Drug Design Approaches to Manipulate the Agonist-Antagonist

PDF hosted at the Radboud Repository of the Radboud University Nijmegen The following full text is a publisher's version. For additional information about this publication click this link. http://hdl.handle.net/2066/101060 Please be advised that this information was generated on 2021-09-27 and may be subject to change. Molecular Determinants of Partial Agonism by Progesterone Receptor Modulators Receptor Progesterone by Agonism of Partial Molecular Determinants Molecular Determinants of Partial Agonism by Progesterone Receptor Modulators Implications for Drug Design & Discovery Scott J Lusher Scott J Lusher Molecular Determinants of Partial Agonism by Progesterone Receptor Modulators Implications for Drug Design & Discovery Proefschrift ter verkrijging van de graad van doctor aan de Radboud Universiteit Nijmegen op gezag van de rector magnificus prof. mr. S.C.J.J. Kortmann, volgens besluit van het college van decanen in het openbaar te verdedigen op maandag 18 februari 2013 om 13:30 uur precies door Scott James Lusher geboren op Barking, London te 18 februari 1976 Promotor Prof. dr. J. de Vlieg Copromotor Dr. R. McGuire (Bioaxis Research) Manuscriptcommissie Prof. dr. L.M.C. Buydens Prof. R.C. Glen (University of Cambridge) Prof. dr. F.P.J.T. Rutjes Title: Molecular Basis for Ligand-Mediated Partial Agonism in the Progesterone Receptor: Implications for Drug Design & Discovery Copyright © 2012 Scott James Lusher, `s-Hertogenbosch, The Netherlands Printing by: Graphical Stroke, www.graphicalstroke.com Cover design: Scott J. Lusher & Tomasz Karawajczyk Lay-out: Raoul Frijters ISBN/EAN: 978-90-9027302-0 "...we balance probabilities and choose the most likely. It is the scientific use of the imagination,..." Sherlock Holmes For Mum & Dad…. for all I am. For Rita…. for all I can be. For Samuel & Tomas…. for all I want to be. Table of contents Chapter 1 A general introduction to computer-assisted drug discovery 6 Chapter 2 A general introduction to steroid receptor structural biology 22 Chapter 3 Drug design approaches to manipulate the agonist-antagonist 32 equilibrium in steroid receptors Chapter 4 Structural basis for agonism and antagonism for a set of chemically 50 related progesterone receptor modulators Chapter 5 X-ray structures of the progesterone receptor ligand-binding domain 70 in its agonist state reveal differing mechanisms for the mixed profiles of 11β-substituted steroids Chapter 6 Peptide-recruitment profiles for progesterone receptor modulators 96 differentiate the classic ligand classes and generally correlate to the 909 degree of clash between ligand and Met Chapter 7 A molecular informatics view on best practice in multi-parameter 126 compound optimization Summary Overview and structure of thesis 155 Samenvatting Overzicht en opzet van de dissertatie 163 Curriculum vitae 170 Additional publications and bibliography 172 Words of thanks 178 CHAPTER1 GeneralA general introduction to computer- assisted drug discovery Computer-Assisted Drug Discovery 9 1.1. The Complexity of Drug Discovery & Design On January 20th 1961 John F. Kennedy was sworn in as the 35th President of the USA. In his inaugural address he committed the US to sending a man to the moon, and returning him safely, before the end of that decade. As we all know, Neil Armstrong fulfilled this 1 Chapter promise in July 1969, this incredible undertaking taking less than 9 years. On average, it takes the pharmaceutical industry 10-15 years and more than $800 million to bring a new prescription drug to the market [1,2]. Whilst it may not be a truly fair comparison, we might be tempted to conclude that drug design is not rocket science, but actually a far more difficult undertaking. It is certainly amongst the most intellectually challenging and risky pursuits in any commercial enterprise, and judging by falling output compared to expenditure, it is becoming more difficult all the time [3]. So why is the pharmaceutical industry finding it so difficult to maintain output? The conclusion from most observers is that the questions are becoming more difficult. We demand more effective and safer drugs that can be administered for longer periods of time, at the same time as regulatory authorities grow increasingly more stringent. In many areas, drug designers are working on 4th and 5th generation medications for particular diseases, and therefore efficacy and safety alone are not sufficient. These new drugs also have to be better and more cost effective than the earlier treatments. In the future, personalized medicine (beyond the one size fits all approach of current therapeutics) will further challenge an industry already struggling to reach its required output [4]. Science, and the way we undertake research, is changing. The