Role of Mechanotransduction in Periodontal Homeostasis and Disease.

Novel findings broaden the concept of mechanotransduction (MT) in biophysically stimulated tissues such as the periodontium by considering nuclear MT, convergence of intracellular MT pathways, and mechanoresponsive cotranscription factors such as Yes-associated protein 1 (YAP1). Regarding periodontal disease, recent studies have elucidated the role of bacterial gingipain proteases in disturbing the barrier function of cadherins, thereby promoting periodontal inflammation. This leads to dysregulation of extracellular matrix homeostasis via proteases and changes the cell's biophysical environment, which leads to alterations in MT-induced cell behavior and loss of periodontal integrity. Newest experimental evidence from periodontal ligament cells suggests that the Hippo signaling protein YAP1, in addition to integrin-FAK (focal adhesion kinase) mechanosignaling, also regulates cell stemness. By addressing mechanosignaling-dependent transcription factors, YAP1 is involved in osteogenic and myofibroblast differentiation and influences core steps of autophagy. Recent in vivo evidence elucidates the decisive role of YAP1 in epithelial homeostasis and underlines its impact on oral pathologies, such as periodontitis-linked oral squamous cell carcinogenesis. Here, new insights reveal that YAP1 contributes to carcinogenesis via overexpression rather than mutation; promotes processes such as apoptosis resistance, epithelial-mesenchymal transition, or metastasis; and correlates with poor prognosis in oral squamous cell carcinoma. Furthermore, YAP1 has been shown to contribute to periodontitis-induced bone loss. Mechanistically, molecules identified to regulate YAP1-related periodontal homeostasis and disease include cellular key players such as MAPK (mitogen-activated protein kinase), JNK (c-Jun N-terminal kinase), Rho (Ras homologue) and ROCK (Rho kinase), Bcl-2 (B-cell lymphoma 2), AP-1 (activator protein 1), and c-myc (cellular myelocytomatosis). These findings qualify YAP1 as a master regulator of mechanobiology and cell behavior in human periodontal tissues. This review summarizes the most recent developments in MT-related periodontal research, thereby offering insights into outstanding research questions and potential applications of molecular or biophysical strategies aiming at periodontal disease mitigation or prevention.

N-Acetylcysteine modulates the effects of composites on human gingival keratinocytes.

OBJECTIVE: The aim of this study was to evaluate, if antioxidants, like N-Acetylcysteine, can modulate effects of composite eluates on human gingival keratinocytes. METHODS: Composite samples of ceram.x® universal, Filtek™ Supreme XTE, and Admira® Fusion were stored 72h in cell culture medium to prepare eluates, according to ISO 10993-12:2012. Human gingival keratinocytes were exposed to these eluates with or without 3mM N-Acetylcysteine. Following cell observation by iCELLigence®, exposure periods were determined at 1d and 4d. Cell morphological analysis combined with live/dead staining was performed. Tissue-specific biomarkers of terminal differentiation, Involucrin and Filaggrin, were analyzed by indirect immunofluorescence (IIF) and Western blot (WB). qPCR profiling was performed on genes encoding for: inflammation, apoptosis, turn-over of extracellular matrix, adhesion, proliferation and differentiation. For statistical analysis one-way Anova was used (p<0.05). RESULTS: Cells exposed to N-Acetylcysteine exhibited morphological changes but no cell death. After adding 3mM N-Acetylcysteine to HGK cultures, increased fluorescence intensity and protein amounts of Involucrin and Filaggrin indicated enhanced differentiation (p<0.05). Gene expression was modulated by: (i) composition of the composite eluates, (ii) NAC and (iii) exposure time. Filtek™ Supreme XTE showed a significant increased gene expression in inflammatory genes (p<0.05), which was amplified by the addition of NAC at 1d. Concerning exposure time, modulated gene expression showed eluate dependency, substantiated by Filtek™ Supreme XTE modulation at day 1 and Admira® Fusion at day 4. SIGNIFICANCE: N-Acetylcysteine-emerging effects on gingival keratinocytes were threefold: (i) increase of differentiation, (ii) modulation of composite-related effects and (iii) in parts counteraction of eluate-induced effects.

Comparative differentiation analysis of distinct oral tissue-derived cells in response to osteogenic stimulation.

