Chronic immune thrombocytopenic purpura (ITP) is due to platelet destruction by circulating antiplatelet antibody. Although autoantibodies against the platelet glycoprotein IIb/IIIa (GPIIb/IIIa) complex and GPIb have been demonstrated using various methods, practical assays for detection of platelet-associated or plasma autoantibodies have not been available. We studied 59 patients with chronic immune thrombocytopenic purpura in whom platelet-associated and plasma autoantibodies against the GPIIb/IIIa complex and GPIb were measured using a newly developed immunobead assay and a previously reported microtiter-well assay. Platelet-associated autoantibody was detected using the immunobead assay in 21 of 28 patients (75.0%; 13 with anti-GPIIb/IIIa, 8 with anti-GPIb). Plasma autoantibodies were noted in 34 of 59 patients (57.6%; 21 with anti-GPIIb/IIIa, 11 with anti-GPIb, and 2 with both). Positive results were noted in 30 of 59 patients using the immunobead assay and in only 14 of 59 using the microtiter-well assay, suggesting that solubilization of the platelets prior to antibody addition, as in the microtiter-well assay, alters epitope stability. Of the 31 thrombocytopenic control patients studied, all gave negative results using both assays. We conclude that these clinically adaptable assays allow detection of autoantibodies in most patients with chronic ITP, confirming the presence of an autoimmune process.

Summary. Chronic immune thrombocytopenic purpura (ITP) is an autoimmune disease characterized by antibody-induced platelet destruction. To better define the role of antigen-specific assays in adult chronic ITP, we prospectively measured plateletassociated autoantibody against either glycoprotein (GP) IIb/ IIIa or GPIb/IX in 282 patients with chronic ITP and 289 patients with thrombocytopenia of other causes. We divided chronic ITP into four subgroups: presplenectomy, mild (platelet count >30 000 mL À1 requiring no therapy), presplenectomy, severe (platelet count <30 000 mL À1 requiring therapy but not splenectomy), postsplenectomy, remission (postsplenectomy partial or complete remission without further therapy) and postsplenectomy refractory (required therapy after splenectomy failure). Positive results: total ITP group, 55.4%; presplenectomy, mild, 31.1%; presplenectomy, severe, 42.6%; postsplenectomy, remission, 50.0%; and postsplenectomy, refractory, 87.8%. In addition, the degree of positivity increased with the severity of the patient's disease. The assay had a minimum specificity of 84.4% if clinical factors, consistent with immune thrombocytopenia, were not considered in patients with thrombocytopenia associated with other diseases. However, if clinical factors consistent with immune thrombocytopenia were considered and only patients with questionable immune thrombocytopenia and patients 'lost to follow-up' were included in the false-positive group the specificity was 93.1%. We conclude that the presence of immune thrombocytopenia is highly probable if the immunobead assay is positive and that antigenspecific assays are diagnostically useful in adult chronic ITP.

A disintegrin and metalloprotease (ADAM) 1 is a family of recently identified cell surface and secreted glycoproteins that possess both proteolytic and adhesive properties (1, 2). Prototypical members of this family are composed of a prodomain, metalloprotease, disintegrin-like, cysteine-rich, EGF-like, transmembrane, and cytoplasmic domains. The metalloprotease domain is homologous to the reprolysins, members of the metazincin superfamily that include the matrix metalloproteases, the astacins, and serralysins (3). Although the ADAMs may serve other proteolytic functions, their role in the ectodomain shedding of specific cell surface proteins has been well documented (4, 5). ADAMs 9 (meltrin-␥), 10 (kuzbian), and 17 (tissue necrosis factor-␣-converting enzyme) have been shown to function in ectodomain shedding of membrane-anchored heparin-binding EGF-like growth factor, amyloid precursor protein, and tissue necrosis factor-␣, respectively (6 -10). These ADAMs and other members of this protein family have also been implicated in the shedding of other proteins such as L-selectin, FasL, and VCAM-1 (11-13). In addition to the proteolytic functions mentioned, several ADAMs have been shown to interact with the integrin family of cell surface receptors. Integrins are a widely distributed superfamily of noncovalent heterodimeric glycoproteins that play a vital role in cellular adhesion, migration, and signal transduction (14, 15). The ADAM disintegrin-like domains are homologous to small non-enzymatic peptides isolated from the venom of snakes that function as antagonists of integrins (16,17). The direct interaction of the snake venom disintegrin peptides with integrins led to the hypothesis that the disintegrin-like domains of ADAMs may function as integrin ligands. The disintegrin-like domains of ADAMs 1-3 expressed on the surface of sperm interact with the integrin ␣ 6 ␤ 1 in association with CD9 and CD98 on the egg surface (18 -23). Recognition of ADAMs 2 and 3 by ␣ 6 ␤ 1 requires the residues DECD located within a region of the disintegrin domain, designated the disintegrin loop, that corresponds to an extended loop in the snake venom peptides that typically contains an RGD sequence required for integrin binding (21-23). Human ADAM 15 contains an RGD sequence within the disintegrin loop region and is recognized by the integrins ␣ v ␤ 3 and ␣ 5 ␤ 1 (24 -26). However, mouse ADAM 15 lacks the RGD sequence found in the human homologue. Both human and mouse ADAM 15 as well as ADAM 12 were shown to be recognized by the integrin ␣ 9 ␤ 1 via residues outside the disintegrin loop (27). Other ADAM disintegrin-like domains reported to bind integrins are ADAM 9, which is recognized by ␣ 6 ␤ 1 and ␣ v ␤ 5 (28, 29), and ADAM 23, which is recognized by ␣ v ␤ 3 (30). * This work was supported by National Institutes of Health Grant AI47314. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section...

