Rhizospheric soil samples were collected from wheat plants cultivated in an experimental field in the botanical garden (28° 52′ 44” NL and 76° 37′ 19″ EL) at Maharshi Dayanand University, Rohtak, Haryana, India. Wheat rhizospheric bacteria were isolated following previously standardized conditions (Sharma et al., 2024). The phosphate solubilizing potential of wheat rhizosphere bacteria was screened as per the Pikovskaya agar plate screening (Pikovskaya, 1948).

The alkali lysis method (Chauhan et al., 2009) was employed for DNA isolation of PSB. Qualitative and quantitative analyses of the DNA were performed using agarose gel electrophoresis and Qubit HS DNA estimation kits (Invitrogen, USA), respectively. The 16S rRNA gene was amplified and sequenced to determine the taxonomic affiliation of the microbes following standardized protocols (Yadav et al., 2023). The Gram staining kit (K001-1KT, Himedia) was used to characterize phosphate solubilizing bacteria (PSB). Optimal growth conditions of PSB were examined across various pH (3, 4, 5, 7, 8, 9, 10, 11 and 12) and temperature ranges (10°C, 15°C, 20°C, 25°C, 30°C, 35°C, 40°C, 45°C, 50°C, 55°C, 60°C) (Rasul et al., 2021). The bacterial growth pattern was monitored in LB broth for 48 h at 37°C with continuous shaking at 200 rpm to assess their doubling time (Wang et al., 2015). The substrate utilization preferences of the identified microbes were evaluated using the Hi Carbo kit (Himedia, KB009A-1KT, KB009B-1KT, and KB009C-1KT). Furthermore, the biochemical properties of the strains were assessed through activities for amylase (Swain et al., 2006), catalase (Iwase et al., 2013), pectinase (Oumer and Abate, 2018), cellulase (Kasana et al., 2008), esterase (Ramnath et al., 2017), and protease (Vijayaraghavan et al., 2017). The antibiotic susceptibility profiles of the microbial isolates were determined using the Combi IV kit (Himedia, OD023) and G-VI-plus (Himedia, OD034). Stress response physiology was assessed by subjecting them to salt, arsenic, and oxidative stress (Yadav et al., 2023). Salt stress tolerance was checked after assessing the growth of PS1 and PS2 in LB broth (5 mL) supplemented with different concentrations of NaCl, KCl, and LiCl (Yadav et al., 2023). Oxidative stress tolerance of PS1 and PS2 was assessed after observing their growth in LB broth (5 mL) supplemented with different concentrations of H₂O₂ (0, 1.0 mM, 2.5 mM, 5.0 mM, 7.5 mM, 10 mM, 12.5 mM. 15.0 mM, 20.0 mM, and 25 mM) (Sharma et al., 2024). Arsenic stress tolerance of PSB was checked after observing their growth in LB broth (5 mL) supplemented with different concentrations of sodium arsenite and sodium arsenate (Sharma et al., 2024).

PS1 and PS2 bacterial genomes were sequenced using Illumina MiSeq with the Nextera XT DNA Library Prep kit. Sequence curation, genome assembly, basement of genome completeness, genome annotation, and genome map creation were performed as described previously (Rasul et al., 2021; Sharma et al., 2024). CRISPR/Cas in the genome was identified using a CRISPR identifier, and antibiotic-resistance genes were identified using the CARD identifier. Protein features responsible for phosphate solubilization, antibiotic resistance, metal/metalloid resistance, and oxidative stress resistance were identified using the Rapid Annotation using Subsystem Technology (RAST) server.¹ Pathogenesis potential was assessed using Island Viewer 4 with default parameters. Phylogenomic characterization of PS1 and PS2 with other Pantoea sp. was plotted using roary_plots.py v0.1.0.² The core multiple sequence alignment was used to infer the phylogenomic tree using FastTree v2.1.10 (Price et al., 2010).

The phosphate solubilizing capacity of the bacteria was evaluated using the previously described protocol (Behera et al., 2017). The bacterial strains were cultured overnight in the National Botanical Research Institute’s phosphate growth medium (NBRIP) at 37°C with continuous shaking at 200 rpm to assess alkaline phosphatase and acid phosphatase activities.

