Transport and phosphorylation of choline in higher plant cells. Phosphorus-31 nuclear magnetic resonance studies

When sycamore cells were suspended in basal medium containing choline, the latter was taken up by the cells very rapidly. A facilitated diffusion system appertained at low concentrations of choline and exhibited Michaelis-Menten kinetics. At higher choline concentrations simple diffusion appeared to be the principal mode of uptake. Addition of choline to the perfusate of compressed sycamore cells monitored by 31P NMR spectroscopy resulted in a dramatic accumulation of P-choline in the cytoplasmic compartment containing choline kinase and not in the vacuole. The total accumulation of P-choline over a 10-h period exhibited Michaelis-Menten kinetics. During this period, in the absence of Pi in the perfusion medium there was a marked depletion of glucose-6-P, and the cytoplasmic Pi resonance disappeared almost completely. When a threshold of cytoplasmic Pi was attained, the phosphorylation of choline was sustained by the continuous release of Pi from the vacuole although at a much lower rate. However, when 100 microM inorganic phosphate was present in the perfusion medium, externally added Pi was preferentially used to sustain P-choline synthesis. It is clear, therefore, that cytosolic choline kinase associated with a carrier-mediated transport system for choline uptake appeared as effective systems for continuously trapping cytoplasmic Pi including vacuolar Pi entering the cytoplasm.     

Control of phosphate transport across the plasma membrane of Chara corallina

This paper examines the control of phosphate uptake into Chara corallina. Influxes of inorganic phosphate (Pi) into isolated single internodal cells were measured with ³²Pi. Pretreatment of cells without Pi for up to 10 d increased Pi influx. However, during this starvation the concentrations of Pi in both the cytoplasm and the vacuole remained quite constant. When cells were pre-treated with 0.1 mM Pi, the subsequent influx of Pi was low. Under these conditions the Pi concentrations in the cytoplasm was almost the same as that of Pi-starved cells, but vacuolar Pi increased with time. Transfer of cells from medium containing 0.1 mM Pi to Pi-free medium induced an increase of Pi influx within 3 d irrespective of the concentration of Pi in the vacuole.During Pi starvation, neither the membrane potential nor the cytoplasmic pH changed. Manipulation of the cytoplasmic pH by weak acids or ammonium decreased the Pi influx slightly.Pi efflux was also measured, using cells loaded with ³²Pi. Addition of a low concentration of Pi in the rinsing medium rapidly and temporarily induced an increase in the efflux.The results show that Pi influx is controlled by factors other than simple feedback from cytoplasmic or vacuolar Pi concentrations or thermodynamic driving forces for H⁺-coupled Pi uptake. It is suggested that uptake of Pi is controlled via the concentration of Pi in the external medium through induction or repression of two types of plasma membrane Pi transporters.Key words: Chara corallina, membrane transport, phosphate influx, phosphate starvation      

Biochemical changes during sucrose deprivation in higher plant cells. Phosphorus-31 nuclear magnetic resonance studies

An experimental arrangement was described that enables nuclear magnetic resonance spectra of compressed plant cells to be recorded while circulating a medium through the sample. The system provided a convenient arrangement for monitoring by 31P NMR the behavior of plant cells over a long period of time under different conditions such as sucrose starvation. Perfusion of compressed sycamore cells with sucrose-free culture medium triggered a progressive decrease in the glucose 6-P and uridine-5'-diphosphate-alpha-D-glucose resonances over 30 h. When almost all the intracellular carbohydrate pool had disappeared the nucleotide triphosphate resonances decline progressively. These changes were accompanied by a Pi accumulation in the vacuole and a phosphorylcholine (P-choline) accumulation in the cytoplasm. The very long lag phase observed for ATP and P-choline evolution was comparable with that observed for the progressive intracellular digestion of cytoplasmic constituents (Journet, E., Bligny, R. and Douce, R. (1986) J. Biol. Chem. 261, 3193-3199). Addition of sucrose in the circulating system after a long period of sucrose starvation led to a disappearance of the cytoplasmic Pi resonance and a marked increase in that of glucose 6-P. Under these conditions the vacuolar Pi pool did not fluctuate to buffer the Pi in the cytoplasm. The results suggest that Pi which has been sequestered in the vacuole during the course of sucrose starvation is not restored to the cytoplasm for rapid metabolic processes. Furthermore, the presence of P-choline in plant cells in large excess should be considered as a good marker of membrane utilization after a long period of sucrose starvation and is very likely related to stress.     

