Abstract Background Xanthomonas citri subsp. citri is the causal agent of citrus canker, which causes substantial losses in citrus production. Here, we report the role of a polyketide cyclase (PKC) on the virulence in X. citri subsp. citri. Methods The structure of PKC was precisely predicted using Alphafold3. Promoter GUS fusion constructs and real-time quantitative reverse transcription (qRT-PCR) were employed to study the pattern of expression of the polyketide gene. A deletion mutation was created to explore the role of PKC in virulence and metabolic change. Results The PKC was determined to have a signal peptide, a START/RHO_alpha_C/PITP/Bet_v1/CoxG/CalC (SRPBCC) domain, and a GyrI-like small molecule binding domain. The expression of the PKC gene was induced in planta, as well as under stress by CuSO[4] and SDS. An in-frame deletion mutation resulted in a loss of virulence on the citrus hosts, which was restored by the SRPBCC domain. Furthermore, there as a remarkable reduction in the expression of type III genes, such as hrpG and hrpX. In the mutant carrying the pkc deletion, ketoleucine and acetone cyanohydrin were downregulated, and four metabolites, including d-ribose, creatine, polyoxyethylene dioleate, and cohibin C, were upregulated. Conclusions The overall data indicate that the PKC affects bacterial virulence by modulating the type III secretion system, possibly through the biosynthesis of particular metabolites. Supplementary Information The online version contains supplementary material available at 10.1186/s12866-025-03749-3. Keywords: Xanthomonas citri subsp. citri, Polyketide cyclase, Virulence, Type III gene, Citrus canker Introduction The Gram-negative bacterium Xanthomonas citri subsp. citri is the causal pathogen of citrus canker disease worldwide and one of the major causes of economic losses in citrus production [[32]1]. Citrus canker has characteristic disease symptoms, such as hypertrophy and hyperplasia. It has become one of the most important bacterial diseases because most commercial citrus varieties are susceptible [[33]2, [34]3]. Xanthomonas citri subsp. citri disperses to healthy host tissue via water splashed off infected plants and then directly enters the hosts through wounds or stomatal openings [[35]4]. The pathogen replicates rapidly within the apoplast of a compatible host to achieve a sufficient density and elicits its characteristic hypertrophy and hyperplasia symptoms [[36]5]. Xanthomonas citri subsp. citri possesses a type III secretion system (T3SS) responsible for injecting a set of effectors into the plant cells, especially the disease effector PthA4 responsible for activating the canker susceptibility gene CsLOB1 [[37]6, [38]7]. The biosynthesis of T3SS is controlled by a hrp gene cluster located on the chromosome, which is composed of six operons, hrpA–F. As in other pathogenic species of Xanthomonas, the expression of the hrp gene cluster is positively regulated by the OmpR family regulator hrpG and the AraC-type transcriptional activator hrpX [[39]8]. Furthermore, the expression of the hrp gene is induced in planta, as well as in a hrp-inducing medium XVM2 that mimics the bacterial milieu in the plant intercellular spaces [[40]9, [41]10]. It has been reported that a Lon protease plays a critical role in modulating the levels of the HrpG regulator. This results in high degradative activity on HrpG when cultured in nutrient-rich media, which results in the low expression of T3SS [[42]11]. The disruption of T3SS blocks the injection of effectors into the host cells and thus, impairs the induction of the hypersensitive response (HR) of nonhosts and the pathogenicity of susceptible hosts [[43]8]. Polyketides are a large class of natural products that are widely synthesized in bacteria, fungi, plants, and animals. The polyketide synthases (PKSs) responsible for the biosynthesis of polyketides are classified into three types according to the structural organization of their functional domains [[44]12, [45]13]. The type II PKS proteins that catalyze the cyclization of linear poly-β-ketone intermediates to cyclized products, generate aromatic polyketides with versatile pharmacological properties, including tetracycline, doxorubicin, nogalamycin, and Tcm C [[46]14]. The type II PKSs are multi-enzyme complexes mainly found in bacteria that are composed of five to eight discrete monofunctional proteins [[47]15]. During the biosynthesis of polyketides, these proteins