spacer gif spacer gif spacer gif spacer gif spacer gif
 QUICK SEARCH:   [advanced]


spacer gif
     Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents    

First published online 27 May 2003
doi: 10.1242/jcs.00522


This Article
Right arrow Summary Freely available
Right arrow Full Text
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Related articles in JCS
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schmidt, M. H. H.
Right arrow Articles by Bögler, O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schmidt, M. H. H.
Right arrow Articles by Bögler, O.

SETA/CIN85/Ruk and its binding partner AIP1 associate with diverse cytoskeletal elements, including FAKs, and modulate cell adhesion

Mirko H. H. Schmidt, Baihua Chen, Lisa M. Randazzo and Oliver Bögler*

William and Karen Davidson Laboratory of Brain Tumor Biology, Hermelin Brain Tumor Center, Department of Neurosurgery, Henry Ford Hospital, 2799 West Grand Blvd, Detroit, MI 48202, USA



View larger version (95K):

[in a new window]
 
Fig. 1. SETA co-localizes with microfilaments and microtubules. Confocal microscopy of astrocytes stained with polyclonal anti-SETA antibody together with FITC-conjugated secondary antibodies (A,D,G,J), TRITC-phalloidin (B,E), or ß-tubulin followed by TRITC-conjugated secondary antibodies (H,K). The analysis revealed partial co-localization of SETA and actin microfilaments, and focal adhesions (arrows in A-C). However, areas without co-localization were also detected (arrowheads in D-F). Furthermore, partial overlap of SETA and ß-tubulin staining was detected in the peri-nuclear region and along individual bundles of microtubules (arrows in G-L). As shown for actin, the overlap was not complete (arrowheads in J-L).

 


View larger version (32K):

[in a new window]
 
Fig. 2. AIP1 but not SETA co-immunoprecipitates with cytoskeletal proteins. SETA or Flag-tagged AIP1 were transiently transfected into HEK293 cells and various cytoskeletal proteins were immunoprecipitated with specific antibodies, as indicated. The immunoprecipitates were subsequently analyzed by PAGE and immunoblotting with polyclonal anti-SETA antibody or monoclonal anti-Flag antibody. Although SETA was present in the lysates of the cells (lane 11), and was immunoprecipitated with the positive control protein EGFR (lane 10), it was not found in any of the immunoprecipitates made with antibodies to cytoskeletal proteins (lanes 1 to 9). In contrast, AIP1 revealed a strong signal after immunoprecipitation of most cytoskeletal proteins, with actin, {alpha}-tubulin, ß-tubulin and phosphotyrosine tubulin, as well as the microtubule-associated proteins (MAPs), showing particularly strong signals, and was also found in EGFR immunoprecipitates.

 


View larger version (39K):

[in a new window]
 
Fig. 3. SETA associates with FAK and PYK-2 as a dimer. To further analyze SETA's co-localization with focal adhesions, two major components of these structures, FAK (A) and PYK-2 (B), were immunoprecipitated from HEK293 cells that had been transiently transfected with SETA. Western blotting revealed that SETA could be recovered in both immunoprecipitates (A, lane 1; B, lanes 1,2), but was not recovered in control transfections (A, lanes 3,4; B, lanes 5,6). SETA appeared predominantly at about 160 kDa in the immunoprecipitates, while equal amounts of SETA at 85 kDa were present in lysates. Direct analysis (C) of SETA under strong (s; lanes 1,2) and weak (w; lane 3) denaturating PAGE conditions, as well as comparison with SETA{delta}cc (lane 4) lacking the coiled-coil domain, revealed that the 160 kDa band observed probably represents a SETA dimer mediated by homophilic interaction by this C-terminal domain, as shown previously (Borinstein et al., 2000Go; Watanabe et al., 2000Go). No SETA band at 160 kDa was found under strong denaturing conditions even when longer exposures (lane 1) were examined. This suggests that SETA preferentially associated with focal adhesion kinases as a dimer.

 


View larger version (30K):

[in a new window]
 
Fig. 4. The SETA binding partner AIP1 is associated with focal adhesion kinases. Analysis of the presence of the SETA-binding partners AIP1 and c-Cbl within focal adhesions revealed that AIP1 could be recovered by immunoprecipitation of FAK (A, lane 1) and PYK-2 (B, lanes 1,2) from transiently transfected HEK293 cells, whereas c-Cbl was exclusively and weakly associated with PYK-2 (B, lanes 3,4). Changing the cell density from low or non-confluent (l) to high or confluent (h) increased the amount of AIP1 associated with PYK-2, but did not influence the amount of c-Cbl. Neither protein had an influence on the concentration of PYK-2 under these conditions. Finally, the AIP1-binding partner ALG-2 was also found to co-precipitate with PYK-2 (C), indicating that an extended SETA-associated protein complex was intimately associated with focal adhesions. Please note that relative mobilities of proteins in Nu-PAGE gels is different from conventional Laemmli gels with stronger denaturation (see Materials and Methods).

