A PP2A-B55-Mediated Crosstalk between TORC1 and TORC2 Regulates the Differentiation Response in Fission Yeast

Abstract

Extracellular cues regulate cell fate, and this is mainly achieved through the engagement of specific transcriptional programs. The TORC1 and TORC2 complexes mediate the integration of nutritional cues to cellular behavior, but their interplay is poorly understood. Here, we use fission yeast to investigate how phosphatase activity participates in this interplay during the switch from proliferation to sexual differentiation. We find that loss of PP2A-B55^Pab1 enhances the expression of differentiation-specific genes and leads to premature conjugation. pab1 deletion brings about a transcriptional profile similar to TORC1 inactivation, and deletion of pab1 overcomes the repression of differentiation genes in cells overexpressing TORC1. Importantly, we show that this effect is mediated by an increased TORC2-AKT (Gad8) signaling. Under nutrient-rich conditions, PP2A-B55^Pab1 dephosphorylates Gad8 Ser546, repressing its activity. Conversely, TORC1 inactivation upon starvation leads to the inactivation of PP2A-B55^Pab1 through the Greatwall-Endosulfin pathway. This results in the activation of Gad8 and the commitment to differentiation. Thus, PP2A-B55^Pab1 enables a crosstalk between the two TOR complexes that controls cell-fate decisions in response to nutrient availability.

Keywords: B55; Gad8; PP2A; S. pombe; TORC1; TORC2; nitrogen starvation; sexual differentiation.

Graphical abstract

Figure 1

Deletion of pab1 Results in an Exacerbated Mating Response (A) Homothallic WT, pab1Δ, par1Δ, and dis2Δ cells were incubated at 25°C in the absence of nitrogen, and their mating ability was determined at 0, 4, 6, 8, 24, and 48 hr. (B) DIC pictures of homothallic WT and pab1Δ cells in the absence of nitrogen for 6 hr. (C) FACS analysis of the DNA content of homothallic WT and pab1Δ cells treated as in (A). (D) Phosphorylation of the mating pheromone responsive MAPK Spk1 was used as a measure of the pheromone signaling in homothallic WT and pab1Δ cells starved for nitrogen. Cdc2 (PSTAIR) served as a loading control. (E) mRNA expression of mei2 in homothallic WT and pab1Δ cells that had been incubated for 4 hr in EMM (control) or EMM-N. Expression is relative to actin and was determined by qPCR. Mean and SEM of three biological replicates are shown. See also Figure S1.

Figure 2

Deletion of pab1 Leads to Major Transcriptional Changes (A) Volcano plot graph showing pab1Δ transcriptional profile in comparison to WT in nitrogen-rich conditions, the genes upregulated with an adjusted p value <0.01 and fold change >3 are highlighted. In the right panel, enrichment analysis of genes upregulated in pab1Δ by GO Biological Process and Gene expression is shown, (p = adjusted p value). (B) Venn diagram illustrating the overlaps between pab1Δ-upregulated genes in nitrogen-rich conditions, WT-induced genes upon 4 hr of nitrogen starvation, and genes upregulated in a tor2-ts6 strain at restrictive temperature. In both cases, there is significant overlap with a p value = 3.83e^–25 when comparing to the gene set induced in the WT strain in response to nitrogen starvation, and a p value = 1.75e^–37 when comparing to the gene set induced in the tor2-ts6 strain. In the right panel, enrichment analysis of the genes shared in pab1Δ and tor2-ts6 by GO Biological Process and Gene expression (p = adjusted p value). (C) Venn diagram showing the overlaps between pab1Δ-upregulated genes in nitrogen-rich conditions, WT-induced genes, and tor1Δ-induced genes upon 4 hr of nitrogen starvation. See also Figure S2.

Figure 3

PP2A-B55^Pab1 Functions Downstream of TORC1 Preventing Sexual Differentiation (A) Homothallic WT, pab1Δ, nmt1-tor2, and nmt1-tor2 pab1Δ cells that had been grown in the absence of thiamine for 18 hr (to induce the overexpression of tor2 in the nmt1-tor2 strains) were incubated at 25°C in the absence of nitrogen, and their mating ability was determined at 0, 4, 8, 24, and 48 hr. (B) mRNA expression of mei2 in cells from (A) after incubation for 4 hr in EMM (control) or EMM-N. Expression is relative to actin and was determined by qPCR. Mean and SEM of three biological replicates are shown. (C) Phosphorylation of the mating pheromone responsive MAPK Spk1 was used as a measure of the pheromone signaling in homothallic WT, pab1Δ, nmt1-tor2, and nmt1-tor2 pab1Δ cells (grown as in A) starved for nitrogen. Phosphorylated Psk1 was used as a measure of TORC1 activity. Cdc2 (PSTAIR) served as a loading control. (D) Homothallic WT, pab1Δ, tsc2Δ, and tsc2Δ pab1Δ cells were incubated at 25°C in the absence of nitrogen and their mating ability was determined at 0, 4, 8, 24, and 48 hr. (E) mRNA expression of mei2 in cells from (D) after incubation for 4 hr in EMM (control) or EMM-N. Expression is relative to actin and was determined by qPCR. Mean and SEM of three biological replicates are shown. (F) Phosphorylation of the mating pheromone responsive MAPK Spk1 was used as a measure of the pheromone signaling in cells in (D). Phosphorylation of Psk1 served as a readout for TORC1 activity. Cdc2 (PSTAIR) served as a loading control. (G) mRNA expression of mei2 in heterothallic WT, pab1Δ, tsc2Δ, and tsc2Δ pab1Δ cells treated as in (D) after incubation for 4 hr in EMM (control) or EMM-N. Expression is relative to actin and was determined by qPCR. Mean and SEM of three biological replicates are shown. See also Figure S3.

