Advertisement

Keratins Mediate Localization of Hemidesmosomes and Repress Cell Motility

  • Kristin Seltmann
    Affiliations
    Division of Cell and Developmental Biology, Translational Center for Regenerative Medicine (TRM), Institute of Biology, University of Leipzig, Leipzig, Germany
    Search for articles by this author
  • Wera Roth
    Affiliations
    Division of Cell and Developmental Biology, Translational Center for Regenerative Medicine (TRM), Institute of Biology, University of Leipzig, Leipzig, Germany
    Search for articles by this author
  • Cornelia Kröger
    Affiliations
    Division of Cell and Developmental Biology, Translational Center for Regenerative Medicine (TRM), Institute of Biology, University of Leipzig, Leipzig, Germany
    Search for articles by this author
  • Fanny Loschke
    Affiliations
    Division of Cell and Developmental Biology, Translational Center for Regenerative Medicine (TRM), Institute of Biology, University of Leipzig, Leipzig, Germany
    Search for articles by this author
  • Marcell Lederer
    Affiliations
    Section for Molecular Cell Biology, Institute of Molecular Medicine, Martin Luther University of Halle, Halle, Germany
    Search for articles by this author
  • Stefan Hüttelmaier
    Affiliations
    Section for Molecular Cell Biology, Institute of Molecular Medicine, Martin Luther University of Halle, Halle, Germany
    Search for articles by this author
  • Thomas M. Magin
    Correspondence
    Division of Cell and Developmental Biology, Translational Center for Regenerative Medicine (TRM), Institute of Biology, University of Leipzig, Philipp-Rosenthal-Straße 55, D-04103 Leipzig, Germany
    Affiliations
    Division of Cell and Developmental Biology, Translational Center for Regenerative Medicine (TRM), Institute of Biology, University of Leipzig, Leipzig, Germany
    Search for articles by this author
      The keratin (K)–hemidesmosome (HD) interaction is crucial for cell-matrix adhesion and migration in several epithelia, including the epidermis. Mutations in constituent proteins cause severe blistering skin disorders by disrupting the adhesion complex. Despite extensive studies, the role of keratins in HD assembly and maintenance is only partially understood. Here we address this issue in keratinocytes in which all keratins are depleted by genome engineering. Unexpectedly, such keratinocytes maintain many characteristics of their normal counterparts. However, the absence of the entire keratin cytoskeleton leads to loss of plectin from the hemidesmosomal plaque and scattering of the HD transmembrane core along the basement membrane zone. To investigate the functional consequences, we performed migration and adhesion assays. These revealed that, in the absence of keratins, keratinocytes adhere much faster to extracellular matrix substrates and migrate approximately two times faster compared with wild-type cells. Reexpression of the single keratin pair K5 and K14 fully reversed the above phenotype. Our data uncover a role of keratins, which to our knowledge is previously unreported, in the maintenance of HDs upstream of plectin, with implications for epidermal homeostasis and pathogenesis. They support the view that the downregulation of keratins observed during epithelial–mesenchymal transition supports the migratory and invasive behavior of tumor cells.

      Abbreviations

      BP
      bullous pemphigoid
      GFP
      green fluorescent protein
      HD
      hemidesmosome
      K
      keratin
      KO
      knockout
      WT
      wild type

