WO2000066619A2

WO2000066619A2 - Means and methods for altering the functional properties in eukaryotic cells

Info

Publication number: WO2000066619A2
Application number: PCT/EP2000/004389
Authority: WO
Inventors: Frans Van Roy; Stefan Bonne
Original assignee: Vlaams Interuniversitair Instituut Voor Biotechnologie Vzw
Priority date: 1999-04-30
Filing date: 2000-04-28
Publication date: 2000-11-09
Also published as: AU4922400A; WO2000066619A3

Abstract

The present invention relates to the characterization of a novel, human catenin-like protein, denominated as plakophilin-3, which is present in desmosomes and nuclei of epithelial cells. The nucleic acids encoding said protein, the protein itself and antibodies against said protein can be used to diagnose and/or treat diseases, in particular skin diseases and wounds. Also the promoter sequence regulating plakophilin-3 gene expression and methods to modulate the regulatory function of said promoter are disclosed.

Description

Means and'methods for altering the functional properties in eukaryotic cells

The invention relates to the field of medicine. More in particular the invention relates to the field of diagnosis and therapy of disease. Cadherins and cadherin-like molecules are glycosylated proteins, which form an integral part of the cellular membrane of many cell types. The cadherins and cadherin-like proteins traverse the plasma membrane once and have an N-terminal extracellular domain. Cadherins and related proteins such as desmoglein are involved in cell-cell adhesion en play as such a role in morphogenesis, tumour suppression, tumour metastasis and inflammation. Expression of cadherins is spatiotemporarily controlled, dependent upon the phase of development and tissue- specific.

Catenins and catenin-like molecules are proteinaceous molecules that form complexes with cadherin or cadherin-like molecules and play an essential role in the function of said cadherins. Expression of catenins is tissue specific. Some catenins such as β-catenin, plakoglobin, p120^ctn and plakophilin-1 and -2 are Armadillo-like proteins. Armadillo-like (Arm) proteins are characterised by a series of repeats of 42-45 amino acid residues (AA), originally described in the Drosophila protein Armadillo (Riggleman et al., 1989). These Arm repeats have meanwhile been detected in a wide variety of proteins in a number of organisms, even including plants (Lo and Frasch, 1998; Peifer et al., 1994; Vithalani et al., 1998; Wang et al., 1998). β-catenin is probably the most extensively studied Arm protein at the moment. The protein acts as a mediator of cell-cell adhesion through the E-cadherin/catenin complex in epithelial cells (Aberle et al., 1994; Ozawa et al., 1989). However, β-catenin is also a regulator of gene transcription via its interaction with LEF-1/TCF transcription factors (Behrens et al., 1996; Molenaar et al., 1996). In the E-cadherin/catenin cell adhesion complex, either β-catenin or plakoglobin (also known as γ-catenin) links the cytoplasmic domain of E-cadherin to the actin cytoskeleton via αE-catenin (Herrenknecht et al., 1991 ; Knudsen et al., 1995; McCrea and Gumbiner, 1991 ). Another Arm protein present in this cell-cell adhesion complex is p120^ctπ, which interacts with the membrane-proximal part of the cytoplasmic domain of E-cadherin but not with αE-catenin (Daniel and Reynolds, 1995; Finnemann et al.,

1

CONFIRMATION COPT 1997; Lampugnani et al., 1997; Reynolds et al., 1994; Yap et al., 1998). The p120^c,n protein comprises 10 Arm repeats, while β-catenin and plakoglobin each comprise 12.5 Arm repeats. It serves as a prototype of a novel subfamily of Arm proteins (Reynolds and Daniel, 1997). Already five proteins have been described to share the p120^ctn Arm repeat organisation, namely plakophilin-1 (also known as "band 6 protein" of bovine muzzle desmosome fractions (Heid et al., 1994; Schmidt et al., 1997), plakophilin— 2 (Mertens et al., 1996), the ARVCF (Armadillo repeat gene deleted in velo cardio-facial syndrome) gene product (Sirotkin et al., 1997), p0071 (Hatzfeld and Nachtsheim, 1996), and δ-catenin, which is also known as the neural plakophilin-related arm-protein (NPRAP) (Paffenholz and Franke, 1997; Zhou et al., 1997). These proteins are generally expressed in a wide variety of cell types, except for δ-catenin/NPRAP, which is so far the only protein of this family displaying a restricted expression pattern, in this case neural tissue (Paffenholz and Franke, 1997). For several members of the p120^ct7plakophilin family a desmosomal localisation has been shown. Desmosomes (maculae adhaerentes) are epithelial adhering junctions involved in cell-cell adhesion, differentiation and signal transduction (Bornslaeger et al., 1997; Hatzfeld, 1997). They are assembled on a scaffold of transmembrane glycoproteins of the cadherin superfamily, i.e. the desmogleins and desmocollins (King et al., 1997; Koch et al., 1992). Both types of desmosomal cadherins are essential for cell-cell adhesion (Chitaev and Troyanovsky, 1997). Desmosomal cadherins are linked to the intermediate filament cytoskeleton through desmosomal plaque proteins, which commonly include plakoglobin and desmoplakin-l. Desmoplakin-ll, encoded by the same gene as desmoplakin-l but generated by alternative splicing, exhibits a more cell-type specific expression pattern (Bornslaeger et al., 1997). Additional desmosomal plaque components are continuously being detected and include plakophilin-1 , plakophilin-2 and p0071 , which share with plakoglobin the membership of the superfamily of Arm proteins (Hatzfeld and Nachtsheim, 1996; Heid et al., 1994; Mertens et al., 1996; Moll et al., 1997). Despite their initial identification as desmosomal plaque proteins, the plakophilins were shown to be generally expressed as nuclear proteins translocated to desmosomes only in certain stages of differentiation (Mertens et al., 1996; Schmidt et al., 1997). This was a surprising observation, especially in the case of plakophilin-1 , which had been assumed for a long time to be restricted to desmosomal cell-cell contacts of stratified and complex epithelia (Kapprell et al., 1988). These results suggest, besides a mechanical function for plakophilins in desmosomes of certain cell types, their involvement in signal transduction pathways between the plasma membrane and the cell nucleus. In cells deprived of desmosomes, both plakophilins may display an exclusively nuclear function. Originally, the p0071 protein has been localised to desmosomes in cultured cell lines (Hatzfeld and Nachtsheim, 1996). However, it has also been detected in cell nuclei, which suggests that p0071 might exert a role in signalling pathways (Hatzfeld, 1997).

Functional cadherin, catenin and/or cadherin/catenin complexes and related molecules and complexes are crucial for the correct behavior and functionality of a cell. Stabilizing mutations in β-catenin are associated with several forms of cancer such as colorectal cancer (Morin et al., 1997), sporadic hepatoblastoma (Koch et al., 1999) and human pilomatricomas, a common human skin tumor (Chan et al., 1999). Mutant forms of desmoplakin are involved in keratosis palmoplantaris striata (Keith et al., 1999). Mutants in plakophilin-1 have shown to cause the ectodermal dysplasia / skin fragility syndrome (McGrath et al., 1999; McGrath et al., 1997).

Due to the loss of correct cadherin-like, catenin-like and/or cadherin/catenin-like function in a cell, that cell is seriously affected in its function. Loss of correct function in a cell of a body may lead amongst others to tumorigenesis, metastasis, and skin diseases.

In one aspect the present invention provides in a method for providing a cell with an artificial cadherin/catenin-like molecular pathway. Provided with said artificial pathway said cell is capable of a novel means for modulating, i.e. upregulating or downregulating the expression of genes, thereby altering at least in part the function of said cell in response to a signal. Said signal may be an artificial signal not present in the natural environment of said cell. Said signal may be a natural signal that can be present in the environment of said cell. Said signal may even be a signal that said cell prior to being provided with said artificial pathway could respond to. However, said artificial pathway renders said cell capable of responding in a different way to said signal. Providing cells with a novel means for modulating the expression of genes can, in one aspect of the invention, at least in part alter, preferably decrease, the propensity with which said cell can metastasize. In another aspect said means can at least in part alter, preferably decrease the neoplastic properties of a cell. In particular, modulation of the expression of desmosomal catenin-like molecules may alter, preferably improve the wound healing characteristics of the skin.

In one aspect the invention provides in a method for altering an undesirable functional property of a cell comprising providing said cell with an artificial pathway suitable for transmitting and/or modifying signals in a cell. Preferably said cell is provided with an additional proteinaceous molecule capable of at least in part transmitting and/or modifying a signal of a cadherin/catenin-like signaling pathway. Preferably said pathway comprises at least one catenin-like and/or at least one cadherin-like molecule. Said undesirable functional property preferably comprises a neoplastic property and/or a metastatic property. Preferably said proteinaceous molecule is plakophilin-3 or a functional part, derivative and/or analogue thereof. More preferably a human, mouse or Xenopus laevis plakophilin-3 or a functional part, derivative and/or analogue thereof.

In one embodiment the invention provides an isolated or recombinant nucleic acid encoding a plakophilin-3, which in humans comprises a nucleic acid sequence as depicted in figure 2, and which in mice comprises a nucleic acid sequence as depicted in figure 10 and which in Xenopus laevis comprises a nucleic acid sequence as depicted in figure 12 or a functional part, derivative and/or analogue thereof.

The invention also provides a nucleic acid delivery vehicle comprising a nucleic acid according to the invention. Said vehicle may be any vehicle for the introduction of nucleic acid in a cell such as calciumphosphate precipitation, liposomes or viral vector medicated nucleic acid delivery. Preferably said nucleic acid delivery vehicle comprises an adenovirus particle, an adeno-associated virus particle, a retrovirus particle or a liposome particle. In one embodiment the invention provides a cell and/or the progeny thereof provided with a nucleic acid according to the invention. Preferably said cell is provided with said nucleic acid through contacting said cell with a nucleic delivery vehicle of the invention and incubating said in order to allow delivery of said nucleic acid to said cell. Incubation may be performed in vitro in for instance a culture dish or in vivo upon administration of said delivery vehicle to a body.

In another aspect the invention provides a proteinaceous molecule or a functional part, derivative and/or analogue thereof, derived from a nucleic acid according to the invention or a cell according to the invention. In one embodiment the invention provides in a proteinaceous molecule or a functional part, derivative and/or analogue thereof, wherein said molecule comprises a plakophilin-3 which in humans comprises an amino acid sequence as depicted in figure 2 in mice comprises an amino acid sequence as depicted in figure 11 and which in Xenopus laevis comprises an amino acid sequence as depicted in figure 13. In another aspect the invention provides an antibody or a functional part, derivative and/or analogue thereof, specific for a protein of the invention.

In another aspect the invention provides an anti idiotypic antibody or a functional part, derivative and/or analogue thereof, of an antibody according to the invention. The invention also provides a nucleic acid or a functional part, derivative and/or analogue thereof, encoding an antibody according to the invention. In one embodiment the invention provides a nucleic acid delivery vehicle comprising a nucleic acid encoding an antibody or a functional part, derivative and/or analogue thereof, of the invention. The invention further provides the use of a nucleic acid delivery vehicle of the invention in a Gene Therapy application. With the term "Gene Therapy application" as used herein is meant an application involving a treatment of an individual suffering from a disease or at risk of developing a disease, wherein said treatment comprises providing cells of said individual with nucleic acid of the invention, preferably through contacting said cells with a nucleic acid delivery vehicle of the invention, and incubating said cells in order to allow delivery of said nucleic acid to said cell. The invention further provides a method for diagnosing an epithelial tissue disease, either inherited or sporadic, comprising obtaining a sample of cells of fetal origin or of an affected area of the postnatal body and detecting the up- or down- regulation of the plakophilin-3 messenger RNA and/or the plakophilin-3 protein level and/or the presence or absence of a mutated plakophilin-3 and/or plakophilin-3 encoding nucleic acid, or a functional part, derivative and/or analogue thereof. Preferably said epithelial tissue comprises skin tissue. Such diagnosis can be realized, as a non-limiting example, by DNA-RNA hybridization or by a PCR-based technique, as known to the people skilled in the art, using sequence with Genbank ID nr. AF053719 (figure 2) or parts thereof as probe or primer, or by ELISA and related techniques, know to the people skilled in the art, using a monoclonal and/or polyclonal antibody or antibodies, raised against a protein with sequence with Genbank ID nr. AF053719 (figure 2) or fragments thereof. Healthy skin tissue may be used as reference material. The invention further provides nucleic acid probes and antibodies that can be used in said diagnosis.

In another aspect the invention provides a method for the treatment of skin disease such as ectodermal dysplasia/skin fragility, kerastosis palmoplantaris striata or skin cancer comprising providing skin cells of an affected skin area of an individual with a nucleic acid delivery vehicle of the invention.

In another aspect the invention provides a method for the treatment of skin disease such as ectodermal dysplasia/skin fragility, kerastosis palmoplantaris striata or skin cancer comprising providing skin cells of an affected skin area of an individual with a chemical compound affecting the expression of the plakophilin-3 gene and/or the function of the plakophilin-3 protein.

In another aspect the invention provides a method for the stimulation of wound healing comprising providing skin cells of an affected skin area or other epithelial cells of other affected organs of an individual with a chemical compound affecting (i.e. decreasing or increasing) the expression or the functionality of the plakophilin-3 gene and/or the function of the plakophilin-3 protein. It is clear that the antibodies of the present invention can be used in a method for the treatment of skin diseases or for the stimulation of wound healing as indicated above.

In one aspect the invention provides a cell, and/or the progeny thereof, contacted with a nucleic acid delivery vehicle according to the invention.

In one aspect the invention provides a method for at least in part altering functional properties of a cell comprising providing said cell with an artificial pathway suitable for transmitting or modifying signals in a cell, said pathway comprising at least one catenin-like and at least one cadherin-like molecule and said method comprising providing said cell with an additional proteinaceous molecule capable of at least in part transmitting a signal of a cadherin/catenin-like signaling pathway.

A functional property of a cell that may be altered by a method of the invention may include but is not limited to a metastatic property, a neoplastic property, a wound healing property.

