PROGRAM

Sponsored by

2

THURSDAY OCTOBER 1ST

8:30-9:00 Registration 9:00-9:15 Opening session:

Dr. Jordi Alberch, Vice Chancellor, Universitat de Barcelona

Dr. Pere L. Cabot, Dean, faculty of Chemistry, universitat de

Barcelona Dr. Francesc Illas, Universitat de Barcelona and XRQTC Director,

Spain

Session 1. Chairwoman: Dr. Nuria Lopez, Senior Group Leader, ICIQ, Spain 9:15-9:55 Titania in cement and construction industry: the contribution of

modeling

Dr Gianfranco Pacchioni, Professor, University of Milano Bicocca,

Italy. 9:55-10:35 Multiscale modelling of advanced materials with applications in

energy storage, military defence and carbon nanoscience Dra. Elena Bichoutskaia, Professor, University of Nottingham,

United Kingdom 10:35-11:10 Electron transport in low-dimensional systems for electronic and

optoelectronic device simulations Dr. Albert Cirera, Professor, University of Barcelona, Spain

11:15-11:45 Coffee-Break

Session 2. Chairman: Dr. Daniel Fernández Hevia, INAEL Electrycal Systems

S.A., Toledo, Spain 11:45-12:25 Modeling-Guided Catalyst Design for Fischer-Tropsch synthesis:

Structure, Activity, Selectivity and Stability Dr. Mark Saeys, Professor, University of Ghent. 12:25-13:05 Modeling the industrially relevant heterogeneous Ziegler-Natta

catalytic systems Dr. Luigi Cavallo, Professor, King Abdullah University of Science

and Technology

3

13:05-13:45 DFT studies of organic compounds over metal oxide surfaces

Dr. Hicham Idriss, Research Fellow, SABIC Saudi Basic Industries

Corporation, Saudi Arabia.

13:45-15:15 Lunch networking and Poster Session Session 3. Chairwoman: Dra. Mariona Sodupe, Professor, Universitat

Autònoma Barcelona, Spain 15:15-15:55 Quantum chemistry meets polymer reaction engineering.

Dr. Peter Deglmann, Manager research scientist, BASF SE,

Germany. 15:55-16.35 Molecular dynamics explorations of active site structure in

designed and evolved enzymes Dra. Silvia Osuna, Research Scientist, IQCC Institute, University of

Girona, Spain 16:35-17:00 Coffee-Break

Session 4. Chairman: Dr. Ramón Crehuet, Scientist, IQAC-CSIC, Spain 17:00-17:40 Biohybrids as novel catalysts: On the role of molecular modeling

Dr. Jean Didier Marechal, Lecturer of Physical Chemistry,

Universitat Autonoma de Barcelona, Spain 17:40-18:20 Chemical interaction of DNA with complex minerals

(hydroxyapatite)

Dr. Pau Turon, R&D, Regulatory Affairs and Quality Director, B.

Braun Surgical, S.A, Spain.

4

FRIDAY OCTOBER 2nd

9:00-9:30 Registration

Session 1. Chairwoman: Dr. Mireia Olivella García, Professor, Vic

University, Spain

9:30-10:10 Understanding ligand binding, receptor selectivity and

pharmacological profiles on GPCRs from computer simulations

Dr. Hugo Gutierrez de Teran, Researcher, Uppsala Universitet,

Sweden.

10:10-10:50 Structural and dynamic basis of G protein-coupled receptor

activation

Dr. Xavier Deupí, Senior Scientist, Paul Scherrer Institute,

Switzerland.

10:50-11:20 Coffee-Break

Session 2. Chairman: Dr. Leonardo Pardo, Professor, Universitat Autonoma

Barcelona, Spain

11:20-12:00 Integrating Chemical and Biological Data for Compound

Selection and Mode-of-Action Analysis

Dr. Andreas Bender, Lecturer, University of Cambridge, United

Kingdom.

