Formation of subepithelial dense deposits in rats induced by a monoclonal antibody against the glomerular cell surface antigen

We developed a monoclonal antibody, H5H3, of IgG1 subclass by hybridization technique using spleen cells of mice immunized with plasma membrane fraction of isolated rat glomeruli. H5H3 recognized main bands at about 220 kD by immuno-overlay technique and bound to the glomerulus as well as brush border of proximal tubules by indirect immunofluorescence (IF) microscopy on normal rat kidney frozen sections. By immunoelectron microscopy (IEM) it bound to the surface of mainly glomerular epithelial cell and weakly to the endothelial cell. After injection to Wistar rats it remained granularly in the glomerulus for more than 2 weeks seen by IF. When rats were preimmunized with murine IgG 4 days before the injection of H5H3, mouse IgG, rat IgG and C3 were strongly visible granularly in the glomerulus in 14 days by IF. Numerous dense deposits were formed at subepithelial area seen by transmission electron microscopy. Perfusion experiment of H5H3 into rat left kidney showed granular distribution of mouse IgG in 48 h, indicating that the reaction occurred in situ. H5H3 bound diffusely in fine granular pattern on the surface of cultured glomerular epithelial cells (GEC) studied by IF and IEM. Antigenic redistribution occurred on GEC after incubation of H5H3 at 37 C. These results suggested the required conditions to form subepithelial immune dense deposits, namely that H5H3 after reaction with antigen could stay for long time in the glomerulus; that H5H3 became an antigen in autologous phase to induce large immune complexes; and H5H3 could induce antigenic modulation.     

Hybridomas using athymic nude mouse injected with Crohn's disease (CD) tissue filtrate. Immunoreactivity of the hybridomas with CD sera.

Injections of Crohn's disease (CD) tissue filtrates produce lymphoma and hyperplastic lymph nodes from plasma cell hyperplasia (PCH) in athymic nude (nu/nu) mice; these lymphoid tissue contain an antigen(s) recognized by CD serum/gamma G immunoglobulin (IgG). To immortalize the "CD-reactive antigen(s)," the authors fused the lymphoid cells from a CD tissue filtrate primed nu/nu mouse with nonsecretory mouse myeloma cells. Hybrids were screened and selected based on their reactivity with CD serum IgG, but not with control serum IgG in an indirect immunofluorescence assay (IF). Two CD-positive hybridomas were examined by IF with sera from 47 CD, 38 ulcerative colitis (UC), 13 controls with other gastrointestinal diseases, 19 with autoimmune diseases, and 21 normal subjects. Sera from 16 CD patients (34%) reacted with the two hybridomas, but only one of 38 UC sera and none of the 53 other disease or normal control sera reacted. The immunoreactivity of CD sera was significantly higher than UC sera (P less than 0.01) and each of the other groups (P less than 0.007). Using immunoperoxidase techniques at light and electron microscopic levels, the authors localized CD-associated antigen(s) in the plasma membrane of the two hybridomas. Further characterization of these hybridomas and the immunoreactive protein(s) may provide an important probe(s) for the diagnosis and the understanding of the pathogenesis of CD.     

Post Hoc Analysis of a Randomized Double-Blind Trial of the Correlation of Functional and Binding Antibody Responses Elicited by 13-Valent and 7-Valent Pneumococcal Conjugate Vaccines and Association with Nasopharyngeal Colonization

In a randomized double-blind trial in healthy Israeli infants in Israel who received the 13-valent or 7-valent pneumococcal conjugate vaccine (PCV13 or PCV7, respectively) at 2, 4, 6, and 12 months, PCV13 significantly reduced nasopharyngeal (NP) colonization of serotypes 1, 6A, 7F, 19A, cross-reacting 6C, and the common PCV7 serotype 19F, from ages 7 to 24 months. No differences were observed between the vaccine groups for serotype 3 or for the remaining common PCV7 serotypes. For serotype 5, too few events were observed to draw an inference. Generally consistent with these findings, PCV13 elicited significantly higher enzyme-linked immunosorbent assay (ELISA) IgG-binding antibody responses than did PCV7 for the additional PCV13 serotypes 1, 3, 5, 6A, 7F, 19A, and for the common serotype 19F, with similar or lower responses for the remaining common serotypes. To further assess immunogenicity and colonization, we conducted a post hoc analysis of PCV13 functional antibody responses measured by opsonophagocytic activity (OPA) assays in a randomly selected subset of subjects. The pattern of functional antibody OPA responses elicited by PCV13 relative to PCV7 was similar to that of the ELISA anticapsular IgG-binding antibody responses described above. In addition, the OPA responses generally correlated positively with IgG responses for all 13 serotypes among the PCV13 recipients and for all 7 common serotypes and the additional serotype 6A but not for 19A or the other serotypes unique to PCV13 among the PCV7 recipients. This post hoc analysis supports an association between serum OPA functional and IgG-binding antibody levels, allowing for a transfer of inferred associations between IgG responses and NP colonization to OPA responses.     