form transient complexes in the solution, with the acyl carrier protein (ACP) serving as a central hub [[48]16]. AknH and SnoaL2 are small PKCs that catalyze the biosynthesis of aclacinomycin and nogalamycin in Streptomyces, respectively [[49]17, [50]18]. An olivetolic acid cyclase from Cannabis sativa is the plant PKC involved in the biosynthesis of cannabinoids [[51]19]. For most PKCs, their catalytic mechanisms, associated intermediates, and biological functions remain unclear. Since the first complete genome sequence of X. citri subsp. citri was released, numerous multiple studies have reported the virulence genes from this pathogen based on a genome-wide analysis [[52]20, [53]21]. The necessity of PKC for citrus canker development was determined in both X. citri subsp. citri 306 and 29 − 1 strains using Tn5-based random mutagenesis [[54]22, [55]23]. In this study, a non-polar deletion mutant of the PKC gene was created to verify its necessity for virulence in hosts and induction of the HR in nonhost plants. Based on a sequence analysis, the conserved SRPBCC and GyrI binding-domains were evaluated for their contribution for virulence. In addition, the levels of expression of the type III genes and changes in the biosynthesis of metabolites were studied in the deletion mutant, thus, significantly advancing the current understanding of the role of PKC. Materials and methods Bacterial strains, plasmids and growth conditions The bacterial strains and plasmids used in this study are listed in Supplement F1 (Table [56]S1). The X. citri subsp. citri strains were cultivated in nutrient-rich broth (NB) media or NB with 1.5% agar (NA) at 28 °C [[57]24]. Strains of Escherichia coli were cultured in LB media at 37 °C. Antibiotics were applied at the following concentrations: kanamycin (Km) at 50 µg mL^-1, spectinomycin (Sp) at 50 µg mL^-1, and gentamycin (Gm) at 10 µg mL^-1. Mutant construction and complementation analysis Based on the genome sequence of X. citri subsp. citri strain 29 − 1 (No. [58]CP004399), the specific primers 12301.F and 12301.R were designed to amplify a 280 bp DNA fragment upstream of the PKC gene. A 468 bp DNA fragment containing 105 bp PKC gene 3′ terminus was amplified using the primer set 12302.F/12302.R (Supplement F1, Table [59]S2). The two fragments were individually cloned into the suicide vector pKMS1, which generated pKMS-XAC1230 (Supplement F1, Table [60]S1). The recombinant plasmid was introduced into X. citri subsp. citri 29 − 1 to create deletion mutants by two steps of homologous recombination [[61]24]. After selection on NA plates supplemented with 10% sucrose, the colonies sensitive to kanamycin were identified by PCR with the primer pair 12301.F/12302.R. This produced a 748 bp PCR product with a deletion of the 1062 bp coding sequence. The deletion of targeted fragments was confirmed by sequencing the PCR products obtained from mutants. The primer pair CXAC1230.F/CXAC1230.R was used to amplify a 1447 bp DNA fragment that contained the PKC gene and its promoter region (Supplement F1, Table [62]S2). The PCR product was cloned into the vector pBBR1MCS-5 to obtain the construct pBB-XAC1230. The pBB-XAC1230 construct was transformed into the deletion mutant for complementation analysis. Complementation with the PKC mutants was additionally studied by designing specific primers to combine with CXAC1230.F to generate constructs expressing 1–100, 1–180, and 1–255 amino acids of the PKC protein. The corresponding generated constructs M[1 − 255], M[1 − 180], and M[1 − 100] were transformed into the deletion mutant to evaluate their ability to restore pathogenicity. Expression of the PKC gene under stress conditions The approximately 500 bp DNA sequence upstream of the PKC gene was retrieved from the 29 − 1 genome (No. [63]CP004399). The putative promoter sequence was predicted using Softberry ([64]http://www.softberry.com/). According to the promoter location, the primer pair PXAC1230.F/PXAC1230.R was used to amplify a 280 bp DNA fragment composed of the promoter region. The PCR product was inserted into the pRG960 vector for fusion with GUS at the PstI and BamHI sites (Supplement F1, Table [65]S2). The generated PXAC1230-GUS construct was electrotransformed into 29 − 1 to analyze the activity of its promoter. The 29 − 1 and 29 − 1/PXAC1230-GUS strains were cultured in NB broth at 28 °C for 36 h and sub-cultured (1:100) in 5 mL of fresh NB broth until their OD[600] ≈ 1.0. The cultures