD unit in CDM−ILV compared to CDM+ILV in the lrp mutant is due to the lack of growth of the mutant in CDM−ILV (Table 3). Expression from the iviI promoter was strongly up-regulated in CDM−ILV in the wild-type and complemented mutant strains containing Lrp and was unresponsive in the Lrp mutant. These data suggest that Lrp is directly involved in the regulation of the A. pleuropneumoniae ilvIH promoter within pIviI in response to BCAA limitation.

FIG. 7.

Relative expression from iviI and iviG promoters in wild-type and lrp mutant backgrounds in CDM−ILV and CDM+ILV. Bacterial strains were inoculated into CDM+ILV and CDM−ILV broth to an OD₅₂₀ of ∼0.1, and luciferase activity expressed from the promoter:luxAB fusions was measured over time. Luciferase activity was first normalized to RLU per OD₅₂₀, and relative expression for each culture at each time point was calculated as RLU/OD unit in CDM−ILV divided by RLU/OD unit in CDM+ILV. (A) Expression from the iviI promoter in wild-type (AP225/pGZRS39/pIviI, white bars), lrp mutant (AP359/pGZRS39/pIviI, dark gray bars), and complemented lrp mutant (AP359/pTW415/pIviI, light gray bars) strains. (B) Expression from the iviG promoter in wild-type (AP225/pGZRS39/pIviG, white bars), lrp mutant (AP359/pGZRS39/pIviG, dark gray bars), and complemented lrp mutant (AP359/pTW415/pIviG, light gray bars) strains. Data are presented as the means ± standard deviations from three experiments in panel A and from six experiments in panel B.

The luciferase activities of strains containing pIviG were also examined. The luciferase activity expressed from the iviG promoter in wild-type, the lrp mutant, and the complemented mutant showed an initial level of expression of ∼100 RLU/OD unit in CDM+ILV, that increased two- to threefold within 3 h in all three backgrounds In the wild-type background, expression from the iviG promoter was further increased twofold in CDM−ILV compared to CDM+ILV (Fig. 7B). This additional increase in expression from iviG was not seen in the lrp mutant and was partially restored in the complemented mutant. Again, the apparent decrease in relative RLU/OD unit in CDM−ILV compared to CDM+ILV in the lrp mutant is due to the lack of growth of the mutant in CDM−ILV (Table 3). These data suggest that Lrp is also directly involved in the regulation of the iviG promoter in response to BCAA limitation, although the increase in expression for iviG is distinctly lower and slower than the rapid and robust response seen with iviI.

DISCUSSION

In this study, we report the identification of the A. pleuropneumoniae lrp gene and its cloning, sequencing, protein purification, and mutation by deletion disruption using a chloramphenicol selection/sucrose counterselection procedure. The A. pleuropneumoniae Lrp is similar in amino acid sequence and function to the extensively studied Lrp from E. coli. The functional domains for DNA binding, transcriptional activation, and leucine response identified in E. coli Lrp (43) are also highly conserved in A. pleuropneumoniae and suggest that the functions of A. pleuropneumoniae Lrp may be similar. This is further supported by the in vitro binding of purified A. pleuropneumoniae His₆-Lrp binding in vitro to the E. coli ilvI promoter in an EMSA experiment (data not shown). These results demonstrate that A. pleuropneumoniae does have an lrp gene and that its protein function is similar to that of E. coli Lrp.

However, differences may exist between these similar proteins. A study by Platko and Calvo (43) showed that multiple mutations in Lrp can affect DNA binding, activation, and the leucine response of Lrp. While A. pleuropneumoniae has the same amino acids at 21 of the 22 sites identified as critical for these functions, E. coli has a serine at position 125 of E. coli Lrp whereas A. pleuropneumoniae has an alanine at the homologous position (Fig. 1). The serine was shown to be important in the activation of transcription by Lrp in E. coli. The unconserved alanine in A. pleuropneumoniae Lrp suggests that a difference in regulation could exist between E. coli and A. pleuropneumoniae Lrp. In Haemophilus influenzae, a bacterium closely related to A. pleuropneumoniae, Lrp was shown to affect the expression of fewer proteins (18) than in E. coli (15), which suggests that a difference between the roles of Lrp in E. coli and A. pleuropneumoniae could be expected.

Our main hypothesis guiding this study was that a subset of ivi genes responding similarly to BCAA limitation are also regulated by a similar mechanism. Since Lrp had been identified as a regulator of ilvI in E. coli (44, 46), we speculated that an A. pleuropneumoniae Lrp would regulate not only the A. pleuropneumoniae ivi gene ilvI (19) but also other ivi genes we previously identified as being up-regulated by BCAA limitation (59).

Since it had been previously shown that the E. coli His₆-Lrp behaves as the native protein does (36), we began to address our hypothesis by using purified A. pleuropneumoniae His₆-Lrp to determine if A. pleuropneumoniae Lrp has a role in the regulation of ivi genes shown to induce under BCAA limitation. The observation that two of the eight identified clone inserts bound A. pleuropneumoniae His₆-Lrp supports that a subset of clones shown to induce under BCAA limiting conditions are regulated similarly. The remaining six ivi clone inserts that did not bind A. pleuropneumoniae His₆-Lrp could be regulated by countless other mechanisms, such as by other protein regulators or amino acid attenuation, or could be regulated indirectly by Lrp.

