The cortactin-binding domain of WIP is essential for podosome formation and extracellular matrix degradation by murine dendritic cells

European Journal of Cell Biology 90 (2011) 213–223

Contents lists available at ScienceDirect

European Journal of Cell Biology journal homepage: www.elsevier.de/ejcb

The cortactin-binding domain of WIP is essential for podosome formation and extracellular matrix degradation by murine dendritic cells a ˜ Inmaculada Banón-Rodríguez , James Monypenny b , Chiara Ragazzini a , Ana Franco a , Yolanda Calle c , b Gareth E. Jones , Inés M. Antón a,∗ a b c

Cellular and Molecular Department, Centro Nacional de Biotecnología – CSIC, Madrid 28049, Spain Randall Division of Cell & Molecular Biophysics, King’s College London, London SE1 1UL, UK Haematology Dept., King’s College London, London SE5 9NU, UK

a r t i c l e

i n f o

Article history: Received 29 March 2010 Received in revised form 5 August 2010 Accepted 1 September 2010 Keywords: WIP (WASP interacting protein) Cortactin Podosome Metalloproteinase Dendritic cell Matrix degradation Gelatin Fibronectin

a b s t r a c t In immature dendritic cells (DCs) podosomes form and turn over behind the leading edge of migrating cells. The Arp2/3 complex activator Wiskott-Aldrich Syndrome Protein (WASP) localises to the actin core of forming podosomes together with WASP-Interacting Protein (WIP). A second weaker Arp2/3 activator, cortactin, is also found at podosomes where it has been proposed to participate in matrix metalloproteinase (MMP) secretion. We have previously shown that WIP−/− DCs are unable to make podosomes. WIP binds to cortactin and in this report we address whether WIP regulates cortactin-mediated MMP activity. Using DCs derived from splenic murine precursors, we found that wild-type cells were able to localise MMPs at podosomes where matrix degradation takes place. In contrast, WIP−/− DCs remain able to synthesise MMPs but do not degrade the extracellular matrix. Infection of WIP KO DCs with lentivirus expressing WIP restored both podosome formation and their ability to degrade the extracellular matrix, implicating WIP-induced podosomes as foci of functional MMP location. When WIP KO DCs were infected with a mutant form of WIP lacking the cortactin-binding domain (WIP110-170) DCs were only able to elaborate disorganised podosomes that were unable to support MMP-mediated matrix degradation. Taken together, these results suggest a role for WIP not only in WASP-mediated actin polymerisation and podosome formation, but also in cortactin-mediated extracellular matrix degradation by MMPs. © 2010 Elsevier GmbH. All rights reserved.

Introduction Normal cell behaviour relies on the capacity to respond to environmental cues including modifications in extracellular matrix (ECM) composition. Cell attachment to the ECM induces intracellular signalling that initiates at foci of cell-matrix contacts. Such contacts are fundamental aspects of cell and tissue organisation, and are mediated through specific adhesion receptors, normally integrins, that are linked to the cytoskeleton through defined signalling pathways (Adams, 2002). During cell migration, actin-rich dynamic protrusions known as filopodia and lamellipo-

Abbreviations: DCs, dendritic cells; Arp, actin related protein; WASP, Wiskott-Aldrich syndrome protein; WIP, WASP interacting protein; MMP, metalloproteinase; ECM, extracellular matrix; SH3, Src homology 3; N-WASP, neural WASP; PLL, poly-l-lysine; FN, fibronectin; MT1-MMP, membrane type 1-MMP; HS1, hematopoietic lineage cell-specific protein-1. ∗ Corresponding author at: CNB-CSIC, Darwin 3, Campus Cantoblanco, 28049 Madrid, Spain. Tel.: +34 91 585 5312; fax: +34 91 585 4506. ˜ E-mail addresses: [email protected] (I. Banón-Rodríguez), james.monypenny of [email protected] (J. Monypenny), [email protected] (C. Ragazzini), [email protected] (A. Franco), [email protected] (Y. Calle), [email protected] (G.E. Jones), [email protected] (I.M. Antón). 0171-9335/$ – see front matter © 2010 Elsevier GmbH. All rights reserved. doi:10.1016/j.ejcb.2010.09.001

dia are both generated and retracted at the leading edge where cells make transient cell-substratum contacts (focal complexes) that may develop into more stable adhesions (focal adhesions) linked to the actomyosin-based contractile machinery. Although these are the most investigated adhesion structures, in the case of monocytes and lineage-related leukocytes, adhesion and migration on a planar matrix is achieved through formation of podosomes. These actin-rich structures can also be induced in endothelial cells, smooth muscle cells and in Src-transformed fibroblasts (Gimona et al., 2008). The dense, filamentous actin core of podosomes contain Wiskott-Aldrich syndrome protein (WASP) and Actin Related Protein 2/3 (Arp 2/3) complex proteins surrounded by a ring of adhesion related molecules including vinculin, paxillin, talin, fimbrin, gelsolin, vimentin, and numerous adaptor molecules associated with integrin signalling (Burns et al., 2001; Linder, 2007). In migrating macrophages, osteoclasts and dendritic cells (DCs), podosomes localise behind the leading edge, playing a role in cell polarity, locomotion and ECM degradation (Brunton et al., 2004; Calle et al., 2004). In migrating immature DCs podosomes are highly dynamic structures with a half-life between 30 s and 10 min depending upon migratory status (Burns et al., 2004). Although the mechanisms of regulation of this cyclical turnover are still largely unknown, regulated proteolysis of the underlying matrix

214

I. Ba˜ nón-Rodríguez et al. / European Journal of Cell Biology 90 (2011) 213–223

(Adams, 2002), by secreted and membrane bound matrix metalloproteinases (MMPs), at sites of podosome formation have been implicated in their function (Linder, 2007). MMPs are endopeptidases capable of cleaving a wide range of extracellular substrates, thus regulating the activity of such moieties (Gill and Parks, 2008). MMP activity favours the release of growth factors from the cell membrane or the ECM, the cleavage of growth factor receptors from the cell surface, the shedding of cell adhesion molecules and the activation of other MMPs. It is now well established that MMP-mediated degradation contributes to essential physiological functions related to cell migration such as tissue repair, immune response, regulation of chemokine activity, diapedesis and inflammation (Clark et al., 2008). Cortactin is an actin assembly protein that functions in both the activation and stabilisation phases of branched actin assembly by the Arp2/3 complex (Tehrani et al., 2007; Uruno et al., 2001), and it is present at sites of dynamic F-actin assembly in cellular protrusions such as lamellipodia, invadopodia (Yamaguchi and Condeelis, 2007) and podosomes (Tehrani et al., 2006; Webb et al., 2006). It has been previously shown that the inhibition of invadopodia structure assembly by cortactin depletion results in the inhibition of matrix degradation due to a failure of invadopodia formation (Artym et al., 2006). More recently it has been proposed that cortactin has a critical function promoting the secretion of MMPs in invadopodia (Clark and Weaver, 2008). Cortactin has a Src homology 3 (SH3) domain that interacts with WASP interacting protein (WIP) and this interaction increases the efficiency of cortactin-mediated Arp 2/3 complex activation of actin polymerisation in a concentration-dependent manner (Kinley et al., 2003). Apart from cortactin, WIP interacts with other actin-linked proteins including WASP, N-WASP, Nck and myosin, as well as with actin itself (Anton et al., 2007; Krzewski et al., 2006). WIP localization at podosomes was first described in aortic endothelial cells (Moreau et al., 2003) and later confirmed in monocytes, DCs and osteoclasts (Chabadel et al., 2007; Chou et al., 2006; Tsuboi, 2006). WIP is essential for podosome formation, more specifically for the formation of actin cores containing WASP and cortactin (Chou et al., 2006). However, to date no information on the potential role of WIP in protease activity and ECM degradation is available. In the present work, we investigate the possible role of WIP in regulated proMMP2/MMP2 and proMMP9/MMP9 secretion, where cortactin seems to have an essential role, and the function of the WIP–cortactin interaction in podosome formation and function in DCs. Our results demonstrate that in contrast to WT DCs, DCs derived from WIP KO mouse spleen are able to synthesise MMPs but do not secrete these MMPs. The observed defects in podosome formation and MMP activity are restored by infection of WIP KO DCs with lentivirus encoding WIP but not by virus encoding a mutant form of WIP lacking the cortactin-binding domain. We conclude that WIP is necessary for podosome formation and its interaction with cortactin is essential for ECM degradation.

