Bioisosteres in Medicinal Chemistry E-Book
144,99 €
Mehr erfahren.
- Herausgeber: Wiley-VCH
- Kategorie: Wissenschaft und neue Technologien
- Serie: Methods and Principles in Medicinal Chemistry
- Sprache: Englisch
Written with the practicing medicinal chemist in mind, this is the first modern handbook to systematically address the topic of bioisosterism.
As such, it provides a ready reference on the principles and methods of bioisosteric replacement as a key tool in preclinical drug development.
The first part provides an overview of bioisosterism, classical bioisosteres and typical molecular interactions that need to be considered,
while the second part describes a number of molecular databases as sources of bioisosteric identification and rationalization. The third part
covers the four key methodologies for bioisostere identification and replacement: physicochemical properties, topology, shape, and overlays of
protein-ligand crystal structures. In the final part, several real-world examples of bioisosterism in drug discovery projects are discussed.
With its detailed descriptions of databases, methods and real-life case studies, this is tailor-made for busy industrial researchers with little time for reading, while remaining easily accessible to novice drug developers due to its systematic structure and introductory section.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 379
Veröffentlichungsjahr: 2012
Ähnliche
Contents
Cover
Methods and Principles in Medicinal Chemistry
Title Page
Copyright
List of Contributors
Preface
A Personal Foreword
Part One: Principles
Chapter 1: Bioisosterism in Medicinal Chemistry
1.1 Introduction
1.2 Isosterism
1.3 Bioisosterism
1.4 Bioisosterism in Lead Optimization
1.5 Conclusions
References
Chapter 2: Classical Bioisosteres
2.1 Introduction
2.2 Historical Background
2.3 Classical Bioisosteres
2.4 Nonclassical Bioisosteres
2.5 Summary
References
Chapter 3: Consequences of Bioisosteric Replacement
3.1 Introduction
3.2 Bioisosteric Groupings to Improve Permeability
3.3 Bioisosteric Groupings to Lower Intrinsic Clearance
3.4 Bioisosteric Groupings to Improve Target Potency
3.5 Conclusions and Future Perspectives
References
Part Two: Data
Chapter 4: Bioster: A Database of Bioisosteres and Bioanalogues
4.1 Introduction
4.2 Historical Overview and the Development of Bioster
4.3 Description of Bioster Database
4.4 Examples
4.5 Applications
4.6 Summary
4.7 Appendix
Acknowledgment
References
Chapter 5: Mining the Cambridge Structural Database for Bioisosteres
5.1 Introduction
5.2 The Cambridge Structural Database
5.3 The Cambridge Structural Database System
5.4 The Relevance of the CSD to Drug Discovery
5.5 Assessing Bioisosteres: Conformational Aspects
5.6 Assessing Bioisosteres: Nonbonded Interactions
5.7 Finding Bioisosteres in the CSD: Scaffold Hopping and Fragment Linking
5.8 A Case Study: Bioisosterism of 1H-Tetrazole and Carboxylic Acid Groups
5.9 Conclusions
Acknowledgments
References
Chapter 6: Mining for Context-Sensitive Bioisosteric Replacements in Large Chemical Databases
6.1 Introduction
6.2 Definitions
6.3 Background
6.4 Materials and Methods
6.5 Results and Discussion
6.6 Conclusions
Acknowledgments
References
Part Three: Methods
Chapter 7: Physicochemical Properties
7.1 Introduction
7.2 Methods to Identify Bioisosteric Analogues
7.3 Descriptors to Characterize Properties of Substituents and Spacers
7.4 Classical Methods for Navigation in the Substituent Space
7.5 Tools to Identify Bioisosteric Groups Based on Similarity in Their Properties
7.6 Conclusions
References
Chapter 8: Molecular Topology
8.1 Introduction
8.2 Controlled Fuzziness
8.3 Graph Theory
8.4 Data Mining
8.5 Topological Pharmacophores
8.6 Reduced Graphs
8.7 Summary
References
Chapter 9: Molecular Shape
9.1 Methods
9.2 Applications
9.3 Future Prospects
References
Chapter 10: Protein Structure
10.1 Introduction
10.2 Database of Ligand–Protein Complexes
10.3 Generation of Ideas for Bioisosteres
10.4 Context-Specific Bioisostere Generation
10.5 Using Structure to Understand Common Bioisosteric Replacements
10.6 Conclusions
References
Part Four: Applications
Chapter 11: The Drug Guru Project
11.1 Introduction
11.2 Implementation of Drug Guru
11.3 Bioisosteres
11.4 Application of Drug Guru
11.5 Quantitative Assessment of Drug Guru Transformations
11.6 Related Work
11.7 Summary: The Abbott Experience with the Drug Guru Project
Acknowledgments
References
Chapter 12: Bioisosteres of an NPY-Y5 Antagonist
12.1 Introduction
12.2 Background
12.3 Potential Bioisostere Approaches
12.4 Template Molecule Preparation
12.5 Database Molecule Preparation
12.6 Alignment and Scoring
12.7 Results and Monomer Selection
12.8 Synthesis and Screening
12.9 Discussion
12.10 SAR and Developability Optimization
12.11 Summary and Conclusion
Acknowledgments
References
Chapter 13: Perspectives from Medicinal Chemistry
13.1 Introduction
13.2 Pragmatic Bioisostere Replacement in Medicinal Chemistry: A Software Maker's Viewpoint
13.3 The Role of Quantum Chemistry in Bioisostere Prediction
13.4 Learn from “Naturally Drug-Like” Compounds
13.5 Bioisosterism at the University of Sheffield
References
Index
Methods and Principles in Medicinal Chemistry
Edited by R. Mannhold, H. Kubinyi, G. Folkers
Editorial Board
H. Buschmann, H. Timmerman, H. van de Waterbeemd, T. Wieland
Previous Volumes of this Series:
Gohlke, Holger (Ed.)
