Friday, June 8
8:00 AM Chairperson’s Remarks
Robert A. Copeland, Ph.D., Executive Vice President of R&D and Chief Scientific Officer, EpiZyme, Inc.
8:10 Computational Target Evaluation and Lead Generation Strategies
Diane Joseph-McCarthy, Associate Director, Infection Computational Sciences, AstraZeneca Pharmaceuticals, Inc.
Computational strategies for target evaluation and lead generation will be presented. Various computational approaches for assessment of binding site druggability, in particular with a view toward designing broad spectrum antibacterials, will be described. In addition, the expanding role of virtual and fragment-based screening for lead generation in the drug discovery process will be discussed. Specifically, the combined use of computational and biophysical-based methods, the development of screening libraries, and the challenges overall will be highlighted.
8:40 Drug-Target Residence Time: An Alternative Approach to Lead Optimization
Robert A. Copeland, Ph.D., Executive Vice President of R&D and Chief Scientific Officer, EpiZyme, Inc.
Durable pharmacologic activity depends not merely on the affinity of drug-target interactions but also on the lifetime of the drug-target complex in vivo. Residence time provides an experimental measure that related directly to the lifetime of the drug-target complex. The residence time concept will be introduced and examples of its application for improving lead efficacy and safety will be presented. Structural features that facilitate long drug-target residence time will be discussed in the context of a new theoretical framework for drug dissociation: the retrograde induced-fit model.
9:10 Kinetically Informed and Structure-Based Drug Design
Rumin Zhang, Ph.D., Biophysics Team Lead, In Vitro Pharmacology, Merck Research Laboratories
It is now well recognized that binding kinetics (BK) plays a pervasive role not only in the ultimate drug action in patients, but in the very process of drug discovery as well. This talk will address the potential synergy between structure-based drug design (SBDD) and BK in developing best in class drugs that engage biological targets with kinetically optimized potency and selectivity. Discussed will be the concept and experimental measurement of drug-target residence time, and the roles of SBDD in elucidating orthosteric and allosteric mode of binding and in guiding the design of slow binding inhibitors. Finally, a future that combines the spatial resolution of SBDD and the temporal clarity of BK is envisioned.
9:40 Coffee Break in the Exhibit Hall with Poster Viewing
10:20 Keynote Presentation
Binding Kinetics in Drug Discovery
David Swinney, Ph.D., CEO, Institute for Rare and Neglected Diseases Drug Discovery (iRND3)
Binding kinetics are the rate of association (kon) and dissociation (koff) of a drug to a physiological target and are integral to a medicine’s efficacy, safety and selectivity. Greater understanding and application of binding kinetics will to add value to drug discovery. Opportunities where value may be added include experimental design, molecular descriptors for chemical optimization, molecular mechanisms of action that increase efficacy, safety and the therapeutic index, duration of action and differentiation of medicines.
10:50 Water-Shielded Hydrogen Bonds: A Handle on Binding Kinetics of Protein-Ligand Complexes
Xavier Barril, Ph.D., ICREA Research Professor, Physical Chemistry Department, University of Barcelona
With growing awareness about the fundamental role of binding kinetics for drug efficacy, there is a pressing need to understand and predict structure-kinetic relationships. We have recently shown that water-shielded hydrogen bonds act as kinetic traps, a knowledge that creates opportunities to modulate the kinetic behavior of drug candidates (Schmidtke et al., JACS 2011; 133(46):18903). An overview of the principle will be presented along with our more recent research in the field.
11:20 Ligand-Target Residence Time Based Drug DiscoveryNils Hansen, Ph.D., CEO, Vipergen ApSThe lifetime of the drug-target complex quantified by its residence time is recognized as an important predictor of in vivo efficacy but is difficult to address in early drug discovery efforts at desired throughputs. Binder Trap Enrichment® is a novel concept for identifying small molecules based on residence time from DNA-encoded chemical libraries. Here binding complexes of a DNA-encoded ligand and DNA-tagged target protein are trapped within emulsion droplets, thus allowing instant ligand ranking by residence time.
