Directed Evolution

DIRECTED EVOLUTION

The protein engineering techniques described to this point require knowledge of protein structure plus detailed knowledge of active-site function. Very few enzymes have been studied this intensively. Directed evolution is a powerful technique to alter the function of an enzyme without the need for exhaustive structural and functional data. Directed evolution can be used to change substrate specificity, either changing the enzyme to recognize a totally different substrate, or making subtle changes where the substrate is slightly different. The main premise of directed evolution is the random mutagenesis of the gene of interest, followed by a selection scheme for the new desired function.

As already described in Previews Pages, new ribozymes are isolated using a similar principle.

Directed evolution screens for new enzyme activities by constructing a library of different enzymes derived from the same original protein. Each protein in the library is slightly different because of random mutagenesis. Random mutants may be generated over the entire length of the gene sequence. Alternatively, certain target amino acids can be replaced by random amino acids. The third main method for generating mutants relies on recombination (homologous or nonhomologous).

These mutant genes are then screened for the new, desired enzyme activity after insertion into a suitable expression vector and host cell.

Random mutagenesis usually starts with a gene whose function is close to that desired. The gene is randomly mutated throughout the entire sequence using error-prone PCR (epPCR). Different methods exist to induce errors during PCR amplification. The most straightforward is to add MnCl₂ to the PCR reaction. Taq polymerase has a fairly high rate of incorporating the wrong nucleotide, and MnCl₂ stabilizes the mismatched bases. The error will be copied in subsequent rounds of amplification. Adding nucleotide analogs such as 8-oxo-dGTP and dITP, which form mismatches on the opposite strand, can also enhance the PCR error rate. These analogs in combination with MnCl₂ can induce a wide variety of different mutations along the length of a gene. Some random mutations that occur outside the active site may cause subtle changes with profound effects on substrate recognition and enzyme function.

Target mutagenesis is much more focused and requires some knowledge of the enzyme structure, including the active site. For example, tyrosyl-tRNA synthetase from Methanococcus jannaschii was mutated through directed evolution. The gene for this enzyme had been sequenced, but no structural data were available. By comparing the sequence with that of another tyrosyl-tRNA synthetase, whose structure had been solved, the researchers identified amino acids potentially involved in tyrosine recognition. These residues were then randomly mutagenized. Altering residues with known functions in substrate recognition is more likely to have potent effects.

The third method to form new enzymes via directed evolution involves various schemes for recombining different domains. These are based on homologous or nonhomologous sequences and encompass a variety of different protocols, including DNA shuffling and combinatorial protein libraries. Some of these are discussed in detail later.