J Biol Chem, Vol. 273, Issue 45, 29524-29529, November 6, 1998

Identification of the Region of -Catenin That Plays an Essential Role in Cadherin-mediated Cell Adhesion^*

Masayuki Ozawa

From the Department of Biochemistry, Faculty of Medicine, Kagoshima University, Kagoshima 890-8520, Japan

	ABSTRACT

Top Abstract Introduction Procedures Results Discussion References

-Catenin is an intrinsic component of the cadherin adhesion complex and is a 102-kDa protein with multiple interaction sites, including homodimerization sites, and binding sites for - and -catenin (plakoglobin), -actinin, and actin. Besides the binding to - or -catenin, it is unknown, however, which interaction is critical for the function of cadherins. By expressing a series of E-cadherin--catenin chimeric molecules on leukemia cells (K562), we have identified the region of -catenin that confers aggregation inducing activity to nonfunctional tail-less E-cadherin. The region has been mapped to the carboxyl-terminal 295 amino acids of -catenin. Consistent with this result, expression in -catenin-deficient cells (DLD-1/) of a mutant -catenin molecule consisting of the amino-terminal -/-catenin-binding site and the carboxyl-terminal cell adhesion region identified in the above experiments induced E-cadherin-mediated cell aggregation and compaction. Cells expressing E-cadherin chimeric molecules with the homologous carboxyl-terminal region of vinculin, which contains the actin-binding site of vinculin, did not, however, aggregate as strongly as ones expressing E-cadherin--catenin chimeric molecules.

	INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

Cadherins comprise a family of Ca²⁺-dependent transmembrane molecules that play essential roles in the initiation and stabilization of cell-cell contacts (1, 2). The extracellular domain of cadherins is responsible for specific homophilic binding (3), and the conserved cytoplasmic domain facilitates adhesion through binding to intracellular proteins, termed catenins (4-6). Each cadherin molecule can bind to either -catenin or -catenin (plakoglobin), which in turn binds to -catenin (7-10). Cadherins cannot mediate strong cell-cell adhesion in cells that lack -catenin (11). PC9 cells that lack -catenin because of deletion of a part of the -catenin gene exhibit cadherin-dependent adhesion upon introduction of -catenin, identifying the latter as an indispensable molecule for the cadherin adhesion complex to be functional (12). In addition, reintroduction of -catenin into the same cell line has been shown to induce a polarized phenotype typical of epithelial cells and to alter the growth rate (13).

-Catenin is a 102-kDa multifunctional protein with multiple interaction sites, including amino-terminal -/-catenin-binding site (14-18) and homodimerization sites (18), a central region for -actinin binding (15), and amino-terminal as well as the carboxyl-terminal actin-binding sites (19). -Catenin also binds to ZO-1, a 220-kDa actin-binding protein found at tight junctions in epithelial cells and at the cadherin-based adhesion sites in non-epithelial cells (20). The role of these interactions, except for the binding to -/-catenin, in cadherin-mediated adhesion is, however, unknown.

-Catenin exhibits sequence similarity to vinculin (21, 22), a highly conserved 117-kDa cytoskeletal protein found in both cell-cell and cell-extracellular matrix adherens-type junctions (23-25). In such junctions, vinculin is thought to be one of a number of interacting proteins which link the cytoplasmic face of adhesion receptors of the cadherin or integrin family to the actin cytoskeleton. Similarity between -catenin and vinculin is restricted to three regions in their amino-terminal, central, and carboxyl-terminal regions, that for the latter being the highest. The amino-terminal region of vinculin contains a tailin-binding site (26), whereas its carboxyl-terminal region contains a binding site for actin (27, 28). Recently, it was shown that vinculin associates with E-cadherin complexes via -catenin (29).

While the discovery and characterization of catenins provided a major insight into the molecular interactions of cadherins, it is possible that the actual junctional complexes contain a multitude of additional proteins. To examine the homophilic adhesive properties of cadherins, we recently developed a model system which involves the transfection of K562 leukemia cells with the cDNA of interest and analysis of the adhesive properties of the resulting transfectants (30). The expression of functional cadherin changes non-adhesive cells into cells that grow as aggregates. In those studies we transfected cells with an expression vector encoding an E-cadherin--catenin chimeric molecule consisting of (a) the entire extracellular and transmembrane domains of E-cadherin as well as the first 80 amino acids of its cytoplasmic domain, excluding the region shown to associate with - or -catenin (6), and (b) amino acids 301-906 of -catenin, which include the domains necessary for association with -actinin and actin (15, 19), but not the domain essential for association with -catenin and -catenin (14-18). The cell clones expressing the chimeric protein on their surface were found to form aggregates in an E-cadherin-dependent manner. These findings provided us with an opportunity to identify the minimum region of -catenin, besides the -/-catenin-binding site, required for its function in cadherin-mediated cell adhesion. In this study we present evidence that the carboxyl-terminal region of -catenin (residues 612-906) is enough to trigger the adhesive activity of E-cadherin provided it is covalently linked to E-cadherin or associated with the E-cadherin adhesion complex through its interaction with -/-catenin via its amino-terminal -/-catenin-binding site.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