BACKGROUND: Mixed cell populations from oral tissues may be superior to pure stem cells for regenerative approaches. Therefore, the aim of the present study was to explore the osteogenic potential of mixed cells derived from oral connective tissues compared to alveolar osteoblasts. MATERIALS AND METHODS: Primary cells were isolated from the alveolar bone, periodontal ligament and gingiva. Following characterization by colony formation, growth capacity and flow cytometry, all cells were subjected to osteogenic differentiation induction and screened for a large panel of osteogenic markers using western blots, qPCR arrays, and matrix mineralization and alkaline phosphatase quantification. RESULTS: Non-induced mixed cells from gingiva showed higher colony formation efficiency but decreased proliferation compared to non-induced periodontal mixed cells, while both entities revealed similar surface markers tested in this setup. Following osteogenic induction, all cell populations individually expressed receptors with distinctively activated downstream effectors. Gene expression of induced periodontal mixed cells was similar to alveolar osteoblasts, but was differently modulated in gingival mixed cells. The latter failed to achieve osteogenic differentiation in terms of matrix mineralization and alkaline phosphatase activity, which was well observed in periodontal mixed cells and osteoblasts. CONCLUSION: Mixed cells from periodontal ligament but not from gingiva feature an inherent osteogenic capacity in vitro. From these results, it can be concluded that periodontal cells do not require further stem cell enrichment in order to qualify for bone regeneration. CLINICAL RELEVANCE: Our data contribute to the development of novel cell-based therapies using mixed cells from the periodontal ligament in regenerative periodontics.

Modulation of gingival cell response towards dental composites.

OBJECTIVE: This aim of this study was to investigate the cell behavioural response on clinically applied dental composites in exposition-relevant target cells, i.e. human gingival fibroblasts (HGF) and epithelial keratinocytes (HGK). METHODS: HGF and HGK were exposed to eluates of Ceram X™, Filtek™ Silorane, Filtek™ Supreme XTE, Fusio™ Liquid Dentin and Vertise™ Flow. Eluates were created by storing material samples in respective cell culture medium, for 24h and 72h (n=17), according to ISO 10993-12:2012. Cell response was evaluated at eluate exposure periods of 24h and 72h by (i) impedance analysis-based real-time monitoring of adhesion and proliferation, (ii) semi-quantitative indirect immunofluorescence (sq-IIF) detection of tissue-specific biomarkers, and (iii) ELISA-detection of pro-inflammatory interleukin (IL)-6. RESULTS: Generally, cell behavioural response towards the eluates was gradual in HGK and HGF, the latter exhibiting a less pronounced modulation per se. In HGK, ERK 1/2 was mainly activated after 24h by Fusio™ Liquid Dentin and Vertise™ Flow, while an increase in biomarker expression occurred time-delayed. A 72h exposure of HGK to eluates of Filtek™ Supreme XTE, Fusio™ Liquid Dentin and Vertise™ Flow significantly decreased secreted IL-6 amounts. In HGK, the impedance analysis revealed less proliferation and/or adhesion in case of Fusio™ Liquid Dentin and Vertise™ Flow with matched other composites. SIGNIFICANCE: In detail, protein expression and secretion is modulated particularly in terms of signal transduction, differentiation and inflammation. On cell biological level, all tested materials modulated the analysed features of cell behaviour with emphasis on the self-adhering composites.

Environmental biomechanics substantiated by defined pillar micropatterns govern behavior of human mesenchymal stem cells.

While evidence on the impact of the biomechanical environment elasticity on human mesenchymal stem cell (hMSC) behavior is growing, the aspect of micropatterning is still poorly understood. Thus, the present study aimed at investigating the influence of defined environmental micropatterning on hMSC behavior. Following characterization, hMSCs were grown on defined pillar micropatterns of 5, 7, 9, and 11 µm. With respect to cell behavior, primary hMSC adhesion was detected by indirect immunofluorescence (iIF) for paxillin, vinculin, integrin αV, and actin, while proliferation was visualized by histone H3. Morphogenesis was monitored by scanning electron microscopy and the expression of stem cell-specific biomarkers by real-time PCR. Favoritism of primary adhesion of hMSCs on pillar tops occurred at smaller pillar micropatterns, concomitant with cell flattening. While vinculin, integrin αV, and paxillin appeared initially more cytoplasmic, high pillar micropatterns favored a progressive redistribution with polarization to cell tension sites and at cell borders. Accomplishment of morphogenesis at day 3 revealed establishment of fully rotund cell somata at 5 µm, while hMSCs appeared progressively elongated at rising micropatterns. The hMSC proliferation capacity was influenced by pillar micropatterns and gene expression analysis of stem cell- and differentiation-associated biomarkers disclosed clear modulation by distinct pillar micropatterns. In response to environmental biomechanics, our results show that hMSC behavior is governed by pillar micropatterning. In turn, these findings may form the basis to prospectively direct lineage specificity of hMSCs in a customized fashion.