Protein-Ligand Interactions
2012
ISBN: 978-3-527-32966-3
Vol. 53
Kappe, C. Oliver / Stadler, Alexander / Dallinger, Doris
Microwaves in Organic and Medicinal Chemistry
Second, Completely Revised and Enlarged Edition
2012
ISBN: 978-3-527-33185-7
Vol. 52
Smith, Dennis A. / Allerton, Charlotte / Kalgutkar, Amit S. / van de Waterbeemd, Han / Walker, Don K.
Pharmacokinetics and Metabolism in Drug Design
Third, Revised and Updated Edition
2012
ISBN: 978-3-527-32954-0
Vol. 51
De Clercq, Erik (Ed.)
Antiviral Drug Strategies
2011
ISBN: 978-3-527-32696-9
Vol. 50
Klebl, Bert / Müller, Gerhard / Hamacher, Michael (Eds.)
Protein Kinases as Drug Targets
2011
ISBN: 978-3-527-31790-5
Vol. 49
Sotriffer, Christoph (Ed.)
Virtual Screening
Principles, Challenges, and Practical Guidelines
2011
ISBN: 978-3-527-32636-5
Vol. 48
Rautio, Jarkko (Ed.)
Prodrugs and Targeted Delivery
Towards Better ADME Properties
2011
ISBN: 978-3-527-32603-7
Vol. 47
Smit, Martine J. / Lira, Sergio A. / Leurs, Rob (Eds.)
Chemokine Receptors as Drug Targets
2011
ISBN: 978-3-527-32118-6
Vol. 46
Ghosh, Arun K. (Ed.)
Aspartic Acid Proteases as Therapeutic Targets
2010
ISBN: 978-3-527-31811-7
Vol. 45
Ecker, Gerhard F. / Chiba, Peter (Eds.)
Transporters as Drug Carriers
Structure, Function, Substrates
2009
ISBN: 978-3-527-31661-8
Vol. 44
All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.
Library of Congress Card No.: applied for
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library.
Bibliographic information published by the Deutsche Nationalbibliothek
The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.d-nb.de.
© 2012 Wiley-VCH Verlag & Co. KGaA, Boschstr. 12, 69469 Weinheim, Germany
All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.
Composition Thomson Digital, Noida, India
Cover Design Schulz Grafik-Design, Fußgönheim
Print ISBN: 978-3-527-33015-7
ePDF ISBN: 978-3-527-65433-8
ePub ISBN: 978-3-527-65432-1
mobi ISBN: 978-3-527-65431-4
oBook ISBN: 978-3-527-65430-7
List of Contributors
Frank H. Allen
Cambridge Crystallographic Data Centre (CCDC)
12 Union Road
Cambridge CB2 1EZ
UK
Karam B. Alsayyed Ahmed
University of North Carolina at Greensboro
Department of Chemistry & Biochemistry
Center for Drug Design
Greensboro, NC 27410
SUA
Pedro J. Ballester
European Bioinformatics Institute
Wellcome Trust Genome Campus
Hinxton, Cambridge CB10 1SD
UK
David A. Bardwell
Cambridge Crystallographic Data Centre (CCDC)
12 Union Road
Cambridge CB2 1EZ
UK
Caterina Barillari
The Institute of Cancer Research
Cancer Research UK Cancer Therapeutics Unit
15 Cotswold Road
Sutton SM2 5NG
UK
Nicholas P. Barton
GlaxoSmithKline Pharmaceuticals
New Frontiers Science Park (North)
Coldharbour Road
Harlow, Essex CM15 5AD
UK
Michael J. Bodkin
Eli Lilly Limited
Erl Wood Manor
Windlesham, Surrey GU20 6PH
UK
J. Phillip Bowen
University of North Carolina at Greensboro
Department of Chemistry & Biochemistry
Center for Drug Design
Greensboro, NC 27410
USA
and
Mercer University
College of Pharmacy and Health Sciences
Department of Pharmaceutical Sciences
3001 Mercer University Drive
Atlanta, GA 30341
USA
Nathan Brown
The Institute of Cancer Research
Cancer Research UK Cancer Therapeutics Unit
15 Cotswold Road
Sutton SM2 5NG
UK
Ian J. Bruno
Cambridge Crystallographic Data Centre (CCDC)
12 Union Road
Cambridge CB2 1EZ
UK
Mike Devereux
University of Basel
Klingelbergstrasse 80
4056 Basel
Switzerland
Peter Ertl
Novartis Institutes for BioMedical Research
Novartis Campus
4056 Basel
Switzerland
Marcus Gastreich
BioSolveIT
An der Ziegelei 79
53757 St. Augustin
Germany
Valerie J. Gillet
The University of Sheffield
Information School
Regent Court
211 Portobello
Sheffield S1 4DP
UK
Colin R. Groom
Cambridge Crystallographic Data Centre (CCDC)
12 Union Road
Cambridge CB2 1EZ
UK
Julian Hayward
Digital Chemistry Ltd
30 Kiveton Lane
Todwick, Sheffield S26 1HL
UK
John W. Liebeschuetz
Cambridge Crystallographic Data Centre (CCDC)
12 Union Road
Cambridge CB2 1EZ
UK
Nicholas A. Meanwell
Bristol-Myers Squibb Pharmaceutical Research and Development
Department of Medicinal Chemistry
5 Research Parkway
Wallingford, CT 06492
USA
David Millan
Sandwich Laboratories
Pfizer Global Research and Development
Ramsgate Road
Sandwich, Kent CT13 9NJ
UK
James E. J. Mills
Sandwich Laboratories
Pfizer Global Research and Development
Ramsgate Road
Sandwich, Kent CT13 9NJ
UK
Tjelvar J. Olsson
Cambridge Crystallographic Data Centre (CCDC)
12 Union Road
Cambridge CB2 1EZ
UK
George Papadatos
Eli Lilly Limited
Erl Wood Manor
Windlesham, Surrey GU20 6PH
UK
Paul L.A. Popelier
University of Manchester
Manchester Interdisciplinary Biocentre (MIB)
131 Princess Street
Manchester M1 7DN
UK
and
University of Manchester
School of Chemistry
Oxford Road
Manchester M13 9PL
UK
Matthias Rarey
ZBH University of Hamburg
Bundesstrasse 43
20146 Hamburg
Germany
Gisbert Schneider
ETH Zurich
Institute of Pharmaceutical Sciences
8093 Zurich
Switzerland
Jason Shanley
Abbott Laboratories
Global Pharmaceutical Research and Development
Department of Structural Biology
100 Abbott Park Road
Abbott Park, IL 60031
USA
Dennis A. Smith
Sandwich Laboratories
Pfizer Global Research and Development
Ramsgate Road
Sandwich, Kent CT13 9NJ
UK
Kent D. Stewart
Abbott Laboratories
Global Pharmaceutical Research and Development
Department of Structural Biology
100 Abbott Park Road
Abbott Park, IL 60031
