INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

The primary pathway of carbon fixation in C₃ plants is the reductive pentose-phosphate (Calvin) cycle located in the chloroplast stroma. The Calvin cycle can be divided into three distinct phases: carboxylation of ribulose-1,5-bisphosphate, reduction of 3-phosphoglycerate, and regeneration of the CO₂ acceptor molecule. The enzyme sedoheptulose-1,7-bisphosphatase (SBPase; EC 3.1.3.37) functions in the regenerative phase of the Calvin cycle where it catalyzes the dephosphorylation of SBPase. SBPase activity increases in the light by more than 10-fold as a result of light-modulated activation by thioredoxin f (Breazeale et al., 1978; Wirtz et al., 1982). Light-induced changes in stromal Mg²⁺ levels and pH also regulate SBPase activty (Portis et al., 1977; Purczeld et al., 1978; Nishizawa and Buchanan, 1981; Woodrow and Walker, 1982; Woodrow et al., 1984). The highly regulated catalytic activity of SBPase, together with modeling studies, has led to the suggestion that this enzyme may play an important role in the control of carbon flux through the Calvin cycle (Petterson and Ryde-Petterson, 1989; Poolman et al., 2000). In addition, the location of SBPase at the branch point between regeneration of the CO₂ acceptor molecule and biosynthesis of starch and Suc could potentially influence the distribution of carbon between these three competing pathways. An antisense approach recently was taken to modify levels of SBPase in transgenic plants to assess the contribution of this enzyme to the control of carbon flux through the Calvin cycle. Transgenic tobacco (Nicotiana tabacum L. cv Samsun) plants with levels of SBPase activity less than wild type were found to have decreased rates of photosynthetic carbon fixation and altered carbohydrate levels in mature source leaves (Harrison et al., 1998; Raines et al., 2000). Quantitative flux control analysis (Kacser and Burns, 1973; Fell, 1997) of the photosynthetic data from the SBPase antisense plants showed that SBPase exerted considerable control on carbon assimilation, particularly under saturating light and CO₂ conditions (Raines et al., 2000).

The analysis of the transgenic plants with reduced SBPase activity has revealed that SBPase plays an important role in regulating carbon flow into the regenerative phase of the Calvin cycle in mature source leaves (Harrison et al., 1998; Raines et al., 2000). However, in dicot species photosynthetic capacity changes dependent on the stage of leaf development, increasing during leaf expansion to a maximum in the mature fully expanded leaves, which is then followed by a rapid decline during post-maturity and senescence (Gepstein, 1988). At present we have no information on the effect of reduced SBPase activity on photosynthetic carbon fixation during leaf expansion and maturation. Studies on transgenic tobacco plants containing reduced levels of Rubisco have been extended to investigate the role of Rubisco in the regulation of photosynthesis at different stages of leaf development (Jiang and Rodermel, 1995). This analysis revealed that the overall pattern of change in photosynthetic rates in these leaves was similar in wild-type and Rubisco antisense plants, indicating that the leaf developmental program was insensitive to even sharp reductions in Rubisco content. However, more recently, growth analysis has revealed that reductions in Rubisco of 80% altered the timing of shoot development, specifically delaying an early phase, possibly due to reduced source strength or to changes in the sink-source balance (Jiang and Rodermel, 1995; Tsai et al., 1997). It is clear that this analysis of the Rubisco antisense plants has added a new dimension to the analysis of transgenic plants and provided further insight into the interactions between primary metabolism and development throughout the life cycle of the plant. Given the impact of small reductions in SBPase on the photosynthetic capacity and patterns of carbohydrate accumulation in fully expanded leaves, we have extended our study of these antisense plants to examine the effects of reduced SBPase activity in developing leaves. The results of this analysis are presented here.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Experimental Design

The SBPase antisense transgenic tobacco (Tsa) plants used in these experiments were T1 progeny from four independent lines, previously shown to segregate plants with a range of reductions in SBPase activity (Harrison et al., 1998). The growth rates of the transgenic plants were slower than the wild-type plants. To allow comparisons to be made between leaves at the same developmental stage, wild-type and SBPase antisense plants were grown during the summer season in the greenhouse until 23 leaves were produced. At this stage the height of the majority of the transgenic plants was similar to wild type, with the exception of two antisense plants that were significantly shorter due to a reduction in the internode distance.

