Selective modulation of bacterial attachment to oral epithelial cells by enzyme activities associated with poor oral hygiene.

The present investigation explored the hypothesis that elevated levels of certain enzymes in the gingival crevicular environment of individuals with poor oral hygiene and/or gingival inflammation may modify the surfaces of epithelial cells and thereby modulate the types of bacteria which attach and colonize. Buccal epithelial cells treated with neuraminidase and certain proteases were used as a model for study. Bacteria studied included Streptococcus sanguis and Streptococcus mitis which have been associated with gingival health, Actinomyces species which are increased in plaque associated with developing gingivitis, and Bacteroides gingivalis, Bacteroides intermedius, and Actinobacillus actinomycetemcomitans which are associated with destructive periodontal diseases. Treatment of epithelial cells with the enzymes studied produced selective effects on their receptivity for bacteria. Neuraminidase treatment of epithelial cells greatly reduced the attachment of all strains of S. sanguis and S. mitis studied. In contrast, the number of Actinomyces viscosus, A. naeslundii and A. israelii cells which attached was significantly increased. Neuraminidase treatment also appeared to enhance attachment of B. intermedius and B. gingivalis. Treatment of buccal cells with trypsin, chymotrypsin or papain also selectively affected bacterial attachment. Such protease treatment greatly reduced the numbers of streptococci and A. viscosus cells which attached, while the numbers of B. gingivalis and B. intermedius were significantly increased. Treatment of epithelial cells with preparations of lysosomal enzymes derived from human PMNs produced similar selective effects. The changes in bacterial adhesion observed by the enzyme treatments studied are consistent with the shifts in the composition of the gingival crevice flora which occur when oral hygiene is terminated and gingivitis develops.

Role of cryptic receptors (cryptitopes) in bacterial adhesion to oral surfaces.

Progress in characterizing the receptors that promote bacterial attachment to teeth and oral epithelial cells has suggested that hidden molecular segments may frequently be involved. Such cryptic receptors, referred to as 'cryptitopes', may become exposed by several mechanisms. Hidden segments of salivary acidic proline-rich proteins evidently become exposed when the molecules undergo a conformational change as they adsorb to apatitic mineral. Adhesins of Actinomyces viscosus and certain other prominent dental plaque bacteria are able to bind to these cryptitopes, and this enables these organisms to bind to proline-rich proteins on apatitic surfaces while avoiding interactions with these proteins in solution. Cryptitopes may also become exposed as a result of enzymatic action. Thus, several bacteria, including Fusobacterium nucleatum, Eikenella corrodens, A. viscosus, A. naeslundii and Bacteroides intermedius, have adhesins that bind to galactosyl receptors which become exposed after treatment with neuraminidase. Similarly, the adhesion of some Gram-negative bacteria, such as Bact. gingivalis, is enhanced when tissue surfaces are treated with certain proteases, or lysosomal enzymes derived from human polymorphonuclear leucocytes. It seems likely that elevated levels of enzymes present in gingival fluid as sequelae of poor oral hygiene and gingivitis may generate cryptitopes for potentially periodontopathic bacteria, and thereby contribute to modulation of the gingival flora.

The predominant cultivable microbiota of crevicular epithelial cells.

The purpose of the present investigation was to compare the composition of the predominant cultivable microbiota associated with gingival crevicular epithelial cells with that of the unattached microbiota recovered from the same site. Samples were taken from 2 diseased sites from 8 periodontal patients, by scraping the epithelial lining of the pocket with a curette. The epithelial cells were separated from the unattached subgingival bacteria by centrifugation in a reduced 50% Percoll density gradient. Epithelial cells formed a band at the top of the gradient and were removed separately from the unattached bacteria located at the base. Each layer was dispersed, diluted and plated on Trypticase soy agar with 5% sheep blood and 50 isolates were characterized from each sample. The microorganisms associated with the epithelial layer harbored 5- to 20-fold higher mean percentages of Bacteroides gingivalis, Bacteroides intermedius and Peptostreptococcus micros. The layer of unattached organisms exhibited 4- to 10-fold higher mean percentages of Streptococcus uberis, Capnocytophaga ochracea, Eikenella corrodens and Veillonella parvula.

