Protein mutations are often introduced at the DNA level in the corresponding gene, which takes advantage of the central dogma of biology (DNA is transcribed to mRNA; mRNA is translated to proteins. The introduction of mutations in target genes has become trivial with the advances in synthetic biology. Major classes of library construction include (1) random mutagenesis, (2) recombination, (3) semi-rational design, and (4) scanning mutagenesis. Together, these classes of library construction are termed “combinatorial design” and require little to no prior knowledge of the protein, albeit more knowledge can help design libraries more efficiently. The counter technique is termed “rational design” and requires detailed information about the protein for iterative computational mutagenesis. Both design processes have been instrumental in protein engineering, but this section focuses on the main classes of combinatorial design. For a more detailed description of methods for library construction, Packer and Liu recently published an excellent review (Packer and Liu 2015). Random mutagenesis is a method where substitutions, deletions, or insertions are introduced at random in the target gene, mimicking random mutations introduced during natural evolution, but at an accelerated and tunable rate. Random mutagenesis is particularly appropriate for studying proteins with a lack of structural or functional information. This method can highlight particular “hot spots” in the protein where mutations are most likely to be advantageous to the desired fitness, which can then be targeted for future rounds of mutagenesis. Several techniques have been used to create randomly mutated libraries including, ultra violet light exposure, chemical mutagens, error-prone strains, and error-prone polymerase chain reaction (ep-PCR) (Cirino et al. 2003). The latter has been the dominant method over the past 25 years owing to its specificity, tunability, and accessibility (preassembled mutagenesis kits are available through several biotechnology companies). Both low and high error-rates for ep-PCR have been successful for engineering functional protein variants (Daugherty et al. 2000; Hamamatsu et al. 2006; Kunichika et al. 2002; Takase et al. 2003; Zaccolo and Gherardi 1999), but the choice of error-rate is a balance between uniqueness and function and is dependent on the protein and the mutagenesis protocol (Drummond et al. 2005). Another technique that can mimic natural evolution is recombination, where fragments of DNA are swapped between similar genes using homologous recombination. In general, library design through recombination techniques involves the creation of DNA fragments from a single gene or multiple related genes, pooling all the fragments, and piecing back together full length genes. Each full-length gene will be a mixture of the different parts and constitute a unique member in the library. The DNA fragments can be derived from orthologs, genes from the same family, or even genes of previously isolated mutants. Several sophisticated and unique techniques have been developed for recombination library design including DNA shuffling (Stemmer 1994a; Stemmer 1994b), staggered extension process (StEP) (Zhao et al. 1998), random chimeragenesis on transient templates (RACHITT) (Coco et al. 2001), incremental truncation for the creation of hybrid enzymes (ITCHY), and SCRATCHY (a combination of DNA shuffling and ITCHY) (Lutz et
…show more content…
Based on knowledge about hot spots in the protein (e.g. binding pockets, catalytically active sites, dimerization domains), specific residues are selected for mutagenesis due to their importance in the hot spot. Residues are substituted randomly using techniques such as overlap- extension PCR (oe-PCR) and QquickC change mutagenesis PCR (QC-PCR). The former is popular for introducing multiple mutations in the gene. Degenerate oligonucleotides are used as primers in PCR to amplify fragments of the gene. Each amplified fragment contains a region of homology with the appropriate adjacent gene fragment. Fragments are mixed, similarly to DNA shuffling, in the absence of primer and allowed to extend until the full length gene has been amplified. Though this technique has proven successful, it can be tedious with multiple fragments and introduce bias during each round of PCR amplification. Recently with the advances in DNA synthesis, an entire gene can be synthesized with the proper degenerate sites. The preferred method in academia remains oe-PCR because of the cost of degenerate gene synthesis, but the prices of DNA synthesis is constantly declining and may lead to more attractive pricing when compared with the time and labor involved with in-house library
Based on knowledge about hot spots in the protein (e.g. binding pockets, catalytically active sites, dimerization domains), specific residues are selected for mutagenesis due to their importance in the hot spot. Residues are substituted randomly using techniques such as overlap- extension PCR (oe-PCR) and QquickC change mutagenesis PCR (QC-PCR). The former is popular for introducing multiple mutations in the gene. Degenerate oligonucleotides are used as primers in PCR to amplify fragments of the gene. Each amplified fragment contains a region of homology with the appropriate adjacent gene fragment. Fragments are mixed, similarly to DNA shuffling, in the absence of primer and allowed to extend until the full length gene has been amplified. Though this technique has proven successful, it can be tedious with multiple fragments and introduce bias during each round of PCR amplification. Recently with the advances in DNA synthesis, an entire gene can be synthesized with the proper degenerate sites. The preferred method in academia remains oe-PCR because of the cost of degenerate gene synthesis, but the prices of DNA synthesis is constantly declining and may lead to more attractive pricing when compared with the time and labor involved with in-house library