Deciphering the mechanisms of microbe mediated drought stress alleviation in wheat

Drought stress adversely influences the crop plants. Herein, present research was designed to elucidate the role of plant growth promoting bacteria for amelioration of water stress in wheat. A pot experiment was conducted for screening the microorganisms on the basis of plant growth, chlorophyll and proline content under water stress. Bacillus sp. BT3 and Klebsiella sp. HA9 were found to be more promising strains that positively influenced the plant growth, chlorophyll and proline status of seedlings under water stress condition. Further, Bacillus sp.BT3 and Klebsiella sp. HA9 along with check strain (BioNPK) were used for elucidating their detailed effect on morphological, biochemical, physiological and molecular traits to mitigate water stress in wheat. Microbial inoculation significantly improved plant height, fresh weight and dry weight of root and shoot, relative water content, chlorophyll content and root morphological parameters over the uninoculated water stressed (30% FC) control. Likewise, sugar content, protein content and antioxidant enzymes were also significantly enhanced due to microbial inoculation under water stress (30% FC). Microbial inoculation significantly decreased proline, glycine betaine, lipid peroxidation, peroxide and superoxide radicals in wheat over the uninoculated water stressed (30%FC) control. Quantitative real-time (qRT) PCR analysis revealed that Bacillus sp. BT3, Klebsiella sp. HA9 and BioNPK inoculation significantly upregulated stress responsive genes (DHN, DREB, L15 and TaABA-8OH) over the uninoculated water stressed (30% F.C.) control. The study reports the potential of Bacillus sp. BT3 and Klebsiella sp. HA9 along with BioNPK in water stress alleviation in wheat which could be recommended as effective bioinoculants.


Introduction
Drought stress is thought to be one of the crucial problems in agriculture because it changes the morphology, biochemistry and metabolic activities of plants. Plant cells make free radicals and reactive oxygen species (ROS) in the event of a drought. A higher amount of ROS could cause cellular degeneration and lower the amount of chlorophyll (Anjum et al. 2011;Hasanuzzaman et al. 2018;Ngumbi and Kloepper 2016;Rahdari et al. 2012). ROS also cause lipid peroxidation, protein denaturation, membrane leakage, and oxidative damage to nucleic acids (Hasanuzzaman et al. 2020;Impa et al. 2012). Water scarcity causes diminished turgor pressure and water potential resulting in stomata closure leading to reduced rate of photosynthesis (Martin-StPaulet al. 2017;Rodriguez-Dominguez and Brodribb 2020). Water limitation can also influence nutrient absorption and translocation, as nutrients move to the roots through water (Bista et al. 2018;Rouphael et al. 2012). Biosynthesis of ethylene gas is the main physiological response of plants to drought stress that slows their growth. Therefore, drought stress has a detrimental impact on the growth, biomass, and yield of the plant (Ullah et al. 2017(Ullah et al. , 2018aKumar and Verma 2018). Plants have adaptive mechanisms, such as the production and storage of osmoprotectants and antioxidants, to counter the negative impacts of drought (Anjum et al. 2017;Hasanuzzaman et al. 2018;Kaushal and Wani 2016;Kim et al. 2017). Phytohormones also help plants to withstand under abiotic stresses (Ullahet al. 2018b). Abscisic acid (ABA) levels rise in response to water stress, initiating droughtresponsive signalling pathways that, in turn, govern the plant's morphological, physiological, and biochemical responses, (Egamberdieva et al. 2017;Khan et al. 2018;Wilkinson et al. 2012;Zonget al. 2020). Plants respond to drought stress by modulating stress-related genes such as DHN, DREB, NAC, Hsp, EREBPs, LOXs, L15, WRKY, etc. (Gontia-Mishra et al. 2016;Li et al. 2012;Sallam et al. 2019;Zhu 2016). Drought tolerance genes respond to signal transduction, stress response, and drought stress by producing protein products (Zhou et al. 2010). Functional proteins, such as LEA protein, chaperones, and osmotic regulators, and regulatory proteins, which are components of signalling pathways and gene transcription, are most commonly involved in stress management (Takahashi and Shinozaki 2019).
Rice, maize, and wheat (Triticum aestivum L.) are the world's three primary food grains. Abiotic stresses like soil salinity, water shortage, and micronutrient insufficiency threaten wheat productivity (Kumar and Verma 2018). Daryanto et al. (2016) reported that wheat productivity was declined by 21% with a 40% reduction in water availability. To fulfil global food demand, improving wheat's drought tolerance is a major area of research. To mitigate drought's effects, conventional plant breeding, biotechnology, and agronomical methods have been developed. Plant breeding and biotechnology create drought-tolerant cultivars to combat drought stress (Bodner et al. 2015;Dolferus et al. 2019;Kazemia and Safaria 2018;Mohammadi and Amri 2011). Agronomic strategies such as water saving irrigation, shifting the crop calendars, mulching and resource management practices etc. are used to improve the capacity of plants to bear with the water stress (Venkateswarlu and Shanker 2009). However, most of these technologies are highly complex, expensive, labour intensive and non-renewable. Thus, environment friendly methods that improve soil production are needed. Plant growth-promoting microbes can reduce drought stress in a sustainable and eco-friendly way.
Plant growth-promoting microorganisms can help plants cope with drought stress by increasing plant growth and improving nutrient uptake and translocation. These microbes can mitigate the drought response of crop plants by improving water potential of cells through modulating accumulation of osmoprotectants, antioxidants, up-or down-regulation of stress responsive genes (Singh et al. 2020;Ullah et al. 2019). Plant growth-promoting microorganisms accelerate biosynthesis and metabolism of ACC (1-aminocyclopropane-1-carboxylate) deaminase enzymes, phytohormones, and volatile organic molecules, enabling crop plants endure drought (Danish et al. 2020;Gontia-Mishra et al. 2016;Ngumbi and Kloepper 2016;Vurukonda et al. 2016). ACC deaminase enzymes lower plant ethylene content and protect against environmental stress (Ali and Kim 2018). Plant growth-promoting bacteria increased biochemical and functional features like effective quantum yield of PSII photochemistry (YII)), electron transport rate (ETR), photochemical efficiency, transpiration rate, and stomatal conductance under drought conditions in plants (Yaghoubi Khanghahi et al. 2021;Saleem et al. 2021).
Crops including mung bean, pea, maize, wheat, and chickpea, etc., have been shown to benefit from the presence of microorganisms in a number of investigations (Kasim et al. 2013;Mayak et al. 2004;Naveed et al. 2014;Sandhya et al. 2010;Sarma and Saikia 2014;Tiwari et al. 2016;Zahiret al. 2008). Plant growth-promoting rhizobacteria (PGPR) are thought to have a substantial role in reducing wheat's vulnerability to water stress; however there is scant evidence to support these claims. Consequently, it is necessary to conduct research that sheds more insight on the mechanisms that work in microbial species with the potential to mitigate drought stress in a sustainable manner. The primary goals of this research were to assess the efficacy of various bacterial strains in mitigating the effects of drought stress on wheat and to explore the interaction of plant growth promoting bacteria (PGPB) with the wheat crop to improve drought tolerance. Various morphological, physiological, and biochemical features, as well as the expression levels of stress-related genes (DHN,DREB,L15, were examined under drought stress to get a clear picture of plant-microbe interactions.

