The Vam6 and Gtr1-Gtr2 pathway activates TORC1 in response to amino acids in fission yeast

Abstract

The Rag family of GTPases has been implicated in the TORC1 activation in Drosophila and in mammalian cells in response to amino acids. We have investigated the role of the Rag GTPases Gtr1 and Gtr2 in TORC1 regulation in Schizosaccharomyces pombe. Fission yeast Gtr1 and Gtr2 are non-essential proteins that enhance cell growth in the presence of amino acids in the medium. The function of Gtr1 and Gtr2 in nutrient signaling is further supported by the observation that even in rich medium the deletion of either gene results in the promotion of mating, meiosis and sporulation, consistent with the downregulation of TORC1. We show that Gtr1 and Gtr2 colocalize with TORC1 in vacuoles, where TORC1 is presumably activated. Epistasis analyses indicated that Gtr1 and Gtr2 function downstream of Vam6 and upstream of TORC1 in response to amino acid signals. Our data demonstrate the existence of an evolutionarily conserved pathway with the Vam6 and Gtr1-Gtr2 pathway activating TORC1, which in turns stimulates cell growth and inhibits sexual differentiation.

Fig. 1.

Role of Gtr1 and Gtr2 in cell growth and sexual differentiation. (A) Wild-type (wt), gtr1Δ and gtr2Δ cells were grown in Edinburgh minimal medium (EMM) at 30°C, and then transferred into fresh EMM and EMM supplemented with amino acids. At the same time, leu1-32, gtr1Δ leu1-32 and gtr2Δ leu1-32 cells were grown in EMM supplemented with leucine. This experiment was performed three times and the number of cells per ml was counted every 2 hours to calculate the doubling time. (B) gtr1Δ and gtr2Δ cells of opposite mating types were incubated on YES plates for 2 days at 25°C. The gtr1Δ and gtr2Δ mutant cells were able to mate and form spores in rich medium. Scale bar: 10 μm. (C) Mating efficiency in rich medium was determined by counting the number of zygotes observed in YES plates.

Fig. 2.

Subcellular localization of Gtr1 and Gtr2. (A) gtr1-gfp and gtr2-rfp cells were grown in EMM at 30°C in the presence and in the absence of amino acids. Gtr1–GFP and Gtr2–RFP localized to structures similar to the vacuolar membranes, independently of the presence of amino acids. (B) gtr1-gfp cells were grown in EMM at 30°C, and FM4-64 staining was used to identify vacuoles and corroborate that Gtr1–GFP localized to the vacuole membranes. (C) Gtr1–GFP and Gtr2–RFP colocalized to the vacuolar surface in the presence and in the absence of amino acids. (D) gtr1-flag, gtr2-rfp and gtr1-flag gtr2-rfp cells were grown in EMM at 30°C and transferred into EMM in the presence and in the absence of amino acids. Exponentially growing cells were collected after 1 hour and cell lysates were pulled down using anti-Flag M2 beads. Gtr1–Flag physically interacted with Gtr2–RFP and this interaction increased in the presence of amino acids. The relative interaction between Gtr1–Flag and Gtr2–RFP was estimated by densitometry using ImageJ software and normalized to the amount of Gtr2–RFP pulled down with anti-Flag antibodies in the presence of amino acids, which was set as 1. Scale bars: 10 μm.

Fig. 3.

Subcellular localization of TORC1. gfp-tor2, pop3-gfp and gfp-mip1 cells were grown in EMM containing thiamine (EMMT) at 30°C and transferred into EMMT with (A) and without amino acids (B). Cells were stained with FM4-64 to observe the vacuolar membranes. The three subunits of the TORC1 complex (GFP–Tor2, Pop3–GFP and GFP–Mip1) localized to the vacuolar membranes. Scale bars: 10 μm.

Fig. 4.

TORC1 and Gtr1 interact in vivo. (A) Wild-type (wt), gtr1Δ, nmt-tor2 and nmt-tor2 gtr1Δ cells of opposite mating types were grown on YES plates at 25°C for 2 days. Overexpression of Tor2 rescued the mating derepression observed in gtr1Δ cells. (B) Double-mutant gfp-tor2 gtr1-rfp cells were grown EMM containing thiamine (EMMT) at 30°C and transferred into fresh EMMT, EMMT supplemented with amino acids and EMMT without nitrogen for 1 hour. GFP–Tor2 and Gtr1–RFP colocalized on the vacuolar membrane, regardless of the presence of amino acids or nitrogen in the medium. Scale bar: 10 μm. (C) mip1-myc, gtr1-flag and mip1-myc gtr1-flag cells were grown in EMM and transferred into EMM in the presence or absence of amino acids for 1 hour. Cell lysates (Input) and Flag pull-down fractions were subjected to SDS-PAGE and immunoblots were incubated with anti-Myc and anti-Flag antibodies, as indicated. Gtr1–Flag interacted physically with Mip1–Myc (the Raptor ortholog) in an amino-acid-dependent manner. The relative interaction between Gtr1–Flag and Mip1–Myc was estimated by densitometry using ImageJ software and normalized to the Mip1–Myc pulled down in the presence of amino acids, which was set as 1. (D) Wild-type, gtr1Δ and gtr2Δ cells were grown in EMM and transferred into EMM with and without amino acids for 90 minutes. Cell extracts were subjected to SDS-PAGE and immunoblotted to detect phosphorylated Rps6 (P-Rps6) as a measurement of TORC1 activity. The relative amount of phosphorylated Rps6 was estimated by densitometry and normalized to tubulin expression. The amount of phosphorylated Rps6 in wild-type cells growing in the presence of amino acids was set as 1. (E) rps601Δ myc-rps602 cells were grown in the presence and in the absence of amino acids. The amount of Myc–Rps6 was examined with the anti-Myc antibody. Rps6 phosphorylation (Rps602-P) was used as a measurement of TORC1 activity. Tubulin was used as a loading control.

