HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Park, S. J. Articles by Takenawa, T. Search for Related Content PubMed Articles by Park, S. J. Articles by Takenawa, T.

¹ Department of Biochemistry, Institute of Medical Science, University of Tokyo, Tokyo 108-8539, Japan; 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8539, Japan
² PRESTO, Japan Science and Technology Corporation (JST), Institute of Molecular and Cellular Biosciences, University of Tokyo, Tokyo, Japan
³ Fine Morphology Laboratory, Department of Basic Medical Science, Institute of Medical Science, University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8539, Japan

	Abstract

Top Abstract Introduction Results Discussion Experimental procedures References

 
N-WASP induces filopodial actin cytoskeleton through activation of the Arp2/3 complex. Here, we show that heat shock protein 90 (HSP90) regulates the structure of actin filaments induced by N-WASP and the Arp2/3 complex. HSP90 binds to N-WASP and to F-actin and bundles actin filaments. Bundling activity of HSP90 does not affect actin filament nucleation induced by N-WASP and the Arp2/3 complex. HSP90 is co-localized with N-WASP at branching points of actin filaments produced by the Arp2/3 complex and thereby bundles branched filaments; this bundled actin structure is inhibited by blocking direct binding between HSP90 and N-WASP. Furthermore, HSP90 converts branched actin filaments on N-WASP-coated beads to filopodia-like star-shaped bundles. These findings indicate that HSP90 promotes the formation of N-WASP/Arp2/3 complex-induced unbranched filopodial actin structures.

	Introduction

Top Abstract Introduction Results Discussion Experimental procedures References

 
Reorganization of the actin cytoskeleton underlies diverse cellular processes including morphologic change, cell movement, and motility of intracellular vesicles and pathogenic organisms. Wiskott–Aldrich syndrome protein (WASP) family proteins such as WASP, N-WASP (Derry et al. 1994; Miki et al. 1996, 1998a) and WAVEs 1, 2 and 3 (Miki et al. 1998b; Suetsugu et al. 1999) activate the Arp2/3 complex, resulting in rapid actin polymerization at sites where rearrangement of the actin cytoskeleton is needed (Miki et al. 1998a,b; Machesky et al. 1999; Rohatgi et al. 1999).

Actin filaments nucleated by the Arp2/3 complex generate branched actin filament networks that are found in lamellipodia. In vitro studies have shown that WASP/WAVE proteins create branched actin filaments (Blanchoin et al. 2000; Pantaloni et al. 2000). However, actin cytoskeleton structures composed of these proteins show two patterns in vivo. N-WASP is believed to be involved in the generation of unbranched and bundled actin filaments in filopodia (Miki et al. 1998a; Svitkina et al. 2003), whereas WAVE contributes to the formation of branched actin filament networks in lamellipodia (Miki et al. 1998b; Suetsugu et al. 2003). N-WASP- and WAVE-mediated formation of filopodia and lamellipodia are known to be strictly controlled by the small GTPases Cdc42 and Rac, respectively (Symons et al. 1996; Miki et al. 1998a,b; Miki & Takenawa 2003). However, filopodial and lamellipodial actin-containing structures are frequently observed together at the leading edge of migrating cells, suggesting that they act cooperatively for efficient protrusive movement of the cell membrane (Wear et al. 2000; Condeelis 2001; Pollard & Borisy 2003). It has been reported that filopodia are formed by reorganization of dendritic branched filaments into bundled filaments (Svitkina et al. 2003; Vignjevic et al. 2003). Thus, remodeling of branched actin networks after nucleation of branched actin filaments is necessary for filopodial unbranched actin networks.

We previously reported that heat shock protein 90 (HSP90) is involved in N-WASP-mediated actin polymerization (Park et al. 2005). HSP90 associates with N-WASP; interaction between HSP90 and the basic region of N-WASP increases activation of N-WASP in combination with PIP₂ or tyrosine phosphorylation. It has been reported that HSP90 is an actin-binding protein (Koyasu et al. 1986; Nishida et al. 1986), although its function with respect to actin structures is unclear. We have shown that HSP90 and N-WASP are co-localized in cortical actin filaments of invasive podosomes and extending neurites (Park et al. 2005).

Here, we investigated whether HSP90 is directly involved in N-WASP-induced actin filament assembly. We found that HSP90 associates not only with N-WASP but also with actin filaments, leading to transition of branched actin filaments into bundled structures. Blocking of the interaction between HSP90 and N-WASP inhibited bundling of actin filaments induced by N-WASP and the Arp2/3 complex. Purified recombinant HSP90 converted dendritic actin clouds on beads coated with N-WASP into star-shaped actin filament bundles, indicating that HSP90 plays a role in the formation of N-WASP-mediated filopodial actin bundles.