P. agglomerans PS1 and PS2 were screened for nitrate reductase activity (Kim and Seo, 2018), Indole-3-acetic acid (IAA) production (Ehmann, 1977), ammonia production (Bhattacharyya et al., 2020), and siderophore biosynthesis (Himpsl and Mobley, 2019) for the assessment of their bio-fertilization potential.

Drought stress tolerance of PS1 and PS2 was assessed as described previously (Elizabeth Mustamu et al., 2023). PS1 and PS2 were cultured at 37°C for 24 h with continuous shaking at 200 rpm in the nutrient broth (NB). A 100ul of microbial culture was inoculated in LB broth (2x) supplemented with a different concentration of Polyethylene glycol (PEG) (0, 5, 10, 20, 30, and 40% (w/v)). The total reaction volume was adjusted to 5 mL by adding sterile double distilled water. Absorbance was read at 600 nm after an incubation of 24 h at 37°C with continuous shaking at 200 rpm. ACC deaminase production activity of microbial isolates was checked with a standardized methodology (Maheshwari et al., 2020) to evaluate their salt-induced oxidative stress tolerance. In brief, microbial cells were initially induced in a 5 mL minimal media (Dworkin and Foster, 1958) followed by harvesting the cell by centrifugation at 16,000 x g for 5 min. Cells were washed with 0.1 M Tris–HCl (pH 7.6) and resuspended in 600 μL of 0.1 M Tris–HCl (pH 8.5). 30 μL of toluene was added to disrupt the cells, followed by vortexing for 30 s. 200 μL of the toluenized cell suspensions were mixed with 20 μL of 0.5 M ACC and incubated at 30°C for 15 min. One mL of 0.56 N HCl was added after the incubation. The mixture was vortexed to remove cell debris by centrifugation at 16,000 rpm for 5 min. One mL of the culture supernatant was combined with 800 μL of HCl (0.56 N) and freshly prepared 300 μL of 2,4-dinitrophenyl hydrazine (DNPH) reagent (0.1 g DNPH in 100 mL of 2 N HCl). The reaction mixture was incubated at 30°C for 30 min followed by the addition of 2 mL of NaOH (2 N). The absorbance was measured at 540 nm. Enzyme activity was calculated using the α-Ketoglutarate standard curve (R = 0.9998).

The wheat seed germination assays were performed in the presence of P. agglomerans PS1 and PS2. Seed pretreatment with bacterial agents to assess their protective effect was performed as described previously (Sharma et al., 2024). Seeds were soaked in overnight-grown bacterial cultures (density of 10¹¹ cells/ml) supplemented with varying concentrations of NaCl ranging from 0 to 1.0 M and incubated at 37°C for 16 h. The control seeds were soaked directly at different concentrations of NaCl ranging from 0 to 1.0 M for 16 h at 37°C. Subsequently, the seeds were wrapped in germination sheets and placed in 50 mL culture tubes containing 5 mL Hoagland solution to calculate seed germination percentage (Sharma et al., 2024). Tubes were incubated for 7 days in the dark at room temperature. Seed germination percentage, alpha-amylase activity, root length, and shoot lengths were measured after incubation (Singh and Kayastha, 2014).

Wheat cultivar 306 seeds treated with P. agglomerans PS1 and PS2 were cultivated in an experimental field in the botanical garden at Maharshi Dayanand University, Rohtak (28° 52′ 44” NL and 76° 37′ 19″ EL), Haryana, India. Wheat Growth and development stages are categorized into various Feeks stages.³ Wheat roots were harvested at different feeks stages (1.0, 2.0, 3.0, 6.0, 9.0, and 10.5) to assess the impact of P. agglomerans PS1 and PS2 across plant growth stages. (Here: Feeks 1.0 is the seedling emergence stage, Feeks 3.0 is the plant tillering stage, Feeks 6.0 is the first node appearance stage, Feeks 9.0 is flag leaf is the visible stage, while Feeks 10.5 is the wheat growth stage when heading complete). Metagenomic DNA of wheat rhizosphere was extracted using the alkaline lysis methodology (Kumar et al., 2016). Metagenomic DNA was purified using a soil DNA purification kit (Himedia, HiPurA). The 16S rRNA gene sequences were amplified from wheat rhizosphere metagenomic DNA using universal primers (27F 5’-AGAGTTTGATCCTGGCTCAG-3′; 1492R 5’-GGTTACCT TGTTACGACTT-3′) (James, 2010). Sequencing of 16S rRNA gene amplicons was performed on nanopore minion MK1C following midnight protocol. Wheat rhizospheric 16S rRNA sequence data was analyzed using Commander 2.0⁴ following the 16S rRNA sequence analysis pipeline.

