Silencing of sterol glycosyltransferases modulates the withanolide biosynthesis and leads to compromised basal immunity of Withania somnifera

Sterol glycosyltransferases (SGTs) catalyse transfer of glycon moiety to sterols and their related compounds to produce diverse glyco-conjugates or steryl glycosides with different biological and pharmacological activities. Functional studies of SGTs from Withania somnifera indicated their role in abiotic stresses but details about role under biotic stress are still unknown. Here, we have elucidated the function of SGTs by silencing SGTL1, SGTL2 and SGTL4 in Withania somnifera. Down-regulation of SGTs by artificial miRNAs led to the enhanced accumulation of withanolide A, withaferin A, sitosterol, stigmasterol and decreased content of withanoside V in Virus Induced Gene Silencing (VIGS) lines. This was further correlated with increased expression of WsHMGR, WsDXR, WsFPPS, WsCYP710A1, WsSTE1 and WsDWF5 genes, involved in withanolide biosynthesis. These variations of withanolide concentrations in silenced lines resulted in pathogen susceptibility as compared to control plants. The infection of Alternaria alternata causes increased salicylic acid, callose deposition, superoxide dismutase and H2O2 in aMIR-VIGS lines. The expression of biotic stress related genes, namely, WsPR1, WsDFS, WsSPI and WsPR10 were also enhanced in aMIR-VIGS lines in time dependent manner. Taken together, our observations revealed that a positive feedback regulation of withanolide biosynthesis occurred by silencing of SGTLs which resulted in reduced biotic tolerance.

Scientific RepoRts | 6:25562 | DOI: 10.1038/srep25562 of specific glycosyltransferases (GTs), which sequentially transfer the monosaccharide moieties of their activated sugar donor (usually nucleotide donor) to the required acceptor such as lipids and proteins resulting in the formation of a glycosidic bond 9,10 .
Sterol glycosyltransferases (SGTs) in plants catalyze the transfer of glycon moieties to the sterols and their related compounds to generate steryl glycosides (SGs) or saponins. These glyco-conjugated sterols play very important role in defense of the plant against abiotic stresses and plant pathogen interactions 11 . The saponins produced by oats and tomato have been studied to their potential role in the defense of plants against phytopathogenic fungi 12 . SGs are membrane associated sterols and comprised of a sugar moiety attached to the hydroxyl (-OH) group at C-3 position of a sterol molecule 13 . The hydroxyl group at C-3 position is the most preferred one by SGTs for the enzymatic catalysis, followed by 27β -OH position or -OH group present at the side chain of a modified sterol backbone 14,15 . SGTs have been reported to occur as membrane associated as well as cytosolic. In addition, SGTs glycosylate steroidal hormones which function as growth and development regulator in plants 16,17 . In Solanum aculeatissimum, the accumulation of SaGT4A gene for the glycosylation of steroidal sapogenins in response to wounding stress indicates their role in plant defense 18 . WsSGTL1, one of the members of WsSGTs, has been examined to be involved in response to different abiotic stresses and insect resistance in heterologous expression systems 2 . However, extensive analysis of the role of SGTs in the glycosylation of terpenoids and their effect in the plant basal immunity is still lacking.
Virus-induced gene silencing (VIGS) is a quick method for functional analysis by knocking down the gene expression without the need of genetically transforms the plants [19][20][21][22] . Conventional VIGS assays initiate with a large fragment of gene which was converted and modified into small RNAs by the endogenous siRNA-based machinery causing off target gene silencing of the plant. Single artificial miRNA can provide better specificity by minimizing off-target effects 23 . We have developed artificial miRNA and VIGS (aMIR-VIGS) system for the functional characterization of WsSGTL genes under biotic stress. In the present study, an efficient protocol has been developed for the down-regulation of phytoene desaturase (PDS) gene by a combination of aMIR-VIGS system. We have applied this aMIR-VIGS system for the SGTLs members down-regulation of W. somnifera. Silenced lines led to significant accumulation of withanolides in leaves as compared to control plants. In addition, SGTLs down-regulation of W. somnifera affected the early gene transcript of the MVA and MEP pathway. Subsequently, the silneced lines lost their immunity and became susceptible to Alternaria alternata infection.