Materials and methods Cell culture DCs were generated from mouse spleens as previously described (West et al., 1999). Briefly, spleens from 6- to 12-week-old WT and WIP KO SV129/BL6 mice (Anton et al., 2002) were homogenised through a cell strainer to obtain a cell suspension. Cells were washed twice with RPMI (Sigma, UK) containing 1% heat-inactivated foetal bovine serum (FBS) and then resuspended in DC medium (RPMI supplemented with 10% FBS, 1 mM pyruvate (Sigma, UK), 1× non-essential amino acids (Sigma, UK), 2 mM glutamine (Sigma, UK), 50 ␮M 2-ME (Gibco BRL), 20 ng/ml recom-

binant mouse GM-CSF (R&D Systems) and 1 ng/ml recombinant human TGF-␤ (R&D Systems)) and plated at a density of 2 × 106 cells/ml in 75 cm2 culture flasks at 37 ◦ C in a 5% CO2 atmosphere. After 5 days of culture, 7 ml fresh medium were added per flask and at day 8, the cells in suspension were collected, replated and kept in suspension in fresh medium. After a total of 17–18 days ex vivo, 80–90% of the cells in culture were DCs as determined by the expression of CD11c and DEC205 by FACS analysis (data not shown). Cell viability before experimental assays was tested by Trypan Blue exclusion. The work was carried out in accordance with EC Directive 86/609/EEC for animal experiments. Murine DCs, unlike murine platelets that express both cortactin and HS1 (Thomas et al., 2007), only express cortactin. The 293T cells used for lentivirus production were cultured using DMEM (Sigma, UK) supplemented with 10% FBS and 2 mM glutamine at 37 ◦ C in a 5% CO2 atmosphere. The THP-1 human monocytic leukaemia cell line was cultured using RPMI (Sigma, UK) supplemented with 10% FBS and 2 mM glutamine at 37 ◦ C in a 5% CO2 atmosphere. This cell line expresses the variant cortactin HS1 (hematopoietic lineage cell-specific protein-1) but does not express cortactin (data not shown). Reagents and antibodies Monoclonal antibody to cortactin (clone 4F11) was purchased from Millipore and monoclonal antibody to human HS1 (clone 9/HS1) from BD Transduction Laboratories. Anti-vinculin (hVIN1) was purchased from Sigma. Cy5-conjugated anti-mouse IgG antibody and Alexa Fluor 568 phalloidin and Alexa Fluor 488 phalloidin were obtained from Molecular Probes. GAPDH monoclonal antibody was purchased from AbD Serotec. GFP antibody (11814460001) was purchased from Roche. MMP2 rabbit antibody (MMP2/2C1) was from Abcam, MMP9 (GE-213) was from Genetex and MT1-MMP (M5808) from Sigma. Horseradish peroxidase (HRP)-labelled anti-mouse and anti-rabbit antibodies were purchased from Dako. Rhodamine B-isothiocyanate (R-1755) and gelatin (type A from porcine skin, G-2500) were purchased from Sigma and rhodamine fibronectin (FN) from Cytoskeleton. Zymography In vitro differentiated murine DCs were plated in 24-well plates (5 × 104 per well) previously coated with 10 ␮g/ml poly-l-lysine (PLL) or 10 ␮g/ml FN. After 2 h, growth medium was substituted for serum-free medium. Cells were maintained overnight (ON) at 37 ◦ C in a 5% CO2 atmosphere. After 24 h, medium was collected and boiled with non-reducing sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) buffer. SDS-PAGE fractionation of protein was carried out on 0.75 mm, 10% (w/v) acrylamide gels containing 0.075% (w/v) gelatin. Gels were soaked in 2.5% (v/v) Triton X-100 with gentle shaking at room temperature (RT) for 30 min to remove the SDS and to allow protein renaturation. The gel was rinsed once with substrate buffer (0.05 M Tris–HCl, pH 8.0; 1 mM CaCl2 ; 0.02% (w/v) NaN3 ) and incubated in fresh substrate buffer at 37 ◦ C ON. After incubation the gel was stained in Coomassie blue stain with gentle shaking for 30 min, followed by destaining in distilled water until suitable visualisation of digested gelatin bands was achieved. Quantification of pixels intensity in zymographies was performed using Image J Software. Western blotting DCs plated onto PLL or FN were lysed in lysis buffer containing 0.1% Triton X-100, 150 mM NaCl, 50 mM Tris–HCl pH 7.4, 1 mM EDTA, 1 mM EGTA, with protease (Complete from Merck) and

