Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

RNAi and Homologous Over-Expression Based Functional Approaches Reveal Triterpenoid Synthase Gene-Cycloartenol Synthase Is Involved in Downstream Withanolide Biosynthesis in Withania somnifera

  • Smrati Mishra ,

    Contributed equally to this work with: Smrati Mishra, Shilpi Bansal

    Affiliation Department of Metabolic and Structural Biology, CSIR-Central Institute of Medicinal and Aromatic Plants (CSIR-CIMAP), P.O. CIMAP, Lucknow, India

  • Shilpi Bansal ,

    Contributed equally to this work with: Smrati Mishra, Shilpi Bansal

    Affiliation Department of Metabolic and Structural Biology, CSIR-Central Institute of Medicinal and Aromatic Plants (CSIR-CIMAP), P.O. CIMAP, Lucknow, India

  • Bhawana Mishra,

    Affiliation Department of Metabolic and Structural Biology, CSIR-Central Institute of Medicinal and Aromatic Plants (CSIR-CIMAP), P.O. CIMAP, Lucknow, India

  • Rajender Singh Sangwan,

    Affiliations Department of Metabolic and Structural Biology, CSIR-Central Institute of Medicinal and Aromatic Plants (CSIR-CIMAP), P.O. CIMAP, Lucknow, India, Center of Innovative and Applied Bioprocessing (CIAB) (A National Institute under Department of Biotechnology, Government of India), C-127, Phase-8, Industrial Area, S.A.S. Nagar, Mohali, Punjab, India

  • Asha,

    Affiliation Department of Metabolic and Structural Biology, CSIR-Central Institute of Medicinal and Aromatic Plants (CSIR-CIMAP), P.O. CIMAP, Lucknow, India

  • Jyoti Singh Jadaun,

    Affiliation Department of Metabolic and Structural Biology, CSIR-Central Institute of Medicinal and Aromatic Plants (CSIR-CIMAP), P.O. CIMAP, Lucknow, India

  • Neelam S. Sangwan

    Affiliation Department of Metabolic and Structural Biology, CSIR-Central Institute of Medicinal and Aromatic Plants (CSIR-CIMAP), P.O. CIMAP, Lucknow, India

RNAi and Homologous Over-Expression Based Functional Approaches Reveal Triterpenoid Synthase Gene-Cycloartenol Synthase Is Involved in Downstream Withanolide Biosynthesis in Withania somnifera

  • Smrati Mishra, 
  • Shilpi Bansal, 
  • Bhawana Mishra, 
  • Rajender Singh Sangwan, 
  • Asha, 
  • Jyoti Singh Jadaun, 
  • Neelam S. Sangwan


Withania somnifera Dunal, is one of the most commonly used medicinal plant in Ayurvedic and indigenous medicine traditionally owing to its therapeutic potential, because of major chemical constituents, withanolides. Withanolide biosynthesis requires the activities of several enzymes in vivo. Cycloartenol synthase (CAS) is an important enzyme in the withanolide biosynthetic pathway, catalyzing cyclization of 2, 3 oxidosqualene into cycloartenol. In the present study, we have cloned full-length WsCAS from Withania somnifera by homology-based PCR method. For gene function investigation, we constructed three RNAi gene-silencing constructs in backbone of RNAi vector pGSA and a full-length over-expression construct. These constructs were transformed in Agrobacterium strain GV3101 for plant transformation in W. somnifera. Molecular and metabolite analysis was performed in putative Withania transformants. The PCR and Southern blot results showed the genomic integration of these RNAi and overexpression construct(s) in Withania genome. The qRT-PCR analysis showed that the expression of WsCAS gene was considerably downregulated in stable transgenic silenced Withania lines compared with the non-transformed control and HPLC analysis showed that withanolide content was greatly reduced in silenced lines. Transgenic plants over expressing CAS gene displayed enhanced level of CAS transcript and withanolide content compared to non-transformed controls. This work is the first full proof report of functional validation of any metabolic pathway gene in W. somnifera at whole plant level as per our knowledge and it will be further useful to understand the regulatory role of different genes involved in the biosynthesis of withanolides.


Withania somnifera (Indian ginseng) is one of the most famous and widely used medicinal plants [1]. W. somnifera, commonly known as Ashwagandha, is an ancient medicinal herb that has been used in various pharmacological preparations traditionally. The main constituents for its pharmacological activities are withanolides [2].Withanolides has been shown to exert many pharmacological effects, including immune system modulation, antistress activities, antihyperglycemic activities, anti-inflammatory, antioxidant, and anti-cancer effects [36]. Withanolides are basically of the terpenoid origin and are one of the the largest group of natural products with diverse molecular structures. Withanolides or 22-hydroxy ergostane-26-oic acid 26, 22-δ-lactones are C-28 steroidal lactones based on ergostane framework which when oxidized at C-22 and C-26 form a δ-lactone ring [7]. Till now several withanolides have been isolated from different parts of W. somnifera including major withanolides i.e. withaferin A, withanone and withanolide A, withanolide D etc [8]. Withanolides are synthesized through isoprenoid pathway via both mevalonate and non-mevalonate pathways [9]. The first committed step in the triterpenoid biosynthesis occurs by cyclization of epoxysqualene, leading to the formation of C30H50O product with 1–5 rings [10]. Plants contain different oxidosqualene cyclases (OSCs) for a single substrate to synthesize different set of sterols and other triterpenoids [11].

Cycloartenol synthase [(S)-2, 3-epoxysqualene mutase, EC] belongs to the 2–3 oxidosqualene cyclases (OCSs) gene family along with α-amyrin synthase (AAS), β-amyrin synthase (BAS), lupeol synthase (LS), and lanosterol synthase (LAS) [1214]. Cycloartenol synthase (CAS) is one of the key enzymes involved in withanolide biosynthesis in W. somnifera (Fig 1). CAS is involved in cyclization of 2, 3-oxidosqualene to form cycloartenol which acts as the key precursor for the biosynthesis of phytosterols as well as withanolides through a series of desaturation, hydroxylation, epoxidations, cyclization, chain elongation, and glycosylation steps [15]. CAS from other plant species such as Arabidopsis thaliana [16] Glycyrrhiza glabra [17], Pisum sativum [18], Rhizophora stylosa [19], Kandelia candel [20] are also reported earlier. β-amyrin synthase (OCS) has been shown to be involved in the synthesis of glycyrrhizin and soyasaponin in Licorice [21]. Another OSC, shionone synthase is known to be associated with formation of a triterpene shionone in Aster tataricus [22]. Though some of the genes involved in withanolide biosynthesis have been cloned and characterized from W. somnifera [2330]. Their further functional validation at plant level is still awaited. A close association of phytochemical profiling of medicinally important withanolides, organogenesis and expression of pathway genes has been established [3132]. Transcriptomic data of withanolide biosynthetic pathway (leaf and root specific) were also reported which identified putative genes involved in withanolide biosynthesis [3334].

