Fragment-based Drug Discovery Literature to September 2007           © IOTA Pharmaceuticals Ltd

Authors

Journal

Title

Abstract

Abad-Zapatero C, Metz JT.

Drug Discov Today. 2005 Apr 1;10(7):464-9. No abstract available.

Ligand efficiency indices as guideposts for drug discovery.

Successful drug discovery involves the optimization of many variables, such as compound potency, selectivity and solubility. Drug development is an arduous process, beginning with lead selection and continuing through to preclinical testing in animal models. Efficient navigation through these steps should be facilitated by reducing the number of variables.

Adrian FJ, Ding Q, Sim T, Velentza A, Sloan C, Liu Y, Zhang G, Hur W, Ding S, Manley P, Mestan J, Fabbro D, Gray NS.

Nat Chem Biol. 2006 Feb;2(2):95-102. Epub 2006 Jan 15.

Allosteric inhibitors of Bcr-abl-dependent cell proliferation.

Chronic myelogenous leukemia (CML) is a myeloproliferative disorder characterized at the molecular level by the expression of Bcr-abl, a 210-kDa fusion protein with deregulated tyrosine kinase activity. Encouraged by the clinical validation of Bcr-abl as the target for the treatment of CML by imatinib, we sought to identify pharmacological agents that could target this kinase by a distinct mechanism. We report the discovery of a new class of Bcr-abl inhibitors using an unbiased differential cytotoxicity screen of a combinatorial kinase-directed heterocycle library. Compounds in this class (exemplified by GNF-2) show exclusive antiproliferative activity toward Bcr-abl-transformed cells, with potencies similar to imatinib, while showing no inhibition of the kinase activity of full-length or catalytic domain of c-abl. We propose that this new class of compounds inhibits Bcr-abl kinase activity through an allosteric non-ATP competitive mechanism.

Andrews MJ, Kontopidis G, McInnes C, Plater A, Innes L, Cowan A, Jewsbury P, Fischer PM.

Chembiochem. 2006 Dec;7(12):1909-15.

REPLACE: a strategy for iterative design of cyclin-binding groove inhibitors.

We describe a drug-design strategy termed REPLACE (REplacement with Partial Ligand Alternatives through Computational Enrichment) in which nonpeptidic surrogates for specific determinants of known peptide ligands are identified in silico by using a core peptide-bound protein structure as a design anchor. In the REPLACE application example, we present the effective replacement of two critical binding motifs in a lead protein-protein interaction inhibitor pentapeptide with more druglike phenyltriazole and diphenyl ether groups. These were identified through docking of fragment libraries into the volume of the cyclin-binding groove of CDK2/cyclin A vacated through truncation of the inhibitor peptide-binding determinants. Proof of concept for this strategy was obtained through the generation of potent peptide-small-molecule hybrids and by the confirmation of inhibitor-binding modes in X-ray crystal structures. This method therefore allows nonpeptide fragments to be identified without the requirement for a high-sensitivity binding assay and should be generally applicable in replacing amino acids as individual residues or groups in peptide inhibitors to generate pharmaceutically acceptable lead molecules.

Aronov AM, Bemis GW.

Proteins. 2004 Oct 1;57(1):36-50.

A minimalist approach to fragment-based ligand design using common rings and linkers: application to kinase inhibitors.

We present a novel method for stepwise scaffold assembly that integrates fragment-by-fragment ligand design approaches with high-throughput virtual library screening (COREGEN). As an extension of our earlier studies of common features present in drug molecules, we investigate the hypothesis that most pharmaceutically interesting ligands can be expressed in terms of the ring-linker frameworks that comprise them. Analysis of 119 published kinase inhibitors from at least 18 different targets illustrates that a basis set of 4 rings and 8 linkers is sufficient to describe approximately 90% of ring and linker occurrences, respectively. A similar result was derived from a larger set of approximately 40,000 kinase inhibitors from curated patents. A method for ring-linker-based assembly of scaffold libraries that uses experimental information to guide the placement of anchor fragments is validated using a set of reported kinase inhibitors of Bcr-Abl, Cdk2, and Src. In every case, the predominant structural motif of reported ligand cores is reproduced and variations are suggested. To underscore generality of this approach, a novel scaffold for a cyclooxygenase-2 (COX-2) selective ligand is proposed.

Babaoglu K, Shoichet BK.

Nat Chem Biol. 2006 Dec;2(12):720-3. Epub 2006 Oct 29.

Deconstructing fragment-based inhibitor discovery.

Fragment-based screens test multiple low-molecular weight molecules for binding to a target. Fragments often bind with low affinities but typically have better ligand efficiencies (DeltaG(bind)/heavy atom count) than traditional screening hits. This efficiency, combined with accompanying atomic-resolution structures, has made fragments popular starting points for drug discovery programs. Fragment-based design adopts a constructive strategy: affinity is enhanced either by cycles of functional-group addition or by joining two independent fragments together. The final inhibitor is expected to adopt the same geometry as the original fragment hit. Here we consider whether the inverse, deconstructive logic also applies--can one always parse a higher-affinity inhibitor into fragments that recapitulate the binding geometry of the larger molecule? Cocrystal structures of fragments deconstructed from a known beta-lactamase inhibitor suggest that this is not always the case.

Bailey D, Zanders E, Dean P.

Pharmacogenomics J. 2001;1(1):38-47.

