The soil and plant samples were initially collected from the coastal area of Parangipettai, Tamil Nadu, India. The S. portulacastrum was mostly found on the sandy shores of the backwaters (Figure 1). The collected halophyte was successfully established in the soil salinized by paper and pulp mill effluent irrigation at Karur, Tamil Nadu, India. The rhizospheric soil from two soil series, Thulukanur and Vannapatti soil series in the paper mill effluent irrigated area was collected. The samples were collected in sterile plastic bags and transferred to laboratory for further analysis. The organic matter was analyzed by the Walkley-Black method while Electrical Conductivity (EC) and pH were analyzed by the saturated paste extract method (Murtaza et al., 2017). The physico chemical properties are tabulated in Supplementary Table S3. Soil samples required for the study were collected from the soils of long-term treated paper and pulp mill effluent irrigated area located at 11° 01′24.9″ N and 77° 59′59″ E. Adequate amount of soil was shade dried, large debris was removed and subsequently 10 kg of soil was transferred to perforated pots for secondary evaluation. The pH and EC of the experimental soil were found to be 8.18 and 2.62 dS m⁻¹, respectively (Supplementary Table S3). The Exchangeable Sodium Percentage (ESP) of the soil was 13.54 per cent with an organic carbon content of 0.63 per cent.

To isolate rhizospheric bacteria, plant roots were gently shaken to remove the clumps of loosely adhering soil to the roots, leaving behind the root-firmly adhering soil particles (rhizospheric soil), which were then suspended and vortexed in 100 mL of sterile 1% NaCl solution. Thereafter, 10 folded serial dilutions (10×, 20×, 30×, 40×, and 50×) were prepared and 1 ml from each diluent was plated on nutrient agar (NA) medium (Supplementary Table S1) supplemented with 2% NaCl. The plates were incubated at 28°C and monitored for colony formation for up to 1 week (Fisher et al., 2007). For the isolation of endophytic bacteria, the roots of the collected samples were washed carefully under running tap water for 10 min to remove adhering soil particles. The roots were disinfected with 70% ethanol for 1 min, then rinsed three times with sterile distilled water. The roots were then surface sterilized with 3% sodium hypochlorite solution containing a few drops of Tween 20 R (Sigma-Aldrich, Steinheim, Germany) for 10 min followed by six rinses with sterile distilled water. To confirm root surface sterilization efficiency, an aliquot (100 mL) from the sixth wash solution was spread on NA medium and incubated at 28°C for 5 days. Thereafter, 1 g of the surface sterilized root tissue was macerated with a sterilized mortar and pestle in 10 mL distilled water and 1 mL from the tissue extracts and the serial diluents (10×, 20×, and 30×) were spread on NA medium supplemented with 2% NaCl. The plates were incubated at 28°C and monitored for up to 1 week for bacterial colony formation (Ramadoss et al., 2013).

The colonies of rhizospheric and endophytic bacterial isolates were examined morphologically for their shape, size, margin, elevation, appearance, texture, and pigmentation. In addition, cellular morphology, shape, Gram staining and biochemical characters were also examined (Sandhya and Ali, 2018; Supplementary Table S2). Colonies with distinct morphological characteristics were selected and purified by subculturing three times on NA media supplemented with 3% NaCl, before their storage in a 40% glycerol solution at −20°C till further use. Each sample was labeled representing the plant source (Sesuvium portulacastrum: SP) followed by the abbreviation of collection site (P: Parangipettai coastal soil; TV: TEWLIS area Vannapatti series and TT: TEWLIS area Thulukanur series), E in case of Endophytic, and the isolate number.

Initially, all bacterial isolates were screened for halotolerance using NA media supplemented with 2 and 3% NaCl (14.7 and 21.6 dS m⁻¹, respectively). The cultures that were found to grow in 3% were selected for further evaluation with growth curve experiment. The 48 h old cultures of the isolates from the broth were transferred with an equal quantity of inoculum (Optical density of 660 nm = 0.10) to 100 mL NA broth supplemented with 3, 5 and 7% NaCl (21.6 dS m⁻¹, 32.3 dS m⁻¹ and 41.1 dS m⁻¹, respectively). The OD 660 nm values were measured once in 4 h for all the isolates until the stationary phase is achieved. The OD 660 values were plotted against the time to obtain the growth curve under each level of NaCl.

