GpaR Drives Type III Secretion and Virulence Through Direct Activation of the HrpG‐HrpX Cascade in Xanthomonas oryzae pv. oryzicola

ABSTRACT

Xanthomonas oryzae pv. oryzicola (Xoc), the causal agent of bacterial leaf streak of rice, relies on a functional type III secretion system (T3SS) for full virulence and for induction of the hypersensitive response (HR) in nonhost plants. Although the HrpG‐HrpX regulatory cascade is central to T3SS gene expression in Xanthomonas, the upstream regulatory mechanisms controlling this pathway in Xoc remain insufficiently understood. In the present study, we identified GpaR as a previously unrecognised transcriptional activator required for T3SS‐associated pathogenicity in Xoc. Deletion of gpaR significantly attenuated virulence in rice and resulted in delayed and weakened HR in Nicotiana benthamiana. Reverse transcription‐quantitative PCR analysis showed that transcript levels of hrpG, hrpX, and multiple T3SS‐associated hrp genes were markedly reduced in the ΔgpaR mutant. Consistent with these results, β‐glucuronidase (GUS) reporter assays demonstrated that promoter activities of hrpG and hrpX were substantially decreased in the absence of GpaR. Furthermore, constitutive expression of hrpG or hrpX in the ΔgpaR background largely restored virulence and HR, indicating that these regulators function downstream of GpaR. Electrophoretic mobility shift assays (EMSA) and chromatin immunoprecipitation‐quantitative PCR (ChIP‐qPCR) demonstrated that GpaR binds directly to the promoter regions of hrpG and hrpX in vitro and associates with them in vivo. In addition, in vitro transcription assays showed that GpaR directly activates transcription from both promoters. Collectively, these findings establish GpaR as a direct upstream activator of the HrpG‐HrpX cascade and identify a new regulatory layer controlling T3SS expression and virulence in Xoc.

GpaR directly activates hrpG and hrpX, triggering the HrpG–HrpX cascade to promote virulence in rice and induce the hypersensitive response in nonhost plants, revealing a key upstream regulator of type III secretion and pathogenicity in Xanthomonas oryzae pv. oryzicola.

1. Introduction

Rice ( Oryza sativa ), a globally critical cereal crop and staple food for nearly half the world's population, is central to global food security, yet its production is persistently plagued by various diseases that severely limit improvements in yield and quality (Wang et al. 2025). Xanthomonas oryzae pv. oryzicola (Xoc) is a bacterial pathogen responsible for bacterial leaf streak (BLS), which causes significant losses in rice worldwide (Niño‐Liu et al. 2006). A central determinant of disease in many gram‐negative plant pathogens is the type III secretion system (T3SS), a syringe‐like nanomachine that translocates type III effectors (T3Es) into plant cells to manipulate immunity and cellular physiology (Deng et al. 2017). In plant‐associated pathogens, T3SS expression is often low in rich media but strongly induced under host‐mimicking conditions or in planta (Zou et al. 2006). In xanthomonads, T3SS transcription is classically regulated by HrpG/HrpX. HrpG functions as a master response regulator that activates expression of hrpX, which in turn drives transcription of hrp/hrc structural genes and subsets of effector genes (Zou et al. 2006). HrpX is an AraC‐type transcriptional activator that recognises plant‐inducible promoter (PIP) box motifs (commonly summarised as TTCGC–N₁₅–TTCGC) in target promoters, thereby coupling upstream regulation to coordinated activation of T3SS‐associated operons (Koebnik et al. 2006).

The hrpG gene encodes an orphan OmpR‐family response regulator, whereas most OmpR‐type regulators function within canonical two‐component systems (TCSs) (Wengelnik et al. 1996). Li et al. (2014) identified a cognate histidine kinase, HpaS, in Xanthomonas campestris pv. campestris (Xcc) that forms a functional TCS pair with HrpG. Although the HrpG/HrpX module constitutes a conserved core of T3SS regulation in Xanthomonas, accumulating evidence indicates that this pathway is embedded within a broader physiological network that integrates diverse environmental and host‐derived cues. Diffusible signal factor (DSF)‐mediated quorum sensing (QS), governed by the rpf gene cluster, modulates multiple virulence‐associated traits and has been shown to intersect with T3SS control; notably, the DSF sensor kinase RpfC positively influences hrpX expression and the broader hrp/T3SS regulon in Xcc (Jiang et al. 2018). Proteostasis and post‐translational regulation provide an additional layer of host‐dependent control, as host‐induced phosphorylation of the Lon protease stabilises HrpG in Xanthomonas citri , thereby enhancing hrpX transcription and promoting T3SS activation in citrus (Zhou et al. 2018). These findings underscore that additional regulators may operate upstream of, or directly at, the hrpG/hrpX tier to integrate diverse signals into a coherent T3SS transcriptional response.

Accumulating evidence further indicates that multiple transcriptional regulators modulate T3SS gene expression by specifically recognising the hrpG promoter. In Xanthomonas oryzae pv. oryzae (Xoo), the histone‐like nucleoid‐structuring protein (H‐NS) XrvA binds directly to the hrpG promoter and activates its expression (Feng et al. 2009), whereas XrvB represses hrpG transcription (Kametani‐Ikawa et al. 2011). In X. citri subsp. citri, the global post‐transcriptional regulator RsmA binds directly to the 5′ untranslated region (UTR) of hrpG, stabilising the hrpG mRNA and promoting HrpG protein accumulation, which subsequently activates expression of the hrp/hrc genes (Andrade et al. 2014). In addition to HrpG, numerous regulators have been reported to activate hrpX transcription by directly targeting the hrpX promoter. The sigma factor RpoE1 affects both virulence of Xcc on host plants and the hypersensitive response (HR) on non‐host plants. RpoE1 binds the hrpX promoter and positively regulates hrpX expression, thereby indirectly activating the hrp gene cluster and promoting T3SS output (Yang et al. 2018). The Fis‐like protein Flp is also required for full virulence in Xcc. Flp directly associates with the hrpX promoter and induces its transcription (Leng et al. 2019). In the Xcc 8004 strain, the zinc homeostasis regulator Zur positively regulates expression of the hrp gene cluster by activating hrpX (Huang et al. 2009). Likewise, in Xoc, the TetR‐family transcriptional regulator TfmR binds the hrpX promoter and stimulates hrpX transcription (Chang et al. 2024). Notably, several regulators can target both the hrpG and hrpX promoters, enabling coordinated control of the HrpG/HrpX cascade. In Xoo, several regulators, including the σ⁷⁰ factor RpoD (Xu et al. 2025), the transcription factor Sar (Shao et al. 2023) and the LysR‐type activator GamR (Rashid et al. 2016), directly bind the promoters of both hrpG and hrpX, thereby coordinately activating the HrpG/HrpX regulatory cascade and promoting downstream hrp gene cluster expression. Collectively, these examples highlight a regulatory strategy in xanthomonads in which diverse transcriptional regulators converge on the HrpG/HrpX module. By activating hrpG, hrpX or both, these regulators drive expression of the hrp gene cluster and ultimately potentiate T3SS‐dependent virulence.

