Lasp-1 Regulates Podosome Function

Eukaryotic cells form a variety of adhesive structures to connect with their environment and to regulate cell motility. In contrast to classical focal adhesions, podosomes, highly dynamic structures of different cell types, are actively engaged in matrix remodelling and degradation. Podosomes are composed of an actin-rich core region surrounded by a ring-like structure containing signalling molecules, motor proteins as well as cytoskeleton-associated proteins. Lasp-1 is a ubiquitously expressed, actin-binding protein that is known to regulate cytoskeleton architecture and cell migration. This multidomain protein is predominantely present at focal adhesions, however, a second pool of Lasp-1 molecules is also found at lamellipodia and vesicle-like microdomains in the cytosol. In this report, we show that Lasp-1 is a novel component and regulator of podosomes. Immunofluorescence studies reveal a localization of Lasp-1 in the podosome ring structure, where it colocalizes with zyxin and vinculin. Life cell imaging experiments demonstrate that Lasp-1 is recruited in early steps of podosome assembly. A siRNA-mediated Lasp-1 knockdown in human macrophages affects podosome dynamics as well as their matrix degradation capacity. In summary, our data indicate that Lasp-1 is a novel component of podosomes and is involved in the regulation of podosomal function.


Introduction
Podosomes are highly dynamic adhesion structures that are constitutively formed in monocytic cells such as macrophages, dendritic cells, or osteoclasts, [1,2,3]. In addition, endothelial cells, smooth muscle cells, as well as glomerular podocytes, have been shown to form podosomes upon stimulation with cytokines, growth factors or phorbol esters [4,5,6,7,8]. Podosomes play a role in cell adhesion and matrix degradation, but their precise function in tissue invasion is still under consideration [9,10]. On the molecular level, podosomes consist of an actin-rich core and associated proteins embedded in a ring-like structure of plaque molecules and signalling proteins.
Recent studies have shown that the formation of podosomes occurs very rapidly and starts with a clustering of podosome initiation factors. Subsequently in these cell type a recruitment of the actin polymerization machinery and a later maturation of podosomes occur [11,12,13].
The Lasp-1 protein is composed of an aminoterminal LIM domain, followed by two F-actin-binding nebulin (NEBU) domains, a Linker region and a carboxyterminal SH3 domain [15,22]. The binding to F-actin is mainly mediated by the nebulin repeats of Lasp-1 [23]. The Lasp-1 SH3 domain interacts with zyxin, a mechanosensitive regulator of actin assembly, and controls its recruitment to focal adhesions [24,25]. A second identified SH3-binding partner of Lasp-1 is palladin, a protein which is involved in actin assembly, too [8,26].
Lasp-1 displays several phosphorylation motifs for cAMPdependent serine/threonine kinases (PKA/PKC) as well as a substrate-recognizing sequence for the Abelson tyrosine kinase (Abl) [14,27]. Furthermore, the subcellular distribution and physiological activity of Lasp-1 is controlled by phosphorylation at several sites [14,28]. For example, induced Lasp-1 phosphorylation in fibroblasts prevents its localization at focal contacts and promotes its perinuclear enrichment [17].
The data presented in this study demonstrate that Lasp-1 is a component of podosomes in primary human macrophages and activated rat smooth muscle (A7r5) cells. Live cell imaging analysis, in combination with a siRNA-mediated knockdown approach demonstrates that Lasp-1 influences several parameters of podosome dynamics and also regulates podosome function by influencing their matrix degradation capacity.