Construction of Expression Vectors-- The expression vector for the wild-type E-cadherin or an E-cadherin--catenin chimeric protein (EMC)¹ was as described previously (30). The cDNA encoding an E-cadherin mutant protein, EC71 (5), was cloned into the same expression vector, pCAGGS neo (31) (a gift from Dr. K. Yamamura, Kumamoto University). A full-length cDNA clone for human -catenin was described previously (32). For the expression of E-cadherin--catenin chimeric proteins, the ClaI-ClaI fragment of E-cadherin cDNA that encodes the 71 amino acids including the catenin-binding domain of E-cadherin was replaced with the following cDNA fragments encoding various regions of -catenin generated using convenient restriction enzyme sites within the cDNA clones or by means of the polymerase chain reaction using Pwo DNA polymerase (Boehringer) (Fig. 1, A and C). The combinations of restriction enzymes used were: BglII and StuI (for construction of EN), Eco47III and ClaI (EM), ClaI and ClaI (EC), ClaI and PmaCI (ECC1), ClaI and PstI (ECC4), ClaI and HindIII (ECC5), SmaI and EcoRV (ECN2), BalI and EcoRV (ECN3), BalI and EcoRV (ECN4), and BamHI and BamHI (ECN5). For the polymerase chain reaction, three combinations of sense and antisense primers (5'-CCCATCGATACCCCTGAGGAGTTG and 5'-CCCATCGATTAGATGCTGTCCATAGC) (for construction of ECN1), (5'-GAGTTTATCGATGCTTCCCGC and 5'-CCCATCGATTAACCCTGTGACTTTTG) (ECC2), and (5'-GAGTTTATCGATGCTTCCCGC and 5'-CCCATCGATTACCCAGAGACAACAAG) (ECC3), containing a ClaI recognition sequence at the 5'-end were used. Expression vectors for E-cadherin-vinculin chimeric molecules were constructed in the same way using cDNA for chicken vinculin (25) (a kind gift from Dr. B. Geiger, The Weizmann Institute of Science). The restriction enzymes used were: ClaI and ClaI (for construction of EVC1), and Sau3AI and ClaI (EVC2). In the latter case, the reading frame was adjusted using an oligonucleotide, ATCGAT. To construct the expression vector for an E-cadherin chimera with the full-length vinculin, a ClaI recognition sequence was introduced into the vinculin cDNA at positions 6 to 1 of the initiation codon. In the case of the carboxyl-terminal truncation, the termination codon was introduced using either oligonucleotide ATCGGCTACCCCTACGACGTCCCCGACTACGCCGGCGTCTAGATCAAGCTTATCG (for construction of EN), ATATCGGCTACCCCTACGACGTCCCCGACTACGCCGGCGTCTAGATCAAGCTTATCG (EM), GCTTAATTAATTAAGC (ECC1 and ECC4), or GTGA (ECC5). The wild-type and mutant -catenin polypeptides were expressed using the same expression vector. cDNA encoding a mutant -catenin polypeptide with a deletion in the carboxyl-terminal one-third of the amino-terminal region and the entire middle portion (N'M) was constructed by deleting a 1231-base pair BalI-ClaI fragment. cDNA encoding another mutant -catenin polypeptide with a further deletion in the carboxyl-terminal region (N'M-2) was constructed by replacing a 1460-base pair BalI-BalI fragment with an oligonucleotide, GGGGG.

Cells and Transfection-- Human leukemia K562 cells (kindly provided by Dr. K. Sekiguchi, Research Institute, Osaka Medical Center for Maternal and Child Health) were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum. K562 cells expressing wild-type E-cadherin (EK cells) or an E-cadherin--catenin chimeric protein (EMCK cells) were as described previously (30). K562 cells (5 × 10⁶) were transfected with the expression vectors (10 µg) by electroporation using a Bio-Rad Gene Pulser set at 280 V and 960 microfarads. Human colon cancer DLD-1/ cells, DLD-1 cells deficient in -catenin expression, were kindly provided by Dr. S. T. Suzuki (Institute for Developmental Research, Aichi Human Service Center), and were grown as described above. Transfection of DLD-1/ cells with the expression vectors was carried out by the calcium phosphate method as described previously (5).

Antibodies-- Monoclonal antibodies against -, -, and -catenin were purchased from Transduction Laboratories. DECMA-1, a monoclonal antibody to E-cadherin (33), was kindly provided by Dr. R. Kemler (Max-Planck-Institut für Immunbiologie). HECD-1, a monoclonal antibody to human E-cadherin, was purchased from Takara Shuzou Co.

