POSTER SESSION

P1

Binding of the heme-Dependent activator YC-1 into the active catalytic DOMAIN of soluble guanylate cyclase Luis Agulló1, Ignasi Buch2, Hugo Gutiérrez de Terán3, Gianni de Fabritis2, David Garcia-Dorado4, Jordi Villà-Freixa1 1U_Science Tech, Universitat de Vic, Vic, ES, 2Hospital del Mar Medical Research Institute (IMIM), Barcelona, ES, 3Uppsala University, Uppsala, SE, 4Vall d'Hebron Research Institute, Barcelona. E-mail: luis.agullo@uvic.cat Soluble guanylate cyclase (sGC), the main target of nitric oxide (NO), has been proven to have a significant role in coronary artery disease, pulmonary hypertension, erectile dysfunction and myocardial infarction. Several drugs that increase the activity of this enzyme are now in clinical phase of development: some of them are heme-dependent and might interact with the catalytic domain and others are heme-independent and supposedly bind to the sensory domain. The absence of reliable structural information is one of the factors that have precluded knowledge of the precise site of interaction of these molecules and of the mechanism of activation of the enzyme. Homology models of the catalytic domain of sGC in 'inactive' or 'active' conformation were constructed using, for the β-chain, the structure of recently published crystal of a nonphysiological homodimer of β subunits of human guanylate cyclase (2WZ1), for the α-chain, a similar domain of the green algae Chlamydomonas reinhardtii (3ET6) and, for monomer arrangement, the sGC 'inactive' structure (3ET6) or the 'active' catalytic domain of adenylate cyclase (1CJU). Molecular dynamics simulations of about 1μs each where run on all relevant models (NAMD/ACEMD, Amber99SB). In the different trajectories, sGC conformation varied between having 1CJU- and 3ET6-like structures. One of these trajectories maintained extremely stable relative positions of the aminoacids in the catalytic site, being very similar to those described in 1CJU. The observed conformational transitions suggest a possible mechanism for the transmission of the cooperativity signal between the pseudo-symmetric and the catalytic site, in which Arg-92 (α-chain) and Arg-539 (β-chain) and the loop β2-β3 seem to play a critical role. Docking of YC-1, a classic heme-dependent activator, to all frames of this trajectory and absolute binding free energies with the linear interaction energy method (LIE) for selected poses revealed one potential binding site located between pseudo-symmetric and catalytic sites just over the loop β2-β3. This site would be compatible with the binding of a second GTP or an inhibitory ATP to the pseudo-symmetric site.

P2 Triarylmethyl radicals: potential building blocks for molecular spintronics Isaac Alcón,* Daniel Reta, Iberio de P.R. Moreira, Stefan T. Bromley Departament de Química Física and Institut de Química Teòrica i Computacional (IQTC-UB), Universitat de Barcelona, 08028 Barcelona, Spain ialcon@ub.edu Triarylmethyls (TAMs) are the oldest family of organic radicals and, since their discovery by Gomberg in 1900,1,2 hundreds of triarylmethyls have been synthesized and widely utilized for many different applications.3–5 All TAMs are composed of three aryl rings bonded to a central carbon atom, where their unpaired electron mainly resides (see figure below). In the last few years their potential for Molecular Spintronics has been probed.6–9 This field tries to take advantage of the electronic spin and charge in organic systems for the preparation of new devices and materials.10,11 As demonstrated in those studies, due to the aromatic radical nature of TAMs, these molecules present relatively high electrical conductivities8 and, moreover, their unpaired electron interacts with the pi-conduction channel giving rise to very interesting phenomena for Spintronics such as magnetoresistance effects.9 In the first part of our project, we have studied what factors mainly influence the distribution (and thus energy) of the unpaired electron in triarylmethyls. Our results demonstrate the existence of a general and robust spin/structure relationship in these radicals that could be extremely useful for the design and preparation of organic materials with controllable magnetic and electrical properties. In the second part of our project, we have designed a TAM-based 2D material where, by external strain, its magnetic properties can be finely tuned. Its unexpected large band gap predicts an outstanding chemical stability and robustness of its inherent properties, overcoming in this way one of the main difficulties that highly conjugated organic materials (such as graphene) present for their applicability under realistic conditions.

Bibliography (1) Gomberg, M. J. Am. Chem. Soc. 1900, 22, 757–771. (2) Gomberg, M. J. Am. Chem. Soc. 1903, 25, 1274–1277. (3) Gabellieri, C.; Mugnaini, V.; Paniagua, J. C.; Roques, N.; Oliveros, M.; Feliz, M.; Veciana, J.; Pons, M. Angew. Chem. Int. Ed. Engl. 2010, 49, 3360–3362. (4) Veciana, J.; Ratera, I. In Stable Radicals: Fundamentals and Applied Aspects of Odd-Electron Compounds (ed R. G. Hicks); Hicks, R. G., Ed.; John Wiley & Sons, Inc., 2010; pp. 33–80. (5) Shishlov, N. M. Russ. Chem. Rev. 2006, 75, 863–884.

(6) Mas-Torrent, M.; Rovira, C.; Veciana, J. Adv. Mater. 2013, 25, 462–468. (7) Simão, C.; Mas-Torrent, M.; Casado-Montenegro, J.; Otón, F.; Veciana, J.; Rovira, C. J. Am. Chem. Soc. 2011, 133, 13256–13259. (8) Crivillers, N.; Mas-Torrent, M.; Rovira, C.; Veciana, J. J. Mater. Chem. 2012, 22, 13883. (9) Frisenda, R.; Gaudenzi, R.; Franco, C.; Mas-Torrent, M.; Rovira, C.; Veciana, J.; Alcon, I.; Bromley, S. T.; Burzurí, E.; van der Zant, H. S. J. Nano Lett. 2015. (10) Rocha, A. R.; García-Suárez, V. M.; Bailey, S. W.; Lambert, C. J.; Ferrer, J.; Sanvito, S. Nat. Mater. 2005, 4, 335–339. (11) Sanvito, S. Chem. Soc. Rev. 2011, 40, 3336–3355.

P3 Molecular Dynamic Simulations of Gadolinium-doped Ceria and Yttria-Stabilized Zirconia interfaces for electrolyte applications Xavier Aparicio Anglès,1 and Nora H. de Leeuw1,2 1 Department of Chemistry, University College London, London, WC1H 0AJ, United Kingdom. 2 School of Chemistry, Cardiff University, Cardiff, CF10 3AT, United Kingdom. Gadolinium-doped ceria (GDC) and yttria-stabilised zirconia (YSZ) are well-known solid oxides materials used in solid-oxide fuel cells (SOFC). They can be used as electrolyte or cathode materials, being the former the one of our interest. They can be found in the SOFC device, for example either as GDC/YSZ bi-layered electrolyte,1 where their role is directly related to the oxygen transport along the cathode-electrolyte-anode system. This oxygen transport undergoes via migration of oxygens through oxygen vacancies, generated by doping ceria and zirconia with gadolinia and yttria respectively. The mobility of oxygens across the materials is influenced by many different factors, like the dopant concentration, partial oxygen pressure, or vacancy-dopant interactions among others. Interestingly, they have been well studied in bulk materials, but real systems are actually polycrystalline. In this context, the presence of grain boundaries can definitely influence the chemistry of the materials and therefore affect the oxygen mobility.2 Due to the high complexity of grain boundaries, little is known about them from a theoretical point of view. What is clear is that dopants tend to segregate towards them and then blocking the oxygen conduction.3 In the present study, our aim is to model realistic grain boundaries between gadolinium-doped ceria and yttria-stabilised zirconia using molecular dynamic simulations and interatomic potentials. Based on the amorphisation & recrystallization strategy (A&R) proposed by Sayle et al.,4 we have been able to generate realistic models that we used to explore and analyse the oxygen conductivity in GDC-YSZ interfaces, with high concerns in the structure, diffusion, and activation energies of oxygen migration in each parent material and at the interface.

Figure. Example of one model of grain boundary (left) and a slice of bulk (right where a dislocation can be observed. Colour legend: red for O, green for Ce, pink for Gd, grey for Zr, and blue for Y. References (1) Timurkutluk, B.; Timurkutluk, C.; Mat, M. D.; Kaplan, Y. J. Power Sources 2011, 196, 9361. (2) Aidhy, D. S.; Zhang, Y.; Weber, W. J. J. Mater. Chem. A 2014, 2, 1704. (3) Dholabhai, P. P.; Aguiar, J. A.; Wu, L.; Holesinger, T. G.; Aoki, T.; Castro, R. H. R.; Uberuaga, B. P. Phys. Chem. Chem. Phys. 2015, 17, 15375. (4) Sayle, T. X. T.; Parker, S. C.; Sayle, D. C. J. Mater. Chem. 2006, 16, 1067.

P4 Computational evaluation of different materials for CO2 capture Daniel Bahamon1 and Lourdes F. Vega 1,2* 1MATGAS Research Center, Campus UAB, Bellaterra, 08193, Spain 2*Alya Technology & Innovation, campus UAB, Bellaterra, 08193, Spain *lvega@alyatech.com

Although separation with amines is the usual process used at industrial scale for post-combustion CO2 capture, it carries an energy penalty of about 30% [1] which reinforces the search for more efficient processes both in economic and environmental terms. In this sense, alternative Swing Adsorption Processes (i.e. PSA, TSA) advance in order to achieve a more dynamic production, however, there is still work needed to find the most suitable material [2]. A variety of promising sorbents such as activated carbonaceous materials, silicas or zeolites have been studied by different authors due to their specific properties [3]. Moreover, Metal-Organic Frameworks (MOFs) have attracted significant attention in recent years due to their versatile crystalline structures to enable a “materials by design” approach [4]. Regarding the characterization, Grand Canonical Monte Carlo (GCMC) simulations can be used as a screening method for adsorption properties. Its value in this field has been previously proven through the interest for storage of H2, CH4 and CO2 [5], however the investigation of the effects from coexisting components such as water is less explored [6]. Here, we present a systematic computational comparison of several representative types of MOFs vs. zeolites and materials tipically used for gas storage and CO2 purification. The main objective was to find the most suitable adsorbent for a typical TSA process. A first screen was done based on adsorption isotherms, Henry's constants and isosteric heats, calculated from GCMC simulations. Predicted selectivities were evaluated and complemented with breakthrough curves among the different frameworks, with and without water traces. Results indicate that Mg-MOF-74 shows a great potential to become a material for this type of separation, with even better performance than the most traditionally used zeolite 13X [7]. This is a step forward in moving Mg-MOF-74 to industrial applications. References [1] Sumida, K.; Rogow, D. L.; Mason, J. A.; McDonald, T. M.; Bloch, E. D.; Herm, Z. R.; Bae, T.-H.; Long, J. R. Chem. Rev. 2012, 112 (2), 724-781. [2] Hedin, N.; Andersson, L.; Bergström, L.; Yan, J. Appl. Energ. 2013, 104, 418-433. [3] Samanta, A.; Zhao, A.; Shimizu, G. K. H.; Sarkar, P.; Gupta, R. Ind. Eng. Chem. Res. 2012, 51. [4] O'Keeffe, M.; Yaghi, O. M. Chem. Rev. 2012, 112 (2), 675-702. [5] Duren, T.; Bae, Y.-S..; Snurr, R. Q. Chem. Soc. Rev., 2009, 38, 1237-1247. [6] Yu, J.; Ma, Y.; Balbuena. P. B. Langmuir. 2012, 28, 8064. [7] Bahamon, D.; Vega, L.F., Chemical Eng. J. 2015, in press (10.1016/j.cej.2015.08.098) Partial support for this work has been provided by the Generalitat of Catalonia (project 2014SGR-1582) and by Carburos Metálicos/Air Products Group.

P5 In-vivo-like study of the excluded volume effects on the kinetics of enzymatic reactions

Cristina Balcells1, Claudia Hernández1, Mireia Via1, Isabel Pastor2, Josep Lluís Garcés3, Sergio Madurga1, Eudald Vilaseca1, Marta Cascante4 and Francesc Mas1

1 Department of Physical Chemistry and Research Institute of Theoretical and Computational Chemistry (IQTCUB) of Barcelona University, Barcelona (Spain); crisgatsu@gmail.com

2 Small Biosystems Lab, Department of Fundamental Physics, University of Barcelona, Barcelona (Spain) and CIBER-BBN, Carlos III Health Institute, Madrid (Spain);

3 Department of Chemistry, and AGROTECNIO, University of Lleida (UdL), Lleida (Spain); 4 Department of Biochemistry and Molecular Biology and Institute of Biomedicine (IBUB) of Barcelona University,

Barcelona (Spain)

The cell is a heterogeneously distributed and highly crowded medium in which a wide variety of physical and chemical processes take place. Until recently, each single process had been studied as an independent and isolated event, as close to ideality as possible. Still, this is quite unrealistic, both in terms of intermolecular interactions and in the fraction of occupied volume, which can be up to 300-400 g/L. Volume exclusion is one of the most relevant entropic effects occurring inside the cell, since it gives raise to steric repulsions, depletion forces and directly impacts on diffusion, interactions, kinetics and conformational equilibriums of biopolymers [1].

