Plant-pathogen interactions involve intricate molecular communications facilitated by various signaling molecules, notably phytohormones. These bacterial phytohormones have emerged as significant virulence factors that allow pathogens to influence the host’s physiological and immune responses (Radouane et al., 2023). Indole-3-acetic acid (IAA), the predominant naturally occurring auxin, regulates numerous aspects of plant growth and development and also modulates immune responses (Xu et al., 2024). Growing evidence suggests that many plant-pathogenic bacteria produce IAA as a virulence factor, which alters host hormone balance, reduces defensive responses, and promotes infection and colonization (Kunkel and Harper, 2018).
Xanthomonas oryzae pv. oryzae (Xoo), the causal agent of bacterial leaf blight of rice, poses a serious threat to global rice production and food security (Qi et al., 2025; Sanya et al., 2022). This vascular, xylem-colonizing pathogen employs a suite of virulence strategies, including the delivery of type III secreted effectors, secretion of extracellular cell wall-degrading enzymes, production of exopolysaccharides, and modulation of phytohormone signaling, to overcome host defenses and establish successful infections (Ramasetty et al., 2023; Zhou et al., 2023). Recent studies have highlighted a critical role for bacterial IAA in pathogenicity, showing that pathogen-derived auxin can attenuate salicylic acid (SA)-mediated immunity and enhance auxin-responsive gene expression in host tissues (Tian et al., 2025; Zhang et al., 2024). These findings suggest that the ability of Xoo to interfere with rice hormone signaling may be integral to its virulence.
Xanthomonas oryzae consists of two primary pathovars: Xoo and Xoc. Although they are closely related genetically, they are responsible for different diseases in rice. Phylogenetic analyses demonstrate that Xoo and Xoc originate from a shared ancestral lineage, yet they have diverged due to host tissue specialization and the development of unique virulence repertoires (Niño-Liu et al., 2006; Ryan et al., 2011). Xoo serves as the causal agent of bacterial blight, primarily colonizing xylem vessels and resulting in systemic infection. In contrast, Xoc is responsible for bacterial leaf streak and is predominantly located within the intercellular spaces of mesophyll tissues (Niño-Liu et al., 2006; Oliva et al., 2019). The observed differences correlate with variations in secretion systems, transcription activator-like (TAL) effector repertoires, and host target genes, which ultimately influence infection strategies and symptom development (Ryan et al., 2011; Wilkins et al., 2015). Both pathovars significantly influence host physiology, particularly hormone signaling pathways, to facilitate disease progression. Auxin production and modulation are implicated in Xanthomonas-plant interactions, indicating that an investigation of auxin biosynthesis in Xoo may yield insights into both conserved and divergent virulence mechanisms among X. oryzae pathovars, as well as elucidate the role of hormonal manipulation in tissue-specific pathogenicity (Supplementary Table 1).
Bacterial IAA biosynthesis occurs via various tryptophan-dependent and -independent pathways, with the nitrilase-mediated pathway being significant for auxin production (Tang et al., 2023; Zhang et al., 2024). Nitrilases hydrolyze nitriles to carboxylic acids and ammonia and can convert indole-3-acetonitrile (IAN) into IAA in both plants and microbes. In Xanthomonas oryzae pv. oryzicola (Xoc), two nitrilase genes, NIT24 and NIT29, are essential for IAA biosynthesis and virulence. (Zhang et al., 2024). However, the functions of nitrilase homologs in Xoo, especially in strain PXO99A, remain poorly understood. Understanding nitrilase function in Xoo is crucial for revealing potential auxin-mediated virulence mechanisms shared or varied among these closely related pathovars.
Successful phytopathogens depend on various physiological traits for colonization and persistence in plants, including biofilm formation, motility, and enzyme secretion (Antar et al., 2020; Han et al., 2025). Biofilms enhance survival by offering protection and adherence while increasing resistance to environmental stresses. Extracellular enzymes aid in tissue penetration and nutrient acquisition. Chemotaxis allows bacteria to move towards favorable conditions. Research highlights that IAA influences the expression of genes associated with virulence and biofilm formation in pathogens like Pseudomonas syringae, acting as both a virulence factor and an internal signaling molecule. (Djami-Tchatchou et al., 2020; Kunkel and Harper, 2018; Linda et al., 2024; Thakur and Yadav, 2024).
This hormonal antagonism creates a dynamic molecular conflict in which pathogens exploit auxin production to reprogram host physiology and suppress defense, while plants attempt to mount effective SA-mediated responses. However, how nitrilase-mediated IAA biosynthesis in Xoo integrates with bacterial virulence traits and host hormone signaling, and to what extent it contributes to the auxin-SA balance during infection, remains unclear (Han et al., 2019; Rawat and Laxmi, 2025; Tian et al., 2025). In this study, we identify NIT24 and NIT29 orthologs in Xoo PXO99A and examine whether they participate in nitrilase-dependent IAA production. Using deletion and plasmid complementation, we examine how NIT-dependent auxin production impacts bacterial virulence and in planta growth, host auxin-SA balance and defense-marker expression, and key bacterial fitness traits including biofilm formation, cellulase activity, chemotaxis, and virulence gene expression. Together, these analyses clarify how nitrilase-linked metabolism connects hormone manipulation to virulence programs in a major rice vascular pathogen.
Materials and Methods
Bacterial strains and plant material
Xoo strain PXO99A and its derivative strains were routinely cultured on peptone sucrose agar (PSA; peptone 10 g L−1, sucrose 10 g L−1, glutamic acid 1 g L−1, agar 15 g L−1, pH 7.7) at 28°C for 3 days. The rice (Oryza sativa L.) cultivar Kitaake was used for all pathogenicity and hormone analyses. Seeds were surface-sterilized, germinated in water for 4 days, and then transferred to medium-sized plastic pots (approximately 10-12 cm in diameter) containing a standard potting soil mixture composed of peat moss, vermiculite, and perlite. Plants were grown in a growth chamber under controlled conditions at 28°C, 60% relative humidity, with a 12 h light/12 h dark photoperiod and a light intensity of 200 μmol m−2 s−1 for 28 days (Yoo et al., 2023).
