Dia1 and IQGAP1 interact in cell migration and phagocytic cup formation

Abstract

The Diaphanous-related formin Dia1 nucleates actin polymerization, thereby regulating cell shape and motility. Mechanisms that control the cellular location of Dia1 to spatially define actin polymerization are largely unknown. In this study, we identify the cytoskeletal scaffold protein IQGAP1 as a Dia1-binding protein that is necessary for its subcellular location. IQGAP1 interacts with Dia1 through a region within the Diaphanous inhibitory domain after the RhoA-mediated release of Dia1 autoinhibition. Both proteins colocalize at the front of migrating cells but also at the actin-rich phagocytic cup in macrophages. We show that IQGAP1 interaction with Dia1 is required for phagocytosis and phagocytic cup formation. Thus, we identify IQGAP1 as a novel component involved in the regulation of phagocytosis by mediating the localization of the actin filament nucleator Dia1.

Figure 1.

Identification of IQGAP1 as a Dia1-binding protein. (A) HeLa cell lysate was applied to glutathione–Sepharose bead resin columns loaded with GST, GST-FH2, or GST-FH3. Bound proteins were eluted by 200 and 500 mM NaCl and analyzed by SDS-PAGE and Coomassie staining. (B) An aliquot of the eluate was analyzed by immunoblotting using antibodies specific for IQGAP1. (C) HeLa cells were treated with 10% FBS or left untreated before immunoprecipitation for Dia1 or control IgG. 5% of total cell extracts and 30% of immunoprecipitation (IP) reactions were analyzed by immunoblotting as indicated. (D) Myc-Dia1-FH3 plasmids were microinjected into NIH3T3 cells located in the front row of a wounded monolayer 30 min after wounding. 3 h later, cells were stained for myc, F-actin, and IQGAP1. Bar, 10 μm.

Figure 2.

IQGAP1 binds the armadillo repeat of Dia1. (A) Schematic representation of Dia1 and IQGAP1. (B) Flag-IQGAP1 constructs were transfected in HEK293 cells with myc-FH3. (C) Indicated Dia1 mutants were transfected with GFP-IQGAP1. (D) Dia1 mutants were transfected with GFP-IQGAP1. (E) Flag-Dia1 constructs were transfected with GFP-IQGAP1. (B–E) Cell extracts were incubated using α-Flag-agarose and analyzed by immunoblotting as indicated. (F) Increasing concentrations of recombinant His–Dia1-nt were incubated with 60 μM of immobilized GST-DBR, the amounts of bound Dia1-nt were visualized by Coomassie staining, and the K_D value was calculated by nonlinear regression analysis. The results are representative of three independent experiments. (G) Association of GST-DBR to Dia1-nt depends on Dia1-ct–mediated autoinhibition and its release by RhoAV14. Purified proteins were analyzed by immunoblotting (α-His). Coomassie staining for GST-DBR is shown (bottom). One representative experiment out of three independent experiments is shown.

Figure 3.

IQGAP1 is required for localization but does not activate or inhibit Dia1. (A) Coomassie-stained SDS-PAGE of 1–2 μg of proteins used in B. (B and C) Pyrene actin assembly assays containing 4 μM actin (5% pyrene labeled) and purified components as indicated. (D) NIH3T3 cells were transfected with either control or siRNAs specific for IQGAP1 (siRNA #2) for 48 h before wounding. Cells were fixed 4 h after wounding and analyzed for IQGAP1 and Dia1 using specific antibodies. (E) Dia1 localization at the wound edge was determined by scoring a minimum of 100 cells. (F) Cells were treated as in D, lysed, and the indicated proteins were analyzed by immunoblotting. (G) NIH3T3 cells were transfected with either control or siRNA specific for Dia1 for 48 h before wounding. Cells were fixed 4 h later and analyzed for IQGAP1 and Dia1 using specific antibodies. (H) IQGAP1 localizations to the wound edge were quantified as in E. (I) Cells were treated as in G, lysed, and the amount of Dia1 as well as the indicated controls were analyzed by immunoblotting as indicated. (E and H) Values represent means ± SD (error bars) of three independent experiments. Bar, 10 μm.

Figure 4.

IQGAP1, Rho-GTP, and Dia1 are components of the phagocytic cup. RAW macrophages were incubated with 3-μm avidin-coated beads for 15 min before fixation and immunostaining. Beads were visualized by phase contrast. All images represent a single confocal plan. (A) IQGAP1 (green) and F-actin (red) colocalize at the phagocytic cup during bead internalization (arrows). (B and C) Dia1 (B) and active Rho, which was detected by using recombinant GFP-RBD (C), are enriched at the phagocytic cup. Arrows indicate localization at the phagocytic cup. (D) The colocalization of Dia1 (red) and IQGAP1 (green) at the phagocytic cup is shown (merge). Boxed areas are magnifications of the phagocytic cup. Bars, 5 μm.

Figure 5.

Association of IQGAP1 with Dia1 is indispensable for phagocytosis. (A and B) RAW macrophages were transfected with GFP (A) as a control or GFP-IQGAP1 (B; green) before incubation with 3-μm avidin-coated beads for 15 min. GFP expression does not interfere with bead binding at the membrane (red) or phagocytosis (internalized beads observed by phase contrast). IQGAP1 is recruited at the phagocytic cup (arrow) but is not detected around internalized beads (asterisk). (C) GFP-IQGAP1– and GFP-actin–expressing macrophages showed the recruitment of actin and IQGAP1 in the tail of 1-μm rocketing phagosomes. Asterisks indicate the positions of the rocketing phagosome in each video still. (D) RAW macrophages were transfected with GFP-DBR (green), incubated with 3-μm avidin- coated beads for 1 h, fixed, and immunostained for Dia1 or IQGAP1 (red). Membrane- attached beads (noninternalized) are indicated by arrows (top, α-avidin in blue). (E) The percentage of positive transfected cells either with GFP or GFP-DBR displaying one or more than two internalized beads per cell. (F) IQGAP1 protein expression in RAW cells transfected with control siRNA or two different siRNA specific for IQGAP1 (#1 and #2). (G) The graph shows the percentage of siRNA-treated cells with at least one internalized avidin-coated bead after 1 h of incubation. All quantifications shown are the means ± SD (error bars) of three independent experiments. Bars, 5 μm.