lable at ScienceDirect

Plant Physiology and Biochemistry 48 (2010) 310e318

Contents lists avai

Plant Physiology and Biochemistry

journal homepage: www.elsevier .com/locate/plaphy

Research article

Induced biosynthesis of resveratrol and the prenylated stilbenoids arachidin-1and arachidin-3 in hairy root cultures of peanut: Effects of culture mediumand growth stage

Jose Condori a, Ganapathy Sivakumar a, John Hubstenberger a, Maureen C. Dolan a,b,Victor S. Sobolev c, Fabricio Medina-Bolivar a,b,d,*aArkansas Biosciences Institute, Arkansas State University, P.O. Box 639, State University, AR 72467, USAbNature West Inc., Jonesboro, AR 72401, USAcNational Peanut Research Laboratory, Agricultural Research Service, U.S. Department of Agriculture, P.O. Box 509, Dawson, GA 39842, USAdDepartment of Biological Sciences, Arkansas State University, P.O. Box 599, State University, AR 72467, USA

a r t i c l e i n f o

Article history:Received 21 October 2009Accepted 12 January 2010Available online 21 January 2010

Keywords:Hairy rootResveratrolArachidin-1Arachidin-3StilbenoidPeanutPrenylation

* Corresponding author at: Arkansas BioscienceUniversity, P.O. Box 639, State University, AR 72467, Ufax: þ1 (870) 680 4348.

E-mail address: fmedinabolivar@astate.edu (F. Me

0981-9428/$ e see front matter � 2010 Elsevier Masdoi:10.1016/j.plaphy.2010.01.008

a b s t r a c t

Previously, we have shown that hairy root cultures of peanut provide a controlled, sustainable andscalable production system that can be induced to produce stilbenoids. However to leverage peanut hairyroots to study the biosynthesis of this polyphenolic biosynthetic pathway, growing conditions andelicitation kinetics of these tissue cultures must be defined and understood. To this end, a new peanut cv.Hull hairy root (line 3) that produces resveratrol and its prenylated analogues arachidin-1 and arachidin-3 upon sodium acetate-mediated elicitation was established. Two culture media were compared forimpact on root growth and stilbenoid biosynthesis/secretion. The levels of ammonium, nitrate, phos-phate and residual sugars were monitored along growth and elicitation period. A modified MS (MSV)medium resulted in higher root biomass when compared to B5 medium. The stilbenoid profile afterelicitation varied depending on the age of the culture (6, 9, 12, and 15-day old). After elicitation at day9 (exponential growth in MSV medium), over 90% of the total resveratrol, arachidin-1 and arachidin-3accumulated in the medium. Our studies demonstrate the benefits of the hairy root culture system tostudy the biosynthesis of stilbenoids including valuable prenylated polyphenolic compounds.

� 2010 Elsevier Masson SAS. All rights reserved.

1. Introduction

Plants arewell-known for their ability to synthesize awide arrayof low molecular weight specialized compounds known assecondary metabolites [1]. However, the production levels of thesechemicals in nature can be quite low, often transient and highlydependent upon the physiological and developmental stage of thesource plant [2]. Dramatic increases in the levels of these secondarymetabolites often results when a plant's defense system is activatedby pathogens and environmental stresses [3,4].

Hairy root cultures have emerged as a sustainable, tissue-basedsystem that preserves the multiple specialized cell types importantin maintaining consistent and reproducible synthesis of complex

s Institute, Arkansas StateSA. Tel.: þ1 (870) 680 4319;

dina-Bolivar).

son SAS. All rights reserved.

small molecules without the use of externally-supplied hormones.These genetically and biochemically stable organ cultures arecapable of producing substantial biomass that is linkedwith greaterlevels of secreted, bioactive small molecules when compared tofield-grown plants or other plant tissue/cell culture systems [5e8].Indeed, we and others [9e12] have shown that these roots can beselectively manipulated to produce various classes of small bioac-tive molecules. Recently, we successfully established peanut hairyroots (Arachis hypogaea) with biosynthetic capacity for producingresveratrol [12]. Resveratrol belongs to the class of phenyl-propanoid compounds known as stilbenoids or stilbenes. Thesepolyphenolic compounds have been shown to confer protection tothe peanut plant in countering pathogens and therefore arebelieved to function as phytoalexins. Because of their additionalantioxidant properties, stilbenoids may provide peanuts and otherplants supporting this metabolic pathway, protection againstoxidative stress and other environmental challenges includingozone and UV light exposure [13].

J. Condori et al. / Plant Physiology and Biochemistry 48 (2010) 310e318 311

In addition to their purported function as phytoalexins inplanta, the stilbenoids exhibit significant antioxidant and anti-inflammatory properties in mammalian cells [14]. We havepreviously shown that hairy roots exposed for 24 h to sodiumacetate resulted in at least a 60-fold increase of resveratrol levelswhen compared to non-treated roots. Furthermore, most of thisresveratrol was secreted into the culture media [12]. Interestingly,in addition to resveratrol peanut plants challenged with fungalpathogens are known to produce several other stilbenoids,including a number of unique prenylated resveratrol analogues[15,16]. Therefore, to enable us to explore the complex biosyn-thesis of this class of phytochemicals produced by the peanuthairy root system, we have characterized the production ofresveratrol in comparison to two prenylated stilbenoids, arach-idin-1 and arachidin-3 (Fig. 1). In the present work, a hairy rootline from a peanut cultivar different from that described in ourprevious study [12] was established and the growth kineticsand factors affecting the biosynthesis of resveratrol, arachidin-1and arachidin-3 were characterized. Furthermore, parametersincluding nutrient conditions of the media, pH and conductivityof the culture system, nutrient metabolism and induction of theculture at different ages were measured and the concomitantimpact of these factors on the biosynthesis and secretion ofresveratrol and its prenylated analogues was evaluated. Asrecently reviewed by Yazaki et al. [17], prenylation plays a majorrole in the diversification of aromatic compounds. The presenceof the isoprenoid side chain in the prenylated stilbenoids mayprovide a key determinant of their biological activity as describedfor prenylated flavonoids [18,19]. As far as we know, the peanuthairy root culture system developed in this study provides for thefirst time a controlled and sustainable tissue culture system forcharacterizing the biosynthesis of prenylated stilbenoids andenabling discovery of stilbenoid prenyltransferases. The cultureparameters that ensure a reproducible model system for studyingthese polyphenolic compounds are discussed.