LuxR‐family transcription factors are best known as QS receptors in LuxI/LuxR systems, but many proteobacteria encode “LuxR solos” (or orphan LuxRs) lacking a cognate LuxI synthase (Niu et al. 2022). These regulators can sense exogenous N‐acyl‐homoserine lactone (AHL), endogenous non‐AHL ligands, or host‐derived small molecules, enabling intra‐, interspecies and interkingdom communication that reshapes bacterial gene expression during host association (Bez et al. 2023). REC–LuxR (LuxR/FixJ or NarL‐type) response regulators are widely distributed across bacterial phyla. Well‐characterised examples include NarL/NarP in enterobacteria (Huynh et al. 2015) and FixJ‐type regulators in rhizobia (Liu et al. 2021), in which phosphorylation of the N‐terminal REC domain modulates the activity of the C‐terminal LuxR‐like HTH output domain. Increasing evidence from plant‐associated bacteria suggests that REC–LuxR (LuxR/FixJ‐ or NarL‐type) response regulators are involved in pathogenicity‐related traits in plant pathogens. AcrR regulates flagellar assembly and virulence in Acidovorax citrulli (Guan et al. 2020). In Azorhizobium caulinodans , both AclR1 and AcfR enhance the competitive colonisation and tumour formation capabilities of A. caulinodans on host plants (Liu et al. 2020, 2021). In Xoc, VmsR contributes to virulence, cell motility, extracellular polysaccharide (EPS) production and biofilm formation (Zhang et al. 2024).

In our preliminary genetic screening of LuxR‐family regulators in Xoc, we observed that deletion of gpaR resulted in markedly attenuated pathogenicity relative to the wild‐type (WT) strain, which motivated us to further dissect its underlying molecular regulatory mechanism. Here, we investigate a LuxR‐family regulator (hereafter GpaR) in Xoc that directly targets the T3SS master tier. We aim to define (i) the contribution of GpaR to T3SS‐associated phenotypes and virulence‐related outputs; (ii) its regulatory impact on hrpG and hrpX transcription and promoter activity; and (iii) the underlying molecular mechanism, focusing on direct promoter binding and transcriptional control at hrpG or hrpX. By establishing mechanistic links between a LuxR‐family regulator and the hrpG or hrpX master tier, our study expands the architecture of T3SS regulation in xanthomonads.

2. Results

2.1. gpaR of Xoc Is Involved in Pathogenicity

We identified a novel pathogenicity‐associated regulator in the Xoc GX01 genome. The XOC gx_0993 gene (named gpaR in this study) is annotated to encode a LuxR‐type transcriptional regulator with an N‐terminal REC domain and a C‐terminal helix‐turn‐helix (HTH) DNA‐binding domain (Figure 1A). To confirm the function of GpaR, we generated a gpaR deletion mutation (∆gpaR) in Xoc. The sequence comprising positions +10 to +663 of the gpaR gene (657 bp in length) was deleted (Figure S1A). Virulence assays were first conducted to compare the WT strain and ∆gpaR. The ∆gpaR strain demonstrated significantly attenuated virulence relative to the WT strain. Pathogenicity was restored to WT levels in a complemented strain (C∆gpaR) (Figure 1B). The statistical results of the lesion lengths for Xoc strains are shown in Figure 1C. To identify the factors underlying the attenuated virulence of the ∆gpaR, we examined several key pathogenicity‐related phenotypes. Compared with the WT strain, the ∆gpaR exhibited significantly reduced EPS production (Figure S1B). In contrast, no significant differences were observed between the ∆gpaR and the WT strain in other virulence‐associated phenotypes, including extracellular protease activity, swarming motility and swimming motility (Figure S1B,C). The ∆gpaR exhibited a prolonged lag phase, but its exponential phase and stationary phase showed no significant differences compared to the WT (Figure 1D). These results indicate that gpaR contributes to the pathogenicity of Xoc.

FIGURE 1.

Domain organisation of GpaR and its contribution to virulence in Xanthomonas oryzae pv. oryzicola (Xoc). (A) Schematic diagram of the predicted secondary structure of XOCgx_0993 from the Pfam database. HTH, helix‐turn‐helix domain; REC, phosphoacceptor receiver domain. (B) Disease symptoms caused by Xoc strains wild‐type (WT), ∆gpaR and C∆gpaR at 14 days post‐inoculation. (C) Lesion lengths were measured in rice leaves inoculated with Xoc strains WT, ∆gpaR and C∆gpaR. Data are presented as mean ± SD; n = 30; ***p < 0.001 by one‐way ANOVA. Representative results from three independent experiments are shown. (D) Growth curves of strains WT, ∆gpaR and C∆gpaR. Bacterial growth was monitored by measuring the optical density at 600 nm. Data are presented as mean ± SD of three replicates.

2.2. GpaR Promotes the HR‐Inducing Capacity of Xoc in Non‐Host Plant

To determine whether GpaR is involved in T3SS function, HR symptoms and electrolyte leakage were examined in Nicotiana benthamiana leaves infiltrated with different Xoc strains (phosphate‐buffered saline [PBS] was used as a negative control). Compared with the WT and C∆gpaR strains, the ∆gpaR induced a significantly delayed and attenuated necrotic response, with weaker necrosis at 24–36 h post‐inoculation (hpi) (Figure 2A). Consistently, electrolyte leakage in ∆gpaR‐inoculated leaves was significantly lower at early time points and increased later than that in WT and C∆gpaR (Figure 2B). These results indicate that deletion of gpaR delays and attenuates HR development in N. benthamiana.

