Rhizosphere soil samples of A. truncatum Bunge were taken from three sites in Jiulongshan Mountain Preserve, Beijing, People’s Republic of China (39°57′48″ N, 116°05′00″ E). At each sampling site, lateral roots of four Acer plants in the 10-30 cm soil layer were collected using a sterilized shovel and scissors. Samples of approximately 100 g of soil tightly adhering to lateral roots were collected in sterilized plastic bags, immediately placed on dry ice, and transferred to the laboratory for further work. Initial soil properties of the three sites were as follows: total N 1.3 g kg⁻¹, total P 0.9 g kg⁻¹, pH 8.1 (1:2.5, soil:water), Olsen-P 4.2 mg kg⁻¹, available N (NH₄⁺-N plus NO₃^–N) 17.3 mg kg⁻¹, and exchangeable K 67 mg kg⁻¹.

The serially diluted soil samples were plated on the TCP medium containing (per 1 liter): 5.0 g Ca₃(PO₄)₂, 0.50 g (NH₄)₂SO₄, 0.30 g NaCl, 0.30 g KCl, 0.30 g MgSO₄, 0.03 g FeSO₄, 0.03 g MgSO₄, 0.50 g yeast extract, 10.0 g glucose, and 15.0 g agar. The PSB in the sampling rhizosphere were identified by clear halo zones around their colonies after 3 days of incubation at 30 °C. Experiments were performed in four replicates. The capacity of PSB to solubilize the water-insoluble phosphate was studied by the determination of solubilization index [the ratio of the total diameter (colony + halozone) and the colony diameter] (Li et al., 2020).

Single colonies with clear halos indicating P solubilization were selected and purified (Wang et al., 2020 and Bouizgarne et al., 2023). The purified strain designated MQR6^T was obtained and maintained (i) on tryptone soy agar (TSA) plates at 4°C in a refrigerator for further characterization and (ii) as suspensions supplemented with 30% (w/v) glycerol at ‐80°C. Strain MQR6^T was deposited at the China General Microbiological Culture Collection Center (CGMCC No. 23609^T).

The reference strains Pantoea vagans DSM 23078^T and Pantoea allii DSM 25133^T were obtained from the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSMZ), and the strain Pantoea ananatis CICC 10283^T was obtained from the China Center of Industrial Culture Collection (CICC). These strains were cultured under the same conditions as described above for comparative analyses.

Colony morphology of purified bacterial isolate was studied by streaking on TSA medium, followed by incubation of the plates at 30°C for 24 h. Cell morphology was examined by light microscopy (model 50i, Nikon), and cellular morphology was observed by a scanning electron microscope (FEI Quanta 250 FEG, USA). Gram staining was performed using a bioMérieux Gram stain kit (Hangzhou Tianhe Microorganism Reagent Co.) according to the manufacturer’s instructions.

The temperature range for growth was determined in tryptone soy broth (TSB) liquid medium at 4, 10, 15, 20, 25, 28, 30, 37, 40, and 45°C. Growth at different pH values (3.0–12.0 at 1.0 pH unit increments) was evaluated in the TSB medium for 24 h, using the following buffer systems: pH 3.0–5.0, 0.1 M citric acid/0.1 M sodium citrate; pH 6.0–8.0, 0.1 M KH₂PO₄/0.1 M NaOH; and pH 9.0–12.0, 0.1 M NaHCO₃/0.1 M Na₂CO₃. The range of NaCl concentrations for growth was determined in the TSB medium containing 0–10% NaCl at increments of 1% (w/v). Bacterial growth was measured by an increase in turbidity at 600 nm using a spectrophotometer.

