Kevin Hestir

Discovery of a cytokine and its receptor by functional screening of the extracellular proteome.

To understand the system of secreted proteins and receptors involved in cell-cell signaling, we produced a comprehensive set of recombinant secreted proteins and the extracellular domains of transmembrane proteins, which constitute most of the protein components of the extracellular space. Each protein was tested in a suite of assays that measured metabolic, growth, or transcriptional responses in diverse cell types. The pattern of responses across assays was analyzed for the degree of functional selectivity of each protein. One of the highly selective proteins was a previously undescribed ligand, designated interleukin-34 (IL-34), which stimulates monocyte viability but does not affect responses in a wide spectrum of other assays. In a separate functional screen, we used a collection of extracellular domains of transmembrane proteins to discover the receptor for IL-34, which was a known cytokine receptor, colony-stimulating factor 1 (also called macrophage colony-stimulating factor) receptor. This systematic approach is thus useful for discovering new ligands and receptors and assessing the functional selectivity of extracellular regulatory proteins.

Deorphanization of the human leukocyte tyrosine kinase (LTK) receptor by a signaling screen of the extracellular proteome

Significance Secreted factors and their cell-surface receptors play important roles in the communication between cells in normal and pathological conditions. There are many transmembrane receptor-like proteins whose ligands have not been identified (also known as orphan receptors). Knowledge of the ligand should help in understanding the biological role of the receptor. We used a strategy of screening the extracellular proteome, one protein at a time, to identify ligands for such receptors. We discovered the ligands for the orphan receptor leukocyte tyrosine kinase. To our knowledge, this is the first case in which secreted factor ligands were identified for an orphan receptor with this technique. This approach is especially valuable when little is known about the ligand. There are many transmembrane receptor-like proteins whose ligands have not been identified. A strategy for finding ligands when little is known about their tissue source is to screen each extracellular protein individually expressed in an array format by using a sensitive functional readout. Taking this approach, we have screened a large collection (3,191 proteins) of extracellular proteins for their ability to activate signaling of an orphan receptor, leukocyte tyrosine kinase (LTK). Only two related secreted factors, FAM150A and FAM150B (family with sequence similarity 150 member A and member B), stimulated LTK phosphorylation. FAM150A binds LTK extracellular domain with high affinity (KD = 28 pM). FAM150A stimulates LTK phosphorylation in a ligand-dependent manner. This strategy provides an efficient approach for identifying functional ligands for other orphan receptors.

Controlling feature selection in random forests of decision trees using a genetic algorithm: classification of class I MHC peptides.

Feature selection is an important challenge in many classification problems, especially if the number of features greatly exceeds the number of examples available. We have developed a procedure--GenForest--which controls feature selection in random forests of decision trees by using a genetic algorithm. This approach was tested through our entry into the Comparative Evaluation of Prediction Algorithms 2006 (CoEPrA) competition (accessible online at: http://www.coepra.org). CoEPrA was a modeling competition organized to provide an objective testing for various classification and regression algorithms via the process of blind prediction. In the competition GenForest ranked 10/23, 5/16 and 9/16 on CoEPrA classification problems 1, 3 and 4, respectively, which involved the classification of type I MHC nonapeptides i.e. peptides containing nine amino acids. These problems each involved the classification of different sets of nonapeptides. Associated with each amino acid was a set of 643 features for a total of 5787 features per peptide. The method, its application to the CoEPrA datasets, and its performance in the competition are described.

Abstract 5449: FP-1039/GSK3052230, an FGF ligand trap, enhances VEGF antagonist therapy in preclinical models of RCC and HCC

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA An increasing body of evidence has implicated FGF2 as one of the drivers of resistance to various inhibitors of VEGF-mediated angiogenesis. This resistance may play a role as a key limitation to the efficacy of therapies targeted at VEGF and its receptors. We investigated the potential for FP-1039/GSK3052230, a ligand trap that sequesters FGFs and inhibits their signaling, to enhance the activity of VEGF antagonist therapies in certain preclinical models of renal cell (RCC) and hepatocellular (HCC) carcinomas. First, we examined whether FP-1039/GSK3052230 has single agent efficacy against human RCC and HCC xenografts that express relatively high levels of FGF2, a profile that would mimic FGF2-driven resistance to VEGF therapy. We determined that this expression profile represents 34% of clear cell RCC (ccRCC) and 31% of HCC patients, based on the cancer genome atlas (TCGA) data. Human ccRCC xenografts with high FGF2 expression and low VEGFA expression demonstrated a significant inhibition in tumor growth when treated with FP-1039/GSK3052230 alone (TGI: 39-81%). In addition, we show that the high FGF2 expression profile is similarly predictive for the anti-tumor response of a human HCC model to single-agent FP-1039/GSK3052230 (TGI: 31-55%). In contrast, RCC models with low FGF2 expression, representing 66% of all ccRCC in the TCGA, are relatively insensitive to FP-1039/GSK3052230 as a single-agent. However, combination therapy of FP-1039/GSK3052230 with pazopanib in these tumors is significantly more effective than either agent alone. FP-1039/GSK3052230 not only slows tumor growth, but can induce ∼25% tumor regression when administered to mice bearing ccRCC xenografts that have become resistant to pazopanib. Together, our data demonstrate that FP-1039/GSK3052230 may be an effective therapy against RCC and HCC, both as a single agent in disease driven by FGF2 and in combination with VEGF antagonist therapies that represent the current standards of care for advanced disease. Citation Format: David I. Bellovin, Servando Palencia, Kevin Hestir, Ernestine Lee, M. Phillip DeYoung, Thomas Brennan, Gerrit Los, Kevin Baker. FP-1039/GSK3052230, an FGF ligand trap, enhances VEGF antagonist therapy in preclinical models of RCC and HCC. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 5449. doi:10.1158/1538-7445.AM2014-5449