FIGURE 2.

Deletion of gpaR impairs hypersensitive response (HR) induction and ion leakage in nonhost plants. (A) HR in Nicotiana benthamiana leaves induced by infiltration with wild‐type (WT), ∆gpaR and C∆gpaR, using phosphate‐buffered saline (PBS) as a negative control. Photographs were taken at 24, 36 and 48 h post‐inoculation (hpi). Results are representative of three independent experiments, each with three biological replicates. (B) Ion leakage measured as electrolyte conductivity in leaf tissues infiltrated with the indicated strains at 24, 36 and 48 hpi. Data are presented as mean ± SD; n = 3 independent samples per time point.

2.3. GpaR Positively Impacts the Expression of hrp Gene Cluster

Reverse transcription‐quantitative PCR (RT‐qPCR) analysis showed that the expression levels of the key regulatory genes hrpG and hrpX, as well as several other hrp genes, were significantly reduced in the ∆gpaR compared with the WT strain after 16 h of incubation in the hrp‐inducing medium XOM3 (Figure 3A,B). In C∆gpaR, the expression levels of hrpG and hrpX were essentially identical to those in the WT strain (Figure 3A). β‐glucuronidase (GUS) activity assays demonstrated that hrpG and hrpX expression were markedly decreased in the ∆gpaR (Figure 3C). These results indicate that GpaR positively regulates hrp gene expression in Xoc.

FIGURE 3.

GpaR activates hrpG and hrpX expression and promotes downstream hrp gene transcription in Xanthomonas oryzae pv. oryzicola (Xoc). (A) Reverse transcription‐quantitative PCR (RT‐qPCR) analysis of hrpG and hrpX mRNA expression in Xoc GX01, ∆gpaR and C∆gpaR. Bacterial 16S rDNA was used as the internal reference. Data are presented as mean ± SD; n = 3; ***p < 0.001 by unpaired Student's t‐test. (B) Relative expression of hrp genes within the hrpA–F operon in Xoc GX01 and ∆gpaR, analysed by RT‐qPCR with 16S rDNA as the internal control. Data are presented as mean ± SD; n = 3; *p < 0.05, ***p < 0.001, ns p > 0.05 by one‐way ANOVA. (C) β‐glucuronidase (GUS) activity assays measuring promoter activity of hrpG and hrpX in Xoc GX01 and ∆gpaR. Data are presented as mean ± SD; n = 3; **p < 0.01, ***p < 0.001 by unpaired Student's t‐test.

2.4. Constitutive Expression of hrpX / hrpG Restores the Pathogenicity and HR of ΔgpaR

To determine whether the reduced virulence of ∆gpaR in rice and the delayed HR in N. benthamiana are associated with hrpG and hrpX, we constructed constitutive expression strains of hrpG and hrpX in the ∆gpaR background. The hrpG and hrpX genes were cloned into the pXUK vector and introduced into ∆gpaR by triparental mating, generating ∆gpaR‐pK‐G and ∆gpaR‐pK‐X, respectively. Xoc strains were inoculated into rice and N. benthamiana leaves by syringe infiltration. At 14 days post‐inoculation (dpi), lesion length analysis showed that both ∆gpaR‐pK‐G and ∆gpaR‐pK‐X largely restored the virulence of ∆gpaR, producing lesions comparable to those of the WT and C∆gpaR strains, and significantly longer than those caused by ∆gpaR and the vector control ∆gpaR‐pK (Figure 4A). Statistical analysis of lesion length measurements was consistent with the observed disease phenotypes (Figure 4B). In N. benthamiana, HR symptoms were observed earlier and were stronger in areas inoculated with WT, C∆gpaR, ∆gpaR‐pK‐G and ∆gpaR‐pK‐X, whereas ∆gpaR and ∆gpaR‐pK showed a significantly delayed and attenuated necrotic response (Figure 4C). Consistently, electrolyte leakage in these inoculated areas showed the same trend (Figure 4D), supporting that constitutive expression of hrpG or hrpX compensates for the ∆gpaR defect. In addition, the ∆gpaR‐pK‐G and ∆gpaR‐pK‐X strains also restored EPS production in the ∆gpaR (Figure S1D).

FIGURE 4.

hrpG and hrpX function downstream of GpaR to mediate virulence and hypersensitive response (HR) in Xanthomonas oryzae pv. oryzicola (Xoc). (A) Disease symptoms on rice leaves at 14 days post‐inoculation (dpi) caused by different Xoc strains. ΔgpaR‐pK‐G, the ΔgpaR strain carrying hrpG; ΔgpaR‐pK‐X, the ΔgpaR strain carrying hrpX; and ΔgpaR‐pK, the ΔgpaR strain carrying the empty vector pXUK as a negative control. PBS, phosphate‐buffered saline; WT, wild type. (B) Lesion lengths were measured in rice leaves inoculated with different Xoc strains. Data are presented as mean ± SD; n = 30; *p < 0.05, ***p < 0.001, ns p > 0.05 by one‐way ANOVA. Representative results from three independent experiments are shown. (C) HR in nonhost Nicotiana benthamiana plants infiltrated with the indicated strains at 24, 36 and 48 h post‐inoculation (hpi). Images are representative of three independent experiments, each with three biological replicates. (D) Ion leakage measured as electrolyte conductivity in inoculated leaf tissues at 24, 36 and 48 hpi. Data are presented as mean ± SD; n = 3 independent samples per time point.

2.5. GpaR Directly Activates the Expression of hrpG and hrpX

To determine whether GpaR directly regulates T3SS gene expression, we examined the binding of GpaR to the promoter regions of hrpG and hrpX using both in vitro and in vivo approaches. First, electrophoretic mobility shift assays (EMSA) were performed using a 5′ 6‐FAM‐labelled DNA probe corresponding to the putative promoter region of hrpG and hrpX (Figure 5A). Purified His‐tagged GpaR (GpaR‐His) caused a distinct mobility shift, indicating that GpaR directly binds to the hrpG and hrpX promoters in vitro. The gyrB promoter fragment was included as a nonspecific competitor control and showed no detectable binding, confirming the specificity of the interaction.

FIGURE 5.

