УДК 581.1
Starch-Related Enzymes during Potato Tuber Dormancy and Sprouting1
© 2012 L. I. Sergeeva*,**,***2, M. M. J. Claassens*2, D. C. L. Jamar*,
L. H. W. van der Plas*, D. Vreugdenhil*,**
* Laboratory of Plant Physiology, Wageningen University, Wageningen, The Netherlands
** Centre for BioSystems Genomics, Wageningen, The Netherlands
*** Timiryazev Institute of Plant Physiology, Russian Academy of Sciences, Moscow, Russia
Received June 3, 2011
Activities of enzymes presumably involved in starch biosynthesis (ADP-glucose pyrophosphorylase, AGPase) and/or breakdown (starch phosphorylase, STP; amylases) were determined during potato (Solanum tuberosum L.) tuber dormancy and sprouting. Overall activities of all these enzymes decreased during the first stage of tuber dormancy. No clear changes were detected at the time of dormancy breaking and sprouting. However, when AGPase activity was monitored by in situ staining during the entire dormancy period, a clear decrease during the dormant period and a large increase before visible sprouting could be observed. This increase was especially evident near the vascular tissue and at the apical bud, which showed a very intensive staining. In situ staining of STP activity in sprouting tubers showed that the tissue distribution of STP was the same as for AGPase. As a possible explanation, direct starch cycling is suggested: STP produces glucose-1-phosphate during starch breakdown, which can be directly used as a substrate by AGPase for starch synthesis. Gene expression studies with the AGPaseS promoter coupled to the firefly luciferase reporter gene also clearly showed a higher activity in sprouting tubers as compared to dormant tubers, with the highest expression levels observed around the apical buds. The presence of amylase activity at dormancy initiation and AGPase activity persistent at the sprouting stage suggest that starch was cycling throughout the entire dormancy period. According to the in situ studies, the AGPase activity increased well before visible sprout growth and could therefore be one of the first physiological determinants of dormancy breakage.
----------------------------------
1 This text was submitted by the authors in English.
2 These authors contributed equally to the article.
----------------------------------
Abbreviations: AGPase  ADP-glucose pyrophosphorylase; AGPPase  ADP-glucose pyrophosphatase; BA  benzyladenine; NBT  nitroblue tetrazolium; 3PGA  3-phosphoglycerate; PVP  polyvinylpyrrolidone; STP  starch phosphorylase.
Corresponding address: L. I. Sergeeva. Laboratory of Plant Physiology, Wageningen University, Droevendaalsesteeg 1, 6708 PB Wageningen, The Netherlands. Fax: +31 317 418094; e-mail: lidiya.sergeeva@wur.nl
Keywords: Solanum tuberosum  dormancy  sprouting  ADP-glucose pyrophosphorylase  amylase  starch phosphorylase  starch cycling

INTRODUCTION
Potato is among the most important crop species. However, the regulation of potato tuber dormancy and sprouting, as part of potato life cycle, is still far away from clear understanding.
Potato tubers are formed underground at so-called stolons, which develop as lateral shoots at the base of the stems of potato plants. In order to form tubers, the plants have to become induced to tuberize, e.g., by photoperiod length, temperature, or nitrogen limitation. Tuber induction leads to tuber initiation and is followed by tuber growth and maturation. The tuber then enters a period of dormancy, which presumably is already initiated during tuber development [1]. The length and depth of this dormancy depend on cultivar and environmental factors occurring during tuber development and storage. At the end of the dormancy period, the tuber starts sprouting, and a new plant is formed, thus completing the vegetative life cycle of the potato plant.
Potato tubers can also form in vitro, and they mainly differ in size from soil-grown tubers. In many other aspects, e.g., the ultrastructure and the relative activities of enzymes involved in carbohydrate metabolism, such microtubers are similar to soil-grown tubers [2]. The developing potato tuber, a storage/sink organ by definition, accumulates starch during its development. Starch accumulation occurs already at the tuber initiation and is accompanied by an increase in enzymes involved in its biosynthesis [3]. At maturity, starch represents about 1525% of the tuber fresh weight. This reserve carbohydrate provides energy and substrates (glucans, maltose, and others) for future plant metabolism during sprouting. The former sink thus turns into a source: starch is broken down by diverse enzymes and used to support growth of the new plant [4].
The rate-limiting step in starch biosynthesis is the conversion of glucose-1-phosphate to the precursor for starch synthesis, ADP-glucose. This reversible conversion is catalyzed by the enzyme ADP-glucose pyrophosphorylase (AGPase) [5]. Starch degradation coincides with enhanced activities of α-amylase, β-amylase, starch phosphorylase (STP), maltase, and debranching enzymes, as well as the accumulation of their products. Among all these enzymes, STP catalyzes a reversible reaction: the conversion from starchn to starchn-1 and glucose-1-phosphate (using inorganic phosphate) and vice versa, implying that this enzyme could be involved in both starch synthesis and breakdown [6].
