More particularly, the present invention relates to novel proteins that are involved in transduction of a plakoglobin related-signal to the nucleus.

The cadherin superfamily represents several cadherins with function in cell-cell adhesion, morphogenesis and tissue homeostasis (Takeichi, 1991; Kemler, 1992; Suzuki, 1996; Nollet et al., 2000). The transmembrane glycoprotein E-cadherin is the best-studied prototype of this family and has been identified as a potent suppressor of invasion (Behrens et al., 1989; Frixen et al., 1991; Vleminckx et al., 1991). Recent studies provided proof for a tumor suppressor role of human E-cadherin as the CDH1 gene behaves according to the two-hit model of Knudson (Knudson, 1985) in infiltrative lobular cancers (Berx et al., 1995; Berx et al., 1996) and diffuse gastric cancers (Becker et al., 1994; Becker et al., 1996). Cadherins function as cell-cell adhesion molecules by homophilic interactions with other cadherin molecules, but linkage to the actin cytoskeleton is also essential. The latter is achieved by the catenins (catena means chain) (Ozawa et al., 1990; Cowin, 1994), which comprise the Armadillo proteins (e. g. ß-catenin, plakoglobin and p120) and the vinculin-like a-catenins.

These catenins were also found to be associated with the cytoplasmic tumor suppressor gene product APC (adenomatous polyposis coli) (Peifer, 1993; Su et al., 1993). The APC protein is linked to the microtubular cytoskeleton.

The armadillo protein plakoglobin is a structural component of the intercellular cadherin/catenin adhesion complex. Desmosomal cadherins can also bind to plakoglobin, localizing the plakoglobin in the desmosomes. (Mathur et al., 1994; Troyanovsky et al., 1994a; Troyanovsky et al., 1994b; Chitaev et al., 1996). In the desmosomes, plakoglobin mediates a link between desmosomal cadherins and the cytokeratin cytoskeleton via desmoplakin (Korman et al., 1989; Kowalczyk et al., 1997). Other Armadillo-like components of the desmosomal plaque are the plakophilins. So far, three plakophilins have been identified (Heid et al., 1994; Mertens et al., 1996; Bonne et al., 1999; Schmidt et al., 1999). Their exact role in desmosomes is still under study. One hypothesis is that they provide additional binding sites for intermediate filaments in the upper layers of the epidermis, where they are most needed (Kowalczyk et al., 1999; Green and Gaudry, 2000). In addition, it was shown that plakoglobin, like the homologous armadillo protein ß-catenin, is found in a

macromolecular complex with the tumor suppressor protein APC (Rubinfeld et al., 1993; Su et al., 1993; Shibata et al., 1994; Rubinfeld et al., 1995).

The P-catenin protein shows a high homology with plakoglobin, especially in the Arm domain, while the amino-and carboxyterminal sequences are less conserved (McCrea et al.,'1991, Butz, et al., 1992). Knock-out mice for the plakoglobin gene are lethal, revealing a specialized and indispensable function for plakoglobin in the desmosomes, particularly in heart (Ruiz et al., 1996; Bierkamp et al., 1996). Knock-out mice for p- catenin are as well lethal, but differ significantly from the plakoglobin knock-out phenotype (Haegel et al., 1995). Truncation of the plakoglobin protein can result in Naxos disease (McKoy et al., 2000). For the P-catenin knock-out mice, it was shown that the development of the embryonic ectoderm was affected (Haegel et al., 1995).

One of the most intriguing discoveries in this field is the recently described association of LEF-1 (lymphocyte enhancer-binding factor-1), an architectural transcription factor (Love et al., 1995), with ß-catenin. This was initially found by the use of the two-hybrid system (Behrens et al., 1996). The interaction between ß-catenin and LEF-1 occurs in the nucleus, implicating a central role for P-catenin in the transcriptional regulation of target genes, which can lead to tumorigeneity (Huber et al., 1996; Peifer, 1997).

Among the target genes induced by the ß-catenin/LEF-1 complex are the c-myc proto- oncogene (He et al., 1998) and the cyclin-D1 gene (Shtutman et al., 1999; Tetsu and McCormick, 1999).

The difference in function despite the high homology between these two vertebrate armadillo proteins P-catenin and plakoglobin prompted us to look for new, known or unknown interaction partners of plakoglobin that may play a role in signaling or nuclear localisation. Interestingly, novel plakoglobin interacting polypeptides were found of which at least one is involved in transduction of plakoglobin related signals to the nucleus.

It is a first aspect of the present invention to provide novel plakoglobin interacting polypeptides. Preferentially, said proteins are involved in transduction of a plakoglobin mediated and/or a plakoglobin related signal to the nucleus. One embodiment of the invention is a plakoglobin interacting polypeptide comprising the novel amino acid sequence depicted in SEQ ID N° 6 or an amino acid sequence at least 60%, preferably at least 70%, more preferably at least 80% identity to SEQ ID N° 6, as calculated on the whole sequence using an advanced Blast search (Altschul et al., 1997), or a functional fragment thereof. Examples of functional fragments are plakoglobin

interacting polypeptides, comprising SEQID N°2 and/or SEQID N°4. Preferably, said functional fragments are essentially consisting of SEQ ID N°2 or SEQ ID N° 4.

Another embodiment of the invention is a plakoglobin interacting polypeptide comprising a novel amino acid sequence as depicted in SEQ ID N° 8.

It is another aspect of the invention to provide a nucleic acid sequence encoding a plakoglobin interacting protein according to the invention. One preferred embodiment is a nucleic acid sequence comprising a nucleic acid sequence as depicted in SEQ ID No 5. Another preferred embodiment is a nucleic acid sequence comprising a nucleic acid sequence as depicted in SEQ ID N° 1 and/or comprising a nucleic acid sequence as depicted in SEQ ID N° 3. Still another embodiment is a nucleic acid sequence comprising a nucleic acid sequence as depicted in SEQ ID No 7.

Another aspect of the invention is an expression vector, comprising a nucleic acid sequence encoding a plakoglobin interacting polypeptide according to the invention.

Still another aspect of the invention is a host cell, transformed or transfected with said expression vector; said host cell can be a prokaryotic host cell, such as Escherichia coli or Bacillus subtilis, or it can be an eukaryotic host cell. A eukaryotic host cell may be any eukaryotic host cells including but not limited to yeast cells, insect cells and mammalian cells.

Another aspect of the invention is a method to produce a plakoglobin interacting polypeptide, according to the invention. Preferentially, this method is comprising the use of a host cell according to the invention. Methods to produce said polypeptide using said host cell are known to the person skilled in the art.

Another aspect of the invention is a novel plakoglobin interacting protein, according to the invention, for the use as a medicament.

Still another aspect of the invention is the use of a plakoglobin interacting protein, according to the invention, for the manufacture of a medicament to treat plakoglobin related diseases. Such proteins may be novel proteins as described above, or known proteins for which the plakoglobin interaction has been demonstrated for the first time, such as NRF1 or NDP52. Plakoglobin related diseases include, but are not limited to, cancers, especially skin carcinomas such as basal cell carcinoma, squamous cell carcinoma or extramammary Paget's disease, Naxos disease, heart diseases, skin blistering and acantholytic diseases such as subcorneal acantholysis, Grover's disease, Hailey-Hailey's disease or Darier's disease. Still another aspect of the invention is a plakoglobin interacting protein, that is also interacting with plakophilin.

Another aspect of the invention is the use of said plakoglobin interacting protein to treat plakophilin related diseases, such as ectodermal dysplasia/skin fragility syndrome (McGrath et al., 1997; Whittock et al., 2000). Preferably, said plakoglobin interacting protein comprises SEQ ID N°6, SEQ ID N° 2 or SEQ ID N° 4, or a functional fragment thereof.

The terms'medicament'or'use for the manufacture of a medicament to treat'or pharmaceutical composition' (see below) relate to a composition comprising plakoglobin interacting proteins according to the invention, or homologues, derivatives or functional fragments as described above and a pharmaceutical acceptable carrier or excipient (both terms can be used interchangeably) to treat diseases as indicated above. Suitable carriers or excipients known to the skilled man are saline, Ringer's solution, dextrose solution, Hank's solution, fixed oils, ethyl oleate, 5% dextrose in saline, substances that enhance isotonicity and chemical stability, buffers and preservatives. Other suitable carriers include any carrier that does not itself induce the production of antibodies harmful to the individual receiving the composition such as proteins, polysaccharides, polylactic acids, polyglycolic acids, polymeric amino acids and amino acid copolymers. The'medicament'may be administered by any suitable method within the knowledge of the skilled man. The dosage and the method of administration will depend upon the individual. Generally, the medicament is administered so that the compound of the present invention is given at a dose between 1 ug/kg and 10 mg/kg, more preferably between 10 lug/kg and 5 mg/kg, most preferably between 0.1 mg/kg and 2 mg/kg.

Still another aspect of the invention is the use of a plakoglobin interacting polypeptide according to the invention and/or a functional fragment thereof to screen compounds that interfere with the interaction of said plakoglobin interacting polypeptide with plakoglobin. As a non-limiting example, polypeptides as depicted in SEQ ID 2,4,6, or 8 can be used, as well as the polypeptides NRF1 or NDP52. Still another aspect of the invention is the use of a plakoglobin interacting polypeptide according to the invention and/or a functional fragment thereof to screen compounds that interfere with the interaction of said plakoglobin interacting polypeptide with plakophilin. As a non-limiting example, polypeptides as depicted in SEQ ID 2,4, or 6 can be used Screening methods for such compounds are known to the person skilled in the art and have been described, as a non limiting example, in W09813502 and in W09923116.

Another aspect of the invention is said screening method, comprising a plakoglobin interacting polypeptide according to the invention.

Still another aspect of the invention is a compound, isolated with said screening method.

Another aspect of the invention is a pharmaceutical composition comprising one or more plakoglobin interacting polypeptides according to the invention, or one or more said compounds, isolated with the screening method according to the invention. Such pharmaceutical compositions may be used to treat cancers, especially skin carcinomas, or acantholytic diseases such as subcorneal acantholysis, Grover's disease, Hailey-Hailey's disease or Darier's disease, or ectodermal dysplasia/skin fragility syndrome.

DEFINITIONS The following definitions are set forth to illustrate and define the meaning and scope of various terms used to describe the invention herein.

Plakoglobin interacting polypeptide means that said protein may interact with plakoglobin, preferentially human plakoglobin, or a functional fragment of said plakoglobin as can be measured as a non limiting example by a yeast two hybrid test. A functional fragment of human plakoglobin is a fragment that comprises at least amino acid residues 37-404 and/or amino acid residues 405-570 of the human plakoglobin sequence.

