SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Tight junctions (TJs) and adherens junctions (AJs) form an apical junctional complex at the apical side of the lateral membranes of epithelial cells, in which TJs are aligned at the apical side of AJs. Many cell adhesion molecules (CAMs) and cell polarity molecules (CPMs) cooperatively regulate the formation of the apical junctional complex, but the mechanism for the alignment of TJs at the apical side of AJs is not fully understood. We developed a cellular system with which epithelial-like TJs and AJs were reconstituted in fibroblasts and analyzed the cooperative roles of CAMs and CPMs. We exogenously expressed various combinations of CAMs and CPMs in fibroblasts that express negligible amounts of these molecules endogenously. In these cells, the nectin-based cell–cell adhesion was formed at the apical side of the junctional adhesion molecule (JAM)-based cell–cell adhesion, and cadherin and claudin were recruited to the nectin-3- and JAM-based cell–cell adhesion sites to form AJ-like and TJ-like domains, respectively. This inversed alignment of the AJ-like and TJ-like domains was reversed by complementary expression of CPMs Par-3, atypical protein kinase C, Par-6, Crb3, Pals1 and Patj. We describe the cooperative roles of these CAMs and CPMs in the apico-basal alignment of TJs and AJs in epithelial cells.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Cellular polarity is a common feature in a variety of cell types in multicellular organisms (Lecuit & Pilot 2003; Nelson 2003). Cell–cell and cell–matrix junctions play important roles in the formation and maintenance of apico-basal polarity. For instance, epithelial cells are polarized along their apical/basal axis, making the structural and functional characteristics of the apical side of the lateral membranes distinct from those of the basal side. Tight junctions (TJs) and adherens junctions (AJs) are formed at the apical side of the lateral membranes, and TJs are aligned at the apical side of AJs (Farquhar & Palade 1963). TJs have two functions: one is a barrier function that prevents the passage of soluble molecules through the gaps between cells, and the other is a fence function that keeps the cell surface lipids at the baso-lateral region separated from those at the apical region (Tsukita et al. 2008; Anderson & Van Itallie 2009), although the latter function remains controversial (Ikenouchi et al. 2012). AJs play a role in mechanically connecting adjacent cells to resist strong contractile forces and to maintain tissue structure (Cavey & Lecuit 2009). Although AJs exist in both epithelial cells and fibroblasts, in fibroblast TJs are missing and apico-basal polarity is less developed. Another marked difference in cell–cell junctions between epithelial cells and fibroblasts is actin filaments (F-actin) that undercoat AJs. In epithelial cells, the actin structure shows a circumferential and continuous belt-like structure, whereas in fibroblasts, it shows a noncircumferential and discontinuous structure (Yonemura et al. 1995).

The major cell–cell adhesion molecules (CAMs) that constitute TJs are claudins, occludin and junctional adhesion molecules (JAMs) (Tsukita et al. 2001). Claudins are Ca2+-independent CAMs with four transmembrane segments and comprise a family consisting of over 20 members. Claudins form TJ strands to seal the apposed plasma membranes of adjacent cells. The structure of occludin is similar to that of claudins, but its function has not yet been established. JAMs are Ca2+-independent immunoglobulin (Ig)-like CAMs with a single transmembrane segment and comprise a family consisting of four members. Besides these CAMs, several peripheral membrane proteins, such as ZO-1, ZO-2 and ZO-3, are localized at TJs and interact with claudins, occludin and JAMs to link these CAMs to F-actin that undercoat TJs.

At AJs, cadherins and nectins are two major CAMs (Takeichi 1991; Takai et al. 2003). Cadherins are key Ca2+-dependent CAMs with a single transmembrane segment, comprising a family consisting of over 100 members, and are expressed in various kinds of cells including epithelial cells and fibroblasts. E-cadherin is mainly expressed in epithelial cells, whereas N-cadherin is mainly expressed in fibroblasts. E- and N-cadherins directly bind β-catenin and p120ctn (Takeichi 2007). β-Catenin in turn interacts with α-catenin, which binds many peripheral membrane proteins, such as vinculin, α-actinin and EPLIN. Cadherins are linked with F-actin through these proteins to reinforce the cell–cell junctional connection. Nectins are Ca2+-independent Ig-like CAMs with a single transmembrane segment and comprise a family consisting of four members: nectin-1, nectin-2, nectin-3 and nectin-4. Nectins are associated with F-actin through afadin. Our series of studies have showed that nectins first form cell–cell adhesion and recruit cadherins to the nectin-based cell–cell adhesion site to establish AJs (Takai et al. 2003; Nakanishi & Takai 2004; Ogita & Takai 2006). Afadin binds ponsin, ADIP and LMO7, and α-catenin binds β-catenin, α-actinin and vinculin, all of which are necessary for the mutual association of two CAMs, nectins and cadherins. It was recently shown that PLEKHA7 binds to p120ctn and is associated with E-cadherin in epithelial cells (Meng et al. 2008). We have found that PLEKHA7 binds to not only p120ctn but also afadin and that this binding is involved in the association of the nectin–afadin system with the E-cadherin–catenin system and the proper formation of AJs in epithelial cells (Kurita et al. in press).

Nectins are involved in the formation of not only AJs but also TJs and in the establishment of cell polarization (Takai et al. 2003; Nakanishi & Takai 2004). While and/or after the formation of AJs, nectins first recruit JAMs and then recruit claudins and occludin to the apical side of AJs, resulting in the formation of TJs. It remains unknown how the trans-interactions of nectins lead to the recruitment of the TJ components, but the interaction of afadin and ZO-1 is essential for the recruitment of CAMs at TJs (Ooshio et al. 2010). The actin cytoskeleton is also required for this recruitment. It had been believed that cadherins are required for the formation of TJs (Gumbiner et al. 1988), but it was recently shown that the cadherin-based cell–cell adhesion is not absolutely necessary for the formation of TJs (Harris & Peifer 2004; Okamoto et al. 2005; Yamada et al. 2006; Capaldo & Macara 2007).

Genetic and cell biological studies have identified many factors that are required for the apico-basal polarization of epithelial cells and are called cell polarity molecules (CPMs) (Shin et al. 2006; Suzuki & Ohno 2006). They form a variety of complexes dynamically and sequentially in the processes of apico-basal polarization. There are three major polarity complexes, the apical Crumbs complex consisting of three proteins, Crumbs3 (Crb3), protein associated with Lin-7 (Pals1) and Pals1-associated tight junction protein (Patj); the Par complex consisting of Par-3, Par-6 and atypical protein kinase C (aPKC); and the lateral Scribble complex consisting of Scribble, discs large (Dlg1) and lethal giant larvae protein (Lgl). These complexes together serve as the core proteins establishing epithelial apico-basal polarity (Roh & Margolis 2003; Suzuki & Ohno 2006). Activated Cdc42 small G protein (GTP-Cdc42) binds to Par-6, which is bound to aPKC and Lgl2, and induces the phosphorylation of Lgl2 by aPKC, leading to replacement of phosphorylated Lgl2 with Par-3 and the formation of a complex of GTP-Cdc42-Par-6-Par-3. Par-6 directly binds to Crb3 or via Pals1, and Patj binds to ZO-3, which is bound to the cytoplasmic tails of claudins and occludin. Par-3 directly binds to the cytoplasmic tails of JAMs, nectin-1 and nectin-3 (Ebnet et al. 2001; Itoh et al. 2001; Takekuni et al. 2003).

