Molecular characterization of three CYP450 genes reveals their role in withanolides formation and defense in Withania somnifera, the Indian Ginseng

The medicinal properties of Ashwagandha (Withania somnifera) are attributed to triterpenoid steroidal lactones, withanolides, which are proposed to be derived from phytosterol pathway, through the action of cytochrome P450 (CYP450) enzymes. Here, we report the characterization of three transcriptome-mined CYP450 genes (WsCYP749B1, WsCYP76 and WsCYP71B10), which exhibited induced expression in response to methyl jasmonate treatment indicating their role in secondary metabolism. All three WsCYP450s had the highest expression in leaf compared to other tissues. In planta characterization of WsCYP450s through virus induced gene silencing (VIGS) and transient overexpression approaches and subsequent metabolite analysis indicated differential modulation in the accumulation of certain withanolides in W. somnifera leaves. While WsCYP749B1-vigs significantly enhanced withaferin A (~ 450%) and reduced withanolide A (~ 50%), its overexpression drastically led to enhanced withanolide A (> 250%) and withanolide B (> 200%) levels and reduced 12-deoxywithastramonolide (~ 60%). Whereas WsCYP76-vigs led to reduced withanolide A (~ 60%) and its overexpression increased withanolide A (~ 150%) and reduced 12-deoxywithastramonolide (~ 60%). Silencing and overexpression of WsCYP71B10 resulted in significant reduction of withanolide B (~ 50%) and withanolide A (~ 60%), respectively. Further, while VIGS of WsCYP450s negatively affected the expression of pathogenesis-related (PR) genes and compromised tolerance to bacteria P. syringae DC3000, their overexpression in W. somnifera and transgenic tobacco led to improved tolerance to the bacteria. Overall, these results showed that the identified WsCYP450s have a role in one or several steps of withanolides biosynthetic pathway and are involved in conferring tolerance to biotic stress.

www.nature.com/scientificreports/ respectively [6][7][8] . CYP450 enzymes are a large class of heme-thiolate proteins that are ubiquitous in their presence across all genera of organisms and catalyze diverse reactions in pivotal molecular pathways. The CYP450 enzymes have been studied and classified by Nelson, DR (2009) 9 and available on https:// drnel son. uthsc. edu/ Cytoc hrome P450. html and an estimated 80 CYP450s have been assigned biochemical functions related to the plant triterpene metabolism 10 . Efforts have been made to identify and characterize CYP450s in Withania somnifera: Elicitor responsive WsCYP98A and WsCYP76A have been found to express abundantly in stalk and root, respectively, correlating positively to withanolides accumulation 11 . In another study, it was shown that WSCYP93Id protein catalyzed the conversion of withaferin A to an unidentified hydroxylated product 12 . Functional characterization of WsCYP85A69 through miRNA and transient overexpression resulted in modulation of castasterone, stigmasterol and withanolides 13 . Further, it was shown that WsCYP85A69 catalyzed in vitro conversion of 6-deoxocastasterone into castasterone 13 . Despite some of these above studies, the knowledge on the in planta role of CYP450s in withanolides biosynthesis and defense is limited. Here, we present the molecular characterization of three genes encoding CYP450s and show the in planta involvement in withanolides biosynthesis and defense using silencing and overexpression studies.

Materials and methods
All methods were carried out in accordance with relevant guidelines and regulations.
Plant material, elicitor treatment and tissue collection. Seeds of Ashwagandha used in the study were obtained from our institutional gene bank. Ashwagandha variety (Poshita) developed by CSIR-Central Institute of Medicinal and Aromatic Plants (CSIR-CIMAP) was used in this study. Seeds were sown and plants were grown as described in Singh et al. (2015;. Withania somnifera leaves of 3 rd developmental stage were plucked from 10-week-old plants and kept in Petri dishes containing solutions of methyl jasmonate (MeJA; 500 mM) with respective controls. Samples were harvested at different time (h) intervals and stored at −80 °C until further use. For tissue-specific expression analysis berries, flowers, leaves, roots and stems were collected from field-grown W. somnifera plants 14,15 . Phylogenetic analysis. CYP450s belonging to respective families shown to be involved in secondary metabolism were used to construct the phylogenetic tree. Neighborhood joining method was used for tree construction using MEGAX with default setting and bootstrap value changed to 1000 16 . RNA isolation, cDNA synthesis and qRT-PCR analysis. Total RNA isolation, cDNA preparation and qRT-PCR were performed as reported earlier 14,15 . The cDNA was normalized using 18S rRNA as endogenous control. qRT-PCR analysis was carried out with gene-specific primers (Table S1) that were designed outside the gene region used for cloning into pTRV2. The reaction was performed using 2X SYBR green mix (Thermo Scientific, USA) and run in StepOne Real-Time PCR System (Applied Biosystems, USA). The qRT-PCR conditions were: 94 °C for 10 min, followed by 40 cycles of 94 °C for 15 s and 60 °C for 45 s. Fold-change differences in gene expression were analyzed using the comparative cycle threshold (C t ) method.

