SEARCH

SEARCH BY CITATION

Keywords:

  • phosphate transporters;
  • OsPT2;
  • OsPT6;
  • rice;
  • cell membrane potential;
  • RNA interference

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Plant phosphate (Pi) transporters mediate the uptake and translocation of this nutrient within plants. A total of 13 sequences in the rice (Oryza sativa) genome can be identified as belonging to the Pi transporter (Pht1) family. Here, we report on the expression patterns, biological properties and the physiological roles of two members of the family: OsPht1;2 (OsPT2) and OsPht1;6 (OsPT6). Expression of both genes increased significantly under Pi deprivation in roots and shoots. By using transgenic rice plants expressing the GUS reporter gene, driven by their promoters, we detected that OsPT2 was localized exclusively in the stele of primary and lateral roots, whereas OsPT6 was expressed in both epidermal and cortical cells of the younger primary and lateral roots. OsPT6, but not OsPT2, was able to complement a yeast Pi uptake mutant in the high-affinity concentration range. Xenopus oocytes injected with OsPT2 mRNA showed increased Pi accumulation and a Pi-elicited depolarization of the cell membrane electrical potential, when supplied with mM external concentrations. Both results show that OsPT2 mediated the uptake of Pi in oocytes. In transgenic rice, the knock-down of either OsPT2 or OsPT6 expression by RNA interference significantly decreased both the uptake and the long-distance transport of Pi from roots to shoots. Taken together, these data suggest OsPT6 plays a broad role in Pi uptake and translocation throughout the plant, whereas OsPT2 is a low-affinity Pi transporter, and functions in translocation of the stored Pi in the plant.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Phosphorus (P) is one of the major essential mineral nutrients, and is involved in many key metabolic pathways in plants. Within plant cells, the inorganic phosphate (Pi) concentration is generally >10 mm (>3 g kg−1, on a biomass dry-weight basis), and yet the concentration in the soil solution is typically <10 μm (Bieleski, 1973; Marschner, 1995). Because of the low concentration of the soluble form and slow rate of diffusion to the root surface, plants have therefore evolved a range of strategies to increase the availability and uptake of soil Pi. The high-affinity Pi transporters (PTs) are assumed to be the predominant system responsible for Pi acquisition by plant roots (Marschner, 1995; Raghothama, 1999).

The first genes encoding putative plant PTs were isolated from Arabidopsis (Muchhal et al., 1996). Since then, a large number of PT genes have been identified from different plant families, including cereals, legumes and Solanaceous species (Chen et al., 2007; Chiou et al., 2001; Glassop et al., 2005; Harrison et al., 2002; Javot et al., 2007a; Leggewie et al., 1997; Liu et al., 1998a,b; Maeda et al., 2006; Mitsukawa et al., 1997; Mudge et al., 2002; Nagy et al., 2005; Paszkowski et al., 2002; Rae et al., 2003; Smith et al., 1997; Xu et al., 2007). The majority of the plant PTs belong to the Pht1 family, and they are predicted to have 12 transmembrane (TM) domains (Saier, 2000), containing two partially duplicated subdomains of six TM segments (Lagerstedt et al., 2004). Most of them showed expression either exclusively or predominately in the roots, and transcript levels were strongly induced by a low-Pi supply or by inoculation with arbuscular mycorrhiza (Bucher, 2007; Javot et al., 2007b; Rausch and Bucher, 2002), consistent with their role in uptake from the soil by roots or via mycorrhizal fungi. Analysis of the Arabidopsis genome has revealed that there are nine genes in the Pht1 family (Mudge et al., 2002), whereas in barley at least eight members have been identified (Rae et al., 2003). Expression analysis of promoter fusions to GUS and GFP reporter genes has showed that some Pht1 family members are expressed in plant parts other than the roots, including the stem, leaves, cotyledons, tubers and flowers (Karthikeyan et al., 2002; Mudge et al., 2002). It was speculated that these PTs were involved in the translocation of Pi within the plant (Mudge et al., 2002; Raghothama and Karthikeyan, 2005). Analysis of the T-DNA insertional knock-out plants revealed that AtPht1;1 and AtPht1;4, the most highly expressed PTs in Arabidopsis, play significant roles in Pi acquisition from both low- and high-Pi environments (Misson et al., 2004; Shin et al., 2004). In addition, knock-down or knock-out of arbuscular mycorrhiza (AM)-specific or -inducible PTs resulted in a significant decrease of Pi uptake via AM under P-limiting conditions (Javot et al., 2007a; Maeda et al., 2006; Xu et al., 2007), and a decrease or early death of the arbuscules (Javot et al., 2007a; Maeda et al., 2006).

Uptake of Pi into plant cells results in the transient depolarization of the plasma membrane, which is accompanied by an increase in the extracellular pH and in the acidification of the cytoplasm (Mimura, 1995; Sakano, 1990; Ullrich-Eberius et al., 1981). The biochemical analysis of plant PTs functionally expressed in yeast has provided further evidence that Pi uptake is mediated by proton/Pi co-transport (Daram et al., 1998; Leggewie et al., 1997; Liu et al., 1998b). Although a stoichiometry of between two and four H+ ions per Pi molecule has been suggested for symport across the plant–cell membrane (Sakano, 1990; Ullrich-Eberius et al., 1981, 1984), mechanistically, OH efflux (one OH molecule per Pi molecule antiport) cannot be excluded (Sakano, 1990). The electrical activity of the transport cycle of PTs cannot be easily studied in the yeast expression system, and so the mechanism is difficult to elucidate.