In the present work, the volume exclusion problem, also known as macromolecular crowding, has been applied to the field of enzyme kinetics. It has been approached by adding neutral, relatively inert polymers, which act as crowding agents or obstacles, in the media of given enzymatic reactions. The concentration and size of these obstacles have been changed systematically while studying the kinetic behavior of four differently-sized enzymes: α-Chymotrypsin (α-Chy, 25 kDa) [2], Horseradish Peroxidase (HRP, 42 kDa) [3], Alkaline Phosphatase (ALKP, 104 kDa) [4] and Lactate Dehydrogenase (LDH, 140 kDa) [5].

Results, in both experiment [6] and simulation [7], indicate that the performance of a certain enzyme depends on the amount of excluded volume, regardless of the enzymatic system. However, only large, oligomeric proteins display an obstacle size-dependent behavior. In this regard, the enzyme-crowding agent ratio can have a significant impact on the kinetics of a given reaction. Besides, it has been shown that such crowding can hinder diffusion to the extent of being capable of altering reaction control from activation to diffusion.

References:

[1] HX. Zhou, G. Rivas, A.P. Minton, Annu. Rev. Biophys. 2008, 37, 375-397; I. M. Kuznetsova, K.K. Turoverov and V.N. Uversky, Int. J. Mol. Sci. 2014, 15, 23090–23140.

[2] I. Pastor, E. Vilaseca, S. Madurga, M. Cascante, F. Mas J. Phys. Chem. B. 2011 115(5), 1115–21. [3] L. Pitulice, I. Pastor, E. Vilaseca, S. Madurga, A. Isvoran, M. Cascante, F. Mas., J. Biocatal. Biotransform., 2013, 2, 1-5. [4] C. Balcells., C. Hernández, M. Via, I. Pastor, J.L. Garcés, S. Madurga, M. Cascante, F. Mas, in preparation 2015. [5] C. Balcells., I. Pastor, E. Vilaseca, S. Madurga, M. Cascante, F. Mas J. Phys. Chem. B. 2014, 118 , 4062-4068. [6] I. Pastor, L. Pitulice, C. Balcells, E. Vilaseca, S. Madurga, A. Isvoran, F. Mas, Biophys. Chem., 2014, 185, 8–13. [7] L. Pitulice, E. Vilaseca, I. Pastor, S. Madurga, J.L. Garcés, A. Isvoran, F. Mas, Math. Bios., 2014, 251, 72-82.

P6 Brownian dynamics simulation of reaction-diffusion processes of proteins in intracellular environaments Pablo M. Blanco1, Mireia Via1, Sergio Madurga1, Josep Lluís Garcés2, Eudald Vilaseca1 and Francesc Mas1 1Department of Physical Chemistry and Research Institute of Theoretical and Computational Chemistry (IQTCUB), University of Barcelona (UB), Spain 2Department of Chemistry and AGROTECNIO, University of Lleida (UdL), Spain The cellular cytosol is a very dense medium, with huge concentrations of biological macromolecules that, by means of non-specific interactions, has a considerable effect in processes like diffusion and reactivity. Therefore, theoretical models that describe these processes in homogeneous media are no longer valid in crowded conditions. Due to that, the mechanisms of the diffusion and reaction processes should be studied in these conditions in order to obtain more reliable and realistic results. In that way, new theoretical models that consider the presence of macromolecules can be proposed. Computer simulations are a powerful tool to study reaction and diffusion processes. In this work, enzymatic reactions are simulated by means of the ReaDDy software package [1]. The stochastic movement of particles is generated by a Brownian motion algorithm, whereas the reactivity is studied with a Monte Carlo algorithm [2],[3]. The space-time evolution analysis of the system allows the calculation of the diffusion coefficient of the enzyme and the study of the Michaelis-Menten behavior of its reaction in different kinetic and excluded volume conditions.

Graphic example of a simulation of an enzymatic reaction with ReaDDy with obstacles (green), enzymes (red), complexes (orange), substrates (yellow) and products (black).

[1] Schöenberg, J.; Noé, F. PLoS ONE. 2013, 8, e74261. [2] Vilaseca, E., Isvoran, A., Madurga, S., Pastor, I., Garcés, J.L., Mas, F. Phys Chem Chem Phys, 2011, 13, 7396-7407. [3] Pitulice, L., Vilaseca, E., Pastor, I., Madurga, S., Garcés, J.L., Isvoran, A., Mas, F. Mathematical Biosciences, 2014, 251, 72-82.

P7 Molecular determinants for selective C25-hydroxylation of vitamins D2 and D3 by fungal peroxygenases Marina Cañellas,1,a,c Fátima Lucas,1,a Esteban D. Babot,1,b José C. del Río,b Lisbeth Kalum,d René Ullrich,e Martin Hofrichter,e Victor Guallar,a,f Angel T. Martínezg and Ana Gutiérrezb 1 These three authors equally contributed to this work a Joint BSC-CRG-IRB Research Program in Computational Biology, Barcelona Supercomputing Center, Jordi Girona 29, E-08034 Barcelona, Spain. E-mail: fati.lucas@gmail.com b Instituto de Recursos Naturales y Agrobiología de Sevilla, CSIC, Reina Mercedes 10, E-41012 Seville, Spain. E-mail: anagu@irnase.csic.es; Fax: +32 954624002 ; Tel: +32 954624711 c Anaxomics Biotech, Balmes 89, E-08008 Barcelona, Spain d Novozymes A/S, Krogshoejvej 36, 2880 Bagsvaerd, Denmark e TU Dresden, Department of Bio- and Environmental Sciences, Markt 23, 02763 Zittau, Germany f ICREA, Passeig Lluís Companys 23, E-08010 Barcelona, Spain g Centro de Investigaciones Biológicas, CSIC, Ramiro de Maeztu 9, E-28040 Madrid, Spain Selective oxygenations of aliphatic compounds are among the most challenging reactions in organic chemistry for the regio and/or stereo specific synthesis of pharmaceuticals and fine chemicals. Recently, a new peroxidase type, which shares the active-site architecture and reaction mechanism of cytochromes P450, but has the advantage of being activated directly by H2O2, was isolated from Agrocybe aegerita (Ullrich, 2004), and later identified in a variety of sequenced basidiomycete genomes, including that of Coprinopsis cinerea (Floudas, 2012). The A. aegerita heme-thiolate (HTP) enzyme has been the most widely investigated basidiomycete peroxygenase, and recent studies have shown that the C. cinerea enzyme has comparative advantages related to its high conversion yield/selectivity for some hydroxylation reactions, and its production as a recombinant protein in an industrial expression host (Babot, 2013; 2014). 25-monohydroxylated vitamin D3 (cholecalciferol) and D2 (ergocalciferol), compounds of high interest in human health and animal feeding, can be obtained through reaction with both A. aegerita and C. cinerea enzymes. In the present study, PELE computational analyses are used to rationalize the underlying mechanisms responsible for differences in reactivity and selectivity observed experimentally between both enzymes (Lucas, 2015). References Babot, E. D., et al., Biotechnol. Bioeng., (2013), 110, 2332. Babot, E. D., et al., Chemcatchem, (2014), online. Floudas, D., et al. Science, (2012), 336, 1715-1719. Lucas, F., et al. Catal. Sci. Tech., (2015), online. Ullrich, R., et al., Appl. Environ. Microbiol. (2004), 70, 4575-4581.

P8 Exploring the protonation properties of photosynthetic phycobiliprotein pigments from molecular modeling and spectral line shapes M. Corbellaa, Z. S. D. Toab, G. D. Scholesb, F. J. Luquea, C. Curutcheta

aDepartament de Fisicoquímica, Facultat de Farmàcia, Universitat de Barcelona, Av. Joan XXIII s/n, 08028 Barcelona, Spain bDepartment of Chemistry, Princeton University, Washington Road, Princeton, New Jersey 08544, United States e-mail: marinacorbella@ub.edu In photosynthesis, specialized light harvesting pigment-protein complexes (PPCs) are used to capture incident sunlight and funnel its energy to the reaction center. The PPCs of cryptophyte algae use tunable linear tetrapyrrole chromophores (bilins) covalently bounded to the protein scaffold, which structure and disposition inside the protein have evolved to increase the spatial and spectral cross section for absorption of incident light.1 These proteins are suspended in the lumen, where the pH ranges between ~5-7, depending on the prolongation of the incident sunlight. Many theoretical and experimental studies have been done in order to uncover the basic mechanisms that drive electronic energy transfers in these organisms.2–5 However, the pKa of the several kinds of bilin chromophores encountered in these complexes and the effect of its protonation state on the energy transfer process is still unknown. Here, we combine quantum chemical and continuum solvent calculations to estimate the intrinsic aqueous pKas of different bilin pigments: phycocyanobilin (PCB), phycoerythrobilin (PEB), 15,16-dihydrobiliverdin (DBV) and mesobiliverdin (MBV). We then use APBS classical electrostatic calculations to estimate the change in protonation free energies when the bilins are embedded inside five different phycobiliproteins (PE545, PC577, PC612, PC630 and PC645), and critically assess our results by analysis of the changes in the absorption spectral line shapes measured within a pH range from 4.0 to 9.4. Our results suggest that each individual protein environment strongly impacts the intrinsic pKa of the different chomophores, being the final responsible of their protonation state. Our study paves the way for accurate structure-based studies on the light harvesting properties of cryptophyte antenna complexes. (1) Harrop, S. J.; Wilk, K. E.; Dinshaw, R.; Collini, E.; Mirkovic, T.; Teng, C. Y.; Oblinsky, D. G.; Green, B. R.; Hoef-Emden, K.; Hiller, R. G.; Scholes, G. D.; Curmi, P. M. G. Proc. Natl. Acad. Sci. U. S. A. 2014, 111 (26), E2666. (2) Curutchet, C.; Kongsted, J.; Muñoz-Losa, A.; Hossein-Nejad, H.; Scholes, G. D.; Mennucci, B. J. Am. Chem. Soc. 2011, 133 (9), 3078. (3) Curutchet, C.; Novoderezhkin, V. I.; Kongsted, J.; Muñoz-Losa, A.; van Grondelle, R.; Scholes, G. D.; Mennucci, B. J. Phys. Chem. B 2013, 117 (16), 4263. (4) Viani, L.; Curutchet, C.; Mennucci, B. J. Phys. Chem. Lett. 2013, 4 (3), 372. (5) Viani, L.; Corbella, M.; Curutchet, C.; O’Reilly, E. J.; Olaya-Castro, A.; Mennucci, B. Phys. Chem. Chem. Phys. 2014, 16 (30), 16302.

P9 Molecular basis of the exciton-phonon interactions in the PE545 and WSCP pigment-protein complexes C. Curutchet,a A. Rosnik,a L. Viani,b M. Corbella,a E. O’Reilly,c A. Olaya-Castroc and B. Mennuccib aDepartament de Fisicoquímica, Facultat de Farmàcia, Universitat de Barcelona, Spain bDipartimento di Chimica e Chimica Industriale, Università di Pisa, Italy cDepartment of Physics and Astronomy, University College London, United Kingdom e-mail: carles.curutchet@ub.edu

The observation of long-lived quantum coherence effects in several pigment-protein complexes has spurred a heated debate on their actual role in the photosynthetic light-harvesting process.[1] Besides their impact, for example, in the overall efficiency of the process, the more fundamental question addressing their molecular basis in a biological disordered environment remains unclear, thus precluding the design of novel light-harvesting devices incorporating such potentially interesting property. The role of the environment in sustaining quantum coherence has been convincingly recognized. However, the actual mechanism remains controversial, and several hypotheses have been suggested, such as the presence of protein-induced correlated fluctuations or the role of the structured nature of the spectral density of pigment-protein coupling.

Fig. 1. Spectral densities simulated for the bilin pigments in the PE545 antenna from cryptophytes. Here, we present a detailed theoretical investigation of the spectral density of exciton-phonon interactions in the phycoerythrin 545 (PE545) antenna of cryptophyte algae[1-3] and the water-soluble chlorophyll-binding protein (WSCP) of Virginia pepperweed.[4-5] We use a strategy we have recently developed with combines classical MD simulations with QM/MM calculations of the relevant excited-states of the complex, where explicit environment polarization effects are fully taken into account in atomic detail.[2] We find that low-frequency features in the spectral densities of the PE545 bilin pigments, expected to play an important role in sustaining quantum coherence, are strongly modulated by the local protein environment, features that are largely smoothed out when the spectral density is averaged over the complex.[3] This underscores the limits of adopting a common spectral density for different pigments, a common assumption in theoretical studies of photosynthetic complexes. Moreover, we find that the description of high-frequency intramolecular features in the spectral density of WSCP is significantly improved when the analysis is performed over QM/MM MD trajectories compared to classical simulations.[4] References [1] a) Engel et al. Nature 2007, 446, 782; b) Collini et al. Nature 2010, 463, 644. [2] a) Curutchet et al. J. Am. Chem. Soc. 2011, 133, 3078; b) Curutchet et al. J. Phys. Chem. B 2013, 117, 4263. [3] a) Viani et al. J. Phys. Chem. Lett. 2013, 4, 372; b) Viani et al. Phys. Chem. Chem. Phys. 2014, 16, 16302. [4] Pieper et al. J. Phys. Chem. B 2011, 115, 4042. [5] Rosnik & Curutchet, in preparation.