were supplied with 1% NaCl, 5% sorbitol, 0.008% SDS, and 0.2 mM CuSO[4]. At 2 h post-incubation, the 29 − 1 and 29 − 1/PXAC1230-GUS cells were subjected to a real-time quantitative reverse transcription PCR (qRT-PCR) analysis to evaluate the levels of transcripts of both PKC and gusA. The experiments were repeated three times. Pathogenicity, hypersensitive response, and replication assays in planta The X. citri subsp. citri strains were cultured in NB broth for 48 h and sub-cultured (1:100) in 5 mL of fresh NB broth until the OD[600] reached 1.0. The cultured cells were suspended in sterile distilled water to a final concentration of 10^7 CFU mL^− 1 (OD[600] = 0.03). The bacterial suspensions were infiltrated into grapefruit (Citrus × paradisi) leaves with a needleless syringe for pathogenicity assay. Disease symptoms were scored and documented as images at 5 d post-inoculation [[66]25]. For the growth assay in planta, 0.8-cm-diameter leaf discs were collected at 4 d post-infiltration. The discs were completely ground in 1 mL of sterile ddH[2]O. Serial dilutions of the suspension were spread on NA plates, and the individual colonies were recorded to determine the CFUs per cm^2 leaf. The SD was calculated based on the colony counts of four replicates. The HR response was analyzed by infiltrating the cell suspension into tomato (Solanum esculentum) leaves. The reaction was viewed 24 h post-inoculation. The tests for pathogenicity and the HR were repeated three times, and the replication assay in planta was repeated four times. Analysis of the expression of the gene for T3SS The expression of the T3SS gene was analyzed by separately electrotransforming PhrpG-GUS and PhrpX-GUS into mutant and complemented strains. The wild type 29 − 1, mutant, and complemented strains that harbored PhrpG-GUS or PhrpX-GUS were cultured in NB broth until the OD[600] reached 1.0. After centrifugation (5000 × g for 10 min), the cells were re-suspended with the hrp-inducing medium XVM2 [[67]26]. A 2 µL volume of cells was spotted onto 1.5% agar XVM2 plates supplied with 20 µg/ml p-nitrophenol-β-D-glucuronide (X-gluc), and the activity of GUS was determined via colony color at 3 d post-inoculation. The levels of gusA transcripts were quantified by incubating the cells in XVM2 liquid media for 6 h and then subjected to qRT-PCR. The levels of the transcripts of hrpG, hrpX, hrpD6, and hrcV in wild type 29 − 1, deletion mutant, and complemented strain were studied after cell suspensions (10^7 CFU mL^− 1) were infiltrated into citrus leaves. At 2 d post-inoculation, qRT-PCR was conducted to detect the levels of transcripts of the four hrp genes. Real-time quantitative reverse transcription PCR RNA was extracted from the X. citri subsp. citri cells using an RNAprep Pure Kit for Cells/Bacteria (Tiangen Biotech, Beijing, China). A Plant RNA Kit (Omega Bio-Tek, Norcross, GA, USA) was used to isolate RNA from the grapefruit leaves. A total of 2 µg RNA was reverse-transcribed into single-stranded cDNA synthesized with HiScript QRT SuperMix (Vazyme, Nanjing, China). All the primers used for qRT-PCR are listed in Supplement F1 (Table [68]S2). The PCR thermal cycling conditions were as follows: 95 °C for 5 min, followed by 40 cycles of denaturation at 95 °C for 5 s and annealing at 60 °C for 20 s. The level of expression of gyrA was evaluated as the internal control. All the experiments included three biological replicates, and each had three technical replicates. Identification of the differentially expressed metabolites using UPLC-MS/MS The metabolomic analysis was conducted by Shanghai BioTree Biotech Co., Ltd. (Shanghai, China). The X. citri subsp. citri cells were cultured in NB broth and resuspended in XVM2 liquid for 6 h of incubation. The cells were then centrifuged and suspended in 1 mL of extract solution (acetonitrile: methanol: water, 2:2:1 [v:v:v]) to extract the metabolites. Liquid chromatography with tandem mass spectrometry (LC-MS/MS) analyses were performed using an ultra-high performance liquid chromatography (UHPLC) system (Vanquish, Thermo Fisher Scientific, Waltham, MA, USA) with a UPLC BEH Amide column (2.1 mm × 100 mm, 1.7 μm) coupled to a Q-Exactive HFX mass spectrometer (Orbitrap MS, Thermo Fisher Scientific). The MS/MS spectra were acquired in the information-dependent acquisition (IDA) mode using acquisition software (Xcalibur, Thermo Fisher Scientific). In this mode, the ESI source conditions were set as follows: sheath gas flow rate, 50 Arb; Aux gas flow rate, 10 Arb; capillary temperature, 320 ℃; full MS resolution, 60,000; MS/MS resolution, 7,500; collision energy, 10/30/60 in NCE mode; spray voltage, 3.5 kV (positive ion) or -3.2 kV (negative ion). The raw data were converted to the mzXML format using ProteoWizard and processed with an in-house program, which was developed using R and based on XCMS, for peak detection, extraction, alignment, and integration. An in-house MS2 database (BiotreeDB) was then applied to annotate the metabolites. The p- and fold-change values were set to 0.05 and 2.0, respectively, to screen for the differentially accumulated metabolites. In addition, commercial databases, including KEGG ([69]http://www.genome.jp/kegg/) and MetaboAnalyst ([70]http://www.metaboanalyst.ca/), were used for the pathway enrichment analysis. Quantification and statistical analysis The statistical analyses were conducted using GraphPad Prism (GraphPad Software, San Diego, CA, USA), including two sample Student’s t-tests and a one-way analysis of variance (ANOVA) with a Tukey’s multiple comparison test at the 95% level. The values are presented as the mean ± SD. Results Structural traits of PKC in Xanthomonas In the reference genome sequence of X. citri subsp. citri strain 306 that was first released (No. [71]AE008923), two predicted protein products were annotated as polyketide synthase. The first one, FrnE, is encoded by XAC1331, which corresponds to a DsbA-like protein that contains a CXXC motif. The second one is annotated as a PKC encoded by XAC1230. The XAC1230 gene has 1167 bp and encodes a 388 aa protein. After the prediction of the 3-D structure in AlphaFold3 ([72]https://golgi.sandbox.google.com/), the XAC1230 protein was found to have three distinctive domains. From residue position 1 to 27 aa is a putative signal peptide; from residue position 28 to 180 is a conserved SRPBCC domain, and from residue position 200 to 365 is a GyrI-like substrate-binding domain with a β-barrel fold (Fig. [73]1A). The SRPBCC is composed of seven β-sheets and one α-helix. The GyrI domain is composed of six β-sheets and two α-helices, which forms a triangle structure. This domain, from positions 273 to 295, forms a curve out from the main body of the triangle structure (Fig. [74]1B). The XAC1230 homolog in the 29 − 1 strain is annotated as XAC29_06185. Notably, XAC1230 is only found in species of Xanthomonas, including plant pathogens. Comparison with the homologs in 14 plant pathogenic species of Xanthomonas, including X. oryzae pv. oryzae, X. campestris pv. campestris, and X. alfalfa, revealed amino acid sequence similarities of over 95% (Supplement F2, Figure [75]S1). The most genetic diversity occurred near position 286, where additional amino acid residues were found among the homologs. They corresponded to the curve out from the triangle structure in the GyrI domain (Fig. [76]1B). Fig. 1. [77]Fig. 1 [78]Open in a new tab Sequence analysis of polyketide cyclase in Xanthomonas citri subsp. citri. (A) Schematic diagram of polyketide cyclase. The position of a SRPBCC domain is indicated in brown, and a putative small molecule-binding domain is indicated in red. (B) Predicted three-dimensional structure of polyketide cyclase. The structure was analyzed using AlphaFold3 ([79]https://golgi.sandbox.google.com/). The superposition-free score of predicted Local Distance Difference Test (pLDDT) is shown upper Expression of PKC is environmentally induced To determine the pattern of expression of XAC1230, its promoter region was first predicted to be located at approximately − 45 bp in front of the coding sequence using Softberry ([80]http://www.softberry.com/). The promoter has a putative − 10 box (-TGCTAGGAT-) and a putative − 35 box (-TGAACA-) (Fig. [81]2A). After the promoter region was fused with GUS in the pRG960 vector, the wild type 29 − 1 strain that harbored this fusion construct was cultured under diverse conditions. When inoculated on citrus plants, the promoter–driven transcripts of gusA increased 80-fold relative to culture in NB liquid media (Fig. [82]2B). Moreover, the transcripts of XAC1230 increased 2.8-fold (Fig. [83]2C). There was no difference in the expression of XAC1230 when NaCl or sorbitol was added to the NB media to create an osmotic environment. The addition of CuSO[4] or SDS resulted in a significant increase in the expression of XAC1230. Under the stress conditions imposed by 0.2 mM CuSO[4], the