By continuing to browse this site you agree to us using cookies as described in About Cookies
Wiley Online Library is migrating to a new platform powered by Atypon, the leading provider of scholarly publishing platforms. The new Wiley Online Library will be migrated over the weekend of February 24 & 25 and will be live on February 26, 2018. For more information, please visit our migration page: http://www.wileyactual.com/WOLMigration/
We recently discovered that post-translational redox modulation of ADP-glucose pyrophosphorylase (AGPase) is a powerful new mechanism to adjust the rate of starch synthesis to the availability of sucrose in growing potato tubers. A strong correlation was observed between the endogenous levels of sucrose and the redox-activation state of AGPase. To identify candidate components linking AGPase redox modulation to sugar supply, we used potato tuber discs as a model system. When the discs were cut from growing wild-type potato tubers and incubated for 2 h in the absence of sugars, redox activation of AGPase decreased because of a decrease in internal sugar levels. The decrease in AGPase redox activation could be prevented when glucose or sucrose was supplied to the discs. Both sucrose uptake and redox activation of AGPase were increased when EDTA was used to prepare the tuber discs. However, EDTA treatment of discs had no effect on glucose uptake. Feeding of different glucose analogues revealed that the phosphorylation of hexoses by hexokinase is an essential component in the glucose-dependent redox activation of AGPase. In contrast to this, feeding of the non-metabolisable sucrose analogue, palatinose, leads to a similar activation as with sucrose, indicating that metabolism of sucrose is not necessary in the sucrose-dependent AGPase activation. The influence of sucrose and glucose on redox activation of AGPase was also investigated in discs cut from tubers of antisense plants with reduced SNF1-related protein kinase activity (SnRK1). Feeding of sucrose to tuber discs prevented AGPase redox inactivation in the wild type but not in SnRK1 antisense lines. However, feeding of glucose leads to a similar activation of AGPase in the wild type and in SnRK1 transformants. AGPase redox activation was also increased in transgenic tubers with ectopic overexpression of invertase, containing high levels of glucose and low sucrose levels. Expression of a bacterial glucokinase in the invertase-expressing background led to a decrease in AGPase activation state and tuber starch content. These results show that both sucrose and glucose lead to post-translational redox activation of AGPase, and that they do this by two different pathways involving SnRK1 and an endogenous hexokinase, respectively.
ADP-glucose pyrophosphorylase (AGPase) is a key regulatory enzyme of starch biosynthesis in higher plants (Geigenberger, 2003; Martin and Smith, 1995; Preiss, 1988). It has recently been discovered that AGPase is subjected to post-translational redox regulation (Ballicora et al., 2000; Fu et al., 1998), which provides a powerful new mechanism to adjust the rate of starch synthesis in potato tubers to sucrose supply (Tiessen et al., 2002). The mechanism involves the reversible formation of an intermolecular cysteine bridge between the two small subunits in the AGPase heterotetramer (Fu et al., 1998), which affects the kinetic properties of the enzyme by changing its substrate affinities and by modifying its sensitivity to allosteric effectors (Tiessen et al., 2002). This regulatory mechanism was found to be operating in vivo in a range of conditions that modify carbon supply and sucrose levels in growing potato tubers (Tiessen et al., 2002). Redox activation of AGPase was high at early stages of tuber development and decreased as the tubers aged and matured. The decrease in AGPase redox state was paralleled by a decrease in sucrose levels and by a decline in the rate of starch accumulation in these tubers. Sudden interruption of sucrose supply also led to a decrease in redox activation of AGPase in growing potato tubers within 24 h after detachment, resulting in a strong inhibition of starch biosynthesis. Ectopic expression of sucrose phosphorylase in the cytosol of transgenic potato tubers led to a decrease in tuber sucrose levels, which was again accompanied by a decrease in the redox-activation state of AGPase and an inhibition of starch accumulation in these lines. Furthermore, antisense inhibition of AGPase led to increased tuber-sucrose levels and an increased redox activation of the remaining AGPase. Crucially, across all these treatments, a strong correlation was observed between the endogenous levels of sucrose and the redox-activation state of AGPase (Tiessen et al., 2002), suggesting a link between sucrose and redox modulation of this enzyme.
