Soil samples were gathered from rhizospheres of eight different trees (Abies nordmanniana, Corylus avellana, Pyrus elaeagnifolia, Acer pseudoplatanus, Robinia pseudoacacia, Juglans regia, Malus domestica, Ficus carica) at a soil depth of 2–5 ft within different locations ( Table 1 , Figures 1 , 2 ) of Uzungöl forest, Trabzon, Turkey. The soil bound to the root surface was extracted and identified as a potential habitat of PGPR bacteria. The soil samples were kept in polyethylene bags and stored at −20°C for later analysis.

To isolate PGP bacteria, 1 g of the soil sample was dissolved in sterilized distilled water, followed by creating serial dilutions (10⁻¹–10^–5) of the soil solutions. From each dilution, 30 μL was spread onto nutrient agar (NA) medium plates, which were then incubated at 30°C for 2 days. After incubation, diverse bacterial colonies differing in shape, color, and size appeared on NA plates. Single bacterial colonies with distinct morphological characteristics were then selected on the basis of colony form, elevation, margin, and color and were transferred to fresh NA plates through restreaking. Individual colonies were then preserved on NA slants at 4°C and stored in 30% glycerol stocks at −80°C for later studies ( Figure 1 ).

To isolate endophytic bacteria, 20 g of rhizosphere soil from each sample was taken in separate pots and mixed with sterilized vermiculite ( Figure 1 ). Seeds of the Kocamaninci rice variety were sterilized by washing with 70% ethanol for 30 s, followed by 3 min of dipping in 3% sodium hypochlorite solution, and then washed five times with sterilized distilled water. Sterilized rice seeds were sown in trapezoidal pots (6.5 cm top width × 4.5 cm bottom width × 6.5 cm height) containing soil inoculum from a specific rhizosphere. These pots were placed in a growth chamber under controlled conditions (16-h light/8-h dark cycle and day/night temperatures of 25°C/18°C). After 21 days, plants were harvested, rinsed in tap water to eliminate vermiculite, and then washed five times with sterile distilled water within a laminar flow cabinet. Leaves, stems, and roots were segmented into 1–2 cm pieces and crushed in a sterile mortar and pestle. The resulting turbid solution was serially diluted (10⁻¹–10⁻⁵) using sterile distilled water. A loop from each diluted sample was then transferred into a screw-capped tube containing 4 mL of nitrogen-free bromothymol blue (NFB) semisolid media (Habibi et al., 2014; Baldani et al., 2014). Tubes were incubated at 30°C for 1 week, during which bacterial growth appeared as veil-like pellicles. Subsequently, 100 μL from each pellicle layer was further diluted (10⁻¹–10⁻⁵), and 30 μL from each dilution was spread onto NFB agar plates. These plates were then incubated at 30°C for 2 days in an anaerobic jar, after which distinct colonies were formed. Single colonies were selected and transferred to fresh NFB agar plates by streaking. The purified cultures were preserved in nutrient agar slants at 4°C and stored in 30% glycerol stocks at −80°C for later studies.

To observe the morphological features of isolates, bacterial colonies were cultured on NA agar plates. A loop from each colony was diluted with 1 mL of sterilized distilled water in a microtube and mixed well. Subsequently, 5 μL of this mixture was spread onto an NA agar plate and incubated at 30°C for 3 days. The morphological characteristics of bacteria, such as colony shape, elevation, edge, and color, were observed under a stereo microscope ( Figure 1 ).

For the detection and quantification of indole-3-acetic acid (IAA), each bacterial strain was cultured in a test tube containing 3 mL of King’s B medium (comprising 17.2 mL of 87% glycerol, 1.15 g of K₂HPO₄, 20 g of peptone, 0.5 g of L-tryptophan, and 1.5 g of MgSO₄.7H₂O per liter). The cultures were incubated for 3 days at 30°C, and then 1 mL from each culture was transferred into a 1.5-mL microcentrifuge tube. The samples were then centrifuged at 5,400×g for 10 min, and 500 μL of the supernatant was carefully pipetted into a new microcentrifuge tube. Subsequently, 500 μL of Salkowski reagent was added to each tube and incubated in the dark for 30 min. Then, 300 μL from each tube was transferred to a 96-well plate, and absorbance was measured at 550 nm using a microplate reader ( Figure 3A ). The IAA concentration in the supernatant was determined using an equation developed from a standard curve calibrated against known concentrations of IAA (Abdelwahed et al., 2023).

The ability of bacterial strains to produce siderophores was estimated using the universal CAS assay with slight modifications (Arora and Verma, 2017). At first, to eliminate iron contamination, all glassware was rinsed with 3 mol/L of hydrochloric acid (HCl) and then washed with deionized water. The CAS reagent was prepared as follows: 121 mg of CAS was dissolved in 100 mL of distilled water parallel to this, and a 20-mL solution of 1 mM ferric chloride (FeCl₃.6H₂O) was prepared in 10 mM of HCl. These two solutions were mixed, and the mixture was then added to 20 mL of a hexadecyl trimethyl ammonium bromide (HDTMA) solution under continuous stirring. The HDTMA solution was prepared by dissolving 729 mg of HDTMA in 400 mL of distilled water. The resulting CAS-HDTMA solution was sterilized prior to use.

For estimating siderophore production, the procedure was carried out using a microtiter plate. Each bacterial isolate was cultured in a test tube containing 3 mL of King’s B medium for 3 days at 30°C. Later, 1 mL from each tube was transferred to a 1.5-mL microcentrifuge tube and centrifuged at 5,400×g for 10 min. Then, 100 μL of the supernatant from each bacterial culture was pipetted into separate wells of the microtiter plate, followed by the addition of 100 μL of CAS reagent to each well. The plates were incubated in the dark for 20 min. The optical density of each sample was measured at 630 nm using a microplate reader ( Figure 3B ). Siderophore production was quantified in percent siderophore units (psu), calculated according to the formula (Payne, 1993).

where Ar is the absorbance of the reference (CAS solution and uninoculated media), and As is the absorbance of the sample (CAS solution and cell-free supernatant of the sample).

Bacterial isolates were cultured in the nutrient broth medium for 48 h at 28°C. Subsequently, 5 μL from each culture was dropped into Pikovskaya’s agar medium supplemented with tricalcium phosphate (Pikovskaya, 1948). These plates were then incubated for 7 days at 28°C. The formation of a clear zone around each bacterial colony indicated the ability of the isolate to solubilize phosphate ( Figure 3C ). In order to estimate the phosphate-solubilizing activity, the size of the clear zone, or halozone, was measured using the formula:

The HCN production of bacteria was evaluated using the Bakker and Schippers method. Each isolate was streaked on a nutrient agar plate supplemented with 4.4% glycine. A Whatman filter paper No. 1 wetted with 2% sodium carbonate in 0.5% picric acid solution was attached to the underside of the plate lids. The plates were then sealed with parafilm and incubated for 5 days at 28°C ± 2°C. A color change in the filter paper from yellow to slightly brown, medium brown (++), or strong brown (+++) served as an indicator of hydrocyanic acid production (Bakker and Schippers, 1987).

The ability of bacterial isolates to produce ammonia was evaluated using the method described by Di-Benedetto. Briefly, bacterial cultures were inoculated into test tubes filled with 10 mL of peptone water broth and incubated at 30°C for 3 days. Following the incubation, 0.5 mL of Nessler’s reagent was added to each tube. The emergence of a brown color from yellow indicated the production of ammonia (Mir et al., 2022).

Protease activity was assessed by streaking bacterial cultures onto sterile skim milk agar plates. These plates were then incubated for 48–72 h at 30°C. The presence of a halo zone around the bacterial colonies indicated protease production ( Figure 4A ). Protease activity was marked by the formation of a halo zone around bacterial growth. The size of the halo zone indicated the extent of activity (Vermelho et al., 1996).

For the estimation of amylase production, bacterial cultures were streaked onto starch agar medium and incubated at 28°C ± 2°C for 48–72 h. After incubation, the plates were exposed to iodine solution for a minute. The formation of colorless zones around the colonies ( Figure 4B ) indicated amylase activity (Gupta et al., 2003).

Cellulase production was determined using the method of Hendricks. Briefly, 5 μL of actively grown bacterial cultures were spotted onto carboxymethylcellulose Congo red media, containing 0.5 g of carboxymethylcellulose, 0.099 g of K₂HPO₄, 0.049 g of MgSO₄.7H₂O, 0.05 g of yeast extract, 0.05 g of Congo red, and 20 g of agar per liter and adjusted to pH 7.2. The plates were then incubated for 72 h at 28°C ± 2°C, the plates were then washed with saline water to highlight the clear zones. The formation of a halo zone around bacterial growth ( Figure 4C ) indicated positive cellulase activity (Hendricks et al., 1995).

Bacterial isolates were screened for their salt tolerance using nutrient broth (NB) media supplemented with different NaCl (w/v) concentrations, such as 3% (0.513 M), 5% (0.855 M), and 7% (1.196 M). The tubes containing media were inoculated with a loop full of fresh bacterial culture and incubated at 30°C on a shaker (120 rpm). The controls were maintained at 0% (0 M) NaCl (w/v) in NB medium, with and without bacterial inoculation. Bacterial growth in NaCl-supplemented broths was observed and compared with the controls after time intervals of 24 h, 48 h, and 72 h.

