JXB Advance Access originally published online on March 21, 2006
Journal of Experimental Botany 2006 57(6):1399-1411; doi:10.1093/jxb/erj120
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RESEARCH PAPER |
Down-regulation of cinnamoyl-CoA reductase in tomato (Solanum lycopersicum L.) induces dramatic changes in soluble phenolic pools
UMR 5546 CNRS-Université Paul Sabatier ‘Surfaces Cellulaires et Signalisation chez les Végétaux’, Pôle de Biotechnologie Végétale, 24 chemin de Borderouge, BP 42617, F-31326 Castanet, France
* To whom correspondence should be addressed. E-mail: rochange{at}scsv.ups-tlse.fr
Received 21 September 2005; Accepted 13 January 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Health-beneficial properties of many secondary plant metabolites have created much interest into the control of their biosynthesis in crop species. Phenolic compounds, including flavonoids, hydroxycinnamates, and tannins, make up an important group of such phytonutrients. They are formed via the phenylpropanoid pathway and share common precursors with lignin, an insoluble cell wall-associated polymer. In this study, the aim was to reduce lignin biosynthesis so as to enhance the availability of these precursors and, thereby, stimulate the production of soluble, potentially health-promoting, phenolic compounds in tomato (Solanum lycopersicum L.). First two tomato genes encoding cinnamoyl-CoA reductase (CCR), a key enzyme in the formation of lignin monomers, were identified and characterized. Transgenic plants exhibiting a reduced lignin content were subsequently obtained through an RNAi strategy targeting one of these genes. As anticipated, the total level of soluble phenolics was higher in stems and leaves of the transformants as compared with control plants. This was correlated with an increased antioxidant capacity of the corresponding plant extracts. Analysis of the soluble phenolic fraction by HPLC-MS revealed that vegetative organs of CCR down-regulated plants contained higher amounts of chlorogenic acid and rutin, and accumulated new metabolites undetectable in the wild type, such as N-caffeoyl putrescine and kaempferol rutinoside. In fruits, CCR down-regulation triggered the moderate accumulation of two new compounds in the flesh, but the total phenolic content was not affected. Although the prospects of exploiting such a strategy for crop improvement are limited, the results provide further insight into the control of the phenylpropanoid pathway in the Solanaceae.
Key words: Antioxidant, cinnamoyl-CoA reductase, flavonoid, hydroxycinnamate, lignin, Solanum lycopersicum, phenolic, tomato
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
There is growing awareness that a diet rich in fruit and vegetables contributes to the prevention of chronic diseases. Evidence coming from epidemiology, feeding studies, and in vitro analyses suggests that particular classes of phytonutrients exert protective effects against cancer or cardiovascular disease (for example, Liu, 2004
). Among putative beneficial phytonutrients, ascorbic acid (vitamin C), tocopherol (vitamin E), carotenoids, and phenolics (including flavonoids, hydroxycinnamates, lignans, etc.) are found at high levels in a wide variety of plants. This makes it difficult to assign a protective role to a particular compound, and the issue is further complicated by the extensive modifications undergone by these molecules prior to and during human absorption (Rice-Evans et al., 2000; Stahl et al., 2002). Likewise, the molecular mechanisms underlying these health-promoting effects are still largely unknown. For example, the beneficial effects of many phytonutrients have long been attributed to their antioxidant properties, but evidence is gathering that some antioxidant compounds such as flavonoids may in fact exert their effect by interfering with signalling pathways (Williams et al., 2004). Nevertheless, the interest for such compounds and their occurrence in the diet is growing.
In the past decade, there have been several successful attempts to increase the content of phytonutrients in crop species, as providing plant products enriched in certain compounds is considered more efficient than the use of food supplements (Cooper, 2004). Biotechnology is often considered the method of choice for the improvement of such characters, although traditional breeding can offer an interesting alternative by recruiting favourable genes/alleles from wild species closely related to crop plants (Willits et al., 2005). Recent achievements in the design of crops with added nutritional value include enhanced vitamin content in various crops through metabolic engineering (reviewed by Herbers, 2003), production of tomatoes with high flavonol content following overexpression of two key transcription factors (Bovy et al., 2002), and the production of phytoestrogens in non-legume species by expression of a heterologous isoflavone synthase gene (Yu et al., 2000). Recently, Davuluri et al. (2005) reported enhancement of both carotenoid and flavonoid content in tomato via silencing of the transcription factor DET1.
These approaches usually involve modifying the expression of key enzymes catalysing the synthesis of target compounds (e.g. chalcone isomerase for the production of flavonols; Muir et al., 2001) and/or of transcription factors that control the expression of these enzymes. The latter approach is often more efficient as whole biosynthetic pathways can be controlled by a small number of transcription factors (Borevitz et al., 2000; Bovy et al., 2002). By contrast, little is known about the possibility of enhancing availability of precursors involved in the synthesis of the requested molecule(s) by rerouting metabolic flow.
Lignin represents a major carbon sink in vascular plants, and is built from three main monomers termed monolignols. The biosynthesis of monolignols involves two specific steps branching off the general phenylpropanoid pathway. Therefore, decreasing lignin synthesis should make more precursors available for the synthesis of other phenolic compounds. Down-regulating cinnamoyl-CoA reductase (CCR; EC 1.2.1.44
[EC]
), the first committed enzyme of the monolignol biosynthetic pathway (Lacombe et al., 1997), appears to be one of the most efficient ways to decrease lignin biosynthesis (Piquemal et al., 1998; Anterola and Lewis, 2002). Limiting carbon flow down the monolignol pathway should enhance the availability of coumaroyl-CoA esters, which are some of the substrates of chalcone synthase, the first catalytic step towards flavonoid synthesis.
Tomato is one of the preferred targets for metabolic engineering because it is consumed widely in many Western countries, which makes it one of the principal sources of phytonutrients (Canene-Adams et al., 2005), and it is amenable to biotechnological modifications. In tomato fruits, the main carotenoid is lycopene. Phenolics are represented by a wide variety of compounds, including chlorogenic acid isomers (Niggeweg et al., 2004). Flavonoids, mainly represented by naringenin chalcone and rutin, are found in the skin in relatively small amounts (Bovy et al., 2002).