Carbon source utilization tests, enzyme activity tests, and additional physiological and biochemical tests were performed using API-20NE, API 50CH (BioMérieux), and Biolog GEN III MicroPlate systems (Reis et al., 2004). The bacterial inoculation was performed according to the manufacturer’s instructions. The type strains of P. vagans DSM 23078^T, P. allii DSM 25133^T and P. ananatis CICC 10283^T were used as reference strains. The results for API 20NE and API 50CH were obtained after 48 h of incubation as recommended by the manufacturer. When the Biolog system was used, strains were incubated on biological universal growth medium (Biolog) at 30 °C for 24 h. GEN III microplates were inoculated according to the manufacturer’s instructions and incubated at 30 °C for 22 h. Results were captured and analyzed based on an extensive species library in the Biolog GEN III database (Woźniak et al., 2019).

For cellular fatty acid analysis, cell mass of strain MQR6^T was harvested from TSA plates after incubation for 24 h at 30 °C. The fatty acid methyl esters were extracted and prepared according to the methods described by Sasser (1990). The fatty acids methyl ester mixtures were separated and analyzed on an Agilent GC-6890N gas chromatograph using the Sherlock Microbial Identification System with standard MIS Library Generation Software (version 6.0; Microbial ID Inc., Newark, DE, USA).

Genomic DNA was extracted using a Bacterial DNA Kit (Tiangen, Beijing, China) following the manufacturer’s instructions. The 16S rRNA gene was amplified by PCR using the universal primers 27F (5’-AGAGTTTGATCCTGGCTCAG-3’) and 1492R (5’-GGTTACCTTGTTACGACTT-3’) (He and Wan, 2022). and the purified PCR products were sequenced by Sangon Biotech (Shanghai, PR China). The 16S rRNA gene sequences were assembled by using the SeqMan package (DNAStar). The 16S rRNA gene sequence of strain MQR6^T was compared with the sequences available in the National Center for Biotechnology Information (NCBI) GenBank database (https://blast.ncbi.nlm.nih.gov/Blast.cgi) and EzBioCloud (www.ezbiocloud.net/identify) (Kim et al., 2012). Multiple alignments were carried out by using CLUSTAL_X software (Thompson et al., 1997). The phylogenetic trees were constructed using the neighbour joining (Saitou and Nei, 1987), maximum likelihood (Felsenstein, 1981) and maximum parsimony (Tamura et al., 2011) methods with the MEGA version 7.0 program (Kumar et al., 2016). The evolutionary distances were calculated using the Maximum Composite Likelihood Method (Tamura et al., 2004). Bootstrap values were calculated based on 1000 bootstrap replications in each case.

For genome sequencing of strain MQR6^T, Illumina Hiseq TM2500 sequencing was performed at Shanghai Personal Biotechnology Co. (Beijing, PR China). The raw data were filtered and trimmed by AdapterRemoval (version 2.1.7) (Schubert et al., 2016). The trimmed reads were assembled using A5-miseq v20150522 with default parameters (Coil et al., 2015). CheckM v1.0.3 was used to estimate the completeness of the genome (Chen et al., 2020). Protein-coding open-reading frames were predicted by using Glimmer v3.02 (Delcher et al., 2007). Contigs longer than 1 kb and with read coverage of more than 10 were kept for further analysis. The G+C content of the chromosome was determined according to a draft genome sequence. The tRNA genes were predicted by tRNAscan-SE 94 (ver. 1.3.1) and the rRNA genes by Barrnap (0.9-dev) 95 (https://github.com/tseemann/barrnap) (Lowe and Eddy, 1997). Gene function prediction was performed by the Rapid Annotations using Subsystems Technology (RAST v.2.0) server (http://rast.nmpdr.org) (Overbeek et al., 2014) and eggNOG-mapper v2 (http://eggnog-mapper.embl.de) (Cantalapiedra et al., 2021). Metabolic pathways were analyzed by using the KEGG’s Blast KOALA service (Kanehisa et al., 2016).