	Results

Top Abstract Introduction Results Discussion Experimental procedures References

 
HSP90 binds to and cross-links F-actin

We first confirmed that HSP90 binds to actin filaments by co-sedimentation assay. Both His-HSP90 expressed in Sf9 cells and bovine brain HSP90 co-sedimented with F-actin in a concentration-dependent manner (Fig. 1A, B). HSP90 alone did not sediment under these conditions, indicating that sedimentation was due to binding to F-actin. The equilibrium dissociation constant (K_d) for binding of HSP90 to F-actin was 1.64–1.82 µM (Fig. 1B), similar to previous data obtained with HSP90 purified from mouse lymphocytes (Koyasu et al. 1986). F-actin binding of HSP90 was detected in precipitates after low-speed centrifugation (12 000 g) (Fig. 1C). Electron microscopy showed packed actin filament bundles in the presence of HSP90 but not in the absence of HSP90 (Fig. 1D,E). These results indicate that bundling was predominantly responsible for F-actin pelleting with HSP90 during low-speed centrifugation.

View larger version (104K): [in this window] [in a new window]   Figure 1  HSP90 binds to and bundles F-actin. (A) For the binding assay, 4 µM actin was polymerized and mixed with 2 µM of His-HSP90 expressed in Sf9 cells or bovine brain HSP90 for 1 h at room temperature. Association with F-actin was examined by co-sedimentation assay after ultracentrifugation at 70 000 r.p.m. Monomeric (supernatant) actin and filamentous (precipitate) actin were analyzed by SDS-PAGE followed by Coomassie Blue staining. (B) Binding to F-actin. Increasing amounts of F-actin were incubated either alone or with 2 µM HSP90 and subjected to co-sedimentation assay by high-speed ultracentrifugation. The equilibrium dissociation constant for binding of HSP90 to F-actin was calculated with the Lineweaver–Burk reciprocal plot and was 1.82 µM for His-HSP90 and 1.64 µM for bovine HSP90. (C) For cross-linking/bundling assay, 2 µM F-actin was mixed with 2 µM His-HSP90 or bovine brain HSP90 for 1 h and then centrifuged at 12 000 g. (D and E) A concentration of 2 µM F-actin was incubated with (E) or without (D) 2 µM His-HSP90 for 2 h. The actin solution was observed by electron microscopy. Scale bar, 0.1 µm. We observed packed actin bundles in the presence of HSP90 but no in the absence of HSP90, although in low density.

HSP90 binds to N-WASP (Park et al. 2005; Fig. 2C). Therefore, binding of HSP90 to F-actin may affect the interaction of HSP90 and N-WASP and the assembly of N-WASP-induced actin filaments. We first induced polymerization of G-actin for 15 min in the presence or absence of HSP90, N-WASP and/or the Arp2/3 complex, during which actin polymerization induced by N-WASP/Arp2/3 complex reaches a plateau (Fig. 2D), centrifuged the actin solutions, and analyzed the F-actin pellets (Fig. 2A). Addition of HSP90 did not affect the concentration of assembled F-actin. When HSP90 was added in the presence of wild-type (WT) N-WASP and Arp2/3 complex, we found that F-actin binding of HSP90 was enhanced. The IQ-Basic N-WASP mutant, which constitutively activates Arp2/3 complex but lacks HSP90 binding (Park et al. 2005; Fig. 2C), also induced excessive F-actin formation in the presence of Arp2/3 complex, similar to the amount induced by full-length WT N-WASP in the presence of PIP₂ and Cdc42. However, N-WASP IQ-Basic mutant did not increase the level of HSP90 in the F-actin pellet, as WT N-WASP did, indicating that the increase in HSP90-F-actin binding after addition of N-WASP/Arp2/3 complex was accomplished not only by F-actin binding of HSP90 but also by interaction of HSP90 with N-WASP. N-WASP in the absence of Arp2/3 complex did not significantly affect the HSP90-F-actin interaction, suggesting cooperative function of HSP90, N-WASP and the Arp2/3 complex.

