Microbial inoculation in rice regulates antioxidative reactions and defense related genes to mitigate drought stress

Microbial inoculation in drought challenged rice triggered multipronged steps at enzymatic, non-enzymatic and gene expression level. These multifarious modulations in plants were related to stress tolerance mechanisms. Drought suppressed growth of rice plants but inoculation with Trichoderma, Pseudomonas and their combination minimized the impact of watering regime. Induced PAL gene expression and enzyme activity due to microbial inoculation led to increased accumulation of polyphenolics in plants. Enhanced antioxidant concentration of polyphenolics from microbe inoculated and drought challenged plants showed substantially high values of DPPH, ABTS, Fe-ion reducing power and Fe-ion chelation activity, which established the role of polyphenolic extract as free radical scavengers. Activation of superoxide dismutase that catalyzes superoxide (O2−) and leads to the accumulation of H2O2 was linked with the hypersensitive cell death response in leaves. Microbial inoculation in plants enhanced activity of peroxidase, ascorbate peroxidase, glutathione peroxidase and glutathione reductase enzymes. This has further contributed in reducing ROS burden in plants. Genes of key metabolic pathways including phenylpropanoid (PAL), superoxide dismutation (SODs), H2O2 peroxidation (APX, PO) and oxidative defense response (CAT) were over-expressed due to microbial inoculation. Enhanced expression of OSPiP linked to less-water permeability, drought-adaptation gene DHN and dehydration related stress inducible DREB gene in rice inoculated with microbial inoculants after drought challenge was also reported. The impact of Pseudomonas on gene expression was consistently remained the most prominent. These findings suggested that microbial inoculation directly caused over-expression of genes linked with defense processes in plants challenged with drought stress. Enhanced enzymatic and non-enzymatic antioxidant reactions that helped in minimizing antioxidative load, were the repercussions of enhanced gene expression in microbe inoculated plants. These mechanisms contributed strongly towards stress mitigation. The study demonstrated that microbial inoculants were successful in improving intrinsic biochemical and molecular capabilities of rice plants under stress. Results encouraged us to advocate that the practice of growing plants with microbial inoculants may find strategic place in raising crops under abiotic stressed environments.

photorespiration, photosynthesis and mitochondrial respiration lead to produce excessive ROS 5 that disturbs intrinsic cellular homeostasis 6 . Environmental stresses also trigger activity of monoamine oxidase (MAO), xanthine oxidase (XOD) and NADPH oxidase that balance production and accumulation of ROS 7 . The consequences are observed in terms of negative cellular metabolic functions that damage nucleic acid, protein, lipid and carbohydrate metabolism 2 .
Plants are evolved with a sophisticated system to overcome ROS burden within the cells through prominent antioxidative defense mechanisms 8 . Enzymatic antioxidative mechanisms include regulation of the enzymes like superoxide dismutase, catalase, peroxidase, glutathione reductase, glutathione S-transferase and guaiacol peroxidase. These enzymes prevent or repair the oxidative damage caused due to disrupted cellular homoeostasis under stress conditions 5 . Cells also synthesize diverse antioxidant molecules that regulate signal pathways in redox mechanisms to overcome oxidative damage 4 . Increased production of antioxidative enzymes like SOD, POD, CAT, GPX and GST 9 and the accumulation of antioxidant compounds such as carotenoids 10 and phenylpropanoids 11 successfully help plants reduce their load of ROS within the cells. These processes cumulatively help plants mitigate burden of oxidative mechanisms while maintaining their growth and development under stressful conditions.
Among various devastating environmental stresses for plants, drought conditions, either moderate to intense or short to prolonged, have remained a challenge for crop productivity 12 . Drought adaptation, avoidance and/ or mitigation strategies in crop plants lie with their intrinsic metabolic and molecular mechanisms which, when triggered by environmental stimulus strengthen plant growth, development and productivity 13 . Beneficial microbial interactions with plants either under normal growth conditions or in stressful environment manifest diverse physiological, biochemical and molecular roles [14][15][16] . Microbial communities, the most natural inhabitants of the soils and the rhizosphere, the specific ecological niche associated with the root vicinity, tremendously influence plant growth and productivity 17,18 . Their interaction with the plant root system constitutes the most complex and intricate biological phenomenon that helps plant activate their adaptive capabilities against drought stress through induced defense mechanisms 19,20 . Plant growth promoting rhizobacteria (PGPR) colonize rhizosphere to promote growth and induce systemic drought tolerance 21,22 through phytohormone, epoxypolysaccharides and ACC deaminase production [23][24][25][26] . Plant responses to Trichoderma inoculation as a biocontrol agent are manifested by early escape of abiotic stresses through activation of antioxidant machinery 27,28 . Inoculation of T. harzianum helped plants alleviate water deficit in tomato 29 and rice 28 through enhanced activation of ascorbate and glutathione-related defense enzymes 30 . Cumulatively, microbe-plant interaction and the resultant metabolic changes are being realized as a real time stress tolerance strategy in the plants for their survival and sustainable productivity 31 .