Figure 4

Nitrogen Starvation Leads to an Increased TORC2-Gad8 Signaling that Depends on the Inactivation of TORC1 and PP2A-B55^Pab1 (A) Homothallic WT and pab1Δ cells were incubated at 25°C in the absence of nitrogen, and samples were collected at the indicated time points. Phosphorylation of Gad8 at Ser546 was followed over the time course by western blot. Total Gad8 and Cdc2 (PSTAIR) served as loading controls. (B) Homothallic WT and nmt1-pab1 cells were grown in the absence of thiamine for 18 hr prior to shifting them to minimal medium without nitrogen (EMM-N). Samples were collected at the indicated time points and phosphorylation of Gad8 at Ser546 was monitored by western blot. Phosphorylation of Psk1 served as readout of TORC1 activity and total Gad8 and Cdc2 (PSTAIR) served as loading controls. (C–E) Control cells (containing the auxin inducible degron background, Padh15-skp1-At-Tir1-2NLS-Padh15-sk1-Os-Tir1) and nmt41-3PK-miniAID-pab1 cells in the same genetic background (referred to as AID-pab1) were treated with thiamine (0.5 μM) and NAA (0.5 mM), and samples were collected at the indicated time points. (C) Gad8 phosphorylation at Ser546 was monitored by western blot. Depletion of B55^Pab1 was detected by western blot against its N-terminal 3PK tag. Total Gad8 and Cdc2 (PSTAIR) served as loading controls. (D) Mock-treated and thiamine and NAA-treated cells were stained with Calcofluor to visualize the septa. (E) mRNA expression of mei2 after 6 hr in the presence of thiamine and NAA. Expression is relative to actin and was determined by qPCR. Mean and SEM of three biological replicates are shown. (F and G) nmt41-3PK-miniAID-pab1 and nmt41-3PK-miniAID-pab1 cdc10-V50 cells were incubated at 36°C and treated with thiamine (0.5 μM) and NAA (0.5 mM) for 6 hr or mock treated. Control cells were kept at 25°C for the same time. (F) Gad8 phosphorylation at Ser546 was monitored by western blot. Depletion of B55^Pab1 was detected by western blot against its N-terminal 3PK tag. Cdc13 was used to monitor the cell-cycle arrest. Total Gad8 and Cdc2 (PSTAIR) served as loading controls. (G) mRNA expression of mei2 in cells from (F). Expression is relative to actin and was determined by qPCR. Mean and SEM of three biological replicates are shown. (H and I) tor2-51, tor2-51 igo1Δ, WT, and wee1-50 cells were grown at 25°C (permissive temperature) and then shifted to 36°C in order to inactivate TORC1 and Wee1. Samples were collected at the indicated time points. For comparability between strains, the samples corresponding to the tor2-51 mutant and the WT strain were run in the two gels. (H) Gad8 phosphorylation at Ser546 and Igo1 phosphorylation at Ser64 (P-ENSA) were detected by western blot. Psk1 phosphorylation was used as readout of TORC1 activity. Total Gad8 and Cdc2 (PSTAIR) served as loading controls. The asterisk in the Gad8 panels indicates remaining signal form the phosphorylated Psk1 western blot. (I) mRNA expression of mei2 at permissive temperature and after incubation for 4 hr at 36°C. Expression is relative to actin and was determined by qPCR. Mean and SEM of three biological replicates are shown. See also Figure S4.