Seeds were soaked with 5 mL of overnight-grown bacterial cultures (density of 10¹¹ cells/ml). Preincoucalted seeds were planted in the experimental soil bed and allowed to grow till harvest stages. Wheat cultivar WC-306 plant roots treated with P. agglomerans PS1 and PS2 were harvested at different feeks and assessed for phosphate solubilizing ability (Behera et al., 2017), nitrate reductase activity (Kim and Seo, 2018), total sugar content (Ludwig and Goldberg, 1956) and reducing sugar content (Khatri and Chhetri, 2020) in reference to non-treated WC-306 plant roots. Additionally, the number of tillers formed, the number of leaves per plant, the number of spikes per plant, spike length, the number of spikelets per spike, and grain yield were also estimated in both treated and non-treated WC-306 plants to assess the impact of Pantoea agglomerans strains on plant growth and yield.

All experiments were carried out in replicates. SIGMA plot 15 was employed to perform statistical evaluations and for graphical representations of the datasets. ANOVA analysis was performed to calculate significance among microbial treated and non-treated groups using Systat Software SigmaPlot 15.

The wheat rhizospheric soil had a pH of 7.3 ± 0.15275, a temperature of 22.6°C ± 0.50332, and a moisture content of 11.5% ± 1.10014 (w/w). Twelve morphologically distinct bacteria were isolated from the wheat rhizosphere. Plate screening assay for phosphate solubilization property showed that only two isolates, PS1 and PS2 led to the development of halo-zones of 23 ± 0.327 mm and 25 ± 0.47735 mm, respectively.

BLASTn analysis of the 16S rRNA gene sequences of PS1 (1,446 bp) and PS2 (1,429 bp) exhibited 98.62 and 98.46% similarity, with P. agglomerans NCTC 9381 and P. agglomerans ATCC 27155 in rRNA/ITS databases (NCBI). Phylogenetic analysis with 16S rRNA gene sequences of PS1 and PS2 with rRNA/ITS databases homologs further supported these findings (Figure 1). The 16S rRNA gene sequences of PS1 and PS2 showed a similarity of 99.48%, indicating their subspecies-level distinctiveness. Bacterial isolates PS1 and PS2 were designated as P. agglomerans PS1 and P. agglomerans PS2, respectively, for subsequent analyses. Microscopic investigation indicated P. agglomerans PS1 to be gram-negative, rod-shaped, and motile, while P. agglomerans PS2 was observed as gram-negative, rod-shaped, and non-motile. P. agglomerans PS1 showed optimum growth at 37°C (Figure 2A) within the pH range of 6.0–7.0 (Figure 2B). P. agglomerans PS2 showed optimum growth at pH 7.0 and 37°C (Figures 2A,B). Growth pattern analysis indicated that P. agglomerans PS1 and PS2 attain a log phase after 12.5 and 12 h, respectively. The doubling time of P. agglomerans PS1 (Supplementary Figure S1A) and PS2 (Supplementary Figure S1B) was calculated as 32.42 and 54.31 min, respectively. P. agglomerans PS1 showed 0.396 OD at 600 nm when grown anaerobically for 24 h at 37°C, indicating its facultative anaerobic nature. P. agglomerans PS2 did not show any growth under identical conditions and thus inferred as strict aerobe.