Out of 129 isolates, 41 with significant plant growth-promoting traits were tested in a germination trial to evaluate their effectiveness as seed biopriming agents. Briefly, rice seeds (Oryza sativa L.) were surface-sterilized as previously described and then immersed for 30 min in a 2-day-old active bacterial culture having a cell density of 10⁹ colony-forming units/mL (CFU/mL), grown at 28°C in King’s B broth medium. For the control, seeds were soaked in non-inoculated media. From each treatment, seeds were then transferred onto 1% agar plate, and three replications of each treatment were maintained. The plates were initially kept in the dark for 7 days, followed by 3 days under controlled conditions (16-h light/8-h dark cycle at 25°C/18°C day/night temperatures) in a growth chamber. Germination data were recorded at 24-h intervals. At the end of the experiment, seedlings were removed carefully, and measurements were taken for emerging root length (ERL), emerging shoot length (ESL), emerging root mass (ERM), and emerging shoot mass (ESM). Additional parameters like germination energy (GE), seedling vigor index (SVI) 1, and seedling vigor index (SVI) 2 were calculated with the help of data obtained from the germination test using the following equations:

Similar to the germination trial, 41 bacterial isolates were also evaluated for their effectiveness as root inoculants to promote the growth of rice seedlings in hydroponic conditions. Briefly, sterilized rice seeds were placed in petri dishes containing wetted filter papers and allowed to germinate. After 7 days, young seedlings with uniform growth were taken and their roots were inoculated for 1 h in a 2-day-old active bacterial culture having a cell density of 10⁹ CFU/mL, grown at 28°C in King’s B broth medium. Seedlings treated with non-inoculated media were used as the baseline control. The seedlings were then transferred onto sterilized plastic boxes (dimensions 17 cm × 9 cm × 4.5 cm) designed for hydroponic culture, and three replications were maintained for each treatment in a completely randomized design (CRD) and were kept in a growth chamber controlled at a 16-h light/8-h dark cycle, 25°C/18°C day/night temperatures. Sterilized Hoagland solution was provided as a nutrient supplement. Plants were harvested after 3 weeks, and the roots were washed thoroughly in tap water. The shoot, root length, and fresh weight of the plants were recorded. Plant samples were dried at 60°C in an oven, and then the dry root and shoot masses were recorded.

The roots of 1-week-old seedlings were rinsed with distilled water for 30 s in a laminar flow cabinet. Specimens were prepared for the examination under a scanning electron microscope (SEM) as follows. Immediately after rinsing, seminal roots were cut into several segments of 3–5 mm lengths and were then fixed with 2.5% glutaraldehyde dissolved in 0.2 M of sodium phosphate buffer (pH 7.2) for 12 h at 4°C. The segments were subsequently dehydrated using a series of ethanol (30%, 50%, 75%, and 96%) solutions, each for 15 min. Following dehydration, the samples were dried using a Hitachi Critical Point Dryer HCP-I with liquid carbon dioxide and were then coated with gold using a Giko IB-3 Ion Coater. The prepared samples were examined under a Hitachi SEM (SSM-2A) at an accelerating voltage of 10 kV (Asanuma et al., 1979).

In order to evaluate the effectiveness of bacterial treatments in comparison with the control, parameter scoring was performed. The corresponding equations were used to calculate the actual scores of individual parameters for each treatment ( Table 2 ). Scores for every single trait were considered to be zero in the control, and maximum scores were given to the highest value within each trait.

Of the 41 selected isolates, 14 bacterial isolates were found to be effective for both biopriming and root inoculation of rice along with their diverse PGP traits. These isolates were identified through 16S rDNA sequencing. Briefly, universal bacterial primers were used in colony PCR to amplify the 16S rDNA sequence. The forward primer F (5′-AGAGTTTGATCMTGGCTCAG-3′) and the reverse primer R (5′-TACGGYTACCTTGTTACGACTT-3′) positioned at 8–27 and 1492–1507, respectively, on the 16S rDNA sequence of Escherichia coli were designed. The reaction mixture for PCR had a total volume of 50 μL, including 1 μL of DNA template, 5 μL of 10× reaction buffer, 2.5 μL each of 10 μM F and R primers, 0.5 μL of Dream Taq DNA Polymerase (Thermo Fisher Scientific, Foster City, California, US), and 4 μL of dNTP mixture. The PCR program was as follows: 94°C initial denaturation for 2 min, 94°C denaturation for 30 s, 55°C annealing for 1 min 15 s, 72°C extension for 1 min 30 s, 34 cycles, and 72°C final extension for 5 min. The PCR fragments were analyzed using gel electrophoresis with 1% (w/v) agarose gel. Bands corresponding to the 16S rRNA gene were confirmed, and the PCR fragments were sent to Macrogen Europe for sequencing. The 16S rDNA sequences were obtained and submitted to the GenBank database using the BLAST software to get accession numbers. Sequences homologous to bacterial isolates were compared in the EzBioCloud database, and alignments were performed using the Clustal-W program. The neighbor-joining method was used to construct phylogenetic trees, and evolutionary analysis was performed using the MEGA X software ( Supplementary Figures 1S–4S ).

All the experiments were repeated in triplicate, and data were analyzed using GraphPad Prism 8, IBM SPSS Statistics 27, and Origin 2024. The comparison between treatment means and control was made using Dunnett’s test at P<0.05. Principal component analysis (PCA) and hierarchical cluster analysis were performed using the Origin 2024 software.

We examined bacterial isolates obtained from the rhizosphere soils of eight different trees to assess their colony morphology ( Table 3 ). The analysis of the colony form revealed that isolates from A. nordmanniana exhibited circular (84.2%) and filamentous (15.8%) colonies. All of the isolates from the rhizospheres of C. avellana and F. carica were circular in form. Regarding P. elaeagnifolia, isolates formed circular (79.3%) and irregular (20.7%) colonies. For R. pseudoacacia, isolates showed an almost similar trend with circular (78.6%) and irregular (21.4%) colony forms. Isolates from J. regia exhibited circular (50%) and irregular (50%) colonies. For A. pseudoplatanus sp., isolates were circular (70%) and irregular (30%) in form. Parallel to this, isolates from M. domestica formed circular (64.3%) and irregular (35.7%) colonies.

In terms of colony color, A. nordmanniana isolates displayed 31.6% whitish, 52.6% yellowish, and 15.8% creamy colonies. The isolates from C. avellana, P. elaeagnifolia, and R. pseudoacacia predominantly formed whitish (58.3%) and creamy (41.7%) colonies. Acer pseudoplatanus rhizospheric isolates produced whitish (70%), creamy (15%), and transparent (15%) colonies. Malus domestica isolates were 64.3% whitish and 35.7% creamy, whereas F. carica isolates had a more diverse color distribution, showing 28.6% each of whitish, yellow, and creamy colonies.

In terms of colony elevation, isolates from A. nordmanniana exhibited convex (68.4%), flat (15.8%), or raised (15.8%) elevations. All isolates from J. regia exhibited raised elevations, and 45% of A. pseudoplatanus isolates displayed raised or umbonate elevations, while 15% showed flat elevations. In contrast, 78.6% of R. pseudoacacia isolates exhibited umbonate elevations, with the remaining (21.4%) showing raised elevations. Corylus avellana isolates displayed 16.4% umbonate or crateriform, while the rest (58.3%) exhibited raised elevations.

Regarding colony margins, 68.4% of isolates from A. nordmanniana exhibited entire margins, while the remaining isolates had either filiform (15.8%) or curled (15.8%) margins. The majority of isolates from C. avellana (83.3%) and M. domestica (64.3%) also exhibited entire margins. Acer pseudoplatanus isolates showed entire (55%), undulate (15%), and curled margins (30%), while isolates from M. domestica had entire (78.6%) and curled (21.4%) margins. All the isolates from F. carica exhibited entire margins.

Overall, the variations in colony morphology such as form, color, elevation, and margins came from differences in genetic makeup, environmental adaptability, and interactions with the host plant.

We examined a total of 129 bacteria isolated from different rhizospheres for their IAA production. The findings are detailed in Tables 4 , 5 . The results indicated that a significant majority, 109 isolates, demonstrated the ability to synthesize IAA, highlighting the widespread potential of these bacteria to influence plant growth through hormonal interactions.