USA
István Ujváry
iKem BT
Búza u. 32
1033 Budapest
Hungary
Peter Willett
University of Sheffield
Information School
Sheffield S1 4DP
UK
Preface
Bioisosteric replacement of substituents, ring atoms, linkers, and other groups aims to generate chemical substitutes with related biological properties, in the hope that the new analogues may have somewhat better properties. Such replacements are the toolbox of medicinal chemists to optimize their lead structures with respect to lipophilicity, solubility, activity, selectivity, absorption, metabolism, and lack of toxic and other side effects. Whenever an analogue with some improved properties is observed, the new compound is taken as the starting point for further modification. In this evolutionary procedure, either a preclinical or a clinical candidate results or the project has to be terminated, without success. Whereas the whole process quite often follows a trial and error procedure, certain empirical rules developed in medicinal chemistry. Very simple ones are, for example, the replacement of a hydrogen atom in the para-position of a benzene ring, to avoid rapid metabolic degradation, or, on the other hand, the introduction of an aromatic methyl group instead of a chlorine atom, to avoid too long biological half-life. More sophisticated rules exist for modification of the ligands of certain targets, for example, proteases or kinases.
The organization of this book follows a logical sequence, starting with Part One on the principles of bioisosterism, including an introductory chapter, and chapters on classical bioisosteres in medicinal chemistry and the logical but often surprising consequences of bioisosteric replacement. Part Two presents a database on bioisosteres and bioanalogues and discusses the search for bioisosteres, using the Cambridge Structure Database of 3D structures of small molecules, as well as the mining of bioisosteric pairs. Part Three presents methods to identify bioisosteres under the aspect of physicochemical properties, molecular topology, molecular shape, and protein 3D structures. Part Four describes a computer program for drug design, using medicinal chemistry rules, discusses the bioisosteric modification of a receptor antagonist, and ends with a concluding chapter on perspectives from medicinal chemistry.
Whereas some reviews on bioisosteres are found in the literature, as well as chapters in medicinal chemistry books, no dedicated monograph on bioisosteres has been published so far. Thus, we are very grateful to Nathan Brown for editing such a book, which will help novices in the field as well as experienced scientists to manage lead structure optimization in an even more rational manner. In addition, we are very much indebted to Frank Weinreich and Heike Nöthe, both at Wiley-VCH. Their support and ongoing engagement, not only for this book but also for the whole series “Methods and Principles in Medicinal Chemistry,” adds to the success of this excellent collection of monographs on various topics, all related to drug research.
March 2012Raimund MannholdDüsseldorfHugo KubinyiWeisenheim am SandGerd FolkersZürichA Personal Foreword
“Hamlet: Do you see yonder cloud that's almost in shape of a camel?
Polonius: By th' Mass, and 'tis like a camel, indeed.
Hamlet: Methinks it is like a weasel.
Polonius: It is backed like a weasel.
Hamlet: Or like a whale.
Polonius: Very like a whale.”
Hamlet, Act III, Scene II
William Shakespeare
The essence of design is the identification of appropriate constituents and their careful arrangement in sympathy with the requirements of the desired object. The same principles apply in drug design, where the components are elements and elemental groups, and their arrangement is achieved through the synthetic organic chemistry that is undertaken. The ultimate requirement in the design of new drugs is an entity that summons a physiological response of benefit to the patient.
In this book, we cover the key aspects of drug design through the identification and replacement of bioisosteric groups within the context of the drug design ethic. Bioisosterism is a phenomenon where molecular groups are functionally similar, that is, they have a similar biological effect, while modulating other properties.
This is the first book to provide a general overview of the field of bioisosterism at a time when its application has become a formal process. There are now many information sources and design tools available to assist the medicinal chemist in the identification of relevant bioisosteres.
The first part of this book covers the historical aspects of bioisosterism, from its founding principles of isosterism from Langmuir through defined sets of classical isosteres and bioisosteres, to the potential consequences of bioisosteric replacement in context.
A considerable amount of knowledge has been collated in recent years, in large molecular databases with metadata that can be analyzed and brought to bear in bioisosteric replacement. Knowledge-based methods form the second part, covering experimentally determined bioisosteric replacements from the medicinal chemistry literature; small-molecule crystal identification of bioisosteres; and mining unknown bioisosteres from these databases through the application of recently developed methods for their identification.
