Wastewater samples were collected from the Effluent Treatment Plant, Amrita Vishwa Vidyapeetham, Amritapuri campus, Kerala. Ammonium Mineral Salt (AMS) agar supplemented with 20 μM LaCl₃ and 8% methanol (Skovran and Martinez-Gomez, 2015) was used for isolation and culture. Amphotericin B (1.5 μg/mL) was used as an antifungal agent in the media. ASBT-KP1 is one among 11 priorly isolated bacterial strains from the wastewater collected from the effluent treatment plant (ETP) collection tank of the university campus of Amrita Vishwa Vidyapeetham, Amritapuri, Kerala, India. The collection tank receives sewage and sullage from the septic tanks of different campus departments. ASBT-KP1 was chosen for its maximum arsenic tolerance level when compared to others.

The isolate was characterized based on its morphological and physiological characteristics. The bacterial growth curve was studied as described by Anes et al. (2017). The growth optima of the strain ASBT-KP1 was studied at different pH (3–11) and temperatures (4–70°C) after 24 h incubation (Mukherjee et al., 2020). The sugar fermentation profile of the isolate was analyzed using the HiCarbo Kit (KB009A/KB009B1/KB009C) Hi-Media, India, as per the instructions given by the manufacturer. The Kirby-Bauer disk diffusion method was followed to test the antibiotic susceptibility of the isolate (Drew et al., 1972). The following antibiotics were used for the study; Ceftazidime (30 μg), Ceftazidime/clavulanic acid (30/10 μg), Ticarcillin (75 μg), Ciprofloxacin (5 μg), Levofloxacin (5 μg), Co-Trimoxazole (25 μg), Tobramycin (10 μg), Aztreonam (30 μg), Meropenem (10 μg), Colistin (10 μg), Imipenem (10 μg), Gentamicin (120 μg), Piperacillin (30 μg), Minocycline (30 μg), Amikacin (30 μg).

The genomic DNA (gDNA) of ASBT-KP1 was isolated using the protocol described in Sambrook and Russel, Fourth edition. The isolated DNA was used for whole genome shotgun sequencing and analysis using the Illumina HiSeq 2,500 Sequencing platform at DNAxperts Pvt. Ltd. (India). The gDNA sample was used to construct pair-end sequencing libraries using the Illumina Truseq protocol v3. The average fragment sizes for the pair-end libraries were 2 × 100/2 × 150 bp. The de novo assembly and annotation of the gDNA sample were carried out using the tools from PATRIC (Wattam et al., 2017, 2018) as per the user’s manual. The strain identification and annotation were carried out using RASTtk. Moreover, the predicted and annotated genes were compared manually with Klebsiella sp. D5A genome data (Liu et al., 2016), a known PGP bacteria for the presence of genes responsible for the PGP traits.

The BLASTN analysis¹ for identifying the closely related bacterial species was performed with the 16S rDNA sequence of ASBT-KP1 (Accession number – MT815532.1). The known and closely related bacterial species of ASBT-KP1 and known PGP K. pneumoniae strains were first aligned using ClustalW with the Gap opening penalty and Gap extension penalty for Pairwise and multiple alignments set at 10.00 and 1.00, respectively. The aligned sequence was then used for phylogenetic tree construction using the Neighbour-Joining method (Saitou and Nei, 1987) with the Partial deletion value set at 80%. The evolutionary distances were computed using the maximum composite likelihood method (bootstrap values, as a percentage of 1,000 replicates) (Tamura et al., 2004) using the MEGA11 software (Tamura et al., 2021). Erwinia persicina strain LMG 11254 was used as the outgroup.

Plant Growth Promoting traits, including Indole Acetic Acid (IAA) production, siderophore production, precursor 1-aminocyclopropane-1-carboxylate (ACC) deaminase enzyme production, phosphate solubilization, Hydrocyanic acid (HCN) production and production of extracellular ammonia for the strain ASBT-KP1 were analyzed. The ability of ASBT-KP1 to produce IAA was detected by the Salkowski method (Glickmann and Dessaux, 1995; Gang et al., 2019). The siderophore production was tested using a modified blue agar chrome azurol sulfonate (CAS) assay protocol by Schwyn and Neilands (Louden et al., 2011; Arora and Verma, 2017). Production of an orange-yellow zone around the inoculated bacterial colonies indicated a positive reaction. For measuring the ACC deaminase activity, the method described by the researchers Penrose and Glick was followed (Penrose and Glick, 2003; Mukherjee et al., 2020). Bradford method was used for estimating the protein concentration of the samples (Kruger, 2009). The phosphate solubilization ability of ASBT-KP1 was tested on Pikovskaya agar medium (M520, HiMedia, India) plates. The presence of a halo zone around the colony indicated a positive result. Phosphate solubilization was estimated using the modified phospho-molybdenum method described by the researchers (Wang Z. et al., 2017). The resultant blue color was measured at 700 nm. The ability of ASBT-KP1 to produce ammonia was tested in peptone water following the protocol of Ahmad et al. (2008). A change in color from yellow to brown with the precipitate formation on adding Nessler’s reagent indicated a positive result. The HCN production was studied as per the protocol (Indiragandhi et al., 2008). A color change from yellow to reddish brown indicated HCN production.