Fig. 1. Chemicals structures of trans-resveratrol (1), trans-arachidin-1 (2) and trans-arachidin-3 (3).

2. Materials and methods

2.1. Establishment and molecular characterization of hairy roots

Seeds of peanut (A. hypogaea) cv. Hull [20], kindly provided byDr. Daniel Gorbet (University of Florida), were surface sterilized andgerminated under aseptic conditions as previously described [12].Cotyledonary leaves were excised from in vitro seedlings andinoculated with Agrobacterium rhizogenes strain ATCC 15834 togenerate hairy roots according to a methodology previouslydescribed by our laboratory [12]. Among several hairy root linesestablished, line 3 was selected for its vigorous and sustainedgrowth and used in further analysis. Genomic DNA was extractedfrom line 3 hairy roots using the DNeasy� Plant Mini kit (Qiagen,USA). PCR analyses of rolC, aux1 and virD2 genes were performedaccording to Medina-Bolivar et al. [12].

2.2. Tissue culture media

Two culture media containing 3% sucrose were tested. In addi-tion to the standard Gamborg's B5 medium [21], a Murashige andSkoog's (MS) medium [22] modified from [23] to contain 7.7 mMNH4þ (fromNH4NO3) and 22.85 mMNO3

� (fromNH4NO3 and KNO3)was used and hereafter referred to as MSV medium. Among otherdifferences between MSV and MS [22] are pH values of the mediaprior to autoclaving (pH6 .0 for MSV and pH 5.7 for MS).

2.3. Growth kinetics

To establish growth kinetics of the hairy root line 3, ten root tips(2e3 cm long) were inoculated into 250 ml flasks containing 50 mlof B5 or MSV medium. The flask cultures were grown on an orbitalshaker (Innova� 44R, New Brunswick Scientific) at 90 rpm, 28 �Cand maintained under continuous darkness. Three culture flaskswere harvested every 3 days through day 30. Harvested roots wererinsed with tap water three times, blotted on paper towels andfresh weights (FW) measured. The roots were placed in 60 �C ovenfor�2 days to obtain the dry weight (DW). The specific growth rate(m) was calculated as:

m ¼Ln

DWi

DW0Dt

where DWi, is average dry weight of the roots (g) at the end of theexponential growth (12 days for MSV and 15 days for B5); DW0, isaverage dry weight of inoculi (g) at the start of the exponentialgrowth (day three); t is interval of time (in days) between 0 and i(9 days for MSV and 12 days for B5). Conductivity and pH of themedium were measured using a SevenEasy conductivity meter(Mettler Toledo) and an Accumet� Basic AB15 pH meter (FisherScientific), respectively. An aliquot of mediumwas stored at �20 �Cfor nutrient analyses.

2.4. Elicitation

Sodium acetate (10.2 mM) was used as elicitor as describedpreviously [12]. In this study, elicitations were performed at variousphases of hairy root culture growth as indicated. Prior to elicitation,the spent medium was collected from each culture and pH andconductivity measured as detailed above. Fresh medium, contain-ing sodium acetate, was added to each culture. All elicitations werecarried out for 24 h under continuous darkness at 28 �C. Followingelicitation, roots and media were collected. The roots were har-vested, FWmeasured, and immediately frozenwith liquid nitrogen

J. Condori et al. / Plant Physiology and Biochemistry 48 (2010) 310e318312

and stored at �80 �C. Frozen roots were ground with a mortar/pestle in the presence of liquid nitrogen. The processed tissue wascollected in a falcon tube (50 mL), previously weighed, andlyophilized in a FreeZone 4.5 freeze drying apparatus (Labconco).The weight of the falcon tube with the dried tissue was recordedand the difference with the previous value was used as the mass ofthe DW. Lyophilized tissue was kept at �80 �C for further analyses.The pH and conductivity of the media collected after elicitationwasmeasured as described above. An aliquot of medium was stored at�20 �C for nutrient analysis. As controls, fresh media withoutelicitor were used. All experiments were performed using threebiological replicates per treatment.

2.5. Nutrient analyses of the medium

The concentrations of ammonium, nitrate, phosphate andresidual sugars in themediumwere analyzed using a 96-well plate-based method as outlined below:

Ammonium concentration was determined by a modification ofthe colorimetric method described in the Technical Bulletin No. 640from Sigma. Briefly, in a flat bottom 96-well plate (Fisher12565501), 50 mL of nitroprusside solution (Sigma P6994) wasmixed with 25 mL of standard or sample. Then, 50 mL of alkalinehypochloride solution (Sigma A1727) was added and mixed bypipetting. Finally 175 mL of Millipore water was added and mixed.Samples were incubated at room temperature for 20 min andabsorbance measured at 570 nm. NH4NO3 (Fisher A676) was usedin establishing standard solutions (from 2 mg/L to 40 mg/L).