      Introduction

      Cell-matrix adhesion is crucial for a variety of biological processes in skin, including skin development, wound healing, inflammation, and malignant progression (
      • Ridley A.J.
      • Schwartz M.A.
      • Burridge K.
      • et al.
      Cell migration: integrating signals from front to back.
      . In the skin, matrix adhesion is maintained by actin-associated focal adhesions and keratin-dependent hemidesmosomes (HDs). Although the former are relatively well characterized, little is known about the mechanisms regulating the interactions of keratins during HD formation and maintenance. Disruption of the keratin–HD multiprotein complex gives rise to epidermolysis bullosa, a group of severe skin disorders caused by mutations in genes encoding either hemidesmosomal proteins, or keratins K5 and K14 (
      • Fine J.D.
      Inherited epidermolysis bullosa: past, present, and future.
      . The binding of plectin to α6β4-integrin is regarded to be essential for the assembly and stability of HDs (
      • de Pereda J.M.
      • Lillo M.P.
      • Sonnenberg A.
      Structural basis of the interaction between integrin alpha6beta4 and plectin at the hemidesmosomes.
      . Plectin interacts with at least three binding sites to β4-integrin mainly through the N-terminal actin-binding domain (
      • Koster J.
      • van W.S.
      • Kuikman I.
      • et al.
      Role of binding of plectin to the integrin beta4 subunit in the assembly of hemidesmosomes.
      . Binding to keratin intermediate filaments involves the C-terminal plakin repeat domain. Owing to its binding sites for actin, intermediate filament proteins, and microtubules, plectin qualifies to coordinate the dynamics between HDs and focal adhesion dynamics by rearranging the actin and keratin cytoskeletons (
      • Andra K.
      • Kornacker I.
      • Jorgl A.
      • et al.
      Plectin-isoform-specific rescue of hemidesmosomal defects in plectin (−/−) keratinocytes.
      ;
      • Ozawa T.
      • Tsuruta D.
      • Jones J.C.
      • et al.
      Dynamic relationship of focal contacts and hemidesmosome protein complexes in live cells.
      ;
      • Tsuruta D.
      • Hashimoto T.
      • Hamill K.J.
      • et al.
      Hemidesmosomes and focal contact proteins: functions and cross-talk in keratinocytes, bullous diseases and wound healing.
      .
      HDs connect the keratin cytoskeleton (K5/K14) with the extracellular matrix (
      • Jones J.C.
      • Green K.J.
      Intermediate filament-plasma membrane interactions.
      . The transmembrane proteins bullous pemphigoid (BP)180, CD151, and α6β4-integrin directly interact with the extracellular ligand laminin-332 (
      • Jones J.C.
      • Hopkinson S.B.
      • Goldfinger L.E.
      Structure and assembly of hemidesmosomes.
      ;
      • Borradori L.
      • Sonnenberg A.
      Structure and function of hemidesmosomes: more than simple adhesion complexes.
      ;
      • Sterk L.M.
      • Geuijen C.A.
      • Oomen L.C.
      • et al.
      The tetraspan molecule CD151, a novel constituent of hemidesmosomes, associates with the integrin alpha6beta4 and may regulate the spatial organization of hemidesmosomes.
      . In addition, they contact keratin-associating proteins plectin and BP230 (
      • Green K.J.
      • Virata M.L.
      • Elgart G.W.
      • et al.
      Comparative structural analysis of desmoplakin, bullous pemphigoid antigen and plectin: members of a new gene family involved in organization of intermediate filaments.
      ; Figure 1). The interaction of plectin with α6β4-integrin via the actin-binding domain and with keratins via the plakin repeat domains on the C-terminal site is necessary for HD assembly (
      • Koster J.
      • van W.S.
      • Kuikman I.
      • et al.
      Role of binding of plectin to the integrin beta4 subunit in the assembly of hemidesmosomes.
      .
      Figure thumbnail gr1
      Figure 1Structural organization of hemidesmosomes. Hemidesmosomes connect the intracellular keratin cytoskeleton (K5/K14) with the extracellular matrix. The participating transmembrane proteins bullous pemphigoid (BP)180, CD151, and α6β4-integrin directly interact with the extracellular ligand laminin-332. In addition, they stay in contact with keratin-associating proteins plectin and BP230. Therefore, the interaction of plectin with α6β4-integrin mainly over the actin-binding domain and the association with keratin cytoskeleton by plakin repeat domain are essential for the assembly of hemidesmosomes. Here we focus on the influence of the keratin cytoskeleton on plectin, which binds to the C-terminal site via the intermediate filament (IF)–binding domain to keratins.
      Keratins form the intermediate filament cytoskeleton in all epithelial cells and act as supramolecular scaffolds by interacting with cell-matrix and cell–cell contacts (
      • Fuchs E.
      • Cleveland D.W.
      A structural scaffolding of intermediate filaments in health and disease.
      ;
      • Simpson C.L.
      • Patel D.M.
      • Green K.J.
      Deconstructing the skin: cytoarchitectural determinants of epidermal morphogenesis.
      . Depending on their primary amino-acid sequence, they are grouped into type I and type II keratin gene families with 28 and 26 members, respectively. Most epithelia express between 6 and 10 different keratins as pairs of distinct type I and type II proteins (
      • Schweizer J.
      • Bowden P.E.
      • Coulombe P.A.
      • et al.
      New consensus nomenclature for mammalian keratins.
      ;
      • Magin T.M.
      • Vijayaraj P.
      • Leube R.E.
      Structural and regulatory functions of keratins.
      . In the basal epidermis, the keratin pair K5/K14 is expressed, which upon terminal differentiation becomes sequentially replaced by K1/K10 in suprabasal keratinocytes, where they support cornified envelope formation (
      • Coulombe P.A.
      • Wong P.
      Cytoplasmic intermediate filaments revealed as dynamic and multipurpose scaffolds.
      . In addition to the cytoskeletal function performed by all keratins, certain keratin isotypes act as regulators of cell size, proliferation, protein biosynthesis, and organelle distribution (
      • Gu L.H.
      • Coulombe P.A.
      Keratin function in skin epithelia: a broadening palette with surprising shades.
      ;
      • Vijayaraj P.
      • Kroger C.
      • Reuter U.
      • et al.
      Keratins regulate protein biosynthesis through localization of GLUT1 and -3 upstream of AMP kinase and Raptor.
      . Wound healing and closure are characterized by profound changes in keratin organization and isotype expression. In stratified epithelia, migration into the wound starts from suprabasal epidermal layers (
      • Coulombe P.A.
      Towards a molecular definition of keratinocyte activation after acute injury to stratified epithelia.
      . During reepithelialization, expression of K6, K16, and K17 is induced, accompanied by the downregulation of differentiation-specific K1 and K10. This correlates with alterations in cell morphology and migratory properties (
      • Paladini R.D.
      • Takahashi K.
      • Bravo N.S.
      • et al.
      Onset of re-epithelialization after skin injury correlates with a reorganization of keratin filaments in wound edge keratinocytes: defining a potential role for keratin 16.
      ;
      • Patel G.K.
      • Wilson C.H.
      • Harding K.G.
      • et al.
      Numerous keratinocyte subtypes involved in wound re-epithelialization.
      . In addition, it was shown that alterations in K6, K16, and K17 expression also influence wound healing in vivo and keratinocyte migration ex vivo (
      • Wawersik M.
      • Coulombe P.A.
      Forced expression of keratin 16 alters the adhesion, differentiation, and migration of mouse skin keratinocytes.
      ;
      • Mazzalupo S.
      • Wong P.
      • Martin P.
      • et al.
      Role for keratins 6 and 17 during wound closure in embryonic mouse skin.
      ;
      • Wong P.
      • Coulombe P.A.
      Loss of keratin 6 (K6) proteins reveals a function for intermediate filaments during wound repair.
      . Owing to keratin redundancy, however, isotype-restricted functions of keratins during adhesion and migration, which are critical in wound healing, are not well understood.
      During migration and adhesion, HDs and focal adhesions are continuously remodeled (
      • Geuijen C.A.
      • Sonnenberg A.
      Dynamics of the alpha6beta4 integrin in keratinocytes.
      ;
      • Tsuruta D.
      • Hopkinson S.B.
      • Jones J.C.
      Hemidesmosome protein dynamics in live epithelial cells.
      . In this context, serine phosphorylation by protein kinase-Cα of β4-integrin leads to destabilization and relocalization of hemidesmosomal proteins into lamellipodia (
      • Litjens S.H.
      • de Pereda J.M.
      • Sonnenberg A.
      Current insights into the formation and breakdown of hemidesmosomes.
      . Intriguingly, β4-integrin signaling promotes keratinocyte migration by activation of protein kinase-B/Akt kinase and members of the mitogen-activated protein kinase family (
      • Kippenberger S.
      • Loitsch S.
      • Muller J.
      • et al.
      Ligation of the beta4 integrin triggers adhesion behavior of human keratinocytes by an “inside-out” mechanism.
      ;
      • Nikolopoulos S.N.
      • Blaikie P.
      • Yoshioka T.
      • et al.
      Targeted deletion of the integrin beta4 signaling domain suppresses laminin-5-dependent nuclear entry of mitogen-activated protein kinases and NF-kappaB, causing defects in epidermal growth and migration.
      . Deletion of β4-integrin in mouse keratinocytes enhances migratory properties (
      • Raymond K.
      • Kreft M.
      • Janssen H.
      • et al.
      Keratinocytes display normal proliferation, survival and differentiation in conditional beta4-integrin knockout mice.
      . Moreover, focal adhesions can act as mechanosensors and thereby have a major influence on migration (
      • Schwartz M.A.
      Integrins and extracellular matrix in mechanotransduction.
      .
      Here we address for the first time the role of the keratin cytoskeleton in localization and regulation of HDs during cell adhesion and migration. To that end, we compare murine keratinocyte cell lines lacking the entire set of keratins or reexpress the single keratin pair K5/K14, and investigate HD maintenance and their adhesive and migratory properties in comparison with normal keratinocytes (
      • Vijayaraj P.
      • Kroger C.
      • Reuter U.
      • et al.
      Keratins regulate protein biosynthesis through localization of GLUT1 and -3 upstream of AMP kinase and Raptor.
      .