The factors and components of the putative pathways connecting sucrose and post-translational redox modulation of AGPase have not yet been identified. More is known on the sugar-signalling components involved in the transcriptional regulation of gene expression (Koch, 1996). It has been suggested that SNF1-related protein kinase (SnRK1) is involved in the sucrose-induced expression of sucrose synthase in potato plants (Purcell et al., 1998). Based on previous work on yeast (Johnston, 1999), hexokinase has frequently been proposed to act as a sugar sensor in various plant species (Jang and Sheen, 1997; Jang et al., 1997). However, so far, there is no evidence that hexokinase is involved in sugar-signalling pathways in potato tubers (Veramendi et al., 1999, 2002).
To identify candidate components linking AGPase redox modulation to sucrose supply, we used potato tuber discs as a model system. Previous experiments showed that the redox-activation state of AGPase decreased when tuber discs were incubated for 2 h without sugar supply, which could be partially prevented by providing sucrose in the medium (Tiessen et al., 2002). In the present paper, we (i) asked whether a similar response is observed when glucose is supplied instead of sucrose, (ii) used different sucrose and glucose analogues to investigate whether metabolism or phosphorylation is an essential step of the signal perception and transduction leading to redox activation of AGPase, (iii) used lines with antisense reduction of SnRK1 to test whether this protein kinase is involved in the redox activation of AGPase in response to the supply of different sugars and (iv) used transgenic lines with ectopic expression of invertase alone or in combination with a bacterial glucokinase to evaluate the effect of opposing changes in the levels of glucose and sucrose on the activation state of AGPase in planta. Our results show that both sucrose and glucose lead to redox activation of AGPase, and that they do this by two different pathways involving SnRK1 and endogenous hexokinase, respectively.
Influence of feeding glucose or sucrose on redox activation of AGPase in wild-type potato tuber discs
Previous studies showed that AGPase redox-activation state and sucrose levels both decrease within 2 h when discs from wild-type potato tubers are incubated in buffer solution, and that this decrease can be partially prevented by external feeding of sucrose (Tiessen et al., 2002). In the following experiments, we investigated whether a similar effect is observed when glucose, instead of sucrose, is supplied to the discs. We also compared the sugar-uptake rate in tuber discs prepared without or with EDTA. Figure 1(a,b) compares the uptake rates of glucose and sucrose in a range between 2 and 200 mm by potato tuber discs. Glucose-uptake rates were two- to threefold higher than that of sucrose uptake. Interestingly, addition of EDTA to the incubation medium increased sucrose-uptake rates whereas glucose uptake was not substantially changed. Figure 1(c) compares the AGPase redox-activation state in intact tubers (time 0) and discs incubated for 2 h without sugar (control) or with 200 mm sucrose and glucose. Glucose feeding led to a strong increase in AGPase redox activation compared to sucrose feeding, buffer control or intact tubers at the start of the experiment. A similar increase in AGPase activation was measured when glucose was supplied to tuber discs treated with or without EDTA. Also, in the buffer-control incubations, the EDTA treatment alone had no significant effect on the activation state of AGPase. In contrast to this, the influence of sucrose feeding on AGPase activation was dependent on the presence of EDTA in the medium. The EDTA treatment increased sucrose-uptake rates (Figure 1b), and only in this condition, the decrease in AGPase activation state after 2 h incubation with 200 mm sucrose was completely prevented (Figure 1c). In the following disc experiments, 1 mm EDTA was therefore routinely used to prepare tuber discs, and was included in the incubation solutions.