It still remains to be fully understood how CAMs and CPMs cooperatively form TJs at the apical side of AJs to establish apico-basal polarity in epithelial cells. Development of the reconstruction system in which epithelial-like TJs and AJs are formed in fibroblasts is useful to promote our understanding of the cooperative roles of these molecules in the apico-basal polarization in epithelial cells. Here, we attempted to reconstruct TJs and AJs with apico-basal polarity by expressing various combinations of CAMs and CPMs in L and NIH3T3 cells that did not express endogenously or express only faintly these molecules.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Apico-basal alignment of nectin-3-based and JAM-A-based domains in L cells

L cells express nectin-1 and nectin-2 to small extents and lack any types of cadherins, claudins or occludin (Nagafuchi et al. 1987; Saitou et al. 1997; Furuse et al. 1998; Honda et al. 2003). JAM-A was expressed to a small extent (W. Ikeda and Y. Takai, unpublished data). We first obtained L cells stably expressing various combinations of CAMs at AJs and TJs and used them for the study on the formation of AJs and TJs. When L cells expressing various combinations of CAMs were cultured under conventional conditions in which the cells were cultured on tissue culture dishes or cover glasses, apico-basal polarity, which is observed in epithelial cells, was not markedly developed, at least partially, because the apico-basal length of these cell lines was not long enough (Kuramitsu et al. 2008; W. Ikeda and Y. Takai, unpublished data). It was previously shown that epithelial cells cultured on Transwell filters developed clearer apico-basal cell polarity when compared with those cultured under conventional conditions (Umeda et al. 2004). Therefore, we cultured L cells on Transwell filters, so that their apico-basal length became long enough to develop polarity.

When L cells expressing both nectin-3 and JAM-A (nectin-3-JAM-A-L cells) were cultured on Transwell filters, the immunofluorescence signals for nectin-3 and JAM-A were concentrated at cell–cell contact sites and the signal for JAM-A was concentrated at the basal side of the signal for nectin-3 in the vertical section (Fig. 1A). This distribution pattern was different from that observed in epithelial cells in which the signal for JAM-A was concentrated at the apical side of the signal for nectin-3. These results indicate that nectin-3 and JAM-A form distinctly separated cell–cell adhesion membrane domains, although this alignment was inversed as compared to that of epithelial cells. AJs of both epithelial cells and fibroblasts are undercoated with F-actin (Yonemura et al. 1995). Although the AJs of epithelial cells form a circumferential and continuous belt-like structure, AJs in fibroblasts are discontinuous and not circumferential. In nectin-3-JAM-A-L cells, the nectin-3-based and JAM-A-based domains were discontinuous and not circumferential (W. Ikeda and Y. Takai, unpublished data).

image

Figure 1. Regular apico-basal alignment of nectin-3- and JAM-A-based cell–cell adhesion domains in L cells. L cells expressing indicated combinations of CAMs were cultured on Transwell filters for 72 h and stained with various combinations of the anti-JAM-A pAb, the anti-nectin-3 mAb, the anti-E-cadherin mAb and the anti-claudin-1 pAb. Each panel shows a respective confocal microscopic Z-section of the apico-basal localization of the proteins (apical at the top, basal at the bottom). (A) Nectin-3-JAM-A-L cells, (B) nectin-3-E-cadherin-L cells, (C) JAM-A-E-cadherin-L cells, (D) nectin-3-JAM-A-E-cadherin-L cells, (E) nectin-3-claudin-1-L cells, (F) E-cadherin-claudin-1-L cells, (G) JAM-A-claudin-1-L cells, (H) nectin-3-JAM-A-claudin-1-L cells. Bars: 5 μm. The results shown are representative of three independent experiments.

Download figure to PowerPoint

Unnecessity of E-cadherin or claudin-1 for apico-basal alignment of nectin-3-based and JAM-A-based domains

We next examined the effect of E-cadherin on the formation of the apico-basal alignment of the nectin-3-based and JAM-A-based domains, using L cells expressing nectin-3 and E-cadherin (nectin-3-E-cadherin-L cells), L cells expressing JAM-A and E-cadherin (JAM-A-E-cadherin-L cells) and L cells expressing nectin-3, JAM-A and E-cadherin (nectin-3-JAM-A-E-cadherin-L cells). In nectin-3-E-cadherin-L cells, the immunofluorescence signals for nectin-3 and E-cadherin were concentrated at cell–cell contact sites, overlapped with each other and distributed along the lateral plasma membranes in the vertical section (Fig. 1B). In JAM-A-E-cadherin-L cells, the signals for JAM-A and E-cadherin were concentrated at cell–cell contact sites but were randomly aligned along the lateral plasma membranes in the vertical section (Fig. 1C). In nectin-3-JAM-A-E-cadherin-L cells, the signals for nectin-3, JAM-A and E-cadherin were all concentrated at cell–cell contact sites, the colocalized signals for nectin-3 and E-cadherin were separated from the signal for JAM-A, and the signal for JAM-A was concentrated at the basal side of the signals for nectin-3 and E-cadherin in the vertical section (Fig. 1D). These results indicate that the establishment of proper apico-basal alignment needs both the nectin-3-based and JAM-A-based domains, but not the E-cadherin-based domain. It is likely that E-cadherin forms the same domain with nectin-3, which results in the apico-basal alignment of the nectin- and E-cadherin-based and JAM-A-based domains.

We then examined the effect of claudin-1 on the apico-basal alignment of the nectin-3-based and JAM-A-based domains, using L cells expressing nectin-3 and claudin-1 (nectin-3-claudin-1-L cells), L cells expressing E-cadherin and claudin-1 (E-cadherin-claudin-1-L cells), L cells expressing JAM-A and claudin-1 (JAM-A-claudin-1-L cells) and L cells expressing nectin-3, JAM-A and claudin-1 (nectin-3-JAM-A-claudin-1-L cells). In nectin-3-claudin-1-L cells, the signals for nectin-3 and claudin-1 were concentrated at cell–cell contact sites, and the signal for nectin-3 was concentrated at the lateral plasma membranes, whereas the signal for claudin-1 was not regularly aligned along the lateral plasma membranes in the vertical section (Fig. 1E). In E-cadherin-claudin-1-L cells, the signals for E-cadherin and claudin-1 were concentrated at cell–cell contact sites and were separated but were not regularly aligned along the lateral plasma membranes in the vertical section (Fig. 1F). In JAM-A-claudin-1-L cells, the signals for JAM-A and claudin-1 were concentrated at cell–cell contact sites, the signal for JAM-A was concentrated at the lateral plasma membranes and the signal for claudin-1 partially overlapped with that for JAM-A (Fig. 1G). In nectin-3-JAM-A-claudin-1-L cells, the signals for nectin-3, JAM-A and claudin-1 were concentrated at cell–cell contact sites, and the signals for JAM-A and claudin-1 were colocalized and separated from that for nectin-3 (Fig. 1H). The signal for nectin-3 was concentrated at the apical side of the lateral plasma membranes, whereas the signals for JAM-A and claudin-1 overlapped and concentrated at the basal side of the lateral plasma membranes (Fig. 1H). These results indicate that claudin-1 is not sufficient to establish the formation of the apico-basal alignment of the nectin-3-based and JAM-A-based domains. In addition, they also indicate that E-cadherin is not sufficient to establish the apico-basal alignment of the nectin-3-based, JAM-A- and claudin-1-based domains. JAM-A and claudin-1 formed the same domain efficiently at the basal side of the nectin-based domain, consistent with our previous observation that the efficient colocalization of JAM-A and claudin-1 requires the nectin-3-based domain (Kuramitsu et al. 2008).

