Purification of soluble immune complexes from serum using polymethylmetacrylate beads coated with conglutinin or C1q. Application to the analysis of the components of in vitro formed immune complexes and of immune complexes occurring in vivo during leishmaniasis.

A procedure for the isolation of immune complexes from human sera has been developed. Two steps are involved: (1) lipid-free serum is precipitated by polyethylene glycol; (2) the solubilized precipitate is absorbed on a column of polymethylmethacrylate beads coated with conglutinin (K) or C1q; the column is washed, the complexes are then eluted, using 0.02 M EDTA (for K column) or 0.5 M NaCl (for C1q column). This procedure permitted the purification and the characterization of soluble 125I-BSA-anti-BSA, 125 I-tetanus toxoid-anti-tetanus toxoid, and 125-I-HBsAg-anti-HBsAg complexes made in vitro in the presence of fresh human serum. The isolated complexes were shown to contain antigen, antibody, C1q, C1r, C1s and C3. When normal human serum was submitted to such a procedure, no detectable amount of protein was present in the final eluted fraction. Immune complexes formed in vivo were also purified by conglutinin column from the serum of a patient with disseminated leishmaniasis. The isolated material was found to contain IgM, IgG, C1q, C1r, C1s, C3c and C3d. The purified complexes dissociated at acid pH were found to contain anti-IgG and anti-leishmania antibodies.     

Solid-phase enzyme immunoassay or radioimmunoassay for the detection of immune complexes based on their recognition by conglutinin: conglutinin-binding test. A comparative study with 125I-labelled C1q binding and Raji-cell RIA tests

Bovine conglutinin was used in a solid-phase assay for the detection of immune complexes. In a first step, the tested serum sample is incubated in polypropylene tubes coated with conglutinin to allow C3-coated immune complexes to bind to solid-phase conglutinin. In a second step, the conglutinin-bound complexes are detected using an enzyme-conjugated or radiolabelled anti-immunoglobulin antibody.

The conglutinin-binding (KgB) test does not suffer from the interference of DNA, heparin or endotoxins. Its limit of sensitivity for aggregated IgG is 3 μg/ml undiluted human serum. Immune complexes prepared in vitro using tetanus toxoid, or DNA, and corresponding antibodies in human sera could be detected at various antigen/antibody ratios and at antibody concentrations lower than 8 μg/ml. The KgB test allowed for the detection of immune complexes in sera from patients with systemic lupus erythematosus, rheumatoid arthritis, idiopathic vasculitis, leprosy and leukemia. These sera were also tested using the ¹²⁵I-labelled Clq-binding activity (BA) test and the KgB test simultaneously, and a significant rank order correlation was observed. In patients with leukemia, a significant correlation was observed using three tests, KgB, ¹²⁵I-labelled Clq BA and Raji-cell radioimmunoassay (RIA).

Therefore, the KgB test appears as a simple and reproducible method, utilizing a very stable reagent, with a sensitivity and specificity comparable to the other tests studied and allowing for clinical application.

     

Chronic liver disease: the detection and characterization of circulating immune complexes.

Circulating immune complexes containing IgG, IgM and hepatitis B surface antigen (HBsAg) in sera from groups of patients with various liver diseases were detected by both the C1q and conglutinin solid phase assays. Elevated levels of antigen non-specific immune complexes were observed in sera from all groups and complexes containing IgG were present to a greater extent than were IgM-containing complexes. Higher levels of complexes were generally obtained using the conglutinin assay than the C1q assay and the two assays were shown to preferentially bind complexes of different size ranges and antigen-antibody ratios. Only sera from HBsAg-positive patients had complexes containing HBsAg, and although serum HBsAg titres and levels of HBsAg-containing complexes were correlated, the correlation coefficient was low. The mean levels of immune complexes and the frequency of positive sera varied between different disease categories, but there was little correlation between levels of the three types of complexes detected by the two tests. Assay of immune complexes in sequential serum samples from an individual patient revealed considerable variation in the levels of the three complex types, demonstrating that the measurement of complexes in single serum samples is of limited value in assessing the potential significance of circulating immune complexes in hepatitis B.     

