SPS expression in growing tubers
The growing tubers of transformants contained decreased amounts of SPS transcript (Fig. 1a) and decreased SPS protein (Fig. 1b). To allow identification of the SPS forms, a sample from cold-stored tubers was included. Growing wild-type tubers contained mainly SPS-1a, with a small amount of SPS-2, whereas SPS-1b was not detectable. The band corresponding to SPS-1a was strongly decreased in the antisense lines 5–15 and 5–7, and very low in lines 1–74 and 5–59. As in leaves (see above), SPS-2 was less strongly affected. To assess the effect on overall SPS activity, an assay was used that contained saturating concentrations of hexose phosphates (Fig. 1c). Compared to wild-type tubers, overall SPS activity is decreased to 44%, 38%, 20% and 14% of the wild-type activity in the transformant lines 5–15, 5–7, 1–74 and 5–59, respectively. ADP glucose pyrophosphorylase (AGPase) and sucrose synthase activities were similar in wild-type and transformant tubers (data not shown).
Figure 1. SPS expression in tubers.
(a) Steady-state Sps transcript. Twenty micrograms of total RNA were loaded per lane.
(b) SPS protein detected by immunoblotting. Top arrow indicates SPS-2; second arrow, SPS-1b; third arrow, SPS-1a. Bottom arrow on left side, molecular weight marker, 116 kDa.
(c) Activity under optimal assay conditions: Vmax assay (12 mm Fru6P, 36 mm Glc6P and 6 mm UDPGlc).
(d) Activity under limiting assay conditions: Vsel assay (2 mm Fru6P, 6 mm Glc6P, 6 mm UDPGlc and 5 mm Pi).
Download figure to PowerPoint
SPS activity was also measured in an assay that contained limiting hexose phosphate concentrations and phosphate (Vsel; Fig. 1d). In these assay conditions, the reduction of SPS activity in the transformants was much smaller. A similar effect was found in the leaves (data not shown). This weak inhibition of SPS activity in the transformants in the Vsel assay could be explained by post-translational activation of the remaining enzyme leading to changes in the kinetic properties. Alternatively, it could be explained if SPS-2, which represents a large portion of the residual SPS protein in the transformants (Fig. 1b), has a higher affinity for hexose phosphates or a lower sensitivity to inhibition by phosphate. To distinguish between these possibilities, tuber slices from line 5–59 were incubated with the protein phosphatase inhibitors 5 nmol g–1 FW calyculin or 19 nmol g–1 FW okadaic acid for 3 h, and then assayed for SPS activity in the presence of saturating hexose phosphates (Vmax assay) or the presence of limiting hexose phosphates plus phosphate (Vsel assay). The inhibitor concentrations were high enough to reverse the activation of potato tuber SPS by mannose or high mannitol in wild-type tubers (Reimholz et al. 1994; Reimholz et al. 1997). The ratio of Vsel activity to Vmax activity was higher in 5–59 than in the wild-type tubers in discs prepared directly from the tubers (0.20 ± 0.034 compared to 0.10 ± 0.004, mean ± SE, n = 3, original data not shown) and remained high at the end of a 3 h incubation, irrespective of whether the discs were incubated in the absence of protein phosphatase inhibitors (0.28 ± 0.03) or with okadaic acid (0.28 ± 0.04) or calyculin (0.24 ± 0.03). These results indicate that the high ratio found in the transformant is not primarily due to protein phosphorylation.
Metabolism in water-stressed tuber discs
Discs from wild-type tubers and four transformant lines (5–15, 5–7, 1–74 and 5–59) were incubated for 2 h with 2 mm14C glucose and 10, 100 or 300 mm mannitol, corresponding to an external water deficit of –0.02, –0.24 and –0.72 MPa, respectively. SPS activity responded to water stress in wild-type material as described previously (Geigenberger et al. 1997). Maximum SPS activity (measured in the assay that contained saturating hexose phosphates; Vmax) remained unaltered (Fig. 2a). SPS activity (measured in the assay that contained limiting hexose phosphates and phosphate, Vsel) (Fig. 2b) and the ratio of Vsel activity/Vmax activity (not shown) increased twofold as the mannitol concentration increased. The increase of Vsel in water-stressed wild-type tubers is prevented by protein phosphatase inhibitors (Geigenberger et al. 1997).
Figure 2. Metabolism of 2 mm[14C]glucose in potato slices of plants with reduced SPS incubated at 10, 100 and 300 mm mannitol.
After 90 min pre-incubation, [14C]glucose was added for 2 h until slices were harvested to determine radioactivity in sucrose and starch.