With the term "artificial pathway" as used herein is meant a pathway for transmitting signals in a cell not present in said cell prior to providing said cell with a means for said pathway. Preferably, said pathway leads to a modulation of the expression of a set of genes in said cell. Modulation of the expression of genes capable of modulating the expression of other genes is, of course, possible and for the present invention such proteins are considered to be part of the artificial pathway. A proteinaceous molecule of the invention may be any kind of proteinaceous molecule as long as said molecule is capable of transmitting or modifying a signal in a cadherin/catenin-like signal transduction pathway. For instance said proteinaceous molecule may comprise a peptide or a polypeptide. A (poly)peptide may be post-translationally and/or peri-translationally modified and/or may be synthesized artificially. Preferably said proteinaceous molecule comprises a catenin-like or a cadherin-like molecule or a functional part, derivative and/or analogue thereof. Preferably said proteinaceous molecule comprises plakophilin-3 or a functional part, derivative and/or analogue thereof. In the present invention novel plakophilin-3 proteins are presented, for which the protein expression pattern in humans seems to be largely restricted to epithelial cell types. The protein was localised in the desmosomal plaque and in the cell nucleus, and therefore it is likely to be involved in plasma membrane/cell nucleus signal transduction pathways. Sequence alignment with other p120^ct7plakophilin subfamily members suggests that plakophilin-3 might exert some specific functions, or a combination of different functions displayed by other family members.

In one embodiment the invention provides a plakophilin-3, which in humans comprises an amino acid sequence as depicted in figure 2 or a functional part, derivative and/or analogue thereof. A functional part of a plakophilin-3 of the invention may be the whole of the aminoterminal sequence stretches that are conserved between the plakophilin-3 proteins of man, mouse and Xenopus laevis, and which are not conserved in human plakophilin-1 or -2, and which are therefore unique for plakophilin-3 proteins. These conserved stretches are underlined in figure 4 and include the following amino acid residues (using the human plakophiiin-3 protein as reference): 6-13; 91-96; 119-127; 159-167; 173-189; 240-261 and 277- 284. Moreover, the mutual amino acid similarity of the Arm-repeat domains of plakophilin-3 proteins of man, mouse and Xenopus laevis is 85%, which is much higher than the similarity of the same domain between human plakophilins 1 , 2 and 3 (Table 2), which ranges from 51 % to 61 %. Also, the carboxyterminal domain of the plakophilin-3 proteins of man, mouse and Xenopus laevis is remarkably similar, i.e. about 90%. Such stringent interspecies conservation of sequence stretches, not present in plakophilin-1 and -2, may indicate that these regions fulfil an essential role in the maintenance of plakophilin-3-specific functions.

In another embodiment the invention provides a nucleic acid encoding a plakophilin-3 according to the invention, which in humans comprises a nucleic acid sequence as depicted in figure 2 or a functional part, derivative and/or analogue thereof.

In one embodiment the invention provides a plakophilin-3, which in mice comprises an amino acid sequence as depicted in figure 11 or a functional part, derivative and/or analogue thereof.

In another embodiment the invention provides a nucleic acid encoding a plakophilin-3 according to the invention, which in mice comprises a nucleic acid sequence as depicted in figure 10 or a functional part, derivative and/or analogue thereof.

In one embodiment the invention provides a plakophilin-3, which in Xenopus laevis comprises an amino acid sequence as depicted in figure 13 or a functional part, derivative and/or analogue thereof.

In another embodiment the invention provides a nucleic acid encoding a plakophilin-3 according to the invention, which in Xenopus laevis comprises a nucleic acid sequence as depicted in figure 12 or a functional part, derivative and/or analogue thereof. In another aspect the invention provides an antibody specific for plakophilin-3 of the invention according the invention, or a functional part, derivative and/or analogue thereof.

In another aspect the invention provides an anti idiotypic antibody of an antibody according to the invention or a functional part, derivative and/or analogue thereof.

An (anti-idiotype) antibody of the invention may be an antibody or a functional part, derivative and/or analogue thereof. A non-limiting example of a suitable part is a FAB-fragment. A non-limiting example of a suitable derivative is a single chain antibody. A non-limiting example of a suitable analogue is a synthetic antibody selected from a recombinant antibody library. A proteinaceous molecule with similar binding characteristics, in kind not necessarily in amount, an (anti-idiotype) antibody of the invention is a non-limiting example of a suitable analogue of an (anti- idiotype)antibody of the invention. Such a proteinaceous molecule may be obtained by performing screening assays in which many different proteinaceous molecules are tested for their binding specificity's. In a preferred embodiment an (anti- idiotype)antibody of the invention is a monoclonal antibody or a functional part, derivative and/or analogue thereof.

The invention specifically relates to the monoclonal antibodies 23E3/4 and

12B11F8, which are deposited with the Belgian Coordinated Collections of Microorganisms - BCCM™ represented by the Laboratorium voor Moleculaire Biologie - Plasmidencollectie (LMBP), University of Ghent, K.L. Ledeganckstraat 35, B - 9000 Ghent, Belgium on April 26, 2000 and have accession numbers LMPB 5482CB and LMBP 5481 CB, respectively.

In another aspect the invention provides a nucleic acid encoding an (anti- idiotype) antibody according to the invention, or a functional part, derivative and/or analogue thereof.

In yet another aspect the invention provides a nucleic acid delivery vehicle comprising a nucleic acid according to the invention. Providing a cell with a proteinaceous molecule of the invention is preferably performed by providing said cell with an expressible nucleic acid encoding said proteinaceous molecule and incubating said cell in order for said proteinaceous molecule to be expressed. A preferred method for providing a cell with a nucleic acid of the invention is by contacting said cell with a nucleic acid delivery vehicle comprising said nucleic acid and incubating said cell in order for said delivery vehicle to deliver said nucleic acid to said cell. Incubating a cell may be performed ex vivo or in a body.

In one embodiment the invention provides the use of a nucleic acid delivery vehicle according to the invention in a Gene Therapy application.

In another aspect the invention provides in a method for diagnosing a disease involving epithelial tissue comprising, obtaining a sample of cells of an affected area of the body and detecting the up- or down regulation of plakophilin-3 messenger RNA and/or the up- or down regulation of the level of plakophilin-3 protein and/or the presence or absence of a mutated plakophilin-3 and/or plakophilin-3 encoding nucleic acid, or a functional part, derivative and/or analogue thereof. Preferably said epithelial tissue comprises skin tissue.

In yet another aspect the invention provides a method for the treatment of skin disease comprising providing skin cells of an affected skin area of an individual with a nucleic acid delivery vehicle according to invention.

In yet another aspect the invention provides a method for the treatment of skin disease comprising providing skin cells of an affected skin area of an individual with a chemical compound or an antibody according to the present invention affecting the expression or functionality of the plakophilin-3 gene and/or the expression or function of the plakophilin-3 protein. In yet another aspect the invention provides a method for the stimulation of wound healing comprising providing skin cells of an affected skin area of an individual with a chemical compound or an antibody according to the present invention affecting the expression or functionality of the plakophilin-3 gene and/or the expression or function of the plakophilin-3 protein.

The dosage and mode of administration of a chemical compound or an antibody according to the present invention will depend on the individual. Generally, the medicament comprising a chemical compound and/or an antibody of the present invention and a pharmaceutically acceptable carrier of excipient is administered so that the chemical compound and/or antibody of the present invention is given at a dose between 1 μg/kg and 10 mg/kg, more preferably between 10 μg/kg and 5 mg/kg, most preferably between 0J and 2 mg/kg.

In one embodiment the invention provides a cell, and/or the progeny thereof, provided with a plakophilin-3 or a functional part, derivative and/or analogue thereof.

In another embodiment the invention provides a cell, and/or the progeny thereof, provided with a nucleic acid of the invention or a functional part, derivative and/or analogue thereof.

In yet another embodiment the invention provides a cell, and/or the progeny thereof, contacted with a nucleic acid delivery vehicle according to the invention.

In another embodiment the invention provides for the production of transgenic non-human animal models in which a mutant or wild type protein according to the invention is expressed, or in which a nucleic acid according to the invention has been inactivated for the study of skin diseases and/or cancer.

In still another aspect, the invention provides a protein or functional part, derivative and/or analogue thereof, that shows a binding capacity with a protein of the invention. In yet another aspect the invention provides for the production of transgenic non-human animal models in which a mutant or wild type protein according to the invention has been expressed, or in which said protein has been inactivated (knock-out deletion). Animal species suitable for use in the animal models of the present invention include, but are not limited to, rat, mice, hamster, guinea pigs, rabbits, dogs, cats, goats, sheep, pigs, and non-human primates. To create such animal model, a nucleic acid according to the invention can be inserted into a germ line or stem cell using standard techniques of oocyte microinjection or transfection or microinjection into embryonic stem cells. Similarly, if it is desired to inactivate or replace an endogenous gene, homologous recombination using embryonic stem cells may be employed.

In yet another aspect the invention provides a protein with a binding capacity to a protein according to the invention. In light of the present disclosure, one ordinary skill of the art is enabled to practice new screening methodologies which will be useful in the identification of proteins and/or other compounds which bind to, or otherwise directly interact with a protein according to the invention. As a non-limiting example, such screening methodology can consist of the so-called yeast two hybrid screening. The proteins and compounds identified will include endogenous cellular compounds, which interact with e.g. plakophilin-3 in vivo and which, therefore, provide new targets for pharmaceutical and therapeutic interventions.

In one aspect the invention describes a novel Armadillo protein, designated plakophilin-3, as it shows significant similarity to plakophilin-1 and -2 (Heid et al., 1994; Mertens et al., 1996; Schmidt et al., 1997). The human plakophilin-3 protein is encoded by a 2.8-kb mRNA and has a calculated M_τ of 87 kDa. It is composed of a central Arm repeat domain, containing 10 Arm repeats in the configuration (4+1 +1 +4), typical of the p120^ct plakophilin subfamily of Arm proteins (Reynolds and Daniel, 1997). The very short C-terminal domain (27-AA) is typical of all plakophilins reported so far. At the protein level the full-length mouse plakophilin-3 sequence (Genbank ID nr. AF136719; figure 11 ) is 94% identical and 96% similar to the human plakophilin-3 protein (figures 2 and 21 ). One plakophilin-3-specific EST clone present in public databases (Genbank-ID AA515691 ; origin: colon) contains an 81-bp insert with respect to our cDNA sequence. In silico translation results in an additional 27-AA fragment which displays no similarity to any known protein. We believe, however, this fragment to be intronic, as it displays typical intron features such as the gt/ag splice donor/acceptor sites, a branch site and a C/T stretch at its 3' end. Moreover, Western blot experiments on colon-derived cell lines did not reveal evidence of alternative splicing, and at least in small intestine mRNA this additional fragment was not detectable by an appropriate RT-PCR reaction.

The human plakophilin-3 gene (PKP3) was mapped to the chromosomal region 11 p15 by FISH. The gene is not linked to either the plakophilin-1 gene (PKP1), mapped on 1 q32, or the plakophilin-2 gene (PKP2) mapped on 12p13, or any other gene encoding a p120^c7plakophilin family member: p120^cln, ARVCF, p0071 and δ- catenin/NPRAP (Bonne et al., 1998). Hence, these mapping data do not reveal particular evolutionary links between these various gene family members.

Multiple sequence alignments of either the full-size protein sequences or Armadillo repeat regions revealed (1 ) that plakophilin-3 belongs to the p120^ct7plakophilin subfamily of Armadillo proteins, and (2) that the plakophilins themselves form a distinct subgroup within this subfamily. Although the overall sequence of plakophilin-3 is most related to those of plakophilin-1 and -2, in the Arm repeat region the protein shows equal similarity to each of the other p12O^ct7plakophilin family members. This might indicate that plakophilin-3 combines functions from these different proteins of this family, mirrored by several locally conserved protein regions. Alternatively, the non-conserved sequences of plakophilin-3 may specify particular functions of this protein not showed by the other p120^ct7plakophilin family members. Nearby their amino-terminus domain, the plakophilins contain a well-conserved sequence stretch, which we denoted the plakophilin HR2 or PKP-HR2 domain. For p120^ctn, the ARVCF gene product, p0071 and δ-catenin/NPRAP, a short conserved amino-terminal domain was previously denoted Homology Region 2 (HR2; Fig. 4) (Reynolds and Daniel, 1997). This HR2 is suggested to have a common function in the different proteins or to correspond with a common binding motif

(Reynolds and Daniel, 1997). An analogous domain albeit with low homology, is also present in the plakophilins. The function of these HR2 regions in the otherwise poorly conserved amino-terminal domain is unknown. The PKP2-HR2 sequence might, however, mediate an important biological function, as it is even conserved in sequence data we obtained from a Xenopus laevis plakophilin-3 cDNA clone (figure 21 ).

Using an anti-peptide rabbit polyclonal antibody, the plakophilin-3 expression pattern was investigated in a variety of cell lines. Several cell lines originating from epithelial tissues expressed plakophilin-3. We were, however, unable to detect expression of this new plakophilin in fibroblast-derived cell lines, including different sarcoma cell lines. This is apparently different from the expression patterns reported for plakophilin-1 and -2, which are widespread nuclear proteins, also detectable in nuclei of fibroblast cells (Mertens et al., 1996; Schmidt et al., 1997). The p120^ctn protein is also known to be widely expressed (Keirsebilck et al., 1998; Mo and Reynolds, 1996). In contrast, plakophilin-3 and the neural tissue-specific δ-catenin/NPRAP (Paffenholz and Franke, 1997) seem to be members of the p120^ct7plakophilin family whose expression is restricted to certain cell types, indicating that they might fulfil more specialised functions in these cell types.

Our immunodetection experiments revealed a dual location for plakophilin-3, i.e. desmosomal and nuclear, a situation which is also described in detail for plakophilin-1 and -2 (Mertens et al., 1996; Schmidt et al., 1997). Plasma membrane-associated plakophilin-3 was co-localized with desmoglein-2, a well-described desmosomal cadherin. Plakophilin-3 thus plays a mechanical role in desmosomes, assuring formation of tight cell-cell contacts. Mutations in or complete loss of desmosomal components, such as desmosomal cadherins or proteins of the intracellular desmosomal plaque, are responsible for loosened cell-cell adhesion. The desmosomal cadherins desmogleins and desmocollins are both necessary for proper desmosomal functionality (Chitaev and Troyanovsky, 1997). Decreased functionality of either desmoglein-1 or -3 can be caused by autoimmune diseases, resulting in affected keratinocyte cell-cell adhesion and ultimately skin blistering (Emery et al., 1995; Karpati et al., 1993; Koch et al., 1997). Mutations in the desmosomal plaque protein plakophilin-1 have recently been described too (McGrath et al., 1997). A patient with two independent mutations in the plakophilin-1 alleles, resulting in a functional knockout, displayed features of skin fragility and congenital ectodermal dysplasia, affecting skin, hair and nails. Remarkably, no significant abnormalities were detected in other epithelia or tissues. The phenotype described is apparently due to the loss of desmosome-associated plakophilin-1 , known to be present in stratified and complex epithelia (Schmidt et al., 1997). No defect in this patient could, however, be linked to the absence of the plakophilin-1 fraction, which normally is present in the nuclei of a wide variety of cell types and tissues (Schmidt et al., 1997). One may speculate that this is due to a functional redundancy of plakophilins-1 , -2 and -3 in the nucleus but not in desmosomes, although no evidence has been presented yet.