Kinetic analysis of Pi uptake in many plant species shows that plants have both low- and high-affinity uptake systems (Bieleski, 1973; Furihata et al., 1992; Nandi et al., 1987; Ullrich-Eberius et al., 1984). The complementation of a yeast Pi transport mutant has revealed the function of PTs. The tomato transporter LePT1 showed expression that was enhanced by Pi-starvation, and was able to complement a yeast mutant with a Michaelis constant, Km, of 31 μm (Daram et al., 1998). Similarly, three potato PTs, StPT1, StPT2 and StPT3, also mediated Pi transport in yeast with apparent Km values of 280, 130 (Leggewie et al., 1997) and 64 μm (Rausch et al., 2001), respectively. In another yeast complementation experiment, a rather high Km value (493 or 668 μm) was obtained for MtPT4, a mycorrhizal-specific PT in Medicago (Harrison et al., 2002). This suggested that MtPT4 is likely to function as a low-affinity Pi transporter in acquiring Pi released from mycorrhizal fungi. The Km values from yeast mutant complementation experiments are higher than those observed in physiological experiments in plants (Raghothama, 1999). By expressing PT in tobacco cells, a Km of only 3 μm was measured for AtPht1;1, which is an Arabidopsis PT (Mitsukawa et al., 1997). Using a similar plant-cell expression strategy, Km values of 9 and 385 μm were obtained for HvPht1;1 and HvPht1;6, two barley PTs expressed in rice suspension cells (Rae et al., 2003). However, the plant cell culture expression systems are limited for the analysis of the kinetic properties of transporters, as a result of the activity of endogenous PTs.

Rice is one of the most important crops, feeding about one half of the world’s population. In the genome of rice, there are a total of 13 genes encoding proteins that belong to the Pht1 high-affinity PT family (Goff et al., 2002; Paszkowski et al., 2002). However, their roles in the acquisition and translocation of P remain unclear, except for OsPht1;11 (OsPT11), which was shown to be specifically activated during mycorrhizal symbiosis (Paszkowski et al., 2002). PCR techniques were used to show that OsPht1;2 (OsPT2) and OsPht1;6 (OsPT6) are two of the most abundant transcripts among the Pi-starvation-regulated Pht1 genes in roots (Paszkowski et al., 2002; Jianning Zhao, Shubin Sun, Penghui Ai and Guohua Xu, unpublished data). For these two rice PTs, we report on their tissue-localized expression patterns, biochemical and biophysical properties in heterologous expression systems, and in planta roles. In addition, we provide electrophysiological evidence specifically demonstrating the electrical activity of plant PTs in a heterologous expression system. Our results are consistent with the idea that the PT in root epidermal cells acquires Pi from the low concentrations in the soil solution, whereas the PT in the stele operates at higher concentrations, translocating Pi within the plant.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

OsPT2 and OsPT6 are abundantly expressed under Pi starvation in rice

We have used reverse transcription (RT)-PCR to examine the specific expression pattern of all 13 members of the rice Pht1 family, during changes in P supply (data not shown). The transcripts of both OsPT2 and OsPT6 were predominately expressed in the roots in response to Pi deprivation, but expression in the leaves also increased upon Pi starvation, especially for OsPT6 (Figure 1). The RT-PCR was carried out by comparing expression with the housekeeping gene, actin (OsRac1, accession number AB047313). We also analyzed the expression of the two PT genes by quantitative real-time RT-PCR. The relative expression levels of OsPT2 and OsPT6 were 4.12 and 3.49 in the roots and 0.021 and 0.68 in the leaves, respectively, under P-starvation conditions.

image

Figure 1.  The phosphate-regulated transcriptional patterns of OsPT2 and OsPT6 in the roots and leaves of rice (Oryza sativa ssp. Japonica cv. Nipponbare). Total RNA was extracted from rice plants grown for 21 days in the presence of 0.3 mm Pi (+P), or in the absence of Pi (−P). RT-PCRs were performed with specific primers (Table S1) for OsPT2 and OsPT6 (27 cycles) in the roots, and for OsPT2 (35 cycles) and OsPT6 (27 cycles) in the leaves. A housekeeping gene, actin (OsRac1, accession number AB047313), was used as the internal standard.

Download figure to PowerPoint

Characterization of GUS expression patterns driven by the native promoters of OsPT2 and OsPT6 in rice

To determine the promoter regions for tissue specificity and Pi responsiveness, the sequences immediately upstream of the translation start for OsPT2 and OsPT6 were amplified. The 5′ fragment of 1670 bp for OsPT2 and 2380 bp for OsPT6 was used in expression analysis. The GUS gene (Jefferson et al., 1987) was used as the reporter, to examine the spatial distribution of expression and in the quantitative analysis of promoter strength. The promoter-reporter gene fusions were introduced into Nipponbare, the cultivar used for sequencing the rice genome (Goff et al., 2002), and progeny of the transgenic lines were assayed for the spatial distribution of expression under low- and high-Pi supply.

