Sugar conformations that enhance cleavage of glycosidic bonds in carbohydrate-active enzymes

~ 97 ~ Sugar Conformations that Enhance Cleavage of Glycosidic Bonds in Carbohydrate-Active Enzymes Lluís Raich 1 and Carme Rovira 1,2 1. Departament de Química Orgànica and Institut de Química Teòrica i Computacional (IQTCUB), Universitat de Barcelona, Martí i Franquès 1-11, 08028 Barcelona, Spain. 2. Institució Catalana de Recerca i Estudis Avançats (ICREA), Passeig Lluís Companys 23, 08018 Barcelona, Spain. [email protected] Abstract- Carbohydrate-active enzymes select particular conformations of one sugar of their carbohydrate substrates to enhance catalysis. The evolution of this conformation during catalysis, known as the catalytic itinerary, is of fundamental interest in glycobiology. Here we calculate the free energy landscape of βxylose to predict the possible catalytic itineraries of β-xylan enzymes. Our results discard one of the three catalytic itineraries proposed on the basis of X-ray structures of mutant enzymes. Additional classical and quantum mechanics/molecular mechanics (QM/MM) calculations indicate that complexes obtained from modified enzymes do not necessarily represent the structures that take place in the reaction coordinate. I. I NTRODUCTION The covalent bond that joins a simple carbohydrate molecule (also known as sugar) to another group, which may or may not be another carbohydrate, is known in chemistry as glycosidic bond. These bonds are extremely stable in solution and display half-lives of several million years. However, in the presence of enzymes such as glycoside hydrolases (GHs), the half-life of glycosidic bonds can decay up to the millisecond range [1]. How do these enzymes enhance the breakage of glycosidic bonds is a topic that has been extensively studied since 1894 Emil Fisher's 'lock and key' model [2]. It is nowadays accepted that, for beta oligosaccharides (carbohydrate molecules bound by beta glycosidic bonds), one of the sugar molecules changes shape upon binding to the GH enzyme [3], facilitating catalysis by placing the leaving group of the reaction in an axial position. In particular, a key sugar molecule deforms from its ground state chair conformation in solution to a new conformation (typically a boat, skew-boat or half-chair) that is less populated in the same conditions (Fig. 1). Knowing the precise conformation of the sugar molecule in the active site of the enzyme and how this shape changes during catalysis (the catalytic itinerary) is important not only to understand how these enzymes work, but also to design new drugs for GHs involved in human health and disease [4], as well as engineering GHs for their use in food, detergent, oil, gas and biotechnological industries. In particular, β-xylanases are GH enzymes 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. An intriguing aspect of these enzymes is that, unlike other GHs, their β-xylan carbohydrate substrates are assumed to adopt more than one conformation in the enzyme active site, thus leading to several conformational itineraries during catalysis (Fig. 2 B). Based on crystallographic structures, some βxylanases have even been predicted to adopt conformations that disobey previously established rules for enzymatic glycosidic bond cleavage [5]. Fig. 1. A) β-xylose sugar in the most populated conformation in solution: chair 4 C 1 . B) β-xylose oligosaccharide bound to a family 11 GH enzyme. The key sugar, which is located between the two glutamic catalytic residues, displays a 2 S O skew-boat conformation. Almost all hydrogens have been omitted for the sake of clarity. In recent years, our group has contributed to the understanding of how GHs tune carbohydrate conformations to enhance catalysis [6,8]. Specifically, we demonstrated that the computed conformational free energy landscape (FEL) of isolated monosaccharides (e.g. glucose and mannose) correlate with the observed X-ray structures of enzyme-ligand complexes [7]. In other words, one can predict the conformation and electronic properties that a given carbohydrate will exhibit in the active sites of GHs from gas phase calculations [8,9]. Here we investigate all possible conformations of a β-xylose sugar molecule using state-of-the- art computational techniques based on molecular dynamics (MD). In particular, we compute the conformational FEL of βxylose by ab initio metadynamics and use it to predict the most likely catalytic itineraries for β-xylanases. Moreover, we employ classical and quantum mechanics/molecular mechanics (QM/MM) MD simulations to check whether common enzyme modifications used in experiments can affect the observed xylose conformation, leading to incorrect assignments of catalytic itineraries. II. M ETHODS Metadynamics [10] is a molecular dynamics based technique that enhances phase space exploration by the addition of repulsive potentials on a set of collective