Advertisement

Plakoglobin as a Regulator of Desmocollin Gene Expression

      Desmosomes are cell adhesion junctions required for the normal development and maintenance of mammalian tissues and organs such as the skin, skin appendages, and the heart. The goal of this study was to investigate how desmocollins (DSCs), transmembrane components of desmosomes, are regulated at the transcriptional level. We hypothesized that differential expression of the Dsc2 and Dsc3 genes is a prerequisite for normal development of skin appendages. We demonstrate that plakoglobin (Pg) in conjunction with lymphoid enhancer-binding factor 1 (Lef-1) differentially regulates the proximal promoters of these two genes. Specifically, we found that Lef-1 acts as a switch activating Dsc2 and repressing Dsc3 in the presence of Pg. Interestingly, we also determined that NF-κB pathway components, the downstream effectors of the ectodysplasin-A (EDA)/ ectodysplasin-A receptor (EDAR)/NF-κB signaling cascade, can activate Dsc2 expression. We hypothesize that Lef-1 and EDA/EDAR/NF-κB signaling contribute to a shift in Dsc isoform expression from Dsc3 to Dsc2 in placode keratinocytes. It is tempting to speculate that this shift is required for the invasive growth of placode keratinocytes into the dermis, a crucial step in skin appendage formation.

      Abbreviations

      ChIP
      chromatin immunoprecipitation
      DSC
      desmocollin
      DSG
      desmoglein
      HF
      hair follicle
      Lef-1
      lymphoid enhancer-binding factor 1
      MDCK
      Madin Darby canine kidney
      Pg
      plakoglobin
      TCF
      T-cell factor

      INTRODUCTION

      Desmosomes are cell adhesion complexes that are assembled at the plasma membrane where they serve as membrane anchors for intermediate filament proteins (
      • Cheng X.
      • Koch P.J.
      In vivo function of desmosomes.
      ;
      • Cheng X.
      • Den Z.
      • Koch P.J.
      Desmosomal cell adhesion in mammalian development.
      ;
      • Dubash A.D.
      • Green K.J.
      Desmosomes.
      ). The importance of this cell junction for organ stability and function is demonstrated by the severe acquired and inherited diseases of the skin, skin appendages (e.g., hair), and the heart that result from impaired desmosome function (see, e.g.,
      • Amagai M.
      • Stanley J.R.
      Desmoglein as a target in skin disease and beyond.
      ;
      • Petrof G.
      • Mellerio J.E.
      • McGrath J.A.
      Desmosomal genodermatoses.
      ;
      • Swope D.
      • Li J.
      • Radice G.L.
      Beyond cell adhesion: the role of armadillo proteins in the heart.
      ). These observations in human patients are further supported by the severe skin and heart phenotypes of mice with mutations in genes encoding desmosomal components (see, e.g.,
      • Koch P.J.
      • Mahoney M.G.
      • 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.
      ,
      • Koch P.J.
      • Mahoney M.G.
      • Cotsarelis G.
      • et al.
      Desmoglein 3 anchors telogen hair in the follicle.
      ;
      • Chen J.
      • Den Z.
      • Koch P.J.
      Loss of desmocollin 3 in mice leads to epidermal blistering.
      ;
      • Ganeshan R.
      • Chen J.
      • Koch P.J.
      Mouse models for blistering skin disorders.
      ;
      • Li J.
      • Swope D.
      • Raess N.
      • et al.
      Cardiac tissue-restricted deletion of plakoglobin results in progressive cardiomyopathy and activation of {beta}-catenin signaling.
      ,
      • Li D.
      • Zhang W.
      • Liu Y.
      • et al.
      Lack of plakoglobin in epidermis leads to keratoderma.
      ). Desmosomes also contribute to cell sorting, morphogenetic cell movement, and the formation of the proper histoarchitecture during embryonic development (
      • Cheng X.
      • Koch P.J.
      In vivo function of desmosomes.
      ).
      Desmosomes contain transmembrane adhesion molecules (desmosomal cadherins, desmogleins (DSGs) and desmocollins (DSCs)) and associated plaque proteins (reviewed in
      • Cheng X.
      • Den Z.
      • Koch P.J.
      Desmosomal cell adhesion in mammalian development.
      ;
      • Dubash A.D.
      • Green K.J.
      Desmosomes.
      ). The main plaque proteins belong either to the armadillo family of structural and signaling proteins (plakoglobin (Pg) and plakophilins) or the plakin family (desmoplakin). The mouse genome encodes three Dsc and six Dsg genes. All DSG and DSC proteins are synthesized in the epidermis of the skin and in skin appendages: e.g., Dsc2 and Dsc3 are present in the basal and immediate suprabasal layers, whereas Dsc1 is mainly restricted to the granular layer of the mouse epidermis (references in
      • Chen J.
      • Den Z.
      • Koch P.J.
      Loss of desmocollin 3 in mice leads to epidermal blistering.
      ). It is thought that the specific complement of DSG and DSC proteins affects the adhesive and potentially the signaling properties of desmosomes (see, e.g.,
      • Schmidt A.
      • Koch P.J.
      Desmosomes: just cell adhesion or is there more?.
      ;
      • Muller E.J.
      • Williamson L.
      • Kolly C.
      • et al.
      Outside-in signaling through integrins and cadherins: a central mechanism to control epidermal growth and differentiation?.
      ). However, little is known about the gene-regulatory mechanisms that control the expression of these desmosomal genes (see, e.g.,
      • Oshiro M.M.
      • Watts G.S.
      • Wozniak R.J.
      • et al.
      Mutant p53 and aberrant cytosine methylation cooperate to silence gene expression.
      ;
      • Smith C.
      • Zhu K.
      • Merritt A.
      • et al.
      Regulation of desmocollin gene expression in the epidermis: CCAAT/enhancer-binding proteins modulate early and late events in keratinocyte differentiation.
      ;
      • Oshiro M.M.
      • Kim C.J.
      • Wozniak R.J.
      • et al.
      Epigenetic silencing of DSC3 is a common event in human breast cancer.
      ). In this study, we focused on the regulation of the Dsc2 and Dsc3 genes; their coexpression in the deep layers of the mouse epidermis suggests that they may have a role in keratinocyte differentiation and skin appendage formation (see, e.g.,
      • Chidgey M.A.
      • Yue K.K.
      • Gould S.
      • et al.
      Changing pattern of desmocollin 3 expression accompanies epidermal organisation during skin development.
      ;
      • Cheng X.
      • Koch P.J.
      In vivo function of desmosomes.
      ). Furthermore, we questioned whether signaling cascades known to have important roles during skin appendage formation affect the expression of Dsc2 and Dsc3, respectively.
      Both the WNT and the ectodysplasin-A (EDA)/ ectodysplasin-A receptor (EDAR)/NF-κB pathways have been shown to have critical roles in hair follicle (HF) formation (
      • Schmidt-Ullrich R.
      • Aebischer T.
      • Hulsken J.
      • et al.
      Requirement of NF-kappaB/Rel for the development of hair follicles and other epidermal appendices.
      ;
      • Zhang Y.
      • Tomann P.
      • Andl T.
      • et al.
      Reciprocal requirements for EDA/EDAR/NF-kappaB and Wnt/beta-catenin signaling pathways in hair follicle induction.
      ). The canonical Wnt cascade ultimately affects target gene expression via β-catenin/T-cell factor (TCF)/lymphoid enhancer-binding factor (Lef) complexes. Pg, which is sequence related to β-catenin, has also been shown to act as a transcription factor, both in conjunction and independent of TCF/Lef factors (see e.g.,
      • Simcha I.
      • Shtutman M.
      • Salomon D.
      • et al.
      Differential nuclear translocation and transactivation potential of beta-catenin and plakoglobin.
      ;
      • Zhurinsky J.
      • Shtutman M.
      • Ben-Ze’ev A.
      Differential mechanisms of LEF/TCF family-dependent transcriptional activation by beta-catenin and plakoglobin.
      ;
      • Shtutman M.
      • Zhurinsky J.
      • Oren M.
      • et al.
      PML is a target gene of beta-catenin and plakoglobin, and coactivates beta-catenin-mediated transcription.
      ;
      • Maeda O.
      • Usami N.
      • Kondo M.
      • et al.
      Plakoglobin (gamma-catenin) has TCF/LEF family-dependent transcriptional activity in beta-catenin-deficient cell line.
      ;
      • Teuliere J.
      • Faraldo M.M.
      • Shtutman M.
      • et al.
      beta-Catenin-dependent and -independent effects of DeltaN-plakoglobin on epidermal growth and differentiation.
      ;
      • Williamson L.
      • Raess N.A.
      • Caldelari R.
      • et al.
      Pemphigus vulgaris identifies plakoglobin as key suppressor of c-Myc in the skin.
      ;
      • Galichet A.
      • Strauss C.
      • Sayar B.S.
      • et al.
      Desmosomal cadherins crosstalk to the nucleus via a novel potent transcriptional regulator of the Wnt pathway: Plakoglobin (PG, JUP, γ-catenin).
      ). Similar to β-catenin, Pg does not possess a DNA-binding domain, i.e., it requires cofactors (such as TCF/Lef factors) to bind to promoter sequences.
      It is well established that Pg, independent of β-catenin signaling, is a key regulator of various cellular processes such as migration (
      • Rieger-Christ K.M.
      • Ng L.
      • Hanley R.S.
      • et al.
      Restoration of plakoglobin expression in bladder carcinoma cell lines suppresses cell migration and tumorigenic potential.
      ;
      • Yin T.
      • Getsios S.
      • Caldelari R.
      • et al.
      Plakoglobin suppresses keratinocyte motility through both cell-cell adhesion-dependent and -independent mechanisms.
      ;
      • Franzen C.A.
      • Todorovic V.
      • Desai B.V.
      • et al.
      The desmosomal armadillo protein plakoglobin regulates prostate cancer cell adhesion and motility through vitronectin-dependent Src signaling.
      ), proliferation (
      • Li D.
      • Zhang W.
      • Liu Y.
      • et al.
      Lack of plakoglobin in epidermis leads to keratoderma.
      ), apoptosis (
      • Dusek R.L.
      • Godsel L.M.
      • Chen F.
      • et al.
      Plakoglobin deficiency protects keratinocytes from apoptosis.
      ), and gene expression (see above).
      Here we demonstrate that Pg, in conjunction with Lef-1, can control Dsc2 and Dsc3 gene expression: in the presence of Lef-1, Pg activates Dsc2 and represses Dsc3 expression. Furthermore, we show that NF-κB proteins, the downstream effectors of EDA/EDAR signaling (
      • Schmidt-Ullrich R.
      • Aebischer T.
      • Hulsken J.
      • et al.
      Requirement of NF-kappaB/Rel for the development of hair follicles and other epidermal appendices.
      ), activate Dsc2 as well. Both the EDA/EDAR/NF-κB and the TCF/Lef signaling have crucial roles in the early steps of HF formation. Our results predict that these two pathways shift expression from Dsc3 to Dsc2 in placode keratinocytes invading the dermis, a process that might be required for effective keratinocyte migration and thus appendage formation. To the best of our knowledge, these results provide previously unreported evidence that the catenin/TCF/Lef and NF-κB signaling cascades control Dsc gene expression in keratinocytes.