Fig 1. Withanolide biosynthetic pathway.

IPP, isopentenyl pyrophosphate; MVA, mevalonate; MEP, 2-C-methyl-D-erythritol 4-phosphate GPPS, geranyl pyrophosphate synthase; FPPS, farnasyl pyrophosphate synthase; SS, squalene synthase; SE squalene epoxidase; CAS, cycloartenol synthase; LAS, lanosterol synthase; LS, lupeol synthase; BAS, beta amyrin synthase; AAS, alpha amyrin synthase.

Different genetic engineering methods, such as overexpression of transgenes, multiple expressions of transgenes, gene silencing, and transcription factors are powerful tools for the engineering of biosynthetic pathway [35]. In order to determine the gene function via transgenic approach two methods are very effective one of which involves loss of gene function i.e. RNAi mediated mechanism and other through gain of function or over-expression approaches. Both of these mechanisms together serve as an important tool to validate the function of a geneRNA-mediated gene silencing is initiated by the delivery of double-stranded RNA (dsRNA), which is processed into small interference RNA molecules (siRNA) of 21–25 bp by Dicer (DCR), which depends on ATP and RNaseIII. The siRNA is then incorporated into the RNAi-induced silencing complex (RISC), which in turn degrades the target mRNA [36]. As RNAi involves sequence-specific interaction between siRNA and mRNA, siRNAs can be tailored to silence a particular gene or even all the members of the gene family. -RNAi of specific target genes based on the stable expression of dsRNA has been successfully used in many economically important crop species [3741]. Therefore, the study of functional role of candidate gene(s) dedicated to withanolide production via genetic transformation and functional genomics becomes an important issue in Withania research. In this study RNAi binary vector pGSA1131 expressing hairpin RNAs of WsCAS and pIGCAS (over-expression construct) were utilized to generate stable transgenic lines of W. somnifera to demonstrate the relation between WsCAS gene and withanolide production. Several transgenic silenced lines with reduced withanolide content were obtained as compared to control. In the same way transgenic Withania plants over-expressing WsCAS accumulated enhanced level of withanolides. Our study demonstrates that this transgenic approach could be efficiently used to improve and understand withanogenesis through functional genomics.

Materials and Methods

Plant Material

Withania somnifera Dunal plants were grown in glasshouse as well as in field in CIMAP-Experimental Farm at Lucknow, India, following standard agronomic practices. Seedlings were raised and planted in pots and field in the months of July and August and samples were harvested as per the experimental requirement from time to time and various experimental observations were recorded.

Molecular cloning of WsCAS

Total RNA from W. somnifera leaf tissue was extracted by Trizol method and the first strand cDNA was synthesized using superscript II reverse transcriptase (Invitrogen), according to the manufacturer’s instructions. Sequences of primers were designed from the conserved domains identified in CAS gene reported from other plants. The PCR parameters were: initial denaturation at 94°C for 3 min, followed by 35 cycles at 94°C (20 sec), annealing at 56°C (40 sec), extension at 72°C (2 min) and final extension at 72°C for 7 min. The PCR products were electrophoresed on 1.2% agarose gel. The resolved amplicons were purified and cloned in vector pTZ57R/T (Fermentas, Life Sciences) and sequenced and deposited in NCBI database. On the basis of sequence information, a set of gene specific primers for 5’ and 3’ RACE were designed. The RACE amplified fragments were sequenced. On the basis of sequence information, a composite sequence of WsCAS encoding a full length cDNA (2.277 kb) was established. Another set of primers (WsCASF0 and WsCASR0) were designed to amplify complete strand of WsCAS and this full length was cloned in vector pJET1.2/blunt (Fermentas, Life Sciences) following standard protocols. Information related to all the primers used in this study was provided in S1 Table.

Stress treatment

Seeds of W. somnifera were sown in field and one month old germinated seedlings were utilized for various stress treatments such as salicylic acid (SA), methyl jasmonate (MeJA), heat-shock, cold and wounding. For SA and MeJA treatment, seedlings were dipped in distilled water supplemented with 2 mMand 5 mM SA and 200 μM and 500 μM MeJA for 4 hrs. For heat shock treatment, seedlings were kept at 65°C for 1hr and for cold treatment seedling were kept at 4°C for 4 hr. Mechanical injury was caused by rubbing the leaves, and seedlings were dipped in distilled water for 4 hours.

Silencing constructs preparation

The binary vector, pGSA (double-stranded RNAi vector) (ABRC Stock Number CD3-449), was used in this study to generate the WsCAS RNAi constructs. The pGSA vector contains an intron of the GUS gene of 372 bp to express CAS RNAi harpins. Three fragments of 261bp, 250bp and 323bp from WsCAS coding sequence from 5’, middle and 3’ prime regions respectively were amplified using the primers incorporating Spe I and Asc I in forward primer and Bam HI and Swa I in reverse primer as shown in bold letter in Table 1 and cloned in pJET vector and further sequenced. The amplification conditions were initial denaturation at 94°C for 3 min, followed by 35 cycles at 94°C (30 s), annealing at 55°C (40 s), extension at 72°C (1min) and final extension at 72°C for 7 min. The amplified fragment was initially digested with AscI and SwaI and cloned in sense orientation into the binary pGSA1131 immediately after the CaMV 35S promoter to generate an intermediary vector. The same WsCAS fragment was then digested with SpeI and BamHI and cloned in antisense orientation next to the GUS intron spacer region of the intermediary vector resulting in silencing construct WsRNAi 1, WsRNAi 2 and WsRNAi 3 respectively. The vector was sequenced to confirm the authenticity and correct orientation of the two cloned inserts.

Preparation of WsCAS over-expression construct

The full-length WsCAS sequence was amplified from leaf cDNA of W. somnifera with the forward WsCASF0 and reverse WsCASR0 primers incorporating BamHI site at 5’ end as shown in bold letter in S1 Table. The amplified fragment was digested with BamHI and cloned into pJET cloning vector (Fermentas). After sequence confirmation the desired insert was cloned at BamHI site of binary vector pIG121Hm (AB 489142.1) resulting into overexpression vector pIGCAS.

Generation of transgenic lines of WsCAS in W. somnifera

Node explants of in vitro grown seedling of W. somnifera were used for infection. Silencing constructs alongwith RNAi vector pGSA1131 and pIGCAS (overexpression construct) were used to transform Withania node explants. Stable transgenic Withania plants harboring these constructs were generated via optimized transformation protocol developed in our lab [42]. The pGSA1131 binary vector confers BASTA resistance in plants and hence transgenic silenced lines were selected by 50 μg L-1 BASTA. pIGCAS is harboring kanamycin as plant selectable marker gene thus overexpressing lines were selected on medium supplemented with 50 mg L-1 kanamycin.

