 |
|
|
|
Insights into FAAH catalysis by QM/MM modelling study
Alessio Lodola (1), Adrian J. Mulholland (2), Johannes C. Hermann (2) and Marco Mor (1)
1. Dipartimento Farmaceutico, Università degli Studi di Parma, 43100 Parma, Italy
2. School of Chemistry, University of Bristol, Bristol, BS8 1TS, UK
Fatty acid amide hydrolase (FAAH) is a member of the amidase signature family that inactivates neuromodulatory lipid amides, including the endogenous cannabinoid anandamide and the sleep-inducing substance oleamide. [1] A new class of carbamic acid aryl esters [2,3] which exhibit anxiolytic effects in rodents by FAAH inhibition, has been described: this suggests that this enzyme represents an attractive therapeutic target for the treatment of anxiety and other CNS disorders. In this light, improved mechanistic understanding of the hydrolytic reaction catalyzed by fatty acid amide hydrolase should assist in the development of new inhibitors and new anxiolytic drugs. The crystal structure of FAAH [4], and kinetic studies of mutants [5] suggest that the core catalytic machinery is composed by serine-serine-lysine catalytic triad, in contrast to the serine-histidine-aspartate one, typical of most serine hydrolases. Combined quantum mechanics/molecular mechanics (QM/MM) [6] approach (PM3/CHARMM) allowed us to describe at atomic level the first step of the acylation reaction between FAAH and the substrate oleamide. Adiabatic mapping approach was employed to calculate potential energy surfaces (PESs) and for generating approximate models of the transition states and intermediates. Reaction energetics were also studied by high level hybrid density-functional theory (DFT) calculations. The proposed pathway [7] demonstrates that Lys142 and Ser217 play a cooperative role in activating the nucleophilic Ser241, in agreement with experimental data on FAAH mutants. [5] The calculated barrier is consistent with the experimental rate. Key interactions in the enzyme have been identified and analyzed. The proposed mechanism provides new and detailed insight into FAAH catalysis that will be useful in current and future inhibitor design.
References:
1. Piomelli, D. Nat. Rev. Neurosci. 2003, 4, 873-884.
2. Tarzia, G.; Duranti, A.; Tontini, A.; Piersanti, G.; Mor, M.; Rivara, S.; Plazzi, P. V.; Park, C.; Kathuria, S.; Piomelli, D. J. Med. Chem. 2003, 46, 2352-2360.
3. Mor, M.; Rivara, S.; Lodola, A.; et al. J. Med. Chem. 2004, 47, 4998-5008.
4. Bracey, M. H.; Hanson, M. A.; Masuda, K. R.; Stevens, R. C.; Cravatt, B. F. Science 2002, 298, 1793-1796.
5. McKinney, M. K.; Cravatt, B. F. J. Biol. Chem. 2003, 278, 37393-37399.
6. Field, M. J.; Bash, P. A.; Karplus, M. J. Comput. Chem. 1990, 11, 700-733.
7. Lodola, A.; Mor, M.; Hermann, J. C.; Tarzia, G.; Piomelli, D.; Mulholland, A. J. J. Am. Chem. Soc. 2004, submitted.
A Retrospective Docking Study Of PDE4B Ligands And An Analysis Of The Behaviour Of Selected Scoring Functions
Chido P. Mpamhanga (1), Beining Chen (1), Iain McLay (2), Daniel Ormsby (2) and Mika K. Lindvall (2)
1. Department of Chemistry, University of Sheffield, Dainton Building, Broockhill, Sheffield, S3 7HF, UK
2. GlaxoSmithKline Medicines Research Centre, Computational and Structural Sciences, Stevenage, Hertfordshire, UK, SG1 2NY
In general fast scoring functions fall short in their ability to determine the relative affinities of ligands for their receptors. This study demonstrates a successful docking and scoring methodology for PDE4B which is based on combining the results of several scoring functions using consensus scoring. A series of known inhibitors of PDE4B were docked into the PDE4B/ Pyrazolo[3,4-b]pyridine binding site using LigandFit a fast shape matching docking algorithm. This work was done as a preliminary study, to verify the suitability of the LigandFit/DockScore protocol for virtual screening for another project which required a method that could enrich the top 5% of a database by a factor of at least four. An RMSD comparison of the LigandFit/DockScore (in virtual screening mode) generated poses with the crystallographic poses of 19 inhibitors whose x-ray structures were available, revealed a reasonable success rate. However, the main objective was to investigate the effectiveness of five available scoring functions (PMF, JAIN, PLP2, LigScore2 and DockScore) to enrich the top ranked fractions of nine artificial databases constructed by seeding 20 randomly selected inhibitors (pIC50 > 6.5) into 1980 inactive ligands (pIC50 < 5). PMF and JAIN showed high average enrichment factors (greater than 4) in the top 5-10% of the ranked databases. Rank-based consensus scoring was also investigated and the rational combination of three scoring functions resulted in more robust and generalisable scoring schemes, consensus scores DPmJ (DockScore, PMF and JAIN) and PPmJ (PLP2, PMF and JAIN) yielded particularly good results. Finally a brief analysis of the behaviour of the scoring functions across different chemo-types or chemical classes followed. This revealed the inherent bias of the docking and scoring method towards the initial crystal structure binding mode (PDE4B/Pyrazolo[3,4-b]pyridine). This in turn suggests a need to consider multiple binding modes in docking studies. Thus future work will be focused on the docking of ligands into multiple binding sites of the PDE4B. Another lesson learnt from this investigation is that scoring functions can be used to a limited extent for lead optimization.