I. Ba˜ nón-Rodríguez et al. / European Journal of Cell Biology 90 (2011) 213–223

phosphatase inhibitors (50 mM NaF, 1 mM Na3 VO4 and 1 mM okadaic acid). Twenty micrograms of total cell lysate protein was loaded per lane in an 8% gel and subjected to reducing SDSPAGE. Proteins were blotted onto nitrocellulose membranes using a Bio-Rad Mini protein II transfer apparatus. For the detection of MMP2, MT1-MMP and GAPDH, blots were blocked with 5% dried milk solution diluted in TBS-T (10 mM Tris–HCl, pH 7.5, 100 mM NaCl, 0.1% Tween 20), and incubated with indicated antibody. For MMP9 detection the samples were prepared in nonreducing buffer and the blots were blocked with 2% BSA solution diluted in TBS-T. In all cases, signal was detected with horseradish peroxidase-conjugated secondary antibodies and enhanced chemiluminescence (ECL) detection system. Matrix degradation assay Fluorescent gelatin Gelatin (Sigma) was labelled with rhodamine B Isothiocyanate (Sigma) by dissolving the gelatin in sodium borohydrate buffer (50 mM Na2 B4 O7 (Aldrich), 61 mM NaCl, pH 9.3) for 1 h at 37 ◦ C and incubation with rhodamine (36 ␮g/ml) for 2 h at RT in the darkness. The buffer was changed to phosphate buffered saline (PBS) by extensive dialysis at 4 ◦ C over 2 days, followed by a quick spin to remove insoluble material. Sucrose was added to the sample to a final concentration of 2.5% gelatin/2.5% sucrose, aliquoted and stored for up to 21 days at 4 ◦ C. To coat coverslips, gelatin/sucrose in PBS was warmed to 37 ◦ C and added as a fine film, followed by crosslinking with 0.5% glutaraldehyde in PBS. Coverslips were washed 3× with PBS, and then incubated with 5 mg/ml sodium borohydride (Aldrich) in PBS for 3 min. Coverslips were then washed gently 3× in PBS and sterilised in 70% ethanol for 5 min, dried and quenched in RPMI for 1 h at 37 ◦ C. Gelatin from pig skin, Oregon Green 488 conjugate was purchased from Invitrogen. Fluorescent fibronectin After centrifugation at 16,000 rpm for 5 min to remove aggregates, a 50 ␮g/ml solution of rhodamine FN (Cytoskeleton) was prepared in PBS and incubated in the dark for 1 h at RT onto the coverslips coated with cross-linked gelatin as described above. The coverslips were sterilized with 70% ethanol, washed in RPMI, and equilibrated with DC medium for 30 min before the addition of the cells. For matrix degradation assays, 5 × 104 cells were suspended in 2 ml of DC medium, and seeded onto gelatin or FN-coated coverslips (that had been previously distributed into the wells of a 24-well plate) for 6 h at 37 ◦ C. The cells were fixed in 4% paraformaldehyde in PBS, permeabilised with 0.5% Triton X-100 in PBS and blocked with 3% bovine serum albumin in PBS + 0.1% Tween-20 and incubated with appropriate primary and secondary antibodies or fluorescent phalloidin. Coverslips were mounted onto slides using Vectashield mounting medium (Vector Laboratories, UK) and visualised using a Zeiss LSM 510 Meta confocal laser scanning head attached to a Zeiss Axioplan 2 microscope. LSM 510 software was used to collect four sequential images from four separate optical sections in the z axis 0.5 ␮m apart. The same software was used to obtain merged confocal images. Immunoprecipitation and pull down assays For co-immunoprecipitation experiments, DCs were plated on FN and lysed 24 h later in lysis buffer containing 150 mM NaCl, 50 mM Tris–HCl pH 7.4, 0.4% Triton X-100, 1 mM EDTA, 1 mM EGTA, with protease (Complete from Merck) and phosphatase inhibitors (50 mM NaF, 1 mM Na3 VO4 and 1 mM okadaic acid). Cell lysates were incubated with 4F11 antibody (anticortactin) coupled to Sepharose A/G resin (Sigma) ON at 4 ◦ C.

215

For co-immunoprecipitation experiments from THP-1 lysates, cells infected with recombinant lentivirus (pLVeGFP, pLVWIP-eGFP or pLVWIP110-170WIP-eGFP constructs) were resuspended in lysis buffer and the soluble extracts incubated with anti-GFP antibody coupled to Sepharose A resin (Sigma) for 2 h at 4 ◦ C. Infection of DCs using a lentiviral system Recombinant lentiviral stocks were produced in 293FT cells by co-transfecting the transfer vector pLVeGFP, pLVWIP-eGFP, pLNT/Sffv-mCherry, pLNT/Sffv-WIP-mCherry or pLVWIP110170-eGFP, the envelope plasmid pMD.2G, and the packaging plasmid pCMVR8.91, as previously described (Zufferey et al., 1997). Cells (1.5 × 107 ) were seeded onto 150 cm2 flasks and transfected with 10 ␮g DNA envelope, 30 ␮g DNA packaging and 40 ␮g DNA transfer vector by precomplexing with 0.125 ␮M PEI (22 kDa) for 15 min at RT in OptiMEM. After 4 h at 37 ◦ C the medium was replaced with fresh DMEM 10% FBS and virus particles were harvested 48 and 72 h post-transfection. After filtering through a 0.45 ␮m-pore-size filter, the virus suspension was concentrated by centrifugation at 50,000 × g for 2 h at 4 ◦ C. The resulting pellet was resuspended in RPMI (Sigma, UK) and stored at −80 ◦ C until use. The desired number of DCs were plated in complete culture medium as described above using phenol-free RPMI (Sigma, UK) and concentrated lentivirus was added to the cells at an MOI (multiplicity of infection) of 100 and incubated for 24 h. Medium was replaced with complete DC culture medium after 24 h and cells were cultured for a further 48 h to allow maximal expression of lentiviral vectors before being used for experiments. Seven to 10% of the DCs were infected according to GFP expression. Results WIP is required for the secretion of proMMP9 and proMMP2 in DCs In murine DCs endogenous WIP localises to the core of podosomes and its absence modifies membrane and adhesion structure dynamics such that podosomes are not formed at all, or in a small minority of cases, poorly organised podosome-like structures are elaborated (Chou et al., 2006). Instead, WIP−/− DCs form large focal adhesion-like structures (Chou et al., 2006). Podosomes are specialised adhesion structures where ECM degradation takes place (Linder, 2007) after presumptive localised MMP secretion. In order to test WIP participation in the secretion of pro- or active MMPs to the extracellular medium, we performed zymography analysis. Supernatants from spleen-derived control or WIP-null DC cultures were subjected to non-reducing electrophoresis in a gelatin-polyacrylamide gel. The results shown in Fig. 1A and C reveal that the absence of WIP results in a defect in gelatinase activity present in the supernatants of DC cultures. Based on their gelatinase activity and their apparent molecular weight, the two stronger bands should correspond to proactive forms of MMP9 (pro-MMP9, 92 kDa) and MMP2 (pro-MMP2, 72 kDa). In order to confirm their nature, control supernatants were incubated in the presence of APMA (aminophenylmercuric acetate) to activate latent zymogen and induce pro-MMP to MMP transition. APMA treatment clearly promotes the digestion of pro-MMP9 (or pro-MMP2) to render active MMP9 (or MMP2) (Fig. 1B). The quantification of pro-MMP secretion (as determined by relative pixel intensity using Image J software) indicates that in the case of WT DCs the FN-dependent activation of integrins induces the secretion of both pro-MMP9 (1.52-fold increase; p < 0.01) and pro-MMP2 (1.26-fold increase; p < 0.05) (Fig. 1A and C). Integrin activation also promotes MMP2 secretion by WT DCs. In contrast, WIP KO

216

I. Ba˜ nón-Rodríguez et al. / European Journal of Cell Biology 90 (2011) 213–223

Fig. 1. WIP is implicated in pro-MMP9 and pro-MMP2 secretion but not in MMP production. Spleen-derived WT and WIP-deficient (KO) DCs are plated onto PLL or FN and the resultant culture supernatants are subjected to gelatin zymography. WT DCs secrete gelatinases (pro-MMP9 and pro-MMP2) when they are plated on PLL and this secretion is incremented by integrin activation when cells are plated on FN. Minor amounts of active MMP9 and MMP2 are detected in supernatants from FN-plated WT DCs. However, WIP KO DCs has a severely reduced capability for pro-MMP/MMP secretion even when integrins are activated. (B) Supernatants from WT DCs plated onto FN-coated wells are incubated in the presence or absence of APMA and subjected to gelatin zymography. APMA treatment increases the generation of active MMP9 and MMP2. (C) Quantification of pro-MMP9 and pro-MMP2 secretion by WT and KO DCs (n = 5). (D) Total cell lysates from DCs plated on PLL or FN were separated by SDS-PAGE and Western blotted. Immunoblotting with the corresponding antibodies showed that total intracellular levels of MMP9, MMP2 or MT1-MMP were not different between the WT DCs and the KO DCs. GAPDH represents the protein loading control.

