The Role Of Academe-based HTS

---

The need for new pharmaceutical agents has resulted in the birth of high throughput screening (HTS). Technological advances in robotics, computing hardware and software, and detectors, coupled with advances in genomics, proteomics and combinatorial chemistry has enabled the development of large-scale, automated testing of many compounds against a single target.

In its most basic form, HTS is a means of carrying out thousands or even millions of experiments in a short period of time. Its primary use has been in the pharmaceutical sector where it has been utilized as a method of hit identification in the drug-discovery process. In this context, a large number of chemical compounds are tested for their ability to perturb the function of a target of interest. Any compounds that are found to alter the activity of the target, once confirmed, are then classified as hits, and can form the basis of a lead identification project with the goal of identifying a new chemical entity, and eventually a new drug. It wasn’t until the late 1990s that HTS began to move into academe. The first dedicated university-based HTS lab was the Institute of Chemistry and Cell Biology (ICCB) at Harvard Medical School (Boston, MA).

The first Canadian university-based HTS lab opened in 2001 at McMaster University (Hamilton, ON). At that time, the McMaster HTS Lab was one of four or five academe-based HTS labs in the world. Since then, a large number of screening labs have been formed in academe, with at least four more being formed in Canada alone. In the United States, the National Institutes of Health (NIH) has recently set up the Molecular Libraries Screening Centers Network with $89 million US in funding for nine screening labs across the U.S.

HTS utilizes automation to test multiple samples in microwell-plate format. It has been commoditized to the point where there are currently a number of equipment and software vendors that provide off-the-shelf automation and informatics platforms for HTS. It is still possible to build customized solutions, but the costs associated with standard laboratory automation, control systems and informatics have decreased to the point that it is within the reach of academic research centres. Using off-the-shelf solutions and an open-door policy to researchers across Canada, the McMaster HTS Lab has carried out over 20 large scale screens (>100,000 compounds) and over 30 smaller scale screens (1,000 to 10,000 compounds) over the past four years, generating in excess of 10 million data points. While modest in relation to throughput in a pharmaceutical setting, it is exciting for Canadian academic scientists and researchers to have access to small molecule libraries, and screening equipment and expertise.

The role of screening in academe is not to replicate the use of HTS in drug discovery as pioneered by the pharmaceutical industry. Universities are ill equipped for drug discovery as the cost and infrastructure requirements of taking a compound from the hit stage through to filing of a new chemical entity and subsequent clinical trials is most often well beyond the capabilities of a single university. Where academe-based HTS labs excel are in the areas of chemical biology, chemical genetics, target identification and validation, assay development and the open and free exchange of knowledge. Large-scale genomics projects have led to the realization that there are a large number of genes that code for proteins for which no function is known, and consequently proteomics efforts are required to characterize these proteins. Thus, there is a need for small molecules that are able to specifically activate or deactivate a protein of interest. The primary proponent of this approach is Stuart Schreiber, PhD at Harvard University (Cambridge, MA) and the Broad Institute (Cambridge, MA) — a collaboration of researchers from Harvard and Massachusetts Institute of Technology (Cambridge, MA). Schreiber has published a number of papers on the use of small molecules to probe biology.1-3

Academe is well-suited for study in this area as it encompasses more basic research, asking fundamental questions about proteins and their function, structure and localization, etc. As well, it can lead to new and exciting cross-disciplinary collaborations, as there is now a real need for chemists and biologists to talk to one another. Also, one of the primary goals of the academic screening community is the free and open sharing of data. The results of HTS campaigns in the private sector are intellectual property that is never released to the public. In the academic sector, the goals are somewhat different. While both sectors have similar goals with regards to improving human and animal health, a key driver for industry is profit. While there is increasing pressure for universities to commercialize their discoveries, one of the main goals of academic research is discovery, and as such, the publication and free and open sharing of data and information, with prudent protection of intellectual property, is crucial.