In vivo measurements of CO₂ assimilation rates were made on selected alternate leaves, including young expanding leaves (leaf numbers 14 and 16), new fully expanded leaves (10 and 12), and a mature, fully expanded leaf (8). Immediately following photosynthetic measurements, leaf tissue was harvested directly into liquid N₂ for SBPase activity assays, western-blot analysis, and chlorophyll and carbohydrate measurements. None of the plants used in this analysis displayed the leaf veinal chlorosis observed previously in antisense plants with very severe reductions in SBPase activity (Harrison et al., 1998).

The levels of SBPase protein in wild-type and transgenic plants during leaf maturation and expansion were established using western-blot analysis (Fig. 1). The results show that in wild-type plants SBPase protein levels remained constant in the younger leaves but decreased slightly in the two oldest leaves analyzed. A similar developmental pattern was seen in the SBPase antisense plant; however, the decrease in the mature leaves was more pronounced, particularly in the plants with the lowest levels of SBPase.

View larger version (46K): [in this window] [in a new window]   Figure 1.   Calvin cycle protein levels in leaves of greenhouse-grown wild-type (WT) and Tsa plants at the 23-leaf stage. Samples were taken from alternate leaves, from leaf 16 (young expanding leaf) to leaf 8 (fully expanded mature leaf), immediately following photosynthesis measurements and total leaf soluble proteins extracted. Protein extracts were separated by SDS-PAGE, loaded on an equal leaf area basis. Western blots were probed with polyclonal antibodies raised against SBPase, fructose-1,6-bisphosphatase (FBPase), and phosphoribulokinase (PRKase), and proteins detected using enhanced chemiluminescence (Amersham International, Little Chalfont, Buckinghamshire, UK).

We had observed previously that, in fully expanded leaves, reductions in SBPase protein levels did not affect the levels of other Calvin cycle enzymes (Harrison et al., 1998). To determine if this was also the case during leaf expansion and maturation, SBPase protein levels were compared with those of two other Calvin cycle enzymes, FBPase and PRKase (Fig. 1). The results indicated that in the wild-type plants, the relative amounts of these three Calvin cycle enzymes remained constant throughout leaf development. In the antisense plants the levels of FBPase and PRKase followed a similar pattern to that observed in the wild-type plants. No reduction in the levels of either FBPase or PRKase was observed in the antisense plants; however, quantification of these data suggested that the levels of PRKase increased slightly in response to reduced SBPase protein levels (data not shown). No significant differences were observed in the protein content per unit leaf area between wild-type and SBPase antisense plants during leaf development.

Photosynthetic Capacity during Leaf development in Wild-Type and Antisense SBPase Tobacco Plants

Total SBPase activity was measured in alternate leaves of wild-type and SBPase antisense plants, sampled as described above. The antisense plants contained lower levels of SBPase activity in all of the leaves examined when compared with equivalent wild-type leaves (Fig. 2). The trend observed in both the wild-type and antisense plants was for SBPase activity to be highest in the youngest leaves and to decline (on a leaf area basis) by between 30% to 50% during leaf maturation. In the wild-type plants this decrease in activity occurred consistently between leaves 12 and 10; however, in the antisense plants the pattern of decline was more variable (Fig. 2).

View larger version (24K): [in this window] [in a new window]   Figure 2.   SBPase activity during leaf development in wild-type and SBPase antisense tobacco plants. Enzyme activity measurements were made using total leaf-soluble proteins extracted from greenhouse grown wild-type (WT) and Tsa plants sampled at the 23-leaf stage, immediately following photosynthesis measurements. Alternate leaves, from leaf 16 (young expanding leaf) to leaf 8 (fully expanded mature leaf) were assayed and data points for WT plants are the mean ± SE (n = 5) and for the Tsa plants are the mean from duplicate extracts from individual leaves on each plant. Measurements from individual extracts were made in triplicate.

The impact of decreased SBPase activity on light-saturated CO₂ assimilation rates (A_sat) was determined during leaf expansion and maturation in wild-type and antisense plants (Fig. 3). The response of A_sat to reductions in SBPase activity unexpectedly changed during leaf development. The rate of photosynthetic carbon assimilation in the two youngest leaves (16 and 14) examined was unaffected, or even increased, when SBPase activity was reduced by up to 50%. However, similar reductions in SBPase activity in the newly fully expanded leaves (10 and 12) and the mature source leaf (8) resulted in significant reductions in A_sat. These differences in the response of photosynthetic carbon assimilation to reductions in SBPase activity were quantified by plotting the data for each leaf (data in Fig. 3) on a doubled logarithmic scale (Fell, 1997). The gradients of the resulting straight lines, corresponding to the flux control coefficients for SBPase, changed during leaf development from 0.2 in the youngest expanding leaves to between 0.35 and 0.5 in fully expanded mature leaves.

