The phytohormone indole-3-acetic acid induces quorum sensing signal DSF turnover via a positive feedback biosynthetic loop in the phytopathogen Xanthomonas campestris

Abstract

Indole-3-acetic acid (IAA) accumulates in host plants following infection by Xanthomonas campestris pv. campestris (Xcc), the causal agent of cruciferous black rot. How exposure to IAA affects the invading Xcc remains unclear. Here, we demonstrate that either exogenous addition of IAA or endogenous production of IAA induced turnover of the quorum sensing (QS) signal diffusible signaling factor (DSF) in a RpfB-dependent manner. IAA addition prevented the cytoplasmic and culture pH decline. Transcriptomic analyses revealed four IAA-regulated gene clusters. Specifically, IAA induced the expression of trpB-A, enhancing tryptophan biosynthesis and intracellular IAA accumulation, and thereby establishing a self-reinforcing synthesis loop. IAA upregulated F₀F₁ ATP synthases and a resistance-nodulation-cell division (RND)-family efflux pump HepABCD to induce a pH-dependent DSF turnover. Moreover, IAA downregulated another RND family efflux pump IaepABCDE to induce a pH-independent DSF turnover. Finally, the IAA-regulated gene clusters were transcribed during the XC1 infection of cabbage. Collectively, these findings reveal a previously unrecognized role of IAA in modulating bacterial QS, underscoring the importance of IAA in the molecular dialogue between the pathogen and its host.

Supplementary Information

The online version contains supplementary material available at 10.1007/s44154-026-00298-1.

Introduction

Xanthomonas campestris pv. campestris (Xcc) is the causal agent of black rot, a devastating bacterial disease that threatens global production of major cruciferous crops, including Chinese cabbage, cabbage, kale, mustard, and cauliflower (Vicente and Holub 2013). Xcc enters host plants through hydathodes or wounds caused by mechanical damage or insect activity, and subsequently colonize the xylem, leading to systemic infection (Onsando 1992; Dai et al. 2025). Over the past 60 years, a wide range of virulence factors have been identified in Xcc (Büttner and Bonas 2010; Vicente and Holub 2013; Timilsina et al. 2020). Because of its agricultural importance and well-characterized virulence mechanisms, Xcc has been established as a model bacterial pathogen for studying plant–pathogen interactions (Mansfield et al. 2012; Timilsina et al. 2020).

Xcc relies on quorum sensing (QS) to regulate production of virulence factors (He and Zhang 2008; Deng et al. 2011). In this mechanism, the diffusible signal factor (DSF), chemically identified as cis−11-methyl-dodecenoic acid, acts as the QS signal (Wang et al. 2004). DSF is synthesized via the classical fatty acid elongation cycle and the bifunctional enzyme RpfF (Barber et al. 1997; Zhou et al. 2015b). Once synthesized, DSF is sensed by the RpfC/RpfG two-component system and mediated by the global regulator Clp (Slater et al. 2000; He et al. 2006a; Tao et al. 2010). At the late stage of growth, a fatty acyl-CoA ligase, RpfB, is activated to degrade DSF through the fatty acid β-oxidation pathway (Bi et al. 2014; Zhou et al. 2015a). This DSF-dependent QS system orchestrates collective behaviors of Xcc that are essential for host colonization and pathogenicity, including virulence factor production, biofilm formation, and adaptation to plant defense mechanisms (He et al. 2006b; He and Zhang 2008; Zhou et al. 2017).

Indole-3-acetic acid (IAA) is the most abundant natural auxin in plants, playing a critical role in regulating physiological processes such as cell division, elongation, differentiation, and fruit development (Zhao 2010; Etesami and Glick 2024). A variety of phytopathogens manipulate host IAA biosynthesis or directly synthesize IAA during infection to suppress plant immunity (Duca and Glick 2020; Zhang et al. 2024). IAA concentrations in planta are dynamic and context-dependent. The determination of site-specific IAA levels in plant tissues is constrained by current detection methods. Reported IAA levels primarily fall within the micromolar range (Tang et al. 2023; Zhang et al. 2025), though certain subcellular compartments can accumulate IAA to millimolar levels (Herud-Sikimić et al. 2021). Furthermore, elevated IAA levels are commonly observed in plant tissues infected by Xanthomonas species (O'Donnell et al. 2003; Fu et al. 2011). Our previous work showed that Xcc harbors an L-tryptophan (L-Trp)-dependent IAA biosynthetic pathway and that IAA positively regulates Xcc virulence toward cabbage. One of the IAA signaling cascades in Xcc includes the ilvCGM-leuA gene cluster and branched-chain amino acids, and this cascade regulates reactive oxygen species production and colony formation (Li et al. 2025a). These findings suggest that IAA may act as a signaling molecule in the interaction network between Xcc and its host plants.

In this study, we investigated the role of IAA in DSF-mediated QS signaling and explored its underlying molecular mechanisms in Xcc. We uncovered that IAA induces RpfB-dependent DSF turnover and prevents culture pH decrease. IAA also was found to activate a self-reinforcing IAA biosynthetic loop by inducing the expression of the L-tryptophan biosynthesis gene cluster trpB-A. Furthermore, our data indicate that IAA-regulated efflux pumps and ATPase activity likely contribute to RpfB-dependent DSF turnover. These findings provide new insights into the role of IAA in QS and reveal a previously unrecognized IAA-mediated interplay between host-derived hormones and the invading pathogen Xcc.

Result

Exogenous addition of IAA induces DSF turnover in Xcc

To assess whether IAA affects DSF turnover in Xcc under host-mimicking conditions, we supplemented Xylem-Sucrose (XYS) medium, which was designed to simulate in planta conditions, with IAA at final concentrations of 1 to 100 μM. While exogenous IAA had no significant effect on the growth of the wild-type strain XC1 (Fig. 1a), it caused a marked reduction in DSF levels at both 24- and 36-h post-inoculation (hpi) (Fig. 1b). At 36 hpi, DSF concentrations declined to 0.0032, 0.0019, and 0.0003 μM in the presence of 1, 10, and 100 μM IAA, respectively, corresponding to 11.9%, 9.2%, and 1.4% of that in the control (no exogenous IAA) (0.0207 μM). To investigate whether this effect persisted in a DSF-overproducing background, we treated the ΔrpfC mutant with the same IAA concentrations (Song et al. 2025). IAA did not significantly affect ΔrpfC growth (Fig. 1c), but 10 and 100 μM IAA caused substantial reductions in DSF levels at 36 hpi, whereas 10 μM IAA showed no impact at 24 hpi (Fig. 1d). Collectively, these results demonstrate that exogenous IAA promotes DSF turnover in both XC1 and ΔrpfC under host-mimicking conditions.

Fig. 1.

