Plant Cell Rep (2009) 28:589–599 DOI 10.1007/s00299-008-0664-0
GENETIC TRANSFORMATION AND HYBRIDIZATION
Functional characterization of a stilbene synthase gene using a transient expression system in planta Jose Condori Æ Giuliana Medrano Æ Ganapathy Sivakumar Æ Vipin Nair Æ Carole Cramer Æ Fabricio Medina-Bolivar
Received: 2 December 2008 / Revised: 5 December 2008 / Accepted: 9 December 2008 / Published online: 31 December 2008 Ó Springer-Verlag 2008
Abstract The expression and functionality of a resveratrol synthase (RS) gene from peanut (Arachis hypogaea) was studied using an Agrobacterium tumefaciens-mediated transient expression system in Nicotiana benthamiana leaves. Functional analysis of RS was demonstrated by tracking its expression during 96 h. To measure the transcripts levels of RS transgene, real-time qRT-PCR was used and revealed that the highest level of transcripts was at 48 h post-transfection. Western blot analyses showed that RS protein was accumulated to the highest levels at 72 h post-transfection. Finally, HPLC and mass spectrometry analyses revealed the production of trans-piceid (resveratrol glucoside) as the major stilbenoid compound confirming the functional activity of the RS enzyme in planta. No activity of RS transgene was detected in negative controls. This strategy showed advantages over conventional systems because it does not require establishment of cell cultures, feeding with appropriate substrates or generation of stable transgenic plants. This transient system proved to be a rapid and direct approach to perform functional analysis of stilbene synthases, such as
Communicated by G. Phillips.
Electronic supplementary material The online version of this article (doi:10.1007/s00299-008-0664-0) contains supplementary material, which is available to authorized users. J. Condori G. Medrano G. Sivakumar V. Nair C. Cramer F. Medina-Bolivar (&) Arkansas Biosciences Institute, Arkansas State University, State University, P.O. Box 639, AR 72467, USA e-mail:
[email protected] F. Medina-Bolivar Department of Biological Sciences, Arkansas State University, State University, AR 72467, USA
resveratrol synthase. Furthermore, this approach can be useful to study the metabolic effects of over-expressing or silencing specific genes within a short period of time. Keywords Resveratrol synthase Transient expression Nicotiana benthamiana Resveratrol Piceid Abbreviations RS Resveratrol synthase CHS Chalcone synthase STS Stilbene synthase
Introduction Stilbenoids are specialized plant metabolites derived from the phenylpropanoid/acetate pathway, which have been described in taxonomically unrelated species including agronomically important crops such as grape, peanut and blueberry. In addition to their in planta function as phytoalexins, they have shown to possess bioactive properties in mammalian cells (Roupe et al. 2006) and health promoting benefits in mammals (Baur et al. 2006). The best-characterized stilbenoid is trans-resveratrol (trans-3,40 ,5-trihyhydroxystilbene). The final step in transresveratrol biosynthesis involves the condensation of one molecule of p-coumaroyl-CoA and three molecules of malonyl-CoA, a reaction catalyzed by resveratrol synthase (RS) (Lanz et al. 1990). Chalcone synthase (CHS), a ubiquitous enzyme, also uses the same substrates and the same type of catalytic reaction as RS, but forms a different product (Austin and Noel 2003). Resveratrol synthase belongs to the group of enzymes known as stilbene synthases (STS), which in most
123
590
stilbenoid-producing species is represented by a multigene family. For instance, 15–20 copies of STS per haploid genome have been proposed to be present in grape (Sparvoli et al. 1994). So far, studies of STS have been done expressing multiple enzymes from the phenylpropanoid pathway in heterologous transgenic systems (Zhang et al. 2006) or by using stable transformation in transgenic plants (Hain et al. 1990; Hipskind and Pavia 2000). The main drawback of these approaches is that they require establishing of cell cultures, feeding with appropriate substrates or generation of transgenic plants which is time consuming. Therefore, the need for effective and quick methodologies for functional characterization of these STS will greatly facilitate the elucidation of how these genes function in planta. To address these issues, we developed an Agrobacterium-mediated transient expression system in intact leaves of Nicotiana benthamiana to characterize a RS from peanut (Arachis hypogaea cv. Andru II). Transient expression has the advantage of being rapid, without the need to regenerate transformed cells, and therefore useful for evaluating multiple expression constructs (Hellens et al. 2005). N. benthamiana was chosen because there is no report of RS as an endogenous gene in this species. Nevertheless, as expected a CHS gene has been reported in this organism (GenBank accession number EF421432). Therefore, the substrates for the synthesis of trans-resveratrol are present and available to CHS. Earlier studies revealed that synthesis of trans-resveratrol is possible in N. tabacum by stable expression of RS (Hain et al. 1990). One additional advantage of N. benthamiana as a platform is its high efficiency of transference of T-DNA (*100%) in mesophyll cells by Agrobacterium tumefaciens infiltration as reported by Marillonnet et al. (2004). Herein we demonstrated that the N. benthamiana transient expression system can be expanded for metabolic engineering studies before committing to the generation of transgenic plants by stable transformation. Indeed, we transiently expressed a cloned STS gene and characterized the profile of stilbenoids produced, therefore confirming the functionality of this gene in planta.