Rice (Oryza sativa L.) is the most important crop that feeds almost half of the world's population 32 . Being a crop of tropical and subtropical origin, rice is usually sensitive to abiotic stresses, especially to drought conditions 33 . Water deficit is amongst the major limiting factors to produce rice in many parts of the world 34 . High sensitivity to drought and water deficit poses serious threat towards enhanced productivity of this crop 35 . Microbial communities are the dominant natural inhabitants of the plant rhizosphere 36,37 including rice crop 38,39 . Their colonization and interaction with the rice roots impart beneficial plant growth promotion and abiotic stress mitigation impacts 40,41 . We demonstrated that the individual and combined inoculation of rice with Pseudomonas fluorescens and Trichoderma asperellum (T42) have contributed to strengthen intrinsic mechanisms in rice plants, thereby offering protective support against drought. Enzymatic and non-enzymatic antioxidant reactions in plants grown with microbial inoculation under non-drought and drought conditions were improved. The expression of defence-related genes that helped plants regulate ROS as key steps in microbe-mediated stress mitigation processes was explored. The study reveals that growing plants under microbe-inoculated conditions leads to modulate intrinsic biochemical and molecular mechanisms to help plants mitigate drought conditions. The observations warrant microbial inoculation as an efficient stress mitigation strategy for rice crop challenged with drought stress in the fields.

Materials and Methods
Seeds, microbial inoculants and experimental conditions. Seeds of rice variety Pusa Basmati (PB) 1612 were obtained from the seed bank of ICAR-Indian Institute of Seed Science, Mau, India. Rhizosphere compatible bioagents namely Pseudomonas fluorescens (Pf) (OKC; Genbank accession No. JN128891) and Trichoderma asperellum (Th) (T42; GenBank accession No. JN128894) were obtained from the Department of Mycology and Plant Pathology, Institute of Agricultural Sciences, Banaras Hindu University, Varanasi, India. Rice seeds were treated with both the cultures as described by Patel et al. 42 . For seed treatment, spore suspension of Th (spore count 1.3 × 10 8 ml −1 ) and cell suspension of Pf (1.2 × 10 8 cells ml −1 , optical density equivalent to 0.39) was prepared in 0.5% sterilized carboxymethylcellulose (CMC). For combined application, equal proportion of fungal spores and bacterial cell suspension was mixed together and applied. Rice seeds (variety PB 1612) were surface sterilized with 0.1% HgCl 2 solution for 2 min followed by washing thrice with sterilized distilled water. Dried seeds were coated with the inoculant suspension (individual and in combination) and kept for 3 h in air under sterilized conditions. Microbe-coated seeds were sown in earthen pots (10 inch diameter) containing 5.4 kg sterilized soil mixed with 20% vermicompost. Pots were kept in well-ventilated glasshouse throughout the Kharif season of 2017 from mid-June to November. Temperature ranged from 16.4 to 31.5 (min) to 30.1 to 39.2 °C (max) with gradual decrease as the plant development approached maturity. Regular watering was applied prior to flowering stage, before the onset of which, 7 days of continuous drought was given to one set of pots sown with the microbe-inoculated and non-inoculated (NI) rice seeds. All the plants were harvested after completion of drought period and leaves were collected for further experimentation.
Quantification of H 2 O 2 . Leaf samples (0.1 g) from each treatment were homogenized in 2.0 ml 0.1% (w/v) trichloroacetic acid (TCA) and centrifuged (12,000 g, 15 min). The supernatant (0.5 ml) was added with 10 mM phosphate buffer (pH 7.0). Afterwards, potassium iodide solution (1 M, 1 ml) was added following incubation for 5 min. The oxidation product formed was examined at 390 nm 44 . The concentration of H 2 O 2 formed was determined as nMol H 2 O 2 g −1 fresh weight (FW).