P10 Atomically dispersed M species (M = Pd, Ni, Cu) in ceria nanoparticles: Stability and red-ox processes

Alberto Figueroba,1 Konstantin M. Neyman1,2

1 Departament de Química Física & Institut de Química Teòrica i Computacional, Universitat de Barcelona, Barcelona, Spain 2 Institució Catalana de Recerca I Estudis Avançats (ICREA), Barcelona, Spain

afigue89@gmail.com

We studied various properties of cationic M species (M = Pd, Ni, Cu) in nanostructured ceria by means of density functional calculations. Following recent work on M = Pt [1], a cuboctahedral Ce40O80 nanoparticle is used to investigate the origin and stability of M species. The effect of the reducing conditions on the fuel cell canoed catalyst during operation is also investigated. For that we explored the formation of oxygen vacancies in the nanoparticle and the adsorption of hydrogen atoms on the M-Ce40O80 model.

We found that all deposited M atoms prefer to be anchored featuring the 2+ state, with surface substitutions being the most favoured ones. Upon deposition, two Ce4+ cations are reduced to Ce3+. The two reduction mechanisms considered lead to structures with the M cation in the centre of a square formed by four O2- anions.

References:

1. A. Bruix, V. Matolín, J. Libuda, K. M. Neyman et al. Angew. Chem. Int. Ed. 53 (2014) 10525

P11 A dynamical model for 13C fluxomics Carles Foguet1, Silvia Marin1, Vitaly Selivanov1, Eric Fanchon2, Pedro De Atauri1 and Marta Cascante1 (1)Department of Biochemistry and Molecular Biology, Universitat de Barcelona, Barcelona, Spain (2)TIMC-IMAG Laboratory, UJF-Grenoble 1 - CNRS, Grenoble, France Fluxomics is the field of omics that studies metabolic fluxes, that is to say, the rate at which enzymes and transporters catalyse and facilitate chemical reactions and transport processes in living beings. Fluxomics has a key role in our understanding of any metabolic systems since the fluxome, the set of metabolic fluxes, is a direct manifestation of the metabolic phenotype of the system. However, the fluxome cannot be directly quantified through experimental measurements, instead it must be estimated through the integration of experimental data in complex mathematical models referred as metabolic models. 13C fluxomics is a field of fluxomics that uses data from 13C experiments. In such experiments, cells are incubated with substrates labelled with 13C and afterwards the enrichment of 13C is quantified in metabolic products. Such measurements, when integrated into a mathematical model can be used to accurately predict the fluxome in the experimental conditions. Our group has developed a metabolic model of central metabolism which simulates with great level of detail glycolysis, gluconeogenesis, glycogen metabolism, the pentose phosphate pathway, the Krebs cycle, fatty acid metabolism and energy metabolism. Additionally a key feature of the model is that is capable of performing dynamical simulations allowing to predict the evolution of the metabolic system over time. This is possible because the kinetic properties of each individual enzymes included in the model have been modelled. Furthermore the model is capable of integrating data from 13C experiments allowing it to perform 13C fluxomics. Our model has countless potential application including identifying vulnerabilities in the metabolic reprogramming associated with neoplastic progression, which can be then exploited on therapy, or assessing possible side effects of drugs on metabolic function.

P12 Predictive DFT-based QSAR model for the nucleophilic reactivity of trivalent boron compounds Diego García-López1*, Jessica Cid1, Elena Fernández1, Jorge J. Carbó1. 1Department de Química Física i Inorgànica, Universitat Rovira I Virgili (URV), Tarragona, Spain diego.garcia@urv.cat Most trivalent boron compounds are electrophiles due to the lack of two electrons to fill the outer orbital of the boron atom. However, these compounds can change their reactivity from an electrophilic character to a nucleophilic behaviour by changing the nature of the substituents on the boron atom [1]. This tuneable reactivity is specially remarked when the boron atom is bonded to different transition metals and shows potential application in organic synthesis. Our group had defined computational descriptors to build a tendency map that classifies the boryl fragments according to their nucleophilic character [2]. In a step forward, we pursue to set a quantitative relationship between the steric and electronic properties of boryl compounds and their nucleophilic reactivity. The study covers a full range of boryl fragments that are bonded to main-group metals and coordinated to transition-metals. To define a quantitative dependent variable that measures the nucleophilic activity, we made use of the boryl addition to formaldehyde and computed its free energy barrier. Using multivariate regression techniques, we were able to generate mathematical models that correlate well with computed reactivity and that show predictive ability.

1. [1] Cid J, Gulyás H, Carbó J J, Fernández E: Trivalent boron nucleophile as a new tool in organic synthesis: reactivity and asymmetric induction. Chem Soc Rev 2012, 41:3558-3570. 2. [2] Cid J, Carbó J J, Fernández E: Trivalent Disclosing the structure/activity correlation

in trivalent boron-containing compounds: a tendency map. Chem Eur J: 2012, 40:12794-12802.

P13 IMPACT OF VAN DER WAALS INTERACTIONS ON THE WATER gas shift reaction over copper surfaces

Hèctor Pratsa,* Leny Álvareza, Pablo Gamalloa, Francesc Illasa and Ramón Sayósa a: Departament de Química Física and Institut de Química Teòrica I Computacional, Universitat de Barcelona, C. Martí I Franquès 1, 08028, Barcelona, Spain e-mail: hpratsga@ub.edu The effect of van der Waals (also known as dispersion) interactions on adsorption properties has been the focus of an intense research in the past few years, especially after the landmark contributions of Grimme and co-workers [1] which has triggered many new theoretical developments and the appearance of a plethora of new functionals aiming to account for these terms in an accurate and non-empirical way. In spite of the large number of articles devoted to study the importance of dispersion terms in adsorbate-surface interactions, there is almost no information regarding the effect of dispersion terms in the energy profile of heterogeneously catalysed reactions, especially for complex mechanisms involving several elementary steps. An important catalysed reaction with special technological relevance is the WGSR transforming CO and H2O into CO2 and H2 [2]. This process takes place in two stages, high and low temperature, respectively, and copper-based catalysts are very used in the second stage. There have been studies in depth for the Cu(111) [3] and Cu(321) [4] surfaces, the latter one containing different low coordinated sites offers a more realistic model of the catalyst. Moreover in the latter case, there is detailed information regarding the structure of the many transition state structures involved in the mechanism and thus constitutes and excellent system to test the impact of the dispersion terms on the overall energy profile and rate constants. This is precisely the goal of the present work. We will provide compelling evidence that while the qualitative picture of the overall reaction scheme is not largely affected by the inclusion of the dispersion terms, there are significant differences in the calculated rate constants, which has important implications in the macroscopic description of the overall process via microkinetic or kinetic Monte Carlo simulations. 1) S. Grimme, J. Comput. Chem. 2004, 25, 1463 2) D.S. Newsome, Catal. Rev. Sci. Eng. 1980, 21, 275 3) A.A. Gokhale, J.A. Dumesic, M. Mavrikakis. J. Am. Chem. Soc. 2008, 130, 1402. 4) J.L.C. Fajin, M.N.D.S Cordeiro, F. Illas, J.R.B. Gomes, J. Catal. 2009, 268, 131

P14 The importance of framework cations in the selective adsorption of CO2 compared to O2 and N2 in Faujasite structures. Gerard Alonso, Ramón Sayós, Xavier Giménez, Pablo Gamallo Departament de Química Física and Institut de Química Teórica i Computacional, Universitat de Barcelona, C. Martí i Franquès 1, 08028, Barcelona, Spain e-mail: g.alonso@ub.edu Zeolites are adsorbent aluminosilicate materials with many applications. Depending on their structure they can serve as water adsorbents, reversible cation exchangers, hydrocarbon separators, gas separators or catalysts. There are three main advantages that make zeolites attractive for industrial applications: They are solid structures that can be easily regenerated in discontinuous processes; the zeolite engineering allows tuning almost completely the structure, and thus the properties of the zeolite; and finally their low cost. Regarding the zeolite engineering, the main factor that controls the activity of the zeolite is the so-called Si/Al ratio. Analyzing the structure, each unit of [AlO2

-] is paired with an alkali or alkali earth cation, which serves as a new site of adsorption[1]. The family of Faujasites contain zeolites widely used in the study of gas separation due to its capacity of adsorbing selectively some gases into their highly porous structure[2]. In order to study the optimum structure of Faujasite to capture CO2, in front of the most common gases present in dry air (O2 and N2), periodic dispersion corrected Density Functional Theory calculations (DFT-D2) have been performed, at the Perdew, Burke and Ernzerhof (PBE) level of theory. To evaluate the effect of the pairing cations many structures with different Na+ amounts have been studied, from the fully silicated Faujasite, until a Faujasite with Si/Al ratio = 3.34 The main sites of adsorption are analyzed through DFT calculations. Our results reveal that the interactions between Faujasite and the three gases are weak. However, the presence of Na+ cations enhances the adsorption properties of Faujasite, especially towards CO2, improving the selective separation of this specie. The diffusion of the three gases within Faujasite is also analyzed denoting the importance of Na+ cations in this property. Additionally, the interpretation of the physics of adsorption in Faujasite understood by means of DFT is compared with the interpretation led by popular Grand Canonical Monte Carlo results[3-4]. 1) Ch. Baerlocher, L.B. McCusker, D.H. Olson. Atlas of Zeolite Framework Types (6th Ed). 2007, Elsevier. 2) A. Ghoufi, L. Gaberova, J. Rouquerol, D. Vincent, P.L. Llewellyn, G. Maurin. Microporous Mesoporous Mater. 2009, 119, 117 3) A. Mateo. Unpublished Degree Theses. University of Barcelona, 2014. 4) G. Maurin, P.L. Llewellyn, R.G. Bell. J. Phys. Chem. B, 2005, 109, 16084.

P15 THE REGIOSELECTIVITY OF ARACHIDONIC ACID PEROXIDATION MODIFIED BY in silico MUTATIONS OF RABBIT 15-LIPOXYGENASE P. Saura,a,b L. Masgrau,b J. M. Lluch,a,b and À. González-Lafont*a,b

aDepartament de Química and bInstitut de Biotecnologia i de Biomedicina (IBB), Universitat Autònoma de Barcelona, 08193 Bellaterra, Barcelona, Spain. E-mail: Angels.Gonzalez@uab.cat Phone: (34)935811672 The regio- and stereospecific catalysis of lipoxygenases (LOXs) is a key feature to understand how each LOX isoform manages to generate a different lipid mediator, with quite different (even opposite) physiological roles, from a common substrate, arachidonic acid (AA). The proximity of one particular AA methylene to the Fe(III)-OH- cofactor has been used as an structural feature that determines the regioselectivity of the rate-determining H-abstraction step. In this study we explore the molecular basis of rabbit 15-LOX-1 reaction specificity by means of two different in silico mutations that correspond to: i) modification of residues at the bottom of the active site cavity (Ile418) in accordance with mutagenesis experiments and, ii) modification of residues which hinder the evolution of AA during the reactive process (Leu597 and Ile663). Molecular dynamics simulations and quantum mechanics/molecular mechanics calculations have been carried out on the Ile418Ala and Leu597Ala/Ile663Ala mutants of rabbit 15-LOX-1:AA complex. The lower average energy barriers for H10-abstraction versus H13-abstraction in both mutants confirm the change of regioselectivity from a 15-lipoxygenating wild type enzyme (H13-abstraction favored) into a 12-lipoxygenating mutant enzyme (H10-abstraction favored). However, the analysis of the generated productive mutant structures shows that the H10/H13-OH distances are neither the unique nor the most relevant factor for explaining the molecular basis of 15-LOX-1 regioselectivity. So, our in silico calculations provide a new mechanistic basis for the experimental “triad concept”. In the wild type enzyme,[1] which is predominantly 15-lipoxygenating, H10 abstractions, but not the H13 ones, are sterically hindered by the bulky conserved residues Leu597 and C-terminal Ile663. However, in its Ile418Ala mutant arachidonic acid slides in deeper into the binding pocket, in such a way that C10 moves to the region occupied by C13 in wild type, and H10-abstraction can occur without steric disturbance, the mutant becoming predominantly 12-lipoxygenating in agreement with experiment. When those two bulky residues are mutated in silico to smaller ones in the Leu597Ala/Ile663Ala mutant, H10-abstraction is not impeded anymore and regioselectivity is also clearly inverted. [1] P. Saura, R. Suardíaz, L. Masgrau, J.M. Lluch, À. González-Lafont. ACS Catalysis 2014, 4, 4351-4363.