Plants expressing the reporter gene were supplied with differing levels of Pi, and various tissues were excised and stained for GUS detection (Figures 2–4). After Pi starvation for 3 weeks, the total P in the leaves decreased from 3.2 g kg−1 dry matter to <1.7 g kg−1 dry matter; meanwhile, the GUS activity driven by OsPT2 and OsPT6 promoters dramatically increased, and reached a maximum after 2–3 weeks (Figure S1). Both OsPT2 and OsPT6 were predominantly expressed in root tissues, and were strongly induced by Pi deprivation (Figures 2 and 3), which was consistent with their expression patterns analyzed by RT-PCR (Figure 1).

image

Figure 2.  Localization of OsPT2 promoter-GUS expression in transformed rice. Expression in different tissues of rice supplied with 0.3 mm Pi (+Pi) (a–d) and without Pi (−Pi) (e–k) for 21 days: (a, e) root tips; (b, f) root lateral branching zones; (c, g) hand-cut cross sections of the root–shoot junction; (d, h) leaves; (i) root-tip cross sections of the P-deficient plant, showing GUS activity in the phloem (Ph) and xylem (X), but not in the epidermis (Ep), cortex (Co) or endodermis (En); (j) cross sections of the lateral root branching zone of the P-deficient plant, showing GUS activity only in the phloem (Ph), xylem (X) and lateral root primordium (LRP); (k) cross section of leaf from the P-deficient plant. Leaf cell-types showing GUS expression include the mesophyll (Me), phloem (Ph) and xylem (X) cells, but not the epidermal (Ep) cells.

Download figure to PowerPoint

image

Figure 3.  Localization of OsPT6 promoter-GUS expression in transformed rice. Expressions in different tissues of the rice supplied with 0.3 mm Pi (+Pi) (a–d) and without Pi (−Pi) (e–k) for 21 days: (a, e) root tips; (b, f) root lateral branching zones; (c, g) hand-cut cross section of the root–shoot junction; (d, h) leaves; (i) root-tip cross section of the P-deficient plant, showing GUS activity in the epidermis (Ep), cortex (Co), endodermis (En), phloem (Ph), and xylem (X) cells; (j) cross sections of the lateral root branching zone of the P-deficient plant, showing GUS activity only in the phloem (Ph), xylem (X) and lateral root primordium (LRP) cells; (k) cross section of a leaf from the P-deficient plant. Leaf cell-types showing GUS expression include the phloem (Ph), xylem (X) and mesophyll (Me) cells.

Download figure to PowerPoint

image

Figure 4.  Functional expression of OsPT6 in yeast. (a) Staining test for acid phosphatase activity in the yeast strains MB192 (control), Yp112-OsPT6, which contains OsPT6 in MB192, and wild type (WT). The culture medium contains 0.02, 0.06 and 0.10 mm Pi, respectively. (b) The effects of different pH in the culture medium on the growth of the three yeast strains: Yp112-OsPT6, MB192 and WT. (c) Velocity of 32Pi transport by Yp112-OsPT6 as a function of Pi concentration. The non-linear regression of Pi uptake of strain Yp112-OsPT6 versus external Pi concentration at pH 6 was used to estimate the apparent Km value for Pi uptake.

Download figure to PowerPoint

Under P-deprived conditions, the reporter gene driven by the OsPT2 promoter was expressed throughout the stele in primary roots and lateral roots, but not in epidermal and cortical cells (Figure 2e–g,i,j), whereas it was not detectable in the roots under Pi-replete conditions (Figure 2a,b). Strong OsPT6 promoter-GUS expression was observed in the younger primary and lateral roots under P deficiency (Figure 3e,f). Furthermore, the expression of OsPT6 was found not only in epidermal cells, and throughout the young primary and lateral roots, which are the dominant sites for uptake of Pi from the soil solution, but was also found throughout the cortical and stele cells, under Pi-starved conditions (Figure 3g,i,j), thereby indicating that this transporter might play a wider role in Pi uptake and translocation. As the age of the primary roots increased, the degree of OsPT6 promoter-GUS expression decreased (Figure 3f,j), whereas expression was particularly abundant at the site of lateral root emergence (Figure 3j). Provision of high levels of Pi in the culture solution suppressed the expression of OsPT6 in the root system (Figure 3a–c), and a similar effect for OsPT2 could also be seen (Figure 2a,b). No expression of either OsPT2 or OsPT6 promoter-GUS fusion was observed in the root cap, either with or without a supply of Pi (Figures 2a,e and 3a,e).

We also checked the level of GUS expression driven by either the OsPT2 or OsPT6 promoters in the leaves of Pi-starved rice (Figures 2h and 3h). A high supply of Pi in the culture solution decreased the expression of the OsPT2 gene (Figure 2c,d), and almost completely suppressed the expression of OsPT6 in the leaves (Figure 3c,d). Both OsPT2 and OsPT6 promoters also directed the expression of GUS in some mature organs (Figure S2).

Functional characterization of OsPT6 in a yeast strain defective in Pi-uptake

We analyzed the function of OsPT2 and OsPT6 using the yeast MB192 strain for complementation analysis. The cells transformed with either OsPT2 or OsPT6, or with empty vector, were grown in yeast nitrogen base (YNB) medium containing various concentrations of Pi for 10 h. Bromocresol Purple was used as a pH indicator, giving a color shift, from purple to yellow, during the acidification of the liquid medium: this change correlated well with the growth of the yeast cells. In comparison with the cells of both wild type plants and the mutant MB129, the mutant cells expressing OsPT6 (Yp112-OsPT6) could partially restore their growth at 20 μm Pi, and grew well at 60 μm Pi (Figure 4a). At mm Pi concentrations, there was no obvious difference in growth between the three cell lines, because of the activity of the endogenous low-affinity Pi-uptake system (Tamai et al., 1985). In contrast, the yeast cells transformed with p112-OsPT2 did not functionally complement the mutants (Figure S3), and thus could not be used for further studies. This observation suggested that OsPT2 might be a low-affinity PT.

