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

First published online 28 September 2004
doi: 10.1242/jcs.01410

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 Related articles in JCS 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 De Corte, V. Articles by Gettemans, J. Search for Related Content PubMed PubMed Citation Articles by De Corte, V. Articles by Gettemans, J.

Increased importin-ß-dependent nuclear import of the actin modulating protein CapG promotes cell invasion

Veerle De Corte^1,*, Katrien Van Impe^1,*, Erik Bruyneel², Ciska Boucherie¹, Marc Mareel², Joël Vandekerckhove¹ and Jan Gettemans^1,

¹ Department of Medical Protein Research, Flanders Interuniversity Institute for Biotechnology (V.I.B.), Ghent University, Faculty of Medicine and Health Sciences, Albert Baertsoenkaai 3, 9000 Ghent, Belgium
² Laboratory of Experimental Cancerology, Department of Radiotherapy and Nuclear Medicine, Ghent University Hospital (1P7), De Pintelaan 185, 9000, Ghent, Belgium

View larger version (87K): [in a new window]   Fig. 1. CapG-EGFP expression triggers invasion of MDCK-AZ cells. (A-A') Epifluorescence microscopic images of MDCK-C cells and corresponding Texas-red phalloidin (F-actin) staining. (B-B') Distribution of endogenous CapG in MDCK-AZ cells and corresponding F-actin staining; (C) Control staining for CapG in the presence of excess antigen; (C') DAPI; (D) EGFP-expressing MDCK-AZ cells (MDCK-E); (D') F-actin staining. (E) CapG localization in HeLa cells. Bar in A and D, 40 µm; bar in B, C and E, 50 µm. (F) Western blot analysis on cytosolic proteins from MDCK-C clones. CapG-EGFP expression levels relative to endogenous CapG are indicated. (G) Collagen invasion index of seven independent MDCK-C clones (lanes 5-11). MDCK-AZ (lane 1) and MDCK-E cells (lanes 2-3) were used as negative controls; DHD-FIB = positive control (lane 4). cl1 = clone1; clC = clone C. Data are means ± s.e.m. of three independent experiments.

View larger version (59K): [in a new window]   Fig. 2. Invasion of MDCK-C cells in precultured chick heart fragments and inhibition of cell compaction. (A-D) Sections of confrontation cultures of MDCK-E clone A control cells (A,B) and MDCK-C clone W cells (C,D), fixed and stained with haematoxylin-eosin (A,C) or with antiserum against MDCK cells (B-D). Arrowheads in A and B highlight the epithelioid cell layer. Arrowheads in C and D mark invasive MDCK-C-EGFP clW cells. Bar, 100 µm. (E-H) Slow aggregation assay. Phase contrast images of MDCK-E clA control cells in the absence (E) or presence (F) of DECMA-1 antibody. (G,H) MDCK-C cl 3 (G) and MDCK-C cl W (H) in the absence of DECMA-1 antibody. Bar, 500 µm. (I) Fast aggregation assay. Plotted curves of relative volume distribution (y-axis) as a function of particle diameter (x-axis) are shown for MDCK-E clB without DECMA-1 antibody after 30 minutes (), MDCK-E clB with DECMA-1 antibody at 30 minutes () and MDCK-C cl5 without DECMA-1 antibody at 30 minutes ().

View larger version (75K): [in a new window]   Fig. 3. CapG nuclear import requires energy and importin-ß. (A) Western blot: recombinant Myc-CapG-V5-His6 was probed with anti-myc, anti-V5 or anti-CapG antibodies. (B-D) CapG-V5 import in the absence of RRL and energy (B), in the presence of energy but without RRL (C), or in the presence of GTP--S and RRL (D). Staining was performed with anti-V5 antibody. B'-D', DAPI stain. (E-E') Nuclear import of NLS-TRITC-BSA in the presence of RRL and GTP--S. (F,G) Import of CapG-V5 (F) and NLS-TRITC-BSA (G) in the presence of energy and RRL. (F'-G') DAPI stain. (H,I) Negative controls: anti-V5 antibody staining in the absence of CapG-V5 (H) and TRITC-BSA lacking the SV40 NLS (I). (H'-I') DAPI stain. (J) Importin-ß is required for CapG import. J1, CapG localization; J2, importin-ß localization; J3, merged image of J1 and J2; J4, DAPI stain. Note that CapG localizes to the nucleus only in cells that do not react with anti-importin-ß antibodies. (K) Anti-importin-ß antibody prevents NLS-TRITC-BSA nuclear import. (K') DAPI stain. The SV40 large T-antigen NLS peptide did not inhibit nuclear accumulation of CapG (L). (L') DAPI stain. Bar, 50 µm.