P16 Towards multiscale investigation of base-pair level properties of chromatin Jürgen Walther1,2, Pablo D. Dans1,2, Modesto Orozco1,2,3 1 Institute for Research in Biomedicine (IRB Barcelona), Baldiri Reixac 10-12, 08028 Barcelona, Spain. 2 Joint BSC-IRB Research Program in Computational Biology, Baldiri Reixac 10-12, 08028 Barcelona, Spain. 3 Department of Biochemistry and Molecular Biology, University of Barcelona, 08028 Barcelona, Spain Multiple features of chromatin are expected to strongly depend on sequence specific properties of nucleosomal and linker DNA. However, detailed analysis of chromatin characteristics with its dependence on the DNA sequence still constitutes an open question. We developed an efficient coarse-grained Monte Carlo (MC) algorithm to investigate sequence dependent DNA properties. In our model, DNA is represented intrinsically at base pair level with an elastic potential representing the interactions between adjacent base pairs. Coupling terms between base pairs are extracted from high detail all atom molecular dynamics (MD) simulations with the improved parmbsc1 force field. A comparison of DNA structures of the same sequence generated by MC and by all atom MD reveals striking similarity of important ensemble features such as distribution of helical parameters, bending, persistence length and essential dynamics. We further developed a method to locate the phosphate in the backbone and estimate its dynamics just based on the set of intrinsic base pair parameters. With this procedure the major groove width can be determined at an accuracy of less than 1Å. Summarizing, the coarse-grained MC model (simulation times ~10⁵ times faster than conventional all atom MD simulations) can be used to accurately describe the dynamics of DNA. This result makes it possible to study the characteristics of sequence specific chromatin fibers at base pair level accuracy by extending the model beyond the scale of DNA.

P17 Accurate determination of chemical ordering in several nm large bimetallic nanoparticles Gábor Kovács,a Sergey M. Kozlov,a Riccardo Ferrando,b Konstantin M. Neymana,c a Departament de Química Física and Institut de Química Teòrica i Computacional (IQTCUB), Universitat de Barcelona, c/ Martí i Franquès 1, 08028 Barcelona, Spain b Dipartimento di Fisica and CNR-IMEM, Via Dodecaneso 33, 16146 Genova, Italy c Institució Catalana de Recerca i Estudis Avançats (ICREA), 08010 Barcelona, Spain The search for the lowest-energy chemical ordering, that is, the most stable arrangement of atoms of different elements within a given nanoparticle structure is crucial in the assessment of the behavior of bimetallic nanoalloys, used widely in heterogenous catalysis amongst other applications. Chemical ordering governs the most important NP properties, such as surface composition and electronic structure, which are crucial for surface reactivity in heterogeneous catalysis. Herein, we present results obtained by the application of a method (called TOP method in the following) newly developed in our group for the optimization of chemical ordering in bimetallic nanoalloys. TOP means the method is based on topological expressions, whose coefficients (called descriptors) are fitted to DFT-energies calculated for 20-30 selected structures, obtained previously via MC simulation. The TOP method is shown to be a powerful new methodology for the optimization of chemical ordering in NPs of different metals, shapes, sizes and composition. The method was shown to enable the optimization of the chemical ordering in nanoparticles of a few thousand atoms via descriptors obtained from DFT calculations carried out on smaller NPs.

Figure 1. The structure of Pt1097Co366 with optimized chemical ordering.

P18 Effect of Charge Transfer between CeO2(111) Support and Pt Nanoparticles on Their Properties Sergey M. Kozlov,1 Konstantin M. Neyman1,2 1 Departament de Química Física and Institut de Química Teòrica i Computacional (IQTCUB), Universitat de Barcelona, c/ Martí i Franquès 1, 08028 Barcelona, Spain 2 Institució Catalana de Recerca i Estudis Avançats (ICREA), 08010 Barcelona, Spain Many industrially used heterogeneous catalysts consist of transition metal nanoparticles on oxide supports. The ability of oxide support to dramatically affect the activity of such catalysts is known from numerous experiments. In particular, reducible oxides like CeO2 are usually referred to as active supports, which are likely to modify the course of reaction. Although the mechanism causing these changes is not yet well understood, the charge transfer between metal nanoparticles and oxide supports is thought to be very important for catalytic activity. In this work we investigate how the interaction and the electron transfer between Pt nanoparticles and CeO2(111) support affect properties of the former. This particular system has been chosen because of 1) the application of Pt/ceria catalysts in fuel cells, and 2) the ability to monitor and control the magnitude of metal-oxide charge transfer by the number of reduced Ce3+ cations in the system. Namely, we characterize the effect of CeO2(111) support on geometric and electronic structure of ~1.5 nm large Pt95 and Pt122 particles in the presence and in the absence of electron transfer. Previously, Pt nanoparticles of this size were shown to be representative of much larger species commonly dealt with in experiment and application.

Figure 1. Structures of investigated Pt95 and Pt122 nanoparticles interacting with CeO2(111) support through {100} and {111} facets, respectively. (Edge) Pt atoms are displayed in (dark) blue, (surface) O atoms – (dark) red, Ce4+ cations – beige, Ce3+ cations – cyan.

P19 Surface Contact Engineering in Photoactive ZnO Nanostructures Oriol Lamiel-García,1 Francesc Viñes,1 Ana Iglesias-Juez,2 Marcos Fernández-García,2 and Francesc Illas,1*

1. Departament de Química Física & Institut de Quimica Teòrica i Computacional (IQTCUB), Universitat de Barcelona, c/Martí i Franquès 1, 08028 Barcelona, Spain.

2. Instituto de Catálisis y Petroleoquímica, CSIC, c/Marie Curie 2, Cantoblanco, 28049 Madrid, Spain.

*corresponding author: francesc.illas@ub.edu A series of ZnO nanostructures with variable morphology were prepared by a microemulsion method and their structural, morphological, and electronic properties were investigated by a combined experimental and theoretical approach using microscopy (high resolution transmission electron microscopy) and spectroscopic (X-ray diffraction, Raman, and UV-visible) tools, together with density functional theory calculations. The present experimental and computational study provides a detailed insight into the relationship between surface-related physicochemical properties and the photochemical response of ZnO nanostructures. Specifically, the present results provide conclusive evidence that light-triggered photochemical activity of ZnO nanostructures is intimately governed by a synergistic effect between polar and nonpolar surfaces rather than by the predominance of highly-active (polar) surfaces, bandgap sizes, carrier mobilities, and other variables usually mentioned in the literature. Computational results highlight the oxidative capability of polar surfaces, independently of the degree of hydration and the presence of surface defects.

P20 Characterization of the biodiesel production process using a generalized predictive approach with a molecular-based equation of state F. Llovell1,2, M.B. Oliveira3, J.A.P. Coutinho3, L. F. Vega1,4 1MATGAS 2000 AIE, Campus de la UAB, 08193, Bellaterra, Barcelona, Spain 2IQS School of Engineering, Universitat Ramon Llull,. Via Augusta, 390, 08017 Barcelona, Spain 3CICECO, Chemistry Department, University of Aveiro, 3810-193 Aveiro, Portugal 4Alya Technology & Innovation, Campus UAB, 08193 Bellaterra, Barcelona, Spain *felix.llovell@iqs.edu Biodiesel is a renewable, biodegradable fuel that can be manufactured from vegetable oils, animal fats or even food residues, becoming the easiest replacement for petroleum fluids. However, the characterization of the different types of biodiesel, made of mixtures of fatty acid esters in different compositions, is a requirement to approve their use for different processes. In this work, a simple but reliable theoretically-based model, the soft-SAFT EoS,1 is used as a tool for the development, design and optimization of the biodiesels production processes, through the description of their thermodynamic properties, as well as the phase equilibria of systems formed along the industrial units. A molecular model within the soft-SAFT EoS framework has been proposed for the fatty acid esters. The Density Gradient Theory, coupled with soft-SAFT, is used for the description of interfacial properties, while viscosities are calculated through the Free-Volume Theory, in an integrated model. Density, surface tension and viscosity data for fatty acid methyl and ethyl esters, ranging from C8:0 to C24:0, with up to three unsaturated bonds, are successfully reproduced for pressures up to 150 MPa, and temperatures up to 423.15 K. The high pressure densities and viscosities for 8 biodiesels are then successfully predicted without further information.2 Also, the water solubility in fatty acid esters, and the vapor-liquid equilibria of fatty acid ester + methanol/ethanol systems, formed during the process, have been studied through a new association scheme to take into account the solvation phenomenon between esters and water/alcohols. The selected model is able to describe the water solubility and the VLE of binary systems of fatty acid esters and methanol/ethanol using only one binary interaction parameter.3 Finally, the solubility of supercritical carbon dioxide in fatty acid esters is also addressed as a possible media during the production and purification steps.4

Acknowledgements: Financial support from the Spanish Government (project CTQ2014-53987-R) and the Catalan Government (project 2014SGR-1582) is acknowledged. 1. F.J. Blas, L.F Vega, Mol. Phys. 92 (1997) 135-150. 2. M.B. Oliveira, S.V.D. Freitas, F. Llovell, L.F. Vega, J.A.P. Coutinho, ChERD, 92 (2014) 2898-2911. 3. M. B. Oliveira, F. Llovell, M. Cruz, L. F. Vega, J. A. P. Coutinho. Fuel, 129, (2014) 116–128 4. F. Llovell, L.F. Vega, J. Supercritical Fluids, 96 (2015) 86-95.

P21 Incorporation of conserved water molecules in protein structures. Eduardo Mayol1, Arnau Cordomí1, Mireia Olivella2, Leonardo Pardo1. 1 Laboratori de Medicina Computacional, Unitat de Bioestadística, Facultat de Medicina, Universitat Autònoma de Barcelona, Bellaterra, Spain 2 Department de Biologia de Sistemes, Universitat de Vic, Vic, Barcelona, Spain Water molecules located in interstitial cavities of proteins often have structural and functional importance, but unfortunately, these are not always resolved in the crystal structures, especially in those with lower resolution. The situation is even worse for homology models, where there is often complete lack of information regarding the localization of such interstitial waters. We have developed a protocol to incorporate internal water molecules (and ions) in structural models of proteins relying entirely on crystal structures of other members of the same family. The protocol tries to hierarchically incorporate water molecules (and ions) from other members that do not clash with the protein atoms. This is done guided by a family phylogenetic tree and taking into account the resolution of the crystal structures and the B-factors of the water oxygens. In the present report we focused on members of the G protein-coupled receptors family as a test case, and explored the effect in molecular dynamics simulations of introducing or not internal water molecules in molecular models. The results show that incorporation of such analogous waters provides more stable systems and that this is particularly relevant for isolated water sites, where bulk water has more difficulties to access them during simulations.

P22 Small Peptides As Copper Chelators In Alzheimer’s Disease Andrea Mirats,a Luis Rodríguez-Santiagoa and Mariona Sodupea

aUniversitat Autònoma de Barcelona, Departament de Química, Edifici C Facultat de Ciències, Cerdanyola del Vallès - SPAIN e-mail: andrea.mirats.qf@gmail.com Alzheimer’s disease (AD) is the most common form of neurodegenerative dementia and one of the main hallmark is the presence of extracellular deposits formed by the aggregation of the amyloid-b peptide (Ab) and its deposition is the causative agent of AD.1 On the other hand, in vitro studies have revealed that transition metal cations such as Fe3+, Cu2+ and Zn2+ may promote Ab aggregation. Specially, the presence of redox-active Cu2+ may be associated with the increased oxidative stress present in the brains of AD patients, due to the formation of reactive oxygen species (ROS).2 Therefore, copper is a potential target to develop drugs against AD. In the last years, metal-ion chelation therapy with the so-called Metal Protein Attenuating Compounds (MPACs) has emerged as a promising treatment strategy. The main goal of the original MPAC compounds was to inhibit the interaction between Aβ and the metal. However, during the development of these chelators it has been observed that the inhibition of Aβ–metal interactions was insufficient to explain all the observed effects of the drug. Thus, recent studies have focus on the development of new multifunctional metal chelators, that is, chelators designed with ligands that are not only capable of efficiently coordinate the metal cations but also have additional properties.3 In the present work, we plan to advance in the development of biocompatible and effective copper chelators in collaboration with the experimental group of Dr. Patrick Gámez of the Universitat de Barcelona. These chelators will be based on small histidine-containing peptides (Figure 1) with high Cu2+

affinity and selectivity. There are several experimental results about this metal chelators but the sphere metal coordination is not known. Quantum chemical methods allowed us the study of the coordination of these peptides to the metal by computing the Cu2+ affinity of the different possible coordination environments and the redox potentials of the complexes formed. 1 A. Rauk, Chem. Soc. Rev., 2009, 38, 2698–715. 2 C. Opazo, X. Huang, R. A. Cherny, R. D. Moir, A. E. Roher, A. R. White, R. Cappai, C. L. Masters, R. E. Tanzi, N. C. Inestrosa and A. I. Bush, J. Biol. Chem., 2002, 277, 40302–40308. 3 M. A. Telpoukhovskaia and C. Orvig, Chem. Soc. Rev., 2013, 42, 1836–46.