As the Pht1 transporters belong to the H+/Pi symporter family, we assayed the pH dependence of OsPT6 Pi transport by measuring the optical density of the yeast cell lines at a range of pH values, from 4 to 8. Both the wild-type and the MB192 yeast strain expressing OsPT6 grew much faster at pH 4–6 when compared with pH 7 and 8 (Figure 4b). The pH optima for the yeast mutant cells carrying p112-OsPT6 was 6, whereas in comparison, for the wild type it was pH 4–5, indicating the differing proton dependence of OsPT6 and the endogenous high-affinity yeast PTs. The growth rate of the wild type at pH 8 was about an eighth of the rate at pH 4 and 5. However, the strain expressing OsPT6 could tolerate alkaline conditions much more than the wild type (growth rate at pH 8 was about 0.5–0.2 of the rate at pH 4–6), indicating that OsPT6, in contrast with the wild-type, high-affinity PTs, was better able to function and sustain cell growth at pH 8 in the culture medium (Figure 4b).

In order to determine the kinetic properties of the OsPT6 transporter, Pi-uptake experiments using 32Pi were performed using the transformed yeast. A Lineweaver–Burk plot was drawn using the reciprocal uptake velocities at 5 min after the addition of 32Pi, and these graphs indicated that Pi uptake mediated by OsPT6 followed Michaelis–Menten kinetics, with an apparent mean Km of 97 μm Pi, found from the average of three independent experiments (Figure 4c).

Functional assay of OsPT2 in oocytes by measurement of Pi accumulation and changes in membrane potential

As OsPT2 failed to complement the yeast Pi-uptake mutant, we attempted to heterologously express the protein in Xenopus oocytes. The oocytes injected with mRNA encoding OsPT2 were compared with water-injected controls in Pi-uptake assays. Expression of OsPT2 resulted in a more than sevenfold increase in the Pi content of the oocyte, in comparison with the control (Figure 5a). This result demonstrated that OsPT2 could mediate the uptake of Pi from the external solution. Electrophysiological measurements showed that the plasma membrane potential became less negative when NaH2PO4 was supplied in the external solution of oocytes injected with OsPT2 mRNA (Figure 5b). This extent of the electrical depolarization was dependent on the external NaH2PO4 concentration, and increased from 1 to 10 mm. In contrast, water-injected oocytes did not show any response to the supply of NaH2PO4 at these concentrations (Figure 5c). This result indicated that the entry of Pi into the cell occurred with a net positive charge, supporting the idea that OsPT2 is a co-transporter, with an H+:Pi stoichiometry ratio of 2–4:1 (Sakano, 1990). The data also suggest that OsPT2 has a low Pi affinity, mediating Pi-uptake in the mm range (Figure 5b).

image

Figure 5.  The accumulation of P in oocytes injected with OsPT2 mRNA (a), and the responses of the oocyte membrane potential injected with OsPT2 mRNA to NaH2PO4 (b), and the responses of the oocyte membrane potential injected with water to NaH2PO4 and 20 mm NH4Cl (c). The cytosolic Pi values were shown as the mean ± SD for 16–20 oocytes incubated in 0.5 mm NaH2PO4 modified Barth’s saline (pH 7.4) overnight (a). Warol. The control oocyte was treated with 20 mm NH4Cl to show a characteristic endogenous response of the cell, and this response was also shown by mRNA-injected cells (data not shown).

Download figure to PowerPoint

Knock-down of OsPT2 and OsPT6 through in planta RNAi decreases P-uptake and transport within the plant

To knock-down (KD) the expression of OsPT2 and OsPT6, an OsPT2- and an OsPT6-RNAi construct, driven by a Cauliflower mosaic virus (CaMV) 35S promoter, was generated and used to transform rice. The constructs contained inverted repeats of fragments of either OsPT2 or OsPT6 specific coding regions that are not conserved among the 13 rice Pht1 members. In the two selected OsPT2-RNAi transgenic lines, the expression of OsPT2, but not of OsPT6, was reduced significantly in the line r2-1, and was moderately reduced in the r2-2 plants (Figure 6a). In the roots, relative to wild-type controls, the transcript level of OsPT2 was around 18% in line r2-1, and 54% in line r2-2 (Figure S4). In the three OsPT6-RNAi KD lines (r6-1, r6-2 and r6-3) treated with 10 μm Pi, the expression of OsPT6 relative to controls was about 17–23% in the roots, and 27–32% in the shoots (Figures 6b and S4). The expression of OsPT2 in the KD lines r6-2 and r6-3 was also greatly decreased relative to wild-type controls (Figures 6b and S4).

image

Figure 6.  RT-PCR analysis in the roots and 32P uptake assay (a, b), and total P uptake (c) of the wild type, Ospt2 and Ospt6 RNAi transgenic plants. r2-1 and r2-2 represent different OsPT2 RNAi transgenic lines. r6-1, r6-2, r6-3 represent different OsPT6 RNAi transgenic lines. The wild-type and Ospt2 (a)/Ospt6 (b) RNAi transgenic plants were grown at 10 μm Pi solution for three weeks. The seedlings were transferred into Hoagland's nutrient solution containing KH232PO4 and 10 μm additional KH2PO4 for 24 h, with a day/night cycle of 16/8 h, and with temperature settings of 28/22°C. Autoradiographs of the seedlings were developed in three replicated experiments (a, b). The total P in shoots and roots of the wild-type and Ospt2/Ospt6 RNAi plants grown in 10 μm Pi solution was measured after 3 weeks (c). Error bars indicate SE (n = 4).

