Figures
Abstract
Some potato species require a short-day (SD) photoperiod for tuberization, a process that is negatively affected by gibberellins (GAs). Here we report the isolation of StGA3ox2, a gene encoding a GA 3-oxidase, whose expression is increased in the aerial parts and is repressed in the stolons after transfer of photoperiod-dependent potato plants to SD conditions. Over-expression of StGA3ox2 under control of constitutive or leaf-specific promoters results in taller plants which, in contrast to StGA20ox1 over-expressers previously reported, tuberize earlier under SD conditions than the controls. By contrast, StGA3ox2 tuber-specific over-expression results in non-elongated plants with slightly delayed tuber induction. Together, our experiments support that StGA3ox2 expression and gibberellin metabolism significantly contribute to the tuberization time in strictly photoperiod-dependent potato plants.
Citation: Bou-Torrent J, Martínez-García JF, García-Martínez JL, Prat S (2011) Gibberellin A1 Metabolism Contributes to the Control of Photoperiod-Mediated Tuberization in Potato. PLoS ONE 6(9): e24458. https://doi.org/10.1371/journal.pone.0024458
Editor: Miguel A. Blazquez, Instituto de Biología Molecular y Celular de Plantas, Spain
Received: May 26, 2011; Accepted: August 10, 2011; Published: September 22, 2011
Copyright: © 2011 Bou-Torrent et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by grants from the Ministerio de Educación y Ciencia - FEDER and the Generalitat de Catalunya to SP (BIO96-0532-C02-02, BIO99-0961, BIO2005-07284), to JLGM (BIO2009-07968) and to JFMG (BIO2008-00169, 2005SGR-00284, Xarxa de Referència en Biotecnologia). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Potato tubers differentiate at the tip of stolons (underground stems), which upon tuber induction stop growing longitudinally and start swelling. They serve as vegetative propagation organs, as well as energy reservoir for the future new plant. Potato tubers are a rich source of carbohydrates widely used in food and industry, being the third crop in economical importance after wheat and rice (faostat.fao.org). Hence, it is of great agronomic interest to unveil signals that control tuber formation.
The control of tuber formation is complex and environmental factors such as photoperiod, temperature, and shortage in nitrogen supply are known to have an important effect [1]. Short days (SD, 8 h light/16 h dark) favor tuberization whereas long days (LD, 16 h light/8 h dark) delay this developmental process. Sensitivity to day length depends on the genetic background: whereas most commercial cultivars of Solanum tuberosum were bred to be relatively independent on day length for tuberization, wild potato species like S. demissum and several varieties of S. tuberosum ssp. andigena are strictly dependent on SD for tuber formation [2]. These wild species tuberize under SD conditions, but do not form tubers when grown under LDs.
Current evidence supports the existence of at least a photoperiod- and a GA-dependent pathway in controlling potato tuberization [3], [4], [5]. Somehow these independent pathways interact and the balance between their positive and negative effects determines tuberization. Therefore, some cross-talk between both pathways is occurring.
Gibberellins (GAs) have also been shown to regulate tuberization. For instance, greater GA content is observed in stolons of plants grown under non-inductive LDs, whereas a decrease in GA activity is found when leaves are exposed to inductive SD [6], [7]. Increased GA activity has also been observed in response to high temperatures or continuous nitrate supply, conditions that prevent tuber formation [8], [9]. The dwarf ga1 mutant of the photoperiod-dependent S. tuberosum ssp andigena, which appears to be blocked in the GA biosynthesis 13-hydroxylation step (Figure 1A), forms tubers after culture for several months under LD conditions [10]. Treatment of potato ssp. andigena plants with GA synthesis inhibitors induces tuberization in LD [11]. Altogether, the content of GAs is greater in conditions that inhibit tuberization and conversely a reduction in GA content might induce tuberization even under non-inductive conditions, which led to the generally accepted conclusion that GAs inhibit tuberization.
(A) The pathway is classified into three groups based on the enzymatic activity and subcellular localization. Plastidial terpene cyclases convert the precursor geranylgeranyl diphosphate (GGDP) into ent-copalyl diphosphate (CDP), and subsequently into ent-kaurene, in a two-step process catalyzed by ent-copalyl diphosphate synthase (CPS) and ent-kaurene synthase (KS). Cytochrome P-450 mono-oxygenases such as ent-kaurene oxidase (KO), ent-kaurenoic acid oxidase (KAO) yield GA12. Cytoplasmic dioxygenases are responsible for the latest steps and conversion to bioactive gibberellins and their inactive catabolites. GA12 in potato is early 13-hydroxylated by GA 13-oxidase (GA13ox). As indicated by curved dotted T-lines or arrows emerging from the bioactive GA1, GA20ox and GA3ox corresponding transcripts are under negative feed-back regulation, whereas GA2ox mRNA is under positive feed-forward regulation by GA1. (B) Neighbor-joining tree obtained with the alignment of the amino acid sequences: St3ox1 and St3ox2 correspond to potato StGA3ox1 (FJ792644) and StGA3ox2 (FJ792643); At3ox1 and At3ox2 to arabidopsis AtGA3ox1 (NM101424) and AtGA3ox2 (NM106683); Ps3ox1 to pea PsGA3ox1 (AF010167), Ls3h1 and Ls3h2 to lettuce LsGA3ox1 (AB012205) and LsGA3ox2 (AB012206); Le3ox1 and Le3ox2 to tomato SlGA3ox1 (AB010991) and SlGA3ox2 (AB010992), tobacco Nty is NtGA3ox1 (AB032198), and Os3ox1 and Os3ox2 are rice genes OsGA3ox1 and OsGA3ox2 (AB054084 and AB056519).
GA 20-oxidases and GA 3-oxidases catalyze the last two steps of active GA biosynthesis, and GA 2-oxidase their conversion to inactive catabolites (Figure 1A). These three types of enzymes are encoded by small families of GA20ox, GA3ox and GA2ox genes [12], [13]. All three enzymatic steps are thought to be main sites of GA biosynthesis regulation (Figure 1A) [14]. GA20ox and GA3ox expression is under feed-back regulation by GA1/GA3, whereas GA2ox expression is under feed-forward regulation by bioactive GAs [12]. In addition, GA20ox expression is also regulated in response to day length in several plant species [12], [15], [16], [17]. In Arabidopsis and spinach, for example, bolting in response to LD conditions is associated to greater levels of GA20ox expression [18], [19]. In potato, by contrast, levels of the three StGA20ox transcripts were not greater in the leaves of potato plants grown under non-inductive (SD+NB or LD) conditions compared to plants grown under tuber-inducing (SD) conditions. Therefore, it was concluded that changes in GA levels observed during tuber induction do not result from regulated expression of this biosynthetic activity [20].