Generation of VIGS and overexpression constructs.
Plasmids pTRV1 and pTRV2 17 procured from TAIR (www. arabi dopsis. org) were used to generate VIGS constructs. A 500 bp gene-specific fragment was amplified from leaf cDNA using gene-specific primers (Table S1). The amplicon for each gene was cloned into a pGEMT-easy vector and sub-cloned into XbaI and XhoI sites of pTRV2 yielding pTRV2::WsCYP450 constructs (Fig. S1). Next, the open reading frame (ORF) of each gene was PCR amplified using leaf cDNA and specific primers (Table S1). The amplicons were cloned into pJET1.2/vector and confirmed by nucleotide sequencing and subcloned into XbaI and SacI sites of pBI121 vector to form pBI121::WsCYP450 constructs (Fig. S1). The pTRV2-and pBI121-derived constructs were transformed into A. tumefaciens by freeze-thaw method.
VIGS and transient overexpression of WsCYP450s. VIGS of WsCYP450s was performed as described by Singh et al. 14 . Infiltration was performed using a 1:1 ratio of Agrobacteria cultures harboring TRV1 and TRV2 or its derivatives using a 1-mL needleless syringe on the abaxial side of leaves of 4-leaf-staged plants. Thirty days post-infiltration, tissues exhibiting viral infection phenotype were harvested and used for transcript and metabolite analyses. For transient overexpression, overnight grown Agrobacterium culture was pelleted by centrifugation and resuspended in infiltration buffer (50 mM MES, 2 mM Na 3 PO 4 , 0.5% glucose, and 100 µM acetosyringone with pH adjusted to 5.6) to an OD 600 of 0.2 and incubated at 28 °C for 3-4 h. Leaf infiltration was performed as described above into the first pair leaves of 6-8 leaf-staged plants and placed in the dark. The infiltrated leaves were harvested after 48 h for further metabolite and transcript analysis.
Extraction and analysis of withanolides. Withanolides extraction and analysis were performed following Singh et al. 14 . Twenty mg oven-dried (55 °C) leaf tissue was ground using 1-mL absolute methanol, sonicated and the supernatant was collected. The ground tissue was re-extracted twice and obtained methanolic extracts were pooled in a scintillation vial and evaporated. The dried residue was resuspended in 3-mL of 70% methanol. This was extracted thrice with 3-mL chloroform. The lower layer of a chloroform extract (9-mL) was decolorized using charcoal and dried. To the dried samples, 200 µL of 7:3 chloroform: methanol was added. Withanolides were analyzed using HPLC (Model: Nexera, Shimadzu, Kyoto, Japan) fitted with C-18 column as per the program described previously 14  www.nature.com/scientificreports/ area of individual withanolides was determined after normalizing with the peak area of the internal standard, catharanthine (Sigma-Aldrich).
Bacterial growth curve assay. The model plant pathogen P. syringae pv. tomato DC3000 was cultured in 5 mL nutrient broth and grown overnight in an incubator shaker at 28 °C. The culture was centrifuged and the bacterial pellet was resuspended in 10 mM MgCl 2 and the cell density was adjusted to OD 600 = 0.002. Bacterial infiltration and CFU determination were performed according to our earlier reports 14,15 . 3 days after inoculation, leaf discs from infected and buffer infiltrated control plants were collected from the infiltration zone. Later the cfu/cm 2 was determined by plating serial dilutions of leaf extracts on Pseudomonas-specific agar plates after homogenization with 10 mM MgCl 2 .