Materials and methods Gene cloning A search was done in the National Center for Biotechnology Information (NCBI, GenBank) database for sequences (DNA and cDNA) encoding RS and CHS in Arachis hypogaea. Thirteen sequences related to RS and two sequences to CHS were found as of February 2007 (Supplementary table S1). To design gene-specific primers, these sequences were aligned using AlignX software (Vector NTIÒ version
123
Plant Cell Rep (2009) 28:589–599
10, Invitrogen). In order to clone the entire genes, primers included the ATG start site and the common stop codon (TAA). Conserved nucleotide regions at the 50 - and 30 -end of RS and CHS sequences were selected to design a set of primers for cloning of RS and CHS. Primers were flanked with restriction enzyme sites (BamHI or SacI) used in further cloning steps. These sites are not present in any of the coding and noncoding regions of known peanut RS or CHS (Supplementary table S1). Primers for RS were 50 -CCGGATCCAT GGTGTCTGTGAGTGGAAT-TCGCAAGGT-30 (BamHI site underlined) and 50 -CCGAGCTCTTATATGGCCA CA-CTGCGGAG-30 (SacI site underlined). Primers for CHS were 50 -CCGGATCCATGGTGAATGTATCTGA G-30 (BamHI site underlined) and 50 -CCGAGCTCTT ATATAGCCACACTATGCAA-30 (SacI site underlined). Specificity at the 30 -end of the primers was considered in order to maximize discrimination between the RS and CHS targets. Genomic DNA from previously established peanut cv. Andru II hairy root Line 2 (Medina-Bolivar et al. 2007) was obtained using DNeasyÒ Plant Mini kit (Qiagen). PCR was carried out employing the two set of primers described above and two DNA polymerases: Taq polymerase using puReTaq Ready-To-Go PCR beads (Amersham Biosciences) and Pfu polymerase (Strategene). The latter has 30 – 50 exonuclease activity (proofreading). PCR products were cloned into pBC SK(-) vector and transformed into TOP10 chemically competent E. coli (Invitrogen) according to manufacturer’s protocol. Plasmids that had the PCR products were sent to DNA sequencing facility (University of Chicago Cancer Research Center) for sequencing. Sequence analysis Sequence analysis of cloned genes was done in silico. BLASTN (Basic Local Alignment Search Tool-Nucleotide) software (NCBI) was used to compare cloned sequences. Prediction of exon/intron regions was done employing GENSCAN software (http://genes.mit.edu/GENSCAN. html). Previously, this software was tested with an available genomic CHS sequence (GenBank accession number AY192572) and the output (exon/intron) was identical to that reported in GenBank (data not shown). A narrow comparison with RS and CHS sequences from peanut (Supplementary table S1) was done by alignment of these sequences with our cloned sequences using AlignX (Vector NTIÒ version 10). A sequence for RS (GenBank accession number EF620775), available since June 2007, was also taken in consideration for the analysis. Only coding regions of RS and CHS genes were considered for alignment analysis. Two sequences: GenBank accession numbers L00953 and L00954 were not considered, because of their short sequences.
Plant Cell Rep (2009) 28:589–599
Preparation of Agrobacterium inoculum JC1 cloned from peanut was subcloned as a XhoI/SacI fragment into a modified R8-2 binary vector (MedinaBolivar and Cramer 2004) containing double-enhanced 35S CaMV (cauliflower mosaic virus) constitutive promoter and 30 UTR of the nopaline synthase gene. Then the vector was mobilized into A. tumefaciens strain LBA4404 using freeze/thaw method (Holsters et al. 1978). The engineered A. tumefaciens was grown in 8 ml of YEP medium [10 g/l Bacto-peptone (Difco), 10 g/l yeast extract (Difco), 5 g/l NaCl (Sigma-Aldrich), pH 7.0], containing 0.1 g/l of kanamycin (Sigma-Aldrich) and 0.03 g/l of streptomycin (Sigma-Aldrich) for antibiotic selection. Bacteria were grown at 28°C on an orbital shaker at 225 rpm for 2 days. The 8 ml of bacterial suspension was used as inoculum of 50 ml fresh YEP medium containing the antibiotics. The bacteria were allowed to grow for one additional day under the same conditions. Bacteria were pelleted by centrifugation at 2,504g for 10 min, resuspended in 400 ml of induction medium (IM) as described by Medrano et al. (2009). Cultures were incubated for 5 h at 28°C on an orbital shaker at 225 rpm. Concentration of the culture was measured by spectrophotometry. Cultures with an OD600 ranging from 0.35 to 0.45 were used for infiltration as detailed below. Infiltration of N. benthamiana with Agrobacterium Agrobacterium-mediated vacuum infiltration was performed according to Medrano et al. (2009) using 4-weekold N. benthamiana (accession PI 555478). For IM samples (negative control), IM was employed without bacteria. Leaf tissue of infiltrated plants was harvested 48, 72 and 96 h post-infiltration. Leaves from the ‘‘middle tier’’ were used for comparative analyses as they provided sufficient biomass to carry out the different analysis from the same tissue. Whole leaves were weighed (fresh weight), frozen in liquid nitrogen and stored at -80°C for further analysis. Real-time quantitative reverse transcriptase-polymerase chain reaction (qRT-PCR) Total RNA was extracted from leaves of Agrobacteriuminfiltrated N. benthamiana with TrizolÒ reagent (Invitrogen) according to manufacturer’s protocol. In order to maintain the integrity of RNA by degrading any potential remaining genomic DNA, RNA preparations were treated with RQ1 RNAse-Free DNase (Promega). Total RNA was resuspended in DEPC-treated water and quantified by spectrophotometer (ND-1000, NanodropÒ). cDNA was synthesized using iScriptTM Select cDNA Synthesis kit (Bio-Rad) using oligo(dT) primer.