Necessity of afadin and ZO-1 for apico-basal alignment of nectin-3-based and JAM-A-based domains

We previously reported that JAM-A is recruited to the nectin-based cell–cell adhesion sites through the interaction between their respective cytoplasmic tail-binding proteins, ZO-1 and afadin (Fukuhara et al. 2002a; Ooshio et al. 2010). We therefore examined whether these molecules are required for the apico-basal alignment of the nectin-3-based and JAM-A-based domains. We first examined the localization of afadin and ZO-1 in L cells expressing nectin-3-ΔC, which lacks the C-terminal PDZ-binding motif of nectin-3 (nectin-3-ΔC-L cells) or L cells expressing JAM-A-ΔC, which lacks the C-terminal PDZ-binding motif of JAM-A (JAM-A-ΔC-L cells). In these cells, the immunofluorescence signal for nectin-3-ΔC or JAM-A-ΔC was concentrated at cell–cell contact sites, whereas those for afadin and ZO-1 were not concentrated there (Fig. 2A,B). We next made L cells expressing both nectin-3-ΔC and JAM-A-ΔC (nectin-3-ΔC-JAM-A-ΔC-L cells). In nectin-3-ΔC-JAM-A-ΔC-L cells, the signals for nectin-3-ΔC and JAM-A-ΔC were concentrated at cell–cell contact sites but separated from one another with a partial overlap (Fig. 2C,a). The signal for afadin or ZO-1 was not concentrated at cell–cell contact sites (Fig. 2C,b,c). In the vertical section, the signals for nectin-3-ΔC and JAM-A-ΔC were concentrated at the lateral plasma membranes, but were not regularly aligned along the lateral plasma membranes (Fig. 2D).

image

Figure 2. Necessity of the C-terminal PDZ-binding motifs of nectin-3 and JAM-A for regular apico-basal alignment of nectin-3- and JAM-A-based cell–cell adhesion domains. (A–C) Cell culture under a conventional condition. Cells were cultured on cover slips for 72 h and stained with various combinations of the anti-JAM-A pAb, the anti-nectin-3 mAb, the anti-afadin mAb and the anti-ZO-1 mAb. (A) Nectin-3-ΔC-L cells. (Aa) Nectin-3-ΔC and afadin; (Ab) nectin-3-ΔC and ZO-1. (B) JAM-A-ΔC-L cells. (Ba) JAM-A-ΔC and afadin; (Bb) JAM-A-ΔC and ZO-1. (C) Nectin-3-ΔC-JAM-A-ΔC-L cells. (Ca) JAM-A-ΔC and nectin-3-ΔC, (Cb) Afadin and nectin-3-ΔC, (Cc) JAM-A-ΔC and ZO-1. (D) Cell culture on Transwell filters. Nectin-3-ΔC-JAM-A-ΔC-L cells were cultured on Transwell filters for 72 h and stained with the anti-JAM-A pAb and the anti-nectin-3 mAb. Each panel shows a respective confocal microscopic Z-section of the apico-basal localization of the proteins (apical at the top, basal at the bottom). Bars: 5 μm. The results shown are representative of three independent experiments.

Download figure to PowerPoint

We then prepared nectin-3-JAM-A-L cells in which afadin or ZO-1 was knocked down by the siRNA (afadin-KD-nectin-3-JAM-A-L or ZO-1-KD-nectin-3-JAM-A-L cells, respectively). The amounts of afadin or ZO-1 proteins were markedly reduced in afadin-KD-nectin-3-JAM-A-L or ZO-1-KD-nectin-3-JAM-A-L cells, respectively (Fig. 3A,a,b). In both cell lines, the signals for nectin-3 and JAM-A were concentrated at cell–cell contact sites and separated from one another with a partial overlap (Fig. 3B,a,b). In the vertical section of afadin-KD-nectin-3-JAM-A-L and ZO-1-KD-nectin-3-JAM-A-L cells, the signals for nectin-3 and JAM-A were concentrated at the lateral plasma membranes, but were not regularly aligned along the lateral plasma membranes (Fig. 3C,a,b). Taken together, these results indicate that afadin and ZO-1 are not necessary for the formation of the separate domains of nectin-3 and JAM-A, but are necessary for the apico-basal alignment of their domains.

image

Figure 3. Necessity of afadin and ZO-1 for regular apico-basal alignment of nectin-3- and JAM-A-based cell–cell adhesion domains. (A,B) Confirmation of the knockdown of afadin and ZO-1. (A) Western blotting. Nectin-3-JAM-A-L cells were transfected with the siRNA against afadin, ZO-1 or control siRNA and cultured for 72 h. The cell lysates were subjected to SDS-PAGE, followed by Western blotting with the indicated Abs. (Aa) Knockdown of afadin; (Ab) knockdown of ZO-1. (B) Immunofluorescence images. Nectin-3-JAM-A-L cells were transfected with the siRNA against afadin or ZO-1 by reverse transfection method and plated on cover slips. After 72 h, the cells were stained with various combinations of the anti-JAM-A pAb, the anti-nectin-3 mAb, the anti-afadin mAb and the anti-ZO-1 mAb. (Ba) Afadin-knockdown cells; (Bb) ZO-1-knockdown cells. (C) Effect of afadin or ZO-1 knockdown on regular apico-basal alignment of nectin-3- and JAM-A-based cell–cell adhesion membrane domains. Nectin-3-JAM-A-L cells were transfected with the siRNA against afadin or ZO-1 by reverse transfection method and plated on Transwell filters. After 72 h, the cells were stained with various combinations of the anti-JAM-A pAb, the anti-nectin-3 mAb, the anti-afadin mAb and/or the anti-ZO-1 mAb. Each panel shows a respective confocal microscopic Z-section image of the apico-basal localization of the proteins (apical at the top, basal at the bottom). (Ca) Afadin-knockdown cells; (Cb) ZO-1-knockdown cells. Bars: 5 μm. The results shown are representative of three independent experiments.

Download figure to PowerPoint

Necessity of CPMs for normal alignment of nectin-3-based and JAM-A- and claudin-1-based domains in NIH3T3 cells

To convert the inversed alignment of the nectin-3-based and JAM-A- and claudin-1-based domains in nectin-3-JAM-A-claudin-1-L cells to the epithelial alignment, we attempted to express additionally a variety of combinations of CPMs, such as Par-6, Crb3 and Patj, which were not endogenously expressed in nectin-3-JAM-A-claudin-1-L cells. However, we could not obtain stable transformants expressing many CPMs in nectin-3-JAM-A-claudin-1-L cells, because the transfection efficiency to this cell line was markedly decreased. We then used NIH3T3 cells that were easily infected with a number of recombinant retrovirus, whereas L cell lines were not infected with them. We successfully established NIH3T3 cell lines that stably expressed more than 10 exogenous molecules and one endogenous molecule was stably knocked down. As for CAMs, NIH3T3 cells endogenously express N-cadherin, and small amounts of nectin-1, nectin-2 and nectin-3 (Fujito et al. 2005; W. Ikeda and Y. Takai, unpublished data). However, epithelial CAMs and the related proteins, E-cadherin, JAM-A, claudin, Necl-2, integrin α6β4, PLEKHA7 and Nezha were not expressed. In contrast, NIH3T3 cells strongly expressed extracellular matrix, fibronectin. To create epithelial-like cells from NIH3T3 cells, we exogenously expressed eight CAMs, E-cadherin, JAM-A, claudin-1, Necl-2, nectin-1, nectin-3, and integrin α6β4 together with two AJ regulator PLEKHA7 and Nezha, and stably knocked down fibronectin in NIH3T3 cells (10 genes-FNkd-NIH3T3 cells). As for CPMs, NIH3T3 cells endogenously expressed Par-3, aPKC and Pals1, but not Par-6, Crb3 and Patj. Therefore, we exogenously expressed three CPMs Myc-Crb3, Par-6b and Patj in 10 genes-FNkd-NIH3T3 cells (13 genes-FNkd-NIH3T3 cells). In Western blotting, exogenously expressed Par-6b, Patj and Myc-Crb3 were detected in 13 genes-FNkd-NIH3T3 cells (Fig. 4A). The expression levels of aPKC and Pals1 were increased. Exogenous expression of Myc-Crb3, Par-6b and Patj might affect the stability of these molecules by their interactions.