Metabolic pathways may include so-called futile cycles, i.e., concomitant metabolite synthesis and breakdown [7, 8]. Such cycles are thought to offer rapid, flexible, and sensitive regulation of fluxes through these pathways. Such futile cycles have been suggested for starch metabolism as well [9, 10]. If amylases, STP, and AGPase would participate in these cycles, in this report, we will thus focus on these enzymes. The literature presents discrepant data on the activity of amylases during tuber development, dormancy, and sprouting. Davies and Viola [11] reported a decrease in total amylase activity during sprouting, although the decrease in α-amylase was only transient. Bailey et al. [12] reported a transient increase for α-amylase at the time of sprouting, and Biemelt et al. [13] showed an increase in both α- and β-amylase activities in the sub-eye tissue after the onset of sprouting. The latter authors also found β-amylase activity to decline during the first phase of dormancy, but did not discuss this phenomenon further.
Also for STP, conflicting data are presented. Several authors agree on enhanced STP activity during dormancy initiation [14, 15]. However, the latter authors concluded that it was not clear whether in this case STP catalyzed the net starch degradation or synthesis. During dormancy breaking/sprouting, a temporary rise in the STP activity was mentioned by Bruinsma [14]. Davies and Viola [11] found a decline in STP activity during the same period; such decline was also described by Biemelt et al. [13].
AGPase activity was enhanced during tuberization [3, 15, 16] in parallel with starch accumulation. AGPase activity was higher in the vascular bundles and parenchyma of dormant tubers treated by bromoethane, a dormancy-breaking agent. No changes in AGPase activity were found in untreated tubers or in tubers treated with gibberellin [17].
The aim of this work was to study AGPase and STP activities from the end of tuber development to dormancy, then dormancy breakage and the subsequent sprouting. For comparison, the activities of amylases were also determined. The discrepancies mentioned above for STP activities might partly be explained by tissue-specific changes in the activity of this enzyme. Therefore, we used histochemical analyses of STP and AGPase activities in order to detect local changes, possibly overlooked when whole tubers were sampled for enzyme assays. Additionally, preliminary AGPase gene expression studies are shown to support the histochemical analyses of AGPase.

MATERIALS AND METHODS
In vitro tuberization system. The protocol to produce microtubers is described by Appeldoorn et al. [18]. In vitro plants (Solanum tuberosum L.), cv. Bintje for data on enzymes, cv. Desiree for data on gene activity, were grown at a 16-h light period (50 W/m2, 20°C), and after 4 weeks these plantlets were transferred to soil. They were maintained in a growth chamber for 4 weeks at a 16-h light period (80 W/m2, 20°C). To induce tubers in these plants, the light regime was changed to short days (8-h light period) for three weeks more. Single-node cuttings were taken and after surface sterilization placed on a tuber-inducing medium consisting of a modified MS medium (containing 0.1 part of the standard amounts of KNO3 and NH4NO3), 8% sucrose, 5 µM N6-benzyladenine (BA), and 0.8% agar at a final pH 5.8. The explants were incubated in the dark at 20°C, and tuber formation commenced on the 5th day. After growing for approximately 1 month, the tubers were considered mature and dormant. The diameter of these microtubers was around 8 mm, and their fresh weight was approximately 500 mg. When stored in the dark at room temperature, the tubers stayed dormant for approximately 4 to 5 months. Tubers used for enzyme analysis were stored attached to stem cuttings and on the nutrition medium (cv. Bintje). Tubers used for the analysis of gene activity were air-dried for 4 to 5 days at 20°C, in darkness, at relative humidity of approximately 80%. In this way, the stem cuttings (and stolons if present) dried out and could be cut off, leaving a minimal wound area.
Quantitative measurements of enzyme activities. For the measurements of AGPase, STP and α- and β-amylase activities, samples were taken from various batches differing in physiological state, ranging from late tuber development to dormancy, dormancy breakage, and subsequent sprouting. The first sample consisted of 23-day-old tubers, which were considered as still developing. The next samples were 86-, 149- and 184-day-old tubers, the latter tubers were considered dormant. Approximately half of the batch of 212-day-old tubers was already sprouting, and both sprouting and nonsprouting tubers were sampled and analyzed separately. At day 240, all tubers were sprouting. In the nonsprouting samples, the apical part was sampled separately from the rest of the tuber. In the sprouting samples, the sprouts were sampled separately. All data are based on 3 to 6 samples, each consisting of 2 to 4 (parts of) tubers or sprouts.
