VTC- and 5-IP7-dependent polyP synthesis by isolated vacuoles

Vacuoles were isolated from logarithmic cultures of BY4742 wildtype cells (WT) or from isogenic vtc4Δ cells strains. They were incubated in polyphosphate synthesis assays without (A) or (B) in presence of 50 µM 5-InsP7. At indicated times, aliquots were withdrawn, solubilized in Triton X-100 and polyphosphate was quantified through the polyP-dependent fluorescence of added DAPI. Means ± SD (standard deviation) of three independent experiments are shown. * p<0.05, ** p<0.005 for a paired t-test comparing WT with and without 5-InsP7 at each time point.

VTC-dependent accumulation of Pi in isolated vacuoles

Vacuoles were isolated from the indicated, logarithmically growing BY4742 (WT) cells. PolyP overproduction was achieved either through (A, B) overexpressing VTC5 from the strong GPD promotor, or (C,D) by expressing the hyperactivating vtc3k126A and vtc4K129A alleles form their native promotors as the sole source of these two proteins. The vacuoles were incubated with an ATP-regenerating system and 50 µM 5-InsP7 under conditions allowing polyP synthesis. At indicated times, the vacuoles were solubilized with Triton X-100 and Pi was assayed through malachite green. Graphs represent the mean ± SEM of three independent experiments for each strain. * p<0.05, ** p<0.005, *** p<0.001 from a paired t-test comparing each time WT vs VTC5 overexpression or vtc3k126A vtc4K129A.

Effect of Pi on polyphosphatase activity.

A. Pi titration. Vacuoles were isolated from logarithmically growing wildtype cells (BY4742). The organelles were diluted in polyphosphatase assay reaction buffer, which contained 0.1% Triton X-100 and hence liberated the luminal polyphosphatases. This lysate was incubated with 30 μM polyP300 as a substrate and supplemented with 1 mM ZnCl2 and the indicated concentrations of K-Pi pH 6.8. After the indicated times of incubation, the remaining polyP was quantified through DAPI. The DAPI signal at the beginning of the incubation served as 100% reference. Graphs represent the means ± SEM of three independent experiments. B, C. Differentiation of Pi effects on Ppn1 and Ppn2. Vacuoles were isolated from the indicated, logarithmically growing strains, lysed and used in polyP degradation assays with 300 µM polyP300 as in (A). These assays were performed in the presence or absence of 30 mM K-Pi pH 6.8 and 1 mM MgCl2 (B) or, instead of this, with 1 mM ZnCl2 (C) as cation supporting catalytic activity.

Accumulation of Pi in isolated vacuoles

VTC- and Pho91-dependence. Vacuoles were isolated from the indicated, logarithmically growing strains. The purified organelles were incubated as in Fig. 1, i.e. in a buffer with an ATP-regenerating system that allows the synthesis of polyP, and either (A) without (w/o) or (B) in the presence of 50 µM 5-IP7. After the indicated periods of incubation at 27°C, an 80 µl aliquot was withdrawn, the vacuoles were sedimented by centrifugation, washed and lysed. Released vacuolar Pi was determined by malachite green assay. Graphs represent the mean ± SEM of three independent experiments. p<0.05 from a paired t-test comparing each time WT versus pho91Δ.

PolyP accumulation in pho91 mutant vacuoles.

Vacuoles from BY4742 cells (WT) and isogenic pho91Δ cells were isolated and incubated under conditions supporting polyP synthesis and Pi accumulation as in Fig. 4, in the absence (w/o) or presence of 50 µM 5-InsP7. At the indicated timepoints, aliquots were withdrawn, the vacuoles were lysed in detergent, and polyP was assayed through DAPI fluorescence. Graphs show the means and SEM from three independent experiments. p>0.05 from a paired t-test for all differences between WT and pho91Δ.