Molecular analysis of putative transformants

Plant genomic DNA was isolated using cetyl trimethylammonium bromide (CTAB) protocol with some modification from randomly selected silenced and overexpressing lines along with empty vector pGSA and untransformed plant. Genomic DNA from selected lines after BASTA/kanamycin selection was examined by PCR amplification for the presence of bar (phosphinothricin acetyl transferase gene conferring BASTA resistance) / nptII (neomycin phosphotransferase gene conferring kanamycin resistance) gene. The amplification conditions were initial denaturation at 94°C for 3 min, followed by 35 cycles at 94°C (30 s), annealing at 55°C (40 s), extension at 72°C (1:30 min) and final extension at 72°C for 7 min. Genomic DNA (15 μg) from randomly selected PCR-positive plantlets and non-transformed plant were digested with HindIII (silenced lines)/XbaI (over expressing lines), and the resulting fragments were separated by electrophoresis on 0.8% (w/v) agarose gel and immobilized on H+ nylon membrane (Sigma, USA) by capillary transfer method described by Sambrook and Russell [43]. Blots were hybridized with PCR-generated probes for bar/nptII gene (amplified as above), labeled with DIG-high prime, hybridization, washing, and detection were performed according to DIG-DNA labeling and detection kit according to manufacturer’s instruction (Roche Applied Science, Germany). The bar/nptII PCR product and genomic DNA from untransformed plants served as positive and negative controls, respectively.

Quantitative Real-time PCR (qRT-PCR) analysis

Total RNA was isolated from shoot tissues using TRI reagent (Sigma, USA) and treated with RNase-free DNaseI (Fermentas). The first strand cDNA was synthesized from 5 μg of total RNA using RevertAid Premium first strand cDNA synthesis kit (Fermentas) according to manufacturer’s instructions (Fermentas). qRT-PCR was performed in transgenic silenced lines and over expressing WsCAS gene using the ABI real time DNA amplification system (Applied Biosystem, USA) under the following conditions: 95°C for 10 s; 40 cycles of 95°C for 5 s, 52°C for 20 s using WsCAS RT specific primers (Table 1). Melting curve analysis was included to verify specificity of the DNA amplification. The actin gene was used as reference gene. Ten-microliter reactions were set up for each sample in triplicate.

HPLC analysis of withanolide content

Silenced lines, overexpressing lines of WsCAS and untransformed shoots were separated from the medium, frozen in liquid nitrogen, ground to a fine powder and were mixed with 1 ml 50% methanol and place for overnight with intermediate vigorous mixing. The extracted withanolides were dissolved in HPLC-grade methanol & used for HPLC analysis as reported earlier [44]. Total withanolide content (TWC) was defined as μg mg-1 of withanolides in fresh tissues used to assess the effects of WsCAS activity in withanogenesis.


Cloning of WsCAS

Cycloartenol synthase gene occupies central position in the withanolide biosynthetic pathway (Fig 1). Different set of degenerate primers were designed on the basis of other known plant CAS gene sequences (S1 Table). Three partial WsCAS cDNAs were obtained using different pairs of degenerate primers and were designated as WSTS1 (GenBank: GU295060.1), WSTS2 (GenBank: GU295059.1) and WSTS3 (GenBank: GU295058.1). These partial WsCAS sequences showed very high sequence similarity with each other as well as exhibited close homology with CAS sequences in the NCBI database (S1 Fig). To make it full length gene 5’and 3’RACE was carried out from these partial fragments resulting in a single cDNA, WsCAS, of 2277 bps which was cloned using specific primer WsCASF0 and WsCASR0 (Table 1). The alignment of the WsCAS sequence with the sequences of reported CASs revealed that WsCAS contained a putative open reading frame of 2277 bp encoding a polypeptide of 758 amino acid residues. The predicted polypeptide had a molecular weight of ~83 kDa. Deduced polypeptide showed sequence elements typical for CAS’s such as a DCTAE motif, QW motifs (S1 Fig). A phylogenetic tree was generated from the alignment of the deduced full-length amino acid sequences of twenty four CAS sequences reported from other plants. Despite being high level of homology among all CAS proteins, the phylogenetic analysis showed maximum similarity (92%) with SlCAS (S2 Fig).

Modulation of WsCAS under different elicitor conditions

Withania seedlings under different abiotic stress were shown in Fig 2A. SA 2 mM and 5 mM and MeJA 200 μM and 500 μM enhanced the WsCAS transcript expression levels upto 2.49, 1.72, 1.65 and 1.14 folds respectively while cold, heat and wounding lowered the WsCAS expression levels upto 0.25, 0.51and 0.59 fold respectively. To establish a link between WsCAS transcript accumulation and withanolide levels under the above conditions, phytochemical analysis was also carried out. The overall content of withanolide was higher in case of SA and MeJA treatments at both the concentrations while lower in cold, heat and wounding in comparison with control respectively. Percentage of total withanolide at SA 2 mM and 5 mM was found increased upto 2.0 and 1.75 folds, respectively as compared to control. In case of MeJA treatment withanolide content was increased upto 1.42 fold at 200 μM and upto 1.31 fold at 500 μM while decreased by 0.2 fold under cold stress conditionand 0.63 folds under both, heat and wounding treatments. Our results revealed a direct correlation between the WsCAS expression and accumulation of withanolide in treated seedlings under elicitor (SA and MeJ) treatment and heat, cold and wounding stress conditions (Fig 2).

Fig 2. Profiling of WsCAS expression and withanolide in Withania seedlings under different elicitor conditions.

(A) Seedling before and after treatment of cold, wounding, SA 2mM, SA 5mM, MeJ 200μM, MeJ 500μM and heating (a-i) respectively. (B) Real time-PCR analysis of WsCAS. (C) Quantitative withanolide profile.

Designing of silencing constructs

Since higher silencing efficiencies were desired, the full length WsCAS cDNA sequence was analyzed for strong putative silencing sites. Three sites were selected and tested in small inverted repeats, first inverted repeat (IRs) comprising 261 bp from the 5’ region, second one of 250 bp from middle region and the third 323 bp from 3’ region respectively (Fig 3). For silencing construct preparation a binary RNAi vector pGSA1131 was used. The three selected sites were amplified using Forward primer incorporating SpeI and AscI sites and reverse primer having BamH1 and SwaI restriction enzyme sites (shown in bold letters in Table 1). The amplified products were cloned in cloning vector pJET (Fermentas) and sequenced. The cloned plasmids were characterized via colony PCR and restriction digestion. These cloned WsCAS fragments digested with AscI and SwaI of 271bp, 260 bp and 333 bp respectively from 5’, middle and 3’ region were cloned in sense orientation into the binary double-stranded RNA vector pGSA1131 immediately after the CaMV 35S promoter to generate an intermediary vector. The same WsCAS fragment was then digested with SpeI and BamHI of 287 bp, 276 bp and 349 bp respectively from 5’, middle and 3’ region and cloned in antisense orientation next to the GUS intron spacer region of the intermediary vector resulting in RNAi silencing constructs WsRNAi 1, WsRNAi 2 and WsRNAi 3. Transcription of these construct would be expected to give a hairpin RNAs (Fig 4). The total cassette including sense and antisense fragments along with GUS spacer region were checked by cutting it with AscI restriction enzyme and the total length of the inserted fragment were 923, 901, and 1047 bp, as expected respectively from 5’, middle and 3’ region of WsCAS. DNA sequencing confirmed that these WsCAS RNAi vectors were constructed successfully. The T-DNA regions of these silencing vectors were shown in Fig 5A.