The metalloprotease disintegrin cysteine-rich (MDC) proteins are a recently identified family of transmembrane proteins that function in proteolytic processing of cell surface molecules and in cell adhesion. Since lymphocytes must interact with a constantly changing environment, we hypothesized that lymphocytes would express unique MDC proteins. To identify MDC proteins expressed in human lymph node, a polymerase chain reaction-based strategy combined with degenerate oligonucleotide primers was employed. We report here MDC,1 or ADAMs, are a family of transmembrane glycoproteins with a unique domain structure (1, 2). The prototypical MDC protein comprises pro, metalloprotease, disintegrin, cysteine-rich, EGF, transmembrane, and cytoplasmic domains. The metalloprotease domain is homologous to the reprolysins, members of the zinc binding metzincin superfamily that also includes the matrix metalloproteases (MMPs), the astacins, and serralysins (3). The prodomain regulates the activity of the metalloprotease by blocking access to the zinc ion; removal of the prodomain or inactivation of a conserved cysteine thiol group results in a gain of proteolytic activity. The disintegrin domains of MDC proteins are homologous to small nonenzymatic peptides isolated from the venom of snakes. Snake venom disintegrin peptides interfere with platelet aggregation by inhibiting binding of fibrinogen to the integrin ␣ IIb ␤ 3 (4). The functions previously attributed to the individual domains suggests roles for MDC proteins in cell adhesion and the proteolysis of extracellular proteins.The first mammalian MDC proteins identified were the fertilins (5). These proteins function in the attachment and fusion of the sperm and egg during fertilization. A similar cell fusion function for the MDC protein meltrin-␣ (MDC-12) was shown for the formation of multinucleated myotubes (6). These MDC proteins have, in addition to the domains described above, a fusion peptide-like sequence not found in all members of the MDC protein family. However, the interaction of the fertilin disintegrin domain on the sperm surface with the integrin ␣ 6 ␤ 1 on the surface of the egg is also required for membrane adhesion and fusion (7-9). Additional evidence for the functionality of the disintegrin domain comes from studies using recombinantly expressed metargidin (MDC-15) disintegrin domain, which was shown to specifically interact with the integrin

Chronic immune thrombocytopenic purpura (ITP) is an autoimmune disorder due to antiplatelet autoantibodies, many of which are directed against platelet membrane glycoprotein (GP) IIb-IIIa or GPIb-IX. In a recent study, we described plasma autoantibodies from 13 selected ITP patients, which required the presence of the putative GPIIIa cytoplasmic region for antibody binding. Since this region may not be available for antibody binding under physiologic conditions, we evaluated the frequency of binding to this or other regions of GPIIb- IIIa by platelet-associated and plasma autoantibody from a group of chronic ITP patients. We studied platelet-associated autoantibodies in 27 patients and plasma antibodies in 21 patients; in 15 patients, both were studied. To determine if autoantibodies were directed to the cytoplasmic portion of GPIIIa or to another portion of the GPIIb-IIIa molecule, antibody eluted from patient platelets or plasma antibody was tested in an antigen capture assay for binding to GPIIb-IIIa obtained from Chinese hamster ovary (CHO) cells transfected with GPIIb and either intact GPIIIa or GPIIIa lacking the carboxy terminal 35 residues. Of the 21 plasma autoantibodies tested, 13 bound primarily to the carboxy terminus of GPIIIa and eight to other epitopes. Conversely, all 26 platelet-associated autoantibodies, including eight of the 13 with anti-carboxy terminus antibodies, bound to epitopes in other regions of GPIIb-IIIa. Comparison of the degree of antibody adsorption by intact or lysed platelets indicated that epitopes on the c-terminal region of GPIIIa are relatively inaccessible on the surface of intact washed platelets when compared with other epitopes. We conclude that the importance of plasma autoantibodies in chronic ITP patients should be interpreted cautiously, since their specificity may differ from that of antibodies bound to the platelet. Whether antibodies against the c- terminus of GPIIIa are of pathogenetic importance remains to be determined, although patients with these antibodies have particularly severe disease.