Site-specific molecular design and its relevance to pharmacogenomics and chemical biology.

The emergence of the new discipline of pharmacogenomics reflects the growing convergence of chemical and genomic space. The massive information-driven growth in both computational chemistry and structural biology is leading to unprecedented opportunities in both chemical and biological design. In this paper we relate current opinion in structural biology to recent developments in computational drug design. Sequence information now permits protein structure prediction and, together with experimental protein structure determination, a complete database of ligand-binding sites and protein-protein interactions can be assembled. When aligned with site exploration and virtual screening, this information provides a foundation for structure-based pharmacogenomics. In association with chemical genomics, structure-based design will allow major new insights into a compound's biological and pharmaceutical properties.

Bailey D, Zanders E, Dean P.

Nat Biotechnol. 2001 Mar;19(3):207-9. No abstract available.

The end of the beginning for genomic medicine.

[No abstract]

Barakat MT, Dean PM.

J Comput Aided Mol Des. 1995 Oct;9(5):457-62.

The atom assignment problem in automated de novo drug design. 5. Tests for envelope-directed fragment placement based on molecular similarity.

The fragment placement method has been successfully extended to the problem of envelope-directed design. The atom assignment paradigm was based on molecular similarity between two molecular structures. A composite supersurface is defined to form the surface onto which the molecular fields are projected. The assignment process is then determined by using molecular similarity in the objective function to be optimized. In principle, this procedure is closely similar to that outlined in the previous paper for site-directed design. The rationale has been extensively tested on two benzodiazepine antagonists believed to bind to the same site.

Barakat MT, Dean PM.

J Comput Aided Mol Des. 1995 Oct;9(5):448-56.

The atom assignment problem in automated de novo drug design. 4. Tests for site-directed fragment placement based on molecular complementarity.

Three previous papers in this series have outlined an optimization method for atom assignment in drug design using fragment placement. In this paper the procedure is rigorously tested on a selection of five ligand-protein co-crystals. The algorithm is presented with the molecular graph of the ligand, and the electrostatic/hydrophobic potential of the site, with the aim of creating a placement on the molecular graph which is as electrostatically complementary or hydrophobically similar to the site as possible. Various designer options were tested, including, where appropriate, hydrogen bonding and a restricted number of halogens. In most cases, the placement obtained was at least as good as the native ligand, if not significantly better.

Barakat MT, Dean PM.

J Comput Aided Mol Des. 1995 Aug;9(4):359-72.

The atom assignment problem in automated de novo drug design. 3. Algorithms for optimization of fragment placement onto 3D molecular graphs.

Atom assignment onto 3D molecular graphs is a combinatoric problem in discrete space. If atoms are to be placed efficiently on molecular graphs produced in drug binding sites, the assignment must be optimized. An algorithm, based on simulated annealing, is presented for efficient optimization of fragment placement. Extensive tests of the method have been performed on five ligands taken from the Protein Data Bank. The algorithm is presented with the ligand graph and the electrostatic potential as input. Self placement of molecular fragments was monitored as an objective test. A hydrogen-bond option was also included, to enable the user to highlight specific needs. The algorithm performed well in the optimization, with successful replications. In some cases, a modification was necessary to reduce the tendency to give multiple halogenated structures. This optimization procedure should prove useful for automated de novo drug design.

Barakat MT, Dean PM.

J Comput Aided Mol Des. 1995 Aug;9(4):351-8.

The atom assignment problem in automated de novo drug design. 2. A method for molecular graph and fragment perception.

If atom assignment onto 3D molecular graphs is to be optimized, an efficient scheme for placement must be developed. The strategy adopted in this paper is to analyze the molecular graphs in terms of cyclical and non-cyclical nodes; the latter are further divided into terminal and non-terminal nodes. Molecular fragments, from a fragments database, are described in a similar way. A canonical numbering scheme for the fragments and the local subgraph of the molecular graph enables fragments to be placed efficiently onto the molecular graph. Further optimization is achieved by placing similar fragments into bins using a hashing scheme based on the canonical numbering. The graph perception algorithm is illustrated in detail.

Barakat MT, Dean PM.

J Comput Aided Mol Des. 1995 Aug;9(4):341-50.

The atom assignment problem in automated de novo drug design. 1. Transferability of molecular fragment properties.

This paper is the first of a series which examines the problems of atom assignment in automated de novo drug design. In subsequent papers, a combinatoric optimization method for fragment placement onto 3D molecular graphs is provided. Molecules are built from molecular graphs by placing fragments onto the graph. Here we examine the transferability of atomic residual charge, by fragment placement, with respect to the electrostatic potential. This transferability has been tested on 478 molecular structures extracted from the Cambridge Structural Database. The correlation found between the electrostatic potential computed from composite fragments and that computed for the whole molecule was encouraging, except for extended conjugated systems.

Barker J, Courtney S, Hesterkamp T, Ullmann D, Whittaker M

Exp Opin Drug Discov. 2006 1, 225-236

Fragment screening by biochemical assay.

[No abstract]

Bartoli S, Fincham CI, Fattori D.

Curr Opin Drug Discov Devel. 2007 Jul;10(4):422-9.

Fragment-based drug design: combining philosophy with technology.

