¹Department of Molecular Biology and Biochemistry, Osaka University Graduate School of Medicine/Faculty of Medicine, Suita 565-0871, Japan

²Department of Geriatric Medicine, Faculty of Medicine, Kyoto University, Kyoto 606-8501, Japan

Correspondence to: Y Takai, E-mail:ytakai@molbio.med.osaka-u.ac.jp

Junctional adhesion molecule (JAM) is a Ca²⁺-independent immunoglobulin-like cell-cell adhesion molecule which localizes at tight junctions (TJs). Claudin is a key cell-cell adhesion molecule that forms TJ strands at TJs. JAM is associated with claudin through their cytoplasmic tail-binding protein, ZO-1. JAM is furthermore associated with Par-3, a cell polarity protein which forms a ternary complex with Par-6 and atypical protein kinase C. Nectin is another Ca²⁺-independent immunoglobulin-like cell-cell adhesion molecule which localizes at adherens junctions (AJs). Nectin is associated with E-cadherin through their respective cytoplasmic tail-binding proteins, afadin and catenins, and involved in the formation of AJs cooperatively with E-cadherin. We show here that nectin is furthermore involved in the localization of JAM at TJs. During the formation of the junctional complex consisting of AJs and TJs in Madin-Darby canine kidney (MDCK) cells, JAM was recruited to the nectin-based cell-cell adhesion sites. This recruitment of JAM was inhibited by nectin inhibitors, which inhibited the trans-interaction of nectin. Microbeads coated with the extracellular fragment of nectin, that interacted with cellular nectin, also recruited JAM to the bead-MDCK cell contact sites. Furthermore, when cadherin-deficient L fibroblasts stably expressing both exogenous JAM and nectin (nectin-JAM-L cells) were co-cultured with L fibroblasts expressing only nectin (nectin-L cells), JAM was concentrated at the cell-cell adhesion sites between nectin-JAM-L and nectin-L cells without the trans-interaction of JAM. Analyses of the localization and immunoprecipitation of JAM revealed that it was associated with nectin through afadin and ZO-1. These results suggest that nectin has a role in the localization of JAM at TJs in the process of the formation of the junctional complex in epithelial cells.

Oncogene (2002) 21, 7642-7655. doi:10.1038/sj.onc. 1205875

nectin; JAM; cadherin; adherens junctions; tight junctions

Introduction

Junctional adhesion molecule (JAM) is a Ca²⁺-independent immunoglobulin-like cell-cell adhesion molecule which localizes at tight junctions (TJs) in epithelial cells (Martin-Padura et al., 1998; Bazzoni et al., 2000; Liu et al., 2000; Palmeri et al., 2000; Itoh et al., 2001). At TJs, claudin is a key Ca²⁺-independent cell-cell adhesion molecule that forms TJ strands (Furuse et al., 1998a,b; Tsukita and Furuse, 1999; Tsukita et al., 1999). Occludin is another transmembrane protein at TJs, but its function has not been established (Tsukita and Furuse, 1999; Tsukita et al., 1999). Claudin and occludin interact with actin-filament (F-actin)-binding scaffold molecules, ZO-1, -2, and -3 (Itoh et al., 1993, 1997, 1999a,b; Willott et al., 1993; Furuse et al., 1994; Haskins et al., 1998; Wittchen et al., 2000). JAM also interacts with ZO-1 (Bazzoni et al., 2000; Ebnet et al., 2000; Itoh et al., 2001). ZO-1 and -2 form a dimer with ZO-3 (Haskins et al., 1998; Itoh et al., 1999b; Wittchen et al., 1999). Furthermore, JAM directly interacts with one of the cell polarity proteins, Par-3, which forms a ternary complex with Par-6 and atypical protein kinase C (aPKC), suggesting that JAM is involved in the formation of cell polarity through these proteins (Ebnet et al., 2001; Itoh et al., 2001). JAM modulates monocyte transmigration (Martin-Padura et al., 1998), and serves as a ligand of the ₂ integrin LFA-1 involved in transendothelial migration of leukocytes (Ostermann et al., 2002). JAM furthermore serves as a receptor for reovirus (Barton et al., 2001). Two other JAM homologs, VE-JAM/JAM-2 and JAM-3, have been identified (Cunningham et al., 2000; Palmeri et al., 2000; Arrate et al., 2001; Aurrand-Lions et al., 2001). VE-JAM/JAM-2 and JAM-3 are predominantly expressed in endothelial cells of different vascular beds and are absent in epithelial cells (Palmeri et al., 2000; Aurrand-Lions et al., 2001).

The formation and maintenance of TJs are dependent on the cell-cell adhesion activity of E-cadherin in epithelial cells (Gumbiner and Simons, 1986; Gumbiner et al., 1988; Pasdar and Nelson, 1988a,b; Watabe-Uchida et al., 1998). E-Cadherin is a key Ca²⁺-dependent cell-cell adhesion molecule at adherens junctions (AJs) which localize just at the basal side of TJs (Takeichi, 1988, 1991; Geiger and Ginsberg, 1991; Gumbiner, 1996, 2000). E-Cadherin is associated with the actin cytoskeleton through peripheral membrane proteins, including -, -, and -catenins, -actinin, and vinculin (Ozawa et al., 1989; Nagafuchi et al., 1991; Tsukita et al., 1992). Several experiments in which E-cadherin function is inhibited have revealed that E-cadherin is important for the assembly of the junctional complex in epithelial cells: incubation of epithelial cells in a low Ca²⁺ medium or in a medium containing E-cadherin function-blocking antibodies (Abs) prevents assembly of AJs and TJs (Gumbiner and Simons, 1986; Gumbiner et al., 1988; Pasdar and Nelson, 1988a,b; Siliciano and Goodenough, 1988); and the junctional complex does not assemble in the cells genetically deficient of -catenin (Watabe et al., 1994). However, the mechanism of the organization of the junctional complex remains unknown.

Nectin is another immunoglobulin-like Ca²⁺-independent cell-cell adhesion molecule at AJs (Aoki et al., 1994; Lopez et al., 1995; Takahashi et al., 1999; Miyahara et al., 2000; Satoh-Horikawa et al., 2000). Nectin comprises a family consisting of at least four members, nectins-1, -2, -3, and -4 (Morrison and Racaniello, 1992; Aoki et al., 1994; Lopez et al., 1995; Eberle et al., 1995; Cocchi et al., 1998; Warner et al., 1998; Satoh-Horikawa et al., 2000; Reymond et al., 2001). Nectins-1, -2, and -3 have two or three splicing variants. Nectin-1 has originally been identified as the poliovirus receptor-related protein and has recently been shown to serve as the -herpes virus entry and cell-cell spread mediator (Lopez et al., 1995; Eberle et al., 1995; Cocchi et al., 1998, 2000; Geraghty et al., 1998; Warner et al., 1998; Lopez et al., 2000; Sakisaka et al., 2001). Each member of the nectin family forms homo-cis-dimers, followed by formation of homo-trans-dimers, causing cell-cell adhesion (Lopez et al., 1998; Takahashi et al., 1999; Miyahara et al., 2000; Satoh-Horikawa et al., 2000; Sakisaka et al., 2001). Nectin-3 furthermore forms hetero-trans-dimers with either nectin-1 or -2 and this formation of each hetero-trans-dimer is much stronger than that of each homo-trans-dimer (Satoh-Horikawa et al., 2000). Nectin-4 also forms hetero-trans-dimers with nectin-1 (Reymond et al., 2001). Nectin is associated with the actin cytoskeleton through afadin, an F-actin-binding protein (Mandai et al., 1997, 1999; Takahashi et al., 1999). Most of the nectin family members have a C-terminal conserved motif of four amino acid residues which interacts with the PDZ domain of afadin (Takahashi et al., 1999; Satoh-Horikawa et al., 2000; Reymond et al., 2001). Afadin has at least two splicing variants, l- and s-afadins. l-Afadin, the larger splicing variant, is an F-actin-binding protein with one PDZ domain and three proline-rich domains and connects nectin to the actin cytoskeleton (Mandai et al., 1997; Takahashi et al., 1999). s-Afadin, the smaller splicing variant, has one PDZ domain but lacks the F-actin-binding domain and the third proline-rich domain (Mandai et al., 1997). Human s-afadin is identical to the gene product of AF-6, a gene that has been identified as an ALL-1 fusion partner involved in acute myeloid leukemias (Prasad et al., 1993). In this study, l-afadin is simply referred to as afadin. The nectin-afadin system is expressed not only in cells having TJs but also cells lacking TJs, such as fibroblasts, neurons, and spermatids (Lopez et al., 1995; Eberle et al., 1995; Mandai et al., 1997; Takahashi et al., 1999; Satoh-Horikawa et al., 2000; Nishioka et al., 2000; Mizoguchi et al., 2002; Ozaki-Kuroda et al., 2002).