AGPase and STP. The extraction procedure for AGPase and STP was modified after Appeldoorn et al. [18]. In short, 1.2 ml of extraction buffer (50 mM Hepes/KOH (pH 7.4), 5 mM MgCl2, 1 mM EGTA, 1 mM EDTA, 10% glycerol, 0.1% BSA, 5 mM DTT, and 2% insoluble PVP was added to 10 mg of powdered freeze-dried tissue. After shaking (1 min) and centrifugation (5 min, 15 000 g), the supernatant was divided into small portions (0.2 ml), frozen in liquid nitrogen, and stored at 75°C until required. All extraction procedure was carried out in a cold room at 4°C.
Activities of AGPase and STP were determined in 250 µl assay medium (in a 96-well plate) at 30°C, for 10 min, using 30 µl of extract [15]. The reactions were started by adding extract or substrate and were monitored spectrophotometrically at 340 nm using a microtiter plate reader (“Versamax, Molecular Devices”, United States).
α-Amylase and β-amylase. The extraction procedure for α-amylase and β-amylase was modified after Cochrane et al. [19]. In short, 0.5 ml of extraction buffer (30 mM ethylenediamine dihydrochloride (pH 7.0 with KOH), 3 mM CaCl2, 20% glycerol, 10 mM DTT, and 0.5% Nonidet P-40) was added together with 10 mg of insoluble PVP to 10 mg of powdered freeze-dried tissue. After centrifugation (20 min, 6000 g), the supernatant was removed, frozen in liquid nitrogen and stored at 75°C until required. After shaking (1 min) and centrifugation (20 min, 6000 g), the supernatant was divided into portions (0.1 ml), frozen in liquid nitrogen, and stored at 75°C until required. All extraction procedure was carried out in a cold room at 4°C.
For measuring α- and β-amylase activity, the ‘Ceralpha’ and the ‘Betamyl’ kits (“Megazyme”, Ireland) were used. All solutions were prepared according to the instruction of the kits, and both assays were performed in a 96-well plate. For α-amylase, 20 µl of substrate solution (blocked p-nitrophenyl maltoheptaoside, glucoamylase, and α-glucosidase) was added to 15 µl of extract, and this mixture was incubated at 30°C for 90 min. For β-amylase, 20 µl of substrate solution (p-nitrophenyl maltopentaoside, α-glucosidase, and stabilizers) was added to 10 µl of extract, and this mixture was incubated at 30°C for 60 min. After incubation, both reactions were stopped by adding 200 µl of stopping reagent (1% Trizma base), and the absorbance was read at 405 nm with a microtiter plate reader (“Versamax, Molecular Devices”, Sunnyvale, CA). Blanks were carried out without substrate solution.
In situ staining of enzyme activities. Staining was based on the coupling of the oxidation NADH to the reduction of nitroblue tetrazolium (NBT), which results in precipitation of the blue tetrazolium salt. The whole procedure was carried out as described by Sergeeva and Vreugdenhil [20]. In short, sections of 120 µm thickness were cut with a sledge microtome. Sections were immediately fixed in a mixture of 2% paraformaldehyde, 2% PVP 40, and 1 mM DTT, pH 7.0, at 4°C for 1 h. After fixation, sections were rinsed overnight with water at 4°C and refreshed at least five times to remove soluble carbohydrates.
AGPase staining was performed by incubating the tissue sections at 30°C for 30 min in 1 ml of reaction medium. This medium was similar to the reaction mixture for the quantitative measurements, with addition of 0.03% NBT. The control reaction was carried out by omitting the substrate ADP-glucose [20]. Sections were cut, stained, and analyzed from tubers grown in three independent experiments, with 5 tubers per experiment.
STP staining was performed by incubating the sections at 30°C for 30 min in 1 ml of reaction medium, similar to the one used in the quantitative measurements, with the addition of 0.03% NBT. Controls were carried out without soluble starch and/or Na3PO4. For STP staining tubers were used, grown in two independent experiments, with 4 tubers per experiment.
Stained sections were studied under a Leica binocular or a Nikon Optiphot microscope in bright field mode. A digital Panasonic Color Video Camera or a Sony CCD camera DKR 700 was used to take photographs.
AGPase gene expression. For monitoring AGPase gene promoter expression, transgenic plants were used containing the luciferase reporter gene (from the American firefly, Photinus pyralis) fused with the AGPaseS promoter [21]. The AGPaseSluc transgenic tubers were sprayed with the substrate solution (1 mM D-luciferin, sodium salt, “Duchefa”; 0.01% Tween 80) 24 h and 3 h before measurement. For monitoring the light emitting reaction of luciferase, tubers were placed under an intensified CCD camera (C2400-77, “Hamamatsu Photonics”, Japan). Images were generated by integrating emission of photons for 15 min. The pixel intensity of these images is directly proportional to the activity of the luciferase enzyme present in the tubers. These measurements were done at the range of time points during dormancy and sprouting of one batch of tubers.