GpaR directly binds the hrpG and hrpX promoters and activates their transcription. (A) Electrophoretic mobility shift assay (EMSA) analysis for the binding of GpaR to the promoters of hrpG, hrpX and gyrB. FAM‐labelled DNA probes were used; unlabelled probe served as competitor. “+” and “–” indicate presence or absence of the respective component. Free DNA and protein–DNA complexes are indicated. (B) Chromatin immunoprecipitation (ChIP)‐quantitative PCR analysis of GpaR binding to target promoters in vivo, using Flag antibody for ChIP and IgG as mock control. Data are presented as mean ± SD; **p < 0.01, ***p < 0.001, ns p > 0.05. (C) In vitro transcription analysis of hrpG, hrpX and gyrB promoters in the presence of increasing concentrations of GpaR (0, 10, 20 and 40 nM). Reactions contained Escherichia coli RNA polymerase holoenzyme (RNAP). Each template spans the region from upstream of the transcription start site to +150 bp.

To further verify the association of GpaR with the hrpG and hrpX promoters in vivo, chromatin immunoprecipitation‐quantitative PCR (ChIP‐qPCR) analysis was performed using Xoc GX01 and the C‐terminally 3 × Flag‐tagged strain GX01‐GpaR3Flag. Western blotting confirmed successful construction and expression of GpaR‐Flag in GX01‐GpaR3Flag (Figure S2A). The quality of chromatin fragmentation after sonication was verified by agarose gel electrophoresis (Figure S2B), and immunoprecipitated ChIP input and elution fractions were further validated by western blotting (Figure S2C). In Figure 5B, Mock‐ChIP represents the sample obtained from the WT strain subjected to the same ChIP procedure, whereas ChIP represents the sample obtained from the GX01‐GpaR3Flag strain. ChIP‐qPCR analysis showed that the hrpG and hrpX promoter fragments were significantly enriched in the ChIP sample compared with the Mock‐ChIP sample, whereas no significant enrichment was observed for the gyrB promoter. These results support that GpaR specifically associates with the promoter regions of hrpG and hrpX in vivo.

Moreover, in vitro transcription assays showed that the presence of GpaR enhanced transcription from both the hrpG and hrpX promoters (Figure 5C), indicating that GpaR not only binds these promoter regions but also functions as a transcriptional activator. Collectively, these results demonstrate that GpaR directly targets the promoters of hrpG and hrpX and positively regulates their expression.

2.6. Identification of the Core‐DNA Sequence Bound by GpaR

To map the GpaR‐binding sites in the hrpG and hrpX promoters, we performed EMSA using promoter probes of different lengths. For hrpG (Figure 6A,C), probes spanning −240/−175/−160 to −1 showed clear mobility shifts, whereas binding was weak with the −140 to −1 probe and absent with the −110 to −1 probe, suggesting that the GpaR‐binding site lies within −160 to −110. For hrpX (Figure 6B,C), probes −240/−174/−145/−135 to −1 retained strong shifts, while −115 to −1 bound weakly and −110 to −1 did not bind, indicating a binding region within −135 to −110. We further modelled GpaR–DNA interactions using the online AlphaFold 3 server (https://alphafoldserver.com/) (Abramson et al. 2024). The predicted complexes showed high confidence for hrpG (−155 to −136; ipTM = 0.83, pTM = 0.87) and hrpX (−138 to −115; ipTM = 0.82, pTM = 0.88). Interface visualisation in PyMOL suggested specific residue–base hydrogen‐bond contacts consistent with the predicted binding regions (Figure 6D,E). To identify the key nucleotides required for GpaR binding, we designed three mutant unlabelled competitors for each of the predicted GpaR‐binding regions in the hrpG and hrpX promoter fragments; the corresponding primer sequences are listed in Table S3. As shown in Figure 6F, EMSA competition assays showed that a 100‐fold excess of unlabelled WT competitors abolished the shifted bands for both the hrpG and hrpX probes. By contrast, the MUT1 competitor failed to compete with the labelled probes, while the MUT2 and MUT3 competitors also demonstrated competitive binding. For the hrpG promoter, the MUT1 substitution corresponds to the CTGTCCG region, while for the hrpX promoter, the MUT1 substitution corresponds to the GTGAGAT region. These results indicate that CTGTCCG and GTGAGAT are critical for GpaR binding to the hrpG and hrpX promoters, respectively.

FIGURE 6.

Identification and validation of the GpaR‐binding sites in the hrpG and hrpX promoters. (A) Electrophoretic mobility shift assay (EMSA) of GpaR binding to sequentially truncated hrpG promoter fragments (positions relative to the start codon are indicated). (B) EMSA of GpaR binding to truncated hrpX promoter fragments. (C) Binding activity of GpaR to truncated hrpG and hrpX promoter fragments. The bars represent promoter fragments of different lengths. Red bars indicate promoter fragments that did not bind to GpaR, whereas dark green bars indicate promoter fragments that bound to GpaR. The yellow shaded region indicates the critical regulatory region required for GpaR binding. The magenta and light green arrows indicate the transcriptional directions of hrpG and hrpX, respectively. The symbols denote: “+”, binding; “–”, non‐binding. (D) Predicted interaction between GpaR (blue) and the hrpG promoter segment from −155 to −136 bp (yellow), visualised using the online AlphaFold 3 server and the PyMOL software. Key nucleotides are highlighted in red; yellow dashed lines indicate hydrogen bonds with lengths (in Å) labelled. (E) Predicted complex of GpaR (blue) and the hrpX promoter segment from −138 to −115 bp (yellow). Red nucleotides denote key interaction sites; hydrogen bond lengths (Å) are shown. (F) Competitive EMSA using cold and mutated cold probes. (G) The putative GpaR binding motifs in the promoter region of hrpG and hrpX.

A recent study identified the HrpG‐binding consensus motif as [ATT(C/T)(C/T)(G/C/A)(T/A)T] in Xcc 8004 (Zhang et al. 2020). Using this motif as a query, Shao et al. (2023) searched the hrpX promoter of Xoo PXO99^A and identified a matching sequence, ATTCCATT. Sequence comparisons showed that HrpG from Xoc GX01 shares 72.9% amino‐acid identity with HrpG _Xcc and 97.7% identity with HrpG _Xoo (Figure S3). Consistent with this high conservation, we detected the same HrpG‐binding motif (ATTCCATT) in the promoter region of hrpX (Figure 6G). In addition, we identified the putative GpaR‐binding sites within the promoter regions of hrpG and hrpX (Figure 6G).

