The ColR/ColS Two-Component System Plays Multiple Roles in the Pathogenicity of the Citrus Canker Pathogen Xanthomonas citri subsp. citri

Abstract

Bacterial citrus canker disease, which is caused by Xanthomonas citri subsp. citri, is one of the most devastating diseases of citrus plants. In this study, we characterized the role of the two-component regulatory system ColR/ColS in the pathogenicity of X. citri subsp. citri. colS mutants (256A10 and 421E7), colR mutants (386C6 and 417E10), and a colR colS double mutant (306DSR) all lost pathogenicity and produced no symptoms on grapefruit leaves inoculated by either pressure infiltration or the spray method. The pathogenicity defect of the colS, colR, and colR colS mutants could be complemented using the wild-type colS, colR, and colR colS genes, respectively. Mutation of colS or colR significantly reduced X. citri subsp. citri growth in planta. The ColR/ColS system also played important roles in bacterial biofilm formation in glass tubes and on leaf surfaces, lipopolysaccharide (LPS) production, catalase activity, and tolerance of environmental stress, including phenol, copper, and hydrogen peroxide. Furthermore, quantitative reverse transcription-PCR assays demonstrated that the ColR/ColS system positively regulated the expression of important virulence genes, including hrpD6, hpaF, the O-antigen LPS synthesis gene rfbC, and the catalase gene katE. Overall, our data indicate that the two-component regulatory system ColR/ColS is critical for X. citri subsp. citri virulence, growth in planta, biofilm formation, catalase activity, LPS production, and resistance to environmental stress.

Xanthomonas citri subsp. citri (syn. X. axonopodis pv. citri, X. citri, or X. campestris pv. citri) (5, 35, 42) is a Gram-negative bacterium which is the causal agent of citrus canker disease and has become one of the model organisms for studying plant-microbe interactions (4). Citrus canker is widely distributed in many tropical and subtropical citrus-growing areas and has been one of the most serious diseases of most commercial citrus cultivars, resulting in significant losses worldwide (37). In citrus-producing areas without citrus canker, X. citri subsp. citri is considered a quarantine pest and is subject to strict international regulation. Thus, citrus canker has a significant impact on national and international citrus markets and trade (14). X. citri subsp. citri is spread by rain splash and enters its hosts through openings such as stomata and wounds and forms distinctively raised, necrotic lesions surrounded by oily, water-soaked margins and yellow chlorotic rings on leaves, stems, and fruits (15). Progress has been made in understanding the pathogenicity and virulence mechanism of X. citri subsp. citri. It has been reported that the type III secretion system (TTSS) and the effector PthA are critical for the development of citrus canker symptoms (11, 39). Epiphytic survival and biofilm formation have been suggested to be important for X. citri subsp. citri before it invades intercellular spaces (34). The monofunctional catalase KatE (40) and the rpf cell-to-cell quorum-sensing system, which is involved in exopolysaccharide, endoglucanase, and protease production (36), also contribute to the development of canker symptoms.

Bacteria have evolved complicated regulatory mechanisms to coordinate their virulence factors to adapt to different environments. Among them, the two-component regulatory system is one of the basic stimulus-response-coupling mechanisms for bacteria to sense and respond to changes in many different environmental conditions (38). One of the intriguing phenomena uncovered by analysis of the X. citri subsp. citri genome is a large number of genes (n = 114) belonging to the two-component regulatory systems in this pathogen (7, 33). In a typical two-component system, the environmental signal is detected by the transmembrane kinase sensor, which can be autophosphorylated at a conserved histidine residue and then transfers the phosphoryl group to the cognate response regulator, which results in a conformational change in the regulatory protein and activates the regulatory effect. Even though the two-component system has been suggested to play crucial roles in multiple cellular functions of X. citri subsp. citri, only a few have been studied for their contributions to virulence, such as HrpG (46). In this study, we present evidence that the two-component system ColR/ColS plays important roles in the virulence, growth in planta, biofilm formation, lipopolysaccharide (LPS) production, catalase activity, and environmental stress resistance of the citrus canker pathogen X. citri subsp. citri by controlling the expression of important virulence genes, including hrpD6, hpaF, the O-antigen LPS synthesis gene rfbC, and the catalase gene katE.

MATERIALS AND METHODS

Bacterial strains, growth conditions, and plasmids.

The bacterial strains and plasmids used in this work are listed in Table 1. The Escherichia coli strain was cultured at 37°C in Luria-Bertani (LB) medium. Wild-type X. citri subsp. citri strain 306 (kindly provided by the Division of Plant Industry of the Florida Department of Agriculture and Consumer Services, Gainesville) was stored at −80°C. X. citri subsp. citri strains were cultured at 28°C in nutrient broth (NB; Difco, Detroit, MI), on nutrient agar (NA; Difco, Detroit, MI) plates, and in XVM2 medium (44). When required, growth media were supplemented with ampicillin (50 μg ml⁻¹), chloramphenicol (20 μg ml⁻¹), gentamicin (5 μg ml⁻¹), kanamycin (50 μg ml⁻¹), rifamycin (50 μg ml⁻¹), and spectinomycin (100 μg ml⁻¹).

TABLE 1.