From the A. nordmanniana rhizosphere, isolate R1(B) stood out with the highest IAA production at 353.5 ± 5.7 µg/mL followed by S1(E) at 180 ± 14.5 µg/mL, S1(B) at 144.5 ± 6.8 µg/mL, SS1(1) at 136 ± 57.2 µg/mL, SS1(4) at 121.4 ± 12.4 µg/mL, and S1(G) at 107.9 ± 3.1 µg/mL. All 19 bacterial isolates derived from this rhizosphere were able to produce IAA. From the C. avellana rhizosphere, all 12 bacterial isolates produced IAA. Particularly, isolate SS2(4) recorded a high IAA production level of 353.5 ± 95 µg/mL followed by SS2(1)2 with 259.3 ± 32 µg/mL, L2(6)2 with 169.6 ± 93.6 µg/mL, R2(2)2 with 135.6 ± 119.6 µg/mL, and SS2(4)2 with 59.4 ± 1.2 µg/mL. For P. elaeagnifolia, 90% of the 29 isolates were capable of producing IAA. The highest production of 511.9 ± 133.8 µg/mL was observed in isolate SS3(8) followed by SS3(5)1 with 447 ± 11.7 µg/mL, SS3(6)1 with 403 ± 23 µg/mL, SS3(9) with 256 ± 28.5 µg/mL, S3(1) with 224.5 ± 23.2 µg/mL, S3(3) with 185.6 ± 7.7 µg/mL, SS3(5)2 with 130 ± 26.5 µg/mL, SS3(2)2 with 121.2 ± 50.5 µg/mL, SS3(2)1 with 104.7 ± 44.4 µg/mL, and SS3(4)1 with 84.2 ± 37.6 µg/mL, indicating a varying potential of bacteria to modulate plant growth through hormonal interaction. The A. pseudoplatanus rhizosphere showed that 95% of its 20 bacterial isolates were capable of synthesizing IAA. R4(1)2 exhibited the highest IAA production at 739.9 ± 251.5 µg/mL, followed by SS4(10) at 183.5 ± 10.5 µg/mL, SS4(8) at 170.3 ± 89.8 µg/mL, SS4(5) at 146.3 ± 72.1 µg/mL, S4(2) at 116.7 ± 14.2 µg/mL, L4(1) at 102.7 ± 11.1 µg/mL, and SS4(7) at 53 ± 4.4 µg/mL. From R. pseudoacacia, 71% of the 14 isolates produced IAA, with R5(1) showing the maximum production of 148.6 ± 12.2 µg/mL followed by S5(1) with 91 ± 64.7 µg/mL, SS5(5) with 73.8 ± 12.1 µg/mL, R5(1)1 with 39.4 ± 0.9 µg/mL, and S5(2)1 with 28.9 ± 2 µg/mL. Juglans regia and M. domestica both showed that 57% of their 14 isolates produced IAA. Notably, R6(3) from J. regia showed a peak production of 125.5 ± 4.2 µg/mL followed by SS6 (5) with 117.5 ± 2.8 µg/mL. Isolate R7(3) from M. domestica exhibited the highest production of 122.1 ± 35.2 µg/mL followed by R7(1) with 121.1 ± 3.1 µg/mL and SS7(5) with 85 ± 15 µg/mL. Ficus carica had 86% of its 7 isolates producing IAA, with isolate S8 producing the highest at 108.3 ± 8.6 µg/mL followed by SS8(6) with 84.5 ± 12.1 µg/mL and SS8(2) with 32.4 ± 4.6 µg/mL.

These results highlight the significant phytohormonal influence these isolates may have on their host plants. These bacterial isolates are potential growth stimulators to enhance the growth of plant through natural hormonal supplementation. The variability in IAA production of different isolates represented the diversity in the metabolic capabilities of rhizosphere-associated bacteria.

In examining siderophore production among bacterial isolates derived from various tree rhizospheres, only a small fraction of the 129 tested isolates demonstrated this capability, with just 16 isolates showing positive results. The results, which are detailed in Tables 4 , 5 , highlighted significant variability in siderophore production across different rhizospheres.

Among the isolates from C. avellana, 17% were capable of producing siderophores, with isolate L2 (6)2, showing notably high production at 62.2% ± 33.4% followed by R2(2)2 at 14.5% ± 4.3%. In contrast, only 10% of the bacterial isolates from the rhizosphere of P. elaeagnifolia were able to produce siderophores, with the peak production recorded at 22.2% ± 5.2% by isolate SS3(2)1 followed by SS3(9) at 12.9% ± 7.5%, and SS3(2)2 at 3.8% ± 1.7%. A slightly higher proportion of A. pseudoplatanus-associated isolates, 20%, produced siderophores, with the highest level observed in isolate L4 (1) at 26.5% ± 1.3%, SS4(10) at 15.8% ± 5%, SS4(8) at 9.3% ± 3.9%, and S4(2) at 3.7% ± 1.8%.

Isolates from the rhizosphere of R. pseudoacacia showed a relatively better capability, with 28% isolates producing siderophores. Among these isolates, R5(1) recorded the highest production rate at 91.1% ± 0.8% followed by SS5(5) at 38.7% ± 10.1%, S5(1) at 37.4% ± 1%, and S5(2)1 at 2.5% ± 0.7%. Both M. domestica and F. carica had 14% of their isolates capable of producing siderophores, with the highest outputs being 9.4% ± 1.1% in isolate R7(1) from M. domestica followed by R7(3) at 8.9% ± 1.8%. Isolate S8 showed peak production of siderophores at 50.1% ± 2.6% from the F. carica rhizosphere.

Interestingly, none of the isolates from A. nordmanniana and J. regia exhibited siderophore production. This distinct absence highlights the adaptive traits of microbial communities associated with a specific rhizosphere. The ability to produce siderophores helps in iron chelation which enhances plant growth by influencing the dynamics of nutrient uptake within these rhizospheric ecosystems.

The inorganic phosphate solubilization activity of 129 bacterial isolates was assessed using a plate assay. A total of nine isolates exhibited clear zones around bacterial growth, indicating phosphate solubilization ( Figure 3C , Tables 4 , 5 ). Approximately 8.3% of the isolates from the rhizosphere of C. avellana showed solubilization activity, with isolate SS2(4)2 forming a clear zone of 2.6 ± 0.3. In the rhizosphere of P. elaeagnifolia, 6.9% of the isolates demonstrated phosphate solubilization, with the highest activity recorded in isolate SS3(4)1 at 2.9 ± 0.3 followed by SS3(6)1 at 2.7 ± 0.2.

Among the isolates from the rhizosphere of A. pseudoplatanus sp., 5% exhibited phosphate solubilization, with the maximum zone size observed in isolate SS4(7) at 3.1 ± 0.3. From the rhizosphere of R. pseudoacacia, 21.4% (highest frequency) of the isolates were positive for phosphate solubilization, with isolate S5(2)1 showing the peak zone size of 3.8 ± 0.5 followed by S5(1) at 3.1 ± 0.5 and R5(1)1 at 2.8 ± 0.3.

From the rhizosphere soil of M. domestica, 7.1% of the isolates were capable of forming clear zones, with isolate R7(1) producing a zone of 1.6 ± 0.3. Among the isolates from the rhizosphere of F. carica, 14.3% exhibited phosphate solubilization, with isolate SS8(2) forming a zone size of 2.3 ± 0.1.

None of the isolates from the rhizospheres of A. nordmanniana and J. regia showed phosphate solubilization activity. These results highlight variability in phosphate solubilization among bacteria associated with a particular rhizosphere.

Identification of potential endogenous nitrogen-fixing diazotrophs in different parts (leaf, stem, and root) of rice was performed using semisolid NFB media. N-fixing diazotrophic bacteria fix atmospheric nitrogen which was confirmed through a color change in media from green to blue along with clear pellicle formation.

Briefly, of the 129 total isolates derived from the root zones of different trees, 50.4% were able to localize in different organs of rice plant (7.8% in the leaf, 17.1% in the stem, and 25.6% in the root) and were identified as potential endogenous nitrogen-fixing diazotrophs of rice Tables 4 , 5 ). Out of the 19 isolates derived from A. nordmanniana, 73.7% showed positive growth on NFB media and were able to reside inside various parts of rice plant (36.8% in the stem including S1(B), S1(E), and S1(G) and 36.8% in the root including R1(B)). Out of the 12 isolates from C. avellane, 66.7% showed their potential for nitrogen fixation inside rice plants (16.7% in the leaf including L2(6)2 and 50% in the root including R2(2)2).

Of the 29 isolates from the P. elaeagnifolia rhizosphere, 55.2% exhibited their potential for nitrogen fixation inside rice plants (20.7% in the leaf, 10.3% in the stem including S3(1) and S3(3), and 50% in the root). Out of the 29 isolates from the root zone of A. pseudoplatanus, 40% revealed their potential for nitrogen fixation inside rice plants (10% in the leaf including L4(1), 15% in the stem including S4 (2), and 15% in the root including R4(1)2). Of the 14 isolates from R. pseudoacacia, 50% showed their potential ability for N-fixation inside rice plants (28.6% in the stem including S5(1) and S5(2)1 and 21.4% in the root including R5(1)1).

Out of the 14 isolates from J. regia, 35.7% showed their potential for N-fixation inside rice plants [7.1% in the stem and 28.6% in the root including R6(3)]. Of the 14 isolates from M. domestica, 42.9% showed their potential for N-fixation inside rice plants [21.4% in the stem and 21.4% in the root including R7(1) and R7(3)]. Out of the 14 isolates from F. carica, 14.3% including S8(2) showed their potential ability for N-fixation inside rice plants. This extensive identification and evaluation process underscores the diverse localization potential of forest-isolated diazotrophic microbes inside rice plants.

