Abstract
Numerous impediments exist in the identification and development of new small molecules for malignancy. One challenge has been the limitations of current phenotype-based and target-based small molecule screening approaches. Although phenotype-based assays have identified the majority of cytotoxic chemotherapy agents used today, they are generally limited to the measurement of simple state changes and require unique assay customization. In addition, the identification of the protein target of emerging hits is often an impasse to subsequent drug development. In contrast, target-based screening is limited by the requirement for target knowledge a priori and for a pharmacologically tractable target. Moreover, the average time to develop a new drug is over 10 years – a long road from the laboratory to the clinic. In order to begin to address these challenges, we developed an alternative approach to small molecule screening: gene expression-based high-throughput screening (GE-HTS). GE-HTS, based on the premise that gene expression signatures can be used for high-throughput screening, relies on ligation-mediated amplification (LMA) of up to 500 transcripts and the quantification of transcript by a fluorescent bead-based technology. Our initial proof-of-concept experiments have applied GE-HTS to biological state modulation and to the alteration of intractable cancer targets. We have intentionally focused our efforts, in part, on small molecule libraries enriched for bioactive molecules and FDA-approved drugs with the idea of repurposing drugs already approved for another indication.
We first applied GE-HTS to the identification of new compounds inducing acute myeloid leukemia (AML) differentiation. One class of molecules emerging from our screening efforts was epidermal growth factor (EGFR) inhibitors. These EGFR inhibitors induce differentiation and inhibit cell viability with case reports now of clinical responses, including two complete remissions, via a non- EGFR, off-target mechanism. We next sought to identify the mechanism of action of these EGFR inhibitors by integrating proteomic and RNAi-based approaches. Spleen tyrosine kinase (SYK) emerged at the intersection. SYK is a cytoplasmic tyrosine kinase critical to B-cell development and signaling in hematopoietic cells and has been recently implicated in multiple hematopoietic malignancies. Genetic and pharmacological inactivation of SYK with a drug in clinical trial for other indications promoted differentiation of AML cells, inhibited AML cell viability, and attenuated leukemia growth in multiple in vivo models of AML.
We next sought to apply GE-HTS to the identification of compounds modulating intractable protein targets, such as transcription factor abnormalities in cancer. We turned our attention to Notch1 in T-cell acute lymphoblastic leukemia (T-ALL). Gain of function mutations in Notch1, which encodes a signaling protein that is converted into a transcription factor upon activation, are the most common genetic abnormality in human T-ALL. Although inhibiting Notch1 activity represents a potential therapeutic opportunity, the discovery of new Notch1 pathway antagonists poses a challenge because the development and execution of high-throughput assays to measure binding of protein to DNA have been difficult. Instead, we derived a 32-gene Notch1 expression signature from genome-wide expression profiling of seven different Notch1 mutant T-ALL cell lines treated with vehicle (Notch1 on) versus a Notch1 inactivating – secretase inhibitor (GSI; Notch1 off). We then screened a small molecule library for compounds inducing the Notch1 off state and identified numerous calcium ion flux modulators as top hits. Among these top hits was bepridil, a compound previously FDA-approved for the treatment of cardiac disease. Bepridil induced the Notch1 off state in numerous Notch1 mutant T-ALL cell lines, and similar to the phenotypic effects of GSI, bepridil induced a G0/G1 cell cycle arrest, inhibited cellular viability, and decreased cell size in multiple T-ALL cell lines. Similar to GSI, bepridil treatment decreased levels of intracellular Notch1 (ICN1), but in contrast to GSI, it decreased levels of the furin-processed extracellular and transmembrane forms of Notch1. The full length Notch1 precursor form, however, accumulates upon bepridil treatment.
These results illuminate the potential of integrating chemical genomic, proteomic, and genetic screening approaches to identify new therapeutic strategies for cancer. Moreover, chemical genomic small molecule screening approaches can lead to discovery of new clinically relevant small molecules for malignancy with the potential for more rapid translation to the clinic.
Citation Information: Clin Cancer Res 2010;16(14 Suppl):IA6-2.