Transcriptional Regulation of the virR Operon of the Intracellular Pathogen Rhodococcus equi

Abstract

The virR operon, located on the virulence plasmid of the intracellular pathogen Rhodococcus equi, contains five genes, two of which (virR and orf8) encode transcriptional regulators. The first gene of the operon (virR), encoding a LysR-type transcriptional regulator, is transcribed at a constitutive low level, whereas the four downstream genes are induced by low pH and high growth temperature. Differential regulation of the virR operon genes could not be explained by differential mRNA stability, as there were no major differences in mRNA half-lives of the transcripts representing each of the five genes within the virR operon. Transcription of virR is driven by the P_virR promoter, with a transcription start site 53 bp upstream of the virR initiation codon. The four genes downstream of virR are transcribed from P_virR and from a second promoter, P_orf5, located 585 bp downstream of the virR initiation codon. VirR binds to a site overlapping the initiation codon of virR, resulting in negative autoregulation of the virR gene, explaining its low constitutive transcription level. The P_orf5 promoter is induced by high temperature and low pH, thus explaining the observed differential gene expression of the virR operon. VirR has a positive effect on P_orf5 activity, whereas the response regulator encoded by orf8 is not involved in regulating transcription of the virR operon. The P_virR promoter is strikingly similar to those recognized by the principal sigma factors of Streptomyces and Mycobacterium, whereas the P_orf5 promoter does not share sequence similarity with P_virR. This suggests that P_orf5 is recognized by an alternative sigma factor.

The mycolic acid-containing actinomycete Rhodococcus equi is an intracellular pathogen of macrophages and the causative agent of foal pneumonia. Although foals are the primary host of this pathogen, other animals, including pigs, goats, and cattle, are sporadically infected. In addition to these animal hosts, R. equi infections are increasingly encountered in immunocompromised humans, in particular in those diagnosed with AIDS (20, 24). Virulence relies on the ability of R. equi to replicate inside macrophages (13), a process which is dependent on the pathogen interfering with endosomal maturation following phagocytosis and preventing acidification of the vacuole in which it resides (9, 37, 38). Eventually, intracellular proliferation of the pathogen leads to necrosis of the macrophage, accompanied by massive damage to lung tissue characterized by cavitation and granuloma formation (18, 20, 24).

All R. equi strains isolated from foals harbor an 81-kb plasmid that has been shown to be essential for intracellular replication and virulence (10, 35, 36). This virulence plasmid contains a region of 27.5 kb that is characterized by a lower G+C content than the remainder of the virulence plasmid and the R. equi genome, indicating that it probably was acquired via lateral gene transfer (33). It was proposed that this region represents a pathogenicity island, a hypothesis that was lent credence by the fact that the transcription of genes located within it is upregulated following phagocytosis of R. equi (25, 33). Among these is a group of genes encoding a family of small proteins, one of which, virulence-associated protein A (VapA), has been shown to be a virulence factor (14).

Transcriptional regulation of pathogenicity island genes is subject to at least six different environmental parameters, including temperature and pH (3, 25, 32, 34). In contrast to this apparent regulatory complexity, only two genes encoding transcriptional regulators, located in a five-cistron operon (the virR operon, containing virR, orf5, vapH, orf7, and orf8), have been identified in the pathogenicity island (28, 33). It therefore is likely that regulators encoded by the R. equi genome play an important role in controlling the expression of pathogenicity island genes. This may be accomplished, at least in part, through controlling the expression levels of the two transcriptional regulators within the pathogenicity island. The virR gene encodes a LysR-type transcriptional regulator (LTTR) that is required for transcription of the vapA gene (28). The orf8 gene encodes a response regulator; interestingly, a cognate sensor kinase protein is not encoded on the virulence plasmid (33). Disruption of either gene attenuates R. equi, demonstrating their importance in controlling virulence (26). A recent report describing a low level of constitutive transcription of the virR gene while the transcript levels of the four downstream genes responded to changes in growth conditions (25) appeared to contradict our previous finding that the five genes constitute an operon (28). Considering the importance of the virR operon in controlling the virulence of R. equi, this apparent paradox was examined further. This paper reports that the observed difference in regulation of the virR operon genes is not due to differential mRNA stability of the virR operon transcript but to the presence of a regulated promoter within the virR gene driving transcription of orf5, vapH, orf7, and orf8.

MATERIALS AND METHODS

Bacterial strains and plasmids.

Bacterial strains and plasmids used in this study are listed in Table 1.