12:00-12:40 Discovery of non-competitive pharmacological chaperones

using structure-based methods

Dra. Elena Cubero, Senior Scientist, Minoryx Therapeutics,

Barcelona, Spain

12:40-14:00 Cocktail Lunch networking and Poster Session

Session 2. Chairman: Dr. Gianni De Fabritiis, Associate Professor,

Universitat Pompeu Fabra, Spain

14:00-14:40 Structural chemogenomics databases to better understand

protein-ligand interactions

Dr. Iwan de Esch, Full Professor, Medicinal Chemistry, VU

University of Amsterdam.

5

14:40-15:20 Fragment Libraries at Astex and their Application to PPI Targets

Dr. Gianni Chessari, Director of Computational Chemistry, Astex

Pharmaceuticals, Cambridge, UK.

15:20-15:30 Delivery of Prizes for the Poster Session Contest and Concluding

remarks

6

Titania in cement and construction industry: the contribution of modeling

Gianfranco Pacchoni Dipartimento di Scienza dei Materiali, Università di Milano-Bicocca, Via R. Cozzi, 55, 20125, Milano, Italy – gianfranco.pacchioni@unimib.it Titania is an essential component of devices of new generation for photocatalysis and solar energy conversion. Its special behavior under illumination is at the basis of several practical applications like self-cleaning and self-sterilizing surfaces, superhydrophilicity, corrosion protection, etc. Most of these effects are observed under ultra-violet (UV) light and efforts are now oriented to the preparation of visible light photoactive titania via doping and nanostructuring. In this talk we will review the most recent advances in this field, with particular attention to concrete examples of applications of this material in buildings, roads, hospitals, smart-windows for self-cleaning and energy savings. We will address the basic mechanisms which are responsible for the photoactivity of titania and the role of ab initio modeling for the cement industry. In particular, we will discuss the current strategies to obtain doped titania nanoparticles as additives for cement active under visible solar light.

7

Multiscale modelling of advanced materials with applications in energy storage, military defence and carbon nanoscience

Elena Bichoutskaia School of Chemistry, University of Nottingham, University Park, Nottingham NG7 2RD, Nottingham, United Kingdom Elena.Bichoutskaia@nottingham.ac.uk; +44 115 846 8465 In this talk we overview recent advances in multiscale modelling of novel materials with particular emphasis on applications in (1) selective sorption of carbon dioxide and methane gas in metal-organic frameworks1,2; (2) mechanical behavior of lightweight high performance ceramic/metal composites under static and dynamic (shock wave propagation) loading3,4; electron beam irradiation induced processes in transmission electron microscopy5,6. References 1) S. Yang, X. Lin, W. Lewis, M. Suyetin, E. Bichoutskaia et al. Nature Materials 11, 710 (2012). 2) Y. Yan, M. Suyetin, E. Bichoutskaia, A. J. Blake, D. R. Allan, S. A. Barnett, M. Schröder, Edge Article, Chemical Science 4, 1731 (2013). 3) E. I. Volkova, A. Jones, R. Brooks, Y. Zhu, E. Bichoutskaia, Physical Review B 86, 104111 (2012). 4) E. I. Volkova, A. Jones, R. Brooks, Y. Zhu, E. Bichoutskaia, Composite Structures 96, 601 (2013). 5) S. T. Skowron, I. Lebedeva, A. Popov, E. Bichoutskaia, Feature Article, Nanoscale 5, 6677 (2013). 6) A. Santana, A. Zobelli, J. Kotakoski, A. Chuvilin, E. Bichoutskaia, Physical Review B 87, 094110 (2013).