Influence of feeding different sugar analogues on redox activation of AGPase in wild-type potato tuber discs
Sugar analogues are useful tools by which information relating to signalling mechanisms can be derived because, when used carefully, they allow the discrimination of the effects of the analogous sugar per se and the effect of a metabolic product of this sugar (Koch, 1996; Martin et al., 1997; Pego et al., 1999; Roitsch et al., 1995). In a further set of experiments, we incubated tuber discs with different sugar analogues to investigate whether glucose or sucrose per se or their metabolism are important for the redox modulation of AGPase. Freshly cut discs from growing wild-type tubers were either frozen in liquid N2 immediately (time 0) or incubated for a further 2 h in a medium lacking sugars (MES control) or containing 200 mm glucose, fructose, mannose, 2-deoxy-glucose or 3-O-methyl-glucose. In these experiments, the analysis of the redox-activation state of AGPase was performed in two different ways: (i) by separation of the protein on non-reducing SDS gels with subsequent Western blotting using small subunit of AGPase (AGPB)-specific antibody (Figure 2a) and (ii) by activity assay (Figure 2b). Incubation of discs for 2 h in the absence of sugars led to a strong decrease of the monomeric form of AGPB and a 50% decrease in the redox-activation state of AGPase. External feeding of glucose or fructose prevented both the inactivation and dimerisation of AGPase. These sugars are rapidly absorbed, phosphorylated and metabolised in potato tubers (Geiger et al., 1998). External feeding of mannose and 2-deoxy-glucose also prevented the redox inactivation of AGPase in the discs (Figure 2a,b). These sugars are similar to glucose and fructose in being rapidly phosphorylated by hexokinase, but are also different from glucose and fructose in not (or only very slowly) being metabolised in potato-tuber tissue (Renz and Stitt, 1993; Schnarrenberger, 1990). Interestingly, external feeding of 3-O-methyl-glucose did not prevent the inactivation and dimerisation of AGPase in tuber discs (Figure 2a,b). This glucose analogue can be taken up by the cell, but cannot be further phosphorylated by hexokinase (see recent review by Rolland et al., 2002). Similar results were obtained in a second set of experiments, where tuber discs were prepared and incubated in the absence of EDTA (data not shown). Moreover, no activation of AGPase was observed in incubations with other non-phosphorylatable hexose analogues (6-deoxy-glucose and l-glucose) or with substrates (glycerol and choline) that are phosphorylated by other enzymes than hexokinase (data not shown). These results suggest that the phosphorylation step catalysed by hexokinase is a necessary component in the hexose-induced redox activation of AGPase.
The levels of phosphorylated glucose and fructose (sum of glucose-6P, glucose-1P and fructose-6P) were not substantially increased after feeding glucose and fructose to the discs (Figure 2c). This was probably because of a stimulation of glucose-P- and fructose-P-consuming pathways by (i) increased activation of AGPase (Figure 2a,b) and subsequent stimulation of starch synthesis, (ii) increased rates of respiration (Figure 2d; Geiger et al., 1998) and (iii) increased rates of sucrose re-synthesis (Geiger et al., 1998). Increased activation of AGPase (Figure 2a,b) and subsequent stimulation of starch synthesis probably led to a decrease in the levels of glucose-P and fructose-P in discs incubated with mannose and 2-deoxy-glucose (Figure 2c, data exclude mannose-6P and 2-deoxy-glucose-6P), whereas the decrease in AGPase activation (Figure 2a,b) and starch synthesis in discs incubated with 3-O-methyl-glucose probably led to an accumulation of glucose-P and fructose-P (Figure 2c). Analysis of the respiration rate of tuber discs incubated in parallel for 1 h showed that glucose and fructose led to an increase, and mannose led to a decrease, in respiration rates, whereas all other conditions did not lead to significant changes in respiration rates as compared to freshly prepared tuber discs or buffer control (Figure 2d). Similar results were obtained when respiration rates were analysed after 3 h incubation (data not shown). The sustained respiration rates indicate that there were no toxic effects of the sugar analogues within the time-frame of incubation.
In a separate experiment, we also incubated tuber discs in the presence of EDTA with the non-metabolisable sucrose analogue, palatinose (Fernie et al., 2001), to address the question whether palatinose was able to partially prevent the decrease in AGPase redox activation in the discs (Table 1). Whereas non-sugar incubations led to inactivation of AGPase in discs after 2 h, AGPase inactivation was partially prevented by external feeding of sucrose or 100 mm palatinose (Table 1). These results suggest that metabolism including cleavage of sucrose into glucose and fructose is not a necessary component in the sucrose-induced redox activation of AGPase.