Immunoblotting and Immunoprecipitation-- For immunoblot analysis, cells (1 × 10⁵) were boiled for 5 min in Laemmli SDS gel sample buffer, run on 8% polyacrylamide gels, and then electroblotted onto nitrocellulose membranes. The membranes were blocked with 5% nonfat milk in phosphate-buffered saline, and then incubated with monoclonal antibodies and finally peroxidase-conjugated antibodies (Jackson ImmunoResearch Laboratories). After washing with the buffer containing 0.1% Tween 20, the protein bands were visualized with an ECL detection kit (Amersham). To detect E-cadherin chimeric molecules in the detergent-insoluble fraction of cells, cells (1 × 10⁶) were lysed in 0.5 ml of 10 mM Tris-HCl buffer, pH 7.6, containing 1% Triton X-100, 0.5% Nonidet P-40, 150 mM NaCl, 1 mM CaCl₂, 0.1 mM sodium orthovanadate, 1 mM phenylmethylsufonyl fluoride, 10 µg/ml leupeptin, and 25 µg/ml aprotinin. After centrifugation, the precipitates were boiled for 5 min in the SDS gel sample buffer and then subjected to immunoblot analysis. Immunoprecipitation was carried out as described previously (30) with the following modifications. The E-cadherin-catenin complex was collected using HECD-1 anti-E-cadherin monoclonal antibodies which had been preabsorbed to protein G-Sepharose 4B (Sigma).

Cell Aggregation-- The cell aggregation assay was performed as described previously (6) except that the K562 cell transfectants were passed through Pasteur pipettes several times to obtain single cells. The transfected DLD-1/ cells were dissociated with 0.01% trypsin in HEPES-buffered saline containing 2 mM CaCl₂. After the incubation, the cells were fixed by the addition of an equal volume of 6% formaldehyde in phosphate-buffered saline. Immunofluorescence staining was performed as described previously (5) using DECMA-1 and fluorescein isothiocyanate-labeled anti-rat IgG.

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

Expression of E-Cadherin--Catenin Chimeric Molecules on K562 Cells-- To be fully functional in cell-cell adhesion, E-cadherin is believed to become associated with catenins through its carboxyl-terminal region (Fig. 1A). Either -catenin or -catenin associates directly with the cadherin; -catenin binds to -catenin/-catenin. A carboxyl terminus-truncated cadherin (EC71) cannot associate with catenins and therefore is nonfunctional, i.e. cells expressing this protein cannot form aggregates. By covalently linking the amino-terminal half or the carboxyl-terminal half of the -catenin polypeptide to the nonfunctional cadherin, it has been shown that the carboxyl-terminal half of -catenin has the ability to restore the adhesive activity of nonfunctional E-cadherin (34). On expressing, in a leukemia cell line (K562), a similar E-cadherin chimeric molecule containing the carboxyl-terminal two-thirds of -catenin, we observed that, like E-cadherin expressing K562 cells, cells expressing this chimeric E-cadherin not only formed aggregates but also showed compaction (30). To identify the region of -catenin that confers aggregation and compaction inducing activities to the nonfunctional tail-less E-cadherin, I constructed a series of cDNA encoding E-cadherin chimeric molecules containing various regions of -catenin (Fig. 1C), and expressed them on K562 cells. K562 cells grow as non-adhesive single cells, with no endogenous cadherin. They are more advantageous than other types of cells, such as L cells, because the aggregates formed by K562 cells expressing E-cadherin can be dissociated into single cells by passage several times through Pasteur pipettes without the use of trypsinization in the presence of Ca²⁺.

View larger version (40K): [in this window] [in a new window]   Fig. 1.   Structures of E-cadherin chimeric proteins. A, schematic representation of the E-cadherin-catenin complex (top), the nonfunctional mutant E-cadherin polypeptide lacking the carboxyl-terminal catenin-binding domain (middle), and the E-cadherin chimeric molecule covalently linked with the polypeptide sequences to be analyzed (bottom). B, comparison of -catenin and vinculin. Homologous regions are indicated by shaded boxes together with the percentage identity. C and D, the regions of -catenin (C) or vinculin (D) used for the construction of E-cadherin chimeric proteins. The numbers refer to amino acid positions in -catenin or vinculin. The nomenclature for each construct is given on the right.

Cells stably expressing these molecules were selected and examined by SDS-polyacrylamide gel electrophoresis followed by immunoblotting with an anti-E-cadherin antibody, DECMA-1 (Fig. 2A). The E-cadherin--catenin chimeric molecules migrated as polypeptides of the sizes expected from their constructs. The chimeric protein levels did not differ by more than 20% among the K562 cell lines expressing the different E-cadherin--catenin chimeric molecules except for ENK cells and EMK cells; in these cells the protein levels decreased during the culture. The expression of these two chimeric proteins seemed to be unstable because two types of cells, one positive and the other negative for DECMA-1 staining, were present even after recloning of the cells.

