Cancer Research Meeting Calendar Frontiers in Basic Cancer Research
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Cancer Research 67, 5275, June 1, 2007. doi: 10.1158/0008-5472.CAN-07-0318
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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 Google Scholar
Google Scholar
Right arrow Articles by Ramsay, A. G.
Right arrow Articles by Marshall, J. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ramsay, A. G.
Right arrow Articles by Marshall, J. F.

Cell, Tumor, and Stem Cell Biology

HS1-Associated Protein X-1 Regulates Carcinoma Cell Migration and Invasion via Clathrin-Mediated Endocytosis of Integrin {alpha}vß6

Alan G. Ramsay1,2, Melanie D. Keppler3, Mona Jazayeri1,2, Gareth J. Thomas1,2, Maddy Parsons3, Shelia Violette4, Paul Weinreb4, Ian R. Hart1,2 and John F. Marshall1,2

1 Centre for Tumour Biology, Institute of Cancer and Cancer Research UK Clinical Centre; 2 John Vane Science Centre, Barts and The London, Queen Mary's School of Medicine and Dentistry; 3 Randall Division of Cell and Molecular Biophysics, King's College London, London, United Kingdom; and 4 Biogen Idec, Cambridge, Massachusetts

Requests for reprints: Ian R. Hart, Centre for Tumour Biology, Institute of Cancer and Cancer Research UK Clinical Centre, John Vane Science Centre, Ground Floor, London EC1M 6BQ, United Kingdom. Phone: 44-207-014-0402; Fax: 44-207-014-0401; E-mail: ian.hart{at}cancer.org.uk or John F. Marshall, Phone: 44-207-014-0407; Fax: 44-207-014-0401; E-mail: john.marshall{at}cancer.org.uk.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Enhanced expression levels of integrin {alpha}vß6 have been linked to more aggressive invasive carcinoma cell behavior and poorer clinical prognosis. However, how {alpha}vß6 determines invasion and the dynamics of integrin {alpha}vß6 regulation in tumor cells are poorly understood. We have identified the 35-kDa HS1-associated protein X-1 (HAX-1) protein as a novel binding partner of the ß6 cytoplasmic tail using a yeast two-hybrid screen. We show that {alpha}vß6-dependent migration is blocked following small interfering RNA (siRNA)–mediated depletion of HAX-1 in oral squamous cell carcinoma cell lines. Using both siRNA and membrane-permeable peptides, we show that {alpha}vß6-dependent migration and invasion require HAX-1 to bind directly to ß6 and thereby regulate clathrin-mediated endocytosis of {alpha}vß6 integrins. Progression of oral cancer is associated with enhanced expression of {alpha}vß6 and HAX-1 proteins in patient tissue. This report establishes that integrin endocytosis is required for {alpha}vß6-dependent carcinoma cell motility and invasion and suggests that this process is an important mechanism in cancer progression. [Cancer Res 2007;67(11):5275–84]


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Integrins are heterodimeric cell-surface receptors, comprising noncovalently associated {alpha} and ß subunits (1). They mediate cell-cell and cell-extracellular matrix interactions involved in cell migration, proliferation, differentiation, and apoptosis (1–4), processes essential in tumor cell invasion (5). The ß6 subunit first was identified in cultured epithelial cells as part of the {alpha}vß6 heterodimer, the only heterodimer it is known to form (6). Whereas {alpha}vß6 expression is low or absent on resting epithelial cells, de novo or increased expression occurs during wound healing responses (7) and in cancers such as oral (8), ovarian (9), and colorectal (10) carcinomas. The {alpha}vß6 heterodimer, which mediates keratinocyte migration on fibronectin, vitronectin, and transforming growth factor ß (TGF-ß) latency-associated peptide (LAP), is expressed preferentially at the leading edge of migrating cells and at the tumor cell-stroma interface (11). Because {alpha}vß6 promotes invasion in vitro, this localization suggests an active role in tumor cell invasion (12–15) and increased expression has been correlated with a dramatic reduction in survival of patients with colorectal cancer (10).

The ß-subunit cytoplasmic domains link integrins to the cytoskeleton (1), determine their correct subcellular localization, and regulate ligand affinity (16). Three regions of highly conserved sequence within ß-subunit cytoplasmic domains, Cyto 1, Cyto 2, and Cyto 3 (17), are essential for integrin binding to many cell signaling molecules and other cellular proteins. The Cyto 2 and Cyto 3 domains comprise NXXY motifs similar to those that affect clathrin-mediated endocytosis of several membrane receptors (18).

Numerous proteins bind to one or more ß-integrin tails (19) but, to date, only one molecule, ERK2, has been reported to bind to the ß6 cytoplasmic domain (20). We speculated that additional binding partners of the ß6 cytoplasmic domain might well have a role in this integrin's proinvasive activity.

We screened a human keratinocyte cDNA library and identified HS1-associated protein X-1 (HAX-1) as an interaction partner for ß6, an interaction confirmed by coimmunoprecipitation. We defined the regions of HAX-1 and ß6 required for this interaction. We also showed that HAX-1 expression is up-regulated in more advanced oral carcinoma.

We report that reduction of HAX-1 levels by small interfering RNA (siRNA) suppresses {alpha}vß6-dependent carcinoma cell migration by down-regulating {alpha}vß6 endocytosis via a clathrin-mediated pathway. Moreover, this clathrin-mediated activity is a prerequisite for regulation of invasive behavior by {alpha}vß6-expressing carcinoma cells. Finally, using membrane permeable peptides to block binding of HAX-1 to the ß6 subunit, we abrogated {alpha}vß6 endocytosis, which caused complete suppression of {alpha}vß6-dependent invasion.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Antibodies and reagents. The monoclonal antibodies (mAb) used in this study were as follows: P1F6, antihuman integrin {alpha}vß5, and 10D5, antihuman {alpha}vß6 integrin, from Chemicon International; antihuman HAX-1 and anti–clathrin heavy chain (CHC), from BD Biosciences; antihuman transferrin receptor from Zymed Laboratories (Invitrogen Ltd.); TS2/16, antihuman integrin ß1, from American Type Culture Collection; and B-6, antihuman HSC70, from Santa Cruz Biotechnology. 7.1C5, 6.2G2, and 6.3G9, antihuman {alpha}vß6 integrin, were generated by Biogen Idec. 53a rat mAb against human {alpha}vß6 was produced in-house. Polyclonal antibodies C-19, antihuman integrin ß6, and FL-279, antihuman HAX-1, were from Santa Cruz Biotechnology. Human LAP, plasma fibronectin, type 1 collagen, and laminin were all from Sigma. Human vitronectin was from Chemicon International.