Exogenous addition of IAA induces DSF turnover in Xcc strains. a XC1 growth in XYS medium in the presence of 1–100 μM IAA. b DSF levels in XC1 culture in the presence of 1–100 μM IAA. c ΔrpfC growth in XYS medium in the presence of 1–100 μM IAA. d DSF levels in ΔrpfC culture in the presence of 1–100 μM IAA. Results are presented as the averages of three technical repeats with standard deviations. *p ≤ 0.05, **p ≤ 0.01, ns: not significant

Endogenous production of IAA triggers DSF turnover in Xcc

Next, we investigated whether endogenous IAA production boosts DSF turnover as well. Because the iaaHM gene cluster is responsible for IAA production in Pseudomonas chlororaphis (Jiang et al. 2014), we integrated the groES promoter-driven iaaHM gene cluster into Xcc strains to construct high-IAA-producing strains XC1::iaaHM and ΔrpfC::iaaHM (Fig. 2a–d). Quantitative high-performance liquid chromatography (HPLC) analysis revealed significantly elevated IAA production in XC1::iaaHM cultures, reaching 81.213 μM at 12 hpi, 113.181 μM at 24 hpi, and 133.022 μM at 36 hpi (Fig. 2c). By contrast, IAA levels in wild-type XC1 cultures were only 0.012 μM at 12 hpi, 0.004 μM at 24 hpi, and 0.005 μM at 36 hpi (Fig. 2c). DSF levels in XC1::iaaHM were significantly lower than those in XC1 (Fig. 2e). Similarly, in the DSF-overproducing background, the ΔrpfC::iaaHM strain exhibited markedly increased IAA production (Fig. 2c) and significantly reduced DSF levels compared with ΔrpfC in XYS medium (Fig. 2f). These results demonstrate that endogenous IAA production promotes DSF turnover in both wild-type and DSF-overproducing Xcc strains.

Fig. 2.

Endogenous production of IAA induces DSF turnover in Xcc strains. a The iaaHM gene cluster from Pseudomonas chlororaphis subsp. aurantiaca JD37. b XC1, XC1::iaaHM, ΔrpfC, and ΔrpfC::iaaHM growth over time in XYS medium. c IAA production in the XYS cultures of the XC1, XC1::iaaHM, ΔrpfC, and ΔrpfC::iaaHM strains. d UPLC-TOF MS analysis of IAA in an XYS culture of Xcc. e DSF levels in XYS cultures of the XC1 and XC1::iaaHM strains. f DSF levels in XYS cultures of the ΔrpfC and ΔrpfC::iaaHM strains. Results are presented as the averages of three technical repeats with standard deviations. *p ≤ 0.05

IAA-induced DSF turnover depends on RpfB

To determine whether IAA-induced DSF turnover involves rpfB, which encodes a long-chain fatty acid CoA ligase that converts DSF to DSF-CoA (Fig. 3a) (Zhou et al. 2015a), we examined the IAA response in strains with different genetic backgrounds, including ΔrpfB and ΔrpfB::rpfB (a ΔrpfB mutant complemented with a single chromosomal copy of rpfB). Exogenous addition of 100 μM IAA failed to reduce DSF levels in the ΔrpfB strain, indicating that RpfB is essential for IAA-induced DSF turnover (Fig. 3b). Conversely, IAA significantly reduced DSF levels in the complemented strain ΔrpfB::rpfB (Fig. 3b), demonstrating that rpfB restoration rescues IAA-induced DSF turnover. These results establish that in Xcc IAA-induced DSF turnover depends on RpfB.

Fig. 3.

IAA-induced DSF turnover is dependent on RpfB. a RpfF and RpfB are respectively responsible for DSF biosynthesis and turnover in Xcc. b DSF levels in the strains ΔrpfB and ΔrpfB::rpfB in the absence and presence of 100 μM IAA. c,d The relative GUS activity of the XC1::P_rpfB-gusA and XC1::P_rpfF-gusA reporter strains in the presence of 1–100 μM IAA. e Purified His-tagged RpfB proteins. f RpfB enzyme activity in the absence and presence of IAA. Results are presented as the averages of three technical repeats with standard deviations. *p ≤ 0.05, ns: not significant

To investigate the underlying mechanism, we assessed whether IAA affects the transcription of rpfB or rpfF, the latter encoding the key enzyme for DSF biosynthesis (Fig. 3a) (Zhou et al. 2015b). Using previously constructed promoter–gusA fusion reporter strains, XC1::P_rpfB-gusA and XC1::P_rpfF-gusA (Song et al. 2022), we observed no significant changes in transcriptional activity for either gene following treatment with 1–100 μM IAA (Fig. 3c and d). We next purified RpfB to determine whether IAA directly enhances RpfB enzymatic activity (Fig. 3e). In vitro assays using DSF as the substrate, with or without 100 μM IAA, showed no significant differences in DSF levels after 15 or 30 min incubation at 37 °C (Fig. 3f), suggesting that IAA does not directly stimulate RpfB activity under these conditions.

IAA treatment prevents decreases in culture and cytoplasmic pH

Because elevated cytoplasmic pH promotes DSF turnover via an RpfB-dependent mechanism (Song et al. 2022), we next examined whether the promotive effect of IAA on DSF turnover is via affecting pH dynamics. We monitored both culture and cytoplasmic pH over time in XYS medium supplemented with 0–100 μM IAA. In the absence of IAA, the culture pH of strain XC1 declined progressively from 6.75 at 12 hpi to 4.67 at 24 hpi and 4.40 at 36 hpi (Fig. 4a), indicating progressive environmental acidification during growth. Consistently, cytoplasmic pH also decreased steadily from 12 to 36 hpi (Fig. 4b).

Fig. 4.

Addition of IAA prevents decreases in XC1 culture and cytoplasmic pH. a The culture pH of XC1 grown in XYS medium supplemented with 1–100 μM IAA. b The cytoplasmic pH of XC1 grown in XYS medium supplemented with 1–100 μM IAA. Results are presented as the averages of three technical repeats with standard deviations. *p ≤ 0.05, ns: not significant

Treatment with 1 μM IAA had no significant effects on culture or cytoplasmic pH (Fig. 4). However, 10 μM IAA moderately mitigated acidification of both culture and cytoplasmic pH at 36 hpi (Fig. 4). Notably, 100 μM IAA significantly elevated culture and cytoplasmic pH as early as 24 hpi, with effects sustained at 36 hpi (Fig. 4).

IAA regulates the expression of gene clusters for tryptophan synthases, efflux pumps, and F₀F₁ATP synthase

To assess the impact of IAA on global gene expression in Xcc, we performed transcriptome sequencing (RNA-seq) of Xcc strains cultured in XYS medium with or without 100 μM IAA at 24 hpi. Statistically significant differentially expressed genes (DEGs) were selected based on a fold-change magnitude of log₂ greater than 1.5 and an adjusted P-value inferior to 0.05 (Fig. S1a). KEGG pathway enrichment analysis indicated that IAA treatment significantly impacted several metabolic pathways, most notably photosynthesis, tryptophan metabolism, and oxidative phosphorylation (Fig. S1b). In-depth analysis of differentially expressed genes (DEGs) aligned with our mechanistic hypotheses. First, genes involved in the biosynthesis of IAA's precursor were markedly upregulated. Specifically, the L-tryptophan synthase gene cluster (trpBA, Xcc2541-Xcc2543) was induced 21- to 45-fold. Second, IAA treatment increased the expression of the atpABCDEFGH operon (Xcc0548-Xcc0555), encoding the F₀F₁ ATP synthase, by 2.6- to 5.2-fold. This enzyme is crucial for cytoplasmic pH homeostasis. Third, given that resistance-nodulation-cell division (RND) efflux pumps modulate QS in Xcc (Song et al. 2024), we further investigated the effect of IAA on these transporters. IAA strongly upregulated the RND-family efflux pump gene cluster hepABCD (Xcc4168-Xcc4171) by 6.0- to 13.0-fold, while concurrently downregulating another RND transporter cluster, iaepABCDE (Xcc1438-Xcc1442), to 0.06- to 0.35-fold of the control level.