Microorganisms
Ten bacterial isolates procured from National Agriculturally Important Microbial Culture Collection (NAIMCC), ICAR-NBAIM, India, five bacterial isolates obtained from Microbial Technology Unit II, ICAR-NBAIM, India, along with one BioNPK formulation were used in this study ( Table 1). All of the bacterial cultures obtained were chosen based on their plant growth promoting activities (data unpublished). In this study, BioNPK was utilised as a check which has found potent in stimulating plant growth in water limiting conditions (Saxena et al. 2020).

Pot experiment for screening of microorganisms for drought stress alleviation
An initial screening was conducted with all the microorganisms used in the study on the basis of chlorophyll content, proline content and various plant growth parameters. The best isolates obtained from the study would be used later for carrying out detailed investigation. Therefore, one month pot trial was conducted in a glass house at ICAR-National Bureau of Agriculturally Important Microorganisms, Mau, Uttar Pradesh, India ( Supplementary Fig. 3). Each pot (4″ diameter) was filled with 0.5 kg sterilized sand: soil mixture (1:3) which was sterilized in autoclave by tyndallization technique. Soil pH and electrical conductivity (EC) were analysed and found to be 8.34 and 1.21 dSm −1 , respectively (Singh et al. 1999). The pots were weighed and 100% saturated with water to calculatethe field capacity (FC). Thereafter, the weight of each pot (sand: soil and water) was calculated for 50% and 30% FC which was then used to maintain the water levels at FC according to the treatments. Drought stress levels were assessed using 30% (Stressed Control) and 50% (Un-stressed Control) FC. Log phase broth cultures (containing10 9 CFU/ mL) were mixed with 0.2% carboxymethyl cellulose (CMC) carrier and coated on wheat (var. HD2967) seeds. Seeds treated with nutrient broth and CMC were used as uninoculated control. Prior to seed treatment, the wheat seeds were surface sterilised for 1 min with 70% ethanol and then for 5 min with a 1.5% sodium hypochlorite solution (Rudolph et al. 2015). Eight seeds were sown in each pot which were trimmed to four plants after germination. All treatments were taken in triplicates and randomised. Recommended dose of NPK (60:30:20) mg kg −1 of soil was applied in all the treatments. The pots were weighed everyday and water was simply supplied to maintain field capacity (FC) according to drought stress levels of treatments. The experiment was set up using the following treatments -(i) 30% FC-Uninoculated stressed control; (ii) 50% FC-Uninoculated non-stressed control; (iii) 30% FC + microbial inoculation. A total of 15 bacteria and one BioNPK formulation were used in our investigation, resulting in a total of eighteen treatments (Tables 3 and 4).

Analyses of plant growth and biomass
Three plants were uprooted from each treatment replicate after 30 days of sowing. Root and shoot length were measured using inch tape. Fresh weight of root and shoot were determined using a weighing balance. To determine the dry weight, wheat roots and shoots were incubated in hot air oven at 80 °C for three days.

Analysis of root morphology
Root studies were carried out by collecting plants from three replicates after 30 days of sowing. The adherence of the soil to the roots was detached by the method of Costa et al. (2002). The LA2400 (3rd Gen.) scanner was used to measure root length, surface area, projected area, volume, average diameter, number of root tips, number of forks and number of links. Thereafter, WIN RHIZO Programme v.2017 software (Regent Instruments Inc. Ltd., Quebec, Canada) was used to analyse the actual values of each root parameters.

Relative water content
Weatherly's method was used to analyse the relative water content (RWC) (Weatherly 1950). Leaves were harvested from the plants and weighed. Thereafter, leaves were transferred in distilled water for 24 h. After the leaves were fully turgid, they were weighed again. After weighing, the leaves were put in oven at 80 °C for 72 h and dry weight was noted. The following formula was used to determine RWC.

