Molecular dynamics simulations of the transport of ligands through pathways of haloalkane dehalogenases DhaAwt and DhaA31

Zuzana Dunajova, Sergio M. Marques, Jan Brezovsky*, Jiri Damborsky

Loschmidt Laboratories, Department of Experimental Biology and Research Centre for Toxic Compounds in the Environment RECETOX, Faculty of Science, Masaryk University, Brno

*brezovsky@mail.muni.cz

DhaA31 is a five-point mutant of the haloalkane dehalogenase DhaAwt, with 32-fold improvement of the overall catalytic rate towards the anthropogenic substrate 1,2,3-trichloropropane (TCP). The higher activity of DhaA31 was achieved by introducing bulky residues and thus narrowing the tunnels that connect the active site with the bulk solvent, what demonstrated the importance of the access pathways for the catalytic efficiency [1]. To understand the performance of enzymes at the molecular level, molecular dynamics (MD) simulations proved to be very beneficial because of the possibility to analyse details of the catalytic cycle. Previous computational studies that were focused on the ligand transport were performed with DhaA31 before solving its crystal structure, and were also significantly limited by the accessible time-scales [1, 2]. Now, with the availability of high quality crystal structures [3], GPU enabled computation [4], and generic accelerating methods [5], a more realistic study of individual steps of the catalytic cycle can be performed.

In this study, MD simulations of DhaAwt and DhaA31 were carried out with either the substrate TCP, or the products 2,3-dichloropropan-1-ol (DCP) and Cl- ion. The purpose was to monitor binding and release steps of the catalytic cycle, respectively. These simulations revealed that the substrate and the products strongly influence the tunnel opening in both enzymes and the main tunnel (known as p1) seems to be relevant for the transport of studied ligands in both enzymes. The binding and release of ligands was much slower in DhaA31, what is in agreement with the product release being the rate-limiting step. The TCP release from the active sites of both enzymes was much slower compared to DCP, which matches its more hydrophobic nature. Interestingly, several TCP molecules were observed to bind the active site of DhaAwt simultaneously, suggesting higher substrate inhibition of this enzyme. Simulations of TCP bound to the active site sampled more than three times larger amount of reactive positions with DhaA31 than with the DhaAwt, which could explain the better performance of this variant. The Tyr176 and Phe245 residues introduced in the DhaA31 notably contributed to the binding of TCP in the reactive positions. In conclusion, our MD simulations provided a mechanistic description of ligand transport in DhaAwt and DhaA31. This knowledge could help with designing the next generation of TCP-degrading enzymes.

1. M. Pavlova, M. Klvana, Z. Prokop, R. Chaloupkova, P. Banas, M. Otyepka, et al., Nat. Chem. Biol., 5, (2009), pp. 727–733.

2. M. Klvana, M. Pavlova, T. Koudelakova, R. Chaloupkova, P. Dvorak, Z. Prokop, et al., J. Mol. Biol., 392, (2009), pp. 1339–1356.

3. M. Lahoda, J. R. Mesters, A. Stsiapanava, R. Chaloupkova, M. Kuty, J. Damborsky, et al., Acta. Cryst., 70, (2014), pp. 209–217.

4. M. S. Friedrichs, P. Eastman, V. Vaidyanathan, M. Houston, S. Legrand, A. L. Beberg, et al., J. Comput. Chem., 30, (2009), pp. 864–872.

5. D. Hamelberg, J. Mongan, J. A. McCammon, J. Chem. Phys., 120, (2004), pp. 11919–11929.