This paper reports on experiments aimed at reorientating the carbon flow from lignin formation to the synthesis of soluble phenolic compounds in transgenic tomato plants. It is shown that down-regulating CCR through an RNAi strategy leads to important quantitative and qualitative changes in the soluble phenolic content of extracts from fruits and vegetative organs. In stems and leaves, the observed modifications are strongly correlated with an increase in antioxidant capacity. Factors that limit the applicability of this approach for crop improvement are discussed.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Plant material and culture conditions
All experiments were performed on Solanum lycopersicum L. cv. Microtom, a miniature tomato cultivar commonly used as a model system (Meissner et al., 1997
). Plants were grown in soil in a culture room (16 h/8 h light/dark regime, 25 °C day/22 °C night, with 80% humidity) unless stated otherwise.
Cloning of LeCCR1 and LeCCR2
Total RNA was extracted from stems, flowers, and leaves using Extract'All reagent (Eurobio, France) according to the manufacturer's instructions, and cDNAs were obtained by reverse transcription using an oligo-dT primer. The coding sequence for LeCCR2 was amplified by PCR on the basis of the EST (expressed sequence tag) sequences found in the TIGR database (http://www.tigr.org). The 3' end of LeCCR1 was amplified using an oligo-dT primer and a primer derived from the partial coding sequence deduced from ESTs. Following sequencing of the 3' end, the whole coding sequence of LeCCR1 could be amplified. Full-length amplification products for both LeCCR1 and LeCCR2 were sequenced, and the sequences were deposited into GenBank under accession numbers DQ019125 and DQ019126, respectively.
RNAi construct
An expression cassette comprising the CaMV 35S promoter and CaMV polyadenylation signal was introduced into pGreen0029 binary vector (Hellens et al., 2000). A 220 bp intron (Vancanneyt et al., 1990) was then inserted between the promoter and the polyadenylation signal. The resulting plasmid, pSR02, allows the generation of RNAi constructs designed to produce self-complementary RNA which will be targeted for degradation.
A 420 bp fragment corresponding to the 5' region of LeCCR1 cDNA was amplified by PCR using upstream primer 5'-GTCGTCTGTGTTACCGGTG-3' and downstream primer 5'-ATCAGGATCGCTCCAGCA-3'. It was then introduced upstream and downstream of the intron in pSR02 in the antisense and sense orientations, respectively. The resulting vector, pVDR01, was introduced by the freeze–thaw method (Holsters et al., 1978) into Agrobacterium tumefaciens strain LBA4404 containing pSoup (Hellens et al., 2000).
Plant transformation
Tomato transformation through A. tumefaciens with vectors pSR02 and pVDR01 was performed as described in Ling et al. (1998). Rooted primary (T0) transformants were transferred to soil, grown in the greenhouse, and allowed to self-pollinate. Non-fertile transgenic plants were maintained by vegetative propagation; stem cuttings were dipped in Rootone F® (Bayer) hormone rooting powder and transferred to soil.
Recombinant proteins
Full-length coding sequences for LeCCR1 and LeCCR2 were amplified from mixed leaf and stem cDNAs. The PCR products were cloned into pGEX 4T-1 vector (Amersham) to obtain GST-CCR fusion proteins. These constructs were then introduced into Escherichia coli strain BL21 (Invitrogen). Bacteria were grown in liquid LB medium supplemented with 50 mg l–1 ampicillin for a few hours at 37 °C until OD reached 0.5. Production of recombinant protein was then induced by 0.2 mM IPTG, at 23 °C for 3 h. Bacterial cells were harvested by centrifugation, resuspended in PBS buffer, and lysed by sonication. The supernatant (crude extract) containing soluble recombinant protein was collected after centrifugation at 10 000 g. Recombinant proteins were purified on a glutathione resin (Amersham) according to the manufacturer's instructions.
Activity of purified recombinant proteins was measured spectrophotometrically according to Wengenmayer et al. (1976) using 2 µg of purified recombinant protein for each test. The reaction was carried out in 100 mM sodium citrate buffer pH 6.2 with 70 µM hydroxycinnamoyl-CoA and 40 µM NADPH. The activity was deduced from the decrease of absorbance corresponding to the transformation of hydroxycinnamoyl-CoA (for feruloyl-CoA: 
366=13.0 mM–1 cm–1) into its corresponding hydroxycinnamaldehyde (for coniferaldehyde: 366=10.0 mM–1 cm–1) and to the oxidation of NADPH (366=3.0 mM–1 cm–1). The four substrates, feruloyl-CoA, sinapoyl-CoA, caffeoyl-CoA, and coumaroyl-CoA, were synthesized according to the method of Stöckigt and Zenk (1975).
RNA gel blot analysis
Total RNA was extracted from plant samples using Extract'All reagent (Eurobio, France). Ten micrograms of RNA of each sample were run on a formaldehyde denaturing gel and blotted onto a nylon membrane (Hybond N+; Schleicher and Schuell). Blots were hybridized according to Church and Gilbert (1984) with a [
-32P]-labelled DNA probe corresponding to the 200 bp 3'-end of the LeCCR1 coding sequence.
Histochemistry
Stem transverse sections obtained with a vibratome (100 µm thick for phloroglucinol, 200 µm thick for autofluorescence) and fruit longitudinal sections were cleared by overnight incubation in 95% ethanol, and were either mounted directly on a glass side for fluorescence microscopy or stained for 30 min with phloroglucinol-HCl reagent (Prolabo, France). Sections were observed using an inverted microscope (Leitz DMIRBE, Leica, Heidelberg, Germany), as described by Chabannes et al. (2001).
Determination of lignin content
Whole stems were harvested from 3-month-old wild-type and transgenic plants, frozen in liquid nitrogen, and freeze-dried. Cell wall residue was prepared as described previously (Chabannes et al., 2001), and Klason lignin content was determined using a micro-Klason technique (Whiting et al., 1981). Two plants from each line were pooled for extraction of the cell wall residue, and measurements were performed in triplicate for each cell wall residue sample.