Chlorophyll and carotenoid content
Chlorophyll a, b, total chlorophyll and carotenoid were analysed following the method described by Arnon (1949). One gram fresh leaves were grounded in liquid nitrogen followed byhomogenizationin 80% acetone. A spectrophotometer was RWC = fresh weight − dry weight turgid weight − dry weight × 100 used to measure the absorbance of the supernatant at 663, 645, and 480 nm (Analytik Jena).

Proline content
The acid-ninhydrin approach established by Bates et al. (1973) was used to determine the proline content of root and leaf tissues using spectrophotometry. Plant material (0.5 g) was crushed with liquid nitrogen and homogenized in 10 ml of 3.0% sulfosalicylic acid. Homogenate was centrifuged at 10,000 rpm for 20 min. Supernatant (2 ml) was mixed with equal volumes of acid ninhydrin and glacial acetic acid and heated at 100 °C for 1 h. Reaction was stopped by putting the tubes in ice. After that, toluene (4 ml) was added and mixed thoroughly by vortexing for 15 to 20 min. Absorbance of the upper layer was recorded at 520 nm using a spectrophotometer (Scan Drop, Analytik Jena). A standard curve was prepared using L-proline (10-100 µg/ml) as a standard.

Sugar content
Method described by Dubois et al (1951) was followed to calculate sugar content. Plant samples (200 mg) were cooked for one hour with 10 ml of 80% ethanol. The extract was filtered through Whatman no. 1 filter paper after cooling. One ml of filtrate was added with 1.0 ml of 5% phenol and 5 ml concentrated H 2 SO 4 and vortexed well. Absorbance was recorded at 490 nm and the concentration of sugar was calculated with reference to standard curve made using varied concentration (0-100 µg/ml) of glucose.

Protein content
The Bradford assay was used to analyse the protein content of plant roots and leaves (Bradford 1976). Plant materials (0.5 g) were crushed in liquid nitrogen and homogenised in 5 ml sodium phosphate buffer (10.0 mM, pH 7.0). The suspension was then centrifuged for 20 min at 12,000 rpm, and 0.5 ml of clear supernatant was mixed with 3 ml of Bradford reagent. A spectrophotometer (Scan Drop, Analytik Jena) was used to measure the absorbance at 595 nm. The concentration of protein in unknown sample was calculated with reference to standard curve made from Bovine serum albumin (100-1000 µg ml −1 ). Grieve and Grattan (1983) method was followed to analyse the glycine betaine in plant tissues.

Lipid peroxidation
Lipid peroxidation was assayed following the methods described by Heath and Packer (1968). Plant samples were finely ground in liquid nitrogen and homogenised in 10.0 ml of 0.1% trichloroacetic acid (TCA). The homogenate was centrifuged at 1500 g for 15 min. One ml of supernatant was added to 4.0 ml of 0.5% thiobarbituric acid in 20% TCA. The mixture was then heated for 30 min at 95 °C before being chilled in an ice bath. The mixture was centrifuged for 10 min at 10,000 g after cooling. The absorbance of the supernatant was measured at 532 nm and 600 nm. The extinction coefficient for thiobarbituric acid reactive substance is 155 mM −1 cm −1 . The results were represented as nmol malondialdehyde (MDA) equivalents per gram of fresh weight.

Superoxide dismutase (SOD)
The methodology reported by Dhindsa et al. (1981) was used to determine SOD activity. Plant samples (100 mg) were crushed and centrifuged (15,000 g, 20 min.) in 0.1 M phosphate buffer (pH 7.5). A reaction mixture (3.0 ml) was made with 0.2 ml of 200.0 mM methionine, 0.1 ml of 2.25 mM nitroblue tetrazolium chloride (NBT), 0.1 ml of 3 mM EDTA, 1.5 ml of 100 mM phosphate buffer (pH 7.8), 0.1 ml of 1.5 M sodium carbonate, and 0.1 ml of enzyme extract. Finally, water was used to make up the final volume to 3 ml. Thereafter, 0.4 ml of 2 μmol l −1 riboflavin was added and exposed to light (15W fluorescent lamp, 15 min). After deactivating the enzyme activity in the dark, the absorbance was measured at 560 nm. One unit of SOD was represented by a 50% decrease in absorbance when compared to the control, which lacked enzyme extract.

Peroxidase (POD)
POD activity was determined using the method reported by Castillo et al. (1984). Plant samples (100 mg) were ground, homogenised, and centrifuged (15,000 g, 20 min.) in 0.1 M phosphate buffer (pH 7.5). A 3.0 ml reaction mixture was made up with 0.5 ml of 96 mM guaiacol, 1.0 ml of 100.0 mM phosphate buffer (pH 6.1), 0.5 ml of H 2 O 2 (12.0 mM), and 0.1 ml enzyme extract. The change in absorbance at 470 nm was recorded at every 30 s interval and the enzyme activity was calculated as Units (U) (tetra guaiacol) min −1 g −1 fresh weight. Tetra guaiacol has an extinction coefficient of 26.6 mM −1 cm −1 .

Ascorbate peroxidase (APX)
APX was estimated following Nakano and Asada method (Nakano and Asada 1981). Plant samples (100 mg) were ground, homogenised, and centrifuged (15,000 g, 20 min.) in 0.1 M phosphate buffer (pH 7.5) containing 1 mM ascorbic acid and 0.5 mM EDTA. A 3.0 ml reaction mixture was made with 0.1 ml of EDTA (3 mM), 1.5 ml of 100.0 mM phosphate buffer (pH 7.0), 0.1 ml of 0.1 mM H 2 O 2 , 0.5 ml of 3.0 mM ascorbic acid, and 0.1 ml enzyme extract. After 60 s, absorbance was taken at 290 nm, and activity was represented as U min −1 g −1 fresh weight.