The salt tolerance for the isolate was studied by culturing it in R2-A broth supplemented with NaCl at different concentrations – 0, 0.1, 0.5, 1, 2, 5, 10, and 20% for 24 h at 37 ± 2°C. Absorbance at 600 nm was measured following incubation. Heavy metal tolerance of the isolate was studied in R2-A broth supplemented with various concentrations (1 μM–10 mM) of heavy metals. The heavy metals used were copper (II) (CuSO₄·5H₂O), arsenate (V) (Na₂HAsO₄. 7H₂O), arsenite (III) (NaHAsO₄) and mercury (II) (HgCl₂). Growth of ASBT-KP1 was estimated by serially diluting and plating the culture suspension onto LB plates to determine the colony count after incubation for 48 h at 37°C (Mukherjee et al., 2020). The resistance exhibited by ASBT-KP1 to Arsenite (III) and Arsenate (V) was studied in R2-A broth supplemented with increasing concentrations of As (III), from 0 to 100 mM and As (V), from 0 to 150 mM. The negative control for the experiment was the bacterial cultures without As (III) and As (V) treatment. The treated and untreated cells were incubated at 37°C for 48 h. The growth was measured at OD_600nm in a microplate reader (BioTek, United States) and serially diluted and plated onto LB plates to determine the colony count. The minimal inhibitory concentrations (MIC) of As (III) and As (V) were determined for ASBT-KP1.

Scanning electron micrographic images were used to analyze the morphological changes in the bacterial population in the presence of 1 mM of arsenate. Untreated bacterial cells were kept as a control. The samples for SEM were prepared as described by Chakravarty and Banerjee (2008). Following incubation for 48 h, the treated and control cells were harvested (7,000 × g, 5 min, 4°C) and washed 3 times with 0.1 M phosphate-buffered saline (PBS, pH 7.4). The cells were then fixed overnight with 2% glutaraldehyde at 4°C. The cells were again washed thrice with 0.1 M PBS and then dehydrated with alcohol concentrations of 10, 30, 50, 70, and 90% and absolute ethanol. Before imaging, the samples were placed on glass slides (1 cm × 1 cm) and sputter-coated with gold. The Jeol 6390LAv was used for the SEM imaging (Focardi et al., 2010; Mallick et al., 2018). An energy dispersive X-ray spectrometry (EDX) was carried out with an Oxford XMX N (Oxford Instruments, United Kingdom) probe to estimate the adsorption of As onto the cells.

The arsenic bioaccumulation capability of ASBT-KP1 was determined by ICP-MS analysis of the bacterial sample grown in 25 mL of R2-A broth supplemented with 1 mM of As (V) for 72 h. ASBT-KP1 culture was pelleted down and washed with sterile water, followed by a wash with 0.1 M of EDTA to remove any arsenic adsorbed to the surface of the cells. The sample was washed twice with sterile water and dried overnight at 70°C. The sample was weighed and digested with concentrated nitric acid before being analyzed in an iCAP RQ (Thermo Scientific) using Helium KED mode.

The isolate was studied for its ability to form biofilm using the quantitative biofilm assay described by the researchers (Gómez et al., 2016). On a sterile 24-well plate (Tarson, Korea), R2-A broth supplemented with arsenic (50 mM) and bacterial suspension of ASBT-KP1 of OD_600nm 0.5 ± 0.05 was added to each well with four replicates and incubated overnight at 37°C. The plates were washed thrice with 2 mL of sterile distilled water, after which 2 mL of methanol was added to each well to fix the biofilm for 15 min. After incubation, methanol was removed, and 2% (v/v) of crystal violet was added and stained for 5 min. The plate was placed under gently running tap water to remove excess stains. The stain from the adherent cells was released by adding 2 mL of 33% (v/v) glacial acetic acid. The absorbance in each well was measured at 595 nm in a microplate reader (BioTek, United States). R2-A broth devoid of culture was kept as the negative control. The negative control’s mean OD_600nm value was kept as the OD cut-off value (ODC). Based on the OD_600nm, the strain was classified as strong biofilm producers if their OD > ODC × 4. The strains are said to be moderate biofilm producers if 2 × ODC < OD ≤ 4 × ODC, weak biofilm producers if ODC < OD ≤ 2 × ODC, and non-biofilm producers if OD ≤ ODC (Borges et al., 2012).

