A chloroplast phosphate transporter, PHT2;1, influences allocation of phosphate within the plant and phosphate-starvation responses

Abstract

The uptake and distribution of Pi in plants requires multiple Pi transport systems that must function in concert to maintain homeostasis throughout growth and development. The Pi transporter PHT2;1 of Arabidopsis shares similarity with members of the Pi transporter family, which includes Na(+)/Pi symporters of fungal and animal origin and H(+)/Pi symporters of bacterial origin. Sequence comparisons between proteins of this family revealed that plant members possess extended N termini, which share features with chloroplast transit peptides. Localization of a PHT2;1-green fluorescent protein fusion protein indicates that it is present in the chloroplast envelope. A Pi transport function for PHT2;1 was confirmed in yeast using a truncated version of the protein lacking its transit peptide, which allowed targeting to the plasma membrane. To assess the in vivo role of PHT2;1 in phosphorus metabolism, we identified a null mutant, pht2;1-1. Analysis of the mutant reveals that PHT2;1 activity affects Pi allocation within the plant and modulates Pi-starvation responses, including the expression of Pi-starvation response genes and the translocation of Pi within leaves.

Figure 1.

PHT2;1 Expression in Wild-Type Plants. RNA gel blot analysis was performed on 5 μg of RNA isolated from root and leaf tissues of plants either deprived of Pi (−Pi) or grown under control conditions (+Pi) on washed sand. Plants were harvested between 1 and 2 h after the onset of the light phase of the photoperiod (lanes 1 to 4) or at the midpoints of the light and dark phases (lanes 5 and 6, respectively; 10 h of light/14 h of dark). Blots were hybridized with a PHT2;1 cDNA probe, stripped, and hybridized with an 18S rRNA probe to verify equivalent sample loading.

Figure 2.

Expression of PHT2;1 Is Induced in Response to Light. RNA gel blot analysis was performed on 5 μg of total RNA isolated from plants held in complete darkness for 6 days with and without subsequent reexposure to light. Blots were hybridized with PHT2;1 and TPT cDNA probes, stripped, and hybridized with an 18S rRNA probe to verify equivalent sample loading.

Figure 3.

PHT2;1-GFP Fusion Protein Colocalizes with Chloroplasts. A chimeric PHT2;1-GFP construct was introduced into Arabidopsis leaves by particle bombardment, and the localization of fluorescent signals was examined 24 h after transformation. All green signals detected colocalized with red signals. The green and red fluorescent signals indicate GFP and chlorophyll, respectively. Bar in (C) = 50 μm for (A) to (C).

Figure 4.

Subcellular Localization of PHT2;1-GFP and PHT2;1(Δ1-71)-GFP Fusion Proteins in Yeast. PAM2 yeast cells carrying PHT2;1-GFP ([A] to [D]) or PHT2;1(Δ1-71)-GFP ([E] to [H]) constructs were incubated with and without the mitochondrion-selective dye MitoTracker Red. The localization of green and red fluorescent signals, indicating GFP and mitochondria, respectively, was examined by confocal microscopy. Bar in (H) = 5 μm for (A) to (H).

Figure 5.

Transport Properties of PHT2;1 and PHT2;1(Δ1-71). (A) Pi uptake rate of yeast strain PAM2 carrying PHT2;1, PHT2;1(Δ1-71), or a vector control plotted as a function of external Pi concentration at pH 4.0. Values for one of three independent experiments are shown (means ± se for three replicates). (B) Effect of pH on the rate of Pi uptake. Uptake medium contained 25 mM citrate buffer to maintain the indicated pH and 1 mM Pi. Values from one of three independent experiments are shown (means ± se for three replicates). (C) Effect of sodium on the rate of Pi uptake. Uptake medium contained 1 mM Pi and 25 mM citric acid and was titrated to pH 4.0 with either KOH or NaOH. Uptake values for each experiment are pmol Pi/min for 0.5 mL of yeast cell suspension with an absorbance of 1 unit at 600 nm. Values from one of three independent experiments are shown (means ± se for three replicates). (D) Growth of transformed PAM2 cells in synthetic dextrose medium containing 25 mM citrate buffer, pH 4.0, and 0.22 mM Pi. OD₆₀₀ was monitored as a measurement of growth.

Figure 6.

Molecular Characterization of the pht2;1-1 locus. (A) Scheme of the PHT2;1::T-DNA insertion characterized in pht2;1-1. The structure of the pht2;1-1 locus was deduced from PCR, DNA gel blot, and genomic sequence analyses. (B) Agarose gel separation of PCR samples. DNA from the wild type (lane 5) and mutants containing the PHT2;1::T-DNA insertion (lanes 2, 3, 6, and 7) were amplified using the primer combinations indicated. Lanes 6 and 7 show the amplification products from plants hemizygous and homozygous for the PHT2;1::T-DNA insertion, respectively. Lanes 1 and 4 show 1-kb ladder standards. (C) DNA gel blot analysis of BamHI-digested genomic DNA from the wild type (lane 1) and mutants hemizygous (lane 2) and homozygous (lanes 3 and 4) for the T-DNA insertion. Blots were hybridized with probes as indicated. Sizes of the detected bands are indicated in kb. (D) RNA gel blot analysis of total RNA isolated from wild-type (lane 1) and pht2;1-1 (lane 2) leaf tissues hybridized with probes as indicated.

Figure 7.

Effect of Pi Supply on Plant Growth and Pi Accumulation. Roots and rosette leaves from 8 to 10 plants grown under defined conditions were harvested and weighed ([A] and [B]). The Pi content was determined from three replicate samples ([C] and [D]). Values shown are means ± se for two independent experiments. WT, wild type.

Figure 8.

Expression of Pi-Responsive Genes in Wild-Type and pht2;1-1 Mutant Plants Grown under Low-Pi and High-Pi Conditions. Sequential RNA gel blot analyses were performed with 5 μg of total RNA from root and leaf tissues and hybridized with AtPT1, AtPT2, and PAP1 probes (A) and a TPT probe (B). Signal intensities were quantified using a PhosphorImager and normalized to that of the corresponding 18S rRNA. The numbers beneath each gel reflect the ratio (mutant to wild type) of the normalized signal intensities for the corresponding probe. Dashed lines indicate ratios that could not be determined because of signal intensities below the limit of detection. Separate blots were used in (A) and (B), but the samples were derived from the same tissues. The results shown represent one of two independent experiments. −Pi, low Pi; +Pi, high Pi; m, mutant; WT, wild type.

Figure 9.

Pi Retranslocation in Wild-Type and pht2;1-1 Leaves during Pi Deprivation. For each time point, young and old leaves, defined as the eight newest and eight oldest leaves in a rosette, respectively, were harvested from eight plants. The Pi content (nmol Pi/mg fresh weight) was determined from three to five replicate samples. The ratio of young-to-old leaf Pi content is plotted as a function of time after Pi deprivation. Values shown are means ± se. WT, wild type.