Rational design, analysis, and potential utility of GM-CSF antagonists.

Granulocyte-macrophage colony-stimulating factor (GM-CSF) is an important cytokine involved in many immune and inflammatory processes and is believed to act in the early stages of immune responses. GM-CSF stimulates antigen-presenting cells, enhancing antigen presentation and inducing macrophage tumoricidal activity. GM-CSF binds to specific cellular receptors that are potential targets for pharmacological design. Rational design of small-molecule mimics is an important approach to pharmacophore design. One of the strategies in the development of small-molecular mimics of larger polypeptyde ligands is analysis of alternative ligands that bind the same site as does the native ligand. Molecular studies of GM-CSF-receptor interactions have led to the development of interaction site analogs and the development of an "anti-anti-GM-CSF" recombinant antibody (rAb) analog of a site on GM-CSF important for biological activity and receptor binding. This rAb and a peptide derived from the rAb first complementarity determining region (CDR) sequence bind to a monoclonal anti-GM-CSF antibody that mimics the GM-CSFR alpha chain, compete with GM-CSF for binding to GM-CSF receptor alpha chain (GM-CSFR alpha), and are specific biological antagonists. Molecular modeling of the rAb suggests structural similarity with a site previously implicated in GM-CSF binding to the GM-CSFR alpha. Two cyclic peptides, 1785 and 1786, also were developed on the basis of structural analysis of the GM-CSF region mimicked by anti-anti-GM-CSF recombinant antibody (rAb) 23.2. These peptides were designed to mimic structurally the positions of specific residues on the B and C helicies of human GM-CSF implicated in receptor binding and bioactivity. Both 1785 and 1786 were recognized specifically by polyclonal anti-GM-CSF antibody. 1786 also competitively inhibited binding of GM-CSF to the GM-CSF receptor and demonstrated antagonist bioactivity, as shown by its reversal of GM-CSF's ability to inhibit apoptosis of the GM-CSF-dependent cell line MO7E. These studies support the role of residues on the GM-CSF B and C helicies in receptor binding and bioactivity and suggest strategies for mimicking binding sites on four-helix bundle proteins with cyclic peptides.

Recombinant antibodies in bioactive peptide design.

Granulocyte-macrophage colony-stimulating factor (GM-CSF) is important in many immune and inflammatory processes. GM-CSF binds to specific cellular receptors which belong to a recently described supergene family. These receptors are potential targets for pharmacologic design, and such design depends on a molecular understanding of ligand-receptor interactions. One approach to dissecting out critical intermolecular interactions is to develop analogs of specific interaction sites of potential importance. Monoclonal antibodies have been employed for these purposes in prior studies. Here we present application of recombinant antibody technology to the development of analogs of a site on GM-CSF bound by a neutralizing anti-GM-CSF monoclonal antibody. Polyclonal antisera with high titer neutralizing activity against human GM-CSF were developed in BALB/c mice. Purified immunoglobulins were prepared and used to immunize syngeneic mice. Anti-anti-GM-CSF was developed which demonstrated biological antagonist activity against GM-CSF-dependent cellular proliferation. RNA was extracted from spleen cells of mice with biologically active anti-anti-GM-CSF, cDNA synthesized, and polymerase chain reaction performed with primers specific for murine kappa light chain V regions. Polymerase chain reaction products were cloned into the pDABL vector and an expression library developed. This was screened with anti-GM-CSF neutralizing mAb 126.213, and several binding clones isolated. One clone (23.2) which inhibited 126.213 binding to GM-CSF was sequenced revealing a murine kappa light chain of subgroup III. Comparison of the 23.2 sequence with the human GM-CSF sequence revealed only weak sequence similarity of specific complementarity determining regions (CDRs) with human GM-CSF. Structural analysis revealed potential mimicry of specific amino acids in the CDR I, CDR II and FR3 regions of 23.2 with residues on the B and C helices of GM-CSF. A synthetic peptide analog of the CDR I was bound by 126.213, specifically antagonized GM-CSF binding to cells and blocked GM-CSF bioactivity. These studies indicate the feasibility of using recombinant antibody libraries as sources of interaction site analogs.

Development of GM-CSF antagonist peptides.

