JXB Advance Access originally published online on July 26, 2006
Journal of Experimental Botany 2006 57(12):3007-3018; doi:10.1093/jxb/erl061
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2006 The Author(s).
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.This paper is available online free of all access charges (see http://jxb.oxfordjournals.org/open_access.html for further details)
RESEARCH PAPER |
Overexpression of a bacterial 1-deoxy-D-xylulose 5-phosphate synthase gene in potato tubers perturbs the isoprenoid metabolic network: implications for the control of the tuber life cycle
1Quality, Health and Nutrition, Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, UK
2Rothamsted Research, Harpenden, Hertfordshire AL5 2JQ, UK
3Gene Expression, Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, UK
To whom correspondence should be addressed. E-mail: mtaylo{at}scri.sari.ac.uk
Received 11 January 2006; Accepted 25 May 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Potato tubers were engineered to express a bacterial gene encoding 1-deoxy-D-xylulose 5-phosphate synthase (DXS) in order to investigate the effects of perturbation of isoprenoid biosynthesis. Twenty-four independent transgenic lines out of 38 generated produced tubers with significantly elongated shape that also exhibited an early tuber sprouting phenotype. Expression analysis of nine transgenic lines (four exhibiting the phenotype and five showing a wild-type phenotype) demonstrated that the phenotype was strongly associated with dxs expression. At harvest, apical bud growth had already commenced in dxs-expressing tubers whereas in control lines no bud growth was evident until dormancy was released after 56–70 d of storage. The initial phase of bud growth in dxs tubers was followed by a lag period of
56 d, before further elongation of the developing sprouts could be detected. Thus dxs expression results in the separation of distinct phases in the dormancy and sprouting processes. In order to account for the sprouting phenotype, the levels of plastid-derived isoprenoid growth regulators were measured in transgenic and control tubers. The major difference measured was an increase in the level of trans-zeatin riboside in tubers at harvest expressing dxs. Additionally, compared with controls, in some dxs-expressing lines, tuber carotenoid content increased 2-fold, with most of the increase accounted for by a 6–7-fold increase in phytoene. Key words: Cytokinin, 1-deoxy-D-xylulose 5-phosphate synthase, dormancy, isoprenoid, potato, sprouting
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Isoprenoid biosynthetic pathways provide a wide range of metabolites that are essential for both plant development and storage organ food quality. Over 22 000 different isoprenoids have been identified in plant species (Harborne, 1991), forming a structurally and functionally diverse group of metabolites. Examples include the carotenoids which act as photosynthetic pigments and are important micronutrients in plant-derived food; sterols, essential for membrane function; tocopherols and tocotrienols (vitamin E); chlorophylls that contain a C20 isoprenoid side chain; the isoprenoid-derived phytohormones, gibberellins, brassinosteroids, cytokinins, and abscisic acid (ABA); and monoterpenes, sesquiterpenes, and diterpenes involved in plant defence, aroma, and flavour.
There has been a rapid escalation in our knowledge of the isoprenoid biosynthetic pathways, particularly at the molecular level (see, for example, Lange and Ghassemian, 2003; Laule et al., 2003). Isoprenoids are synthesized from the five-carbon intermediates isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP). In plants, there are two distinct pathways for the synthesis of IPP. In one pathway, localized in the cytosol/endoplasmic reticulum, IPP is synthesized from acetyl-CoA via mevalonic acid (McGarvey and Croteau, 1995). IPP is then converted to a range of isoprenoid products, most notably phytosterols, ubiquinone, and the sesquiterpenes. More recently, the plastidic pathway of isoprenoid biosynthesis has been identified and characterized (Rohmer et al., 1993; Arigoni et al., 1997; Rodriguez-Concepcion and Boronat, 2002; Fig. 1). In this pathway, the five-carbon building blocks of isoprenoid metabolism, IPP and DMAPP, are synthesized from glyceraldehyde-3-phosphate and pyruvate via the methylerythritol phosphate (MEP) pathway (also known as the deoxy-D-xylulose 5-phosphate or the mevalonate-independent pathway). 1-Deoxy-D-xylulose 5-phosphate (DXP), the product of the initial reaction, is the precursor of thiamine and pyridoxol as well as IPP. In many plant species there are two distantly related genes encoding 1-deoxy-D-xylulose 5-phosphate synthase (DXS; Walter et al., 2002). On the basis of their different expression patterns, it has been speculated that DXS1 and DXS2 may encode isoforms with roles in the biosynthesis of different classes of isoprenoid.
|
The requirement for a functional MEP pathway was demonstrated by the phenotype of cla1 mutants of Arabidopsis, in which the DXS gene is disrupted (Mandel et al., 1996; Estevez et al., 2000). In these mutants, chloroplast development is arrested at an early stage, and the low levels of carotenoid and chlorophyll produced give the plants an albino appearance. Constitutive overexpression of the endogenous DXS gene in Arabidopsis (Estevez et al., 2001) results in increases in the levels of many plastidic isoprenoids when compared with wild-type levels, including chlorophyll (134–142%),
-tocopherol (154–215%), carotenoids (112–131%), and ABA (295–397%). Despite these changes in isoprenoid levels, there were only slight changes in growth and germination rates. However, these experiments demonstrated that the IPP pool is rate limiting for isoprenoid biosynthesis. Based on the expression pattern of DXS and the results of 1-deoxy-D-xylulose feeding experiments, it was concluded that, in ripening tomato fruit, DXS activity is rate limiting for carotenoid biosynthesis (Lois et al., 2000). Transgenic tomato plants expressing a bacterial dxs gene driven by a fruit-specific promoter confirmed that it was possible to obtain fruit with elevated carotenoid levels when DXS activity was increased. An increase in total fruit carotenoid level of up to 1.6-fold was measured, with phytoene and ß-carotene increases accounting for the increased total level (Enfissi et al., 2005). In this study, changes in plastidic isoprenoids other than carotenoids were not reported and no effects of elevated DXS activity on plant development were observed. Although carotenoid levels in potato tubers are generally much lower than in tomato fruit, potato germplasm has been identified which contains nutritionally significant levels of certain carotenoids (Nesterenko and Sink, 2003; Morris et al., 2004). Transgenic manipulation of potato has also been used successfully to elevate carotenoid levels and increase the spectrum of carotenoids that accumulate to significant levels (Römer et al., 2002; Ducreux et al., 2005). For example, expression of a bacterial crtB gene encoding phytoene synthase led to up to 6-fold higher carotenoid levels, and nutritionally significant levels of ß-carotene accumulated in these tubers, whereas barely detectable levels of this major provitamin A carotenoid were measured in control lines (Ducreux et al., 2005).