Nuclear plakophilin-3 was detected as bright speckles, dispersed throughout the nuclear volume but excluding the nucleoli. The presence of plakophilin-3 in the cell nucleus was confirmed by confocal microscopy. A similar nuclear appearance has been described for plakophilin-1 (Klymkowsky, 1999; Schmidt et al., 1997). The mechanism whereby the plakophilins are transported into the nucleus is still unclear, as these proteins lack obvious nuclear localization signals. Using green fluorescence protein-tagged human plakophilin-1 and fragments thereof, Klymkowsky (1999) investigated the nuclear accumulation of these proteins. Full-size plakophilin-1 accumulated almost exclusively in cell nuclei, both in Xenopus embryos and in epithelial Xenopus A6 cells, and was often present in similar nuclear structures as reported here for plakophilin-3. Also according to Klymkowsky (1999), both the amino-terminal head domain and the carboxy-terminal Armadillo repeat domain can enter nuclei, though the amino-terminal domain and the full-size polypeptide were present mainly in the nucleus of Xenopus embryos, while the Armadillo domain localized also in the cytoplasm with enrichment at the cell cortex and cell-cell contacts.

The occurrence of the novel plakophilin-3 in both desmosomes and nuclei of epithelial cells indicates that this Armadillo protein is involved in signal transduction pathways between the plasma membrane and the nucleus, besides a mechanical role in cell-cell attachment by desmosomes. Such a dual role is reminiscent to the one well documented for β-catenin, an Arm protein present in adherens junctions but also involved in the Wnt signalling pathway (Willert and Nusse, 1998). In the nucleus, β-catenin associates with LEF-1/TCF transcription factors and modulates gene transcription (Behrens et al., 1996; Huber et al., 1996; Molenaar et al., 1996). Target genes of the β-catenin/LEF-1 or β-catenin/TCF transcription factor complexes comprise c-myc, c-jun, fra-1, the genes encoding cyclin-D1 , urokinase-type plasminogen activator receptor or ZO-1 , the latter one being inhibited by the β- catenin containing transcriptional complex (He et al., 1998; Mann et al., 1999; Tetsu and McCormick, 1999). Nevertheless, injection of full-size plakophilin-1 mRNA had no obvious effect on early embryonic development, indicating that at least plakophilin-1 is not involved in the Wnt-like pathway in the early Xenopus embryo (Klymkowsky, 1999). The injection of mouse p120^ctn isoform 1 B into ventral Xenopus blastomeres was also reported not to mimic the Wnt pathway (Geis et al., 1998).

The present invention also relates to an isolated or recombinant nucleic acid encoding a promoter sequence of plakophilin-3, or a functional part, derivative and/or analogue thereof. The sequences of the human and mouse promoter are depicted in figure 23. The term 'promoter' refers to a combination of start sequence elements to which RNA polymerase binds in order to initiate transcription of a gene. The promoter of the present invention is further characterized by having SIP1 (see Remade et al., 2000 and Verschueren et al., 1999) and Snail (see Battle et al., 2000 and Cano et al., 2000) binding sites. The latter molecules downregulate plakophilin-3 expression and can thus be used to modulate plakophilin-3 mRNA levels in a cell as is exemplified further (see examples section).

EXAMPLES

Example 1 : Discovery and cloning of plakophilin-3

MATERIALS AND METHODS Database searches and DNA sequencing

BLASTP searches (Altschul et al., 1990) were performed at GenomeNet in Japan (http://www.blast.genome.ad.jp/). EST clones encoding unknown Arm-like proteins were ordered from the UK HGMP Resource Centre (Hinxton, UK) and from Genome Systems (St. Louis, MO). The sequences of the full-size inserts of these clones and all other DNA sequences were determined by an ABI 377 automated DNA sequencer (Applied Biosystems, Foster City, CA). Sequence data were processed using the DNAstar software (DNAstar, Madison, Wl) and the Staden Package (Bonfield et al., 1995); see also the Staden Package WWW site at http://www.mrc- lmb.cam.ac.uk/pubseq/.

Reverse transcription (RT) - PCR

Total RNA from human uterus was purchased from Clontech Laboratories (Palo Alto, CA). cDNA synthesis was performed as described (Keirsebilck et al., 1998). PCR primers were designed using the Oligo 5.0 Primer Analysis software (NBI, Plymouth, MN) and purchased from Gibco BRL system (Gibco BRL Life Technologies, Paisley, UK). PCR was performed with the Advantage GC KlenTaq Polymerase Mix (Clontech) on a PTC-200 Peltier Thermal Cycler PCR system (MJ Research, Watertown, MA). The PCR reaction mixture contained template cDNA, 25 pmol of each primer, 10 μl KlenTaq 5χ PCR buffer, 10 μl GC Melt, 4 μl 5 mM dXTP's and 1 μl KlenTaq in a final volume of 50 μl. Cycling conditions were 3 minutes at 94°C (initial denaturation), followed by 35 cycles of 30 seconds at 94°C, 45 seconds at 60°C and 2 minutes at 72°C. Final extension was for 10 minutes at 72°C. A plakophilin-3 specific RT-PCR fragment, used for Northern blot hybridisation and cDNA library screening, was generated accordingly. Forward primer sequence (5'-»3'): cgccctggtcacctctatca; reverse primer sequence (5'-»3'): ttctcactgtcggggctgtc. The predicted 774-basepair (bp) PCR product was checked on a 1 % agarose gel for the presence of a specific reaction products, followed by purification with the Qiaquick PCR purification kit (Qiagen, Chatsworth, CA) and sequencing.

Northern blot analysis

Total RNA was prepared with the RNeasy kit (Qiagen) following the manufacturer's protocol. Total RNA (25 μg) was glyoxylated, size-fractionated on a 1% agarose gel and transferred onto a Hybond-N⁺ membrane (Amersham Pharmacia Biotech, Rainham, UK). Hybridizations were performed as described before (Bussemakers et al., 1991 ). A 774-bp RT-PCR product (see above) was [³²P]-labeled using the RadPrime DNA labelling system (Life Technologies). For data collection Phosphorlmager 425 equipment was used (Molecular Dynamics, Sunnyvale, CA). 5' Rapid amplification of cDNA ends (5' RACE)

5' RACE experiments were performed using either the Marathon cDNA amplification kit (Clontech) or a 5' RACE system (Life Technologies) following the manufacturer's instructions and using different human messenger RNA (mRNA) sources. None of these experiments resulted in completion of the human plakophilin-3 mRNA sequence. Using the Marathon cDNA amplification kit only aspecific products were amplified, while the Gibco 5' RACE kit yielded specific, though very short products. The latter observation indicated that reverse transcription was suboptimal, probably due to the formation of secondary structures in the G/C-rich 5' end of the mRNA.

cDNA library screening

A human fetal kidney 5' stretch cDNA library in vector λDR2 (Clontech) was screened with a [³²P]-labeled 774-bp RT-PCR product as mentioned above. Five positive plaques were identified upon screening of approximately 800,000 plaques. After a second screening cycle, plaques were cut out and converted in vivo to pDR2-derived plasmids according to the manufacturer's instructions. Restriction digestion and sequence analysis revealed only one clone (13H5B) containing a full-length cDNA insert. To facilitate sequencing of the unknown 5' region of the cDNA, a BamHl restriction fragment of about 1 ,330 bp was subcloned into the pGEM-11Zf(+) vector (Promega, Madison, Wl) and sequenced. The fetal mouse Rapid-Screen cDNA library (OriGene, Rockville, MD) was screened by PCR according to the manufacturer's instructions. Forward primer sequence (5'→3'): tacccagcccactccaccta; reverse primer sequence (5'->3'): ttgcgtagctcatcatcctg. Reaction conditions of the PCR, which generates a 813-bp product, were as follows: 2 minutes at 95°C, followed by 35 cycles of 40 seconds at 94°C, 45 seconds at 60°C and 1 minute at 72°C. Final extension was for 5 minutes at

72°C. In vitro transcription/translation assay

The in vitro transcription/translation assay was performed using the TNT Coupled Reticulocyte Lysate System kit (Promega) according to the manufacturer's instructions. The full-length human plakophilin-3 cDNA was obtained by Hind\\\ digestion of clone pDR2 13H5B and ligated into the pGEM-11Zf(+) vector (Promega). Depending on the orientation of the insert, either T7 or SP6 RNA polymerase was used for transcription. Clone Q13 contained the full-length human plakophilin-3 cDNA under control of the T7 promoter. The translation products were [³⁵S]methionine labeled and separated by SDS-PAGE on an 8% gel followed by drying. The labeled proteins were detected using a Phosphorlmager 425 (Molecular Dynamics).

Human genomic DNA library screening

A BAC (Bacterial Artificial Chromosome) human genomic DNA library (Genome Systems) was screened by PCR as recommended by the supplier. The PCR reaction was performed with the Taqf PCR Core kit (Qiagen) supplemented with GC Melt (Clontech) on a PTC-200 Peltier Thermal Cycler PCR system (MJ Research). The reaction mixture contained 100 ng template DNA, 25 pmol of each primer, 5 μl 10χ PCR buffer, 2 μl 25 mM MgCI₂, 7 μl GC Melt, 1 μl 10 mM dXTP's and 0.5 μl (2.5 units) Taq DNA polymerase in a final volume of 50 μl. Forward primer sequence (5'-»3'): cgccctggtcacctctatca; reverse primer sequence (5'- 3'): tcgtcgtagaggcggtagga. PCR reaction conditions were as follows: 3 minutes at 94°C, followed by 35 cycles of 30 seconds at 94°C, 45 seconds at 60°C and 1 minute at 72°C. Final extension was for 5 minutes at 72°C. This reaction yielded a 192-bp product, containing an 81 -bp intron sequence. Products were analyzed on a 2% agarose gel which revealed one BAC clone (245A8) containing the screened region. DNA of this clone was purified using KB-100 columns (Magnums, Genome Systems). Chromosomal localisation of the human plakophilin-3 gene by fluorescent in situ hybridisation (FISH)

FISH analysis using BAC clone 245A8 specific for the human plakophilin-3 gene was performed according to standard procedures (Kievits et al., 1990) with some minor modifications. DNA of the BAC clone was biotinylated using the BioNick- kit (Life Technologies) according to the manufacturer's protocol. Fluorescent image results were captured by a Photometries Image Point CCD camera (Photometries, Mϋnchen, Germany) mounted on a Zeiss Axiophot microscope (Carl Zeiss, Jena, Germany). Image processing was performed and chromosome G-banding was obtained by reverse 4'-6-diamidino-2-phenylindole (DAPI) banding, using the MacProbe v3.4.1 software (Perceptive Scientific International, League City, TX).

Chromosomal localisation of the human plakophilin-3 gene using a monochromosomal cell hybrid mapping panel

PCR was performed on a monochromosomal cell hybrid mapping panel (NIGMS Human/Rodent Somatic Cell Hybrid Mapping Panel #2, Coriell Cell Repositories, Camden, NJ) using the same primers and PCR reaction conditions as for the BAC human genomic DNA library screening. All cell hybrid templates were diluted to a final DNA concentration of 100 ng/μl, using 1 μl as PCR template. PCR products were analyzed on a 2% LSI MP agarose gel (Life Science International, Zellik, Belgium).

Protein alignment and calculation of interprotein sequence similarities

AA similarities of protein sequences, aligned using the CLUSTAL W program (Higgins et al., 1996; Thompson et al., 1994), were calculated at the Belgian EMBnet Node (ben.vub.ac.be) by the Homologies program. The latter program is part of the EGCG software (egcg@embnet.org) and was written by Jack A.M. Leunissen (jackl@caos.kun.nl). For calculation of AA similarities, the Dayhoff table (Schwartz and Dayhoff, 1979) and the default threshold of comparison (0.6) were used. The alignments were shaded using the WWW-BOXSHADE server (http://ulrec3.unil.ch/software/BOX_form.html). The phylogenetic non-rooted trees were obtained using the TreeView program (Page, 1996). Bootstrap values of the phylogenetic trees were calculated by the CLUSTAL W program using default settings. Bootstrap values are commonly used to test tree branch reliability, and are calculated by resampling the data (aligned sequences) and predicting trees from these resampled sequences (Felsenstein, 1988). In this case, 1000 iterations were performed. For branches in the predicted tree topology to be significant, the resampling data sets should frequently predict the same branches. Bootstrap values greater than 700 are assumed to indicate reliable bifurcations.

Cell cultures

Most of the cell lines used were purchased from the American Type

Culture Collection (ATCC, Rockville, MD): colon adenocarcinoma cell lines HT29 (HTB-38), SW1116 (CCL-233), SW480 (CCL-228), SW620 (CCL-227) and LoVo (CCL-229), leiomyosarcoma cell line SK-LMS1 (HTB-88), epidermoid carcinoma cell line A431 (CRL-1555), osteosarcoma cells HOS (CRL-1543), liposarcoma cells SW872 (HTB-92), SV40-transformed lung fibroblasts WI-38-VA13-subline 2RA (CCL75J , abbreviated below as VA13) and breast carcinoma cells SK-BR-3 (HTB-30). Ileocecal adenocarcinoma cell lines HCT8/E8 and HCT8/R1 were obtained by subcloning of cell line HCT8 (CCL-224), where E stands for epitheloid and R for round cell variants (Vermeulen et al., 1995). Similarly, DLD1/R2/7 was a round cell variant subcloned from DLD1 (CCL-221 ) (Vermeulen et al., 1995; Watabe-Uchida et al., 1998). GLC34 is derived from a small cell lung carcinoma (De Leij et al., 1985). Cell lines LICR-HN2, LICR-HN3 and LICR-HN6 are derived from head and neck squamous cell carcinomas (Easty, 1981 ). MKN45 is a gastric carcinoma cell line (Motoyama and Watanabe, 1983), PC AA/C1 , abbreviated below as PC, is a colon adenocarcinoma-derived cell line (Paraskeva et al., 1984). MCF-7/AZ and MCF-7/6 cell lines are derived from the MCF-7 (HTB-22) human mammary carcinoma cell line (Bracke et al., 1991 ). HaCaT is a human keratinocyte cell line (Boukamp et al., 1988). FS4 is a human foreskin fibroblast cell line, and HEK293 is a human embryonic kidney fibroblast cell line.