Localization of Fc receptors on liver sinusoidal endothelium. A histological study by electron microscopy

The localization of Fc receptors for IgG on mouse liver sinusoidal endothelium was studied in tissue sections by light and electron microscopy using peroxidase-antiperoxidase IgG complexes as ligands. Light microscopy revealed that Fc receptors were continuously present along the sinusoidal wall, but absent on the endothelium of the central vein and that of the portal area blood vessels. Electron microscopy revealed Fc receptors on both the luminal and basal aspects of the plasma membrane, far more being present on the former, on the walls of cytoplasmic fenestrae, and abundantly on the walls of coated pits and vesicles of sinusoidal endothelial cells. Fc receptors were found more frequently on sinusoidal endothelial cells than on Kupffer cells. Fat-storing cells lacked the receptors and hepatocytes showed an ambiguous result which was considered to be nonspecific.     

LOCALIZATION OF Fc RECEPTORS ON LIVER SINUSOIDAL ENDOTHELIUM A Histological Study by Electron Microscopy

The localization of Fc receptors for IgG on mouse liver sinusoidal endothelium was studied in tissue sections by light and electron microscopy using peroxidase-antiperoxidase IgG complexes as ligands. Light microscopy revealed that Fc receptors were continuously present along the sinusoidal wall, but absent on the endothelium of the central vein and that of the portal area blood vessels. Electron microscopy revealed Fc receptors on both the luminal and basal aspects of the plasma membrane, far more being present on the former, on the walls of cytoplasmic fenestrae, and abundantly on the walls of coated pits and vesicles of sinusoidal endothelial cells. Fc receptors were found more frequently on sinusoidal endothelial cells than on Kupffer cells. Fat-storing cells lacked the receptors and hepatocytes showed an ambiguous result which was considered to be nonspecific.     

The subcellular localization of immunoglobulin in mouse plasma cells,as studied with immunoferritin cytochemistry on ultrathin frozen sections

Cells of mineral oil-induced plasmacytomas (MOPC) and normal plasma cells from mouse lymph nodes were used to study the intracellular localization of IgG by means of immunoferritin cytochemistry on ultrathin frozen sections. IgG was demonstrated in the rough endoplasmic reticulum, in virus-containing smooth ER of the tumor cells, in peripheral elements at the cis-side of the Golgi complex, and in all Golgi cisternae. It is suggested that the peripheral elements transfer IgG molecules from the RER to the Golgi complex. Vacuoles showing a strong immunoreaction occurred at the trans-side of the Golgi complex. These vacuoles were normal in lymph node plasma cells and were occasionally seen in the tumor plasma cells. It is proposed that these vacuoles carry IgG from the Golgi complex to the cell membrane and, hence, can be considered as secretory vacuoles.     

Hybrids from normal, germ free, nude and neonatal mice produce monoclonal autoantibodies to eight different intracellular structures.

The supernatants of hybrids produced by fusion of splenocytes from normal, germ free, nude and neonatal BALB/c mice with mouse NS-1 myeloma cells were examined for immunofluorescence reactivity with viable and acetone fixed monolayers of syngeneic mouse fibroblasts and mouse 3T3 cells and with viable cell suspensions of syngeneic mouse erythrocytes and thymocytes. The supernatants of 70 out of 419 (16.7%) hybrids from normal mice, 87 out of 627 (13.9%) from germ free mice, 28 out of 240 (11.7%) from nude mice and 42 out of 1020 (4%) from neonatal mice reacted with 8 different intracellular structures in mouse fibroblasts and 3T3 cells, and with rat and human fibroblasts. The reactive intracellular structures comprised stress fibres, intermediate filaments, cell membrane associated, golgi complex, cytoplasm, cytoplasmic vesicles, mitochondria and nuclei. Of 45 stable clones derived by limiting dilution, 43 produced IgM antibodies and two IgG2b antibodies. Only one out of the 2,306 (0.04%) hybrids produced an autoantibody to the surface membrane of mouse thymocytes. These results show that B cells with reactivity towards different intracellular structures are present in the normal B cell repertoire whereas surface reactive B cells may be deleted or more profoundly suppressed or anergic and unable to form viable Ig producing hybridomas.     

