P. citrinum strain B9 stored as filter paper stocks was revived on Potato dextrose agar medium (PDA; VWR, Singapore), and fresh mycelial blocks from the hyphal margins were used for sub-culturing. Strain B9 was routinely cultivated in prune juice agar (PA) medium (Gu et al., 2023) or PDA plates while potato dextrose broth (PDB; VWR, Singapore) was used for culture filtrate extraction and IAA estimation. The plates were incubated at 28°C in the dark for 7 days.

Seedlings of A. thaliana ecotype Columbia (Col-0) wild type, taa1 mutant line, and the transgenic line expressing the auxin reporter DR5rev:GFP (Benková et al., 2003; Friml et al., 2003) were used. Arabidopsis seeds were generally stored at 4°C in the dark to allow stratification and were surface sterilized by soaking in 70% ethanol for 5 min, treated with 10% commercial bleach for 2 min followed by five times rinsing in sterile distilled water. Seeds were germinated on plates containing full Murashige and Skoog basal salts (Sigma-Aldrich; no. M5524) with 1% sucrose and 1% agar (MS-agar). Plates with germinating seeds were incubated and grown vertically in a plant growth chamber AR95L (Percival Scientific, USA) under long-day conditions (16-h/8-h light/dark photoperiod) at 22°C with a relative humidity of 60% and 120 μmol m⁻² s⁻¹ light intensity.

Certified standards of IAA, auxin biosynthesis inhibitor L-Kynurenine, and L-tryptophan (Trp) were purchased from Sigma-Aldrich (Darmstadt, Germany). Standard stock solution (1,000 μg/mL) of IAA was prepared in methanol and stored at −20°C in the dark. Stock solution was used to prepare working standard solutions for analytical experiments. For L-Kynurenine, a 1-mM stock solution was prepared in DMSO, which also served as a solvent only or mock control in parallel. Trp dissolved in distilled water (1 g/mL stock solution) was added to PDB at the indicated final concentration. Propidium iodide was purchased from Invitrogen (Carlsbad, USA). Stock solution (1 mg/mL) was prepared in distilled water with 10 μg/mL propidium iodide used as the working concentration for plant cell wall visualization. For LC-MS analysis, acetonitrile with 0.1% formic acid (Optima LC-MS grade) and methanol (Optima LC-MS grade) were obtained from Fluka Honeywell (Charlotte, NC, USA). Milli-Q water was used for the preparation of the mobile phase (Millipore, Burlington, MA, USA).

A plant–fungus interaction was fostered on a sucrose-free MS agar medium to assess the plant growth promotion by P. citrinum B9. Seven-day-old seedlings were transferred to sucrose-free MS agar medium (square dishes 17 cm × 17 cm) with or without mycelial plugs of B9 grown in PDA for 4 days, by locating their primary roots a few centimeters above the fungal inoculum and grown for another 7 days. For diffusible compounds (IAA), 7-day-old Arabidopsis seedlings were transferred to the vertical split-agar plates prepared with sucrose-free MS agar medium, which was segmented horizontally into the shoot and root sections and inoculated with B9 for 7 days in the root region supplemented with or without the inhibitor L-Kynurenine at the required concentration (Meier et al., 2020). Seven days after the treatment, the media plates were scanned using the EPSON V330 photo scanner, and the root length and area were measured using ImageJ software as previously described (Liu et al., 2018).

For IAA extraction, fresh mycelial plugs of P. citrinum B9 grown in PDA or PA for 4 days were inoculated into PDB supplemented with or without 1 mg/mL Trp and/or L-Kynurenine. The flasks were incubated at 28°C, 180 rpm on an orbital shaker in the dark for the prescribed period. Culture supernatant was collected by filtering through four layers of Miracloth and centrifuged at 10,000 rpm for 10 min. Such cell-free extract was then used for the estimation of IAA using LC-MS analysis. For the estimation of IAA, the cell-free extract was then passed through a sterile filter membrane (0.2 μm), and the resulting extract was vacuum-dried by a CentriVap concentrator. The samples were reconstituted in 80% methanol (v/v) before analysis.