Construction and verification of PXO99A ΔNIT24 and ΔNIT29 mutant and complemented strains
Targeted deletion mutants of NIT24 and NIT29 were generated in the Xoo PXO99A via homologous recombination using a suicide pUC18-based vector with kanamycin selection. Approximately 1 kb upstream and downstream flanking regions of each gene were PCR-amplified, cloned into pGEM-T Easy, and disrupted by insertion of a kanamycin resistance cassette at an engineered BamHI site within the coding sequence. The disrupted fragments were excised with HindIII and KpnI and subcloned into pUC18 to produce gene replacement constructs. Recombinant plasmids were introduced into PXO99A by electroporation (1,500 V, 25 μF, 400 Ω), recovered in PSA broth, and kanamycin-resistant transformants were selected on PSA agar. Gene disruption was confirmed by colony PCR using flanking primers, followed by diagnostic PCR and sequence verification. For genetic complementation, full-length NIT24 and NIT29 coding sequences, including their native ribosome binding sites, were amplified and cloned downstream of the lacZ promoter in the broad-host-range vector pBBR1-MCS5 using KpnI and HindIII. The resulting constructs were sequence-verified and introduced into the respective knockout strains by electroporation, generating complemented strains maintained on PSA supplemented with kanamycin and gentamicin. Expression of complemented nitrilases was validated by PCR and Western blot analysis. His-tagged NIT24 and NIT29 proteins expressed from pBBR1-MCS5 were detected following SDS-PAGE separation of total proteins from sonicated cell lysates, transfer to PVDF membranes, blocking with 5% skim milk in TBST, incubation with anti-His primary and HRP-conjugated secondary antibodies, and chemiluminescent detection. Plasmid and primer details are listed in Supplementary Table 2, Supplementary Figs. 1 and 2.
Quantification of bacterial IAA production
IAA production by Xoo strains was quantified using a colorimetric assay with Salkowski reagent as previously described (Mohite, 2013). Strains were inoculated into nutrient broth supplemented with L-tryptophan (1 g L−1) and containing meat extract (1 g L−1), yeast extract (2 g L−1), peptone (5 g L−1), and NaCl (5 g L−1), and incubated at 28°C with shaking. Culture supernatants were collected at 24, 48, 72, and 96 h by centrifugation to remove cells.
For IAA determination, 1 mL of culture supernatant was mixed with 2 mL of Salkowski reagent (2% 0.5 M FeCl3 in 35% HClO4) and incubated in the dark for 30 min. Absorbance was measured at 535 nm using a spectrophotometer. A standard curve was generated using authentic IAA solutions (0-100 μg mL−1) treated in parallel with Salkowski reagent, and the resulting linear regression (y = ax + b) was used to calculate IAA concentrations in culture supernatants from their optical density values.
Virulence assays and in planta sampling
Pathogenicity assays were performed on rice cultivar Kitaake using the standard leaf-clipping method (Antar et al., 2020; Han et al., 2019). Xoo (wild type), ΔNIT24, ΔNIT29, their complemented strains CC11-NIT24 and CC3-NIT29, and mock controls were evaluated in experiments. Bacterial strains were grown on PSA with antibiotics at 28°C for 3 days, and then diluted to 1 × 107 CFU mL−1. Leaves of 4-5-week-old plants were inoculated with the bacterial suspensions or sterile water. Disease severity was measured by lesion lengths 14 days post-inoculation. For genetic and hormone analysis, leaf segments near the inoculation site were collected at various intervals (0-, 12-, 24-, and 48-hpi and 7 and 14 days of inoculation) and stored for IAA and SA extraction. All virulence assays with knockout and complemented strains were reproduced at least three times, confirming consistent findings.
qRT-PCR analysis of rice gene expression
For analysis of rice gene expression, total RNA was extracted from frozen leaf tissue using RNAiso reagent (Takara, Shiga, Japan) according to the manufacturer’s instructions. Residual genomic DNA was removed by DNase I treatment (Qiagen, Hilden, Germany). RNA concentration and purity were determined using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA).
First-strand cDNA was synthesized from 2 μg of total RNA using the SuPrime-Script cDNA Synthesis Kit (Takara RNA to cDNA EcoDry premix) following the manufacturer’s protocol. Quantitative real-time PCR (qRT-PCR) was performed with 2× Prime Q-Mastermix with SYBR Green I (GeNet Bio, Daejeon, Korea) on a Rotor-Gene Q system (Qiagen). Relative transcript levels were calculated using the 2−ΔΔCt method (Livak and Schmittgen, 2001), with OsUBQ5 (LOC_Os01g22490) serving as the internal reference gene (Phan and Schläppi, 2025). Fold changes were determined relative to mock- or wild-type-inoculated controls as indicated in each experiment. Each assay included three independent biological replicates, each with two technical replicates. Primer sequences are listed in Supplementary Table 3.
qRT-PCR analysis of genes involved in virulence
To assess transcriptional changes in key virulence-associated pathways of Xoo, quantitative reverse transcription PCR (RT-qPCR) was performed on representative genes from major regulatory networks. Target genes included components of the type III secretion system (T3SS) regulatory cascade and effector machinery (hrpG, hrpX, hrcC, hrpB1, and selected xop genes), extracellular polysaccharide (EPS) biosynthesis genes (e.g., gum genes), and DSF-dependent quorum-sensing regulators (rpfF, rpfC, and rpfG). Bacterial cultures were grown under conditions identical to those used for phenotypic assays, including XOM2 virulence-inducing medium when appropriate, and harvested at late logarithmic phase. Cells were rapidly chilled, collected by centrifugation, and total RNA was extracted using RNAiso reagent (Takara). Genomic DNA contamination was removed by RNase-free DNase I treatment. RNA concentration and purity were assessed spectrophotometrically. For cDNA synthesis, 1 μg of total RNA was reverse-transcribed using random primers. RT-qPCR was conducted with SYBR Green chemistry and gene-specific primers on a real-time PCR platform. The cycling program consisted of an initial denaturation step followed by 40 amplification cycles (denaturation and annealing/extension). Melting curve analysis was performed to confirm amplification specificity. Relative gene expression levels were calculated using the 2−ΔΔCt method, with 16S rRNA as the internal reference gene. Each experiment included at least three independent biological replicates with technical duplicates or triplicates. Data are presented as mean ± standard deviation (SD). Complete primer sequences and corresponding amplicon details for all genes analyzed in this study are provided in Supplementary Table 4.