Generation of transgenic tobacco. Tobacco was transformed using A. tumefaciens harboring pBI121 and pBI121-derived pBI121::CYP450 constructs by leaf disc co-cultivation method as described in Singh et al. 15 . Transformed shoots regenerated on MS (Murashige & Skoog) 18 selection medium supplemented with kanamycin (100 mg/L −1 ) were transferred to ½ MS rooting media. Plantlets with well-developed roots were hardened and PCR-screened positive lines were transferred to the growth chamber. T 1 plants were PCR-screened and positive lines were used for transcript and bacterial growth analyses 15 .
Statistical analysis. The statistical analysis was performed using GraphPad 9.2. The results are represented as mean ± standard error (SE). Error bars denote SE of results obtained from at least three independent experiments. One-way ANOVA (analysis of variance) was performed followed by Tukey's comparison test. The p-value 0.12 was considered as not significant (ns), 0.033 as significant (*), 0.002 as moderately significant (**), and 0.001 as highly significant (***).
Ethical approval. Appropriate permissions for use of Ashwagandha variety Poshita have been obtained from National Gene Bank, CSIR-CIMAP, Lucknow, India.

Results and discussion
Identification of WsCYP450s involved in secondary metabolism. In Ashwagandha, few previous studies have reported molecular and biochemical characterization of some CYP450s and suggested their possible involvement in withanolides biosynthesis. For example, two CYP450s (WsCYP98A and WsCYP76A) were found to be highly expressed in stalk and root, respectively, and their expression was positively correlated with the accumulation of withanolides 11 . Another CYP450, WsCYP93Id was shown to convert withaferin A to an unidentified hydroxylated product under in vitro conditions 12 . In a recent study, in vitro biochemical studies showed that WsCYP85A69 is involved in catalyzing the conversion of 6-deoxocastasterone to castasterone, and in planta studies modulated the accumulation of castasterone, stigmasterol, and withanolides 13 . Despite these studies, continued efforts to identify more genes is needed to characterize CYP450 families to elucidate the highly complex withanolides biosynthesis as the pathway may involve multiple CYP450s. In this study, to identify WsCYP450s possibly involved in withanolides biosynthesis, Withanomics (www. im-crop. snu. ac. kr) database was searched resulting in 474 sequences annotated as CYP450. Screening of these 474 sequences resulted in a subgroup of 159 sequences belonging to CYP71, CYP76, CYP82, CYP90, CYP96, CYP716, and CYP87 families that have been proposed to be involved in the secondary metabolic pathways 19,20 . Among them, 32 sequences showed a significant leaf-specific FPKM (> 10) value. Further, candidates with identical locus and sequence identity were eliminated resulting in 14 gene sequences (Fig. S2). In addition, a search of the in-house generated W. somnifera transcriptome showed 157 sequences annotated as CYP450 in which 107 sequences were annotated to be involved in secondary metabolisms such as sesqui-and triterpene biosynthesis, alkaloid, brassinosteroid, and steroid biosynthesis. Finally, 17 gene sequences remained after removal of identical and partial length candidates, followed by cross -referencing with shortlisted sequences from the Withanomics database (Fig. S2).