Download figure to PowerPoint

The differences in the total Pi uptake, and long-distance transport from roots to shoots, between the controls and the RNAi KD lines were evaluated by supplying the 32P radioisotope in the culture solution for 1 day, after 3 weeks of provision at 10 μm Pi, and by the continuous supply of 10 μm Pi for 3 weeks. Even after carefully washing the residual 32P from the root surfaces using nutrient solution several times, both the wild-type and the RNAi lines showed very strong 32P signals in their roots. Whereas in the shoot, the 32P signal was abundant in the wild type, but it decreased dramatically from the base to the top of the plant in four of the five RNAi lines; one exception was r2-2, which also showed much less suppression of OsPT2 expression (Figures 6a and S4). In comparison with the wild type, we observed that the concentrations of total P were slightly, but not significantly, decreased in the KD plant roots, whereas the concentrations of P in the shoots was much lower, particularly in the KD r2-1 line (Figure S5). The KD lines contained almost the same level of total P as the control roots, but only 60–83% of that measured in the shoots of control plants (Figure 6c). The significant decrease of total P in the KD plants relative to controls (Figure 6c) demonstrates that both OsPT2 and OsPT6 are required for rice to maximize P uptake from low-P culture medium. The ratio of total P uptake in the shoot to that in the roots of the KD lines r6-2, r6-3 and particularly in r2-1 was very significantly lower than that in the wild type. Whereas, by contrast, the same ratio was not significantly different from wild-type plants in the r2-2 and r6-1 lines (Figure S6). In view of the differing degree of KD of the two genes in these lines, we could conclude that OsPT2, represented by the r2-1 line, is required for the long-distance transport of Pi from roots to shoots in rice. Whereas OsPT6, represented by the r6-1 line, plays a more important role in the direct uptake from the culture medium, rather than in long-distance Pi transport.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The low concentration of Pi commonly found in the soil solution (Marschner, 1995) has led to the hypothesis that only high-affinity PTs can function for the uptake of Pi across the plasma membrane of root epidermal cells, whereas the low-affinity PTs may mediate transport of Pi within the plant (Rae et al., 2003). Using the analysis of GUS-reporter tissue-specific expression, and the functional characterization in two different heterologous expression systems, together with RNAi KD lines, we show that both OsPT2 and OsPT6 are proton/Pi co-transporters, with distinct roles in acquisition and transport of Pi within the rice plant. An in planta plasma membrane localization of these two transporters was not confirmed, and still requires further research. The P1BS-like and W-box elements are present in the promoter sequences of a range of Pi-regulated genes in monocot and dicot plant species (Devaiah et al., 2007; Schunmann et al., 2004; Tittarelli et al., 2007). The putative promoter regions of both OsPT2 and OsPT6 contain P1BS-like and W-box cis-elements (data not shown), supporting the fact that Pi deficiency enhanced the expression of these two PTs in both roots and leaves (Figures 1–3). Furthermore, we report that plant members of the Pht1 family can be functionally expressed in Xenopus oocytes, and that the uptake is demonstrated by both the accumulation of Pi and the change in membrane electrical potential of the cell (Figure 5).

Deciphering OsPT2 and OsPT6 transport activity

The PT proteins are present in all living organisms, and are reckoned to use either the TM Na+ (animals and fungi) or H+ (plants and bacteria) gradients to drive Pi transport (Ravera et al., 2007). It has long been assumed that plant PTs are proton/Pi co-transporters, with a stoichiometry of 2–4 H+/Pi (Rausch and Bucher, 2002). The uptake of Pi into plant cells should then result in a transient depolarization of the plasma membrane, which is accompanied by an increase in the extracellular pH and an acidification of the cytoplasm. By complementation of a knock-out of endogenous high-affinity PT and pharmacological testing of yeast mutants, or by measuring the increase of Pi uptake in transformed plant cell lines, several PTs from different plant species have now shown properties suggesting a pH-dependent co-transport that is energized by the electrochemical proton gradient across the plasma membrane (Daram et al., 1998; Harrison et al., 2002; Leggewie et al., 1997; Liu et al., 1998b; Rae et al., 2003; Rausch et al., 2001). We have successfully demonstrated the Pi transport activity of OsPT2, providing direct proof that this low-affinity transporter mediates electrogenic influx across the plasma membrane, and is therefore dependent on the electrochemical gradient of protons (Figure 5). This result suggests that the internal transport of Pi within tissues (see Figure 2) requires co-transport with protons, even at relatively high concentrations outside the cell. Therefore, within the vascular tissue of the rice plant there must be high concentrations, in the mm range, of Pi in the apoplast, which are likely to be maintained in solution by the expected acidic environment.

There have been several papers deciphering the characteristics of mammal Na+-coupled Pi co-transporters expressed in Xenopus oocytes (Bacconi et al., 2007; Ravera et al., 2007; Saliba et al., 2006), and most of the Na+/Pi co-transporters have shown low affinity in voltage clamp experiments. We observed a significant depolarization of the membrane potential of the oocytes when both 1 and 10 mm Pi was added to the external solution (Figure 5b). This result demonstrates that OsPT2 functions as a low-affinity PT, a result that is consistent with its specific localization in the root stele (Figure 2), where a concentration of as much as 10 mm Pi may be possible (Bieleski, 1973; Mimura, 1995; Poirier et al., 1991). The failed functional yeast complementation by OsPT2, in contrast to the high-affinity Pi transporter OsPT6 (Figure 4, discussed below), further supports the idea that it is a low-affinity PT.

