Plastidial a-Glucan Phosphorylase Is Not Required for Starch ...

3 downloads 45 Views 475KB Size Report
Samuel C. Zeeman*, David Thorneycroft, Nicole Schupp, Andrew Chapple, Melanie ...... Edwards A, Fulton DC, Hylton CM, Jobling SA, Gidley M, Rossner U,.
Plastidial a-Glucan Phosphorylase Is Not Required for Starch Degradation in Arabidopsis Leaves But Has a Role in the Tolerance of Abiotic Stress1 Samuel C. Zeeman*, David Thorneycroft, Nicole Schupp, Andrew Chapple, Melanie Weck, Hannah Dunstan, Pierre Haldimann, Nicole Bechtold, Alison M. Smith, and Steven M. Smith Institute of Plant Sciences, University of Bern, CH–3013 Bern, Switzerland (S.C.Z., P.H.); Institute of Cell and Molecular Biology, University of Edinburgh, Edinburgh EH9 3JH, United Kingdom (D.T., H.D., S.M.S.); John Innes Centre, Norwich NR4 7UH, United Kingdom (N.S., A.C., M.W., A.M.S.); and Institut National de la Recherche Agronomique, Lab Ge´ne´tique et Ame´lioration des Plantes, F–78026 Versailles, France (N.B.)

To study the role of the plastidial a-glucan phosphorylase in starch metabolism in the leaves of Arabidopsis, two independent mutant lines containing T-DNA insertions within the phosphorylase gene were identified. Both insertions eliminate the activity of the plastidial a-glucan phosphorylase. Measurement of other enzymes of starch metabolism reveals only minor changes compared with the wild type. The loss of plastidial a-glucan phosphorylase does not cause a significant change in the total accumulation of starch during the day or its remobilization at night. Starch structure and composition are unaltered. However, mutant plants display lesions on their leaves that are not seen on wild-type plants, and mesophyll cells bordering the lesions accumulate high levels of starch. Lesion formation is abolished by growing plants under 100% humidity in still air, but subsequent transfer to circulating air with lower humidity causes extensive wilting in the mutant leaves. Wilted sectors die, causing large lesions that are bordered by starch-accumulating cells. Similar lesions are caused by the application of acute salt stress to mature plants. We conclude that plastidial phosphorylase is not required for the degradation of starch, but that it plays a role in the capacity of the leaf lamina to endure a transient water deficit.

a-Glucan phosphorylase (EC 2.4.1.1) is a key enzyme in glucan catabolism in animals, fungi, and prokaryotes (Newgard et al., 1989). It is also considered to be an important enzyme in the degradation of starch in plants, but definitive evidence for this role is lacking (Preiss, 1982; Beck and Ziegler, 1989; Sonnewald et al., 1995; Trethewey and Smith, 2000). The enzyme catalyzes the phosphorolysis of the terminal residue from the nonreducing ends of a-1,4-linked glucan chains, liberating Glc-1-phosphate (Glc-1-P). Although this reaction is reversible, it has been argued that the relatively low levels of Glc-1-P and high levels of inorganic phosphate in plant cells favor the glucandegrading reaction (Kruger and ap Rees, 1983a). All plants studied so far have plastidial and cytosolic isoforms of phosphorylase, which are encoded by separate genes (Nakano et al., 1989; Mori et al., 1991; Van Berkel et al., 1991; Buchner et al., 1996). Starch is 1

This work was supported by the National Centre of Competence in Research (Plant Survival), National Science Foundation, Switzerland, and by the Biotechnology and Biological Science Research Council (BBSRC) of the U.K. (grant nos. D11089 and D11090 and a competitive strategic grant to the John Innes Centre). M.W. was supported by a Leonardo award from the European Union. * Corresponding author; e-mail [email protected]; fax 41–31–631–5222. Article, publication date, and citation information can be found at www.plantphysiol.org/cgi/doi/10.1104/pp.103.032631.

synthesized exclusively in plastids, so only the plastidial isoform is implicated in its metabolism. The amino acid sequence of a-glucan phosphorylase is conserved between prokaryotes and eukaryotes (Newgard et al., 1989). However, plastidial isoforms of starch phosphorylase differ significantly from all other forms due to the presence of an insert of 80 amino acids in length. This domain contributes toward the marked difference in substrate preference between the plastidial and cytosolic isoforms (Mori et al., 1993). The plastidial form (L-form) has a low affinity for large branched glucans and a high affinity for small linear maltodextrins while the cytosolic isoform (H-form) has the opposite relative affinities. Chloroplasts contain amylases in addition to phosphorylase (Stitt et al., 1978; Okita et al., 1979; Lin et al., 1988; Li et al., 1992). Starch degradation in isolated chloroplasts results in either phosphorylated or nonphosphorylated products or both depending upon incubation conditions (Stitt and Heldt, 1981; Kruger and ap Rees, 1983b). However, the relative contributions of the phosphorolytic and hydrolytic pathways in vivo are not known. Stitt et al. (1985) suggested that the products of phosphorolysis and hydrolysis could have different metabolic fates, supporting respiration and Suc synthesis, respectively. Antisense repression of a gene encoding a plastidial isoform of phosphorylase in potato (Solanum tuberosum) decreased the detectable phosphorylase activity in leaves but had

Plant Physiology, June 2004, Vol. 135, pp. 849–858, www.plantphysiol.org Ó 2004 American Society of Plant Biologists

849

Zeeman et al.

no major impact on the accumulation of starch (Sonnewald et al., 1995). However, the nature of the antisense technique and the fact that potato contains a second plastidial isoform, which is expressed in leaves (Albrecht et al., 2001), mean that an important function for phosphorylase in starch metabolism could not be ruled out. The availability of knockout mutants in Arabidopsis now allows us to explore the precise role of plastidial a-glucan phosphorylase in starch degradation and synthesis. Here we report that complete loss of the enzyme does not cause a significant change in the overall accumulation of starch during the day, or its remobilization at night in healthy plants. However, we present evidence that phosphorylase-deficient plants are more sensitive to transient water and salt stress and speculate that, by providing substrates for chloroplast respiratory metabolism, the phosphorolysis of starch may play an important role in stress tolerance.