To study the plant growth promotion effect of ASBT-KP1 on V. radiata, the seeds of V. radiata were procured from the local market in bulk to keep the sample uniform for all the experiments. The seeds and the resulting plants were identified as V. radiata and assigned an accession number SNCH 4518 by Dr. Kiranraj MS, Curator of Herbarium, Sree Narayana College Herbarium, Sree Narayana College, Kollam, Kerala. The study was conducted according to the local and national guidelines for the cultivation of plants. Surface sterilized seeds of V. radiata were placed in Petri plates with Murashige and Skoog medium (PT103, HiMedia) for germination at 25–27°C for 3 days in the dark. The seedlings were bacterized with ASBT-KP1 for 8 h and then planted in seedling trays. The soil for the study was collected from a previously wastewater-irrigated site in Kollam, Kerala, India.

The basic properties of the soil are as follows: pH, 5.5; total N, 0.14%; total organic carbon, 0.16%; available phosphate, 16.7 mg/kg; total soil As content was below the detection limit of 0.1 mg/kg (Supplementary Table S3). The soil was sieved and sterilized before being used for the study. To study the effect of arsenic toxicity, seedlings were grown in seedling trays filled with soil supplemented with 10 mg/kg of sodium arsenate, which was relevant to the As concentrations recorded in polluted soil environments (Majumder et al., 2013). The soil was left for proper acclimatization of arsenic for 1 week before use. The planted seedlings were rhizoinoculated with ASBT-KP1 (10⁸ CFU/mL); in the soil, the bacterial load was 10⁷ CFU/g. Soil, without culture on the seedlings, was kept as control. The plants were harvested after 21 days, and their morphological and biochemical growth parameters were analyzed. The morphological parameters studied were shoot length, root length and dry biomass. Biochemical parameter, chlorophyll (Chl-a. chl-b, and total chl.) content was measured. The arsenic uptake by V. radiata from the soil was determined by ICP-MS analysis of the digested plant samples.

The rhizospheric soil from ASBT-KP1 treated plants and untreated control was collected to study the ability of ASBT-KP1 to establish itself in the rhizospheric soil. The soil was serially diluted and cultured on nutrient-rich media, LB and Klebsiella spp. specific media, Klebsiella Blue agar (KBA) (Prasad et al., 2022). The systemic uptake of ASBT-KP1 was tested by inoculating 7-day-old V. radiata seedlings grown in unsterile soil with 2 mL of ASBT-KP1 (10⁸ CFU/mL). Uninoculated plants were kept as control. Six plants were randomly selected from each group and surface sterilized with 2% sodium hypochlorite supplemented with 0.1% Tween 20, followed by three washes with sterile distilled water (Sahu et al., 2022). The samples were macerated and plated onto LB and KBA media and incubated at 37°C for 24 h. Growth in KBA media indicates the presence of ASBT-KP1, owing to the specificity of the media.

Seven-day-old O. sativa Kym 2 – Bhagya variety (seeds procured from Onattukara Regional Agricultural Research Station, Kerala Agricultural University, Kayamkulam, Kerala, India) saplings were replanted onto non-sterile soil spiked with 10 mg/kg of sodium arsenate. Three sets of 12 saplings each were employed for test and control experiments. Rhizoinoculation of ASBT-KP1 was performed as described in section 2.11, involving the V. radiata experiment. The plants were harvested after 15 days, and their morphological and biochemical growth parameters were analyzed. The morphological parameters studied were shoot length, root length and dry biomass. The arsenic uptake by O. sativa from the soil was determined by ICP-MS analysis of the digested plant samples.

The ability of ASBT-KP1 to establish itself in the rhizospheric soil of O. sativa was studied as previously described in section 2.11, for V. radiata. The systemic uptake of ASBT-KP1 was tested by plating the surface sterilized parts of plants onto KBA media. Growth in KBA media indicates the presence of ASBT-KP1, owing to the specificity of the media.