A functional fragment of Plakoglobin interacting protein is a fragment that still can interact with plakoglobin, or with a functional fragment of plakoglobin as defined above. Polypeptide as used here means any proteineous structure, independent of the length and includes molecules such as peptides, phosphorylated proteins and glycosylated proteins.

Interacting or interaction means any interaction, be it direct or indirect. A direct interaction implies a contact between the binding partners. An indirect interaction means any interaction whereby the interaction partners interact in a complex of more than two compounds. This interaction can be completely indirect, with the help of one or more bridging compounds, or partly indirect, where there is still a direct contact that is stabilized by the interaction of one or more compounds.

Interfere with the interaction can be any way of interference, both positive or negative.

It can mean an enhancement of the interaction, a weakening of the interaction or a complete inhibition of the interaction.

Compound means any chemical of biological compound, including simple or complex inorganic molecules, peptides, peptido-mimetics, proteins, antibodies, carbohydrates, nucleic acids or derivatives thereof.

BRIEF DESCRIPTION OF THE FIGURES Figure 1 : Retransformation of yeast cells with plakoglobin fragments (bait plasmids) and PLA_2H12 (prey plasmid). Protein-protein interactions, indicated by + or ++, were demonstrated by expression of both the HIS3 and the LacZ reporter genes. Minus sign, no interaction is detected. Lines indicate the plakoglobin fragments (initiating and terminating amino acid residues are indicated by the appropriate codon numbers), as described at the right and encoded by the plasmids listed at the left. Boxes covering the lines represent armadillo repeats.

Figure 2: Northern blot analysis of the PLA2H12 mRNA in several human tumor cell lines and human tissues. 28S and 18S are ribosomal size markers.

Figure 3: Sequence of human PLA2H12 cDNA. Spaces separate blocks of 10 nucleotides (nt). The predicted amino acid sequence of the ORF is indicated in bold (one-letter code). The start codon and the stop codon are boxed in black. Also indicated is the sequence of the predicted bipartite nuclear localisation signal (shaded box). The predicted SH3 domain is indicated with open arrows. Two leucine zipper motifs are underlined. The leucine residues that are part of the consensus sequence of a leucine zipper (L-X6-L-X6-L-X6-L) are marked in gray. The poly-A signal is indicated in bold. Exon-exon junctions are marked. Between black arrows is indicated the original cDNA fragment that was found by us to interact with plakoglobin in a yeast 2- hybrid screen.

Figure 4: Alignment of the SH3 domains of PLA_2H12 (human), ZO-1 (human and mouse), ZO-2 (human and dog), ZO-3 (dog). The conserved amino acids are in bold (one letter code). The SH3 domain of human PLA 2H12 was detected by BLAST using the ProDom database (Altschul et al., 1997).

Figure 5 : Genomic structure of the human PLA2H12 gene

Figure 6: Rapid-scan RT-PCR expression analysis of the PLA2H12 mRNA. M= Marker (BstEll digested k DNA), + = positive control on plasmid DNA,-no cDNA added.

Figure 7: Sequence of the human PLA2H34 cDNA fragment, found to encode a polypeptide interacting with plakoglobin in a in a yeast 2-hybrid screen. Spaces separate blocks of 10 nucleotides (nt). The predicted amino acid sequence of the ORF is indicated in bold (one-letter code). A stop codon is boxed.

Figure 8: Retransformation of yeast cells with bait plasmids, encoding plakoglobin fragments (listed at the right; see also Fig. 1) and prey plasmids, encoding various protein fragments (listed on top; see also Table 6). Protein-protein interactions, indicated by + to +++, were demonstrated by expression of both the HIS3 and the LacZ reporter genes. Minus signs, no interaction is detected. Lines indicate the plakoglobin fragments (initiating and terminating amino acid residues are indicated by the appropriate codon numbers). Boxes covering the lines represent armadillo repeats.

Figure 9: Retransformation of yeast AH109 with plakoglobin encoding bait plasmids, listed on the left, and prey plasmids listed on the right, encoding various parts of the human PLA_2H12 protein. Growth and blue-staining on SD-LWHAde+alphaX-gal plates indicate interacting protein (fragments) (a, a'). Panel b and b'represent control plates indicative for the ability of the transformed yeasts to grow on medium independent of any interaction between bait and prey. For the numbering see Figure 1.

EXAMPLES Materials and methods to the examples Bacterial strains and cell lines Escherichia coli DH5a (supE44, hsdR17, deoR, recA1, endA1,/acZDM15) and E. coli HB101 (supE44, mort, mrr, hsdS20, recA1) were used for transformations, plasmid propagation and isolation. The bacteria were grown in LB medium supplemented with 100 ug/ml ampicillin. For selection of the two-hybrid cDNA-library plasmid, transformed HB101 bacteria were grown on minimal M9 medium, supplemented with 50 ug/ml ampicillin, 40 ug/ml proline, 1 mM thiamine-HCI and 1% of the appropriate amino acid drop-out solution. After selection the bacteria were maintained in LB medium supplemented with 50 pg/ml ampicillin.

Colon adenocarcinoma cell lines SW620 (CCL-227) and SW1116 (CCL-233) were obtained from the American Tissue Culture Collection (ATCC, Rockville, MD).

DLD1/R2/7 was a round cell variant subcloned from DLD1 (CCL-221). (Vermeulen et al., 1995; Watabe-Uchida et al., 1998). Cell lines LICR-HN2, LICR-HN3, LICR-HN5 and LICR-HN6 are derived from head and neck squamous cell carcinoma (Easty et al., 1981). MKN45 is a gastric carcinoma cell line (Motoyama and Watanabe, 1983).

Plasmids and gene assembly Restriction enzymes were purchased from Gibco BRL Life Technologies (Paisley, UK) or from New England Biolabs (Beverly, MA, USA). Restriction enzymes were used according to manufacturers'recommendations. All PCR reactions were performed using Vent (Biolabs) DNA polymerase. The primers for PCR amplification were either home made (University of Ghent) or obtained from Gibco BRL. The standard PCR mixture, in a reaction volume of 100 ul, contained template cDNA (plasmid), 25 pmol of both specific primers, 200 uM dXTPs and the PCR buffer supplied with Vent DNA polymerase. Unless otherwise stated, no additional MgSO4 was added. Vents DNA polymerase was used at 1 U/reaction. The DNA amplification was performed in the PTC-200 Peltier Thermal Cycler PCR System (MJ Research, Watertown, MA). The PCR program started with a DNA denaturating step at 94°C for 3 min, followed by 80°C for 1 min. Cycling conditions were 94°C for 1 min, 50-60°C for 30 sec and 72°C for 2 min. This was repeated for a total of 35 cycles and was followed by a final extension step at 72°C for 10 min.

Construction of the plasmid encoding the plakoglobin-GAL4-DBD hybrid protein for two-hybrid screening The plasmid pHPG Ca2.1, containing full-length human plakoglobin cDNA was obtained from Dr. Goldschmidt M. and Dr. W. Franke (German Cancer Research Center, Heidelberg) (Genbank Acc. No. M23410) (Franke et al., 1989). A fragment was amplified from this plasmid by PCR using the forward primer (= FVR133F; Table 2) 5'- GGTGAATTCGTCAGCAGCAAGGGCATCAT-3'containing an EcoRl restriction site (underlined and bold) and the compatible reverse primer (FVR134R ; Table 2) 5'- GGTTTGATGCAGGGTCCACAGGCAGTTCT-3'. Subsequently, the PCR product was digested with the EcoRl and Sacl restriction enzymes. Another fragment of human plakoglobin cDNA was generated by digestion of the plasmid pHPG Ca2.1 with Sacl and Bglll restriction enzymes. Digestion of the plasmid with the restriction enzymes Bglll and Pstl resulted in the isolation of a third fragment. Ligation of these three

fragments into the EcoRl-Pstl digested pGBT9 vector (Matchmaker, Clontech, Palo Alto, CA) resulted in the construction of the pGBT9Plako (227-2340) plasmid.

Prior to the construction of the pGBT9Plako (227-1853) plasmid, these three plakoglobin cDNA fragments were ligated into the pSE280 vector (Invitrogen ; Brosius et al., 1981) to yield the pSE280Plako (227-2340) plasmid. Subsequently, this plasmid was digested with the EcoRl and the Xmal restriction enzymes. This fragment was ligated into the pSE280 to yield the pSE280Plako (227-1853) plasmid. Digestion of the pSE280Plako (227-1853) plasmid with EcoRl and BamHl was followed by ligation of this fragment into the EcoRl-BamHl digested pGBT9 vector. This plasmid was called the pGBT9Plako (227-1853) and was used as a bait in a two-hybrid library screening.

The in-frame cloning of the fragments was confirmed by DNA sequence analysis using the vector-specific forward primer (= FVR175F; Table 3) with sequence 5'- ATCATCGGAAGAGAGTAGTA-3'. The inserts were also sequenced with the vector- specific reverse primer (= FVR217R ; Table 3) with sequence 5'- AAAATCATAAATCATAAGAA-3'.

Construction of plasmids encoding fragments of plakoglobin fused to the GAL4-DBD A fragment was amplified from plasmid pHPG Ca2.1 by PCR using the forward primer (= FVR133F ; Table 2) with sequence 5'-GGTGAATTCGTC- AGCAGCAAGGGCATCAT-3'containing an EcoRl restriction site (underlined and bold) and the compatible reverse primer (= FVR134R; Table 2) with sequence 5'- GGTTTGATGCAGGGTCCACAGGCAGTTCT-3'. Subsequently, the PCR product was digested with the EcoRl and Sacl restriction enzymes. The fragment was ligated into the EcoRl-Sacl digested pBluescriptlIKS-vector (Stratagene, La Jolla, CA) to construct the pBSKSIIPlako (227-560). This plasmid was digested with EcoRl-Pstl and the fragment was ligated into the EcoRl-Pstl digested pGBT9 vector to yield the pGBT9Plako (227-461) plasmid.

The pBSKSIIPlako (227-1335) plasmid was constructed by ligation of the EcoRl-Hincil fragment of the pSE280Plako (227-2340) plasmid into the EcoRl-Hincll digested pBluescriptilKS-vector (Stratagene). This plasmid was used to isolate an EcoRI-Sall fragment. This fragment of human plakoglobin cDNA was finally ligated into the EcoRl- Sall digested pGBT9 to construct the pGBT9Plako (227-1335).

The pBSKSIIPlako (1335-2340) plasmid was constructed by ligation of the Hincil-Xhol fragment of pSE280Plako (227-2340) into the Hincll-Xhol digested pBluescriptilKS-

vector (Stratagene). The pBSKSIIPlako (1335-2340) plasmid was digested with the EcoRl-Xmal restriction enzymes and the resulting fragment was ligated into the EcoRl- Xmal digested pGBT9 vector. The final plasmid was called pGBT9Plako (1335-1853).