Fig 3. Full length WsCAS gene with three selected silencing sites for silencing constructs preparation in Withania somnifera.

19 bp silencing sites were enclosed in red colored boxes. Arrows indicate the region of start of amplification of particular fragment incorporating silencing sites.

Fig 4. Sequence analysis of WsCAS.

Alignment of predicted amino acid sequence of newly identified WsCAS cDNA with the cycloartenol synthases reported in the gene bank from various plants such as Lotus japonicus (AB181246), Arabidopsis thaliana (NM_126681), Solanum lycopersicum (EU449280), Panax notoginseng (EU342419) and Bupleurum kaoi (AY514456) showing conserved domains. The predominant DCTAE motif (highlighted in yellow), five QW motifs (underlined) and three conserved amino acids- Tyr410, His477, Ile481 (depicted by symbol♦) are appropriately shown.

Fig 5.

(A) T-DNA region of constructs. BAR, gene encoding resistance to the herbicide BASTA; OCS 3', poly adenylation signal sequence; MAS 3', poly adenylation signal sequence; MAS 2', plant promoter; CaMV 35S’ Viral promoter; LB, left border; RB, right border; GUS, 360 base pair fragment from the GUS gene; P promoter; T terminator; nos nopaline synthase; nptII neomycin phosphotransferase gene. (B) Generation of W. somnifera transgenic lines.

Preparation of WsCAS over-expression constructs

The full-length WsCAS sequence was amplified from cDNA of Withania with the forward WsCASF0 and reverse WsCASR0 primer incorporating BamHI site at 5’ end (Table 1). The amplified fragment of WsCAS full length (2277 bp) was digested with BamHI and cloned at BamHI site of binary vector pIG121Hm (AB 489142.1) resulting into over-expression vector pIGCAS. The T-DNA region of pIGCAS vector was also shown in Fig 5A.

Generation of transgenic plants harboring silencing and over-expression constructs

These silencing constructs along with vector control pGSA and over-expression construct were finally transformed in Agrobacterium strain GV3101 and after characterization used for plant transformation experiments. Stable transgenic Withania plants harboring these constructs were generated via developed transformation protocol using binary vector pIG121Hm from node explants [42]. Briefly, explants after 10–12 days of infection, transferred to selection media having BASTA 50 μg l-1 or kanamycin 50 mg l-1 for three consecutive selection cycles. All transformants were morphologically similar to untransformed one. The only difference is that silencing lines generated after BASTA selection, while in case of over expressing lines selection was on kanamycin (Fig 5B).

PCR and southern blot analysis

The plants after three consecutive cycles of BASTA/kanamycin selection were used for molecular analysis. The PCR amplifications with genomic DNA from silenced lines as well as vector pGSA transformed lines showed 418 bp amplicon for bar gene (phosphinothricin acetyl transferase gene conferring BASTA resistance) on agarose gel (1.0%, w/v) electrophoresis of the amplification reaction mixture (Fig 6A). In the same way PCR analysis of over expressing lines of randomly selected putative T0 transgenic plant showed desired amplicon of nptII of 500 bp (Fig 6C). No amplification was detected in the untransformed (control) plants. To confirm the integration and copy number of transgene in W. somnifera plantlets, Southern blot hybridization was performed on total genomic DNA isolated from BASTA resistant (silenced lines along with vector pGSA transformed line) and kanamycin resistant (over expressing lines) and untransformed (control) plantlets. bar and nptII PCR products were loaded as positive control. Genomic DNA was digested with HindIII, which has unique site within the T-DNA region of silencing constructs and with XbaI in case of overexpressing lines. The blot was probed with the PCR-amplified product of bar (418bp) and nptII (500bp) gene there by generating a unique fragment for each integrated copy. Genomic DNA of randomly selected silenced lines alongwith vector pGSA transformed lines as well as overexpressing lines showed hybridization signal with digoxigenin-labelled bar and nptII gene (PCR product 418 bp & 500 bp respectively) probe (Fig 6B and 6D). No hybridization signal was observed in the control plant. The position of hybridization signal varies among different regenerated transgenic plantlets indicating these transgenic plants arose as independent transformation events.

Fig 6. Molecular characterization of Withania silenced lines and overexpressing lines of WsCAS gene.

(A) PCR amplification with bar primers showing amplicon of ≈400 bp. Lane M 100 bp ladder; lane–C untransformed shoot; lane +C plasmid DNA; lane 1–9, in vitro transformed shoots; (B) Southern blot analysis of transgenic silenced lines of Withania with bar PCR probe. (C) PCR amplification of pIGCAS transformants with nptII primers showing amplicon of 500bp. Lane- 1 to 6 transformed lines. (D) Southern blot analysis of overexpresing lines.

Expression analysis WsCAS transcript level in transformants

Quantitative real time-PCR was carried out to check the transcription level of WsCAS in the transgenic silenced lines generated with different RNAi constructs and in over-expressing lines as well. The results showed that the transcription level of WsCAS was reduced by approximately 80–97% as compared to the control from 5’, middle and 3’ region (Fig 7A). WsRNAi 1 showed the maximum suppression of mRNA levels upto 96.7% followed by WsRNAi 3 (92.8%) and WsRNAi 2 (79.9%) on average basis. As expected, the transgenic WsCAS overexpressing lines showed enhancement of 1.2 to 7 fold of WsCAS transcript level, as compared to untransformed control plants (Fig 7C).

Fig 7. WsCAS transcript expression analysis and withanolide accumulation in Withania silencing and overexpressing lines.

(A) RealTime PCR analysis of WsCAS silenced lines. (B) Withanolide content in silenced lines. (C) Real-Time PCR analysis of WsCAS overexpressing lines. (D) Withanolide content in WsCAS overexpressing lines.

