Portal de Periódicos da CAPES

Trangenic Misexpression of the Differentiation-Specific Desmocollin Isoform 1 in Basal Keratinocytes

2001; Elsevier BV; Volume: 116; Issue: 1 Linguagem: Inglês

10.1046/j.1523-1747.2001.00234.x

1523-1747

Frank Henkler, Molly Strom, Kathleen Mathers, Hayley C. Cordingley, Kate Sullivan, Ian King,

Polysaccharides and Plant Cell Walls

Keratinocytes undergoing terminal differentiation are characterized by well-defined changes in protein expression, which contribute towards the transformation of cytoarchitecture and epithelial morphology. Characteristic patterns of desmosomal cadherins are tightly regulated and distinct isoforms are expressed during development and differentiation of epithelial tissues. Desmocollin-1 is strictly confined to suprabasal layers of epidermis, but it is absent in mitotically active, basal keratinocytes. This raises the question of whether basal desmocollin-1 could alter desmosomal functions and compromise keratinocyte proliferation, stratification, or early differentiation in skin. In this study, we misexpressed human desmocollin-1 in mouse epidermis, under control of the keratin-14 promoter. Transgenic animals were generated, which showed a specific expression of transgenic human desmocollin-1 in epidermal basal cells. High level transgenic expression, which was equal to or greater than endogenous protein levels, was observed in mice with multiple copy integration of the transgene. A punctate distribution of desmocollin-1 was demonstrated at the cell membrane by indirect immunofluorescence. Transgenic human desmocollin-1 colocalized with endogenous desmosomal marker proteins, indicating efficient incorporation into desmosomes. Transgenic mice did not display any obvious abnormalities, either in the histology of skin and hair follicles, or in the ultrastructure of desmosomes. These observations suggest that desmocollin-1 can function as a desmosomal cadherin both in basal and suprabasal cells. We propose that the differentiation-specific desmocollin isoforms desmocollin-1 and desmocollin-3 are functionally equivalent in basal epidermal cells and suggest that their changing expression patterns are markers, but not regulators, of the initial steps in keratinocyte differentiation. Keratinocytes undergoing terminal differentiation are characterized by well-defined changes in protein expression, which contribute towards the transformation of cytoarchitecture and epithelial morphology. Characteristic patterns of desmosomal cadherins are tightly regulated and distinct isoforms are expressed during development and differentiation of epithelial tissues. Desmocollin-1 is strictly confined to suprabasal layers of epidermis, but it is absent in mitotically active, basal keratinocytes. This raises the question of whether basal desmocollin-1 could alter desmosomal functions and compromise keratinocyte proliferation, stratification, or early differentiation in skin. In this study, we misexpressed human desmocollin-1 in mouse epidermis, under control of the keratin-14 promoter. Transgenic animals were generated, which showed a specific expression of transgenic human desmocollin-1 in epidermal basal cells. High level transgenic expression, which was equal to or greater than endogenous protein levels, was observed in mice with multiple copy integration of the transgene. A punctate distribution of desmocollin-1 was demonstrated at the cell membrane by indirect immunofluorescence. Transgenic human desmocollin-1 colocalized with endogenous desmosomal marker proteins, indicating efficient incorporation into desmosomes. Transgenic mice did not display any obvious abnormalities, either in the histology of skin and hair follicles, or in the ultrastructure of desmosomes. These observations suggest that desmocollin-1 can function as a desmosomal cadherin both in basal and suprabasal cells. We propose that the differentiation-specific desmocollin isoforms desmocollin-1 and desmocollin-3 are functionally equivalent in basal epidermal cells and suggest that their changing expression patterns are markers, but not regulators, of the initial steps in keratinocyte differentiation. desmocollin Desmosomes play an important role in cell adhesion and maintenance of tissue structure in the epidermis. Previous studies have identified two structurally related but distinct families of desmosomal cadherins, named desmocollins (Dsc) and desmogleins, which both comprise three different isoforms (for review seeGarrod et al., 1999Garrod D.R. Tselepis C. Runswick S.K. North A.J. Wallis S.R. Chidgey M.A.J. Desmosomal adhesion.Adv Mol Cell Biol. 1999; 28: 165-202Crossref