Reciprocal regulation of TORC signaling and tRNA modifications by Elongator enforces nutrient-dependent cell fate

Abstract

Nutrient availability has a profound impact on cell fate. Upon nitrogen starvation, wild-type fission yeast cells uncouple cell growth from cell division to generate small, round-shaped cells that are competent for sexual differentiation. The TORC1 (TOR complex 1) and TORC2 complexes exert opposite controls on cell growth and cell differentiation, but little is known about how their activity is coordinated. We show that transfer RNA (tRNA) modifications by Elongator are critical for this regulation by promoting the translation of both key components of TORC2 and repressors of TORC1. We further identified the TORC2 pathway as an activator of Elongator by down-regulating a Gsk3 (glycogen synthase kinase 3)-dependent inhibitory phosphorylation of Elongator. Therefore, a feedback control is operating between TOR complex (TORC) signaling and tRNA modification by Elongator to enforce the advancement of mitosis that precedes cell differentiation.

Fig. 1. The Elongator complex regulates the expression of components of TORC1/TORC2 signaling pathways.

(A) A schematic of the TORC1 and TORC2 signaling pathways in fission yeast. The quality and availability of the nitrogen source control the balance between TORC1 that enforces growth and TORC2 that enforces cell differentiation. (B) Top: Average expression level of the proteome in the elp3-deleted strain compared to wild type set to 100%. Average expression level within the GO functional group (cellular response to starvation) was significantly more affected than the average in the absence of Elongator (*P < 0.05; **P < 0.01; Student’s t test). Bottom: Codon usage for lysine (AAA-AAG), within the GO functional group (cellular response to starvation) compared to the average ORFeome lysine (AAA-AAG) codon usage. Note that the 0 to 20% and 40 to 100% of the y axis are not shown to highlight variations from the ORFeome codon usage. (*P < 0.05; **P < 0.01; ***P < 0.001; Student’s t test). wt, wild type. (C) Hierarchical clustering and heat map analysis of the expression level within the GO functional group (cellular response to starvation) in the ctu1 and elp3 deletion strains compared to wild type. The data are presented as log 2 mutant/wild-type ratio of immunoblotting signals and are color-coded as indicated in the key.

Fig. 2. The Elongator complex regulates the balance between TORC1 and TORC2 activities.

(A) Western blot analysis of the level of expression of Ste20-TAP (Tandem Affinity Purification) (Rictor in TORC1), Mip1-HA (Raptor in TORC2), and Tsc2 in the indicated strains showing the specific down-regulation of Ste20 and Tsc2 when Elongator is inactivated and the rescue when the tRNA^LYS_UUU is overexpressed. Tubulin is used as a loading control. (B) Growth assay of the indicated strains in the presence or absence of rapamycin. The tor1 S1834E and tor2 S1837E are rapamycin-resistant alleles. nmt1-tor2 is overexpressing the genes from the nmt1 promoter. EMM, Edinburgh minimal medium. (C) Growth assay of the indicated strains in the presence or absence of canavanine. The can1-50 allele harbors a mutation of the canavanine transporter Can1, resulting in resistance. nmt1-tor1 and nmt1-tor2 are overexpressing the genes from the nmt1 promoter. (D) Growth assay of the indicated prototrophic or auxotrophic strains grown on minimal media supplemented with adenine and uracil.

Fig. 3. The Elongator complex is required for the nutrient-dependent shift between the TORC1 and TORC2 complexes.

(A) The percentage of dividing cells was determined during a shift from an efficient (glutamate) and a poor (proline) nitrogen source in wild-type or Elongator-deficient strains. A glutamate-to-glutamate shift is shown as control. Each column represents the averaged value ± SEM (n = 3). Right: Averaged values for clarity. Note that an elp4 mutant was used, as Elp4 will be the focus of the rest of the manuscript, but that the deletion of any subunit of Elongator results in the inactivation of the whole complex. (B) The percentage of dividing cells was determined after addition of rapamycin that mimics the mitotic advancement following a nutritional shift. The indicated strains including rapamycin-resistant alleles of tor1 and tor2 were used. Each column represents the averaged value ± SEM (n = 3). Right: Averaged values for clarity. (C) The percentage of dividing cells was determined after a shift from a poor (proline) to an efficient (glutamate) nitrogen source in wild-type or Elongator-deficient strains. A proline-to-proline shift is shown as control. Each column represents the averaged value ± SEM (n = 3). Right: Averaged values for clarity.