Plant growth promoting traits of bacterial isolates were assessed for ammonia production, inorganic phosphate solubilization, siderophore production, IAA production and ACC deaminase activity. All assays were carried out in triplicates and the activity was assessed.

The bacterial isolates were confirmed by molecular analysis of the 16S rRNA gene for bacteria. DNA isolation from microbial samples was done using the EXpure Microbial DNA isolation kit developed by Bogar Bio Bee stores Pvt. Ltd. DNA concentrations were measured by Qubit fluorometer 3.0. The PCR amplification was done by adding 5 μL of isolated DNA in 25 μL of PCR reaction solution (1.5 μL of Forward Primer and Reverse Primer, 5 μL of deionized water, and 12 μL of Taq Master Mix). The DNA template is heated to 95°C for 2.5 min and cooled at 55°C for 30 s. Then it is heated to 72°C, the optimal temperature for DNA polymerization. The unincorporated primers and dNTPs from PCR products were removed by using Montage PCR Clean up kit (Millipore). Then the DNA sequencing was performed using an ABI PRISM^® Big Dye™ Terminator Cycle Sequencing Kit with AmpliTaq^® DNA polymerase (FS enzyme) (Applied Biosystems). Single-pass sequencing was performed using below 16 s rRNA universal primers. The samples were resuspended in distilled water and subjected to electrophoresis in an ABI 3730xl sequencer (Applied Biosystems). The 16 s rRNA sequence was blasted using NCBI blast similarity search tool. The phylogeny analysis with the closely related sequence of blast results was performed by multiple sequence alignment. The MUSCLE 3.7 was used for multiple alignments of sequences (Edgar, 2004). Poorly aligned positions and divergent regions were cured using the program G blocks 0.91b (Talavera and Castresana, 2007). Finally, the program PhyML 3.0 aLRT was used for phylogeny analysis and HKY85 as substitution model. The program Tree Dyn 198.3 was used for tree rendering (Dereeper et al., 2008).

The salt tolerant isolates were evaluated for their effect on germination and growth attributes of Vigna mungo L. under salinity (2% NaCl). An initial standardization experiment was carried out to fix the level of initial tolerance of Vigna mungo L. Accordingly, germination and growth of Vigna mungo L. under three different salinity was evaluated (0.5, 1, 1.5, 2, 3% NaCl). As per ISTA (International Seed Testing Association), the NaCl concentration at which the germination percentage falls below 65% should be selected to observe the impact of microbial inoculation. Hence 2% NaCl concentration was selected for the in vivo evaluation of isolates. The Vigna mungo L. seeds were surface sterilized in 0.1% sodium hypochlorite for 3 min and repeatedly washed with distilled water. After this, the 28 h old inoculum of all the isolates was individually used for seed priming. The NA broth without isolate was used as a control. The microbial inoculum with an OD of 1 at 660 nm was used for seed priming. After which the seeds were grown in germination sheets placed in 2% NaCl solution. The germination percentage, root length, shoot length and total dry weight were measured after 20 days. The growth vigor was calculated using Equation 3.

The culture compatibility among the bacterial isolates was assessed by cross streaking method using NA medium. The bacterial strain was streaked at one end of the plate followed by streaking the other bacterial strains perpendicular to it and incubated at 30°C for 24–48 h. The inhibition in growth between the cultures was noted and the compatible microbes were selected for further confirmation. The selected compatible strains were streaked in a triangular pattern so that all the streaks overlap each other as a confirmation test. Any two perpendicular steaks that showed inhibition were incompatible and the absence of inhibition proved that the cultures were compatible (Thomloudi et al., 2019).