Expression profiling of WsCYP450s. Generally, pathway genes involved in secondary metabolism including genes related to withanolides formation are induced in response to phytohormone elicitor MeJA with a corresponding increase of target secondary metabolites 15,21,[21][22][23] . Hence, expression analysis of WsCYP450 genes in leaves of MeJA treated plants was performed. This analysis showed that of the 17 analyzed genes, 8 genes were highly induced upon MeJA treatment. In this study, we have cloned and characterized 3 of the induced CYP450 candidates viz., WsCYP749B1, WsCYP76 and WsCYP71B10. All three genes exhibited induced expression with maximum induction at 6 h upon exposure to MeJA (Fig. 1A). There was about a 14-and16-fold increase in the expression of WsCYP76 and WsCYP749B1, respectively, with an eightfold increase in WsCYP71B10 transcripts. After the peak induction at 6 h, while WsCYP749B1 showed continued induction of ~four-fivefold at 12 h and 24 h post-MeJA treatment and reached basal level expression at 48 h, WsCYP76 exhibited continued induction of sixfold at 12 h reaching basal level of expression thereafter (Fig. 1A). In contrast, WsCYP71B10 exhibited induced expression only at 6 h and its expression remained at basal level in other time points (Fig. 1A). Similar to our results, two CYP450 genes of W. somnifera (WsCYP98A and WsCYP76A) also showed induced expression in response to MeJA treatment with increased withanolide accumulation 11 . However, no in planta or functional studies were reported for these genes. In another study, WsCYP85A69 involved in castasterone formation also exhibited induced expression in response to MeJA and its transient silencing/overexpression in W. somnifera reduced/enhanced castasterone, stigmasterol and withanolides accumulation 24  www.nature.com/scientificreports/ expression of WsCYP749B1, WsCYP76 and WsCYP71B10 in this study also indicates the involvement of these genes in secondary metabolism and possibly in withanolides biosynthesis. Further, it was found that all three WsCYP450s exhibited a similar pattern of expression in different tissues with the highest expression in leaves compared to other tissues. Among the three candidates, WsCYP749B1 showed highest expression in leaves followed by WsCYP71B10 and WsCYP76 (Fig. 1B). Both WsCYP749B1 and WsCYP71B10 transcripts exhibited siginificant differences in their expression pattern among all different tissues, whereas WsCYP76 showed significant differences between only a few tissues. Moreover, with respect to tissues other than leaf, while WsCYP749B1 showed comparatively higher expression in stem, flower and roots with least expression in berries, WsCYP71B10 had higher levels of transcripts in stem followed by berry and In each graph, the tissue having the least Ct was set to 1 to determine the relative abundance of transcripts in other tissues. 18s rRNA was used as an internal reference for normalization. Error bars show mean ± SE of three independent experiments. Statistical significance represented as *p < 0.033, **p < 0.002, ***p < 0.001. www.nature.com/scientificreports/ flower with least expression in roots. In the case of WsCYP76, there was a negligible expression in all tissues other than leaves (Fig. 1B). Similarly, phytosterol pathway genes such as WsHMGR, WsFPPS, and WsDWF5 showed differential expression in tissues with maximal expression in W. somnifera leaves 3,25 . In another study, expression level of obtusifoliol-14-demthylase (CYP51) and sterol methyl transferase (SMT-1) was shown to be preponderant in leaves of W. somnifera than in other tissues 26 . In our previous study, a transcription factor (WsWRKY1) regulating the biosynthesis of phytosterols and withanolides also exhibited higher expression in leaves 15 . It has been reported that withanolides are differentially present in different plant parts: however, some show tissue-specific accumulation. For instance, withaferin A is primarily synthesized in leaf tissue while withanolide A biosynthesis is seen specifically in roots 7 . Since CYP450s have the inherent ability to act on a range of substrates, their differential expression in tissues with maximal expression in leaves suggests that they may have a higher degree of involvement in the formation of leaf-specific withanolides. Whereas, they may have lesser role in the formation of withanolides specific to other tissues. The differential expression of WsCYP450s in tissues with maximal expression in leaves suggest that they may have a higher degree of involvement in formation of leaf-specific withanolides.  (Fig. 2). The percent identities of WsCYP76 and WsCYP71B10 were in the range of 34-41% and 37-48%, respectively, with other analyzed proteins of CYP76, and CYP71 families that have been shown to have a role in secondary metabolism. While WsCYP76 exhibited the highest (~ 41%) identity to geraniol 8-hydroxylase of Catharanthus roseus 27 and Swertia mussotii 28 followed by santalene/bergamotene oxidase (40%) of Santalum album 29 , WsCYP71B10 showed highest homology to Populus trichocarpa PtCYP71B40v3 involved in the conversion of aldoxime to defensive nitriles 30 , followed by ~ 45% and ~ 42% homology to Persea Americana transcinnamic acid 4-hydrolase (PaCYP71A1) and Artemisia annua amorpha-4,11-diene 12-monooxygenase (AaCY-P71AV1), respectively 31,32 (Fig. 2). Concerning WsCYP749B1, no enzymes for the CYP749 family (reported to be evolved only in Asteroids, Rosides and Ranunculales members) have been functionally characterized. However, the tomato CYP749A20 gene was up-regulated in red and orange fruit with unknown functions 33 . WsCYP749B1 exhibited ~ 48-56% homology to analyzed sequences with highest homology to Capsicum annuum CYP749A22 with unknown function of Solanaceae family. Further, prediction of subcellular localization using different tools indicated that all three WsCYP450s are membrane targeted, which is a characteristic of CYP450 (Table S3).