Nectin has a potency to recruit the E-cadherin--catenin complex to the nectin-based cell-cell adhesion sites through afadin and -catenin in fibroblasts (Tachibana et al., 2000). Nectin has furthermore a potency to recruit ZO-1 to the nectin-based cell-cell adhesion sites through afadin in a cadherin-independent manner in fibroblasts (Yokoyama et al., 2001). In epithelial cells of afadin (-/-) mice and (-/-) embryoid bodies, the proper organization of AJs and TJs is impaired (Ikeda et al., 1999). Nectin-1 has recently been determined by positional cloning to be responsible for cleft lip/palate-ectodermal dysplasia, which is characterized by cleft lip/palate, syndactyly, mental retardation, and ectodermal dysplasia (Suzuki et al., 2000). These results suggest that the nectin-afadin system plays an important role in the organization of both AJs and TJs, but the direct evidence for this role of the nectin-afadin system has not yet been obtained.

In this study, we have examined the possible interaction between nectin and JAM by the use of Madin-Darby canine kidney (MDCK) cells and cadherin-deficient L cells stably expressing nectin and JAM, and have found that nectin is involved in the localization of JAM at TJs in epithelial cells.

Results

Remaining of nectin-1-based cell-cell adhesion in the absence of the cadherin-based cell-cell adhesion

We examined the localization of JAM in comparison with that of nectin-1 or E-cadherin during the disruption of cell-cell AJs in MDCK cells stably expressing FLAG-nectin-1 (nectin-1-MDCK cells). When the Ca²⁺ concentration in the culture medium is switched from 2 mM to 2 M, the cells gradually detached from each other as described in wild-type MDCK cells (Kartenbeck et al., 1991). Wild-type MDCK cells expressed nectin-1 and -2 as estimated by Western blotting, but none of them were stained by any currently available Abs (data not shown). Thus, we utilized nectin-1-MDCK cells in this study. The staining patterns for afadin, E-cadherin, and JAM in wild-type MDCK cells cultured at 2 mM Ca²⁺ were similar to those in nectin-1-MDCK cells (data not shown). The immunofluorescence signals for nectin-1, E-cadherin, and JAM were highly concentrated at cell-cell adhesion sites of nectin-1-MDCK cells cultured at 2 mM Ca²⁺ as described previously (Takahashi et al., 1999; Asakura et al., 1999) (Figure 1; 0 min). When these cells were cultured at 2 M Ca²⁺ for 15, 30, 60, and 120 min, the immunofluorescence signals for E-cadherin and JAM gradually disappeared and partly appeared on intracellular vesicles (Figure 1; 15, 30, 60, 120 min). The signals were first observed just beneath the plasma membrane and gradually translocated to the perinuclear regions. The signal for E-cadherin first disappeared and the signal for JAM then disappeared (Figure 1; 15, 30 min). In contrast, the signal for nectin-1 mostly remained on the plasma membrane and formed a ring-like structure. The extracellular domain of nectin-1 was exposed to the outside of the cells which had been cultured at 2 M Ca²⁺ for 120 min, as estimated by immunostaining of non-permeabilizing cells with the anti-FLAG monoclonal antibody (mAb) which recognized the extracellular domain of FLAG-nectin-1 (data not shown).

The disappearance of the immunofluorescence signals for E-cadherin and JAM from the plasma membrane might be due to their internalization, their degradation, or to their diffusion on the plasma membrane. To examine which is the case, the homogenate was prepared from nectin-1-MDCK cells cultured at 2 M Ca²⁺ for indicated periods of time and subjected to SDS-polyacrylamide gel electrophoresis (PAGE), followed by Western blotting with an Ab against each protein. The total amount of nectin-1, E-cadherin, or JAM was not changed (Figure 2). The post-nuclear supernatant was then prepared from the cells precultured at 2 M Ca²⁺ for 120 min and subjected to sucrose density gradient ultracentrifugation, followed by Western blotting of each fraction with an Ab against each protein. Nectin-1, E-cadherin, and JAM were partly detected in the intracellular vesicle fraction where early or late endosomes were enriched, but mostly detected in the plasma membrane fraction even after the cells were cultured at 2 M Ca²⁺ for 120 min (data not shown). These results indicate that the disappearance of the immunofluorescence signals for E-cadherin and JAM from the plasma membrane was mainly due to the lateral diffusion of these proteins on the plasma membrane and partly due to their internalization.

We examined in detail whether the nectin-1-based cell-cell adhesion remained after the cadherin-based cell-cell adhesion was disrupted by culturing nectin-1-MDCK cells at 2 M Ca²⁺ for 120 min (Figure 3). The immunofluorescence signal for nectin-1 was diffusely observed along the free surface of the plasma membrane at the apical side, but it was concentrated at the cell-cell adhesion sites at the basal side of the cells (Figure 3; arrow). The signals for afadin and ZO-1 colocalized with that for nectin-1 not only at the free surface of the plasma membrane but also at the cell-cell adhesion sites in the cells which had been cultured at 2 M Ca²⁺ for 120 min (Figure 3). These results indicate that nectin-1 forms cell-cell adhesion even in the absence of the cadherin-based cell-cell adhesion, and afadin and ZO-1 still remain at the nectin-1-based cell-cell adhesion sites.

Recruitment of JAM to the nectin-1-based cell-cell adhesion sites by increase of Ca²⁺ from a low to a high concentration

Nectin-1-MDCK cells precultured at 2 M Ca²⁺ for 120 min were recultured at 2 mM Ca²⁺ for 15, 30, and 60 min, and the time courses of the accumulation of E-cadherin and JAM at the nectin-1-based cell-cell adhesion sites were examined in detail. Before the incubation at 2 mM Ca²⁺, the immunofluorescence signal for E-cadherin or JAM was not observed at the nectin-1-based cell-cell adhesion sites where the signals for nectin-1, afadin, and ZO-1 were observed (Figure 4A-C; 0 min). After the incubation at 2 mM Ca²⁺, the signals for E-cadherin and JAM were concentrated at the nectin-1-based cell-cell adhesion sites (Figure 4A; 30 min, arrow and arrowhead). After 60 min of the incubation at 2 mM Ca²⁺, the signals for nectin-1, E-cadherin, JAM, afadin, and ZO-1 were concentrated at the cell-cell adhesion sites (Figure 4A-C; 60 min). The signals for both JAM and ZO-1 were more concentrated at the apical side than at the basal side, whereas the signals for nectin-1 and E-cadherin were more concentrated at the basal side than at the apical side, indicating that the signals for both JAM and ZO-1 were localized at the apical side of the nectin-1-based cell-cell adhesion sites (Figure 4A-C; 60 min).