Catalase (CAT)
CAT activity in the plant samples was assayed following Aebi method (Aebi 1983). Plant samples (100 mg) were ground in 0.1 M phosphate buffer (pH 7.5), homogenized and centrifuged (15,000 g, 20 min.). To make a 3.0 ml reaction mixture, 1.5 ml of 100.0 mM (pH 7.0) buffer, 0.5 ml of 75.0 mM H 2 O 2 , and 50 l enzyme extract were mixed. The final volume of the reaction mixture was made up by adding water. At 30 s intervals, the change in absorbance at 240 nm was measured, and the enzyme activity was expressed as Umin −1 g −1 fresh weight.

Glutathione reductase (GR)
Method described by Smith et al. (1988) was used for measuring glutathione reductase activity. Plant samples (100 mg) were ground, homogenised, and centrifuged (15,000 g, 20 min.) in 10 ml of 0.1 M phosphate buffer (pH 7.5). A 3.0 ml reaction mixture was made with 1.0 ml of 0.2 M phosphate buffer containing 1 mM EDTA, 0.5 ml of 3.0 mM 5,5-dithiobis [2-nitrobenzoic acid] (DTNB), 0.1 ml of 2.0 mM NADPH, 0.1 ml of 20 mM glutathione disulphide (GSSG), and 0.1 ml of enzyme extract. Spectrophotometric measurements were taken to determine the increase in absorbance at 412 nm. The extinction coefficient of NADPH is 6.22 mM −1 cm −1 . The activity was expressed as U min −1 g −1 fresh weight.

Histo-chemical detection of peroxide and superoxide radicals
Method described by Fryer et al. (2002) was followed for staining of superoxide and peroxide radicals in plant leaves. Plant leaves were placed in tubes and immersed in nitroblue tetrazolium (NBT) (0.2%) and 3, 3′-diaminobenzidine (DAB) (1.0 mg/ ml, pH 3.8) staining solution for staining of superoxide and peroxide radicals respectively. The tubes were placed in the desiccator and attached to a vacuum pump for increasing the infiltration of staining solution.
Tubes were rolled up in aluminium foil and left overnight at room temperature. After incubation period, staining solution was drained off. Stained leaves were thoroughly washed for 5 min in an acetic acid-glycerol-ethanol (1:1:3) solution at 100 °C. Leaves were transferred onto a paper towel saturated with 60% glycerol. Superoxide radicals were visualised as a dark blue stains due to NBT precipitation and peroxide radicals were visualised as reddish brown stains due to DAB polymerization.

Analyses of expression of genes associated with drought response in root and shoot of wheat
Thirty-day-old root and shoot samples were collected, and total RNA was extracted using the Trizol method (Rio et al. 2010). qPCR was used to validate the expression of genes (DHN, DREB, L15, and TaABA-8OH) with potential roles in drought stress response. Three independent samples of each were used. TOPscript TM cDNA synthesis kit (Enzynomics, Republic of Korea) was used synthesize cDNA from 2 μg of total RNA, according to the manufacturer's protocol. β-actin gene was used as an endogenous control to normalize the expression levels. Table 2 lists the gene-specific primers used for qPCR. The Agilent Mx3000P™ PCR platform and Maxima SYBR Green qPCR kit Master Mix (2X) Universal (Thermo Fisher Scientific) were used for the qPCR following the manufacturer's instructions. The 2 −ΔΔCt method was used to calculate the relative expression levels of the selected genes normalised to the expression level of β-actin from cycle threshold values. Three independent biological replicates with three technical replicates were used in the experiment.

Statistical analyses
The results of the experiment were presented as the average of three replications. The results of each experiment were statistically analysed using MiniTab 17's one-way analysis of variance (ANOVA). Tukey's test was used to compare mean values of acquired data between treatments (P ≤ 0.05).

Screening of microorganisms
On the basis of plant growth, chlorophyll content, and proline concentration in wheat leaves, microorganisms were evaluated for their ability to alleviate water stress. When compared to their counterparts at 50% FC, un-inoculated stressed plants at 30% FC experienced a significant reduction in plant growth metrics such as root length, shoot length (plant height), fresh and dry weight. Similarly, there was a reduction in the amount of chlorophyll (Table 3) (Table 3). Almost all bioinoculants significantly enhanced the chlorophyll content over the uninoculated stressed control (30% FC). Highest chlorophyll content was recorded with Bacillus sp. BT-3 (12.35 mg g −1 FW) followed by Klebsiella sp. HA9 (10.86 mg g −1 FW). Highest proline content in wheat leaves (8.50 mg g −1 FW) and root (3.67 mg g −1 FW) was recorded for uninoculated water stressed plants. Whereas, lowest proline content in leaves (2.87 mg g −1 FW) and root (1.20 mg g −1 FW) was found with uninoculated non-stressed plants (50% FC). Proline content was higher in the leaves as compared to roots. Proline accumulation significantly decreased due to microbial inoculation as compared to uninoculated stressed control (30% FC). Highest percentage reduction in proline accumulation was recorded with Klebsiella sp. HA9 followed by Bacillus sp. BT3 under 30% FC (Table 4). Thus, Bacillus sp. BT3 and Klebsiella sp.
HA9 were found to be more promising bioagent to enhance the plant growth and were further selected for detailed investigation.