Extraction and quantification of IAA and SA from rice leaves
Endogenous IAA and SA levels were quantified from rice seedling leaves using methanol-based extraction protocols with minor modifications of established methods (Cho et al., 2013; Matsuda et al., 2005; Mustafa, 2025).
Approximately 0.5 g of fresh leaf tissue per sample was frozen in liquid nitrogen and ground to a fine powder before extraction. For IAA analysis, tissue was extracted with 5 mL of 80% ethanol (1:25, w/v), vortexed for 1-2 min, and incubated at room temperature in the dark for 3-4 h with intermittent mixing. Extracts were centrifuged (3,000-5,000 × g, 10 min), and the pellet was re-extracted once under identical conditions. Combined supernatants were filtered through a 0.45 μm syringe filter. For SA analysis, ground tissue was extracted twice overnight at 4°C with 5 mL of pre-chilled 90% methanol under gentle shaking. Following centrifugation (10,000 × g, 10 min, 4°C), supernatants were pooled and evaporated to dryness under reduced pressure or nitrogen. Dried residues containing IAA and SA were reconstituted in 1 mL of 5% methanol and filtered (0.45 μm). Hormone separation and quantification were performed by HPLC using a C18 reverse-phase column. For SA, a binary mobile phase consisting of solvent A (0.1% formic acid in water) and solvent B (acetonitrile) was applied at a flow rate of 1 mL min−1. Detection was carried out using fluorescence settings of 305 nm excitation and 407 nm emission. Quantification was based on calibration curves generated with authentic IAA and SA standards, and recovery efficiency was monitored using an internal standard.
Biofilm formation assay
Biofilm formation was quantified in borosilicate glass tubes as described previously (Antar et al., 2020), with minor modifications. Xoo strains were cultured on PSA before being transferred to XOM2 broth with antibiotics and incubated at 28°C until mid-log phase. Aliquots were set to an initial OD600 of 0.05 and incubated statically for 4 days. Afterward, OD600 measurements of planktonic cells were taken, followed by washing the tubes to remove non-adherent cells. The tubes were stained with crystal violet, and excess stain was rinsed off. The retained stain was solubilized in ethanol, and absorbance was measured at 550 nm. Biofilm formation was calculated as the ratio of OD550 (crystal violet) to OD600 (planktonic growth), with experiments conducted in triplicate.
Extracellular cellulase activity assay
Extracellular cellulase activity was assayed on PSA plates supplemented with 0.5% (w/v) carboxymethyl cellulose (CMC) as previously described (Zhang et al., 2024). Fresh Xoo strains were inoculated on PSA-CMC plates, air-dried, and incubated at 28°C for 48 hours for enzyme secretion. Following incubation, plates were treated with 0.1% Congo Red for 30 minutes, rinsed, and washed with 1 M NaCl to visualize clear halos around colonies, indicating cellulase activity. The experiment was consistently repeated three times.
Chemotaxis assay
Chemotactic motility toward glucose was evaluated on semi-solid agar plates using a previously described method (Bae et al., 2018). Soft agar medium (0.3% agar) in XOM2 was solidified in Petri dishes. Xoo strains from overnight cultures were diluted to an OD600 of 0.5 and spotted at the center of each plate. Sterile 6 mm paper discs were positioned 1.5 cm from the center, with one disc loaded with 10 μL of 15% glucose solution as the chemoattractant and the other with sterile distilled water as the control. After 24-hour incubation at 28°C, the diameters of the migration halos were measured, and chemotactic response was determined by the difference in halo radius between the glucose and water discs.
Statistical analysis
All experiments were performed with at least three independent biological replicates, each including two to three technical replicates, unless stated otherwise. Data were analyzed using GraphPad Prism 5.0 (GraphPad Software, San Diego, CA, USA). Statistical significance was determined by one-way ANOVA followed by Tukey’s honest significant difference (HSD) post hoc test. Differences were considered statistically significant at P < 0.05. Data are presented as mean ± SD of biological replicates. For time-course experiments, two-way ANOVA with Bonferroni’s multiple comparisons test was used when appropriate.
Results
Identification of NIT24 and NIT29 orthologs in Xoo
Nitrilase-dependent IAA biosynthesis mediated by NIT24 and NIT29 was recently demonstrated in Xoc, so we next investigated whether orthologous genes are present in the genome of Xoo. Orthologs of NIT24 (AKO15524.1) and NIT29 (AKO15829.1), which are essential for nitrilase-dependent IAA biosynthesis in Xoc, were identified in Xoo strains through NCBI BLAST analysis. Using stringent search parameters (E-value < 1 × 10−10 and sequence identity > 70%), high-confidence matches were detected in multiple Xoo genomes, notably in strain PXO99A.
Construction and validation of NIT24 and NIT29 knock-out mutant and complemented strains
To functionally characterize the putative nitrilases in Xoo, deletion/disruption mutants were generated by homologous recombination using pUC18-based suicide constructs carrying ~1-kb flanking regions interrupted by a kanamycin cassette, resulting in targeted gene disruption of NIT24 or NIT29. Colony PCR and sequencing validated the insertion of the kanamycin resistance cassette at the NIT24 and NIT29 loci, confirming effective mutagenesis. For complementation, full-length copies of each gene were cloned under the control of the lac promoter in an expression vector and introduced into the corresponding mutant strains, and the presence of the plasmid constructs was confirmed by colony PCR. Western blot analysis using an anti-His antibody further demonstrated expression of N-terminally His-tagged NIT24 and NIT29 proteins in the complemented strains, whereas no specific signal was detected in the wild-type or mutant backgrounds (Supplementary Figs. 1, 2). Collectively, these results establish a validated genetic framework for investigating the roles of NIT24 and NIT29 in IAA biosynthesis and virulence in Xoo.