RESULTS
Quantitative measurements of enzyme activities

Activities of AGPase and STP during tuber development, dormancy, and sprouting are shown in fig. 1. Activities were determined separately in the tuber apical part, comprising approximately 10% of the tuber tissue around and including the apical bud, and in the rest of the tuber tissue. When tubers were sprouting, the sprouts were sampled separately, viz., at days 212 (50% sprouting) and 240 (100% sprouting).
It is clear that the AGPase activity of both the apical eye and the rest of the tuber tissue declined during the initial phase of dormancy. At the first two days of sampling (days 23 and 86), the AGPase activity in the apical part was higher than in the rest of the tuber. This difference disappeared when the tubers were stored for a longer time. No significant differences in AGPase activity were observed between sprouting and nonsprouting tubers at the end of the dormancy period (i.e., nonsprouting tubers of 184- and 212-day-old and sprouting tubers of 212- and 240-day-old). In tubers that started to sprout (day 212), the AGPase activity in the sprout did not differ significantly from the activity in the rest of the tuber. However, in older tubers (day 240), the activity in the sprout increased.
The STP activity, as shown in fig. 1b, also manifested a decline in activity during the initial phase of dormancy, most noticeably in the apical part of the tuber. At the end of the dormancy period, no notable difference in STP activity was found between dormant and sprouting tubers, although a slight tendency to increase was observed for the apical parts of sprouting tubers. The STP activity in the sprouts was significantly lower than in the apical part and the rest of the tuber tissue.
The activity of α-amylase (fig. 2a) also decreased during the initial phase of dormancy. In developing tubers (day 23), a higher activity in the apical part was seen compared to the rest of the tuber tissue. In dormant tubers, α-amylase activity gradually declined until the end of the dormant period (day 212, nonsprouting). During sprouting, an increase was observed, but only in the apical parts of the tubers.
The decrease in activity in the first phase of dormancy was also observed for β-amylase, both in the apical part and the rest of the tuber tissue. During later phases of dormancy and early at sprouting, the levels of β-amylase activity did not change significantly. In the sprouting tubers (day 240), β-amylase activity increased in the apical part (fig. 2b).

In situ staining of enzyme activities
The same batches of tubers as used for the quantitative measurements of enzyme activities were also used for in situ staining of AGPase enzyme activity. For localization of STP activity, only sprouting tubers were stained. Figures 3A-1 to 3A-9 represent tubers through a complete dormancy period, including sprouting and the sprout itself. At the end of tuber development (23-day-old tuber, fig. 3A-1), substantial activity of AGPase was seen in the whole tuber, especially around the vascular bundles. During the first phase of dormancy (58-day-old), activity in the tuber tissue decreased and was mainly visible around the apical bud (fig. 3A-2). In 86-day old tubers, i.e., dormant ones, activity was hardly detectable (fig. 3A-3), and at day 149, no activity was observed (fig. 3A-4). When the tubers were 184-day-old, and yet did not show visible sprouting, high AGPase activity was visible in the vascular tissue towards and in the apical part of the tuber and also in the vascular tissue in the rest of the tuber (fig. 3A-5). Figure 3A-9 is an enlargement of fig. 3A-5 revealing very dark staining in the apical bud of the tuber before sprouting. For comparison, a control (without ADP-glucose) completely lacking blue staining is shown in fig. 3A-10. Figure 3A-7 shows an enlargement of the base of a sprout, in which staining in vascular tissue and meristems are visible. The vascular tissue of the sprout was also stained as shown in fig. 3A-8.
Fig. 3S-1 presents the data on STP staining in a sprouting tuber. Like AGPase activity, it was present in the vascular tissue in tuber and sprout. Besides, a band of activity appeared in or around the phellogen layer. A control staining (without soluble starch and phosphate) is shown in fig. 3S-5: here some staining is still visible in the apical bud and around the phellogen layer, although much lower than in the presence of the STP substrates. Fig. 3S-2 is an enlargement of fig. 3S-1 clearly showing high enzyme activity in vascular tissue, storage parenchyma, and meristems. Figures 3S-3 and S-4 are enlargements of the band of staining around the phellogen layer revealing that the first layer of the phelloderm is stained rather than the phellogen layer itself.
AGPase gene expression
For measuring AGPase gene expression, microtubers were obtained from an AGPase-luc transformant, carrying the firefly luciferase gene under control of the AGPaseS promoter. The microtubers were stored in a 24-well plate, the apical buds facing the camera. The same tubers were measured every few weeks and fig. 4 shows a luciferase picture of dormant tubers and the same tubers but sprouting (a, 128-day-old, and b, 191-day-old). The tubers measured here are part of a batch with the sprouting level of 87%. The light emitted by the luciferase enzyme in fig. 4 is presented in a grey color. There was a large difference in luciferase activity between dormant and sprouting tubers, as represented in fig. 4. Hardly any of the 48 dormant tubers (fig. 4a) showed luciferase activity. Of the 23 tubers, still present in the wells after 191 days, at least 20 showed high luciferase activity, the remaining three showed lower activity. Luciferase activity was mainly found at the apical bud of the tuber (fig. 4; personal communication by J. Verhees). This is in agreement with the observation that some of the tubers showed luciferase activity at two spots indicating the sprouts from the lateral bud or the axillary bud at the main sprout.

