Transport of ligands between buried active sites and bulk solvent is a key step in the catalytic cycle of many enzymes. The absence of evolutionary optimized transport tunnels is an important barrier limiting the efficiency of biocatalysts prepared by computational design. Creating a structurally defined and functional “hole” into the protein represents an engineering challenge. Here we describe the computational design and directed evolution of a de novo transport tunnel in haloalkane dehalogenase. Mutants with a blocked native tunnel and newly opened auxiliary tunnel in a distinct part of the structure showed dramatically modified properties. The mutants with blocked tunnels acquired specificity never observed with native family members: up to 32 times increased substrate inhibition and 17 times reduced catalytic rates. Opening of the auxiliary tunnel resulted in specificity and substrate inhibition similar to those of the native enzyme and the most proficient haloalkane dehalogenase reported to date (kcat = 57 s–1 with 1,2-dibromoethane at 37 °C and pH 8.6). Crystallographic analysis and molecular dynamics simulations confirmed the successful introduction of a structurally defined and functional transport tunnel. Our study demonstrates that, whereas we can open the transport tunnels with reasonable proficiency, we cannot accurately predict the effects of such change on the catalytic properties. We propose that one way to increase efficiency of an enzyme is the direct its substrates and products into spatially distinct tunnels. The results clearly show the benefits of enzymes with de novo transport tunnels, and we anticipate that this engineering strategy will facilitate the creation of a wide range of useful biocatalysts.