11:50 Luncheon Presentation (Opportunity Available) or Lunch on Your Own
1:25 PM Chairperson’s Remarks
Florian Nigsch, Ph.D., Investigator I, Developmental and Molecular Pathways, Novartis Institutes for BioMedical Research
1:30 Human Pocketome for Predicting Activities and Adverse Reactions
Ruben Abagyan, Ph.D., Professor, Skaggs School of Pharmacy & Pharmaceutical Sciences, San Diego Supercomputer Center, University of California, San Diego
Ligand docking and scoring to multiple conformations of the binding pocket of a target is a powerful and chemically unbiased predictor of compound activity. We developed a comprehensive set of experimentally validated pockets in human proteome. Further, we optimized protocols of conformational selection, docking, scoring to those pockets or related ligand fields. The method can be used for screening, predicting subtype specificity, adverse effects and can guide a medicinal chemistry program.
2:00 Some Novel Methodologies that Complement Structure-Based Approaches
Kiyoyuki Omoto, Ph.D., Computational Scientist, WorldWide Medicinal Chemistry, Pfizer Neusentis
While some successful applications of structure-based drug design (SBDD) have been reported, the methodology still has not been very reliable as it often ignores essential processes for a protein-ligand interaction, such as induction of conformational change in a binding site. In this talk, a few approaches will be introduced which complements the SBDD; (1) 3D diversity analysis for docking poses, (2) an ensemble docking approach which utilizes more than one protein structures to provide diverse potential binding poses and (3) an idea generator for scaffold swapping.
2:30 Highly Parallelized Structural Biology Informatics as One Component towards the Development of Novel Poly-Pharmacology Compounds
Susan Thomas, Ph.D., Scientist B, Biomedical Informatics Center of Indian Council of Medical Research, National Institute for Research on Reproductive Health (ICMR)
We have developed and implemented highly parallelized workflows for protein structure modeling, binding site prediction, alignment, and computation of receptor site similarities, and generation of novel ligand diversity by functionality shuffling across similar binding sites. We applied these automated protocols to amplify the structural coverage of the human Kinome and the diversity of existing ATP site inhibitors with the goal to prioritize compounds with desirable selectivity profiles. We also applied our technology to the human drug targets to identify potential unrecognized off-targets of approved therapeutics.
3:00 Refreshment Break
3:15 GPCR Solvent Binding Thermodynamic Properties: Any Impact on Design?
Andrea Bortolato, Ph.D., Senior Computational Chemist, Heptares Therapeutics Ltd.
Exploiting a structure-based design approach for GPCRs is now possible for these important targets because of the recent tremendous technological developments in GPCR stabilization and novel structural biology. We applied a variety of computational approaches to evaluate the role of the solvent in protein druggability, structure-activity relationship elucidation and ligand free energy of binding prediction. Using recent crystallographic information we analyzed the potential impact of these innovative computational methods on SBDD, including the perturbation of receptor hydration properties (water energetics) upon ligand binding, for these targets and a set of reference enzyme targets.
3:45 Druggability and Ligandability: Structural Genomics Insights Suggest in Excess of 10,000 Potentially Druggable Proteins in the Human Proteome
Florian Nigsch, Ph.D., Investigator I, Developmental and Molecular Pathways, Novartis Institutes for BioMedical Research
We present our recent efforts to exploit the information from comprehensive structural genomics and chemogenomics resources. To identify the set of currently ligandable domains we identified all InterPRO signatures that contain ligand-binding residues across all 75,000+ PDB structures. The occurrence of the identified InterPRO signatures in the human proteome suggests that for at least 25% of the human proteome there is chemical matter with high affinity available. We present applications of the derived data in target identification and chemical tractability assessment.
4:15 Biological Networks in Three Dimensions
Yu (Brandon) Xia, Ph.D., Assistant Professor, Bioinformatics Program, Department of Chemistry, Department of Biomedical Engineering, Boston University
Two studies are highlighted where three-dimensional modeling of protein networks provides insights into the biophysics and evolution of biological systems. First, a structural analysis of the yeast protein-protein interaction network reveals that protein evolution is constrained by the biophysics of protein folding, the mutational robustness of proteins, and the biophysics and function of protein-protein interactions. Second, a structural analysis of human-virus protein-protein interaction networks reveals distinct principles governing antagonism versus cooperation in host-pathogen interactions.
4:45 End of Conference