      RESULTS

      Dsc gene regulation in skin appendage formation

      It is thought that changing the molecular composition of desmosomes is one mechanism used by keratinocytes to adapt their cell adhesion system to their specific environment. For example, in the early stages of skin appendage formation, desmocollins are downregulated in placode keratinocytes invading the dermis (
      • Nanba D.
      • Hieda Y.
      • Nakanishi Y.
      Remodeling of desmosomal and hemidesmosomal adhesion systems during early morphogenesis of mouse pelage hair follicles.
      ,
      • Nanba D.
      • Nakanishi Y.
      • Hieda Y.
      Changes in adhesive properties of epithelial cells during early morphogenesis of the mammary gland.
      ). However, little is known regarding gene-regulatory mechanisms that control Dsc gene expression in skin. To begin to identify gene-regulatory pathways controlling Dsc gene expression, we cloned and sequenced the proximal promoters of mouse, Dsc2 and Dsc3. These two Dsc genes are expressed in the basal layers of the interfollicular epidermis, the cellular compartment that maintains the epidermis and that forms HFs.
      DNA sequence analysis revealed the presence of putative TCF/Lef (Figure 1a and b) and NF-κB (see below) target sites in the Dsc2 and Dsc3 promoters. Considering that the Wnt pathway, which affects gene expression via catenin/TCF/Lef transcription factors, and the NF-κB pathways have been shown to have major roles in HF formation, we decided to focus on the role of these two signaling cascades in Dsc gene regulation.
      Figure thumbnail gr1
      Figure 1Identification and functional characterization of T-cell factor/lymphoid enhancer-binding factor (TCF/Lef)–binding sites in the desmocollin 2 (Dsc2) and desmocollin 3 (Dsc3) promoters. (a, b) Schematic representation of putative transcription factor–binding sites (TCF/Lef) in the proximal Dsc promoters. The arrows indicate translation start sites (ATG; A is defined as position +1). The DNA sequences of wild-type (Wt) and mutant (Mut) transcription factor–binding sites are shown. (c, d) Chromatin immunoprecipitation (ChIP) assays demonstrating TCF/Lef binding to the predicted target sites in the Dsc promoters. Note that the point mutations introduced into the TCF/Lef target sequences (Dsc2 Mut, Dsc3 Mut) abrogate binding of the transcription factors. Input, chromatin used for immunoprecipitation; IgG, immunoprecipitation (IP) with unspecific IgG. (e, f) Reporter assays in mouse keratinocytes. Note that inactivation of the TCF/Lef-binding sites in the Dsc3 construct increases reporter activity significantly. Error bars, SD. Star indicates a statistically significant result (P<0.05).