OsPT2 is responsible for the root to shoot transport of Pi

Most of the Pi taken up by the root is transported in the xylem to growing leaves in Pi-sufficient plants, whereas a limited supply of Pi from the root to the shoot is augmented by an increase in mobilization of stored Pi, in the older leaves, for re-translocation to both younger leaves and growing roots in Pi-deficient plants (Himelblau and Amasino, 2001; Jeschke et al., 1997). It is presumed that Pi is loaded into the xylem, where the concentration may be as high as 10 mm, by an active Pi transport system (Poirier et al., 1991; Rausch et al., 2004). It was once assumed that Pht2;1, the only PT belonging to the low-affinity Pht2 family in plants, was a plasma membrane protein involved in the Pi loading of shoots (Daram et al., 1999). However, this hypothesis has been abandoned upon the discovery of the chloroplast localization of Pht2;1 (Rausch et al., 2004). Low-affinity PTs in the Pht1 family are now thought to play this role in loading transport to the shoot, and this has largely been inferred from the spatial expression of these genes in several different plant species (Chiou et al., 2001; Daram et al., 1998; Mudge et al., 2002). Here, we show that GUS reporter activity driven by the OsPT2 promoter was observed exclusively in the stele, the vascular tissue of all primary and lateral roots, as well as in leaves, but not in epidermal and cortex cells (Figure 2). As OsPT2 has a low Pi affinity (Figure 5), and expression increases during starvation (Figures 1 and 2), it is reasonable to conclude that OsPT2 probably functions in the transport of Pi from roots to shoots. The relative abundance of expression of OsPT2 at the site of later root development supports this role for the transporter (Figure 2j), as Pi may be mobilized to supply the developing root. The dramatic decrease in total P transport from the roots to the shoots in the OsPT2 KD lines supplied with low levels of Pi, relative to control plants (Figure 6a,c), confirms this hypothesis. However, the noticeable expression in Pi-sufficient leaves (Figures 1 and 2) and mature organs (Figure S2) may suggest other functions for this transporter, although this is likely to depend on leaf age. Interestingly, OsPT6 was also expressed in the stele of Pi-starved younger primary roots and leaves, besides epidermal and cortex cells (Figure 3), suggesting some overlap between OsPT2 and OsPT6 function within rice plants. This overlap of expression pattern may occur when apoplastic concentrations of Pi are low, when OsPT6 can scavenge to maintain supply to the meristem.

Dual functions of OsPT6 in the uptake and translocation of Pi in rice

Under conditions of Pi starvation, the major mechanism for Pi uptake by roots is usually considered to be the high-affinity system (Raghothama, 1999). We observed that the OsPT6 promoter directed expression to the root epidermis, and was strongly induced when grown under Pi starvation (Figure 3), suggesting that this PT is involved in Pi uptake directly from the soil solution. It therefore seems that OsPT6 functions in high-affinity uptake to enable rice to acquire Pi from low soil concentrations. However, OsPT6 was also expressed in almost all cell types, except for the epidermis of younger primary roots, including vascular tissue cells, and generally occurs in tissues thought to contain quite high concentrations of Pi (Figure 3). This pattern of distribution is more characteristic of a low-affinity PT. Similarly, some of the Pht1 transporters in Arabidopsis are not only involved with the uptake of Pi from the soil, but may also be involved with its transfer into the vascular systems (Karthikeyan et al., 2002; Mudge et al., 2002). The lack of expression of OsPT6 in the epidermis of young primary roots (Figure 3e), together with the strong expression of both genes in the vascular tissue where a lateral root emerges (Figures 2j and 3j), suggests that the young root tissues may be strong sinks for Pi, depending upon supply from other parts of the plant.

The high-affinity Pi uptake system in plants operating at low Pi concentrations has an apparent uptake affinity, with Km values ranging from 3 to 10 μm, whereas the low-affinity system operating at high Pi concentrations has a Km value ranging from 50 to 300 μm (Furihata et al., 1992; Nandi et al., 1987). The yeast expression system has been successfully used for characterizing the kinetic properties of many PTs from dicotyledonous plants. We also successfully obtained functional expression of OsPT6, a monocot PT, in MB192 (Figure 4a), a yeast stain lacking endogenous high-affinity PT activity (Bunya et al., 1991). The transporter had a Km, detected by the uptake of isotope 32P, of 97 μm Pi (Figure 4c). Previously, it has been shown that the Km values obtained in the yeast system ranged between 31 and 280 μm for non-mycorrhizal PTs from several dicots (Daram et al., 1998; Leggewie et al., 1997; Rausch et al., 2001), and up to 668 μm for a mycorrhizal specific PT in Medicago (Harrison et al., 2002). Although the data from this heterologous expression system may not truly represent the physiological data in plants, OsPT6 does appear to function as a high-affinity PT. Studies of Arabidopsis gene knock-out plants showed that AtPht1;1 and AtPht1;4, the two most abundant PTs in Pi-starved roots, play significant roles in Pi acquisition from both low- and high-Pi supply (Misson et al., 2004; Shin et al., 2004). Combining the diverse expression patterns both in roots and shoots (Figure 3), and the estimated kinetics (Figure 4c), we suggest that OsPT6 in rice might have similar functions to AtPhT1;1 and AtPhT1;4 in Arabidopsis, playing a broad role in Pi uptake, translocation and internal transport throughout the plant, to enable adaptation to changing levels of P in the soil.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

RT-PCR and quantitative real-time RT-PCR (qRT-PCR) analysis

RNA was extracted from root and shoot tissue samples using TRIzol reagent (Invitrogen, http://www.invitrogen.com). RT-PCR for the target genes, OsPT2 (accession number AF536962) and OsPT6 (accession number AF536966), and an actin (OsRac1, accession number AB047313), using gene-specific primers (Table S1), was performed following the protocol described by Li et al. (2006). qRT-PCR of the three genes (OsPT2, OsPT6 and OsRac), using the primers in Table S1, was carried out in the MyiQTM Single Color Real-Time PCR Detection System (Bio-Rad, http://www.bio-rad.com), and the products were labeled using the SYBR green master mix (SYBR®Premix Ex TagTM II; TaKaRa Bio Inc., http://www.takara-bio.com), according to the method described by Chen et al. (2007).