The elite compatible microbial isolates were mass multiplied as pure cultures in the NA media for further formulation of the microbial consortium. The mass culturing was carried out in 250 mL Erlenmeyer flask, which serves as seed culture for further multiplication. The culture broth was autoclaved in 15 lbs for 20 min and after cooling the microbial isolates were inoculated as loopful cultures in their respective broths prepared. The inoculated flasks were maintained as shake flask cultures by incubating them at room temperature at 120 rpm for 2 days for mass production. The 48 h old cultures were serially diluted and plated to count the population in the broth. To attain a uniform population in the consortium, 5 mL inoculum was centrifuged at 10,000 rpm for 15 min at 10°C and resuspended in 10 mL of deionized water to obtain 4 × 10⁹ CFU (Jain and Srivastava, 2012). The OD 660 nm value was also recorded for the resuspended water used for consortium development. Talc was used as carrier material for the consortium due to its local availability and high Magnesium and Calcium content. Talc is a fine, light-weight powder that is easily soluble in water and has been shown to retain viable bio-inoculant propagules (Tripathi et al., 2015). A 100 g of talc was taken in autoclavable plastic bags and autoclaved. A 10 mL of the resuspended solution of each selected isolate was mixed to obtain the consortium. A total volume of 30 mL of the bacterial isolates was added to the carrier material and left for drying in shade (1 day), after which it was evaluated as a soil inoculant.

The potential of microbial consortium dosage on alleviation of salt stress in Vigna mungo L. was assessed through a pot culture experiment and the details of treatment are given in Table 1. The treated effluent with EC of 2.5 dS m⁻¹ was used for irrigation. The pots were irrigated at an interval of 5–7 days based on the field capacity of the soil (22.5 per cent) which was estimated using pressure plate apparatus (Obi, 1974). After the application of amendments and microbial consortium, five seeds were sown in each 3 kg pot and irrigated with treated effluent. Thinning was carried out after 15 days to maintain a uniform population of 3 plants per pot. The morphological, physiological and biochemical parameters were recorded at three growth stages viz., vegetative (25th day), flowering (45th day) and harvest stage. The height of the plant from the ground level to the tip of the main stem was measured and expressed in centimeters. The physiological and biochemical parameters including chlorophyll content, leaf free proline content and catalase were carried out using standard procedures.

The data on various characteristics studied during the investigation were statistically analyzed by the method given by Gomez and Gomez (1984) using SPSS Version 16.0. The critical difference was worked out at 5 per cent (0.05) probability levels.

A total of eight morphologically different bacterial strains were isolated from rhizospheric soil samples of Sesuvium portulacastrum collected from coastal areas of Parangipettai. Bacterial isolates were labeled as SPP 1 to SPP 8. The rhizosphere soil samples from paper and pulp mill effluent irrigated soil (TEWLIS area) were collected and 16 strains with varying morphology were isolated. Out of 16 isolates, 9 were from Thulukanur soil series (SPTT 1 – SPTT 9) and 7 were from Vannapatti soil series (SPTV 1 – SPTV 7). Similarly, eight endophytic bacterial isolates were obtained from roots of profusely grown S. portulacastrum cultivated in two soil series of paper and pulp mill effluent irrigated areas. Two strains from Thulukanur soil series were named SPTTE 1 and SPTTE 2. Similarly, five strains from Vannapatti soil series were isolated and named SPTVE 1 to SPTVE 5. The isolates were screened for salt tolerance in NA media supplemented with 3% NaCl, since identifying halotolerant PGPR with high tolerance potential is essential. The inoculated plates were observed for colony growth after 48 h of incubation. Among 31 bacterial isolates, profuse growth was observed in 9 isolates (SPP 2, SPP 5, SPP 6, SPTT 3, SPTT 7, SPPTT 8, SPTV 3, SPTVE 3, and SPTVE 4), whereas other strains failed to grow in NA media supplemented with 3% NaCl. Among the isolates, two (SPTVE 3 and SPTVE 4) were endophytes.