Fig. 4. The TORC1/TORC2 pathway controls the level of the Elongator-dependent modifications.

(A) Left: Hierachical clustering analysis of changes in the relative levels of a set of tRNA ribonucleoside modifications in mutants of genes required for the synthesis of wobble s² (ctu1) and mcm⁵ (elp3) uridine moieties. The bottom color bar displays the range of log 2 fold change values. Right: Similar to that in the left, except that all detectable Elongator-dependent modified ribonucleosides are shown and the following conditions were applied: treatment with rapamycin, overexpression of tor1 or tor2, deletion of the trm9 gene encoding the enzyme required for the methylation of the cm⁵ moiety of the mcm⁵U and mcm⁵s²U modifications. (B) Phosphorylation assay of peptides corresponding to the amino acids 107 to 120 of fission yeast Elp4 harboring the indicated mutations or phosphorylation (S118^P) using immunoprecipitated Gsk3-flag. (C) Phosphorylation assay of endogenous immunoprecipitated Elp4-TAP protein (wt or S114A as indicated) mixed with immunoprecipitated Gsk3-flag before the kinase assay. The band corresponding to Elp4 (40 kDa) is indicated by an arrow. (D) Western blot controlling the level of Gsk3 or Elp4 from the experiment shown in (C). Half of the immunoprecipitation (IP) products were used. Note that the Elp4 S114A mutant is expressed at higher level (see below).

Fig. 5. The S114A mutant is the hyperactive version of Elp4.

(A) A schematic of the disrupted allele retaining the first 91 amino acids of Elp4. bp, base pair. (B) Growth assay of the elp4 disruption strain transformed with plasmids expressing the indicated wild-type or elp4 mutants or an empty vector (ctr) in the presence or absence of rapamycin. (C) Western blot analysis of the level of expression of the indicated elp4 mutants. Tubulin is used as a loading control. (D) Western blot analyses of the phosphorylation status of Gad8 and Psk1 in a wild-type strain or the Elp4 S114A mutant. Tubulin is used as a loading control. *Note that the anti–phospho-Gad8 antibody also recognizes Elp4 because of the presence of an immunoglobulin G binding domain within the TAP tag. (E) Iodine staining and microscopic observations of the sterility in the indicated homothallic elp4-deleted strain transformed with plasmids expressing the indicated wild type, elp4 S114A mutant, or an empty vector (ctr). Clones were grown on low-nitrogen medium for 48 hours. The dark iodine staining indicates the presence of gametes (asci) as confirmed by 4′,6-diamidino-2-phenylindole (DAPI) staining. The mating efficiency is indicated as percentage.

Fig. 6. The balance between TORC1 and TORC2 pathways is regulated by Elongator.

(A) Cluster analysis visualization of changes in the relative levels of all detectable Elongator-dependent modified ribonucleosides in the indicated strains. The bottom color bar displays the range of log 2 fold change values. (B) Growth assay of the indicated strain in the presence of rapamycin/caffeine (top), rapamycin (bottom), or vehicle. DMSO, dimethyl sulfoxide. (C) Phosphorylation level of the TORC1 substrate Psk1 and the TORC2 substrate Gad8 during nitrogen starvation. Cells were filtered and resuspended in a medium deprived of nitrogen at time 0 and analyzed for 90 min in the indicated strains. Each point represents the averaged normalized value ± SEM (n = 3). (D) Phosphorylation level of the TORC1 substrate Psk1 and the TORC2 substrate Gad8 after addition of nitrogen on cells starved for 2 hours. Cells were filtered and resuspended in a medium deprived of nitrogen for 2 hours. Nitrogen (NH₄CL) was added back at time 0, and indicated strains were analyzed for 60 min. Each point represents the averaged normalized value ± SEM (n = 3).