(a,b) SPS activity measured in the slices after 3 h incubation, (a) Vmax activity, (b) Vsel activity. (c) Uptake of [14C]glucose. (d,e) Distribution of label in sucrose (d) and starch (e) as a percentage of uptake. (f) Relation of label between sucrose and starch. Data are means ± SE, n = 3–4.
Download figure to PowerPoint
Maximum SPS activity was inhibited by 35–40%, 40–55%, 63–72% and 67–78% in 5–15, 5–7, 1–74 and 5–59, respectively. The magnitude of the inhibition was not modified by the mannitol treatment (Fig. 2a). Increasing mannitol led to an increase of the Vsel activity in 5–15 and 5–7, but not in 1–74 and 5–59 (Fig. 2b). In the latter two lines, the Vsel/Vmax ratio was already between 0.3 and 0.4 in 10 mm mannitol (data not shown) and did not increase further at high mannitol.
Glucose uptake was stimulated at high mannitol (Fig. 2c) (see also Geigenberger et al. 1997). Discs from wild-type tubers and the transformants showed similar rates of [14C]glucose uptake (Fig. 2c). In the following figure panels, the labelling of the different fractions is given as a percentage of the absorbed label. In wild-type material, label incorporation into sucrose increased (Fig. 2d) and label incorporation into starch decreased (Fig. 2e) as the mannitol concentration was increased (see also Geigenberger et al. 1997). This response was modified in the discs from transformants. When discs in 10 mm mannitol are compared, decreased expression of SPS inhibited label incorporation into sucrose (Fig. 2d). However, only a small portion of the label was converted to sucrose in non-stressed discs, and the inhibition of sucrose synthesis was not accompanied by a significant increase of label incorporation into starch (Fig. 2e). When discs in 100 mm or 300 mm mannitol are compared, the transformants contained less label in sucrose (Fig. 2d), and more label in starch (Fig. 2e) than wild-type discs. This change was seen in all four transformant lines at 100 mm mannitol, but only in transformant lines 1–74 and 5–59 with the largest decrease in expression of SPS at 300 mm mannitol. The increase in partitioning of label to sucrose that is typical for water-stressed wild-type material (see Geigenberger et al. 1997) was partially inhibited in line 1–74 and almost totally prevented in line 5–59 (Fig. 2d). The accompanying decrease in partitioning of label to starch was almost totally suppressed in 1–74 and 5–59 (Fig. 2e). The water stress-induced increase of the sucrose-to-starch labelling ratio was partly or completely suppressed in 1–74 and 5–59 (Fig. 2f).
Metabolism was investigated at a wider range of mannitol concentrations in discs from wild-type tubers and the antisense line 5–59 (Fig. 3). SPS activity was decreased by 75% in this batch of 5–59 tubers, compared to wild-type tubers (data not shown). Tuber slices were incubated in 2 mm[14C]glucose with 0, 50, 150, 250 or 500 mm mannitol, corresponding to an external water deficit of 0, –0.12, –0.36, –0.60 and –1,2 MPa, respectively. Discs incubated on buffer gained marginally in fresh weight (see also Geigenberger et al. 1997), discs incubated on 150–250 mm mannitol lost 6–10% of their fresh weight, and discs incubated on 500 mm mannitol were under severe stress and lost 34% of their initial fresh weight (Fig. 3a). Similar changes in disc weight were found for wild-type and transformant material.
Figure 3. Metabolism of 2 mm[14C]glucose in potato slices of plants with reduced SPS incubated at 0, 50, 150, 250 and 500 mm mannitol as in Fig. 2
(a) Fresh weight per disc. (b) Uptake of [14C]glucose. (c) Label remaining in glucose as percentage of uptake. (d,e) Incorporation of label into (d) sucrose and (e) starch. (f) Relation of label between sucrose and starch. Data are means ± SE, n = 3–4.
Download figure to PowerPoint
Discs from wild-type and 5–59 tubers had similar rates of glucose uptake and metabolism (Fig. 3b,c). As the mannitol concentration was increased, wild-type discs incorporated more label into sucrose and less label into starch (Fig. 3d,e). This response was altered in 5–59. At mannitol concentrations up to 250 mm (corresponding to –0.6 MPa), the stimulation of sucrose synthesis was almost completely suppressed in 5–59 (Fig. 3d), and the inhibition of starch synthesis (Fig. 3e) was much weaker than in wild-type discs. Whereas the sucrose-to-starch labelling ratio increased dramatically in the wild-type, it remained at a low value in the transformant (Fig. 3f). More severe water deficits (500 mm mannitol, corresponding to –1.2 MPa) led to a stimulation of sucrose synthesis and inhibition of starch synthesis, although the changes were less marked than in wild-type discs.