HPLC analysis of phenolic compounds
When the first fruits reached maturity, leaves, stems, and fruits at different stages of ripening were collected, ground in liquid nitrogen, and freeze-dried. Tissue samples were taken from at least two individual plants from each line and extracted separately. Soluble phenolics were extracted from 50 mg of lyophilized powder by stirring the sample three times in 500 µl of methanol/water (2:1, v/v), so that the final extract volume was 1.5 ml. Prior to injection, a 50 µl aliquot of each extract was rotary evaporated under vacuum and subsequently resuspended in the same volume of water/methanol (4:1; v/v). For each run, 20 µl of sample was injected into the HPLC column. HPLC separation was performed in duplicate for each extract.
HPLC analysis was carried out on a Spectra Physics system with an SP8800 ternary pump. UV spectra from 240 nm to 380 nm were monitored with a UV-3000 (Spectra-system) detector. Analyses were performed using a 250 x 4.6 mm RP-C18 Nova-Pak column (Waters). Mobile phase A consisted of water/methanol/5 M HCl (80:20:0.1; v/v/v) and solvent B was acetonitrile. The following linear gradient was run at a flow rate of 0.6 ml min–1 for 60 min: 95A:5B from 0–10 min, decreasing linearly to 50A:50B by 35 min, held for a further 5 min, and back to 95A:5B by 45 min. The last 15 min (45–60 min) at 95A:5B ensured re-equilibration of the column. Compound quantification was based on the area under peak, determined at 320 nm.
LC-MS and LC-MS/MS characterization
HPLC separation was carried out with an Agilent 1100 system with detection at 320 nm. The C18 Nova-Pak (Waters) column and the mobile phase were identical to those used for HPLC-UV analysis with the following exceptions: 0.05% aqueous formic acid was substituted for 0.1% 5 M HCl in the mobile phase A and the volume of sample injected was reduced to 5 µl. Atmospheric pressure ionization mass spectrometry analysis was performed on a Q-Trap mass spectrometer operated using the electrospray ionization interface in the negative mode (ESI–). Capillary voltage was set at 4200 V and source temperature at 400 °C, and the data were analysed using Analyst 1.4 software (Applied Biosystems). For LC-MS, mass spectrometric data were acquired in the full scan mode over the m/z 150–800 range. Sensitivity of the mass spectrometer was optimized using a chlorogenic acid standard (Acros) in the 150–400 range and a quercetin-rutinoside standard (Extrasynthese, France) in the 400–800 range. For LC-MS/MS, selected ions were fragmented using a collision energy of –25 V, except compound 6 which was fragmented at –15 V.
Total phenolic content and antioxidant capacity
Methanol/water extracts (2:1; v/v) obtained as described previously were evaporated under vacuum, the dry residue was resuspended in the same volume of water, and centrifuged for 2 min at 12 000 g. Total phenolic content was estimated by the Folin-Ciocalteu method adapted from Dicko et al. (2002). Assays were performed in duplicate for each extract by mixing 20 µl supernatant, 20 µl water, and 100 µl 50% Folin-Ciocalteu's reagent. After 5 min incubation at room temperature, 100 µl of 20% (w/v) Na2CO3 and 560 µl water were added. The mixture was incubated for a further 30 min at room temperature and absorption at 720 nm was determined using a Cary100-UV spectrophotometer (Varian, Australia). A calibration curve was obtained with freshly prepared solutions of gallic acid. Results were calculated as micrograms gallic acid equivalents per millilitre of extract.
Antioxidant capacity was determined in the same extracts by a TEAC assay, based on the capacity of samples to scavenge 2,2'-azinobis(3-ethylbenzothiozoline-6-sulphonate) (ABTS+) radical cations. Measurements were performed in triplicate for each extract as described by Braca et al. (2003). Antioxidant capacity was determined relative to the reactivity of Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid; Aldrich) as a standard, and expressed as nanomoles Trolox equivalent per microlitre of extract.
For both measurements (total phenolic content and antioxidant capacity), each extract represented an individual plant from one of the lines, and the whole experiment was repeated with similar results.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Cloning of two tomato CCR genes and characterization of recombinant proteins
The aim of this study was to reduce lignin biosynthesis in tomato plants through down-regulation of CCR activity, and to evaluate subsequent impacts on phenylpropanoid metabolism. The first gene encoding CCR was cloned from Eucalyptus gunnii (Lacombe et al., 1997
). Since then, a number of CCR genes have been identified in dicot as well as monocot species (reviewed by Boudet et al., 2004). Despite the existence of a large set of sequences annotated as ‘CCR-like proteins’ (Raes et al., 2003), bona fide CCR enzymes seem to be encoded by a small number of closely related genes; so far, no more than two genes have been demonstrated experimentally to encode CCR in one species. The E. gunnii CCR sequence was used to search the TIGR EST database (http://tigrblast.tigr.org) for tomato homologues. This search yielded two contigs with a very high homology to the target sequence (80% identity at the amino acid level), plus a few other contigs that were less similar (45–48% identity). The two sequences with highest homology to E. gunnii CCR were named LeCCR1 and LeCCR2. While it is likely that LeCCR2 corresponded to a full-length CCR sequence, LeCCR1 lacked the 3' end, which was cloned by a 3' RACE approach. Complete reconstituted sequences of LeCCR1 and LeCCR2 were used to design oligonucleotide primers at the 3' and 5' ends. Full-length cDNAs corresponding to both genes were then amplified by RT-PCR. Their exact sequences were determined and deposited in Genbank under accession numbers DQ019125 and DQ019126. Both coding sequences are 999 bp long, and encode proteins with a predicted molecular mass of 36.8 kDa. They share 83% and 87% identity at the nucleotide and amino-acid levels, respectively. By comparison with CCR genes from other plant species, it can be concluded that LeCCR1 and LeCCR2 are closer to each other than to any other CCR gene (data not shown).