View larger version (48K): [in this window] [in a new window]   Figure 2  HSP90 cross-links N-WASP/Arp2/3 complex-induced actin filaments. (A and B) Concentration of 1 µM HSP90 and/or 0.4 µM N-WASP (NW) were added to 1 µM G-actin polymerization reaction mixtures in the absence or presence of Arp2/3 complex (100 nM). For activation of N-WASP, 5 µM PIP2 and 0.2 µM GTPs-Cdc42 were also added. After actin polymerization for 15 min, the actin mixtures were centrifuged at 70 000 r.p.m. (A) or 12 000 g (B), and subjected to SDS-PAGE and Coomassie Blue staining. F-actin pellets were also analyzed by Western blotting with anti-Arp3 antibody. (C) Pull-down assay with His-N-WASP or GST-HSP90 protein. A concentration of 5 µg His-N-WASP WT or IQ-B protein was incubated with 1 µg His-HSP90 in the presence or absence of 100 nM Arp2/3 complex and then precipitated with Ni-NTA beads. Bound HSP90 and Arp3 were assayed with anti-HSP90 and anti-Arp3 antibodies (left panel). Five µM PIP2 and 0.2 µM GTPs-Cdc42 were used. N-WASP IQ-Basic mutant constitutively activates Arp2/3 complex, but lacks HSP90 binding. Similarly, 5 µg GST-HSP90 was incubated with 2 µg His-N-WASP WT and 200 nM Arp2/3 complex and then precipitated with glutathion sepharose beads (light panel). (D) Actin polymerization was performed with 0.5 µM actin containing 10% pyrene, 60 nM Arp2/3, and 100 nM WT N-WASP in the presence or absence of 1 µM HSP90. F-actin was added at a final concentration of 0.2 µM. GTPs-Cdc42 and PIP2 were added at 500 nM and 5 µM, respectively. After preincubation for 5 min, actin was added, and fluorescence was monitored. Concentrations of barbed-end of actin filaments at time points where polymerization was 80% complete were calculated from elongation rates according to the following equation: [barbed ends] = elongation rate/(K+[actin monomers]), where K+ is the rate constant for the association of G-actin with filament ends.

To examine the possibility that HSP90 affects the kinetics of N-WASP/Arp2/3 complex-induced actin polymerization, we analyzed the effect of HSP90 on N-WASP/Arp2/3 complex-induced actin polymerization in the presence or absence of F-actin (Fig. 2D). Addition of HSP90 to the actin polymerization reaction mixture had no effect on the lag time of polymerization, the growing rate of actin filaments or the concentration of polymerized actin (Fig. 2D). We did not detect a significant difference in the concentration of barbed ends after addition of HSP90 (Fig. 2D). F-actin binding of HSP90 did not affect the association of the Arp2/3 complex with actin filaments (Fig. 2A,D). Similar results were obtained when prepared F-actin was added (Fig. 2D). The actin cross-linking activity of HSP90 did not significantly affect the nucleation of actin filaments induced by N-WASP and the Arp2/3 complex.

HSP90 induces the formation of long unbranched actin filaments

We next examined the structure of N-WASP/Arp2/3 complex-induced actin filaments in the presence or absence of HSP90. After N-WASP/Arp2/3 complex-mediated actin polymerization, actin filaments were visualized by staining with Alexa 488-conjugated phalloidin (Fig. 3A). We imaged fluorescent F-actin and carefully quantified the numbers of branches and lengths of mother filaments (Fig. 3B–F). In the absence of HSP90, a 3-min actin polymerization reaction containing 0.1 µM WT N-WASP and 60 nM Arp2/3 complex showed branching of 58% of filaments, with 0.42 branches/filament µM (Fig. 3B,C). In the presence of HSP90, actin polymerization reaction generated long and unbranched actin filaments (Fig. 3Ab), of which bundling was confirmed by electron microscopy (Fig. 5Cb and Supplementary Fig. S2B). Addition of 2 µM HSP90 decreased the total number of actin filaments by 20% and decreased the number of branched filaments by ~40% (Fig. 3B). HSP90 decreased actin filament branching, whereas it increased the length of actin filaments 1.35-fold compared to that in the absence of HSP90, leading to a branching density of 0.24 branches/filament µM (Fig. 3C). HSP90 decreased the number of WT N-WASP-mediated branched actin filaments at reaction times > 3 min, in which the slope of growing rate of actin filament is linear (Fig. 3D). Under these conditions in which filament branching is decreased, the effect of HSP90 on increased actin filament length was marked (Fig. 3E). Inhibition of actin filament branching by HSP90 was inversely correlated with filament elongation (Fig. 3D–F).

View larger version (33K): [in this window] [in a new window]   Figure 3  HSP90 induces the formation of long unbranched actin filaments. (A) Visualization of actin filaments after actin polymerization for 3 min. Actin polymerization was performed with 2 µM actin, 60 nM Arp2/3 and 100 nM N-WASP in the presence (b) or absence (a) of 2 µM of His-HSP90 or GST-fascin (c). GTPs-Cdc42 and PIP2 were added at 500 nM and 1 µM, respectively. F-actin products were visualized by light microscopy after Alexa Flour 488-conjugated phalloidin staining. Scale bar, 2 µm. (B) Numbers of total and branched filaments produced after actin polymerization for 3 min. (C) Values for mother filaments length and branches per mother filament µM after actin polymerization for 3 min were calculated as the mean of at least three hundred mother filaments in each five independent experiment. (D–F) Time courses of actin polymerization. Plots show number of branches on mother filament, mother filament length and branch density per mother filament µM vs. time. The data shown are means of at least three independent experiments.