Elevation of collagen type I in fibroblast-keratinocyte cocultures emphasizes the decisive role of fibroblasts in the manifestation of the phenotype of cyclosporin A-induced gingival overgrowth.

BACKGROUND AND OBJECTIVE: Collagen type I elevation in cyclosporin A-induced gingival overgrowth supports evidence that gingival fibroblasts play a decisive role in the manifestation of the phenotype. To analyze the role of gingival fibroblasts under more in vivo-like conditions, we evaluated the effect of cyclosporin A on collagen type I gene and protein expression in gingival overgrowth-derived gingival fibroblasts established as cocultures with gingival keratinocytes as well as in matched gingival fibroblast monolayers. MATERIAL AND METHODS: Monolayers and cocultures of primary gingival fibroblasts were treated with cyclosporin A for 6 and 72 h. The expression of collagen type I mRNA was analyzed by quantitative real time polymerase chain reaction, while expression and secretion of collagen type I protein was analyzed by indirect immunofluorescence and western blotting. RESULTS: Compared with controls, significant elevation of collagen type I mRNA was restricted to cocultures after 6 and 72 h of treatment with cyclosporin A. In keratinocytes, collagen type I remained undetectable. In monolayers and cocultures, indirect immunofluorescence showed a slightly higher level of collagen type I protein in gingival fibroblasts in response to stimulation with cyclosporin A. Semiquantitative detection of collagen type I by western blotting demonstrated a nonsignificant increase for cell extracts in monolayers and cocultures. For secreted collagen type I, western blot analysis of the supernatants revealed elevated protein levels in cultures stimulated with cyclosporin A. Compared with the corresponding monolayers, the stimulatory effect of cyclosporin A on protein secretion was significant only in coculture. CONCLUSION: Our results indicate that collagen type I is a target of cyclosporin A and that gingival fibroblasts are decisive for the manifestation of the gingival overgrowth-phenotype. Furthermore, the results suggest that cocultures of gingival overgrowth-derived gingival fibroblasts and gingival keratinocytes permit analysis of cyclosporin A-induced effects under more in vivo-like conditions.

Analysis of interleukin-1beta-modulated mRNA gene transcription in human gingival keratinocytes by epithelia-specific cDNA microarrays.

BACKGROUND/OBJECTIVES: Proinflammatory cytokines such as interleukin-1beta are known to be synthesized in oral gingivitis and periodontitis and lead to the activation of the transcription factor nuclear factor-kappaB (NF-kappaB). Although numerous effects of interleukin-1beta on mesenchymal cells are known, e.g. up-regulation of intercellular adhesion molecule-1 in endothelial cells, little is known of the effects of interleukin-1beta on oral keratinocytes. The purpose of the present study was to seek interleukin-1beta-mediated alterations in mRNA gene transcription and a putative activation of NF-kappaB in oral gingival keratinocytes. METHODS: As an in vitro model for gingivitis and periodontitis, immortalized human gingival keratinocytes (IHGK) were stimulated with the proinflammatory cytokine interleukin-1beta. An epithelia-specific cDNA microarray was used to analyze mRNA expression profiles from IHGK cells treated with 200 units interleukin-1beta/ml for 3, 6, 9, 12, and 24 h. Indirect immunofluorescence was carried out to detect NF-kappaB in IHGK following interleukin-1beta treatment. RESULTS: Detailed analysis revealed distinct patterns of time-dependent changes, including genes induced or repressed early (3-6 h) or late (12-24 h) after interleukin-1beta treatment. Differentially expressed genes were involved in (i) cell stress, (ii) DNA repair, (iii) cell cycle and proliferation, (iv) anti-pathogen response, (v) extracellular matrix turnover, and (vi) angiogenesis. A large number of genes were responsive to NF-kappaB and induction was concomitant with nuclear translocation of the p65 RelA subunit of NF-kappaB. Interestingly, many of these genes contain multiple NF-kappaB binding sites in their promoters. CONCLUSION: Analysis of altered gene expression allows identification of gene networks associated with inflammatory responses. In addition to a number of well-known genes involved in gingivitis and periodontitis, we identified novel candidates that might be associated with the onset and maintenance of an inflammatory disease.

