TY - JOUR
AU - Jung, Gyoo Yeol
AB - Abstract Evolutionary approaches have been providing solutions to various bioengineering challenges in an efficient manner. In addition to traditional adaptive laboratory evolution and directed evolution, recent advances in synthetic biology and fluidic systems have opened a new era of evolutionary engineering. Synthetic genetic circuits have been created to control mutagenesis and enable screening of various phenotypes, particularly metabolite production. Fluidic systems can be used for high-throughput screening and multiplexed continuous cultivation of microorganisms. Moreover, continuous directed evolution has been achieved by combining all the steps of evolutionary engineering. Overall, modern tools and systems for evolutionary engineering can be used to establish the artificial equivalent to natural evolution for various research applications. Sungho Jang and Minsun Kim have contributed equally to this work. Introduction Evolution enables living organisms to survive and adapt to changing environments. The mechanism underlying this process, descent with modification, not only shapes nature but also help researchers to engineer useful phenotypes. Adaptive laboratory evolution reproduces natural evolution in a research laboratory by passaging microbial cultures under selective pressures to which the cell must adapt [22, 48, 59, 66]. The evolutionary process has also been mimicked through directed evolution of enzymes [1, 14, 79], in which variants of a target gene are generated in vitro and transformed to cells and selected for the desired phenotype such as binding affinity. These evolutionary engineering approaches have emerged as important strategies for bioengineering. Numerous technologies are used in each step, with thenext-generation sequencing (NGS) technology enabling determination of genotype–phenotype relationships after the evolution [40, 71, 95]. In addition to the traditional technologies, recent advances in synthetic biology have improved the efficiency of evolutionary engineering and broadened the field to which these approaches can be applied. In vivo mutagenesis methods increased the library size and reduced the time and labor required to generate a library [105]. Genetically encoded biosensors have expanded the range of phenotypes that can be selected beyond the tolerance to stress and binding affinity of a ligand [51]. Microfluidic systems enabled high-throughput, single-cell screening [6, 43]. Synthetic genetic circuits and continuous evolution systems facilitated whole evolution cycles by reducing or even eliminating interventions by researchers. In this review, we describe recent progress in evolutionary engineering of biomolecules and microorganisms (Fig. 1). The main objective of this article is to provide an updated view of the tools and systems useful for accelerating evolutionary engineering. Therefore, we focus on the methods used in each step of evolutionary engineering: diversity generation and phenotype screening. Thereafter, a more advanced concept for an efficient evolution known as continuous directed evolution will be explained that requires least interventions by researchers such as PCR, cloning, and transformation. Fig. 1 Open in new tabDownload slide Scheme of evolutionary engineering Tools and systems for generating diversity Directed evolution guides a mutant pool towards desired phenotypes by associating the phenotype with the fitness of a variant in the iterative cycle of diversity generation, selection, and amplification. A concept of the fitness landscape is generally used to illustrate the effect of a genetic variation to a given phenotype of interest. In the fitness landscape, the horizontal plane represents different genetic variations and the magnitude of the phenotype of interest for each genetic variation is plotted as height. Because the fitness landscape resembles a mountain range, we can interpret the goal of evolutionary engineering as climbing up the fitness landscape to reach a maximum peak where the genetic variation at such a point results in the phenotype of interest at its maximum [73]. To successfully evolve biomolecules, genetic pathways, or even genomes in a directed manner, it is crucial that the evolution system meets three fundamental criteria. (i) The mutant library should be designed to contain sufficient genetic diversity to increase the probability of reaching a maximum peak in the fitness landscape [73]. (ii) Appropriate selection pressure should be applied to eliminate