Projects

Small-Molecule Microarrays: A Tool for High-Throughput Ligand Discovery

Uncovering the function of several thousands of gene products, in various states of post-translational modification, will be a key challenge in the post-genome era. Cell permeable small molecules that bind and perturb the function(s) of these gene products should prove useful as tools in studies that require temporal or spatial control over the protein target.  Specific small-molecule probes may uncover novel therapeutic targets for human disease as well as serve as templates for therapeutic design. In order to identify small-molecule probes for each protein function, high-throughput methods for discovery or design of ligands will be needed.

Our group has worked to build a technology platform for high-throughput ligand discovery that is centered on the use of small-molecule microarrays. We aimed to make microarrays including compounds that were not intentionally synthesized with reactive appendages for attachment, such as natural products and FDA-approved drugs. After sampling several covalent-capture methods, we identified simple and robust conditions that were used to immobilize hundreds of natural products and FDA-approved drugs alongside products from diversity-oriented synthesis. The microarrays, which contain ~10,000 different compounds, are compatible with protein-binding screens directly from cellular lysates, thereby removing the need for protein purification. We collaborate with groups participating in the Broad Institute Disease Initiatives, such as the Pyschiatric Disease Initiative and the Infectious Disease Initiative, interested in identifying ligands to proteins related to specific diseases. Additionally, we screen small-molecule microarrays against several proteins that play a role in cancer, sponsored by the National Cancer Institute's Initiative for Chemical Genetics (ICG), with a bias towards proteins involved in transcriptional regulation. We use secondary binding assays involving surface plasmon resonance (SPR) or fluorescence-based thermal shifts to prioritize compounds for additional in vitro or in vivo studies. We are also pursuing small-molecule microarrays as a platform for proteomic profiling using protein-small molecule interactions as a signature of cell type or cell state.

NCI Initiative for Chemical Genetics-Sponsored Cancer Collaborations:

• Professor Paolo Dotto, Massachusetts General Hospital
• Professor Peter Howley, Harvard Medical School
• Professor Ronen Marmorstein, Wistar Institute
• Professor Karl Münger, Harvard Medical School
• Professor Paul Nghiem, University of Washington
• Professor David Sacks, Brigham and Women's Hospital
• Professor Dean Tantin, The University of Utah
• Professor Robert Tjian, UC Berkeley

Building A Small-Molecule Toolbox to Study Transcriptional Regulation

Transcription factors that become overactive in cancers are promising yet untested targets for cancer therapeutics. These proteins mediate the excessive transcription of genes whose products are required for tumor growth and metastasis. Inhibiting the function of a transcription factor requires specific disruption of DNA–protein or protein–protein interactions. The discovery or design of small molecules that disrupt these interactions specifically has thus far proven to be a significant challenge. Using small-molecule microarrays, we identified a ligand to the yeast protein Hap3p, a 16 kDa subunit of the Hap2/3/4/5p transcription factor complex involved in aerobic respiration and the nutrient-response signaling network. The compound, called haptamide, reversibly inhibits activation of a GDH1-lacZ reporter gene in dose-dependent fashion. Whole-genome transcriptional profiling was used to compare treatment with a more potent compound, called haptamide B, to a hap3Δ deletion strain and provided further evidence that the compound selectively inhibits Hap2/3/4/5p-mediated transcription in cells. So far, the mechanism for inhibition by haptamide B is unknown. We propose to identify which biomolecular interaction(s) in the transcription factor complex are disrupted by haptamide B using biochemical and structural approaches. We will also examine whether the compound inhibits the human NF-Y complex, a ubiquitously expressed transcription factor that regulates the transcription of genes implicated in myeloid differentiation and osteosarcoma, and investigate functional consequences of inhibition. Finally, compounds related to haptamide will be synthesized in an effort to improve activity against the yeast and human complexes.

Building upon experience with haptamide, we are screening for ligands to over 50 human transcription factors including HDACs, STATs, and NFκB. The high-throughput and miniaturized nature of the microarray-based screen allows us to probe large panels of proteins for binding in a relatively short period. The proteins are screened against microarrays containing bioactive compounds, natural products, and products of diversity-oriented synthesis. We also hope to screen roughly 200 annotated transcriptional regulators from S. cerevisiae using lysates prepared from strains containing GFP-tagged proteins. Putative small-molecule ligands will be characterized in secondary assays involving SPR, reporter genes, and phenotypic studies. Assay results will be made publicly available through ChemBank. In this fashion, we hope to create a chemical biology resource for members of the transcription research community. Finally, through screening, we hope to gain insight into the types of compounds that bind and modulate the functions of varying structural and functional classes of transcription factors. Hopefully, the results will advance efforts to design compounds that target this very important set of proteins.

Structural Genomics Consortium:
Small-Molecule Aids in Crystallization

Obtaining diffraction-quality crystals can be a rate-limiting step in determining three-dimensional protein structures when using x-ray crystallography. Proteins are often stabilized when in complex with small molecules such as drugs, cofactors, or substrates. This stabilization can promote the macromolecular crystallization process. In collaboration with the Structural Genomics Consortium, we are screening several proteins of medical relevance that are considered to be high-value structural targets for ligands using the small-molecule microarray platform. Small molecules that bind to these proteins may be used as aids in crystallization trials or they may serve as probes of protein function. Structures of selected protein-small molecule complexes may prove to be useful in the area of drug design.

Library-Based Development of
New Optical Imaging Probes

Small molecule and GFP-based fluorophores have demonstrated value in optical imaging of cells and have been applied to the study of individual biomolecules. However, severe limitations with respect to their photophysical properties and their specificity hinder application in high-resolution live cell imaging. A library-based approach focused on the development of new fluorescent probes with optimized properties for single-molecule resolution optical imaging in living cells is proposed. Libraries of cyanine, rhodamine, and rosamine-type fluorophores, synthesized using combinatorial methods in the laboratory of Professor Young-Tae Chang at NYU, are screened for the ability to target specific RNA or protein sequences in vitro and in live cells with high specificity. For example, RNA aptamers and protein targets of interest are screened against small-molecule microarrays containing the fluorophores. Additional solution-based and cell-based screens are performed in the Chang lab as well as the laboratories of Professor Alice Ting at MIT and Dr. Paul Clemons of the Broad Institute. This research project is supported through our Exploratory Center for the Development of High Resolution Probes for Cellular Imaging.