Nitrate concentration was determined by a modification of themethod described by Kim et al. [24]. A volume of 25 mL of standardor sample was added to 250 mL of 0.5% Szechrome NAS (Poly-sciences, Inc.). Samples were mixed and incubated at roomtemperature for 20 min prior to measuring absorbance at 570 nm.KNO3 (Sigma P8291) was employed to make standard solutions(from 6 mg/L to 48 mg/L).

Phosphate concentration was measured using a modification ofthe method described by Towler et al. [25]. A volume of 20 mL ofstandard or sample was added to 280 mL of phosphate reagent [25].Samples were mixed by pipetting, and incubated at room temper-ature for 10 min prior to measuring absorbance at 650 nm. KH2PO4(Sigma P5379) was employed to make standard solutions (from14 mg/L to 280 mg/L).

Residual sugars were analyzed using a modified spectophoto-metric method described by Dubois et al. [26]. Briefly, 20 mL ofstandard or mediumwas added to 40 mL of 5% phenol andmixed bypipetting. Concentrated sulfuric acid (200 mL) was added andmixedby pipetting. Samples were incubated at room temperature for20 min and absorbance was measured at 490 nm. Glucose (SigmaG5400) was used in making standard solutions (from 200 mg/L to1000 mg/L).

All readings were carried out at the specific wavelength indi-cated using a PowerWave�Microplate spectophotometer (Bio-Tek)with the KCjunior software (Bio-Tek). Some medium samples werediluted with Millipore water in order to be within the standardrange. All standards weremadewithMilliporewater andwere usedfresh. Each sample was read in triplicate.

2.6. Extraction of stilbenoids

Lyophilized root tissue samples were vortexed for 1 min in ethylacetate (EtOAc) (ratio of 0.05 g DW tissue to 1 mL solvent); soni-cated for 30 s and followed by centrifugation for 8 min at 16,100� gto remove undissolved material. Supernatant fractions weretransferred to amber glass tubes and evaporated to dryness undernitrogen stream using the Reacti-Vap� III apparatus (Pierce). To

extract stilbenoids from the media samples, the latter was parti-tioned twice with EtOAc (80% v/v and 60% v/v). The combinedEtOAc phases were evaporated to dryness under vacuum at 40 �C ina rotary evaporator (Büchi, rotavapor R-200). Extracts were resus-pended in 1.5 mL of EtOAC, transferred into amber vials and dried tocompleteness under nitrogen stream using the Reacti-Vap� IIIapparatus (Pierce). The extracts were resuspended in 300 mL(medium samples) and 200 mL (root tissue samples) of methanol(MeOH) and filtered through 0.2 mmnylon filter. Samples were keptat �20 �C until they were analyzed. Ten mL of the filtrate was usedfor HPLC analyses.

2.7. Preparation of arachidin-1 and arachidin-3 reference standards

2.7.1. Plant material and processingPeanut seeds of the 31e1314 peanut runner breeding line (2007

harvest) from the National Peanut Research Laboratory (Dawson,GA) were processed as described [27].

2.7.2. Purification of standardsChromatographically pure arachidin-1 and arachidin-3 were

obtained from elicited peanut seeds as described [28] except thatpreparative HPLC was used as a final purification step rather thanpreparative TLC. HPLC separation was achieved by using isocraticmobile phase 3 [27]. Trans-arachidin-1 and trans-arachidin-3 wereisolated as off-white solids. Their identities were confirmed by ESI-MS2 and UV spectrometry (these data are given in parenthesis as[M�H]� values followed by UV absorption maxima): trans-arach-idin-1 (m/z 311; 339 nm); trans-arachidin-3 (m/z 295; 335 nm).

2.8. HPLC-fluorescence/photodiode array detection of stilbenoids

Analyses were performed by reverse phase HPLC as describedpreviously [29]. Concentrations of trans-resveratrol, trans-arach-idin-1, and trans-arachidin-3 in the root and culture media extractswere determined based on peak areas of corresponding purestandards. Calibration curves were done using either a fluorescencedetector (excitation 335 nm and emission 374 nm) or absorbance(320 nm). A 280 nm absorbance calibration curve was used for twosamples in order to quantify trans-resveratrol, because their valueswere over the detection range of fluorescence (374 nm) andabsorbance (320 nm). Reference compounds were trans-resveratrol(trans-3,40,5-trihydroxystilbene) purchased from Sigma. Trans-arachidin-1 and trans-arachidin-3 purified as detailed above.Standards solutions for HPLC were prepared by dissolving thereference compounds in MeOH. Dilutions of the standards solu-tions were done with MeOH to obtain concentrations from 5 to80 ng/mL for trans-resveratrol; 1.5 to 800 ng/mL for trans-arachidin-1, and 1.25 to 800 ng/mL for trans-arachidin-3.

2.9. Mass-spectrometry-electrospray ionization (MS-ESI) analysis

For mass spectrometry analyses a Varian LC-MS (Walnut Creek,CA) instrument consisting of a ProStar 420 autosampler, ProStar 210pump, 212-LC solvent delivery system and a 320 triple quadrupolemass spectrometer equipped with an electrospray ionizationsource was used. The system was monitored using a DELL PCcomputer equipped with Varian MS Workstation software version6.9 for instrument control, data acquisition and data processing,respectively. The chromatographic separation was done asrecently described by our laboratory [29]. The temperatures of theelectrospray source and desolvation gas were 60 �C and 250 �C,respectively. MS-ESI detection was done in the negative ions[M�H]� mode by scanning over the 100 - 2000 m/z range withunit mass resolution. The optimized electrospray capillary voltage

J. Condori et al. / Plant Physiology and Biochemistry 48 (2010) 310e318 313

was set to 30 V and spray voltage to 4700 V. Nitrogen was used asdrying gas for solvent evaporation. Prior to sample analysis, the MSresponse was optimized using a solution of trans-resveratrol inMeOH15mg/ml. Themost parent ion [M�H]�m/z227wasobtainedwith the aforementioned condition. These conditions were estab-lished prior to a software controlled autotune on this ion. Trans-arachidin-1 and trans-arachidin-3 were detected in the samplesbased on the retention time and UV spectrum of the referencestandards and the ion chromatogram extracted at m/z values cor-responding to the deprotonated species [M�H]�molecular ionm/z311 for trans-arachidin-1 and m/z 295 for trans-arachidin-3.

