The Plant Journal (2002) 32, 685–699 An Arabidopsis thaliana knock-out mutant of the chloroplast triose phosphate/phosphate translocator is severely compromised only when starch synthesis, but not starch mobilisation is abolished Anja Schneider1,
, Rainer E. Häusler1, Üner Kolukisaoglu2,y, Reinhard Kunze1, Eric van der Graaff1, Rainer Schwacke1, Elisabetta Catoni3, Marcelo Desimone2 and Ulf-Ingo Flügge1 1 Botanisches Institut der Universität zu Köln, Gyrhofstrasse 15, D-50931 Köln, Germany 2 Max-Delbrück-Laboratorium in der Max-Planck-Gesellschaft, Carl-von-Linné-Weg 10, D-50829 Köln, Germany, and 3 Plant Physiology, Zentrum für Molekularbiologie der Pflanzen (ZMBP), Universität Tübingen, Auf der Morgenstelle 1, D-72076 Tübingen, Germany, Received 25 July 2002; revised ?? XXX 2002; accepted 8 August 2002.
For correspondence (fax þ49 221 4705039; e-mail anja.schneider@uni-koeln.de). yPresent adress: Institut für Molekulare Physiologie und Biotechnologie, Universität Rostock, Albert-Einstein-Strasse 3, D-18059 Rostock, Germany. Summary The Arabidopsis thaliana tpt-1 mutant which is defective in the chloroplast triose phosphate/phosphate translocator (TPT) was isolated by reverse genetics. It contains a T-DNA insertion 24 bp upstream of the start ATG of the TPT gene. The mutant lacks TPT transcripts and triose phosphate (TP)-specific transport activities are reduced to below 5% of the wild type. Analyses of diurnal variations in the contents of starch, soluble sugars and phosphorylated intermediates combined with 14CO2 labelling studies showed, that the lack of TP export for cytosolic sucrose biosynthesis was almost fully compensated by both continuous accelerated starch turnover and export of neutral sugars from the stroma throughout the day. The utilisation of glucose 6-phosphate (generated from exported glucose) rather than TP for sucrose biosynthesis in the light bypasses the key regulatory step catalysed by cytosolic fructose 1,6-bisphosphatase. Despite its regulatory role in the feed-forward control of sucrose biosynthesis, variations in the fructose 2,6-bisphosphate content upon illumination were similar in the mutant and the wild type. Crosses of tpt-1 with mutants unable to mobilise starch (sex1) or to synthesise starch (adg1-1) revealed that growth and photosynthesis of the double mutants was severely impaired only when starch biosynthesis, but not its mobilisation, was affected. For tpt-1/sex1 combining a lack in the TPT with a deficiency in starch mobilisation, an additional compensatory mechanism emerged, i.e. the formation and (most likely) fast turnover of high molecular weight polysaccharides. Steady-state RNA levels and transport activities of other phosphate translocators capable of transporting TP remained unaffected in the mutants. Keywords: reverse genetics, carbohydrate metabolism, photosynthesis, fructose 2, 6-bisphosphate, adg1, sex1. Introduction The triose phosphate/phosphate translocator (TPT) of the inner envelope membrane of chloroplasts represents the major interface for the distribution of photoassimilates between the chloroplast and the cytosol. In the light, triose phosphates (TP) are exported from the chloroplast stroma in strict counter exchange with inorganic phosphate (Pi), generated during sucrose biosynthesis in the cytosol (Fliege et al., 1978; Flügge, 1999; Flügge et al., 1989). The ß 2002 Blackwell Publishing Ltd Pi released in the cytosol is required for the synthesis and re-plenishment of ATP in the stroma. If sucrose biosynthesis slows down during the day, the limitation of Pi import redirects photosynthetic carbon flow into starch biosynthesis (Stitt et al., 1983). Hence, a decline in the TPT transport activity would favour starch biosynthesis. This view was reinforced by the analysis of potato plants with antisense mRNA re-pressed TPT (Heineke et al., 1994; Riesmeier et al., 685 686 Anja Schneider et al. 1993). These transgenic potato plants accumulated four times the amount of starch at the end of the light period compared to wild-type plants with only a 30% reduction in TP transport capacity, indicating a strong control by the TPT on the partitioning of carbon between sucrose and starch. These data were recently used for calculating the control TPT exerts on photosynthetic CO2 assimilation (Poolman et al., 2000). The transgenic potato plants lacked a clear phenotype (despite a slight growth retardation in axenic culture) mainly because they were capable of mobilising starch at a higher rate during the dark period and, thereby, compensating for the limitation on TP export during the day (Heineke et al., 1994; Riesmeier et al., 1993). Only when TP export and starch biosynthesis were inhibited simultaneously by antisense mRNA repression of ADP-glucose pyrophosphorylase (AGPase) and the TPT, potato plants exhibited stunted growth, a down regulation of photosynthetic enzymes and eventually chlorotic lesions on the leaves (Hattenbach et al., 1997). More recently, transgenic tobacco plants with an antisense repression of the TPT, which resulted in a decrease in TPT transport activity down to 30% of the wild-type levels, were analysed (Häusler et al., 1998, 2000b,c). Surprisingly, these transgenic lines were neither affected in growth or phenotypical appearance, nor did they show a large accumulation of starch at the end of the light period. The large decrease in TPT transport capacity could be compensated for by an increased rate of starch mobilisation commencing in the light (Häusler et al., 1998, 2000c). Increased rates of amylolytic starch degradation were accompanied by higher transport capacities for glucose across the envelope, a transient increase in the activity of chloroplastic a-amylase (in the middle of the photo period) and a higher activity of hexokinase. Thus, the synthesis of starch and its simultaneous breakdown results in an increased export of starch degradation products from the chloroplast. It was proposed that glucose 6-phosphate (Glc6P), generated from exported glucose, rather than TP acts as precursor for sucrose biosynthesis in the antisense TPT tobacco plants, thereby bypassing the reaction catalysed by cytosolic fructose 1,6-bisphosphatase (FBPase), which is regarded as a major feed-forward control step for sucrose biosynthesis in the light. Moreover, a detailed control analysis using tobacco transformants with decreased, as well as increased TPT activities, revealed that control by the TPT on sucrose biosynthesis, photosynthetic CO2 assimilation and electron transport, could be dissected from the contribution of compensatory effects of starch turnover (Häusler et al., 2000b). From control plots, it was extrapolated that the complete absence of the TPT would slow down the rate of sucrose biosynthesis to zero (without compensatory mechanisms) or only to 70% of the wild-type level (assuming compensatory starch turnover) and that the CO2-satu- rated rate of CO2 assimilation would decline to 40–50%. CO2 assimilation would, hence, adapt to the maximum capacity for starch turnover in the tobacco system. Since no TPT knock-out mutants of tobacco are available, this hypothesis could not be tested yet. In contrast, the Arabidopsis thaliana system offers the possibility to search directly for knock-out mutants by reversed genetics and, in addition, to challenge the system by creating double- or multiple-mutants affected in carbohydrate metabolism. In A. thaliana, as in potato or tobacco, the amylolytic rather than the phosphorolytic pathway of starch mobilisation predominates in leaves (Lin et al., 1988a,b;c). This view was supported by investigations on high-starch mutants lacking a chloroplast a-amylase, which is supposed to initiate nocturnal starch degradation (Zeeman et al., 1998a). The regulation of starch breakdown is by far less understood than the regulation of its synthesis. The mobilisation of starch in potato plants has been shown to correlate with the abundance of the R1 protein bound to starch granules (Lorberth et al., 1998). In A. thaliana, the analysis of a starchexcess mutant (sex1) revealed that the mutated gene is orthologous to the potato R1 gene (Yu et al., 2001). It has been suggested that the SEX1 (R1) protein regulates starch mobilisation by