8

Optoelectronic Simulations for Quantum Dot based Devices

Illera S., Garcia N., Prades J. D and Cirera A. MIND/IN2UB Departament d’Electrònica, Universitat de Barcelona, c/ Martí i Franquès 1, E-08028 Barcelona, Spain. acirera@ub.edu Strong confined structures, such as the Quantum Dots (QDs), were used as promising nanostructures to develop the third generation of photovoltaic solar cells due to their size dependent energy band gap and the increased optical absorption. These devices are composed by a large array of embedded QDs in an insulator matrix as a top structure of a classical p-n junction creating a tandem solar cell. Within this structure, two different optical absorption energy edges are obtained increasing the efficiency of the cell. However, the efficient extraction of the photogenerated carriers in the QD matrix imposes new technological and material requirements. Besides, an electron transport capable to deal with these large QDs arrays is also necessary. Here, we present an electronic transport model based on Transfer Hamiltonian Formalism and rate equations to describe the ballistic charge transport in QDs embedded in an insulator matrix1. This methodology was compared to NEGFF reproducing the same theoretical trends (2) and it also demonstrated that can be easily scalable for large systems (3). Two unique features of this transport model are: (i) the model is based on just a few basic material parameters and on the device geometry; (ii) it is simple enough to tackle problems involving large number of Qd, which is the case of real devices. Moreover, it is compatible with ab initio theories taking advantage of the atomistic calculations in order to accurately describe the Qd electrical properties as we have demonstrated (4,5). Once the electronic transport model was presented and validated, the light interaction was also included making possible to describe and simulate optoelectronic QD based devices. Thus, a complete and valuable theoretical tool based on low-level material and geometrical parameters are developed which can be used to design and predict the optoelectronic response of these devices. 1) S. Illera, J. D. Prades, A. Cirera and A. Cornet EPL 98, 17003 (2012).

9

2) S. Illera, N. Garcia-Castello, J. D. Prades and A. Cirera, J. Appl. Phys 112, 093701 (2012). 3)S. Illera, J. D. Prades and A. Cirera, J. Appl. Phys. 117, 174307 (2015). 4)N. Garcia-Castello, S. Illera, R. Guerra, J. D. Prades, S. Ossicini and A. Cirera, Phys. Rev B 88, 075322 (2013). 5)N. Garcia-Castello, S. Illera, J.D. Prades,, S. Ossicini, A. Cirera and R. Guerra, Nanoscale 7, 12564 (2015).

10

Modeling-Guided Catalyst Design for Fischer-Tropsch synthesis: Structure, Activity, Selectivity and Stability

Mark Saeys Laboratory for Chemical Technology, Ghent University, Belgium Catalyst design and kinetic modeling often start from molecular-scale hypotheses about the reaction mechanism, the structure of the active sites and the nature of the rate and selectivity determining steps. Computational catalysis has become a crucial tool to analyze molecular-scale concepts and elucidate their electronic origin. In combination with characterization and experimental kinetic validation, insights gained from computational catalysis can be translated all the way to the industrial scale. This pas-de-deux between experiment and theory is becoming the new paradigm in catalyst design and kinetic modeling, both in academia and in industry. In this presentation, I will illustrate how this approach can contribute to different aspect of catalysis research. The nature of catalytically active sites under reaction conditions often differs dramatically from the clean ideal surface. Using operando computational catalysis and insights into chemical bonding, we showed that the spontaneous formation of Co nano-islands during FT synthesis is driven by the stability of unusual sigma-aromatic, square planar carbon species. Insight into the structure of the active sites provides the basis to elucidate the reaction mechanism. Again using operando computational catalysis, we developed a novel kinetic model that agrees with the experimentally measured kinetic parameters. This in turn forms the basis for to design catalyst with enhanced selectivity and stability. The success of this approach is illustrated with the discovery of a boron promotor that enhances the stability of cobalt catalysts during Fischer-Tropsch synthesis of clean fuels by an order of magnitude. References Zhuo, Borgna, Saeys, J Catal 297, 217, (2013) Banerjee, Kuipers, Van Bavel, Saeys, ACS Catal., (2015) Tan, Chang, Borgna, Saeys, J. Catal., 280, 50 (2011)

