Stimulated human T cells from healthy volunteers demonstrate attenuated early interleukin (IL)-2 receptor (R) signaling in the presence of daclizumab (Dac). Aiming to confirm that this ex-vivo effect of Dac is also observed in-vivo, we studied T cells from 3 kidney transplant recipients before and 2–3 weeks and 4–6 months after transplantation. We found by flow cytometry that T cells obtained pre-transplant and stimulated ex-vivo with phytohemeagglutinine upregulated the IL-2R α-(CD25) and β-(CD122) chains as expected. Moreover, exogenous IL-2 induced characteristic tyrosine phosphorylation events detectable by immunoblotting in these cells. However, T cells studied post-transplant neither exhibited CD25 or -122 upregulation nor IL-2-induced tyrosine phosphorylation events, indicating broad, persistent suppression of the IL-2R signaling machinery which thus appears largely inaccessible for Dac in actual transplant recipients. We therefore conclude that the clinical efficacy of this agent may depend on additional mechanisms in-vivo other than those identified ex-vivo.

1021 CD4 ligation (CD4L) with anti-CD4 antibodies (αCD4s), gp120 or IL-16 can down-modulate T lymphocyte activation and promote long-term immunological hyporesponsiveness, resulting in prolonged allograft survival in organ transplantation under αCD4. T cell depletion is not obligatory for these effects, and they are as of yet poorly understood. We have previously shown that a gp120-derivative can interfere with interleukin-2 receptor (IL-2R) signaling by attenuating IL-2-induced JAK3-tyrosine phosphorylation and subsequent formation of DNA-binding protein containing STAT5 (Transpl Proc, in press). Therefore, we have now examined whether αCD4 may have similar capabilities. Freshly isolated human T lymphocytes, after stimulation with PHA (2μg/ml) to express high-affinity IL-2R and a subsequent rest period of 1 to 2 days, were incubated with αCD4 16H5 (Lancet 338:702, 1991) at various concentrations and for different time periods. Preliminary 3H-incorporation studies revealed dose-dependent inhibition of T cell proliferation by the antibody. Subsequent electrophoretic mobility shift assays (EMSAs) to track the effect of 16H5 on IL-2(100U/ml, 20-30 min.)-induced formation of DNA-binding protein in human T cells again documented an antibody dose-dependent attenuation of such formation after 16H5-pretreatment for 3 hours. This effect was not detectable in control experiments with non-specific immunoglobulin in equal concentrations. Furthermore, we examined the capability of αCD4 to interfere with cell proliferation in CMO cells, a CD3−, CD4+ and CD25+ human lymphoid cell line transformed with HTLV-1 and thus exhibiting constitutive, IL-2-independent activation of the JAK3/STAT5 signaling cascade. Accordingly, ongoing CMO cell proliferation could not be augmented further by IL-2 in 3H-incorporation studies, but 16H5 dramatically decreased it. Lastly, we have shown in complementary EMSAs that CD4L also attenuates the expression of STAT5-containing DNA-binding protein in CMO cells. We therefore conclude that CD4L with αCD4s can directly interfere with IL-2R signal transduction. Our results provide a novel molecular mechanism of action underlying αCD4-induced attenuation of T cell activation, long-term immunonogical hyporesponsiveness and possibly transplantation tolerance.

