Immunoglobulin G Binding Research Articles

Molecular Weights and Isoelectric Points of Sperm Antigens Relevant to Autoimmune Infertility in Men

Purpose We determined the molecular weights and isoelectric points of antigens in the uncapacitated and capacitated spermatozoa of fertile men binding to the serum immunoglobulin G (IgG) from 8 autoimmune infertile men and 8 fertile nonautoimmune men. Materials and Methods We used double fluorochrome cytotoxicity and immunobead binding assays to determine the sperm antibody status of the study subjects. 2-Dimensional gel electrophoresis and Western blot analysis were used to determine the molecular weights and isoelectric points of sperm antigens binding to serum IgG from these men. Amino acid sequencing of the digested peptides of chosen proteins was accomplished. Immune reactivity to the proteins in autoimmune infertile men was further verified. Results Serum IgG from fertile men failed to react significantly. Serum IgG from all autoimmune men (100 percent) showed significant binding to proteins with a molecular weight of 92 kDa. and isoelectric points of 3.5 to 4.0 in the capacitated spermatozoa. Six of 8 infertile men (75 percent) had serum IgG binding to capacitated sperm antigens with a molecular weight of 18 kDa. and isoelectric points of 4.5 to 5.2. Amino acid sequencing of peptides of the 92 kDa. protein matched complement component 1 (C1) inhibitor, with noted differences in the amino acid sequencing from the latter. The 18 kDa. protein matched calmodulin. We verified that serum IgG from autoimmune infertile men bound with C1 inhibitor and ascertained that the 92 kDa. protein in the spermatozoa was C1 inhibitor-like protein. Conclusions Significant antibody responses to C1 inhibitor-like protein and calmodulin were noted in autoimmune men. Both of these proteins may be of testicular origin and these autoimmune responses may be highly relevant to infertility.

Silicate antibodies in women with silicone breast implants: development of an assay for detection of humoral immunity

Silicon, in the form of sodium silicate (Na2SiO3), adsorbed onto bovine serum albumin (BSA)-precoated plates served as the solid-phase antigen in an enzyme immunoassay to detect silicate-reactive antibodies in the plasma of 40 symptomatic women with silicone breast implants, 91 asymptomatic women with silicone breast implants, 50 healthy control women, and 52 women with rheumatic diseases and without silicone breast implants, Silicate-reactive antibodies of immunoglobulin G (IgG) or IgM isotypes were detected in the plasma of 30% (12 of 40) of the symptomatic women with silicone breast implants; 9% (8 of 91) of the asymptomatic women with silicone breast implants; 5% (1 of 20) of the women without implants who had systemic lupus erythematosus; and 0% (0 of 32) of the women without implants who had either Sjögren syndrome, scleroderma, or rheumatoid arthritis. Only 2% (1 of 50) of the sera from the healthy control women contained silicate-reactive antibodies. Preincubation of sera with silicate and eight other metal compounds (including SiO2) demonstrated that the IgG and IgM antibodies bound specifically to silicate, because preincubation with Na2SiO3 inhibited more than 90% of the activity, whereas CrO3, Li2SO4, MgSO4, NiSO4, HgCl2, ZrOCl2, BeSO4, and SiO2 failed to inhibit the IgG or IgM antibody binding to the silicate-BSA plates. Furthermore, the F(ab')2 portion and not the Fc portion of the silicate-reactive IgG was reactive with BSA-bound silicate in the enzyme immunoassay. The assay for silicate-reactive antibodies was quantified by assigning arbitrary units to a standard curve composed of serial twofold dilutions of high-positive (ten times higher than the cutoff) silicate antibody sera. This novel assay is a useful method for detecting and quantifying humoral immune response to silicate.

Binding properties of Streptococcus suis for immunoglobulin G and other plasma proteins.

Immunoglobulin G (IgG) binding proteins on the surface of Streptococcus suis could be readily detected by direct cultivation of the bacteria on nitrocellulose membranes and subsequent treatment of the membranes with human IgG. Among the 75 S. suis isolates tested two cultures (S. suis P43, S. suis P143) caused a blue colouration of the membranes indicating IgG binding activities. The IgG binding proteins could be solubilized by heat treatment of the bacteria at an acid pH and also by mutanolysin treatment. Western blot analysis revealed numerous protein bands with IgG binding activities. The IgG binding proteins were also released into the culture supernatant of the bacteria. This could be detected for 51 of the 75 S. suis using a microfiltration assay. In binding studies with 125I-IgG S. suis P43 and S. suis P143 but none of the other S. suis isolates showed a significant binding of the protein. These two cultures additionally bound 125I-albumin, 125I-alpha 2-macroglobulin and 125I-fibrinogen all from humans but not 125I-chicken IgG or 125I-human haptoglobin 2-1. The binding profiles of the two S. suis cultures tested indicate a close relation of these binding proteins with streptococcal protein G.

