Generating segmental mutations in haloalkane dehalogenase: a novel part in the directed evolution toolbox

Abstract

Directed evolution techniques allow us to genuinely mimic molecular evolution in vitro. To enhance this imitation of natural evolutionary processes on a laboratory scale in even more detail, we developed an in vitro method for the generation of random deletions and repeats. The pairwise fusion of two fragments of the same gene that are truncated by exonuclease BAL-31 either at the 3' or 5' side results in a deletion or a repeat at the fusion point. Although in principle the method randomly covers the whole gene, it can also be limited to a predefined area in the sequence by controlling the level of the initial truncation. To test the procedure and to illustrate its potential, we used haloalkane dehalogenase from Xanthobacter autotrophicus GJ10 (DhlA) as a model enzyme, since the adaptation of this enzyme towards new substrates is known to occur via the generation of this type of mutation. The results show that the mutagenesis method presented here is an effective tool for accessing formerly unexplorable sequence space and can contribute to the success of future directed evolution experiments.

Figure 1

Schematic illustration of the segmental mutagenesis procedure. The vector is first linearized either at the 5′ or 3′ end of the dehalogenase gene. Progressive exonuclease action and removal of the remaining vector DNA yields two batches of either 3′ or 5′ truncated gene fragments. After assembly of these two ends, the segmental mutagenesis library is ligated into a fresh vector, transformed to E.coli and can be screened for activity.

Figure 2

Proposed route for the evolution of DhlA. The putative predecessor is indicated as DhlAS/Dh1AT, denoting both the serine- and threonine-containing species. The order of the mutations that occurred after the initial duplication of an 11-amino acid stretch (1) is not clear. Event (2) represents a substitution of either a T→S or S→T, depending on which amino acid was present in the original sequence, and (3) indicates the substitution of W→F. The last step (4) shows the deletion of 2 amino acids in the first half of the repeat.

Figure 3

Sequence alignment of the primitive dehalogenase (DhlAS/T), the hypothetical mutant arising after the duplication of the 11-amino acid stretch, wild-type DhlA and the 10 selected mutants. The sequences are arranged according to an increasing length of the N-terminal part of the fusion. Repeats in the mutants are boxed, the part of the sequence involved in the repeat leading to the actual wild type is underlined.