RESULTS

pools by PCR using multiple primer combinations (Krysan et al., 1996). The second (Atphs1-2) was identified by the Genoplante FLAGdb/FST initiative. T-DNA-specific PCR primers close to the left border gave products with AtPHS1-specific primers for both Atphs1-1 and Atphs1-2. Sequence analysis revealed the precise location of each T-DNA insertion within the AtPHS1 gene (Fig. 1). We used PCR to identify plants that were homozygous for each T-DNA insertion. The disruption of the AtPHS1 gene in Atphs1-1 is expected to result in a null mutation. Reverse transcription (RT)PCR analysis confirmed that no functional AtPHS1 mRNA is produced, although hybrid transcripts containing both AtPHS1 and T-DNA sequences are produced (not shown). In Atphs1-2, the insertion site is upstream of the ATG translational start codon. RT-PCR and RNA gel-blot analyses established that this allele is transcribed from a promoter within the T-DNA, producing a transcript larger than AtPHS1 mRNA (not shown).

Isolation of Plastidial a-Glucan Phosphorylase Mutants

Loss of Plastidial Starch Phosphorylase Activity in Atphs1-1 and Atphs1-2

The Arabidopsis genome contains two genes on chromosome 3 encoding isoforms of a-glucan phosphorylase (At3g29320 and At3g46970). Full-length cDNAs (GenBank accession nos. AY049235 and BT003012, respectively) confirm the gene predictions. The protein encoded by At3g46970 is 841 amino acids long and does not have a predicted chloroplast transit peptide. This most likely encodes a cytosolic isoform (AtPHS2). The protein encoded by At3g29320 is 962 amino acids long including a domain of approximately 80 amino acids characteristic of plastidial isoforms of phosphorylase, and a putative aminoterminal chloroplast transit peptide of 62 amino acids (ChloroP; Emanuelsson et al., 1999). This indicates that At3g29320 encodes the plastidial isoform of a-glucan phosphorylase (AtPHS1). Two Arabidopsis lines containing T-DNA insertions in the AtPHS1 gene were identified from the collection produced at INRA Versailles (Bechtold et al., 1993; Bouchez et al., 1993). The first (Atphs1-1) was identified by screening DNA

We used biochemical assays and native PAGE techniques to analyze the activity of phosphorylase and other starch metabolizing enzymes in wild-type and homozygous Atphs1 plants. Enzymes were assayed under conditions previously determined to be optimal, giving activity that was linear with respect to time and volume of extract added (Zeeman et al., 1998a; Critchley et al., 2001). Phosphorylase was measured by the generation of Glc-1-P from a glucan substrate and Pi. Total activity was significantly lower in Atphs1-1 than in the wild type, using three different glucan substrates (amylopectin, glycogen, and maltoheptaose). The difference was most pronounced using maltoheptaose and least pronounced using glycogen (Table I). This is consistent with a reduction in plastidial phosphorylase. The plastidial isoform from other species has a marked preference for short linear malto-oligosaccharides over large branched substrates, whereas the converse is true for the cytosolic isoform (Shimomura et al., 1982; Steup, 1988).

Figure 1. T-DNA insertion mutations at the AtPHS1 locus. Structure of the AtPHS1 gene; exons are depicted as closed boxes. T-DNA left border sequence is on a hatched background. Underlined sequence is unidentified DNA lacking similarity to the AtPHS1 locus or the T-DNA sequence. The T-DNA insertion in Atphs1-1 starts 12 bp into intron 6, whereas the insertion site in Atphs1-2 is 99 bp upstream of the ATG translational start codon. 850

Plant Physiol. Vol. 135, 2004

Function of a-Glucan Phosphorylase in Starch Metabolism

Table I. The activities of starch metabolizing enzymes in wild-type and Atphs1-1 plants Measurements are the mean 6 SE from four samples, each from an individual plant. Asterisks mark Atphs1-1 values that are significantly different from wild-type values (*, P # 0.1; **, P # 0.05; ***, P # 0.01). Enzyme

Activity Wild Type

Atphs1-1

nmol min21 g21 fresh weight

Starch phosphorylase Amylopectin substrate Glycogen substrate Maltoheptaose substrate a-Amylase b-Amylase D-enzyme Pullulanase Maltase Starch synthase Branching enzymea

76.3 38.6 77.5 3.6 4,840 321 34.8 19.6 107 4.4

6 6 6 6 6 6 6 6 6 6

3.0 2.9 4.0 0.7 231 7 3.4 0.6 9 0.1

57.5 26.6 35.6 5.7 5,201 215 39.3 19.4 116.7 4.6

6 6 6 6 6 6 6 6 6 6

6.5* 4.3* 3.3*** 0.4** 336 15*** 5.9 3.1 7 0.3

a Branching enzyme activity is measured as the stimulation of 14C incorporation from 14C-Glc-1-P into glucan by phosphorylase a (mmol min21 g21 fresh weight).