Fragment-based drug design began more than ten years ago and has been steadily gaining in popularity. This review discusses how fragments have been used to choose druggable targets, and what parameters need to be evaluated if a fragment hit is to be considered a suitable ligand for development. Examples of fragment-based screening from the recent literature are reviewed to highlight the various approaches used, along with the possible application of additional techniques to fragment screening against immobilized targets. Finally, mention is made of two different areas, multi-target drug discovery and selective tumor cell targeting, where fragment-based approaches may play an important role in the future.

Baurin N, Aboul-Ela F, Barril X, Davis B, Drysdale M, Dymock B, Finch H, Fromont C, Richardson C, Simmonite H, Hubbard RE.

J Chem Inf Comput Sci. 2004 Nov-Dec;44(6):2157-66.

Design and characterization of libraries of molecular fragments for use in NMR screening against protein targets.

We have designed four generations of a low molecular weight fragment library for use in NMR-based screening against protein targets. The library initially contained 723 fragments which were selected manually from the Available Chemicals Directory. A series of in silico filters and property calculations were developed to automate the selection process, allowing a larger database of 1.79 M available compounds to be searched for a further 357 compounds that were added to the library. A kinase binding pharmacophore was then derived to select 174 kinase-focused fragments. Finally, an additional 61 fragments were selected to increase the number of different pharmacophores represented within the library. All of the fragments added to the library passed quality checks to ensure they were suitable for the screening protocol, with appropriate solubility, purity, chemical stability, and unambiguous NMR spectrum. The successive generations of libraries have been characterized through analysis of structural properties (molecular weight, lipophilicity, polar surface area, number of rotatable bonds, and hydrogen-bonding potential) and by analyzing their pharmacophoric complexity. These calculations have been used to compare the fragment libraries with a drug-like reference set of compounds and a set of molecules that bind to protein active sites. In addition, an analysis of the overall results of screening the library against the ATP binding site of two protein targets (HSP90 and CDK2) reveals different patterns of fragment binding, demonstrating that the approach can find selective compounds that discriminate between related binding sites.

Baurin N, Baker R, Richardson C, Chen I, Foloppe N, Potter A, Jordan A, Roughley S, Parratt M, Greaney P, Morley D, Hubbard RE.

J Chem Inf Comput Sci. 2004 Mar-Apr;44(2):643-51.

Drug-like annotation and duplicate analysis of a 23-supplier chemical database totalling 2.7 million compounds.

We have implemented five drug-like filters, based on 1D and 2D molecular descriptors, and applied them to characterize the drug-like properties of commercially available chemical compounds. In addition to previously published filters (Lipinski and Veber), we implemented a filter for medicinal chemistry tractability based on lists of chemical features drawn up by a panel of medicinal chemists. A filter based on the modeling of aqueous solubility (>1 microM) was derived in-house, as well as another based on the modeling of Caco-2 passive membrane permeability (>10 nm/s). A library of 2.7 million compounds was collated from the 23 compound suppliers and analyzed with these filters, highlighting a tendency toward highly lipophilic compounds. The library contains 1.6 M unique structures, of which 37% (607,223) passed all five drug-like filters. None of the 23 suppliers provides all the members of the drug-like subset, emphasizing the benefit of considering compounds from various compound suppliers as a source of diversity for drug discovery.

Baurin N, Mozziconacci JC, Arnoult E, Chavatte P, Marot C, Morin-Allory L.

J Chem Inf Comput Sci. 2004 Jan-Feb;44(1):276-85.

2D QSAR consensus prediction for high-throughput virtual screening. An application to COX-2 inhibition modeling and screening of the NCI database.

Using classification (SOM, LVQ, Binary, Decision Tree) and regression algorithms (PLS, BRANN, k-NN, Linear), this paper details the building of eight 2D-QSAR models from a 266 COX-2 inhibitor training set. The predictive performances of these eight models were subsequently compared using an 88 COX-2 inhibitor test set. Each ligand is described by 52 2D descriptors expressed as van der Waals Surface Areas (P_VSA) and its COX-2 binding IC50. One of our best predictive models is the neural network model (BRANN), which is able to select a subset, from the 88 ligand test set, that contains 94% COX-2 active inhibitors (pIC50>7.5) and detects 71% of all the actives. We then introduce a QSAR consensus prediction protocol that is shown to be more predictive than any single QSAR model: our C3 consensus approach is able to select a subset from the 88 ligand test set that contains 94% active inhibitors and 83% of all the actives. The 2D QSAR consensus protocol was finally applied to the high-throughput virtual screening of the NCI database, containing 193,477 organic compounds.

Baurin N, Vangrevelinghe E, Morin-Allory L, Merour JY, Renard P, Payard M, Guillaumet G, Marot C.

J Med Chem. 2000 Mar 23;43(6):1109-22.

3D-QSAR CoMFA study on imidazolinergic I(2) ligands: a significant model through a combined exploration of structural diversity and methodology.