Scopus (15) Google Scholar). The genes of both families are closely linked and organized as a single cluster that is localized on chromosome 18 in both mouse and human. Due to alternative splicing, all three Dsc are expressed as two variants. Only the longer a-form binds plakoglobin, whereas the function of the shorter b-form is not yet understood. In epidermal tissue different isoforms of desmosomal cadherins are expressed during development and in adult differentiating skin. These complex expression patterns have been well described in several studies (Arnemann et al., 1993Arnemann J. Sullivan K.H. Magee A.I. King I.A. Buxton R.S. Stratification-related expression of isoforms of the desmosomal cadherins in human epidermis.J Cell Sci. 1993; 104: 741-750Crossref PubMed Google Scholar;Legan et al., 1994Legan P.K. Yue K.K.M. Chidgey M.A.J. Holton J.L. Wilkinson R.W. Garrod D.R. The bovine desmocollin family; a new gene and expression pattern reflecting epithelial cell proliferation.J Cell Biol. 1994; 126: 507-518Crossref PubMed Scopus (65) Google Scholar;King et al., 1995King I.A. Sullivan K.H. Bennet R. Buxton R.S. The desmocollins of human foreskin epidermis: identification and chromosomal assignment of a third gene and expression pattern of the three isoforms.J Invest Dermatol. 1995; 105: 314-321Crossref PubMed Scopus (69) Google Scholar;King et al., 1997King I.A. Angst B.D. Hunt D.M. Kruger M. Arnemann J. Buxton R.S. Hierarchical expression of desmosomal cadherins during stratified epithelial morphogenesis in mouse.Differentiation. 1997; 62: 83-96Crossref PubMed Google Scholar;Nuber et al., 1995Nuber U.A. Schäfer S. Schmidt A. Koch P.J. Franke W.W. The widespread desmocollin DSC2 and tissue specific patterns of synthesis of various desmocollin subtypes.Eur J Cell Biol. 1995; 66: 69-74PubMed Google Scholar;North et al., 1996North A.J. Chidgey M.A.J. Clarke J.P. Bardsley W.G. Garrod D.R. Distinct desmocollin isoforms occur in the same desmosomes and show reciprocally graded distributions in bovine nasal epidermis.Proc Natl Acad Sci USA. 1996; 93: 7701-7705https://doi.org/10.1073/pnas.93.15.7701Crossref PubMed Scopus (90) Google Scholar;Yue et al., 1995Yue K.K.M. Holton J.L. Clarke J.P. Hyam J.M.L. Hashimoto T. Chidgey M.A.J. Garrod D.R. Characterisation of a desmocollin isoform (bovine DSC3), exclusively expressed in lower layers of stratified epithelia.J Cell Sci. 1995; 108: 2163-2173PubMed Google Scholar;Schäfer et al., 1996Schäfer S. Stumpp S. Franke W.W. Immunological identification and characterisation of the desmosomal cadherin DSG2 in coupled and uncoupled epithelial cells and in human tissues.Differentiation. 1996; 60: 90-108Crossref Google Scholar). The biologic significance of the different isoforms and whether they have divergent properties or functions in desmosomal adhesion is not known. Dsc1 and Dsc3 are specifically expressed in stratified epithelia, and expression of Dsc1 is further restricted to the epidermis, tongue, and lymph node (King et al., 1996King I.A. O'brien T. Buxton R.S. Expression of the ‘‘skin-type’' desmosomal cadherin DSC1 is closely linked to the keratinisation of epithelial tissues during mouse development.J Invest Dermatol. 1996; 107: 531-538Crossref PubMed Scopus (35) Google Scholar;Nuber et al., 1996Nuber U.A. Schäfer S. Stehr S. Rackwitz H.R. Franke W.W. Patterns of desmocollin synthesis in human epithelia; immunolocalisation of desmocollins 1 and 3 in special epithelia and in cultured cells.Eur J Cell Biol. 1996; 71: 1-13PubMed Google Scholar). Dsc1 expression is most prominent in the higher, terminally differentiating cell layers (Legan et al., 1994Legan P.K. Yue K.K.M. Chidgey M.A.J. Holton J.L. Wilkinson R.W. Garrod D.R. The bovine desmocollin family; a new gene and expression pattern reflecting epithelial cell proliferation.J Cell Biol. 1994; 126: 507-518Crossref PubMed Scopus (65) Google Scholar). In contrast, high levels of Dsc3 are detectable in the basal and lower spinous layers. Although Dsc1 and Dsc3 can occur within the same desmosome, their overall distribution is reciprocally graded throughout the thickened epidermis of the bovine nose (North et al., 1996North A.J. Chidgey M.A.J. Clarke J.P. Bardsley W.G. Garrod D.R. Distinct desmocollin isoforms occur in the same desmosomes and show reciprocally graded distributions in bovine nasal epidermis.Proc Natl Acad Sci USA. 1996; 93: 7701-7705https://doi.org/10.1073/pnas.93.15.7701Crossref PubMed Scopus (90) Google Scholar). It was speculated that desmosomes, which incorporate variant isoforms, are distinguished by differential adhesiveness, and define a gradient that may be