Impact of the vacuolar polyP/Pi cycle on cytosolic Pi signalling

A) Illustration of nucleo-cytoplasmic relocation of Pho4-GFP in response to Pi availability. Wildtype yeast cells were grown in SC medium under Pi replete conditions. During exponential phase (OD600nm=1), cells were transferred for 30 min to synthetic complete media with 200 µM phosphate (-Pi) or 7.5 mM Pi (+Pi) and imaged by fluorescence microscopy. B) The indicated yeast strains were logarithmically grown over night in SC medium with 7.5 mM Pi,, harvested at OD600nm=1 and immediately imaged by fluorescence microscopy as in A. The graph shows the means and SEM of the percentage of cells showing Pho4 predominantly in the nucleus. n=3 independent experiments with 200 cells quantified per sample. ** p<0.005, *** p<0.001, ****p<0.0001 from a paired t-test comparing WT with the mutants.

Impact of Pho91 and vacuolar polyphosphatases on InsPP levels.

pho91Δ and ppn1Δppn2Δ cells, as well as their isogenic wildtypes, were logarithmically grown in Pi-replete SC medium as in Fig. 6. At OD600nm=1 the cells were extracted with perchloric acid as previously described (Wilson et al, 2015) and analysed for the indicated InsPs through capillary electrophoresis coupled to mass spectrometry (CE-MS) as described (Qiu et al, 2023). n=3. * p<0.05 from a paired t-test comparing WT with the mutants.

Working model of acidocalcisome-like vacuoles as Pi buffering systems.

(A) Under Pi-replete conditions, ATP drives the conversion of Pi into polyP and its translocation into the organelle. Here, polyP is degraded by the vacuolar polyphosphatases Ppn1 and Ppn2 to establish a vacuolar pool of free Pi. Feedback inhibition of Pi gradually reduces polyP degradation, enabling the buildup of a vacuolar polyP stock. Red lines and SPX colouring indicate inhibitory action, green colouring stimulation. (B) Cytosolic Pi scarcity decreases InsPP levels, which triggers two compensatory, SPX-controlled effects: The transfer of Pi from the cytosol into vacuoles through VTC ceases; and Pho91-dependent export of Pi from vacuoles is activated. Both measures synergize to stabilize cytosolic Pi. The export of Pi from the vacuole lifts product inhibition on the polyphosphatases Ppn1 and Ppn2 and stimulates a compensatory degradation of polyP.

Isolated vacuoles during polyP synthesis

BJ 4742 (WT) vacuoles were incubated in vitro as in Fig. 1B. After 10 min, 0.1μM of the fluorophore FM4-64 was added to stain the membranes. The suspension was immediately imaged on a Yokogawa spinning disc microscope with a Gataca LiveSR unit and a 100x 1.45 NA objective. Stacks were taken at 0.2 µM z-distance. The image shows an optical section out of such a stack. To determine vacuole diameters, the stacks were subjected to maximum intensity projection in ImageJ. The largest visible diameter for 100 organelles was measured manually using ImageJ, yielding an average diameter of 0.8 µm with an SD of 0.2 µm Scale bar: 1 µm

Effects of N- and C-terminal fluorescent protein tags on the localization of Pho91.

PHO91 was fused with a variety of N- or C-terminal protein tags and peptide spacers as indicated. They were expressed from the endogenous PHO91 promotor or, where indicated, from the overexpressing ADH1 or GPD1 promotors. Cells were logarithmically grown overnight in SC medium, harvested at OD600nm of 1-2, and analysed by fluorescence microscopy.

Pho91 cannot replace other Pi transporters to support growth of yeast.

We generated a BY4741 strain with a quintuple deletion of the genes for the known Pi transporters, PHO84, PHO87, PHO89, PHO90 and PHO91 (D5m). These cells were kept alive by expressing the gene for the plasma membrane Pi transporter PHO89 from a URA3-based centromeric (single copy) plasmid (pRS416). PHO91 was expressed from a HIS3-based centromeric plasmid (pRS315). Cells were plated in a dilution series on SC lacking histidine (SC-HIS) to verify that the cells had the HIS3-based PHO91 plasmid, or on SC with 5-fluoro-orotic acid (5-FOA), a drug that forces cells to lose the URA3-based pRS416 and thus to live without PHO89. PHO91 as the sole Pi transporter gene (on SC + 5-FOA) does not allow cells to grow.