3. Discussion

The T3SS is a pivotal determinant of pathogenicity in xanthomonads, and its expression is orchestrated by the conserved HrpG/HrpX cascade, which acts as a master‐tier transcriptional switch controlling the hrp/hrc gene cluster and many associated virulence outputs (Teper et al. 2021). Multiple transcriptional regulators have been shown to directly target the promoters of hrpG and/or hrpX across different xanthomonads, enabling coordinated control of the T3SS regulon (Büttner and Bonas 2010; Teper et al. 2021). A wide range of regulatory factors (Chang et al. 2024; Leng et al. 2019; Rashid et al. 2016; Shao et al. 2023; Xu et al. 2025), including TCSs (Li et al. 2014, 2020; Subramoni et al. 2012), QS system (Jiang et al. 2018), and transcriptional regulators, have been shown to directly or indirectly target hrpG or hrpX, thereby modulating the initiation of the virulence programme. Together with reports of other promoter‐targeting regulators, these findings indicate that diverse regulatory pathways converge at the hrpG/hrpX tier to modulate downstream hrp gene cluster expression and shape T3SS‐dependent infection dynamics.

Our results identify GpaR as an upstream activator of the HrpG/HrpX cascade and a key determinant of T3SS‐dependent pathogenicity in Xoc. Loss of GpaR markedly attenuated disease development, as evidenced by significantly reduced lesion length on rice, and delayed the HR in the non‐host N. benthamiana system, consistent with compromised activation of T3SS‐associated outputs (Guo et al. 2011; Noël et al. 2001; Wengelnik et al. 1996). The ΔgpaR also exhibited a longer lag phase than the WT, although its growth during the exponential phase was not markedly affected, suggesting that loss of gpaR may delay bacterial adaptation to new conditions rather than impair overall growth capacity. Such a delay could influence the initial stages of infection, as successful disease development requires efficient early colonisation and multiplication in host tissues (Hirano and Upper 2000; Xin et al. 2018). Thus, the reduced virulence of ΔgpaR may be partly related to slower early growth.

In addition, ΔgpaR showed reduced EPS production, which may also contribute to the attenuated virulence, as EPS is an important virulence‐associated factor in Xanthomonas and is involved in bacterial colonisation, biofilm formation and disease development (He et al. 2010; Zhang et al. 2013). Notably, constitutive expression of either hrpG or hrpX in the ΔgpaR background restored EPS production to WT levels, suggesting that the effect of GpaR on EPS synthesis is at least partly connected with the HrpG/HrpX regulatory cascade. Therefore, the reduced lesion length of ΔgpaR may reflect combined defects in T3SS activity, early adaptation and EPS production. Nevertheless, given that T3SS‐related gene expression/phenotypes were clearly reduced in ΔgpaR and restored in CΔgpaR, the attenuated virulence is more consistent with impaired T3SS activity, a major determinant of pathogenicity in gram‐negative plant pathogens (Alfano and Collmer 2004; Büttner and He 2009). Therefore, delayed early growth may contribute to the phenotype, but T3SS suppression likely plays a more central role.

Although most tested hrp/hrc/hpa genes were down‐regulated in the ΔgpaR, several genes showed unchanged or even increased transcript levels. This differential response is not unexpected, because the hrp regulatory network in Xoc is not controlled by a simple linear pathway. Previous studies have shown that, in addition to HrpG and HrpX, HrpD6 also modulates the expression of a subset of hrp‐hrc‐hpa genes, including hpa1, hpa2, hpaB, hrcC and hrcT (Li et al. 2011). Moreover, some genes, such as hrcC, hrpE and hpa3, are not completely dependent on HrpG/HrpX, suggesting the presence of additional regulatory inputs (Jiang et al. 2009). HrcT has also been reported to positively regulate hrpX, adding an internal feedback layer to the HrpG/HrpX cascade (Liu et al. 2014). Therefore, loss of GpaR may mainly impair the HrpG/HrpX‐dependent branch of T3SS regulation, whereas genes controlled by additional regulators, feedback loops or promoter‐specific mechanisms may respond differently. These results support a positive role of GpaR in overall T3SS regulation, while also indicating that individual hrp/hrc/hpa genes are differentially integrated into a broader regulatory network.

Mechanistically, we demonstrate that GpaR directly binds the promoters of both hrpG and hrpX and activates their transcription. Moreover, restoring hrpG or hrpX expression in the GpaR‐deficient background was sufficient to rescue virulence and HR phenotypes. Together, these findings establish GpaR as a previously uncharacterised high‐level regulator that links LuxR‐family transcriptional control to the top of the T3SS regulatory hierarchy, providing a mechanistic framework by which xanthomonads can rapidly amplify virulence programmes via direct activation of hrpG and hrpX. Zhang et al. (2020) confirmed by ChIP‐seq that the HrpG‐binding motif in Xcc 8004 is [ATT(C/T)(C/T)(G/C/A)(T/A)T]. In the promoter region of the hrpX in Xoo PXO99^A, the Sar‐binding site (ATCGTGTGCGCCAGC) is located downstream of the HrpG‐binding site (ATTCCATT), and the two sites are separated by 16 bp (Shao et al. 2023). Notably, the same ATTCCATT motif is present at 81 nt upstream of the hrpX coding region in Xoc GX01, and the HrpG amino acid sequence similarity between Xoc GX01 and Xoo PXO99^A reaches 97.7% (Figure S3). These observations suggest that HrpG from Xoc GX01 is highly likely to recognise and bind the ATTCCATT motif in the hrpX promoter. Furthermore, our results demonstrated by EMSA and ChIP‐qPCR that GpaR specifically binds to a region 121 nt upstream of the hrpX coding sequence. This binding site is spatially independent of the previously reported HrpG‐binding site (−81 nt), with an intersite distance of 32 bp. We therefore hypothesise that GpaR may function as an auxiliary activator that enhances the regulatory efficiency of HrpG; however, the molecular basis of their coordinated action remains to be elucidated.