Development of efficient aMIR-VIGS constructs.
To develop aMIR-VIGS system for W. somnifera, we primarily targeted the PDS gene by designing amiRNA primer from the conserved region of the gene from closely related species of Solanaceae family ( Supplementary Fig. 1a). The amipds construct was prepared by PCR based mutagenesis of miRNA159a of Arabidopsis thaliana and cloned into VIGS vector ( Supplementary Fig. 1b) Syringe infiltration of this amipds-VIGS construct into W. somnifera plants developed bleaching (photobleaching) in the systemic leaves 15 to 20 days post inoculation (DPI) due to PDS gene silencing (Fig. 1a). First, the bleached regions were restricted to the veins of the leaves, later the symptoms extended to most of the leaf tissues 24 . The positive lines were also confirmed by coat protein specific primers (Fig. 1b). We have checked the level of PDS, by qRT-PCR analysis (Fig. 1c) which concludes that PDS mRNA expression was 75 to 90% less in the systemic tissue of amipds-VIGS lines than in control plants (empty vector).
Silencing of sterol glycosyltransferases. In order to investigate the role of SGTLs in W. somnifera, initially three different amiRNAs were designed from conserved sequences of WsSGTL1, WsSGTL2 and WsSGTL4, named as, 2misgt, 4misgt and 6misgt ( Supplementary Fig. 2). To examine their silencing efficiency, amiRNAs were first cloned into pBI121 ( Supplementary Fig. 3a) and then transformed into leaves of 3-weeks-old plants ( Supplementary Fig. 3b). After 48 h and 72 h, qRT-PCR of 2misgt-pBI121 and 4misgt-pBI121 showed better silencing efficiency against members of SGTLs expression than 6misgt-pBI121 ( Supplementary Fig. 3c,d). Further, 2misgt and 4misgt amiRNAs were utilized to clone into VIGS vector for the development of aMIR-VIGS system named as 2misgt-VIGS and 4misgt-VIGS ( Supplementary Fig. 4a,b). The aMIR-VIGS construct of WsSGTLs members were transformed into leaves of 3-weeks-old plants. After 4 weeks of transformation, positive aMIR-VIGS lines were selected through PCR of CP specific primers ( Supplementary Fig. 4c). The qRT-PCR were performed by using SGTL1, SGTL2 and SGTL4 gene specific primers and observed that their transcript level was decreased in the aMIR-VIGS plants. Interestingly, both aMIR-VIGS constructs showed almost same level of down-regulation of SGTLs members (Fig. 2a,b) as compared to control plants.
We have performed comparative nonradioactive sterol glycosyltransferase assay in the aMIR-VIGS lines of W. somnifera 25 and control plants. Stigmasterol and solanidine were taken as acceptor molecule which have free 3β -OH moiety group. Accepter molecules were modified by donor substrate, uridine 5′-diphosphoglucose (UDP-glucose) with the help of SGT enzyme. This method takes advantage of a specific phosphatase to remove inorganic phosphate quantitatively from the leaving nucleotide diphosphate, such as UDP or GDP, of glycosyltransferase reactions. Total phosphate output/well in the protein reaction of 2misgts-VIGS and 4misgts-VIGS lines decreased significantly (upto 90%) as compared to control plants (Fig. 2c,d). This confirms that due to down-regulation of members of WsSGTs, silenced lines posses lesser SGT activity due to decreased expression of the genes.
Analysis was carried out to study whether the down-regulation of members of WsSGTL gene family affected the expression of genes involved in intermediate steps of the withanolide biosynthesis. To study this, transcript level of WsHMGR, WsDXR, WsFPPS, WsCYP710A1, WsDWF5 and WsSTE1 genes involved in MVA pathway were analysed by qRT-PCR in the aMIR-VIGS lines and control plants ( Supplementary Fig. 7). There was a significant up-regulation (approx 3 fold) of WsCYP710A1 expression which might be responsible for the increase in the concentration of sitosterol and stigmasterol in silenced lines. Up-regulation of WsSTE1 (approx 3 fold) and WsDWF5 (approx 3.5 fold) was observed which might be responsible for increased withanolide A and withaferin A content in silenced lines. The expression of intermediate genes, WsHMGR, WsDXR, WsFPPS in aMIR-VIGS lines were also up-regulated upto 2.8, 10 and 2.9 fold, respectively. Our metabolite content and gene expression analysis conclude that down-regulation of SGTLs modulates the accumulation of withanolides by increasing expression of genes involved in MVA pathway (Fig. 4).
