View metadata, citation and similar papers at core.ac.uk brought to you by CORE providedORIGINAL by Elsevier - ARTICLE Publisher Connector See related Commentary on page 1215 Di¡erential E¡ects of Desmoglein 1 and Desmoglein 3 on Desmosome Formation

Yasushi Hanakawa, Masayuki Amagai,n Yuji Shirakata, Yoko Yahata, Sho Tokumaru, Kenshi Yamasaki, Mikiko Tohyama, Koji Sayama, and Koji Hashimoto Department of Dermatology, School of Medicine, Ehime University, Ehime, Japan; nDepartment of Dermatology, School of Medicine, Keio University, Tokyo, Japan

The desmoglein plays an important part in the forma- desmoglein 3DEC. Therefore, we conclude that the tion of desmosomes. We constructed recombinant ade- dominant-negative e¡ect of desmoglein 1DEC is not noviruses containing desmoglein 1 and desmoglein 3 simply due to plakoglobin sequestration. On the other derivatives partly lacking the extracellular domain (des- hand, during low-level expression of full-length desmo- moglein 1DEC and desmoglein 3DEC, respectively), and glein 3 and desmoglein 1, they both colocalized with full-length desmoglein 1 and desmoglein 3 and studied desmoplakin. During high-level expression, however, the involvement of desmoglein 1 and desmoglein 3 in keratin insertion at cell^cell contact sites was inhibited desmosome formation. During low-level expression in desmoglein 1 but not in desmoglein 3, and desmopla- of desmoglein 3DEC in transduced HaCaTcells, keratin kin was stained at cell^cell contact sites in desmoglein 3 insertion at cell^cell contact sites was only partially but not in desmoglein 1. These data suggest desmoglein inhibited and desmoplakin was partially stained at 1 and desmoglein 3 expressed at low level were incorpo- cell^cell contact sites. Low-level expression of desmo- rated into desmosome but at high-level expression, des- glein 1DEC, however, resulted in complete inhibition moglein 1 disrupted desmosomes but desmoglein 3 did of keratin insertion at the cell^cell contact sites, and not. Our ¢ndings provide biologic evidence that des- desmoplakin was stained in perinuclear dots. These moglein 1 and desmoglein 3 play a di¡erent functional results indicate the dominant-negative e¡ect of desmo- role in cell^cell adhesion of keratinocytes. Key words: glein 1DEC on desmosome formation was stronger than desmosomes/desmoglein 1/desmoglein 3. J Invest Dermatol that of desmoglein 3DEC. Desmoglein 1DEC coprecipi- 119:1231 ^1236, 2002 tated plakoglobin to approximately the same extent as

esmosomes are cell^cell adhesion complexes that is a disease caused by autoantibodies directed against Dsg3 in provide mechanical integrity to keratinocytes by which skin lesion biopsies exhibit suprabasilar acantholysis (Udey linking to keratin intermediate ¢laments. Desmo- and Stanley, 1999). On the other hand, pemphigus foliaceus is somes are composed of two major transmembrane caused by autoantibodies directed against Dsg1 and in this case, proteins, desmoglein (Dsg) and desmocollin (Ama- lesion biopsies exhibit subcorneal acantholysis. Dsg3 knockout Dgai, 1996a; Kowalczyck et al, 1999; Green and Gaudry, 2000). In mice phenotypically mimicked pemphigus vulgaris patients humans, three desmoglein isoforms have been identi¢ed: Dsg1, (Koch et al, 1997) in that they displayed oral erosions and loss of Dsg2, and Dsg3. They are encoded by individual genes and dif- intercellular adhesion of suprabasal layers of the mucosal epithe- ferentially distributed in tissue. Dsg2 is expressed in all desmo- lium and epidermis. some-containing tissues, including simple epithelium and Several studies have suggested that both the extracellular and myocardium. In contrast, Dsg1 and Dsg3 are expressed cytoplasmic domains of desmogleins are critical for normal des- in strati¢ed squamous epithelia. Dsg3 is found in the basal and mosome formation. N-terminally truncated Dsg3 caused domi- suprabasal layers of stratifying epithelia, whereas Dsg1 is nant-negative e¡ects on desmosome formation in HaCaT cells dominantly expressed in the di¡erentiated upper layers of (Hanakawa et al, 2000). Expression of chimeric molecules con- epithelia (Arnemann et al, 1993; Shimizu et al,1995;Amagaiet al, taining the transmembrane domain of connexin and cytoplasmic 1996b). domain of Dsg1 disrupted desmosomes in A431 cells (Troyanovs- Desmogleins play important parts in the formation and main- ky et al, 1993). Similarly, a chimeric molecule containing the ex- tenance of desmosomes. Autoantibodies against desmogleins lead tracellular domain of E-cadherin and cytoplasmic domain of to impairment of epidermal tissue integrity. Pemphigus vulgaris Dsg1 (Ecad-Dsg1) disrupted desmosomes in A431 cells (Norvel and Green, 1998). In contrast, a recent report indicated that a chi- meric molecule of the extracellular domain of E-cadherin and Manuscript received January 10, 2002; revised June 24, 2002; accepted for cytoplasmic domain of Dsg3 (Ecad-Dsg3) was incorporated into publication August 29, 2002 Reprint requests to: Yasushi Hanakawa, MD, Ehime University School desmosomes of A431 cells (Andl and Stanley, 2001). This di¡er- of Medicine, Department of Dermatology, Shitukawa, Shigenobu, Onsen- ence between Ecad-Dsg1 and Ecad-Dsg3 indicates functional dif- gun, Ehime 791-0295, Japan. Email: [email protected] ferences of cytoplasmic domains of Dsg1 and Dsg3. In addition, Abbreviations: Dsg, desmoglein; MOI, multiplicity of infection. full-length Dsg1 disrupted desmosome when the expression level

0022-202X/02/$15.00 Copyright r 2002 by The Society for Investigative Dermatology, Inc.