View larger version (44K): [in this window] [in a new window]   Fig. 2.   Immunoblot analysis of E-cadherin chimeric molecules. A, immunoblot detection of E-cadherin chimeric molecules. K562 cells expressing wild type E-cadherin, nonfunctional E-cadherin (EC71), or E-cadherin chimeric proteins covalently linked with the regions of -catenin or vinculin shown in Fig. 1 were lysed in SDS sample buffer and then subjected to immunoblot analysis with an E-cadherin monoclonal antibody, DECMA-1. Small amounts of high molecular weight materials found for some chimeric constructs represent intracellular proprotein forms, because they were not digested with 0.01% trypsin treatment in the presence of 1 mM EGTA. The major bands of each chimeric protein were expressed on the cell surface because they were degraded on the same digestion but not in the presence of 2 mM Ca2+. B, immunoblot detection of E-cadherin chimeras associated with the detergent-insoluble cytoskeletal fraction of K562 cells. K562 cells expressing E-cadherin chimeras were lysed with detergents as described under "Experimental Procedures." The detergent-insoluble fractions were subjected to immunoblot analysis with DECMA-1. EVK cells are K562 cells expressing an E-cadherin chimera with the full-length vinculin.

Aggregation of Cells Expressing E-Cadherin--Catenin Chimeric Molecules-- To quantitatively compare the cell adhesion activities of different E-cadherin--catenin chimeric molecules, and that of wild-type E-cadherin, cells expressing the chimeric proteins together with cells expressing wild-type E-cadherin (EK cells) or nonfunctional E-cadherin (EC71K cells) were subjected to cell aggregation assaying (Fig. 3). As described previously (30), EK cells expressing the intact form of E-cadherin aggregated in an E-cadherin-dependent manner; i.e. it was inhibited by the presence of the E-cadherin antibody, DECMA-1. EC71K cells expressing nonfunctional E-cadherin did not aggregate as extensively as EK cells. Among the cells expressing different E-cadherin--catenin chimeric proteins, cells expressing the chimeric molecules, EMC (EMCK cells) and EC (ECK cells), were able to aggregate to a similar extent to EK cells (Fig. 3), and showed a morphological change, so-called compaction (data not shown). ECC1K cells and ECN1K cells aggregated, however, reproducibly to a lower extent as compared with EMCK cells and ECK cells. In the case of ECC1K cells, the size of the aggregates formed was significantly smaller (less than half size in diameter) than in the cases of the other cells, such as EMCK and ECK cells (data not shown). Cells transfected with the other constructs, ENK cells, EMK cells, ECC2K cells, ECC3K cells, ECC4K cells, ECC5K cells, ECN2K cells, ECN3K cells, ECN4K cells, and ECN5K cells, did not aggregate significantly under the same conditions (Fig. 3). As mentioned above, cultures of ENK or EMK cells contained two populations of cells, one strongly positive and the other negative as to the expression of E-cadherin chimeras. Despite the strong expression of the E-cadherin chimeras on their surface, these cells did not aggregate under the conditions used. Therefore, it seemed that the region of -catenin that confers aggregation and compaction inducing activities to nonfunctional tail-less E-cadherin is localized to residues 612-906. Like the aggregation of EK cells, the aggregation of K562 cells expressing the different E-cadherin--catenin chimeric molecules was inhibited in the presence of the E-cadherin antibody (Fig. 3). The aggregation of these cells is also Ca²⁺-dependent, since no aggregation was observed in the presence of 5 mM EGTA (data not shown).

View larger version (44K): [in this window] [in a new window]   Fig. 3.   Aggregation of K562 cells expressing E-cadherin chimeric molecules. K562 cells expressing the wild-type E-cadherin (EK cells), nonfunctional E-cadherin, i.e. EC71 (EC71K cells), with a deletion in the cytoplasmic domain, or E-cadherin chimeric polypeptides with various regions of -catenin or vinculin shown in Fig. 1, or K562 cells transfected with the control vector (nK cells) were allowed to aggregate for 30 min in the absence (filled boxes) or presence (open boxes) of an anti-E-cadherin monoclonal antibody, DECMA-1. EVK cells are K562 cells expressing an E-cadherin chimera with the full-length vinculin.