To obtain the evidence that the nectin-afadin system is necessary for the recruitment of JAM, we utilized the inhibitors of nectin-1, the chimeric proteins of extracellular fragments of glycoprotein D and nectin-3 fused with human IgG Fc, gD and Nef-3. Glycoprotein D is an envelope protein of herpes simplex virus type 1, one of -herpes viruses (Campadelli-Fiume et al., 2000; Spear et al., 2000). gD binds to nectin-1 and inhibits not only the formation of homo-trans-dimers of nectin-1 but also the formation of hetero-trans-dimers between nectin-1 and -3 (Sakisaka et al., 2001). We have recently found that Nef-3 has also similar activity (Honda et al., submitted). As the mixture of gD and Nef-3 has more inhibitory activity than gD or Nef-3 alone in nectin-1-MDCK cells (data not shown), we utilized the mixture of gD and Nef-3 as nectin inhibitors in this study. Nectin-1-MDCK cells were first cultured at 2 M Ca²⁺ for 120 min and then cultured at 2 mM Ca²⁺ in the presence or absence of the mixture of gD and Nef-3. In the absence of gD or Nef-3, the immunofluorescence signals for nectin-1, E-cadherin, and JAM were detected at 93, 78, and 79% of the cell-cell adhesion sites, respectively (n=105) (Figure 5; control and see Figure 4A; 60 min). However, in the presence of the mixture of gD and Nef-3, the immunofluorescence signal for nectin-1 was reduced to 20% of the cell-cell adhesion sites (n=164) (Figure 5; inhibitors, Nectin-1). The signal for E-cadherin at the cell-cell adhesion sites was reduced to 36% of the cell-cell adhesion sites (n=164) in the presence of the mixture of gD and Nef-3 (Figure 5; inhibitors, E-Cad). The signal for JAM at the cell-cell adhesion sites was also reduced to 31% of the cell-cell adhesion sites (n=164) in the presence of the mixture of gD and Nef-3 (Figure 5; inhibitors, JAM). When the cells were incubated at 2 mM Ca²⁺ for a longer time, the immunofluorescence signals for nectin-1, E-cadherin, and JAM were re-concentrated at the cell-cell adhesion sites even in the presence of the mixture of gD and Nef-3 (data not shown). These results indicate that the mixture of gD and Nef-3 inhibits the formation of the trans-dimers of nectin-1 and thereby inhibits the recruitment of E-cadherin and JAM to the nectin-1-based cell-cell adhesion sites.

Phorbol ester-induced recruitment of JAM to thenectin-1-based cell-cell adhesion sites

It has been shown that when nectin-1-MDCK cells precultured at 2 M Ca²⁺ for 120 min is incubated with a phorbol ester, 12-o-tetradecanoyl-phorbol-13-acetate (TPA), a TJ-like structure is formed, and that nectin-1, afadin, ZO-1, but not E-cadherin, accumulate there (Balda et al., 1993; Asakura et al., 1999). We next examined whether JAM accumulates at the TPA-induced TJ-like structure in nectin-1-MDCK cells. The immunofluorescence signal for JAM as well as that for nectin-1, but not that for E-cadherin, accumulated at the TPA-induced TJ-like structure (Figure 6; 30 min, 60 min). The sites of recruited JAM were apparently indistinguishable from the nectin-1-based cell-cell adhesion sites by confocal microscopy, because the cells became flattened by the action of TPA and the cell-cell adhesion sites were not thick.

Then we examined whether this TPA-induced recruitment of JAM to the nectin-1-based cell-cell adhesion sites was inhibited by the mixture of gD and Nef-3. In the absence of gD or Nef-3, the immunofluorescence signals for nectin-1 and JAM were detected at 96 and 89% of the cell-cell adhesion sites, respectively (n=46) (Figure 7; control and see Figure 6; 60 min). However, in the presence of the mixture of gD and Nef-3, the immunofluorescence signal for nectin-1 was reduced to 19% of the cell-cell adhesion sites (n=32) (Figure 7; inhibitors, Nectin-1). The immunofluorescence signal for JAM was also reduced to 31% of the cell-cell adhesion sites (n=32) (Figure 7; inhibitors, JAM). These results indicate that the mixture of gD and Nef-3 inhibits the formation of the trans-dimers of nectin-1 and thereby inhibits the recruitment of JAM to the TPA-induced TJ-like structure.

Recruitment of JAM to the Nef-3-coated bead-cell contact sites

To obtain more definitive evidence that nectin-1 recruits JAM to the nectin-1-based cell-cell adhesion sites, we utilized microbeads coated with Nef-3. We have recently shown that the microbeads coated with Nef-3 recruit first the nectin-afadin complex and then the E-cadherin-catenin complex to the bead-cell contact sites using L fibroblasts stably expressing both E-cadherin and nectin-1 (Honda et al., submitted). Nectin-1-MDCK cells were cultured at 2 M Ca²⁺ for 120 min in the presence of the Nef-3-coated beads, and further incubated for various periods of time at 2 mM Ca²⁺ (Figure 8). The cells were then fixed, followed by immunostaining for cellular nectin-1, afadin, ZO-1, and JAM. The immunofluorescence signals for cellular nectin-1 and ZO-1 were concentrated at the bead-cell contact sites at 60 min (Figure 8). The signals for afadin and JAM were also concentrated at the bead-cell contact sites (Figure 8). The time courses of the concentration of the signals of nectin-1, ZO-1, and afadin at the bead-cell contact sites were similar, but the signal for JAM at the bead-cell contact sites was not observed at 60 min, but started to be concentrated at 180 min and gradually increased (data not shown). The signal for nectin-1, ZO-1, afadin, or JAM at the bead-cell contact sites was not observed when the control microbeads coated with human IgG were used (Figure 8 and data not shown). These results suggest that Nef-3 on the beads first forms trans-dimers with cellular nectin-1, which might form a complex with afadin and ZO-1, and that JAM then accumulates at the bead-cell contact sites. Taken together, these results indicate that JAM is recruited to the nectin-1-mediated cell-cell or cell-bead adhesion sites, and that the trans-interaction of nectin-1 is necessary and sufficient for this recruitment of JAM.

Afadin- and ZO-1-dependent recruitment of JAM to the nectin-based cell-cell adhesion sites without the trans-interaction of JAM

Next we examined whether JAM is recruited to the nectin-based cell-cell adhesion sites through their cytoplasmic tail-binding proteins, afadin and ZO-1. For this purpose, we utilized cadherin-deficient L cells stably expressing both full-length nectin-2 and full-length JAM (nectin-2-JAM-L cells), both C-terminal four amino acids (aa)-deleted nectin-2 and full-length JAM (nectin-2- C-JAM-L cells), or both full-length nectin-2 and C-terminal two aa-deleted JAM (nectin-2-JAM-C-L cells). Nectin-2-C-L cells forms nectin-2-C-based cell-cell adhesion sites, but afadin does not accumulate at the adhesion sites, while afadin accumulates at the nectin-2-based cell-cell adhesion sites in nectin-2-L cells (Miyahara et al., 2000; Tachibana et al., 2000). Similarly, endogenous ZO-1 is recruited to the JAM-based cell-cell adhesion sites in JAM-L cells, whereas ZO-1 is not recruited to the JAM-C-based cell-cell adhesion sites in JAM-C-L cells (Itoh et al., 2001).

When nectin-2-JAM-L cells were co-cultured with L cells stably expressing both full-length nectin-3 (nectin-3-L cells), the immunofluorescence signal for nectin-2 was concentrated at the adhesion sites between two nectin-2-JAM-L cells and between nectin-2-JAM-L and nectin-3-L cells (Figure 9A). The signal for JAM was concentrated at the adhesion sites between two nectin-2-JAM-L cells. In addition, the signal for JAM was concentrated at 42% of the adhesion sites between nectin-2-JAM-L and nectin-3-L cells (n=106) (Figure 9A). These results indicate that JAM is recruited to the nectin-2/-3-based cell-cell adhesion sites without the trans-interaction of JAM in L cells. When nectin-2-C-JAM-L cells were co-cultured with nectin-3-L cells, the immunofluorescence signal for nectin-2-C was also concentrated at the adhesion sites between two nectin-2-C-JAM-L cells and between the nectin-2-C-JAM-L and nectin-3-L cells (Figure 9B). However, the signal for JAM at the adhesion sites between nectin-2-C-JAM-L and nectin-3-L cells was significantly reduced (6% of the adhesion sites between nectin-2-C-JAM-L and nectin-3-L cells, n=62) when compared to that of nectin-2-JAM-L cells. These results indicate that the binding of nectin-2 to afadin is required for the recruitment of JAM to the nectins-2/-3-based cell-cell adhesion sites. When nectin-2-JAM-C-L cells were co-cultured with nectin-3-L cells, the immunofluorescence signal for JAM-C was not concentrated at the adhesion sites between nectin-2-JAM-C- and nectin-3-L cells (less than 2% of the adhesion sites between nectin-2- C-JAM-L and nectin-3-L cells, n=53) (Figure 9C). These results indicate that the binding of JAM to ZO-1 is required for the recruitment of JAM to the nectins-2/-3-based cell-cell adhesion sites. Taken together, JAM is recruited to the nectins-2/-3-based cell-cell adhesion sites without the trans-interaction of JAM, and this recruitment of JAM requires afadin and ZO-1.