TABLE 1.

Bacterial strains and plasmids used in this study

Strain, plasmid, or oligonucleotide	Genotype or characteristics	Source or reference
Strains
E. coli DH5α	supE44 ΔlacU169 (φ80lacZΔM15) hsdR17 recA1 endA1 gyrA96 thi-1 relA1	Bethesda Research Laboratories
E. coli BL21(DE3)pLysE	F−ompT hsdSB(rB− mB−) dcm gal λ(DE3) (pLysE) Cmr	31
R. equi ATCC 33701	Virulent strain with 81-kb virulence plasmid p33701	American Type Culture Collection
R. equi ATCC 33701 (P−)	Avirulent strain; virulence plasmid cured	35
R.equi orf8 mutant	orf8 deletion	26
Plasmids
pBluescript II KS(+)	Apr αlacZ′	Stratagene
pREV5	Aprr oriV(pMF1) oriV(ColE1)	19
pBlueRegP1	pBluescript KS with 6,680-bp fragment containing orf3 to orf8 from the R. equi ATCC 33701 virulence plasmid	28
pET3bvirRhis	pET3b with 931-bp fragment containing virR with a six-His tag at the 3′ end	28
pJOE814.2	Ampr ColE1 ori xylE reporter gene	23
pREV6	pREV5 with t tag located within two terminators derived from pJOE814.2	This study
pORF3PEX	pREV6 with a 2,705-bp DNA fragment containing promoters PvirR and Porf5, virR, and orf5	This study
pVirRT	pREV6 with a 2,183-bp DNA fragment containing Porf5, virR, and orf5	This study
p004	pREV6 with a 1,840-bp DNA fragment containing virR and orf5	This study
pVirRT-LysR	pVirRT with virR located outside the transcriptional terminators	This study
pInterVir	pBluescript KS with a 459-bp DNA fragment containing the virR-orf3 intergenic region	This study
pCodingVir	pBluescript KS with an 846-bp DNA fragment containing 690 bp of the virR coding region	This study
pREV531	pREV5 with a 1.5-kb DNA fragment containing orf3 and virR′ derived from pBlueRegP1	This study
pREV5341	pREV5 with a 3.1-kb DNA fragment containing orf3, virR, and orf5′ derived from pBlueRegP1	This study

Media and growth conditions.

Escherichia coli DH5α (Bethesda Research Laboratories) and E. coli BL21(DE3)pLysE (Novagen) were used for general cloning procedures and for expression of virR-his, respectively. Bacterial strains were grown in Luria-Bertani (LB) broth (29). Growth of R. equi at 37°C and pH 6.5 (inducing conditions) was used to induce virulence plasmid gene expression, whereas these genes were transcribed at low levels following growth at 30°C and pH 8.0 (noninducing conditions). Where appropriate, the following supplements were added: kanamycin, 50 μg ml⁻¹ (E. coli) or 200 μg ml⁻¹ (R. equi); ampicillin, 50 μg ml⁻¹; apramycin, 30 μg ml⁻¹ (E. coli) or 80 μg ml⁻¹ (R. equi); chloramphenicol, 30 μg ml⁻¹; 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-Gal), 20 μg ml⁻¹; isopropyl-β-d-thiogalactopyranoside (IPTG), 0.1 mM. For solid medium, agar was added at 1.5% (wt/vol).

DNA manipulations.

Chromosomal DNA was isolated as described previously (28). Plasmid DNA was isolated via the alkaline lysis method of Birnboim and Doly (5) or by using a Wizard Plus SV miniprep kit as described by the manufacturer (Promega). DNA fragments were isolated from agarose gels by using a Genelute DNA purification kit as described by the manufacturer (Sigma-Aldrich). PCR was carried out using Taq DNA polymerase (Promega) or Deep Vent DNA polymerase (New England Biolabs) as described by the manufacturer. Other DNA manipulations were done in accordance with standard protocols (29).

Plasmid construction.

A region located on plasmid pJOE814.2 containing two transcriptional terminators and a xylE gene was amplified with Deep Vent DNA polymerase, using primers XylF and XylR. This 1,424-bp fragment was cloned into DraI-digested pREV5 to make pREV81. The xylE gene was subsequently excised by digesting pREV81 with ClaI, followed by religation of the 4,948-bp fragment, resulting in pREV6.