The average nucleotide identity based on blase (ANIb) among the strain MQR6^T and related species was calculated using JspeciesWS online (Richter et al., 2016). The estimated digital DNA–DNA hybridization (dDDH) values among the strains were calculated by Genome-to-Genome Distance Calculator (GGDC2.0) with the alignment method of BLAST+ (Auch et al., 2010). The partial genome files were uploaded to the GGDC 2.0 web interface (http://ggdc.dsmz.de/ggdc.php#), and Formula 2 was used as recommended for the calculation of dDDH values. The proposed and generally accepted species boundary for ANIb and dDDH values are 95~96 and 70%, respectively (Meier-Kolthoff et al., 2013).

Two methods were used to construct phylogenetic trees of strain MQR6^T and the closely related Pantoea species. The first method used the classification workflows in Genome Taxonomy Database toolkit version 2.0.0 (GTDB-Tk) to identify and concatenate 120 single-copy bacterial marker genes (Chaumeil et al., 2020). The ML phylogenetic tree was established using IQ-TREE 2.2.0 program, SH-aLRT test, 1000 repeated ultrafast guidance, and ModelFinder to determine the best-fit model (Nguyen et al., 2015). The second method uploads the genome sequence data to the Type (Strain) Genome Server (https://tygs.dsmz.de/) (Meier-Kolthoff et al., 2022) for the whole-genome-based taxonomic analysis. The Bacterial Pan Genome Analysis (BPGA) pipeline was used for the pan-genome analyses of strain MQR6^T and the closely related Pantoea species (Chaudhari et al., 2016).

The ability of the strain to solubilize water-insoluble phosphate was measured in liquid media containing either TCP or powdered RP. The RP medium was modified from the TCP medium by adding 5 g L^-1 of rock phosphate instead of TCP. One milliliter of MQR6^T culture (approximately 1×10⁸ cfu mL^-1) was transferred to a 300-mL flask containing 100 mL of medium, followed by shaking (150 rpm) at 30 °C. The non-inoculated TCP and RP media were used as controls. Quadruplicate cultivations were conducted for each medium. The suspensions were sampled up to 96 hours at 12-hour intervals. At every sampling time, 3 mL of culture liquids was sampled and centrifuged at 5000 g (Anke LXJ-IIB) for 20 min to remove biomass and insoluble matter, and the supernatants were used for determination of pH and soluble P concentration. The pH value of the medium was measured with a pH meter. Phosphorus in the supernatant was determined by the molybdenum-blue method using a spectrophotometer at 700 nm (Watanabe and Olsen, 1965).

The secretion of plant growth hormone IAA by strain MQR6^T was measured by colorimetry. The test was performed in the presence and absence of L-tryptophan as the precursor of IAA. One milliliter of bacterial culture (approximately 1×10⁸ cfu·mL^-1) was added to 100 mL TSB medium (with 5 mM L-tryptophan or without) in 250 mL Erlenmeyer flasks, and then incubated on a shaker (30 °C, 150 rpm) for 96 h. The suspensions were sampled at 12-hour intervals. The method to collect culture supernatant was the same as that in detecting P-solubilizing capacity. The production of IAA was screened by mixing 100 µL of bacterial suspension droplets with 100 µL of Salkowski reagent (50 mL 35% v/v HClO_{4 +} 1 mL 0.5 mol L^-1 FeCl₃) on a white ceramic plate based on the color change after 30 min of reaction in the dark at room temperature (Naqqash et al., 2016). Indole compounds react with Salkowski reagent to form a pink chromophore in absorbance at 530 nm using spectrophotometer (UV 3200, Shanghai, China). IAA concentrations were determined using a standard curve made from commercial IAA (Sigma), with the sterile medium as the blank.

Siderophore production was assayed qualitatively as described by Schwyn and Neilands (Schwyn and Neilands, 1987). Briefly, overnight culture of strain MQR6^T was spot-inoculated onto a chrome azurol S (CAS) agar plate and incubated for 5 days at 30° C. The basic principle underlying the test is that when a strong ligand (for example, siderophore) is added to a highly coloured dye-Fe³⁺ complex, the Fe³⁺-ligand complex is formed, and the release of free dye is accompanied by a colour change. When a strong chelator removes the iron from the dye, its color turns from blue to orange.

