In vertebrates, cell-cell-adhering proteins of both the punctum adhaerens and the desmosome type are characterized by transmembrane glycoproteins of the cadherin superfamily, and by cytoplasmic plaques, which comprise the common plaque protein plakoglobin (Cowin et al., 1986; Franke et al., 1989), as well as proteins that are specific for either the desmosome, such as desmoplakin (Franke et al., 1982; Virata et al., 1992) and one or two members of the plakophilin 1 to plakophilin 3 group of proteins, or the adherens junctions such as α- and β-catenin (Ozawa et al., 1989; Nagafuchi and Takeichi, 1989), protein p120 (Reynolds et al., 1994) and protein ARVCF (Mariner et al., 2000; Walter et al., 2008). In addition, another plaque protein with the molecular mass of 135 kDa that also belongs to the family of armadillo-type plaque proteins (Peifer and Wieschhaus, 1990; Peifer et al., 1994) has been discovered by Hatzfeld and Nachtsheim (Hatzfeld and Nachtsheim, 1996) and reported to also be a `dual-junction' molecule occurring in the plaques of both desmosomes and adherens junctions (Calkins et al., 2003; Hatzfeld et al., 2003; Setzer et al., 2004) (for a review, see Hatzfeld, 2005). Molecular binding partners of this protein – named p0071 – that are involved in the recruiting of the protein to plaques and its integration into plaques have also been described (e.g. Deguchi et al., 2000; Hatzfeld et al., 2003; Izawa et al., 2002; Jaulin-Bastard et al., 2002; Ohno et al., 2002).
To elucidate the distribution and functional contribution(s) of such plaque proteins in various developmental situations or in malignant growth and spread, one needs to identify the proteins at question sensitively and unequivocally in biochemical and cell biological experiments. Unfortunately, specific, well-defined and reliable antibodies against p0071 are not generally available, and the cell-type- and cell-structure-specificity as well as the nature of its molecular partners in functional complexes have not yet been adequately determined. One consequence of this lack of defined antibodies is the fact that the application of p0071 as a cell-type marker has so far not found its way into diagnostic pathology. Therefore, we have generated a number of cross-species-reactive antibodies that specifically recognize certain epitopes of protein p0071 and are useful in biochemical and cell biological studies. Using immunofluorescence and immunoelectron microscopy, as well as immunoprecipitation experiments, these antibodies have allowed the exclusion of a desmosomal location and the definition of a specific subtype of adherens junctions.
We have systematically prepared antibodies against various domains of p0071 (cf. Hatzfeld and Nachtsheim, 1996; Deguchi et al., 2000). The epitopes of those antibodies, which we have found most useful, are systematically presented in Fig. 1A, including murine monoclonal antibodies [designated clones p0071-1 to p0071-3 against peptide 1-1, clone p0071-4 against peptide 2-2, clones p0071-5 and p0071-6 against peptide 2-4, as well as guinea pig polyclonal antibodies against the cocktail of peptides 1-1, 1-2, 1-3 and 1-4 of the human (accession number NP_003619) and the murine (accession number AAH79848) sequence of p0071]. The antibodies were characterized and compared with antibodies against other junction proteins by using immunoblotting, coimmunoprecipitation, as well as light and electron microscopic immunolocalization techniques [for methods see Borrmann et al. (Borrmann et al., 2006)]. All murine monoclonal as well as guinea pig polyclonal antibodies mentioned here react with the p0071 polypeptide and localize to the same structures. The antibodies will soon be commercially available as single antibodies and as antibody cocktails covering several p0071 epitopes (for more information contact Progen Biotechnik, Heidelberg, Germany).