level of XAC1230 transcripts and the transcription of the promoter-driven gusA increased by 2.2- and 4.6-fold, respectively (Fig. [84]2D, E). The levels of XAC1230 and gusA transcripts were increased by 8.4- and 45.6-fold, respectively, when subjected to the stress of 0.008% SDS (Fig. [85]2D, E). Fig. 2. [86]Fig. 2 [87]Open in a new tab The induced expression of the polyketide cyclase gene. (A) Promoter region of polyketide cyclase. The predicted − 10 and − 35 box elements are indicated with underlining. The putative translation initiation site is shown by an arrow. (B) qRT-PCR analysis of the induced gusA gene. This analysis was conducted in the wild type strain that harbored a polyketide cyclase gene promoter GUS fusion construct. (C) qRT-PCR analysis of the induced expression of polyketide cyclase in planta. The level of expression was assayed in the wild type cells cultured in nutrient-rich broth (NB) and cells inoculated in citrus plants at 3 d post-inoculation. The level of expression in the NB was set as 1, and the levels of expression in planta were relative to that value. (D) qRT-PCR analysis of the levels of transcripts of polyketide cyclase under diverse stress conditions. The cultured Xanthomonas citri subsp. citri cells were incubated in NB media and then resuspended in NB broth supplemented with various nutrients for 2 h. The level of expression in NB was set as 1, and the levels of expression in different stress conditions were relative to that value. (E) qRT-PCR analysis of the induced gusA gene under diverse stress conditions. This performance was assayed in the wild type strain that harbored the polyketide cyclase gene promoter GUS fusion construct. All the experiments were repeated three times. Error bars represent the standard deviation from three independent experiments (Student’s t-test, **p < 0.001) PKC is necessary for pathogenicity on citrus A deletion mutation was constructed using PCR to amplify a 280 bp left segment and 468 bp right segment, which was then cloned separately into the suicide vector pK18mobsacB. The in-frame deletion mutation of the PKC gene was created based on the sequence of 29 − 1 as the original strain through homologous recombination. By using the forward primer for the left segment and reverse primer for the right segment, a 748 bp PCR product was amplified from the mutant, which included a deletion of the 1062 bp coding region of PKC (Supplement F2, Figure [88]S2; Supplement F3, Figure [89]S1). In comparison with the canker symptoms caused by WT 29 − 1 at 5 d post-infiltration inoculation, the mutant caused no symptoms on the leaves (Fig. [90]3A). There was no difference in the growth of the mutant compared to the WT in nutrient-rich media (Fig. [91]3B). At 4 d post-inoculation, the mutant cells collected from the citrus leaves were reduced by 60-fold compared with those from the WT (Fig. [92]3C). Furthermore, the mutant did not induce a hypersensitive reaction on nonhost plants (Fig. [93]3D). Fig. 3. Fig. 3 [94]Open in a new tab The phenotypes of a deletion mutant of the polyketide cyclase gene. (A) The loss of virulence on citrus plants. The wild type Xanthomonas citri subsp. citri 29–1, mutant, and complemented strains (10^7 CFU/mL) were infiltrated into citrus leaves to analyze their induction of canker symptoms. The canker symptoms were recorded at 5 d post-inoculation. The areas of infiltration are indicated by dotted lines. (B) Growth of the mutant in medium. Prior to initiation of the assay, the cultured cells were prepared to OD[600] = 0.05 in nutrient broth media. The growth was assayed every 6 h during cultivation at 28 °C. Columns labeled with different letters indicate a significant difference among the means (ANOVA, P < 0.05). (C) Growth of the mutant in citrus leaves. Cell cultures (10^7 CFU/mL) were infiltrated into the citrus leaves. At 4 d post-inoculation, bacteria were recovered from the leaves and counted on nutrient-rich agar plates. Error bars represent the SD from four independent experiments (Student’s t-test, **p < 0.001). (D) The hypersensitive response of tomato (Lycopersicon esculentum). Cell cultures (10^7 CFU/mL) were infiltrated into tomato leaves. The hypersensitive response was scored at 24 h post-inoculation Deletion of PKC impairs the expression of T3SS genes To evaluate the expression of the type III secretion system in the XAC1230 mutant, the hrpG and hrpX promoter GUS fusion constructs (PhrpG-GUS, PhrpX-GUS) were introduced