View larger version (32K): [in this window] [in a new window]   Figure 5  HSP90 is co-localized with N-WASP at branch points. (A and B) Visualization of actin filaments after actin polymerization. Actin polymerization with 2 µM actin, 60 nM Arp2/3 complex, and 10% Alexa 633-labeled N-WASP (100 nM) was performed in the absence (A) or presence (B) of 2 µM HSP90 containing 10% Alexa 488-labeled HSP90. GTPs-Cdc42 and PIP2 were added at 500 nM and 5 µM, respectively. After actin polymerization for 3 min, actin filaments were visualized by rhodamine-conjugated phalloidin staining, Scale bar, 1 µm (C) A concentration of 2 µM F-actin was polymerized for 3 min with 2 µM actin, 60 nM Arp2/3 and 100 nM WT N-WASP in the presence (b) or absence (a) of 2 µM HSP90. Negatively stained F-actin was observed by electron microscopy. Arrow heads indicate branch points. Scale bar, 0.2 µm. Branched filaments are enlarged in the small panels. (D and E) Total internal reflection fluorescent images of actin filaments. Concentration of 0.7 µM actin, 30 nM Arp2/3 complex and 50 nM N-WASP were incubated in the absence (D) or presence (E) of 2 µM HSP90. Proteins were mixed quickly in TIRF polymerization buffer and transferred immediately to flow cells. Arrow heads indicate newly nucleated branch points. Scale bar, 1 µm.

Interaction of HSP90 and N-WASP is important for formation of long unbranched actin filaments

We next examined whether interaction of HSP90 and N-WASP affects the formation of long unbranched filament structures. The VCA region of N-WASP, the catalytic region for Arp2/3 complex activation which shows little binding to HSP90 (Park et al. 2005), induced the formation of branched actin filaments similar to those formed by N-WASP activated by PIP₂ and Cdc42 (data not shown). Addition of HSP90 to the VCA N-WASP-induced actin polymerization mixture decreased branching density by 20% (Fig. 4A). However, in the combination of the VCA N-WASP and HSP90, actin filaments length was less than that induced by WT N-WASP activated by PIP₂ and Cdc42 (Fig. 4B). Similar results were obtained in the experiments with IQ-Basic N-WASP. In polymerization mixture containing the IQ-Basic N-WASP, addition of HSP90 did not induce a marked increase in filament length. These results indicate that direct binding between HSP90 and N-WASP is required for efficient bundling of filament and elongation.

View larger version (23K): [in this window] [in a new window]   Figure 4  Interaction of HSP90 and N-WASP is important for the formation of long unbranched filaments. (A and B) Mother filaments length and branches per filament µM after actin polymerization for 3 min were quantified by light microscopy. All assays were performed with WT N-WASP or N-WASP mutants at 100 nM and Arp2/3 complex at 60 nM. GTPs-Cdc42 and PIP2 were added at 500 nM and 1 µM, respectively. His-HSP90 or GST-fascin was added at a final concentration of 2 µM.

HSP90 and N-WASP are co-localized at branching points of actin filaments

To confirm that HSP90 acts together with N-WASP to regulate the structure of actin filaments, we examined the distribution of HSP90 and N-WASP. F-actin was polymerized in the presence of Arp2/3 complex, fluorescently labeled N-WASP, and HSP90 (Fig. 5). In actin polymerization assays in the absence of HSP90, N-WASP was undetectable in filaments (Fig. 5A). In contrast, addition of HSP90 induced the production of unbranched actin filaments with HSP90 localized along the actin filaments and N-WASP dotting the actin filaments (Fig. 5B). Interestingly, under this condition, we frequently identified HSP90 co-localized with N-WASP at branch points of branched actin filaments. We occasionally observed nodes of strong phalloidin staining on long filaments where both HSP90 and N-WASP were clearly co-localized. Immunostaining with Arp2 antibody coincided with the nodes, indicating that the nodes are formed from actin branch structures and that HSP90 interacts with N-WASP at branch points (Fig. 5B).

Electron microscopy revealed branched filaments with 70° angles to the direction of mother filament growth in the absence of HSP90 (Fig. 5Ca). In the presence of HSP90, we observed bundled actin filaments in which daughter filaments were aligned along mother filaments, particularly with no gaps near the branch points, and in looser bundles distal to the branch points, suggesting that bundling by HSP90 is initiated at branch points rather than at other sites (Fig. 5Cb). We then observed N-WASP/Arp2/3 complex-induced actin polymerization directly by time-lapse microscopy (Fig. 5D,E). In the absence of HSP90, branched actin filaments appeared within 1 min and grew, forming daughter branches (Fig. 5D). In the presence of HSP90, branch formation showed a similar time course. However, these branches were soon assembled into the mother filaments, overlapping F-actin fluorescence (Fig. 5E). After bundling of branched actin filaments, length of the mother filaments was elongated compared to length of the aligned daughter filaments, indicating that bundling of actin filaments causes elongation of mother filaments. When we induced G-actin polymerization in the absence of Arp2/3 complex or in the presence of a capping protein (CapZ), addition of HSP90 had no effect on length of actin filaments (Supplementary Fig. 1). HSP90 did not directly regulate growing rate of actin filament via association with actin filament ends.

