The dento-epithelial junction: cell adhesion by type I hemidesmosomes in the absence of a true basal lamina.

BACKGROUND: The junctional epithelium (JE) is a unique structure that makes contact with both a non-renewable hard tooth surface and with a basement membrane (BM) facing the connective tissue. Ultrastructurally, this attachment occurs through hemidesmosomes (HD) and a basal lamina-like extracellular matrix which, on the tooth side, is termed the internal basal lamina. In this study we investigated the expression of basal cell markers in the tooth-facing (TF) cells of JE. METHODS: Samples of healthy marginal gingiva were removed by careful dissection. The expression of laminin-5 was used to indicate TF cell preservation in double immunofluorescence labeling and confocal laser scanning microscopy. RESULTS: The results show that integrin alpha6beta4 and laminin-5 colocalize unequivocally in the TF cells. The results also show the specific expression of the basal cytokeratin 14 and the alpha(v) integrin subunit in the TF cells. All 3 major hemidesmosomal components BP180, BP230, and HD1 antigen are likewise present. On the other hand, type IV collagen, laminin-1/10, type VII collagen, and the BM proteoglycan perlecan are all absent from the dento-epithelial junction. CONCLUSIONS: The results indicate that the epithelium-tooth interface is a unique structure wherein epithelial cells adhere by means of bona fide hemidesmosomes to an epithelium-derived extracellular matrix lacking most of the common BM components. Moreover, TF cells differ from connective tissue facing (CTF) cells, not only by their cell surface molecules and their production of extracellular matrix, but also by their cytoskeletal architecture.

Laminin synthesis and the adhesion characteristics of immortalized human corneal epithelial cells to laminin isoforms.

We have studied the synthesis of laminins (Ln) and determined the specific integrins mediating the adhesion of immortalized human corneal epithelial cells to mouse Ln-1, and human Lns-5 and -10. Immunofluorescence microscopy of the cells demonstrated integrin alpha(2), alpha(3), alpha(6), beta(1)and beta(4)subunits, integrins alpha(6)and beta(4)being found in a typical 'leopard-skin' like manner. Immunoprecipitation studies showed that the cells produced alpha 3, beta 3 and gamma 2 chains of Ln-5, but not Lns-1 or -10. In culture Ln-5 was found as small plaques beneath the adhering cells within 1 hr, while in 4 hr widely spread Ln-5 plaques were observed in colocalization with beta(4)integrin subunit. By using a quantitative cell adhesion assay and function-blocking monoclonal antibodies we showed that integrin beta(1)subunit plays a role in mediating corneal epithelial cell adhesion to mouse Ln-1. However, none of the available function-blocking antibodies to integrin alpha-subunits inhibited the adhesion. Integrin alpha(3)beta(1)complex mediated the adhesion of corneal epithelial cells to human Lns-5 and -10. Integrin complex alpha(3)beta(1), as well as laminin alpha(3)chain, was also shown to mediate cell adhesion to newly produced endogenous Ln-5. The present results show that integrin alpha(3)beta(1)complex mediates the adhesion of corneal epithelial cells to Lns-5 and -10, while a yet unknown integrin alpha subunit appears to play a role in the adhesion to Ln-1. The results also show that among corneal basement membrane laminins, Ln-5 is synthetized by epithelial cells while Ln-10 may be a product of keratocytes.

The junctional epithelium around murine teeth differs from gingival epithelium in its basement membrane composition.

It is not known whether epithelial differentiation patterns are reflected in the composition of gingival basement membranes (BMs). We have investigated the expression of laminin isoforms and associated BM components in the murine dento-epithelial junction by using immunofluorescence microscopy. Our results show that chains of laminins 5/6/7/10/11 are expressed in the BM of outer gingival epithelium. The external BM between junctional epithelium (JE) and connective tissue differs from gingival BM by lacking laminin-7 and -11 chains. The internal basal lamina (IBL) between JE and tooth contains only laminin-5. Collagen chains alpha1,2(IV) and nidogen-1 are present in other BMs except the IBL. The dento-epithelial junction thus has a unique BM composition, suggesting that epithelial cells are able to secrete two extracellular matrices in a polarized manner. The exclusive expression of the non-self-polymerizing laminin-5 indicates that the IBL is not a BM by definition, but rather a simple extracellular matrix lacking network structure.