While LuxR‐family transcription factors have been linked to virulence‐associated traits in several plant pathogens, including xanthomonads, the mechanistic connection between LuxR‐family regulation and master‐tier control of the T3SS cascade has remained underexplored (Zhang et al. 2024). By demonstrating direct promoter binding and transcriptional activation of hrpG and hrpX, our work extends the functional scope of LuxR‐family regulators beyond peripheral virulence traits and positions them as high‐level nodes capable of directly gating the T3SS regulatory hierarchy. Given the conservation of GpaR among Xanthomonas species and other bacterial genera, this study adds a new layer to the regulatory architecture of the HrpG/HrpX cascade and advances our understanding of the fine‐tuned control of pathogenicity in Xanthomonas.

Our work establishes GpaR as a key response regulator within the Xoc virulence regulatory network. Notably, genome annotation analysis indicates that GpaR is not flanked by an obvious cognate histidine kinase gene, raising important questions about how this regulator is activated during infection. In particular, it remains to be determined whether GpaR is controlled by non‐canonical signal perception or phosphotransfer pathways and which environmental or host‐derived cues trigger its activation in planta. Future studies should therefore focus on identifying potential kinase partners or alternative upstream sensory components and defining the signal molecules that initiate these pathways. Such efforts will not only refine the architecture of signal integration and virulence regulation in Xoc but may also reveal precise molecular targets for the development of anti‐virulence biopesticides aimed at disrupting key signal transduction circuits and attenuating disease.

4. Experimental Procedures

4.1. Bacterial Strains and Plant Materials

Strains and plasmids used in this study are listed in Tables S1 and S2. Xoc strains were cultured at 28°C in nutrient broth (NB) (beef extract, 3 g/L; yeast extract, 1 g/L; polypeptone, 5 g/L) or on nutrient agar (NA) plates (NB supplemented with 1% agar). Escherichia coli strains were cultured at 37°C in Luria Bertani (LB) medium. For hrp induction, the minimal medium XOM3 (D‐xylose, 1.8 g/L; D, L‐methionine, 670 μM; sodium L‐glutamate, 10 mM; NaFe²⁺‐EDTA, 240 μM; MgCl₂, 5 mM; KH₂PO₄, 14.7 mM; MnSO₄, 40 μM, pH 6.0) was used (Cai et al. 2022). The minimal medium XOM3 was also used to examine the GUS enzyme activity. All media were supplemented with the following antibiotics when necessary: kanamycin (Km), 25 μg/mL; rifampicin (Rif), 50 μg/mL.

Oryza sativa subsp. japonica ‘Nipponbare’ rice was used in this study. Rice plants were grown in greenhouses with a 14‐h light (30°C)/10‐h dark (28°C) photoperiod. N. benthamiana was grown at 22°C with a 16‐h/8‐h light/dark photoperiod.

4.2. Construction of the gpaR Deletion Mutant and the Complemented Strain

The construction of the ΔgpaR was performed as previously described (Zhang et al. 2024). Briefly, the upstream and downstream flanking regions of gpaR were amplified from the Xoc GX01 genome and cloned into the suicide vector pK18mobsacB. The resulting plasmid was propagated in E. coli DH5α and introduced into Xoc GX01 by electroporation. Mutants were selected on NA plates containing 10% (w/v) sucrose and appropriate antibiotics, and verified by PCR using external and internal primer sets. The complementation of the ∆gpaR was constructed by fusion PCR as previously described (Lin et al. 2021). Briefly, primers were designed based on the Xoc GX01 genome to amplify the upstream and downstream homologous arms, with a ~20 bp 3 × Flag overlap introduced at the inner primers for fusion PCR. The fused fragment was amplified, digested with the same restriction enzymes as pK18mobsacB and ligated using T4 DNA ligase. The recombinant plasmid was introduced into the ∆gpaR, and transformants were selected to obtain the complemented strain. The primers used in this study are listed in Table S3.

To construct the complemented strains ∆gpaR‐pK‐G and ∆gpaR‐pK‐X, the recombinant plasmids were first generated. The coding sequences of hrpG and hrpX were amplified and cloned into the broad‐host‐range vector pXUK, in which gene expression is controlled by the lacZ promoter, to obtain pXUKhrpG and pXUKhrpX, respectively. The constructs were verified by PCR and sequencing. Subsequently, the recombinant plasmids pXUKhrpG and pXUKhrpX were introduced into the ∆gpaR strain via triparental mating, yielding the complemented strains ∆gpaR‐pK‐G and ∆gpaR‐pK‐X, respectively. Transconjugants were selected on appropriate antibiotic‐containing media and further confirmed by PCR. The primers used for plasmids construction are listed in Table S3.

4.3. Pathogenicity Assay, HR Assay and Growth Curve

To assess the pathogenicity of Xoc in rice, the infiltration method was employed, as previously outlined (Zhang et al. 2024). The Xoc strains, collected from overnight cultures, were washed and adjusted to a consistent final density (OD₆₀₀ of 0.5, approximately 10⁸ CFU/mL). These resuspended bacterial cells were inoculated into 6‐week‐old rice leaves under appropriate conditions, and the resulting lesions and symptoms were evaluated 14 dpi. The lesion length was also recorded on at least 15 leaves.

For the HR assay, bacterial suspensions (OD₆₀₀ = 0.5) were infiltrated into the abaxial side of N. benthamiana leaves using a needleless syringe. HR symptoms were observed and recorded under white light. PBS and Xoc GX01 were used as negative and positive controls, respectively. Electrolyte leakage was measured at 24, 36 and 48 hpi. Leaf discs (0.4 cm²) were excised and incubated in 10 mL sterile ultrapure water, followed by vacuum treatment for 30 min. Conductivity was measured using a conductivity meter. Four infiltration sites per leaf were considered one replicate, with three independent biological replicates.

Bacterial growth was monitored using an automated growth curve analyser (Bioscreen C, Oy Growth Curves Ab). Cultures were adjusted to OD₆₀₀ = 0.01 and inoculated into XOM3 medium in 96‐well plates (100 μL per well) with three biological replicates. Growth was recorded under appropriate conditions.

4.4. EPS and Extracellular Protease Assays

EPS and extracellular protease assays were conducted, following previously established protocols (Zhang et al. 2024). For EPS production assays, Xoc strains were spotted onto NA plates supplemented with 2% (w/v) sucrose and incubated for 3 days. Extracellular protease activities were assessed by spotting Xoc strains onto NA plates supplemented with skim milk, followed by incubation for 3 days.

