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

First published online June 20, 2006
doi: 10.1242/10.1242/jcs.02996

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 Za, L. Articles by de Curtis, I. Search for Related Content PubMed PubMed Citation Articles by Za, L. Articles by de Curtis, I.

ßPIX controls cell motility and neurite extension by regulating the distribution of GIT1

Lorena Za¹, Chiara Albertinazzi^1,*, Simona Paris¹, Mariacristina Gagliani², Carlo Tacchetti² and Ivan de Curtis^1,

¹ Cell Adhesion Unit, Department of Molecular Biology and Functional Genomics, San Raffaele Scientific Institute, Via Olgettina 58, 20132 Milano, Italy
² MicroScoBio Research Center and IFOM Center for Cell Oncology and Ultrastructure, Department of Experimental Medicine, University of Genoa, Via deToni 14, 16132, Genoa, Italy

View larger version (34K): [in a new window]   Fig. 1. ßPIX-induced recruitment of GIT1 at cytoplasmic structures. CEFs (A-D) and COS7 cells (E-H) were transfected to overexpress GIT1 (A,E), ßPIX (B,F), or both proteins (C,D,G,H), and analyzed by immunofluorescence. Same fields are shown in C,D, and in G,H. Co-expression of GIT1 and ßPIX enhanced the localization of both proteins at large structures (arrowheads). Bars, 20 µm.

View larger version (79K): [in a new window]   Fig. 2. Effects of GIT1 and ßPIX overexpression. (A) Overexpression of GIT1 and ßPIX does not affect the organization of intermediate filaments. Confocal images were taken of COS7 cells cotransfected for GIT1 and ßPIX expression and immunostained for GIT1 (red) and vimentin (green). (B,C) Overexpression of GIT1 and ßPIX partially affects the recycling of transferrin (Tf). A431 cells cotransfected with GIT1 and ßPIX were incubated for 1 hour at 37°C with 60 µg/ml of Alexa Fluor 488-labelled human Tf (green, B), and chased for 2 hours at 37°C (C). Two different confocal planes along the z-axis are shown in B. (D) Overexpression of ßPIX prevented Tf internalization in A431 cells. Cells were fixed, permeabilized and immunostained with antibodies (red) for ßPIX (C,D) or GIT1 (B). Bars, 20 µm.

View larger version (131K): [in a new window]   Fig. 3. GIT1 colocalizes with transferrin receptor (TfR) in endosome-derived structures. COS7 cells transfected for GIT1 and ßPIX. (A-C) Double immunofluorescence on ultrathin cryosections. GIT1 (A) and endogenous TfR (B) colocalize (arrows) within large intracellular structures present in the cytoplasm (C) (GIT1, green; TfR, red; nuclei, blue). (D-G) Immunogold labelling of ultrathin cryosection for GIT1, identified by antibodies to Flag, (FLAG, gold 15 nm), and TfR (gold 10 nm). GIT1 associates with tubulovesicular endosomes (D) and intracellular electron dense structures (E), often displaying bounding membranes (F,G arrows). GIT1 and endogenous TfR associate to the same electron dense structures (E-G), often within small vesicles (G). Bars, (D) 253 nm; (E) 212 nm; (F) 137 nm; (G) 57 nm.

View larger version (104K): [in a new window]   Fig. 4. COS7 cells transfected for ßPIX-HA (A,B) or GIT1-Flag (C,D). (A,B) Double immunogold labelling on ultrathin cryosections. ßPIX-HA (gold, 15 nm), identified by antibodies to HA, labels extensively large intracellular structures. ßPIX-HA (gold, 15 nm) and endogenous TfR (gold, 10 nm) scarcely colocalize within these structures (arrows in B). (C) Immunogold labelling of ultrathin cryosection for GIT1 (gold, 15 nm), identified by antibodies to Flag. GIT1 labels extensively large intracellular structures and endosomes (arrow). (D) Transferrin receptor (TfR, gold 10 nm) associates with similar electron dense structures. Bars, (A) 470 nm; (B) 180 nm; (C) 140 nm; (D) 57 nm.