A series of DNA fragments serially shortened from the 5′ end was constructed by PCR with the oligonucleotide BCMRev in conjunction with either ORF3PEX2, VirRT1, or 004F, using pBlueRegP1 as a template. These fragments were ligated into the unique EcoRV restriction site of pREV6 to yield pORF3PEX, pVirRT, and p004, respectively. An intact virR gene was amplified from pBlueRegP1 by using oligonucleotides 005R and Orf3BlnI. The resulting 1,717-bp product was digested with BlnI, and the 1,296-bp fragment was cloned into SpeI-digested pVirRT.

pInterVir was constructed by amplifying a 459-bp fragment containing the orf3-virR intergenic region from pBlueRegP1, using oligonucleotides GM-Lys-F and GM-Lys-R; this fragment was subsequently ligated into EcoRV-digested pBluescript II KS. pCodingVir was constructed by amplifying an 846-bp fragment containing the virR coding region, using oligonucleotides VirRT1 and −70−50REV; this fragment was subsequently ligated into EcoRV-digested pBluescript II KS. The sequences of the oligonucleotides used in the construction of these plasmids are listed in Table 2.

TABLE 2.

Oligonucleotides used in this study

Oligonucleotide	Sequence (5′ to 3′)	Purpose	Reference
XylF	TATATAACTAGTGCCTCTTCGCTATT	Plasmid construction	This study
XylR	TATAGGCCGGCCCGATTCATTAATGCA	Plasmid construction	This study
BCMRev	GTTATCTAGGGGCAGGCGACAG	Plasmid construction	This study
VIRRT1	CGCATTGAACGACAGGTTG	Plasmid construction	This study
ORF3PEX2	TGACGAGCACCAATGTTTTC	Plasmid construction	This study
GM-Lys-R	CGGCTGCCGCAATATGAC	Plasmid construction	This study
GM-Lys-F	CACCCCAACCTGTCGTTC	Plasmid construction	This study
−70−50 REV	GACCACCTGCGATGCGTGAT	Plasmid construction	This study
GyrCOOH1	GTCGAGCAGGGTCAAGTGTA	Quantification of gyrB	This study
GyrCOOH25′	AGCTCCTTGGCGTTCATCT	Quantification of gyrB	This study
16SrRNAF200	ACGAAGCGAGAGTGACGGTA	Quantification of 16S rRNA	21
16SrRNAR200	ACTCAAGTCTGCCCGTATCG	Quantification of 16S rRNA	21
003F	GTTTCGTCTTCCACCGTCTT	Quantification of orf3	This study
003R	AGCCTTATCGTCGCAACTGT	Quantification of orf3	This study
004F	CGGACGAGTTCGACTGGTAT	Quantification of virR	This study
004R	CAAAGACGATTTGGGGTACG	Quantification of virR	This study
005F	CTCTTCCTGATCGGAGTTGC	Quantification of orf5	This study
005R	GAGTCGCAGACGAGGTAAGC	Quantification of orf5	This study
006F	AGGGTTATGCAGGTGGATTG	Quantification of vapH	This study
006R	TACCGATTACGGAGCTCACC	Quantification of vapH	This study
007NF	ATGCACTCCCTGAAAACTATC	Quantification of orf7	This study
007NR	GGTGGGCTGGATTGACGCGCA	Quantification of orf7	This study
008F	GAACAACTGGGAATGGTGGT	Quantification of orf8	This study
008R	GTTCGCCGTTTCTAGACGAA	Quantification of orf8	This study
D4-VIR5022	GATGTGCAGGGCGTCAGC	Primer extension	This study
VIR5022	GATGTGCAGGGCGTCAGC	Sequencing	This study
D4-2IntPex5564	GGTACGACGCCAGCAGCCGC	Primer extension	This study
2IntPex5564	GGTACGACGCCAGCAGCCGC	Sequencing	This study

Electroporation of R. equi.

Plasmids were introduced into R. equi strains by electroporation as described previously (19).

RNA isolation and real-time RT-PCR.