591
Primers for real-time qPCR were designed based on sequence obtained from the cloned JC1: 50 -TATGTATTT AACAGAAGAAATAC-30 and 50 -AGTTGCAGCCTCTT TTCCAA-CT-30 . An amplicon size of *100 bp was targeted in order to obtain efficient amplification (Bustin 2000). PCR reactions were run in triplicate using: 0.8 lM of each primer, 19 iQTM SYBR Green Supermix (BioRad) and 100 ng cDNA (12.5 ll final volume of reaction). These modifications where shown to give the same results as the manufacturer’s protocol (Bio-Rad) (data not shown). Real-time qPCR was carried out using an iCycler iQ (BioRad). The following qPCR cycling program was used for all sets of primers: denaturation at 95°C for 3 min, 30 cycles of amplification (95°C for 10 s, 55°C for 45 s) and a melting curve program (from 55°C with an increase set point temperature after cycle two by 0.5°C). Actb (b-actin) gene was employed as an endogenous housekeeping gene (Burger et al. 2003). Efficiencies of JC1 and Actb genes were determined to be two according to Pfaffl et al. (2002) with R2 values of at least 0.99. cDNA from transfected plant with JC1 was used for this calibration. Calculation of relative gene expression of JC1 in relation to housekeeping gene was performed using REST-MSC (Relative Expression Software Tool-Multiple Condition Solver) software version 2 (Pfaffl et al. 2002). Cycle threshold (Ct) values were employed for analysis. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) Total soluble protein (TSP) was extracted from leaves of Agrobacterium-infiltrated N. benthamiana with SDS (sodium dodecyl sulfate) extraction buffer (150 mM Tris– HCl pH 6.8, 30% glycerol, 6% SDS, 5 mM EDTA) in a ratio of 1:2 (w:v). Concentration of TSP was determined using Bradford protein assay (Bradford 1976) with BSA (bovine serum albumin) (Pierce) as the standard. Samples were run in duplicate using Advanced Protein Assay Reagent 59 (Cytoskeleton) as reagent. Absorbance at 595 nm was measured with PowerWaveTM Microplate Spectophotometer (Bio-Tek) using the KCjunior software (Bio-Tek). TSP from N. benthamiana leaves (NB and pBIB-Kan samples) were also extracted with PBS (phosphate buffered saline) (137 mM NaCl, 2.7 mM KCl, 10.1 mM Na2HPO4, 1.76 mM KH2PO4, pH 7.4). Samples extracted with PBS and SDS buffers were analyzed with Bradford protein assay as below. The ratio of the TSP value using PBS versus SDS buffers was 3–1 (data not shown). TSP values in this work refer to determinations using PBS. Twenty-one micrograms of TSP was mixed with 19 NuPAGEÒ LDS Sample Buffer and 19 NuPAGEÒ Reducing Agent containing 50 mM dithiothreitol (DTT) (Invitrogen), and then heated at 70°C for 10 min. The
123
592
denaturated protein extracts were loaded on a NuPAGEÒ NovexÒ Bis–Tris 10% mini gel (Invitrogen) and the proteins were separated using 19 MOPS SDS Running Buffer (Invitrogen) with 0.25% NuPAGEÒ Antioxidant (Invitrogen) in a XCell SureLockTM (Invitrogen) horizontal gel electrophoresis unit. Protein extracted from peanut hairy root line 2 (elicited with sodium acetate for stilbenoid production as described by Medina-Bolivar et al. 2007) was used as positive control. Protein extraction from the root tissue was done as described above. SimplyBlueTM SafeStain (Invitrogen) was employed to stain proteins in SDS-PAGE gels following manufacturer’s basic protocol. Gels were visualized with VersaDocTM Imaging System Model 4000 (Bio-Rad) using Quantity One software Version 4.5.1 (Bio-Rad). Western blotting After electrophoresis, the proteins were electroblotted to 0.2 lm nitrocellulose membrane (Bio-Rad) in transfer buffer [10% methanol (Fisher), 19 NuPAGEÒ Transfer Buffer (Invitrogen), 0.1% NuPAGEÒ Antioxidant] using an XCellTM Blot Module (Invitrogen). Protein blot was blocked with PBS (phosphate buffered saline) (137 mM NaCl, 2.7 mM KCl, 10.1 mM Na2HPO4, 1.76 mM KH2PO4, pH 7.4) supplemented with 3% BSA Fraction V (Roche) and 0.1% TweenÒ 20 (Sigma) for 1 h at room temperature (RT). Immunoblot detection of JC1 was done with rabbit-anti RS3 antibody (Lanz et al. 1991) (kindly provided by Dr. Scott Baerson), as the primary antibody diluted at 1:300 (best dilution after standardization, data not shown) in PBS supplemented with 3% BSA Fraction V and 0.1% TweenÒ 20 for 1 h at RT. Protein blot was washed three times (10 min per wash) with PBS plus 0.1% TweenÒ 20. Goat–anti rabbit IgG (H ? L)-alkaline phosphatase conjugated (Bio-Rad) was used as secondary antibody diluted at 1:5,000 in PBS supplemented with 3% BSA Fraction V and 0.1% TweenÒ 20 for 45 min at RT. Protein blot was washed twice (10 min per wash) with PBS plus 0.1% TweenÒ 20 followed by a final wash with PBS buffer for 10 min at RT. Detection was carried out with CDP-Star (Roche) and Nitroblock Enhancer II (Tropix) in accordance with manufacturers’ procedures. Densitometric analysis Semi-quantitative densitometric analysis of Western blot bands was done with VersaDocTM Imaging System Model 4000 (Bio-Rad) using Quantity One software Version 4.5.1(Bio-rad). TSP from JC1-B were run in SDS-PAGE under reducing conditions, and processed as above. JC1-B samples were chosen because this plant showed the highest production of the target metabolite at 96 h.