DISCUSSION
Quantitative measurements of enzyme activities

During tuber development, high AGPase activity is required, as the tuber is a sink organ accumulating large amounts of starch [3]. In our previous experiments on tuber development, we found that AGPase activity in 9-day-old tubers was 4 times higher than at day 23 (data not shown). During the early phases of dormancy, the activity declined further (fig. 1a).
STP and amylase activities (figs. 1b, 2) also showed a clear decline at the beginning of dormancy. Compared to AGPase, which decreased from 3 nmol/(min mg dry wt) during tuber early development (day 9) to 0.2 during dormancy; the decrease of STP activity was not as steep, viz., from 2.5 nmol/(min mg dry wt) during tuber development (day 9, data not shown) to approximately 1.0 during tuber dormancy. Although it is clear that STP activity is present, it is not known whether it is involved in starch synthesis or breakdown [6], but during tuber development breakdown is likely to occur through amylases, which also showed high activities.
In general, none of the enzymes discussed above showed a clear change in activity during the sprouting period. Only, some marginal effects could be seen. These changes were either insignificant or occurred after visible sprouting. Thus, they may be related to sprouting and not to breakage of dormancy. A similar conclusion was drawn by Biemelt et al. [13] for β-amylase.

In situ staining of enzyme activities
Sergeeva and Vreugdenhil [20] showed high AGPase staining in developing 12-day-old tubers, resembling the young tuber in fig. 3A-1, with clear staining in the tuber and around the vascular bundles. The AGPase activity, as visualized by the histochemical method, decreased to very low levels during dormancy (figs. 3A-2, 3A-3, 3A-4). So, the initial decrease of AGPase activity during and after tuber development as determined in extracts (fig. 1b) is also visible in situ.
However, in contrast to the quantitative activity measurements, an elevated level of AGPase activity is clearly shown before and during the sprouting period, especially at the apical eye (figs. 3A-5, 3A-6, 3A-9). Also the sprout itself showed a blue precipitate, indicating AGPase activity. The detection of AGPase staining before visible sprouting indicates that AGPase is in some way connected to dormancy breaking or is at least one of the first signs of the breaking of dormancy in potato tubers.
It is unlikely that NBT is precipitated due to activity of another enzyme than AGPase. Fig. 3A-10 shows a control staining without the substrate ADP-glucose, in which no blue precipitate could be observed, indicating that the enzyme giving the blue precipitate uses ADP-glucose as a substrate. Several years ago, another enzyme breaking down ADP-glucose has been described, viz. ADP-glucose pyrophosphatase (AGPPase), which breaks down ADP-glucose to glucose-1-P and AMP [22]. However, the ADP-glucose breakdown catalyzed by AGPPase is inhibited in a competitive manner by PPi and 3PGA and its optimal activity occurs at pH 6.0, whereas the AGPase in our experiments was assayed at pH 8.0 in the presence of PPi and 3PGA. Moreover, an additional control for AGPase was carried out without both PPi and 3PGA, which gave no staining at all (data not shown). Thus, the precipitate shown in figs. 3A-1 to 3A-9 is very likely the result of AGPase activity only.
The remarkable increase in the AGPase activity at the apical eye, observed well before visible sprouting, was not evident from the quantitative measurements, not even in extracts of the apical part. The difference between the quantitative and in situ measurements could be caused by the presence of different isoforms of AGPase (for review see [23]); during the extraction procedure for the quantitative activity measurements, the isoform responsible for the in situ staining might not be extracted due to inactivation or tight association with subcellular structures. This would also mean that different isozymes of AGPase are active during tuber development and dormancy breaking/sprouting. On the other hand, the extraction of AGPase enzyme may be simply more difficult from amyloplasts in older tubers, or there could be a component present in the older tubers, which interferes with AGPase activity measurements. Research by Tiessen [5] suggests that the AGPase enzyme is subjected to activation/deactivation in response to oxygen availability. This would involve redox activation of AGPase via a mechanism, depending on oxygen availability. Such a mechanism would possibly link with the differential expression of oxidative-stress related genes during both tuberization and sprouting.