Cell culture. Several oral squamous cell carcinoma (SCC) cell lines were used. VB6 cells were engineered by retroviral transduction for high {alpha}vß6 expression (12, 13). H400, which express high endogenous {alpha}vß6 levels, and H357, which are {alpha}v negative, were from Prof. Stephen Prime (University of Bristol). Lines were grown in keratinocyte growth medium (KGM) as described (12, 13). Organotypic culture experiments required human foreskin fibroblasts (supplied by Cell Services, Cancer Research UK, London Research Institute) maintained in fibroblast growth medium (DMEM supplemented with 10% FCS).

Yeast two-hybrid screens. Yeast two-hybrid assays used the MATCHMAKER two-hybrid kit (Clontech) according to the manufacturer's instructions. The yeast strain AH109 was transformed with pGBKT7-ß6. A human keratinocyte cDNA library (5 x 106 independent clones), prepared in the activation domain vector pGAD10 (Clontech), was transformed into AH109 containing pGBKT7-ß6. Approximately 1.5 x 107 transformants were screened. HaeIII digestion patterns of PCR amplified library inserts eliminated duplicate positive clones.

Biochemical methods. For immunoprecipitations of VB6 or H400 cells, 5 x 106 cells were lysed for 20 min at 4°C with 1 mL of a buffer containing 20 mmol/L HEPES (pH 7.8), 1% NP40 (or 1% octyl ß-D-glucopyranoside), 50 mmol/L NaCl, 1 mmol/L CaCl2, 3 mmol/L MgCl2, and 0.3 mol/L sucrose, with protease and phosphatase inhibitors. After centrifugation at 13,000 rpm for 20 min, supernatants were incubated with 50-µL protein A/G PLUS-agarose (Santa Cruz Biotechnology) for 1 h at 4°C with constant rotation. Precleared supernatants were aliquoted and incubated with 5 µg of indicated primary antibody bound to 30 µL of protein A/G PLUS agarose for 24 h at 4°C with constant rotation. Beads were washed thrice with lysis buffer and immune complexes were eluted by boiling for 10 min in reducing SDS sample buffer (Invitrogen). Alternatively, magnetic beads (Dynal, Invitrogen) conjugated to primary antibodies were incubated with lysates overnight at 4°C with constant rotation. Beads were washed six times with lysis buffer before immune complexes were eluted. Extracts were electrophoresed on SDS-polyacrylamide gels for Western blotting with indicated antibodies.

Plasmids, siRNAs, and transfections. Dominant-negative Eps15E{Delta}95/295-GFP and Eps15DIII-GFP (21) were kind gifts from Dr. Alexandre Benmerah (Cochin Institute, Paris, France).

Custom SMARTpool siRNA reagents targeting HAX-1, CHC, and control (nontargeting siRNA pool) were from Dharmacon RNA Technologies.

Cells were transfected transiently with plasmid DNA using Lipofectamine 2000 (Invitrogen) or FuGENE 6 (Roche Diagnostics). Transfection efficiencies, estimated with control GFP-expressing plasmids, were 60% to 80%. Cells, transfected with 100 µmol/L of siRNA pool using Oligofectamine reagent (Invitrogen), did not increase their IFN-{alpha} response compared with untransfected cells as quantified using an ELISA kit (PBL Biomedical Laboratories). After 48 h, transfected cells were harvested with trypsin and used in the indicated assays.

Western blot analysis. Cell lysates were prepared with radioimmunoprecipitation assay buffer plus protease inhibitors, and insoluble material was cleared by centrifugation for 10 min at 4°C. Supernatants were corrected to equivalent protein concentrations using the DC Protein Assay kit (Bio-Rad Laboratories) and loaded on 4 % to 12% Novex Bis-Tris or 3% to 8% Novex Tris-acetate gels (Invitrogen) under reducing or nonreducing conditions and transferred to nitrocellulose membranes (Hybond-ECL, Amersham Biosciences, GE Healthcare). Membranes were blocked with nonfat milk, then probed with the indicated primary antibody, and developed with horseradish peroxidase (HRP)–conjugated secondary antibody (DakoCytomation) and enhanced chemiluminescence reagents (Amersham Biosciences). HSC70 was used as loading control. Exposures of blots in the linear range were quantified by densitometry software (Scion Corp.).

Internalization and recycling assay. The endocytosis assay was done as described (22) with minor modifications. Transfected cells grown in 10-cm culture dishes were surface labeled at 4°C with 0.2 mg/mL EZ-Link Sulf-NHS-SS-Biotin (Pierce), then transferred to KGM at 37°C for internalization. Primaquine (0.6 µmol/L; Sigma) was added as required. At indicated times, cell-surface biotinylated protein was removed by treatment with a non–cell-permeant reducing agent (MesNa; Sigma) at 4°C. To determine total surface biotinylated integrin, samples were not treated with reducing agent. Cells then were lysed on ice and protein concentration was determined and equalized. Levels of internalized biotinylated integrin were determined by immunoprecipitation (using Dynal magnetic beads conjugated to anti-integrin antibodies) and Western blot analysis (nonreducing SDS-PAGE) with streptavidin conjugated to HRP (Biosource), or by capture-ELISA.

Integrin recycling assays were done as described (22). Transfected cells grown in 10-cm culture dishes were labeled with biotin and transferred to KGM for 10 or 30 min at 37°C (early and late internalization, respectively) before being returned to ice and treated with MesNa. The internalized integrin pool was then chased from the cells by returning them to 37°C in KGM for the indicated times. Cells were then returned to ice, biotin was removed by a second treatment with MesNa, lysed on ice, equalized for protein concentration, and levels of biotinylated integrin were determined by capture-ELISA. The proportion of {alpha}vß6 recycled back to the plasma membrane is expressed as the percentage of the integrin pool labeled during the internalization period.

Capture-ELISA assay. Detection of biotinylated receptors by capture ELISA was done as described (22). Microtiter wells were coated with 5 µg/mL of indicated mAb and proteins were captured by overnight incubation of 50-µL cell lysate at 4°C. Bound material was incubated with streptavidin conjugated to HRP and biotinylated protein was detected by incubation with a substrate-chromogen (TMB+ Substrate-Chromogen, DakoCytomation).

Migration and invasion assays. Haptotactic migration assays were done as described (12) using extracellular matrix protein–coated polycarbonate filters (8 µm pore size, Transwell, Costar). VB6 cells migrated for 16 h and H400 cells for 6 h. Cells were trypsin detached separately from both the top and bottom chambers and counted on a CASY-1 cell counter (Scharfe System GmbH). Percentage migration was determined by the number of cells in the lower chamber relative to the total number of cells in the well. Data are expressed as mean percentage of cells migrating ±SD. Note that we used LAP in migration assays to allow unequivocal identification of the involvement of {alpha}vß6 (14).

Invasion assays were done over 72 h using polycarbonate filters coated with Matrigel (BD Biosciences) as described (13). Cells that had invaded through to the lower chamber were trypsinized and counted on a CASY-1 counter.