LC-MS of the analyzed samples was acquired on an Agilent 1290 Infinity II coupled to an Agilent 6400 series Triple Quadrupole (Agilent, Santa Clara, CA, USA). The ultrahigh-performance liquid chromatography (UHPLC) system Agilent 6490 was equipped with a Kinetex C18 column (100-mm length, 2.1-mm inner diameter, 1.7-μm particle size, and 100 Å, Phenomenex, Torrance, CA, USA) heated at 35°C and the auto-sampler temperature was set at 4°C.

For the detection of IAA, 10 µL of fungal culture filtrate extracts was injected and separation was performed at a constant flow rate of 0.3 mL/min, in a gradient of Solvent A (water acidified with 0.1% formic acid) and B (acetonitrile acidified with 0.1% formic acid). The gradient program was as follows: 5% B (0–1 min); 5%–100% B (1–10.5 min); 100% B (10.5–13.4 min); re-equilibration from 100% to 5% B (13.4–13.5 min); and reconditioned with 5% B (13.5–16.5 min). Electrospray ionization was performed in negative ion mode in scheduled Multiple Reaction Monitoring (MRM) modes, with the following source parameters: a drying gas (N₂) temperature of 250°C with a flow of 12 L/min, a nebulizer gas pressure of 35 psi, a sheath gas temperature of 350°C with a flow of 11 L/min, and a capillary voltage of 4,000 V. Data processing was performed using the associated MassHunter software B.10.00.

In order to investigate the impact of fungal inoculation on the distribution of auxin maxima in roots, seedlings of DR5rev:GFP Arabidopsis reporter line were co-cultivated with B9 with or without the auxin inhibitor L-Kynurenine as mentioned above (Meier et al., 2020). DMSO was used as the mock control in parallel for each treatment. To assess the effect of P. citrinum B9-derived IAA on plant growth, a 100-μL aliquot of B9 exudates or culture filtrate was applied to the roots of Arabidopsis DR5rev:GFP line for 15 min. For imaging, roots of the DR5rev:GFP reporter line were stained with 10 μg/mL propidium iodide for 2 min, rinsed in water, and mounted on a glass slide with a coverslip. Imaging was done on a Leica TCS SP8 X inverted confocal system equipped with an HC Plan Apochromat 20×/0.75 CS2 Dry objective or a 63×/1.40 CS2 Oil objective. Green fluorescence (GFP) was excited at 488 nm and detected at 500–530 nm and red fluorescence (propidium iodide) was excited at 561 nm and detected at 600–700 nm. All parts of the system were under the control of the Leica Application Suite X software package (release version 3.5.5.19976).

Statistical analysis was carried out using GraphPad prism (GraphPad Software, San Diego, CA) and the values of the treatments were represented as mean with standard error. The significance of differences between the treatments was statistically evaluated using Student’s t-test and post-hoc Tukey’s HSD (honestly significant difference) test and significance was considered at a probability level of p < 0.05 (*), p < 0.01(**), and p < 0.001 (***).

Our previous studies confirmed that P. citrinum isolate B9 promotes significant plant growth under both gnotobiotic and in vivo conditions in the Brassicaceae green leafy vegetable Choy Sum (Gu et al., 2023). A. thaliana serves as a model plant system to unravel the molecular and physiological responses of plants to microbial inoculation. Here, we started with co-cultivation experiments using MS agar medium initially to study the responses of A. thaliana to P. citrinum B9 using plant–fungal interaction assays ( Figure 1A ). Inoculation of B9 increased the primary root growth and the lateral root formation that ultimately resulted in a threefold increase in the root area of Arabidopsis wild-type (WT) Col-0 seedlings at 7 days of growth ( Figure 1B ). Inoculation of B9 further promoted a twofold increase in leaf area, together with root growth correlating to a doubling of the total fresh weight of inoculated plants ( Figure 1B ).

