The 15-member SynCom was created using results from two interconnected experiments. In the first study (Acharya et al., 2023b), the microbe recruitment along the root surface was studied in young Brachypodium plants grown in natural soil. In the subsequent study (Chen et al., 2024), over 750 stable reduced community consortia (RCC) were enriched by growing root-attached microbes from the previous study over either 3 or 7 days in media with carbon substrates generally present in Brachypodium root exudates. In this study, based on high species diversity as denoted by the richness and Shannon diversity index, enrichments grown on two media, i.e., 0.1X R2A (BD Diagnostics), and RCH2 minimal media (Chakraborty et al., 2017) supplemented with carbon sources (either Glutamine or mixed carbon), were used to isolate colonies on the corresponding agar media plates by incubating the plates at 30°C in the dark for 7 days. Morphologically distinct colonies were picked and streaked for further purification. For species identification of the isolates, genomic DNA extracted using a PureLink Genomic DNA Mini Kit (Invitrogen, United States), was amplified using the universal 16S rRNA eubacterial primer pairs, i.e., 8F/27F and 1492R. DNA sequencing was carried out at the UC Berkeley DNA Sequencing Facility. Geneious Prime v2020.2.5 was used to process the 16S reads and the resulting consensus sequences were taxonomically classified using the SILVA database (Gurevich et al., 2013).

The whole genome sequencing was performed at Novogene (Illumina Novaseq 6,000 platform). Genome assembly and annotation were performed using the KBase platform (Arkin et al., 2018). Raw reads quality was assessed with FastQC v0.11.9, followed by trimming using Trimmomatic v0.36 (Bolger et al., 2014). The reads were then assembled using Spades v3.15.3 (Prjibelski et al., 2020) and the genome quality was evaluated using CheckM v1.0.18 (Parks et al., 2015). DRAM v0.1.2 (Shaffer et al., 2020) and Blastkoala (Kanehisa et al., 2016) were used to assign KEGG orthology (KO) numbers and functional annotation to the protein-coding genes and KEGG mapper (Kanehisa and Sato, 2020) was used to visualize metabolic pathways. Annotations from the metabolic assembly output files obtained from DRAM and KEGG were used to search for specific properties or genes such as PGP traits, transporters, and osmoprotectant-related genes. For the taxonomic classification, a concatenated alignment of 120 single-copy marker proteins was created using the GTDB-tk workflow (Parks et al., 2022; Chaumeil et al., 2022). Subsequently, 4–5 representative sequences closest to the SynCom strains were selected and aligned using clustalo (Sievers et al., 2011). This was followed by the tree construction with FastTree (Price et al., 2009) and visualization using iTOL (Letunic and Bork, 2007).

We evaluated in vitro stability of the 15-member SynCom by tracking changes in the relative abundances of the strains over a three-week period. For this, we cultured 5 mL of the inoculum, containing equal cell numbers (4×10⁷ cells) of each SynCom member, in 45 mL of 0.2X MS (Murashige and Skoog) media (M0404, Sigma Aldrich, United States). The mixture was incubated in a plant growth chamber with conditions set at 16-h light, 24°C temperature, and 50% humidity. At the end of each week, 5 mL of the culture was transferred to 45 mL of fresh 0.2X MS media, and 5 mL of the culture was sampled for 16S rRNA community analysis.

Approximately 150 Brachypodium distachyon (Bd 21–3) seeds were dehusked and sterilized by following the procedures as described in the protocol (Novak et al., 2024). Seeds were immersed in 70% ethanol for 30 s, 50% bleach for 5 min, and then washed five times with autoclaved MQ water. The sterilized seeds were stored in sterile water at 4°C in the dark for 1 week. Subsequently, the seeds were germinated on semisolid plates containing 0.5X MS media with 0.4% w/v phytagel and placed in a plant growth chamber (Percival Scientific AR-41 L3, United States) set to 16 h of light, 24°C temperature, and 50% humidity for 3–5 days (Ingram et al., 2012; Li et al., 2019). For potting, sterilized plastic pots (3.9”D x 3.94” W x 3.15” H, Manufacturer: MiMiLai) lined with coffee filters and filled with autoclaved calcined clay (PROFILE Products LLC, United States) were used. The clay was saturated with 0.2X MS media before transplanting one seedling per pot. The weight of each pot, once filled with clay and plants and watered to clay saturation, was recorded as the saturation weight.