Interaction of human immunoglobulin G with l-histidine immobilized onto poly(ethylene vinyl alcohol) hollow-fiber membranes

l-Histidine as pseudobiospecific ligand was immobilized onto poly(ethylene vinyl alcohol) hollow-fiber membranes to obtain an affinity support for immunoglobulin G (IgG) purification. The interaction of human IgG with the affinity membranes was studied by chromatography and equilibrium binding analysis. Adsorption was possible over a broad pH range and was found to depend strongly on the nature of the buffer ions rather than on ionic strength. With zwitterionic buffers like morpholinopropanesulfonic acid (Mops) and hydroxyethylpiperazineethanesulfonic acid (Hepes), much higher adsorption capacities were obtained than with other buffers like Tris-HCl and phosphate buffers. An inhibition analysis revealed that non-zwitterionic buffers competitively inhibit IgG binding, whereas Mops and Hepes in their zwitterionic form do not. By choosing the appropriate buffer system, it was possible to adsorb specifically different IgG subsets. The IgG molecules were found to adsorb on membrane immobilized histidine via their F ab part. Determination of dissociation constants at different temperatures allowed calculation of thermodynamic adsorption parameters. Decrease in K D with increasing temperature and a positive entropy value between 20 and 35°C (in Mops buffer) indicated that adsorption is partially governed by hydrophobic forces in that temperature range, whereas at lower temperatures, electrostatic forces are more important for adsorption.

Lung cancer-related changing profile of B-cell epitopes recognized on mucosal antigens. The Der p1 model

The natural immunoglobulin G (IgG) response towards antigens presented at the mucosal level in patients with lung cancer, the most frequent mucosal malignancy in humans was studied. Compared with healthy control subjects, patients with lung cancer display lower IgG antibody titers toward antigen p1 of the house dust mite, Dermatophagoides pteronyssinus (Der p1), chronically presented to the respiratory mucosa. The present study further characterizes this defect in terms of antibody specificity. Antibody specificity was studied by comparing the IgG binding to native Der p1 (nDer p1) and its products of pepsin hydrolysis (dDer p1) in a solid phase enzyme-linked immunosorbent assay (ELISA) in 148 patients with lung carcinoma, 148 healthy control subjects, and 50 patients with chronic bronchitis. Antibody specificity was also evaluated before and after (5 +/- 1 weeks) surgical excision of the tumor in competition ELISA using streptavidin-biotin technology. Lung cancer sera had a higher degree of binding to dDer p1 compared with nDer p1, whereas control group sera bound similarly to the two forms of the antigen. Lung cancer sera and control group sera showed distinct capacities to prevent the binding of pooled IgG from each group to nDer p1. The inhibition capacity displayed by cancer sera is changed 5 weeks after cancer removal. These results, somewhat similar to those observed for bovine betalactoglobulin, document the recognition of a different set of epitopes on Der p1 and more generally on mucosal antigens by lung cancer IgG compared with the IgG from control patients. This distinct profile of epitope specificity changes partially soon after cancer removal, suggesting a tumor-dependent disturbance and opens the way to a new class of markers in lung cancer. Furthermore, this study documents a surprising but striking influence of the clinical status on the choice of B-cell immunodominant epitopes in mucosal responses.

Species-specific long range interactions between receptor/ligand pairs

Total internal reflection microscopy (TIRM) monitors Brownian fluctuations in elevation as small as 1 nm by measuring the scattering of a single sphere illuminated by an evanescent wave when the sphere is levitated by colloidal forces such as electrostatic double-layer repulsion. From the Boltzmann distribution of elevations sampled by the sphere over time, the potential energy profile can be determined with a resolution of approximately 0.1 of the thermal energy kT. Thus, the interaction between a receptor-coated (goat, horse, or rabbit immunoglobulin G (IgG)) latex sphere and a protein A (SpA)-coated glass microscope slide was studied. A typical TIRM potential energy profile measured between a bare sphere and a bare glass plate, where the sphere fluctuates around the secondary potential energy minimum formed between double-layer repulsion and gravitational attraction, agrees well with DLVO theory. The interactions measured between IgG-coated spheres and SpA-coated slides, on the other hand, displayed a weaker repulsion compared with that observed between bare surfaces under the same conditions. Analysis of the results obtained between the coated surfaces suggests an additional attractive force. The decay length of this attraction correlates with the known dissociation constants for the binding of IgG with SpA in free solution.