The ability of bacterial isolates to produce clear zones in skim milk agar plate signified protease activity ( Figure 4A ). The results are detailed in Tables 4 , 5 . Briefly, 31.6% of the bacteria isolated from the rhizosphere soil of A. nordmanniana exhibited protease activity, with the highest activity observed in SS1(1) at 2.9 ± 0.5 followed by SS1(4) at 1.2 ± 0. Similarly, 41.7% of the isolates from C. avellana were capable of catalyzing protein, with the highest catalytic activity found in R2(2)2 at 3.5 ± 0.4 followed by SS2(1)2 at 2.3 ± 0.2. Approximately 20.7% of the isolates from P. elaeagnifolia showed protease activity, with the highest (3.9 ± 0.4) in SS3(4)1 followed by SS3(5)2 at 3.5 ± 0.4. The most significant incidence (70%) of protease production was noted in isolates from A. pseudoplatanus, with the maximum activity measured at 4 ± 0.3 in SS4(7) followed by SS4(10) at 3.8 ± 0.9, R4(1)2 at 1.9 ± 0.3, L4(1) at 1.7 ± 0.2, and SS4(5) at 1.3 ± 0.1. Additionally, 50% of the bacteria derived from the root zone of J. regia and 35.7% from M. domestica demonstrated protease activity, with a notable activity of 5.5 ± 0.6 in isolate SS6(5) and 1.7 ± 0.3 in isolate R7(3). Almost 71.4% of the isolates from the rhizosphere of F. carica were positive for protease production, with the highest activity (2.9 ± 0.1) in isolate SS8(2) followed by SS8(6) at 2.7 ± 0.3. Out of the 41 shortlisted isolates, protease activity was confirmed in 15 bacteria.

Starch digestion was analyzed to assess amylase production ( Tables 4 , 5 , Figure 4B ). The frequencies of amylase-producing bacteria were noted as 15.8%, 16.7%, 20.7%, and 30% from A. nordmanniana (including SS1(1), +), C. avellana (including L2(6)2, +), P. elaeagnifolia (including SS3(4)1, +; SS3(5), +), and A. pseudoplatanus (including SS4(7), +; L4(1), +) as mentioned in Tables 4 , 5 . Of the 41 selected bacteria, 6 were positive for amylase activity.

Cellulase activity was assessed using a carboxymethylcellulose Congo red media plate ( Tables 4 , 5 , Figure 4C ). Approximately 52.6% of the isolates from A. nordmanniana displayed varying levels of cellulase activity from the highest (+++) in S1(G), moderate (++) in R1(B), to low (+) in SS1(1). Almost 16.7% of the isolates (including SS2(4)2, +++) from C. avellana and approximately 10.3% from P. elaeagnifolia (including S3(3), +) exhibited cellulase activity. In contrast, 30% of the bacteria (including L4(1), +++; R4(1)2, +++) from A. pseudoplatanus and 21.4% from R. pseudoacacia (including R5(1), ++) exhibited cellulase activity. Approximately 71.4% of bacteria originating from the rhizosphere soil of F. carica (including SS8(6) and S8) showed activity, but in a low range as given in Table 4 . Of the 41 selected isolates, 10 showed cellulase activity ( Table 5 ).

Variations in enzyme production among rhizosphere-associated bacteria highlight their specialized functions to specific rhizospheric environments.

Hydrogen cyanide (HCN) production was assessed in bacterial isolates, as indicated by the browning of filter paper. Approximately 31.6% of the isolates from the rhizosphere of A. nordmanniana were capable of producing HCN, with isolates SS1(4) and S1(G) exhibiting low production levels (+). Approximately 16.7%% of bacteria isolated from the root zone of C. avellana produced HCN, with moderate production (++) noted in isolate L2(6)2. For P. elaeagnifolia, 31% of the bacterial isolates produced HCN, with isolate SS3(9) showing significant activity (+++), confirmed by a dark brown reaction on the filter paper followed by SS3(2)1 (+) and SS3(2)2 (+), exhibiting low production as given in Tables 4 , 5 .

From A. pseudoplatanus, 45% of the isolates (including SS4(5), +; L4(1), +; and R4(1)2, +) produced HCN. Among the isolates from R. pseudoacacia and J. regia, 21.4% and 50%, respectively, showed HCN production (including R5(1), +and R6(3), +). No HCN production was detected in the isolates from M. domestica or F. carica. Among the 41 selected isolates, 11 were tested positive for HCN ( Table 5 ).

These results underscore the potential of these bacteria for biological control applications, given the inhibitory effects of HCN on plant pathogens.

Ammonia production was detected through a color change in peptone water from yellow to brown. The frequencies of ammonia producers varied among isolates derived from different rhizosphere soils as detailed in Tables 4 , 5 . Specifically, 31.6% of the bacterial isolates [including S1(B) and S1(G)] from the rhizosphere of A. nordmanniana produced ammonia. Similarly, 41.7% of the isolates [including SS2(1)2 and R2(2)2) from C. avellana, 58.6% from P. elaeagnifolia (including SS3(4)1, SS3(5)2, SS3(6)1, and SS3(8)], and 71.4% from F. carica (including SS8(6) and S8) exhibited ammonia production capabilities. In contrast, only 21.4% of the isolates [including R5(1)] from R. pseudoacacia were capable of producing ammonia, while no ammonia-producing bacteria were found among the isolates from A. pseudoplatanus, J. regia, and M. domestica. Ammonia production was confirmed in 13 out of 41 selected bacterial isolates ( Table 5 ).

Particularly, the rhizosphere of P. elaeagnifolia and F. carica showed higher frequencies of ammonia producers, suggesting their potential role in nitrogen cycling within their respective ecosystems.

The study of bacterial isolates revealed a continuous decline in growth with increasing concentrations of NaCl. Approximately 84.2% of the isolates [including SS1(1), SS1(4), S1(B), S1 (G), R1(B)] from A. nordmanniana were able to survived at 3% NaCl concentration. Isolate S1(G) notably withstood NaCl up to 7% concentration. Similarly, 58.3% of the isolates (including SS2(4)2, L2(6)2, and R2(2)2) from C. avellana tolerated 3% NaCl. Particularly, isolate L2(6)2 grew up to 7% NaCl (detailed in Table 6 ).

All the isolates from P. elaeagnifolia survived 3% NaCl, 69% isolates (including SS3(2)1, SS3(2)2, SS3(4)1, SS3(5)1, and SS3(5)2) survived 5% NaCl, and 41.4% isolates [including SS3(4)1, SS3(5)2, S(1), and S3(3)] tolerated as high as 7% NaCl. For A. pseudoplatanus, all the isolates grew at 3% NaCl, with 70% (including SS4(7), SS4(10), L4(1), S4(2), and R4(1)2) surviving up to 7% NaCl (detailed in Table 6 ). From R. pseudoacacia, all the bacteria were viable at 3%, 5%, and 7% NaCl.

Approximately 50% of the isolates from J. regia tolerated NaCl up to 7%. All the isolates from M. domestica tolerated 3% NaCl, though 64.3% managed to survive 7% NaCl including R7(1) and R7(3). Every isolate from F. carica survived at 3% NaCl, but 71.4% of the isolates including SS8(2) and SS8 (6) tolerated up to 7% NaCl concentration (detailed in Table 6 ).

The bacteria associated with the R. pseudoacacia rhizosphere exhibited the highest tolerance at 7% NaCl concentration as shown in Table 6 with 100% survival rate. These results highlighted the potential adaptations and tolerance of the microbial community associated with the rhizosphere of R. pseudoacacia under high salt conditions.

Rhizospheric bacteria associated with R. pseudoacacia, A. pseudoplatanus, and F. carica were highly tolerant; however, J. regia- and M. domestica-associated bacteria exhibited moderate tolerance, while P. elaeagnifolia-, C. avellane-, and A. nordmanniana-associated bacteria displayed the least tolerance, respectively, when provided with high NaCl concentrations.

Based on significant PGP traits, 41 bacterial isolates out of 129 were selected and tested in the germination trial in order to check their effectiveness as rice seed biopriming agents.

No significant change in final germination percentage was recorded with the application of bacterial isolates as compared to the control; however, germination traits such as emerging shoot height (ESH), ERL, ESM, ERM, SVI, SVI2, GE, emerging root shoot (ERS) ratio, and Avg. no. of roots exhibited significant difference from the control as detailed in Table 7 .