Metabolite Analysis

HPLC analysis was performed to estimate the withanolides in stable transgenic silenced lines as per protocol developed and used in our earlier reports [16]. The silenced lines had altered withanolide profiles, as revealed by HPLC (S3 Fig). These results indicate that there was 33–91% reduction in withanolide content in silenced lines (Fig 7B).Variation was seen in withanolide content among three different target sites i.e. WsRNAi 1, WsRNAi 2 and WsRNAi 3 and also within the transgenic lines of these sites. Withanolide content reduced in WsRNAi 1 lines from 33–91% followed by WsRNAi 2(30–69%), WsRNAi 3 (42–91%). Although variation in withanolide content was seen among different independent lines transformed with same construct but generalization of the result showed that among three sites withanolide content was lowest in WsRNAi 1 followed by WsRNAi 3 and WsRNAi 2. In case of over-expressing lines withanolide content was increased to the extent of 1.06 to 1.66 fold (Fig 7D).


Oxidosqualene cyclases (OSC’s) serve as key enzymes in secondary metabolite synthesis by driving the metabolic pool of isoprenoids and converting it to a range ofhigher terpenoid- secondary metabolites via cyclization of 2, 3 oxidosqualene [1214]. Cycloartenol synthases (CAS), an important OCS have been cloned from other plant species also such as Glycyrrhiza glabra, Oryza sativa licorice, pea etc. and its involvement in terpenoid biosynthesis has also been studied [1620]. It has been established that under stress related conditions elicitors help in increasing the concentration of metabolites and provides resistance against various biotic and abiotic stresses. However when a fungal elicitor protein, cryptogein encoding gene was co-transformed in hairy roots of W. somnifera, withanolide content and expression of pathway genes such as HMGR, FPPS were down-regulated. The co-transformation resulted in shift of metabolic pool from withasteroid formation to phenylpropanoid formation [45]. Secondary metabolites also have defensive role so an attempt was made to understand the linkage between elicitor treatment and expression of CAS gene. Real time PCR results showed that WsCAS was up regulated by SA and MeJA (signalling molecules) while down regulated by cold, heat and wounding stress.Expression of WsDXS and WsDXR in response to exogenous application of SA, MeJA, and wound treatments were also studied earlier showing their involvement in early steps of withanolide biosynthesis [24]. A similar result was observed in case of WsFPPS gene in response to different elicitors [23]. Modulations in the expression of WsHMGR in response to exogenous applications of SA and MJ as well as to mechanical injury was studied earlier [Akhtar et al. 2013 26]. Our findings suggest that expression level of WsCAS was modulated in response to heat cold, wound, MeJA, and SA treatments in condition specific manner. HPLC analysis showed the relative change in withanolide content also (Fig 2). These finding supports that withanolide biosynthesis is tightly regulated by WsCAS in W. somnifera.

Despite being indicated several times CAS as an important enzyme in withanolide biosynthetic pathway by various researchers, hardly attempts were made to validate the function of the gene in planta. Transgenic approaches provide a powerful tool for metabolic engineering and gene function studies in Withania plants. Based on the improved transformation system and development of frequent changes and progression in the field of RNAi, we carried out the study on the gene function investigation by RNAi technology. RNAi mediated gene silencing studies was an effective measure to carry out gene function investigation and further improvement in crop plant species [46] In tobacco 9.1–96.7% reduction in nicotine levels was observed on silencing of putrescine N-methyltransferase (PMT) gene [47]. Similarly transgenic RNAi silenced lines for caffeine synthase showed reduction in levels of caffeine and theobromine in Camellia sinensis [48]. When an OSC, dammarenediol synthase (DDS) which is involved in ginsenoside synthesis was silenced by RNAi in Panax ginseng, significant reduction in transcript level and subsequently ginsenoside production was observed [49]. In present study we constructed three silencing vectors for WsCAS gene spanning various regions of the gene (WsRNAi 1, WsRNAi 2 and WsRNAi 3) of full length WsCAS (Fig 5A). Most importantly the entire cassette including sense and antisense fragments along with GUS spacer region were transferred to any other plant transformation vector by cutting single enzyme Asc I (Fig 5A).

Silencing of related genes is highly efficient after infiltrating leaves with Agrobacterium transformed with the RNAi vector containing two inverted-repeat sequences separated by an intron in many earlier reports [50]. RNAi-mediated silencing of the flavanone 3-hydroxylase gene in transiently transformed strawberry fruits was studied and found downregulation of F3H gene by approximately 70% in the agroinfiltrated fruits compared with the control and also showed reduced flavonol and anthocyanin content [37]. These transformation methods gave first clue for understanding of both biosynthetic and regulatory genes and provide potential means to improve the specificity and effectiveness of particular metabolite biosynthetic pathway. But further such metabolic engineering efforts augment well after generation of stable transgenic lines within short time for the manipulation of metabolic flux towards efficient biosynthesis of desired secondary metabolites. RNA interference of squalene epoxidase 1 gene in transgenic Panax ginseng completely suppressed PgSQE1 transcription. Concomitantly, the interference of PgSQE1 resulted in 50% reduction of ginsenoside production [39]. As silencing of a gene helps in determination of gene function another way for characterization of a gene is its over-expression. It offers the advantage that apart from functional validation of a gene it helps in increased production of metabolites. Over-expression involves increasing the level of genes either by transcriptional enhancers mediated random activation of endogenous genes or by the expression of individual transgenes by transformation [51]. Keeping this in view WsCAS gene was over-expressed using pIG121 as binary vector. This gene was selected for testing RNAi in Withania because of catalyzing penultimate step of withanolide biosynthesis. Large numbers of independent transformants were produced for each construct. During the course of experiment no significant visible changes were observed in growth and development pattern of transgenic lines in comparison to untransformed lines (Fig 5B).

Molecular analysis of putative T0 transformants was carried out by PCR amplification of the bar/nptII fragment for the silenced transgenic lines and over-expressed lines respectively. Amplification of bar was observed in all silenced transgenic lines along with pGSA vector and over-expressing lines showed positive amplification for nptII gene. The Southern blots were performed with HindIII (in case of silenced lines) or XbaI (overexpressing lines) digested genomic DNA from the shoot of T0 transgenic plants, and probed with a DIG-labeled DNA probe of partial PCR amplified fragment of bar/nptII gene. The results showed the bar/nptII sequence was present in all T0 transgenic plants analyzed, and mostly each line contained single integrated loci of the bar/nptII gene (Fig 6). Southern blot also showed that the integration of transgene is a random event and not site specific and hence may result in different pattern of enzymatic activity. Agrobacterium mediated transformation generally results in integration of low copy number of T-DNA. Single copy of transgene provides stability to transgenic lines.