We next examined whether JAM is associated with nectin-2 through afadin and ZO-1. For this purpose, we performed immunoprecipitation analysis of nectin-2-JAM-L, nectin-2-C-JAM-L, and nectin-2-JAM-C-L cells. When the cell extracts of nectin-2-JAM-L and nectin-2-C-JAM-L cells were subjected to immunoprecipitation by the use of the nectin-2 mAb, similar amounts of nectins-2 and -2-C were precipitated (Figure 10A, lanes 3, 4, Nectin-2). Afadin and ZO-1 were co-immunoprecipitated with nectin-2, whereas they were not co-immunoprecipitated with nectin-2-C (Figure 10A, lanes 3, 4, Afadin and ZO-1), as described previously (Yokoyama et al., 2001). It should be noted that afadin was more efficiently co-immunoprecipitated than ZO-1 with nectin-2 (Figure 10A, lanes 3, 4, Afadin and ZO-1). However, JAM was not co-immunoprecipitated with nectin-2 (Figure 10A, lanes 3, 4, JAM). When the cell extracts of nectin-2-JAM-L and nectin-2-JAM-C-L cells were subjected to immunoprecipitation by the use of the JAM mAb, similar amounts of JAM and JAM-C were precipitated (Figure 10B, lanes 3, 4, JAM). Afadin and ZO-1 were co-immunoprecipitated with JAM, whereas they were not co-immunoprecipitated with JAM-C (Figure 10B, lanes 3, 4, Afadin and ZO-1). In this case, ZO-1 was more efficiently co-immunoprecipitated than afadin with JAM (Figure 10B, lanes 3, 4, Afadin and ZO-1). Nectin-2 was not co-immunoprecipitated with JAM (Figure 10B, lanes 3, 4, Nectin-2). Although nectin-2 was not co-immunoprecipitated with JAM, the co-immunoprecipitation of afadin and ZO-1 with both nectin and JAM suggest that JAM is associated with nectin-2 through afadin and ZO-1.

It has been reported that JAM directly binds AF-6, the splicing variant of afadin, through the interaction of the C-terminal PDZ binding motif of JAM with the PDZ domain of AF-6 (Ebnet et al., 2000). This result suggests that JAM directly binds afadin and nectin may be associated with JAM through afadin only. Thus, we next examined whether JAM directly binds afadin. A GST-fusion protein of the cytoplasmic region of JAM (GST-JAM-CP) did not bind an MBP-fusion protein of the PDZ domain of afadin (MBP-afadin-PDZ) immobilized on amylose resin beads (Figure 10C, lane 3). A GST-fusion protein of the cytoplasmic region of nectin-2 (GST-nectin-2-CP) bound MBP-afadin-PDZ and the stoichiometry of this interaction was 1 : 1 (Figure 10C, lane 4), as described previously (Yokoyama et al., 2001). These results indicate that JAM does not directly bind afadin, and suggest that JAM is indirectly associated with nectin-2 through afadin and ZO-1.

Discussion

We have previously shown that nectin has a potency to recruit the E-cadherin--catenin complex and ZO-1 to the nectin-based cell-cell adhesion sites in fibroblasts (Tachibana et al., 2000; Yokoyama et al., 2001). In this paper, we show several lines of evidence that nectin has a potency to recruit JAM to the nectin-based cell-cell adhesion sites. First, JAM was recruited to the nectin-1-based cell-cell adhesion sites during the formation of the junctional complex in nectin-1-MDCK cells. This recruitment of JAM was inhibited by the nectin inhibitors, gD and Nef-3, which inhibited the trans-interaction of nectin-1. Secondly, microbeads coated with the extracellular fragment of nectin-3, Nef-3, that interacts with cellular nectin-1, also recruited ZO-1 and JAM to the bead-cell contact sites. Thirdly, when nectin-2-JAM-L cells were co-cultured with nectin-3-L cells, JAM was concentrated at the adhesion sites between nectin-2-JAM-L and nectin-3-L cells without the trans-interaction of JAM. Analyses of the localization and immunoprecipitation of JAM revealed that it was associated with nectin through afadin and ZO-1. Taken together, our results indicate that the trans-interaction of nectin is necessary and sufficient for the recruitment of JAM, and that the nectin-afadin system plays a role in the recruitment of JAM in the process of the formation of the junctional complex in epithelial cells.

In the initial stage during the formation of AJs and TJs, primordial spot-like cell-cell junctions are first formed at the tips of the cellular protrusions radiating from adjacent cells (Yonemura et al., 1995; Adams et al., 1998; Ando-Akatsuka et al., 1999; Vasioukhin et al., 2000). The spot-like junctions fuse to form short belt-like junctions. The cadherin-catenin and nectin-afadin systems colocalize to the spot-like junctions where occludin is not concentrated (Ando-Akatsuka et al., 1999; Asakura et al., 1999; Sakisaka et al., 1999). Claudin is not concentrated at the spot-like junctions (Fukuhara et al., manuscript in preparation), but ZO-1 and JAM appear early at the spot-like junctions (Ebnet et al., 2001). Our results that nectin has a potency to recruit JAM to the nectin-based cell-cell adhesion sites suggest that nectin first accumulates at the spot-like junction, and then JAM accumulates there. However, when we compared the concentrations of nectin and JAM at the spot-like junction, we could not detect which one accumulates first in our conventional staining procedure. More detailed analyses with time-lapse imaging should be required for the analysis of the ordered assembly of the junctional complex.

How does nectin recruit JAM to the cell-cell adhesion sites? JAM interacts with ZO-1 through its C-terminal (Bazzoni et al., 2000; Ebnet et al., 2000; Itoh et al., 2001). We have previously found that nectin recruits ZO-1 to the nectin-based cell-cell adhesion sites through afadin in a cadherin-independent manner (Yokoyama et al., 2001). Our studies here show that the recruitment of JAM requires afadin and ZO-1. Afadin and ZO-1 were co-immunoprecipitated with either nectin or JAM, and JAM did not directly bind to afadin. Thus, nectin recruits JAM through their cytoplasmic-tail binding proteins, afadin and ZO-1. As afadin does not directly interact with ZO-1 (Yokoyama et al., 2001), both proteins may interact through an unidentified factor. When nectin-1-MDCK cells were cultured at 2 M Ca²⁺ for 120 min, nectin-1 still formed cell-cell adhesion at the basal sites of the cells and afadin and ZO-1 colocalized with nectin-1. However, the signal for JAM mostly disappeared from the nectin-1-based cell-cell adhesion sites. Thus, the interaction between JAM and ZO-1 could be regulated in response to Ca²⁺ switch. In addition, in polarized epithelial cells, ZO-1 and JAM localize at TJs, while nectin localizes at AJs, indicating that there must be unknown mechanisms that regulate the translocation of ZO-1 and JAM from the nectin-based cell-cell adhesion sites to TJs. It is of crucial importance to clarify the molecular linkages among the nectin-afadin complex and the JAM-ZO-1 complex to understand how the junctional complex is formed in epithelial cells.