Use of Percoll density gradients for studying the attachment of bacteria to oral epithelial cells.

An assay for studying the attachment of bacteria to oral epithelial cells has been developed which utilizes Percoll density gradient centrifugation to separate bacteria and epithelial cells. 3H-thymidine-labeled bacteria were incubated with suspensions of buccal epithelial cells in microtitration plates for 2.5 hr at 35 degrees C. The mixtures were then subjected to density gradient centrifugation in 50% Percoll. Epithelial cells with attached bacteria formed a band near the top of the tube, while unattached bacteria formed a band near the bottom. The epithelial cells were collected on membrane filters, and the number of attached bacteria was determined by scintillation counting. Binding of S. mitis C5 was found to increase with time, and equilibrium was attained within two hr. Saturation of available binding sites occurred when 10(7) S. mitis cells were incubated with 1.5 x 10(4) buccal epithelial cells. The numbers of streptococci which attached as determined with this assay were in good agreement with values obtained by direct microscopic counts. Adsorption of S. mitis C5 cells was adequately described by a Langmuir isotherm (correlation coefficient 0.998). This permitted calculation of estimates of the number of binding sites and the affinity of the organism. The assay proved reliable even when as few as 1000 epithelial cells were used. Treating the epithelial cells with neuraminidase or trypsin significantly decreased the number of S. mitis C5 cells which attached. In contrast, binding of A. naeslundii 12104 to neuraminidase-treated cells was increased, and attachment of B. gingivalis 381 was also enhanced, especially to epithelial cells which had been treated with trypsin.

Biosynthesis of D-alanyl-lipoteichoic acid by Lactobacillus casei: interchain transacylation of D-alanyl ester residues.

Lipoteichoic acid (LTA) from Lactobacillus casei contains poly(glycerophosphate) substituted with D-alanyl ester residues. The distribution of these residues in the in vitro-synthesized polymer is uniform. Esterification of LTA with D-alanine may occur in one of two modes: (i) addition at random or (ii) addition at a defined locus in the poly(glycerophosphate) chain followed by redistribution of the ester residues. A time-dependent transacylation of these residues from D-[14C]alanyl-lipophilic LTA to hydrophilic acceptor was observed. The hydrophilic acceptor was characterized as D-alanyl-hydrophilic LTA. This transacylation requires neither ATP nor the D-alanine incorporation system, i.e., the D-alanine activating enzyme and D-alanine:membrane acceptor ligase. No evidence for an enzyme-catalyzed transacylation reaction was observed. We propose that this process of transacylation may be responsible for the redistribution of D-alanyl residues after esterification to the poly(glycerophosphate). As a result, it is difficult to distinguish between these proposed modes of addition.

Biosynthesis of D-alanyl-lipoteichoic acid: role of diglyceride kinase in the synthesis of phosphatidylglycerol for chain elongation.

Lipophilic and hydrophilic D-alanyl-lipoteichoic acids are elongated in Lactobacillus casei by the transfer of sn-glycerol 1-phosphate units from phosphatidylglycerol to the poly(glycerophosphate) moiety of the polymer. These sn-glycerol 1-phosphate units are added to the end of the poly(glycerophosphate) which is distal to the glycolipid anchor; 1,2-diglyceride results from this addition. The presence of a diglyceride kinase was suggested by the ATP-dependent phosphorylation of 1,2-diglyceride to phosphatidic acid. Inorganic phosphate was used to initiate the synthesis of lipophilic lipoteichoic acid (LTA) and the elongation of both lipophilic and hydrophilic LTA. Three observations suggest that phosphate and other anions play a role in the in vitro synthesis of LTA and its precursors. First, the conversion of 1,2-diglyceride to phosphatidic acid by diglyceride kinase was stimulated. Second, the synthesis of phosphatidylglycerol was increased. Third, the elongation of lipophilic and hydrophilic LTA was enhanced. These observations indicated that one effect of phosphate might be to enhance the utilization of 1,2-diglyceride for the synthesis of phosphatidic acid. This phospholipid is a precursor of phosphatidylglycerol, the donor of sn-glycerol 1-phosphate for elongation of LTA.