controlling the phosphate content of starch (Lorberth et al., 1998). Recently, the catalytic function of the R1 protein has been unravelled. The catalytic domain of the R1 protein acts as a starch:water dikinase introducing the b-phosphate moiety of ATP into the C3- or C6-position of glucose in the amylopectin backbone of starch (Ritte et al., 2002), thus supporting the proposed mechanism. Moreover, starch isolated from sex1 mutant alleles completely deprived of the R1 protein lacks any phosphorylation (Yu et al., 2001). Here, we report on the identification of the tpt-1 mutant of A. thaliana and its compensatory changes of carbohydrate metabolism and photosynthetic performance. Moreover, the crucial role of starch turnover was analysed in double mutants, in which the absence of the TPT in tpt-1 was combined with a deficiency in starch mobilisation (sex11) or biosynthesis (adg1-1) by genetic crosses. Data obtained with these double mutants will allow deeper insights in the compensatory flexibility of plant carbohydrate metabolism. Results Isolation and characterisation of the tpt-1 mutant A reverse genetic approach was applied to screen for plants with a knock-out of the TPT gene (At5g46110) in a population of T-DNA insertional mutants of A. thaliana (Feldmann, ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 Transport activities of tpt-1 mutant in Arabidopsis thaliana 687 TPT-specific transport rates are close to the detection limit Figure 1. Characterisation of the tpt-1 mutant. (a) Exon/intron structure of the TPT gene with the T-DNA insertion located 24 bp upstream of the ATG. Primers used for the reverse genetic screen were TPT-R and LB, Primers used for genotyping were TPT-F and TPT-R. (b) RNA gel blot analysis of wild type and mutant and plants using RNA of plants, grown under controlled environment conditions. As a loading control, the ethidium bromide stained gel is shown. (c) RT-PCR analysis of the same plants, using TPT-RT-F and TPT-RT-R primers and actin-specific primers as control. (d) Relative transport rates of tpt-1 (dark grey) and the wild type (light grey) are given as a percentage of the activity measured for reconstituted wildtype proteoliposomes preloaded with 3-PGA. The 100% exchange activity was 65 nmol mg Chl1 min1. Mean values from three to five different experiments, whereby in all cases the individual values deviated less than 4% from the mean value. 1991). A single line containing a T-DNA insertion 24 bp upstream of the start ATG (tpt-1) was identified (Figure 1a). The T-DNA insertion was flanked by 14 bp of foreign DNA at the site of integration. The inserted T-DNA represents a rearrangement of at least two T-DNA copies as it contains left borders on both sites (Figure 1a). The structure of the TDNA insertion was verified by Southern blot analysis using a left border-specific probe (data not shown). To test the effects of the insertion on the expression of TPT, RNA gel blot analysis and transport measurements (see below) were conducted. Steady-state contents of TPT-specific mRNA was below the detection limits on RNA gel blots in the tpt-1 mutant (Figure 1b). However, RT-PCR revealed a residual TPT expression in the tpt-1 mutant (Figure 1c). Most likely, the T-DNA insertion just upstream of the gene causes the observed residual expression. ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 More conclusive than mRNA expression studies are comparisons of specific transport activities between mutant and wild-type plants, in particular, as other inner envelope phosphate translocators, such as the glucose 6-phosphate/phosphate translocator (GPT) and the xylulose 5phosphate/phosphate translocator (XPT) share TP as a common substrate. The XPT has recently been shown to be expressed in the leaf mesophyll and to exhibit a low Km for TP (Eicks et al., 2002). Furthermore, transcripts of the GPT1 gene are also present in leaf mesophyll cells (unpublished observations). As shown in Figure 1(d), both 3-phosphoglycerate (3-PGA)- and TP-specific transport was reduced by about 95% in tpt-1, while Glc6P-specific transport activities were almost absent in both the wild type and the mutant. Thus, compensatory changes in the activity of other phosphate translocators (PT) are not likely to occur in tpt-1. This is consistent with the observation that the expression of other PT genes remained unaffected in the mutant (see below, Figure 10). It is likely that the residual transport rates measured in the tpt-1 mutant are due to the remaining TPT-activity, which most probably result from residual TPT expression (Figure 1c). The lack in TP transport in tpt-1 is completely compensated by an accelerated turnover of transitory starch The A. thaliana tpt-1 mutant shares many common physiological and biochemical features compared to the reported TPT antisense lines of tobacco and potato (see Introduction for details). However, this is the first report on a mutant surviving an almost complete reduction in TPT transport activity without major effects on growth, biomass or seed production. Surprisingly, in the tpt-1 mutant starch contents were persistently elevated by a factor of 2 during the course of a day and did not show any transient increase in starch contents (Figure 2a) as was observed for the potato or tobacco system. Moreover, there were only small differences in the rate of starch accumulation, estimated from the slope of starch contents versus time. Furthermore, the decline in starch contents during the dark period remained unaffected in the mutant compared to the wild type. Sucrose contents remained low in the mutant and even slightly declined during the light period (Figure 2b). This decline in sucrose contents suggests higher rates of export from the leaf to the sinks compared to its biosynthesis in the leaf. Contents of glucose (Glc) and fructose (Fru) were lowered in tpt-1 and lacked pronounced variations during the day. In particular, the increase in Glc content during the early light period in the wild type was absent in tpt-1 (Figure 2c,d). Contents of maltose (deriving from the 688 Anja Schneider et al. Figure 3. Pulse-chase 14C-labelling experiments. Time course of changes in the percentage 14C-labelling in starch and the soluble fraction in leaves of tpt-1 (*) and wild-type (*) plants. Leaves were labelled with 14CO2 30 min after the beginning of the photoperiod. The average cpm g FW1 declined from 837 009 to 451 550 in the wild type and from 651 833 to 388 021 in the tpt-1 mutant, respectively. Mean values of three different experiments  SD. Figure 2. Starch and soluble sugar content. (a) Starch, (b) sucrose, (c) glucose and (d) fructose contents were determined from leaves of tpt-1 mutant (*) and wild-type (*) plants. Black bars indicate the time in the dark. Mean values of three different experiments  SD. action of chloroplastic b-amylase) as a potential starch degradation product, which could be exported from the chloroplast were equally low in the wild type (52  6 nmol g FW1) and the mutant (64  7 nmol g FW1) and increased to a similar level at the beginning of the dark period (data not shown). Further information on carbon fluxes were obtained from pulse-chase experiments with 14CO2. The mutant accumulated consistently more 14C in the starch fraction than in the soluble fraction and was capable of mobilising starch faster than the wild type (Figure 3), irrespective of the time of the day, the experiment was conducted (not shown). This suggests that tpt-1 is capable of mobilising starch continuously in the light, thereby providing precursors for sucrose biosynthesis. Moreover, the relative proportion of labelled starch remained constant over the course of the day (Table 1). A substantial labelling of the neutral fraction (mainly sucrose) was found in the tpt-1 mutant, despite the lack of sucrose accumulation in intact plants (Figure 2). Apart from a slight increase in the labelling of the anionic fraction (i.e. phosphorylated intermediates and organic acids) in the mutant from 4.7% at the beginning of the light period to 7.4% after 8 h in the light there were no pronounced changes in the cationic fraction (mainly amino acids). Contents of metabolic intermediates are differentially affected in the tpt-1 