Construction and transfection of a plasmid encoding plakophilin-3

Plasmid DNA from clone pDR2 13H5B was purified with a plasmid mini kit (Qiagen) and used as template in two PCR reactions, generating the appropriate cDNA fragments for in-frame ligation in the pEFHOBES eukaryotic expression vector downstream of an E-tag-encoding cDNA fragment. The pEFHOBES vector was a kind gift from M. Van de Craen (DMB, University of Gent, Belgium), consisting of the expression vector pEF-BOS (Mizushima and Nagata, 1990), in which the E-tag from plasmid pCANTAB5E (Pharmacia) is inserted. The following primer pairs were used (A) forward primer (5'-»3'), containing an additional Not\ site (underlined) aagcggccgcgcaggacggtaacttcctg, plus reverse primer (5'- 3') ctgaggaagccggtggcgttgtagaagat; and (B) forward primer (5'→3') gtgaagctcttcaaccacgccaaccag, plus reverse primer (5'→3'), containing an additional Kpn\ site (underlined): atggtaccacagccaacccccacctct. PCR was performed using the Advantage GC KlenTaq Polymerase Mix (Clontech) on a PTC-200 Peltier Thermal Cycler PCR system (MJ Research). The PCR reaction mixture contained approximately 50 ng plasmid DNA, 25 pmol of each primer, 10 μl KlenTaq 5χ PCR buffer, 10 μl GC Melt, 4 μl 5 mM dXTP's and 1 μl KlenTaq in a final volume of 50 μl. Cycling conditions were 3 minutes at 94°C (initial denaturation), followed by 25 cycles of 30 seconds at 94°C, 2 seconds at 96°C, 30 seconds at 66°C and 1 minute at 72°C. Final extension was for 5 minutes at 72°C. In this way, two partially overlapping fragments of respectively 1 ,456-bp (A) and 1 ,585-bp (B) were generated. These two fragments, which comprise the complete plakophilin-3 coding sequence besides start codon and part of the 3' UTR, contain a unique SatnHI site (position 1 ,329 of the plakophilin-3 cDNA) in their overlapping sequence. Hence, the PCR reaction products A and B were purified using the Qiaquick PCR purification kit (Qiagen), ligated in the pGEM-T vector (Promega) and electroporated in DH5α host bacteria. Plasmids were purified using the plasmid mini kit (Qiagen) and double-digested with Λ/o-l/SamHI (PCR product A), Kpn\/BamH\ (PCR product B); or Not\/Kpn\ (pEFHOBES vector). Finally, all fragments were purified and ligated together. In this construct, transcription of the cDNA is under control of the human EF1 α-promoter. Transfection was performed using Lipofectamin Reagent (Life Technologies) according to the manufacturer's instructions with some minor modifications.

Antibodies, antibody production and purification

Antibodies specific for both the human and mouse plakophilin-3 protein were raised by immunization of rabbits with 200 μg of a synthetic peptide with sequence NH₂-KLHRDFRAKGYRKED-COOH. This peptide was coupled to keyhole limpet hemacyanin via an additional cysteine residue at the NH₂-terminal end. Immunization was followed by boost injections after two weeks. After another two weeks, antisera were collected on a biweekly basis for one month and tested in ELISA assays, using the synthetic peptide. Antibodies were affinity-purified using the synthetic peptide covalently bound to p-hydroxymercuribenzoate-agarose (Sigma, St. Louis, MO). Purified antibodies were then tested on in vitro transcribed and translated products separated by SDS-PAGE and blotted as described. Recognition of plakophilin-3 by the antibodies was inhibited by incubation of the polyclonal antibody with the antigenic peptide for 1 hour prior to use. This pre-treatment also abrogated detection of plakophilin-3 when tested on Western blots.

The following mouse monoclonal antibodies were used: anti-desmoglein antibody DG3J 0 (Cymbus Bioscience, Southampton, UK), human keratin 18 antibody RGE 53 (Euro-Diagnostics, The Netherlands) and anti-plakophilin-1 and -2 antibodies (Progen, Heidelberg, Germany).

Secondary antibodies used in immunofluorescence microscopy were coupled to either FITC (Amersham Life Science, Buckinghamshire, UK), Alexa594 or Alexa488 (Molecular Probes, Eugene, OR) anti-rabbit Ig, anti-rat Ig or anti-mouse Ig antibodies. Western blot analysis

Total protein lysates were prepared by washing subconfluent cell cultures twice with 1 χ PBS, followed by scraping of the cells in 1 χ Laemmli sample (Laemmli, 1970) and sonication. Protein concentration was measured using the DC protein assay kit (Biorad, Richmond, CA). Forty μg of total protein were boiled in 5% 2-mercaptoethanol and separated by 8% SDS-PAGE. Proteins were transferred onto Immobilon-P membranes (Millipore, Bedford, MA) and blocked with 5% nonfat dry milk, 0.1% Tween-20 in PBS (pH 7.4) prior to incubation with the primary antibody. Secondary anti-rabbit Ig and anti-mouse Ig antibodies coupled to horseradish peroxidase (Amersham Pharmacia Biotech) were used for detection of proteins on Western blots. The ECL Western blotting detection system (Amersham Pharmacia Biotech) was used for detection of the secondary antibodies after extensive washing of the blots with PBS (pH 7.4).

Immunofluorescence assays of cultured cell lines

Cell cultures were grown on glass coverslips and briefly washed in PBS, 2 mM MgCI₂, 2 mM CaCI₂ (complete PBS) followed by fixation with either methanol or paraformaldehyde. In the first case, cells were treated with ice-cold 100% methanol, followed by a 10-minutes treatment with ice-cold methanol. After brief air-drying, coverslips were incubated for 5 minutes in complete PBS, 0.2% Triton X-100. The cells were then incubated for 1 hour at room temperature with primary antibodies diluted in 0.4% gelatin in complete PBS. Cells were washed 3 times for 10 minutes with complete PBS, followed by incubation of the appropriate secondary antibody diluted in 0.4% gelatin in complete PBS. Finally, cells were washed again and incubated with DAPI solution. Specimens were subsequently mounted with Vectashield (Vector Laboratories) to prevent photobleaching. Alternatively, cells were fixed for 20 minutes with 3% paraformaldehyde in complete PBS at room temperature. Fixation was followed by quenching in a 50-mM NH₄CI solution in complete PBS for 5 minutes and by permeabilization in complete PBS, 0.2% TX-100 for 5 minutes at room temperature. All antibodies were diluted in 0.3% skim milk in complete PBS and incubated for 1 hour at room temperature. After incubation with secondary antibodies, cells were washed again and mounted as described before.

Fluorescent image results were captured by a Photometries Image Point CCD camera (Photometrics-GmbH, Germany) mounted on a Zeiss Axiophot microscope, or by a Zeiss LSM 410 confocal laser-scanning immunofluorescence microscope. Image processing was performed using the MacProbe v3.4J software (Perceptive Scientific International LTD.). Alternatively, pictures were taken using a standard 35-mm camera.

RESULTS

Discovery and cloning of the human plakophilin-3 mRNA

In our search for novel Arm proteins, BLAST searches (Altschul et al., 1990) were performed in non-redundant databases using the coding sequences of various known Arm proteins. These searches revealed a number of human expressed sequence tags (EST) displaying significant sequence similarity with human p120^ctπ and plakophilin-1 proteins. Complete sequencing of the inserts of these cDNA clones allowed the assembly of a contig representing a 1 ,970-bp cDNA fragment, which may encode a protein fragment highly related to plakophilin-1 and -2. This sequence, still incomplete at the 5' end, appeared to be derived from a genuine mRNA with one long open reading frame (ORF) and a 3' untranslated region (UTR), containing a polyadenylation signal (AATAAA) 18-bp in front of a poly-A tail. To determine the total length of the corresponding mRNA, a Northern blot hybridization experiment was performed on total RNA extracted from various cell lines, some of which originated from the same tissue type as the EST clones used to construct the contig, for instance colon. This indicated the corresponding full-length mRNA to be approximately 3-kilobase (kb) in size (Fig. 1 ). However, despite several attempts, we were unable to clone the lacking 5' end of the mRNA by RACE experiments (see Materials and Methods). This is probably due to the relative high GC content of the 5' end of the mRNA (up to 85%, while its overall GC content is 67%), which stimulates the formation of secondary structures and thereby inhibits efficient reverse transcription (Zhang and Frohman, 1998). Therefore, a human fetal kidney cDNA library was screened using a 770-bp RT-PCR product as probe. This resulted in the identification of five specific cDNA clones of which one (clone 13H5B) contained a probable full-length insert. The completed 2,786-bp plakophilin-3 cDNA sequence (Genbank Accession No: AF053719) contains an ORF of 2,391 -bp, a very short 5' UTR (54-bp) and a 3' UTR of 341 -bp (Fig. 2). The ORF encodes a 797-AA protein, with a calculated M_r of 87 kDa. The ATG codon at position 55 is very likely the correct translation initiation site, as it lies in a sequence context favoring such initiation of translation and because no upstream ATG codons are present (Kozak, 1996; Kozak, 1997). This was consolidated by an in vitro transcription/translation assay, using this putative full-length plakophilin-3 cDNA as template. Only one major polypeptide was produced and this showed the correct predicted size (results not shown).

Cloning of the mouse plakophilin-3 mRNA

Starting from mouse EST clones, we sequenced and assembled a partial mouse plakophilin-3 cDNA, which was about 2,320-bp in size but still lacked the 5' end. In order to complete this sequence, a fetal mouse cDNA library was screened by

PCR. This resulted in the identification of two clones containing a full-length mouse plakophilin-3 cDNA insert. The 2,829-bp murine sequence (Genbank Accession No: AF136719) encodes a 797-AA protein, which is 94% identical and 96% similar to the human plakophilin-3 protein. The overall nucleotide identity between human and mouse plakophilin-3 mRNA was found to be 81 %.

Mapping of the human plakophilin-3 gene to chromosomal region 11 p15

We mapped the human plakophilin-3 gene (proposed gene symbol: PKP3) by FISH to chromosomal region 11 p15 (Fig. 3A). The assignment to chromosome 11 was confirmed by PCR performed on a human monochromosomal cell hybrid- mapping panel (Fig. 3B). The 192-bp P P3-specific fragment was detected only in the lanes containing the positive control and human chromosome 1 1. The PKP-3 gene was not found in GeneMap'98 (http://www.ncbi.nlm.nih.gov/genemap98) using "plakophilin" as search string (most of the plakophilin-3-specific EST clones present in public databases are annotated with "similar to plakophilin"). According to GeneMap'98 several EST clones annotated "plakophilin-2" were mapped to different chromosomes, but none to chromosome 11. However, the similarity of these EST clones to plakophilin-2 is probably based on an Alu repeat present in an exon, specific for the plakophilin-2b isoform (our unpublished observations).

Plakophilin-3, a novel member of the p120^ct7plakophilin subfamily of

Armadillo proteins

The plakophilin-3 protein contains a central Armadillo domain composed of 10 repeats preceded by a 293-AA amino-terminal region and a short (27-AA) carboxy-terminal region (Figs 2, 4). The AA sequence and the organization of these repeats (4+1+1 +4) are very similar to these of the proteins belonging to the p12O^ctNplakophilin Arm subfamily (Reynolds and Daniel, 1997). Multiple alignment of plakophilin-3 with the previously reported proteins of this subfamily, namely p120^ctn, the ARVCF protein, p0071 , δ-catenin/NPRAP, and plakophilin-1 and -2 was performed using the CLUSTAL W program (Higgins et al., 1996; Thompson et al.,

1994) (Figs 4, 5A). A phylogenetic tree generated by the CLUSTAL W alignment of the Arm-repeats of the p120^ct7plakophin family members, β-catenin, plakoglobin and the Drosophila protein Armadillo is presented as Fig. 5B. As evident from Fig. 5A, the plakophilins form a somewhat distinct subgroup within the p120^ct7plakophin protein family. The interprotein similarities of the full-size proteins and of the central Arm repeat regions are presented in, respectively, Tables 1 and 2. The plakophilin-3 protein shows equal similarity to both plakophilin-1 and -2 (44% to 45%), but also substantial similarity to the other Arm proteins (40%). The same observation holds true when identities instead of similarities were calculated (data not shown). However, markedly higher intermolecular resemblance is obvious for the p120^ctn and ARVCF proteins, plakophilin-1 and -2, and p0071 and δ-catenin/NPRAP (Table 1 , underlined). When the sequences of the central Arm repeat regions are compared to each other (Table 2, underlined), the former observations were largely corroborated. Together, these data indicate that three pairs of very similar proteins can be distinguished, i.e. the p120^ctn and the ARVCF proteins, which are also quite similar to the p0071 plus δ-catenin/NPRAP pair, and finally the more distantly related plakophilin-1 and -2 proteins. In between these protein pairs stands plakophilin-3, whose overall sequence is slightly more related to that of plakophilin-1 or -2, but whose Arm domain shows equal similarity to the Arm domains of any other p120^ct7plakophilin protein. Nevertheless, we preferred, for the sake of simplicity, to designate our new family member plakophilin-3. Moreover, both the amino- and carboxy-terminal domains of plakophilin-3 share some striking features with the other plakophilins, although this is hardly reflected by interprotein AA-similarity analysis (Tables 3 and 4). Besides the central Arm domain, a second region of homology can be observed, i.e. the plakophilin HR2 domain nearby the amino-terminus (Fig. 4). An analogous second region of homology has been described for the other p120^ct7plakophilin family members, but its sequence is very different from the plakophilin HR2 domain (Reynolds and Daniel, 1997). On the other hand, the short plakophilin-3 carboxy-terminal region downstream of the central Arm repeats is both in length and local sequence composition quite similar to that of plakophilin-1 and -2 (Fig. 4).

Western blot analysis of plakophilin-3

A rabbit polyclonal antibody was raised against a carboxy-terminal plakophilin-3-specific peptide and affinity-purified. The sequence of this 15-AA peptide is poorly conserved in either plakophilin-1 or -2 (Fig. 4), which makes cross-reactivity of the antibody with the latter plakophilins very unlikely. This was confirmed by a Western blot analysis using mouse monoclonal antibodies against plakophilin-1 and -2 and the rabbit polyclonal anti-plakophilin-3 antibody (Fig. 6A). In order to confirm that our plakophilin-3 cDNA was complete, HEK293 cells were transfected with a cDNA encoding a tagged plakophilin-3 protein with a predicted mass of 89 kDa. Untransfected HEK293 cells do not express endogenous plakophilin-3 protein (Fig. 6B, lane 1 ). Using the polyclonal anti-plakophilin-3 antibody the exogenously expressed 89-kDa plakophilin-3 was detected as a band with slightly slower migration in SDS-polyacrylamide gels than the 87-kDa endogenous plakophilin-3 from HaCaT cells (Fig. 6B, lanes 2 and 3). This result indicated our plakophilin-3 cDNA sequence to be correct.