Isoforms of phosphorylase in crude extracts of leaves were separated using native PAGE with glycogen-containing gels. Glycogen retards the cytosolic isoform relative to the plastidial one (Steup, 1990). Wild-type plants contained both forms of phosphorylase, but Atphs1-1 and Atphs1-2 contained no detectable plastidial activity (Fig. 2). We conclude that although the coding region is not disrupted in Atphs1-2, the aberrant transcript produced from this gene is either not translated, or results in greatly reduced amounts of AtPHS1 protein, or produces an aberrant protein. We assayed other enzymes of starch metabolism to discover whether there were any pleiotropic changes. b-Amylase, pullulanase, maltase, starch synthase, and starch branching enzyme were unchanged in Atphs1-1 relative to the wild type. a-Amylase activity was 60% higher and disproportionating enzyme 30% lower in the mutant relative to the wild type (Table I). Activity gels were also used to analyze the isoforms of starch hydrolyzing enzymes, starch synthases, and starch branching enzymes. No differences between wild type and mutants were observed (not shown).

3A). The results indicate that under the growth conditions used, the loss of phosphorylase has little or no impact on the overall starch and sugar metabolism in the leaves. Loss of Plastidial Phosphorylase Causes Lesion Formation on Leaves

We observed a consistent phenotypic difference between wild type and both Atphs1 lines. The leaves of mature mutant plants consistently had small, white lesions on the tips or margins of fully expanded leaves. Both independent mutant lines displayed the lesionforming phenotype and in each line, lesion formation cosegregated with the loss of plastidial phosphorylase. The lesions did not increase in size after appearance and were not bordered by chlorotic tissue (Fig. 4A). The frequency and severity of lesions varied between batches of plants, but lesion-free mutant plants were rare. Further examination of the lesions on leaves of the mutant plants revealed that high levels of starch were present in the living cells bordering each lesion at the end of the night, when the rest of the leaf had metabolized its starch (Fig. 4B). This starch accumulation occurred in a highly cell-specific manner (Fig. 4C) with adjacent starch-rich and starch-free cells. At the end of the day, all cells contained starch and could not be distinguished by qualitative iodine staining (not shown). Because the regions of starch accumulation were so small compared with the total leaf area, the measurements of total leaf starch (Fig. 3A) did not detect this difference between the wild type and mutant. The localized accumulation of starch around the lesions of Atphs1 leaves led us to investigate whether activities of a-amylase and D-enzyme (Table I) were altered specifically in lesion-bearing leaves. A separate batch of plants was grown and a-amylase and

Loss of Plastidial Phosphorylase Does Not Alter Total Starch Content or Starch Structure

We measured the starch content in wild type and Atphs1-1 leaves over the diurnal cycle. No difference between wild-type and mutant plants was observed (Fig. 3A). There was also no difference in the distribution of chain lengths of the amylopectin fraction (Fig. 3B), or in the amylose to amylopectin ratio of starch extracted from leaves at the end of the day (Fig. 3C). Comparison of the Suc and free hexose contents of the leaves over the diurnal cycle revealed only minor differences between the wild type and mutant (Fig. Plant Physiol. Vol. 135, 2004

Figure 2. Loss of plastidial phosphorylase activity in Atphs1-1 and Atphs1-2. Soluble proteins from crude extracts of leaves were subjected to native PAGE and stained to reveal phosphorylase isoforms. The cytosolic form (H-form) is present in all extracts. The chloroplast form (L-form) is missing in extracts of Atphs1-1 and Atphs1-2. 851

Zeeman et al.

Figure 3. Starch and sugar contents of wild-type and Atphs1-1plants during the diurnal cycle. A, Samples comprising all the leaves of individual wild-type (white symbols, solid line) or Atphs1-1plants (black symbols, dashed line) plants were harvested and immediately frozen in liquid N2. Leaf material was ground to a powder and extracted in ice-cold perchloric acid. Starch in the insoluble material and sugars in the soluble fraction were measured. Each point is the mean 6 SE from four replicate samples. B, Starch samples (1 mg) from leaves of the wild type (gray bars) and Atphs1-1 (black bars) were debranched with isoamylase, derivatized with the fluorophore APTS, and chains of different lengths separated by PAGE in an Applied Biosystems DNA sequencer (Foster City, CA). Peak areas of chains between 6- and 41-Glc units in length were summed and the areas of individual peaks expressed as a percentage of the total. Representative data from one of three independent experiments are shown. C, Starch samples (0.5 mg) of leaves from wild type (gray circles, solid line) and Atphs1-1 (black circles, dashed line) were separated by Sepharose CL2B chromatography. The absorbance of each fraction at 595 nm was determined after the addition of an iodine solution. To normalize each data set, the absorbance of each fraction was divided by the summed absorbencies of all the fractions. Representative data from one of three independent experiments are shown.

D-enzyme were measured in wild-type leaves and in Atphs1-1 leaves that were either healthy or that contained visible lesions. The results (Table II) indicate that the changes in these enzymes are associated with lesion formation. The activities were the same in wildtype leaves and mutant leaves without lesions, but in mutant leaves with lesions, the activity of a-amylase was increased 80% and that of D-enzyme was decreased 50%, relative to the wild type. Atphs1 Mutants Are Susceptible to Transient Water Deficit

We observed that the growth conditions influenced the occurrence of lesions on the mutant plants. A particularly high frequency was observed after seedlings had been transplanted from germination medium to individual pots. After transplantation, trays of individual pots were well watered and covered for 5 d with a clear plastic lid to increase the humidity near 852

to 100%. Lesions developed on the mutant leaves over a period of 3 to 5 d after the lid was removed. We reasoned that the sudden reduction in humidity might be promoting lesion formation. Accordingly, plants were grown from seed at close to 100% humidity in still air for 5 weeks. Transplantation to potting compost was carried out after 3 weeks. No lesions were visible at 3 weeks or at 5 weeks on either wildtype or mutant plants (Fig. 5A). After 5 weeks, the plants were shifted abruptly to circulating air with 60% 6 5% relative humidity at the end of their normal photoperiod, then photographed and scored for lesions on successive days. After 1 d the fully expanded leaves of the mutant plants were severely wilted. Wilted sectors died, leading to lesions similar to those seen under normal growth conditions but more severe (Fig. 5, B and C). Minor symptoms were visible on some fully expanded wild-type leaves. Leaves present at the time of the shift to lower humidity were scored Plant Physiol. Vol. 135, 2004