DCs secrete minimal amounts of either of these pro-MMPs even in the presence of integrin-activating FN ligand (Fig. 1A and C). To discount the possibility that the impaired secretion is a secondary effect derived from reduced synthesis of MMP in WIP KO DCs, we measured the intracellular levels of MMP9 and MMP2 in total cell lysates by western blotting. Intracellular levels of MMP9 and MMP2 are equivalent in WT and WIP KO DCs (Fig. 1D). Latent form of MMP2 (pro-MMP2) requires digestion by MT1MMP (membrane type 1-MMP) in order to be rendered active (Barbolina and Stack, 2008). Therefore, we studied if the reduced MMP2 activity in the supernatants from WIP KO DCs was due to a defect in the production of MT1-MMP. By Western blotting we saw that, as with MMP9 and MMP2, there was no defect in the production of pro-MT1-MMP/MT1-MMP by WIP KO DCs (Fig. 1D). Taken together these results support a role for WIP in pro-MMP/MMP secretion and exclude the possibility that the impaired gelatinase activity observed in WIP−/− supernatants was a secondary effect due to abnormal RNA transcription or protein synthesis of MMPs. WIP and cortactin are implicated in the gelatinase activity of MMPs by DCs In WIP-null DCs, podosomes are substituted by adhesion complexes and focal adhesions, which are significantly less dynamic than podosomes. MMP-mediated ECM degradation is an essential component for the correct turnover of cell adhesion structures at the substratum (Adams, 2002). In order to study the matrix degrading capability of WIP-null DCs, we used fluorescent gelatin as a substratum for cell attachment. We confirmed that MMP activ-

ity in WT DCs is concentrated at podosomes, which is believed to be important for the rapid dissolution of substratum adhesions required for efficient cell migration (Fig. 2A and B). The majority of WT DCs (74.8 ± 12.7%, Fig. 2E) were able to degrade the extracellular gelatin and the quantification of the overall degradation capability (as determined by relative pixel intensity using Image J software) reveals that WT DCs degrade 44.7 ± 28.5 arbitrary units per cell. In contrast, only 23.3 ± 12.8% of WIP−/− DCs (Fig. 2E) degrade the labelled gelatin to such a minor extent such that overall degradation capability is reduced to 1.3 ± 6.5 units in the case of WIP KO DCs (p = 0.01; two-tailed Student’s t-test). In WIP KO DCs, none of the alternative actin-rich structures observed in 80% of the cells, such as the actin puncta thought to be malformed podosome-like structures or focal complexes and focal adhesions (Chou et al., 2006), were able to secrete MMPs and degrade the underlying gelatin matrix (Fig. 2C and D) to significant levels. These data support the argument that the presence of WIP is necessary not only for the correct organisation of the actin cytoskeleton but it is also controlling the appropriate targeting of MMP activity. It has been recently proposed that cortactin is essential for the localised secretion of key proteases at podosome-related structures named invadopodia that are found in invasive tumour cells (Clark and Weaver, 2008). Immunofluorescence analysis of WT DCs plated onto fluorescent gelatin using a monoclonal antibody against cortactin, shows the presence of a high concentration of cortactin in the core of the podosomes (Fig. 2A). Since an interaction between WIP and cortactin has also been demonstrated (Kinley et al., 2003), we considered if absence of WIP expression would disrupt the localisation of cortactin to adhesion structures. In WIP KO DCs cortactin is found distributed throughout the cell and is not concentrated in

I. Ba˜ nón-Rodríguez et al. / European Journal of Cell Biology 90 (2011) 213–223

217

Fig. 2. WIP deficiency prevents gelatin degradation by murine DCs without affecting cortactin-MMP2 complex formation. Scale bar, 5 ␮m. (A) WT DCs are plated on fluorescent gelatin (red) and subjected to immunofluorescence analysis with phalloidin-Alexa488 (green) and anti-cortactin (blue). MMPs from WT DCs can degrade gelatin at the podosome-rich area. Cortactin (blue) is localised throughout the cell but specially concentrated in these structures where ECM degradation by MMPs occurs. (B) Z stack showing how WT DCs can go into the gelatin. (C) WIP KO DCs are plated on fluorescent gelatin (red) and subjected to immunofluorescence analysis as in A. WIP KO DCs cannot degrade the gelatin and cortactin (blue) is localised throughout the cell. (D) Z stack showing how WIP KO DCs cannot penetrate the gelatin. (E) Graph showing the quantification of the percentage of DCs from WT and KO origin with capability to degrade gelatin (n = 80). (F) Co-immunoprecipitation analysis of cortactin-MMP2 interaction. DC lysates (WT or KO) are subjected to immunoprecipitation (IP) analysis with anti-cortactin or negative control antibody (IgG). Immunoadsorption indicates that cortactin binds MMP2 even in the absence of WIP and, although WIP KO DCs cannot secrete MMP2, cortactin is still associated with it.

the various actin-rich structures that formed in these cells (Fig. 2C and Chou et al., 2006). Hence, we decided to test: (a) if cortactin associates with MMPs, and (b) if the absence of WIP affects this association and by extension the activity or localisation of MMPs via the actin rich structures (podosomes) missing in WIP KO DCs. By

co-immunoprecipitation analyses we found that cortactin is in fact associated with MMP2 either in the presence or in the absence of WIP (Fig. 2F). Consequently, the inability of WIP KO DCs to degrade ECM via gelatinase activity of MMPs is not due to an impaired capacity for cortactin to bind MMPs.

218

I. Ba˜ nón-Rodríguez et al. / European Journal of Cell Biology 90 (2011) 213–223

WIP is essential for podosome formation, but its association with cortactin is dispensable to the initiation of podosomes We have previously shown that in the absence of WIP murine DCs fail to form podosomes and instead focal adhesions are assembled (Chou et al., 2006). To test if reconstitution of WIP expression was enough to recover podosome formation in WIP KO DCs, we infected DCs with a recombinant lentivirus expressing either the fusion protein WIP-eGFP (pLVWIP-eGFP) or only eGFP (pLVeGFP) as a control (Fig. 3A and B). As expected, the infection of WIP KO DCs with the control lentivirus pLVeGFP had no effect on the cytoskeleton or morphology of these cells (Fig. 3A). Infected cells are not able to induce podosome formation or to polarise. Aggregates of F-actin are occasionally observed which are distributed randomly throughout the cell but none of the infected cells are able to recover the phenotype of a WT DC. Similarly, vinculin remained distributed throughout the cell and it did not organise itself around F-actin aggregates (Fig. 3A). In contrast, infection of WIP KO DCs with pLVWIP-eGFP results in the recovery of the normal DC phenotype. Sixty percent of the cells that are infected with the lentivirus and thus express WIP-eGFP can polarise as normal and form normally organised clusters of podosomes localised at the leading edge (Fig. 3B). Using immunofluorescence techniques we confirmed the appropriate protein organisation of the podosome core and ring. The WIPrescued DCs form podosomes equivalent to those found in control murine DCs containing an actin core (red in Fig. 3B) surrounded by a well-defined vinculin ring at its base (blue in Fig. 3B). Moreover, for the first time, we describe the distribution of WIP-eGFP in murine DC podosomes co-localising with the actin core (green in Fig. 3B). It has previously been demonstrated the in vivo interaction of cortactin and WIP (Chou et al., 2006; Kinley et al., 2003; Tehrani et al., 2007). The proline-rich region of WIP is the binding region for the cortactin SH3 domain as deletion of amino acids 110–170 of WIP prevents their interaction (Kinley et al., 2003). The WIP-cortactin complex has been implicated in podosome formation, however the role of each protein in this process is still unclear. To elucidate how the association of both proteins affects podosome formation, we generated a recombinant lentivirus expressing a mutant form of WIP in which the cortactin-binding domain has been deleted (pLVWIP110-170-eGFP) (Fig. 3E). By co-immunoprecipitation we confirm that this deletion impairs the association of WIP with cortactin in 293T cells (data not shown) and with HS1, the hematopoietic cell-specific homolog of cortactin expressed by the monocytic cell line THP-1 that lacks cortactin (Fig. 3D). WIP110170-eGFP mutant still retains WASP binding capability (Fig. 3D). After infecting WIP KO DCs with this mutant virus we observed that a substantial minority of the infected cells (30.77%) completely fail to form podosomes, whereas the remaining 69.23% (a representative cell is shown in Fig. 3C) form podosome-like structures with actin cores and vinculin rings around them. However the structures formed by pLVWIP110-170-eGFP infected WIP KO DCs are smaller in size and often more poorly organised (lacking the honeycomb distribution of control vinculin rings) in comparison to podosomes in WT DCs or WIP KO DCs expressing full length WIP-eGFP. Quantification of the area of the core after phalloidinTRITC staining (44.4 ± 7.9 a.u. in WIP-eGFP versus 19.2 ± 9.6 a.u. in WIP110-170-eGFP, p < 0.001) and of the F-actin intensity in the core (73.7 ± 8.9 a.u. in WIP-eGFP versus 20.4 ± 1.8 a.u. in WIP110170-eGFP, p = 0.007) showed that both parameters are significantly lower for the mutant podosomes. Thus from our data it seems likely that cortactin is not involved in podosome initiation but instead implicated in later steps of podosome maturation or stabilisation.