Transwell assay experiments were repeated thrice, with quadruplicate wells for each treatment. For blocking experiments, anti-{alpha}vß6 antibody 6.3G9 (10 µg/mL), added to the cells for 30 min before plating, was present during the entire assay.

Tat-linked peptide assays. Peptides were synthesized in-house (Cancer Research UK, London Research Institute). Cells were treated with 10 µmol/L of peptide in serum-free {alpha}-MEM for 1 h at 37°C before use in the biochemical or biological assays.

Organotypic culture and quantitative analysis. Organotypic cultures were prepared as described (23). Gels were composed of a 50:50 mixture of Matrigel and type 1 collagen (Upstate, Millipore) containing 5 x 105 human foreskin fibroblasts, to which were added 5 x 105 carcinoma cells. KGM was changed every 2 days and, after 10 days, the gels were fixed in formal saline, bisected, and embedded to paraffin.

Invasion in organotypics was quantified as described (23). The "invasion index" was calculated using the product of the average depth of invasion, the number of invading particles, and the area of the invading particles.

Immunohistochemistry. Organotypic sections were stained with H&E or immunostained with the following antibodies: anti-cytokeratin AE1/AE3 (dilution, 1:50; DakoCytomation), anti-{alpha}vß6 6.2G2 (0.5 µg/mL), or anti–HAX-1 FL-279 (10 µg/mL), followed by biotin-streptavidin labeling (Vectastain Elite ABC reagent, Vector Laboratories). {alpha}vß6 staining required 5 min of antigen retrieval with Digest-ALL 3 Pepsin Solution (Zymed). Detection was done with DAB+ (DakoCytomation) and counterstained with Mayer's hematoxylin (Sigma).

Ten oral SCC, 10 epithelial dysplasias, and 12 polyps showing fibroepithelial hyperplasia were stained for {alpha}vß6 and HAX-1. Staining intensity was scored from 1 to 3 (1, weak; 2, moderate; 3, strong), and the proportion of epithelial cells staining positively was scored from 1 to 4 (1, <25%; 2, 25–50%; 3, 51–75%; 4, 76–100%). The score for intensity was added to the score for proportion to give a score in the range of 0 to 7 and grouped as – (score = 0), + (score = 1–3), ++ (score = 4–5), or +++ (score = 6–7).

Statistical analysis. Statistical differences between experimental groups were evaluated by Student's t test (two tailed).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Identification of HAX-1 as a binding partner for ß6 in a yeast two-hybrid screen. To identify novel ß6 cytoplasmic tail (pGBKT7-ß6) binding proteins, we did a yeast two-hybrid analysis and isolated HAX-1 from a human keratinocyte cDNA library. HAX-1 was identified originally as an intracellular, 35-kDa hemopoietic specific protein-1 (HS1)–interacting protein (24). Analysis of the transformants positive for pGBKT7-ß6 interaction identified three different fragments of HAX-1, denoted clone types A, B and C, which did not interact with empty pGBKT7 vector in the yeast two-hybrid analysis (Fig. 1A ) or with the nonspecific controls, pGBKT7-laminC or pGBKT7-p53 (data not shown).


Figure 1
View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Characterization of HAX-1-integrin {alpha}vß6 association. A, schematic diagrams of the yeast two-hybrid interaction of three HAX-1 clones with integrin ß6 cytoplasmic tail and the respective binding regions for the ß6 and HAX-1 interaction. Yeast growth on -4DO media following cotransfection of pGBKT7-ß6 or pGBKT7 and each of the three HAX-1 clones. + or – denote positive or negative interactions, respectively, with HAX-1 or ß6 tail. B, endogenous HAX-1 and ß6 proteins interact physiologically. VB6 and H400, left, extracts were immunoprecipitated with antibodies against {alpha}vß6 (mAb; 7.1C5) or isotype-matched immunoglobulin G (IgG) and Western blots probed with anti–HAX-1 mAb (top) or anti-ß6 polyclonal antibody (bottom). VB6 and H400, right, reverse immunoprecipitation experiments with antibodies against HAX-1 (polyclonal antibody; FL-279) or isotype-matched IgG and Western blots with anti-ß6 polyclonal antibody (top) or anti–HAX-1 mAb (bottom) for the Western blot analysis.

 
Identification of the ß6-HAX-1 interaction sites by a yeast two-hybrid screen. Truncation mutants of ß6 defined the region required for the interaction with HAX-1. Removal of the terminal 11 amino acids did not disrupt the ability of ß6 to bind HAX-1 (Fig. 1A). Removal of the conserved Cyto 3 and Cyto 2 domains also failed to abolish binding to HAX-1. Therefore, ß6 binds HAX-1 between residues 731 and 758, which incorporates the juxtamembrane region and cytodomain 1.

The ß6 binding site within HAX-1 was located initially within the COOH-terminal 118 amino acids as shown by the interaction of clone C with ß6. Truncations of HAX-1 from the COOH terminus were made in 10-amino-acid steps. Interaction occurred between ß6 and both full-length HAX-1 (1–279 amino acids) and clone C (162–279 amino acids); deletion of the terminal nine amino acids of HAX-1 abolished this binding (Fig. 1A).

The interaction between HAX-1 and {alpha}vß6 in carcinoma cells. Coimmunoprecipitation assays used two oral SCC cell lines: (a) VB6, retrovirally engineered to express high {alpha}vß6 levels (12, 13), and (b) H400, expressing high levels of {alpha}vß6 endogenously. In both lines when {alpha}vß6 was immunoprecipitated with an anti-{alpha}vß6 antibody, HAX-1 was also coprecipitated (Fig. 1B). Reciprocally, when anti–HAX-1 antibody was used to immunoprecipitate HAX-1, blotting with anti-ß6 antibody showed coprecipitation of ß6 (Fig. 1B). This association of ß6 with HAX-1 was also observed in the lung adenocarcinoma cell line H441 (see Supplementary Fig. S1A) where requirement for the terminal nine amino acids of HAX-1 for association with ß6 was confirmed biochemically (Supplementary Fig. S1B). Myc-tagged HAX-1 was transfected transiently into this cell line and also into 3T3 mouse fibroblasts transduced with human ß6 (3T3 ß6). In both lines, ß6 was coprecipitated with anti-myc antibody (Supplementary Fig. S1C).