Effect of sucrose starvation on sycamore (Acer pseudoplatanus) cell carbohydrate and Pi status.

The mobilization of stored carbohydrates during sucrose starvation was studied with sycamore (Acer pseudoplatanus) cells. When sucrose was omitted from the nutrient medium, the intracellular sucrose pool decreased rapidly during the first hours of the experiment, whereas the starch content remained practically unchanged. After 10h of sucrose starvation, starch hydrolysis replaced sucrose breakdown. From this moment, the phosphate-ester pool and respiration rate decreased with time. Conversely, the intracellular Pi concentration increased. 31P n.m.r. of intact sycamore cells indicated that, under these conditions, most of the Pi accumulated in the vacuole. These results strongly suggest that starch breakdown, in contrast with sucrose hydrolysis, is not rapid enough to maintain a high cellular metabolism.     

31P NMR studies of spinach leaves and their chloroplasts

An experimental arrangement is described which enables high quality 31P NMR spectra of compressed spinach leaf pieces to be continuously recorded in which all the resonances observed (cytoplasmic and vacuolar Pi, glycerate-3-P, nucleotides) were sharp and well resolved. 31P NMR spectra obtained from intact chloroplasts showed a distinct peak of stromal Pi. An upfield shift of the stromal Pi resonance was associated with a decrease in the external Pi and vice versa. Nucleotides were largely invisible to NMR in intact chloroplasts, whereas the same nucleotides reappeared in a typical 31P NMR spectrum of an acid extract of intact chloroplasts. Perfusion of compressed spinach leaf pieces with a medium containing Pi triggered a dramatic increase in the vacuolar Pi over 12 h. Addition of choline to the Pi-free perfusate of compressed leaf pieces resulted in a steady accumulation of phosphorylcholine in the cytoplasmic compartment at the expense of cytoplasmic Pi. When a threshold of cytoplasmic Pi concentration was attained, Pi was drawn from the vacuole to sustain choline phosphorylation. In spinach leaves, the vacuole represents a potentially large Pi reservoir, and cycling of Pi through vacuolar influx (energy dependent) and efflux pathways is an efficient system that may provide for control over the cytosolic-free Pi and phosphorylated intermediate concentrations. 31P NMR spectra of neutralized perchloric acid extracts of spinach leaves showed well defined multipeak resonances (quadruplet) of intracellular phytate. The question of cytosolic Pi concentration in green cells is discussed.     

Phosphorus compartmentation in Pinus serotina Michx. (pond pine); observations from in vivo nuclear magnetic resonance spectroscopy

³¹P nuclear magnetic resonance (NMR) spectroscopy was used to estimate the amount of inorganic phosphate (Pi) present in the cytoplasm and vacuole of root tips and subapical root segments of pond pine ( Pinus serotina Michx.). In root tips of seedlings grown with 100 mmol m^–3P (HP) the cytoplasmic Pi content, on a root volume basis, was ≈ 1·5 μ mol cm^–3 and the vacuolar Pi content, on a root volume basis, was ≈ 3·4 μ mol cm^–3. In root tips from Pi starved seedlings the cytoplasmic Pi content, on a root volume basis, was ≈ 0·75 μ mol cm^–3; vacuolar Pi was too low to be reliably estimated. Similar results were obtained with subapical root segments; the Pi concentration in the cytoplasm was maintained at around 2 mol m^–3 while that in the vacuole varied with Pi supply. This work demonstrates for the first time that quantitative measurements of the subcellular compartmentation of Pi can be made in young tissues of a woody species. The results indicate that cytoplasmic Pi levels are maintained across a range of external Pi supplies probably by withdrawing Pi stored in the vacuole.     

Is the Cytosolic Pi Concentration a Limiting Factor for Plant Cell Respiration?

The substrate-dependent O₂ uptake by sycamore (Acer pseudoplatanus L.) cell mitochondria in the presence of ADP and limiting Pi concentrations has been measured. The Pi concentration for half-maximum O₂ uptake rate was found to be in the range 20 to 50 micromolar for all the substrates tested. ³¹P NMR of intact sycamore cells indicated that the Pi concentration in the cytoplasm was in the range 5 to 6 millimolar, approximately 100-fold higher than the Pi concentration required for maximum O₂ uptake rates by isolated mitochondria. When sycamore cells were transferred to a culture medium devoid of Pi, the cytoplasmic Pi concentration decreased from 6 to less than 3 millimolar, but the intact cell respiration remained practically constant for at least 4 days. These results strongly suggest that, in vivo, the respiration rate of sycamore cells is not limited by the quantity of Pi supplied to the mitochondria.     