ASBT-KP1 was assessed for its virulence using two different methods. Firstly, the string test was performed to check for hypervirulence, as described by the researchers (Hagiya et al., 2014). Secondly, Caenorhabditis elegans was used for studying the virulence of ASBT-KP1 (Salim et al., 2022). Caenorhabditis elegans strain N2 (wildtype strain) were grown in nematode growth media (NGM; Peptone 2.5 g/L, NaCl 2.9 g/L, Bacto-Agar 17 g/L, CaCl₂ 1 mM, cholesterol 5 μg/mL, KH₂PO₄ 25 mM, MgSO₄ 1 mM) having a lawn culture of E. coli OP50. The C. elegans larvae were synchronized by isolating the eggs from gravid adults using a sucrose density-gradient-based separation technique (Shaham, 2006). The isolated eggs were hatched overnight in M9 buffer supplemented with cholesterol (1 mg/mL). The L1-stage worms were resuspended in M9 buffer, and the test bacteria –ASBT-KP1 (washed with sterile distilled water and resuspended in M9 buffer at an OD of 0.1). Uninfected C. elegans with E. coli OP50 was used as a control. The survival of the L1 worms to adults for the next 5 days was observed and plotted as a survival curve using Kaplan and Meier survival plot using Graph Pad Prism (v9.0.0).

The data collected were analyzed statistically using Graph Pad Prism (V 9.0). All experiments were conducted in triplicates or otherwise specified. The statistical difference for the plant growth studies was assessed using the One-way ANOVA followed by Dunnett’s multiple comparisons test for multiple comparisons between the control and treated samples. The O. sativa growth studies were tested using Welch’s t-test to compare control and test samples. The data were represented as the mean ± standard deviation of replicates, and the significance level was studied at a p-value <0.05.

ASBT-KP1 is a non-motile bacillus of a size range of 1.2–2.3 μm long. The sugar utilization profile of ASBT-KP1 includes lactose, xylose, fructose, dextrose, trehalose, sucrose, l-arabinose, mannose, galactose, inositol, maltose, adonitol, and arabitol but not d-arabinose, sorbose, inulin, glycerol, and dulcitol (Supplementary Table S1). The generation time of the isolate is 29 min (Figure 1A). The organism grew in a pH and temperature range of 5–9 pH (Figure 1B) and 4–50°C, respectively (Figure 1C). ASBT-KP1 showed sensitivity to all the tested antibiotics except ceftazidime, ciprofloxacin and ticarcillin (Figure 1D). Based on the whole genome sequence analysis, ASBT-KP1 is identified as K. pneumoniae. Throughout the manuscript, ASBT-KP1 is used as the strain designation, implying that it is Klebsiella pneumoniae. The phylogenetic tree of ASBT-KP1 was constructed using 16S rRNA gene sequences of known and closely related bacterial species and those of PGP K. pneumoniae strains (Figure 1E) (Tamura et al., 2021). The evolutionary relatedness of ASBT-KP1 reiterates its potential as a PGPB.

The genome of K. pneumoniae ASBT-KP1 has a total of 5,533,283 bp with an average G + C content of 57.10% (Figure 2A). The genome has 5,564 coding sequences, of which 4,775 genes of the predicted CDS were assigned biological roles while 789 coding sequences were classified as proteins with hypothetical functions. ASBT-KP1 has five rRNA genes comprising two 5S rRNA, two 16S rRNA, and a 23S rRNA. Moreover, it has 84 tRNA genes representing 21 amino acids and one pseudo tRNA.