Requirement for HAX-1 in {alpha}vß6-mediated carcinoma cell migration. Levels of HAX-1 expression were reduced by 70% to 90% in both VB6 and H400 following 48-h treatment with HAX-1 siRNA (Fig. 2A ). Growth studies, up to 12 days posttransfection with HAX-1 siRNA, revealed no differences in cell number, relative to control siRNA treated cells, whether cells were cultured under two-dimensional or three-dimensional conditions (data not shown). We examined VB6 and H400 cell migration toward different matrix ligands, particularly toward TGF-ß LAP where migration of VB6 cells is modulated solely through {alpha}vß6 (14). Inhibition of HAX-1 expression significantly decreased migration toward LAP (Fig. 2B, 68% reduction, P < 0.001) and H400 cells (Fig. 2C, 65% reduction, P < 0.001), comparable to the effect of anti-{alpha}vß6 antibody (6.3G9; 81% and 76% reduction in migration of VB6 and H400 cells toward LAP, respectively). Migration toward fibronectin, laminin, type 1 collagen, and vitronectin, mediated by non-{alpha}vß6 integrins in our carcinoma cell lines (data not shown), was unaffected by HAX-1 depletion (Fig. 2B and C). Although {alpha}vß6 is a fibronectin receptor, we previously showed that migration of VB6 cells toward fibronectin is modulated through {alpha}5ß1 and {alpha}vß6, requiring inhibition of both integrins to abrogate cell movement toward this substrate completely (12). Similar HAX-1 reduction in H357 cells, an {alpha}vß6-negative SCC cell line, failed to affect migration toward LAP or other ligands (Fig. 2D).


Figure 2
View larger version (27K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. HAX-1 knockdown specifically inhibits {alpha}vß6-dependent SCC migration toward the {alpha}vß6-specific substrate LAP. A, SCC cell lines transfected (48 h) with control (CON) or HAX-1 siRNA were analyzed by Western blotting with the indicated antibodies (anti–HAX-1 mAb, anti-HSC70 mAb, and anti-ß6 polyclonal antibody). VB6 cells (B), H400 cells (C), and H357 cells (D) were transfected with control or HAX-1 siRNA and tested in migration assays across Transwell membrane coated with 0.5% bovine serum albumin (BSA) or 10 µg/mL extracellular matrix substrate: fibronectin (FN), type 1 collagen (COL1), laminin (LAM), vitronectin (VN), or 0.5 µg/mL LAP. Columns, mean derived from three separate experiments done in quadruplicate; bars, SD. *, P < 0.001.

 
Cellular migration to LAP was inhibited by HAX-1 down-regulation but carcinoma cell line adherence to LAP and fibronectin, as well as to the non-{alpha}vß6 ligands, type 1 collagen, laminin, and vitronectin, was unaffected by siRNA knockdown (Supplementary Fig. S2).

HAX-1–depleted cells exhibit an impaired ability to internalize {alpha}vß6 integrin. We examined trafficking of {alpha}vß6 using established assays for integrin internalization and recycling (22). Both integrin subunits were internalized with similar kinetics in control siRNA–transfected cells, with internal pools achieving a steady-state level by 30 min in VB6 cells (Fig. 3A ) and by 15 min in H400 cells (Supplementary Fig. S3A), values comparable to those for untreated control cells (data not shown). Conversely, both extent and rate of {alpha} and ß integrin subunit endocytosis were reduced significantly in VB6 cells (Fig. 3A) and H400 cells (Supplementary Fig. S3A) exposed to HAX-1 siRNA. Densitometric analysis confirmed that HAX-1 knockdown caused a significant reduction in accumulated {alpha}v and ß6 integrin subunits over time (Fig. 3A; 46 ± 13% and 58 ± 11%, respectively, in VB6 cells; Supplementary Fig. S3B; 34 ± 9% and 47 ± 15%, respectively, in H400 cells). Using capture-ELISA assays, cells transfected with HAX-1 siRNA had a dramatically decreased internalization rate of {alpha}vß6 integrin compared with control siRNA–transfected cells. Thus, ~50% less surface {alpha}vß6 was accumulated after 60 min in VB6 cells (Fig. 3B) and after 15 min in H400 cells (Supplementary Fig. S3C). We examined {alpha}vß6 recycling (exocytosis) rates in our carcinoma cell lines. Internalization was allowed to proceed for 30 min and accumulated integrin was then chased from the cells for various times and quantified by capture-ELISA. Figure 3C shows that HAX-1 siRNA treatment did not alter the rate of {alpha}vß6 recycling compared with control siRNA–treated cells (Supplementary Fig. S3D; H400 cells). To monitor recycling of early endocytosis, we chased accumulated integrin after 10-min internalization and still showed no change in the {alpha}vß6 recycling rate in HAX-1–depleted cells (data not shown). Moreover, internalization assays using the recycling inhibitor primaquine (0.6 mmol/L; Sigma; ref. 25) confirmed that HAX-1 knockdown inhibited {alpha}vß6 endocytosis but had no effect on recycling (data not shown).


Figure 3
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. HAX-1 depletion inhibits the clathrin-mediated endocytic pathway of {alpha}vß6 integrin. A, {alpha}vß6 internalization assays on VB6 cells transfected with either control or HAX-1 siRNA. Integrin was immunoprecipitated with anti-{alpha}vß6 mAb (7.1C5) and the internalized {alpha}vß6 fractions, for both treatments, were revealed on identically treated Western blots with HRP-conjugated streptavidin. The {alpha}v and ß6 integrin subunits are indicated. Representative blots of three independent experiments. Relative band intensity (5–30 min) was determined by comparing the band intensities from HAX-1 siRNA–treated VB6 cells to that obtained from control siRNA–treated cells, which were assigned a value of 100%. Right, columns, mean of three individual assays; bars, SD. *, P < 0.005. B, capture-ELISA {alpha}vß6 internalization assays on VB6 cells following transfection with control or HAX-1 siRNA. Internalized biotinylated integrin was captured using microtiter wells coated with anti-{alpha}vß6 mAb (7.1C5). The proportion of {alpha}vß6 internalized is expressed as the percentage of the total surface integrin pool. Points, mean from three separate experiments; bars, SD. C, capture-ELISA {alpha}vß6 recycling assays of transfected VB6 cells (control versus HAX-1 siRNA). Recycling integrin was quantified using microtiter wells coated with anti-{alpha}vß6 mAb (7.1C5). Points, mean from three separate experiments; bars, SD. D, i, {alpha}vß6 internalization assays using capture-ELISA. The internalized fractions from VB6 cells treated with control or CHC siRNA were prepared for the times indicated and biotinylated integrin was measured using microtiter wells coated with anti-{alpha}vß6 mAb (7.1C5) as for (B). Points, mean from three separate experiments; bars, SD. ii, HAX-1 and CHC siRNA suppress {alpha}vß6 internalization to a similar extent. VB6 cells were transfected with CHC or HAX-1 siRNA alone or both together (double CHC/HAX-1 siRNA knockdown) before internalized integrin was determined as described. Points, mean from three separate experiments; bars, SD. iii, inhibition of clathrin-mediated endocytosis suppresses {alpha}vß6-mediated carcinoma cell motility toward LAP. VB6 cells transfected with control or CHC siRNA were tested in migration assays across Transwell membrane coated with 0.5% BSA or 0.5 µg/mL LAP. Columns, mean derived from three separate experiments done in quadruplicate; bars, SD. *, P < 0.005.