View larger version (48K): [in a new window]   Fig. 5. EGF stimulation induces vesicle accumulation in cells expressing an ArfGAP mutant of GIT1. (A) Schematic representation of the GFP-tagged GIT1 constructs utilized in this study: p95-GFP including the full length GIT1; p95-C2-GFP including residues 229-740 of GIT1; p95-C-GFP, including residues 346-740 of GIT1, and GFP alone. ANK, ankyrin repeats; SHD, Spa2 homology domain; LZ, leucine zipper; PBS, paxillin binding subdomain. (B,C) A431 cells were stimulated for the indicated times with 100-200 ng/ml of EGF: representative frames from time-lapse observation of cells expressing p95-C2-GFP (B) and p95-C-GFP (C); Bar, 10 µm. Arrowheads in B indicate newly formed GFP-positive structures following EGF stimulation. Arrows in C indicates growth factor-induced lamellipodial extensions. Bar, 10 µm. (D) Frames from time-lapse observation of cells expressing p95-C2-GFP. Arrowheads and arrows indicate two examples of fusion between GFP-positive structures during stimulation with EGF. Bar, 5 µm. (E) Cells expressing p95-C2-GFP (n=24 with EGF; n=21 without EGF), p95-C-GFP (n=18), p95-GFP (n=24), or GFP alone (n=24) were analyzed by time-lapse digital analysis, and scored after 40 minutes for the formation of GFP-labelled vesicle during stimulation with EGF. Bars represent the percentages of cells with EGF-induced vesicles.

View larger version (74K): [in a new window]   Fig. 6. EGF stimulation induces retraction in cells expressing an ArfGAP mutant of GIT1. A431 cells were stimulated for the indicated times with 100-200 ng/ml of EGF. (A) EGF-induced lamellipodia protrusion is inhibited in a cell expressing the ArfGAP defective p95-C2-GFP protein (asterisk), whereas large lamellipodia are formed (arrowheads) in two neighbouring non-transfected cells (white dots). Same fields are shown in the frames of the upper and lower rows, respectively. Bar, 10 µm. (B) Cells with contact-free edges expressing p95-C2-GFP (n=23), p95-GFP (n=20), p95-C-GFP (n=18), or GFP (n=18) were analyzed by time-lapse digital analysis, and scored during the first 15 minutes of stimulation with EGF for formation of ruffles. Bars represent the percentages of cells showing inhibition of EGF-induced ruffling. (C) Retraction of two cells expressing p95-C2-GFP (asterisks) during EGF stimulation. No strong retraction was evident in the non-transfected cell. Bar, 20 µm. (D) Retraction was scored 30 minutes after stimulation with EGF in cells with contact-free edges expressing p95-C2-GFP (n=24 with EGF; n=20 no EGF), p95-GFP (n=23), p95-C-GFP (n=18), or GFP (n=22). Bars represent the percentages of retracted cells. (E) A431 cells were treated with 10 µM Y-27632 for 45 minutes before stimulation with 200 ng/ml EGF for the indicated times. The asterisk indicates a transfected cell; arrows indicate retractions. No strong retraction (arrowheads) was evident in the non-transfected cells. Bar, 20 µm. (F) Cells with contact-free edges expressing p95-C2-GFP treated with Y-27632 (30 cells), with EGF (23 cells), or with Y-27632 followed by EGF (25 cells) were analyzed by time-lapse digital analysis. Retraction was scored 30 minutes after EGF (or mock) stimulation. Bars represent the percentage of retracted cells. In B,D,F the standard error of the percentages is shown. Statistical significance was assessed by the 2 test, incorporating Yates' correction for continuity, with P<0.05 considered significant. In B and D the values obtained from cells transfected with each of the GIT1 constructs were compared with that from cells transfected with GFP alone. *P<0.001; **P<0.0005. In F there was no significant difference in terms of retraction after EGF stimulation between Y-27632 treated and untreated cells.