Exocellular and intracellular accumulation of lead in Pseudomonas fluorescens ATCC 13525 is mediated by the phosphate content of the growth medium

Pseudomonas fluorescens appears to elicit disparate lead detoxification mechanisms in phosphate-rich and phosphate-deficient media. When grown in the presence of 0.1 mM Pb2+ complexed to citrate, the sole source of carbon, only a slight diminution in cellular yield was observed in the former medium. However, in a phosphate-deficient milieu, lead imposed approximately a 30% reduction in bacterial multiplication. At stationary phase of growth, 72% of the metal was found in the bacterial cells from the phosphate-deficient medium, while that from phosphate-rich broth contained only 12.5%. The latter medium was characterized by an insoluble pellet that accounted for 73.5% of the lead. Although no citrate was detected in the phosphate-rich media after 40 h of incubation, only 72% of citrate was consumed even after 70 h of growth in the phosphate-deficient cultures. The inclusion of lead did not appear to enhance the production of either extracellular proteins or carbohydrates.     

Subcellular Distribution of Inorganic Phosphate, and Levels of Nucleoside Triphosphate, in Mature Maize Roots at Low External Phosphate Concentrations: Measurements with 31P-NMR

Maize plants were grown in nutrient solution without phosphate,or in which inorganic phosphate (Pi) was maintained at nearlyconstant concentrations of 1 µM, 10µM or 0·5mM. In vivo ³¹P-NMR measurements showed that there was no discernibledifference in the cytoplasmic Pi content (µmol cm^–3root volume) of the mature roots of plants exposed to 1 µM,10µM or 0·5 mM external phosphate for up to 12d. However, the vacuolar Pi content of the mature roots variedabout 10-fold between these three groups. The cytoplasmic Pi content of roots receiving no external phosphatedecreased significantly after about 7 d total growth, and atabout this time the vacuolar pool of Pi became too small foraccurate measurement. The presence of 1 µM Pi in the nutrientsolution completely prevented this decline in cytoplasmic Pi,and there was some evidence that it also raised the Pi contentof the root vacuoles above the almost undetectable level foundin the totally P-starved roots. During the first 7–9 d of growth, the nucleoside triphosphatecontent of the mature roots was unaffected by the concentrationof phosphate in the nutrient solution. The results highlight the close control of cytoplasmic concentrationsof certain important phosphorus metabolites in roots growingin soil of normal agricultural fertility. Key words: Vacuole, cytoplasm, intracellular compartmentation, NTP, P-nutrition     

Effect of Elicitation and Changes in Extracellular pH on the Cytoplasmic and Vacuolar pH of Suspension-Cultured Soybean Cells

We have employed both ³¹P nuclear magnetic resonance spectroscopy and two intracellular fluorescent pH indicator dyes to monitor the pH of the vacuole and cytoplasm of suspension-cultured soybean cells (Glycine max Merr cv Kent). For the ³¹P nuclear magnetic resonance studies, a flow cell was constructed that allowed perfusion of the cells in oxygenated growth medium throughout the experiment. When the perfusion medium was transiently adjusted to a pH higher than that of the ambient growth medium, a rapid elevation of vacuolar pH was observed followed by a slow (approximately 30 minute) return to near resting pH. In contrast, the concurrent pH changes in the cytoplasm were usually fourfold smaller. These data indicate that extracellular pH changes are rapidly communicated to the vacuole in soybean cells without significantly perturbing cytoplasmic pH. When elicitors were dissolved in a medium of altered pH and introduced into the cell suspension, the pH of the vacuole, as above, quickly reflected the pH of the added elicitor solution. In contrast, when the pH of either a polygalacturonic acid or Verticillium dahliae elicitor preparation was adjusted to the same pH as the ambient medium, no significant change in either vacuolar or cytoplasmic pH was observed during the 35 minute experiment. These results were confirmed in experiments with pH-sensitive fluorescent dyes. We conclude that suspension-cultured soybean cells do not respond to elicitation by significantly changing the pH of their vacuolar or cytoplasmic compartments.     