References:
1. C. M. Venkatachalam, X. Jiang, T. Oldfield, M. Waldman. Journal of Molecular Graphics and Modeling, 21, 2003, 289-30.
2. D. G. Allen, D.M. Coe, C.M. Cook, M. D. Dowle, C. D. Edlin, J. N. Hamblin, M.R Johnson, P.S. Jones, R. G. Knowles, M. K. Lindvall, C. J. Mitchell, A.J. Redgrave, N. Trivedi, P. Ward, Pyrazolo[3,4-b]pyridine compounds, and their use as phosphodiesterase inhibitors. PCT Int. Appl. (2004), 293.
3. Mpamhanga C.P., Chen. B. Ormsby. D., Lindvall. M. K., McLay. I. A Retrospective Docking and Scoring Study of PDE4 Ligands and the Consequences of Consensus Scoring. (Under review).
The Role of Water in Protein-Ligand Interactions: Implications for Rational Drug Design
Caterina Barillari, J.W. Essex, R. Viner, C. Neylon (advisor)
Department of Chemistry, University of Southampton, Highfield, Southampton, SO17 1BJ
 |
| Figure 1. Buried waters in BPTI |
Water molecules play a crucial role in mediating the interaction between a ligand and a macromolecular receptor. Some water molecules located in the binding pocket of the protein can be displaced upon ligand binding and this process is favoured by a gain in entropy of the system; other water molecules can stabilize the complex by acting as a bridge between the two components.[1] If we knew a priori which waters are likely to be displaced by the ligand and which ones are instead tightly bound to the protein, molecules could be designed specifically with hydrophilic moieties positioned in such a way to displace only targeted waters, thus increasing the binding affinity for the protein. If a water molecule is tightly bound to a protein, it is likely to have a more favourable free energy of binding than a displaceable water. In this work, three methods for the calculation of the free energy of binding of water molecules buried in proteins using Monte Carlo simulations were compared: finite difference thermodynamic integration (FDTI) [2], cavity biased grand canonical Monte Carlo (CB-GCMC) [3,4] and replica exchange thermodynamic integration (RETI). [5]
Two proteins, containing well identified buried water molecules, were used as test systems: BPTI and the Subtilisin Carlsberg-Eglin C complex. The free energy of binding of one water in BPTI (W122) and 10 waters in Subtilisin Carlsberg-Eglin C were previously calculated using thermodynamic integration in Molecular Dynamics simulations [6,7] and were used as comparison for our calculations. CB-GCMC was not able to reproduce the results reported in the literature and, in particular, it was not able to discriminate between waters forming a cluster, as it yielded an apparent identical free energy of binding for all of them. FDTI and RETI instead gave very similar results, which were in good agreement with those reported in the literature; the advantage of RETI over FDTI is that better convergence in the energy is obtained and the errors in the calculations are reduced.
RETI is now being applied to the calculation of the free energy of binding of key water molecules mediating the interaction between a protein and a ligand in a dataset of 7 proteins each in complex with 6 different ligands. The existence of a relationship between these calculated free energy values and the structure of the protein-ligand complexes will be investigated in future.
References:
1. C.Poornima, P.Dean, J. Comp. Aid. Mol. Des. 9 (1999) 500.
2. C.J.Woods, 'The development of Free Energy Methods for Protein-Ligand Complexes', PhD thesis, University of Southampton (2003).