Light regulation of GA biosynthetic gene expression or activities other than GA 20-oxidases has also been reported [15]. For instance, enhanced synthesis of ent-kaurene under long photoperiods has been observed in the LD plants spinach (Spinacia oleracea) and Agrostemma githago [21]. In Arabidopsis and lettuce (Lactuca sativa) seeds, expression of the GA 3-oxidases AtGA3ox2 (GA4H) and Ls3ox1 genes is induced by seed exposure to red light and this effect is reversed by far-red light treatment. Hence, red light appears to promote GA1 synthesis in these seeds by inducing GA 3-oxidase expression via phytochrome action [22], [23]. Transfer of etiolated pea (Pisum sativum) seedlings to light, in its turn, was found to down-regulate expression of PsGA3ox1 (Mendel's Le gene) and up-regulate PsGA2ox2, encoding for a deactivating GA 2-oxidase [24], [25]. In cowpea (Vigna sinensis), end-of-day far-red treatments, that stimulate epicotyl elongation, resulted in an increase in GA1 levels by inhibiting at least GA 2-oxidase activity in the epicotyl tissues [26]. In potato, StGA2ox1 expression is strongly up-regulated during the early stages of potato tuber development, prior to visible swelling. Characterization of transgenic potato plants with altered levels of StGA2ox1 led to propose a role for this gene in early tuber initiation by reducing GA levels in the subapical stolon region at the onset of tuberization, thereby facilitating normal tuber development and growth [27]. Therefore, the regulated expression of a potato biosynthetic activity different from GA 20-oxidase may modulate reduced GA synthesis under SD conditions and/or contribute to tuberization.
To address the aforementioned possibility, we set out to clone additional GA biosynthetic genes from S. tuberosum ssp. andigena. In this manuscript, we report the functional characterization of StGA3ox2, a gene encoding a GA 3-oxidase from potato, whose expression is differentially regulated in aerial parts and stolons after transfer to SD inductive conditions. Our experiments suggest the importance of this activity in the control of photoperiod-induced potato tuberization.
Materials and Methods
Plant material and growth conditions
Photoperiodic Solanum tuberosum ssp. andigena plants were propagated in vitro in MS media supplemented with 20 g·L−1 sucrose before being transferred to soil, or from tubers. Plant height analysis was performed with plants at the 14-leaf stage, grown in the greenhouse under LD conditions (16 h light/8 h darkness, 22°C). Internode length was measured as reported before [28]. Transformation of potato plants was performed as described elsewhere [29]. About 40 independent transgenic lines were regenerated for each construct, transferred to soil and analyzed at the RNA level. For tuber induction measurements, plants were grown in the greenhouse under LD conditions until they reached a 14-leaf stage and then were transferred to growth chambers under SD conditions (8 h light/16 h dark, 22°C, SD inducing conditions) or to SD with a light treatment of 30 min or night-break in the middle of the dark period (SD+NB, non-inducing conditions). Light intensity in the growth chambers was about 200 µmol·m−2·s−1 provided by high-pressure sodium lamps SON-T AGRO 400 (Philips).
Potato GA3ox PCR amplification and plasmid constructs
Since potato and tomato cDNA collections were very limited at the time this work was initiated, in order to amplify potato cDNAs encoding GA 3-oxidases, degenerated oligonucleotides were designed upon conserved sequences of the AtGA3ox1 (GA4) and PsGA3ox2 (Le) genes from Arabidopsis and pea [30], [31], [32] and used in PCR. Primers H5 (5′-TGG-GGI-(AG)(CT)I-TT(CT)-CA(AG)-(AG)T-3′) and H7 (5′-GT(AG)-TGI-G(GC)I-GCI-A(AG)I-CCC-AT-3′), complementary to the regions 78-WGAFQI-83 and 227-MGLAAHT-233 in AtGA3ox1 respectively yielded a band of expected size. PCR fragments were cloned in pGEM T-easy (Promega) and sequenced. A first strand cDNA obtained from mRNA extracted from leaves of the potato GA deficient mutant ga1 [33] was used as template for PCR amplification. The subcloned B3ox PCR product, which corresponded to a transcript with negative feed-back regulation (Figure S1), was used to screen a cDNA library constructed from leaves of the potato ga1 mutant [20]. Several positives clones were isolated and sequenced (Results S1, Figure S2A), resulting in the isolation of one single full cDNA clone, closely related to the later reported tomato GA 3-oxidase SlGA3ox2 [34]. Hence, the isolated potato clone was designated as StGA3ox2 (gb|ACN89834). Specific primers B8 (5′-AGT-CCT-TCA-AGA-ATC-3′) and B9 (5′-GTA-AGT-GTC-ACT-AGA-GAA-T-3′), complementary to the tomato SlGA3ox1 [34], were used to amplify a second potato GA 3-oxidase-encoding gene, which was designated as StGA3ox1 (gb|ACN89835).
To produce StGA3ox2 protein in E. coli its coding sequence was amplified using oligonucleotides 3B5 (5′-AAT-TTT-CTA-CTC-ATA-TGC-CTT-CAA-3′) and 3B3 (5′-GCC-GCT-CGA-GGC-CTA-CTT-GGA-GAC-3′) bearing NdeI and XhoI restrictions sites. The PCR product was conveniently digested and ligated to the pET28 (Novagen) expression vector restricted with the same pair of enzymes. The clone pET3ox2 was isolated after PCR selection and its sequence was confirmed.
To over-express StGA3ox2 in transgenic potato plants the isolated cDNA clone number 10 (Figure S2A) was digested with EcoRI (containing the complete coding region and part of the 3′ non-coding region), filled-in and cloned into three different plant binary vectors: pBinAR [35], pBinA6 [36], and pB33 [29], directing expression under control of the constitutive CaMV35S promoter (lines 35S:3ox2), the green-tissue specific promoter StLS1 (lines LS1:3ox2), and the tuber-specific patatin promoter (lines Tub1:3ox2) respectively.