4.5. Swimming and Swarming Motility Assays

Swimming motility was assessed on semisolid agar plates (0.3 g/L polypeptone, 0.3 g/L yeast extract, 0.3% agar, pH 7.0) without antibiotics. Bacterial cultures were adjusted to OD₆₀₀ = 1.0, and a sterile toothpick was used to inoculate the medium by vertically stabbing to approximately half the depth of the agar without penetrating the bottom. Plates were incubated upright for 3–4 days under appropriate conditions. Colony diameters were measured and recorded, and images were captured. Each assay included three biological replicates and was repeated three times independently. Swimming motility was quantified based on colony diameter.

For swarming motility, bacterial suspensions were adjusted to OD₆₀₀ = 0.2, and 2 μL aliquots were spotted onto the surface of semisolid agar plates (3 g/L beef extract, 5 g/L polypeptone, 1 g/L yeast extract, 10 g/L sucrose, 0.6% agar, pH 7.0) without antibiotics. Plates were incubated upright for 3–4 days. Colony expansion was documented by measuring colony diameters and photographing the plates. Each experiment was performed with three biological replicates and repeated three times. Swarming motility was quantified based on colony diameter.

4.6. RNA Extraction and RT‐qPCR

Total RNA was extracted from Xoc strains using the TransZol Up Plus RNA Kit (TransGen Biotech) following the manufacturer's instructions. For RT‐qPCR analysis, bacterial cells were harvested by centrifugation at 13,000 g for 5 min at 4°C and washed three times with sterile water prior to RNA extraction. First‐strand cDNA was synthesised from total RNA using RT SuperMix for qPCR (Vazyme Biotech). Quantitative real‐time PCR was performed using SYBR‐based chemistry (SYBR Premix Ex Taq or 2 × RealStar Green Fast Mixture) on an Applied Biosystems 7500 real‐time PCR system with a three‐step amplification protocol. The 16S rRNA gene was used as an internal reference to normalise gene expression levels. Relative transcript abundance was calculated using the 2^−ΔΔCt method. All primers used for RT‐qPCR are listed in Table S3, and each experiment was performed with three independent biological replicates.

4.7. GpaR Purification and Western Blot

To express recombinant GpaR for purification, the gpaR open reading frame (ORF) was cloned into the pRSFDuet‐1 expression vector to generate pRSFDuet‐1‐gpaR. The resulting recombinant plasmid was transformed into E. coli BL21 (DE3), and transformants carrying the construct were selected to obtain the expression strain for subsequent protein production and purification. Recombinant His‐tagged proteins were expressed in E. coli and purified by Ni–NTA affinity chromatography (Kanoh et al. 2023). Briefly, induced cells were harvested, resuspended in lysis buffer (e.g., 20 mM Tris–HCl, 300 mM NaCl, 10 mM imidazole, pH 8.0), and disrupted by sonication on ice. After centrifugation to remove cell debris, the clarified supernatant was loaded onto a Ni–NTA resin pre‐equilibrated with lysis buffer. The resin was washed with wash buffer containing 20–40 mM imidazole to remove non‐specific proteins, and the His‐tagged protein was eluted with elution buffer containing 250–300 mM imidazole. Eluted fractions were analysed by SDS–PAGE, and the purified protein was desalted or buffer‐exchanged into PBS as needed.

Proteins separated by SDS‐PAGE were electrotransferred onto polyvinylidene difluoride (PVDF) membranes (Millipore). Membranes were blocked with 1% (w/v) milk for 1 h at room temperature and then incubated with an anti‐Flag rabbit monoclonal antibody (CST) diluted 1:1000. After washing with Tris‐buffered saline with Tween (TBST; 20 mM Tris–HCl, 0.3 M NaCl, 0.08% Tween 20), membranes were incubated with a horseradish peroxidase (HRP)‐conjugated goat anti‐rabbit IgG secondary antibody (CST) diluted 1:1000. Signals were visualised using chemiluminescent detection according to the manufacturer's instructions.

4.8. GUS Activity Assay

GUS activity in Xoc strains was assayed as previously described (Leng et al. 2019). Briefly, the promoter regions of hrpG or hrpX were fused to the promoterless plasmid pLgus, generating pLgus‐hrpG and pLgus‐hrpX, which were introduced into Xoc strains by transformation. Strains harbouring the reporters were cultured in XOM3 at 28°C. For GUS assays, 1 mL of bacterial suspension was mixed with 40 μL toluene and shaken for 10 min, and OD₆₀₀ was measured. After phase separation for 5 min, 125 μL of the aqueous phase was mixed with 375 μL GUS reaction buffer (50 mM sodium dihydrogen phosphate, 0.1% Triton X‐100, 10 mM β‐mercaptoethanol, pH 7.2) and incubated at 37°C for 10 min. The reaction was stopped with 200 μL stop solution (2.5 M 2‐amino‐2‐methyl‐1,3‐propylene glycol), and the absorbance of 200 μL supernatant was measured at 415 nm. A p‐nitrophenol standard curve (triplicate measurements, R ² ≥ 0.99) was generated by linear regression. Total GUS activity was calculated as (A₄₁₅ − α) × 0.5/(β × 10 × 0.125 × OD₆₀₀), where α and β are the coefficients from the standard curve (α = 0.0117, β = 0.26).

4.9. EMSA

The EMSA procedure followed a previously described protocol (Peng et al. 2023) with minor modifications. Briefly, DNA fragments corresponding to the promoter regions of the target genes relative to the translational start codon were amplified and used as probes for EMSA. The 5′ ends of the probes were labelled with 6‐carboxyfluorescein (6‐FAM) (Sangon Biotech). Binding reactions were performed in a total volume of 20 μL containing 25 μg/mL sheared salmon sperm DNA, 0.5 mg/mL bovine serum albumin (BSA), 100 mM Tris–HCl (pH 7.5), 25% (v/v) glycerol, 250 mM KCl and 5 mM dithiothreitol (DTT). Increasing amounts of purified protein were incubated with the labelled probes at room temperature for 20 min. The reaction mixtures were then separated by electrophoresis on a 4% native polyacrylamide gel (acrylamide/bisacrylamide, 29:1) in 0.5× Tris‐borate‐EDTA (TBE) buffer (44.5 mM Tris base, 44.5 mM boric acid, 1 mM EDTA, pH 8.0). After electrophoresis, fluorescent signals were detected using a gel imaging system. For competition EMSAs, excess unlabelled probes without 6‐FAM labelling were included in the binding reactions as specific competitors.