Figure 1. Cu2+ chelator

P23 First approach to the diffusion of hydrogen confined in Single-walled Carbon Nanotubes using 6D quantum dynamics Manel Mondelo-Martell, Fermín Huarte-Larrañaga Departament de Química Física & Institut de Química Teòrica i Computacional, Universitat de Barcelona, Barcelona 08028, Spain e-mail:manel.mondelo@ub.edu Hydrogen has been largely regarded as an environmentally friendly alternative to the nowadays ubiquitous fossil fuels due to its high energetic efficiency and the low environmental impact of its combustion, which yields only water as secondary product1. However, its use is not still extended due to, among others, the difficulty to store this gas2. A lot of research has been made to develop new storage devices for low-density gases, and the most promising technology available nowadays is based on nanostructured materials, such as carbon nanotubes and metal-organic frameworks. Research on this field lead to the discovery of distortions on the dynamics and structure of the molecules trapped in the nanostructure, known as quantum confinement effects. This phenomena allows for potential applications, as they could be used for designing better storage devices for different compounds, separate isotopes or even tailoring properties of chemical reactions3-8. In the present work we show a first study of the diffusion of a hydrogen molecule inside an (8,0) single-walled carbon nanotube. In order to account for all possible quantum effects, including tunneling and quantum confinement, the simulations are made using a full quantum method, the Multiconfigurational Time-dependent Hartree approach9. A 6D Hamiltonian is applied in order to fully consider the couplings between the different degrees of freedom, which has been shown to be of great importance. The propagations are carried out using 5D stationary states as reference and letting them evolve freely in time in 6D. The analysis of these propagations is made using overlap functions in order to compare the initial state with the different time steps. This analysis shows that the diffusion along the nanotube is an almost separable coordinate, which opens the way to a development of novel strategies to improve the model without an increase of the computational effort10. [1] S. Dunn, Int. J. Hydrogen Energy 27, 235 (2002). [2] L. Schlapbach and A. Züttel, Nature 414, 353 (2001). [3] T. K. Nielsen, U. Bösenberg, R. Gosalawit, M. Dornheim, Y. Cerenius, F. Besenbacher, and T. R. Jensen, ACS Nano 4, 3903 (2010). [4] J.-R. Li, R. J. Kuppler, and H.-C. Zhou, Chem. Soc. Rev. 38, 1477 (2009). [5] I. Matanović, J. L. Belof, B. Space, K. Sillar, J. Sauer, J. Eckert, and Z. Bačić, J. Chem. Phys. 137, 014701 (2012). [6] M. Xu, L. Ulivi, M. Celli, D. Colognesi, and Z. Bačić, Chem. Phys. Lett. 563, 1 (2013). [7] G. Garberoglio, Chem. Phys. Lett. 467, 270 (2009). [8] M. Mondelo-Martell and F. Huarte-Larrañaga, J. Chem. Phys. 142, 084304 (2015). [9] M. Beck, A. Jäckle, G. A. Worth, and H.-D. Meyer, Phys. Rep. 324, 1 (2000). [10] M. Mondelo-Martell and F. Huarte-Larrañaga, Chem. Phys. (2015), 10.1016/j.chemphys.2015.07.029. (In press)

P24 Computer Aided Laccases Engineering: new directions Emanuele Monza,a Maria Fatima Lucas,a,b Ferran Sancho, a,b Gerard Santiago,a Victor Guallara,c a. Joint BSC-CRG-IRB Research Program in Computational Biology, Barcelona Supercomputing Center, Carrer Jordi Girona 29, 08034 Barcelona, Spain b. Anaxomics Biotech, Carrer Balmes 89, 08008 Barcelona, Spain c. ICREA, Passeig Lluís Companys 23, 08010 Barcelona, Spain emanuele.monza@bsc.es Laccases are multicopper oxidases whose catalytic core is organized in two copper centers: the T1 site, where substrate is oxidized, and the trinuclear cluster (TNC), where molecular oxygen is reduced to water.1 The broad specificity of these proteins, the use of oxygen as sole final electron acceptor and its conversion into water make them ideal candidates for sustainable industrial processes.2 Techniques, such as directed evolution, which mimic the natural selection process by evolving proteins towards the improvement of a given property, have unquestionably demonstrated their value and are routinely used in large industrial companies.3 Nevertheless, the brute force employed in these methods, could significantly gain from an all-atom description of the underlying catalytic mechanisms, to center the efforts on more limited areas of the protein. We have recently developed a computational tool4 which combines PELE's protein-ligand conformational sampling5 with the electronic structure description of QM/MM methods,6 to study the details of a variety of reactions. The application of this method has shown its potential at the descriptive and predicting level, paving the way for in silico evolution of fit-for-purpose laccases. This work was done in collaboration with: Centro de Investigaciones Biológicas, Institut des Sciences Moleculaires de Marseille and Novozymes A/S. Funded by the INDOX (KBBE-2013-7-613549) European Project. (1) Giardina, P.; Faraco, V.; Pezzella, C.; Piscitelli, A.; Vanhulle, S.; Sannia, G. Cell. Mol. Life Sci. 2010, 67 (3), 369. (2) Riva, S. Trends Biotechnol. 2006, 24 (5), 219. (3) Mate, D. M.; Alcalde, M. Biotechnol. Adv. 2015, 33 (1), 25. (4) Monza, E.; Lucas, M. F.; Camarero, S.; Alejaldre, L. C.; Martínez, A. T.; Guallar, V. J. Phys. Chem. Lett. 2015, 6 (8), 1447. (5) Borrelli, K. W.; Vitalis, A.; Alcantara, R.; Guallar, V. J. Chem. Theory Comput. 2005, 1 (6), 1304. (6) Senn, H. M.; Thiel, W. Angew. Chem. Int. Ed. 2009, 48 (7), 1198.

P25

Importance of Surface Morphology in Interstellar H2 Formation Javier Navarro-Ruiz,a Mariona Sodupe,a Piero Ugliengob and Albert Rimolaa a Universitat Autònoma de Barcelona, Departament de Química, Campus Universitari s/n (Edifici Cn), 08193 Bellaterra – SPAIN b Università degli Studi di Torino, Dipartimento di Chimica and NIS Centre, Via P. Giuria 7, 10125 Torino – ITALY E-mail: jnavarro@qf.uab.cat Among the molecules in space, H2 is one of the most relevant of the universe. It is the most abundant one in the interstellar medium and is a key intermediate for the formation of more complex molecules. Its formation is complex, but due to its inherent relevance understanding its interaction and its formation can be considered as a paradigm of the astrophysical process. The association of two H atoms on the surfaces of cosmic dust, which consists partly of Mg,Fe-silicates, is thought to be the main reaction channel for H2 formation in these regions. It involves first the adsorption of two H atoms on the surface, the subsequent diffusion of the ad-atoms and the final recombination to form H2. In this presentation1,2 H2 formation has been studied by means of B3LYP-D2* periodic calculations performed with the CRYSTAL09 code. The reaction has been studied on the (010), (001) and (110) surfaces of forsterite (Mg2SiO4), which represent surfaces of different stability and surface morphology. The most stable H adsorption states sport the H atom on surface O atoms, forming a SiOH group and transferring the H spin density to the neighbouring Mg ion. Because of that, the second H adsorption occurs on the neighbouring Mg ion, forming an Mg–H surface group, which enables the formation of a di-hydride OH···H–Mg hydrogen bond. The formation of a H2 molecule adopting a Langmuir–Hinshelwood mechanism does not occur equally at the different forsterite surfaces. On the (010) surface, the formation takes place either starting from two physisorbed H atoms through a radical–radical coupling driven reaction or from a hydride-proton coupling with barriers surmountable at very low temperatures. On the (110) surface, the unique route is through a hydride-proton coupling, whereas on the (001) surface the reaction is disfavoured because the doubly-H-adsorbed initial states are more stable than the final product. (1) Navarro-Ruiz, J.; Sodupe, M.; Ugliengo, P.; Rimola, A. Interstellar H Adsorption and H2 Formation on the Crystalline (010) Forsterite Surface: A B3LYP-D2* Periodic Study. Phys. Chem. Chem. Phys. 2014, 16 (33), 17447–17457. (2) Navarro-Ruiz, J.; Martínez-González, J. Á.; Sodupe, M.; Ugliengo, P.; Rimola, A. Relevance of Silicate Surface Morphology in Interstellar H2 Formation. Insights from Quantum Chemical Calculations. Mon. Not. R. Astron. Soc. 2015, 453 (1), 914–924.

P26 Electronic-Structure-Based Chemical Descriptors: (In)dependence on Self-Interaction and Hartree-Fock Exchange Almudena Notario-Estévez, Sergey M. Kozlov, Francesc Viñes,* and Francesc Illas Departament de Química Física & Institut de Quimica Teòrica i Computacional (IQTCUB), Universitat de Barcelona, c/Martí i Franquès 1,08028 Barcelona, Spain. The interaction of atoms and molecules with solid surfaces is ubiquitous in surface and materials science, tribology, nanotechnology, and eventually, heterogeneous catalysis. The bonding between an adsorbate and a transition metal surface is inherently governed by the quantum interaction between their orbitals and bands, respectively. In principle, to accurately calculate the magnitude of such interaction one has to resort to time-consuming computational techniques, which cannot be applied to large-scale screening involving hundreds of materials. Hence, various easy to calculate parameters of metal´s electronic structure were proposed to be used as descriptors for different properties of interest, such as adsorption energies and reaction energy barriers. In turn, electronic structure of transition metal surfaces is typically calculated with non-exact density functionals containing various degrees of approximations. The purpose of this work is to evaluate how the choice of a particular type of exchange-correlation functional affects the value of obtained electronic-structure-based descriptors. Namely, we calculated such descriptors as d-band center, width-corrected d-band center and the position of the highest peak in Hilbert transform of density of states with VWN, PBE, TPSS, and HSE06 functionals of LDA, GGA, meta-GGA and hybrid families, respectively. Whereas VWN and PBE functionals suffer from so-called self-interaction error, TPSS and HSE06 functionals intent to correct by additional term dependent on kinetic energy density of electrons or explicit Fock-exchange. Here we show for all 30 transition metals that original or width-corrected d-band centers as well as descriptors based on Hilbert transform are unaffected by self-interaction, while poor treatment of static correlation by hybrid functionals leads to an unbalanced description. Hence, hybrid functionals are not advised when computing electronic structure based descriptors for transition metal materials.

P27 IN SILICO STRUCTURAL AND FUNCTIONAL INVESTGATION OF CYSTEINYL LEUKOTRIENE RECEPTORS Ajay Pal1, Alison L. Reynolds1, Brendan N.Kennedy1, Denis Shields1, Anthony J. Chubb2

1UCD School of Biomolecular & Biomedical Science, Conway Institute, University College Dublin, Belfield, Dublin 4, Ireland. (ajay.pal@ucdconnect.ie) 2 RCSI School of Postgraduate Studies, Royal College of Surgeons in Ireland, Dublin 2, Ireland. Previously, a phenotype-based drug discovery study to uncover novel anti-angiogenic compounds identified Quininib, a cysteinyl leukotriene antagonist, opening up new possibilities for developing potential drug treatments for various diseases. Therefore, insight into ligand-receptor interactions is of pivotal importance for the design of new ligands with therapeutic potential. In order to study these interactions, three-dimensional structural information about the receptor structure can be most helpful. The predicted homology models of CysLTRs have the signature residues and motifs expected of GPCRs, and were used for carrying out molecular docking studies with an endogenous ligand and selective antagonists. Three sets of explicit molecular dynamics simulations were carried out in an all-atom lipid bilayer to analyse the stability of the predicted model, differences in structural features such as transmission/activation switch ”WxP” on Helix-6 (TM-6), helical tilts and hydrogen bonding pattern that help distinguish between the agonist- and antagonist-bound receptor forms. In absence of a crystal structure, this study may help to shed light on the structural features of CysLTRs, and their behaviour in the presence of an agonist and an antagonist, which might be helpful in the future drug discovery process for asthma, ischemic heart disease, atherosclerosis and ocular neovascular disorder.