P. agglomerans PS1 was positive for amylase, esterase, lipase, protease, and catalase activity, while P. agglomerans PS2 was positive for amylase, esterase, protease, and catalase activity. Substrate utilization assay of P. agglomerans PS1 and PS2 indicates their substrate utilization profile is similar to other P. agglomerans species (Supplementary Table S1). Antibiotic susceptibility assay indicated that P. agglomerans PS1 was resistant toward amikacin, novobiocin, cefotaxime, lincomycin antibiotics while showing sensitivity towards amoxicillin, cephalothin, erythromycin, oxytetracyclin, vancomycin, ceflnaxone, ceflazidime, netillin, ofloxacin and bacitracin (Supplementary Table S2). Similarly, P. agglomerans PS2 was found resistant towards amikacin, ceflnaxone, vancomycin, cephalothin while showing sensitivity towards novobiocin, cefotaxime, lincomycin, amoxicillin, erythromycin, oxytetracyclin, ceflazidime, netillin, ofloxacin, and bacitracin (Supplementary Table S2). The antibiotic resistance profile of P. agglomerans PS1 and PS2 was similar to other Pantoea agglomerans species (Supplementary Table S2). Similar biochemical, substrate utilization, and antibiotic resistance profiles of P. agglomerans PS1 and PS2 to other Pantoea species strengthen the 16S rRNA gene-based taxonomic observations. Stress response physiology indicated that P. agglomerans PS1 and PS2 can successfully grow in the presence of salts (Figures 3A,B), arsenic (Figures 3C,D), and oxidizing agents (Figure 3E), respectively, as observed for other Pantoea sp. (Supplementary Table S3).

Genome sequencing of P. agglomerans PS1 and PS2 resulted in 1,020,610 and 683,599 paired-end raw reads, respectively. P. agglomerans PS1 and PS2 reads were assembled into 96 and 406 contigs accounting for 4,987,053 bps and 5,177,646 bps, respectively (Supplementary Table S4). Functional annotation of the P. agglomerans PS1 genome identified 4,860 protein-coding sequences, 09 rRNA genes, 71 tRNA genes, and 01 tmRNA gene. P. agglomerans PS2 genome encoded 5,275 protein-coding sequences, 11 rRNA genes, and 74 tRNA genes. The average ANI among different species of Pantoea ranged from 74–99%, indicating significant interspecific genomic variations. Furthermore, the ANI scores of P. agglomerans PS1 and P. agglomerans PS2 with P. agglomerans strain AR1a and P. agglomerans CFSAN047153 were 99.59 and 99.83, respectively, which were comparatively higher than those with other members of the species (Supplementary Table S5). The affiliation of P. agglomerans PS1 and PS2 as a member of P. agglomerans species was further reconfirmed using terra correlation. P. sp. CFSAN033090 had been awarded 0.99951 z-score against both P. agglomerans PS1 and PS2 during terra-correlation, confirming their similarity with P. agglomerans. Other Pantoea species exhibited good similarity (z-score ∼0.95–0.99) (Supplementary Table S6).

After the ANIb and tetra confirmation, the P. agglomerans PS1 and PS2 genomes were compared with genomes of Pantoea species to analyze genome-wide similarities and distinctiveness. The matrix generated using the Roary tool showed the comprehensive nature of the genome in which the P. agglomerans PS1 and PS2 showed the highest similarity with Pantoea sp. CFSAN033090 (Figure 4). It also revealed that all Pantoea genomes share only a few numbers of genes as their core genome. Shell and cloud genome collectively forms the central part of the genomes. P. agglomerans PS1 and PS2 genome has neither a pathogenic gene/island nor any virulence-related genes, indicating their non-pathogenic behavior. The genomic surveillance of both microbes revealed the presence of genes for phosphate solubilization and transport (Table 1). In addition to the genes for phosphate solubilization, P. agglomerans PS1 and PS2 genomes harbor genes encoding proteins for plant growth promotion activities like auxin biosynthesis, nitrogen assimilation, and siderophore biosynthesis (Table 2). P. agglomerans PS1 and PS2 genomes also harbor genes responsible for arsenic resistance, oxidative stress tolerance, metal stress tolerance, and salt tolerance (Supplementary Table S7) explaining its stress response physiology. The thirty-eight CAZymes clusters in the P. agglomerans PS1 genome (Supplementary Table S8) and twenty-nine CAZymes clusters within the P. agglomerans PS2 (Supplementary Table S9) genome indicate their diverse carbohydrate utilization profile. Several proteins were identified to be essential for effective colonization in plant rhizosphere (Kumar et al., 2023). An in-depth analysis of the P. agglomerans PS1 and PS2 genomes identifies the presence of genes encoding proteins for the synthesis of Type 1 & IV pili, exopolysaccharide (Table 3) essential for plant surface adhesion, auto-aggregation, and early biofilm formation (Kumar et al., 2023).

