QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed 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 Pfaff, M. Articles by Jurdic, P. Search for Related Content PubMed PubMed Citation Articles by Pfaff, M. Articles by Jurdic, P.

Podosomes in osteoclast-like cells

structural analysis and cooperative roles of paxillin, proline-rich tyrosine kinase 2 (Pyk2) and integrin Vß3

Martin Pfaff^*, and Pierre Jurdic

Ecole Normale Supérieure de Lyon, 46 allée d'Italie, 69364 Lyon Cedex 07, France
^* Present address: Institut Albert-Bonniot, LEDAC (UMR5538), Faculté de Medecine, F 38706 La Tronche Cedex, France
Author for correspondence (e-mail: martin.pfaff{at}ujf-grenoble.fr )

View larger version (38K): [in a new window]   Fig. 6. Binding of paxillin and Pyk2 to synthetic peptides containing the C terminus of the integrin ß3 cytoplasmic domain. (A) Integrin-derived sequences of the peptides: the C terminus of the chicken integrin ß3 tail, two variants of this sequence containing single amino acid substitutions (S752P (SP), Y759A (YA)), and the corresponding region of the integrin ß1 tail were linked at their N termini to penetratin, which was itself biotinylated at its N terminus. (B) Affinity precipitation experiment performed with these peptides and cell lysates of chicken osteoclast precursors. Proteins bound to the peptides were eluted together with these peptides into reducing sample buffer for SDS-polyacrylamide electrophoresis, and analyzed by immunoblotting with antibodies against talin, Pyk2 and paxillin. Peptides (Mr approx. 4.5 kDa) were detected on separate Coomassie-stained 15% SDS-polyacrylamide gels (Coom.). Supernatants of the binding reaction were also analyzed (unbound). Note the partial depletion in the ß3-unbound fraction of Pyk2 and paxillin, but not of talin. (C) Affinity precipitation with wild-type and S752P-mutated integrin ß3-tail peptides performed with chicken osteoclast-like cells that had been lysed either after 2 hour suspension culture (Susp) or after 40 minutes adhesion to serum-coated plastic (Adh). Bound and non-bound fractions were analyzed by immunoblotting with an antiphosphotyrosine mAb (PY99), or with antibodies against paxillin and Pyk2. Note that equal amounts of paxillin and Pyk2 were bound, although the levels of tyrosine phosphorylation differed strongly in the lysates of suspended versus adhered cells. (D) Affinity precipitation with the ß3-tail peptide performed with lysates of chicken osteoclast-like cells. Half of the lysates had been subjected to two rounds of immunoprecipitation with an anti-paxillin monoclonal antibody before the incubation with the ß3-tail peptide. Lysates and bound fractions were analyzed by immunoblotting with antibodies against Pyk2 and paxillin. Note that equal amounts of Pyk2 bound from paxillin-containing and paxillin-depleted lysates.

View larger version (55K): [in a new window]   Fig. 8. Binding of recombinant structural mimics of human integrin ß1 and ß3 cytoplasmic tails to proteins in lysates of chicken osteoclast precursor cells. Integrin sequences of the recombinant proteins are depicted at the bottom, including those containing a tyrosine-alanine point mutation at a conserved NPXY motif. Via an N-terminal His-Tag sequence, these proteins were bound to a Ni2+-resin and incubated with cell lysates. Bound cellular proteins were eluted together with the recombinant proteins from the resin and analyzed by western blotting with antibodies against Pyk2, Talin and Paxillin. Coomassie Blue staining (Coom.) was used to verify that comparable amounts of recombinant proteins were present in each experiment. Supernatants of the binding reactions were also analyzed (non-bound). (C, control Ni2+-resin used without any added His-tagged protein; n.d., not done).

View larger version (78K): [in a new window]   Fig. 1. The actin cytoskeleton in chicken osteoclast precursors during in vitro osteoclast differentiation. After an initial trypsinization (see Materials and Methods), macrophagelike cells were cultured on glass coverslips for one day (left panel) or for additional four days in the absence (middle panel) or presence (right panel) of RANKL-ODF. Subsequently, cells were fixed, permeabilized, stained with rhodamine-phalloidin and photographed on a Zeiss immunofluorescence microscope. Note the profound changes in the cellular distribution and in the density of podosomes. Scale bar: 10 µm.

View larger version (100K): [in a new window]   Fig. 3. Phosphotyrosine-detection in podosomes. Chicken (A,B) or human (C) osteoclast precursors cultured in the absence (A) or presence (B,C) of RANKL-ODF were double-stained with the antiphosphotyrosine antibody 4G10 and rhodamine-phalloidin and analyzed by confocal microscopy. (B) The cell peripheries of three chicken osteoclasts, which surround an undifferentiated mononuclear cell located at the upper left side. 4G10 and rhodamine staining are depicted separately and merged (4G10, green; rhodamine phalloidin, red). Insets at the upper left of panels in A and C are examples of high power views of podosomes obtained by scanning in the z-axis to show the bright phosphotyrosine signal at podosome tips. The bottom side of these insets corresponds to the cell side which is in contact with the extracellular substrate. Panels at the far right are fivefold magnifications of image regions marked by rectangles in the adjacent panels. Merged channels (upper half) and the 4G10 channel alone (bottom half) are represented. Arrows in the right-hand panels indicate cases where strongly rhodamine-stained structures did not co-distribute with a strong phosphotyrosine staining. Arrowheads indicate co-distribution of strong phosphotyrosine and strong F-actin signals. Scale bars: 10 µm in A,B; 2 µm in insets.