Function of a-Glucan Phosphorylase in Starch Metabolism

Figure 4. The formation of lesions bordered by starch-accumulating cells in Atphs1 mutants. A, A typical 5-week-old Atphs1-1 plant grown in a growth chamber with a 12-h light/12-h dark regime (175 mmol photons m22 s21), 20°C, 75% relative humidity. Plants were sown onto germination medium and transplanted after 2 to 3 weeks into individual pots. The arrowhead marks a visible lesion. Bar represents 1 cm. B, The same plant as in A, decolorized with 80% ethanol and stained with iodine at the end of the dark period. The arrowhead marks the local accumulation of starch around the lesion site. Bar represents 1 cm. C, Light micrograph of leaf lamina bordering the lesion site shown in B revealing cell-specific accumulation of starch. Bar represents 0.25 mm.

as containing no lesion, minor lesion (#30% leaf area), major lesion ($30% leaf area), and dead. In the mutant, 75% of the leaves developed lesions, the majority of which were severe. In the wild type only 25% of the leaves developed lesions, very few of which were severe (Fig. 6). Lesion-bearing leaves were stained with iodine 6 d after the shift to lower humidity. Starch accumulating cells were visible bordering the lesions in the mutant plants, but not in the wild type (not shown). Leaves of Atphs1-1 that emerged after the shift to lower humidity exhibited far fewer lesions than the mature leaves. No lesions were visible on new wildtype leaves (Fig. 5B). A second batch of plants treated identically gave very similar results. The susceptibility of wild-type and mutant plants to acute salt stress was also investigated. Plants were grown under high humidity as described above to prevent lesion formation. After 5 weeks the soil was thoroughly irrigated with 500 mM NaCl (see ‘‘Materials and Methods’’). The humidity was maintained near 100%. This salt treatment caused a loss of turgor in wild type and both mutant lines. Symptoms were more severe in mutant plants and, after 6 d, severe lesions had developed on mutant plants but not on the wild type (Fig. 5D). We investigated the stomatal response to an abrupt exposure to water deficit. First, we detached mutant and wild-type rosettes from their root systems and weighed them at intervals over 60 min. There was appreciable water loss at first, but the rate rapidly decreased and was 20% to 25% of the initial value after 60 min (Fig. 7A). The rate of water loss was the same from both mutants and wild-type plants. Second, we investigated gas exchange of individual mutant and wild-type leaves using infrared gas analysis. Leaves that had attained a steady state of gas exchange in the cuvette were detached to stop their water supply. This caused an appreciable and essentially identical decrease in stomatal conductance in mutant and wild type plants (Fig. 7B). Abruptly decreasing the relative Plant Physiol. Vol. 135, 2004

humidity of the air around attached leaves by 75% resulted in increased transpiration, but not stomatal closure, in both the mutants and in the wild type (not shown). Subsequent detachment of the leaf did cause stomatal closure. We used light microscopy to examine the starch content of guard cells in iodine-stained epidermal peels. There was no visible difference between the wild type and Atphs1-1 in the starch content of guard cell chloroplasts 1 h before the start and 1 h before the end of the photoperiod. Both lines contained starch at both time points (not shown). Together, these results suggest that stomatal function in the mutants does not differ from that of the wild type.

DISCUSSION Starch Degradation in Arabidopsis Is Primarily Hydrolytic

Our results show that the loss of plastidial a-glucan phosphorylase does not impact significantly on the metabolism of starch when viewed on a whole plant basis. Leaves of two independently isolated knockout mutants accumulate and degrade the same amount of

Table II. The activities of a-amylase and D-enzyme in healthy and lesion-bearing leaves of Atphs1-1 plants Measurements are the mean 6 SE from four samples, each from an individual plant. Different letters indicate that values are significantly different to each other. Activity Enzyme

Atphs1-1 Leaves

Wild-Type Leaves

Healthy

nmol min

a-Amylase D-Enzyme

2.3 6 0.3 (a) 269 6 8.4 (c)

21

g

21

Lesion-Bearing

fresh weight

2.1 6 0.1 (a) 4.2 6 0.5 (b) 230 6 22 (c) 140 6 12 (d) 853

Zeeman et al.

Figure 5. Lesion formation in Atphs1 can be induced by transient water stress. A, Wild-type (top two rows) and Atphs1-1 plants (bottom two rows) grown as described in Figure 4A, except that still air with a humidity near 100% was maintained with a clear plastic cover. No lesions were visible on mutant plants at this stage of growth. The plastic cover was then removed and the plants exposed to circulating air with 60% 6 5% relative humidity. B, The same plants as in A 3 d after the plastic cover was removed. Wild-type plants exhibit only infrequent wilt symptoms at leaf margins and continue to grow. Mutant plants exhibit severe wilt symptoms, with the death of entire leaves, and an overall suppression of growth. C, Enlargement of a single mature leaf of an Atphs1-1 plant 1, 2, 3, and 6 d after the switch from high to low humidity described in A. D, Wild-type and Atphs1-2 plants grown as described in A were irrigated with 0.5 M NaCl as described in ‘‘Materials and Methods.’’ Wild-type and mutant plants exhibited a loss of turgor in existing leaves and the growth of new, pale green, wrinkled leaves. Plants were photographed 6 d after the beginning of the salt treatment. Identical results were obtained with Atphs1-1. 854