The results showed that the transcription level of WsCAS was reduced by approximately 80–97% as compared to the control from different target regions of the gene. WsRNAi 1 showed the maximum suppression of mRNA levels followed by WsRNAi 3 and WsRNAi 2 on average basis. The 5’ IR construct (WsRNAi 1) gave a higher silencing efficiency in range from 93.3 to 98.9% than the middle (WsRNAi 2) and 3’ IR (WsRNAi 3) construct. Withanolide content from the transformed plants was analyzed by HPLC analysis. Several silenced lines with low levels of withanolide content were obtained, in which, withanolide content decreased considerably showing the down regulation mediated by RNAi of WsCAS. The results demonstrated that the withanolide content of transformants differed significantly in silenced lines with varying levels of individual withanolide as compared to the untransformed one. Manipulation of saponin biosynthesis by RNA interference-mediated silencing of b -amyrin synthase gene in soybean exhibited a stable reduction in seed saponin content, correlated with the b -amyrin synthase mRNA depletion [41]. In tomato silencing of lycopene β and ε-cyclase genes was resulted in reduction of gene expression upto 8.95% and 13.16%, respectively [52]. Similarly, RNAi-mediated suppression of p-coumaroyl-CoA 3’-hydroxylase in hybrid poplar resulted in down-regulation of C3’H in a number of transformed lines, which generally correlated very well with reduced total cell wall lignin content [53]. From this we concluded that the down-regulation of WsCAS gene expression was due to the formation of the WsCAS RNA hairpin in vivo, which resulted in a decrease in WsCAS enzyme activity, ultimately resulting in a reduction in the biosynthesis of withanolides (Fig 7). In the same way WsCAS overexpressing lines showed enhancement of upto 7 fold of WsCAS transcript levels. An increase in withanolide content was maximally upto 1.66 fold was noticeable in over-expressed transformants (Fig 7). In Catharanthus roseus over-expression of a regulatory gene (ORCA3) and a structural gene (G10H) involved in MIA pathway increased the accumulation of monoterpenoid indole alkaloids [54]. Overexpression of CrtR-b2(carotene beta hydroxylase 2) in transgenic tomato plants resulted in elevated xanthophyll content [55].

Therefore, inverted repeat silencing construct under the control of the constitutive promoter were able to effectively down-regulate the pathway gene expression in Withania. However variation in direct linkage between CAS transcript level and relative withanolide content was observed in transgenic lines. The reason for that withanolide is not the immediate product of cycloartenol synthase. The withanolide biosynthesis involves various post-transcriptional and post- translation modification steps [3]. Difference in withanolides arises due to modification in carbocyclic skeleton or the side chain. The plant exhibits enzymatic machinery which can oxidize all the carbon atoms in steroid nucleus [9]. Studies on biosynthetic pathways and metabolic engineering already showed that the fluxes through the pathways are controlled not only by gene expression levels, but also by post-translational regulation of enzyme activity and enzyme and metabolite compartmentalization and transport [56]. Also the variation within these lines may be attributed to variation in expression levels of transgenes.

In conclusion, the present study was an attempt to generate transgenic lines containing a construct which impairs the normal functioning of a key pathway enzyme cycloartenol synthase (CAS) thereby suppressing withanolide biosynthesis in W. somnifera and also generating over-expressed lines with elevated levels of withanolides. Strategies utilized in both the experiments were different but it complemented with each other in the sense that both helped in determining the function of the WsCAS gene. The observed decrease in WsCAS transcription level in Withania silenced lines very likely resulted in a reduction in WsCAS enzyme activity in vivo, which in turn responsible for reduction in the biosynthesis of withanolides by regulationg a key position in the biosynthetic pathway by WsCAS.

Supporting Information

S3 Fig. HPLC chromatograms of withanolide analysis in Withania transgenic lines.

(a) silencing lines. (b) over expressing lines.


S1 Table. Primers used for molecular cloning of cycloartenol synthase from Withania somnifera.



Authors thank grant No BT/PR10715/AGR/36/602/2008 from Department of Biotechnology, New Delhi, NMITLI, Government of India as financial support granted to NSS and RSS. The authors are also thankful to BSC 203 grant for supporting completion of work. We thank Dr. D Nadegoda, for vector pGSA. SB is thankful to CSIR, New Delhi for Senior Research Fellowship. Authors thank Director CSIR-CIMAP for his constant support and encouragement.

Author Contributions

Conceived and designed the experiments: NSS RSS. Performed the experiments: SM SB BM A. Analyzed the data: SM SB BM RSS NSS. Contributed reagents/materials/analysis tools: SM SB BM RSS A JSJ NSS. Wrote the paper: SM SB BM RSS NSS.