Metabolism in intact non-stressed growing tubers attached to the plant
Growing tubers of well-watered plants were harvested at the start of the photoperiod and analysed for metabolite levels (Table 1). A similar sucrose level was found in tubers from wild-type plants and the transformants. Decreased expression of SPS was accompanied by a marked and significant increase of UDPGlc, a slight increase of Glc6P and Fru6P, and a slight and non-significant increase of 3PGA especially in lines 1–74 and 5–59 where SPS expression was most strongly inhibited.
Table 1. Metabolite levels in intact growing tubers of potato plants with reduced expression of SPS, harvested at the start of the photoperiod
|Sucrose||13.8 ± 4.8||15.3 ± 0.8||10.7 ± 2.9||14.1 ± 4.0|
|UDPGlc||35.8 ± 2.5||97.0 ± 17.4||59.0 ± 18.8||83.1 ± 20|
|Glc-1-P||6.0 ± 1.4||5.6 ± 1.1||6.7 ± 1.3||8.0 ± 1|
|Glc-6-P||90.3 ± 7.3||88.1 ± 27.8||107.6 ± 9.4||114.7 ± 13.1|
|Fru-6-P||19.0 ± 2.2||17.5 ± 2.6||20.7 ± 0.7||23.0 ± 2.4|
|3-PGA||51.6 ± 2.6||51.3 ± 5.7||57.1 ± 2.3||65.5 ± 2.0|
Fluxes were investigated by injecting carrier-free [14C]glucose into intact tubers that were attached to the plant. This experiment was carried out with line 5–59. SPS activity was decreased by 70% in 5–59 (445 ± 5 nmol gFW–1 min–1 in wild-type tubers, 135 ± 10 nmol gFW –1 min–1 in transformant tubers, original data not shown). The experiment commenced 4 h into the photoperiod. Wild-type and 5–59 tubers metabolised the 14C glucose at the same rate (data not shown). Tubers from 5 to 59 showed a small significant decrease in the labelling of sucrose and a slight but non-significant increase in the labelling of starch (original data not shown). The ratio of label in sucrose to label in starch showed a small but significant decrease from 0.19 ± 0.005 in the wild-type tubers to 0.12 ± 0.021 in the transformant tubers (significant, P < 0.05).
Starch content was investigated in five harvests from independently grown batches of well-watered plants (Table 2). No significant differences in the starch content were observed between the genotypes when the plants were grown in the winter season in a greenhouse in Heidelberg under low light conditions (Table 2). When grown under higher irradiance in a growth chamber, there was a trend to a slightly higher tuber starch content in the transformants compared to wild-type tubers, but the increase was only significant in one experiment with 5–59.
Table 2. Starch content in developing tubers from 8- to 10-week-old plants with reduced expression of SPS
|Greenhouse, January||n = 10–14||346 ± 18||375 ± 25||– *||288 ± 10||326 ± 16|
|Greenhouse, November||n = 10–12||387 ± 44||322 ± 29||367 ± 31||326 ± 31||387 ± 12|
|Growth chamber, 12 h light, 250μE||n = 6–8||463 ± 32||497 ± 43||567 ± 29||519 ± 18||521 ± 29|
|Growth chamber, 14 h light, 250μE||n = 8–10||476 ± 34||592 ± 41||– ||553 ± 37||552 ± 34|
|Growth chamber, 14 h light, 350 μE||n = 15||618 ± 20||– ||– ||673 ± 65||824 ± 58|
Metabolism in intact water-stressed growing tubers on the plant
Wild-type potato plants and two transformants with a strong reduction of SPS (1–74 and 5–59) were grown in a growth cabinet under 350 μmol photons m–2 s–1 irradiance with daily watering for 6 weeks, by which time tubers had been initiated and then divided into three groups. One group was watered daily until harvest, 17 days later. A second group was not watered during the first 12 days, by which time the relative water content of the soil had decreased to 14% and the leaves were visibly wilted. For the next 5 days, the pots were weighed and the daily water loss substituted by rewatering, to keep the water content of the soil at a low but constant level. During this period, the leaves remained wilted. Water stress will decrease the rate of photosynthesis and therefore, as a control, a third group of wild-type plants were transferred to low light (20 μmol photons m–2 s–1) for the last 17 days.
Water stress led to a 38% increase of Vmax SPS activity in wild-type tubers (Fig. 4a). There was an even larger increase of SPS activity when it was assayed with limiting hexose phosphates and phosphate (Fig. 4b), and the ratio of Vsel to Vmax activity increased (not shown) as in water-stressed discs (see above). This increase of SPS activity and of the Vsel to Vmax ratio did not occur in low light. The transformants showed a strong reduction of SPS activity that was retained under water stress (Fig. 4a).
Figure 4. Metabolism of high specific activity [14C]glucose injected into intact tubers attached to non-stressed or water-stressed potato plants.