P. agglomerans PS1 and PS2 were observed to grow on Pikovskaya agar plates, which was further validated through quantitative phosphatase assay. P. agglomerans PS1 exhibited extracellular alkaline phosphatase (420.51 IU) and acid phosphatase activity (57 IU), while P. agglomerans PS2 displayed extracellular alkaline phosphatase (476.089 IU) and acid phosphatase activity (63 IU). Similarly, P. agglomerans PS1 and PS2 showed intracellular alkaline phosphatase (22 IU and 15 IU) and acid phosphatase activity (10.1 IU and 6.5 IU), respectively.

P. agglomerans PS1 and PS2 exhibited nitrate reductase activity (486 IU and 526 IU) and produced IAA (0.671 IU and 0.725 IU), respectively. The qualitative assessment of ammonia-producing activity indicated that both strains were involved in ammonia production. Siderophore biosynthesis assay indicated the appearance of an orange color and zones 22 ± 0.058 mm and 17 ± 0.025 mm for P. agglomerans PS1 and PS2, respectively. These results indicated the siderophore biosynthesis property of P. agglomerans PS1 and PS2. Furthermore, they were capable of producing and secreting plant growth-promoting hormones into the surrounding environment. The presence of genes responsible for plant growth promotion and their demonstrated bioactivity suggested that P. agglomerans PS1 and PS2 possess nutrient assimilation properties to boost plant growth.

The microbial isolates PS1 and PS2 exhibited robust growth up to 40% of polyethylene glycol (PEG) concentration. Interestingly, PS2 showed greater activity compared to PS1. These findings highlight the plant growth-promoting features under drought stress (Figure 5). Microbial isolate PS2 (0.532 EU) was found to produce more ACC deaminase enzyme as compared to PS1 (0.436 EU). ACC deaminase activity in both strains confirms their oxidative stress-mitigating properties.

A seed germination rate of 70.66% ± 0.57735 was observed in the control group. However, seeds pre-treated with P. agglomerans PS1 and PS2 exhibited germination efficiencies of 95.6 ± 0.57735 and 94% ±0.57705%, respectively (Figure 6A). P. agglomerans PS1 and PS2 were found to increase seed germination by ~135-and ~ 133-fold, respectively. Furthermore, pre-treatment with P. agglomerans PS1 and PS2 not only enhanced seed germination but also significantly improved it compared to the control (p < 0.001). Pre-treatment with P. agglomerans PS1 and PS2 also significantly increased alpha-amylase activity (0.956 IU (p = 0.0001) and 0.94 IU (p = 0.0001)), respectively, compared to the control (0.253 IU). The significant increase in alpha-amylase activity in wheat seeds after pre-treatment with P. agglomerans PS1 and PS2 could be a possible factor for enhanced seed germination. Seed germination significantly decreased (p = 0.020) with increasing salt concentration. P. agglomerans PS1 and PS2 pretreatment also exhibited enhanced seed germination under high salinity conditions (p = 0.004) (Figure 6B). PS2 not only improved seed germination but also promoted the growth of wheat plantlets. Wheat seeds pre-treated with P. agglomerans PS1 and PS2 demonstrated significantly increased shoot length (p = 0.001) and root length (p = 0.0018) compared to untreated seeds under high salinity conditions.