Effect of phosphorus deficiency on levels of phosphorus compounds in spirodela

When Spirodela plants are transferred to a phosphate-deficient medium, growth slows down immediately, and ceases after 14 days. During this time, inorganic phosphate content falls from 30 to 0.7 μmoles/g fresh weight of tissue, phosphate ester content from 3.5 to 0.6 μmoles/g, phospholipid content from 3.5 to 1.2 μmoles/g, and residual phosphate (mainly RNA) content from 7.5 to 2.0 μmoles/g. Relative proportions of the various phosphate esters, and relative proportions of the various phospholipids, are not markedly affected by phosphate deficiency. Turnover rates of phosphate esters are somewhat higher in phosphate-deficient tissue. In control tissue, inorganic phosphate is present in 2 pools; a metabolic (12%) and a non-metabolic pool (88%). In phosphate-deficient tissues, most of the inorganic phosphate (>90%) is in the metabolic pool. Non-metabolic phosphate is presumably stored in the vacuole, and is not readily accessible to the tissue, so that growth normally occurs at the expense of external phosphate. During deficiency, growth is limited by the rate at which phosphate can be transported through the tonoplast and tissue to the growing point. Growth ceases when the supply of non-metabolic phosphate is exhausted. Metabolic phosphate is presumably located in the cytoplasm: it can not be used for growth. Nor can the plant respond to deficiency by making some phosphorus compounds at the expense of others. In this respect, phosphorus deficiency and nitrogen deficiency are dissimilar.     

Intracellular distribution of phosphate in cultured Humulus lupulus cells growing at elevated exogenous phosphate concentrations

The distribution of inorganic phosphate (Pi) between the cytoplasm and the vacuole of Humulus lupulus L. cells grown in suspension culture at different exogenous Pi levels was examined by 31-P nuclear magnetic resonance. In growing cells excess Pi accumulated in the vacuoles and the inhibitory effect of high exogenous Pi was not associated with a change in the cytoplasmic Pi level or with a change in the cytoplasmic pH.Abbreviations MES 2-(N-morpholino)ethanesulphonic acid - NMR nuclear magnetic resonance - Pi inorganic phosphate - ppm parts per million     

Biochemical changes during sucrose deprivation in higher plant cells

The mobilization of stored carbohydrates (sucrose and starch) during sucrose starvation was studied with sycamore (Acer pseudoplatanus) cells. When sucrose was omitted from the nutrient medium, vacuolar sucrose was first consumed. When a threshold of intracellular sucrose concentration was attained the cytoplasmic phosphorylated compounds decreased whereas cytoplasmic Pi increased symmetrically. Such a situation triggered starch breakdown. When almost all the intracellular sucrose pool had disappeared, the cell respiration rates (normal and uncoupled) declined progressively. The decrease in the rate of respiration triggered by sucrose starvation was attributable neither to the availability of substrate for mitochondrial respiration nor to a decrease in the maximal rate of O2 consumption by mitochondria expressed in terms of nanomole of O2 consumed per min/mg of mitochondrial protein. In fact, the uncoupled respiration rates decreased in parallel with the decrease in total intracellular cardiolipin or cytochrome aa3. These results demonstrate therefore that after a long period of sucrose starvation the progressive decrease in the uncoupled rate of O2 consumption by sycamore cells was attributable to a progressive diminution of the number of mitochondria/cell.     

31P NMR Measurements of the Cytoplasmic and Vacuolar Pi Content of Mature Maize Roots: Relationships with Phosphorus Status and Phosphate Fluxes

The vacuolar and cytoplasmic inorganic phosphate (Pi) contentof the mature regions of maize roots was measured by a ³¹P NMRtechnique which used an external standard to avoid the needfor tissue extraction and which exploited the relatively rapidrelaxation of cytoplasmic Pi in order to improve the detectionof this pool in fully-vacuolated cells. In mature roots of maize growing with abundant external phosphate,the concentration of Pi in the cytoplasm was approximately 6.5mol m^–3. When these plants were deprived of external phosphate,the vacuolar Pi content of the roots decreased rapidly, butthe cytoplasmic Pi concentration initially remained constantand did not begin to decline until P-stress became severe. Calculationsshow that withdrawal of Pi from the vacuoles into the cytoplasmunder these conditions would be against an electrochemical gradient. During P-starvation, an increased capacity for Pi influx developed,preceding any detectable change in the cytoplasmic Pi contentof the roots. This response is considered in terms of paralleleffects on transport sites for phosphate at the plasmalemmaand at the tonoplast. Comparisons of simultaneous rates of influxand net uptake implied that phosphate efflux accounted for <10% of influx in plants of a steady or declining P-status. However,direct measurements of efflux suggested that this process maybe temporarily accelerated when plants are recovering from P-stress. Key words: P-nutrition, subcellular compartmentation     