      Results

      Characterization of keratin-free keratinocytes

      To analyze the functional importance of keratins for keratinocyte morphology, adhesion, and migration, we deleted all type II keratin genes, which at the protein level led to loss of the entire keratin protein family (
      • Vijayaraj P.
      • Kroger C.
      • Reuter U.
      • et al.
      Keratins regulate protein biosynthesis through localization of GLUT1 and -3 upstream of AMP kinase and Raptor.
      . Here we report the first characterization of murine keratinocyte lines without keratins (knockout (KO) cells), one that stably reexpresses the single keratin pair K5/K14, and a wild-type (WT) keratinocyte line (a full description of cells will be presented elsewhere). Absence of all keratins in keratin-free keratinocytes was confirmed by immunofluorescence and western blotting (Figure 2). Rescue cell lines showed colocalization of K5/K14, forming long intermediate filament, whereas K6 was absent (Figure 2a). At the biochemical level, the amount of K5 and K14 reached ∼13% of corresponding amounts in WT cells (Figure 2b). Unexpectedly, loss of keratins had only a slight effect on the morphology of keratinocytes grown under low-calcium conditions (Figure 2c).
      Figure thumbnail gr2
      Figure 2Characterization of keratin-free keratinocytes. (a) Immunofluorescence analyses of the cells stained for keratin 5, keratin 6, and keratin 14 were performed. (b) Absence of keratins in keratin-free cells and reexpression of keratin 5 (K5) in rescued cells were proved by western blotting. (c) Morphology of wild-type (WT), keratin-free, and K5-reexpressing keratinocytes is shown by phase-contrast images. GFP, green fluorescent protein; KO, knockout. Bar=10μm.

      Altered distribution of hemidesmosomal proteins in keratin-free keratinocytes

      To investigate whether the keratin cytoskeleton affected HDs, we analyzed the distribution of plectin, β4-integrin, BP180, BP230, and the extracellular ligand laminin-332 during cell migration, at the migrating front, where HDs display a dynamic behavior, and in the middle of the cell monolayer, where static adhesion prevails. Most remarkably, plectin was completely dissociated from β4-integrin in all settings (Figure 3a). At the migrating front of KO cells, staining for β4-integrin, BP180, BP230, and the extracellular ligand laminin-332 revealed that hemidesmosomal proteins were no longer clustered. Furthermore, β4-integrin had lost its patchy localization in migrating KO cells but maintained its patchy localization in static cells (Figure 3c). At the same time, colocalization of β4-integrin and extracellular laminin-332 was maintained in KO cells (Figure 3b). Similar to its β4-integrin binding partner, α6-integrin failed to colocalize with plectin in KO keratinocytes (Supplementary Figure S1a online). At the same time, distribution of BP180 and BP230 showed a partial colocalization with β4-integrin in WT keratinocytes (Supplementary Figure S1b and c online). In contrast, in KO keratinocytes, the colocalization of cytosolic BP230 with keratins at HD was reduced compared with that of WT cells. Loss of keratins did not influence the partial colocalization of transmembrane BP180 with β4-integrin.
      Figure thumbnail gr3
      Figure 3Altered localization of hemidesmosomes in keratin-free keratinocytes. Immunofluorescence analysis of wild-type (WT), keratin-free (KO), and keratin 5–reexpressing (rescue) keratinocytes stained against keratin 5, plectin, β4-integrin, and laminin-332 to visualize the structures of hemidesmosomes. (a) Intermediate filament (IF) analyses of WT and keratin 5–reexpressing keratinocytes showed characteristic patchy patterns of hemidesmosomal proteins. Note that KO cells have a clustered localization of β4-integrin in the cell layer, but no colocalization with plectin. (b) Staining of extracellular ligand laminin-332 and β4-integrin shows no difference among the three cell types. (c) Keratinocytes immunolabeled for hemidesmosomal proteins on the migratory front. Interestingly, β4-integrin is missing in hemidesmosome structures of migrating keratin-free cells. (d, e) Total cell lysates analyzed by western blotting using (d) β4-integrin and (e) plectin antibodies. (f) Loading control of protein lysates by Coomassie gel. KO, knockout. Bar=10μm.
      Stable reexpression of K5 and K14 in KO keratinocytes reconstituted the typical localization of plectin to β4- and α6-integrin, and clustering of hemidesmosomal proteins, demonstrating keratin dependence of the phenotype (Figure 3). Plectin again colocalized with the keratin cytoskeleton in K5-reexpressing keratinocytes, similar to WT cells. In addition, BP180 and BP230 displayed partial colocalization with hemidesmosomal patches, as seen in WT keratinocytes (Supplementary Figure S1 online). By western blotting, presence or absence of keratins did slightly decrease the amount of β4-integrin but did not affect plectin (Figure 3d and e). Together, these results suggest that the presence of keratins is necessary for the maintenance of plectin in HDs and for the proper clustering of the HDs itself.