Cell culture, drug treatment and transfection
Human peripheral blood leukocytes were isolated by centrifugation of heparinized blood or buffy coats (kindly provided from F. Bentzien, University Medical Center Hamburg-Eppendorf, Germany) in Ficoll (Biochrom, Berlin, Germany). Monocytic cells were isolated with magnetic anti-CD14 antibody beads and MS+ separation columns (Miltenyi Biotec, Auburn, CA), according to the manufacturer's instructions and seeded onto glass coverslips at a density of 5610 4 or 35 mm plastic wells at a density of 1610 6 . Cells were cultured in RPMI containing 20% cell-free autologous serum at 37uC, 5% CO 2 and 90% humidity and differentiated into macrophages through plate adherence over 6 days. Autologous cell-free serum, which was isolated from whole blood using Serum Separation tubes (S-MonovetteH from Sarstedt, Nümbrecht, Germany), filtrated through StericupH Filter Units (Millipore) and used without previous heat inactivation. Medium was changed every 3-4 days.
A7r5 rat smooth muscle cells (ATCC) were grown on glass coverslips in high glucose Dulbecco's Modified Eagle Medium (DMEM) without phenol red, containing 10% fetal calf serum and 16 penicillin/streptomycin. For the expression of EGFP, mRFP or DsRed fusion proteins, A7r5 cells were transfected with plasmid DNA using Fugene HD reagent (Roche, Mannheim, Germany) following the manufacturer's instructions. 48 hours after transfection, podosome formation was induced by PDBu treatment for the indicated times. Transfection of macrophages was described previously [29,30].

Western blotting
Western blotting was performed using standard techniques [31]. Antibodies were diluted in blocking buffer containing 5% skim milk powder in PBS/Tween. After washing, the membranes were incubated with respective secondary antibodies coupled to horseradish peroxidase. Finally, the membranes were washed and developed using the Lumilight chemiluminiscence detection kit (Roche) and X-ray film developer. Podosome formation in A7r5 Lasp-1 knockdown cells was induced by PDBu treatment (1 mM for 30 min) 72 h posttransfection. Cells were fixed and stained with phalloidin-Alexa594 to visualize the actin core of podosomes. The number of podosomes per cell after siRNA treatment was determined and quantitatively analyzed using a student t-test. At least 35 cells per coverslip from three different experiments were counted.
Preparation of lysates from siRNA-transfected cells was carried out 78 h and 96 h posttransfection to determine the efficiency of the siRNA-mediated Lasp-1 knockdown using Western blotting analysis.

Immunofluorescence analysis
A7r5 cells grown on coverslips were fixed in 4% paraformaldehyde in PBS. Human macrophages were fixed with 3.7% formaldehyde in PBS and permeabilized in ice-cold acetone and 0.5% Triton X-100, respectively. Endogenous proteins were detected by subsequent incubation of the cells with primary and fluorochrome-coupled secondary antibodies for one hour at room temperature. After washing, specimen were mounted in Vectastain mounting medium (Vector Laboratories, Burlingame, CA) or Mowiol (Calbiochem, Schwalbach, Germany) containing pphenylendiamine (Sigma-Aldrich) as anti-fading reagent, and sealed with nail polish. Samples were analyzed on a Leica photomicroscope (Leica, Wetzlar, Germany) attached to a Spot 2 Slider digital camera (Meyer Instruments, Houston, TX). Confocal microscopy was performed on a confocal laser scanning microscope (Leica) or a LSM510 Meta microscope (Carl Zeiss, Jena, Germany). Substitution of primary antibodies by nonimmune serum served as negative controls.
For analyzing the number and diameter of podosomes, primary human macrophages were seeded on glass coverslips in a density of 1610 5 cells. After 6 h, the cells were fixed, permeabilized and stained with Alexa 488-coupled phalloidin for highlighting podosome cores. The numbers of podosomes were evaluated using ImageJ and GraphPad Prism software. In total, 168 cells were used from three different donors. For the diameter of the podosome cores, in total 24 cells from three different donors were analysed using Volocity 3D Image Analysis Software (PerkinElmer, Rodgau, Germany) and evaluated with Excel and GraphPad Prism software.
For analysing the number of podosomes contacted by MT1-MMP containing vesicles, MT1-MMP-mCherry was overexpressed for 18 h in primary human macrophages, then fixed, permeabilized and stained with Alexa 488 coupled phalloidin. The fluorescence intensity of single podosomes was measured using ImageJ software and evaluated with Excel and GraphPad Prism software. In total, 12 cells from three different donors were analysed.

