Roles of plakoglobin end domains in desmosome assembly

2359Journal of Cell Science 110, 2359-2371 (1997)Printed in Great Britain © The Company of Biologists Limited 1997JCS4445

Roles of plakoglobin end domains in desmosome assembly

Helena L. Palka1,2,3 and Kathleen J. Green1,2,3,*1Department of Pathology, 2Robert H. Lurie Cancer Center and 3Department of Dermatology, Northwestern University MedicalSchool, 303 E. Chicago Avenue, Chicago, IL 60611, USA

*Author for correspondence (e-mail: [email protected])

Plakoglobin, a member of the armadillo family of proteins,is a component of intercellular adhesive junctions. Thecentral domain of plakoglobin comprises a highlyconserved series of armadillo repeats that facilitate its asso-ciation with either desmosomal or classic cadherins, or withcytosolic proteins such as the tumor suppressor geneproduct adenomatous polyposis coli. Sequences in the N-and C-terminal domains of plakoglobin are less highlyconserved, and their possible roles in regulating plakoglo-bin’s subcellular distribution and junction assembly arestill unclear. Here we have examined the role of plakoglo-bin end domains by stably expressing constructs lackingthe N and/or C terminus of plakoglobin in A-431 cells. Ourresults demonstrate that myc-tagged plakoglobin lackingeither end domain is still able to associate with the desmo-somal cadherin desmoglein and incorporate into desmo-somes. In cell lines that express an N-terminal truncationof plakoglobin, an increase in the cytosolic pool of en-

dogenous and ectopic plakoglobin was observed that mayreflect an increase in the stability of the protein. Deletionof the N terminus did not have a dramatic effect on thestructure of desmosomes in these cells. On the other hand,striking alterations in desmosome morphology wereobserved in cells expressing C-terminal truncations ofplakoglobin. In these cell lines, ectopic plakoglobin incor-porated into desmosomes, and extremely long junctions orgroups of tandemly linked desmosomes which remainedwell attached to keratin intermediate filaments, wereobserved. Together, these results suggest that plakoglobinend domains play a role in regulating its subcellular dis-tribution, and that the presence of the C terminus limits thesize of desmosomes, perhaps through regulating protein-protein interactions required for assembly of the desmoso-mal plaque.

Key words: Plakoglobin, Armadillo family, Desmosome

SUMMARY

INTRODUCTION

The junction associated protein plakoglobin (Pg) is a memberof the growing armadillo family of proteins which includes theadherens junction protein β-catenin and its Drosophilahomologue, the segment polarity gene product Armadillo(Peifer et al., 1994a). β-catenin was first recognized as a celljunction protein by virtue of its localization to adherensjunctions, where it serves as a bridge linking classicalcadherins to α-catenin (Aberle et al., 1996b). α-catenin in turninteracts directly or indirectly with the actin cytoskeleton(Knudsen et al., 1995; Rimm et al., 1995). However, in recentyears it has been realized that, like Armadillo, which was orig-inally defined as a downstream effector of the Drosophilawingless signaling pathway, β-catenin also plays a moreglobal role in intracellular signaling pathways and may evenact as a transcriptional transactivator (for reviews seeGumbiner, 1995; Huber et al., 1996; Miller and Moon, 1996;Peifer, 1995).

Plakoglobin, a close relative of β-catenin, has been reportedas a constituent both of desmosomes, adhesive junctions thatprovide cell surface attachment sites for intermediate filaments(IF), and adherens junctions (Cowin, 1994; Cowin and Burke,1996; Cowin et al., 1986). Although β-catenin’s contribution

to junction structure and adhesive function as well as signaltransduction has been firmly established, the role of Pg in inter-cellular junctions and signaling is not as well understood. Withregard to signaling, like β-catenin, Pg can promote axis dupli-cation in Xenopus (Karnovsky and Klymkowsky, 1995). Inaddition, a structural role for this molecule in recruiting IF tothe desmosomal plaque has been suggested (Troyanovsky etal., 1994b). As Pg preferentially incorporates into desmosomesin cells that assemble both adherens junctions and desmosomes(Adams et al., 1996; Nathke et al., 1994) this molecule mayplay a particularly important role in cytoskeletal attachmentand adhesive function in these junctions. Finally, Pg knock outmice have recently been shown to exhibit an embryonic lethalphenotype due to devastating structural defects of intercalateddiscs in the developing heart muscle (Bierkamp et al., 1996;Ruiz et al., 1996).

The functions of armadillo family members are dependenton their interactions with a diverse group of partner proteinspresent in different subcellular compartments. So far theseprotein interactions appear to be mediated in large part by over-lapping regions of the central armadillo repeats for which thefamily is named. Plakoglobin itself interacts with the cyto-plasmic tails of the desmosomal cadherins, the desmogleins(Dsg) and the desmocollins (Dsc) (Korman et al., 1989;

2360 H. L. Palka and K. J. Green

Kowalczyk et al., 1994; Mathur et al., 1994; Peifer et al., 1992;Roh and Stanley, 1995; Troyanovsky et al., 1994a,b), as wellas classical cadherins (Butz and Kemler, 1994; Jou et al., 1995;Knudsen and Wheelock, 1992; Peifer et al., 1992; Piepenhagenand Nelson, 1993). Plakoglobin also interacts with cytoplas-mic proteins such as α-catenin (Aberle et al., 1994; Hulsken etal., 1994; Rubinfeld et al., 1995) and the tumor suppressor geneproduct adenomatous polyposis coli (APC) (Shibata et al.,1994; Hulsken et al., 1994; Rubinfeld et al., 1995). The generalconsensus emerging from recent deletion analyses is that themore centrally located armadillo repeats are required for asso-ciation with classical cadherins, whereas the repeats flankingthe central region are essential for desmosomal cadherinbinding (Chitaev et al., 1996; Ozawa et al., 1995; Sacco et al.,1995; Troyanovsky et al., 1996; Wahl et al., 1996; Witcher etal., 1996). In particular, Dsg binding to Pg is dependent on theintegrity of sequences in repeats 1-3, a site that partiallyoverlaps with the region required for α-catenin binding (Aberleet al., 1996a; Ozawa et al., 1995; Sacco et al., 1995; Troy-anovsky et al., 1996; Wahl et al., 1996; Witcher et al., 1996).This overlap in binding domains of desmosomal cadherinsand α-catenin may be important in regulating the cytoskeletallinkages associated with desmosomal versus classicalcadherins.

Although Pg, β-catenin and Armadillo share a high degreeof sequence similarity in their central domains, the N- and Ctermini of the family members are more divergent (Peifer etal., 1992; Peifer and Wieschaus, 1990). This sequence diver-gence suggests that the end domains may confer specific regu-latory functions on the armadillo family members. We haveaddressed the functions of Pg end domains by generating stableA-431 cell lines expressing full length myc tagged Pg or trun-cations lacking the N terminus (Pg∆N), C terminus (Pg∆C) orboth N- and C termini (Pg∆N/∆C) (Fig. 1). Our results suggestthat Pg end domains contain regulatory sequences that governthe balance of Pg in various subcellular pools and also affectthe assembly of intercellular junctions. In particular, the Pg Cterminus appears to limit the length of desmosomes, perhapsby regulating the extent of protein-protein interactions that giverise to the desmosomal plaque.

Fig. 1. Schematic representation of Pg constructs. Pg constructsencoding full length Pg (Pg), an N-terminal truncated Pg (Pg∆N), aC-terminal truncated Pg (Pg∆C) and an N and C-terminal truncatedPg (Pg∆N/∆C) are shown. The shaded boxes represent the armadillorepeats found in Pg and its family members (Peifer et al., 1994a).The numbers designate amino acids expressed for each construct.Constructs are myc epitope tagged.