Bacterial strains and plasmids used in this study

Strain or plasmid	Relevant characteristica	Reference or source
E. coli EC100D pir+	F−mcrA Δ(mrr-hsdRMS-mcrBC) φ80dlacZΔM15 ΔlacX74 recA1 endA1 araD139 Δ(ara-leu)7697 galU galK λ−rpsL nupG pir+ (DHFR)b	Epicentre
X. citri subsp. citri
306	Syn. X. axonopodis pv. citri strain 306; wild type Rifr	DPIc
256A10	colS::EZ-Tn5 derivative of strain 306; Rifr Kmr	This study
421E7	colS::EZ-Tn5 derivative of strain 306; Rifr Kmr	This study
386C6	colR::EZ-Tn5 derivative of strain 306; Rifr Kmr	This study
417E10	colR::EZ-Tn5 derivative of strain 306; Rifr Kmr	This study
306DSR	colS and colR double deletion mutant, derivative of strain 306; Rifr	This study
Plasmids
pCR2.1	Cloning vector, pUC ori, f1 ori, Kmr Apr	Invitrogen
pCR-colSR	2,290-bp SacI-BamHI fragment containing wild-type colR colS ligated into pCR2.1; Kmr Apr	This study
pCR-ΔcolSR	pCR2.1 containing colS colR double deletion; Kmr Apr	This study
pOK1	Suicide vector in X. citri subsp. citri; sacB sacQ r6k ori Spcr	21
pOK-ΔcolSR	Suicide vector containing colS colR double deletion on pOK1; Spcr	This study
pUFR053	E. coli-X. citri subsp. citri shuttle vector; IncW Mob+mob(P) lacZ+ Par+ Cmr Gmr	12
p53-colS	718-bp SacI-BamHI fragment containing wild-type colS gene ligated into pUFR053; Cmr Gmr	This study
p53-colR	1,595-bp SacI-BamHI fragment containing wild-type colR gene ligated into pUFR053; Cmr Gmr	This study
p53-colSR	2,290-bp SacI-BamHI fragment containing wild-type colR colS ligated into pUFR053; Cmr Gmr	This study

^aAp^r, Cm^r, Gm^r, Km^r, Rif^r, and Spc^r indicate resistance to ampicillin, chloramphenicol, gentamicin, kanamycin, rifamycin, and spectinomycin, respectively.

^bDHFR, dihydrofolate reductase.

^cDPI, Division of Plant Industry of the Florida Department of Agriculture and Consumer Services, Gainesville.

Identification of EZ-Tn5 insertion sites in X. citri subsp. citri mutants.

Rescue cloning was conducted to determine the insertion sites of the EZ-Tn5 transposon in the mutants according to the manufacturer's instructions (Epicentre, Madison, WI). Briefly, 1 μg genomic DNA of the mutant of interest was digested with PstI overnight at 37°C. The digested DNA was purified, self-ligated, and transformed into E. coli strain EC100D by electroporation. Transformant cells were spread onto LB plates containing kanamycin and kept overnight at 37°C. The plasmid DNA was extracted from kanamycin-resistant colonies and sequenced using primer R6KAN-2 RP-1 (Epicentre, Madison, WI). BLAST analysis and alignment with the genome sequence of strain 306 (7) were performed to determine the insertion site of the EZ-Tn5 transposon.

Construction of a colR colS double mutant strain and cloning of colR and colS.

To construct a colS colR double deletion mutant of strain 306, a 2,032-bp colR colS fragment was amplified using primers C8F1 (5′AAGAGCTCAAAGACAGCTTAACGAACGAGG3′ [underlining shows the SacI restriction enzyme site]) and C9R1 (5′ATGGATCCCCTGTAGAGTTTGAACGG3′ [underlining shows the BamHI restriction enzyme site]) (Fig. 1) and cloned into vector pCR2.1. The resulting plasmid, pCR2.1-colSR, was digested with SalI and self-ligated to discard the 424-bp SalI fragment, which resulted in plasmid pCR-ΔcolSR (Fig. 1). A 1,658-bp BamHI-XbaI fragment was cut from pCR-ΔcolSR and ligated into suicide vector pOK1, which was digested with the same restriction enzymes. The resulting vector, pOK-ΔcolSR, was used to construct a cols colR double deletion mutant based on the homologous recombination method as described elsewhere (28). The double deletion mutant was confirmed by PCR and sequencing.

FIG. 1.

Schematic diagram of colS and colR in the genome of X. citri subsp. citri strain 306. The single-headed arrows represent the locations and orientations of the genes in the genome. The EZ-Tn5 insertion positions in the mutants are indicated by inverted triangles. The construction of colR colS double mutant 306DSR and complementary plasmids p53-colS, p53-colR, and p53-colSR is described in Materials and Methods. The primers used to construct plasmids for gene complementation are shown on the left, and the SalI enzyme sites used to construct the double mutant are shown below the diagram.