Expression of cell adhesion complexes in epithelial cells seeded on biomaterial surfaces.

Clinical studies indicate that soft tissue responses around dental implants vary, depending on the material used. It is therefore also possible that there are differences in how epithelial cells attach to various biomaterial surfaces. We studied the adhesion of cultured epithelial cells to five different dental material surfaces and to glass. The efficacy of adhesion was evaluated by using scanning electron microscopy (SEM) and immunofluorescence microscopy (IF) with antibodies to vinculin and alpha(6)beta(4) integrin, two cell surface molecules that are functional in epithelial cell adhesion. Our results indicate that epithelial cells adhere and spread more avidly on metallic surfaces (titanium, Ti(6)Al(4)V titanium alloy, dental gold alloy) than on ceramic surfaces (dental porcelain, aluminum oxide). As revealed by SEM, cells on metallic surfaces had a flattened morphology and formed multicellular islands. On porcelain and aluminum oxide most cells were round and adhesion occurred as single cells. Surface coverage was over twofold on metallic surfaces as compared to ceramic surfaces. IF of cells grown on metallic surfaces revealed vinculin in well-organized focal contacts and alpha(6)beta(4) integrin in punctate patterns typical of prehemidesmosomes. On porcelain and aluminum oxide surfaces the cells were mostly round and showed less well-organized adhesion complexes. Our results indicate that smooth metallic biomaterial surfaces are optimal for epithelial cell adhesion and spreading. These findings may have clinical implications in the design of transgingival implant structures.

Laminin gamma2 expression is developmentally regulated during murine tooth morphogenesis and is intense in ameloblasts.

Mutations in the laminin gamma2 gene cause junctional epidermolysis bullosa, and enamel hypoplasias are frequently seen in these patients. Laminin gamma2 is one of the three polypeptide chains forming the basement membrane glycoprotein laminin-5. We have localized the expression of the laminin gamma2 gene by in situ hybridization during mouse tooth development from early morphogenesis to completion of crown development. The expression was restricted to epithelial cells. During the early morphogenesis of the tooth germ, laminin gamma2 was expressed by the outer dental epithelium and by the stellate reticulum cells. No expression was detected in the cells of the inner dental epithelium giving rise to ameloblasts. The pre-ameloblasts remained negative during the early bell stage, but, interestingly, expression was very prominently upregulated as the cells differentiated into ameloblasts. This upregulation appeared to coincide with the start of enamel matrix secretion. The ameloblasts expressed laminin gamma2 intensely throughout the period of active enamel deposition. The expression continued at a lower level in the maturation-stage ameloblasts covering the enamel surface. Immunolocalization of laminin-5 with polyclonal antibodies indicated that the protein formed a continuous lining at the basal surfaces of the cells expressing the laminin gamma2 transcripts. We suggest that the role of laminin-5 during enamel formation may be to strengthen the anchorage of the ameloblasts to the enamel matrix, and that the pathogenesis of enamel hypoplasias in cases of laminin-5 mutations could be associated with detachment of the ameloblast cell layer from the enamel surface.

The epithelium-tooth interface--a basal lamina rich in laminin-5 and lacking other known laminin isoforms.

The attachment of the marginal gingiva to the tooth surface is mediated by a thin nonkeratinized epithelium termed the junctional epithelium (JE). Ultrastructural studies have revealed that the attachment of the JE to the tooth surface occurs through hemidesmosomes (HD) and a basal lamina-like extracellular matrix termed the internal basal lamina (IBL). We have previously shown that neither type IV collagen nor prototypic laminin, two common components of basement membranes (BM), is present in the IBL between the epithelium and the tooth. In the present study, we show that laminin-5 is a major component of the IBL in both rodent and human tissues. By using in situ hybridization, we also show that the cells of the JE express the LAMC2 gene of laminin-5. In other parts of gingival epithelium, LAMC2 gene expression is less prominent. Our results indicate that the epithelium-tooth interface is a unique structure wherein epithelial cells are induced to secrete a basal lamina containing laminin-5 and no other presently known laminin isoform.