A fragment of the plakoglobin cDNA was isolated by digestion of the pSE280Plako (227-2340) plasmid with Xmal and Pstl restriction enzymes. This fragment was ligated into the Xmal-Pstl digested pBluescriptlIKS-vector to obtain the pBSKSIIPlako (1832- 2340) plasmid. This plasmid was subsequently digested with BamHl and Pstl restriction enzymes and ligation of this fragment into the BamHI-Pstl digested pGBT9 vector resulted in the construction of the pGBT9Plako (1853-2340) plasmid.

To construct plasmid pGBT9Plako (227-1028), the EcoRi-Sau3AI fragment of the pBSKSIIPlako (227-1335) plasmid was ligated into the EcoRl-BamHl digested pGBT9 vector.

The pBSKSIIPlako (227-1335) plasmid was digested with Ec113611 and Xhol restriction enzymes. This fragment was ligated into the Smal-Xhol digested pBluescriptilKS- plasmid to yield the pBSKSIIPlako (560-1332) plasmid. A BamHI-Pstl restriction fragment of this plasmid was isolated and ligated into the pGBT9 vector digested with the appropriate restriction enzymes. The constructed plasmid was called pGBT9Plako (558-998) plasmid. The pGBT9-Plako (1335-2120) plasmid was constructed by ligation of the EcoRI-Sacl fragment of the pGBT9Plako (1335-2340) plasmid into the pGBT9 vector. To construct the pGBT9Plako (558-2120), the Sacl fragment of pGBT9Plako (227-2340) was ligated into the pGBT9 vector.

RNA isolation RNA was isolated from various human tumor cell lines by using the RNA-zol BT"" procedure (Wak Chemie-Medical, Bad Homburg, Germany). RNA from normal tissue (fetal brain, mammary gland, uterus and small intestine) was purchased from Clontech (Clontech Laboratories, Palo Alto, CA).

Yeast strains and media The Saccharomyces cerevisiae strain HF7c (Mata, ura3-52, his3-A200, ade2-101, lys2-801, trp1-901, leu2-3, 112, gal4-542, gal80-538, lys2 :: GAL1-HIS3, URA3:: GAL4 17-mers)-CYC1-LacZ) (Matchmakerm, Clontech, Palo Alto, CA) was used for most of the assays. The HF7c yeast strain carries two reporter genes, HIS3 and LacZ, both integrated into the yeast genome and under the control of GAL4 responsive elements,

the GAL1 UAS and the UASG-17mer. It has also two auxotrophic markers, trp1 and/eu2, which are used for plasmid selection upon transformation. Yeast cultures were grown at 30°C in either complete YPD medium (1% yeast extract, 2% peptone and 2% glucose) or SD minimal medium (0.5% yeast nitrogen base without amino acids, 2% glucose, and 1 % of the appropriate amino acid drop-out solution).

Alternatively, the yeast strain AH109 (MATa, trp1-901, leu2-3, 112, ura3-52, his3-200, gal4A, gal80A, LYS2:: GAL1UAS-GAL1TATA-HIS3, GAL2uAs-GAL2TATA-ADE2, URA3:: MEL1uAs-MEL1TATA-iaczX Clontech) was used. The AH109 strain carries three reporter genes (ADE2, HIS3, and lacZ) under the control of distinct GAL4 upstream activating sequences (UASs) and TATA boxes. These promoters yield strong and specific responses to GAL4. The strain has two auxotrophic markers, trp1 and leu2, used for plasmid selection upon transformation. Yeast cultures were grown in either complete YPD medium (see above) supplemented with adenine to a final concentration of 0.003% adenine (YPDA), or SD minimal medium (see above).

Yeast transformation and, 8-galactosidase assays Plasmids encoding the GAL4 hybrid proteins were introduced into the HF7c yeast reporter strain by the lithium acetate (LiAc) transformation procedure (Gietz et al., 1992). Transformants were selected for the presence of the plasmids by growing on appropriate media at 30°C. They were allowed to grow until the colonies were large enough to perform a ß-galactosidase filter assay, usually for 3-4 days. The transformed cells were then transferred onto an 82-mm nitrocellulose membrane (Sartorius, Goettingen, Germany), permeabilized by freezing the membranes in liquid nitrogen for one minute, followed by thawing at room temperature. The membranes are soaked with 1.5 mi of Z-buffer containing 5-bromo-4-chloro-3-indolyl--D-galactosidase (X-gal) and incubated at 30°C until the appearance of blue colonies. This takes usually 30 min to 12 h. The membranes are then dried, analyzed and stored.

Two-hybrid cDNA library screening and DNA sequence analysis The plasmid pGBT9Plako (227-1853) was used to screen a human fetal kidney cDNA library, cloned in the GAL4 activation domain vector pGAD10 (Clontech, Palo Alto, CA). The plasmids were introduced into the HF7c yeast strain by using sequential transformation by the lithium acetate (LiAc) method described by Gietz et al. (Gietz et al., 1992). The interaction screen was carried out in the yeast strain HF7c on media

lacking leucine, tryptophan, histidine, and containing 5 mM 3-aminotriazole (3-AT). The P-galactosidase activity in yeast was measured using a ß-galactosidase filter assay.

Yeasts harboring interacting proteins were used for plasmid isolation. The obtained plasmid mixture was transformed into Escherichia coli HB101 electrocompetent cells.

HB101 has a defect in the leuB gene, which can be complemented by LEU2 from yeast. So, this strain can be used for selection of the library plasmid, which carries the yeast LEU2 transformation marker. From these transformants, the library plasmids were isolated and introduced into Escherichia coli DH5a. Plasmids isolated from this strain were further characterized by DNA sequence. This was done by the dideoxy chain termination method (Sanger, 1981) using fluorescent dye terminators and an ABI 373A or an ABI 377 automated DNA sequencer (Applied Biosystems, Foster City, CA).

The sequences were further analyzed by BLAST search (Altschul et al., 1990) and the DNAstar software packages (DNASTAR Inc, Madison, USA), and by the Staden gap4 software (Bonfield et al., 1995). To sequence the cDNA insert of the library plasmid, two primers were designed on the pGAD10 vector sequence flanking the cDNA insert: a forward primer 5'-ACCACTACAATGGATGATGT-3' (= FVR174F; Table 3) and a reverse primer 5'-TAAAAGAAGGCAAAACGATG-3' (FVR 192R; Table 3).

Constructs made in the two-hybrid vectors pGBKT7 and pGADT7 The insert of the PLA_2H12 plasmid (pGAD10hPLA_2H12) as isolated from the two- hybrid screen was EcoRl cut and cloned in the EcoRl sites of either pGBKT7 or pGADT7, resulting in pGBKT7hPLA_2H12Original and pGADT7hPLA_2H12Original, respectively. Clones of interest were identified by restriction analysis and sequencing.

In order to clone the full length human PLA 2H12 open reading frame (ORF) in two- hybrid vectors, an RT-PCR was performed using primers FVR2486F and FVR1039R using MKN45 cDNA as template. Primer FVR2486F contains an artificial EcoRl site (Table 1). The amplified fragment was EcoRl/Xhol digested and a 530 bp fragment was purified. The PLA_2H12-specific cDNA clone 11b10 isolated from the Rapid- Screen cDNA library was cut with Spel, blunted, cut with Xhol and a 2860bp fragment was purified. The cDNA insert of clone 11 b10 comprises nt 324-3948 of the full length PLA_2H12 (Fig. 3). Both fragments were simultaneously ligated into the pGBKT7 vector, which was BamHl cut, blunted and Sall cut, yielding a full length PLA_2H12 cDNA in frame with the GAL4-DNA binding domain (pGBKT7hPLA2H12FL). Both the PCR fragment and the cloning sites of this construct were sequenced to ensure no

mutations had occurred. The pGBKT7hPLA_2H12FL plasmid was subsequently cut with EcoRl and Sall, the insert of interest was purified and cloned in the EcoRl and Xhol sites of pGADT7, yielding pGADT7hPLA_2H12FL. Again, the cloning sites were sequenced to ensure no mutations had occurred and the insert was in frame with the GAL4-DNA activation domain. The full length human plakophilin-1 (PKP1) ORF cloned in the pEGFP-C2 vector (Clontech) was a gift from dr. M. Hatzfeld (Halle, Germany).

This construct was cut with EcoRl and Sall, the insert of interest was purified and cloned in the EcoR and Sall sites of the pGBKT7 vector yielding pGBKT7hPKP1.

Sequencing of the cloning sites was performed. The full-length human plakophilin-3 (PKP3) ORF was amplified using primers FVR1846F and FVR1847R (Table 2b).

These primers contain an artificial EcoRl and Sall site, respectively. The amplified product was purified, cut with EcoRl and Sall and cloned in the EcoR and Sall sites of the pGBKT7 vector yielding pGBKT7hPKP3. This construct was fully sequenced to ensure no mutations had occurred. All constructs were checked for auto-activation in the two-hybrid system.

Northern blot analysis The total length of the complete mRNA encoding the new plakoglobin-binding protein PLA_2H12 was estimated by Northern blot analysis. Total RNA (30 ug) was glyoxylated, size fractioned on a 1% agarose gel and transferred to a Hybond-N+ membrane (Amersham). The probe was radioactively labeled using random priming (RadPrime DNA labeling system; Gibco BRL Life Technologies, Gent, Belgium). The probe used was the EcoRl-generated insert of the originally isolated two-hybrid prey clone PLA_2H12. Hybridizations were performed as described elsewhere (Bussemakers et al., 1991). The blot was exposed for 10 d to a P-imager screen (Molecular Dynamics).

Isolation of the full-length cDNA of PLA_2H12 and DNA sequence analysis In order to complete the cDNA of the clone isolated by the two-hybrid screen, we used several techniques. In a first approach, conventional screening of a phage library of human fetal kidney did not yield any PLA_2H12-specific sequence. Second, 5'and 3' RACE experiments were performed according to the Marathon cDNA amplification protocol (Clontech Laboratories, Palo Alto, CA) using a Marathon library of human testis. For the 5'RACE, two PLA_2H12-specific primers were used: FVR1039R

(GSP1 ; 5'-AGCTGCCTCCCCAGC-TGCTTCTG-3' ; Table 1) and FVR1040R (GSP2; 5'-GGCCTGGTCTCGCTCCTTCTCAA-3'). The 3'RACE primers used were FVR1041F (GSP1 ; 5'-GGCCCAACTGGAGGAGATTGAGA-3', Table 1) and FVR1042F (GSP2; 5'-GCAAC-TGGGGAGGCAGCTGTCAT-3'). The fragments were further characterized by cloning into pGEM@-T cloning system (Promega, Madison, WI) and by subsequent DNA sequence analysis by using the M13 forward (5'- CGCCAGGGTTTTCCCAGTCACGAC-3' ; FVR283; Table 4) and M13 reverse primer (5'-TCACACAGGAAACAGCTATGAC-3' ; FVR284). The specificity of the fragments was determined using the DNAstar software packages (DNASTAR Inc, Madison, USA), and by the Staden gap4 software (Bonfield et al., 1995).