Table 1. Effects of the non-metabolisable sucrose analogue, palatinose, on AGPase activation
AGPase % activation (Vsel/Vred)
Sucrose and palatinose have a similar activation effect on AGPase. Tubers discs were prepared in 10 mm MES, 1 mm EDTA (pH 6.5) and incubated for 2 h with continuous shaking. After incubation, the discs were frozen in liquid nitrogen and the AGPase activation was determined with the spectrophotometric activity assay. Values are the mean from several measurements on a pool of incubated tuber discs ± SE (n = 3).
55 ± 0.0
2-h incubation + EDTA
32 ± 1.0
200 mm sucrose
42 ± 1.2
100 mm palatinose
42 ± 0.4
The activating effect of sucrose on AGPase is diminished in tuber slices of SnRK1 antisense plants
Two SnRK1 antisense lines were used (Pat 1.3 and Pat 1.10) with a tuber-specific 50–70% reduction of SnRK1 protein kinase activity (Purcell et al., 1998). These lines had no visual phenotype (data not shown). Transgenic tubers had slightly reduced starch contents (13 ± 1, 9.3 ± 0.6, and 10 ± 0.6% starch per gram FW for the wild type, Pat 1.3 and Pat 1.10, respectively, mean ± SE; n = 8), but no major changes occurred in the levels of sucrose, glucose, hexose-P, pyruvate, phosphoenol pyruvate (PEP) or various nucleotides (data not shown). The SnRK1 antisense lines had reduced steady-state levels of sucrose synthase (SuSy) transcript in their tubers, but SuSy activity was only reduced in one line (Pat 1.10; Purcell et al., 1998, and data not shown).
Discs from wild-type and SnRK1 tubers were either frozen immediately (time 0) or cut into a medium containing 1 mm EDTA and 10 mm MES-KOH (pH 6.5), and incubated for 1 h without sugars or by supplying 200 mm sucrose or glucose. Discs were either frozen in liquid N2 to analyse AGPase redox activation (Figure 3a) or washed briefly to remove exogenous sugars before freezing in liquid nitrogen and analysing the levels of sucrose (Figure 3b), glucose (Figure 3c) and total hexose phosphates (Figure 3d) within the discs. Redox activation of AGPase in intact tubers was slightly higher in antisense lines compared to the wild type (Figure 3a). When the discs were incubated in the absence of exogenous sugars, the levels of sucrose and glucose (Figure 3b,c) and the AGPase redox-activation state (Figure 3a) decreased in both wild-type and transgenic tuber discs. When 200 mm glucose was supplied, there was a marked increase in glucose levels in wild-type and transgenic discs (Figure 3c), and the decrease in AGPase-activation state could be prevented to a similar extent in wild-type and the antisense lines. After supplying 200 mm sucrose, there was a similar increase in sucrose levels in wild-type and transgenic tuber discs; however, the increase in AGPase redox activation was much stronger in the wild type than in the antisense lines (Figure 3a). Compared to the buffer control, incubation with sucrose led to a slight decrease and incubation with glucose led to a slight increase in hexose-phosphate levels, as has been seen in previous experiments (Geiger et al., 1998; Loef et al., 2001). These results show that wild-type and antisense lines with reduced SnRK1 respond differently in respect to AGPase activation when fed with sucrose and glucose in the presence of EDTA. The data are consistent with sucrose leading to redox modulation of AGPase via SnRK1-dependent signalling, and with glucose acting independently of SnRK1.
The data presented in Figure 3(a) indicate that a decrease in SnRK1 leads to a more rapid inactivation of AGPase in tuber discs upon incubation in buffer. This was confirmed in a second experiment, where a more detailed time-course was analysed (data not shown). In this case, within 10–40 min of incubation, AGPase redox activation decreased dramatically in transgenic tuber discs, which was not affected or delayed by sucrose. Discs from wild-type tubers showed no significant changes in AGPase activation state within this time-frame.
