High-throughput screening’s ability to perform thousands of experiments efficiently and under carefully controlled conditions has made it an important tool for basic and translational biological research. At Columbia University, the JP Sulzberger Columbia Genome Center and the Chemical Probe Synthesis Facility provide a flexible platform for researchers interested in applying high-throughput experimentation in their work. On December 17, 2012, the Genome Center hosted a symposium to spotlight its capabilities in high-throughput screening, to explain the important role that synthetic chemistry plays in high-throughput screening, and to describe some recent research projects at Columbia that have utilized these tools.

As High-Throughput Screening Core Scientific Director Charles Karan explained, the Genome Center operates a suite of advanced technologies for automated liquid handling, robotic assay implementation, and high-throughput, high-content microscopy. The Genome Center also offers Columbia University researchers access to several large collections for conducting high-throughput screens. These include the Columbia Cell Line Encyclopedia, which includes 850 cancer cell lines collected from around the world, as well as a chemical diversity library curated by researchers in the Chemical Probe Synthesis Facility. This “tool chest” gives Columbia investigators access to a pre-selected set of compounds that have been predicted to result in the highest quality potential hits. Karan also reported that the Genome Center recently negotiated an arrangement with Sigma Aldrich to give access to the company’s shRNA clones to researchers at Columbia at greatly discounted rates.

Karan emphasized that the high-throughput screening facility encourages close collaboration with Columbia University investigators. “We want you to be involved. We want you to be as proactive as possible with your screens. We want you to understand the decisions we’re making and why we’re making them,” he stressed. Researchers can access their data through an online software platform called Pipeline Pilot, which can both generate simple reports and enable advanced data analysis. Ultimately, the goal of the staff at the High-Throughput Screening Core is to support researchers by developing assays that help to identify specific molecules of interest, and best address their scientific problems.

As physician and organic chemist Donald Landry pointed out, however, merely identifying a possible target is often just the first step in a longer process of identifying a unique chemical that will produce a desired effect in vivo. He and colleagues in his lab work closely with researchers to “transform hits into probes,” developing optimized molecules for biological research. Chemists can help biologists in many ways, analyzing research and patent literature for information about related approaches that have been tried in the past; identifying and acquiring compounds for testing structure-activity relationships; dissecting chemical domains in a molecule of interest to optimize activity, eliminate adverse effects, and promote selectivity; and generating libraries of molecules that can be screened in follow-up experiments aimed at refining a hypothesis. If these efforts achieve promising results, chemists at Columbia also assist biomedical research by producing sufficient amounts of a compound for testing in animal models or human cohorts. In addition to providing this overview, Landry discussed examples of compound optimization work his laboratory has undertaken.

Another important tool for conducting high-throughput screening is the creation of small molecule libraries. At Columbia, Brent Stockwell directs the Chemical Probe Synthesis Facility, which collaborates closely with the High-Throughput Screening Core to generate and curate small molecules used in high-throughput screens. The Chemical Probe Synthesis Facility can also investigate hits from screens to improve upon available compounds, synthesize new compounds, perform molecular modeling, and conduct in vivo studies of small molecules. Stockwell discussed work that he has led that has identified a previously unknown cell death mechanism called ferroptosis. Using an approach called modulatory profiling, researchers in his laboratory identified a cluster of 62 compounds that cause cell death but are not active in previously identified cell death mechanisms such as apoptosis and necrosis. One of these compounds is erastin, which was independently determined to selectively kill engineered RAS mutant tumor cells through an iron-dependent, oxidative, non-apoptotic pathway. In his talk, Stockwell described efforts his lab has taken to identify genetic regulators of ferroptosis, to tease out the upstream signaling pathway that induces ferroptosis, to develop compounds capable of inducing or suppressing this cell-death mechanism, and to develop in vivo models that will help to improve compounds for potential clinical applications.

High-throughput screening has also been an important tool in work by Chris Henderson, director of the Motor Neuron Center, to understand the molecular mechanisms involved in amyotrophic lateral sclerosis (ALS, or Lou Gehrig’s disease). In particular, he is interested in understanding the causes of axon degeneration in motor neurons in patients with ALS, the primary cause of the debilitation that is characteristic in patients with the disease. After harvesting motor neurons derived from embryonic stem cells, he worked with Brent Stockwell’s lab to screen 50,000 compounds against motor neurons in suboptimal growth conditions, in order to identify molecules that promoted axon growth. They found four lead compounds that stimulated axon growth, the most promising of which were in the statin family. Because statins have anti-cholesterol effects, and the spinal cord cannot generate its own cholesterol, Henderson worked with Donald Landry to find a similar compound with the same known target. Their goal is to synthesize analogues that could potentially promote axon growth while avoiding the adverse side effects of anti-cholesterol medications to the central nervous system. Henderson’s team has also been looking for downstream targets in the same signaling pathway, and working with Hai Li in the Genome Center, found that blocking prenylation in motor neuron cells can achieve a similar effect without interfering with cholesterol levels.

Andrea Califano described Columbia University’s participation in the Library of Integrated Network-Based Cellular Signatures (LINCS), a National Institutes of Health initiative to compile a publicly accessible library of molecular signatures associated with specific cellular perturbations. Working in collaboration with researchers at the Broad Institute and other institutions, Califano is leading two centers at Columbia, one focusing on technology development and the second on experimentation. Under one grant, researchers are developing graph theoretical methods to study LINCS profile data from other centers in order to identify mechanisms of action that are either the same or different when comparing in vitro and in vivo data. This project will also generate new tools for identifying cell line-specific compound mechanisms of action, as well as novel algorithms for cataloguing gene-gene, gene-compound, and compound-compound synergies. In the experimental grant, Califano’s team is developing reporters for measuring synergy and viability in different phenotypes, and establishing a baseline understanding of the synergies that develop when researchers screen for synergistic compounds whose mechanisms of action are as orthogonal as possible. Identifying orthogonal mechanisms is important because many compound mechanisms of action are the same. The team will also screen 1,000 compound combinations that have been predicted to be the most synergistic. In his talk, Califano described the computational and experimental techniques that his team is using to achieve these goals.

The event also featured a presentation by Paul Fletcher of Perkin-Elmer, who discussed Alpha Technology and time-resolved fluorescence, two technologies that his company has developed that can be used in the context of high-throughput screening. Alpha Technology provides a method for detecting whether cellular events have occurred. It can be used, for example, for exploring molecular phosphoproteins, protein-protein interactions, proteases, kinases, and post-translational modifications. Fletcher also briefly described time-resolved fluorescence, which takes advantage of the long half-lives of fluorescent molecules to track cellular events at multiple time points.

All of the researchers stressed that Columbia has developed a strong expertise in high-throughput screening and chemistry, and invited Columbia researchers to contact the Columbia Genome Center for guidance on how these methodologies can support their work.

The symposium was organized by Olivier Couronne, executive director of genome sequencing and analysis at the Columbia Genome Center.

— Chris Williams