11

Modeling the industrially relevant heterogeneous Ziegler-Natta catalytic systems

Luigi Cavallo King Abdullah University of Science and Technology (KAUST), Physical Sciences and Engineering Division, Kaust Catalysis Center, Thuwal 23955-6900, Saudi Arabia. luigi.cavallo@kaust.edu.sa Despite of polyolefin commodities have a huge economic impact, with total yearly volume of 106 tons and a billionaire market, knowledge of the active site of Ziegler-Natta (ZN) catalysts at molecular level remains elusive. This limited comprehension is due to the complex nature of ZN-catalysts, although only four ingredients are fundamental to compose an industrial catalyst: i) the inert MgCl2 support; ii) the catalytically active TiCl4 adsorbed on the MgCl2 surfaces; iii) an Al-alkyl, typically AlEt3, to activate the adsorbed TiCl4; iv) a Lewis base, either monodentate or bidentate, to improve catalytic performance by increasing the amount and stereoregularity of the produced polypropylene. Due to the difficulties in the experimental characterization of these catalysts, remarkable advances in our comprehension of Ziegler-Natta catalysts have been achieved via computational chemistry, which has been used to test the validity of models proposed on the basis of experiments, as well as to explore new models driven by computational evidences. In this communication we will give an overview of the computational work in the field, from the stability of different perfect and defective (104) and (110) facets of the MgCl2 support, on the adsorption of Lewis bases on these facets of the MgCl2 support with a possible impact on the morphology of the formed MgCl2 crystallites, to the adsorption of TiCl4 on perfect and defective facets of the MgCl2 support, to the stereoselectivity of possible active sites on both perfect and defective surfaces.

12

Figure 1. Model of a MgCl2 monolayer, with indication of the (104) and (110) lateral cuts.

13

DFT studies of organic compounds over metal oxide surfaces

Hamdan Al-Ghamdia, Y. Al-Salika, Paul Bagus (b) and Hicham Idriss (a) a) SABIC-CRI at KAUST, Saudi Arabia. b) University of Northern Texas, Denton, TX, USA Surface reactions of metal oxides are relevant to many catalytic processes related to renewables and petrochemical industries. In this work, examples of our studies linking computational results to experimental ones are given for metal oxides model surfaces. In particular, we are focusing on TiO2, CeO2, and UO2 surface or bulk interactions with organic compounds or water. TiO2 is used as it is the prototype semiconductor for photocatalytic water splitting [1], CeO2 is mostly considered for its oxygen transport properties [2] and UO2 because of its unique chemical reactions (due in part to its 5f orbitals) [3]. The mode of interaction of organic adsorbates, used as sacrificial agents, with the surface of TiO2, during water splitting to hydrogen, is crucial to understand the relevant surface coverage under reaction conditions and its effect on the overall reaction rate. In the case of CeO2 we focus on the effect of doping with metal cations in order to reduce the energy needed to create oxygen vacancies via charge transfer mechanism [4, 5]. A correlation is established between the effect of dopants on lowering the energy needed for the creation of oxygen vacancies and thermal hydrogen production from water [6]. In addition, an example of comparing the electronic core level of the Ce3d,4d lines experimentally and those computed using ab initio methods [4, 7] is given. [1] A. Kudo, Y. Miseki, Chem. Soc. Rev., 2009, 38, 253–278 [2] M. B. Watkins, A. S. Foster, A. L. Shluger, J. Phys. Chem. C 2007, 111, 15337-15341 [3] H. Idriss, Surf. Sci. Rep. 2010, 5, 67-109 [4] Y. Al-Salik, I. Al-Shankiti, H. Idriss, J. Electron Spectrosc Relat Phenom, 2013, 194, 66–73 [5] B. E. Hanken, T. Y. Shvareva, N. Grønbech-Jensen, C. R. Stanek, M. Asta, A. Navrotsky, Phys. Chem. Chem. Phys., 2012, 14, 5680–5685 [6] J. Scaranto, H. Idriss Top. Catal., 2015, 58, 143–148

14

[7] P. S. Bagus, C. J. Nelin, Y. Al-Salik, E. S. Ilton, H. Idriss, Surf. Sci. in press doi:10.1016/j.susc.2015.06.002