MATERIALS AND METHODS

Generation of plakoglobin cDNA constructsFull length human Pg cDNA was isolated and subcloned into themammalian expression vector LK 444 under the control of the humanβ-actin promoter (Kowalczyk et al., 1994). The nucleotide numberslisted for all Pg constructs listed below correspond to the human Pgsequence submitted to GenBank (accession number Z68228; Franke etal., 1989).

An N-terminal deletion of Pg was generated using the previouslydescribed Pg construct p236 (Kowalczyk et al., 1994) as a PCRtemplate and primers LN123 (5′ACGCGTCGACCCACGATGCT-CAAGTCGGCCATTG) and LN52 (5′CTGCAGACCACATGCTG-GCCGTGGAGCAAAAGCTCATTTCTGAAGAGGACTTGTAGA-AGCTTGGCC). The resulting PCR product, Pg∆N (nucleotides486-2,354), contains a SalI restriction site and Kozak consensussequence at the 5′ end, and an 11 amino acid c-myc epitope tag anda HindIII restriction site at the 3′ end. Using the engineered restric-tion sites, tagged Pg∆N was subcloned into the SalI/HindIII site inpBluescript. To ensure that the Pg sequence was correct, the resultingplasmid, p464, was digested with StyI (nucleotide 598) and BglII(nucleotide 1,855) and replaced with a StyI/BglII fragment from p236.In addition, the remaining PCR generated ends of p464 weresequenced. Using the SalI/HindIII sites in p464, Pg∆N was subclonedinto the SalI/HindIII sites of LK 444 generating plasmid p465. p465was used as an expression vector with the Pg∆N cDNA expressiondriven by the β-actin promoter.

A C-terminal deletion of Pg, which also removes a small portionof the 13th armadillo repeat, was generated using p236 as a PCRtemplate and primers LN51 (5′ACGTGTCGACCCACGATGGAG-GTGATGAACCTGA) and LN120 (5′TGTTCCGCATCTCCGAG-GTGGAGCAAAAGCTCATTTCTGAAGAGGACTTGTAGAAGC-TTGCGCA). The resulting PCR product, Pg∆C (nucleotides120-2,081), contains a SalI restriction site and Kozak consensussequence at the 5′ end, and an 11 amino acid c-myc epitope tag anda HindIII restriction site at the 3′ end. Using the engineered restric-tion sites, tagged Pg∆C was subcloned into the SalI/HindIII site inpBluescript to generate plasmid p470. An internal replacement of theStuI (nucleotide 261)/BglII fragment of p470 with a StuI/BglIIfragment from p236 was performed and the remaining PCR generatedends of p470 were sequenced to ensure that no PCR generated errorswere present in the final construct. Using the SalI/HindIII sites inp470, Pg∆C was subcloned into the SalI/HindIII sites of LK 444 gen-erating plasmid p511. p511 was used as an expression vector with thePg∆C cDNA expression driven by the β-actin promoter.

An N- and C-terminal deletion of Pg was generated using p236 asa PCR template and primers LN123 (5′ACGCGTCGACCCACGAT-GCTCAAGTCGGCCATTG) and LN120 (5′TGTTCCGCATCTCC-GAGGTGGAGCAAAAGCTCATTTCTGAAGAGGACTTGTAGA-AGCTTGCGCA). The resulting PCR product, Pg∆N/∆C (nucleotides486-2,081), contains a SalI restriction site and Kozak consensussequence at the 5′ end, and an 11 amino acid c-myc epitope tag anda HindIII restriction site at the 3′ end. Using the engineered restric-tion sites, tagged Pg∆N/∆C was subcloned into the SalI/HindIII sitein pBluescript to generate plasmid p467 and the same replacementstrategy described for Pg∆N was performed. Using the SalI/HindIIIsites in p467, Pg∆N/∆C was subcloned into the SalI/HindIII sites ofLK 444 generating plasmid p468. p468 was used as an expressionvector with the Pg∆N/∆C cDNA expression driven by the β-actinpromoter.

Stable cell linesA-431 epithelial cells (a gift from Dr M. Wheelock, University ofToledo) were cultured in DMEM containing 10% fetal bovine serum,100 i.u./ml penicillin, and 100 µg/ml streptomycin. cDNA constructsencoding full length Pg (p330), Pg∆N (p465), Pg∆C (p511) andPg∆N/∆C (p468), with the neomycin resistance gene on the same

2361Functions of plakoglobin end domains

plasmid, were transfected into A-431 cultures in duplicate asdescribed previously (Bornslaeger et al., 1996). Individual clones ofeach cell line were isolated and maintained in medium containing 350µg/ml (active concentration) G418. Expression of Pg proteins in thecell lines was assayed by SDS-PAGE (6-12% acrylamide gradient gel)and immunoblot analysis.

Densitometric analysisUsing the Bio-Rad model GS-670 imaging densitometer and theMolecular Analyst image analysis software, 2-5 immunoblots ofwhole cell extracts, probed with Pg specific antibodies, were analyzedto determine the levels of endogenous and ectopic Pg in the Pg celllines. Once the total amount of Pg (endogenous + ectopic) in the celllines was determined, the percentage endogenous versus ectopic Pgwas calculated. As the Pg specific antibodies used in the immunoblotsdo not recognize Pg∆N/∆C an indirect comparison of Pg to Pg∆N/∆Cwas made by first determining the ratio of Pg∆C detected by the N-terminal Pg specific antibody compared to the Pg∆C detected by themyc antibody. This ratio was then used to generate an estimate forPg∆N/∆C, based on the amount of myc reactive Pg∆N/∆C.

AntibodiesA rabbit polyclonal antibody directed against the c-myc epitope, wasa gift from Dr J. Stanley (University of Pennsylvania). The followingantibodies were previously described: the mouse monoclonal antibodyagainst the c-myc epitope tag, 9E10.2 (Evans et al., 1985), the mousemonoclonal antibody 11E4, directed against the N terminus of plako-globin (Kowalczyk et al., 1994), a rabbit polyclonal antibody directedagainst the C terminus of plakoglobin (Hinck et al., 1994), the rabbitpolyclonal antibody NW6, directed against desmoplakin (Angst et al.,1990), the mouse monoclonal antibody 6D8, directed against Dsg2(Bornslaeger et al., 1996; Wahl et al., 1996), the mouse monoclonalantibody 1G5, directed against α-catenin (Johnson et al., 1993). 11E4,6D8 and 1G5 were provided by Dr M. Wheelock (University ofToledo) and the Pg C-terminal antibody was provided by Dr J. Papkoff(Megabios Corp., Burlingame, CA). The mouse monoclonal anti-cytokeratin peptide 18 antibody (KSB17.2) was purchased fromSigma (St Louis, Missouri). For immunoblotting the polyclonalantibody directed against the c-myc tag was used at 1:500, concen-trated 11E4 supernatant at 1:2,000, the polyclonal antibody againstthe Pg C terminus at 1:4,000, and 6D8 ascites at 1:1,000. Primary anti-bodies were diluted in PBS, 5% milk. Secondary antibodies, goat anti-mouse peroxidase and goat anti-rabbit peroxidase (Kirkegaard &Perry Laboratories, Gaithsburg, MD), were used at a dilution of1:5,000 in PBS, 5% milk. Antibodies were detected using EnhancedChemiluminescence (Amersham).