The selected 9 strains were characterized for cell morphology, Gram behavior, pigment production and motility (Supplementary Table S2). Out of 9 isolates, 6 tested positive in Gram test and the remaining were Gram negative. The isolate SPTT 8 was coccoid in shape and all others were rod shaped. The isolates SPP 2 and SPTT 8 exhibited orange and light green colonies, respectively, whereas no distinguishable colors were noticed in the other isolates. The orange or red pigmentation, due to carotenoids might help the bacteria against damaging UV radiation (Khaneja et al., 2010). Strains SPTT 3, SPTT 7, SPTVE 3, and SPTVE 4 were motile which has diffused zone of growth extended out from the line of inoculation. Both flagellar motility and citrate utilization are thought to be important in bacterial root colonization and maintenance (Weisskopf et al., 2011). Biochemical characters such as casein, starch hydrolysis, citrate, amylase, protease, urease, catalase and oxidase activities, indole production, nitrate reduction and Methyl Red test – Voges Proskauer (MR-VP) are given in Supplementary Table S2. Among these traits, protease, catalase and amylase were positive for all the isolates. MR-VP test was positive for all isolates except SPP 2, SPP 5 and SPTVE 3. Isolates SPP 6, SPTT 8 and SPTVE 3 tested positive for urease activity. The strains SPP 5, SPTT 3, SPTT 7 and SPTVE 3 had the ability to reduce nitrate to nitrite. All the isolates tested positive for catalase test which indicates all are aerobic bacteria and would neutralize hydrogen peroxide by producing catalase enzyme (Facklam and Elliott, 1995). Protease test was positive for all the isolates which indicates their ability to breakdown protein into amino acids.

The nine selected bacterial isolates were identified by 16S rRNA sequencing and blasting in the NCBI database. The isolate SPP 2 has a similarity (99.9%) with that of Metabacillus indicus and SPP 5 was 99.7% similar to Neobacillus niacini. The SPP 6 was 100% similar to Staphylococcus warneri and SPTT 3 was 99.8% similar to Bacillus velezensis. Similarly, SPTT 7, SPTT 8 and SPTV 3 were similar to Bacillus circulans, Kocuria rhizophila and Bacillus oleronius, respectively. The isolate SPTVE 3 had 99.9% similarity with Serratia marcescens and SPTVE 4 was 99.8% similar to Kosakonia radicincitanas. The phylogenetic tree of isolates used in microbial consortium II was obtained by the Neighbor Joining method with a boot strap value of 1,000 and depicted in Figure 2. The derived nucleotide sequence was submitted to NCBI database and the GenBank accession numbers are given in Table 2.

The salinity levels had a profound influence on the growth of isolates tested for tolerance (Figure 3). All the isolates had shown profound growth in control. The highest growth rate was observed in isolate SPTT 7. Including control, this isolate had grown well at both concentrations of NaCl (5 and 7%). In 7% NaCl, the stationary phase was attained at 52 h after inoculating. The isolates SPP 5 and SPTT 8 were unable to grow at 7% NaCl concentration and the growth rate slackened at 5% NaCl. M. indicus and B. oleronius had very slow growth rate at both the NaCl concentrations (5 and 7%) than the control. But the growth rate declined after 64 h in 5 and 7% NaCl. At 5% NaCl, S. marcescens had a slow and steady growth rate, while at 7% no growth was observed. In control, the lowest OD 660 nm value of 0.35 was recorded in the isolate B. velezensis at 60 h. The growth rate declined after 64 h at both NaCl concentrations (5 and 7%). The isolate S. warneri attained stationary phase after 40 and 60 h in control and 5% NaCl, respectively. High concentration of NaCl (7%) restricted the growth of the isolate. Isolates S. warneri, K. rhizophila and S. marcescens failed to grow at 7% NaCl, while they had a steady growth rate at 5%. Singh and Jha (2016b) also reported that S. marcescens grew well up to 6% NaCl. Similarly, K. rhizophila isolates had been reported to grow substantially in 5% salt medium, however, when salt concentration increased, the isolates showed a decreasing trend (Shi et al., 2021). B. circulans and B. oleronius had a shorter lag phase of 8 h than the other isolates, which indicated the potential of the microbe to grow under saline conditions (Finkel and Kolter, 1999). All the Bacillus genera (B. oleronius, B. circulans, N. niacini, and M. indicus) had long stationary phase that could be due to the synthesis of protective factors and adaptation of current environmental conditions at higher NaCl concentrations. In a study conducted by Duc et al. (2006), B. indicus tolerated higher levels of NaCl (8%), similarly Metabacillus indicus isolated in this study also tolerated 7% NaCl. Identifying halotolerant PGPR that could tolerate high levels of salinity is imperative for improving crop productivity under severe stress.