Response of selected microbial inoculants on plant growth
In the next set of experiments, we further evaluated Bacillus sp. BT3, Klebsiella sp. HA9 and BioNPK for their performance to alleviate water stress with detailed morpho-physiological and biochemical analyses. During this experiment again, shoot length (plant height) and fresh weight of uninoculated stressed (30% FC) control plants significantly decreased over the uninoculated non-stressed (50% FC) plants. In case of shoot length, all the microbial inoculations were statistically at par with each other and significantly better than the uninoculated stressed (30% FC) control. Inoculation of Bacillus sp. BT3 recorded the highest shoot length (19.33 cm). Microbial application significantly enhanced the fresh weight and dry weight of shoot over uninoculated stressed (30% FC) control. Inoculation of Klebsiella sp. HA9 recorded maximum shoot fresh weight (317.33 mg). Inoculation of Klebsiella sp. HA9 (57.33 mg) recorded significantly higher root dry weight. In case of fresh and dry

Response of selected microbial inoculants on root architecture
Under water stressed (30% FC) conditions, total root length, number of root tips and number of forks in wheat plants decreased significantly as compared to the non-stressed uninoculated treatment ( Fig. 1A and B). Almost all microbial inoculations significantly increased the total root length, number of root tips, number of links ( Fig. 1A and B) and surface area ( Fig. 2A) over the uninoculated stressed (30% FC) control. There was no significant effect of microbial inoculation on the projected area ( Fig. 2A), root volume and root diameter (Fig. 2B) under 30% FC. Highest number of root tips, numbers of forks and number of links was recorded with inoculation of BioNPK followed by Klebsiella sp. HA9 under water stressed (30% FC) ( Fig. 1A-B, Supplementary  Fig. 1).

Response of selected microbial inoculants on relative water content (RWC)
Relative water content (RWC) significantly reduced in uninoculated water stressed (30% FC) control as compared to the uninoculated non-stressed control (50% FC). Microbial inoculation significantly increased the RWC over uninoculated water stressed (30% FC) control (

Response of selected microbial inoculants on chlorophyll and carotenoids
Uninoculated water stressed (30% FC) control had lower chlorophyll a, b, carotenoids, and total chlorophyll than the non-stressed control. All microbial inoculations significantly increased chlorophyll a, total Chlorophyll, and carotenoids over the uninoculated water stressed (30% FC) control ( Table 6).

Response of selected microbial inoculants on osmoprotectant levels
Total soluble sugar, protein, glycine betaine and proline enhanced significantly in both roots and leaves of the uninoculated wheat plants growing at 30% FC as compared to uninoculated non-stressed (50% FC) control (Figs. 3A-B, 4A-B and 5). The build-up of osmoprotectant was found to be greater in the leaves than in the roots. Microbial inoculated plants had significantly higher leaf total soluble sugar and protein content than the uninoculated water stressed (30% FC) control. Inoculation of Bacillus sp. BT3 significantly increased the protein content of leaves over the other treatments (Fig. 3A). In case of roots, only Bacillus sp. BT3 and Klebsiella sp. HA9 significantly enhanced the sugar content over the uninoculated water stressed (30% FC) control (Fig. 4A). Glycine betaine and proline accumulation in both root and leaves significantly decreased due to microbial inoculation over the uninoculated water stressed (30% FC) control (Figs. 3B, 4B and 5). There were no significant differences among the microbial inoculants with respect to glycine betaine content in both leaves and root. However, Fig. 1 Response of microbial inoculation in relation to wheat root morphology under drought stress. A Root length, number of root tips; B Number of forks and number of links. Data are the average of three replicates ± SD; Grouping information between mean values of obtained data was carried out by Tukey's test and 95% confidence (P ≤ 0.05) Fig. 2 Response of microbial inoculation in relation to wheat root morphology under drought stress. A Projected area and surface area; B Average diameter and root volume. Data are the average of three replicates ± SD; Grouping information between mean values of obtained data was carried out by Tukey's test and 95% confidence (P ≤ 0.05) significantly higher reduction of glycine betaine in leaves and roots were recorded with inoculation of Klebsiella sp. HA9 and Bacillus sp. BT3 respectively, over uninoculated water stressed (30% FC) control (Figs. 3B and 4B). There were no significant differences among the microbial inoculations with respect to proline content in leaves. Both in roots and leaves, inoculation of Bacillus sp. BT3 significantly decreased the proline content over uninoculated water stressed (30% FC) control (Fig. 5).

Lipid peroxidation (MDA content)
Lipid peroxidation in wheat tissues enhanced significantly in uninoculated water stressed (30% FC) control over the uninoculated non-stressed (50% F.C.) control. Microbial inoculation significantly reduced the lipid peroxidation in leaves. Lowest lipid peroxidation in both leaves (29.94 mg g −1 F.W.) and roots (6.33 mg g −1 F.W.) was recorded with Klebsiella sp. HA9 followed by Bacillus sp.
In case of roots, only SOD and POD significantly increased in uninoculated water stressed (30% FC) control over the uninoculated non-stressed (50% FC) control.

Histo-chemical staining
Histological staining showed maximum superoxide radicals (as a dark blue spots) accumulation in uninoculated water stressed (30% FC) control plants (Supplementary Fig. 2A). Likewise, peroxide radicals as reddish brown spots were also visualised more in uninoculated water stressed (30% FC) control ( Supplementary Fig. 2B). Inoculation with Bacillus sp. BT3, Klebsiella sp. HA9and BioNPK reduced the intensity of dark blue spots and reddish brown spots over the uninoculated water stressed (30% FC) control. The intensity of dark blue spots and reddish brown spots was very low for the uninoculated non-stressed (50% F.C.) control plants.