      TCF/Lef regulation of the Dsc2 and Dsc3 promoters

      We cloned 2kb of the genomic DNA sequences immediately upstream of the Dsc2 and Dsc3 translation start codons (Figure 1a and b). These DNA fragments were cloned into a promoter-less Luciferase reporter vector designed to assess transcriptional activity of promoter fragments (reporter assays). Both the Dsc2 and Dsc3 constructs showed reporter activity in primary mouse keratinocytes (data not shown). Next, we introduced point mutations abrogating DNA binding of TCF/Lef factors to their target sites into both promoters (Figure 1a and b). The wild-type, but not the mutant TCF/Lef sites, showed binding of Lef-1 and TCF-4 as demonstrated by chromatin immunoprecipitation (ChIP) assays (Figure 1c and d). TCF3 behaved identical to TCF-4 in all assays performed in this study. Consequently, only the Lef-1 and TCF-4 data are shown. All ChIP assays and biochemical experiments were conducted in canine MDCK (Madin Darby canine kidney) cells, whereas reporter assays and gene expression studies were conducted in primary mouse keratinocytes. MDCK cells express Dsc2 and Dsc3 (data not shown). Owing to differences in the promoter DNA sequences of dog and mouse and the use of species-specific antibodies, we were able to investigate protein–DNA and protein–protein interactions in MDCK cells without the interference of endogenous canine genes and proteins.
      Next, we assessed whether loss of functional TCF/Lef sites affected reporter activity in primary keratinocytes. As shown in Figure 1, Dsc2 promoter activity was unchanged when wild-type and mutant reporter constructs lacking a functional TCF/Lef site were compared (Figure 1e). Dsc3 promoter activity was significantly increased in the mutant construct, indicating that this site is repressive (Figure 1f).

      β-Catenin expression has no effect on the proximal Dsc2 and Dsc3 promoters

      TCF/Lef sites are the primary target sites of β-catenin/TCF/Lef transcription factor complexes, the principal effectors of the Wnt signaling pathway. ChIP assays demonstrated that β-catenin can bind to the Dsc3 but not the Dsc2 promoter (Supplementary Figure S1 online; these experiments were conducted in MDCK cells expressing TCF-4 as a cofactor, data not shown). To determine whether β-catenin can influence Dsc promoter activity, we transfected the Dsc reporter constructs with different combinations of TCF/Lef factors into keratinocytes. We did not observe a statistically significant effect of these β-catenin/TCF/Lef factors on Dsc2 or Dsc3 reporter activity (Supplementary Figure S1 online). A slight repression of Dsc3 reporter activity by β-catenin and Lef-1 was often observed but did not reach statistical significance.

      Pg is a key regulator of Dsc2 and Dsc3 promoter activity

      Pg has been shown to regulate gene expression in keratinocytes and other epithelial cell types (see references in the Introduction section). To determine whether Pg can affect Dsc promoter activity, we first assessed the ability of this protein to bind the Dsc2 and Dsc3 promoter fragments. As shown in Figure 2a and b, Pg binds to the Dsc2 promoter but not the Dsc3 promoter in the presence of Lef-1. Reporter assays demonstrated that Pg can affect both promoters, although its effects are dependent on Lef-1 (Figure 2c and d); in the presence of Lef-1, Pg activates the Dsc2 promoter, whereas Dsc3 promoter activation occurs in the absence of ectopic TCF/Lef expression. These results suggest that Lef-1 can act as a switch that shifts activity from the Dsc3 to the Dsc2 promoter in the presence of Pg.
      Figure thumbnail gr2
      Figure 2Effects of plakoglobin (Pg) on desmocollin (Dsc) reporter activities. (a, b) Chromatin immunoprecipitation (ChIP) assays demonstrating direct binding of Pg to the Dsc2 promoter in the presence of lymphoid enhancer-binding factor 1 (Lef-1). Note that Pg does not bind the Dsc3 promoter in the presence of Lef-1 (b) or T-cell factor 4 (TCF-4; data not shown). Input, chromatin used for immunoprecipitation (IP); IgG, IP with unspecific IgG. (cf) Reporter assays in keratinocytes. (c) Coexpression of Pg and Lef-1 markedly increases Dsc2 reporter activity, whereas all other combinations of Pg and TCF/Lef factors have no effect. (d) The Dsc3 promoter is activated in the presence of Pg. This activation is reversed in cells coexpressing Pg and TCF-4 or Lef-1. (e) Dsc2 promoter activation by Pg and Lef-1 is dependent on the presence of a functional TCF/Lef-binding site. Mut, mutant. (f) Pg overexpression and loss of a functional TCF/Lef–binding site increase Dsc3 reporter activity to a similar extent. Note that expression is normalized to the baseline expression of the WT promoter (set to 1). (g, h) Endogenous Dsc gene expression in keratinocytes transfected with different combinations of Pg and TCF/Lef factors. Error bars, SD. Stars indicate statistically significant results (P<0.05).
      Neither Lef-1, TCF-3, nor TCF-4 had any effect on the Dsc2 or Dsc3 promoter in single transfection experiments (data not shown). These results are expected given that keratinocytes express endogenous TCF/Lef factors, but show very low cytoplasmic and nuclear Pg levels (
      • Williamson L.
      • Raess N.A.
      • Caldelari R.
      • et al.
      Pemphigus vulgaris identifies plakoglobin as key suppressor of c-Myc in the skin.
      ), conditions that favor low Dsc2 and Dsc3 reporter activity.
      As shown in Figure 2e, Dsc2 promoter activation by Pg and Lef-1 was dependent on the presence of a functional TCF/Lef-binding site. In the case of the Dsc3 promoter, we found no additive or synergistic effects of Pg and the TCF/Lef mutation (Figure 2f). These findings are consistent with the hypothesis that Pg asserts its activating effect on Dsc2 by binding to the TCF/Lef site, whereas this was not the case for the Dsc3 promoter.
      We next assessed the effects of Pg and Lef-1 coexpression on the endogenous Dsc2 and Dsc3 gene activity in primary keratinocytes by quantitative reverse transcriptase–PCR. As shown in Figure 2g and h, both genes responded to ectopic Pg/Lef expression as predicted by our reporter assays, confirming the validity of our conclusions for the regulation of these genes in keratinocytes.
      To gain further insights into the mechanisms by which Pg and Lef-1 control Dsc gene expression, we assessed whether Pg can bind to Lef-1. As shown in Figure 3a, Lef-1 and Pg can form a complex as shown by coimmunoprecipitation experiments using Lef-1 antibodies. Most interestingly, ChIP competition experiments suggested that the interaction of Pg and Lef-1 interferes with the ability of Lef-1 to bind to the Dsc3 promoter (Figure 3b). Considering that TCF/Lef-1 complexes can act as transcriptional repressors (see, e.g.,
      • Hoverter N.P.
      • Waterman M.L.
      A Wnt-fall for gene regulation: repression.
      ), these results raise the possibility that Pg could activate Dsc3 expression by interfering with the binding of a TCF/Lef repressor complex to the Dsc3 promoter. Further support for this hypothesis is provided by western blot experiments (Figure 3c and d), demonstrating that ectopically expressed Pg can interfere with the nuclear accumulation of Lef-1 and thus potentially suppress the formation of a repressor complex at the Dsc3 promoter. On the other hand, Lef-1 is required to shuttle Pg into the cell nucleus, where it activates the Dsc2 promoter as shown in Figure 3d and e. The data summarized above demonstrate that Lef-1 and Pg localize to the nucleus, which is predicted to lead to an activation of the Dsc2 gene and a suppression of the Dsc3 gene.
      Figure thumbnail gr3
      Figure 3Plakoglobin (Pg) colocalizes in the nucleus with lymphoid enhancer-binding factor 1 (Lef-1) and disrupts T-cell factor (TCF)/Lef transcription factor binding to the desmocollin 3 (Dsc3) promoter. (a) Co-immunoprecipitation (Co-IP) assays demonstrating an interaction between Pg and Lef-1. (b) Chromatin immunoprecipitation (ChIP) assays demonstrating that increasing amounts of Pg (measured in μg plasmid transfected) interfere with the binding of Lef-1 to the Dsc3 promoter. Input, chromatin used for immunoprecipitation; IgG, IP with unspecific IgG. (c, d) Western blot analysis of Madin Darby canine kidney (MDCK) cells transfected with Pg and Lef-1 (transfection constructs shown on top; NT, not transfected). The nuclear (N) and cytoplasmic (C) distribution of the proteins is shown. Antibodies used to detect Lef-1 and the KT3-tagged plakoglobin construct are shown on the left sides of the blots. Our Lef-1 antibody does not detect endogenous Lef-1 expression in MDCK cells. Lamin B1 (nuclear fraction) and α-tubulin (cytoplasmic fraction) antibodies were used as controls. (e) Immunofluorescence microscopy of MDCK cells transfected with Pg and Lef-1. The antibodies used for staining are indicated. Note the nuclear colocalization of Pg and Lef-1 in several cells (arrows). Bar=50μm.

