A reverse immuno-glycoproteomic approach for the discovery of new immunotherapeutic targets

A major challenge for the advancement of immunotherapeutic treatment of hematologic neoplastic conditions remains the identification of selective and fully targetable cell surface antigens. Transcriptomic and proteomic data, supported by computational analyses and propelled by experimentation at the single-cell level, are gradually becoming the mainstream approach to unveil novel immunotherapeutic targets. However, most of the discovered targets fail to pass the clinical, or even the preclinical, experimental validation because of not being sufficiently cancer-specific and/or because of the lack of (or failure to produce) suitable immunological agents for their effective targeting. Meanwhile, it is becoming increasingly clear that aberrant post-translational modifications of integral and cell surface-associated/exposed proteins of neoplastic lymphocytes may delineate the outermost cancer-selective antigens for antibody-based therapeutic approaches. Bearing this in mind, the present study has explored the potential to adopt an alternative drug discovery strategy having as a starting point unique monoclonal antibodies (mAbs) raised against highly glycosylated molecules of the cancer cell membrane, followed by determination of the validity of the recognized antigen(s) as putative immunotherapeutic targets. The strategy that we have denominated “reversed immuno-glycoproteomics” was assayed using a follicular lymphoma cell line as model for the initial production of a panel of unique mAbs, which were generated through the conventional murine hybridoma technology. In parallel, a previously available battery of mAbs, originally raised the human cartilage proteoglycan aggrecan, was investigated to confirm a previously proposed cross-recognition of highly glycanated lymphoma cell-associated molecules by a set of similar mAbs. Immunization of mice with complex mixtures of enriched, heavily glycosylated proteins derived from the lymphoma cell line Sci-1 yielded a total of 25 hybridoma clones that were selected for their reactivity against Sci-1 cell extracts. This panel of mAbs was initially assayed against a panel of 98 lymphoma and leukemia cell lines by flow cytometry to select the ones reacting with native, surface-exposed antigens. This screen identified a subset of 25 antibodies which were differentially tested on healthy peripheral blood mononucleated cells and immortalized lymphocytes to determine their cancer cell-specificity. Four lead mAbs were derived from these tests and these antibodies were surveyed by flow cytometry, immunocytochemistry and immunoblotting against a panel of >100 cell lines representing all major lymphoma and leukemia subtypes, such as to categorize and prioritize the pathologies of potentially highest clinical interest. A parallel flow cytometry-based screening of 17 mAbs originally raised against cartilage aggrecan confirmed the previously proposed expression on lymphomas of highly glycosylated surface antigens. A first effort to disclose the antigens recognized by mAbs raised against plasma membrane-associated macro-complexes was pursued by conventional immunoprecipitation (IP)/shot-gun, label-free mass spectrometry and screening of protein arrays spotted with >20,000 recombinant human proteins. While refinement of the sample preparation and IP procedures are still needed to yield trustable and reproducible information on the nature of the recognized antigen, surprisingly, protein array screening identified as a primary candidate protein believed to be transiently exposed on the outer cancer cell surface. Treatment of model cell lines of diverse pathological origin with the four lead mAbs was found to induce cell aggregation, interfere with proliferation and induce regulated cell death, in an antibody versus cell phenotype-specific manner, suggesting that the recognized surface antigens are functionally linked to cell survival pathways. Accordingly, combined treatment of model cells with low doses of doxorubicin and the most effective pro-cytotoxic antibody strongly accentuated the cell death-inducing effect exerted by each of the agents alone. While a full structural-functional characterization of the antigens recognized by the unique antibodies investigated in this study awaits to be completed, several of these immunological agents seem to possess the proper attributes to further explored preclinically as putative immunotherapeutic drugs