mutant compared to the wild type Changes in relevant metabolic intermediates were analysed to elucidate the consequences of an altered allocation of photoassimilates on the contents of metabolites, which undergo high fluxes. As shown in Figure 4, there was a clear increase in both 3-PGA and TP in the mutant compared to the wild type (Figure 4g,h). The increase in TP contents most likely reflects the restricted export via the TPT and suggests an accumulation within the chloroplast. Likewise, the 3-PGA content was increased indicating an adaptation of the metabolic status to increased rates of starch synthesis (i.e. a relief of Pi inhibition at the site of AGPase by an increase in the 3-PGA/Pi ratio). The pattern of diurnal changes of TP and fructose 1,6-bisphosphate (Fru1,6P2) ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 Transport activities of tpt-1 mutant in Arabidopsis thaliana Table 1 Incorporation of phase 14 CO2 into starch and soluble components following a 20-min pulse with approximately 5% 14 689 CO2 in the gas 14 CO2 incorporated (%) Plant-line Starch Soluble fraction Neutral fraction Anionic fraction Cationic fraction tpt-1 0h 4h 8h 51.0  1.1 50.1  1.5 56.0  2.3 49.0  1.1 49.9  1.5 44  2.3 30.1  2.1 28.4  1.8 29.1  1.8 4.7  0.5 6.6  0.9 7.4  1.0 14.2  1.7 14.9  2.4 7.5  1.6 Wild type 0h 4h 8h 39.2  2.3 40.5  0.5 38.9  1.7 60.8  2.3 59.5  0.5 61.1  1.7 38.7  2.9 42.6  3.1 41.4  1.6 5.2  0.4 3.8  0.7 3.6  0.8 16.9  2.2 13.1  0.6 16.1  2.3 The sum of 14CO2 incorporation into the starch and soluble fraction was set to 100%. The soluble fraction was separated into neutral (sugars), anionic (organic acids, phosphorylated intermediates) and cationic (predominantly amino acids) components. Plant leaves were taken at the onset of photoperiod (0 h), in the middle of the light period (4 h) and towards the end of the light period (8 h). were very similar in the mutant. As TP and Fru1,6P2 are in equilibrium via the reaction catalysed by aldolase, this observation suggests that both metabolites are present in the same cellular compartment. Hexose monophosphates and Fru1,6P2 showed only little variation in the wild type. In contrast, in tpt-1 the contents of both hexose monophosphates were elevated and increased towards the end of the light period. Contents of glycolytic intermediates such as phosphoenolpyruvate (PEP) and pyruvate showed only little variation in the mutant compared to the wild type. The moderate changes in metabolic intermediates suggest that metabolism can be perfectly adapted to the re-direction of carbon fluxes into sucrose via accelerated starch turnover in tpt-1. Diurnal changes of fructose 2,6-bisphosphate content is not severely altered in tpt-1 Figure 4. Determination of metabolic intermediates in wild-type and tpt-1 plants during a light/dark cycle. Variation in the content of metabolic intermediates in leaves of tpt-1 (*) and wild-type (*) plants. Black bars indicate the time in the dark. Mean values of three different plants  SD. ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 Fructose 2,6-bisphosphate (Fru2,6P2) is an effective regulator controlling the partitioning between sucrose and starch biosynthesis in leaves (Larondelle et al., 1986; Stitt et al., 1984). It strongly inhibits cytosolic FBPase, which converts Fru1,6P2 derived from TPs exported from the chloroplast to Fru6P as starting point for sucrose biosynthesis. The rapid decline in Fru2,6P2 levels upon illumination is proposed to act as a feed-forward regulator of sucrose biosynthesis as it gives way to TP utilisation in the cytosol (Stitt, 1990). As cytosolic FBPase is supposed to be by passed in the mutant, variations in the contents of Fru2,6P2 were analysed during the course of the day, which had not been addressed in previous investigations on antisense TPT plants. Interestingly, the initial drop of Fru2,6P2 contents upon illumination is identical in the mutant and the wild type (Figure 5). For the mutant, this drop is unlikely to be correlated with an increase in cytosolic TP because the export of TP in the tpt-1 mutant can be regarded to be negligibly low. Thus, despite the fact that cytosolic FBPase is bypassed in the mutant, the initial diurnal pattern of Fru2,6P2 content mimics regulatory necessities, which are not required in the mutant. 690 Anja Schneider et al. Figure 5. Diurnal changes of Fru2,6P2 levels the in wild-type and tpt-1 plants. Fru2,6-P2 level in leaves of wild-type (*) and tpt-1 (*) plants. The Fru2,6P2 levels dropped after 1 h in the light and recovered during the rest of the light period. Black bars indicate the time in the dark. Mean values of three different plants  SD. CO2 assimilation is altered only at elevated CO2 concentration and high light There were only marginal differences in the Ci dependency of CO2 assimilation between the wild type and tpt-1 (Figure 6a). The initial slope of the A/Ci curves were similar, indicating that Rubisco activities (and/or activation states) are not affected in the mutant compared to the wild type. The A/Ci curves of the mutant and the wild-type diverged at Ci values of above 500 ml l1. In the mutant, CO2 assimilation rates remained unchanged at about 8 mmol m2 sec1, whereas the shape of the A/Ci dependency allows to extrapolate that CO2 saturated rates of CO2 assimilation would approach 12–13 mmol m2 sec1 in the wild type at CO2 concentration above 1500 ml l1 in the external gas mixture. The attainment of a higher Ci in the wild type was restricted by stomatal closure as a response to high CO2. This stomatal response was less pronounced in the mutant. In ambient air, light curves of CO2 assimilation were indistinguishable between the two genotypes, whereas in elevated CO2, light saturated rates of CO2 assimilation increased from 6 mmol m2 sec1 in tpt-1 to 8 mmol m2 sec1 in the wild type (Figure 6b,c). Generation of double mutants affected in TP transport and starch metabolism The data presented in this report suggest that in the tpt-1 mutant the major flux of carbon export from the chloroplast occurs via starch biosynthesis and simultaneous breakdown. To test this hypothesis, double mutants with a block in TP-export combined with a deficiency to synthesise or mobilise starch were generated. The tpt-1 mutant was crossed with the starch-free mutant agd1-1 lacking ADPglucose pyrophosphorylase activity (Lin et al., 1988a). Furthermore, tpt-1 was crossed with sex1-1, a starch excess Figure 6. Gas exchange characteristics of individual leaves. The dependencies of the CO2 assimilation rates from the intercellular CO2 concentration (a) comprise data from five individual measurements with wild-type (*) and tpt-1 plants (*) at a PFD of 505 mmol m2 sec1. Light dependencies of CO2 assimilation were measured for wild-type (b) and tpt-1 (c) plants either in air (&) or a gas mixture containing 1490 ppm CO2 and 2% O2 balanced with N2 (&) at leaf temperatures between 22 and 258C. Mean values of five individual determinations  SD. Curves were fitted to the data points using curve functions implemented in the SIGMA Plot program (version 5.0, SPSS, Germany). mutant (Caspar et al., 1991) with a single nucleotide substitution in the R1 gene, causing a Gly-1628 to Glu1628 transition (Yu et al., 2001). The homozygous double mutants sex1-1/tpt-1 were identified by their starch excess phenotype and by a PCR-based assay for tpt-1. After preselection of double mutants, the presence of the mutation in sex1-1 was confirmed by sequencing (data not shown). ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 Transport activities of tpt-1 mutant in Arabidopsis thaliana Growth and development of tpt-1 and the double mutants adg1-1/tpt-1 and sex1-1/tpt-1 The vegetative growth of tpt-1 as judged from the rosette diameter and changes in fresh weight (DFW) remained unaffected under controlled environmental conditions (Figures 7a,d; Table 2). At a light/dark cycle of 12 h/12 h, the onset of flowering was initiated on average two days earlier in the tpt-1 mutant than in wild-type plants. Both double mutants had similar points of time for the onset of flowering compared to their respective starch-free or starchexcess parental lines (not shown). In contrast, the double mutant adg1-1/tpt-1 exhibited a severe growth retardation. The rosette diameter declined by 70% and DFW (between days 35 and 50 after germination) was only 10% compared to the adg1-1 parental line (Figures 7c,f; Table 2), indicating that a limitation on TP export combined with the deficiency to synthesise starch is fatal for plant development (see also Hattenbach et al., 1997). However, it is remarkable that this double mutant is still viable. 