RNAs were isolated from R. equi as described previously (28). Reverse transcriptase (RT) reactions using random primers (Promega) were performed with 1 U Improm II RT following the manufacturer's recommendations, with 100 ng of total RNA as the template in a final volume of 20 μl. The product was subsequently amplified using a QuantiTect SYBR green real-time kit following the manufacturer's instructions (QIAGEN). Reaction mixtures were subjected to 40 cycles of 95°C for 15 s, 60°C for 30 s, and 72°C for 30 s in a LightCycler instrument (Roche) with a temperature transition rate of 20°C/s. Melting curve analysis was performed at 50 to 95°C (temperature transition rate, 0.2°C/s), with stepwise fluorescence detection following amplification. Cycle threshold (C_T) values were obtained and used to calculate the number of RNA copies per μg of total RNA, using a standard curve of known amounts of DNA target with r² coefficients of >0.997 in the range of 5 × 10³ to 5 × 10⁸ molecules per reaction. gyrB mRNA was used as a housekeeping gene to compare the amounts of RNA in each reaction. The data reported in this paper represent the results of three independent experiments in which each sample was analyzed in duplicate. The sequences of oligonucleotides used for real-time PCR are listed in Table 2.

Fluorescent primer extension and DNA sequencing.

The D4-labeled oligonucleotides D4-VIR5022 and D4-2IntPex5564 (Table 2), complementary to sequences 62 to 80 bp and 624 to 643 bp, respectively, downstream of the initiation codon of virR, were used in primer extension reactions to determine the transcriptional start sites of the virR operon. Total RNA (2 μg) and 1 μM of D4-VIR5022 were incubated at 70°C for 5 min, followed by reverse transcription at 42°C for 60 min, using 5 U of Superscript III RT in a volume of 20 μl as recommended by the manufacturer (Invitrogen). After treatment of the sample with 20 μg of RNase A at 37°C for 30 min, cDNA was precipitated and dissolved in 12 μl of nuclease-free water. The primer extension product (0.5 ng) was combined with 0.5 μl of DNA size standard kit 600 (Beckman Coulter) and 40 μl of CEQ sample loading solution (Beckman Coulter) and analyzed with a CEQ 8000 fragment analysis system on a CEQ 8000 DNA sequencer (Beckman Coulter). In addition, dideoxy sequencing reactions using either VIR5022 or 2IntPex5564 (Table 2) and 60 ng of NheI-digested pORF3PEX (Table 1) were performed using a CEQ DCTS kit as described by the manufacturer (Beckman Coulter). The D4-labeled primer extension products (50 pg) were added to the sample prior to analyzing the sequence on a CEQ 8000 DNA sequencer to identify transcriptional start sites.

mRNA half-life determination.

RNAs were isolated from R. equi in the mid-logarithmic phase of growth (optical density at 600 nm = 0.5) at selected intervals following inhibition of transcription by the addition of 200 μg ml⁻¹ rifampin (Sigma). The number of mRNA copies was determined using real-time RT-PCR, followed by linear regression to determine the half-life. 16S rRNA was used for internal normalization in this study.

EMSA.

Expression and purification of virR-his were done as previously described (28). A DNA fragment containing the virR promoter region, pInterVir, was digested with EcoRI and HindIII and labeled with [α-³²P]dATP, using Klenow DNA polymerase as described previously (28). The labeled fragment was subsequently purified using a QIAquick PCR purification kit according to the manufacturer's instructions (QIAGEN). Radiolabeled DNA fragments (2 ng) were incubated with purified VirR-His at 30°C for 30 min in electrophoretic mobility shift assay (EMSA) binding buffer, 20 μg of bovine serum albumin, and 1 μg of poly(dI-dC) DNA (Amersham Biosciences) in a volume of 20 μl. The samples were separated by electrophoresis in a prerun 5% nondenaturing polyacrylamide gel containing TBE (45 mM Tris base, 45 mM boric acid, 1 mM EDTA) and run at 4°C and 10 V cm⁻¹. Following drying, the gel was analyzed by autoradiography.

DNA restriction protection assay.

A 459-bp fragment containing the virR promoter region was amplified using the oligonucleotides GM-Lys-F and GM-Lys-R, with pInterVir as the template. To obtain a negative control DNA region, an 846-bp fragment was amplified using the oligonucleotides VirRT1 and −70−50REV. DNA fragments (500 ng) were incubated with purified VirR-His at 30°C for 30 min in EMSA binding buffer. HincII (10 units) was added to each sample, which was further incubated at 30°C for 30 min. The samples were separated by electrophoresis and visualized using SYBR green (Sigma) per the manufacturer's instructions.

RESULTS

Transcription of the virR operon is controlled by temperature and pH.

