Photosynthesis Research 14:211-227 (1987) © Martinus Nijhoff Publishers, Dordrecht - Printed in the Netherlands R e g u l a r paper
Regulation of photosynthetic carbon metabolism during phosphate limitation of photosynthesis in isolated spinach chloroplasts C H R I S T O P H G I E R S C H l & S I M O N P. R O B I N S O N CSIRO, Division of Horticultural Research, GPO Box 350, Adelaide, 5001, Australia; Permanent address to which correspondence should be sent: Botanisches lnstitut, Universiti~t Diisseldorf, Universitiitsstrafle 1, D-4000 Diisseldorf, FRG Received 9 June 1987; accepted in revised form 15 July 1987
Key words: carbon metabolism, chloroplasts, phosphate, photosynthesis Abstract. Intact chloroplasts isolated from spinach were illuminated in the absence of inorgan-
ic phosphate (Pi) or with optimum concentrations of Pi added to the reaction medium. In the absence of Pi photosynthesis declined after the first 1-2rnin and was less than 10% of the maximum rate after 5 min. Export from the chloroplast was inhibited, with up to 600 of the ~4Cfixed being retained in the chloroplast, compared to less than 20% in the presence of Pi. Despite the decreased export, chloroplasts depleted of Pi had lower levels of triose phosphate while the percentage of total phosphate in 3-phosphoglycerate was increased. Chloroplast ATP declined during Pi depletion and reached dark levels after 3 4 min in the light without added Pi. At this point, stromal Pi concentration was 0.2 mM, which would be limiting to ATP synthesis. Addition of Pi resulted in a rapid burst of oxygen evolution which was not initially accompanied by net CO2 fixation. There was a large decrease in 3-phosphoglycerate and hexose plus pentose monophosphates in the chloroplast stroma and a lesser decrease in fructose-l,6-bisphosphate. Stromal levels of triose phosphate, ribulose-l,5-bisphosphate and ATP increased after resupply of Pi. There was an increased export of t4C-labelledcompounds into the medium, mostly as triose phosphate. Light activation of both fructose-l,6-bisphosphatase and ribulose-l,5-bisphosphate carboxylase was decreased in the absence of Pi but increased following Pi addition. It is concluded that limitation of Pi supply to isolated chloroplasts reduced stromal Pi to the point where it limits ATP synthesis. The resulting decrease in ATP inhibits reduction of 3-phosphoglycerate to triose phosphate via mass action effects on 3-phosphoglycerate kinase. The lack of Pi in the medium also inhibits export of triose phosphate from the chloroplast via the phosphate transporter. Other sites of inhibitionof photosynthesis during Pi limitation may be located in the regenerative phase of the reductive pentose phosphate pathway. FBP- Fructose- 1,6-bisphosphate, FBPase- Fructose- 1,6-bisphosphatase, MP-Hexose plus pentose monophosphates, PGA-3-phosphoglycerate, Pi inorganic orthophosphate, RuBP- ribulose- 1,5-bisphosphate, RuBPCase- ribulose- 1,5-bisphosphate carboxylase, TP Triose Phosphate Abbreviations:
212 Introduction
Inorganic phosphate (Pi) has a major role in the regulation of photosynthetic carbon metabolism. The chloroplast is predominantly a Pi-consuming, triose phosphate-exporting organelle and the chloroplast pool is unable to support continued CO2 fixation for more than a few minutes. This Pi-dependence of CO: fixation is obvious with isolated chloroplasts, where photosynthesis soon ceases in the absence of added Pi but can be restarted by addition of Pi to the suspending medium (Cockburn et al. 1967). In vivo, triose phosphate is exported from the chloroplast in exchange for the uptake of Pi, via the phosphate transporter, which maintains the chloropolast Pi pool and allows photosynthesis to continue. The triose phosphate is used in the cytosol predominantly for sucrose synthesis, thereby releasing Pi to maintain the cytosolic Pi pool and allow the process to continue. Manipulation of in vivo Pi pools suggests that this recycling of Pi from the cytosol is a major regulatory factor in the photosynthetic process. By feeding sugars such as mannose to leaves, cytosolic Pi is sequestered as non-metabolized sugar phosphates, photosynthesis is inhibited, and starch synthesis is increased (Chen-She et al. 1975, Herold et al. 1976). Photosynthesis is also inhibited in plants grown with an insufficient supply of Pi, but the rate can be significantly increased within a short time by feeding Pi to the leaves from these Pi-deficient plants (Brooks 1986, Dietz and Foyer 1986). Sivak and Walker (1986) reported that photosynthesis can also be significantly increased by feeding Pi to the petiole of leaves from plants grown in complete nutrient media. This suggests that supply of Pi to the chloroplast may be a limiting factor in photosynthesis even under normal conditions. The vacuole of leaf cells normally contains a large pool of phosphate but this does not appear to equilibrate rapidly with the cytosolic Pi pool (Foyer and Spencer 1986, Woodrow et al. 1984). Because the turnover time of the cytosolic Pi pool during photosynthesis is short, any decrease in the rate of Pi release into the cytosol would rapidly decrease the concentration of Pi in the cytosol. This in turn would decrease Pi import into the chloroplast and photosynthesis would become Pi-limited. To identify inhibitory and regulatory steps of CO: fixation under such short-term Pi limitation, we have studied depletion of Pi and the dynamical transition induced by its resupply by measuring oxygen evolution, CO2 fixation, metabolite pools and enzyme activities.
213 Materials and methods
Plant material Spinach seeds (Spinacia oleracea L. cv. Hybrid 102) were germinated in moist peat and after 10 days the seedlings were transferred to hydroponic culture. Four seedlings were placed in each 6-1itre container with the following nutrient solution: 6 m M KNO3, 2 m M MgSO4, l m M KH2PO4, 4 m M Ca(NO3)2, 50/~M FeNa(EDTA)2, 50pM H3BO3, 10pM MnCI2, l pM ZnSO4, 0.5pM CuSO4 and 0.1#M Na2MoO4. Plants were grown in a glasshouse but the day length was restricted to 12 hours by a mechanical shutter and supplementary lighting was provided by fluorescent and incandescent lights to ensure a minimum light intensity of 100/~mol-m-2"s -~ (PAR). The pots were topped up with de-ionized water as required and were aerated continuously.
Chloroplast &olation Intact chloroplasts were isolated from spinach leaves and purified on a two-step Percoll gradient. All procedures were carried out at 0 °C. Leaves (30-70 g) were ground for 3 seconds in a Polytron blender with 200 ml of 330mM sorbitol, 5 m M MgC12, 10mM NaaP2OT, 2 m M isoascorbate, and 0.1% BSA (pH 6.5). The brei was squeezed through two layers of Miracloth containing a layer of cotton wool and the filtrate was centrifuged at 1700 g for 1 minute. The pellets were resuspended in 6 ml of 330 mM sorbitol, 2 mM EDTA, 1 mM MgC12, 1 mM MnC12, 50 mM Hepes-KOH, and 0.2% BSA (pH 7.6) and placed into two centrifuge tubes. Each was underlayered with 4 ml of the same medium plus 40% (v/v) Percoll, then the tubes were again centrifuged at 1700 g for 1 min. Broken chloroplasts formed a band at the top of the Percoll layer, whereas intact chloroplasts were pelleted by this procedure. The supernatants were discarded and the pellets of intact chloroplasts were resuspended in the above medium. The total isolation took 15 to 20min and 02 evolution by the chloroplasts could be measured within 30min of harvesting the leaves. The chloroplasts were greater than 95% intact based on penetration of ferricyanide (Lilley et al. 1975) and exhibited rates of CO2-dependent 0 2 evolution of 150 to 270/~mol.mg Chl ~-h- ~. Chlorophyll was determined by the method of Arnon (1949).
02 evolution CO2-dependent 02 evolution was measured at 20 °C using Hansatech 02 electrodes. Unless stated otherwise, the assay medium contained 330 mM
214 sorbitol, 2 m M EDTA, 1 mM MgC12, 1 mM MnC12, 50mM Hepes-KOH (pH 7.6), 4 mM NaHCO3, 0.4 mM Pi, 1000 units-ml-1 catalase and chloroplasts equivalent to 50 pg Chl.m1-1 . The suspension was illuminated with white light (1500 #mol. m- 2. s- 1 PAR).
14C02fixation Chloroplasts were incubated under the same conditions as for 02 evolution with NaH14CO3 of specific activity 0.19 MBq-mo1-1. Samples were transferred to microcentrifuge tubes containing 50 pl of silicone oil (Wacker type AR 180) on top of 20#1 of 1 N HC10 4. The tubes were centrifuged in a Beckman Microfuge for 20 s and illumination of the samples was continued during this separation procedure. The pellet fraction of each sample was resuspended in 100 #1 H20, and aliquots of the supernatant and pellet fractions were acidified by adding the five-fold volume of 10% (v/v) propionic acid. The samples were dried down and the residues resuspended in 1 ml H20; 10ml scintillation cocktail were added, and the amount of acidstable 14C in each fraction was determined by liquid scintillation counting. The percentage of total 14C incorporated which was retained within the chloroplasts was calculated from these results.
Determination of metabolites Chloroplasts were separated from the medium by silicone oil filtering centrifugation (Heldt 1980) and labelled metabolites determined by HPLC (Giersch 1979, Robinson and Giersch 1987). The following were successively layered in 0.4 ml microcentrifuge tubes: 20 #l 1 N HC104, 75 pl silicone oil (Wacker type AR 180), 200 pl chloroplast suspension. After centrifuging for 20s the supernatant was acidified by addition of 10#l of 12NHCIO4. Illumination was continued during the separation procedure. The supernatant and pellet fractions were removed, centrifuged to remove protein, then neutralized with K2CO 3 and centrifuged again to remove the precipitate of KCIO4. Metabolites were separated by ion exchange chromatography using a 25 cm x 4.6 mm column of Partisil SAX, strongly basic anion exchanger, 10#m particle size. Metabolites were eluted with the following gradient: 5mM KH2PO 4 (Buffer A) for 10min, then a linear gradient from 5 to 200mM KHzPO4 (Buffer B) for 55min followed by isocratic elution in Buffer B for a further 15 min. The pH of Buffer B was adjusted to 2.8 with H 3PO4 and Buffer A was made by diluting Buffer B with water. The flow rate was 0.5 ml.min -1. Radioactivity was detected on-line using a Berthold LB 504 monitor.