Plant growth promotion by microorganisms is often attributed to the production of secondary metabolites, particularly the phytohormones such as auxins, cytokinins, and gibberellins, of which the IAA is the main auxin implicated in the regulation of cell division and elongation and the remodeling of root architecture in plants. Since P. citrinum B9 inoculation improved the root architecture and the growth of Arabidopsis, we reasoned whether this beneficial fungus produces bioactive auxin as a regulatory component that aids in root and/or plant growth promotion. Initially, we started with an analysis of the gnotobiotic culture filtrate of P. citrinum B9 by LC-MS together with the IAA standard ( Figure 2 ). The MRM mode was used to assess the direct comparison of retention time associated with precursor-to-product ion m/z transition signals between the IAA standard (1 µg/ml / mL) and the corresponding compound(s) in the fungal exudate samples. Compound identification was confirmed through a product ion scan of the IAA standard and the corresponding fungal metabolite(s). IAA in P. citrinum B9 culture filtrate was determined at a retention time of 5.54 min in extracted single ion chromatogram with a precursor ion of m/z 174 [M-H] (negative ionization) and a product ion of m/z 130 ( Figure 2B ), which perfectly matched the results from the IAA standard.

As the LC-MS analysis confirmed that B9 unequivocally produces IAA and secretes it in the culture medium, we next analyzed the time course of IAA production in this beneficial fungal isolate. The main precursor for IAA synthesis is Trp and reports thus far have implied that IAA biosynthesis mostly occurs via the indole-3-acetamide (IAM) and indole-3-pyruvate (IPA) pathways in fungi while only a few species employ alternate or multiple pathways (Numponsak et al., 2018) for such auxin biosynthesis. Nevertheless, both Trp-dependent and Trp-independent IAA biosynthetic pathways coexist in plants and microbes (Fu et al., 2015). However, in the absence of Trp, IAA produced by B9 was in negligible amounts whereas, upon addition of Trp, there was a tremendous hike in the production of IAA as detected by LC-MS analyses ( Figures 2C–E ). Furthermore, when L-Kynurenine, an IPA pathway inhibitor that selectively targets TAA1/TAR (Trp aminotransferase of the TAA/TAR family) and effectively suppresses auxin biosynthesis (He et al., 2011; Won et al., 2011), was added to the growth medium along with Trp, there was a significant reduction in the auxin. However, L-Kynurenine had no effect on IAA production in P. citrinum B9 grown without Trp ( Figures 2C–E ). In addition, we used quantitative spectrophotometric assays using the Salkowski reagent to determine the total indolic compounds in the cell-free supernatant to analyze the temporal profile of IAA production in P. citrinum B9 (Gordon and Weber, 1951; Tarroum et al., 2022). Consistent with LC-MS analysis data, the addition of Trp led to increased IAA production in B9, whereas treatment with the auxin inhibitor significantly decreased the IAA production and secretion in P. citrinum B9 supplemented with Trp, and there was no significant difference in B9 growth per se in the absence of Trp, nor with or without L-Kynurenine ( Supplementary Figure S1A ). In general, the fungal IAA production reached a maximum at 6 days of B9 growth and declined at 10 days of culture when analyzed periodically ( Supplementary Figure S1B ). Taken together, these results revealed that IAA biosynthesis occurs in P. citrinum B9 via the Trp-dependent pathway and since L-Kynurenine treatment was able to reduce but not completely abolish the IAA levels, we further inferred that multiple pathways might be involved in auxin biosynthesis in P. citrinum B9 in addition to the partly shared IPA pathway that exists in Arabidopsis.