Using a microarray to study the transcription levels of pathogenicity island genes, Ren and Prescott (25) showed that the transcription of four genes (orf5, vapH, orf7, and orf8) downstream of virR, but not that of virR itself, is upregulated when cells are grown at pH 6.5 and 37°C compared to that at pH 8.0 and 30°C. Since this observation appeared to contradict our previous findings that these genes are organized in an operon (28), mRNAs from R. equi grown under these two conditions were isolated and quantified using real-time RT-PCR (Fig. 1). The transcription levels of the four genes downstream of virR were significantly increased (threefold) when cells were grown at pH 6.5 and 37°C compared to those for cells grown at pH 8.0 and 30°C. In contrast, there was no significant difference in transcription levels of virR, the first gene of the operon, thus confirming the earlier microarray data (25).

FIG. 1.

Regulation of virR operon gene transcription by temperature and pH. mRNAs were isolated from Rhodococcus equi cells grown under noninducing (30°C, pH 8.0) or inducing (37°C, pH 6.5) growth conditions, followed by absolute quantification of the mRNA molecules using real-time RT-PCR. Transcription levels are indicated for each gene of the virR operon (virR, orf5, vapH, orf6, orf7, and orf8) and the divergently transcribed orf3 gene following growth under noninducing (gray bars) and inducing (black bars) conditions.

mRNA stability of the virR operon transcript.

Although virR, orf5, vapH, orf7, and orf8 are organized in an operon (28), their transcription is not coordinately regulated (25). A possible explanation for this apparent paradox may be that mRNAs derived from orf5, vapH, orf7, and orf8 are more stable than that of virR, allowing the former four transcripts to accumulate. To examine this possibility, the mRNA half-lives of these genes were determined following inhibition of transcription by the addition of rifampin (Table 3). The half-lives of the transcripts representing each of the five genes within the virR operon varied from 1.3 to 2.3 min. Since the virR transcript had a half-life of 1.8 min, differential mRNA stability of the virR operon transcript does not account for the observed differences in transcript levels of virR and its four downstream genes.

TABLE 3.

Transcript half-lives of genes within the virR operon

Gene	Half-life (min)
virR	1.8 ± 0.2
orf5	1.3 ± 0.3
vapH	2.3 ± 0.3
orf7	2.2 ± 0.1
orf8	1.3 ± 0.1

The virR operon contains an internal promoter regulated by pH and temperature.

An alternative explanation for the observed differential regulation of the virR operon genes is the presence of a promoter located downstream of the virR operon promoter (P_virR) driving transcription of the orf5, vapH, orf7, and orf8 genes. Transcription levels were analyzed using the promoter probe vector pREV6, which contains a small DNA fragment (t tag) located between two transcriptional terminators which is not present in R. equi (Fig. 2A). Since there is no promoter, transcription of the t tag does not occur (Fig. 2B). Introduction of a DNA fragment containing the orf3-virR intergenic region as well as orf5 into pREV6 resulted in transcription of the t tag, demonstrating the presence of an active promoter. Deletion of the orf3-virR intergenic region, including the 5′ end of virR (1 to 111 bp), reduced but did not abolish transcription of the t tag. Transcription of the t tag was completely abolished when a larger 5′ fragment (1 to 404 bp) of virR was deleted, demonstrating that a promoter (P_orf5) or sequences required for its activity are located between bp 111 and 404 of the virR gene (Fig. 2B).

FIG. 2.

Determination of P_virR and P_orf5 promoter activities by real-time RT-PCR. (A) Schematic representation of inserts in the promoter probe vector pREV6. The hairpin structure marked with a “T” represents the transcriptional terminators. Genes and their direction of transcription are indicated by black arrows. The open white box represents the t tag. The transcription level of the t tag was measured by real-time RT-PCR. (B) mRNA transcript levels of the t tag in R. equi cells harboring pORF3PEX, containing both P_virR and P_orf5 (1); pVirRT, containing P_orf5 (2); p004 (3); or pREV6 (4). (C) mRNA transcript levels of the t tag in R. equi cells harboring pVirRT containing only P_orf5 grown at 37°C and pH 6.5 (1) or at 30°C and pH 8.0 (2) and in R. equi cells harboring pREV6 grown at 37°C and pH 6.5 (3) or 30°C and pH 8.0 (4).

To examine whether the internal promoter P_orf5 is regulated by temperature and pH, the transcription levels of the t tag were examined following growth of R. equi(pVirRT) under either inducing (pH 6.5, 37°C) or noninducing (pH 8.0, 30°C) growth conditions. Transcription of the t tag increased 67-fold under inducing compared to noninducing conditions, whereas transcription of the t tag did not occur from pREV6 (Fig. 2C).

Mapping of the transcriptional start sites of the virR operon.

