HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 42, 39703-39712, October 18, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/42/39703    most recent M204320200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bourguignon, L. Y. W. Articles by Zhou, B. Search for Related Content PubMed PubMed Citation Articles by Bourguignon, L. Y. W. Articles by Zhou, B. Social Bookmarking                      What's this?

Hyaluronan Promotes Signaling Interaction between CD44 and the Transforming Growth Factor  Receptor I in Metastatic Breast Tumor Cells^*

Lilly Y. W. Bourguignon, Patrick A. Singleton^§, Hongbo Zhu, and Bo Zhou

From the Department of Medicine, University of California, San Francisco, and the Endocrine Unit, Veterans Affairs Medical Center, San Francisco, California 94121

Received for publication, May 2, 2002, and in revised form, June 13, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In this study we have examined the interaction between CD44 (a hyaluronan (HA) receptor) and the transforming growth factor  (TGF-) receptors (a family of serine/threonine kinase membrane receptors) in human metastatic breast tumor cells (MDA-MB-231 cell line). Immunological data indicate that both CD44 and TGF- receptors are expressed in MDA-MB-231 cells and that CD44 is physically linked to the TGF- receptor I (TGF-RI) (and to a lesser extent to the TGF- receptor II (TGF-RII)) as a complex in vivo. Scatchard plot analyses and in vitro binding experiments show that the cytoplasmic domain of CD44 binds to TGF-RI at a single site with high affinity (an apparent dissociation constant (K_d) of ~1.78 nM). These findings indicate that TGF-RI contains a CD44-binding site. Furthermore, we have found that the binding of HA to CD44 in MDA-MB-231 cells stimulates TGF-RI serine/threonine kinase activity which, in turn, increases Smad2/Smad3 phosphorylation and parathyroid hormone-related protein (PTH-rP) production (well known downstream effector functions of TGF- signaling). Most importantly, TGF-RI kinase activated by HA phosphorylates CD44, which enhances its binding interaction with the cytoskeletal protein, ankyrin, leading to HA-mediated breast tumor cell migration. Overexpression of TGF-RI by transfection of MDA-MB-231 cells with TGF-RIcDNA stimulates formation of the CD44·TGF-RI complex, the association of ankyrin with membranes, and HA-dependent/CD44-specific breast tumor migration. Taken together, these findings strongly suggest that CD44 interaction with the TGF-RI kinase promotes activation of multiple signaling pathways required for ankyrin-membrane interaction, tumor cell migration, and important oncogenic events (e.g. Smad2/Smad3 phosphorylation and PTH-rP production) during HA and TGF--mediated metastatic breast tumor progression.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CD44, a hyaluronan (HA)¹ receptor (1), belongs to a family of transmembrane glycoproteins that exist as several isoforms (2). Cell surface expression of certain CD44 isoforms is closely correlated with breast tumor development and metastasis (3-8). Most often, CD44 isoforms are up-regulated in breast carcinomas (3-8). In fact, the presence of a high level of various CD44 isoform (particularly CD44s (the standard form), CD44v3, and CD44v10) expression is emerging as an important metastatic tumor marker in a number of carcinomas and is also implicated in the unfavorable prognosis for a variety of cancers (7). Carcinomas expressing high levels of CD44 isoforms are more malignant than those carcinomas with a low level of CD44 isoform expression (3-8). Cells expressing a high level of CD44 isoforms also display enhanced HA binding that increases their migration capability (9-12). Recently, a number of studies indicate that interaction of certain extracellular matrix components (e.g. HA) with cells triggers the cytoplasmic domain of various CD44 isoforms to bind unique downstream oncogenic signaling molecules: Tiam1 (9), Vav2 (10), RhoA-activated ROK (11), c-Src kinase (12), and p185^HER2 (13) and to coordinate intracellular signaling pathways (e.g. Rho/Ras signaling and receptor-linked/non-receptor-linked tyrosine kinase pathways) leading to the onset of multiple cellular functions (e.g. tumor cell growth, migration, and invasion) and breast tumor progression.

CD44 isoforms are also directly involved in the binding of cytoskeletal proteins such as ankyrin (14, 15). Deletion mutation analyses indicate that at least two sub-regions within the CD44 cytoplasmic domain contribute to the ankyrin binding: region I (e.g. the high affinity ankyrin-binding region) and region II (e.g. the regulatory region). In particular, the region I ankyrin-binding domain (e.g. NGGNGTVEDRKPSEL between amino acids 306 and 320 in the mouse CD44 (14) and NSGNGAVEDRKPSGL between amino acids 304 and 318 in the human CD44 (15)) is required for hyaluronan-mediated binding and cell adhesion (14, 15). An ankyrin-binding domain of CD44 isoforms has also been shown as necessary for oncogenic signaling and tumor cell transformation (15, 16). Moreover, certain ankyrin fragments (e.g. the ankyrin repeat domain (ARD) and/or the subdomain 2 (S2) of ARD)) have been identified as an ankyrin-binding region for both CD44 (16) and Tiam1 (17). Overexpression of these ankyrin fragments promotes hyaluronan-dependent and CD44-specific tumor cell migration (16). These observations support the notion that CD44-ankyrin interaction is not only very important for presenting CD44 properly for hyaluronan binding but is also required for cytoskeleton activation during hyaluronan signaling.

Cytokines, such as the transforming growth factor  (TGF-) superfamily (18, 19), are multifunctional peptides that are known to regulate a diverse set of cellular processes by binding to their specific surface receptors (18, 19). Three mammalian TGF- isoforms (TGF-1, TGF-2, and TGF-3), coded by different genes, have been identified (20). TGF- interacts with three surface receptors known as type I (TGF-RI), type II (TGF-RII), and type III (TGF-III) receptors (18, 19). TGF-1 mediates its activity by high affinity binding to the type II (TGF-II) receptor, which has been identified as a 70-80-kDa transmembrane protein with a cytoplasmic serine/threonine kinase domain (18-22). For cellular signaling, the TGF-RII requires both its kinase activity and association with members of a series of related 55-kDa TGF-RIs (designated as activin receptor-like kinase-ALK (1-6 different subtypes)). Of these, only ALK5 has been shown to represent a functional TGF-RI (18-23). Subsequently, the TGF- signal is propagated from the plasma membranes (via TGF-RII/TGF-RI kinases) by phosphorylation of the Smad proteins that belong to a class of intracellular mediators known to regulate transcriptional responses and gene expression in the nucleus (24-26). The type III receptor (TGF-RIII) also binds TGF- and may function in capturing TGF- for presentation to the signaling receptors (27, 28). In cancers, the TGF- receptors on tumor cells are often mutated or functionally defective (29). For example, defective ligand binding to the cell surface caused by the absence of TGF-RII, or expression of a truncated form or splice variant of TGF-RII, may account for the resistance to activated TGF- in cancer cells (30-32). Some studies also indicate that decreased expression of TGF-RII may contribute to breast cancer progression and is related to a more aggressive phenotype in both in situ and invasive carcinomas (33-36).

TGF- is known to increase parathyroid hormone-related protein (PTH-rP) production by cancer cells (37). PTH-rP shares many, but not all, properties of parathyroid hormone (PTH). Both of these hormones share homology in 8 of the first 13 amino acids and bind to the type 1 PTH receptor (38-40, 44, 45). PTH-rP, like PTH, is a potent activator of bone resorption but unlike PTH does not appear to stimulate bone formation (38-40, 44, 45). This makes PTH-rP a particularly potent osteolytic agent (38-40, 44, 45). Thus, cells expressing PTH-rP in bone are likely to gain a foothold thereby stimulating the removal of the calcified matrix (38-40, 44, 45). However, buried within the matrix of bone are high concentrations of certain cytokines (e.g. TGF-) that can feed back on the metastases to promote their tumor growth (33-36). Mice inoculated with breast tumor cells (e.g. MDA-MB-231 cells) engineered to express a dominant-negative form of the TGF- receptor had fewer and smaller osteolytic metastases (35, 36, 41). The net result in this situation is that PTH-rP production by breast cancers increases metastasis of breast cancer to bone. Because both CD44 and TGF--mediated signaling events are important in breast tumor progression, the question of whether the interaction between CD44 and TGF- receptor(s) plays a significant role in regulating metastatic breast tumor cell-specific behaviors (e.g. Smad activation, PTH-rP production, membrane-cytoskeleton interaction, and tumor cell migration) is the primary focus of this study.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell Culture-- The breast tumor cell line (MDA-MB-231 cells) was obtained from the American Type Culture Collection (ATCC) and grown in Eagle's minimum essential medium supplemented with Earle's salt solution, essential and non-essential amino acids, vitamins, and 10% fetal bovine serum.