Fig. 5.

vam6Δ cells show a defect in cell growth. (A) Wild-type (wt) and vam6Δ mutant cells were grown exponentially in EMM, in EMM plus amino acids and in EMM plus leucine. Cell numbers were counted every 2 hours to establish the doubling time in each medium. Wild-type, vam6Δ, gtr1Q61L and vam6Δ gtr1Q61L cells were grown on EMMT (EMM supplemented with thiamine) plates (B) and on plates supplemented with Phloxine B (C) in the presence of thiamine to induce expression of low levels of gtr1Q61L. vam6Δ cells formed small colonies (B) containing a high number of dead cells, generating dark red colonies (Phloxine B positive) (C). The viability of vam6Δ cells expressing the active form of Gtr1 (gtr1Q61L) increased; these cells formed larger colonies (B and C) and the number of dead cells decreased (C). (D) Wild-type, gtr1Δ, gtr2Δ, vam6Δ, gtr1Q61L and vam6Δ gtr1Q61L cells were grown in EMMT (i.e. in the presence of thiamine) at 30°C and membrane vacuoles were stained with FM6-64. vam6Δ cells showed a defect in membrane fusion; FM4-64 was unable to stain membrane vacuoles in vam6Δ cells, and the fluorescence appeared as small dots corresponding to endosomal vesicles. gtr1Δ, gtr2Δ or gtr1Q61L cells did not show this phenotype and FM4-64 was able to stain the vacuolar membrane correctly. The same defect of vam6Δ cells in membrane fusion was observed in vam6Δ gtr1Q61L, indicating that the active form of Gtr1 is unable to restore the vacuolar structure of vam6Δ. (E) Wild-type, vam6Δ, gtr1Q61L and vam6Δ gtr1Q61L cells were grown at 30°C in EMMT and transferred to EMMT in the presence and in the absence of amino acids (supplemented with thiamine in both cases). Cells were collected after 90 minutes and the extracts were subjected to SDS-PAGE and immunoblotting to detect Rps6 phosphorylation (P-Rps6) as a measurement of TORC1 activity. The relative amount of phosphorylated Rps6 was estimated by densitometry and normalized to tubulin expression. The amount of phosphorylated Rps6 in vam6Δ gtr1Q61L cells growing in the presence of amino acids was normalized to 1. In the presence of amino acids the amount of phosphorylated Rps6 increased in wild-type cells. The vam6 mutant was unable to respond to amino acids and the activity of TORC1 was low. By contrast, in cells expressing gtr1Q61L (the active form of Gtr1) Rps6 phosphorylation increased. The mutant vam6 showed high Rps6 phosphorylation when Gtr1Q61L was expressed, indicating that the constitutively active form of Gtr1 is able to restore TORC1 activity in vam6Δ cells. Scale bars: 1 cm (B); 1 mm (C); 10 μm (D).

Fig. 6.

Vam6 and Gtr1 interact in vivo. (A) gfp-vam6 cells were grown in EMM with thiamine at 30°C and vacuolar membranes were stained with FM4-64. GFP–Vam6 localized to the vacuolar membranes in growing cells. (B) gfp-vam6 gtr1-rfp cells were grown at 30°C in EMM and transferred to EMM, EMM supplemented with amino acids and EMM without nitrogen (always in the presence of thiamine). GFP–Vam6 and Gtr1–RPF colocalized to the vacuolar membranes in all conditions. (C) gtr1-flag, vam6-tap and gtr1-flag vam6-tap cells were grown in EMM and transferred into EMM supplemented with amino acids. Cells were collected after 20 and 40 minutes. Cell lysates (Input) and TAP pulldown fractions were subjected to SDS-PAGE and immunoblots were incubated with anti-TAP and anti-Flag antibodies, as indicated. The interaction between Gtr1–Flag and Vam6–TAP was estimated by densitometry using ImageJ software and the Gtr1–Flag pulled down in the presence of amino acids was normalized to 1. Gtr1–Flag physically interacted with Vam6–TAP in an amino-acid-dependent manner. Scale bars: 10 μm.

Fig. 7.

Hypothetical model showing how Vam6, Gtr1–Gtr2 and TORC1 respond to amino acids. Amino acids activate Tor2 by means of Vam6 and the Rag proteins to induce translation and promote cell growth. In the presence of amino acids, Gtr1 interacts strongly with Gtr2 and this heterodimer binds Vam6. Vam6 would activate Gtr1 and this, in turn, would bind to the TORC1 complex by means of Mip1. The active TORC1 complex induces Rps6 phosphorylation and, hence, induces cell growth and inhibit sexual differentiation. This activation would occur in the vacuolar membrane, where all the components would interact strongly in an amino-acid-dependent manner.