A phosphorus-31 nuclear magnetic resonance study of phosphate uptake and storage in cultured Catharanthus roseus and Daucus carota plant cells

High resolution 31P NMR spectra (103.2 MHz) of oxygenated Catharanthus roseus and Daucus carota cells grown in suspension cultures were obtained using a solenoidal perfusion probe. The spectra showed resonances for various phosphorylated metabolites such as ATP, ADP, NAD(P)(H), nucleoside diphosphoglucose, and sugar phosphates. The relative levels of the phosphorylated metabolites remained constant throughout the growth curve. No resonances for storage compounds such as polyphosphates, pyrophosphate, or phytates were observed. Two resolved resonances for Pi indicated an intracellular pH of 7.3 and 5.7 (or below) for the cytoplasm and vacuoles, respectively. The time course of Pi uptake and storage during growth in fresh culture medium was followed by studying the level of vacuolar Pi with 31P NMR (145.7 MHz). Simultaneously, the level of Pi in the culture medium was followed with radioactive 32P. C. roseus quickly takes up all the Pi from the culture medium (maximum rate 1.7 mumol min-1 g-1 (dry weight of cells]. The Pi is first stored in the vacuoles; subsequently, one part of this pool is used to keep a constant cytoplasmic Pi level while another part is apparently accumulated as an NMR invisible Pi store, probably in another cell organelle. In contrast, D. carota does not accumulate Pi in the vacuoles and consequently it takes up Pi from the medium at a much slower rate (0.05 mumol min-1 g-1 (dry weight of cells].     

Seed Dormancy in Red Rice : VI. Monocarboxylic Acids: A New Class of pH-Dependent Germination Stimulants

The addition of an elicitor (glucan) to Phaseolus vulgaris cell suspension cultures increased the formation of the phytoalexin phaseollin. Intracellular pH and phosphate concentrations were studied with ³¹P nuclear magnetic resonance spectroscopy on elicitor-treated cells which were aerated during the nuclear magnetic resonance measurement. The pH of the vacuole and to a lesser extent the pH of the cytoplasm were affected at 10 minutes after elicitor addition; a decrease in pH from 5.3 to 4.8 was noted in the vacuole and from 7.46 to 7.28 in the cytoplasm. The ratio between the amount of Pi in the vacuole to that in the cytoplasm also changed within 10 minutes after elicitor addition. The signal for ATP (β-ATP) was low after elicitor addition and was high again 23 hours after elicitation. Forty-eight hours after elicitor addition, vacuolar and cytoplasmic pH had almost returned to their initial values. The rapid change in vacuolar and cytoplasmic pH may cause the change of metabolism that occurs in elicitor-treated P. vulgaris cells.     

Lack of Control in Inorganic Phosphate Uptake by Catharanthus roseus (L.) G. Don Cells (Cytoplasmic Inorganic Phosphate Homeostasis Depends on the Tonoplast Inorganic Phosphate Transport System?)

Inorganic phosphate (Pi) uptake by Catharanthus roseus (L.) G. Don cells was studied in relation to its apparent uncontrolled uptake using 31P-nuclear magnetic resonance spectroscopy. Kinetics of Pi uptake by the cells indicated that apparent Km and Vm were about 7 [mu]M and 20 [mu]mol g-1 fresh weight h-1, respectively. Pi uptake in Murashige-Skoog medium under different Pi concentrations and different initial cell densities followed basically the same kinetics. When supplied with abundant Pi, cells absorbed Pi at a constant rate (Vm) for the first hours and accumulated it in the vacuole. As the endogenous pool expanded, the rate of Pi uptake gradually decreased to nil. Maximum Pi accumulation was 100 to 120 [mu]mol g-1 fresh weight if cell swelling during Pi uptake (about 2-fold in cell volume) was not considered. Results indicated that (a) the rate of Pi uptake by Catharanthus cells was independent of initial cell density and was constant over a wide range of Pi concentrations (2 mM to about 10 [mu]M) unless the cells were preloaded with excess Pi, and (b) there was no apparent feedback control over the Pi uptake process in the plasma membrane to avoid Pi toxicity. The importance of the tonoplast Pi transport system in cytoplasmic Pi homeostasis is discussed.     