Displaying an unprecedented structural diversity, 119 I(2) ligands, and their pK(i) values, were collected and submitted to a comparative molecular fields analysis (CoMFA) study. They were discerned into three structural subsets (A, B, C), to explore the I(2) 3D-QSARs from finite structural systems (A, B, C) to more complex ones (AB, AC, BC, ABC). In addition, various key steps of the CoMFA methology were explored. The applied method used two pharmacophore templates and seven molecular field combinations (electrostatic, lipophilic, steric), as well as eight alignment methods (two point-by-point and six similarity-based variations). That way, 644 CoMFA models were obtained and further selected according to their predictive ability through two filters. The first filter was mainly based on the q(2), which internally evaluates the predictive ability from the training set. For the second filter, the predictive ability was externally evaluated through the prediction of test sets. Finally, one model was extracted from the whole data as the best. Indeed, it combines three features of upmost importance for the further design of ligands endowed with high I(2) affinity: structural diversity (n = 73), robustness (N = 9, r(2) = 0.96, s = 0. 28, F = 148), and a great fully assessed predictive ability (q(2) = 0.50, r(2)(test set) = 0.81, n(test set) = 46). On the basis of structural data and CoMFA isocontours, some elements of the I(2) tridimensional pharmacophore are also suggested.

Bender A, Mussa HY, Glen RC.

J Biomol Screen. 2005 Oct;10(7):658-66. Epub 2005 Sep 16.

Screening for dihydrofolate reductase inhibitors using MOLPRINT 2D, a fast fragment-based method employing the naive Bayesian classifier: limitations of the descriptor and the importance of balanced chemistry in training and test sets.

A fragment-based similarity searching method, MOLPRINT 2D, was employed for virtual screening of Escherichia coli dihydrofolate reductase inhibitors. Using the original training set of 50,000 compounds, only marginal enrichment factors (between 1 and 3) could be achieved on the test library. The active structures contained in the training and test libraries represented different types of "chemistry", that is, different substructural features associated with activity. Training and test sets were pooled in a 2nd step and randomly split into training and test of equal size, with the objective of smoothing out the different chemical characteristics of both libraries. In a 10-fold cross-validation study on the new training and test sets, typically 10-fold enrichment could be found in the first 96 positions, 4-fold enrichment in the first 384 positions, and 3-fold enrichment in the first 1536 positions, corresponding to 6, 10, and 28 hits, respectively (out of a total of 307; activity defined as average residual activity of less than 80%). The conclusions are 2-fold. On one hand, the exact fragment-matching similarity searching method employed here is not capable of finding completely novel hit structures. On the other hand, this study emphasizes the requirement for a comparable distribution of chemical features of the training and test sets. MOLPRINT 2D is freely downloadable from http://www.cheminformatics.org.

Blume A, Angulo J, Biet T, Peters H, Benie AJ, Palcic M, Peters T.

J Biol Chem. 2006 Oct 27;281(43):32728-40. Epub 2006 Aug 21.

Fragment-based screening of the donor substrate specificity of human blood group B galactosyltransferase using saturation transfer difference NMR.

Saturation transfer difference NMR experiments on human blood group B alpha-(1,3)-galactosyltransferase (GTB) for the first time provide a comprehensive set of binding epitopes of donor substrate analogs in relation to the natural donor UDP-Gal. This study revealed that the enzyme binds several UDP-activated sugars, including UDP-Glc, UDP-GlcNAc, and UDP-GalNAc. In all cases, UDP is the dominant binding epitope. To identify the minimum requirements for specific binding, a detailed analysis utilizing a fragment-based approach was employed. The binding of donor substrate to GTB is essentially controlled by the base as a "molecular anchor." Uracil represents the smallest fragment that is recognized, whereas CDP, AMP, and GDP do not exhibit any significant binding affinity for the enzyme. The ribose and beta-phosphate moieties increase the affinity of the ligands, whereas the pyranose sugar apparently weakens the binding, although this part of the molecule controls the specificity of the enzyme. Accordingly, UDP represents the best binder. The binding affinities of UDP-Gal, UDP-Glc, and UMP are about the same, but lower than that of UDP. Furthermore, we observed that beta-D-galactose and alpha-D-galactose bind weakly to GTB. Whereas beta-D-galactose binds to the acceptor and donor sites, it is suggested that alpha-D-galactose occupies a third hitherto unknown binding pocket. Finally, our experiments revealed that modulation of enzymatic activity by metal ions critically depends on the total enzyme concentration, raising the question as to which of the bivalent metal cations Mg(2+) and Mn(2+) is more relevant under physiological conditions.

Blume A, Benie AJ, Stolz F, Schmidt RR, Reutter W, Hinderlich S, Peters T.

J Biol Chem. 2004 Dec 31;279(53):55715-21. Epub 2004 Oct 21.

Characterization of ligand binding to the bifunctional key enzyme in the sialic acid biosynthesis by NMR: I. Investigation of the UDP-GlcNAc 2-epimerase functionality.

The bifunctional enzyme UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase is the key enzyme for the biosynthesis of sialic acids. As terminal components of glycoconjugates, sialic acids are associated with a variety of pathological processes such as inflammation and cancer. For the first time, this study reveals characteristics of the interaction of the epimerase site of the enzyme with its natural substrate, UDP-N-acetylglucosamine (UDP-GlcNAc) and derivatives thereof at atomic resolution. Saturation transfer difference NMR experiments were crucial in obtaining ligand binding epitopes and to rank ligands according to their binding affinities. Employing a fragment based approach, it was possible to assign the major component of substrate recognition to the UDP moiety. In particular, the binding epitopes of the uridine moieties of UMP, UDP, UDP-GalNAc, and UDP-GlcNAc are rather similar, suggesting that the binding mode of the UDP moiety is the same in all cases. In contrast, the hexopyranose units of UDP-GlcNAc and UDP-GalNAc display small differences reflecting the inability of the enzyme to process UDP-GalNAc. Surprisingly, saturation transfer difference NMR titrations show that UDP has the largest binding affinity to the epimerase site and that at least one phosphate group is required for binding. Consequently, this study provides important new data for rational drug design.

