SEROLOGIC AND OPSONIC STUDIES OF BACTEROIDES FRAGILI

An experimental model for the production of Bacteroides abscesses in mice was used to evaluate the ability of counterimmunoelectrophoresis (CIEP) to detect specific Bacteroides antigens in these abscesses. The model was also used to investigate the ability of factors present in serum from abscess bearing mice to opsonize B. fragilis for phagocytosis by mouse peritoneal macrophages.;In the initial experiments, mice were inoculated with various Bacteroides species to induce subcutaneous abscesses. The success rate of abscess induction depended upon the species of Bacteroides inoculated and ranged from 78.9 to 100 percent. Bacteroides antigens were heat-extracted from the purulent contents of the abscesses and were evaluated by CIEP analysis against homologous and heterologous rabbit antisera produced against whole Bacteroides cells. Precipitation bands indicated the presence of antigens in 18.8 to 100 percent of the abscesses tested. No reactions were evident when extracts were tested against heterologous antisera. The frequency of antigen recovery was associated with the abscess-inducing ability of the particular organism tested.;Subsequent attempts were made to detect the presence of B. fragilis antigens in 38 clinical specimens by CIEP testing. Eight of 11 specimens positive for B. fragilis by culture also yielded positive CIP tests. The detection of B. fragilis antigens in human clinical specimens demonstrated a sensitivity of 72.7 percent, a specificity of 58.3 percent, and a predictive value of 44.4 percent.;In a second series of experiments, comparisons were made between the opsonic activity of sera collected from normal mice or mice that were infected with B. fragilis (immune serum). Immune serum demonstrated a significantly greater amount of opsonization than normal serum, although normal serum also possessed a significant amount of opsonizing activity compared to serum-free controls. Macrophages collected from either infected or normal mice ingested B. fragilis to the same extent.;Heat inactivation and bacterial adsorption of sera indicated that the opsonic activity of immune serum was dependent upon the presence of specific antibody and complement, although either alone was capable of significant opsonization. Normal serum was devoid of antibody detectable by passive hemagglutination techniques and relied entirely on heat labile components, presumably complement, for its opsonic activity. The opsonic activity of immune serum was associated with increased levels of IgG1, IgG2a, and IgG2b in these sera. Reduction of immune serum with 2-mercaptoethanol demonstrated that IgM played a minor, but significant, role in opsonization.;A stock strain of B. fragilis was ingested equally well by mouse macrophages under aerobic or anaerobic conditions. However, a clinical isolate of this organism was more resistant to phagocytosis under anaerobic compared to aerobic conditions. Furthermore, the stock culture was more easily opsonized by normal mouse serum than the clinical isolate.;In vitro, mouse macrophages ingested and killed greater numbers of B. fragilis organisms in the presence of immune mouse serum than in normal serum or serum-free media. Normal serum also demonstrated significant ability to mediate ingestion and killing of the organism when compared to serum-free controls. B. fragilis was incapable of survival within macrophages for prolonged lengths of time.