Sequential detergent extractionCells were grown to confluence on 60 mm culture dishes, rinsed incomplete PBS and scraped into 150 µl cold saponin buffer (0.01%saponin, 10 mM Tris-HCl, pH 7.5, 140 mM NaCl, 5 mM EDTA, 2mM EGTA, 1 mM PMSF). Proteins were extracted for 20 minutes onice and samples centrifuged at approximately 14,000 g for 30 minutesat 4°C. After centrifugation, the saponin soluble pool was transferredto a fresh tube and 50 µl reducing SDS-PAGE sample buffer wasadded. The remaining pellet was extracted in 150 µl cold Triton X-100 buffer (1% Triton X-100, 10 mM Tris-HCl, pH 7.5, 140 mMNaCl, 5 mM EDTA, 2 mM EGTA, 1 mM PMSF). Samples werevortexed for 1 minute and centrifuged at approximately 14,000 g for30 minutes at 4°C. The Triton X-100 soluble pool was transferred toa fresh tube and 50 µl reducing SDS-PAGE sample buffer was added.The Triton X-100 insoluble proteins were solubilized in 200 µlSDS/urea buffer (1% SDS, 8 M urea, 10 mM Tris-HCl, pH 7.5, 5 mMEDTA, 2 mM EGTA, 1 mM PMSF). The saponin and Triton X-100soluble samples were heated at 95°C for 10 minutes and all thesamples were centrifuged at approximately 14,000 g for 5 minutesand subjected to SDS-PAGE on a 6-12% gradient gel. Proteins were

transferred to nitrocellulose and immunoblot analysis was performed.Densitometric analysis using the Bio-Rad model GS-670 imagingdensitometer and the Molecular Analyst image analysis software wasperformed for three independent experiments to determine the distri-bution of endogenous and ectopic Pg among the three pools.

ImmunoprecipitationFor immunoprecipitations from the saponin pool, cells were grown on60 mm culture dishes, rinsed in complete PBS and scraped into 1 mlcold saponin buffer (see above). Proteins were extracted for 20minutes on ice and samples centrifuged at approximately 14,000 g for30 minutes at 4°C. For immunoprecipitations from the Triton X-100pool, cells were grown on 100 mm culture dishes, rinsed in completePBS and scraped into 1 ml cold Triton X-100 buffer (see above).Samples were vortexed for 1 minute and centrifuged at approximately14,000 g for 30 minutes at 4°C. Samples were precleared with 50 µlGamma Bind Plus Sepharose beads (Pharmacia, Uppsala, Sweden)for at least 1 hour at 4°C and centrifuged for 5 minutes to removenon-specific complexes. Protein complexes from the saponin poolwere immunoprecipitated using the polyclonal myc antibody andTriton X-100 soluble complexes were immunoprecipitated using theantibody 6D8, directed against Dsg2. Antibodies were added for onehour after which 50 µl Gamma Bind Plus Sepharose beads were addedfor an additional hour. Beads were washed 4 times each with 1 ml 1%Triton X-100 wash buffer, resuspended in 70 µl reducing SDS-PAGEsample buffer and heated at 95°C for 10 minutes. Samples weresubjected to SDS-PAGE on a 7.5% gel and immunoblot analysis.

ImmunofluorescenceCells were grown on coverslips in 35 mm culture dishes, rinsed inPBS and fixed in cold methanol for 2 minutes. To detect ectopic Pg,concentrated supernatant of the monoclonal antibody against the c-myc tag, 9E10, was used at a 1:10 dilution or, the polyclonal antibodyagainst the c-myc tag was used at a 1:900 dilution. 9E10 staining wasdetected using a rhodamine-conjugated goat anti-mouse IgG and thepolyclonal myc tag antibody was detected using a rhodamine-conju-gated goat anti-rabbit IgG. Endogenous Pg was detected using con-centrated supernatant of the Pg specific monoclonal antibody 11E4diluted 1:30 in PBS. 11E4 staining was detected using a fluorescein-conjugated goat anti-mouse IgG. Desmoplakin protein was detectedusing the polyclonal antibody NW6 diluted 1:50 in PBS. NW6staining was detected using a fluorescein-conjugated goat anti-rabbitIgG. Keratin filaments were detected using the monoclonal antibodyKSB17.2 diluted 1:200 in PBS. The keratin antibody was detectedusing a rhodamine-conjugated goat anti-mouse IgG. α-catenin wasdetected using the monoclonal antibody 1G5 diluted 1:1 in PBS. 1G5staining was detected using a rhodamine-conjugated goat anti-mouseIgG. All secondary antibodies were purchased from Kirkegaard &Perry Laboratories, Gaithsburg, MD and diluted 1:50 in PBS.

Electron microscopyCells were grown to confluence on 60 mm culture dishes andprocessed for conventional electron microscopy as described previ-ously (Green et al., 1991). Briefly, cells were fixed in 1% glutaralde-hyde and post-fixed in 1% osmium tetroxide. Samples were stainedwith aqueous 1% uranyl acetate, dehydrated in a graded ethanol seriesand embedded in epon-Araldite. Ultra-thin sections were viewedusing a JEOL 100CX.

RESULTS

Generation of stable cell lines expressing myc-tagged full length Pg or Pg lacking one or both enddomainsTo address the function of Pg end domains in the subcellular


Fig. 2. Immunoblot analysis of endogenous and ectopic Pg expression in stable A-431cell lines. One or two representative clones of stable A-431 cell lines expressing the various Pg truncations were analyzed by immunoblot using Pg specific antibodies directed against the Pg Nterminus (Pg NT), the Pg C terminus (Pg CT), or a polyclonal antibody directed against the myc tag (myc pAb). (A) Immunoblots of A-431 celllines using Pg specific antibodies. Endogenous Pg is detected in all of the cell lines using either the Pg N or C-terminal antibody. Pg.mycmigrates slower than endogenous Pg and is detected using the Pg NT antibody. Pg∆N is detected using the Pg CT antibody and migrates at thepredicted molecular mass (~70 kDa). Pg∆C also migrates at the predicted molecular mass (~72 kDa) and is detected using the Pg NT antibody.Pg∆N/∆C is not recognized by either Pg specific antibody and only endogenous Pg is detected using the Pg N-terminal antibody.(B) Immunoblots of A-431 cell lines using the myc tag antibody. The ectopically expressed Pg proteins were detected using a polyclonalantibody directed against the myc tag (myc pAb). Using the myc antibody Pg∆N/∆C was detected at the predicted molecular mass (~55 kDa).(C) Comparison of endogenous and ectopic Pg levels in the A-431 cell lines. The levels of endogenous and ectopic Pg were compared in eachof the cell lines using the Bio-Rad model GS-670 imaging densitometer and the Molecular Analyst image analysis software as described inMaterials and Methods. The total Pg level for each clone was determined and the percentage endogenous and ectopic Pg was calculated withstandard deviations (Pg.myc (A5) 86±1.5%, 14±1.5%; Pg.myc (B7) 76±7.8%, 24±7.8%; Pg∆N (A7) 43±3%, 57±3%; Pg∆N (B1) 53±2.3%,47±2.3%; Pg∆C (10) 49±2.1%, 51±2.1%; Pg∆C (B17) 78±11.3%, 22±11.3%; Pg∆N/∆C (A1) 46±8.5%, 54±8.5%).