Gene expression
Four drought responsive genes DHN, L15, DREB and TaABA-8OH were targeted for expression studies using gene specific primers. All genes were expressed in wheat root and shoot under drought. All the genes were up-regulated at variable extent both in roots and leaves due to microbial inoculations. In case of roots, highest up-regulation (2.05 folds) of DHN was recorded due to inoculation of BioNPK. While, 1.70 folds and 1.42 folds up-regulation of L15 and TaABA-8OH genes were recorded in case of plants inoculated with Bacillus sp. BT3 and Klebsiella sp. HA9 respectively (Fig. 6A). In case of leaves, highest up-regulation (10.35 folds) of DHN was recorded due to inoculation of Klebsiella sp. HA9 while ~ 5 folds up-regulation of both L15 and TaABA-8OH genes were observed in case of plants inoculated with Bacillus sp. BT3. In leaves, highest up-regulation (1.50 folds) of DREB gene was observed in plants treated with BT3 (Fig. 6B).

Discussion
Global climate change has brought about an increase in the temperature along with decline in rainfall resulting in drought stress adversely affecting crop productivity. Drought stress disrupts normal plant functions causing serious manifestations in the physiological and morphological traits of the plant (Hsiao 2000). Moreover, the induction of reactive oxygen species (ROS) and free radicals hampers the various Fig. 5 Response of microbial inoculation in relation to proline content in wheat roots and leaves under drought stress.Data are the average of three replicates ± SD; Grouping information between mean values of obtained data was carried out by Tukey's test and 95% confidence (P ≤ 0.05) levels of organization mainly by membrane degradation, lipid peroxidation and disruption of various biomolecules in the plant (Bartels and Sunkar 2005;Hasanuzzaman et al. 2018;Meena et al. 2017;Ngumbi and Kloepper 2016). The application of beneficial microbes is being considered as a possible approach to overcome the harmful consequences of water deficit in a faster, more sustainable and cost-effective way. Agriculturally important microorganisms can boost the plant growth through nutrient mobilization and solubilisation, secretion of growth hormones, disease suppression along with strengthening the induced systemic resistance (ISR) thereby improving its yield and productivity.
In the present study, two bacterial cultures viz. Bacillus sp. BT3 and Klebsiella sp. HA9 were found to be the best among the screened microorganisms which positively modulated plant growth parameters, chlorophyll content and proline status that enhanced the potential of wheat plant to endure the imposed drought. Further, promising strains (Bacillus sp. BT3 and Klebsiella sp. HA9) along with check strain (BioNPK biofertilizer) were used for elucidating their detailed impact on plant morphological, physiological, biochemical and molecular traits to alleviate the drought stress in wheat. Again, higher plant growth was also recorded with Bacillus sp. BT3 and Klebsiella sp. HA9 strains than BioNPK under water stress (30% FC). Similar results were also acquired in wheat plants by Chakraborty et al. (2013) who investigated that inoculation of Bacillus safensis and Ochrobactrum pseudogregnonense increased the plant biomass, plant height as well as photosynthetic pigments under water deficit. Rathod et al. (2011) found that water deficit dramatically decreased photosynthetic activities such as performance indices (PI abs ), quantum efficiencies and absorption per reaction centre (ABS/RC) in soybean. However, soybean plants treated with arbuscular mycorrhizal fungi exhibited higher photosynthetic activities and plant growth than uninoculated water stressed control plants.  Sheteiwy et al. (2021a) recently reported that arbuscular mycorhiza and Bradyrhizobium japonicum treated soybean plants had higher leaf chlorophyll, plant biomass and seed yield than uninoculated plants under drought conditions. Different strains of PGPRs like Azospirillum sp., Azotobacter sp., and Pseudomonas fluorescens are distinguished for their beneficial role on plants in water scarcity (Zhu et al. 2020). Singh et al. (2020) also reported that drought decreased the rice growth but treatment of rice plants with Trichoderma and Pseudomonas minimised the negative impact of drought. Khan and Bano (2019) found that Azospirillum, Pseudomonas, Bacillus and Azotobacter inoculation improved the plant growth and biomass of field crops under water deficit. During the second pot experiment in the current study, equal enhancement of chlorophyll content was recorded with Bacillus sp. BT3 and Klebsiella sp. HA9 (88%) over the uninoculated water stressed(30% FC). Chakraborty et al. (2013) and Naveed et al. (2014) also reported that application of microbes enhanced the chlorophyll content in wheat and boosted the plant growth in water deficit. Increased photosynthetic pigment content in the leaves due to application of plant growth promoting bacteria could be because of increased availability of nutrients. Actually, when Bacillus sp. BT3 and Klebsiella sp. HA9 were practiced to wheat plants in water stress, the increase in leaf chlorophyll (Chlorophyll a, Chlorophyll b, total chlorophyll) and carotenoids could improve the photosynthetic efficiency and protect the photosynthetic machinery, which might be linked to wheat plants' drought tolerance.
The relative water content (RWC) of a plant is a diagnostic of the plant's water balance and could be used as a way to make plants more resistant to water stress (Nounjan et al. 2018). In this study, inoculation of Bacillus sp. BT3 and Klebsiella sp. HA9 significantly enhanced the RWC in wheat leaves by 14.33% and 17.67% respectively over the uninoculated water stressed (30% FC) control. The contribution of beneficial microbes in maintaining the RWC for alleviating drought stress has been reported earlier as well (Ngumbi and Kloepper 2016;Naseem and Bano 2014). Abdela et al. (2020) reported that co-inoculation of Mesorhizobium ciceri and Pseudomonas fluorescens increased the RWC in chickpea by 22.2% over the uninoculated drought stressed control.
The present study demonstrated the ability of Bacillus sp. BT3, Klebsiella sp. HA9 and BioNPK inoculation to promote plant growth by improving root architecture. Root system architecture is most important traits of plants among the many adaptive traits for enduring drought stress (Huang and Gao 2000;Huang et al. 2014). Individual inoculation of Bacillus sp. BT3 or Klebsiella sp. HA9 increased the root length by 2.0 folds over the uninoculated stressed (30% FC) control. Bacillus sp. BT3 and Klebsiella sp. HA9 inoculation enhanced the root surface area by 48% and 70% respectively over the uninoculated water stressed (30% FC).). Likewise, number of root tips and number of forks were also enhanced by more than 100%. Under drought stress, an increase in the number of root tips, root hairs and lateral roots not only leads expansion of surface area for uptake, but also improves the root's capillary pressure (Miyahara et al. 2011). Earlier reports suggested that microbial inoculation positively altered root architecture and enhanced the absorption sites for water and nutrients (Gouda et al. 2018;Hosseini et al. 2017;Khan et al. 2020). Jochum et al. (2019) found that inoculation of maize and wheat with Bacillus and Enterobacter significantly enhanced the survivability of seedlings through modification in root architecture under moisture deficit conditions. MDA, a lipid peroxidation by-product, represents the level of oxidative stress to the cell membrane (Gontia-Mishra et al. 2016). In our present investigation, wheat plants treated with Klebsiella sp. HA9 and BioNPK under water stress had remarkably low MDA content in roots and shoots, suggesting that inoculation of beneficial microbes shielded plant cellular homeostasis against the deleterious impacts of water stress. Our findings are in consistent with previous studies indicating that microbial