 
To measure cell-surface expression of {alpha}vß6 following exposure to HAX-1 siRNA, we surface labeled cells with biotin and captured {alpha}vß6 integrin by capture-ELISA. Surface expression of biotinylated {alpha}vß6 following exposure of cells to HAX-1 siRNA increased moderately compared with control cells: 28 ± 4% and 25 ± 5% in VB6 and H400 cells, respectively. Western blot analyses of protein extracts showed that the introduction of HAX-1 siRNA in the cells resulted in no change in relative ß6 levels compared with control treatment cells (Fig. 2A).

Capture ELISA assays showed that neither ß1 (Supplementary Fig. S4A and C) nor {alpha}vß5 (Supplementary Fig. S4B and D) had their internalization rate affected by HAX-1 depletion.

Endocytosis of {alpha}vß6 integrin via the clathrin-mediated endocytosis pathway is required for carcinoma cell migration toward LAP. Clathrin-dependent endocytosis regulates cellular uptake of foot-and-mouth disease virus where {alpha}vß6 is the known virus receptor (26). Thus, we examined if knockdown of CHC, essential for clathrin-coated pit formation, affected {alpha}vß6 endocytosis in our lines. Western blot analyses showed that 48-h CHC siRNA treatment reduced CHC levels effectively (~70% knockdown) in both cell lines but {alpha}vß6 levels were unaffected (Supplementary Fig. S5A). Growth studies over 12 days showed no differences between cells transfected with CHC-siRNA and control treated cells (data not shown). Internalization assays, in combination with Western blotting (Supplementary Fig. S5B), identified a reduction in {alpha}v and ß6 integrin subunit accumulation (Supplementary Fig. S5C; 56 ± 16% and 51 ± 15%, respectively). The internalization rate of {alpha}vß6 quantified by capture-ELISA was reduced in VB6 (Fig. 3D-i) and H400 (Supplementary Fig. S5D) cells to about that observed with HAX-1 knockdown. Double knockdown, using CHC and HAX-1 siRNA, revealed no difference in {alpha}vß6 internalization rate compared with single knockdown with HAX-1 or CHC siRNA, suggesting that both molecules act through the same endocytic pathway (Fig. 3D-ii). Cells treated with CHC siRNA had moderately increased levels of biotinylated {alpha}vß6 surface expression compared with control treated cells as measured by capture-ELISA (15 ± 5% and 14 ± 2% increase in VB6 and H400 cells, respectively).

We analyzed the ability of cells treated with HAX-1 siRNA to internalize the transferrin receptor. As measured by capture-ELISA, HAX-1 depletion caused a decrease in the internalization rate which, although smaller than the effect of CHC depletion, still was significant (Supplementary Fig. S6A–D). These data indicate that HAX-1 contributes to the clathrin-mediated endocytosis pathway. Knocking down CHC with siRNA significantly reduced migration toward LAP by 45% (P < 0.005) and 40% (P = 0.05) in VB6 and H400 cells, respectively (Fig. 3D-iii and Supplementary Fig. S7A). Cells were also transfected with a dominant-negative mutant of Eps15 (E{Delta}95/295), which inhibits clathrin-coated pit formation when overexpressed (21). This dominant-negative mutant blocked migration toward LAP by 43% (P < 0.005) in VB6 cells and by 35% (P < 0.005) in H400 cells (Supplementary Fig. S7B and C). A second dominant-negative mutant of Eps15 (DIII) gave similar results (Supplementary Fig. S7B and C).

Together, these findings suggest that HAX-1 is involved in regulating the clathrin-mediated endocytosis of integrin {alpha}vß6 and, thereby, {alpha}vß6-expressing carcinoma cell motility.

Competitive inhibition of the direct association between HAX-1 and the ß6 integrin subunit blocks integrin internalization and prevents carcinoma cell migration. The Tat sequence was linked to a cargo 14-mer peptide corresponding to the ß6 binding site of HAX-1 (Tat-HAX-1) or a control peptide containing a scrambled version of the same peptide sequence labeled Tat-CON (Fig. 4A ). Coimmunoprecipitation experiments showed that prior exposure of VB6 cells to the Tat-HAX-1 peptide completely disrupted the ß6-HAX-1 interaction; no such disruption followed treatment with Tat-CON (Fig. 4B). Cells exposed to these peptides were used in capture-ELISA assays to monitor {alpha}vß6 internalization (Fig. 4C). As shown, {alpha}vß6 internalization rate was reduced in cells treated with Tat-HAX-1 compared with those treated with equivalent doses of Tat-CON peptide. Equally, cells pretreated with the Tat-HAX-1 peptide displayed a drastic reduction in their migratory ability toward LAP, comparable to that observed after pretreatment of cells with an anti-{alpha}vß6 antibody; cells pretreated with Tat-CON peptide showed no such reduction (Fig. 4D and Supplementary Fig. S7D).


Figure 4
View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Competitive inhibition of the direct association of HAX-1 with integrin {alpha}vß6 inhibits integrin endocytosis and cell motility. A, schematic representation of the Tat (GRKKRRQRRRPPQ)-linked peptide for Tat-HAX-1 (terminal 14 amino acids of HAX-1 containing the binding region for {alpha}vß6) and Tat-CON (14 amino acids of scrambled sequence). B, the association of HAX-1 and {alpha}vß6 was blocked competitively by a Tat-linked peptide containing the {alpha}vß6 binding sequence within HAX-1. Monolayer VB6 cells exposed to Tat-HAX-1 or Tat-CON at 10 µmol/L for 60 min at 37°C were lysed and subjected to immunoprecipitation with anti-{alpha}vß6 mAb (7.1C5) and then Western blotting for HAX-1 (mAb) and ß6 (polyclonal antibody). As control, Tat-CON–treated VB6 cells were immunoprecipitated with isotype-matched IgG. C, VB6 cells were treated with Tat-CON or Tat-HAX-1 peptide, as described, before internalized biotinylated integrin fractions were determined by capture-ELISA. Points, mean from three separate experiments; bars, SD. D, untreated VB6 cells and VB6 cells pretreated with either Tat-CON or Tat-HAX-1 as described were compared for their ability to migrate across Transwell membrane coated with 0.5% BSA or 0.5 µg/mL LAP. Blocking anti-{alpha}vß6 mAb (6.3G9) was added to cells as a control where indicated. Columns, mean derived from three separate experiments done in quadruplicate; bars, SD. *, P < 0.005.