Generation of expression vectors and transformation of the genes

For the isolation of the OsPT2 and OsPT6 promoters, the regions upstream of the coding regions were PCR amplified from Oryza sativa L. ssp. Japonica cv. Nipponbare genomic DNA, using the primers listed in Table S2. The restriction sites were incorporated in the primers to facilitate cloning into the expression vectors. The amplified fragments were cloned into the pMD19-T vector (TaKaRa), and were confirmed by restriction enzyme digestion and DNA sequencing. After digesting with the pMD19-T vector, the promoter fragments were cloned with the GUS reporter genes into the binary vectors pS1aG-3 (kindly provided by Dr Delhaize, CSIRO Plant Industry, http://www.pi.csiro.au).

To generate the OsPT2/OsPT6 RNAi constructs, a 255-bp fragment of the OsPT2 coding sequence and a 258-bp fragment of the OsPT6 coding sequence were amplified using the specific primers listed in Table S2. The fragments were placed upstream and downstream of a 478-bp rice intron, in opposite directions, under a maize ubiquitin promoter, according to the methods described previously (Wang et al., 2004).

The expression vectors were transferred to Agrobacterium tumefaciens strain EHA105 by electroporation, and were transformed into rice (O. sativa L. ssp. Japonica cv. Nipponbare), as described by Upadhyaya et al. (2000).

Plant growth conditions

The seed sterilization procedure and the nutrient solution for seedling growth in a glasshouse were as described previously (Li et al., 2006). After 10 days of growth, the plants were transferred to nutrient solution, either with 0.3 mm Pi (Pi sufficient) or without the inclusion of Pi (Pi deficient). The solution pH was adjusted to 5.5, and the solution was replaced every 3 days. The plants were grown for a further 3 weeks.

For detecting histochemical localization of the reporter gene in the stamen, caryopsis and seeds, the transgenic rice plants selected by hygromycin were grown in normal soil until the materials were harvested at different reproductive stages.

For determining the total P uptake in roots and shoots of wild-type and RNAi lines, the transgenic plant seeds were germinated and screened in a solution containing 25 mg l−1 hygromycin for 7 days before being transferred to the hydroponics system, as describe above. Both the wild-type and the RNAi seedlings were grown for 3 weeks in the culture solution containing 10 μm Pi until harvest. The plant samples were dried for the chemical analysis of P concentration.

Histochemical localization of GUS expression

The histochemical analysis of GUS activity was examined as described previously (Raghothama et al., 1997). The samples were submersed in GUS reaction mix [0.05 mm sodium phosphate buffer, pH 7.0, 1 mm X-gluc and 0.1% (v/v) Triton X-100], and were incubated at 37°C overnight. Green tissues were destained with ethanol prior to observation. The stained tissues were photographed using an OLYMPUS MVX10 steromicroscope, with a color CCD camera (Olympus, http://www.olympus-global.com). To investigate subcellular expression patterns, the stained tissues were rinsed and fixed in FAA [formalin:acetic acid:70% ethanol (1:1:18)] for 24 h, were embedded in Paraffin and were then sectioned. The sections (15-μm thick) were transferred onto a slide and visualized under an Olympus BX51T stereomicroscope, with a color CCD camera (Olympus).

Yeast manipulations

The yeast Pi uptake-defective mutant MB192 (Bunya et al., 1991), and the expression vector in MB192, p112A1NE (kindly provided by Dr M. Feng, Fudan University, http://www.fudan.edu.cn/englishnew), were used for these experiments. The coding sequence of OsPT6 was cut and cloned into p112A1NE. Yeast strains Yp112-OsPT6, wild-type and Yp112A1NE were grown to the logarithmic phase on YNB medium, and were then harvested and washed in Pi-free medium (YNB). Then, the YNB liquid media containing three different Pi concentrations (0.02, 0.06 and 0.1 mm) were used to incubate the yeast strains at 30°C for 10 h. Bromocresol purple was used as a pH indicator, with a purple to yellow color shift reflecting the acidification of the liquid medium, which correlated with the growth of the yeast cells.

To substantiate the pH dependence of Pi uptake, different extracellular pHs ranging from 4.0 to 8.0 were used at a fixed concentration of 0.08 mm Pi, using 2-(N-morpholino)ethanesulfonic acid (MES) buffer in the YNB medium for the yeast strains Yp112-OsPT6, wild-type and MB192.

The functional assay of OsPT2 in oocytes

mRNA synthesis of OsPT2 cDNA.  The full-length cDNA of OsPT2 was amplified with primers (Table S1) and then subcloned into the BglII and SpeI sites of the oocyte expression vector pT7TS (Cleaver et al., 1996). Briefly, the PCR products were purified by gel extraction, and then digested by BglII and SpeI for 3 h at 37°C. pT7Ts plasmid was digested under the same conditions. Both digested products were purified by gel extraction, and were ligated at 4°C overnight. All of the subclones were sequenced to confirm their authenticity. mRNA synthesis was performed as described previously (Tong et al., 2005).