View larger version (36K): [in a new window]   Fig. 4. Collagen invasion is mediated by nuclear CapG. (A) Amino acid sequence alignment between the N-termini of fragminP, severin and CapG; aa 1-27 mark the severin/fragminP nuclear export sequence. Shaded residues are conserved. Residues marked in black are hydrophobic. (B) Nuclear CapG promotes collagen invasion (upper panel). Fusion of the FrgP NES to CapG-EGFP blocks collagen invasion (lane 3, `control'). Mutation of FrgP I9 or I14 to alanine shows no effect on collagen invasion (lanes 4-5). Mutation of L17, L21 or L27 to alanine restores nuclear localization and collagen invasion (lanes 6-8). Lower panel: western blot showing equal expression of the fusion proteins.

View larger version (15K): [in a new window]   Fig. 5. CapG represses transcriptional activity of VP16 in a dose-dependent manner. HEK 293T cells were transiently transfected with a Gal4-dependent reporter plasmid (p(Gal4)250hu.IL-6P-luc+) and pGal4-VP16, either alone or in combination with increasing amounts of pNLS-Gal4-CapG (pGal4-CapG), as indicated. Increasing amounts of NLS-Gal4-DBD-CapG caused a decrease in reporter activity. Luciferase activities were normalized using ß-galactosidase as an internal control. Data are means ± s.e.m. of three independent experiments.

View larger version (24K): [in a new window]   Fig. 6. Invasion and signal transduction. (A,B) Co-transfection of CapG-EGFP with RhoAN19 or Cdc42N17 (lanes 4-5), but not with Rac1N17 (lane 3), inhibits CapG-induced collagen type I invasion of HEK293T cells (A) (mean ± s.e.m., n=3, EV = empty vector) or HCT8/E-11 colon adenocarcinoma cells (B). (C) Co-expression of RasGAP (lane 5), p85iSH2N (`PI3KDN', lane 6) or dominant-negative MEK (lane 7) with CapG-EGFP blocks invasion. RasV12C40 shows no inhibition of invasion (lane 10) in contrast to RasV12S35 (lane 8) and RasV12E38 (lane 9). Values are representative of three independent experiments (mean ± s.e.m.). (D) Upper: western blot on transfected HEK293T c ells. Numbers correspond to the numbers in the histogram in (C). Lower: tubulin western blot = control. (E) Western blots showing expression of RasV12, RasGAP, p85iSH2N, RasV12E38, RasV12C40, RasV12S35 and MEKDN in cells co-transfected with CapG-EGFP. (F) Collagen invasion triggered by CapG-EGFP is counteracted by MMP inhibitors (mean ± s.e.m., n=3): aprotinin (10 µg/ml), galardin (10 µM) and the corresponding negative control (10 µM), TIMP-1 (500 ng/ml). CTTHWGFTLC = decapeptide (50 µM).

View larger version (18K): [in a new window]   Fig. 7. Overview of the CapG-induced invasion pathway. The figure summarizes co-transfection experiments in HEK293T cells of CapG-EGFP with dominant-negative or partial loss-of-function mutants of signalling proteins. Arrows indicate activation. The crossed arrows mean that this pathway is not implicated. The broken arrows point to signalling to the nucleus. The thick arrow indicates parallel or branching pathways. PAK1 83-149 (encoding the auto-inhibitory domain) and PAK1 K299R (a kinase dead mutant) did not block CapG-induced invasion. Therefore the Cdc42 (PAK1RhoA) hierarchical ordering as depicted does not contribute to this invasion pathway. Instead, RhoA is either linked to Cdc42 via another route, or acts independently of Cdc42.