Transporter-mediated depletion of extracellular proline directly contributes to plant pattern-triggered immunity against a bacterial pathogen

Plants and animals detect pathogens via pattern-recognition receptors (PRRs) that sense pathogen- or microbe-associated molecular patterns (M/PAMPs). In plants, individual tissue cells must locally restrict pathogens, many of which reside extracellularly in the apoplast. PRR activation triggers apoplast-localized defenses (ROS production, antimicrobial secretion, cell wall reinforcement), resulting in pattern-triggered immunity (PTI), but the mechanisms by which PTI limits apoplastic pathogen growth remain unclear. Pseudomonas syringae uses a type III secretion system (T3SS) to deliver effectors that suppress PTI; T3SS genes are induced by host-exuded metabolites (specific sugars and amino acids). Host genetic control of exudation affects infection outcomes (e.g., Arabidopsis mkp1 exudes fewer T3SS-inducing metabolites and is more resistant). Prior observations show MAMP pretreatment restricts P. syringae effector injection and reduces T3SS gene expression during infection. The research question addressed here is whether MAMP-induced changes in the apoplastic metabolite composition causally restrict T3SS deployment and growth of P. syringae, and which specific metabolite(s) and plant transporter(s) mediate this effect.

Existing work establishes that: (1) PRR-mediated PTI in plants restricts pathogen infection by apoplastic defenses, yet the direct growth-limiting mechanisms are not fully defined. (2) P. syringae T3SS induction depends on host-derived metabolites, including certain sugars and amino acids, sensed via defined bacterial receptors; mutants in these pathways show reduced virulence. (3) Arabidopsis mkp1 mutants with reduced exudation of T3SS-inducing metabolites show enhanced resistance, highlighting the role of host exometabolome in disease. (4) MAMP pretreatment limits effector injection and downregulates T3SS gene expression in P. syringae during infection. (5) Transporters such as STP13 and LHT1 modulate sugar and amino acid uptake during immune responses, and prior seedling/suspension cell studies reported broad MAMP-induced changes in exuded metabolites, though leaf apoplast dynamics may differ. Collectively, these studies suggest that modulation of extracellular nutrients/signals is a vulnerability that plants could exploit to suppress bacterial virulence.

The study combined metabolomics, bacterial genetics, plant genetics, and functional assays. Key components: (1) Plant materials and treatments: Arabidopsis thaliana Col-0 and mutant lines (fls2, efr-2, dde2 ein2 pad4 sid2 QKO, sid2-1, lht1-5, lht1-7, lht1-7 sid2-1, prot1-1, prot2-3, prot2-2, prot3-2) grown under controlled conditions. Leaves were syringe-infiltrated with 100 nM flg22, 100 nM elf26, or mock (DMSO). (2) In planta T3SS assays: Leaves pretreated with MAMPs were infiltrated with P. syringae pv. tomato DC3000; AvrPto accumulation assessed by immunoblot; hrpLpromoter::gfp reporter strains used to quantify T3SS gene expression in planta. (3) Apoplastic wash fluid (AWF) extraction: Following MAMP pretreatment (8 h), leaves were infiltrated with H2O containing ribitol internal standard; AWF collected by low-speed centrifugation of stacked leaves in parafilm, clarified, chloroform partitioned to obtain aqueous/organic fractions, lyophilized and reconstituted. MDH enzyme assays assessed cytosolic contamination; pH measured. (4) Bacterial responses in AWF: DC3000 hrpLpromoter::gfp cultured in buffered AWF (adjusted to pH 6.0 with minimal medium) to measure hrpL induction; AvrPto immunoblotting after incubation; growth measured by OD600. Chloroform fractionation localized hrpL- and growth-inducing activities. (5) Metabolomics: GC-MS profiling (derivatization with methoxyamine and MSTFA; Agilent GC-MS; AMDIS analysis; FiehnLib matching). Eight independent Col-0 experiments and four QKO experiments; volcano plots, categorization (amino acids, organic acids, sugars). Absolute quantification for metabolites with differential abundance using external standards normalized to ribitol. (6) Functional metabolite tests: Minimal medium with fructose plus individual metabolites at measured apoplastic concentrations tested for hrpL induction; proline supplementation (100 µM) added to AWF to test rescue of hrpL and growth. (7) 13C-proline uptake kinetics: Leaves infiltrated with 500 µM 13C-proline plus ribitol; AWF collected at 0–60 min; GC-MS distinguished 13C vs 12C proline; time courses compared in mock vs flg22-pretreated Col-0 and QKO; detached leaf controls assessed vascular transport contributions. (8) Bacterial genetics: Construction of DC3000 ΔputA via allelic exchange; putApromoter::gfp reporter; hrpLpromoter::gfp in ΔputA; growth tests in M9 with individual amino acids; hrpL induction by metabolites; in planta virulence/growth and AvrPto abundance for WT vs ΔputA and putA::Tn5 strains. (9) Transporter genetics: Measured apoplastic proline in AWF of prot1, prot2, prot3 mutants; in planta putApromoter::gfp and hrpLpromoter::gfp induction; DC3000 growth in mutants; flg22 impact on apoplastic proline and growth in prot2 backgrounds; 13C-proline uptake assays. (10) LHT1 role: AWF proline levels and 13C-proline depletion in lht1-7 and lht1-5 with/without flg22; AWF bioactivity on hrpL; in planta DC3000 vs ΔputA growth in Col-0 vs lht1-7 under mock or flg22 pretreatment; SA dependence tested in sid2-1 and lht1-7 sid2-1 double mutants. (11) Statistics: Two-tailed t-tests; ANOVA with Tukey’s HSD for multiple comparisons; data pooled across independent experiments where indicated.