1231 1232 HANAKAWA ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY was high; however, full-length Dsg1 was incorporated into des- (murine anti-desmoplakin; PROGEN); PG5.1 (murine anti-plakoglobin; mosome when the expression level was low in A431 cells (Norvel PROGEN); PP1-5C2 (murine anti-plakophilin 1; PROGEN); anti- and Green, 1998). These data suggest that Dsg1 and Dsg3 expres- plakophilin 2 (murine; PROGEN); anti-involucrin (Abcam, Cambridge, sion have di¡erent e¡ects on desmosome formation and integrity. U.K.); and 6H6 (murine anti-Dsg3). The following rabbit anti-sera were used: Z622 (anti-pan-keratin; DAKO, Copenhagen, Denmark); anti-myc To characterize further and de¢ne these di¡erences, we con- (a kind gift from Dr John Stanley, University of Pennsylvania, PA); anti- structed full-length and N-terminally truncated mutants of keratin 1 (CONVENCE, Richmond, CA); anti-loricrin (CONVENCE); Dsg1 and Dsg3, and introduced them into cultured keratinocytes and anti-FLAG (Abcam). Fluorescein isothiocyanate-conjugated goat anti- using an adenovirus vector. Taking advantage of the adenovirus mouse or goat anti-rabbit antibodies (Kirkegaard and Perry Laboratories, vector to control expression level by changing multiplicity of Gaithersburg, MD) and rhodamine-conjugated goat anti-rabbit antibody infection (MOI), we evaluated the e¡ects of expression levels of (BioSource, Camarillo, CA) were used. these isoforms on the integrity of a desmosome^keratin struc- tured complex. Immunohistochemistry Cells grown on Laboratory-Tech 4-well culture slides (Nalge-Nunc, Napierville, IL) were ¢xed with methanol at ^201C for 20 min and permeabilized with 0.05% Triton X-100 in Tris- 2 þ bu¡ered saline containing 1 mM CaCl2 (TBS-Ca ) at room temperature MATERIALS AND METHODS for 5 min. After incubation with 1% bovine serum albumin in TBS-Ca2 þ for 20 min at room temperature, the cells were incubated with various Cell culture The human embryonic kidney cell line 293 was obtained antibodies at the appropriate dilution in 1% bovine serum albumin in from American TypeTissue Culture (ATCC; Rockville, MD). The human TBS-Ca2 þ for 1 h at room temperature. The samples were further naturally immortalized keratinocyte HaCaTcell line was a kind gift from incubated with £uorescein isothiocyanate- or rhodamine-conjugated Dr Norbert Fusenig (German Cancer Research Center, Heidelberg, secondary antibodies at the appropriate dilution in 1% bovine serum Germany). These cells were cultured in Dulbecco modi¢ed Eagle’s albumin in TBS-Ca2 þ for 1 h at room temperature. Stained cells were medium supplemented with 10% fetal bovine serum. examined using a confocal laser microscope (Zeiss, Oberkachen, Germany). Plasmid construction The full-length cDNA encoding human Dsg1 was a kind gift from Dr Kathleen Green (North-western University, Immunoblotting and immunoprecipitation For immunoblotting, Chicago, IL). The construction of the Dsg3 mutant, in which a large part cells were lyzed in sodium dodecyl sulfate sample bu¡er (62.5 mM Tris^ of the extracellular domain was deleted and a seven c-myc tag was inserted HCl, pH 7.5, 1% sodium dodecyl sulfate, 0.0025% bromophenol blue, £anking the C-terminal end, has been described previously (Hanakawa 10% glycerol, 2.5% 2-mercaptoethanol) and subjected to immunoblotting et al, 2000). A mutant of Dsg1 (Dsg1DEC), with a large part of the using enhanced chemi£uorescence according to the manufacturer’s extracellular domain deleted and with a seven c-myc tag £anking the protocol (Amersham Pharmacia Biotech, Uppsala, Sweden). For soluble C-terminal end, was also constructed. Dsg1 DNA fragments (nucleotides and insoluble fractionation, the cells were scraped in 1% Nonidet P-40 13^402 and 1726^3360) were generated by polymerase chain reaction and 1% Trion-X-100 in TBS-Ca2 þ on ice. Following centrifugation at using primers DG1F150 (50 -TTTTAGGGTGGGGATCCAGAC-30), DG1F130 1 0 0 0 15,000 r.p.m. (30,000g) for 10 min at 4 C, the supernatant was collected as (5 -GGGGTACCGTCTTCACCTTCACGACAGGC-3 , KpnI site at the 3 the soluble fraction. The pellet was dissolved in sodium dodecyl sulfate end), DG1F250 (50 -GGGGTACCGGCAGACAAGAAAGTACT-30, KpnI 0 0 0 sample bu¡er as the insoluble fraction. Immunoprecipitation was carried site at the 5 end), and DG1F23 (5 -CGGAATTCCTTGCTATAT- out on adenovirus-infected cells extracted with 1% Nonidet P-40, 1% TGCACGGT-30, EcoRI site at the 30 