Identification of novel immune regulators of tumor growth using RIPPS screening in vivo

Identification of novel targets in cancer immunotherapy is needed to address the significant number of patients that either do not respond to current therapies or encounter unacceptable toxicities. The discovery of such targets, including novel checkpoint regulators and the counter-receptors for previously “orphan” checkpoints, has been limited by a lack of a comprehensive collection of proteins suitable for functional screening and methods for assessing their function in high-throughput. We have generated a comprehensive library of substantially all human extracellular proteins, encompassing nearly every target for protein therapeutics. Our library contains more than 5700 proteins, including secreted protein ligands and the extracellular domains of membrane-bound receptors in soluble forms. The library proteins represent therapeutic targets and in some cases may act as therapeutics themselves. A portion of this library we call the immunome contains ~500 proteins with structural features characteristic of immune-activators and checkpoints that we selected. RIPPSSM technology is a robust method whereby FivePrimes library of soluble secreted proteins can be tested in vivo in virtually any disease model. Each cDNA representing a unique protein is administered to a cohort of mice and results in high circulating levels of the encoded protein. RIPPSSM also allows us to rapidly confirm activity identified by other in vitro screening approaches. Here, we have exploited RIPPSSM technology to screen for new immuno-oncology therapeutics and targets for therapeutic development. As positive controls we performed RIPPSSM on CT-26 tumor-bearing mice using known agonists and antagonists of the immune response, which resulted in decreased or increased tumor growth, respectively. Subsequently, we screened over 250 immunome proteins by RIPPSSM in the CT-26 tumor model and have identified proteins that enhance and inhibit tumor growth and display changes in TIL (tumor-infiltrating lymphocyte) profiles. In addition, the immunome was screened in vitro for the ability to modulate anti-CD3-stimulated human T cell proliferation[1]. This screen identified a number of known checkpoint regulators, which validated the assay, and identified numerous novel co-inhibitory proteins – the majority of which have no published link to immune cell regulation. Proteins identified in the in vitro screens have been validated using RIPPSSM and other in vivo methods. These data demonstrate the power of our discovery platform to discover and validate novel therapeutic targets and protein therapeutics for immuno-oncology.

Identification of a novel immune checkpoint regulator and potential therapeutic antibody target in oncology

Antibody blockade of immune checkpoint regulators such as PD-1 and CTLA4 has been shown to be an effective cancer treatment strategy; however, a large percentage of patients still do not respond to existing therapies. Discovery of additional immune checkpoints and development of antibody therapeutics against them are likely critical to address this unmet patient need. We generated a comprehensive library of essentially all human extracellular proteins and screened proteins in this library in vitro and in vivo for the ability to modulate immune responses or tumor growth. As a result of these screens, we identified a number of novel immune checkpoints1. One such protein, referred to herein as Novel Checkpoint 1, was originally identified through its inhibitory activity on anti-CD3-stimulated human T cell proliferation. To confirm its activity as an immune checkpoint, we demonstrated that the native protein expressed on an antigen-presenting cell line could inhibit antigen-stimulated CD8+ T cell activation. Furthermore, blocking antibodies against this protein relieved the inhibition. This inhibitory activity translated to a murine system, as the mouse ortholog and blocking antibodies behaved similarly in murine T cell activation assays. Overexpression of the protein in mouse syngeneic tumor models resulted in increased tumor growth, consistent with inhibition of anti-tumor immune responses. Novel Checkpoint 1 is expressed primarily on activated and regulatory T cells in humans and mice – an expression profile similar to those of PD-1 and CTLA4. Additionally, it is expressed on 40-70% of tumor-infiltrating T cells while only on 10-15% of circulating T cells from those tumor-bearing mice. We are currently evaluating the anti-tumor activity of blocking antibodies in mouse tumor models, either alone or in combination with other checkpoint blocking antibodies. Taken together, we believe that these data demonstrate that this newly discovered protein may act as a checkpoint regulator in tumors and that blocking antibodies against it have potential as a novel cancer immunotherapeutic.