The experiment was carried out in a greenhouse located at the Experimental Centre of Forestry in North China, Chinese Academy of Forestry, to evaluate the effects of strain MQR6^T on plant growth and nutrient uptake by A. truncatum seedlings. The soil was obtained from the Acer forest in Jiulongshan Mountain Preserve, with the properties as described above. The soil samples were air-dried, passed through a 2-mm sieve and filled into the pots at bulk density of 1.32 g cm^-3. To ensure that the supply of other nutrients was adequate for plant growth, soil was supplemented with basal nutrients at the following rates (mg kg⁻¹ soil): 200 N (NH₄NO₃), 50 Ca (as CaCl₂), 150 K (as KCl), 28 Mg (as MgSO₄), 4 Zn (as ZnSO₄), and 1 Fe (as EDTA-Fe).

There were four treatments in the present study: (1) control, non-inoculated and without the application of P fertilizer (CK), (2) application of P fertilizer only, non-inoculated (P), (3) inoculation with MQR6^T only, without the application of P, and (4) inoculation with MQR6^T plus the application of P (MQR6+P). Monopotassium phosphate (KH₂PO₄) is a highly water-soluble inorganic salt, widely used as a P fertilizer in agricultural soils. The fertilizer KH₂PO₄ was used as P source, and the concentration was 50 mg P kg^-1 soil.

The Acer truncatum seeds were surface-sterilized with sodium hypochlorite (1% w/v) for 30 min and then rinsed extensively with sterilized distilled water. The suspension of overnight bacterial culture (TSB) was diluted in sterile distilled water to a final concentration 10⁸ cfu mL^-1, and the resulting suspensions were used to treat seeds and seedlings. The surface-sterilized seeds were dipped in the inoculum (containing 10⁸ cfu mL^-1) for 15 min and then placed in pots containing 800 g of soil on 25 June 2021. Seeds dipped in medium not containing the strain were used for the non-inoculated treatments. A second inoculation was done at days 45 after seedling emergence at rates of 5 mL of bacterial suspension described above per pot. Plants were watered weekly to maintain 70–80% of field capacity.

Plants were harvested at 290 days after sowing on 10 April 2022. Plant height was recorded by measuring the length from soil surface to the tip of the main stem. Trunk diameter was measured at the base (5 cm from the ground) using a digital vernier caliper with an accuracy of 0.01 mm (Wuxi Kaibaoding Tools Co., Ltd., China). Chlorophyll content was read in the youngest fully-developed leaves using a chlorophyll meter (SPAD-502, Minolta, Osaka, Japan). The plants were separated into shoot and roots. The roots were kept in an icebox, transported to the lab, rinsed with water, and scanned by a scanner at resolution of 400 dpi. Root images were analyzed using WinRhizo Pro 2009b software (Regent Instruments Inc., Quebec, Canada) to calculate the root length and surface area. The shoots and roots were oven-dried at 105°C for 30 min and then at 70°C for 3 days to constant weight to determine dry weights and P uptake. P contents were assayed using the dry ashing digestion method.

One-way analysis of variance was performed using SAS statistical software (SAS 8.1, USA), and significant differences among means were assessed using Tukey’s test at 5% probability (P ≤ 0.05).

Colonies were circular, smooth, mucoid, convex with clear edges, and 0.8–2.0 mm in diameter after 24 h of incubation at 30°C on TSA plates (Figure 1A). Cells were rod-shaped, Gram-stain negative, single, non-spore-forming, measuring 0.5–1.4 μm×1.0–3.0 μm (Figures 1B–D). Bacterium features one or more long flagella (Figure 1D). Growth was found to occur at 10–40°C (optimum, 28–30°C) and at pH 4–11 (optimum, pH 7–9). In TSB medium, growth occurred in the presence of 0–7% (w/v) NaCl (optimum, 0–1%).

