Steven Rust

Combining phenotypic and proteomic approaches to identify membrane targets in a ‘triple negative’ breast cancer cell type

BackgroundThe continued discovery of therapeutic antibodies, which address unmet medical needs, requires the continued discovery of tractable antibody targets. Multiple protein-level target discovery approaches are available and these can be used in combination to extensively survey relevant cell membranomes. In this study, the MDA-MB-231 cell line was selected for membranome survey as it is a ‘triple negative’ breast cancer cell line, which represents a cancer subtype that is aggressive and has few treatment options.MethodsThe MDA-MB-231 breast carcinoma cell line was used to explore three membranome target discovery approaches, which were used in parallel to cross-validate the significance of identified antigens. A proteomic approach, which used membrane protein enrichment followed by protein identification by mass spectrometry, was used alongside two phenotypic antibody screening approaches. The first phenotypic screening approach was based on hybridoma technology and the second was based on phage display technology. Antibodies isolated by the phenotypic approaches were tested for cell specificity as well as internalisation and the targets identified were compared to each other as well as those identified by the proteomic approach. An anti-CD73 antibody derived from the phage display-based phenotypic approach was tested for binding to other ‘triple negative’ breast cancer cell lines and tested for tumour growth inhibitory activity in a MDA-MB-231 xenograft model.ResultsAll of the approaches identified multiple cell surface markers, including integrins, CD44, EGFR, CD71, galectin-3, CD73 and BCAM, some of which had been previously confirmed as being tractable to antibody therapy. In total, 40 cell surface markers were identified for further study. In addition to cell surface marker identification, the phenotypic antibody screening approaches provided reagent antibodies for target validation studies. This is illustrated using the anti-CD73 antibody, which bound other ‘triple negative’ breast cancer cell lines and produced significant tumour growth inhibitory activity in a MDA-MB-231 xenograft model.ConclusionsThis study has demonstrated that multiple methods are required to successfully analyse the membranome of a desired cell type. It has also successfully demonstrated that phenotypic antibody screening provides a mechanism for rapidly discovering and evaluating antibody tractable targets, which can significantly accelerate the therapeutic discovery process.

Identification of anti-tumour biologics using primary tumour models, 3-D phenotypic screening and image-based multi-parametric profiling.

BackgroundMonolayer cultures of immortalised cell lines are a popular screening tool for novel anti-cancer therapeutics, but these methods can be a poor surrogate for disease states, and there is a need for drug screening platforms which are more predictive of clinical outcome. In this study, we describe a phenotypic antibody screen using three-dimensional cultures of primary cells, and image-based multi-parametric profiling in PC-3 cells, to identify anti-cancer biologics against new therapeutic targets.MethodsScFv Antibodies and designed ankyrin repeat proteins (DARPins) were isolated using phage display selections against primary non-small cell lung carcinoma cells. The selected molecules were screened for anti-proliferative and pro-apoptotic activity against primary cells grown in three-dimensional culture, and in an ultra-high content screen on a 3-D cultured cell line using multi-parametric profiling to detect treatment-induced phenotypic changes. The targets of molecules of interest were identified using a cell-surface membrane protein array. An anti-CUB domain containing protein 1 (CDCP1) antibody was tested for tumour growth inhibition in a patient-derived xenograft model, generated from a stage-IV non-small cell lung carcinoma, with and without cisplatin.ResultsTwo primary non-small cell lung carcinoma cell models were established for antibody isolation and primary screening in anti-proliferative and apoptosis assays. These assays identified multiple antibodies demonstrating activity in specific culture formats. A subset of the DARPins was profiled in an ultra-high content multi-parametric screen, where 300 morphological features were measured per sample. Machine learning was used to select features to classify treatment responses, then antibodies were characterised based on the phenotypes that they induced. This method co-classified several DARPins that targeted CDCP1 into two sets with different phenotypes. Finally, an anti-CDCP1 antibody significantly enhanced the efficacy of cisplatin in a patient-derived NSCLC xenograft model.ConclusionsPhenotypic profiling using complex 3-D cell cultures steers hit selection towards more relevant in vivo phenotypes, and may shed light on subtle mechanistic variations in drug candidates, enabling data-driven decisions for oncology target validation. CDCP1 was identified as a potential target for cisplatin combination therapy.

Improved drug-like properties of therapeutic proteins by directed evolution

Many natural human proteins have functional properties that make them useful as therapeutic drugs. However, not all these proteins are compatible with large-scale manufacturing processes or sufficiently stable to be stored for long periods prior to use. In this study, we focus on small four-helix bundle proteins and employ ribosome display in conjunction with three parallel selection pressures to favour the isolation of variant proteins with improved expression, solubility and stability. This in vitro evolution strategy was applied to two human proteins with known drug development issues, granulocyte colony-stimulating factor (G-CSF) and erythropoietin (EPO). In the case of G-CSF, the soluble expression levels in Escherichia coli were improved 1000-fold, while for EPO the level of aggregation in an accelerated shelf-life study was reduced from over 80% to undetectable levels. These results exemplify the general utility of our in vitro evolution strategy for improving the drug-like properties of therapeutic proteins.

Development of a Novel Ectonucleotidase Assay Suitable for High-Throughput Screening

5′-Ectonucleotidase (NT5E) catalyzes the conversion of adenosine monophosphate to adenosine and free phosphate. The role of this ectonucleotidase and its production of adenosine are linked with immune function, angiogenesis, and cancer. NT5E activity is typically assayed either by chromatographic quantification of substrates and products using high-performance liquid chromatography (HPLC) or by quantification of free phosphate using malachite green. These methods are not suitable for robust screening assays of NT5E activity. HPLC is not readily suitable for the rapid and efficient assay of multiple samples and malachite green is highly sensitive to the phosphate-containing buffers common in various media and sample buffers. Here the development and validation of a novel high-throughput ectonucleotidase screening assay are described, which makes use of a luciferase-based assay reagent, the Promega CellTiter-Glo kit, to measure the catabolism of AMP by NT5E. This multiwell plate-based assay facilitates the screening of potential ectonucleotidase antagonists and is unaffected by the presence of contaminating phosphate molecules present in screening samples.