VIGS of WsCYP450s modulates withanolides accumulation. VIGS approach has been successfully
utilized for gene function studies in W. somnifera 3,14,15,25 . To verify the in planta role of WSCYP749B1, WSCYP76 and WSCYP71B10 in withanolide biosynthesis, VIGS was performed. As VIGS is prone to off target silencing 34 , the sequence region unique to each WsCYP450s was chosen to generate the silencing constructs. Thirty days post-infiltration leaves of similar developmental stages exhibiting typical viral infection symptoms were collected along with empty vector control for transcript and metabolite analysis. qRT-PCR analysis showed that the extent of silencing among the three CYP450 candidate genes ranged between 75 and 85% (Fig. 3). Among the three genes subjected to VIGS, WsCYP71B10 exhibited maximum silencing effect (85%), followed by WsCYP76 (80%), and WsCYP749B1 (75%) (Fig. 3A,C,E). The degree of silencing of WsCYP450s in this study was comparable to the VIGS of other related genes of W. somnifera reported in earlier studies 3,14,15 . Subsequent analysis of withanolides through HPLC indicated that VIGS of all individual WsCYP450s exhibited modulation in the accumulation of one or more withanolides in comparison to control plants. Though VIGS of all three genes individually showed similar modulation profiles of withanolides, only withaferin A, withanolide A and withanolide B were significantly affected (Fig. 3B,D,F). While VIGS of WsCYP749B1 significantly enhanced withaferin A content by ~ 450% and reduced withanolide A by 50% (Fig. 3B), silencing of WsCYP76 and WsCYP71B10 significantly decreased the accumulation of withanolide A (70%) and withanolide B (50%), respectively (Fig. 3D,F). The modulation in withanolide profile in VIGS samples indicated the functional role of WsCYP749B1, WsCYP76 and WsCYP71B10 genes in withanolide biosynthesis.

Overexpression of WsCYP450s differentially affects withanolides accumulation. To further
investigate the effect of overexpression on withanolides accumulation, all three WsCYP450 genes were individually overexpressed in W. somnifera leaves by Agroinfiltration. Analysis of gene expression by qRT-PCR showed a significant increase in the levels of transcripts ranging between 2.5-to 35-fold. WsCYP749B1 had the highest expression (35-fold) followed by WsCYP71B10 (10 fold) and WsCYP76 (2.5-fold) compared to vector control samples (Fig. 4A,C,E). This increase in transcript levels of WsCYP450s resulted in differential accumulation of withanolides. Overexpression of WsCYP749B1 and WsCYP76 exhibited a similar accumulation profile with a significant increase of withanolide A by 250% and 150% and a drastic decrease of 12-deoxywithastramonolide by 60% and 70%, respectively (Fig. 4B,D). WsCYP749B1 overexpression also lead to increased withanolide B content by 180% compared to control. Whereas, overexpression of WsCYP71B10 significantly reduced (60%) withanolide A levels without affecting any other analyzed withanolides (Fig. 4F). There was a positive correla- www.nature.com/scientificreports/ tion with respect to transcript levels of WsCYP749B1 and WsCYP76, and withanolide A content in VIGS and overexpression background (Figs. 3,4). CYP450s have inherent ability to catalyze a diverse range of oxidative reactions exhibiting versatile substrate specificities and promiscuity. For instance, A. annua CYP71AV1 is involved in catalyzing three successive reactions in artemisinin pathway 35 . In Salvia miltiorrhiza the promiscuous nature of CYP76AH3 and CYP76AK1 bifurcate the tanshinones biosynthetic pathways and are involved in multiple steps suggesting it comprises a complex metabolic network 36 . Whereas Coleus forskohlii CYP76AH15 is involved in catalyzing reactions in two different pathways leading to ferruginol and forskolin 37 . In this study also, in planta silencing and overexpression of WSCYP749B1, WSCYP76 and WSCYP71B10 resulted in modulation in the accumulation of certain withanolides (Figs. 3, 4). From VIGS and overexpression results, it can be construed that the three characterized genes may be involved in one or several steps of the pathway though not necessarily in a sequential manner. While WSCYP749B1 may utilize withaferin A and 12-deoxywithastromonolide for formation of certain yet to be identified intermediates in the pathway, and could also have a role in formation of withanolide A and withanolide B. Likewise, WsCYP76 may be involved in utilizing 12-deoxywithastromonolide and could also have a role in withanolide A formation. Whereas WSCYP71B10 might have a role in utilizing withanolide A and forming withanolide B (Table S4).  