215 For 14C-labelled metabolites, chloroplasts were incubated with 2.8 mM NaH14CO3 of specific activity 2.14MBq.mo1-1. To determine 32p-labelled metabolites, chloroplasts (1 mg Chl) were first prelabelled by incubation for 10min at 0°C in l m l of resuspension medium containing 5mM 32pi (1.85 M Bq-mol-l). The suspension was then diluted ten-fold with resuspension medium and layered over 5 ml of resuspension medium containing 40% (v/v) Percoll, centrifuged for 1 min at 1700 g, and the chloroplasts resuspended. Enzyme activities
FBPase activity was measured spectrophotometrically by ading 0.05ml chloroplast suspension to a 1 cm lightpath semimicro cuvette containing in a total volume of 0.45 ml: 10mM MgCI2, 1 mM EDTA, 100mM Tris-HC1 (pH 8.0), 0.2% Triton X-100, 0.3mM NADP, 0.6mM FBP, 2.4units/ml glucose-6-phosphate dehydrogenase and 4units/ml phosphoglucose isomerase. The change in absorbance at 340 mm was recorded for 1 min. The increase in absorption was linear for at least 1 min and proportional to the amount of added chlorophyll. RuBPCase activity was assayed by adding 10/~1of the chloroplast suspension to an Eppendorf tube containing in a total volume of 190#1: 10mM MgCI2, 1 mM EDTA, 10mM KC1, 0.1% Triton X-100, 50mM Hepes, pH (KOH) 8.0, 0.55 mM RuBP and 5.2 mM NaHl4CO3 (0.17 MBq//~mol). CO2 fixation was stopped after 30 s by addition of 100/A 50% propionic acid. The acidified samples were dried down, and the acid stable lable was counted in a liquid scintillation counter after redissolving the dried residue in 100#1 H2 O and adding 1 ml scintillation cocktail. RuBPCase activity was linear for at least 60 s. Parallel experiments where the RuBPCase activity was measured by following the rate of PGA formation enzymically (Lilley and Walker 1979), gave essentially the same results. A TP determination
Chloroplasts were incubated as for 02 evolution measurements and samples (80 #1) were injected into 40 #1 of 3 N H C 1 0 4 with rapid stirring. Illumination of the samples was continued during the killing procedure. The acidified samples were centrifuged to remove protein, neutralized with K2CO 3 and then centrifuged again to remove the precipitate of K C 1 0 4. ATP was determined in a 10 pl aliquot by the luciferin-luciferase bioluminescence method using Boehringer CLS reagent in an LKB 1250 luminometer.
216 Stromal pH The pH of the chloroplast stroma and intrathylakoid spaces were determined as described by Robinson (1985). Chemicals Silicone oil was a gift from Wacker (Australia), RuBP was kindly provided by J. Andrews, Res. School of Biological Sciences, Canberra. Radioisotopes were obtained from Amersham, UK, and enzyme solutions from Boehringer, Mannheim, West Germany. All other biochemicals were obtained from Sigma Chemical Co., St. Louis, USA.
Results
Oxygen evolution and CO 2fixation Intact isolated chloroplasts illuminated in a medium with optimal concentrations of Pi (0.2-0.5 mM) exhibited high rates of CO2-dependent 02 evolution (150-270 #mol. mg Chl- ~. h - ~). Following the normal induction phase, a linear rate of 02 evolution was maintained for several minutes. If Pi was omitted from the reaction medium, the induction phase was shorter and a lower maximum rate was achieved, after which the rate declined due to depletion of Pi. This decline in O: evolution had the characteristic kinetics shown in Fig. 1, which is typical of a number of experiments with purified intact chloroplasts. The maximum rate was achieved after 1-2 min, then there was a sharp transition to a lower rate which was maintained for a further 1-2 min before 02 evolution again declined to a steady-state rate of 5-20 ~mol" mg Chl- ~"h-~ (Fig. 1). During the second phase, there was often a secondary increase in 02 evolution which is evident from the plot of rate of 02 evolution against time in the upper panel of Fig. 1. The final rate was maintained for several minutes and probably reflects the rate of starch synthesis, which does not result in net consumption of Pi. Addition of Pi at this point caused an immediate burst of 02 evolution, lasting 15-30 s, which was often followed by a transitory decline in the rate of O2 evolution before a new steady-state rate was achieved 0.5-1 min after adding Pi (Fig. 1). Measurement of CO2 fixation (incorporation of ~4C) indicated that it paralleled 02 evolution during the depletion of Pi, whereas the burst of oxygen evolution immediately following the addition of Pi (Fig. 1) was not accompanied by any net fixation of ~4CO2, which only started 0.5-1 min after
217 J~
7 ~40~
~20~ IO0'! ~o~ ~0) ~0~ ~oi ,
•
oi
4
5
6
7
0,35~ 0.30 /
0.25
/11~
~ 0
1
2
3 Time
4 5 In L~h~( nlln )
0
7
Fig. 1. Oxygen evolution by intact isolated chloroplasts illuminated in a reaction medium without added Pi. Oxygen electrode trace is shown in the lower panel with numbers indicating the rate of oxygen evolution in ~mol O2 • mg Chl-~ • h ~. The rate of oxygen evolution is plotted in the upper panel. Chloroplasts equivalent to 50 ~tg Chl were added in a total volume o f 1 ml. After 5 rain illumination, Pi was added to a final concentration of 0.4 raM. The dotted line in the lower panel shows oxygen evolution in a parallel experiment where no addition of Pi was made.
Pi was added (data not shown). Export of ]4C-labelled products into the medium was inhibited in the absence of added Pi. After 5 min in the light, nearly 60% of the ~4C fixed was retained within the chloroplasts whereas with optimum Pi in the reaction medium less than 20% of the ]4C incorporated was in the chloroplast by the time the maximum rate of photosynthesis was achieved (Fig. 2). Upon addition of Pi to Pi-depleted chloroplasts there was an immediate export of ~4C into the medium. The amount of 14C in the chloroplast decreased by 40% and there was an equivalent increase in the amount of ~4C in the medium, resulting in a decrease in the percentage of ]4C in the chloroplast (Fig. 2). This was followed by a small but reproducible increase in the percentage of ~4C in the chloroplast then a gradual decline over the next 2-4min as the amount of incorporated ~4C found in the medium increased (Fig. 2).
218 80
60
0
40
6
o
,~
'
~
'
,'o
T i m e in L i g h t ( rain )
Fig. 2. Export of ~4C-label from chloroplasts illuminated in a medium with 0.4 mM Pi added ( + Pi) and in the absence of added Pi ( - Pi). After 5 min illumination, Pi was also added to the - Pi medium to give a final concentration of 0.4 mM.
Metabolite pools To determine metabolites, chloroplasts were incubated with 32pi to label the endogenous phosphate pools then washed and illuminated in a medium without Pi until photosynthesis had virtually ceased due to Pi depletion. For comparison, chloroplasts prelabelled with 32p were also illuminated in a medium containing 0.4 mM 32pi until the maximum rate of photosynthesis was achieved (Table 1). In the absence of Pi in the reaction medium, the total Table 1.32p-labelled metabolites in chloroplasts illuminated in the absence and presence of Pi in the reaction medium. The chloroplasts were prelabelled with 32pi, washed, then incubated in the absence or presence of 0.4mM 32Pi of the same specific activity. The chlorophyll concentration was 100pg.ml t. Samples were taken after 5min ( - P i ) or 3min ( + P i ) illumination. Total 32p in the chloroplast fraction was 307 ( - Pi) and 560 ( + Pi) natom, mg Chl- ~. Rates of oxygen evolution were 5 ( - Pi) and 163 ( + Pi)/lmol. mg Chl- ~-h - t. -Pi
+Pi
Metabolite
% of total label
nmol- mg Chl- ~
% of total label
nmol" mg Chl-
MP TP Pi PGA FBP RuBP
28.9 1.4 1.5 47.0 2.7 8.7
89.0 4.4 4.5 144.0 4.2 13.4
28.9 5.5 8.6 28.6 2.8 14.7
162.0 31.0 48.0 159.0 7.9 42.0
219 label in the chloroplast was 45% lower and there was a redistribution of label between individual metabolites. There was a marked decrease in the amount of Pi and triose phosphate (TP) and also decreases in the amounts ofhexose plus pentose monophosphates (MP), FBP and RuBP. The amount of PGA was not significantly decreased and on a percentage basis there was a significant increase in the proportion of 32p in PGA. The ratio PGA/TP increased from 5.1 in chloroplasts with optimal Pi to 32.7 in Pi-depleted chloroplasts. Similar results were obtained when metabolite pools were determined by labelling with 14CO2 (data not shown). Figure 3 shows changes in 14C-label in various metabolites in the chloroplast following addition of Pi to chloroplasts after 5 min illumination without added Pi. Upon addition of Pi, label in PGA and MP was dramatically reduced and that in FBP to a lesser extent. There was a subsequent rise in label in these compounds and 3 min after adding Pi the new steady-state levels were, except for PGA, similar to those found before Pi addition. In contrast, label in TP and RuBP gradually increased after Pi addition then subsequently declined again. The new steady-state levels of TP and RuBP were higher than before Pi addition while PGA was lower. The drastic drop
MP
Pi A d d e d
400
A
300
Cl
200
T 0 100
V
30
q'
B
20
10 Pi
A~ded
3 Time
( rain )
Fig. 3. Metabolites in the stroma of chloroplasts following addition of Pi to the medium. Chloroplasts were illuminated in a medium with 2.8mM NaHJ4CO3 without added Pi for 5 min. Then Pi was added to give a final concentration of 0.2 mM. At the times indicated, chloroplasts were separated from the medium by centrifugation through silicone oil and ~4C-label in each metabolite was determined by HPLC.
