In vitro experimentation indicates that periodontitis-associated bacteria contain potent polyclonal B-cell activators (PBA). We reasoned that if PBA were operative in vivo, plasma cells specific for nonoral antigens should be present in the inflamed gingival tissues, which are characterized by a plasma cell infiltrate. To test this, rabbits with experimental periodontitis were immunized in the hind legs with the histochemically detectable antigen horseradish peroxidase (HRP) or glucose oxidase (GO). At various times after secondary immunization, inflamed gingival tissue was removed, sectioned, and treated histochemically to reveal plasma cells that specifically bound HRP or GO. Remarkably, by 9 days after secondary immunization, hundreds of HRPor GO-binding plasma cells were found in the inflamed gingival tissue of immunized rabbits. The presence of these plasma cells, observed 7 to 10 days after booster immunization, was further substantiated by the presence of large amounts of locally produced HRPor GO-specific antibody in gingival crevicular fluid. By 1 month after secondary immunization, the number of antigen-binding plasma cells had decreased dramatically, but a small number of antigen-specific plasma cells were detected for as long as 9 months after secondary immunization. The large number of HRPor GO-specific plasma cells observed 9 days after immunization led us to see whether recently stimulated cells were more susceptible to PBA. Peripheral blood lymphocytes (PBL) were obtained at different times after booster immunization and cultured in the presence or absence of a PBA from Fusobacterium nuckatum. At 7 days after immunization, PBL spontaneously differentiated into antibody-forming cells in culture, and this process was enhanced by PBA. In contrast, PBL taken months after immunization produced little antibody in culture, and enhancement by PBA was difficult to detect. Compared with resting B cells, the recently stimulated B cells clearly differentiated more readily into antibody-forming cells. In conclusion, antibody synthesis specific for nonoral antigens did occur in inflamed gingival tissue, and a number of mechanisms, including PBA, probably contributed to this phenomenon.

Plasma cells are common in chronically inflamed sites, including periodontal lesions. The aim of this study was to determine which factors contribute to this local accumulation of plasma cells. Specifically, we sought to evaluate the effects of specific antigen and nonspecific activators from an infectious agent associated with chronic inflammation (Fusobacterium nucleatum, an organism prominent in chronic periodontal lesions) and the effect of the chronic inflammation itself. Chronic inflammation (14 to 17 days) was induced in horseradish peroxidase (HRP)-immune rabbits by subcutaneous injection of 50 ,ul of sterile alum in several sites in their backs. Controls included sites injected with saline or more acute sites examined after 3 days of alum inflammation. Sites were challenged with HRP (the antigen), sonicated F. nucleatum (the nonspecific activator), or both together to see whether F. nuckatum has an adjuvant effect. Three days after challenge, HRP-specific antibody-forming cells (AFC) were enumerated after peroxidase histochemistry. In noninflamed sites or sites with acute inflammation, virtually no HRP-specific AFC were evident. In contrast, chronic inflammation alone was sufficient to elicit a specific AFC response (=10 cells per mm2). Addition of either F. nucleatum or HRP to the chronic lesion about doubled the number of HRP-specific AFC. However, a dramatic 8-to 15-fold (80 to 150/mm2) increase was seen in chronically inflamed sites challenged with antigen and activator together. Interestingly, the activator did not have this adjuvant effect in the acute sites or in normal skin. In short, accumulation of plasma cells in inflamed sites is promoted by chronic inflammation, activators of microbial origin, and specific antigen. This milieu can be expected to develop in some periodontal lesions and could help explain why gingival crevicular fluid from some sites may contain extraordinary levels of locally produced specific antibodies for certain antigens. * Corresponding author. showed that, in recently primed rabbits, chronic inflammation, antigen, or nonspecific activator alone or in any combination could induce local accumulations of AFC. However, by far the most potent recruiter of antigen-specific AFC was the combination of antigen, activator, and chronic inflammation. The fact that combinations of antigen, non

No abstract