691 Table 2 Fresh weight of whole plants was determinate 35 days after germination (A) and 50 days after germination (B) grown under 12 h light/12 h dark cycle Fresh weight (g) Plant-line A B DFW (g) adg1-1 sex1-1 tpt-1 Ws-2 Col-0 adg1-1/tpt-1 sex1-1/tpt-1 F2-WT individuals 0.158  0.06 0.114  0.02 0.235  0.02 0.244  0.03 0.376  0.05 0.015  0.002 0.076  0.004 0.205  0.07 0.617  0.09 0.536  0.02 0.739  0.08 0.730  0.06 0.779  0.09 0.054  0.01 0.296  0.04 0.808  0.2 0.459 0.422 0.504 0.486 0.403 0.039 0.220 0.603 Mean values from 6 to 8 individuals  SD. Interestingly, the double mutant sex1-1/tpt-1 showed a relatively minor retardation in growth (about 20% of rosette diameter and 50% of DFW) with respect to sex1-1 (Figures 7b,e; Table 2). Apparently, the deficiency in starch mobilisation in the tpt-1 background has a much smaller impact on the development of the double mutant than would have been expected from the proposed compensatory effect of increased starch turnover (i.e. its mobilisation). Hence, effective secondary compensatory mechanisms must come into operation when starch mobilisation, but not its synthesis, is abolished. The analysis of growth and development is hampered by the fact, that adg1-1 and sex1-1 are in the Col-0 and tpt-1 is in the Ws-2 background. We analysed the FW of plants from segregating F2 populations of adg1-1/tpt-1 and sex1-1/tpt-1. The DFW of those plants which are WT at the ADG1, SEX1 and TPT loci was slightly increased compared to Col-0 and Ws-2 (Table 2). Thus, the retardation in growth segregated perfectly with the double mutation adg1-1/tpt-1 and sex1-1/ tpt-1. Photosynthetic parameters support the ‘growth phenotype’ of the double mutants Figure 7. Growth phenotypes of Ws-2, tpt-1, sex1-1, sex1-1/tpt-1, adg1-1, and adg1-1/tpt-1. Pictures were taken of 5 weeks old plant grown under 12-h light/12-h dark cycles. ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 The photosynthetic light reaction was quantified by determinations of modulated chlorophyll a fluorescence parameters in single and double mutants. As light dependencies of photosynthesis revealed, neither photosynthetic electron transport (ETR) nor photochemical (qP) or nonphotochemical (qN) fluorescence quenching were significantly altered in tpt-1 compared to the wild type (Figure 8a–c). Interestingly, in the double mutants, electron transport was decreased significantly with sex1-1/tpt-1 exhibiting an intermediate and adg1-1/tpt-1 a severe drop in ETR compared to the respective single mutants sex1-1 or adg1-1 (Figure 8d–i). The decline in ETR was reflected in an increased redox state of QA (i.e. 1  qP ¼ QAred) in both double 692 Anja Schneider et al. Figure 8. Determination of photosynthetic parameters. Dependency of photosynthetic electron transport (a,d,g), photochemical quenching, qP (b,e,h) and non-photochemical quenching, qN (c,f,i) on the photon flux density (PFD) in ambient air at leaf temperatures between 22 and 258C in Ws-2 (*), tpt-1 (*), sex1-1 (&), sex1-1/tpt-1 (&), adg1-1 (~) and adg1-1/tpt-1 (~). The data comprise data of light curves obtained from leaves of five individual plants per line  SD. The curves were fitted to the data points using curve functions implemented in the SIGMA Plot program (version 5.0, SPSS, Germany). mutants. Moreover, qN was dramatically increased only in adg1-1/tpt-1 and approached values close to the maximum of 1.0 with increasing PFDs. For the double mutant adg1-1/ tpt-1, the severe drop in ETR and the large increase in qN, particularly at PFDs exceeding those the plants had experienced during growth indicates large perturbations in photosynthetic energy transduction. This is also reflected by a sharp drop in the Fv/Fm ratio of dark adapted leaves of between 0.77 and 0.79 in the wild type, the single mutants and sex1-1/tpt-1 to 0.53 in adg1-1/tpt-1. A decrease in the Fv/Fm ratio indicates a decline in the intactness of PSII energy transduction. Furthermore, adg1-1/tpt-1 was severely photoinhibited after illumination with a PFD of about 1300 mmol m2 sec1. However, this photoinhibitory effect was completely reverted and plants recovered after 12–16 h of darkness. ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 Transport activities of tpt-1 mutant in Arabidopsis thaliana Is the lack of starch mobilisation in sex1-1/tpt-1 compensated by a higher turnover of non-starch polysaccharides? Surprisingly, the double mutant sex1-1/tpt-1 exhibited only a moderate growth phenotype when compared to adg1-1/ tpt-1. Preliminary analyses revealed that sex1 mutants contained increased contents of non-starch high molecular weight soluble polysaccharides, HMWP (A. Weber, personal communication). These non-starch polysaccharides can be composed of homoglycans or heteroglycans, and are localised in different compartments of the cell (Yang and Steup, 1990). In sex1-1, only the portion of glucose in nonstarch polysaccharides was increased (A. Weber, personal communication). In order to address this alternative way for a transient deposition of photoassimilates, the contents of starch and glucose derived from non-starch polysaccharides in tpt-1, Ws-2 (background of tpt-1), sex1-1, Col-0 (background of sex1-1), and sex1-1/tpt-1 were compared. As the adg1-1 mutant and the double mutant adg1-1/tpt-1 lacked starch (and HMWP) completely, these mutants are not displayed. As shown in Figure 9(a), there was no significant difference in the starch contents between the two ecotypes Ws-2 and Col-0. In both parental single mutant lines tpt-1 and sex1-1, starch contents were increased compared to the respective wild-type plants. The starch content in the double mutant sex1-1/tpt1 was even further increased compared to the respective single mutants. Figure 9(b) shows that there were no significant differences in HMWP contents between Ws-2, Col-0, and tpt-1. However, the portion of glucose in HMWP was increased 10-fold in sex1-1 and even 15-fold in sex1-1/tpt-1 compared to the respective wild types. Consequently, the ratio of HMWP derived glucose to starch was two- to three-fold higher in sex1-1 and sex1-1/tpt-1 compared to tpt-1 and the wild Figure 9. Determination of carbohydrate contents in wild-type and mutant plants. (a) Starch was extracted at three time points during a light/dark cycle. The bars represent the mean  SD obtained with three different plants per genotype. (b) High molecular weight polysaccharides (HMWP) were prepared from the same samples as in (a). (c) HMWP/starch-ratios were calculated from data shown in (a) and (b). ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 693 types (Figure 9c). These data suggest that the formation and (most likely) a fast turnover of HMWP may act as an additional compensating mechanism for photosynthate export once the degradation of starch granules is impaired. The expression of plastidic metabolite transporters remained unaffected in the single and double mutants As shown above, a restriction in TP transport triggers compensatory changes in carbohydrate metabolism. These changes may include modulations in the expression of chloroplast envelope metabolite transporters. For instance, transgenic tobacco plants with an antisense repression of the TPT exhibited a higher transport capacity for glucose determined with isolated chloroplasts (Häusler et al., 1998). Meanwhile, all relevant sequences for members of the PT family in Arabidopsis are available, including TPT (At5g46110), PPT (At5g33320 and At3g01550; Fischer et al., 1997; Streatfield et al., 1999), GPT (At5g54800 and At1g61800; Kammerer et al., 1998) and XPT (At5g17630; Eicks et al., 2002). Besides the TPT, both the GPTs and the XPT are capable of transporting TP. Hence, an increase in the expression of both genes could partially compensate for a lack in TPT. The expression of the PT genes as well as the plastidic