Abstract 4218: Phenotypic selection screening reveals cancer stem cell therapeutic targets

Cancer stem cells (CSCs) represent an important subpopulation of tumor cells that are untouched and perhaps even expanded after chemotherapy and radiation. This population has been shown in murine models of cancer to be solely responsible for the regrowth of tumors following chemotherapy, underscoring the importance of therapeutically targeting this population. Therefore, we sought to find novel targets against CSCs using a phenotypic selection screen. Using a designed ankyrin repeat (DARpin) library, we panned cells disaggregated from tumor spheres from three patient-derived xenograft models of late stage pancreatic cancer to enrich for DARPins that bind CSCs. Once released from the cells, the DARPins were further analyzed for binding to both normal, pancreatic cancer total tumor cells and pancreatic CSCs in order to choose targets that showed either pancreatic cancer-specific or pancreatic CSC-specific binding. These DARPins were then tested in a sphere formation assay, an EZH2-based high content imaging screen, and a standard proliferation assay. Those DARPins showing good cancer-specific binding profiles were also tested for internalization and killing using a saporin-conjugated secondary antibody. Using this approach, we identified 30 DARPins of interest for either direct inhibitory antibody approaches or for potential antibody drug conjugate or nanoparticle delivery approaches. Identification of the proteins bound by each of the DARPins was done using either immunopreciptation followed by liquid chromatography tandem-mass spectrometry or through Retrogenix9s cell expression microarray technology. Identified targets included known CSC antigens such as integrin alpha6 (CD49f) as well as other potentially novel targets. These results highlight the possibilities of finding therapeutic targets for CSCs and opens up an exciting possibility for novel therapies. Citation Format: Elaine M. Hurt, Matt Flynn, Suneetha Thomas, Lilian van Vlerken-Ysla, Alan Sandercock, Steven Rust, Minter Ralph, Robert Hollingsworth. Phenotypic selection screening reveals cancer stem cell therapeutic targets. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4218. doi:10.1158/1538-7445.AM2015-4218

Abstract 4328: Phenotypic selection for identification of functional antibodies and development of high throughput screening assays.

Discovery of therapeutic antibodies requires the continued identification of tractable targets, and we have employed a phenotypic screening strategy to this end. In this study, antibodies from phage display libraries were screened for their abilities to inhibit MDA-MB-231 (‘triple negative’) breast cancer cells. The screening cascade included tests for specific cell binding, antibody internalization, and cytotoxicity as an antibody-drug conjugate. Targets of these antibodies then were identified using immunoprecipitation and mass spectrometry, and confirmed by showing diminished antibody binding after siRNA-mediated knockdown of the putative target gene. One target was identified as NT5E (also known as CD73), a 5’-ectonucleotidase. NT5E catalyzes the conversion of adenosine monophosphate to adenosine, and its function has been linked with immunity, angiogenesis, and cancer. To screen for additional inhibitors of this target, a novel, high-throughput ectonucleotidase assay was developed. NT5E activity is typically assayed either by high performance liquid chromatography or by quantification of free phosphate using malachite green, and neither of these methods is suitable for high-throughput screening. One inhibitory antibody was found to significantly slow growth of MDA-MB-231 xenograft tumors in mice. In summary, our phenotypic screening approach provides a mechanism for rapidly discovering and evaluating new antibody targets and is being used to accelerate the discovery of new cancer drugs. Citation Format: Carl Hay, Steven Rust, Erin Sult, Lori Clarke, Kim Rosenthal, Sandrine Guillard, David Lowne, Matt Flynn, Lutz Jermutus, Ralph Minter, Robert Hollingsworth, Kris Sachsenmeier. Phenotypic selection for identification of functional antibodies and development of high throughput screening assays. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4328. doi:10.1158/1538-7445.AM2013-4328

Abstract 974: New target discovery for oncology utilizing phenotypic selection and primary NSCLC patient tumors

Lung cancer is the most common type of cancer, with approximately 222,000 new cases expected each year in the United States. Approximately 72% of the individuals diagnosed with lung cancer will fall victim to the disease. Non-small cell lung cancer (NSCLC) is the most common type of lung cancer, but current treatments will not cure the disease for most patients. In an effort to identify novel therapeutics to address this unmet need, a phenotypic screening campaign was initiated. This type of target identification approach allows the discovery of novel drug targets by selection of antibodies based on their ability to elicit a specific desired phenotype. Once antibodies with the desired function are isolated, the targets to which the “hit” antibodies bind can be identified using a variety of approaches including IP-mass spec. Although phenotypic screens have been performed previously, the current effort has important advantages over those described previously. Specifically, this included the use of several unique screening assays run in parallel to allow the evaluation of multiple phenotypes simultaneously. Also unique was the use of primary material from NSCLC patients for both selection and functional screening. Specifically, we demonstrate the use of primary tumor cells in high throughput screening assays. We show the utility of incorporating novel screening assays including anoikis and spheroids with multiple endpoint readouts. Finally we provide data describing a series of “hit” antibodies isolated in the screen which had a variety of activities across the screening assays. These studies show the utility of function-first screening in the identification of unique potential therapeutic targets for oncology. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 974. doi:1538-7445.AM2012-974