Antigen detection in immune complexes by a modified staphylococci binding assay and Western blot analysis

The combination of a modified Staphylococci binding assay for immune complexes and Western blot analysis is described for the isolation and detection of antigen in immune complexes from human sera. The strategy of the procedure is to preclear immune complexes from other serum components by sequential polyethylene glycol precipitation and incubation with insoluble protein A under conditions in which immune complexes are preferentially bound. Immune complexes are eluted from protein A in sodium dodecyl sulfate buffer and dissociated by acrylamide electrophoresis. Resolved proteins are then transblotted to nitrocellulose, and antigen is detected with specific antibody. Immune complexes were prepared in vitro with an antigen (keyhole limpet hemocyanin) that focuses well on electrophoresis and for which a potent immunologic probe (antibody) was available. In this system, antigen could be detected when immune complexes were present in sera in concentrations as low as 20 micrograms aggregated-IgG eq/ml, regardless of antigen-antibody ratio. We demonstrate the detection of horse globulin in immune complexes from a patient with acute serum sickness and hepatitis B virus antigen in complexes from a patient with vasculitis.     

Antigen-specific detection of soluble immune complexes in conglutinin-binding assays.

A modification of the conglutinin-binding assay has been used for antigen identification in circulating immune complexes (IC). The complexed antigen was identified, in conglutinin-bound complexes, by antibodies or antibody fragments, specifically directed against the antigen. These antibody preparations were either radiolabelled or detected with 125I-protein A. This system was found efficient using in vitro-formed bovine serum albumin--anti-bovine serum albumin and tetanus toxoid--anti-tetanus toxoid soluble immune complexes at equivalence or at slight antigen or antibody excess. In addition, with 125I-IgG--anti-HBs and with 125I-F(ab')2--anti-HBs, we examined 44 sera from 41 patients with viral type B hepatitis and the presence of HBs was observed in 18 (40.9%) IC containing tested samples. Therefore, this method appears applicable to clinical investigation.     

Isolation and characterization of circulating immune complexes in cystic fibrosis.

A methodology for the isolation and immunologic characterization of IgG-containing circulating immune complexes (IgG-CIC) as detected by the 125IClq-binding assay (ClqBA) is described. We applied this methodology to sera from patients with cystic fibrosis (CF), both positive and negative for IgG-CIC. We used latex-fixation-positive rheumatoid arthritis sera and normal human sera as positive and negative controls, respectively. All IgG-CIC-positive serum samples from CF patients were found to contain antibodies against Pseudomonas aeruginosa in the isolated complexes. Some patients also had antibodies in CIC specific for Staphylococcus aureus and Candida albicans. CIC specificity corresponded to respiratory tract colonization for each patient.     

Detection and characterization of immune complexes by the platelet aggregation test. II. Circulating complexes

Model experiments with preformed antigen–antibody complexes have shown that the platelet aggregation test can be used to characterize immune complexes. The platelet aggregating activity of four sera behaved like antigen–antibody complexes with respect to the effect of added antigen, added antibody and sedimentation in density gradient centrifugation. Two of the sera were from patients with fever of unknown origin. The findings suggest that the two sera contained circulating immune complexes which had both measles hyperimmunization induced and normal specificities. The serum of a patient with fatal subacute hepatitis seemed to contain immune complexes with Australia antigen. It was also positive in many of the autoimmunity tests. One of the sera was from a patient with haemorrhagic varicella in the acute phase of the disease. The patient had received large amounts of γ-globulin. The results suggested the transient presence of circulating immune complexes with varicella specificity.     

Detection of circulating immune complexes in human sera by simplified assays with polyethylene glycol

The search for circulating immune complexes by precipitation tests using polyethylene glycol (PEG) was performed on a series of normal and pathological sera. Various factors affecting PEG precipitation were studied. Immunoglobulins and complement factors precipitated by PEG (3.5%) were quantified and their significance was discussed in relation to serum levels. The PEG test was compared to labeled C1q binding test with a fairly good correlation. The direct evaluation of the amount of C4 precipitated with IgG by 3% PEG (C4 test) provided a simpler routine assay than the C1q binding test for detecting complement-fixing immune complexes. The direct PEG test and the C4 test gave positive results in patients with diseases generally presumed to be associated with immune complexes including systemic lupus erythematosus, acute glomerulonephritis, bacterialsub-acute endocarditis and chronic active hepatitis. The demonstration of HBs antigen and antibody after acid dissociation of PEG precipitates from heptitis B seronegative sera illustrated the fact that PEG does precipitate and thus concentrates circulating immune complexes.     

Characterization of the Specificity of Monoclonal Antibodies against Mycobacterium tuberculosis by Crossed Immunoelectrophoresis

Isopycnic ultracentrifugation is frequently applied for preparative isolation of macromolecules. Using bovine serum albumin (BSA)-anti-BSA antibody complexes as a model system, isopycnic banding of complexes was observed in CsCl, Nycodenz, and sucrose gradients. In CsCl gradients, free antigen or antibody could not be separated from the immune complexes. Variations in antigen to antibody ratio from equivalence and in the amount of complement present during complex formation resulted in zone broadening and banding at slightly lower densities in Nycodenz. This was not observed in sucrose. Serum immune complexes were isolated from patients with rheumatoid arthritis or ankylosing spondylitis. Banding of in vivo-formed immune complexes was observed more frequently in sucrose than in Nycodenz.     