Individual SynCom isolates were cultivated in R2A media for 24–72 h until they reached log phase (Supplementary Figure S8). After cultivation, the isolates were checked for purity, washed three times, and resuspended in a 30 mM phosphate buffer. For each isolate suspension, cells were stained with the nucleic acid stain SYBR green and counted using a flow cytometer (AttuneNxT®, ThermoFisher Scientific) following the manufacturer’s instructions. Suspensions for individual isolates were adjusted to equal cell numbers, specifically 4.1 × 10⁷ cells, for the final inoculum used in subsequent in planta experiments.

This experiment comprised three conditions: Drought, Rewatered drought, and Salinity, to reflect natural environmental stresses. Each condition included two experimental sets; (1) SynCom-amended plants where 1 mL of 30 mM phosphate buffer containing equal cell numbers of each SynCom strain were inoculated at the base of plant shoot 3 days before inducing stress conditions, and (2) unamended plants where 1 mL of 30 mM phosphate buffer was inoculated at the base of plant shoot 3 days before inducing stress conditions. Each experimental set contained 7 plant replicates. After transplanting seedlings into pots, they were watered with 0.2X MS media to maintain 80% saturation weight for 4 days, after which they were inoculated with either SynCom or buffer, as explained earlier.

In addition to the three stress conditions, there was a control set with both SynCom-amended and unamended plants, maintained at 80% saturation weight by watering with 0.2X MS media. Stress induction began 3 days after SynCom inoculation. The ‘Drought’ condition involved maintaining plants at 40% saturation weight (Gombos et al., 2023) by watering with 0.2X MS media until they were harvested 21 days after stress induction. The ‘Rewatered drought’ condition involved 14 days of drought treatment, followed by restoring saturation weight to 80% for the last week before harvest. The ‘Salinity’ condition involved watering the plants at 80% saturation weight with 0.2X MS media containing 60 mM NaCl. All these plants were incubated in a plant growth chamber (Percival Scientific AR-41 L3, United States) set to 16 h of light, 24°C temperature, and 50% humidity. After 3 weeks of stress induction, plants were harvested by gently removing them from the pots for phenotypic measurements, including shoot wet weight, root length, leaf count, and shoot length (longest leaf). In addition, root tip and root base samples were aseptically harvested as 2 cm cuttings and suspended in 5 mL of 5 mM sodium pyrophosphate + 30 mM phosphate buffer. The sample was sonicated for 10 min and subsequently left on a benchtop for 5 min to allow soil particles to settle down. The supernatant was pipetted out and pelleted to collect microbial cells. DNA extracted from these samples were sent to Novogene, United States, for amplicon sequencing. Quantitative Insights Into Microbial Ecology (QIIME2; Bolyen et al., 2019) was used to process 16S rRNA amplicon data. Within QIIME2, DADA2 was employed for quality filtering, chimera checking, and paired-end read joining. The resulting sequences were trimmed to obtain V4 region (515F-806R), and taxonomic classification was performed using the SILVA database (Quast et al., 2013).

Indole-3-Acetic Acid (IAA) assay: The protocol for IAA assay was adapted from Gilbert et al. (2018). Briefly, the isolates were cultured in 5 mL R2A media, with and without 0.5 mg/mL L-tryptophan, for 3 days at 30°C. Following incubation, cultures were centrifuged at 14,000 rpm for 5 min, and 100 μL supernatant was transferred to a 96-well microplate. Subsequently, 200 μL of Salkowski reagent (2% of 0.5 M FeCl₃ in a 35% HClO₄ solution) was added to each well. IAA standards (0–10 μg/mL) were also prepared simultaneously. The microplate was incubated in the dark for 30 min, and absorbance values at OD₅₃₀ was recorded to quantify IAA production using the standard curve.

ACC-deaminase activity: The method for measuring 1-Aminocyclopropane-1-carboxylate (ACC) deaminase production in bacterial isolates was adapted from the colorimetric ninhydrin assay described by Li et al. (2011). Briefly, SynCom isolates were cultivated in 2 mL of DF salinity minimum media supplemented with 3 mM ACC for 24 h. Following incubation, 1 mL of the culture was centrifuged at 8,000 x g for 5 min. Subsequently, 100 μL of the resulting supernatant was diluted tenfold, and the ninhydrin assay, with ACC standards ranging from 0.5 mM to 0.005 mM, was used to quantify bacterial ACC consumption.