inferior mutants, so that propagation of false-positives or cheater mutants is minimized [86, 102]. (iii) The evolution cycle should be easily iterated for efficient evolution of a desired phenotype. In vitro mutagenesis technologies such as error-prone PCR and DNA shuffling have been broadly utilized for directed evolution [82, 103, 104], as they allow a tunable mutation window in a targeted region. Nevertheless, in vitro mutagenesis is labor-intensive and involves sub-cloning and transformation, which leads to significant loss of the mutant pool because of the limited transformation efficiency [4, 18]. To overcome the drawbacks of in vitro methods and repeat iterative cycles of directed evolution, in vivo approaches have been developed by creating mutagenesis machinery in the cell. However, typical in vivo mutagenesis such as hypermutator strains [29, 30] or mutator plasmids [3, 78] involve the risk of accumulating detrimental mutations in the host genome, because every genetic information in the cell is equally affected by mutagenesis [70]. As random mutagenesis (non-targeted, global) threatens the viability of the host cell and eventually directed evolution, orthogonality, the ability to focus mutations only to a target site (targeted) not to non-target sites (Fig. 2a), has become an integral feature of in vivo mutagenesis while showing a good mutation rate, window, and spectrum. Here, we reviewed several mutagenesis tools (Table 1) according to their strategies to achieve the orthogonality of mutagenesis by (i) implementing an orthogonal cis–trans pair, (ii) detouring the central dogma, (iii) repurposing a pair of programmed guide RNA and CRISPR-associated endonuclease, and (iv) exploiting the infectious cycle of phage. Fig. 2 Open in new tabDownload slide In vivo mutagenesis methods. a Orthogonality of mutagenesis methods. b Targeted mutagenesis based on orthogonal tetO-tetR pair. Mag1-sctetR fusion protein expressed from the plasmid binds tetO array and excises bases on the target locus (upper). In contrast, Mag1-sctetR fusion cannot excise non-target genes located alongside non-specific operators (lower). Mag1, 3-methyladenine glycosylase; sctetR, single chain tetR. c Synthetic retroelement constructed using retrotransposon Ty1 and mutagenic T7 RNA polymerase-retron system. Mutagenesis target is cloned into the genome as a genetic cargo between Ty1RT (Gag-Pol) and 3′-LTR (Long Terminal Repeat). Transcribed mRNA is converted into mutation-containing cDNA as a result of error-prone reverse transcription by reverse transcriptase and re-integrated into the genome. The red box indicates mutation and the black boxes indicate LTRs (upper). Error-prone T7 RNA polymerase synthesizes the transcript of targeting sequence, generating mutations. Then, the transcript is converted into cDNA. Subsequently, msDNA complex comprised of gRNA and error-containing target cDNA edits homologous genomic locus (lower). Orange box and red line indicates reverse transcriptase recognition site and mutation, respectively. d CRISPR-guided mutagenesis tools. Error-prone DNA polymerase I introduces mutations to the target sequence guided by gRNA and nCas9 (left). CRISPR-mediated double-strand break induces homologous recombination with pooled error-prone PCR products (right). Green, purple, and red boxes indicate homologous sequences and mutation, respectively. e Mutator plasmid. Genes located downstream of PBAD promoter are overexpressed upon arabinose induction to disrupt replication maintenance and repair pathways Mutagenesis methods Location . Orthogonality . Name . Machinery . Mutagenesis . Machinery . Mutational range . Host organism . Disadvantage . Reference . In vivo Orthogonal pair ColE1/PolI Orthogonal plasmid/polymerase pair Error-prone DNAPI Replication < 2 kb E. coli Not fully orthogonal Host cell is limited to E. coli [10] OrthoRep Orthogonal plasmid/polymerase pair Error-prone TP-DNAPI 22 kb Yeast S. cerevisiae Difficult genetic manipulation Works only in yeasts [69, 70] Muta T7 Phage derived promoter/polymerase pair Cytidine deaminase Repair Multi-kb Narrow mutation spectrum (C:G → T:A) [58] TaG-TEAM tetO/tetR DNA glycosylase-driven error-prone homologous recombination 20 kb Eukaryotic tetO array should be inserted in genome Mutational range is fixed to 20 kb [26] Compartmentalized PACE Separation of space (phage-assisted) Mutator plasmid MP6 Repair deficient Phage-susceptible Specialized turbidostat is required [3, 23] Detour central dogma ICE Retrotransposon Error-prone reverse