View larger version (12K): [in this window] [in a new window]   Figure 3.   The response of Asat to reduced SBPase activity during leaf development was measured in alternate leaves (leaf 16, young expanding leaf, to leaf 8, fully expanded mature leaf) of wild-type (WT) and Tsa plants at the 23-leaf stage. Plants were grown in the greenhouse with supplementary lighting giving between 700 to 1000 µmol m2 s1 and Asat measurements using an open gas exchange system under saturating light of 1,000 µmol m2 s1 with 24°C leaf temperature at ambient CO2. Measurements were made on the same day on all the leaves of an individual plant. Data points for the wild-type (black circles) plants are the mean ± SE, n = 5 and for the individual transgenic plants (white circles) are the mean of two measurements.

The pattern of development of photosynthetic capacity during leaf expansion and maturation in the antisense SBPase plants was different from wild-type plants. In wild-type plants a pattern typically found during dicot leaf development was observed: A_sat was lowest in the youngest leaf examined (16) and then increased to a peak in leaf 12, just prior to attaining full expansion. A slight decrease in A_sat was then observed in the mature, fully expanded leaves. In contrast, the trend observed in the transgenic plants was for A_sat to be highest in the youngest leaves and decrease as the leaves matured. The chlorophyll content and chlorophyll a/b ratio were similar in all the leaves of both the wild-type and SBPase antisense plants (data not shown).

Carbohydrate Status during Leaf Development in Antisense SBPase and Wild-Type Plants

The effect of reduced SBPase levels on the carbohydrate profile during leaf expansion and maturation was determined in leaves of wild-type and SBPase antisense plants. The pattern of carbohydrate accumulation was broadly similar in all the plants, with the highest concentrations of Glc, Suc, and starch in the youngest leaves (16 and 14) examined and lowest levels in fully expanded leaves (12, 10, and 8; Fig. 4). However, there were significant differences in the amounts of these carbohydrates found in the antisense plants dependent on the level of SBPase activity. In the antisense plants with the smallest reductions in SBPase activity, the levels of Glc, Suc, and starch were close to or higher than in wild-type plants, most noticeably in the youngest leaves (16 and 14; Fig. 4). In contrast, plants with more severe reductions in SBPase activity had low levels of Glc and starch in all the leaves studied, although Suc was maintained at near wild-type levels.

View larger version (34K): [in this window] [in a new window]   Figure 4.   Carbohydrate content during leaf development. Glc, Fru, Suc, and starch were measured in samples harvested at the end of the light period from the same leaves used for SBPase activity and photosynthesis measurements. Data points for the wild-type plants (black circles) are the mean ± SE (n = 4), and for the transgenic SBPase antisense plants (white circles) are the mean of triplicate measurements of single extracts from individual leaves on each plant.

The relationship between the rates of photosynthetic carbon assimilation and carbohydrate levels was also examined (Fig. 5A). In the mature leaves (8, 10, and 12) of the SBPase antisense plants, levels of starch declined in parallel with reductions in photosynthetic rate. In contrast, in the young expanding leaves (14 and 16) the levels of starch varied dramatically over a relatively narrow range of A_sat values. Levels of soluble carbohydrates (Fru, Glc, and Suc) showed more variability but appeared to be less sensitive to reductions in photosynthetic capacity than starch, with the exception of the youngest leaf.

View larger version (29K): [in this window] [in a new window]   Figure 5.   The relationship between carbohydrate levels and photosynthetic carbon assimilation. A, Starch and Suc contents of leaves 8 through 16 of wild-type (black symbols) and SBPase antisense plants (white symbols) were plotted against the values for Asat obtained for each leaf (data shown in Fig. 3). B, Levels of Fru, Glc, Suc, and starch in leaf 16 plotted against Asat in leaves 8 (circles) and 10 (squares) of the same plant.