Antibodies and Reagents-- Monoclonal rat anti-human CD44 antibody (Clone, 020; isotype, IgG_2b; obtained from CMB-TECH, Inc., San Francisco) used in this study recognizes a common determinant of the CD44 class of glycoproteins. For the preparation of polyclonal rabbit anti-CD44v3 antibody, specific synthetic peptides (~15-17 amino acids unique for the v3 sequence of CD44) were prepared, respectively, by the Peptide Laboratories using an Advanced Chemtech automatic synthesizer (model ACT350). All CD44 antibodies were prepared using conventional DEAE-cellulose chromatography and tested to be monospecific (by immunoblot assays). Mouse monoclonal anti-ankyrin was prepared as described previously (42). Monoclonal mouse anti-HA1 (hemagglutinin epitope) antibody (clone 12 CA5) and rabbit anti-phospho-Smad2 (Ser-465/467)/Smad3 IgG were obtained from Roche Molecular Biochemicals and Upstate Biotechnology, Inc., respectively. Both rabbit anti-TGF-RI IgG (specific for the ALK-5 form of TGF-RI p55) and rabbit anti-TGF-RII IgG (specific for TGF-RII p70) were purchased from Santa Cruz Biotechnology. Rabbit anti-phosphothreonine antibody and rabbit anti-phosphoserine antibody were purchased from Zymed Laboratories Inc..

Cloning, Expression, and Purification of CD44 Cytoplasmic Domain (CD44cyt) from Escherichia coli-- The cytoplasmic domain of human CD44 (CD44cyt) was cloned into pFLAG-AST using the PCR-based cloning strategy. By using human CD44 cDNA as template, one PCR primer pair (left, FLAG-EcoRI; right, FLAG-XbaI) was designed to amplify complete CD44 cytoplasmic domain. The amplified DNA fragments were one-step cloned into a pCR2.1 vector and sequenced. Then the DNA fragments were cut out by double digestion with EcoRI and XbaI and subcloned into EcoRI/XbaI double-digested pFLAG-AST (Eastman Kodak Co.) to generate FLAG-pCD44cyt construct. The nucleotide sequence of FLAG/CD44cyt junction was confirmed by sequencing. The recombinant plasmids were transformed to BL21-DE3 to produce FLAG-CD44cyt fusion protein. The FLAG-CD44cyt fusion protein was further purified by anti-FLAG M2 affinity gel column (Kodak). The nucleotide sequence of primers used in this cloning protocol is as follows: FLAG-EcoRI, 5'-GAGAATTCGAACAGTCGAAGAAGGTGTCTCTTAAGC-3'; FLAG-XbaI, 5'-AGCTCTAGATTACACCCCAATCTTCAT-3'.

Cell Transfection-- The cDNA encoding human TGF-RI (full-length) (43) is driven by the cytomegalovirus promoter and preceded by the hemagglutinin epitope (HA1) tag in the expression vector pCGN-Bam, which contains the hygromycin-resistant gene as a selection marker. To establish a transient expression system, cultured cells (e.g. MDA-MB-231 cells or COS-7 cells) were transfected with two plasmid DNAs (e.g. HA1-tagged TGF-RIcDNA or vector alone) using LipofectAMINE 2000. These transfectants were then analyzed for their protein expression (e.g. TGF-RI-related proteins) by immunoprecipitation/immunoblot, TGF-RI kinase activity, TGF-RI interaction with CD44 and ankyrin, as well as breast tumor cell migration assays as described below.

Immunoblotting and Immunoprecipitation Techniques-- Unlabeled MDA-MB-231 cells (or surface-biotinylated) were solubilized in 50 mM HEPES (pH 7.5), 150 mM NaCl, 20 mM MgCl₂, 1.0% Nonidet P-40, 0.2 mM Na₃VO₄, 0.2 mM phenylmethylsulfonyl fluoride, 10 µg/ml leupeptin, and 5 µg/ml aprotinin. The sample was then centrifuged at 14,927 × g for 15 min, and the supernatant was analyzed by SDS-PAGE in a 5 or 7.5% polyacrylamide gel. Separated polypeptides were then transferred onto nitrocellulose filters. After blocking nonspecific sites with 2% bovine serum albumin, the nitrocellulose filters were incubated with each of the specific immuno-reagents (e.g. rat anti-CD44 IgG (5 µg/ml), rabbit anti-CD44v3 IgG (5 µg/ml), rabbit anti-TGF-RI IgG (5 µg/ml), and rabbit anti-TGF-RII IgG (5 µg/ml)) followed by incubating with horseradish peroxidase-labeled immunoreagents (e.g. goat anti-rat IgG, goat anti-rabbit IgG, or goat anti-mouse IgG) or ExtrAvidin peroxidase (to detect surface-biotinylated proteins). The blots were then developed by the ECL system (Amersham Biosciences). For analyzing the complex formation between endogenous TGF-RI, TGF-RII, or ankyrin into CD44v3 complex, MDA-MB-231 cells treated with various reagents (e.g. HA (50 µg/ml; Sigma) or TGF-1 (50 ng/ml; R & D Systems) or pre-treated with anti-CD44 antibody followed by HA (50 µg/ml) or TGF-1 (50 ng/ml) treatment or without any treatment) were solubilized by 1.0% Nonidet P-40 and immunoprecipitated with rat anti-CD44 antibody followed by anti-TGF-RI (or anti-TGF-RII or anti-ankyrin)-mediated immunoblot, or anti-phosphoserine/anti-phosphothreonine-mediated immunoblot, or immunoprecipitated with anti-TGF-RI antibody followed by anti-CD44-mediated immunoblot or anti-phosphoserine/anti-phosphothreonine-mediated immunoblot, respectively. In some experiments, MDA-MB-231 cells (e.g. untransfected or transfected with HA1-tagged TGF-RIcDNA or vector only) (either treated with HA (50 µg/ml) or TGF-1 (50 ng/ml) or without any treatment) were immunoprecipitated with rabbit anti-CD44v3 IgG followed by immunoblotting with rat anti-CD44 (or mouse anti-HA1 (hemagglutinin epitope) IgG or mouse anti-ankyrin IgG) for 1 h at room temperature followed by incubation with horseradish peroxidase-conjugated goat anti-rabbit IgG (or goat anti-mouse IgG) (1:10,000 dilution) at room temperature for 1 h.

In some experiments, Nonidet P-40-solubilized cell lysate (isolated from cells treated with HA (50 µg/ml) or TGF-1 (50 ng/ml) or pre-treated with anti-CD44 antibody followed by HA (50 µg/ml) or TGF-1 (50 ng/ml) treatment or without any treatment) was analyzed by SDS-PAGE followed by immunoblotting with rabbit anti-phospho-Smad2 (50 µg/ml) or immunoprecipitated with anti-Smad3 followed by anti-phosphothreonine, anti-phosphoserine, and anti-Smad3-mediated immunoblot, respectively. These blots were then treated with peroxidase-conjugated goat anti-rabbit IgG and ECL chemiluminescence reagent.

In Vitro Binding of CD44cyt to the TGF-RI-- Aliquots (0.5-1 ng of protein) of HA1-tagged TGF-RI (isolated from COS-7 or MDA-MB-231 cells)-conjugated Sepharose beads were incubated in 0.5 ml of binding buffer (20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.1% bovine serum albumin, and 0.05% Triton X-100) containing various concentrations (10-800 ng/ml) of ¹²⁵I-labeled cytoplasmic domain of CD44 (CD44cyt) fusion protein (5,000 cpm/ng protein) at 4 °C for 4 h. Specifically, equilibrium-binding conditions were determined by performing a time course (1-10 h) of ¹²⁵I-labeled CD44cyt binding to TGF-RI at 4 °C. The binding equilibrium was found to be established when the in vitro CD44-TGF-RI binding assay was conducted at 4 °C after 4 h. Following binding, the CD44-TGF-RI-conjugated beads were washed extensively in binding buffer, and the beads-bound radioactivity was counted. Nonspecific binding was determined using a 50-100-fold excess of unlabeled CD44cyt in the presence of the same concentration of ¹²⁵I-labeled CD44cyt. Nonspecific binding, which represented ~20% of the total binding, was always subtracted from the total binding. Our binding data are highly reproducible. The values expressed under "Results" represent an average of triplicate determinations of 3-5 experiments with an S.D. less than ±5%.

