Phosphorus Uptake by Plants: From Soil to Cell

Abstract

P is an important plant macronutrient, making up about 0.2% of a plant's dry weight. It is a component of key molecules such as nucleic acids, phospholipids, and ATP, and, consequently, plants cannot grow without a reliable supply of this nutrient. Pi is also involved in controlling key enzyme reactions and in the regulation of metabolic pathways (Theodorou and Plaxton, 1993). After N, P is the second most frequently limiting macronutrient for plant growth. This update focuses on P in soil and its uptake by plants, transport across cell membranes, and compartmentation and redistribution within the plant. We will concentrate on P in higher plants, although broadly similar mechanisms have been shown to apply in algae and fungi. Although the total amount of P in the soil may be high, it is often present in unavailable forms or in forms that are only available outside of the rhizosphere. Few unfertilized soils release P fast enough to support the high growth rates of crop plant species. In many agricultural systems in which the application of P to the soil is necessary to ensure plant productivity, the recovery of applied P by crop plants in a growing season is very low, because in the soil more than 80% of the P becomes immobile and unavailable for plant uptake because of adsorption, precipitation, or conversion to the organic form (Holford, 1997). Plant acquisition of soil P. Plant root geometry and morphology are important for maximizing P uptake, because root systems that have higher ratios of surface area to volume will more effectively explore a larger volume of soil (Lynch, 1995). For this reason mycorrhizae are also important for plant P acquisition, since fungal hyphae greatly increase the volume of soil that plant roots explore (Smith and Read, 1997). In certain plant species, root clusters (proteoid roots) are formed in response to P limitations. These specialized roots exude high amounts of organic acids (up to 23% of net photosynthesis), which acidify the soil and chelate metal ions around the roots, resulting in the mobilization of P and some micronutrients (Marschner, 1995). The uptake of P poses a problem for plants, since the concentration of this mineral in the soil solution is low but plant requirements are high. The form of P most readily accessed by plants is Pi, the concentration of which rarely exceeds 10 μm in soil solutions (Bieleski, 1973). Therefore, plants must have specialized transporters at the root/soil interface for extraction of Pi from solutions of micromolar concentrations, as well as other mechanisms for transporting Pi across membranes between intracellular compartments, where the concentrations of Pi may be 1000-fold higher than in the external solution. There must also be efflux systems that play a role in the redistribution of this precious resource when soil P is no longer available or adequate. The form in which Pi exists in solution changes according to pH. The pKs for the dissociation of H3PO4 into H2PO4−and then into HPO42− are 2.1 and 7.2, respectively. Therefore, below pH 6.0, most Pi will be present as the monovalent H2PO4−species, whereas H3PO4 and HPO42− will be present only in minor proportions. Most studies on the pH dependence of Pi uptake in higher plants have found that uptake rates are highest between pH 5.0 and 6.0, where H2PO4−dominates (Ullrich-Eberius et al., 1984: Furihata et al., 1992), which suggests that Pi is taken up as the monovalent form. Under normal physiological conditions there is a requirement for energized transport of Pi across the plasma membrane from the soil to the plant because of the relatively high concentration of Pi in the cytoplasm and the negative membrane potential that is characteristic of plant cells. This energy requirement for Pi uptake is demonstrated by the effects of metabolic inhibitors, which rapidly reduce Pi uptake. The precise mechanics of membrane transport are still not clear, although cotransport of Pi with one or more protons is the favored option based on the following observations. The addition of Pi to starved roots results in both depolarization of the plasma membrane and acidification of the cytoplasm (Ullrich and Novacky, 1990). The depolarization indicates that Pi does not enter simply as H2PO4− or HPO42−, both of which would lead to membrane hyperpolarization. From these results it is likely that Pi is co-transported with positively charged ions. Cotransport of Pi with a cation involving a stoichiometry of more than 1 C+/H2PO4−or more than 2 C+/HPO42−would result in a net influx of positive charge and hence lead to the observed membrane depolarization. The cytoplasmic acidification associated with Pi transport would suggest that the cation is H+, but acidification would occur regardless of the nature of the cation if the transported species were H2PO4−, since it would undergo a pH-dependent dissociation in the cytoplasm to HPO42− and H+. To verify H+cotransport requires simultaneous or at least comparable measurements of Pi influx and the change induced in cytoplasmic pH. Estimates of the cytoplasmic buffering capacity would then allow calculation of the Pi-associated H+ flux, from which the stoichiometry could be deduced. Pi uptake across the plasma membrane in animal cells normally involves cotransport with Na+. Na-energized, high-affinity Pi uptake systems have also been found in cyanobacteria and green algae. In some organisms, such as Saccharomycescerevisiae, both Na+- and H+-dependent Pi uptake systems have been described (Roomans et al., 1977). Dependence of Pi uptake on Na+ has not yet been demonstrated in higher