Rapid spreading and mature hemidesmosome formation in HaCaT keratinocytes induced by incubation with soluble laminin-5r.

HaCaT cells, an immortalized keratinocyte line, incubated in plastic wells in the presence of conditioned medium from 804G cells adhered and spread rapidly in less than 30 min. In contrast, cells plated in fibroblast or keratinocyte conditioned medium adhered poorly and remained rounded at 30 min. Immunodepletion of 804G conditioned medium with polyclonal antisera to laminin-5r, but not control antisera, abolished rapid cell spreading. Electron microscopy of HaCaT cells spread by incubation in 804G conditioned medium, but not control medium, revealed mature hemidesmosomes after 24 h. Rapid spreading was also observed in wells precoated with 804G conditioned medium or 804G cell-deposited matrix, but not with fibronectin, vitronectin, or laminin-1. Immunoblotting of 804G conditioned medium with anti-laminin-5r antibodies unveiled polypeptides of 150, 140, 135, and 100 kDa, identical by electrophoretic mobility to immunoreactive polypeptides in 804G deposited matrix. Our results suggest that addition of laminin-5r in a soluble form is sufficient to promote rapid spreading and hemidesmosome assembly in keratinocytes. The mechanism of soluble laminin-5r action may include efficient surface "priming" for cell adhesion. Soluble laminin-5r may have a physiologic role in morphogenesis and repair of the epidermis and may be of use for therapeutic applications.

Immunolocalization of fibronectin and vitronectin receptors in human gingival fibroblasts spreading on titanium surfaces.

The adhesion and spreading of human gingival fibroblasts on glass and differently processed titanium surfaces was studied by immunolocalization of vinculin and the alpha and beta subunits of the fibronectin (alpha 5 beta 1) and (alpha v beta 3) receptors. Vinculin-containing focal contacts were present both at 4 and 24 h of spreading in cells grown on glass or electropolished or etched titanium surfaces but not in cells spreading on sandblasted titanium surfaces. Immunostaining for the alpha 5 and beta 1 subunits of the fibronectin receptor showed only a diffuse membrane fluorescence after 4 h of cell spreading irrespective of the growth surface. The alpha v and beta 3 subunits of the vitronectin receptor were at this stage detected in focal contacts in cells spreading on glass or electropolished or etched titanium surfaces. In cells spreading on sandblasted titanium surfaces, however, the vitronectin receptor had only a diffuse distribution. In cells that had been allowed to spread for 24 h on glass or electropolished or etched titanium surfaces the alpha 5 and beta 1 integrin subunits were either diffusely distributed or showed a localization typical of extracellular matrix contacts. The alpha v and beta 3 integrin subunits were, as earlier, localized to typical focal contacts in cells grown on glass or electropolished or etched titanium surfaces. Cells attached to sandblasted titanium surfaces still expressed all the integrin subunits only diffusely. The results show that the surface texture of the substratum can affect the expression of integrin subunits in human gingival fibroblasts.(ABSTRACT TRUNCATED AT 250 WORDS)

Increased rate of salivary epidermal growth factor secretion in patients with juvenile periodontitis.

We compared salivary epidermal growth factor (EGF) concentrations in patients with juvenile periodontitis (JP) and periodontally healthy controls. In initial screening of 45 JP patients and a group of healthy controls, significantly higher salivary EGF concentrations were measured in the JP patients. Subsequently, 17 JP patients who had high EGF concentrations in some of their salivary samples were chosen, and a group of age- and sex-matched controls was selected. We then examined their EGF concentrations and EGF secretion rates under standardized conditions in stimulated and unstimulated saliva and studied the expression of EGF receptor (EGF-R) in their gingival tissues. The results showed that the mean EGF concentration (pmol/ml) was slightly higher in JP patients than in controls. However, the difference was statistically significant only in stimulated saliva and when calculated per milligram salivary protein. When EGF release was measured as the rate of EGF secretion (pg/min), significantly higher values were observed in JP patients than in controls both in unstimulated and stimulated saliva. Immunofluorescence microscopy (IF) of gingival samples from JP patients and their controls revealed no quantitative or qualitative differences in the expression of EGF-R. Our results demonstrate the complex nature of salivary EGF release. The elevated rate of salivary EGF secretion in JP patients may be associated with the pathogenetic mechanisms of juvenile periodontitis.