Previous studies have shown that one of the two CCR genes is usually predominant in lignin biosynthesis, and that down-regulating that gene can efficiently trigger a decrease in lignin content (Jones et al., 2001; Lauvergeat et al., 2001; Goujon et al., 2003). Therefore, an investigation that was carried out to find out which of the newly cloned tomato CCR genes would be more appropriate as a silencing target. Towards this end, recombinant GST-fusion proteins were produced in E. coli and purified to >90% homogeneity, as estimated by SDS-PAGE (data not shown). Specific activities of purified, soluble fusion proteins were determined for four potential CCR substrates: feruloyl-CoA, sinapoyl-CoA, caffeoyl-CoA, and coumaroyl-CoA (Table 1). These results clearly show that LeCCR1 is able to use the four substrates albeit with a strong preference for feruloyl-CoA, while LeCCR2 only uses feruloyl-CoA, with a lower efficiency than LeCCR1. Data obtained on recombinant proteins are likely to reflect the properties of the corresponding native proteins, as was shown for E. gunnii CCR enzymes (Lacombe et al., 1997).
|
Expression patterns for LeCCR1 and LeCCR2 were examined by RT-PCR. While LeCCR2 was found to be expressed only in roots and stems, LeCCR1 was expressed in stems, petioles, leaves, and fruits at the expanding green and red stages (data not shown). These observations combined with the higher CCR activity exhibited by LeCCR1 recombinant protein led this gene to be selected as a target for RNAi silencing.
Phenotype of CCR down-regulated tomato plants
Generation of transgenic lines:
In order to obtain RNAi-mediated silencing of LeCCR1, a construct designed to produce dsRNA under the control of the cauliflower mosaic virus (CaMV) 35S promoter was introduced into tomato plants through Agrobacterium tumefaciens. Sixteen independent transformants were regenerated. Three of them were selected for further analysis on the basis of a strongly reduced LeCCR1 expression (Fig. 1) and the ability to produce fruits. However, in these selected lines, seed formation was strongly affected and the very small seeds produced were unable to germinate. Therefore, these lines had to be maintained by vegetative propagation. In these plants, steady-state levels of LeCCR2 transcript were also reduced, albeit to a lesser extent than LeCCR1 (Supplementary Fig. S1 can be found at JXB online.). A transgenic line carrying the empty vector pSR02 was also generated as a control and taken to the T2 generation. Transformants were subsequently compared with both wild-type and control plants.
|
Plant morphology and development:
All transgenic plants with reduced LeCCR1 expression exhibited a dramatic reduction of size as compared with the wild-type or control plants (Fig. 2A). Furthermore, the leaves of CCR down-regulated lines were darker and smaller (they contained one to five leaflets versus up to nine in the wild type). The three selected lines, L2, L5, and L15, could only produce a limited number of small fruits (Fig. 2B).
|
Microscopic examination and histochemistry:
Sections were taken from stems and fruits and observed either for lignin autofluorescence under UV light or after phloroglucinol staining. Under UV light, xylem vessels in stems and petioles of CCR down-regulated plants appeared partially collapsed with the cell walls apparently buckling under mechanical stress, as compared with control vessels which exhibited a round-open shape (Fig. 3A, B). The differences between CCR down-regulated and wild-type or control lines were even more obvious after staining with phloroglucinol; xylem tissue was hardly stained in the CCR-deficient plants while it exhibited a strong pink-red coloration in wild-type and control stems (Fig. 3C, D). Similar observations were made on vascular bundles from fruits (Fig. 3E, F). Another distinctive feature of CCR down-regulated plants was the brown-orange coloration of the xylem ring in the basal part of the stem, visible without any staining (not shown).
|
Lignin content:
Klason lignin content was determined in the cell wall residue extracted from whole stems. This experiment allows the quantification of sulphuric acid-insoluble lignin material in plant samples (Whiting et al., 1981
). Table 2 shows that Klason lignin content is markedly decreased (by 20–37%) in CCR-deficient plants as compared with wild-type and control lines.
|
Analysis of soluble phenolic compounds
In order to evaluate the potential impact of CCR down-regulation on different pathways of phenolic metabolism, methanol-soluble phenolic compounds were quantified by a global assay, then separated by HPLC. Vegetative organs (stems and leaves), as well as ripe fruits, were studied separately.
Determination of total phenolic content:
Following reduction of CCR expression and of lignin synthesis, the phenylpropanoid metabolic flow was expected to be reoriented towards the production of soluble phenolics. To investigate whether this was the case, the total soluble phenolic content of extracts from different organs was estimated using Folin-Ciocalteu's method (Dicko et al., 2002). The results (illustrated in Fig. 4) clearly show the accumulation of soluble phenolics at higher levels in stems and leaves of CCR down-regulated lines as compared with control plants. Differences are significant according to Student's t test (=0.05), and correspond to 3.1- and 1.4-fold increases in the total phenolic content of stems and leaves, respectively. By contrast, no significant difference could be observed in total phenolic contents of fruit extracts (P=0.09).
|
Global comparison of phenolic profiles:
To further characterize modifications of phenolic metabolism in CCR down-regulated plants, methanol-soluble fractions were subjected to HPLC separation. As important differences were detected, the study focused on the most abundant compounds. Major peaks detected in extracts from all organs were numbered consecutively according to their order of elution on the HPLC column. Chromatograms from wild-type plants (Fig. 5) show that some of the major phenolic compounds are present in stems, leaves, and fruits, while others are specific to a particular organ (e.g. compound 3). Differences between wild-type and CCR down-regulated plants can be observed by comparing HPLC profiles, and were confirmed by quantification of individual compounds (Table 3). In stems, some of the major compounds were hyper-accumulated in the transformants (compounds 1, 4, and 9), and four new compounds appeared that were undetectable in the wild type (2, 5, 6, and 11; Fig. 5; Table 3) even when analysed by LC-MS. Leaf extracts also exhibited differences to the wild-type, albeit to a lesser extent. The same compounds specifically observed in CCR down-regulated stems were also present (peaks 2, 5, 6, and 11), together with at least three new compounds (7, 8, and 12; Fig. 5; Table 3). By contrast, HPLC profiles from fruits did not show significant differences between lines. Amounts of specific phenolic compounds in the fruit varied to a much larger extent than in vegetative organs; important variations were observed between several fruit samples from the same line, whereas extracts from vegetative parts were essentially similar. Technical replicates (i.e. repeated extraction and analysis of the same sample) gave consistent results, indicating that variation was not due to the analytical procedure. This observation suggests that phenolic compound accumulation in the fruit is more subject to biological variation, and could be strongly influenced by slight differences in stage of development. For this reason, it was difficult to detect significant differences between transgenic and control fruits. Yet, compounds 5 and 6 were consistently detected in the fruits from CCR-deficient lines and absent from wild-type and control fruits (Fig. 5; Table 3). Separate analysis of the phenolics extracted from the flesh and skin of these fruits revealed that this trait was more conspicuous in the flesh, while several other fruit phenolics were more abundant in the skin (Fig. 6).