      Topology of Lef-1 and Dsc3 expression in developing hair follicles is mutually exclusive

      Our results so far suggest that Lef-1 can act as a switch between Dsc2 and Dsc3 expression and that the presence of Lef-1 in keratinocytes can suppress the Dsc3 gene. We therefore compared the distribution of DSC3 and Lef-1 in transgenic mice expressing a nuclear LacZ reporter under the control of the Dsc3 promoter (Dsc3-LacZ mice). These animals were designed to identify cells and tissues that express low levels of Dsc3, or cells in which antigen masking prevented protein detection via antibodies. In all developmental stages and tissues examined so far, DSC3 antibody staining and LacZ transgene activity perfectly overlapped (data not shown).
      We conducted whole-mount β-galactosidase staining experiments using Dsc3-LacZ embryos (Figure 4a). At embryonic day 15.5 (E15.5), β-galactosidase activity was prominent in whisker pads (vibrissae follicles), in mammary gland buds, and in developing HFs over the entire body surface (Figure 4a, and data not shown). Strikingly, the overall DSC3-LacZ expression pattern appeared similar to the staining patterns observed in transgenic embryos of the same age that expressed a TCF/Lef-regulated promoter driving a LacZ reporter (Figure 4b; Bat-Gal;
      • Maretto S.
      • Cordenonsi M.
      • Dupont S.
      • et al.
      Mapping Wnt/beta-catenin signaling during mouse development and in colorectal tumors.
      ). However, a histochemical analysis revealed that the expression patterns of the Wnt pathway components (Lef-1; TCF-3/-4; data not shown) and the DSC3-LacZ transgene did not overlap at the cellular level. In fact, Wnt activity (Lef-1 expression) and Dsc3 expression were mutually exclusive (Figure 4c and d). We observed that LacZ-positive (Dsc3-expressing) cells were found in the suprabasal layer on top of the newly forming HFs. Lef-1, on the other hand, was observed in the leading edge of keratinocytes growing down into the dermis and throughout the basal cell layer. These findings were confirmed by staining E16.5 wild-type mouse epidermis with Lef-1 and DSC3 antibodies (Figure 4e–h). Strong Lef-1 antibody staining correlated with reduced or absent staining for DSC3. Unfortunately, we were not able to assess the distributions of DSC2, as antibodies that recognize mouse DSC2 are not available. Pg antibodies stained cell–cell borders in placode keratinocytes (data not shown). Nuclear Pg staining was not observed, possibly because of epitope masking or low nuclear Pg levels.
      Figure thumbnail gr4
      Figure 4Wnt activity and desmocollin 3 (Dsc3) expression during mouse appendage development. (a) Whole-mount in situ staining for β-galactosidase (β-gal) activity of a transgenic mouse (Dsc3-LacZ BAC, embryonic day 15.5 (E15.5)) expressing β-gal under the control of the Dsc3 promoter. Note that whisker pads, hair follicles, and mammary glands (not shown) are strongly stained. (b) Whole-mount β-gal staining of a BAT-gal transgenic mouse (
      • Maretto S.
      • Cordenonsi M.
      • Dupont S.
      • et al.
      Mapping Wnt/beta-catenin signaling during mouse development and in colorectal tumors.
      ) at E15.5. Note the similar staining patterns of the transgenic mice shown in (a) and (b). (c) Immunohistochemical staining of a skin section from a Dsc3-LacZ BAC mouse at E15.5 with (c) β-gal antibodies (Abs) and (d) Lef-1 Abs. The LacZ transgene contains a nuclear localization signal. Note that β-gal and Lef-1 expression appears to be mutually exclusive. Arrows point to Lef-1-expressing cells in a placode and a dermal papilla of a forming hair follicle. (e, g) Immunofluorescence staining of a developing hair follicle in wild-type mouse skin at E16.5 with antibodies against Lef-1 and DSC3. The area demarcated by the dotted box is shown at higher magnification in (h). (h) The large arrow points to DSC3-positive keratinocytes in the basal cell layer. The short arrow points to DSC3-negative and Lef-1-positive cells in a forming hair placode. Dotted lines in e and h indicate basement membrane area. Bar=50μm.