The phenotypic properties differentiating between strain MQR6^T and its closest phylogenetic neighbors are shown in Table 1. According to API 50CH tests, strain MQR6^T showed negative results with glycerin and sucrose, which were positive (or weakly positive) for P. vagans DSM 23078^T, P. allii DSM 25133^T and P. ananatis CICC 10283^T. Acid was produced from the fermentation of D-glucose, ribose, D-xylose, galactose, fructose, rhamnose, mannitol, N-acetylglucosamine, maltose, and trehalose by strain MQR6^T and the other three reference strains. With API 20NE, strain MQR6^T was negative for sodium citrate, L-arabinose, inositol, melibiose, and rhamnose but the other three reference strains were positive (at least weakly). Strain MQR6^T grew on D-maltose, but the other three reference strains did not.

The fatty acid analysis revealed that all strains contained C_12:0, C_14:0, C_16:0, C_17:0 cyclo, summed feature 3 (C_16:1 ω7c and/or C_16:1 ω6c), and summed feature 8 fatty acids (C_18:1ω6c and/or C_18:1ω7c) as the major components (Table 2). The major cellular fatty acid profile (>5% of total) of strain MQR6^T was summed feature 3 (C_16:1ω7c and/or C_16:1ω6c), summed feature 8 (C_18:1ω6c and/or C_18:1ω7c), C_14:0, C_16:0, and C_17:0 cyclo.

The 16S rRNA gene sequence (1440 bp) of strain MQR6^T was deposited in GenBank under the accession number OM826981. Based on the analysis of the EzBioCloud database, the strain MQR6^T was related closely to P. vagans DSM 23078^T (98.47%) and P. ananatis CICC 10283^T (98.51% similarity). Phylogenetic trees were reconstructed using the neighbor joining, maximum likelihood, and maximum parsimony methods (Figure 2, Figures S1, S2). All three treeing methods yielded similar phylogeny. Strain MQR6^T was located within the genus Pantoea and had a separate clade, indicating that strain MQR6^T represented a member of a novel species of genus Pantoea.

A total of 7,989,160 reads were obtained from genome sequencing of strain MQR6^T, yielding a genome of 7,674,100 reads in length. The genome was predicted to contain a total of 4548 genes, which included 4473 protein-coding genes, 4 rRNA genes and 71 tRNA genes. There were 31 contigs in strain MQR6^T. The genomic DNA G+C content of strain MQR6^T was 51.3%. The dDDH values between strain MQR6^T and the type strains of the genus Pantoea were 19.5-23.3%, and the average nucleotide identity based on blast (ANIb) between them was lower than 79.7% (Table 3). The phylogenomic tree based on the Type (Strain) Genome Server (TYGS) web also revealed the distinct phylogeny of strain MQR6^T and its close relationship with P. ananatis LMG 2665^T, P. allii LMG 24248^T and P. dispersa DRS002603^T (Figure 3). In a phylogenetic tree based on 120 single-copy genes and the whole genome, strain MQR6^T forms a separate evolutionary branch (Figure 4). We conducted a preliminary analysis of the pan-genome, which showed that 1350 shared orthologous coding sequences were clustered into the core genome of Pantoea, 65,787 were represented in the accessory genome, and 8030 were identified as strain-unique genes. The total number of genes increased in the pan-genome of Pantoea with the rise in the analyzed genome number, suggesting that the pan-genome was open (Figure S3). The previous reports showed that the gene number in the core genomes was highly conserved, while many strain-unique genomes and accessory genomes were thought to contribute to species diversity, indicating that species in the genus Pantoea were multifarious.

The whole-genome shotgun sequencing output has been deposited at DDBJ/ENA/GenBank under the accession JAKYXB000000000. The version described in this paper is JAKYXB010000000.

Functional analysis of the strain MQR6^T genome identified genes associated with the solubilization of phosphate and production of IAA, siderophores and phytohormones that are conducive to plant growth promotion. Eight key genes responsible for IAA production were found in MQR6^T genome, including seven genes related to tryptophan operon (trpS, trpB, trpH, trpR, trpA, trpC, trpE) and the ipd C gene encoding indole pyruvate decarboxylase (Table S1). These results indicate that IPyA pathway may be the main pathway for IAA production in the strain MQR6^T.