Endothelial Cell Damage in Sarcoidosis and Neurosarcoidosis: Autoantibodies to Endothelial Cells

Damage to the vascular system, of which endothelial cells are the main constituent, may occur in sarcoidosis. Evidence of blood-brain barrier (BBB) disruption is shown in sarcoidosis with central nervous system (CNS) involvement by means of magnetic resonance imaging or computed tomography. We investigated the presence of antiendothelial cell antibodies by culture of endothelial cells derived from human umbilical cord vein and from human brain using an enzyme-linked immunosorbent assay, and the presence of circulating immune complexes using a Raji cell assay, in the sera of patients with sarcoidosis. The patients with sarcoidosis displayed significantly high levels of immunoglobulin G (IgG) binding to endothelial cells after blocking Fc receptor compared to controls (p < 0.01). The sera of neurosarcoid patients with CNS involvement still showed significantly increased levels of IgG binding to brain endothelial cells after blocking Fc receptor compared to controls and those of pulmonary sarcoidosis (p < 0.001 and p < 0.01 respectively). These results suggest that the presence of autoantibodies to endothelial cells may be involved in endothelial cell damage, including BBB disruption.

Identification and expression of a murine cytomegalovirus early gene coding for an Fc receptor.

Several herpesviruses, including cytomegalovirus, induce receptors for the Fc domain of murine immunoglobulin G (IgG) molecules. Viral genes coding for these receptors have been characterized only for alphaherpesviruses. In this report, we describe a new approach that led to the identification of an Fc receptor (FcR) of murine cytomegalovirus (MCMV). The Fc fragment of IgG precipitated glycoproteins (gp) of 86 to 88 and 105 kDa from MCMV-infected cells. Deglycosylation by endoglycosidase F resulted in a protein with a molecular mass of 64 kDa. Injection of complete MCMV DNA or of DNA fragments, and the subsequent testing of cytoplasmic binding of IgG by immunofluorescence microscopy, was used to search for the coding region in the MCMV genome. The gene was located in the HindIII J fragment, map units 0.838 to 0.846, where an open reading frame of 1,707 nucleotides predicts a gp of 569 amino acids with a calculated molecular mass of 65 kDa. The sequence of this gp is related to those of the gE proteins of herpes simplex virus type 1 and varicella-zoster virus. The defined length of the mRNA, 1,838 nucleotides, was in agreement with that of a 1.9-kb RNA expressed throughout the replication cycle, starting at the early stages of infection. Expression of the gene fcr1 by recombinant vaccinia virus resulted in the synthesis of gp86/88 and gp105, each with FcR properties, and the correct identification of the gene encoding the FcR was confirmed by the DNA injection method.

The inability of placental alkaline phosphatase antibodies to block immunoglobulin-G binding to cultured term trophoblast cells and to microvillous membranes

The inability of placental alkaline phosphatase antibodies to block immunoglobulin-G binding to cultured term trophoblast cells and to microvillous membranes

Investigation of the interaction between the class I MHC-related Fc receptor and its immunoglobulin G ligand

The neonatal Fc receptor (FcRn) is structurally similar to class 1 major histocompatibility molecules. FcRn transports maternal immunoglobulin G (IgG) from ingested milk into the blood. IgG is bound at the pH of milk (pH 6.0–6.5) in the gut and released at the pH of blood (pH 7.5). We find that alteration of a histidine pair within the α3 domain of FcRn and of a nearby loop (the FcRn counterpart of the class I CD8-binding loop) affects the affinity for IgG. Inhibition studies suggest the involvement of the FcRn B 2-microglobulin domain in IgG binding. Fragment B of protein A inhibits FcRn binding to IgG, localizing the binding site on Fc for FcRn to the C H2-C H 3 domain interface. Three histidines present at the C H2-C H3 domain interface of Fc could be partially responsible for the pH-dependent interaction between FcRn and IgG.