Phytate solubilization assay: Isolates were inoculated into 3 mL of filter-sterilized phytase-specific medium (PSM; Hosseinkhani and Hosseinkhani, 2009) supplemented with 0.5 mM of Sodium phytate (Fisher Scientific). The cultures were then incubated for 48 h at 30°C in a shaker incubator set at 80 rpm. Following incubation, the OD₆₀₀ was measured, and the cultures were centrifuged at 10,000 rpm for 15 min. The resulting supernatant was retained for a colorimetric assay using the QuantiChrom™ Phosphate Assay Kit (BioAssay Systems, CA, United States). Absorbance was measured at 620 nm and compared against the provided phosphate standards included with the Phosphate Assay Kit.

Siderophore production assay: The capability of siderophore production by the isolates was assessed using the CAS overlaying medium assay (Acharya et al., 2023a; Louden et al., 2011; Pérez-Miranda et al., 2007). Briefly, each isolate was cultured on R2A agar plates at 30°C for 3 days. Subsequently, blue CAS medium with 0.9% agar was overlaid onto the plates. The plates were then further incubated in the same conditions and monitored for color changes over an additional 4 days. The presence of a yellow halo surrounding colonies indicated positive siderophore production. As a positive control, Burkholderia sp. PA-E8, a known siderophore producer isolated at our laboratory, was included (Darling et al., 1998; Ong et al., 2016).

Biofilm formation assay: The crystal violet assay for quantifying biofilm formation was adapted from Haney et al. (2021). Briefly, each isolate was grown in R2A media, washed, and resuspended in a 30 mM phosphate buffer to a final OD₆₀₀ of 0.2. The cultures were inoculated into 96-well microtiter plates containing 180 μL of 0.2X MS media, at a 1:10 (v/v) ratio to achieve a final volume of 200 μL (initial OD₆₀₀ of 0.02). The plates were incubated statically at 30°C for 3 days. Post incubation, the liquid fraction was removed by inverting the plates, and each well was washed three times with MilliQ water and air-dried. Subsequently, 100 μL of a 0.1% crystal violet solution (0.1% v/v crystal violet, 1% v/v methanol, and 1% v/v isopropanol in MilliQ water) was added to each well, followed by a 30-min incubation at room temperature. After discarding the staining solution, wells were rinsed three times with MilliQ water. Biofilms were destained with 100 μL of a 30% acetic acid solution and incubated at room temperature for 30 min. OD₅₉₅ of the destaining solution was measured for biofilm quantification.

Gibberellin (GA) assay: The protocol for gibberellin quantification was adapted from Graham and Henderson (1961). Briefly, microbial isolates were cultured in R2A broth at 30°C for 48 h until they reached late-log phase. Cultures were centrifuged at 8,000 rpm for 5 min, and the supernatant was acidified to pH 1.5–2.0 using HCl. An equal volume of ethyl acetate was added, mixed thoroughly, and allowed to separate into organic and aqueous layers. The upper ethyl acetate layer, containing gibberellins, was collected and extracted twice. Combined extracts were vortexed to homogenize, and 50 μL was mixed with 750 μL phosphomolybdic acid reagent in 2 mL tubes. Samples were incubated in a boiling water bath for 1 h, cooled to room temperature, and 200 μL was transferred to a 96-well microplate for absorbance measurement at 780 nm. Gibberellin concentrations were determined using a standard curve prepared from 0 to 10 mg/mL gibberellic acid in absolute alcohol, processed alongside samples.

Acid and Alkaline Phosphatase assay: The protocol for assessing acid and alkaline phosphatase activity was adapted from Sinsabaugh (2001) and Allison (2007). Microbial isolates were cultivated in phytate minimal medium at 30°C until reaching mid-log phase. Cultures were then subjected to OD600 measurements and centrifuged at 8,000 rpm for 5 min. The supernatant was discarded, and the cell pellets were washed three times with, and subsequently resuspended in, 50 mM boric buffer (pH 9) for alkaline phosphatase assays or 50 mM acetate buffer (pH 5) for acid phosphatase assays. For each assay, 750 μL of the cell suspension was mixed with 750 μL of 5 mM p-nitrophenyl phosphate (pNPP) prepared in the corresponding buffer. A substrate control was prepared by combining 750 μL of 5 mM pNPP with 750 μL of buffer alone. All reactions, including isolate samples and controls, were incubated at room temperature for 45 min. Following incubation, samples were centrifuged at 10,000 rpm for 1–2 min. A 750 μL aliquot of the resulting supernatant was transferred to a fresh tube, and 75 μL of 1.0 M NaOH was added to terminate the reaction and facilitate color development. A 300 μL portion of this solution was then transferred to a 96-well microplate, and absorbance was measured at 410 nm. Phosphatase activity was quantified by comparing absorbance values to a standard curve of p-nitrophenol prepared in the corresponding buffer.