Plastidic isoprenoid-derived hormones (particularly gibberellins, cytokinins, and ABA) are important in the regulation of the potato tuber life cycle (reviewed in Fernie and Willmitzer, 2001). The timing of tuber initiation, establishment and maintenance of tuber dormancy, initiation of tuber sprouting, and rate of sprout growth are of biological and commercial importance. Although implicated in the control of the tuber life cycle, the precise role of the cytokinins, gibberellins, and ABA have not yet been defined fully as studies to date have largely correlated levels of the hormones with developmental events in the life cycle (Suttle, 2004a).
In this study the tuber isoprenoid metabolic network was perturbed by expressing a bacterial dxs gene specifically in the tuber. Changes in the levels of a range of tuber isoprenoids resulting from this increased activity are described. Transgene expression also corresponds with altered tuber phenotype: as well as giving rise to significantly elongated tubers, dxs expression results in changes in tuber dormancy and sprouting characteristics, demonstrating the role of the isoprenoid-derived metabolites in the control of aspects of the tuber life cycle.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Plant material
Solanum tuberosum L. (cv. Desiree) were grown from seed tubers or in vitro propagated tissue-culture plants in 30 cm diameter pots containing a compost mix of: Irish moss peat (bedding grade), 1200.0 l; Pavoir sand, 100.0 l; limestone (magnesium), 2.5 kg; limestone (calcium), 2.5 kg; Sincrostart base fertilizer, 1.5 kg (William Sinclair, Lincoln, UK); Celcote wetting agent, 1.5 kg (LBS Horticulture, Lancashire, UK); Perlite, 100.0 l; Osmocote mini controlled-release fertilizer, 2.0 kg (Scotts, Humberside, UK), and Intercept insecticide, 0.39 kg (Bayer CropScience, UK). Plants were raised in a glasshouse maintained at a daytime temperature of 20 °C and a nocturnal temperature of 15 °C. Light intensity (photosynthetic photon flux density) ranged from 400 to 1000 µmol m–2 s–1 and the mean day length was 16 h. Developing tubers were harvested after 8–10 weeks and mature tubers after the surface foliage had fully senesced, typically 16–20 weeks.
In vitro tuberization
After 28 d culture, potato plants were excised into single nodes (10 per plantlet) and transferred to 90 mm Petri dishes containing Murashige and Skoog medium (Murashige and Skoog, 1962) supplemented with 0.8% (w/v) bacto microagar (Difco), 8% (w/v) sucrose and, where included, 9.8 µM N-6-benzyladenine (BA). All dishes were sealed with Nescofilm and placed in a 16 °C incubator with an 8 h photoperiod at 80 µmol m–2 s–1 for 7 d prior to incubation at 16 °C, in total darkness, for a further 28 d.
Binary vector construction and transformation of potato
The dxs gene from Escherichia coli was kindly provided by Georg Sprenger, University of Munich, Germany, in the form of plasmid pUCBM20dxs (Sprenger et al., 1997). Polymerase chain reaction (PCR) primers were designed to amplify the dxs open reading frame in this plasmid corresponding to nucleotides 17 765–19 627 in EMBL accession number u82664, while engineering a 5' end PstI and a 3' end EcoRI restriction site (underlined) to facilitate subcloning. Primer sequences were: forward primer (5dxsEcoRI), 5'-AAGAATTCTTATGCCAGCCAGGCCTTGAT-3'; and reverse primer (3dxsPstI), 5'-AACCTGCAGATATTGCCAAATACCCGACC-3'. The resultant fragment was subcloned into pGEM-T Easy (Promega, UK) and sequenced on a 377 automated DNA sequencer (Applied Biosystems), using a cycle sequencing protocol and the BigDye Terminator Cycle Sequencing Kit (Applied Biosystems), in order to determine the authenticity of the cloned fragment. The open reading frame of the dxs gene was cloned in-frame behind the pea rbcS transit peptide in vector pJIT117 utilizing EcoRI–PstI restriction sites (Guerineau et al., 1988) to target the DXS protein to the plastid. This intermediate construct was designated pJIT117dxsEP. A 1252 bp HindIII fragment, consisting of the transit peptide and an incomplete 5' dxs gene fusion, was excised from the resultant plasmid and cloned into the HindIII site of pBluescript SK(–) so as to gain BamHI and KpnI restriction sites. For tuber-specific expression, the sense recombinant plasmid was selected (after screening by restriction digestion), digested with BamHI and KpnI, and cloned behind a patatin promoter contained in binary vector pBI140.5 [essentially pBIN19 containing the 3.5 kb class I patatin cassette from pBI240.7 (Bevan et al., 1986)]. The resulting plasmid was digested with SalI and KpnI, and the excised fragment was replaced with the SalI and KpnI fragment from pJIT117dxsEP, resulting in a binary construct containing patatin promoter–pea rbcS transit peptide–sense E. coli dxs gene–cauliflower mosaic virus (CaMV) poly(A) terminator. The binary vector was transformed into Agrobacterium tumefaciens strain LBA4404 by electroporation. Transformed Agrobacterium cells were selected by their resistance to kanamycin (100 µg ml–1) and rifampicin (100 µg ml–1). Potato transformation (S. tuberosum L. cultivar Desiree) was carried out as described previously (Ducreux et al., 2005).
Analysis of tuber carotenoids
Peeled whole tuber samples were frozen in liquid nitrogen, freeze-dried, and stored at –80 °C prior to analysis. Triplicate samples (three independent tubers) from the different transgenic and control lines were examined. Total potato tuber carotenoids were extracted and analysed by reverse-phase high-performance liquid chromatography (HPLC) essentially as detailed in Morris et al. (2004), but with the addition of a coupled Gilson 170 diode array detector.