One can describe a molecule in many ways and the same applies to bioisosteres. Molecular descriptor methods are covered in the third part by the application of different representations. A number of computational approaches to bioisosteric replacement are covered in chapters on physicochemical properties, molecular topology, molecular shape, and the use of protein structure information. Each chapter covers many common methods and overviews of when best to apply these methods, and where they have been successfully applied.
This book concludes with two case studies of where bioisosteric replacement strategies have been applied in drug discovery, to provide demonstrable evidence of their utility. Finally, a few leading scientists in this field have kindly provided personal perspectives on bioisosterism and its relevance to drug discovery.
My sincere wish is that you enjoy reading this book as much as I did working with the very talented team of scientists who contributed chapters. I would also like to thank the publishing team and the series editors for their help in bringing this book together.
London, 2012
Nathan Brown
Part One
Principles
Chapter 1
Bioisosterism in Medicinal Chemistry
Nathan Brown
1.1 Introduction
One of the key challenges for the medicinal chemist today is the modulation and mediation of the potency of a small-molecule therapeutic against its biological target. In addition, it is essential to ensure that the molecule reaches its target effectively while also ensuring that it satisfies necessary safety requirements. One of the most significant approaches to assist in efficiently navigating the available chemistry space is that of bioisosteric replacement.
This book, the first dedicated solely to the subject of bioisosterism, covers the field from the very beginning to its development as a reliable and well-used approach to assist in drug design. This book is split into four parts. The first part covers the principles and theory behind isosterism and bioisosterism. The second part investigates methods that apply knowledge bases of experimental data from a variety of sources to assist in decision making. The third part reports on the four main computational approaches to bioisosteric identification and replacement using molecular properties, topology, shape, and protein structure. This book concludes with real-world examples of bioisosterism in application and a collection of reflections and perspectives on bioisosteric identification and replacement from many of the current leaders in the field.
This chapter provides an overview of the history of bioisosterism from its beginning in the early twentieth century to the present day. We also provide an overview of the importance of judicious bioisosteric replacement in lead optimization to assist in the path toward a viable clinical candidate and, ultimately, a drug.
1.2 Isosterism
James Moir [1] first considered isosterism in all but name, in 1909. It was not until 1919 that the term isosterism was given to this phenomenon by Irving Langmuir [2] in his landmark paper “Isomorphism, isosterism and covalence.” The focus of this early isosterism work was on the electronic configuration of atoms. Langmuir used experiment to identify the correspondence between the physical properties of different substances. Langmuir, in accordance with the octet rule where atoms will often combine to have eight electrons in their valence shells, compared the number and arrangement of electrons between nitrogen, carbon monoxide, and the cyanogen ion and identified that these would be the same. This relationship was demonstrated to be true between nitrogen and carbon monoxide in terms of their physical properties. The same similarities were also reported between nitrous oxide and carbon dioxide when taking experimental data from Landolt–Börnstein's tables and Abegg's handbook (Table 1.1).
Table 1.1 Experimental data from Landolt–Börnstein's tables and Abegg's handbook for nitrous oxide (N2O) and carbon dioxide (CO2)
PropertyN2OCO2Critical pressure (atm)7577Critical temperature (°C)35.431.9Viscosity at 20 °C148 × 10−6148 × 10−6Heat conductivity at 100 °C0.05060.0506Density of liquid at −20 °C0.9961.031Density of liquid at + 10 °C0.8560.858Refractive index of liquid, D line, 16 °C1.1931.190Dielectric constant of liquid at 0 °C1.5981.582Magnetic susceptibility of gas at 40 atm, 16 °C0.12 × 10−60.12 × 10−6Solubility in water at 0 °C1.3051.780Solubility in alcohol at 15 °C3.253.13However, Langmuir identified one distinct property that is substantially different between nitrous oxide and carbon dioxide, the freezing point: −102 and −56 °C, respectively. Evidence for this was assumed to be due to the freezing point being “abnormally sensitive to even slight differences in structure.”
With this observation of the correlation between the structure and arrangement of electrons with physical properties, Langmuir defined the neologism calling them isosteres, or isosteric compounds. Langmuir defined isosterism as follows:
“Comolecules are thus isosteric if they contain the same number and arrangement of electrons. The comolecules of isosteres must, therefore, contain the same number of atoms. The essential differences between isosteres are confined to the charges on the nuclei of the constituent atoms. Thus in carbon dioxide the charges on the nuclei of the carbon and oxygen atoms are 6 and 8, respectively, and there are 2 × 8 + 6 = 22 electrons in the molecule. In nitrous oxide the number of charges on the nitrogen nuclei is 7, but the total number of electrons in the molecule is again 2 × 7 + 8 = 22. The remarkable similarity of the physical properties of these two substances proves that their electrons are arranged in the same manner.”
The list of isosteres that Langmuir described in 1919 is given in Table 1.2. Langmuir extended his concept of isosterism to predicting likely crystal forms using sodium and fluorine ions as exemplars, these having been solved by William Henry Bragg and William Lawrence Bragg – father and son who were together awarded the Nobel Prize for Physics in 1915. Since the magnesium and oxygen ions are isosteric with the sodium and fluorine ions, it follows that magnesium oxide will have a crystal structure that is identical to that of sodium fluoride.
Table 1.2 List of isosteres defined by Langmuir in 1919.