Demonstration of Antigenic Sites in Glomeruli of Patients with Acute Poststreptococcal Glomerulonephritis by Immunofluorescein and Immunoferritin Technics

The presence and localization of antigenic sites in glomeruli of 14 patients with acute poststreptococcal glomerulonephritis (AGN) were studied by immunofluorescein and immunoferritin technics. Labeled IgG fractions from the same patients were used for the identification of antigenic sites. The staining capacity of these IgG fractions depended on the time when sera were obtained. Staining was minimal during the first week, and increased up to the fourth or fifth week. Glomeruli, however, stained only when renal tissue was obtained during the early phase of the disease. Precise localization of antigenic sites was determined with ferritin-conjugated patients' IgG. Segmental deposition of ferritin was observed in the mesangial matrix and on the endothelial side of the glomerular basement membrane. Subepithelial electron-dense deposits contained no or very few ferritin particles. In contrast, ferritin-conjugated antihuman IgG was distributed diffusely in the mesangial matrix, on the endothelial side of the basement membrane and in subepithelial deposits. These findings suggest that, during the early stage of acute poststreptococcal glomerulonephritis, free antigen is present in the glomeruli of patients with this disease.     

Shedding of the guinea pig peritoneal macrophage Fc gamma receptor. Effect on apparent association constant and on the number of IgG-binding sites on the cells

During our studies on the structure and properties of guinea pig peritoneal macrophage Fc gamma receptor we observed that these cells spontaneously release an IgG-binding material into supernatant. Although the shedding was accompanied by a decrease in IgG-binding ability of macrophages, the number of IgG-binding sites/cell before and after shedding was similar. However, macrophages after the shedding interacted with IgG with a lower apparent association constant, Ka. Therefore, we assume that the decrease of IgG-binding ability of cells was an effect of decrease in value of Ka. Experiments with protein synthesis inhibitors showed that the shed receptor is replaced by "de novo" synthesized receptor molecules.     

Antibody recognition of ragweed allergen Ra3: localization of the full profile of the continuous antigenic sites by synthetic overlapping peptides representing the entire protein chain

A comprehensive synthetic approach, for the localization of the full profile of the continuous antigenic sites on proteins, previously introduced by this laboratory, was applied here to localize the continuous antigenic sites of ragweed allergen, Ra3. The following 10 consecutive peptides, each comprising 15 residues (except for peptide 91-101) and overlapping each of its neighbors by 5 residues, were synthesized and purified: 1-15, 11-25, 21-35, 31-45, 41-55, 51-65, 61-75, 71-85, 81-95 and 91-101. Quantitative radiometric titrations of protein and peptide adsorbents were performed with 125I-labeled anti-Ra3 IgG antibodies from rabbit, outbred mouse and human antisera. The specificity of antibody binding to peptide adsorbents was confirmed by inhibition experiments. These studies established the full profile of antigenic (IgG-binding) sites of Ra3 and permitted comparison with the allergenic (IgE-binding) sites recently localized. It was found that the recognition by IgG antibodies was independent of the host species in which the antibodies were raised. Furthermore, the regions recognized by human IgE antibodies coincided with those recognized by IgG antibodies in three different species. Thus, Ra3 was found to have 4 continuous antigenic sites which occupy the same locations as the allergenic sites.     

