Variable region structure and staphylococcal protein A binding specificity of a mouse monoclonal IgM anti-laminin-receptor antibody.

Staphylococcal protein A is a cell wall-attached polypeptide that acts as a B-lymphocyte superantigen. This activation correlates with specific VH gene segment usage in the B-cell receptor. B-cell receptor assembled from members of the VH3 family in humans, or S107 family in mice, has an intrinsic affinity for protein A. Human VH3-derived antibodies bind to domain D of protein A. We have characterized a mouse IgM monoclonal antibody that binds protein A. The sequencing of the variable region suggests an almost germline-encoded VH derived from S107 family and a V kappa 8-derived VL. The binding specificity of the monoclonal antibody was tested with various recombinant constructions derived from protein A. We show that, unlike human VH3-derived antibody, mouse S107-derived immunoglobulin binds to the B domain of the bacterial superantigen.

Molecular analysis of spontaneous nephrotropic anti-laminin antibodies in an autoimmune MRL-lpr/lpr mouse.

To explore the genetic relationship between anti-laminin and anti-DNA autoantibodies (autoAb), VH gene and gene family expression were determined among autoAb derived from an individual 6-mo-old MRL-lpr/lpr mouse. Whereas 85% of the anti-DNA Ig were identified by one of two VH family probes, 7183 and VHJ558, none of the anti-laminin antibodies (Ab) examined were recognized by these probes. Subsequent V region sequence analysis of three of the anti-laminin Ab revealed that they in fact utilized a J558 VH gene (VH50). Furthermore, FR2 and CDR2 oligonucleotide probes complementary to VH50 recognized multiple anti-laminin Ab by Northern blot analysis; the FR2 probe recognized two control anti-DNA Ab, but neither probe recognized anti-DNA Ab from the same mouse. Polymerase chain reaction amplification of MRL-lpr/lpr genomic liver DNA using primers generated from VH50 and Vk50 sequences indicated that all three anti-laminin Ig have a single replacement mutation in both their VH and Vk genes. Search of the nucleic acid databases revealed that both germline VH and Vk genes are expressed unmutated by murine lupus anti-dsDNA autoAb, previously sequenced in other laboratories. Sequence comparisons suggest that differences in anti-DNA and anti-laminin reactivity may be dependent upon somatically generated differences in the CDR3 regions of the H and L chains. The results indicate that lupus anti-laminin Ab can arise from distinct B cell populations but express the same unmutated germline V region genes as lupus anti-dsDNA autoAb. They further raise the possibility that these distinct B cell populations may be activated and expanded either: independently, by distinct Ig receptor ligands such as the Ag, laminin and DNA; or simultaneously, by a common ligand such as an anti-Id recognizing a common V region epitope.

Acute agranulocytosis after prolonged high-dose usage of intravenous dipyrone--a different mechanism of dipyrone toxicity?

Two seriously injured trauma patients presenting with intense and progressive neutropenia are described. Bone marrow examination in both cases showed virtually absent granulopoiesis but normal erythropoiesis and megakaryopoiesis, allowing the diagnosis of acute agranulocytosis. Discontinuation of only one drug (dipyrone) with no further treatment was required for normalization of blood parameters. The association of dipyrone with neutropenia is still debatable. The recent medical literature on dipyrone generation of agranulocytosis is reviewed.

Anti-platelet autoantibodies from ITP patients recognize an epitope in GPIIb/IIIa deduced by complementary hydropathy.

Idiopathic thrombocytopenic purpura (ITP) is a frequent platelet disorder due to the presence of anti-platelet autoantibodies. Recently a fibronectin/fibrinogen receptor in platelets, integrin GPIIb/IIIa, has been implicated as the antigen in chronic ITP. To examine the epitopes involved in the autoimmune response against GPIIb/IIIa we have used concepts from the complementary hydropathy principle. We used the peptide Trp-Thr-Val-Pro-Thr-Ala, WTVPTA (deduced from the complementary nucleotide sequence to that which codes for the Arg-Gly-Asp, RGD, domain in fibronectin), to test the immunologic activity of ITP sera. Sera from 31 patients with clinically defined ITP were tested in ELISA for reactivity towards WTVPTA and affinity purified GPIIb/IIIa. Seventeen sera (57%) reacted strongly with the glycoprotein complex, five of which reacted with the peptide. By affinity chromatography of one of these sera, we were able to show that antibodies that bind to the peptide are within the population that binds to GPIIb/IIIa. Liquid phase competition experiments revealed that binding of ITP serum to WTVPTA was inhibited only by a hydropathically compatible peptide. Our data indicate that autoantibodies can bind to hydropathically generated antigenic determinants and thus, render these peptides clinically important as diagnostic tools.

A "primary" thrombotic syndrome: absence of anti-phospholipid antibodies.

