Academic Outreach
NIH Chemical Genomics Initiative Sparks New Roles for Academia and SBS

By Richard M. Eglen
Member, SBS Board of Directors
Chief Scientific Officer, DiscoveRx Corp.

The involvement of academia in high-throughput screening has historically been limited by resources, lack of access to large compound libraries, and infrastructure. Nonetheless, the situation has changed over the last five years or so, and compound screening groups are now established at several academic institutions.1 Moreover, the National Institutes of Health (NIH: www.nih.gov) has established a Chemical Genomics initiative as a key aspect of the NIH Roadmap for Medical Research initiative.2 Compound screening in the non-commercial sector thus complements high-throughput screening approaches traditionally used in most pharmaceutical and biotech companies, presenting a clear opportunity for the Society for Biomolecular Sciences (SBS) to extend its reach into academia, broaden the membership base, and increase its impact. This brief report is a personal perspective on the current status of screening in academia and the emerging role of SBS.

The use of chemical compounds to modulate gene function and elucidate biological pathways involved in human health and disease is often referred to as “chemical genomics.” Under the aegis of the Roadmap for Medical Research, the National Institutes of Health (NIH) has established the NIH Chemical Genomics Center (NCGC; www.genome.gov/12512295), based in the National Human Genome Research Institute’s Division of Intramural Research in Bethesda, MD. The NCGC is the first center in the developing Molecular Libraries Screening Center Network (MLSCN), a consortium of six to 10 centers distributed throughout the United States to be funded in Fiscal Year 2005 (http://grants1.nih.gov/grants/guide/rfa-files/RFA-RM-04-017.html)3

Issues & Opportunities
The MLSCN is part of the “Molecular Libraries and Imaging” component of the NIH Roadmap (http://nihroadmap.nih.gov/molecularlibraries/index.asp), and is being led by the National Institute of Mental Health and National Human Genome Research Institute, with representation from all NIH Institutes and Centers. The MLSCN will screen a collection of compounds being established in a repository that will grow to between 500,000 and one million compounds, and will be complemented by other molecular libraries and imaging initiatives in cheminformatics and technology development. Assays screened at the MLSCN centers will be chosen from the research community by a peer-review process that is currently in development. Together, these initiatives will provide the academic community with access to high-throughput, automated screening of assays on large collections of small molecules.

In some respects, the techniques used in the NIH Chemical Genomics Center and other MSLCN centers will mirror those of screening laboratories in the pharmaceutical industry. However, there are some key differences in focus and outcome.4

First, data generated by the MLSCN will be deposited in a centralized database, called PubChem (http://pubchem.ncbi.nlm.nih.gov/), access to which is freely available to the scientific community.4

Second, assays screened by the consortium will address targets not generally considered by commercial drug-discovery organizations, either because of the esoteric nature of the assay, the target’s perceived “drugability,” or the low commercial potential of the therapeutic lead. This is because the major objective of the MLSCN centers will be to develop research tools for understanding genes, pathways, and phenotypes, rather than necessarily identifying the start of a campaign to make a drug. These “chemical probes” will have potency and aqueous solubility suitable for in vitro research applications, and it is fully intended and that the probes be used as starting points for analoguing optimization chemistry campaigns for in vivo use.

Of the hundreds of thousands of proteins encoded by the human genome, less than 500 have a chemical compound with which they can interact.4 This suggests that there is a large proportion of the genome that remains to be explored in terms of the interactions of proteins and pathways with small molecules. It is likely, therefore, that the structural diversity of the chemical library in the NIH consortium will be broader than that of many pharmaceutical company libraries and include compounds that are non-druglike or toxic, as well as natural products.5

Once all the centers and the database are in place, the impact of the initiative on the academic community may be substantial. Routine access to compounds as research tools will enable better understanding of cellular pathways and/or gene function. Derivatization of compound hits from active precursors to novel therapeutics—i.e., hit-to-lead optimization—will augment the involvement of government and academic scientists in the initial stages of drug discovery.6

As a result, investment in this initiative is substantial, as was recently illustrated in two announcements. The NIH Chemical Genomics Center has entered into an agreement with Kalypsys Inc, valued at up to $30 million over four years, as a means of providing a highly automated robotic screening system.2,4 The NIH also recently awarded approximately $24 million to Discovery Partners International (DPI) to set up and maintain the small chemical molecule repository, and create and maintain a web site that contains various informatic tools and will interface with PubChem.7

Some issues remain, and are still under debate, however, perhaps reflecting the difference in approach in the initiative from that of a commercial organization. One is ownership of intellectual property associated with the data deposited in databases such as PubChem. The NIH has suggested that members of the consortium immediately deposit screening data and waive some patent rights, so that the data and compound structures are publicly available.2 Some university technology transfer officers have alternatively suggested that researchers obtain property rights before submitting the data, but not enforce them for academic use.

The NIH has now issued guidance for the sharing of data and plans for intellectual property.8 These guidelines emphasize the NIH’s view that “enforcement of patents on HTS hit or chemical probes could have a chilling effect on the development of future substantive inventions .....and interfere with the broad utilization of early-stage biological and chemical information.”