AGPase redox activation is changed in transgenic potato tubers with altered glucose to sucrose ratios
A further set of experiments was carried out with intact tubers of transgenic potato lines with tuber-specific ectopic overexpression of invertase in the cytosol (U-INV2-30) or apoplast (U-INV1-33; Sonnewald et al., 1997) to provide further evidence that hexoses lead to increased redox activation of AGPase independent of sucrose. These transgenic lines have been extensively characterised in the past, and have been documented to be representative lines (Hajirezaei et al., 2000; Sonnewald et al., 1997; Tauberger et al., 1999). Ectopic expression of invertase led to a marked decrease in sucrose and a marked increase in hexose levels, but unaltered or slightly decreased starch levels in developing tubers of both transgenic lines, compared to the wild type (Table 2, and references above). Also hexose-P levels increased, the increase being higher when invertase was expressed in the cytosol (Table 2). Overexpression of invertase also led to an increase in the redox-activation state of AGPase in both of the transgenic lines (Table 2), whereas overall activity reflecting AGPase protein level was not substantially changed. This provides in planta evidence that glucose can redox-activate AGPase in the presence of low levels of sucrose.
Table 2. Effects of ectopic overexpression of invertase in the cytosol or in the apoplast
Overexpression of yeast invertase in potato tubers leads to reduced sucrose but to increased hexoses and AGPase activation state. Developing tuber samples were taken from 9-week-old plants and sugars, hexose-P and starch were measured after ethanol extraction. Values are mean ± SE (n = 3). U unit of enzyme activity (µmol min−1).
46.4 ± 1.2
65.6 ± 1.9
69 ± 0.7
U g−1 FW
0.55 ± 0.05
0.60 ± 0.03
0.49 ± 0.06
µmol g−1 FW
364 ± 14
277 ± 16
377 ± 12
µmol g−1 FW
43.4 ± 1.0
12.8 ± 0.3
5.6 ± 1.6
µmol g−1 FW
14.3 ± 2.9
117.1 ± 2.4
65.7 ± 8.7
µmol g−1 FW
0.1 ± 0.01
9.43 ± 1.05
1.33 ± 0.26
nmol g−1 FW
230 ± 53
279 ± 108
1455 ± 72
We also investigated the changes in AGPase redox state in transgenic lines overexpressing a heterologous invertase in combination with a bacterial glucokinase (Trethewey et al., 1998). Combined overexpression in the cytosol led to an increased hexose-phosphorylation activity in the transgenic glucokinase (GK3) tubers, GK3-29, GK3-38 and GK3-41, which increased 2.4-fold when measured with glucose as substrate, but increased only slightly (1.2-fold) when measured with fructose (Table 3). As reported for the different hexokinase and fructokinase isoforms in potato tuber (Renz and Stitt, 1993), activity is highest for the fructose and UTP substrate combination, whereas activity with glucose is lower and prefers ATP as substrate (Table 3). Compared to the wild type, sucrose levels were markedly decreased, and the hexose levels were not or only moderately increased in developing GK3 tubers (Table 3). Compared to the parental invertase line (see Table 2), overexpression of the bacterial glucokinase led to a marked decrease in the hexose levels. This was accompanied by a decrease in overall AGPase expression and the redox-activation state of AGPase in the GK3 tubers, both in comparison to the wild type (Table 3) and the parental U-INV2-30 line (compare with Table 2). The GK3 tubers also had decreased starch contents compared to both the wild type (Table 3) and the parental line (see Table 2), which is consistent with previous studies on these lines (Fernie et al., 2000; Trethewey et al., 1998, 1999).
Table 3. Effects of combined overexpression of invertase and a bacterial glucokinase in the cytosol
Invertase + glucokinase
Overexpression of a bacterial glucokinase in the INV2-30 background in potato tubers leads to inactivation of AGPase and decreased AGPase activity compared to the wild type and the parental line. Developing tuber samples were taken from 10-week-old plants to analyse AGPase activity and metabolites levels. Sugars, hexose-P and starch were measured after TCA extraction. Hexokinase activity was assayed with 1.5 mm hexose and 2.5 mm nucleotide, respectively. Values are mean ± SE (n = 3). U unit of enzyme activity (µmol min−1).