distribution of Pg and in intercellular junction assembly, A-431 cell lines expressing myc epitope tagged Pg∆N, Pg∆C andPg∆N/∆C (Fig. 1) were generated. To determine if the myctag affected the subcellular distribution of Pg or its incorpo-ration into intercellular junctions, A-431 cell lines expressinga full length myc-tagged Pg construct (Pg.myc) were alsogenerated. Each of the ectopically expressed proteins,detectable by Pg antibodies or the myc antibody, migrated atthe predicted molecular mass (Fig. 2A,B). The level of ecto-pically expressed Pg.myc, Pg∆N and Pg∆C was directlycompared to endogenous Pg levels by using Pg specific anti-bodies directed against the Pg N or C terminus (Fig. 2A).Because Pg∆N/∆C is not detected by the Pg specific anti-bodies the level of ectopically expressed Pg∆N/∆C was in-directly compared to endogenous Pg levels as described inMaterials and Methods. Although the level of ectopic andendogenous Pg varied somewhat among the cell lines andamong the clones of each cell line (Fig. 2C), the specific phe-notypes observed for the different cell lines is unlikely to besolely dependent on expression level, since clones such asPg∆N (A7), Pg∆C (10) and Pg∆N/∆C (A1), with comparablepercents of ectopic Pg (57%, 51%, 54% of total Pg, respec-

tively), exhibited dramatically different junctional phenotypes(see Fig. 6).

The subcellular distribution of full length myc-tagged Pg is similar to endogenous Pg in controlcellsTo determine any possible effects of the C-terminal myc tag onPg, the subcellular distribution of full length myc-tagged Pgand endogenous Pg were compared both biochemically andmorphologically in stable transfectants and control cells. Incontrol A-431 cells endogenous Pg was distributed between acytosolic saponin-soluble pool, a membrane bound Triton-soluble pool, and a Triton-insoluble pool, thought to be largelyjunction or cytoskeleton-associated (Fig. 3). Similar to en-dogenous Pg from control cells, a small proportion of fulllength myc-tagged Pg was observed in the cytosolic pool andthe majority of the protein was in the Triton-soluble and Triton-insoluble pools (Fig. 3). Although there appeared to be amodest decrease in the amount of endogenous and ectopic Pgin the Triton-insoluble pool compared to control cells (Fig. 3),endogenous Pg and Pg.myc were nevertheless able to incor-porate into junctions as assessed by co-localization with


endogenous cell junction components such as DP (see Fig.6A,B), Dsg and Dsc (data not shown).

Deletion of the Pg N terminus results in a largercytosolic pool of ectopic and endogenous Pg, butneither end domain is required for the incorporationof Pg into a final Triton X-100 insoluble poolTo determine the role of Pg end domains in the regulation ofthe subcellular distribution of the Pg protein, the distributionof both endogenous and ectopic Pg in cell lines expressing thevarious Pg truncations was analyzed. Unlike endogenous Pg incontrol cells or Pg.myc in which a small proportion of the totalprotein was in the cytosolic pool, a greater proportion of Pg∆Nand Pg∆N/∆C was observed in the cytosolic pool. Densito-metric analysis of three independent extractions revealed only14% of the total Pg in control cells, and 19% of Pg.myc, in thecytosolic pool. In contrast, 42% of Pg∆N and 46% of Pg∆N/∆Cwas observed in the cytosolic pool. Interestingly, the differ-ences in the subcellular distribution of Pg∆N and Pg∆N/∆Cwere accompanied by similar alterations in the distribution ofendogenous Pg in each of these A-431 cell lines. EndogenousPg in the cytosolic pool accounted for 33% of total en-dogenous Pg in the mutant cell lines compared to 14% incontrol cells (Fig. 3). In addition to the differences seen in thecytosolic pool, there was a smaller proportion of endogenousPg in the Triton-insoluble pool in Pg∆N and Pg∆N/∆C celllines. Nevertheless, deletion of the N terminus did not preventectopic or endogenous Pg in these cell lines from incorporat-ing into intercellular junctions as assessed by immunofluores-cence analysis (see Fig. 6).

Although the deletion of the Pg C terminus also resulted inan increase in the cytosolic pool of Pg∆C compared to en-dogenous Pg in control cells, it was not as dramatic an increase

as was seen for Pg∆N or Pg∆N/∆C (Fig. 3). In addition,although there was an apparent decrease of endogenous Pg inthe Triton-insoluble pool in Pg∆C cell lines (Fig. 3), thisdecrease was not detectably different from that seen in fulllength Pg.myc cell lines. As in the case of Pg∆N, deletion ofthe C terminus did not prevent Pg∆C or endogenous Pg in thesecell lines from incorporating into junctions (see Fig. 6).

To determine if the increase in the cytosolic pool of en-dogenous Pg seen in cell lines expressing N-terminal trunca-tions of Pg was due to the formation of complexes betweenendogenous and ectopic Pg, cytosolic protein complexes wereanalyzed. A polyclonal antibody directed against the mycepitope tag co-immunoprecipitated endogenous Pg in cell linesexpressing Pg∆N, Pg∆C and Pg∆N/∆C (Fig. 4). In cell linesexpressing Pg.myc, endogenous Pg was barely detectable inthe cytosolic protein complex precipitated with the mycantibody (Fig. 4).

Pg lacking the N or C terminus is found in a complexwith desmogleinThe Triton X-100 soluble pool of proteins from the A-431 celllines was analyzed to determine if the N and/or C termini affectthe ability of Pg to interact with desmoglein 2 (Dsg2). Anantibody directed against the Dsg2 extracellular domain co-immunoprecipitated both Pg∆N and Pg∆C (Fig. 5). Further-more, the truncated Pg proteins did not prevent complexes fromforming between endogenous Pg and Dsg since endogenous Pgin all of the cell lines was able to co-immunoprecipitate withDsg. Pg∆N and Pg∆C also co-immunoprecipitated with desmo-collin and classical cadherins (data not shown). In addition, datafrom our laboratory has shown that both Pg∆N and Pg∆C canco-immunoprecipitate with another desmoglein isoform, Dsg1,when they are co-expressed in mouse L-cell fibroblasts

Fig. 3. Subcellular distribution ofendogenous and ectopic Pg. A-431 celllines expressing the various Pgconstructs were extracted in saponindetergent to release cytosolic proteinsfollowed by a Triton X-100 extractionto release membrane bound proteins.Triton X-100 insoluble, presumablyjunction- or cytoskeleton-associatedproteins were solubilized in SDS/ureasample buffer. Endogenous and ectopicPg were detected using the appropriatePg antibody. Pg∆N/∆C was detectedusing the polyclonal antibody againstthe myc tag. The ectopic Pg proteinsare denoted by an asterisk in the clonesPg.myc (B7), Pg∆N (A7), Pg∆C (10)and Pg∆N/∆C (A1).


Fig. 4. Co-immunoprecipitation of endogenous Pg with ectopic Pg.Using the myc polyclonal antibody, protein complexes wereimmunoprecipitated from the saponin soluble pool of proteins fromthe various A-431 cell lines. Immunoprecipitates from each cell linewere run, in duplicate, on SDS-PAGE gels, transferred tonitrocellulose and immunoblotted with the Pg N-terminal antibody orthe Pg C-terminal antibody. Endogenous Pg was barely detectable inimmunoprecipitates from cell lines expressing Pg.myc, however,endogenous Pg co-immunoprecipited with the myc antibody in celllines expressing Pg∆N, Pg∆C and Pg∆N/∆C. The A-431 clonesshown are Pg.myc (B7), Pg∆N (A7), Pg∆C (10) and Pg∆N/∆C (A1).