The colR and colS genes were cloned by the PCR method with strain 306 genomic DNA as the template. The colR gene of strain 306 was amplified using primers C8F1 and C8R1 (5′AAGGATCCTCGCTCGTACTTTCAGGC3′ [underlining shows the BamHI restriction enzyme site]) (Fig. 1), and the 718-bp PCR product was digested with SacI and BamHI and ligated into the complementary vector pUFR053 (12) to construct plasmid p53-colR (Fig. 1). The colS gene was amplified using primers C9F1 (5′ AAGAGCTCGATGCCTGAAAGTACGAGCG3′ [underlining shows the SacI restriction enzyme site]) and C9R1 (Fig. 1), and the 1,595-bp PCR product was digested with SacI and BamHI and then ligated into vector pUFR053 to construct plasmid p53-colS (Fig. 1). The same method was used to construct complementary plasmid of p53-colSR, which contains both the colR and colS genes, by using primers C8F1 and C9R1 (Fig. 1). The three constructs obtained, p53-colR, p53-colS, and p53-colSR, were confirmed by sequencing.

Pathogenicity assays.

Pathogenicity assays were conducted as described previously (16). Briefly, fully expanded, immature grapefruit (Citrus paradise Macf. cv. Duncan) leaves were prepared in a quarantine greenhouse facility at the Citrus Research and Education Center, Lake Alfred, FL. The X. citri subsp. citri strains were cultured for 2 days on NA plates at 28°C and resuspended in sterile tap water. For the pathogenicity assays, a bacterial suspension (10⁸ CFU ml⁻¹) was inoculated onto the leaf surface by the spray method and a bacterial suspension (10⁶ CFU ml⁻¹) was injected into the intercellular spaces of leaves with a needleless syringe. All tests were repeated three times independently.

Growth assays in planta.

X. citri subsp. citri strains were cultured on NA plates for 2 days at 28°C. The bacterial cells were resuspended in sterile tap water at a concentration of 10⁶ CFU ml⁻¹ and used to infiltrate cv. Duncan grapefruit leaves with a needleless syringe (16). To evaluate the growth of strain 306 and its derivatives in the intercellular spaces of plants, leaf disks (leaf area, 1 cm² in diameter) from inoculated leaves (four leaves for each sample) were excised with a cork borer and then ground in 1 ml sterile tap water. The samples were serially diluted and plated on NA plates with appropriate antibiotics. The bacterial colonies were counted after incubation at 28°C for 3 days. In planta growth was measured in quadruplicate, and the assays were repeated three times independently.

Biofilm formation assays.

Assays of biofilm formation in glass tubes were performed as described previously (16). Briefly, X. citri subsp. citri strains were cultured on NA plates at 28°C for 2 days, and cells were resuspended in fresh XVM2 liquid medium to a final concentration of 10⁶ CFU ml⁻¹. One milliliter of the cell suspension was added to borosilicate glass tubes and incubated at 28°C for 48 h without shaking. The biofilm that formed on the glass tube was visualized by staining with 0.1% crystal violet and washed twice with sterilized water. The stain remaining in cells was dissolved in 95% ethanol and quantified by measuring the optical density at 590 nm using an Agilent 8453 UV-visible light spectrophotometer (Agilent Technologies, Inc., Santa Clara, CA). The average of four replicates was used for quantitative measurement. Assays of biofilm formation on leaf surfaces were conducted as described previously (13). Briefly, 20 μl of each bacterial suspension (10⁸ CFU ml⁻¹) was incubated on citrus leaves and the leaves were kept at 28°C for 24 h in a humidified chamber. Biofilm formation on the leaf surfaces was visualized using crystal violet staining. The assays were repeated three times.

LPS analysis.

X. citri subsp. citri strains were cultured overnight at 28°C in XVM2 liquid medium with shaking (220 rpm). Ten-milliliter samples of cultures in the exponential stage were collected, and LPS samples were extracted as described previously (30). LPS was separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and visualized using silver staining following the manufacturer's instructions (Bio-Rad Laboratories, Inc., Hercules, CA). Standard LPS from Salmonella enterica serovar Typhimurium was obtained from Sigma. The test was performed three times independently.

Assays of X. citri subsp. citri resistance to hydrogen peroxide, phenol, and copper.

Resistance of X. citri subsp. citri strains to hydrogen peroxide and phenol was tested by a MIC method. Briefly, X. citri subsp. citri strains were cultured to stationary phase (24 h) in XVM2 medium and dilutions of the bacterial cells were spread on NA plates supplemented with different concentrations of each reagent. The surviving colonies on the plates were counted after 3 days of incubation at 28°C. Copper resistance was determined by growing X. citri subsp. citri strains in liquid NB with different concentrations of CuSO₄. The growth of each strain was monitored by measuring the optical density at 600 nm after growth for 24 h at 28°C with shaking at 200 rpm. All experiments were performed in triplicate and repeated three times.

Cell extract preparation and catalase activity assays.

Cell extracts were prepared from X. citri subsp. citri strains which were cultured for 24 h in XVM2 medium to stationary phase. Bacteria were washed twice and resuspended in ice-cold 50 mM potassium phosphate buffer (pH 7.0) containing 1 mM phenylmethylsulfonyl fluoride and then sonicated for 2 min in an ice water bath with a sonicator (Misonix Sonicator 3000). The sonicate was clarified by centrifugation at 13, 000 × g for 20 min at 4°C. The concentration of soluble proteins was determined using a bicinchoninic acid protein assay kit (Novagen, San Diego, CA). The catalase activity of cell extracts was measured spectrophotometrically by following the rate of decrease in absorbance at 240 nm caused by the decomposition of hydrogen peroxide (2). The reaction was performed with 50 mM potassium phosphate buffer (pH 7.0) containing 16 mM hydrogen peroxide at 25°C. The coefficient of absorption at 240 nm for hydrogen peroxide was taken to be 43.6 M⁻¹ cm⁻¹ (17) to calculate the specific activity. One unit of catalase activity was defined as the amount of protein required to decompose 1 μmol H₂O₂ min⁻¹ under the assay conditions used.