Protein Phosphorylation Assay in Vitro-- The kinase reaction was carried out in 50 µl of the reaction mixture containing 40 mM Tris-HCl (pH 7.5), 2 mM EDTA, 1 mM dithiothreitol, 7 mM MgCl₂, 0.1% CHAPS, 0.1 µM calyculin A, 100 µM [-³²P]ATP (15-600 mCi/mmol), purified enzymes (e.g. 100 ng of TGF-RI kinase isolated from MDA-MD-231 cells either treated with HA (50 µg/ml) or TGF-1 (50 ng/ml) or without any treatment), and 1 µg of cellular proteins (e.g. myelin basic protein and purified CD44v3). After incubating at 30 °C for 2 h, the reaction mixtures were boiled in SDS-sample buffer and subjected to SDS-PAGE. The protein bands were revealed by silver stain, and the radiolabeled bands were visualized by fluorography or analyzed by liquid scintillation counting.

Binding of ¹²⁵I-Labeled Ankyrin to CD44v3-- Purified ¹²⁵I-labeled ankyrin (0.35 nM protein, 1.5 × 10⁴ cpm/ng) was incubated with purified CD44v3 (bound to anti-CD44v3-conjugated beads) (0.80 µg of protein in TGF-RI phosphorylated or unphosphorylated form, prepared according the methods described above) in 0.5 ml of the binding buffer (20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.1% (w/v) bovine serum albumin, and 0.05% Triton X-100). Binding was carried out at 4 °C for 5 h under equilibrium conditions. Equilibrium conditions were determined by performing a time course (e.g. 1-10 h) of the binding reaction. Following binding, the beads were washed in the binding buffer, and the bead-bound radioactivity was determined. Nonspecific binding was determined in the presence of either a 100-fold excess of unlabeled ankyrin or using bovine serum albumin-conjugated Sepharose beads. Nonspecific binding was ~20-30% of the total binding and was subtracted from the total binding.

Measurement of PTH-rP Production-- Breast tumor cells (MDA-MB-231 cells) were washed three times with serum-free (SF)-DMEM and incubated in 3 ml of SF-DMEM containing various reagents (e.g. HA (50 µg/ml) or TGF-1 (50 ng/ml) or anti-CD44 antibody plus HA (50 µg/ml) or TGF-1 (50 ng/ml) treatment or without any treatment) for 24 h at 37 °C in a 5% CO₂ humidified chamber. Subsequently, PTH-rP concentrations in the conditioned medium and cells were determined using a two-site immunoradiometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA) that detects concentrations as low as 0.3 pM PTH-rP (35, 36). Statistical analysis was done using the Student's t test. All data were expressed as the mean ± S.D.

Cell Migration Assay-- Twenty four transwell units were used for monitoring in vitro cell migration as described previously (9-12). Specifically, the 8-µm porosity polycarbonate filters were used for the cell migration assay (9-12). MDA-MB-231 cells (~1 × 10⁴ cells/well in phosphate-buffered saline, pH 7.2) (in the presence or absence of HA (50 µg/ml) or TGF-1 (50 ng/ml) rat anti-CD44 antibody (50 µg/ml) or cytochalasin D (20 µg/ml)) were placed in the upper chamber of the transwell unit. In some cases, MDA-MB-231 cells were transfected with either HA1-tagged TGF-RI bcDNA or vector alone. The medium containing high glucose DMEM supplemented with 50 µg/ml hyaluronan was placed in the lower chamber of the transwell unit. After 18 h of incubation at 37 °C in a humidified 95% air, 5% CO₂ atmosphere, vital stain 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (Sigma) was added at a final concentration of 0.2 mg/ml to both the upper and the lower chambers and incubated for an additional 4 h at 37 °C. Migratory cells at the lower part of the filter were removed by swabbing with small pieces of Whatman filter paper. Both the polycarbonate filter and the Whatman paper were placed in dimethyl sulfoxide to solubilize the crystal. Color intensity was measured in 570 nm. Cell migration was determined by measuring the percent of total cells that migrated to the lower side of the polycarbonate filters by standard cell number counting methods as described previously (9-12). The CD44-specific cell migration was determined by subtracting nonspecific cell migration (i.e. cells migrate to the lower chamber in the presence of rat anti-CD44 antibody treatment) from the total migratory cells in the lower chamber. Each assay was set up in triplicate and repeated at least 3 times. All data were analyzed statistically using the Student's t test, and statistical significance was set at p < 0.01.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Characterization of CD44 and TGF- Receptor Expression in Breast Tumor Cells-- Breast cancer cells overexpress several variant isoforms of the transmembrane protein, CD44 (3-8). These different CD44 variant (CD44v) isoforms appear to confer on breast cancer cells the malignant properties of increased invasion, migration, and proliferation (46).

To examine CD44 expression on the surface of breast tumor cells (MDA-MB-231 cells), we have utilized surface biotinylation techniques and a specific monoclonal rat anti-CD44 antibody (recognizing a common determinant of the CD44 class of glycoproteins, including various variant isoforms)-mediated immunoprecipitation (Fig. 1, lane 1). Our results indicate that multiple surface-biotinylated polypeptides (~125, 85, 70, and 55 kDa) are selectively immunoprecipitated with the monoclonal rat anti-CD44 antibody (Fig. 1, lane 1). In order to further identify the presence of particular CD44 isoform(s) in MDA-MB-231 cells, we immunoblotted these rat anti-CD44-precipitated surface proteins with a specific rabbit antibody against CD44v3. Our data show that a single band of the CD44v3 protein is expressed in MDA-MB-231 cells (Fig. 1, lane 2) which corresponds to the surface-labeled 85-kDa polypeptide (Fig. 1, lane 1). No CD44-containing material is observed in control samples when normal rat IgG or pre-immune rabbit IgG is used in these experiments (data not shown).

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Analysis of CD44v3·TGF- receptor complex in human breast tumor cells (MDA-MB-231 cells). Unlabeled MDA-MB-231 cells (or surface-biotinylated) were solubilized in 50 mM HEPES (pH 7.5), 150 mM NaCl, 20 mM MgCl2, and 1% Nonidet P-40 buffer followed by SDS-PAGE analyses and immunoblot and/or immunoprecipitation by various immuno-reagents (e.g. rabbit anti-CD44v3 and/or anti-TGF-RI and -RII antibodies) as described under "Materials and Methods." Lane 1, immunoprecipitation of surface-biotinylated MDA-MB-231 cells using monoclonal rat anti-CD44 antibody (recognizing a common determinant of the CD44 class of glycoproteins, including variant isoforms). Lane 2, immunoblot of rat anti-CD44-immunoprecipitated materials with rabbit anti-CD44v3 antibody. Lane 3, immunoprecipitation of surface-biotinylated MDA-MB-231 cells using rabbit anti-CD44v3 antibody (note that a 55-kDa polypeptide is in the complex with CD44v3). Lane 4, detection of TGF-RI in the CD44v3 complex by anti-CD44v3-mediated immunoprecipitation followed by immunoblotting with anti-TGF-RI-specific antibody (note that the TGF-RI is detected in the complex with CD44v3). Lane 5, detection of TGF-RII in the CD44v3 complex by anti-CD44v3-mediated immunoprecipitation followed by immunoblotting with anti-TGF-RII-specific antibody (note that the TGF-RII is not detected in the complex with CD44v3). Lane 6, detection of CD44v3 in the TGF-RI complex by anti-TGF-RI-mediated immunoprecipitation followed by immunoblotting with anti-CD44v3-specific antibody. Lane 7, immunoblot of MDA-MB-231 cell lysate with anti-TGF-RII antibody.

Cytokines, such as TGF-, are known to regulate cellular processes by binding to their specific surface receptors (18, 19). Previous studies (47) have shown that both TGF- receptor I (RI, ~55-kDa polypeptide) and TGF- receptor II (RII, ~70-80-kDa polypeptide) are expressed in breast tumor cells such as MDA-MB-231 cells. In this study, using surface biotinylation techniques and specific anti-CD44v3 immunoprecipitation, we have determined that the 85-kDa surface CD44v3 and a 55-kDa surface protein are closely associated in a complex in MDA-MB-231 cells (Fig. 1, lane 3). We have also analyzed these anti-CD44v3-precipitated immunocomplexes by immunoblotting with either anti-TGF-RI (Fig. 1, lane 4) or TGF-RII antibody (Fig. 1, lane 5). Our results reveal the presence of the TGF-RI protein (~55-kDa polypeptide) (Fig. 1, lane 4) but not the TGF-RII protein (Fig. 1, lane 5) in the anti-CD44v3-immunoprecipitated materials. Furthermore, we have carried out anti-TGF-RI-mediated immunoprecipitation followed by anti-CD44v3 immunoblot (Fig. 1, lane 6). The results of this procedure indicate that the 85-kDa CD44v3 band is also present in anti-TGF-RI-immunoprecipitated materials (Fig. 1, lane 6). In order to confirm that the failure of the TGF-RII association with CD44v3 is not due to the lack of TGF-RII expression in MDA-MB-231 cells, we have conducted an immunoblot analysis of MDA-MB-231 cell lysate using anti-TGF-RII antibody. Our results clearly indicate that the 70-80-kDa TGF-RII is expressed in MDA-MB-231 cells (Fig. 1, lane 7). These findings clearly establish the fact that CD44v3 is physically linked to the TGFRI in vivo in the breast tumor cells (MDA-MB-231 cells). The fact that CD44v3 forms a complex with TGF-RI (but not TGF-RII) suggests that a specific interaction occurs between CD44v3 and TGF-RI.