Involvement of NIT24 and NIT29 in the biosynthetic pathway of IAA in Xoo
The quantification of IAA production in culture supernatants of wild-type PXO99A, ΔNIT24, ΔNIT29, and their complemented strains revealed a time-dependent increase in the wild type, reaching ~31-32 μg/mL at 96 h (from ~13 μg/mL at 24 h). In contrast, both ΔNIT24 and ΔNIT29 mutants showed consistently reduced IAA levels throughout the time course, with differences becoming more pronounced at later stages (Fig. 1). By 96 h, ΔNIT29 accumulated ~11-12 μg/mL (~38-40% of wild-type levels), whereas ΔNIT24 reached ~16-17 μg/mL (~52-54%). Complementation of both mutants restored IAA production to near wild-type levels, supporting that the observed reductions were attributable to disruption of the respective NIT genes (Fig. 1).
Virulence assay
We investigated the pathogenicity of NIT gene deletion mutants in rice to determine the role of NIT-dependent IAA synthesis in Xoo virulence. Rice plants inoculated with wild-type PXO99A showed typical blight lesions (22-25 cm), whereas ΔNIT24 and ΔNIT29 mutants produced 32-73% shorter lesions and had markedly lower bacterial populations (~106 CFU/leaf vs. 4-5 × 108 CFU/leaf in wild type). Complemented strains (CC11-NIT24, CC3-NIT29) restored lesion length and bacterial growth, confirming that loss of NIT genes specifically impairs Xoo virulence (Fig. 2).
Xoo infection alters the expression of auxin- and salicylic acid-related genes in rice
For auxin-related genes, expression remained close to basal levels in all treatments during the early phase of infection [0-24 hours post-inoculation (hpi)], indicating that auxin signaling was not strongly activated at these time points (Fig. 3A). By 48 hpi, however, plants inoculated with the wild-type strain showed a clear induction of auxin-responsive genes, with OsIAA1 and OsGH3-8 displaying the most prominent increases compared with mock, reaching approximately 25-31-fold and 15-32-fold higher expression, respectively, in wild-type- and complement-infected leaves. In contrast, induction of OsIAA1 and OsGH3-8 in ΔNIT24- and ΔNIT29-infected plants was markedly weaker, reaching only about 10-fold and ~9-fold above mock, respectively. OsPIN1 and OsAUX1, which encode auxin transport components, also showed increased expression at 48 hpi in wild-type and complemented treatments, with OsPIN1 transcripts rising to ~6-14-fold and OsAUX1 to ~3-18-fold over mock. In ΔNIT24- and ΔNIT29-infected plants, however, OsAUX1 induction was markedly reduced, reaching only ~1-3-fold compared with mock. A similar pattern was observed in plants infected with the complemented strains CC3-NIT29 and CC11-NIT24, which again showed strong upregulation of these auxin-related genes at 48 hpi (Fig. 3A). At later time points (7 and 14 dpi), expression of most auxin-related genes declined toward basal levels in all treatments, approaching those of the mock control (Fig. 3A).
SA-related genes also showed time-dependent changes in expression following Xoo infection (Fig. 3B). OsPAL1, which encodes a key enzyme in SA biosynthesis, was induced after infection with all Xoo strains, with higher transcript levels generally observed at 48 hpi than at earlier time points. In wild-type-infected leaves, OsPAL1 transcripts increased to ~20-21-fold above mock, whereas in ΔNIT24- and ΔNIT29-infected plants, OsPAL1 induction was lower, at approximately 4-6-fold (Fig. 3B). For defense marker genes, including OsPR1 and OsNPR1, expression increased in response to Xoo infection relative to mock controls, and the induction tended to be stronger in plants inoculated with the IAA-deficient ΔNIT24 and ΔNIT29 mutants than in those infected with wild-type PXO99A or the complemented strains at 48 hpi. Specifically, OsPR1 reached ~5-8-fold induction in wild-type infections but ~15-19-fold in ΔNIT24- and ΔNIT29-infected leaves, while OsNPR1 increased to ~4-10-fold in wild-type infections and to ~10-15-fold in the IAA-deficient mutants (Fig. 3B). OsPBZ1, encoding a WRKY-type transcription factor associated with SA signaling, was also strongly induced upon infection, with high expression levels detected in all Xoo treatments at 48 hpi; OsPBZ1 transcripts reached ~17-19-fold above mock in wild-type infections and approximately 10-fold in the mutants (Fig. 3B). In contrast, OsSAH2, which is implicated in SA catabolism, showed higher expression in leaves infected with wild-type and complemented strains than in those inoculated with the IAA-deficient mutants at 48 hpi: OsSAH2 transcripts increased to ~12-13-fold in wild-type- and CC11-NIT24-infected plants, but only to ~6-7-fold and ~4-5-fold in ΔNIT29- and ΔNIT24-infected plants, respectively (Fig. 3B). For most SA-related genes, transcript levels decreased again by 7 and 14 dpi and approached mock levels, consistent with a transient activation of SA-associated defense responses during the early stages of infection (Fig. 3B).
Changes in IAA and salicylic acid levels in Xoo-infected rice leaves
In mock-inoculated plants, IAA contents remained within a basal range of approximately 3-7 μg/g FW throughout the experiment (Fig. 4A). Leaves infected with wild-type PXO99A showed a progressive increase in IAA from ~5 μg/g FW at 12 hpi to ~10 μg/g FW at 24 hpi, reaching peak levels of ~19-20 μg/g FW at 48 hpi. The complemented strains CC11-NIT24 and CC3-NIT29 accumulated IAA to levels comparable to or slightly higher than the wild type at 48 hpi (~21-22 μg/g FW). Conversely, leaves inoculated with the IAA-deficient ΔNIT24 or ΔNIT29 mutants failed to exhibit a notable transient increase; IAA contents remained close to mock levels (~2-5 μg/g FW) at most time points and were only ~2-3 μg/g FW at 48 hpi. From 7 dpi onward, IAA levels in wild-type and complemented strain treatments decreased to ~6-7 μg/g FW and by 14 dpi approached the basal range (~2-5 μg/g FW) across all treatments (Fig. 4A).