end). The polymerase chain reaction Triton X-100, 2 mM CaCl2,and150mM NaCl in 10 mM Tris^HCl, pH products were subcloned into pcDNA1^7myc/Amp to generate 7.4, in the presence of protease inhibitors (2 mg aprotinin per ml, 2 mg pcDsg1DEC. pcDNA1^7myc/Amp was a kind gift from Dr Kathleen leupeptin per ml, phenylmethylsulfonyl £uoride 1 mM). Aliquots of cell Green (North-western University; Roth et al,1991). lysates were preadsorbed with normal mouse IgG or normal rabbit IgG and protein G-Sepharose (Amersham Pharmacia Biotech), then Adenovirus vector construction and infection The cosmid cassette R precipitated with various antibodies and protein G-Sepharose. pAxCAw, pAxCALNLw (Kanegae et al, 1996), the loxP-Neo -loxP unit Immunoprecipitates were eluted from protein G-Sepharose with sodium under a CA promoter consisting of a cytomegalovirus enhancer and dodecyl sulfate sample bu¡er and subjected to immunoblotting. The chicken b-actin promoter (Niwa et al, 1991), the nuclear localizing signal- immunoblots were analyzed by using ImageQuantTM (Amersham tagged Cre recombinase-expressing adenovirus (AxCANCre), the control Pharmacia Biotech) software to determine the relative ratio in the adenovirus Ax1w, and the parent virus Ad5-dLX (Miyake et al,1996) immunoprecipitated complex. were all kind gifts from Dr Izumu Saito (Tokyo University, Japan). Fragments of Dsg1DEC were subcloned into the adenovirus cosmid cassette pAxCALNLw. Adenovirus containing CALNL and Dsg1DEC RESULTS (AxCALNLDsg1DEC) were generated by the COS-TPC method (Miyake et al, 1996). The cosmid DNA was mixed with the EcoT22I- digested DNA-terminal protein complex of Ad5-dLX, and used to Desmosome disruption is greater with Dsg1DEC than with cotransfect 293 cells. Recombinant viruses were generated through Dsg3DEC A Dsg1 mutant was constructed by deleting a large homologous recombination in 293 cells. Virus stocks were prepared by a part of the extracellular domain and adding a seven c-myc tag standard procedure (Miyake et al, 1996). Concentrated, puri¢ed virus (Dsg1DEC; Fig 1). The corresponding Dsg3 mutant has been stocks were prepared by the CsCl gradient method and the virus titer was previously described (Hanakawa et al,2000;Fig 1). To prevent checked using the plaque formation assay. HaCaT cells were doubly the expression of toxic, truncated products that interfered with infected with AxCANCre and Ax1w, AxCALNLDsg1DEC, or the production of recombinant virus, the loxP gene sequence, AxCALNLDsg3DEC. HaCaT cells were infected with AxCANCre at a disrupted by a neomycin-resistance cassette, was interposed MOI of 5 and Ax1w, AxCALNLDsg1DEC, or AxCALNLDsg3DEC at a MOI of 5 (low expression) or were infected with AxCANCre at a MOI between the CAG promoters and the coding regions in the of 15 and Ax1w, AxCALNLDsg1DEC, or AxCALNLDsg3DEC at a MOI mutants (Kanegae et al, 1996). Co-infection of keratinocytes of 15 (high expression). (HaCaT cells) with adenovirus-expressing Cre recombinase Adenovirus expressing mouse Dsg1-FLAG and mouse Dsg3-FLAG have removed the stu¡er sequence and activated the expression of the already been shown elsewhere (Amagai et al, 2002). Brie£y, cDNA mutant desmogleins. The expression of mutant proteins was encoding mouse Dsg1-FLAG and mouse Dsg3-FLAG were subcloned detected at 6 h and reached a plateau between 24 and 36 h after into the adenovirus cosmid cassette pAxCAw. Adenovirus containing CA infection (data not shown). Almost all cells expressed the mutant promoter and cDNA encoding mouse Dsg1-FLAG and mouse Dsg3- proteins when HaCaT cells were coinfected with adenovirus FLAG (AxmDsg1F and AxmDsg3F) were generated by the COS-TPC carrying mutant desmogleins and Cre recombinase, £anked method (terminal protein complex). Puri¢ed, concentrated, and titer- with the nuclear localization signal (NCre) at a MOI of 30 checked viruses were infected to HaCaTcells at a MOI of 10 and 30. (mutant: Cre ¼1 : 1) or a MOI of 10. The relative expression Antibodies The following monoclonal antibodies were used: AE1 and levels of both mutants were essentially the same when HaCaT AE3 (murine anti-keratin intermediate ¢lament; PROGEN, Heidelberg, cells were infected at the same MOI, with a MOI of 30 causing Germany); 9E10.2 (murine anti-myc; American TypeTissue Culture); anti- greater expression than MOI of 10 (Fig 2). No toxicity was b-catenin (murine; Transduction Laboratories, Lexington, KY); DPI/II apparent in cells infected with control adenovirus (Ax1w) or VOL. 119, NO. 6 DECEMBER 2002 DIFFERENCE BETWEEN DSG 1 AND DSG3 1233