Carbon source utilization assay: Each isolate’s growth was assessed on 18 Brachypodium root-exudate carbon sources that are significantly enriched during drought stress (Skalska et al., 2021; Fisher et al., 2016; Ahkami et al., 2019; Supplementary Table S2). Cultures were incubated aerobically in 0.2 × MS medium supplemented with 5 mM of each carbon source at 30°C in the dark, with shaking, in biological triplicate. Cell density (OD600) was recorded at 0, 24, 48, 92, 120, 144, and 168 h. The growth was considered positive when the OD600 increased by more than 0.07 relative to the 0-h measurement.

Network analysis and SynCom member interactions: To understand the interspecies interactions from 0.1X R2A enrichments which have the highest diversity and most isolates, the top 50 most abundant ASVs among the core ASVs (present in >75% samples from the 4 generations) from this enrichment were selected. Their abundance matrix across all samples was analyzed using “NetCoMi” package (Peschel et al., 2021) in R. Pearson correlation coefficients greater than 0.3 and Student’s t-test results with p-values less than 0.05 were used to generate a sparse matrix for the network analysis. In the correlation network, each node represents an individual ASV, with different colors representing the corresponding modules. The edges connecting the nodes indicate a strong and significant correlation between the ASVs. The clustering in this network was used to group ASVs into modules that are densely interconnected internally but have sparse connections with other modules.

SynCom strain abundances in the rhizosphere under drought and salinity stress: To identify the changes in individual SynCom member’s abundances under stress, natural log fold differences between the control versus drought, rewatered drought, and salinity conditions were assessed using Analysis of Compositions of Microbiomes with Bias Correction (R package “ANCOMBC”; Lin and Peddada, 2020; Martins et al., 2023). False discovery rates were controlled using the Benjamini-Hochberg method.

Rhizosphere functional redundancy of SynCom members in stress conditions: To test whether the plants under stress modulate rhizosphere microbiome toward similar functions, functional redundancy (FR) across different environments (de Bello et al., 2021) estimated and compared using the R package “SYNCSA” (Debastiani and Pillar, 2012).

In our previous research, we had developed a number of reduced-complexity microbial enrichments from the rhizosphere of Brachypodium distachyon (Chen et al., 2024). We isolated 175 bacterial strains from those enrichments, which we classified by their 16S rRNA genes using the SILVA database (Pruesse et al., 2012). These isolates represented six phyla: Alphaproteobacteria (15), Gammaproteobacteria (124), Actinobacteria (15), Firmicutes (13), Bacteroidetes (7), and Acidobacteria (1).

Out of those isolates, we downselected 15 isolates that showed a 100% match in their 16S rRNA gene sequences with both top abundant ASVs from these enrichments (Chen et al., 2024) and with sequences from their rhizosphere origins (Acharya et al., 2023b). To build a SynCom that is truly representative of the Brachypodium rhizosphere microbiome, we considered the following criteria: (1) relative abundance of different microbial taxa within the enrichments, (2) representation of phylogenetic diversity among the most abundant microbes, and (3) network interactions among the taxa. Together, the ASVs corresponding to these 15 isolates made up 79.93% of the total reads in the enrichments. We also conducted network analyses on the top 50 ASVs to identify key interactions, which revealed six distinct modules (Figure 1) and a total of 38 positive and 17 negative correlations—24 positive and 9 negative of which occurred among our SynCom members. For the ASVs that matched with selected isolates, six were in Module 1, five in Module 3, three in Module 4, and two each in Modules 5 and 6, which cover most network modules. Some isolates, like Mesorhizobium sp. RCC-202 (degree = 8) and Leifsonia sp. RCC-180 (degree = 5), showed high network degrees (number of significant correlations connecting to the node), suggesting they play important roles in maintaining network structure. All but one strain, Bosea sp. RCC-152.1, had positive connections with at least one other strain across different modules. Informed by all these different analyses, we assembled a SynCom with these 15 isolates that represent taxonomically diverse 5 distinct bacterial phyla (Proteobacteria, Bacteroidota, Acidobacteriota, Actinobacteriota and Firmicutes; Figure 2).