TypeIsosteres1H−, He, Li+2O2−, F−, Ne, Na+, Mg2+, Al3+3S2−, Cl−, A, K+, Ca2+4Cu+, Zn2+5Br−, Kr, Rb+, Sr2+6Ag+, Cd2+7I−, Xe, Cs+, Ba2+8N2, CO, CN−9CH4, NH4+10CO2, N2O, N3−, CNO−11NO3−, CO32−12NO2−, O313HF, OH−14ClO4−, SO42−, PO43−15ClO3−, SO42−, PO43−16SO3, PO3−17S2O62−, P2O64−18S2O72−, P2O74−19SiH4, PH4+20MnO4−, CrO42−21SeO42−, AsO43−In 1925, H.G. Grimm [3] extended the concept of isosterism, introduced by Langmuir, with Grimm's hydride displacement law:
“Atoms anywhere up to four places in the periodic system before an inert gas change their properties by uniting with one to four hydrogen atoms, in such a manner that the resulting combinations behave like pseudoatoms, which are similar to elements in the groups one to four places, respectively, to their right.”
Therefore, according to this law, the addition of hydrogen to an atom will result in a pseudoatom with similar properties to the atom of the next highest atomic number. So, CH is isosteric with N and NH is isosteric with O and so on.
Beginning in 1932, Friedrich Erlenmeyer [4, 5] extended the concepts from Grimm further and the first applications of isosterism to biological systems. Erlenmeyer redefined isosteres as:
“. . .elements, molecules or ions in which the peripheral layers of electrons may be considered identical.”
In addition, Erlenmeyer also proposed the following three additions to the concept of isosteres:
It was with Erlenmeyer that the concept of bioisosterism was introduced to differentiate from classical isosteres, ensuring its relevance to medicinal chemistry. The introduction of ring equivalences is significant. This was the formalization of what we consider to be a bioisosteric comparison and is the first definition of most relevance to medicinal chemistry.
1.3 Bioisosterism
Classical isosteres are traditionally categorized into the following distinct groupings [6]:
A number of classical bioisosteric examples are provided in Table 1.3 that illustrate typical replacements possible in each of these five groups.
Table 1.3 Some examples of classical bioisosteres – groups in each row are equivalent.
Monovalent bioisosteresF, HOH, NHF, OH, NH, or CH3 for HSH, OHCl, Br, CF3Divalent bioisosteres–C=S, –C=O, –C=NH, –C=C–Trivalent atoms or groups–CH=, –N=–P=, –As=Tetrasubstituted atomsRing equivalentsHowever, more recent definitions of isosterism, and more specifically bioisosterism, relax these constraints and permit bioisosteric pairings between moieties that do not necessarily contain the same number of atoms. Specifically, nonclassical bioisosteres include the addition of the following two groups:
The origins of classical isosterism focused largely on the electronic similarity of groups rather than their functional similarity. As investigation into the field progressed, it became obvious that these very defined rules on isosterism, although powerful, were restrictive in particular to medicinal chemistry. The addition of the latter two groups for nonclassical bioisosteres permitted the mimicking of spatial arrangements, electronic properties, or another physicochemical property that is important for biological activity.
In extending and broadening the purer rules of classical isosterism, two scientists are credited with progressing the field of bioisosterism: Friedman and Thornber. In 1951, Friedman [7] provided the first definition closest to what we call bioisosterism today:
“[bioisosteres are structural moieties] which fit the broadest definition of isosteres and have the same type of biological activity.”
With this definition, the generalization of what constitutes bioisosterism was formed. However, this definition really only considers the macromolecular recognition of bioisosteres, which is of course highly important, but largely ignores the specifics of the numerous other physicochemical properties that are optimized in a medicinal chemistry project. Friedman's definition was followed in 1979 with the much less specific definition from Thornber [8] of bioisosteres and nonclassical bioisosteres:
“Bioisosteres are groups or molecules which have chemical and physical similarities producing broadly similar biological properties.”
At first reading, this definition looks somewhat similar to Friedman's, but it is the relevant importance of chemical and physical similarities that differentiates this from Friedman's definition. In addition to this definition, Thornber also defined eight parameters that could be considered in making an alteration to a structural moiety to elicit a bioisosteric pairing:
Depending on the particular property that is modified by a bioisosteric replacement, the result will typically fall into one or more of the following:
These four key generalized parameters, with specific properties governing the optimization of each, provide what can be formalized as the changes that may be made in lead optimization to provide guidance on the optimization of functional groups that are bioisosteric.
In 1991, Alfred Burger [9] defined bioisosterism as:
“Compounds or groups that possess near-equal molecular shapes and volumes, approximately the same distribution of electrons, and which exhibit similar physicochemical properties. . .”
Burger's definition succinctly defines bioisosteres including all of the aforementioned extensions defined by other scientists in the field. The next section focuses on the specific improvements in lead optimization that can be gained by prudent application of the concepts of bioisosterism.
1.4 Bioisosterism in Lead Optimization
One of the processes where bioisosteric replacement can have a substantial impact, particularly in the discovery of a novel small-molecule therapeutic, is in the lead optimization stage of a drug discovery project. Once a lead molecule has been identified, the medicinal chemist is faced with the considerable challenge of making small, defined changes to an identified core structure (also chemotype or scaffold) by the addition or substitution of functional groups to test specific hypotheses. While the challenge of scaffold hopping (the replacement of the functional or specific exit geometries of a molecular scaffold) is important, this challenge will only be considered as a subset of bioisosteric replacement in this book [14–18].
1.4.1 Common Replacements in Medicinal Chemistry
When considering a medicinal chemistry project where a lead molecule has been identified, and also chemical handles, to permit the synthesis of many analogues, the project team will identify substituents that are potential bioisosteric replacements using a number of different methods. Many of these methods will be discussed in Parts Two and Three of this book from the literature and in silico modeling approaches, respectively. Southall and Ajay [10] reported a number of common medicinal chemistry bioisosteric replacements from kinase drug candidates (Table 1.4). Sildenafil (Viagra) Vardenafil (Levitra) [PDE5 Inhibitor: Pfizer Bayer AG, SP, GSK] Ciprofloxacin (Proquin) Levofloxacin (Tavanic) [Antibacterial: Bayer AG Sanofi-Aventis] Gefitinib (Iressa) Erlotinib (Tarceva) [EGFR Inhibitor: AZ Roche/ISI].
