Genome-wide analysis reveals the vacuolar pH-stat of Saccharomyces cerevisiae

Abstract

Protons, the smallest and most ubiquitous of ions, are central to physiological processes. Transmembrane proton gradients drive ATP synthesis, metabolite transport, receptor recycling and vesicle trafficking, while compartmental pH controls enzyme function. Despite this fundamental importance, the mechanisms underlying pH homeostasis are not entirely accounted for in any organelle or organism. We undertook a genome-wide survey of vacuole pH (pH(v)) in 4,606 single-gene deletion mutants of Saccharomyces cerevisiae under control, acid and alkali stress conditions to reveal the vacuolar pH-stat. Median pH(v) (5.27±0.13) was resistant to acid stress (5.28±0.14) but shifted significantly in response to alkali stress (5.83±0.13). Of 107 mutants that displayed aberrant pH(v) under more than one external pH condition, functional categories of transporters, membrane biogenesis and trafficking machinery were significantly enriched. Phospholipid flippases, encoded by the family of P4-type ATPases, emerged as pH regulators, as did the yeast ortholog of Niemann Pick Type C protein, implicated in sterol trafficking. An independent genetic screen revealed that correction of pH(v) dysregulation in a neo1(ts) mutant restored viability whereas cholesterol accumulation in human NPC1(-/-) fibroblasts diminished upon treatment with a proton ionophore. Furthermore, while it is established that lumenal pH affects trafficking, this study revealed a reciprocal link with many mutants defective in anterograde pathways being hyperacidic and retrograde pathway mutants with alkaline vacuoles. In these and other examples, pH perturbations emerge as a hitherto unrecognized phenotype that may contribute to the cellular basis of disease and offer potential therapeutic intervention through pH modulation.

Figure 1. A genome-wide screen for mutations that disrupt vacuole pH in yeast.

(a) Distribution of vacuole pH values at external pH 4.0. Top panel, a histogram showing the distribution of vacuole pH values observed in the ResGen collection of single gene knockout strains (n = 4606). Outliers (middle panel) fall beyond the distribution of vacuole pH values measured in wild type cells (bottom panel, n = 46). (b) Cumulative probability plots showing the distribution of vacuole pH values measured in the mutant collection at external pH 2.7 (acidic conditions, n = 4469; red), pH 4.0 (control conditions, n = 4606; black) or pH 7.0 (alkaline conditions, n = 4593; blue). Median values are shown for each population (indicated by the horizontal line). (c) Vacuole pH values from the mutant collection grown under acidic conditions (top) or alkaline conditions (bottom) were compared to values measured under control conditions. Minimum (red lines) and maximum (blue lines) pH values observed in the wild type population (light grey points) are indicated for each condition. Mutants with abnormally acidic (red) or alkaline (blue) vacuole under both conditions are highlighted; black closed circles indicate mutants with normal vacuole pH values, black open circles are mutants with abnormal vacuole pH values observed under a single condition. (d) Venn diagram displaying the distribution of outliers identified under all three conditions. ∼20% of the mutants displayed either hyperacidic vacuoles (red, n = 77) or alkaline vacuoles (blue; n = 27) in at least two growth conditions with different pH; these are listed in Table S3.

Figure 2. P4-ATPase flippases contribute to vacuole pH regulation.

(a) Vacuolar pH is defective in P4-ATPase mutants. Vacuolar pH was measured in wild type BY4742 cells, strains lacking one or more of the P4-ATPases (drs2Δ, dnf1Δ, dnf2Δ, dnf3Δ) or with a temperature sensitive allele of the essential NEO1 gene (neo1-1) without or with (+NEO1) a plasmid containing wild type NEO1. The neo1 strains were shifted to the nonpermisive temperature for 1 hour (34°C). Data represent average of duplicates and one of three independent experiments. (b) Overexpression of VMA genes suppresses the neo1-1 mutant growth defect. The neo1-1 strain was transformed with an empty vector, a single copy VMA11, a multi-copy VMA11, or multi-copy VMA5 plasmids. Serial dilutions of the transformants were spotted on YPD plates and tested at 27°C and 34°C. (c) Inhibition of V-ATPase function suppresses neo1-ts growth defect at nonpermissive temperature. Serial dilution of wild type or two neo1-ts mutant cultures were tested on YPD plates containing 5 µM bafilomycin A1. (d) Compartmental hyperacidification in neo1-1 or triple dnf1,2,3Δ mutants. Micrographs show quinacrine staining of wild type (wt), neo1-1 or dnf1,2,3Δ triple mutants at permissive (27°C) and nonpermissive (37°C) temperatures. Fluorescence intensity (a.u., arbitrary units) was quantified by flow cytometry in the absence or presence of 5 µM bafilomycin A1 (baf A). Negative controls include drs2Δ cells. 30,000 cells were analyzed for each strain.

Figure 3. Lysosomal hyperacidification accompanies a cellular Niemann-Pick type C phenotype.

(a) Vacuoles in ergosterol accumulating ncr1Δ cells are hyperacidic, whereas erg mutants lacking ergosterol have abnormally alkaline vacuoles. (b) Intracellular sterol accumulates in ncr1Δ yeast. Yeast cells were stained with filipin to detect sterol in the presence or absence of NCR1. Bar, 2 µm. (c) Fibroblasts harboring pathogenic alleles of NPC1 (P237S/I1061T) have hyperacidified lysosomes. Wild type cells from a control subject (wt) or mutant cells from a patient with Niemann-Pick type C disease (NPC1^−/−) were stained with acridine orange in the absence or presence of 10 nM concanamycin A1 (concan A), an inhibitor of the V-ATPase. Bar, 10 µm. Cellular acridine orange fluorescence was quantified relative to untreated wild type cells; bars represent standard deviations (n = 100 cells). (d) Nigericin prevents aberrant accumulation of cholesterol in lysosomes of NPC1^−/− mutant cells. Cells were stained with filipin to detect cholesterol in the presence or absence of 10 nM nigericin, a K⁺/H⁺ ionophore.

Figure 4. Transporter dysfunction in vps36Δ cells.

(a) Nhx1 is mislocalized in vps36Δ cells. Micrographs show the location of Nhx1-GFP in nhx1Δ or vps36Δ cells transformed with the 2 µ plasmid pRin82 containing NHX1:GFP behind its endogenous promoter. Bar, 2 µm. The number of Nhx1-GFP stained puncta per cell was quantified; >100 cells were analyzed for each strain. (b) vps36Δ phenocopies nhx1Δ. Growth of wild type BY4742 (wt), nhx1Δ or vps36Δ cultures was measured after 19 hrs at 30°C under control conditions (APG, pH 4.0), acid stress (pH 2.7), or in the presence of 10 µg/ml hygromycin B (Hygro B), 1.5 M KCl or 1.5 M NaCl. Bars represent mean +/− S.E.M. (c) The pH homeostasis defect in nhx1Δ is downstream of Vps36 function. Micrographs show the cellular location of Ste3-GFP in wild type BY4742 (wt), nhx1Δ or vps36Δ cells grown in APG pH 4.0 medium at 30°C with or without methylamine (MA, 10 mM) treatment. Note that pH amelioration fails to correct trafficking dysfunction in vps36Δ cells. Bar, 2 µm.