image

Figure 4. Necessity of CPMs for the epithelial alignment of AJ-like and TJ-like domains in NIH3T3 cells. (A) Expressions of the CPMs in 13 genes-FNkd-NIH3T3 cells. NIH3T3 cells that stably expressed exogenous 8 CAMs (nectin-1, nectin-3, E-cadherin, JAM-A, claudin-1, Necl-2, integrin α6 and integrin β4) and 2 AJ regulators (PLEKHA7 and Nezha) and in which fibronectin was stably knocked down (10 genes-FNkd-NIH3T3 cells) were stably transfected with 3 CPMs (Par-6b, Patj and Myc-Crb3) to establish 13 genes-FNkd-NIH3T3 cells. The cell lysate from 10 genes-FNkd-NIH3T3 or 13 genes-FNkd-NIH3T3 cells was subjected to Western blotting with the indicated Abs. (B) Localization patterns of CAMs and CPMs in 13 genes-FNkd-NIH3T3 cells. NIH3T3 cells and 13 genes-FNkd-NIH3T3 cells were cultured and stained with the indicated Abs. Bar: 10 μm. (C) Electron microscopic analysis of the apico-basal alignment of AJ-like and TJ-like domains in 13 genes-FNkd-NIH3T3 cells. Thirteen genes-FNkd-NIH3T3 cells were cultured and sampled for thin-section transmission electron microscope. The AJ-like and TJ-like domains were morphologically identified in each section, and the cell–cell adhesion sites were classified into four categories: TJ only, only a TJ-like domain was observed at the cell–cell adhesion sites; AJ only, only an AJ-like domain was observed at the cell–cell adhesion sites; AJ and TJ (normal), both AJ-like and TJ-like domains were observed at the cell–cell adhesion sites, and the TJ-like domain was localized at the apical side of the AJ-like domain; AJ and TJ (inversion), both AJ-like and TJ-like domains were observed at the cell–cell adhesion sites, whereas the TJ-like domain was localized at the basal side of the AJ-like domain. Twelve cell–cell adhesion sites from three independent experiments were examined. (Ca) Typical images of AJ and TJ (normal) and AJ and TJ (inversion). Bars: 500 nm. (Cb) Circular chart of the epithelial alignment of AJ and TJ (n = 12). AJ and TJ indicate AJ-like and TJ-like domains, respectively.

Download figure to PowerPoint

The immunofluorescence signals for nectin-3, E-cadherin, JAM-A and claudin-1 were all concentrated at cell–cell contact sites (Fig. 4B). In addition, the signals for Par-6b, Par-3, Myc-Crb3, Pals1 and Patj were similarly concentrated at the same cell–cell contact sites (Fig. 4B). Moreover, the signal for Crb3 was localized at the apical membrane. In these cells, the nectin-3- and E-cadherin-based and JAM-A- and claudin-1-based domains became more circumferential and continuous, although they were still incomplete as compared with AJs and TJs seen in epithelial cells. In the section of electron microscope, in approximately 42% of the cells, both TJ-like domains and AJ-like domains were formed, and the TJ-like domain was located at the apical side of the AJ-like domain. In approximately 25% of the cells, the AJ-like and TJ-like domains were formed, but the TJ-like domain was located at the basal side of the AJ-like domain (Fig. 4C,a,b). In approximately 33% of the cells, the AJ-like domain was formed, but the TJ-like domain was not observed (Fig. 4C,a,b). Electron microscopically, the nectin-3- and E-cadherin-based membrane domain resembled AJs, and the JAM-A- and claudin-1-based membrane domain resembled TJs. When either one of Par-6, Crb3 and Patj was not expressed or when endogenously expressed Par-3 or aPKC was knocked down, the frequency of the normal apico-basal alignment of the AJ-like and TJ-like domains formed in this NIH3T3 cell line was markedly reduced (W. Ikeda and Y. Takai, unpublished data). These results indicate that the CPMs over-expressed here convert the inversed alignment of AJ-like and TJ-like domains in NIH3T3 cells.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

In epithelial cells, the two different cell–cell adhesion membrane domains, TJs and AJs, are clearly separated when visualized by electron microscope (Tsukita et al. 2001). In addition, TJs are formed at the apical side of AJs. These junctions are undercoated with a circumferential and continuous belt-like F-actin structure, are continuously and circumferentially encircled along cell–cell contact sites and show a belt-like structure particularly in epithelial cells, but this continuity is less complete in endothelial cells as compared with that in epithelial cells (Yonemura et al. 1995; Bazzoni & Dejana 2004). We attempted here to reconstruct the apico-basal cell polarization by expressing various combinations of the molecular components of TJs and AJs and CPMs in L and NIH3T3 cells that have AJs, but not TJs, and do not have well-developed apico-basal polarity.

We showed here that nectin-3-JAM-A-L cells formed separate cell–cell adhesion domains, although the nectin-3-based domain was inversely formed at the apical side of the JAM-A-based domain. The same results were obtained when nectin-1 instead of nectin-3 was co-expressed with JAM-A in L cells (W. Ikeda and Y. Takai, unpublished data). Of the many combinations of homophilic and heterophilic trans-interactions of nectins, the trans-interaction between nectin-1 and nectin-3 is the strongest (Nakanishi & Takai 2004). When L cells that expressed nectin-1 and JAM-A and L cells that expressed nectin-3 and JAM-A were co-cultured, the essentially same results were obtained (W. Ikeda and Y. Takai, unpublished data). These results indicate that any member of the nectin family and JAM-A are necessary and sufficient for the apico-basal alignment of their cell–cell adhesion domains.

It is noted that when both E-cadherin and JAM-A were co-expressed in L cells, they formed separate cell–cell adhesion domains, but they were randomly aligned along the lateral plasma membranes. The additional expression of nectin-3 resulted in the formation of apico-basally aligned domains although the alignment of these domains was reversed as compared to that observed in epithelial cells. These results indicate that E-cadherin and JAM-A do not have the ability to form the apico-basal alignment independently of nectin-3. We previously showed that TJs are formed without the formation of the cadherin-based cell–cell adhesion under some conditions in cultured Madin-Darby canine kidney (MDCK) cells (Yamada et al. 2006). Under these conditions, both the nectin-based cell–cell adhesion and afadin are necessary for the formation of TJs. Similar results were obtained in the studies of MDCK cells and Drosophila models (Harris & Peifer 2004; Capaldo & Macara 2007).