Fig. 5. Co-immunoprecipitation of ectopic Pg with Dsg2. Using theDsg 2 specific antibody 6D8, protein complexes wereimmunoprecipitated from the Triton X-100 soluble pool of proteinsfrom the various A-431 cell lines. Immunoprecipitates from each cellline were run, in triplicate, on SDS-PAGE gels, transferred tonitrocellulose and immunoblotted with the Dsg antibody 6D8, the PgN-terminal antibody or the Pg C-terminal antibody. Pg.myc (openarrow), which migrates slower than endogenous Pg, is detected byboth Pg antibodies. Pg∆C (open arrowhead) and Pg∆N (filledarrowhead) are detected by the Pg N or C-terminal antibodies,respectively. A non-specific band (open circle) is detected in all theimmunoprecipitates, including those from control cells, with the PgC-terminal antibody. The A-431 clones shown are Pg.myc (B7),Pg∆N (A7), Pg∆C (10) and Pg∆N/∆C (A1).
(Kowalczyk et al., 1997). Because the IgG heavy chain used in
the immunoprecipitations migrated at the same relativemolecular mass as Pg∆N/∆C (~55 kDa), we were unable toconfirm Pg∆N/∆C’s presence in a Dsg complex. However, asPg∆N/∆C incorporated into junctions (see Fig. 6) it is likely todo so by associating with members of the cadherin family. Theability of Pg∆N and Pg∆C to co-immunoprecipitate withdesmosomal cadherins is consistent with the results of othersshowing that Pg armadillo repeats, and not the end domains, arerequired for the interaction of Pg with both desmosomal andclassical cadherins (Chitaev et al., 1996; Ozawa et al., 1995;Sacco et al., 1995; Troyanovsky et al., 1996; Wahl et al., 1996;Witcher et al., 1996).

Cells expressing C-terminal deletions of Pg exhibit amore continuous, less punctate, distribution ofdesmosomal componentsIn order to assess the possible effects of removing the Pg enddomains on desmosome assembly, indirect double labelimmunofluorescence using the 9E10 antibody directed againstthe myc tag and an antibody against the desmosomal plaqueprotein desmoplakin (DP) was performed. In cells expressingPg.myc the ectopic protein exhibited a punctate stainingpattern at cell-cell borders that co-localized with endogenousDP (Fig. 6A,B). The staining pattern for DP in Pg.myc cellswas similar to the punctate staining pattern observed for DP incontrol A-431 cells (not shown). Similar to Pg.myc, Pg∆Nexhibited a punctate staining pattern at cell-cell borders thatco-localized with DP staining (Fig. 6C,D). In addition, a cyto-plasmic myc staining pattern could be detected that did not

always co-localize with DP, which may reflect in part thecytosolic pool of Pg∆N observed biochemically (see Fig. 3).In cell lines expressing Pg∆C, the ectopic protein was distrib-uted more continuously at some borders and DP in these cellshad a more continuous staining pattern (Fig. 6E,F) comparedto control or Pg.myc cells. The altered distribution of desmo-somal components was most dramatic in cell lines expressingPg∆N/∆C in which Pg∆N/∆C and DP frequently appearedmore continuous at cell-cell borders (Fig. 6G,H). In additionto DP, other desmosomal components, such as Dsg and Dsc,exhibited a more continuous staining pattern in cell linesexpressing C-terminal truncations of Pg (data not shown).

To determine if the phenotype in cell lines expressing C-terminal truncated Pg was in part due to the exclusion ofendogenous Pg from these long junctions, indirect double labelimmunofluorescence using an antibody against the mycepitope tag and a Pg specific antibody recognizing onlyendogenous Pg was performed. In cell lines expressingPg∆N/∆C, endogenous Pg exhibited a more continuousstaining pattern in areas that co-localized with ectopic Pgstaining (Fig. 7A,B). Endogenous Pg appeared more punctatein areas where ectopic Pg staining was barely detectable.

Expression of Pg truncations does not disruptintermediate filament attachment to the cell surfaceor induce mixing of desmosomal and adherensjunction componentsAlthough Pg has been implicated in playing a role in recruit-


ing intermediate filaments (IF) to the desmosomal plaque(Troyanovsky et al., 1994b), how Pg contributes to IF attach-ment at the molecular level is not understood. To determinewhether truncation of either the Pg N or C terminus compro-mised IF attachment to the cell surface, double label immuno-fluorescence using an antibody against DP and an antibodyagainst keratin IF was performed. In all the cell lines express-ing ectopic Pg proteins, IF bundles remained associated withthe cell surface and terminated at sites of DP staining (Fig. 8).

Fig. 6. Ectopic Pg co-localizeswith DP. Indirect double labelimmunofluorescence wasperformed to detect ectopic Pgpolypeptides and endogenousDP. The localization of ectopicPg in A-431 clones Pg.myc(B7) (A and B), Pg∆N (A7) (Cand D), Pg∆C (B7) (E and F)and Pg∆N/∆C (A1) (G and H)was determined by using themyc antibody 9E10 (A,C,E,G).DP in these cells was detectedusing the DP antibody NW6(B,D,F,H). Note the morecontinuous staining pattern ofDP in cell lines expressing Pgprotein lacking the C terminus(F,H). All panels are at the samemagnification. Bar, 10 µm.

In addition to playing a role in recruiting IF to the desmo-somal plaque, Pg may also be involved in the segregation ofjunctional components. Junctions in the hearts of Pg null micecontain components of both desmosomes and adherensjunctions suggesting that Pg may be involved in the sorting ofdesmosomal and adherens junction proteins (Ruiz et al., 1996).To determine if the N and/or C terminus of Pg is involved inthis segregation, double label immunofluorescence using anantibody against DP and an antibody against the adherens


Fig. 7. Ectopic Pg co-localizes with endogenous Pg. To compare thelocalization of endogenous Pg and Pg∆N/∆C, double labelimmunofluorescence was performed on the A-431 clone Pg∆N/∆C(A1). Ectopic Pg was detected using a polyclonal antibody against themyc tag (A) and endogenous Pg was detected using the Pg N-terminalantibody 11E4 (B). Note the co-localization of ectopic and endogenousPg in the extended junctions, but not in the more normal punctatestructures. Both panels are at the same magnification. Bar, 10 µm.

junction protein α-catenin was performed. DP and α-cateninexhibited distinct staining patterns in control cells and in all ofthe cell lines expressing ectopic Pg (Fig. 9).

Longer desmosomes and groups of tandemly linkeddesmosomes are present in cell lines expressing C-terminal deletions of PgUltrastructural analysis was performed to determine if the con-tinuous staining pattern observed by immunofluorescence inPg∆C and Pg∆N/∆C cell lines represented continuous desmo-somes. Desmosomes observed in cell lines expressing Pg∆N(Fig. 10B) were similar to desmosomes observed in controlcells (Fig. 10A) and cell lines expressing Pg.myc (data notshown). However, in cell lines expressing Pg∆C and Pg∆N/∆Cmany desmosomes appeared to occupy more of the plasmamembrane and groups of tandemly linked desmosomes wereobserved (Fig. 10C,D). Consistent with the immunofluor-escence analysis, IF attachment to the cell surface was notdisrupted in any of the cell lines expressing ectopic Pg proteins.