RNA isolation and gel-blot analysis
Total RNA was extracted from different potato plant tissues according to the method described [37]. Electrophoresis, northern analysis and labeling of the probes were carried out as described [28].
Gibberellin content analysis
GA content was determined from the apical part of potato shoots (including the shoot apex and the four youngest internodes and leaves). Tissue was frozen in liquid N2, kept at −80°C until use for GA extraction and quantification by HPLC and GC-MS as described elsewhere [38].
GA 3-oxidase recombinant protein and activity assay
BL21 E. coli competent cells were transformed with pET3ox2. Cultures were grown at 30°C until OD600 = 0.4 and recombinant StGA3ox2 protein accumulation was induced with 1 mM IPTG for 2 h shaking at 30°C. Aliquots were taken before and after induction, to check the accumulation of the recombinant protein in the soluble fraction. Cell lysates producing StGA3ox2 were prepared by resuspending initial 50 mL pelleted cultures in 2 mL TrisHCl pH = 7.5, 4 mM DTT and 1 mg/mL lysozime. After 15 min incubation at room temperature and sonication, extracts were ultracentrifuged at 35.000 rpm for 30 min at 4°C. Supernatants were frozen for activity assays. 85 µL aliquots of these cell lysates were incubated at 30°C for 2 h with either [3H]-GA20 and [2D]-GA20 or with [14C]-GA9 in a total volume of 100 µL containing 4 mM 2-oxoglutarate, 4 mM ascorbate, 0.5 mM ferrous sulphate, 4 mM DTT and 2 mg/mL BSA, as described in [39]. The reaction mixture was analyzed by HPLC, detecting the radioactivity of each sample in a Beckman scintillation counter. Fractions at elution times corresponding to GA1 and GA4 were analyzed by GC-MS for confirmation of the reaction products.
Results
Potato GA 3-oxidase is encoded by at least two genes
When this work was initiated, potato and tomato cDNA collections were very limited and no information was available on GA 3-oxidase-encoding genes from these species. Therefore cDNA prepared from leaves of the S. tuberosum ssp andigena ga1 mutant was used as template to amplify StGA3ox genes, because this mutant is supposed to accumulate high levels of the GA 3-oxidase-encoding transcripts due to negative feed-back regulation. As a result (check Methods for detailed information), we identified a full length clone with a higher percentage of homology to the pea PsGA3ox1 (Le) and the Arabidopsis AtGA3ox2 (GA4) genes [30], [31], [32] than to any other 2-oxoglutarate dependent dioxygenase sequence available at that moment, thus suggesting that it encoded a potato GA 3-oxidase. The sequences from tomato (Solanum lycopersicum), SlGA3ox1 and SlGA3ox2, were later reported [34], and because the potato full length clone shared 92% identity with SlGA3ox2, it was designated as StGA3ox2. Based on the nucleotide sequence of the SlGA3ox1 gene and using RT-PCR and specific primers (B8 and B9) on potato RNA as a template, a new cDNA was cloned, named as StGA3ox1, which included the complete coding region corresponding to this gene (Figure S2B). From these results we conclude that potato, like tomato, has at least two independent genes encoding GA 3-oxidase enzymes as shown in the phylogenetic tree of Figure 1B.
StGA3ox1 and StGA3ox2 show a differential spatial pattern of expression
The expression of the StGA3ox1 and StGA3ox2 genes was analyzed by Northern blot using RNAs from different organs of S. tuberosum ssp andigena plants (leaf-14 stage) grown under non-inducing LD conditions (Figure 2A). Because at this developmental stage plants had very few or no stolons, these organs were not harvested in our analyses. Different patterns of expression were observed for the two StGA3ox genes. Relatively high levels of the StGA3ox2 transcript were found in the apex, internodes and stem nodes, lower levels in flowers and leaves and almost no expression in roots. StGA3ox1 transcripts were detected mainly in flowers and, to a lower extent, in apex, nodes, internodes and roots. Almost no expression was detected in leaves (Figure 2A). Thus, whereas StGA3ox2 was expressed in most vegetative tissues, StGA3ox1 transcripts seemed to be more abundant in floral organs. Consequently, we focused on the study of StGA3ox2. Consistent with our results, data on the expression levels of the two StGA3ox genes in non-tuberizing stolon tips detected expression of only StGA3ox2 [27].
(A) Analysis of tissue-specific pattern of expression of clones StGA3ox1 and StGA3ox2, at 14-leaf developmental stage of plants grown in the greenhouse (LD). (B) Expression of StGA3ox2 in different organs from plants grown in LD until the 14-leaf stage and then entrained to SD+NB or to SD for three weeks. 30 µg of total RNA from each sample was electrophoresed and equal loading on the gel was assessed by ethidium bromide staining, representative pictures are shown out of at least two independent Northern analysis.
StGA3ox2 encodes a functional potato GA 3-oxidase
In order to assess the GA 3-oxidase activity of the StGA3ox2 protein, we obtained pET3ox2 clone, containing the coding sequence of StGA3ox2 in frame in a pET28 (Novagen) expression vector. E. coli BL21 strain was transformed with the pET3ox2 clone and grown at 30°C. The recombinant protein was allowed to accumulate by induction with IPTG for 2 h (Figure 3A). Crude extracts of the expression cultures were incubated for 2 h at 30°C in the presence of the radiolabeled substrates, either [3H]-GA20 or 14C-GA9. HPLC analysis of the reaction products resulted with radioactivity peaks at the same elution time as 3H-GA1 and 14C-GA4 in each reaction (Figure 3B). These fractions were further identified by GC-MS as GA1 and GA4 (data not shown), thus confirming that StGA3ox2 encodes a functional GA 3-oxidase.
(A) Recombinant StGA3ox2 accumulated in BL21 E.coli growing at 30°C. After 2 h induction with IPTG, we detected a band with a molecular weight of 43 kDa corresponding to StGA3ox2. (B) HPLC chromatogram after 2 h incubation of E.coli lysates expressing recombinant StGA3ox2 feeded with the radiolabeled putative substrates 3H-GA20 and 14C-GA9. The corresponding products peaked at the 3H-GA1 and 14C-GA4 corresponding elution times.