 
HAX-1 is required for integrin {alpha}vß6–dependent carcinoma invasion. We have established that VB6 invasion through Matrigel is {alpha}vß6 dependent (12–14, 27). HAX-1 siRNA–treated VB6 cells invaded significantly less (P < 0.05) through Matrigel, relative to their control siRNA–transfected counterparts (Fig. 5A-i ). Inhibition of invasive function following HAX-1 gene silencing seemed to be specific for {alpha}vß6-mediated invasion. Thus, HAX-1 siRNA treatment of H357 cells, which invade through an {alpha}3- and {alpha}6-dependent mechanism (data not shown), had no effect on their capacity to invade through Matrigel (Fig. 5A-ii). Prior treatment of VB6 cells with the Tat-HAX-1 peptide also inhibited invasion through Matrigel significantly (P < 0.05), comparable to pretreatment with an {alpha}vß6 blocking antibody (Fig. 5A-iii).


Figure 5
View larger version (34K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. Effect of HAX-1 knockdown on carcinoma {alpha}vß6-mediated tumor cell invasion. A, HAX-1 knockdown or blocking the direct association of HAX-1 with the ß6 subunit inhibited {alpha}vß6-dependent carcinoma cell invasion through Matrigel. Effect of control and HAX-1 siRNA transfection on invasion of {alpha}vß6-expressing VB6 cells (i) or {alpha}v-negative H357 cells (ii). iii, VB6 cell invasion following treatment with a Tat-linked peptide (Tat-HAX-1) that blocks binding of HAX-1 to ß6 or a scrambled control (Tat-CON). *, P < 0.05. B, HAX-1 depletion suppresses three-dimensional tumor cell invasion. Cytokeratin staining of 10-d VB6 organotypic culture following transfection with control siRNA (left) or HAX-1 siRNA (right). Bar chart comparing the quantitative invasion index (AU) of control siRNA–transfected versus HAX-1 siRNA–transfected VB6 cells in organotypic assays. Columns, mean of seven individual organotypic experiments; bars, SD. *, P < 0.001. C, HAX-1 and {alpha}vß6 in human oral SCC tumor islands. Immunohistochemistry showing representative {alpha}vß6 (left) and HAX-1 (right) expression (brown) in normal oral epithelium (i) and oral SCC (ii). Bottom, enlargements of the indicated areas in (ii).

 
We also examined invasion using a quantitative organotypic gel assay, which more closely mimics the tumor microenvironment (23), where VB6 invasion is {alpha}vß6 specific (27). HAX-1 siRNA achieved knockdown of HAX-1 expression levels (~90% reduction) up to 12 days posttransfection (Supplementary Fig. S8B). Organotypics were harvested after 10 days (12 days posttransfection) and analyzed. HAX-1 siRNA–treated cells invaded dramatically less than their control siRNA–transfected counterparts, with a reduction of ~90% relative invasion (Fig. 5B; P < 0.001). CHC siRNA caused a similar inhibition of invasive function, compared with control siRNA–treated cells (Supplementary Fig. S8A), suggesting that clathrin-mediated endocytosis is an integral component of the invasive process.

HAX-1 is highly expressed in human oral SCC. We evaluated HAX-1 and {alpha}vß6 expression in normal/hyperplastic, dysplastic epithelium, and oral SCC tumor tissues by immunostaining. Expression of {alpha}vß6 and HAX-1 occasionally occurred at a low level in normal epithelium tissue, but both proteins were up-regulated in SCC, with moderate to strong expression in 90% of tumors (summarized in Table 1 ). Additionally, {alpha}vß6 and HAX-1 were up-regulated in premalignant epithelial dysplasia relative to normal epithelial tissue (Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1. HAX-1 and {alpha}vß6 are highly expressed in human oral SCC tumor islands

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
HAX-1, a ubiquitously expressed protein of 35 kDa, was originally identified as an interaction partner for HS1, a substrate of the Src family of tyrosine kinases (24). Yeast two-hybrid screens also identified HAX-1 as a binding partner for several virus-associated proteins (28–30) and the polycystic kidney disease protein (31). More recently, HAX-1 has been reported to interact physically with the {alpha} subunit of G13 heterotrimeric G protein, an interaction required for G{alpha}13-stimulated cell migration (32).

Although the reported binding profile for HAX-1 is wide, it often seems to be involved in mammalian cell migration. The demonstration that HAX-1 is a binding partner for the ATP-binding cassette–type proteins and that it, together with cortactin, is involved in their internalization via clathrin-coated vesicles suggests a role in receptor endocytosis (33).

Depletion of HAX-1 levels in two carcinoma cell lines impaired the ability to internalize {alpha}vß6 via the clathrin-mediated pathway (Fig. 3 and Supplementary Figs. S3 and S5). Integrin heterodimers have been known to be internalized continually from the plasma membrane into endosomal compartments, and thence back to the cell surface, since the late 1980s (34, 35). However, the precise mechanisms of trafficking remain unknown for most integrin subunits (18). The effects of CHC depletion (Fig. 3 and Supplementary Fig. S5) and dominant-negative mutants of Eps15 (Supplementary Fig. S7) suggest that ß6 is internalized via the clathrin pathway, a possibility supported by the reduction in transferrin receptor internalization subsequent to HAX-1 down-regulation (Supplementary Fig. S6). Interestingly, {alpha}vß5, which has been localized in clathrin-coated pits by electron microscopy (36) and considered the best-characterized of integrin heterodimers with regard to uptake by clathrin-dependent mechanisms (18), was not affected by HAX-1 knockdown (Supplementary Fig. S4). The NXXY motifs in the ß6 cytoplasmic tail, which effect clathrin-dependent internalization of other membrane receptors, were not required for binding to HAX-1 (Fig. 1A). It seems that {alpha}vß6, for which definite clathrin-mediated regulation of endocytosis has been proved (ref. 26, and work herein), does not use NXXY for this direct interaction with HAX-1. We have yet to establish whether {alpha}vß6 internalization can occur via clathrin-independent as well as clathrin-dependent mechanisms as reported for molecules such as TGF-ß receptor II (37).

Chimeric Tat-linked peptides, which mimic the ß6 binding site of HAX-1, were capable of disrupting coimmunoprecipitation with ß6 (Fig. 4B) and disrupting {alpha}vß6-dependent migration and invasion. Integrin internalization can modulate cancer cell migration (38, 39) and Rab11-dependent recycling is implicated in {alpha}6ß4-dependent invasion (40). As membrane type 1 matrix metalloproteinase endocytosis affected extracellular matrix degradation (41, 42), such data suggested that endocytosis might play a key role in regulating invasion-promoting activity of various transformed cells (43). However, our report seems to be the first direct evidence that integrin internalization is essential for the invasive activity of epithelial cancer cells. This conclusion was arrived at using both the Matrigel-Transwell assay and the organotypic assay; in both of which invasion is completely {alpha}vß6 dependent. These findings, in conjunction with a report that increased expression levels of the small GTPase RAB25 that controls integrin recycling determine the aggressiveness of epithelial cancers (44), provide compelling support for the importance of receptor trafficking in transformed cell invasive behavior.