High salt concentration in the rhizospheric zone leads to osmotic stress and disrupts cell ion homeostasis due to the accumulation of Na⁺ and Cl⁻ ions (Paranychianakis and Chartzoulakis, 2005). This interferes with the uptake of essential nutrients and water thereby inhibiting plant growth. It affects shoot growth as well as xylem and root architecture. The root architecture in sense, the biomass, root network length and root length density are significantly altered, amplifying the impacts of salinity in the plants (Machado and Serralheiro, 2017). Plants cope with this by secretion of metabolites like proline and upregulation of potassium uptake that regulate homeostasis. The proline is synthesized from L-glutamate which requires ATP and NADH. Additionally, oxidative stress due to generation of ROS alters various biochemical pathways by inhibiting the binding of enzymes with substrate (Singh et al., 2022). Both osmotic stress and ionic stress result in oxidative damage to membrane components and cell organelles especially, chloroplasts. Owing to this the chlorophyll content in the plants under saline stress is reduced (Fatma et al., 2021). This inhibits photosynthetic activity and protein production, followed by inactivation of critical enzymes in the ATP synthesis (Singh et al., 2022). Nucleic acids, the structural component of proteins and DNA are impaired, leading to seizure of replication or transcription process (Hasanuzzaman et al., 2021). The chlorophyll degradation due to salinity is very high with the prolongation of stress (Hossain et al., 2017). In general, about 80% of the growth reduction at high salinity could be attributed to ethylene build up, resulting in reduction of leaf area, loss of chlorophyll content and decline in photosynthesis. The remaining 20% could be likely explained by a decrease in stomatal conductance and nutrient non-availability (Isayenkov and Maathuis, 2019). Ethylene is a plant hormone that regulates root hair, root growth, fruit ripening, leaf abscission, seed germination and ROS in the plant system (Fatma et al., 2021). Consequently, these changes in the plant’s metabolism could reduce the growth and yield of crops. Hence, the alleviation of above impacts with the aid of microbiome could help the crop to tolerate the salinity and increase yield even under high saline environments.

The results of in vitro experiment also revealed that when inoculated with the microbial isolates, the test plant’s growth improved significantly even under salt stress (Table 4). Significant difference in germination percentage, root length, shoot length and vigor index of Vigna mungo L. seeds was noted. Each isolate showed varying degrees of enhancement in crop growth attributes. Among the treatments, the highest germination percentage (88.5%) of Vigna mungo L. seeds was reported in S. marcescens inoculated treatment (T₁₀) and the lowest was in (T₂) uninoculated control (2% NaCl) with 65.3%. Treatments inoculated with N. niacini (T₄) and B. velezensis (T₆) recorded 83.4 and 85.2% germination, respectively. Though germination percentage was high when inoculated with S. marcescens (T₁₀), root length was observed to be high (6.8 cm) when inoculated with B. velezensis (T₆). The highest vigor index of 1819 was recorded with B. velezensis (T₆), followed by N. niacini (T₄) inoculation (1785). The uninoculated control (2% NaCl) had the lowest vigor index of 780 followed by S. warneri (T₅) inoculation (792). In an experiment conducted by Singh and Jha (2016b), S. marcescens (T₁₀) inoculation improved wheat plant growth under salinity stress (150–200 mM) and effectively reduced the suppression of plant growth due to salt stresses. Under salt stress (2% NaCl), the lengths of shoots and roots in uninoculated plants were severely reduced, whereas their lengths increased significantly in the presence of PGPR. The availability of higher auxin concentrations as IAA by these inoculants is likely to be the cause of plants’ increased root length (Rajkumar et al., 2005; Chakraborty et al., 2006). Shukla et al. (2012) also revealed that inoculation of halotolerant PGPR strains isolated from Salicornia brachiata improved the groundnut crop growth in hydroponics experiment with 100 mM NaCl than the control. The shoot length (14.6 cm) was observed to be higher in S. marcescens and B. velezensis inoculated plants, than in the uninoculated control (8.2 cm). Similarly, tomato seeds coated with B. velezensis strain had higher plant height and stem diameter at 100 mmol L⁻¹ NaCl than the control (Medeiros and Bettiol, 2021). In this study, K. radicincitans inoculated treatment recorded high shoot and root growth of 62.2 and 79%, respectively, than the uninoculated control. Similarly, Berger et al. (2013) also reported 80 and 50% increase in root and shoot growth of 5 weeks old tomato plants, emphasizing K. radicincitans significance as a strong PGPR for variety of crops.