For SA, mock-inoculated leaves maintained low basal levels of approximately 3-4 μg/g FW throughout the time course (Fig. 4B). Infection with wild-type PXO99A resulted in a moderate increase in SA to ~4-5 μg/g FW at 12 hpi and ~6-7 μg/g FW at 24-48 hpi. In contrast, leaves infected with ΔNIT24 or ΔNIT29 exhibited substantially higher SA contents, reaching ~16-19 μg/g FW at 48 hpi. The complemented strains CC11-NIT24 and CC3-NIT29 restored SA levels to values comparable to the wild type at 48 hpi (~5-7 μg/g FW). At 7 dpi, SA contents in ΔNIT24 and ΔNIT29 remained elevated (~13-15 μg/g FW) relative to wild-type and complemented strain treatments (~5-8 μg/g FW), whereas by 14 dpi SA levels declined in all treatments to ~1-6 μg/g FW, approaching the basal range (Fig. 4B).
Bacterial IAA promotes biofilm formation
Qualitative crystal violet staining showed that wild-type PXO99A formed abundant biofilm with intense purple staining and substantial surface-associated biomass, whereas the ΔNIT29 mutant exhibited faint staining and markedly reduced biofilm, and the ΔNIT24 mutant displayed an intermediate phenotype. In both cases, introduction of the respective complementation plasmid (CC3-NIT29 or CC11-NIT24) restored staining to levels visually comparable to the wild type. Quantitative analysis (OD550/OD600) supported these observations: wild-type PXO99A reached ~7.5-8.0, while ΔNIT29 and ΔNIT24 showed reduced values of ~2.0-2.5 and ~2.5-3.0, respectively, representing an approximately 60-70% decrease in biofilm biomass relative to the wild type. Complemented strains recovered to ~6.0-7.5, approaching wild-type levels. Overall, these results demonstrate that NIT-dependent IAA production significantly contributes to biofilm formation in Xoo, with loss of NIT29 exerting a somewhat stronger effect than loss of NIT24, although both mutants retain measurable residual biofilm under the tested conditions (Fig. 5).
Extracellular cellulase activity is reduced in NIT mutants
Visual inspection of Congo red-stained plates showed that wild-type PXO99A formed well-defined, clear zones around colonies, indicative of high extracellular cellulase activity, whereas the ΔNIT29 mutant produced only very small or barely visible halos, and the ΔNIT24 mutant produced smaller but still clearly detectable halos (Fig. 6A, 6C). Correspondingly, measured halo diameters for wild-type PXO99A were around 0.85 cm, while ΔNIT29 and ΔNIT24 showed reduced mean diameters of approximately 0.28 cm and 0.58 cm, respectively (Fig. 6B, 6D). The complemented strains CC3-NIT29 and CC11-NIT24 exhibited halo diameters of about 0.80-0.84 cm, approaching those of the wild type (Fig. 6B, 6D). These results indicate that NIT-dependent IAA biosynthesis in Xoo is closely associated with the level of extracellular cellulase activity under the conditions tested.
Chemotactic motility toward glucose is reduced in NIT mutants of Xoo
On glucose-containing plates, wild-type PXO99A displayed a distinct yellow growth front extending from the central inoculation point toward the glucose source, indicating strong chemotactic motility, while only minimal growth was observed toward the water control (Fig. 7A, 7B). In contrast, the ΔNIT29 mutant displayed normal growth at the inoculation site but little or no directional expansion toward glucose, suggesting a significant reduction in chemotactic response (Fig. 7A). The ΔNIT24 mutant showed an intermediate phenotype, with a detectable but smaller migration halo toward glucose compared with the wild type (Fig. 7B). In both cases, introduction of the corresponding complementation plasmid restored directional spreading toward glucose to a level similar to that of wild-type PXO99A (Fig. 7A, 7B). Together, these observations indicate that NIT-dependent IAA production in Xoo is closely associated with efficient chemotactic motility toward glucose under the conditions tested.
Expression of virulence-associated genes in NIT mutants of Xoo
To explore the molecular basis for lower IAA biosynthesis and virulence in ΔNIT24 and ΔNIT29 mutants on rice, we analyzed the expression of specific virulence-associated genes in wild-type PXO99A compared to the NIT mutants and complemented strains. Bacterial RNA from late-log-phase cultures was extracted, and the transcript levels of genes related to the type III secretion system, non-TAL and TAL effectors, extracellular polysaccharide biosynthesis enzymes, and quorum-sensing regulators were measured using qRT-PCR, with wild-type PXO99A expression normalized to 1.0 (Fig. 8).
Expression levels of T3SS-associated genes, including the master regulator hrpG, σ-factor hrpX, outer-membrane component hrcC, and needle complex component hrpB1, were substantially reduced in NIT mutants relative to wild type. The ΔNIT29 mutant showed reductions of hrpG and hrpX to 0.4-0.5 and 0.2-0.3 of wild-type levels, respectively, while ΔNIT24 exhibited milder decreases (hrpG at 0.6-0.7 and hrpX at 0.5-0.6). Similar reductions were noted for hrcC and hrpB1, with ΔNIT29 having lower expression (0.2-0.4) than ΔNIT24 (0.4-0.9). Complemented strains CC3-NIT29 and CC11-NIT24 showed increased transcription (1.5-4.5-fold) versus wild type (Fig. 8A). Among non-TAL T3SS effectors, xopX, xopN, and xopQ displayed gene-specific expression responses. In strains ΔNIT29 and ΔNIT24, xopX and xopQ transcripts were reduced to approximately 0.6-0.9 relative to wild type, whereas xopN expression was stable at around 1.0. In complemented strains, xopX and xopQ expression significantly increased, with xopX rising to about 4.8-5.0-fold and xopQ reaching 6.2-10.0-fold compared to wild type, while xopN exhibited modest increases (Fig. 8B). For EPS biosynthesis, three gum genes (gumB, gumC, gumD) were studied in ΔNIT24 and ΔNIT29 mutants. Expression levels were generally comparable to wild type, indicating maintained EPS gene expression despite the absence of NIT24 or NIT29. In complemented strains, transcripts for gumB and gumC were elevated (~10.0-10.5-fold and ~6.0-7.0-fold, respectively), while gumD had a moderate increase (~3.0-3.5-fold) (Fig. 8C).