Expression of StGA3ox2 in the stolons is strongly repressed under tuber-inducing (SD) conditions
In previous studies no major changes were detected in the levels of expression of the StGA20ox genes between potato plants induced (SD) and non-induced (SD+NB or LD) to tuberize [20]. Therefore, we have investigated whether photoperiod regulated the expression of StGA3ox2 in different organs of 14-leaf plants grown for 3 additional weeks under inductive (SD) and non-inductive (SD+NB) conditions (Figure 2B). In plants grown under non-inductive conditions (SD+NB or LD) levels of StGA3ox2 transcript were similar but not identical (compare Figs. 2A and 2B). Remarkably high levels of StGA3ox2 mRNA were detected in the underground stolons. By contrast, in plants grown under inductive SD conditions, StGA3ox2 expression was clearly up-regulated in apex, leaves and nodes. The most conspicuous change in gene expression however was found in the stolons, where StGA3ox2 expression was completely repressed, with no transcript detected either in stolons of plants induced to tuberize or tubers already formed (Figure 2B). Since exposure of potato plants to SD tuber-inducing conditions results in a large decrease in stolon GA levels [7], our correlative data led us to hypothesize that the repression of this GA 3-oxidase activity participates in the reduction of stolon GA content associated with tuber induction.
Expression of StGA3ox2 in the aerial part of the plant is diurnally regulated and is increased under tuber-inducing (SD) conditions
The levels of StGA3ox2 transcript in leaves were found to be dependent on the time of the day at which leaves were sampled. To analyze whether this was due to a diurnal rhythm of expression, we studied the accumulation of StGA3ox2 messenger in potato leaves of plants grown under inducing (SD) and non-inducing (SD+NB) conditions, over a period of 24 h. Similar patterns of transcript abundance were observed in plants entrained to SD or SD+NB conditions within the 24 h period (Figure 4). The expression of StGA3ox2 fluctuated during the day, with transcript levels reaching a peak around 3 h after starting the light period (lights were switched on at 11:00 h and maximal levels of transcript were detected at 14:30 h) and a valley around 2 h after lights switch-off, after which they gradually recovered during the night period (Figure 4). Although the pattern of StGA3ox2 transcript accumulation was similar in leaves of plants entrained to SD or SD+NB conditions, levels of StGA3ox2 mRNA were higher in SD- than SD+NB-grown plants over the entire 24 h period. This agrees with the results shown in Figure 2B and suggests that, in contrast with that observed in the underground stolons, increased levels of StGA3ox2 mRNA occur in leaves and other aerial part of the shoot of plants entrained to SD conditions compared to plants grown under SD+NB non-inducing conditions.
Plants at 14-leaf stage were entrained to SD (8 h light/16 h dark) or SD+NB conditions (8 h light/16 h dark with a 30 min light break in the middle of the night) for 3 weeks. Leaf samples were harvested at the time points indicated, approximately every 2 to 3 hours, for a period of 24 hours. The lanes come from the same gel, a dividing line (in SD panels) indicates intervening lanes have been removed. Films were quantified by densitometric scanning and levels of transcript were normalized to the sample with the strongest hybridization signal (14:30 h in SD), to which we assigned an arbitrary value of 100. 30 µg of total RNA from each sample was electrophoresed and equal loading on the gel was assessed by ethidium bromide staining, representative pictures are shown out of at least two independent Northern analysis.
Constitutive over-expression of StGA3ox2 in the aerial parts results in elongated plants and early tuberization under SD conditions
The putative regulatory role of GA biosynthesis catalyzed by StGA3ox2 led us to investigate its function in the photoperiod-dependent tuberization. Plants over-expressing the StGA3ox2 transcript under the control of the constitutive 35S promoter (35S:3ox2 lines) or the green tissue specific StLS1 promoter (LS1:3ox2 lines) were obtained. The StLS1 promoter drives expression to green tissues of the leaves and stem [36]. Several transformants were regenerated that accumulated higher levels of StGA3ox2 transcript than the controls (Figure 5A). These lines exhibited an elongated phenotype as compared to the wild-type controls (Figure 5B and 6), with taller stems caused by an increase in internode length and not by an elevated number of internodes (see Figure 5C). GA content was quantified in shoots (leaves and stems) of four different StGA3ox2 over-expressing lines. Higher levels of GA1 (ranging from 1.3 to 2.2 fold-increase), and much lower levels of its 13-hydroxylated precursor GA20 and of GA29 (a GA20 catabolite) were found in both types of over-expressing lines (35S:3ox2 and LS1:3ox2) compared to wild-type plants (Table 1). Transformants with a more elongated phenotype also exhibited the highest levels of GA1, thus evidencing a positive correlation between increased levels of bioactive GAs and internode growth. No apparent differences in GA53, GA44 and GA19 (GA1 precursors) were detected (Table 1).
(A) RNA blot analysis from LD-grown shoots, 30 µg of total RNA from each sample was electrophoresed and equal loading on the gel was assessed by ethidium bromide staining, a representative picture is shown out of at least two independent Northern analysis. (B) Stem height and (C) Internode length of overexpressing lines, wt and cosupressed line grown under LD conditions. Mean value obtained from 10–15 individual plants from each line. Errors bars indicate SE, asterisks (*) highly significant differences (P<0,01) and circles ( ) significant differences (P<0,05) compared to wt and for the 5th internode in C. (D) Tuber induction data, represent the average of 10–15 independent replicates for each line from a representative experiment out of three with similar results.
Plant height of the constitutive 35S:3ox2–10 (35S:3ox2) and the tuber-specific Tub1:3ox2–30 (Tub1:3ox2) over-expressers is shown in comparison to that of lines over-expressing the GA 20-oxidase activity (35S:20ox1) or wild-type (wt) plants. The number of nodes was the same in all plants.
None of the StGA3ox2 over-expressing lines formed tubers under LD conditions but, when transferred to inducing SD conditions, they tuberized earlier than the controls (Figure 5D). The early tuberization response was significantly observed in both types of over-expressing lines (35S:3ox2 and LS1:3ox2) (Table S1). Consistently, the yield in mass of tubers per plant was significantly higher in the over-expressing lines than in the controls (Table 2).