HSP90 induces N-WASP-mediated filopodia-like actin filament bundles

Given the multiple links between N-WASP/Arp2/3 complex and filopodial actin structure, we finally induced the formation of filopodia-like bundles with purified proteins (Fig. 6). WT N-WASP was coated on to beads and then mixed with actin and Arp2/3 complex, which led to the formation of clouds of actin filaments on the beads. When HSP90 was added to the assay system, ~25% of the actin clouds were assembled into star-shaped filopodia-like bundles. When beads coated with VCA N-WASP were used, these filopodial structures were not observed. Instead, adjacent filaments were cross-linked as in a spider web. Beads in the absence of N-WASP showed no formation of star-shaped bundles. Addition of HSP90 induced only low-density cross-linked bundles. These results indicate that association between HSP90, N-WASP and branched actin filaments are required for the formation of filopodia-like actin bundles. Fascin formed excessive filopodia on beads coated with VCA or with WT N-WASP. Bundles formed by fascin were strong and straight and were longer than those formed by HSP90. It is likely that the bundling activity of fascin is not dependent on an interaction with nucleating factor on the beads and can occur anywhere along the filaments.

View larger version (41K): [in this window] [in a new window]   Figure 6  HSP90 induces filopodia-like actin bundles on beads coated with N-WASP. WT N-WASP or N-WASP VCA was coated on to polystyrene beads. A volume of 0.5 µL of beads was preincubated with 60 nM Arp2/3 complex, and 2 µM actin was added. The mixture was incubated on ice for 20 min and then brought to room temperature to allow for actin assembly for 1 h. To produce filopodia-like stars, 2 µM HSP90 was added to N-WASP-coated beads prior to incubation at room temperature for 1 h. Scale bar, 5 µm. Enlarged examples are shown in the panels below.

	Discussion

Top Abstract Introduction Results Discussion Experimental procedures References

We found that the middle domain of HSP90 is required for N-WASP binding (Park et al. 2005) and F-actin binding but is not sufficient for the cross-linking activity of HSP90 (data not shown). We attempted to create HSP90 mutant lacking F-actin or N-WASP binding. A large number of point-mutated full-length HSP90 mutants of the middle domain were created. However, no information was obtained due to impaired ATP binding/ATPase activity. Treatment with geldanamycin, a specific inhibitor of HSP90 that inhibits ATP binding of HSP90, decreased the interaction between HSP90 and N-WASP and also decreased F-actin binding by HSP90 (data not shown). When we used a large excess of actin in F-actin binding assays, a substantial fraction of HSP90 still did not bind to F-actin, suggesting that HSP90 may be present in a largely inactive form. HSP90 undergoes conformation changes in an ATP binding-dependent manner (Huai et al. 2005; Ali et al. 2006). Although the molecular mechanism with regard to actin cross-linking of HSP90 is still unclear, we propose that flexible conformational change of full-length HSP90 contributes to F-actin cross-linking. In response to extracellular stimuli, HSP90 would associate F-actin together with N-WASP/Arp2/3 complex and lead to rapid actin cytoskeletal reorganization.

We succeeded in reconstituting filopodia-like star-shaped bundles on beads coated with WT full-length N-WASP in a simple solution containing actin, Arp2/3 complex and HSP90. Combination of VCA N-WASP, and HSP90 induced the formation of mesh-like actin filament structures. In the absence of N-WASP, filopodial bundles were not formed. These data strongly indicate that HSP90 allows filopodial bundles to arise from actin nucleation points associated with N-WASP and rapidly links activation of N-WASP to actin filament assembly. HSP90 binds N-WASP at branch points generated by the Arp2/3 complex, and this allows actin filament bundling to occur in parallel with actin nucleation. Real-time imaging revealed that branching and consecutive bundling are responsible for the elongation of actin filaments in the presence of HSP90.

Many studies have suggested that fascin is a major actin cross-linking protein within filopodial bundles (Bartles 2000; Svitkina et al. 2003). In our results, fascin did not bind to N-WASP (data not shown), and the strong ability of fascin to bundle actin filaments was not associated with the presence or activation of N-WASP. In an in vitro actin polymerization system, fascin restricted the actin nucleation rate and actin filament length (Figs 3 and 4). Thus, to bundle actin filaments and to persistently elongate filopodia, fascin should be recruited to actin filaments at the later stages of filopodia formation. In the case of fascin, bundling occurs separately from actin filament nucleation (Vignjevic et al. 2003). It is not clear how fascin is recruited to filopodial sites and activated. Bundles induced by fascin differ from those induced by HSP90. Bundles induced by fascin were thick and straight, and were composed mainly of unbranched parallel filaments (Supplementary Fig. S2). This finding strongly support that HSP90 plays a crucial role in formation of N-WASP-related unbranded actin structures.