plants, but this may be partly because few studies have actually tested this possible mode of energized Pi uptake. Transfer of Pi from the cytoplasm to the vacuole involves a different set of thermodynamic parameters to those applying to the plasma membrane, mainly because of the millimolar concentrations in the cytoplasm and vacuole compared with the micromolar concentrations in the soil. Few estimates of cytosolic and vacuolar Pi concentrations are available. However, when maize was grown at Pi concentrations similar to those found in soils (i.e. 10 μm), the root cell cytoplasmic Pi concentration was estimated to be higher than the vacuolar concentration (Lee and Ratcliffe, 1993). Soybean leaf cell cytoplasmic Pi concentrations were also found to be higher than concentrations in the vacuole when plants were grown in solutions containing 50 to 100 μm Pi (Lauer et al., 1989). Since the membrane potential of the vacuole is usually slightly positive with respect to the cytoplasm under these realistic conditions, Pi transfer to the vacuole need not be energized. In plants supplied with higher concentrations of P, Pi appears to be close to electrochemical equilibrium across the tonoplast. In one of the few studies in which tonoplast transport has been examined, Pi uptake into vacuoles isolated from P-sufficient barley leaves was shown to follow a monophasic, almost linear concentration dependence up to at least 20 mm, and was independent of ATP supply (Mimura et al., 1990). However, in vacuoles isolated from Pi-starved cells, Pi uptake rates were found to be much higher and ATP dependent, despite the fact that the lower Pi concentrations in the vacuoles would favor passive Pi accumulation. This suggests the de-repression or activation of a second transporter in the tonoplast in response to Pi starvation. The concentration dependence of Pi uptake in vacuoles from Pi-starved cells has not been reported; a biphasic response would support the presence of a second transporter that might play an important role in maintaining Pi homeostasis when the Pi supply is limited. The process of vacuolar Pi mobilization following Pi starvation is likely to require energy-dependent transport across the tonoplast, the mechanism of which is not understood, although an H+/H2PO4−symport would be thermodynamically feasible. There is clearly a great deal more to understand about the specific mechanisms of vacuolar Pi transport in higher plants and the role these mechanisms play in buffering cytoplasmic Pi concentration. The question of whether there are several Pi transporters with different functional characteristics in plant cell membranes or only one transporter with characteristics that vary with internal Pi status or external concentration has been addressed using kinetic analysis of uptake. In this type of analysis a transporter's affinity (Km) for a particular mineral is estimated by measuring the rate of uptake at different external concentrations of an ion. Results from kinetic studies have been variously interpreted to support the existence of only one uptake system in barley roots (Drew and Saker, 1984) or up to seven in maize roots (Nandi et al., 1987). The most common interpretation of these kinetic studies is that two Pi uptake systems exist, one with a high affinity and activity that is either increased or de-repressed by Pi starvation, and one with a lower affinity and activity that is constitutive. Estimates of theKm for high-affinity uptake range from 3 to 7 μm, whereas for low-affinity transporters theKm estimates are more variable, from 50 to 330 μm in several different tissues and plant species (Ullrich-Eberius et al., 1984; McPharlin and Bieleski, 1987; Furihata et al., 1992). Recent advances in the molecular biology of putative plasma membrane and tonoplast Pi transporters confirm that plants have multiple transporters for Pi. Thus far, four different transporter genes have been cloned from Arabidopsis, three from potato, and two from tomato. Putative plasma membrane or tonoplast phosphate transporters in higher plants were cloned by probing the database of translated expressed sequence tags with fungal phosphate transporter peptide sequences. This approach identified at least three expressed sequence tags from randomly sequenced Arabidopsis cDNAs with translational products that were similar to the fungal phosphate-transporter proteins. Using the expressed sequence tags, full-length clones have been isolated from cDNA and genomic libraries (Muchhal et al., 1996; Leggewie et al., 1997; Smith et al., 1997). One putative phosphate transporter gene was expressed in tobacco cells (Mitsukawa et al., 1997). High-affinity Pi uptake was detected in the cells in which this gene was overexpressed, demonstrating that at least one member of this gene family encodes a high-affinity plasma membrane Pi transporter. The proteins encoded by these genes contain large regions that are identical to each other (Table I). The gene family appears to be clustered in the Arabidopsis genome with at least three members (APT1, APT2, andAtPT4) mapping to a specific region of chromosome 5 (Lu et al., 1997; Smith et al., 1997). These multiple Pi-transporter genes are differentially expressed. Some are strongly up-regulated by Pi starvation, whereas the expression of others is constitutive (Leggewie et al., 1997). In the cases of APT1 and APT2, the deduced amino acid sequences are 99% identical, which suggests that the proteins have the same functional characteristics. Although these proteins are almost identical, the promoter regions are completely different and may contain specific information that controls the