Table 1.4 Common replacements in medicinal chemistry taken from the literature [10]
1.4.2 Structure-Based Drug Design
It is becoming increasingly common that protein–ligand cocrystal structures are available to assist early on in a drug design project. The inclusion of structural information allows the design of molecules that take into account what may or may not be tolerated in a particular position, according to the conformations of key protein structure residues. This is in contrast to only using the information within the ligands that have already been synthesized and tested. The latter can lead to the assumption that the bioisosteric replacement must have the same bulk properties as the original group or, more frequently, lead to inefficiency in the design process through the unnecessary synthesis of molecules that function only to probe functional group tolerability at different positions on a molecule.
The application of protein structures to suggest bioisosteric replacements will be covered more fully in Chapter 10.
1.4.3 Multiobjective Optimization
As has been discussed previously, lead optimization involves the separate, although sometimes simultaneous, optimization of multiple parameters. When considering replacement of key functional groups around a common molecular scaffold, the chemical space of potential molecules that could be synthesized (assuming no issues in terms of synthetic accessibility, stability, etc.) is the product of the number of feasible replacement groups at each substitution point on the molecular scaffold. For example, a project with one chemical scaffold that has three points of variation, using a conservative set of 50 possible monomers at each substitution point, generates a potential project chemical space (i.e., the set of all molecules that could be synthesized) of 125 000. Typically, a medicinal chemistry project can only realize the synthesis of a small proportion of these virtual compounds, for example, approximately 1%. Therefore, the design of which molecules to synthesize and test is of great importance to ensure that those molecules are most likely to fulfill the design objectives.
To effectively and efficiently propose the most appropriate molecules for synthesis, two key points should be considered by the project team: exploration and exploitation. Exploration uses a molecular diversity measure to efficiently cover the space of virtual molecules with an even distribution of known properties. This leads to a high confidence that the entirety of the space is represented with as few molecules as necessary to demonstrate regions of specific interest. This can be achieved using a wide variety of diversity selection algorithms [11]. Here, the question being asked is that of the entirety of the chemical space.
The coverage of diversity must also be balanced with the synthesis of very close analogues to finesse those properties that are important for that specific project, many of which have been defined already in this chapter. Here, the investigation is directed on small and specific changes, most often a number of single alterations that enhance the understanding of the local structure–activity relationship (SAR). It is with this part of the lead optimization process that bioisosteric replacements are most important, as opposed to the diversity design where bioisosteric replacements will not necessarily provide sufficient information about the global chemical space [13].
Bioisosteric replacement is often considered when the aims are to maintain enzyme potency while optimizing additional properties, such as cellular penetration, solubility, metabolism, toxicity, and so on. This principle is often referred to as multiobjective optimization (MOOP) or multiparameter optimization (MPO) [12]. There are many ways in which one can address multiple objectives, but it is important to understand the landscape of the trade-off surface between each of the important objectives, including an understanding of parameters that may be correlated with each other (Figure 1.1).
Figure 1.1 Schematic of multiobjective optimization in a drug discovery project optimizing potency and pharmacokinetic properties over time. Initially, the emphasis is on potency in this schematic and significant improvement was made in this respect. As time progresses and additional characterization of the molecules is realized, optimizing the pharmacokinetics becomes increasingly important. However, the various different parameters should be optimized simultaneously to ensure progression to a clinical candidate.
The combination of identifying bioisosteric replacements in a lead molecule together with the multiobjective prioritization of virtual molecules in that chemical series for synthesis provides the medicinal chemist with the key information for making design decisions in a therapeutic project. The approaches to identifying these replacements will be covered in Parts Two and Three of this book, but they can all be applied in this challenge.
1.5 Conclusions
The origins of isosterism have been traced back to the early twentieth century, most notably in the work of Langmuir, which also gave the concept its name. The extension of isosterism through Grimm and Erlenmeyer paved the way to the definition of bioisosterism, largely promulgated by Friedman and Thornber. Moving from a definition of isosterism that focused specifically on the electronic makeup of isosteres to a more functional outlook in terms of biological properties was a major step forward toward what we today call bioisosterism.
Bioisosterism is now one of the most important tools that medicinal chemists have at their disposal. Through shrewd application of bioisosteres that have experimental precedent or have been identified by theoretical calculations, the medicinal chemist is now well prepared with highly effective tools that have been demonstrated to be of great utility in therapeutic design programs. The remaining chapters in this part will detail the key theories behind bioisosteres and their replacement.
References
1. Meanwell, N.A. (2011) Synopsis of some recent tactical application of bioisosteres in drug design. Journal of Medicinal Chemistry, 54, 2529–2591.
2. Langmuir, I. (1919) Isomorphism, isosterism and covalence. Journal of the American Chemical Society, 41, 1543–1559.
3. Grimm, H.G. (1925) On construction and sizes of non-metallic hydrides. Zeitschrift fur Elektrochemie und Angewandte Physikalische Chemie, 31, 474; 1928, 34, 430; 1934, 47, 553–594.
4. Erlenmeyer, H. and Berger, E. (1932) Studies on the significance of structure of antigens for the production and the specificity of antibodies. Biochemical Zoology, 252, 22–36.
5. Erlenmeyer, H., Berger, E., and Leo, M. (1933) Relationship between the structure of antigens and the specificity of antibodies. Helvetica Chimica Acta, 16, 733–738.