Claudin-1 is colocalized with JAM-A at TJs in epithelial cells (Itoh et al. 2001; Suzuki et al. 2002). We previously showed that in L cells co-expressing JAM-A and claudin-1, the immunofluorescence signals for JAM-A and claudin-1 did not fully overlap (Kuramitsu et al. 2008). Co-expression of nectin-3 with JAM-A and claudin-1 enhanced the colocalization of the signals for JAM-A and claudin-1, indicating that JAM-A alone is not sufficient for the colocalization of claudin-1 with JAM-A and that nectin-3 is required for this colocalization. We showed here that nectin-3-JAM-A-claudin-1-L cells formed the apico-basal alignment of the nectin-3-based and JAM-A- and claudin-1-based domains. In contrast, nectin-3-claudin-1-L cells did not form the apico-basal alignment of the nectin-3-based and claudin-1-based domains. Moreover, co-expression of claudin-1 and nectin-3 or E-cadherin did not result in the apico-basal alignment of the nectin-3- and E-cadherin-based and claudin-1-based domains in the absence of JAM-A. These results indicate that JAM-A as well as nectin-3 is essential for the apico-basal polarization and are consistent with our previous observations that nectins regulate the formation of not only AJs but also TJs in MDCK cells (Fukuhara et al. 2002a,b; Ooshio et al. 2010).

We further investigated here how regular apico-basal alignment of the nectin-3-based and JAM-A-based domains is formed in L cells and showed that afadin and ZO-1, which bind to the C-terminal region of nectin-3 and JAM-A, respectively, were necessary for this apico-basal alignment. These results are consistent with previous observations that afadin and ZO-1 are necessary for the recruitment of JAM-A to the apical side of the nectin-based cell–cell adhesion sites in MDCK cells (Fukuhara et al. 2002a; Ooshio et al. 2010) and that ZO-1 and ZO-2 are necessary for the formation of TJs in EpH4 cells (Umeda et al. 2006).

To convert the inversed alignment of the nectin-3-based domain and JAM-A- and claudin-1-based domains in Nectin-3-JAM-A-Claudin-1-L cells to the epithelial alignment, we established 10 genes-FNkd-NIH3T3 cells and first attempted to analyze the vertical section by immunofluorescence microscopy. However, AJ-like and TJ-like domains became condensed by the expression of CPMs and we could not observe the clear vertical section due to the height of this NIH3T3 cell line. We then attempted to analyze the vertical section by electron microscope and found that the TJ-like domain was frequently localized at the apical side of the AJ-like domain. These results indicate that the CPMs used here convert the inversed alignment of AJ-like and TJ-like domains in NIH3T3 cells and that both CAMs and CPMs cooperatively form the apico-basal polarization. When either one of Par-6, Crb3 and Patj was not expressed or when endogenously expressed Par-3 or aPKC was knocked down, the frequency of the normal apico-basal alignment of AJ-like and TJ-like domains formed in this NIH3T3 cell line was markedly reduced (W. Ikeda and Y. Takai, unpublished data), indicating that all of these CPMs are essential for the apico-basal polarization.

It remains unknown how CPMs are localized at the JAM-A-based domain, which is localized at the apical side of the nectin-3-based domain in 13 genes-FNkd-NIH3T3 cells. Crb3 is localized at the apical plasma membrane of epithelial cells and interacts with Patj through Pals1 (Shin et al. 2006). Patj further directly binds to the cytoplasmic tail of JAMs (Adachi et al. 2009). Par-6 directly binds to Crb3 or indirectly via Pals1, and Patj binds to ZO-3, which binds to the cytoplasmic tail of occludin and claudins. ZO-3 binds to ZO-1 to form a heterodimer (Haskins et al. 1998). Par-3 directly binds to the cytoplasmic tails of JAMs and nectin-1 and nectin-3 (Ebnet et al. 2001; Itoh et al. 2001; Takekuni et al. 2003). ZO-1 as well as afadin is essential for the recruitment of JAMs to the apical side of the nectin-based cell–cell adhesion site in epithelial cells (Ooshio et al. 2010). We showed here that ZO-1 as well as afadin is also essential for the recruitment of JAM-A to the basal side of the nectin-based membrane domain in the reconstruction assay system using L cells.

Based on all of the previous and present findings, the most plausible explanation for the cooperative roles of the CPMs and the CAMs in the apico-basal alignment of TJs and AJs is as follows: (i) nectins first form cell–cell adhesion and recruits E-cadherin to the nectin-based cell–cell adhesion site through afadin, α-catenin and their binding proteins, resulting in the formation of AJs; (ii) Crb3 is transported to the apical membrane; (iii) Crb3 recruits JAM-A as the Pals1-Patj-ZO-3/ZO-1-JAM-A complex and/or the Pals1-Patj-JAM-A complex to the apical side of the nectin-based membrane domain; and (iv) thereafter, claudin-1 is recruited to the JAM-A-based domain with a help of the nectin-based domain, forming TJs at the apical side of AJs.

In conclusion, the present results indicate that nectins and JAMs, but not cadherins or claudins, play critical roles in the apico-basal polarization in epithelial cells in cooperation with the CPMs. In addition, not only the cell–cell adhesion mediated by the extracellular regions of nectins and JAMs but also their respective interactions with afadin and ZO-1 through their cytoplasmic tails are essential for the apico-basal cell polarization. Of the many CPMs identified thus far, Crb3 is localized at the apical plasma membrane. It is crucially important to elucidate the molecular mechanism of when and how this molecule is specifically transported to the apical plasma membrane of epithelial cells, in order to understand the whole picture of the cooperative roles of CAMs and CPMs to establish apico-basal polarity in epithelial cells.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Antibodies

A mouse anti-afadin monoclonal antibody (mAb) was prepared as described (Sakisaka et al. 1999). A rat anti-nectin-3 mAb was prepared as described (Satoh-Horikawa et al. 2000). A rat anti-Necl-2 mAb was prepared as described (Shingai et al. 2003). A rat anti-E-cadherin mAb was supplied by Dr Masatoshi Takeichi (Center for Developmental Biology, RIKEN, Kobe, Japan). A rat anti-JAM-1 mAb (H202) was supplied by Dr M. Aurrand-Lions (Malergue et al. 1998). A rabbit anti-JAM-A polyclonal antibody (pAb), a rabbit anti-claudin-1 pAb, anti-ZO-1 pAb and a rabbit anti-Par-3 pAb were purchased from Invitrogen. A mouse anti-ZO-1 mAb was purchased from Sanko Junyaku. A mouse anti-JAM1 mAb, a goat anti-nectin-3 pAb, a rabbit anti-Par-6b pAb and a rabbit anti-Pals1 pAb were purchased from Santa Cruz Biotechnology (Santa Cruz). A mouse anti-actin mAb (C4), a mouse anti-myc mAb (4A6) and a mouse anti-integrin β4 mAb (ASC-3) were purchased from Merck-Millipore. A mouse anti-aPKC mAb was purchased from BD transduction. A rat anti-integrin α6 mAb (GoH3) was purchased from Abcam. Rabbit antiserum against PATJ was raised against the recombinant PATJ protein fused to GST. Horseradish peroxidase-conjugated secondary Abs was purchased from GE Healthcare. Fluorophore (FITC, Cy3 and Cy5)-conjugated secondary Abs were purchased from Jackson ImmunoResearch.