transcriptase Reverse transcription 5 kb/cargo Eukaryotes Efficiency is limited by homologous recombination [15] Retron Retron Error-prone RNAP Transcription Within 30 bp (11 bp) Prokaryotes Mutational range is limited [81] CRISPRguided EvolvR gRNA targeted Nickase + error-prone DNAPI Repair 200 bp Off-target effect [31] In vitro CasPER DSB + error-prone PCR Recombination 300–600 bp Unwanted mutation could be introduced by the competition with NHEJ [35] CREATE DSB + oligonucleotide synthesis [28] Location . Orthogonality . Name . Machinery . Mutagenesis . Machinery . Mutational range . Host organism . Disadvantage . Reference . In vivo Orthogonal pair ColE1/PolI Orthogonal plasmid/polymerase pair Error-prone DNAPI Replication < 2 kb E. coli Not fully orthogonal Host cell is limited to E. coli [10] OrthoRep Orthogonal plasmid/polymerase pair Error-prone TP-DNAPI 22 kb Yeast S. cerevisiae Difficult genetic manipulation Works only in yeasts [69, 70] Muta T7 Phage derived promoter/polymerase pair Cytidine deaminase Repair Multi-kb Narrow mutation spectrum (C:G → T:A) [58] TaG-TEAM tetO/tetR DNA glycosylase-driven error-prone homologous recombination 20 kb Eukaryotic tetO array should be inserted in genome Mutational range is fixed to 20 kb [26] Compartmentalized PACE Separation of space (phage-assisted) Mutator plasmid MP6 Repair deficient Phage-susceptible Specialized turbidostat is required [3, 23] Detour central dogma ICE Retrotransposon Error-prone reverse transcriptase Reverse transcription 5 kb/cargo Eukaryotes Efficiency is limited by homologous recombination [15] Retron Retron Error-prone RNAP Transcription Within 30 bp (11 bp) Prokaryotes Mutational range is limited [81] CRISPRguided EvolvR gRNA targeted Nickase + error-prone DNAPI Repair 200 bp Off-target effect [31] In vitro CasPER DSB + error-prone PCR Recombination 300–600 bp Unwanted mutation could be introduced by the competition with NHEJ [35] CREATE DSB + oligonucleotide synthesis [28] Open in new tab Mutagenesis methods Location . Orthogonality . Name . Machinery . Mutagenesis . Machinery . Mutational range . Host organism . Disadvantage . Reference . In vivo Orthogonal pair ColE1/PolI Orthogonal plasmid/polymerase pair Error-prone DNAPI Replication < 2 kb E. coli Not fully orthogonal Host cell is limited to E. coli [10] OrthoRep Orthogonal plasmid/polymerase pair Error-prone TP-DNAPI 22 kb Yeast S. cerevisiae Difficult genetic manipulation Works only in yeasts [69, 70] Muta T7 Phage derived promoter/polymerase pair Cytidine deaminase Repair Multi-kb Narrow mutation spectrum (C:G → T:A) [58] TaG-TEAM tetO/tetR DNA glycosylase-driven error-prone homologous recombination 20 kb Eukaryotic tetO array should be inserted in genome Mutational range is fixed to 20 kb [26] Compartmentalized PACE Separation of space (phage-assisted) Mutator plasmid MP6 Repair deficient Phage-susceptible Specialized turbidostat is required [3, 23] Detour central dogma ICE Retrotransposon Error-prone reverse transcriptase Reverse transcription 5 kb/cargo Eukaryotes Efficiency is limited by homologous recombination [15] Retron Retron Error-prone RNAP Transcription Within 30 bp (11 bp) Prokaryotes Mutational range is limited [81] CRISPRguided EvolvR gRNA targeted Nickase + error-prone DNAPI Repair 200 bp Off-target effect [31] In vitro CasPER DSB + error-prone PCR Recombination 300–600 bp Unwanted mutation could be introduced by the competition with NHEJ [35] CREATE DSB + oligonucleotide synthesis [28] Location . Orthogonality . Name . Machinery . Mutagenesis . Machinery . Mutational range . Host organism . Disadvantage . Reference . In vivo Orthogonal pair ColE1/PolI Orthogonal plasmid/polymerase pair Error-prone DNAPI Replication < 2 kb E. coli Not fully orthogonal Host cell is limited to E. coli [10] OrthoRep Orthogonal plasmid/polymerase pair Error-prone TP-DNAPI 22 kb Yeast S. cerevisiae Difficult genetic manipulation Works only in yeasts [69, 70] Muta T7 Phage derived promoter/polymerase pair Cytidine deaminase Repair Multi-kb Narrow mutation spectrum (C:G → T:A) [58] TaG-TEAM tetO/tetR DNA glycosylase-driven error-prone homologous recombination 20 kb Eukaryotic tetO array should be inserted in genome Mutational range is fixed to 20 kb [26] Compartmentalized PACE Separation of space (phage-assisted) Mutator plasmid MP6 Repair deficient Phage-susceptible Specialized turbidostat is required [3, 23] Detour central dogma ICE Retrotransposon Error-prone reverse transcriptase Reverse transcription 5 kb/cargo Eukaryotes Efficiency is limited by