A Rapid-Screen cDNA library panel of human lung was used (OriGene Technologies, Inc., Rockville, MD) to elongate further the PLA2H12 cDNA. The screening was performed according to the manufacturers'protocol with the PLA_2H12-specific primers FVR731F (5'-CGCAGGACAGCAGGCAGGAG-3' ; Table 1) and FVR733R (5'- CAGGATGGAAAGCCGATTGA-3'). The PCR, in a reaction volume of 25 ul, contained 5 p. ! of template cDNA, 50 pmoles of the PLA_2H12-specific primers FVR731F and FVR733R (Table 1), 50 RM dXTPs, 5 U of Taq DNA polymerase (Boehringer) and the buffer supplied with the DNA polymerase. A programmable thermal cycler (GeneAmp PCR system 9700, Perkin Elmer Cetus) was used. The PCR program started with a DNA denaturating step at 94°C for 3 min. Cycling conditions were 94°C for 1 min, 60 for 1 min and 72°C for 1 min 30 sec. This was repeated for a total of 35 cycles and followed by a final extension step of 10 min at 72°C. Positive PCR reactions finally resulted in the isolation of plasmids containing the cDNA of interest. These plasmids were further sequenced using the following primers FVR733R, FVR1042F, FVR1498R, FVR1499F, FVR1500F, FVR1520F, FVR1521R and FVR1575 (Table 1). The sequences were further analyzed by BLAST search (Altschul et al., 1990) and by software packages DNAstar (DNASTAR Inc, Madison, USA) and Staden gap4 (Bonfield et al., 1995).

Finally, we used 5'RACE technology (GIBCO BRL, Life Technologies, Gent, Belgium) to further elongate the cDNA. The lacking 5'fragment was isolated using a gene- specific primer 5'-TCAGGGGCAGGTAAGGTGAC-3' (GSP1 = FVR1748R ; Table 1) to synthesize the first strand of the cDNA. We then performed a PCR with the primer set GSP2 5'-CCGTCGCACCTCTGTCATCA-3' (= FVR1520F ; Table 1) and anchor primer 5'-GAATTCGTCGACTAGTACGGGIIGGGIIGGGIIG-3' (= FVR239F; Table 5), followed

by a nested PCR with the gene-specific primer GSP3 5'- CCACGGCAGCGCAAGATGTC-3' (= FVR1749R; Table 1) and primer UAP 5'- GAATTCGTCGACTAGTAC-3' (= FVR240F; Table 5). This yielded 2 bands from LICR- HN6 mRNA and 1 from SW620 mRNA. The fragments were further characterized by cloning into pGEM@-T cloning system (Promega, Madison, WI) and by subsequent DNA sequence analysis by using the M13 forward (5'-CGCCAGGG- TTTTCCCAGTCACGAC-3' ; FVR283; Table 4) and M13 reverse primer (5'- TCACACAGGAAACAGCTATGAC-3' ; FVR284). Again, the specificity of the fragments was determined using the software packages DNAstar and Staden gap4.

Since 5'RACE experiments did not enable us to complete the 5'end of the human PLA2H12 cDNA sequence, another strategy was followed. Using. the Genscan software (http ://genes. mit. edu/GENSCAN. html) putative exons were identified in PLA_2H12-specific human genomic sequences available in public databases (Genbank Acc. No. AL022315). Based upon these sequences, primers were designed for RT-PCR amplification of the PLA2H12 cDNA. Forward primers used were FVR2313F, FVR2381F and FVR2382F, each in combination with either FVR1498R or FVR1039R. For primer sequences, see Table 1. PCR reactions were performed on cDNA from human cell lines SW620 (CCL-227) or MKN45 using pfu DNA polymerase (Stratagene). A typical PCR reaction contained 25 pmol of forward and reverse primers, 1 p110mM dXTP, 5 pLI 10 x reaction buffer, 5 ll1 SW620 cDNA, 2.5 U pfu DNA polymerase and deionized water to a final volume of 50 pI. PCR reactions were run on a PTC-200 thermal cycler (MJ Research) and PCR programs were as follows : an initial denaturation step at 95°C for 3 min, followed by 38 cycles of 45 sec at 94°C, 40 sec at 65°C and 90 sec at 75°C. Final extension was 10 min at 75°C. After purification, fragments were incubated with TAQ DNA polymerase and dXTP for 25 min at 72°C and this allowed, after purification of the fragments, to clone these fragments using the pGEM-T vector system (Promega). Cloned fragments were checked by subsequent sequencing. Alternatively, PCR products were directly sequenced using internal primers. DNA sequence analysis was performed as described above.

Human RAPID-ScanT"" ! n order to determine the expression profiles of the newly identified PLA_2H12, we used the human Rapid-Scan Gene expression panel (OriGene Technologies, Inc., Rockville, MD). This PCR-based system provides first strand cDNAs derived from 24

different tissues and/or developmental stages. The PCR, in a reaction volume of 25 ul, contained 5 Ll of a 100-times diluted template cDNA, 25 pmoles of the PLA_2H12- specific primers FVR731 F (Table 1; 5'-CGCAGGACAGCAGGCAGGAG-3') and FVR733R (5'-CAGGATGGAAAGCCGATTGA-3'), 200 tM dXTPs, 1.25 U of AmpliTaq DNA polymerase (Perkin Elmer) and the buffer supplied with the DNA polymerase. A programmable thermal cycler (GeneAmp PCR system 9700, Perkin Elmer Cetus) was used. The PCR program started with a DNA denaturating step at 94°C for 10 min.

Cycling conditions were 94°C for 30 sec, 60°C for 30 sec and 72°C for 1 min. This was repeated for a total of 35 cycles and followed by a final extension step of 10 min at 72°C. A band of 584 bp can be visualized on a 1% agarose gel when PLA_2H12 is expressed in the tissue examined. Nested PCR was done using primers FVR729R (Table 1; 5'-GCTGCCTCCCCAGTTGCTTC-3') and FVR1041F (Table 1; 5'- GGCCCAACTGGAGGAGATT-GAGA-3') yielding a PLA 2H 12-specific fragment of 332 bp. The nested PCR was done on 2.5 Ll (1/10 of the end volume) of the first PCR reaction. PCR reactions using ß-actin-specific primers, provided in the kit, were performed as a positive control on each cDNA, derived from the different tissues, examined.

Example 1: Two-hybrid cDNA library screening For the initial two-hybrid screen, the pGBT9Plako (227-1853) plasmid was used as bait. This comprises a large cDNA fragment, encoding amino acid residues 37-570 of the human plakoglobin, fused to the GAL4 DNA binding domain in the pGBT9 vector (Clontech). This bait plasmid was assayed for interaction with proteins encoded by a GAL4 activation domain cDNA library from human fetal kidney (Clontech). In a first experiment, three H/S3-positive, LacZ-positive clones were isolated. For two of these clones, the library plasmids could be isolated and the sequences of the inserts were determined. The first plasmid, designated by us PLA 2H01 (PLAkoglobin 2Hybrid clone 1), encoded almost the entire cytoplasmic domain of human N-cadherin. The other library plasmid, PLA 2H03, contained a fragment of human P-cadherin. We then performed a larger two-hybrid screen using the same bait to screen the same prey cDNA library from human fetal kidney (Clontech). About 2x106 clones were screened and examined for interaction with plakoglobin on the basis of induction of two genes: the selection gene HIS3 and the reporter gene LacZ. Thirty-two clones, designated PLA 2H04 to PLA_2H35, which exhibited the desired HlS3-positive, ß-galactosidase-

positive phenotype, were isolated out of this screen and were further examined.

Among these 32 clones, several known cadherins, like cadherin-11 and P-cadherin were identified (see also Table 6). One plasmid contained cDNA encoding a fragment of the tumor suppressor gene product APC. An interaction between plakoglobin and these cadherin and APC molecules was reported before (Rubinfeld et al., 1993; Su et al., 1993; Shibata et al., 1994; Rubinfeld et al., 1995). Several library plasmids contained fragments of cytokeratin 8 cDNA. The extracellular domain of human Notch 2 was also isolated as a putative interaction partner of plakoglobin. Two cDNA clones encoded nuclear proteins such as NRF1 (nuclear respiratory factor 1; Chan et al., 1993) and NDP52 (nuclear dot protein 52), also known as nuclear domain 10 protein (Korioth et al., 1995). These prey plasmids comprised, respectively, nt 1431-1984/aa 168-322 of human NRF-1 cDNA/protein (GenbankAcc. No. HUMNRF1A ; genpeptAcc.

No. A49672) and nt 556-1366/aa171-438 of human NDP52 (Genbank Acc. No. HSU22897) Interaction of plakoglobin with these protein fragments points at a nuclear function of plakoglobin. Interestingly, NRF1 belongs to the CNC-bZIP family of transcription factors forming a complex with transcription factors of the Fos/Jun family (Novotny et al., 1998 ; Venugopal and Jaiswal, 1998), whereas NDP52 has been reported to localize mainly in the cytoplasm besides the nucleus (Sternsdorf et al., 1997). However, the three most intriguing clones (PLA2H12, PLA2H15 and PLA2 H34) contained unknown cDNAs. Two of these (PLA2H12 and PLA_2H15) turned out to be identical ; their sequence (indicated between black arrows in Fig. 3) is disclosed in SEQ ID N° 5. These two identical clones were chosen for further examination. The sequence of PLA 2H34 (Figure 7) is disclosed as SEQ ID N° 7.

Example 2: Retransformation of the PLA2H12 clone in the yeast two-hybrid system The primary purpose of this test was to check the specificity of the interaction between plakoglobin and the polypeptides encoded by the abovementioned clones, more particularly NRF-1, NDP52, PLA_2H12 and PLA_2H34. The isolated two-hybrid plasmids were retransformed into the HF7c strain and were combined with various fragments of plakoglobin fused to the GAL4-DBD (Figures 1 and 8). Assuming that only few of the fragments of plakoglobin will interact, we considered the non-interacting fragments as negative and the interacting fragments as positive controls. In this way, we could also narrow down the PLA_2H12 binding region of plakoglobin.