Silencing of WsSGTLs affect growth and development of plant. The down-regulation of WsSGTLs resulted delay in the normal growth of aMIR-VIGS lines. After 3 weeks of transformation, plant height and leaf area of the silenced lines was measured. The height of the aMIR-VIGS lines was significantly shorter than control plants. This difference was maintained even after 10 weeks of growth (Fig. 5a,b), however, there was no change in the number of nodes and internodes. Total leaf area of the silenced lines was lesser than the control plants ( Fig. 5c,d). These results suggest that the ratio of free sterol and its glycosylated forms may work as signalling molecules and regulate the growth of the plants. For the confirmation of this hypothesis, we have measured the ratio of free sitosterols and stigmasterol vs its glycosylated forms by acid hydrolysis 26 . The ratio of free phytosterol vs glycosylated phytosterols was higher in silenced lines than control plants (Fig. 5e,f). This experiment gave us indication of sterol imbalance in VIGS lines of W. somnifera and might be one of the reasons of shorting the height and leaf area of plants.  (Fig. 6b) and chlorosis in the leaves of 2misgt-VIGS and 4misgt-VIGS lines was much higher than control plants. The extracts isolated from the infected leaves were used for total colony forming unit count (cfu) which was significantly higher in silenced lines than control plants (Fig. 6c).
Furthermore, callose deposition in the leaves of infected plants is another important parameter to check disease susceptibility against the biotic stress. After 7DPI of A. alternata infection, the leaf of the silenced lines and control plants was stained with aniline blue. Under confocal microscope, the higher amount of callose deposition was observed in the leaves of 2misgt-VIGS and 4misgt-VIGS lines as compared to the leaves of control plants (Fig. 6d). In addition to evaluate A. alternata induced disease symptoms, microscopic studies were conducted in down-regulated lines for cellular and intercellular reactive oxygen intermediates (ROIs). Hydrogen peroxide (H 2 O 2 ) was detected by non-fluorescent (3, 3′-diaminobenzidine [DAB]) staining method. Microscopic analysis after 48 h infection revealed a strong dark brown precipitation in the leaves of 2misgt-VIGS and 4misgt-VIGS lines whereas no such precipitation was observed in the leaves of control plants (Fig. 6e). This was further accomplished by quantitative analysis of H 2 O 2 where it increased (upto 1.8 fold) in the leaves of aMIR-VIGS lines after 48 h of fungal infection (Fig. 6f). We also quantified SOD level at 0 h and after 48 h post A. alternata infection in 2misgt-VIGS, 4misgt-VIGS lines and control plants. At 0 h, no significant differences were observed in the level of SOD, whereas, SOD level significantly increased upto 1.1, 2.7 and 2.1 for EV, 2misgt-VIGS and 4misgt-VIGS lines, respectively after 48 h of A. alternata infection, (Fig. 6g). The ratio of stigmasterol/sitosterol also maintains the innate immunity by increasing the disease susceptibility of the plant leaves. We have measured the stigmasterol/ sitosterol ratio in plant before and after 7 days of A. alternata infection. After 7 days of stress, the concentration of sitosterol was decreased whereas, stigmasterol concentration increases in silenced lines and control plants. Although, stigmasterol/sitosterol ratio were higher in silenced lines than control plant (Fig. 6h). These results indicated that after down-regulation of members of WsSGTLs, the plants lose their basal immunity and became more prone to infection with A. alternata.

Salicylic acid level increases the expression of defence related genes in silenced lines.
To investigate the contribution of defense related genes against A. alternata infection in silenced lines, real time expression analysis of some important genes, namely, WsPR1, WsDFS, WsSPI and WsPR10 was performed on the basis of earlier published transcriptome analysis 27 . The maximum enhancement in expression of WsPR1 was found upto 406 to 833 fold after 72 h of fungus infection in 2misgt-VIGS and 4misgt-VIGS respectively, whereas, upto 83 fold in EV as compared to mock plant (Fig. 7a[A] and 7b[A]). The increase in the expression level of WsDFS and WsSPI was also similar to the expression level of WsPR1 (Fig. 7a[B-C] and 7b[B-C]). Finally, it was concluded that after A. alternata infection, the expression of defense related genes in silenced lines of W. somnifera was highly up-regulated as compared to control plants.