RNA extraction and quantitative reverse transcription-PCR (QRT-PCR).

X. citri subsp. citri strains were cultured in XVM2 medium at 28°C with shaking at 200 rpm, and 1-ml samples of the bacterial cells were collected at 18 h after inoculation. RNA was stabilized immediately by mixing with 2 volumes of RNAprotect Bacteria Reagent (Qiagen, Valencia, CA) and extracted by using an RNeasy minikit (Qiagen, Valencia, CA). Contaminated genomic DNA was removed by treatment with a TURBO DNA-free kit (Ambion, Austin, TX). The concentration of RNA was determined with a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE) and adjusted to 50 ng ml⁻¹.

The RNA obtained was subjected to a one-step QRT-PCR assay with a 7500 fast real-time PCR system (Applied Biosystems, Foster City, CA) using a QuantiTect SYBR green RT-PCR kit (Qiagen, Valencia, CA). The gene-specific primers listed in Table 2 were designed to generate products of 100 to 250 bp based on the genome sequence of strain 306. The 16S rRNA gene was used as an endogenous control. The relative fold change in target gene expression was calculated by using the formula 2^−ΔΔCT (29).

TABLE 2.

Primers used for QRT-PCR

Gene	Product	Primer direction,a sequence (5′→3′)
16S rRNA	Ribosome component	F: CGCTTTCGTGCCTCAGTGTCAGTGTTGG
		R: GGCGTAAAGCGTGCGTAGGTGGTGGTT
hpaF	TTSS component	F: ACGCGCCTGTCCAATCTCA
		R: CGGCATGCGCAACTCGGTCAAATC
hrpD6	TTSS component	F: ATGTTCGATGCCATGACCGATACG
		R: TCATGCACGGCATTGAAGTCGTTG
katE	Monofunctional catalase	F: TCAATGAGAAAGGCGAGAGCACCT
		R: AGATCGCGACGGTGAAAGTCTTGA
rfbC	LPS O-antigen biosynthesis protein	F: ATCCATCACCAGCACCTGTTCGTA
		R: GAATCCGCCAATGGCATCGAAGTT
pthA	TTSS effector	F: TGGCGTCGGCAAACAGTGGTC
		R: TGCTCCGGGGTCAGGTTCAGG

^aF, forward; R, reverse.

RESULTS

Isolation of virulence-deficient mutants of X. citri subsp. citri strain 306.

To isolate virulence-deficient mutants of strain 306, an EZ-Tn5 transposon mutagenesis library was screened as described in our previous study (16). Four mutants, 256A10, 421E7, 386C6, and 417E10, which were virulence deficient following inoculation onto citrus leaves were isolated (Fig. 1).

The nucleotide sequences were obtained from the cloned DNA fragment flanking the EZ-Tn5 transposon in the isolated mutants. The insertion site of EZ-Tn5 in the mutants was determined using BLAST analysis and alignment with the genome sequence of strain 306 (7). Our results revealed that EZ-Tn5 was inserted into the colS (XAC3249) gene in mutants 256A10 (between nucleotides 247 and 248 downstream of the colS translation start site) and 421E7 (between nucleotides 621 and 622 downstream of the colS translation start site) and into the colR (XAC3250) gene in mutants 386C6 (between nucleotides 404 and 405 downstream of the colR translation start site) and 417E10 (between nucleotides 593 and 594 downstream of the colR translation start site) (Fig. 1). These two genes encode the two-component regulatory system ColR/ColS, which was reported to be important for root colonization by Pseudomonas fluorescens (9). Since the colS mutant 256A10 showed the same phenotypes as 421E7 and the colR mutant 386C6 showed the same phenotypes as 417E10, only the results obtained with 256A10 and 386C6 are presented here.

There are three pairs of paralogs of ColR/ColS systems in the whole genome of X. citri subsp. citri strain 306 (7); they are ColR_XAC0834/ColS_XAC0835, ColR_XAC1221/ColS_XAC1222, and ColR_XAC3249/ColS_XAC3250. Only ColR_XAC3249/ColS_XAC3250 was identified by our screening to be required for the pathogenicity of strain 306. The percent identities between the amino acid sequences of ColR/ColS of P. fluorescens WCS365 and ColR_XAC3249/ColS_XAC3250 of X. citri subsp. citri were of 57.7% for ColR and 27.9% for ColS, respectively. Similar results had been reported for the phytopathogen X. campestris pv. campestris 8004, in which only one (ColR_XC1049/ColS_XC1050) out of three pairs of the ColR/ColS two-component system was involved in pathogenicity (47). In addition, ColR_XC1049/ColS_XC1050 of X. campestris pv. campestris shows the highest levels of identity (100% and 95.8% for ColR and ColS, respectively) with ColR_XAC3249/ColS_XAC3250 of X. citri subsp. citri, while it shows only 50.0% and 26.1% identity with ColR_XAC0834/ColS_XAC0835 and 51.6% and 22.9% identity with ColR_XAC1221/ColS_XAC1222.