Survival and tissue localization of xenogeneic human lymphocytes in hypothymic nude mice: observations in normal subjects and multiple sclerosis

The feasibility of using the hypothymic nude mouse as a transfer system for the study of cellular autoimmunity, by injection and tracing of 51chromium (Cr) or fluorescein labelled lymphocytes from humans, allogeneic or syngeneic mice was investigated. The localization of the injected cells was studied by radioisotopic counting or microscopy. For mouse cells there was better survival of syngeneic than allogeneic cells, and the pattern of distribution of injected mouse lymphocytes differed markedly from that of xenogeneic human lymphocytes. Thus, the lymph nodes and spleen were the predominant sites of localization of transferred syngeneic (BALB/c) lymphocytes, indicative of long recirculation of these cells, whereas the liver was the predominant site of localization of transferred human lymphocytes, and there was no evidence for recirculation of these cells. In studies on patients with multiple sclerosis, lymphocytes did not localize in neural tissue. The data indicate that the nude mouse is unlikely to be useful for the study of the tissue distribution of dispersed xenogeneic (human) lymphocytes.     

Binding of native alpha 2-macroglobulin to human group G streptococci.

Binding of human alpha 2-macroglobulin (alpha 2M) to group G streptococci and to their immunoglobulin G (IgG)-binding proteins (protein G) was investigated. Native alpha 2M bound specifically to strain G-148 with an apparent dissociation constant of (2.2 +/- 1.5) x 10(-9) M. Proteinase-complexed alpha 2M did not compete for the binding sites, and 125I-labelled proteinase-complexed alpha 2M did not bind to the bacteria. Binding of native alpha 2M to the cells was not affected by IgG or protein G consisting of only IgG-binding domains. 125I-labelled recombinant protein G did not bind to native or proteinase-complexed alpha 2M. However, a lysate of G-148 cells inhibited binding of alpha 2M to the bacteria, and immobilized wild-type protein G bound alpha 2M directly from fresh human plasma. In 13 group G streptococcal isolates, IgG-binding proteins were immunologically identified as protein G. In 11 isolates, these molecules reacted also with alpha 2M and human serum albumin (HSA). Western blots (immunoblots) of two wild-type protein G variants revealed identical bands reactive with goat IgG, HSA, and native alpha 2M. Digestion of wild-type protein G with clostripain destroyed in both variants the binding sites for alpha 2M but not for albumin and IgG. N-terminal fragments of protein G (lacking the IgG-binding region) bound both alpha 2M and HSA, whereas a similar HSA-binding peptide lacking the first 80 amino acids did not react with alpha 2M. Our findings are consistent with a specific binding site for native alpha 2M in the N-terminal region of protein G and suggest that binding of alpha 2M via IgG-binding proteins may be a general feature of human group G streptococci.     

The roles of T cells in experimental crescentic glomerulonephritis]

A crescentic glomerulonephritis (GN) model was induced by intravenous injection of rabbit anti-mouse glomerular basement membrane (GBM) antiserum and lipopolysaccharide (LPS) in BALB/c mice, heterozygous mice and nude mice respectively, in order to detect the T-cells effects on the development of crescentic GN. The immunofluorescence and morphological changes of glomeruli in different groups of animals were compared. Intense fluorescence (4+) of rabbit IgG could be found along the GBM in liner pattern in all the animals. Intense (3(+)-4+) mouse IgG was also found along the GBM in liner pattern in the normal and heterozygous mice, but could not be identified in the nude mice. The normal mice developed typical crescentic GN, characterized by serious degeneration and destruction of GBM, fibrin deposition and crescents formation, 3-6 weeks after the injection. The heterozygous mice only developed mild proliferation of the mesangial cells in the glomeruli but there was no glomerular lesion detected in the nude mice. It suggests that the glomerular immune damage requires the participation of functional T-cell.     

A new site-directed transgenic rheumatoid factor mouse model demonstrates extrafollicular class switch and plasmablast formation