A different profile of epitopic dominance in the immunoglobulin G response to bovine betalactoglobulin in lung cancer

The authors previously documented a quantitative defect in the immunoglobulin G (IgG) response toward bovine betalactoglobulin (BLG), the major cow's milk antigen, and antigen p1 of the house dust mite, Dermatophagoides pteronyssinus (Der p1), in patients with lung cancer. In the Der p1 model, the authors documented at the IgG level an epitope specificity that differed between patients with lung cancer (preferential specificity for cryptic epitopes) and healthy control subjects and patients with mite allergy. The current study investigated whether this varying specificity might be extended to the IgG response toward BLG. The authors compared the IgG binding to native BLG (nBLG) and its products of pepsin hydrolysis (dBLG) in a solid-phase enzyme-linked immunosorbent assay (ELISA) using peroxidase-conjugated protein A in 120 patients with lung cancer, 52 patients with chronic obstructive pulmonary disease (COPD) who were closely matched for age, sex, and smoking habits with the patients with cancer, and 120 healthy control subjects (blood donors). Expressing the ratio between optical densities observed for dBLG and nBLG, respectively, the authors documented two groups: patient with lung cancer with higher levels of binding on dBLG (mean ratio +/- SD, 1.66 +/- 0.26) and healthy control subjects and patients with COPD with similar levels of retention for dBLG and nBLG (mean ratios +/- SD, 1.00 +/- 0.10 and 1.01 +/- 0.07, respectively). Influence of population characteristics could be excluded. The histologic type of cancer and its extent had no influence on the defined ratio. These results suggest a preferential recognition of epitopes unmasked by pepsin hydrolysis (cryptic epitopes?) by lung cancer IgG, contrasting with the preferential specificity of IgG from healthy control subjects and patients with COPD for structural epitopes unaffected by the proteolysis. These findings are similar to those observed previously with Der p1 and indicate a varying, and possibly specific, profile of epitopic dominance in the IgG response to antigens naturally presented at the mucosal level in patients with lung carcinoma, a model of mucosal cancer.

Detection of immunoglobulin-G-binding proteins in Streptococcus suis.

This study was undertaken to search for the presence of immunoglobulin G (IgG)-binding proteins in Streptococcus suis, an important swine pathogen. Whole bacterial cells were incubated with human or pig IgG conjugated to gold particles and examined by transmission electron microscopy. Cells of some S. suis strains were labelled as were cells of the positive control strain, Staphylococcus aureus Cowan I. Binding of pig and human IgG to five different bacterial species of group D streptococci, to reference strains representing the 29 capsular types of S. suis, and to 12 S. suis capsular type 2 strains was then examined using Western blotting. All strains interacted with pig and human IgG, although the binding profiles were slightly different. A 52 kDa protein was observed in all capsular types of S. suis. This protein, absent in other group D streptococcal species, was observed in all capsular type 2 isolates originating from diseased or clinically healthy pigs, and was shown to bind human IgG-Fc fragments. The IgG-binding activity was also observed in the culture supernatant and was sensitive to proteolysis.

Vitiligo autoantibodies are effective against melanoma

Vitiligo is a dermatologic disease characterized by local, dispersed, or diffuse white patches on the skin. The disease is defined as an autoimmune disorder because autoantibodies against membranal components of melanocytes are found in the patients' sera. The current study examined whether the autoantibodies reacting with the normal melanocytes could be a potent therapy against melanoma cells. The three in vitro assays used to determine the antibody reactivities using a mouse melanoma cell line B-16-F10 and M-14 human melanoma cells as targets are as follows: enzyme-linked immunosorbent assay (ELISA), proliferation assay, and morphologic examination in the presence of antibodies purified from sera of patients with vitiligo. In the in vivo studies, experimental melanoma was intravenously induced in C57BL/6J mice, and the mice were treated by daily intraperitoneal injections with purified immunoglobulin G (IgG) fraction derived either from patients with vitiligo or from healthy subjects. The binding of IgG derived from patients with vitiligo was demonstrated by ELISA: Exposure of melanoma cells to the vitiligo autoantibodies was followed by inhibition of their proliferation capacity. In addition, morphologic alterations exemplified by detachment of the cells from their solid support associated with melanin release were observed in the B-16-F10 cells. Less metastatic foci developed in the lungs of the mice treated with the purified IgG fraction from the sera of patients with vitiligo compared with those treated with purified IgG fraction from healthy subjects. The results of this study point to the presence of anti-melanoma autoantibodies in the sera of patients with localized and diffuse vitiligo. These antibodies have a destructive effect on melanoma cells in vitro and in vivo.