3. M.Mezei, Molecular Physics, 40 (1980) 901.
4. M.Mezei, J. Chem. Phys., 86 (1987) 7084.
5. C.J.Woods, J.W.Essex, M.A.King, J. Phys. Chem. B, 107 (2003) 13703
6. L.R.Olano, S.W.Rick, J. Am. Chem. Soc., 126 (2004) 7991.
7. L.Zhang, J.Hermans, Proteins, 24 (1996) 433.
Investigating Fundamental Principles of Enzyme Catalysis
Kara Ranaghan, Lars Ridder, and Adrian Mulholland
Centre for Computational Chemistry, School of Chemistry, University of Bristol, Cantock's Close, Bristol BS8 1TS.
Chorismate mutase catalyses the Claisen rearrangement of chorismate to prephenate, a rare example of a biochemically-catalysed pericyclic reaction. Catalysis does not involve covalent bond formation with the enzyme and the rearrangement occurs via the same mechanism in solution and in the enzyme. This makes it an ideal test case for studying the principles of enzyme catalysis. This enzyme has been the focus of many theoretical and experimental studies over the past 20 years, but the nature of catalysis still remains a matter of debate. Until recently it has been widely accepted that preferential stabilisation of the transition state (TS) through electrostatic interactions with the enzyme is the major contributor to catalysis in this enzyme, with 'strain' or substrate compression possibly also playing a role. In contrast to these widely accepted ideas, Bruice and co-workers [1] have recently proposed that catalysis in this enzyme is due to the ability of the enzyme to bind 'near attack conformations' of the substrate (NACs). QM/MM calculations have been carried out on B. subtilis chorismate mutase at the semiempirical (AM1) and ab initio (RHF/6-31G(d)) QM/MM levels of theory [2]. The best estimates of the potential energy barrier in the enzyme are 7.4-11.0 kcal mol-1 (MP2/6-31+G(d)//6-31G(d)/CHARMM22) and 12.7-16.1 kcal mol-1 (B3LYP/6-311+G(2d,p)//6-31G(d)/CHARMM22). These values are comparable to the experimental estimate of ΔH‡ = 12.7±0.4 kcal mol-1. Significant TS stabilisation is predicted at both the AM1 and RHF/6-31G(d) CHARMM22 QM/MM levels. Important residues for TS stabilisation are: Arg90, Arg7, Arg63, Glu78, and some water molecules. A QM/MM free energy perturbation study (FEP) at the AM1 level of theory indicates that NACs may contribute 3.8-4.6 kcal mol-1 to catalysis [3], somewhat lower than the 8.4 kcal mol-1 suggested by Bruice et al. from MD simulations. Results from different types of QM/MM calculations show that TS stabilisation is a key factor in catalysis in chorismate mutase.
References:
1. S. Hur, and T. C. Bruice, J. Am. Chem. Soc., 2003, 125, 5964-5972.
2. K. E. Ranaghan, L. Ridder, B. Szefczyk, et al., Org. Biomol. Chem., 2004, 2, 968-9.
3. K. E. Ranaghan, and A. J. Mulholland, Chem. Commun., 2004, 1238-1239.
Design and synthesis of combinatorial libraries of pyridine dicarbonitriles, potent binding ligands to cellular prion protein, PrPC
T. R. Reddy, V. Gillet, B. Chen
Department of Chemistry, University of Sheffield, Daintons Building, Brookhill, Sheffield S3 7HF
Prion diseases are a group of fatal neurodegenerative disorders caused by conformational change of cellular prion proteins in humans and a variety of other animals. When abnormal prion protein (PrPSc) entering the body, it is able to convert their cellular counterparts into the abnormal forms. The difference between the normal and abnormal proteins lies in their folding. The abnormal PrPSc proteins are folded in a way that resists normal protease degradation leading to the build-up of aggregates of PrPSc. Aggregation of PrPSc results in neurological dysfunction accompanied by neuronal vacuolation and astrocytic gliosis.
Discovering lead compounds which can stabilize native prion protein (PrPC) to prevent the disease progression, is the key in the project. Pyridine dicarbonitrile compound, Cp-60 and three of its analogues inhibit the formation of newly synthesized PrPSc and cure infected cells. This class of compounds were also screened using our direct binding assay and showed binding to PrPC. Hence it was selected as a lead compounds for combinatorial optimisation.