Dynamic aspects of vacuolar and cytosolic amino acid pools of Saccharomyces cerevisiae.

By using the Cu2+ method (Y. Ohsumi, K. Kitamoto, and Y. Anraku, J. Bacteriol. 170:2676-2682, 1988) for differential extraction of the vacuolar and cytosolic amino acid pools from yeast cells, the amino acid compositions of the two pools extracted from Saccharomyces cerevisiae cells, grown in synthetic medium supplemented with various amino acids, were determined. Histidine and lysine in the medium expanded the vacuolar pool extremely. Glutamate also accumulated in the cells, but mainly in the cytosol. The composition of amino acids in the cytosolic pool was fairly constant, in contrast to that in the vacuolar pool. Cells grown in synthetic medium supplemented with 10 mM arginine accumulated arginine in the vacuoles at a concentration of about 430 mM. This large arginine pool was metabolically active and was effectively utilized during nitrogen starvation. Arginine efflux from the vacuoles was coupled with K+ influx, with an arginine/K+ exchange ratio of 1, as judged by the initial rate. The vacuolar arginine pool was exchangeable with lysine added to the medium and was decreased by treatment of the cells with the mating pheromone, alpha-factor.     

In vivo grapevine anthocyanin transport involves vesicle-mediated trafficking and the contribution of anthoMATE transporters and GST

In cells, anthocyanin pigments are synthesized at the cytoplasmic surface of the endoplasmic reticulum, and are then transported and finally accumulated inside the vacuole. In Vitis vinifera (grapevine), two kinds of molecular actors are putatively associated with the vacuolar sequestration of anthocyanins: a glutathione-S-transferase (GST) and two MATE-type transporters, named anthoMATEs. However, the sequence of events by which anthocyanins are imported into the vacuole remains unclear. We used MYBA1 transformed hairy roots as a grapevine model tissue producing anthocyanins, and took advantage of the unique autofluorescence of anthocyanins to study their cellular trafficking. In these tissues, anthocyanins were not only visible in the largest vacuoles, but were also present at higher concentrations in several vesicles of different sizes. In the cell, small vesicles actively moved alongside the tonoplast, suggesting a vesicular trafficking to the vacuole. Subcellular localization assays revealed that anthoMATE transporters were closely related with these small vesicles, whereas GST was localized in the cytoplasm around the nucleus, suggesting an association with the endoplasmic reticulum. Furthermore, cells in hairy roots expressing anthoMATE antisense did not display small vesicles filled with anthocyanins, whereas in hairy roots expressing GST antisense, anthocyanins were accumulated in vesicles but not in the vacuole. This suggests that in grapevine, anthoMATE transporters and GST are involved in different anthocyanin transport mechanisms.     

Novel mannose-sequestration technique reveals variation in subcellular orthophosphate pools do not explain the effects of phosphorus nutrition on photosynthesis in Eucalyptus globulus seedlings

Although only a small proportion of plant phosphorus (P) is used for photosynthesis, the relationships between P and photosynthesis can be strong. It was hypothesized, in this study, that variation in the allocation of orthophosphate (Pi) between active (cytoplasmic) and nonactive (vacuolar) pools would underpin differences in rates of photosynthesis in 4-month-old Eucalyptus globulus seedlings grown with a varying P supply. Photosynthetic biochemistry was assessed by the response of net photosynthesis to increasing intercellular [CO2]. Cytoplasmic Pi was sequestered as mannose 6-phosphate. Total P and the proportion of P as Pi were positively related to P supply. The ratios of active : stored Pi (10-24%) varied little over the range of treatments. Active Pi was positively related to P supply, as was photosynthesis (7 micromol CO2 m(-2) s(-1) with 0 mM P vs. 16 micromol CO2 m(-2) s(-1) with 0.32 mM P). Positive relationships between P supply and photosynthesis were explained best by leaf P content, not by active pools of Pi. The distribution of Pi between the vacuole and the cytoplasm had little impact on the photosynthetic phosphorus-use efficiency (PPUE), and reductions in cytoplasmic Pi had little effect on photosynthesis. Hence, PPUE is an unsuitable guide for assessing plant responses to increasingly unavailable P in the environment.