The Lrp regulon in E. coli is extensive, including ∼10% of all E. coli genes (54). We identified A. pleuropneumoniae homologues of several genes known to be regulated by Lrp in E. coli, including ilvG (45), serA (67), and lrp itself (61), and tested these by EMSA. Lrp also regulates a variety of virulence-associated genes, including those for fimbriae in E. coli (5) and hemolysin in Xenorhabdus nematophila (10, 26). Therefore, we also tested binding of Lrp to the putative A. pleuropneumoniae promoters of Apx toxin genes, two fimbrial operons, and the nqr gene. A. pleuropneumoniae Lrp did bind to its own promoter, suggesting that regulation of the Lrp gene may be similar in A. pleuropneumoniae and E. coli. However, Lrp failed to bind to the ilvG or serA promoter, which suggests that regulation of BCAA biosynthesis by Lrp in A. pleuropneumoniae is different from, or at least not as complex as, that in E. coli (18). A. pleuropneumoniae Lrp did not bind to the apxI, apxII, apxIV, apf, or flp-rcp-tad promoters under the assay conditions used, suggesting that either Lrp does not regulate these genes in A. pleuropneumoniae, the effect of Lrp is indirect, or the assay conditions established for binding to the ilvIH promoter are not optimal for all DNA fragments. Lrp did bind to the DNA fragment containing the nqr operon promoter. However, it should be noted that this fragment potentially also contains the promoter region for a divergently transcribed upstream open reading frame. A. pleuropneumoniae nqrA has been shown to be strongly expressed and antigenic in vivo (11) and has been shown to be essential for survival during infection (52). In Vibrio cholerae, mutations in Nqr affect virulence gene expression (25). However, there is no nqr operon in E. coli. This is the first report of potential regulation of nqr by Lrp. These results indicate that the A. pleuropneumoniae Lrp regulon, while not as extensive as that characterized for E. coli, is not limited to genes involved in BCAA biosynthesis and does include both in vivo-induced and virulence-associated genes.

While these experiments were not designed to allow accurate calculation of binding affinities of Lrp, we were able to estimate the binding constant for the ilvIH (iviI) promoter to be 125 nM. This estimate is greater than the calculated K_ds for E. coli Lrp binding to the E. coli ilvIH gene (8 nM) and to the lrp gene (35 nM) (61) but still within a reasonable range for a DNA-binding protein.

The ability of A. pleuropneumoniae His₆-Lrp to bind to the DNA inserts of ivi clones that did not respond to the limitation of BCAAs was not analyzed. Given that certain promoters can be regulated by E. coli Lrp in the presence of leucine and other promoters in the absence of leucine (15, 34, 40), it is distinctly possible that Lrp from A. pleuropneumoniae may bind to additional ivi clone DNA inserts from clones that failed to induce under BCAA limitation. If this is the case, there may be additional ivi genes that are regulated by Lrp that have not been identified within the scope of this study.

While the demonstrated binding of recombinant A. pleuropneumoniae His₆-Lrp to DNA fragments in vitro suggests regulation by Lrp, it is not proof that expression of these genes is controlled by Lrp. To complement these data, we analyzed expression from the iviI and iviG promoters in wild-type A. pleuropneumoniae, an lrp mutant, and a complemented mutant. In the presence of leucine, isoleucine, and valine, there was minimal expression from the iviI promoter in all three backgrounds regardless of the presence of Lrp. In the absence of leucine, isoleucine, and valine, expression from the iviI promoter was strongly and rapidly up-regulated in the wild-type and complemented mutant strains containing Lrp but not in the Lrp mutant. These results suggest that the iviI promoter shows Lrp-dependent activation that is antagonized by leucine or by a combination of leucine, isoleucine, and/or valine, with a low basal level of expression in the absence of Lrp, and that Lrp is critical for the regulation of the ilvIH (iviI) in A. pleuropneumoniae.

The role of A. pleuropneumoniae Lrp in the expression of the iviG promoter is less clear. In CDM+ILV, there is an increase in basal activity from this promoter with increased growth (or possibly growth rate) in all three backgrounds, which was not seen with the iviI promoter. In the absence of leucine, isoleucine, and valine, expression from the iviG promoter was up-regulated in the wild-type and complemented mutant strains containing Lrp but not in the Lrp mutant. This pattern of up-regulation in response to BCAA limitation is similar to, but dramatically smaller and less rapid than, the response seen with the iviI promoter. These results suggest that the iviG promoter also shows Lrp-dependent activation that is antagonized by leucine, or by a combination of leucine, isoleucine, and/or valine, but in addition shows expression independent of Lrp. How the Lrp-independent expression of iviG is regulated has not been determined.