Blundell TL, Sibanda BL, Montalvao RW, Brewerton S, Chelliah V, Worth CL, Harmer NJ, Davies O, Burke D.

Philos Trans R Soc Lond B Biol Sci. 2006 Mar 29;361(1467):413-23. Review.

Structural biology and bioinformatics in drug design: opportunities and challenges for target identification and lead discovery.

Impressive progress in genome sequencing, protein expression and high-throughput crystallography and NMR has radically transformed the opportunities to use protein three-dimensional structures to accelerate drug discovery, but the quantity and complexity of the data have ensured a central place for informatics. Structural biology and bioinformatics have assisted in lead optimization and target identification where they have well established roles; they can now contribute to lead discovery, exploiting high-throughput methods of structure determination that provide powerful approaches to screening of fragment binding.

Braisted AC, Oslob JD, Delano WL, Hyde J, McDowell RS, Waal N, Yu C, Arkin MR, Raimundo BC.

J Am Chem Soc. 2003 Apr 2;125(13):3714-5. No abstract available.

Discovery of a potent small molecule IL-2 inhibitor through fragment assembly.

Using a site-directed fragment discovery method called tethering, we have identified a 60 nM small molecule antagonist of a cytokine/receptor interaction (IL-2/IL2R) with cell-based activity. Starting with a low micromolar hit, we employed a combination of tethering, structural biology, and computational analysis to design a focused set of 20 compounds. Eight of these compounds were at least 5-fold more active than the original hit. One of these compounds showed a 50-fold enhancement and represents the highest affinity inhibitor reported against this protein-protein target class. This method of coupling selected fragments with a low micromolar hit shows great potential for generating high-affinity lead compounds.

Bright J.

IDrugs. 2007 May;10(5):308-9. No abstract available.

American Chemical Society--233rd National Meeting. Protein fragment screening in novel drug design. 25-29 March 2007, Chicago, IL, USA.

[No abstract]

Capdeville R, Buchdunger E, Zimmermann J, Matter A.

Nat Rev Drug Discov. 2002 Jul;1(7):493-502. Review.

Glivec (STI571, imatinib), a rationally developed, targeted anticancer drug.

In the early 1980s, it became apparent that the work of pioneers such as Robert Weinberg, Mariano Barbacid and many others in identifying cancer-causing genes in humans was opening the door to a new era in anticancer research. Motivated by this, and by dissatisfaction with the limited efficacy and tolerability of available anticancer modalities, a drug discovery programme was initiated with the aim of rationally developing targeted anticancer therapies. Here, we describe how this programme led to the discovery and continuing development of Glivec (Gleevec in the United States), the first selective tyrosine-kinase inhibitor to be approved for the treatment of a cancer.

Card GL, Blasdel L, England BP, Zhang C, Suzuki Y, Gillette S, Fong D, Ibrahim PN, Artis DR, Bollag G, Milburn MV, Kim SH, Schlessinger J, Zhang KY.

Nat Biotechnol. 2005 Feb;23(2):201-7. Epub 2005 Jan 30.

A family of phosphodiesterase inhibitors discovered by cocrystallography and scaffold-based drug design.

Cyclic nucleotide phosphodiesterases (PDEs) comprise a large family of enzymes that regulate a variety of cellular processes. We describe a family of potent PDE4 inhibitors discovered using an efficient method for scaffold-based drug design. This method involves an iterative approach starting with low-affinity screening of compounds followed by high-throughput cocrystallography to reveal the molecular basis underlying the activity of the newly identified compounds. Through detailed structural analysis of the interaction of the initially discovered pyrazole carboxylic ester scaffold with PDE4D using X-ray crystallography, we identified three sites of chemical substitution and designed small selective libraries of scaffold derivatives with modifications at these sites. A 4,000-fold increase in the potency of this PDE4 inhibitor was achieved after only two rounds of chemical synthesis and the structural analysis of seven pyrazole derivatives bound to PDE4B or PDE4D, revealing the robustness of this approach for identifying new inhibitors that can be further developed into drug candidates.

Carr RA, Congreve M, Murray CW, Rees DC.

Drug Discov Today. 2005 Jul 15;10(14):987-92. Review.

Fragment-based lead discovery: leads by design.

Fragment-based lead discovery (also referred to as needles, shapes, binding elements, seed templates or scaffolds) is a new lead discovery approach in which much lower molecular weight (120-250 Da) compounds are screened relative to HTS campaigns. Fragment-based hits are typically weak inhibitors (10 microM-mM), and therefore need to be screened at higher concentration using very sensitive biophysical detection techniques such as protein crystallography and NMR as the primary screening techniques, rather than bioassays. Compared with HTS hits, these fragments are simpler, less functionalized compounds with correspondingly lower affinity. However, fragment hits typically possess high 'ligand efficiency' (binding affinity per heavy atom) and so are highly suitable for optimization into clinical candidates with good drug-like properties.