|
|
|
Compound identification:
Identification of phenolic compounds, which had differentially accumulated between control and transformed lines, was attempted using information derived from UV/Vis absorption spectra, HPLC retention times, and LC-MS or LC-MS/MS data. Table 3 summarizes the data available for major compounds: abundance in CCR-deficient lines relative to the wild type, retention time on HPLC, UV/Vis absorption maxima, molecular mass, MS/MS fragmentation ions, and compound name if determined. LC-MS and LC-MS/MS spectra of tentatively identified compounds are presented in Supplementary figures S2 and S3 at JXB online. The identity of compounds was confirmed by spiking with the corresponding standard when available.
- Peaks 1, 3, and 4 exhibited UV spectral characteristics similar to chlorogenic acid isomers, and displayed, after LC-MS analysis, an m/z of 353 [M-H]–1 and a major fragmentation ion at 191 [M-H]–1 which is likely to correspond to the quinic acid moiety.
- Peak 4 was confirmed to correspond to chlorogenic acid (5-caffeoylquinic acid) by spiking with a standard.
- Compounds 1 and 3 therefore are likely to correspond to chlorogenic acid isomers. On the basis of elution times and by analogy with previous work (Chabannes et al., 2001
), it was proposed that compound 1 is 3-chlorogenic acid (neochlorogenic acid) and compound 3 is 4-chlorogenic acid (cryptochlorogenic acid).
- Compound 2, though abundant in the CCR down-regulated lines, could not be identified by LC-MS since the modification of the elution protocol (the use of formic acid instead of HCl) resulted in the disappearance of this compound from the LC chromatogram.
- Compound 5, with an m/z of 355 [M-H]–1 and fragmentation ions of 193 [M-H]–1 and 179 [M-H]–1 was tentatively identified as a hexoside of ferulic acid.
- Compound 6 exhibited an m/z of 251 [M+H]+1. Its UV spectral properties are very similar to those reported for hydroxycinnamic acid amide conjugates (Whitaker and Stommel, 2003
). On the basis of this observation and the molecular mass of this compound, it is proposed that it is N-caffeoyl putrescine.
- UV spectra of compounds 7, 8, and 12 were very similar to that of chlorogenic acid, suggesting that they may contain a hydroxycinnamoyl moiety; however, their identity could not be determined precisely by LC-MS.
- Peak 9 exhibited an m/z of 609 [M-H]–1 and was identified as quercetin-O-rutinoside (rutin) by co-elution with a standard.
- Compounds 10 and 13, with an m/z of 515 [M-H]–1, fragmentation ions at 353 [M-H]–1 and 191 [M-H]–1, and UV spectra similar to chlorogenic acid were tentatively identified as two isomers of dicaffeoylquinic acid.
- Peak 11, with an m/z of 593 [M-H]–1, was identified as kaempferol-O-rutinoside by comparison with a standard.
- Peak 4 was confirmed to correspond to chlorogenic acid (5-caffeoylquinic acid) by spiking with a standard.
Antioxidant properties of plant extracts
Since the general aim of this work was to improve the antioxidant content in tomato, and despite the fact that few changes were observed in the phenolic patterns of the edible part of the tomato plant, the fruit, antioxidant properties of methanol extracts from different organs were measured using a TEAC assay. This method measures the capacity of a solution to scavenge ABTS+ radicals and is suitable for the determination of antioxidant properties of phenolics such as hydroxycinnamates and flavonoids (Re et al., 1999). As shown in Fig. 7A, total antioxidant capacity is increased, respectively, 3.4- and 1.6-fold in stem and leaf extracts from CCR down-regulated plants as compared with control lines. As these results are very similar to those reported for total phenolic content (Fig. 4) a possible correlation between these two traits was examined. Figure 7B shows a plot of TEAC values versus total phenolic content in stem extracts from various plants, and reveals a remarkable correlation between these two measurements (R2=0.999). A similar observation can be made on leaf extracts (R2=0.986). In fruit samples, antioxidant capacity did not differ significantly between control and transformed lines, as was also observed for total phenolic content.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Two tomato CCR genes, LeCCR1 and LeCCR2, were cloned on the basis of a high sequence homology with E. gunnii CCR. The identity of these genes was subsequently confirmed by analysis of the catalytic properties of corresponding recombinant proteins produced in E. coli. The two enzymes exhibited contrasting activities on the four cinnamoyl-CoA substrates tested; while LeCCR1 could use all four substrates, LeCCR2 could only reduce feruloyl-CoA. Furthermore, LeCCR1 was more active than LeCCR2 on feruloyl-CoA, and was expressed in a wider range of organs. This situation is similar to that reported in Arabidopsis thaliana by Lauvergeat et al. (2001)
; one of the Arabidopsis CCR enzymes, encoded by At1g15950, exhibits higher activity on different cinnamoyl-CoA thioesters than the other (At1g80820). Expression studies, as well as other reports based on the analysis of mutants or transgenic plants, revealed that the former is most likely to be involved in constitutive lignification, while the latter seems to be preferentially expressed when the plant is challenged with a bacterial pathogen. Based on the results of expression and recombinant protein studies, LeCCR1 was chosen as a target for the silencing strategy. Of the 16 transformants obtained following tomato transformation with a LeCCR1 RNAi construct, many were unable to set fruit. Since this organ was of particular interest for the present study, a decision was made to select three transgenic lines that exhibited a strong reduction of LeCCR1 expression but were still able to produce fruits. Lignin content determined in vegetative organs by the micro-Klason method (Whiting et al., 1981) was 20–37% lower in these transgenic lines as compared with control plants. The lignin content could not be quantified in the fruits, but absence of staining of vascular traces by phloroglucinol revealed that lignin synthesis was also affected in this organ. This reduction in lignin content was an expected outcome of transformation with the LeCCR1 RNAi construct, since extinction of CCR genes in other species has already been reported to bring about important decreases in lignin content. In the present case, the differences were not as important as those reported for tobacco plants (about 50% decrease in Klason lignin content; Piquemal