220 in PGA upon Pi addition appeared to be mainly due to Pi-induced export of label into the medium. However, label in PGA in the medium only increased by 20 natom 14C-mgChl-] whereas the drop in PGA in the chloroplast was ten-fold higher. Moreover, TP in the chloroplast was not lowered upon Pi addition (Fig. 3) even though the amount of TP in the medium increased from 150 natom ]4C.mg Chl-] at 4.5 min to 590 natom 14C.mg Ch1-1 at 5.5 min. The observation that, despite the large export of TP upon Pi addition the stromal pool of this metabolite actually increased, suggests that a major limitation of carbon flux during Pi limitation was the reduction of PGA to TP. The data suggest that photosynthesis in Pidepleted chloroplasts was limited by the phosphoglycerate kinase or glyceraldehyde phosphate dehydrogenase reactions rather than by low export rates due to a lack of Pi in the medium. This view is supported by the observed burst in 02 evolution in the absence of net 14CO2fixation when Pi was added which indicates a dramatic increase in the rate of PGA reduction. Activities of FBPase and RuBPCase
The metabolite data of Fig. 3 suggests the existence of another limitation of carbon flux under Pi deficiency, residing in the regenerative phase of the reductive pentose phosphate cycle. Release of this inhibition would explain the observed decline in pentose plus hexose monophosphates and increase in RuBP following addition of Pi. FBPase activity in darkened chloroplasts was low but increased upon illumination of chloroplasts both in the absence and presence of Pi (Fig. 4). In the presence of Pi, full activity was obtained ÷Pb
150
O
T" U
o
,oo
I~
L Pi Added
0 T i m e in L i g h t ( rain )
Fig. 4. FBPase activity of chloroplasts incubated in the absence ( - Pi) and presence ( + Pi) of 0.4mM Pi. After 4.75min illumination, Pi was added to the - P i medium to give a final concentration of 0.4 mM. Rates of oxygen evolution in/~mol'mg Chl- ~.h-~ at various times are indicated by the numbers in boxes.
221 after 3-4 min in the light which coincided with achievement of the maximum rate of 02 evolution. In the absence of Pi, activation of FBPase was initially more rapid but ceased after 1 min. After 5 min illumination, the activity of FBPase in Pi-depleted chloroplasts was only about half of that in chloroplasts in the presence of Pi. Upon addition of Pi there was a gradual increase in FBPase activity and within 3 min it reached the same value as chloroplasts illuminated in the presence of Pi from the outset. It should be noted that the maximum rate of 02 evolution was achieved within 0.5-1 min of adding Pi (Fig. 1) but at this point FBPase activity had only increased by 30% (Fig: 4). The activity of RuBPCase was also measured in chloroplasts illuminated in the absence and presence of added Pi (Fig. 5). In the presence of Pi, RuBPCase increased 1.3-2.2-fold upon illumination, with maximum activity achieved after 2-3 min in the light when the maximum rate of 02 evolution was obtained. In the absence of Pi, RuBPCase was often slightly higher in the dark but activation in the light was greatly reduced. In six separate experiments the activity only increased 1.1-1.2-fold upon illumination in the absence of Pi. Addition of Pi partially restored the light activation of RuBPCase (Fig. 5). The extent of the increase was variable but full activation (to the same level as the + P i control) was not observed. As with FBPase, achievement of the maximum rate of 02 evolution normally preceded full activation of RuBPCase after addition of Pi. //-
.....
+P~ / / la=
400
7-
/ /
'
la=
/ I
7" ÷Pi
o
0
0
0
150
© ~
o o
100
f
//
/
....
200
~
/-,(
100
/
SO ,
/'
/J
x 0
/
"~
/
'~-~",
r/
i" i i I
0
/.-" O
I
21
I
--I
___t
I
4 T i m e in L i g h t ( rain
I
6
r
I
I
0
8
)
Fig. 5. Rate of oxygen evolution and RuBP carboxylase activity (Rubisco) of chloroplasts incubated in the absence ( - Pi) and presence ( + Pi) of 0.4 mM Pi. After 4.75 min illumination, Pi was added to the - P i medium to give a final concentration of 0.4 mM.
222 A TP levels
In both the presence and absence of Pi chloroplasts contained low levels of ATP in the dark (5-8nmol.mgChl -~) but after 15s in the light ATP increased to 30-35nmol'mgCh1-1 (Fig. 6). In the presence of Pi, ATP declined after about 1 min in the light, reaching about 20 nmol.mg Chlafter 3 min when the maximum rate of 02 evolution occurred, then decreasing more slowly over the next 3-4 min while high rates of 02 evolution were maintained. In the absence of Pi, ATP declined much more rapidly and reached dark levels after 3-4min in the light, when photosynthesis had virtually ceased. Dark ATP levels of 5-8 nmol.mg Ch1-1 are frequently observed in isolated chloroplasts and this is considered to be a thermodynamically unavailable pool, possibly bound to the chloroplast coupling factor (Inoue et al. 1978, Robinson 1985). Thus the low level of ATP in Pi-deficient chloroplasts indicates virtually complete exhaustion of the available ATP pool. Addition of Pi led to a rapid increase in ATP, resulting in levels above those in chloroplasts incubated in the presence of Pi. The level of ATP subsequently declined slightly over the next 2-3 min (Fig. 6). Similar effects of Pi depletion on stromal ATP were observed by Kaiser and Urbach (1977) and M/ichler et al. (1984). Stromal pH
The pH of the chloroplast stroma is involved in the regulation of photosynthetic carbon metabolism both by its effect on the activity and activation
÷Pi
30
u
o
E
,o
4 v
v
f
Pi A d d e d o
;
,
,
,
/
,
,
,
6
, 8
T i m e in L i g h t ( rain )
Fig. 6. ATP levels in chloroplasts incubated in the absence ( - P i ) and presence ( + Pi) of 0 . 2 m M P i . After 5min illumination, Pi was added to the - P i concentration of 0.2 mM.
medium to give a final
223 state of enzymes and via effects on transport across the chloroplast envelope (Robinson and Walker 1981). As shown in Table 2 the pH of the stroma was not decreased in the absence of Pi. There was a slight decrease in the pH of the intrathylakoid space resulting in a small increase in the calculated pH gradient across the thylakoid membrane, possibly caused by the low rate of photophosphorylation under Pi-limited conditions.
Discussion
In the absence of added Pi, oxygen evolution by isolated chloroplasts is only maintained for a few minutes before declining because of Pi depletion. The kinetics illustrated in Fig. 1 were characteristic of purified intact chloroplasts. Isolated chloroplasts normally contain 300-500 nmol Pi. mg Chl- 1 (Kaiser and Urbach 1977, Robinson et al. 1983, Robinson and Giersch 1987). The oxygen evolution which occurs before endogenous Pi is depleted reflects the utilisation of this Pi for synthesis of the sugar phosphate intermediates of the reductive pentose phosphate cycle. If all the Pi is converted to triose phosphate this would allow 0.9-1.5 pmol 02-mg Chl-1 to be evolved before photosynthesis ceased. The first decline in 02 evolution after 1-2 mins in the light occurred after 1-2 pmol 02 • mg Chl- 1had been evolved (Fig. 1) and probably reflects utilisation of the endogenous Pi pool. The subsequent 02 evolution would depend on recycling of this phosphate pool (via starch synthesis) or release of Pi from other metabolites not normally a part of the reductive pentose phosphate cycle (e.g. pyrophosphate, glucose phosphates etc). Recycling of Pi could also result from phosphatase activity, which can be present as a contaminant in chloroplast preparations (Robinson 1982). The total phosphate pool of the chloroplasts was lower in the absence of Pi in the medium (Table 1) suggesting that there was some net leakage of phosphates out of the chloroplast. Export of metabolites was severely Table 2. Photosynthesis and pH of the chloroplast stroma and intrathylakoid spaces in chloroplasts after 5 rain illumination in the absence ( - P i ) and presence ( + Pi) of 0.4 mM Pi. The reaction medium was pH 7.6.
Rate O: evolution (#mol'mg Chl- ~"h- z) Stromal pH Intrathylakoid pH ApH envelope membrane ApH thylakoid membrane
- Pi
+ Pi
4.0 7.85 5.38 0.25 2.48
204.0 7.86 5.48 0.26 2.38
224 inhibited in the absence of exogenous Pi (Fig. 2). This is to be expected since export of triose phosphate requires uptake of Pi for counter-exchange via the phosphate transporter. Despite the decreased export of triose phosphate into the medium, there was not an accumulation of triose phosphate in chloroplasts in the absence of Pi. On the contrary, triose phosphate was lower in the absence of Pi whereas the percentage of label in PGA was increased (Table 1). This indicates that lack of Pi inhibited reduction of PGA to triose phosphate as well as export of triose phosphate to the medium. Similar changes in PGA and triose phosphate pools were reported by Heldt et al. (1978) and M/ichler and N6sberger (1984) for chloroplasts in the presence of limiting amounts of Pi. The equilibrium position of phosphoglycerate kinase lies in the direction of PGA formation and for the phosphorylation of PGA, as required for photosynthesis, a high ratio of substrates (PGA and ATP) to products (glycerate-1,3-bisphosphate and ADP) is necessary (Robinson and Walker 1981). In chloroplasts, phosphorylation of PGA and reduction to triose phosphate is very sensitive to changes in the concentrations of ATP and ADP (Robinson and Walker 1979). A decrease in stromal Pi would limit ATP synthesis which would decrease ATP, increase ADP and inhibit phosphoglycerate kinase. The decrease in ATP (Fig. 6) paralleled the decrease in photosynthesis (Figs. 1, 5) and after 3-4min illumination in the absence of Pi the chloroplast ATP pool was similar to that in the dark. At this point, the chloroplasts contained less than 5 n m o l P i . m g C h l -~ (Table 1). Assuming a chloroplast volume of 25 #1- mg Chl- ~(Heldt 1980) the stromal Pi concentration would be 0.2 mM, which is considerably below the Km of the chloroplast ATP synthase (Selman and Selman-Reimer 1981). Low Pi may also cause inhibition of PGA reduction via effects on electron transport of the NADP system: if glycerate-l,3-bisphosphate falls low due to inhibition of the phosphoglycerate kinase under low Pi, NADPH will pile up, and electron transport will slow down because of the lack of acceptor. This view is supported by the observation that electron transport is not limited by the high proton gradient built up under nonphosphorylating ( - P i ) conditions in illuminated chloroplasts (Furbank et al. 1987). The decrease in the rate of electron transport by lack of acceptor will reinforce the inhibition of photosynthesis caused by the effect of ATP on phosphoglycerate kinase activity. Thus it is clear that limitation of Pi supply to the chloroplasts was perceived as a decrease in stromal Pi, resulting in lowered ATP which inhibited PGA reduction and probably electron transport. This was a major site of inhibition of the reductive pentose phosphate cycle as indicated by the large increase in ATP and decrease in PGA upon addition of Pi (Figs. 3, 6).