2.10. Statistical analyses

Analyses were done using GraphPad Prim 5 software (GraphPadSoftware, Inc). In the case of comparison of different treatmentsversus a specific condition, a one-way analysis of variance (ANOVA)using a Dunnett's multiple comparison test was performed (witha¼ 0.01). Comparison between all pairs of treatmentwas donewithANOVA using Tukey's multiple comparison test (with a¼ 0.01).

3. Results and discussion

3.1. Establishment of hairy root cultures

Infection of peanut cv. Hull explants with A. rhizogenes strainATCC 15834 produced several hairy root lines. Among them, line 3

Fig. 2. Comparison of growth of hairy root of peanut cv. Hull line 3 in MSV and B5 media. (grown in two different culture media. Pictures are representative of triplicate cultures. Scale30. Each value in the curve represents is the average of three independent cultures; errordifferent stages of growth. Values represent the average of three independent flasks; error balegends, the reader is refered to the web version of this article.)

was selected based on its sustained growth in liquid culture. PCRanalysis of line 3 was used in confirming the integration of the twoT-DNAs from the Ri plasmid of A. rhizogenes strain ATCC 15834(pRi15834) into the peanut genome. As two T-DNAs, the left (TL-DNA) and right (TR-DNA), have been described for agropine typeAgrobacterium strains such as ATCC 15834, it was important thatour PCR analyses target a fragment of the rolC and aux1 genes thatare located on independent TL- and TR-DNAs, respectively. Asshown in Supplementary Fig. S1, the rolC and aux1 were present inline 3, indicating the co-integration of both T-DNAs. Furthermore,PCR targeting a portion of virD2 gene, which is located outside bothof these T-DNAs, was negative and thus confirmed the absence ofany remaining Agrobacterium.

3.2. Effect of MSV and B5 culture media on root growth

Media composition can have a significant impact on root growthin culture systems [23]. To this end, we compared the effects of twoculture media on the growth of peanut cv. Hull line 3 hairy roots.Gamborg's B5 medium served as a reference, as we previously usedB5 to grow peanut hairy roots of cv. Andru II [12]. In this study wecompared B5 with a modified MS medium (MSV) that differs fromMS and B5 media in the ratio of ammonia to nitrate concentration.A growth kinetics analysis was carried out on line 3 hairy rootsgrown over a period of 30 days. All culture flasks in the study wereinitiated with 10 tips (average of 0.28 g FW/flask). While thephenotype of root tissue in the two media was very similar from

A) Depicts root phenotype and coloration throughout the 30 day culture period whenbar¼ 2 cm. (B) Growth kinetics in MSV and B5 media at select time points through daybars represent standard deviation. (C) Medium pH (red) and conductivity (green) atrs represent standard deviation. (For interpretation of the references to colour in figure

J. Condori et al. / Plant Physiology and Biochemistry 48 (2010) 310e318314

day 0 to day 18 (Fig. 2A), a yellowish color was observed at day 18 inthe roots grown in B5 medium. Interestingly, this color was notobserved until day 30 for roots grown in MSV medium (Fig. 2A).A corresponding color change in the culture media was alsoobserved that is likely associated with the secretion of newlysynthesized root metabolites into the culture medium. It is possiblethat at certain developmental stages, the roots are more prone toactivate metabolic pathways associated with the production andsecretion of these metabolites.

Fig. 2B compares the growth curves for line 3 roots grown inMSV and B5 media. Under either media condition, a lag phase up today 3 was observed. A longer exponential growth phase wasobserved in roots cultured in B5 medium (days 3e15) in compar-ison to MSV medium (days 3e12). Consequently, the roots main-tained longer stationary phase in MSV medium as compared to B5medium. In addition, hairy roots grown in MSV medium presenteda higher specific growth rate (m) when compared to roots grown inB5 medium. The m calculated during the exponential growth phasewas 0.318 day�1 in MSV medium and 0.258 day�1 in B5 medium.

3.3. Media analysis during root growth

To better understand the nutrient utilization of peanut hairyroot cultures, the pH, conductivity and nutrient composition of themedia was monitored throughout the growth of the hairy rootcultures. The pH measurement detects the ionized H (Hþ) in themedium, whereas conductivity measures the total ionic composi-tion change (Hþ and other ions) in the medium [30]. In ourexperiments we established that the original pH values for MSV(pH 6.0) and B5 (pH 5.7) media were reduced to 5.540 and 5.270,respectively post-autoclave. Autoclaving has been described tocause a pH decrease of 0.3 units in MS medium [31] and similarvalues were observed for the two media evaluated in this study.

Fig. 3. Nutrient composition analysis of the culture media over the growth period of the peaand residual sugars (D) was monitored during the growth of peanut hairy root cultures in Merror bars representing the standard deviation.

Fig. 2C shows that in both media, the pH dropped in the first threedays and then increased, as reported for cell suspension of coffee[32] and peanut hairy root cv. Andru II [12].

As shown in Fig. 2C, the conductivity of the media decreasedinversely to increases in the biomass (DW), suggesting that nutri-ents were taken up by the roots from the media. A high inversecorrelation (r2¼ 0.98) between conductivity and biomass throughexponential growth was observed for bothmedia tested. No changein conductivity was observed when the cultures reached theirstationary phase.