Reduced levels of StGA3ox2 in the aerial parts results in shorter plants and delayed tuberization under SD conditions
An exception in our analyses was the LS1:3ox2–16 line that was significantly shorter than the controls and in which reduced levels of StGA3ox2 transcript were detected (Figure 5A–C), likely due to a co-suppression effect. Interestingly, the LS1:3ox2–16 plants tuberized significantly later than wild-type controls (Figure 5D; Table S1B). In addition, the co-suppressed LS1:3ox2–16 line produced smaller tubers and a lower tuber yield per plant (Table 2).
Restricted over-expression of StGA3ox2 in tubers slightly delays tuberization under SD conditions
The inverse correlation between the levels of StGA3ox2 mRNA and the number of inductive SDs required for tuber formation found in StGA3ox2 over-expressing and co-suppressed lines (Figure 5) is in contrast with the inhibitory role of GAs in regulating tuberization. To address whether high levels of StGA3ox2 in the stolons inhibit tuberization, transgenic lines that over-expressed the StGA3ox2 transcript only in tubers were obtained, by expressing its coding region under the patatin B33 promoter control [29]. Several transformants (Tub1:3ox2 lines) were regenerated that accumulated in tubers higher levels of the StGA3ox2 mRNA than the controls (Figure 7A). Stem lengths of these lines were slightly longer or very similar to wild-type, with similar results obtained in measurements done in plants grown under LD conditions or after transferring them to SD to induce tuber formation (Figure 6 and 7B). Tuberization onset in the Tub1:3ox2 over-expressers occurred at the same time or slightly delayed (but not earlier) compared to wild-type (Figure 7C; Table S1C). Tuber yield in these lines measured as mass of tubers per plant was similar to that of the controls. These plants, however, formed more tubers per plant compared to the controls (Table 2). Tubers were smaller and more elongated than those of control plants (data not shown), thus evidencing that the GA 3-oxidase activity does effectively over-accumulate in the transgenic tubers, with high levels of GA 3-oxidase activity affecting tuber growth and apical dominance by primarily formed tubers.
(A) RNA blot analysis from SD-grown tuber tissues, 30 µg of total RNA from each sample was electrophoresed and equal loading on the gel was assessed by ethidium bromide staining, a representative picture is shown out of at least two independent Northern analysis. (B) Stem heights of Tub1:3ox2 lines and wt (10 individual plants per line), measured after 3 weeks under SD tuber-inducing conditions. Errors bars indicate SE, asterisks (*) highly significant differences (P<0,01) and circles (°) significant differences (t-test; P<0,05) compared to wt. (C) Tuber induction data represent the average of five to nine independent replicates for each line from a representative experiment out of three with similar results.
Discussion
The GA 3-oxidase activity plays a role in controlling both early and late responses after transfer to SD inductive conditions
We report the isolation of StGA3ox2 and StGA3ox1, two potato cDNA clones encoding GA 3-oxidases (Figure 1A). The recombinant product of StGA3ox2 is a functional enzyme catalyzing the conversion of both 3H-GA20 and 14C-GA9 into 3H-GA1 and 14C-GA4, respectively (Figure 3B). This agrees with pea PsGA3ox1 and Arabidopsis AtGA3ox2 genes encoding GA 3-oxidase proteins catalyzing the synthesis of active GAs both in the early- and non-C13-hydroxylation pathways [30], [31], [32]. The higher StGA3ox2 expression in the shoots of the 35S:3ox2 and LS1:3ox2 transgenic plants resulted in lower levels of GA20 (the substrate) and higher levels of GA1 (the product) than in control plants (Table 1), demonstrating that StGA3ox2 encodes an enzyme with GA 3-oxidase activity also in planta. In addition, GA 3-oxidase activity might have a role in the control of bioactive GA synthesis during tuber induction, as we detected a strong down-regulated expression of StGA3ox2 in stolons induced to tuberize, possibly contributing to the drop in GA1 levels observed in the stolons of plants exposed to SD tuber-inducing conditions (Figure 2B) [27].
One of the earlier events associated with the onset of tuberization is a change in the plane of cell division, from longitudinal to transversal, of the cells at the sub-apical region of the stolon. As a consequence of this switch, stolon growth is arrested and lateral tuber expansion begins [40], [41]. Polarized cell expansion in plant cells is controlled by cortical microtubules localized below the plasma membrane [42]. Orientation of these microtubules has been shown to be regulated by GAs, which promote a perpendicular distribution of the microtubules to the growing axis of the cell, thereby directing cell elongation along the longitudinal axis [43]. In induced stolons, decrease in GA1 levels partially caused by down-regulated expression of the StGA3ox2 gene expression is thus likely to contribute to tuber formation, in concert with changes in the expression of other GA metabolism genes [27]. A complex regulatory mechanism at the transcriptional level has been proposed for such coordinated induction of StGA2ox1, StGA20ox1, and StGA20ox3 expression and repression of StGA3ox2 expression, after the switch from LD to SD conditions, possibly mediated by a common mechanism [27].
Transfer of plants to SD conditions induces at the same time a drop in the StGA3ox2 mRNA levels in the stolons and a rise in the levels of transcript in the aerial part of the plant (shoot apex, leaves and the nodes; Figure 2B). These changes in gene expression are accompanied by an increase in stem elongation as a result of the reduced number of hours of light that the plants receive per day. While stem elongation is already observed after 1 week of transfer the plants to SD conditions, tuber formation requires at least 2 weeks under inductive conditions and, thereby, stem elongation was considered to be a short-term adaptive response independent of tuberization [3]. Our results suggest that up-regulated expression of StGA3ox2 in the leaves might mediate this short-term elongation response. Alternatively, it might reflect the positive feed-back of a general reduction of bioactive GA levels in plants that have already stopped growing and started to senesce [3].
Over-expression of StGA3ox2 in the leaves results in elongated plants and early tuberization
We have investigated the function of a GA 3-oxidase in the regulation of tuber induction by generating transgenic potato plants that over-expressed StGA3ox2 under the control of the constitutive 35S CaMV (35S:3ox2) or the potato green tissue-specific StLS1 (LS1:3ox2) promoters. As expected, transformants accumulating high levels of the transgene exhibited a taller phenotype due to longer internodes (Figure 5). Consistently, the co-suppressed line LS1:3ox2–16, with reduced StGA3ox2 mRNA levels, showed reduced plant height with shorter internodes. Increase of plant height was similar in both constitutive (35S:3ox2) and leaf-specific (LS1:3ox2) over-expressers but always smaller than that observed in transformants over-expressing StGA20ox1 [28]. This suggests that GA 20-oxidase activity may be limiting in potato plants (hence, the GA20 precursor) and/or that negative feed-back regulation of the genes encoding this enzyme is tighter than that of genes encoding GA 3-oxidase activity.