into the mutant to examine the promoter-driven GUS activity. The GUS activity was first assayed on hrp-inducing medium XVM2 plates supplemented with 25 mM X-gluc. No GUS activity was observed in the mutant colonies after cultivation on the plates (Fig. [95]4A). The activity of GUS was quantified after culturing the cells in liquid XVM2. The GUS activity driven by the hrpG promoter was reduced by 85%, and the GUS activity driven by the hrpX promoter was reduced by 73% compared with the WT (Fig. [96]4B). This demonstrated that the expression of hrpG and hrpX was impaired in the PKC mutant. To verify the expression of the type III genes, qRT-PCR was conducted to assess the transcription of hrpG, hrpX, hrpD6, and hrcV in planta. The levels of transcripts of the four genes were all substantially downregulated in the mutants (Fig. [97]4C). Fig. 4. [98]Fig. 4 [99]Open in a new tab The necessity of polyketide cyclase for the expression of type III genes. (A) Expression of hrpG and hrpX promoter (PhrpG, PhrpX) GUS fusion constructs in mutant 29 − 1 on XVM2 plates. Wild type 29 − 1 (WT), mutant and complemented strains that harbor hrpG or hrpX promoter GUS fusion constructs were cultivated on XVM2 plates supplemented with 25 mM X-gluC. (B) Quantification of the activity of the hrpG or hrpX promoter GUS fusion construct in the mutant. The activities of GUS were quantified in the cells that harbored PhrpG-GUS or PhrpX-GUS fusion construct transformants cultured in hrp-inducing medium XVM2. The value was measured with p-nitrophenol-β-d-glucuronide as substrate and recorded as nanomoles of product per min per OD unit. Columns labeled with different letters indicate a significant difference among the means (ANOVA, P < 0.05). (C) qRT-PCR analysis of the levels of transcripts of hrpG, hrpX, hrpD6, and hrcV. RNA was extracted from Xanthomonas citri subsp. citri cells inoculated onto citrus plants at 3 d post-inoculation. The level of expression in the WT cells was set as 1, and the levels of expression in the mutant and complemented strains were calculated relative to that value. Error bars represent the SD from three independent experiments (Student’s t-test, **p < 0.001) Deferentially expressed metabolites in the PKC gene mutant Based on UHPLC-QE-MS/MS, 3071 individual peaks were observed in the WT vs. mutant comparison after data filtering, missing value recoding, and normalization. In the negative ion mode, three metabolites, including proline, acetone cyanohydrins, and Cer(d18:0/16:0), were downregulated, and 14 metabolites were upregulated in the mutant. In the negative ion mode, two metabolites (proline and ketoleucine) were downregulated, and 10 metabolites were upregulated in the mutant (Supplement F2, Figure [100]S3). Compared to the complemented strain, three metabolites were downregulated, and five metabolites were upregulated in the mutant in the negative ion mode. In the positive ion mode, six metabolites were downregulated, and seven metabolites were upregulated in the mutant (Supplement F2, Figure [101]S4). Among those differentially expressed metabolites, six metabolites had the same pattern of expression in the mutant compared to the WT and complemented strain. Among these metabolites, ketoleucine and acetone cyanohydrin were downregulated, while d-ribose, creatine, polyoxyethylene dioleate, and cohibin C were upregulated (Fig. [102]5). In the KEGG enrichment analysis, d-ribose was identified to be part of the [103]C00121 pathway involved in the biosynthesis of organic oxygen compounds, while creatine was identified to be part of the [104]C00300 pathway involved in the biosynthesis of carboxylic acids and their derivatives. Fig. 5. [105]Fig. 5 [106]Open in a new tab Differential metabolites in the polyketide cyclase mutant. The data show the changes in six independent repeats. NEG, negative ion mode; POS, positive ion mode. The horizontal axis in the figure represents different experimental groups, while the vertical axis represents the differential metabolites compared in each group. Red and blue colors represent relative levels of expression that are high and low, respectively The SRPBCC domain is sufficient for the restoration of pathogenicity Because the 388-aa PKC has two distinctive substrate-binding domains, we studied the role of the SRPBCC or GyrI domains individually in pathogenicity. A series of mutants were constructed to evaluate this. The M[1 − 255] construct had a 133 aa deletion at its