4.10. ChIP‐qPCR

The GX01‐GpaR3Flag strain was constructed following the method described (Lin et al. 2021). Briefly, a 3 × Flag tag was fused to the C terminus of GpaR by fusion PCR, by inserting the tag immediately upstream of the gpaR stop codon. For allelic exchange via double‐crossover homologous recombination, the left and right homologous arms were amplified from Xoc GX01 genomic DNA. The left homologous arm was 460 bp and ended just before the stop codon, whereas the right homologous arm was 395 bp and contained the 3 × Flag sequence together with the downstream sequence of gpaR.

The ChIP‐qPCR assay as previously described (Ge et al. 2024). The bacterial cultures were grown to an OD₆₀₀ of 0.8 at 28°C. Sonication fragmented the crosslinked chromatin into 100–500 base pairs. Anti‐3 × Flag antibody (DYKDDDDK Tag [D6W5B] rabbit monoclonal antibody; Cell Signalling Technology), rabbit IgG (30000‐0‐AP; Proteintech), and Protein A/G Magnetic Beads (HY‐K0202; MedChemExpress) were used to enrich the protein–DNA complexes according to the manufacturer's instructions. Chromatin immunoprecipitation was also performed using the other aliquot without any added antibodies, which was used as a negative control (Mock‐ChIP samples). The RNA in the DNA sample was removed by incubation with RNase A for 2 h at 37°C, and proteins were then removed by incubation with proteinase K for 2 h at 55°C. The DNA sample was subsequently purified with a PCR purification kit (Qiagen). To measure the enrichment of P _hrpG , P _hrpX and P _gyrB (negative control) in the ChIP DNA samples, qPCR was performed using the SYBR Green PCR Master Mix (Applied Biosystems) and an ABI 7300 Real‐Time PCR System. The relative enrichment was calculated as the fold change using the ΔΔC _t method (Livak and Schmittgen 2001).

4.11. In Vitro Transcription

In vitro transcription assay was performed as described previously (Feng et al. 2022). In vitro transcription reactions contained 50 ng template DNA (containing the promoter and coding sequence regions of hrpG or hrpX), 0.1 U/mL E. coli RNA polymerase holoenzyme (New England Biolabs), the designed concentration of GpaR (0, 10, 20, 40 nM), 1 M KCl, and 0.1 M DTT in 5× transcription buffer. After incubation for 30 min at 25°C, 2.5 mM NTP mix, 2.5 mM biotin‐UTP, RNase inhibitor and Holo RNAP were added into the reaction mixtures in a total volume of 20 μL and incubated for another 1 h at 37°C. Two microlitres of RNase‐free DNase I (20 U/mL) was added to remove template DNA, by incubating for 30 min at 37°C, and the reaction was finally stopped by adding 20 μL 2× RNA loading dye. Samples were then separated by 7 M urea 8% polyacrylamide gel electrophoresis, and transferred to nylon membranes. Chemiluminescent detection of biotin‐labelled mRNA was performed using a D3308 kit (Beyotime Biotechnology).

Author Contributions

Huajun Qin: investigation, methodology, writing – review and editing. Yaqi Zhang: conceptualization, methodology, software, data curation, investigation, validation, formal analysis, visualization, writing – original draft, writing – review and editing. Xiyao Zhao: methodology, validation, writing – review and editing. Jiuxiang Wang: validation, methodology, investigation, writing – review and editing, data curation. Sheng Huang: conceptualization, methodology, supervision, resources, project administration, funding acquisition, writing – review and editing. Lei Liu: data curation, investigation, writing – review and editing.

Funding

This research was supported by the National Natural Science Foundation of China (Grant 32360045).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Verification and phenotypic characterisation of the ΔgpaR mutant. (A) Agarose gel electrophoresis showing confirmation of the ΔgpaR using internal and external primer pairs. M, GenRuler 100 bp Plus DNA Ladder. Lane 1, external primer verification of ΔgpaR, showing a 797‐bp fragment. Lane 2, external‐primer verification of Xanthomonas oryzae pv. oryzicola (Xoc) GX01, showing a 1318‐bp fragment. Lane 3, internal‐primer verification of ΔgpaR, showing no amplification product. Lane 4, internal‐primer verification of Xoc GX01, showing a 363‐bp fragment. (B) Plate assays showing extracellular polysaccharide (EPS) production, extracellular protease activity, swimming motility and swarming motility of the Xoc strains. (C) Quantification of swimming and swarming motility of the indicated strains. The diameters of the motility halos were measured in the respective assays. Data are presented as mean ± SD from three independent replicates. No significant differences were observed among the strains (ns, p > 0.05). (D) EPS production of wild type (WT), CΔgpaR, ΔgpaR, ΔgpaR‐pK, ΔgpaR‐pK‐G and ΔgpaR‐pK‐X. Representative colony phenotypes are shown, and colony diameters were measured for quantitative analysis. Data are presented as mean ± SD from six independent replicates. Statistical significance was determined by one‐way ANOVA followed by Tukey's multiple‐comparison test. ns, not significant; *p < 0.05; **p < 0.01.

Figure S2: Construction and validation of the GX01‐GpaR3Flag strain for chromatin immunoprecipitation (ChIP) analysis. (A) Western blot analysis of three randomly selected GX01‐GpaR3Flag clones constructed for ChIP experiments. (B) DNA fragments were sheared by sonication to sizes ranging from 100 to 500 bp. M, GenRuler 100 bp Plus DNA Ladder. Lanes 1 and 2, sonicated DNA fragments from GX01‐GpaR3Flag samples. (C) Western blot analysis of the input and elution fractions from the ChIP assay.

Figure S3: Multiple sequence alignment of HrpG proteins. Amino acid sequences of HrpG from Xanthomonas campestris pv. campestris (Xcc), Xanthomonas oryzae pv. oryzae (Xoo) and X. oryzae pv. oryzicola (Xoc) were aligned using ClustalW/Clustal Omega. A red background indicates fully conserved residues, whereas white or light blue backgrounds indicate variable positions. Numbers denote amino acid positions.

Table S1: Strains used in this study.

Table S2: Plasmids used in this study.

Table S3: Primers used in this study.

Data Availability Statement

The data that supports the findings of this study are available in the Supporting Information of this article.