Figure1. Mutant and full-length Dsg1 and Dsg3. Mutants of human Dsg1 and Dsg3 (Dsg1DEC and Dsg3DEC, respectively) were constructed by deleting a large part of the extracellular domain (EC1^EC4) and £ank- ing the C-terminal end with a seven c-myc tag. Full-length mouse Dsg1 and Dsg3 were tagged with FLAG epitope in C-terminal end. S, signal peptide; P, prosequence; EC, extracellular; EA, extracellular anchor do- main; IA, intracellular anchor domain; ICS, intracellular cadherin-speci¢c domain; IPL, intracellular proline-rich domain; RUD, repeating unit domain; DTD, desmoglein-speci¢c terminal domain.

Figure 3. Mutant protein expression level and desmosome forma- tion. Double label immuno£uorescence staining of Dsg1DEC and Dsg3DEC adenovirus-infected HaCaT cells in low-power view (a), and high-power view (b). Cells infected at di¡erent adenovirus MOI (10 or 30) were doubly stained with anti-keratin (J^N) and anti-myc (O^R)or anti-desmoplakin (A^E) and anti-myc (F^I) antibodies. When Dsg1DEC and Dsg3DEC were expressed at high levels (MOI ¼ 30), keratin insertion at the cell^cell contact sites was inhibited (L,N) and desmoplakin was stained in perinuclear dots (C, E). When Dsg3DEC was expressed at low Figure 2. Expression of mutant desmogleins in HaCaTcells. Immu- levels (MOI ¼ 10), however, keratin insertion was only partially inhibited noblots using anti-myc antibody show the expression of Dsg1DEC and (M) and desmoplakin stained at cell^cell contact sites (E), whereas in Dsg3DEC in HaCaT cells 36 h after adenovirus infection. The expression Dsg1DEC adenovirus-infected cells keratin insertion was inhibited com- levels of Dsg1DEC and Dsg3DEC were stronger at a MOI of 30 than at a pletely (K) and desmoplakin was stained in perinuclear dots (B). MOI of 10.

NCre adenovirus at MOI of 10 or 30 in phase microscope. We after infection with control adenovirus, Dsg1DEC adenovirus, or examined the e¡ects of these di¡erent expression levels of Dsg3DEC adenovirus at a MOI of 30 (data not shown). mutant desmogleins on HaCaT cells, at 36 h after infection. We next studied desmosomal changes in HaCaTcells induced There were no apparent morphologic changes in cells infected by each mutant by immuno£uorescence localization of keratin with Dsg1DEC adenovirus or Dsg3DEC adenovirus when and desmoplakin. Dsg1DEC and Dsg3DEC adenovirus compared with cells infected with the control adenovirus, using infections e¡ectively abolished keratin insertion and phase microscopy (data not shown). The e¡ects of mutant desmoplakin expression at cell^cell contact sites when the desmogleins on the formation of adherens junctions were expression levels of the mutant were high (MOI ¼ 30; Fig examined by immuno£uorescence staining using double 3a,b,C,L,E,N). Keratin was found in perinuclear regions, and staining for b-catenin (a marker of adherens junctions) and myc. desmoplakin was seen in punctuate locations throughout the No di¡erences were observed in b-catenin staining patterns 36 h cytoplasmic region. No di¡erences were observed in the keratin 1234 HANAKAWA ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Figure 5. Expression of full-length desmogleins in HaCaTcells. Im- munoblots using anti-FLAG antibody show the expression of full-length Dsg1 and Dsg3 in HaCaTcells 36 h after adenovirus infection. The expres- sion levels of Dsg1 and Dsg3 were stronger at a MOI of 30 than at a MOI Figure 4. Coprecipitation of plakoglobin with Dsg1DEC and of 10. Dsg3DEC. Extracts from HaCaT cells infected with the desmoglein mu- tants were immunoprecipitated with polyclonal anti-myc (A) or anti-pla- koglobin (B) antibodies, and immunoblotted with either monoclonal anti-myc or anti-plakoglobin antibodies. Plakoglobin coprecipitated with both Dsg1DEC and Dsg3DEC to the same extent and this was also true when the order of addition of the antibodies was reversed. or desmoplakin staining patterns 36 h after infection with control adenovirus (Fig 3a,b,A,J). In contrast, when the expression level was low (MOI ¼10), Dsg3DEC adenovirus partially a¡ected on keratin insertion and desmoplakin expression (Fig 3a,b,D,M), whereas Dsg1DEC adenovirus completely inhibited keratin insertion and desmoplakin expression at the cell^cell contact site (Fig 3a,b,B,K). In low-level expression, the Dsg3DEC mutant