Effect of surface processing on the attachment, orientation, and proliferation of human gingival fibroblasts on titanium.

The adhesion, orientation, and proliferation of human gingival fibroblasts was studied on electropolished (elpTi), etched (etchTi), and sandblasted (sblTi) titanium surfaces. The texture, chemical state, and composition of the titanium surfaces were analyzed using a surface tracing instrument and electron spectroscopy for chemical analysis. Considerable differences were evident in the surface texture and chemical composition of the differently treated titanium plates. Electropolishing produced the smoothest and cleanest surface. Human gingival fibroblasts attached, spread, and proliferated on all titanium surfaces. However, cells on elpTi exhibited an extremely flat morphology and seemed to form cellular bridges with adjacent cells, whereas the etchTi and sblTi surfaces harbored both round and flat cells with many long processes. Cells on elpTi appeared to grow in thick layers with no specific orientation, whereas on etchTi surfaces they were migrating along the parallel, irregular minor grooves caused by mechanical polishing, and on sblTi surfaces they seemed to grow in clusters. Stress-fiber type actin bundles and vinculin-containing focal adhesions were present in cells spreading on elpTi and etchTi surfaces but not in cells spreading on sblTi surfaces. Cell shape, orientation, and proliferation appear to depend on the texture of the titanium surface and probably also on the properties of the oxide layer and adjacent bulk material. Our findings suggest that smooth or finely grooved titanium surfaces could be optimal in implants adjacent to soft tissues as they support the attachment and growth of human gingival fibroblasts.

Immunolocalization of integrin alpha 6 beta 4 in mouse junctional epithelium suggests an anchoring function to both the internal and the external basal lamina.

The localization of the integrin alpha 6 beta 4, a transmembrane adhesion molecule associated with hemidesmosomes, was studied in mouse junctional epithelium (JE) by the use of monoclonal antibodies in indirect immunofluorescence microscopy. The results showed that the integrin a6 subunit was expressed throughout the JE and was localized to the cell membranes, including the aspects facing the internal and external basal laminae. The beta 4 subunit had a more restricted distribution. It was expressed only in cells facing the internal and the external basal laminae and had a basally polarized distribution. In other parts of gingival epithelium, both integrin subunits were mainly expressed at the basal aspects of basal epithelial cells. The basement membrane components, type IV collagen and laminin, could be detected only in the external basal lamina and in other basement membranes of gingival epithelium. The results indicate that the a6 beta 4 integrin, expressed in mouse JE, has a role in mediating the attachment of the cells to the basement membranes facing the connective tissue and the tooth.

Immunolocalization of proteins specific for adhaerens junctions in human gingival epithelial cells grown on differently processed titanium surfaces.

The localization of desmoplakins 1 and 2 (DP 1&2), components of desmosomes, vinculin, and actin, was studied in gingival epithelial cells grown on cell culture glass and on titanium plates with various surface topography. The results showed that epithelial cells attached and spread more readily on smooth than on rough, sandblasted titanium surfaces. Moreover, the cells appeared to develop more granular DP 1&2 immunoreactivity at their ventral surfaces when grown on smooth or etched titanium as compared to glass. In cells grown on sandblasted titanium surfaces, DP 1&2-specific immunoreactivity was primarily located at cell-cell contacts. Cells grown on smooth titanium surfaces harbored a fine network of actin filaments with apparent cell-to-cell organization. Vinculin was confined to cell-cell contact areas. No vinculin-containing focal adhesions could be detected, suggesting that the cells adhere either by means of close contacts, extracellular matrix contacts, or by means of hemidesmosomes. The findings suggest that smooth of finely grooved titanium surfaces could be optimal in maintaining the adhesion and specialized phenotype of gingival epithelial cells.

Immunohistochemical localization of epidermal growth factor receptors in human gingival epithelia.