P28 A ReaxFF based MD simulations of MgCl2 hydrates Amar Deep Pathak, Silvia Nedea, David Smulders, Herbert Zondag, and Camilo Rindt Department of Mechanical Engineering (Energy Technology), Eindhoven University of Technology, Den Dolech 2, 5612 AZ Eindhoven, The Netherlands Continuously rising energy demand and global warming associated with CO2 emission bring the need for clean and renewable source of energy. Energy demand in thermal comfort of building consumes a major fraction of energy in Europe [1]. Solar energy is one of the important pillar of renewable energy systems and can suffice the energy demand in this sector. There is however a mismatch between the heat demand in winter and solar energy availability in summer. Seasonal heat storage is a technology to improve the applicability of solar energy. Thermochemical materials (TCM) works on reversible thermo-chemical reaction and used for seasonal solar heat storage. Salt hydrates are one class of TCM materials. MgCl2 hydrates has high storage capacity and fast kinetics hence offers compact energy storage material for seasonal heat storage. MgCl2 hydrates stores solar energy in chemical form via hydration-dehydration cycle. Hydrolysis is an irreversible competitive reaction along with dehydration reaction in the charging cycle and affects the durability. Hydrolysis can be minimized under a specific temperature range and HCl partial pressure and by appropriate chemical mixing of other salts [2]. Heat and mass transport (H2O diffusion) inside the solid salt hydrate affects the reaction rate and sometime became rate determining step. A molecular insight of hydrolysis and dehydration reaction is an essential step in designing the MgCl2 hydrate based storage system. Reactive force field (ReaxFF) based Molecular dynamics (MD) are the key to gain the molecular insight of chemical reactions and mass transport. Chemical reactivity has been incorporated using bond order dependent ReaxFF. A new force field has been developed to describe the dehydration/hydrolysis reactions of MgCl2 hydrates. The diffusive transport of H2O molecule inside the salt hydrate can affect the rate of dehydration reaction. We perform MD simulations under 300K to 500K to measure the reaction kinetics. Reference: 1. K.E.N’Tsoukpoe, T Schmidt, H U Rammelberg, B.A. Watts and W K.L Ruck A systematic multi-step screening of numerous salt hydrates for low temperature thermochemical storage Applied Energy 2014 124 pp 1-16 2. Shoval S., Yariv S., Kirsh Y. and Peled H. The Effect of Alkali-Halides on Thermal Hydrolysis of Magnesium-Chloride and Magnesium Bromide. Thermochimica Acta 1986 109 pp 207-226

P29 Molecular mechanism of positive allosteric modulation of the metabotropic glutamate receptor 2 by JNJ-46281222 Maarten L J Doornbos, Laura Pérez-Benito, Gary Tresadern, Thea Mulder-Krieger, Ilse Biesmans, Andrés A Trabanco, Jose María Cid, Hilde Lavreysen, Adriaan P IJzerman & Laura H Heitman The metabotropic glutamate receptor 2 (mGlu2R) belongs to class C G protein-coupled receptors (GPCRs). Activation of the mGlu2 receptor is a potential strategy for the treatment of psychiatric disorders such as schizophrenia, anxiety and depression. Due to the highly conserved orthosteric binding site, the exploration of positive allosteric modulators (PAMs) has increased tremendously over the last decade. Here we study the binding of the PAM JNJ-46281222 in order to gain a better understanding of the relation between orthosteric and allosteric binding sites at the mGlu2 receptor. We present JNJ-46281222 docked into an active state model of the 7TM of the mGlu2R. We use molecular dynamics simulations to further analyze the binding mode of JNJ-46281222, which was validated experimentally in subsequent mutagenesis experiments. The results obtained with JNJ-46281222 in unlabeled and tritiated form further contribute to our understanding of mGlu2 allosteric modulation. The computational simulations and mutagenesis provide a plausible binding mode with indications of how the ligand permits allosteric activation. Altogether, this work offers new insights into the functioning of the mGlu2 receptor, which might contribute to development of new and improved PAMs

P30 The model of amorphization in inorganic eutectic alloys and composites Aleksandr A. Petrov1,2,3, Denis S. Ostroumov1,3, Natalia V. Kireeva1,3, Vladislav S. Pervov2

1 Moscow Institute of Physics and Technology, Moscow 2 Kurnakov Institute of General and Inorganic Chemistry RAS, Moscow 3 Frumkin Institute of Physical Chemistry and Electrochemistry RAS, Moscow E-mail: petrov.aa@phystech.edu The amorphization and eutectic colonies formation in inorganic alloys are a complex processes which are governed by numerous of parameters. In order to determine and to estimate these parameters the two-dimensional simplified model of binary alloys solidification is developed. We believe that the set of parameters represented in the model will allow us to determine the way the incommensurability and possible interactions affect mode and trajectories of self-organization processes of metastable states during solidification [1, 2]. Using model provides us with estimation of the impact of size characteristics, molar excess volume of forming metastable phases, which characterize the degree of amorphization, and determine how excess volume depends on incommensurability and mutual concentration [3]. We have determined that excess volume depends on the mutual concentration, but the dependency varies from insignificant at small incommensurability to large when incommensurability reaches 10 percent. Incommensurability causes fast growing of excess volume, but then excess volume reaches a plateau. Further increase in incommensurability does not affect excess volume. The maximum excess volume is observed at the concentration ratio 1:1. Other studies have shown that the considered system consists of clusters separated with amorphised boundaries until the incommensurability reaches 10 percent, while 20 percent of the incommensurability corresponds to completely amorphous system. Because of the distinctions in the properties of the amorphised phase boundaries and those of the corresponding substructures the special descriptors' parameters considering these differences have to be proposed to accelerate and to improve the development of new functional composite inorganic materials involving computational chemistry and chemoinformatics approaches. Acknowledgements Financial support by the Russian Foundation for Basic Research (projects 14-03-00223 and 14-29-04084) and by President Grant for Support of Young Russian Scientists (MK-1003.2014.3) is gratefully acknowledged. References [1] Pervov V.S. et al. ChemPhysChem, 2013, vol. 14, no. 17, p. 3865. [2] Maier J. Chem. Mater. 2014, vol. 26, no. 1, p. 348. [3] Pervov V.S. et al. Rus. Chem. Rev., 2003, vol. 72, no. 9, p. 759.

P31 Assessing drug-protein binding by simulation of stereoselective energy transfer dynamics: electronic interactions between tryptophan and flurbiprofen Silvana Pinheiro,a and Carles Curutcheta a)Departament de Fisicoquímica, Facultat de Farmàcia, Universitat de Barcelona, Spain e-mail: sdesouza@ub.edu The fluorescence of proteins is a complex process, often involving several electronic energy transfer (EET) reactions between aromatic amino acids before light emission, typically arising from tryptophan. In protein-ligand complexes, the ligand can also modify the fluorescence properties by participating in these EET processes, as well as by contributing to electron transfer reactions or the formation of exciplexes. The complex interpretation of optical experiments, however, typically precludes a full explotation of the structural information encapsulated in such experiments and related to the drug-binding events observed. In this study, we simulate how energy transfers involving different flurbiprofen enantiomers modulate the fluorescence properties of model tryptophan-flurbiprofen (TRP-FBP) and flurbiprofen-HSA (human serum albumin) complexes, where stereoselective dynamic quenchings have been recently observed.1 To this aim, we combine classical MD techniques with the a polarizable QM/MM methodology we have recently developed2 and applied to study the light-harvesting properties of photosynthetic systems.3,4 On the basis of our results, we discuss the potential of structure-based simulations in the study of drug-binding events through fluorescence techniques. [1] I. Vayá, P. Bonancía, M. C. Jiménez, D. Markovitsi, T. Gustavsson and M. A. Miranda; Excited state interactions between flurbiprofen and tryptophan in drug-protein complexes and in model dyads. Fluorescence studies from the femtosecond to the nanosecond time domains. Phys. Chem. Chem. Phys. 2013, 15, 4727-4734. [2] C. Curutchet, A. Muñoz-Losa, S. Monti, J. Kongsted, G. D. Scholes and Benedetta Mennucci; Electronic energy transfer in condensed phase studied by a polarizable QM/MM model, J. Chem. Theory Comput. 2009, 5, 1838-1848. [3] C. Curutchet, J. Kongsted, A. Muñoz-Losa, H. Hossein-Nejad, G. D. Scholes and B. Mennucci; Photosynthetic light-harvesting is tuned by the heterogeneous polarizable environment of the protein, J. Am. Chem. Soc. 2011, 133, 3078-3084. [4] C. Curutchet, V. I. Novoderezhkin, J. Kongsted, A. Muñoz-Losa, R. Grondelle, G. D. Scholes and B. Mennucci; Energy Flow in the Cryptophyte PE545 Antenna is Directed by Bilin Pigment Conformation, J. Phys. Chem. B 2013, 117, 4263-4273.

P32 QM/MM Study of the Reaction Mechanism of Fructose-6-Phosphate Aldolase Ferran Planas, Jordi Bujons, Pere Clapés, Ramon Crehuet The aldol reaction is a key chemical reaction for the formation of new C-C bonds. The carboligation between two different carbonyl substrates is relevant for the synthesis of polyhydroxylated molecules, and aldolases are enzymes that catalyze this reaction with regioselectivity, stereoselectivity and enantioselectivity1. Fructose-6-phosphate aldolase (FSA) was first described catalyzing the the cleavage of Fructose-6-phosphate (Fru-6P) into Dihydroxyacetone (DHA) and Glyceraldehyde-3-phosphate (Ga-3P)2.. FSA is an aldolase of paramount interest because of its promiscuity, as it catalyzes the aldol reaction between a wide range of substrates3.

The specific roles of the active site residues are being currently discussed4, however, a complete computational study of FSA has not been done until this work. In this work, done with QM/MM methodology, we describe the mechanism of FSA with two different substrates: Fructose-6-phosphate and D-Threose, which cleavage produce two molecules of Glycolaldehyde (GO). We characterize the plausible transition states and intermediates, and we discuss some alternatives in the acid-base steps, as there are two different acid-basic residues in the active site (Tyrosine 131 and Aspartic Acid 6). Our study complements the experimental result of our collaborators to unveil the most plausible mechanism for the reaction and its energetic profile. 1.. Samland, A. K., Rale, M., Sprenger, G. a. & Fessner, W. D. The Transaldolase Family: New Synthetic Opportunities from an Ancient Enzyme Scaffold. ChemBioChem 12, 1454–1474 (2011). 2. Thorell, S., Schürmann, M., Sprenger, G. a. & Schneider, G. Crystal structure of decameric fructose-6-phosphate aldolase from Escherichia coli reveals inter-subunit helix swapping as a structural basis for assembly differences in the transaldolase family. J. Mol. Biol. 319, 161–171 (2002). 3. Clapés, P. & Garrabou, X. Current Trends in Asymmetric Synthesis with Aldolases. Adv. Synth. Catal. 353, 2263–2283 (2011). 4. Stellmacher, L. et al. Acid-Base Catalyst Discriminates between a Fructose 6-Phosphate Aldolase and a Transaldolase. ChemCatChem n/a–n/a (2015). doi:10.1002/cctc.201500478

P33 Performance of Density functional theory based methods in predicting core-level binding energies and the physical meaning of Kohn-Sham orbital energies. Noèlia Pueyo Bellafont,1 Paul S. Bagus 2 and Francesc Illas1 1) Departament de Química Física & Institut de Química Teòrica i Computacional (IQTCUB), Universitat de Barcelona, C/Martí i Franquès 1, 08028 Barcelona, Spain 2) Department of Chemistry, University of North Texas, Denton, TX 76203-5017, USA. A rigorous and general decomposition of core level binding energy (BE) values into initial and final state contributions to the BE’s is proposed that can be used within either Hartree-Fock (HF) and Density Functional Theory (DFT) methods. For HF wavefunctions, Koopman’s Theorem (KT) ensures that the initial state BE = -ei is rigorous. However, a systematic study of the N(1s) BE’s in nitrogen containing molecules at the HF, B3LYP, PBE0, LC-BPBE levels of theory allows us to conclude that KT does not hold for the Kohn-Sham (KS) core orbital energies. In fact, these do not provide either initial or final state estimate to the BE’s. On the other hand, the calculated results show that the shifts of the KS orbital energies with respect to an arbitrary reference follow the trend of the properly calculated initial state shifts, hence, while the HF and KS orbital energies have different meanings, the shifts of the KS orbital energies allow us to obtain the initial state contributions to the BE shifts. Finally, the properly calculated core level BE’s values including initial and final state contributions are in very good agreement to experiment, especially for the DFT based methods.

P34 SAFAN-ISP: A NEW SMALL MOLECULE FRAGMENT INFORMATICS APPROACH FOR IN-SILICO PROFILING Luisa Pugliese (1), Marco Botta (2) (1) S.A.F.AN. BIOINFORMATICS. via Fulvio Croce 23/B 10137 Turin Italy luisa.pugliese@safan-bioinformatics.it (2) Università degli Studi di Torino Dipartimento di Informatica Corso Svizzera 185 10149 Turin Italy botta@di.unito.it Background: "Ligandability" describes a protein target's ability to bind small molecules with high affinity. The pharmaceutical industry would be highly interested in any predictive tool that helps distinguish ligandable from non-ligandable targets. Many known computational methods for predicting ligandability require knowing the 3D structure of the protein target and/or understanding what ligand space is compatible with the target. Fragments (molecules of low complexity) sample chemical space more effectively than drug-sized molecules. Methods: SAFAN-ISP combines fragment sampling capabilities and "revisited" publicly available ligand target information concurrently predicting a compound's "ligandability" to 2900 protein targets belonging to 12 different classes, yielding binding constant as output. Target prediction are coupled to diseases:target and disease:side effect data to output drug repositioning information on selective drugs. Results: The method has been validated on the prediction of 18000 drug:target predictions that quantitatively correlate to the experimental data yielding a Pearson correlation coefficient of 0.8. The average error on the predicted binding constants, using a logarithmic scale is 0.5. Conclusion: SAFAN-ISP predictions are more precise than most of the other available methods and allow to find highly selective drugs. Using molecular fragments the technology obtains reliable predictions where other methods fail: using a unique weighted combination of whole-molecules and fragments match computations, consistently outperforms other tools based only on either methods, as measured by correlation with known experimental results. Using this ISP technology, molecules with lower similitude scores by either methods can be shown to nonetheless adhere closely to experiments, unlocking their potential as active principles in drug repositioning studies. The developed SAFAN-ISP technology can be used for: 1. Drug repositioning using the target: disease database 2. Side Effect prediction using target: side effect database 3. Target identification in phenotypic screening outputs. 4- Potential therapeutic indications for natural compounds.