Plant Physiol. Vol. 135, 2004

Function of a-Glucan Phosphorylase in Starch Metabolism

also be the case in potato, starch degradation in mesophyll chloroplasts is primarily hydrolytic. The Role of Plastidial Phosphorylase

Although analysis of total leaf starch and sugar contents revealed no differences between mutant and wild-type plants, the mutants did display a consistent lesion-forming phenotype, with a marked accumulation of starch around the lesion sites. This starch accumulation was very localized and would not contribute significantly to the whole plant measurements. Lesions were observed in both of the independently isolated Atphs1 mutants and cosegregated with the loss of phosphorylase. The localized starch accumulation is consistent with the mutant phenotypes resulting directly from disruption of the Atphs1 gene, rather than from a secondary effect of T-DNA insertion in the AtPHS1 gene affecting neighboring genes. The failure to mobilize starch via phosphorylase in certain cells under specific environmental conditions may result in cell death and be the basis of the lesion formation. By altering the growth conditions and imposing salt stress we have provided evidence that the resilience of Arabidopsis plants to acute water stress conditions is compromised in Atphs1 mutants. Lesion formation Figure 6. Quantification of lesion formation on leaves of wild type and Atphs1-1. A, Five-week-old wild-type plants (pictured in Fig. 5, A and B) were scored for lesions (day 0) then shifted from high to low humidity as described. Leaves present on day 0 were scored for lesions on days 1, 2, 3, and 6. The percentage of leaves bearing no lesion (white bars), lesions covering less that 30% of the lamina (light gray bars), lesions covering more than 30% of the lamina (dark gray bars), and dead leaves (black bars) is presented. Leaves that developed after day 0 were not scored for lesion formation. SE were lower than 5% (n 5 8). B, Fiveweek-old Atphs1-1 plants (pictured in Fig. 5, A and B) were scored as described in A.

starch as do wild-type plants. There are no detectable alterations in the amylopectin structure or the composition of the starch, no major changes in the activities of other starch-metabolizing enzymes, and the contents of Suc and free hexoses do not differ appreciably from the wild type. If starch phosphorylase catalyzes a major degradative (or synthetic) flux in vivo, our results indicate that it can be replaced in this role, and is thus redundant. Although this result is surprising given the frequent suggestions that phosphorylase is important in starch degradation in leaves (Preiss, 1982; Beck and Ziegler, 1989; Trethewey and Smith, 2000), it is consistent with a growing body of evidence to the contrary. The activity of plastidial phosphorylase in Arabidopsis is low when compared with the amylolytic activities and alone is insufficient to catalyze the observed rate of starch degradation (Lin et al., 1988; Zeeman et al., 1998a). Furthermore, reduction of the plastidial isoforms of phosphorylase in potato using antisense techniques had no reported effects on leaf starch metabolism (Sonnewald et al., 1995; also see introduction). Thus, we conclude that in Arabidopsis, as may Plant Physiol. Vol. 135, 2004

Figure 7. Stomatal function in wild type (white symbols), Atphs1-1 (gray symbols), and Atphs1-2 (black symbols). A, Rate of water loss from detached 5-week-old rosettes of wild type, Atphs1-1, and Atphs1-2 (black symbols). Plants were detached and weighed at intervals over the course of 60 min. Each point is the mean 6 SE of measurements on four replicate plants. B, Stomatal conductance of single leaves from wild type, Atphs1-1, and Atphs1-2. Gas exchange was measured over the course of 60 min following the detachment of the leaf and used to calculate stomatal conductance. Each point is the mean 6 SE of measurements from four replicate leaves. 855

Zeeman et al.

was not observed when plants were grown in still air with high humidity, but was strongly promoted by transfer to low humidity with circulating air and by treatment of the roots with a salt solution. Both changes in conditions are likely to cause a rapid transient water stress in leaves. The combination of still air and high humidity would promote a high stomatal conductance. A sudden increase in transpiration rate due to air circulation and a drop in humidity could result in the loss of excessive amounts of water, particularly from the tips and margins of the leaves, which represent the extremities of the transpirational path. Indeed, the consequence of this stress was visible on the wild-type plants as the margins of some of the largest, mature leaves wilted and died after the change in conditions. The acute salt treatment would also result in the loss of excessive water from the leaf lamina, but in this case through the reduced uptake into the roots and transpirational flow. In addition this treatment would cause a long-term stress as the plant takes up salt. We suggest two possible reasons for the increased sensitivity of Atphs1 plants to these conditions. First, the loss of plastidial phosphorylase could affect stomatal function such that transpirational water loss from the mutant leaves is greater than from the wild type, causing indirect symptoms in the leaf lamina. The metabolism of starch in guard cells has long been viewed as an important source of carbohydrate for potassium counter-ion synthesis or sugar accumulation during stomatal opening (Lloyd, 1908; Outlaw and Manchester, 1979). However, we do not favor this hypothesis because a block in starch degradation would result in a failure of stomatal opening rather than a failure of closure. Furthermore, our results indicate that stomatal closure occurs at the same rate in the mutant as in the wild type (Fig. 7). Second, the loss of plastidial phosphorylase could compromise the ability of the mutant palisade and mesophyll cells to tolerate the transient water stress, leading to cell death and wilting. Such a role is consistent with the local accumulation of starch around the lesion sites. We suggest that in these cells, a shift away from starch hydrolysis toward phosphorolysis has been triggered. The failure of the phosphorolytic pathway in these conditions could thus cause both lesion formation and local starch accumulation. Although we favor this second hypothesis, at this stage we cannot exclude the possibility that lesion phenotype is pleiotropic and only indirectly linked to the loss of plastidial phosphorylase.