Isolate S1(E) derived from A. nordmanniana exhibited superior performance as a biopriming agent (see Figure 5 ) with the highest score of 81.6 followed by SS1(4) with 48.3, SS1(1) with 46.6, S1(G) with 39, S1(B) with 38.3, and R1(B) with 26.5 as shown in Table 7 . Application of bacterial isolates increased the germination energy (GE) (S1(E), 53.8%; SS1(4), 30.9%; SS1(1), 7.8%; S1(G), 30.9; S1(B), 30.9%; and R1(B), 30.9%), SVI1 (S1(E), 190.5%; SS1(4), 101.1%; SS1(1), 117%; S1(G), 56.1%; S1(B), 64.4%; and R1(B), 49%), SVI2 (S1(E), 114.2%; SS1(4), 66.3%; SS1(1), 80%; S1(G), 42.1%; S1(B), 45%; and R1(B), 22%), average number of roots (S1(E), 12.8%; SS1(4), 21.3%; SS1(1), 27.7%; S1(G), 27.7%; S1(B), 27.7%; and R1(B), 0%), ESH (S1(E), 43.9%; SS1(4), 29.3%; SS1(1), 36.6%; S1(G), 36.6%; S1(B), 22%; and R1(B), 9.8%), ERL (S1(E), 1214.3%; SS1(4), 585.7%; SS1(1), 642.9%; S1(G), 357.1%; S1(B), 485.7%; and R1(B), 300%), ESM (S1(E), 40.6%; SS1(4), 31.3%; SS1(1), 43.8%; S1(G), 34.4%; S1(B), 25%; and R1(B), 0%), ERM (S1(E), 193.5%; SS1(4), 103.2%; SS1(1), 119.4%; S1(G), 64.5%; S1(B), 66.7%; and R1(B), 61.3%), and ERS ratio (S1(E), 87.5%; SS1(4), 50%; SS1(1), 50%; S1(G), 12.5%; S1(B), 25%; and R1(B), 37.5%) with respect to control. All the selected isolates from A. nordmanniana showed overall positive impact on germination.

Among the bacterial isolates obtained from the rhizosphere of C. avellane, isolate R2(2)2 got the highest score of 62.1 followed by SS2(1)2 with 47.2, SS2(4) with 37.8, L2(6)2 with 36.4, and SS2(4)2 with 28.3 as mentioned in Table 7 . Treatment with bacterial isolates increased the different germination parameters such as germination energy (R2(2)2, 46.2%; SS2(1)2, 15.3%; SS2(4), 23.1%; L2(6)2, 38.4%; and SS2(4)2, 15.3%), seedling vigor index 1 (R2(2)2, 101.1%; SS2(1)2, 115.8%; SS2(4), 55%; L2(6)2, 49%; and SS2(4)2, 40.1%), seedling vigor index 2 (R2(2)2, 74.2%; SS2(1)2, 69.7%; SS2(4), 31.7%; L2(6)2, 37.4%; and SS2(4)2, 23.7%), average number of roots (R2(2)2, 21.3%; SS2(1)2, 27.7%; SS2(4), 97.9%; L2(6)2, 6.4%; and SS2(4)2, 42.6%), emerging shoot height (R2(2)2, 58.5%; SS2(1)2, 43.9%; SS2(4), 22%; L2(6)2, 29.3%; and SS2(4)2, 29.3%), emerging root length (R2(2)2, 900%; SS2(1)2, 471.4%; SS2(4), 271.4%; L2(6)2, 442.9%; and SS2(4)2, 328.6%), emerging shoot mass (R2(2)2, 46.9%; SS2(1)2, 31.3%; SS2(4), 15.6%; L2(6)2, 25%; and SS2(4)2, 18.8%), emerging root mass (R2(2)2, 103.2%; SS2(1)2, 129%; SS2(4), 64.5%; L2(6)2, 51.6%; and SS2(4)2, 54.8%), and emerging root shoot ratio (R2(2)2, 25%; SS2(1)2, 50%; SS2(4), 25%; L2(6)2, 12.5%; and SS2(4)2, 12.5%) as compared to control. All the selected isolates from C. avellana presented overall positive affect as biopriming agents on the germination traits of rice (see Figure 5 ).

The application of selective bacterial isolates derived from the root zone of P. elaeagnifolia showed overall positive results on germination parameters (see Figures 5 , 6 ) except a few isolates like SS3(5)1, SS3(4)1, and S3(3) showing little, no, or negative impact on some of the germination parameters as mentioned in Table 7 . The highest score of 54.8 was achieved by the isolate SS3(2)2 followed by SS3(6)1 with 35, SS3(8) with 26.1, S3(1) with 25.7, SS3(9) with 23.9, SS3(2)1 with 20, SS3(5)2 with 18.5, S3(3) with 14.4, SS3(4)1 with 12.7, and SS3(5)1 with 7.3. Application of the bacterial isolate improved the germination energy (SS3(2)2, 53.8%; SS3(6)1, 38.4%; SS3(8), 15.3%; S3(1), 23.1%; SS3(9), 30.9%; SS3(2)1, 23.1%; SS3(5)2, 15.3%), seedling vigor index 1 (SS3(2)2, 80.9%; SS3(6)1, 44.7%; SS3(8), 47.4%; S3(1), 23.6%; SS3(9), 27.7%; SS3(2)1, 13.2%; SS3(5)2, 21%; S3(3), 49%; SS3(4)1, 37.1%; and SS3(5)1, 10.7%), seedling vigor index 2 (SS3(2)2, 60.5%; SS3(6)1, 32.6%; SS3(8), 38.2%; S3(1), 19.2%; SS3(9), 15.3%; SS3(2)1, 11.5%; SS3(5)2, 2.2%; S3(3), 37.4%; SS3(4)1, 36.4%; and SS3(5)1, 5.8%), average number of roots (SS3(2)2, 12.8%; SS3(6)1, 27.7%; SS3(8), 21.3%; S3(1), 42.6%; SS3(9), 42.6%; SS3(2)1, 12.8%; SS3(5)2, 6.4%; S3(3), 34%; SS3(4)1, 34%; and SS3(5)1, 48.9%), emerging shoot height (SS3(2)2, 43.9%; SS3(6)1, 14.6%; SS3(8), 22%; S3(1), 12.2%; SS3(2)1, 24.4%; SS3(5)2, 24.4%; S3(3), 22%; and SS3(4)1, 9.8%), emerging root length (SS3(2)2, 571.4%; SS3(6)1, 342.9%; SS3(8), 171.4%; S3(1), 171.4%; SS3(9), 157.1%; SS3(2)1, 142.9%; SS3(5)2, 300%; S3(3), 228.6%; SS3(4)1, 342.9%; and SS3(5)1, 171.4%), emerging shoot mass (SS3(2)2, 40.6%; SS3(6)1, 21.9%; SS3(8), 28.1%; S3(1), 25%; SS3(9), 3.1%; SS3(2)1, 15.6%; SS3(5)2, 3.1%; S3(3), 25%; and SS3(4)1, 40.6%), emerging root mass (SS3(2)2, 83.9%; SS3(6)1, 45.2%; SS3(8), 48.4%; S3(1), 38.7%; SS3(9), 29%; SS3(2)1, 19.4%; SS3(5)2, 51.6%; S3(3), 51.6%; SS3(4)1, 45.2%; and SS3(5)1, 12.9%), and emerging root shoot ratio (SS3(2)2, 25%; SS3(6)1, 25%; SS3(8), 12.5%; S3(1), 12.5%; SS3(9), 37.5%; SS3(5)2, 12.5%; S3(3), 12.5%; SS3(4)1, 37.5%; and SS3(5)1, 37.5%) with respect to control.

Most of the selected isolates from A. pseudoplatanus showed overall positive effect ranging from low to moderate on the germination of rice (see Figure 6 ); however, few isolates—SS4(10), SS4(7), L4(1), and S4(2)—were found to influence some of the germination parameters negatively. Among the tested isolates, SS4(5) recorded with the highest score of 48.6 followed by SS4(8) with 43.3, SS4 (10) with 38, R4(1)2 with 35.6, S4(2) with 24.8, L4(1) with 11.6, and SS4(7) with 5.8 as mentioned in Table 7 . Biopriming with bacterial isolates enhanced the seedling vigor index 1 (SS4(5), 133%; SS4(8), 75.6%; SS4(10), 129.8%; R4(1)2, 82.3%; S4(2), 8.7%; L4(1), 2.5%; and SS4(7), 9.6%), seedling vigor index 2 (SS4(5), 90%; SS4(8), 53.7%; SS4(10), 81.6%; R4(1)2, 51.1; S4(2), 22.1%; and SS4(7), 11.6%), average number of roots (SS4(5), 27.7%; SS4(8), 34%; SS4(10), 21.3%; R4(1)2, 12.8%; S4(2), 21.3%; L4(1), 27.7%; and SS4(7), 34%), emerging shoot height (SS4(5), 36.6%; SS4(8), 31.7%; SS4(10), 36.6%; R4(1)2, 39%; S4(2), 31.7%; L4(1), 19.5%; and SS4(7), 2.4%), emerging root length (SS4(5), 742.9%; SS4(8), 500%; SS4 (10), 914.3%; R4(1)2, 700%; S4(2), 300%; L4(1), 171.4%; and SS4(7), 142.9%), emerging shoot mass (SS4(5), 46.9%; SS4(8), 31.3%; SS4(10), 34.4%; R4(1)2, 25%; S4(2), 50%; and SS4(7), 12.5%), emerging root mass (SS4(5), 135.5%; SS4(8), 77.4%; SS4(10), 132.3%; R4(1)2, 93.5%; S4(2), 22.6%; L4(1), 12.9%; and SS4(7), 9.7%), and emerging root shoot ratio (SS4(5), 62.5%; SS4(8), 25%; SS4 (10), 62.5%; and R4(1)2, 25%) in comparison with control.

