JXB Advance Access originally published online on October 18, 2006
Journal of Experimental Botany 2006 57(14):3801-3811; doi:10.1093/jxb/erl152
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RESEARCH PAPER |
Transport and metabolism of raffinose family oligosaccharides in transgenic potato
1 Max-Planck-Institut für Molekulare Pflanzenphysiologie, D-14424 Potsdam, Germany
2 Biologisches Institut, Abt. Botanik, Universität Stuttgart, Pfaffenwaldring 57, D-70569 Stuttgart, Germany
* To whom correspondence should be addressed. E-mail: arnd.heyer{at}bio.uni-stuttgart.de
Received 27 July 2006; Accepted 28 July 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Raffinose family oligosaccharides (RFOs) are involved in the storage and transport of carbon and serve as compatible solutes for protection against abiotic stresses like drought or cold. RFOs are usually transported in plant species that load sugars symplastically into the phloem. Loading probably occurs by a polymer trapping mechanism which establishes a concentration gradient of assimilates between the mesophyll and the vasculature. Transgenic approaches have demonstrated phloem transport of small molecules produced in the companion cells of apoplastic loading species, but these molecules have been non-native transport substances to plants. In this study, transgenic potato plants with constitutive or companion cell specific overexpression of galactinol synthase (GS) or GS plus raffinose synthase (RS) are characterized, which together provide new insights into the metabolism and transport of RFOs in plants. It is demonstrated that raffinose and galactinol are both transported in the phloem and that, whilst the effect of GS overexpression is promoter-independent, that of RS is dependent on the promoter used. The presence of significant amounts of galactinol in the phloem is shown and also that transgenic potato is unable to transport large amounts of raffinose despite high RS expression and substrate concentrations. These data indicate that there may be additional features of intermediary cells, the specialized companion cells of RFO transporting plants, required for significant RFO synthesis and transport that are currently not well-understood.
Key words: Galactinol, phloem loading, potato, raffinose family oligosaccharides, sugar transport
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Raffinose family oligosaccharides (RFOs) have diverse roles in plants, being used for the transport and storage of carbon and as compatible solutes for protection against abiotic stress (Bachmann et al., 1994; Haritatos et al., 1996; Taji et al., 2002). They are synthesized from sucrose by the sequential addition of galactose moieties donated by galactinol. The trisaccharide raffinose and tetrasaccharide stachyose are the first and best-studied products of this pathway and their production is catalysed by raffinose synthase (RS) and stachyose synthase (SS), respectively (Peterbauer and Richter, 2001). Galactinol is formed from UDP-galactose and myo-inositol via the activity of galactinol synthase (GS) which is considered a key regulator in the pathway (Keller and Pharr, 1996). Expression of GS isoforms is increased during drought and cold exposure in Arabidopsis and GS overexpression increased drought tolerance (Liu et al., 1998; Taji et al., 2002; Cunningham et al., 2003). The presence of high raffinose levels during abiotic stress is well-established, however, raffinose was recently demonstrated to be neither necessary nor sufficient for improved freezing tolerance in Arabidopsis (Zuther et al., 2004a). In many plant species, raffinose is also stored in high amounts in the seeds where it is thought to have an additional role in desiccation tolerance (Obendorf, 1997). Although there is evidence for a correlation between GS activity and the RFO content of plant organs, and GS overexpression increases RFO content in most cases, there are some examples indicating that alternative factors can modulate RFO accumulation (Karner et al., 2004). Karner and colleagues showed that, rather than GS activity alone, the concentration of the initial substrates myo-inositol and sucrose control RFO accumulation in seeds.
Sucrose-transporting type 2 plants like potato and Arabidopsis are characterized by low frequencies of plasmodesmata between mesophyll and companion cells, and phloem loading of sucrose occurs by an apoplastic route via proton symport (Turgeon, 1996). RFOs are transported in type 1 plants such as cucurbits, which have high numbers of plasmodesmata and load assimilates via a symplastic route proposed to involve a polymer trap mechanism (Turgeon, 1996). Sucrose is thought to diffuse through the numerous plasmodesmata linking mesophyll and companion cells and is converted into RFOs, particularly stachyose, which are then too large to diffuse back (Turgeon, 1996). In these species there are specialized, structurally distinct, companion cells with a high frequency of plasmodesmata called intermediary cells where RFO synthesis occurs (Beebe and Turgeon, 1992; Turgeon, 1996; Turgeon et al., 1993; Turgeon and Medville, 2004). Other type 1 plant species without intermediary cells do not transport high amounts of RFOs and have been shown to load via the apoplast (Turgeon and Medville, 2004). The presence of sucrose in the phloem of some RFO transporting species and the absence of galactinol transport, despite its high concentration in intermediary cells, were highlighted as aspects of the polymer trap hypothesis that are not fully understood (Turgeon, 1996). Recently, it was proposed that small molecules in the companion cells enter the sieve elements indiscriminately but that differential retention and retrieval mechanisms are responsible for differences in the efficiency of their transport (Ayre et al., 2003). Labelling studies in Cucumis blumei, which transports RFOs, showed that galactinol was transferred to the apoplast and had limited phloem transport whilst sucrose, raffinose and particularly stachyose were effectively retained and transported (Ayre et al., 2003). Companion cell specific expression of galactinol synthase in tobacco, a type 2 species, led to only minor transport of galactinol (Ayre et al., 2003). Ayre and colleagues used a GS promoter from Cucumis melo to drive gene expression in the minor-vein companion cells of both transgenic tobacco and Arabidopsis although neither of these plants normally transport RFOs (Haritatos et al., 2000). This implies that the promoter responds to a minor-vein-specific regulatory cascade that is highly conserved across a broad range of eudicotyledons. Quantification of RFO in the cytoplasm of different cell types in C. melo indicates that although raffinose and stachyose are massively enriched in intermediary cells, galactinol is likely to be evenly distributed throughout the leaf (Haritatos et al., 1996), probably located in the vacuole (Voitsekhovskaja et al., 2006). In Cucurbita pepo the enzymes of RFO biosynthesis have all been localized to intermediary cells where synthesis predominantly occurs, although RFO synthesis can be detected in the mesophyll (Beebe and Turgeon, 1992). The mesophyll may also be the main site for RFO synthesis as demonstrated for Ajuga reptans, however, this species also stores large amounts of RFOs in the leaf (Sprenger and Keller, 2000).