Expression levels of avrBs3/PthA family member (PXO_00567) and pthXo1 were lower in NIT mutants compared to wild-type PXO99A, with ΔNIT29 showing a greater reduction than ΔNIT24. For the avrBs3/PthA family member, expression dropped to ~0.35-0.40 in ΔNIT29 and ~0.50-0.55 in ΔNIT24, while in complemented strains, expression increased to about 2.4-2.5-fold. pthXo1 also exhibited reduced levels of ~0.25-0.30 in ΔNIT29 and ~0.35-0.40 in ΔNIT24, recovering partially to ~1.25-1.55 in complemented strains (Fig. 8D). Finally, key genes in the DSF-dependent quorum-sensing system were analyzed, including rpfF (DSF synthase), rpfC (DSF sensor kinase), and rpfG (c-di-GMP phosphodiesterase). In the ΔNIT29 strain, transcript levels of rpfF and rpfG were reduced to approximately 0.30-0.40 of wild-type levels, while ΔNIT24 showed intermediate reductions at around 0.60-0.70. In complemented strains, rpfF and rpfG expression increased to roughly 2.25-2.50-fold relative to wild type, whereas rpfC expression remained close to 1.0 across all strains, indicating minor variation (Fig. 8E).
Because complementation was plasmid-borne under a lac promoter, the elevated transcript levels observed in complemented strains likely reflect copy-number/promoter effects rather than a physiological overshoot. Therefore, complementation is interpreted primarily as restoration of the direction of regulation, not an exact return to wild-type magnitude.
Discussion
Our findings indicate that NIT24 and NIT29 serve as important contributors to IAA production in Xoo PXO99A. In vitro, IAA accumulated over time in the wild type, whereas both ΔNIT24 and ΔNIT29 showed a pronounced attenuation, particularly at later time points, and complementation restored the phenotype (Fig. 1). This pattern is consistent with the nitrilase-dependent route proposed for Xanthomonas, in which nitrilase-family enzymes hydrolyze IAN to IAA. Notably, a related study in Xoc reported a 38-49% reduction of IAA upon nitrilase deletion (Zhang et al., 2024), suggesting that nitrilase-linked metabolism contributes to auxin output across X. oryzae pathovars, although the magnitude can vary by strain and experimental context. Xoo is one of the most devastating bacterial pathogens of rice, responsible for bacterial blight disease that causes severe yield losses across major rice-growing regions worldwide (Younas et al., 2025). Despite extensive research on this pathogenic system, the molecular basis of auxin-mediated virulence remains largely undefined, with limited insight into how bacterial hormone biosynthesis integrates with pathogenicity mechanisms (Zhong et al., 2024). In this study, we performed comprehensive genomic analyses of two nitrilase family genes, originally identified in Xoc, to investigate their homologs in Xoo. Nitrilase family genes are broadly conserved in plants, including rice, and their presence alone does not constitute evidence for host-pathogen coevolution. Instead, these conserved plant enzymes highlight that pathogen-derived nitrilase activity and IAA production could plausibly interface with conserved auxin-SA regulatory circuits that shape disease outcomes (Zhang et al., 2024). While the broader evolutionary framework of the nitrilase family has been described previously (Zhang et al., 2024), our comparative analysis further supports that the NIT24 and NIT29 orthologs identified in Xoo are highly conserved relative to their Xoc counterparts (Supplementary Fig. 3). Phylogenetic mapping placed these proteins within closely related clades (Letunic and Bork, 2024) (Supplementary Fig. 3A), and Circoletto-based comparison showed strong sequence conservation without major rearrangement in the analyzed regions (Darzentas, 2010) (Supplementary Fig. 3B). Pairwise SIAS analysis likewise indicated very high sequence similarity between pathovars (99.31% for NIT24 and 98.87% for NIT29; Supplementary Fig. 3C). Together, these observations support strong conservation of these nitrilases between Xoo and Xoc and are consistent with the possibility that they retain similar biochemical roles. At the same time, the phenotypic relevance of this conservation should be interpreted cautiously, as sequence similarity alone does not establish complete functional equivalence in planta (Supplementary Fig. 3).
Xoo primarily infects rice leaves via hydathodes, circumventing immune responses associated with stomatal closure. The continuously open water pores of the hydathode permit direct bacterial access to the xylem, thereby facilitating systemic colonization. The type III secretion system (T3SS) promotes infection via a needle-like apparatus that delivers effector proteins into host cells. The expression of T3SS is regulated by the master regulators HrpG and HrpX, which coordinate a minimum of 24 genes responsible for encoding secretion machinery and effectors. The effectors comprise TAL effectors and non-TAL (Xop) effectors. TALs bind specifically to host promoters through repeat-variable di-residues (RVDs), thereby activating susceptibility genes like SWEET14, which facilitates nutrient release for bacterial proliferation. Secretion systems, including T3SS (and in some bacteria T6SS), have been implicated in modulating host processes, highlighting the broader potential for bacterial interference with hormone-associated signaling (Gupta et al., 2024; Mijatović Scouten et al., 2025). IAA, a plant hormone, functions in various capacities, including as a virulence factor, a regulator of bacterial gene expression, and a suppressor of plant immunity. Pathogens such as Xoo, Xoc, and Pseudomonas syringae produce IAA through nitrilase-dependent pathways, specifically NIT24 and NIT29 in Xoc, which facilitate colonization and suppress host defenses. Bacterial IAA production illustrates a complex strategy whereby pathogens may manipulate plant hormonal networks to promote their growth and diminish the plant’s immune responses (Cerboneschi et al., 2016; Cerutti et al., 2017; Kunkel and Harper, 2018; Tan et al., 2024; Zhang et al., 2024).
The successful construction of ΔNIT24 and ΔNIT29 mutants in Xoo represents an essential step toward investigating the role of nitrilase-like enzymes in bacterial auxin biosynthesis and virulence-related processes (Zhang et al., 2024). The genes NIT24 and NIT29 were inactivated using a kanamycin resistance cassette, and their function was restored by a plasmid carrying His-tagged nitrilase proteins, confirmed through immunodetection assays. The absence of the NIT29 signal in the ΔNIT29 mutant and its reappearance in complemented strains validates the targeted gene disruption and recovery approach. These complemented strains serve as a robust model to explore the roles of NIT24 and NIT29 in IAA biosynthesis, EPS production, and virulence in rice hosts. Overall, the analyses affirm the reliability of these mutants and complemented strains for further phenotypic and biochemical studies, linking nitrilase-mediated auxin metabolism to host defense signaling and the pathogenic success of Xoo.