After confirming the IAA production in P. citrinum B9, we were interested to assess the growth-promoting effect of B9-derived auxin on root growth and its relationship with host auxin signaling in A. thaliana. To check if P. citrinum can alter auxin-regulated gene expression, we used seedlings of the auxin biosensor DR5rev:GFP containing an auxin-inducible marker gene and measured the response to inoculation with culture filtrate or exudates of P. citrinum B9 grown in the presence (B9-CF-Trp) or absence of Trp (B9-CF). DR5rev:GFP expression was strictly restricted to the center of the meristematic region in control plants and despite the fact that B9-CF contained a low amount of IAA, its exogenous addition failed to induce the DR5rev:GFP activity and the auxin distribution pattern was similar to the uninoculated control ( Figure 3A ). This might be due to negligible amounts of IAA that are not enough to trigger the auxin signaling. Nonetheless, the period of exposure should also be considered here, as auxin signaling was observed within a short period of exposure of the B9-CF to the root, which is not the case when plant roots are inoculated with B9. In contrast, inoculation with B9-CF-Trp enhanced the auxin signaling in DR5rev:GFP, and interestingly, in addition to enhancing the signaling in the meristematic region, the DR5rev:GFP signal also expanded into the columella (COL) and lateral root caps (LRC) of the reporter line that was similar to exogenous treatment with IAA ( Figure 3B ). As expected, Trp alone did not show any effect on DR5rev:GFP, implying that P. citrinum B9-derived IAA produced upon supplementation with Trp plays a role in modulating auxin signaling in the host plants. In parallel with the distribution of auxin, the DR5rev:GFP epifluorescence further remained comparable under the supply of exogenous IAA and B9-CF-Trp ( Figure 3C ). We conclude that the presence and the form of P. citrinum B9-derived IAA are most likely biologically active in the auxin signaling network of A. thaliana.

The enhanced DR5rev:GFP activity in the auxin biosensor provides evidence for the bioactive functionality of P. citrinum B9-derived IAA. To define more precisely the impact of this beneficial fungus on the host auxin network, we used the inhibitor L-Kynurenine to distinguish between the effect of endogenous auxin and B9-derived IAA. Since L-Kynurenine inhibits IAA biosynthesis via the IPA pathway in Arabidopsis, we inoculated B9 to the DR5rev:GFP seedlings grown with and without the auxin inhibitor. Inoculation of B9 showed an increased level of expression in the primary root apex similar to the B9-CF-Trp enhancement of the DR5rev:GFP expression at the COL and LRC and the relative epifluorescence ( Figure 3D ). On the other hand, treatment with L-Kynurenine depleted the endogenous auxin biosynthesis in Arabidopsis, thus decreasing the DR5rev:GFP signals dramatically. However, P. citrinum B9 inoculation could restore this impaired auxin biosynthesis, and a higher level of concomitant DR5rev:GFP expression and relative GFP epifluorescence was observed in the root apex of B9 inoculated and L-Kynurenine-treated plants ( Figure 3D ). Confocal epifluorescence imaging clearly provided evidence that B9-derived IAA is biologically active and can substitute or compensate for the loss of auxin signaling in the host.

Nevertheless, to assess whether the P. citrinum B9-derived IAA can simulate endogenous auxin and restore growth and development in plants lacking auxin, we designed a co-cultivation experiment with the Arabidopsis taa1 mutant and observed the growth-promoting effect, if any, of P. citrinum B9 ( Figure 4 ) therein. TAA1 is involved in the biosynthesis of IPA, a major precursor for auxin production. Defects in auxin biosynthesis lead to dwarfism and abnormal root development (e.g., shorter primary root and fewer lateral roots) in the taa1 mutant plants (Stepanova et al., 2008). Interestingly, the alleviation of phenotypic defects associated with such auxin-deplete conditions was found in the taa1 mutant plants upon B9 inoculation, thus indicating the production and functionality of fungal auxin during host interaction and its beneficial effects on plant growth and development. A similar suppressive trend via fungal auxin was also found in the L-Kynurenine-treated A. thaliana Col-0 seedlings. The inhibitory effect of L-Kynurenine in root growth was evident in split agar plate assays with Arabidopsis seedlings ( Supplementary Figure S2A ). Treatment with auxin inhibitor showed a negative impact on the overall root length and architecture, while inoculation with P. citrinum B9 could significantly suppress the defects imparted by L-Kynurenine ( Supplementary Figure S2B ). Notably, the treatment with such auxin inhibitor showed no discernible effect on fungal growth or colony morphology ( Supplementary Figure S3 ). Taken together, these results confirm that the beneficial fungus P. citrinum B9-derived IAA is biologically relevant or functional to the host plant and enhances the host auxin signaling. In conclusion, our results unambiguously show that the plant growth-promoting mechanism of the P. citrinum isolate B9 includes the production of auxin/IAA to enhance the nascent auxin signaling and significantly induce growth in the host plant species.