The placement of a ligature around the second maxillary molar of the conventional rat caused osteoclastic bone resorption and simultaneously, alveolar bone formation. The number of peripheral mononuclear cells in the blood and lymphoblastic transformation of spleen cells in response to concanavalin A increased. Cyclophosphamide (CY), an immunosuppressive agent, given shortly after placing the ligature suppressed the lymphoid reactions, spleen size, and bone formation and enhanced bone destruction. CY given in higher doses also suppressed the number of PMN cells. Septicemia developed in several of these animals. Pseudomonas aeruginosa, Klebsiella pneumoniae and Escherichia coli were isolated from the blood and/or ligature. Antibiotics prevented bone destruction. Without placing a ligature, the high dose of CY did not result in bone loss. These findings suggest that 1) bone destruction of the ligature‐treated rat is of bacterial origin, 2) CY suppresses proliferation of osteoblasts but does not seem to interfere with the activity of osteoclasts, and 3) suppression of the host defenses greatly facilitates bone destruction.

Oral isolates of Actinobacillus actinomycetemcomitans (strain Y4) release spherical microvesicles in large numbers during normal growth. The biological activities of these products were studied, and it was estimated that approximately 1/10 of their dry weight was made up of heat-and proteolysis-resistant endotoxin. The chicken embryo lethality and bone-resorbing activity of the microvesicles were heat stable but proteolysis sensitive. Other laboratories have reported the presence of a heat-and proteolysis-sensitive leukotoxin in similar preparations. Accordingly, the microvesicles released by strain Y4 may contain, in addition to endotoxin, several potent substances which are highly toxic and active in bone resorption, and these may be significant factors in the pathogenesis of periodontal diseases.

The development of humoral and cell-mediated immune responses to Eikenella corrodens (a bacterium that causes periodontal lesions in gnotobiotic rats) was measured and compared with the rate of appearance of macroscopic lesions. A possible inverse relationship was found. A strong cell-mediated immune response, as measured by skin reactivity and lymphocyte mitogenesis, occurred between 4 and 6 weeks after infection and subsided soon thereafter to a low response level. Humoral antibodies to endotoxin from E. corrodens could not be detected at any time. The disease developed only after the cell-mediated immune response diminished, thus suggesting that lack of an efficient immune response may permit the development of the disease. This is seemingly in contradiction to the assumption that tissue destruction in such cases is caused by the immune response and its products. We are inclined to believe, based on our findings reported here, that the lack of immune responsiveness to the bacterium and/or its products is the major causative factor in the development of periodontitis. At the same time, we wish to emphasize that occurrence of both phenomena during the long development of periodontal disease is possible.

Eikenella corrodens can induce periodontitis-like disease in gnotobiotic rats. Some components of this bacterial cell elicit measurable humoral immune response during the development of the disease, but in this system endotoxin is not among the efficient immunogens. Because no humoral immune response could be seen to the endotoxin of Eikenella corrodens it is assumed that this endotoxin can act uncontrolled in monoinfected rats. Accordingly, the lack of protective humoral immune response to pathogenic components of Eikenella corrodens may be the major factor permitting the development of the disease described here. The possibility that both cell-mediated immunity and uncontrolled endotoxic action are parts of the pathomechanism of the disease is supported by our observations.