225 A second site of inhibition in the regenerative phase of the reductive pentose phosphate cycle was indicated from the changes in metabolite pools following addition of Pi (Fig. 3). One third of the ATP consumed during CO2 fixation is used in the conversion of ribulose-5-phosphate to RuBP by phosphoribulokinase. This enzyme has a greater affinity for ATP than phosphoglycerate kinase but is inhibited by ADP, PGA and a number of other stromal metabolites (Robinson and Walker 1981, Flfigge et al. 1982, Gardemann et al. 1983). The decrease in MP upon addition of Pi (Fig. 3) probably reflects increased phosphoribulokinase activity as a result of increased stromal ATP and decreased PGA. Light activation of RuBPCase was inhibited in the absence of Pi (Fig. 5) as has been reported previously (Heldt et al. 1978, M~ichler and N6sberger 1984). It should be noted, however, that the activity was still well in excess of that required to maintain the observed rates of photosynthesis. Furthermore, the increase in rate of photosynthesis following Pi addition preceded the increase in RuBPCase. The increase in RuBPCase was less than 50% despite a 20-fold increase in the rate of 02 evolution. This suggests that lack of activation of this enzyme was not a limiting factor to photosynthesis in Pi-depleted chloroplasts. Light activation of FBPase was also decreased in the absence of Pi (Fig. 4). As with RuBPCase, extractable FBPase activity was well in excess of that required for the observed rates of photosynthesis (a rate of 02 evolution of 100/~mol • mg Chl- J" h- J requires FBPase activity of 33#mol-mgChl -~ "h -J) and the increase in photosynthesis following Pi addition preceded the activation of FBPase (Fig. 4). There was not an increase in FBP in the absence of Pi (Table 1) but addition of Pi resulted in a transient decrease in FBP (Fig. 3) suggesting a release of limitation by FBPase. Activation of this enzyme requires reduction by thioredoxin and is dependent on the pH and concentration of Mg 2÷ and FBP (Leegood et al. 1985). There was no decrease in stromal pH in the absence of Pi (Table 2) but there was some decrease in the concentration of FBP (Table 1). In the absence of Pi, electron transport is likely to be greatly decreased, as suggested by the low rates of oxygen evolution. It seems likely that the inhibition of light activation of FBPase in the absence of Pi and its increase following Pi addition reflect changes in the rate of electron transport which would alter the redox state of the enzyme. The results presented here suggest that limitation of Pi supply to the chloroplast leads to a decrease in stromal Pi concentration to the point where ATP synthesis becomes Pi-limited. One major consequence of the decrease of ATP is an inhibition of PGA reduction via mass action effects on phosphoglycerate kinase. Because of the high turnover rate of Pi and
226 phosphorylated metabolites in the reductive pentose phosphate cycle relative to the chloroplast pool sizes, carbon flux trough the cycle increases rapidly following resupply of Pi. Thus the chloroplast is very responsive to changes in the rate of supply of Pi from the cytosol which is consistent with the major role recycling of Pi has in the regulation of photosynthesis.
Acknowledgements We wish to thank Marcia McFie for her excellent technical assistance. Financial support for C. Giersch was provided by the Deutsche Forschungsgemeinschaft and CSIRO.
References Arnon DI (1949) Copper enzymes in isolated chloroplasts. Polyphenol oxidase in Beta vulgaris. Plant Physiol 24:1-15 Brooks A (1986) Effects of phosphorus nutrition on ribulose-l,5-bisphosphate carboxylase activation, photosynthetic quantum yield and amounts of some Calvin cycle metabolites in spinach leaves. Aust J Plant Physiol 13:221-237 Chen-She SH, Lewis DH and Walker DA (1975) Stimulation of photosynthetic starch formation by sequestration of cytoplasmic orthophosphate. New Phytol 74:373-392 Cockburn W, Baldry CW and Walker DA (1967) Oxygen evolution by isolated chloroplasts with carbon dioxide as the hydrogen acceptor. A requirement for orthophosphate or pyrophosphate. Biochim Biophys Acta 131:594-596 Dietz KJ and Foyer C (1986) The relationship between phosphate status and photosynthesis in leaves. Reversibility of the effects of phosphate deficiency on photosynthesis. Planta 167: 376-381 Fl/igge UI, Stitt M, Freisl M and Heldt HW (1982) On the participation of phosphoribulokinase in the light regulation of CO 2 fixation. Plant Physiol 69:263-267 Foyer C and Spencer C (1986) The relationship between phosphate status and photosynthesis in leaves. Effects of intracellular orthophosphate distribution, photosynthesis and assimilate partitioning. Planta 167:369-375 Furbank RT, Foyer CH and Walker DA (1987) The site of limitation of photosynthesis by inorganic phosphate in isolated spinach chloroplasts. Biochim Biophys Acta, in press Gardemann A, Stitt M and Heldt HW (1983) Control of CO 2 fixation. Regulation of spinach ribulose-5-phosphate kinase by stromal metabolite levels. Biochim Biophys Acta 722:51-60 Giersch C (1979) Quantitative high performance liquid chromatographic analysis of ~4Clabelled photosynthetic intermediates in isolated intact chloroplasts. J Chromatography 172:153 161 Heldt HW (1980) Measurement of metabolite movement across the envelope and of the pH in the stroma and the thylakoid space in intact chloroplasts. Methods in Enzymolgy 69: 604-613 Heldt HW, Chon CJ and Lorimer GH (1978) Phosphate requirement for the light activation of ribulose-l,5-bisphosphate carboxylase in intact spinach chloroplasts. FEBS Lett 92: 234-240
227 Herold A, Lewis DH and Walker DA (1976) Sequestration of cytoplasmic orthophosphate by mannose and its differential effect on photosynthetic starch synthesis in C3 and C4 species. New Phytol 76:397-407 Inoue Y, Kobayashi Y, Shibata K and Heber U (1978) Synthesis and hydrolysis of ATP by intact chloroplasts under flash illumination and in darkness. Biochim Biophys Acta 505: 142-152 Kaiser WM and Urbach W (1977) The effect of dihydroxyacetone phosphate and 3-phosphoglycerate on O2 evolution and on the levels of ATP, ADP and Pi in isolated intact chloroplasts. Biochim Biophys Acta 459:337 346 Leegood RC, Walker DA and Foyer CH (1985) Regulation of the Benson-Calvin cycle. In: Barber J and Baker NR (eds) Photosynthetic mechanisms and the environment, pp 189 258. Amsterdam: Elsevier Lilley RMcC and Walker DA (1974) An improved spectrophotometric assay for ribulose diphosphate carboxylase. Biochim Biophys Acta 358:226-229 Lilley RMcC, Fitzgerald MP, Rienits KG and Walker DA (1975) Criteria of intactness and the photosynthetic activity of spinach chloroplast preparations. New Phytol 75:1 10 M/ichler F and N6sberger J (1984) Influence of inorganic phosphate on photosynthesis of wheat chloroplasts. II Ribulose bisphosphate carboxylase activity. J Exp Bot 35:488-494 M/ichler F, Schnyder H and N6sberger J (1984) Influence of inorganic phosphate on photosynthesis and assimilate export at 4 ~'C and 25 ~C. J Exp Bot 35:481-487 Robinson SP (1982) 3-phosphoglycerate phosphatase activity in chloroplast preparations as a result of contamination by acid phosphatase. Plant Physiol 70:645-648 Robinson SP (1985) The involvement of stromal ATP in maintaining the pH gradient across the chloroplast envelope in the light. Biochim Biophys Acta 806:187 194 Robinson SP and Walker DA (1979) The control of 3-phosphoglycerate reduction in isolated chloroplasts by the concentrations of ATP, ADP and 3-phosphoglycerate. Biochim Biophys Acta 545:528 536 Robinson SP and Walker DA (1981) Photosynthetic carbon reduction cycle. In: Hatch MD and Boardman NK (eds) The Biochemistry of Plants, Vol 8 pp 193 236. NY: Academic Press Robinson SP, Downton WJS and Millhouse JA (1983) Photosynthesis and ion content of leaves and isolated chloroplasts of salt-stressed spinach. Plant Physiol 73:238-242 Robinson SP and Giersch C (1987) Inorganic phosphate concentration in the stroma of isolated chloroplasts and its influence on photosynthesis. Aust J Plant Physiol, in press Selman BR and Selman-Reimer S (1981) The steady gtate kinetics of photophosphorylation. J Biol Chem 256:1722 1726 Sivak MN and Walker DA (1986) Photosynthesis in vivo can be limited by phosphate supply. New Phytol 102:499-512 Woodrow IE, Ellis JR, Jellings A and Foyer CH (1984) Compartmentation and fluxes of inorganic phosphate in photosynthetic cells. Planta 161:525-530
Regulation of photosynthetic carbon metabolism during phosphate limitation of photosynthesis in isolated spinach chloroplasts C H R I S T O P H G I E R S C H l & S I M O N P. R O B I N S O N CSIRO, Division of Horticultural Research, GPO Box 350, Adelaide, 5001, Australia; Permanent address to which correspondence should be sent: Botanisches lnstitut, Universiti~t Diisseldorf, Universitiitsstrafle 1, D-4000 Diisseldorf, FRG Received 9 June 1987; accepted in revised form 15 July 1987
Key words: carbon metabolism, chloroplasts, phosphate, photosynthesis Abstract. Intact chloroplasts isolated from spinach were illuminated in the absence of inorgan-