A patient presenting fetal loss, livedo reticularis, severe migraine crises, myocardial infarction and thrombotic vasculopathy of both external iliac arteries is described. The serologic study showed absence of antiphospholipid antibodies (aPL), suggesting that in some cases the presence of these antibodies may be a consequence of tissue damage.

Differences in human laminin B2 sequences.

A cDNA clone encoding the B2 chain of laminin has been isolated from a human endothelial lambda gt11 cDNA library. The nucleotide and deduced amino acid sequences of the clone were determined and showed one amino acid substitution (Ser1519 instead of Asn) when compared to lung laminin B2 chain and one silent nucleotide change (G4200 instead of A) in relation to the human placenta laminin B2 chain. Other differences in the 3'-untranslated region were also found.

Human and murine anti-DNA antibodies induce the production of anti-idiotypic antibodies with autoantigen-binding properties (epibodies) through immune-network interactions.

To examine the potential role of immune-network interactions in the production of lupus autoantibodies, normal NZW rabbit antibody responses were analyzed after immunization with one of the following Ig preparations: human lupus serum anti-dsDNA antibodies, human lupus serum anti-ssDNA antibodies, a mixture of human lupus serum anti-dsDNA and anti-ssDNA antibodies, the MRL-lpr/lpr anti-dsDNA mAb H241, and the MRL-lpr/lpr anti-ssDNA mAb H130. Four of five rabbits produced Ig typical of lupus autoantibodies: individual rabbit Ig cross-reacted with multiple autoantigens including nucleic acids, cardiolipin, SmRNP, glomerular extract, laminin, and exogenous Ag. Rabbit anti-Id against human anti-dsDNA antibodies were highly specific for dsDNA. Notably, in each serum the autoantibody activity was confined to the anti-Id Ig fraction. A similar spontaneously occurring Id-anti-Id interaction was also found between anti-ssDNA and anti-dsDNA antibodies isolated from an individual lupus patient. These results indicate that lupus autoantibodies which share Ag binding properties with pathogenic Ig, including both cross-reactive and anti-dsDNA antibodies, can induce the production of Ig with similar autoantigen binding properties through immune-network interactions. This phenomenon, if unregulated, could lead to the amplification of pathogenic autoantibody production in individuals with systemic lupus.

Comparative study of idiotypes on monoclonal antibodies derived from patients with lupus and leprosy and from normal individuals.

A collaborative study was performed to compare the expression of a series of idiotypes defined on human anti-DNA and other autoantibodies. Three panels of human monoclonal antibodies were tested: eight derived from patients with systemic lupus erythematosus (SLE); 13 from an individual with lepromatous leprosy; and 38 from normal subjects. The following rabbit anti-idiotype sera were used: one (RId16/6) raised against the lupus-derived monoclonal anti-DNA antibody 16/6, four (RId8E7, RId4G7, RId4D5 and RIdTH9) against leprosy-derived monoclonal antibodies of various specificities, and one (anti-4.6.3) against a normal-derived anti-DNA monoclonal (KIM 4.6). In addition, two other anti-idiotypes were used--one a murine monoclonal (3I), the other a rabbit polyclonal (RIdD)--which had been raised against polyclonal anti-DNA antibodies from lupus serum. Further experiments were performed with immunoabsorbed fractions of RId8E7. Direct-binding and competition assays were used. All of the anti-idiotypes produced different patterns of positivity among the three panels of human monoclonal antibodies, with the exception of RId8E7 and RId4G7, which showed considerable concordance. There was a tendency towards anti-idiotypes being disease- or group-specific: thus anti-4.6.3 failed to bind to any of the lupus or leprosy-derived monoclonals, while RId16/6 and RId8E7 bound most strongly to the lupus- and leprosy-derived antibodies respectively. KIM 4.6 itself was bound only weakly by RId16/6, while 16/6 was not recognized by anti-4.6.3; 16/6 was, however, bound by 3I, while KIM 4.6 was not. 3I bound to several other monoclonals but RIdD, which has been shown to be specific for the anti-DNA fraction of lupus serum, did not bind to any of them. These results indicate that the majority of these anti-idiotype preparations recognize largely separate sets of determinants. The monoclonal antibodies which bind to DNA may be only partly representative of anti-DNA antibodies in the serum of lupus patients.

Are synthetic peptides suitable for the detection of continuous epitopes only?

Present-day methods for the definition of antibody binding sites on antigenic polypeptides show a strong bias toward identification of sequential epitopes. We have employed anti-sequence monoclonal antibodies raised against short synthetic peptides to screen a random-primed cDNA library constructed in an expression vector. Sequence data analysis performed on three clones thus isolated showed that the clones did not cross-hybridize and that the antigenic peptide sequence was found in none of them. Our findings suggest therefore that sequential epitopes may be preferentially identified because of an inherent bias in commonly used epitope mapping protocols which masks available conformational determinants.

Cross-reactivity distinguishes serum and nephritogenic anti-DNA antibodies in human lupus from their natural counterparts in normal serum.