The phosphonate-related phn gene cluster is responsible for the release of biologically available phosphate through the bacterial degradation of phosphonates. Our study revealed that strain MQR6^T carries several phn genes, including phnN, phnM etc., showing the capacity to hydrolyze phosphonate into phosphate and alkane (Table S2).

Gluconic acid (GA) is an organic acid that is largely responsible for the solubilization of mineral phosphates. GA biosynthesis is carried out by glucose-1-dehydrogenase along with its co-factor pyrrolo-quinolone quinine. Accordingly, MQR6^T genome annotation indicated the presence of several genes related to gluconic acid biosynthesis and its co-factor genes, including pqqBDEF and gcd. Another organic acid identified in the strain MQR6^T that is relevant to the phosphate-solubilizing trait is 2-ketogluconic acid produced by gluconate 2-dehydrogenase alpha/beta chain and 2-keto-D-gluconate reductase. Moreover, the strain MQR6^T was found to produce other organic acids such as lactic, acetic, glycolic, and succinic (Table S3).

Genomic study showed that MQR6^T may synthesize an enterobactin siderophore involving the entABCEF genes. The siderophore is then exported from the cell using entS and is responsible for recovering iron by complexing. Having several siderophore receptor genes (Table S4), strain MQR6^T may take up siderophores produced by other organisms as well.

Clear halozones were formed around the colonies of strain MQR6^T on inorganic phosphate media, with solubilization index (SI) values from 3.20 to 3.98 (Figure 1C). The strain MQR6^T was able to solubilize water-insoluble TCP and powdered rock phosphate (RP); however, amount of P solubilized was significantly higher in the TCP medium as compared to the RP medium (Figure 5A). The soluble P concentration in the TCP medium ranged between 232 and 559 mg L^-1, with variations over time. By contrast, the soluble P concentration in the RP medium exhibited a range of 159–339 mg L^-1. The highest concentration of soluble P in the two media was observed after 24 h incubation, then P solubilization gradually decreased over time.

The pH of the culture filtrates of strain MQR6^T decreased from an initial level of 7.22 to 3.84 in the TCP medium and to 3.52 in the RP medium (Figures 5C, D). There was a negative correlation between soluble P concentration and pH value of the TCP medium (R² = 0.36, P < 0.01) and RP medium (R² = 0.51, P < 0.01).

Strain MQR6^T showed pink color reaction with Salkowski reagent which indicated the production of IAA (Figure 1F). The highest (166 mg L^-1) concentration of IAA in the medium with L-tryptophan was observed after 24 h incubation (Figure 5B). The siderophore production was detected using CAS, showing orange colonies after incubation due to siderophore-dependent removal of Fe from the dye (Figure 1E), indicating the capacity of strain MQR6^T to exude siderophore.

Inoculation with strain MQR6^T significantly enhanced the shoot and root growth, dry weight accumulation and P uptake of A. truncatum seedlings compared to those grown in non-inoculated soils (Table 4 and Figure 6). Compared with the non-inoculated control treatment, plant parameters in the MQR6^T inoculation treatment increased by 47% (height), 53% (trunk diameter), 15% (SPAD chlorophyll content), 100% (shoot dry weight) and 133% (root dry weight). Phosphorus accumulations in shoots and roots were, respectively, 102% and 79% greater in the MQR6^T inoculated treatment than those in the non-inoculated control. Similarly, P addition alone significantly increased plant height, trunk diameter, dry weight and P uptake when compared to the non-inoculated control treatment. Moreover, combined P addition and inoculation treatment significantly increased plant height, trunk diameter, biomass accumulation and P uptake when compared to the control, the P addition only and the inoculation only treatments. No significant difference was found in root/shoot ratio among the four treatments, and there was no significant difference in shoot and root biomass, SPAD chlorophyll content and P uptake between treatments with P addition and inoculation only.