Carter TA, Wodicka LM, Shah NP, Velasco AM, Fabian MA, Treiber DK, Milanov ZV, Atteridge CE, Biggs WH 3rd, Edeen PT, Floyd M, Ford JM, Grotzfeld RM, Herrgard S, Insko DE, Mehta SA, Patel HK, Pao W, Sawyers CL, Varmus H, Zarrinkar PP, Lockhart DJ.

Proc Natl Acad Sci U S A. 2005 Aug 2;102(31):11011-6. Epub 2005 Jul 26.

Inhibition of drug-resistant mutants of ABL, KIT, and EGF receptor kinases.

To realize the full potential of targeted protein kinase inhibitors for the treatment of cancer, it is important to address the emergence of drug resistance in treated patients. Mutant forms of BCR-ABL, KIT, and the EGF receptor (EGFR) have been found that confer resistance to the drugs imatinib, gefitinib, and erlotinib. The mutations weaken or prevent drug binding, and interestingly, one of the most common sites of mutation in all three kinases is a highly conserved "gatekeeper" threonine residue near the kinase active site. We have identified existing clinical compounds that bind and inhibit drug-resistant mutant variants of ABL, KIT, and EGFR. We found that the Aurora kinase inhibitor VX-680 and the p38 inhibitor BIRB-796 inhibit the imatinib- and BMS-354825-resistant ABL(T315I) kinase. The KIT/FLT3 inhibitor SU-11248 potently inhibits the imatinib-resistant KIT(V559D/T670I) kinase, consistent with the clinical efficacy of SU-11248 against imatinib-resistant gastrointestinal tumors, and the EGFR inhibitors EKB-569 and CI-1033, but not GW-572016 and ZD-6474, potently inhibit the gefitinib- and erlotinib-resistant EGFR(L858R/T790M) kinase. EKB-569 and CI-1033 are already in clinical trials, and our results suggest that they should be considered for testing in the treatment of gefitinib/erlotinib-resistant non-small cell lung cancer. The results highlight the strategy of screening existing clinical compounds against newly identified drug-resistant mutant variants to find compounds that may serve as starting points for the development of next-generation drugs, or that could be used directly to treat patients that have acquired resistance to first-generation targeted therapy.

Caskey CT.

Annu Rev Med. 2007;58:1-16. Review.

The drug development crisis: efficiency and safety.

Despite advancements in genetics, chemistry, and protein engineering, recent years have seen fewer approvals of new drugs, increases in development costs, and high-profile drug withdrawals. This article focuses on technologic methods for improving drug development efficiency. These technologies include high-content cell screening, expression profiling, mass spectroscopy, mouse models of disease, and a post-launch screening program that enables investigations of adverse drug effects. Implementation of these new technologies promises to improve performance in drug development and safety.

Chau PL, Dean PM.

J Comput Aided Mol Des. 1994 Oct;8(5):527-44.

Electrostatic complementarity between proteins and ligands. 2. Ligand moieties.

Drug design strategies consider factors governing intermolecular interactions to build up putative ligands. In many strategies, the ligand is constructed using fragments which are placed in the site sequentially. The optimization is then performed with each fragment. We would like to examine if this optimization strategy could generate ligands with optimal electrostatic interactions. The electrostatic complementarities between constituent moieties and the receptor site have been calculated. The whole-ligand complementarity does not appear to be the mathematical mean of the individual complementarities, nor have we found a simple relationship between the moiety and whole-ligand complementarities. The results demonstrate clearly that, using a simple model, it is very difficult to predict the electrostatic potential complementarity of the whole ligand from the complementarities of its constituent chemical moieties. This means that ligand design strategies must optimize the electrostatic complementarity globally, and not moiety by moiety.

Chau PL, Dean PM.

J Comput Aided Mol Des. 1992 Aug;6(4):407-26.

Automated site-directed drug design: an assessment of the transferability of atomic residual charges (CNDO) for molecular fragments.

Department of Pharmacology, University of Cambridge, U.K.

In this paper a database of atomic residual charges has been constructed for all the molecular fragments defined previously in a combinatorial search of the Cambridge Structural Database. The charges generated for the atoms in each fragment are compared with charges calculated for whole molecules containing those fragments. The fragment atomic charges lie within 1 S.D. of the mean for 68%, and within 2 S.D. for 91%, of the atoms whose charges were computed for whole molecules. The actual charges on any atom are strongly influenced by the adjacent connected atoms. There is a large spread of atomic residual charge within the fragments database.

Chau PL, Dean PM.

J Comput Aided Mol Des. 1992 Aug;6(4):397-406.

Automated site-directed drug design: searches of the Cambridge Structural Database for bond lengths in molecular fragments to be used for automated structure assembly.

In this paper a database of small frequently occurring molecular fragments is used for the determination of fragment bond lengths from the Cambridge Structural Database. A large number of bond types are described that have not been reported previously.

Chau PL, Dean PM.

J Comput Aided Mol Des. 1992 Aug;6(4):385-96.

Automated site-directed drug design: the generation of a basic set of fragments to be used for automated structure assembly.

If a method is to be developed to assemble putative ligand structures in site-directed drug design, from molecular graphs generated in the site, then basic building blocks are needed. Structure assembly is a combinatoric process that needs to be optimised if it is to be tractable. What has to be determined is whether small molecular fragments can have transferable properties from one molecule to another. In this paper we determine all possible combinations of 3-, 4- and 5-atom aliphatic fragments from a small set of atoms H, C, N, O, F or Cl. The frequency of occurrence of these candidate fragments is searched for in the Cambridge Structural Database. A similar analysis is performed on charged fragments. A more restricted search is carried out for P and S and aromatic structures. A basic set of fragments can be derived that have a significant frequency in known crystal structures. The transferability of fragment properties is discussed in subsequent papers.