To distinguish the properties of anti-DNA antibodies in patients with lupus from those in normal individuals, we compared the ligand binding, idiotypic and charge properties of serum anti-DNA antibodies derived from: patients with active lupus; normal individuals; and among Ig eluted from the kidneys of two patients with active lupus nephritis (one with mesangial proliferation and the other with membranous nephropathy). The kidney eluate anti-DNA antibodies were the most cross-reactive; they cross-reacted with ssDNA, poly(GdC), poly(dT), poly(dG), poly(dC), ZDNA, SmRNP and the phospholipids cardiolipin and phosphatidyl serine. Lupus serum anti-DNA antibody cross-reacted with polynucleotides but not with phospholipids, whereas anti-DNA antibodies derived from normal serum reacted only with poly(dT). An anti-idiotype (anti-IdD; produced against serum anti-DNA antibodies from one patient) reacted with: anti-DNA antibodies in 8/9 lupus sera; antibodies in both kidney eluates; and anti-DNA antibodies from 5/7 normal sera. Anti-IdD did not react with Ig that did not bind to DNA. Isoelectric focusing of Ig showed that the charge of anti-DNA antibodies from lupus serum and normal serum were similar and unrestricted (pI 5.4-9.0); Ig in kidney eluates varied: membranous lupus pI 4.5-8.6; mesangial lupus pI 8.1-9.1. We conclude that idiotypically related anti-DNA antibodies in tissue lesions, lupus serum and normal serum from different individuals can be distinguished on the basis of their cross-reactive antigen-binding properties. Furthermore the cross-reactive properties of lupus auto-antibodies may influence their capacity to form glomerular immune deposits.

Identification and characterization of highly conserved antigenic determinants in the laminin molecule.

1. Fragments P1 and E8, the result of two different enzymatic digestions of the laminin molecule, represent interaction sites of laminin with specific cell receptors. By using negative and positive affinity purification of a rabbit antiserum against mouse laminin we have generated antibodies to these two fragments. 2. Antibodies against P1 were able to immunoprecipitate fragment E8 from elastase-digested laminin. By liquid phase competition experiments we demonstrated that the epitopes shared by P1 and E8 are a minor portion of the antigenic determinants of P1. When we checked for the presence of these shared epitopes in the human laminin molecule, they were the major fraction of the interspecies antigenic conservation. 3. A similar approach using polyclonal antibodies against human laminin has confirmed these results. 4. The shared epitopes present in both mouse and human laminin molecules seem to be spatially determined, because antibodies against these sites did not bind to fully denatured laminin. 5. Since human and mouse laminin bind to cell receptors and to other extracellular matrix proteins from both species, we conclude that these antigenic determinants may represent the actual sites for at least some of these interactions.

A murine nephritogenic monoclonal anti-DNA autoantibody binds directly to mouse laminin, the major non-collagenous protein component of the glomerular basement membrane.

The interaction of the murine monoclonal anti-DNA antibody H241 with extracellular glomerular antigens was found to be due to the binding of this antibody to laminin, the major non-collagenous protein constituent of the glomerular basement membrane. This interaction is specific, since it is inhibited by laminin, double-stranded DNA and single-stranded DNA in solution. Furthermore, the binding of H241 to mouse laminin is mediated by conformational properties of the antigen because mild denaturation of laminin strongly decreases the binding capacity of H241, while exposure of laminin to sodium dodecyl sulfate, completely abolishes this interaction. H241 is able to bind to both, human and mouse laminin. These findings are in agreement with the ligand binding specificities of the autoantibodies spontaneously produced, that differ from those generated by artificial immunization. We conclude that the polyreactivity of H241 that confers to it the capacity to bind laminin, may account for its ability to form immune deposits by binding directly to non-DNA glomerular antigens.

Idiotypic markers of polyclonal B cell activation. Public idiotypes shared by monoclonal antibodies derived from patients with systemic lupus erythematosus or leprosy.

We investigated idiotypic markers of monoclonal antibodies derived from patients with polyclonal B-cell activation. Four monoclonal antibodies with different ligand binding specificities derived from a patient with lepromatous leprosy and three monoclonal anti-DNA antibodies from two patients with SLE were studied. Three new public idiotopes, which were common to monoclonal antibodies from all three patients, were defined by five polyclonal rabbit antiidiotypes, two monoclonal mouse antiidiotopes, and a monoclonal mouse antibody against a synthetic peptide that contains residues of the heavy chain CDR-1 of a monoclonal lupus anti-DNA antibody. The antibody against the synthetic idiotype was found to react with native immunoglobulins in solution. One idiotope was found to be consistently immunogenic in all animals tested. Since the three patients are of different ethnic origins, these shared idiotypes are probably encoded by germline V genes. These genes may be recurrently expressed in states of polyclonal B-cell activation, regardless of etiology. The results suggest that some autoantibodies arise by expansion of a pool of precursors in the normal antibody repertoire.