Materials and methods

Materials and chemicals

Rabbit anti-nectin-1 and anti-nectin-2 pAb were prepared as previously described (Takahashi et al., 1999). Rat anti-nectin-2 and anti-nectin-3 mAbs were prepared as previously described (Takahashi et al., 1999; Satoh-Horikawa et al., 2000). A mouse anti-afadin mAb was prepared as described (Sakisaka et al., 1999). A mouse anti-ZO-1 mAb (Itoh et al., 1993) was kindly supplied from Dr Tsukita (Kyoto University, Kyoto, Japan). A rat anti-E-cadherin mAb (ECCD-2) was supplied by Dr M Takeichi (Center for Developmental Biology, RIKEN, Kobe, Japan). A rabbit anti-human JAM pAb and a mouse anti-human JAM mAb were raised and characterized previously (Ozaki et al., 1999; Itoh et al., 2001). A mouse anti-FLAG mAb was from SIGMA. Mouse anti-EEA1 and anti-E-cadherin mAbs were from Transduction. A rabbit anti-Rab11 pAb was from Santa Cruz. Secondary Abs for immunofluorescence microscopy were obtained from Chemicon International, Inc. gD and Nef-3 (the chimeric proteins of extracellular fragments of glycoprotein D and nectin-3 fused with human IgG Fc) were prepared as described (Tachibana et al., 2000; Sakisaka et al., 2001; Honda et al., submitted).

Cell culture and DNA transfection

MDCK cells were kindly supplied by Dr W Birchmeier (Max-Delbruck-Center for Molecular Medicine, Berlin, Germany). Nectin-1-MDCK cells were prepared as described (Takahashi et al., 1999). Briefly, MDCK cells were transfected with pCAGI-puro-FLAG-nectin-1 using the LipofectAMINE reagent (Gibco BRL). The cells were then cultured for 24 h, replated, and selected by culturing in the presence of 5 g/ml puromycin (SIGMA). JAM-L and JAM-C-L cells were kindly supplied by Dr S Tsukita (Kyoto University, Kyoto, Japan). Nectin-3-L cells were prepared as described (Satoh-Horikawa et al., 2000). Nectin-2-JAM-L or nectin-2-JAM-C-L cells were prepared as follows: JAM-L or JAM-C-L cells were transfected with pPGKIH-FLAG-nectin-2 (Miyahara et al., 2000) using LipofectAMINE reagent. The cells were cultured for 24 h, replated, and selected by culturing in the presence of 500 g/ml hygromicin B (Gibco BRL). Nectin-2-C-JAM-L cells were prepared as follows: JAM-L cells were transfected with pPGKIH-FLAG-nectin-2-C (Miyahara et al., 2000) using LipofectAMINE reagent. The cells were cultured for 24 h, replated, and selected by culturing in the presence of 500 g/ml hygromicin B (Gibco BRL).

Ca²⁺ switch assay

Ca²⁺ switch experiments using nectin-1-MDCK cells were done as described (Kartenbeck et al., 1991). Briefly, nectin-1-MDCK cells (1´10⁵) were seeded on 18-mm glass cover slips in 12-well culture dishes. Forty-eight hours later, the cells were washed with phosphate buffered saline (PBS) and cultured at 2 mM Ca²⁺ in DMEM without serum for 60 min. The cells were then cultured at 2 M Ca²⁺ (DMEM with 5 mM EGTA) in the presence or absence of 60 g/ml gD and 60 g/ml Nef-3 for indicated periods of time. After the culture, cells were washed with PBS and cultured at 2 mM Ca²⁺ in DMEM without serum in the presence or absence of 60 g/ml gD and 60 g/ml Nef-3 for indicated periods of time. When the cells were treated with TPA, the cells were washed with PBS and cultured at 2 mM Ca²⁺ in DMEM without serum for 60 min. The cells were then cultured at 2 M Ca²⁺ in DMEM with 5 mM EGTA in the presence or absence of 60 g/ml gD and 60 g/ml Nef-3 for 120 min. After the culture, 100 nM TPA was added to the medium and the cells were further cultured in the presence or absence of 60 g/ml gD and 60 g/ml Nef-3 for indicated periods of time.

Immunofluorescence microscopy

Immunofluorescence microscopy was done as described (Mandai et al., 1997, 1999). Briefly, the cells were fixed in the mixture of 50% acetone and 50% methanol at -20^°C for 1 min. The fixed cells were then washed three times with PBS. After blocking in PBS containing 1% BSA for 60 min, the cells were incubated in the same buffer with various combinations of the anti-afadin, anti-E-cadherin, anti-FLAG, anti-JAM, anti-nectin-2, anti-nectin-3 and anti-ZO-1 mAbs, and the anti-JAM, anti-nectin-1 and anti-nectin-2 pAbs for 60 min. The samples were washed three times with PBS for 5 min and incubated for 30 min in PBS containing 1% BSA with the secondary pAbs. The samples were then washed three times with PBS for 5 min and mounted in Immuno-fluore mounting medium (ICN Biomedicals, Inc.). The samples were analysed by a Radiance 2000 confocal laser scanning microscope (Bio-Rad Laboratories).

Preparation of Nef-3-coated beads

Latex-sulfate microspheres (3´10⁸, 5.2-m diameter; Interfacial Dynamics Corporation, Portland, OR, USA) were washed, resuspended in 0.2 ml of 0.1 M borate buffer, pH 8.0, and incubated with 100 g of a goat anti-human IgG (Fc-specific) Ab (Sigma) with gentle mixing at room temperature for 18 h. The beads were then centrifuged at 16 000 g at 4°C for 10 min and washed three times with 1 ml PBS. The beads were then incubated with 0.2 ml PBS containing 5 mg/ml BSA (BSA/PBS) at room temperature for 3 h. Aliquots of 0.2 ml of the bead suspension (6´10⁷ microspheres) were then incubated with 30 g of Nef-3 or human IgG at room temperature for 3 h. After the incubation, the beads were washed three times with 1 ml of BSA/PBS and re-suspended in 0.2 ml of BSA/PBS.

Bead-cell adhesion assay

Nectin-1-MDCK cells (3´10⁴) were seeded on 14-mm glass cover slips in 24-well culture dishes. Forty-eight hours later, the cells were washed with PBS and cultured at 2 mM Ca²⁺ in DMEM without serum for 60 min. The cells were then cultured at 2 M Ca²⁺ (DMEM with 5 mM EGTA) with latex-sulfate microspheres (3´10⁴) coated with human IgG or Nef-3 for 120 min. After the culture, the cells were washed with DMEM and cultured at 2 mM Ca²⁺ in DMEM with 10% FCS at 37°C for indicated periods of time.

Co-culture of various L cells

Nectin-3-L, nectin-2-JAM-L, nectin-2-C-JAM-L, and nectin-2-JAM-C-L cells were washed with PBS, incubated with 0.2% trypsin and 1 mM EDTA at 37°C for 5 min, and dispersed by gentle pipetting to obtain single cell suspensions. The cell number of each cell line was counted, and nectin-2-JAM-L, nectin-2-C-JAM-L, or nectin-2-JAM-C-L cells were mixed with an equal number of nectin-3-L cells. Three combinations of nectin-3-L and nectin-2-JAM-L cells, nectin-3-L and nectin-2-C-JAM-L cells, and nectin-3-L and nectin-2-JAM-C-L cells were prepared. These mixed cells (6´10⁴) were seeded on 18-mm glass cover slips in 12-well culture dishes and cultured for 4 h.

Immunoprecipitation

Nectin-2-JAM-L, nectin-2-C-JAM-L, or nectin-2-JAM-C-L cells were sonicated in an ice-cold lysis buffer (25 mM Tris/HCl at pH 7.5, 1 mM CaCl₂, 1 mM MgCl₂, 100 mM NaCl, 1% Triton X-100, 2 g/ml Aprotinin, 10 g/ml leupeptin, and 100 g/ml PMSF), followed by centrifugation at 100 000´g at 4°C for 15 min. The supernatant (500 g of protein) was incubated with the anti-FLAG mAb or the anti-JAM mAb bound to protein G-Sepharose beads (20 g of Ab/20 l of beads) (Amersham Pharmacia Biotech) at 4°C overnight. After the beads were extensively washed with the lysis buffer, the bound proteins were eluted by boiling the beads in an SDS sample buffer (60 mM Tris/Cl at pH 6.7, 3% SDS, 2% (v/v) 2-mercaptoethanol, and 5% glycerol), and subjected to SDS-PAGE (10% polyacrylamide gel), followed by Western blotting.