Wild-type potato plants and two lines of transformants were grown in a growth cabinet under 350 μE m–2 s–1 irradiance with daily watering for 6 weeks, at which point tubers had already been initiated. The plants were then divided into three groups. One group was watered daily through to harvest 17 days later. A second group was not watered during the first 12 days. After this time interval leaves were visibly wilted and the relative water content of the soil was decreased down to 14%. For the last 5 days, pots were weighted and the daily water loss substituted by rewatering to keep the water content of the soil at a low but constant level. During this time, leaves remained in a wilted condition. A third group of wild-type plants were transferred to low light (20 μE m–2 s–1) and harvested 17 days later.
(a,b) SPS activity in the tubers, (a) Vmax activity, (b) Vsel activity. (c–f) Metabolism of carrier-free [14C]glucose injected into intact tubers attached to the plants: (c) label metabolised as percentage of total injected, (d–f) label incorporation into (d) sucrose, (e) starch, (f) glycolytic intermediates (anions+cations), as percentage of label metabolised. Data are the mean ± SE, n = 4 separate tubers from four different plants.
Download figure to PowerPoint
To investigate fluxes, high specific activity [14C]glucose was injected into intact tubers attached to the plants (Fig. 4c–f). In daily watered plants, the transformants showed a decrease in labelling of sucrose (Fig. 4d), a slight increase in labelling of starch (Fig. 4e), and the ratio of label in sucrose:label in starch decreased from 0.08 ± 0.01 in the wild-type to 0.042 ± 0.007 and 0.034 ± 0.011 in 1–74 and 5–59, respectively (mean ± SE, n = 4). Water stress led to a threefold stimulation of sucrose synthesis in wild-type plants, but did not stimulate sucrose synthesis in the transformants. Water stress led to a 33% inhibition of label incorporation into starch in wild-type tubers, but had a much weaker inhibitory effect in 1–74 and 5–59 (20% and 10% inhibition of starch synthesis, respectively). During water stress, the sucrose:starch labelling ratio increased approximately fivefold to 0.456 ± 0.228 in wild-type tubers, whereas in 1–74 and 5–59 it remained very low (0.043 ± 0.029 and 0.039 ± 0.006, respectively). During water stress, a larger proportion of the label was directed into glycolysis leading to increased labelling of ionic components (mainly phosphate esters, organic acids and amino acids; see Fig. 4f).
In tubers of wild-type plants transferred to low light, label in starch decreased and labelling of glycolytic products and intermediates markedly increased but, in contrast to water-stressed wild-type tubers, no significant increase in the labelling of sucrose was observed and the sucrose-to-starch labelling ratio remained low (0.148 ± 0.025, n = 4).
Effect of water stress on biomass allocation in whole potato plants
Sucrose levels remained high in tubers growing on water-stressed wild-type plants, whereas they decreased by 75% in low light (Fig. 5a). Tubers of the transformants contained marginally less sucrose than wild-type tubers, and this trend became more marked in water-stressed transformants (Fig. 5a). Tubers of the transformants had a significantly lower water content compared to the wild-type, and this trend was also sustained under water stress (Fig. 5b). Water stress led to a slight but non-significant increase of the starch content on a dry-weight basis in wild-type and transformants tubers (Fig. 5f).
Figure 5. Effect of water-stress on biomass allocation in whole potato plants.
Plants used in the same experiment as in Fig. 4 were harvested to determine: (a) the sucrose content of the tubers, (b) the water content of the tubers, (c) shoot dry-weight, (d) tuber dry-weight, (e) the ratio between shoot and tuber dry-weights, and (f) the starch content of the tubers on a dry-weight basis. Data for sucrose, starch and water content of the tubers are the mean ± SE, n = 12 separate tubers from four different plants. Data for biomass allocation are the mean ± SE, n = 4 different plants.
Download figure to PowerPoint
Changes in whole plant allocation were also investigated. In wild-type plants, water stress led to a larger reduction of shoot dry weight (50%) (Fig. 5c) than of tuber dry weight (29%) (Fig. 5d). As a result, the ratio of tuber dry weight to shoot dry weight rose in wild-type plants under water stress (Fig. 5e). This increase of the tuber-to-shoot ratio was not seen in wild-type plants in low light (Fig. 5e), even though there was a similar overall decrease of plant size (Fig. 5c,d). Well-watered transformants had a slightly higher tuber to shoot ratio than wild-type plants (Fig. 5e). However, water stress led to a larger inhibition of tuber growth in the transformants (43–48%) than in wild-type plants (30%) (Fig. 5d). The tuber:shoot ratio of the transformants therefore remained unaltered or decreased slightly in response to water stress (Fig. 5e), rather than increasing as in wild-type plants.