The salicylic acid (SA) signalling cascades are the master regulator of defense process in plants which control the gene expression of large number of genes. Modulation of defense related gene expression encouraged us to study the possible role of SA. To know this, we measured SA levels before and after 7 days of A. alternata infection in aMIR-VIGS lines and control plants. Before the infection of A. alternata, no significant difference in the level of SA was observed in silenced lines and control plants. However after infection with A. alternata, SA level increased upto 3 fold in 2misgt-VIGS and 4misgt-VIGS lines as compared to control (Fig. 7c, Supplementary Fig. 10a-d).
Altogether, these observations indicated that increased level of SA was responsible for modulation of defence related gene expressions which ultimately altered the basal disease resistance in aMIR-VIGS lines.

Discussion
W. somnifera is a medicinal plant rich in sterols, sterol glycosides and steroidal lactones which play a crucial role in adaptation of the plants to stress conditions 28 . In our previous studies, we reported that over-expression of WsSGTL1 in heterologous systems enhances germination, salt tolerance, heat tolerance, cold tolerance and insect resistance as compared to control plants 2,28 . In the present report, silencing of members of WsSGTs was done using the amiRNA technology 29 . However, Virus-induced gene silencing (VIGS) is also a quick method for knocking down the gene expression 21 . We designed artificial miRNA and developed efficient aMIR-VIGS system for the functional analysis of SGTLs members in W. somnifera. The gene silencing by VIGS within a plant is highly dependent on the environmental factors and method of inoculation [30][31][32] . The marker gene used for the initiation of virus symptoms is GFP or PDS 24,33 . Gene silencing of PDS gene of W. somnifera through amipds-VIGS causes suppression of carotenoid biosynthesis where the affected plants became susceptible to photobleaching 20,24,34,35 . This protocol of aMIR-VIGS system has been used for the down-regulation of WsSGTs members to obtain a comprehensive understanding of their role in plants.
The members of WsSGTL are important modifying enzymes which participate in isoprenoid biosynthetic pathway, specifically in the glycosylation of phytosterols and withanolides. We have developed silenced lines of W. somnifera using aMIR-VIGS system. The down-regulation of members of WsSGTs was confirmed by comparative qRT-PCR analysis and SGT enzymatic assay of aMIR-VIGS lines and control plants. WsSGTLs down-regulated lines accumulated high amount of withanolide A, withaferin A, sitosterol and stigmasterol, whereas, as expected less amount of withanoside V in the leaves of silenced plants. It has been already reported that over-expression of glycosyltransferase genes leads to a significant increase in their respective glucosides 36 ,. Recently, it was cleared that after over-expression of WsSGTL1 gene in homologous system increased the production of glycowithanolides 37 . To develop a better understanding of the effect of down-regulation of WsSGTLs, qRT-PCR of WsHMGR, WsDXR, WsFPPS, WsCYP710A1, WsSTE1 and WsDWF5 (genes of MVA pathway) was performed. From this analysis, it was concluded that WsSGTLs down-regulation positively affected the upstream genes of MVA pathway (Fig. 4, Supplementary Fig. 7). Recently, it was published that WsSQS leads to positive feedback and negative feed-forward regulation of sterol biosynthetic genes. Down-regulation of WsSQS significantly modulates the expression of MVA pathway genes located upstream and downstream of SQS 5 . From these results, we could conclude that the accumulation of withanolides and increased expression of biosynthetic pathway genes were positively affected by the down-regulation of members of WsSGTLs.