involved in keratinocyte positioning during differentiation or development (North et al., 1996North A.J. Chidgey M.A.J. Clarke J.P. Bardsley W.G. Garrod D.R. Distinct desmocollin isoforms occur in the same desmosomes and show reciprocally graded distributions in bovine nasal epidermis.Proc Natl Acad Sci USA. 1996; 93: 7701-7705https://doi.org/10.1073/pnas.93.15.7701Crossref PubMed Scopus (90) Google Scholar;Garrod et al., 1999Garrod D.R. Tselepis C. Runswick S.K. North A.J. Wallis S.R. Chidgey M.A.J. Desmosomal adhesion.Adv Mol Cell Biol. 1999; 28: 165-202Crossref Scopus (15) Google Scholar). This may further suggest a possible regulatory role for Dsc in tissue morphogenesis or cell migration (Garrod et al., 1996Garrod D. Chidgey M. North A. Desmosomes: differentiation, development, dynamics and diseases.Curr Opin Cell Biol. 1996; 8: 670-678Crossref PubMed Scopus (129) Google Scholar;King et al., 1996King I.A. O'brien T. Buxton R.S. Expression of the ‘‘skin-type’' desmosomal cadherin DSC1 is closely linked to the keratinisation of epithelial tissues during mouse development.J Invest Dermatol. 1996; 107: 531-538Crossref PubMed Scopus (35) Google Scholar). The importance of classical cadherins in this respect is well established (Takeichi, 1991Takeichi M. Cadherin cell adhesion receptors as morphogenic regulators.Science. 1991; 251: 1451-1455Crossref PubMed Scopus (2981) Google Scholar). Putative functions of their desmosomal counterparts as regulators in morphogenesis or differentiation have not yet been evaluated, however. The expression pattern of Dsc1 is particularly interesting, as it is exclusively expressed in suprabasal cells but not in the basal layer of dividing keratinocytes (Legan et al., 1994Legan P.K. Yue K.K.M. Chidgey M.A.J. Holton J.L. Wilkinson R.W. Garrod D.R. The bovine desmocollin family; a new gene and expression pattern reflecting epithelial cell proliferation.J Cell Biol. 1994; 126: 507-518Crossref PubMed Scopus (65) Google Scholar;Nuber et al., 1996Nuber U.A. Schäfer S. Stehr S. Rackwitz H.R. Franke W.W. Patterns of desmocollin synthesis in human epithelia; immunolocalisation of desmocollins 1 and 3 in special epithelia and in cultured cells.Eur J Cell Biol. 1996; 71: 1-13PubMed Google Scholar). Dsc1 expression was also demonstrated to correlate with the onset of keratinization (King et al., 1996King I.A. O'brien T. Buxton R.S. Expression of the ‘‘skin-type’' desmosomal cadherin DSC1 is closely linked to the keratinisation of epithelial tissues during mouse development.J Invest Dermatol. 1996; 107: 531-538Crossref PubMed Scopus (35) Google Scholar), suggesting further that this isoform is associated with keratinocytes that are committed towards terminal differentiation. This raises the question of whether Dsc1 expression in basal cells could alter desmosomal functions, such as their interactions with the intermediate filament network, and whether basal Dsc1 is compatible with keratinocyte proliferation or migration of stratifying cells into the spinous layer. This question also has wider implications. Desmosomal cell adhesion has been suggested as a potential inhibitor of tumor cell invasion (Tselepis et al., 1998Tselepis C. Chidgey M. North A. Garrod D. Desmosomal adhesion inhibits invasive behaviour.Proc Natl Acad Sci USA. 1998; 95: 8064-8069https://doi.org/10.1073/pnas.95.14.8064Crossref PubMed Scopus (154) Google Scholar), implying that desmosomal cadherins could constitute potential candidates for gene therapeutic intervention in skin-related cancers. Dsc1 could prove particularly useful in this regard, as it is strongly expressed in differentiated keratinocytes and is thought to increase desmosomal adhesion during differentiation. In addition, Dsc1 was shown to inhibit invasive behavior in L929 cells (Tselepis et al., 1998Tselepis C. Chidgey M. North A. Garrod D. Desmosomal adhesion inhibits invasive behaviour.Proc Natl Acad Sci USA. 1998; 95: 8064-8069https://doi.org/10.1073/pnas.95.14.8064Crossref PubMed Scopus (154) Google Scholar). In this study we have investigated the consequences of human Dsc1 expression in basal keratinocytes in vivo, using a novel transgenic mouse model. This approach has allowed a functional evaluation of Dsc1 expression in basal keratinocytes and an investigation of the possible consequences for epidermal keratinocyte differentiation and skin morphogenesis. The K14 expression cassette, pG3ZK14, was kindly provided by Elaine Fuchs (Wang et al., 1997Wang X. Zinkel S. Polonsky K. Fuchs E. Transgenic studies with a keratin promoter-driven growth hormone transgene: prospects for gene therapy.Proc Natl Acad Sci USA. 1997; 94: 219-226https://doi.org/10.1073/pnas.94.1.219Crossref PubMed Scopus (150) Google Scholar). The single EcoRI site at the 5′ side of the promoter (Figure 1a) was used to introduce additional XhoI and BglII sites, to allow the recovery of the K14-DSC1a fragment for pronuclear injection. A full-length human Dsc1a cDNA was obtained by linking the 3′ sequences of clone K1 to K55/24, using an internal ApaI site (King et al., 1993King I.A. Arnemann J. Spurr K.N. Buxton R.S. Cloning of the cDNA (DSC1) coding for human type 1 desmocollin and its assignment to chromosome 18.Genomics. 