      Keratins slow down matrix adhesion of keratinocytes

      In view of the altered distribution and composition of HDs in the absence of keratins, we analyzed keratinocyte adhesion. WT keratinocytes spread and attached completely in ∼3hours (Figure 4a) and displayed HDs, visualized by colocalization of plectin, β4-integrin, and laminin-332 (Supplementary Figure S2 online). In contrast, KO keratinocytes were fully adherent after ∼1hour but lacked complete HDs in which plectin and β4-integrin would colocalize in patches. Both in WT and keratin-free keratinocytes, the spreading started with the formation of membrane protrusions. In this setting, the transmembrane β4-integrin is pushed forward first, followed by plectin. Following primary attachment, β4-integrin and laminin-332 became incorporated into nascent HDs. Quantification of the adhesion revealed remarkable differences between WT and KO keratinocytes, underlining the faster adhesion of the latter (Figure 4b). Similar to WT keratinocytes, green fluorescent protein (GFP)–K5-reexpressing cells adhered completely in ∼3hours. Thus, GFP-K5-reexpressing cells have adhesion properties similar to those of WT cells. In summary, absence of keratins supports the faster adhesion of keratinocytes.
      Figure thumbnail gr4
      Figure 4Keratin-free cells adhere much faster. Results of immunofluorescence analyses of wild-type (WT) and keratin-free keratinocytes in the course of the adhesion process after 30minutes, and 1, 2, and 3hours are shown. Therefore, cells were immunolabeled for hemidesmosomal proteins plectin, β4-integrin, the extracellular ligand laminin-332, and keratin 5. (a) Hemidesmosomal structures visualized by plectin and β4-integrin can be seen after 3hours in WT keratinocytes. In contrast, keratin-free cells are already adherent after 1hour. Moreover, plectin and β4-integrin colocalization is missing in keratin-free keratinocytes. (b) Quantitative analysis of nonadherent cells on specific time points. KO, knockout. Bar=10μm.

      Keratins repress cell migration in keratinocytes

      Cell-matrix contacts have a vital role in cell motility processes. Thus, we further analyzed whether the altered composition and distribution of HDs affected migration properties of WT and KO keratinocytes, applying an established in vitro gap closure assay. KO keratinocytes migrated nearly twofold faster than WT cells, taking ∼16hours to close the gap, whereas WT cells required ∼28hours (Figure 5a and b). The difference became noticeable already ∼8hours into the assay. Next, we directly compared migration of KO and WT keratinocytes using time-lapse video microscopy. Tracking of individual cells to determine directionality of migration revealed differences between WT and KO keratinocytes, as displayed in vector diagrams (Figure 5c). Further, the mean values of distance and velocity of individual cells were determined (Figure 5d and e). Intriguingly, KO cells migrated faster and displayed an altered directionality compared with WT cells.
      Figure thumbnail gr5
      Figure 5In vitro wound-healing assay of wild-type (WT) and keratin-free keratinocytes. (a) The cell gap was closed after 28hours by WT keratinocytes, whereas keratin-free keratinocytes closed the gap already after 16hours. (b) Wound areas were measured and plotted against the time point. (c) Vector diagram of in vitro wound-healing assay of WT (black) versus keratin-free keratinocytes (red) depicting migration tracks of 10 individual cells by video analysis. Axes of vector diagrams=150μm. (d) The average traveled distance of WT and keratin-free keratinocytes indicates a longer migrated way of keratin-free cells. (e) The average velocity of keratin-free cells is enhanced as well. Values are mean±SEM of three independent experiments. *P<0.01. KO, knockout. Bar=10μm.

      Genetic rescue confirms keratin dependence

      Given the complexity of genetic and epigenetic alterations in KO and control keratinocytes, we were concerned whether changes in plectin distribution, adhesion, and migration resulted from the keratin status of both cell lines or from other experimental manipulations. To address this, we conducted gain-of-function experiments by stable expression of GFP-K5 in KO keratinocytes, using lentiviral vectors. This leads to a stabilization of endogenous K14 (Figure 2a). As a control, KO cells were transfected with the same vector expressing only GFP. Subsequent experiments were conducted with cell lines homogenously expressing either GFP-K5 or GFP only. Therefore, we investigated migratory properties of K5-reexpressing keratinocytes versus KO keratinocytes, using the established in vitro gap closure assay (Figure 6a). Consistent with a major role of keratins in restricting migration, K5-reexpressing KO keratinocytes closed the gap in 28hours similar to WT cells. In support, KO cells expressing only GFP closed the gap in 16hours. Next, video analysis of migrating K5-reexpressing keratinocytes versus GFP control keratinocytes in the same approach was performed (Figure 6b). K5-reexpressing cells are illustrated by red lines and GFP control cells by black lines. This revealed that K5-reexpressing cells displayed an altered directionality, restoring the mode of WT cells. Moreover, reexpression of K5 reduced the distance and velocity of migrating cells (Figure 6c and d). Collectively, our data show that the single keratin pair K5 and K14 is sufficient to maintain keratinocyte adhesion and migration, and to localize plectin to β4-integrin.
      Figure thumbnail gr6
      Figure 6Functional properties of rescued keratin 5 (K5)–reexpressing keratinocytes. Cells are either transfected with a green fluorescent protein (GFP)–K5 or a GFP-control vector. (a) Adhesion assay on laminin-332. 100,000 Cells were seeded and their adhesion was monitored over 120min. (b) In vitro wound-healing assay of K5-reexpressing and GFP-control keratinocytes. GFP-control keratinocytes close the cell gap after 16hours, whereas GFP-K5 cells need 26hours. (c) Wound areas were measured and plotted against the time point. (d) Vector diagram illustrating tracked individual cells of in vitro wound-healing assay of K5-reexpressing cells (red) versus GFP control keratinocytes (black) by video analysis. Axes of vector diagrams=100μm. (e) Average of the traveled distance and velocity (f) of tracked GFP-K5 and GFP control keratinocytes. Values are mean±SEM of three independent experiments. *P<0.01. Bar=10μm.
      In summary, we demonstrate a requirement of the keratin cytoskeleton in the maintenance and distribution of HDs upstream of plectin, which has implications for keratinocyte adhesion and migration.