STP activity (figs. 3S-1 to 3S-5) showed approximately the same staining pattern as AGPase activity, except for the first phelloderm layer. In this layer, AGPase activity was not present, while STP activity clearly was. Moreover, starch granules had disappeared from this layer. The presence of the STP activity here indicates that the enzyme may be involved in starch breakdown. The relatively low level of STP activity found in the quantitative assay with tubers and sprouts (fig. 1b) can be explained by the localized activity at the base of the sprouts (fig. 3S-2), which is diluted in whole-sprout extracts.
The presence of a very low level of STP activity (blue precipitate) in the control experiment (fig. 3S-5) can be explained by the fact that, although soluble starch was omitted, endogenous starch is always present in the tissue slices as it cannot be washed out. Phosphate can be chemically or enzymatically released from starch. In this way all necessary substrates may be available for the STP reaction.
The experiments described above indicate that in situ staining is more sensitive than detection of enzyme activities in extracts and show the importance of accurate localization studies [20].

AGPase gene promoter expression
The AGPase enzyme is a heterotetramer composed of two large and two small subunits, AGPaseS and AGPaseB, respectively. There are three genes in potato encoding the large subunit and one gene encoding the small subunit and all four are expressed in the tuber. Furthermore, the expression of these genes can be differentially regulated, e.g., AGPaseS2 is expressed stronger in sink leaves than in source leaves, and its expression is induced by exogenous sucrose [23].
The AGPaseS-luc transformants used in this study showed an elevated expression in sprouting tubers (fig. 4). The presence of luciferase activity in the apical bud (fig. 4) supports the results of the staining experiments (figs. 3: A-5, A-6, A-7, A-9). However, whether this expression increased before visible sprouting remains unclear, mostly due to the asynchronicity of sprouting. This gene expression study indicates regulation of AGPase at the transcriptional level rather than at the translational level.

Starch cycling?
Appeldoorn et al. [15] already suggested starch cycling during tuber development as a consequence of the STP activity found during this period. In the research described here, active amylases could be extracted from developing tubers, strongly suggesting starch breakdown during this period. The increase in AGPase enzyme activity in situ, well before visible sprouting, suggests starch synthesis before and during the sprouting period. Taken together, these findings suggest starch cycling during all periods of potato tuber life cycle.
The staining patterns in sprouting tubers for AGPase and STP are similar, which might suggest a direct cycling of starch, STP degrading starch, forming glucose-1-phosphate, which is the substrate for AGPase. On the other hand, STP itself may also be involved in starch synthesis [6], and in this case, both enzymes would form starch in the direct neighbourhood of the vascular tissue. Duwenig et al. [24] reported parallels in the regulation of expression of phosphorylase and AGPase genes in spinach and potatoes and use this finding to support their suggestion that STP is involved in starch synthesis. However, such evidence can equally well indicate that STP participates in starch cycling. If it is so, such cycling is likely to involve the plastid isoform of STP, since Duwenig et al. [24] presented evidence that the cytosolic isoform was not involved in starch breakdown during storage. One may then hypothesize that amylases degrade starch in storage parenchyma cells, the resulting sugars being transported towards the vascular tissue, and starch is resynthesized around the vascular tissue. The high activity of AGPase and STP found in the bud could be explained by the converging of vascular bundles towards the bud. Viola et al. [4] showed that during the dormancy period the apical bud is symplastically isolated from the rest of the tuber, but at the onset of bud outgrowth, this symplastic isolation is terminated. The authors suggested that sucrose unloading into the buds was a key prerequisite for bud outgrowth.
Local increases in sucrose could up-regulate AGPase expression [23] and subsequent enzyme activity. In that case, the AGPase activity reported here would be an early result of the dormancy breaking process, but not a trigger. On the other hand, the increase in AGPase activity as seen in fig. 3A-5 is relatively early and should be studied in more detail.
Recently, a new mechanism for starch degradation was discovered: two enzymes, namely glucan water dikinase and phosphoglucan water dikinase, from leaves of potato and Arabidopsis, are shown to be involved in a novel mechanism of starch degradation via phosphorylation [6, 25]. Another recent study demonstrated that characteristics of some enzymes of starch metabolism may alter when they act as the components of different multienzyme complexes, and such assembly is essential for starch synthesis and degradation [25]. Further investigations are needed to reveal if this new evidences [6, 25] is relevant for understanding the regulation of potato tuber dormancy and sprouting.
We gratefully acknowledge the contribution of J. Krassenburg, who did the experiments on AGPase gene expression.
The work presented here was funded by the European Union, Project ‘Biology of tuber dormancy and sprouting’ (BIO4-CT96-0529) and partly by the Program of the Presidium of RAS “Molecular and Cell Biology” and the grant from the Russian Foundation for Basic Research (no 10-04-00638).
REFERENCES
1.Claassens M.M.J., Vreugdenhil D. Is Dormancy Breaking of Potato Tubers the Reverse of Tuber Initiation? // Potato Res. 2000. V. 43. P. 347–369.