The detergent insolubility of cadherin has been shown to be an indication of complex association with the actin cytoskeleton (6, 35). This association is a prerequisite for the cell adhesive activity of cadherins (4, 6, 12). Therefore, we assessed the solubility of the E-cadherin--catenin chimeric molecules in buffer containing 1% Triton X-100 and 0.5% Nonidet P-40. Approximately 90% of the wild-type E-cadherin expressed in K562 cells was solubilized in the buffer (data not shown). In EK cells, however, a fraction (~10%) of E-cadherin was detected in the detergent-insoluble fraction, whereas there was almost no mutant tail-less E-cadherin (EC71) (Fig. 2B). A similar amount (approximately 10%) of E-cadherin--catenin chimeric molecules containing the carboxyl-terminal two-thirds of -catenin (EMC) was also detected in the insoluble fraction. Of the E-cadherin--catenin chimeric molecules containing either the amino-terminal domain (EN), middle part (EM), or carboxyl-terminal domain (EC) of -catenin, only EC was detected in the insoluble fraction (Fig. 2B). The progressive deletions from the carboxyl terminus of the carboxyl-terminal domain significantly reduced the amounts of the chimeric proteins recovered in the insoluble fraction (ECC1 and ECC2), and further deletions completely prevented the interaction with the actin cytoskeleton (ECC3, ECC4, and ECC5). The three deletions from the amino-terminal side of the carboxyl-terminal domain did not affect the interaction with the actin cytoskeleton (ECN1, ECN2, and ECN3), but further deletions resulted in a lack of association with the actin cytoskeleton (ECN4 and ECN5). From these results it seemed that the region of the -catenin polypeptide required for the interaction with the actin cytoskeleton resides in residues 689-906. This region is much smaller than the region that confers aggregation and compaction inducing activities to nonfunctional tail-less E-cadherin (residues 612-906), but it corresponds to the carboxyl-terminal region of -catenin showing the highest degree of homology with vinculin (Fig. 1B). These results suggested that the anchorage to the actin cytoskeleton is not enough to rescue the nonfunctional E-cadherin and that the carboxyl-terminal region of -catenin plays an additional role other than binding to the actin cytoskeleton.

The Carboxyl-terminal Region of Vinculin Cannot Substitute for the Role of -Catenin-- The carboxyl-terminal region of -catenin having the ability to rescue nonfunctional cadherin as an adhesion molecule when covalently attached to the nonfunctional E-cadherin contains the region showing the highest degree (34%) of homology with vinculin (Fig. 1B). Therefore, we next examined whether or not an analogous region of vinculin can rescue the nonfunctional E-cadherin through covalently linking to the latter protein. Two cDNAs encoding two E-cadherin-vinculin chimeric molecules were constructed (Fig. 1D). These constructs contained the carboxyl-terminal 350 (amino acid residues 717-1066) or 257 (residues 810-1066) amino acids of vinculin, respectively, thus both included the actin-binding domain (27, 28). Cells stably expressing the respective chimera proteins (EVC1K cells and EVC2K cells) were isolated and examined by SDS-polyacrylamide gel electrophoresis, followed by immunoblotting with DECMA-1 (Fig. 2A). The E-cadherin-vinculin chimeric molecules (EVC1 and EVC2) migrated as polypeptides of the sizes expected from their constructs. As expected from their known ability to bind to actin filaments, ~10% of these chimeric polypeptides were recovered in the detergent-insoluble cytoskeletal fraction (Fig. 2B). The cells expressing these chimeras did not, however, aggregate to the same extent as EK cells or ECK cells did under the conditions used (Fig. 3). Furthermore, an E-cadherin chimera with the full-length vinculin (EV) expressed on K562 cells (Fig. 2A) could not support E-cadherin-mediated adhesion (Fig. 3), although ~50% of the chimera was partitioned in the detergent-insoluble fraction (Fig. 2B). Thus, the results seemed to support the idea that the carboxyl-terminal region of -catenin plays an additional role other than binding to the actin cytoskeleton.

Expression of a Mutant -Catenin Molecule Consisting of the Amino-terminal -/-Catenin-binding Site and the Carboxyl-terminal Region Induced E-cadherin-mediated Cell Aggregation and Compaction in -Catenin-deficient Cells-- In a normal situation, -catenin becomes associated with the cadherin-adhesion complex through binding to -catenin or -catenin in the complex. Its amino-terminal binding site for -catenin or -catenin is responsible for the binding. It is therefore of importance to determine whether or not the -catenin region identified in the above experiments can activate the E-cadherin adhesion complex when connected to either -catenin or -catenin via its own binding site. For this, we constructed a cDNA encoding a mutant -catenin polypeptide consisting of the amino-terminal -/-catenin-binding site and the carboxyl-terminal 295 amino acids (N'M) by deleting the sequence encoding the carboxyl-terminal one-third of the amino-terminal region and the entire middle part (amino acids 203-611) (Fig. 4A). As a control, another truncated -catenin (N'M-2) that lacks amino acids 203-688 was also constructed. A human cell line (DLD1/ cells) that did not express endogenous -catenin was transfected with the expression vector containing cDNA for either wild-type -catenin or the mutant -catenin polypeptides, 1/3NM or 1/3NM-2, or the control neo vector.

View larger version (47K): [in this window] [in a new window]   Fig. 4.   Expression of mutant -catenin polypeptides in -catenin-deficient cells. A, schematic representation of -catenin (top), a mutant -catenin polypeptide with a deletion in the carboxyl-terminal one-third of the amino-terminal region and the entire middle portion (N'M, middle), and a mutant -catenin polypeptide with a further deletion in the carboxyl-terminal region (N'M-2, bottom). B, immunoblot detection of -catenin polypeptides. DLD1/ cells expressing wild-type -catenin (D cells) or the mutant -catenin polypeptides (N'MD cells or N'M-2D cells), or DLD1/ cells transfected with the control vector (nD) were lysed in the SDS sample buffer and then subjected to analysis with an -catenin monoclonal antibody. C, immunoprecipitation analysis of the E-cadherin adhesion complex. Cells labeled overnight with [35S]methionine were lysed and then subjected to immunoprecipitation with an E-cadherin antibody, HECD-1. The immunoprecipitates were analyzed by SDS-polyacrylamide gel electrophoresis and fluorography. The positions of E-cadherin, -catenin, -catenin, and the mutant -catenin polypeptides are indicated on the right.