The two-component system ColR/ColS plays an important role in the virulence of X. citri subsp. citri strain 306.

To rule out the possibility of multiple EZ-Tn5 insertions in the genome and a polar effect of the transposon on the expression of a downstream gene(s) in our selected mutants, complementation assays were conducted for both colS and colR mutants. The complementary plasmids with intact colR or colS (Fig. 1) were transformed into the mutants, and they were assayed for virulence following pressure inoculation of bacterial suspensions into grapefruit leaves. The results showed that the pathogenicity of the colS or colR mutant could be restored to the wild-type level by plasmid p53-colS or p53-colR, respectively, but not by the empty vector (Fig. 2A).

FIG. 2.

The ColR/ColS system is involved in the virulence of X. citri subsp. citri strain 306. (A) Wild-type strain 306 and its derivatives (approximately 10⁶ CFU ml⁻¹) were inoculated into the intercellular spaces of fully expanded, immature grapefruit (C. paradise Macf. cv. Duncan) leaves by pressure infiltration with a needleless syringe. A representative leaf from four replicates was photographed at 12 dpi. The pathogenicity assays were repeated three times with similar results. (B) Bacterial cell suspensions (approximately 10⁸ CFU ml⁻¹) of wild-type strain 306 and its derivatives were inoculated onto fully expanded, immature grapefruit cv. Duncan leaves by the spray method. WT, wild-type strain 306/pUFR053; M1, colS mutant 256A10/pUFR053; C1, colS⁺ strain 256A10/p53-colS; M2, colR mutant 386C6/pUFR053; C2, colR⁺ strain 386C6/p53-colR; M3, colR colS mutant 306DSR/pUFR053; C3, colR⁺ colS⁺ strain 306DSR/p53-colSR. A representative leaf from four replicates was photographed at 14 dpi. The assay was repeated three times with similar results, and only one representative result is presented.

To further confirm the role of the ColR/ColS system in pathogenicity and understand the relationship between the sensor kinase ColS and the regulator ColR, a colS colR double deletion mutant, named 306DSR (Fig. 1), was constructed with wild-type strain 306. The double mutant showed the same virulence deficiency as the colS or colR mutant (Fig. 2A). The pathogenicity of the double mutant could be complemented to the wild-type level by p53-colSR, which contains both the colR and colS genes (Fig. 1).

To mimic the natural invasion conditions under which the canker pathogen enters its host by rain splash, wild-type strain 306 and its derivatives were inoculated onto citrus leaves using the spray method. Obvious symptoms were observed on the leaves 2 weeks postinoculation with wild-type strain 306. The colS, colR, and colR colS mutants produced no symptoms on the inoculated leaves. Symptom production by the colS or colR mutant could be restored to the wild-type level by complementary plasmid p53-colS or p53-colR, respectively, but not by the empty vector (Fig. 2B).

The ColR/ColS system contributes to X. citri subsp. citri growth in planta.

To check whether mutation of colS or colR affects X. citri subsp. citri growth in planta, wild-type strain 306 and the colR, colS, and colR colS mutants were inoculated into grapefruit leaves. As shown in Fig. 3, the populations of mutants 256A10 (colS), 386C6 (colR), and 306DSR (colS colR) were much smaller than that of wild-type strain 306 at each time point assayed. After declining by 2 orders of magnitude at 2 days postinoculation (dpi), the colS mutant maintained a low concentration of 10³ CFU cm⁻² in leaves during the remainder of the assay. The populations of the colR mutant and the colR colS double mutant decreased quickly after inoculation, and viable cells were not detectable at 3 and 4 dpi, respectively. However, the bacterial populations increased to a level similar to that of the colS mutant at 6 dpi (Fig. 3). The experiments were repeated three times with similar results.

FIG. 3.

The ColR/ColS system contributes to the growth of X. citri subsp. citri strain 306 in planta. Wild-type strain 306 and the mutant and complementary strains (colR [A], colS [B], and colR colS [C]) (approximately 10⁶ CFU ml⁻¹) were inoculated into grapefruit (C. paradise Macf. cv. Duncan) leaves using a needleless syringe. Bacterial cells were recovered from the inoculated leaves at different time points, and the values shown are means of three technical repeats ± standard deviations.

The growth of the colR or colS mutant was assayed in XVM2 medium. XVM2 medium was selected for this research because it was reported to mimic the environment of plant intercellular spaces (44). The results showed that the colR or colS mutant grew similarly to the wild-type strain in XVM2 medium and formed colonies similar to those of the wild-type strain on M9 minimal medium plates (data not shown), which suggested that the colR or colS mutant is not auxotrophic.

The ColR/ColS system positively regulates biofilm formation by X. citri subsp. citri strain 306.

Biofilm formation is a well-characterized, virulence-related trait observed in many bacteria and plays important roles in host-pathogen interactions (6). To investigate whether the two-component system ColR/ColS is involved in the regulation of biofilm formation, biofilm formation was assayed both in glass tubes and on citrus leaf surfaces. The colR, colS, and colR colS mutants showed a significant decrease in biofilm formation in glass tubes compared with that of the wild-type, and the complemented strains were restored to levels similar to that of the wild-type strain (Fig. 4A). Consistent with the observations in glass tubes, the colR, colS, and colR colS mutants were defective in biofilm formation on citrus leaf surfaces (Fig. 4B), suggesting that ColR/ColS is involved in the attachment of X. citri subsp. citri cells to citrus leaves. These results indicate that the ColR/ColS two-component system regulates biofilm formation by strain 306.