Over-expression of StGA3ox2 induced early tuberization under SD inductive conditions, with a higher yield in mass of tubers per plant in the transgenic lines as compared to the wild-type controls (Table 2). Consistently, the co-suppressed line LS1:3ox2–16 tuberized later and formed smaller tubers than the controls. Thus, a direct correlation between increased StGA3ox2 expression in the leaves and early tuberization in SD was found in these lines, with those accumulating the highest levels of StGA3ox2 mRNA (lines 35S:3ox2–10 or LS1:3ox2–18) also exhibiting the earliest tuberization onset. This observation is in apparent contradiction with the widely accepted notion that GAs inhibit tuberization. However, it is important to note that levels of GA1 in these transgenics were only determined in the shoots (Table 1) and we have been unable to quantify them in the stolons. To bypass this technical problem, we obtained transgenic lines that over-expressed StGA3ox2 in tubers (Tub1:3ox2 lines; Figure 7). These lines should accumulate higher levels of GA1 in tubers. Indeed, lines exhibiting increased levels of the StGA3ox2 transcript in tubers showed a slight delay in the tuberization onset under SD conditions (Figure 7C). The patatin B33 promoter is a marker for tuber formation assumed to become active once the stolon starts to differentiate into a tuber [29]. Hence, the Tub1:3ox2 lines should begin to accumulate GA1 in the stolons during their early transition into tubers, with this late increase in GA1 content having a mild impact on the tuberization onset. The Tub1:3ox2 lines, however, formed a higher number of tubers per plant than the controls, but with reduced weight and with several small tubers of elongated shape attached to each stolon (Table 2). This phenotype indicates that the StGA3ox2 product accumulated in the tubers of these lines, probably leads to increased GA1 content affecting tuber size and shape, as well as tuber apical dominance, with growth of secondary tubers observed in these lines. Although there are reports indicating that the patatin promoter is somehow leaky and some reporter expression can be observed in “pre-tuberizing” stolons [44], the subtle tuberization time phenotypes observed in Tub1:3ox2 lines are likely to relate to the effect of altered GA levels on tuber development, whereas highly specific stolon promoters should be used to asses StGA3ox2 photoperiodic control of tuber induction. Altogether, our results seem to suggest that increased levels of GA1 in stolons induced to tuberize have an inhibitory effect on tuber formation, in agreement with the accepted negative role of GAs in this process. Recent studies have suggested that GAs can move inside the plant, for instance a basipetal transport was observed in Populus and from leaves to the shoot apical meristem in Arabidopsis [45], [46], whereas upwards (acropetal) transport was described in pea stems [47].
Despite the subtle but significant phenotypes of our StGA3ox2 over-expression lines, we performed grafting experiments between the wt and 35S:3ox2–5 transgenic line (data not shown). We could not detect any significant difference between the grafted specimens, possibly due to technical problems, such as earlier tuberization time in all grafts induced in older plants or by a tight regulation of GA1 metabolism and degradation possibly mediated by StGA2ox1 [27].
Contrasting effects on tuber induction time in StGA3ox2 and StGA20ox1 over-expressers: the importance of the different mobility of GA20 and GA1 in the plant and the balance between shoots and tubers in metabolizing GA20
Both StGA20ox1 and StGA3ox2 over-expressers display elongated stems and high levels of the bioactive GA1 in the shoots (Table 1; [28]). A relevant difference between these transgenic lines is the different accumulation in the shoots of GA20, the immediate precursor of the bioactive GA1: it was high in StGA20ox1 over-expressers and it was low in StGA3ox2 over-expressers. In addition, the early tuberization observed in the 35S:3ox2 and LS1:3ox2 lines (Figure 5) is in clear contrast with previous results showing that over-expression of StGA20ox1 results in delayed tuber formation under SD inductive conditions [28].
A possible explanation for this paradox is that GA20 and GA1 have different mobility within the potato plants, as occurs in pea. In this species, using le (blocked in GA 3-oxidase) and na (blocked prior conversion to GA12-aldehyde) mutants it was found that while GA1 application induces elongation of le lines, the stature of le is not altered when grafted to wild-type rootstock, indicating that GA1 is not transported in sufficient amount from the wild-type rootstock to the le scion. By contrast, stem elongation of the na line increased after GA application or by grafting to either wt or le rootstocks, indicating that GA20 or earlier precursors are readily transported from the wt or le rootstocks to the na scions, where they can be metabolized to GA1 [47]. In our transgenic Tub1:3ox2 lines (Figure 7), StGA3ox2 is over-expressed in tubers, thus likely accumulating higher local levels of GA1 (as deduced from the elongated tubers phenotype) and delaying tuberization; they however show slightly longer stem lengths or very similar to wt either under LD or after transferring them to SD conditions (Figs. 6, 7B), in agreement with less efficient acropetal GA1 transport.
In 35S:3ox2 and LS1:3ox2 lines, over-expressing organs (aerial shoots) may act as a strong sink of GA20 in the plant transported acropetally from the stolons and roots, which results in increased GA1 levels and stem elongation; this sink effect lowers the level of this GA1 precursor in the induced stolons. As a consequence of this substrate decrease, the GA 3-oxidase activity in the stolons may produce less amount of the bioactive GA1 and tuberization occurs earlier. In Tub1:3ox2 lines, by contrast, the absence of the GA20 sink effect by the aerial organs results in a local increase of GA1 delaying tuberization most likely by altering tuber development.
Together, our experiments provide evidence in support of different mobility for GA20 vs. GA1 that might contribute to tuberization time in potato. Furthermore, these conclusions would support that increased rates of stem elongation may occur concomitantly with an induction of tuber formation, as it was previously observed in phyB antisense plants, in which very elongated internodes (high levels of GA1 in shoots) and pale green leaves, co-existed with a strong induction of tuber formation [3], [48]. Conversely, reduced rates of stem elongation may occur concomitantly with an inhibition of tuberization, as it was observed in potato plants over-expressing Arabidopsis CONSTANS (CO) [4]. Both PHYB and CO are components of the photoperiod pathway that seem to interact with GA metabolism. However, since StGA3ox2 over-expressing or co-suppressing lines still require SD photoperiods to produce tubers, this indicates that impairment of this transcript levels does not constitutively block and/or activate the photoperiod pathway, but modulates it.
