A FAM21-containing WASH complex regulates retromer-dependent sorting

Abstract

The Arp2/3 complex regulates endocytosis, sorting, and trafficking, yet the Arp2/3-stimulating factors orchestrating these distinct events remain ill defined. WASH (Wiskott-Aldrich Syndrome Protein and SCAR Homolog) is an Arp2/3 activator with unknown function that was duplicated during primate evolution. We demonstrate that WASH associates with tubulin and localizes to early endosomal subdomains, which are enriched in Arp2/3, F-actin, and retromer components. Although WASH localized with activated receptors, it was not essential for endocytosis. However, WASH did regulate retromer-mediated retrograde CI-MPR trafficking, which required its association with endosomes, Arp2/3-directed F-actin regulation, and tubulin interaction. Moreover, WASH exists in a multiprotein complex containing FAM21, which links WASH to endosomes and is required for WASH-dependent retromer-mediated sorting. Significantly, without WASH, retromer tubulation was exaggerated, supporting a model wherein WASH links retromer-mediated cargo containing tubules to microtubules for Golgi-directed trafficking and generates F-actin-driven force for tubule scission.

Figure 1. WASH localizes along microtubules with F-actin-rich endosomes through its N-terminus

(A) Endogenous WASH localization (green) in HeLa, Jurkat, and U-87MG cells observed by immunofluorescence (DNA staining with Hoechst 33342, blue). (B) Localization of transiently expressed low-level YFP-WASH in HeLa (left panel) and Jurkat cells (middle panel) under basal conditions or when Jurkat were spread on anti-TCR coated coverslips (right panel). Live cells were analyzed. (C) Localization of low-level YFP-WASH N-terminal deletion mutants in living U-87MG cells. (D–L) Jurkat cells, either resting (D, G, H, and I) or conjugated with superantigen-pulsed/CMAC-stained Raji B cells (blue) (E, F, and J–L), were stained as indicated and imaged. Arrows in D and E indicate MTOC position. APC, antigen presenting cell.

Figure 2. WASH co-localizes with early endosomes during EGFR endocytosis

(A–D) U-87MG cells stained with anti-WASH (green), anti-EEA1 (red) and Hoechst 33342 (blue). (B) Distinct points of WASH accumulation on early endosomes are highlighted; squares denote points at which WASH is accumulated as two distal spots on EEA1-stained rings (magnified in C). Circles denote single WASH puncta with smaller endosomes (magnified in D). (E) HeLa cells stained with anti-WASH (green), anti-LAMP1 (red) and Hoechst 33342 (blue). (F–H) U-87MG cells were stimulated with rhodamine-EGF (Rh-EGF; red) and stained with anti-WASH (green), anti-EEA1 (purple) and Hoechst 33342 (blue). (G) Early WASH and Rh-EGF co-localization is magnified. (I and J) Low-level YFP-WASH (green) transfected U-87MG cells were stimulated with Rh-EGF (red) and analyzed by live cell imaging.

Figure 3. WASH interacts directly with tubulin, but is not involved in endocytosis

(A) Jurkat cells were stimulated with anti-TCR over time and then lysed, immunoprecipitated with control IgG (NRS) or anti-WASH, and immunoblotted as indicated. (B) Jurkat cells were incubated on ice with anti-TCR and then cross-linked with TRITC-labeled secondary (red) at 37°C to induce TCR internalization, immediately fixed, co-stained with anti-WASH (green) and imaged. (C) WASH GST-fusion proteins were used to co-precipitate αtubulin from Jurkat lysate. (D) GST-WASH-WHD2 was tested for direct binding to purified α/βtubulin. In (C) and (D), fusion protein loading was visualized with Coomassie. (E) EGFR internalization was analyzed by surface biotinylation and streptavidin-based precipitation in EGF-stimulated control and suppressed HeLa cells (cyclohexamide, CHX). Lysates were also immunoblotted as indicated. (F) HeLa cells were transfected as indicated (green), stimulated with Rh-EGF (red) for 15 minutes, fixed, and stained with Hoechst 33342 (blue).