Oocytes preparation, mRNA injection, Pi uptake and electrophysiology.  Oocyte preparation, mRNA injection and the electrophysiology testing were as described previously (Tong et al., 2005). Small modifications in the oocyte preparation and mRNA injection were followed, as indicated. The collagenase concentration was reduced from 0.03 to 0.05 g ml−1, and oocytes were pre-treated in Ca2+-free ND96 (Liu and Tsay, 2003) for 8–15 min. The treatment time course was dependent on the individual batch of oocytes and the thickness of the cell coat. The treatment was stopped by adding Modified Barth's Medium (MBS) (with Ca2+) as soon as the oocyte was softened. The oocytes were injected then, or after incubation at 18°C overnight. For gene expression in oocytes, 50 ng of mRNA was injected, whereas 50 ng of water was injected as a control. After injection, all oocytes were incubated in MBS, pH 7.4, at 18°C for 3–5 days, before testing for Pi transport.

The Pi uptake of oocytes was measured at 18°C for 14 h in an incubator, adding 0.5 mm NaH2PO4 into the MBS (adjusted to pH 7.4). Between 16 and 20 oocytes were used for each Pi-uptake experiment. At the end of uptake, oocytes were washed in ice-cold MBS, and were then put into 0.25 ml of water for the calibration. All oocytes were lysed using a sonicator for 10 sec, and a 10-fold diluted sample was assayed to measure the Pi accumulated in each cell by the method described by Vanveldhoven and Mannaerts (1987). The membrane potential of the oocytes was recorded by the method described previously (Tong et al., 2005).

Radioactive 32P uptake assay, and measurement of total P concentration in plants

For 32P-uptake experiments, washed and Pi-starved cells were re-suspended in a 3% glucose solution to energize the plasma membrane. The Pi uptake by intact Saccharomyces cerevisiae cells was assayed by the addition of 2 μl of [32P]orthophosphate (0.091 Ci μmol−1; 1 mCi = 37 MBq) to 50-μl aliquots of cells, suspended in 25 mm Tris–succinate (pH 6) solution, and then supplemented with 3% glucose, to give final Pi concentrations varying from 0 to 100 μm, in the presence of 15 mm NaCl. The data were analyzed using the software SigmaPlot (v10.0) to determine the Km value of the OsPT6 protein for Pi uptake.

The seeds of the wild-type and OsPT2/OsPT6 RNAi plants were sterilized, germinated and then screened in a solution of 25 mg l−1 hygromycin for 7 days. After 3 weeks of growth in a 10 μm Pi solution, the seedlings were transferred to 200 ml of Hoagland solution, which contained 8 μCi of KH232PO4 (2 mCi ml−1) and 10 μm of additional KH2PO4 for 24 h, with a day/night cycle of 16/8 h, and with temperature settings of 28/22°C. Plants were rinsed 11 times in nutrient solution until the radioactivity could hardly be detected in the liquid, and were then blot-dried on 3M filter paper. Autoradiographs of the radioactive seedlings were then developed using photographic film plates (Kodak X-omat, Kodak BT size 25.4 × 30.5 cm, Kodak, http://www.kodak.com). The total concentrations of P in the plants were measured as described previously (Chen et al., 2007).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank Prof. Huixia Shou from Zhejiang University, Prof. Nava Moran from Hebrew University of Jerusalem, Prof. Raghothama KG from Purdue University and Prof. Eaton DC from Emory University, for valuable comments on the manuscript. This work was supported by the China 973 Programme (2005CB120903), China National Natural Science Foundation and the Basic Research Programme of Jiangsu Province (BK2005089), 111 project (No. B07030). Rothamsted Research is grant-aided by the Biotechnology and Biological Sciences Research Council (BBSRC) of the UK.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Figure S1. GUS activity both in the roots (a) and shoots (b) driven by the native promoters of OsPT2 and OsPT6 and also in wild type plants treated with removal of Pi supply (-P) and controls (+P) for 21 days.

Figure S2. Histochemical localization of OsPT2 and OsPT6 promoter-GUS expression in stamens, caryopses and germinated seeds. (a–c) OsPT2 promoter-GUS expression; (d–f) OsPT6 promoter-GUS expression.

Figure S3. Growth curves of the yeast wild type (WT), mutant (MB192) and the cells transformed with p112-OsPT2 and p112-OsPT6 cultured in 60 μm Pi medium.

Figure S4. Quantitative real time RT-PCR analysis for knockdown of OsPT2 and OsPT6 expression in the roots (a) and leaves (b) of RNAi transformed lines of rice. A housekeeping gene, Osactin (OsRac1, accession number AB047313), was used as the internal standard.

Figure S5. The concentrations of total P in the roots and shoots of the RNAi-knockdown mutants and their wild type (WT).

Figure S6. The ratio of total P uptake in the shoots to that in the roots in the RNAi-knockdown mutants and wild type (WT).

Table S1. Primers used to amplify the OsPT2 and OsPT6 cDNA.

Table S2. Primers used to generate the OsPT2/OsPT6 expression vector and restriction site sequences are underlined.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
TPJ_3726_sm_figs_S1-S6.pdf88KSupporting info item
TPJ_3726_sm_tables_S1-S2.doc53KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.