Supplementary Table S2 summarizes the genes in the ASBT-KP1 attributed to phosphate solubilization and uptake, Indole Acetic Acid production, siderophore synthesis, resistance to oxidative stress, and resistance to heavy metals like arsenic, copper, zinc, and cadmium. Phosphate solubilization genes include those that encode glucose dehydrogenase (GDH) and redox cofactor Pyrrolo-quinolone quinine (pqq) and pqqBDEF, responsible for the synthesis of gluconic acid and PstBACS and PhnCDE2E1.The strain encodes 8 copies of genes for siderophore receptors and genes responsible for the synthesis of siderophores viz., TonB-dependent receptors, IroN receptors, ferrichrome-iron receptor, TonB-dependent hemin ferrichrome receptor, TonB-dependent ferric enterobactin, colicins B, D receptors, entBF and entS. Additionally, the strain has 49 genes that encode for iron transport proteins. The presence of ABC transporter and associated genes indicate that the strain can heterologously obtain siderophores produced by other soil bacteria. Genes that confer resistance to heavy metals like arsenic (arsCBRD), copper (copDC and related genes), and zinc are present in the strain. The genome has an ars gene cluster that confers arsenic resistance through the production of arsenate reductase. From the circos plot (Figure 2B), it could be deduced that the genes for arsenical resistance operon repressor (arsR), arsenite permease (arsB) and arsenate reductase (arsC) exhibit strong synteny (contig 4, fig|1162296.25.peg.1694–1,696). These genes code for a transcription repressor, an expulsion pump, and a reductase enzyme essential in the arsenic bioremediation potential of ASBT-KP1. Genes that protect plants against phytopathogens are hosted by ASBT-KP1, such as phzF, involved in phenazine synthesis, which functions as an antibiotic. A gene which encodes chitinase enzyme, known to degrade the cell walls of pathogenic fungal and insect pests, is also present.

Additionally, the strain has gabR and associated genes involved in gamma-aminobutyric acid production. The genome also encoded proteins that protect the cell from oxidative stress: three superoxide dismutases, five glutathione-S-transferases, and six peroxidases. The antibiotic resistance profile of ASBT-KP1 indicates its sensitivity to most tested antibiotic classes (Figure 1E). This observation was confirmed by an in vitro antibiotic susceptibility test followed by validation using the ResFinder tool hosted at CGE (Supplementary Data Sheet 1). Additionally, none of the classic antibiotic-resistance point mutations indicative of functional ramR, acrR, ompK36, ompK35, ompK37, gyrA, gyrB, rpsL or parC genes are identified that accounts for the sensitivity profile of ASBT-KP1.

ASBT-KP1 produced 120 ± 5 μg/mL of IAA (Table 1). The strain showed an increase in the production of IAA in the presence of arsenite (III), 165 μg/mL, and arsenate (V), 270 ± 5 μg/mL. The strain solubilized the calcium phosphate in the medium with a distinct zone of clearance (Supplementary Figure S1A). Qualitatively, 210 ± 5 mM of phosphate was solubilized. ASBT-KP1 expressed ACC deaminase with a specific 4.04 nmol/mg/h activity. HCN production was observed with a color change in the media from yellow to darker yellow-orange (Supplementary Figure S1B).

The strain showed differential tolerance levels toward mercury, arsenate, arsenite and copper at different concentrations. ASBT-KP1 tolerated >75 mM of As (V) (Figure 3A), >25 mM of As (III) (Figure 3B), ~250 μM of Cu (II) (Figure 3C), and ~ 10 μM of HgCl₂ (Figure 3D). ASBT-KP1 showed a tolerance of up to 7.5% of NaCl (Figure 3E).

The minimum inhibitory concentration for the strain against arsenate (V) and arsenite (III) was 120 and 70 mM, respectively. ASBT-KP1 has a relatively reduced tolerance to arsenite (III) compared to arsenate (V) (Figures 4A,B).

Scanning electron micrograph revealed that the arsenic-treated cells showed changes in the morphology and size of the cells. The average size of untreated cells was 1.7 ± 0.5 μm (Figure 5A), while the treated cells shrunk in size with an average decrease of 0.79 ± 0.25 μm (Figure 5B). The outer membrane of the cells did not show any significant changes, while there was visible vacuolation due to As (V) stress (Figure 5B).

The surface of the treated and control cells was further analyzed using SEM–EDX to study the arsenic adsorption by the bacterial cells. The surface of neither treated nor control cells showed any peak corresponding to arsenic (Figures 5C,D), indicating the absence of arsenic adsorption onto the cell surface. The buffer solution, glass slide and coating material contributed to the few unrelated peaks obtained in the analysis. The bioaccumulation potential was determined using ICP-MS analysis of the cell pellet. The analysis showed that the strain accumulated 30 ± 2 μg/g of bacterial dry mass when incubated in 1 mM of arsenate (V) for 72 h.

The microtiter plate assay for biofilm formation showed that ASBT-KP1 was a strong biofilm producer with an OD_600nm of 1.21, 14 times greater than the OD cut-off value (Borges et al., 2012). The isolate also formed biofilm in the presence of arsenate (Supplementary Figure S2).