www.nature.com/scientificreports/ VIGS and overexpression of WsCYP450s affects PR genes and defense against P. syringae. In our previous studies in W. somnifera, it was observed that VIGS and overexpression of genes related to phytosterol and withanolides pathway modulate the expression of defense-related PR genes 14,15 . Moreover, withanolides have been shown to possess antibacterial and antifungal properties 38 . Since withanolides are derived from the phytosterol pathway and WsCYP749B1, WsCYP76 and WsCYP71B10 show their role in withanolides biosynthesis, we checked whether VIGS and overexpression of these genes have any effect on PR genes. Both salicylic acid (SA)-dependent PR1 and jasmonate-dependent PR3 exhibited a significantly reduced transcripts in all three WsCYP450-vigs samples in comparison to vector control. While PR1 exhibited 40% downregulation in WsCYP749B1-vigs, and ~ 70-80% reduced transcripts in WsCYP76-vigs and WsCYP71B10-vigs samples, PR3 downregulation was in the range of 60-70% in all three WsCYP450-vigs backgrounds (Fig. 5A). www.nature.com/scientificreports/ CYP450s play important roles in plant defense through their involvement in phytoalexin biosynthesis, hormone metabolism and the biosynthesis of some other secondary metabolites 39 . As silencing and transient overexpression of WsCYP450s resulted in modulation of withanolides content and PR gene expression, bacterial growth assay was performed to determine WsCYP450s' role in defense. Leaves from all three WsCYP450-vigs backgrounds developed severe disease symptoms and sustained more tissue damage than EV leaves when inoculated with P. syringae DC3000 (Fig. S4). Further, bacterial growth assay using extract isolated from inoculted leaves 3 dpi showed that WsCYP450-vigs samples exhibited adrastic increase in P. syringae DC3000 growth than that of EV control. The log cfu/cm 2 for WsCYP749B1-vigs, WsCYP76-vigs and WsCYP71B10-vigs was found to be 7.4, 7.2 and 7.5, respectively, whereas it was ~ 6 cfu/cm 2 for EV control (Fig. 5B).
In contrast to the effect of VIGS, overexpression of WsCYP450s led to upregulation of both PR1 and PR3 genes and enhanced the tolerance to bacterial growth. While PR1 exhibited 10-to 30-fold enhanced expression, PR3 showed ~ 14 to 18-fold increased expression in different WsCYP450 overexpression samples in comparison to control (Fig. 5C). Further, leaves overexpressing WsCYP450-vigs remained healthy and exhibited no necrotic phenotype compared to control at 4 dpi of P. syringae DC3000. Subsequent bacterial growth assay revealed a significant reduction in the growth of P. syringae DC3000. The log cfu/cm 2 in WsCYP749B1, WsCYP76 and www.nature.com/scientificreports/ WsCYP71B10 overexpressing tissue was 5.5, 6.9 and 6, respectively, in comparison to 7.2 in vector control. The observed phenotype and bacterial growth in P. syringae infiltrated samples correlated with elevated and reduced expression of WsCYP749B1, WsCYP76 and WsCYP71B10 (Fig. 5D). It is interesting to note that PR genes were affected in both VIGS and OE backgrounds indicating that the products (one or more withanolides) formed by WsCYP749B1, WsCYP76 and WsCYP71B10 enzymes could be linked to defense signaling. In our earlier studies also, silencing of WsSQS of phytosterol pathway and silencing and overexpression of WsWRKY1 involved in regulation of phytosterol and withanolides formation, showed positive correlation with expression of PR genes 14,15 . Furthermore, AtCYP76C2 and AtCYP71A12 from Arabidopsis were found to be associated with defense mechanism against P. syringae infection 40,41 . Hence, it is clear that the three identified genes are involved in bacterial defense, however, how the change in expression of WsCYP450s affected PR gene expression is quite intriguing and interesting and needs further investigation. The enzymatic products of these genes (one or more withanolides) may play a role in signaling as well as providing tolerance through their antibacterial activity.
