A Ras signaling complex controls the RasC-TORC2 pathway and directed cell migration

Abstract

Ras was found to regulate Dictyostelium chemotaxis, but the mechanisms that spatially and temporally control Ras activity during chemotaxis remain largely unknown. We report the discovery of a Ras signaling complex that includes the Ras guanine exchange factor (RasGEF) Aimless, RasGEFH, protein phosphatase 2A (PP2A), and a scaffold designated Sca1. The Sca1/RasGEF/PP2A complex is recruited to the plasma membrane in a chemoattractant- and F-actin-dependent manner and is enriched at the leading edge of chemotaxing cells where it regulates F-actin dynamics and signal relay by controlling the activation of RasC and the downstream target of rapamycin complex 2 (TORC2)-Akt/protein kinase B (PKB) pathway. In addition, PKB and PKB-related PKBR1 phosphorylate Sca1 and regulate the membrane localization of the Sca1/RasGEF/PP2A complex, and thereby RasC activity, in a negative feedback fashion. Thus, our study uncovered a molecular mechanism whereby RasC activity and the spatiotemporal activation of TORC2 are tightly controlled at the leading edge of chemotaxing cells.

Figure 1. Aimless is found in a preformed complex assembled by the scaffold protein Sca1

(A and B) Silver staining of the SDS-PAGE-resolved proteins pulled down with HHF-Aimless, -Aimless^LisH, -RasGEFH, or -Sca1, expressed in their respective null background, from either vegetative (V) or developed cells stimulated or not with 1 µM cAMP for the time indicated. Wild-type cells (WT; AX2) were used as control. The most abundant purified proteins identified by mass spectrometry, and their molecular mass, are indicated. Aimless (70.7 kDa) and RasGEFH (69.7 kDa) normally migrate at a similar molecular mass; addition of the HHF tag underlies the observable shifts (HHF-Aimless, ~73.7 kDa; HHF-RasGEFH, ~72.7 kDa). See also Table S1 and Figure S1A. (C–H) HHF- or V5-tagged Aimless (HHF-Aim, V5-Aim) and RasGEFH (HHF-GEFH, V5-GEFH), their LisH domain mutant forms [Aim^LisH−, Aimless (L79E, E94A); GEFH^LisH−, RasGEFH (F127R, E142A)] as well as myc-Sca1 and T7-PP2A–A were assessed for interaction in co-immunoprecipitation and detected by immunoblotting. IP: Immunoprecipitation, IB: Immunoblot. (I) Sca1 deletion mutants. (J) Pull-downs performed with the Sca1 deletion mutants compared to full-length Sca1. See also Figure S1B. (K) Deduced architecture of the complex. Data are representative of at least 2 independent experiments.

Figure 2. The Sca1 complex regulates aggregation, chemotaxis and vegetative cell motility

(A) Development of the different knockout strains compared to wild-type. Shown pictures were taken at 12h after starvation. Data are representative of 3 independent experiments. See also Figure S2. (B and C) DIAS analysis and traces of representative developed cells performing chemotaxis to cAMP (B) or randomly moving vegetative cells (C). Data represent analysis performed on 10 traces from 3 independent experiments ± SD. Speed refers to the speed of the cell’s centroid movement along the total path; directionality indicates migration straightness; Persistence indicates the persistence of movement in a given direction; direction change refers to the number and frequency of turns; and roundness indicates cell polarity.

Figure 3. The Sca1 complex controls the RasC-TORC2-PKB/PKBR1 pathway

(A) FLAG-RasC was expressed in the indicated strains and cAMP-induced RasC activity assessed following pull-down of active RasC (RasC-GTP) with GST-RBD(Byr2). Pulled-down and total RasC were revealed by FLAG immunoblotting. See also Figure S3 (B) cAMP-induced kinase activity of immunoprecipitated PKB was assessed using H2B as a substrate. H2B phosphorylation was detected by autoradiography and PKB revealed by immunoblotting. (C) Basal (inset; expressed as % of wild-type) and cAMP-induced F-actin polymerization. (D) Total cAMP production in response to stimulation by 10 µM 2’-deoxy-cAMP for the time indicated. (E) Imaging of PHc-GFP in wild-type and scaA⁻ performing chemotaxis to cAMP. (F) Translocation of PHcrac-GFP to the plasma membrane upon uniform cAMP stimulation. The graph depicts the relative fluorescence intensity of membrane-localized PHcrac-GFP as a function of time after cAMP stimulation. Data represent the measured membrane fluorescence intensity of 10 different cells. (G) cAMP-induced phosphorylation of immunoprecipitated PKB at threonine 473 (T^P473), and of PKBR1, from total cell lysates, at threonine 470 (T^P470). Immunoprecipitated PKB was revealed by immunoblotting, and Coomassie blue (CB) staining was used as loading control for total cell lysates. Data are representative, or represent the mean ± SD (C and D) or SEM (F), of at least 3 independent experiments.