Genomic analysis of nearly complete genomes for all 15 isolates showed that they exhibited a wide range of genome sizes (3.7 Mbp − 8 Mbp), GC content (35–69%), and numbers of genes (3343–7,740). The key general genome features of the SynCom isolates are outlined in Table S1. Most SynCom isolates encoded either a complete set of genes or the essential genes involved in pathways known to be associated with PGP traits (Olanrewaju et al., 2021; Bruto et al., 2014; Gupta et al., 2014; Bhattacharyya and Jha, 2012), potentially contributing to overall plant growth (Table 1). Some of the important PGP traits encoded in the genomes of multiple isolates included 1-aminocyclopropane-1-carboxylate (ACC) deaminase production (9 isolates), Indole-3-acetic acid (IAA) for auxin synthesis (2 isolates), synthesis of antimicrobial compound, i.e., gamma-aminobutyric acid (GABA; 12 isolates), genes associated with plant hormones like amidases (13 isolates), synthesis of volatile organic compounds (VOCs) i.e., 2,3 butanediol and acetoin (14 isolates), trehalose synthesis (13 isolates) siderophore production or transportation (11 isolates), and EPS production (15 isolates; Table 1).

We tested eight PGP traits essential for plant’s growth by conducting phenotypic assays. These include IAA production, ACC deaminase activity, phytate solubilization, siderophore production, biofilm formation, Gibberellin production, and Acid and Alkaline Phosphatase assay (Table 2). Six SynCom isolates produced IAA in the presence of L-tryptophan, with concentrations ranging from 0.25 to 23 μg/mL. The highest IAA levels (~23 μg/mL) were observed in Enterobacter sp. RCC-40 and Mucilaginibacter sp. RCC-168. All SynCom isolates exhibited ACC deaminase activity, with nine strains degrading more than 50% of the supplied 3 mM ACC. Siderophore production was detected in seven isolates, indicated by the formation of a yellow halo in the o-CAS assay (Table 2). Ability to form biofilms, as assessed by crystal violet assay, was significant (p-value < 0.05) by Student’s t-test in nine isolates. All the isolates produced Gibberellins, with Bacillus sp. RCC-6.1 and Enterobacter sp. RCC-40 showing the highest levels. Five isolates, including Luteibacter sp. RCC-6.2, Terriglobus sp. RCC-193, Mucilaginibacter sp. RCC-168, Leifsonia sp. RCC-180, and Rhizobium sp. RCC-161.2, demonstrated phytate solubilization with phosphorus release in the range of 0.0003–0.006 mg/dL. Six isolates produced acid phosphatase, with Bosea sp. RCC-152.1 and Enterobacter sp. RCC-40 exhibited the highest activity (~600 uM), while the remaining isolates produced <200 uM. Lastly, five isolates produced alkaline phosphatase, with Enterobacter sp. RCC-40 and Arthrobacter sp. RCC-34 displaying the highest concentrations (146–375 uM; Table 2). Notably, Mucilaginibacter sp. RCC-168 tested positive for all eight PGP traits. We also evaluated each strain’s growth on 18 drought-enriched Brachypodium exudates (Skalska et al., 2021; Fisher et al., 2016; Ahkami et al., 2019). Eleven of these compounds were metabolized by every isolate (Supplementary Figure S1). Paraburkholderia sp. RCC-158 grew on all 18 substrates and exhibited the most rapid growth as indicated by OD600 increase (Supplementary Figure S2).