DNA extraction and Southern analysis
Plant genomic DNA was extracted from leaves as described previously (Draper and Scott, 1988). A 10 mg aliquot of DNA was digested with KpnI, as this enzyme only cut once within the transfer DNA region, and resolved by electrophoresis on 0.8% agarose gels. DNA was transferred to nylon membranes (Hybond-N+, Amersham) as previously described (Sambrook et al., 1989). Filters were hybridized with dxs cDNA labelled to high specific activity (1x109 cpm µg–1) with [-32P]dCTP using random primers (HiPrime, Boehringer). Following hybridization, filters were washed at low stringency [0.5x standard saline citrate (SSC), 0.1% sodium dodecylsulphate (SDS) at 45 °C] and exposed to X-ray film for 24 h at –70 °C with intensifying screens.
Northern blot analysis
Northern blots were performed using total RNA extracted from developing and mature tuber samples using an RNA isolation kit (Qiagen GmbH, Hilden, Germany). A 10 mg aliquot of RNA per track was analysed by denaturing agarose gel electrophoresis. Nucleic acids were transferred to positively charged nylon membrane (Hybond-N+, Amersham Biosciences) as previously described (Sambrook et al., 1989). Filters were probed with random-primed (HiPrime, Boehringer) [-32P]dCTP-labelled DNA for 16 h at 42 °C in formamide hybridization buffer (Sambrook et al., 1989). A potato 18S rRNA probe was used as a control. Filters were washed in low and high stringency SSC/SDS solution at 42 °C until an acceptable signal to noise ratio was achieved. Relative gene expression was determined by autoradiography.
Quantitative RT–PCR analysis
Total RNA (10 µg) was treated with DNase I (Ambion Inc., Austin, TX, USA) before undergoing reverse transcription, using random hexamers as primer and SuperScriptTM II reverse transcriptase (Invitrogen Life Technologies, Carlsbad, CA, USA) to generate the first-strand cDNA template. Samples (1 µl of a 10-fold dilution) were amplified in 25 µl using a Perkin Elmer ABI Prism 7700 sequence detector in conjunction with the Quantitect SYBR green PCR kit (Qiagen GmbH, Hilden, Germany). Thermal cycling conditions were: 15 min denaturation at 95 °C followed by 40 cycles (15 s at 95 °C, 30 s at 58 °C, 30 s at 72 °C). Relative expression levels were calculated and the primers validated using the 
Ct method (Livak, 1997) using data obtained with the ubiquitin-specific primers as an internal control. Primer sequences were: PSYF, 5'-TGGCTTCAATCTCATCGAGAATC-3'; PSYR, 5'-AGGAGTGACAGAATTAAGCGCAG-3'; PDSF, 5'-GCCGAGATGGTTGCTTGC-3'; PDSR, 5'-CGAAGCTTTAACTTATGACCCATTG-3'; UBIF, 5'-ACACCATTGATAATGTCAAGGCTAAG-3'; and UBIR, 5'-GCCATCCTCCAATTGCTTTC-3'.
Phenotypic analysis
In order to gain a statistically significant measure of length to width ratios, five mature tubers from 10 individual plants were examined. For rate of sprout growth experiments, the first three sprouts from 12 individual tubers were measured for each line under investigation.
Gibberellin analysis
Peeled whole tuber samples (mature tubers at harvest) were frozen in liquid nitrogen, freeze-dried, milled to a fine powder, and stored at –80 °C prior to analysis. Triplicate samples (three independent tubers) from the different transgenic and control lines were analysed. Samples (500 mg dry weight) were extracted in the presence of 2H-labelled internal standards (5 ng) and 3H-labelled gibberellins GA1, GA19, and GA20 (833 Bq each), as markers, at 4 °C overnight in 80 ml of 80% (v/v) aqueous methanol. After filtration, the residue was re-extracted with 50 ml of 80% (v/v) aqueous methanol for 2 h and refiltered, and the extracts were combined and concentrated to 5 ml in vacuo at 35 °C. The pH of this solution was adjusted to 8.0 prior to loading onto a 5 ml bed volume QAE Sephadex A-25 anion exchange column (Pharmacia) pre-equilibrated with water at pH 8.0 (6x5 ml). After loading, the column was washed with 3x5 ml water at pH 8.0 and gibberellins were eluted with 20 ml of 0.2 M formic acid directly onto a C18 bond elute cartridge (500 mg; Varian) pre-equilibrated with 100% methanol (5 ml) and water at pH 3.0 (5 ml). Gibberellins were eluted from the C18 cartridge with methanol (5 ml) and the eluates taken to dryness in vacuo. Samples were methylated with ethereal diazomethane and evaporated to dryness under a gentle stream of nitrogen. Methylated extracts were redissolved in 1 ml of ethyl acetate and twice partitioned against 1 ml of water. The ethyl acetate phases were applied to a Bond Elut aminopropyl ion exchange cartridge (100 mg; Varian) pre-equilibrated with ethyl acetate (2 ml). The ethyl acetate eluates were evaporated to dryness in vacuo, and the gibberellin methyl esters resolved by C18 reverse phase HPLC (Croker et al., 1999). Quantitative analysis of the pooled fractions by gas chromatography–mass spectrometry (GC–MS) of the trimethylsilyl ethers was as previously described (Coles et al., 1999), except that peak areas were obtained from full scans.
Cytokinin measurements
Peeled whole tuber samples (2.0 g fresh weight, mature tubers at harvest or tubers stored for 10 weeks at 4 °C in darkness) were frozen in liquid nitrogen and mechanically homogenized in the presence of 10 vols of 80% (v/v) aqueous ethanol. The homogenates were left to stand at –20 °C overnight prior to filtration through a 0.45 µm acrodisc filter (Pall, Farlington, UK). The clarified homogenates were concentrated to the aqueous phase by rotary flash evaporation (RFE) at 35°C. The pH of this solution was adjusted to 6.5 and it was applied to a pre-equilibrated [first with methanol then with 10 mM ammonium acetate (pH 6.5)] Strata-X 33 µm polymeric sorbent column (2 g of column packing g–1 tissue fresh weight; Phenomenex, Macclesfield, UK). After loading, the column was washed with 10 vols of ammonium acetate buffer (pH 6.5). Cytokinins were eluted with 10 vols of 60% (v/v) aqueous methanol, transferred to round-bottomed flasks, and concentrated to the aqueous phase by RFE at 35 °C. Samples were then frozen at –80 °C and freeze-dried. Dried cytokinin extracts were reconstituted in TRIS-buffered saline. The cytokinins, trans-zeatin riboside (tZR) and isopentenyl adenosine (IPA), were quantified using plant growth regulator detection kits (Sigma, UK) using extracts diluted 10-fold in TRIS-buffered saline. Cytokinin standards tZR and IPA were purchased from Sigma (Poole, UK).