Using transgenic plants, phloem transport of small molecules produced specifically in the companion cells has been demonstrated for the non-native small molecule octopine and the fructan 6-kestose formed by the action of octopine synthase and invertase, respectively (Ayre et al., 2003; Zuther et al., 2004b). The establishment of companion cell specific production or transport of RFO in transgenic plants has not been investigated. In this study, transgenic potato plants with constitutive or companion cell-specific overexpression of either GS or GS plus RS are characterized, which together provide new insights into the metabolism and transport of RFOs in plants.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Plant material
The coding region from a GS cDNA from Cucumis sativus was isolated and cloned into the binary vector pBin AR (pBin19 with 35S promoter) (Bevan, 1984) as described (Zuther et al., 2004a). In addition, the same fragment was cloned into rolC-pBin19, a derivative containing the Agrobacterium tumefaciens rolC companion cell-specific promoter (Kuhn et al., 1996). These constructs were used to transform Solanum tuberosum cv. Désirée. Plants were screened for GS expression by quantitative RT-PCR and by carbohydrate analysis. Selected GS overexpressing plants were retransformed with RS constructs. Briefly, a 2355 bp sequence (ATG to stop) coding for C. sativus RS was PCR amplified from cDNA and TOPO-TA cloned into the pCR®2.1 vector (Invitrogen). The fragment was excised and non-directionally cloned into the Asp718 site of both pBinAR-HYG (hygromycin resistance in plants) (Dormann and Benning, 1998) and the derivative rolC-pBinAR-HYG (rolC instead of 35S promoter). Orientation was checked by restriction digest and the fragment was sequenced. Comparison to the database sequence for C. sativus RS (Genbank accession AF073744 [GenBank] ) revealed three base changes (T-C, A-T, and G-A at 500, 1598, and 1602 bp from ATG, respectively) resulting in the amino acid changes F–L, N–Y, and D–N, respectively at amino acids 133, 499, and 500. However, as the amino acids identified are conserved in RS from both Glycine max and Vicia faba it was concluded that these must represent errors in the database sequence for C. sativus (see Peterbauer et al., 1999, for sequence alignment).
Plants were grown in tissue culture on MS medium (Murashige and Skoog, 1962) containing 2% sucrose, under 16/8 h light/dark regime. For greenhouse experiments, plants were transferred into 20 cm pots and cultivated under high light conditions (400–600 µmol m–2 s–1) with a light/dark cycle of 16/8 h at 20/17 °C. Phloem exudation and leaf sugars were measured using 10–12-week-old plants harvested around midday. Plant and tuber biomass and tuber sugars were determined after 12 weeks of growth. Grafted plants were harvested approximately 2-weeks later due to the time taken for graft recovery. A completely randomized block design was used for all experiments.
Phloem exudate collection
Phloem exudation experiments were performed using slight modifications of a previously described method (Zuther et al., 2004b). Briefly, leaves were cut from the plant, immediately cut again in 5 mM EDTA and then placed into vials containing 10 ml of 5 mM EDTA (pH 7.5). Leaf-containing vials were placed in trays covered with a transparent cover to prevent excess transpiration and kept under high light conditions at 20 °C for 6 h. Transpiration was measured by volume changes of the EDTA solution and sugar amounts were corrected to the final volume. Exudation is expressed on the basis of leaf fresh weight to correct for size differences.
Grafting
Four-week-old plants were grafted as described by Balachandran et al. (1995) and Zuther et al. (2004b). However, instead of using tape to support grafts, two perpendicular metal clips were used as previously described for bean (Hannah et al., 2000), giving an improved survival rate of 100%. Plants were covered with plastic bags for 14 d to minimize transpiration. Rootstocks were covered with aluminium foil to prevent leaf growth and photosynthesis. Autografts were avoided by using scions from different plants of the same line. Nomenclature of the graft combinations is scion/rootstock.
Carbohydrate analysis
Sugar analysis was performed as described previously (Zuther et al., 2004a, b). Samples were immediately frozen in liquid nitrogen, homogenized using a ball mill and extracted twice in 1 ml of 80% ethanol at 80 °C. Extracts were dried, dissolved in 1 ml of distilled water and then deionized (AG 501-X8 resin, Bio-Rad). Glucose, fructose, sucrose, and raffinose were separated using a CarboPac PA-100 column, and galactinol using a MA-1 column, for high-performance anion exchange chromatography (HPAEC) using a Dionex DX-500 gradient chromatography system coupled with pulsed amperometric detection (PAD) by a gold electrode.