      Discussion

      The role of keratins in HD maintenance, cell adhesion, and migration has been understood only partially. Comparing keratinocytes that either lack the entire keratin protein family or express the single keratin pair K5 and K14 with normal keratinocytes, we uncovered a new role of keratins in HD integrity, which is mediated through plectin and β4-integrin.

      Keratins are necessary to maintain HDs

      Our data provide the first evidence for a crucial role of keratins in the localization of hemidesmosomal proteins, based on the observation that loss of keratins caused an altered distribution of plectin and β4-integrin. In line with the established function of plectin in HD assembly and maintenance, our data support a novel role of keratins acting through plectin (
      • Koster J.
      • Geerts D.
      • Favre B.
      • et al.
      Analysis of the interactions between BP180, BP230, plectin and the integrin alpha6beta4 important for hemidesmosome assembly.
      . In apparent contrast, plectin KO keratinocytes displayed an altered keratin filament network (
      • Osmanagic-Myers S.
      • Gregor M.
      • Walko G.
      • et al.
      Plectin-controlled keratin cytoarchitecture affects MAP kinases involved in cellular stress response and migration.
      . Possibly, lack of keratins triggers a conformational change of plectin at its C-terminal intermediate filament–binding site, which might inhibit the N-terminal binding to β4-integrin (
      • de Pereda J.M.
      • Lillo M.P.
      • Sonnenberg A.
      Structural basis of the interaction between integrin alpha6beta4 and plectin at the hemidesmosomes.
      . At the same time, the interaction between β4-integrin and laminin-332 was unaffected in KO keratinocytes, which is consistent with the binding to laminin-332 of a β4-integrin mutant unable to bind plectin (
      • Geuijen C.A.
      • Sonnenberg A.
      Dynamics of the alpha6beta4 integrin in keratinocytes.
      . The altered distribution of β4-integrin in KO keratinocytes could be caused by an enhanced endocytosis and recycling of cell-matrix contacts during migration (
      • Ridley A.J.
      • Schwartz M.A.
      • Burridge K.
      • et al.
      Cell migration: integrating signals from front to back.
      . Our data are not at odds with the maintenance of core HDs in epidermolysis bullosa simplex patients or in K5 and K14 KO mice (
      • Lloyd C.
      • Yu Q.C.
      • Cheng J.
      • et al.
      The basal keratin network of stratified squamous epithelia: defining K15 function in the absence of K14.
      ;
      • Peters B.
      • Kirfel J.
      • Bussow H.
      • et al.
      Complete cytolysis and neonatal lethality in keratin 5 knockout mice reveal its fundamental role in skin integrity and in epidermolysis bullosa simplex.
      ;
      • Fine J.D.
      Inherited epidermolysis bullosa: past, present, and future.
      . In these settings, residual keratins are still expressed, whereas here keratins are absent. Collectively, our results show that keratins are necessary for the maintenance of intact HDs, most likely upstream of plectin. It is noteworthy that the single keratin pair K5 and K14 was sufficient for rescuing this function. Determining whether maintenance of HDs relies directly or indirectly on keratins requires more detailed investigation.