ic phosphate (Pi) or with optimum concentrations of Pi added to the reaction medium. In the absence of Pi photosynthesis declined after the first 1-2rnin and was less than 10% of the maximum rate after 5 min. Export from the chloroplast was inhibited, with up to 600 of the ~4Cfixed being retained in the chloroplast, compared to less than 20% in the presence of Pi. Despite the decreased export, chloroplasts depleted of Pi had lower levels of triose phosphate while the percentage of total phosphate in 3-phosphoglycerate was increased. Chloroplast ATP declined during Pi depletion and reached dark levels after 3 4 min in the light without added Pi. At this point, stromal Pi concentration was 0.2 mM, which would be limiting to ATP synthesis. Addition of Pi resulted in a rapid burst of oxygen evolution which was not initially accompanied by net CO2 fixation. There was a large decrease in 3-phosphoglycerate and hexose plus pentose monophosphates in the chloroplast stroma and a lesser decrease in fructose-l,6-bisphosphate. Stromal levels of triose phosphate, ribulose-l,5-bisphosphate and ATP increased after resupply of Pi. There was an increased export of t4C-labelledcompounds into the medium, mostly as triose phosphate. Light activation of both fructose-l,6-bisphosphatase and ribulose-l,5-bisphosphate carboxylase was decreased in the absence of Pi but increased following Pi addition. It is concluded that limitation of Pi supply to isolated chloroplasts reduced stromal Pi to the point where it limits ATP synthesis. The resulting decrease in ATP inhibits reduction of 3-phosphoglycerate to triose phosphate via mass action effects on 3-phosphoglycerate kinase. The lack of Pi in the medium also inhibits export of triose phosphate from the chloroplast via the phosphate transporter. Other sites of inhibitionof photosynthesis during Pi limitation may be located in the regenerative phase of the reductive pentose phosphate pathway. FBP- Fructose- 1,6-bisphosphate, FBPase- Fructose- 1,6-bisphosphatase, MP-Hexose plus pentose monophosphates, PGA-3-phosphoglycerate, Pi inorganic orthophosphate, RuBP- ribulose- 1,5-bisphosphate, RuBPCase- ribulose- 1,5-bisphosphate carboxylase, TP Triose Phosphate Abbreviations:
212 Introduction
Inorganic phosphate (Pi) has a major role in the regulation of photosynthetic carbon metabolism. The chloroplast is predominantly a Pi-consuming, triose phosphate-exporting organelle and the chloroplast pool is unable to support continued CO2 fixation for more than a few minutes. This Pi-dependence of CO: fixation is obvious with isolated chloroplasts, where photosynthesis soon ceases in the absence of added Pi but can be restarted by addition of Pi to the suspending medium (Cockburn et al. 1967). In vivo, triose phosphate is exported from the chloroplast in exchange for the uptake of Pi, via the phosphate transporter, which maintains the chloropolast Pi pool and allows photosynthesis to continue. The triose phosphate is used in the cytosol predominantly for sucrose synthesis, thereby releasing Pi to maintain the cytosolic Pi pool and allow the process to continue. Manipulation of in vivo Pi pools suggests that this recycling of Pi from the cytosol is a major regulatory factor in the photosynthetic process. By feeding sugars such as mannose to leaves, cytosolic Pi is sequestered as non-metabolized sugar phosphates, photosynthesis is inhibited, and starch synthesis is increased (Chen-She et al. 1975, Herold et al. 1976). Photosynthesis is also inhibited in plants grown with an insufficient supply of Pi, but the rate can be significantly increased within a short time by feeding Pi to the leaves from these Pi-deficient plants (Brooks 1986, Dietz and Foyer 1986). Sivak and Walker (1986) reported that photosynthesis can also be significantly increased by feeding Pi to the petiole of leaves from plants grown in complete nutrient media. This suggests that supply of Pi to the chloroplast may be a limiting factor in photosynthesis even under normal conditions. The vacuole of leaf cells normally contains a large pool of phosphate but this does not appear to equilibrate rapidly with the cytosolic Pi pool (Foyer and Spencer 1986, Woodrow et al. 1984). Because the turnover time of the cytosolic Pi pool during photosynthesis is short, any decrease in the rate of Pi release into the cytosol would rapidly decrease the concentration of Pi in the cytosol. This in turn would decrease Pi import into the chloroplast and photosynthesis would become Pi-limited. To identify inhibitory and regulatory steps of CO: fixation under such short-term Pi limitation, we have studied depletion of Pi and the dynamical transition induced by its resupply by measuring oxygen evolution, CO2 fixation, metabolite pools and enzyme activities.
213 Materials and methods
Plant material Spinach seeds (Spinacia oleracea L. cv. Hybrid 102) were germinated in moist peat and after 10 days the seedlings were transferred to hydroponic culture. Four seedlings were placed in each 6-1itre container with the following nutrient solution: 6 m M KNO3, 2 m M MgSO4, l m M KH2PO4, 4 m M Ca(NO3)2, 50/~M FeNa(EDTA)2, 50pM H3BO3, 10pM MnCI2, l pM ZnSO4, 0.5pM CuSO4 and 0.1#M Na2MoO4. Plants were grown in a glasshouse but the day length was restricted to 12 hours by a mechanical shutter and supplementary lighting was provided by fluorescent and incandescent lights to ensure a minimum light intensity of 100/~mol-m-2"s -~ (PAR). The pots were topped up with de-ionized water as required and were aerated continuously.
Chloroplast &olation Intact chloroplasts were isolated from spinach leaves and purified on a two-step Percoll gradient. All procedures were carried out at 0 °C. Leaves (30-70 g) were ground for 3 seconds in a Polytron blender with 200 ml of 330mM sorbitol, 5 m M MgC12, 10mM NaaP2OT, 2 m M isoascorbate, and 0.1% BSA (pH 6.5). The brei was squeezed through two layers of Miracloth containing a layer of cotton wool and the filtrate was centrifuged at 1700 g for 1 minute. The pellets were resuspended in 6 ml of 330 mM sorbitol, 2 mM EDTA, 1 mM MgC12, 1 mM MnC12, 50 mM Hepes-KOH, and 0.2% BSA (pH 7.6) and placed into two centrifuge tubes. Each was underlayered with 4 ml of the same medium plus 40% (v/v) Percoll, then the tubes were again centrifuged at 1700 g for 1 min. Broken chloroplasts formed a band at the top of the Percoll layer, whereas intact chloroplasts were pelleted by this procedure. The supernatants were discarded and the pellets of intact chloroplasts were resuspended in the above medium. The total isolation took 15 to 20min and 02 evolution by the chloroplasts could be measured within 30min of harvesting the leaves. The chloroplasts were greater than 95% intact based on penetration of ferricyanide (Lilley et al. 1975) and exhibited rates of CO2-dependent 0 2 evolution of 150 to 270/~mol.mg Chl ~-h- ~. Chlorophyll was determined by the method of Arnon (1949).
02 evolution CO2-dependent 02 evolution was measured at 20 °C using Hansatech 02 electrodes. Unless stated otherwise, the assay medium contained 330 mM
214 sorbitol, 2 m M EDTA, 1 mM MgC12, 1 mM MnC12, 50mM Hepes-KOH (pH 7.6), 4 mM NaHCO3, 0.4 mM Pi, 1000 units-ml-1 catalase and chloroplasts equivalent to 50 pg Chl.m1-1 . The suspension was illuminated with white light (1500 #mol. m- 2. s- 1 PAR).
14C02fixation Chloroplasts were incubated under the same conditions as for 02 evolution with NaH14CO3 of specific activity 0.19 MBq-mo1-1. Samples were transferred to microcentrifuge tubes containing 50 pl of silicone oil (Wacker type AR 180) on top of 20#1 of 1 N HC10 4. The tubes were centrifuged in a Beckman Microfuge for 20 s and illumination of the samples was continued during this separation procedure. The pellet fraction of each sample was resuspended in 100 #1 H20, and aliquots of the supernatant and pellet fractions were acidified by adding the five-fold volume of 10% (v/v) propionic acid. The samples were dried down and the residues resuspended in 1 ml H20; 10ml scintillation cocktail were added, and the amount of acidstable 14C in each fraction was determined by liquid scintillation counting. The percentage of total 14C incorporated which was retained within the chloroplasts was calculated from these results.
Determination of metabolites Chloroplasts were separated from the medium by silicone oil filtering centrifugation (Heldt 1980) and labelled metabolites determined by HPLC (Giersch 1979, Robinson and Giersch 1987). The following were successively layered in 0.4 ml microcentrifuge tubes: 20 #l 1 N HC104, 75 pl silicone oil (Wacker type AR 180), 200 pl chloroplast suspension. After centrifuging for 20s the supernatant was acidified by addition of 10#l of 12NHCIO4. Illumination was continued during the separation procedure. The supernatant and pellet fractions were removed, centrifuged to remove protein, then neutralized with K2CO 3 and centrifuged again to remove the precipitate of KCIO4. Metabolites were separated by ion exchange chromatography using a 25 cm x 4.6 mm column of Partisil SAX, strongly basic anion exchanger, 10#m particle size. Metabolites were eluted with the following gradient: 5mM KH2PO 4 (Buffer A) for 10min, then a linear gradient from 5 to 200mM KHzPO4 (Buffer B) for 55min followed by isocratic elution in Buffer B for a further 15 min. The pH of Buffer B was adjusted to 2.8 with H 3PO4 and Buffer A was made by diluting Buffer B with water. The flow rate was 0.5 ml.min -1. Radioactivity was detected on-line using a Berthold LB 504 monitor.