The isolated two-hybrid clone PLA_2H12 was retransformed into the HF7c strain. This PLA-2H 12 prey clone was then combined with either one of the following bait plasmids (Fig. 1): the original bait plasmid pGBT9Plako (227-1853), the pGBT9-Plako (227-461), the pGBT9Plako (227-1028), the pGBT9Plako (227-1335), the pGBT9-Plako (1335- 1853), the pGBT9Plako (1335-2120), the pGBT9Plako (1853-2340), the pGBT9Plako (558-998) and the pGBT9-Plako (558-2120). The PLA_2H12 plasmid transformants exhibited the desired H/S3-positive,-galactosidase-positive phenotype upon combination with pGBT9Plako (227-1853) as expected, but also upon combination with the pGBT9Plako (227-1335), the pGBT9Plako (1335-1853) or the pGBT9Plako (558-2120) bait plasmids (Fig. 1). No interaction was seen upon combination with either pGBT9Plako (227-461), pGBT9-Plako (227-1028), pGBT9Plako (1335-2120), pGBT9Plako (1853-2340) or pGBT9Plako (558-998). The minimal binding domain of plakoglobin is therefore situated between halfway arm-repeat 7 and the first 5 aa of arm-repeat 11. However, there is a discrepancy between the results of the retransformations with either pGBT9Plako (1335-1853) or pGBT9Plako (1335-2120).

Upon combination of the first plasmid with PLA_2H12, both HIS3 and LacZ reporter genes are expressed, while the latter plasmid encoding a larger fragment of plakoglobin does not. A possible explanation lies in the three-dimensional structure of the fusion proteins. This is in line with the observation that combination of pGBT9Plako (1335-1853) with the other prey plasmids besides PLA_2H12 is also not indicative of interaction, in contrast to the interacting plasmids pGBT9Plako (227-1335) and pGBT9Plako (558-2120) (Figures 1 and 8). Stronger expression of the reporter genes was also seen when the original bait plasmid pGBT9Plako (227-1853) was retransformed together with PLA_2H12. or the other prey plasmids. The GAL4-DBD fusion proteins encoded by the pGBT9Plako (227-1335) and the pGBT9-Plako (1335- 1853) plasmids both interact with the polypeptide encoded by the prey plasmid PLA 2H12. On the other hand, the GAL4-DBD fusion proteins encoded by the pGBT9Plako (227-1335) and the pGBT9Plako (1335-2120) plasmids both interact with the polypeptides encoded by the prey plasmids NRF1 and NDP52. This can be explained by a bipartite binding site of plakoglobin for PLA 2H 12 (Fig. 1). Stronger binding is achieved when both domains are present. However, one of both domains is sufficient for the binding of PLA 2H12. The same holds true for interaction with the NRF1 and NDP52 polypeptides.

Example 3: Northern Blot Analysis In order to estimate the length of the complete mRNA of PLA_2H12, we performed a Northern blot analysis. Total RNA of various human cell lines and human tissues was hybridized with a 32P-labeled EcoRl-fragment of the PLA2H12 cDNA. A very weak signal could be detected in most cell lines or normal tissues (Fig. 2). However, a stronger signal was seen in the lanes with SW620, SW1116, DLD1R2/7 and LICR- HN3 RNA. The positions of 28S and 18S ribosomal RNA on the blot were visualized using methylene blue staining. Using these positions as markers, we could estimate the size of the PLA2H12 mRNA to be about 3.7 Kb.

Example 4 : Isolation of the full-length cDNA of PLA 2H12 and in silico analvsis of the predicted protein sequence The PLA_2H12 plasmid, as isolated from the yeast two-hybrid screen, contained a cDNA insert of about 800 bp (flanked by black arrows in Fig. 3). This cDNA fragment revealed a complete open reading frame, but neither a start, nor a stop codon was present in this sequence. As the size of the full-length cDNA was estimated by Northern blot analysis to be 3.7 Kb, this cDNA fragment needed to be elongated at both sides. In a first approach, conventional screening of a phage library of human fetal kidney cDNA did not yield any PLA2H12-specific sequence. Subsequently, 5' and 3'RACE experiments were performed. Several specific fragments were isolated, but most fragments contained intronic sequences, and were therefore derived from genomic contamination of the cDNA preparation.

Screening a master plate of a Rapid-Screen cDNA library panel of human lung resulted in the identification of two PLA_2H12-positive subplates. One subplate revealed two signals upon PCR with PLA_2H12-specific primers. Three plasmids were isolated, containing PLA2H12-specific inserts. Analysis of the inserts was done using the DNAstar and Staden gap4 software packages. Further completion of the PLA2H12 cDNA by RT-PCR experiments resulted in a 3,948-nt cDNA molecule, in agreement with the size of the mRNA estimated by Northern blotting detection (Fig. 2).

The ATG codon at position 45 appears to be a genuine translation initiation site, including surrounding DNA sequence corresponding partially to the optimal Kozak context (Kozak, 1997). The open reading frame encodes a protein of 1032 aa. A stop

codon was clearly detected, together with a poly-A signal followed by a poly-A tail. This suggests that the 3'untranslated region (3'UTR) is also completely isolated.

The cDNA sequence of PLA_2H12 as depicted in Fig. 3, did not correspond with any cloned full-size cDNA in the public-domain databases. However, several EST (Expressed Sequence Tag) clones with high degree of sequence similarity (low P- score) are present in the databases of human sequences (Table 7) and mouse sequences (Table 7b). The same is true for the partial cDNA'sequence of PLA 2H34 (Table 9).

The amino acid sequence, predicted on the basis of the cDNA sequence, was further analyzed using the PSORT program (http ://sort. nibb. ac. jp/cgi-bin/runpsort. pl). The most interesting feature predicted by the program was a stretch rich in basic amino acids (Arg, Lys) that was annotated as a putative bipartite nuclear localization signal (residues 252-273; Fig. 3). Indeed, the cellular localization as predicted by this program was nuclear, with a reliability of 89%. Moreover, the program also predicts the presence of two leucine zipper domains (Fig. 3). Further study, using the ProDom database of the WU-BLASTP program (http ://protein. toulouse. inra. fr/prodom/cgi- bin/NewBIastProdomil. pl) revealed homology of residues 707-770 with an SH3 (Src homology 3) domain, known as a protein-protein interaction module (Morton and Campbell, 1994). SH3 domains mediate assembly of protein complexes through binding of proline-rich peptides. The highest homology was observed with the SH3 domain of the tight junction proteins ZO-1, ZO-2 and ZO-3 (Fig. 4) (Anderson, 1995; Anderson, 1988; Haskins et al., 1998).

The small SH3 domains of about 60 aa in length were first identified in the non- catalytic part of several cytoplasmic protein tyrosine kinases, as Src, Abl and Lck.

Later on, it has been found in a great variety of other intracellular or membrane- associated proteins (Pawson and Schlessinger, 1993; Pawson, 1995b). SH3 domains may regulate protein localization or enzymatic activity, and often participate in the formation of multiprotein signaling complexes (Morton and Campbell, 1994 ; Pawson, 1995a). This protein interaction module can be implicated, like other signaling domains such as SH2, PTB, PDZ and WW domains, in a diverse array of signaling events (Pawson, 1995a). The interaction of plakoglobin with the newly identified protein PLA_2H12 with interesting domain structure, might explain the reported but unexplained signaling activity of plakoglobin. Like-catenin, plakoglobin can bind to LEF/TCF transcription factors and influence transcription of target genes (Behrens et

a/., 1996; Huber et al., 1996 ; Molenaar et al., 1996). In contrast, it was shown that even membrane-anchored plakoglobin can induce a Wnt-like phenotype in Xenopus (Merriam et a/., 1997). The latter study shows that nuclear translocation of plakoglobin is not required to induce axis duplication in the early Xenopus embryo. Moreover, this membrane-anchored plakoglobin did not induce an increase in cytoplasmic, nor in nuclear P-catenin (Merriam et al., 1997). Hence, the plakoglobin-mediated signal should be transduced into the nucleus via other mechanisms, such as the interaction with the PLA 2H12 protein. The original PLA 2H12 clone did not contain the putative nuclear localization signal, nor the SH3 domain. Moreover, one of both leucine zipper motifs was also excluded from the clone, leaving the second leucine zipper motif as the only domain present. The PLA_2H12 protein might function in this signaling pathway through its SH3 and leucine zipper domains. Another possible mechanism implies the nuclear translocation of very low levels of the plakoglobin protein through its interaction with PLA 2H12, as the latter contains a putative nuclear localization signal.

Example 5 : In silico analysis of two human genomic clones specific for PLA 2H12 Two human PLA_2H12-specific genomic sequences (Genbank Acc. No. AL049851 and AL022315) were identified on the basis of a BLASTN search analysis of GenBank (Fig. 5). The PLA_2H12 gene is partly represented in these two BAC clones, which reside within the human chromosomal region 22q13. 1. We performed an in silico analysis of these human genomic sequences for the PLA 2H12 gene. The exon-intron boundaries of the PLA_2H12 gene could be determined by comparison between the cDNA and genomic sequences (Table 8). They all turned out to be consistent with the ag-gt rule (Mount, 1982). The splice donor/acceptor probability scores were determined according to Shapiro and Senapathy (Shapiro and Senapathy, 1987) (Table 8). The gene consists of 21 exons identified so far, ranging in size from 27 nt (exon 11) to 1,032 nt (exon 21). The length of intron 11 and intron 12 could not be determined, because the PLA_2H12 gene is only partly represented in these two non- overlapping BACs (Table 8).

Example 6: PLA 2H12 expression analysis

Using the human Rapid-Scan panel, we tried to investigate the expression pattern of the novel plakoglobin-binding protein PLA_2H12 (Fig. 6). No PLA_2H12-specific bands could be detected in the cDNA pool of brain or fetal brain, nor in adrenal gland, muscle, stomach, bone marrow and fetal liver. A weak signal was visualized using heart, placenta, ovary and prostate cDNA as a template. Strong to very strong signals were detectable in cDNA from kidney, spleen, liver, lung, colon, small intestine, testis, salivary gland, thyroid gland, pancreas, uterus, skin and PBL (peripheral blood lymphocytes) (Fig. 6).