et al., 1998). This may be due to the selection of lines with an intermediate level of CCR down-regulation, since the lines exhibiting the most severe phenotype were discarded because they did not set fruit. Alternatively, this discrepancy may be related to the fact that Piquemal et al. (1998) analysed isolated xylem tissue, which was not possible for very small tomato plants. When whole stems from Arabidopsis CCR antisense lines were studied (Goujon et al., 2003), the reduction in Klason lignin content was similar to the present results. In agreement with their decreased lignin content, transgenic lines L2, L5, and L15 exhibited a morphological phenotype typical of CCR down-regulated plants: stunting, collapsed vessels, absence of staining by phloroglucinol of vascular tissue, and coloration of xylem in unstained basal stems. Collapsing of vessels can be explained by the lower resistance to negative pressure of cell walls containing less lignin. This in turn may affect plant growth and development by decreasing water and solute transport efficiency (Piquemal et al., 1998). Effects of CCR down-regulation on fertility observed in tomato are the most severe described so far, although reduced seed viability and decreased fertility at high temperatures have already been reported in Arabidopsis (Jones et al., 2001; Goujon et al., 2003). At the moment it is not known if the altered fertility and seed production is linked to the reduction in lignin content or to other induced metabolic changes that occur in these plants. As a result of success in decreasing lignin synthesis in transgenic tomato plants, the consequences of this modification on the soluble phenolic content were next examined. Total phenolic content was markedly increased in stems and leaves of transformants as compared with control plants, indicating that the pool of precursors left unused by the monolignol biosynthetic pathway had been reallocated to the synthesis of other phenolic compounds. However, this was not the case in fruits, most likely because this organ is poorly lignified. This observation suggests that soluble phenolic compounds accumulated in vegetative organs as a result of CCR down-regulation are not transported towards the fruit.
In order to further characterize the modifications that occurred in phenolic metabolism, methanol extracts were subjected to separation by HPLC. In stems and leaves, the observed accumulation of soluble phenolic compounds corresponded to both quantitative and qualitative changes; some compounds already present in wild-type plants accumulated to higher levels in the transgenics (e.g. chlorogenic acid and rutin), and new compounds undetectable in the wild type appeared as major contributors to the phenolic profile in the transformants.
Most of the differentially accumulated compounds that could be tentatively identified on the basis of UV-Vis spectral properties and LC-MS data are hydroxycinnamate derivatives, with the exception of two flavonoids, rutin and kaempferol rutinoside. Interestingly, the majority of these compounds are potent radical scavengers, as has been documented for several hydroxycinnamate derivatives (Sawa et al., 1999; Niggeweg et al., 2004). This is also illustrated by the remarkable correlation observed between the increases in total phenolic content and antioxidant capacity measured in extracts from transgenic plants. Indeed, stimulating the synthesis of some of these compounds is considered the target of interest to improve fruit nutritional quality (Niggeweg et al., 2004). In the present case, however, the strong morphological phenotype of transformants, as well as the absence of soluble phenolic accumulation in fruits, makes CCR down-regulation an inappropriate approach.
From a more fundamental point of view, the identity of the differentially accumulated compounds provides new insight into the control of phenylpropanoid metabolism in tomato. First, all hydroxycinnamate derivatives identified in this study comprise a caffeic acid or ferulic acid moiety (Table 3). While the accumulation of a ferulic acid derivative could be expected, feruloyl-CoA being the preferred substrate of both tomato CCR isoforms (Table 1), the predominance of caffeate derivatives was more intriguing. Caffeate and caffeoyl-CoA are synthesized upstream of CCR, and are methylated into ferulate and feruloyl-CoA, respectively, by a caffeate O-methyltransferase and a caffeoyl-CoA O-methyltransferase. The abundance of caffeate derivatives in CCR down-regulated plants could therefore be attributed to a regulation of the metabolic flow (either by feedback inhibition or transcriptional control) at this O-methylation step. By contrast, the absence of identifiable coumarate derivatives suggests that the flow entering the 3'-hydroxylation step (from coumarate to caffeate) is poorly affected by CCR down-regulation. This may explain the limited accumulation of end-products deriving from coumaroyl-CoA, such as flavonoids. Alternatively, other limiting factors may restrain flavonoid synthesis: availability of malonyl-CoA, chalcone isomerase activity (Verhoeyen et al., 2002), or activity of other flavonoid biosynthetic enzymes. Indeed, up-regulation of the corresponding genes via overexpression of key transcription factors leads to the accumulation of flavonoids regardless of precursor availability (Schijlen et al., 2004).
Secondly, several of the differentially accumulated compounds have been described as predominant phenolic metabolites in Solanaceae. For example, chlorogenic acid (compounds 1, 3, and 4) is the most abundant phenolic compound in tobacco leaves and stems, and may correspond to a storage form that can be used when necessary to build lignin or other downstream phenolics. Its increased abundance in CCR down-regulated transformants may involve hydroxycinnamoyl-CoA:shikimate/quinate hydroxycinnamoyl transferase (Hoffmann et al., 2003), since this enzyme can synthesize chlorogenate back from caffeoyl-CoA, a substrate of CCR. Caffeoyl putrescine (compound 6), on the other hand, is not constitutively present in tomato extracts but it has been shown to accumulate to important amounts in Solanaceae in response to jasmonate treatment and fungal infection (Keller et al., 1996; Keinänen et al., 2001). In the present case, its accumulation may either correspond to re-routing of the metabolic flow following blocking of the monolignol pathway, or to a stress response due to the severe developmental alterations observed in the transformants.
