PMC

Association of IRSp53 with lipids. (A) Domain structure of IRSp53. (B) Quantification of the phosphoinositide-binding abilities of the RCB domain. ELISA was performed with GST as a negative control and GST-AktPH as positive controls. 1.0 μg ml⁻¹ GST-tagged proteins were overlaid on lipid vesicles (PE/PC/PIXP, PA, or phosphatidylserine [PS] = 1:1:0.2 μg). Error bars represent SD of at least three independent experiments. PA, phosphatidic acid; PE, phosphatidyl ethanolamine. (C) Liposome binding by the RCB domain. Liposomes (PC/PI/PIP₃ = 48:48:4 or PC/PI = 50:50) were mixed with the RCB protein at the indicated concentrations. The proteins that cosedimented with the liposomes were visualized. (D) Negative control of C without liposome at a 1.5-μM RCB concentration.

Role of IRSp53 in WAVE2 localization. (A–D) Cells were fractionated by ultracentrifugation, and the proteins in each fraction were analyzed by Western blotting. (A) Proteins in membrane and cytosol fractions of serum-starved A431 cells. The ratios of the protein levels were examined by densitometry and expressed with SD. (B) Proteins in the membrane fraction of A431 cells transfected with expression vectors for constitutively active Rac (Rac CA), dominant-negative Rac (Rac DN), control RNAi, and/or IRSp53 RNAi pSuper vector. Stealth RNAi was applied for cells treated with EGF. (C) Proteins in the membrane fraction of EGF-stimulated or unstimulated A431 cells. (D) Quantification of WAVE2 in the membrane fractions. (E) Proteins in the membrane fraction of EGF-stimulated A431 cells transfected with stealth RNAi targeted to WAVE2. *, P < 0.005; **, P < 0.05 by t test. Error bars represent SD.

Localization of WAVE2 and IRSp53 in A431 cells. (A) Localization of WAVE2 and IRSp53 in EGF-stimulated A431 cells treated with control, IRSp53, or IRTKS stealth RNAi was examined by immunofluorescence with anti-WAVE2 antibody or anti-IRSp53 antibody (M051-3). Actin filaments stained with phalloidin are also shown. WAVE2 and IRSp53 were localized at ruffles (arrowheads) and cell–cell junctions (arrows). (B) Localization of WAVE2-GFP in A431 cells with reduced IRSp53 or Abi1 expression. A431 cells were transfected with WAVE2-GFP expression vector and pSuper vectors for control, IRSp53, or Abi1 RNAi. GFP signal was enhanced by anti-GFP staining. Bar, 40 μm. (C) Western blot analysis of whole cell lysates from control, IRSp53, or Abi1 stealth RNAi-treated A431 cells. Western blot with anti-actin antibody is shown as a loading control. (D) Quantification of the speed of the leading edge extension of EGF-stimulated A431 cells. Significant difference was analyzed by the t test. Error bars represent SD.

Rac- and PIP₃-dependent actin polymerization mediated by IRSp53 and WAVE2. (A) 40 nM of purified recombinant WAVE2 was preincubated with 60 nM of recombinant IRSp53, 300 nM GTPγS-loaded Rac, 300 nM GTPγS-loaded Cdc42, 8 μM PIP₃-containing PC/PI liposomes (PIP₃/PC/PI), and/or PC/PI liposomes. Arp2/3 complex activation was determined with 50 nM Arp2/3 complex and 2 μM of 10% pyrene-labeled actin. (B) 40 nM of purified recombinant WAVE2 coexpressed and copurified with Abi1 that was incubated with recombinant PIR121 copurified with 250 nM Nap1. Rac-, PIP₃-, and IRSp53-dependenent Arp2/3 activation was examined as in A. (C) Concentration response of filament elongation rate dependent on IRSp53 concentration. Purified recombinant WAVE2 was incubated with various concentrations of IRSp53 in the absence or presence of PIP₃/PC/PI liposome and GTPγS-loaded Rac or PC/PI liposomes alone. (D) Purified recombinant wild-type or PIP3 binding-defective A5 mutant of WAVE2 was incubated with IRSp53 in the absence or presence of PIP₃/PC/PI liposome and GTPγS-loaded Rac. Filament elongation rate is shown with SD (error bars). (E) 100 nM of purified recombinant WAVE2, 250 nM Abi1, PIR121 copurified with 250 nM Nap1, and/or 500 nM IRSp53 were mixed, and Arp2/3 complex activation was measured in the absence of liposomes. (F) Effect of 15 μM GTPγS-loaded Rac on Arp2/3 activation in the absence of liposomes.

Arp2/3 complex activation by the WAVE2 complex. (A and D) WAVE2 complex purified from the Triton X-100 or cytosol fraction with (EGF) or without stimulation (starved) of A431 cells (A) and from the Triton X-100 fraction of control or IRSp53 RNAi-treated cells (D) was analyzed for Arp2/3 complex activation ability with 50 nM Arp2/3 complex and 2 μM of 10% pyrene-labeled actin. Actin polymerization was monitored by pyrene fluorescence. The VCA fragment of WAVE2 was also analyzed for concentration-dependent activation of the Arp2/3 complex. (B and E) Dose-response curves for A and D show the change in filament elongation rate as a function of increasing concentrations of WAVE2 complex. Elongation rate was determined by the slope of fluorescence change at 15% actin polymerization. (C) Dose-response curves show the change in filament elongation rate as a function of increasing concentrations of WAVE2 complex purified from the Triton X-100 or cytosol fraction with (EGF) or without stimulation (starved) of WAVE2 knockout MEFs stably expressing FLAG-WAVE2. (F) WAVE2 complex purified from the cytosol fraction was mixed with recombinant IRSp53 or GTPγS-loaded Rac, and Arp2/3 complex activation was measured as in A.