1993; 18: 185-194https://doi.org/10.1006/geno.1993.1453Crossref PubMed Scopus (46) Google Scholar). This fragment was subcloned into the BamH1 site of pG3ZK14, using the BamH1 site of pBluescript SK and an internal site within the 3′-untranslated sequence of the gene. A fragment containing the K14 promoter, β-globin intron, full-length human Dsc1 cDNA, and the polyadenylation site of the K14 gene was recovered by digestion with XhoI and HindIII (Figure 1a). This fragment was further purified, using the GENECLEAN II kit (Anachem) and Spin-X centrifuge tube filters (Costar). The DNA was dissolved in injection buffer [5 mM Tris pH 7.4; 0.15 mM ethylenediamine tetraacetic acid (EDTA)] and the final concentration was adjusted to 5 ng per μl. Isolation of fertilized mouse eggs and pronuclear injection of DNA fragments was carried out as described previously (Hogan et al., 1994Hogan B. Beddington R. Constantini F. Lacy E. Manipulating the Mouse Embryo. 2nd edn. Cold Spring Harbor Laboratory Press, 1994Google Scholar), using (CBA/Ca × C57BL/10)F1 mice. M2 and M16 media for embryo culture were obtained from Sigma. Embryos were reimplanted into pseudopregnant recipients, either after microinjection or at the two cell stage, as described previously (Hogan et al., 1994Hogan B. Beddington R. Constantini F. Lacy E. Manipulating the Mouse Embryo. 2nd edn. Cold Spring Harbor Laboratory Press, 1994Google Scholar). Genomic DNA was isolated from tails, digested in 100 mM Tris pH 8.0, 5 mM EDTA, 200 mM NaCl, 0.2% sodium dodecyl sulfate (SDS), 100 μg per ml proteinase K at 55°C overnight, purified by phenol/chloroform extraction and dissolved in water. Samples of genomic mouse DNAs were analyzed for transgene integration by polymerase chain reaction (PCR), using a primer that binds within the β-globin intron (5′-ACCATGTTCATGCCTTCTTC-3′) and a Dsc1-specific primer (5′-GTGTAATTCTGTGCAGCA-3′). In addition, DNA slot blot hybridization analysis was carried out, using a labeled fragment (XbaI/HindIII) containing the 3′-untranslated region of Dsc1 and the K14 polyadenylation site as probe. Blots were analyzed by autoradiography and signals were quantified by densitometry, using a Melico Photolog Transmission densitometer. Transgenic Dsc1 was detected with the human specific monoclonal antibody DSC1-U100 (Progen) and with monoclonal antibody A4 (King et al., 1996King I.A. O'brien T. Buxton R.S. Expression of the ‘‘skin-type’' desmosomal cadherin DSC1 is closely linked to the keratinisation of epithelial tissues during mouse development.J Invest Dermatol. 1996; 107: 531-538Crossref PubMed Scopus (35) Google Scholar), which also detects endogenous rodent Dsc1. These antibodies were localized with fluorescein isothiocyanate (FITC) conjugated rabbit antimouse IgG (Dako). Polyclonal antiserum (antiPG7) against the N-terminus of plakoglobin (residues 5–304) was kindly provided by Tony Magee. The polyclonal antiserum against desmoplakin has been published before (Arnemann et al., 1993Arnemann J. Sullivan K.H. Magee A.I. King I.A. Buxton R.S. Stratification-related expression of isoforms of the desmosomal cadherins in human epidermis.J Cell Sci. 1993; 104: 741-750Crossref PubMed Google Scholar). The FITC conjugated donkey antimouse IgG and Texas red conjugated donkey antirabbit IgG used in double staining experiments were obtained from Jackson Immunoresearch Laboratories. Mouse skin was heated for 1 min at 65°C in phosphate-buffered saline (PBS). The epidermis was separated from the dermis using a fresh razorblade and lyzed in 60 mM Tris pH 6.8, 5% β-mercaptoethanol, 1% SDS, 10% glycerol. Western blot analysis was carried out as described before (King et al., 1995King I.A. Sullivan K.H. Bennet R. Buxton R.S. The desmocollins of human foreskin epidermis: identification and chromosomal assignment of a third gene and expression pattern of the three isoforms.J Invest Dermatol. 