Chavatte P, Yous S, Marot C, Baurin N, Lesieur D.

J Med Chem. 2001 Sep 27;44(20):3223-30.

Three-dimensional quantitative structure-activity relationships of cyclo-oxygenase-2 (COX-2) inhibitors: a comparative molecular field analysis.

The three-dimensional quantitative structure-activity relationship (3D-QSAR) approach using comparative molecular field analysis (CoMFA) was applied to an extensive series of 305 varied diarylheterocyclic derivatives known as COX-2 selective inhibitors. X-ray crystal structure of COX-2 bound with SC-558, a selective COX-2 inhibitor, was used to derive the putative bioactive conformation of these inhibitors. Five statistically significant models were obtained from the randomly constituted training sets (229 compounds) and subsequently validated with the corresponding test sets (76 compounds). The best predictive model (n = 229, q(2) = 0.714, N = 8, r(2) = 0.905, s = 0.291, F = 261.545) was selected for further comparison of the CoMFA contour maps obtained for steric, electrostatic, and lipophilic fields with the enzyme structure. The high level of compatibility with the COX-2 enzyme topology shows the great accuracy of this model that can predict inhibitory activities for a wide range of compounds and offers important structural insight into designing novel antiinflammatory drugs prior to their synthesis.

Chen J, Zhang Z, Stebbins JL, Zhang X, Hoffman R, Moore A, Pellecchia M.

ACS Chem Biol. 2007 May 22;2(5):329-36. Epub 2007 Apr 27.

A fragment-based approach for the discovery of isoform-specific p38alpha inhibitors.

In this study, we describe a novel approach for lead discovery against protein kinases, pharmacophore by interligand nuclear Overhauser effect (ILOE), in which a pair of ligands that bind to adjacent pockets on the target surface is identified by the detection of protein-mediated ILOEs. We demonstrate that a pharmacophore-based search guided by experimental binding data of weakly interacting fragments can be rapidly and efficiently used to identify (or synthesize) high-affinity, selective ligands. Targeting the inactive state of protein kinases represents a promising approach to achieve selectivity and cellular efficacy. In this respect, when we apply the method for the discovery of potent p38alpha inhibitors, we also demonstrate that the resulting bidentate compounds are highly selective and exhibit a cellular activity that parallels their in vitro binding to the inactive form of the kinase. The method is relatively simple and of general applicability, and as such we anticipate its potential implementation against a variety of macromolecular targets, including not only protein kinases but also those involved in protein-protein interactions or even nucleic acids.

Chen X, Reynolds CH.

J Chem Inf Comput Sci. 2002 Nov-Dec;42(6):1407-14.

Performance of similarity measures in 2D fragment-based similarity searching: comparison of structural descriptors and similarity coefficients.

2D fragment-based similarity searching is one of the most popular techniques for searching a large database of chemical structures and has been widely applied in drug discovery. However, its performance, especially its effectiveness in retrieving active structural analogues, has not been adequately studied. We report a series of computational experiments, where we systematically studied the influence of structural descriptors and similarity coefficients on the effectiveness of similarity searching. The study was conducted using two public large data sets, NCI anti-AIDS and MDDR. Four sets of 2D linear fragment descriptors, based on the original definitions of atom pairs and atom sequences, were compared. The effect of using the Tanimoto coefficient and the Euclidean distance was studied as a function of descriptor set. The results clearly indicate that the Tanimoto coefficient is superior to the Euclidean distance in 2D-fragment based similarity searching, in terms of hit rate, while atom sequences demonstrate the best overall performance among the structural descriptors we studied.

Cherry M, Williams DH.

Curr Med Chem. 2004 Mar;11(6):663-73. Review.

Recent kinase and kinase inhibitor X-ray structures: mechanisms of inhibition and selectivity insights.

Recent years have seen an explosion in the number of publicly available x-ray crystal structures of protein kinases. These structures have provided a wealth of information on the regulatory mechanisms, conformational plasticity and drugability of this important family of enzymes. Drawing upon structural information, new insights into the development of protein kinase inhibitors are discussed including de-novo design, molecular templates for ATP competitive inhibitors and alternative mechanisms of inhibition. The highly conserved nature of the ATP binding site is of central concern to drug development and the concept of a selectivity profile has arisen with structure-based design emerging as a key tool for addressing the challenges of specificity. In addition, protein-ligand complexes, where the enzyme is in an inactive conformation, signify an alternate approach to protein kinase inhibition. The belief that an inactive kinase presents a less conserved target is reviewed using observations on the structural changes occurring during protein kinase regulation.

Choong IC, Lew W, Lee D, Pham P, Burdett MT, Lam JW, Wiesmann C, Luong TN, Fahr B, DeLano WL, McDowell RS, Allen DA, Erlanson DA, Gordon EM, O'Brien T.

J Med Chem. 2002 Nov 7;45(23):5005-22.

Identification of potent and selective small-molecule inhibitors of caspase-3 through the use of extended tethering and structure-based drug design.