In addition, WsSGTLs down-regulated lines also showed phenotypic changes which suggested that WsSGTLs activity may be affecting the growth hormone signalling pathway in plants. In the silenced lines, height and leaf area were significantly shortened as compared to control plants (Fig. 5a-d). It was already proved that one members of WsSGTL gene family i.e., WsSGTL1 over-expression causes the alteration in phenotype of the plants such as enhanced growth and expansion of leaves 37 . Many reports suggested that glycosylation of important hormones, brassinosteroids and strigolactone was being carried out by the members of SGTs in plants 13,38 . The SDG8i glucosyltransferase and UGT74D1 were capable of influencing the biological activity of plant hormones such as strigolactone -oxindole-3-acetic acid (OxIAA) via glycosylation 38 . Gene mutants of UGT80B1and SMT1 have revealed roles of sterols modifications in embryo, vein, shoot and root patterning, cell expansion, polarity and proliferation, fertility, cellulose level maintenance, gravitropism, and hormone signalling 39,40 . After HPLC analysis of silenced lines it was concluded that the ratio of free phytosterol and their respective glycoside higher than control plants (Fig. 5e,f). It may be concluded that WsSGTLs down-regulation causes imbalance of glycosylated and nonglycosylated sterol which results in phenotypic changes of silenced lines. Besides this, the down-regulation might be affecting the signalling pathway of hormones which hinders the growth of plants.
Earlier reports showed the role of WsSGTL1 in response to abiotic stresses in heterologous system 2,28 but no report is available regarding biotic stress. It was already reported that expression of SGTs were differentially modulated in different organs in response to external stimuli such as treatment of SA and methyl jasmonate in W. somnifera 41 , suggesting their role in defense. Here, we report that the down-regulation of WsSGTLs leads to increased susceptibility of plants against the infection of A. alternata.
Necrotrophic fungus, A. alternata, produces an AAL toxin (TA and TB) that causes necrosis in more than 200 species of Solanaceae family 42,43 . In biotic stress experiments, the leaves of silenced lines have 2-3 fold higher total cfu mL −1 count than control plant leaves. Callose also plays an important role in plant defense during interactions with pathogens. It rapidly synthesized and deposited just beneath the sites of attempted pathogen penetration [44][45][46] . The susceptibility was further evident by the massive callose deposition in the leaves of aMIR-VIGS lines as compared to control plants after one week of A. alternata infection. Further, we quantified H 2 O 2 and SOD accumulation in silenced lines and control plants after the infection. It was reported earlier that SOD and H 2 O 2 production plays a crucial role in restricting the growth of pathogen during hypersensitive response [47][48][49][50][51] . In the present study, WsSGTLs silenced lines accumulated high amount of H 2 O 2 and SOD after the fungus infection. Several reports are available which emphasizes that exposure of stress resulted in increase in SOD activity due to increase in the ROS and H 2 O 2 levels 52,53 . Since the high SOD level causes the high rate of H 2 O 2 production in cell 54 . After pathogenic infection, plants induced array of metabolites change that potentially contributed in the susceptibility of the plants against pathogen infection. The conversion of sitosterol to stigmasterol is one of the significant marker of susceptibility against pathogenic infection and stigmasterol/sitosterol ratio become rise 55,56 . After the A. alternata infection, the conversion of sitosterol to stigmasterol was reported in the WsSGTLs silenced lines and control plants. This conversion was little high in the aMIR-VIGS lines than control plants. Resultant, the C22 desaturation of sterol was high in the silenced plant membrane after fungal infection and plants showed more disease susceptibility.
The level of SA was estimated before and after A. alternata infection in WsSGTLs silenced lines and control plants. SA plays important role in plant defense and involved in the activation of defense responses against biotrophic and hemi-biotrophic pathogens as well as in the establishment of systemic acquired resistance (SAR) 57,58 . It was also reported that programmed cell death was directly induced by level of SA 58 . This was one of the reasons of nercosis and callose deposition in the leaves of silenced lines in our study. Several previous reports demonstrated that the level of SA induced the expression level of PR proteins and other defense related genes 59-62 . To check this hypothesis, qRT-PCR of WsPR1, WsDFS, WsSPI and WsPR10 was performed in silenced and control plants and compared with the mock plants. Expression analysis showed up-regulation of these genes in time dependent manner 63,64 .