Example 7: Transformation of the full length human PLA 2H12 and fragments thereof, in the yeast two-hybrid system using yeast strain AH109 In order to assess whether the full-length PLA_2H12 was still interacting with human plakoglobin fragments, transformation assays in the yeast two-hybrid system were performed. Interaction of plasmid products was assessed on SD plates minus leucine (L), tryptophan (W), histidine (H) and adenine (Ade), and including blue staining as reporter assay. Plating on SD-LW was performed to assess growth of the transformed yeast clones. Since all PLA 2H12 constructs in pGBKT7 were auto-activating, interaction assays were performed using the PLA 2H12 constructs in pGADT7. In interaction assays in yeast strain AH109 using the original PLA_2H12 prey plasmid, a clear-cut interaction could be observed with plasmids pGBT9Plako (227-1335), pGBT9Plako (227-1853) and pGBT9Plako (558-2120) (Fig. 9). In contrast, using the pGADT7hPLA_2H12Original prey plasmid in these assays, interaction could only be observed with plasmid pGBT9Plako (227-1335) (Fig. 9). This can be explained by absence of growth of several of the transformed yeast clones assayed here, which is clear from the SD-LW plates (Fig. 9b'). Most convincingly, the full-length human PLA_2H12 protein (pGADT7hPLA_2H12FL) still interacts with plakoglobin fragments, as can bee seen in Fig. 9. The full-length human PLA_2H12 interacts clearly with pGBT9Plako (227-1335), pGBT9Plako (227-1853), which is the original bait plasmid, while a weaker interaction is observed with pGBT9Plako (227-1028) and pGBT9Plako (558-2120). Based on these results, part of the Armadillo repeats of plakoglobin appears to be responsible for the interaction with the PLA_2H12 protein. Intriguingly, all PLA 2H12 fragments tested in the yeast two-hybrid assay appear to interact with the human Armadillo-like proteins plakophilin-1 (PKP1) and plakophilin-3 (PKP3). The PKP1 and PKP3 proteins belong to the superfamily of Armadillo-related proteins, to

which also plakoglobin belongs. This also suggests the Armadillo repeats to be important for interaction with the PLA 2H12 protein, since this is the most conserved domain in these proteins.

Table 1: Primers used for the characterization of the human PLA 2H12 cDNA Primera Application Sequence 5' 3'" FVR702F SE gcaggacagcaggcaggagc FVR703R SE aagtgctgctgaggttggag FVR704F SE gagatggaagacctgcggct FVR707R SE aggggaactcgctcaggctc FVR727F SE atggaaaacctgcggctcaa FVR728F SE ttgagaaggagcgagaccag FVR729R SE gctgcctccccagttgcttc FVR730R SE ggcctggtctcgctccttct FVR731 F SE, PCR cgcaggacagcaggcaggag FVR732F SE tggaggcggagcgggatgag FVR733R SE, PCR caggatggaaagccgattga FVR1039R 5'RA agctgcctccccagctgcttctg FVR1040R 5'RA ggcctggtctcgctccttctcaa FVR1041F 3'RA ggcccaactggaggagattgaga FVR1042F SE, 3'RA gcaactggggaggcagctgtcat FVR1498R SE ccgccagtcatgctctagga FVR1499F SE cccgcaccccctaagagatc FVR1500F SE caagcggaggcaggaatggttc FVR1520F 5'RA, SE ccgtcgcacctctgtcatca FVR1521R SE gggccatccaggagtctgtt FVR1575 SE tgcccactcccctctacttg FVR1595F PCR ccgggtcatcgacgagcag FVR1640F RT-PCR agtggggacacgcagaggag FVR1641R RT-PCR tcaggggcaggtaaggtgac FVR1748R 5'RA tcaggggcaggtaaggtgac FVR1749R 5'RA ccacggcagcgcaagatgtc FVR1902R SE cacgggaagcggtaggt FVR1941F SE gggcgggcggcaatgtgg FVR2313F RT-PCR gtctgaggcggaggaggac FVR2381 F RT-PCR gagcccgaggcgactgtag FVR2382F RT-PCR gaggacacggccatgccg FVR2486F RT-PCR ataagaattcggagcccgaggcgactgta Legend to Table 1: a F (Forward) and R (Reverse) refers to the sense or antisense orientation of the <BR> <BR> primers.<BR> <BR> <BR> <BR> <P> Application of the primers: 5'RA, 5'RACE; 3'RA, 3'RACE; SE, sequencing; RT- PCR, reverse transcription polymerase chain reaction.

° Restriction sites added are underlined and in bold.

Table 2a: Primers used for in-frame cloning of human plakoqlobin cDNA into pGBT9 vector Primera Positionb Applicationc Sequence 5'-> 3 d FVR133F 228 PCR/CL ggtgaattcgtcagcagcaagggcatcat FVR134R 1228 PCR/CL ggtttgatgcagggtccacaggcagttct a R (Reverse) and F (Forward) refers to the sense or antisense orientation of the <BR> <BR> primers.<BR> <BR> <BR> <P> The position of the most 5'nucleotide is given. Sequences are numbered according to the cDNA starting from the putative transcription initiation site. c PCR/CL, cloning of cDNA fragments obtained by PCR d Restriction sites added are underlined and in bold.

Table 2b: PCR Primers used for in-frame cloning of the full length human plakophilin-3 open reading frame into the pGBKT7 and pGADT7 vectors <BR> <BR> <BR> <BR> <BR> <BR> Primera Sequence 5'#3'b<BR> <BR> <BR> <BR> <BR> FVR1846F atacqaattccaggacggtaacttcctg<BR> <BR> <BR> <BR> FVR1847Ratacatcgacacagccaacccccacctct a R (Reverse) and F (Forward) refers to the sense or antisense orientation of the <BR> <BR> primers.<BR> <BR> <BR> <P> Restriction sites added are underlined and in bold.

Table 3: Primers used for sequencing the inserts of two-hvbrid vectors Primer Vector Sequence 5'# 3' FVR174F pGAD10 accactacaatggatgatgt FVR175F pGBT9 atcatcggaagagagtagta FVR192R pGAD10 taaaagaaggcaaaacgatg FVR217R pGBT9 aaaatcataaatcataagaa Table 4: M13F/R primers used for sequencing the inserts of pGEM@-T clones Primer Application Sequence 5'-'-3' FVR283 SE cgccagggttttcccagtcacgac FVR284 SE tcacacaggaaacagctatgac Table 5: primers used for 5'RACE Primer Application Specificty Sequence 5'-> 3' FVR239F 5'RACE Anchor gaattcgtcgactagtacgggiigggiigggiig FVR240F 5'RACE UAP gaattcgtcgactagtac Table 6: PLA 2H clones isolated from a GAL4-DBD yeast two-hybrid fetal kidney cDNA library IDENTIFICATION FRAGMENT (NT) FRAGMENT (AA) PLA_2H01 human N-cadherin nt 1801-2696 aa 601-stop PLA_2H03 human P-cadherin nt 2138-2858 aa 695-stop PLA_2H04 human P-cadherin nt 2138-2858 aa 695-stop PLA_2H05 human NRF1 protein nt 1431-1984 aa 168-322 PLA_2H07 human cytokeratin 8 nt 419-1221 aa 119-387 PLA_2H08 human OB-cadherin = cadherin-11 nt 2507-3035 aa 689-stop PLA_2H10 human OB-cadherin = cadherin-11 nt 2484-3556 aa 681-stop PLA_2H12 unknown cDNA, slight homology to human envoplakin 837 nt 270 aa PLA_2H14 human OB-cadherin = cadherin-11 nt 2371-3661 aa 643-stop PLA_2H15 unknown cDNA, slight homology to human enveoplakin 837 nt 270 aa PLA_2H17 rat Notch2 (extracellular EGF-like domain) nt 4426-5472 aa 1305-1654 PLA_2H21 human nuclear domain 10 protein (ndp52) nt 556-1366 aa 171-438 PLA_2H34 unknown cDNA ~ 2000 nt - PLA_2H35 human APC tumor suppressor nt 5037-6194 aa 1673-2058 Table 7: Human EST clones with high sequence homology to PLA 2H12 (Part 1 of 2) EST-ID CloneID NCBI-ID Genbank-ID P-score Tissue wv03b02.x1 2528427 3' 5877877 AW024347 0.0 kindey wx06g05.x1 2583512 3' 6038303 AW083151 0.0 colon UI-H-BII-acu-c-12-UI.s1 2715646 3' 6141140 AW137007 0.0 mixture of 21 libraries wi54c04.x1 2394054 3' 5177871 AI762204 0.0 colon td15a02.x1 2075690 3' 5436724 AI817645 0.0 colon UI-H-BI1-acy-a-06-UI.s1 2715922 3' 6142529 AW138129 0.0 mixture of 21 libraries tg07b10.x1 2108059 3' 4393114 AI492111 0.0 B-cell chronic lymphotic leukemia UI-H-BI1-aem-h-09-0-UI.s1 2720104 3' 6504625 AW205153 0.0 mixture of 21 libraries wd06f03.x1 2327357 3' 4988245 AI700345 0.0 colon wc35f02.x1 2317179 3' 5364095 AI798623 0.0 prostate wi28h04.x1 2391607 3' 5101394 AI739413 0.0 colon wl18b06.x1 2425235 3' 5450122 AI829451 0.0 uterus nr94g08.s1 1175678 2568151 AA642933 0.0 prostate oJ70a11.s1 1503644 3' 3042360 AA906900 0.0 fetal lung, testis, B-cell UI-H-BI1-acs-g-06-0-UI.s1 2715466 3' 6141753 AW137435 0.0 mixture of 21 libraries zt86a03.s1 729196 3' 2052662 AA399648 0.0 testis qg83g10.x1 1841826 3' 3802165 AI219962 0.0 fetal lung, testis, B-cell zd35d05.s1 342633 3' 1377564 W68684 0.0 heart ah74e03.s1 1321372 3' 2807054 AA759191 0.0 testis wq17d03.x1 2471525 3' 5741224 AI948914 0.0 kidney zm27e04.s1 526878 3' 1664731 AA113168 0.0 pancreas th34h12.x1 2120231 3' 4307692 AI432082 0.0 pancreas ty98h02.x1 2287155 3' 4685798 AI634468 0.0 uterus yj11c07.r1 148428 5' 877182 H1362 0.0 human qz36f05.x1 2028993 3' 3871418 AI263215 E-175 kidney qz36e08.x1 2028998 3' 3871410 AI263207 E-172 kidney zd53e07.s1 344388 3' 1383521 W73388 E-169 fetal heart yj11c07.s1 148428 3' 877129 H12309 E-166 human wx78c01.x1 2549760 3' 5746299 AI953989 E-163 ovary Table 7: Continued (Part 2 of 2) wq17c03.x1 2471524 3' 5741214 AI948904 E-163 kidney zd53e07.r1 344388 5' 1383581 W73449 E-150 fetal heart af11a11.s1 1031324 3' 2457513 AA609085 E-147 testis tr96g08.x1 2226974 3' 4569319 AI583422 E-141 pancreas zt86a04.r1 729198 5' 2050625 AA397734 E-129 testis Legend: EST-ID: identification number of EST (Expressed Sequence Tag); Clone ID: identification number of the clone; NCBI-ID:<BR> identification number according to NCBI database; Genbank-ID: identification number according to the Genbank datatase; P-score:<BR> index which indicates sequence homologies to the BLAST algorithm (Altschul et al., 1990); Tissue: tissue used to isolate EST<BR> Table 7b: Mouse DNA clones with high sequence homology to PLA 2H12 ID Genbank-ID P-score Mus musculus chromosome 11 clone RP23-430J8 AC026386 E-104 Mus musculus chromosome 11 clone RP23-324E9 AC025911 3E-52 Table 8: Overview of the exons and introns of the human PLA 2H12 gene (part 1 of 2) Intron Exon Scoreb Intron # (bp) Exon # (bp) Splice donora Splice acceptora Splice Splice donor acceptor 1 n.d. 27 28 1 125 ACTGTAGCGTGCGGG gtgagtcgcg ..gcctcttgcccgcag ACCCTGAGGAC 92 89 5'UTR 5'UTR 2 252 279 280 82 82 2 857 GTC-AAC-CGC-ACC-G-- gtgagccgcg.. ..cccactactctgtag -GG-CGC-CTG-ATG 82 82 V N R T G R L M 3 137 417 418 3 1,663 TCC-ATG-ATC-CTC-G-- gtgagtggat.. ..gttctgtcctggcag -AT-GAG-GAG-GGG 88 87 S M I L D E E G 4 325 743 744 4 5,552 GAC-CTG-CAG-CTG-GCG gtaggccctg.. ..tgtcttgggctccag GTG-GAT-CAG-CTC 75 86 D L Q L A V D Q L 5 209 954 68 87 5 1,530 GAG-GGC-CTG-CAG-CAG gtaccagggc.. ..ccctgtgatccccag GAG-CGC-AGC-CGG 68 87 E G L Q Q E A S R 6 155 1109 1110 8 179 6 572 GAG-CTG-CGC-GAC-CAG gtggggccct.. ..ccgtgccaccctcag TAC-CTG-CAG-GAG 81 79 E L R D Q Y L Q E 7 125 1235 1236 7 1,446 AAG-GAG-CGA-GAC-CAG gtgagcccag.. ..ctccccaaccctag GCC-ATC-CAG-AGC 91 79 K E R D Q A I Q S 8 191 1427 1428 8 1,422 GGC-ACC-TGC-CTC-AAG gtgagcgggt.. ..ctctccctcttccag GCC-TGT-GCC-TCC 91 95 G T C L K A C A S 9 134 1562 1563 9 334 GAG-CCT-CAC-AAC-TCG gtaagacaac.. ..gtcatggcattgcag GAG-GAA-GCC-ACA 86 86 E P H N S E E A T 10 116 1679 1680 10 973 GCA-CCC-CCT-AAG-AG- gtgagatggc.. ..cctctctctgcccag --A-TCC-TTC-AGC 91 89 A P P K R S F S 11 27 1707 1708 11 460 TCA-GAC-ATC-ACA-G-- gtaacagccc.. ..attctttttccccag -GG-AGT-GTG-ACA 80 97 S D I T G S V T Table 8: Overview of the exons and introns of the human PLA 2H12 gene (part 2 of 2) 12 123 1831 1832 12 5424 GAC-TTC-CTC-AAC-AG- gtactgtagc.. ..tttctcccttctcag --G-TCT-CTG-GCT 69 96 D F L N R S L A 13 159 1991 1992 13 457 ATG-GAA-CCA-AGA-GAG gtgaggccct.. ..tgggttttctcccag CAA-AGG-GTG-GAA 92 87 M E P R E Q R V E 14 111 2103 2104 14 447 ATG-GAC-TCG-AAG-G-- gtaatctggc.. ..cccatccatgcccag -CC-TGC-CAG-TCC 86 79 M D S K A C Q S 15 243 2347 2348 15 226 CCC-AAT-TAT-CAG-AG- gtgacgcaga.. ..cactccctcccacag --A-GCC-CAG-CAG 78 87 P N Y Q R A Q Q 16 69 2417 2418 78 85 16 1,329 GGC-CCC-CGC-AGT-AAT gtgagcacgg.. ..cctgcaccctcccag CTG-AAG-AAG-AGA 78 85 G P R S N L K K R 17 102 2520 2521 96 85 17 1,281 GAG-CCC-TGT-GCA-G-- gtgagtgcgc.. ..cctcccctgccacag -AG-CCG-GAG-CGG 96 85 E P C A E P E R 18 159 2680 2681 18 93 GTG-TGC-CCA-GCG-G-- gtgagactat.. ..ccactccctgtgcag --AA-AGC-CTC-TCT 83 87 V C P A E S L S 19 114 2795 2796 19 566 GAG-TGT-GTT-GGG-AAG gtgggtgctg.. ..cgatcctccaagcag AAG-CAC-TGC-CTG 86 77 E S V G K K H C L 20 111 2907 2908 20 327 GTC-CGG-GAA-GTC-AG- gtgaggtggg.. ..ctcttctctcggcag --G-GGT-CTG-CTG 92 96 V R E V R G L L 21 1032 TGF TGA (3141) - 3' END (polyA) Legeng:<BR> aIntron sequences in lowercase letters. Exon sequeces in capital letters. Splice consensus sequences (ag-gt) in bold. Amino acid residues<BR> (single letter codes) are indicated below the corresponding codons.<BR> bScores for donor and acceptor splice sites according to the method of Shapiro and Senapathy (Shapiro and Senapathy, 1987). n.d.: not<BR> determined.