FIG. 4.

The ColR/ColS system regulates biofilm formation by X. citri subsp. citri strain 306. Biofilm formation in glass tubes (A) and on citrus abaxial leaf surfaces (B) was visualized using crystal violet staining. Biofilm formation in glass tubes was quantified by measuring the optical density at 590 nm after dissolution in 95% ethanol. Columns: 1, wild-type (WT) strain 306/pUFR053; 2, colS mutant 256A10/pUFR053; 3, colS⁺ strain 256A10/p53-colS; 4, colR mutant 386C6/pUFR053; 5, colR⁺ strain 386C6/p53-colR; 6, colR colS mutant 306DSR/pUFR053; 7, colR⁺ colS⁺ strain 306DSR/p53-colSR. All experiments were performed in quadruplicate and repeated three times with similar results, and only one representative result is presented. Means ± standard deviations are shown.

The ColR/ColS system is involved in LPS production.

LPS is the major component of the outer membrane of Gram-negative bacteria and serves as a structural permeability barrier against toxic plant defense compounds. To determine whether knockout of colS or colR affects LPS production by X. citri subsp. citri, SDS-PAGE analysis was performed with LPS extracted from different strains. Our results showed that the wild type produced O antigen, LPS core, and lipid A, while the LPS produced by the colS, colR, and colR colS mutants was remarkably different from that of the wild type. All three mutants were defective in O-antigen biosynthesis and produced less LPS than did the wild type (Fig. 5). The LPS production of the complemented strains was similar to that of the wild-type strain.

FIG. 5.

The ColR/ColS system regulates LPS synthesis by X. citri subsp. citri strain 306. The LPSs produced by strain 306 and its derivatives were extracted, subjected to SDS-PAGE analysis, and visualized by silver staining. Lanes: 1, wild-type (WT) strain 306/pUFR053; 2, colR mutant 386C6/pUFR053; 3, colR⁺ strain 386C6/p53-colR; 4, colS mutant 256A10/pUFR053; 5, colS⁺ strain 256A10/p53-colS; 6, colR colS mutant 306DSR/pUFR053; 7, colR⁺ colS⁺ strain 306DSR/p53-colSR. S, LPS standard from S. enterica serovar Typhimurium (10 μg; Sigma). Experiments were repeated three times with similar results, and the results of only one experiment are presented.

The ColR/ColS system contributes to the tolerance of X. citri subsp. citri to hydrogen peroxide, phenol, and copper.

The rapid decrease in the colR or colS mutant bacterial population in grapefruit leaves (Fig. 3) prompted an investigation of the role of the ColR/ColS two-component system in tolerance to environmental stress. The resistance of the colS and colR mutants to hydrogen peroxide, phenol, and copper was measured using a MIC method. As shown in Fig. 6, the MICs of all three reagents for both the colS and colR mutants were remarkably lower than those for the wild-type and complemented strains, indicating that ColR/ColS plays an important role in the hydrogen peroxide, phenol, and heavy metal tolerance of X. citri subsp. citri.

FIG. 6.

The ColR/ColS system is involved in the resistance of X. citri subsp. citri to phenol, hydrogen peroxide, and copper. X. citri subsp. citri strains were cultured in XVM2 medium for 24 h to stationary phase, and bacterial cells were diluted and plated on an NA plate supplemented with different concentration of phenol (A) and hydrogen peroxide (B). Bacterial colonies were counted after incubation at 28°C for 3 days. For copper resistance assay (C), X. citri subsp. citri cells were inoculated into NB containing different concentration of CuSO₄. Growth was measured by determining optical density (OD) at 600 nm after 24 h. All experiments were performed in triplicate and repeated three times with similar results, and representative results of only one experiment are presented. WT, wild type.

The ColR/ColS system positively regulates catalase activity in X. citri subsp. citri.

Sensitivity of colR and colS mutants to hydrogen peroxide (Fig. 6A) could result from a deficiency in LPS production (Fig. 5) or decreased catalase activity. To determine whether the ColR/ColS system is involved in the regulation of catalase activity in X. citri subsp. citri, the catalase activity of crude extracts of stationary-phase cells was investigated. Results showed that catalase activity of the colR and colS mutants was remarkably lower than that of the wild-type and complemented strains (Fig. 7). This result agreed with the MIC assay of hydrogen peroxide (Fig. 6A) and suggested that the ColR/ColS system positively regulates catalase activity in X. citri subsp. citri.

FIG. 7.

The ColR/ColS system regulates the catalase activity of X. citri subsp. citri strain 306. X. citri subsp. citri strains were cultured in XVM2 medium, and crude cell lysates prepared from stationary-phase cultures (24 h) were monitored for total catalase activity. The values presented are the means and standard deviations of three replicates. WT, wild type.

The ColR/ColS system regulates the expression of virulence genes.