2.Fernie A.R., Willmitzer L. Molecular and Biochemical Triggers of Potato Tuber Development // Plant Physiol. 2001. V. 127. P. 1459–1465.
3.Visser R.G.F., Vreugdenhil D., Hendriks T., Jacobsen E. Gene Expression and Carbohydrate Content during Stolon to Tuber Transition in Potatoes (Solanum tuberosum) // Physiol. Plant. 1994. V. 90. P. 285–292.
4.Viola R., Pelloux J., van der Ploeg A., Gillespie T., Marquis N., Roberts A.G., Hancock R.D. Symplastic Connection Is Required for Bud Outgrowth Following Dormancy in Potato (Solanum tuberosum L.) Tubers // Plant Cell Environ. 2007. V. 30. P. 973–983.
5.Tiessen A., Hendriks J.H.M., Stitt M., Branscheid A., Gibon Y., Farré E.M., Geigenberger P. Starch Synthesis in Potato Tubers Is Regulated by Post-Transcriptional Redox Modification of ADP-Glucose Pyrophosphorylase: A Novel Regulatory Mechanism Linking Starch Synthesis to Sucrose Supply // Plant Cell. 2002. V. 14. P. 2191–2213.
6.Rathore R.S., Garg N., Garg S., Kumar A. Starch Phosphorylase: Role in Starch Metabolism and Biotechnological Applications // Crit. Rev. Biotechnol. 2009. V. 29, P. 214–224.
7.Nquven-Quoc B., Fover C.H. A Role for ‘Futile Cycles’ Involving Invertase and Sucrose Synthase in Sucrose Metabolism of Tomato Fruit // J. Exp. Bot. 2001. V. 358. P. 881–889.
8.Rossouw D., Bosch S., Kossmann J., Botha F.C., Groenewald J.-H. Downregulation of Neutral Invertase Activity in Sugarcane Cell Suspension Cultures Leads to a Reduction in Respiration and Growth and an Increase in Sucrose Accumulation // Funct. Plant Biol. 2007. V. 34. P. 490–498.
9.Geigenberger P., Merlo L., Reimholz R., Stitt M. When Growing Potato Tubers Are Detached from Their Mother Plant There Is a Rapid Inhibition of Starch Synthesis, Involving Inhibition of ADP-Glucose Pyrophosphorylase // Planta. 1994. V. 193. P. 486–493.
10.Sweetlove L.J., Burrell M.M., ap Rees T. Starch Metabolism in Tubers of Transgenic Potato (Solanum tuberosum) with Increased ADPglucose Pyrophosphorylase // Biochem. J. 1996. V. 320. P. 493–498.
11.Davies H.V., Viola R. The Effect of Gibberellic Acid on Starch Breakdown in Sprouting Tubers of Solanum tuberosum L. // Ann. Bot. 1988. V. 61. P. 689–693.
12.Bailey K.M., Phillips I.D.J., Pitt D. The Role of Buds and Gibberellin in Dormancy and the Mobilization of Reserve Materials in Potato Tubers // Ann. Bot. 1978. V. 42. P. 649–657.
13.Biemelt S., Hajirezaei M., Hentschel E., Sonnewald U. Comparative Analysis of Abscisic Acid Content and Starch Degradation during Storage of Tubers Harvested from Different Potato Varieties // Potato Res. 2000. V. 43. P. 371–382.
14.Bruinsma J. A Survey of Recent Japanese Research on Dormancy of Potato Tubers // Eur. Potato J. 1962. V. 8. P. 195–203.
15.Appeldoorn N.J.G., de Bruijn S.M., Koot-Gronsveld E.A.M., Visser R.G.F., Vreugdenhil D., van der Plas L.H.W. Developmental Changes in Enzymes Involved in the Conversion of Hexose-Phosphate and Its Subsequent Metabolites during Early Tuberisation of Potato // Plant Cell Environ. 1999. V. 22. P. 1085–1096.
16.Sowokinos J.R. Pyrophosphorylases in Solanum tuberosum. I. Changes in ADP-Glucose and UDP-Glucose Pyrophosphorylase Activities Associated with Starch Biosynthesis, during Tuberization, Maturation and Storage of Potatoes // Plant Physiol. 1976. V. 57. P. 63–68.
17.Alexopoulos A.A., Aivalakis G., Akoumianakis K.A., Passam H.C. Bromoethane Induces Dormancy Breakage and Metabolic Changes in Tubers Derived from True Potato Seed // Postharv. Biol. Technol. 2009. V. 54. P. 165–171.
18.Appeldoorn N.J.G., de Bruijn S.M., Koot-Gronsveld E.A.M., Visser R.G.F., Vreugdenhil D., van der Plas L.H.W. Developmental Changes of Enzymes Involved in the Conversion of Sucrose to Hexose-Phosphate during Early Tuberisation of Potato // Planta. 1997. V. 202. P. 220–226.
