Local production of IgA- and IgM-rheumatoid factors in adult periodontal disease.

The enzyme-linked immunospot assay was used to enumerate both the number and the frequency of spontaneous IgG, IgA, and IgM immunoglobulin-secreting cells and IgA- and IgM-rheumatoid factor (RF)-producing cells present in the gingivae and peripheral blood of adult periodontitis patients. Cells from 29 patients were incubated on plates coated with human IgG, Fc, or F(ab')2 fragments and on plates coated with class-specific anti-human antibodies and secreted antibodies were subsequently visualized by means of an immunoenzymatic procedure. The data indicate that (1) IgA-RF- and IgM-RF-secreting cells are frequently present in the gingiva of adult periodontitis patients; (2) production of RF in gingivae of adult periodontitis patients occurs in the absence of demonstrable RF production by simultaneously obtained peripheral blood mononuclear cells, suggesting that local autoimmune reactions may occur in this disease; and (3) lack of correlation between IgA-RF and IgM-RF production in diseased gingiva suggests that the two RF isotypes are regulated independently of each other.

Analysis of human IgG and IgA subclass antibody-secreting cells from localized chronic inflammatory tissue.

The chronic inflammatory diseases in humans have been intensively investigated, however the immune mechanisms underlying diseases such as rheumatoid arthritis (RA), inflammatory bowel disease, and periodontal disease (PD) remain elusive. In this study, we have analyzed the distribution of IgM, IgG, and IgA secreting cells with emphasis on the IgG and IgA subclasses among mononuclear cell populations isolated from gingiva at different stages of PD. Surgically removed tissues were treated with Dispase to gently dissociate cells and the Ficoll-Hypaque gradient centrifugation was used to enrich for viable mononuclear cells rich in lymphocytes, macrophages, and plasma cells. The total numbers of plasma cells increased with the severity of disease. Immunofluorescence analysis showed that most Ig-containing cells were of the IgG isotype; however, significant numbers of IgA-positive cells but few IgM-positive cells were seen. This isolation procedure allowed analysis, at the single cell level, of the distribution of IgG and IgA subclasses of antibody-secreting cells with monoclonal antibodies to human IgG and IgA subclasses. For this, we selected four monoclonal anti-IgG subclass (anti-gamma 1, -gamma 2, -gamma 3, and -gamma 4) antibodies with no subclass cross reactivity for use in the enzyme-linked immunospot assay. Analysis of slight, moderate, and advanced stages of PD showed a progressive increase in spotforming cells (SFC) numbers, and the major isotype of SFC was IgG followed by IgA. The major IgG subclass SFC seen was IgG1 followed by IgG2 whereas similar numbers of IgG3 and IgG4 SFC were observed, a pattern also seen with cells from synovium of RA patients and in mitogen-triggered spleen and PBMC. In terms of the IgA subclass distribution, IgA1 predominated in moderate stages, whereas a selective increase in IgA2 SFC were seen in the more advanced stage of PD. These results show that significant numbers of viable plasma cells/Ig-secreting cells can be isolated from inflamed gingival tissues. Further, careful analysis has shown that IgG subclass responses in gingiva are similar to those found in synovia of RA subjects, and in stimulated PBMC and spleen. However, it should be noted that the number of IgG4- and IgA2-secreting cells increased in the advanced stage of PD.

Autoimmunity to collagen in adult periodontal disease.

Although it has been documented that exogenous antigens of microbial origin are involved in the induction of the local inflammatory responses in human adult periodontitis (AP), endogenous antigens may contribute to the chronicity of this common disease. In this study, we used the enzyme-linked immunospot (ELISPOT) test to enumerate antibody-secreting cells to human collagen Types I-VI by cells isolated from the gingivae and peripheral blood of AP patients. Analyses of dissociated cells from gingivae of 39 AP patients revealed the presence of high numbers of cells that secrete antibodies to Type I collagen, and to a lesser extent, Type III. Although the majority of such cells produced specific antibodies of the IgG class, IgA- and IgM- anti-collagen -secreting cells were also detected. When compared to the total antibody-producing cells, the numbers of cells forming specific antibodies to collagen Type I were surprisingly high. In contrast, anti-collagen antibody-producing cells were rarely detected in the peripheral blood of patients with adult periodontal disease and only low levels of anti-collagen antibodies were present in the serum. The finding of local production of anti-collagen antibodies in AP suggests that autoimmunity may contribute to the pathogenesis of this common disease.

HNK-1+ (Leu-7) cells and natural killer cell activity in inflamed human gingival tissue.