The autoimmune nature of chronic immune thrombocytopenic purpura (ITP) in adults is widely accepted. In contrast, the pathogenetic mechanism in acute and chronic ITP in children is not known. In this report, we studied 39 children with destructive thrombocytopenia, 15 patients with acute ITP and 24 patients with chronic ITP. Platelet autoantibodies to platelet glycoprotein IIb/IIIa were detected in 14 of 24 patients (58.3%) in the chronic ITP group and in four of 15 (26.7%) with acute ITP. Binding ratios (+/- SD) of positive patients were significantly greater (P = .01) in chronic ITP (8.0 +/- 9.1) when compared with those of acute ITP where the binding ratios were only slightly above the normal range (1.9 +/- 0.4). The results show that autoantibodies against platelet glycoproteins are present in the majority of children with chronic ITP confirming the autoimmune nature of this disorder. The minimal elevation seen in the positive children with acute ITP suggests a different pathogenetic mechanism. These data suggest that this approach may be useful in differentiating acute from chronic ITP patients.

Although increased platelet destruction and elevated platelet-associated IgG have been shown in patients with lymphomas and various autoimmune diseases, such as systemic lupus erythematosus (SLE), there have been few studies evaluating autoantibodies against platelet-specific antigens. We evaluated 24 patients retrospectively with disease-related thrombocytopenia (12 with lymphoproliferative diseases and 12 with various autoimmune disorders) using a recently reported antigen-specific assay. Autoantibodies against platelet GPIIb/IIIa or GPIb/IX were noted in 15 of the 24 patients (10 of 12 with autoimmune disease and five of 12 with lymphoproliferative disorders). Platelet-associated autoantibodies were present in 60% and plasma autoantibodies in 33%. Anti-GPIIb/IIIa autoantibodies were much more common than those against GPIb/IX. In one patient each with thrombocytopenia and either SLE or myasthenia gravis, absorption of plasma with platelets completely removed the anti-GPIIb/IIIa autoantibodies, but did not affect the level of anti-cochlear autoantibody involved with immune-mediated hearing loss in the SLE patient or the anti-acetylcholine receptor autoantibody in the myasthenic patient. These findings show that, in some cases of disease-related immune thrombocytopenia, autoantibodies against GPIIb/IIIa or GPIb/IX can be detected similar to those seen in chronic ITP. As shown in two patients with multiple autoimmune manifestations, the various autoantibodies have diverse specificities and do not crossreact.

ADAMs (a disintegrin and metalloprotease) are a family of proteins that possess functional adhesive and proteolytic domains. ADAM 28 (MDC-L) is expressed by human lymphocytes and contains a disintegrin-like domain that serves as a ligand for the leukocyte integrin, alpha4beta1. To elucidate which residues comprise the alpha4beta1 binding site in the ADAM 28 disintegrin domain, a charge-to-alanine mutagenesis strategy was utilized. Each alanine substitution mutant was evaluated and compared to the native sequence for its ability to support cell adhesion of the T-lymphoma cell line, Jurkat. This approach identified ADAM 28 residues Lys(437), Lys(442), Lys(455), Lys(459), Lys(460), Lys(469), and Glu(476) as being essential for alpha4beta1-dependent cell adhesion. The epitope for a function-blocking monoclonal antibody, Dis 1-1, was localized to the N-terminal end of the ADAM 28 disintegrin domain using these same charge-to-alanine mutants. Three distinct molecular models based upon the known structures of snake venom disintegrins suggested that residues contributing to alpha4beta1 recognition are aligned on one face of the domain. This study demonstrates that residues located outside of the disintegrin loop participate in integrin recognition of mammalian disintegrins.