Cells expressing these proteins were cloned and examined by SDS-polyacrylamide gel electrophoresis and immunoblotting with an anti--catenin monoclonal antibody (Fig. 4B). The mutant proteins migrated as a polypeptide of 54 and 46 kDa, respectively, the expected molecular mass of the proteins. Association of the mutant -catenin proteins with the E-cadherin adhesion complex was studied by means of co-immunoprecipitation experiments. Cells were metabolically labeled with [³⁵S]methionine, and E-cadherin was immunoprecipitated from cell lysates with an E-cadherin antibody, HECD-1. A protein migrating to positions corresponding to 88 kDa was coprecipitated with E-cadherin (120 kDa) in the case of DLD1/ cells transfected with the control vector (nD cells) (Fig. 4C). The coprecipitated protein was identified as -catenin by subjecting the immunoprecipitates to immunoblot analysis with -catenin antibodies (data not shown). In the case of DLD1/ cells expressing wild-type -catenin (D cells), a protein of 102 kDa was also coprecipitated in addition to these proteins. The same analysis of DLD1/ cells expressing mutant -catenin polypeptides, i.e. N'M or N'M-2 (N'MD cells or N'M-2D cells), revealed that polypeptides of 54 or 46 kDa, respectively, which correspond to the size of the mutant -catenin proteins, were coprecipitated together with E-cadherin and -catenin (Fig. 4C). These protein bands were identified as the wild-type and mutant -catenin polypeptides by immunoblot analysis of the immunoprecipitates with the -catenin antibody (data not shown).

These cells were dissociated with 0.01% trypsin in the presence of 2 mM Ca²⁺, and then subjected to the aggregation assay (Fig. 5A). Although DLD1/ cells expressing N'MD-2 protein (N'M-2D cells) and DLD1/ cells transfected with the control neo vector (nD cells) as well as parental DLD1/ cells (not shown) showed a low degree of aggregation, cells expressing either wild-type -catenin (D cells) or N'M protein (N'MD cells) showed a significantly enhanced level of aggregation (Fig. 5A). The cadherin-mediated cell aggregation is accompanied by a morphological change, so-called compaction. The aggregates of D cells and N'MD cells showed extensive compaction (Fig. 5B), whereas the aggregates of N'M-2D cells and nD cells did not show such a morphological change, and each cell in the aggregates was easily distinguishable (Fig. 5B).

View larger version (74K): [in this window] [in a new window]   Fig. 5.   Aggregation of DLD1/ cells expressing wild-type -catenin or mutant -catenin polypeptides. A, DLD1/ cells expressing wild-type -catenin (D cells) or mutant -catenin polypeptides (N'MD cells or N'M-2D cells), or DLD1/ cells transfected with the control vector (nD cells) were allowed to aggregate for 30 min in the presence of Ca2+ (filled boxes) or EGTA (open boxes). B, compaction, a morphological change, is induced in D cells or N'MD cells, but not in N'M-2D cells or nD cells. Cells were allowed to aggregate for 30 min in the presence of Ca2+. Bar, 50 µm.

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

To be fully functional in cell-cell adhesion, the cadherin molecule is believed to become associated with the actin cytoskeleton via cytoplasmic catenins. Through binding to - or -catenin, -catenin becomes associated with the cadherin adhesion complex. Although -catenin itself is an actin-binding protein (19), it has been shown that -catenin interacts with at least two actin-binding proteins, i.e. -actinin (15) and ZO-1 (20), both proteins being co-precipitated with and colocalized with E-cadherin (20, 36). Because the amino acid sequence of the carboxyl-terminal domain of -catenin is similar to that of the self-association domain of vinculin, it has been suggested that vinculin may interact with -catenin at adhesion sites (22). Therefore, it is possible that the molecular linkage between cadherin and the actin cytoskeleton may include multiple actin-binding proteins.