The pH of the soil, total suspended solids, organic carbon, nitrogen, total phosphate, and arsenic levels were measured (Supplementary Table S3) to characterize the soil used for the study. The soil pH was slightly acidic (pH 5.5), and the total organic carbon was 0.78%. The plant growth study revealed that the mean dry plant weight (26.34 ± 0.402 mg) of plants inoculated with ASBT-KP1 (Figure 6D) was significantly higher than (p < 0.001) than the uninoculated plants (23.31 ± 0.257 mg). Similarly, the chlorophyll content of the plants inoculated with ASBT-KP1 (8.81 ± 0.125 μg/g) was significantly increased than the uninoculated plants (2.74 ± 0.089 μg/g) (Figure 6E). Statistical analysis of growth, in terms of mean shoot and root length, revealed that the inoculated V. radiata plants exhibited a more significant growth (9.6 ± 0.506 cm) when compared to the uninoculated plants (12.0 ± 1.40 cm) (p < 0.001). In the presence of Arsenic, V. radiata showed an increased mean shoot length (11.7 ± 1.32 cm) mean dry biomass (28.6 ± 2.24 mg) and chlorophyll content (10.6 ± 0.59 μg/g) compared to those without bacterial inoculation (8.37 ± 1.33 cm; 20.9 ± 2.77 mg; 7.77 ± 1.57 μg/g respectively) (Figures 6B–E), (p < 0.001). It is evident from the results that the strain protects the plants from the toxic effects of arsenic and improves growth compared to the uninoculated control. Even under normal physicochemical conditions, i.e., in the arsenic-free condition, ASBT-KP1 inoculated plants showed significant improvement in growth. The colonization of the ASBT-KP1 in the rhizospheric environment of the V. radiata plants helped significantly decrease the plants’ arsenic uptake by 91% (Figure 6F).

The bacterial load in the test soil was higher (5.5 × 10⁶ CFU/mL) in comparison to that in the control soil (3.5 × 10⁶ CFU/mL) when quantified in nutrient-rich media. ASBT-KP1 was effectively established after its introduction since its presence was detected (1.4 × 10⁵ CFU/mL) after 8 days of incubation. Additionally, to test the infiltration of ASBT-KP1 into the various parts, the plant’s root, stem and leaves were tested for its presence in both LB and KBA. Except in roots (8.2 × 10³ CFU/mL), none of the parts showed the presence of the strain. However, other bacterial endophytes were uniformly present in both the control (root – 1.93 × 10⁵ CFU/mL; stem – 1.33 × 10⁵ CFU/mL; leaves – 8.67 × 10⁴ CFU/mL) and test (root – 2.4 × 10⁵ CFU/mL; stem – 1.4 × 10⁵ CFU/mL; leaves – 8.67 × 10⁴ CFU/mL).

The study on O. sativa (Supplementary Figure S3) revealed that the mean dry plant weight (42.8 ± 6.31 mg) of plants inoculated with ASBT-KP1 in the presence of arsenate was higher (Supplementary Figure S4D) than the uninoculated plants (37.6 ± 6.06 mg). The analysis of growth in terms of mean shoot and root length, revealed that the inoculated O. sativa exhibited a significantly higher shoot and root length (19.9 ± 2.51 cm; 6.83 ± 1.46 cm) (Supplementary Figures S4A,B) when compared to uninoculated plants (17 ± 2.08 cm; 5.38 ± 1.28 cm) (p < 0.05). Thus, from the results, we can ascertain the ability of ASBT-KP1 to improve the growth of the O. sativa plants under arsenic-stressed conditions. By colonizing the roots of O. sativa, ASBT-KP1 significantly reduced the plants’ arsenic uptake by 74.19% (Supplementary Figure S4E).

The soil inoculated with ASBT-KP1 showed a higher total bacterial load (1.11 × 10⁶ CFU/mL) than untreated soil (9.5 × 10⁵ CFU/mL) when cultured in a nutrient-rich media. The test soil had 5.9 × 10⁴ CFU/mL of ASBT-KP1 indicating its ability to establish itself in the rhizosphere of the O. sativa plants.

The negative result for the string test indicated a reduced virulent nature of ASBT-KP1. The Caenorhabditis elegans infected with ASBT-KP1 5 days post-infection were viable and healthy compared to Escherichia coli OP50-treated control (Figure 7A). The worms in the ASBT-KP1 infected wells showed active pharyngeal pumping and movement and were reproductively active and laid eggs (Figure 7C). The worms that were scored dead appeared rigid (Figure 7D) and did not show any movement compared to the E. coli OP50-treated worms (Figure 7B). Therefore, the strain is non-pathogenic in the surrogate worm model.