Finally, it may be concluded that after down-regulation of SGTL1, SGTL2 and SGTL4 genes of W. somnifera, withanolide contents were increased, whereas withanoside content was decreased. Also, this modulation was analysed with the increase in the expression of some important intermediate genes of MVA pathway. Silencing of WsSGTL members, modulation was observed in the free and glycosylated phytosterols which made the phenotypic variations in plants. Due to this modulation of withanolides,withanosides and ratio of nonglycosylated vs glycosylated phytosterols the plants might have lost their first line of defense. Further, the infection of A. alternata increased levels of SOD, H 2 O 2 and SA significantly in down-regulated lines as compared to control plants. Higher level of SA also modulated the expression of WsPR1, WsDFS, WsSPI and WsPR10 genes of plant defense system. Altogether, we could demonstrate that the members of WsSGTLs play a significant role in maintaining the metabolic balance of the cell which facilitated the proper growth and defense system of the plants.

Methods
Preparation of artifitial miRNA. Expression of amiRNAs requires a mature miRNA backbone sequence, of which the stem region was replaced with desired sequences. A 545-bp fragment containing the entire sequence of the Arabidopsis miR159a was cloned into pTZ57R/T vector by PCR amplification using primers miR159-F1 (5′-ATATCTCCTTCATAGCTCTAATG-3′), miR159R1 (5′-AAATAACACGCTAAACATTGCTTCG-3′) and utilized for mutagenesis to yield amiRNA for WsPDS and members of WsSGTLs using oligonucleotide primers. For the preparation of these oligonucleotide primers, PDS sequences of closely related species of Solanaceae family and SGTLs sequences of W. somnifera were downloaded from National Centre for Biotechnology Information (NCBI) database. 21 bp conserve region for PDS and WsSGTLs were selected by multalin online tool ( Supplementary Figs 1 and 2). Three conserve sequences of WsSGTLs 2misgt, 4misgt, 6misgt and one PDS named amipds (Supplementary Table 1) were selected on the basis of their specificity for the target. Three constructs have been prepared in pBI121 by using pTZ-miRNA159a as template named 2misgt-pBI121, 4misgt-pBI121 and  Fig. 4a,b) and grown on Luria-Bertani agar medium (50 mg mL −1 kanamycin, 50 mg mL −1 gentamycin and 25 mg mL −1 rifampicin). Two days before infiltration, 5-mL primary cultures of Agrobacterium strains were inoculated from single colonies on plates and grown for 16 h at 28 °C. 20 mL secondary culture was inoculated from the primary culture and grown for 5-6 h for obtaining OD 600 0.4. The culture was harvested by centrifugation at 3000 × g at room temperature for 5 min. Cell pellet was resuspended in same volume of induction buffer, 10 mM MES (pH 5.5), 200 μ M acetosyringone and incubated for 3-4 h at 28 °C at 80 rpm. Then pallet the culture again at 3000 × g and resuspended in the infiltration buffer containing 5 mM MES (pH 5.5) and OD 600 was adjusted to 0.8-0.9. TRV1 + GOI: TRV2 cultures were mixed at 1:1 (v/v) ratio. A needleless syringe was used to inoculate the abaxial side of the lower leaves with TRV1 + GOI::TRV2 mixture of Agrobacterium culture. Infiltration of TRV1::TRV2 in the plants served as control (empty vector). The inoculated plants in the greenhouse were maintained at 21 ± 2 °C until the observation period was achieved.

RNA isolation and RT-PCR.
Total RNA was extracted from upper leaves of 3-weeks-infiltered plants by using the RNA isolation kit (Sigma-Aldrich) and treated with RNase-free DNase I (Ambion). First-strand complementary DNA was synthesized using 5 μ g of total RNA with oligo (dT) primer (Fermentas). The qRT-PCR was done in the StepOnePlus Real-Time PCR System (Applied Biosystems 7500). Primers used for qRT-PCR are listed in Supplementary Table S2. Three independent biological replicates and for each biological replicate, two technical replicates were analyzed by qRT-PCR analysis.