Quantum Chemistry Meets Polymer Reaction Engineering

Peter Deglmann, Volker Settels BASF SE, Advanced Materials & Systems Research, Carl-Bosch-Str. 38, 67056 Ludwigshafen, Germany The architecture and thus the properties of a polymer depend in a complex way on the conditions applied within the polymerization process, e.g. temperature, reaction medium, feed of reactants, reactor type (batch, semibatch, continuous). Therefore it is not surprising that a simulation of polymerization processes is of enormous industrial importance. One key ingredient in such simulations represents the knowledge of rate coefficients of all potentially relevant elementary reactive steps involved. Whereas it is a standard task to monitor quantities like the overall conversion during a polymerization process, it is highly difficult – if not impossible – to experimentally measure all elementary rate coefficients individually. For this reason it is very desirable to be able to predict them via ab-initio calculations. In this presentation, strategies and challenges for such a computation of rate coefficients in the condensed phase (encountered e.g. upon solution or bulk polymerization) are discussed. One huge obstacle for a direct quantitative prediction of reaction kinetics represent the exponential dependence on both energy and entropy of activation. Whereas for gas phase reactions of small molecules quantum chemical predictions have reached and even well bypassed the limit of chemical accuracy (1 kcal/mol), this is typically not the case for reactions in solution. Thus, a reliable calculation of solvation thermodynamics represents a critical issue when applying quantum chemistry to real world problems. The COSMO-RS method is presented here as a convenient access to Gibbs free energies of solvation as – like the quantum chemical calculation itself – it does not require any chemistry specific parameterization. Furthermore, as will be shown, automation of the resulting elaborate workflows represents an important feature to productively employ quantum chemistry in an industrial environment.

15

16

Molecular Dynamics Investigations of Active Site Structure in Designed and Laboratory-generated Enzymes

Sílvia Osuna,a,b) Gonzalo Jiménez-Osés,b) and K.N. Houkb) a) Institut de Química Computacional i Catàlisi and Departament de Química, Universitat De Girona, Spain b) Department of Chemistry and Biochemistry, University of California, Los Angeles (UCLA), USA In this study, the use of molecular dynamics (MD) simulations to reveal how mutations alter the structure and organization of enzyme active sites is presented.[1] As initially proposed by Pauling, and elaborated by many others since then, biocatalysis is efficient when the catalytic residues in the active site of an enzyme are in optimal positions for transition state stabilization. Using MD simulations, we explore the dynamical pre-organization of the active sites of designed and evolved enzymes, by analyzing the fluctuations between active and inactive conformations normally concealed to static crystallography. MD shows how the various arrangements of active site residues influence the free energy of the transition state, and relates the populations of the catalytic conformational ensemble to enzyme activity (see Figure 1). The importance of dynamics in evaluating a series of computationally designed and experimentally evolved enzymes for the Kemp elimination, a popular subject in the enzyme design field, is first presented. Finally, we show how microsecond MD has been used to uncover the role of remote mutations on the active site dynamics and catalysis of a transesterase, LovD, a useful commercial catalyst for the production of the drug simvastatin.[2]

17

Figure 1: Correlation between the population of the triad in catalytically competent conformations, and catalytic activities of the LovD mutants. [1] S. Osuna, G. Jiménez-Osés, E. L. Noey, K. N. Houk, Acc. Chem. Res. 2015, 48, 1080-1089. [2] G. Jiménez-Osés, S. Osuna, X. Gao, M. R. Sawaya, L. Gilson, S. J. Collier, G. W. Huisman, T. O. Yeates, Y. Tang, K. N. Houk, Nat. Chem. Biol. 2014, 10, 431-436.