Calcium-dependent and calcium-independent signals in the conglutinin-binding assay (KgBa) for immune complexes. Influence of anti-collagen-antibodies.

A solid phase ELISA conglutinin-binding assay (KgBa) was evaluated for the detection of circulating immune complexes. ELISA wells were coated with purified bovine conglutinin and incubated with test sera. Bound IgG was detected with enzyme labelled anti-immunoglobulin. Heat aggregated IgG which had been "solubilized" (i.e., complement treated by incubation with serum) was employed as a reference. The binding of the complement-reacted IgG to solid phase conglutinin was found to be calcium-dependent and inhibitable with N-acetyl-D-glucosamine (GlcNAc). Prolonged incubation (4 days) of aggregated IgG with serum at 37 degrees C abolished the binding to conglutinin, a finding consistent with the complete degradation of deposited C3b to C3c and C3d. The solubilized IgG that bound to solid phase conglutinin was found by gel chromatography to be of high molecular weight (greater than 600 kDa). Binding of IgG to solid phase bovine conglutinin was also observed to a variable degree in normal and pathological sera. However, in this situation the IgG binding was largely calcium-independent, was not inhibited by GlcNAc and did not decrease after prolonged incubation of the serum at 37 degrees C. The reactive IgG eluted on gel chromatography at the position of monomeric IgG suggesting binding via the antigen binding sites. Binding of this IgG was inhibited by both collagen type II and purified conglutinin. These observations suggest that the assay detects cross-reacting autoantibodies against collagen epitopes, or, alternatively, antibodies against the dietary antigen, bovine conglutinin.     

Characterization of antigen moiety of HBsAg in the complement fixing immune complexes of hepatitis B virus positive chronic active hepatitis.

Circulating immune complexes (CIC) are frequently found in hepatitis B virus-induced chronic active hepatitis. Since antigen and antibody moieties of complexes are critical in determining many of its pathogenic factors, the constituents of these complexes were investigated with particular attention to the quantity and nature of the HBs antigen moiety of the complexes. Complement fixing immune complexes were isolated from sera of 14 patients with chronic active hepatitis by utilizing conglutinin's unique property to bind C3-fixed complexes. Low pH (2.6) was used to dissociate the complexes. Components of dissociated complexes were separated into antigen and antibody fractions using immobilized Protein A. Both fractions were analysed by electrophoresis in polyacrylamide gel with SDS. The antibody fractions showed heavy and light chains of IgG and IgM. The antigen fractions demonstrated six to 10 protein bands with mol. wt ranging between 17,000 and 120,000 daltons. To define precisely the polypeptide antigen moiety involved in the immune complex formation, a transfer blotting technique was used employing human anti-HBs globulin as probe. Polypeptides with mol. wt 97,000 and 49,000 reacted as antigen moieties of HBsAg. In addition, the levels of HBsAg in the antigen fractions were significantly greater (P less than 0.005) compared to sera from patients with acute hepatitis. Implications of these findings are discussed.     

The isolation of the antibody moieties of immune complexes from serum by the pepsin digestion of conglutinin-anti-conglutinin complexes.

A technique is described which allows the antibodies of circulating immune complexes to be isolated as their F(ab')2 fragments. The method is based on the precipitation of the complexes by the sequential addition of conglutinin and anti-conglutinin, and the subsequent digestion of these precipitates by pepsin. Using this technique it has been possible to show antibodies to Epstein-Barr (EB) virus antigens in the immune complexes of patients with Burkitt's lymphoma and to microbial antigens in two patients with nephritis. By substituting DNAase for pepsin it has also been possible to show antibodies to DNA-containing nuclear antigens in the serum of patients with systemic lupus erythematosus.     

Demonstration of HBsAg as the antigen component in circulating immune complexes detected by peg-solid phase test

The development of an enzyme-linked immunosorbent assay to identify HBsAg as the antigen component within circulating immune complexes using immobilized polyethylene glycol (PEG) is described. The method utilizes, on one hand, the ability of PEG to bind stably to plastic supports and, on the other, to precipitate circulating macromolecules. This method is easily performed, very cheap, quick and, above all, it helps define the biological nature of the immune complexes. HBsAg can be revealed as the antigen component of HBsAg/anti-HBs soluble immune complexes at concentrations of at least 20 ng/ml and either in antigen or antibody excess. Our results indicate that HBsAg circulates in a complexed form in 47% of HBsAg chronic carriers and in 10.7% of patients with liver disease who are positive for serum antibody to hepatitis B surface antigen (anti-HBs) and to core antigen (anti-HBc). None of the other groups of patients in the study had circulating HBsAg in the complexed form.     