tions but is also crucial for controlling the levels of reactive oxygen intermediates via the ascorbate-glutathione cycle, thereby avoiding oxidative stress. Reactive oxygen intermediates are produced from multiple sources in the cell during abiotic stress conditions (including water deficit) and can diffuse between cellular compartments (Smirnoff, 1993; Mittler, 2002). By limiting the operation of the chloroplast OPPP, the loss of phosphorylase could also limit the scavenging of reactive oxygen species. Mesophyll chloroplasts are not able to obtain hexose-phosphate from the cytosol, as they lack the ability to translocate hexosephosphates across the inner envelope (Fliege et al., 1978). Furthermore, chloroplasts are reported to lack hexokinase preventing the conversion of Glc to Glc-6-P (Stitt et al., 1978; Wiese et al., 1999), although this point has recently been questioned (Olsson et al., 2003). The import of five-carbon sugars via the recently discovered xylulose5-P/Pi exchange transporter (Eicks et al., 2002) could potentially provide substrates for the OPPP. We speculate that when the demand for reducing power in the chloroplast at night is high or is suddenly increased by imposition of conditions that generate an oxidative stress, direct provision of hexose-phosphate within the chloroplast via phosphorolytic starch degradation may be essential. Other pathways that could supply substrates for the OPPP may be inadequate. The hypothesis that the lesions on leaves of plants lacking plastidial phosphorylase are a consequence of impaired functioning of chloroplastic OPPP in the dark requires further investigation. We observed that the sensitivity of the mutants to water stress was restricted to existing mature leaves. New leaves that developed after the application of stress conditions did not develop lesions to the same extent, indicating that adaptive changes occurred to enhance tolerance to the new conditions. Thus, phosphorylase probably acts to provide a very rapid response to a sudden environmental change, and is followed by altered gene expression for complete acclimation. Interestingly, a hexose-phosphate transporter that could provide substrates for the oxidative pentose phosphate pathway in the dark is induced by water stress in spinach leaves (Quick et al., 1995). The induction of additional tolerance mechanisms could explain why the penetrance of the lesion-forming phenotype was dependent on the precise growth conditions and why lesion-bearing phenotypes have not been reported for starchless mutants. Producing double mutants lacking both starch and phosphorylase will provide a means to test this theory.

The Role of Hexose Phosphates Supplied by Phosphorylase

MATERIALS AND METHODS

The role of phosphorylase may be to provide hexose phosphates as substrates for the oxidative pentose phosphate pathway (OPPP) inside the chloroplast at night. This pathway utilizes Glc-6-P to provide reducing power (NADPH) for many biosynthetic reac856

Plants and DNA A population of T-DNA-transformed Arabidopsis (Wassilewskija ecotype) was obtained from Institut National de la Recherche Agronomique-Versailles. DNA was prepared from pools of 100 lines in collaboration with P. Benoist and M. Thomas, University of Paris-Sud, Orsay, France. PCR screening was carried

Plant Physiol. Vol. 135, 2004

Function of a-Glucan Phosphorylase in Starch Metabolism

out as described by Germain et al. (2001) and allele Atphs1-1 found in pool 15/5. The mutant line 125B01 was obtained from the Flagdb/FST initiative (http://flagdb-genoplante-info.infobiogen.fr/projects/fst/) and designated Atphs1-2.

Growth Conditions and Gas Exchange Measurements Plants were grown in a greenhouse or a controlled environment chamber. Unless otherwise stated the conditions were as follows. Greenhouse temperature was maintained above 10°C and natural illumination supplemented to provide a minimum photoperiod of 16 h. The controlled environment chamber provided a constant 20°C, 75% relative humidity, and a 12-h/12-h light/dark cycle, with uniform illumination of 175 mmol photons m22 s21. Sown seeds were covered with a clear plastic lid and stratified (4°C, 2 d). Lids were removed 10 d after sowing when the cotyledons had emerged. Seeds were sown either directly onto a peat-based potting compost or germinated first on fine grade seed compost, then transplanted after 2 to 3 weeks into individual pots (5 3 5 3 5 cm) containing potting compost. Transplanted seedlings were again covered with a clear plastic lid for 5 d. To administer acute salt stress, pots containing mature rosettes were immersed in a solution of 500 mM NaCl for 15 min. The rosettes were kept dry. Excess solution was drained and the application repeated twice, each time with fresh salt solution. Plants were further watered on subsequent days with salt solution. To determine kanamycin resistance, seeds were germinated on 0.7% (w/v) agar plates containing 50 mg/L kanamycin and the seedlings scored for resistance after 2 weeks. Gas exchange measurements were made using a CIRAS 1 infrared gas analyser (PP Systems, Hitchin, UK). Mature leaves were placed in the cuvette and illuminated using lights of the growth cabinet. When the leaves had equilibrated, achieving a steady state of gas exchange, the petioles were cut and changes in gas exchange measured over the course of 1 h.

Materials All enzymes and biochemicals were from Roche (Poole, Dorset, UK) except Taq polymerase, isoamylase, and starch azure, which were from SigmaAldrich Chemical (Poole, Dorset, UK).

Native PAGE Native PAGE for isoforms of a-glucan phosphorylase was adapted from Steup (1990). Leaves harvested midway through their photoperiod were extracted in 100 mM 3-[N-morpholino] propanesulfonic acid), pH 7.2, 1 mM dithiothreitol, 1 mM EDTA, and 10% (v/v) ethanediol (extraction buffer). Gels were run exactly as described previously (Zeeman et al., 1998a) except that the incubation medium contained 100 mM Tris-HCl, pH 7.0, 1 mM dithiothreitol, and 20 mM Glc-1-P. Native PAGE for the detection of starch hydrolyzing activities and branching enzymes isoforms were performed exactly as described previously (Zeeman et al., 1998b). Native PAGE for the detection of starch synthases was performed as described by Edwards et al. (1995).