homologous recombination [15] Retron Retron Error-prone RNAP Transcription Within 30 bp (11 bp) Prokaryotes Mutational range is limited [81] CRISPRguided EvolvR gRNA targeted Nickase + error-prone DNAPI Repair 200 bp Off-target effect [31] In vitro CasPER DSB + error-prone PCR Recombination 300–600 bp Unwanted mutation could be introduced by the competition with NHEJ [35] CREATE DSB + oligonucleotide synthesis [28] Open in new tab Orthogonal pair: ColE1/Pol1, OrthoRep, TaG-TEAM, Muta T7 It is pivotal to mutate the target sequence while maintaining the genomic integrity and error-prone capacity of the mutagenesis machinery to sustain directed evolution continuously within the cell. To minimize the interaction of error-generating machinery with unwanted genetic loci, orthogonal replication pairs are employed as in vivo mutagenesis tools. The ColE1 origin found in bacterial plasmid requires initial transient catalyzation of DNA polymerase I (Pol I) for its distinct replication machinery. Given the complete reliance on Pol I, ColE1 provides the cis element for targeted mutation and amplification [10]. Through catalysis by error-prone Pol I encoded in another plasmid, the gene of interest localized downstream of the ColE1 origin can acquire mutations and replicate simultaneously. Although unwanted chromosomal mutation is inevitable because of the contribution of DNA polymerase I to Okazaki fragment joining and DNA repair [46], mutagenesis performed under saturated growth conditions maximizes the gap between the plasmid and chromosomal mutation rate. Despite the even distribution of mutations, the switch from the error-prone Pol I to the endogenous Pol III during the target plasmid replication limits the mutation window to 650 base pairs (bp) downstream, which has a minimal chance of encoding a single gene. Another orthogonal pair of plasmid and polymerase is applied in a yeast-specific mutagenesis tool named as OrthoRep [70]. The linear cytoplasmic plasmid pGKL-1 (p1), derived from Kluyveromyces lactis used in OrthoRep, necessitates dimerization of the terminal protein (TP) anchored at the terminal end and TP-DNAP for initiation of replication [87]. Deduced from the distinctive protein-primed replication machinery and spatial segregation from the chromosome, Ravikumar et al. hypothesized that p1 could be extended as a vector of the target gene for orthogonal mutagenesis and observed an elevated mutation rate on p1 using error-prone TP-DNAP1 with minimal cross-talk between the host chromosome [69]. Recently, they further engineered TP-DNAP1 to reduce fidelity without impairing processivity, which dramatically increased the mutation rate. Error-prone TP-DNAP1 is an indispensable enzyme for sustaining the replication of p1 and leads to the inevitable introduction of mutations on its pair of plasmids, preventing impairment of the trans element of mutagenesis on the nuclear plasmid. The low off-target mutation rate enables the replication over 90 generations with a stable mutation rate, which is an important property for in vivo mutagenesis to facilitate continuous evolution. However, the applicable host being confined to the yeast strain and meticulous genetic manipulation limits the applications of this method. Finney-Manchester and Maheshri constructed targeting glycosylases to embedded arrays for mutagenesis (TaG-TEAM) system based on base excision repair (BER) and the specific interaction between tetR and its cognate operator, tetO (Fig. 2b) [26]. 3-Methyladenine DNA glycosylase derived from Saccharomyces cerevisiae has a broad substrate range and excises alkylated bases in the first step of BER [62]. Apyrimidinic/apurinic (AP) sites caused by glycosylase are further removed from the phosphate backbone through the action of AP endonuclease. However, a previous study reported that overexpressed glycosylase excises normal base pairs in AP endonuclease knock-outs, increasing the spontaneous mutation rate [96, 97]. This is because unprocessed AP sites cause stalling of DNA synthesis machinery and recruit error-prone trans-lesion polymerase [7]. The authors designed a fusion of non-specific yeast glycosylase Mag1 with single-chain tetR (sctetR). The fusion protein bound to one of the tetO arrays near the mutagenesis target and excised the normal bases nonspecifically. As a result, mutations were observed spanning 10 kb bi-directionally from the tetO array. TaG-TEAM provides a fixed mutation range long enough to contain several genes, albeit the mutation rate declines as the distance from