Search

Anaerobic, agar-dilution, minimal inhibitory concentrations (MICs) of 18 antibiotics are given for the numerically important bacterial groups from the human oral cavity. Strains are divided into susceptibility categories using the guidelines for interpretation of MICs suggested by the National Committee for Clinical Laboratory Standards. These guidelines are based on data on antibiotic concentrations attainable in serum following various dosage regimens. MICs are also compared with attainable gingival fluid levels where these are known. The highest percentages of strains were susceptible to tetracycline, with 89% of the 139 strains tested susceptible to serum levels and 97% conditionally susceptible to attainable gingival fluid levels. Ninety-eight percent of strains were conditionally susceptible to attainable gingival fluid levels of minocycline, but many strains, including Actinobacillus actinomycetemcomitans, were only moderately susceptible to attainable serum levels of this tetracycline analogue. Carbenicillin was effective against most groups of organisms, with the important exception of A. actinomycetemcomitans, at serum levels attainable with oral formulations of carbenicillin. Only 2% of the total strains tested were resistant to penicillin, while 33% of strains were categorized as moderately susceptible. Clindamycin was active against many strains of Gram-negative bacteria but was not active against A. actinomycetemcomitans, some Bacteroides, Eikenella corrodens, or the anaerobic vibrios. Metronidazole was active against A. actinomycetemcomitans, all five groups of oral Bacteroides tested, and against Capnocytophaga species. Chloramphenicol was active against A. actinomycetemcomitans, but not against most of the other groups of oral organisms. Nearly all groups contained strains non-susceptible to serum levels attainable with the usual doses of erythromycin, spiramycin, vancomycin, kanamycin, neomycin, streptomycin, doxycycline, oxytetracycline, or chlortetracycline; several strains were resistant to maximum attainable serum levels of each of these antibiotics except doxycycline.

Potential infectivity of subgingival plaque of Macaca arctoides in guinea pig leg muscle was examined before, during, and after chlorhexidine treatment. The regime of chlorhexidine administration did not appear to alter the monkey subgingival microflora markedly and had no demonstrable effects on the pathogenicity of the subgingival plaque. These data confirm previous suggestions that chlorhexidine essentially does not penetrate into the deeper part of periodontal pockets following a supragingival mode of administration.

Immunological tests including immunofluorescence microscopy, latex agglutination, flow cytometry, and enzyme-linked immunosorbent assays are being developed for the detection of carcinogenic microorganisms and periodontal pathogens. In the present study, indirect immunofluorescence microscopy was used to examine two hundred eighty-three subjects for Actinobacillus (“Haemophilus”) actinomycetemcomitans and Bacteroides gingivalis in subgingival dental plaque samples pooled from the mesial surface of the four first molar teeth. Alveolar bone height on panoramic radiographs was measured at each sample site using a Schei ruler and correlated with the results of immunofluorescence microscopy. Over ninety percent of the subjects did not demonstrate either putative periodonlal pathogen or harbored only trace amounts. Only 5% of the subjects demonstrated 2% or greater A. actinomycclemcomitans and no subject harbored more than 5% consistent with the finding that there were no localized juvenile periodontitis patients in this subject group. Nine percent of the subjects revealed 2% or more, and 5% demonstrated 5% or more of the total cell count as B. gingivalis. K. gingivalis levels, however, did exhibit a significant positive correlation with loss of alveolar bone height. The use of immunofluorescence microscopy for B. gingivalis exhibited 19% sensitivity, 96% specificity, 69% positive predictive value, and 69% negative predictive value when correlated with alveolar bone height measurements. The present data indicate that indirect immunofluorescence microscopy is a useful, rapid method for the detection of A. actinomycetemcomitans and B. gingivalis in subgingival dental plaque which can be used in the clinical diagnosis, treatment, and monitoring of periodontal disease. The sensitivity of microbiological assays as indicators of periodontal disease can likely be increased by utilization of additional Immunological tests for other putative periodontal pathogens.

Several antibiotics were found to adsorb to saliva-coated enamel and to inhibit in vitro plaque formation by pure cultures of oral bacteria: Actinomyces viscosus, Actinomyces naeslundii and Streptococcus mutans. Tetracycline, minocycline and oxytetracycline adsorbed to the greatest degree, showing 100-fold higher adsorption than spiramycin, the test antibiotics with least adsorption. Inhibition of in vitro plaque formation was found to require both drug substantivity (capacity for adsorption) and antimicrobial activity. Inhibition of plaque formation in the in vitro assay employed correlated well with clinical efficacy.