  1. 1. Sangwan RS, Chaurasiya ND, Misra LN, Lal P, Uniyal GC, Sharma R, et al. Phytochemical variability in commercial herbal products and preparations of Withania somnifera (Ashwagandha). Curr Sci. 2004; 86: 461–465.
  2. 2. Sangwan NS, Sabir F, Mishra S, Bansal S, Sangwan RS. Withanolides from Withania somnifera Dunal: development of cellular technology and their production. Recent Patents on Biotechnology. 2013; 8: 25–35.
  3. 3. Sangwan NS, Sangwan RS. Secondary metabolites of traditional medicinal plants- a case study of Ashwagandha (Withania somnifera) L. In: Opatrny Z, Nick P, editors. Applied Plant Cell Biology Tools and Approaches for Plant Biotechnology in series Plant Cell Monographs Eds Springer. 2014; 325–367.
  4. 4. Mondal S, Mandal C, Sangwan RS, Chandra S, Mandal C. Withanolide D induces apoptosis in leukemia by targeting the activation of neutral sphingomyelinase-ceramide cascade mediated by synergistic activation of c-Jun N-terminal kinase and p38 mitogen-activated protein kinase. Mol Cancer. 2010; 9: 239. pmid:20836852
  5. 5. Khedgikar V, Kushwaha P, Gautam J, Verma A, Changkija B, Kumar A, et al. Withaferin A: a proteasomal inhibitor promotes healing after injury and exerts anabolic effect on osteoporotic bone. Cell Death Dis. 2013; 4: e778 pmid:23969857
  6. 6. Sangwan RS, Chaurasiya ND, Lal P, Misra L, Tuli R, Sangwan NS. Withanolide A is inherently de novo biosynthesized in roots of the medicinal plant Ashwagandha (Withania somnifera). Physiol Plant. 2008; 133: 278–287. pmid:18312497
  7. 7. Chatterjee S, Srivastava S, Khalid A, Singh N, Sangwan RS, Sidhu OP, et al. Comprehensive metabolic fingerprinting of Withania somnifera leaf and root extracts. Phytochemistry. 2010; 71: 1085–1094. pmid:20483437
  8. 8. Mirjalili MH, Moyano E, Bonfill M, Cusido RM, Palazon J. Steroidal lactones from Withania somnifera, an ancient plant for novel medicine. Molecules. 2009; 14: 2373–2393. pmid:19633611
  9. 9. Chaurasiya ND, Sangwan NS, Sabir F, Misra L, Sangwan RS. Withanolide biosynthesis recruits both mevalonate and DOXP pathways of isoprenogenesis in Ashwagandha Withania somnifera L. (Dunal). Plant Cell Rep. 2012; 31:1889–1897. pmid:22733207
  10. 10. Madina BR, Sharma LK, Chaturvedi P, Sangwan RS, Tuli R. Purification and characterization of a novel glucosyltransferase specific to 27β-hydroxy steroidal lactones from Withania somnifera and its role in stress responses. Biochim Biophys Acta (BBA)—Proteins & Proteomics. 2007; 1774: 1199–1207.
  11. 11. Wang Z, Yeats T, Han H, Jetter R. Cloning and characterization of oxidosqualene cyclases from Kalanchoe daigremontiana: enzymes catalyzing up to ten rearrangement steps yielding friedelin and other triterpenoids. J Biol Chem. 2010; 285: 29703–29712. pmid:20610397
  12. 12. Gas-Pascual E, Berna A, Bach TJ, Schaller H. Plant oxidosqualene metabolism: cycloartenol synthase–dependent sterol biosynthesis in Nicotiana benthamiana. PLoS ONE. 2014; 9: e109156. pmid:25343375
  13. 13. Mosesa T, Polliera J, Shenf Q, Soetaertg S, Reede R, Erffelinck M-L, et al. OSC2 and CYP716A14v2 catalyze the biosynthesis of triterpenoids for the cuticle of aerial organs of Artemisia annua. The Plant Cell. 2015; 27: 286–301. pmid:25576188
  14. 14. Xue Z, Duan L, Liu D, Guo J, Ge S, Dicks J, et al. Divergent evolution of oxidosqualene cyclases in plants. New Phytologist. 2012; 193: 1022–1038. pmid:22150097
  15. 15. Sangwan NS, Mishra LN, Tripathi S, Kushwaha AK. Omics of secondary metabolic pathways in Withania somnifera Dunal (Ashwagandha). Omics Applications: In: Barh D, Taylor Francis (eds) Crop Science. 2013
  16. 16. Corey EJ, Matsuda SPT, Bartelt B. Isolation of an Arabidopsis thaliana gene encoding cycloartenol synthase by functional expression in a yeast mutant lacking lanosterol synthase by the use of a chromatographic screen. P Natl Acad Sci. 1993; 90: 11628–11632.
  17. 17. Hayashi H, Hiraoka N, Ikeshiro Y, Kushiro T, Morita M, Shibuya M, et al. Molecular cloning and characterization of a cDNA for Glycyrrhiza glabra cycloartenol synthase. Biol Pharm Bull. 2000; 23: 231–234. pmid:10706391
  18. 18. Morita M, Shibuya M, Lee MS, Sankawa U, Ebizuka Y. Molecular cloning of pea cDNA encoding cycloartenol synthase and its functional expression in yeast. Biol Pharm Bull. 1997; 20: 770–775. pmid:9255418
  19. 19. Basyuni M, Oku H, Tsujimoto E, Baba S. Cloning and functional expression of cycloartenol synthases from mangrove species Rhizophora stylosa Griff. and Kandelia candel (L.) Druce Biosci Biochem. 2007; 71: 1788–1792.
  20. 20. Shinozaki J, Shibuya M, Masuda K, Ebizuka Y. Squalene cyclase and oxidosqualene cyclase from a fern. FEBS Lett. 2008; 582: 310–318. pmid:18154734
  21. 21. Hayashi H, Huang P, Kirakosyan A, Inoue K, Hiraoka N, Ikeshiro Y, et al. Cloning and characterization of a cDNA encoding β-amyrin synthase involved in glycyrrhizin and soyasaponin biosynthesis in licorice. Biol Pharm Bull. 2000; 24: 912–916.
  22. 22. Sawai S, Uchiyama H, Mizuno S, Aoki T, Akashi T, Ayabe S, et al. Molecular characterization of an oxidosqualene cyclase that yields shionone, a unique tetracyclic triterpene ketone of Aster tataricus. FEBS Lett. 2011; 585: 1031–1036. pmid:21377465
  23. 23. Gupta P, Akhtar N, Tewari SK, Sangwan RS, Trivedi PK. Differential expression of farnesyl diphosphate synthase gene from Withania somnifera in different chemotypes and in response to elicitors. J Plant Growth Regul. 2011; 65: 93–100.
  24. 24. Gupta P, Agarwal AV, Akhtar N, Sangwan RS, Singh SP, Trivedi PK. Cloning and characterization of 2-C-methyl-D-erythritol-4-phosphate pathway genes for isoprenoid biosynthesis from Indian Ginseng, Withania somnifera. Protoplasma. 2013; 250: 285–295. pmid:22526204
  25. 25. Bhat WW, Lattoo SK, Razdan S, Dhar N, Rana S, Dhar RS, et al. Molecular cloning, bacterial expression and promoter analysis of squalene synthase from Withania somnifera (L.) Dunal. Gene. 2012; 499: 25–36. pmid:22425978
  26. 26. Akhtar N, Gupta P, Sangwan NS, Sangwan RS, Trivedi PK. Cloning and functional characterization of 3-hydroxy-3-methylglutaryl-Coenzyme A reductase gene from Withania somnifera: an important medicinal plant. Protoplasma. 2013; 250: 613–622. pmid:22936023
  27. 27. Srivastava S, Sangwan RS, Tripathi S, Mishra B, Narnoliya LK, Misra LN, et al. Light and auxin responsive cytochrome P450s from Withania somnifera Dunal: cloning, expression and molecular modelling of two pairs of homologue genes with differential regulation. Protoplasma. 2015. 252: 1421–1437. pmid:25687294
