In the Spotlight
Molecular Libraries Initiative: Application to Tox21 and the NIH Chemical Genomics Center
Published: August 2, 2010
As this new “21st Century Toxicity” vision is being explored, it is important to take stock of ongoing, related efforts to assess the impact of molecules on biological pathways and their downstream impacts. Here we examine one such effort, the Molecular Libraries Initiative (MLI). As noted below, the National Institutes of Health Chemical Genomics Center (NCGC) carries out work to advance both the MLI and 21st century toxicity testing, underscoring the connection between the two initiatives. Specifically, the NCGC is a hub of the federal government’s “Tox21” program to advance 21st Century Toxicology.
The U.S. National Institutes of Health (NIH) Roadmap began as a process designed to ask probing questions regarding the re-definition of priorities and scientific challenges during a series of meetings in 2002 initiated by then NIH director Elias Zerhouni (Zerhouni, 2003). As a result of talking to the nation’s top biomedical leaders from academia, government, and the private sector, three key themes, comprising various initiatives within them, emerged as the foundation for the NIH Roadmap. The first theme, “New Pathways to Discovery,” set out to provide a toolbox for biomedical researchers, capitalize on the sequencing of the Human Genome Project, and provide access to scientific resources such as new technologies and databases. The second theme, “Research Teams of the Future,” set out to combine skills and expertise of scientists in order to foster novel discoveries and partnerships between public and private sectors. The third theme, “Re-engineering the Clinical Research Enterprise,” set out to develop new partnerships among various patient communities, community-based physicians, and researchers. The ultimate goal of this theme was to create better networks of academic centers working jointly on clinical trials and organization of clinical research protocols and strategies pertaining to the clinical research workforce (Zerhouni, 2003).
With the aforementioned themes propelling the development of the NIH Roadmap, one of the initiatives launched within “New Pathways to Discovery” was the Molecular Libraries Initiative (MLI) in 2003. This initiative was established to provide researchers the tools and infrastructure to identify chemical probes for exploring biological pathways and provide a starting point for drug discovery (“Dealing with,” 2008; Kaiser, 2008). More specifically, the MLI was brought about to empower research communities to utilize chemical probes and small molecules as tools to perturb genes and pathways (Lazo, 2006), while using data from the Human Genome Project and exploiting the development of new therapeutics for rare diseases. Keeping in line with the “big picture” goals set forth by the NIH Roadmap (chemical diversity, assay diversity, chemoinformatics, instrumentation, and predictive ADME/toxicology), a screening phase (pilot phase) and probe production phase of the MLI were established (Austin, et al., 2004; MLI Technology Development).
The pilot phase was introduced around 2004 to focus on technology development and to establish 10 screening centers, including the NIH Chemical Genomics Center, with a diverse array of screening/robotic capabilities (Thomas, et al., 2009). The screening paradigm was envisioned as assays assigned to individual screening centers and high-throughput assays conducted with a large chemically diverse screening collection with known and unknown biological activities (part of the Molecular Libraries Small Molecule Repository (MLSMR)). It was noted in 2004 that chemical probes identified from screens would be subjected to refinement based on potency, selectivity, and solubility, making them more suitable for additional studies. The resulting probes would then be made available to all researchers without intellectual property restrictions (Kaiser, 2008). Screening data would also be deposited into PubChem, a publicly available database that provides structure, screening, and probe information from the different centers (Lazo, 2006; “Dealing with,” 2008). Around 2008, the NIH officially announced the second phase of the MLI, where three funded centers perform over 25 screens per year (Kaiser, 2008), leading to the synthesis of new compounds and secondary assays (Lazo, 2006).
As aforementioned, one component of the Roadmap/MLI during the pilot phase of testing included predictive ADME/Toxicology, which has been integral to the paradigm shift in toxicity testing in the 21st century (Shukla, 2009). A connection between the MLI and the Tox21 includes the robotic tools and technologies utilized for screening a diverse set of compounds, such as natural products, metabolic intermediates, and agents with potential in vivo toxicity across a battery of toxicity and pathway-specific endpoints (Austin, et al., 2004). In other words, the scientific expertise and resources born from the MLI will support continual “development, standardization, and validation of novel approaches to obtain comprehensive ADME and toxicological profiles that could better predict how new molecular entities will perform in humans” through alternative in vitro-based approaches to animal testing.