1995; 105: 314-321Crossref PubMed Scopus (69) Google Scholar), using 8% polyacrylamide gels. Cryostat sections embedded in OCT embedding medium were prepared from frozen tissue (5–8 μm), using a cryotome 620 (Shandon). Sections were blocked in 1% bovine serum albumin/PBS. Incubation with primary antibodies was carried out for 1 h. All monoclonal antibodies were directly applied as hybridoma supernatants; the rabbit polyclonal antisera were diluted 1:200. Secondary rabbit antimouse conjugates were diluted 1:50 and were applied for a further 45 min. Secondary donkey antibodies were diluted 1:20 and applied for 1 h. Sections were washed in PBS/0.1% Nonidet P-40 and mounted with Citifluor (Agar Scientific). Fluorescent specimens were analyzed using an Olympus microscope with CCD camera, which was operated by Delta Vision computer software (Applied Precision). Alternatively, double-stained specimens were examined using a Leica DM RXE confocal microscope. Cytoplasmic tails of human Dsc1b and Dsc1a were amplified from pG3ZK14-Dsc1a using an oligonucleotide corresponding to amino acids 591–597 (5′-GCCGAATTCCCAGAAGACATAGCCGAGC-3′) and an antisense oligonucleotide in the 3′-untranslated region (5′-GCCGGATCCGCAGATGCTGCTAACATTCTGC-3′). The amplified fragment was digested with EcoR1/BamH and cloned in frame to the Gal4 DNA binding domain in plasmid pGBD (James et al., 1996James P. Hallerday J. Craig E.A. Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast.Genetics. 1996; 144: 1425-1436Crossref PubMed Google Scholar), thus creating pAS-Dsc1a. Full-length human plakoglobin was cloned into pACTII (Clontech); the resulting plasmid was named pACT-PG. Yeast strain CG 1945 was transformed with combinations of pGBD and pGAD constructs and colonies were assayed for their ability to grow on 3-aminotriazole (Sigma) and for β-galactosidase activity, as described inDurfee et al., 1993Durfee T. Becherer K. Chen L.P. et al.The retinoblastoma protein associates with the protein phosphatase type 1 catalytic subunit.Genes Dev. 1993; 7: 666-669Crossref Scopus (1298) Google Scholar. The Dsc1a cytoplasmic tail was also used as bait to screen an HaCaT cDNA library (Clontech). cDNA inserts of positive clones were amplified by PCR, sequenced using the BigDye Seq sequencing kit (Perkin Elmer), and identified by sequence homology using the Basic logical alignment search tool (BLAST,Altschul et al., 1990Altschul S.F. Gish W. Miller W. Myers E.W. Lipman D.J. Basic local alignment search tool.J Mol Biol. 1990; 215: 402-410Crossref Scopus (70353) Google Scholar). The human Dsc1a gene was subcloned into an expression cassette that is regulated by the K14 promoter (Figure 1a), thus facilitating its specific expression in basal epidermal cells. Pronuclear injections were carried out and eight founder mice, which harboured copies of the integrated transgene, were identified by PCR and subsequently by slot blot DNA hybridization (Figure 1b). A further transgenic animal (indicated as + in Figure 1b) died postnatally at day 13. The copy numbers of integrated transgene were determined from the signal intensities and are indicated in brackets in Figure 1(b). Heterozygous and homozygous mice were bred from the founder with the highest copy number of integrated transgene (Figure 1b, mouse 6), and from two founder mice of intermediate copy number (mice 27 and 28). The expression of transgenic Dsc1a was analyzed by immunofluorescence staining of frozen tail sections. The human-specific monoclonal antibody DSC1-U100 was used, as this allowed the transgenic human and endogenous mouse Dsc1 proteins to be distinguished. Transgenic Dsc1a was detected in the heterozygous progeny of founder mice 6, 27, and 28 (Figure 2a, b), which carried the transgene, but not in wild-type littermates (Figure 2d). The expression of transgenic Dsc1a was patchy in the epidermis of animals derived from founders 27 and 28, and expression was absent in up to a third of basal cells (Figure 2b). In contrast, heterozygous carriers that were bred from the high copy founder (Figure 1b, mouse 6) showed strong and specific staining for human Dsc1a in 80%-100% of basal keratinocytes (Figure 2a). These mice were used in subsequent analysis of the effects of Dsc1 misexpression in basal epidermal cells. Transgenic Dsc1a expression was restricted to basal keratinocytes, hair follicle (outer root sheath), and sebaceous gland (Figure 2a, c), with very little transgenic Dsc1a detectable in suprabasal cells. To estimate the relative expression levels of transgenic Dsc1a, compared with the endogenous protein, tail sections of transgenic and wild-type animals were analyzed using the monoclonal antibody A4 (Figure 3). This antibody is specific for Dsc1, but recognizes both the human and murine protein. No Dsc1 was detected in basal keratinocytes