Live cell imaging
24 h posttransfection, A7r5 cells were seeded onto 35 mm mdishes (ibidi, Martinsried, Germany) and intracellular dynamics of recombinant EGFP-, mRFP-or DsRed-tagged proteins were analyzed with the live cell imaging system Biostation IM (Nikon, Duesseldorf, Germany). Therefore, an internal sealed chamber with humidified 5% CO 2 and 37uC was used. Pictures from selected cells were taken every 30 sec over time periods of 5-30 min.
Primary human macrophages were seeded onto 12 mm glas bottom dishes (Willco Wells BV, Amsterdam, Netherlands). The dynamics of single podosomes or the area of all podosomes, were analysed 18 h posttransfection of Lifeact-TagGFP2. Pictures from single cells were taken every 10 sec over time periods of 30 min. For evaluation, in total eleven cells,from three different donors, were analysed, using Volocity 3D Image Analysis Software and GraphPad Prism. Rework of movies was done using After Effects CS5 (Adobe, Neu-Isenburg, Germany).

Matrix degradation assay
A matrix degradation assay was performed as described earlier [32]. In brief, gelatin (from swine, Roth, Karlsruhe, Germany) was fluorescently labeled with NHS-rhodamine (ThermoScientific, Rockford, IL). Coverslips were coated with labeled matrix solution, fixed in 0.5% glutaraldehyde and washed with 70% ethanol and medium. SiRNA-transfected human macrophages were seeded on coated coverslips with a density of 8610 4 cells and fixed after 6 and 24 h, respectively.
Quantification of matrix degradation was performed using ImageJ software. Values were determined by measurement of the fluorescence intensity of the matrix in a single cell mode. Values of cells transfected with control siRNA were set at 100%. For comparability, laser intensity was not changed between measurements. For each value, 3630 cells were evaluated. When indicated, differences between mean values were analyzed using the Student's t test. P,0.05 was considered as statistically significant.

Lasp-1 is a component of the podosomal ring
To analyze a putative podosomal localization of Lasp-1, we used indirect immunofluorescence and confocal microscopy ( Fig. 1A-E). In transfected human macrophages, Lasp-1 tagged with the enhanced green fluorescent protein (EGFP-Lasp-1) colocalized with endogenous vinculin at the ring structure of podosomes surrounding the actin core (Fig. 1A). Similar results were obtained with macrophages encoding vinculin-tagged enhanced yellow fluorescent protein (EYFP) and stained for endogenous Lasp-1 and F-actin (Fig. 1B). A more detailed analysis of single podosomes from macrophages using the Volocity software package confirmed that both endogenous as well as overexpressed Lasp-1 localize to the podosomal ring structure, where they partially colocalize with vinculin ( Fig. S1A/B).
Whereas macrophages generate podosomes constitutively, in A7r5 smooth muscle cells, podosome formation can be induced by stimulation of protein kinase C with the phorbol ester PDBu. Using this sytem, we observed that in unstimulated A7r5 cells, Lasp-1 was mainly found at focal adhesions that anchors stress fibres to the basal membrane (Fig. 1E). Concomitant with the PDBu-induced cytoskeletal rearrangement, Lasp-1 was enriched at podosomes in PDBu-treated A7r5 cells, where it colocalized with the podosome component zyxin (Fig. 1E).

Podosomal localization of Lasp-1 truncation mutants
Lasp-1 is a multidomain protein that can bind different proteins including F-actin as well as the known podosomal components zyxin and palladin ( Fig. 2A). To determine the specific Lasp-1 domain(s) that mediate(s) its podosomal recruitment, we analyzed the distribution of various EGFP-tagged Lasp-1 truncation mutants in transfected macrophages. EGFP fusion proteins encoding only isolated Lasp-1 domains (SH3, LIM, NEBU) displayed a diffuse cytosolic distribution indicating that a single Lasp-1 domain is not sufficient to establish a podosomal localization (data not shown). In contrast, EGFP-Lasp-1 truncation mutants that lack either the aminoterminal LIM domain, the internal NEBU repeats and the adjacent linker region, or the carboxyterminal SH3 domain all localized to podosomes of transfected macrophages (Fig. 2B), and also to vinculin-positive focal adhesions in A7r5 cells (Fig. S2A/B). These data demonstrate that a combination of at least two distinct functional domains is involved in the recruitment of Lasp-1 to podosomes.
The podosomal component cortactin is known to be crucial for early steps in podosome biogenesis as the clustering of initiation factors occurs [13]. When we compared DsRed-cortactin and EGFP-Lasp-1 in double-transfected, PDBu-treated A7r5 cells, both recombinant proteins displayed comparable dynamics during the rapid assembly process (Fig. 3C).