From the rhizosphere of R. pseudoacacia, isolate R5(1) (score −3.1) exhibited an overall negative impact on germination as shown in Table 7 and Figure 7 ; opposite to this, isolate S5(2)1 showed the highest score of 50 followed by S5(1) with 16.4, R5(1)1 with 13.9, and SS5(5) with 11.1 as detailed in Table 7 . Biopriming with bacterial isolates increased the germination energy (S5(2)1, 46.2%; and R5(1), 23.1%), seedling vigor index 1 (S5(2)1, 101.1%; S5(1), 66%; R5(1)1, 42.6%; and SS5(5), 35.4%), seedling vigor index 2 (S5(2)1, 56.3%; S5(1), 35.8%; R5(1)1, 30.5%; and SS5(5), 33.5%), average number of roots (S5(2)1, 34%; S5(1), 21.5%; R5(1)1, 21.3%; and SS5(5), 12.8%), emerging shoot height (S5(2)1, 7.3%; R5(1)1, 24.4%; and SS5(5), 17.1%), emerging root length (S5(2)1, 528.6%; S5(1), 414.3%; R5(1)1, 300%; and SS5(5), 285.7%), emerging shoot mass (S5(2)1, 12.5%; S5(1), 6.2%; R5(1)1, 18.8%; and SS5(5), 37.5%), emerging root mass (S5(2)1, 103.2%; S5(1), 67.7%; R5(1)1, 45.2%; and SS5(5), 45.2%), and emerging root shoot ratio (S5(2)1, 75%; S5(1), 87.5%; R5(1)1, 12.5%; and SS5(5), 12.5%) with respect to control.

Isolate R6(3) got the highest score of 56.2 followed by SS6(5) with 40.3 in the germination trial among the isolates obtained from the rhizosphere soil of J. regia as shown in Table 7 and Figure 7 , and their application improved the GE (R6(3), 30.9%; SS6(5), 23.1%), SVI1 (R6(3), 149.3%; SS6(5), 76.6%), SVI2 (R6(3), 79.9%; SS6(5), 66.3%), Avg. no. of roots ((R6(3), 34%; SS6(5), 34%), ESH ((R6(3), 39%; SS6(5), 26.8%), ERL (R6(3), 485.7%; SS6(5), 442.9%), ESM (R6(3), 56.3%; SS6(5), 18.8%), ERM (R6(3), 164.5%; SS6(5), 77.4%), and ERS ratio (R6(3), 75%; SS6(5), 37.5%) as compared to the control.

Isolate SS7(5) showed the best performance (score 59.7) followed by R7(1) with 31.4 and R7(3) with 21.2 among the selected isolates derived from the root zone of M. domestica as mentioned in Table 7 and Figure 7 . Treatment with bacterial isolates enhanced the GE (R7(1), 15%; R7(3), 38%), SVI1 (SS7(5), 171.6%; R7(1), 37.1%), SVI2 (SS7(5), 118.6%; R7(1), 37.1%; R7(3), 0.3%), Avg. no. of roots (SS7(5), 21.3%; R7(1), 42.6%), ESH (SS7(5), 29.3%; R7(1), 31.7%; R7(3), 22%), ERL (SS7(5), 814.3%; R7(1), 328.6%; R7(3), 200%), ESM (SS7(5), 75%; R7(1), 43.8%; R7(3), 18.8%), ERM (SS7(5), 187.1%; R7(1), 45.2%; R7(3), 6.5%), and ERS ratio (SS7(5), 112.5%) with respect to control.

Bacterial isolate SS8(2) showed the most positive outcome (score 54) followed by S8 with 25.5 and SS8(6) with 24.8, among the isolates derived from the rhizosphere soil of F. carica in the germination trial as shown in Table 7 and Figure 7 . Seed biopriming with these isolates enhanced the GE (SS8(2), 23.1%; S8, 23.1%; SS8(6), 15.3%), SVI1 (SS8(2), 137.6%; S8, 28.7%; SS8(6), 29.8.3%), SVI2 (SS8(2), 80.4%; S8, 16.2%; SS8(6), 33.2%), Avg. no. of roots (SS8(2), 27.7%; S8, 42.6%; SS8(6), 27.7%), ESH (SS8(2), 43.9%; S8, 24.4%; SS8(6), 26.8%), ERL (SS8(2), 428.6%; S8, 142.9%; SS8(6), 85.7%), ESM (SS8(2), 31.3%; S8, 9.4%; SS8(6), 37.5%), ERM (SS8(2), 151.6%; S8, 38.7%; SS8(6), 32.3%), and ERS ratio (SS8(2), 62.5%; S8, 12.5%) with respect to control.

The application of selective bacterial isolates derived from different tree rhizospheres significantly improved the germination traits. These results demonstrated the potential of bacterial biopriming as an effective strategy for enhancing crop emergence and early growth.

By analyzing the images taken by an SEM, one can gain insights into the interactions between rice roots and plant growth-promoting bacteria. The study on the efficiency of bacterial colonization and potential effects on root morphology is essential for understanding the beneficial relationships in plant–microbe interactions. SEM examination of rice root sections indicated a uniform distribution of isolate S1(E) in a large quantity across the root surface as shown in Figure 8D . In contrast, isolates SS1(4), SS4(10), S1(G), SS7(5), and S3(1) were mainly found segregated or clustered in the surface furrows at the junctions of epidermal cells ( Figures 8A-C, E, F ). Rice seedlings that were not inoculated with bacteria displayed a smooth and intact epidermal surface ( Figure 8G ). Observations revealed that roots inoculated with isolates S1(E) and SS7(5) exhibited high bacterial population colonized to the roots as compared to those inoculated with isolates SS1(4), SS4(10), S1(G), and S3(1), as depicted in Figure 8 .

The root inoculation with selected bacterial isolates exhibited remarkable influence on the growth of 21-day-old rice seedlings in a hydroponic culture, as detailed in Table 8 and Figure 9 .

Root inoculation with the selected bacteria isolated from the root zone of A. nordmanniana showed overall positive impact on rice growth; however, a few isolates like SS1(1) and R1(B) were found to influence some of the growth parameters negatively. Isolate S1(E) proved to be the most effective (see Figure 9 ) with the highest score of 56.2 followed by S1(B) with 54.7, S1(G) with 52.7, SS1(4) with 41.0, R1(B) with 35.7, and SS1(1) with 29.7 as detailed in Table 8 . Inoculation of rice root with bacterial isolates boosted different growth parameters such as plant fresh weight (PFW) (S1(E), 147.6%; S1(B), 96.4%; S1(G), 77.1%; SS1(4), 54.5%; R1(B), 89.7%; and SS1(1), 47.2%), shoot height (SH) (S1(E), 25.3%; S1(B), 39.1%; S1(G), 40.4%; SS1(4), 14.7%; and R1(B), 32.9%), root length (RL) (S1(E), 5.1%; S1(B), 30.4%; S1(G), 2.5%; and SS1(4), 36.7%), dry shoot mass (DSM) (S1(E), 55.7%; S1(B), 73.8%; S1(G), 86.9%; SS1(4), 32.1%; R1(B), 57.4%; and SS1(1), 15.6%), dry root mass (DRM) (S1(E), 60%; S1(B), 48.6%; S1(G), 53.2%; SS1(4), 52.2%; R1(B), 25.4%; and SS1(1), 53.3%), and root shoot (RS) ratio (S1(E), 25%; SS1(4), 25%; and SS1(1), 50%) as compared to the control (see Table 8 , Figure 9 ).

Generally, the inoculation with the selective isolates from C. avellana improved the overall growth scoring except isolate R2(2)2 (score −4.9). However, some of the growth parameters including root length were negatively influenced with the application. Isolate SS2(1)2 exhibited the highest score of 41.1 followed by SS2(4)2 with 30.5, SS2(4) with 26.1, and L2(6)2 with 22.8 as mentioned in Table 8 . Treatment with bacterial isolates increased plant fresh weight (SS2(1)2, 99.7%; SS2(4)2, 47.6%; SS2(4), 49.6; L2(6)2, 15%; and R2(2)2, 5.3%), shoot height (SS2(1)2, 7.6%; SS2(4)2, 32.9%), dry shoot mass (SS2(1)2, 54.4%; SS2(4)2, 39.2%; and L2(6)2, 7.6%), dry root mass (SS2(1)2, 78.1; SS2(4)2, 31.4%; L2(6)2, 47.6%; and R2(2)2, 4.8%), and root shoot ratio (SS2(1)2, 25%; L2(6)2, 50%; and R2(2)2, 25%) as compared to the control.