Expression analysis
RNA extraction and expression analysis by quantitative RT-PCR was as described previously (Rohde et al., 2004; Zuther et al., 2004a). CsGS primers were as described by Zuther et al. (2004a), whilst CsRS primers were: CsRS-F, 5'-TTCGATAGACAAGTCCCCGGT-3'; CsRS-R, 5'-CCCAATCGAAACAACATGTCG-3'. PCRs were performed using a Gene Amp 5700 Sequence Detection System using a 1-step QuantiTect SYBR Green RT-PCR kit (Qiagen). An equal amount of total RNA (50 ng) was used for all reactions and threshold cycle numbers were directly compared between wild-type and transformants as it was found that normalization to a reference gene did not affect the results (data not shown).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Constitutive non-specific or companion cell-specific expression of GS in potato
GS from C. sativus was constitutively overexpressed in transgenic potato under the control of the CaMV 35S promoter (35S:GS) and the transformants compared to a set of transgenic plants expressing GS under the control of the Agrobacterium tumefaciens rolC promoter (rolC:GS), which has previously been shown to direct companion cell specific expression (Kuhn et al., 1996). Plants were screened for leaf galactinol and independent lines were selected.
The amounts of sucrose, glucose, and fructose, and of galactinol and raffinose were determined in leaves, tubers, and phloem exudates of 10–12-week-old 35S:GS, rolC:GS, and wild-type (WT) potato plants by HPAEC coupled with pulsed amperometric detection. There were no significant changes in the content of sucrose, glucose, or fructose in any tissue in 35S:GS or rolC:GS lines. Raffinose was only occasionally detected and changes were not consistently observed in 35S:GS or rolC:GS lines. Galactinol, however, accumulated to high levels in several independent lines and was present at equal levels in the leaves of both 35S:GS and rolC:GS plants, reaching concentrations in the same range as sucrose (6–8 µmol g–1 FW) (Fig. 1A). Galactinol was also increased in the tubers but to a lesser degree (Fig. 1C), typically 5–10% of the tuber sucrose concentration and around 10% of the concentration detected in the leaves.
|
To investigate assimilate transport, sugars were measured in phloem exudates collected from the petioles of cut leaves placed into 5 mM EDTA for 6 h. Sucrose was the main transported sugar in potato, and glucose and fructose, which were both present at equal concentration (approximately 10–20% of the sucrose concentration) were highly correlated with sucrose levels. It has previously been shown in Arabidopsis that the presence of glucose and fructose in phloem exudates is probably due to sucrose cleavage by extracellular invertase (Corbesier et al., 1998). Therefore, sucrose exudation was calculated from sucrose plus the average of glucose and fructose concentration. Galactinol was easily detected in phloem exudates from all 35S:GS and rolC:GS lines and exuded at a rate of 3–5 nmol g–1 FW h–1 (Fig. 1B). Average sucrose exudation was much higher at 550 nmol g–1 FW h–1. Galactinol exudation was significantly (e.g. for rolC:GS P=1x10–15, n=72) correlated with that of sucrose both in 35S:GS and rolC:GS lines.
Constitutive non-specific or companion cell-specific expression of RS
Consistent with the finding that potato does not produce significant amounts of RFOs, GS overexpression alone was not sufficient to produce much raffinose in any tissue, while galactinol accumulated to significant amounts. To achieve raffinose production, the GS expressing lines 35S:GS 41 and rolC:GS 25 and 31 were retransformed with RS from C. sativus under the control of the same promoters to produce the genotypes 35S:GS/RS and rolC:GS/RS. Several independent lines for each genotype were selected based on quantitative RT-PCR followed by rescreening for raffinose content among those with the highest expression. Leaves were harvested and pooled from 5-week-old plants for six independent transformants for each line and leaf raffinose was measured. There was a large increase in raffinose in all 35S:GS/RS lines ranging from 4–15-fold of the parental background (Fig. 2). All 12 independent rolC:GS/RS lines were very similar to their parental lines with the maximal raffinose increase just 1.3-fold (Fig. 2).
|
As for the GS-expressing plants, tissue content and transport of sugars were analysed in 35S:GS/RS and rolC:GS/RS plants and compared with the parental lines. The best four 35S:GS/RS lines and the best five rolC:GS/RS lines were selected from the initial screen for these measurements. No significant changes were detected in any tissue for sucrose, glucose, or fructose. However, raffinose accumulation was easily detected in leaves of 35S:GS/RS lines, reaching levels of 1 µmol g–1 FW (Fig. 3A). By contrast, no significant increase in leaf raffinose was observed in any of the rolC:GS/RS lines.
|
Although raffinose was not detectable in rolC:GS/RS leaves, two rolC:GS/RS lines, 25-39 and 31-38, had significant amounts of raffinose in phloem exudates. Levels of phloem raffinose in these lines were similar to those found in the 35S:GS/RS lines 41-10 and 41-19, while line 41-28 contained around 50% more (data not shown). Raffinose was correlated with sucrose exudation both in the 35S:GS/RS and rolC:GS/RS lines.