To further test whether TGF- receptors such as TGF-RI are involved in the direct binding to CD44 in vitro, we have used purified recombinant TGF-RI and the FLAG-tagged cytoplasmic domain of CD44 (FLAG-CD44cyt) fusion protein to identify the TGF-RI-binding site on the CD44 molecule. Specifically, we have tested the binding of TGF-RI to ¹²⁵I-labeled FLAG-CD44cyt under equilibrium binding conditions. The results of a Scatchard plot analysis presented in Fig. 2 demonstrate that the cytoplasmic domain of CD44 (CD44cyt) binds to TGF-RI at a single site with high affinity with an apparent dissociation constant (K_d) of ~1.78 nM. These findings further support the notion that a strong binding interaction occurs between CD44 and TGF-RI.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Scatchard plot analysis of the binding interaction between 125I-labeled FLAG-CD44cyt and TGF-RI. Various concentrations of 125I-labeled FLAG-CD44cyt were incubated with the TGF-RI-coupled beads at 4 °C for 4 h. Following binding, beads were washed extensively in binding buffer, and the bead-bound radioactivity was counted. As a control, 125I-labeled FLAG-CD44cyt was also incubated with uncoated beads to determine the binding observed due to the nonspecific binding of the ligand. Nonspecific binding, which represented ~20% of the total binding, was always subtracted from the total binding. Our binding data are highly reproducible. The values expressed under "Results" represent an average of triplicate determinations of 3-5 experiments with an S.D. less than ±5%.

HA-activated CD44/TGF-RI Kinase and Signaling Events-- HA is known to be involved in certain pathophysiological processes. For example, high levels of HA in solid tumors (e.g. breast tumors) appear to be closely associated with tumor progression and metastasis (48, 49). In this study, we have determined that CD44v3-associated TGF-RI serine/threonine kinase is significantly up-regulated by HA treatment as detected by anti-phosphoserine immunoblot (Fig. 3A, lane 2) or anti-phosphothreonine immunoblotting, respectively (Fig. 3B, lane 2). The level of TGF-RI serine/threonine phosphorylation is relatively low in untreated cells (Fig. 3, A, lane 1, and B, lane 1) or those cells pre-treated with anti-CD44 followed by HA treatment (Fig. 3, A, lane 3, and B, lane 3). As a positive control, we have confirmed that TGF- activates TGF-RI serine and threonine kinases (Fig. 3, A, lane 4, and B, lane 4). No significant inhibition of serine/threonine phosphorylation on TGF-RI is observed in cells treated with anti-CD44 followed by TGF- treatment (Fig. 3, A, lane 5, and B, lane 5). These observations strongly support the conclusion that HA-mediated TGF-RI kinase activity is CD44-dependent, whereas TGF--stimulated TGF-RI kinase activity does not involve CD44. Of course, we cannot preclude the possibility that HA is also capable of interacting with other binding protein(s) which is(are) linked to TGF--regulated signaling pathways.

View larger version (34K): [in this window] [in a new window]   Fig. 3.   Detection of TGF-RI phosphorylation. MDA-MB-231 cells treated with various reagents (e.g. HA (50 µg/ml) or TGF-1 (50 ng/ml) or pre-treated with anti-CD44 antibody followed by HA (50 µg/ml) or TGF- (50 ng/ml) treatment or without any treatment) were solubilized by 1.0% Nonidet P-40 and immunoprecipitated (IP) with anti-TGF-RI antibody followed by anti-phosphoserine- (A) and anti-phosphothreonine (B)-mediated immunoblot as described under "Materials and Methods." Lane 1, untreated cells; lane 2, cells treated with HA (50 µg/ml); lane 3, cells pre-treated with anti-CD44 antibody followed by HA (50 µg/ml); lane 4, cells treated with TGF-1 (50 ng/ml); lane 5, cells pre-treated with anti-CD44 antibody followed by TGF-1 (50 ng/ml) treatment.

Both Smad2/Smad3 phosphorylation (24-26, 37) and PTH-rP production (35-37, 41) are known to be closely associated with TGF- signaling. In this study, we have observed that both Smad2/Smad3 phosphorylation (Fig. 4, A, lane 2, and B, lane 2) and PTH-rP production (Fig. 5B) occur during HA activation of TGF-RI serine/threonine kinases (Fig. 3, A, lane 2, and B, lane 2). As a positive control, we have confirmed that activation of TGF-RI serine/threonine kinases by TGF- also promotes Smad2/Smad3 phosphorylation (Fig. 4, A, lane 4, and B, lane 4) and PTH-rP production (Fig. 5D) in MDA-MB-231 cells. We believe that HA-mediated TGF-RI kinase activation (Fig. 3, A, lane 2, and B, lane 2) leading to Smad2/Smad3 phosphorylation and PTH-rP production is CD44-specific because control samples (either without HA treatment (Fig. 4, A, lane 1, and B, lane 1, and Fig. 5A) or pre-treatment with anti-CD44 followed by HA addition (Fig. 4, A, lane 3, and B, lane 3, and Fig. 5C)) display very low levels of CD44-associated TGF-RI kinase activity (Fig. 3, A, lane 2 and B, lane 2). Consequently, no significant amount of Smad2/Smad3 phosphorylation (Fig. 4, A, lanes 1 and 3, and B, lanes 1 and 3) and PTH-rP (Fig. 5, A and C) is detected under these conditions. It is also noted that no significant reduction of Smad2/Smad3 phosphorylation (Fig. 4, A, lane 5, and B, lane 5) or PTH-rP production (Fig. 5E) occurs in cells pre-treated with anti-CD44 followed by TGF- treatment. Therefore, we believe that these results provide strong evidence that the physiological ligand for CD44v3, HA, plays an important role in activating CD44v3-associated TGF-RI kinase activity required for the onset of Smad2 (or Smad3)-mediated nuclear activities and PTH-rP production during the progression of breast cancers.

View larger version (54K): [in this window] [in a new window]   Fig. 4.   Detection of Smad protein phosphorylation. MDA-MB-231 cells treated with various reagents (e.g. HA (50 µg/ml) or TGF-1 (50 ng/ml) or pre-treated with anti-CD44 antibody followed by HA (50 µg/ml) or TGF-1 (50 ng/ml) treatment or without any treatment) were immunoblotted with anti-phospho-Smad2 (A) or solubilized by 1.0% Nonidet P-40 and immunoprecipitated (IP) with anti-Smad3 (B) followed by anti-phosphothreonine (a), anti-phosphoserine (b), and anti-Smad3 (c)-mediated immunoblot as described under "Materials and Methods." Lane 1, untreated cells; lane 2, cells treated with HA (50 µg/ml); lane 3, cells pre-treated with anti-CD44 antibody followed by HA (50 µg/ml); lane 4, cells treated with TGF-1 (50 ng/ml); lane 5, cells pre-treated with anti-CD44 antibody followed by TGF-1 (50 ng/ml) treatment.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Measurement of PTH-rP production. Breast tumor cells (MDA-MB-231 cells) were washed three times with SF-DMEM and incubated in 3 ml of SF-DMEM containing various reagents (e.g. HA (50 µg/ml) or TGF-1 (50 ng/ml) or anti-CD44 antibody plus HA (50 µg/ml) or TGF-1 (50 ng/ml) treatment or without any treatment) for 24 h at 37 °C in a 5% CO2 humidified chamber. Subsequently, both cells and the conditioned medium will be collected and analyzed for the production of PTH-rP using 125I-labeled anti-PTH-rP antibodies and radioimmunoassay according to the procedures described under "Materials and Methods." Statistical analysis was done using the Student's t test. All data are expressed as the mean ± S.D. A, untreated cells; B, cells treated with HA (50 µg/ml); C, cells pre-treated with anti-CD44 antibody followed by HA (50 µg/ml); D, cells treated with TGF-1 (50 ng/ml); E, cells pre-treated with anti-CD44 antibody followed by TGF-1 (50 ng/ml) treatment.

