PDB COMMUNITY FOCUS: Stephen K. Burley, Structural GenomiX, Inc.

Stephen K. Burley (M.D., D.Phil., F.R.S.C.) is the chief scientific officer of Structural GenomiX, Inc. (SGX; http://www.stromix.com/), located in San Diego, California. SGX is an oncology focused drug discovery and development company, with Troxatyl® in clinical trials and multiple protein kinase inhibitors in preclinical development. Prior to joining SGX, he was the Richard M. and Isabel P. Furlaud professor and chief academic officer at The Rockefeller University, and a full investigator in the Howard Hughes Medical Institute. Burley received an M.D. degree from Harvard Medical School in the joint Harvard-MIT Health Sciences and Technology program and, as a Rhodes Scholar, received a D.Phil. in Molecular Biophysics from Oxford University. He trained in internal medicine at the Brigham and Women's Hospital, and did post-doctoral work with Gregory A. Petsko at the Massachusetts Institute of Technology and William N. Lipscomb at Harvard University. With William J. Rutter and others at the University of California, San Francisco and The Rockefeller University, Burley co-founded Prospect Genomics, Inc., which was subsequently acquired by SGX. He is a fellow of the Royal Society of Canada and of the New York Academy of Sciences.

Q: What made you decide to leave academia for a position in a biotechology firm?

A: Although my much publicized departure from The Rockefeller University and the Howard Hughes Medical Institute was initially greeted with surprise, one of my closest colleagues told me that on reflection he realized it was inevitable. He was referring to my training in both structural biology and medicine, followed by internship and residency in internal medicine. He was also reflecting on my long-standing commitment to exploring new areas of science and technology. In short, I value the breadth of my training and am continually looking for new opportunities to learn.

The advent of high-throughput X-ray crystallography and improvements in computational chemistry convinced me that a leadership role in the right industrial environment would allow me to help change the way in which new drugs are discovered. For much of the late 1990s, I wondered which company and when. My involvement in Structural GenomiX, Inc. (SGX) as a member of the scientific advisory board and my role as a co-founder of Prospect Genomics, Inc. (an in silico drug discovery company) made it clear that the time was ripe for such a move. Shortly after SGX acquired Prospect Genomics, I volunteered myself for the post of Chief Scientific Officer.

Three plus years later there have been many changes, both personal and professional. I left the excitement of New York City for southern California, where my family and I live in idyllic surroundings complete with palm trees, the Pacific ocean, and as much fly fishing as I can fit in to an extremely busy professional schedule. At SGX, I have helped guide our transition from a gene-to-structure platform company to an oncology drug discovery and development company. We have combined high-throughput X-ray crystallography with combinatorial synthesis and state-of-the-art computational chemistry tools to create a best-in-class lead discovery engine that is delivering drug development candidates targeted at protein kinases that cause cancer. On the clinical front, SGX now has a compound in clinical trials for acute myeloid leukemia and various solid tumor malignancies.

Q: Chairing the RCSB PDB Advisory Committee requires a lot of time and energy - what is your interest in this board?

A: My involvement with the PDB coincided with its transition from Brookhaven National Laboratory to the RCSB, when Helen Berman asked me to chair the advisory committee. Unlike some invitations to do committee work, this one was easy to accept.

The PDB is a mission critical global archive on which all of biomedical research relies. As an industrial scientist, I also appreciate the enormous economic benefit that the PDB provides to pharmaceutical and biotechnology companies. Without facile access to this comprehensive, single archive of experimentally-derived three-dimensional structures of biological macromolecules, our drug discovery activities would be considerably handicapped. Finally, the PDB represents a vitally important intellectual and educational resource. During the next decade, I foresee that the wealth of structures in the PDB will play a decisive role in integrating our understanding of distinct chemical and signaling steps as we build up "systems biology" views of how cells and organ systems work at the molecular level.

I would add that the quality of the RCSB leadership and the committee makes my job as chair of the RCSB PDB Advisory Committee both interesting and fun. My fellow committee members were recruited from among the top ranks of industrial and academic scientists. Working closely with Helen, John Westbrook, and Phil Bourne and the advisory committee, we have forged a strong partnership that provides the best possible advice to the RCSB and serves as a vocal advocate for PDB user community.

Q: You are also a member of the advisory committee of the wwPDB how do you see the future of this organization?

A: In my answer to your previous question, I described the PDB as a global archive. I view the PDB as a resource that must grow and develop for the collective benefit of all humanity. I share HelenĘs vision for the wwPDB, and feel honored to be playing an advisory role now that the RCSB, the MSD, and PDBj are working closely together to make it happen. As the keeper of the single, global archive, the RCSB and the US government funding agencies together have a central role that brings with it a special responsibility to ensure the security and stability of the archive. I believe that the wwPDB advisory committee provides an important venue to help ensure that the wwPDB membership effectively coordinates effort and funding from different parts of the globe.

Q: There are now ~30,000 structures in the PDB. How many more structures do you think there would be if the drug companies deposited all the structures they have been working on. Should the drug companies be encouraged to deposit these structures?

A: Extrapolating from structure determination efforts at SGX over the past five years, I suspect that there are well in excess of ten thousand protein structures in the proprietary databases of biotech and pharma companies world wide. Even before my move to industry, I was a strong advocate of including depositions from drug companies in the PDB. These proprietary databases contain a wealth of structures illuminating how protein targets recognize small molecule ligands. If present in the PDB, I believe that this enormous body of information could be brought together and used to synthesize a more predictive, quantitative understanding of protein-drug interactions that would serve to accelerate drug discovery activities in both academe and industry. Companies are being encouraged to deposit their structures, and many do so to a limited extent. I am aware of three impediments to increasing the number of such depositions.

First, there are understandable intellectual property concerns relating to proprietary small molecules that can only be resolved by deferring depositions until the necessary composition-of-matter patents pertaining to the small molecules have been issued. Once full disclosure has occurred, the structures can and should be included in the PDB.

Second, there are practical limits to resources that company scientists can devote to depositing structures in the PDB. The business of these companies is drug discovery and development, not publications and PDB depositions. The RCSB is committed to helping companies get their data into the PDB with as little effort as possible. Working together, SGX and the RCSB have developed a facile method for porting large numbers of structures from our company database to the PDB. I am optimistic that this advance will be used by other companies.

Finally, some companies have taken the position that de novo protein structures should be kept confidential to avoid enabling competitors. From my experience at SGX, I can tell you that this strategy rarely yields a sustainable competitive advantage. We routinely determine structures that others companies are holding in confidence, thereby leveling the playing field. I would like to see biotechnology and pharmaceutical companies come to a common recognition that de novo protein structures represent pre-competitive information that should be shared, just as with single nucleotide polymorphism data. Structural genomics programs in the US, Europe, and Japan should help the industry come to this view sooner rather than later. With the enormous flood of new structures coming into the PDB over the next decade, pre-competitive structural information will soon be freely available for many drug discovery targets. Once the deluge starts, companies will have nothing to gain by keeping their de novo structures confidential and much publicity value to be had by depositing them to the PDB.