Abscisic acid quantification
Peeled whole tuber samples (pooled samples of three tubers) were freeze-dried and stored at –80°C prior to analysis. Triplicate samples from the different transgenic and control lines were analysed. Approximately 0.2 g of developing tuber material was analysed for each replicate. Samples were extracted in 2 ml of 80% acetone and were treated according to Artsaenko et al. (1995). ABA was quantified using a Phytodetek-ABA kit (Agdia, Elkhart, IN, USA) using extracts diluted 1:10 in TRIS-buffered saline.
Statistical analysis
The Student's t-test method was used to test the statistical relationship between control and transgenic lines for growth characteristics, hormone concentrations, and carotenoid levels. Analysis of variance was used to compare genotypes, the effect of BA, and their interaction for the in vitro tuberization assay. A log transformation was used for the stolon length data to achieve equal variances, but neither percentage score needed any transformation.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Generation of transgenic potato plants expressing a bacterial dxs gene in the tuber
In order to perturb plastidic isoprenoid metabolism, the dxs gene from E. coli was expressed specifically in the potato tuber. The patatin promoter was chosen as a well-characterized tuber-specific promoter (Verhees et al., 2002) and the dxs open reading frame was fused to the pea Rubisco targeting sequence (Guerineau et al., 1988) to ensure the DXS product was directed to the tuber plastid (Fig. 2A).
|
Changes in tuber phenotype associated with up-regulated dxs expression
A total of 38 independent dxs transgenic lines were generated and grown to maturity in a glasshouse, alongside five empty vector control lines. At harvest, the shape of the tubers from 24 of the 38 dxs transgenic lines was markedly different from that of empty vector and wild-type controls (typical examples shown in Fig. 3A). Measurement of the tuber length to width ratio (Fig. 3B) quantified the difference in shape apparent in Fig. 3A and demonstrated that the change in tuber shape was statistically significant (P <0.05). In addition to the elongated shape, it was also observed that in tubers from the same 24 transgenic lines, sprout growth had already commenced at harvest, and the apical bud of the tubers had grown to a length of
1–2 mm. In control tubers, grown under identical conditions, the tuber apical meristem was fully dormant as expected for this cultivar (Fig. 3C). Thus it appears that in the dxs overexpressing lines, the tubers had already emerged from dormancy at harvest. The dxs expression level was determined in developing tubers from four lines that exhibited similar tuber shape and sprouting phenotypes (even in line 3 where dxs expression was highest) and compared with the level in five lines that did not show the phenotype. For all lines, the occurrence of the phenotype was associated with dxs expression and, in tubers with a wild-type phenotype, no dxs expression could be detected by reverse transcription–PCR (RT–PCR) using ubiquitin primers as a control (Fig. 2B, C). In two lines that exhibited the phenotype, RNA blot analysis demonstrated that expression was higher in developing tubers than in mature tubers (Fig. 2D), consistent with the activity of the patatin promoter (Verhees et al., 2002). This phenotype was maintained through at least two subsequent cycles of vegetative propagation, in which plants were grown from tubers. As the phenotype was robustly associated with dxs expression, two lines that exhibited the phenotype (dxs1 and dxs2) and two that did not (dxsne1 and dxsne2) were selected for detailed analysis using plants grown from tubers. Southern analysis confirmed that the dxs gene had been successfully transformed into potato, with a wide range in the number of transgene copies, varying from 12 (line dxs1) to one (line dxsne1, data not shown).
|
The sprouting behaviour of the dxs-expressing tubers was compared with that of control tubers in more detail. Mature tubers were harvested from dxs-expressing lines and controls and immediately replanted in compost and grown to maturity in a glasshouse. Sprouts from lines overexpressing dxs emerged 20–30 d earlier than empty vector or non-expressing tubers (Table 1). Additionally the time to tuber initiation was reduced in the dxs-expressing lines by
30 d (Table 1). No significant differences in plant height or tuber yield were measured between control and dxs-expressing tubers (data not shown). Further experiments compared the rate of sprout growth in the dxs-expressing tubers with control tubers (wild type, empty vector, and a non-expressing line dxsne2, Fig. 4). Prior to the experiment, the control tubers were stored at 4 °C for 120 d and had started to sprout. These control tubers, in which visible sprout growth had commenced and possessed sprouts of 1–2 mm in length, were compared with dxs-expressing tubers, freshly harvested but also having sprouts of 1–2 mm. All tubers were placed in a growth cabinet maintained at 25 °C and 90% relative humidity, conditions optimal for sprout growth (Wiltshire and Cobb, 1996). Sprouts from the different control tubers all continued to grow under these conditions. For the dxs-overexpressing tubers, however, there was a significant lag (of 56 d) before any increase in sprout length could be detected (Fig. 4). Following the lag period, the mean rate of sprout growth was considerably lower than the initial rate of control tuber sprout growth (Fig. 4).
|
|
Levels of the plastid-derived isoprenoid growth regulators in dxs-expressing tubers
In order to account for the early-sprouting phenotype evident on overexpression of the dxs gene, the levels of the major plastidic isoprenoid growth regulators were compared in dxs-expressing tubers and controls. As gibberellins, some cytokinins, and ABA are all plastid isoprenoids, it was considered possible that overexpression of dxs could have an effect on their levels. Expression of the dxs transgene was driven by a patatin promoter, active in most major tuber cell types (Verhees et al., 2002), and so the levels of the growth regulators were measured in samples from whole tubers.