There were several ADME/Tox projects which were funded during the pilot phase of the MLI, including gene expression studies in drug inflammation models as predictive of adverse drug reactions (Shaw, et al., 2009), predictive ADME-Tox modeling (Rodgers, et al., 2010), and the use of human stem cells for toxicity screening (Kim, et al., 2009). The implementation of the MLI has also been demonstrated through a range of Tox21 assays performed at NCGC (Shukla, 2009). These assays involved quantitative high throughput screening (qHTS), robotic technology, and over 1,300 chemicals from the National Toxicology Program (NTP) to obtain a better understanding of mechanisms of toxicity in human and rodent cell-based assays. In addition, computational resources have been put into place for improved toxicity prediction and chemical prioritization (Huang, et al., 2009). Overall, targets that have been screened as part of the MLI can also be utilized for toxicity testing. Furthermore, an antagonist vs. agonist mode of screening can be taken depending upon the pathway in question, which will differentiate drug targets and toxicity targets (Miller, et al., 2010).
The strength of the connection between the MLI and Tox21 initiatives rests upon the aforementioned goals initially set forth by the Roadmap. The increasing assay, chemical, and instrumentation diversity has set the foundation for the current high-throughput toxicity assays conducted at NCGC, and additional computational resources will enable better predictive power with regard to chemical perturbation of toxicity pathways. Those seeking to advance “21st Century Toxicity” would do well to take stock of the MLI and similar programs.
Austin, C.P., Brady, L.S., Insel, T.R. & Collins, F.S. (2004). NIH Molecular Libraries Initiative. Science. 306, 1138-1139.
Huang, R., Southall, N., Xia, M., Cho, M.H., Jadhav, A., Nguyen, D.T., Inglese, J., Tice, R.R. & Austin, C.P. (2009). Weighted feature significance: A simple, interpretable model of compound toxicity based on the statistical enrichment of structural features. Toxicol. Sci. 112, 385-393.
Kaiser, J. (2008). Molecular biology. Industrial-style screening meets academic biology. Science. 321, 764-766.
Kim, S., Dere, E., Burgoon, L.D., Chang, C.C. & Zacharewski, T.R. (2009). Comparative analysis of AhR-mediated TCDD-elicited gene expression in human liver adult stem cells. Toxicol. Sci. 112, 229-244.
Lazo, J.S. (2006). Roadmap or roadkill: A pharmacologist’s analysis of the NIH Molecular Libraries Initiative. Mol. Interv. 6, 240-243.
Miller, S.C., Huang, R., Sakamuru, S., Shukla, S.J., Attene-Ramos, M.S., Shinn, P., Van Leer, D., Leister, W., Austin, C.P. & Xia, M. (2010). Identification of known drugs that act as inhibitors of NF-kappaB signaling and their mechanism of action. Biochem. Pharmacol. 79, 1272-1280.
Rodgers, A.D., Zhu, H., Fourches, D., Rusyn, I. & Tropsha, A. (2010). Modeling liver-related adverse effects of drugs using knearest neighbor quantitative structure-activity relationship method. Chem. Res. Toxicol. 23, 724-732.
Shaw, P.J., Ditewig, A.C., Waring, J.F., Liguori, M.J., Blomme, E.A., Ganey, P.E. & Roth, R.A. (2009). Coexposure of mice to trovafloxacin and lipopolysaccharide, a model of idiosyncratic hepatotoxicity, results in a unique gene expression profile and interferon gamma-dependent liver injury. Toxicol. Sci. 107, 270-280.
Shukla, S. (2009). A New Perspective on the Past, Present and Future of Toxicity Testing. AltTox.org. Available here.
Thomas, C.J., Auld, D.S., Huang, R., Huang, W., Jadhav, A., Johnson, R.L., Leister, W., Maloney, D.J., Marugan, J.J., Michael, S., Simeonov, A., Southall, N., Xia, M., Zheng, W., Inglese, J. & Austin, C.P. (2009). The pilot phase of the NIH Chemical Genomics Center. Curr. Top. Med. Chem. 9, 1181-1193.
Zerhouni, E. (2003). Medicine. The NIH Roadmap. Science. 302, 63-72.