• PTI restricts P. syringae T3SS. flg22 or elf26 pretreatment reduced AvrPto accumulation in Col-0 but not in fls2 or efr mutants; QKO plants failed to restrict AvrPto and hrpL expression upon flg22 treatment.
• AWF bioactivity changes with MAMPs: AWF from flg22-treated Col-0 (flg22-AWF) showed reduced hrpL induction and reduced bacterial growth compared to mock-AWF. Differences localized to the aqueous fraction. Apoplastic pH increased modestly after flg22 (6.17 to 6.63) in both Col-0 and QKO, but buffering eliminated pH effects, indicating pH was not causal.
• Metabolomics: Among 96 detected features (62 identified), in Col-0 flg22-AWF 4 features increased and 6 decreased (p < 0.05, ≥2-fold). Salicylic acid increased ~13-fold; threonic acid increased ~4-fold. Changes in QKO overlapped only for threonic and succinic acids, suggesting these are not causal for reduced T3SS induction.
• Proline as a key apoplastic signal: Absolute concentrations in mock-AWF showed proline ~120 µM and serine ~194 µM; most other changed metabolites <20 µM. At measured concentrations, only proline and serine induced hrpL, with proline more potent. flg22 reduced apoplastic proline by ~90 µM. Adding 100 µM proline restored hrpL induction in flg22-AWF to mock levels and partially rescued bacterial growth.
• flg22 accelerates apoplastic proline depletion: 13C-proline levels in AWF declined rapidly; after flg22 pretreatment in Col-0, 13C-proline was >35% lower than mock at 20 and 40 min, indicating increased removal. No flg22-mediated change occurred in QKO.
• Bacterial PutA is essential for proline-dependent virulence: ΔputA failed to grow on proline as sole carbon source but grew on other amino acids; proline-induced hrpL expression required putA, whereas other T3SS-inducing metabolites (e.g., aspartate, citric acid) induced hrpL independently of putA. In planta, ΔputA showed reduced AvrPto accumulation and reduced growth.
• PROT2 regulates baseline apoplastic proline: prot2-3 leaves had ~4-fold higher AWF proline; DC3000 putApromoter::gfp and hrpLpromoter::gfp were higher in prot2-3; DC3000 growth and AvrPto increased in prot2 mutants. The growth enhancement was abolished with DC3000ΔputA, confirming causality of proline. However, flg22 still reduced apoplastic proline and bacterial growth in prot2 backgrounds; 13C-proline uptake increased upon flg22 even in prot2, indicating PROT2 is not required for flg22-induced depletion.
• LHT1 is required for flg22-induced proline depletion and contributes to PTI: In lht1-7 (and lht1-5), flg22 did not reduce AWF proline; flg22 failed to accelerate 13C-proline depletion; flg22-AWF from lht1-7 did not reduce hrpL induction vs mock-AWF. In planta, flg22 reduced DC3000 growth strongly in Col-0 but significantly less in lht1-7; in flg22-treated lht1-7, DC3000 outgrew ΔputA, indicating residual proline signaling. The LHT1 contribution to PTI was independent of salicylic acid: flg22 reduced apoplastic proline similarly in sid2-1; in sid2-1 vs lht1-7 sid2-1, flg22 resistance was weaker in the double mutant, showing LHT1’s SA-independent role.