Affinity chromatography

To examine the interaction of afadin with nectin-2 or JAM, MBP-afadin-PDZ (20 g of protein) was immobilized on amylose resin beads (20 l of wet volume). GST-Nectin-2-CP or GST-JAM-CP (each 100 g of protein) was applied to the MBP-fusion protein-immobilized beads equilibrated with PBS. After the beads were extensively washed with PBS, elution was performed with PBS containing 10 mM maltose. Each eluate was subjected to SDS-PAGE (15% polyacrylamide gel), followed by protein staining with Coomassie brilliant blue.

Other procedures

Protein concentrations were determined with BSA as a reference protein (Bradford, 1976). SDS-PAGE was done as described (Laemmli, 1970).

Acknowledgements

We thank Dr M Takeichi (Center for Developmental Biology, RIKEN, Kobe, Japan) for providing us with the anti-E-cadherin mAb, Dr Sh Tsukita (Kyoto University, Kyoto, Japan) for providing us with the JAM-L and JAM-C-L cells and the anti-ZO-1 mAb, and Dr W Birchmeier (Max-Delbruck-Center for Molecular Medicine, Berlin, Germany) for providing us with MDCK cells. The investigation at Osaka University Medical School was supported by grants-in-aid for Scientific Research and for Cancer Research from the Ministry of Education, Science, Sports, Culture, and Technology, Japan (2000, 2001).

Adams CL, Chen YT, Smith SJ, Nelson WJ. (1998). J. Cell Biol., 142: 1105-1119. Article MEDLINE

Ando-Akatsuka Y, Yonemura S, Itoh M, Furuse M, Tsukita S. (1999). J. Cell Physiol., 179: 115-125. MEDLINE

Aoki J, Koike S, Ise I, Sato-Yoshida Y, Nomoto A. (1994). J. Biol. Chem., 269: 8431-8438. MEDLINE

Arrate MP, Rodriguez JM, Tran TM, Brock TA, Cunningham SA. (2001). J. Biol. Chem., 276: 45826-45832.

Asakura T, Nakanishi H, Sakisaka T, Takahashi K, Mandai K, Nishimura M, Sasaki T, Takai Y. (1999). Genes Cells., 4: 573-581. MEDLINE

Aurrand-Lions M, Duncan L, Ballestrem C, Imhof BA. (2001). J. Biol. Chem., 276: 2733-2741. MEDLINE

Balda MS, Gonzalez-Mariscal L, Matter K, Cereijido M, Anderson JM. (1993). J. Cell Biol., 123: 293-302. MEDLINE

Barton ES, Forrest JC, Connolly JL, Chappell JD, Liu Y, Schnell FJ, Nusrat A, Parkos CA, Dermody TS. (2001). Cell, 104: 441-451. MEDLINE

Bazzoni G, Martinez-Estrada OM, Orsenigo F, Cordenonsi M, Citi S, Dejana E. (2000). J. Biol. Chem., 275: 20520-20526. Article MEDLINE

Bradford MM. (1976). Anal. Biochem., 72: 248-254. Article MEDLINE

Campadelli-Fiume G, Cocchi F, Menotti L, Lopez M. (2000). Rev. Med. Virol., 10: 305-319.

Cocchi F, Lopez M, Menotti L, Aoubala M, Dubreuil P, Campadelli-Fiume G. (1998). Proc. Natl. Acad. Sci. USA., 95: 15700-15705. Article MEDLINE

Cocchi F, Menotti L, Dubreuil P, Lopez M, Campadelli-Fiume G. (2000). J. Virol., 74: 3909-3917. MEDLINE

Cunningham SA, Arrate MP, Rodriguez JM, Bjercke RJ, Vanderslice P, Morris AP, Brock TA. (2000). J. Biol. Chem., 275: 34750-34756. MEDLINE

Eberle F, Dubreuil P, Mattei MG, Devilard E, Lopez M. (1995). Gene, 159: 267-272. Article MEDLINE

Ebnet K, Schulz CU, Meyer Zu Brickwedde MK, Pendl GG, Vestweber D. (2000). J. Biol. Chem., 275: 27979-27988. MEDLINE

Ebnet K, Suzuki A, Horikoshi Y, Hirose T, Meyer Zu Brickwedde MK, Ohno S, Vestweber D. (2001). EMBO J., 20: 3738-3748. Article MEDLINE

Furuse M, Itoh M, Hirase T, Nagafuchi A, Yonemura S, Tsukita S, Tsukita S. (1994). J. Cell Biol., 127: 1617-1626. MEDLINE

Furuse M, Fujita K, Hiiragi T, Fujimoto K, Tsukita S. (1998a). J. Cell Biol., 141: 1539-1550. Article MEDLINE

Furuse M, Sasaki H, Fujimoto K, Tsukita S. (1998b). J. Cell Biol., 143: 391-401. Article MEDLINE

Geiger B, Ginsberg D. (1991). Cell Motil. Cytoskeleton., 20: 1-6. MEDLINE

Geraghty RJ, Krummenacher C, Cohen GH, Eisenberg RJ, Spear PG. (1998). Science, 280: 1618-1620. Article MEDLINE

Gumbiner B, Simons K. (1986). J. Cell Biol., 102: 457-468. MEDLINE

Gumbiner B, Stevenson B, Grimaldi A. (1988). J. Cell Biol., 107: 1575-1587. MEDLINE

Gumbiner BM. (1996). Cell, 84: 345-357. MEDLINE

Gumbiner BM. (2000). J. Cell Biol., 148: 399-403. Article MEDLINE

Haskins J, Gu L, Wittchen ES, Hibbard J, Stevenson BR. (1998). J. Cell Biol., 141: 199-208. Article MEDLINE

Ikeda W, Nakanishi H, Miyoshi J, Mandai K, Ishizaki H, Tanaka M, Togawa A, Takahashi K, Nishioka H, Yoshida H, Mizoguchi A, Nishikawa S, Takai Y. (1999). J. Cell Biol., 146: 1117-1132. Article MEDLINE

Itoh M, Nagafuchi A, Yonemura S, Kitani-Yasuda T, Tsukita S, Tsukita S. (1993). J. Cell Biol., 121: 491-502. MEDLINE

Itoh M, Nagafuchi A, Moroi S, Tsukita S. (1997). J. Cell Biol., 138: 181-192. Article MEDLINE

Itoh M, Morita K, Tsukita S. (1999a). J. Biol. Chem., 274: 5981-5986. Article MEDLINE

Itoh M, Furuse M, Morita K, Kubota K, Saitou M, Tsukita S. (1999b). J. Cell Biol., 147: 1351-1363. Article MEDLINE

Itoh M, Sasaki H, Furuse M, Ozaki H, Kita T, Tsukita S. (2001). J. Cell Biol., 154: 491-498. Article MEDLINE

Kartenbeck J, Schmelz M, Franke WW, Geiger B. (1991). J. Cell Biol., 113: 881-892. MEDLINE

Laemmli UK. (1970). Nature, 227: 680-685.