Cortactin-dependent MMP activity requires WIP We have previously demonstrated the importance of WIP in podosome formation (Chou et al., 2006), while others have suggested an important role for cortactin in MMP secretion (Clark and Weaver, 2008). Here we show that although WIPcortactin binding has a role in defining the formation of proper podosome, it is not essential for initiating podosome-like formation. Hence, we decided to examine the functionality of the structures formed when WIP KO DCs are infected with the cortactin binding domain mutant lentivirus pLVWIP110-170-eGFP. We again used fluorescent gelatin to determine the ability of these podosome-like structures to degrade ECM and therefore to elucidate the role of the WIP-cortactin association in MMP-mediated ECM degradation. The infection of WIP KO DCs with pLNT/Sffv-WIPmCherry restores the formation of well-defined podosomes and 68.75 ± 10.18% of the infected cells (n = 30, 3 independent experiments) are able to actively degrade fluorescent gelatin (Fig. 4A). In contrast, WIP KO DCs infected with the control lentivirus pLNT/Sffv-mCherry or pLVeGFP (data not shown) or with the mutant lentivirus pLVWIP110-170-eGFP lacking the cortactinbinding domain were not able to degrade the ECM (28.57 ± 10.4% of infected cells degraded gelatin, n = 30, 3 independent experiments; Fig. 4B and C). It is unfortunate that the reduced efficiency of the lentiviral infection (only 7–10% of the DC are transduced at detectable levels) prevents the biochemical analysis via zymography of the culture supernatants to confirm the activity of secreted MMP. These results suggest that WIP is essential not only for correct podosome formation, but it is also implicated in the functionality of these structures through its ability to bind cortactin, thus bringing MMPs to the correct location at cell-substratum adhesions where they will be localised and can degrade the matrix. WIP contributes to the degradation of extracellular fibronectin Our zymographic analysis indicated that WIP deficiency diminished the capability of murine DCs to secrete proMMP/MMPs to culture supernatants while the degradation assay supported a role for WIP in the localised digestion of fluorescent gelatin (Figs. 1 and 2). Although these results could suggest that the gelatinolytic effect observed is mostly due to MMP2 and MMP9, from these data we cannot exclude the possibility that the function of other MMPs with gelatinase activity is affected by the absence of WIP. MT1-MMP is a transmembrane protein able to degrade substrates such as gelatin, FN and laminin (Ohuchi et al., 1997). In contrast, MMP2 and MMP9 lack the proteolytic activity required for FN degradation. In order to test if WIP contribution to ECM degradation is broader and not limited to gelatin-rich matrices, we analysed if WIP−/− DCs retained the capability to degrade FN. Control and WIP−/− DCs were plated onto coverslips coated with fluorescent FN and their degradation capability imaged. While 48.8 ± 16.3% of control cells degrade the labelled FN, only 16.5 ± 4.6% of WIP-deficient DCs were able to do so (Fig. 5). Therefore, the vast majority of WIP−/− DCs lack the capability to degrade extracellular FN. These data indicate that WIP contributes to the degradation of ECM of different composition e.g. gelatin or FN. Discussion In this study we present data supporting the participation of WIP in regulating ECM degradation by MMPs synthesised by murine DCs. Using a combination of imaging and biochemical techniques, we have demonstrated that DCs derived from WIP KO mouse spleen were able to synthesise MMPs but not able to degrade

I. Ba˜ nón-Rodríguez et al. / European Journal of Cell Biology 90 (2011) 213–223

219

Fig. 3. The cortactin-binding domain of WIP is essential for the formation of defined podosomes. Scale bar, 5 ␮m. (A) WIP KO DCs infected with pLVeGFP (green) maintain the typical morphology of a WIP KO DC devoid of podosomes, and they show actin dots (red) and disorganised vinculin (blue). Lateral images represent magnified views of the selected boxed areas. (B) WIP KO DCs infected with pLVWIP-eGFP can restore WIP expression (green) and by this way the formation of well-defined podosomes with an actin core (red) and a vinculin ring (blue). WIP is localised in the podosome core (green). The non-infected WIP KO DC (lower area) maintains the formation of vinculin-rich focal contacts instead of podosomes. (C) Infection of WIP KO DCs with pLVWIP110-170-eGFP induces the expression of a WIP mutant (green) that cannot bind cortactin but can induce podosome-like formation with actin dots (red) and diffuse vinculin rings (blue). (D) GFP immunoprecipitation (IP) of soluble lysates from THP-1 cells infected with pLVeGFP, pLVWIP-eGFP or pLVWIP110-170-eGFP lentivirus and posterior blotting with anti-HS1 (the hematopoietic cell-specific homologue of cortactin expressed by THP-1 cells) showing that the mutant WIP110-170-eGFP cannot bind it, with anti-WASP (showing that both proteins can bind endogenous WASP), and with anti-GFP (control showing protein precipitation). GFP and fusion proteins are marked with asterisks. (E) Schematic representation of the lentivirus constructs used in the study.