6. Lemke, T.L. and Williams, D.A. (2007) Foye's Principles of Medicinal Chemistry, 6th edn, Lippincott Williams & Wilkins, Baltimore, MD.
7. Friedman, H.L. (1951) Influence of isosteric replacements upon biological activity. NAS-NRS Publication No. 206, NAS-NRS, Washington, DC, pp. 295–358.
8. Thornber, C.W. (1979) Isosterism and molecular modification in drug design. Chemical Society Reviews, 8, 563–580.
9. Burger, A. (1991) Isosterism and bioisosterism in drug design. Progress in Drug Research, 37, 288–362.
9. Southall, N.T. and Ajay (2006) Kinase patent space visualization using chemical replacements. Journal of Medicinal Chemistry, 49, 2103–2109.
11. Gillet, V.J. (2011) Diversity selection algorithms. Wiley Interdisciplinary Reviews: Computational Molecular Science, 1, 580–589.
12. Nicolaou, C.A., Brown, N., and Pattichis, C.K. (2007) Molecular optimization using multi-objective methods. Current Opinion in Drug Discovery & Development, 10, 316–324.
13. Brown, N. and Lewis, R.A. (2006) Exploiting QSAR methods in lead optimization. Current Opinion in Drug Discovery & Development, 9, 419–424.
14. Ciapetti, P. and Giethlen, B. (2008) Molecular variations based on isosteric replacements, in The Practice of Medicinal Chemistry, 3rd edn (ed. C.G. Wermuth), Elsevier.
15. DiMasi, J. and Faden, L.B. (2011) Competitiveness in follow-on drug R&D: a race of imitation? Nature Reviews Drug Discovery, 10, 23–27.
16. Langdon, S.R., Ertl, P., and Brown, N. (2010) Bioisosteric replacement and scaffold hopping in lead generation and optimization. Molecular Informatics, 29, 366–385.
17. Patani, G.A. and LaVoie, E.J. (1996) Bioisosterism: a rational approach in drug design. Chemical Reviews, 96, 3147–3176.
18. Wermuth, C.G. (2006) Similarity in drugs: reflections on analogue design. Drug Discovery Today, 11, 348–354.
Chapter 2
Classical Bioisosteres
Caterina Barillari and Nathan Brown
2.1 Introduction
The discovery and development of a candidate for clinical evaluation is a long process that involves small modifications to a lead compound to improve some of its properties, such as pharmacological activity, selectivity, and pharmacokinetics. This is often achieved by the medicinal chemists by replacing a functional group with groups sharing similar physical or chemical properties and maintaining similar activity, which are defined as bioisosteres. We will hereby provide a historical overview of the development and evolution of the concepts of isosterism and bioisosterism, followed by a selection of successful examples of bioisosteric modifications reported in the literature.
2.2 Historical Background
The concept of isosterism was first introduced by Langmuir in 1919 to describe molecules that contain the same number and arrangement of electrons and have similar physicochemical properties [1]. Langmuir identified 21 types of isosteres, a few examples of which are reported in Table 2.1.
Table 2.1 Examples of isosteres identified by Langmuir.
TypeIsosteres1H−, He, Li+2O2−, F−, Ne, Na+, Mg2+, Al3+3S2−, Cl−, A, K+, Ca2+8N2, CO, CN−9CH4, NH4+21SeO42−, AsO43−In 1925, Grimm formulated his “hydride displacement law,” which states that the addition of a hydride to an atom produces a pseudoatom with the same physical properties as those present in the column immediately behind in the periodic table, as shown in Table 2.2 [2].
Table 2.2 Isosteres based on Grimm's hydride displacement law.
The concept of isosterism was later broadened by Erlenmeyer in 1932 to include elements, ions, or molecules with the same number of electrons at the valence level (Table 2.3) [3]. Erlenmeyer stated that elements in the same column of the periodic table are isosteres among themselves and also introduced the concept of electronically equivalent rings.
Table 2.3 Isosteres as defined by Erlenmeyer.
The term bioisosterism was introduced in 1952 by Friedman to describe structurally related substances with similar or antagonistic biological properties [4]. This term was later broadened by Thornber to include “groups or molecules that have chemical and physical similarities producing broadly similar biological properties” [5].
Finally, in 1970, Alfred Burger classified bioisosteres into classical and nonclassical [6]. The former include atoms or groups of the same valence as well as ring equivalents, while the latter are basically those that do not fit the first definition. Several reviews on bioisosteres have been reported in the literature over the years [7–11], and in the next sections a selection of examples for each of the two categories will be provided.
2.3 Classical Bioisosteres
2.3.1 Monovalent Atoms and Groups
One of the most common monovalent isosteric replacements is the substitution of hydrogen with fluorine [7]. These atoms have similar van der Waals radii but different electronic effects, fluorine being the most electronegative element in the periodic table. Due to the high strength of the C–F bond, fluorine is often introduced to achieve metabolic stability. Moreover, due to its high electronegativity, fluorine can be introduced to reduce basicity of proximal amines or increase acidity of proximal acids and also to introduce a conformational bias in molecules.
One of the most well-known examples of effective replacement of hydrogen with fluorine is observed in the antineoplastic drug 5-fluorouracil (Figure 2.1). This compound is metabolized in vivo to 5-fluoro-2′-deoxyuridylic acid (5-fluoro-dUMP), which is the active drug that covalently binds to thymidylate synthase, the enzyme responsible for the essential conversion in DNA synthesis of uridylic acid to thymidylic acid.
Figure 2.1 Bioisosteric H/F replacement in 5-fluorouracil.