18

Biohybrids as novel catalysts: On the role of molecular modeling

Jean-Didier Maréchal Departament de Química, Universitat Autònoma de Barcelona, Edifici C.n., 08193 Bellaterra, Spain Merging synthetic compounds with biological entities is a concept increasingly explored to expand the scope of enzymatic reactions. Amongst the most interesting biohybrids constructed to date are those that consist in the integration of organometallic scaffolds into biological frameworks (i.e., protein, DNA or peptides). The resulting artificial metalloenzymes are interesting at least in two aspects: 1) they lead to biocompatible catalysts that absent from the biological reign and 2) they allow catalytic specificities and selectivities that are unreachable by standard homogenous approaches. Both advantages are opening new horizons for greener approaches in chemical synthesis. Despite the major successes of several groups in developing efficient biometallic hybrids, the prediction and analysis of their molecular behavior still represents a complex exercise. The lack of evolutionary pressure generally leads to a first generation of molecules with relatively low stability and difficult structural characterization. Moreover, the identification of the best complementarities among biological receptor, organometallic cofactors and substrates implies a major combinatorial space that challenges biochemical and chemical intuitions. Virtually, molecular modeling could be of the best allies in this field since computational methodologies can deal with processes related to molecular recognition and catalytic mechanisms. However, the modeling of artificial metaloenzymes stands out of the scope of standard approaches and novel methodologies are needed. In the recent years, our group designed, tested and applied a series of computational strategies in the field of artificial bioinorganics. From protein-ligand dockings to multi-scale approaches, our work allowed to better understand the molecular mechanism of artificial metalloenzymes, provided information on how they mechanistically diverge from natural

19

ones and gave some hints on how we could computationally guide the design of new candidates. In this talk, I will briefly present the underpinning concepts of our strategies and the most important results obtained so far both from pure computational works and in collaboration with experimentalists.

Literature: V. Muñoz Robles, E. Ortega-Carrasco, L. Alonso-Cotchico, J. Rodríguez-Guerra, A. Lledós, A. and J.-D. Maréchal, ACS Catal. 2015, 5, 2469.

20

Chemical interaction of DNA with complex minerals (hydroxyapatite)

Pau Turon Dols Research and Development, Regulatory Affairs and Quality Managament dept. Centre of Excellence Closure Technologies. B. Braun Surgical, S.A. Ctra. Terrassa, 121, Rubí, Barcelona, Spain – pau.turon@bbraun.com In the last decade, the combination of hydroxyapatite (HAp) with biopolymers has captured the attention of the researchers, mainly due to its medical applications. The system DNA-HAp (hydroxyolite) presents outstanding features. Recently, it has been confirmed at atomistic level that DNA templates HAp crystal and, additionally, HAp includes ions in its lattice, such as Mg2+, that are able to stabilize DNA modulating its structure and dynamics. We highlight that hydroxyapatite encapsulates DNA and protects it against environmental aggressions. The main consequence is that the information that DNA contains is preserved through time. DNA remains functional, encapsulated inside HAp, ready to be reintroduced into the mainstream of life. Cell transfection process, intended to incorporate a plasmid into the cell nucleus, can be achieved through hydroxyolites but its efficiency is low. Atomistic molecular dynamics simulations can shed light on how these processes work and how they can be improved.

21

Understanding ligand binding, receptor selectivity and pharmacological profiles on GPCRs from computer simulations

Hugo Gutiérrez de Terán Department of Cell and Molecular Biology, Uppsala University. BMC, Box 596, 751 24, Uppsala (Sweden). Hugo.gutierrez@icm.uu.se Receptors belonging to the superfamily of G-protein-coupled receptors (GPCRs) play a key role in the cellular communication, a reason why they are of the highest interest as drug targets in a widespread repertoire of diseases. The recent availability of crystal structures, combined with the outcome of molecular biology experiments such as site-directed mutagenesis, provide excellent starting points to examine in deep detail the molecular determinants of ligand binding. In our research group, we combine homology modeling, protein-ligand docking and molecular dynamics simulations to assess in the design of compounds with specific selectivity and pharmacological profiles for different GPCRs. In this presentation, I will introduce our recently developed computational scheme, based on free energy perturbation (FEP) simulations, to evaluate the effects of single point mutations on ligand binding.1-3 This method is generalized to deal with any amino acid mutation, and is illustrated in the investigation of both agonist and antagonist binding in different GPCRs, namely adenosine A2A receptor and the neuropeptide Y1 receptor, with excellent results. This computational protocol is useful not only for the characterization of ligand selectivity in GPCRs, but also to address problems of drug resistance in pathogens as well as individual responses to drug treatment due to genetic variation. Finally, I will show how a combination of accurate homology modeling, ligand docking and molecular dynamics simulations can successfully explain the pharmacological profiles (i.e., agonist versus antagonist) and selectivity issues on GPCRs, which is illustrated in the case of angiotensin receptors.4