220

I. Ba˜ nón-Rodríguez et al. / European Journal of Cell Biology 90 (2011) 213–223

Fig. 4. Binding of WIP to cortactin has a critical role in MMP activity to degrade gelatin. Infected KO DCs are plated on fluorescent gelatin and subjected to immunofluorescence analysis. (A) WIP KO DCs infected with pLNT/Sffv-WIP-mCherry can restore WIP expression (pseudocolour green) and degrade the underlying fluorescent gelatin (Oregon Green 488, pseudocolour red). Actin staining is shown in blue. (B) Infection of WIP KO DCs with pLVWIP110-170-eGFP (green) does not restore fluorescent gelatin degradation (red). Actin staining is shown in blue. (C) Graph showing the quantification of the percentage of KO DCs expressing WIP or the mutant WIP110-170 with capability to degrade gelatin (p = 0.0085).

gelatin or FN. The fact that the matrix-degrading function of both secreted and matrix-associated MMPs is affected, may suggest that WIP contribute to correctly target MMPs to the cell-matrix interface resulting in reduced capability to degrade ECM. The gelatin degradation defect was restored by lentivirus-mediated expression of full length WIP but not of a mutant form of WIP lacking the cortactin-binding domain. Therefore, we propose that formation of properly organised podosomes, where matrix degradation takes place, depends on WIP binding to cortactin. Cortactin has been reported as an actin assembly scaffolding protein that functions in both the activation and stabilisation phases of branched actin assembly mediated by the Arp 2/3 complex (Tehrani et al., 2007; Uruno et al., 2001). Cortactin localises to sites of dynamic actin assembly in cellular invasive protrusions such as podosomes and invadopodia (Chou et al., 2006; Webb et al., 2006; Yamaguchi and Condeelis, 2007) where WIP is also found in abundance. WIP is capable of modulating WASP-dependent actin dynamics through controlling WASP stability, activation, and subcellular localisation (Anton et al., 2007). Cortactin and WIP display an overlapping cortical localisation in cells and it has been previously demonstrated that cortactin binds to a sequence (aa 136–205) of WIP through its SH3 domain (Kinley et al., 2003). A deletion variant of WIP lacking the proline-rich stretch spanning amino acids 110–170 does not interact with endogenous cortactin, indicating the importance of this domain in WIP binding to cortactin (Kinley et al., 2003). Potential mechanisms through which cortactin may link vesicular trafficking and dynamic branched actin assembly to regulated protease secretion in invadopodia-associated ECM degradation have been previously proposed (Clark and Weaver, 2008). Cortactin overexpression enhances cell migration (Bryce et al., 2005; Hill et al., 2006; Huang et al., 1998; Patel et al., 1998), and cells where cortactin has been knocked down using siRNA show motility

defects in both two-dimensional and three-dimensional migration assays (Clark and Weaver, 2008; Clark et al., 2007). Similarly, the conditional genetic disruption of murine cortactin causes defects in random and directed cell migration (Lai et al., 2009). While cortactin knockdown is accompanied by a decreased number of invadopodia puncta per cell (40% decrease), the ability of these cells to degrade ECM is completely inhibited, a finding that fits with our observations for those cells expressing WIP110-170-eGFP in WIP KO DCs. This truncated form of WIP does not bind cortactin (or HS1 in THP-1 cells), contributes to podosome initiation but does not support MMP activity (Figs. 3 and 4). These results may point to a common pathway for WIP and cortactin that is required for late steps involved in podosome maturation and MMP function. Lentivirus-mediated WIP-eGFP expression in primary mouse DCs corroborates previous immunofluorescence data demonstrating localisation of WIP in the actin-rich podosome core (Chou et al., 2006). Furthermore, we have demonstrated that expression of WIP-eGFP or WIP-mCherry in WIP KO DCs restores normal cell morphology and functionality through the restoration of podosomes capable of active ECM degradation. In addition to its role in podosome formation, a role for WIP in monocyte/macrophage chemotaxis has been reported (Tsuboi, 2006, 2007). However, no reports on the contribution of WIP to DC motility exist. The turnover of adhesions in WT DC, as previously determined by IRM (interference reflection microscopy), is far greater than that found for the highly stable focal complexes/adhesions formed in WIP KO DCs (Chou et al., 2006). Our unpublished data suggest that the higher stability of the focal adhesion structures formed in the absence of WIP results in an impaired migratory phenotype, regardless of integrin activation. As MMP secretion modulates podosome lifetime in osteoclasts (Goto et al., 2002) and also DC migration, we sought to determine the potential role of WIP in MMP secretion, since no previous link has been reported between WIP and MMPs. Zymography

I. Ba˜ nón-Rodríguez et al. / European Journal of Cell Biology 90 (2011) 213–223

221

Fig. 5. Fibronectin degradation at podosomes requires WIP expression. (A) Scale bar, 25 ␮m. WT DCs are plated on fluorescent FN (red) and subjected to immunofluorescence analysis with phalloidin-Alexa488 (green) and anti-cortactin (blue). MMPs from WT DCs can degrade FN at the podosome-rich area. Cortactin (blue) is localised throughout the cell but specially concentrated in these structures where MMPs degrade ECM. (B) Scale bar, 10 ␮m. WIP KO DCs are plated on fluorescent FN (red) and subjected to immunofluorescence analysis as in A. WIP KO DCs cannot degrade the FN (red) and cortactin (blue) is localised throughout the cell. (C) Graph showing the quantification of the percentage of DCs from WT and KO origin with capability to degrade FN (n = 80).

analysis revealed that in the absence of WIP the capability of DCs to secrete pro-MMP9 and pro-MMP2 is severely reduced independently of integrin activation. Even when integrins are activated by plating cells on FN, a condition that increases basal MMP secretion in control DCs, WIP KO DCs show almost identical and minimal levels of secreted pro-MMPs. MT1-MMP can proteolytically activate MMP2 (van Hinsbergh et al., 2006) and it is detected together with MMP9 at podosomes in osteoclasts (Delaisse et al., 2000; Sato et al., 1997) and with MMP2 at podosomes in endothelial cells (Osiak et al., 2005; Tatin et al., 2006). However we found neither difference in MT1-MMP synthesis between WT and WIP KO DCs nor in MT1-MMP localisation at plasma membrane (Fig. 1 and data not shown). In spite of the lack of MMP secretion in WIP−/− cells, intracellular levels of MMPs are not increased (Fig. 1) so no storage of pro-MMPs is seen in these cells. This is probably due to the regulatory feedback mechanism that exists for regulating appropriate levels of MMP production (Linder, 2007). It is already known that WIP binds cortactin through its prolinerich domain and one of the proposed roles of cortactin at sites of cell adhesion to ECM is to promote the secretion of proteases. Using fluorescent gelatin and FN we observed that in the presence of WIP, cortactin localises to podosomes where the underlying ECM (gelatin or FN) is degraded. We also observed that in the absence of WIP, DCs were unable to degrade gelatin or FN, and cortactin is found distributed throughout the cell. Moreover, in this study we have demonstrated that cortactin and MMP2 are in the same complex even in WIP KO DCs (Fig. 2F) so WIP plays no role in structurally linking cortactin to MMPs within cells. Cortactin localises to endocytic vesicles and Golgi membranes (Cao

et al., 2005; Kaksonen et al., 2000) and interacts with membraneassociated signalling proteins and dynamic actin on endosomal vesicles. It is therefore possible that a transient interaction between cortactin and complexes containing MMPs occurs at vesicles. This interaction could be indirect and studies are required to define its nature. There are reports showing the importance of cortactin in podosome formation (Tehrani et al., 2006). Using a lentivirus encoding a mutant form of WIP lacking the cortactin-binding domain, we examined the role of cortactin-WIP binding in podosome formation. Our findings favour a role for cortactin in podosomes acquiring their completely defined structure rather than a direct role in podosome initiation, since WIP KO DCs expressing this cortactin-binding deficient mutant can produce podosome-like structures. The actin cores of these structures were smaller and the vinculin rings were more diffused than those found in WT DCs. The functionality of these reconstituted podosomes was assessed using the fluorescent gelatin assay, and our results support the hypothesis that binding of WIP to cortactin is essential for the recruitment of MMPs to podosomes and consequently for the localised degradation of gelatin. At this time we cannot exclude the possibility that MMP activity in WIP KO DCs is impeded by the absence of fully formed podosomes rather than the specific lack of WIP-cortactin interaction. The identification of novel WIP mutants unable to bind cortactin but able to reconstitute complete podosomes would be an excellent tool to address this question. The present study has focused on DC-mediated ECM degradation, where podosomes, which represent highly specialised adhesion structures, enable rapid, invasive cell migration due to