Digital PCR validation of four representative genes, including trpB (Xcc2543), hepB (Xcc4169), atpB (Xcc0548), and iaepA (Xcc1438), confirmed the IAA-induced patterns in RNA-seq results. Exogenous addition of 1, 10 and 100 μM IAA significantly increased trpB expression by 3.63-, 9.41- and 15.09-fold; hepB by 1.27-, 2.03- and 9.23-fold; and atpB by 1.23-, 2.24- and 4.96-fold, while suppressing iaepA to 0.81-, 0.50- and 0.06-fold of control levels (Fig. 5b–e).

Fig. 5.

IAA regulates the transcriptional expression of genes involved in ATP biosynthesis, transporter activity, and L-Trp biosynthesis in Xcc. a Fold-change values for L-Trp synthase, transporter, and F₀F₁ ATP synthase transcript levels in response to IAA treatment as identified by RNA-seq. b Digital PCR analysis of the tryptophan synthase gene Xcc2543 in the absence and presence of 100 μM IAA. c Digital PCR analysis of the transporter gene Xcc4169 in the absence and presence of 100 μM IAA. d Digital PCR analysis of the F₀F₁ ATP synthase gene Xcc0548 in the absence and presence of 100 μM IAA. e Digital PCR analysis of the transporter gene Xcc1438 in the absence and presence of 100 μM IAA. Results are presented as the averages of three technical repeats with standard deviations. *p ≤ 0.05, ns: not significant

The IAA-induced trpB-A expression enhances Trp and IAA levels

Transcriptomic analysis revealed that the Xcc2541-2543 gene cluster is strongly upregulated by IAA. Within this cluster, Xcc2541 and Xcc2543 were annotated to encode the α and β subunits of tryptophan synthase, respectively, and Xcc2542 was annotated as a hypothetical protein. Sequence comparison and domain analysis showed that Xcc2541 and Xcc2543 share 41.8% and 61.9% identity with TrpA and TrpB from Pseudomonas aeruginosa PAO1, respectively (Fig. 6a; Fig. S2; Fig. S3), suggesting that they encode the tryptophan synthase genes trpA and trpB in Xcc. To validate their functions, we constructed and characterized individual deletion mutants ΔtrpA and ΔtrpB. These mutants exhibited impaired growth in XYS medium, and this growth defect was partially rescued by 100 μM L-Trp and fully restored by 300 μM L-Trp (Fig. 6b). When trpA or trpB was complemented into the mutants, wild-type growth was restored (Fig. 6b). In contrast, deletion of Xcc2542 had no notable effect on growth, indicating that Xcc2542 is not involved in tryptophan biosynthesis. Furthermore, overexpression of trpB-A produced an approximately 1.6-fold increase in intracellular L-Trp levels compared with wild type at 24 hpi (Fig. 6c). These results confirm that trpB-A plays a crucial role in tryptophan synthesis in Xcc.

Fig. 6.

IAA treatment upregulates trpB-A expression to promote L-Trp and IAA biosynthesis. a The trp gene cluster in Xcc and its homologous gene cluster in Pseudomonas aeruginosa strain PAO1. The numbers below each gene indicate the percentage of sequence identity. b The growth of the ΔtrpA and ΔtrpB deletion strains in the absence and presence of 100–300 μM L-Trp, and the XC1, ΔXcc2542, ΔtrpA::trpA, and ΔtrpB::trpB strains in the absence of L-Trp. c The relative endogenous L-Trp level in the XC1 and XC1::trpB-A strains grown in XYS medium at 24 hpi. d The relative endogenous IAA level in the XC1 and XC1::trpB-A strains grown in XYS medium at 24 hpi. e DSF levels in XYS cultures of the ΔrpfC, ΔrpfC::trpB-A, and ΔrpfCΔtrpB strains. f Relative GUS activity for ΔtrpI::P_trpB-gusA and ΔtrpI(trpI)::P_trpB-gusA in the absence and presence of 100 μM IAA at 24 hpi. g The ability of the TrpI protein (50 ng) to bind Cy5-P_trpB in the presence of 0.5 mM to 5 mM IAA. Results are presented as the averages of three technical repeats with standard deviations. *p ≤ 0.05, **p ≤ 0.01, ns: not significant

Our previous study demonstrated that L-Trp serves as a direct precursor for IAA synthesis in Xcc (Li et al. 2025a). To investigate whether IAA-induced trpB-A expression enhances IAA biosynthesis, we compared intracellular IAA levels between wild-type and trpB-A overexpression strains. The XC1::trpB-A strain exhibited significantly higher IAA levels than wild-type XC1 (Fig. 6d). Collectively, these findings suggest that IAA initiates a self-reinforcing loop in Xcc by inducing trpB-A expression, and that during this process, intracellular L-Trp levels are promoted to induce subsequent IAA biosynthesis.

To examine whether the IAA-induced trpB-A expression contributes to DSF turnover, we measured DSF levels in ΔrpfCΔtrpB and ΔrpfC::trpB-A strains. Deletion of trpB in the ΔrpfC background substantially increased DSF accumulation, whereas overexpression of trpB-A significantly reduced DSF levels (Fig. 6e). At 36 hpi, DSF concentrations were 20.15 μM in ΔrpfCΔtrpB and 0.88 μM in ΔrpfC::trpB-A, corresponding to approximately 672% and 29% of the levels evident in ΔrpfC (3.00 μM), respectively (Fig. 6e). Together, these findings demonstrate that IAA-induced expression of trpB-A enhances intracellular IAA biosynthesis and facilitates DSF turnover, establishing a key regulatory node linking IAA signaling and QS in Xcc.

IAA-dependent trpB-A induction is TrpI-dependent

Previous research has demonstrated that the expression of trpBA is regulated by the adjacent trpI gene in P. aeruginosa (Olekhnovich and Gussin 1998). Xcc2544 shares 57.4% amino acid identity with P. aeruginosa TrpI and contains similar conserved domains, suggesting that Xcc2544 encodes a TrpI homolog (Fig. 6a; Fig. S2; Fig. S3). Further analysis of this locus indicated that the genes trpB, Xcc2542, trpA form a compact cluster with short overlaps (GTGA and ATGA) at the junctions (Fig. 6a), suggesting that trpB-Xcc2542-trpA (trpB-A) is transcribed as a single operon.

To test whether TrpI regulates trpB-A expression in Xcc, we constructed the reporter strain XC1(pBBR)::P_trpB-gusA, in which a 516 bp region upstream of trpB drives gusA expression. A trpI deletion mutant (ΔtrpI::P_trpB-gusA) and its complementation strain ΔtrpI(trpI)::P_trpB-gusA were subsequently generated. In the absence of IAA, the expression level of trpB-A (measured based on β-glucuronidase [GUS] activity) decreased by 69.5% in the ΔtrpI::P_trpB-gusA strain compared to XC1(pBBR)::P_trpB-gusA (Fig. S5a). Conversely, the complementation strain ΔtrpI(trpI)::P_trpB-gusA exhibited elevated expression levels approaching twice those in the wild-type (Fig. S5a), suggesting that TrpI positively regulates trpB-A operon expression.