56.7 ± 0.4
48.0 ± 3.2
47.1 ± 3.1
45.0 ± 1.7
U g−1 FW
0.75 ± 0.02
0.55 ± 0.02
0.39 ± 0.01
0.47 ± 0.00
Glucose + ATP
U g−1 FW
0.11 ± 0.01
0.26 ± 0.01
0.26 ± 0.01
0.24 ± 0.01
Glucose + UTP
U g−1 FW
0.04 ± 0.01
0.13 ± 0.02
0.10 ± 0.01
0.11 ± 0.02
Fructose + ATP
U g−1 FW
0.61 ± 0.01
0.85 ± 0.01
0.69 ± 0.01
0.80 ± 0.03
Fructose + UTP
U g−1 FW
0.86 ± 0.01
1.11 ± 0.03
0.89 ± 0.00
1.09 ± 0.02
µmol g−1 FW
718 ± 11
556 ± 40
544 ± 52
498 ± 38
µmol g−1 FW
25.2 ± 1.4
0.3 ± 0.0
0.0 ± 0.0
0.6 ± 0.1
µmol g−1 FW
9.3 ± 0.4
17.7 ± 0.5
3.4 ± 0.1
27.5 ± 0.1
µmol g−1 FW
1.0 ± 0.0
1.7 ± 0.1
1.0 ± 0.1
2.4 ± 0.0
nmol g−1 FW
209 ± 16
1932 ± 65
1798 ± 89
1458 ± 64
We have shown in previous studies that AGPase is regulated by post-translational redox modulation in a sucrose-dependent manner, leading to a stimulation of starch biosynthesis in growing potato tubers (Tiessen et al., 2002). The present paper further investigates the link between sugars and redox regulation in potato tubers. Results show that both sucrose and glucose modulate the redox regulation of AGPase, and that they do this by different pathways involving SnRK1 and hexokinase, respectively.
A SNF1-related kinase is involved in the signalling pathway leading to sucrose-dependent redox activation of AGPase
Results of Figure 3 show that decreased expression of SnRK1 modifies AGPase redox regulation in tuber discs in response to sucrose, but not to glucose. Decreased expression of SnRK1 also led to a more rapid inactivation of AGPase, when internal sucrose concentrations decreased in discs incubated in the absence of sugars (data not shown). This implies that SnRK1 acts as a positive regulator component in the response of AGPase redox state to sucrose. Interestingly, the influence of external sucrose, but not of glucose, on AGPase redox activation was strongly dependent on the treatment of the tuber discs with low concentrations of EDTA (Figure 1). Moreover, the uptake of sucrose, but not of glucose, by the tuber discs was stimulated in response to EDTA. The reason for this phenomenon is unknown, but might involve specific modulations of transporter activities by changes in Ca2+ concentrations. Nevertheless, the EDTA treatment allowed SnRK1-related processes to be uncovered and further dissected.
Interestingly, the redox state of AGPase in tubers of wild-type and SnRK1 antisense lines was not substantially different at the time of the start of these experiments (Figure 3a). This suggests that SnRK1 is modulating AGPase redox responses to dynamic and rapid changes in sucrose levels rather than determining the general AGPase redox state of a tuber. In the tuber disc system, decreased expression of SnRK1 led to a more rapid AGPase inactivation when sucrose decreased and to a lag in the activation kinetics when sucrose was supplied. It is tempting to speculate whether in planta alterations in sucrose supply (such as those occurring during day–night transitions) could lead to temporal drops in AGPase redox activation and starch synthesis in SnRK1 antisense tubers. This might explain that on a long-term scale, starch accumulation was slightly lower in growing tubers of SnRK1 plants as compared to the wild type.