Constructs

The cDNA fragments of mouse nectin-1, nectin-3, claudin-1, JAM-A, E-cadherin, Necl-2, myc-tagged Crb3, Patj, Par-6b, PLEKHA7 and Nezha were obtained from various mouse tissues by RT-PCR and cloned into pDONR vector series for MultiSite Gateway Pro system (Invitrogen) using BP recombination reactions according to manufacturer's protocol. The cDNA fragments of human integrin α6 and integrin β4 were obtained from a human tissue cDNA library by PCR and cloned into pDONR vector series as described above. The PCR fragments of PGK promoter, HA-tag and T7-tag were cloned into pDONR vector series as described above. We constructed destination vectors using Gateway rfA Cassette (Invitrogen) for Gateway system based on the retroviral vector pCX4 or pCSX4 with or without various drug-resistant genes, such as puromycin, blasticidin S, bleomycin and L-histidinol dihydrochloride, and obtained pCX4-DEST (no drug), pCX4-puro-DEST, pCX4-bsd-DEST, pCX4-bleo-DEST, pCX4-hisD-DEST and pCSX4-DEST (no drug). The pCX4 and pCSX4 vectors were kindly provided by Dr Akagi (Akagi et al. 2003). pCX4-claudin-1/PGK promoter/JAM-A, pCX4-E-cadherin/PGK promoter/Nectin-3, pCX4-Necl-2/PGK promoter/Nectin-1 and pCX4-bsr-mycCrb3/PGK promoter/Par-6b were generated by 3-fragment system of MultiSite Gateway Pro system using LR recombination reaction according to manufacturer's protocol. pCX4-puro-HA/PLEKHA7 and pCX4-hisD-T7/Nezha were generated by the 3-fragment system of MultiSite Gateway Pro system. pCX4-integrin α6, pCX4-integrin β4 and pCX4-bleo-Patj were generated by the single fragment system of Gateway system. For the knockdown of fibronectin (FN1), the BLOCK-iT Pol II miR RNA interference technology (Invitrogen) was used. Oligonucleotides targeting mouse fibronectin (5′-TGCTGATAAGTGTCACCCACTTTGTAGTTTTGGCCACTGACTGACTACAAAGTGTGACACTTAT-3′) were cloned into the pcDNA6.2-GW/EmGFP-miR vector according to manufacturer's protocol and we obtained pcDNA6.2GW/EmGFP-FN1 miR. Next, EmGFP of pcDNA6.2GW/EmGFP-FN1 miR was replaced by TagBFP (evrogen), and then, the region of TagBFP to FN1 miR was subcloned into pCSX4-ND-DEST by Gateway system and we obtained pCSX4-ND-TagBFP-FN1 miR.

Establishment of NIH3T3 cell lines by recombinant retrovirus infection

Ten genes-FNkd-NIH3T3 cells were obtained by retrovirus infection derived from pCX4-claudin-1/PGK promoter/JAM-A, pCX4-E-cadherin/PGK promoter/Nectin-3, pCX4-Necl-2/PGK promoter/Nectin-1, pCX4-integrin α6, pCX4-integrin β4. pCX4-puro-HA/PLEKHA7, pCX4-hisD-T7/Nezha and pCSX4-Fn1-miR-BFP. For the production of retroviruses, the retrovirus pCX4 or pCSX4 vector series were cotransfected with a packaging vector pE-eco and pGp into G3T-hi cells using TransIT-LT1 (Takara bio) and converted to retroviruses. Objective cells were obtained by FACS sorting and drug selection. After the infection, claudin-1- and JAM-A-expressing cells were enriched by FACS using the rat anti-JAM-A mAb. E-Cadherin and nectin-3-expressing cells were enriched by FACS using the rat anti-E-cadherin mAb. Necl-2- and nectin-1-expressing cells were enriched by FACS using the rat anti-Necl-2 mAb. Integrin α6β4-expressing cells were enriched by FACS using the mouse anti-integrin α6 mAb. PLEKHA7-expressing cells were selected by puromycin. Nezha-expressing cells were selected by L-histidinol dihydrochloride. FN1-miR-expressing cells were enriched by FACS using TagBFP as a marker. Claudin-1 and JAM-A were components of TJs. E-cadherin, nectin-1 and nectin-3 were components of AJs. Necl-2 was a CAM of epithelial lateral membrane. The reason why integrin α6β4 was expressed was because it is expressed in epithelial cells to form hemidesmosomes that are involved in the cell–matrix adhesion, but not in NIH3T3 cells (Wilhelmsen et al. 2006; Folgiero et al. 2007). The reason why PLEKHA7 and Nezha were exogenously expressed was because we recently found that PLEKHA7 binds to not only p120ctn but also afadin and that this binding is involved in the association of the nectin–afadin system and the E-cadherin–catenin system to form AJs properly in epithelial cells (Kurita et al. in press). The reason why fibronectin was knocked down was because it was more abundantly expressed in NIH3T3 cells than in epithelial cells and secreted into the matrix. To establish 13 genes-FNkd-NIH3T3 cells, 10 genes-FNkd-NIH3T3 cells were infected with recombinant retrovirus derived from pCX4-bsr-mycCrb3/PGK promoter/Par-6b and pCX4-bleo-Patj. These expressing cells were selected by blasticidin S and zeocin. Expressions of 13 genes and knockdown of fibronectin were confirmed by immunofluorescence staining, Western blotting or FACS analysis (Fig. 4 and data not shown).

Cell culture

Wild-type L, JAM-A-L, JAM-A-ΔC-L, claudin-1-L, E-cadherin-L and E-cadherin-claudin-1-L cells were kindly provided by Dr Shoichiro Tsukita (Kyoto University, Kyoto, Japan). Nectin-3-L cells were prepared as previously described (Satoh-Horikawa et al. 2000). Nectin-3-JAM-A-L, nectin-3-E-cadherin-L, JAM-A-E-cadherin-L, nectin-3-claudin-1-L, E-cadherin-claudin-1-L, JAM-A-claudin-1-L and nectin-3-JAM-A-claudin-1-L cells were prepared as previously described (Kuramitsu et al. 2008). To establish nectin-3-JAM-A-E-cadherin-L cells, pCAGIZeo-nectin-3 (Kuramitsu et al. 2008) was transfected into JAM-A-E-cadherin-L cells. Transfection into L cells was performed using Lipofectamine and Plus reagents (Invitrogen), and transfected L cells were selected by Zeocin (InvivoGen). These cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum. NIH3T3 cell lines were maintained in Dulbecco's modified Eagle's medium supplemented with 10% calf serum.

Immunofluorescence microscopy

For microscopy of conventional culture of L cell lines, 1 × 105 cells were plated on cover slips in 24-well culture dishes. To examine the vertical-sectional view of L cell lines, 2 × 105 cells were plated on Transwell filters (0.4-μm-pore polyester membrane insert in 12-well culture dishes; Corning). After 72 h of culture, the cells grown on cover slips or Transwell filters were fixed in a mixture of 50% acetone and 50% methanol at −20 °C for 1 min. The cells were then washed with PBS twice, incubated with 1% BSA in PBS for 1 h and then incubated with primary Abs for 1 h in a moist chamber. After being washed with PBS three times, the cells were incubated with the appropriate fluorophore-conjugated secondary Abs for 30 min. The cells were washed three times with PBS and then mounted in Prolong Gold mount gel (Invitrogen). The samples were analyzed by a confocal laser scanning microscope C1si-ready system (Nikon).

For microscopy of conventional culture of NIH3T3 cell lines, NIH3T3 cells were plated at density of 2 × 104 cells on diluted Matrigel-coated 13-mm cover slip onto 24-well plates and cultured in 5% CO2 for 3 days at 37 °C. The cells were fixed with acetone/methanol (1 : 1) at −20 °C or 1% PFA in PBS and subsequently permeabilized with 0.1% Triton X-100. The cells were incubated with 20% BlockAce (Dainihon Sumitomo Seiyaku) in PBS for 1 h. The samples were stained with various combinations of the primary Abs or Alexa-labeled Abs and then with appropriate fluorophore-conjugated secondary Abs (Jackson ImmunoResearch) or Alexa-labeled Abs. The samples were analyzed by a confocal laser scanning microscope A1Rsi system (Nikon).