The AM14 rheumatoid factor (RF) transgenic (Tg) mouse has been valuable for studying how self-reactive B cells are regulated beyond central tolerance, because they remain ignorant in normal mice. AM14 B-cell activation can be studied on autoimmune-prone strains or by inducing activation with IgG2a anti-chromatin antibodies (Abs). Despite the utility of conventional Ig-Tg mice, site-directed Ig-Tg (sd-Tg) mice provide a more physiological model for B-cell responses, allowing class switch and somatic hypermutation. We report here the creation of an AM14 sd-Tg mouse and describe its phenotype on both normal and autoimmune-prone backgrounds. AM14 sd-Tg B cells develop normally but remain unactivated in the BALB/c background, even after significant aging. In contrast, in the autoimmune-prone strain MRL/lpr, AM14 sd-Tg B cells become activated and secrete large amounts of IgG RF Ab into the serum. Class-switched Ab-forming cells were found in the spleen and bone marrow. IgG RF plasmablasts were also observed in extrafollicular clusters in the spleens of aged AM14 sd-Tg MRL/lpr mice. Class switch and Ab secretion were observed additionally in AM14 sd-Tg BALB/c B cells activated in vivo using IgG2a anti-chromatin Abs. Development of IgG auto-Abs is a hallmark of severe autoimmunity and is related to pathogenesis. Using the AM14 sd-Tg, we now show that switched auto-Ab-forming cells develop robustly outside germinal centers, further confirming the extrafollicular expression of activation induced cytidine deaminase (AID). This model will allow more physiological studies of B-cell biology in the future, including memory responses marked by class switch.     

Isolation of immunoglobulin G by affinity chromatography using an IgG Fc receptor protein from Streptococcus dysgalactiae coupled to a solid phase

A culture of Streptococcus dysgalactiae (C 26) was shown to bind only to 125I-IgG, whereas another S. dysgalactiae culture (C 12) bound both 125I-IgG and 125I-albumin. The IgG-binding proteins could be readily solubilized by lysozyme treatment of the bacteria and isolated by affinity chromatography on IgG Sepharose. The purified IgG-binding protein from S. dysgalactiae C 26, which lacked simultaneous albumin binding activity, precipitated with IgG preparations from man, cow, horse, pig and mouse but not with chicken IgG. This IgG-binding protein was coupled to CNBr-activated Sepharose and subsequently used for the purification of IgG from both bovine and human serum. SDS-PAGE and immunoelectrophoretic studies confirmed the purity of the eluted proteins.     

Binding of immunoglobulin G and albumin to Streptococcus dysgalactiae

All 24 cultures of Streptococcus dysgalactiae investigated bound 125I-IgG, 13 cultures additionally interacted with 125I-albumin. Inhibition experiments with unlabelled IgG and albumin preparations from humans and various animal species indicated the specificity of the binding sites which showed characteristics of IgG Fc-receptors of type III and albumin-receptors of type c. IgG and albumin-binding proteins could be removed from the streptococcal surface by solubilization and subsequently isolated by affinity chromatography. The isolated binding proteins of S. dysgalactiae strains C 12 and C 8 obtained from IgG and albumin sepharose precipitated with IgG in immunodiffusion reactions, and in immunoelectrophoretic studies, and they reacted, after transfer onto nitrocellulose, with 125I-IgG or 125I-albumin and vice versa. Antisera produced against IgG-binding proteins of S. dysgalactiae C 12 inhibited binding of 125I-IgG and 125I-albumin. Solubilization of binding proteins by trypsinization yielded low molecular weight fragments with 125I-IgG but not with 125I-albumin binding activities. IgG-binding proteins isolated from S. dysgalactiae C 26 reacted with 125I-IgG but not with 125I-albumin, indicating the presence of 2 groups of type III Fc-receptors among S. dysgalactiae strains.     

Ultrastructural studies on the localization of IgG in the aortic endothelium and subendothelial intima of atherosclerotic and nonatherosclerotic rabbits

Immunoelectron microscopy with peroxidase-conjugated Fab fragments of anti-IgG was used for studying the localization of IgG in the aortic endothelium and subendothelial intima of atherosclerotic and nonatherosclerotic rabbits. Small amounts of IgG were found in the cell coat, in caveolae and vesicles, and also in intercellular clefts of endothelial cells from normocholesterolemic rabbits. Injured endothelial cells exhibited prominent accumulations of IgG in the cytoplasmic matrix, possibly due to leakage through plasma membrane defects. In atherosclerotic lesions from hypercholesterolemic rabbits, there was a striking increase in the amount of IgG-reactive material in the cell coat and vesicles of intact endothelial cells. Also in these animals, injured endothelial cells were characterized by a cytoplasmic IgG accumulation. There were prominent IgG depositions in the subendothelial zone of the lesions. IgG was adhering to collagen fibers, and also coating the surfaces of subendothelial foam cells. The pathophysiological significance of an interaction between such intimal IgG and phagocytes is discussed.     

Natural mouse IgG reacts with self antigens including molecules involved in the immune response.

