© 2009 Nature America, Inc. All rights reserved. NATURE CHEMICAL BIOLOGY ADVANCE ONLINE PUBLICATION 1 ARTICLES Combining rational protein design with directed evolution provides a very efficient two-step approach for engineering proteins with desired activities 1–7 . Computer modeling is initially used to identify residues participating in substrate binding and transition state stabilization, then directed evolution is used for systematic mutagenesis of these “hot spot” residues and for fine-tuning constructs prepared by com- puter-guided site-directed mutagenesis 8 . The amino acid residues that are most frequently predicted as hot spots for mutagenesis are located in the active site or its close vicinity 9 . However, in the study presented here we targeted residues located in the access tunnels connecting the active site of the engineered protein (DhaA haloalkane dehaloge- nase from Rhodococcus rhodochrous NCIMB13064) with bulk solvent. The results demonstrate the power of combining rational design with directed evolution and highlight the importance of access tunnels as both evolutionary loci and protein engineering targets. Living organisms have typically had millions of years to evolve enzymes capable of metabolizing naturally occurring substances, including potential toxins. In contrast, anthropogenically gener- ated substrates introduced into the biosphere by humans since the Industrial Revolution are often recalcitrant and persist in the environ- ment because microorganisms have not had sufficient time to evolve enzymes capable of catalyzing their conversion 10,11 . One such toxic, non-natural compound is TCP (1), which is released into the environ- ment as a result of its manufacture, formulation and use as a solvent and extractive agent. TCP is used as a chemical intermediate in the pro- duction of polysulfone liquid polymers and 1,3-dichloroprop-1-ene (2). In addition, it is produced in significant quantities as a byproduct in the manufacture of other chlorinated compounds—for example, 2-(chloromethyl)oxirane (also called epichlorohydrin, 3) 12 . TCP has been detected in low concentrations in surface water, drinking water and groundwater, and is anticipated to be a human carcino- gen. Further, given that its half-life under groundwater conditions is estimated to extend up to 100 years, TCP is likely to persist for long periods of time in the groundwater environment 13 . However, under laboratory conditions, TCP can be slowly (k cat = 0.08 s -1 ) converted to 2,3-dichloropropane-1-ol (DCL, 4) by the action of microbial halo- alkane dehalogenase enzymes 14 , which cleave carbon-halogen bonds by a hydrolytic mechanism. The first dehalogenase reported to have detectable activity against TCP is the DhaA haloalkane dehaloge- nase 15 , which has been isolated from various strains of Rhodococcus spp. worldwide 16–20 . The active site of DhaA is an occluded cavity located between two protein domains 21 , with two major access tun- nels (referred to as the main tunnel and the slot tunnel 22 ) through which ligands can exchange between the active site and its surround- ing environment. The dehalogenation reaction proceeds in the active site via nucleophilic attack of a nucleophile on a carbon atom of the halogenated substrate, leading to cleavage of the carbon-halogen bond, displacement of a halide and formation of a covalent alkyl- enzyme intermediate. This alkyl-enzyme intermediate is subsequently hydrolyzed by a water molecule activated by a catalytic base 23 . DNA shuffling and error-prone PCR have been applied to the dhaA gene to improve the kinetic properties of DhaA for TCP con- version 14,24 . Two evolved double point mutants (G3D C176F and C176Y Y273F (M2)) were obtained that were 4 times and 3.5 times Redesigning dehalogenase access tunnels as a strategy for degrading an anthropogenic substrate Martina Pavlova 1,5 , Martin Klvana 1,5 , Zbynek Prokop 1 , Radka Chaloupkova 1 , Pavel Banas 2 , Michal Otyepka 2 , Rebecca C Wade 3 , Masataka Tsuda 4 , Yuji Nagata 4 & Jiri Damborsky 1 Engineering enzymes to degrade anthropogenic compounds efficiently is challenging. We obtained Rhodococcus rhodochrous haloalkane dehalogenase mutants with up to 32-fold higher activity than wild type toward the toxic, recalcitrant anthropogenic compound 1,2,3-trichloropropane (TCP) using a new strategy. We identified key residues in access tunnels connecting the buried active site with bulk solvent by rational design and randomized them by directed evolution. The most active mutant has large aromatic residues at two out of three randomized positions and two positions modified by site-directed mutagenesis. These changes apparently enhance activity with TCP by decreasing accessibility of the active site for water molecules, thereby promoting activated complex formation. Kinetic analyses confirmed that the mutations improved carbon-halogen bond cleavage and shifted the rate-limiting step to the release of products. Engineering access tunnels by combining computer-assisted protein design with directed evolution may be a valuable strategy for refining catalytic properties of enzymes with buried active sites. 1 Loschmidt Laboratories, Institute of Experimental Biology and National Centre for Biomolecular Research, Faculty of Science, Masaryk University, Brno, Czech Republic. 2 Department of Physical Chemistry and Center for Biomolecules and Complex Molecular Systems, Palacký University, Olomouc, Czech Republic. 3 Molecular and Cellular Modeling Group, EML Research gGmbH, Heidelberg, Germany. 4 Department of Environmental Life Sciences, Graduate School of Life Sciences, Tohoku University, Sendai, Japan. 5 These authors contributed equally to this work. Correspondence should be addressed to J.D. (jiri@chemi.muni.cz). Received 20 January; accepted 28 May; published online 23 August 2009; doi:10.1038/nchembio.205