Figure 5

Enhanced TORC2-Gad8 Signaling Contributes to the Premature Differentiation Response in pab1Δ Cells (A and B) tor2-51, tor2-51 igo1Δ, and tor2-51 igo1Δ ryh1QL were incubated at 25°C (permissive temperature) or 32°C for 4 hr in order to inactivate TORC1. (A) Gad8 phosphorylation at Ser546 and Igo1 phosphorylation at Ser64 (P-ENSA) were detected by western blot. Psk1 phosphorylation was used as readout of TORC1 activity. Total Gad8 and Cdc2 (PSTAIR) served as loading controls. (B) mRNA expression of mei2 at permissive temperature and after incubation for 4 and 6 hr at 32°C. Expression is relative to actin and was determined by qPCR. Mean and SEM of three biological replicates are shown. (C and D) Homothallic WT and pab1Δ cells were incubated at 25°C in the absence of nitrogen and in the presence of Torin1 (25 μM) or in its absence (treated with DMSO instead), and samples were collected at the indicated time points. (C) phosphorylation of Gad8 at Ser546 was monitored by western blot. Phosphorylation of Psk1 served as readout of TORC1 activity. Total Gad8 and Cdc2 (PSTAIR) served as loading controls. (D) The mating counting indicates that Torin1 completely abrogates the mating ability of both WT and pab1Δ cells. (E and F) Homothallic WT, pab1Δ, and pab1Δ gad8Ser546Ala cells grown in EMM were shifted to medium devoid of nitrogen (EMM-N) at 25°C and samples were collected at the indicated time points. (E) mating counting indicates that mutation of Gad8 Ser546Ala completely abrogates the exacerbated conjugation of pab1Δ cells. (F) mRNA expression of mei2 before and upon shift to EMM-N. Expression is relative to actin and was determined by qPCR. Mean and SEM of three biological replicates are shown. (G) Homothallic WT, pab1Δ, fkh2Ser321Ala, and pab1Δ fkh2Ser321Ala cells were incubated at 25°C in the absence of nitrogen, and their mating ability was determined at the indicated time points. The mating counting indicates that mutation of Fkh2 Ser321 to Ala reduces the enhanced mating phenotype of pab1Δ cells. See also Figure S5.

Figure 6

PP2A-B55^Pab1 Interacts with Tor1 and Gad8 and Directly Dephosphorylates Gad8 at Ser546 (A) N-terminally TAP-tagged B55^Pab1 was purified from cells expressing flag-tagged versions of Tor1 (3flag-Tor1) and Gad8 (Gad8-5flag). Western blot against the flag tag was used to detect Gad8 and Tor1 that co-purify with Pab1. A strain expressing 3flag-Tor1 and Gad8-5flag was used as a negative control in the TAP-purification. B55^Pab1 was monitored by western blot against calmodulin-binding peptide (CBP). (B) nmt41-3PK-miniAID-pab1 sin1-13myc gad8-5flag cells and nmt41-3PK-miniAID-pab1 sin1-13myc cells (all containing the auxin inducible degron background) were mock treated or treated with thiamine (15 μM) and NAA (0.5 mM) for 5 hr in order to deplete B55^Pab1. Gad8 was immunoprecipitated via its C-terminal 5flag tag, and the interaction between Sin1 and Gad8 was subsequently assessed by western blot against the flag tag and the myc tag. (C) nmt1-3HA-pab1 sin1-13myc gad8-5flag cells and nmt1-3HA-pab1 sin1-13myc cells were grown in the absence of thiamine for 18 hr in order to induce the overexpression of pab1. Gad8 was immunoprecipitated via its C-terminal 5flag tag, and the interaction between Sin1 and Gad8 was subsequently assessed by western blot against the flag tag and the myc tag. Pab1 overexpression was monitored by western blot against the 3HA tag. (D) PP2A- B55^Pab1 purified from par1Δ cells via an N-terminal TAP tag in B55^Pab1 was used in a phosphatase assay using as substrate immunopurified Gad8-5flag. Phosphorylation of Gad8 at Ser546 in the reactions was monitored by western blot. Okadaic acid (1 and 10 nM) was added in two of the reactions in order to show that the dephosphorylation observed was PP2A dependent. Immunoprecipitated Gad8 Ser546Ala was also used as control for the non-phosphorylated form of Gad8. Western blots against the flag tag (Gad8), the CBP tag (B55^Pab1), and against Ppa2 (the major catalytic subunit of the complex) served as controls for the purifications and showed the even presence of each component in the reactions. (E) Gad8-5flag or Gad8S546A-5flag was immunopurified from WT and pab1Δ cells and used in a kinase assay against a recombinant fragment of Fkh2 (Gln291-Pro370) fused to GST. Gad8K259R-5flag (kinase dead) was used as negative control. Phosphorylation of this fragment was determined by western blot using an antibody that recognizes the phosphorylated AKT consensus sequence (anti-PAS) and was used as readout of Gad8 activity. Western blots against the flag tag (Gad8) and the GST tag (recombinant Fkh2) served as controls for the purifications. See also Figure S6.

Figure 7

Working Model PP2A-B55^Pab1 represses the activity of the module TORC2-Gad8 by directly reverting the phosphorylation of Gad8 at Ser546. In nitrogen-rich conditions, this mechanism prevents untimely expression of mating genes. Upon nitrogen starvation, TORC1 inactivation leads to the inhibition of PP2A-B55^Pab1, through a mechanism that depends on the Greatwall-Endosulfin pathway (Ppk18-Igo1). Consequently, PP2A-B55^Pab1 can no longer exert its negative effect on the TORC2-Gad8 module, which results in the increased phosphorylation of Gad8 at Ser546 and its enhanced activity toward Fkh2Ser321. Phosphorylation of Fkh2 (and probably other unknown targets of Gad8) leads to the induction of the expression of mating genes. Other unknown targets of PP2A-B55^Pab1 might also participate in the transcriptional regulation of mating genes and sexual differentiation. The relative importance of Gad8 compared to other targets cannot be evaluated with current data. This effect is aided by other events also regulated by TORC1, e.g., cell shortening and G1 arrest (see main text for details).