DISCUSSION

Plakoglobin’s ability to interact with junctional proteins such

as desmosomal cadherins, classical cadherins and α-catenin isdependent largely on the highly conserved armadillo repeatsfound in Pg and its family members (Chitaev et al., 1996;Ozawa et al., 1995; Sacco et al., 1995; Troyanovsky et al.,1996; Wahl et al., 1996; Witcher et al., 1996). Less is knownregarding the function of the more divergent end domains ofPg, and their possible roles in regulating Pg’s association withvarious proteins or assembly into intercellular junctions. Herewe demonstrate by stably expressing Pg with deletions of oneor both end domains, that these domains contain informationthat regulates Pg’s subcellular distribution as well as normalassembly of the desmosomal plaque.

Deletion of the Pg N terminus led to an accumulation of bothectopic and endogenous Pg in the cytosolic pool (Fig. 3). Pre-viously, our laboratory has shown that Pg is unstable andrapidly degraded when expressed in L-cell fibroblasts in theabsence of a desmosomal cadherin (Kowalczyk et al., 1993).It is possible that accumulation of Pg lacking the N terminusin the cytosol reflects an increase in the stability of Pg∆N andPg∆N/∆C in a non-cadherin associated pool. The stability ofthe Pg family member β-catenin is dependent upon a glycogensynthase kinase 3 (GSK3) consensus site in the N terminus ofβ-catenin, and mutations or deletions of the N-terminal GSK3site in β-catenin result in an increased stability of β-catenin(Barth et al., 1997; Munemitsu et al., 1996; Yost et al., 1996).A similar GSK3 consensus site is found in the N terminus ofPg (Peifer et al., 1994b) and deletion of this site results in theaccumulation of Pg in Xenopus embryos (Rubenstein et al.,1997). Intriguingly, the expression of N-terminal truncations ofPg in A-431 cells resulted in an increase in the level of endoge-nous Pg in the cytosolic pool (Fig. 3). Previous investigatorshave suggested that Pg may be able to self-associate (Kapprellet al., 1987). If N-terminally truncated, stable Pg and en-dogenous Pg formed a complex in the cytosolic pool, this couldpotentially explain the observed accumulation of endogenousPg. Although we did not observe obvious interactions betweenendogenous Pg and Pg.myc, our results suggest that deletionof either end domain, and particularly the N terminus, greatlyenhances complex formation between ectopic and endogenousPg (Fig. 4). Therefore the increase in stability of N-terminallytruncated constructs, along with their enhanced ability tointeract with full length Pg, may lead to endogenous Pg’s sta-bilization.

Recently, Barth et al. (1997) have shown that the expressionof N-terminal truncations of β-catenin affected the morphol-ogy and migration of MDCK cells. Although we have notobserved obvious effects on A-431 cell behavior, furtherstudies will be required to determine whether expression of Pgdeletions affects cell adhesion, migration or morphology.

As might be predicted by previous studies that mappeddesmosomal cadherin binding sites to the Pg armadillo repeats(Chitaev et al., 1996; Ozawa et al., 1995; Troyanovsky et al.,1996; Wahl et al., 1996; Witcher et al., 1996) and our own dataindicating that Pg∆N and Pg∆C still associate with Dsg2 (Fig.5), deletion of either Pg end domain did not prevent incorpo-ration of Pg into desmosomes (Fig. 6). Decreases in the Triton-insoluble pool were observed (Fig. 3), but this was the caseeven in cells expressing full length myc-tagged Pg, suggestingthat the observed decrease may reflect a modest inhibition ofincorporation due to the myc epitope tag that is on the Cterminus of each of the ectopic Pg proteins. In spite of the rel-


Fig. 8. Expression of ectopic Pg does not disrupt IF attachment. Dual color double label immunofluorescence was performed on control cells(A) and A-431 clones Pg.myc (B7) (B), Pg∆N (A7) (C), Pg∆C (10) (D) and Pg∆N/∆C (A1) (E, F). The DP antibody, NW6, was detected usinga fluorescein-conjugated secondary antibody while the keratin antibody, KSB17.2 (Sigma) was detected using a rhodamine-conjugatedsecondary antibody. In areas where DP and keratin filaments overlap the staining pattern appears yellow. All panels are at the samemagnification. Bar, 10 µm.

atively low level of protein that entered a Triton-insoluble pool,particularly in the case of Pg∆N/∆C, the observed effects ondesmosomal plaque assembly were dramatic. In cell linesexpressing Pg∆C or Pg∆N/∆C desmosomes appeared muchlonger and tandemly linked desmosomes were observed (Figs6 and 10). Although the underlying basis for this effect ondesmosomes in not yet clear, one possibility is that deletion ofthe Pg C terminus may enhance interactions with other desmo-somal components. Consistent with this idea is the reportedidentification of a stretch of ten amino acids in Pg armadillorepeat thirteen, the most C-terminal repeat, that interacts withand masks upstream cadherin binding sites within Pg itself(Troyanovsky et al., 1996). Although this stretch of amino

acids is deleted in the C-terminal truncated Pg constructs usedin the present study, we did not observe an increase in the inter-action between Pg∆C and the desmosomal cadherin Dsg2 (Fig.5). Deletion of the end domains did, however, enhance theinteraction between endogenous and ectopic Pg, possibly con-tributing to the enhanced desmosomal plaque assembly seen inPg∆C and Pg∆N/∆C cell lines.

Exposure of cryptic binding sites in the armadillo repeatsmight also enhance interactions with other plaque components.One candidate protein is the desmosomal component DP. DPis involved in anchoring IF to the cell surface via its C terminus(Bornslaeger et al., 1996; Kouklis et al., 1994; Stappenbeck etal., 1993; Stappenbeck and Green, 1992). Although the N


Fig. 9. Desmosomal and adherens junction components remain segregated in cell lines expressing Pg truncations. Dual color double labelimmunofluorescence was performed on control cells (A) and A-431 clones Pg.myc (B7) (B), Pg∆N (A7) (C), Pg∆C (10) (D) and Pg∆N/∆C(A1) (E, F). The DP antibody, NW6, was detected using a fluorescein-conjugated secondary antibody while the α-catenin antibody, 1G5, wasdetected using a rhodamine-conjugated secondary antibody. DP and α-catenin have distinct staining patterns that do not appear to overlapextensively in any of the cell lines. All panels are at the same magnification. Bar, 10 µm.

terminus of DP is required for targeting the molecule to theplaque (Bornslaeger et al., 1996; Stappenbeck et al., 1993), theproteins with which this DP domain interacts directly at thecell surface have not been identified. Recent work from ourlaboratory suggests that the DP N terminus associates directlywith Pg and clusters Pg-desmosomal cadherin complexes whenco-expressed in L cell fibroblasts (A. P. Kowalczyk et al.,1997). Furthermore, the Pg end domains were not required forthis association. These observations are consistent with thework presented here demonstrating that when the same Pgtruncations were expressed in A-431 cells, the ectopic Pg wasable to co-localize with DP (Fig. 6). Since expression of thetruncated proteins did not disrupt IF attachment to the cell

surface (Fig. 8) the Pg N and C termini are apparently notrequired for linking the DP-IF complex to the cell surface. Infact, if sequences in the Pg C terminus mask cryptic bindingsites in the armadillo repeats that constitute the DP bindingregion, deletion of these sequences may enhance interactionswith DP and could contribute to the enhanced plaque formationseen in this study.