A product based library, which relies on the core structure of Cp-60, was designed and ligand selection was performed taking into account synthetic accessibility and flexibility. Inspired from the elegant one pot reaction, a reaction-based library was also created using commercially available aldehydes and thiols. Both libraries were generated and the individual compounds were docked into the binding pocket of PrPC using various software packages to guide the selection of final library for chemical synthesis. Initial screening of the synthesised library showed a number of compounds having improved binding to PrPC compared to the parent compound.
A Computer Program for the Prediction and Analysis of Organic Reactivity
Ingrid M. Socorro (1), Jonathan M. Goodman (1) and Keith T. Taylor (2)
1. Unilever Cambridge Molecular Informatics, Department of Chemistry, Lensfield Road, Cambridge, CB2, 1EW, UK
2. Elsevier MDL, 14600 Catalina Street, San Leandro, CA 94577, USA
Our work is focused on the development of a computer program with the purpose of predicting and analysing organic reactivity. The system developed is able to predict the outcome of organic reactions given starting materials and conditions. The program achieves reaction prediction based on a new approach that combines general knowledge of organic chemistry and molecular modeling. In this way, the program applies these series of rules to determine the reactivity of the molecules, making decisions on primary aspects of organic reactivity such as where are the reactive sites, which bonds are to be broken or made, etc. Moreover, a mechanistic approach is taken when considering the reactions. Thus, the program goes from the reactants to the products through a series of intermediate structures, which are filtered and selected according the coded rules and the results given by molecular and quantum mechanic calculations. Therefore, new reactivity could be found and analysed when considering novel reactions. We have applied the program to investigate biosynthetic pathways as illustrated in the example below.
Aromatic Interactions in Molecular Recognition
Christopher M. Baker and Guy H. Grant
Department of Chemistry, Physical and Theoretical Chemistry Laboratory, South Parks Road, Oxford, United Kingdom
Recent studies have revealed a number of cases in which aromatic residues play an important role in molecular recognition at the binding site [1], these cases include neuroreceptors, GPCRs, transporters and enzymes. It follows that the correct representation of the electrostatic interactions of aromatic molecules is fundamental to the understanding of many important molecular recognition processes.
Within the framework of molecular mechanics simulations the electrostatic interaction is typically represented by atom-centred point charges. For a polar molecule this will often be adequate, and if one is concerned solely with dynamic behaviour, averaged electrostatic energies, which are slowly changing with distance, will be reasonable.
Non-polar molecules, however, present a greater challenge. Consider, for example, benzene. Within the point charge representation, partial charges are placed on the carbon and hydrogen positions. The individual bond dipoles created by this charge distribution will sum to give the required zero dipole moment, however, this ignores the fact that benzene has an anisotropic charge distribution that results in a molecular quadrupole moment. And it is this quadrupole that determines the correct interaction geometry for the aromatic ring.
Hunter and Sanders [2] proposed a simple model that would take into account the non-atom centred charge distribution within aromatic systems. In the present work, this model is adapted for the treatment of benzene and in particular its dynamic behaviour via molecular dynamics simulation. Initial results suggest that this model offers a better reproduction of experimental results than is found for atom-centred potentials and as such offers a new insight into the dynamic behaviour of systems in which aromatic residues play an important role in molecular recognition
Following on from this the model will be applied to the modelling of dynamic behaviour in cases in which aromatic interactions play an important role in biomolecular recognition
References:
1. N. Zacharias and D. A. Dougherty, Trends in Pharmacological Sciences, 2002, 23, 6.
2. A. Hunter and J. K. M. Sanders, J. Am. Chem. Soc., 1990, 112, 5525
Development and Application of Novel Free Energy Methods to Rational Drug design
Sébastien Foucher, Dr. J. W. Essex, Dr. M. A. King, Dr. J. Frey (advisor)
Department of Chemistry, University of Southampton, Highfield, Southampton, SO17 1BJ
Over the past decades, free energy methods have become increasingly popular. [1] They offer valuable insights into a wide range of physical and biochemical processes. Owing to both theoretical advances and increasing computational power, they are now widely applied to complex biological systems such as protein/ligand interactions. [2] However, problems still occur (principally related to sampling and force-field), preventing them from being used routinely in rational drug design. The goal of this project is to develop novel methods and apply them to protein systems relevant to pharmaceutical projects. More specifically, a wide range of methods to calculate free energy differences have been described, such as the popular Free Energy Perturbation (FEP) [3], Thermodynamic Integration (TI) [4], λ Dynamics [5] or the recent non-equilibrium techniques [6]. A study conducted in the context of this project was to compare an efficient equilibrium method, Replica Exchange TI (RETI) [7] and a more recent and promising non-equilibrium method called Fast Growth (FG).