P35 THE CONFORMATIONAL FREE ENERGY LANDSCAPE OF β-XYLOSE REVEALS A TWO-FOLD CATALYTIC ITINERARY FOR β-XYLANASES Javier Iglesias-Fernández1, Lluís Raich1, Albert Ardèvol2 and Carme Rovira1,3

1. Department of Chemistry. University of Barcelona. Martí i Franquès 1. 08028 Barcelona. 2. ETH Zürich, USI Campus, 6900 Lugano, Switzerland. 3. Institució Catalana de Recerca i Estudis Avançats (ICREA), 08018 Barcelona, Spain. lraich@ub.edu Unraveling the conformational catalytic itineraries[1] of glycoside hydrolases (GHs) is a growing topic of interest in glycobiology, with major impact in the design of GH inhibitors. β-xylanases are responsible for the hydrolysis of glycosidic bonds in β-xylans, a group of hemicelluloses of high biotechnological interest that are found in plant cell walls. The precise conformations followed by the substrate during catalysis in β-xylanases have not been unambiguously resolved, with three different pathways being predicted from structural analyses. In this work, we compute the conformational free energy landscape (FEL) of β-xylose to predict the most likely catalytic itineraries followed by β-xylanases. The calculations are performed by means of ab initio metadynamics, using the Cremer-Pople puckering coordinates as collective variables.[2] The computed FEL supports only two of the previously proposed itineraries, 2SO → [2,5B]ǂ → 5S1 and 1S3 → [4H3]ǂ

→ 4C1, which clearly appear in low energy regions of the FEL. Consistently, 2SO and 1S3 are conformations preactivated for catalysis in terms of free energy/anomeric charge and bond distances. The results[3] however exclude the OE → [OS2]ǂ

→ B2,5 itinerary that has been recently proposed for a family 11 xylanase.[4] Classical and ab initio QM/MM molecular dynamics simulations reveal that, in this case, the observed OE conformation has been enforced by enzyme mutation. These results add a word of caution on using modified enzymes to inform on catalytic conformational itineraries of glycoside hydrolases. 1) G. J. Davies, A. Planas and C. Rovira, Acc. Chem. Res. 2012, 45, 308-316. 2) X. Biarnés, A. Ardèvol, A. Planas, C. Rovira, A. Laio and M. Parrinello, J. Am. Chem. Soc. 2007, 129, 10686-10693. 3) J. Iglesias-Fernández, L. Raich, A. Ardèvol and C. Rovira, Chem. Sci. 2015, 6, 1167-1177. 4) Q. Wan, Q. Zhang, S. Hamilton-Brehm, K. Weiss, M. Mustyakimov, L. Coates, P. Langan, D. Graham and A. Kovalevsky, Acta Crystallogr. D 2014, 70, 11-23.

P36 UNDERSTANDING THE BEHAVIOUR OF MULTIFUNCTIONAL GOLD NANOPARTICLES (AuNPs) Asli Raman 1, Victor F. Puntes 2 and Carlos Jaime1 1 Department of Chemistry, Universitat Autònoma de Barcelona, 08193 Bellaterra, Barcelona, Spain. 2 Institut Català de Nanociència i Nanotecnologia (ICN2), Campus UAB, 08193 Bellaterra, Barcelona, Spain. Gold nanoparticles (AuNPs) have suitable properties for controlled drug delivery, cancer treatment, biomedical imaging, diagnosis and many others. This study allowed a better understanding on the most widely used mechanisms of AuNPs stabilization in biological media: pegylation and protein corona. Polyethylene glycol (PEG) is a regular coating of nanomedicines and that conformational changes of the PEG attached to the NP have been correlated to altered biodistribution and appearance of toxicity. Our aim is to understand the change of the conformation from mushroom to brush in the mixed monolayer, which will allow further engineering of the NP surface. The dissipative particle dynamics (DPD) method was used to simulate a system containing AuNPs functionalized with 11-Mercaptoundecanoic acid (MUA), thiolated polyethylene glycol (SH-PEG) and bovine serum albumin (BSA). The composition of the layer determines the physicochemical properties of the AuNPs. The competition between the MUA and SH-PEG on the surface of the nanoparticle is the key point to understand these systems.

Figure 1. A system of 15000 gold particles functionalized with 120 MUA and 120 SH-PEG (water particles were removed for clarity) - Ghosh, P.; Han, G.; De, M.; Kim, C. K.; Rotello, V. M., Gold nanoparticles in delivery applications. Advanced Drug Delivery Reviews 2008, 60, 1307-1315. - Sharma, P.; Brown, S.; Walter, G.; Santra, S.; Moudgil, B., Nanoparticles for bioimaging. Advances in Colloid and Interface Science 2006, 123–126, 471-485. - McMahon, S. J.; Hyland, W. B.; Muir, M. F.; Coulter, J. A.; Jain, S.; Butterworth, K. T.; Schettino, G.; Dickson, G. R.; Hounsell, A. R.; O'Sullivan, J. M.; Prise, K. M.; Hirst, D. G.; Currell, F. J., Biological consequences of nanoscale energy deposition near irradiated heavy atom nanoparticles. Sci. Rep. 2011, 1, 18, 9 pp.

P37 - Moeendabary, E.; Ng, T. Y.; Zangeneh, M., Dissipative particle dynamics: Introduction, Methodology and complex fluid applications – a review. Int. J. Appl. Mechanics 2009, 1,737.

GPCRtm: An amino acid substitution matrix for the transmembrane region of class A G Protein-Coupled Receptors Authors: Santiago Rios†, Marta F. Fernandez†, Gianluigi Caltabiano, Mercedes Campillo, Leonardo Pardo and Angel Gonzalez* *Correspondence: Angel.Gonzalez@uab.es †Equal contributors Laboratori de Medicina Computacional, Unitat de Bioestadística, Facultat de Medicina, Universitat Autònoma de Barcelona, 08193 Bellaterra, Barcelona, Spain Background: Protein sequence alignments and database search methods use standard scoring matrices calculated from amino acid substitution frequencies in general sets of proteins. These general-purpose matrices are not optimal to align accurately sequences with marked compositional biases, such as hydrophobic transmembrane regions found in membrane proteins. In this work, an amino acid substitution matrix (GPCRtm) is calculated for the membrane spanning segments of the G protein-coupled receptor (GPCR) rhodopsin family; one of the largest transmembrane protein family in humans with great importance in health and disease. Results: The GPCRtm matrix reveals the amino acid compositional bias distinctive of the GPCR rhodopsin family and differs from other standard substitution matrices. These membrane receptors, as expected, are characterized by a high content of hydrophobic residues with regard to globular proteins. On the other hand, the presence of polar and charged residues is higher than in average membrane proteins, displaying high frequencies of replacement within themselves. Conclusions: Analysis of amino acid frequencies and values obtained from the GPCRtm matrix reveals patterns of residue replacements different from other standard substitution matrices. GPCRs prioritize the reactivity properties of the amino acids over their bulkiness in the transmembrane regions. A distinctive role is that charged and polar residues seem to evolve at different rates than other amino acids. This observation is related to the role of the transmembrane bundle in the binding of ligands, that in many cases involve electrostatic and hydrogen bond interactions. This new matrix can be useful in database search and for the construction of more accurate sequence alignments of GPCRs.

P38 Toward accurate solvation energies of large biological systems S. Romero,a F. J. Luque,a X. Barril,a F. Lipparini,b B. Mennucci,c C. Curutchet,a a: Departament de Fisicoquímica, Facultat de Farmàcia, Universitat de Barcelona, Spain b: Laboratoire Jacques-Louis Lions - Laboratoire de Chimie Theorique - Institut du Calcul et de la Simulation, Sorbonne Universites, Paris, France c: Dipartamento de Chimica e Chimica Industriale, Universita di Pisa, Italy. e-mail: sromerte11@gmail.com Continuum solvation models, such as the widely used Poisson-Boltzmann or the Generalized Born techniques, have become a key tool in the molecular modeling community in order to understand how solvation/desolvation effects impact a variety of biological processes, e.g. drug binding or protein aggregation events.[1] A different family of continuum solvation models, like PCM or COSMO and known as apparent surface charge (ASC) methods,[1] which are the standard tool in the context of the quantum chemistry community, have been typically limited to small biological molecules because of the large computational cost associated to such calculations. Recently, however, a new implementation of COSMO based on a domain decomposition strategy (ddCOSMO) has been presented, which speeds up calculations by several orders of magnitude, thus paving the way for its application to very large systems. Here, we report the parametrization of ddCOMSO to the prediction of hydration free energies based on the MST solvation model developed in Barcelona,[1][2] which has been previously successfully applied to a variety of biological problems, e.g. tautomeric equilibria, octanol-water partition coefficients, or the definition of 3D hydrophobic/hydrophilic profiles of biomolecules. The parametrization is based on the recently developed PM6 semiempirical Hamiltonian and is based on a set of over 200 experimental hydration free energies. The new model opens the way to the accurate prediction of hydration free energies of very large biomolecules, thus going beyond the usual classical MM-PBSA or MM-GBSA approaches.

[1] M. Orozco, F. J. Luque, Chem. Rev. 100 (2000), 4187. [2] F. Lipparini, L. Lagardère, G. Scalmani, B. Stamm, E. Cancès, Y. Maday, J.-P. Piquemal, M. J. Frisch, B. Mennucci, J. Phys. Chem. Lett. 5 (2014) 953. [3] C. Curutchet, M. Orozco, F.J. Luque, J. Comput. Chem. 22 (2001), 1180; C. Curutchet, A. Bidon-Chanal, I. Soteras, M. Orozco, F.J. Luque, J. Phys. Chem. B 109 (2005), 3565.

P39

P40 Improving Docking Performance with input from MDMix Simulations

Ruiz-Carmona, Sergio1,2; Alvarez-Garcia, Daniel1,2 and Barril, Xavier1,2,3 1 Departament de Fisicoquímica, Facultat de Farmàcia, Universitat de Barcelona, Barcelona, Spain

2 Institut de Biomedicina de la Universitat de Barcelona (IBUB), Barcelona, Spain

3 Catalan Institution for Research and Advanced Studies (ICREA), Barcelona, Spain

Our group has developed MDMix, a method based on Molecular Dynamics of aqueous/organic solvent mixtures, initially as a means of detecting druggable cavities [1] and more recently as a tool capable of identifying pharmacophoric points and predicting water displaceability [2]. The method is based on the empirical observation that organic solvents tend to localize on binding sites [3,4], revealing binding hot spots. This information can be used at different levels in drug discovery, and is potentially complementary with docking, as the interaction hot spots identified with MDMix can be used as pharmacophoric restraints to supplement the scoring functions and guide the docking process.

On the other hand, we have recently released rDock: a free and open source docking program which can be used to dock small molecules against proteins and nucleic acids [5]. Thanks to its design and implementation, rDock can be installed in a computation cluster and be used in High Throughput Virtual Screening (HTVS) campaigns for docking thousands to millions of compounds in a matter of days. Important features of rDock are the ability to accept user-supplied pharmacophoric restraints in order to bias docking and a flexible definition of interstitial water molecules, which can have variable degrees of freedom.

In this work, we bring the two different methods together, allowing seamless introduction of MDMix-derived information into the rDock docking protocol. Water displaceability information is included at the level of cavity definition, while the binding hot spots are introduced in rDock as a new type of non-mandatory restraints. Initial results with test systems using the DUD benchmark set [6] demonstrate significant improvements over unbiased docking.

[1] Seco, J.; Luque, F. J.; Barril, X. J. Med. Chem. 2009, 52, 2363-2371.

[2] Alvarez-Garcia, D; Barril, X. J. Med. Chem. 2014, 23; 57(20):8530-9.

[3] Liepinsh, E.; Otting, G. Nat. Biotechnol. 1997, 15, 264-268.

[4] English, A. C.; Groom, C. R.; Hubbard, R. E. Protein Eng. 2001, 14, 47-59.

[5] Ruiz-Carmona, S. et al. PloS Comput Biol. 2014, 10 (4): e1003571

P41 Steroid Hydroxylation by Basidiomycete Peroxygenases: a Combined Experimental and Computational Study Sancho F,b,c Babot ED,a del Río JC,a Cañellas M,b,c Lucas F,b,c Guallar V,b,d Kalum L,e Lund H,e Gröbe G,f Scheibner K,f Ullrich R,g Hofrichter M,g Martínez AT,h Gutiérrez Aa

Instituto de Recursos Naturales y Agrobiología de Sevilla, CSICa; Joint BSC-CRG-IRB Research Program in Computational Biology, Barcelona Supercomputing Centerb; Anaxomics Biotechc; ICREAd; Novozymes A/Se; JenaBios GmbHf; TU Dresden, Department of Bio- and Environmental Sciencesg; Centro de Investigaciones Biológicas, CSICh ferran.sanchojodar@bsc.es Steroids comprise a group of cyclic organic compounds with a four-ring structure. They represent an important class of natural compounds that are widely spread in the nature underlying structural functions and many biological processes. Moreover, steroids have a huge variety of pharmacological properties, being the second largest category among medical products. The structure of these compounds has a direct relation with the physiological activity2. Thus, steroids can be naturally modified using fungal enzymes in order to search for new and more active compounds3. With this purpose, peroxygenases from three basidiomycete species were tested for the hydroxylation of a variety of steroidal compounds, using H2O2 as the only cosubstrate. The enzymatic reactions on free and esterified sterols, steroid hydrocarbons and ketones were monitored by gas chromatography, and the products were identified by mass spectrometry. Hydroxylation at the side chain over the steroidal rings was preferred, with the 25-hydroxyderivatives predominating. Interestingly, antiviral and other biological activities of 25-hydroxycholesterol have been reported recently4. To understand the yield and selectivity differences between the different steroids, a computational study was performed using Protein Energy Landscape Exploration (PELE) software for dynamic ligand diffusion. These simulations showed that the active-site geometry and hydrophobicity favors the entrance of the steroid side chain, while the entrance of the ring is energetically penalized. Also, a direct correlation between the conversion rate and the side chain entrance ratio could be established that explains the various reaction yields observed. Funded by the INDOX (KBBE-2013-7-613549) European Project. (1) Babot ED et al., Appl Environ Microbiol 2015, 81, 4130 –4142. (2) Donova MV et al., Appl Microbiol Biotechnol,2012,94, 1423–1447 (3) Beneventi E et al., J Mol Catal B Enzym 2009, 58, 164 –168.