Bacterial IAA synthesis encompasses multiple biochemical pathways, including the indole-3-acetamide (IAM), indole-3-pyruvic acid (IPA), tryptamine (TAM), and tryptophan side-chain oxidase (TSO) pathways. Each pathway plays a role in auxin biosynthesis depending on specific physiological and environmental conditions, highlighting the metabolic flexibility of bacteria in IAA production. In this study, we focused on a nitrilase-dependent route where tryptophan is converted to IAN and then hydrolyzed by nitrilase enzymes NIT24 and NIT29 (Batista-Silva et al., 2022; Rico-Jiménez et al., 2023; Tang et al., 2023; Zhang et al., 2024). IAA production in PXO99A increased progressively during in vitro growth, rising from ~13 μg/mL at 24 h to ~31-32 μg/mL at 96 h, indicating active and growth-associated auxin biosynthesis. In contrast, disruption of NIT24 or NIT29 significantly attenuated IAA accumulation throughout the time course, with the effect more pronounced at later stages. At 96 h, ΔNIT29 produced only ~11-12 μg/mL (~38-40% of wild-type levels), whereas ΔNIT24 reached ~16-17 μg/mL (~52-54% of WT), indicating that the phenotype associated with ΔNIT29 appeared more pronounced under the tested conditions. Restoration of IAA levels upon complementation confirms that the observed reductions were specifically due to loss of the respective NIT genes. Collectively, these findings support that both NIT24 and NIT29 contribute to IAA biosynthesis in Xoo. Complementation analyses further supported the contribution of these nitrilase genes to IAA biosynthesis. Under the tested conditions, the phenotype associated with ΔNIT29 appeared stronger, although the present data do not support a definitive conclusion regarding the relative contribution of the two genes. Although this study focuses on nitrilase-mediated IAA biosynthesis, bacteria can utilize multiple pathways for auxin production, including the indole-3-acetamide (IAM) and indole-3-pyruvate (IPyA) pathways (Ansari and Sridhar, 2000). The presence of residual IAA in the ΔNIT24 and ΔNIT29 mutants suggests that these alternative routes may also contribute to IAA production in Xoo. Such pathways could partially compensate for the loss of nitrilase activity and may influence the observed phenotypes. Therefore, while our results support an important role for nitrilase-mediated IAA production, they do not exclude the contribution of other pathways, which warrants further investigation (Ansari and Sridhar, 2000).
To determine whether the observed phenotypic differences were influenced by general growth defects (Supplementary Fig. 4), we compared the in vitro growth kinetics of the ΔNIT24 and ΔNIT29 mutants with the wild-type strain. The results showed that ΔNIT24 exhibited growth comparable to the wild type, while ΔNIT29 displayed a slightly faster growth rate under the tested conditions. These findings indicate that the reduced virulence, altered IAA production, and associated phenotypic changes observed in the mutants are unlikely to be attributable to impaired bacterial growth. Instead, they more likely reflect specific effects of nitrilase disruption on metabolic and regulatory pathways. Notably, the enhanced growth of ΔNIT29 further suggests that the attenuation in virulence is not due to a general fitness cost but may arise from disrupted coordination between bacterial physiology and host-interaction processes.
Auxin participates in plants’ responses to various biotic and abiotic stresses, and bacterial synthesis of auxin plays diverse regulatory roles in plant-bacteria interactions. Our study indicates that NIT24 and NIT29 affect bacterial fitness, influencing factors beyond IAA production. Notably, impaired biofilm formation highlights the relevance of nitrilase-associated processes in bacterial virulence and lifestyle. (Chattaraj et al., 2025; Djami-Tchatchou et al., 2022; Fahad et al., 2025; Giri et al., 2025; McClerklin et al., 2018; Rico-Jiménez et al., 2023; Zhang et al., 2024). Additionally, decreased cellulase production suggests a link between auxin biosynthesis and virulence factor expression, optimizing tissue degradation when plant immunity is weakened. Taken together, our results suggest that NIT24 and NIT29 contribute to the pathogenicity of Xoo on rice. The mechanisms of reduced virulence involve complex physiological defects beyond mere IAA deficiency. Although the phenotypes correlate with reduced IAA, nitrilase loss may also affect broader metabolic states; future work using exogenous IAA rescue (or catalytic-dead nitrilase controls) would help disentangle IAA-dependent vs IAA-independent effects. Restoration of pathogenicity-related phenotypes in complemented strains supports that the observed effects are associated with nitrilase gene function, further indicating an important role for these enzymes in bacterial pathogenesis (Zhang et al., 2024; Zhong et al., 2024).
Continually evolving research has consistently indicated that the amplification of auxin signal transduction and biosynthesis in plants enhances the colonization proficiency of pathogens (Aragón et al., 2014; Hossain et al., 2024; Yang et al., 2007; Zhang et al., 2024). Our hormone analysis using HPLC showed an interaction between bacterial auxin production and plant SA-associated immunity. At 48 hpi, WT and complemented strains showed a transient increase of IAA, whereas ΔNIT24 and ΔNIT29 remained near basal levels. In contrast, SA accumulation was higher in ΔNIT24/ΔNIT29 than in WT/complements, consistent with an association between reduced bacterial IAA and enhanced SA-related responses. These patterns are consistent with the idea that bacterial IAA influences plant hormone homeostasis during infection and may affect host immune responses. The auxin-SA crosstalk mechanism highlights a complex regulatory interference where bacterial IAA acts as a molecular mimic, activating auxin signaling pathways while inhibiting SA biosynthesis and immune responses (Chen et al., 2023; Han et al., 2025; Rawat and Laxmi, 2025; Tian et al., 2025). Our qRT-PCR analysis shows that WT and complemented strains induced auxin-responsive genes more strongly, whereas ΔNIT24/ΔNIT29 triggered higher expression of selected SA-marker genes (e.g., OsPR1 and OsNPR1), consistent with elevated SA accumulation in the mutants. Notably, not all SA-associated markers (OsPBZ1) exhibited the same expression pattern, indicating additional complexity in SA signaling responses during infection. These results suggest that bacterial IAA may contribute to the modulation of host immune responses, which could partially explain the reduced pathogenicity observed in the mutant strains.