2.3.2 Bivalent Atoms and Groups
A successful use of bivalent bioisosteres can be found in derivatives of the antihypertensive drug rilmenidine (Figure 2.2), where they were employed as a way of finding similar compounds with reduced side effects [12]. Rilmenidine exerts its activity by binding to the I1 imidazoline receptors (I1Rs), but it also binds to the α2-adrenoreceptors (α2ARs), which is considered responsible for the side effects. As shown in Figure 2.2, several bivalent bioisosteres of rilmenidine were able to maintain I1R binding, while losing affinity on the α2-adrenoreceptors, thus reducing undesired side effects.
Figure 2.2 Bivalent bioisosteric derivatives of rilmenidine.
2.3.3 Trivalent Atoms and Groups
An application of trivalent bioisosterism can be found in the development of hypocholesterolemic agents, where Counsell et al. replaced two CH groups present in cholesterol with two nitrogen atoms (Figure 2.3) [13]. This resulted in 20,25-diazacholesterol, which is a potent inhibitor of cholesterol biosynthesis.
Figure 2.3 Cholesterol and its bioisosteric derivative 20,25-diazacholesterol.
2.3.4 Tetravalent Atoms
A recent example of bioisosteric replacement of a carbon atom with silicon was reported by Warneck and colleagues at Paradigm Therapeutics, who synthesized the silicon derivative of the p38 MAP kinase inhibitor BIRB-796 (Figure 2.4), which is in clinical evaluation for several inflammatory diseases, such as rheumatoid arthritis, Crohn's disease, and psoriasis [14]. This silicon isostere, compound 1, was found to be unusually less lipophilic than BIRB-796, of comparable potency, and more metabolically stable in human liver microsomes.
Figure 2.4 Example of silicon bioisosteric replacement.
2.3.5 Ring Equivalents
The substitution of phenyl ring with pyridine is widely used to improve metabolic stability. One efficient example of this can be recently found in HIV-1 inhibitors developed at Bristol-Myers Squibb and currently in clinical development. Compound 2 (Figure 2.5) was identified as lead in the discovery of drugs that inhibit the attachment of HIV to CD4 host cells but, due to high metabolism and low solubility, isosteres of the phenyl ring were evaluated. This led to the identification of BMS-488043 (Figure 2.5), which retained potency against HIV-1 attachment, but had better pharmacokinetic profile than compound 2 and was thus advanced in clinical trials [15].
Figure 2.5 Phenyl/pyridine bioisosteric replacement leading to clinical candidate BMS-488043.
2.4 Nonclassical Bioisosteres
2.4.1 Carbonyl Group
Modafinil [16] (Figure 2.6) is a widely used drug in the treatment of excessive sleepiness caused by narcolepsy, shift work sleep disorder, and obstructive sleep apnea. The mechanism of action of this drug is still uncertain, but it is believed to work in a localized manner using hypocretin, histamine, epinephrine, γ-aminobutyric acid, and glutamate. This compound has an asymmetric sulfoxide group, and it is currently marketed as a racemate. Studies have shown that the sulfone derivative of modafinil retains similar activity to the parent compound, as well as not showing increase in toxicity. De Risi et al. investigated the replacement of the sulfoxide group with a carbonyl to facilitate synthesis and to remove problems associated with chirality. Compound 3 showed a slight loss of activity compared to modafinil, but this was restored when the amide function was modified in compound 4 (Figure 2.6).
Figure 2.6 Bioisosteric replacement of modafinil.
2.4.2 Carboxylic Acid
CO2HSO2NHRSO3HPO(OH)NH2PO(OH)OEtCONHCNIsosteres of carboxylic acids are often sought to enhance pharmacokinetic properties, by reducing polarity and increasing lipophilicity in order to increase membrane permeability. In 2010, Hadden et al. at Merck Research Laboratories reported the synthesis and activity of agonists of the G-protein-coupled receptor bombesin receptor subtype-3 (BB3), the lack of which has been associated with obesity, hypertension, and diabetes in genetically altered mice [17]. Compound 5 (Figure 2.7) is a potent inhibitor of BB3 obtained from a combination of high-throughput screening and SAR development. In order to improve oral bioavailability and brain penetration, a series of bioisosteres of the carboxylic acid in compound 5 were later synthesized. Despite maintaining good in vitro potency and improving oral bioavailability, all compounds failed to increase brain penetration: a partially successful bioisosteric replacement.
Figure 2.7 Carboxylic acid bioisosteres in BB3 inhibitors.
2.4.3 Hydroxyl Group
In 2005, Wu et al. at the Schering-Plough Research Institute reported the synthesis and biological evaluation of phenol bioisosteric analogues of benzazepine D1/D5 antagonists (Figure 2.8) [18]. SCH 23390 and SFK 38393 represented a major breakthrough in the pharmacology of dopamine receptors, the first being a high-affinity and selective D1/D5 antagonist and the second being a partial agonist. SCH 39166 has undergone several clinical trials, including schizophrenia, cocaine addiction, and obesity. However, all three compounds present pharmacokinetic issues.
Figure 2.8 Benzazepine D1/D5 antagonists.
For SCH 39166, this liability is mainly due to O-glucuronidation of the phenol and N-dealkylation of the NCH3 group; thus, several heterocyclic rings containing an N–H hydrogen bond donor as isosteres of the phenol group were investigated. Four compounds (Figure 2.9) showed good pharmacokinetic profiles, while maintaining good antagonist activity on the D1/D5 receptors.
Figure 2.9 Bioisosteric D1/D5 antagonists with good pharmacokinetic profiles.
2.4.4 Catechol
Catechol bioisosteres are often utilized to overcome pharmacokinetic and toxicological issues linked to this moiety. Successful examples of bioisosteric replacement of catechols can be found in catecholamines. Benzimidazole analogues of the adrenergic agonists norepinephrine and isoproterenol (Figure 2.10) were synthesized by Arnett et al.