The data in Figure 5A suggested that the photosynthetic capacity of the young leaves could not account for the changes in carbohydrate levels in these leaves. One explanation for this could be that it was the photosynthetic capacity of the mature leaf that was determining the carbohydrate status of the young leaves. To examine this hypothesis the carbohydrate concentrations in leaf 16 were plotted against A_sat in the mature source leaves (8 and 10; Fig. 5B). When Suc and starch levels in leaf 16 were below that of wild type, a strong correlation between photosynthetic capacity of the source leaves and the carbohydrate status of the young sink leaf (16) was evident. In contrast, in the plants with higher starch and Suc levels than wild type, no relationship was observed between A_sat in the source leaves (8 and 10) and carbohydrate levels in leaf 16.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

In this paper we have examined the relationship between SBPase activity and photosynthetic capacity during the expansion and maturation stages of leaf development. One striking difference between the SBPase antisense and wild-type plants was found in the photosynthetic rates obtained for leaves approaching and attaining full expansion. In wild-type plants photosynthetic capacity followed the expected developmental pattern, with lowest rates in the youngest leaves and maximum rates in the mature fully expanded leaves (Gepstein, 1988; Dietz and Heilos, 1990; Harn et al., 1993). In contrast, in the SBPase antisense plants the youngest leaves had the highest photosynthetic capacity, attaining rates as high, and in some cases higher than, in equivalent leaves of wild-type plants. This result indicated that SBPase was not limiting carbon assimilation in these young expanding leaves where the transition from sink to source status was ongoing. In keeping with this, a small negative value was obtained for the flux control coefficient of SBPase on CO₂ assimilation rates in these young leaves. However, as leaf expansion was completed, photosynthetic rates decreased in the antisense plants in response to reduced SBPase activity. Quantitative flux control analysis of these data gave a control coefficient for SBPase on photosynthetic carbon assimilation in the mature leaves of between 0.35 and 0.50. These results are consistent with previous data showing that SBPase shares control of carbon assimilation in source leaves with Rubisco and plastid aldolase (Stitt et al., 1991; Stitt and Schulze, 1994; Haake et al., 1998; Haake et al., 1999; Raines et al., 2000). In addition, the data presented here demonstrate that the control SBPase exerts on photosynthetic carbon assimilation changes significantly during development. This may be due to changes in the demand for photosynthate that occur in leaves as they make the transition from being net importers of carbon (sinks) to net exporters (sources).

Analysis of nonstructural carbohydrates in leaves of the antisense plants revealed that reductions in SBPase activity had differential effects on carbohydrate levels, dependent on the developmental stage of the leaf and the magnitude of decrease in SBPase activity. In the plants with the greatest reductions in SBPase activity the trend is for carbohydrate levels to decline; however, Suc levels were generally less sensitive to reductions in SBPase activity and it was only in the youngest leaf (16) that a significant decrease in Suc was evident. In contrast, starch levels declined in parallel with the rate of photosynthetic carbon assimilation rates in the mature and young expanding leaves. In the young leaves of these plants photosynthetic carbon assimilation rates were reduced; however, the decrease in starch levels was greater than would have been expected given the reduction in photosynthetic capacity in these leaves. It is interesting that a linear relationship was observed between starch and Suc levels in these young leaves and the rate of photosynthetic carbon assimilation in the mature source leaves of these plants. This suggested that the source capacity of the mature leaves was limited and that this was directly affecting the availability of carbohydrate for the young leaves. These results are in keeping with the proposal that source metabolism dominates the control of carbon flux from source to sink (Sweetlove et al., 1998). Further evidence demonstrating the dynamic relationship between source strength and sink activity comes from our analysis of plants with small reductions in SBPase activity. The pattern of carbohydrate accumulation in this group of antisense plants was significantly different from plants with more severe reductions. Suc and starch levels in the young leaves unexpectedly were higher than in the wild-type plants, whereas in the mature leaves they were close to wild-type levels, a situation that might be associated more with sink limitation. In addition, there appeared to be no correlation between carbohydrate levels in the young leaves of these plants and the photosynthetic capacity of the mature source leaves. These results suggested that small changes in the photosynthetic capacity of the source leaves was being compensated for by reduced rates of leaf and shoot growth, leading temporarily to sink limitation. Growth analysis of a separate cohort of SBPase antisense plants would support this suggestion because significant reductions in leaf area and plant height were observed in response to small reductions in SBPase activity (Bryant, 1999). Further evidence for this argument comes from the increased rates of carbon fixation in the young sink leaves of this group of SBPase antisense plants. It is interesting that this was also observed in the young sink leaves of transgenic plants overexpressing invertase, where source limitation due to futile cycling of Suc led to a significant reduction in sink growth (von Schaewen et al., 1990; Sonnewald and Willmitzer, 1992).