Here, we have shown that HAX-1 and {alpha}vß6 are up-regulated in human epithelial tumor islands. Elevated expression levels of HAX-1 also have been identified in hypoxic tumor progression (45), metastatic pancreatic cancer (46), and in liver, lung, and breast cancer.5 HAX-1 has long been known to share partial sequence similarity with prosurvival members of the BCL2 family (24), and a recent report has shown that HAX-1 deficiency results in an increased apoptosis of myeloid cells, which underlies severe congenital neutropenia (47). Interestingly, neutrophils from such patients show defective migratory responses and impaired cytoskeletal organization (48). We do not believe that the impaired invasive activity we have observed in HAX-1 RNAi–treated tumor cells is due to increased apoptosis. First, we see impaired migration in short-term assays (6 h) where apoptosis is unlikely to play any modifying role. Second, cell counts from three-dimensional cultures revealed no differences between HAX-1–treated and control siRNA–treated populations (data not shown). Thus, effects on invasion seem to be independent of apoptotic events but operate through {alpha}vß6 endocytosis.


   Acknowledgments
 
Grant support: The Association of International Cancer Research (J.F. Marshall) and Cancer Research UK (I.R. Hart).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   Footnotes
 
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

5 http://www.oncomine.org. Back

Received 1/26/07. Revised 2/26/07. Accepted 3/26/07.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Hynes RO. Integrins: bidirectional, allosteric signaling machines. Cell 2002;110:673–87.[CrossRef][Medline]
  2. Aplin AE, Howe AK, Juliano RL. Cell adhesion molecules, signal transduction and cell growth. Curr Opin Cell Biol 1999;11:737–44.[CrossRef][Medline]
  3. Naylor MJ, Li N, Cheung J, et al. Ablation of ß1 integrin in mammary epithelium reveals a key role for integrin in glandular morphogenesis and differentiation. J Cell Biol 2005;171:717–28.[Abstract/Free Full Text]
  4. Reddig PJ, Juliano RL. Clinging to life: cell to matrix adhesion and cell survival. Cancer Metastasis Rev 2005;24:425–39.[CrossRef][Medline]
  5. Yamaguchi H, Wyckoff J, Condeelis J. Cell migration in tumors. Curr Opin Cell Biol 2005;17:559–64.[CrossRef][Medline]
  6. Sheppard D, Rozzo C, Starr L, Quaranta V, Erle DJ, Pytela R. Complete amino acid sequence of a novel integrin ß subunit (ß6) identified in epithelial cells using the polymerase chain reaction. J Biol Chem 1990;265:11502–7.[Abstract/Free Full Text]
  7. Breuss JM, Gallo J, DeLisser HM, et al. Expression of the ß6 integrin subunit in development, neoplasia and tissue repair suggests a role in epithelial remodelling. J Cell Sci 1995;108:2241–51.[Abstract]
  8. Jones J, Watt FM, Speight PM. Changes in the expression of {alpha}v integrins in oral squamous cell carcinomas. J Oral Pathol Med 1997;26:63–8.[CrossRef][Medline]
  9. Ahmed N, Pansino F, Clyde R, et al. Overexpression of {alpha}(v)ß6 integrin in serious epithelial ovarian cancer regulates extracellular matrix degradation via the plasminogen activation cascade. Carcinogenesis 2002;23:237–44.[Abstract/Free Full Text]
  10. Bates RC, Bellovin DI, Brown C, et al. Transcriptional activation of integrin ß6 during the epithelial-mesenchymal transition defines a novel prognostic indicator of aggressive colon carcinoma. J Clin Invest 2005;115:339–47.[CrossRef][Medline]
  11. Thomas GJ, Nystrom ML, Marshall JF. {alpha}vß6 integrin in wound healing and cancer of the oral cavity. J Oral Pathol Med 2006;35:1–10.[CrossRef][Medline]
  12. Thomas GJ, Lewis MP, Hart IR, Marshall JF, Speight PM. {alpha}vß6 integrin promotes invasion of squamous carcinoma cells through up-regulation of matrix metalloproteinase-9. Int J Cancer 2001;92:641–50.[CrossRef][Medline]
  13. Thomas GJ, Lewis MP, Whawell SA, et al. Expression of the {alpha}vß6 integrin promotes migration and invasion in squamous carcinoma cells. J Invest Dermatol 2001;117:67–73.[CrossRef][Medline]
  14. Thomas GJ, Hart IR, Speight PM, Marshall JF. Binding of TGF-ß1 latency-associated peptide (LAP) to {alpha}(v)ß6 integrin modulates behaviour of squamous carcinoma cells. Br J Cancer 2002;87:859–67.[CrossRef][Medline]
  15. Xue H, Atakilit A, Zhu W, Li X, Ramos DM, Pytela R. Role of the {alpha}(v)ß6 integrin in human oral squamous cell carcinoma growth in vivo and in vitro. Biochem Biophys Res Commun 2001;288:610–8.[CrossRef][Medline]
  16. O'Toole TE, Ylanne J, Culley BM. Regulation of integrin affinity states through an NPXY motif in the ß subunit cytoplasmic domain. J Biol Chem 1995;270:8553–8.[Abstract/Free Full Text]
  17. Reszka AA, Hayashi Y, Horwitz AF. Identification of amino acid sequences in the integrin ß1 cytoplasmic domain implicated in cytoskeletal association. J Cell Biol 1992;117:1321–30.[Abstract/Free Full Text]
  18. Caswell PT, Norman JC. Integrin trafficking and the control of cell migration. Traffic 2006;7:14–21.[CrossRef][Medline]
  19. Arnaout MA, Mahalingam B, Xiong JP. Integrin structure, allostery, and bidirectional signaling. Annu Rev Cell Dev Biol 2005;21:381–410.[CrossRef][Medline]
  20. Ahmed N, Niu J, Dorahy DJ, et al. Direct integrin {alpha}vß6-ERK binding: implications for tumour growth. Oncogene 2002;21:1370–80.[CrossRef][Medline]
  21. Benmerah A, Poupon V, Cerf-Bensussan N, Dautry-Varsat A. Mapping of Eps15 domains involved in its targeting to clathrin-coated pits. J Biol Chem 2000;275:3288–95.[Abstract/Free Full Text]
  22. Roberts M, Barry S, Woods A, Van Der Sluijs P, Norman J. PDGF-regulated rab4-dependent recycling of {alpha}vß3 integrin from early endosomes is necessary for cell adhesion and spreading. Curr Biol 2001;11:1392–402.[CrossRef][Medline]