Expression of mRNAs encoding for growth factors, ECM molecules, and MMP13 in mono-cultures and co-cultures of human periodontal ligament fibroblasts and alveolar bone cells.

Although the function and effects of many growth factors and extracellular matrix (ECM) molecules have been described for several periodontal tissues in vivo and in vitro, the molecular interactions involved in the communication between cells of the periodontal ligament and the alveolar bone are poorly understood. To contribute to the identification of such interactions, we have generated co-cultures (CCs) of periodontal ligament fibroblasts (PDLs) and alveolar bone cells (ABCs) and compared mRNA expression for various growth factors, ECM molecules, and matrix metalloproteinase13 (MMP13) after 1 and 2 weeks with matched mono-cultures (MCs) by reverse transcription/polymerase chain reaction. Compared with CCs of 1 week, PDLs and ABCs after 2 weeks revealed relatively high levels of all analyzed mRNAs, viz., for EGF, HGF, VEGF, TGFbeta1, collagen-I (COL1), osteonectin (ON), fibronectin (FN1), and MMP13. At week 2, when compared with MCs, CCs showed an elevation of all tested mRNAs in PDLs and ABCs, except for TGFbeta1 and FN1, which only increased in PDLs. After 1 week, when CCs were compared with MCs, mRNAs for HGF and TGFbeta1 were less abundant in PDLs and ABCs, whereas the other genes exhibited lower expression levels in only one of the cell types. Analysis of our data indicated that the expression of mRNAs for growth factors and for COL1, ON, FN1, and MMP13 was modulated in the distinct cell types with respect to culture time and culture type. The differences in the mRNA expression patterns between CCs and MCs suggest that the respective genes are involved in the molecular interactions of PDLs and ABCs.

Morphogenesis and proliferation in mono- and organotypic co-cultures of primary human periodontal ligament fibroblasts and alveolar bone cells.

Cells of the periodontal ligament and the alveolar bone lie in close vicinity in the periodontium. The goal of this study was to create an in vitro model to facilitate the study of the morphogenesis and proliferation of these two cell types under more in-vivo-like conditions. This was accomplished by the generation of organotypic co-cultures of primary human periodontal ligament fibroblasts (PDL) and alveolar bone cells (BC) and matched mono-cultures after 1, 2 and 3 weeks. Indirect immunofluorescence (IIF) for vimentin indicated that PDL cells exhibited sustained stratification only in the presence of BC cells, suggesting an important role for BCs in maintaining the stratification of PDL cells. In mono-cultures, only BC cells showed progressing stratification. They also displayed the most pronounced contraction of the cell culture matrix. Moreover, Ki-67 antigen detection by IIF revealed that these features coincided with cell proliferation localized on the matrix surface at the onset of cell stratification. These findings suggest that, in addition to proliferation, a further prerequisite for stratification may be cell migration. Furthermore, the maintained cell stratification, proliferation, and compartmentalization noted for PDL cells in organotypic co-cultures and BCs in mono-cultures can only be observed in a three-dimensional culture system. Thus, our system represents a novel experimental tool to further elucidate the underlying mechanisms of the growth and differentiation of PDL and bone tissue.

Metastatic growth of squamous cell carcinomas is correlated with upregulation and redistribution of hemidesmosomal components.