The bacterial isolate in the microbial consortium must be able to proliferate in the presence of each other without hindering the growth and development of other microorganisms. The phytohormone develops a signaling network which mutually regulates several metabolic systems between the microorganisms (Patel et al., 2016). When isolates in a consortium have antagonistic relationships with each other, it makes the consortium unstable, and the intended function is not attained (Sarkar and Chourasia, 2017). Compatibility analysis of different PGPR cultures with each other in line streak assay revealed the isolates antagonistic activity against each other. The absence of an inhibition zone indicated that the isolates were compatible with each other. A zone of inhibition was observed when isolates B. circulans and B. oleronius were streaked indicating their high antagonistic activity. Isolates M. indicus, N. niacini, and S. marcescens were compatible with each other, since there was no inhibition zone observed between these three isolates. Similarly, B. velezensis, K. rhizophila, B. oleronius, and K. radicincitans had compatibility with each other. In secondary confirmation study, B. oleronius had shown inhibition zone with B. velezensis and K. radicincitans. In this study, B. circulans had antagonistic effect with all the other isolates. Abada et al. (2014) noted the presence of Permetin A, an antibiotic chemical in B. circulans culture filtrate that had activity against Gram negative bacteria. Hence, this could be the reason for incompatibility with other isolates. M. indicus, N. niacin and S. marcescens had synergism between them and were combined as Microbial consortium I. Similarly, Microbial consortium II was formed with B. velezensis, K. rhizophila, and K. radicincitans. These isolates were tested again for their compatibility by streaking them in triangular shape and no inhibition zone was observed between the isolates in both the consortia (Figure 4).

Several reports have shown that halotolerant PGPRs effectively improve the growth of various agricultural crops under salinity stress conditions (Qin et al., 2016; Singh and Jha, 2016a; Etesami and Noori, 2019; Orhan and Demirci, 2020). Mechanisms by which they impact growth are (i) triggering plant antioxidant defense by upregulating vital enzymes that scavenge excess ROS, (ii) alleviating nutrient deficiency by fixing atmospheric nitrogen and producing ammonia, solubilizing P, producing siderophores for Fe, Zn and Ln uptake, (iii) increasing the uptake of selected ions for maintaining a high K⁺/Na⁺ ratio (John and Lakshmanan, 2018), (iv) decreasing plant Na⁺ accumulation by excreting exopolysaccharide to bind cations (especially Na⁺) in roots and prevent their translocation to leaves, and (v) reduction in ROS production by counteracting ethylene synthesis (Figure 8). Probably, different mechanisms are involved in each interaction that leads to plant growth stimulation under saline conditions. The prevention of nucleic acid and chlorophyll damage by quenching the ROS plays a major role in improving the salt tolerance in the crops. Nevertheless, the variation in chloroplast structure and SOD activity among the different plants interferes with the PGPR ability in stress alleviation. Hence, a bacterial strain that stimulates the growth of a particular plant species in the presence of salt may not cause similar effect in other plants. Furthermore, many intricate mechanisms are involved when endophytes come into the picture (Santoyo et al., 2016; Khan et al., 2017). The endophyte, Burkholderia phytofirmans enhanced the growth of six switchgrass cultivars out of the eight that were tested (Khan et al., 2017). Inoculation with this strain was found to induce wide-spread changes in gene expression in the host plant, including transcription factors that are known to regulate the expression of some plant stress factor genes (Lara-Chavez et al., 2015). It is likely that changes in plant gene expression could also be induced by the halotolerant consortium used in this study to inoculate Vigna mungo L. Hence further investigation of the genomic interaction with the plants is in process. This would have a significant influence in identifying microbial inoculums that guarantees the successful cultivation of crops under variety of stressful environments.