666 Administration of non-depleting anti-CD4 antibody can lead to long-term allograft survival and possibly transplantation tolerance in rodents, but mechanisms underlying in vivo immunomodulation by CD4 ligands are incompletely understood. Using peptomer, a CD4-binding peptide multimer, we have previously identified attenuation of IL-2-induced nuclear translocation of STAT5 as a downstream effect of CD4 ligation (CD4L). To further characterize this interference of CD4L with IL-2 receptor (IL-2R) signaling, we examined proximal elements of IL-2-activated signal transduction cascades as potential molecular targets of such ligation. Purified primed human T cells were stimulated with IL-2 (100U/ml, 10-60 min.) alone or after pretreatment with peptomer (1 hr.). By flow cytometry, IL-2R α-chain expression, which is regulated by IL-2 via STAT5, was unaffected in the time frame studied. In anti-phosphotyrosine immunoblots, IL-2-induced tyrosine phosphorylation of IL-2R β-chain immunoprecipitates was also unaltered by preceding CD4L with peptomer. However, anti-phosphotyrosine immunoblots of whole cell lysates revealed that peptomer attenuates IL-2-induced tyrosine phosphorylation of 130 kD protein(s) indistinguishable from JAKs 1 and 3 on reprobing with antibodies against these kinases. As IL-2R-associated JAK3 is essential for STAT5 activation, we examined JAK3 immunoprecipitates from the cells and found that peptomer pretreatment indeed inhibits JAK3 tyrosine phosphorylation by IL-2. Cytoskeletal sequestration of another kinase, p56lck, which associates with CD4 and the IL-2R, also correlates with CD4L-induced immune modulation and is preventable by pre-incubation with cytochalasin D (CyD) to inhibit cytoskeletal actin polymerization. We thus asked if CyD could block the attenuation of IL-2R signaling by antecedent CD4L, and if p56lck was required for this effect. Nuclear extracts of human T cells treated as above and with or without CyD-pre-incubation (0.5μM, 2-3 hrs.) were analyzed in gel shift assays. CyD was found to protect IL-2-induced nuclear translocation of STAT5 from attenuation by CD4L. Identical results were obtained with a p56lck-deficient human lymphoid cell line. These studies identify JAK3 as a novel target of CD4 ligands and indicate that CD4L can interfere with IL-2R signaling via attenuation of IL-2-induced JAK3 tyrosine phosphorylation. Polymerized cytoskeletal actin, but not p56lck, appears to be required for this interference. Our observations delineate new molecular mechanisms underlying CD4L-induced immune modulation and thus provide fresh insights into the possible genesis of transplantation tolerance induced by therapies targeting CD4.

Bone and bone marrow are important sites of metastasis formation in breast cancer. Extracellular matrix proteins with attachment properties are generally believed to play a key role in tumorigenesis and metastasis formation. We have investigated whether mammary carcinoma cells (MDA-MB-231) can recognize constructs of the fairly bone-specific human bone sialoprotein, which encompass the RGD sequence (EPRGD-NYR). Exogenously added bone sialoprotein peptides with this amino acid sequence in their backbone structure, but not the more common fibronectin-derived GRGDS peptide, strongly inhibited breast cancer cell adhesion to extracellular bone matrix at micromolar concentrations. Most cyclic derivatives with the EPRGDNYR sequence were more effective inhibitors of tumor cell adhesion to bone than their linear equivalents. Furthermore, changes in the RGD-tripeptide of the backbone structure of the constructs, removal of the NYR flanking sequence, or a different tertiary cyclic structure significantly decreased their inhibitory potencies. In addition, the RGE-analogue EPRGENYR was capable of inhibiting breast cancer cell adhesion to bone, albeit to a lesser extent. We conclude therefore, that the inhibitory potency of the bone sialoprotein-derived peptides on breast cancer cell adhesion to bone is not solely due to a properly positioned RGD-motif alone but is also determined by its flanking regions, together with the tertiary structure of the EPRGDNYR peptide. Synthetic cyclic constructs with the EPRGDNYR sequence may, therefore, be potentially useful as antiadhesive agents for cancer cells to bone in vivo.