P42 Molecular Dynamics Analysis of the Interactions between Hen Egg-White Lysozyme and Several Polyoxometalates Acting as Selective Proteases.

Albert Solé-Daura,† Vincent Goovaerts,‡ Karen Strootbants,‡ Gregory Absillis,‡ Pablo Jiménez-Lozano,† Tatjana N. Parac-Vogt,‡ Josep M. Poblet,† Jorge J. Carbó*† †Departament de Química Física i Inorgànica, Universitat Rovira I Virgili (URV), Tarragona, Spain ‡Laboratory of Bioinorganic Chemistry, KU Leuven, Heverlee, Belgium albert.soled@estudiants.urv.cat Polyoxometalates (POMs) are a versatile and tunable class of inorganic polynuclear metal-oxygen clusters that have shown potential applications in biochemistry and medicine,1 including the ability to hydrolyze peptide bonds in proteins selectively. Notably, hen egg-white lysozyme (HEWL) is specifically cleaved in the presence of a 3- charged Ce(IV)-substituted Keggin POM (W11Ce) at Trp28–Val29 and Asn44–Arg45 peptide bonds (sites I and II respectively).2 Although the computational modeling has been largely employed to study POM chemistry,3-5 the study of POM···protein interactions is still an unexplored field for computational chemistry. Thus, we have run MD simulations to shed some light on the unexplored physicochemical foundations of the POM···protein interactions. On a step further, we have performed a comparative study between W11Ce and other transition metal-substituted POMs in order to understand how POM features such as the shape, the size, the charge or the type of incorporated metal ion and its ligands influence on the POM···protein interactions MD simulations revealed two regions of the protein in which the POM interacts strongly that can be related with the observed selectivity in the hydrolytic activity. There, the negatively charged polyoxoanion chiefly interacts with the positively charged and the polar H-donor residues of HEWL. In addition, we concluded that the main factors governing these interactions are the charge and the size of the POM and therefore, a strong and long-lived POM···protein interaction may balance them to occur. [1] Chen, Q.; Yang, L.; Zheng, C.; Zheng, W.; Zhang, J.; Zhou, Y.; Liu, J. Nanoscale 2014, 6, 6886-6897 [2] Strootbants, K.; Moelants, E.; Ly, H. G. T.; Proost, P.; Bartik, K.; Parac-Vogt, T. N. Chem. Eur. J. 2013, 19, 2848-2858 [3] Antonova, N. S.; Carbó, J. J.; Kortz, U.; Kholdeeva, O. A.; Poblet, J. M. J. Am. Chem. Soc. 2010, 132, 7488–7497 [4] López, X.; Carbó, J. J.; Bo, C.; Poblet, J. M. Chem. Soc. Rev. 2012, 41, 7537–7571 [5] Jiménez-Lozano, P.; Carbó J. J.; Chaumont, A.; Poblet, J. M.; Rodríguez-Fortea, A.; Wipff, G. Inorg. Chem. 2014, 53, 778–786

P43 EXPLORING THE CONFORMATIONAL PREFERENCES OF SMALL MOLECULES WITH THE MULTILEVEL STRATEGY. AUTHORS: ANTONIO VIAYNA, JORDI JUÁREZ, XAVIER BARRIL, F. JAVIER LUQUE Exploring the conformational preferences of small and flexible compounds, as well as predicting the conformational penalty due to the selection of the bioactive conformation, is a challenging question in drug design. Recently, we have developed a Multilevel Strategy1,2 to explore the conformational space of flexible and small ligands and to obtain the relative free energy of the conformation wells. The Multilevel methodology relies on the predominant-state approximation, which establish that the conformational space can be divided into different conformational wells (M) and that the total configurational integral is equal to the sum of the configurational integral of all the wells. In this context, the free energy (G) of a flexible molecule can be expressed as the addition of the contributions of such wells (Gj):

𝐺 = ∑ 𝐺𝑗 = −𝑀

𝑗=1𝑅𝑇 ∑ ln [𝜈𝑗

𝑁𝑗 ∑ 𝑒−𝐸𝑘 𝑅𝑇⁄

𝑁𝑗

𝑘=1𝑘∈𝑗

]𝑀

𝑗=1

The Multilevel strategy relies on the combination of different theoretical methods to sample the conformational space and evaluate the stability of the wells. Thus, Low-Level (LL) methods are used to identify the conformational minima of the molecule and obtain the local free energy of each conformational well (Glocal). Then, High-Level (HL) methods are used to re-scale the conformational minima (Emin). In this work, we use the Multilevel method to explore the conformational preferences of a set of bioactive ligands in water and to predict the conformational penalty of selecting the bioactive species.

(1) Forti, F.; Cavasotto, C..; Orozco, M.; Barril, X.; Luque, F. J. A Multilevel Strategy for the Exploration of the Conformational Flexibility of Small Molecules. J. Chem. Theory Comput. 2012, 8, 1808-1819

(2) Juárez-Jiménez, J.; Barril, X.; Orozco, M.; Pouplana, R.; Luque, F.J. Assessing the Suitability of the Multilevel Strategy for the Conformational Analysis of Small Ligands. J.Phys. Chem. B 2015, 119, 1164-1172

P44 Yeast Mitochondrial Threonyl-tRNA Synthetase (MST1) as a Model Enzyme for Studying High Fidelity Information: First Part of a Multi-Scale Computational Chemistry Study Wanlei Wei1,2; Gerald Monard2; James W Gauld1

1 Department of Chemistry and Biochemistry, University of Windsor. 401 Sunset Ave, Windsor, ON N9B 3P4, Canada, gauld@uwindsor.ca 2 Faculté des Sciences – UMR 7565 SRSMC, Université de Lorraine. Boulevard des Aiguillettes B.P. 70239 Postal Code: 54506 Vandoeuvre-les-Nancy, France, gerald.monard@univ-lorraine.fr Accurate translation of genetically encoded information into proteins, the workhorses of cells, is a fundamental biological process critical for maintaining normal cellular functions. A key step in this process is the attachment of amino acids to their cognate tRNA in preparation of incorporation into proteins. This step catalyzed via two half-reactions by the ancient class of enzymes, aminoacyl-tRNA synthase (aaRS). First, the amino acid is activated by reaction with ATP to yield the aminoacyladenylate derivative. Then, within the same active, and using a substrate-assisted mechanism, they catalyze the transfer of the aminoacyl moiety onto its cognate tRNA. It should be noted that aaRS are important to a diverse array of physiological processes including viral assembly, cancer, and porphyrin biosynthesis. Remarkably, aaRSs achieve this with an exceptionally high-fidelity, having an intrinsic error rate of about 1 in 10 000, and have been referred to as “paradigms” of molecular specificity. There are three approaches used by aaRS to achieve such accuracy: (i) absolute fidelity in substrate binding, (ii) pre-transfer editing, and (iii) post-transfer editing. The latter two are of particular interest since as they involve the catalyst proofing and editing against errors. Post-translational editing sites, found in distal regions of several aaRSs, have been studied quite extensively in many organisms. However, the importance of pre-transfer editing has been largely overlooked until recently, and the detailed mechanisms of by which it may occur are unknown. MST1 is a particularly good enzyme to study the pre-transfer editing process as it achieves high fidelity attachment of tRNA to cognate amino acids in the absence of a post-translational editing site. Experimentally, MST1 was able to selectively hydrolyze against the non-cognate Seryl-AMP substrate, while retaining Threonyl-AMP. This is remarkable because there is a mere difference of a methyl group between the two substrates. The mechanism by which MST1 distinguishes between residues with such small differences is still unclear. In the first part of the study, we have applied molecular mechanics simulations (MM MD) on MST1, bound to either Ser-AMP or Thr-AMP using the AMBER14 program package with PMEMD GPU acceleration. We discovered some interesting differences in the enzyme when bound to different substrates, which give evidence and may explain why the non-cognate Ser-AMP is more easily hydrolyzed than the cognate substrate. These findings will be presented at the conference, which is the first part of the two-part project. In the near future, in order to gain further understanding of the enzymatic reactivity of MST1, QM/MM free energy calculations will be employed.

P45 A new hydrophobicity scale of amino acids based on IEF-MST calculated log P and log DpH William J. Zamora R, Josep M. Campanera & F. Javier Luque Departament de Fisicoquimica and Institut de Biomedicine (IBUB), Facultat de Farmacia, Universitat de Barcelona In the vast world of naturally occurring peptides, where more than 7000 peptides are known and approximately 140 peptide therapeutics are currently being evaluated in clinical trials (Fosgerau & Hoffmann, 2015), the rapid and accurate determination of their physicochemical properties is key in peptide drug discovery. Among these properties, hydrophobicity is crucial for understanding molecular recognition and biomolecular aggregation. Hence, there is a great interest in determining hydrophobicity scales for amino acid structures. In this work, octanol/water partition (log P) and octanol/water distribution (log DpH, Fig. 1) of N-acetyl-L-amino-acid methyl amides were determined by means of quantum mechanical IEF-MST solvation calculations taking into account the intrinsic conformational preferences of each amino acid according to Dunbrack`s libraries (Dunbrack & Karplus, 1993;1994). The results reveal log D7.4 differences for α-helical and β-sheet conformations in Arg, Lys, Hid, Asn, Gln, Met, Cys, Leu and Ile. Furthermore, by decomposing the octanol/water transfer free energy into electrostatic and non-electrostatic components, we estimated that the non-electrostatic cost of transferring the amino acid side chain amounts to 23.9 r 3.0 cal/mol.Å2, in agreement with previous estimates reported in the literature. Comparison of our scale with other theoretical and experimental hydrophobicity scales yields satisfactory results, leading to correlation coefficients ranging from 0.61 to 0.94. Additionally, the MST-derived hydrophobicity scale led to significant correlations with the RP-HPLC retention factors measured for eight decapeptides (r = 0.97) and for 195 influenza virus hemagglutinin 13-mer (Ac-YPYDVPDYASLRS-Amide) peptides (r = 0.80). Finally, the hydrophobicity scale was able to reproduce the experimental log P for 118 random neutral peptides (r = 0.92) and log D7.4 for 115 random charged peptides (r = 0.95), Fig. 2. Future studies will address the application of this methodology to nonproteogenic amino acids, the prediction of peptide hydrophobicity at global and atomic level in peptides, and the scoring of peptide-protein interactions.

Figure 1. Representation of log D7.4 values for twenty-one amino acid residues. Black circle, red box and blue triangle represent the log D7.4 values for total, D-helix and E-sheet conformers respectively. Residues in bold yield significantly values between D-helix and E-sheet conformers.

Figure 2. Representation of experimental log D7.4 for 115 random peptides versus hydrophobicity using our computed log P values (correlation in yellow) and also using our computed log D7.4 values (correlation in red).

P46 MULTI-LEVEL STRATEGY FOR ANALYSIS OF BIOACTIVE DRUG CONFORMATIONS Authors: Sanja Zivanovic1,2; Adam Hospital1,2, Modesto Orozco1,2,3 1 Institute for Research in Biomedicine (IRB Barcelona), Baldiri Reixac 10-12, 08028 Barcelona, Spain 2 Joint BSC-IRB Research Program in Computational Biology, Baldiri Reixac 10-12, 08028 Barcelona, Spain. 3 Department of Biochemistry and Molecular Biology, University of Barcelona, 08028 Barcelona, Spain e-mail: sanja.zivanovic@irbbarcelona.org Exploring the conformational preferences of small flexible ligands plays an increasingly important role in drug design. Estimating the relative free energy of a ligand in its target-bound state (i.e. the bioactive conformation) is necessary to understand the process of molecular recognition, to optimize the potency of bioactive molecules and to improve the accuracy of structure-based drug design methods. A set of 100 crystal structures of pharmaceutically relevant drug-like molecules was tested using multi-level computational strategy. Taking into account the flexible nature of these molecules, protonation state and tautomeric forms, makes our task even more challenging. We combined low-level method for sampling the conformational minima and high-level ab-initio calculations for estimating their relative stability in order to examine the conformational space of flexible ligands and to obtain the relative free energy of the conformational wells. In the future, this strategy may represent an efficient tool for predicting the conformational landscape of drugs while keeping a reasonable balance between chemical accuracy and computational cost.