Lasp-1 knockdown, podosome formation and matrix degradation
To investigate the functional relevance of Lasp-1 for podosome formation, we used a siRNA-based approach to knock down Lasp-1 in PDBu-treated A7r5 smooth muscle cells. Compared to cells transfected with control siRNA, there was a notable reduction of the protein level of Lasp-1 in cells transfected with specific siRNA against Lasp-1 (Fig. S3A/B). Interestingly, podosomes were still apparent after Lasp-1 knockdown in PDBu-treated cells (Fig. S3B), indicating that Lasp-1 is not an essential factor for podosome assembly in A7r5 cells.

Lasp-1 Regulates Podosomes
PLoS ONE | www.plosone.org were still able to form podosomes with a core region that contains F-actin and the marker proteins Arp2 and gelsolin (Fig. S4B) as well as a podosome ring structure that contains vinculin and paxillin (Fig. S4C). These data suggest that the main structure of podosomes is not altered in Lasp-1-deficient macrophages.
Next, we measured whether a Lasp-1 knockdown in macrophages affects podosome size and number (Fig. 4). Cells that were treated with specific siRNA against Lasp-1 showed slightly more podosomes with a larger diameter of the core structure ( Fig. 4 D/ E). However, the number of podosomes per cell was decreased compared to cells treated with control siRNA (Fig. 4 F/G). Furthermore, overall dynamic of the podosome field per cell was quantified by measuring the total podosome-covered area per cell over time (Fig. 4 A-C; Video S1). Macrophages transfected with Lasp-1-specific siRNA (Oligo A or C) showed a higher variability of this value, compared to control cells, indicating increased overall podosome dynamics. By contrast, the size of these cells showed no difference to controls (Fig. 4