Actinobacillus actinomycetemcomitans is an oral bacterium which is being encountered with increasing frequency in infective endocarditis. This organism occurs in high numbers in periodontitis lesions of patients with localized juvenile periodontitis (periodontosis). It is present infrequently, and only in low numbers in most other individuals. Its common resistance to penicillin, erythromycin and vancomycin represents a clinical problem in patients at risk of developing endocarditis after dental treatment. However, the high activity of tetracyclines against A. actinomycetemcomitans may be useful in prophylactic endocarditis considerations by allowing a suppression of the organism prior to the institution of recommended prophylactic protocols. In this study, we determined the effect of systemic tetracycline-HCl therapy (1 gm/day) on the oral A. actinomycetemcomitans population in five localized juvenile periodontitis patients who were heavily infected with the organism. A. actinomycetemcomitans could not be detected in samples of subgingival and supragingival dental plaque and cheek mucosal surfaces following 14 days of administration of systemic tetracycline. The organism was still undetectable 3 weeks after therapy but it reappeared at a few oral sites at week 8 post-treatment. On the basis of this data, it is proposed that the prophylactic endocarditis therapy of patients with high numbers of penicillin-resistant A. actinomycetemcomitans include a two-stage approach: first, the systemic administration of tetracycline for 14 days, and second, institution of a conventional prophylactic protocol during the time of dental treatment.

Shortly after strains of Streptococcus mutans were shown to be cariogenic, they were found to produce dextran and levan polymers from sucrose. These polymers are responsible for the ability of S. mutans to adhere to, and colonize smooth surfaces of teeth (1–3; cf. reviews 4,5). The large molecular weight dextrans and levans are necessary for the formation of plaques on teeth which in turn, appears to be a requirement for cariogenicity. This is supported by the finding that a mutant of S. mutans, lacking the ability to produce these polymers, is no longer cariogenic (6). We have shown the adherence of S. mutans to wires can be inhibited by antiserum to whole S. mutans cells (7). In the same experiments we found that the production of cell-associated polysaccharide was inhibited by antisera, while cell growth was actually increased. The adherence of S. mutans to a glass surface has also been shown to be inhibited by antiserum to whole cells (8). Experiments in our laboratory have shown that antisera to S. mutans will inhibit the glucosyltransferases present in S. mutans culture fluids. These enzymes synthesize polymeric dextrans from sucrose (9). Recently, evidence has been presented (10) showing that the extracellular or soluble glucosyltransferases of S. mutans become cell-associated as insoluble dextran is synthesized. These results suggest that there is a receptor on the bacterial cell surface which binds glucosyltransferase-dextran complexes. The finding that whole cell antisera inhibit the adherence of S. mutans to wire and glass surfaces, inhibit the production of cell-associated polysaccharide synthesis while not inhibiting cell growth, and inhibit the synthesis of polyglucan-containing polysaccharides from sucrose by S. mutans glucosyltransferase enzymes suggests that antibodies prevent S. mutans from colonizing dental surfaces by interfering with the production of adherent polysaccharide polymers. The studies to be reported here show that the specificity of antibody-mediated inhibition of S. mutans adherence, and inhibition of glucosyltransferase activity do not always follow the specificity of surface antigens which are used to classify strains of S. mutans (11). The evidence presented indicates that adherence does not depend upon an antigen common to all S. mutans and hence if caries immunization via inhibition of adherence is to be successful, multivalent vaccines may be required.

Enzymatic proteolysis of human immunoglobulin A yields Fabα fragments but the Fcα is degraded to small peptides. The unavailability of Fcα has hampered to some extent certain studies of IgA, e.g., its primary structure and the attachment site of secretory component and J chain. Recently, while studying human intestinal immunoglobulins, it was observed by Mehta et al. (1) that normal feces contain an IgA fragment which was shown to be Fcα, and that feces contain a proteolytic enzyme which cleaves both serum and secretory IgA to yield Fabα and Fcα. Of considerable interest is that the enzyme is produced by bacteria in those regions of the alimentary tract heavily populated by bacteria, principally the mouth and the colon. We have termed the enzyme IgA protease because enzymatic activity has thus far been found only against IgA.