proteins were detected with anti-myc antibodies at cell^cell boundaries and in the cytoplasm (Fig 3b,H,Q); however, the Dsg1DEC mutant proteins were detected mainly in the Figure 6. Full-length Dsg1 and Dsg3 were incorporated into des- cytoplasm but not at cell^cell contact sites (Fig 3b,F,O). In mosome at low-level expression. HaCaT cells were transduced with high-level expression, both the Dsg3DEC and Dsg1DEC mutant adenovirus containing full-length Dsg1 and Dsg3 at a MOI of 10. Double proteins were detected at cell^cell boundaries and in the immuno£uorescence staining of anti-FLAG (A,D) and anti-desmoplakin cytoplasm (Fig 3b,G,I,P,R). (B, E) showed colocalization of Dsg1 and Dsg3 with desmoplakin at cell^ cell contact site. C Shows a merged picture of A and B,andF shows a Plakoglobin binds equally to Dsg1DEC and Dsg3DEC We merged picture of D and E. combined immunoprecipitation with immunoblotting in order to detect plakoglobin bound to either Dsg1DEC or Dsg3DEC (Fig 4). Cell extracts from HaCaT cells expressing mutant antibody showed the relative expression levels of Dsg1 and Dsg3 desmogleins were immunoprecipitated with an anti-myc were essentially the same when HaCaTcells were infected at the antibody, then subsequently immunoblotted with anti-myc and same MOI (Fig 5). Double immuno£uorescence staining of anti- anti-plakoglobin antibodies. Plakoglobin coprecipitated with the FLAG and anti-desmoplakin showed colocalization of Dsg1 and Dsg1DEC and Dsg3DEC proteins expressed by the two mutants Dsg3 with desmoplakin at cell^cell contact site when the with similar e⁄ciencies (Fig 4A). When the order in which the expression level was low, although in the merged picture red antibodies were used was reversed, i.e., immunoprecipitation staining of anti-FLAG was stronger compared with desmoplakin with anti-plakoglobin followed by immunoblotting with anti- staining (Fig 6). These results show the incorporation of Dsg1 myc antibody, plakoglobin was again coprecipitated with and Dsg3 into desmosomes when their expression levels were Dsg1DEC and Dsg3DEC at similar levels (Fig 4B). We could low. not detect any association of plakophilin 1 and plakophilin 2 with Dsg1DEC or Dsg3DEC (data not shown). In high-level expression, full-length Dsg1 disrupts Several published data indicate that disruption of cell adhesion desmosome formation but Dsg3 does not On the other leads to altered di¡erentiation in keratinocytes (Allen et al,1996; hand, when the expression levels were high, Dsg1 and Dsg3 Zhu and Watt, 1996). To test whether Dsg1DEC and Dsg3DEC showed di¡erent e¡ects on desmosome formation. In high level might a¡ect di¡erentiation in HaCaT cells, we analyzed the full-length expressing HaCaT cells, double immuno£uorescence di¡erentiation marker of keratinocytes after transduction with staining of anti-FLAG and anti-keratin showed insertion of Dsg1DEC and Dsg3DEC. Western blotting of Dsg1DEC and keratin to cell^cell contact site in Dsg3 (Fig 7A,B). Disruption Dsg3DEC expressing HaCaT cell extracts stained with anti- of insertion to cell^cell contact site and shrinking around keratin 1, anti-involucrin, and anti-loricrin antibody antibodies nucleus of keratin, however, was observed in Dsg1. Double showed no change (data not shown), suggesting dominant- staining of FLAG and desmoplakin showed dot staining of negative e¡ects on desmosomes did not a¡ect di¡erentiation in desmoplakin at cell^cell contact site in Dsg3, suggesting Dsg3 is our system. still incorporated into desmosomes at high-level expression. In contrast, high-level expression of Dsg1 changed staining of Full-length Dsg1 and Dsg3 are incorporated into desmoplakin to a di¡use cytoplasmic pattern. These results desmosome in low-level expression We also tried to ¢nd the suggest that when full-length protein expression levels were functional di¡erence in full-length Dsg1 and Dsg3 in desmosome high, Dsg1 disrupts desmosome formation but Dsg3 is still formation. We constructed adenovirus containing full-length incorporated into desmosomes. In high-level expression, Dsg1 and Dsg3 with FLAG epitope tag on the C-termini and full-length Dsg3 localized mainly at cell^cell contact site transduced HaCaT cells. Western blots stained with anti-FLAG (Fig 7A,C); however, Dsg1 localized both cell^cell contact site VOL. 119, NO. 6 DECEMBER 2002 DIFFERENCE BETWEEN DSG 1 AND DSG3 1235