Figure 4. Retrograde transport of CI-MPR is WASH- and Arp2/3-dependent

(A) U-87MG cells were stained with anti-WASH (green), anti-SNX1 (red) and Hoechst 33342 (blue). (C and D) HeLa cells were transfected as indicated (green) and stained with anti-CI-MPR (red) and Hoechst 33342 (blue). Pseudocoloring of CI-MPR staining demonstrates pixel intensity. (C) Transfectants were scored for either compact or dispersed CI-MPR distributions. Bars represent mean ±StDev from four independent experiments (*P <0.01 and **P <0.005).

Figure 5. A FAM21-containing WASH complex associates with the retromer

(A) Jurkat cell lysates were immunoprecipitated with rabbit IgG (NRS), anti-WASH, or anti-FAM21 and immunoblotted as indicated. (B) HA-YFP and HA-YFP-SNX2 were transiently expressed in HeLa cells, immunoprecipitated with anti-HA, and immunoblotted as indicated. (C) YFP-WASH-expressing HeLa cells were stained with anti-FAM21. (D) Inset displays co-localization. (E) WASH mutants used in shWASH/HA-YFP-WASH rescue vectors. (F) HeLa cells were transfected as indicated, and reconstituted HA-YFP-WASH proteins were analyzed for FAM21 association via immunoprecipitation. (G) FAM21 mutants used in shFAM21/HA-YFP-FAM21 rescue vectors. The FAM21C sequence used in this study corresponds to GenBank accession NM_015262 plus the indicated FAM21A inserts after AA623 and AA935 making the FAM21C coding sequence 1341 amino acids. The FAM21 antibody was generated as indicated. (H) HeLa cells were transfected as indicated, and reconstituted FAM21 mutants were analyzed for WASH association via immunoprecipitation. In (F) and (H), lysates were also immunoblotted to verify suppression and re-expression. Black asterisks denote reconstituted FAM21 proteins and red asterisks indicate FAM21 mutants unrecognized by anti-FAM21. n.s., non-specific; degrad., degradation; endog., endogenous FAM21. (I and J) Jurkat cells were suppressed, lysed and immunoblotted as indicated.

Figure 6. WASH and FAM21 collaboratively regulate retromer-mediated sorting of CI-MPR

(A) HeLa cells were transfected as indicated (green), fixed, and stained with anti-CI-MPR (red) and Hoechst 33342 (blue). (B) HeLa cells were transfected as indicated (green) and stained as in (A). Pseudocoloring of CI-MPR staining demonstrates pixel intensity. (C) Transfectants were scored for either compact or dispersed CI-MPR distributions. Bars represent mean ±StDev from four independent experiments (all P values <0.05).

Figure 7. WASH depletion leads to exaggerated retromer-mediated tubulation

(A and B) HeLa cells were transfected with YFP-VPS29-expressing control or shWASH/YFP-VPS29 dual suppression/expression vectors (here YFP-VPS29 is a marker of suppression). (C and D) HeLa cells were transfected as indicated (green), incubated at 37°C with antibody to the luminal domain of CI-MPR for the indicated times, fixed, and stained to monitor CI-MPR internalization. (E and F) Compact juxtanuclear accumulation of internalized CI-MPR and enrichment of uptake CI-MPR on tubular structures was quantified as indicated and bars represent mean ±StDev from three independent experiments (all P values <0.05). (A–D and G) Arrows indicate distinct tubulation events between endosomes. (H) The functional architecture of WASH, and a proposed model for how WASH and FAM21 cooperatively regulate retromer-mediated trafficking. See Discussion for details. FAM21-mediated Endosomal Localization (FEL) domain; Tubulin Binding Region (TBR); Proline-Rich (PR); Verprolin, Connecting and Acidic (VCA).