In situ examination of cell death. In situ cell death determination was carried out by treating plant leaves with 0.1% Evans blue solution. After 15 min, leaves were dipped in 95% boiling ethanol (30 min) for depigmentation. Necrotic spots were identified as indigo blue lesions at the leaf surface 45 .
Determination of total polyphenolic content (TPC). TPC was determined following the method of Zheng and Shetty 46 with modifications. Leaf tissues (0.1 g) were macerated in 5 ml water:methanol (1:1, v/v) at 4°C and extracted for 48 h. Homogenized samples were centrifuged at 15000 g (10 min). Polyphenolic content was quantified using Folin-Ciocalteau reagent. The extract (1 ml) was mixed with water:methanol (1:1, 1 ml, v/v), distilled water (3 ml) and Folin-Ciocalteau regent (0.5 ml) followed by thorough mixing. The reaction mixture containing 5% sodium carbonate (1 ml) was kept for 30 min and examined at 725 nm. TPC was calculated as mg gallic acid equivalents (GAE) per g FW.
Quantitative determination of enzymes. One g of fresh rice leaves were washed with the sterilized distilled water and macerated with 5 ml phosphate buffer (pH 7.8) in ice cooled pestle-mortar kept at 4 °C. The extract was centrifuged at 15,000 rpm for 15 min at 4 °C and used for enzymatic assays.
Peroxidase (PO). PO (EC 1.11.1.7) was estimated in the reaction mixture containing 1.5 ml pyrogallol (0.05 mol), 0.05 ml enzyme extract and 0.5 ml H 2 O 2 (1%; v⁄v) 48 . The change at 420 nm was determined at every 30 s intervals and the enzyme activity was recorded as U per min per g FW.
Catalase (CAT). CAT (E.C. 1.11.1.6) was assayed by Aebi method 50 . Reaction mixture consisting of phosphate buffer (300 µM, pH 7.2) and H 2 O 2 (100 µM) in enzyme extract (1 ml) was allowed to release O 2 by enzymatic dissociation of H 2 O 2 in the dark for 1 min. O 2 produced due to enzyme reaction was determined at 240 nm (extinction coefficient of H 2 O 2 is 0.036 mM −1 cm −1 ). The activity of the enzyme was expressed as µM H 2 O 2 oxidized U min −1 g −1 FW.
Phenylalanineammonia lyase (PAL). Powdered leaf samples (0.5 g) were homogenized in 5 ml of ice-cold phosphate buffer (100 mM; pH 7.0 and 0.5 mM EDTA and mixed with 1.4 mmol l −1 β-mercaptoethanol 53 . The homogenate was centrifuged (15000 g, 15 min) and the supernatant was added with 0.1 mol l −1 l-phenylalanine (pH 8.7, 1 ml) along with the mixture of 0.5 ml 0.2 mol l −1 phosphate buffer (pH 8.7), 0.2 ml enzyme extract and 1.3 ml distilled water following incubation for 30 min. Trichloroacetic acid (TCA, 0.5 ml, 1 mol l −1 ) was added to terminate the reaction. The observations were recorded at 290 nm and activity was expressed in terms of µmol t-cinnamic acid g −1 FW. estimation of non-enzymatic antioxidative reactions. Free radical scavenging activity (FRSA). The free radical scavenging activity was evaluated by 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging method using the stable radical DPPH 54 . Plant extract with TPC (100 μl) was mixed with 2.9 ml freshly prepared DPPH solution (60 μM in MeOH). The reduction in DPPH radical was determined at 515 nm till 1 h until stable values were obtained.
ABTS activity. The ABTS activity in TPC from the rice leaf was determined using the ABTS• + decolorization method 55 . The reaction mixture containing 10 ml ABTS• + radical (ABTS 9.5 mL, 7 mM) and potassium persulfate (245 μL, 100 mM) was kept in the dark for 18 h and then diluted with potassium phosphate buffer (0.1 M, pH 7.4) to an absorbance of 0.70 (±0.02) at 734 nm. TPC from rice leaves (50 μL) was mixed thoroughly with 2.95 mL ABTS radical solution. The absorbance was recorded at 734 nm and expressed as % inhibition of the activity.