215 For 14C-labelled metabolites, chloroplasts were incubated with 2.8 mM NaH14CO3 of specific activity 2.14MBq.mo1-1. To determine 32p-labelled metabolites, chloroplasts (1 mg Chl) were first prelabelled by incubation for 10min at 0°C in l m l of resuspension medium containing 5mM 32pi (1.85 M Bq-mol-l). The suspension was then diluted ten-fold with resuspension medium and layered over 5 ml of resuspension medium containing 40% (v/v) Percoll, centrifuged for 1 min at 1700 g, and the chloroplasts resuspended. Enzyme activities
FBPase activity was measured spectrophotometrically by ading 0.05ml chloroplast suspension to a 1 cm lightpath semimicro cuvette containing in a total volume of 0.45 ml: 10mM MgCI2, 1 mM EDTA, 100mM Tris-HC1 (pH 8.0), 0.2% Triton X-100, 0.3mM NADP, 0.6mM FBP, 2.4units/ml glucose-6-phosphate dehydrogenase and 4units/ml phosphoglucose isomerase. The change in absorbance at 340 mm was recorded for 1 min. The increase in absorption was linear for at least 1 min and proportional to the amount of added chlorophyll. RuBPCase activity was assayed by adding 10/~1of the chloroplast suspension to an Eppendorf tube containing in a total volume of 190#1: 10mM MgCI2, 1 mM EDTA, 10mM KC1, 0.1% Triton X-100, 50mM Hepes, pH (KOH) 8.0, 0.55 mM RuBP and 5.2 mM NaHl4CO3 (0.17 MBq//~mol). CO2 fixation was stopped after 30 s by addition of 100/A 50% propionic acid. The acidified samples were dried down, and the acid stable lable was counted in a liquid scintillation counter after redissolving the dried residue in 100#1 H2 O and adding 1 ml scintillation cocktail. RuBPCase activity was linear for at least 60 s. Parallel experiments where the RuBPCase activity was measured by following the rate of PGA formation enzymically (Lilley and Walker 1979), gave essentially the same results. A TP determination
Chloroplasts were incubated as for 02 evolution measurements and samples (80 #1) were injected into 40 #1 of 3 N H C 1 0 4 with rapid stirring. Illumination of the samples was continued during the killing procedure. The acidified samples were centrifuged to remove protein, neutralized with K2CO 3 and then centrifuged again to remove the precipitate of K C 1 0 4. ATP was determined in a 10 pl aliquot by the luciferin-luciferase bioluminescence method using Boehringer CLS reagent in an LKB 1250 luminometer.
216 Stromal pH The pH of the chloroplast stroma and intrathylakoid spaces were determined as described by Robinson (1985). Chemicals Silicone oil was a gift from Wacker (Australia), RuBP was kindly provided by J. Andrews, Res. School of Biological Sciences, Canberra. Radioisotopes were obtained from Amersham, UK, and enzyme solutions from Boehringer, Mannheim, West Germany. All other biochemicals were obtained from Sigma Chemical Co., St. Louis, USA.
Results
Oxygen evolution and CO 2fixation Intact isolated chloroplasts illuminated in a medium with optimal concentrations of Pi (0.2-0.5 mM) exhibited high rates of CO2-dependent 02 evolution (150-270 #mol. mg Chl- ~. h - ~). Following the normal induction phase, a linear rate of 02 evolution was maintained for several minutes. If Pi was omitted from the reaction medium, the induction phase was shorter and a lower maximum rate was achieved, after which the rate declined due to depletion of Pi. This decline in O: evolution had the characteristic kinetics shown in Fig. 1, which is typical of a number of experiments with purified intact chloroplasts. The maximum rate was achieved after 1-2 min, then there was a sharp transition to a lower rate which was maintained for a further 1-2 min before 02 evolution again declined to a steady-state rate of 5-20 ~mol" mg Chl- ~"h-~ (Fig. 1). During the second phase, there was often a secondary increase in 02 evolution which is evident from the plot of rate of 02 evolution against time in the upper panel of Fig. 1. The final rate was maintained for several minutes and probably reflects the rate of starch synthesis, which does not result in net consumption of Pi. Addition of Pi at this point caused an immediate burst of 02 evolution, lasting 15-30 s, which was often followed by a transitory decline in the rate of O2 evolution before a new steady-state rate was achieved 0.5-1 min after adding Pi (Fig. 1). Measurement of CO2 fixation (incorporation of ~4C) indicated that it paralleled 02 evolution during the depletion of Pi, whereas the burst of oxygen evolution immediately following the addition of Pi (Fig. 1) was not accompanied by any net fixation of ~4CO2, which only started 0.5-1 min after
217 J~
7 ~40~
~20~ IO0'! ~o~ ~0) ~0~ ~oi ,
•
oi
4
5
6
7
0,35~ 0.30 /
0.25
/11~
~ 0
1
2
3 Time
4 5 In L~h~( nlln )
0
7
Fig. 1. Oxygen evolution by intact isolated chloroplasts illuminated in a reaction medium without added Pi. Oxygen electrode trace is shown in the lower panel with numbers indicating the rate of oxygen evolution in ~mol O2 • mg Chl-~ • h ~. The rate of oxygen evolution is plotted in the upper panel. Chloroplasts equivalent to 50 ~tg Chl were added in a total volume o f 1 ml. After 5 rain illumination, Pi was added to a final concentration of 0.4 raM. The dotted line in the lower panel shows oxygen evolution in a parallel experiment where no addition of Pi was made.
Pi was added (data not shown). Export of ]4C-labelled products into the medium was inhibited in the absence of added Pi. After 5 min in the light, nearly 60% of the ~4C fixed was retained within the chloroplasts whereas with optimum Pi in the reaction medium less than 20% of the ]4C incorporated was in the chloroplast by the time the maximum rate of photosynthesis was achieved (Fig. 2). Upon addition of Pi to Pi-depleted chloroplasts there was an immediate export of ~4C into the medium. The amount of 14C in the chloroplast decreased by 40% and there was an equivalent increase in the amount of ~4C in the medium, resulting in a decrease in the percentage of ]4C in the chloroplast (Fig. 2). This was followed by a small but reproducible increase in the percentage of ~4C in the chloroplast then a gradual decline over the next 2-4min as the amount of incorporated ~4C found in the medium increased (Fig. 2).
218 80
60
0
40
6
o
,~
'
~
'
,'o
T i m e in L i g h t ( rain )
Fig. 2. Export of ~4C-label from chloroplasts illuminated in a medium with 0.4 mM Pi added ( + Pi) and in the absence of added Pi ( - Pi). After 5 min illumination, Pi was also added to the - Pi medium to give a final concentration of 0.4 mM.
Metabolite pools To determine metabolites, chloroplasts were incubated with 32pi to label the endogenous phosphate pools then washed and illuminated in a medium without Pi until photosynthesis had virtually ceased due to Pi depletion. For comparison, chloroplasts prelabelled with 32p were also illuminated in a medium containing 0.4 mM 32pi until the maximum rate of photosynthesis was achieved (Table 1). In the absence of Pi in the reaction medium, the total Table 1.32p-labelled metabolites in chloroplasts illuminated in the absence and presence of Pi in the reaction medium. The chloroplasts were prelabelled with 32pi, washed, then incubated in the absence or presence of 0.4mM 32Pi of the same specific activity. The chlorophyll concentration was 100pg.ml t. Samples were taken after 5min ( - P i ) or 3min ( + P i ) illumination. Total 32p in the chloroplast fraction was 307 ( - Pi) and 560 ( + Pi) natom, mg Chl- ~. Rates of oxygen evolution were 5 ( - Pi) and 163 ( + Pi)/lmol. mg Chl- ~-h - t. -Pi
+Pi
Metabolite
% of total label
nmol- mg Chl- ~
% of total label
nmol" mg Chl-
MP TP Pi PGA FBP RuBP
28.9 1.4 1.5 47.0 2.7 8.7
89.0 4.4 4.5 144.0 4.2 13.4
28.9 5.5 8.6 28.6 2.8 14.7
162.0 31.0 48.0 159.0 7.9 42.0
219 label in the chloroplast was 45% lower and there was a redistribution of label between individual metabolites. There was a marked decrease in the amount of Pi and triose phosphate (TP) and also decreases in the amounts ofhexose plus pentose monophosphates (MP), FBP and RuBP. The amount of PGA was not significantly decreased and on a percentage basis there was a significant increase in the proportion of 32p in PGA. The ratio PGA/TP increased from 5.1 in chloroplasts with optimal Pi to 32.7 in Pi-depleted chloroplasts. Similar results were obtained when metabolite pools were determined by labelling with 14CO2 (data not shown). Figure 3 shows changes in 14C-label in various metabolites in the chloroplast following addition of Pi to chloroplasts after 5 min illumination without added Pi. Upon addition of Pi, label in PGA and MP was dramatically reduced and that in FBP to a lesser extent. There was a subsequent rise in label in these compounds and 3 min after adding Pi the new steady-state levels were, except for PGA, similar to those found before Pi addition. In contrast, label in TP and RuBP gradually increased after Pi addition then subsequently declined again. The new steady-state levels of TP and RuBP were higher than before Pi addition while PGA was lower. The drastic drop
MP
Pi A d d e d
400
A
300
Cl
200
T 0 100
V
30
q'
B
20
10 Pi
A~ded
3 Time
( rain )
Fig. 3. Metabolites in the stroma of chloroplasts following addition of Pi to the medium. Chloroplasts were illuminated in a medium with 2.8mM NaHJ4CO3 without added Pi for 5 min. Then Pi was added to give a final concentration of 0.2 mM. At the times indicated, chloroplasts were separated from the medium by centrifugation through silicone oil and ~4C-label in each metabolite was determined by HPLC.