      NF-κB regulation of the Dsc2 gene

      Our initial analysis of the Dsc promoters indicated the presence of putative NF-κB (Rel)-binding sites in both Dsc promoters. We thus performed ChIP assays and reporter assays to determine the effect of Rel factors and a dominant-negative (DN) construct blocking NF-κB signaling (IκB-DN) on Dsc promoter activity. The results shown in Figure 5a–c demonstrate that c-Rel specifically binds to and activates the proximal Dsc2 promoter, whereas we did not observe any effect of the NF-κB factors on the proximal Dsc3 promoter (Figure 5d). We then transfected primary keratinocytes with the NF-κB factors and assessed endogenous Dsc gene expression (Figure 5e and f). We confirmed the specific activation of the endogenous Dsc2 gene by ectopic expression of c-Rel.
      Figure thumbnail gr5
      Figure 5Effects of NF-κB transcription factors on the proximal desmocollin 2 (Dsc2) and desmocollin 3 (Dsc3) promoters. (a) Chromatin immunoprecipitation (ChIP) assays demonstrating that both c-Rel and Rel A bind to a Dsc2 promoter fragment that contains two predicted Rel-binding sites. Input, chromatin used for immunoprecipitation (IP); IgG, IP with unspecific IgG. (b) Schematic representation of putative NF-κB transcription factor–binding sites in the proximal Dsc2 promoter. The arrows indicate translation start sites (ATG; A is defined as position +1). (c) c-Rel markedly increases the reporter activity of the Dsc2 construct, whereas Rel A does not. The IκB-DN construct encodes a dominant-negative (DN) inhibitor of the NF-κB pathway, i.e., this construct can block endogenous NF-κB activity. The promoter activities in the absence of ectopically expressed transcription factors are defined as relative expression level 1. (d) The Dsc3 reporter construct does not show any significant changes in activity in response to NF-κB transcription factors. (e, f) Endogenous Dsc gene expression in mouse keratinocytes. (e) c-Rel ectopic expression significantly increases endogenous Dsc2 expression. (f) None of the NF-κB transcription factors affects endogenous Dsc3 expression. (g) Simplified model of signaling pathways active in hair follicle placodes and their proposed effects on Dsc2 and Dsc3 gene expression. Note that open arrows symbolize the upregulation and downregulation of gene expression, respectively. Error bars, SD. Stars indicate statistically significant results (P<0.05). Lef-1, lymphoid enhancer-binding factor 1; Tcf, T-cell factor.

      DISCUSSION

      It has been shown that deregulated expression or loss of desmosomal cadherins, including Dsc, can lead to abnormal differentiation of the epidermis and defects in HFs (see, e.g.,
      • Chidgey M.
      • Brakebusch C.
      • Gustafsson E.
      • et al.
      Mice lacking desmocollin 1 show epidermal fragility accompanied by barrier defects and abnormal differentiation.
      ;
      • Elias P.M.
      • Matsuyoshi N.
      • Wu H.
      • et al.
      Desmoglein isoform distribution affects stratum corneum structure and function.
      ;
      • Merritt A.J.
      • Berika M.Y.
      • Zhai W.
      • et al.
      Suprabasal desmoglein 3 expression in the epidermis of transgenic mice results in hyperproliferation and abnormal differentiation.
      ;
      • Hardman M.J.
      • Liu K.
      • Avilion A.A.
      • et al.
      Desmosomal cadherin misexpression alters beta-catenin stability and epidermal differentiation.
      ), suggesting that tight control of Dsc gene expression is required for normal epidermal development and homeostasis. In this study, we focused on the regulation of Dsc2 and Dsc3, the two DSCs expressed in the deep epidermis, the compartment that maintains the skin and which has a crucial role in skin appendage development.
      Little is known regarding gene-regulatory pathways that control the expression of Dsc genes in processes such as epidermal differentiation and HF formation. Epidermal appendage formation requires extensive remodeling of cell adhesion systems, including desmosomes (
      • Kurzen H.
      • Moll I.
      • Moll R.
      • et al.
      Compositionally different desmosomes in the various compartments of the human hair follicle.
      ). The first step in the development of these appendages is placode formation. Keratinocytes in these structures segregate from the surrounding epidermis and then begin to invade the dermis.
      • Nanba D.
      • Hieda Y.
      • Nakanishi Y.
      Remodeling of desmosomal and hemidesmosomal adhesion systems during early morphogenesis of mouse pelage hair follicles.
      ,
      • Nanba D.
      • Nakanishi Y.
      • Hieda Y.
      Changes in adhesive properties of epithelial cells during early morphogenesis of the mammary gland.
      ) have demonstrated that DSCs are downregulated both in the hair and mammary gland placodes. However, the antigen specificity of the antibody used by these authors was not determined, i.e., it was not known which of the DSC proteins was actually downregulated. Given the results presented in this study, it is likely that the main desmosomal cadherin isoform downregulated in this process is Dsc3.
      The expression of classical cadherins is also switched during hair follicle formation. E-cadherin is downregulated in HF placodes, whereas P-cadherin is upregulated. Furthermore, it was shown that forced expression of E-cadherin inhibits HF formation (
      • Jamora C.
      • DasGupta R.
      • Kocieniewski P.
      • et al.
      Links between signal transduction, transcription and adhesion in epithelial bud development.
      ), demonstrating that control of cadherin isoform expression is crucial for HF formation.
      Given our results, it is tempting to speculate that an analogous switch occurs from Dsc3 (TCF/Lef factor-mediated repression) to Dsc2 (Pg/Lef-1-mediated induction) in placode keratinocytes (Figure 5g). Unfortunately, owing to a lack of appropriate antibodies, we currently do not have the tools required to assess DSC2 expression in mouse epidermis. Nevertheless, expression studies in human embryonic epidermis indicated increased DSC2 and reduced DSC3 expression in bulbous hair pegs, suggesting that the above postulated switch from Dsc3 to Dsc2 expression is likely to occur in mammalian skin (
      • Kurzen H.
      • Moll I.
      • Moll R.
      • et al.
      Compositionally different desmosomes in the various compartments of the human hair follicle.
      ).
      Pg appears to activate both the Dsc2 and the Dsc3 promoter. Previous studies have suggested that Pg can signal by changing the levels of signaling active β-catenin in cells. Our experiments failed to demonstrate a role of β-catenin in regulating the two Dsc genes, suggesting that the signaling we observed is a specific function of Pg.
      The Dsc2 promoter requires Lef-1 for activation and this regulation is dependent on the TCF/Lef-binding site in the Dsc2 promoter. Interestingly, this effect is specific for Lef-1 as TCF3 and TCF4 do not appear to be able to substitute for Lef-1. It is noteworthy that Lef-1 expression facilitates nuclear accumulation of Pg, a prerequisite for activation of the Dsc2 gene.
      Pg activated the Dsc3 promoter without direct binding. Interestingly, ectopic expression of TCF/Lef factors completely blocked Pg-mediated activation of the Dsc3 promoter. This suggests that Pg interferes with the activity of a repressor complex containing TCF/Lef factors that inhibits the Dsc3 promoter. A likely mechanism that could explain this observation is that binding of Pg to TCF/Lef proteins causes a depletion of the TCF/Lef pool, thus preventing these factors from forming an inhibitory complex on the Dsc3 promoter.
      Previous experiments in mouse models suggested that Pg is not required for the formation of HFs (
      • Teuliere J.
      • Faraldo M.M.
      • Shtutman M.
      • et al.
      beta-Catenin-dependent and -independent effects of DeltaN-plakoglobin on epidermal growth and differentiation.
      ;
      • Li D.
      • Zhang W.
      • Liu Y.
      • et al.
      Lack of plakoglobin in epidermis leads to keratoderma.
      ). These results suggest the existence of alternative mechanisms that can regulate a switch of Dsc isoforms, specifically the activation of Dsc2. NF-κB signaling might be such an alternative mechanism. Our experiments revealed a role of c-Rel, a NF-κB protein, in activating the Dsc2 gene. NF-κB proteins are downstream effectors of the EDA/EDAR/NF-κB pathway, which interact with the Wnt pathway, a process crucial for HF formation during embryogenesis (
      • Zhang Y.
      • Tomann P.
      • Andl T.
      • et al.
      Reciprocal requirements for EDA/EDAR/NF-kappaB and Wnt/beta-catenin signaling pathways in hair follicle induction.
      ). The observation that both catenin/TCF/Lef and NF-κB signaling are required for pelage hair development, and our in vitro data demonstrating differential regulation of Dsc genes by these two pathways, suggest the possibility that both signaling cascades partially function through modulating desmosomal cell adhesion.
      On the basis of the observation summarized above, we suggest the following mechanisms for the regulation of Dsc2 and Dsc3 gene expression during the early stages of skin appendage formation: Wnt activation in the early placode leads to an increase accumulation of Lef-1 in placodes. Lef-1, potentially in conjunction with other corepressors (
      • Hoverter N.P.
      • Waterman M.L.
      A Wnt-fall for gene regulation: repression.
      ;
      • Arce L.
      • Pate K.T.
      • Waterman M.L.
      Groucho binds two conserved regions of LEF-1 for HDAC-dependent repression.
      ), binds to the Dsc3 promoter and inhibits expression, thus leading to a reduction in the desmosomal adhesion receptors present at the plasma membrane. In turn, this could lead to a transient increase of cytoplasmic Pg, which is usually bound to the carboxy-terminal domain of DSC3. Given the abundance of Lef-1 in placode keratinocytes, Pg would then form a transcription complex with Lef-1 and activate the Dsc2 promoter. This chain of events would lead to a shift from DSC3 to DSC2 as the main DSC synthesized in placode keratinocytes, which would be consistent with the proposed distribution of these two proteins in skin placodes (our data and
      • Kurzen H.
      • Moll I.
      • Moll R.
      • et al.
      Compositionally different desmosomes in the various compartments of the human hair follicle.
      ). Activation of the EDA/EDAR pathway, via its NF-κB effectors, would then further facilitate the shift from DSC3 to DSC2. It is possible that this change from DSC3 to DSC2 is better suited to support invasive keratinocytes growth. In this context, it is noteworthy that we have previously shown in a mouse model that invasive growth of skin squamous cell carcinoma is associated with a specific loss of Dsc3 expression in tumor cells (
      • Chen J.
      • O’Shea C.
      • Fitzpatrick J.E.
      • et al.
      Loss of desmocollin 3 in skin tumor development and progression.
      ). Similar results have also been reported in other types of cancer, such as breast cancer (
      • Klus G.T.
      • Rokaeus N.
      • Bittner M.L.
      • et al.
      Down-regulation of the desmosomal cadherin desmocollin 3 in human breast cancer.
      ).
      Further experiments will be required to unequivocally prove this model. It is tempting to speculate that changes in desmosomal cell adhesion might affect the signaling pool of plakoglobin and thus, in a feedback loop, control the expression of genes that encode central desmosomal components. This represents a previously unreported and exciting concept that can now be tested in vivo.

