Tuning Up the Assays
Both the selection of assays and the small molecules being screened through the NIH’s Molecular Libraries Screening Centers will generate experiments never before imagined, says director of the Penn Center for Molecular Discovery, Scott Diamond. “The NIH screening network has the opportunity to consider assays that would never make it past an HTS [high-throughput screening] oversight committee in pharma,” he says. “For example, assays could include human bone marrow or human blood as the cellular background.”
At the University of Pennsylvania, the small-molecule effort might include microarray-based HTS, zebrafish screening, and metallo-organic chemistries. Diamond has developed a nanodroplet technique to screen thousands of compounds on a single microarray in homogeneous assays.
The microarray approach is already paying dividends. “We had used microarrays to profile a library against more than 20 proteases,” Diamond says. “And then, when we learned that our colleague Paul Bates in microbiology had assays demonstrating that cathepsin L was involved in SARS [severe acute respiratory syndrome] entry, we were able to hand him a 10 nM inhibitor of cathepsin L.”1
Elsewhere in the Penn Center for Molecular Discovery, Eric Meggers has developed a technique to rapidly assemble pharmacophores on a metal atom. “Metallo-organics are rarely explored by pharmaceutical companies,” says Diamond, “but they allow rapid access to a diversity space that is difficult to explore with standard combinatorial or parallel synthetic approaches.” For example, Meggers recently reported the generation of a low nanomolar inhibitor of GSK-3 kinase that is extremely selective.2"Inhibitors of cathepsin L prevent severe acute respiratory syndrome coronavirus entry," Simmons G, Proc Natl Acad Sci Vol 102, 11876-81 Aug. 16, 2005"An organometallic inhibitor for glycogen synthase kinase 3," Bregman H, J Am Chem Soc Vol 126, 13594-5 Oct. 27, 2004
Jason Varney Photography
The National Institutes of Health has placed the heft of a large academic collaboration, on par in scale with the Human Genome Project, behind a task usually performed by pharmaceutical companies. When it launched its Molecular Libraries Screening Centers Network this year, it embarked on an enormous needle-in-a-haystack challenge, says Scott Diamond, director of a new center established under the plan. In search of small molecules that selectively interact with specific biological targets, scientists in the initiative will screen hundreds of thousands of candidates in scores of high-throughput assays.
The screening centers network provides some $88.9 million in grants during the next three years to nine academic institutions – including the University of Pennsylvania's new Penn Center for Molecular Discovery, led by Diamond.
While the payoff in drug targets for pharmaceutical companies employing such technologies has not been exactly overwhelming, the NIH expects such interrogations will produce a wealth of biological knowledge. The aim is to begin building a free-access database of the biological activities of small molecules, hopefully to stimulate research and hasten drug development.
"I think we will benefit from the bumps in the road that the pharmaceutical industry has experienced," Diamond says. "The technology is there, and the timing is just right."
The NIH Chemical Genomics Center, established last year, is now joined by centers at Columbia University, Emory University, Southern Research Institute, the Burnham Institute, The Scripps Research Institute, the University of New Mexico Albuquerque, Diamond's group at the University of Pennsylvania, the University of Pittsburgh, and Vanderbilt University.
The 10 centers will screen molecules supplied from a repository established under contract at Discovery Partners International, a drug discovery research firm in San Francisco, using a selection of assays submitted by scientists around the country.
Broadly speaking, the criteria for selecting the appropriate molecules is suitably pragmatic, according to Douglas Livingston, senior vice president, chemistry, for the company, and the purpose is definitely not to discover new drugs. "We're most interested in the information we get out of the screening," he says. "What I'm interested in is getting as many biological readouts as possible, as quickly as possible."
The initial focus will be on drug-like molecules, but at a later date more globular, natural products will hopefully be introduced into the repository, says Diamond. "I think the natural products will be a very important part to add to the library." Practically speaking, each of the collaborating centers will receive 75,000 to 100,000 molecules every six months from the repository for screening.
In mid-August, the first large shipments of molecules were being made. The goal for the initial three-year project is that each center should be able to conduct 10 to 20 screens on the complete library of molecules as it exists two- and-a-half years from now – perhaps more than 500,000 molecules.
The launching of the NIH centers certainly occurred at a time when chemical biology is coming into its own. "It's clear something is happening," says Stuart Schreiber, from the Broad Institute in Boston. As director of the National Cancer Institute-sponsored Initiative for Chemical Genetics, he has been a proponent of systematically interrogating living systems with small molecules. "I don't know why it took so long, to be honest. I suppose it takes time for the community to come together," Schreiber adds.
Roughly a year ago, the United Kingdom's Biotechnology and Biological Sciences Research Council (BBSRC), together with the Royal Society of Chemistry, and the Engineering and Physical Sciences Research Council, held a meeting to gauge the interest of the scientific community to chemical biology research. "Surprise, surprise, there was a lot of interest," says Colin Miles, whose group at the BBSRC is responsible for developing new areas of funding. The research councils launched a new funding stream to the tune of €10 million, inviting applications from groups interested in running projects using selective chemical interventions to investigate biological processes.
Seventy-six applications were received. "Many of them were for fairly substantial projects involving collaborations between people working in the biological and chemical sciences," says Miles. The successful applications are expected to be made public this month, destined to spill even more data into the public realm.
The key will be to relate the objects in these repositories to each other and also to information and objects in genetic databases, like the International HapMap Project, or databases of chromatin structure, Schreiber explains. "I think what we're seeing is the very beginning of an open-source approach to drug discovery," he says. "It's an exciting prospect to think how scientists in the future will generate starting points for new drugs and hypotheses for drug targets that would not be possible otherwise."