Liu Y, Nusrat A, Schnell FJ, Reaves TA, Walsh S, Pochet M, Parkos CA. (2000). J. Cell Sci., 113: 2363-2374. MEDLINE

Lopez M, Eberle F, Mattei MG, Gabert J, Birg F, Bardin F, Maroc C, Dubreuil P. (1995). Gene, 155: 261-265. Article MEDLINE

Lopez M, Aoubala M, Jordier F, Isnardon D, Gomez S, Dubreuil P. (1998). Blood, 92: 4602-4611. MEDLINE

Lopez M, Cocchi F, Menotti L, Avitabile E, Dubreuil P, Campadelli-Fiume G. (2000). J. Virol., 74: 1267-1274. MEDLINE

Mandai K, Nakanishi H, Satoh A, Obaishi H, Wada M, Nishioka H, Itoh M, Mizoguchi A, Aoki T, Fujimoto T, Matsuda Y, Tsukita S, Takai Y. (1997). J. Cell Biol., 139: 517-528. Article MEDLINE

Mandai K, Nakanishi H, Satoh A, Takahashi K, Satoh K, Nishioka H, Mizoguchi A, Takai Y. (1999). J. Cell Biol., 144: 1001-1017. Article MEDLINE

Martin-Padura I, Lostaglio S, Schneemann M, Williams L, Romano M, Fruscella P, Panzeri C, Stoppacciaro A, Ruco L, Villa A, Simmons D, Dejana E. (1998). J. Cell Biol., 142: 117-127. Article MEDLINE

Miyahara M, Nakanishi H, Takahashi K, Satoh-Horikawa K, Tachibana K, Takai Y. (2000). J. Biol. Chem., 275: 613-618. MEDLINE

Mizoguchi A, Nakanishi H, Kimura K, Matsubara K, Ozaki-Kuroda K, Katata T, Honda T, Kiyohara Y, Heo K, Higashi M, Tsutsumi T, Sonoda S, Ide C, Takai Y. (2002). J. Cell Biol., 156: 555-565. MEDLINE

Morrison ME, Racaniello VR. (1992). J. Virol., 66: 2807-2813. MEDLINE

Nagafuchi A, Takeichi M, Tsukita S. (1991). Cell, 65: 849-857. MEDLINE

Nishioka H, Mizoguchi A, Nakanishi H, Mandai K, Takahashi K, Kimura K, Satoh-Moriya A, Takai Y. (2000). J. Comp. Neurol., 424: 297-306. Article MEDLINE

Ostermann G, Weber KS, Zernecke A, Schroder A, Weber C. (2002). Nat. Immunol., 3: 151-158. Article MEDLINE

Ozaki H, Ishii K, Horiuchi H, Arai H, Kawamoto T, Okawa K, Iwamatsu A, Kita T. (1999). J. Immunol., 163: 553-557. MEDLINE

Ozaki-Kuroda K, Nakanishi H, Ohta H, Tanaka H, Kurihara H, Mueller S, Irie K, Ikeda W, Sakai T, Wimmer E, Nishimune Y, Takai Y. (2002). Curr. Biol., 12: 1145-1150.

Ozawa M, Baribault H, Kemler R. (1989). EMBO J., 8: 1711-1717. MEDLINE

Palmeri D, van Zante A, Huang CC, Hemmerich S, Rosen SD. (2000). J. Biol. Chem., 275: 19139-19145. Article MEDLINE

Pasdar M, Nelson WJ. (1988a). J. Cell Biol., 106: 677-685. MEDLINE

Pasdar M, Nelson WJ. (1988b). J. Cell Biol., 106: 687-695.

Prasad R, Gu Y, Alder H, Nakamura T, Canaani O, Saito H, Huebner K, Gale RP, Nowell PC, Kuriyama K, Miyazaki Y, Croce CM, Canaani E. (1993). Cancer Res., 53: 5624-5628. MEDLINE

Reymond N, Fabre S, Lecocq E, Adelaide J, Dubreuil P, Lopez M. (2001). J. Biol. Chem., 276: 43205-43215. MEDLINE

Sakisaka T, Nakanishi H, Takahashi K, Mandai K, Miyahara M, Satoh A, Takaishi K, Takai Y. (1999). Oncogene, 18: 1609-1617. Article MEDLINE

Sakisaka T, Taniguchi T, Nakanishi H, Takahashi K, Miyahara M, Ikeda W, Yokoyama S, Peng YF, Yamanishi K, Takai Y. (2001). J. Virol., 75: 4734-4743. MEDLINE

Satoh-Horikawa K, Nakanishi H, Takahashi K, Miyahara M, Nishimura M, Tachibana K, Mizoguchi A, Takai Y. (2000). J. Biol. Chem., 275: 10291-10299. MEDLINE

Siliciano JD, Goodenough DA. (1988). J. Cell Biol., 107: 2389-2399.

Spear PG, Eisenberg RJ, Cohen GH. (2000). Virology, 275: 1-8. Article MEDLINE

Suzuki K, Hu D, Bustos T, Zlotogora J, Richieri-Costa A, Helms JA, Spritz RA. (2000). Nat. Genet., 25: 427-430. Article MEDLINE

Tachibana K, Nakanishi H, Mandai K, Ozaki K, Ikeda W, Yamamoto Y, Nagafuchi A, Tsukita S, Takai Y. (2000). J. Cell Biol., 150: 1161-1176. MEDLINE

Takahashi K, Nakanishi H, Miyahara M, Mandai K, Satoh K, Satoh A, Nishioka H, Aoki J, Nomoto A, Mizoguchi A, Takai Y. (1999). J. Cell Biol., 145: 539-549. Article MEDLINE

Takeichi M. (1988). Development, 102: 639-655. MEDLINE

Takeichi M. (1991). Science, 251: 1451-1455. MEDLINE

Tsukita S, Tsukita S, Nagafuchi A, Yonemura S. (1992). Curr. Opin. Cell Biol., 4: 834-839. MEDLINE

Tsukita S, Furuse M. (1999). Trends. Cell Biol., 9: 268-273.

Tsukita S, Furuse M, Itoh M. (1999). Curr. Opin. Cell Biol., 11: 628-633. Article MEDLINE

Vasioukhin V, Bauer C, Yin M, Fuchs E. (2000). Cell, 100: 209-219. MEDLINE

Warner MS, Geraghty RJ, Martinez WM, Montgomery RI, Whitbeck JC, Xu R, Eisenberg RJ, Cohen GH, Spear PG. (1998). Virology, 246: 179-189. Article MEDLINE

Watabe M, Nagafuchi A, Tsukita S, Takeichi M. (1994). J. Cell Biol., 127: 247-256. MEDLINE

Watabe-Uchida M, Uchida N, Imamura Y, Nagafuchi A, Fujimoto K, Uemura T, Vermeulen S, van Roy F, Adamson ED, Takeichi M. (1998). J. Cell Biol., 142: 847-857. MEDLINE

Willott E, Balda MS, Fanning AS, Jameson B, Van Itallie C, Anderson JM. (1993). Proc. Natl. Acad. Sci. USA, 90: 7834-7838. MEDLINE

Wittchen ES, Haskins J, Stevenson BR. (1999). J. Biol. Chem., 274: 35179-35185. MEDLINE

Wittchen ES, Haskins J, Stevenson BR. (2000). J. Cell Biol., 151: 825-836.

Yokoyama S, Tachibana K, Nakanishi H, Yamamoto Y, Irie K, Mandai K, Nagafuchi A, Monden M, Takai Y. (2001). Mol. Biol. Cell, 12: 1595-1609. MEDLINE

Yonemura S, Itoh M, Nagafuchi A, Tsukita S. (1995). J. Cell Sci., 108: 127-142. MEDLINE

Figure 1 Disappearance of the immunofluoresence signals for E-cadherin and JAM from the plasma membrane by reduction of medium Ca²⁺ from a normal to a low concentration. Nectin-1-MDCK cells were cultured at 2 M Ca²⁺ for indicated periods of time. The cells were triple stained with the anti-FLAG mAb, the anti-JAM polyclonal Ab (pAb), and the anti-E-cadherin mAb. Arrows, nectin-1-based cell-cell adhesion sites remained at 2 M Ca²⁺; and Bars, 10 m. The results are representative of three independent experiments

Figure 2 Stability of nectin-1, E-cadherin, and JAM during the culture at low Ca²⁺. Nectin-1-MDCK cells were cultured at 2 M Ca²⁺ for indicated periods of time. The cells were homogenized and the lysates of the cells (10 g of protein each) were subjected to SDS-PAGE (8% polyacrylamide gel), followed by Western blotting with the anti-FLAG mAb, the anti-E-cadherin mAb, and the anti-JAM pAb. The results are representative of three independent experiments

Figure 3 Remaining of the nectin-1-based cell-cell adhesion sites and colocalization of afadin and ZO-1 with nectin-1 during the culture at low Ca²⁺. Nectin-1-MDCK cells were cultured at 2 M Ca²⁺ for 120 min, and the cells were double stained with the anti-nectin-1 pAb (a1, b1) and the anti-afadin mAb (a2, b2) (left panel) or with the anti-nectin-1 pAb (c1, d1) and the anti-ZO-1 mAb (c2, d2) (right panel). Two optical sections are shown for each staining. Confocal images of b1, b2, b3, d1, d2, and d3 are the apical side of the images of a1, a2, a3, c1, c2, and c3, respectively. Arrows, nectin-1-based cell-cell adhesion sites remained at 2 M Ca²⁺; and Bars, 10 m. The results are representative of three independent experiments