Specificity for polypetides with the molecular mass of 135 kDa – i.e. p0071 – in a diversity of human, rat and mouse cells and tissues is evident from Fig. 1B,C. Analysis of our immunoprecipitation experiments showed that constitutive proteins of adherens junctions co-immunoprecipitated (Fig. 1D shows examples for E-cadherin and β-catenin), whereas desmosomal proteins did not appear in such precipitates (Fig. 1D) [for immunoprecipitation methods see, for example, Hofmann et al. or Borrmann et al. (Hofmann et al., 2000; Borrmann et al., 2006)]. In addition, we noticed in carefully conducted immunoelectron microscopy experiments using various epithelial and non-epithelial tissues as well as cultured cells that all of our antibodies against p0071 reacted intensively and specifically with the plaques of adherens junctions, including even tiny puncta adhaerentia, but not at all with those of desmosomes (Fig. 1E). This exclusive specificity for adherens junctions and the exclusion of desmosomes was also demonstrated by high-resolution double-label immunofluorescence microscopy, in particular using monolayer cultures of epithelial cells. Fig. 2A presents an example of a human breast carcinoma cell (MCF-7 cells) monolayer culture, which shows mutually exclusive labeling of desmosomes and puncta adhaerentia.
We obtained corresponding immunolocalization results on cryostat sections of frozen tissues and – using some of the antibodies and the antigen-retrieval technique – also on sections of fixed and paraffin-embedded tissue samples. As an example, Fig. 2B-B″ presents lactating alveolar epithelial cells of a bovine mammary gland, showing colocalization of p0071 with β-catenin on epithelial plasma membranes. Similar results were obtained when analyzing a wide selection of single- and multi-layered epithelia (data not shown) or blood vessel endothelial cells (Fig. 2B-B″, arrows).
It has been reported previously that the junctional plaque protein p0071 also occurs specifically in sarcomeric I-bands of cross-striated muscles (Schröder et al., 2000). Therefore, we examined this unexpected localization using our antibodies in cardiac and skeletal muscle tissues from human, bovine and rat. As shown for the sarcomeric muscles of rat and human hearts in Fig. C-D″, protein p0071 does only occur in the plaques of the composite junctions (area compositae) of the cardiac intercalated disks [for the specific molecular composition of this type of junction plaque see Borrmann et al. (Borrmann et al., 2006)] in wide-spread colocalization with both desmoplakin (Fig. 2D-D″) and N-cadherin (Fig. 2C-C″), whereas the sarcomeres were negative.
The most important result of our study is that, in contrast to previous reports, protein p0071 does not occur at detectable levels in normal desmosomes, i.e. in cell-cell adhering junctions formed by specifc cadherins of both the desmocollin and desmoglein group, or by a distinct set of plaque proteins – desmoplakin and plakophilin (for reviews, see Getsios et al., 2004; Green and Gaudry, 2000). The reported accumulation of genuine p0071 in natural desmosomes (Hatzfeld and Nachtsheim, 1996; Hatzfeld et al., 2003) could not be confirmed by us for any of the epithelial, carcinoma and fetal myocardial cells of tissues and in cell cultures. Reports of accumulation of p0071 at desmosmal plaques that occurred upon transfections of certain cultured cells with constructs that express the p0071 gene (Hatzfeld et al., 2003) are, of course, not necessarily contradictory to our findings in non-transfected cells. However, essentially in agreement with several previous reports (Borrmann et al., 2006) we demonstrated that protein p0071 is a major constituent of the myocardial composite junctions and of the diverse morphological forms of adherens junctions, in which it forms stable complexes with specific catenins and `classic' cadherins (see also Calkins et al., 2003; Hatzfeld et al., 2003). It is obvious that the precise immunolocalization of p0071 that is now possible, as well as its precipitation, will significantly widen the application of junctional markers for cell biological research and for diagnostics in pathology.
We thank Jennifer Hermes for technical support, Mechthild Hatzfeld for helpful comments in the generation of antibodies and Lutz Langbein for critical discussion. This study was supported by project grants from the German Ministry for Education and Research (BMBF) in a cooperative research program entitled `Standardization of mesenchymal stem cells for regenerative medicine (START-MSC)' and from the Deutsche Krebshilfe to W.W.F. (project 10-2049).