The observation that NIT24 and NIT29 influence the expression of the type III secretion system (T3SS) and quorum-sensing pathways through IAA biosynthesis aligns with findings in other Xanthomonas species, where endogenous IAA has been shown to modulate virulence gene expression and suppress plant basal immunity (Zhang et al., 2024). Similar regulatory roles for IAA have been documented in Pseudomonas syringae pv. tomato DC3000, where auxin signaling is integrated with the T3SS to promote pathogenesis by dampening salicylic acid-mediated defenses (Djami-Tchatchou et al., 2020). Furthermore, the discrepancy between high transcriptional abundance in complemented strains and wild-type levels of virulence, attributed here to copy-number or promoter-driven effects, is a well-documented phenomenon in bacterial genetics; for instance, studies on Agrobacterium tumefaciens have noted that over-expression of virulence regulators does not always yield a linear increase in disease severity due to the rate-limiting nature of the host-pathogen interface (Gelvin, 2000). These comparative insights reinforce the conclusion that transcriptional levels alone are insufficient to predict functional pathogenicity, as virulence is ultimately governed by a complex, multi-layered regulatory network.
Auxin response factors (ARFs) serve as pivotal transcription factors within the auxin signalling pathway. Our analysis shows that Xoo strains modulate auxin-responsive gene expression in rice, with wild-type infections activating auxin signaling pathways more effectively than nitrilase mutants. NIT24 and NIT29 contribute to changes in the expression of genes in the IAA signaling pathway, potentially facilitating bacterial colonization through effects on host hormone balance (Cancé et al., 2022). IAA, as an endogenous plant hormone, is vital for regulating growth and development, including cell division, elongation, root and shoot development, and environmental responses (Gomes and Scortecci, 2021; Mazzoni-Putman et al., 2021). The bacterial production of IAA through nitrilase-mediated pathways represents an evolutionary adaptation that may allow pathogens to influence plant developmental programs (Zhang et al., 2024). This influence promotes plant growth in ways that aid pathogen establishment by creating more stable habitats within plant tissues while diverting host resources away from immune responses toward growth processes, thereby weakening plant resistance. Taken together, our findings suggest that nitrilase genes contribute to multiple virulence-related processes. The coordinated changes in auxin production, biofilm formation, enzyme secretion, and motility suggest that these processes may be influenced by overlapping regulatory networks, providing several evolutionary advantages, including optimal virulence factor combinations, regulatory efficiency, and functional redundancy that enhances pathogen robustness under varying environmental conditions (Chiquito-Contreras et al., 2024; Mahto et al., 2022). The discrepancy between elevated gene expression and stable phenotypic output in the complemented strains likely reflects a non-linear relationship between transcript abundance and functional outcomes. Although copy-number and promoter effects may increase mRNA levels, these changes do not necessarily translate into proportional phenotypic differences. This may be due to additional layers of regulation and the involvement of complex gene networks that govern traits such as virulence, biofilm formation, and motility. Consequently, increased transcription of specific genes may not lead to enhanced phenotypes, particularly if these genes are not rate-limiting within the broader regulatory framework.
Our findings demonstrate that the nitrilases NIT24 and NIT29 contribute to IAA production and are associated with multiple virulence-related traits in Xoo. Disruption of these genes alters bacterial fitness, host hormone balance, and the expression of key virulence determinants, supporting a role for nitrilase-mediated auxin biosynthesis in modulating host-pathogen interactions. However, these conclusions should be interpreted with some limitations. Although single gene mutants revealed important functional contributions, residual IAA production suggests potential redundancy or involvement of additional biosynthetic pathways. Attempts to generate a double knockout mutant were unsuccessful, limiting our ability to assess the combined effects of complete nitrilase disruption. In addition, transcriptional changes observed in complemented strains may be influenced by copy-number or promoter effects and do not necessarily reflect proportional changes in virulence. Furthermore, variation among SA-associated markers indicates complexity in host defense signaling that warrants deeper investigation. Future studies employing alternative genetic strategies and broader pathway analyses will be important to further clarify the mechanistic role of bacterial IAA in virulence.
Conclusion and Future Prospects
This study shows that the nitrilases NIT24 and NIT29 contribute to IAA production in Xoo and are associated with several virulence-related traits in rice. Genetic and physiological analyses indicate that disruption of either gene reduces IAA accumulation, lesion development, in planta bacterial growth, and multiple bacterial fitness-associated phenotypes, including biofilm formation, cellulase activity, and chemotactic motility. Under the tested conditions, the phenotype associated with ΔNIT29 appeared more pronounced, although this trend should be interpreted cautiously. Our data further support an association between nitrilase-mediated IAA biosynthesis and changes in host hormone-related responses during infection. Overall, these findings support a role for nitrilase-linked auxin biosynthesis in Xoo virulence and provide a basis for future mechanistic studies. Targeting nitrilase-mediated IAA biosynthesis may therefore be useful to explore in future disease-management strategies for bacterial leaf blight in rice, including chemical or genetic approaches. Evaluating the conservation of this mechanism across related pathosystems may also help inform broader management strategies (Fig. 9).
Future research should clarify the molecular mechanisms linking nitrilase activity and auxin levels to observed phenotypic changes. Key areas include understanding auxin signal perception and its integration with virulence regulators, and exploring nitrilase functions in various rice cultivars. Genome-wide analyses and biochemical studies may uncover further pathways, while nitrilase-mediated auxin biosynthesis presents a potential target for managing bacterial leaf blight. NIT24 and NIT29 may be viable targets for intervention through inhibitors or genetic modifications. Comparative studies across bacterial species can reveal the conservation of these strategies, aiding in the development of sustainable disease management practices for rice cultivation.
PDF Links
PubReader
Full text via DOI
Full text via PMC
Download Citation
Supplement1
Print