	Experimental procedures

Top Abstract Introduction Results Discussion Experimental procedures References

 
Proteins and antibodies

WT and IQ-Basic of N-WASP were produced with the Bac-to-Bac baculovirus expression system (Gibco BRL), either with His or GST tag (Park et al. 2005). His-HSP90 ß and GST-fascin were produced in a baculovirus expression system. Bovine brain HSP90 was purchased from Sigma-Aldrich. His-HSP90 ß was used as a protein of HSP90 unless indicated otherwise. Capping protein ß1 (CapZ) was purified as previously described (Soeno et al. 1998).

Anti-N-WASP polyclonal antibody was produced as previously described (Miki et al. 1998a). Anti-HSP90 monoclonal and polyclonal antibodies were from Santa Cruz Biotechnology, Inc.

Actin polymerization assay

Pyrene-actin assays were performed as previously described (Rohatgi et al. 1999). Polymerization reaction mixtures contained 0.5 or 2 µM G-actin, 0.2 µM pyrene-labeled actin, 0.2 µM ATP, various proteins in 80 µL of X buffer (10 mM HEPES, pH 7.0, 100 mM KCl, 1 mM MgCl₂, 0.1 mM EDTA, 1 mM DTT) and were preincubated for 5 min. N-WASP and HSP90 were added at final concentrations of 100 nM and 2 µM, respectively, unless indicated otherwise. The reaction was initiated by adding a mixture of actin and pyrene-labeled actin to the preincubated protein mixtures. Fluorescence changes were measured at 407 nm with a 365 nm excitation wavelength in a fluorescence spectrometer (Jasco).

Visualization of F-actin

After actin polymerization, 2 µL of the reaction mixture was diluted in 100 µL of staining solution composed of 0.002% BSA, 0.5% methylcellulose, 3 mg/mL glucose, 20 µg/mL catalase, 100 µg/mL glucose oxidase and 2 µM Alexa Fluor 488-conjugated phalloidin (Molecular Probes) in X buffer. Two milliliters of the solution containing actin filaments was placed on to 11 x 22-mm poly L-lysine-coated coverslips and observed by an Olympus IX70 microscope with  x100 (numerical aperture 1 : 35) objective lens.

F-actin binding and cross-linking/bundling assay

Binding of HSP90 to F-actin was analyzed by co-sedimentation assay. Actin was prepared from rabbit muscle as previously described (Miki et al. 1996). F-actin (4 µM) was polymerized by adding 50 mM KCl, 1 mM ATP and 1 mM MgCl₂ to G-actin solution, and incubating for 2 h at room temperature (RT). Prior to binding assay, HSP90 was clarified by ultracentrifugation at 750 000 r.p.m. for 30 min at 4 °C to remove aggregates. F-actin and HSP90 were mixed and incubated for 1 h at RT. The mixture was then centrifuged at 70 000 r.p.m. for 30 min. Supernatants and pellets were then analyzed by SDS-PAGE. For cross-linking/bundling assays, 2 µM HSP90 was reacted with 2 µ freshly prepared F-actin and then centrifuged at 12 000 g for 30 min.

Actin polymerization bead assay

Carboxylated polystyrene beads (Polysciences, Inc.) were coated with N-WASP as previously described (Cameron et al. 1999). In brief, 15 µL of 0.5 µM N-WASP was mixed with 0.5 µL of 0.5-µm carboxylated polystyrene beads. The mixture was incubated for 30 min at RT. Beads were washed and resuspended in 10 µL of buffer containing 10 mM HEPES, pH 7.0, 50 mM KCl, 1 mM MgCl₂ and 1 mM EGTA. For reconstitution, 0.5 µL of coated beads was first mixed with 60 nM of Arp2/3 complex. After incubation for 5 min at RT, 2 µM actin was added, and the mixture was incubated on ice for 20 min before the addition of 2 µM HSP90, allowing for actin assembly at RT for 1 h. A concentration of 2 µM actin, added for actin polymerization, contained 0.2 µM rhodamine-conjugated actin. Rhodamine-conjugated actin was prepared by labeling actin with N-hydroxy succinimidyl-rhodamine (Molecular Probes) as previously described (Isambert et al. 1995) and stored at –80 °C. To visualize actin polymerization on beads, 1 µL of sample solution was placed between a microscope slide and a 22-mm square cover glass to create a chamber.