      Keratins negatively regulate adhesion and migration

      HDs provide stable attachment, and therefore localization of HD is necessary for proper adhesion of keratinocytes (
      • Jones J.C.
      • Green K.J.
      Intermediate filament-plasma membrane interactions.
      ;
      • Borradori L.
      • Sonnenberg A.
      Structure and function of hemidesmosomes: more than simple adhesion complexes.
      ;
      • Litjens S.H.
      • de Pereda J.M.
      • Sonnenberg A.
      Current insights into the formation and breakdown of hemidesmosomes.
      . Altered distribution and dynamics of hemidesmosomal proteins can result in a faster adhesion of keratin-free cells, consistent with the reduced, plectin-dependent clustering of β4-integrin that was observed in dynamic, i.e., migrating, keratinocytes. In fact, KO keratinocytes adhered faster than WT cells, migrated twice as fast, and showed a reduced directionality during migration. Mechanistically, this could be because of an increased turnover of HDs in KO keratinocytes. Alternatively, altered distribution of HD might affect the distribution and functionality of focal adhesions, possibly through plectin (
      • Ozawa T.
      • Tsuruta D.
      • Jones J.C.
      • et al.
      Dynamic relationship of focal contacts and hemidesmosome protein complexes in live cells.
      ;
      • Tsuruta D.
      • Hashimoto T.
      • Hamill K.J.
      • et al.
      Hemidesmosomes and focal contact proteins: functions and cross-talk in keratinocytes, bullous diseases and wound healing.
      . Further experiments are needed to clarify the mechanism.
      The first hemidesmosomal protein pushed forward along the cell membrane by lamellipodia formation is β4-integrin, followed by plectin. In KO keratinocytes, colocalization of β4-integrin and plectin did not take place. However, laminin-332 was established in hemidesmosomal patches. Although a previous study suggested that α6β4-integrin forms hemidesmosomal clusters depending on laminin-332 deposition along the cell surface (
      • Geuijen C.A.
      • Sonnenberg A.
      Dynamics of the alpha6beta4 integrin in keratinocytes.
      , our data support a novel role of keratins during this process.
      Increased cell motility in KO keratinocytes can be caused by several mechanisms including altered localization of hemidesmosomal proteins. In support, both plectin and β4-integrin KO keratinocytes displayed enhanced migration (
      • Raymond K.
      • Kreft M.
      • Janssen H.
      • et al.
      Keratinocytes display normal proliferation, survival and differentiation in conditional beta4-integrin knockout mice.
      ;
      • Osmanagic-Myers S.
      • Gregor M.
      • Walko G.
      • et al.
      Plectin-controlled keratin cytoarchitecture affects MAP kinases involved in cellular stress response and migration.
      . Relocalization of BP proteins, BP180 and BP230, can also contribute to a migratory phenotype (
      • Guo L.
      • Degenstein L.
      • Dowling J.
      • et al.
      Gene targeting of BPAG1: abnormalities in mechanical strength and cell migration in stratified epithelia and neurologic degeneration.
      ;
      • Tasanen K.
      • Tunggal L.
      • Chometon G.
      • et al.
      Keratinocytes from patients lacking collagen XVII display a migratory phenotype.
      . Furthermore, α6β4-integrin was shown to stabilize lamellipodia and therefore affects cell motility (
      • Rabinovitz I.
      • Mercurio A.M.
      The integrin alpha6beta4 functions in carcinoma cell migration on laminin-1 by mediating the formation and stabilization of actin-containing motility structures.
      . In addition, it was suggested that clustering of β4-integrin and its interaction with laminin-332 determine whether β4-integrin inhibits cell migration (
      • Geuijen C.A.
      • Sonnenberg A.
      Dynamics of the alpha6beta4 integrin in keratinocytes.
      . Recent data support a role of HDs as signaling platforms in which α6β4-integrin signaling is mediated through plectin. Plectin interacts with RACK1, which is linked to the protein kinase-Cδ signaling pathway (
      • Osmanagic-Myers S.
      • Gregor M.
      • Walko G.
      • et al.
      Plectin-controlled keratin cytoarchitecture affects MAP kinases involved in cellular stress response and migration.
      . Furthermore, a role of HDs in upstream signaling of the kinases, protein kinase-B and extracellelar signal-regulated kinase, was shown (
      • Kippenberger S.
      • Hofmann M.
      • Zoller N.
      • et al.
      Ligation of beta4 integrins activates PKB/Akt and ERK1/2 by distinct pathways-relevance of the keratin filament.
      . It is conceivable that loss of the keratin cytoskeleton, and therefore decreased maintenance of HDs, also influences intracellular signaling pathways, which affects the migration of keratinocytes. The fact that plectin, in addition to keratins, additionally interacts with the actin cytoskeleton provides an alternative hypothesis. Intriguingly, it cannot bind to β4-integrin and actin at the same time, as both bind through the actin-binding domain (
      • de Pereda J.M.
      • Lillo M.P.
      • Sonnenberg A.
      Structural basis of the interaction between integrin alpha6beta4 and plectin at the hemidesmosomes.
      . Therefore, loss of keratins might affect plectin’s interaction with the actin cytoskeleton, suggesting enhanced adhesion and migration of keratinocytes through additional mechanisms. In support of our data, high expression of K16 in cultured skin explants was accompanied by decreased migration (
      • Wawersik M.
      • Coulombe P.A.
      Forced expression of keratin 16 alters the adhesion, differentiation, and migration of mouse skin keratinocytes.
      . In contrast, K17-null embryos showed delayed wound closure (
      • Mazzalupo S.
      • Wong P.
      • Martin P.
      • et al.
      Role for keratins 6 and 17 during wound closure in embryonic mouse skin.
      . Moreover, skin explant cultures of K6α/K6β KO mice migrated faster compared with WT counterparts (
      • Wong P.
      • Coulombe P.A.
      Loss of keratin 6 (K6) proteins reveals a function for intermediate filaments during wound repair.
      . The apparent differences in previous experiments are resolved by our study, which precludes compensation by other keratins.
      The current study provides the first evidence for a major role of keratins in HD maintenance through plectin, with implications for cell adhesion and migration. An attractive theory to be tested is whether the keratin–HD complex has a role in mechanotransduction.

      Materials and Methods

      Cell culture

      Isolation of primary mouse keratinocytes of KO and WT cells has been described elsewhere (
      • Vijayaraj P.
      • Kroger C.
      • Reuter U.
      • et al.
      Keratins regulate protein biosynthesis through localization of GLUT1 and -3 upstream of AMP kinase and Raptor.
      . Keratinocytes were grown in FAD Medium Chelex-treated (
      • Brennan J.K.
      • Mansky J.
      • Roberts G.
      • et al.
      Improved methods for reducing calcium and magnesium concentrations in tissue culture medium: application to studies of lymphoblast proliferation in vitro.
      ; FAD low Ca, Biochrom, Berlin, Germany), supplemented with 10% FCS Gold (PAA, Pasching, Austria), 0.18mM adenine, 0.5μgml−1 hydrocortison, 5μgml−1 insulin, 100pM choleratoxin (Sigma, St. Louis, MO), 10ngml−1 EGF, 100Uml−1, and 100μgml−1 penicillin/streptomycin, and 2mM glutamax (Invitrogen, Darmstadt, Germany), in 5% CO2 at 32°C. Cells were cultured on collagen I (rat tail, Invitrogen)–coated cell culture dishes. Keratin-free keratinocytes stably expressing GFP-K5 and GFP were generated by lentiviral transduction essentially as described (
      • Stöhr N.
      • Köhn M.
      • Lederer M.
      • et al.
      IGF2BP1 promotes cell migration by regulating MK5 and PTEN signaling.
      .