220 in PGA upon Pi addition appeared to be mainly due to Pi-induced export of label into the medium. However, label in PGA in the medium only increased by 20 natom 14C-mgChl-] whereas the drop in PGA in the chloroplast was ten-fold higher. Moreover, TP in the chloroplast was not lowered upon Pi addition (Fig. 3) even though the amount of TP in the medium increased from 150 natom ]4C.mg Chl-] at 4.5 min to 590 natom 14C.mg Ch1-1 at 5.5 min. The observation that, despite the large export of TP upon Pi addition the stromal pool of this metabolite actually increased, suggests that a major limitation of carbon flux during Pi limitation was the reduction of PGA to TP. The data suggest that photosynthesis in Pidepleted chloroplasts was limited by the phosphoglycerate kinase or glyceraldehyde phosphate dehydrogenase reactions rather than by low export rates due to a lack of Pi in the medium. This view is supported by the observed burst in 02 evolution in the absence of net 14CO2fixation when Pi was added which indicates a dramatic increase in the rate of PGA reduction. Activities of FBPase and RuBPCase
The metabolite data of Fig. 3 suggests the existence of another limitation of carbon flux under Pi deficiency, residing in the regenerative phase of the reductive pentose phosphate cycle. Release of this inhibition would explain the observed decline in pentose plus hexose monophosphates and increase in RuBP following addition of Pi. FBPase activity in darkened chloroplasts was low but increased upon illumination of chloroplasts both in the absence and presence of Pi (Fig. 4). In the presence of Pi, full activity was obtained ÷Pb
150
O
T" U
o
,oo
I~
L Pi Added
0 T i m e in L i g h t ( rain )
Fig. 4. FBPase activity of chloroplasts incubated in the absence ( - Pi) and presence ( + Pi) of 0.4mM Pi. After 4.75min illumination, Pi was added to the - P i medium to give a final concentration of 0.4 mM. Rates of oxygen evolution in/~mol'mg Chl- ~.h-~ at various times are indicated by the numbers in boxes.
221 after 3-4 min in the light which coincided with achievement of the maximum rate of 02 evolution. In the absence of Pi, activation of FBPase was initially more rapid but ceased after 1 min. After 5 min illumination, the activity of FBPase in Pi-depleted chloroplasts was only about half of that in chloroplasts in the presence of Pi. Upon addition of Pi there was a gradual increase in FBPase activity and within 3 min it reached the same value as chloroplasts illuminated in the presence of Pi from the outset. It should be noted that the maximum rate of 02 evolution was achieved within 0.5-1 min of adding Pi (Fig. 1) but at this point FBPase activity had only increased by 30% (Fig: 4). The activity of RuBPCase was also measured in chloroplasts illuminated in the absence and presence of added Pi (Fig. 5). In the presence of Pi, RuBPCase increased 1.3-2.2-fold upon illumination, with maximum activity achieved after 2-3 min in the light when the maximum rate of 02 evolution was obtained. In the absence of Pi, RuBPCase was often slightly higher in the dark but activation in the light was greatly reduced. In six separate experiments the activity only increased 1.1-1.2-fold upon illumination in the absence of Pi. Addition of Pi partially restored the light activation of RuBPCase (Fig. 5). The extent of the increase was variable but full activation (to the same level as the + P i control) was not observed. As with FBPase, achievement of the maximum rate of 02 evolution normally preceded full activation of RuBPCase after addition of Pi. //-
.....
+P~ / / la=
400
7-
/ /
'
la=
/ I
7" ÷Pi
o
0
0
0
150
© ~
o o
100
f
//
/
....
200
~
/-,(
100
/
SO ,
/'
/J
x 0
/
"~
/
'~-~",
r/
i" i i I
0
/.-" O
I
21
I
--I
___t
I
4 T i m e in L i g h t ( rain
I
6
r
I
I
0
8
)
Fig. 5. Rate of oxygen evolution and RuBP carboxylase activity (Rubisco) of chloroplasts incubated in the absence ( - Pi) and presence ( + Pi) of 0.4 mM Pi. After 4.75 min illumination, Pi was added to the - P i medium to give a final concentration of 0.4 mM.
222 A TP levels
In both the presence and absence of Pi chloroplasts contained low levels of ATP in the dark (5-8nmol.mgChl -~) but after 15s in the light ATP increased to 30-35nmol'mgCh1-1 (Fig. 6). In the presence of Pi, ATP declined after about 1 min in the light, reaching about 20 nmol.mg Chlafter 3 min when the maximum rate of 02 evolution occurred, then decreasing more slowly over the next 3-4 min while high rates of 02 evolution were maintained. In the absence of Pi, ATP declined much more rapidly and reached dark levels after 3-4min in the light, when photosynthesis had virtually ceased. Dark ATP levels of 5-8 nmol.mg Ch1-1 are frequently observed in isolated chloroplasts and this is considered to be a thermodynamically unavailable pool, possibly bound to the chloroplast coupling factor (Inoue et al. 1978, Robinson 1985). Thus the low level of ATP in Pi-deficient chloroplasts indicates virtually complete exhaustion of the available ATP pool. Addition of Pi led to a rapid increase in ATP, resulting in levels above those in chloroplasts incubated in the presence of Pi. The level of ATP subsequently declined slightly over the next 2-3 min (Fig. 6). Similar effects of Pi depletion on stromal ATP were observed by Kaiser and Urbach (1977) and M/ichler et al. (1984). Stromal pH
The pH of the chloroplast stroma is involved in the regulation of photosynthetic carbon metabolism both by its effect on the activity and activation
÷Pi
30
u
o
E
,o
4 v
v
f
Pi A d d e d o
;
,
,
,
/
,
,
,
6
, 8
T i m e in L i g h t ( rain )
Fig. 6. ATP levels in chloroplasts incubated in the absence ( - P i ) and presence ( + Pi) of 0 . 2 m M P i . After 5min illumination, Pi was added to the - P i concentration of 0.2 mM.
medium to give a final
223 state of enzymes and via effects on transport across the chloroplast envelope (Robinson and Walker 1981). As shown in Table 2 the pH of the stroma was not decreased in the absence of Pi. There was a slight decrease in the pH of the intrathylakoid space resulting in a small increase in the calculated pH gradient across the thylakoid membrane, possibly caused by the low rate of photophosphorylation under Pi-limited conditions.
Discussion
In the absence of added Pi, oxygen evolution by isolated chloroplasts is only maintained for a few minutes before declining because of Pi depletion. The kinetics illustrated in Fig. 1 were characteristic of purified intact chloroplasts. Isolated chloroplasts normally contain 300-500 nmol Pi. mg Chl- 1 (Kaiser and Urbach 1977, Robinson et al. 1983, Robinson and Giersch 1987). The oxygen evolution which occurs before endogenous Pi is depleted reflects the utilisation of this Pi for synthesis of the sugar phosphate intermediates of the reductive pentose phosphate cycle. If all the Pi is converted to triose phosphate this would allow 0.9-1.5 pmol 02-mg Chl-1 to be evolved before photosynthesis ceased. The first decline in 02 evolution after 1-2 mins in the light occurred after 1-2 pmol 02 • mg Chl- 1had been evolved (Fig. 1) and probably reflects utilisation of the endogenous Pi pool. The subsequent 02 evolution would depend on recycling of this phosphate pool (via starch synthesis) or release of Pi from other metabolites not normally a part of the reductive pentose phosphate cycle (e.g. pyrophosphate, glucose phosphates etc). Recycling of Pi could also result from phosphatase activity, which can be present as a contaminant in chloroplast preparations (Robinson 1982). The total phosphate pool of the chloroplasts was lower in the absence of Pi in the medium (Table 1) suggesting that there was some net leakage of phosphates out of the chloroplast. Export of metabolites was severely Table 2. Photosynthesis and pH of the chloroplast stroma and intrathylakoid spaces in chloroplasts after 5 rain illumination in the absence ( - P i ) and presence ( + Pi) of 0.4 mM Pi. The reaction medium was pH 7.6.
Rate O: evolution (#mol'mg Chl- ~"h- z) Stromal pH Intrathylakoid pH ApH envelope membrane ApH thylakoid membrane
- Pi
+ Pi
4.0 7.85 5.38 0.25 2.48
204.0 7.86 5.48 0.26 2.38
224 inhibited in the absence of exogenous Pi (Fig. 2). This is to be expected since export of triose phosphate requires uptake of Pi for counter-exchange via the phosphate transporter. Despite the decreased export of triose phosphate into the medium, there was not an accumulation of triose phosphate in chloroplasts in the absence of Pi. On the contrary, triose phosphate was lower in the absence of Pi whereas the percentage of label in PGA was increased (Table 1). This indicates that lack of Pi inhibited reduction of PGA to triose phosphate as well as export of triose phosphate to the medium. Similar changes in PGA and triose phosphate pools were reported by Heldt et al. (1978) and M/ichler and N6sberger (1984) for chloroplasts in the presence of limiting amounts of Pi. The equilibrium position of phosphoglycerate kinase lies in the direction of PGA formation and for the phosphorylation of PGA, as required for photosynthesis, a high ratio of substrates (PGA and ATP) to products (glycerate-1,3-bisphosphate and ADP) is necessary (Robinson and Walker 1981). In chloroplasts, phosphorylation of PGA and reduction to triose phosphate is very sensitive to changes in the concentrations of ATP and ADP (Robinson and Walker 1979). A decrease in stromal Pi would limit ATP synthesis which would decrease ATP, increase ADP and inhibit phosphoglycerate kinase. The decrease in ATP (Fig. 6) paralleled the decrease in photosynthesis (Figs. 1, 5) and after 3-4min illumination in the absence of Pi the chloroplast ATP pool was similar to that in the dark. At this point, the chloroplasts contained less than 5 n m o l P i . m g C h l -~ (Table 1). Assuming a chloroplast volume of 25 #1- mg Chl- ~(Heldt 1980) the stromal Pi concentration would be 0.2 mM, which is considerably below the Km of the chloroplast ATP synthase (Selman and Selman-Reimer 1981). Low Pi may also cause inhibition of PGA reduction via effects on electron transport of the NADP system: if glycerate-l,3-bisphosphate falls low due to inhibition of the phosphoglycerate kinase under low Pi, NADPH will pile up, and electron transport will slow down because of the lack of acceptor. This view is supported by the observation that electron transport is not limited by the high proton gradient built up under nonphosphorylating ( - P i ) conditions in illuminated chloroplasts (Furbank et al. 1987). The decrease in the rate of electron transport by lack of acceptor will reinforce the inhibition of photosynthesis caused by the effect of ATP on phosphoglycerate kinase activity. Thus it is clear that limitation of Pi supply to the chloroplasts was perceived as a decrease in stromal Pi, resulting in lowered ATP which inhibited PGA reduction and probably electron transport. This was a major site of inhibition of the reductive pentose phosphate cycle as indicated by the large increase in ATP and decrease in PGA upon addition of Pi (Figs. 3, 6).