Biosynthesis of D-alanyl-lipoteichoic acid in Lactobacillus casei: D-alanyl-lipophilic compounds as intermediates.

D-Alanyl-lipoteichoic acid (D-alanyl-LTA) from Lactobacillus casei contains a poly(glycerol phosphate) moiety that is selectively acylated with D-alanine ester residues. To characterize further the mechanism of D-alanine substitution, intermediates were sought that participate in the assembly of this LTA. From the incorporation system utilizing either toluene-treated cells or a combination of membrane fragments and supernatant fraction, a series of membrane-associated D-[14C]alanyl-lipophilic compounds was found. The assay of these compounds depended on their extractability into monophasic chloroform-methanol-water (0.8:3.2:1.0, vol/vol/vol) and subsequent partitioning into chloroform. Four lines of evidence suggested that the D-alanyl-lipophilic compounds are intermediates in the synthesis of D-alanyl-LTA. First, partial degradation of the poly(glycerol phosphate) moiety of D-alanyl-LTA by phosphodiesterase II/phosphatase from Aspergillus niger generated a series of D-alanyl-lipophilic compounds similar to those extracted from the toluene-treated cells during the incorporation of D-alanine. Second, enzymatic degradation of the D-alanyl-lipophilic compounds by the above procedure gave D-alanyl-glycerol, the same degradation product obtained from D-alanyl-LTA. Third, the incorporation of D-alanine into these compounds required the same components as the incorporation of D-alanine into membrane-associated D-alanyl-LTA. Fourth, the phosphate-induced loss of D-[14C]alanine-labeled lipophilic compounds could be correlated with the stimulation of phosphatidylglycerol synthesis in the presence of excess phosphate. We interpreted these experiments to indicate that the D-alanyl-lipophilic compounds are D-alanyl-LTA with short polymer chains and are most likely intermediates in the assembly of the completed polymer, D-alanyl-LTA.

Biosynthesis of D-alanyl-lipoteichoic acid: characterization of ester-linked D-alanine in the in vitro-synthesized product.

d-Alanyl-lipoteichoic acid (d-alanyl-LTA) contains d-alanine ester residues which control the ability of this polyer to chelate Mg(2+). In Lactobacillus casei a two-step in vitro reaction sequence catalyzed by the d-alanine-activating enzyme and d-alanine:membrane acceptor ligase incorporates d-alanine into membrane acceptor. In this paper we provide additional evidence that the in vitro system catalyzes the covalent incorporation of d-[(14)C]alanine into membrane acceptor which is the poly([(3)H]glycerol phosphate) moiety of d-alanyl-LTA. This conclusion was supported by the observation that the d-[(14)C]alanine and [(3)H]glycerol labels of the partially purified product were co-precipitated by antiserum containing globulins specific for poly(glycerol phosphate). The isolation of d-[(14)C]alanyl-[(3)H]glycerol from d-[(14)C]alanine.[(3)H]glycerol-labeled d-alanyl-LTA synthesized in the in vitro system indicated that the d-alanine was linked to the poly(glycerol phosphate) chain of the LTA. A comparison of the reactivities of the d-alanine residues of d-alanyl-glycerol and d-alanyl-LTA supported the conclusion that the incorporated residue of d-alanine was attached by an ester linkage. Thus, the data indicated that the in vitro system catalyzes the incorporation of d-alanine covalently linked by ester linkages to the glycerol moieties of the poly(glycerol phosphate) chains of d-alanyl-LTA. New procedures are presented for the partial purification of d-alanyl-LTA with a high yield of ester-linked d-alanine and for the sequential degradation of the poly(glycerol phosphate) moiety substituted with d-alanine of d-alanyl-LTA with phosphodiesterase II/phosphatase from Aspergillus niger.