Antigen-specific detection of soluble immune complexes by a solid phase specific antibody system.

An antigen-specific immune complex assay based on the following principle has been developed: xenogeneic, antigen-specific antibodies are attached to a solid phase. During first incubation with patient's serum, immune complexes in antigen excess are bound to the xenogeneic antibody by their free antigenic determinants. In a second incubation a labeled antibody, specific for human immunoglobulins, is combined with the antibody part of the immune complex. Quantitation of the label allows the determination of the immune complexes. The principle of the method has been varied using artificial soluble immune complexes of tetanus toxoid and human anti-tetanus toxoid antibody, and immune complexes prepared by mixing patient sera containing either HBsAg or anti-HBsAg antibodies. The reliability of the results is shown by their coefficient of variation (2.5%). With the method described soluble immune complexes predominantly in slight antigen excess, which are thought to be responsible for development of immune complex disease, have been detected.     

The alternative complement pathway control protein H binds to immune complexes and serves their detection

During solubilization of immune complexes C3b becomes fixed to the immunoglobulin part and serves as a receptor for the alternative complement pathway control protein H. The H-C3b immune complex interaction can be made detectable using 4% polyethyleneglycol to separate free from bound 125I-H. Tetanus toxoid (Te)/anti-Te complexes kept soluble with fresh serum and containing 125 IU of specific antibody bound 18% of 125I-H; when fresh serum was chelated with 10 mM EDTA, 125I-H binding was only 5%. On sucrose density gradients, the H-binding material sedimented in the range of 12 to 30 S. In 36 serum samples from rheumatoid arthritis (RA) patients and in 12 serum samples from patients with systemic lupus erythematosus (SLE), 125I-H binding was significantly elevated to 9.5 +/- 4.7% (mean +/- 1 SD) and 13.3 +/- 5.6%, respectively, while 125I-H binding by 36 normal human sera was 4 +/- 2%. RA samples (17/36, 47%) and SLE samples (9/12, 75%) had H-binding values increased by more than 2 SD above the normal mean. The serum samples were also assessed for conglutinin- and C1q-binding activities; a significant correlation between H and C1q binding was observed (P less than 0.001); there was no correlation between H and conglutinin binding. Although binding to immune complexes through its interaction with C3b, H clearly detects a population of complexes other than conglutinin, thus expanding the possibilities of further characterizing pathological complexes.     

Circulating immune complexes in patients with usual interstitial pulmonary fibrosis: partial characterization and relationship with Thermoactinomyces vulgaris.

Sixty-six sera were analysed by solid-phase conglutinin binding assay, to detect the levels of circulating immune complexes (CIC), and by enzyme-linked immunosorbent assay (ELISA) to show a correlation with antibodies to Thermoactinomyces vulgaris. Sixty per cent of patients with usual interstitial pulmonary fibrosis (UIP), were positive for CIC; and T. vulgaris antibodies were detected in 60% of the same patients. In comparison, there was a low frequency of positive results in bronchitis patients (5% for CIC and 35% for T. vulgaris), and in normal blood donors (0% for CIC and 30% for T. vulgaris). Furthermore 31% of patients with lung cancer were found positive for CIC, but not for T. vulgaris. Immune complexes purified on Protein A-Sepharose and by sucrose density gradient from patients with UIP, showed a sedimentation coefficient higher than 19 S. The purified material was found to contain IgG and IgM as antibodies. Binding of immune complexes, purified by sedimentation on sucrose gradient, to conglutinin was inhibited by the presence of T. vulgaris antigen; thus suggesting that this antigen might be present in the complexes.     

Isolation and characterization of soluble insulin–anti-insulin immune complexes formed in vitro and in vivo in sera from patients with diabetes mellitus