in nontransgenic animals, using A4 or DSC1-U100. This observation confirms that endogenous Dsc1 is excluded from the basal layer in mouse skin. In contrast, transgenic animals showed strong staining for transgenic human Dsc1 in the basal layer, using both antibodies (Figure 3a, b). The expression of transgenic Dsc1a was comparable to endogenous suprabasal levels of mouse Dsc1, as indicated by a similar staining intensity, using the cross-reactive antibody A4 (compare Figure 3b with 3d). We conclude that the graded distribution of Dsc1 that was previously speculated to control keratinocyte positioning or to regulate epidermal differentiation (North et al., 1996North A.J. Chidgey M.A.J. Clarke J.P. Bardsley W.G. Garrod D.R. Distinct desmocollin isoforms occur in the same desmosomes and show reciprocally graded distributions in bovine nasal epidermis.Proc Natl Acad Sci USA. 1996; 93: 7701-7705https://doi.org/10.1073/pnas.93.15.7701Crossref PubMed Scopus (90) Google Scholar) had been eliminated in these mice. Expression of the transgenic Dsc1a protein was further analyzed by Western blotting, using the human-specific antibody DSC1-U100 as well as the cross-reactive monoclonal antibody A4. A single band, corresponding in size to mouse Dsc1a, was specifically detected by DSC1-U100 in transgenic mice. This result suggests that only full-length human Dsc1a is expressed within the basal layer (Figure 1c). To analyze whether transgenic Dsc1a was incorporated into desmosomes, its cellular localization was studied by confocal microscopy of cryostat sections that had been costained with the human Dsc1-specific monoclonal DSC1-U100 and with DP121, a polyclonal antiserum against desmoplakin (Figure 4). Examination at lower magnification again showed that the transgenic human Dsc1 was restricted to basal keratinocytes and that it had a generally similar distribution to endogenous desmoplakin (Figure 4a). Again, no transgenic Dsc1 was found within suprabasal layers. The staining was predominantly punctate, which would be consistent with incorporation of exogenous Dsc1 into the desmosomes of basal cells. Analysis at higher magnification, however, showed that the Dsc1 and desmoplakin staining were not completely identical and did not overlap entirely (Figure 4b). This partial overlap may reflect the very different location of the two proteins within the structure of the desmosome and also the high resolution of confocal microscopy. This interpretation is supported by the observation of comparable, partially overlapping patterns for endogenous Dsc3 and desmoplakin (Figure 4c). The distribution of transgenic Dsc1a and endogenous plakoglobin was also analyzed in transgenic mice (Figure 4d), using DSC1-U100 and the rabbit polyclonal antiserum antiPG7. Dsc1 and plakoglobin were again detected in partially overlapping, punctate patterns at the surface of basal keratinocytes similar to the colocalization found for Dsc1 and desmoplakin. These data are consistent with the incorporation of transgenic Dsc1a into desmosomes in basal epidermal keratinocytes. All the transgenic animals examined appeared to be normal. No gross pathologic or morphologic abnormalities were observed either in founder mice or in heterozygous or homozygous transgenic progeny. This observation suggests that expression of Dsc1a within the basal layer is compatible with normal stratification, development of epidermis, and differentiation. Homozygous transgenic animals, aged 17 and 34 d, were analyzed for possible histologic and morphologic alterations. Sections of tail, back skin, footpad, and tongue were examined by conventional hematoxylin and eosin staining and compared with tissues from nontransgenic littermates. No obvious abnormalities were found in any of these tissues (data not shown), suggesting that expression of Dsc1a in basal cells does not interfere with epidermal development or differentiation in adult skin or tongue. Although transgenic Dsc1a was strongly expressed in the outer root sheath, there was no evidence for morphologic changes or abnormalities in either hair follicle or hair. Ultrastructural analysis of tail, back skin, tongue, and footpad from transgenic mice and noncarrier littermates was carried out by conventional electron microscopy. The ultrastructure of desmosomes appeared to be normal in tissues from 17-d-old and 34-d-old transgenic mice and no significant variations were found in desmosome numbers (data not shown). Further, no alterations were apparent in the density and distribution of cytokeratin filaments or in their associations with desmosomes. Although transgenic Dsc1a was strongly detectable by