To investigate how TrpI modulates trpB-A expression, His-tagged TrpI was expressed via the pET-14b vector and purified with Ni–NTA resins. Electrophoretic mobility shift assays (EMSA) were subsequently performed to assess the ability of TrpI to bind to a 288-bp DNA fragment encompassing the trpB promoter region, P_trpB (4 ng). Increasing concentrations of TrpI (10–200 ng) progressively shifted DNA mobility (Fig. S5b). At 50 ng, an additional, slower-migrating band appeared, and its intensity increased with further TrpI addition. No shift occurred with excess unlabeled competitor probe (900 ng) (Fig. S5b). Thus, our EMSA results indicate a specific binding interaction between TrpI and P_trpB.

Since IAA induces trpB-A expression (Fig. 5a and b) and TrpI positively regulates trpB-A (Fig. S5), we subsequently tested whether IAA induction requires TrpI. GUS activity levels were measured for the ΔtrpI::P_trpB-gusA and ΔtrpI(trpI)::P_trpB-gusA strains in the presence or absence of 100 μM IAA. The addition of IAA had no significant effect on the expression of trpB-A in the ΔtrpI::P_trpB-gusA strain (Fig. 6f). However, IAA significantly induced the expression of the trpB-A gene cluster in the ΔtrpI(trpI)::P_trpB-gusA strain (Fig. 6f). These results indicate that TrpI is essential for IAA-mediated induction of trpB-A. The addition of IAA directly to the EMSA reaction did not alter the binding between TrpI and P_trpB (Fig. 6g), suggesting that TrpI does not function as a direct IAA sensor.

The IAA-repressed iaepABCDE is involved in DSF turnover

Transcriptomic analysis revealed significant repression of the Xcc1438–Xcc1442 gene cluster by IAA (Fig. 5a and e). Within this cluster, Xcc1438 and Xcc1441 were annotated as short-chain dehydrogenases/reductase family oxidoreductases, whereas Xcc1439, Xcc1440, and Xcc1441 were annotated as a periplasmic adaptor subunit, a permease subunit, and an outer membrane subunit of an RND family efflux transporter, respectively. Genomic and domain organization analyses revealed that Xcc1439-Xcc1440-Xcc1442 is homologous to mexEF-oprN in P. aeruginosa PAO1 (Fig. 7a; Fig. S2; Fig. S4), an RND family efflux pump involved in metabolite export (Köhler et al. 1997). Accordingly, we designated Xcc1438-Xcc1442 as iaepA-E as it encodes an “IAA-associated efflux pump.”

Fig. 7.

The iaepABCDE (Xcc1438-Xcc1442) gene cluster mediates a RpfB-dependent DSF turnover pathway. a The iaepABCDE gene cluster in Xcc and its homologous gene cluster in Pseudomonas aeruginosa strain PAO1. The numbers below each gene indicate the percentage of sequence identity. b DSF levels in the XC1, ΔiaepABCDE, and ΔiaepABCDE::iaepABCDE strains in XYS medium. c DSF levels in theΔrpfC, ΔrpfCΔiaepABCDE, and ΔrpfCΔiaepABCDE::iaepABCDE strains in XYS medium. d DSF levels in the ΔrpfB, ΔrpfBΔiaepABCDE, ΔrpfB::rpfB, and ΔrpfBΔiaepABCDE::rpfB strains in XYS medium. e DSF levels in the ΔrpfC, ΔrpfCΔhepABCD, and ΔrpfCΔiaepABCDE strains in the absence and presence of 100 μM IAA. Results are presented as the averages of three technical repeats with standard deviations. *p ≤ 0.05, ns: not significant

IAA treatment induced DSF turnover while repressing iaepABCDE expression. To determine whether this repression contributes to DSF turnover, we measured DSF levels in strains XC1, ΔiaepABCDE, and ΔiaepABCDE::iaepABCDE. At 24 hpi, DSF levels in the ΔiaepABCDE were reduced by 61.96% compared to XC1. Complementation by the chromosomal integration of iaepABCDE restored DSF production (Fig. 7b). In a DSF-overproducing background (ΔrpfC), deletion of iaepABCDE reduced DSF levels at 24 and 36 hpi, and this was reversed by iaepABCDE complementation (Fig. 7c). These findings indicate that iaepABCDE maintains DSF levels in Xcc.

Given that the IAA-induced DSF turnover was dependent on rpfB (Fig. 3b), we investigated the degree to which iaepABCDE influences DSF levels via rpfB. Deletion of iaepABCDE failed to alter DSF levels in ΔrpfB, but significantly reduced DSF production in ΔrpfB::rpfB (Fig. 7d). Specifically, DSF levels in ΔrpfBΔiaepABCDE::rpfB were reduced by 62.98% and 72.25% compared to ΔrpfB::rpfB at 24 and 36 hpi, respectively (Fig. 7d). These results indicate that iaepABCDE suppresses RpfB-dependent DSF turnover. To further clarify whether this effect depends on pH, we monitored both the culture and cytoplasmic pH of strains XC1, ΔiaepABCDE, and ΔiaepABCDE::iaepABCDE in XYS medium. No significant differences were observed in either culture or cytoplasmic pH among these strains (Fig. S6), indicating that iaepABCDE suppresses DSF turnover via a pH-independent mechanism. Notably, exogenous IAA still significantly reduced DSF levels in both the ΔrpfCΔhepABCD and ΔrpfCΔiaePABCDE mutants at 24 hpi (Fig. 7e). This result demonstrated that DSF turnover involves additional pathways other than those mediated by the hepABCD and iaepABCDE clusters.

The IAA-regulated gene clusters were transcribed during XC1 infection of cabbage

To elucidate the roles of key IAA-regulated gene clusters (trpB-A, atpABCDEFGH, hepRABCD, and iaepABCDE) in Xcc pathogenesis, the expression of these cluster was determined during disease development. To this end, we constructed four reporter strains XC1::P_trpB-gusA, XC1::P_atpB-gusA, XC1::P_hep-gusA, XC1::P_iaep-gusA. Histochemical staining detected promoter-driven GUS expression in infected leaf tissues, with XC1::gusA (lacking promoter elements) serving as the negative control. Subsequent GUS quantitative assays showed functional expression of all four clusters during host infection (Fig. 8).

Fig. 8.

The IAA-regulated gene clusters were transcribed during XC1 infection of cabbage. a β‐Glucuronidase (GUS) histochemical staining in infected cabbage leaves at 5 dpi. The numbers below indicate the bacterial colony formation units (CFUs) in the infected leaf tissues. b Quantitative analysis of different promoter‐driven GUS activities per 10.⁸ CFUs of the Xcc strains inside Chinese radish using 4‐methylumbelliferyl‐β‐d‐glucuronide as a substrate at 5 dpi. XC1::gusA represents a negative control strain containing a nonpromoter‐driven gusA gene. Results are presented as the averages of three technical repeats with standard deviations. *p ≤ 0.05, **p ≤ 0.01

Discussion

Xanthomonas is a group of plant pathogens capable of infecting approximately 400 plant species (Büttner and Bonas 2010). Our previous work showed that Xcc synthesizes IAA via a novel L-Trp-dependent pathway and exploits IAA to enhance viability, EPS production, protease activity, and virulence on cabbage, while also suppressing ROS production (Li et al. 2025a). This study demonstrates that host plants employ IAA as a signal to induce DSF turnover and disrupt QS-regulated virulence factor production. Collectively, these findings reveal a dual role of IAA in Xcc–host plant interaction: Xcc employs IAA to enhance infection and adaptation, whereas host plants deploy IAA as a defense mechanism to counter pathogen invasion.