The class I major histocompatibility complex related Fc receptor shows pH-dependent stability differences correlating with immunoglobulin binding and release.

Maternal immunoglobulin G (IgG) in milk is transported to the bloodstream of newborn rodents via an Fc receptor (FcRn) expressed in the gut. The receptor shows a striking structural similarity to class I major histocompatibility complex (MHC) molecules, being composed of a related heavy chain and the identical light chain (beta 2-microglobulin). FcRn binds IgG at the pH of milk in the proximal intestine (pH 6.0-6.5) and releases it at the pH of blood (pH approximately 7.5). We have compared the stability of a soluble form of FcRn in these two pH ranges and find that the heterodimer is markedly more stable at the permissive pH for IgG binding. Using the rate of beta 2m exchange as a correlate of heterodimer stability, we find that exchange is more than 10 times slower at pH 6.1 compared to pH 7.8. Thermal denaturation profiles of FcRn heterodimers at pH 8.0 indicate a two-step, sequential heavy-chain (Tm = 52 degrees C) and beta 2m (Tm = 67 degrees C) denaturation. By contrast, at pH 6.0, a single transition is observed, centered at 62 degrees C, corresponding to denaturation of both chains. The striking difference in stability does not appear to be correlated with the binding of peptide as in class I MHC molecules, because analysis of purified FcRn by acid dissociation and sequencing suggests that FcRn is not associated with cellular peptides. These results are indicative of pH-dependent conformational changes in the FcRn heterodimer, which may be related to its physiological function.

Antigens in Capacitated Spermatozoa Eliciting Autoimmune Resposes

We studied the possible qualitative antigenic differences between native and capacitated spermatozoa pertaining to immunological infertility. Western blot analysis was used to test serum and seminal plasma immunoglobulin G (IgG) from 18 infertile couples with sperm antibodies, and 14 fertile men and 9 fertile women without significant sperm antibodies against native and capacitated spermatozoa from fertile and infertile men. More fertile and infertile subjects had serum and seminal plasma IgG binding to antigens with molecular weights of 34/36, 46, 68, 97, 105, 115, 120, 130, 150 and 190kDa. in capacitated versus native spermatozoa of all men. Of the other hand, immune reactivity to antigens with molecular weight 22/24, 30, 32, 50, 80, 88/92 and 180kDa. found in native autoimmune spermatozoa was strongest in infertile couples with sperm antibodies. This reactivity significantly increased against capacitated autoimmune spermatozoa. Native spermatozoa from few fertile men had these antigens but they appeared after capacitation. It seems that certain antigens normally appearing or enhanced after capacitation are already present in native spermatozoa from infertile men, probably due to an inherent aberration or premature capacitation. This might account for the observed enhanced immune responses in infertile couples to sperm antigens from infertile husbands.

Autoantibodies to the GLUT-2 glucose transporter of beta cells in insulin-dependent diabetes mellitus of recent onset.

Purified immunoglobulin G (IgG) from the serum of patients with insulin-dependent diabetes mellitus (IDDM) of recent onset inhibits high-Km uptake of 3-O-methyl-beta-D-glucose by rat pancreatic islets. To determine if the inhibition is the result of antibodies against GLUT-2, the high-Km glucose transporter of beta cells, we incubated IDDM sera with rat islet cells and with AtT-20ins cells transfected to express GLUT-2. IDDM sera inhibited glucose uptake in islet cells and in GLUT-2-expressing AtT-20ins cells but not in AtT-20ins cells transfected to express the low-Km isoform, GLUT-1. In 24 of 30 (77%) patients with newly diagnosed IDDM, IgG binding as measured by immunofluorescence and flow cytometry of the cells transfected to express GLUT-2 was > 2 standard deviations from the mean of the nondiabetic population; 29 of 31 (96%) of nondiabetic children were negative (P < 0.0001). Increased IgG binding could be removed by absorption with GLUT-2-expressing cells but not with GLUT-1-expressing cells. We conclude that most patients with IDDM of recent onset have autoantibodies to GLUT-2.

Quantitation of granulocyte antibodies in sera and determination of their binding sites.