In conclusion, for the first time down-regulation of CCR expression is reported in a fruit species. This modification leads to the accumulation of antioxidant soluble phenolic compounds in vegetative organs, but not in the fruit, and the present results highlight the tight relationship between soluble phenolic abundance and antioxidant properties of plant extracts. Precursors made available by CCR down-regulation are only partly used for the synthesis of flavonoids, which shows the existence of complex regulatory networks within phenylpropanoid metabolism. Finally, the present work confirms that, as previously demonstrated for other metabolic pathways, rerouting metabolic flows can lead to unexpected effects at the biochemical and/or developmental levels. As illustrated by the present results, this can include the synthesis of novel compounds not detected in normal plants. This remarkable plasticity of plant secondary metabolism could potentially be exploited in future metabolic engineering studies for the production of high-value pharmaceutical compounds.
|
Supplementary data |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Supplementary data are available at JXB online.
|
Acknowledgements |
|---|
This work was funded by the European Community (PROFOOD program, contract no. QLK1-CT-2001-01080 to BvdR).
|
Footnotes |
|---|
Abbreviations: CCR, cinnamoyl-CoA reductase; ESI, ElectroSpray ionization; ABTS, 2,2'-azinobis(3-ethylbenzothiozoline-6-sulphonate); Trolox, 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Anterola AM, Lewis NG. 2002. Trends in lignin modification: a comprehensive analysis of the effects of genetic manipulations/mutations on lignification and vascular integrity. Phytochemistry 61, 221–294.[CrossRef][ISI][Medline]
Borevitz JO, Xia Y, Blount J, Dixon RA, Lamb C. 2000. Activation tagging identifies a conserved MYB regulator of phenylpropanoid biosynthesis. The Plant Cell 12, 2383–2394.
Boudet AM, Hawkins S, Rochange S. 2004. The polymorphism of the genes/enzymes involved in the last two reductive steps of monolignol synthesis: what is the functional significance? Comptes Rendus Biologies 327, 837–845.[ISI][Medline]
Bovy A, de Vos R, Kemper M, et al. 2002. High-flavonol tomatoes resulting from the heterologous expression of the maize transcription factor genes LC and C1. The Plant Cell 14, 2509–2526.
Braca A, Politi M, Sanogo R, Sanou H, Morelli I, Pizza C, De Tommasi N. 2003. Chemical composition and antioxidant activity of phenolic compounds from wild and cultivated Sclerocarya birrea (Anacardiaceae) leaves. Journal of Agricultural and Food Chemistry 51, 6689–6695.[CrossRef][ISI][Medline]
Canene-Adams K, Campbell JK, Zaripheh S, Jeffery EH, Erdman Jr JW. 2005. The tomato as a functional food. Journal of Nutrition 135, 1226–1230.
Chabannes M, Barakate A, Lapierre C, Marita JM, Ralph J, Pean M, Danoun S, Halpin C, Grima-Pettenati J, Boudet AM. 2001. Strong decrease in lignin content without significant alteration of plant development is induced by simultaneous down-regulation of cinnamoyl CoA reductase (CCR) and cinnamyl alcohol dehydrogenase (CAD) in tobacco plants. The Plant Journal 28, 257–270.[CrossRef][ISI][Medline]
Church GM, Gilbert W. 1984. Genomic sequencing. Proceedings of the National Academy of Sciences, USA 81, 1991–1995.
Cooper DA. 2004. Carotenoids in health and disease: recent scientific evaluations, research recommendations and the consumer. Journal of Nutrition 134, 221–224.
Davuluri GR, van Tuinen A, Fraser PD, et al. 2005. Fruit-specific RNAi-mediated suppression of DET1 enhances carotenoid and flavonoid content in tomatoes. Nature Biotechnology 23, 890–895.[CrossRef][ISI][Medline]
Dicko MH, Hilhorst R, Gruppen H, Traore AS, Laane C, van Berkel WJH, Voragen AGJ. 2002. Comparison of content in phenolic compounds, polyphenol oxidase, and peroxidase in grains of fifty sorghum varieties from Burkina Faso. Journal of Agricultural and Food Chemistry 50, 3780 –3788.[CrossRef][ISI][Medline]
Goujon T, Ferret V, Mila I, Pollet B, Ruel K, Burlat V, Joseleau JP, Barriere Y, Lapierre C, Jouanin L. 2003. Down-regulation of the AtCCR1 gene in Arabidopsis thaliana: effects on phenotype, lignins and cell wall degradability. Planta 217, 218–228.[ISI][Medline]
Hellens RP, Edwards EA, Leyland NR, Bean S, Mullineaux PM. 2000. pGreen: a versatile and flexible binary Ti vector for Agrobacterium-mediated plant transformation. Plant Molecular Biology 42, 819–832.[CrossRef][ISI][Medline]
Herbers K. 2003. Vitamin production in transgenic plants. Journal of Plant Physiology 160, 821–829.[CrossRef][ISI][Medline]
Hoffmann L, Maury S, Martz F, Geoffroy P, Legrand M. 2003. Purification, cloning, and properties of an acyltransferase controlling shikimate and quinate ester intermediates in phenylpropanoid metabolism. Journal of Biological Chemistry 278, 95–103.
Holsters M, de Waele D, Depicker A, Messens E, van Montagu M, Schell J. 1978. Transfection and transformation of Agrobacterium tumefaciens. Molecular and General Genetics 163, 181–187.