spatial expression of these genes in different cell types, such as epidermal or cortical cells in the roots. Comparison matrix of phosphate transporter polypeptides Arabidopsis (Smith et al., 1997). Arabidopsis (Lu et al., 1997). Solanum (Leggewie et al., 1997). Arabidopsis (Muchhal et al., 1996). Glomus (Harrison and van Buuren, 1995). Saccharomyces (Bun-Ya et al., 1991). % identity of aligned sequences including gaps. Comparison matrix of phosphate transporter polypeptides Arabidopsis (Smith et al., 1997). Arabidopsis (Lu et al., 1997). Solanum (Leggewie et al., 1997). Arabidopsis (Muchhal et al., 1996). Glomus (Harrison and van Buuren, 1995). Saccharomyces (Bun-Ya et al., 1991). % identity of aligned sequences including gaps. A cDNA encoding a Pi transporter from potato, which is expressed in roots under conditions of Pi starvation, was characterized in thepho84 yeast mutant (Leggewie et al., 1997). TheKm for Pi uptake was 130 μm, much higher than would be expected if it were involved in Pi uptake from soils, where concentrations rarely exceed 10 μm. Various reasons were suggested (Leggewie et al., 1997) for the highKm values, but perhaps the most interesting is that phosphate transporters may contain a number of different protein subunits. The normal function of phosphate transporters may require subunits that are absent when this plant cDNA is expressed in yeast. Genetic evidence from Saccharomyces cerevisiaeindicates that several proteins containing putative membrane-spanning domains may interact to form a Pi-transporter complex (Bun-ya et al., 1991, 1996; Yompakdee et al., 1996). Although the association between these proteins has not been directly demonstrated, and one protein (Pho84) has been shown to be sufficient to catalyze phosphate transport in proteoliposomes (Berhe et al., 1995), the genetic evidence supports the idea that phosphate transporters are comprised of multiple subunits. In summary, kinetic and molecular data show that higher plants have multiple transporters for Pi across cellular membranes. The molecular data show that there are at least four genes that encode Pi transporters, and the kinetic data suggests the presence of two types of transporters with different affinities for Pi. The recent advances in the molecular biology of these transporters provide powerful tools for understanding how their function is integrated into plant physiological processes. More work will be required to gain a comprehensive picture of the location (cellular and subcellular) and precise function of the multiple phosphate transporters in plants. Maintenance of stable cytoplasmic Pi concentrations is essential for many enzyme reactions. This homeostasis is achieved by a combination of membrane transport and exchange between various intracellular pools of P. These pools can be classified in a number of different ways. First, according to their location in physical compartments such as the cytoplasm, vacuole, apoplast, and nucleus. The pH of these compartments will determine the form of Pi. The second pKa for H3PO4 is 7.2, so Pi in the cytoplasm will be approximately equally partitioned between the ionic forms H2PO4−and HPO42−, whereas in the more acidic vacuole and apoplast, H2PO4−will be the dominant species. Second, by the chemical form of P, such as Pi, P-esters, P-lipids, and nucleic acids. The proportion of the total P in each chemical form (except P in DNA) changes with tissue type and age and in response to P nutrition. Third, according to physiological function, as metabolic, stored, and of cells. The of the several including and cytoplasmic vacuolar of of and and of from et al., are for Pi and other in the cytoplasm or in the acidic vacuole is at present the only to directly the cytoplasmic and vacuolar pools of Pi in In an the of the in the an of the amounts of the different The the of Pi from which concentrations can be and Ratcliffe, 1993). studies that a rapidly of Pi of total is in the cytoplasm and a larger is in the vacuole studies have a to of the of the cytoplasmic and vacuolar pools of Pi within the plant. Pi is at concentrations more or of external Pi concentrations, under P (Lee et al., and Ratcliffe, 1995). In vacuolar Pi concentrations vary under conditions of P starvation, vacuolar Pi may be almost Pi in the vacuole also more readily than other P in response to P However, it does not to increase about (Lee et al., and Ratcliffe, 1995). the supply of Pi is plants grow more roots, increase the rate of uptake by roots from the Pi from and the vacuolar of Pi. In may more the roots. when plants have an supply of Pi and are it at rates that exceed a number of to the of Pi These the conversion of Pi into organic a in the Pi uptake rate from the outside solution (Lee et al., and Pi by which can be between and of the influx and or of these may be for the of intracellular Pi It is from both kinetic and molecular studies that the capacity to transport Pi across cellular membranes involves several different transporters and is in some by the external supply of et expression of phosphate transporters using kinetic in which the but not the system was by high concentrations of Pi. The expression of certain members of the putative plasma membrane or tonoplast phosphate-transporter gene family of Pi starvation. In Arabidopsis at least three genes encoding phosphate transporters are expressed in roots and are up-regulated by Pi starvation. in one gene was induced in roots and by the plants of Pi, whereas a second gene was expressed the plant under conditions of high or low in activity and phosphate-transporter gene expression show that plant cells to changes in the Pi