123
Plant Cell Rep (2009) 28:589–599
Extraction and analyses of stilbenoids Two solvents were used for extraction: ethyl acetate (Fisher) and methanol (Fisher). Frozen leaves from Agrobacterium-infiltrated N. benthamiana were ground to powder in liquid nitrogen in a mortar and transferred to a tube with solvent [ratio of 1 g fresh weight (FW) tissue to 10 ml solvent]. Samples were mixed with vortex for 2 min and centrifuged for 30 min at 2,504g at 4°C. Supernatants were transferred into glass tubes and dried to completeness under nitrogen stream using RapidVap N2 system (Labconco) at 40°C and 40% rotor speed. Dried extracts were resuspended in methanol and analyzed as below. Root tissues were extracted with ethyl acetate, and the extracts were dried and resuspended in methanol as above. Samples were filtered through 0.2 lm nylon filter and analyzed by reverse phase high performance liquid chromatography (HPLC). The HPLC system (Dionex) consisted of a P680 pump, an Ultimate3000TM column compartment and an ASI100TM autosampler. Two detectors, a PDA100TM photo diode array detector and a RF2000TM fluorescence detector were serially connected to this system. Chromatography was done in a SunFireTM C18 (25 cm 9 4.6 mm, 5 lm) column (Waters) at 30°C and a flow rate of 0.5 ml/min. The column was initially equilibrated in 99:1 (v/v) acetonitrile (Fisher) and water with 0.05% (v/v) formic acid (Fluka) as mobile phase. The following gradient was used: 0–8 min linear gradient from 10 to 18% (v/v) acetonitrile, 2 min isocratic with 18% (v/v) acetonitrile, 5 min linear gradient from 18 to 25% (v/v) acetonitrile, 3 min linear gradient from 25 to 35% (v/v) acetonitrile, 10 min isocratic with 35% (v/v) acetonitrile, 20 min isocratic 35% (v/v) acetronitrile, 59 min linear gradient from 35 to 60% (v/v) and 6 min linear gradient from 60 to 10% (v/v) acetonitrile. Detection of stilbenoids was done with coupled photodiode array (PDA) and fluorescence (excitation 330; emission 374 nm) detectors. trans-Resveratrol (Sigma) and transpolydatin (also known as trans-piceid; Alexis) were used as stilbenoid standards. To confirm identity of trans-piceid, the extracts were analyzed by mass spectrometry (MS) using a VarianÒ 1200L Quadrupole LC/MS/MS system (Walnut Creek). The system consisted of a ProStar 420 autosampler, ProStar 210 pumps, 212-LC solvent delivery system and a 320 triple quadrupole mass spectrometer equipped with an electrospray ionization (ESI) source. The HPLC/MS system was operated and the data were collected using Prostar/ DynamaxTM software (version 2.4). The HPLC separation conditions were as detailed above. LC/ESI–MS was performed in negative mode from m/z 100–800. The heated capillary was set to 200°C and the electrospray voltage to 2.8 kV. The sheath and auxiliary gas were adjusted at 65 and 20 arbitrary units, respectively. The API source was
Plant Cell Rep (2009) 28:589–599
maintained at 65°C. The mass spectrometer was used in scan mode and this was followed by collision induced dissociation (CID) of the highest abundant ion selected from the full scan. Prior to sample analysis, the MS response was optimized using trans-resveratrol and transpiceid as standards. Enzymatic hydrolysis of trans-resveratrol glucosides Extracts from Agrobacterium-infiltrated N. benthamiana at 72 h post-infiltration were treated with b-D-glucosidase (Fluka). Dried extracts (methanol and ethyl acetate) resuspended in methanol were brought to a final volume of 1 ml with phosphate buffer 20 mM, pH 6.0 and then b-Dglucosidase (1 mg/ml) was added to diluted extracts. Hydrolysis was carried out by incubation at 50°C for 9 h (La Torre et al. 2004). Reaction was stopped by placing the mixtures in an ice bath. Subsequent extraction was done with ethyl acetate. Supernatants were transferred into glass vials and dried under nitrogen flow. Dried extracts were resuspended in 50 ll of methanol, filtered through 0.2 lm nylon filter and analyzed by reverse phase HPLC as describe above.