inoculation counters oxidative stress imposed by a lack of water (Gontia-Mishraet al. 2016;Tiwari et al. 2016). Osmoprotective substances are marked biochemical signals of plant stress tolerance. Osmoprotectants preserve cell turgidity under drought stress and protect the plant from oxidative stress by adjusting the plant cell's osmotic potential (Ullah et al. 2017;Wang et al. 2019). The results of our study revealed that microbial inoculation significantly influenced the amount of osmoprotectants. For example-inoculation of Bacillus sp. BT3 and Klebsiella sp. HA9 decreased the glycine betaine and proline content in leaves by over 40% over the uninoculated water stressed (30% FC) control. It was obvious that plant growth promoting bacteria treated plants were not subjected to as much drought stress, consequently, less proline and glycine betaine were accumulated in the presence of beneficial bacteria. Gontia-Mishra et al. (2016) reported that protein and sugars concentration were enhanced in wheat tissues under water deficit condition. However, microbial inoculation significantly decreased the proline concentration in wheat plants over the uninoculated wheat plants under water stress. Sheteiwy et al. (2021b) recently investigated that soybean plants inoculated with Bacillus amyloliquifaciens and/or arbuscular mycorhiza exhibited higher α-amylase and β-amylase activities and induced expression of protein synthesising genes under water deficit, resulting in more soluble sugars and protein accumulation than uninoculated control plants. Grover et al. (2014) and Tiwari et al. (2016) also found similar results in their respective research. Many researchers believe that inoculating plants with plant growth promoting microbes (PGPM) can improve water potential of plant cells by increasing total soluble sugar, protein, glycine 81 Page 14 of 18 betaine, and proline content (Asghari et al. 2020). This however can be due to different plant genotypes and microbial species.
Our results revealed that uninoculated water stressed (30% FC) control wheat plants had higher levels of antioxidant enzymes including POD, SOD, APX, CAT, and GR than uninoculated un-stressed (50% FC) control wheat plants. Similar results were also found by Kaushal and Wani (2016), Mishra et al. (2020) and Tiwari et al. (2016). Inoculation of Bacillus sp. BT3 and Klebsiella sp. HA9 significantly enhanced the concentration of antioxidant enzymes (SOD, APX, POD, GR and CAT) in wheat leaves under water stress (30% FC) over the uninoculated water stressed (30% FC) plants. Results of antioxidant assay showed higher reactive oxygen species (ROS) production in uninoculated water stressed (30% FC) plants as compared to uninoculated un-stressed (50% FC) control plants. However, it was decreased in the presence of Bacillus sp. BT3, Klebsiella sp. HA9 and BioNPK. The application of microbes under drought stress conditions may increase antioxidant enzymes by upregulating their related genes. Sheteiwy et al. (2021a) discovered that application of arbuscullar mycorrhiza and/ or Bradyrhizobium japonicum to drought-stressed soybean plants enhanced the relative expression of the CAT and POD genes. Thus, the current study revealed that the use of microorganisms may have the potential to boost the antioxidant defence system and prevent oxidative damage under drought stress conditions. Similarly, Yaseen et al. (2020) investigated that inoculation of Pseudomonas moraviensis significantly increased the APX and CAT accumulation in wheat leaves over the uninoculated water stressed wheat plants. Batool et al. (2020) reported that Bacillus subtilis HAS31 inoculation with potato plants notably enhanced CAT, POD and SOD accumulation over the uninoculated control plants under water stress. Our findings are also consistent with previous reports on the impact of plant growth promoting rhizobacteria in modulating antioxidant enzymes in rice, maize, and tomato and improving crop plant drought tolerance (Haddidi et al. 2020;Narayanasamy et al. 2020;Sood et al. 2020;Tsai et al. 2020). Our histo-chemical staining results revealed that microbial inoculation decreased the accumulation of peroxide and superoxide radicals in wheat leaves under water stress which further supports the higher levels of ROS scavenging enzymes like SOD and POD. Our results are in compliance with the observation by Gontia-Mishra et al. (2016) who found that microbial inoculation significantly decreased H 2 O 2 accumulation in wheat seedlings under drought stress. It could be because inoculated plants sense less water stress or accumulate more antioxidant enzymes.
Furthermore, we also analysed the expression of drought stress associated genes (DHN, DREB, L15 and TaABA-8OH genes) in wheat plants under water stress either with or without microbial inoculation. Our results revealed that microbial inoculation enhanced the relative expression levels of DHN, DREB, L15 and TaABA-8OH genes over the uninoculated water stressed (30% FC) control. Microbial mediated upregulation of stress responsive genes (DHN, DREB, L15 and TaABA-8OH) might have helped wheat plants in tolerating the moisture deficit. PGPM have recently been identified to mediate drought tolerance in many crop plants via the induction of various genes linked to abiotic stresses (Kasim et al. 2013;Nautiyal et al. 2013;Naveed et al. 2014;Sarma and Saikia 2014;Saakre et al. 2017). Wu et al. (2021) investigated that application of Azospirillium brasilense and Bacillus amyloliquefaciens alleviated the moisture deficit in wheat crop by up-regulating the water stress associated genes such as APX1, HSP17.8 and SAMS1 over the uninoculated control under moisture deficit. Ahmad et al. (2019) observed that inoculation of Pseudomonas fluorescens enhanced the expression level of DHN1 gene of maize over the uninoculated control during prolonged stress at 6 days after sowing. Singh et al. (2020) reported that Pseudomonas fluorescens OKC and Trichoderma asperellum T42 inoculation with rice plants enhanced expression levels of stressresponsive genes including OSPiP1, DHN, and DREB over the uninoculated control plants under drought stress.
Dehydrins (DHN) store water molecules, neutralize reactive oxygen species, and bind to DNA, RNA, protein, and lipid to preserve plant biological activity in drought stress (Liu et al. 2017). Dehydration-responsive element binding proteins (DREB) activate drought-tolerant genes in plants (Agarwal et al. 2006). DREB genes also regulate abiotic stress-associated gene expression in ABA-independent stress-tolerance pathways (Stockinger et al. 1997). The TaABA-8OH gene is implicated in abscisic acid catabolism, which converts ABA to 8'-hydroxy-ABA. Subsequently, the 8′-hydroxy ABA is transformed to phaseic acid (Kushiro et al. 2004).This particular gene significantly contributes towards ABA catabolism and especially to reduce the endogenous levels of ABA promptly after the dehydration stress is removed (Umezawa et al. 2006). Over expression of the TaABA-8OHgene may encode a higher amount of ABA-8'-hydroxylase, which reduced endogenous ABA content in plants and improved the drought stress response of wheat plants. Cura' et al. (2017) discovered that inoculating maize plants with Herbaspirillum seropedicae reduced ABA content by approximately 70% as compared to the uninoculated stressed control and improved maize drought stress response. Drought stress induces genes that produce essential metabolic proteins and regulate downstream signal transduction pathways. For example-DREB gene overexpression increases downstream target gene expression, including dehydrins and COR protein genes (Lata and Prasad 2011). From the above study, it can be concluded that BT3, HA9 and BioNPK could be attributed towards decreasing the level Page 15 of 18 81 of drought stress, improved root architecture (root length, surface area, number of root tips), induced drought tolerance along with modulation of osmoprotective substances and antioxidant enzymes.