Among the selected isolates from the rhizosphere of P. elaeagnifolia, the impact of inoculating rice roots was positive on overall growth scoring except the isolates SS3(2)1 (score −15.5) and S3(3) (score −11.1). Some bacterial applications negatively affected certain growth parameters. Isolate S3(1) exhibited the highest score of 66.9 (see Table 8 , Figure 9 ) followed by SS3(5)1 with 60.3, SS3(4)1 with 39.6, SS3(2)2 with 35.5, SS3(9) with 39.1, SS3(6)1 with 22.2, SS3(5)2 with 19.8, and SS3(8) with 6. Treatment with bacterial isolates showed notable improvements in PFW (S3(1), 68.4%; SS3(5)1, 47.5%; SS3(4)1, 49.9%; SS3(2)2, 44.7%; SS3(9), 82.1%; SS3(6)1, 33.3%; SS3(5)2, 15.3%; S3(3), 28.3%; SS3(2)1, 9.3%; and SS3(8), 6.6%), shoot height (S3(1), 13.8%; SS3(4)1, 17.3%; SS3(2)2, 17.3%; SS3(9), 28.9%; SS3(6)1, 20.9%; and SS3(5)2, 1.8%), root length (S3(1), 40.5%; SS3(5)1, 94%; SS3(4)1, 19%), dry shoot mass (S3(1), 58.2%; SS3(5)1, 43.9%; SS3(4)1, 39.7%; SS3(2)2, 44.3%; SS3(9), 65.4%; SS3(6)1, 30.4%; SS3(5)2, 48.5%; and SS3(8), 5.9%), dry root mass (S3(1), 112.4%; SS3(5)1, 86.7%; SS3(4)1, 47.6%; SS3(2)2, 55.2%; SS3(9), 44.8%; SS3(6)1, 23.8%; SS3(5)2, 47.6%; and SS3(8), 21.9%), and root shoot ratio (S3(1), 50%; SS3(5)1, 50%; SS3(4)1, 25%; SS3(2)2, 25%; SS3(8), 25%; S3(3), 25%; and SS3(2)1, 25%) in comparison with the control as detailed in Table 8 .

From the A. pseudoplatanus rhizosphere, isolate L4(1) was particularly effective with the highest score of 54 followed by SS4(8) with 45.3, SS4(5) with 32, SS4(7) with 22.4, R4(1)2 with 19.5, SS4(10) with 10.4, and S4(2) with 8.1 as mentioned in Table 8 . Treatments with different bacterial inoculations resulted in overall better growth; however, few bacterial isolates showed negative impacts on some of the growth parameters. The application of bacterial isolates improved plant fresh weight (L4(1), 55.5%; SS4(8), 66.7%; SS4 (5), 25.4%; SS4(7), 62.9%; R4(1)2, 31.1%; SS4(10), 6.3%; and S4(2), 16.1%), shoot height (L4(1), 11.1%; SS4(8), 11.6%; SS4(5), 12.4%; R4(1)2, 10.2; and SS4(7), 12.4), root length (L4(1), 19%; SS4(5), 5.1%; and SS4(10), 6.3%), dry shoot mass (L4(1), 71.3%; SS4(8), 31.6; SS4(5), 35.4%; R4(1)2, 13.1%; and SS4(7), 40.9%), dry root mass (L4(1), 97.1%; SS4(8), 69.5%; SS4(5), 46.7%; R4(1)2, 22.9%; SS4(7), 37.1; S4(2), 18.1%; and SS4(10), 19%), and root shoot ratio (L4(1), 25%; SS4(8), 50%; SS4(5), 25%; R4(1)2, 25%; S4(2), 50%; and SS4(10), 50%) as compared to the control as detailed in Table 8 .

Among the selected isolates from the rhizosphere of R. pseudoacacia, root inoculation with isolate R5(1)1 exhibited better growth (see Figure 9 ) with the highest score of 57.7 followed by S5(1) with 51.1, SS5(5) with 26.4, S5(2)1 with 4.2, and R5(1) with −1.5. Treatment with bacterial isolates improved the plant fresh weight (R5(1)1, 108.8%; S5(1), 82.8%; SS5(5), 60.3%; and R5(1), 24.6%), shoot height (R5(1)1, 22.2%; S5(1), 25.8%; SS5(5), 29.8%; and R5(1), 8.9%), root length (R5(1)1, 17.7%; S5(1), 10.1%), dry shoot mass (R5(1)1, 58.6%; S5(1), 48.5%; SS5(5), 43.5%; S5(2)1, 0.4%; and R5(1), 10.1%), dry root mass (R5(1)1, 81.9%; S5(1), 72.4%; SS5(5), 24.8%; and S5(2)1, 12.4%), and root shoot ratio (25% increase for R5(1)1, S5(1), and S5(2)1) in comparison with the control. Isolates like S5(2)1 and R5(1) exhibited negative impacts on some of the growth parameters as mentioned in Table 8 .

Isolate SS6(5) from the root zone of J. regia scored 44.6 (see Table 8 , Figure 9 ). Its application enhanced PFW (44.6%), RL (6.3%), DSM (44.7%), DRM (88.6%), and RS ratio (50%) as compared to the control. In contrast, root inoculation with isolate R6(3) resulted in growth reduction (score −6.9).

Isolate SS7(5) from the rhizosphere of M. domestica exhibited the highest score of 49.4 (see Table 8 , Figure 9 ) followed by R7(1) with 44.4, while isolate R7(3) exhibited negative effects (score −11.6) on rice growth. Treatment with isolates SS7(5) and R7(1) enhanced the plant fresh weight (118.2%, 80.4%), shoot height (37.8%, 28.9%), dry shoot mass (52.7%, 71.7%), and dry root mass (41.9%, 69.5%), respectively, as compared to the control. Isolate SS7(5) also exhibited an increase of 24.1% in root length than control.

Among the selected isolate from the root zone of F. carica, root inoculation with isolate SS8(2) showed the highest performance as a growth-improving agent with a score of 27.7 followed by SS8(6) with 11.6 and S8 with 3.9 as mentioned in Table 8 . Treatment with SS8(2) increased PFW (41.8%), DSM (14.3%), DRM (44.8%), and RS ratio (50%).

These results underscore the potential effects of these bacterial isolates on enhancing various growth parameters of rice seedlings in hydroponic culture, demonstrating their effectiveness as growth-stimulating agents.

Of the 41 selected isolates, 14 bacterial isolates of different rhizospheric origins exhibited a significant impact on germination and growth traits of rice along with their diverse PGP traits. These isolates were selected for their molecular identification through 16S rRNA gene sequencing. The 16S rDNA sequences of the respective isolates were compared in the EzBioCloud database to find out similar sequences ( Table 9 ). The homologous sequences were aligned to construct their phylogenetic trees for the examination of their relatedness with other close members ( Supplementary Figures 1S–4S ). The 16S rDNA gene sequences of the respective isolates were submitted to NCBI GenBank. The following accession numbers were obtained: PQ605795, S1(E); PQ605785, SS1(4); PQ605786, SS2(1)2; PQ605787, SS3(2)2; PQ605796, S3(1); PQ605788, SS4(5); PQ605789, SS4(8); PQ605794, R5(1)1; PQ605790, SS6(5); PQ605791, SS7(5); PQ605793, R(7)1; and PQ605792, SS8(2) (see Table 9 ). The sequence of the S1(E) isolate (source: A. nordmanniana rhizosphere) was found to be 99.85% similar to Herbaspirillum huttiense subsp. huttiense, isolate SS1(4) (source: A. nordmanniana rhizosphere) was 99.06% similar with Viridibacillus arenosi, SS2(1)2 isolate (source: C. avellana rhizosphere) showed 99.01% similarity with Psychrobacillus faecigallinarum, isolate SS3(2)2 (source: P. elaeagnifolia rhizosphere) exhibited 99.71% similarity with Lysinibacillus fusiformis, isolate S3(1) (source: P. elaeagnifolia rhizosphere) was 99.05% similar to Lelliottia amnigena, SS4(5) isolate (source: A. pseudoplatanus rhizosphere) showed 99.64% similarity with Bacillus siamensis, isolate SS4(8) (source: A. pseudoplatanus rhizosphere) was 99.93% similar with Microbacterium phyllosphaerae, isolate R4(1)2 (source: A. pseudoplatanus rhizosphere) was 100% similar with Bacillus altitudinis, isolates S5(2)1 and R5(1)1 (source: R. pseudoacacia rhizosphere) exhibited 100% and 99.85% similarity with Acinetobacter calcoaceticus, isolate SS6(5) (source: J. regia rhizosphere) displayed 99.42% similarity with Micrococcus luteus, SS7(5) and R7(1) isolates (source: M. domestica rhizosphere) were found to be 99.26% and 99.28% similar to Pseudomonas mohnii and Lelliottia sp. respectively, while isolate SS8(2) (source: F. carica rhizosphere) exhibited 99.43% similarity with Staphylococcus succinus ( Table 9 ). It was found that selective isolates in the study belong to diverse genera, and the genus Bacillus was the single most dominant genus among the isolates ( Supplementary Figures 1S–4S ). Genetic diversity among isolates sourced from rhizosphere soils of different trees can be associated with their broad range in PGP traits and resulting effects on rice growth.

Principal component (PC) 1 highly contributed (23.4%) to illustrate the variability in PGP traits of bacterial isolates followed by PC2 (18.2%), PC3 (15.9%), PC4 (13.1%), and PC5 (9.7%). The cumulative percentage of variance for the first five PCs was 80.2% along with the first four PCs exhibiting eigenvalues >1 ( Figure 10A ). PC1 was positively affected by SDP formation, HCN production, cellulase activity, and IAA, while it was negatively affected by amylase activity, protease activity, and PS. PC2 was positively influenced by almost all PGP traits except IAA and PS ( Figure 10B ).