Ferric reducing power antioxidant assay. The Fe-ion reducing power assay was performed with the leaf extracts taking quercetin as the standard compound 56 . To 200 and 500 μl aliquots, 1.0 ml MeOH, 2.5 ml phosphate buffer (pH 6.6) and 1% (w/v) potassium ferricyanide were added. Reaction mixture was incubated at 50 °C for 20 min and 2.5 ml TCA (10% w/v) was added to terminate the reaction. Absorbance was recorded at 700 nm and percent increase in Fe reducing activity was calculated.
Total RNA isolation and cDNA synthesis. Total RNA was isolated from 0.2 g fresh rice leaves using TRIzol ™ LS reagent (Invitrogen; http://www.invitrogen.com). Three µg of the total RNA was digested with RNase-free DNase I (Thermo Scientific) to remove genomic DNA contamination. The poly(A)-RNA concentration was determined using NanoDrop 2000 spectrophotometer (Thermo Scientific). Samples with a 260/280 ratio of 1.9-2.1 and a 260/230 ratio ≥ 2.0 were chosen to determine the quality and purity of the RNA preparations. The integrity of the purified RNA was checked on 2% formamide denaturing gel. Subsequently, first-strand cDNA was synthesized in a 20 μL reaction mixture by using RevertAid H minus kit (Fermentas) following the manufacturer's instructions and stored at −20 °C until use.
Quantitative qRT-PCR assay. Gene specific primer sequences for the defense-related genes as listed in Table 1 were obtained from TIGR Rice Genome Annotation Resource 58 with the help of BLASTn and were synthesized from Helix Biosciences, India. qRT-PCR amplification was performed in 96-well plates with a iQ5 RT-PCR Detection System (BioRad Laboratories, Germany) using Green Supermix Kit Eva Green SYBR ® (BioRad). Expression of the gene specific primers at a concentration of 0.1 µM was analyzed 42 . In brief the qPCR conditions were: denaturation at 95 °C for 2 min followed by 40 repeats at 95, 60 and 72 °C temp for 20, 30 and 25 s. The sense/antisense primer sequences for actin (5'-TCCATCTTGGCATCTCTCAG-3'/5'-GTACCCTCATCAGGCATCTG-3') and rRNA (5'-CTTCGGGATCGGAGTAATGA-3'/5'-AACTAAGAACGGCCATGCAC-3'), respectively were used as internal controls for normalizing relative gene expression levels in technically independent and triplicate biological experiments 59 . The threshold cycle (Ct) was measured automatically by the software.
Statistical analysis. Data were subjected to Two Way ANOVA using PRISM version 5.0. Tests for normality of data and for homogeneity of variances were performed before running ANOVAs. PCA analysis was carried out using R-program. Except for the real time experiments using qRT-PCR, for which three replications were used, all the experiments were performed in complete randomized block design having six replications (n = 6). For the gene expression analyses, the expression values of the two housekeeping genes (actin and rRNA) were subjected to Two-way ANOVA using geometrical means of the internal controls and based on the mean values, the expression profile of all the genes was normalized. For all the experiments, the data were expressed as the mean value of the replicates. Standard error for each mean value was represented separately in the table and figures.

Results and Discussion
Plant responses to abiotic stresses are growth dependent and complex 60 . The underlying array of mechanisms for stress avoidance, tolerance and adaptation are conditional constraints involving multiple cellular physiological, metabolic and molecular alterations 31 . Stress induced antioxidative conditions within the cells generate reactive oxygen species (ROS) and lead to accumulation of free radicals that disrupt cellular homeostasis and adversely affect cell viability 61 . Stressed plants undergo multiple intrinsic equilibrations for early stress perception, signal channeling, gene expression and metabolic modifications to refrain from unfavorable conditions 62 . Microbial interactions with plants elicit modulation in molecular mechanisms to activate metabolic networks at gene, enzyme and metabolite level. This works in parallel to enhance plant's intrinsic strength to support stress mitigation 63  Pair-wise analysis indicated significant differences between control and Th, Pf and Th + Pf inoculated non-drought plants. Further, the protein concentration in control plants was also significantly different than those grown under drought condition or in the plants challenged with the drought and given microbial inoculation ( Table 2). Drought or desiccation tolerance in plants is known to promote accumulation of biomolecules including proteins 64 .