593
for Pfu one-step mediated-amplification and no PCR product was detected, further suggesting that a secondary structure in this sequence could interfere with Pfu-mediated amplification. In contrast, CHS (JC3) was cloned using Pfu polymerase without any difficulty. Sequence analyses of JC1 and JC3 genes revealed that they have a single intron in their sequences, which is a feature of RS and CHS genes. Comparison at the nucleotide and amino acid levels of JC1 and JC3 versus available peanut RS and CHS sequences (Fig. 1) clustered JC1 together with RS; a distant cluster of CHS was formed by JC3. JC1 showed the highest identity (98%) with the GenBank accession number AB027606 at the nucleotide level. At the amino acid level, they differ at a single amino acid close to the 30 -end (V387 M). The sequence of JC3 is 99.9% identical to the GenBank accession numbers AY735111 (DNA) and AY192572 (cDNA), at the nucleotide level. When considering the amino acid sequence, JC3 is identical to AY192572 and different to AY735111
Statistical analysis Statistical analysis was done in order to compare relative mRNA transcript levels between treatments. A condition was considered to be significantly different when P calculated by T-test was \0.05. T-test was performed using XLSTAT software version 2008.3.01 (Addinsoft).
Results and discussion Cloning of RS and CHS We cloned a putative genomic RS gene designated JC1 (GenBank accession number EU384706) and a putative genomic CHS gene designated JC3 (GenBank accession number EU418492) from A. hypogaea cv. Andru II. Cloning of JC1 was done using Taq polymerase. To confirm the fidelity of this sequence, DNA sequences derived through different events of PCR and different templates (root cDNA, genomic DNA, genomic DNA digested with BamHI/PstI) were employed (data not shown). Interestingly, Pfu polymerase was not able to amplify RS in onestep cloning, even when different conditions and templates were used (data not shown). One explanation could be that the RS sequence forms a secondary structure which does not allow Pfu to utilize this substrate for DNA synthesis. This phenomenon was not observed with Taq polymerase. A cloning vector harboring JC1 also was used as a template
Fig. 1 Neighbor-joining consensus tree (Tamura-Nei model) estimated after a search with 2,000 replicates, using DNA (a) and amino acid (b) sequence alignments of A. hypogaea resveratrol synthase (RS) and chalcone synthase (CHS) sequences from GenBank (Supplementary table S1) with sequences of JC1 and JC3 cloned in this study. Alignment was generated using AlignX (Vector NTI Version 10, Invitrogen). Consensus tree was done with Geneious Pro software (Biomatters)
123
594
only in one amino acid (I43T). To date, only two CHS sequences have been reported for peanut in GenBank (Supplementary table S1). Nonetheless, this gene belongs to a gene family in many species (Austin and Noel 2003). The identity between JC1 and JC3 was 78.4% (amino acid level), this difference had been indicated to have an effect on substrate specificity and regiospecificity for the cyclization and condensation reaction of RS and CHS respectively (Schro¨der and Schro¨der 1990). Amino acid homology between RS and CHS suggests that RS has evolved in a limited number of phylogenetically distinct plants via gene duplication and subsequent mechanistic divergence from CHS (Austin et al. 2004). Transient expression of RS in N. benthamiana To determine if JC1 is a functional RS, we decided to express it transiently in N. benthamiana. JC1 was cloned downstream of the constitutive dual enhanced 35S
Fig. 2 Real-time qRT-PCR of JC1 in post-infiltrated N. benthamiana leaves. Each sample was analyzed in triplicate. Error bars represent standard deviation among the triplicate readings. JC1-A and JC1-B independent events of infiltration with the construct harboring JC1 gene. Negative controls: pBIB-Kan plant infiltrated with empty vector, IM plant infiltrated with induction media without bacteria, NB intact noninfiltrated plant. Leaves were harvested after 48, 72 and 96 h post-infiltration. Relative mRNA values are based on JC1A 96 h (reference sample). Actb was used as housekeeping gene to normalize mRNA expression. Amplification of JC1 was not detected in negative controls or root tissue. * Significantly different (T-test, P \ 0.05) to JC1-A 96 h
123
Plant Cell Rep (2009) 28:589–599