Effects of TGF- Receptor Kinase-mediated CD44 Phosphorylation on Ankyrin Binding and Tumor Cell Migration-- A number of serine/threonine kinases have been shown to be involved in the regulation of CD44 phosphorylation during HA signaling (11, 50-52). In MDA-MB-231 cells, the level of CD44v3 phosphorylation in the absence of HA treatment is very low (Fig. 6, A, lane 1, and B, lane 1), whereas the amount of CD44v3 phosphorylation increases significantly during HA treatment (Fig. 6, A, lane 2, and B, lane 2) as detected by anti-threonine and anti-serine antibody, respectively. To test whether CD44 functions as a possible cellular substrate(s) of TGF-RI kinases in MDA-MB-231 cells during HA signaling, we have examined the ability of TGFRI kinase to phosphorylate CD44v3. Specifically, we have analyzed the stoichiometry of CD44 phosphorylation by TGF-RI kinase, along with myelin basic protein (MBP) phosphorylation as a positive control (Table I). Our results indicate that approximately ~1.2 mol of phosphate becomes maximally incorporated into 1 mol of CD44v3 using TGF-RI kinase isolated from MDA-MB-231 cells treated with HA (Table I). We have also found that ~1 mol of phosphate becomes maximally incorporated into 1 mol of MBP by HA-activated TGF-RI kinase (Table I). In contrast, phosphorylation of CD44v3 and MBP appears to be minimal (at most ~0.1 mol of phosphate incorporated into per mol of CD44v3 or ~0.15 mol of phosphate incorporated into per mol of MBP) using TGF-RI kinase isolated from MDA-MB-231 cells without any HA treatment (Table I). Because the stoichiometry of CD44 phosphorylation by HA-activated TGF-RI is comparable with that of MBP phosphorylation (by HA-activated TGF-RI), we conclude that CD44v3 is a good cellular substrate for TGF-RI.

View larger version (26K): [in this window] [in a new window]   Fig. 6.   Detection of CD44 phosphorylation. MDA-MB-231 cells treated with various reagents (e.g. HA (50 µg/ml) or TGF-1 (50 ng/ml) or pre-treated with anti-CD44 antibody followed by HA (50 µg/ml) or TGF-1 (50 ng/ml) treatment or without any treatment) were solubilized by 1.0% Nonidet P-40 and immunoprecipitated (IP) with anti-CD44v3 followed by anti-phosphothreonine (A) or anti-phosphoserine (B) as described under "Materials and Methods." Lane 1, untreated cells; lane 2, cells treated with HA (50 µg/ml); lane 3, cells pre-treated with anti-CD44 antibody followed by HA (50 µg/ml); lane 4, cells treated with TGF-1 (50 ng/ml); lane 5, cells pre-treated with anti-CD44 antibody followed by TGF-1 (50 ng/ml) treatment.

View this table: [in this window] [in a new window]   Table I Stoichiometry analysis of CD44v3 phosphorylation by TGF-RI kinase The kinase reaction used in these experiments was the same as described under "Materials and Methods." The amount of [-32P]ATP incorporated into CD44v3 and myelin basic protein (MBP) by TGF-RI kinase (isolated from HA-treated or -untreated cells) was measured as described under "Materials and Methods."

Phosphorylation of the cytoplasmic domain of CD44 has been shown to be important for its interaction with certain cytoskeletal proteins such as ankyrin (11, 14, 15, 50, 53-55). In this study we have examined the effect of TGF-RI kinase-mediated CD44 phosphorylation on ankyrin binding. Specifically, the highly phosphorylated form of CD44v3 (by TGF-RI kinase isolated from HA-activated MDA-MB-231 cells) (as shown in Table I) was incubated with ¹²⁵I-labeled ankyrin. Our results indicate that the total amount of ¹²⁵I-ankyrin binding to the TGF-RI kinase-phosphorylated form of CD44v3 (Fig. 7A) is significantly higher than that unphosphorylated form of CD44v3 (Fig. 7B). These results clearly support the notion that phosphorylation of the cytoplasmic domain of CD44v3 by activated TGFRI kinase enhances its binding interaction with ankyrin. It is likely that this interaction is required for the activation of membrane-associated cytoskeleton function.

View larger version (57K): [in this window] [in a new window]   Fig. 7.   125I-Ankyrin binding to TGF-RI-phosphorylated CD44v3. Purified 125I-labeled ankyrin (~0.35 nM protein, 1.5 × 104 cpm/ng) was incubated with CD44v3 (bound to anti-CD44v3-conjugated beads) (~0.80 µg of protein in TGF-RI-phosphorylated or unphosphorylated form) in 0.5 ml of the binding buffer (20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.1% (w/v) bovine serum albumin, and 0.05% Triton X-100) as described under "Materials and Methods." Following binding, the beads were washed in the binding buffer, and the bead-bound radioactivity was determined. Nonspecific binding was determined in the presence of a 100-fold excess of unlabeled ankyrin. A, the amount of 125I-ankyrin binding to highly phosphorylated CD44v3 (by TGF-RI kinase). B, the amount of 125I-ankyrin binding to minimally phosphorylated CD44v3 (in the absence of TGF-RI kinase).

We have also demonstrated that in the absence of HA a low amount of ankyrin is associated with CD44v3 as analyzed by anti-CD44v3-mediated immunoprecipitation followed by anti-ankyrin immunoblot in MDA-MB-231 cells (Fig. 8, lane 1). HA treatment of cells recruits a significant amount of ankyrin (Fig. 8, lane 2) into a complex with CD44v3 (Fig. 8, lane 2). When cells were pre-treated with anti-CD44 antibody followed by HA treatment, the recruitment of ankyrin into CD44v3 is greatly reduced (Fig. 8, lane 3). These results are consistent with previous findings showing HA is capable of inducing the accumulation of ankyrin into CD44 complexes (42). Interestingly, TGF- is also causing ankyrin recruitment into CD44v3 (Fig. 8, lane 4). No obvious reduction of CD44v3-ankyrin complex formation is observed in MDA-MB-231 cells pre-treated with anti-CD44 antibody followed by TGF- treatment (Fig. 8, lane 5). These results strongly suggest that the TGF- receptor (in particular, TGF-RI) is not only physically complexed with CD44v3 but also functionally coupled to CD44v3-ankyrin-based cytoskeleton functions.

View larger version (14K): [in this window] [in a new window]   Fig. 8.   Analysis of CD44v3-ankyrin complex in human breast tumor cells (MDA-MB-231 cells). Unlabeled MDA-MB-231 cells treated with various reagents (e.g. HA (50 µg/ml) or TGF-1 (50 ng/ml) or pre-treated with anti-CD44 antibody followed by HA (50 µg/ml) or TGF-1 (50 ng/ml) treatment or without any treatment) were solubilized by 1.0% Nonidet P-40 and immunoprecipitated with anti-CD44v3 followed by anti-ankyrin as described under "Materials and Methods." Lane 1, untreated cells; lane 2, cells treated with HA (50 µg/ml); lane 3, cells pre-treated with anti-CD44 antibody followed by HA (50 µg/ml); lane 4, cells treated with TGF-1 (50 ng/ml); lane 5, cells pre-treated with anti-CD44 antibody followed by TGF-1 (50 ng/ml) treatment).

Effects of TGF-RI Overexpression on CD44-Ankyrin Interaction and HA-mediated Breast Tumor Migration-- In order to correlate CD44-TGF-RI kinase signaling with breast tumor cell-specific behaviors (e.g. membrane-cytoskeleton interaction and tumor cell migration), we have transiently transfected the breast tumor cells (MDA-MB-231 cells) with a HA1-tagged TGF-RIcDNA (Fig. 9A) and vector alone (Fig. 9B). By using anti-CD44v3-mediated immunoprecipitation of MDA-MB-231 cells transfected with HA1-tagged TGF-RIcDNA followed by immunoblotting with various antibodies (e.g. anti-CD44, anti-HA1, or anti-ankyrin antibody), we have determined that CD44v3 is expressed at comparable levels in these two transfectants (Fig. 9, A, lane a, and B, lane a) and that only HA1-tagged TGF-RI (Fig. 9B, lane b), but not the vector control sample (Fig. 9A, lane b), is co-precipitated with CD44v3 (Fig. 9B, lane b). By using the same anti-CD44v3-mediated immunoprecipitation procedures, only a low level of ankyrin (Fig. 9A, lane c) was detected in the CD44v3 immunocomplex isolated from cells transfected with vector alone. Overexpression of TGF-RI by transfecting MDA-MB-231 cells with TGF-RIcDNA promotes a significant increase in ankyrin recruitment into CD44v3·TGF-RI complex. These findings suggest that TGF-RI overexpression mimics HA and/or TGF- signaling in the induction of CD44v3-ankyrin complex formation.