In conclusion, we have shown that differential pattern of StGA3ox2 expression in stolons plays a role in tuberization induction, unveiling a possible mechanism by which photoperiod- and GA-dependent pathways cross-talk in controlling potato tuberization. Our study also points out that the role of GAs in the regulation of photoperiod-controlled tuberization results from the combination of the enzymatic activities involved in their local production and metabolism, together with the differential distribution of GA1 and its precursors within the plant. It also highlights the regulatory role played by the integration of local levels of GAs with the photoperiod-dependent signal that induces tuberization.
Supporting Information
Results S1.
Isolation of potato GA 3-oxidase clones.
https://doi.org/10.1371/journal.pone.0024458.s001
(PDF)
Figure S1.
Negative feed-back regulation of the corresponding mRNA hybridizing to PCR product B3ox.
https://doi.org/10.1371/journal.pone.0024458.s002
(PDF)
Figure S2.
Structure of the StGA3ox2 clones and amino acid sequence comparison to other GA 3-oxidases.
https://doi.org/10.1371/journal.pone.0024458.s003
(PDF)
Table S1.
Statistical analysis on the tuberization time in transgenic lines over-expressing StGA3ox2.
https://doi.org/10.1371/journal.pone.0024458.s004
(PDF)
Acknowledgments
We thank P. Fontanet, L. Etxarri, M.C. Badías, E. Saavedra, and A. Sanz for excellent care of the plants.
Author Contributions
Conceived and designed the experiments: SP. Performed the experiments: JB-T. Analyzed the data: JB-T SP JFM-G JLG-M. Contributed reagents/materials/analysis tools: JLG-M JFM-G. Wrote the paper: JB-T SP. Manuscript corrections: JLG-M JFM-G. Financial help: SP JFM-G.
References
- 1.
Krauss A (1985) Interaction of nitrogen nutrition, phytohormones, and tuberization. In: Li PH, editor. Potato Physiology. Orlando: Academic Press, Inc. pp. 209–230.
- 2. Ewing EE, Struik PC (1992) Tuber formation in potato: induction, initiation and growth. Horticultural Review 14: 89–197.
- 3. Martínez-García JF, García-Martínez JL, Bou J, Prat S (2001) The interaction of gibberellins and photoperiod in the control of potato tuberization. Journal of Plant Growth Regulation 20: 377–386.
- 4. Martínez-García JF, Virgós-Soler A, Prat S (2002) Control of photoperiod-regulated tuberization in potato by the Arabidopsis flowering-time gene CONSTANS. Proceedings of the National Academy of Sciences of the United States of America 99: 15211–15216.
- 5. Rodríguez-Falcón M, Bou J, Prat S (2006) Seasonal control of tuberization in potato: conserved elements with the flowering response. Annu Review of Plant Biology 57: 151–180.
- 6. Machackova I, Konstantinova TN, Seergeva LI, Lozhnikova VN, Golyanovskaya SA, et al. (1998) Photoperiodic control of growth, development and phytohormone balance in Solanum tuberosum. Physiology Plantarum 102: 272–278.
- 7. Xu X, van Lammeren AAM, Vermeer E, Vreughdenhil D (1998a) The role of gibberellin, abscisic acid and sucrose in the regulation of potato tuber formation in vitro. Plant Physiology 117: 575–584.
- 8. Woolley DJ, Wareing PF (1972) Environmental effects on endogenous cytokinins and gibberellin levels in Solanum tuberosum. New Phytologist 71: 1015–1025.
- 9. Menzel CM (1983) Tuberization in potato at high temperatures: gibberellin content and transport from buds. Annals of Botany 52: 697–702.
- 10. van den Berg JH, Simko I, Davies PJ, Ewing EE, Halinska A (1995) Morphology and (14C)gibberellin-A12-aldehyde metabolism in wild-type and dwarf Solanum tuberosum ssp. andigena grown under long and short photoperiods. Journal of Plant Physiology 146: 467–473.
- 11. Jackson SD, Prat S (1996) Control of tuberization in potato by gibberellins and phytochrome B. Physiology Plantarum 98: 407–412.
- 12. Yamaguchi S, Kamiya Y (2000) Gibberellin biosynthesis: its regulation by endogenous and environmental signals. Plant Cell Physiology 41: 251–257.
- 13. Hedden P, Phillips AL (2000) Gibberellin metabolism: new insights revealed by the genes. Trends in Plant Science 5: 523–530.
- 14. Hedden P, Kamiya Y (1997) Gibberelin biosynthesis: enzymes, genes and their regulation. Annual Review of Plant Physiology and Plant Molecular Biology 48: 431–460.
- 15. Kamiya Y, García-Martínez JL (1999) Regulation of gibberellin biosynthesis by light. Current Opinion in Plant Biology 2: 398–403.
- 16. Martin DN, Proebsting WM, Parks TD, Dougherty WG, Lange T, et al. (1996) Feed-back regulation of gibberellin biosynthesis and gene expression in Pisum sativum L. Planta 200: 159–166.
- 17. Ross JJ, MacKenzie-Hose A, Davies PJ, Lester DR, Twitchin B, et al. (1999) Further evidence for feedback regulation of gibberellin biosynthesis in pea. Physiology Plantarum 105: 532–538.
- 18. Wu K, Li L, Gage DA, Zeevaart JAD (1996) Molecular cloning and photoperiod-regulated expression of gibberellin 20-oxidase from the long-day plant spinach. Plant Physiology 110: 547–554.
- 19. Xu YL, Gage DA, Zeevaart JAD (1997) Gibberellins and stem growth in Arabidopsis thaliana: effects of photoperiod on expression of the GA4 and GA5 loci. Plant Physiology 114: 1471–1476.
- 20. Carrera E, Jackson SD, Prat S (1999) Feedback control and diurnal regulation of gibberellin 20-oxidase transcript levels in potato. Plant Physiology 119: 765–774.