Drought reduced shoot and root dry weight although microbial inoculation substantially supported plant growth. Tukey's pairwise tests indicated significant differences (p < 0.05) in between non-drought (control) plants and both Pf and Th + Pf inoculated plants, but no significant differences either between control and Th inoculated drought treated plants or between non-drought Pf and Pf + Th inoculated drought treated plants (Table 2). Similarly, on root dry wt, the impact of drought [F(1,40) = 16.67, p = 0.0002] and microbial inoculation [F(3,40) = 62.89, p < 0.0001] was statistically significant but the impact of interaction was non-significant [F(3,40) = 1.598, p < 0.2049]. Reduction in growth parameters in rice is the most obvious negative impact of drought and water deficit 65 . We reported that despite drought, microbial inoculation supported growth and development of shoot and root of rice plants in almost similar way as was evidenced under non-drought condition. Therefore, the negative impact of one factor (drought) is substantially being compensated by the other factor (microbial inoculation). Since growth promoting microorganisms enhance nutrient uptake by the plants, produce phytohormones and stimulate plant's immune system 14 , the observed effect of microbial inoculation on developmental parameters, even in stress challenged plants, seems natural. These observations provided evidence that microbial inoculation may protect plants by bringing positive changes at physiological and morphological level under drought challenged condition.

Sl. No.
Gene Name Primer 5′-3′   (Fig. 1a, Supplementary Table 1). We showed that although drought led to high H 2 O 2 level, microbial inoculation lowered the magnitude of accumulation and thereby, lowered the toxic effect of H 2 O 2 in the cells. This is further evidenced from the in situ hypersensitive reaction in the leaves of the rice plants (Fig. 1b). Leaves of non-stressed plants (Fig. 1b,A) grown with microbial inoculation (Fig. 1b,B,C,D) remained almost free from the lesions. Leaves of the plants grown under drought showed maximum stained lesions (Fig. 1b,E). However, microbial inoculation helped stressed plants minimize hypersensitive spots on the leaves (Fig. 1b,F,G) and minimum lesions were seen over the leaves of the plants inoculated with Th + Pf (Fig. 1b,H). Higher accumulation of H 2 O 2 in plant cells is a toxic phenomenon leading to hypersensitive cell death. Microbial inoculation not only reduced the level of H 2 O 2 in drought stressed plants, but it also minimized lesion development due to hypersensitive cell death in plant leaves. Drought as an unfavorable condition leads to the overproduction of H 2 O 2 that eventually increased phytotoxicity leading to cell necrosis. Existing reports further confirm such processes in plants experiencing stressed conditions [66][67][68] .  Table 2. Impact of microbial inoculation on protein concentration and shoot and root dry weight of rice plants grown under non-drought and drought-challenged conditions. p values in Bold are significantly different.  (Table S1). Drought-stressed plants had always significantly higher total polyphenol concentration than non-stressed plants (Fig. 2a). On the other hand, one-way ANOVAs and post hoc Tukey's tests on both the drought stressed and non-stressed plant cohorts showed that significantly the lowest total polyphenol concentration was always in uninoculated plants. Among the three inoculation treatments in the cohort of drought-stressed plants, combined inoculation resulted in significantly high polyphenol concentration. Also, in the cohort of non-stressed plants, plants doubly inoculated with Trichoderma and Pseudomonas (Th + Pf) had significantly higher (p < 0.05) total polyphenol concentration than singly inoculated plants (Fig. 2a).
Microbial inoculation resulted in enhanced activity of PAL enzyme in rice leaves. One way ANOVA and Tuckey's test results on drought and non-drought plants indicated that significantly low PAL activity was always reflected in stressed plants (Fig. 2b). In the cohort of non-stressed plants that always showed higher PAL activity than stressed plants, those with combined inoculation of Th + Pf had significantly high PAL activity than any other single microbial inoculation.   (Fig. 2b). Microbial inoculation to plants under stressed condition influences accumulation of polyphenolics and activates PAL enzyme activity [72][73][74][75] . Since polyphenolics are strong antioxidants and PAL is a defense-related enzyme, high accumulation of polyphenolics and enhanced PAL enzyme activity in the leaves are supposed to strengthen plants under drought challenged condition. Having shown that the microbial inoculation enhanced polyphenolic accumulation and improved PAL enzyme activity, the expression of PAL gene was checked in plant leaves (Fig. 2c). Microbial inoculation enhanced PAL gene expression in the non-drought plants. In the cohort of plants grown under drought following microbial inoculation, expression of PAL gene was multi-fold enhanced (Fig. 2c) Table 2). Stressed conditions usually activate phenylpropanoid pathway, in which PAL is a key gene to offer physiological and structural support to the plants 76,77 . A correlative activation pattern of the PAL gene, the enzyme activity and accumulation of polyphenolics in the leaves of rice plants grown with microbial inoculation was found under drought stress. Such biochemical and molecular strategies are presumed to confer cumulative support to rice to tolerate the adverse impact of stress.