Using the latter antibody, plakophilin-3 protein expression was investigated in a variety of human cell lines. Based on human and mouse EST clone data, the plakophilin-3 mRNA is expressed in skin, colon, mammary gland, fetal heart, placenta, ovary, thymus and T-cells. The keratinocyte cell line HaCaT was found to express strongly plakophilin-3 (Fig. 6C). Other epithelial cell lines expressing plakophilin-3 include the epidermoid carcinoma cell line A431 ; HT29, PC, LoVo, SW480, SW1116, HCT8/E8, all derived from colon, LICR-HN6 and LICR-HN3 derived from squamous carcinomas of head and neck, cell lines SK-BR-3 and MCF7 derived from mammary gland adenocarcinomas (Fig. 6C). Other cell lines did not detectably express the plakophilin-3 protein. These include human embryonic kidney cells (HEK293), FS4 fibroblasts and the SV-40 transformed fibroblastoid cell line VA13. The SK-LMS1 leiomyosarcoma, SW872 sarcoma and HOS osteosarcoma cell lines also did not express plakophilin-3 (Fig. 6C).

Immunodetection of nuclear and desmosome-associated plakophilin-3

In order to determine the intracellular location of plakophilin-3, immunostainings were performed on cultured human cell lines using the same rabbit polyclonal antibody mentioned above. The staining results varied dramatically depending on the fixation and permeabilization protocols used. In methanol-fixed human ileocecal adenocarcinoma cells HCT8/E8, plakophilin-3 was detected along cell-cell borders in a punctuate staining pattern typical for desmosomal proteins (Fig. 7A). A desmosome-like immunostaining was also detected in methanol fixed HaCaT cells (Fig. 7B). Co-localisation of the plakophilin-3 protein with the desmosomal protein desmoglein was demonstrated in HCT8/E8 cells, transfected with a cDNA plasmid encoding a full-length plakophilin-3 protein (Fig. 7C,D,E). When the plakophilin-3 antibody was pre-incubated prior to use with the peptide against it was raised, the punctuate staining along cell-cell borders was no longer detectable (result not shown). Methanol fixed HEK cells were negative for plakophilin-3 by immunofluorescence, in accordance with our Western blot experiments (Fig. 6).

In addition to the desmosomal immunolocalisation, immunostaining was observed as bright nuclear speckles in HCT8/E8 cells, especially upon appropriate focusing (Figs 7A, 8A). The plakophilin-3 specificity of this nuclear staining was shown by its complete inhibition upon pre-incubation of our antibody with the immunogenic peptide (Fig. 8B). Such plakophilin-3 detection in HCT8/E8 nuclei was insensitive to the fixation protocol used, whereas desmosomal staining was abolished by a paraformaldehyde fixation protocol used. The nuclear plakophilin-3 staining was not detectable in HEK cells (not shown). In contrast, it was easily detectable in paraformaldehyde fixed HaCaT cells (Fig. 8C,D), but poorly detectable in methanol fixed HaCaT cells. As can be seen in Fig. 8C, nucleoli are not stained. The nuclear presence of plakophilin-3 was confirmed by confocal microscopy on A431 cells, showing the presence of plakophilin-3 as nuclear spheres (Fig. 9).

Example 2: Production and Use of Mouse Anti-Human plakophilin-3 Monoclonal Antibodies

Generation of mouse anti-human plakophilin-3 antibodies

Using the Lasergene Protean software, the antigenicity and surface probability of the amino acids composing the human plakophilin-3 protein were analyzed. Furthermore, identities with plakophilin-1 and -2 proteins were checked. As a result, two plakophilin-3-specific peptides, #748 and #926, as depicted in Fig. 4, were selected for immunization of mice. Peptide #748 (one-letter amino acid sequence: KLHRDFRAKGYRKED) is located at the extreme carboxyterminus of the plakophilin-3 protein, whilst peptide #926 (sequence: FTPQSRRLRELPLAADALTF) comprises spacer region 3 between armadillo repeats 6 and 7 of the human plakophilin-3 protein and also small sequence parts of both repeats 6 and 7. An additional cysteine residue was added to the aminoterminus of the peptides to enable coupling to Keyhole limpet haemocyanin. Immunization of mice using these peptides was performed according to Schafer et al. (1996). Spleen cells of immunized mice were fused with cell line SP2/0-Ag14 and supematants were tested in ELISA assays using the appropriate antigenic peptide as coating. Supematants of clones that scored positive in the ELISA assays were used in a Western blot detection assay, allowing identification of the clones producing plakophilin-3-specific antibodies. Such clones were further subcloned, resulting in hybridomas producing genuine plakophilin-3-specific monoclonal antibodies. Two of such clones are described below.

Monoclonal hybridoma 12B11 F8, resulting from immunization with the human plakophilin-3-specific peptide #748

Hybridoma 12B11 F8 produces anti-human plakophilin-3 antibodies directed against the carboxyterminal epitope KLHRDFRAKGYRKED (Fig. 4), and was generated upon immunization of mice with peptide #748. This antibody is deposited with the Belgian Coordinated Collections of Microorganisms - BCCM™ represented by the Laboratorium voor Moleculaire Biologie - Plasmidencollectie (LMBP), University of Ghent, K.L. Ledeganckstraat 35, B - 9000 Ghent, Belgium on April 26, 2000 and has accession number LMBP 5481 CB. Western blot results show that supernatant from this clone specifically detects plakophilin-3, and does not crossreact with either plakophilin-1 or -2 (Fig. 14A). Furthermore, using this supernatant little or no background is observed in Western blot detection experiments. Detection of plakophilin-3 with supernatant from clone 12B11 F8 can specifically be inhibited by preincubation of the antibodies with peptide #748 against which it was raised (Fig. 14B). According to the Isostrip mouse isotyping kit (Boehringer Mannheim, Mannheim, Germany), the antibody isotype is lgG2a.

Supernatant of the 12B11 F8 hybridoma was also used in immunofluorescence detection assays performed on methanol-fixed cells from human ileocaecal adenocarcinoma HCT8/E8, human epidermoid carcinoma A431 and human keratinocyte HaCaT cell lines. Cells were permeabilized for 5 min. with 0.2% Triton X- 100 before incubation with primary antibody 12B11 F8. Plakophilin-3 was detected as a punctuate staining along cell-cell borders, which is typical for desmosomal components (Fig. 15). Also, some weak cytoplasmic immunoreactivity was observed, which has been reported before for plakophilin-3 immunolocalizations (Schmidt et al., 1999). Unlike the rabbit polyclonal anti-human plakophilin-3 antibody described above, monoclonal 12B1 1 F8 antibodies do not detect speckle-like nuclear structures possibly containing plakophilin-3 (Fig. 15), despite the fact that both polyclonal and monoclonal antibodies were raised against the same peptide.

Monoclonal hybridoma 23E3/4, resulting from immunization with the human plakophilin-3-specific peptide #926

Hybridoma 23E3/4 produces anti-human plakophilin-3 antibodies directed against the peptide FTPQSRRLRELPLAADALTF, covering mainly the spacer sequence between the armadillo repeats 6 and 7 of the plakophilin-3 protein (Fig. 4). It was generated upon immunization of mice with peptide #926. This antibody is deposited with the Belgian Coordinated Collections of Microorganisms - BCCM™ represented by the Laboratorium voor Moleculaire Biologie - Plasmidencollectie (LMBP), University of Ghent, K.L. Ledeganckstraat 35, B - 9000 Ghent, Belgium on April 26, 2000 and has accession number LMBP 5482CB. Western blot results show that supernatant from this clone specifically detects plakophilin-3, and does not crossreact with either plakophilin-1 or -2 (Fig. 16). Furthermore, using this supernatant little or no background is observed in Western blot detection experiments. According to the Isostrip mouse isotyping kit (Boehringer), the antibody isotype is lgG2b. Supernatant of the 23E3/4 hybridoma was used in immunofluorescence detection assays on methanol-fixed human ileocaecal adenocarcinoma HCT8/E8 cells. Cells were pretreated for 5 min. with 0.2% Triton X-100 before incubation with primary antibody 23E3/4. Plakophilin-3 was detected as a punctuate staining along cell-cell borders, which is typical for desmosomal components (Fig. 17A,B). The 23E3/4 antibody can also be used in immunofluorescent detection assays of plakophilin-3 in paraformaldehyde-fixed HCT8/E8 cells pretreated with 0.2% Triton X-100 for 15 min. prior to incubation with primary antibody (Fig. 17C). Unlike the polyclonal anti-human plakophilin-3 antibody described above, the present monoclonal antibodies do not detect speckle-like nuclear structures possibly containing plakophilin-3 (Fig. 17).

Concentration of the 23E3/4 hybridoma supematants by loading on Protein-G Sepharose resulted in antibody preparations that can be diluted up to 1 :4000 for immunofluorescent detection of plakophilin-3 (Fig. 18A). When the primary antibody was omitted, no signal could be detected (Fig. 18B).

Example 3: Cross-Reactivity of Monoclonal Antibodies 12B11 F8 and 23E3/4 with plakophilin-3 Proteins From Other Species Besides Man

Monoclonal antibodies 12B11 F8 and 23E3/4 yield double plakophilin-3 bands in Western blot experiments using protein samples from various mouse body parts

Mouse foot sole, ear and tail were dissected, dissolved in Laemmli buffer (Laemmli, 1970), mixed and used in Western blot detection assays. Antibodies 23E3/4 and 12B11 F8 clearly detected two bands in these mouse protein samples (Fig. 19A). The second band may result from alternative splicing in vivo, which needs further investigation by RT-PCR analysis and sequencing. However, these double signals may also result from a different phosphorylation status of various plakophilin-3 protein pools. In the human cell line A431 , clearly only one band can be observed (Fig. 19A). Detection of plakophilin-1 and -2 proteins using antibodies obtained from Progen results in signals that differ in electrophoretic mobility and expression pattern from the plakophilin-3 proteins (Fig. 19B). Use of secondary antibody only results in absence of any signal. Monoclonal antibody 20C10D3 (Fig. 19A) is directed against the same peptide as 12B11 F8 and is included as an additional control. The 12B11 F8 but not the 23E3/4 antibody detects the Xenopus laevis plakophilin-3 protein

The full length Xenopus laevis plakophilin-3 cDNA was cloned in the pGEM11 vector (Promega), yielding plasmid GB3. In vitro transcription/translation of this plasmid was performed using the TnT Coupled Reticulocyte Lysate System (Promega). As illustrated in panel A of Fig. 20, the 23E3/4 antibody does not detect the Xenopus laevis plakophilin-3 TnT product, while a similarly made human plakophilin-3 TnT product is clearly detected. However, both 12B11 F8 and 20C10D3 antibodies can be used to detect the Xenopus laevis plakophilin-3 protein (Fig. 19A). An autoradiograph obtained from the TnT products clearly shows that both human and Xenopus laevis plakophilin-3 proteins were synthesized efficiently (Fig. 20B).

As can be concluded from the alignments of the corresponding human, mouse and Xenopus laevis plakophilin-3 proteins (Fig. 21 ), the sequence of peptide #748 is much more conserved between those species than the amino acid sequence of peptide #926. This might explain why the 23E3/4 antibody does not recognize the

Xenopus laevis plakophilin-3 protein.

Example 4: The mouse PKP3 gene: sequence and organization, and analysis of the human and mouse PKP3 promoter

Sequence and exon/intron structure of the mouse PKP3 gene

The mouse PKP3 gene has been fully sequenced by us and so far 11 ,689 basepairs (bp) of sequence are available (Fig. 22). Of these, 2,149-bp comprise 5' nontranscribed region, and 365-bp are 3' nontranscribed region. The gene consists of 13 exons and 12 introns, the sizes of which are presented in Table 5. All introns follow the gt-ag splicing rule. Comparison of human and mouse promoter sequences reveals possible deltaEF-1/SIP1 and Snail binding sites

The nontranscribed upstream sequences 5' proximal of the human and mouse PKP3 genes contain stretches of conserved residues, as depicted in the Clustal alignment of both sequences (Fig. 23). The double underlined human sequence is also transcribed according to Schmidt et al. (1999), and therefore no part of the promoter, although it was not found in the human plakophilin-3 cDNA clone characterized by us (Bonne et al., 1999). One remarkable feature is the presence in both human and mouse PKP3 promoters of three possible SIP1 binding sites (Remade et al., 1999; Verschueren et a!, 1999), of which two are genuine E-boxes (Fig. 23) and therefore possible Snail binding sites (Batlle et al., 2000; Cano et al., 2000). Expression of SIP1 induces invasion and loss of cell aggregation by transcriptional downregulation of the epithelium-specific gene E-cadherin (Comijn et al., manuscript in preparation). Since plakophilin-3 expression is restricted to epithelial cell types, and since SIP1 and Snail binding site consensus sequences are present in both mouse and human promoters, expression of plakophilin-3 is negatively regulated by SIP1 and Snail in nonepithelial tissues and in epithelial tumors with downregulated E-cadherin levels.

Evidence for negative regulation of PKP3 gene expression by SIP1

The canine MDCK-Tetoff cell line (Clontech) was stably transfected with an inducible SIP1 -encoding expression plasmid to create the MDCK-Tetoff-SIP1 cell line (Comijn et al., in preparation). The SIP1 cDNA in this construct is under control of a tTA-dependent promoter. The MDCK-Tetoff cell line expresses the tet-off transactivator, tTA (Gossen et al., 1995), which is unable to activate the tTA- dependent promoter in the presence of tetracycline. In the absence of tetracycline, however, tTA can activate this promoter and subsequently induces SIP1 expression. In consequence, E-cadherin mRNA and protein expression are downregulated in MDCK-Tetoff-SIP1 cells without tetracycline (Comijn et al., in preparation). This downregulation is initiated at the mRNA level through repression of E-cadherin promoter activity by SIP1 binding.

MDCK cells do express plakophilin-3, as expected for epithelial cells, as detected by the monoclonal antibodies 23E3/4, 12B11 F8 and 20C10D3 (Fig. 24A). However, upon activation of SIP1 expression (MDCK-Tetoff-SIP1 -tet), plakophilin-3 protein expression is downregulated dramatically (Fig. 24A). Unlike plakophilin-3, plakophilin-2 expression is far less affected (Fig. 24B). It has been described that plakophilin-3 expression is conferred to epithelial cell types, while plakophilin-2 is abundantly expressed in a wide variety of cell types (Bonne et al., 1999; Mertens et al., 1996).