The 16S rRNA gene analysis of wheat rhizosphere at different growth stages revealed variations in the microbiota with wheat growth stages (Figures 7A–F; Supplementary Table S10). At Feeks 1.0 (emergence stage), Alphaproteobacteria (Figure 8A), Betaproteobacteria (Figure 8B), and Gammaproteobacteria (Figure 8C) were abundant microbial groups in treated and non-treated wheat seedlings. However, the abundance of these microbial groups in treated seedlings was significantly (p < 0.001), different from non-treated ones (Figure 7). Among all bacterial groups, Pantoea sp. were significantly abundant (p < 0.001) in treated seeds with PS1 and PS2 than the non-treated ones. At Feeks 2.0 (beginning of tillering), rhizosphere microbiota primarily comprised Alphaproteobacteria (12.48% in PS1, 12.12% in PS2 and 6.47% in not-treated plants) (Figure 8A) Betaproteobacteria (Figure 8B) (34.35% in PS1, 32.42% in PS2 and 26% in not-treated plants) and Gammaproteobacteria (Figure 8C) (20.04% in PS1, 22% in PS2 and 9.45% in not-treated plants). The 16S rRNA gene sequences taxonomically affiliated to Pantoea sp. were observed in both treated as well as non-treated groups. However, 16S rRNA gene sequences taxonomically affiliated to Pantoea sp. were significantly higher (p < 0.001) in treated (0.38% in PS1 and 0.55% in PS2) than in the non-treated ones (0.15%).

At Feeks 3.0, Actinobacteria (46.15% in non-treated and 63.92 and 72.92% in PS1 and PS2 treated, respectively) (Figure 8D) was observed as a highly abundant microbial group, followed by Alphaproteobacteria (Figure 8A) (12.15% in non-treated, 15.36% in PS1 treated, and 18.31% PS2 treated), Betaproteobacteria (Figure 8B) (31.09% in non-treated, 38.52% in PS1 treated and 31.52% in PS2 treated plants), Deltaproteobacteria (Figure 8E) (15.16% in non-treated, 19% in PS1 treated, and 21.64% in PS2 treated), and Gammaproteobacteria (Figure 8C) (11.96% in non-treated, 26.4% in PS1 treated, and 26.55% in PS2 treated). Despite similar microbial abundance profiles, PS1 and PS2 showed significantly different (p < 0.01) rhizosphere microbiota compared to the untreated ones. Likewise, microbiota profiling at Feeks 2, the 16S rRNA gene sequences affiliated to Pantoea species were significantly abundant (p < 0.001) in PS1 (2.14%) and PS2 (2.47%) in comparison to untreated plants (0.16%). Additionally, Flavobacteria (1.95 and 4.15%), Clostridia (2.98, 4.33%), Chitinophagia (4.7 and 4.31%), and Bacilli (3 and 4.7%) were exclusively associated with plants treated with Pantoea agglomerans PS1 and PS2, respectively.

At Feeks 6.0 (internode formation), Actinobacteria (12.84% in non-treated and 51.8, 57.06% in PS1 and PS2 treated) (Figure 8D) was the most abundant bacterial group, followed by, Alphaproteobacteria (16.03% in non-treated, 17.16% in PS1 treated, and 22.9% in PS2 treated) (Figure 8A), Betaproteobacteria (11.24% in non-treated, 22.72% in PS1 treated and 22.42% in PS2 treated) (Figure 8B), Deltaproteobacteria (0.003, 4.18 and 6.52% in non-treated, PS1, and PS2 treated respectively) (Figure 8E), and Gammaproteobacteria (1.43% in non-treated, 15.08% in PS1 treated and 19.74% in PS2 treated) (Figure 8C). Among Gammaproteobacteria, the 16S rRNA gene sequences affiliated with Pantoea were present in plants inoculated with only PS1 (0.74%) and PS2 (1.92%). Moreover, Chitinophagia, Deltaproteobacteria, and Flavobacteria were specifically associated with plants treated with Pantoea agglomerans PS1 and PS2.