Granulocyte/macrophage colony stimulating factor (GM-CSF) is both a hematopoietic growth factor and a cytokine implicated in inflammatory disease. The development of GM-CSF antagonist peptides corresponding to the GM-CSF native sequence should allow their modification into higher affinity analogs, but this is hampered by the low affinity of linear peptides. To adequately evaluate such low affinity peptides, the use of several independent assays should allow specific versus nonspecific inhibitors to be distinguished. In this study, inhibition of GM-CSF-dependent cell growth, inhibition of GM-CSF binding and immunologic cross-reactivity between GM-CSF-derived peptides and native protein by neutralizing antibodies have been used to evaluate peptide analogs with potential bioactivity. The GM-CSF sequence was divided into 6 peptides ranging in size from 15-24 amino acids. Antisera were raised to these peptides in mice and assayed for immunologic cross-reactivity. 4/6 anti-peptide antisera bound GM-CSF on ELISA and 3/6 on immunoprecipitation. Antisera to two of the peptides (corresponding to residues 17-31 and 96-112) inhibited GM-CSF-dependent cellular proliferation in two cell lines, with one peptide derived from residues 17-31 demonstrating inhibition of GM-CSF binding and direct biological inhibitory activity. A peptide that did not elicit native GM-CSF reactive antibodies, corresponding to residues 54-78, was recognized by two neutralizing monoclonal antibodies. It exhibited inhibition of GM-CSF binding and direct biological antagonist activity. These studies implicate two sites in mediating GM-CSF biological activity, and indicate that biological antagonists can be developed based on these sites.

Molecular cloning of a soluble form of the granulocyte-macrophage colony-stimulating factor receptor alpha chain from a myelomonocytic cell line. Expression, biologic activity, and preliminary analysis of transcript distribution.

OBJECTIVE: To analyze the molecular and functional characteristics of a soluble form of the granulocyte-macrophage colony-stimulating factor receptor alpha chain (sGM-CSFR alpha), and analyze transcript expression in immune cells and the cellular constituents of rheumatoid arthritis synovial tissue. METHODS: We amplified, cloned, and expressed the sGM-CSFR alpha and transmembrane form of the receptor (tmGM-CSFR alpha) from complementary DNA derived from a human myelomonocytic cell line. Competitive polymerase chain reaction assays were developed to determine the absolute and relative amounts of tmGM-CSFR alpha versus sGM-CSFR alpha message synthesized by various cell lines and tissues. RESULTS: sGM-CSFR alpha transcripts were detected in bone marrow, monocyte/macrophages (cultured in GM-CSF), rheumatoid synovial tissue, and rheumatoid synovial tissue T cell lines, and represented the predominant transcript in synovial fibroblasts and osteoarthritis synovial tissue. Levels of expression in monocyte/macrophages and some synovial fibroblast and T cell lines approached those seen in transfected cell lines producing functional sGM-CSFR alpha. CONCLUSION: sGM-CSFR alpha represents a functional antagonist of GM-CSF activity in vitro. Expression of sGM-CSFR alpha in bone marrow, rheumatoid synovial tissue T cells, and synovial fibroblasts suggests an important role in vivo, both in regulating myelopoiesis and in modulating the immune response.

Tyrosine kinase signal transduction in rheumatoid synovitis.

Explants of synovial cells in rheumatoid arthritis display a transformed phenotype with focus formation and anchorage-independent growth. Many of the cytokines that activate these fibroblasts mediate their action through tyrosine kinase growth factor receptors. Mechanisms of signal transduction via such tyrosine kinases are therefore relevant to the pathogenesis of rheumatoid lesions. Data are presented using the neu oncogene product p185neu as a model system to explore signal transduction by receptor tyrosine kinases. Evidence is shown that increased tyrosine kinase activity in the oncogenic form of this protein may result from dimerization of the tyrosine kinase receptor. In the normal cellular counterpart of p185neu, dimerization appears to be mediated by the action of an as yet unidentified ligand. Dimerization also appears to be important in signal transduction mediated by epidermal growth factor, platelet-derived growth factor, and colony-stimulating factor 1. These cytokines also alter the phenotype of rheumatoid synovial fibroblasts to resemble transformed fibroblasts. Additionally, preliminary data that suggest increased tyrosine kinase activity in rheumatoid arthritis synovia compared with osteoarthritis synovia are presented. Molecular characterization of tyrosine kinase receptors will be an important direction for future studies of the pathogenesis of rheumatoid disease.