In conclusion, the data presented here suggest that source strength contributes significantly to the development of maximal photosynthetic carbon assimilation during leaf expansion and maturation. Our results also indicate that plants can adjust their sink/source balance, most likely through reductions in growth rates, to compensate for changes in carbon availability. It is interesting that analysis of antisense Rubisco plants suggested that source strength was an important factor in determining shoot development and leaf longevity possibly as a strategy to maintain the photosynthetic capacity of the whole plant (Tsai et al., 1997). These results demonstrate the importance of primary carbon metabolism in the regulation/modulation of growth and development, and transgenic plants, such as the SBPase and Rubisco antisense plants, will continue to be powerful tools to investigate how changes in carbon availability affects whole-plant processes.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Plant Material and Growth Conditions

Wild-type and T1 transgenic (Nicotiana tabacum L. cv Samsun) seeds were germinated on sterile Murashige and Skoog media supplemented with 3% (w/v) Suc (Murashige and Skoog, 1962). For transgenic seedlings the media also contained kanamycin (100 µg mL¹). Three-week-old seedlings were transferred to soil (Levington F2, Fisons, Ipswich, UK) and grown in the greenhouse with a 14-h photoperiod at 25°C light/18°C dark. The plants were illuminated with natural light supplemented with high-pressure sodium lamps that gave 700 to 1200 µmol m² s¹ from pot level to the top of the plant. Plants were watered daily with Hoagland solution (Hoagland and Arnon, 1950) and harvested when they had formed 23 leaves. Alternate leaves were sampled, from leaf 16 (young expanding leaf) to leaf 8 (fully expanded mature leaf). Plants were numbered Tsa1 to indicate the transformation using a construct containing an SBPase tobacco partial cDNA, cloned in the reverse orientation between the cauliflower mosaic virus 35S promoter and the nos termination sequence (Harrison et al., 1998). Following this is the number of the parent line and the last number indicates the individual progeny plant. All experiments described here used T1 progeny.

Photosynthesis Measurements

Rates of CO₂ uptake were measured using a portable open gas exchange system (CIRAS-1, PP-Systems, Hitchin, UK), incorporating an infrared CO₂ and water analyzer that was calibrated against a known CO₂ standard (Linde Gas Ltd., UK). The steady-state rate of CO₂ uptake was determined under saturating light (1,000 µmol m² s¹) and A_sat. Net photosynthesis per unit leaf area and intercellular CO₂ concentration were determined using the equations of von Caemmerer and Farquhar (1981). Photon flux density was measured with a quantum sensor (Skye Instruments Ltd., Wales, UK).

Chlorophyll Analysis

Discs were removed from the leaves used for photosynthesis measurements and immediately frozen in liquid N₂. Pigments were extracted in 80% (v/v) acetone and the chlorophyll concentration was determined spectrophotometrically according to Hill et al. (1985).

SBPase Activity and Western Blotting

In parallel with sampling for gas exchange, leaf discs were removed for protein extraction and immediately frozen in liquid N₂. SBPase activity was determined as described previously (Harrison et al., 1998). Amounts of SBPase, FBPase, PRKase, and Rubsico protein were determined by separation on 12% (w/v) SDS-PAGE followed by western blotting. Polyclonal antibodies raised against SBPase and FBPase were a gift from Tristan Dyer (Department of Plant Sciences, University of Cambridge, UK), PRKase a gift from Bob Buchanan (University of California, Berkeley), and Rubsico a gift from Martin Parry (IACR Rothamsted, Harpenden, UK). Proteins were detected using horseradish peroxidase-conjugated second antibody and an enhanced chemiluminescence kit (Amersham International Public Limited Company; Harrison et al., 1998).

Carbohydrate Analysis

Leaf discs were from leaves used for photosynthetic measurements and frozen immediately in liquid N₂. Carbohydrates were extracted from the leaf discs in 80% (v/v) ethanol for 30 min at 80°C, followed by two washes with 80% (v/v) ethanol. Glc, Fru, and Suc levels were determined using an enzyme-based protocol (Stitt et al., 1989). Starch was measured in the ethanol-insoluble pellet according to Stitt et al. (1978) with the exception that instead of autoclaving, the samples were boiled for 1 h.

Metabolic Control Analysis

The flux control coefficient of SBPase on photosynthetic carbon fixation was determined from the gradient of a plot of the natural logarithm of CO₂ assimilation against the natural logarithm of SBPase activity using the data from the experiment in Figure 2. These data were fitted to the equation for a straight line, the slope of which corresponded to the flux control coefficient (Fell, 1997).