Therefore, we chose to analyze the samples by 2D-DIGE in an effort to resolve sCTLA-4 from contaminating abundant proteins. gene, and that the CTLA-4 protein is not part of this molecule. These results may explain why the relationship of sCTLA-4 to immune system activity has been difficult to elucidate. Background Alternate splicing of the CTLA-4 mRNA transcript can give rise to at least three mRNA species that encode different polypeptides [1]. The most well characterized of these is a type I transmembrane protein (CTLA-4-TM) expressed on activated T-lymphocytes [2,3]. CTLA-4-TM is a co-receptor for the B7.1 (CD80) and B7.2 (CD86) molecules expressed on antigen presenting cells [4,5]. CTLA-4-TM inhibits immune activity in multiple ways. It regulates signaling through the T-cell receptor [6,7], induces expression of immunoregulatory factors such as TGF- and ICAM-1 [8,9], alters the organization of the immunological synapse [10], increases tryptophan catabolism by antigen presenting cells [11,12] and AMG 487 binds B7.1 and B7.2 preventing activation of lymphocytes through the costimulatory lymphocyte receptor CD28 [13]. Another transcript of the CTLA-4 gene encodes a molecule lacking the transmembrane domain, AMG 487 thus producing a soluble CTLA-4 polypeptide referred to as sCTLA-4 [14,15]. Like CTLA-4-TM, sCTLA-4 appears to bind B7.1 and B7.2, and may AMG 487 have immunomodulatory properties [15-17]. RCBTB2 Finally, a variant transcript [18] has been identified in mouse (although not humans) that encodes a membrane-spanning molecule with an intact cytoplasmic tail, but lacks the extracellular domain. As such, the molecule does not bind the B7 family ligands [19] and has been referred to as ligand-independent CTLA-4 (liCTLA-4). The expression of these three CTLA-4 transcripts and their polypeptide products has been associated with immunoregulatory function, and differences in their expression have been associated with immune-mediated disease. For example, CTLA-4 knockout mice express none of the possible alternate transcripts and show a profound lymphoproliferative disorder with fatal multiorgan destruction [20,21]. Although it is commonly believed that the absence of the CTLA-4-TM transcript is solely responsible for the observed immunological disorder in CTLA-4-knockout mice, the role(s) of the other transcripts have not been studied as intensively. LiCTLA-4 may have immunoregulatory functions, as transfection of it into CTLA-4 deficient T cells partially corrects the tendency for hyperresonsiveness [19], and the liCTLA-4 transcript has been associated with the development of insulin dependent diabetes mellitus in the NOD mouse [18]. Finally, a variety of reports implicate a role for sCTLA-4 in human autoimmune disease. The CT60 single nucleotide polymorphism of the CTLA-4 gene has been associated with autoimmunity and with reduced levels of the sCTLA-4 transcript [18]. Various studies have demonstrated elevated levels of the sCTLA-4 protein in the blood of patients with a variety of immunologically mediated diseases including autoimmune thyroid disease [22], systemic lupus erythematosis [23,24], cutaneous systemic sclerosis AMG 487 [25], allergic asthma [26,27], and psoriasis vulgaris [28]. This apparent inverse relationship between levels of sCTLA-4 mRNA and circulating levels of the sCTLA-4 protein is not understood. Several years ago, we [22] and others [14] described immunoassays for the detection of sCTLA-4 in human plasma. Presumably, such material was the gene product of the sCTLA-4 transcript; however, this was never formally proven. In order to characterize sCTLA-4 in human blood, we performed biochemical analyses of blood-derived molecules that are recognized by multiple CTLA-4-specific antibodies. Our results suggest that the immunoreactive material in human blood is not the direct product of the sCTLA-4 alternate transcript and has several biochemical features of human immunoglobulin. In addition, CTLA-4 immunoreactive material from human plasma binds the B7.1 and B7.2 proteins, and may have immunomodulatory function. Methods Monoclonal Antibodies and fusion proteins The following monoclonal antibodies against CTLA-4 (CD152) were used in these studies: BNI3 (BD Pharmingen, SanDiego, CA), AS32 and AS33 (Antibody Solutions, Palo Alto, CA), are monoclonal antibodies that recognize extracellular epitopes in the CTLA-4 molecule. The MOPC-21C antibody (Sigma-Aldrich, St. Louis, MO) was used as a negative control. ELISA assays for CTLA-4 were done as described previously [25]. B7.1-Ig (CD-80) and B7.2-Ig (CD86) fusion proteins (R&D Systems, Minneapolis, MN) were biotinylated with the use of a commercially available kit (Pierce Immunochemical, Rockford, IL). The Muc18-Ig protein (Muc-Ig) fusion protein was produced by transfecting CHO cells with a commercially available plasmid (Novagen) encoding AMG 487 a Muc18-Ig fusion as described [15,29]. Purification of blood-derived sCTLA-4 Plasma samples used in these studies were obtained from humans undergoing.