Jones L, Ennos AR, Turner SR. 2001. Cloning and characterization of irregular xylem4 (irx4): a severely lignin-deficient mutant of Arabidopsis. The Plant Journal 26, 205–216.[CrossRef][ISI][Medline]
Keinanen M, Oldham NJ, Baldwin IT. 2001. Rapid HPLC screening of jasmonate-induced increases in tobacco alkaloids, phenolics, and diterpene glycosides in Nicotiana attenuata. Journal of Agricultural and Food Chemistry 49, 3553–3558.[CrossRef][ISI][Medline]
Keller H, Hohlfeld H, Wray V, Hahlbrock K, Scheel D, Strack D. 1996. Changes in the accumulation of soluble phenolics in elicitor-treated cell suspension cultures and fungus-infected leaves of Solanum tuberosum. Phytochemistry 42, 389–396.[CrossRef]
Lacombe E, Hawkins S, Van Doorsselaere J, Piquemal J, Goffner D, Poeydomenge O, Boudet AM, Grima-Pettenati J. 1997. Cinnamoyl CoA reductase, the first committed enzyme of the lignin branch biosynthetic pathway: cloning, expression and phylogenetic relationships. The Plant Journal 11, 429–441.[CrossRef][ISI][Medline]
Lauvergeat V, Lacomme C, Lacombe E, Lasserre E, Roby D, Grima-Pettenati J. 2001. Two cinnamoyl-CoA reductase (CCR) genes from Arabidopsis thaliana are differentially expressed during development and in response to infection with pathogenic bacteria. Phytochemistry 57, 1187–1195.[CrossRef][ISI][Medline]
Ling HQ, Kriseleit D, Ganal MW. 1998. Effect of ticarcillin/potassium clavulanate on callus growth and shoot regeneration in Agrobacterium-mediated transformation of tomato (Lycopersicon esculentum Mill.). Plant Cell Reports 17, 843–847.[CrossRef]
Liu RH. 2004. Potential synergy of phytochemicals in cancer prevention: mechanism of action. Journal of Nutrition 134, 3479S–3485S.
Meissner R, Jacobson Y, Melamed S, Levyatuv S, Shalev G, Ashri A, Elkind Y, Levy AA. 1997. A new model system for tomato genetics. The Plant Journal 12, 1465–1472.[CrossRef][ISI]
Muir SR, Collins GJ, Robinson S, Hughes S, Bovy A, Ric De Vos CH, van Tunen AJ, Verhoeyen ME. 2001. Overexpression of petunia chalcone isomerase in tomato results in fruit containing increased levels of flavonols. Nature Biotechnology 19, 470–474.[CrossRef][ISI][Medline]
Niggeweg R, Michael AJ, Martin C. 2004. Engineering plants with increased levels of the antioxidant chlorogenic acid. Nature Biotechnology 22, 746–754.[CrossRef][ISI][Medline]
Piquemal J, Lapierre C, Myton K, O'Connell A, Schuch W, Grima-Pettenati J, Boudet AM. 1998. Downregulation of cinnamoylCoA reductase induces significant changes of lignin profiles in transgenic tobacco plants. The Plant Journal 13, 71–83.
Raes J, Rohde A, Christensen JH, Van de Peer Y, Boerjan W. 2003. Genome-wide characterization of the lignification toolbox in Arabidopsis. Plant Physiology 133, 1051–1071.
Re R, Pellegrini N, Proteggente A, Pannala A, Yang M, Rice-Evans C. 1999. Antioxidant activity applying an improved ABTS radical cation decolorization assay. Free Radical Biology & Medicine 26, 1231–1237.
Rice-Evans C, Spencer JP, Schroeter H, Rechner AR. 2000. Bioavailability of flavonoids and potential bioactive forms in vivo. Drug Metabolism and Drug Interactions 17, 291–310.[Medline]
Sawa T, Nakao M, Akaike T, Ono K, Maeda H. 1999. Alkylperoxyl radical-scavenging activity of various flavonoids and other phenolic compounds: implications for the anti-tumor-promoter effect of vegetables. Journal of Agricultural and Food Chemistry 47, 397–402.[CrossRef][ISI][Medline]
Schijlen EG, Ric de Vos CH, van Tunen AJ, Bovy AG. 2004. Modification of flavonoid biosynthesis in crop plants. Phytochemistry 65, 2631–2648.[CrossRef][ISI][Medline]
Stahl W, van den Berg H, Arthur J, et al. 2002. Bioavailability and metabolism. Molecular Aspects of Medicine 23, 39–100.[CrossRef][Medline]
Stockigt J, Zenk MH. 1975. Chemical syntheses and properties of hydroxycinnamoyl-coenzyme A derivatives. Zeitschrift für Naturforschung 30, 352–358.
Vancanneyt G, Schmidt R, O'Connor-Sanchez A, Willmitzer L, Rocha-Sosa M. 1990. Construction of an intron-containing marker gene: splicing of the intron in transgenic plants and its use in monitoring early events in Agrobacterium-mediated plant transformation. Molecular and General Genetics 220, 245–250.
Verhoeyen ME, Bovy A, Collins G, Muir S, Robinson S, de Vos CH, Colliver S. 2002. Increasing antioxidant levels in tomatoes through modification of the flavonoid biosynthetic pathway. Journal of Experimental Botany 53, 2099–2106.
Wengenmayer H, Ebel J, Grisebach H. 1976. Enzymic synthesis of lignin precursors: purification and properties of a cinnamoyl-CoA:NADPH reductase from cell suspension cultures of soybean (Glycine max). European Journal of Biochemistry 65, 529–536.[ISI][Medline]
Whitaker BD, Stommel JR. 2003. Distribution of hydroxycinnamic acid conjugates in fruit of commercial eggplant (Solanum melongena L.) cultivars. Journal of Agricultural and Food Chemistry 51, 3448–3454.[CrossRef][ISI][Medline]
Whiting P, Favis BD, St-Germain FGT, Goring DAI. 1981. Fractional separation of middle lamella and secondary tissue from spruce wood. Journal of Wood Chemistry and Technology 1, 29–42.
Williams RJ, Spencer JP, Rice-Evans C. 2004. Flavonoids: antioxidants or signalling molecules? Free Radical Biology & Medicine 36, 838–849.
Willits MG, Kramer CM, Prata RT, De Luca V, Potter BG, Steffens JC, Graser G. 2005. Utilization of the genetic resources of wild species to create a non-transgenic high flavonoid tomato. Journal of Agricultural and Food Chemistry 53, 1231–1236.[CrossRef][ISI][Medline]
Yu O, Jung W, Shi J, Croes RA, Fader GM, McGonigle B, Odell JT. 2000. Production of the isoflavones genistein and daidzein in non-legume dicot and monocot tissues. Plant Physiology 124, 781–794.