Moreover, several cell lines have been identified by us that display an inverse correlation between on the one hand E-cadherin and plakophilin-3 mRNA expression, and on the other hand SIP1 mRNA expression (Table 6). We previously showed that cell lines FS4 and SK-LMS1 do not express the plakophilin-3 protein, whilst cell lines HCT8/E8, MCF7 and SW480 do express plakophilin-3 protein (Bonne et al., 1999). In Fig. 24, it is shown that colon adenocarcinoma COLO320DM cells do not express the epithelium-specific proteins plakophilin-3 and E-cadherin (C,D), but express the widespread plakophilin-2 protein (B). To further investigate the possible regulation of the plakophilin-3 promoter by SIP1 , the mouse plakophilin-3 promoter is cloned in reporter constructs and effects of SIP1 expression on the promoter activity are studied.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1. Detection of plakophilin-3 mRNA by Northern blot hybridization of total RNA extracted from various human cell lines.

Fig. 2. Sequences of the human plakophilin-3 cDNA and protein. Both start and stop codons of the 2,786-bp cDNA are boxed, as well as the AATAAA polyadenylation signal. Translation of the open reading frame of 2,391 -bp results in a

797-AA polypeptide. The central Arm-domain of 10 repeats is interrupted by three short sequence stretches (indicated as spacers 1 to 3 in Fig. 5), i.e. between repeats 4 and 5, between repeats 5 and 6, and between repeats 7 and 8.

Fig. 3. Mapping of the human plakophilin-3 gene (PKP3) to chromosomal region 11 p15. (A) Typical FISH result (arrows point at gene-specific hybridisation signals). (B) Gene localisation on chromosome 11 idiogram. (C) PCR amplification of a 192-bp plakophilin-3-specific fragment using genomic DNA from a human monochromosomal cell-hybrid mapping panel as template. PCR was performed on cell hybrids containing human chromosomes 1 , 2, 5, 11 , 12, or 22; +, total human genomic DNA used as positive control template; -, total mouse genomic DNA; -, total hamster genomic DNA; — , no template added; *, marker.

Fig. 4. Box shaded Clustal W alignment of the human plakophilin-1 a, -2a and -3 proteins displaying the interprotein identities (boxed black) and similarities (boxed gray). The high interprotein similarity in the Arm repeat domain is obvious, while the amino-terminal and the very short carboxy-terminal regions are less well conserved. The sequences of the carboxyterminal plakophilin-3 peptide #748 (KLHRDFRAKGYRKED) and the centrally located peptide #926 (FTPQSRRLRELPLAADALTF) are underlined. From this alignment it is clear that the selected plakophilin-3 peptides show little sequence identity with either plakophilin-1 or -2. Peptide #748 was used for the generation of both rabbit polyclonal and mouse monoclonal antibodies.

Fig. 5. Graphical display of CLUSTAL W alignment results using a TreeView program (Page, 1996). The lengths of the branches are drawn proportional to the number of sequence changes between proteins; scale bar: 1 AA substitution per 10 AA residues. Bootstrap values indicate the reliability of each bifurcation (1000 iterations performed). (A) Phylogenetic tree of the full-size human proteins of the p^O^/plakophilin subfamily of Arm proteins. Database Accession Nos: ARVCF: U51269; p120^ctn isoform 1ABC: AF062341 ; δ-catenin/NPRAP: U96136 or AB013805; p0071 : X81889, plakophilin-1 a: Z34974; plakophilin-2a: X97675; plakophilin-3: AF053719. (B) Phylogenetic tree of the central Arm domains of human proteins of the p120^ct7plakophilin subfamily, human β-catenin and plakoglobin, and the Drosophila Armadillo protein. Additional Database Accession Nos: β-catenin: X87838; plakoglobin: Z68228; Armadillo: X54468.

Fig. 6. Western blot analysis using a plakophilin-3-specific anti-peptide polyclonal antibody. (A) Immunodetection of plakophilin-1 , -2 and -3 in a total protein lysate of HaCaT cells. Lane 1 , plakophilin-1 ; lane 2, plakophilin-2 (isoform a and b); lane 3, plakophilin-3. (B) Immunodetection of a 89-kDa exogenous plakophilin-3 in transfected HEK293 cells as compared to the 87-kDa endogenous plakophilin-3 from HaCaT cells. Protein lysates were prepared from: lane 1 , HEK293 cells; lane 2, HEK293 cells transfected with a cDNA encoding a plakophilin-3 protein fused to an amino-terminal tag; lane 3, HaCaT cells. (C) Immunodetection of the human plakophilin-3 protein in various cell lysates. The latter were loaded as equal protein amounts.

Fig. 7. Immunofluorescence microscopy of endogenously and exogenously expressed plakophilin-3 in human epithelial HCT8/E8 and HaCaT cells, using a polyclonal plakophilin-3-specific antibody. The plakophilin-3 protein co-localises with the desmosomal cadherin desmoglein-2. (A,B) Endogenously expressed plakophilin-3 is detected as a linear punctuate pattern along cell-cell contacts, which is reminiscent of the typical expression pattern of desmosomal components. Note the bright fluorescent speckles in the nucleus, which are plakophilin-3-specific. (C-E) Double immunofluorescence for desmoglein-2 (C) and overexpressed plakophilin-3 (D) in HCT8/E8 cells transfected with a plasmid encoding plakophilin-3. (E) Merged image.

Fig. 8. Nuclear localisation of plakophilin-3 by immunofluorescence in methanol-fixed HCT8/E8 cells (A,B) or paraformaldehyde-fixed HaCaT cells (C,D). Cell nuclei were stained with DAPI. (A) Detection of plakophilin-3 as spherical nuclear particles. Due to focusing at the nuclear level, desmosome-associated plakophilin-3 is not visible. (B) Pre-incubation of the antibody with the antigenic peptide abrogates completely the detection of nuclear plakophilin-3. (C) Immunofluorescence detection of nuclear plakophilin-3, clearly leaving the nucleoli unstained. (D) After incubation with secondary antibodies only, no specific signals were detected. Fig. 9. Detection of nuclear plakophilin-3 by confocal laser-scanning microscopy. Nuclear plakophilin-3 is detected as speckles surrounded by cytoplasmic keratin.

Fig. 10. Mouse plakophilin-3 cDNA sequence. The ATG-codon (bold and underlined) for translation initiation is at position 93. The stop codon TAG (bold and underlined) is at position 2484. The ORF is from 93 to 2483 and encodes a 797-AA protein. A poly-adenylation signal (bold and underlined) is present 15-bp in front of the poly-A tail.

Fig. 11 Mouse plakophilin-3 protein sequence. The amino-, carboxy- and central Armadillo domains of this protein are indicated in Fig. 4.

Fig. 12. Xenopus laevis plakophilin-3 cDNA sequence. The ATG-codon

(bold and underlined) for translation initiation is at position 248. The stop codon TGA (bold and underlined) is at position 2711. The ORF is from 248 to 2710 and encodes a 821-AA protein.

Fig. 13. Xenopus laevis plakophilin-3 protein sequence. The amino-, carboxy- and central Armadillo domains of this protein are indicated in Fig. 4.

Fig. 14. Detection of plakophilins in Western blot experiments. (A) Western blot detection of plakophilin-1 (lane 1), plakophilin-2 isoform a and b (broad band in lane 2) and plakophilin-3 (lane 3) in total protein lysates of HaCaT cells. Calculated molecular weights are 81 kDa (plakophilin-1 ), 93 and 98 kDa (plakophilin -2a and -2b) and 87 kDa (plakophilin-3). Plakophilin-1 and -2 antibodies were obtained from Progen (Heidelberg, Germany). Plakophilin-3 antibody was 12B11 F8 supernatant that clearly does not crossreact with either plakophilin-1 or -2. All primary antibodies were diluted 1 :10. (B) Specific inhibition of plakophilin-3 recognition by 12B11 F8 hybridoma supernatant in Western blots by competition with the antigenic peptide #748. Detection of plakophilin-3 in A431 (lane 1 ) and HaCaT (lane 3) total protein lysates by 12B11 F8 antibodies was abrogated upon preincubation of the antibodies with the antigenic peptide #748 at a final concentration of 0.01 μg/μl (lane 2, A431 ; lane 4, HaCaT). Secondary antibody was alkaline phosphatase-conjugated goat anti- mouse IgG antibody (Sigma). Molecular weight markers are indicated by dots and expressed in kDa.

Fig. 15 Immunofluorescent detection of endogenously expressed plakophilin-3 in methanol-fixed human epithelial HCT8/E8 (A, B), A431 (C) and HaCaT cells (D) using 1 :10 diluted supernatant from hybridoma 12B11 F8. Endogenously expressed plakophilin-3 is detected as a linear punctuate pattern along cell-cell contacts. This corresponds with the typical expression pattern of desmosomal components. Also, some cytoplasmic immunoreactivity is observed. In (D), cell nuclei were stained with DAPI. Secondary antibody was goat anti-mouse IgG FITC-conjugated antibody (Amersham Life Sciences).

Fig. 16 Western blot detection of plakophilin-3 using supernatant produced by hybridoma clone 23E3/4. Plakophilin-3 is detected in total protein lysates from either HEK293 cells transfected with an eukaryotic expression plasmid encoding human plakophilin-3 (lane 1 ), or from A431 cells (lane 3). Untransfected HEK293 cells do not express plakophilin-3 (lane 2), as shown before (Bonne et al., 1999). The electrophoretic mobilities of plakophilin-1 and -2 proteins are, respectively, higher and lower than this of plakophilin-3 (see Fig. 14A). Primary antibody dilution was 1 :100. Secondary antibody was alkaline phosphatase-conjugated goat anti-mouse IgG antibody (Sigma). Molecular weight markers are indicated by dots and expressed in kDa.

Fig. 17 Immunofluorescent detection of endogenously expressed plakophilin-3 in methanol- (A,B) or paraformaldehyde-fixed (C) human epithelial HCT8/E8 cells, permeabilized with Triton X-100, using monoclonal supernatant 23E3/4. Plakophilin-3 appears in a linear punctuate pattern along cell-cell contacts, which is reminiscent of the typical expression pattern of desmosomal components.

Also, some cytoplasmic immunoreactivity is observed. Primary antibody dilution in these experiments was 1 :5. Secondary antibodies used were alexa594-conjugated goat anti-mouse IgG (Molecular Probes) (A,C) or FITC-conjugated goat anti-mouse IgG (Amersham Life Sciences) (B).

Fig. 18 Immunofluorescent detection of endogenously expressed plakophilin-3 in methanol-fixed human epithelial HCT8/E8 cells, using Protein-G Sepharose concentrated monoclonal supernatant 23E3/4. (A) Endogenously expressed plakophilin-3 is detected as a linear punctuate pattern along cell-cell contacts. Also, some cytoplasmic immunoreactivity is observed. Primary antibody dilution in this experiment was 1 :4000. Secondary antibody was FITC-conjugated goat anti-mouse IgG antibody (Amersham Life Sciences). (B) No signal is detected when the primary antibody is omitted.

Fig. 19 Western blot detection of plakophilin-3 from mouse and Xenopus laevis using various monoclonal antibodies. (A) Antibodies 23E3/4, 12B11 F8 and 20C10D3 all detect double plakophilin-3-specific signals in protein lysates of mouse foot sole, ear and tail. Antibodies 12B11 F8 and 20C10D3 clearly also detect a Xenopus laevis plakophilin-3 protein, produced by a coupled in vitro transcription/translation reaction (TnT, Promega) performed on plasmid GB3. (B) Using plakophilin-1 and -2-specific antibodies (Progen), bands with clearly different electrophoretic mobility and expression pattern were obtained. Upon omission of primary antibodies, no specific signals are detected. Lanes indicated by — contain no protein lysates. Molecular weight markers are indicated by dots and expressed in kDa. Secondary antibody was alkaline phosphatase-conjugated goat anti-mouse IgG antibody (Sigma).

Fig. 20 Inability of the 23E3/4 antibody to recognize Xenopus laevis plakophilin-3 in Western blot experiments. (A) The 23E3/4 antibody does not detect a Xenopus laevis plakophilin-3-specific TnT product, while a TnT product of human plakophilin-3 is clearly and specifically detected. (B) An autoradiograph of the same blot clearly shows efficient translation of both human and Xenopus laevis plakophilin-

3 proteins. Fig. 21 Clustal W alignment of plakophilin-3 proteins. hPKP3, mPKP3, xPKP3 are human, mouse and Xenopus laevis plakophilin-3 protein sequences, respectively. Identical residues are boxed in black, similar residues in gray. The sequence of peptide #748, against which both 12B11 F8 and 20C10D3 antibodies are directed, is much more conserved across these species than the #926 peptide against which the 23E3/4 antibody is directed. This might explain the absence of recognition of the Xenopus laevis plakophilin-3 protein by the latter.

Fig. 22 Sequence of the mouse PKP3 gene and surrounding genomic fragments. Exons are boxed in black. Both the start codon in exon 1 and the stop codon in exon 13 are boxed in gray. Repeats were found with the RepeatMasker software (http://ftp.genome.washington.edu/cgi-bin/RepeatMasker). The program was set to identify simple repeats, rodent SINEs, LINEs, MIR and LINE2 repetitive sequences, retroviral sequences, tough LINEI s and low complexity DNA. All DNA repeats found here are confined to introns, and are depicted in bold and double underlined.

Fig. 23 Clustal alignment of mouse (mPKP3) and human (hPKP3) plakophilin-3 promoter sequences. Conserved DNA residues are boxed in black. The double underlined human sequence is assumed to be transcribed as well (Schmidt et al., 1999), though it was not present in our human plakophilin-3 cDNA clone described earlier (Bonne et al., 1999). Boxed in gray: three possible SIP-1 binding sites (CACCT or AGGTG), fully conserved between mouse and human sequences, of which two are genuine E-boxes (CACCTG or CAGGTG), known to be possible Snail binding sites.

Fig. 24 Detection of plakophilin-2, plakophilin-3 and E-cadherin in protein lysates of various cell lines. (A) The MDCK-Tetoff cell line, stably transfected with an expression vector containing the SIP1 cDNA under control of the tTA-dependent promoter (MDCK-Tetoff-SIP1 ), does not express plakophilin-3 when SIP1 expression is induced (MDCK-Tetoff-SIP1 -tet). However, in the absence of exogenous SIP-1 expression (MDCK-Tetoff-SIP1 +tet), the plakophilin-3 protein is expressed. Plakophilin-3 protein was revealed in Western blots using three different monoclonal antibodies. (B) The effects on plakophilin-2 protein expression are much less dramatic. COLO320DM cell line expresses the plakophilin-2 protein (B), but lacks both plakophilin-3 (C) and E-cadherin protein (D). Cell lines HaCaT and A431 are included as positive controls. Monoclonal antibodies used were from Progen (anti- plakophilin-2) or Zymed (anti-E-cadherin antibody HECD-1 , Zymed Laboratories Inc., San Francisco, CA). All lysates were loaded as equal protein amounts. Molecular weight markers are indicated by dots and expressed in kDa. Secondary antibody was alkaline phosphatase-conjugated goat anti-mouse IgG antibody (Sigma).