interactions occur between the mutant cadherins and plakophilin 1 and plakophilin 2. The other possibility for the disruption of desmosome is in- stability of desmosomes at cell^cell contact sites. Exfoliative toxin from Staphylococcus aureus cleaves the extracellular domain of Dsg1 and causes dysfunction of Dsg1 (Amagai et al, 2001, 2002). Keratinocytes in neonatal mice injected with exfoliative toxin A or B showed internalization of Dsg1 and resultant blister forma- tion in the upper layers of the epidermis. We speculate that cleaved Dsg1 might be unstable in the desmosome because it does not bind its partner, and such unstabilized desmosomes could be easily internalized; however, these observed internalizations only happened in living epidermis but were not seen in the cryosec- tion of epidermis (Amagai et al, 2002). It is probable that the dy- Figure 7. Full-length Dsg1 disrupted desmosome but Dsg3 did not namics of a living cell are a prerequisite for desmosome at high-level expression. HaCaT cells were transduced with adenovirus internalization after the dysfunction of desmogleins. These data containing full-length Dsg1 and Dsg3 at a MOI of 30. Double immuno- are consistent with our ¢ndings of a retracted, perinuclear stain- £uorescence staining of anti-FLAG and anti-keratin showed insertion of ing of keratin ¢lament network in HaCaT cells expressing high keratin to cell^cell contact site in Dsg3 (A,B). Keratin insertion to cell^cell amounts of mutant desmogleins. The reason for the disruption contact site was inhibited in Dsg1 (E,F). Double staining of FLAG and des- of the desmosome with full-length Dsg1 might be similar. We moplakin showed dot staining of desmoplakin at cell^cell contact site and speculate that Dsg1 might associate with more desmosomal mo- linear cell^cell staining of FLAG in Dsg3 (C, D ). In contrast, high-level ex- lecules than Dsg3, based on our results with HaCaT cells with pression of Dsg1 changed staining of desmoplakin to di¡use cytoplasmic regard to basal layer keratinocytes in the epidermis. When Dsg1 pattern (G, H). cannot ¢nd enough amount partners to stabilize on the cell sur- face, Dsg1 might be easily internalized with already interacted desmosomal proteins. A recent report that Dsg2 e⁄ciently incor- and cytoplasmic region (Fig 7E,G). Although full-length Dsg1 porated into desmosomes but Dsg1 did not in MDCK and A431 in cytoplasm showed reticular pattern, we cannot explain the cells (Ishii et al, 2001) support the idea that cell and di¡erentiation reason for such a staining pattern. speci¢c capacity for isotype dependent incorporation of desmo- gleins into desmosome. This study has shown a functional di¡erence between Dsg1 DISCUSSION and Dsg3 on desmosome formation in keratinocytes; however, why does the human body need di¡erent isoforms of Dsg1 and We investigated the roles of Dsg1 and Dsg3 in desmosome for- Dsg3 in strati¢ed epithelia? Dsg3 knockout mice showed erosions mation in HaCaT keratinocytes with adenovirus vectors contain- in the oral mucosa and in areas subject to mechanical irritation ing dominant-negative mutants and full-length forms of Dsg1 (Koch et al, 1997), although the epidermis was intact in nonme- and Dsg3. When the mutant desmoglein expression levels were chanically stressed areas. This result suggests that Dsg1 may com- high, both Dsg1DEC and Dsg3DEC inhibited the formation of pensate functionally for a de¢ciency in Dsg3 (Mahoney et al, desmosomes. When expression levels were low, however, 1999; Wu et al, 2000). If this hypothesis is correct, small amounts Dsg3DEC only partially inhibited the formation of desmosomes, of Dsg1 in the lower part of the epidermis are enough to produce whereas Dsg1DEC completely inhibited the formation of desmo- desmosomes, although normal desmosomal function may be somes. These ¢ndings indicate that the dominant-negative e¡ect compromised in such situations. Although expression levels of on desmosome formation of the mutated Dsg1 was stronger than Dsg1 are lower than Dsg3, Dsg1 may play an important part in that of mutated Dsg3. On the other hand, when the full-length desmosome formation in basal layer keratinocytes in epidermis desmoglein expression levels were low, both Dsg1 and Dsg3 in- and in certain circumstances, compensate for the lack of Dsg3 corporated into desmosomes. In contrast, when expression levels function. The distribution of Dsg1 and Dsg3 is di¡erent between were high, Dsg1 inhibited the formation