By expressing a series of E-cadherin--catenin chimeric molecules on leukemia cells (K562) that have no endogenous cadherin, we have identified the region of -catenin that confers aggregation and compaction-inducing activities to nonfunctional tail-less E-cadherin. The region has been mapped to the carboxyl-terminal 295 amino acids (amino acids 612-906) of -catenin. Consistent with this result, expression in -catenin-deficient cells (DLD1/) of a mutant -catenin molecule consisting of the amino-terminal -/-catenin-binding site and the carboxyl-terminal cell adhesion region identified in the above experiments induced E-cadherin-mediated cell aggregation and compaction. This region seems to contain two functional domains, a carboxyl-terminal one (amino acid residues 689-906) involved in the association with the actin cytoskeleton, and an amino-terminal one (amino acid residues 612-688) with an unknown function. Both of the domains are required for the cell adhesion activity. Although carboxyl-terminal amino acid residues 689-906 of -catenin seem to be sufficient to anchor the chimeric molecule to the actin cytoskeleton, they were not able to restore the adhesive activity to nonfunctional tail-less E-cadherin. Together with the observation that cells expressing the E-cadherin chimeric molecules covalently linked with the carboxyl-terminal region of vinculin, which contains the actin-binding site of vinculin, do not aggregate to the same extent as cells expressing the E-cadherin chimeric molecule with the carboxyl-terminal region of -catenin, these results seem to suggest that the simple linkage of the cadherin molecules to the actin cytoskeleton is not enough to activate the nonfunctional E-cadherin, and that the carboxyl-terminal region of -catenin could play a role besides actin binding.

The role of the region including amino acid residues 612-688 of -catenin remains to be determined in future experiments. This region does not, however, correspond to the -actinin-binding site. Amino acid residues 325-394 of -catenin have been shown to be sufficient for the interaction with -actinin (15). Our results suggest that the interaction of -catenin with -actinin is not essential for the cadherin-mediated cell adhesion in the cells used in the present study. Although the binding site on -catenin for ZO-1 has not been identified, ZO-1 seems not to be expressed in K562 cells (37). Therefore, the E-cadherin-mediated cell adhesion analyzed in the present study takes place in the absence of ZO-1. Thus, it is less likely that this region is the site for ZO-1-binding.

Recently, gene trap screening of mice revealed a fusion between the amino-terminal 632 amino acids of -catenin and the -geo reporter. Embryos homozygous for this mutant allele were shown to exhibit deficits in cell adhesion resulting in embryonic lethality (38). Overexpression of an -catenin mutant lacking the carboxyl-terminal 230 amino acids in Xenopus embryos causes severe developmental defects that reflect impaired Ca²⁺-dependent blastomere adhesion (39). These observations suggested the importance of the carboxyl-terminal region of -catenin in cadherin-based cell adhesion. Although in these studies why the carboxyl-terminal deletion of -catenin resulted in a deficiency of cadherin-mediated cell adhesion was not determined, our finding provides the molecular basis for these observations.

	ACKNOWLEDGEMENTS

We thank Drs. Rolf Kemler, Shintaro T. Suzuki, Kiyotoshi Sekiguchi, Ken-ichi Yamamura, Noriyuki Kioka, and Benjamin Geiger for providing reagents, and Kumiko Sato for secretarial assistance.

	FOOTNOTES

^* This work was supported by grants from the Ministry of Education, Science and Culture of Japan, the Naito Foundation for the Promotion of Science, and the Kodama Memorial Foundation.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 should be addressed: Dept. of Biochemistry, Faculty of Medicine, Kagoshima University, Kagoshima 890-8520, Japan. Tel.: 81-99-275-5246; Fax: 81-99-264-5618; E-mail: mozawa{at}med2.kufm.kagoshima-u.ac.jp.

The abbreviations used are: EMC, E-cadherin--catenin chimeric protein; EK cells, K562 cells expressing E-cadherin.