Bullous pemphigoid antigen-1 (BPA1) and alpha(6)beta(4)-integrin colocalize at the hemidesmosomes in basal-layer keratinocytes of normal squamous epithelia. The expression of these genes was analyzed during the process of tumor cell invasion and metastasis on frozen sections of head and neck biopsies, and the structural appearance of hemidesmosomes was analyzed by electron microscopy. Despite a diminution of hemidesmosomal structures as revealed by electron microscopy, gene expression of BPA1 and alpha(6)beta(4)-integrins was distinctly upregulated with the onset of invasive growth, demonstrated at the mRNA level by in situ hybridization. The upregulated gene expression extended to the entire proliferative zone of invasive tumors, including the tumor cells which have lost contact with the basement membrane and no longer display hemidesmosomes. The polarized localization of the BPA1 and alpha(6)beta(4) proteins to the basal aspect of the peripheral tumor cells was largely retained in invasive but nonmetastatic lesions, but was lost upon progression to metastatic growth of head and neck squamous cell carcinomas (SCC), in which pericellular staining extended into many tumor cell layers. The results of this study confirm that expression of BPA1 and alpha(6)beta(4)-integrins is elevated in carcinoma cells but is not directed to intact hemidesmosomes. Importantly, this loss of directed localization is an indicator of the capacity to metastasize.

Topographic changes of focal adhesion components and modulation of p125FAK activation in stretched human periodontal ligament fibroblasts.

Mechanical stress has been shown in vitro to modulate integrin-beta1-mediated activation of p125FAK/FAK. To test the hypothesis whether this also applies to periodontal ligament fibroblasts (PDLs), we subjected human PDLs to mechanical stretch and analyzed stress-induced changes of p125FAK activation by quantitative immunoprecipitation of p125FAK and changes in the topography of molecules localizing in focal adhesions by indirect immunofluorescence. Generally, all components of focal contacts under study-including detection of phosphotyrosine, i.e., integrin-beta1, p125FAK, and paxillin-revealed a relative co-localization during stretch application. Under stretch, we observed a re-distribution of all components from the cell periphery to the cytoplasm following the main axes. Tyrosine phosphorylation of p125FAK was monitored up to 72 hours under stretch. While the amount of p125FAK remained essentially constant, the activation of p125FAK was clearly modulated. Tyrosine phosphorylation of p125FAK increased from 15 minutes up to 1 hour and declined after stretching periods of 24, 48, and 72 hours. The analysis of our data indicated a stretch-induced redistribution of focal adhesion components and a modulation of p125FAK activation, suggesting alterations in focal adhesions and their associated signal cascade.

Connexin 43 expression is downregulated in raft cultures of human keratinocytes expressing the human papillomavirus type 16 E5 protein.

A decrease in gap junction-mediated cell-to-cell communication has previously been observed in monolayer cultures of human keratinocytes (HaCaT cells) expressing the human papillomavirus type 16 E5 (HPV16 E5) gene and attributed to the reduced phosphorylation of connexin 43, the most abundant connexin in HaCaT cells. In line with this observation, we have now analyzed the effect of HPV16 E5 on connexin 43 expression in raft cultures produced by transfected HaCaT cells. These keratinocytes transcribe HPV16 E5 under the control of a dexamethasone-inducible promoter. Our results show that treatment with dexamethasone leads to an almost complete disappearance of connexin 43 in rafts expressing the E5 gene but not in control rafts. In our study we discuss the possible effects of this downregulation on cell-cell communication and cellular malignant transformation.

Modulation of the epidermal growth factor receptor by the human papillomavirus type 16 E5 protein in raft cultures of human keratinocytes.

It has been shown that the E5 protein of the human papillomavirus type 16 modulates epidermal growth factor receptor downregulation in monolayer cultures of human keratinocytes and mouse fibroblasts. We have now analysed the effect of this protein on the expression, the distribution and the activation of EGF receptors in raft cultures derived from an E5-transfected human keratinocyte cell line. The epithelia generated in these cultures were stratified and exhibited suprabasal expression of cytokeratins 1 and 10, which are known markers of early epidermal differentiation. In situ hybridization with an antisense riboprobe to the human papilloma virus type 16 E5 protein revealed a homogeneous gene expression within the entire epithelium of E5-transfected but not empty vector-transfected control cultures. Treatment of serum-starved rafts with EGF for 48 hours led to a strong decrease of suprabasal EGF receptors in control cultures, but not in rafts of E5-expressing cells. Under these conditions, no activated receptors were observed in control cultures, but activated receptors were still present in E5-raft cultures. Our results indicate that human papilloma virus type 16 E5-mediated modulation of EGF receptor expression occurs in a time- and structure-dependent manner in epithelial equivalents of human keratinocytes.