Analysis of Pg null mice has revealed another possiblefunction for Pg in proper junction assembly. Junctions in thehearts of Pg null mice contain components of both desmo-somes and adherens junctions and embryos die due to heartdefects (Ruiz et al., 1996). The end domains of Pg do notappear to be involved in the segregation of junctional compo-


Fig. 10. Ultrastructural analysis of desmosomes in cell lines expressing Pg truncations. Electron micrographs of control cells (A) and A-431clones Pg∆N (A7) (B), Pg∆C (10) (C) and Pg∆N/∆C (A1) (D). Intermediate filament attachment is not disrupted in any of the Pg cell lines.Note the groups of tandemly linked desmosomes in cell lines expressing Pg lacking the C terminus (C,D). All panels are at the samemagnification. Bar, 0.25 µm.


nents as we did not see any mixed junctions in cell linesexpressing Pg truncations (Fig. 9). Although desmosomes wereable to form in epithelial organs of Pg null mice the desmo-somes were abnormal and mutant embryos that died aroundbirth had skin blistering and subcorneal acantholysis suggest-ing that Pg is required for proper desmosome function(Bierkamp et al., 1996).

The Pg N terminus contains regulatory sequences that areinvolved in the distribution of Pg in various subcellular poolswhile the Pg C terminus is involved in the proper assembly ofthe desmosomal plaque. The C terminus of Pg is also the targetof post-translational modifications that may further affect itsregulatory activities. We have preliminary data suggesting thatthe C terminus of Pg contains sequences required for EGF-induced Pg tyrosine phosphorylation (H. L. Palka and K. J.Green, unpublished data). After EGF stimulation, tyrosinephosphorylated Pg is predominantly in the Triton X-100soluble pool. Thus, phosphorylation of the C terminus of Pgmay act to prevent incorporation of soluble Pg complexes intojunctions, thereby inhibiting junction assembly. Further studieswill be required to determine whether such phosphorylationprevents association of Pg with desmosomal plaque compo-nents such as DP, thereby modulating desmosome assembly.

The authors thank Drs M. Wheelock, J. Papkoff and J. Stanley forantibodies provided. The authors also thank Jennifer Lamb andMargaret Ruesch for their assistance in generating Pg deletion con-structs and stable cell lines and Jie Pan for assistance with prepara-tion of sections for electron microscopy. Special thanks to E.Bornslaeger, M. Denning, A. Kowalczyk, S. Norvell and L. Bannonfor critical reading of the manuscript and insightful discussions. Thiswork was supported by NIH grants R01AR43380 and R01AR41836awarded to K. Green. H. Palka was supported in part by the NIHtraining grant 5T32CA09560-10. K. Green is an American CancerSociety Faculty Research Awardee.

REFERENCES

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

Aberle, H., Schwartz, H., Hoschuetzky, H. and Kemler, R. (1996a). Singleamino acid substitutions in proteins of the armadillo gene family abolishtheir binding to α-catenin. J. Biol. Chem. 271, 1520-1526.

Aberle, H., Schwartz, H. and Kemler, R. (1996b). Cadherin-catenin complex:protein interactions and their implications for cadherin function. J. Cell.Biochem. 61, 514-523.

Adams, C. L., Nelson, W. J. and Smith, S. J. (1996). Quantitative analysis ofcadherin-catenin-actin reorganization during development of cell-celladhesion. J. Cell Biol. 135, 1899-1911.

Angst, B. D., Nilles, L. A. and Green, K. J. (1990). Desmoplakin II expressionis not restricted to stratified epithelia. J. Cell Sci. 97, 247-257.

Barth, A. I. M., Pollack, A. L., Altschuler, Y., Mastov, K. E. and Nelson, W.J. (1997). NH2-terminal deletion of β-catenin results in stable colocalizationof mutant β-catenin with adenomatous polyposis coli protein and alteredMDCK cell adhesion. J. Cell Biol. 136, 693-706.

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.

Bornslaeger, E. B., Corcoran, C. M., Stappenbeck, T. S. and Green, K. J.(1996). Breaking the connection: Displacement of the desmosomal plaqueprotein desmoplakin from cell-cell interfaces disrupts anchorage ofintermediate filament bundles and alters intercellular junction assembly. J.Cell Biol. 134, 985-1001.

Butz, S. and Kemler, R. (1994). Distinct cadherin-catenin complexes in Ca2+-dependent cell-cell adhesion. FEBS Lett. 355, 195-200.

Chitaev, N. A., Leube, R. E., Troyanovsky, R. B., Eshkind, L. G., Franke,W. W. and Troyanovsky, S. M. (1996). The binding of plakoglobin todesmosomal cadherins: patterns of binding sites and topogenic potential. J.Cell Biol. 133, 359-369.

Cowin, P., Kapprell, H.-P., Franke, W. W., Tamkun, J. and Hynes, R. O.(1986). Plakoglobin: a protein common to different kinds of intercellularadhering junctions. Cell 46, 1063-1073.

Cowin, P. (1994). Plakoglobin. In Molecular Biology of Desmosomes andHemidesmosomes (ed. J. E. Collins and D. R. Garrod), pp. 1-131. R. G.Landes Co., Austin.

Cowin, P. and Burke, B. (1996). Cytoskeleton-membrane interactions. Curr.Opin. Cell Biol. 8, 56-65.

Evans, G. I., Lewis, G. K., Ramsey, G. and Bishop, J. M. (1985). Isolation ofmonoclonal antibodies specific for human c-myc proto-oncogene product.Mol. Cell Biol. 5, 3610-3616.

Franke, W. W., Goldschmidt, M. D., Zimbelmann, R., Mueller, H. M.,Schiller, D. L. and Cowin, P. (1989). Molecular cloning and amino acidsequence of human plakoglobin, the common junctional plaque protein.Proc. Nat. Acad. Sci. USA 86, 4027-4031.

Green, K. J., Stappenbeck, T. S., Noguchi, S., Oyasu, R. and Nilles, L. A.(1991). Desmoplakin expression and distribution in cultured rat bladderepithelial cells of varying tumorigenic potential. Exp. Cell Res. 193, 134-143.

Gumbiner, B. (1995). Signal transduction by β-catenin. Curr. Opin. Cell Biol.7, 634-640.

Hinck, L., Nelson, W. J. and Papkoff, J. (1994). Wnt-1 modulates cell-celladhesion in mammalian cells by stabilizing β-catenin binding to the celladhesion protein cadherin. J. Cell Biol. 124, 729-740.

Huber, O., Bierkamp, C. and Kemler, R. (1996). Cadherins and catenins indevelopment. Curr. Opin. Cell Biol. 8, 685-691.

Hulsken, J., Birchmeier, W. and Behrens, J. (1994). E-cadherin and APCcompete for the interaction with β-catenin and the cytoskeleton. J. Cell Biol.127, 2061-2069.

Johnson, K. R., Lewis, J. E., Li, D., Wahl, J., Soler, A. P., Knudsen, K. A.and Wheelock, M. J. (1993). P- and E-cadherin are in separate complexes incells expressing both cadherins. Exp. Cell Res. 207, 252-260.

Jou, T. S., Stewart, D. B., Stappert, J., Nelson, W. J. and Marrs, J. A. (1995).Genetic and biochemical dissection of protein linkages in the cadherin-catenin complex. Proc. Nat. Acad. Sci. USA 92, 5067-5071.

Kapprell, H.-P., Cowin, P. and Franke, W. W. (1987). Biochemicalcharacterization of the soluble form of the junctional plaque protein,plakoglobin, from different cell types. Eur. J. Biochem. 166, 505-517.

Karnovsky, A. and Klymkowsky, M. W. (1995). Anterior axis duplication inXenopus induced by the over-expression of the cadherin-binding proteinplakoglobin. Proc. Nat. Acad. Sci. USA 92, 4522-4526.

Knudsen, K. A. and Wheelock, M. J. (1992). Plakoglobin, or an 83-kDhomologue distinct from β-catenin interacts with E-cadherin and N-cadherin. J. Cell Biol. 118, 671-679.