glucose transporter gene pGlcT (At5g16150; Weber et al., 2000) was analysed in tpt-1, Ws-2, sex1-1, adg1-1 and the double mutants sex1-1/tpt-1, adg1-1/tpt-1 by RT-PCR using gene-specific primers. As shown in Figure 10, there were no significant changes in the transcript amount of other transporter genes apart from the strong decline in TPT mRNA in the mutants tpt-1, sex1-1/tpt-1 and adg1-1/ tpt-1. In addition, in sex1-1/tpt-1, the transport activities for 3PGA and TP were reduced to the same low levels as in tpt-1 whereas in sex1-1, the transport activities for 3-PGA and TP 694 Anja Schneider et al. Figure 10. Analyses of transcript levels of various plastidic transporter genes in wild-type and mutant plants using RT-PCR. RNAs from leaves were isolated in the middle of the light period. RT-PCR was done as described in Experimental procedures. PCR primers were placed, if possible, in close proximity to introns to monitor DNA contamination; genomic DNAs were used as controls. PPT1-specific primers spanned an intron, therefore no PCR product was obtained using genomic DNA. Actin-specific primers were used in control reactions to balance differences in first strand-synthesised cDNAs. resembled those of the wild type. Transport of Glc6P could not be detected in either of the mutants. Thus, compensatory mechanisms exerted by other phosphate transporters could not be detected, neither at the RNA level nor at the activity level. Discussion In the present study, an A. thaliana mutant with an almost complete loss of TP transport across the chloroplast envelope has been identified. The residual transport activity of about 5% is unlikely to account for the lack of any visible phenotype of tpt-1 plants grown under controlled environmental conditions. The physiological characteristics of tpt-1 resembled, in many respects to previous reports on transgenic potato and tobacco plants with an antisense repression of the TPT (Häusler et al., 1998, 2000b,c; Heineke et al., 1994). However, these transgenics contained still substantial remaining TPT transport activities. A complete loss of the TPT could only be extrapolated from control plots (Häusler et al., 2000b). As is shown in this study, the almost complete loss of TP export in A. thaliana can indeed be fully compensated by very large changes in the allocation of photosynthates predominantly by higher rates of starch turnover in the light (see Figures 2 and 3). As for the tobacco system, 3-PGA was increased substantially giving way to higher rates of starch biosynthesis through stimulation of plastidic AGPase and the rates of CO2 assimilation were only affected at conditions which promote high rates of photosynthesis (i.e. high light and CO2) suggesting that the system adjusts to maximum possible rates of starch biosynthesis and mobilisation. However, apart from this common overall mechanisms of compensating for a reduction in TP export certain aspects were unique to tpt-1 or have not been reported before: (i) the steady-state content of sucrose even slightly declined in tpt-1 in the light (Figure 2b) suggesting that the rate of sucrose exported from the leaf exceeds the rate of its biosynthesis. (ii) The lowered contents of Glc indicate an increased rate of Glc utilisation for sucrose biosynthesis, presumably via the reaction catalysed by the chloroplast outer envelope-bound hexokinase (Wiese et al., 1999), particularly during the early light period, when Glc contents in the wild type reached maximum values (Figure 2c). This view is supported by the observation that after feeding of detached leaves with 14CO2 in the light, there was a substantial 14C labelling in the neutral metabolite fraction, mainly consisting of sucrose. Interestingly, sucrose contents increased at the transition from light to dark, suggesting either a limitation on the export from the leaf or an increased rate of sucrose biosynthesis. (iii) Apart from 3-PGA (compare Häusler et al., 1998; Häusler et al., 2000c), TP contents increased more than two-fold in tpt-1 compared to the wild type, which most likely reflects the block of TP export from the stroma. (iv) There were only moderate changes in Fru1,6P2 contents during the diurnal cycle between the mutant and the wild type. Following dephosphorylation of Fru1,6P2 by stromal FBPase, the product Fru6P can be fed into starch biosynthesis by subsequent reactions. Interestingly, chloroplastic Fru1,6P2 contents have been reported to be increased in potato leaves with a combined antisense repression of the TPT and the AGPase, i.e. when starch synthesis is abolished (Hattenbach et al., 1997). The only moderate change in the Fru1,6P2 level in tpt-1 suggests that Fru1,6P2 can be efficiently fed into starch biosynthesis. (v) Diurnal variations in Glc6P may reflect changes in cytosolic contents. Interestingly, in the mutant, Glc6P content correlates directly with sucrose contents (Figures 2b and 4a). Both Glc6P and Fru6P can be regarded as direct precursors for the synthesis of sucrose phosphate via sucrose phosphate synthase (SPS). Moreover, in its phosphorylated state, SPS activity is sensitive to the cytosolic Glc6P/Pi ratio. From the data shown in ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 Transport activities of tpt-1 mutant in Arabidopsis thaliana Figures 2(b) and 4(a), it can be assumed that Glc6P relieves the feedback inhibition of SPS by Pi. It is conceivable that most of the sucrose generated during the day is exported andtheincreaseinGlc6Pcontentsattheendofthelightperiod allows significant accumulation of sucrose in the leaf. Is Fru2,6P2 required for the regulation of sucrose biosynthesis in tpt-1? A major regulatory step in sucrose biosynthesis, the cytosolic FBPase, can obviously be bypassed in tpt-1. It is therefore surprising that the diurnal changes of the Fru2,6P2, a key regulator of the cytosolic FBPase, and hence, sucrose biosynthesis were quite similar in tpt-1 and wild-type plants (Figure 5), suggesting that the basic regulatory features of the Fru2,6P2 system are not impaired by a lack of TP export. The content of Fru2,6P2 is controlled by the action of the bifunctional enzyme PFK2 and FBPase2, which in turn is regulated by the levels of Fru6P, Pi, 3-PGA and TP (Larondelle et al., 1986; Stitt et al., 1984). It has been shown recently that A. thaliana lines with reduced Fru2,6P2 contents due to an antisense repression of PFK2/FBPase2 contained higher levels of sugars (sucrose and hexoses) and lower levels of TP and hexose phosphate in the leaves during ongoing photosynthesis (Draborg et al., 2001). The latter suggests an increased utilisation of TP and hexose phosphate for sucrose biosynthesis. Because there is no precise information available on the subcellular distribution of metabolites affecting the Fru2,6P2 system in the tpt-1 mutant, a more detailed analysis is required which is beyond the scope of this paper. In some respects, the tpt-1 mutant resembles transgenic A. thaliana plants with an antisense repression of the cytosolic FBPase (antifbp lines). These transformants show a decline in the rate of sucrose synthesis, an accumulation of phosphorylated intermediates, Pi-limitation of photosynthesis and a stimulation of starch synthesis (Strand et al., 2000). The increase in the (most probably, stromal) content of 3-PGA as well as TPs in the tpt-1 mutant would be consistent with a decline in free Pi. In the anti-fbp lines, recycling of Pi from phosphorylated intermediates is accelerated by increasing the rate of starch synthesis. However, unlike in the tpt-1 mutant, the antisense repression of cytosolic FBPase results in a more than 50% inhibition of plant growth combined with lower leaf protein contents and lowered photosynthetic rates. Manipulation of starch biosynthesis and degradation in the tpt-1 background To further dissect starch metabolism in tpt-1, double mutants were generated, in which either starch biosynthesis (adg1-1/tpt-1) or mobilisation (sex1-1/tpt-1) was blocked. Both parental lines sex1-1 and adg1-1 showed ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 695 late-flowering phenotypes when grown under short day conditions. The onset of flowering time in the tpt-1 mutant was earlier compared to the wild type. However, this