Figure 4 Recruitment of E-cadherin and JAM to the nectin-1-based cell-cell adhesion sites by increase of Ca²⁺ from a low to a high concentration. Nectin-1-MDCK cells were cultured at 2 M Ca²⁺ for 120 min and then incubated with 2 mM Ca²⁺ for indicated periods of time. (A) The cells were triple stained with the anti-FLAG mAb (a1, b1, c1, d1, e1, f1, g1, h1), the anti-JAM pAb (a2, b2, c2, d2, e2, f2, g2, h2), and the anti-E-cadherin mAb (a3, b3, c3, d3, e3, f3, g3, h3); (B) The cells were double stained with the anti-nectin-1 pAb (a1, b1, c1, d1, e1, f1, g1, h1) and the anti-ZO-1 mAb (a2, b2, c2, d2, e2, f2, g2, h2); and (C) The cells were double stained with the anti-nectin-1 pAb (a1, b1, c1, d1, e1, f1, g1, h1) and the anti-afadin mAb (a2, b2, c2, d2, e2, f2, g2, h2). Two optical sections are shown for each staining. Confocal images of b1, b2, b3, d1, d2, d3, f1, f2, f3, h1, h2, and h3 are the apical side of the images of a1, a2, a3, c1, c2, c3, e1, e2, e3, g1, g2, and g3, respectively. Arrows, the signal for E-cadherin concentrated at the nectin-1-based cell-cell adhesion sites; Arrowheads, the signal for JAM concentrated at the nectin-1-based cell-cell adhesion sites; and Bars, 10 m. The results are representative of three independent experiments

Figure 5 Inhibition by the nectin inhibitors of the recruitment of E-cadherin and JAM to the nectin-1-based cell-cell adhesion sites. Nectin-1-MDCK cells were cultured at 2 M Ca²⁺ for 120 min and then incubated with 2 mM Ca²⁺ for 60 min in the absence (Control) or presence (Inhibitors) of 60 g/ml gD and 60 g/ml Nef-3. The cells were triple stained with the anti-FLAG mAb, the anti-JAM pAb, and the anti-E-cadherin mAb. Bars, 10 m. The results are representative of three independent experiments

Figure 6 TPA-induced recruitment of JAM to the nectin-1-based cell-cell adhesion sites. Nectin-1-MDCK cells were cultured at 2 M Ca²⁺ for 120 min and then incubated with 100 nM TPA for indicated periods of time. The cells were triple stained with the anti-FLAG mAb (a1, b1, c1, d1, e1, f1), the anti-JAM pAb (a2, b2, c2, d2, e2, f2), and the anti-E-cadherin mAb (a3, b3, c3, d3, e3, f3). Two optical sections are shown for each staining. Confocal images of b1, b2, b3, d1, d2, d3, f1, f2, and f3 are the apical side of the images of a1, a2, a3, c1, c2, c3, e1, e2, and e3, respectively. Arrows, the signal for JAM, but not that for E-cadherin, concentrated at the nectin-1-based cell-cell adhesion sites; and Bars, 10 m. The results are representative of three independent experiments

Figure 7 Inhibition by the nectin inhibitors of the TPA-induced recruitment of JAM to the nectin-1-based cell-cell adhesion sites. Nectin-1-MDCK cells were cultured at 2 M Ca²⁺ for 120 min and then incubated with 100 nM TPA for 60 min in the absence (Control) or presence (Inhibitors) of 60 g/ml gD and 60 g/ml Nef-3. The cells were double stained with the anti-FLAG mAb and the anti-JAM mAb. Bars, 10 m. The results are representative of three independent experiments

Figure 8 Recruitment of JAM and ZO-1 to the Nef-3-coated bead-cell contact sites. The Nef-3-coated or control microbeads were added to nectin-1-MDCK cells, followed by incubation at 2 M Ca²⁺ for 120 min. After the incubation, the cells were washed with DMEM and further incubated for various periods of time (360 min for nectin-1/JAM staining, 60 min for nectin-1/ZO-1 and nectin-1/afadin staining) at 2 mM Ca²⁺. The cells were then fixed, followed by immunostaining for cellular nectin-1, afadin, JAM, and ZO-1 using the anti-FLAG mAb or the anti-nectin-1 pAb, the anti-afadin mAb, the anti-ZO-1 mAb, and the anti-JAM pAb, respectively. The samples were analysed using a differential interference contrast and confocal laser scanning microscope. The position of the bead is marked with an asterisk. Arrow, bead-cell contact sites; and Bars, 10 m. The results shown are representative of three independent experiments

Figure 9 Afadin- and ZO-1-dependent recruitment of JAM to the nectin-2- and -3-based cell-cell adhesion sites without the trans-interaction of JAM. (A) Nectin-2-JAM-L cells were co-cultured with nectin-3-L cells and triple stained with the anti-nectin-2 pAb, the anti-JAM mAb, and the anti-nectin-3 mAb.; (B) Nectin-2-C-JAM-L cells were co-cultured with nectin-3-L cells and triple stained with the anti-nectin-2 pAb, the anti-JAM mAb, and the anti-nectin-3 mAb.; and (C) Nectin-2-JAM-C-L cells were co-cultured with nectin-3-L cells and triple stained with the anti-nectin-2 pAb, the anti-JAM mAb, and the anti-nectin-3 mAb. Arrow, cell-cell adhesion sites between the same cells; Arrowhead, cell-cell adhesion sites between two different cells; N2/JAM, Nectin-2-JAM-L cell; N3, nectin-3 -L cell; N2C/JAM, Nectin-2-C-JAM-L cell; N2/JAMC, Nectin-2-JAM-C-L cell; and Bars, 10 m. The results shown are representative of three independent experiments

Figure 10 Indirect association of JAM and nectin-2 through afadin and ZO-1. (A) Association of nectin-2 with afadin and ZO-1. The cell extracts of nectin-2-JAM-L and nectin-2-C-JAM-L cells (each 500 g of protein) were separately subjected to immunoprecipitation with the anti-FLAG mAb. The immunoprecipitate was then subjected to SDS-PAGE (10% polyacrylamide gel), followed by Western blotting with the anti-nectin-2 pAb, the anti-ZO-1 mAb, the anti-afadin mAb, and the anti-JAM pAb. 1, 2; total extract, and 3, 4; immunoprecipitates of the anti-FLAG mAb. 1, 3; nectin-2-JAM-L cells; and 2, 4; nectin-2-C-JAM-L cells. (B) Association of JAM with afadin and ZO-1. The cell extracts of nectin-2-JAM-L and nectin-2-JAM-C-L cells (each 500 g of protein) were separately subjected to immunoprecipitation with the anti-JAM mAb. The immunoprecipitate was then subjected to SDS-PAGE (10% polyacrylamide gel), followed by Western blotting with the anti-nectin-2 pAb, the anti-ZO-1 mAb, the anti-afadin mAb, and the anti-JAM pAb. 1, 2; total extract, and 3, 4; immunoprecipitates of the anti-JAM mAb. 1, 3; nectin-2-JAM-L cells; and 2, 4; nectin-2-JAM-C-L cells. (C) Inability of afadin to directly interact with JAM. MBP-afadin-PDZ (20 g of protein) was immobilized on amylose resin beads. GST-Nectin-2-CP and GST-JAM-CP (each 100 g of protein) were applied to the affinity beads. After the beads were extensively washed, elution was performed with 10 mM maltose. The eluate was subjected to SDS-PAGE (15% polyacrylamide gel), followed by protein staining with Coomassie brilliant blue. 1, GST-JAM-CP (input); 2, GST-Nectin-2-CP (input); 3, MBP-afadin-PDZ+GST-JAM-CP (eluate); and 4, MBP-afadin-PDZ+GST-Nectin-2-CP (eluate). The results shown are representative of three independent experiments