SNF-like proteins are central components in the regulation of carbon metabolism and signalling in microbes and animals (Hardie et al., 1998). Recent studies on SnRK1 antisense lines provided evidence for the functional significance of SNF-like kinases in plants, providing valuable information on the influence of SnRK1 on sucrose-induced gene expression (Purcell et al., 1998). However, the biochemical phenotypes of SnRK antisense lines have been difficult to address. Our results provide evidence for a role of SnRK1 in the post-translational regulation of plant metabolism, in addition to its importance in modulating gene expression. Although in yeast, SNF1 may serve as a ‘fuel gauge’ (Hardie and Carling, 1997), we propose that the function of SnRK1 in plants is not restricted to sugar catabolism, but is also involved in regulating central biosynthetic processes like starch synthesis according to the sucrose supply.
Redox activation of AGPase seems also to be mediated by palatinose (Table 1), a sucrose analogue that cannot be metabolised by potato tuber cells but leads to a stimulation of starch synthesis when supplied to wild-type tuber discs (Fernie et al., 2001). This could indicate that the sucrose-dependent redox activation of AGPase is not necessarily linked to sucrose metabolism or sucrose cleavage into glucose and fructose.
An endogenous hexokinase activity is involved in the signalling pathway leading to glucose-dependent redox activation of AGPase
In contrast to the results with the sucrose analogue palatinose, the use of different glucose analogues show that phosphorylation of hexoses by hexokinase is indeed a necessary step in the pathway leading to glucose-dependent AGPase redox regulation in potato tuber discs (Figure 2). This provides evidence for the involvement of hexokinase in the hexose-sensing pathway leading to redox modulation of AGPase.
Ectopic expression of invertase led to low sucrose and high glucose levels and to increased AGPase redox activation within tubers, providing in planta evidence that glucose acts independently of sucrose (Table 2). Interestingly, expression of a bacterial glucokinase in combination with invertase in the cytosol increased the hexose phosphorylation activity, but decreased the AGPase redox state in transgenic tubers (Table 3). This provides evidence that the hexose-phosphorylation-dependent signal transduction leading to redox activation of AGPase needs the activity of a specific endogenous hexokinase, rather than metabolism of glucose per se.
More work will be required to identify and verify the remaining components of the sugar-signalling network leading to redox modulation of AGPase. It will be of special interest to investigate how the sugar signal is transferred from the cytosol to the plastid, and what are the specific steps that transform the sugar to a redox signal. In vitro studies show that potato tuber AGPase can be activated by thioredoxin m and f from leaves (Ballicora et al., 2000). More studies are needed to identify the regulatory thioredoxins in potato tubers and to test their functional significance for storage processes.
Plant material and growth conditions
Wild-type potato plants (Solanum tuberosum L. cv. Desiree) were propagated in tissue culture and grown on soil in the greenhouse as described previously by Tiessen et al. (2002). The generation of transgenic potato plants expressing tuber-specific antisense SnRK1 have been described by Purcell et al. (1998). The generation and extensive characterisation of transgenic potato lines expressing invertase in the apoplast (INV1-33) or cytosol (INV2-30) have been described by Buessis et al. (1997), Hajirezaei et al. (2000), Roessner et al. (2001), Sonnewald et al. (1997) and Tauberger et al. (1999). The generation of the invertase-GK3-overexpression potato plants has been described by Trethewey et al. (1998). The term ‘growing tubers’ is used for tubers (10–80 g FW) attached to plants (6–9 weeks old) that look healthy and have only green leaves.
Preparation of discs and incubation of potato tuber tissue
Tuber tissue for individual experiments was prepared from several tubers derived from one or several plants. After harvesting the tuber, the skin was rapidly and partially removed and the tuber was immersed in 1 l buffer solution (1 mm EDTA, 10 mm MES-KOH (pH 6.5)), and 2-mm slices were cut underwater using a common kitchen slicer. In the experiments described in Figure 1, the slices were prepared as above but including or omitting EDTA in the buffer solution. Rapidly thereafter, subpools of slices (approximately 10 g FW) were dried with tissue paper and frozen in liquid N2 (time 0) or transferred into flasks with 10 ml g−1 FW buffer solution supplemented with sugars and EDTA as specified in the legends and incubated with continuous shaking for a total of 1 or 2 h at 25°C. After incubation, the tuber slices were washed three times with 500 ml water for 10 sec and dried with tissue paper to remove exogenous sugars, frozen in liquid N2 and stored at −80°C until use.
