|
Day
One
Thursday, June 15
7:30am Conference Registration, Morning
Coffee, Exhibit and Poster Set-Up
PROTEIN-TARGETED
DESIGN
8:30 Chairperson's Opening Remarks
| Keynote
Presentation |
 |
8:40 Progress in
Computer-Aided Drug Design
William L. Jorgensen, Ph.D., Whitehead Professor of Chemistry,
Department of Chemistry, Yale University
General issues for structure-based drug design will be covered based on
our experiences with molecule docking, growing, fragment simulations, ADME-properties
evaluation, similarity searching, and free-energy perturbation
calculations. In-house drug development is being pursued through multiple
computer-aided routes, followed by synthesis, and assaying. (a) In one
mode, the design begins with use of the ligand-growing program BOMB, which
rapidly constructs combinatorial libraries given the structure of the
target protein and a selected core and substituents. BOMB grows the
analogs inside the protein's binding site, performs a thorough
conformational search, and estimates the analog's binding affinity or
activity using scoring functions. The QikProp program is applied to filter
all designed molecules to insure that they have drug-like properties
including solubility and cell permeability. MC/FEP simulations are then
performed to refine the predictions for the best scoring leads using
hundreds of explicit water molecules and extensive sampling for the
protein and ligand. (b) Alternatively, screening of known compound
libraries is performed by filtering for drug-likeness and similarity to
known active compounds with QikProp and QikSim, followed by docking with
Glide. (c) Lead compounds are also sought through multiple-copy
simulations of small molecules (fragments) with BOSS. In this case a
protein is saturated with hundreds of copies of a fragment, which are then
annealed to seek consensus binding sites. Recent methodological advances
and representative applications will be presented with emphasis on
inhibitor development for HIV reverse transcriptase. |
|
|
Reaping The
Rewards of
Fragment-Based Drug Design
9:10 Fragments to Clinic: Successful
Applications of Fragment-Based Drug Discovery
Christopher Murray, Ph.D., Director, Computational Chemistry and
Informatics, Astex Therapeutics
This presentation will focus on a structure-based approach to the
design of potent inhibitors starting from weakly binding fragments. The
discovery of cdk2 inhibitors which are currently in phase I clinical
trials will be described. The talk will also outline how the approach was
used to rapidly identify novel hsp90 inhibitors with good in vivo
efficacy. There will be a discussion of fragment linking versus fragment
growing strategies and the approaches will be compared for the serine
proteases, thrombin and Urokinase.
9:40 Assembling Fragments into
Molecules in Structure-Based Drug Design
Jeffrey Wiseman, Ph.D., Vice President and Officer, Technology
& Informatics Department, Locus Pharmaceuticals
Grand canonical ensembles provides a rapid way to calculate
thermodynamically rigorous binding affinities for molecular fragments.
Given this accurate scoring function, the talk is about the
state-of-the-art for the additional factors required in assembling
fragments into molecules with predictable affinities: sampling efficiency,
conformational flexibility, additivity of fragment free energies, tightly
bound water, and protein flexibility. The ability to quantitatively
predict tightly bound water, in particular, is found to be essential in
the accurate prediction of binding poses and energies. |
10:10 Networking Coffee Break, Poster and
Exhibit Viewing
10:55 Combining Flexible Docking and
Multidimensional QSAR from Quantifying Binding Affinities to the Prediction of
Adverse Drug Reactions
Markus Lill, Ph.D., Lecturer, Department of Pharmacy, University of
Basel; Project Lead, ADME Laboratory, Biographics Laboratory 3R
The quantification of ligand-binding affinities to a macromolecular target
represents a major challenge in computer-aided drug discovery. We present a new
technology developed at our laboratory to identify and quantify binding modes
for ligands of biomedical interest. It combines flexible docking at the target
protein with a novel multidimensional QSAR concept (consensus scoring using
Quasar [1,2] and Raptor [3]). In our approach, induced protein fit (receptor-to-ligand
adaptation) is explicitly simulated. The technology has been applied to nuclear
receptors [2,4] and cytochrome P450 [5] and demonstrates the ability to predict
the affinity for large and diverse ligand sets binding to a common target
protein and suggest a potential to predict adverse effects triggered by drugs
and chemicals [6].
 |
11:25 Case Study of Structure-Based Design
for DPP-IV Inhibition
Kenton Longenecker, Ph.D., Protein Crystallographer, Structural Biology
Department, Abbott Laboratories
Pharmaceutical inhibition of dipeptidyl peptidase IV (DPP-IV) is a recently
explored strategy aimed to benefit patients with type-II diabetes by regulating
plasma glucose levels. X-ray crystallographic studies of DPP-IV in complex with
inhibitors aided drug candidate discovery through structure-based evaluation and
design. This presentation will highlight contributions of structural biology to
these efforts at Abbott Laboratories. |
11:55 An Integrated Approach to Library
Design
W. Pat Walters, Ph.D., Senior Research Fellow, Computational
Chemistry and Molecular Modeling, Vertex Pharmaceuticals, Inc.