concentration of the external or in the However, the intracellular and the that gene expression in the cytoplasmic concentrations of Pi relatively are at the molecular may provide into the that phosphate uptake the of genes encoding proteins that interact and Recent studies (Mimura et al., 1996; et al., 1997) provide a picture of of Pi in plants. In P-sufficient plants most of the Pi by the roots is transported in the to the of Pi in the range from 1 in Pi-starved plants to 7 in plants grown in solutions containing μm Pi (Mimura et al., 1996). There is also of Pi in the from leaves to the growing and from the to the roots. In plants the supply of Pi to the from the roots the is by increased mobilization of P in the leaves and to both the leaves and growing roots. This process involves both the of Pi and the of organic P in the A of plants is that approximately of the Pi from the to the roots in the is then to the and to the et al., 1997). In the P is transported almost as Pi, whereas amounts of organic P are found in the A number of that show Pi in leaves have been These may to understand the controlling the of Pi within the plant. One Arabidopsis mutant was isolated based on total phosphate concentrations in the leaf tissue et al., and was shown to have root Pi uptake rates that were the same as the but rates to the In it is not whether a gene encoding a transporter or has been the phosphate-transporter genes that have been cloned not to This the of specialized mechanisms for the transfer of Pi to the Arabidopsis P in its leaves to concentrations, which is of a in the regulation of Pi concentrations in and and the of intracellular There is a that Pi uptake by plants as a of uptake from the soil by root cells. However, in more than of plants, are formed with fungi. In these plants the fungal hyphae play an important role in the acquisition of P for the plant Smith and Read, 1997). can be into two and of which mycorrhizae are the most in the plant (Smith and Read, 1997). The is on the exchange of from the plant in for P and other mineral from the of P in roots by can be 3 to 5 higher than in roots of Smith and Read, 1997). The few studies of the of Pi uptake that roots and isolated hyphae have systems with characteristics similar to those found in roots and other et al., Smith and Read, 1997). of the have two systems μm and et al., 1990). A recent molecular (Harrison and van Buuren, identified the which encodes a high-affinity fungal phosphate transporter in external hyphae that is similar in both and function to high-affinity transporters in plants (Table I). A number of may to the increased rate of Pi uptake in plants (Smith and Read, 1997). of hyphae from the the plant to explore a volume of by the of Pi in the soil. studies have shown that the around plant roots, which is by plant uptake and the immobile nature of Pi, is larger in than in plants 1991). may also be to Pi from the soil solution more effectively than other soil because may be limiting in the is to the by the plant. The association could the plant to more effectively with soil for the amount of available soil Pi. may also be to P from organic that are not available directly to the plant acid and nucleic et al., 1992). is about the transport of P within mycorrhizae or the mechanism of P efflux from the Pi and organic P as could be within the by cytoplasmic or by to the plant root from external hyphae in the soil. The is that Pi is the form by the across the membranes. However, there is also evidence in higher plants that can be outside cells to release Pi. It is possible that is also by the to the Pi would then be taken up by the plant an H+ as in roots. Since it is that the phosphate transporter cloned from Glomus is not expressed in fungal the it cannot be a for the fungal P efflux of P must on a different transporter of The role of P in the regulation of is still understood, in because of In roots for P by the plant may the activity of P transporters in the with efflux from the the limiting However, studies of roots of et al., that although there was an increase in P in roots, the vacuolar Pi of and roots was The plants not Pi in the which suggests that the may be to the efflux of P to the plant. roots are to up Pi from solutions containing up to 100 Pi (Smith and Read, concentrations that likely to be in the soil. external Pi concentrations (up to on and growth of in the et al., 1996). These results suggest that the low of in plants growing in soils with high P status may not be the result of regulation of the activity of the by soil Pi, that specific from the plant the activity of the that P is an essential and often limiting for plant it is that many of P uptake and transport in plants are not studies have a picture of where Pi is in a kinetic studies have the functional characteristics of plasma membrane and tonoplast Pi transporters, and molecular studies have the presence of multiple genes encoding phosphate transporters that are differentially expressed. the important in understanding in this area will from the of these to provide a comprehensive picture of the function of phosphate transporters and how the of their spatial and expression the plant to with A to is that the soil Pi concentration has often been by plant It is common to in which plants were grown in 1 Pi, which may be higher than the Pi concentrations plants in agricultural or To understand how plants Pi from soils and internal Pi concentrations, studies on Pi uptake by plants must more soil conditions, in which the concentration of Pi is low and soil both acquisition and We and Smith for their and We to the were not directly because of limitations.