Organotypic cultures of gingival cells: an epithelial model to assess putative local effects of orthodontic plate and occlusal splint materials under more tissue-like conditions.

This article explores whether organotypic cultures of immortalized gingival keratinocytes constitute a suitable model for assessing the epithelial cell compatibility of two groups of dental resins, each of them representing one group used in orthodontics and temporo-mandibular disorders (TMD) therapy under conditions more closely resembling the actual tissue situation. The resins were tested with the agar diffusion assay (ADA) in conventional monolayer and organotypic cultures. Compared to the control exhibiting a neutral red destaining index of 3, the index of 4 obtained after exposure of monolayers to one soft permanent resin (Durabase) indicated the presence of a non-lytic but physiologically active substance. In contrast, the adaptation of the ADA to organotypic cultures revealed no apparent lesions at the epithelial surface by performing scanning electron microscopy, while histoarchitecture indicated the development of stratified surface epithelia. This was substantiated by undamaged cells in the uppermost cell layers and by the preservation of cell-to-cell contacts. Furthermore, indirect immunofluorescence for Ki-67 and the cytokeratins ck 14 and ck4 revealed that cell proliferation and epithelial structure were maintained, while differentiation was enhanced, possibly increasing epithelial resistance. The results obtained from the organotypic cultures suggest that (i) cell-affecting effects of materials visible in monolayer cultures may not be seen in epithelia resembling that in vivo and that (ii) enhanced differentiation may be associated with increased stability of the epithelial cells. Thus, organotypic cultures of gingival cells constitute a tissue model allowing short-term tissue compatibility studies of dental materials and rendering a potential candidate also for long-term studies.

Assessment of acute cyto- and genotoxicity of corrosion eluates obtained from orthodontic materials using monolayer cultures of immortalized human gingival keratinocytes.

Whilst a patient is undergoing orthodontic treatment, dental appliances based on non-precious metals or titanium remain in the oral cavity for up to several years. Throughout this period the appliance is in either direct or indirect contact with the oral mucosa. To investigate the possibility of cell damage occurring as a result of appliance corrosion, monolayer cultures of immortalized human gingival keratinocytes were assessed for acute cyto- and genotoxicity using the hexosaminidase assay and the Comet assay respectively. The materials tested included 1. a nickel-free wire, 2. a UK-1 bond, 3. nickel-free as well as nickel-containing brackets with and without color signature and 4. a titanium expansion screw. Each of the test materials was corroded in a solution consisting of equal amounts of lactic acid and sodium chloride (0.1 M) for 1, 3, 7 and 14 days. The cell cultures were then exposed to eluates exhibiting the highest ion concentrations. None of the eluates was found to exhibit acute cytotoxicity, regardless of the type of test system used. Qualitative assessment using neutral red dye for live cells and either trypan blue or propidium iodide to disclose dead cells failed to reveal any significant increase in cell damage when exposed cells were compared to control cultures. Unrestricted cell vitality was confirmed by quantifying viable cells through measurement of hexosaminidase enzyme activity. Furthermore, assessment of genotoxicity revealed no apparent DNA damage to immortalized gingival keratinocytes following exposure to the test eluates. Because the materials tested in this study were corroded using the exacting methods normally applied to precious metals or gold-containing alloys, the lack of either acute cyto- or genotoxic effects following exposure to the test eluates indicates that the materials tested exert no adverse effects on cells similar to those of the target tissue exposed to the materials in situ.

Defects of basement membrane and hemidesmosome structure correlate with malignant phenotype and stromal interactions in HaCaT-Ras xenografts.