To further understand how ColR/ColS regulates different traits as mentioned above, QRT-PCR assays were performed to monitor the expression profiles of five genes which were selected based on the phenotypes of interest. Since the two-component response regulator system was responsible for the regulation of target gene expression, the colR mutant was chosen to investigate the effect of the ColR/ColS system on gene expression in X. citri subsp. citri. For the QRT-PCR analysis, ΔΔC_T values of target genes for mutant 386C6 (colR) were obtained by using the wild type as a calibrator, and the n-fold change in gene expression was calculated using the formula 2^−ΔΔCT (29). As shown in Table 3, the TTSS genes hrpD6 and hapF were significantly downregulated in the colR mutant, and similar results were reported for X. campestris pv. campestris, i.e., that ColR/ColS positively regulated the expression of hrpC and hrpE operons (47), which might contribute to the loss of pathogenicity of the colR or colS mutant in the host (Fig. 2A). Additionally, consistent with the observations of the deficiencies in LPS production (Fig. 5) and catalase activity (Fig. 7), the expression of rfbC (encoding an O-antigen biosynthesis protein) and katE (encoding a monofunctional catalase) was significantly decreased in the colR mutant, respectively. The expression of pthA, which was demonstrated to encode a major virulence factor of X. citri subsp. citri to induce canker symptoms in citrus plants (11), was not significantly affected in the colR mutant compared to that in the wild type.

TABLE 3.

Comparison of virulence gene expression in the wild type and the colR mutant 386C6 cultured in XVM2 medium by QRT-PCR

Gene ID	Gene	Function of protein	ΔΔCT ± SDa	Fold change ± SDb
XAC0393	hpaF	TTSS component	3.6583 ± 0.6036	0.0842 ± 0.0375c
XAC0398	hrpD6	TTSS component	3.6077 ± 0.2814	0.0831 ± 0.0159c
XAC1211	katE	Monofunctional catalase	2.6493 ± 0.8975	0.1789 ± 0.0941c
XAC3598	rfbC	LPS O-antigen biosynthesis	1.8980 ± 0.3848	0.2750 ± 0.0767c
XACb0065	pthA	TTSS effector	0.0491 ± 0.3894	0.9905 ± 0.2711

^aThe mean ΔΔC_T was determined using three biological repeats.

^bThe n-fold expression change (mutant/wild type) in mutant 386C6 was calculated by using 2^−ΔΔCT.

^cStatistically significant difference (P < 0.05, analyzed by Student's two-sample t test).

DISCUSSION

This study has significantly advanced our understanding of the roles of the ColR/ColS two-component regulatory system. The ColR/ColS system was first identified in P. fluorescens and found to contribute to its colonization of plant roots (9). ColR/ColS was further reported to be involved in different biological responses, including the transposition of a transposon (18), phenol tolerance (24), and heavy metal resistance (19) of Pseudomonas putida. Furthermore, ColR/ColS was shown to be involved in virulence, to regulate hrpC and hrpE operon expression, to contribute to the tolerance of X. campestris pv. campestris to various stresses, including the antibiotics ampicillin and rifampin, the organic solvent phenol, the heavy metal salt CdSO₄, and the osmotic agent NaCl (47). In this work, the two-component regulatory system ColR/ColS of X. citri subsp. citri was shown to contribute to virulence, growth in planta, biofilm formation, LPS production, catalase activity, and resistance to phenol, hydrogen peroxide, and copper by regulating the expression of important virulence genes, including TTSS component genes hrpD6 and hpaF, the O-antigen LPS synthesis gene rfbC, and the catalase gene katE. Thus, the ColR/ColS system plays critical roles in bacterial adaptation to different environments.

The two-component regulatory system ColR/ColS is required for the virulence of X. citri subsp. citri. After inoculation into citrus leaves, the cell population of wild-type X. citri subsp. citri increased rapidly (Fig. 3) and the leaves showed obvious symptoms at around 5 dpi (Fig. 2A). However, the colR, colS, and colR colS mutants showed significantly reduced populations in the intercellular spaces of citrus leaves (Fig. 3) and no symptom was observed under the assay conditions used (Fig. 2A). This result, along with the restoration of the colR, colS, and colR colS mutants to the wild-type level by complementation with intact colR, colS, and colR colS, respectively (Fig. 2A and 3), strongly suggested that the two-component system ColR/ColS plays an important role in the multiplication and virulence of the X. citri subsp. citri pathogen in citrus plants. A similar result was reported for the phytopathogen X. campestris pv. campestris 8004, the causal agent of black rot disease in cruciferous crops, in which the expression of the hrpC and hrpE operons was positively regulated by the ColR/ColS system (47). In our study, QRT-PCR results showed that the expression of hpaF and hrpD6 was significantly downregulated in the colR mutant (Table 3). The hrpD6 gene belongs to the hrpE operon, which encodes conserved TTSS components in Xanthomonas and has been demonstrated to be required for the pathogenicity of X. campestris pv. vesicatoria in host plants and the hypersensitive response in nonhost plants (43). This result suggested a conserved cross talk between the ColR/ColS system and the TTSS in X. citri subsp. citri, which might contribute to the loss of pathogenicity of the colR, colS, and colR colS mutants. Meanwhile, the ColR/ColS system has no influence on the expression of the TTSS effector gene pthA, which has been proven to encode the major virulence determinant of X. citri subsp. citri to induce canker symptoms in the citrus host (11).