feature disappeared in both double mutants, indicating a link between higher starch mobilisation with the onset of flowering as it has been proposed earlier (Corbesier et al., 1998). The double mutant adg1-1/tpt-1 exhibits a severe growth retardation, which was expected if the fixed carbon can neither be exported by the TPT nor used for starch metabolism (see also Hattenbach et al., 1997). However, the double mutants were still viable and even produced seeds. This is most probably due to the residual expression of TPT gene but not to additional compensatory mechanisms such as the involvement of other transporters capable of transporting TP, in particular the GPT and the XPT. In contrast to adg1-1/tpt-1, the sex1-1/tpt-1 double mutant that is unable to mobilise starch, but in which the AGPase as the starting point for either starch biosynthesis and/or the biosynthesis of HMWP is unaffected, does not show a severe growth phenotype. This finding indicates that in the absence of TPT activity, the decline in the mobilisation of granular starch can be bypassed more efficiently than the complete loss of starch biosynthesis. Both the double mutant sex1-1/tpt-1 and the parental line sex1-1 show a high-starch phenotype and, in addition, possess 15- and 10fold higher levels of non-starch HMWP compared to the wild type. It is likely that (the major fraction of) this watersoluble polysaccharide is located within the chloroplast and that ADP-glucose is used as the immediate precursor, since no HWMP were detectable in the starch-free adg1-1 mutant. It is tempting to speculate that the highly branched watersoluble polysaccharide phytoglycogen represents a variety of these HMWP glucans. An A. thaliana mutant lacking a chloroplastic de-branching enzyme of the isoamylase type was shown to accumulate simultaneously starch and phytoglycogen in the same chloroplast (Zeeman et al., 1998b). It has been suggested that phytoglycogen is not an intermediate of amylopectin synthesis, but rather a separate soluble product produced in the stroma and that the accumulation of these glucans is prevented by the action of the de-branching enzyme. The double mutant sex1-1/tpt-1 can obviously compensate for the deficiency in carbon export from the chloroplasts in form of TP and the concomitant defect in starch mobilisation by directing the fixed carbon into the biosynthesis of HWMP glucans, thereby circumventing the regulation of starch breakdown by the R1 protein. HWMP glucans could be used as a carbon source for hexoses which can be exported from the chloroplast by the pGlcT (Weber et al., 2000). In this respect, the sex1-1/tpt-1 double mutant resembles the tpt-1 mutant except that different pools of polysaccharides serves as carbon sinks, starch in case of the tpt-1 mutant and HMWP in case of the sex1-1/tpt1 double mutant. It will be important to understand more 696 Anja Schneider et al. precisely how the syntheses of HMWP and starch are differentially regulated. An approach to address this issue would be to cross tpt-1 and sex1-1/tpt-1 with mutants that are defective in either the synthesis or the export of starch/ HMWP breakdown products and to analyse these crosses. The observed changes in growth and phenotypical appearances of the double mutants were perfectly underlined by alterations in the photosynthetic capacity. In particular, photosynthetic electron transport decreased intermediately in sex1-1/tpt-1, but severely in adg1-1/tpt-1 compared to the respective single mutants. It is, hence, likely that perturbations in photosynthesis form the basis for the observed growth retardations. For adg1-1/tpt-1, it is likely that the lack of TP export combined with the missing capacity to produce starch feeds back on the rate of photosynthesis. Probably electron transport can be maintained at low rates by the residual TPT activity, though mutants with a tpt null allele are awaited to clarify this assumption. Experimental procedures Plant material and growth conditions Seeds of Arabidopsis thaliana L. (Heynh.) [ecotypes Wassilewskija (Ws-2, N1601) and Columbia (Col-0, N1093), mutant lines adg1-1 (N3094; Lin et al., 1988a) and sex1-1 (N3093; Caspar et al., 1991) and a collection of 6500 T-DNA transformed lines from seed transformation (Forsthoefel et al., 1992) arranged in pool sizes of 100 (N3115 and N6500) and 20 (N3116 and N6400)] were provided by the Nottingham Arabidopsis Stock Centre (NASC, http://nasc.nott.ac.uk/home.html). For metabolite measurements and expression analysis plants were grown on soil under a light/dark cycle of 12 h/12 h, a day/night temperature of 218C/188C and at 40% humidity. The photon flux density (PFD) at plant level was 100 mmol m2 sec1. The leaf material was harvested from 4week-old plants. Plants used for DNA extraction, crosses and propagation were grown in a temperature controlled greenhouse under a light/dark cycle of 16 h/8 h. Isolation of tpt-1 and identification of adg1-1/tpt-1 and sex1-1/tpt-1 DNA prepared from the T-DNA insertion population was combined in six superpools and subjected to PCR using TPT-R and LB primer (see Figure 1a, for primer sequences see Table 3). PCR amplification conditions were 948C for 3 min; 39 cycles at 948C for 1 min, 588C for 1 min and 728C for 2 min, followed by incubation at 728C for 3 min. PCR products obtained were subcloned and sequenced. After identification of a positive PCR product, subsequent pools were analysed and an individual plant was isolated, named tpt-1. To analyse both gene regions flanking the T-DNA integration site in the tpt-1 mutant, PCRs were performed on genomic DNA of the mutant plant using TPT primers in combination with border primers, and the resulting PCR products were again subcloned and sequenced. The mutant line tpt-1 was crossed with adg1-1 and sex1-1 and the resulting F1 generations were allowed to selfpollinate. F2 plants of the crosses were screened for starch-free or starch-excess phenotypes by staining leaves with iodine after a 12-h light or 12-h dark period, respectively. The double mutant Table 3 Primers used for the identification and verification of the tpt-1 mutant, for the generation of probes and RT-PCR analysis Primer name Primer sequence RB LB LB-F LB-R TPT-F TPT-R SEX1-F SEX1-R TPT-RT-F TPT-RT-R PPT1-F PPT1-R PPT2-F PPT2-R XPT-F XPT-R GPT-F GPT-R pGlcT-F pGlcT-R Actin-F Actin-R 50 -TCCTTCAATCGTTGCGGTTCTGTCAGTTC-30 50 -GATGCACTCGAAATCAGCCAATTTTAGAC-30 50 -GGTGTAAACAAATTGACGCTTAGA-30 50 -CTTGCCTATTATGTGAAGGACAATC-30 50 -GTAACTTACGAGTAAACTGGCTAC-30 50 -AGCAGCCGCATTGAAGAATGGCTCAA-30 50 -GAACGAGAGAGCATACTTCAGC-30 50 -AGTCAGTGATCAGAGGATCTG-30 50 -TCCTCCTGCCATCATCGTTG-30 50 -TCTATGCTTTCTTTCCTTGCCG-30 50 -CATTGATGTCTCTCGTTCTGATGG-30 50 -GCGATTCCAGTTCCGAAAGC-30 50 -TCTCTACTTGCTGGTGTTTGCTTG-30 50 -GGATTTGGTTTGACTTGGACTCG-30 50 -TTTCCCGTGGCGATTTTCG-30 50 - GCATTCAGAGGTCTAACAGGATTCC-30 50 -GAAAGTCTGTGAGCGGGATGAAC-30 50 -TGCGGAAGATAATGATGGAGGAG-30 50 -CAGGCACTGCTGTTGCTTCATC-30 50 -CCAAGTAGACACTGCTGATTCCG-30 50 -GGGCAAGTCATCACGATTGG-30 50 -GAAGCAAGAATGGAACCACCG-30 LB and RB represent primers-specific for the T-DNA left and right borders, respectively. sex1-1/tpt-1 was also confirmed by sequencing the sex1-1 allele amplified using SEX1-specific primers. The cross adg1-1/tpt-1 yielded plants, which were homozygous for agd1-1 and heterozygous for tpt-1 in the F2 generation, because both genes are located on chromosome V. A plant homozygous for both adg1-1 and tpt-1 was selected in the F3 generation. RNA gel blot and Southern blot analysis For RNA gel blots, RNA was isolated as described by Logemann et al. (1987). DNA was isolated according to Liu et al. (1995). RNA gel blot analyses (15 mg) of total RNA, hybridised with a 1.3-kb EcoRI fragment of the AtTPT cDNA) and Southern blots (10 mg of DNA) were performed following standard protocols (Sambrook et al., 1989). DNA gel blots were performed to analyse the PCRs in reverse genetic screens, RT-PCR products and to determine the TDNA structure by using a left border-specific probe generated by PCR with LB-F and LB-R primer on superpool-DNA. For RT-PCR analysis total RNA was extracted from 100 mg of fresh tissue using the RNeasy Plant Kit (Qiuagen). Oligo(dt)-primed cDNA from 1 mg of total RNA was synthesised using the SuperScript Reverse Transcriptase system (Gibco/BRL). Primers used for amplification are listed in Table 3. Amplification conditions were as follows: 3 min at 958C; 20–26 cycles of 958C for 30 sec, 558C for 30 sec, and 728C for 30 sec, followed by incubation at 728C for 10 min with 22 cycles for TPT, 24 cycles for XPT, 26 cycles for PPT1, PPT2, GPT, pGlcT, and 20 cycles for actin. PCR products were analysed by DNA gel blotting and hybridisation to PCR-specific probes. Quantification of the signals was performed using a phosphoimager (Storm 860, Molecular Dynamics) and the program Image Quant for MacIntosh (version 1.2; Molecular Dynamics). ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 Transport activities of tpt-1 mutant in Arabidopsis thaliana Transport measurements Transport activities of wild type and the mutants were determined by reconstitution of leaf homogenates (200 mg of FW) into artificial membranes (Flügge and Weber, 1994). The liposomes were preloaded with 20 mM potassium gluconate as a control or with 20 mM 3-PGA, TP or Glc6P serving as counter exchange substrates and incubated with 32Pi (specific acticity 220 GBq mol1); final concentration, 0.3 mM. For the assessment of PT activities a time course of 32Pi uptake was measured. Initial transport rates were calculated from the radioactivity taken up within the first 40 sec expressed on a leaf FW basis. Metabolite determination Leaf contents of starch and soluble sugars were isolated and assayed according to the method described by Lin et al. (1988a). The contents of Glc, Fru and sucrose were determined enzymatically according to Stitt et al. (1989); maltose was determined according to Shirokane et al. (2000). High molecular polysaccharides were isolated according to Yang and Steup (1990) with slight modifications. A. thaliana rosettes (0.2 g) were ground to a fine powder and suspended with 0.5 ml ice-cold water. Water soluble polysaccharides were separated from insoluble material by centrifugation (20 000  g for 5 min). The sediment was used for the determination of starch contents (see above). The aqueous phase was de-proteinised twice with water-saturated phenol and once with chloroform and de-ionised with an anion- and cationexchange resin (AG 501-X8, 20–50 mesh, Bio-Rad). The watersoluble high molecular weight polysaccharides were precipitated with 50 mM KCl and 70% ethanol. The sediment was washed twice with 70% ethanol and re-suspended in 0.1 ml of 2N HCl. An aliquot was checked for glucose contaminations. In none of the experiments, any residual glucose content was observed. For quantitative determination of the polysaccharide fraction, the resuspended material was heated for 1 h at 958C, neutralised, and the glucose content was determined. Metabolite intermediates were determined enzymatically in neutralised perchloric acid extracts (Bergmeyer, 1974; Stitt et al., 1989) and using the Spectrofluor Plus in the fluorescence mode (TECAN, Austria; Häusler et al., 2000a). For the determination of Fru2,6P2 contents, leaf material (0.1 g) was ground in 0.25 ml of ice-cold 10 mM KOH according to (Draborg et al., 2001). The extract was centrifuged at 10 000  g for 1 min and the supernatant was used for the determination of Fru2,6P2 by an assay based on the activation of PFP from potato tubers (Van Schaftingen, 1984). The recovery of added Fru2,6P2 was 90%. Carbon partitioning The incorporation of 14CO2 into A. thaliana leaves was performed as described by Quick et al. (1989). Whole leaves were incubated in a sealed Perspex chamber in the presence of 1 M NaHCO3 solution (pH 9.0) enriched with NaH14CO3 (specific activity, 0.14 MBq mmol1) for 20 min at a PFD of approximately 200 mmol m2 sec1 followed by a 2.5-h chase in the dark. Metabolism in the leaves was quenched in hot 80% ethanol. Starch was separated from the soluble fraction as described above. The soluble fraction was separated into neutral, acidic and basic components by chromotography on Dowex anion (AG 1  8) and cation (AG 50 W  8) exchange resins (mesh size 200–400; Bio-Rad, Germany). The amount of 14C present in the different fractions was determined by liquid scintillation counting. ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 697 Gas-exchange measurements and Chl a fluorescence analysis Gas-exchange characteristics of A. thaliana leaves were measured with an LCA-4 open gas-analyser (Analytical Development, UK) in a custom made-leaf chamber at a leaf temperature of between 22 and 258C. Light was supplied via fibre optics by a Schott KL 1500 projector (Walz, Germany). The air humidity was kept at approximately 30%. The rates of CO2 assimilation (A) as well as the intercellular CO2 concentrations (Ci), were calculated as described (Von Caemmerer and Farquhar, 1981). Gas-exchange parameters of A. thaliana leaves were either measured in air supplied by an air cylinder or in a special gas mixture containing 1490 ml l1 CO2, 2% O2 balanced with N2. Modulated Chl a fluorescence emission from the upper surface of the leaf was measured with a pulse amplitude modulation fluorometer (PAM-2000, Walz, Germany; Schreiber et al., 1986). The ground fluorescence (F0) was measured by exposing leaves, which were dark adapted for at least 20 min to a weak modulated measuring beam. For the determination of the maximum fluorescence (Fm), a flash of saturation light (approximately 5000 mmol m2 sec1; duration 800 msec) was applied. The quantum efficiency of electron flux through photosystem II (FPSII) was assessed according to Genty et al. (1989) from the ratio of (Fm  Fs)/ Fm (Fs ¼ steady-state fluorescence). Rates of electron transport were estimated from FPSII Ia/2, where Ia denotes the portion of photons absorbed by the leaf assuming 0.83 as absorption coefficient. Acknowledgements We thank Darja Henseler for help in isolating the adg1-1/tpt-1 and sex1-1/tpt-1 mutants and Siegfried Werth for photographs. This work was supported by grants from the Deutsche Forschungsgemeinschaft and the Bundesministerium für Bildung und Forschung. Work done in the Max-Delbrück-Laboratorium was continously supported by a grant to U.I.F. from the Ministerium für Schule, Wissenschaft und Forschung des Landes NordrheinWestfalen. References Bergmeyer, H. (1974) Methoden der Enzymatischen Analyse, 3rd edn. Weinheim, Germany: Verlag Chemie. Caspar, T., Lin, T.-P., Kakefuda, G., Benbow, L., Preiss, J. and Somerville, C. (1991) Mutants of Arabidopsis with altered regulation of starch degradation. Plant Physiol. 95, 1181–1188. Corbesier, L., Lejeune, P. and Bernier, G. (1998) The role of carbohydrates in the induction of flowering in Arabidopsis thaliana: comparison between the wild type and a starchless mutant. Planta, 206, 131–137. Draborg, H., Villadsen, D. and Nielsen, T.H. (2001) Transgenic Arabidopsis plants with decreased activity of fructose-6-phosphate, 2-kinase/fructose-2,6-bisphosphatase have altered carbon partitioning. Plant Physiol. 126, 750–758. Eicks, M., Maurino, V., Knappe, S., Flügge, U.I. and Fischer, K. (2002) The plastidic pentose phosphate translocator represents a link between the cytosolic and the plastidic pentose phosphate pathways in plants. Plant Physiol. 128, 512–522. Feldmann, K.A. (1991) T-DNA insertion mutagenesis in Arabidopsis-mutational spectrum. Plant J. 1, 71–82. Fischer, K., Kammerer, B., Gutensohn, M., Arbinger, B., Weber, A., Häusler, R.E. and Flügge, U.I. (1997) A new class of plastidic 698 Anja Schneider et al. phosphate translocators: a putative link between primary and secondary metabolism by the phosphoenolpyruvate/phosphate antiporter. Plant Cell 9, 453–462. Fliege, R., Flügge, U.I., Werdan, K. and Heldt, H.W. (1978) Specific transport of inorganic phosphate, 3-phosphoglycerate and triosephosphates across the inner membrane of the envelope in spinach chloroplasts. Biochim. Biophys. Acta 502, 232–247. Flügge, U.I. (1999) Phosphate translocators in plastids. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50, 27–45. Flügge, U.I., Fischer, K., Gross, A., Sebald, W., Lottspeich, F. and Eckerskorn, C. (1989) The triose phosphate-3-phosphoglyceratephosphate translocator from spinach chloroplasts: nucleotide sequence of a full-length cDNA clone and import of the in vitro synthesized precursor protein into chloroplasts. EMBO J. 8, 39–46. Flügge, U.I. and Weber, A. (1994) A rapid method for measuring organelle-specific substrate transport in homogenates from plant tissues. Planta, 194, 181–185. Forsthoefel, N.R., Wu, Y., Schulz, B., Bennett, M.J. and Feldmann, K.A. (1992) T-DNA Insertion mutagenesis in Arabidopsis: prospects and perspectives. Aust. J. Plant Physiol. 19, 353–366. Genty, B., Briantais, J.M. and Baker, N.R. (1989) The relationship between the quantum yield of photosynthetic transport and quenching of chlorophyll fluorescence. Biochim. Biophys. Acta 990, 87–92. Hattenbach, A., Müller-Röber, B., Nast, G. and Heineke, D. (1997) Antisense repression of both ADP-glucose pyrophosphorylase and triose phosphate translocator modifies carbohydrate partitioning in potato leaves. Plant Physiol. 115, 471–475. Häusler, R.E., Fischer, K.L. and Flügge, U.I. (2000a) Determination of low abundant metabolites in plant extracts by NAD(P)H fluorescence with a microtiter plate reader. Anal. Biochem. 281, 1–8. Häusler, R.E., Schlieben, N.H. and Flügge, U.I. (2000b) Control of carbon partitioning and photosynthesis by the triose phosphate/ phosphate translocator in transgenic tobacco plants (Nicotiana tabacum). Part II. Assessment control. coefficients triose phosphate/phosphate translocator. Planta, 210, 383–390. Häusler, R.E., Schlieben, N.H., Nicolay, P., Fischer, K., Fischer, K.L. and Flügge, U.I. (2000c) Control of carbon partitioning and photosynthesis by the triose phosphate/phosphate translocator in transgenic tobacco plants (Nicotiana tabacum L.). Part I. Comparative physiological analysis of tobacco plants with antisense repression and overexpression of the triose phosphate/ phosphate translocator. Planta, 210, 371–382. Häusler, R.E., Schlieben, N.H., Schulz, B. and Flügge, U.I. (1998) Compensation of decreased triose phosphate/phosphate translocator activity by accelerated starch turnover and glucose transport in transgenic tobacco. Planta, 204, 366–376. Heineke, D., Kruse, A., Flügge, U.I., Frommer, W.B., Riesmeier, J.W., Willmitzer, L. and Heldt, H.W. (1994) Effect of antisense repression of the chloroplast triose-phosphate translocator on photosynthetic metabolism in transgenic potato plants. Planta, 193, 174–180. Kammerer, B., Fischer, K., Hilpert, B., Schubert, S., Gutensohn, M., Weber, A. and Flügge, U.I. (1998) Molecular characterization of a carbon transporter in plastids from heterotrophic tissues: the glucose 6-phosphate/phosphate antiporter. Plant Cell 10, 105– 117. Larondelle, Y., Mertens, E., Van Schaftingen, E. and Hers, H.G. (1986) Purification and properties of spinach leaf phosphofructokinase 2/fructose 2,6-bisphosphatase. Eur. J. Biochem. 161, 351–357. Lin, T.-P., Caspar, T., Somerville, C. and Preiss, J. (1988a) Isolation and characterization of a starchless mutant of Arabidopsis thaliana (L.) lacking ADP-glucose pyrophosphorylase activity. Plant Physiol. 86, 1131–1135. Lin, T.-P., Caspar, T., Somerville, C. and Preiss, J. (1988b) A starch deficient mutant of Arabidopsis thaliana with low ADP-glucose pyrophosphorylase activity lacks one of the two subunits of the enzyme. Plant Physiol. 88, 1175–1181. Lin, T.-P., Spilatro, S.R. and Preiss, J. (1988c) Subcellular localization and characterization of amylases in Arabidopsis leaf. Plant Physiol. 86, 251–259. Liu, Y.G., Mitsukawa, N., Oosumi, T. and Whittier, R.F. (1995) Efficient isolation and mapping of Arabidopsis thaliana T-DNA insert junctions by thermal asymmetric interlaced PCR. Plant J. 8, 457–463. Logemann, J., Schell, J. and Willmitzer, L. (1987) Improved method for the isolation of RNA from plant tissues. Anal. Biochem. 163, 16–20. Lorberth, R., Ritte, G., Willmitzer, L. and Kossmann, J. (1998) Inhibition of a starch-granule-bound protein leads to modified starch and repression of cold sweetening. Nature Biotechn. 16, 473–477. Poolman, M.G., Fell, D.A. and Thomas, S. (2000) Modelling photosynthesis and its control. J. Exp. Bot. 51, 319–328. Quick, P., Siegl, G., Neuhaus, E., Feil, R. and Stitt, M. (1989) Shortterm water stress leads to a stimulation of sucrose synthesis by activating sucrose-phosphate synthase. Planta, 117, 535– 546. Riesmeier, J.W., Flügge, U.I., Schulz, B., Heineke, D., Heldt, H.W., Willmitzer, L. and Frommer, W.B. (1993) Antisense repression of the chloroplast triose phosphate translocator affects carbon partitioning in transgenic potato plants. Proc. Natl Acad. Sci. USA 90, 6160–6164. Ritte, G., Lloyd, J., Eckermann, N., Rottmann, A., Kossmann, J. and Steup, M. (2002) The starch related R1 protein is an aglucan, water dikinase. Proc. Natl. Acad. Sci. USA 99, 7166–7171. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular cloning: A. Laboratory Manual, 2nd edn. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Schreiber, U., Schliwa, U. and Bilger, B. (1986) Continuous recording of photochemical and non-photochemical chlorophyll fluorescence quenching with a new type of modulation fluorometer. Photosynth. Res. 10, 51–62. Shirokane, Y., Ichikawa, K. and Suzuki, M. (2000) A novel enzymic determination of maltose. Carbohydr. Res. 329, 699–702. Stitt, M. (1990) Fructose-2,6-bisphosphate as a regulatory molecule in plants. Annu. Rev. Plant Physiol. Plant. Mol. Biol. 41, 153–185. Stitt, M., Cseke, C. and Buchanan, B.B. (1984) Regulation of fructose 2,6-bisphosphate concentration in spinach leaves. Eur. J. Biochem. 143, 89–93. Stitt, M., Gerhardt, R., Kurzel, B. and Heldt, H.W. (1983) A role of fructose 2,6-bisphosphate in the regulation of sucrose synthesis in spinach leaves. Plant Physiol. 72, 1139–1141. Stitt, M., lilley, R.M.C., Gerhardt, R. and Heldt, H.W. (1989) Determination of metabolite levels in specific cells and subcellular compartments of plant leaves. Meth Enzymol. 174, 518–522. Strand, A., Zrenner, R., Trevanion, S., Stitt, M., Gustafsson, P. and Gardestrom, P. (2000) Decreased expression of two key enzymes in the sucrose biosynthesis pathway, cytosolic fructose-1,6-bisphosphatase and sucrose phosphate synthase, has remarkably different consequences for photosynthetic carbon metabolism in transgenic Arabidopsis thaliana. Plant J. 23, 759–770. Streatfield, S.J., Weber, A., Kinsman, E.A., Häusler, R.E., Li, J., Post-Beittenmiller, D., Kaiser, W.M., Pyke, K.A., Flügge, U.I. and ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 Transport activities of tpt-1 mutant in Arabidopsis thaliana Chory, J. (1999) The phosphoenolpyruvate/phosphate translocator is required for phenolic metabolism, palisade cell development, and plastid-dependent nuclear gene expression. Plant Cell 11, 1609–1622. Van Schaftingen, E. (1984) D-fructose 2,6-bisphosphate. In Methods of Enzymatic Analysis (H.Bergmeyer, ed.). Weinheim, Germany: Verlag Chemie, pp. 335–341. Von Caemmerer, S. and Farquhar, G.D. (1981) Some relationships between the biochemistry of photosynthesis and the gas exchange of leaves. Planta, 153, 376–387. Weber, A., Servaites, J.C., Geiger, D.R., Kofler, H., Hille, D., Gröner, F., Hebbeker, U. and Flügge, U.I. (2000) Identification, purification, and molecular cloning of a putative plastidic glucose translocator. Plant Cell 12, 787–802. Wiese, A., Gröner, F., Sonnewald, U., Deppner, H., Lerchl, J., Hebbeker, U., Flügge, U.I. and Weber, A. (1999) Spinach ß Blackwell Publishing Ltd, The Plant Journal, (2002), 32, 685–699 699 hexokinase I is located in the outer envelope membrane of plastids. FEBS Lett. 461, 13–18. Yang, Y. and Steup, M. (1990) Polysaccharide fraction from higher plants which strongly interacts with the cytosolic phosphorylase isozyme. Plant Physiol. 94, 960–969. Yu, T.S., Kofler, H., Häusler, R.E. et al. (2001) The Arabidopsis sex1 mutant is defective in the R1 protein, a general regulator of starch degradation in plants, and not in the chloroplast hexose transporter. Plant Cell 13, 1907–1918. Zeeman, S.C., Northrop, F., Smith, A.M. and ap Rees, T. (1998a) A starch-accumulating mutant of Arabidopsis thaliana deficient in a chloroplastic starch-hydrolysing enzyme. Plant J. 15, 357–365. Zeeman, S.C., Umemoto, T., Lue, W.-L., Au-Yeung, P., Martin, C., Smith, A.M. and Chen, J. (1998b) A mutant of Arabidopsis lacking a chloroplastic isoamylase accumulates both starch and phytoglycogen. Plant Cell 10, 1699–1711.