TABLES

Table 1. Interprotein AA similarities of the full-length proteins of the p120^ct7plakophilin subfamily. pp-1 = plakophilin-1 ; pp-2= plakophilin-2 ; pp-3= plakophilin-3

ARVCF _{p 1 20}ctn NPRAP p0071 pp-1 pp-2 PP-3

ARVCF 100% 65.7% 51.8% 50.2% 40.8% 40.4% 42.7%

p_{1 20}ctn 100% 50.7% 48.4% 40.4% 38.7% 41.1 %

NPRAP 100% 63.3% 39.6% 40.7% 41.3%

P0071 100% 41.1 % 39.9% 41.6%

pp-1 100% 53.5% 44.3%

pp-2 100% 44.9%

pp-3 100% Table 2. Interprotein AA similarities of the Arm repeat regions of the proteins belonging to the p120^ct7plakophilin subfamily. pp-1 = plakophilin-1 ; pp-2= plakophilin-2 ; pp-3= plakophilin-3

ARVCF p120^ctn NPRAP p0071 pp-1 pp-2 PP-3

ARVCF 100% 75.3% 68.1 % 65.0% 48.1 % 51.1 % 52.0%

_{p 1 20}ctn 100% 68.6% 65.7% 49.5% 50.6% 51.3%

NPRAP 100% 81.5% 50.1 % 53.0% 53.8%

p0071 100% 51.7% 51.5% 52.8%

pp-1 100% 60.9% 51.9%

pp-2 100% 51.0%

pp-3 100%

Table 3. Interprotein AA similarities of the amino-terminal regions (upstream of the central Arm domain) of the proteins belonging to the p120^ct7plakophilin subfamily. pp-1 = plakophilin-1 ; pp-2= plakophilin-2 ; pp-3= plakophilin-3

ARVCF _{p 1 20}ctn NPRAP p0071 pp-1 pp-2 PP-3

ARVCF 100% 59.6% 37.9% 28.8% 29.3% 26.9% 29.9%

p120^ctn 100% 39.9% 32.1 % 27.9% 26.3% 29.6%

NPRAP 100% 41.5% 27.8% 23.2% 29.1 %

p0071 100% 24.7% 23.1 % 27.1 %

pp-1 100% 34.5% 30.2%

pp-2 100% 35.3%

PP-3 100%

Table 4. Interprotein AA similarities of the carboxy-terminal regions (downstream of the central Arm domain) of the proteins belonging to the p120^c,7plakophilin subfamily.

ARVCF p120^ctn NPRAP p0071 pp-1 pp-2 pp-3

ARVCF 100% 41.3% 39.0% 38.5% 46.2% 29.7% 38.5%

_{p 1 20}ctn 100% 27.5% 33.3% 38.5% 32.4% 46.2%

NPRAP 100% 56.9% 35.9% 45.9% 46.2%

p0071 100% 46.2% 32.4% 46.2%

pp-1 100% 33.3% 44.4%

pp-2 100% 48.1 %

PP-3 100%

Table 5. Overview of exon and intron lengths of the mouse PKP3 gene.

No. Exon size (bp) Intron size (bp)

1 324 2240

2 80 133

3 632 76

4 124 668

5 205 1819

6 175 473

7 118 79

8 171 87

9 186 320

10 154 170

11 193 94

12 88 187

13 379 —

Table 6. Comparison of E-cadherin, SIP1 and plakophilin-3 mRNA expression in various human cell lines. Levels of mRNA were measured by Northern blot analysis. ^a COLO320DM is a colon adenocarcinoma cell line, FS4 is derived from foreskin fibroblasts, SK-LMS1 is leiomyosarcoma-derived, MCF7 is derived from a breast carcinoma and SW480 is a colon adenocarcinoma cell line. ^b +: corresponding mRNA can be detected, -: corresponding mRNA is not detected.

REFERENCES

Aberle, H., Butz, S., Stappert, J., Weissig, H., Kemler, R. and Hoschuetzky, H. (1994). Assembly of the cadherin catenin complex in vitro with recombinant proteins. J. Cell Sci. 107, 3655-3663.

Altschul, S. F., Warren, G., Miller, W., Myers, E. W. and Lipman, D. J. (1990). Basic local alignment search tool. J. Mol. Biol. 215, 403-410.

Batlle, E., Sancho, E., Franci, C, Dominguez, D., Monfar, M., Baulida, J. and deHerreros, A. G. (2000). The transcription factor Snail is a repressor of E- cadherin gene expression in epithelial tumour cells. Nature Cell Biology 2, 84-89.

Behrens, J., von Kries, J. P., Kϋhl, M., Bruhn, L., Wedlich, D., Grosschedl, R. and Birchmeier, W. (1996). Functional interaction of beta-catenin with the transcription factor LEF-1. Nature 382, 638-642.

Bonfield, J. K., Smith, K. F. and Staden, R. (1995). A new DNA sequence assembly program. Nucleic Acid Res. 23, 4992-4999.

Bonne, S., van Hengel, J. and van Roy, F. (1998). Chromosomal mapping of human armadillo genes belonging to the p120^ctn/plakophilin subfamily. Genomics 51 , 452-454.

Bonne, S., van Hengel, J., Nollet, F., Kools, P. and van Roy, F. (1999). PIakophilin-3, a novel Armadillo-like protein present in nuclei and desmosomes of epithelial cells. J. Cell Sci. 112, 2265-2276.

Bornslaeger, E. A., Kowalczyk, A. P., Meng, J. J., Ip, W. and Green, K.

J. (1997). The role of the desmosomal plaque protein desmoplakin in intermediate filament anchorage and junction assembly. In Cytoskeletal-membrane interactions and signal transduction, (ed. P. Cowin and M. W. Klymkowsky), pp. 181-198. Austin,

Texas (USA): Landes Company and Chapman & Hall.

Boukamp, P., Petrussevska, R. T., Breitkreutz, D., Hornung, J., Markham, A. and Fusenig, N. E. (1988). Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line. J. Cell Biol. 106, 761-771. Bracke, M. E., Van Larebeke, N. A., Vyncke, B. M. and Mareel, M. M.

(1991 ). Retinoic acid modulates both invasion and plasma membrane ruffling of MCF-7 human mammary carcinoma cells in vitro. Br. J. Cancer 63, 867-872. Bussemakers, M. J. G., Vandeven, W. J. M., Debruyne, F. M. J. and Schalken, J. A. (1991 ). Identification of High Mobility Group Protein l(Y) as potential progression marker for prostate cancer by differential hybridization analysis. Cancer Res. 51 , 606-611. Cano, A., PerezMoreno, M. A., Rodrigo, I., Locascio, A., Blanco, M. J., delBarrio, M. G., Portillo, F. and Nieto, M. A. (2000). The transcription factor Snail controls epithelial-mesenchymal transitions by repressing E-cadherin expression. Nature Cell Biology 2, 76-83.

Chan, E. F., Gat, U., McNiff, J. M. and Fuchs, E. (1999). A common human skin tumour is caused by activating mutations in beta-catenin. Nat. Genet. 21 , 410-413.

Chitaev, N. A. and Troyanovsky, S. M. (1997). Direct Ca2+-dependent heterophilic interaction between desmosomal cadherins, desmoglein and desmocollin, contributes to cell-cell adhesion. J. Cell Biol. 138, 193-201. Comijn, J., Berx, G., Verschueren, K., Van Grunsven, L., Bruyneel R.,

Mareel, M., Huylebroeck, D. and van Roy, F. (2000). SIP1 induces invasion via specific downregulation of the tumor suppressor E-cadherin. Manuscript in preparation.

Daniel, J. M. and Reynolds, A. B. (1995). The tyrosine kinase substrate p120(cas) binds directly to E-cadherin but not to the adenomatous polyposis coli protein or alpha-catenin. Mol. Cell. Biol. 15, 4819-4824.

De Leij, L., Postmus, P. E., Buys, C. H., Elema, J. D., Ramaekers, F., Poppema, S., Brouwer, M., Van Der Veen, A. Y., Mesander, G. and The, T. H. (1985). Characterization of three new variant-type cell lines derived from small cell carcinoma of the lung. Cancer Res. 45, 6024-6033.

Easty, D. M. (1981 ). Five human tumour cell lines derived from a primary squamous carcinoma of the tongue, two subsequent local recurrences and two nodal metastases. Br. J. Cancer 44(3), 363-370.

Emery, D. J., Diaz, L. A., Fairley, J. A., Lopez, A., Taylor, A. F. and Giudice, G. J. (1995). Pemphigus foliaceus and pemphigus vulgaris autoantibodies react with the extracellular domain of desmoglein-1. J. Invest. Dermatol. 104, 323- 328. Felsenstein, J. (1988). Phylogenies from molecular sequences: inference and reliability. Annu. Rev. Genet. 22, 521-565.

Finnemann, S., Mitrik, I., Hess, M., Otto, G. and Wedlich, D. (1997). Uncoupling of XB/U-cadherin-catenin complex formation from its function in cell-cell adhesion. J. Biol. Chem. 272, 11856-11862.

Geis, K., Aberle, H., Kuhl, M., Kemler, R. and Wedlich, D. (1998). Expression of the Armadillo family member p120(cas)1 B in Xenopus embryos affects head differentiation but not axis formation. Dev. Genes. Evol. 207, 471-481.

Gossen, M., Freundlieb, S., Bender, G., Muller, G., Hillen, W. and Bujard, H. (1995). Transcriptional activation by tetracyclines in mammalian cells. Science (Washington DC) 268, 1766-1769.

Hatzfeld, M. (1997). Band 6 protein and cytoskeletal organization. In Cytoskeletal-Membrane interactions and signal transduction., (ed. P. Cowin and M. W. Klymkowsky), pp. 49-59. Austin, Texas (USA): Landes Company and Chapman & Hall.

Hatzfeld, M. and Nachtsheim, C. (1996). Cloning and characterization of a new armadillo family mamber, p0071 , associated with the junctional plaque: evidence for a subfamily of closely related proteins. J. Cell Sci. 109, 2767-2778.

He, T. C, Sparks, A. B., Rago, C, Hermeking, H., Zawel, L., da Costa, L. T., Morin, P. J., Vogelstein, B. and Kinzler, K. W. (1998). Identification of c-MYC as a target of the APC pathway. Science 281 , 1509-1512.

Heid, H. W., Schmidt, A., Zimbelmann, R., Schafer, S., Wintersimanowski, S., Stumpp, S., Keith, M., Figge, U., Schnolzer, M. and Franke, W. W. (1994). Cell type-specific desmosomal plaque proteins of the plakoglobin family: plakophilin 1 (band 6 protein). Differentiation 58, 113-131.

Herrenknecht, K., Ozawa, M., Eckerskorn, C, Lottspeich, F., Lenter, M. and Kemler, R. (1991 ). The uvomorulin-anchorage protein alpha-catenin is a vinculin homologue. Proc. Nat! Acad. Sci. U.S.A. 88, 9156-9160.

Higgins, D. G., Thompson, J. D. and Gibson, T. J. (1996). Using CLUSTAL for multiple sequence alignments. In Computer Methods for

Macromolecular Sequence Analysis. Methods in Enzymology, vol. 266, pp. 383-402. Huber, O., Korn, R., McLaughlin, J., Ohsugi, M., Herrmann, B. G. and Kemler, R. (1996). Nuclear localization of beta-catenin by interaction with transcription factor LEF-1. Mech. Develop. 59, 3-10.

Kapprell, H. P., Owaribe, K. and Franke, W. W. (1988). Identification of a basic protein of Mr 75,000 as an accessory desmosomal plaque protein in stratified and complex epithelia. J. Cell Biol. 106, 1679-1691.

Karpati, S., Amagai, M., Prussick, R., Cehrs, K. and Stanley, J. R. (1993). Pemphigus-vulgaris antigen, a desmoglein type of cadherin, is localized within keratinocyte desmosomes. J. Cell Biol. 122, 409-415. Keirsebilck, A., Bonne, S., Staes, K., van Hengel, J., Nollet, F.,

Reynolds, A. and van Roy, F. (1998). Molecular cloning of the human p120^cu catenin gene (CTNND1 ): Expression of multiple alternatively spliced isoforms. Genomics 50, 129-146.

Keith, D., Armstrong, B., McKenna, K. E., Purkis, P. E., Green, K. J., Eady, R. A. J., Leigh, I. M. and Hughes, A. E. (1999). Haploinsufficiency of desmoplakin causes a striate subtype of palmoplantar keratoderma. Hum. Molec. Genet. 8, 143-148.

Kievits, T., Dauwerse, J. G., Wiegant, J., Devilee, P., Breuning, M. H., Cornelisse, C. J., Van Ommen, G. J. and Pearson, D. L. (1990). Rapid subchromosomal localization of cosmids by non-radioactive in situ hybridization.

Cytogenet. Cell Genet. 53, 134-136.

King, I. A., Angst, B. D., Hunt, D. M., Kruger, M., Arnemann, J. and Buxton, R. S. (1997). Hierarchical expression of desmosomal cadherins during stratified epithelial morphogenesis in the mouse. Differentiation 62, 83-96. Klymkowsky, M. W. (1999). Plakophilin, armadillo repeats, and nuclear localization. Microsc. Res. Tech. 45, 43-54.

Knudsen, K. A., Soler, A. P., Johnson, K. R. and Wheelock, M. J. (1995). Interaction of alpha-actinin with the cadherin/catenin cell-cell adhesion complex via alpha-catenin. J. Cell Bio! 130, 67-77. Koch, A., Denkhaus, D., Albrecht, S., Leuschner, I., vonSchweinitz, D. and Pietsch, T. (1999). Childhood hepatoblastomas frequently carry a mutated degradation targeting box of the beta-catenin gene. Cancer Res. 59, 269-273. Koch, P. J., Goldschmidt, M. D., Zimbelmann, R., Troyanovsky, R. and Franke, W. W. (1992). Complexity and expression patterns of the desmosomal cadherins. Proc. Nat! Acad. Sci. U.S.A. 89, 353-357.

Koch, P. J., Mahoney, M. G., Ishikawa, H., Pulkkinen, L., Uitto, J., Shultz, L., Murphy, G. F., Whitaker-Menezes, D. and Stanley, J. R. (1997).

Targeted disruption of the pemphigus vulgaris antigen (Desmoglein 3) gene in mice causes loss of keratinocyte cell adhesion with a phenotype similar to pemphigus vulgaris. J. Cell Bio! 137, 1091-1102.