Knudsen, K. A., Soler, A. P., Johnson, K. R. and Wheelock, M. J. (1995).Interaction of α-actinin with the cadherin/catenin cell-cell adhesion complexvia α-catenin. J. Cell Biol. 130, 67-77.

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

Kouklis, P. D., Hutton, E. and Fuchs, E. (1994). Making a connection: directbinding between keratin intermediate filaments and desmosomal proteins. J.Cell Biol. 127, 1049-1060.

Kowalczyk, A. P., Palka, H., Nilles, L. A., Anderson, J. E. and Green, K. J.(1993). Expression of plakoglobin protein in L-cells is enhanced by thedesmosomal cadherin desmoglein. Mol. Biol. Cell 4, 438a.

Kowalczyk, A. P., Palka, H. L., Luu, H. H., Nilles, L. A., Anderson, J. E.,Wheelock, M. J. and Green, K. J. (1994). Posttranslational regulation ofplakoglobin expression: Influence of the desmosomal cadherins onplakoglobin metabolic stability. J. Biol. Chem. 269, 31214-31223.

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 clustersdesmosomal cadherin-plakoglobin complexes. J. Cell Biol. (in press).

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

Miller, J. R. and Moon, R. T. (1996). Signal transduction through β-cateninand specification of cell fate during embryogenesis. Genes Dev. 10, 2527-2539.


Munemitsu, S., Albert, I., Rubinfeld, B. and Polakis, P. (1996). Deletion ofan amino-terminal sequence stabilizes β-catenin in vivo and promoteshyperphosphorylation of the adenomatous polyposis coli tumor suppressorprotein. Mol. Cell. Biol. 16, 4088-4094.

Nathke, I. S., Hinck, L., Swedlow, J. R., Papkoff, J. and Nelson, W. J.(1994). Defining interactions and distributions of cadherin and catenincomplexes in polarized epithelial cells. J. Cell Biol. 125, 1341-1352.

Ozawa, M., Terada, H. and Pedraza, C. (1995). The fourth armadillo repeatof plakoglobin (γ-catenin) is required for its high affinity binding to thecytoplasmic domains of E-cadherin and desmosomal cadherin Dsg2, and thetumor suppressor APC protein. J. Biochem. 118, 1077-1082.

Peifer, M. and Wieschaus, E. (1990). The segment polarity gene armadilloencodes a functionally modular protein that is the Drosophila homolog ofhuman plakoglobin. Cell 63, 1167-1178.

Peifer, M., McCrea, P. D., Green, K. J., Wieschaus, E. and Gumbiner, B. M.(1992). The vertebrate adhesive junction proteins β-catenin and plakoglobinand the Drosophila segment polarity gene armadillo form a multigene familywith similar properties. J. Cell Biol. 118, 681-691.

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

Peifer, M., Pai, L. M. and Casey, M. (1994b). Phosphorylation of theDrosophila adherens junction protein Armadillo: roles for wingless signaland Zeste-white 3 kinase. Dev. Biol. 166, 543-556.

Peifer, M. (1995). Cell adhesion and signal transduction: the armadilloconnection. Trends Cell Biol. 5, 224-229.

Piepenhagen, P. A. and Nelson, W. J. (1993). Defining E-cadherin-associatedprotein complexes in epithelial cells: plakoglobin, β- and γ-catenin aredistinct components. J. Cell Sci. 104, 751-762.

Rimm, D. L., Koslov, E. R., Kebriaei, P., Cianci, C. D. and Morrow, J. S.(1995). α(E)-catenin is an actin-binding and -bundling protein mediating theattachment of F-actin to the membrane adhesion complex. Proc. Nat. Acad.Sci. USA 92, 8813-8817.

Roh, J.-Y. and Stanley, J. R. (1995). Plakoglobin binding by human Dsg3(pemphigus vulgaris antigen) in keratinocytes requires the cadherin-likeintracytoplasmic segment. J. Invest. Dermatol. 104, 720-724.

Rubenstein, A., Merriam, J. and Klymkowsky, M. W. (1997). Localizing theadhesive and signaling functions of plakoglobin. Dev. Genet. 20, 91-102.

Rubinfeld, B., Souza, B., Albert, I., Munemitsu, S. and Polakis, P. (1995).The APC protein and E-cadherin form similar but independent complexeswith α-catenin, β-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 micereveals essential functions of desmosomes in the embryonic heart. J. CellBiol. 135, 215-225.

Sacco, P. A., McGranahan, T. M., Wheelock, M. J. and Johnson, K. R.(1995). Identification of plakoglobin domains required for association withN-cadherin and α-catenin. J. Biol. Chem. 270, 20201-20206.

Shibata, T., Gotoh, M., Ochiai, A. and Hirohashi, S. (1994). Association ofplakoglobin with APC, a tumor suppressor gene product, and its regulationby tyrosine phosphorylation. Biochem. Biophys. Res. Commun. 203, 519-522.

Stappenbeck, T. S. and Green, K. J. (1992). The desmoplakin carboxylterminus coaligns with and specifically disrupts intermediate filamentnetworks when expressed in cultured cells. J. Cell Biol. 116, 1197-1209.

Stappenbeck, T. S., Bornslaeger, E. A., Corcoran, C. M., Luu, H. H., Virata,M. L. A. and Green, K. J. (1993). Functional analysis of desmoplakindomains: specification of the interaction with keratin versus vimentinintermediate filament networks. J. Cell Biol. 123, 691-705.

Troyanovsky, S. M., Troyanovsky, R. B., Eshkind, L. G., Krutovskikh, V.A., Leube, R. E. and Franke, W. W. (1994a). Identification of theplakoglobin-binding domain in desmoglein and its role in plaque assemblyand intermediate filament anchorage. J. Cell Biol. 127, 151-160.

Troyanovsky, S. M., Troyanovsky, R. B., Eshkind, L. G., Leube, R. E. andFranke, W. W. (1994b). Identification of amino acid sequence motifs indesmocollin, a desmosomal glycoprotein, that are required for plakoglobinbinding and plaque formation. Proc. Nat. Acad. Sci. USA 91, 10790-10794.

Troyanovsky, R. B., Chitaev, N. A. and Troyanovsky, S. M. (1996). Cadherinbinding sites of plakoglobin: localization, specificity and role in targeting toadherens junctions. J. Cell Sci. 109, 3069-3078.

Wahl, J. K., Sacco, P. A., McGranahan-Sadler, T. M., Sauppe, L. M.,Wheelock, M. J. and Johnson, K. R. (1996). Plakoglobin domains thatdefine its association with the desmosomal cadherins and the classicalcadherins: identification of unique and shared domains. J. Cell Sci. 109,1143-1154.

Witcher, L. L., Collins, R., Puttagunta, S., Mechanic, S. E., Munson, M.,Gumbiner, B. and Cowin, P. (1996). Desmosomal cadherin bindingdomains of plakoglobin. J. Biol. Chem. 271, 10904-10909.

Yost, C., Torres, M., Miller, J. R., Huang, E., Kimelman, D. and Moon, R. T.(1996). The axis-inducing activity, stability, and subcellular distribution of β-catenin is regulated in Xenopus embryos by glycogen synthase kinase 3.Genes Dev. 10, 1443-1454.

(Received 29 April 1997 – Accepted 10 July 1997)

Roles of plakoglobin end domains in desmosome assembly

Documents

Transcript of Roles of plakoglobin end domains in desmosome assembly