A chemist designing a combinatorial library must consider many criteria when
selecting reagents for synthesis. Factors such as target potency, physical
properties, metabolic stability, and off-target activity are among many
parameters that must be optimized. Although computational models exist to aid
the chemist, these models are often poorly validated and are not easily
integrated into the drug discovery process. As part of a continuing effort to
provide library design tools for medicinal chemists, we have created a software
tool known as MedChem2. This software provides an easy means of linking a
virtual combinatorial library with a well-validated set of computational models.
The application of these models can dramatically reduce the size of a virtual
library, and help to focus a chemistry effort on the most relevant compounds.
Models in MedChem2 are constructed using NOMAD, an internally developed software
platform that allows computational chemists to identify optimal combinations of
molecular descriptors and machine learning methods. Models generated using NOMAD
can then be published to MedChem2 where they become part of the medicinal
chemistry workflow. This presentation will provide an overview of NOMAD and
MedChem2, as well as example applications of both programs.
12:25pm Lunch on Your Own
(Technology Workshop Sponsorship Available)
EXPERIMENTAL APPROACHES
1:35 Chairperson's Remarks
1:45 Biospectra Analysis: Model Proteome
Characterizations for Linking Molecular Structure and Biological Response
Robert A. Volkmann, Ph.D., Senior Research Fellow, Pfizer Global Research and
Development
Establishing quantitative relationships between molecular structure and
biological effects has been a long-standing goal in drug discovery. An
operationally simple probabilistic structure-activity relationship (SAR)
approach, termed biospectra analysis, which uses pattern similarity between
biospectra of molecules as determinant, will be described. Comparison of
biospectra derived from in vitro assays yields precise chemical structure
information and is useful for identifying pharmacology and side effect
similarities between medicines. Specific examples will be provided.
2:15 Discovery of Ligands for Nurr1 by
Combined Use of NMR Screening with Different Isotopic and Spin-Labeling
Strategies
Leszek Poppe, Ph.D., Principle Scientist, Molecular Structure Department,
Amgen, Inc.
A comprehensive approach to target screening, hit validation and binding
site determination by nuclear magnetic resonance spectroscopy (NMR) is
presented. Screening by 19F NMR signal perturbation, followed up by
magnetization transfer experiments and second-site screening with spin-labeled
ligand, led to discovery of a molecule which binds to the Ligand Binding Domain
of Nurr1 with dissociation constant ~ 20 uM. With the help of uniform and
residue specific 15N isotope labeling and derivatization of Cys residues with
2-mercaptoethanol-1-13C we were able to determine the binding site location with
knowledge of the APO coordinates.
HOMOLOGY, MODELING, AND
LIGAND-BASED DESIGN
2:45 Success and Lessons from 11 betaHSD1
Homology Models: Using Models of Very Low Homology in Docking and Design
Ying-Duo Gao, Ph.D., Senior Research Fellow, Molecular Systems, Merck
& Co. Inc.
11beta-hydroxysteroid dehydrogenase type1 (11beta-HSD1) is a potential
target for treatment of some of the health problems associated with Metabolic
Syndrome. In assisting medicinal chemistry in lead optimization, we generated
homology models of human and mouse 11betaHSD1 enzymes based on 17betaHSD1 and
7alphaHSD1 structures. These models were used extensively in the program. In
this presentation we demonstrate that models with very low homology (<25%),
that may only partly present the active site correctly, can be highly valuable
for understanding SAR of the ligands and suggesting new designs. In addition,
the recently available crystal structures of human and mouse 11betaHSD1 allowed
us to assess the homology models and discuss how to improve this type of
modeling.