Kozak, M. (1996). Interpreting cDNA sequences: Some insights from studies on translation. Mamm. Genome 7, 563-574.

Kozak, M. (1997). Recognition of AUG and alternative initiator codons is augmented by G in position +4 but is not generally affected by the nucleotides in positions +5 and +6. EMBO J. 16, 2482-2492.

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 ^', 680-685.

Lampugnani, M. G., Corada, M., Andriopoulou, P., Esser, S., Risau, W. and Dejana, E. (1997). Cell confluence regulates tyrosine phosphorylation of adherens junction components in endothelial cells. J. Cell Sci. 110, 2065-2077.

Lo, P. C. H. and Frasch, M. (1998). Bagpipe-dependent expression of vimar, a novel Armadillo-repeats gene, in Drosophila visceral mesoderm. Mech. Develop. 72, 65-75.

Mann, B., Gelos, M., Siedow, A., Hanski, M. L., Gratchev, A., Ilyas, M., Bodmer, W. F., Moyer, M. P., Riecken, E. O., Buhr, H. J. et al. (1999). Target genes of beta-catenin-T cell-factor lymphoid-enhancer-factor signaling in human colorectal carcinomas. Proc. Nat! Acad. Sci. U.S.A. 96, 1603-1608.

McCrea, P. D. and Gumbiner, B. M. (1991 ). Purification of a 92-kDa cytoplasmic protein tightly associated with the cell-cell adhesion molecule E-cadherin (uvomorulin). Characterization and extractability of the protein complex from the cell cytostructure. J. Bio! Chem. 266, 4514-4520. McGrath, J. A., Hoeger, P. H., Christiano, A. M., McMillan, J. R.,

Mellerio, J. E., Ashton, G. H. S., Dopping-Hepenstal, P. J. C, Lake, B. D., Leigh, I. M., Harper, J. I. et al. (1999). Skin fragility and hypohidrotic ectodermal dysplasia resulting from ablation of plakophilin 1. Br. J. Dermatol 140, 297-307. McGrath, J. A., McMillan, J. R., Shemanko, C. S., Runswick, S. K., Leight, I. M., Lane, E. B., Garrod, D. R. and Eady, R. A. J. (1997). Mutations in the plakophilin 1 gene result in ectodermal dysplasia skin fragility syndrome. Nat. Genet. 17, 240-244. Mertens, C, Kuhn, C. and Franke, W. W. (1996). Plakophilins 2a and

2b: constitutive proteins of dual location in the karyoplasm and the desmosomal plaque. J. Cell Bio! 135, 1009-1025.

Mizushima, S. and Nagata, S. (1990). pEF-BOS: a powerful mammalian expression vector. Nucleic Acid Res. 18, 5322. Mo, Y. Y. and Reynolds, A. B. (1996). Identification of murine p120(cas) isoforms and heterogeneous expression of p120(cas) isoforms in human tumor cell lines. Cancer Res. 56, 2633-2640.

Molenaar, M., van de Wetering, M., Oosterwegel, M., Peterson- Maduro, J., Godsave, S., Korinek, V., Roose, J., Destree, O. and Clevers, H. (1996). XTcf-3 transcription factor mediates beta-catenin-induced axis formation in Xenopus embryos. Ce// 86, 391-399.

Moll, I., Kurzen, H., Langbein, L. and Franke, W. W. (1997). The distribution of the desmosomal protein, plakophilin 1 , in human skin and skin tumors. J. Invest. Dermato! 108, 139-146. Morin, P. J., Sparks, A. B., Korinek, V., Barker, N., Clevers, H.,

Vogelstein, B. and Kinzler, K. W. (1997). Activation of beta-eaten in-Tcf signalling in colon cancer by mutations in beta-catenin or APC. Science 275, 1787-1789.

Motoyama, T. and Watanabe, H. (1983). Carcinoembryonic antigen production in human gastric cancer cell lines in vitro and in nude mice. Gann 74(5), 679-686.

Ozawa, M., Baribault, H. and Kemler, R. (1989). The cytoplasmic domain of the cell adhesion molecule uvomorulin associates with three independent proteins structurally related in different species. EMBO J. 8, 1711-1717.

Paffenholz, R. and Franke, W. W. (1997). Identification and localization of a neurally expressed member of the plakoglobin/armadillo multigene family.

Differentiation 61 , 293-304.

Page, R. D. M. (1996). Treeview: an application to display phylogenetic trees on personal computers. Computer Applic. Biosc. 12, 357-358. Paraskeva, C, Buckle, B. G., Sheer, D. and Wigley, C. B. (1984). The isolation and charaterization of colorectal cell lines at different stages in malignant transformation from familial polyposis coli patients. Int. J. Cancer 34(1), 49-56.

Peifer, M., Berg, S. and Reynolds, A. B. (1994). A repeating amino acid motif shared by proteins with diverse cellular roles. Cell 76, 789-791.

Remade, J. E., Kraft, H., Lerchner, W., Wuytens, G., Collart, C, Verschueren, K., Smith, J. C. and Huylebroeck, D. (1999). New mode of DNA binding of multi-zinc finger transcription factors: deltaEFI family members bind with two hands to two target sites. EMBO J. 18, 5073-5084. Reynolds, A. B., Daniel, J., McCrea, P. D., Wheelock, M. J., Wu, J. and

Zhang, Z. (1994). Identification of a new catenin: the tyrosine kinase substrate p120(cas) associates with E-cadherin complexes. Mo! Cell. Biol. 14, 8333-8342.

Reynolds, A. B. and Daniel, J. M. (1997). p120CTN: a SRC-substrate turned catenin. In Cytoskeletal-membrane interactions and signal transduction., (ed. P. Cowin and M. W. Klymkowsky), pp. 1-48. Austin, TX: Landes Company and Chapman & Hall.

Riggleman, B., Wieschaus, E. and Schedl, P. (1989). Molecular analysis of the armadillo locus: uniformly distributed transcripts and a protein with novel internal repeats are associated with Drosophila segment polarity gene. Gene Develop. 3, 96-113.

Schafer, S., Stumpp, S. and Franke, W. W. (1996). Immunological identification and characterization of the desmosomal cadherin Dsg2 in coupled and uncoupled epithelial cells and in human tissues. Differentiation 60, 99-108.

Schmidt, A., Langbein, L., Rode, M., Pratzel, S., Zimbelmann, R. and Franke, W. W. (1997). Plakophilins 1a and 1 b: widespread nuclear proteins recruited in specific epithelial cells as desmosomal plaque components. Cell Tissue Res. 290, 481-499.

Schmidt, A., Langbein, L., Pratzel, S., Rode, M., Rackwitz, H. R. and Franke, W. W. (1999). Plakophilin 3 - a novel cell-type-specific desmosomal plaque protein. Differentiation 64, 291 -306.

Schwartz, R. M. and Dayhoff, M. O. (1979). Dayhoff table. In Atlas of protein sequence and structure, (ed. M. O. Dayhoff), pp. 353-358. Washington D. C: National Biomedical Research Foundation. Sirotkin, H., O'Donnell, H., Dasgupta, R., Halford, S., St. Jore, B., Puech, A., Parimoo, S., Morrow, B., Skoultchi, A., Weissman, S. M. et al. (1997). Identification of a new human catenin gene family member (ARVCF) from the region deleted in Velo-Cardio-Facial syndrome. Genomics 41 , 75-83. Tetsu, O. and McCormick, F. (1999). Beta-catenin regulates expression of cyclin D1 in colon carcinoma cells. Nature 398, 422-426.

Thompson, J. D., Higgins, D. G. and Gibson, T. J. (1994). CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acid Res. 22, 4673-4680.

Vermeulen, S. J., Bruyneel, E. A., Bracke, M. E., De Bruyne, G. K., Vennekens, K. M., Vleminckx, K. L., Berx, G. J., van Roy, F. M. and Mareel, M. M. (1995). Transition from the noninvasive to the invasive phenotype and loss of alpha-catenin in human colon cancer cells. Cancer Res. 55, 4722-4728. Verschueren, K., Remade, J. E., Collart, C, Kraft, H., Baker, B. S.,

Tylzanowski, P., Nelles, W., G., Su, M. T., Bodmer, R., Smith, J. C. et al. (1999). SIP1 , a novel zinc finger/homeodomain repressor, interacts with Smad proteins and binds to 5'-CACCT sequences in candidate target genes. J. Bio! Chem. 27 '4, 20489- 20498. Vithalani, K. K., Parent, C. A., Thorn, E. M., Penn, M., Larochelle, D.

A., Devreotes, P. N. and DeLozanne, A. (1998). Identification of dariin, a Dictyostelium protein with armadillo-like repeats that binds to small GTPases and is important for the proper aggregation of developing cells. Mo! Bio! Cell 9, 3095-3106.

Wang, Y. X., Catlett, N. L. and Weisman, L. S. (1998). Vacδp, a vacuolar protein with armadillo repeats, functions in both vacuole inheritance and protein targeting from the cytoplasm to vacuole. J. Cell Bio! 140, 1063-1074.

Watabe-Uchida, M., Uchida, N., Imamura, Y., Nagafuchi, A., Fujimoto, K., Uemura, T., Vermeulen, S., van Roy, F., Adamson, E. D. and Takeichi, M. (1998). α-Catenin-vinculin interaction functions to organize the apical junctional complex in epithelial cells. J. Cell Bio! 142, 847-857.

Willert, K. and Nusse, R. (1998). Beta-catenin: a key mediator of Wnt signaling. Curr. Opin. Genet. Develop. 8, 95-102. Yap, A. S., Niessen, C. M. and Gumbiner, B. M. (1998). The juxtamembrane region of the cadherin cytoplasmic tail supports lateral clustering, adhesive strengthening, and interaction with p129(Ctn). J. Cell Bio! 141 , 779-789.

Zhang, Y. and Frohman, M. A. (1998). Cloning cDNA ends using RACE. In Molecular diagnosis of infectious diseases, vol. 13 (ed. U. Reischl), pp. 81-105. Totowa.NJ: Humana Press.

Zhou, J. H., Liyanage, U., Medina, M., Ho, C, Simmons, A. D., Lovett, M. and Kosik, K. S. (1997). Presenilin 1 interaction in the brain with a novel member of the Armadillo family. Neuroreport 8, 2085-2090.

Claims

1. A method for altering an undesirable functional property of a cell comprising providing said cell with a natural or artificial pathway suitable for transmitting and/or modulating signals in a cell.

2. A method according to claim 1 , wherein said cell is provided with an additional proteinaceous molecule capable of at least in part transmitting and/or modulating a signal of a cadherin/catenin-like signalling pathway.

3. A method according to claim 1 or claim 2, wherein said pathway comprises at least one catenin-like and/or at least one cadherin-like molecule.

4. A method according to anyone of claims 1-3, wherein said undesirable functional property comprises a neoplastic property.

5. A method according to anyone of claims 1-4, wherein said undesirable functional property comprises a metastatic property.

6. A method according to anyone of claims 1-3, wherein said undesirable functional property comprises a disease of epithelial cells such as skin cells.

7. A method according to anyone of claims 1-6, wherein said proteinaceous molecule is Plakophilin-3 or a functional part, derivative and/or analogue thereof.

8. An isolated or recombinant nucleic acid encoding a plakophilin-3, which in humans comprises a nucleic acid sequence as depicted in figure 2, which in mice comprises a nucleic acid sequence as depicted in figure 10, and which in Xenopus laevis comprises a nucleic acid sequence as depicted in figure 12, or a functional part, derivative and/or analogue thereof.

9. A nucleic acid delivery vehicle comprising a nucleic acid according to claim 8.

10. A cell provided with a nucleic acid according to claim 8 or claim 9 or a derivative thereof.

11. A proteinaceous molecule or a functional part, derivative and/or analogue thereof, derived from a nucleic acid according to claim 8 or a cell according to claim 10.

12. A proteinaceous molecule or a functional part, derivative and/or analogue thereof, according to claim 11 , wherein said molecule comprises a plakophiiin-

13. An antibody or a functional part, derivative and/or analogue thereof, specific for a proteinaceous molecule according to claim 11 or claim 12.

14. An antibody according to claim 13 wherein said antibody is monoclonal antibody 23E3/4 or monoclonal antibody 12B11 F8 as deposited with the Belgian Coordinated Collections of Microorganisms - BCCMTM on April 26, 2000 and having accession numbers LMBP 5482CB and LMBP 5481 CB, respectively.

15. A nucleic acid or a functional part, derivative and/or analogue thereof, encoding an antibody according to claims 13 or 14.

16. A nucleic acid delivery vehicle comprising a nucleic acid according to claim 15.

17. Use of a nucleic acid delivery vehicle according to claim 9 or claim 16 in a Gene

Therapy application.

18. A method for diagnosing an inherited or sporadic epithelial tissue disease comprising obtaining a sample of cells of an affected area of the body and detecting the presence or absence of a mutated plakophilin-3 and/or plakophilin-3 encoding nucleic acid, or a functional part, derivative and/or analogue thereof.

19. A method according to claim 18, wherein said epithelial tissue comprises skin tissue.

20. A method for the treatment of skin disease comprising providing skin cells of an affected skin area of an individual with a nucleic acid delivery vehicle according to claim 9 or claim 16.

21. A method for the treatment of skin disease comprising providing skin cells of an affected skin area of an individual with a chemical compound influencing the expression and/or functionality of the plakophilin-3 gene and/or the expression of the function of the plakophilin-3 protein.

22. A method for the improvement of wound healing comprising providing skin cells of an affected skin area of an individual with a chemical compound influencing the expression and/or functionality of the plakophilin-3 gene and/or the expression and/or function of the plakophilin-3 protein.

23. A cell contacted with a nucleic acid delivery vehicle according to claim 9 or claim 16, or a derivative thereof.

24. A cell, and/or the progeny thereof, provided with nucleic acid encoding a mutant plakophilin-3 or a functional part, derivative and/or analogue thereof.

25. A eukaryotic cell, and/or the progeny thereof, wherein a chromosomal nucleic acid encoding a plakophilin-3 according to claim 11 or 12 has been mutated and/or removed.

26. An antibody according to claims 13 and 14 for use as a medicament.

27. Use of an antibody according to claims 13 and 14 for the manufacture of a medicament to treat skin disease or another disease in which plakophilin-3 is a mediator.

28. An isolated or recombinant nucleic acid encoding a promoter sequence of plakophilin-3 and characterized by having SIP-1 and Snail binding sites which in humans and mice comprises a nucleic acid sequence as depicted in figure 23, or a functional part, derivative and/or analogue thereof.

29. Use of SIP1 and/or snail to modulate plakophilin-3 expression.