Enzyme Measurements Enzyme assays were previously optimized with respect to pH and the concentrations of all components of the assays using extracts of leaves of the Wassilewskija wild-type ecotype (Zeeman et al., 1998a; Critchley et al., 2001). a-Amylase, b-amylase, maltase, disproportionating enzymes, starch synthase, and starch branching enzyme were measured as described in Zeeman et al. (1998a). a-Glucan phosphorylase was assayed using a continuous assay in the direction of Glc-1-P formation, coupled to the production of NADH. Leaves were extracted in extraction buffer (see above). The 1-mL reaction mixture contained 20 mM 3-[N-morpholino] propanesulfonic acid), pH 7.0, 20 mM Na2HPO4/KH2PO4, 10 mM MgCl2, 3.4 mM NAD, 1 unit phosphoglucomutase (from rabbit muscle), 1 unit Glc-6-phosphate dehydrogenase (from Leuconostoc mesenteroides), 2.5 mM Glc-1,6-bisphosphate, and the glucan substrate (amylopectin, glycogen, or maltoheptaose, with final concentrations of 2.5 mg mL21, 1.0 mg mL21, and 1 mM, respectively).

Metabolite Measurements For the extraction and measurements of starch and sugars, samples comprising all the leaves of individual plants were harvested and immedi-

Plant Physiol. Vol. 135, 2004

ately frozen in liquid N2. Extracts were made using 0.7 M perchloric acid as described in Critchley et al. (2001). Starch was measured in the insoluble pellet (Critchley et al., 2001). Suc and hexoses were measured as described in Zeeman and ap Rees (1999). Leaves were stained for starch by killing and decolorizing the tissue in hot 80% ethanol, then staining with Lugols solution (Sigma).

Starch Composition and Structure Separation of amylose and amylopectin using Sepharose CL2B (Sigma) in a 9-mL column was performed as described by Denyer et al. (1995). The distribution of chain lengths of amylopectin was analyzed using the fluorophore assisted PAGE system described by Edwards et al. (1999). Sequence data from this article have been deposited with the EMBL/ GenBank data libraries under accession numbers AP001309 and AL133292.

ACKNOWLEDGMENTS We thank Doris Rentsch and Silvio Arpagaus for critical review of the manuscript, and Martine Trevisan for assistance growing the plants. Received August 31, 2003; returned for revision March 10, 2004; accepted March 10, 2004.

LITERATURE CITED Albrecht T, Koch A, Lode A, Greve B, Schneider-Mergener J, Steup M (2001) Plastidic (Pho1-type) phosphorylase isoforms in potato (Solanum tuberosum L.) plants: expression analysis and immunochemical characterisation. Planta 213: 602–613 Bechtold N, Ellis J, Pelletier G (1993) In planta Agrobacterium-mediated gene transfer by infiltration of adult Arabidopsis thaliana plants. C R Acad Sci Paris Life Sci 316: 1194–1199 Beck E, Ziegler P (1989) Biosynthesis and degradation of starch in higher plants. Annu Rev Plant Physiol Plant Mol Biol 40: 95–117 Bouchez D, Camilleri C, Caboche M (1993) A binary vector based on Basta resistance for in planta transformation of Arabidopsis thaliana. C R Acad Sci Paris Life Sci 316: 1188–1193 Buchner P, Borisjuk L, Wobus U (1996) Glucan phosphorylases in Vicia faba L.: cloning, structural analysis and expression patterns of cytosolic and plastidic forms in relation to starch. Planta 199: 64–73 Critchley JH, Zeeman SC, Takaha T, Smith AM, Smith SM (2001) A critical role for disproportionating enzyme in starch breakdown is revealed by a knock-out mutation in Arabidopsis. Plant J 26: 89–100 Denyer K, Barber LM, Burton R, Hedley CL, Hylton CM, Johnson S, Jones DA, Marshall J, Smith AM, Tatge H, et al (1995) The isolation and characterisation of novel low amylose mutants of Pisum sativum L. Plant Cell Environ 18: 1019–1026 Edwards A, Fulton DC, Hylton CM, Jobling SA, Gidley M, Rossner U, Martin C, Smith AM (1999) A combined reduction in activity of starch synthases II and III of potato has novel effects on the starch of tubers. Plant J 17: 251–261 Edwards A, Marshall J, Sidebottom C, Visser RGF, Smith AM, Martin C (1995) Biochemical and molecular characterisation of a novel starch synthase from potato tubers. Plant J 8: 283–294 Eicks M, MaurinoV, Knappe S, Flu¨gge U-I, Fischer K (2002) The plastidic pentose phopsphate translocator represents a link between the cytosolic and the plastidic pentose phosphate pathways in plants. Plant Physiol 128: 512–522 Emanuelsson O, Nielsen H, von Heijne G (1999) ChloroP, a neural network-based method for predicting chloroplast transit peptides and their cleavage sites. Protein Sci 8: 978–984 Fliege R, Flu¨gge UI, Werdan K, Heldt HW (1978) Specific transport of inorganic-phosphate, 3-phosphoglycerate and triosephosphates across inner membrane of envelope in spinach-chloroplasts. Biochim Biophys Acta 502: 232–247 Germain V, Rylott E, Larson TR, Sherson SM, Bechtold N, Carde J-P, Bryce JH, Graham IA, Smith SM (2001) Requirement for 3-ketoacylCoA thiolase-2 in peroxisome development, fatty acid b-oxidation and