Polyphenolics accumulation enhanced antioxidant profile in inoculated plants. Normal concen-
tration of intracellular ROS regulates redox state in the cells and also acts as signals for defense against stresses 78,79 . Unfavourable conditions enhance production and prolonged accumulation of ROS in cellular compartments 80 , a condition that becomes phytotoxic with deleterious impact due to oxidative damage of cell membrane 81,82 . Small molecule metabolites like phenolics, tocopherol, carotenoids and proline maintain redox state in cells during oxidative damage as ROS scavengers 2,83 . This is why enhanced polyphenolics concentration usually favours ROS scavenging in the plants grown under stress conditions. Two way ANOVA results showed that the effect of microbial inoculation on free radical scavenging activity ( (Fig. 3b). Two way ANOVA results showed that the effects of microbial inoculation on ABTS inhibition was significant [F(3,40) = 11.80, p < 0.0001]. The impact of watering regime was  (Fig. 3c).
The impact of microbial inoculation on Fe +2 chelation activity in plants was statistically significant [F(3,40) = 26.28, p < 0.0001]. The effects of drought was again found to be significant [F(1,40) = 27.63, p < 0.0001] (Fig. 3d, Supplementary Table 3). However, the impact of interaction was statistically non-significant [F(3,40) = 1.255, p0.3029]. Drought induced H 2 O 2 production in plants has been obvious from the results (Fig. 1a) that could lead to high ROS accumulation. We presume that due to high concentration of polyphenolics in leaf extracts, rice plants show ROS scavenging strategy to neutralize the impact of oxidative toxicity. The results apparently describe that polyphenols in leaves of rice plants grown under microbial inoculation has profound non-enzymatic ROS scavenging impact. This strategy appears to be a promising stress tolerance mechanism in plants grown under drought 6,9,84 .  Table 4). Results indicated that microbial inoculation to plants enhanced SOD activity even under drought challenged conditions. It is presumed that SOD is helpful in extending the first line of defense to the plants as they play vital role as ROS scavengers.

Microbial inoculation activate antioxidant defense enzymes in rice.
Glutathione reductase (GR) is a potential enzyme in the antioxidative enzyme system of the plants. Two way ANOVA indicated that the effects of watering regime on GR activity in plants was significant [F(1,40) = 147.2, p < 0.0001] and so was the impact of microbial inoculation [F(3,40) = 44.07, p < 0.0001] and that of interaction [F(3,40) = 12.46, p < 0.0001) (Fig. 4b, Supplementary Table 4). In the cohort of plants challenged with drought and inoculated with the microbial inoculants, the value of GR activity was high in doubly inoculated plants   92 . We demonstrated that in the cohort of non-drought plants, microbial inoculation led to enhance peroxidase activity in rice leaves and maximum activity was found due to doubly inoculation of Th + Pf. Within the cohort of inoculated plants challenged with the drought, again doubly inoculation of Th + Pf showed maximum PO activity than single inoculation or drought plants alone (Fig. 5a). The effects of drought on PO activity was found to be sig-  Table 5).
Catalase possesses high affinity for H 2 O 2 and catalyzes its dismutation into H 2 O and O 2 7,92 . In the cohort of stressed plants, plants doubly inoculated with Th + Pf had high catalase activity than single inoculations. Likewise, within the cohorts of inoculated non-stressed plants, double inoculation again led to high catalase activity (Fig. 5b). The impact of watering regime on catalase activity was significant [F(1,40) = 379.9, p < 0.0001] and so was the significant impact of microbial inoculation in plants [F(3,140) = 30.42, p < 0.0001]. However, the impact of interaction on catalase activity was found to be non-significant [F(3,40) = 0.6272, p0.6017] (Fig. 5b, Supplementary Table 5).
GPX reduces the level of H 2 O 2 in the cells during stress conditions 93,94 . We showed that in the cohort of drought stressed plants, those inoculated with Th + Pf showed high GPX activity than those with single microbial inoculations (Fig. 5c) Table 5).