Benign and malignant HaCaT-ras clones, derived from immortalized HaCaT cells were grown as nude mouse surface transplants rendering a human tumor progression model. Searching for malignancy-related alterations, the deposition, localization and mRNA of basement membrane and hemidesmosome components were analysed by immunofluorescence, in situ hybridization and electron microscopy. Initially, at 1 week epithelia of benign and malignant cells revealed a similarly low polarity and an enlarged 'activated basal' compartment, reflected by partial dislocation and extended pericellular staining of the hemidesmosome constituent integrin alpha 6 beta 4 seen by immunofluorescence. Whereas benign grafts eventually normalized, closely resembling grafts of HaCaT cells, malignant growth was correlated with a persisting epithelial activation state and continuing higher expression of alpha 6 (by immunofluorescence and in situ hybridization). The basement membrane components bullous pemphigoid antigen 1, laminin-5 and collagen IV exhibited a largely linear distribution at 1 week. However, in the malignant cell transplants initially minor basement membrane discontinuities became more severe at around 2 weeks, associated with close stromal cell contacts, angiogenesis and invasion. Most striking were basement membrane alterations seen by electron microscopy. At 1 week stretches of basement membrane had developed in malignant transplants, though to a much lesser extent than in benign specimens. With invasion these basement membrane structures mostly disappeared despite persistent although variable immunofluorescence, suggesting high turnover without ultrastructural assembly. The hemidesmosome structures were defective throughout, completely lacking anchoring plaques with keratin filaments, whereas they were still associated with basement membrane deposits. Thus, malignant HaCaT-ras transplants, while initially resembling regenerating wounds, revealed an increasing loss of tissue polarity and basement membrane structures, which seemed to be accelerated upon stromal cell contacts.

Normalization of keratinocyte-type integrins during the establishment of the oral mucosa phenotype in vitro.

In stratified epithelia, integrins play a fundamental role in mediating basal cell attachment to a variety of extracellular matrix molecules. To assess whether keratinocyte-specific integrins are expressed in a similar way as in the normal situation also under in vivo related conditions, we processed oral mucosa equivalents consisting of keratinocytes and fibroblasts from non-cornified gingiva. In this model histomorphology, the expression of differentiation-specific keratins and keratinocyte-type integrins exhibited similarity to the tissue of origin. The stages of tissue normalization were assessed on frozen sections by indirect immunofluorescence. The initial activated stage (1 week) was characterized by (i) incomplete epithelial organization and a weak presence of the suprabasal mucosa type keratin K4, (ii) diffuse expression of the integrin chains beta 1 and alpha 6 and (iii) abundance of the wound healing-associated integrin alpha v throughout the whole epithelium. After 2 weeks, the increase in epithelial organization was characterized by (i) the presence of a basal and suprabasal cell compartment, (ii) extension of K4 in the suprabasal compartment, (iii) extended expression of the keratinocyte integrins beta 1 and alpha 6 and (iv) concentration of alpha v integrin underneath basal cells. Further normalization of tissue architecture was indicated by (i) a slight increase in K4 extension, (ii) appearance of keratinocyte integrins beta 1 and alpha 6 in basal and parabasal cells and (iii) interruption of the band-like alpha v integrin immunolocalization at the subepithelial site. The findings in the in vitro model system indicate that these oral mucosa equivalents exhibit similarities to the in vivo situation of non-cornified gingiva, thus rendering them a suitable model for the assessment of stages during epithelial reconstruction or in vivo relevant studies on material effects.

Establishment of oral mucosa phenotype in vitro in correlation to epithelial anchorage.

Cell-matrix interactions and the ordered deposition of basement membrane (BM) components are of major importance for the maintenance of tissue homeostasis in complex epithelia. This aspect was studied in vitro in a coculture system designed as an oral mucosa model. As crucial epithelial features the kinetics of proliferation, expression of site-specific keratins as well as integrin patterns in correlation to synthesis of BM components were assessed by immunohistochemistry and in situ hybridization. Comparison with non-cornified gingiva as tissue of origin revealed different stages of epithelial development, eventually leading to complete reconstruction within a time frame of 1-3 weeks. First, the initial activated stage up to 1 week was characterized by (a) high keratinocyte proliferation, (b) extended expression of the basal cell-specific keratin K5 and (c) a patchy pattern of the differentiation-specific keratins K4 and K13. Second, after 2 weeks the improvement of histoarchitecture correlated to (a) predominant K5 expression in the basal and (b) extension of K4 and K13 within the suprabasal cell compartment, (c) high expression of integrins alpha3 beta1 and alpha6 beta4 including their ligand laminin-5 and (d) accumulating deposition of basement membrane components. Third, virtually complete tissue normalization at 3 weeks was indicated by (a) restriction of K5 to the basal cell area, (b) regular suprabasal localization of K4 and K13, (c) polarization of integrins to basal and parabasal cells and (d) linear codistribution of collagen IV, "classical" laminin (-1 or -10) and laminin-5 underneath the basal cells. Thus, these organotypic cocultures represent relevant equivalents for non-keratinized oral mucosa with typical gingival differentiation features and in addition suitable models for preclinical trials such as prospective dental material testing.