Consequently, the guidelines ask that applicants provide rationales for proposed plans that involve principles differing from this general statement of intent. These plans will then be evaluated by a scientific review committee, followed by negotiated agreement between the program staff and the applicant (full details can be found at http://grants1.nih.gov/grants/guide/notice-files/NOT-RM-04-014.html).

A second issue concerns the chemoinformatic tools used to mine the data in the PubChem database. The NIH is still debating whether it will employ commercial chemoinformatic tools or support the development of public sector tools using an open source platform.9

Screening in Academia
The NCGC and the other MLSCN screening centers may be described as “core facilities”—analogous to the Center for Inherited Disease Research (http://www.cidr.jhmi. edu)—aimed at providing extensive HTS infrastructural support for the broad academic community. The approach builds upon the expertise in several screening groups already established in several universities and research institutions, in which large numbers of compounds have been assayed, and extensive data are deposited into searchable databases.1

One of the best known groups in this area is the Institute of Chemistry and Cell Biology (ICCB) at Harvard Medical School. ICCB maintains a database, ChemBank, and is supported by grants from the National Cancer Institute and the National Institute of General Medical Sciences. ChemBank currently contains chemical structures of more than 250,000 bioactive molecules. The ICCB also links to the screening activities in the Laboratory for Drug Discovery and the New England Regional Center of Excellence in Biodefense and Emerging Infectious Diseases.

Other examples include the High-Throughput Bioscience Center at Stanford University, which aims to have 100,000 compounds in the screening collection by the end of 2005; the University of California, San Francisco, which has a database called ZINC that is designed for virtual screening and understanding compound target docking, and which contains over 700,000 structures; the McMaster HTS Laboratory in Canada; the High-Throughput Screening Center at Rockefeller University in New York City; and the High-Throughput Screening Laboratory at Kansas University.

SBS in the Forefront
HTS is thus a focus of government and academic scientists, and will play an increasing role in the identification of targets, assays, research tools, and chemical starting points for drug discovery. Partly as a response to this growing trend, SBS has expanded its relationships with academic screening groups and is meeting with a diverse and enthusiastic reception. The recent regional meeting at the Harvard Medical School was good evidence for this, and it is anticipated that, in the future, there will be further representation of academic screening groups at other SBS regional meetings.

The SBS Academic Outreach Committee, moreover, is very active in developing these interactions, and played a major role in the genesis of the SBS Harvard meeting. The NIH Molecular Libraries Roadmap initiative was discussed at the recent SBS 10th Anniversary Conference in Orlando, and drew a large audience clearly interested in fostering SBS-academic interactions.

Fundamentally, SBS is a forum for biomolecular screeners—both in academia and the commercial sector—to interact, share experiences, and information. As these interactions increase, I expect that there will be dissemination of “best practices” (derived from years of hard-won experience) from the commercial sector groups that will greatly benefit emerging practices in academic screening groups.

The extensive knowledge of many SBS members in the life sciences and the drug-discovery sector is also a valuable resource to share with the academic community, particularly as the aim of the NIH chemical genomics initiative is to produce innovative tools for research and to “contribute in a much more vigorous way to the earliest stages of the drug-discovery pipeline: the identification of drug-discovery targets.”6

Similarly, the public availability of screening data in databases such as PubChem may be of high interest to screeners in either academic or commercial laboratories, particularly when the data (positive and negative hits) will be generated at a very diverse range of assays.

Collectively, the challenge to SBS members is to continue to develop contacts, formally and informally, with the academic community, thereby strengthening ties and fostering interest in SBS. The nexus of the interaction between the emerging screening community in academia and the established community in the commercial sector is where SBS can play a pivotal role.

This can be achieved by developing: additional meetings similar to the recent one at Harvard Medical School, which provided a focused opportunity for interactions and exchange of information; a series of seminars delivered by SBS screeners to new and emerging academic screening groups; internship programs where individuals in the academic group spend a period of time in a commercial lab; and so on.

Collectively, these efforts as well others in the planning stage should expand and diversify the SBS membership, as well as promulgate its decade-old expertise in compound screening, in particular—and in drug discovery, overall.

References & Further Reading:

1. Stein, R (2003) High throughput screening in academia: the Harvard experience. J Biomol. Screen., 8: 615-619.
2. Kaiser, J (2004) NIH gears up for Chemical Genomics, Science, 304, 1728.
3. Borman, S. Charting Better Routes to Drugs. Chem. Eng. News., June 28, 2004.
4. Janssen, D. Small Molecules with Big Promise. Bio.IT, Aug 18, 2004.
5. NIH Press Release, June 9, 2004.
6. Zongoo.com: NIH launches first center in nationwide chemical genomics network. Daily Press and Consumer Information, June 10, 2004.
7. Press Release, Discovery Partners International, August 23, 2004.
8. NIH Guide. Guidance for the sharing of data and resources generated by the molecular libraries screening network (MLSCN) - addendum to RFA RM-04-017. July 22, 2004.
9. Will public chemical genomics projects lead to open source chemoinformatics? BioInform Quarterly Digest 2004.