IAA functions not only as a major hormone regulating diverse aspects of growth and development in plants (Enders and Strader 2015; Yu et al. 2014), but also directly modulates numerous processes in plant-associated bacteria (Kunkel and Johnson 2021; Djami-Tchatchou et al. 2022). For example, in Pseudomonas syringae, IAA affects global gene expression, including genes involved in type III secretion and stress response genes (Djami-Tchatchou et al. 2022). Similarly, IAA reprograms the transcriptome and metabolic profile of the rhizobacterium Serratia plymuthica, enhancing motility while suppressing biofilm formation (Rico-Jiménez et al. 2024). Xanthomonas oryzae pv. oryzicola (Xoc) synthesizes IAA via the nitrilase genes NIT24 and NIT29, which act as virulence determinants during rice infection (Zhang et al. 2024). IAA differentially regulates QS-related genes in Pantoea agglomerans pv. gypsophilae (Chalupowicz et al. 2009), while having no effect on the attKLM quorum-quenching operon in Agrobacterium tumefaciens (Yuan et al. 2008). Together with our findings in Xcc, we propose that IAA-regulated functions are strain-specific rather than universally conserved among bacteria.

The modulation of host pH is a common infection strategy employed by most plant pathogens to enhance their pathogenicity and adaptability (Li et al. 2025b). The results of the present study revealed that plant hormone IAA induces DSF turnover in Xcc through a pH-dependent mechanism. IAA strongly induced expression of hepABCD (encoding an RND-family efflux pump) (Fig. 5a and c), which has been shown to promote RpfB-dependent DSF turnover by generating a proton motive force (PMF). This PMF elevates cytoplasmic pH, thereby enhancing RpfB enzymatic activity (Song et al. 2022, 2024). In addition, IAA induced expression of the atpABCDEFGH gene cluster encoding the F₀F₁ ATP synthase. (Kühlbrandt 2019) (Fig. 5a and d). However, the F₀F₁ ATP synthase is a bifunctional enzyme that catalyzes both forward and reverse reactions. Under acidic stress conditions, bacteria operate this enzyme in reverse (as an ATP hydrolase) to pump protons out of the cell, consuming ATP to maintain cytoplasmic pH homeostasis (Koponen et al. 2012; Krah et al. 2023). Together, these findings demonstrate that IAA triggers DSF turnover via a pH-dependent pathway. Furthermore, IAA was found to induce DSF turnover through a pH-independent mechanism. IAA repressed expression of iaepABCDE, which encodes another RND-family efflux pump (Fig. 5a and e). Although the substrate of this efflux pump remains unknown, its deletion did not alter culture and cytoplasmic pH (Fig. S6). How iaepABCDE contributes to DSF turnover thus warrants further investigation.

The IAA sensor in Xcc remains unidentified. Plants are known to utilize TRANSPORT INHIBITOR RESPONSE 1/AUXIN-SIGNALING F-BOX receptors and auxin-binding protein 1 as universal auxin receptors (Morffy and Strader 2022; Friml et al. 2022); however, it seems distinct sensors are employed by different bacteria species. For example, Serratia plymuthica possesses the IAA sensor AdmX, which activates antibiotic biosynthesis (Gavira et al. 2023; Matilla et al. 2018). Similarly, the MarR-type regulator IadR, which represses IAA degradation loci, binds IAA with high affinity in Variovorax paradoxus CL14 (Conway et al. 2022). In addition, TrpR, an L-Trp repressor regulating tryptophan biosynthesis, has been identified as an IAA sensor in both Escherichia coli and S. plymuthica (Marmorstein et al. 1987; Rico-Jiménez et al. 2024). However, bioinformatic analyses detected no homologs of these known IAA receptors or sensors in Xcc. We had hypothesized that HepR could be an IAA sensor considering that HepR functions as a sensor for salicylic acid (SA) and other phenolic compounds in Xcc (Song et al. 2024, 2025), and that IAA induces expression of the hepRABCD gene cluster. Yet, our unpublished data show that IAA still triggers DSF turnover in the ΔhepR mutant and exhibits no detectable binding affinity for HepR. These results thus eliminate HepR as an IAA sensor. We further demonstrated that TrpI, the transcriptional activator upstream of trpB-A in Xcc, is essential for IAA-induced expression of trpB-A (Fig. 6f), yet TrpI itself does not bind IAA (Fig. 6g). Hence, TrpI supposedly functions downstream of the IAA sensor. Despite these mechanistic insights, identification of the bona fide IAA sensor in Xcc will be the key to fully elucidating the underlying regulatory mechanism.

In plants, IAA concentrations range from micromolar levels (Tang et al. 2023; Zhang et al. 2025) to millimolar levels in specific subcellular compartments (Herud-Sikimić et al. 2021). Compared with the defense signal SA, IAA levels in either healthy or pathogen-infected plant tissues are relatively low. Although many plant-associated pathogens produce IAA, its biosynthesis strictly depends on L-Trp availability (Etesami and Glick 2024; Li et al. 2025a). To achieve stable regulation, microbes have evolved positive-feedback loops that enhance IAA production. For example, in the microalga Chlorococcum humicola, exogenous IAA promotes endogenous tryptophan accumulation, which subsequently stimulates IAA synthesis (Chen et al. 2024), and in Bacillus amyloliquefaciens, IAA upregulates the biosynthetic gene ipdC (Costacurta et al. 1994; Vande Broek et al. 1999). The results from the present study identify a similar IAA-trpBA-L-Trp–IAA feedback loop in XC1. Within this loop, exogenous IAA was found to significantly induce the expression of the L-Trp biosynthesis gene cluster trpB-A, thereby increasing L-Trp production (Fig. 5a, b and Fig. 6c). Given that L-Trp serves as the key precursor for IAA biosynthesis in Xcc (Li et al. 2025a), overexpression of trpB-A substantially elevated intracellular IAA levels (Fig. 6d). Through the IAA feedback loop, Xcc hijacks host-derived IAA as an environmental cue to initiate a more aggressive infection strategy. The feedback loop ensures that even low levels of host IAA are detected and amplified, leading to an increase in bacterial endogenous IAA. This elevation in endogenous IAA subsequently enhances virulence traits, such as exopolysaccharide production and reactive oxygen species (ROS) degradation, directly improving the pathogen's fitness within the plant (Li et al. 2025a, b). The resulting DSF turnover may represent a lifestyle switch, potentially facilitating a more stealthy or persistent infection. The evolution of this IAA feedback loop in the phytopathogen Xcc further highlights the importance of IAA in shaping Xcc-host plant interactions. Whether this feedback loop is conserved across phytopathogens is of future research interest.