Besides the regulatory effect on TTSS genes, our data showed that the ColR/ColS system controls multiple traits for X. citri subsp. citri growth in planta. ColR/ColS positively regulates the catalase activity of X. citri subsp. citri (Fig. 7). Catalase is an antioxidant enzyme and a central component of the detoxification pathways that prevent hydroxyl radical formation by catalyzing the dismutation of hydrogen peroxide to water and oxygen (40). During interactions with potential pathogens, a key component of the initial plant defense responses is rapid production and accumulation of reactive oxygen species (ROS), primarily hydrogen peroxide and superoxide anions (1, 26). The accumulation of hydrogen peroxide at the attempted invasion site is extremely harmful, and pathogens have evolved a variety of enzymes capable of detoxifying ROS to accomplish a successful invasion. The rapid decline in the colR, colS, and colR colS mutant bacterial populations after inoculation into leaves (Fig. 3) indicated that the ColR/ColS system is required for X. citri subsp. citri to overcome the plant defense responses. Indeed, the tolerance of hydrogen peroxide was remarkably hampered by knockout of the ColR/ColS system in X. citri subsp. citri, and the complemented strains showed tolerance levels similar to that of the wild type (Fig. 6A), which demonstrated that the ColR/ColS system is required for X. citri subsp. citri to survive the toxicity of hydrogen peroxide. Importantly, we present evidence that ColR/ColS regulates the catalase activity of X. citri subsp. citri (Fig. 7). This was further supported by the QRT-PCR result that expression of katE, which was demonstrated to encode one of the most active monofunctional catalases required for full virulence of X. citri subsp. citri (40), was significantly reduced in the colR mutant (Table 3). To our knowledge, this is the first report that the two-component system ColR/ColS regulates catalase activity and katE expression in Xanthomonas.

Our data indicate that ColR/ColS positively regulates LPS production, which contributes to X. citri subsp. citri growth in planta. Remarkable changes in the LPS profiles of the colR, colS, and colR colS mutants, compared with that of the wild type, were observed (Fig. 5). The LPS amounts produced by the colR, colS, and colR colS mutants were remarkably lower than that of the wild type; and the LPS O-antigen part was defective by knockout of the ColR/ColS system, indicating the ColR/ColS system is important for the biosynthesis of LPS, particularly O-antigen LPS, in X. citri subsp. citri. Furthermore, the expression of rfbC, which encodes an O-antigen biosynthesis protein, was significantly downregulated in the colR mutant (Table 3). The results strongly suggested that the biosynthesis of LPS, particularly the LPS O antigen, was positively regulated by the ColR/ColS system. LPS is known as one of the major components of the bacterial outer membrane which functions as an important barrier to outside environments, as it restricts the movement of both hydrophilic and lipophilic compounds (31). This is consistent with the reduced tolerance of the colR, colS, and colR colS mutants to phenol, hydrogen peroxide, and copper (Fig. 6). Similar effects of the ColR/ColS system on resistance to phenol and heavy metals have been reported in P. putida (19, 24, 32) and X. campestris pv. campestris (47). Thus, ColR/ColS is involved in maintaining the structural integrity and function of the outer membrane (23, 24) and contributes to X. citri subsp. citri growth in planta. LPS was also reported to play important roles in the attachment of bacteria to surfaces and each other (8). O antigen, together with LPS core and lipid A, constituting the LPS structures in many bacteria, has been reported to be required for attachment and biofilm formation in P. fluorescens (45), Stenotrophomonas maltophilia (20), and Azospirillum brasilense (25). Similarly, mutation of the ColR/ColS system decreased biofilm formation by X. citri subsp. citri in glass tubes and on leaf surfaces (Fig. 4), which might contribute to the loss of virulence of the colR, colS, and colR colS mutants inoculated onto citrus leaves by the spray method (Fig. 2B). These results agree with our recent finding that mutation of rfbC of X. citri subsp. citri resulted in deficiencies in LPS O-antigen biosynthesis, biofilm formation, resistance to environmental stress, and virulence in planta (27). In addition, LPS has been demonstrated to be involved in the pathogenicity of many plant-pathogenic bacteria (3, 10, 22, 41). However, the rfbC mutant showed virulence deficiency in planta only when inoculated by spraying instead of pressure infiltration, while ColR/ColS mutants were not virulent when inoculated by pressure infiltration or spraying (Fig. 2). The difference in virulence between the rfbC mutant and the colR, colS, and colR colS mutants is due mainly to the much broader involvement of the ColR/ColS system in different virulence traits, including TTSS and catalase, besides LPS biosynthesis.

Overall, we have significantly advanced the knowledge of the two-component regulatory system ColR/ColS in virulence; biofilm formation; resistance to phenol, hydrogen peroxide, and copper; and growth in planta of X. citri subsp. citri. ColR/ColS contributes to virulence and growth in planta of X. citri subsp. citri by controlling TTSS, LPS production, and catalase activity by regulating the expression of important virulence genes, including hrpD6, hpaF, the O-antigen LPS synthesis gene rfbC, and the catalase gene katE of X. citri subsp. citri. To understand the broad effects of ColR/ColS, it is necessary to further study the regulatory cascade and downstream genes of colR/colS.