  28. 28. Kushwaha AK, Sangwan NS, Trivedi PK, Negi AS, Misra L, Sangwan RS (2013) Tropine forming tropinone reductase gene from Withania somnifera (Ashwagandha): biochemical characteristics of the recombinant enzyme and novel physiological overtones of tissue-wide gene expression patterns. PLoS ONE. 2013. 8:e74777. pmid:24086372
  29. 29. Razdan S, Bhat WW, Rana S, Dhar N, Lattoo SK, Dhar RS, et al. Molecular characterization and promoter analysis of squalene epoxidase gene from Withania somnifera (L.) Dunal. Mol Biol Rep. 2013; 40: 905–916. pmid:23065254
  30. 30. Rana S, Lattoo SK, Dhar N, Razdan S, Bhat WW, Dhar RS, et al. NADPH-Cytochrome P450 Reductase: molecular cloning and functional characterization of two paralogs from Withania somnifera (L.) Dunal. PLoS ONE. 2013; 8: 1–16.
  31. 31. Dhar N, Rana S, Bhat WW, Razdan S, Pandith SA, Khan S, et al. Dynamics of withanolide biosynthesis in relation to temporal expression pattern of metabolic genes in Withania somnifera (L.) Dunal: a comparative study in two morpho-chemovariants. Mol Biol Rep. 2013; 40: 7007–7016. pmid:24190485
  32. 32. Sabir F, Mishra S, Sangwan RS, Jadaun JS, Sangwan NS. Qualitative and quantitative variations in withanolides and expression of some pathway genes during different stages of morphogenesis in Withania somnifera Dunal. Protoplasma. 2014; 250: 539–549.
  33. 33. Gupta P, Goel R, Pathak S, Srivastava A, Singh SP, Sangwan RS, et al. De Novo assembly, functional annotation and comparative analysis of Withania somnifera leaf and root transcriptomes to identify putative genes involved in the withanolides biosynthesis. PLoS ONE. 2013; 8: 1–12.
  34. 34. Senthil K, Jayakodi M, Thirugnanasambantham P, Lee SC, Duraisamy P, Purushotham PM, et al. Transcriptome analysis reveals in vitro cultured Withania somnifera leaf and root tissues as a promising source for targeted withanolide biosynthesis. BMC Genomics. 2015; 16: 14. pmid:25608483
  35. 35. Zarate R, Jaber-Vazdekis N, Verpoorte R. Metabolic engineering of plant cellular metabolism: methodologies, advances, and future directions. Biotechnology for medicinal plants. 2013; 359–393.
  36. 36. Meister G, Tuschl T. Mechanisms of gene silencing by double stranded RNA. Nature. 2004; 431: 343–349. pmid:15372041
  37. 37. Jiang F, Wang JY, Jia HF, Jia WS, Wang HQ, Xiao M. RNAi-mediated silencing of the flavanone 3-hydroxylase gene and its effect on flavonoid biosynthesis in strawberry fruit. J Plant Growth Regul. 2013; 32: 182–190.
  38. 38. Sunilkumar G, Campbell LM, Puckhaber L, Stipanovic RD, Rathore KS. Engineering cotton seed for use in human nutrition by tissue-specific reduction of toxic gossypol. Proc. Natl Acad Sci. 2006; 103: 18054–18059. pmid:17110445
  39. 39. Han JY, In JG, Kwon YS, Choi YE. Regulation of ginsenoside and phytosterol biosynthesis by RNA interferences of squalene epoxidase gene in Panax ginseng. Phytochemistry. 2010; 71: 36–46. pmid:19857882
  40. 40. Runguphan W, Maresh JJ, Connor SEO. Silencing of tryptamine biosynthesis for production of nonnatural alkaloids in plant culture. P Natl Acad Sci. 2009; 106: 13673–13678.
  41. 41. Takagi K, Nishizawa K, Hirose A, Kita A, Ishimoto M. Manipulation of saponin biosynthesis by RNA interference-mediated silencing of β-amyrin synthase gene expression in soybean. Plant Cell Rep. 2011; 30: 1835–1846. pmid:21630021
  42. 42. Mishra S, Bansal S, Sangwan RS, Sangwan NS. Genotype independent and efficient Agrobacterium-mediated transformation of the medicinal plant Withania somnifera Dunal. J Plant Biochem Biotechnol. 2015.
  43. 43. Sambrook J, Russell DW. Molecular cloning: a laboratory manual, 3rd edn. Cold Spring Harbor Laboratory, New York; 2001.
  44. 44. Sangwan RS, Chaurasiya ND, Lal P, Misra L, Uniyal GC, Tuli R, et al. Withanolide A biogeneration in in vitro shoot cultures of Ashwagandha (Withania somnifera Dunal), a main medicinal plant in ayurveda. Chem Pharma Bull. 2007; 55: 1371–1375.
  45. 45. Sil B, Mukherjee C, Jha S, Mitra A. Metabolic shift from withasteroid formation to phenylpropanoid accumulation in cryptogein-cotransformed hairy roots of Withania somnifera (L.) Dunal. Protoplasma. 2015; 252: 1097–1110. pmid:25534257
  46. 46. Eamens A, Wang MB, Smith NA, Waterhouse PM. RNA Silencing in plants: yesterday, today, and tomorrow. Plant Physiol. 2008; 147: 456–468. pmid:18524877
  47. 47. Wang P, Liang Z, Zeng J, Li W, Sun X, Miao Z, et al. Generation of tobacco lines with widely different reduction in nicotine levels via RNA silencing approaches. Journal of Biosciences. 2008; 33:177–184. pmid:18535352
  48. 48. Mohanpuria P, Kumar V, Ahuja PS, Yadav SK. Agrobacterium-mediated silencing of caffeine synthesis through root transformation in Camellia sinensis L. Molecular Biotechnology. 2011; 48:235–243. pmid:21181507
  49. 49. Han JY, Kwon YK, Yang DC, Jung YR, Choi YE. Expression and RNA interference-induced silencing of the dammarenediol synthase gene in Panax ginseng. Plant Cell Physiol. 2006; 47: 1653–1662. pmid:17088293
  50. 50. Hoffmann T, Kalinowski G, Schwab W. RNAi-induced silencing of gene expression in strawberry fruit (Fragaria x ananassa) by agroinfiltration: a rapid assay for gene function analysis. Plant J. 2006; 48: 818–826. pmid:17092319
  51. 51. Abdeeva I, Abdeev R, Bruskin S, Piruzian E. Transgenic plants as a tool for plant functional genomics, Transgenic Plants—Advances and Limitations, PhD. Yelda Ozden Çiftçi (Ed.), ISBN: 978-953-51-0181-9.
  52. 52. Ma C, Ma B, He J, Hao Q, Lu X, Wang L. Regulation of carotenoid content in tomato by silencing of lycopene β/ε-cyclase genes. Plant Mol Biol Rep. 2011; 29: 117–124.
  53. 53. Coleman HD, Park JY, Nair R, Chapple C, Mansfield SD. RNAi-mediated suppression of p-coumaroyl-CoA 3’-hydroxylase in hybrid poplar impacts lignin deposition and soluble secondary metabolism. Proc. Natl Acad Sci. 2008; 105: 4501–4506. pmid:18316744
  54. 54. Pan Q, Wang Q, Yuan F, Xing S, Zhao J, Choi YH, et al. Overexpression of ORCA3 and G10H in Catharanthus roseus plants regulated alkaloid biosynthesis and metabolism revealed by NMR-metabolomics. PLoS ONE. 2012; 7: e43038. pmid:22916202
  55. 55. Ambrosio GD, Stigliani AL, Giorio G. Overexpression of CrtR-b2 (carotene beta hydroxylase 2) from S. lycopersicum L. differentially affects xanthophyll synthesis and accumulation in transgenic tomato plants Transgenic Res. 2011; 20: 47–60. pmid:20383744
  56. 56. Verpoorte R, Memelink J. Engineering secondary metabolite production in plants. Current Opinion in Biotechnol. 2002; 13: 181–187.