This study establishes in vitro evidence for a novel mechanism of phytohormone indole-3-acetic acid (IAA)-mediated diffusible signal factor (DSF) turnover in Xcc (Fig. 1 and Fig. 2). We further confirmed the expression of a key IAA-regulated gene during Xcc infection (Fig. 8), supporting the physiological relevance of this pathway. However, our experimental approach has limitations. The work primarily utilized defined XYS medium, designed to mimic xylem fluid but inherently simplifying the highly complex and dynamic host environment. Critical factors—including spatiotemporal IAA gradients, concurrent phytohormone fluxes, host immune responses, and plant microbiome interactions—are not fully recapitulated in vitro. Notably, we revealed a synergistic interaction between IAA and SA in regulating DSF turnover and cytoplastic pH (Fig. S7), suggesting co-modulation by multiple hormonal signals in planta. Consequently, the proposed model serves as a foundational framework for host–pathogen signaling interplay. The physiological mechanism and relative contribution of IAA-mediated DSF turnover during infection require validation under conditions more closely approximating the natural context. To fully elucidate this pathway within the native plant-pathogen interface, future work will employ plant mutants with altered IAA homeostasis (e.g., IAA-overexpressing Arabidopsis) in combination with bacterial reporter strains. These reporters, capable of real-time monitoring of DSF dynamics and pathogen gene expression in planta, will be essential for validating the proposed model.

Conclusions

Taken together, we propose a schematic model illustrating plant-derived IAA-induced DSF turnover in Xcc (Fig. 9). During infection, Xcc perceives and takes in host-derived IAA. Inside the bacterium, IAA activates transcription of the tryptophan biosynthesis gene cluster trpB-A via the transcription factor TrpI. This increases tryptophan production, providing precursors for IAA biosynthesis and establishing a self-reinforcing IAA synthesis loop. Consequently, intracellular IAA content is increased via both host uptake and de novo synthesis through this positive-feedback cycle. Elevated intracellular IAA further upregulates two gene clusters: atpABCDEFGH, encoding the F₀F₁ ATP synthase, and hepABCD, encoding an RND-family efflux pump. The combined activities of these systems generate a PMF and elevate cytoplasmic pH under acidic extracellular conditions, thereby stimulating RpfB activity and promoting DSF turnover through a pH-dependent pathway. In parallel, IAA downregulates the RND efflux pump gene cluster iaepABCDE, which also contributes to DSF turnover but through an RpfB-dependent, pH-independent mechanism.

Fig. 9.

Schematic model for plant-derived IAA-mediated induction of DSF turnover in the phytopathogen Xcc. During Xcc infection of the host plant, Xcc senses and takes in host plant-derived IAA. In Xcc, IAA positively regulates the transcription of the tryptophan biosynthesis gene cluster trpB-A to promote L-Trp production. IAA utilizes the precursor L-Trp to synthesize IAA in Xcc, establishing a self-reinforcing IAA synthesis loop. When IAA accumulates at higher concentrations, it positively regulates the transcription of the atpABCDEFGH and hepABCD gene clusters, leading to increased expression of the F₀F₁-ATP synthase and RND efflux pump, respectively. atpABCDEFGH and hepABCD generate a proton-motive force (PMF), resulting in an increase in cytoplasmic pH. The elevated cytoplasmic pH enhances the enzymatic activity of RpfB to induce DSF turnover. IAA negatively regulates the RND efflux pump iaepABCDE, which suppresses RpfB-dependent DSF turnover via a pH-independent pathway. IAA: indole-3-acetic acid, L-Trp: L-tryptophan, DSF: diffusible signaling factor, RND family efflux pump: resistance-nodulation-cell division family efflux pump

Materials and Methods

Bacterial strains and culture conditions

The bacterial strains and plasmids used in this study are detailed in Table S1. The Xcc wild-type strain XC1 and its mutant derivatives were cultured at 28 °C in: XYS medium (0.7 g·L⁻¹ K₂HPO₄, 0.2 g·L⁻¹ KH₂PO₄, 1 g·L⁻¹ (NH₄)₂SO₄, 0.1 g·L⁻¹ MgCl₂·6H₂O, 0.01 g·L⁻¹ FeSO₄·7H₂O, 0.001 g·L⁻¹ MnCl₂·4H₂O, 5 g·L⁻¹ sucrose, and 0.0625% yeast extract; pH 7.0); nutrient yeast glycerol (NYG) medium (5 g·L⁻¹ peptone, 3 g·L⁻¹ yeast extract, and 20 g·L⁻¹ glycerol); or nutrient agar (NA) medium (5 g·L⁻¹, 3 g·L⁻¹ beef extract, 10 g·L⁻¹ sucrose, and 1 g·L⁻¹ yeast extract; pH 7.0). Recombinant plasmid construction was performed in Escherichia coli strains grown in Luria–Bertani (LB) medium (5 g·L⁻¹ yeast extract, 10 g·L⁻¹ peptone, and 10 g·L⁻¹ sodium chloride) at 37 °C. Antibiotics, namely rifamycin (Rif) at 25 μg·mL⁻¹, kanamycin (Kan) at 50 μg·mL⁻¹, carbenicillin at 100 μg·mL⁻¹, and gentamicin (Gm) at 20 μg·mL⁻¹ were added as required. Bacterial growth was assessed by measuring optical density at a wavelength of 600 nm (OD₆₀₀).

Gene deletion and functional complementation analysis

Gene deletion and complementation procedures were conducted according to the methodology established by He et al. (2006a). For gene deletion, a fusion DNA fragment containing the upstream and downstream regions within approximately 500-bp of the target gene was cloned into the suicide vector pK18mobsacB using a ClonExpress MultiS one-step cloning kit (Vazyme Biotech Co., Ltd., Nanjing, China). The recombinant plasmid was then introduced into XC1 via mating. The kan-resistant colonies were selected on NA plates supplemented with 25 μg mL⁻¹ Rif and 5% (w/v) sucrose. The in-frame gene deletion mutants were verified through polymerase chain reaction (PCR) amplification and DNA sequencing. All primers used in this study are listed in Table S2.

To perform single-copy complementation of the target gene, a DNA fragment containing the coding region along with upstream sequence of the translational start codon was amplified via PCR and cloned into the mini-Tn7T-Gm delivery vector. The resulting constructs were electroporated into Xcc strains using the approach described by Jittawuttipoka et al. (2009). For multiple-copy complementation, the full-length coding sequence of the target gene was amplified and cloned into the multiple cloning site of the pBBR1-MCS2 expression plasmid. The resultant construct was transferred into Xcc strains through triparental mating with E. coli pRK2013 as a helper strain. Primers for complementation are provided in Table S2.

Generation of overexpression strains

To construct Xcc strains overexpressing the iaaHM gene, the coding region of iaaHM was PCR-amplified along with a 485-bp region upstream of the translational start codon of groES (a strong promoter from Xcc). This fragment was cloned into the Mini-Tn7T-Gm delivery vector. The recombinant plasmid was introduced into strains XC1 and ΔrpfC via electroporation using published protocols (Jittawuttipoka et al. 2009), generating XC1::iaaHM and ΔrpfC::iaaHM, respectively. Similarly, to generate the trpB-A overexpression strains XC1::trpB-A and ΔrpfC::trpB-A, the trpB-A operon with the same 485-bp groES promoter was cloned into the Mini-Tn7T-Gm vector. Electroporation-competent XC1 and ΔrpfC cells were then transformed with the recombinant plasmid.