  23. Nystrom ML, Thomas GJ, Stone M, Mackenzie IC, Hart IR, Marshall JF. Development of a quantitative method to analyse tumour cell invasion in organotypic culture. J Pathol 2005;205:468–75.[CrossRef][Medline]
  24. Suzuki Y, Demoliere C, Kitamura D, Takeshita H, Deuschle U, Watanabe T. HAX-1, a novel intracellular protein, localized on mitochondria, directly associates with HS1, a substrate of Src family tyrosine kinases. J Immunol 1997;158:2736–44.[Abstract]
  25. Roberts MS, Woods AJ, Dale TC, Van Der Sluijs P, Norman JC. Protein kinase B/Akt acts via glycogen synthase kinase 3 to regulate recycling of {alpha}vß3 and {alpha}5ß1 integrins. Mol Cell Biol 2004;24:1505–15.[Abstract/Free Full Text]
  26. Berryman S, Clark S, Monaghan P, Jackson T. Early events in integrin {alpha}vß6-mediated cell entry of foot-and-mouth disease virus. J Virol 2005;79:8519–34.[Abstract/Free Full Text]
  27. Nystrom ML, McCulloch D, Weinreb PH, et al. Cyclooxygenase-2 inhibition suppresses {alpha}vß6 integrin-dependent oral squamous carcinoma invasion. Cancer Res 2006;66:10833–42.[Abstract/Free Full Text]
  28. Dufva M, Olsson M, Rymo L. Epstein-Barr virus nuclear antigen 5 interacts with HAX-1, a possible component of the B-cell receptor signalling pathway. J Gen Virol 2001;82:1581–7.[Abstract/Free Full Text]
  29. Kawaguchi Y, Nakajima K, Igarashi M, et al. Interaction of Epstein-Barr virus nuclear antigen leader protein (EBNA-LP) with HS1-associated protein X-1: implication of cytoplasmic function of EBNA-LP. J Virol 2000;74:10104–11.[Abstract/Free Full Text]
  30. Sharp TV, Wang HW, Koumi A, et al. K15 protein of Kaposi's sarcoma-associated herpesvirus is latently expressed and binds to HAX-1, a protein with antiapoptotic function. J Virol 2002;76:802–16.[Abstract/Free Full Text]
  31. Gallagher AR, Cedzich A, Gretz N, Somlo S, Witzgall R. The polycystic kidney disease protein PKD2 interacts with Hax-1, a protein associated with the actin cytoskeleton. Proc Natl Acad Sci U S A 2000;97:4017–22.[Abstract/Free Full Text]
  32. Radhika V, Onesime D, Ha JH, Dhanasekaran N. G{alpha}13 stimulates cell migration through cortactin-interacting protein Hax-1. J Biol Chem 2004;279:49406–13.[Abstract/Free Full Text]
  33. Ortiz DF, Moseley J, Calderon G, Swift AL, Li S, Arias IM. Identification of HAX-1 as a protein that binds bile salt export protein and regulates its abundance in the apical membrane of Madin-Darby canine kidney cells. J Biol Chem 2004;279:32761–70.[Abstract/Free Full Text]
  34. Bretscher MS. Endocytosis and recycling of the fibronectin receptor in CHO cells. EMBO J 1989;8:1341–8.[Medline]
  35. Sczekan MM, Juliano RL. Internalization of the fibronectin receptor is a constitutive process. J Cell Physiol 1990;142:574–80.[CrossRef][Medline]
  36. De Deyne PG, O'Neill A, Resneck WG, Dmytrenko GM, Pumplin DW, Bloch RJ. The vitronectin receptor associates with clathrin-coated membrane domains via the cytoplasmic domain of its ß5 subunit. J Cell Sci 1998;111:2729–40.[Abstract]
  37. Le Roy C, Wrana JL. Clathrin- and non-clathrin-mediated endocytic regulation of cell signalling. Nat Rev Mol Cell Biol 2005;6:112–26.[CrossRef][Medline]
  38. Parsons M, Keppler MD, Kline A, et al. Site-directed perturbation of protein kinase C-integrin interaction blocks carcinoma cell chemotaxis. Mol Cell Biol 2002;22:5897–911.[Abstract/Free Full Text]
  39. Lawson MA, Maxfield FR. Ca(2+)- and calcineurin-dependent recycling of an integrin to the front of migrating neutrophils. Nature 1995;377:75–9.[CrossRef][Medline]
  40. Yoon S, Shin S, Mercurio AM. Hypoxia stimulates carcinoma invasion by stabilizing microtubules and promoting the Rab11 trafficking of the {alpha}6ß4 integrin. Cancer Res 2005;65:2761–9.[Abstract/Free Full Text]
  41. Lafleur MA, Mercuri FA, Ruangpanit N, Seiki M, Sato H, Thompson EW. Type I collagen abrogates the clathrin-mediated internalization of membrane type 1 matrix metalloproteinase (MT1-MMP) via the MT1-MMP hemopexin domain. J Biol Chem 2006;281:6826–40.[Abstract/Free Full Text]
  42. Jiang A, Lehti K, Wang X, Weiss SJ, Keski-Oja J, Pei D. Regulation of membrane-type matrix metalloproteinase 1 activity by dynamin-mediated endocytosis. Proc Natl Acad Sci U S A 2001;98:13693–8.[Abstract/Free Full Text]
  43. Polo S, Di Fiore PP. Endocytosis conducts the cell signaling orchestra. Cell 2006;124:897–900.[CrossRef][Medline]
  44. Cheng KW, Lahad JP, Kuo WL, et al. The RAB25 small GTPase determines aggressiveness of ovarian and breast cancers. Nat Med 2004;10:1251–6.[CrossRef][Medline]
  45. Jiang Y, Zhang W, Kondo K, et al. Gene expression profiling in a renal cell carcinoma cell line: dissecting VHL and hypoxia-dependent pathways. Mol Cancer Res 2003;1:453–62.[Abstract/Free Full Text]
  46. Velculescu VE, Zhang L, Vogelstein B, Kinzler KW. Serial analysis of gene expression. Science 1995;270:484–7.[Abstract/Free Full Text]
  47. Klein C, Grudzien M, Appaswamy G, et al. HAX-1 deficiency causes autosomal recessive severe congenital neutropenia (Kostmann disease). Nat Genet 2007;39:86–92.[CrossRef][Medline]
  48. Eisner J, Roesler J, Emmendorffer A, Lohmann-Mathes ML, Welte K. Abnormal regulation in the signal transduction in neutrophils from patients with severe congenital neutropenia: relation of impaired mobilization of cytosolic free calcium to altered chemotaxis, superoxide onion generation and F-actin content. Exp Hematol 1993;21:38–46.[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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 Google Scholar
Google Scholar
Right arrow Articles by Ramsay, A. G.
Right arrow Articles by Marshall, J. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ramsay, A. G.
Right arrow Articles by Marshall, J. F.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online