22

References 1. Keranen, H.; Åqvist, J.; Gutiérrez-de-Terán, H. ChemComm (2015), 51:3522 2. Keranen, H.; Gutiérrez-de-Terán, H.; Åqvist, J. PlOS One (2014)

9:e108492 3. Boukharta, L.; Gutiérrez-de-Terán, H.; Åqvist, J. PlOS Comp. Biol (2014) 10:e1003585. 4. Sallander, J.; Wallinder, C.; Hallberg, A.; Åqvist, J.; Gutiérrez-de-Terán,

H.; submitted.

23

Detection of ligand-induced conformational changes in GPCR activation by NMR

Xavier Deupi Condensed Matter Theory Group and Laboratory of Biomolecular Research, Paul Scherrer Institute, WHGA/114, 5232 Villigen PSI, Switzerland. G protein-coupled receptors (GPCRs) are a large family of transmembrane proteins that trigger cellular signaling responses upon binding of extracellular ligands. Thus, GPCRs act as transmission devices between the environment and the cell interior and, due to this key physiological role, they constitute one of the most important pharmaceutical targets. However, despite recent breakthroughs in GPCR crystallography, the structural and mechanistic aspects of GPCR activation by drugs are not yet well understood. This is partly due to missing dynamical information, which can, in principle, be provided by NMR. However, only limited information of functional relevance on few side chain sites of eukaryotic GPCRs has been obtained to date. Together with the group of Prof. Grzesiek (Biozentrum, University of Basel), we have recently shown that receptor motions can be followed in stabilized mutants of the β1-adrenergic receptor (β1AR). We observe that the response to various ligands is heterogeneous in the vicinity of the extracellular binding pocket, but gets transformed into a homogeneous readout at the intracellular side of the receptor. By analyzing the effect of several mutations, we conclude that even a fully stabilized receptor is able to undergo certain activating motions, but the fully active state is only reached in presence of (a) two specific key residues (Y5.58 and Y7.53), and (b) a stabilizing partner in the cytoplasmic side (e.g. an antibody that mimics the cognate G protein). Our analysis allows us to identify crucial connections in the allosteric signal transmission pathway of ligand-induced GPCR activation, and represents a general experimental method to delineate signal transmission networks at high-resolution in GPCRs.

24

Integrating Chemical and Biological Data for Compound Selection and Mode-of-Action Analysis

Andreas Bender Centre for Molecular Informatics, Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge CB2 1EW. ab454@cam.ac.uk Recent technological advancement in the field of health science brought with it a deluge of both chemical and biological data that is in need of interpretation. This information can be used to generate better understanding of drivers of disease – such as chemical, biological and/or genetic markers – but possibly even more importantly it can be used to design more efficacious and safe medicines prospectively. Still, currently data is often not used effectively to reach this goal, partly due to technical reasons (the sheer amount of data), but even more fundamentally also due to lack of data integration (e.g. inconsistent identifiers), and our often crude understanding of what in particular biological readouts actually mean in order to make better decisions based on them. In our research group, which currently comprises about 20 members, we aim to address the above point by integrating data from across the chemical and biological domains in order to improve decision making in the drug discovery process. To this end, project-specific data is employed (often in collaboration with pharmaceutical companies, such as AstraZeneca, Johnson&Johnson, Eli Lilly, BASF, Unilever, Aboca, and others) in order to address one of two principal aims: Firstly, we aim to decide which compound is most likely to possess desired efficacy and the relatively best side effect profile, given the data at hand. Secondly, we aim to understand the mode-of-action, or more generally biomodulatory capabilities, of a small molecule by integrating diverse data (such as on-target bioactivity data and RNA-Seq data, but also others). Given a large number of collaborations with both academic groups as well as pharmaceutical, chemical, and consumer goods companies we have accumulated significant expertise in the area of chemical and biological

25

data integration in order to support decision making in the drug discovery process in the above two areas (but also beyond). Our presentation will conceptually describe which types of data we are currently integrating, our plans for future projects, as well as case studies where data from across the chemical and biological domains was able to improve compound design, as well as enhance our understanding of a more complete bioactivity profile of chemical structures.