To determine the transcriptional start site of the virR gene, a primer extension reaction using the fluorescently labeled oligonucleotide D4-VIR5022, complementary to the virR gene, was carried out. Using mRNA isolated from R. equi grown under inducing conditions as the template, a 133-bp DNA fragment was observed (Fig. 3A). The transcriptional start site of P_virR was subsequently determined to be an adenine located 53 bp upstream of the virR initiation codon (Fig. 3A and 4D). Interestingly, the −10 sequence (nucleotides −12 to −7 [TAGCAT]) of the P_virR promoter is strikingly similar to that of the consensus σ^hrdB promoter (TAGART), which is recognized by the principal sigma factor of Streptomyces coelicolor (7, 15). However, there is no clear similarity in the −35 region to the consensus σ^hrdB promoter (TTGACA). Using the same approach, the internal transcriptional start site of P_orf5 was mapped to a guanidine 585 bp downstream of the virR initiation codon (Fig. 3B and 4D). The P_virR and P_orf5 promoters do not share any obvious sequence similarity in either the −10 or −35 area.

FIG. 3.

Determination of transcriptional start sites of the virR operon of R. equi. Fluorescent primer extension was carried out with a Cy5-labeled primer and 5 μg of total cellular RNA extracted from R. equi cells grown under inducing conditions (37°C and pH 6.5). The upper panels show the D4-labeled primer extension products combined with DNA size standards and analyzed with a CEQ 8000 fragment analysis system. The lower panels show dideoxy sequencing reactions spiked with the D4-labeled primer extension products. The arrows indicate the transcriptional start sites where the D4-labeled cDNA and sequencing products overlapped. (A) Determination of the P_virR transcriptional start site, using D4-VIR5022. (B) Determination of the P_orf5 transcriptional start site, using D4-2IntPex5564. AU, arbitrary units.

FIG. 4.

Analysis of VirR DNA binding. Various concentrations of VirR were incubated with 2 ng of radiolabeled DNA containing (A) the P_virR promoter region in a 459-bp DNA fragment or (B) the P_orf5 promoter region in an 846-bp DNA fragment. The amount of protein added to each lane was as follows: lanes 1, radiolabeled DNA fragment only; lanes 2, 50 ng VirR-His; lanes 3, 100 ng VirR-His; lanes 4, 200 ng VirR-His; lanes 5, 300 ng VirR-His; and lanes 6, 400 ng VirR-His. The reaction volume was 20 μl. Protein-DNA complexes are indicated with black arrowheads. Nonbound DNA is indicated with a gray arrowhead. (C) HincII restriction protection assay. DNAs were incubated with VirR and HincII, followed by analysis of the restriction digest by gel electrophoresis. Lanes 1 to 3 contain a 459-bp DNA fragment with P_virR, and lanes 5 and 6 contain an 846-bp DNA fragment with P_orf5. Lanes 1 and 5, no VirR or HincII added; lanes 2 and 6, HincII added; lanes 3 and 7, HincII and VirR added. (D) Schematic representation of the 5′ end of the virR operon. Genes and the direction of transcription are indicated by black arrows. The HincII sites used in the restriction protection assay and the positions of P_virR and P_orf5 are indicated above the arrows. The nucleotide sequence of the sequence upstream of virR containing the P_virR promoter is shown. The P_virR transcriptional start site is indicated (+1).

VirR binds to the orf3-virR intergenic region.

To determine whether VirR interacts with the P_virR and P_orf5 promoters, band shift experiments were carried out. Incubation of purified VirR-His with a 459-bp radiolabeled DNA fragment containing the orf3-virR intergenic region resulted in retardation of the DNA fragment. At low VirR-His concentrations, a single retarded band was visible, whereas a second retarded band became visible at higher concentrations (Fig. 4A, lane 5). In contrast, VirR-His failed to bind to a DNA fragment containing only the internal P_orf5 promoter (Fig. 4B).

To further corroborate that VirR-His binds to a DNA fragment containing the P_virR promoter, a restriction enzyme protection assay was carried out. The 459-bp DNA fragment containing the orf3-virR intergenic region was incubated with VirR-His and subsequently subjected to digestion with HincII, whose recognition site is located 59 to 64 bp downstream of the P_virR transcription initiation site. While VirR-His protected this 459-bp DNA fragment from HincII digestion, it failed to prevent digestion by HincII of the adjacent 846-bp DNA fragment. These data show that the VirR binding site overlaps the HincII site downstream of P_virR within the virR coding region (Fig. 4C).