promoter, introduced into a binary vector (Supplementary material S2), and infiltrated in N. benthamiana leaves. The experiment was performed at three different times and the results were reproducible. Here we show the results of one of them, which is representative. The JC1 gene contains one intron and production of mRNA product in the transient system requires RNA splicing, which takes place in the plant cell, but not in A. tumefaciens. In this way, we can ensure that function analysis of JC1 is due to expression of the gene in the plant and not to expression by the bacteria. Different negative controls were considered to guarantee proper characterization of the JC1 gene. N. benthamiana leaves were infiltrated with the empty vector (pBIB-Kan). In addition, infiltration was done with IM without bacteria to test if stress conditions during infiltration produce any stilbenoid compounds and finally, intact non-infiltrated N. benthamiana (NB) leaves were analyzed to see if there was any background of stilbenoids in the plant. Leaves were harvested at 48, 72 and 96 h post-infiltration to analyze the
Plant Cell Rep (2009) 28:589–599
expression of JC1 at the RNA and protein levels and stilbenoid production over time. Expression levels of JC1 transcripts were assessed by real-time quantitative RT-PCR (qRT-PCR). The highest RNA levels of JC1 were observed at 48 h post-infiltration, whereas the lowest levels were found at 96 h. Intermediate values were found at 72 h (Fig. 2). The changes in RNA levels of the transiently expressed JC1 are possibly due to degradation of the transcripts of the foreign DNA. No JC1 products were observed in samples from negative controls. To study if the JC1 transcripts correlated with the production of recombinant JC1 protein, we analyzed total soluble protein (TSP) extracts at 48, 72 and 96 h postinfiltration. TSP was visualized with Coomassie blue stain and an expected band of *42 kDa, size reported for RS protein (Hain et al. 1990), was detected in N. benthamiana transformed with JC1 construct and peanut RS positive control (Fig. 3). The latter band was absent in the negative controls. To determine if the band *42 kDa corresponded to RS protein, Western blot analysis with polyclonal antibody against peanut RS raised in rabbit (Lanz et al. 1991) was used. Detection was observed in N. benthamiana transiently expressing JC1 construct and positive control, but not in the negative controls (Fig. 4a). A band of *90 kDa was also observed in samples, where JC1 protein was detected. This band may correspond to the homodimer form (RS–RS) which has been shown to be the functional form of RS in planta. Time course analyses showed that the highest level of JC1 protein was at 72 h post-infiltration (2.5-fold higher than 48 h) (Fig. 4b).
Fig. 3 Coomassie blue stain of total protein extracted from infiltrated and control plants. Samples (21 lg TSP/lane) were run in SDS-PAGE under reducing conditions. JC1-A and JC1-B independent events of infiltration with the construct harboring JC1 gene. Negative controls: pBIB-Kan plant infiltrated with empty vector, IM plant infiltrated with induction media without bacteria, NB intact non-infiltrated plant.
595
Functional characterization of transiently expressed RS in N. benthamiana Finally to assess whether the JC1 protein was functional in planta, we analyzed the production of stilbenoids after Agrobacterium infiltration. Analyses of the samples were done by HPLC. Three independent experiments were performed showing production of trans-resveratrol glucoside (i.e. trans-piceid) in plants transfected with JC1. Here a representative experiment with two biological samples for JC1 (JC1-A, JC1-B) are shown. These analyses showed the production of the trans-resveratrol glucoside, trans-piceid, only in plants transiently expressing the JC1 gene (Fig. 5). This compound was not detected in the negative controls. Low levels of transpiceid were detected at 48 h post-infiltration, and it increased during the time course. At 96 h, the amount of trans-piceid recovered under methanol extraction was 6.7 lg/g FW (Fig. 5a). Our results are consistent with previous observations using stable transformation of RS gene in plants (Hipskind and Pavia 2000). Since transresveratrol is not a native compound of N. benthamiana, it may be toxic for the plant and might be stored in the vacuole upon glycosylation to form trans-piceid. To further confirm for the presence of trans-resveratrol glucosides in plants transiently expressing the JC1 gene, we treated the extracts with b-glucosidase, which cleaves monosaccharides from trans-resveratrol glucosides (La Torre et al. 2004). In this analysis, trans-resveratrol was detected as the stilbenoid aglycone (Fig. 6).