The nutrient composition analysis of the media was targeted tothree main nutrients involved in the regulation of plant metabo-lism, i.e. nitrogen, phosphorus and carbon. Ammonium and nitrateare inorganic nitrogen sources for plant cells. As shown Fig. 3A,ammoniumwas completely consumed by day 6 in B5 medium andday 9 for MSV medium-grown roots, while nitrate was depleted atday 12 in MSV and day 15 in B5 medium (Fig. 3B). These resultssuggest that ammonium was the preferred nitrogen source as longas it was available in the media. It can be explained by the fact thatin plant cells, ammonium is a more direct nitrogen source thannitrate for the synthesis of amides and amino acids [33,34]. Inaddition, media acidity has been related with an enhancement inthe uptake of nitrate ions [35]. As shown in Fig. 2C, the MSV driftedto and maintained a more acidic pH throughout the growth periodas compared to B5 media.

Phosphate is the inorganic phosphorus source for plants. It playsan important role as a structural element of nucleic acids, high-energy compounds, and phospholipids of biomembranes. In addi-tion, this mineral is important in the energy conservation, carbonassimilation and regulation of enzyme activity through phosphor-ylation/dephosphorylation processes in plants [36]. While MSVmedium contains higher phosphate content than B5, Fig. 3C showsthat the levels of phosphate were similar for both media by day 3.

nut cv. Hull hairy root line 3. Consumption of ammonium (A), nitrate (B), phosphate (C),SV and B5 media. Values represent the average of three replicates and expressed with

Fig. 4. Impact of elicitation on peanut cv. Hull hairy root line 3 culture phenotype and growth performance. (A) Dry weight of line 3 after elicitation at various culture agescomparing roots grown in MSV versus B5 media. Non-elicited cultures established the control samples for comparing elicited cultures. Values represent the average of threereplicates and error bars are calculated based on standard deviation. (B) Pictures represent the root phenotype after 24 h elicitation under both MSV and B5 media growthconditions. Control flasks represent non-elicited cultures. Each culture age monitored and measured represents three independent cultures.

J. Condori et al. / Plant Physiology and Biochemistry 48 (2010) 310e318 315

Residual sugar was used for monitoring the consumption ofsucrose (carbon source) in the media. As shown in Fig. 3D, sucrosewas depleted faster in MSV medium (day 15) than in B5 medium(day21) despite the fact that bothmedium initially contain the sameamount of sucrose (30 g/L). The ratio of driedweight to freshweight(DW/FW)has been used as an indicator of the concentration sucrosein the media as well as an index of the cell water content or cell size[37]. In our study the DW/FW ratio of the roots grown in eithermediawas relatively constant during the exponential growth phase(from days 6e9) and then decreased in the stationary phase, similarto that observed by Park and Kim [37] for cell suspension cultures.

3.4. Effect of elicitation on root culture phenotypeand growth performance

In order to study the effect of culture age on the production ofthe stilbenoids, it was important to characterize the impact ofsodium acetate elicitation at specific growth stages (days 6, 9, 12,

Fig. 5. Comparison of conductivity (A) and pH (B) following elicitation of peanut cv. Hull haiMSV or B5 medium after autoclaving and addition of the elicitor (sodium acetate). Values a(P< 0.01), using Dunnett's multiple comparison test against media was performed.

and 15; Fig. 2B). In general, the root biomass was inhibited underconditions of elicitation compared with matching control rootcultures (non-elicited); this response was independent of media(Fig. 4A). Furthermore, at all stages tested, 24 h-elicited culturesshowed significant indications of stress (browning color in the roottissue) in comparison to the corresponding control cultures (Fig. 4B).

Addition of sodium acetate to the culturemedium also increasedMSV and B5 media pH values by 0.84 and 1.04 units, respectively.Likewise the conductivity values were also increased with elicita-tion (MSV by 1.04 units and B5 by 1.10 units). The mere presence ofthe peanut roots in the MSV medium resulted in significant(p< 0.01) decreases in medium pH, in both the presence andabsence of elicitor (Fig. 5A). A similar observation was reportedwhen Cucumis calli were added to culture media [35]. In this casethe authors suggested that the plant cells have an active role intrying to reach an optimum pH environment. The conductivityvalues at different ages, compared to initial medium value (with orwithout elicitor), decreased significantly (p< 0.01), under both

ry root cultures line 3 in MSV and B5 media. Medium corresponds with the readout forre the average of three replicates and error bars represent standard deviation. ANOVA

J. Condori et al. / Plant Physiology and Biochemistry 48 (2010) 310e318316

culture media conditions (Fig. 5B). The decrease of conductivity isrelated to the increase of root culture biomass which implies thatmore nutrients are being utilized from the media by non-elicited(control) than elicited roots during the 24-h elicitation period.

Fig. 6. Nutrient analysis of media following elicitation of peanut cv. Hull hairy root cultuincubation in fresh MSV (A, E, C, G) or B5 (B, F, D, H) media in the presence or absence ofdeviation of triplicate samples.

Overall nutrient consumption in control (non-elicited) cultureswas higher than in elicited conditions in either MSV or B5 media(Fig. 6), suggesting that under elicited conditions the roots reducedthe nutrient uptake rate. This suggests that plant cells under stress

re line 3. Hairy roots were cultured for 6, 9, 12 and 15 days and subjected to a 24 hthe elicitor. Values are the average of three replicates. Error bars represent standard

Table 1Stilbenoids secreted into the media by peanut hairy roots of different age afterincubation by 24 h with fresh media (control) or fresh media with sodium acetate(elicited).