      MATERIALS AND METHODS

      Animal protocols

      Animal experiments were approved by the institutional animal care and use committee of the University of Colorado Denver (UC Denver; Denver, CO).

      Generation of Dsc promoter constructs and luciferase reporter assays

      A total of 2kb promoter sequences immediately upstream of the Dsc2 and Dsc3 translation start codons were cloned into pGL3-basic vector (Promega, Madison, WI), which contains a promoter-less luciferase reporter cassette. TCF/Lef-1 mutations were generated by PCR using primers DSC2-5F/5R and DSC3-2F/2R (Supplementary Table S1 online). The following expression vectors were used: CMV-βgal (Dennis Roop, UC Denver); pcDNA1.1p65 and pcDNA-Iκβ-DN (Rune Toftgård, Karolinska Institute, Sweden); pcmv4c-Rel (Warner Greene, University of California, San Francisco, San Francisco, CA) (
      • Doerre S.
      • Sista P.
      • Sun S.C.
      • et al.
      The c-rel protooncogene product represses NF-kappa B p65-mediated transcriptional activation of the long terminal repeat of type 1 human immunodeficiency virus.
      ); pCS2 ΔNPβ-catenin (Pamela Cowin, New York University, New York, NY) (
      • Imbert A.
      • Eelkema R.
      • Jordan S.
      • et al.
      Delta N89 beta-catenin induces precocious development, differentiation, and neoplasia in mammary gland.
      ); pRcCMV plakoglobin expression vector tagged with a KT3 epitope (Ansgar Smith, University of Marburg, Marburg, Germany); pCS2-hLef-1 (Rolf Kemler, Max-Planck Institute, Freiburg, Germany) (
      • Huber O.
      • Korn R.
      • McLaughlin J.
      • et al.
      Nuclear localization of beta-catenin by interaction with transcription factor LEF-1.
      ); and pGLOWMYC-hTcf-4 (Hans Clevers, University Hospital, Utrecht, The Netherlands) (
      • Korinek V.
      • Barker N.
      • Morin P.J.
      • et al.
      Constitutive transcriptional activation by a beta-catenin-Tcf complex in APC-/- colon carcinoma.
      ). Plasmids were transiently transfected into the canine kidney epithelial cell line MDCK (ATCC CCL-34) and mouse keratinocytes (MPEK-BL6; Cellntec, Bern, Switzerland), using the Nucleofector Technology (Lonza, Walkersville, MD). Reporter activities were measured with the “Chemiluminescent Reporter Gene Assay System” (Applied Biosystems, Bedford, MA) using a Glomax Multi Detection System (Promega, Madison, WI). The results shown represent average expression levels from three independent experiments, each conducted in triplicate (error bars in all figures: SD, P-values of <0.05 were considered statistically significant).

      ChIP experiments

      ChIP assays were performed essentially as described previously (
      • Koster M.I.
      • Dai D.
      • Marinari B.
      • et al.
      p63 induces key target genes required for epidermal morphogenesis.
      ). The following antibodies were used: Pg (Cell Signaling, Boston, MA), β-catenin (Santa Cruz Biotechnology, Santa Cruz, CA), Lef-1 (Millipore, Billerica, MA), and TCF-4 (Millipore). The following primers (Supplementary Table S1 online) were used for quantitative real-time reverse transcriptase–-PCR: Dsc2-4F/4R (Rel-binding site in Dsc2 promoter), DSC2-6F/6R (TCF/Lef-binding site in Dsc2 promoter), and DSC3-3F/3R (TCF/Lef-binding site in Dsc3 promoter). The data shown are based on three independent experiments.

      Generation of Dsc3-LacZ transgenic mice

      The BAC vector RP23-290M4 (BACPAC Resource, Children’s Hospital Oakland Research Institute, Oakland, CA), which contains the entire mouse Dsc3 gene, including 29kb of 5′ upstream and 90kb of 3′ downstream sequences, was used to generate the Dsc3-LacZ transgene. A promoter-less LacZ cassette (NLS-LacZ-PA cassette; provided by Dr Ming-Jer Tsai, Baylor College of Medicine, Houston, TX) was inserted into exon 1 of the Dsc3 gene immediately downstream of the start codon. The LacZ cassette contains a nuclear localization sequence, i.e., transgenic mice can be identified by the detection of β-galactosidase activity in nuclei.