To investigate raffinose accumulation, leaves at different developmental stages were harvested from the representative lines 35S:GS/RS 41-19 and rolC:GS/RS 31-38. Four leaf types were harvested; the youngest fully unfolded leaf (leaf 1–2 on the main stem), a still expanding leaf at three-quarters plant height (leaf 5–6, sampled in previous experiment), a mature source leaf at half plant height and one of the oldest source leaves at one-quarter plant height. In 35S:GS/RS 41-19 plants the expanding leaf at three-quarters height contained the most raffinose, around twice as much on a FW basis than the older, fully expanded, source leaves sampled from the lower half of the same plants and 60% more than the youngest unfolded leaf (Table 1). In rolC:GS/RS 31-38 plants the maximal raffinose accumulation was observed in the youngest unfolded leaf which contained more than twice as much on a fresh weight basis than any of the lower leaves (Table 1). The levels of raffinose detected were significantly lower in rolC:GS/RS 31-38 than 35S:GS/RS 41-19 plants ranging from around 5–12% depending on the leaf sampled (Table 1).
|
As the primary aim was to identify plants with substantial transport of raffinose, this screening was repeated on additional rolC:GS/RS lines: and the lines 25-39 and 31-38 were reanalysed for leaf raffinose by sampling younger leaves (leaf 3–4 instead of 5–6) which seemed likely to contain higher levels of raffinose in rolC:GS/RS plants. There was a significant increase in leaf raffinose in several rolC:GS/RS lines, including the lines 25-39 and 31-38 (Fig. 4A). However, despite initially screening two independent sets of 50 rolC:GS/RS transformants for RS expression and growing the best 12 lines in multiple experiments, the maximum raffinose accumulation observed was only 20–30% of that in 35S:GS/RS plants and, in some cases, it was not above the background levels that were frequently detected in parental lines (Fig. 4A).
|
The high abundance of raffinose in 35S:GS/RS lines (Fig. 3A) made them suitable to investigate the spatio-temporal distribution of galactinol and raffinose within leaves further. In addition to the previously mentioned harvesting of different leaves from 35S:GS/RS 41-19 plants, samples were harvested from 35S:GS/RS 41-19 and its parent 35S:GS 41 during the light phase of a diurnal cycle (16/8 h light/dark). Galactinol was stable throughout the time-course for both 35S:GS 41 and 35S:GS/RS 41-19 plants, but the level was significantly lower in 35S:GS/RS 41-19 than in 35S:GS 41 plants (Table 2). In the time-course experiment, raffinose was significantly correlated with sucrose and galactinol, whilst during leaf development all three were significantly correlated with each other (Table 3). Galactinol was also positively correlated with raffinose among the four independent 35S:GS/RS lines investigated (Fig. 3A, B).
|
|
Several independent rolC:GS/RS lines were used which had measurable leaf raffinose (Fig. 4A) to characterize RFO metabolism and transport further. By contrast with 35S:GS/RS plants, a significant decrease of leaf galactinol levels in rolC:GS/RS lines in comparison to the parents (compare Figs 3B, 4B) was never detected. In rolC:GS/RS tubers, raffinose was present at a similar concentration to the leaves, although parental background levels were much lower. Different from the situation in leaves, tuber galactinol was significantly reduced to around 50% of the level observed in rolC:GS plants, which was then a concentration similar to that of raffinose (Fig. 4C, D). Raffinose was readily detected in phloem exudates and there was a reduction in galactinol as indicated by a 10–25% decrease in the ratio of galactinol to sucrose in the phloem in all rolC:GS/RS lines (Fig. 4F). In common with galactinol, raffinose in phloem exudates was significantly (P=1.7x10–13, n=40) correlated with sucrose exudation.
To investigate long-distance raffinose transport, reciprocal grafting experiments were performed between scions and rootstocks of WT and two independent rolC:GS/RS lines. The two rolC:GS lines used for retransformation with the RS construct were also included in this experiment. Galactinol was significantly increased in tubers in all cases when scions overexpressing GS (rolC:GS or rolC:GS/RS) were grafted onto WT stocks (Fig. 5B). Tuber raffinose was higher in rolC:GS/RS self grafts than in tubers of rolC:GS/RS stocks grafted to WT scions (Fig. 5A), however, tuber raffinose was not increased in WT stocks grafted to rolC:GS/RS scions (Fig. 5B). The reduction in galactinol transport indicated by the decreased galactinol:sucrose ratio in rolC:GS/RS plants was supported by a reduction of galactinol concentration in the tubers of WT stocks grafted to rolC:GS/RS scions in comparison to rolC:GS (Fig. 5B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Constitutive non-specific or companion cell-specific expression of GS have similar effects
Tobacco plants with companion cell specific expression of GS were shown to accumulate galactinol in the leaf, but only small amounts were transported (Ayre et al., 2003). Companion cell-specific GS expression was compared with non-specific overexpression in potato to identify specific effects of producing galactinol in the companion cell, a tissue specialized for the loading of sucrose into the phloem and the metabolic support of the sieve elements. The typically used CaMV 35S promoter was chosen for constitutive non-specific overexpression, because it has been shown to have high activity in vascular cells (Jefferson et al., 1987) and many other tissues (Benfey et al., 1989; Stockhaus et al., 1989). As a consequence, 35S transformants would display transgene expression in the vasculature and in additional cell types of the leaf, thus allowing identification of the effects of companion cell specific expression brought about by the Agrobacterium rolC promoter (Kuhn et al., 1996). The CaMV 35S promoter has been shown to drive expression of the ß-glucuronidase reporter gene in most tissues in 7-week-old tobacco plants, for example, in mesophyll, vascular and epidermal cells of leaves, the cortex, pith, and vascular cells of the stem, and vascular tissue, pericycle and most other cells of the root (Benfey et al., 1989). In potato and tobacco CaMV 35S driven ß-glucuronidase expression was detected throughout different leaf cell types, while, in tobacco, expression was significantly higher in vascular tissue than cortex or parenchyma in the stem (Stockhaus et al., 1989). By contrast, the Agrobacterium tumefaciens rolC promoter is companion cell specific, driving expression in the phloem of tobacco leaves, stems, and roots (Schmulling et al., 1989). Immunolocalization has revealed that rolC activity is confined to the companion cells in potato (Kuhn et al., 1996).