	REFERENCES

Top Abstract Introduction Procedures Results Discussion References

1. Takeichi, M. (1991) Science 251, 1451-1455[Abstract/Free Full Text]
2. Gumbiner, B. M. (1996) Cell 84, 345-357[CrossRef][Medline] [Order article via Infotrieve]
3. Nose, A., Tsuji, K., and Takeichi, M. (1990) Cell 61, 147-155[CrossRef][Medline] [Order article via Infotrieve]
4. Nagafuchi, A., and Takeichi, M. (1988) EMBO J. 7, 3679-3684[Medline] [Order article via Infotrieve]
5. Ozawa, M., Baribault, H., and Kemler, R. (1989) EMBO J. 8, 1711-1717[Medline] [Order article via Infotrieve]
6. Ozawa, M., Ringwald, M., and Kemler, R. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 4246-4250[Abstract/Free Full Text]
7. Aberle, H., Butz, S., Stappert, J., Weissig, H., Kemler, R., and Hoschuetzky, H. (1994) J. Cell Sci. 107, 3655-3663[Abstract]
8. Hülsken, J., Birchmeier, W., and Behrens, J. (1994) J. Cell Biol. 127, 2061-2069[Abstract/Free Full Text]
9. Sacco, P. A., McGranahan, T. M., Wheelock, M. J., and Johnson, K. R. (1995) J. Biol. Chem. 270, 20201-20206[Abstract/Free Full Text]
10. Jou, T.-S., Stewart, D. B., Stappert, J., Nelson, W. J., and Marrs, J. A. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 5067-5071[Abstract/Free Full Text]
11. Shimoyama, Y., Nagafuchi, A., Fujita, S., Gotoh, M., Takeichi, M., Tsukita, S., and Hirohashi, S. (1992) Cancer Res. 52, 5770-5774[Abstract/Free Full Text]
12. Hirano, S., Kimoto, N., Shimoyama, Y., Hirohashi, S., and Takeichi, M. (1992) Cell 70, 293-301[CrossRef][Medline] [Order article via Infotrieve]
13. Watabe, M., Nagafuchi, A., Tsukita, S., and Takeichi, M. (1994) J. Cell Biol. 127, 247-256[Abstract/Free Full Text]
14. Obama, H., and Ozawa, M. (1997) J. Biol. Chem. 272, 11017-11020[Abstract/Free Full Text]
15. Nieset, J. E., Redfield, A. R., Jin, F., Knudsen, K. A., Johnson, K. R., and Wheelock, M. J. (1997) J. Cell Sci. 110, 1013-1022[Abstract]
16. Huber, O., Krohn, M., and Kemler, R. (1997) J. Cell Sci. 110, 1759-1765[Abstract]
17. Bullions, L. C., Notterman, D. A., Chung, L. S., and Levine, A. J. (1997) Mol. Cell. Biol. 17, 4501-4508[Abstract]
18. Koslov, E. R., Maupin, P., Pradhan, D., Morrow, J. S., and Rimm, D. L. (1997) J. Biol. Chem. 272, 27301-27306[Abstract/Free Full Text]
19. Rimm, D. L., Koslov, E. R., Kebriaei, P., Cianci, C. D., and Morrow, J. S. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 8813-8817[Abstract/Free Full Text]
20. Itoh, M., Nagafuchi, A., Moroi, S., and Tsukita, S. (1997) J. Cell Biol. 138, 181-192[Abstract/Free Full Text]
21. Herrenknecht, K., Ozawa, M., Eckerskorn, C., Lottspeich, F., Lenter, M., and Kemler, R. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 9156-9160[Abstract/Free Full Text]
22. Nagafuchi, A., Takeichi, M., and Tsukita, S. (1991) Cell 65, 849-857[CrossRef][Medline] [Order article via Infotrieve]
23. Geiger, B., Tokuyasu, K. T., Dutton, A. H., and Singer, S. J. (1980) Proc. Natl. Acad. Sci. U. S. A. 77, 4127-4131[Abstract/Free Full Text]
24. Coutu, M. D., and Craig, S. W. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 8535-8539[Abstract/Free Full Text]
25. Price, G. L., Jones, P., Davison, M. D., Patel, B., Bendori, R., Geiger, B., and Critchley, D. R. (1989) Biochem. J. 259, 453-461[Medline] [Order article via Infotrieve]
26. Gilmore, A. P., Jackson, P., Waites, G. T., and Critchley, D. R. (1992) J. Cell Sci. 103, 719-731[Abstract]
27. Menkel, A. R., Kroemker, M., Bubeck, P., Ronsiek, M., Nikolai, G., and Jockusch, B. M. (1994) J. Cell Biol. 126, 1231-1240[Abstract/Free Full Text]
28. Johnson, R. P., and Craig, S. W. (1995) Nature 373, 261-264[CrossRef][Medline] [Order article via Infotrieve]
29. Hazan, R. B., Kang, L., Roe, S., Borgen, P. I., and Rimm, D. L. (1997) J. Biol. Chem. 272, 32448-32453[Abstract/Free Full Text]
30. Ozawa, M., and Kemler, R. (1998) J. Biol. Chem. 273, 6166-6170[Abstract/Free Full Text]
31. Niwa, H., Yamamura, K., and Miyazaki, J. (1991) Gene (Amst.) 108, 193-200[CrossRef][Medline] [Order article via Infotrieve]
32. Ozawa, M., Terada, H., and Pedraza, C. (1995) J. Biochem. (Tokyo) 118, 1077-1082[Abstract/Free Full Text]
33. Vestweber, D., and Kemler, R. (1985) EMBO J. 13, 3393-3398
34. Nagafuchi, A., Ishihara, S., and Tsukita, S. (1994) J. Cell Biol. 127, 235-245[Abstract/Free Full Text]
35. Hirano, S., Nose, A., Hatta, K., Kawakami, A., and Takeichi, M. (1987) J. Cell Biol. 105, 2501-2510[Abstract/Free Full Text]
36. Knudsen, K. A., Soler, A. P., Johnson, K. R., and Wheelock, M. J. (1995) J. Cell Biol. 130, 67-77[Abstract/Free Full Text]
37. Stevenson, B. R., Siliciano, J. D., Mooseker, M. S., and Goodenough, D. A. (1986) J. Cell Biol. 103, 755-766[Abstract/Free Full Text]
38. Torres, M., Stoykova, A., Huber, O., Chowdhury, K., Bonaldo, P., Mansouri, A., Butz, S., Kemler, R., and Gruss, P. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 901-906[Abstract/Free Full Text]
39. Sehgal, R. N. M., Gumbiner, B. M., and Reichardt, L. F. (1997) J. Cell Biol. 139, 1033-1046[Abstract/Free Full Text]