Measurement of sugar-uptake rate
Tuber discs (10 discs per condition) were prepared submerged in either MES-buffer (10 mm MES-KOH, pH 6.5) or MES-EDTA buffer (1 mm EDTA, 10 mm MES-KOH, pH 6.5) and incubated in 4 ml of MES or MES-EDTA buffer supplemented with 2, 10, 50, 100 and 200 mm glucose or sucrose (containing 36 kBq U-14C-radiolabelled sugar). After 3-h incubation, tuber discs were washed three times with 20 ml of water for 30 sec and dried with tissue paper to remove exogenous label. Tuber tissue was extracted in 80% ethanol, homogenised, and the total label (soluble + insoluble) was determined by scintillation counting. Aliquots of the incubation solution were taken to determine the specific activity of the sugar, and the values were used to calculate total sugar uptake in micromole sugar per gram FW.
Frozen tuber tissue was homogenised in liquid N2, and 100 mg was extracted rapidly (1 min) with 1 ml of extraction buffer (50 mm Hepes-KOH (pH 7.8), 5 mm MgCl2 and 3 mm dithiothreitol (DTT)) at 4°C. Although the extraction buffer contained 3 mm DTT, this alone was not enough to change the redox activation state of AGPase after rapid extraction, but did maintain the original activation state for longer and improve the reproducibility of the AGPase measurements (data not shown). The AGPase can only be fully activated by using high concentrations of DTT (10 mm), adding a substrate like ADPglc (at least 300 µm) and incubating for at least 15 min (data not shown). After mixing the fresh extract vigorously for 20 sec and centrifugation for 30 sec at 10 000 g at 4°C, 10 µl of the supernatant was used immediately for assay in microtiter plate format. AGPase activity was measured in the pyrophosphorolysis direction as detailed for Vsel and Vred by Tiessen et al. (2002), with the exception that 10 mm DTT was used for activation (Vred) assay and the pre-incubation procedure was omitted. Instead, the Vred reaction was followed online at 340 nm until AGPase had reached maximal activation after approximately 15 min. The activation state of AGPase is defined as the ratio of Vsel (–DTT) to Vred (+DTT) in the pyrophosphorolysis reaction at a range of concentrations of ADPglc (0.4–1 mm) and 1.3 mm PPi in the absence of Pi and 3PGA. All measurements were performed with freshly and rapidly prepared extracts.
Protein extraction and non-reducing SDS–PAGE and gel blot analysis were carried out as described by Tiessen et al. (2002) using a rabbit antiserum raised against an overexpressed and purified His-tagged AGPB protein from potato (Tiessen et al., 2002).
Measurement of respiration rate of tuber discs
The respiration rate of tuber discs was measured with an oxygen electrode (Hansatech, Kings Lynn, UK) fitted to a recorder. The anode was covered with cigarette paper impregnated with 1 m KCl and isolated with a semiselective membrane. Two discs (2 mg × 82 mg FW) were placed in a chamber (15 mm × 40 mm2) containing 1 ml of the pre-incubation solution containing the sugars as indicated in the figure legend. The solution was stirred vigorously with a small magnetic stirrer. The temperature was kept constant at 20°C. The electrode was calibrated by measuring the oxygen concentration of air-saturated solution (21% O2 = 280 nmol ml−1) and subtracting the value for 0% O2.
This work was generously supported by the Deutsche Forschungsgemeinschaft, Grant Ge 878/1–3 (to A.T., A.B. and P.G.). We wish to thank Mark Stitt for support, John Lunn and Helene Vigeolas for critical readings of the manuscript and Janneke H.M. Hendriks, Joost T. van Dongen and Anna Kolbe for stimulating discussions. We are grateful to Eva M. Farré and Ute Rössner for providing tuber samples of invertase- and GK3-overexpressing plants, to Anja Fröhlich for technical assistance and Karin Köhl, Helga Kulka and Britta Hausmann for taking care of plants. We thank Christine Franck for performing preliminary analysis on the SnRK1 lines during her stay as a visiting scientist in the P.G. laboratory.