View larger version (53K): [in this window] [in a new window]   Fig. 9.   Analysis of the signaling complex formation in MDA-MB-231 cells transfected with TGF-RIcDNA. MDA-MB-231 cells transfected with vector alone (A) or HA1-tagged TGF-RIcDNA (B) were solubilized by Nonidet P-40 (as described above) and immunoprecipitated (IP) with anti-CD44v3 antibody followed by immunoblotting with various immuno-reagents (e.g. anti-CD44 (lane a), anti-HA1 (lane b), or anti-ankyrin (lane c), respectively).

Furthermore, by using in vitro migration assays, we have demonstrated that incubation of untransfected MDA-MB-231 cells with either HA or TGF- stimulates active tumor cell migration (Table II). However, transfection of MDA-MB-231 cells with TGF-RIcDNA also significantly stimulates CD44 cytoskeleton-dependent breast tumor cell migration (Table II) as compared with vector-transfected cells (Table II). Furthermore, treatment of MDA-MB-231 cells with the microfilament inhibitor, cytochalasin D, causes a significant inhibition of HA- and TGF--mediated as well as TGF-RIcDNA-transfected breast tumor cell migration (Table II). Taken together, these findings strongly suggest that both HA and TGF- promote CD44-ankyrin-linked cytoskeleton activation required for metastatic breast tumor cell migration.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CD44 contains a variable extracellular domain, a single spanning 23-amino acid transmembrane domain, and a 70-amino acid cytoplasmic domain (56). Nucleotide sequence analyses reveal that many CD44 isoforms (derived from alternative splicing mechanisms) are variants of the standard form, CD44s (2). CD44 isoforms have been detected on highly metastatic breast tumor cell lines, and transfection of these molecules confers metastatic properties to otherwise non-metastatic cells (9-13, 15, 16, 46). By using CD44-specific antibodies, we have found that metastatic breast tumor cells (e.g. MDA-MB-231 cell line) express several CD44 isoforms including CD44v3 (Fig. 1). The level of CD44v3 isoform expression often increases as the histologic grade of each of the breast tumors progresses. In fact, there is a direct correlation between CD44v3 isoform expression and increased histologic grade of the malignancy (5, 8). These lines of evidence suggest that expression of certain CD44v3 isoform(s) may be an accurate predictor of eventual survival (e.g. nodal status, tumor size, and grade) during breast cancer progression (6). CD44v3 has a heparin sulfate addition site in the membrane-proximal extracellular domain of the molecule that confers the ability to bind heparin sulfate-binding growth factors (58). The attachment of growth factor(s) to the heparin sulfate sites on CD44v3 may be responsible for the onset of breast tumor-associated angiogenesis. In breast tumor cells, CD44v3 is also closely associated with matrix metalloproteinases, MMP-9 (gelatinase B), in the plasma membrane (59). Furthermore, MMP-9 is present in a proteolytically active form and is preferentially localized at the "invadopodia" of the breast tumor cells (59). Therefore, it is likely that the close interaction between CD44v3 and the active form of MMP-9 in the invadopodia structure of breast tumor cells may be required for the degradation of extracellular matrix during breast tumor cell invasion and metastasis.

HA is one of the major components of the extracellular matrix glycosaminoglycan. All CD44 isoforms contain a link module HA-binding site in their extracellular domain (60). Thus, CD44 is considered to be one of the major HA receptors (60). Both CD44 and HA are overexpressed at sites of tumor attachment (48, 49). The binding of HA to CD44 is implicated in the stimulation of a variety of cellular functions including tumor progression (9-13). It is known that CD44 has intricate links to signal transduction processes. In particular, the intracellular domain of CD44 binds to certain cytoskeletal proteins such as ankyrin (11, 14, 15, 50, 53-55) and ERM proteins (ezrin, radixin, and moesin) (61). The transmembrane interaction between CD44 isoforms and ankyrin/ERM provides a direct link between the extracellular matrix and the cytoskeleton. In addition, CD44 couples with tyrosine kinases (e.g. c-Src kinase and p185^HER2 kinase) (12, 13) and serine/threonine kinesis (e.g. protein kinase C and Rho-binding kinase) (11, 50). In cancers, the selective interaction between CD44 and its binding partners has been shown to promote a number of downstream effector functions leading to HA-mediated tumor cell-specific behaviors (1).

TGF- signaling plays a central role in regulating a variety of cellular responses and acts as a growth stimulator or inhibitor, depending on the cellular context (18, 19). It is now generally accepted that TGF- is one of the important regulators in the pathogenesis of human cancers, including breast cancers (33-36, 62, 63). Many late stage or invasive/metastatic breast tumors overexpress TGF- which, perhaps due to autocrine and paracrine effects of TGF-, influences tumor cell growth, invasion, and metastasis (19, 62, 63). Three types of TGF- receptors (e.g. RI, RII, and RIII) belonging to the family of serine/threonine kinase membrane receptors have been identified (18-23, 27, 28). TGF- signaling often involves TGF- binding to TGF-RII which recruits and phosphorylates TGF-RI leading to a series of biological events including phosphorylation of Smad family of proteins (18-26). TGF-RIII, which has no known signaling motif, appears to bind and present TGF- to TGF-RII (27, 28). Alterations of TGF- signaling are generally thought to contribute to the development and progression of human breast cancer (29-32). It has been reported that many TGF-s are secreted in a latent form and are converted to an active form. CD44-associated MMP-9 has been found to be involved in the cleavage of TGF- from a latent form into an active form (64). Therefore, it is clear that a close relationship exists between CD44-associated MMPs and the production of an active form of TGF-. However, the question of whether there is a direct interaction between CD44 and TGF- receptor(s) during breast cancer progression has not been addressed previously.

In this study we have found that both the TGF- type I and II receptors (RI and RII) are expressed in MDA-MB-231 cells (Fig. 1). However, only TGF- type I receptor (RI) (but not type II (RII)) is closely associated with CD44 in metastatic breast tumor cells (MDA-MD-231 cells) (Fig. 1). An in vitro binding assay using two proteins (TGF- type I receptor (TGF-RI) and FLAG-tagged CD44 cytoplasmic domain (FLAG-CD44cyt)) confirms that the TGF-RI is directly involved in the interaction with the cytoplasmic domain of CD44 (Fig. 2). Moreover, HA activates TGF-RI kinase activity (Fig. 3) leading to Smad protein phosphorylation (Fig. 4) in a CD44-dependent manner, whereas TGF--stimulated TGF- kinase activity (Fig. 3) and Smad protein phosphorylation (Fig. 4) do not appear to involve CD44. Thus, HA and TGF- bind to their own specific receptors (e.g. CD44 or TGF- receptors), but their respective downstream signaling pathway(s) appear to be tightly linked in MDA-MB-231 cells. Phosphorylated forms of Smad2 (or Smad3), which often form a complex with Smad4 in the cytosol, have been shown to be translocated into the nucleus for transcriptional activation of many genes (24-26). A recent study (65) demonstrated that human breast carcinoma cells examined by tissue microarrays frequently contain the phosphorylated form of Smad proteins. Therefore, HA and TGF--activated Smad protein phosphorylation may contribute one of the important factors required for the onset of breast cancer progression.

In patients with advanced breast cancers, the malignant cells often metastasize to the bone (66-69). It is now known that TGF- receptor-mediated signaling plays an integral role in stimulating osteolytic bone metastasis by inducing the production of PTH-rP by the tumor cells (66). PTH-rP was originally isolated and cloned from tumors removed from patients with the common paraneoplastic syndrome called humoral hypercalcemia of malignancy (66-69). We have now determined that TGF- receptor-mediated signaling leads to an increase in PTH-rP production by the breast cancer cells (in the MDA-MB-231 cell line) (Fig. 5). The fact that HA activates PTH-rP production in a CD44-specific manner (Fig. 5) suggests that HA-CD44 signaling is also involved in breast tumor-specific hormone production required for breast cancer progression. Other studies have shown that stimulation of PTH-rP by TGF- is regulated through mRNA stabilization (37) or is controlled by a novel Smad3/Ets1 synergism on the P3 promoter of the PTH-rP (70). The question of which mechanism(s) (i.e. mRNA stabilization and/or Smad3-Ets1 interaction) is(are) involved in HA-CD44-mediated PTH-rP expression in the MDA-MB-231 cell awaits future investigation.