Histomorphological and biochemical differentiation capacity in organotypic co-cultures of primary gingival cells.

To establish a three-dimensional in vitro test system mimicking the physiological situation of the oral cavity, organotypic co-cultures consisting of primary gingival cells on a collagen matrix with fibroblasts were generated. The histomorphological development after 7 and 14 d revealed close similarity with the non-keratinized gingiva epithelium. Furthermore, as epithelial specific markers synthesis and localization of keratins as well as the deposition of basement membrane components were assessed on frozen sections by immunofluorescence and keratin expression by in situ hybridization. Primary keratinocytes in conventional culture strained positive for keratin K14 and the mucosal differentiation-specific keratins K4 and K13, while primary fibroblasts, isolated from the same tissue source, and also some keratinocytes, were positive for vimentin. In organotypic co-cultures the keratinocytes formed a multilayered epithelium within 14 d containing basal cells and flattened cells in the uppermost layers. Comparable to native non-keratinized gingiva keratin 14 gene expression was clearly detectable in the basal cell compartment but showed extending immunolocalization. In addition, particularly at the early stage (7 d), basally located keratinocytes were also vimentin positive. According to morphological differentiation K4 and K13 were detectable in suprabasal position a the RNA and protein level. The major basement membrane constituents collagen type IV and laminin increased with time revealing first an interrupted and later a fully extended staining underneath the basal cells. Maintenance of basal cell function was further demonstrated by cell proliferation (BrdU incorporation) which was initially high (7 d) but declined towards the later stages (14-21 d). The results demonstrate i) that this co-culture system leads to a stratified surface epithelium with morphological and biochemical characteristics of the non-keratinized gingiva epithelium and ii) that a state of physiological tissue balance was reached, thus rendering a suitable model for tissue compatibility studies.

Integrin and basement membrane normalization in mouse grafts of human keratinocytes--implications for epidermal homeostasis.

Integrin patterns and formation of basement membrane (BM) were investigated in correlation to epidermal growth and differentiation during skin regeneration in human keratinocyte transplants on nude mice. Immuno-fluorescence and transmission electron microscopy (TEM) showed that different stages of tissue reconstruction were characterized by a sequence of coordinated events. Features of the initial tissue activation, with rapid keratinocyte proliferation around day 4, including cells in a suprabasal position, were: (1) a marked increase in and extended distribution of the integrin chains alpha 2, alpha 3, beta 1 and alpha 6, while beta 4 already showed a preferential basal location; (2) de novo expression of alpha 5 and alpha v; and (3) marked deposition of laminin-5 and nidogen but low levels of other BM components. Tissue normalization during the 2nd week, initiated by a drastic decrease in the number of proliferating cells after day 4, now strictly in basal position, was signified: by (1) orthotopic staining for basal-type keratins (K5, K14) together with a regular pericellular alpha 2 beta 1 and alpha 3 beta 1 distribution, (2) linear, balanced deposition of BM components (e.g. laminin-1, type IV collagen) and (3) colocalization of integrin alpha 6 beta 4 and bullous pemphigoid antigen. Simultaneously at 7 days hemidesmosomes and a defined BM had developed (TEM), becoming continuous at 14 days. This coincided with the regular distribution of suprabasal keratins (K1, K10) as well as intermediate (involucrin) and late differentiation markers (filaggrin, loricrin). Type-VII collagen deposition, still irregular at 14 days, became continuous at 22 days together with developing BM-anchoring fibrils indicating final tissue consolidation. This model mimics principal stages of epidermal wound healing in human skin and implies a linkage between BM assembly, integrin distribution and the compartment of proliferation competent cells, which in turn determines the onset of differentiation. Thus, apart from the balance of diffusible growth regulators, this positional control of keratinocytes, largely accomplished by integrin-matrix interactions, seems to be prerequisite to establishment and maintenance of tissue homeostasis.