Gibberellins were quantified using GC–MS with appropriate internal standards. The levels of gibberellins in mature tuber tissues at harvest were generally very low and for most gibberellins (GA3, GA4, GA8, GA19, and GA20) were below the detection limit of the system. As with previous studies, the major active gibberellin detected was GA1, a product of the 13-hydroxylation pathway (Suttle, 2004b). Although GA4 has been detected in developing tubers (Xu et al., 1998), this gibberellin was not present in mature tubers (Suttle, 2004b). The level of GA29, an inactive metabolite of GA20, was significantly (P <0.05) higher in non-transformed Desiree mature tubers compared with tubers from the dxs-expressing line dxs2 (Table 2); however, the levels of GA1 were not significantly different in the transgenic and control lines at the P=0.05 level. As the tuber shape, dormancy, and sprouting characteristics and transgene expression levels were very similar for tubers from lines dxs1 and dxs2, and the gibberellin levels were similar in three biological replicate tuber samples from Dxs2, this analysis was not extended to other DXS transgenic lines.
|
In order to compare cytokinin levels in dxs-expressing tubers and controls, the levels of tZR and IPA were quantified in extracts of mature tubers using a competitive enzyme-linked immunosorbent assay (ELISA). Although up to eight different cytokinins have been detected in potato tubers, the choice was made to focus on tZR and IPA as previously large increases in the levels of these cytokinins had been measured in potato tubers associated with the release from dormancy (Suttle, 1998). Furthermore, injection of these cytokinins into dormant tubers resulted in the rapid and complete termination of tuber endodormancy (Suttle, 1998). A third consideration was the recent finding that the prenyl group of tZR and IPA is mainly produced through the MEP pathway (Kasahara et al., 2004). In mature tubers from dxs1, the level of tZR at harvest was
6-fold higher than in Desiree controls, whereas the level of IPA in dxs1 tubers was not significantly (at the P=0.05 level) different from that measured in control tubers (Table 3). In tubers stored in darkness for 10 weeks at 4 °C, however, there was no significant difference in the levels of tZr or IPA between dxs-expressing lines dxs1 and dxs2, and controls.
|
ABA levels were also compared in mature dxs-expressing tubers and controls using a competitive ELISA assay. In mature tubers from empty vector lines, ABA levels were similar to those previously reported for potato at harvest (Biemelt et al., 2000; Table 4). A similar level was also measured in tubers from the Southern-positive non-expressing line dxsne1. Although the mean ABA levels in mature tubers at harvest from both the dxs-expressing lines were lower than in controls, t test analysis showed that these differences were not significant at the P=0.05 level (Table 4).
|
In vitro tuberization of dxs-expressing lines compared with controls
In view of the changes in tuber sprouting characteristics observed on overexpression of the bacterial dxs gene (Table 1; Figs 3, 4), the behaviour of the dxs-expressing lines was compared with that of controls in an in vitro microtuberization system. There was considerable variation in the degree of tuberization and stolon length observed in the microtuber system employed, and so a large number (120) of explants were compared for each treatment. Microtubers were produced from nodal cuttings using a standard in vitro tuberization protocol (Faivre-Rampant et al., 2004) omitting the gibberellin inhibitor chlorocholine chloride. Microtubers were produced on growth media in the presence or absence of the synthetic cytokinin BA, added to in vitro tuberization media to accelerate tuberization (Galis et al., 1995). The most notable effect of BA was the reduction in mean stolon length in wild-type Desiree, dxs1, and dxsne2 (Table 5). For dxs1, in the absence of BA, stolons were
5-fold shorter than in both the dxsne2 (P=6.8x10–5) and wild-type explants (P=1.2x10–4). In the presence of BA, stolon length decreased for explants of all three types, the length of the stolons from dxsne2 and wild-type now not being significantly different from the length of stolons from dxs1 in the absence of BA. The inclusion of BA in the tuberization media resulted in an increase in the percentage of nodes that tuberized, for example for dxsne2 the percentage increased from 42±6 to 65±8% (P=0.025), similar to the degree of tuberization observed for dxs1 in the absence of BA (75±5%). BA treatment also resulted in a decrease in the percentage of sessile tubers that formed. The percentage of sessile tubers that formed in wild-type explants (78±6%) was not significantly different (at the P=0.05 level) from that formed in dxsne2 explants in the absence of BA. Overall, the in vitro tuberization characteristics of dxs1 explants in the absence of BA were similar to that observed for wild-type Desiree and non-expressing dxsne2 explants cultured in the presence of BA, as evident in Fig. 5.
|
|
Increases in tuber carotenoid levels associated with dxs overexpression
In tomato fruit, DXS activity has been shown to be rate- limiting for carotenoid biosynthesis (Lois et al., 2000), and transgenic up-regulation of bacterial DXS activity can result in higher (up to 1.6-fold) fruit carotenoid accumulation (Enfissi et al., 2005). Carotenoid levels were measured in developing and mature tubers from dxs transgenic lines and compared with values for controls (empty vector transformants and wild-type Desiree; Table 6). In control tubers, carotenoid levels were similar to those previously measured for Desiree (Morris et al., 2004). In the transgenic dxs lines, where relatively high levels of dxs transcript were measured (lines dxs1 and dxs2), there was an increase in the mean tuber carotenoid content of
2-fold in both developing and mature tubers; however, these values were only significant at the P <0.1 level for the dxs1 developing and mature tubers. At both developmental stages there was an 7-fold increase in the level of phytoene, from 0.4 µg g–1 dry weight (DW) in controls to 3.0 µg g–1 DW in developing tubers of lines dxs1 and dxs2. In the dxs-expressing transgenic tubers, phytoene was the major carotenoid, and the increase in total carotenoid levels was largely attributed to the increase in phytoene level. The levels of the other major carotenoids were not significantly different in dxs-expressing lines compared with controls (P >0.1). The carotenoid levels in the Southern-positive lines in which the dxs transcript could not be detected (e.g. line dxs ne1) were similar to those in wild type and empty vector controls (data not shown), indicating that the changes in carotenoid levels were due to dxs overexpression.