Since all strains were isolated from the plant rhizosphere, we hypothesized that the presence of the plant host would positively impact SynCom growth and stability. To test this, we first examined SynCom growth by combining 15 strains in equal cell volumes and inoculating them into 0.2X MS media, which we generally used to water the plants. Samples were collected weekly, and community composition assessed through 16S rRNA (V3-V4 region) sequencing (Figure 3). The results showed that six of the 15 strains were undetectable or had a relative abundance below 0.1% after 3 weeks. Notably, the in vitro environment did not support the growth of Gram-positive strains, as three (Arthrobacter sp. RCC-34, Leifsonia sp. RCC-180, and Bacillus sp. RCC-6.1) of the four were either absent or present at very low abundance (<0.1%) when sampled. Conversely, members such as MA-N2 sp. RCC-150, Paraburkholderia sp. RCC-158 and Chitinophaga sp. RCC-12, dominated the culture at the end of week 3 with an average relative abundance of 60, 26 and 8%, respectively (Figure 3).

Next, we evaluated the growth and persistence of SynCom members in the presence of Brachypodium. After 3 weeks of growth in planta, the relative abundance of strains was more evenly distributed (Figure 4B). Notably, Gram-positive strains such as Arthrobacter sp. RCC-34 (1.4%) and Leifsonia sp. RCC-180 (0.6%) showed substantial increases in relative abundance compared to in vitro conditions. In contrast, Bacillus sp. RCC-6.1, showed a low relative abundance (<0.1%). Additionally, 11 of the 15 strains displayed higher median relative abundances when grown with the plant than without (Supplementary Figure S3).

We assessed the impact of the SynCom on plant phenotype under drought and salinity stress, two common environmental stressors that plants encounter. No significant differences were observed between SynCom-amended and unamended plants with regards to shoot height, shoot weight, root length, or leaf number under drought stress (Figure 4A; Supplementary Figure S7). However, SynCom-amended plants showed significant improvements across all these above measured parameters (t-test, Figure 4A) on rewatering after drought. Under salinity stress, SynCom-amended plants exhibited a significant increase in shoot height compared to unamended plants (Figure 5A).

When examining SynCom persistence under the above stress conditions, we observed that all strains were persistent throughout drought, rewatered drought, and salinity stress after 3 weeks (Figures 4B, 5B). However, significant differences in relative abundance of the individual strains were observed. Under drought, Paraburkholderia sp. RCC-158 (lfc = 1.29), Rhizobium sp. RCC-161.2 (lfc = 1.91), Mesorhizobium sp. RCC-202 (lfc = 1.82), Terriglobus sp. RCC-193 (lfc = 3.09), and Chitinophaga sp. RCC-12 (lfc = 1.96) showed significant (p < 0.05, two-sided Chi-square test) decreases in log fold change (lfc) relative abundance (Figure 4C) compared to the no-stress control. In rewatered drought conditions, Terriglobus sp. RCC-193 (lfc = 1.74) showed a significant decrease, while MA-N2 sp. RCC-150 (lfc = 1.4), Leifsonia sp. RCC-180 (lfc = 1.04), and Caballeronia sp. RCC-10 (lfc = 1.09) significantly increased (Figure 4D) relative to controls. No significant change in SynCom relative abundance was observed under salinity stress (Figure 5C).

Previous studies have highlighted that plants secrete specific compounds as root exudates to recruit select microbes, especially under stressful conditions (Vives-Peris et al., 2020). Given that exudation might vary between root tips and base (Dragišić Maksimović et al., 2021), we investigated the recruitment and spatial localization of SynCom strains on different parts of the root in the experiments above (drought and salinity stress). The SynCom was applied to the seedling’s growing root base for both these tests, and at the completion of the experiments, we analyzed the community at both the root tip and root base using 16S rRNA-based amplicon sequencing. Distinct colonization patterns emerged for certain strains. Under no stress, strains such as MA-N2 sp. RCC-150, Mucilaginibacter sp. RCC-168, Variovorax sp. RCC-210, and Arthrobacter sp. RCC-34 were significantly enriched (t-test) at the root base compared to their abundance in the inoculum and root tips (Figure 6, control sets). Under drought stress, however, Luteibacter sp. RCC-6.2 and Arthrobacter sp. RCC-34 showed increased abundance at the base, while MA-N2 sp. RCC-150 was significantly enriched at the root tip. In rewatered drought conditions, MA-N2 sp. RCC-150 and Leifsonia sp. RCC-180 were more abundant at the root tip. During salinity stress, MA-N2 sp. RCC-150 was the only strain found to have significantly increased abundance at the root tip. Overall, MA-N2 sp. RCC-150 displayed a notable preference for the root tip, especially under salt and drought stress conditions.