Several independent transformants were screened and selected for both 35S:GS and rolC:GS. Similar to the results of companion cell specific expression reported by Ayre et al. (2003), galactinol accumulated to high levels in the leaves of the rolC:GS plants, reaching concentrations in the same range as sucrose (Fig. 1A). Surprisingly, despite being expressed in many more cell types, galactinol accumulation was no higher in the leaves of 35S:GS plants. Furthermore, galactinol was also present at similar concentrations in tubers and in phloem exudates of 35S:GS and rolC:GS plants (Fig. 1B, C). Such equal accumulation could be explained if the rolC promoter specificity was compensated for by a much higher activity. However, given the small proportion of companion cells, and the established high activity of the CaMV 35S promoter, this seems unlikely. As promoter activity was not monitored under the growth conditions used in this study it cannot be excluded that the CaMV 35S and rolC promoters had a similar expression pattern, with the observed equal accumulation due to the majority of GS being expressed in the companion cells. This seems unlikely based on the extensive literature detailing CaMV 35S expression in most plant organs at most stages of development in solanaceous species (Benfey et al., 1989). The high galactinol accumulation in response to companion cell specific GS expression, shown here and previously (Ayre et al., 2003), demonstrates the substrates for galactinol synthesis myo-inositol and UDP-galactose are readily available and mobile within the leaf. Thus, the similar levels of galactinol observed in 35S:GS and rolC:GS lines probably reflect a limitation of total galactinol production by the availability of GS substrates. The accumulation of galactinol in the leaf of rolC:GS plants is in agreement with its proposed high membrane permeability and leakage out of phloem tissue (Ayre et al., 2003). In addition, the presence of galactinol at equal concentrations in both phloem exudates and leaf samples of 35S:GS and rolC:GS plants suggests that galactinol may also move into the phloem compartment from the mesophyll.
Using pulse-chase labelling with 14C-CO2, Haritatos and co-workers did not detect galactinol in wild-type tobacco leaves after a 30 min pulse (Haritatos et al., 2000), but following substantial galactinol production from GS overexpression small amounts of raffinose could be detected (Ayre et al., 2003). Small amounts of galactinol were found in wild-type (WT) potatoes using HPAEC coupled with pulsed amperometric detection, but raffinose was usually absent in WT and was not consistently increased in 35S:GS and rolC:GS transgenic plants. Given that galactinol production in the short-term is low, as was demonstrated for tobacco (Ayre et al., 2003), the small amounts we detected may be part of a stable pool maintained by low residual activities of RFO biosynthetic enzymes in potato. The inability of potato to metabolize galactinol in significant quantities would, in turn, explain the accumulation to high levels in 35S:GS and rolC:GS plants. Indeed, there are S. tuberosum EST sequences that are similar to GS and RS from other species (e.g. GenBank accessions BG886520 [GenBank] and DR751816 [GenBank] , respectively), indicating low level gene expression, and small amounts of galactinol and raffinose have been detected in potato leaves and tubers by GCMS analysis (Keller et al., 1998). Despite the usually low levels of galactinol, its production and accumulation to high concentrations did not result in obvious phenotypic alterations except for a slight non-significant reduction in biomass in some lines (data not shown).
The effect of RS expression on leaf raffinose is promoter specific
Although 35S:GS overexpression was sufficient to increase raffinose content in Arabidopsis (Taji et al., 2002; Zuther et al., 2004a), a naturally RFO containing species, it did not cause significant raffinose accumulation in potato. As discussed above, in tobacco, a non-RFO-accumulating species, small amounts of raffinose could be detected in tobacco plants overexpressing GS (Ayre et al., 2003), thus indicating a lower but detectable RS activity. However, in potato, GS overexpression did not consistently produce measurable amounts of raffinose in any tissue. RFO have been detected as antinutritional factors in other Solanaceae, for example, in Capsicum annuum (El-Adawy and Taha, 2001) and may therefore have been reduced in cultivated potato by conventional breeding.
Establishing the RFO synthesis pathway by ectopic expression of both GS and RS enzymes has not been previously reported, and it is shown here that the result of this transformation is largely dependent on the promoter used. Retransformation of representative 35S:GS and rolC:GS with RS from C. sativus under the control of the same promoters led to the readily detectable accumulation of raffinose in leaves of 35S:GS/RS lines (Figs 2, 3A), but not in rolC:GS/RS leaves measured in the same experiment (Fig. 2). Furthermore, despite screening the best lines from 100 independent rolC:GS/RS transformants, growing the selected lines in multiple experiments and harvesting leaves at a developmental stage allowing high raffinose accumulation, the maximum level observed was only 20–30% of that in 35S:GS/RS plants (Fig. 4A). Direct comparison between leaves at different developmental stages indicated that raffinose was 8–20-fold lower in the line rolC:GS/RS 31-38 compared with 35S:GS/RS 41-19 (Table 1). There was also a striking reduction in the levels of leaf galactinol in 35S:GS/RS lines, but there was never any significant reduction for rolC:GS/RS lines (Fig. 4B), even when leaf raffinose was clearly detectable (Fig. 4A). One obvious explanation for this is that galactinol produced in companion cells is mobile in the entire leaf and builds a large stable pool mostly unavailable to a cytoplasmic RS located in companion cells. Non-aqueous fractionation of cell compartments has shown that galactinol is, to a large extent, located in the vacuole in species with different phloem loading strategies (Voitsekhovskaja et al., 2006). A vacuolar localization in the mesophyll of potato overexpressing GS would be consistent with it being largely inaccessible to RS localized in the cytosol of companion cells of rolC:GS/RS plants yet more available to RS present in the mesophyll cells of 35S:GS/RS plants.