|
Changes in the expression levels of phytoene synthase and phytoene desaturase in dxs-up-regulated tubers
Phytoene synthase activity is generally regarded as the rate-limiting step for carotenoid biosynthesis in plant storage organs (Fraser and Bramley, 2004). In the case of the dxs-expressing transgenic tubers, however, there was a large accumulation of phytoene, with little change in the downstream carotenoids. This result implies that the rate-limiting step in carotenoid biosynthesis had shifted from phytoene synthase to phytoene desaturase in the dxs-expressing tubers. In order to investigate this possibility further, the expression levels of phytoene synthase and phytoene desaturase were compared in tubers from the dxs-expressing lines. Quantitative RT–PCR analysis (Fig. 6A) demonstrated that there was a large increase (
15-fold) in the steady-state phytoene synthase expression level in mature tubers of dxs-expressing lines (dxs1 and dxs2) compared with empty vector-transformed and wild-type tubers, whereas in a non-expressing dxs line (dxsne1) there was no significant difference from control values. There was no significant difference in the phytoene desaturase transcript level between the control and transgenic lines tested (Fig. 6B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
In this study it is demonstrated that overexpression of a bacterial dxs gene in potato tubers results in large increases in downstream isoprenoids. The perturbation of the isoprenoid biosynthetic pathway in the tuber gives new insights into how this pathway is regulated and the role of isoprenoid-derived plant growth regulators in the control of the tuber life cycle. Previous studies have demonstrated that DXS is rate-limiting for isoprenoid biosynthesis in Arabidopsis (Estevez et al., 2001) and tomato fruit (Enfissi et al., 2005). Thus constitutive overexpression of the endogenous DXS gene (Arabidopsis) or fruit-specific expression of a bacterial dxs gene (tomato) resulted in increased isoprenoid accumulation. In Arabidopsis, overexpression of the endogenous DXS gene, to levels 1.7-fold higher than in the wild type, resulted in relatively modest changes in levels of the plastidic isoprenoids with only small effects on plant development. The tomato fruit study focused on the effects on carotenoids, and showed that the levels of phytoene and ß-carotene increased, with the largest increase (2.4-fold) in phytoene content. The development of the transgenic plants was normal, and no changes in the isoprenoid-derived growth regulators were measured. In the current study, the major change in carotenoid content was also in phytoene, with a 7.5-fold increase measured in the developing tuber. There were no significant changes in carotenoids downstream of phytoene. Measurements of the transcript levels of phytoene synthase and phytoene desaturase indicate that in the dxs-expressing lines the phytoene synthase transcript level increases
15-fold whereas the phytoene desaturase transcript level is unchanged in the dxs-expressing lines compared with controls. The factors that regulate the transcript levels of these genes are unknown, although it has been suggested that the phytoene desaturase gene (PDS) is feedback regulated at the transcriptional level by the accumulation of the end-products of the carotenogenic pathway (Rodermel, 2001). In this study, however, there was no change in PDS transcript level despite the relatively large accumulation of phytoene, whereas there is a large increase in phytoene synthase transcript level. These data would suggest that phytoene desaturase activity becomes rate-limiting when phytoene synthase transcript levels increase, resulting in the accumulation of phytoene. The mechanism that results in the large increase in phytoene synthase transcript level is unknown, but illustrates that feed-forward mechanisms operate to regulate this pathway. A strategy for increasing the levels of carotenoids downstream of phytoene would be to co-transform a bacterial phytoene desaturase gene with the dxs gene, potentially overcoming this metabolic bottleneck and increasing the levels of more nutritionally important carotenoids such as ß-carotene. Such a strategy proved successful in transgenic canola seeds, where co-expression of a bacterial crtI gene (encoding phytoene desaturase) with crtB resulted in an increased seed lycopene content and reduced phytoene content, compared with the crtB transformants (Ravanello et al., 2003). In the study of Lois et al. (2000), direct injection of 1-deoxy-D-xylulose into tomato fruit also resulted in an increase in the transcript level of the gene encoding phytoene synthase. It is possible that the accumulation of 1-deoxy-D-xylulose 5-phosphate, in both the current study and in tomato fruits, results in the increased expression of phytoene synthase. In potato tubers, isoprenoid metabolites are important not only for their nutritional value (e.g. carotenoids), but also because the plastidic isoprenoid-derived plant growth regulators (e.g. cytokinins, ABA, and gibberellins), in addition to ethylene, are important triggers of tuber development (Fernie and Willmitzer, 2001; Suttle, 2004a). Control of the potato tuber life cycle including the processes of tuber initiation, tuber dormancy, and sprouting is important for the commercial exploitation of the potato crop. Suttle (2004a) summarized the state of knowledge regarding the role of plant growth regulators in tuber dormancy and sprouting. ABA and ethylene are required for the initiation of tuber dormancy, but only ABA is required to maintain dormancy. In the early stages of dormancy, cytokinin levels are low and fully dormant tubers are insensitive to exogenous application of cytokinin. As dormancy progresses, tuber ABA levels decline and tubers become more sensitive to exogenous cytokinin. An increase in the level of endogenous cytokinin precedes or coincides with the release from dormancy and sprout growth. Sprout growth is accompanied by an increase in the levels of indole-3-acetic acid (IAA) and gibberellin. Thus the concerted action of several plant growth regulators is involved in the regulation of tuber dormancy and sprouting. Although the role of the plant growth regulators in the control of the potato tuber life cycle is beginning to be understood, few studies have used forward or reverse genetic approaches to manipulate these processes or define the fine detail of the key regulatory steps. Recent attempts to manipulate potato cytokinin content have resulted in tuber life cycle phenotypes (Zubko et al., 2005). However, pleiotropic effects, particularly on root development, have made clear interpretation of these experiments difficult.
In view of the early sprouting phenotype observed on overexpression of dxs, it was of interest to determine whether changes in the plastidic isoprenoid-derived plant growth regulators were evident in these tubers. The most striking difference was in the elevated levels of cytokinins (up to 6-fold for tZR) observed in tubers from a dxs-overexpressing line at harvest. The isoprenoid precursor DMAPP is a precursor of cytokinin biosynthesis, acting as a substrate for ATP/ADP isopentenyltransferase, demonstrated to be the rate-limiting step of cytokinin biosynthesis in Arabidopsis (Kakimoto, 2003). In Arabidopsis seedlings, the prenyl group of tZR and IPA is mainly produced through the MEP pathway (Kasahara et al., 2004) and so the large increase in tZR observed in the dxs-overexpressing tubers may indicate that in potato tubers plastidic DMAPP availability is limiting for cytokinin biosynthesis. An alternative cytokinin biosynthetic pathway has been proposed in which 1-hydroxy-2-methyl-2-(E)-butenyl 4-diphosphate (HMBPP) is the side chain donor (Astot et al., 2000; Krall et al., 2002). As HMBPP is an intermediate in the MEP pathway, it is possible that both HMBPP and DMAPP contribute to the accumulation of tZR in the dxs-overexpressing tubers. Further indirect evidence that cytokinin levels were perturbed in the dxs-overexpressing lines came from an investigation of in vitro tuberization. The mean stolon length, degree of tuberization, and percentage of sessile tubers were similar for explants from dxs-overexpressing lines in the absence of cytokinin to controls supplemented with cytokinin. Although higher tZR levels were measured in tubers from a dxs-expressing line at harvest, no differences in tZR levels were observed after 10 weeks storage at 4 °C, emphasizing the complexity of the regulatory processes and making our conclusions more tentative. However, further support for the hypothesis that the dxs phenotype is due to the enhanced tZR levels measured in these tubers at harvest is given by the absence of significant changes in the other isoprenoid-derived phytohormones. ABA levels were not significantly changed, and the only significant difference in gibberellin content was a decrease in the level of the inactive GA29 in dxs2 tubers compared with the wild type. In fact, the only active gibberellin that could be detected was GA1, and its level did not change in the dxs-expressing line, compared with controls. Generally, regulation of the gibberellin precursor ent-kaurene synthesis is considered to reside with ent-copalyl diphosphate synthase activity (Hedden and Phillips, 2000), and so perturbations in DXS activity are considered unlikely to affect gibberellin levels. However, the possibility cannot be excluded that changes in other active gibberellins, beneath the detection limit of our instrumentation, could be occurring as a result of dxs overexpression, adding a caveat to the interpretation of the data presented here.