26

Discovery of non-competitive pharmacological chaperones using structure-based methods

Dra. Elena Cubero, Senior Scientist, Minoryx Therapeutics, Barcelona, Spain Many monogenic diseases are characterized by the presence of missense mutations which affect the folding and stability of a key enzyme, which is then degraded by the quality control machinery of the cell. This deficiency of enzymatic activity is what originates the diseases. Pharmacological chaperones are a new class of small molecule drugs, which showed great potential for the treatment of such genetic diseases. They prevent the degradation of unstable enzymes with missense mutations, thus producing an enzyme enhancement effect normalizing enzymatic activity, which reverses and/or prevents disease progression. There are several examples where pharmacological chaperones showed excellent efficacy in preclinical models. Nonetheless, success in clinical development has been somehow modest. Pharmacological chaperone therapy has been largely experimented on the lysosomal storage diseases, starting with the pioneer work on Fabry disease in 1999. Most Pharmacological chaperones (PC’s) described until now are substrate analogues which bind to the active site of the target protein. Consequently, such PC’s also inhibit the target protein at higher concentrations thus rendering a narrow therapeutic window and have poor drug-like properties. Through our proprietary technology platform SEE-Tx™, we identify a new generation of non-substrate competitive pharmacological chaperones which potentially offer a much broader therapeutic window. What’s more, such compounds are not substrate analogues, thus presenting much better drug-like properties, particularly for indications with CNS involvement. Here we present our methodology to identify non-competitive pharmacological chaperones applied to the enzyme beta-galactosidase, whose deficiency is related with GM1 Gangliosidosis and Morquio B.

27

Structural chemogenomics databases to better understand protein-ligand interactions

Iwan de Esch Faculteit der exacte wetenschappen, VU University of Amsterdam, Netherland. i.de.esch@vu.nl The high number of protein structures that are co-crystallized with ligands makes it possible to study cross-family ligand binding features in unprecedented detail. We present a thorough analysis of kinase-ligand and phosphodiesterase-ligand interaction patterns. For this extensive analysis, we have constructed a database of all available aligned human kinase-ligand co-crystal structures and all phosphodiesterase-ligand co-crystal structures that are present in the Protein Data Bank (PDB). Next, we used molecular Interaction Finger Prints (IFP) to annotate the different protein-ligand interaction features. The resulting databases contain consistent alignments and numbering these two protein classes and enable the identification of family- or group-specific interaction features and classification of ligands according to their binding modes. We will illustrate how systematic mining of protein-ligand interaction space gives novel insights into how conserved and selective interaction hot spots can accommodate the large diversity of chemical scaffolds. These studies lead to an improved understanding of the structural requirements of ligand binding that will be useful in future drug discovery, drug design studies and poly-pharmacology approaches.

28

Fragment libraries at Astex and their application to protein-protein interaction targets

Gianni Chessari Astex Pharmaceuticals, 436 Cambridge Science Park, Milton Road, Cambridge, CB4 0QA, UK Fragment-based methods are now well established for medicinal chemistry programmes and small molecule drug discovery campaigns. Successful Fragment-Based Drug Discovery (FBDD) requires a reliable and effective fragment library, sensitive screening methods and expertise in structure based drug design to transform weak fragment hits into potent leads. At Astex we have invested time and resources to optimise and integrate each key aspect of FBDD into our screening platform, which has delivered several compounds into the clinic. Here we describe how we have evolved and improved our fragment library over time using information from multiple fragment screening campaigns and thousands of proprietary protein-fragment crystal structures. We will show how the fragment hit rate varies with lipophilicity and heavy atom count. We will also discuss the concept of minimal binding pharmacophore elements for fragments and show how they have been used to drive library enhancement. Finally, we will present examples of Astex Protein-Protein Interaction projects, where fragment hits have been developed into potent lead molecules.

29