      Immunostaining, β-galactosidase staining, western blotting, and Co-Immunoprecipitation

      All experiments were conducted following standard protocols. The following antibodies were used: Lef-1 (Cell Signaling), KT3 antibody (Ansgar Schmidt, University of Marburg); normal IgG (Millipore), Lamin B1 (Santa Cruz Biotechnology), α-tubulin (Sigma, Saint-Louis, MO), DSC3 (
      • Cheng X.
      • Mihindukulasuriya K.
      • Den Z.
      • et al.
      Assessment of splice variant-specific functions of desmocollin 1 in the skin.
      ), plakoglobin (Firtzgerald, North Acton, MA), β-galactosidase (a gift from Dennis Roop, UC Denver); horseradish peroxidase–conjugated and biotinylated secondary antibodies (Vector Laboratories, Burlingame, CA); and Alexa Fluor–coupled secondary antibodies (Invitrogen, Grand Island, NY). Antibody binding was detected and quantified as described previously (
      • Koch P.J.
      • de Viragh P.A.
      • Scharer E.
      • et al.
      Lessons from loricrin-deficient mice: compensatory mechanisms maintaining skin barrier function in the absence of a major cornified envelope protein.
      ).

      Quantitative real-time quantitative reverse transcriptase–PCR

      Real-time reverse transcriptase–PCR was performed using a LightCycler 480 from Roche (Indianapolis, IN), following the manufacturer’s recommendations. Dsc2 and Dsc3 complementary DNAs were amplified with “assay-on-demand probes” Mm00516355_m1 and Mm00492270_m1, respectively, from Applied Biosystems. A glyceraldehyde-3-phosphate dehydrogenase probe set from Applied Biosystems was used as an internal control. The data shown are based on three independent experiments.

      ACKNOWLEDGMENTS

      We thank Charlene O’Shea for expert technical assistance, Maranke Koster and Jason Dinella for critical reading of the manuscript, and Sarah Millar (University of Pennsylvania) for providing Bat-Gal mice with permission from Stefano Piccolo (University of Padova, Padova, Italy). Research reported in this publication was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health under award number 5RO1AR053892. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

      SUPPLEMENTARY MATERIAL

      Supplementary Material is linked to the online version of the paper at http://www.nature.com/jid

