Phosphoinositide 5-phosphatase Fig 4p is required for both acute rise and subsequent fall in stress-induced phosphatidylinositol 3,5-bisphosphate levels

Abstract

Phosphoinositide lipids regulate complex events via the recruitment of proteins to a specialized region of the membrane at a specific time. Precise control of both the synthesis and turnover of phosphoinositide lipids is integral to membrane trafficking, signal transduction, and cytoskeletal rearrangements. Little is known about the acute regulation of the levels of these signaling lipids. When Saccharomyces cerevisiae cells are treated with hyperosmotic medium the levels of phosphatidylinositol 3,5-bisphosphate (PI3,5P(2)) increase 20-fold. Here we show that this 20-fold increase is rapid and occurs within 5 min. Surprisingly, these elevated levels are transient. Fifteen minutes following hyperosmotic shock they decrease at a rapid rate, even though the cells remain in hyperosmotic medium. In parallel with the rapid increase in the levels of PI3,5P(2), vacuole volume decreases rapidly. Furthermore, concomitant with a return to basal levels of PI3,5P(2) vacuole volume is restored. We show that Fig 4p, consistent with its proposed role as a PI3,5P(2) 5-phosphatase, is required in vivo for this rapid return to basal levels of PI3,5P(2). Surprisingly, we find that Fig 4p is also required for the hyperosmotic shock-induced increase in PI3,5P(2) levels. These findings demonstrate that following hyperosmotic shock, large, transient changes occur in the levels of PI3,5P(2) and further suggest that Fig 4p is important in regulating both the acute rise and subsequent fall in PI3,5P(2) levels.

FIG. 1.

Both PI3,5P₂ levels and vacuole volume transiently change after hyperosmotic shock. (A) Wild-type (LWY7235) cells were labeled with [³H]inositol for 12 h and then treated with NaCl for the indicated times. Total cellular phosphatidylinositol was extracted, deacylated, and analyzed by high-pressure liquid chromatography. To normalize each data point for number of cells and [³H]inositol incorporation, each value is shown as the percent of total [³H]phosphatidylinositol extracted. Each data point is an average of at least two independent experiments. (B) Wild-type cells were labeled with FM4-64 dye to visualize vacuole volume and number of vacuole lobes. Cells were then treated with NaCl for the indicated times and viewed by fluorescence microscopy. Representative fields are shown. (C) Vacuole membranes undergo scission as vacuole volume is reduced following hyperosmotic shock. Mutant cells (fig4Δ) which have basal and hyperosmotic shock-induced PI3,5P₂ levels that are lower than normal and thus larger vacuoles were labeled with FM4-64 dye to visualize vacuole volume and number of vacuole lobes. Cells were then treated with NaCl for 10 min and viewed by fluorescence microscopy. Representative fields are shown.

FIG. 2.

PI4P and PI4,5P₂ levels change transiently in response to hyperosmotic shock. Total phosphatidylinositol was extracted from wild-type (LWY7235) cells labeled with [³H]inositol as described in the legend for Fig. 1. The percentages of PI4P and PI4,5P₂ in the cells after hyperosmotic stimulation are shown here. Each data point is an average of at least two independent experiments.

FIG. 3.

Fab1p and its putative activators localize on the vacuole membrane. (A) Wild-type (LWY7235), chromosomal HA-FAB1 (LWY6830), and fig4Δ cells expressing FIG4-HA were lysed in the absence of detergent and extracts were separated into membrane (P) and soluble (S) fractions by centrifugation at 13,000 × g. The P13 fraction contains large organelles such as the vacuole, Golgi apparatus, endoplasmic reticulum, and nucleus. Western blot analysis revealed the levels of the proteins in each fraction. The same strains were also treated with NaCl for 10 min prior to lysis. Data are representative of three independent experiments.(B) Vac14p colocalizes with the vacuole membrane. Wild-type (LWY7235) cells were fixed with formaldehyde and probed with anti-Vac14p and anti-Vph1p sera. (C) Fig4p-HA colocalizes with the vacuole membrane. fig4Δ cells expressing FIG4-HA were treated as described above and probed with anti-Vac8p serum and anti-HA monoclonal antibody.

FIG. 4.

Vac7p and Vac14p play a role in PI3,5P₂ synthesis and possibly turnover. (A to C) fab1Δ (LWY2055), vac7Δ (LWY2054), and vac14Δ (LWY5177) cells were labeled with [³H]inositol for 12 h and total phosphatidylinositol was extracted. PI3P and PI3,5P₂ levels are shown for each strain after exposure to 0.9 M NaCl for the indicated times. These levels are superimposed over wild-type PI3P and PI3,5P₂ levels (dotted lines). Each data point is an average of at least two independent experiments. (D) Data from panel C rescaled to show that the low level of PI3,5P₂ does not decrease after 30 min.

FIG. 5.

Fig4p is required for both PI3,5P₂ synthesis and turnover. (A) PI3,5P₂ levels were determined for wild-type (LWY7235) and fig4Δ (LWY6474) cells. The data are averages of three independent experiments. (B) Wild-type and fig4Δ cells were labeled with FM4-64 to visualize vacuole membranes. Images were captured by fluorescence microscopy. Representative fields are shown. (C) Time course of PI3P and PI3,5P₂ level changes in fig4Δ cells after exposure to NaCl. These levels are superimposed over the PI3P and PI3,5P₂ levels in the wild type (dotted lines). Each data point is an average of at least two independent experiments. (D) Vac14p is partially mislocalized in fig4Δ and fab1Δ cells. Cells were lysed under nondetergent conditions and extracts were spun at 13,000 × g to determine the extent of membrane distribution. Protein levels were determined using Western blot analysis. The data are representative of three independent experiments.

FIG. 6.

Model for the roles of the Vac14p/Fig4p complex in both synthesis and turnover of PI3,5P₂. Hyperosmotic shock increases the levels of PI3P and PI3,5P₂. The PI3,5P₂ increase is dramatically compromised in the absence of Fig4p, Vac7p, or Vac14p, demonstrating that all three proteins are important for Fab1p activation. Turnover of PI3,5P₂ following hyperosmotic stimulation is compromised in the absence of Fig4p and Vac14p. Turnover of PI3P in the absence of any Fab1p activator or Fab1p is normal.