Rac- and PIP₃-dependent actin polymerization mediated by IRSp53 and WAVE2 on PIP₃. (A and B) Protein–lipid association was monitored by cosedimentation assay. Cosedimented proteins with liposomes with or without PIP₃ were examined by Western blot analysis. PIP₃, WAVE2, IRSp53, Rac, and Cdc42 were included at concentrations of 8 μM, 60 nM, 60 nM, 300 nM, and 300 nM, respectively. (C and D) Quantification of proteins bound to liposomes was determined by densitometry for IRSp53 (C) and WAVE2 (D). Error bars represent SD. (E) Lipid association of the WAVE2 complex purified from the high molecular weight fraction of sucrose gradient fractionation from A431 cells was monitored by cosedimentation assay. Cosedimented proteins with liposomes containing or lacking PIP3 were examined by Western blot analysis. GTPγS-loaded Rac and IRSp53 were added, and their effects were examined. sup, supernatant; ppt, precipitate.

Involvement of IRSp53 in Rac and EGF-induced ruffle formation in A431 cells. (A) A431 cells were transfected with constitutively active Rac (Rac CA) and control or IRSp53 RNAi pSuper vector. (left) Actin filament (red; phalloidin) and WAVE2 localization (green; anti-WAVE2) was then analyzed. Rac expression is shown in blue. (right) Quantification of membrane ruffles in transfected cells. Three independent experiments were performed. (B) EGF-stimulated A431 cells transfected with IRSp53 RNAi pSuper vector alone or plus RNAi-resistant GFP-IRSp53 expression vector. (left) Actin filament (red; phalloidin) and GFP localization (green; anti-GFP) were then analyzed. (A and B) Bars, 40 μm. (right) Quantification of membrane ruffles as in A. (C and D) A431 cells stably expressing FLAG-tagged WAVE2 were transfected with Rac CA or Cdc42 CA expression vector. Then, IRSp53 in the FLAG-WAVE2 precipitates was examined by Western blot analysis (C). The amount of IRSp53 relative to WAVE2 in the precipitate was quantified by densitometry (D). Three independent experiments were performed. (E) GST fusion protein of WAVE2 was immobilized, and purified IRSp53 was incubated with GTPγS-loaded Rac or Cdc42. After washing, bound IRSp53 was examined by Western blotting. Error bars represent SD.

WAVE2 complex purified from A431 cells. (A) FLAG-tagged WAVE2 was immunoprecipitated from A431 cells stably expressing FLAG-WAVE2 or from control vector–transfected cells by FLAG affinity immunoprecipitation (IP). Specific association was examined by Western blot analysis. (B) WAVE2 complex was purified from A431 cells stably expressing FLAG-WAVE2 by FLAG affinity immunoprecipitation. The Triton X-100 fraction indicates WAVE2 from both the cytosol and membrane fractions, and the cytosol fraction indicates WAVE2 from the cytosol fraction alone. WAVE2 complex from EGF-treated cells (EGF) or untreated cells (starved) are shown in silver-stained gels. The orders of the samples of Western blots are arranged to fit the order of the samples of silver-stained gels. (C) Molar ratio of the components of the WAVE2 complex shown in B. Molar ratio was determined as the band intensity of Coomassie brilliant blue–stained gels (not depicted) divided by molecular weight. Tx, Triton X-100; cyt, cytosol. (D) The dissociation constants (Kd) of the indicated domains to WAVE2 were determined using a dual polarization interferometer. The representative response curves (top) used for the calculation of Kd (bottom) are shown. Error bars represent SD.

Fractionation of cell lysate and purified WAVE2 from A431 cells by sucrose gradient. (A–C) Cell lysates of control A431 cells (A), A431 cells stably expressing FLAG-tagged WAVE2 (B), and WAVE2 knockout MEF stably expressing FLAG-tagged WAVE2 (C) were fractionated by molecular weight in a 3–30% sucrose gradient, and each fraction was visualized by Western blotting. (D–K) Purified WAVE2 from A431 cells stably expressing FLAG-tagged WAVE2 was fractionated. (D) WAVE2 purified from the Triton X-100 fraction of pSuper vector–transfected cells of control, IRSp53, or Abi1 RNAi. (E) Sequential sucrose gradient fractionation. WAVE2 purified from the Triton X-100 fraction was fractionated, and the high molecular weight fraction (fraction 8–10) was further fractionated. Bar indicates the fractions that were used in the second sucrose gradient. (F) WAVE2 purified from the Triton X-100 fraction of EGF-treated cells. (G) Cytosol fraction of EGF-treated cells. (H) Triton X-100 fraction of serum-starved cells, and (I) cytosol fraction of serum-starved cells. (J and K) WAVE2 from the cytosol fraction of serum-starved cells was incubated with GTPγS-loaded Rac (J) or GTPγS-loaded Rac, IRSp53, and PIP₃ (K). (L and M) Activities of the WAVE2 monomer and WAVE2 complex were compared. (L) Sucrose gradient in the absence of protein was also used as a control. The axis of actin polymerization rate for VCA + Arp2/3 is on the right, and that for Arp2/3 alone is on the left. (M) Each sucrose gradient fraction of WAVE2 from the Triton X-100 fraction of serum-starved cells was analyzed for Arp2/3 complex activation with 50 nM of purified Arp2/3 complex and 2 μM of 10% pyrene-labeled actin. The rate of actin polymerization at 10% polymerization was determined and plotted relative to fraction number. Arrows indicate the peak fraction of the activity of the Arp2/3 complex activation in N. (N) The value of slope in M was normalized to the value of Arp2/3 alone in L as 1 (left axis) and was plotted with WAVE2 concentration determined by Western blotting followed by densitometry (right axis). Error bars represent SD.