857

Zeeman et al.

breakdown of triacylglycerol in lipid bodies of Arabidopsis seedlings. Plant J 28: 1–12 Kruger NJ, ap Rees T (1983a) Properties of a-glucan phosphorylase from pea. Phytochem 22: 1891–1898 Kruger NJ, ap Rees T (1983b) Maltose metabolism by pea chloroplasts. Planta 158: 179–184 Krysan PJ, Young JC, Tax F, Sussman MR (1996) Identification of transferred DNA insertions within Arabidopsis genes involved in signal transduction and ion transport. Proc Natl Acad Sci USA 93: 8145–8150 Li B, Servaites JC, Geiger DR (1992) Characterization and subcellular localization of debranching enzyme and endoamylase from leaves of sugar-beet. Plant Physiol 98: 1277–1284 Lin T-P, Spilatro SR, Preiss J (1988) Subcellular localization and characterization of amylases in Arabidopsis leaf. Plant Physiol 86: 251–259 Lloyd FE (1908) The behaviour of stomata. Carnegie Inst Washington Publ 82: 1–142 Mittler R (2002) Oxidative stress, antioxidants and stress tolerance. Trends Plant Sci 7: 405–410 Mori H, Tanizawa K, Fukui T (1991) Potato tuber type H phosphorylase isozyme. J Biol Chem 28: 18446–18453 Mori H, Tanizawa K, Fukui T (1993) Engineered plant phosphorylases showing extraordinarily high affinity for various a-glucan molecules. Protein Sci 2: 1621–1629 Nakano K, Mori H, Fukui T (1989) Molecular-cloning of cDNA-encoding potato amyloplast alpha-glucan phosphorylase and the structure of its transit peptide. J Biochem (Tokyo) 106: 691–695 Newgard CB, Hwang PK, Fletterick RJ (1989) The family of glycogen phosphorylases: structure and function. Crit Rev Biochem Mol Biol 24: 69–99 Okita TW, Greenberg E, Kuhn DN, Preiss J (1979) Subcellular localization of the starch degradative and biosynthetic enzymes of spinach leaves. Plant Physiol 64: 187–192 Olsson T, Thelander M, Ronne H (2003) A novel type of chloroplast stromal hexokinase is the major phosphorylating enzyme in the moss Physcomitrella patens. J Biol Chem 278: 44439–44447 Outlaw WH, Manchester J (1979) Guard cell starch concentration quantitatively related to stomatal aperture. Plant Physiol 64: 79–82 Preiss J (1982) Regulation of the biosynthesis and degradation of starch. Annu Rev Plant Physiol 33: 455–479 Quick WP, Scheibe R, Neuhaus HE (1995) Induction of hexose-phosphate translocator activity in spinach chloroplasts. Plant Physiol 109: 113–121

858

Shimomura S, Nagai M, Fukui T (1982) Comparative glucan specificities of two types of spinach leaf phosphorylase. J Biochem (Tokyo) 91: 703–717 Smirnoff N (1993) Tansley Review 52. The role of active oxygen in the response of plants to water-deficit and desiccation. New Phytol 125: 27–58 Sonnewald U, Basner A, Greve B, Steup M (1995) A second L-type isozyme of potato glucan phosphorylase: cloning, antisense inhibition and expression analysis. Plant Mol Biol 27: 567–576 Steup M (1988) Starch degradation. In J Preiss, ed, Biochemistry of Plants. Vol 14. Carbohydrates. Academic Press, New York, pp 255–296 Steup M (1990) Starch degrading enzymes. Methods Plant Biochem 3: 103–128 Stitt M, Bulpin PV, ap Rees T (1978) Pathway of starch breakdown in photosynthetic tissues of Pisum sativum. Biochim Biophys Acta 544: 200–214 Stitt M, Heldt HW (1981) Physiological rates of starch breakdown in isolated intact spinach chloroplasts. Plant Physiol 68: 755–761 Stitt M, Wirtz W, Gerhardt R, Heldt HW, Spencer C, Walker D, Foyer C (1985) A comparative study of metabolite levels in plant leaf material in the dark. Planta 166: 354–364 Trethewey RN, Smith AM (2000) Starch metabolism in leaves. In RC Leegood, TD Sharkey, S von Caemmerer, eds, Advances in Photosynthesis. Vol 9. Photosynthesis: Physiology and Metabolism. Kluwer Academic Publishers, Dordrecht, the Netherlands, pp 205–231 Van Berkel J, Conrads-Strauch J, Steup M (1991) Glucan-phosphorylase forms in cotyledons of Pisum sativum L.: localization, developmental change, in vitro translation, and processing. Planta 185: 432–439 Wiese A, Groner F, Sonnewald U, Deppner H, Lerchl J, Hebbeker U, Flu¨gge UI, Weber A (1999) Spinach hexokinase I is located in the outer envelope membrane of plastids. FEBS Lett 461: 13–18 Zeeman SC, ap Rees T (1999) Changes in carbohydrate metabolism and assimilate partitioning in starch-excess mutants of Arabidopsis. Plant Cell Environ 22: 1445–1453 Zeeman SC, Northrop F, Smith AM, ap Rees T (1998a) A starch-accumulating mutant of Arabidopsis thaliana deficient in a chloroplastic starchhydrolysing enzyme. Plant J 15: 357–365 Zeeman SC, Umemoto T, Lue WL, Au-Yeung P, Martin C, Smith AM, Chen J (1998b) A mutant of Arabidopsis lacking a chloroplastic isoamylase accumulates both starch and phytoglycogen. Plant Cell 10: 1699–1711

Plant Physiol. Vol. 135, 2004