Biplot represented each PGP trait with a unique loading vector starting from the origin and its relationship with different bacterial isolates derived from the rhizosphere of various trees. Furthermore, the biplot explained almost 41.6% of the variations in data based on PC1 and PC2 as shown in Figure 10B . All the vectors except NH₃ dispersed away from the origin representing the high variability of PGP traits among bacterial isolates. Isolates L2(6)2, L4(1), and R5(1) showed high, S8 moderate, while SS5(5) and SS3(2)1 exhibited low association with SDPs. Isolate S3(9) demonstrated high, S1(G) moderate, and SS3(2)1, SS3(2)2, and SS1(4) displayed low association with HCN. S1(G) was highly linked, while R5(1) was moderately linked to cellulase activity. Also, three of the vectors (SDPS, HCN, cellulase) were highly interrelated; particularly, HCN and cellulase activity were more closely associated ( Figure 10B ). Majority of the isolates derived from the rhizosphere of eight different trees somehow exhibited high [i.e., SS3(8), SS3(5)1, SS2(4), and R1(B)], moderate, or low association with IAA. Production of IAA was slightly linked with cellulase activity and HCN production. Isolates S5(2)1, SS8(2), S5(1), and R5(1)1 displayed high and R7(1) presented moderate association with the PS vector. Protease activity was highly linked to SS3(5)2, SS3(4)1, SS4(7), SS6(5), and SS1(1); moderately linked to R2(2)2 and SS4(10); and poorly associated with SS8(6) and R7(3). Isolates SS1(1) and SS3(5)2 were closely associated with amylase activity. Three of the vectors (amylase, protease, PS) exhibited interrelationship with each other, where amylase activity was closely linked to protease activity which was closer to PS ( Figure 10B ).

Isolates derived from the rhizosphere of R. pseudoacacia were closely associated with PS activity. Most of the isolates derived from A. nordmanniana and P. elaeagnifolia were linked to IAA ( Figure 10B ). Similarly, isolates derived A. nordmanniana, P. elaeagnifolia, R. pseudoacacia, and J. regia were associated with HCN and cellulase activity. Meanwhile, some isolates from the root zone of C. avellana, A. pseudoplatanus, and R. pseudoacacia were linked to SDPs. Some isolates from these rhizospheres possessed the ability to localize within the root, stem, or leaf of rice plants. A few of the isolates derived from the root zone of A. nordmanniana, P. elaeagnifolia, and A. pseudoplatanus were also linked with protease and amylase activity ( Figure 10B ).

PCA revealed that PC1 highly contributed (59.3%) to explain the variability in germination traits of rice with the application of bacterial isolates followed by PC2 (14%), PC3 (11.1%), PC4 (9.3%), and PC5 (4%). Together, the first five principal components explained 97.6% of the cumulative variance, with the first two PCs having eigenvalues greater than 1 ( Figure 11A ). PC1 was positively influenced by all of the germination traits (ERS ratio, ERM, SVI2, ESM, ESH, SVI1, ERL, GE) except the average numbers of roots. PC2 was positively affected by GE, ESH, ESM, and ERL and negatively influenced by ANR, ERS ratio, ERM, SVI2, and SVI1 ( Figure 11B ).

The biplot displays each germination trait with specific loading vectors starting from the origin in correspondence with the influence of different bacterial isolates on it. Furthermore, the biplot elucidated almost 73.3% of the variability in data based on PC1 and PC2 as revealed in Figure 11B . All the vectors dispersed away from the origin with varying magnitudes displaying variability in germination traits upon biopriming the rice seeds with different bacterial isolates. Based on PC1, all the bacterial isolates except R5(1) showed better association with different germination traits as compared to the control ( Figure 11B ). The biplot illustrated that a large number of isolates were clustered around GE. Isolate R2(2)2 was highly associated, SS3(2)2, SS8(2), R4(1)2, SS6(5), and SS1(1) were moderately associated, and S1(G) and SS3(6)1 were slightly associated with ESH and ESM. Germination traits such as GE, ESH, ESM, and ERL exhibited interrelationship with each other. Isolates S1(E) and SS7(5) were highly associated, whereas SS4(5), SS2(1)2, SS4(10), SS1(4), SS4(8), and SS1(1) were moderately associated with ERL, SVI2, ERM, SVI1, and ERS ratio. Also, parameters such as ERL, SVI2, ERM, SVI1, and ERS ratio were interlinked with each other. Isolates S5(1), SS2(4), SS3(4), SS3(5)1, and SS3(9) were highly associated, S3(1) and SS4(8) were moderately associated, and R5(1)1 and SS3(6)1 were slightly associated with ANR. GE and ANR exhibited a negative relationship with each other ( Figure 11B ).

Isolates almost from every rhizosphere exhibited an association with GE. Some of the isolates derived from the rhizosphere of A. nordmanniana, C. avellana, M. domestica, and A. pseudoplatanus were linked to ERL, SVI2, ERM, SVI1, ESM, ESH, and ERS ratio ( Figure 11B ), and only a few of them possessed the ability to localize within the stem and root of rice plants. A few of the isolates from F. carica and P. elaeagnifolia were also associated with ESM and ESH. Similarly, some isolates derived from C. avellana, P. elaeagnifolia, and R. pseudoacacia were linked with ANR with a few possessing the ability to colonize rice stem ( Figure 11B ).

PCA revealed that PC1 greatly contributed (59.2%) to elucidate the variability in growth parameters of rice upon bacterial inoculation followed by PC2 (25.8%), PC3 (8.1%), PC4 (4.7%), and PC5 (1.9%). The cumulative percentage of variance together for first five PCs was 99.7% along with the first two PCs presenting eigenvalues >1 ( Figure 12A ). PC1 was positively influenced by all growth traits (RL, DRM, DSM, PFW, SH) except the RS ratio. PC2 was positively influenced by RL, DRM, and RS ratio, while it was negatively affected by DSM, PFW, and SH ( Figure 12B ).

The biplot represents each growth trait with a unique vector starting from the origin in relation with bacterial isolates of different rhizospheres. Furthermore, the biplot described almost 85% of the variations in data based on PC1 and PC2 as shown in Figure 12B . All the vectors spread away from the origin with varying magnitudes representing a high degree of variability in growth indices upon the application of different bacterial isolates. Isolates S3(1), L4(1), and R5(1)1 were highly associated, SS4(8), S5(1), SS1(4), and SS3(4)1 were moderately associated, while SS2(1)2 and SS4(5) were slightly associated with RL and DRM. Both RL and DRM vectors were almost aligned in the same direction and showed high interrelationship with each other ( Figure 12B ). Isolates S1(E), S1(B), S1(G), SS7(5), and R7(1) demonstrated high association with DSM, PFW, and SH. Moreover, SS3(9), SS2(4)2, R1(B), and SS2(4) were found to be moderately associated with DSM, PFW, and SH, while SS3(5)2, SS3(6)1, SS4(7), and SS3(2)2 were slightly linked to DSM, PFW, and SH. Three of the vectors (DSM, PFW, and SH) showed a very close interrelationship with each other. On the other side of the biplot, the root shoot ratio of seedlings was found to be moderately associated with isolates L2(6)2, SS1(1), and SS8(2), while it was slightly linked to SS4(5). Moreover, the RS ratio vector was more or less at the 90° angle associated with RL and DRM ( Figure 12B ).

Some of the isolates derived from the rhizosphere of A. pseudoplatanus, P. elaeagnifolia, and R. pseudoacacia were linked to RL and DRM ( Figure 12B ). Most of the isolates derived from A. nordmanniana, M. domestica, and P. elaeagnifolia were associated with DSM, PFW, and SH ( Figure 12B ). Similarly, a few isolates derived from C. avellana, A. nordmanniana, and F. carica were linked to the seedling RS ratio and average numbers of roots ( Figure 12B ). Some of the isolates from these rhizospheres were confirmed for their endophytic nitrogen fixation within the leaf, stem, and root of rice plants.

Cluster hierarchical analysis of isolates derived from the rhizosphere of different trees was performed on the basis of different PGP traits (i.e., IAA, SDPs, PS, amylase, protease, cellulase, NH₃, HCN), germination traits (i.e., GE, ESH, ERL, ESM, ERM, SVI1, SVI2, Avg. no. of roots), and growth traits (i.e., RL, DRM, DSM, PFW, SH, seedlings RS ratio). Cluster analysis of PGP traits grouped the 41 isolates into nine clusters at 90% similarity ( Table 10 ). Cluster I contained 19 isolates followed by 4, 7, 2, 3, 2, 2, 1, and 1, respectively, in clusters II–IX ( Figure 13A ). Cluster analysis of germination traits gathered the 41 isolates into nine clusters at 80% similarity ( Table 11 ). Cluster I enclosed 5 isolates followed by 3, 5, 2, 13, 2, 8, 2, and 1, respectively, in clusters II–IX ( Figure 13B ). Cluster analysis of growth traits grouped the 41 isolates into nine clusters at 74% similarity ( Table 12 ). Cluster I contained 9 isolates followed by 5, 8, 2, 6, 1, 3, and 1, respectively, in clusters II–IX ( Figure 13C ).