The presence of HNK-1 (Leu-7)-positive cells and natural killer (NK) cell activity was determined in human periodontal tissues. Gingival tissues obtained from 25 adult patients were processed for analysis utilizing a HNK-1 (Leu-7) mouse monoclonal antibody. A subpopulation of non-adherent lymphoid cells obtained by collagenase digestion of inflamed gingival tissues from 10 patients was examined for the presence of large granular lymphocytes (LGL) by May-Grünwald-Giemsa staining and for NK cell activity against K562 cells by a 51Cr release cytotoxicity assay. HNK-1+ cells were identified in gingival tissue sections of 21 patients, and were present in or close to discrete foci of plasma cells. HNK-1+ cells were scarce in mildly inflamed or uninflamed tissues sections. LGL were identified in 9 of 10 gingival single-cell suspensions and constituted approximately 5% of the gingival cell population. NK cell-mediated cytolysis, at varying effector/target cell ratios, was observed for 3 of 4 enriched gingival mononuclear cell populations. Gamma-interferon (INF-gamma) preincubation of enriched gingival effector cells from 5 additional patients resulted in a 43% increase in NK cell activity. The finding of increased HNK-1+ cells with gingival inflammation suggests that these cells may play a role in tissue damage, as well as in modulation of B cell activity, in gingivae of patients with periodontal disease.

Diabetic glomerulonephropathy: histopathologic, immunofluorescent, and ultrastructural studies of 16 cases.

Sixteen cases of diabetic glomerulopathy are reported. Direct immunofluorescent and ultrastructural studies of renal biopsy tissues demonstrated that two patients had linear deposits of IgM and C'3 in the absence of IgG, four diabetic patients had sclerosis-induced entrapment of immunoglobulins and complement, and one patient had granular immune complexes in the subepithelial and intramembranous portion of the glomerular basement membrane. In one patient, who had nodular glomerular lesions, diffuse fibrillar deposits of electron-dense material were observed in the mesangium. In this mesangial infiltrate, light microscopy revealed the absence of amyloid and direct immunofluorescence revealed the absence of all immunoglobulins, complement components, and fibrinogen. Our study suggests that the morphologic alterations observed in diabetic glomerulopathy might be mediated by either immune mechanism or by abnormal biochemical or functional factors, such as impairment of the mesangial IgA clearance mechanism.

Membranous glomerulonephritis in a child asymptomatic for hepatitis B virus. Concomitant seropositivity for HBsAG and anti-HBs.

The presence of hepatitis B surface antigen (HBsAg) in association with immunoglobulins and complement components within the glomerular basement membranes of adults having chronic active hepatitis has been well documented. In addition, investigators in Poland have demonstrated HBsAg immune complexes in glomeruli of children who did not have clinical evidence of hepatitis. More recently, a single case of childhood membranous glomerulonephritis in an asymptomatic carrier of hepatitis B virus was cited by observers in Canada. Reported here is the deposition of HBsAg immune complexes in the glomerular basement membranes of a 13-year-old black boy who had membranous glomerulopathy but not clinical evidence of hepatitis. This may be the first reported case in the United States of HbsAg-associated membranous glomerulonephritis in a child asymptomatic for hepatitis B virus, and only the second such case in North America. However, unlike previous studies of childhood glomerulopathy in association with hepatitis B virus, this patient is seropositive for both HBsAg and anti-HBs (antibody for hepatitis B surface antigen). Similar "rare" serologic findings were found for the patient's eldest male sib.

Ultrastructural assessment by colloidal iron of the distribution and localization of anionic sites in human glomerulonephritides.

The distribution of anionic sites within glomerular basement membranes from patients with different glomerulonephritides was examined by the dialyzed colloidal iron (DI) staining technique. The correlation of ultrastructural findings with glomerular disease, renal function, and degree of proteinuria revealed three alterations: 1) speckled DI staining in the lamina densa of patients with decreasing renal function and a proteinuria of greater than 1 g/24 hours; 2) an apparent staining disparity and diminution of DI at the lateral borders of swollen and retracted foot processes with inclination of the foot processes in the direction of the more weakly staining lateral border; and 3) heavy DI reaction on the apical of free surfaces of fused foot processes. Human subjects with a proteinuria of more than 1 g/24 hours display optimal labeling of the endothelial fenestrae, endothelial cell coat, lamina rara interna, and lamina rara externa. The staining observed may be explained either on the basis of direct DI interaction with diffusing plasma proteins secondary to a decrease in the transglomerular charge barrier consequential to a loss of intrinsic anionic sites or on the basis of "unmasking" of anionic moieties within the glomerular basement membrane. Regardless of the mechanism involved, the present study indicates that a threshold proteinuria of 1 g/24 hours is needed to effect staining in the lamina densa.