We describe a novel experimental system in mice for the study of ovarian autoimmune disease, a condition encountered in women with premature ovarian failure. The ovarian autoimmune disease is induced in B6AF1 mice by a 15-amino acid peptide (Cys-Ser-Asn-Ser-Ser-Ser-Ser-Gln-Phe-Gln-Ile-His-Gly-Pro-Arg) from mouse ZP3, the sperm-binding component of the zona pellucida that surrounds growing and mature oocytes. Whereas the peptide induces both T cell and antibody responses, adoptive transfer of CD4+ T cell lines derived from affected animals causes oophoritis without observable antibodies to the zona pellucida peptide. The primacy of the T cell response in the pathogenesis of disease is further substantiated by defining oophoritogenic peptides as small as eight amino acids (Asn-Ser-Ser-Ser-Ser-Gln-Phe-Gln) that do not elicit an antibody response to the full-length ZP3 peptide. The identification of a well characterized peptide as a causative agent of autoimmune oophoritis should facilitate understanding of the pathogenesis of this T cell-mediated autoimmune disease. Because the proteins of the zona pellucida are conserved among mammals (the mouse and human ZP3 proteins are 67% identical), this murine model may lead to better understanding of the pathogenesis of human autoimmune oophoritis.

The zona pellucida surrounding mouse oocytes is an extracellular matrix composed of three sulfated glycoproteins, ZP1, ZP2, and ZP3. It has been demonstrated that a monoclonal antibody to ZP3 injected into female mice inhibits fertilization by binding to the zona pellucida and blocking sperm penetration. A complementary DNA encoding ZP3 was randomly cleaved and 200- to 1000-base pair fragments were cloned into the expression vector lambda gt11. This epitope library was screened with the aforementioned contraceptive antibody, and the positive clones were used to map the seven-amino acid epitope recognized by the antibody. Female mice were immunized with a synthetic peptide containing this B cell epitope coupled to a carrier protein to provide helper T cell epitopes. The resultant circulating antibodies to ZP3 bound to the zona pellucida of immunized animals and produced long-lasting contraception. The lack of ovarian histopathology or cellular cytotoxicity among the immunized animals may be because of the absence of zona pellucida T cell epitopes in this vaccine.

A monoclonal anti-idiotope termed 87.92.6 mimics the neutralization/cell-attachment site of the reovirus type 3 hemagglutinin (HA3). The second complementarity determining regions of the VH and VL of 87.92.6 share sequence similarity with a determinant on the HA3. We have used synthetic peptides (termed VH, VL, and Reo peptides, respectively) to probe the immunologic significance of this sequence similarity. Antibodies specific for Reo peptide or VL peptide neutralized reovirus type 3 infectivity. Although Reo peptide was an effective immunogen by itself, free VL peptide or VH peptide were unable to elicit antibodies unless they were linked to each other (VH-VL peptide). Immunization with reo peptide, 87.92.6, or the HA3 elicited a specific lymphocyte proliferative response to VH peptide, indicating that VH peptide may bear an important TH determinant. As found previously for 87.92.6, VL peptide elicited a delayed-type hypersensitivity response specific for reovirus type 3. Reovirus type 3 specific cytolytic lymphocytes specifically lysed targets coated with VH-VL peptide, but not VH or VL peptide alone. These results suggest that immune cross-reactivity between an external Ag and an internal image antibody can be understood at the primary structural level. These observations may have important implications for understanding the development of autoantibodies, network interactions, and the regulation of immune responses.

Previous studies have identified an area of amino acid sequence similarity shared by the reovirus type 3 cell-attachment protein sigma 1 and an anti-idiotypic/antireceptor monoclonal antibody (mAb) 87.92.6 that mimics reovirus type 3 by attaching to the same cell-surface receptor. We found that synthetic peptides corresponding to this area of primary sequence similarity bind a neutralizing mAb 9BG5 against which the mAb 87.92.6 is directed. The synthetic peptides compete with mAb 87.92.6 and reovirus type 3 for binding by mAb 9BG5 and displace mAb 87.92.6 and reovirus type 3 from binding to the cell-surface reovirus type 3 receptor. Such observations show that the shared primary structure between reovirus type 3 sigma 1 polypeptide and antireceptor mAb 87.92.6 defines the oligopeptide neutralizing/cell-attachment epitope of reovirus type 3. Computer modeling of this epitope, by use of sequence similarities of known immunoglobulin hypervariable loop conformations, permits an examination of the rudimentary three-dimensional structure of this epitope.