immunofluorescence staining, attempts to localize the protein by immunogold labeling were unsuccessful as the epitopes recognized by DSC1-U100 and A4 did not survive fixation. These analyses suggest that transgenic animals do not display abnormalities in morphology, histology, or ultrastructure that could be attributed to overexpression of Dsc1a in basal epidermal keratinocytes. Misexpression of transgenic Dsc1a suggested that this isoform was able to function as a desmosomal cadherin in basal as well as suprabasal epidermal cells. In stratified epithelia, however, individual Dsc could still play distinct roles in fine-tuning desmosome functions or in the recruitment of specific plaque proteins. To address the possible functional diversity of Dsc isoforms, their cytoplasmic domains were analyzed for differential interactions with cellular proteins. We have previously studied putative associations of Dsc3 and Dsc2 with cellular proteins in cultured human foreskin keratinocytes using immune-precipitation analysis (data not shown). In these experiments, associations of both Dsc2 and Dsc3 with plakoglobin were demonstrated. As Dsc1 is not expressed in cultured keratinocytes, the question of differential cytoplasmic associations of this isoform was analyzed using the yeast two-hybrid system. The interaction of the cytoplasmic domain of Dsc1a with plakoglobin was directly demonstrated by assaying yeast that harboured both the Dsc1a-Gal4 bait and pACT-PG for activation of β-galactosidase expression and for their ability to grow on 3-aminotriazole (Table 1). We were unable to examine cytoplasmic interactions of the Dsc3a tail, because the Gal4-binding-domain-Dsc3a fusion protein induced high background expression of both His3 and β-galactosidase. Interactions of the Dsc1a tail were further studied by yeast two-hybrid screening of an HaCaT cDNA library. Sixty to seventy percent of the positive colonies contained full-length plakoglobin, fused in frame to the Gal4 activator domain, confirming that the primary cytoplasmic interaction of Dsc1a was with plakoglobin. No other multiple hits were identified from the remaining colonies. This experiment suggested that Dsc1 interacts primarily with plakoglobin and is equivalent in this respect to the other Dsc isoforms.Table 1Yeast two-hybrid analysis of cytoplasmic interactions of Dsc1aaYeast strain CG 1945 harbouring the indicated plasmids were streaked onto nitrocellulose membranes, grown at 30°C for 2 d, and then assayed for β-galactosidase expression. Assays were allowed to develop for 30 min. They were also tested for their ability to grow on 3-aminotriazole. The Dsc1a-Gal4 fusion protein bait interacted strongly with plakoglobin as indicated by intense blue staining of growing yeast and growth in the presence of high concentrations of 3-aminotriazole. No or very little β-galactosidase expression was detected in the control experiments.β-galactosidase activity[3-aminotriazole]pGBD – DSC1/pACTII– 100 mMpGBD/pACTII – PG–<10 mMa Yeast strain CG 1945 harbouring the indicated plasmids were streaked onto nitrocellulose membranes, grown at 30°C for 2 d, and then assayed for β-galactosidase expression. Assays were allowed to develop for 30 min. They were also tested for their ability to grow on 3-aminotriazole. The Dsc1a-Gal4 fusion protein bait interacted strongly with plakoglobin as indicated by intense blue staining of growing yeast and growth in the presence of high concentrations of 3-aminotriazole. No or very little β-galactosidase expression was detected in the control experiments. Open table in a new tab In this study, the functional significance of differentiation-specific isoforms of Dsc was addressed, using an in vivo transgenic mouse model. Mice that showed strong expression of transgenic Dsc1a in their basal epidermal keratinocytes were generated and analyzed for possible alterations in desmosomal ultrastructure and epidermal tissue abnormalities. No obvious anomalies were found, however, suggesting that basal expression of Dsc1 is compatible with normal epidermal differentiation and development. These observations were somewhat surprising in view of previous transgenic misexpression experiments, which demonstrated the functional diversity of differentiation-specific cytokeratins in the epidermis. Misexpression of K16 in basal epidermal cells caused a dramatic phenotype that involved hyperproliferation of basal keratinocytes and hyperkeratotic skin (Paladini and Coulombe, 1998Paladini R.D. Coulombe P.A. Directed expression of keratin 16 to the progenitor basal cells of transgenic mouse skin delays skin maturation.J Cell Biol. 1998; 