222

I. Ba˜ nón-Rodríguez et al. / European Journal of Cell Biology 90 (2011) 213–223

their rapid turnover and capacity to promote ECM degradation. Efficient migration requires fine coordination between formation and turnover of podosomes. However, the mechanisms regulating the rapid turnover of podosomes are largely unknown. Previous studies have implicated calpain in podosome disassembly through the regulated degradation of proteins such as talin, and WASP, which are responsible for podosome formation (Calle et al., 2006). MMPs have also been implicated in DCs migration (West et al., 2008), and podosomes have been pointed as potential sites of localised MMP secretion (Gimona et al., 2008). WIP, WASP and cortactin are all involved in the two main degradative processes that participate in cell migration: calpain-mediated proteolysis (intracellular) and ECM degradation (extracellular). WASP and cortactin are both calpain substrates during cell migration while WIP protects WASP from digestion (Calle et al., 2006). This information suggests that the temporal and spatial control of both calpain action and MMP secretion need fine coordination to allow for cell migration. From our data it would seem reasonable to postulate a model in which cortactin is dispensable for podosome initiation but once made, cortactin participates in final definition of the invasive structures and has a unique role as a mediator of MMP activity in ECM degradation. In parallel, WIP’s role would be central to podosome formation by binding to and recruiting WASP to foci of podosome initiation, and through its binding to cortactin to also promote stabilisation of actin filaments and function as a mediator of MMPmediated matrix degradation. Integration of integrin-mediated signalling pathways that ultimately promote migration and tissue invasion mediated by ECM degradation following correctly targeted MMP function can only be accomplished in the presence of WIP. In conclusion, we propose an extended role for WIP in DCs where it participates in podosome formation, cell polarisation and MMP activity. WIP is not only essential for the organisation of F-actin (following recruitment of WASP and activation of Arp2/3) and the clustering of integrins and associated proteins that are needed for podosome formation, but it is also required for DC polarisation in cooperation with cortactin (Chou et al., 2006). Possibly the last essential step for matrix degradation includes WIP association with cortactin and MMP recruitment to podosomes, structures responsible for ECM degradation. Acknowledgements We thank Paul Fraylich (Randall Division) for excellent technical assistance, Ester Martín (Randall Division) for help with gelatin assays and Esther García and Alejandra Bernardini for experimental assistance. We are grateful to Angeles García-Pardo and Javier ˜ for helpul suggestions and manuscript comments. Redondo-Munoz This work was supported by grants CSIC-CAM (CCG08-CSIC/SAL3471), CSIC (PIE200720I002), and Spanish Ministry of Education and Science (BFU2007-64144) to IMA. IB is a recipient of a contract from CAM (Comunidad Autónoma de Madrid) and an EMBO short-term fellowship, and AF received a FPU MEC fellowship. GEJ acknowledges Support from the Medical Research Council. YC acknowledges support from The John and Holly Burton myeloma research programme. References Adams, J.C., 2002. Regulation of protrusive and contractile cell-matrix contacts. J Cell Sci 115, 257–265. Anton, I.M., de la Fuente, M.A., Sims, T.N., Freeman, S., Ramesh, N., Hartwig, J.H., Dustin, M.L., Geha, R.S., 2002. WIP deficiency reveals a differential role for WIP and the actin cytoskeleton in T and B cell activation. Immunity 16, 193–204. Anton, I.M., Jones, G.E., Wandosell, F., Geha, R., Ramesh, N., 2007. WASP-interacting protein (WIP): working in polymerisation and much more. Trends Cell Biol 17, 555–562.