In addition, our results indicate that both HA and TGF- are capable of inducing CD44 phosphorylation in vivo (Fig. 6). Moreover, TGF-RI kinase isolated from MDA-MB-231 cells can directly phosphorylate CD44 in vitro (Table I). Biochemical analyses indicate that the stoichiometry of CD44 phosphorylation by TGF-RI isolated from HA-activated MDA-MB-231 cells is comparable with that of MBP phosphorylation (by HA-activated TGF-RI) (Table I). Therefore, we conclude that CD44 may function as one of the cellular substrates for the TGF-RI kinase. Our findings are consistent with previous studies (11, 50-52) showing CD44 can be phosphorylated by several serine-threonine kinases. It is likely that TGF-RI kinase is phosphorylating certain CD44 site(s) such as threonine (amino acids 341, 347, or 351) and serine (amino acids 318, 325, 327, 339, or 356). The identification of the specific phosphorylation site(s) is currently undergoing investigation. Isacke and co-workers (51, 52) have reported that serine (amino acid 325) is the principal CD44 phosphorylation site(s) by serine-threonine kinases and that mutation of this residue blocks CD44-mediated cell migration but not HA binding (51, 52). Thus, the interaction between phosphorylated CD44 and specific intracellular component(s) (e.g. cytoskeletal protein(s)) may be required for cell migration.

Several lines of evidence indicate that the transmembrane interaction between the cytoplasmic domain of CD44 and cytoskeletal proteins (e.g. ankyrin) plays an important role in CD44-mediated oncogenic signaling (15, 16). In particular, the S2 subdomain (but not other subdomains) of the ARD binds to CD44 directly (16); overexpression of the S2 subdomain of ARD promotes CD44-mediated tumor cell migration (16). Ankyrin is also involved in the up-regulation of a Rac1-specific guanine nucleotide (GDP/GTP) exchange factor, Tiam1 (T lymphoma invasion and metastasis), in metastatic breast tumor cell migration (17). In this study we have observed that CD44 phosphorylation by HA-activated TGF-RI kinase stimulates its binding to the cytoskeletal protein ankyrin both in vitro (Fig. 7) and in vivo (Fig. 8). To elucidate further TGF-RI interaction with CD44 and ankyrin in vivo, we have transfected MDA-MB-231 cells with HA1-tagged TGF-RIcDNA (Fig. 9). Our data also confirm that overexpression of the HA1-tagged TGF-RI promotes its association with CD44v3 (Fig. 9) and stimulates recruitment of ankyrin into the CD44v3·TGF-RI complex (Fig. 9) leading to breast tumor cell migration (Table II). Finally, we have found that treatment of MDA-MB-231 cells with cytochalasin D (Table II) induces a reversal of tumor cell-specific phenotypes such as tumor cell migration (Table II). This finding suggests that some actin polymerization or microfilamentous cytoskeleton is required in this event. It is quite possible that the recruitment of ankyrin into CD44v3 induced by either HA/TGF- signaling (Fig. 8) or TGF-RI overexpression (Fig. 9) could contribute to cytoskeleton-mediated breast tumor cell migration (Table II). A previous study (57) has shown that HA is able to bind TGF-1 directly. Therefore, it is possible that the HA·TGF-1 complex in the extracellular matrix also plays a role in stimulating oncogenic signaling and cytoskeletal activation during breast tumor progression

As summarized in Fig. 10, we propose that CD44v3 is tightly complexed with TGF-RI. This CD44v3-associated TGF-RI kinase can be activated by HA and/or TGF- leading to phosphorylation of Smad proteins (Smad2 and Smad 3) and PTH-rP production which is known to cause breast tumor metastasis (in particular, osteolytic bone metastasis). Moreover, HA and/or TGF--activated CD44v3-TGF-RI kinase is also capable of phosphorylating CD44v3. Most importantly, CD44v3 phosphorylation enhances its binding to the cytoskeletal protein ankyrin which, in turn, interacts with the cytoskeleton and induces tumor cell migration. Therefore, we believe that CD44v3-TGF-RI interaction promotes activation of multiple signaling pathways required for ankyrin-membrane interaction, tumor cell migration, and important oncogenic events (e.g. Smad2/Smad3 phosphorylation and PTH-rP production) during HA- and TGF--mediated breast tumor progression.

View larger version (32K): [in this window] [in a new window]   Fig. 10.   A proposed model for the interaction between CD44v3 and TGF- receptor I (RI) during oncogenic signaling and breast tumor progression. CD44v3 (containing the v3 exon-encoded structure) is tightly complexed with TGF-RI. This CD44v3-associated TGF-RI kinase can be activated by HA and/or TGF- leading to phosphorylation of Smad proteins (Smad2 and Smad 3) and PTH-rP production which is known to cause metastasis (e.g. osteolytic bone metastasis). Moreover, HA- and/or TGF--activated CD44v3-TGF-RI kinase is also capable of phosphorylating CD44v3. Most importantly, CD44v3 phosphorylation enhances its binding to the cytoskeletal protein ankyrin which, in turn, interacts with the cytoskeleton and induces tumor cell migration. In conclusion, we believe that CD44v3-TGF-RI interaction plays a pivotal role in the activation of multiple signaling pathways required for ankyrin-membrane interaction, tumor cell migration, and important oncogenic events (e.g. Smad2/Smad3 phosphorylation and PTH-rP production) during HA- and TGF--mediated breast tumor progression.

	ACKNOWLEDGEMENTS

We gratefully acknowledge Dr. Gerard J. Bourguignon for assistance in the preparation of this paper. We also thank Dr. Falko Diedrich for help in data preparation.

	FOOTNOTES

^* This work was supported in part by United States Public Health Service Grants CA66163 and CA78633 and Department of Defense Grant DAMD 17-99-1-9291.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence and reprint requests should be addressed: Endocrine Unit (111N), Dept. of Medicine, University of California, San Francisco, and Veterans Affairs Medical Center, 4150 Clement St., San Francisco, CA 94121. Tel.: 415-221-4810 (Ext. 3321); Fax: 415-383-1638; E-mail:lillyb@itsa.ucsf.edu.

^§ Supported by an American Heart Association predoctoral fellowship.

Published, JBC Papers in Press, July 26, 2002, DOI 10.1074/jbc.M204320200

	ABBREVIATIONS

The abbreviations used are: HA, hyaluronan; TGF-, transforming growth factor ; PTH, parathyroid hormone; PTH-rP, parathyroid hormone-related protein; DMEM, Dulbecco's modified Eagle's medium; SF, serum-free; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid; MMP, matrix metalloproteinase; ARD, ankyrin repeat domain.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				J. Webber, R. H. Jenkins, S. Meran, A. Phillips, and R. Steadman Modulation of TGF{beta}1-Dependent Myofibroblast Differentiation by Hyaluronan Am. J. Pathol., July 1, 2009; 175(1): 148 - 160. [Abstract] [Full Text] [PDF]

				L.-J. Hsu, L. Schultz, Q. Hong, K. Van Moer, J. Heath, M.-Y. Li, F.-J. Lai, S.-R. Lin, M.-H. Lee, C.-P. Lo, et al. Transforming Growth Factor {beta}1 Signaling via Interaction with Cell Surface Hyal-2 and Recruitment of WWOX/WOX1 J. Biol. Chem., June 5, 2009; 284(23): 16049 - 16059. [Abstract] [Full Text] [PDF]

				D. D. Luo, C. Fielding, A. Phillips, and D. Fraser Interleukin-1 beta regulates proximal tubular cell transforming growth factor beta-1 signalling Nephrol. Dial. Transplant., May 6, 2009; (2009) gfp208v1. [Abstract] [Full Text] [PDF]

				J. Webber, S. Meran, R. Steadman, and A. Phillips Hyaluronan Orchestrates Transforming Growth Factor-{beta}1-dependent Maintenance of Myofibroblast Phenotype J. Biol. Chem., April 3, 2009; 284(14): 9083 - 9092. [Abstract] [Full Text] [PDF]

				L. Y. W. Bourguignon, W. Xia, and G. Wong Hyaluronan-mediated CD44 Interaction with p300 and SIRT1 Regulates {beta}-Catenin Signaling and NF{kappa}B-specific Transcription Activity Leading to MDR1 and Bcl-xL Gene Expression and Chemoresistance in Breast Tumor Cells J. Biol. Chem., January 30, 2009; 284(5): 2657 - 2671. [Abstract] [Full Text] [PDF]

				L. Y. W. Bourguignon, K. Peyrollier, W. Xia, and E. Gilad Hyaluronan-CD44 Interaction Activates Stem Cell Marker Nanog, Stat-3-mediated MDR1 Gene Expression, and Ankyrin-regulated Multidrug Efflux in Breast and Ovarian Tumor Cells J. Biol. Chem., June 20, 2008; 283(25): 17635 - 17651. [Abstract] [Full Text] [PDF]