Electron microscopy

Transmission electron microscopy was carried out as described (Mizoguchi et al. 2002; Honda et al. 2006). In brief, cells were fixed with 2% PFA and 2% glutaraldehyde (GA) in 0.1 m phosphate buffer (PB, pH 7.3) for 6 h and then in 1% tannic acid and 1% GA for 1 h after post-fixed 2% osmium tetroxide in PB for 1 h. The samples were dehydrated by passage through a graded series of ethanol and propylene oxide and embedded in epoxy resin. Ultrathin sections were stained 1% uranyl acetate and lead citrate and then observed with an electron microscope (H-7500; Hitachi).

siRNA experiments

For RNAi knockdown, a double-strand 21-nucleotide RNA duplex to afadin (5′-GAUUGGACAUUGAUGAGAATT-3′) was prepared as previously described (Kuramitsu et al. 2008). A double-stranded 21-nucleotide RNA duplex to ZO-1 (5′- GAGGCAUCAUCCCAAAUAATT-3′) and a similar duplex to luciferase (5′-CGUACGCGGAAUACUUCGATT-3′) with no significant homology to any known mammalian gene sequences (control) were purchased from Sigma Genosys. The cells were transfected with siRNAs using Lipofectamine RNAiMAX reagent (Invitrogen) according to manufacturer's protocol of reverse transfection.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We thank Drs Toru Kita, Masatoshi Takeichi and Shoichiro Tsukita for their generous gifts of reagents. This work was supported by the Global COE Program ‘Global Center for Education and Research in Integrative Membrane Biology’ and the Targeted Proteins Research Program from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan, Grants-in-Aid for Scientific Research (S) and (C) from the Japan Society for the Promotion of Science and the grants from the Naito Foundation, the Sagawa Foundation and the Yasuda Medical Foundation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Adachi, M., Hamazaki, Y., Kobayashi, Y., Itoh, M., Tsukita, S. & Furuse, M. (2009) Similar and distinct properties of MUPP1 and Patj, two homologous PDZ domain-containing tight-junction proteins. Mol. Cell. Biol. 29, 23722389.
  • Akagi, T., Sasai, K. & Hanafusa, H. (2003) Refractory nature of normal human diploid fibroblasts with respect to oncogene-mediated transformation. Proc. Natl Acad. Sci. USA 100, 1356713572.
  • Anderson, J.M. & Van Itallie, C.M. (2009) Physiology and function of the tight junction. Cold Spring Harb. Perspect. Biol. 1, a002584.
  • Bazzoni, G. & Dejana, E. (2004) Endothelial cell-to-cell junctions: molecular organization and role in vascular homeostasis. Physiol. Rev. 84, 869901.
  • Capaldo, C.T. & Macara, I.G. (2007) Depletion of E-cadherin disrupts establishment but not maintenance of cell junctions in Madin-Darby canine kidney epithelial cells. Mol. Biol. Cell 18, 189200.
  • Cavey, M. & Lecuit, T. (2009) Molecular bases of cell–cell junctions stability and dynamics. Cold Spring Harb. Perspect. Biol. 1, a002998.
  • Ebnet, K., Suzuki, A., Horikoshi, Y., Hirose, T., Meyer Zu Brickwedde, M.K., Ohno, S. & Vestweber, D. (2001) The cell polarity protein ASIP/PAR-3 directly associates with junctional adhesion molecule (JAM). EMBO J. 20, 37383748.
  • Farquhar, M.G. & Palade, G.E. (1963) Junctional complexes in various epithelia. J. Cell Biol. 17, 375412.
  • Folgiero, V., Bachelder, R.E., Bon, G., Sacchi, A., Falcioni, R. & Mercurio, A.M. (2007) The α6β4 integrin can regulate ErbB-3 expression: implications for α6β4 signaling and function. Cancer Res. 67, 16451652.
  • Fujito, T., Ikeda, W., Kakunaga, S., Minami, Y., Kajita, M., Sakamoto, Y., Monden, M. & Takai, Y. (2005) Inhibition of cell movement and proliferation by cell–cell contact-induced interaction of Necl-5 with nectin-3. J. Cell Biol. 171, 165173.
  • Fukuhara, A., Irie, K., Nakanishi, H., Takekuni, K., Kawakatsu, T., Ikeda, W., Yamada, A., Katata, T., Honda, T., Sato, T., Shimizu, K., Ozaki, H., Horiuchi, H., Kita, T. & Takai, Y. (2002a) Involvement of nectin in the localization of junctional adhesion molecule at tight junctions. Oncogene 21, 76427655.
  • Fukuhara, A., Irie, K., Yamada, A., Katata, T., Honda, T., Shimizu, K., Nakanishi, H. & Takai, Y. (2002b) Role of nectin in organization of tight junctions in epithelial cells. Genes Cells 7, 10591072.
  • Furuse, M., Sasaki, H., Fujimoto, K. & Tsukita, S. (1998) A single gene product, claudin-1 or -2, reconstitutes tight junction strands and recruits occludin in fibroblasts. J. Cell Biol. 143, 391401.
  • Gumbiner, B., Stevenson, B. & Grimaldi, A. (1988) The role of the cell adhesion molecule uvomorulin in the formation and maintenance of the epithelial junctional complex. J. Cell Biol. 107, 15751587.
  • Harris, T.J. & Peifer, M. (2004) Adherens junction-dependent and -independent steps in the establishment of epithelial cell polarity in Drosophila. J. Cell Biol. 167, 135147.
  • Haskins, J., Gu, L., Wittchen, E.S., Hibbard, J. & Stevenson, B.R. (1998) ZO-3, a novel member of the MAGUK protein family found at the tight junction, interacts with ZO-1 and occludin. J. Cell. Biol. 141, 199208.
  • Honda, T., Sakisaka, T., Yamada, T., Kumazawa, N., Hoshino, T., Kajita, M., Kayahara, T., Ishizaki, H., Tanaka-Okamoto, M., Mizoguchi, A., Manabe, T., Miyoshi, J. & Takai, Y. (2006) Involvement of nectins in the formation of puncta adherentia junctions and the mossy fiber trajectory in the mouse hippocampus. Mol. Cell. Neurosci. 31, 315325.
  • Honda, T., Shimizu, K., Kawakatsu, T., Yasumi, M., Shingai, T., Fukuhara, A., Ozaki-Kuroda, K., Irie, K., Nakanishi, H. & Takai, Y. (2003) Antagonistic and agonistic effects of an extracellular fragment of nectin on formation of E-cadherin-based cell–cell adhesion. Genes Cells 8, 5163.
  • Ikenouchi, J., Suzuki, M., Umeda, K., Ikeda, K., Taguchi, R., Kobayashi, T., Sato, S.B., Stolz, D.B. & Umeda, M. (2012) Lipid polarity is maintained in absence of tight junctions. J. Biol. Chem. 287, 95259533.
  • Itoh, M., Sasaki, H., Furuse, M., Ozaki, H., Kita, T. & Tsukita, S. (2001) Junctional adhesion molecule (JAM) binds to PAR-3: a possible mechanism for the recruitment of PAR-3 to tight junctions. J. Cell Biol. 154, 491497.