variables (functions of atomic positions) that enclose the slowest modes that are relevant to the process of interest. We used metadynamics on the phase space of β-xylose described by Density Functional Theory, using θ and Ф Cremer and Pople puckering coordinates [11] as collective variables (Fig. 2 A). All ab initio and hybrid QM/MM molecular dynamics simulations where carried out within the Car-Parrinello approach, as implemented in the CPMD 3.15.1 ~ 98 ~ program [12]. The AMBER [13] force-field was used for the classical molecular dynamics. Fig. 2. A) Cremer and Pople puckering coordinates of a pyranose ring (Q, θ and Ф). This coordinates are able to unequivocally distinguish different conformations of a six-membered ring, having 4 C 1 chair represented at the top, 1 C 4 chair at the bottom, boats and skew-boats in the equator and halfboats and envelopes in between. B) Northern projection (Stoddart representation) with the proposed catalytic itineraries for different families of β-xylanases (the number after GH stands for the family). III. R ESULTS The conformational FEL of β-xylose (Fig. 3) contains a deep minimum in 4 C 1 as well as a wide minimum between 1 S 3 and 2 S O . As previously found for other pyranoses, the puckering coordinates of all the key xylose sugar molecules extracted from experimental structures of β-xylosidase enzyme-ligand complexes are located, without exceptions, in low-energy regions of the FEL. These low-energy regions agree with two of the three conformational routes previously proposed for different β-xylosidase families (orange and green pathways of Fig. 2 B). The excluded one (blue pathway), which passes through a high free energy region, was further studied by classical and QM/MM MD simulations on the crystallographic complex from which was proposed [14]. The results of these calculations suggest that the 4 C 1 / O E conformation found by X-ray experiments is an artifact due to enzyme mutation and that the most probable conformation in the wild type is 2 S O , which places the leaving group in an axial position and is energetically consistent with the computed FEL. Although not all mutant enzymes are expected to behave equally (similar calculations on the complex of a modified GH10 crystal structure [15] that agree with the FEL show both mutant and wild type with the same conformational pattern), our results indicate that complexes obtained from modified enzymes do not necessarily represent the structures that take place in the reaction coordinate. Fig. 3. Conformational free energy landscape of an isolated β-xylose (Stoddart representation of the puckering sphere, Northern hemisphere). Free energy is given in kcal·mol -1 and each contour line of the diagram corresponds to 1 kcal·mol -1 . A CKNOWLEDGMENT This work was supported by the Generalitat de Catalunya (grant 2014SGR-987) and by the Ministerio de Economía y Competitividad (MINECO) (grant CTQ2011-25871). We acknowledge the computer support, technical expertise, and assistance provided by the Barcelona Supercomputing Center- Centro Nacional de Supercomputación (BSC-CNS). L. R. acknowledges the University of Barcelona for a PhD studentship (APIF-2013/2014). R EFERENCES [1] Wolfenden, R. and Snider, M. J.; Acc. Chem. Res. 2001, 34, 938-945. [2] Fischer, E.; Ber. Dtsch. Chem. Ges. 1894, 27, 2985-2993. [3] Vocadlo, D. and Davies, G. D.; Curr. Opin. Chem. Biol. 2008, 12, 539- 555. [4] Gloster and Vocadlo, D.; Nat. Chem. Biol. 2012, 8, 683-694. [5] Iglesias-Fernández, J.; Raich, L.; Ardèvol, A. and Rovira, C.; Chem. Sci. 2015, 6, 1167-1177. [6] Biarnés, X.; Nieto, J.; Planas, A.; Rovira, C.; J. Biol. Chem. 2006, 281, (3), 1432-41. [7] Biarnés, X.; Ardèvol, A.; Planas, A.; Rovira, C.; Laio, A.; Parrinello, M.; J. Am. Chem. Soc. 2007, 129, (35), 10686-93. [8] Davies, G. J.; Planas, A.; Rovira, C.; Acc. Chem. Res. 2012, 45, (2), 308- 16. [9] Thompson, A. J.; Dabin, J.; Iglesias-Fernández, J.; Ardèvol, A.; Dinev, Z.; Williams, S. J.; Bande, O.; Siriwardena, A.; Moreland, C.; Hu, T. C.; Smith, D. K.; Gilbert, H. J.; Rovira, C.; Davies, G. J.; Angew. Chem. Int. Ed. Engl. 2012, 51, (44), 10997-1001. [10] Bussi, G.; Laio, A. and Parrinello, M.; Phys. Rev. Lett. 2006, 96, 090601. [11] Cremer, D. and Pople, J.A.; J. Am. Chem. Soc. 1975, 97, 1354– 1358. [12] CPMD, http://www.cpmd.org/, Copyright IBM Corp 1990-2008, Copyright MPI für Festkörperforschung Stuttgart 1997-2001. [13] Pearlman, D.A.; Case, D.A.; Caldwell, J.W.; Ross, W.S.; Cheatham, T. E.; Debolt, S.; Ferguson, D.; Seibel, G. and Kollman, P.; Comput. Phys. Commun. 1995, 91, 1–41. [14] Wan, Q.; Zhang, Q.; Hamilton-Brehm, S.; Weiss, K.; Mustyakimov, M.; Coates, L.; Langan, P.; Graham, D. and Kovalevsky, A.; Acta Crystallogr., Sect. D: Biol. Crystallogr. 2014, 70, 11–23. [15] Suzuki, R.; Fujimoto, Z.; Ito, S.; Kawahara, S.; Kaneko, S.; Taira, K.; Hasegawa, T. and Kuno, A.; J. Biochem. 2009, 146, 61–70.