				S. Meran, D. W. Thomas, P. Stephens, S. Enoch, J. Martin, R. Steadman, and A. O. Phillips Hyaluronan Facilitates Transforming Growth Factor-{beta}1-mediated Fibroblast Proliferation J. Biol. Chem., March 7, 2008; 283(10): 6530 - 6545. [Abstract] [Full Text] [PDF]

				P. Huebener, T. Abou-Khamis, P. Zymek, M. Bujak, X. Ying, K. Chatila, S. Haudek, G. Thakker, and N. G. Frangogiannis CD44 Is Critically Involved in Infarct Healing by Regulating the Inflammatory and Fibrotic Response J. Immunol., February 15, 2008; 180(4): 2625 - 2633. [Abstract] [Full Text] [PDF]

				A. Ouhtit, Z. Y. Abd Elmageed, M. E. Abdraboh, T. F. Lioe, and M. H.G. Raj In Vivo Evidence for the Role of CD44s in Promoting Breast Cancer Metastasis to the Liver Am. J. Pathol., December 1, 2007; 171(6): 2033 - 2039. [Abstract] [Full Text] [PDF]

				S. Meran, D. Thomas, P. Stephens, J. Martin, T. Bowen, A. Phillips, and R. Steadman Involvement of Hyaluronan in Regulation of Fibroblast Phenotype J. Biol. Chem., August 31, 2007; 282(35): 25687 - 25697. [Abstract] [Full Text] [PDF]

				L. Y. W. Bourguignon, E. Gilad, and K. Peyrollier Heregulin-mediated ErbB2-ERK Signaling Activates Hyaluronan Synthases Leading to CD44-dependent Ovarian Tumor Cell Growth and Migration J. Biol. Chem., July 6, 2007; 282(27): 19426 - 19441. [Abstract] [Full Text] [PDF]

				S. R. Hamilton, S. F. Fard, F. F. Paiwand, C. Tolg, M. Veiseh, C. Wang, J. B. McCarthy, M. J. Bissell, J. Koropatnick, and E. A. Turley The Hyaluronan Receptors CD44 and Rhamm (CD168) Form Complexes with ERK1,2 That Sustain High Basal Motility in Breast Cancer Cells J. Biol. Chem., June 1, 2007; 282(22): 16667 - 16680. [Abstract] [Full Text] [PDF]

				L. Y. W. Bourguignon, K. Peyrollier, E. Gilad, and A. Brightman Hyaluronan-CD44 Interaction with Neural Wiskott-Aldrich Syndrome Protein (N-WASP) Promotes Actin Polymerization and ErbB2 Activation Leading to beta-Catenin Nuclear Translocation, Transcriptional Up-regulation, and Cell Migration in Ovarian Tumor Cells J. Biol. Chem., January 12, 2007; 282(2): 1265 - 1280. [Abstract] [Full Text] [PDF]

				C. Tolg, S. R. Hamilton, K.-A. Nakrieko, F. Kooshesh, P. Walton, J. B. McCarthy, M. J. Bissell, and E. A. Turley Rhamm-/- fibroblasts are defective in CD44-mediated ERK1,2 motogenic signaling, leading to defective skin wound repair J. Cell Biol., December 18, 2006; 175(6): 1017 - 1028. [Abstract] [Full Text] [PDF]

				L. Li, C.-H. Heldin, and P. Heldin Inhibition of Platelet-derived Growth Factor-BB-induced Receptor Activation and Fibroblast Migration by Hyaluronan Activation of CD44 J. Biol. Chem., September 8, 2006; 281(36): 26512 - 26519. [Abstract] [Full Text] [PDF]

				S. Ohno, H.-J. Im, C. B. Knudson, and W. Knudson Hyaluronan Oligosaccharides Induce Matrix Metalloproteinase 13 via Transcriptional Activation of NF{kappa}B and p38 MAP Kinase in Articular Chondrocytes J. Biol. Chem., June 30, 2006; 281(26): 17952 - 17960. [Abstract] [Full Text] [PDF]

				L. Y. W. Bourguignon, E. Gilad, A. Brightman, F. Diedrich, and P. Singleton Hyaluronan-CD44 Interaction with Leukemia-associated RhoGEF and Epidermal Growth Factor Receptor Promotes Rho/Ras Co-activation, Phospholipase C{epsilon}-Ca2+ Signaling, and Cytoskeleton Modification in Head and Neck Squamous Cell Carcinoma Cells J. Biol. Chem., May 19, 2006; 281(20): 14026 - 14040. [Abstract] [Full Text] [PDF]

				K. M.A. Rouschop, N. Claessen, S. T. Pals, J. J. Weening, and S. Florquin CD44 Disruption Prevents Degeneration of the Capillary Network in Obstructive Nephropathy via Reduction of TGF-beta1-Induced Apoptosis J. Am. Soc. Nephrol., March 1, 2006; 17(3): 746 - 753. [Abstract] [Full Text] [PDF]

				P. A. Knepper, A. M. Miller, J. Choi, R. D. Wertz, M. J. Nolan, W. Goossens, S. Whitmer, B. Y. J. T. Yue, R. Ritch, J. M. Liebmann, et al. Hypophosphorylation of Aqueous Humor sCD44 and Primary Open-Angle Glaucoma Invest. Ophthalmol. Vis. Sci., August 1, 2005; 46(8): 2829 - 2837. [Abstract] [Full Text] [PDF]

				S. Misra, S. Ghatak, and B. P. Toole Regulation of MDR1 Expression and Drug Resistance by a Positive Feedback Loop Involving Hyaluronan, Phosphoinositide 3-Kinase, and ErbB2 J. Biol. Chem., May 27, 2005; 280(21): 20310 - 20315. [Abstract] [Full Text] [PDF]

				L. Y. W. Bourguignon, E. Gilad, K. Rothman, and K. Peyrollier Hyaluronan-CD44 Interaction with IQGAP1 Promotes Cdc42 and ERK Signaling, Leading to Actin Binding, Elk-1/Estrogen Receptor Transcriptional Activation, and Ovarian Cancer Progression J. Biol. Chem., March 25, 2005; 280(12): 11961 - 11972. [Abstract] [Full Text] [PDF]

				J. Choi, A. M. Miller, M. J. Nolan, B. Y. J. T. Yue, S. T. Thotz, A. F. Clark, N. Agarwal, and P. A. Knepper Soluble CD44 Is Cytotoxic to Trabecular Meshwork and Retinal Ganglion Cells In Vitro Invest. Ophthalmol. Vis. Sci., January 1, 2005; 46(1): 214 - 222. [Abstract] [Full Text] [PDF]

				L. Y. W. Bourguignon, P. A. Singleton, F. Diedrich, R. Stern, and E. Gilad CD44 Interaction with Na+-H+ Exchanger (NHE1) Creates Acidic Microenvironments Leading to Hyaluronidase-2 and Cathepsin B Activation and Breast Tumor Cell Invasion J. Biol. Chem., June 25, 2004; 279(26): 26991 - 27007. [Abstract] [Full Text] [PDF]

				T. Ito, J. D. Williams, D. J. Fraser, and A. O. Phillips Hyaluronan Regulates Transforming Growth Factor-{beta}1 Receptor Compartmentalization J. Biol. Chem., June 11, 2004; 279(24): 25326 - 25332. [Abstract] [Full Text] [PDF]

				T. Ito, J. D. Williams, D. Fraser, and A. O. Phillips Hyaluronan Attenuates Transforming Growth Factor-{beta}1-Mediated Signaling in Renal Proximal Tubular Epithelial Cells Am. J. Pathol., June 1, 2004; 164(6): 1979 - 1988. [Abstract] [Full Text] [PDF]

				K. M.A. Rouschop, M. E. Sewnath, N. Claessen, J. J.T.H. Roelofs, I. Hoedemaeker, R. van der Neut, J. Aten, S. T. Pals, J. J. Weening, and S. Florquin CD44 Deficiency Increases Tubular Damage But Reduces Renal Fibrosis in Obstructive Nephropathy J. Am. Soc. Nephrol., March 1, 2004; 15(3): 674 - 686. [Abstract] [Full Text] [PDF]

				L. Y. W. Bourguignon, P. A. Singleton, H. Zhu, and F. Diedrich Hyaluronan-mediated CD44 Interaction with RhoGEF and Rho Kinase Promotes Grb2-associated Binder-1 Phosphorylation and Phosphatidylinositol 3-Kinase Signaling Leading to Cytokine (Macrophage-Colony Stimulating Factor) Production and Breast Tumor Progression J. Biol. Chem., August 8, 2003; 278(32): 29420 - 29434. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/42/39703    most recent M204320200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bourguignon, L. Y. W. Articles by Zhou, B. Search for Related Content PubMed PubMed Citation Articles by Bourguignon, L. Y. W. Articles by Zhou, B. Social Bookmarking                      What's this?