Enhanced level of defense related enzymes is directly related to the degree of drought experienced by the plants 95 . Cell-bound peroxidases act as detoxifier of H 2 O 2 produced as a byproduct of antioxidative mechanism 9 . The PO acts in H 2 O 2 -scaveginging and oxidize flavonoid and phenylpropanoids 72 . APX also performs H 2 O 2 scavenging in the cytosol and chloroplast with the help of ascorbate as specific electron donor 91 . Thus, higher activity of both PO and APX is presumed to have a role in detoxification of enhanced H 2 O 2 accumulation in the cells. Enhanced level of GPX and catalase is supposed to support plant's biochemical strategy to mitigate drought under microbial inoculation. The enhanced activity of PO, APX, GPX and CAT enzymes in different cohorts of experiments led us to affirm the role of i) the enzyme activation and activity in imparting protection against stresses and ii) the microbial species in modulating enzyme activity in plants challenged with drought. Enhanced level of the defense related enzymes due to microbial inoculation go in parallel to different molecular mechanisms and strengthen the plant's performance under stressed conditions. Microbial inoculation up-regulates the genes encoding dehydration tolerance. We analysed gene expression of OsPIP1;1, a prominent representative of rice plasma-membrane protein gene family that regulates aquaporin 96 . The impact of inoculation on the expression of OsPIP1;1 was statistically significant [F(3,16) = 12.34, p0.0002] but that of drought was non-significant [F(1,16) = 0.1953, p0.6644] (Fig. 7a, Supplementary Table 6). The interaction effect on the expression of this gene was statistically significant [F(3,16) = 6.054, p0.0059]. Microbial inoculation therefore, up-regulated OsPIP1;1 of the PIP gene family in both the cohorts of stressed and non-stressed plants. OsPIP1;1 is an important gene, the protein product of which is related to less water permeability in the plant cells 97 . We showed that microbial inoculation in plants growing normally (non-stress) led to up-regulation of OsPIP1;1 gene. Within the cohort of stressed plants, maximum upregulation was observed in plants inoculated with Pf alone (Fig. 7a). The results indicate positive role of microbial inoculation in the modulation of OsPIP1 gene, which regulates aquaporin, the water channel protein that mediates stress tolerance in rice plants.
Dehydrins (DHNs) play key role in responding to adaptation against abiotic stresses 98 Table 6). These observations, together with the enzyme activity provided evidences to confirm that microbial inoculation modulates expression of stress responsive genes linked with dehydration. This further makes a clearer understanding on the activation of strategic molecular mechanisms meant for avoidance or adaptation against stress damage in rice due to microbial inoculation.  www.nature.com/scientificreports www.nature.com/scientificreports/ was significant (Fig. 8a, Supplementary Table 7 (Fig. 8c). It was interesting that within the cohort of the three treatments of microbial inoculation in plants growing under stressed condition, inoculation of Pf bacteria showed high upregulation values for all the three genes (Fig. 8). Except for the DREB gene which showed maximum over expression in the cohort of non-stressed plants inoculated with Pf (Fig. 7c), inoculation of plants with the bacteria Pseudomonas alone showed consistently high expression values of OsPIP1;1, DHN and all the three isomorphs of SOD genes in the cohort of stressed plants (Figs. 7a,b and 8a-c). It was concluded that the over-expression of SOD gene isoforms leads to enhanced activity of SOD enzyme in rice plants grown under microbial inoculation and drought challenged condition. Presumably, the enhanced gene expression and subsequent enzyme activity level might have played an important role in reducing the deleterious impact of ROS in rice grown under stress.

