Phosphate (Pi) starvation effect on the cytosolic Pi concentration and Pi exchanges across the tonoplast in plant cells: an in vivo 31P-nuclear magnetic resonance study using methylphosphonate as a Pi analog

Abstract

In vivo (31)P-NMR analyses showed that the phosphate (Pi) concentration in the cytosol of sycamore (Acer pseudoplatanus) and Arabidopsis (Arabidopsis thaliana) cells was much lower than the cytoplasmic Pi concentrations usually considered (60-80 mum instead of >1 mm) and that it dropped very rapidly following the onset of Pi starvation. The Pi efflux from the vacuole was insufficient to compensate for the absence of external Pi supply, suggesting that the drop of cytosolic Pi might be the first endogenous signal triggering the Pi starvation rescue metabolism. Successive short sequences of Pi supply and deprivation showed that added Pi transiently accumulated in the cytosol, then in the stroma and matrix of organelles bounded by two membranes (plastids and mitochondria, respectively), and subsequently in the vacuole. The Pi analog methylphosphonate (MeP) was used to analyze Pi exchanges across the tonoplast. MeP incorporated into cells via the Pi carrier of the plasma membrane; it accumulated massively in the cytosol and prevented Pi efflux from the vacuole. This blocking of vacuolar Pi efflux was confirmed by in vitro assays with purified vacuoles. Subsequent incorporation of Pi into the cells triggered a massive transfer of MeP from the cytosol to the vacuole. Mechanisms for Pi exchanges across the tonoplast are discussed in the light of the low cytosolic Pi level, the cell response to Pi starvation, and the Pi/MeP interactive effects.

Figure 1.

In vivo proton-decoupled ³¹P-NMR spectra of sycamore (A) and Arabidopsis (B) culture cells. Experimental conditions were as follows. Cells cultivated in their respective culture medium supplied with Pi were harvested 5 d after subculture and placed in the NMR tube as described in “Materials and Methods.” They were perfused with a NM containing 50 μm Pi at pH 6.0 and 20°C. Insets show enlarged portions of full spectra centered on cyt-Pi: A1 and B1, standard perfusion conditions; A2, sycamore cells are acidified by the addition of 2 mm propionic acid to NM adjusted at pH 6.4; B2, Arabidopsis cells are illuminated as described in “Material and Methods.” Acquisition time was 1 h (6,000 scans). Peak assignments are as follows: ref, reference (methylenediphosphonate) used to measure chemical shifts and for quantifications; glc-6-P, Glc-6-P; cyt-Pi, cytoplasmic Pi; vac-Pi, vacuolar Pi; NTP, nucleoside triphosphate; UDP-glc, uridine-5′-diphosphate-α-d-Glc. org_mp-Pi and cyt_sol-Pi correspond to the cyt-Pi pools present in the stroma and matrix of plastids and mitochondria and in the cytosol (see text).

Figure 2.

Rapid exchanges of Pi between the cytosol and the organelles (mitochondria and plastids) in Pi-starved sycamore cells transiently supplied with Pi. Experimental conditions were as follows. Prior to NMR analyses, cells were incubated over 5 d in a Pi-free NM in order to decrease Pi and P-compound pools below the threshold of in vivo ³¹P-NMR detection (approximately 20 nmol g⁻¹). A, Portions of in vivo ³¹P-NMR spectra (expanded scale) showing the increase of the cyt_sol-Pi peak shortly after the addition of 50 μm Pi into NM and its decrease with the symmetrical org_mp-Pi increase upon rinsing cells with a Pi-free NM 15 min later. B, Corresponding curves showing the filling of cytosol after the addition of Pi to perfusion NM, and the symmetrical time course changes of cyt_sol-Pi and org_mp-Pi following the cell rinsing with a Pi-free NM. Acquisition time was 4.5 min (450 scans). Values are means ± sd (n = 10).

Figure 3.