      Antibodies

      Immunofluorescence staining

      Cells were fixed for 5minutes in -20°C methanol and for 30seconds in -20°C acetone. Cells were stained with the primary antibody and incubated for 1hour. All antibodies were diluted in Tris-buffered saline containing 1% BSA. Thereafter, cells were incubated with the secondary antibody for 30minutes and mounted with mounting medium (Dianova, Hamburg, Germany). Images were acquired using an AxioImager Z2 equipped with Zeiss Plan-Apochromat × 63/1.4 oil and recorded with an AxiocamMR camera and an Axioplan 2 microscope (Carl Zeiss, Goettingen, Germany). Image analysis and processing were carried out using AxioVision 4.8 software. Samples were analyzed using a fluorescence laser-scanning confocal microscope (LSM 780, Carl Zeiss). Each fluorochrome was scanned individually in single optical sections (“sequential scan”) to avoid cross-talk between channels. For confocal images, Pinhole “airy 1” Zeiss standard settings were used to receive signals only from the focal plane. Analysis and processing of acquired images were carried out using Zen software (Carl Zeiss). Images were cropped and analyzed in Adobe Photoshop CS5 software, and Adobe Illustrator CS5 was used for figure design.

      In vitro wound-healing assay

      For the in vitro wound-healing assay, an Ibidi culture insert was used to create a defined 500μm cell gap (Ibidi, Martinsried, Germany). Briefly, cells were grown in the Ibidi culture insert to confluency on a collagen-coated dish. Migration of cells was analyzed on a Nikon Eclipse Ti-S inverted microscope (Nikon, Melville, NY). Images were taken every 2hours over a total period of 30hours. Images were collected, and percentage of wound closure was determined by digital analysis. In addition, wound closure was monitored by time-lapse video microscopy (Nikon Eclipse Ti-microscope). Images were taken at 15-minute intervals over 24–48hours. Ten single cells per cell type were tracked using Image J, and the mean values of migratory distance and velocity were calculated. Statistical analyses and significance were determined using SigmaPlot 11 software.

      Western blot analysis

      ACKNOWLEDGMENTS

      Work in the Magin lab is supported by the DFG (MA1316-9/3, MA1316-15, INST 268/230-1), the BMBF (network EB), and the Translational Center for Regenerative Medicine, TRM, Leipzig, No. 0315883). We thank the Core Facility Imaging for support with live cell imaging and cell migration analyses.
      Author Contributions
      K.S. conceived the study, designed and conducted the experiments, analyzed data, prepared the figures, and wrote the paper. W.R. designed and assisted in the experiments. F.L. and C.K. performed immunofluorescence analysis and immunoblots to characterize keratinocytes. M.L. and S.H. generated GFP-K5 and GFP control keratin-free keratinocytes stable cell lines. T.M.M. contributed expertise, designed experiments, and wrote the paper.