Mutation of the lrp gene does not appear to affect the growth rate of A. pleuropneumoniae in complex media or in CDM+ILV, except for a slightly longer lag time in broth cultures. However, the lrp mutant fails to grow in chemically defined medium in the absence of BCAAs. Growth in CDM−ILV is restored in the complemented mutant, indicating that the lack of growth is due to the mutation in lrp. Determination of whether this lack of growth CDM−ILV is due solely to the lack of induction of the ilvIH operon in the absence of LRP or to lack of regulation of additional genes involved in BCAA biosynthesis will require more extensive analysis of the Lrp regulon in A. pleuropneumoniae. It should be noted that the lack of growth of the lrp mutant in CDM−ILV could influence the expression assays. The rapid and robust response to BCAA limitation seen with the iviI promoter strongly suggests that the lack of expression in the lrp mutant strain is due to the lrp mutation and not to lack of growth. The lower and slower response seen with the iviG promoter makes it more difficult to draw this conclusion with complete confidence. However, since A. pleuropneumoniae His₆-Lrp binds to the iviG fragment in vitro, the regulation of this promoter by Lrp remains likely.

While the discovery of A. pleuropneumoniae Lrp binding to and regulating the expression of the ilvI promoter in A. pleuropneumoniae is novel because few regulators have been identified and examined in this organism, we were not surprised since Lrp had been shown to regulate ilvI in E. coli. In contrast, the binding of A. pleuropneumoniae Lrp to the iviG promoter is quite surprising. The pIviG insert contains the terminal 3′ end of the cps1A gene and the terminal 5′ end of the cps1B gene of A. pleuropneumoniae serotype 1, but the promoter in pIviG is in an antisense orientation to both cps1A and cps1B. cps1A and cps1B encode putative glycosyl transferases involved in the synthesis of A. pleuropneumoniae capsular polysaccharide. Capsule is required for virulence in this respiratory pathogen. An antisense transcript expressed from the iviG promoter could affect the expression of cps1B alone, both cps1A and cps1B, or the entire capsule biosynthetic operon. While the role of each gene in the biosynthesis of capsule has not been established, it is known that serotype 1 capsule is composed of a repeating N-acetyl-2-dioxy-β-d-glucopyranosyl and α-d-galactopyranosyl disaccharide that is partially O acetylated (2). Two possible roles of a transcript antisense to the capsule biosynthesis operon could be to reduce the total amount of capsule or to alter the antigenic structure by reducing the O acetylation. Reducing the amount of capsule could serve to expose surface adhesins necessary for attachment to respiratory epithelial cells. The fact that A. pleuropneumoniae Lrp binds to this region in vitro and possibly regulates the expression of the iviG promoter raises the possibility that Lrp may play a role in regulation of capsule biosynthesis of A. pleuropneumoniae. Future experiments comparing the amounts or types of capsule produced in wild-type and the lrp mutant strains are needed.

In E. coli, Lrp has been implicated in the regulation of genes involved in virulence such as fimbria genes (5), but to our knowledge, this study is the first time that Lrp has been implicated in the regulation of genes specifically induced during infection of the host. Furthermore, it is interesting to speculate on the affect of an lrp mutation on the virulence of A. pleuropneumoniae and compare it to what is known about the global regulator, Fur. Like Lrp, Fur has been shown to be both a positive and negative regulator (23) but can be modulated by iron rather than leucine. A Fur mutant of A. pleuropneumoniae has recently been shown to have reduced virulence (29). An A. pleuropneumoniae lrp mutant may also be attenuated if Lrp is necessary for the correct regulation of genes important in causing disease. Infection trials with the lrp mutant are needed to address this subject.

This is the first report to identify an A. pleuropneumoniae Lrp homologue. While the role of Lrp in the regulation of ilvI in E. coli has been extensively studied, this work addresses the role of A. pleuropneumoniae Lrp in the regulation of A. pleuropneumoniae virulence-associated genes, in vivo-induced genes, and BCAA biosynthetic genes. A. pleuropneumoniae Lrp was shown to bind to the promoter of A. pleuropneumoniae ilvI and regulate the expression under BCAA limitation. Furthermore, Lrp was shown to bind to the putative nqr promoter and the A. pleuropneumoniae serotype 1 capsule biosynthesis operon, suggesting for the first time that Lrp is involved in the regulation of A. pleuropneumoniae serotype 1 capsule biosynthesis and nqr expression. Our results suggest that the Lrp regulon in A. pleuropneumoniae differs from that found in E. coli and is potentially more extensive than the limited regulon found in H. influenzae.

In summary, our previous IVET studies with A. pleuropneumoniae led to the hypothesis that limitation of branched-chain amino acids is an important environmental cue for respiratory pathogens of mammals, which need to survive and multiply in an anatomical location where these amino acids are in short supply. Analysis of the in vivo-induced gene promoters identified in that work demonstrated that 25% of those ivi promoters were up-regulated on chemically defined medium lacking branched-chain amino acids compared to medium containing BCAAs (59). In this study, we have shown that two of these ivi promoters, as well as two additional genes, are regulated by the global regulatory protein Lrp. These results suggest that the ability to synthesize BCAAs, and the ability to produce a functional Lrp protein, may be required for respiratory pathogens of mammals, which further suggests that inhibition of BCAA synthesis or Lrp function might be a fruitful avenue for the development of new classes of antibiotics that would target respiratory pathogens.