Time-course evolution of the pools of Pi and Glc-6-P in sycamore cells during successive short sequences of Pi supply and starvation. Experimental conditions were as follows. Prior to NMR analyses, cells were incubated as indicated in the legend of Figure 2. At time zero, 50 μm Pi was added to NM. Afterward, cells were perfused alternatively with a Pi-free or with a Pi-supplied NM as indicated. Acquisition time was 4.5 min (450 scans). Abbreviations are as in Figure 1. This experiment was repeated five times. For the sake of readability, values are not given as means ± sd but are from one representative experiment.

Figure 4.

Time-course evolution of the pools of Pi, NTP, and Glc-6-P in sycamore cells following the onset of Pi starvation. Experimental conditions were as follows. Standard cells harvested 5 d after subculture were used. Cells were analyzed by in vivo ³¹P-NMR as described in the legend of Figure 1. At time zero, cells were perfused with a Pi-free NM. Acquisition time was 2 h (12,000 scans). Abbreviations are the same as in Figure 1. Values are from a representative experiment chosen among five.

Figure 5.

Exchanges of MeP between cytosol, organelles (mitochondria and plastids), and vacuole in sycamore cells. Experimental conditions were as follows. Prior to NMR analyses, cells were incubated as indicated in the legend of Figure 2. At time zero, 200 μm MeP was added in the Pi-free NM; at 15 h, cells were rinsed with a MeP-free NM and afterward perfused with a NM containing 50 μm Pi. A, Portions of in vivo ³¹P-NMR spectra (expanded scale) showing the signals corresponding to different MeP pools: spectra 1, 2, and 3 were assigned to times 1.5, 7 h, and 15 h after MeP addition (at the beginning of spectrum 3 acquisition, ext-MeP was rinsed); spectra 4 and 5 were assigned to times 1.5 and 4 h after subsequent cell perfusion with a MeP-free and Pi-supplied NM. B, Curves showing the filling of cyt_sol-MeP, org_mp-MeP, and vac-MeP pools during the incubation with MeP and the drop of cyt_sol-MeP and symmetrical increase of vac-MeP after the subsequent cell perfusion with the MeP-free and Pi-supplied NM. Acquisition time was 15 min (1,500 scans); note that variable delays separate acquisitions on the curves. ext-MeP, cyt_sol-MeP, org_mp-MeP, and vac-MeP represent MeP pools present in external medium, cytosol, organelles, and vacuole, respectively. Values are from a representative experiment chosen among five.

Figure 6.

In vivo proton-decoupled ³¹P-NMR spectra of sycamore cells incubated in a Pi-free NM containing or not containing MeP. Experimental conditions were as follows. Prior to NMR analyses, standard cells were incubated 4 d in a Pi-free NM (A) or in a Pi-free NM supplied with 200 μm MeP (B). During NMR acquisition, cells were perfused with a MeP-free NM at pH 6.0. Acquisition time was 1 h (6,000 scans). Abbreviations are the same as in Figures 1 and 5. This experiment was repeated with various incubation times (data not shown), showing that in the presence of MeP vac-Pi remained stable, contrary to what was observed in the absence of MeP, where it became undetectable after 2 d, and P-compounds became hardly detectable after only 1 d of incubation.

Figure 7.

Proton-decoupled ³¹P-NMR spectra showing the efflux of Pi from isolated sycamore cell vacuoles. Vacuoles were isolated from standard cells as indicated in “Materials and Methods.” Experimental conditions were as follows. For each condition, 0.5 mL of the thick purified vacuole suspension was diluted in 1.5 mL of buffer A adjusted to pH 7.5. Each sample (2 mL) was analyzed using 10-mm NMR tubes and a 10-mm probe. Acquisition time was 1 h (1,000 scans). A, Freshly prepared vacuoles (time zero). B, Vacuoles kept for 12 h in buffer A. C, Vacuoles kept for 12 h in buffer A containing 1 mm MeP. This experiment was repeated three times.