Table 9: Human EST clones with high sequence homology to PLA 2H34 EST-ID CloneID NCBI-ID Genbank-ID P-score Tissue DKFZp434L1816_r1 DKFZp434L1816 5' 5410188 AL041260 0.0 testis zb47c02.r1 306722 5' 1300802 W23919 0.0 fetal lung zt05b06.r1 712211 5' 1921970 AA280332 0.0 germinal center B cell zs24d05.r1 686121 5' 1898404 AA262843 E-164 germinal center B cell HSC2HF111 c-2hf11 574342 Z45140 0.0 normalized infant brain aa35g09.r1 8152960 5' 2211166 AA481614 E-152 germinal center B cell EST28439 613282 T31184 E-141 brain yw64d03.s1 256997 3' 1148754 N30234 E-137 placenta 8 to 9 weeks qf65a10.x1 1754874 3' 3753802 AI201196 E-137 testis xy09b06.x1 2852627 3' 7044173 AW474067 E-137 lymphoma, follicular mixed small and large cell ou28d11.x1 1627605 3' 3231528 AI017192 E-137 fetal lung, testis and B-cell qt26f06.x1 1949123 3' 4079160 AI342233 E-135 pregnant uterus ns53b11.s1 1187325 3' 2589847 AA653693 E-135 normal prostate oo89g05.s1 1573400 3' 3144940 AA969760 E-135 kidney he27g12.x1 2920294 3' 7276339 AW589233 E-135 myeloid cells of blood op28a02.s1 1578122 3' 3147969 AA972789 E-135 fetal lung, testis and B-cell qe14a07.x1 1738932 3' 3739230 AI188021 E-134 testis HSC02H021 c-02h02 668411 F05165 E-132 normalized infant brain yy66g01.s1 278644 3' 1218286 N66161 E-131 multiple sclerosis ne65d01.s1 909121 2211525 AA482680 E-131 alveolar rhabdomyosarcoma np69f09.s1 1131593 3' 2555247 AA633387 E-131 breast yy66c01.s1 278496 3' 1218260 N66135 E-131 multiple sclerosis

REFERENCES Altschul, Stephen F., Thomas L. Madden, Alejandro A. Schaffer, Jinghui Zhang, Zheng Zhang, Webb Miller, and David J. Lipman (1997),"Gapped BLAST and PSI-BLAST : a new generation of protein database search programs", Nucleic Acids Res. 25: 3389- 3402.

Anderson, J. M., Stevenson, B. R., Jesaitis, L. A., Goodenough, D. A., and Mooseker, M. S. 1988. Characterization of ZO-1, a protein component of the tight junction from mouse liver and Madin-Darby canine kidney cells. J. Cell Biol. 106: 1141-1149.

Anderson, J. M., Fanning, A. S., Lapierre, L., and Vanitallie, C. M. 1995. Zonula occludes (ZO)-i and ZO-2 : membrane-associated guanylate kinase homologues (MAGuks) of the tight junction. Biochem. Soc. Trans. 23: 470-475.

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.

Becker, I., Becker, K. F., Rohr, M. H., Minkus, G., Schütze, K., and Hofler, H. 1996.

Single-cell mutation analysis of tumors from stained histologic slides. Lab. Invest.

75: 801-807.

Becker, K. F., Atkinson, M. J., Reich, U., Becker, I., Nekarda, H., Siewert, J. R., and Höfler, H. 1994. E-cadherin gene mutations provide clues to diffuse type gastric carcinomas. Cancer Res. 54: 3845-3852.

Behrens, J., Mareel, M. M., van Roy, F. M., and Birchmeier, W. 1989. Dissecting tumor cell invasion: Epithelial cells acquire invasive properties after the loss of uvomorulin- mediated cell-cell adhesion. J. Cell Biol. 108: 2435-2447.

Behrens, J., von Kries, J. P., Kühl, 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.

Berx, G., Cleton-Jansen, A.-M., Nollet, F., de Leeuw, W. J. F., van de Vijver, M. J., Cornelisse, C., and van Roy, F. 1995. E-cadherin is a tumor/invasion suppressor gene mutated in human lobularbreastcancers. EMBO J. 14: 6107-6115.

Berx, G., Cleton-Jansen, A.-M., Strumane, K., de Leeuw, W. J. F., Nollet, F., van Roy, F. M., and Cornelisse, C. 1996. E-cadherin is inactivated in a majority of invasive human lobular breast cancers by truncation mutations throughout its extracellular domain. Oncogene 13: 1919-1925.

Bierkamp, C., McLaughlin, K. J., Schwarz, H., Huber, O., and Kemler, R. 1996.

Embryonic heart and skin defects in mice lacking plakoglobin. Dev. Biol. 180: 780-785.

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., Nollet, F., Kools, P., and van Roy, F. 1999. Plakophilin-3, a novel Armadillo-like protein present in nuclei and desmosomes of epithelial cells. J.

Cell Sci. 112: 2265-2276.

Brosius, J.; Dull, T. J.; Sleeter, D. D.; Noller, H. F. 1981. Gene organization and primary structure of a ribosomal RNA operon from Escherichia coli, J. Mol. Biol. 148: 107-127.

Bussemakers, M. J. G., Van de Ven, 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.

Butz, S., Stappert, J., Weissig, H., and Kemler, R. (1992) Plakoglobin and beta- catenin: distinct but closely related [letter]. Science 257: 1142-1144.

Chan, J. Y., Han, X.-L., and Kan, Y. W. 1993. Cloning of Nrf1, an NF-E2-related transcription factor, by genetic selection in yeast. Proc. Natl. Acad. Sci. U. S. A. 90: 11371-11375.

Chitaev, N. A., Leube, R. E., Troyanovsky, R. B., Eshkind, L. G., Franke, W. W., and Troyanovsky, S. M. 1996. The binding of plakoglobin to desmosomal cadherins: Patterns of binding sites and Topogenic potential. J. Cela bol 133: 359-369.