As a benchmark system, the alchemical transformation of methane into water within a box of explicit water molecules has been adopted. To our knowledge, this system is larger than any used in the literature on similar studies. All FG variants [6,8,9] described in the literature have been tested. The same simulation length has be used for all simulations. Particular attention has been paid to the study of systematic and statistical errors.
For the same amount of computational resources involved, results show that RETI always performs better than any non-equilibrium FG scheme. FG schemes may, however, become feasible in a context of a resource-rich distributed computing environment, in which a very large number of short calculations may be run on desktop PCs. In cases where computational resources are limited, equilibrium methods seem preferable.
References:
1. Simonson, T., Archontis, G. and Karplus, M., 2002, Acc. Chem. Res., 35, 430.
2. Kollman, P. A., 1993, Chem. Rev., 93, 2395.
3. Zwanzig, R. W., 1954, J. Chem. Phys., 22, 1420.
4. Kirkwood, J. G., 1935, J. Chem. Phys., 3, 300.
5. Kong, X. and Brooks III, C. L., 1996, J. Chem. Phys., 105, 2414.
6. Jarzynski, C., 1997, Phys. Rev. Lett., 78, 2690.
7. Woods, C. J., King, M. A. and Essex, J. W., 2003, J. Phys. Chem. B, 107, 13703.
8. Hummer, G., 2002, Mol. Sim., 28, 81
9. Ytreberg, F. M. and Zuckerman, D. M., 2004, J. Comp. Chem., 25, 1749.
Finding Rearrangement Pathways
Semen A Trygubenko, Tetyana V Bogdan and David J Wales
University Chemical Laboratories, Lensfield Road, Cambridge CB2 1EW
The potential energy surface (PES) governs the observed structure, dynamics and thermodynamics of any molecular system. It is often possible to gain new insight into these properties by expressing them in terms of stationary points of the PES. The number of stationary points on the PES generally scales exponentially with system size, which necessitates an appropriate sampling strategy of some sort for larger systems. In particular, to analyse dynamical properties a database of local minima and the transition states that connect them is usually constructed, which generally involves extensive use of single-ended and double-ended transition state searching techniques. Any path connecting two minima on PES can be broken down into elementary rearrangements, each of which involves a single transition state. The corresponding mechanism can be analysed in detail by calculating the two unique steepest-descent paths that lead downhill from the transition state. The objective of this presentation is to introduce our methods for finding rearrangement pathways, namely DNEB/L-BFGS and Connect algorithms. We also describe several interesting properties of the rearrangement pathways, which may aid new algorithms development and design. The main results of our work include the following: 1. Extensive tests show that the DNEB/L-BFGS combination provides a significant performance improvement over previous implementations. 2. Finding pathways in high-dimensional systems can become a challenging task. Some difficulties have been attributed to instabilities and inefficiencies in transition state searching algorithms, as well as the existence of very different barrier and path length scales. 3. Catastrophe theory predicts quadratic dependence of an energy functional on the order parameter in the vicinity of structural or phase change. 4. Uncooperative rearrangement pathways are usually harder to characterise using double-ended transition state searching algorithms since linear interpolation produces a poor initial guess. 5. It is possible to influence the degree of localisation of the pathways found using single-ended transition state searching algorithms.
References:
1. S. A. Trygubenko and D. J. Wales, 'A Doubly Nudged Elastic Band Method for Finding Transition States', J. Chem. Phys., 120, 2082-2094 (2004).
2. T.V. Bogdan and D.J. Wales. New Results for Phase Transitions from Catastrophe Theory. J. Chem. Phys., 120, 11090-11099 (2004).
Molecular surface point environments: A local descriptor for virtual screening and the elucidation of binding patterns (MOLPRINT 3D)
Andreas Bender, Hamse Y. Mussa, Gurprem S. Gill, Robert C. Glen
Unilever Centre for Molecular Science Informatics, Department of Chemistry, University of Cambridge, Cambridge CB2 1EW, United Kingdom
A novel method (MOLPRINT 3D) for virtual screening and the elucidation of ligand-receptor binding patterns is introduced which is based on environments of points on the molecular surface. Due to the presence or absence of features, fingerprinting of molecular surfaces becomes possible. These fingerprints can be combined with similarity coefficients, which have so far typically been used in combination with 2D descriptors. Employing a Naïve Bayesian Classifier and information-gain based feature selection, knowledge from multiple active structure can be combined.