19.Cochrane M.P., Duffus C.M., Allison M.J., Mackay G.R. Amylolytic Activity in Stored Potato Tubers. 1. Estimation Using P-Nitrophenyloligosaccharides // Potato Res. 1991. V. 34. P. 325–332.
20.Sergeeva L.I., Vreugdenhil D. In Situ Staining of Activities of Enzymes Involved in Carbohydrate Metabolism in Plant Tissues // J. Exp. Bot. 2002. V. 53. P. 361–370.
21.Verhees J., van der Krol A.R., Vreugdenhil D., van der Plas L.H.W. Characterization of Gene Expression during Potato Tuber Development in Individuals and Populations Using the Luciferase Reporter System // Plant Mol. Biol. 2002. V. 50. P. 653–665.
22.Kleczkowski L.A. A New Player in the Starch Field // Plant Physiol. Biochem. 2001. V. 39. P. 759–761.
23.Geigenberger P. Regulation of Starch Biosynthesis in Response to a Fluctuating Environment // Plant Physiol. 2011. V. 155. P. 1566-1577.
24.Duwenig E., Steup M., Kossmann J. Induction of Genes Encoding Plastidic Phosphorylase from Spinach (Spinacia oleracea L.) and Potato (Solanum tuberosum L.) by Exogenously Supplied Carbohydrates in Excised Leaf Discs // Planta. 1997. V. 203. P. 111–120.
25.Kötting O., Kossmann J., Zeeman S., Lloyd J.R. Regulation of Starch Metabolism: The Age of Enlightenment? // Curr. Opin. Plant Biol. 2010. V. 13. P. 321–329.

FIGURE CAPTIONS
Fig. 1. AGPase (a) and STP (b) activities in different parts of potato microtubers during dormancy and sprouting.
Activities were determined in the apical part of the tuber (approximately 0.1 of total tuber mass) (2), the rest of the tuber tissue (1), and in the sprouts (if present) (3). Asterisk (*) indicates that tubers are sprouting. Tuber batches of 212-day-old are divided into sprouting (212*) and nonsprouting tubers (212). Activity is presented as nmol NADH/(min mg dry wt) ± SD. All data are based on 3 to 6 samples, each consisting of 2 to 4 (parts of) tubers or sprouts.

Fig. 2. α-Amylase (a) and β-amylase (b) activities in different parts of potato microtubers during dormancy and sprouting.
Activities were determined in the apical part of the tuber (approximately 0.1 of total tuber mass) (2) and in the rest of the tuber tissue (1). Asterisk (*) indicates that tubers are sprouting. Tuber batches of 212-day-old are divided into sprouting (212*) and nonsprouting tubers (212). Activity is presented as nmol NADH/(min mg dry wt) ± SD. All data are based on 3 to 6 samples, each consisting of 2 to 4 (parts of) tubers.

Fig. 3. Localization of AGPase (A) and STP (S) activities in potato microtubers at different stages of their life cycle. Arrows indicate parts of tubers with intense staining.
A-1: AGPase staining in developing tuber, almost mature, 23-day-old; A-2: dormant tuber, 58-day-old; A-3: dormant tuber, 86-day-old; A-4: dormant tuber, 149-day-old; A-5: dormant tuber, about to break dormancy, 184-day-old; A-6: sprouting tuber, 225-day-old (A1A6: bar = 2 mm); A-7: detail of base of sprout: staining in vascular tissue and meristems of the buds (bar = 0.5 mm); A-8: detail of sprout, showing high staining in vascular tissue, 240-day-old (bar = 0.5 mm); A-9: detail of A-5, very high staining in apical part (eye) and vascular tissue (bar = 1 mm); A-10: control staining of sprouting tuber (without ADP-glucose), no staining visible (bar = 0.5 mm). S-1: STP staining in sprouting tuber, 240-day-old (bar = 1 mm); S-2: detail of S-1, sprouting tuber, staining in vascular tissue (bar = 0.8 mm); S-3: detail of phellogen area of sprouting tuber, high staining in one layer in that area (bar = 0.5 mm); S-4: further detail of phellogen area, staining in first layer of phelloderm (bar = 60 µm); S-5: control staining of sprouting tuber (without soluble starch and Na3PO4), staining visible in vascular tissue, parenchyma cells and in phelloderm cells (bar = 1 mm).

Fig. 4. AGPase gene expression in transgenic potato microtubers.
AGPase-luc expression pattern of the same batch of tubers was measured after 128 days (a; dormant tubers) and after 191 days (b; sprouting tubers). The luciferase activity is displayed in grey colour. In each well, 1 microtuber is present and the apical bud is turned to the camera. In (b), wells marked with a cross did not contain a tuber anymore, since these were sampled for other analyses.