the anchored site increases. Moreover, there are tricky requirements; an array of tetO should be embedded in the vicinity of GOI and DNA repair machinery in the host cell must be partially shut down, which increases the risk non-targeted spontaneous mutations. Adopting a similar concept as a fusion protein, the Muta T7 system utilized an orthogonal pair of the T7 promoter and cognate polymerase [58]. As opposed to TaG-TEAM which induces mutations around tethered sites, the processivity of T7 RNA polymerase delivers mutations on a sliding template. Thus, Muta T7 can establish a mutation window using the T7 promoter and tandem terminator [75]. Moore et al. selected cytidine deaminase as a mutator to promote the transition of cytosine to uracil, which is recognized as thymine by DNA polymerase. However, the accessible mutational spectrum configured by amino acid substitution from cytidine deaminase is 32% of the total case, which confers mutational bias [31]. To fill this gap, an additional T7 promoter is introduced in the opposite orientation to broaden the mutation spectrum. Nevertheless, A and T remain intact, with the system compromising the unreachable landscape of the genotype. Detour central dogma: ICE (Ty1 retrotransposon), retroelement A mutagenesis tool implementing reverse transcription (RT) was established to minimize the interaction with the innate cellular machinery of the host that is not equipped with reverse transcriptase. Crook et al. applied the yeast long terminal repeat transposon Ty1 to enable in vivo continuous evolution (ICE) (Fig. 2c, upper) [15]. The target gene inside the retrotransposon cassette as cargo gains mutations in tandem with error-prone transposition. The authors optimized factors such as the cargo expression level, host factors involved in the transposition rate, and induction conditions to increase targeted mutagenesis by over 50-fold compared to the wild type and, therefore, enlarge the library size in each round. Retroelement-based mutagenesis showed a consistent mutation rate across cargo and moderate mutation bias compared to mutagenic polymerases. Interestingly, by comparing two independent in vivo continuous evolution methods, different mutation profiles were found to be induced from the Ty1 cassette conveying xylose isomerase and one xylulokinase variant with different consumption rates as a pair. This supports that the mutation range comprising multiple genes requires a parameter in the mutagenesis tool that drives a novel evolutionary landscape for metabolic engineering. Moreover, by combining the fitness of evolved mutants from ep-PCR and the retrotransposon element, more than half of the latter surpassed the former. Despite the appealing performances of the Ty1 retrotransposon cassette as a mutagenesis tool, the applicable host is limited to yeast strains, which allows Ty1 transposition. RT-based targeted gene editing machinery operating in prokaryotic taxa was reported, followed by the discovery of retrotransposons in eukaryotes [25]. Based on the previous studies, Simon et al. optimized the editing frequency and length for continuous evolution applications (Fig. 2c, lower) [81]. In contrast to Ty1-based mutagenesis, mutagenic T7 RNA polymerase introduce errors on the transcript of multicopy single-stranded DNA containing a target sequence [55]. Error-containing targeting DNA converted by RT edits the genomic sequence driven by homologous recombination. Simon et al. controlled the promoter strength of the retrotransposon cassette and fidelity of T7 RNA polymerase along with the efficiency of host DNA repair machinery to improve the editing frequency. However, the short mutation range and limited number of multiple point mutations as well as damaged repair machinery prevents successful continuous evolution. CRISPR guided: EvolvR, CREATE, CasPER As a navigator of the CRISPR-Cas9 complex, guide RNA designates the target loci based on sequence homology, which confers orthogonality to CRISPR-based mutagenesis tools. A mutagenesis system named as EvolvR utilizes a fusion protein which consists of a nicking variant of Cas9 (nCas9) and error-prone DNA polymerase I (Fig. 2d, left) [31]. nCas9 guided by gRNA forms a nick on targeted loci where coupled DNA polymerase synthesizes a new strand of up to 200 bp with low fidelity and cleaves the displaced strand. Halperin et al. tuned each component to enhance the mutation rate and range and minimize off-target effects. They aimed to alleviate the affinity of nCas9 with DNA to promote mutagenic