of desmosomes, whereas the epidermis and mucous membrane (Shirakata et al,1998).In Dsg3 did not. These ¢ndings indicate that the functional di¡er- mucous membranes, Dsg1 is less expressed in upper layers com- ence between the full-length form of Dsg1 and Dsg3 at least in pared with epidermis. Recently, involucrin promoter driven less di¡erentiated cells in proliferating in culture. transgenic expression of Dsg3 in mouse showed altered stratum Why is the e¡ect of the Dsg1DEC mutant stronger than that of corneum and increased transepidermal water loss in epidermis Dsg3DECD One possibility is that mutated desmoglein interacts (Elias et al, 2001). These data suggest that Dsg1 and Dsg3 contri- with desmosome components. By combining immunoprecipita- bute not only to the formation of desmosomes but also to tion with subsequent immunoblotting, we found that the cyto- strati¢cation. Strictly organized expression of Dsg1 and Dsg3 plasmic domains of Dsg1DEC and Dsg3DEC interacted with might be needed to architect the di¡erence in epidermis and mu- plakoglobin to the same extent. Recently it has been reported cous membrane, which have di¡erent functions, such as water that the binding ratio of plakoglobin/full-length Dsg1 exhibited loss protection in epidermis and water absorption in mucous less than 2 : 1 but was still higher compared with Dsg3 (Bannon membrane. et al, 2001). Sequestration of plakoglobin by the mutant desmo- In summary, the discovery of the di¡erence in e¡ects on des- glein proteins would make it inaccessible to native desmoglein mosome by Dsg1DEC and Dsg3DEC mutants and by full-length and might partly explain the dominant-negative e¡ect. The idea forms of Dsg1 and Dsg3 support the hypothesis of functional of di¡erential sequestering of plakoglobin, however, is contrain- di¡erences in desmoglein activities during desmosome formation dicated by the ¢nding that there were no di¡erences in plakoglo- in keratinocytes. Further investigations are needed in order to bin-binding capacity between Dsg1DEC and Dsg3DEC. It seems clarify the precise roles of di¡erent desmoglein isotopes in likely that molecules other than plakoglobin contribute to the human tissue. formation of desmosomes. Another armadillo gene family mem- ber of plakophilin (Hatzfeld et al, 1994; Heid et al, 1994; Mertens et al, 1996; Bornslaeger et al, 2001) is now thought to be a major We especially thank Dr John Stanley for insightful discussion and critical reading of desmosomal plaque protein; however, we could not detect bind- this manuscript.We thank MsTerukoTsuda and Mrs Akiko Kon for expert technical ing of plakophilin 1 or plakophilin 2 to Dsg1DEC or Dsg3DEC assistance. We also thank Dr Izumu Saitou for the adenovirus expression system (data not shown).We cannot rule out the possibility that indirect and adenovirus Cre-loxP system, Dr June-ichi Miyazaki for the CAG promoter, 1236 HANAKAWA ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Dr Norbert Fusenig for the HaCaTcells, Dr Kathleen Green for the pcDNA1^7myc/ Heid HW, Schmidt A, Zimbelmann R, et al: Cell type-speci¢c desmosomal plaque Amp and cDNA encoding human Dsg1,and DrJohn Stanley for the polyclonal anti- proteins of the plakoglobin family: plakophilin 1 (band 6 protein). Di¡erentia- myc antibody.This work was supported by a Health Sciences Research Grant for Re- tion 58:113^131, 1994 Ishii K, Norvell SM, Bannon LJ, Amargo EV, Pascoe LT, Green KJ: Assembly of search on Speci¢c Diseases from the Ministry of Health, labor and Welfare of Japan desmosomal cadherins into desmosomes is isoform dependent. J Invest Dermatol and a Grant-in-Aid of Scienti¢c Research from the Ministry of Education, Cluture, 117:26^35, 2001 Sports, Science andTechnology of Japan. Kanegae Y,Takamori K, Sato Y, Lee G, Nakai M, Saito I: E⁄cient gene activation system on mammalian cell chromosomes using recombinant adenovirus producing Cre recombinase. Gene 181:207^212, 1996 Koch PJ, Mahoney MG, Ishikawa H, et al: Targeted disruption of the pemphigus vulgaris antigen (desmoglein 3) gene in mice causes loss of keratinocyte cell adhesion with a phenotype similar to pemphigus vulgaris. JCellBiol REFERENCES 137:1091^1102, 1997 Kowalczyck AP, Bornslaeger EA, Norvell SM, Palka HL, Green KJ: Desmosomes. Allen E, Yu QC, Fuchs E: Mice expressing a mutant desmosomal cadherin exhibit intercellular adhesive junctions specialized for attachment of intermediate