Purification of human autoantibodies from cross-linked antigen immunosorbents

Fine specificity of the autoimmune response to the Ro/SSA and La/SSB ribonucleoproteins

The fine specificity of the Ro and La proteins has been studied by several techniques. In general, there is agreement in a qualitative sense that autoantibodies bind multiple epitopes. For some specific antibody binding, different studies agree quantitatively, for instance, the binding of the carboxyl terminus of 60-kd Ro as described by 2 studies using different techniques and the presence of an epitope within the leucine zipper of 52-kd Ro. In addition, there is general agreement about the location of a prominent epitope at the RRM motif region of the La molecule. On the other hand, the many specific epitope regions of the molecules differ among these studies. These discrepancies are likely the result of using different techniques, sera, and peptide constructs as well as a result of inherent advantages and disadvantages in the individual approaches. Several theories concerning the origin of not only the antibodies, but also the diseases themselves, have been generated from studies of the fine specificity of antibody binding. These include a theory of a primordial foreign antigen for anti-Ro autoimmunity, molecular mimicry with regard to La and CCHB, as well as the association of anti-Ro with HLA. These remain unproven, but are of continuing interest. An explanation for the association of anti-60-kd Ro and anti-52-kd Ro in the sera of patients has sprung from evaluating antibody binding. Data demonstrating multiple epitopes are part of a large body of evidence that strongly suggests an antigen-driven immune response. This means that the autoantigens are directly implicated in initiating and sustaining autoimmunity in their associated diseases. A number of studies have investigated the possibility of differences in the immune response to these antigens in SS and SLE sera. While several differences have been reported, none have been reproduced in a second cohort of patients. Furthermore, none of the reported differences may be sufficiently robust for clinical purposes, such as distinguishing between SS with systemic features and mild SLE, although some might be promising. For instance, in at least 3 groups of SLE patients, no binding of residues spanning amino acids 21-41 of 60-kd Ro has been found. Meanwhile, 1 of those studies found that 41% of sera from patients with primary SS bound the 60-kd Ro peptide 21-41. Perhaps future studies will elaborate a clinical role of such a difference among SS and SLE patients. Study of the epitopes of these autoantigens has, in part, led to a new animal model of anti-Ro and anti-La. Non-autoimmune-prone animals are immunized with proteins or peptides that make up the Ro/La RNP. Such animals develop an autoimmune response to the entire particle, not just the immunogen. This response has been hypothesized to arise from autoreactive B cells. In another, older animal model of disease, the MRL-lpr/lpr mouse, B cells have recently been shown to be required for the generation of abnormal, autoreactive T cells. Thus, there are now powerful data indicating that B cells that produce autoantibodies are directly involved in the pathogenesis of disease above and beyond the formation of immune complexes. Given that the autoreactive B cell is potentially critical to the underlying pathogenesis of disease, then studying these cells will be crucial to further understanding the origin of diseases associated with Ro and La autoimmunity. Hopefully, an increased understanding will eventually lead to improved treatment of patients. Progress in the area of treatment will almost surely be incremental, and studies of the fine specificity of autoantibody binding will be a part of the body of basic knowledge contributing to ultimate advancement. In the future, the animal models will need to be examined with regard to immunology and immunochemistry as well as genetics. The development of these autoantibodies has not been studied extensively because upon presentation to medical care, virtually all patients have a full-

Use of dextrans as long and hydrophilic spacer arms to improve the performance of immobilized proteins acting on macromolecules

New dextran-agarose supports, suitable for covalent immobilization of enzymes and proteins acting on macromolecular substrates, were prepared. The thick internal fibers of agarose gels were covered by a low-density layer of long, flexible, hydrophilic, and inert dextran molecules. Rennin and protein A were immobilized on these novel supports and the resulting derivatives exhibited a very high capacity for biological recognition of soluble macromolecular substrates. Caseinolytic activity of this immobilized enzyme was 15-fold higher than activity of directly immobilized rennin, through short spacer arms, on agarose gels. Similarly, the new derivatives of immobilized protein A were able to adsorb up to 2 molecules of immunoglobulin per each molecule of immobilized protein A. When the immobilized proteins were secluded away from the support surface by using these new long and hydrophilic spacer arms, they exhibit minimal steric hindrances that could be promoted by the proximity of the support surface.

Dual microcolumn immunoassay applied to determination of insulin secretion from single islets of Langerhans and insulin in serum

A dual microcolumn immunoassay (DMIA) was developed and applied to determination of insulin in biological samples. The DMIA utilized a protein G capillary column (150 microns I.D.) with covalently attached anti-insulin to selectively capture and concentrate insulins in a sample. Insulins retained in the capillary immunoaffinity column were desorbed and injected onto a reversed-phase capillary column (150 microns I.D.) for further separation from interferences such as cross-reactive antigens and non-specifically adsorbed sample components. Bovine, porcine and rat insulin all cross-reacted with the antibody and could be determined simultaneously. Using a UV absorbance detector, the dual microcolumn system had a detection limit of 10 fmol or 20 pM for 500-microliter sample volumes. The DMIA system was used to measure glucose-stimulated insulin secretion from single rat islets of Langerhans. Because of the separation in the second dimension, both rat I and rat II insulin could be independently determined. The system was also evaluated for determination of insulin in serum. Using microcolumns instead of conventional HPLC columns resulted in several advantages including use of less chromatographic material and improved mass detection limit.

Anti-La (SS-B) but not anti-Ro52 (SS-A) antibodies cross-react with laminin--a role in the pathogenesis of congenital heart block?