The existence of soluble insulin–anti-insulin immune complexes in the serum of patients with diabetes mellitus was investigated. Formation of such immune complexes in vitro was studied by adding radioiodinated insulin to the sera of patients with anti-insulin antibodies; immune complexes were formed readily, but apparently differed from patient to patient. Immune complexes formed in vitro were precipitated with 5% polyethylene glycol. They eluted in the high-molecular weight fractions when the precipitated material was fractionated by gel filtration, and they remained bound at neutral pH when the high-molecular weight fractions were submitted to affinity chromatography on protein A–Sepharose. When the bound immune complexes were recovered by acid elution and immediately filtered through a Sephadex G-50 column equilibrated with the same acid buffer, free antigen (radiolabelled insulin) and antibody were recovered. This antibody, after neutralization, showed binding capacity when remixed with radioiodinated insulin. When this protocol was applied to a serum that gave positive results in several screening methods for soluble immune complexes, insulin was detected by radioimmunoassay in the high-molecular weight fractions separated from the 5% polyethylene glycol precipitate and in the fractions retained by protein A; however, no free insulin was detected after gel filtration in Sephadex G-50, perhaps due to excessive dilution. The high-molecular weight fraction did have binding capacity for radioiodinated insulin. No insulin-binding protein could be recovered with a similar procedure from a serum negative by all screening tests for soluble immune complexes. These results prove that soluble immune complexes can be formed easily in sera containing anti-insulin antibodies and can be recovered from sera of diabetic patients that show positive results in screening techniques for soluble immune complexes.     

Detection of C1q-bearing immune complexes by a monoclonal anti-C1q ELISA system

A monoclonal antibody directed against the collagenous portion of human C1q was used to detect C1q-bearing immune complexes in patients with rheumatic disorders. Sera of patients with rheumatoid arthritis, systemic lupus erythematosus (SLE), osteoarthritis, as well as normal human sera (NHS) used as controls were tested in an ELISA system. C1q-bearing immune complexes were bound to a solid-phase monoclonal anti-C1q antibody, and detected with F(ab')2 antibodies to human IgG. Heat-aggregated human IgG was adjusted to the same concentration as the WHO standard for immune complexes and used for the standard curve in NHS. The mean value in NHS was 19.5 micrograms/ml equivalents of aggregated IgG. Using 2 SD over the mean as the upper limit for normal values, samples greater than 43 micrograms/ml were considered positive. Patients with osteoarthritis were negative; high levels of C1q-bearing immune complexes were detected in patients with rheumatoid arthritis (up to 800 micrograms/ml equivalents of aggregated IgG). With our assay C1q-bearing immune complexes were detected with high frequency (81%) in the sera of patients with rheumatoid arthritis, while a C1q solid-phase binding assay (C1q SPBA) revealed positive results only in 67% of rheumatoid arthritis sera. Compared to NHS, CH50 titers and C1q values of sera from patients with rheumatoid arthritis were frequently high. In contrast, the sera of SLE patients with low CH50 titers and low C1q levels had IgG immune complexes which could be detected only in the C1q-SPBA. C1q-bearing immune complexes were not detectable in the sera of patients with SLE. Since C1q triggers activation of the classical C pathway, this assay with monoclonal anti-C1q antibody appears to be useful for detecting immune complexes in rheumatoid arthritis patients with normal or elevated CH50 and C1q values, especially in the early stage of the disease.     

Hepatitis B surface antigen containing immune complexes occur in seronegative hepatocellular carcinoma patients

IgG, IgM and hepatitis B surface antigen (HBsAg) containing immune complexes (IC) were detected by the Clq and conglutinin solid phase assays in both HBsAg+ and HBsAg- groups of patients with primary hepatocellular carcinoma (HCC). No differences were observed between the two patient groups either in the levels of antigen non-specific and HBsAg specific complexes or in the immunoglobulin isotype in the complexes. The results show that HBsAg can occur in an IC form in the sera of patients classified as HBsAg- by sensitive commercial assays and provides evidence of a further association of hepatitis B virus (HBV) and HCC in antigen negative patients. Furthermore, the HBsAg IC in HCC patients differ from those in other HBV infected subjects in that they are preferentially detected by the Clq assay.     

Complement activating properties of complexes containing rheumatoid factor in synovial fluids and sera from patients with rheumatoid arthritis.

The relationship between complexes containing rheumatoid factor and complexes activating complement was examined in synovial fluids and sera from patients with rheumatoid arthritis (RA). In each case this was performed by quantifying the amount of rheumatoid factor bound by solid phase Fab'2 anti-C3 and/or solid phase conglutinin. Both anti-C3 coated and conglutinin coated microtitre plates bound high levels of complexes containing rheumatoid factor from sera of RA patients with vasculitis. Unexpectedly, these complexes were detected in synovial fluids from only a minority of RA patients with synovitis. However, RA synovial fluids did contain other complexes as shown by the presence of complement consuming activity, C1q binding material and immunoglobulin attaching to conglutinin. It is considered that in RA synovial fluids the complexes containing RF and those activating complement are not necessarily the same whilst in vasculitic sera the complexes containing rheumatoid factor also activate complement.