Microbial inoculation enhanced expression of genes encoding peroxidation of H 2 o 2 . APX
gene regulates ascorbate-glutathione (AsA-GSH) cycle that plays key role in the reduction of H 2 O 2 to H 2 O 105,106 . Over-expression of APX gene in plants improves oxidative defense and offers tolerance to abiotic stress 105 . In Figure 7. Effect of microbial inoculation and drought stress on the expression of OsPIP1(a), DHN (b) and DREB (c) genes related to less water permeability and dehydration tolerance in rice. Significance values were determined using two-way ANOVA. n = 3; Data are shown as mean ± SEM for each sample; ns is nonsignificant. (2020) 10:4818 | https://doi.org/10.1038/s41598-020-61140-w www.nature.com/scientificreports www.nature.com/scientificreports/ the cohort of plants grown with stress and inoculated with Th, Pf and Th + Pf, single inoculation of Pf showed high overexpression of APX gene than Th or combined inoculation of Th + Pf (Fig. 9a) (Fig. 9a, Supplementary Table 8). The bacterial inoculant Pf showed maximum over-expression of APX gene in plants under drought condition than Th or doubly inoculation of Th + Pf. Inoculating plants with microbial inoculants enhanced expression of the peroxidase (PO) genes (PO D14481 and PO AU076282) in rice. Within the cohort of stressed and non-stressed plants, inoculation resulted in enhanced over-expression than the control. Maximum over-expression was again recorded in plants grown under stressed conditions and inoculated with Pf (Fig. 9b,c) (Fig. 9c, Supplementary Table 8). The effect of the inoculation of bacterial inoculant Pf on the expression of PO AU076282 gene in plants grown under drought was maximum than Th or doubly inoculation of Th + Pf. Results indicated that microbial inoculation helped rice plants in over-expressing peroxidases and the inoculation of Pseudomonas was invariably instrumental in www.nature.com/scientificreports www.nature.com/scientificreports/ highest over-expression of these genes. Peroxidases are the key genes in regulating ROS scavenging and thus, their over-expression in rice can have protective role in plants exposed to drought.
Over-expression of CAT gene enhances oxidative defense response in plants 107 . Inoculation of rice grown under drought condition with Pf resulted in highest level of expression of CAT gene in the cohort of drought stressed and inoculated plants (Fig. 9d). The impact of drought on the gene expression was non-significant  Table 8). The results strongly suggested that microbial inoculation had a positive role in the over-expression of the genes linked with the peroxidation of H 2 O 2 in the plants challenged with the drought. Invariably, the effect of inoculation of Pseudomonas substantially enhanced APX, PO and CAT gene expression in plants grown under stressed condition. These modulations in gene expression may support improved drought tolerance in rice plants.

conclusion
We have shown that although drought suppressed growth of rice plants, as is evident from reduced shoot and root weight, microbial inoculation managed to reduce the impact of drought. There have been multi-pronged mechanisms utilized and adopted by the plants to mitigate and/or minimize the impact of drought if the plants were inoculated with the microbial species. Generation of ROS is a common phenomenon in plant cells under stressed conditions. We reported hyperaccumulation of H 2 O 2 in rice leaves and the resultant hypersensitive cell death responses thereafter. Induced accumulation of the PAL gene transcripts and resultant activation of PAL enzyme facilitated higher accumulation of the phenylpropanoids that have strong ROS scavenging activity and might have helped plants to overcome oxidative burden created due to drought stress.With the activation of the antioxidant enzymes SOD, PO, APX and CAT, rice plants were supposed to minimize tissue damaging impact of high H 2 O 2 levels. Over expression of all the isoforms of SOD, Cu-Zn SOD, Mn-SOD and Fe-SOD genes suggested that microbial inoculation helped plants activate SOD activity as first line of defence at various levels of cellular compartments strongly to overcome ROS burden. Microbial inoculation in plants further improved the activity of the enzymes PO, APX, GPX and GR that have also contributed in reducing ROS burden in the plants following drought challenge. We also observed enhanced regulation of less-water permeability-linked gene, OSPiP1 that regulates aquaporin, drought-adaptation gene DHN and dehydration related DREB gene. Presumably, up-regulation of genes encoding phenylpropanoids, dismutation of superoxide radicals and peroxidation of H 2 O 2 in microbe inoculated and drought challenged condition strongly contributed towards stress mitigation. Enhanced enzymatic and non-enzymatic antioxidant activities were thought to be the repercussions of the enhanced gene expression levels in microbial inoculated plants and have also helped in minimizing antioxidative load to overcome the oxidative stress. We further conclude that the physiological, biochemical and molecular mechanisms contributing to drought mitigation in rice following microbial interaction are multi-faceted, Figure 9. Impact of microbial inoculation on the expression of Chl_sAPX (a), peroxidase D14481 (b) peroxidase AU076282 (c) and Catalase (d) genes in rice plants grown under drought and non-drought conditions. Statistical significance was determined by two-way ANOVA; data are mean ± SEM; n = 3 for transcript analysis. ns is non-significant.