Cross-reactions between maternally derived autoantibodies and fetal cardiac antigens have been postulated to play a role in the pathogenesis of congenital heart block (CHB). We have explored the cross-reactivity of autoantibodies to the small ribonuclear autoantigens, La/SS-B and Ro/SS-A, with laminin, the major component of cardiac sarcolemmal membrane using affinity-purified antibodies from patients with Sjögren's syndrome (SS). Anti-La antibodies purified from eight of 10 patients cross-reacted significantly with mouse laminin by ELISA. In contrast, purified antibodies to Ro52 from the same 10 patients showed little or no binding to laminin. Laminin inhibited up to 70% binding of anti-La antibodies to La antigen, and La inhibited up to 65% binding of anti-La antibodies to laminin. The cross-reaction was further examined on cryosections of 10 human fetal hearts aged from 8.7 to 14.9 weeks of gestation, two normal adult hearts, and one pathological adult heart with a diagnosis of dilated cardiomyopathy. Anti-Ro52 antibodies did not bind to the surface of cardiac cells. However, anti-La antibodies from seven of 10 patients tested bound to the surface of fetal myocytes from hearts aged 9.4 to 14.9 weeks of gestation, and also to the myocytes from the pathological adult heart but not to normal adult hearts. Preincubation with La antigen abolished the binding of anti-La antibodies to the surface of adult heart myocytes with dilated cardiomyopathy, and pre-incubation with mouse laminin could partially block this binding. These results suggest that molecular mimicry between laminin and La, but not Ro52, may act as a target for specific maternal autoantibodies, and contribute to the pathogenesis of CHB at a critical stage during fetal cardiac development.

An [Na+ + K+]coupled L-glutamate transporter purified from rat brain is located in glial cell processes

Polyclonal antibodies were generated against the major polypeptide (73,000 mol. wt) present in a highly purified preparation of the [Na+ + K+]coupled L-glutamate transporter from rat brain. These antibodies were able to selectively immunoprecipitate the 73,000 mol. wt polypeptide as well as most of the L-glutamate transport activity--as assayed upon reconstitution--from crude detergent extracts of rat brain membranes. The immunoreactivity in the various fractions obtained during the purification procedure [Danbolt et al. (1990) Biochemistry 29, 6734-6740] closely correlated with the L-glutamate transport activity. Immunoblotting of a crude sodium dodecyl sulphate brain extract, separated by two-dimensional isoelectric focusing-sodium dodecyl sulphate-polyacrylamide gel electrophoresis, showed that the antibodies recognized one 73,000 mol. wt protein species only. Deglycosylation of the protein gave a 10,000 reduction in molecular mass, but no reduction in immunoreactivity. These findings establish that the 73,000 mol. wt polypeptide represents the L-glutamate transporter or a subunit thereof. The antibodies also recognize a 73,000 mol. wt polypeptide and immunoprecipitate L-glutamate transport activity in extracts of brain plasma membranes from rabbit, pig, cow, cat and man. Using the antibodies, the immunocytochemical localization of the transporter was studied at the light and electron microscopic levels in rat central nervous system. In all regions examined (including cerebral cortex, caudatoputamen, corpus callosum, hippocampus, cerebellum, spinal cord) it was found to be located in glial cells rather than in neurons. In particular, fine astrocytic processes were strongly stained. Putative glutamatergic axon terminals appeared non-immunoreactive. The uptake of glutamate by such terminals (for which there is strong previous evidence) therefore may be due to a subtype of glutamate transporter different from the glial transporter demonstrated by us.

Ro and La antigens and maternal anti-La idiotype on the surface of myocardial fibres in congenital heart block

Congenital complete heart block (CCHB) is a rare but potentially fatal disease of infants born to mothers with autoimmune disease where maternal autoantibodies to Ro (SS-A) are thought to cross the placenta and damage fetal cardiac tissue. We have adopted a novel approach to demonstrate the localization and specificity of maternal autoantibodies deposited in fetal heart. We raised an anti-idiotype against maternal anti-La antibodies, which reacted strongly with the surface immunoglobulin on the myocardial fibres from a CCHB heart but not a control fetal heart of the same age. Maternal immunoglobulin eluted from the CCHB heart reacted with La (SS-B) by ELISA. Using monoclonal and affinity-purified antibodies to La and affinity-purified anti-Ro antibodies, both antigens were identified on the surface of the fibres of the affected heart. Surface co-expression of immunoglobulin, complement and Class II antigen, consistent with a local immune response, was also found. This is the first definitive demonstration of Ro and La antigens and specific maternal anti-La antibody and idiotype on the surface of myocardial fibres in CCHB. It suggests that induction of Ro and La antigens on the surface of myocardial fibres during fetal development may be critical in the localization of the specific autoantibodies and subsequent evolution of congenital complete heart block.

Co-existent anti-La antibodies and rheumatoid factors bear distinct idiotypic markers

This study describes the distribution of isotypes and idiotypes of two autoantibody populations, anti-La and rheumatoid factor, which co-exist in both the sera and saliva of patients with primary Sjögren's syndrome. The two autoantibodies are distinguished not only by their antigenic specificity but also by the nature of their idiotypic markers. IgA anti-La antibodies bearing restricted idiotypes are specifically enriched in saliva compared to serum suggesting their local synthesis. In contrast, rheumatoid factors bear cross-reactive idiotypes and may arise as a direct consequence of the secondary immune response.