 


View larger version (58K):

[in a new window]
 
Fig. 5. Increased calcium concentration causes increased association between AIP1 and PYK-2. Increased intracellular Ca2+ stimulates PYK-2 phosphorylation prompting an investigation of whether increasing the level of free calcium enhanced the AIP1-PYK-2 interaction. PYK-2 immunoprecipitations from transiently AIP1-transfected HEK293 cells were performed in various immunoprecipitation buffers. The IP buffer (control; containing 1 mM EGTA) was supplemented with 1 mM (lane 2) or 10 mM (lane 3) CaCl2; external Ca2+ was complexed by raising the EGTA concentration to 5 mM (lane 4), or protein-associated Ca2+ was complexed by adding 5 mM BAPTA (lane 5). No impact on the AIP1-PYK-2 or actin-PYK-2 interaction was found after moderate increases in the CaCl2 concentration, nor after addition of EGTA or BAPTA. However, raising the CaCl2 concentration to 10 mM resulted in a dramatic increase of AIP1 and actin in PYK-2 immunoprecipitates (lane 3). PYK-2 activity, as measured by phosphotyrosine western blot, is elevated in the presence of 10 mM CaCl2. The AIP1 band appears weaker in lanes 1, 2, 4 and 5 of this figure compared with that in Fig. 4 because of the short exposure time of the film to the blot, as a result of the strong signal in lane 3.

 


View larger version (27K):

[in a new window]
 
Fig. 6. The AIP1 deletion mutant AIP1-{delta}717-784 shows reduced binding to SETA. (A) The C-terminal proline-rich region containing multiple P-x-x-P motifs between residues 717 and 869 (black vertical lines) in full-length AIP1, and regions deleted in two AIP1 mutants AIP1-784Stop and AIP1-{delta}717-784. (B) In vitro binding assays were performed with bacterial GST-SETA fusion proteins encoding either isolated SH3 domains (SH3-1, -2 and -3) or full length SETA (123cc), and in vitro transcribed and translated, radiolabeled, full-length or mutant AIP1. Full-length AIP1 bound to the middle SH3-2 domain, and more strongly to full length SETA, as shown previously (Chen et al., 2000Go). AIP1-784Stop showed a similar binding profile to full-length, while AIP1-{delta}717-784 showed faint, non-specific binding, indicating that the N-terminal half of the P-x-x-P domain mediated SETA binding.

 


View larger version (42K):

[in a new window]
 
Fig. 7. SETA promotes the association between AIP1 and PYK-2. AIP1 and AIP1 mutants were transiently co-transfected into HEK293 cells with SETA or lacZ control plasmid. Endogenous FAK or PYK-2 was immunoprecipitated and the pellet subsequently analyzed by immunoblotting for AIP1 and SETA. When compared with wild-type AIP1, all mutants were significantly reduced in their ability to bind FAK and PYK-2, with the AIP1Y319F mutant, which is altered at a potential src phosphorylation site, showing no detectable binding to either focal adhesion kinase. The PxxP-deficient mutants were also negatively impacted in their binding capacity but remained detectable in the complex. Co-transfection of SETA enhanced wild-type AIP1 binding to PYK-2 but not to FAK. Furthermore, SETA was able to partially restore the binding of AIP1 and the two AIP1 deletion mutants to FAK/PYK-2.

 


View larger version (24K):

[in a new window]
 
Fig. 8. Reduction in cell adhesion by AIP1 proteins correlates with their presence in PYK-2 immunoprecipitates. HEK293 cells were transiently transfected with SETA, AIP1, AIP1-{delta}717-784, AIP1-784Stop, AIP1Y319F c-Cbl and lacZ as a control, as indicated. Two days after transfection cultures were harvested and underwent an ECIS cell attachment assay under confluent conditions on the electrode, over a period of 9 hours (see Materials and Methods for details). The values at each time point for the lacZ-transfected cultures from each experiment were averaged, taken as background and subtracted from the other data sets. Ten values around each hour were then taken, averaged and the standard deviation calculated. The values for each curve were set to zero at the zero time point and changes in resistance over time plotted. After an initial settling period of 3 hours the graphs reflect changes in resistance as cells attach and reach a steady state. (A) The resistance of SETA and AIP1 or c-Cbl-transfected cells moved in opposite directions over time. While SETA increased the resistance of the cell layer up to +1500 ohm, AIP1 and c-Cbl decreased it to about -1500 ohm. Accordingly, SETA mediates a pro-adhesive effect on HEK293 cells, while AIP1 and c-Cbl negatively affect cell adhesion. Cell density at the time of transfection had no major influence on these effects. (B) AIP1-{delta}717-784 and AIP1Y319F induced no major change in cell adhesion, while AIP1 and AIP1-784Stop induced a reduction similar to that observed in AIP1 (A). While SETA cotransfection did not alter the pattern of AIP1 and AIP1-784Stop cell adhesion, it caused an increase in adhesion when co-transfected with AIP1Y319F, which suggests that it's effect was dominant over this mutant, which had no effect on its own. Interestingly, cotransfection of SETA and AIP1-{delta}717-784 caused a reduction in adhesion. Error bars are smaller than plot symbols at some time points.

 


View larger version (63K):

[in a new window]
 
Fig. 9. AIP1 reduces PYK-2 and FAK phosphotyrosine levels. HEK293 cells were transfected with lacZ or AIP1 proteins as indicated, and the endogenous PYK-2 or FAK proteins were immunoprecipitated. Western blots of the immunoprecipitates demonstrate similar levels of focal adhesion kinase proteins, but different levels of phosphotyrosine when AIP1 or, to a lesser extent, AIP1-784Stop are co-transfected.

 





© The Company of Biologists Ltd 2003