Age Mean� SD, nmol/g dry weight root; n¼ 3

trans-resveratrol trans-arachidin-1 trans-arachidin-3

J. Condori et al. / Plant Physiology and Biochemistry 48 (2010) 310e318 317

conditions re-direct their energy away from production of primarymetabolites to secondary metabolism in order to better tolerateand survive the unfavorable environment. Interestingly, in the caseof residual sugars, there was not a significant difference (p< 0.01)between control and elicited cultures at each culture age (Fig. 6Gand H), in either medium.

Control Elicited Control Elicited Control Elicited

MSV6 14.2� 2.6 420.7� 21.1 nda nd nd 35.3� 3.59 8.9� 3.9 393.2� 168.7 nd 107.0� 51.3 nd 124.1� 16.912 1.5� 2.6 147.4� 73.9 nd 2.7� 2.4 nd 42.9� 36.215 2.2� 3.8 92.4� 51.8 nd nd nd 8.5� 8.9

B56 10.8� 2.7 168.7� 22.2 nd 6.1� 5.4 nd 35.1� 10.59 nd 90.3� 17.2 nd 54.3� 50.5 nd 84.2� 23.812 0.8� 1.4 30.3� 16.7 nd 1.5� 1.2 nd 23.4� 22.215 0.9� 1.5 24.6� 14.5 nd 48.1� 36.0 nd 58.3� 33.4

a Not detected.

3.5. Effect of elicitation on biosynthesis and secretion of stilbenoids

The effect of media environment and elicited root culture age onthe levels of three stilbenoids (trans-resveratrol and two prenylatedstilbenoids trans-arachidin-1 and trans-arachidin-3) was assessedusing a combined HPLC separation method coupled with fluores-cence and PDA detection. A typical HPLC profile of the stilbenoidssecreted into the culture medium under non-elicited andelicited conditions is shown in Fig. 7. The stilbenoids, in particulararachidin-1 and arachidin-3, showed higher fluorescence whencompared to resveratrol (Fig. 7B).

HPLC analyses revealed that trans-resveratrol remained themajor secreted stilbenoid in either MSV or B5 media after 24 htreatment with sodium acetate (Table 1). These results are similarto that reported previously for elicited peanut seedlings [38]. In thisprevious study, trans-resveratrol was found to be the majormetabolite produced after 24 h of fungal infection of peanutkernels. In our comparison of MSV and B5 media, we found that inMSVmedium the level of trans-resveratrol was higher at all cultureages studied (Table 1). Roots elicited either at day 6 or 9 were notsignificantly different (p< 0.01) in the production of trans-resver-atrol. In addition to trans-resveratrol, we also analyzed for 2 pre-nylated stilbenoids, i.e. trans-arachidin-1 and trans-arachidin-3.Overall, we observed higher variability in the levels of these twostilbenoids when compared to resveratrol in either B5 or MSVmedium as reflected by the high SD values. Trans-arachidin-3 wasthe second major phytoalexin produced by peanut hairy roots

Fig. 7. HPLC chromatogram of secreted stilbenoids upon elicitation of 9-day peanut cv.Hull hairy root culture line 3. Stilbenoids secreted into the tissue culture media weredetected by absorbance at 320 nm (A) or fluorescence [excitation 335 nm; emission374 nm] (B). Dotted lines represent HPLC profile of non-elicited cultures at the cor-responding culture age.

(at ages 6,12 and 15) (Table 1). These results are quiet similar to thatpreviously reported by Sobolev et al. [39], inwhich trans-arachidin-3 and trans-resveratrol were the main stilbenoids produced byinsect-damaged peanut plants. Trans-arachidin-1 was also detectedin our samples where highest values were observed in MSVmedium of day 9 elicited root cultures (Table 1). Comparison of the4 different ages in which elicitation was performed and the twoculture media studied, supports that day 9 (exponential growth)and MSV medium were the best age and medium to producehighest amount of stilbenoids in the peanut Hull3 line (Table 1).

The root tissue was analyzed to determine if significantquantities of stilbenoids (trans-resveratrol, trans-arachidin-3 andtrans-arachidin-1) were retained in the tissue. In control cultures,stilbenoids were not detectable in the root tissue collected at allages in either medium condition (Table 2). While trace amounts ofstilbenoids were detected in elicited roots, the amounts relative tostilbenoids secreted into the media fraction were quite low(Supplementary Fig. S2).

Neither trans-arachidin-1 nor trans-arachidin-3 was detected inmedia fractions of non-elicited roots cultured in either MSV or B5media (Table 1). Conversely, trans-resveratrol was detected incontrol cultures grown in either media, albeit at very low levelssuggesting that medium exchange is sufficient in triggering theproduction of trans-resveratrol. This observation was previouslyreported in peanut cell suspension cultures diluted with freshmedium which resulted in increased stilbene synthase (theresveratrol-forming enzyme) activity [40]. Taken together thesedata suggest that enzymes responsible for the biosynthesis of

Table 2Stilbenoids retained in the peanut root tissue at different ages after incubation by24 hours with fresh media (control) or fresh media with sodium acetate (elicited).

Age Mean� SD, nmol/g dry weight root; n¼ 3

trans-resveratrol trans-arachidin-1 trans-arachidin-3

Control Elicited Control Elicited Control Elicited

MSV6 nda nd nd 5.7� 0.8 nd 3.2� 0.59 nd 4.3� 0.9 nd 2.4� 0.6 nd 2.4� 0.212 nd 2.5� 0.2 nd 0.8� 0.7 nd 1.9� 0.815 nd 2.1� 0.1 nd 0.5� 0.5 nd 0.8� 0.2

B56 nd 2.2� 3.8 nd 3.6� 0.3 nd 2.6� 0.19 nd 1.9� 1.7 nd 1.6� 0.2 nd 1.5� 0.512 nd 1.6� 1.4 nd 0.4� 0.6 nd 1.1� 0.515 nd nd nd 1.3� 0.1 nd 1.8� 0.2

a Not detected.