3:15 Technology Watch (Sponsorship
Available)
Beyond Scoring Functions: Fast First-Principles Quantum and Molecular Physics Tools for Structure-Based Drug
Design
Kay Denis, MBA., Postion TBA, Timtec, Inc., Business Partner/distributor of Quantum Pharmaceuticals; On behave Peter Fedichev, Ph.D., CSO, Quantum.Corporate Scientific Officer, Quantum
Pharmaceuticals
Quantum is a computational platform aimed at direct pplications of powerful quantum and molecular modeling structure based tools for fast and accurate predictions of protein-ligand binding affinities. The ingredients are: a good, polarizable force field based on quantum mechanics; a solvation energy model; and statistical physics for entropy change evaluation. The transferrability of the vacuum force field for aqueous calculations is ensured by its polarizability and the quality of the water model, but not by the excessive parametrization. This ideology provides a solid and physically motivated ground for the free binding energy calculations. The software has been extensively tested against all available experimental data and has been recently released. Possible areas of application: High Throughput Virtual Screening, Computer Aided Drug Design, Cheminformatics, Computational Chemistry and Biology, ADMET prediction. An unprecedented level of accuracy together with a simple and efficient user interface allows for in silico lead optimization. The technology is expected to speed up pharma R&D radically and irrevocably change the computational chemistry and drug discovery.
3:30 Networking Refreshment Break, Poster
and Exhibit Viewing
4:00 Docking Studies of A3 Agonists and
Antagonists Suggest Activation Mechanism of Adenosine Receptor
Soo-Kyung Kim, Ph.D., Senior Research Associate, Beckman Institute, Molecular Simulation Center, California Institute of Technology
To determine the different binding modes of agonist and antagonist to A3
adenosine receptor (AR), the docking studies of A3 selective nucleoside agonists
and nucleoside/non-nucleoside antagonists were compared by using the FlexX and
the FlexiDock automated docking procedure. There are common binding regions for
the exocyclic amino groups of each 9H-purine ring in agonist Cl-IB-MECA and the
1H-[1,2,4]triazolo[1,5-c]quinazoline ring in non-nucleoside antagonist CGS15943
through H-bonding to the side chain of N6.55. In addition, hydrophobic
interactions of N6-aromatic group were overlapped, interacting with F168 in EL2.
For binding domains of agonist, additional H-bonding of the ribose 3'- and
5'-substituents with the hydrophilic amino acids T3.36, S7.42, and H7.43 and
hydrophobic interaction of the terminal methyl group of the
5-uronamide interacted with the hydrophobic side chain of F6.44
required for the characteristic side-chain movements of TM6 and TM7. Here we
present the novel insights in the putative activation mechanism of A3AR. The
fact that agonist binding disrupts the intramolecular H-bonding network through
W6.48 and H7.42 and occurs the rotation of TM6 suggests that these activation
mechanisms might be extended to other members of the AR family. Collaboration
with Kenneth A. Jacobson, Ph.D., Section Chief.
4:30 Bridging the Gap between Protein
Cavities by Virtual Screening
Hans Briem, Ph.D., Senior Scientist, Compound Design and Compound
Characterization/Computational Chemistry, Schering AG
One strategy to enhance the binding affinity of protein ligands is to
identify sub-pockets on the protein surface to which additional tether groups
linked to a given scaffold may bind. We will demonstrate how we used the new
module FlexxC-Pharm to efficiently dock large combinatorial libraries while
simultaneously considering user-defined pharmacophore constraints. This approach
allows us to prioritize sets of virtual combinatorial libraries by their ability
to bridge the gap between protein cavities of the target of interest.
 |
5:00 Case Presentation of
Strategies for High-Throughput Molecular Docking
Diane Joseph-McCarthy, Ph.D., Principal Research Scientist,
Chemical and Screening Sciences Department, Wyeth Research
In structure-based design, molecular docking techniques are used to
predict the binding of a set of proposed compounds. Accurate molecular
docking of small molecules to a target structure requires adequate
sampling and accurate scoring of each proposed ligand in the target
binding site. The use of our in house pharmacophore-based docking approach
on several therapeutic target projects will be presented and compared to
the use of commercial software. In addition, the exploration of the
inclusion of protein flexibility during docking will be discussed. |
5:30-6:30 Networking Reception in the
Exhibit Hall
For more information, please contact:
Shelley W. Amster, Conference Director
781-972-5473 • samster@healthtech.com
For exhibit and sponsorship information, please contact:
Suzanne Carroll, Manager, Business Development
781-972-5452 • scarroll@healthtech.com
|