View larger version (80K): [in a new window]   Fig. 2. Expression of integrin Vß3 induced by RANKL-ODF treatment. (A) Chicken peripheral blood-derived macrophages were cultured on glass coverslips for four days in the presence (right panels) or absence (left panels) of RANKL-ODF (see Materials and Methods). Cells were fixed, permeabilized and stained with the monoclonal antibody 23C6 (green) and rhodamine-phalloidin (red). Confocal images were acquired using identical channel settings for Vß3 detection and were equally processed in Adobe Photoshop to ensure comparable detection of Vß3 expression. Upper panels, 23C6-staining only; lower panels, 23C6 + rhodamine phalloidin staining. (B) Human osteoclast obtained after a nine day treatment of peripheral blood monocytes with RANKL-ODF. Staining was performed as in A and revealed by confocal microscopy. Upper panel, rhodamine phalloidin-staining (red); lower panel, 23C6- (green) + rhodamine phalloidin-staining. Note the nearly complete absence of integrin Vß3 on cells that were not treated with (A, left panels) or that did not visually respond to (B) the osteoclastogenic factor RANKL-ODF. Scale bar: 20 µm).

View larger version (110K): [in a new window]   Fig. 5. Distribution of paxillin, Pyk2, integrin Vß3 and cortactin in the podosome zones of chicken and human osteoclast-like cells. Confocal immunofluorescence images of cells double-stained with rhodamine phalloidin (red) and with antibodies for paxillin, Pyk2, Vß3 or cortactin (green in colored images or white in grayscale images) are shown. Right-hand panels are high-power views, which were also scanned in the z-axis at positions indicated by the double-arrows. These z-axis scans are shown on top of these panels with the extracellular substrate-facing cell side oriented towards the bottom of the images. Note the co-distribution of paxillin, Pyk2 and Vß3 around and between, but not within, the F-actin core structures of podosomes. By contrast, cortactin shows an inverse distribution, i.e. colocalization with F-actin in xy-scans, but appears also orientated towards the basal podosome side, as indicated by the separation of red (F-actin) and green (cortactin) channels in the inset scanned in the z-axis.

View larger version (27K): [in a new window]   Fig. 4. Tyrosine-phosphorylated proteins in osteoclast-like cells. (A) Tyrosine-phosphorylated proteins in lysates of chicken osteoclast precursors cultured for 4 days in the absence or presence of RANKL-ODF were identified on immunoblots with the antibody 4G10 (an identical protein profile was revealed using the anti-phosphotyrosine antibody, mAb PY99 (see Fig. 6C)). Major bands were detected at 60, 70, 85 and 110-130 kDa. (B) Chicken osteoclast-like cells were trypsinized and kept in suspension for 2 hours at 37°C. Subsequently, the cells were seeded onto plastic coated with serum-containing culture medium and lysed after the indicated times. Equal amounts of lysate protein were added to each lane and analyzed by immunoblotting with monoclonal antibody 4G10. (C) Lysates of adherent chicken osteoclast-like cells were immunoprecipitated twice with anti-paxillin antibodies. Lysate supernatants (S) obtained before the first or after the second round of immunoprecipitation as well as the two immunoprecipitates (IP) were analyzed by immunoblotting with antiphosphotyrosine antibody 4G10. (D) Lysates of chicken osteoclast-like cells were subjected to two rounds of immunoprecipitation with antiphosphotyrosine antibody 4G10 and analyzed by western blotting with 4G10 or anti-paxillin antibodies. Lysate supernatants obtained before the first (S1), and before (S2) and after (S3) the second round of immunoprecipitation, as well as the two immunoprecipitates (IP1, IP2) are shown. Note the mobility shift of tyrosine-phosphorylated paxillin in the immunoprecipitates and the disappearance of a slower mobility fraction of paxillin in the supernatants of the immunoprecipitations. However, this fraction represents only a minor subpopulation of total paxillin. (E) Chicken osteoclast-like cells were trypsinized, kept in suspension (Su) for 2 hours or subsequently adhered (Ad) to serum-coated plastic for 40 minutes. Lysates of these cells were immunoprecipitated with anti-Pyk2 antibodies and analyzed by western blotting using 4G10 and anti-Pyk2 antibodies. S, lysate supernatants; IP, immunoprecipitates; asterisks mark the position of immunoglobulin chains in the immunoprecipitates, which were detected by the secondary antibodies.

View larger version (45K): [in a new window]   Fig. 7. (A) Binding of synthetic peptides (see Fig. 6A) to purified GST-fusion proteins of paxillin (expected Mr approx. 95 kDa) and of the N-terminal domain (amino acids 1-407) of human Pyk2 (expected Mr approx. 70 kDa). Bound and unbound proteins were analyzed by immunoblotting with an anti-GST antibody as well as on Coomassie-stained 15% SDS-polyacrylamide gels. Note the depletion of GST-fusion proteins in the supernatants incubated with the ß3-tail peptide. Note also the nearly complete absence of bovine serum albumin in the bound fractions, which was present in a very large excess over GST-fusion proteins in the binding reactions. (Co, control resin consisting of streptavidin agarose without any added peptide; n.d., not done). (B) Comparison of GST, GST-paxillin and GST-Pyk2NT binding to the ß3-tail peptide. Note the depletion in the unbound fractions of GST-Pyk2NT and GST-paxillin after binding. GST proteins were calibrated to give comparable signals in the supernatants by detection with Anti-GST antibody.