Secretory γA antibodies were induced in the colostrum of New Zealand Red rabbits by inejection of Group A streptococcal vaccine into the mammary glands. Colostral antibodies were purified by elution from Group A cells used as solid immunoadsorbent. Antibodies of both secretory γA and γG class were obtained upon elution with 5 per cent N-acateyl glucosamine (NAG) or pH 2·5 buffer. The eluates were filtered through Sepharose 6B to separate the 11 S γA from the 7 S γG antibodies. From 4·2 to 13·1 μg of γA per ml of original colostrum was obtained by elution with 5 per cent NAG and from 20·4 to 208·8 μg of γA per ml original colostrum was obtained by elution with pH 2·5 buffer. The amount of γG antibody obtained in these eluates varied greatly, ranging from 90 to 10 per cent of the total antibody recovered. Purified γA anti-NAG antibodies isolated from a single rabbit exhibited the following properties: (1) They agglutinated Group A cell was. (2) They bound to Group A cell was as demonstrated by indirect fluorescent antibody straining using monospecific fluorescin-labeled goat anti-rabbit γA. (3) They failed to precipiate with Group A carbohydrate under conditions in which the γG anti-NAG antibody was found to precipitate. (4) Secretory γA antibodies, however, inhibited the precipitation of the γG antibodies with Group A carbohydrate. (5) They contained secretory component as detected by gel diffusion tests. (6) The secretory γA antibodies bound triated p-nitrophenyl-β-N-acetyl glucosaminide in equilibrium dialysis experiments. A plot of r vs c for the binding of the hapten by γA antibodies shows that r increased with increasing hapten concentration and that there was a plateau as r approached 4, indicating that secretory γA may have a valence of 4. An association constant of the order of 10 5 l/m could be estimated from the data; however, because of the low association constant there was considerable sctter of points. The association constant for γG anti-NAG, isolated from the same colostrum was about one log lower when tested with the same hapten. Neither γA or γG antibodies bound detectable amounts of triated N-acetyl glucosamine, suggesting that the β-linkage makes an important contribution to binding. Many of properties of described for secretory γA antibody can be explained by a preference of this antibody to bind most of its combining sites to a single molecule of possessing multiple identical antigenic determinants as opposed to cross-linking two or more such molecules. This property, termed monogamous polyvalency, could significantly enhance certain biologic functions of secretory γA.

Research in oral microbiology since June 1966 is reviewed. The studies have concentrated on microbial colonization of the teeth and dentogingival regions. Much of the material reported here is discussed with reference to the microbiology and pathogenic potential of dental plaque.

THE AIM OF THIS STUDY was to determine the degree to which clinical classifications based on cross-sectional assessments endure in the course of development of early-onset periodontitis (EOP), and to introduce new criteria which might improve the clinical classification of these diseases. Subjects with EOP and a matched group without EOP were identified within a national probability sample examined during the 1986/87 survey of US schoolchildren. Of these, 265 subjects (mean age 16 years) were re-examined during the 1992/93 school year. The clinical attachment level of teeth was assessed, and the individuals were classified into localized juvenile periodontitis (LJP), generalized juvenile periodontitis (GJP), incidental attachment loss (IAL), and no-periodontitis groups using three classification methods previously described. A fourth method that considered the extent and severity of attachment loss and the number of missing teeth was introduced to classify the individuals at baseline and at follow-up as having localized, generalized, or incidental EOP, and no-periodontitis groups. Furthermore, the individuals were classified using criteria based on the rate and pattern of change in attachment loss during 6 years. The results showed low correlations between the baseline classifications and the classifications at the 6-year follow-up examination, irrespective of the method used. In addition, the cross-sectional classifications were not predictive of the rate of progression of periodontal disease in these subjects. In the generalized disease group, two-thirds of the individuals exhibited moderate/rapid disease progression, while one-third had slow or no progression. In the localized disease group, one-half of the individuals had moderate/rapid disease progression and one-half had slow or no progression. In the incidental disease group one-fourth of the individuals had moderate/rapid disease progression and three-fourths had slow or no progression. We propose that the term early-onset periodontitis be used as a generic term to describe periodontal disease before its normal onset. In addition, we suggest that incidental, localized, and generalized EOP are heterogenous groups comprising rapidly and slowly progressing forms within each classification. The findings suggest that a classification system in which subsets of the disease that are defined according to a combination of cross-sectional criteria and the disease progression may be useful in studies of EOP. Furthermore, the findings suggest that clinical classifications of EOP be used as generic descriptors until a full understanding of the pathogenesis of this disease is accomplished.