IgG isolated on protein A-Sepharose from pools of normal sera from various mouse strains were examined by immunoblotting for reaction with self antigens. Homogenates of the major mouse organs, i.e. brain, skin, spleen, kidney, adrenals, thymus, heart, muscle and liver were used as the source of autoantigens. IgG stained at least 220 bands on the immunoblots. The antigens corresponding to these bands were tentatively identified by molecular mass estimation and referenced to computerized mouse protein data banks. IgG mainly recognized enzymes but it also stained intracellular structural constituents and surface molecules implicated in the functioning of the immune system. The validity of this identification was confirmed by analyzing purified antigens from mouse or other animal species by immunoblotting and enzyme immunoassays. Furthermore, extracts of 125I-surface-labeled cells were immunoprecipitated with IgG in the liquid phase or immobilized on beads. The proteins precipitated migrated to the same positions as those precipitated by specific monoclonal antibodies (mAb), such as class I alpha chain and beta 2-microglobulin, class II alpha and beta chains, CD3, CD4 and CD8 antigens. The results obtained with several enzyme immunoassay procedures using cell membrane extracts, specific mAb and normal IgG further supported the specific interaction of IgG with Ia, CD4 and CD8 molecules. Affinity chromatography indicated that at least 20% of normal mouse IgG possess polyreactive autoantibody function. Dissociation constants of these IgG were calculated for some autoantigens and found to be in the range of 2 x 10(-6)-7 x 10(-6) M. It is concluded that normal mouse IgG exhibit autoreactivities similar to those previously described for IgM.     

Natural serum antibodies which bind to damaged erythroblasts: isotypes and light-chain composition

Dimethyl-sulphoxide-induced Friend leukaemia erythroblasts (IFLE) which had been damaged by treatment with inhibitors of protein synthesis (cycloheximide, puromycin) were incubated with normal mouse serum or with doubling dilutions of it. The erythroblasts were subsequently tested for their binding of natural antibodies of all the major immunoglobulin (Ig) isotypes and both light-chain types, using Fc- and light-chain-specific FITC-immunoconjugates, and flow cytometry. After short-term (4-h) exposure of IFLE to puromycin, some binding of IgG but not of IgM or IgA could be demonstrated. By contrast, prolonged (17-h) exposure of IFLE to cycloheximide or puromycin resulted in their reaction with antibodies of all major isotypes (IgM, IgG1, IgG2a, IgG2b, IgG3, IgA) containing kappa-light chains. Under these conditions of damage the percentage of IgM- or IgG-binding IFLE significantly exceeded that of IgA-reactive cells. Moreover, the percentage of damaged IFLE which bound IgG2a or IgG3 was greater than that of cells which bound IgG1 or IgG2b. The percentage of damaged IFLE which bound a certain Ig-isotype did not correlate with the concentration of that Ig-isotype in normal mouse serum. The results suggest that natural antibodies of all major isotypes containing kappa-light chains bind specifically to severely damaged IFLE and could facilitate their interaction with macrophages in a concerted manner.     

Murine soluble Fcγ receptors/IgG-binding factors (IgG-BF): Analysis of the relation to FcγRII and production of milligram quantities of biologically active recombinant IgG-BF

The production of soluble forms of low-affinity Fc gamma R by cells expressing recombinant or natural membrane Fc gamma RII, and the structural relationships between these soluble receptors and membrane Fc gamma RII are described. We show that 37-40 kD soluble Fc gamma RII, corresponding to the two N-terminal domains of Fc gamma RII and binding to IgG, are spontaneously produced in vitro by cleavage of membrane Fc gamma RII. Moreover, we describe methods to produce and purify to homogeneity large quantities of endotoxin-free recombinant IgG-binding factor (rIgG-BF) from the culture medium of a cell line transfected with a mutated Fc gamma RII cDNA. These methods include the use of bioreactors for culturing transfected fibroblasts and the purification of rIgG-BF by ion-exchange chromatography and hydrophobic-interaction chromatography. By using such procedures, about 2.4 mg of rIgG-BF were purified from 1 liter of culture medium of transfected fibroblasts. Like natural IgG-BF, the 95-99% pure rIgG-BF suppressed, in a dose-dependent manner, secondary in vitro IgG antibody responses to sheep red blood cells.