The identification of active structures with minimal 2D similarity is facilitated, commonly referred to as 'scaffold hopping'. The descriptor uses points relative to the coordinates of the molecule which are uniformly binned, thus it is translationally and rotationally invariant. Due to its local nature, conformational variations do not cause major changes in the descriptor. This behaviour minimizes the number of active compounds missed in virtual screening for cases where a distinct conformation of an active compound is not present in the database. Features that are selected by the information-gain based feature selection step can be projected back on the molecular surface. They are shown to be consistent with experimentally determined binding patterns. Examples are given for angiotensin-converting enzyme inhibitors, 3-hydroxy-3-methylglutaryl Coenzyme A reductase inhibitors and Thromboxane A2 antagonists.
Bioinformatics and molecular modelling studies of membrane proteins
Shiva Amiri and Mark S. P. Sansom
Department of Biochemistry, University of Oxford, South Parks Road, Oxford, OX1 3QU
Membrane proteins make up approximately 30% of the human genome and primary targets for the majority of drugs on the market today. They are involved in a variety of significant tasks such as mediating neurotransmission, hormonal secretion, sensory transduction, and control of ion concentrations within a cell. Unfortunately, obtaining high resolution structures of these proteins, to study their function and improve drug design, remains a challenge. To date, there are approximately 80 structures of membrane proteins available, either as full length proteins or fragments in the form of discrete domains. Here, we present a method for taking partial crystallographic, NMR, and microscopic data, to generate complete structures of proteins or protein complexes. The method aligns the different domains of a protein according to a scoring function based on steric clashes and user-defined distance constraints. The model is energy minimized, and the final structure is stereochemically verified and can be used for further analysis of the protein, for example, one might be interested in the motions of the complete structure. We have generated the structure of the α7 nicotinic acetylcholine receptor (nAChR) from homology models of its separate domains. This receptor plays a role in nicotine addiction, cognitive function, and neurological disorders. The α7 nAChR structure was analyzed to investigate possible motions and electrostatic properties of the receptor. Computational coarse-grain motion studies using GNM and CONCOORD have been carried out on the full receptor to gain understanding of its structural variability and its mechanisms of function. The completed models are of great interest in structural studies as they give insight into plausible gating movements and determinants of selectivity. The method is generic and can be extended to other proteins and protein complexes where separate domains have been solved but not the full length structure.
Strategies for ACE2 Structure-Based Inhibitor Design
Monika Rella and Richard M. Jackson
School of Biochemistry and Microbiology, University of Leeds, Leeds, LS2 9JT
Angiotensin-Converting Enzyme (ACE) is an important drug target for hypertension and heart disease. Recently, a new ACE homologue has been identified in human (1), termed ACE2 whose physiological and pathophysiological roles are currently under investigation. ACE2 has been associated with hypertension, heart and kidney disease and a counter-regulatory role to ACE has been proposed which makes it an interesting new cardio-renal disease target.
Based on the solved ACE2 crystal structure with its inhibitor, a structure-based drug design project is undertaken to identify novel potent and selective inhibitors that would help to investigate ACE2's function in vivo. The general long-term strategy comprises computational structure-based drug design approaches, chemical synthesis of promising candidates and bioassay-based potency evaluation. Computational approaches involve (1) crystal structure quality assessment, (2) binding site analysis for favourable interaction sites, (3) combinatorial library design by substituting side chains on a common core as well as (4) de-novo design and (5) pharmacophore-based virtual screening of compound databases.
A small number of synthetically accessible fragments were selected and individually evaluated for improved interaction energy using our in-house rigid-body docking tool Q-fit (2). In a second step, the de-novo design tool SynSPROUT (3) was applied for replacing the original side chain of the core structure with each fragment and docking the whole molecule. Currently, we are preparing a combinatorial library on a larger scale by mimicking a synthetic reaction that involves synthetic starting material such as boronic acid derivatives and any naturally occurring amino acid. R-groups suitable for chemical synthesis will be assessed and prioritised via docking as described above.
References:
1. Tipnis SR, Hooper NM, Hyde R, Karran E, Christie G, Turner AJ. A human homolog of angiotensin-converting enzyme. Cloning and functional expression as a captopril-insensitive carboxypeptidase. 2000, 27:275(43):33238-43
2. Jackson RM. Q-fit: a probabilistic method for docking molecular fragments by sampling low energy conformational space. J Comput Aided Mol Des. 2002, 16(1): 43-57
3. Boda K. SynSPROUT: Generating synthetically accessible ligands by de novo design. PhD thesis 2002, School of Chemistry, University of Leeds