C-terminus, which resulted in the partial deletion of the GyrI domain. M[1 − 180] had a 208 aa deletion at its C-terminus, which resulted in a complete deletion of the GyrI domain. The mutant M[1 − 100] expressed a protein with a partial SRPBCC domain (Fig. [107]6A). The truncated mutants M[1 − 255], M[1 − 180], and M[1 − 100] were expressed in pBBR1MCS-5 for complementation analysis. The M[1 − 255] and M[1 − 180] constructs that expressed the entire SRPBCC domain restored the pathogenicity of the mutant, but the canker symptoms caused by the M[1 − 180] construct were not as severe as those of the wild type. In contrast, the M[1 − 100] construct was unable to restore pathogenicity (Fig. [108]6B). Consistent with these pathogenicity findings, M[1 − 255] restored bacterial replication in planta, while M[1 − 100] did not. The replication was slightly restored by M[1 − 180] but showed no statistical difference from the mutant (Fig. [109]6C). Thus, the SRPBCC domain was the key domain required for PKC to affect pathogenicity. Fig. 6. Fig. 6 [110]Open in a new tab Pathogenicity of the mutant restored by the SRPBCC domain. (A) Schematic diagram of the complementary construction expressing the truncated polyketide cyclase gene. The resulting recombinant constructs were introduced into the mutant to generate CM[1 − 255], CM[1 − 180], and CM[1 − 100]. (B) Symptoms of disease on citrus plants caused by different complemented strains. The cultured Xanthomonas citri subsp. citri cells (10^7 CFU/mL) were infiltrated into citrus leaves, and the canker symptoms were recorded 5 d post-inoculation. The areas of infiltration are indicated by dotted lines. (C) Bacterial growth in the citrus leaves. A volume of 10^7 CFU/mL of X. citri subsp. citri cells were infiltrated into citrus leaves. At 4 d post-inoculation, bacteria were recovered from the leaves and counted on nutrient-rich agar plates. Error bars represent the SD based on four independent observations. Labeling with different letters indicates significant differences between the means. The values are the mean ± SD (n = 4 biological replicates; ANOVA, P < 0.05) Discussion Based on Tn5 random insertion mutagenesis, the necessity of PKC for virulence was identified from the first strain of Xanthomonas citri subsp. citri with a sequenced genome, 306, as well as the Chinese strain 29 − 1 [[111]22, [112]23]. To explore the molecular mechanism mediated by PKC that gives rise to virulence, a deletion mutant was created in this study. The expression of the type III secretion system was found to be remarkably affected in the mutant. Furthermore, a change in the metabolites was found in the mutant relative to the WT strain. These findings support the role of PKC in directing the biosynthesis of certain metabolites required for the activation of T3SS. This not only highlights the critical role of PKC for bacterial virulence but also suggests potential metabolites that affect T3SS during the X. citri subsp. citri infection. The PKC shares high similarity with its homologs in other Xanthomonas species, including several model plant pathogens. Based on an analysis of the amino acid sequences, three domains were found in PKC. Except for the putative signal peptide at its N-terminus, a SRPBCC domain was found at residues 28 to 180. This corresponds to a ligand-binding domain with a deep hydrophobic ligand-binding pocket thought to bind diverse ligands [[113]27]. The residues at positions 200–365 correspond to a GyrI-like small molecule-binding domain in the AraC family of transcriptional regulators. GyrI-like family members have stand-alone domains or are fused to other functional domains, such as DNA-binding or enzymatic domains [[114]28]. Those regulators showed multiple regulatory roles, including sugar catabolism and responses to stress and virulence, and they may also bind small molecules. The N-terminal signal peptide is involved in directing a protein to the plasma membrane [[115]29]. Whether the PKS is transported to plasma membrane will help to elucidate its catalytic mechanism. In addition, PKC may bind diverse substrates for its catalytic function because it contains both SRPBCC and GyrI domains. The protein genes that contain SRPBCC play a wide range of roles, such as fruit ripening in strawberry (Fragaria x ananassa) [[116]30], anthocyanin accumulation [[117]31], and catalysis of the free radical reactions of polyunsaturated fatty acids [[118]32]. 