Leaves were harvested after 48, 72 and 96 h post-infiltration. Roots were harvested 96 h post-infiltration. HR-Line 2 peanut hairy root Line 2 elicited with sodium acetate (positive control for JC1). The dots indicate the position of the stained proteins of 42 kDa. Marker: BenchmarkTM Protein Ladder (Invitrogen)
123
596
Plant Cell Rep (2009) 28:589–599
Fig. 4 Western blot of total protein extracted from infiltrated and control plants. a Samples (21 lg TSP/lane) were run in SDS-PAGE under reducing conditions. Antibody raised against peanut RS was used. Lane designation as in Fig. 3. RS *42 kDa, RS–RS homodimer *90 kDa, HR-Line 2 peanut hairy root Line 2 elicited with sodium
acetate (positive control for JC1). Marker: SeeBlue Plus2 Pre-Stained Standard (Invitrogen). b JC1 levels from a single infiltration (JC1-B) were analyzed by Western blot of proteins from leaves harvested at times indicated. Relative amount of RS were obtained by densitometric analysis
Fig. 5 HPLC analyses of stilbenoids from N. benthamiana infiltrated with Agrobacterium harboring JC1. a Quantification of trans-piceid (trans-resveratrol glucoside) produced in N. benthamiana tissue (leaves and roots) of Agrobacterium-infiltrated with JC1. Each sample was analyzed in triplicate. Error bars represent standard deviation among the triplicate readings. HPLC profiles of leaf extracts after 72 h post-infiltration with JC1 using ethyl acetate (b) or
methanol (c). A peak assigned to trans-piceid. d UV spectra of the transiently produced trans-piceid (A) and trans-piceid standard. JC1A and JC1-B independent events of infiltration with the construct harboring JC1 gene, pBIB-Kan plant infiltrated with empty vector, IM plant infiltrated with induction media without bacteria, NB intact noninfiltrated plant
The identity of trans-piceid in the leaves transiently expressing JC1 was confirmed by MS. trans-Piceid (Fig. 7a, b) fragmentation yielded a peak of m/z 435, which
is a formic acid adduct ion [M–H ? 46]-. The fragmentation of m/z 435, yielded a product ion of m/z 389 by the loss of 46 Da representing formic acid adduct (–HCOOH).
123
Plant Cell Rep (2009) 28:589–599
597
Fig. 6 Assessment of stilbenoid products with and without enzymatic hydrolysis with b-glucosidase. a HPLC profiles of N. benthamiana leaves 72 h after Agrobacterium infiltration. Extractions were done with ethyl acetate. Peak 1 represents trans-piceid. b HPLC profiles of samples shown in (a) after treatment with b-glucosidase, which hydrolyzes trans-resveratrol glucosides. c UV spectra of trans-
resveratrol (Peak 2) produced upon hydrolysis of trans-piceid in JC1B sample and trans-resveratrol standard. JC1-B represents an independent event of infiltration with the construct harboring JC1 gene, pBIB-Kan plant infiltrated with empty vector, IM plant infiltrated with induction media without bacteria, NB intact noninfiltrated plant
The product ion m/z 389 [M–H]- representing trans-piceid was further fragmented to m/z 227 [M–H-162]- due to loss of glucose (M–H–C6H10O5)-. This fragmentation profile has been suggested in the literature (Pu¨ssa et al. 2006; Yu et al. 2008). In the case of the glucosidase treated extracts, trans-resveratrol (Fig. 7c) the most abundant parent ion [M–H]- m/z 227 (deprotonated trans-resveratrol) was obtained. To determine whether JC1-mediated products synthesized in the leaves are transported to and accumulated in roots or if Agrobacterium harboring JC1 construct infiltrated in the leaves can migrate to the roots, we also analyzed root samples after 96 h post-infiltration. Root samples were subjected to the same molecular tools (realtime qRT-PCR, Western blot and HPLC) to track
resveratrol as described above. None of root samples showed JC1 expression (RNA or protein). Therefore, there was not apparent Agrobacterium mobilization to the roots suggesting that vacuum infiltration of the leaves localizes the expression of JC1 in the leaves only. Consequently, stilbenoids were not detected in the roots indicating that stilbenoids in the transient N. benthamiana system were not translocated into the roots within the time frame analyzed under our experimental conditions. In summary, we have demonstrated the enzymatic functionality of a cloned peanut resveratrol synthase, JC1, and the utility of a N. benthamiana transient expression system for rapid assessment not only of the protein product but the impact of the transgene in mediating stilbenoid production. This system can be used for rapid characterization of a
123
598
Plant Cell Rep (2009) 28:589–599 Acknowledgments This work was supported by the Arkansas Biosciences Institute, National Science Foundation-EPSCoR (grant # EPS-0701890; Center for Plant-Powered Production-P3), Arkansas ASSET Initiative and the Arkansas Science and Technology Authority.
References
Fig. 7 Negative LC/MS spectrum of trans-piceid and trans-resveratrol. N. benthamiana leaves 96 h post-infiltration with JC1 were analyzed with LC/MS for the presence of trans-piceid; extractions were done with ethyl acetate (a) and methanol (b). c trans-resveratrol ion obtained in sample JC1-B after treatment with b-glucosidase
variety of STS genes or even for expression, co-expression or silencing of genes from other metabolic pathways, analyzing not only the functional expression of the gene but also the metabolite profile of the transfected plant before commitment to stable transformation.