Epidermal growth factor (EGF) is a small molecular weight polypeptide which is thought to have important functions in epithelial growth and differentiation and in wound healing. EGF exerts its action on cells through binding to a cell surface receptor. Using immunohistochemistry and a monoclonal antibody (mAb) directed against the EGF receptor, we have examined gingival specimens of periodontally healthy individuals and patients with adult adult (AP) and juvenile periodontitis (JP), as well as epithelial cell rests of Malassez. EGF receptors were expressed at high levels on the cell surface of basal cell layers of gingival epithelium. In normal junctional epithelium, on the other hand, specific labeling was faint or negative, indicating that receptors are poorly expressed or absent in these cells. No differences were detected between uninflamed gingival specimens of periodontally healthy subjects and of patients with JP. Instead, in biopsies of inflamed tissue from AP patients, an intense cell surface labeling was revealed in proliferating epithelial cells. Moreover, the epithelial cell rests of Malassez bound the antibody intensely. The results suggest that EGF is involved in control of epithelial growth and differentiation in periodontal tissues.

Expression of integrins in human gingiva.

The distribution of the alpha 1-alpha 6 subunits of beta 1 integrins was studied by using a panel of monoclonal antibodies in indirect immunofluorescence microscopy. The results showed that the beta 1 subunit was expressed at the cell membrane of basal cells of gingival epithelium, throughout the cells of junctional epithelium (JE), and in cells of connective tissue, including endothelial cells and, more faintly, in inflammatory cells in gingival connective tissue. The alpha 4 subunit was expressed selectively in inflammatory cells, and the alpha 5 subunit was expressed in cells throughout gingival connective tissue. An overall cell membrane immunoreactivity for the alpha 2 and alpha 3 subunits was shown in basal cells of gingival epithelium and in cells of JE, corresponding to the epithelial localization of the beta 1 subunit. The alpha 6 subunit was polarized to the basal aspects of basal epithelial cells, but was also present in an overall cell surface distribution in basal cells and in cells of JE. The beta 4 integrin subunit was mainly expressed at the basal aspects of basal cells in gingival epithelium and JE. The results indicate that the alpha 2/beta 1, alpha 3/beta 1, alpha 6/beta 1, and alpha 6/beta 4 integrins are all expressed in human gingival epithelium. Of these, the alpha 6/beta 4 integrin complex is the major candidate for mediation of the attachment of epithelial cells to the basement membrane facing the connective tissue and probably also the tooth.

Psophocarpus tetragonolobus agglutinin reveals N-acetyl galactosaminyl residues confined to endothelial cells and some epithelial cells in human tissues.

We studied the binding of Psophocarpus tetragonolobus agglutinin (PTA) conjugates to human adult tissues. In all kidney specimens studied, PTA bound in a blood group-independent way to endothelia in glomerular and intertubular capillaries as well as in larger vessels. In addition, a heterogeneous binding to collecting duct cells was seen. In specimens of human smooth, cardiac, and skeletal muscle, cerebellum, lung, thyroid gland, liver, proliferative endometrium, and placenta, PTA bound only to endothelial of capillaries and larger vessels. In epidermis and gingiva, PTA conjugates additionally revealed reactivity with keratinocytes. Similarly, in salivary gland, urinary bladder, gastrointestinal tract, mammary gland, and renal pelvis, PTA reacted with some epithelial cell layers. The PTA conjugates gave an even cell surface membrane staining of cultured umbilical vein endothelial cells. Lectin-affinity binding of radioactively surface-labeled endothelial cells showed that PTA and Ulex europaeus I agglutinin (UEA-I) recognized related major cell surface glycoproteins. The results with PTA conjugates show that certain N-acetyl galactosaminyl residues are, in addition to some epithelial cells, confined to endothelial cells in human tissues.

Patterns of expression of intermediate filaments in ameloblastoma and human fetal tooth germ.