3-Hydroxypyridine is an important natural pyridine derivative and is utilized by organisms as sources of carbon, nitrogen, and energy. In Ensifer adhaerens, a four-component dehydrogenase HpdA1A2A3A4 catalyzes the initial hydroxylation of 3-hydroxypyridine to yield the formation of 2,5-dihydroxypyridine. The component HpdA3 contains an SRPBCC ligand-binding domain [[119]33]. This study reported the necessity of the PKC for the pathogenicity of X. citri subsp. citri. Like the Tn5 insertion mutant, a deletion mutation constructed in the 29 − 1 strain resulted in the loss of pathogenicity on citrus plants. By constructing a series of PKC deletion mutants, we confirmed that the SRPBCC domain is sufficient to restore virulence toward citrus plants to the mutant. As PKC catalyzes the closure of rings during the biosynthesis of polyketides, this highlights the importance of catalysis mediated by the SRPBCC domain for pathogenicity. This is the first report to fully elucidate the role of the conserved PKC in Xanthomonas. The T3SS functions as an important mechanism for infection that delivers a repertoire of effectors into its host cells to modulate defense responses or other biological processes [[120]3]. A prominent T3SS trait is that its expression is environmentally induced, especially its high expression in planta [[121]9, [122]10]. By evaluating the level of expression of the PKC gene, we found that it was induced in planta by comparison with its expression in NB media. qRT-PCR and promoter GUS fusion analyses verified that the hrp gene regulators hrpG and hrpX were downregulated in the PKC gene mutant. This and the reduced expression of hrpD6 and hrcV led to the conclusion that the entire T3SS was downregulated in PKC deletion mutant. The expression of PKC was induced when X. citri subsp. citri was inoculated onto citrus plants, which enhanced the biosynthesis of particular metabolites. It is widely accepted that particular signals derived from host plants are specifically recognized by bacterial pathogens [[123]10]. It would be of value to determine whether the differentially expressed metabolites identified in the PKC mutant are associated with T3SS activation in subsequent research. It has been noted that the level of expression of the hrp genes was not fully restored to the wild type in the plasmid-based complementation analysis in this study. An alternative strategy is required to determine this, which may help to determine whether changes in the level of expression of the PKC cause certain effects on the T3SS genes. Based on the UHPLC-QE-MS/MS analysis, ketoleucine and acetone cyanohydrin were found to be downregulated in the mutants relative to the WT and complemented strains. Moreover, d-ribose, creatine, polyoxyethylene dioleate, and cohibin C were upregulated. The exact roles of these metabolites have not been elucidated in X. citri subsp. citri. However, some of them have been demonstrated to be associated with metabolism and growth even in metabolic diseases in animals. For example, binephrectomized rats treated with high doses of ketoleucine exhibit high mortality owing to the progressive inactivation of glycogen synthetase I, which occurs in parallel with the increase in the activity of phosphorylase alpha [[124]34]. Maple syrup urine disease is caused by a deficiency in branched-chain α-keto acid dehydrogenase complex activity and results in the accumulation of branched-chain α-keto acids in tissues, particularly α-ketoleucine [[125]35]. In Annona muricata and Annona nutans, cohibins C and D are two important metabolites involved in the biogenesis of acetogenins [[126]36]. Further studies are merited to elucidate whether cohibin C is directly catalyzed by PKC. Furthermore, the involvement of polyketide synthase FrnE encoded by XAC1331 cannot be ignored. PKC and FrnE may catalyze the biosynthesis of the same chemicals. Conclusions This study determined the role of a PKC in X. citri subsp. citri, which is highly conserved among plant pathogenic species of Xanthomonas. The deletion mutagenesis of PKC impaired the function of T3SS, thereby abolishing its canker symptoms induced on citrus plants. These results highlight the role of PKC in plant pathogenic Xanthomonas strains. Electronic supplementary material Below is the link to the electronic supplementary material. [127]Supplementary Material 1^ (27.1KB, docx) [128]Supplementary Material 2^ (2.2MB, docx) [129]Supplementary Material 3^ (52.2KB, docx) Acknowledgements