Conclusion
We revealed that wheat plants are stunted by drought stress, as assessed by a decrease in shoot length, fresh weight, and dry weight of roots and shoots. Inoculation of plant growth promoting bacteria helped wheat plants to boost their growth under drought stress through multifarious mechanisms. Bacillus sp. BT3 and Klebsiella sp. HA9 along with check strain significantly enhanced the photosynthetic pigments and relative water content as compared to uninoculated water stressed control plants. Total root length, number of root tips, and number of forks in wheat plants decreased significantly under water stressed (30% FC) conditions when compared to the non-stressed uninoculated treatment. Microbial inoculations significantly increased total root length, number of root tips, number of links and surface area over the uninoculated water stressed (30% FC) control. Plants treated with Bacillus sp. BT3 and Klebsiella sp. HA9 did not experience as much drought stress as uninoculated water stressed control plants, and as a result, less proline and glycine betaine were accumulated in wheat plants. Under drought stress, we observed a hyper-accumulation of superoxide and peroxide radicals in wheat leaves. Wheat plants inoculated with microbes minimised tissue damage from high superoxide and peroxide levels by enhancing the buildup of the antioxidant enzymes SOD, POD, APX, and CAT. Moreover, the microbial inoculation also enhanced the relative expression of DHN, DREB, L15 and TaABA-8OH which contributed towards activation stress responsive mechanisms and ABA homeostasis. Our results provided a comprehensive overview on the microbe mediated drought stress alleviation of wheat. Thus, the use of these microorganisms can be recommended as efficient bioinoculants to alleviate water stress in wheat plants.