Previous transgenic manipulations of potato have resulted in increased tuber content of gibberellins and an early sprouting phenotype (Carrera et al., 2000; Bachem et al., 2001), similar in some respects to the phenotype reported here for dxs overexpression. Additionally, in the study of Bachem et al. (2001), transgenic tubers, containing elevated levels of gibberellins, also had an elongated shape. Although gibberellins undoubtedly have important roles in the regulation of the tuber life cycle, Suttle (2004b) concluded that their role was probably in the regulation of sprout growth following dormancy release.
Although the dxs-overexpressing tubers exhibited premature sprout growth, producing sprouts of 2 mm length at harvest, there was a marked lag in the further growth of these sprouts, even under optimal conditions. Thus it appears that although these tubers had exited dormancy, sprout growth was retarded. We presume that sprout growth was arrested until the hormonal or metabolic balance within the tuber was appropriate for sprout growth. Other studies have demonstrated that potato tuber sprouting can be controlled by manipulation of carbohydrate metabolism (reviewed in Sonnewald, 2001). Tuber sprout growth is initially supported by energy captured from sucrose breakdown. As inorganic pyrophosphate is a necessary cofactor for sucrose breakdown, removal of pyrophosphate by expression of a bacterial pyrophosphatase in transgenic tubers increases sucrose content and prevents its use as an energy supplier. Consequently, sprout growth is significantly inhibited when sucrose is limited, but accelerated when sucrose supply is increased. The sprouting phenotype observed in the dxs-overexpressing tubers may be a result of limitation in the availability of sucrose for sprout growth. The extensive separation of the processes of dormancy release and subsequent sprout growth exhibited by the dxs-overexpressing tubers, associated with a large increase in the level of tZR, provides novel mechanistic information about dormancy release and sprouting in potato tubers. It will be of interest to determine whether this mechanism is operative in other plant meristem activation systems. Additionally, the separation of dormancy and sprouting will enable the application of transcript and metabolite profiling technologies to add details of the molecular mechanisms involved in these processes.
|
Acknowledgements |
|---|
This work was funded by the Scottish Executive Environment and Rural Affairs Department, and the Biotechnology and Biological Sciences Research Council of the UK. We are grateful to Paul Hopkins for help with the analysis of gibberellins, Georg Sprenger, Jülich, Germany for providing the E. coli dxs gene, Dr Rob Hancock, SCRI for critically reading the manuscript, and Dr Jim McNicol, BioSS, for statistical advice.
|
Footnotes |
|---|
* Present address: Daniel Rutherford Building, Institute of Molecular Plant Sciences, Kings Buildings, University of Edinburgh, Edinburgh EH9 3JR, UK.
|
Abbreviations |
|---|
ABA, abscisic acid; BA, N-6-benzyladenine; DMAPP, dimethylallyl diphosphate; DXS, 1-deoxy-D-xylulose 5-phosphate synthase; ELISA, enzyme-linked immunosorbent assay; HMBPP, 1-hydroxy-2-methyl-2-(E)-butenyl 4-diphosphate; IPA, isopentenyl adenosine; IPP, isopentenyl diphosphate; MEP, methylerythritol phosphate; tZR, trans-zeatin riboside.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Arigoni D, Sagner S, Latzel C, Eisenreich W, Bacher A, Zenk MH. (1997) Terpenoid biosynthesis from 1-deoxy-D-xylulose in higher plants by intramolecular skeletal rearrangement. Proceedings of the National Academy of Sciences, USA 94:10600–10605.
Artsaenko O, Peisker M, zur Nieden U, Fiedler U, Weiler EW, Muntz K, Conrad U. (1995) Expression of a single-chain Fv antibody against abscisic acid creates a wilty phenotype in transgenic tobacco. The Plant Journal 8:745–750.[CrossRef][ISI][Medline]
Astot G, Dolezal K, Nordstrom A, Wang Q, Kunkels T, Moritz T, Chua N-H, Sandberg G. (2000) An alternative cytokinin biosynthesis pathway. Proceedings of the National Academy of Sciences, USA 97:14778–14783.
Bachem CW, Horvath B, Trindade L, Claassens M, Davelaar E, Jordi W, Visser RG. (2001) A potato tuber-expressed mRNA with homology to steroid dehydrogenases affects gibberellin levels and plant development. The Plant Journal 25:595–604.[CrossRef][ISI][Medline]
Bevan M, Barker R, Goldsbrough A, Jarvis M, Kavanagh T, Iturriaga G. (1986) The structure and transcription start site of a major potato tuber protein gene. Nucleic Acids Research 14:4625–4638.