Monoclonal antibodies (Mab) were used to study the expression of cytokeratins and vimentin in various histological types of ameloblastoma and in human fetal tooth germ. The ameloblastoma and the tooth germ epithelia showed characteristics of both simple glandular and stratified squamous epithelial cells. Cytokeratin No. 18 was detected focally in most ameloblastomas studied but not in fetal odontogenic epithelia. Cytokeratins Nos. 8 and 19 were expressed in all epithelial elements of ameloblastomas and tooth germs. Only two tumors showed focally characteristics of keratinizing epithelia also seen in dental lamina but not in the enamel organ. All tumors except the granular cell ameloblastoma showed a variable coexpression of vimentin and cytokeratins in their neoplastic epithelia. A similar coexpression was detected in the stellate reticulum cells of the developing tooth. Ameloblastoma and human tooth germ epithelia share complex pattern of cytokeratin polypeptides together with coexpression of vimentin. The results strongly support the theory that ameloblastomas are of odontogenic origin and not direct derivatives of basal cells of oral epithelium or epidermis.

Expression of intermediate filaments (IF) in tissues and cultured cells.

Intermediate filaments are found in most nucleated cells as part of their cytoskeleton. Intermediate filaments are formed by different proteins in cells of major tissues types. Therefore, antibodies against intermediate filaments can be used in tissue typing, in the analysis of cell lineages during development and in the elucidation of the origin of unknown tumors.

Saccharide residues in human gingiva as revealed with fluorochrome-coupled lectins.

The histochemical binding of 16 fluorochrome-conjugated lectins to human marginal gingiva was investigated. Of a total of 14 galactose/N-acetylgalactosamine (Gal/GalNAc)-specific lectins, Dolichos biflorus (DBA), Helix pomatia (HPA), and Helix aspersa agglutinins (HAA) were blood group A-reactive whereas Griffonia simplicifolia I-B4 (GSA-I-B4) and Sophora japonica (SJA) agglutinins were blood group B-reactive. HPA, HAA and GSA-I-B4 bound to all suprabasal epithelial cells and to vascular endothelia in tissues with compatible blood groups and detected only upper epithelial cells in tissues lacking the respective blood group antigens. SJA, on the other hand, bound to suprabasal epithelial cells and to endothelial cells in specimens from blood group B, AB and A individuals. DBA gave a heterogeneous labeling of upper epithelial cells in blood group A, AB and B specimens but not in O specimens and did not react with endothelia in any of the tissue samples. DBA bound, instead, consistently to mast cells in gingival lamina propria. Of the other Gal/GalNAc-reactive lectins, 2 bound to suprabasal epithelial cells and 7 to all viable cell layers in gingival epithelium. The binding of these lectins was blood group-independent. Of the fucose-specific lectins, Ulex europaeus agglutinin I (UEA-I) gave an intense suprabasal cell membrane-type of epithelial fluorescence in blood group O specimens and a more diffuse staining in other specimens and recognized endothelial cells in a blood group-independent way. Anguilla anguilla agglutinin (AAA) gave a blood group-independent epithelial staining and bound heterogeneously to endothelial cells only in blood group O samples.(ABSTRACT TRUNCATED AT 250 WORDS)

Platelet glycoprotein IIb/IIIa complex in cultured cells. Localization in focal adhesion sites in spreading HEL cells.

A panel of mouse monoclonal antibodies (MoAbs) was raised that react with platelet glycoproteins (GP) IIb or IIIa. On immunofluorescence, the MoAbs reacted with 30% to 40% of the human erythroleukemia (HEL) cells. When the HEL cells were induced to spread on fibronectin in the presence of 12-O-tetradecanoylphorbol-13-acetate (TPA), the MoAbs reacted with the focal adhesion sites. Only some of the GPIIIa MoAbs reacted with cells other than platelets, megakaryocytes, or HEL cells, and these showed focal adhesion sites in cultured human endothelial cells, fibroblasts, and epithelial cells from normal kidney tubules. They did not react, however, with transformed fibroblasts, fibrosarcoma cells, cultured cells from hypernephromas, or cultured human amnion epithelial cells. The results suggest that the platelet-type GPIIb/IIIa complex is only expressed in cells showing an ability to define megakaryoblastic differentiation. Localization of the GPIIb/IIIa complex at the induced focal adhesion sites in HEL cells and localization of the GPIIIa-like molecules in other cells suggest their direct role in the adhesion process and in the actomyocin organization of adherent cells.