142: 1035-1051Crossref PubMed Scopus (69) Google Scholar). Furthermore, K18, a cytokeratin of simple epithelia, did not rescue the mutant phenotype when it was expressed in the basal layer of K14 knockout mice (Hutton et al., 1998Hutton E. Paladini R.D. Yu Q.C. Yen M. Coulombe P.A. Fuchs E. Functional differences between keratins of stratified and simple epithelia.J Cell Biol. 1998; 143: 487-499Crossref PubMed Scopus (90) Google Scholar). In contrast, we conclude here that the exclusion of Dsc1 from basal keratinocytes has little significance for desmosome function or for specific desmosomal interactions with the basal K5–K14 cytokeratin network. Transgenic Dsc1a was detected as punctate staining that partially colocalized with desmoplakin and plakoglobin. This observation indicated that some of the exogenous basal Dsc1a was associated with desmosomal marker proteins and thus appeared to be incorporated into desmosomes. The cytoplasmic interactions of Dsc1 were further addressed using the yeast two-hybrid system, as immune-precipitation of Dsc1 was not feasible from cultured keratinocytes. These experiments suggested that Dsc1 interacted predominantly with plakoglobin. We speculate that all Dsc are equivalent in this respect and suggest further that Dsc1 functions normally as a desmosomal cadherin when it is misexpressed in basal epidermal keratinocytes. The significance of the different Dsc isoforms is not known. We have demonstrated here that Dsc1 is apparently not involved in the regulation of stratification or early events of epidermal differentiation. It remains unclear, however, whether its upregulation in the suprabasal layers of normal epidermis reflects functional differences. In particular, we cannot exclude the possibility that suprabasal Dsc1 engages in interactions that, though not relevant in basal keratinocytes, may still be important for tissue integrity or desmosomal adhesion in the spinous or granular layer. Associations between Dsc1a and plakophilin-1 have been previously demonstrated by in vitro overlay assays (Smith and Fuchs, 1998Smith E.A. Fuchs E. Defining the interactions between intermediate filaments and desmosomes.J Cell Biol. 1998; 141: 1229-1241Crossref PubMed Scopus (203) Google Scholar). Consistent with these observations, we found weak associations between Dsc1a and plakophilin-1, using the yeast two-hybrid system (Molly Strom, unpublished observations). Plakophilins are nuclear proteins that are recruited to the desmosomal plaque in some tissues. Interestingly, like Dsc1, plakophilin-1 expression is restricted to stratified epithelia and its desmosomal expression is most prominent in the suprabasal layers (Heid et al., 1994Heid H.W. Schmidt A. Zimbelman R. et al.Cell-type specific desmosomal plaque proteins of the plakoglobin family. plakophilin (band 6 protein).Differentiation. 1994; 58: 113-131Crossref PubMed Scopus (158) Google Scholar;Mertens et al., 1996Mertens C. Kahn C. Franke W.W. Plakophilin 2a & 2b: constitutive proteins of dual localisation in the karyoplasm and the desmosomal plaque.J Cell Biol. 1996; 135: 1009-1025Crossref PubMed Scopus (239) Google Scholar;Schmidt et al., 1997Schmidt A. Langbein L. Rode M. Präzel S. Zimbelmann R. Franke W.W. Plakophilins 1a and 1b: widespread nuclear proteins recruited in specific epithelial cells as desmosomal plaque components.Cell Tissue Res. 1997; 290: 481-499https://doi.org/10.1007/s004410050956Crossref PubMed Scopus (149) Google Scholar). Furthermore, novel plaque proteins have been recently identified that are expressed both in tissue- and differentiation-dependent manners (Ruhrberg et al., 1996Ruhrberg C. Hajibagheri M.A. Simon M. Dooley T.P. Watt F.M. Envoplakin, a novel precursor of the cornified envelope that has homology to desmoplakin.J Cell Biol. 1996; 134: 715-729Crossref PubMed Scopus (152) Google Scholar; 1997). It remains to be clarified whether diverse desmosomal cadherins differentially associate with cytoplasmic or plaque proteins and whether this leads to modifications of desmosomal adhesion in suprabasal cells. We are very grateful to Dimitris Kioussis for teaching Frank Henkler pronuclear injections and transgenic technology, as well as to Elaine Fuchs for her generous gift of the K14-expression cassette. We thank Donald Bell and Brigitt Angst for valuable technical advice and support and we are grateful to Tony Magee for many helpful discussions. We are further deeply indebted to Sarah, Sandra, Dave, and Pete Dawson from Dunkin Green for their care and support and we thank Joe Brock for his help with the figures. This work was supported by the U.K. Medical Research Council.