The high abundance of raffinose in 35S:GS/RS lines (Fig. 3A) made them suitable to investigate the spatio-temporal distribution of galactinol and raffinose within leaves. In a diurnal time-course experiment leaf raffinose in 35S:GS/RS plants was significantly correlated with sucrose and galactinol, whilst during leaf development raffinose, sucrose and galactinol were all significantly correlated with each other (Table 3). Raffinose and galactinol were also positively correlated among independent 35S:GS/RS lines (Fig. 3A, B). These data most likely reflect the availability of assimilate, but they also support the assertion of the key role of GS in controlling RFO production and the more recent report that the initial substrates, including sucrose, may limit RFO production (Karner et al., 2004). Sugar distribution during leaf development was also determined for rolC:GS/RS lines, however, in this case the correlation with sucrose levels was weaker and there was no correlation with galactinol levels (Table 1; data not shown). This may be expected, as correlation between galactinol and raffinose in the companion cells would likely be masked by the large stable galactinol pool within the leaf. Another possible reason for the differences in raffinose content between different leaf development stages could be developmental regulation of the promoter activity. There is no sufficient data to assess this possibility, particularly as the galactinol concentration, which is similar among all leaves sampled, is not a reliable marker of promoter activity due to its varying stability, membrane permeability and transport.
Galactinol transport in transgenic potato
Clear evidence was found for the transport of galactinol in all lines overexpressing GS. Galactinol was easily detected in phloem exudates and its concentration was significantly correlated with that of sucrose, the major transport sugar in potato, thus demonstrating that the galactinol is co-transported with the stream of photoassimilates. In addition, grafting experiments showed that tuber galactinol was significantly increased in all cases when rolC:GS or rolC:GS/RS scions were grafted onto WT stocks (Fig. 5B). Compared with sucrose, galactinol generally represented about 1% of the transported carbon in potato expressing GS (Fig. 4F) and in some experiments reached 2% of the total. This is substantially higher than observed in RFO transporting species. In Alonsoa meridionalis, which transports RFOs, galactinol accounts for just 0.2% of the carbon present in the phloem sap (sucrose, raffinose and stachyose=98%) (Voitsekhovskaja et al., 2006). The low concentration of certain sugars in the phloem was recently proposed to result from differential retention and retrieval of different sugars (Ayre et al., 2003). These data support this but indicate that, in addition, specialized features of the intermediary cells in RFO-transporting species may contribute to reducing the concentration of galactinol in the phloem. The vacuolar localization of galactinol (Ayre et al., 2003) and the presence of many small vacuoles in intermediary cells, presumably with a large surface area, may serve to buffer galactinol concentrations in the cytoplasm. The location of GS in the intermediary cell (Beebe and Turgeon, 1992; Haritatos et al., 2000) may also be related to the need for recycling myo-inositol from raffinose and stachyose synthesis rather than a direct need for galactinol production in these cells instead of the leaf mesophyll. A limitation on the substrates for galactinol production is supported by the equal accumulation of galactinol in rolC:GS and 35S:GS lines previously discussed. These data show that galactinol accumulates within the leaf and is exuded from the phloem at the same rate in 35S:GS and in rolC:GS plants (Fig. 1A, B) indicating that galactinol is likely to be mobile within the leaf. As the CaMV 35S promoter drives expression in vascular tissue it cannot be excluded that galactinol produced in the phloem is sufficient to account for the observed export in 35S:GS plants. This is supported by the observation that, although there was a large difference in leaf raffinose, phloem raffinose was generally similar between 35S:GS/RS and rolC:GS/RS lines (data not shown). This could indicate that the CaMV 35S and rolC promoters have similar activities in the companion cells. However, 35S promoter activity is similar in the mesophyll and phloem of the leaf (Stockhaus et al., 1989) and is detected in most leaf cell types (Benfey et al., 1989), and mesophyll cells vastly outnumber those of the phloem. The pronounced difference in leaf raffinose between 35S:GS/RS and rolC:GS/RS plants supports the established differences in CaMV 35S and rolC promoter activities (Benfey et al., 1989; Schmulling et al., 1989; Stockhaus et al., 1989; Kuhn et al., 1996). It is possible that the mechanisms leading to the similarities of phloem galactinol and raffinose are distinct. Alternative routes of raffinose entry into the phloem or the metabolism of sugars in phloem exudates by cells at the cut petiole surface in 35S:GS/RS plants are candidates for such differences.
Alternatively, cell-type influences on GS and RS enzyme activities may also have an effect, but as these activities are co-localized in different tissues (Beebe and Turgeon, 1992; Peterbauer et al., 2001) this seems unlikely.