      SUPPLEMENTARY MATERIAL

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

      REFERENCES

        • Andra K.
        • Kornacker I.
        • Jorgl A.
        • et al.
        Plectin-isoform-specific rescue of hemidesmosomal defects in plectin (−/−) keratinocytes.
        J Invest Dermatol. 2003; 120: 189-197
        • Borradori L.
        • Sonnenberg A.
        Structure and function of hemidesmosomes: more than simple adhesion complexes.
        J Invest Dermatol. 1999; 112: 411-418
        • Brennan J.K.
        • Mansky J.
        • Roberts G.
        • et al.
        Improved methods for reducing calcium and magnesium concentrations in tissue culture medium: application to studies of lymphoblast proliferation in vitro.
        In Vitro. 1975; 11: 354-360
        • Coulombe P.A.
        Towards a molecular definition of keratinocyte activation after acute injury to stratified epithelia.
        Biochem Biophys Res Commun. 1997; 236: 231-238
        • Coulombe P.A.
        • Wong P.
        Cytoplasmic intermediate filaments revealed as dynamic and multipurpose scaffolds.
        Nat Cell Biol. 2004; 6: 699-706
        • de Pereda J.M.
        • Lillo M.P.
        • Sonnenberg A.
        Structural basis of the interaction between integrin alpha6beta4 and plectin at the hemidesmosomes.
        EMBO J. 2009; 28: 1180-1190
        • Fine J.D.
        Inherited epidermolysis bullosa: past, present, and future.
        Ann NY Acad Sci. 2010; 1194: 213-222
        • Fuchs E.
        • Cleveland D.W.
        A structural scaffolding of intermediate filaments in health and disease.
        Science. 1998; 279: 514-519
        • Geuijen C.A.
        • Sonnenberg A.
        Dynamics of the alpha6beta4 integrin in keratinocytes.
        Mol Biol Cell. 2002; 13: 3845-3858
        • Green K.J.
        • Virata M.L.
        • Elgart G.W.
        • et al.
        Comparative structural analysis of desmoplakin, bullous pemphigoid antigen and plectin: members of a new gene family involved in organization of intermediate filaments.
        Int J Biol Macromol. 1992; 14: 145-153
        • Gu L.H.
        • Coulombe P.A.
        Keratin function in skin epithelia: a broadening palette with surprising shades.
        Curr Opin Cell Biol. 2007; 19: 13-23
        • Guo L.
        • Degenstein L.
        • Dowling J.
        • et al.
        Gene targeting of BPAG1: abnormalities in mechanical strength and cell migration in stratified epithelia and neurologic degeneration.
        Cell. 1995; 81: 233-243
        • Jones J.C.
        • Green K.J.
        Intermediate filament-plasma membrane interactions.
        Curr Opin Cell Biol. 1991; 3: 127-132
        • Jones J.C.
        • Hopkinson S.B.
        • Goldfinger L.E.
        Structure and assembly of hemidesmosomes.
        Bioessays. 1998; 20: 488-494
        • Kippenberger S.
        • Hofmann M.
        • Zoller N.
        • et al.
        Ligation of beta4 integrins activates PKB/Akt and ERK1/2 by distinct pathways-relevance of the keratin filament.
        Biochim Biophys Acta. 2010; 1803: 940-950
        • Kippenberger S.
        • Loitsch S.
        • Muller J.
        • et al.
        Ligation of the beta4 integrin triggers adhesion behavior of human keratinocytes by an “inside-out” mechanism.
        J Invest Dermatol. 2004; 123: 444-451
        • Koster J.
        • Geerts D.
        • Favre B.
        • et al.
        Analysis of the interactions between BP180, BP230, plectin and the integrin alpha6beta4 important for hemidesmosome assembly.
        J Cell Sci. 2003; 116: 387-399
        • Koster J.
        • van W.S.
        • Kuikman I.
        • et al.
        Role of binding of plectin to the integrin beta4 subunit in the assembly of hemidesmosomes.
        Mol Biol Cell. 2004; 15: 1211-1223
        • Litjens S.H.
        • de Pereda J.M.
        • Sonnenberg A.
        Current insights into the formation and breakdown of hemidesmosomes.
        Trends Cell Biol. 2006; 16: 376-383
        • Lloyd C.
        • Yu Q.C.
        • Cheng J.
        • et al.
        The basal keratin network of stratified squamous epithelia: defining K15 function in the absence of K14.
        J Cell Biol. 1995; 129: 1329-1344
        • Magin T.M.
        • Vijayaraj P.
        • Leube R.E.
        Structural and regulatory functions of keratins.
        Exp Cell Res. 2007; 313: 2021-2032
        • Mazzalupo S.
        • Wong P.
        • Martin P.
        • et al.
        Role for keratins 6 and 17 during wound closure in embryonic mouse skin.
        Dev Dyn. 2003; 226: 356-365
        • Nikolopoulos S.N.
        • Blaikie P.
        • Yoshioka T.
        • et al.
        Targeted deletion of the integrin beta4 signaling domain suppresses laminin-5-dependent nuclear entry of mitogen-activated protein kinases and NF-kappaB, causing defects in epidermal growth and migration.
        Mol Cell Biol. 2005; 25: 6090-6102
        • Osmanagic-Myers S.
        • Gregor M.
        • Walko G.
        • et al.
        Plectin-controlled keratin cytoarchitecture affects MAP kinases involved in cellular stress response and migration.
        J Cell Biol. 2006; 174: 557-568
        • Ozawa T.
        • Tsuruta D.
        • Jones J.C.
        • et al.
        Dynamic relationship of focal contacts and hemidesmosome protein complexes in live cells.
        J Invest Dermatol. 2010; 130: 1624-1635
        • Paladini R.D.
        • Takahashi K.
        • Bravo N.S.
        • et al.
        Onset of re-epithelialization after skin injury correlates with a reorganization of keratin filaments in wound edge keratinocytes: defining a potential role for keratin 16.
        J Cell Biol. 1996; 132: 381-397
        • Patel G.K.
        • Wilson C.H.
        • Harding K.G.
        • et al.
        Numerous keratinocyte subtypes involved in wound re-epithelialization.
        J Invest Dermatol. 2006; 126: 497-502
        • Peters B.
        • Kirfel J.
        • Bussow H.
        • et al.
        Complete cytolysis and neonatal lethality in keratin 5 knockout mice reveal its fundamental role in skin integrity and in epidermolysis bullosa simplex.
        Mol Biol Cell. 2001; 12: 1775-1789
        • Rabinovitz I.
        • Mercurio A.M.
        The integrin alpha6beta4 functions in carcinoma cell migration on laminin-1 by mediating the formation and stabilization of actin-containing motility structures.
        J Cell Biol. 1997; 139: 1873-1884
        • Raymond K.
        • Kreft M.
        • Janssen H.
        • et al.
        Keratinocytes display normal proliferation, survival and differentiation in conditional beta4-integrin knockout mice.
        J Cell Sci. 2005; 118: 1045-1060
        • Ridley A.J.
        • Schwartz M.A.
        • Burridge K.
        • et al.
        Cell migration: integrating signals from front to back.
        Science. 2003; 302: 1704-1709
        • Schwartz M.A.
        Integrins and extracellular matrix in mechanotransduction.
        Cold Spring Harb Perspect Biol. 2010; 2: a005066
        • Schweizer J.
        • Bowden P.E.
        • Coulombe P.A.
        • et al.
        New consensus nomenclature for mammalian keratins.
        J Cell Biol. 2006; 174: 169-174
        • Simpson C.L.
        • Patel D.M.
        • Green K.J.
        Deconstructing the skin: cytoarchitectural determinants of epidermal morphogenesis.
        Nat Rev Mol Cell Biol. 2011; 12: 565-580
        • Sterk L.M.
        • Geuijen C.A.
        • Oomen L.C.
        • et al.
        The tetraspan molecule CD151, a novel constituent of hemidesmosomes, associates with the integrin alpha6beta4 and may regulate the spatial organization of hemidesmosomes.
        J Cell Biol. 2000; 149: 969-982
        • Stöhr N.
        • Köhn M.
        • Lederer M.
        • et al.
        IGF2BP1 promotes cell migration by regulating MK5 and PTEN signaling.
        Genes Dev. 2012; 26: 176-189
        • Tasanen K.
        • Tunggal L.
        • Chometon G.
        • et al.
        Keratinocytes from patients lacking collagen XVII display a migratory phenotype.
        Am J Pathol. 2004; 164: 2027-2038
        • Tsuruta D.
        • Hashimoto T.
        • Hamill K.J.
        • et al.
        Hemidesmosomes and focal contact proteins: functions and cross-talk in keratinocytes, bullous diseases and wound healing.
        J Dermatol Sci. 2011; 62: 1-7
        • Tsuruta D.
        • Hopkinson S.B.
        • Jones J.C.
        Hemidesmosome protein dynamics in live epithelial cells.
        Cell Motil Cytoskeleton. 2003; 54: 122-134
        • Vijayaraj P.
        • Kroger C.
        • Reuter U.
        • et al.
        Keratins regulate protein biosynthesis through localization of GLUT1 and -3 upstream of AMP kinase and Raptor.
        J Cell Biol. 2009; 187: 175-184
        • Wawersik M.
        • Coulombe P.A.
        Forced expression of keratin 16 alters the adhesion, differentiation, and migration of mouse skin keratinocytes.
        Mol Biol Cell. 2000; 11: 3315-3327
        • Wong P.
        • Coulombe P.A.
        Loss of keratin 6 (K6) proteins reveals a function for intermediate filaments during wound repair.
        J Cell Biol. 2003; 163: 327-337