At Feeks 9.0 (ligule of flag leaf visible), rhizosphere microbiota is represented by Alphaproteobacteria (7.34% in non-treated, 14.22% in PS1 treated and 15.07% in PS2 treated) (Figure 8A), Betaproteobacteria (7.35% in control, 15.66% in PS1 treated and 19.83% in PS2 treated) (Figure 8B), Deltaproteobacteria (4.44% in PS1 treated and 5.73% in PS2 treated) (Figure 8E), and Gammaproteobacteria (2.588% in control, 8.60% in PS1 treated and 11.91% in PS2 treated) (Figure 8C). Despite similar microbial abundance profiles, PS1 and PS2 showed significantly different (p < 0.01) rhizosphere microbiota compared to the untreated ones. Additionally, 16S rRNA gene sequences affiliated with Pantoea sp. were lower in PS1 (0.62%) and PS2 (1.51%) treated plants.

Finally, at Feeks 10.5 (heading and flowering), rhizosphere microbiota is represented by Alphaproteobacteria (6.37% in control, 13.11% in PS1 treated, 19.20% in PS2 treated) (Figure 8A), Betaproteobacteria (4.56% in control, 12.62% in PS1 treated and 9.94% in PS2 treated) (Figure 8B), Deltaproteobacteria (3.44% in PS1 treated and 6.13% in PS2 treated) (Figure 8E), and Gammaproteobacteria (0.36% in control, 11.77% in PS1 treated and 9.90% in PS2 treated) were present (Figure 8C). The 16S rRNA gene sequences affiliated with Pantoea were still present in the case of plants treated with PS1 (0.54%) and PS2 (1.4%) only.

Soil microbiota composition was also assessed to unveil the influence of soil microbiota in the development of wheat rhizosphere microbiota during various wheat growth stages. Proteobacteria was the dominant microbial group observed across all the soil samples collected parallelly with wheat rhizosphere samples during different growth stages (Feeks 1.0, 2.0, 3.0, 6.0, 9.0, and 10.5). However, variability was observed in proteobacterial subgroups. Gammaproteobacteria were the most abundant (33.16%) in the soil sample collected during Feeks 1.0, while the soil sample collected at Feeks 2.0 was marked by a high presence of Betaproteobacteria (17.48%). During Feeks 3.0, Feeks 6.0, Feeks 9.0, and Feeks 10.5, Alphaproteobacteria were the dominant class in all soil samples with relative abundances of 18.55, 18.32, 22.97, and 21.5%, respectively.

Plant growth potential P. agglomerans PS1 and PS2 need to be also validated in field conditions to ensure their efficiency and suitability for agricultural application. Pre-treatment of seeds with P. agglomerans PS1 and PS2 showed a significant increase in the total sugar (13583.21 μg/g and 14490.5 μg/g) and reducing sugar (385 μg/g and 362.97 μg/g), respectively, at feeks 3.0 (Supplementary Figure S2). A respective increase in total sugar (10850.3 μg/g and 10301.5 μg/g) and reducing sugar content (321.7 μg/g and 300.73 μg/g) was observed in seeds treated with P. agglomerans PS1 and PS2 about control at feeks 6.0 (Supplementary Figure S2). Wheat plants treated with P. agglomerans PS1 showed an increase in extracellular alkaline phosphatase (831.83 IU) at feeks 3.0 (Figure 9A) and acid phosphatase (225.54 IU) (Figure 9C) activity at feeks 6.0, respectively. Similarly, P. agglomerans PS2 pre-treatment results in increased extracellular alkaline (653.29 IU) (Figure 9B) and acid phosphatase activity (123.9 IU) at feeks 3.0 (Figure 9D). P. agglomerans PS1 and P. agglomerans PS2 pre-treatment have also increased the nitrate reductase activity (678 IU and 664.5 IU) at feeks 3.0, respectively, compared to control (Figures 9E,F). Wheat plants treated with Pantoea agglomerans PS1 and PS2 showed a significant increase in the number of tillers (p = 0.0018), number of leaves per plant (p = 0.0001), spike length (p = 0.0001), number of spikes per plant (p = 0.0001), number of spikelets per plant (p = 0.0001), grain weight per 1,000 grains (p = 0.0023) and grain yield (p = 0.0001) (Table 4).