abnormalities in desmosomes, proliferation, and epidermal di¡erentiation. ¢laments. Int Rev Cytol 185:237^302, 1999 J Cell Biol 133:1367^1382, 1996 Mahoney MG, Wang Z, Rothenberger K, Koch PJ, Amagai M, Stanley JR: Expla- Amagai M: Pemphigus: autoimmunity to epidermal cell adhesion molecules. Adv nations for the clinical and microscopic localization of lesions in pemphigus Dermatol 11:319^352, 1996a foliaceus and vulgaris. JClinInvest103:461^468, 1999 Amagai M, Koch PJ, Nishikawa T, Stanley JR: Pemphigus vulgaris antigen (desmo- Mertens C, Kuhn C, Franke WW: Plakophilins 2a and 2b. constitutive proteins of glein 3) is localized in the lower epidermis, the site of blister formation in pa- dual location in the karyoplasm and the desmosomal plaque. JCellBiol tients. JInvestDermatol106:351^355, 1996b 135:1009^1025, 1996 Amagai M, Matsuyoshi N,Wang ZH, Andl C, Stanley JR: Toxin in bullous impet- Miyake S, Makimura M, Kanegae Y, et al: E⁄cient generation of recombinant ade- igo and staphylococcal scalded-skin syndrome targets desmoglein 1. Nat Med noviruses using adenovirus DNA-terminal protein complex and a cosmid 6:1275^1277, 2001 bearing the full-length virus genome. Proc Natl Acad Sci USA 93:1320^1324, Amagai M,Yamaguchi T, HanakawaY, Nishifuji K, Sugai M, Stanley JR: Staphylo- 1996 coccal exfoliative toxin B speci¢cally cleaves desmoglein 1. J Invest Dermatol Niwa H, Yamamura K, Miyazaki J: E⁄cient selection for high-expression transfec- 118:845^850, 2002 tants with a novel eukaryotic vector. Gene 108:193^199, 1991 Andl CD, Stanley JR: Central role of the plakoglobin-binding domain for desmo- Norvell SM, Green KJ: Contributions of extracellular and intracellular domains of glein 3 incorporation into desmosomes. J Invest Dermatol 117:1068^1074, 2001 full length and chimeric cadherin molecules to junction assembly in epithelial Arnemann J, Sullivan KH, Magee AI, King IA, Buxton RS: Strati¢cation-related cells. JCellSci111:1305^1318, 1998 expression of isoforms of the desmosomal cadherins in human epidermis. Roth MB, Zahlerm AM, Stolkm JA: A conserved family of nuclear phosphoproteins JCellSci104:741^750, 1993 localized to sites of polymerase II transcription. JCellBiol115:587^596, 1991 Bannon LJ, Cabrera BL, Stack MS, Green KJ: Isoform-speci¢c di¡erences in the size Shimizu H, Masunaga T, Ishiko A, Kikuchi A, Hashimoto T, Nishikawa T: Pemphi- of desmosomal cadherin/catenin complexes. J Invest Dermatol 117:1302^1306, gus vulgaris and pemphigus foliaceus sera show an inversely graded binding 2001 pattern to extracellular regions of desmosomes in di¡erent layers of human Bornslaeger EA, Godsel LM, Corcoran CM, Park JK, Hatzfeld M, Kowalczyk AP, epidermis. J Invest Dermatol 105:153^159, 1995 Green KJ: Plakophilin 1 interferes with plakoglobin binding to desmoplakin, Shirakata Y, Amagai M, Hanakawa Y, Nishikawa T, Hashimoto K: Lack of mucosal yet together with plakoglobin promotes clustering of desmosomal plaque involvement in pemphigus foliaceus may be due to low expression of desmo- complexes at cell^cell borders. JCellSci114:727^738, 2001 glein 1. JInvestDermatol110:76^78, 1998 Elias PM, Matsuyoshi N,Wu H, Lin C,Wang ZH, Brown BE, Stanley JR: Desmo- Troyanovsky SM, Eshkind LG, Troyanovsky RB, Leube RE, Franke WW: Contri- glein isoform distribution a¡ects stratum corneum structure and function. butions of cytoplasmic domains of desmosomal cadherins to desmosome J Cell Biol 153:243^249, 2001 assembly and intermediate ¢lament anchorage. Cell 72:561^574, 1993 Green KJ, Gaudry CA: Are desmosomes more than tethers for intermediate ¢la- Udey MC, Stanley JR: Pemphigusdiseases of antidesmosomal autoimmunity. ments? NatRevMolCellBiol1:208^216, 2000 JAMA 282:572^576, 1999 Hanakawa Y, Amagai M, Shirakata Y, Sayama K, Hashimoto K: Di¡erent e¡ects of WuH,Wang ZH,Yan A,et al: Protection against pemphigus foliaceus by desmoglein dominant negative mutants of desmocollin and desmoglein on the cell^cell 3 in neonates. NEnglJMed343:31^35, 2000 adhesion of keratinocytes. JCellSci113:1803^1811, 2000 Zhu AJ, Watt FM: Expression of a dominant negative cadherin mutant inhibits Hatzfeld M, Kristjansson GI, Plessmann U,Weber K: Band 6 protein, a major con- proliferation and stimulates terminal di¡erentiation of human epidermal stituent of desmosomes from strati¢ed epithelia, is a novel member of the ar- keratinocytes. JCellSci109:3013^3023, 1996 madillo multigene family. JCellSci107:2259^2270, 1994