SGT assay. The SGT activity was measured by the Glycosyltransferase activity kit (R&D system/USA). The assay was carried out by detection of released inorganic phosphate using Malachite Green reagent 25  Biochemical Analysis by HPLC. The extraction procedure of withanolide (withanolide A and withaferin A) and glycowithanolide (withanoside V) has been followed by the protocol 66 . Qualitative and quantitative analysis of withanolides and withanoside were performed by HPLC-PDA with a Shimadzu (Japan) LC-10A system comprising LC-10AT dual-pump system, a SPD-10A PDA detector (operated at 227 nm), and Rheodyne injection valve with 20-μ L sample loop. Compounds were separated on a RP-C18 column (Merck) (4.6 mm × 250 mm, 5-μ m pore size) protected by a guard column containing the same packing. 1 mg mL −1 stock solutions of withaferin A, withanolide A and withanoside V (Sigma) were subsequently diluted to prepare solutions with concentrations in the range of 0.5 mg mL −1 to 50 mg mL −1 for preparation of standard curve. The sitosterol and stigmasterol estimation was carried out by following the protocol 2 . Ratio of these phytosterols of the plant leaves were checked by before and after acid hydrolysis by HCL 26 . For the quantification of SA, 200 mg sample was crushed into 1 mL of methanol and the volume was made to 2 mL. Mobile phase prepared from 20 mM phosphoric acid (A) (pH 2.0 adjusted) and acetonitrile (B) added in 50:50 ratio (A:B). Quantification was performed at 220 nm with PDA detector at 25 °C column temperature and 20 min RT with the flow rate of 1 mL min −1 .
Biotic stress assay with A. alternata. Leaves of W. somnifera were surface sterilized with 1% NaOCl (v/v) and placed at blotting paper. The leaves were transferred into petriplates containing 0.8% agar. For pathogenic assay, 0.5% gelatine was freshly prepared to which 7-10 days old fungal spores were suspended to make the final fungal spore concentration of 10 6 cfu mL −1 . Each leaf was treated with 5 μ l of spore suspension at four different locations on the adaxial surface of the leaves. Control and pathogen infected leaves were kept at 25 °C under 70% relative humidity and 16 h of diurnal light. After 7 days of infection, leaves were sampled for measurement of lesion diameter and for total spore count.
Callose staining. For callose deposition assay, spore suspension sprayed on the leaves of control and transgenic plants with the help of an atomizer and pots were kept in moist chamber. Callose staining were performed after 5-7 DPI of incubation 67 . The infected leaves were incubated in 5 mL buffer [1 vol (water saturated phenol: glycerol: lactic acid: water in 1:1:1:1) + 2 vol ethanol] at 65 °C temperature for 15-30 min. The leaves were transferred into fresh buffer for 25 h, rinsed 3 times each in ethanol and ddH 2 O. Stained in 150 mM K 2 HPO 4 (pH 9.5) containing 0.01% aniline blue for 30 min, four washing was done in buffer (10 min each). The stained leaves were imaged under confocal microscope (Carl Zeiss' LSM 510 META) with 20x Plan-apochromate lens. The excitation was done using diode laser 405 nm and emission was collected using band pass filter 470-500 nm.
Scientific RepoRts | 6:25562 | DOI: 10.1038/srep25562 DAB staining. For in situ detection of H 2 O 2 DAB staining was carried out using leaves of transgenic and control plants 68 . The sample were collected after 48 h of fungal spray and infiltrated under gentle vacuum with 1 mg mL −1 DAB containing 0.05% v/v Tween 20 and 10 mM sodium phosphate buffer pH 7.0. The reaction was completed within 4-5 h post inoculation after visualizing the brown precipitate at the surface of the leaves. Leaves were fixed and then boiled for 15 min in ethanol: acetic acid: glycerol (3:1:1). Bleaching solution was replaced and leaves were incubated until the chlorophyll was completely removed. Leaves were observed by light microscopy under bright field at 4X magnification.
Measurement of H 2 O 2 . The H 2 O 2 content of the leaves was measured spectrophotometrically after reaction with potassium iodide (KI) 69 .The absorbance of the supernatant was measured at 390 nm from the spectrophotometer. The amount of hydrogen peroxide was calculated using a standard curve prepared with known concentrations of H 2 O 2 ( Supplementary Fig. 8a). SOD assay. Superoxide dismutase (SOD, EC 1.15.1.1) activity was measured by nitrotetrazolium (NBT) photochemical assay 70 . The total protein of plants were measured by Bradford 65 method using bovine serum albumin (BSA, Sigma Aldrich, USA) as a standard (Supplementary Fig. 8b).