Total internal reflection fluorescence microscopy

Actin polymerization was initiated by the addition of TIRF buffer (30 mM KCl, 1 mM MgCl₂, 1 mM EGTA, 20 mM DTT, 0.1 mM ATP, 10 mM imidazole, 15 mM glucose, 20 µg/mL catalase, 100 µg/mL glucose oxidase, 0.5% methylcellulose, 20 mM HEPES, pH 7.0). This solution was then immediately transferred to glass-flow cells (24 x 45 x 0.17 mm). Glass-flow cells were incubated with 10 nM N-ethylmaleimide (NEM)-myosin, washed extensively with 1% BSA, and equilibrated with TIRF buffer. Added protein concentrations were 50 nM Arp2/3 complex, 50 nM N-WASP, 2 µM HSP90 and 0.7 µM actin. The proportion of Alexa-488-labeled actin was set to 50%. Images of fluorescent actin filaments were obtained by excitation by total internal reflection at 10 s intervals on an Olympus IX70 microscope.

	Acknowledgements

 
This work was also supported by grants-in-aid for Scientific Research on Priority Areas from the Ministry of Education, Culture, Sports, Science and Technology of Japan and from the Japan Science and Technology Corporation (JST) (T.T. and S.S.).

	Footnotes

 
Communicated by: Eisuke Nishida

^* Correspondence: E-mail: takenawa{at}ims.u-tokyo.ac.jp

	References

Top Abstract Introduction Results Discussion Experimental procedures References

 
Ali, M.M., Roe, S.M., Vaughan, C.K., Meyer, P., Panaretou, B., Piper, P.W., Prodromou, C. & Pearl, L.H. (2006) Crystal structure of an Hsp90-nucleotide-p23/Sba1 closed chaperone complex. Nature 440, 1013–1017.[CrossRef][Medline]

Bartles, J.R. (2000) Parallel actin bundles and their multiple actin-bundling proteins. Curr. Opin. Cell Biol. 12, 72–78.[CrossRef][Medline]

Blanchoin, L., Amann, K.J., Higgs, H.N., Marchand, J.B., Kaiser, D.A. & Pollard, T.D. (2000) Direct observation of dendritic actin filament networks nucleated by Arp2/3 complex and WASP/Scar proteins. Nature 404, 1007–1011.[CrossRef][Medline]

Cameron, L.A., Footer, M.J., van Oudenaarden, A. & Theriot, J.A. (1999) Motility of ActA protein-coated microspheres driven by actin polymerization. Proc. Natl. Acad. Sci. USA 96, 4908–4913.[Abstract/Free Full Text]

Condeelis, J. (2001) How is actin polymerization nucleated in vivo? Trends Cell Biol. 11, 288–293.[CrossRef][Medline]

Derry, J.M., Ochs, H.D. & Francke, U. (1994) Isolation of a novel gene mutated in Wiskott–Aldrich syndrome. Cell 78, 635–644.[CrossRef][Medline]

Huai, Q., Wang, H., Liu, Y., Kim, H.Y., Toft, D. & Ke, H. (2005) Structures of the N-terminal and middle domains of E. coli Hsp90 and conformation changes upon ADP binding. Structure 13, 579–590.[Medline]

Isambert, H., Venier, P., Maggs, A.C., Fattoum, A., Kassab, R., Pantaloni, D. & Carlier, M.F. (1995) Flexibility of actin filaments derived from thermal fluctuations. Effect of bound nucleotide, phalloidin, and muscle regulatory proteins. J. Biol. Chem. 270, 11437–11444.[Abstract/Free Full Text]

Koyasu, S., Nishida, E., Kadowaki, T., Matsuzaki, F., Iida, K., Harada, F., Kasuga, M., Sakai, H. & Yahara, I. (1986) Two mammalian heat shock proteins, HSP90 and HSP100, are actin-binding proteins. Proc. Natl. Acad. Sci. USA 83, 8054–8058.[Abstract/Free Full Text]

Machesky, L.M., Mullins, R.D., Higgs, H.N., Kaiser, D.A., Blanchoin, L., May, R.C., Hall, M.E. & Pollard, T.D. (1999) Scar, a WASp-related protein, activates nucleation of actin filaments by the Arp2/3 complex. Proc. Natl. Acad. Sci. USA 96, 3739–3744.[Abstract/Free Full Text]

Miki, H., Miura, K. & Takenawa, T. (1996) N-WASP, a novel actin-depolymerizing protein, regulates the cortical cytoskeletal rearrangement in a PIP2-dependent manner downstream of tyrosine kinases. EMBO J. 15, 5326–5335.[Medline]

Miki, H., Sasaki, T., Takai, Y. & Takenawa, T. (1998a) Induction of filopodium formation by a WASP-related actin-depolymerizing protein N-WASP. Nature 391, 93–96.[CrossRef][Medline]

Miki, H., Suetsugu, S. & Takenawa, T. (1998b) WAVE, a novel WASP-family protein involved in actin reorganization induced by Rac. EMBO J. 17, 6932–6694.[CrossRef][Medline]