VirR is a negative regulator of the virR gene.

The results of the restriction enzyme protection assay show that VirR protects a HincII site located 59 to 64 bp downstream from the P_virR transcriptional start site. This strongly suggests that VirR acts as a repressor of P_virR driving the transcription of virR. To examine this possibility, plasmids with DNA fragments containing either orf3, an intact virR gene and the 5′ end of orf5 (pREV5341), or orf3 and the 5′ end of virR (pREV531) were electroporated into R. equi P⁻. Transcription of virR was subsequently analyzed by RT-PCR, using primers that amplified a virR fragment located upstream of P_orf5, thus analyzing only the activity of P_virR. While transcription of virR was below the detection limit in strains harboring pREV5341 (Fig. 5, lane 2), it was clearly detectable in the absence of an intact virR gene (pREV531) (Fig. 5, lane 3).

FIG. 5.

Autoregulation of virR transcription. mRNAs were isolated from R. equi P⁻ harboring either pREV531 (orf3-virR′) or pREV5341 (orf3-virR-orf5′). virR and gyrB transcripts were subsequently detected by RT-PCR as 200-bp amplification products. The oligonucleotides used for this experiment (Table 2) amplified a region upstream of P_orf5. Lane 1, 100-bp marker; lane 2, R. equi(pREV5341) (virR); lane 3, R. equi(pREV531) (virR); lane 4, R. equi(pREV5341) (gyrB); lane 5, R. equi(pREV531) (gyrB); lanes 6 and 7, RT-PCR of gyrB genes of R. equi(pREV5341) and R. equi(pREV531), respectively, without RT.

VirR is a positive regulator of the orf5-orf8 cluster.

Since the orf5 and orf8 genes are transcribed from the P_virR promoter as well as the P_orf5 promoter, it is likely that VirR is involved in transcription from P_orf5 as well as that from P_virR. To investigate this possibility, pVirRT, containing P_orf5 and the 3′ end of virR, was introduced into R. equi P⁻. The transcription levels of the t tag in this strain were 14-fold lower than those in the wild-type strain, indicating that the presence of virulence plasmid genes is required for wild-type transcription levels. The transcription levels were restored only partially (twofold increase) following introduction of an intact virR gene into pVirRT (but outside the transcriptional terminators of pREV6), demonstrating that VirR is required but not sufficient for full activity of the P_orf5 promoter.

orf8 does not regulate transcription of the virR operon.

The orf8 gene encodes a response regulator that probably interacts with an as yet unidentified chromosomally encoded sensor kinase protein to regulate virulence plasmid gene expression. To determine whether orf8 is involved in transcriptional regulation of the virR operon, the transcription levels of orf5 were compared in the wild-type strain and an orf8 disruption mutant. The orf5 transcription level in the wild-type strain was 5.6 × 10⁵ molecules mRNA/100 ng RNA ± 7.6 × 10⁴ molecules mRNA/100 ng RNA, whereas this level was 4.7 × 10⁵ molecules mRNA/100 ng RNA ± 5.6 × 10⁴ molecules mRNA/100 ng RNA for the R. equi orf8 mutant, indicating that orf8 does not play a role in regulating transcription of the virR operon.

DISCUSSION

The virR operon contains two genes encoding transcriptional regulators which are required for the virulence of R. equi; one of these, the LTTR protein VirR, is required for expression of the virulence factor VapA (26, 28). The transcriptional organization of virR is unusual in that it is the first gene in a five-cistron operon (28); in most instances, LTTR-encoding genes are transcribed as monocistronic transcripts. In addition, an analysis of virulence plasmid transcript levels following growth under a variety of conditions showed that virR is transcribed at a low and constant level, whereas the transcription of the downstream genes is regulated (25). The aim of this paper was to analyze the transcriptional regulation of the virR operon in order to explain the observed differential regulation of virR compared to the four downstream genes.

Many LTTRs autoregulate their expression by acting as a repressor of the LTTR-encoding gene (30). The data presented here show that in this respect, virR behaves as a typical LTTR. Inactivation of virR in a virulence plasmid-free background resulted in a dramatic increase of virR transcription, which is consistent with the notion that VirR acts as a repressor of virR transcription. This is supported by the finding that VirR binds to a site that overlaps a HincII restriction site located 59 to 64 bp downstream of P_virR within the 5′ end of the virR gene, which is a typical location for a repressor binding site (8). The VirR autoregulatory circuit thus results in a constant low level of transcription of the virR operon from the P_virR promoter, independent of growth temperature and pH.

