The effects of Treponema denticola and its outer membrane-bound chymotrypsin-like proteinase on periodontal ligament epithelial cell cultures at different stages of maturity were studied. In sparse cultures with migrating epithelial cells, large intracellular vacuoles were formed rapidly following exposure to live T. denticola. Treponemes showing structural damage were seen occasionally inside membrane-bound vesicles. Intensive membrane blebbing occurred in infected cells and continued for up to 48 h before the cell died. Blebbing could also be induced by a purified chymotrypsin-like proteinase of T. denticola. Cortical actin and ␣-actinin of the bacterium-treated cells showed disorganization, and pericellular fibronectin was degraded by both whole T. denticola and the isolated proteinase. Epithelial cells with well-formed lateral cell contacts appeared to be more resistant to the effects of T. denticola than migrating isolated cells. In multilayer epithelial cultures, adhesion of T. denticola and membrane blebbing were observed infrequently. There was no evidence of invasion of T. denticola into epithelial multilayers. However, immunogold electron microscopy showed rapid transport of T. denticola chymotrypsin-like proteinase into newly formed large intracellular vacuoles within the epithelial layers. These vacuoles were lined by membranes studded with ribosomes. T. denticola-treated epithelial multilayers had loose cell contacts, collapsed intercellular spaces, and increased permeability. Through its capacity to cause these unique cytopathic effects, the chymotrypsin-like proteinase of T. denticola has the potential to contribute to the initiation of periodontal disease.

The unique features of junctional epithelium involve lack of keratinization, limited differentiation and a relatively permeable structure. In order to study the relationship between differentiation and permeability of stratified epithelium a model system was developed. Porcine periodontal ligament epithelial cells were cultured on the polycarbonate nucleopore membrane of the Transwell two-compartment culture system. Within 5 days of culture the cells formed a confluent multilayered structure. Subsequently, maturation of the structure and differentiation of surface cells took place. Transmission electron microscopy showed that the cells were arranged into basal and suprabasal layers with sparse desmosomal attachments and wide intercellular spaces resembling the organization of junctional epithelium. The basal cells attached to a subepithelial basal lamina through numerous hemidesmosomes. The cytokeratin profile of the cultured epithelium (K5, 6, 14, 16, 19) resembled that of the cells of junctional epithelium attached to the tooth surface. The older cultures expressed differentiation markers, K4, K13 and involucrin, thereby resembling sulcular epithelium. The epithelial permeability, measured by diffusion of phenol red, radioactive dextran or methionine tracers, and as transepithelial electrical resistance, decreased with the increased cell number and maturation of the cultures. The new model provides an organotypic culture system which allows to control differentiation of a multilayered periodontal epithelium. It thus may serve as a valuable new tool for studies on the permeability and behaviour of periodontal epithelium under the influence of exogenous and endogenous factors.

No abstract

Osteopontin is an acidic phosphoprotein containing casein kinase II (CKII) phosphorylatable sites and an acidic amino acid cluster. The metabolically 32P-labelings of both serines and threonines in vitro in osteopontin immunoprecipitated from rat osteoblast-like ROS 17/2.8 cells may suggest that casein kinase II catalyzes this modification. The enzyme occurs in microsomal fractions of rat osteoblast-like ROS 17/2.8 cells. Subcellular fractions containing endoplasmic reticulum and Golgi apparatus were isolated by differential centrifugation and were identified according to their ultrastructures and the presence of marker enzymes such as glucose-6-phosphatase and thiamine pyrophosphatase, respectively. both fractions phosphorylated the partially dephosphorylated osteopontin and the specific substrate peptide RRREEETEEE. Endoplasmic reticulum-catalyzed peptide phosphorylation was 2.7 times lower than that of Golgi although both endoplasmic reticulum- and Golgi-catalyzed peptide reactions were 50% inhibited by 20 and 100 ng/ml heparin, respectively. Western blot analysis revealed that both fractions contained osteopontin and microsomal CKII. Furthermore, microsomal CKII was immunogold-labeled in endoplasmic reticulum and Golgi apparatus. Heparin inhibition and utilization of [gamma-32P]GTP as a phosphate donor by both fractions confirmed their capacity to phosphorylate osteopontin. The results suggest that microsomal CKII modifies the acidic matrix proteins during transportation. These matrix phosphoproteins may participate in the mineralization process of hard tissues.

Microsomal casein kinase II (mCKII) is a membrane-bound enzyme present in the microsomal fractions of ROS 17/2.8 osteoblast-like cells. It phosphorylates acidic matrix phosphoproteins such as phosphophoryn and osteopontin. Addition of 1.0% Nonidet P-40 facilitates extraction of the optimum amount of detergent-solubilized and -activated enzyme from microsomal fractions. mCKII was partially purified over 3000-fold by sequential chromatography over DEAE-cellulose and heparin-agarose. SDS-polyacrylamide gels, showed that mCKII contained 43 kDa and 31 kDa polypeptides, corresponding to the alpha- and beta-subunits of the enzyme, respectively. The alpha subunit was identified by anti-CKII antiserum and the beta subunit, by its ability to undergo autophosphorylation. The enzyme was inhibited by 50% with 0.4 micrograms/ml heparin and stimulated by 100% with 1.0 mM spermine when casein was used as a substrate. The phosphorylation of phosphophoryn was reduced to 50% by 0.8 micrograms/ml heparin, but was increased to 2-2.5 fold by 5 to 15 mM spermine, which may be due to substrate-directed effects. Kinetic analysis showed that the apparent Km values for phosphophoryn (0.39 microM) and for osteopontin (2.1 microM) were lower than that for casein (21.3 microM). Vmax values of phosphophoryn and osteopontin were 2.2-fold and 4.6-fold higher than that of casein. Using the ratio Vmax/Km as a measure of kinetic specificity, osteopontin and phosphophoryn appear to be the more specific substrates than casein for mCKII. Thus, both proteins can be considered as physiological substrates for mCKII.