Cowin, P. 1994. Unraveling the cytoplasmic interactions of the cadherin superfamily.

Proc. Natl. Acad. Sci. U. S. A. 91: 10759-10761.

Easty, D. M., Carter, G. C., Monagham, P., and Butler, L. J. 1981. Ten human carcinoma cell lines derived from squamous carcinomas of the head and neck. Br. J. Cancer 43 : 772-785.

Franke, W. W., Goldschmidt, M. D., Zimbelmann, R., Mutter, H. M., Schiller, D. L., and Cowin, P. 1989. Molecular cloning and amino acid sequence of human plakoglobin, the common junctional plaque protein. Proc. Natl. Acad. Sci. U. S. A. 86: 4027-4031.

Frixen, U. H., Behrens, J., Sachs, M., Eberle, G., Voss, B., Warda, A., Locher, D., and Birchmeier, W. 1991. E-cadherin-mediated cell-cell adhesion prevents invasiveness of human carcinoma cells. J. Cell Biol. 113: 173-185.

Gietz, D., St Jean, A., Woods, R. A., and Schiestl, R. H. 1992. Improved method for high efficiency transformation of intact yeast cells. Nucleic Acid Res. 20: 1425.

Green, K. J., and Gaudry, C. A. 2000. Are desmosomes more than tethers for intermediate filaments ? Nature Rev. Mol. Cell Biol. 1: 208-216.

Haegel, H., Larue, L., Ohsugi, M., Fedorov, L., Herrenknecht, K., and Kemler, R. 1995.

Lack of beta-catenin affects mouse development at gastrulation. Development 121: 3529-3537.

Haskins, J., Gu, L. J., Wittchen, E. S., Hibbard, J., and Stevenson, B. R. 1998. ZO-3, a novel member of the MAGUK protein family found at the tight junction, interacts with ZO-1 and occludin. J. Cell Biol. 141: 199-208.

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.

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.

Kemler, R. 1992. Classical cadherins. Semin. Cell Biol. 3: 149-155.

Knudson, A. G. 1985. Hereditary Cancer, oncogenes, and antioncogenes. Cancer Res.

45: 1437-1443.

Korioth, F., Gieffers, C., Maul, G. G., and Frey, J. 1995. Molecular characterization of NDP52, a novel protein of the nuclear domain 10, which is redistributed upon virus infection and interferon treatment. J. Cell Biol. 130: 1-13.

Korman, N., Eyre, R. W., Klaus-Kovtun, V., and Stanley, J. R. 1989. Demonstration of an adhering junction molecule (plakoglobin) in the autoantigens of pemphigus foliaceous and pemphigus vulgaris. New Engl. J. Med. 321: 631-635.

Kowalczyk, A. P., Bornslaeger, E. A., Borgwardt, J. E., Palka, H. L., Dhaliwal, A. S., Corcoran, C. M., Denning, M. F., and Green, K. J. 1997. The amino-terminal domain of desmoplakin binds to plakoglobin and clusters desmosomal cadherin-plakoglobin complexes. J. Cell Biol. 139: 773-784.

Kowalczyk, A. P., Hatzfeld, M., Bornslaeger, E. A., Kopp, D. S., Borgwardt, J. E., Corcoran, C. M., Settler, A., and Green, K. J. 1999. The head domain of plakophilin-1

binds to desmoplakin and enhances its recruitment to desmosomes-Implications for cutaneous disease. J. Biol. Chem. 274: 18145-18148.

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.

Love, J. J., Li, X., Case, D. A., Giese, K., Grosschedl, R., and Wright, P. E. 1995.

Structural basis for DNA bending by the architectural transcription factor LEF-1. Nature 376: 791-795.

Mathur, M., Goodwin, L., and Cowin, P. 1994. Interactions of the cytoplasmic domain of the desmosomal cadherin Dsg1 with plakoglobin. J. Biol. Chem. 269: 14075-14080.

McCrea, P. D., Turck, C. W., and Gumbiner, B. 1991. A homolog of the armadillo protein in Drosophila (plakoglobin) associated with E-cadherin. Science 254: 1359- 1361.

McGrath, J. A., McMillan, J. R., Shemanko, C. S., Runswick, S. K., Leigh, l. 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. Nature Gen. 17: 240-244.

McKoy, G., Protonotarios, N., Crosby, A., Tsatsopoulou, A., Anastasakis, A., Coonar, A., Norman, M., Baboonian, C., Jeffery, S., and McKenna, W. J. 2000. Identification of a deletion in plakoglobin in arrhythmogenic right ventricular cardiomyopathy with palmoplantar keratoderma and woolly hair (Naxos disease). Lancet 355: 2119-2124.

Merriam, J. M., Rubenstein, A. B., and Klymkowsky, M. W. 1997. Cytoplasmically anchored plakoglobin induces a WNT-like phenotype in Xenopus. Dev. Biol. 185: 67- 81.

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 Biol.

135: 1009-1025.

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. Cell 86 : 391- 399.

Morton, C. J., and Campbell, I. D. 1994. SH3 domains: molecular'velcro'. Curr. Biol.

4: 615-617.

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.

Mount, S. M. 1982. A catalogue of splice junction sequences. Nucleic Acid Res.

10: 459-472.

Nollet, F., Kools, P., and van Roy, F. 2000. Phylogenetic analysis of the cadherin superfamily allows identification of six major subfamilies besides several solitary members. J. Mol. Biol. : In press.

Novotny, V., Prieschl, E. E., Csonga, R., Fabjani, G., and Baumruker, T. 1998. Nrf1 in a complex with fosB, c-jun, junD and ATF2 forms the AP1 component at the TNF alpha promoter in stimulated mast cells. Nucleic Acid Res. 26: 5480-5485.

Ozawa, M., Ringwald, M., and Kemler, R. 1990. Uvomorulin-catenin complex formation is regulated by a specific domain in the cytoplasmic region of the cell adhesion molecule. Proc. Natl. Acad. Sci. U. S. A. 87: 4246-4250.

Pawson, T. 1995a. Protein modules and signaling networks. Nature 373: 573-580.

Pawson, T. 1995b. Protein-tyrosine kinases-getting down to specifics. Nature 373: 477-478.

Pawson, T., and Schlessinger, J. 1993. SH2 and SH3 domains. Curr. Biol. 3: 434-442.

Peifer, M. 1993. Cancer, catenins, and cuticle pattern-a complex connection. Science 262: 1667-1668.

Peifer, M. 1997. Beta-catenin as oncogene : the smoking gun. Science 275: 1752-1753.

Rubinfeld, B., Souza, B., Albert, I., Muller, O., Chamberlain, S. H., Masiarz, F. R., Munemitsu, S., and Polais, P. 1993. Association of the APC gene product with beta- catenin. Science 262: 1731-1734.

Rubinfeld, B., Souza, B., Albert, I., Munemitsu, S., and Polais, P. 1995. The APC protein and E-cadherin form similar but independent complexes with alpha-catenin, beta-catenin, and plakoglobin. J. Biol. Chem. 270: 5549-5555.

Ruiz, P., Brinkmann, V., Ledermann, B., Behrend, M., Grund, C., Thalhammer, C., Vogel, F., Birchmeier, C., Gunthert, U., Franke, W. W., and Birchmeier, W. 1996.

Targeted mutation of plakoglobin in mice reveals essential functions of desmosomes in the embryonic heart. J. Cell Biol. 135: 215-225.

Sanger, F. 1981. Determination of nucleotide sequences in DNA. Science 214 (4526): 1205-1210.

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.

Shapiro, M. B., and Senapathy, P. 1987. RNA splice junctions of different classes of eukaryotes: sequence statistics and functional implications in gene expression. Nucleic Acid Res. 15: 7155-7174.

Shibata, T., Gotoh, M., Ochiai, A., and Hirohashi, S. 1994. Association of plakoglobin with APC, a tumor suppressor gene product, and its regulation by tyrosine phosphorylation. Biochem. Biophys. Res. Commun. 203: 519-522.

Shtutman, M., Zhurinsky, J., Simcha, I., Albanese, C., D'Amico, M., Pestell, R., and Ben-Ze'ev, A. 1999. The cyclin D1 gene is a target of the beta-catenin/LEF-1 pathway.

Proc. Natl. Acad. Sci. U. S. A. 96: 5522-5527.

Sternsdorf, T., Jensen, K., Zuchner, D., and Will, H. 1997. Cellular localization, expression, and structure of the nuclear dot protein 52. J. Cell Biol. 138: 435-448.

Su, L. K., Vogelstein, B., and Kinzler, K. W. 1993. Association of the APC tumor suppressor protein with catenins. Science 262: 1734-1737.

Suzuki, S. T. 1996. Structural and functional diversity of cadherin superfamily : Are new members of cadherin superfamily involved in signal transduction pathway? J. Cell.

Biochem. 61: 531-542.

Takeichi, M. 1991. Cadherin cell adhesion receptors as a morphogenetic regulator.

Science 251: 1451-1455.

Tetsu, O., and McCormick, F. 1999. Beta-catenin regulates expression of cyclin D1 in colon carcinoma cells. Nature 398: 422-426.

Troyanovsky, S. M., Troyanovsky, R. B., Eshkind, L. G., Krutovskikh, V. A., Leube, R. E., and Franke, W. W. 1994a. Identification of the plakoglobin-binding domain in desmoglein and its role in plaque assembly and intermediate filament anchorage. J.

Cell Biol. 127: 151-160.

Troyanovsky, S. M., Troyanovsky, R. B., Eshkind, L. G., Leube, R. E., and Franke, W. W.

1994b. Identification of amino acid sequence motifs in desmocollin, a desmosomal glycoprotein, that are required for plakoglobin binding and plaque formation. Proc. Natl.

Acad. Sci. U. S. A. 91: 10790-10794.

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.

Venugopal, R., and Jaiswal, A. K. 1998. Nrf2 and Nrf1 in association with Jun proteins regulate antioxydant response element-mediated expression and coordinated induction of genes encoding detoxifying enzymes. Oncogene 17: 3145-3156.

Vleminckx, K., Vakaet Jr, L., Mareel, M., Fiers, W., and van Roy, F. 1991. Genetic manipulation of E-cadherin expression by epithelial tumor cells reveals an invasion suppressor role. Cell 66 : 107-119.

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. a-Catenin-vinculin interaction functions to organize the apical junctional complex in epithelial cells. J. Cell Biol. 142: 847-857.

Whittock, N. V., Haftek, M., Angoulvant, N., Wolf, F., Perrot, H., Eady, R. A. J., and McGrath, J. A. 2000. Genomic amplification of the human plakophilin 1 gene and detection of a new mutation in ectodermal dysplasia/skin fragility syndrome. J. Invest.

Dermatology 115: 368-374.