Speeding up Directed Evolution: Combining the Advantages of Solid-Phase Combinatorial Gene Synthesis with Statistically Guided Reduction of Screening Effort

Abstract

Efficient and economic methods in directed evolution at the protein, metabolic, and genome level are needed for biocatalyst development and the success of synthetic biology. In contrast to random strategies, semirational approaches such as saturation mutagenesis explore the sequence space in a focused manner. Although several combinatorial libraries based on saturation mutagenesis have been reported using solid-phase gene synthesis, direct comparison with traditional PCR-based methods is currently lacking. In this work, we compare combinatorial protein libraries created in-house via PCR versus those generated by commercial solid-phase gene synthesis. Using descriptive statistics and probabilistic distributions on amino acid occurrence frequencies, the quality of the libraries was assessed and compared, revealing that the outsourced libraries are characterized by less bias and outliers than the PCR-based ones. Afterward, we screened all libraries following a traditional algorithm for almost complete library coverage and compared this approach with an emergent statistical concept suggesting screening a lower portion of the protein sequence space. Upon analyzing the biocatalytic landscapes and best hits of all combinatorial libraries, we show that the screening effort could have been reduced in all cases by more than 50%, while still finding at least one of the best mutants.

Beating Bias in the Directed Evolution of Proteins: Combining High‐Fidelity on‐Chip Solid‐Phase Gene Synthesis with Efficient Gene Assembly for Combinatorial Library Construction

“…Parallel to these developments, solid‐phase gene synthesis emerged; this potentially offered a chemical alternative to the usual multistep molecular biological mutant library construction. Although an initial step in this direction was made by using the Sloning method for solid‐phase gene synthesis, it was not efficient enough to replace the traditional SM‐based approach, that is, fewer but still too many errors were found in the synthetic libraries. Unfortunately, in that report and in other directed‐evolution investigations, a systematic quality control at the DNA level based on massive sequencing was not made.…”

“…We have also reported the concept of combinatorial library construction based on assembly of designed oligonucleotides (ADO), but the library quality proved to be unsatisfactory at the DNA level . More recently, we compared a traditional PCR‐based SM‐library of P450‐BM3 mutants with the respective chemical library generated by solid‐phase synthesis by using the Sloning method . Although the quality of the Sloning library at the protein level proved to be somewhat higher, it was far from perfect, which suggested that the advantage of such a synthetic approach was limited.…”

Directed evolution 2.0: improving and deciphering enzyme properties

“…A QuikChange experiment would be sufficient to generate 400 (20×20; two positions) or 8000 (three positions) different protein variants. Due to primer length limitations, saturation of more than three positions has to our best knowledge not been reported for the CASTing strategy 31,32,[76][77][78][79] . QuikChange based mutagenesis methods were further advanced to saturate 4 (PFunkel 80 ) to 6 (PFLF-MSDM 81 ) positions simultaneously.…”

Determination of a Screening Metric for High Diversity DNA Libraries

“…These synthetic DNA libraries allow the sequence space of antibodies, enzymes, various other proteins, and genomes to be more thoroughly examined [6–9]. An example of the use of a DNA library in antibody research is the screen of a library of 10 10 variants for the humanization of antibodies [10].…”