The study directly addresses how PTI limits apoplastic pathogen growth by showing that depletion of a specific extracellular metabolite—proline—reduces P. syringae T3SS induction and growth. By profiling AWF and manipulating metabolite levels, the authors demonstrate that proline is a potent T3SS-inducing signal present in the Arabidopsis leaf apoplast at physiologically relevant concentrations and that flg22-triggered PTI reduces its abundance by enhancing uptake from the apoplast. Genetic evidence pinpoints the plant transporter LHT1 as essential for this MAMP-induced depletion, while bacterial genetic analysis shows that P. syringae requires putA to sense and utilize proline for T3SS induction and virulence. These findings imply that plants can achieve effective resistance by starving pathogens of critical signals/nutrients rather than solely by producing antimicrobials or reinforcing physical barriers. The work refines our understanding of PTI mechanics, revealing nutrient/signaling deprivation as a direct defensive strategy. It also clarifies that while other T3SS-inducing metabolites remain in the apoplast and maintain residual induction, proline depletion alone reduces total induction below a threshold necessary for maximal virulence. The LHT1-mediated component of PTI operates independently of salicylic acid, indicating parallel defense sectors converge on metabolic remodeling of the apoplast. This strategy may generalize to other amino acids and pathogens, connecting innate immunity to control of the exometabolome.

This work identifies apoplastic proline as a host-derived virulence-inducing signal and nutrient for Pseudomonas syringae and establishes that MAMP (flg22)-induced, LHT1-dependent depletion of extracellular proline directly contributes to PTI by restricting T3SS gene expression and bacterial growth. Plant transporter LHT1 is necessary for the accelerated removal of apoplastic proline upon immune activation, and bacterial putA is required for proline-driven T3SS induction and virulence. The study highlights a defense paradigm in which depletion of a single extracellular metabolite effectively limits pathogen virulence. Future research directions include: delineating the signaling pathways (JA, ethylene, PAD4 sectors) upstream of LHT1 activation; determining post-transcriptional regulation of LHT1 during PTI; dissecting the relative contributions of proline as signal versus nutrient in vivo; time-resolved apoplast metabolomics during PTI and infection; exploring whether similar transporter-mediated nutrient depletions contribute to effector-triggered immunity and influence commensal microbiota.

Key limitations include: (1) Metabolomics primarily assessed a single time point (8 h) post-flg22, which may not capture full temporal dynamics of apoplastic metabolites during PTI and infection. (2) Due to coupling between proline catabolism (nutrition) and T3SS induction (signaling) via putA, the relative contribution of nutrient limitation versus signal deprivation to growth restriction cannot be fully disentangled here. (3) Although LHT1 is necessary for flg22-induced proline depletion, potential redundancy and regulation with other transporters (e.g., PROT family and unidentified effluxers) remain unresolved. (4) The QKO phenotype indicates multi-hormone sector involvement, but exact upstream signaling components that regulate LHT1 activity during PTI are not identified. (5) Residual T3SS-inducing signals (e.g., sugars, other amino acids) persist; their quantitative contributions and interactions with proline depletion need further delineation.