scale of information generation is now so great that science has to adapt or drown in a data deluge. Additionally, the days of individual researchers working in isolated groups and focused only on their own, increasingly narrow expertise, are also numbered. Breakthroughs are increasingly made at the interface of disciplines by groups of scientists benefitting from the combination of their diverse skills. These changes have allowed some observers to suggest we are entering a "4th paradigm" in science, progressing from the 1st and 2nd paradigms characterized by Observation, Theory and Experimentation and the 3rd paradigm, characterized by Simulation, to a Data-Driven model for research. The basis for data-driven drug-design is that all information and knowledge related to a problem is available so that key parameters/properties requiring improvement can identified and compounds synthesized to address these issues [5]. The one shared aspect of all conventional drug design and optimization strategies is that they are iterative [6]. A compound is designed, synthesized and then tested. The data resulting from this testing is evaluated before influencing the next round of design. We refer to this as the discovery cycle which is dependent on the premise that new experimental data immediately impacts on design. The challenges of this data-driven model center on the need for timely data- 10 Chapter 1 generation, rapid data-dissemination, insightful data-analysis and interpretation (in the context of multiple data sources) as well as the vision and empowerment to act decisively on new insight. 1.3. The Process of Drug Discovery Most pharmaceutical companies follow a similar drug discovery and development pipeline consisting of a few key phases which are introduced below with focus on the role of computing in expediting these tasks: 1.3.1 Target Discovery and Validation Drug discovery begins with the identification and validation of a protein target; whose activity it is believed can be modulated to treat a disease. The use of modern molecular biology including functional genomics (such as micro-array technologies) and proteomics has revolutionized the way in which we identify new targets. Bioinformatics underpins all of this work and is now also a fundamental tool in drug discovery and design. An obvious outcome of the genomics and molecular biology revolution of the last two decades has been the dramatic increase in the number of possible molecular targets for pharmaceutical companies to prioritise. The various -omics technologies are a critical step in determining the role of these proteins in disease and therefore validating their potential as drug targets [7]. However, there is growing realization that whilst a target may play a crucial role in disease, it might not necessarily be a good target for drug discovery. This is the difference between a biological validation of the target and the determination of its suitability to be modulated by a small molecule. This chemical evaluation of the target is often referred to as assessing its "druggability" with protein bioinformatics, including methods like sequence analysis, homology modeling, threading, and the identification and characterization of binding pockets, providing valuable methods to help determine the value of a target. Druggability is often evaluated on a protein family basis as these families share characteristics important for ligand-binding and therefore provide valuable insight into shared ligand chemo-types and selectivity. 1.3.2 Lead Discovery The next step in the process is to identify compounds able to modulate the activity of the target in a desirable fashion whilst retaining a chemical structure suitable for future compound optimization. The use of robotized high-throughput screening (HTS) methods underpins lead discovery and is a process with informatics at its core. Informatics is required to manage screening collections, validate experimental data and store and disseminate results [8]. The source of compounds for screening varies, but to maintain libraries of hundreds of thousands of compounds, pharmaceutical companies must routinely purchase new Computer-Assisted Drug Discovery 11 compounds. The choice of which compounds to purchase is critical to the quality of a screening set and therefore the success of future screening campaigns. Companies tend to purchase a mix of optimally chemically "diverse" compounds to cover all areas of chemical space and "focused" compounds they believe will be particularly suited to the targets 1 Chapter that they work on. If we presume that there are 1060 possible compounds, and estimates of this number vary greatly, and a typical HTS will test ~250000-500000 compounds, then it is clear that we are only scratching the surface of available

Load more