The design, synthesis, and in vitro activities of a series of potent and selective small-molecule inhibitors of caspase-3 are described. From extended tethering, a salicylic acid fragment was identified as having binding affinity for the S(4) pocket of caspase-3. X-ray crystallography and molecular modeling of the initial tethering hit resulted in the synthesis of 4, which reversibly inhibited caspase-3 with a K(i) = 40 nM. Further optimization led to the identification of a series of potent and selective inhibitors with K(i) values in the 20-50 nM range. One of the most potent compounds in this series, 66b, inhibited caspase-3 with a K(i) = 20 nM and selectivity of 8-500-fold for caspase-3 vs a panel of seven caspases (1, 2, and 4-8). A high-resolution X-ray cocrystal structure of 4 and 66b supports the predicted binding modes of our compounds with caspase-3.

Ciulli A, Lobley CM, Tuck KL, Smith AG, Blundell TL, Abell C.

Acta Crystallogr D Biol Crystallogr. 2007 Feb;63(Pt 2):171-8. Epub 2007 Jan 16.

pH-tuneable binding of 2'-phospho-ADP-ribose to ketopantoate reductase: a structural and calorimetric study.

The crystal structure of Escherichia coli ketopantoate reductase in complex with 2'-monophosphoadenosine 5'-diphosphoribose, a fragment of NADP+ that lacks the nicotinamide ring, is reported. The ligand is bound at the enzyme active site in the opposite orientation to that observed for NADP+, with the adenine ring occupying the lipophilic nicotinamide pocket. Isothermal titration calorimetry with R31A and N98A mutants of the enzyme is used to show that the unusual ;reversed binding mode' observed in the crystal is triggered by changes in the protonation of binding groups at low pH. This research has important implications for fragment-based approaches to drug design, namely that the crystallization conditions and the chemical modification of ligands can have unexpected effects on the binding modes.

Ciulli A, Williams G, Smith AG, Blundell TL, Abell C.

J Med Chem. 2006 Aug 10;49(16):4992-5000.

Probing hot spots at protein-ligand binding sites: a fragment-based approach using biophysical methods.

Mapping interactions at protein-ligand binding sites is an important aspect of understanding many biological reactions and a key part of drug design. In this paper, we have used a fragment-based approach to probe "hot spots" at the cofactor-binding site of a model dehydrogenase, Escherichia coli ketopantoate reductase. Our strategy involved the breaking down of NADPH (Kd = 300 nM) into smaller fragments and the biophysical characterization of their binding using WaterLOGSY NMR spectroscopy, isothermal titration calorimetry (ITC), and inhibition studies. The weak binding affinities of fragments were measured by direct ITC titrations under low c value conditions. The 2'-phosphate and the reduced nicotinamide groups were found to contribute a large part of the binding energy. A combination of ITC and site-directed mutagenesis enabled us to locate the fragments at separate hot spots on opposite ends of the cofactor-binding site. This study has identified structural determinants for cofactor recognition that represent a blueprint for future inhibitor design.

Clark M, Guarnieri F, Shkurko I, Wiseman J.

J Chem Inf Model. 2006 Jan-Feb;46(1):231-42.

Grand canonical Monte Carlo simulation of ligand-protein binding.

A new application of the grand canonical thermodynamics ensemble to compute ligand-protein binding is described. The described method is sufficiently rapid that it is practical to compute ligand-protein binding free energies for a large number of poses over the entire protein surface, thus identifying multiple putative ligand binding sites. In addition, the method computes binding free energies for a large number of poses. The method is demonstrated by the simulation of two protein-ligand systems, thermolysin and T4 lysozyme, for which there is extensive thermodynamic and crystallographic data for the binding of small, rigid ligands. These low-molecular-weight ligands correspond to the molecular fragments used in computational fragment-based drug design. The simulations correctly identified the experimental binding poses and rank ordered the affinities of ligands in each of these systems.

Coles M, Heller M, Kessler H.

Drug Discov Today. 2003 Sep 1;8(17):803-10. Review.

NMR-based screening technologies.

NMR-based ligand screening is now an established field in its own right. In recent years, advances in both methodology and hardware have broadened its range of applications and pushed back practical limitations, leading to the growing importance of NMR screening as a tool in industrial drug research. An overview of new screening methods and applications is presented here, and ways in which NMR-screening is being used in cooperation with other screening techniques are discussed.

Congreve M, Aharony D, Albert J, Callaghan O, Campbell J, Carr RA, Chessari G, Cowan S, Edwards PD, Frederickson M, McMenamin R, Murray CW, Patel S, Wallis N.

J Med Chem. 2007 Mar 22;50(6):1124-32. Epub 2007 Feb 22.

Application of fragment screening by X-ray crystallography to the discovery of aminopyridines as inhibitors of beta-secretase.

Fragment-based lead discovery has been successfully applied to the aspartyl protease enzyme beta-secretase (BACE-1). Fragment hits that contained an aminopyridine motif binding to the two catalytic aspartic acid residues in the active site of the enzyme were the chemical starting points. Structure-based design approaches have led to identification of low micromolar lead compounds that retain these interactions and additionally occupy adjacent hydrophobic pockets of the active site. These leads form two subseries, for which compounds 4 (IC50 = 25 microM) and 6c (IC50 = 24 microM) are representative. In the latter series, further optimization has led to 8a (IC50 = 690 nM).

Congreve M, Carr R, Murray C, Jhoti H.