Miki, H. & Takenawa, T. (2003) Regulation of actin dynamics by WASP family proteins. J. Biochem. (Tokyo) 134, 309–313.[Abstract/Free Full Text]

Nishida, E., Koyasu, S., Sakai, H. & Yahara, I. (1986) Calmodulin-regulated binding of the 90-kDa heat shock protein to actin filaments. J. Biol. Chem. 261, 16033–16036.[Abstract/Free Full Text]

Pantaloni, D., Boujemaa, R., Didry, D., Gounon, P. & Carlier, M.F. (2000) The Arp2/3 complex branches filament barbed ends: functional antagonism with capping proteins. Nat. Cell Biol. 2, 385–391.[CrossRef][Medline]

Park, S.J., Suetsugu, S. & Takenawa, T. (2005) Interaction of HSP90 to N-WASP leads to activation and protection from proteasome-dependent degradation. EMBO J. 24, 1557–1570.[CrossRef][Medline]

Pollard, T.D. & Borisy, G.G. (2003) Cellular motility driven by assembly and disassembly of actin filaments. Cell 112, 453–465.[CrossRef][Medline]

Rohatgi, R., Ma, L., Miki, H., Lopez, M., Kirchhausen, T., Takenawa, T. & Kirschner, M.W. (1999) The interaction between N-WASP and the Arp2/3 complex links Cdc42-dependent signals to actin assembly. Cell 97, 221–231.[CrossRef][Medline]

Sasaki, Y., Hayashi, K., Shirao, T., Ishikawa, R. & Kohama, K. (1996) Inhibition by drebrin of the actin-bundling activity of brain fascin, a protein localized in filopodia of growth cones. J. Neurochem. 66, 980–988.[Medline]

Soeno, Y., Abe, H., Kimura, S., Maruyama, K. & Obinata, T. (1998) Generation of functional ß-actinin (CapZ) in an E. coli expression system. J. Muscle Res. Cell Motil. 19, 639–646.[CrossRef][Medline]

Suetsugu, S., Miki, H. & Takenawa, T. (1999) Identification of two human WAVE/SCAR homologues as general actin regulatory molecules which associate with the Arp2/3 complex. Biochem. Biophys. Res. Commun. 260, 296–302.[CrossRef][Medline]

Suetsugu, S., Yamazaki, D., Kurisu, S. & Takenawa, T. (2003) Differential roles of WAVE1 and WAVE2 in dorsal and peripheral ruffle formation for fibroblast cell migration. Dev. Cell 5, 595–609.[CrossRef][Medline]

Svitkina, T.M., Bulanova, E.A., Chaga, O.Y., Vignjevic, D.M., Kojima, S., Vasiliev, J.M. & Borisy, G.G. (2003) Mechanism of filopodia initiation by reorganization of a dendritic network. J. Cell Biol. 160, 409–421.[Abstract/Free Full Text]

Symons, M., Derry, J.M., Karlak, B., Jiang, S., Lemahieu, V., Mccormick, F., Francke, U. & Abo, A. (1996) Wiskott–Aldrich syndrome protein, a novel effector for the GTPase CDC42Hs, is implicated in actin polymerization. Cell 84, 723–734.[CrossRef][Medline]

Vignjevic, D., Yarar, D., Welch, M.D., Peloquin, J., Svitkina, T. & Borisy, G.G. (2003) Formation of filopodia-like bundles in vitro from a dendritic network. J. Cell Biol. 160, 951–962.[Abstract/Free Full Text]

Wear, M.A., Schafer, D.A. & Cooper, J.A. (2000) Actin dynamics: assembly and disassembly of actin networks. Curr. Biol. 10, R891–R895.[CrossRef][Medline]

Yamashiro-Matsumura, S. & Matsumura, F. (1986) Intracellular localization of the 55-kD actin-bundling protein in cultured cells: spatial relationships with actin, -actinin, tropomyosin, and fimbrin. J. Cell Biol. 103, 631–640.[Abstract/Free Full Text]

Received: 15 January 2007
Accepted: 7 February 2007

This article has been cited by other articles:

				R. Liu, M. T. Abreu-Blanco, K. C. Barry, E. V. Linardopoulou, G. E. Osborn, and S. M. Parkhurst Wash functions downstream of Rho and links linear and branched actin nucleation factors Development, August 15, 2009; 136(16): 2849 - 2860. [Abstract] [Full Text] [PDF]

				A. Antonov, C. Snead, B. Gorshkov, G. N. Antonova, A. D. Verin, and J. D. Catravas Heat Shock Protein 90 Inhibitors Protect and Restore Pulmonary Endothelial Barrier Function Am. J. Respir. Cell Mol. Biol., November 1, 2008; 39(5): 551 - 559. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Park, S. J. Articles by Takenawa, T. Search for Related Content PubMed Articles by Park, S. J. Articles by Takenawa, T.