B/C)
To determine the role of Lasp-1 in the dynamics of single podosomes, we analyzed podosome lifetime, which includes 1) appearance to dissolution, 2) appearance to fission, and 3) fusion to dissolution of single podosomes (Fig. 5 A). Importantly, macrophages treated with Lasp-1 specific siRNA (Oligo A or C; Figure 5B) showed a significant decrease (approx. 30%) of mean podosome lifetime, and especially the number of podosomes showing short lifetimes (1-3 min) was significantly increased. (Fig. 5 C/D).
As the degradation of extracellular matrix (ECM) is one of the main functions of podosomes [33,34], we analyzed the putative effect of a Lasp-1 knockdown on the matrix degradation capacity. For this assay, we used macrophages pre-treated with control siRNA or siRNA targeting Lasp-1 (Oligo A or C), seeded on coverslips coated with fluorescent-labeled gelatin and analyzed after 6 hrs of degradation. After this period, a significant reduction of degradation could be observed in cells treated with siRNA against Lasp-1 (Fig. 6A). This could also be quantified by measuring the fluorescence intensity of the degraded area and resulted in a significant difference compared the degradation level of control cells (Fig. 6B).
Matrix degradation at podosomes requires the recruitment of matrix metalloproteinases including MT1-MMP [34]. To test whether Lasp-1 is involved in this recruitment process, we analyzed the amount of MT1-MMP-mCherry-containing vesicles at podosomes from control or Lasp-1 siRNA-treated macrophages. No effect of a Lasp-1 knockdown on the podosomal localization of MT1-MMP was detected (Fig. S5A/B).
The actin-binding protein Lasp-1 is known to localize at stable actin-rich structures like focal adhesions and stress fibres, but can also be found at highly dynamic dorsal membrane ruffles [15,17]. These findings, together with the already known interaction between Lasp-1 and zyxin and palladin, led us to investigate whether Lasp-1 is a component of podosomes, too.
In the current study, we observed Lasp-1 localization at podosomes in smooth muscle cells and human macrophages, respectively ( Fig. 1; Fig. S1). Our immunofluorescence analyses revealed, that Lasp-1 is localized in the ring structure of podosomes and displays a distribution that is similar to that of other adhesion plaque proteins such as vinculin, zyxin and paxillin [1,12,37]. Our data from experiments with Lasp-1 truncation mutants demonstrated that a proper podosomal localization requires the combinantion of at least two functional domains of the protein. Neither the NEBU repeats that are known to associate with F-actin, nor the SH3 domain that binds to paxillin and zyxin were sufficient or necessary to target a EGFP fusion protein to podosomes (Fig. 2). These findings are in line with a recent study demonstrating, that various truncation mutants of Lasp-1 lacking different domains are still recruited to focal adhesions [17].
We observed a comparable localization of Lasp-1 and the early podosome marker cortactin at sites of initial podosome formation (Fig. 3B). Cortactin is known to be crucial for the initiation of actin polymerization at pre-podosome structures [5,12,13,38,39]. As Lasp-1 and cortactin display similar dynamics during podosome biogenesis, we speculated that Lasp-1 is associated with early stages of podosome biogenesis, too. To prove this, we used a siRNA-based approach to knock down Lasp-1 in PDBu-treated A7r5 cells (Fig. S3). Interestingly, we observed no differences in the overall number of podosomes in Lasp-1 knockdown A7r5 cells. However, in human macrophages with a decreased Lasp-1 expression, we observed alterations in several parameters of podosomes: decreased lifetime, smaller diameter and decreased podosome numbers per cell. Moreover, also the degradation capacity of podosomes was diminished in these cells (Fig. 5/6). These data point to potential cell type-specific differences in the recruitment or regulation of both structural and functional podosome components. In this context, it should also be mentioned that macrophages form podosomes constitutively, whereas podosome formation in A7r5 smooth muscle cells is induced by stimulating PKC, and thus not directly comparable. In a similar scenario, knockdown of cortactin in carcinoma cells resulted in decreased matrix degradation ability of the podosomerelated invadopodia [40]. In these cells, secretion of the matrix metalloproteinases (MMP) MMP-2 and MMP-9 as well as cell surface exposure of the transmembrane isoform MT1-MMP was found to closely correlate with cortactin expression levels. Although we did not detect a direct correlation between Lasp-1 expression and the recruitment of MT1-MMP-containing vesicles to podosomes, our results are consistent with the idea that Lasp-1 may be important for the local release of lytic enzymes at matrix degrading podosomes in macrophages.
In summary, our study shows that Lasp-1 is a novel component of the podosomal ring structure. Although Lasp-1 is recruited to podosomes at early stages of their assembly, the protein is probably not necessary for the initiation of podosome formation. However, Lasp-1 is involved in of the regulation of several podosome   parameters including size, number and lifetime and also regulates the matrix degradation capacity of podosomes. These activities reveal Lasp-1 as a novel important regulator of podosomes and also point to Lasp-1 as a potential target for the modulation of invasive cell migration. Figure S1 Lasp-1 is a component of the podosome ring structure. 3D reconstruction of single podosomes of primary human macrophages. Fluorescence micrographs (shown and described in Figure 1 C, D) were converted in silico into an isosurface mode using Volocity 3D Image Analysis Software. Left panels: xyz mode, right panels: xzy mode for the same podosome. It is important to note that common localization of proteins to either core or ring structures of podosomes does not necessarily imply exact colocalization, as apparently subdomains exist within both structures. Thus, both vinculin and Lasp-1 clearly localize to the podosome ring, but they only overlap partially. or two different oligonucleotides (lanes 2 (Oligo A) and 4 (Oligo C)) targeting Lasp-1 mRNA. (B,C) Confocal micrographs of primary human macrophages treated with control siRNA or Lasp-1specific siRNA (Oligo A or C). Cells were fixed and stained with Alexa568-phalloidin for F-actin (highlighting podosome cores; red) and with Arp2-or gelsolin-specific antibodies (with Alexa488conjugated secondary antibody) for Arp2 or gelsolin (podosome cores; green) or with vinculin-or paxillin-specific antibodies (with Alexa488-conjugated secondary antibody) for vinculin and paxillin (podosome ring structure; green), respectively. White bars indicate 10 mm. Video S1 Knockdown of Lasp-1 mediates higher variability of the area of podosomes in macrophages. Confocal time lapse series of primary human macrophages pretreated with control and Lasp-1-specific siRNA (Oligo A or C), respectively, expressing Lifeact-TagGFP2 (green) highlighting Factin (exposure time at 488 nm: 359 ms, acquisition rate: 1 image/30 sec; frame rate: 10 f/s; sequence: 30 min). Note higher variability of the area of podosomes of cells treated with Lasp-1 specific siRNA compared to control cell. (MOV)