Heterologous overexpression of WsCYP450s in transgenic tobacco confers tolerance to P. syringae. Heterologous overexpression of plant CYP450s has been shown to confer tolerance to various stresses including tolerance against bacteria 42 . To further investigate the defensive role of WsCYP450s in a heterologous system, independent transgenic tobacco lines overexpressing WsCYP450s were generated (Fig. S5). Transcript abundance of respective genes in 7, 8 and 9 independent lines of WsCYP749B1, WsCYP76 and WsCYP71B10, respectively, was determined by RT-qPCR. While the expression of WsCYP749B1 in WsCYP749B1-OE tobacco lines ranged from 1-to 45-fold (Fig. 6A), the expression of WsCYP76 and WsCYP71B10 in WsCYP76-OE and WsCYP71B10-OE lines, ranged from 3-to 32-fold (Fig. 6B) and 1-to 12-fold (Fig. 6C), respectively. Three higher expressors along with one least expressor for each WsCYP450 gene and control plants were chosen for bacterial growth assays. Analysis of bacterial growth using extracts isolated from leaf discs of P. syringae infiltrated leaves at 3 dpi revealed that higher WsCYP450 overexpressing tobacco plants supported significantly lower multiplica- Pseudomonas syringae growth assay in WsCYP450s silenced (B) and overexpressing (D) W. somnifera leaves. Growth assay was performed by infiltrating leaves of control and VIGS plants with P. syringae (DC3000). CFU was calculated at 3 dpi of P. syringae infiltration by plating serial dilutions of leaf disc extracts. Error bars show mean ± SE of three independent experiments. Statistical significance represented as *p < 0.033, **p < 0.002, ***p < 0.001, a significant difference. www.nature.com/scientificreports/ tion of P. syringae compared to extracts from leaf discs of pBI121 plants (Fig. S6). While all overexpressing lines of WsCYP749B1 and WsCYP71B10 exhibited a significant reduction in bacterial growth (Fig. 6D,F), only one line of WsCYP76 (line 9.5) showed a drastic decrease in bacterial growth (cfu ~ 7.1) (Fig. 6E). The degree of tolerance to bacterial growth in comparison to control plant was highest in WsCYP71B10-OE lines (cfu 6.0-6.46) followed by WsCYP71B10-OE lines (cfu 6.99-7.12) (Fig. 6). Similar to our results, overexpression of Panax ginseng PgCYP76C9 conferred enhanced resistance to P. syringae in transgenic Arabidopsis 43 .

Conclusion
In this study three MeJA-inducible CYP450s having higher expression in leaves were investigated for their in planta role using silencing and overexpression approaches. The results demonstrated that change in the expression of these genes by silencing or overexpression resulted in significant differential modulation of certain withanolides indicating their role in one or more steps of the withanolides biosynthetic pathway. In addition, our study showed that these three WsCYP450s have a role in defense and confer tolerance against bacteria. Overall, the study demonstrated the in planta role of WsCYP749B1, WsCYP76 and WsCYP71B10 in withanolides biosynthesis and defense against bacteria. Further biochemical characterization of these WsCYP450s could provide insights into their exact role in withanolides formation. The identified genes could also be utilized for metabolic engineering of W. somnifera for targeted enhancement of certain withanolides either in cell cultures or at the whole-plant level. Additionally, these genes could be utilized for generating plants tolerant to bacterial pathogens. WsCYP450s were normalized to internal reference NtEF1α and arerepresented as expression relative to pBI121 control that was set to 1. (B, D, F) Bacterial growth assay with P. syrinage (DC3000) strains. Growth assay was performed by infiltrating leaves of control and transgenic WsCYP450-OE lines with P. syringae (DC3000). The bacterial growth (CFU) at 3 dpi was obtained by plating serial dilutions. Error bars show mean ± SE of three independent experiments. Statistical significance represented as *p < 0.033, **p < 0.002, ***p < 0.001, a significant difference.