  • Kuramitsu, K., Ikeda, W., Inoue, N., Tamaru, Y. & Takai, Y. (2008) Novel role of nectin: implication in the co-localization of JAM-A and claudin-1 at the same cell–cell adhesion membrane domain. Genes Cells 13, 797805.
  • Kurita, S., Yamada, T., Rikitsu, E., Ikeda, W. & Takai, Y. (2009) Binding between the junctional proteins afadin and PLEKHA7 and implication in the formation of adherens junction in epithelial cells. J. Biol. Chem., in press.
  • Lecuit, T. & Pilot, F. (2003) Developmental control of cell morphogenesis: a focus on membrane growth. Nat. Cell Biol. 5, 103108.
  • Malergue, F., Galland, F., Martin, F., Mansuelle, P., Aurrand-Lions, M. & Naquet, P. (1998) A novel immunoglobulin superfamily junctional molecule expressed by antigen presenting cells, endothelial cells and platelets. Mol. Immunol. 35, 11111119.
  • Meng, W., Mushika, Y., Ichii, T. & Takeichi, M. (2008) Anchorage of microtubule minus ends to adherens junctions regulates epithelial cell–cell contacts. Cell 135, 948959.
  • 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) Nectin: an adhesion molecule involved in formation of synapses. J. Cell Biol. 156, 555565.
  • Nagafuchi, A., Shirayoshi, Y., Okazaki, K., Yasuda, K. & Takeichi, M. (1987) Transformation of cell adhesion properties by exogenously introduced E-cadherin cDNA. Nature 329, 341343.
  • Nakanishi, H. & Takai, Y. (2004) Roles of nectins in cell adhesion, migration and polarization. Biol. Chem. 385, 885892.
  • Nelson, W.J. (2003) Adaptation of core mechanisms to generate cell polarity. Nature 422, 766774.
  • Ogita, H. & Takai, Y. (2006) Nectins and nectin-like molecules: roles in cell adhesion, polarization, movement, and proliferation. IUBMB Life 58, 334343.
  • Okamoto, R., Irie, K., Yamada, A., Katata, T., Fukuhara, A. & Takai, Y. (2005) Recruitment of E-cadherin associated with a- and b-catenins and p120ctn to the nectin-based cell–cell adhesion sites by the action of 12-O-tetradecanoylphorbol-13-acetate in MDCK cells. Genes Cells 10, 435445.
  • Ooshio, T., Kobayashi, R., Ikeda, W., Miyata, M., Fukumoto, Y., Matsuzawa, N., Ogita, H. & Takai, Y. (2010) Involvement of the interaction of afadin with ZO-1 in the formation of tight junctions in Madin-Darby canine kidney cells. J. Biol. Chem. 285, 50035012.
  • Roh, M.H. & Margolis, B. (2003) Composition and function of PDZ protein complexes during cell polarization. Am. J. Physiol. Renal Physiol. 285, F377F387.
  • Saitou, M., Ando-Akatsuka, Y., Itoh, M., Furuse, M., Inazawa, J., Fujimoto, K. & Tsukita, S. (1997) Mammalian occludin in epithelial cells: its expression and subcellular distribution. Eur. J. Cell Biol. 73, 222231.
  • Sakisaka, T., Nakanishi, H., Takahashi, K., Mandai, K., Miyahara, M., Satoh, A., Takaishi, K. & Takai, Y. (1999) Different behavior of l-afadin and neurabin-II during the formation and destruction of cell–cell adherens junction. Oncogene 18, 16091617.
  • Satoh-Horikawa, K., Nakanishi, H., Takahashi, K., Miyahara, M., Nishimura, M., Tachibana, K., Mizoguchi, A. & Takai, Y. (2000) Nectin-3, a new member of immunoglobulin-like cell adhesion molecules that shows homophilic and heterophilic cell–cell adhesion activities. J. Biol. Chem. 275, 1029110299.
  • Shin, K., Fogg, V.C. & Margolis, B. (2006) Tight junctions and cell polarity. Annu. Rev. Cell Dev. Biol. 22, 207235.
  • Shingai, T., Ikeda, W., Kakunaga, S., Morimoto, K., Takekuni, K., Itoh, S., Satoh, K., Takeuchi, M., Imai, T., Monden, M. & Takai, Y. (2003) Implications of nectin-like molecule-2/IGSF4/RA175/SgIGSF/TSLC1/SynCAM1 in cell–cell adhesion and transmembrane protein localization in epithelial cells. J. Biol. Chem. 278, 3542135427.
  • Suzuki, A., Ishiyama, C., Hashiba, K., Shimizu, M., Ebnet, K. & Ohno, S. (2002) aPKC kinase activity is required for the asymmetric differentiation of the premature junctional complex during epithelial cell polarization. J. Cell Sci. 115, 35653573.
  • Suzuki, A. & Ohno, S. (2006) The PAR-aPKC system: lessons in polarity. J. Cell Sci. 119, 979987.
  • Takai, Y., Irie, K., Shimizu, K., Sakisaka, T. & Ikeda, W. (2003) Nectins and nectin-like molecules: roles in cell adhesion, migration, and polarization. Cancer Sci. 94, 655667.
  • Takeichi, M. (1991) Cadherin cell adhesion receptors as a morphogenetic regulator. Science 251, 14511455.
  • Takeichi, M. (2007) The cadherin superfamily in neuronal connections and interactions. Nat. Rev. Neurosci. 8, 1120.
  • Takekuni, K., Ikeda, W., Fujito, T., Morimoto, K., Takeuchi, M., Monden, M. & Takai, Y. (2003) Direct binding of cell polarity protein PAR-3 to cell–cell adhesion molecule nectin at neuroepithelial cells of developing mouse. J. Biol. Chem. 278, 54975500.
  • Tsukita, S., Furuse, M. & Itoh, M. (2001) Multifunctional strands in tight junctions. Nat. Rev. Mol. Cell Biol. 2, 285293.
  • Tsukita, S., Yamazaki, Y., Katsuno, T. & Tamura, A. (2008) Tight junction-based epithelial microenvironment and cell proliferation. Oncogene 27, 69306938.
  • Umeda, K., Ikenouchi, J., Katahira-Tayama, S., Furuse, K., Sasaki, H., Nakayama, M., Matsui, T., Tsukita, S. & Furuse, M. (2006) ZO-1 and ZO-2 independently determine where claudins are polymerized in tight-junction strand formation. Cell 126, 741754.
  • Umeda, K., Matsui, T., Nakayama, M., Furuse, K., Sasaki, H., Furuse, M. & Tsukita, S. (2004) Establishment and characterization of cultured epithelial cells lacking expression of ZO-1. J. Biol. Chem. 279, 4478544794.
  • Wilhelmsen, K., Litjens, S.H. & Sonnenberg, A. (2006) Multiple functions of the integrin a6b4 in epidermal homeostasis and tumorigenesis. Mol. Cell. Biol. 26, 28772886.
  • Yamada, A., Fujita, N., Sato, T., Okamoto, R., Ooshio, T., Hirota, T., Morimoto, K., Irie, K. & Takai, Y. (2006) Requirement of nectin, but not cadherin, for formation of claudin-based tight junctions in annexin II-knockdown MDCK cells. Oncogene 25, 50855102.
  • Yonemura, S., Itoh, M., Nagafuchi, A. & Tsukita, S. (1995) Cell-to-cell adherens junction formation and actin filament organization: similarities and differences between non-polarized fibroblasts and polarized epithelial cells. J. Cell Sci. 108(Pt 1), 127142.