      REFERENCES

        • Amagai M.
        • Stanley J.R.
        Desmoglein as a target in skin disease and beyond.
        J Invest Dermatol. 2012; 132: 776-784
        • Arce L.
        • Pate K.T.
        • Waterman M.L.
        Groucho binds two conserved regions of LEF-1 for HDAC-dependent repression.
        BMC Cancer. 2009; 9: 159
        • Chen J.
        • Den Z.
        • Koch P.J.
        Loss of desmocollin 3 in mice leads to epidermal blistering.
        J Cell Sci. 2008; 121: 2844-2849
        • Chen J.
        • O’Shea C.
        • Fitzpatrick J.E.
        • et al.
        Loss of desmocollin 3 in skin tumor development and progression.
        Mol Carcinog. 2011; 51: 535-545
        • Cheng X.
        • Den Z.
        • Koch P.J.
        Desmosomal cell adhesion in mammalian development.
        Eur J Cell Biol. 2005; 84: 215-223
        • Cheng X.
        • Koch P.J.
        In vivo function of desmosomes.
        J Dermatol. 2004; 31: 171-187
        • Cheng X.
        • Mihindukulasuriya K.
        • Den Z.
        • et al.
        Assessment of splice variant-specific functions of desmocollin 1 in the skin.
        Mol Cell Biol. 2004; 24: 154-163
        • Chidgey M.
        • Brakebusch C.
        • Gustafsson E.
        • et al.
        Mice lacking desmocollin 1 show epidermal fragility accompanied by barrier defects and abnormal differentiation.
        J Cell Biol. 2001; 155: 821-832
        • Chidgey M.A.
        • Yue K.K.
        • Gould S.
        • et al.
        Changing pattern of desmocollin 3 expression accompanies epidermal organisation during skin development.
        Dev Dyn. 1997; 210: 315-327
        • Doerre S.
        • Sista P.
        • Sun S.C.
        • et al.
        The c-rel protooncogene product represses NF-kappa B p65-mediated transcriptional activation of the long terminal repeat of type 1 human immunodeficiency virus.
        Proc Natl Acad Sci USA. 1993; 90: 1023-1027
        • Dubash A.D.
        • Green K.J.
        Desmosomes.
        Curr Biol. 2011; 21: R529-R531
        • Dusek R.L.
        • Godsel L.M.
        • Chen F.
        • et al.
        Plakoglobin deficiency protects keratinocytes from apoptosis.
        J Invest Dermatol. 2007; 127: 792-801
        • Elias P.M.
        • Matsuyoshi N.
        • Wu H.
        • et al.
        Desmoglein isoform distribution affects stratum corneum structure and function.
        J Cell Biol. 2001; 153: 243-249
        • Franzen C.A.
        • Todorovic V.
        • Desai B.V.
        • et al.
        The desmosomal armadillo protein plakoglobin regulates prostate cancer cell adhesion and motility through vitronectin-dependent Src signaling.
        PloS One. 2012; 7: e42132
        • Galichet A.
        • Strauss C.
        • Sayar B.S.
        • et al.
        Desmosomal cadherins crosstalk to the nucleus via a novel potent transcriptional regulator of the Wnt pathway: Plakoglobin (PG, JUP, γ-catenin).
        J Investig Dermatol. 2012; 132: S60
        • Ganeshan R.
        • Chen J.
        • Koch P.J.
        Mouse models for blistering skin disorders.
        Dermatol Res Pract. 2010; 2010: 584353
        • Hardman M.J.
        • Liu K.
        • Avilion A.A.
        • et al.
        Desmosomal cadherin misexpression alters beta-catenin stability and epidermal differentiation.
        Mol Cell Biol. 2005; 25: 969-978
        • Hoverter N.P.
        • Waterman M.L.
        A Wnt-fall for gene regulation: repression.
        Sci Signal. 2008; 1: pe43
        • Huber O.
        • Korn R.
        • McLaughlin J.
        • et al.
        Nuclear localization of beta-catenin by interaction with transcription factor LEF-1.
        Mech Dev. 1996; 59: 3-10
        • Imbert A.
        • Eelkema R.
        • Jordan S.
        • et al.
        Delta N89 beta-catenin induces precocious development, differentiation, and neoplasia in mammary gland.
        J Cell Biol. 2001; 153: 555-568
        • Jamora C.
        • DasGupta R.
        • Kocieniewski P.
        • et al.
        Links between signal transduction, transcription and adhesion in epithelial bud development.
        Nature. 2003; 422: 317-322
        • Klus G.T.
        • Rokaeus N.
        • Bittner M.L.
        • et al.
        Down-regulation of the desmosomal cadherin desmocollin 3 in human breast cancer.
        Int J Oncol. 2001; 19: 169-174
        • Koch P.J.
        • de Viragh P.A.
        • Scharer E.
        • et al.
        Lessons from loricrin-deficient mice: compensatory mechanisms maintaining skin barrier function in the absence of a major cornified envelope protein.
        J Cell Biol. 2000; 151: 389-400
        • Koch P.J.
        • Mahoney M.G.
        • Cotsarelis G.
        • et al.
        Desmoglein 3 anchors telogen hair in the follicle.
        J Cell Sci. 1998; 111: 2529-2537
        • Koch P.J.
        • Mahoney M.G.
        • 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.
        J Cell Biol. 1997; 137: 1091-1102
        • Korinek V.
        • Barker N.
        • Morin P.J.
        • et al.
        Constitutive transcriptional activation by a beta-catenin-Tcf complex in APC-/- colon carcinoma.
        Science. 1997; 275: 1784-1787
        • Koster M.I.
        • Dai D.
        • Marinari B.
        • et al.
        p63 induces key target genes required for epidermal morphogenesis.
        Proc Natl Acad Sci USA. 2007; 104: 3255-3260
        • Kurzen H.
        • Moll I.
        • Moll R.
        • et al.
        Compositionally different desmosomes in the various compartments of the human hair follicle.
        Differentiation. 1998; 63: 295-304
        • Li D.
        • Zhang W.
        • Liu Y.
        • et al.
        Lack of plakoglobin in epidermis leads to keratoderma.
        J Biol Chem. 2012; 287: 10435-10443
        • Li J.
        • Swope D.
        • Raess N.
        • et al.
        Cardiac tissue-restricted deletion of plakoglobin results in progressive cardiomyopathy and activation of {beta}-catenin signaling.
        Mol Cell Biol. 2011; 31: 1134-1144
        • Maeda O.
        • Usami N.
        • Kondo M.
        • et al.
        Plakoglobin (gamma-catenin) has TCF/LEF family-dependent transcriptional activity in beta-catenin-deficient cell line.
        Oncogene. 2004; 23: 964-972
        • Maretto S.
        • Cordenonsi M.
        • Dupont S.
        • et al.
        Mapping Wnt/beta-catenin signaling during mouse development and in colorectal tumors.
        Proc Natl Acad Sci USA. 2003; 100: 3299-3304
        • Merritt A.J.
        • Berika M.Y.
        • Zhai W.
        • et al.
        Suprabasal desmoglein 3 expression in the epidermis of transgenic mice results in hyperproliferation and abnormal differentiation.
        Mol Cell Biol. 2002; 22: 5846-5858
        • Muller E.J.
        • Williamson L.
        • Kolly C.
        • et al.
        Outside-in signaling through integrins and cadherins: a central mechanism to control epidermal growth and differentiation?.
        J Invest Dermatol. 2008; 128: 501-516
        • Nanba D.
        • Hieda Y.
        • Nakanishi Y.
        Remodeling of desmosomal and hemidesmosomal adhesion systems during early morphogenesis of mouse pelage hair follicles.
        J Invest Dermatol. 2000; 114: 171-177
        • Nanba D.
        • Nakanishi Y.
        • Hieda Y.
        Changes in adhesive properties of epithelial cells during early morphogenesis of the mammary gland.
        Dev Growth Differ. 2001; 43: 535-544
        • Oshiro M.M.
        • Kim C.J.
        • Wozniak R.J.
        • et al.
        Epigenetic silencing of DSC3 is a common event in human breast cancer.
        Breast Cancer Res. 2005; 7: R669-R680
        • Oshiro M.M.
        • Watts G.S.
        • Wozniak R.J.
        • et al.
        Mutant p53 and aberrant cytosine methylation cooperate to silence gene expression.
        Oncogene. 2003; 22: 3624-3634
        • Petrof G.
        • Mellerio J.E.
        • McGrath J.A.
        Desmosomal genodermatoses.
        Br J Dermatol. 2012; 166: 36-45
        • Rieger-Christ K.M.
        • Ng L.
        • Hanley R.S.
        • et al.
        Restoration of plakoglobin expression in bladder carcinoma cell lines suppresses cell migration and tumorigenic potential.
        Br J Cancer. 2005; 92: 2153-2159
        • Schmidt A.
        • Koch P.J.
        Desmosomes: just cell adhesion or is there more?.
        Cell Adh Migr. 2007; 1: 28-32
        • Schmidt-Ullrich R.
        • Aebischer T.
        • Hulsken J.
        • et al.
        Requirement of NF-kappaB/Rel for the development of hair follicles and other epidermal appendices.
        Development. 2001; 128: 3843-3853
        • Shtutman M.
        • Zhurinsky J.
        • Oren M.
        • et al.
        PML is a target gene of beta-catenin and plakoglobin, and coactivates beta-catenin-mediated transcription.
        Cancer Res. 2002; 62: 5947-5954
        • Simcha I.
        • Shtutman M.
        • Salomon D.
        • et al.
        Differential nuclear translocation and transactivation potential of beta-catenin and plakoglobin.
        J Cell Biol. 1998; 141: 1433-1448
        • Smith C.
        • Zhu K.
        • Merritt A.
        • et al.
        Regulation of desmocollin gene expression in the epidermis: CCAAT/enhancer-binding proteins modulate early and late events in keratinocyte differentiation.
        Biochem J. 2004; 380: 757-765
        • Swope D.
        • Li J.
        • Radice G.L.
        Beyond cell adhesion: the role of armadillo proteins in the heart.
        Cell Signal. 2013; 25: 93-100
        • Teuliere J.
        • Faraldo M.M.
        • Shtutman M.
        • et al.
        beta-Catenin-dependent and -independent effects of DeltaN-plakoglobin on epidermal growth and differentiation.
        Mol Cell Biol. 2004; 24: 8649-8661
        • Williamson L.
        • Raess N.A.
        • Caldelari R.
        • et al.
        Pemphigus vulgaris identifies plakoglobin as key suppressor of c-Myc in the skin.
        EMBO J. 2006; 25: 3298-3309
        • Yin T.
        • Getsios S.
        • Caldelari R.
        • et al.
        Plakoglobin suppresses keratinocyte motility through both cell-cell adhesion-dependent and -independent mechanisms.
        Proc Natl Acad Sci USA. 2005; 102: 5420-5425
        • Zhang Y.
        • Tomann P.
        • Andl T.
        • et al.
        Reciprocal requirements for EDA/EDAR/NF-kappaB and Wnt/beta-catenin signaling pathways in hair follicle induction.
        Dev Cell. 2009; 17: 49-61
        • Zhurinsky J.
        • Shtutman M.
        • Ben-Ze’ev A.
        Differential mechanisms of LEF/TCF family-dependent transcriptional activation by beta-catenin and plakoglobin.
        Mol Cell Biol. 2000; 20: 4238-4252