123
Austin MB, Noel JP (2003) The chalcone synthase superfamily of type III polyketide synthase. Nat Prod Rep 20:79–110 Austin MB, Bowman ME, Ferrer JL, Schro¨der J, Noel JP (2004) An aldol switch discovered in stilbene synthases mediates cyclization specificity of type III polyketide synthases. Chem Biol 11:1179–1194 Baur JA, Pearson KJ, Price NL, Jamieson HA, Lerin C, Kalra A, Prabhu VV, Allard JS, Lopez-Lluch G, Lewis K, Pistell PJ, Poosala S, Becker KG, Boss O, Gwinn D, Wang M, Ramaswamy S, Fishbein KW, Spencer RG, Lakatta EG, Le Courteur D, Shaw RJ, Navas P, Puigserver P, Ingram DK, De Cabo R, Sinclair DA (2006) Resveratrol improves health and survival of mice on a high-calorie diet. Nature 444:337–342 Bradford MM (1976) A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254 Burger C, Sabine R, Benveniste P, Schaller H (2003) Virus-induced silencing of sterol biosynthetic genes: identification of a Nicotiana tabacum L. obtusifoliol-14a-demethylase (CYP51) by genetic manipulation of the sterol biosynthetic pathway in Nicotiana benthamiana L. J Exp Bot 54:1675–1683 Bustin SA (2000) Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays. J Mol Endocrinol 25:169–193 Hain R, Bieseler B, Kindl H, Schro¨der G, Sto¨cker R (1990) Expression of a stilbene synthase gene in Nicotiana tabacum results in synthesis of the phytoalexin resveratrol. Plant Mol Biol 15:325–335 Hellens RP, Allan AC, Friel EN, Bolitho K, Grafton K, Templeton MD, Karunairetnam S, Gleave AP, Laing WA (2005) Transient expression vectors for functional genomics quantification of promoter activity and RNA silencing in plants. Plant Methods 1:13 Hipskind JD, Pavia NL (2000) Constitutive accumulation of a resveratrol-glucoside in transgenic alfalfa increases resistance to Phomo medicaginis. Mol Plant Microbe Interact 13:551–562 Holsters M, De Waele D, Depicker A, Messens E, Van Montagu M, Schell J (1978) Transfection and transformation of Agrobacterium tumefaciens. Mol Gen Genet 163:181–187 La Torre GL, Lagana G, Bellocco E, Vilasi F, Salvo F, Dugo G (2004) Improvement on enzymatic hydrolysis of resveratrol glucosides in wine. Food Chem 85:259–266 Lanz T, Schro¨der G, Schro¨der J (1990) Differential regulation of genes for resveratrol synthase in cell cultures of Arachis hypogaea. Planta 181:169–175 Lanz T, Tropf S, Marner FJ, Schro¨der J, Schro¨der G (1991) The role of cysteines in polyketide synthases. Site-directed mutagenesis of resveratrol and chalcone synthases, two key enzymes in different plant-specific pathways. J Biol Chem 266:9971–9976 Marillonnet S, Giritch A, Gils M, Kandzia R, Klimyuk V, Gleba Y (2004) In planta engineering of viral RNA replicons: efficient assembly by recombination of DNA modules delivered by Agrobacterium. Proc Natl Acad Sci USA 101:6852–6857 Medina-Bolivar F, Cramer C (2004) Production of recombinant protein by hairy roots cultured in plastic sleeve bioreactor. In:
Plant Cell Rep (2009) 28:589–599 Balbas P, Lorence A (eds) Recombinant gene expression: reviews and protocols, vol 267. Humana Press, Totowa, pp 351–363 Medina-Bolivar F, Condori J, Rimando AM, Hubstenberger J, Shelton K, O’Keefe SF, Bennett S, Dolan MC (2007) Production and secretion of resveratrol in hairy root cultures of peanut. Phytochemistry 68:1992–2003 Medrano G, Reidy MJ, Liu J, Ayala J, Dolan MC, Cramer CL (2009) Rapid system for evaluating bioproduction capacity of complex pharmaceutical proteins in plants. In: Faye L, Gomord V (eds) Recombinant proteins from plants: methods and protocols, vol 483. Humana Press, USA, pp 51–67 Pfaffl MW, Horgan GW, Dempfle L (2002) Relative expression software tool (RESTÓ) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res 30:e36 Pu¨ssa T, Floren J, Kuldkepp P, Raal A (2006) Survey of grapevine Vitis vinifera stem polyphenols by liquid chromatography–diode array detection–tandem mass spectrometry. J Agric Food Chem 54:7488–7494
599 Roupe KA, Remsberg CM, Yan˜ez JA, Davies NM (2006) Pharmacometrics of stilbenes: seguing towards the clinic. Curr Clin Pharmacol 1:81–101 Schro¨der J, Schro¨der G (1990) Stilbene and chalcone synthases: related enzymes with key function in plant-specific pathways. Z Naturforsch 45c:1–8 Sparvoli F, Martin C, Scienza A, Gavazzi G, Tonelli C (1994) Cloning and molecular analysis of structural genes for flavonoid and stilbene biosynthesis in grape (Vitis vinifera L.). Plant Mol Biol 24:743–755 Yu CKY, Shih CH, Chu IK, Lo C (2008) Accumulation of transpiceid in sorghum seedling infected with Colletotrichum sublineolum. Phytochemistry 69:700–706 Zhang Y, Li SZ, Li J, Pan X, Cahoon RE, Jaworski JG, Wang X, Jez JM, Cheng F, Yu O (2006) Using unnatural protein fusion to engineer resveratrol biosynthesis in yeast and mammalian cells. J Am Chem Soc 128:13030–13031
123