- 21. Zeevart JAD, Gage DA (1993) ent-kaurene biosynthesis is enhanced by long photoperiods in the long-day plants Spinacea oleracea L. and Agrostemma githago L. Plant Physiology 101: 25–29.
- 22. Toyomasu T, Kawaide H, Mitsuhashi W, Inoue Y, Kamiya Y (1998) Phytochrome regulates gibberellin biosynthesis during germination of photoblastic lettuce seeds. Plant Physiology 118: 1517–1523.
- 23. Yamaguchi S, Smith MW, Brown RG, Kamiya Y, Sun TP (1998) Phytochrome regulation and differential expression of gibberellin 3beta-hydroxylase genes in germinating Arabidopsis seeds. Plant Cell 10: 2115–2126.
- 24. Ait-Ali T, Frances S, Weller JL, Reid JB, Kendrick RE, et al. (1999) Regulation of gibberellin 20-oxidase and gibberellin 3beta-hydroxylase transcript accumulation during de-etiolation of pea seedlings. Plant Physiology 121: 783–791.
- 25. Reid JB, Botwright NA, Smith JJ, O'Neill DP, Kerckhoffs LH (2002) Control of gibberellin levels and gene expression during de-etiolation in pea. Plant Physiology 128: 734–741.
- 26. Martínez-García JF, Santes CM, García-Martínez JL (2000) The end-of-day far-red irradiation increases gibberellin A1 content in cowpea (Vigna sinensis L.) by reducing its inactivation. Physiologia Plantarum 108: 426–434.
- 27. Kloosterman B, Navarro C, Bijsterbosch G, Lange T, Prat S, et al. (2007) StGA2ox1 is induced prior to stolon swelling and controls GA levels during potato tuber development. The Plant Journal 52: 362–373.
- 28. Carrera E, Bou J, García-Martínez JL, Prat S (2000) Changes in GA 20-oxidase gene expression strongly affect stem length, tuber induction and tuber yield of potato plants. Plant Journal 22: 247–256.
- 29. Rocha-Sosa M, Sonnewald U, Frommer W, Stratmann M, Schell J, et al. (1989) Both developmental and metabolic signals activate the promoter of a class I patatin gene. EMBO Journal 8: 23–29.
- 30. Chiang HH, Hwang I, Goodman HM (1995) Isolation of the Arabidopsis GA4 locus. Plant Cell 7: 195–201.
- 31. Lester DR, Ross JJ, Davies PJ, Reid JB (1997) Mendel's stem length gene (Le) encodes a gibberellin 3 beta -hydroxylase. Plant Cell 9: 1435–1443.
- 32. Martin DN, Proebsting WM, Hedden P (1997) Mendel's dwarfing gene: cDNAs from the Le alleles and function of the expressed proteins. Proceedings of the National Academy of Sciences USA 94: 8907–8911.
- 33. Bamberg JB, Hanneman RE (1991) Characterization of a new gibberellin related dwarfism locus in potato (Solanum tuberosum L.). American Potato Journal 68: 45–52.
- 34. Rebers M, Kaneta T, Hawaide H, Yamaguchi S, Yang YY, et al. (1999) Regulation of gibberellin biosynthesis genes during flower and early fruit development of tomato. Plant Journal 17: 241–250.
- 35. Höfgen R, Willmitzer L (1990) Biochemical and genetic analyisis of different patatin isoforms expressed in various organs of potato. Plant Science 66: 221–230.
- 36. Stöckhaus J, Schell J, Willmitzer L (1989) Correlation of the expression of the nuclear biosynthetic gene ST-LS1 with the presence of chloroplasts. EMBO Journal 8: 2445–2451.
- 37. Logemann J, Schell J, Willmitzer L (1987) Improved method for the isolation of RNA from plant tissues. Annals of Biochemistry 163: 16–20.
- 38. García-Martínez JL, Santes C, Croker SJ, Hedden P (1991) Identification, quantitation and distribution of gibberellins in fruits of Pisum sativum L. cv. Alaska during pod development. Planta 184: 53–60.
- 39. García-Martínez JL, López-Díaz I, Sánchez-Beltrán MJ, Phillips AL, Ward DA, et al. (1997) Isolation and transcript analysis of gibberellin 20-oxidase genes in pea and bean in relation to fruit development. Plant Molecular Biology 33: 1073–1084.
- 40. Fujino K, Koda Y, Kikuta Y (1995) Reorientation of cortical microtubules in the subapical region during tuberization in single-node stem segments of potato in culture. Plant Cell Physiology 36: 891–895.
- 41. Xu X, Vreugdenhil D, van Lammeren AMM (1998b) Cell division and cell enlargement during potato tuber formation. Journal of Experimental Botany 49: 573–582.
- 42. Cyr RJ (1994) Microtubules in plant morphogenesis: Role of the cortical array. Annual Review of Cell Biology 10: 153–180.
- 43. Shibaoka H (1994) Plant hormone-induced changes in the orientation of cortical microtubules: Alterations in the cross-linking between microtubules and the plasma membrane. Annual Review of Plant Physiology and Plant Molecular Biology 45: 527–544.
- 44. Stupar RM, Beaubien KA, Jin W, Song J, Lee M-K, et al. (2006) Structural Diversity and Differential Transcription of the Patatin Multicopy Gene Family During Potato Tuber Development. Genetics 172: 1263–1275.
- 45. Björklund S, Antti H, Uddestrand I, Moritz T, Sundberg B (2007) Cross-talk between gibberellin and auxin in development of Populus wood: gibberellin stimulates polar auxin transport and has a common transcriptome with auxin. The Plant Journal 52: 499–511.
- 46. Eriksson S, Bahlenius H, Moritz T, Nilsson O (2006) GA4 Is the Active Gibberellin in the Regulation of LEAFY Transcription and Arabidopsis Floral Initiation. The Plant Cell Online 18: 2172–2181.
- 47. Proebsting WM, Hedden P, Lewis ML, Croker SJ, Proebsting LN (1992) Gibberellin concentration and transport in genetic lines of pea. Plant Physiology 100: 1354–1360.
- 48. Jackson SD, Heyer A, Dietze J, Prat S (1996) Phytochrome B mediates the photoperiodic control of tuber formation in potato. The Plant Journal 9: 159–166.