An enzyme immunoassay using eluates was developed for the quantitation of granulocyte antibodies in sera. Incubation of donor granulocytes with sera containing granulocyte alloantibodies (NA1, NA2, NB1, 5b, HLA) or autoantibodies resulted in a significantly higher amount of immunoglobulin G (IgG) per cell than did incubation with control sera. Ultracentrifugation of the sera prior to testing reduced unspecific binding of IgG to granulocytes. In eight of 10 assessed sera from patients with Felty's syndrome eluted IgG decreased from elevated to normal levels. Ultracentrifugation further allowed determination of the binding sites of granulocyte-reactive alloantibodies. Using sera containing NA1- and NA2-specific alloantibodies 173,000-188,000 binding sites on homozygous and 84,000-110,000 on heterozygous cells were determined. A similar number was found using a Fc gamma receptor III (FcRIII)-specific monoclonal antibody. Granulocytes from patients with paroxysmal nocturnal haemoglobinuria showed greatly reduced binding sites for NA-specific alloantibodies. The binding sites for the human alloantibody NB1 ranged from 36,000 to 318,000 and for the 5b alloantibody from 44,000 to 65,000. Mean values of 139,000 binding sites for HLA antigens and 27,000 binding sites for the FcRII were determined using monoclonal antibodies.

Avid AL, a synthetic ligand affinity gel mimicking immobilized bacterial antibody receptor for purification of immunoglobulin G

Avid AL is an affinity gel designed for the purification of immunoglobulin G (IgG). The gel was prepared by first reacting Sepharose with 3,5-dichloro-2,4,6-trifluoropyridine and 4-dimethylaminopyridine and then with 2-mercaptoethanol. The IgG purified by Avid AL is about 95% pure. The binding parameters of Avid AL for the whole IgG, Fab and Fc fragment and the stability of gel were investigated. The IgG bound to Avid AL can be eluted with an acidic buffer or with a novel neutral buffer containing electron donors. The development of such a mild neutral elution buffer is described. Application of Avid AL in a rapid gram-scale IgG purification was demonstrated. The possible mechanism of IgG binding is discussed.

Evaluation of the mode of binding of immunoglobulin to activated agarose

Determining the orientation of the immobilization of proteins to solid-phase matrices is of critical importance in the development of systems that employ immobilized proteins. Among these are enzyme-linked immunoassays, immobilized enzymes and affinity chromatography matrices. To determine the orientation of immunoglobulin G (IgG) on activated agaraoses, we coupled the immunoglobulin covalently to various activated matrices. The IgG was then cleaved with papain and the liberated fragments collected and analyzed using high-performance liquid chromatography. Only F ab fragments could be detected regardless of the activation method used. This implies that IgG binds to these matrices predominantly via the F c domain. In order to develop a quantitative method of measuring the F ab and F c fragments, we compared the binding of IgG and its papain cleavage fragments to S-Zephyr columns and Mono-S columns. Comparison between these columns showed that IgG is bound more tightly to the S-Zephyr column and, in contrast, its retention on Q-Zephyr is less than on a comparable Mono-Q column. The resolution of IgG and its fragments was better in all cases on S-Zephyr than on Mono-S under the conditions employed.

Neutrophil-induced immunoglobulin binding to erythrocytes involves proteolytic and oxidative injury

Neutrophil-induced alterations in feline erythrocytes were studied to better understand the pathogenesis of erythrocyte destruction associated with inflammatory diseases. As in previous studies, addition of superoxide dismutase/catalase to a coculture of erythrocytes and activated neutrophils attenuated neutrophil-induced immunoglobulin G (IgG) binding. However, incubation of erythrocytes with hydrogen peroxide or neutrophil-derived anuclear cytoplasts (neutroplasts) failed to induce IgG binding. Addition of phenylmethylsulfonyl fluoride, a serine protease inhibitor, to the erythrocyte-neutrophil coculture attenuated IgG binding. These observations suggest that neutrophil-derived serine protease activity is involved in IgG binding to erythrocytes. Further, incubation of erythrocytes with serine proteases, but not metalloproteases or sulfhydryl proteases, induced immunoglobulin binding. Freeze-fracture replicas of the erythrocyte membrane failed to demonstrate clustering of band 3 protein, suggesting that spatial rearrangement of band 3 protein was not the cause of the IgG binding. Neutrophil-induced IgG binding due to the combined action of proteases and oxidants may explain the accelerated destruction of erythrocytes in inflammatory diseases.