Extraction, purification, and quantitative analysis of DSF using ultrahigh-performance liquid chromatography-time of flight mass spectrometry (UPLC-TOF MS) and HPLC

Xcc cultures were adjusted to a pH of 4.0 and extracted with an equal volume of ethyl acetate. Ethyl acetate was subsequently removed via rotary evaporation at 30 °C. After solvent evaporation, the residue was dissolved in 100 μL of HPLC-grade methanol and analyzed using both UPLC-TOF MS (Agilent, Santa Clara, CA, USA) and HPLC (Agilent, Santa Clara, California) with a C18 reversed-phase column (Zorbax XDB; 5 μm, 4.6 mm × 150 mm, Agilent, Santa Clara, CA, USA). For UPLC-TOF MS analysis, samples were eluted isocratically with methanol/water (80:20, v/v) containing 0.1% formic acid at a flow rate of 0.4 mL·min⁻¹. For HPLC analysis, isocratic elution was performed with methanol/water (77:23, v/v) containing 0.1% formic acid at a flow rate of 1.0 mL·min⁻¹.

HPLC-based quantitative analysis of IAA levels in XC1 cultures

The extraction and quantification of IAA in Xcc cultures were conducted as described by Li et al. (2025a). Briefly, Xcc cultures grown in XYS medium were acidified to pH 3.0 with HCl at 12–48 hpi and extracted with an equal volume of ethyl acetate. Ethyl acetate was then removed via rotary evaporation at 30 °C. After evaporation, the residues were dissolved in 100 μL methanol for HPLC analysis using a C18 reverse-phase column (Zorbax XDB, 5 μm, 4.6 × 150 mm; Agilent, Santa Clara, CA, USA). The mobile phase (methanol/water, 90:10, v/v, 0.1% acetic acid) was delivered at 0.8 mL·min⁻¹, followed by a 40-min linear gradient to 65% methanol.

Generation of gusA-fusion reporter strains and GUS activity assay

The gusA-fusion reporter strain used to monitor trpB-A expression was constructed using the approach outlined by Chen et al. (2020). DNA fusion fragments containing the trpB-A promoter region, T0T1 terminator, and gusA gene were cloned into the pMini-Tn7T-Gm plasmid. Subsequently, the construct was electroporated into strain XC1, as previously described.

GUS activity assays were performed using a modified version of the approach described by Chen et al. (2020). Reporter strains were cultured in XYS medium with or without IAA for 12–36 h at 28 °C. Bacterial cells from a 500 μL culture volume were collected by centrifugation (12,000 rpm, 10 min, 4 °C), washed with PBS, and resuspended in 300 μL of bacterial lysis buffer (TieChui, ACE Biotechnology, Nanjing, China). After incubation at 4 °C for 30 min, lysates were clarified by centrifugation (8,000 rpm, 5 min). Then, 10 μL of the supernatant was added to 250 μL of MUG solution (containing 1 mM 4-methylumbelliferyl β-D-glucuronide, 50 mM PBS, 5 mM dithiothreitol, and 1 mM EDTA, pH 8.0) and incubated at 37 °C. Reactions were terminated after 5 min with 800 μL of 0.4 M Na₂CO₃. Fluorescence intensity (Excitation: 365 nm, Emission: 455 nm) was measured the quantify GUS activity using a SpectraMax M5 fluorescence microplate reader (Molecular Devices, San Jose, CA, USA).

RpfB expression, purification, and in vitro DSF turnover activity assays

The expression and purification of RpfB were performed using E. coli BL21(DE3) cells harboring the pET28a-RpfB expression vector, in accordance with previously published protocols (Song et al. 2022). Protein expression was induced with 0.1 mM isopropyl-β-D-thiogalactopyranoside (IPTG) when the culture reached an OD₆₀₀ of 0.5–0.6, followed by incubation at 16 °C for 16 h. Recombinant protein purification was conducted using His-Pur Ni–NTA Resin (Thermo Fisher Scientific, Massachusetts, USA) following the manufacturer's instructions. The elution buffer contained 25 mM HEPES, 100 mM NaCl, 100 mM MgCl₂·6H₂O, 100 mM (NH₄)₂SO₄, and 250 mM imidazole (pH 7.2).

In vitro DSF turnover assays were conducted as previously described (Song et al. 2022). Reaction mixtures (300 μL total volume) contained 100 mM K₂HPO₄-KH₂PO₄, 10 mM MgCl₂, 2 mM EDTA, 0.1% (v/v) Triton X-100, 5 mM ATP, 0.5 mM reduced CoA, 0.2 mM DSF, and 10 μg purified His-tagged RpfB (with 0.1 mM IAA, without IAA, or heat-inactivated). Reactions were incubated at 28 °C for 15 and 30 min, then acidified to pH 4.0 with 1 M HCl. Quantitative analysis of DSF in the reaction mixture was then performed as described above.

Culture and cytoplasmic pH measurement

Culture pH was measured using a calibrated pH meter (METTLER TOLEDO, Shanghai, China). Cytoplasmic pH determination in Xcc cells was performed according using the method established by Song et al. (2022). Recombinant Xcc strains carrying the pBBR-pHluorin-mCherry plasmid were used to measure cytoplasmic pH. Fluorescence intensity levels were recorded using a SpectraMax M5 fluorescence microplate reader (Molecular Devices, San Jose, CA, USA) with the following fluorescence parameters: pHluorin (Excitation: 488 nm, Emission: 510 nm) and mCherry (Excitation: 587 nm, Emission: 610 nm). Measurements were conducted at 12-h intervals throughout the cultivation period.

RNA sequencing (RNA-Seq) and data analysis

The XC1 strain was grown in XYS medium supplemented with or without 100 μM IAA at 28 °C for 24 h. Bacterial cells were collected by centrifugation at 10,000 rpm for 40 min at 4 °C, with three independent biological replicates prepared for each condition. RNA-seq analysis was conducted by Shanghai Personal Biotechnology Co., Ltd. (Shanghai, China) using the Illumina HiSeq system, as previously described (Song et al. 2022).

Digital PCR analysis

XC1 cells were collected from XYS cultures supplemented with 0 or 100 μM IAA at 24 hpi. Total RNA was isolated using a RNeasy Miniprep Kit (Qiagen, Hilden, Germany), followed by genomic DNA (gDNA) removal with a gDNA Eraser. Complementary DNA synthesis was performed using the PrimeScript RT Reagent Kit (Takara Bio, Shiga, Japan). Digital PCR analysis was performed as per the manufacturer’s instructions using the primers listed in Table S2. Amplification was performed using the Probe qPCR Mix MultiPlus (TaKaRa Bio Inc., Shiga, Japan) and an Automated AD3207 Digital PCR System (Pilot Gene Technology (Hangzhou) Co., Ltd., Hangzhou, China).

UPLC-MS/MS-based quantitative analysis of endogenous L-Trp levels in Xcc

Xcc strains were cultured in XYS medium for 24 h. A 5 mL culture aliquot was then centrifuged at 10,000 rpm for 40 min and washed with PBS. Collected bacterial cells were homogenized in 300 μL of extraction solvent (methanol/acetonitrile/water, 2:2:1, v/v/v) using grinding beads (60 Hz, 3 min). After two rounds of vortexing (30 s) and centrifugation (12,000 rpm, 10 min), the supernatant was collected. The extract was mixed with 400 μL of chloroform, vortexed (30 s), and centrifuged (12,000 rpm, 10 min). The aqueous phase was collected and evaporated to dryness. The residues were dissolved in 100 μL of methanol/water (30:70, v/v) for UPLC-MS/MS analysis.