Differential mRNA stability within a polycistronic transcript has been shown to lead to vastly different mRNA levels for individual genes within an operon, resulting in differential gene expression (11, 17). This could account for the observed increase in orf5, vapH, orf7, and orf8 transcripts under inducing conditions. However, the half-life of virR mRNA (1.8 min) was the same as the average half-life of the transcripts of all five genes within the virR operon and therefore does not account for the observed differential regulation of the four genes downstream of virR in the virR operon. The observed short half-lives are typical of the majority of bacterial transcripts, as recently shown for Escherichia coli and Bacillus subtilis, where over 80% of transcripts are unstable, with half-lives of <8 min (4, 12).

Deletion analysis of the virR operon showed that in addition to the P_virR promoter, transcription of the four genes downstream of virR is also driven by a second promoter located within the virR gene. In contrast to P_virR, the activity of the P_orf5 promoter is regulated by temperature and pH, resulting in a 67-fold increase in transcript levels following growth under inducing conditions compared to those obtained under noninducing growth conditions. Interestingly, the activity of the P_orf5 promoter in a virulence plasmid-free R. equi strain was significantly lower than that in the presence of the virulence plasmid, suggesting that one or more components encoded by the virulence plasmid are required for full activity of P_orf5. One of these components is the virR gene, as inclusion of this gene on the plasmid containing P_orf5 resulted in a twofold higher activity, indicating that virR is required but not sufficient for full activity of P_orf5.

The p004 plasmid contains the P_orf5 promoter and its upstream region (−140 nucleotides) but did not display any promoter activity, indicating that a far upstream region required for P_orf5 activity had been deleted. The requirement for far upstream sequences for promoter activity has been shown for many bacterial systems, for example, Pseudomonas aeruginosa and Mycobacterium tuberculosis (22, 27). Often these involve two-component regulatory systems interacting with either σ⁵⁴- or σ⁷⁰-type RNA polymerases (6, 8). However, disruption of orf8, encoding a response regulator, had no significant effect on orf5 transcription levels, showing that it is not involved in controlling transcription of the virR operon under the experimental conditions used. This strongly suggests that genome-encoded transcriptional regulators are involved in regulating the activity of the P_orf5 promoter, and therefore the expression levels of the response regulator Orf8.

The −10 regions of promoters recognized by the principal sigma factors of Streptomyces coelicolor, Mycobacterium smegmatis, and M. tuberculosis (HrdB and MysA) are highly conserved (2, 7, 15). Interestingly, the −10 regions of P_virR and the previously characterized vapA promoter (28) are highly similar to those recognized by HrdB and MysA, indicating that these promoters are recognized by the main principal sigma factor of R. equi. In contrast, the −10 and −35 sequences of P_virR and P_orf5 do not share any obvious sequence similarities, strongly suggesting that these two promoters are recognized by different sigma factors. The involvement of alternative sigma factors in the regulation of virulence factor expression, as proposed here, is increasingly being observed (1, 16). The R. equi genomic sequence was recently completed (http://www.sanger.ac.uk/Projects/R_equi/), and a preliminary genome analysis revealed the presence of at least 20 potential sigma factor-encoding genes (R. J. Fahey and W. G. Meijer, unpublished results). We are currently analyzing the functions of these sigma factors.

The model that emerges from these studies is that the virR operon is transcribed at a low constitutive level under noninducing conditions from the P_virR promoter, which is most likely recognized by the principal sigma factor. The noninducing growth conditions resembled those encountered during saprophytic growth of R. equi, where expression of virulence factors is not required. The constitutive low-level transcription of the virR operon is maintained through binding of VirR to a site overlapping the virR initiation codon, resulting in an autoregulatory circuit. Following a change to inducing growth conditions, which are detected by an as yet unidentified genome-encoded signal transduction pathway, P_orf5 becomes active, resulting in a significant increase of orf5, vapH, orf7, and orf8, but not virR, expression. Transcription from P_orf5 is dependent on an alternative sigma factor and involves far upstream sequences. The orf8 gene encodes a response regulator which is required for virulence (26) but not for P_orf5 activity, whereas the roles of orf5, orf7, and vapH in virulence are not clear. Upregulation of P_orf5 activity by inducing growth conditions may act as a master switch, resulting in increased expression of the response regulator Orf8, leading to subsequent full induction of the virulence genes. The validity of this model for the regulation of virulence gene expression is currently under investigation.