Raffinose transport in transgenic potato
Evidence was found for the transport of raffinose in lines overexpressing GS/RS, with raffinose detected in phloem exudates and its concentration correlated with that of sucrose in 35S:GS/RS and rolC:GS/RS plants. To reach further conclusions on raffinose transport, rolC:GS/RS plants were investigated in more detail. By contrast with galactinol, raffinose was present at much lower concentrations in the leaf (Figs 4A, B, 1A). However, despite being approximately 30 times lower in concentration than galactinol in the leaf, its concentration in phloem exudates was just 2-fold less than galactinol (Figs 4E, 1B). This is in agreement with the hypothesis that raffinose is more effectively retained within the phloem than galactinol leading to a higher transport efficiency (Ayre et al., 2003). Strong evidence for raffinose transport was provided by using data from the large number of independent rolC:GS/RS lines in these experiments which revealed a highly significant (P=1.7x10–13, n=40) correlation of raffinose with sucrose exudation. In addition, tuber raffinose was higher in rolC:GS/RS self grafts than in tubers of rolC:GS/RS stocks grafted to WT scions indicating a contribution of shoot supplied raffinose (Fig. 5A). However, tuber raffinose was not increased in WT stocks grafted to rolC:GS/RS scions (Fig. 5B). The reason for this is unknown, but may indicate that RS activity in companion cells of the stems or tubers could saturate or counteract raffinose metabolism by endogenous enzymes during transport or tuber development in rolC:GS/RS plants. These data indicate that the transport of raffinose itself was successful, complementing previous work showing the transport of octopine (Ayre et al., 2003) and 6-kestose (Zuther et al., 2004b) in the phloem of an apoplastic phloem loader and are the first involving a compound which is naturally transported in plants.
In parallel to the transport of raffinose, there was evidence for the metabolism of galactinol in companion cells. In rolC:GS/RS tubers, galactinol was significantly reduced in several lines and was present at concentrations similar to raffinose (Fig. 4C, D). The ratio of galactinol to sucrose in the phloem was reduced by 10–25% in all rolC:GS/RS lines (Fig. 4F). Finally, galactinol concentration was lower in the tubers of WT stocks grafted to rolC:GS/RS scions compared with rolC:GS, which also indicates that galactinol transport was probably reduced due to its conversion to raffinose (Fig. 5B). However, the continued high accumulation of galactinol in the leaves and its exudation from the phloem indicates that RS activity was not sufficient to convert all galactinol to raffinose (Fig. 4B). Galactinol is also seen in the same concentration range as sucrose in the leaves of RFO transporting species (Ayre et al., 2003; Voitsekhovskaja et al., 2006), but less galactinol is transported in these plants. In the transgenic potatoes, raffinose accounted for a very small proportion of the transported carbon, just 0.5% compared with sucrose, indicating that high expression of RS and the availability of its substrates galactinol and sucrose in companion cells are not sufficient to establish RFO transport as a major pathway in an apoplastic phloem loader. This may indicate that discrimination between galactinol and transport sugars is not working in companion cells of apoplastic loaders allowing entry of galactinol into the phloem, which in turn results in limited galactinol for raffinose synthesis. However, it seems most likely that efficient RFO synthesis additionally requires stachyose synthase or other specific features of intermediary cells for the temporal or spatial coupling of RFO biosynthetic enzyme activity and substrate concentrations.
|
Acknowledgements |
|---|
We thank U Seider for excellent technical assistance. This work was supported by a grant from the Deutsche Forschungsgemeinschaft (SFB 429C2).
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Ayre BG, Keller F, Turgeon R. (2003) Symplastic continuity between companion cells and the translocation stream: long-distance transport is controlled by retention and retrieval mechanisms in the phloem. Plant Physiology 131:1518–1528.
Bachmann M, Matile P, Keller F. (1994) Metabolism of the raffinose family oligosaccharides in leaves of Ajuga reptans L. (cold acclimation, translocation, and sink to source transition: discovery of chain elongation enzyme). Plant Physiology 105:1335–1345.[Abstract]
Balachandran S, Hull RJ, Vaadia Y, Wolf S, Lucas WJ. (1995) Alteration in carbon partitioning induced by the movement protein of tobacco mosaic-virus originates in the mesophyll and is independent of change in the plasmodesmal size-exclusion limit. Plant, Cell and Environment 18:1301–1310.
Beebe DU and Turgeon R. (1992) Localization of galactinol, raffinose, and stachyose synthesis in Cucurbita pepo leaves. Planta 188:354–361.
Benfey PN, Ren L, Chua NH. (1989) The CaMV 35S enhancer contains at least two domains which can confer different developmental and tissue-specific expression patterns. The EMBO Journal 8:2195–2202.[ISI][Medline]
Bevan M. (1984) Binary Agrobacterium vectors for plant transformation. Nucleic Acids Research 12:8711–8721.
Corbesier L, Lejeune P, Bernier G. (1998) The role of carbohydrates in the induction of flowering in Arabidopsis thaliana: comparison between the wild type and a starchless mutant. Planta 206:131–137.[CrossRef][ISI][Medline]
Cunningham SM, Nadeau P, Castonguay Y, Laberge S, Volenec JJ. (2003) Raffinose and stachyose accumulation, galactinol synthase expression, and winter injury of contrasting alfalfa germplasms. Crop Science 43:562–570.
Dormann P and Benning C. (1998) The role of UDP glucose epimerase in carbohydrate metabolism of Arabidopsis. The Plant Journal 13:641–652.[CrossRef][ISI][Medline]
El-Adawy TA and Taha KM. (2001) Characteristics and composition of watermelon, pumpkin, and paprika seed oils and flours. Journal of Agricultural and Food Chemistry 49:1253–1259.[CrossRef][ISI][Medline]
Hannah MA, Iqbal MJ, Sanders FE. (2000) The DL gene system in common bean: a possible mechanism for control of root–shoot partitioning. New Phytologist 147:487–496.[CrossRef]
Haritatos E, Ayre BG, Turgeon R. (2000) Identification of phloem involved in assimilate loading in leaves by the activity of the galactinol synthase promoter. Plant Physiology 123:929–937.