J. Condori et al. / Plant Physiology and Biochemistry 48 (2010) 310e318318

resveratrol may be slightly induced by salt stress effectively createdby addition of fresh media prior to elicitation. In order to producesignificant amounts of resveratrol and other stilbenoids, a moremarked change created with the addition of sodium acetatewas required. Notably acetate (from sodium acetate) could beincorporated into the stilbenoid pathway through the formation ofmalonyl-CoA findings support a synergistic role for sodium acetatein inducing the production of stilbenoids.

4. Conclusions

The present study highlights the importance in optimizing hairyroot culture growth and culturing conditions to better predict anddefine the secondary metabolite biosynthetic capacity of thissystem. Age of the root culture and type of culture media wereshown to have a direct impact on the biosynthesis and secretion ofstilbenoids. Trans-resveratrol was found to be the main stilbenoidproduced under our conditions of 24 h elicitation lending furthersupport for resveratrol serving as the backbone or parent moleculefor other stilbenoid molecules in peanut such as the prenylatedresveratrol analogues arachidin-1 and arachidin-3. The peanuthairy root system provides a promising platform that preserves theproduction of stilbenoids in plants thus making this root culturea valuable and powerful tool for identifying genes involved instilbenoid biosynthesis (such as resveratrol prenyltransferase) andfor studying the complexity and regulation of the stilbenoidbiosynthetic pathway.

Acknowledgments

This work was supported by the National Science Foundation-EPSCoR (grant # EPS-0701890; Center for Plant-Powered Produc-tion-P3), Arkansas ASSET Initiative and the Arkansas Science &Technology Authority and the Arkansas Biosciences Institute. Partof this work was presented as an invited lecture at the 238thAmerican Chemical Society National Meeting in Washington DC(August 2009).

Appendix. Supplementary information

Supplementary data associated with this article can be found inthe online version at doi:10.1016/j.plaphy.2010.01.008.

References

[1] K. Springob, T. Kutchan, Introduction to the different classes of naturalproducts. in: A.E. Osbourn, V. Lanzotti (Eds.), Plant-derived Natural Products.Springer, New York, 2009, pp. 3e50.

[2] R. Bruni, G. Sacchetti, Factors affecting polyphenol biosynthesis in wild andfield grown St. John's Wort (Hypericum perforatum L. Hypericaceae/Guttiferae).Molecules 14 (2009) 682e725.

[3] T.M. Sirvent, S.B. Krasnoff, D.M. Gibson, Induction of hypericins and hyper-forins in Hypericum perforatum in response to damage by herbivores. J. Chem.Ecol. 29 (2003) 2667e2681.

[4] D.E. Gray, S.G. Pallardy, H.E. Garrett, G.E. Rottinghaus, Acute drought stress andplant age effects on alkamide and phenolic acid content in purple coneflowerroots. Planta Med. 69 (2003) 50e55.

[5] G.T. Jeong, D.H. Park, Enhanced secondary metabolite biosynthesis by elici-tation in transformed plant root system: effect of abiotic elicitors. Appl.Biochem. Biotechnol. 129e132 (2006) 436e446.

[6] M.I. Georgiev, A.I. Pavlov, T. Bley, Hairy root type plant in vitro systems as sourcesof bioactive substances. Appl. Microbiol. Biotechnol. 74 (2007) 1175e1185.

[7] B.N.Mishra, R. Ranjan,Growthofhairy-root cultures invarious bioreactors for theproduction of secondarymetabolites. Biotechnol. Appl. Biochem. 49 (2008) 1e10.

[8] S. Srivastava, A.K. Srivatsava, Hairy root culture for mass-production of high-value secondary metabolites. Crit. Rev. Biotechnol. 27 (2007) 29e43.

[9] J.Y. Wu, M. Shi, Ultrahigh diterpenoid tanshinone production throughrepeated osmotic stress and elicitor stimulation in fed-batch culture of Salviamiltiorrhiza hairy roots. Appl. Microbiol. Biotechnol. 78 (2008) 441e448.

[10] E. Ruiz-May, R.M. Galaz-Avalos, V.M. Loyola-Vargas, Differential secretion andaccumulation of terpene indole alkaloids in hairy roots of Catharanthus roseustreated with methyl jasmonate. Mol. Biotechnol. 41 (2009) 278e285.

[11] H.C. Zhang, J.M. Liu, H.Y. Lu, S.L. Gao, Enhanced flavonoid production in hairyroot cultures of Glycyrrhiza uralensis Fisch by combining the over-expressionof chalcone isomerase gene with the elicitation treatment. Plant Cell Rep. 28(2009) 1205e1213.

[12] F. Medina-Bolivar, J. Condori, A.M. Rimando, J. Hubstenberger, K. Shelton, S.F. O'Keefe, S. Bennett, M.C. Dolan, Production and secretion of resveratrol inhairy root cultures of peanut. Phytochemistry 68 (2007) 1992e2003.

[13] K. Tang, J.C. Zhan, H.R. Yang, W.D. Huang, Changes of resveratrol and antiox-idant enzymes during UV-induced plant defense response in peanut seedlings.J. Plant Physiol. 167 (2010) 95e102.

[14] J.C. Chang, Y.H. Lai, B. Djoko, P.L. Wu, C.D. Liu, Y.W. Liu, R.Y. Chiou, Biosynthesisenhancement and antioxidant and anti-inflammatory activities of peanut(Arachis hypogaea L.) arachidin-1, arachidin-3, and isopentadienylresveratrol. J.Agric. Food Chem. 54 (2006) 10281e10287.