Chromatographic separation was performed on an ACQUITY UPLC H-Class system (Waters, Milford, MA, USA) with an ACQUITY UPLC HSS T3 column (100 × 2.1 mm, 1.7 μm; Waters, Milford, MA, USA) at 45 °C. Mobile phase A (0.1% formic acid in water) and B (0.1% formic acid in acetonitrile) were delivered at 0.4 mL/min with a 1–100% B linear gradient over 8 min. A Xevo TQ-XS triple quadrupole mass spectrometer (Waters, Milford, MA, USA) was operated in multiple reaction monitoring (MRM) mode with the following parameters: capillary voltage 1 kV (positive ionization mode), source temperature 150 °C, desolvation gas temperature 450 °C, collision gas argon (> 99.999% purity). Data acquisition was conducted with the MassLynx 4.2 software.

UPLC-MS/MS-based quantitative analysis of endogenous IAA level in Xcc

Xcc strains were cultured in XYS medium for 24 h. Bacterial cells from a 100 mL culture volume were collected by centrifugation (10,000 rpm, 50 min) and washed with PBS. Collected bacterial cells were extracted in 1 mL methanol:formic acid:water (70:1:29, v/v/v) using grinding beads (60 Hz, 3 min). After two rounds of extraction with vortexing (30 s) and centrifugation (12,000 rpm, 10 min), the combined supernatant was evaporated to dryness and dissolved in 100 μL methanol:formic acid:water (70:1:29 v/v/v) for UPLC-MS/MS analysis.

Separation was achieved using an ACQUITY UPLC BEH C18 column (100 × 2.1 mm, 1.7 μm; Waters, Milford, MA, USA) at 45 °C with mobile phases and gradient conditions that were identical to those used for L-Trp analysis. The Xevo TQ-XS MS system was operated in MRM mode with negative ionization (capillary voltage 1 kV). Other parameters were the same as those used for L-Trp detection. Data were processed using MassLynx 4.2.

TrpI protein expression and purification

The trpI coding sequence was PCR-amplified using the primers listed in Table S2 and subsequently cloned into the pET-14b expression vector. The recombinant plasmid was transformed into E. coli BL21(AI) competent cells via heat-shock transformation. For TrpI protein induction, cultures in the exponential growth phase (OD₆₀₀ = 0.5–0.6) were supplemented with 0.1 mM IPTG and 0.2% (w/v) L-arabinose, followed by incubation at 16 °C for 16 h.

His-tagged TrpI was purified using HisPur™ Ni–NTA Resin (Thermo Fisher Scientific, MA, USA) according to the manufacturer’s instructions. The eluted protein was further subjected to buffer exchange and purification through a Superdex 200 gel filtration column (GE HealthCare Technologies Inc. Illinois, USA) equilibrated with 20 mM Tris, 200 mM KCl, 1 mM EDTA, 1 mM dithiothreitol (DTT), and 10% (v/v) glycerol (pH 8.0).

EMSA analysis

EMSA was performed using a modified version of previously published methods (Chen et al. 2020). A 288-bp DNA probe spanning positions –243 to + 45 of the trpB promoter region was PCR-amplified using Cy5-labeled primers (Cy5-P_trpB-F/R). Both Cy5-labeled and unlabeled probes were incubated with serially diluted TrpI protein (0–200 ng) in EMSA binding buffer containing 20 mM Tris (pH 7.9), 2 mM DTT, 10 mM MgCl₂, 5% (v/v) glycerol, 40 μg/mL bovine serum albumin, and 100 ng/mL sonicated salmon sperm DNA. Reaction mixtures were incubated for 30 min at 25 °C and then subjected to electrophoretic separation on 4.5% non-denaturing polyacrylamide gels in 0.5 × Tris–borate-EDTA buffer. Electrophoresis was performed at 125 V for 80 min at 4 °C. Fluorescent DNA was visualized using a Starion FLA-9000 scanner (Fujifilm, Japan).

Plant GUS histochemical staining and in planta bacterial CFU quantification

Xcc strains were cultured overnight in NYG liquid medium and adjusted to an OD₆₀₀ of 0.1 (approximately 1.0 × 10⁸ CFU/mL). Bacterial suspensions were then used to inoculate leaves of cabbage (Brassica oleracea cv. Jingfeng-1) using the leaf-clipping method (Chen et al. 2020). At 5 days post-inoculation, infected leaves were harvested. A subset of leaves was subjected to GUS histochemical staining and β-glucuronidase activity analysis according to Li et al. (2014), while another subset from a parallel experiment was used to quantify in planta Xcc CFUs. For each strain per experiment, three leaves were used for GUS staining and five for CFU quantification, with the entire experiment being performed twice. The strain XC1::gusA served as the negative control.

Statistical analyses

All experimental procedures were performed in triplicate unless otherwise stated. Statistical analyses were conducted using analysis of variance (ANOVA) in the JMP software (version 5.0, SAS Institute Inc., Cary, NC, USA). Treatment effects were deemed statistically significant based on the F statistic at a significance level of p ≤ 0.05.

Abbreviations

Xcc

Xanthomonas campestris pv. campestris

IAA

Indole-3-acetic acid

QS

Quorum sensing

DSF

Diffusible signaling factor

RND

Resistance‐nodulation‐cell division

L-Trp

L-tryptophan

XYS

Xylem-Sucrose

hpi

Hours post-inoculation

HPLC

High-performance liquid chromatography

GUS

β-Glucuronidase

PMF

Proton motive force

S. plymuthica

Serratia plymuthica

NYG

Nutrient yeast glycerol

NA

Nutrient agar

Rif

Rifamycin

Kan

Kanamycin

Gm

Gentamicin

OD₆₀₀

Optical density at a wavelength of 600 nm

PCR

Polymerase chain reaction

EMSA

Electrophoretic mobility shift assays

UPLC-TOF MS

Ultrahigh-performance liquid chromatography-time of flight mass spectrometry

IPTG

Isopropyl-β-D-thiogalactopyranoside

gDNA

Genomic DNA

MRM

Multiple reaction monitoring

DTT

Dithiothreitol

Authors’ contributions

Conceptualization: Si-Nan Li, Kai Song, Lian Zhou, Ya-Wen He; Methodology: Si-Nan Li, Ming-Lei Zhang, Ying Cui, Lin Li, Chitti Thawai, Lian Jiang, Dong-Lan Tian, Yu-Cheng Gu; Formal analysis and investigation: Si-Nan Li, Ming-Lei Zhang, Ying Cui, Lin Li, Chitti Thawai, Lian Jiang, Dong-Lan Tian, Yu-Cheng Gu; Writing-original draft preparation: Si-Nan Li; Writing-review and editing: Si-Nan Li, Kai Song, Lian Zhou, Ya-Wen He; Funding acquisition: Kai Song, Lian Zhou, Ya-Wen He, Ying Cui; Resources: Kai Song, Lian Zhou, Ya-Wen He. Supervision: Kai Song, Lian Zhou, Ya-Wen He.

Funding

This work was financially supported by research grants from the National Natural Science Foundation of China (nos. 32172355 to Ya-Wen He; nos. 32502422 to Kai Song), Startup Fund for Young Faculty at SJTU (SFYF at SJTU to Kai Song), and Syngenta-SJTU-PhD Studentship Project to Ying Cui.

Data availability

The authors confirm that all data generated or analysed during this study are included in this published article and its supplementary information files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors consent for publication.

Competing interests

The authors declare that they have no competing interests.