Figure 4. Sca1 transiently localizes to the plasma membrane in a chemoattractant and actin-dependent manner, and is negatively regulated by TORC2

(A) Live imaging of GFP-Sca1, expressed in either scaA⁻ or gefH⁻/gefA⁻/scaA⁻, upon uniform cAMP stimulation. Numbers represent time after stimulation in seconds. The relative membrane fluorescence intensity of GFP-Sca1 is shown on the right. (B) Cells were treated with 15 µM LatB or 60 µM LY294002 for 30 min prior cAMP stimulation. (C) cAMP-induced RasC activity was assessed as described in the legend to Figure 3A, following LatB and LY294002 treatments. Quantification of data, expressed as % of the 10 sec time point (max) for the control, is shown on the right. Data represent mean ± SD of 2 independent experiments. (D) Live imaging of GFP-Sca1 expressed in wild-type (WT), piaA⁻, or pikA⁻/pikB⁻/pikC⁻ cells. All imaging data represent mean ± SEM of ≥25 measurements performed on ≥20 cells from 3 independent experiments, and scale bars represent 5 µm. See also Figure S4.

Figure 5. Sca1 is enriched at the leading edge of chemotaxing cells

(A) Live imaging of LY294002-treated scaA⁻ cells expressing GFP-Sca1 or wild-type cells expressing soluble GFP and migrating in an exponential gradient of cAMP delivered by a micropipette. *, Position of the micropipette. (B) scaA⁻ cells expressing GFP-Sca1 or wild-type cells expressing either the RasG-GTP reporter (GFP-Raf1RBD) or soluble GFP, and migrating under agar in a cAMP gradient, were imaged by TIRFM. Signal from the soluble GFP had to be intensified in order to be visualized. Arrows mark regions of enriched fluorescence. Scale bars represent 5 µm.

Figure 6. The function of the Sca1 complex is regulated by TORC2 and PKB/PKBR1 in a negative feedback fashion and by PP2A

(A) Phosphoproteomics analysis allowed identification of Sca1 phosphorylation at serine 359 (S359). Two spectra corresponding to the indicated phosphopeptide were obtained from a sample stimulated for 10 s with cAMP. (B–C) cAMP-induced phosphorylation of immunoprecipitated myc-tagged Sca1 expressed in the indicated strains was assessed by immunoblotting with an anti-phospho-Akt/PKB substrate antibody (αP-PKBS). (D–E) cAMP-induced RasC activity in the indicated strains was assessed as described in the legend to Figure 3A. Expression of HHF-Sca1 and -Sca1Δ5 was controlled by HA immunoblotting. (F) cAMP-induced PKB kinase activity was assessed as described in the legend to Figure 3B. (G) Live imaging of GFP-Sca1Δ5 upon uniform cAMP stimulation. Numbers represents time after stimulation in seconds. Scale bar represents 5 µm. Data are representative of at least 2 independent experiments.

Figure 7. Regulation of RasC signaling during chemotaxis

Chemoattractant stimulation promotes the F-actin-dependent recruitment of the Sca1 complex to the plasma membrane and the subsequent activation of the RasC-TORC2-PKB/PKBR1 pathway at the leading edge of chemotaxing cells. This signaling pathway controls cAMP production and leads to modulation of the F-actin cytoskeleton, thereby regulating signal relay and cell motility, respectively. The RasG-PI3K pathway appears to regulate F-actin, ACA (via CRAC), and PKB independently of the RasC-TORC2 pathway, and other regulators and/or Ras proteins control TORC2 activity in addition to RasC. It is also possible that RasC has other effectors. TORC2 and PKB/PKBR1 regulate RasC activity in a negative feedback fashion that involves regulation of the Sca1 complex’s localization through phosphorylation of Sca1. PP2A seems to be necessary for the function of the Sca1 complex, possibly for the dynamic regulation of the complex by phosphorylation.