[15] V.S. Sobolev, T.L. Potter, B.W. Horn, Prenylated stilbenes from peanut rootmucilage. Phytochem. Anal. 17 (2006) 312e322.

[16] V.S. Sobolev, B.W. Horn, T.L. Potter, S.T. Deyrup, J.B. Gloer, Production of stil-benoids and phenolic acids by the peanut plant at early stages of growth. J.Agric. Food Chem. 54 (2006) 3505e3511.

[17] K. Yazaki, K. Sasaki, Y. Tsurumaru, Prenylation of aromatic compounds, a keydiversification of plant secondary metabolites. Phytochemistry 70 (2009)1739e1745.

[18] B. Botta, G. Delle Monache, P. Menendez, A. Boffi, Novel prenyltransferaseenzymes as a tool for prenylation of flavonoids. Trends Pharmacol. Sci. 26(2005) 606e608.

[19] L. Heide, Prenyl transfer to aromatic substrates: genetics and enzymology.Curr. Opi. Chem. Biol. 13 (2009) 171e179.

[20] D.W. Gorbet, Registration of ‘Hull’ peanut. J. Plant Regis. 1 (2007) 125e126.[21] O.L. Gamborg, R.A. Miller, K. Ojima, Nutrient requirements of suspension

cultures of soybean root cell. Exp. Cell Res. 50 (1968) 151e158.[22] T. Murashige, F. Skoog, A revised medium for rapid growth and bioassays with

tobacco tissue cultures. Physiol. Plant. 15 (1962) 473e497.[23] G. Sivakumar, K.W. Yu, K.Y. Paek, Production of biomass and ginsenoides from

adventitious roots of Panax ginseng in bioreactor cultures. Eng. Life Sci. 5(2005) 333e342.

[24] Y.J. Kim, P.J. Weathers, B.E. Wyslouzil, Growth dynamics of Artemisia annuahairy root in three culture systems. Biotechnol. Bioeng. 83 (2003) 428e443.

[25] M.J. Towler, B.E. Wyslouzil, P.J. Weathers, Using an aerosol deposition model toincrease hairy root growth in a mist reactor. Biotechnol. Bioeng. 96 (2007)881e891.

[26] M. Dubois, K.A. Gilles, J.K. Hamilton, P.A. Rebers, F. Smith, Colorimetric methodfor determination of sugars and related substances. Anal. Chem. 28 (1956)350e356.

[27] V.S. Sobolev, S.A. Neff, J.B. Gloer, New stilbenoids from peanut (Arachis hypo-gaea) seeds challenged by an Aspegillus caelatus strain. J. Agric. Food Chem. 57(2009) 62e68.

[28] V.S. Sobolev, R.J. Cole, J.W. Dorner, B. Yagen, Isolation, purification, and liquidchromatographic determination of stilbene phytoalexins in peanuts. J. AOACInt. 78 (1995) 1177e1182.

[29] J. Condori, G. Medrano, G. Sivakumar, V. Nair, C. Cramer, F. Medina-Bolivar,Functional characterization of a stilbene synthase gene using a transientexpression system in planta. Plant Cell Rep. 28 (2009) 589e599.

[30] L. Yang, P.P. Banada, Y.S. Liu, A.K. Bhunia, R. Bashir, Conductivity and pH dualdetection of growth profile of healthy and stressed Listeria monocytogenes.Biotechnol. Bioeng. 92 (2005) 685e694.

[31] I.M.T. Piza, G.P.P. Lima, A. Henrique, O.G. Brasil, Effect of pH on proline contentsand peroxidase activity in metabolism of sugarcane cultivated in vitro. Agron.Trop. 53 (2003) 145e156.

[32] M. Martinez-Estevez, J.A. Muñoz-Sanchez, V.M. Loyola-Vargas, S.M.T. Hernandez-Sotomayor, Modification of the culture medium to producealuminum toxicity in cell suspension of coffee (Coffea arabica L.). Plant CellRep. 20 (2001) 469e474.

[33] H. Kronzucker, M.Y. Siddiqi, A. Class, G. Kirk, Nitrate-ammonium synergism inrice. A subcellular flux analysis. Plant Physiol. 119 (1999) 1041e1045.

[34] O.L. Gamborg, J.P. Shyluk, The culture of plant cells with ammonium salts asthe sole nitrogen source. Plant Physiol. 45 (1970) 598e600.

[35] R.M. Skirvin, M.C. Chu, M.L. Mann, H. Young, J. Sullivan, T. Fermanian, Stabilityof tissue culture medium pH as a function of autoclaving, time, and culturedplant material. Plant Cell Rep. 5 (1986) 292e294.

[36] C.A. Ticconi, S. Abel, Short on phosphate: plant surveillance and counter-measures. Trends Plant Sci. 9 (2004) 548e555.

[37] I.-S. Park, D.-I. Kim, Significance of fresh weight to dry cell weight ratio inplant cell suspension cultures. Biotechnol. Tech. 7 (1993) 627e630.

[38] V.S. Sobolev, Localized production of phytoalexins by peanut (Arachis hypo-gaea) kernels in response to invasion by Aspergillus species. J. Agric. FoodChem. 56 (2008) 1949e1954.

[39] V.S. Sobolev, B.Z. Guo, C.C. Holbrook, R.E. Lynch, Interrelationship of phyto-alexin production and disease resistance in selected peanut genotypes.J. Agric. Food Chem. 55 (2007) 2195e2200.

[40] A. Schöppner, H. Kindl, Purification and properties of a stilbene synthase frominduced cell suspension cultures of peanut. J. Biol. Chem. 259 (1984)6806e6811.