We investigated the binding sites of salivary statherin involved in the interaction with Porphyromonas gingivalis recombinant fimbrillin (r-Fim). Synthetic peptides representing statherin analogs were used to localize the binding domains of statherin. Peptide F4 (residues 29 to 43) significantly bound to r-Fim and inhibited r-Fim binding to statherin-coated hydroxyapatite beads. Successive peptides in which pairs of amino acid residues were deleted starting at the N terminus of peptide F4 were synthesized. Peptide N1 without Leu-29-Tyr-30 had significantly reduced direct binding and inhibition ability. The deletions of residues 31 to 40 had little effect on interaction with r-Fim. The tripeptide N6 representing Tyr-41-Thr-42-Phe-43 retained significant binding to r-Fim. Another set of peptides was synthesized by deleting individual amino acid residues from the C and N termini of peptide F4 to identify functional residues among the five putative functional residues 29, 30, and 41 to 43. Peptide C1 missing Phe-43 lost over 50% of its binding ability. Binding ability was gradually reduced with deletions from the peptides. Peptide C5 (amino acids 31 to 40) weakly affected direct binding and inhibition. Collectively, the results of this study suggests that Leu-29-Tyr-30 and Tyr-41-Thr-42-Phe-43 are important binding regions that mediate the binding of statherin to P. gingivalis fimbrillin.

Fimbriae (the oligomeric form of fimbrillin) are considered important in the adherence and colonization of Porphyromonas gingivalis in the oral cavity. In the present study, we have identified the structural domains of P. gingivalis fimbrillin that mediate the binding to salivary proline-rich protein 1 (PRP1) and statherin. A series of synthetic fimbrillin peptides were used to localize the active fimbrillin domains involved in the binding to PRP1 and statherin. The binding of 125I-labeled 41-r-Fim (whole-length recombinant fimbrillin, amino acid [aa] residues 1 to 337) to PRP1-coated hydroxyapatite beads (HAP) was strongly inhibited by the fimbrillin C-terminal peptides corresponding to aa residues 266 to 286 and 318 to 337 (peptides 266-286, and 318-337, respectively), while the binding to statherin was inhibited by C-terminal peptides 266-286, 293-306 and 307-326. Peptide 126-146 also showed a weak inhibitory effect, about half that of other active peptides, on the binding to both PRP1 and statherin. P. gingivalis whole-cell binding to PRP1- or statherin-coated HAP was inhibited by more than 80% by the same active peptides. To confirm that the C-terminal portion of fimbrillin includes domains responsible for the binding, two C-terminally truncated variants of recombinant fimbrillin were generated and purified. These were designated 34.5-r-Fim, corresponding to aa residues 1 to 286, and 32-r-Fim, corresponding to aa residues 1 to 265. 125I-34.5-r-Fim revealed 35 and 34% loss of binding ability to PRP1 and statherin, respectively. 125I-32-r-Fim had significantly less binding ability to PRP1 and statherin than 125I-34.5-r-Fim, which was reduced 78 and 73%, respectively. Whole-cell binding to PRP1-, statherin-, or whole saliva-coated HAP was inhibited up to 100% by 41-r-Fim, while 32-r-Fim also showed considerable inhibition, possibly due to the region of aa 126 to 146. Collectively, these results suggest that there are separate and multiple binding sites for PRP1 and statherin in the P. gingivalis fimbrillin, and the combination of all of these binding sites may be indispensable in establishing stable bacterial adherence to saliva-coated surfaces in the oral cavity.

Periodontal diseases are characterized by an inflammatory exudate termed gingival fluid (GF). We have examined selected complement components for cleavage products in GF from patients with periodontitis and the less common disease, periodontosis. We hope to establish an objective criterion for diagnosis as well as to provide information on the mechanisms involved in producing periodontal inflammation. GF samples (2 µl) were collected from 18 patients diagnosed as having severe periodontitis and blood samples were collected simultaneously. Serum and GF were stored at -76°C until assayed. Protein concentrations were measured by single radial immunodiffusion for nine plasma proteins, including C3, C4, and Factor B. An antiserum to C3 specific for the B determinant was utilized. It was found that GF contained only 20% of the a-C3 (B determinant) reactive material found in serum, while all other proteins measured were at levels 70 to 80% of that in serum.