Artym, V.V., Zhang, Y., Seillier-Moiseiwitsch, F., Yamada, K.M., Mueller, S.C., 2006. Dynamic interactions of cortactin and membrane type 1 matrix metalloproteinase at invadopodia: defining the stages of invadopodia formation and function. Cancer Res 66, 3034–3043. Barbolina, M.V., Stack, M.S., 2008. Membrane type 1-matrix metalloproteinase: substrate diversity in pericellular proteolysis. Semin Cell Dev Biol 19, 24–33. Brunton, V.G., MacPherson, I.R., Frame, M.C., 2004. Cell adhesion receptors, tyrosine kinases and actin modulators: a complex three-way circuitry. Biochim Biophys Acta 1692, 121–144. Bryce, N.S., Clark, E.S., Leysath, J.L., Currie, J.D., Webb, D.J., Weaver, A.M., 2005. Cortactin promotes cell motility by enhancing lamellipodial persistence. Curr Biol 15, 1276–1285. Burns, S., Cory, G.O., Vainchenker, W., Thrasher, A.J., 2004. Mechanisms of WASpmediated hematologic and immunologic disease. Blood 104, 3454–3462. Burns, S., Thrasher, A.J., Blundell, M.P., Machesky, L., Jones, G.E., 2001. Configuration of human dendritic cell cytoskeleton by Rho GTPases, the WAS protein, and differentiation. Blood 98, 1142–1149. Calle, Y., Carragher, N.O., Thrasher, A.J., Jones, G.E., 2006. Inhibition of calpain stabilises podosomes and impairs dendritic cell motility. J Cell Sci 119, 2375– 2385. Calle, Y., Chou, H.C., Thrasher, A.J., Jones, G.E., 2004. Wiskott-Aldrich syndrome protein and the cytoskeletal dynamics of dendritic cells. J Pathol 204, 460–469. Cao, H., Weller, S., Orth, J.D., Chen, J., Huang, B., Chen, J.L., Stamnes, M., McNiven, M.A., 2005. Actin and Arf1-dependent recruitment of a cortactin-dynamin complex to the Golgi regulates post-Golgi transport. Nat Cell Biol 7, 483–492. Chabadel, A., Banon-Rodriguez, I., Cluet, D., Rudkin, B.B., Wehrle-Haller, B., Genot, E., Jurdic, P., Anton, I.M., Saltel, F., 2007. CD44 and beta3 integrin organize two functionally distinct actin-based domains in osteoclasts. Mol Biol Cell 18, 4899–4910. Chou, H.C., Anton, I.M., Holt, M.R., Curcio, C., Lanzardo, S., Worth, A., Burns, S., Thrasher, A.J., Jones, G.E., Calle, Y., 2006. WIP regulates the stability and localization of WASP to podosomes in migrating dendritic cells. Curr Biol 16, 2337–2344. Clark, E.S., Weaver, A.M., 2008. A new role for cortactin in invadopodia: regulation of protease secretion. Eur J Cell Biol 87, 581–590. Clark, E.S., Whigham, A.S., Yarbrough, W.G., Weaver, A.M., 2007. Cortactin is an essential regulator of matrix metalloproteinase secretion and extracellular matrix degradation in invadopodia. Cancer Res 67, 4227–4235. Clark, I.M., Swingler, T.E., Sampieri, C.L., Edwards, D.R., 2008. The regulation of matrix metalloproteinases and their inhibitors. Int J Biochem Cell Biol 40, 1362–1378. Delaisse, J.M., Engsig, M.T., Everts, V., del Carmen Ovejero, M., Ferreras, M., Lund, L., Vu, T.H., Werb, Z., Winding, B., Lochter, A., Karsdal, M.A., Troen, T., Kirkegaard, T., Lenhard, T., Heegaard, A.M., Neff, L., Baron, R., Foged, N.T., 2000. Proteinases in bone resorption: obvious and less obvious roles. Clin Chim Acta 291, 223–234. Gill, S.E., Parks, W.C., 2008. Metalloproteinases and their inhibitors: regulators of wound healing. Int J Biochem Cell Biol 40, 1334–1347. Gimona, M., Buccione, R., Courtneidge, S.A., Linder, S., 2008. Assembly and biological role of podosomes and invadopodia. Curr Opin Cell Biol 20, 235–241. Goto, T., Maeda, H., Tanaka, T., 2002. A selective inhibitor of matrix metalloproteinases inhibits the migration of isolated osteoclasts by increasing the life span of podosomes. J Bone Miner Metab 20, 98–105. Hill, A., McFarlane, S., Mulligan, K., Gillespie, H., Draffin, J.E., Trimble, A., Ouhtit, A., Johnston, P.G., Harkin, D.P., McCormick, D., Waugh, D.J., 2006. Cortactin underpins CD44-promoted invasion and adhesion of breast cancer cells to bone marrow endothelial cells. Oncogene 25, 6079–6091. Huang, C., Liu, J., Haudenschild, C.C., Zhan, X., 1998. The role of tyrosine phosphorylation of cortactin in the locomotion of endothelial cells. J Biol Chem 273, 25770–25776. Kaksonen, M., Peng, H.B., Rauvala, H., 2000. Association of cortactin with dynamic actin in lamellipodia and on endosomal vesicles. J Cell Sci 113 (Pt 24), 4421–4426. Kinley, A.W., Weed, S.A., Weaver, A.M., Karginov, A.V., Bissonette, E., Cooper, J.A., Parsons, J.T., 2003. Cortactin interacts with WIP in regulating Arp2/3 activation and membrane protrusion. Curr Biol 13, 384–393. Krzewski, K., Chen, X., Orange, J.S., Strominger, J.L., 2006. Formation of a WIP-, WASp, actin-, and myosin IIA-containing multiprotein complex in activated NK cells and its alteration by KIR inhibitory signaling. J Cell Biol 173, 121–132. Lai, F.P., Szczodrak, M., Oelkers, J.M., Ladwein, M., Acconcia, F., Benesch, S., Auinger, S., Faix, J., Small, J.V., Polo, S., Stradal, T.E., Rottner, K., 2009. Cortactin promotes migration and platelet-derived growth factor-induced actin reorganization by signaling to Rho-GTPases. Mol Biol Cell 20, 3209–3223. Linder, S., 2007. The matrix corroded: podosomes and invadopodia in extracellular matrix degradation. Trends Cell Biol 17, 107–117. Moreau, V., Tatin, F., Varon, C., Genot, E., 2003. Actin can reorganize into podosomes in aortic endothelial cells, a process controlled by Cdc42 and RhoA. Mol Cell Biol 23, 6809–6822. Ohuchi, E., Imai, K., Fujii, Y., Sato, H., Seiki, M., Okada, Y., 1997. Membrane type 1 matrix metalloproteinase digests interstitial collagens and other extracellular matrix macromolecules. J Biol Chem 272, 2446–2451. Osiak, A.E., Zenner, G., Linder, S., 2005. Subconfluent endothelial cells form podosomes downstream of cytokine and RhoGTPase signaling. Exp Cell Res 307, 342–353. Patel, A.S., Schechter, G.L., Wasilenko, W.J., Somers, K.D., 1998. Overexpression of EMS1/cortactin in NIH3T3 fibroblasts causes increased cell motility and invasion in vitro. Oncogene 16, 3227–3232. Sato, T., del Carmen Ovejero, M., Hou, P., Heegaard, A.M., Kumegawa, M., Foged, N.T., Delaisse, J.M., 1997. Identification of the membrane-type matrix metalloproteinase MT1-MMP in osteoclasts. J Cell Sci 110 (Pt 5), 589–596.

I. Ba˜ nón-Rodríguez et al. / European Journal of Cell Biology 90 (2011) 213–223 Tatin, F., Varon, C., Genot, E., Moreau, V., 2006. A signalling cascade involving PKC, Src and Cdc42 regulates podosome assembly in cultured endothelial cells in response to phorbol ester. J Cell Sci 119, 769– 781. Tehrani, S., Faccio, R., Chandrasekar, I., Ross, F.P., Cooper, J.A., 2006. Cortactin has an essential and specific role in osteoclast actin assembly. Mol Biol Cell 17, 2882–2895. Tehrani, S., Tomasevic, N., Weed, S., Sakowicz, R., Cooper, J.A., 2007. Src phosphorylation of cortactin enhances actin assembly. Proc Natl Acad Sci USA 104, 11933–11938. Thomas, S.G., Calaminus, S.D., Auger, J.M., Watson, S.P., Machesky, L.M., 2007. Studies on the actin-binding protein HS1 in platelets. BMC Cell Biol 8, 46. Tsuboi, S., 2006. A complex of Wiskott-Aldrich syndrome protein with Mammalian verprolins plays an important role in monocyte chemotaxis. J Immunol 176, 6576–6585. Tsuboi, S., 2007. Requirement for a complex of Wiskott-Aldrich Syndrome Protein (WASP) with WASP interacting protein in podosome formation in macrophages. J Immunol 178, 2987–2995.

223

Uruno, T., Liu, J., Zhang, P., Fan, Y., Egile, C., Li, R., Mueller, S.C., Zhan, X., 2001. Activation of Arp2/3 complex-mediated actin polymerization by cortactin. Nat Cell Biol 3, 259–266. van Hinsbergh, V.W., Engelse, M.A., Quax, P.H., 2006. Pericellular proteases in angiogenesis and vasculogenesis. Arterioscler Thromb Vasc Biol 26, 716–728. Webb, B.A., Eves, R., Mak, A.S., 2006. Cortactin regulates podosome formation: roles of the protein interaction domains. Exp Cell Res 312, 760–769. West, M.A., Antoniou, A.N., Prescott, A.R., Azuma, T., Kwiatkowski, D.J., Watts, C., 1999. Membrane ruffling, macropinocytosis and antigen presentation in the absence of gelsolin in murine dendritic cells. Eur J Immunol 29, 3450–3455. West, M.A., Prescott, A.R., Chan, K.M., Zhou, Z., Rose-John, S., Scheller, J., Watts, C., 2008. TLR ligand-induced podosome disassembly in dendritic cells is ADAM17 dependent. J Cell Biol 182, 993–1005. Yamaguchi, H., Condeelis, J., 2007. Regulation of the actin cytoskeleton in cancer cell migration and invasion. Biochim Biophys Acta 1773, 642–652. Zufferey, R., Nagy, D., Mandel, R.J., Naldini, L., Trono, D., 1997. Multiply attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat Biotechnol 15, 871–875.