Biemelt S, Hajirezaei M, Sonnewald U. (2000) Comparative analysis of abscisic acid content and starch degradation during storage of tubers harvested from different potato varieties. Potato Research 43:371–382.[CrossRef]
Carrera E, Bou J, Garcia-Martinez JL, Prat S. (2000) Changes in GA 20-oxidase gene expression strongly affect stem length, tuber induction and tuber yield of potato plants. The Plant Journal 22:247–256.[CrossRef][ISI][Medline]
Coles JP, Phillips AL, Croker SJ, Garcia-Lepe R, Lewis MJ, Hedden P. (1999) Modification of gibberellin production and plant development in Arabidopsis by sense and antisense expression of gibberellin 20-oxidase genes. The Plant Journal 17:547–556.[CrossRef][ISI][Medline]
Croker SJ, Hedden P, Lenton JR, Stoddart JL. (1990) Comparison of gibberellins in normal and slender barley seedlings. Plant Physiology 94:194–200.
Draper J and Scott RJ. (1988) The isolation of plant nucleic acids. In Draper J, Scott RJ, Armitage P, Walden R (Eds.). Plant genetic transformation and gene expression: a laboratory manual (Blackwell Scientific Publications, Oxford, UK) pp. 199–236.
Ducreux LJ, Morris WL, Hedley PE, Shepherd T, Davies HV, Millam S, Taylor MA. (2005) Metabolic engineering of high carotenoid potato tubers containing enhanced levels of beta-carotene and lutein. Journal of Experimental Botany 56:81–89.
Enfissi EM, Fraser PD, Lois LM, Boronat A, Schuch W, Bramley P. (2005) Metabolic engineering of the mevalonate and non-mevalonate isopentenyl diphosphate-forming pathways for the production of health-promoting isoprenoids in tomato. Plant Biotechnology Journal 3:17–27.
Estevez JM, Cantero A, Reindl A, Reichler S, Leon P. (2001) 1-Deoxy-D-xylulose-5-phosphate synthase, a limiting enzyme for plastidic isoprenoid biosynthesis in plants. Journal of Biological Chemistry 276:22901–22909.
Estevez JM, Cantero A, Romero C, Kawaide H, Jimenez LF, Kuzuyama T, Seto H, Kamiya Y, Leon P. (2000) Analysis of the expression of CLA1, a gene that encodes the 1-deoxy-D-xylulose 5-phosphate synthase of the 2-C-methyl-D-erythritol-4-phosphate pathway in Arabidopsis. Plant Physiology 124:95–104.
Faivre-Rampant O, Gilroy EM, Hrubikova K, Hein I, Millam S, Loake GJ, Birch P, Taylor M, Lacomme C. (2004) Potato virus X-induced gene silencing in leaves and tubers of potato. Plant Physiology 134:1308–1316.
Fernie AR and Willmitzer L. (2001) Molecular and biochemical triggers of potato tuber development. Plant Physiology 127:1459–1465.
Fraser PD and Bramley PM. (2004) The biosynthesis and nutritional uses of carotenoids. Progress in Lipid Research 43:228–265.[CrossRef][ISI][Medline]
Galis I, Macas J, Vlasak J, Ondrej M, Van Onckelen HA. (1995) The effect of an elevated cytokinin level using the IPT gene and N-6-benzyladenine on single node and intact potato plant tuberisation in vitro. Journal of Plant Growth Regulation 16:143–150.
Guerineau F, Woolston S, Brooks L, Mullineaux P. (1988) An expression cassette for targeting foreign proteins into chloroplasts. Nucleic Acids Research 16:11380.
Harborne JB. (1991) Recent advances in the ecological chemistry of plant terpenoids. In Harborne JB and Tomas-Barberan RA (Eds.). Ecological chemistry and biochemistry of plant terpenoids (Clarendon Press, Oxford, UK) pp. 399–426.
Hedden P and Phillips AL. (2000) Gibberellin metabolism: new insights revealed by the genes. Trends in Plant Science 5:523–530.[CrossRef][ISI][Medline]
Hirschberg J. (2001) Carotenoid biosynthesis in flowering plants. Current Opinion in Plant Biology 4:210–218.[CrossRef][ISI][Medline]
Kakimoto T. (2003) Biosynthesis of cytokinins. Journal of Plant Research 116:233–239.[CrossRef][ISI][Medline]
Kasahara H, Takei K, Ueda N, Hishiyama S, Yamaya T, Kamiya Y, Yamaguchi S, Sakakibara H. (2004) Distinct isoprenoid origins of cis- and trans-zeatin biosyntheses in Arabidopsis. Journal of Biological Chemistry 279:14049–14054.
Krall L, Raschke M, Zenk MH, Baron C. (2002) The Tzs protein from Agrobacterium tumefaciens C58 produces zeatin riboside 5'-phosphate from 4-hydroxy-3-methyl-2-(E)-butenyl diphosphate and AMP. FEBS Letters 527:315–318.[CrossRef][ISI][Medline]
Lange BM and Ghassemian M. (2003) Genome organization in Arabidopsis thaliana: a survey for genes involved in isoprenoid and chlorophyll metabolism. Plant Molecular Biology 51:925–948.[CrossRef][ISI][Medline]
Laule O, Furholz A, Chang HS, Zhu T, Wang X, Heifetz PB, Gruissem W, Lange M. (2003) Crosstalk between cytosolic and plastidial pathways of isoprenoid biosynthesis in Arabidopsis thaliana. Proceedings of the National Academy of Sciences, USA 100:6866–6871.
Livak KJ. 1997. Relative quantitation of gene expression. PE Applied Biosystems, Foster City, CA, USA.
Lois LM, Rodriguez-Concepcion M, Gallego F, Campos N, Boronat A. (2000) Carotenoid biosynthesis during tomato fruit development: regulatory role of 1-deoxy-D-xylulose 5-phosphate synthase. The Plant Journal 22:503–513.[CrossRef][ISI][Medline]
McGarvey DJ and Croteau R. (1995) Terpenoid metabolism. The Plant Cell 7:1015–1026.[CrossRef][ISI][Medline]
Mandel MA, Feldmann KA, Herrera-Estrella L, Rocha-Sosa M, León P. (1996) CLA1, a novel gene required for chloroplast development, is highly conserved in evolution. The Plant Journal 9:649–658.[CrossRef][ISI][Medline]
Morris WL, Ducreux L, Griffiths DW, Stewart D, Davies HV, Taylor MA. (2004) Carotenogenesis during tuber development and storage in potato. Journal of Experimental Botany 55:975–982.
Murashige T and Skoog F. (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiologia Plantarum 15:473–497.[CrossRef]
Nesterenko S and Sink K. (2003) Carotenoid profiles of potato breeding lines and selected cultivars.