Haritatos E, Keller F, Turgeon R. (1996) Raffinose oligosaccharide concentrations measured in individual cell and tissue types in Cucumis melo L. leaves: implications for phloem loading. Planta 198:614–622.
Jefferson RA, Kavanagh TA, Bevan MW. (1987) GUS fusions: ß-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO Journal 6:3901–3907.[ISI][Medline]
Karner U, Peterbauer T, Raboy V, Jones DA, Hedley CL, Richter A. (2004) Myo-inositol and sucrose concentrations affect the accumulation of raffinose family oligosaccharides in seeds. Journal of Experimental Botany 55:1981–1987.
Keller F and Pharr DM. (1996) Metabolism of carbohydrates in sinks and sources: galactosyl-sucrose oligosaccharides. In Zamski E and Schaffer AA (Eds.). Photoassimilate distribution in plants and crops(Marcel Dekker, New York) pp. 115–184.
Keller R, Brearley CA, Trethewey RN, Muller-Rober B. (1998) Reduced inositol content and altered morphology in transgenic potato plants inhibited for 1D-myo-inositol 3-phosphate synthase. The Plant Journal 16:403–410.[CrossRef][ISI]
Kuhn C, Quick WP, Schulz A, Riesmeier JW, Sonnewald U, Frommer WB. (1996) Companion cell-specific inhibition of the potato sucrose transporter SUT1. Plant, Cell and Environment 19:1115–1123.
Liu J-JJ, Krenz DC, Galvez AF, de Lumen BO. (1998) Galactinol synthase (GS): increased enzyme activity and levels of mRNA due to cold and desiccation. Plant Science 134:11–20.
Murashige T and Skoog F. (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiologia Plantarum 15:473–479.[CrossRef]
Obendorf RL. (1997) Oligosaccharides and galactosyl cyclitols in seed desiccation tolerance. Seed Science Research 7:63–74.
Peterbauer T, Lahuta LB, Blochl A, Mucha J, Jones DA, Hedley CL, Gorecki RJ, Richter A. (2001) Analysis of the raffinose family oligosaccharide pathway in pea seeds with contrasting carbohydrate composition. Plant Physiology 127:1764–1772.
Peterbauer T, Mucha J, Mayer U, Popp M, Glossl J, Richter A. (1999) Stachyose synthesis in seeds of adzuki bean (Vigna angularis): molecular cloning and functional expression of stachyose synthase. The Plant Journal 20:509–518.[CrossRef][ISI][Medline]
Peterbauer T and Richter A. (2001) Biochemistry and physiology of raffinose family oligosaccharides and galactosyl cyclitols in seeds. Seed Science Research 11:185–197.
Rohde P, Hincha DK, Heyer AG. (2004) Heterosis in the freezing tolerance of crosses between two Arabidopsis thaliana accessions (Columbia-0 and C24) that show differences in non-acclimated and acclimated freezing tolerance. The Plant Journal 38:790–799.[CrossRef][ISI][Medline]
Schmulling T, Schell J, Spena A. (1989) Promoters of the rolA, B, and C genes of Agrobacterium rhizogenes are differentially regulated in transgenic plants. The Plant Cell 1:665–670.
Sprenger N and Keller F. (2000) Allocation of raffinose family oligosaccharides to transport and storage pools in Ajuga reptans: the roles of two distinct galactinol synthases. The Plant Journal 21:249–258.[CrossRef][ISI][Medline]
Stockhaus J, Schell J, Willmitzer L. (1989) Correlation of the expression of the nuclear photosynthetic gene ST-LS1 with the presence of chloroplasts. EMBO Journal 8:2445–2451.[ISI][Medline]
Taji T, Ohsumi C, Iuchi S, Seki M, Kasuga M, Kobayashi M, Yamaguchi-Shinozaki K, Shinozaki K. (2002) Important roles of drought- and cold-inducible genes for galactinol synthase in stress tolerance in Arabidopsis thaliana. The Plant Journal 29:417–426.[CrossRef][ISI][Medline]
Turgeon R. (1996) Phloem loading and plasmodesmata. Trends in Plant Science 1:418–423.[CrossRef]
Turgeon R, Beebe DU, Gowan E. (1993) The intermediary cell–minor-vein anatomy and raffinose oligosaccharide synthesis in the Scrophulariaceae. Planta 191:446–456.
Turgeon R and Medville R. (2004) Phloem loading. A re-evaluation of the relationship between plasmodesmatal frequencies and loading strategies. Plant Physiology 136:3795–3803.
Voitsekhovskaja OV, Koroleva OA, Batashev DR, Knop C, Tomos AD, Gamalei YV, Heldt HW, Lohaus G. (2006) Phloem loading in two Scrophulariaceae species. what can drive symplastic flow via plasmodesmata? Plant Physiology 140:383–395.
Zuther E, Buchel K, Hundertmark M, Stitt M, Hincha DK, Heyer AG. (2004a) The role of raffinose in the cold acclimation response of Arabidopsis thaliana. FEBS Letters 576:169–173.[CrossRef][ISI][Medline]
Zuther E, Kwart M, Willmitzer L, Heyer AG. (2004b) Expression of a yeast-derived invertase in companion cells results in long-distance transport of a trisaccharide in an apoplastic loader and influences sucrose transport. Planta 218:759–766.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||