View larger version (47K): [in a new window]   Fig. 7. Dimerization and the SH3 domain of ßPIX are required for the recruitment of GIT1 at large cytoplasmic structures. (A) Schematic diagram of the PIX constructs used in this study; SH3, Src homology 3 domain; DH, Dbl homology region; PH, pleckstrin homology domain; ABD, ArfGAP binding domain; LZ, leucine zipper. (B-D) Effects of ßPIX expression on GIT1 localization. CEFs were transfected with either GIT1 or PIX (B), cotransfected with both proteins (C), or cotransfected with full length GIT1 and the monomeric SH3 mutant ßPIX-PGLZ (D). Bar, 20 µm. (E) Quantification of the localization of GIT1 at large cytoplasmic structures in CEFs transfected with GIT1 alone (n=300); cotransfected with GIT1 together with each of the following ßPIX constructs: full length PIX (n=385), PIX-PGLZ (n=200), PIXLZ (228), PIX-PG (200), or cotransfected with the monomeric mutant of GIT1 (p95-LZ) and full length PIX (n=200). Error bars represent the s.d. from at least two independent experiments. Statistical significance was assessed by the Student's t-test (P<0.05 considered significant). The values from cells transfected with GIT1 alone or cotransfected with GIT1 and each of the indicated ßPIX mutants were compared to the values from cells cotransfected with GIT1 and wild-type ßPIX. *P<0.005; **P<0.0005. (F) Overexpression of either PIXLZ or PIX-PGLZ alone did not induce the formation of large structures. Bar, 20 µm. (G) Immunoprecipitation with anti-FLAG from lysates of COS7 cells transfected with FLAG-GIT1 and one of the following monomeric mutants of ßPIX: ßPIX-PGLZ (IP1), ßPIX-CLZ (IP2), ßPIX-LZ (IP3), ßPIX-PHLZ (IP4) and ßPIXwt (IP5). 50 µg of each lysate (Lys) were also loaded. Portions of the same filters were immunoblotted for PIX and GIT1, as indicated.

View larger version (58K): [in a new window]   Fig. 8. PAK-Pbd interferes with the recruitment of the PIX/GIT1 complex at large cytoplasmic structures. (A) Scheme of PAK1 and of the PAK-Pbd construct. 1-4 indicate the four proline-rich regions in the amino-terminal portion of PAK1; region 4 (asterisk) is the one involved in binding to the SH3 domain of PIX. (B) Immunoprecipitation from lysates of COS7 cells transfected with ßPIX (IP1) or ßPIX and PAK-Pbd (IP2). Unbound fractions (Ub) and lysates (Lys) from the two samples were also loaded. Portions of the same filters were immunoblotted for PIX (PIXtr), endogenous PAK (PAKe), or PAK-Pbd, as indicated. (C) Immunoprecipitation with anti-Myc antibodies from lysates of COS7 cells transfected with FLAG-GIT1, HA-ßPIX and Myc-PAK (Lys1) or Myc-PAK-Pbd (Lys2). 50 µg of each lysate were also loaded. Portions of the same filters were immunoblotted for GIT1, PIX or PAK, as indicated. (D-G) Effects of PAK expression on the distribution of PIX and GIT1. CEFs were triple transfected with PAK, ßPIX and GIT1 (D,E), or with PAK-Pbd, PIX and GIT1 (F,G). Bar, 20 µm. (H) Quantification of the effects of PAK-Pbd expression on the localization of GIT1 and PIX at large cytoplasmic structures in CEFs cotransfected with GIT1 and ßPIX (n=385), triple transfected with PAK, PIX and GIT1 (n=200), or with PAK-Pbd, PIX and GIT1 (n=200). Error bars represent the s.d. from at least two independent experiments. Statistical significance was assessed by the Student's t-test (P<0.05 considered significant). The values from cells triple transfected with GIT1, ßPIX and PAK-Pbd were compared to those from cells triple transfected with GIT1, ßPIX and wild-type PAK. *P<0.005. (I) A431 cells were cotransfected with PIX and PAK. Endogenous GIT proteins (GITe, green) colocalized with transfected PIX (red) at cytoplasmic structures. Bar, 25 µm.

View larger version (35K): [in a new window]   Fig. 9. Expression of ßPIX constructs in retinal neurons. (A) Immunofluorescence for overexpressed ßPIX on E6 retinal neurons transfected with different ßPIX mutants. Bar, 5 µm. (B) Quantification of the effects of the expression of different ßPIX constructs in retinal neurons. Each bar represents the average percentage obtained from two experiments for each condition; 50 neurons were analyzed in each experiments (total of 100 neurons per condition).

View larger version (50K): [in a new window]   Fig. 10. Co-expression of ßPIX and GIT1-derived constructs in retinal neurons. (A-I) Retinal neurons were fixed 1 day after transfection or cotransfection, and the indicated transfected constructs were detected by immunofluorescence. Bars, 5 µm. (J) Transfected neurons were utilized to quantify the effects on neurite extension. Each bar represents the average percentage obtained from two experiments for each condition; 50 neurons were analyzed in each experiments (total of 100 neurons per condition). (K) Evaluation of the percentages of neurons with long neurites and of neurons with large cytoplasmic structures in neurons transfected with the indicated constructs: 100 neurons per condition were evaluated.