To this end, the McMaster HTS lab has posted complete data sets for screens against E. coli dihydrofolate reductase4 and the SARS coronavirus main protease5 on its website, http://hts.mcmaster.ca. The lab was also the first to host a docking and data mining competition for computational chemists and data miners. The purpose of the contest was to allow competitors to test their tools and methods with a “real world” HTS data set. Given a set of screening data and the compound structures (training set), competitors were to predict the activity of an untested set of compounds for which only structures were available (test set). The submitted results were then compared to the experimental data for the test set. The results of the competition were published in a special edition of the Journal of Biomolecular Screening in October 2005. Over 70 groups and individuals took part in the competition, with one-third of participants equally from the pharmaceutical, small biotech and academic sectors.

A second competition is currently underway with data sets provided by the NIH Chemical Genomics Center and the ICCB, www.sbsonline.org/datamining/index.php. There also exist a number of databases that make available chemical and biological data on hundreds of thousands of compounds. Two of the most comprehensive are PubChem, http://pubchem.ncbi.nlm.nih.gov, maintained by the NIH and National Library of Medicine in the U.S., and ChemBank, http://chembank.broad.harvard.edu, looked after by the Broad Institute. These databases allow a user to look up the effects of a given compound in any assay in which it has been tested. The McMaster HTS Lab is in the process of depositing its screening data in PubChem.

In Canada, the Canadian Institutes of Health Research (Ottawa, ON) has provided funding for a Canadian Chemical Biology Network (CCBN). The CCBN will encourage chemists to submit their compounds for screening by biologists against their target or targets of interest. One of the overriding requirements, and challenges, for taking part in the CCBN is that data is shared publicly, while taking care to adhere to various institutions’ policies regarding intellectual property. This free and open sharing of information will be important to filling the pharmaceutical industry’s drug-discovery pipelines as academe-based efforts identify new targets for their drug-discovery programs. As well, these screening efforts will generate new knowledge that will be used as catalysts for launching new biotech companies.

As mentioned above, the McMaster HTS Lab does not see its role as one of replicating the pharmaceutical industry’s drug-
discovery business. This does not mean, however, that it is not interested in disease therapies and improving health. Academe-based screening ideals are well-suited to tackling diseases of the developing world that don’t satisfy the pharmaceutical industry’s profit motive. Organizations such as the Drugs for Neglected Diseases Initiative (Geneva, Switzerland) aren’t developing drugs themselves, but rather are bringing together the various components needed for research and development of new therapies, a kind of “virtual drug company.” In this way, a HTS lab in one university may work with a chemist at another university to screen natural products against a target identified by a biologist at a third university. Further members of the network are involved in medicinal chemistry efforts, filing of an IND and clinical trials. At these later stages, it is possible for larger, not-for-profit organizations to step in with logistical and financial support. It will not be easy, but it is now within the reach of such a virtual organization to produce a therapeutic agent.

Jonathan Cechetto is manager of the McMaster HTS Lab.

References
1. Schreiber, S.L. “The small-molecule approach to biology: Chemical genetics and diversity-oriented organic synthesis make possible the systematic exploration of biology.” C&E News 81 (2003): 51-61.
2. Strausberg, R.L, and S.L. Schreiber. “From knowing to controlling: a path from genomics to drugs using small molecule probes.” Science 300 (2003): 294-95.
3. Schreiber, S.L. “Small Molecules: The missing link in the central dogma.” Nat. Chem. Biol. 1 (2005): 64-66.
4. Zolli-Juran, M., J.D. Cechetto, R. Hartlen, D.M. Daigle, and E.D. Brown. “High Throughput Screening Identifies Novel Inhibitors of Escherichia coli Dihydrofolate Reductase that are Competitive with Dihydrofolate.” Bioorg. Med. Chem. Lett. 13 (2003): 2493-96.
5. Blanchard, J.E., N.H. Elowe, C. Huitema, P.D. Fortin, J.D. Cechetto, L.D. Eltis, and E.D. Brown. “High throughput screening identifies inhibitors of the SARS coronavirus main proteinase.” Chem. Biol. 11 (2004): 1445-53.