225 A second site of inhibition in the regenerative phase of the reductive pentose phosphate cycle was indicated from the changes in metabolite pools following addition of Pi (Fig. 3). One third of the ATP consumed during CO2 fixation is used in the conversion of ribulose-5-phosphate to RuBP by phosphoribulokinase. This enzyme has a greater affinity for ATP than phosphoglycerate kinase but is inhibited by ADP, PGA and a number of other stromal metabolites (Robinson and Walker 1981, Flfigge et al. 1982, Gardemann et al. 1983). The decrease in MP upon addition of Pi (Fig. 3) probably reflects increased phosphoribulokinase activity as a result of increased stromal ATP and decreased PGA. Light activation of RuBPCase was inhibited in the absence of Pi (Fig. 5) as has been reported previously (Heldt et al. 1978, M~ichler and N6sberger 1984). It should be noted, however, that the activity was still well in excess of that required to maintain the observed rates of photosynthesis. Furthermore, the increase in rate of photosynthesis following Pi addition preceded the increase in RuBPCase. The increase in RuBPCase was less than 50% despite a 20-fold increase in the rate of 02 evolution. This suggests that lack of activation of this enzyme was not a limiting factor to photosynthesis in Pi-depleted chloroplasts. Light activation of FBPase was also decreased in the absence of Pi (Fig. 4). As with RuBPCase, extractable FBPase activity was well in excess of that required for the observed rates of photosynthesis (a rate of 02 evolution of 100/~mol • mg Chl- J" h- J requires FBPase activity of 33#mol-mgChl -~ "h -J) and the increase in photosynthesis following Pi addition preceded the activation of FBPase (Fig. 4). There was not an increase in FBP in the absence of Pi (Table 1) but addition of Pi resulted in a transient decrease in FBP (Fig. 3) suggesting a release of limitation by FBPase. Activation of this enzyme requires reduction by thioredoxin and is dependent on the pH and concentration of Mg 2÷ and FBP (Leegood et al. 1985). There was no decrease in stromal pH in the absence of Pi (Table 2) but there was some decrease in the concentration of FBP (Table 1). In the absence of Pi, electron transport is likely to be greatly decreased, as suggested by the low rates of oxygen evolution. It seems likely that the inhibition of light activation of FBPase in the absence of Pi and its increase following Pi addition reflect changes in the rate of electron transport which would alter the redox state of the enzyme. The results presented here suggest that limitation of Pi supply to the chloroplast leads to a decrease in stromal Pi concentration to the point where ATP synthesis becomes Pi-limited. One major consequence of the decrease of ATP is an inhibition of PGA reduction via mass action effects on phosphoglycerate kinase. Because of the high turnover rate of Pi and
226 phosphorylated metabolites in the reductive pentose phosphate cycle relative to the chloroplast pool sizes, carbon flux trough the cycle increases rapidly following resupply of Pi. Thus the chloroplast is very responsive to changes in the rate of supply of Pi from the cytosol which is consistent with the major role recycling of Pi has in the regulation of photosynthesis.
Acknowledgements We wish to thank Marcia McFie for her excellent technical assistance. Financial support for C. Giersch was provided by the Deutsche Forschungsgemeinschaft and CSIRO.
References Arnon DI (1949) Copper enzymes in isolated chloroplasts. Polyphenol oxidase in Beta vulgaris. Plant Physiol 24:1-15 Brooks A (1986) Effects of phosphorus nutrition on ribulose-l,5-bisphosphate carboxylase activation, photosynthetic quantum yield and amounts of some Calvin cycle metabolites in spinach leaves. Aust J Plant Physiol 13:221-237 Chen-She SH, Lewis DH and Walker DA (1975) Stimulation of photosynthetic starch formation by sequestration of cytoplasmic orthophosphate. New Phytol 74:373-392 Cockburn W, Baldry CW and Walker DA (1967) Oxygen evolution by isolated chloroplasts with carbon dioxide as the hydrogen acceptor. A requirement for orthophosphate or pyrophosphate. Biochim Biophys Acta 131:594-596 Dietz KJ and Foyer C (1986) The relationship between phosphate status and photosynthesis in leaves. Reversibility of the effects of phosphate deficiency on photosynthesis. Planta 167: 376-381 Fl/igge UI, Stitt M, Freisl M and Heldt HW (1982) On the participation of phosphoribulokinase in the light regulation of CO 2 fixation. Plant Physiol 69:263-267 Foyer C and Spencer C (1986) The relationship between phosphate status and photosynthesis in leaves. Effects of intracellular orthophosphate distribution, photosynthesis and assimilate partitioning. Planta 167:369-375 Furbank RT, Foyer CH and Walker DA (1987) The site of limitation of photosynthesis by inorganic phosphate in isolated spinach chloroplasts. Biochim Biophys Acta, in press Gardemann A, Stitt M and Heldt HW (1983) Control of CO 2 fixation. Regulation of spinach ribulose-5-phosphate kinase by stromal metabolite levels. Biochim Biophys Acta 722:51-60 Giersch C (1979) Quantitative high performance liquid chromatographic analysis of ~4Clabelled photosynthetic intermediates in isolated intact chloroplasts. J Chromatography 172:153 161 Heldt HW (1980) Measurement of metabolite movement across the envelope and of the pH in the stroma and the thylakoid space in intact chloroplasts. Methods in Enzymolgy 69: 604-613 Heldt HW, Chon CJ and Lorimer GH (1978) Phosphate requirement for the light activation of ribulose-l,5-bisphosphate carboxylase in intact spinach chloroplasts. FEBS Lett 92: 234-240
227 Herold A, Lewis DH and Walker DA (1976) Sequestration of cytoplasmic orthophosphate by mannose and its differential effect on photosynthetic starch synthesis in C3 and C4 species. New Phytol 76:397-407 Inoue Y, Kobayashi Y, Shibata K and Heber U (1978) Synthesis and hydrolysis of ATP by intact chloroplasts under flash illumination and in darkness. Biochim Biophys Acta 505: 142-152 Kaiser WM and Urbach W (1977) The effect of dihydroxyacetone phosphate and 3-phosphoglycerate on O2 evolution and on the levels of ATP, ADP and Pi in isolated intact chloroplasts. Biochim Biophys Acta 459:337 346 Leegood RC, Walker DA and Foyer CH (1985) Regulation of the Benson-Calvin cycle. In: Barber J and Baker NR (eds) Photosynthetic mechanisms and the environment, pp 189 258. Amsterdam: Elsevier Lilley RMcC and Walker DA (1974) An improved spectrophotometric assay for ribulose diphosphate carboxylase. Biochim Biophys Acta 358:226-229 Lilley RMcC, Fitzgerald MP, Rienits KG and Walker DA (1975) Criteria of intactness and the photosynthetic activity of spinach chloroplast preparations. New Phytol 75:1 10 M/ichler F and N6sberger J (1984) Influence of inorganic phosphate on photosynthesis of wheat chloroplasts. II Ribulose bisphosphate carboxylase activity. J Exp Bot 35:488-494 M/ichler F, Schnyder H and N6sberger J (1984) Influence of inorganic phosphate on photosynthesis and assimilate export at 4 ~'C and 25 ~C. J Exp Bot 35:481-487 Robinson SP (1982) 3-phosphoglycerate phosphatase activity in chloroplast preparations as a result of contamination by acid phosphatase. Plant Physiol 70:645-648 Robinson SP (1985) The involvement of stromal ATP in maintaining the pH gradient across the chloroplast envelope in the light. Biochim Biophys Acta 806:187 194 Robinson SP and Walker DA (1979) The control of 3-phosphoglycerate reduction in isolated chloroplasts by the concentrations of ATP, ADP and 3-phosphoglycerate. Biochim Biophys Acta 545:528 536 Robinson SP and Walker DA (1981) Photosynthetic carbon reduction cycle. In: Hatch MD and Boardman NK (eds) The Biochemistry of Plants, Vol 8 pp 193 236. NY: Academic Press Robinson SP, Downton WJS and Millhouse JA (1983) Photosynthesis and ion content of leaves and isolated chloroplasts of salt-stressed spinach. Plant Physiol 73:238-242 Robinson SP and Giersch C (1987) Inorganic phosphate concentration in the stroma of isolated chloroplasts and its influence on photosynthesis. Aust J Plant Physiol, in press Selman BR and Selman-Reimer S (1981) The steady gtate kinetics of photophosphorylation. J Biol Chem 256:1722 1726 Sivak MN and Walker DA (1986) Photosynthesis in vivo can be limited by phosphate supply. New Phytol 102:499-512 Woodrow IE, Ellis JR, Jellings A and Foyer CH (1984) Compartmentation and fluxes of inorganic phosphate in photosynthetic cells. Planta 161:525-530