sliding of Pol I while introducing a triple combination of fidelity-reducing mutations on Pol I. The processivity of Pol I is also modified to expand the mutation window. Finally, the coding sequence of EvolvR is optimized to fit the codon usage and remove innate strong RBS to diminish off-target mutations. In addition, combinatorial mutagenesis within a single target gene and multiplexed targeting on distant genes can use multiple targeting sequences. Considering the average size of the gene, however, the tunable window of EvolvR requires the design and use of several programmed gRNAs simultaneously. Moreover, non-targeted nickase activity also shows a risk of unintended genomic modification. Another CRISPR-based mutagenesis system coupled with in vitro mutagenesis named as CRISPR-enabled trackable genome engineering (CREATE) was introduced by Garst et al. [28]. CREATE employs homologous repair to integrate a mutation pool by generating site-specific DSBs by programmed gRNA and Cas9 in Escherichia coli. As reported previously, a DSB dramatically increases the efficiency of HR by at least two orders of magnitude [76, 83]. They support a web tool for the automated design of a CREATE cassette which serves as a guiding sequence for targeting, template used in HR, and barcode for tracking. However, the toxicity arising from the DSB decreases transformation efficiency, limiting the expandability of library construction and stability of the barcode in plasmids and restricting the tracking of genotype–phenotype relationships. Ronda et al. introduced another CRISPR-based multiplexed genome engineering method called CRISPR/Cas9 and λ Red recombineering-based MAGE (CRMAGE) [74]. This technology utilized two curable plasmids encoding Cas9, λ Red protein, and target-specific sgRNA under the control of inducible promoters. They also contained self-eliminating genetic circuit for subsequent rounds and efficiency-enhancing elements such as RecX and dam methyltransferase that confers transient MutS and RecA- phenotypes. The CRMAGE system showed higher efficiency compared to the traditional MAGE and was provided with a web-based tool to facilitate the design of an experiment. However, the incomplete killing of non-engineered, wild-type cells remains a challenge. Jakočiūnas et al. developed Cas9-mediated Protein Evolution Reaction (CasPER) which enables the integration of large mutant fragments into genomic loci (Fig. 2d, right). CasPER exploits the specific targeting ability of CRISPR and DSB-induced HR and an error-containing target gene generated by five consecutive ep-PCR [35]. Mutator plasmid Badran and Liu developed a mutator plasmid as an inducible episomal vector, which prevents redundant replication maintenance mechanisms such as proofreading, mismatch repair, and base selection (Fig. 2e) [3]. By combining several genes representing the mutator phenotype, plasmids with diverse mutational spectra were constructed. Among them, MP6 showed the highest mutation rate and unbiased mutation spectrum along with a very large dynamic range. The mutagenic performance of the random mutagenic plasmid was shown to be sustainable in phage-assisted continuous evolution in a subsequent study [2]. Although partially blocked DNA repair machinery shows compromised viability, the mutation rate does not exceed the error threshold of essential genes [9, 45]. Tools and systems for screening and selection Directed evolution and evolutionary engineering have enabled the production of numerous strains to form new chemical bonds, such as carbon–silicon bonding that has never been observed in nature [44], improve the solubilities of heterologous proteins [53, 93], and increase tolerance to chemicals [5, 11, 41, 47]. The previous studies using these applications have already been reviewed [14, 63]. Thus, here, we focus on artificial selection and screening for chemical high-producing cells using synthetic genetic devices consisting of a biosensor including transcription factor or RNA-based regulatory elements and selectable marker or fluorescence reporters with fluidic devices. It is important to create genetic diversity and effectively screen variants showing a desired phenotype (typically chemical high-producer) from the generated library in directed evolution and adaptive laboratory evolution [18, 90]. As a promising industrial strain, E. coli can be used with maximum library sizes of up to 1010; this value is limited by transformation efficiency [33, 42, 91]. However, because the phenotypes are typically inconspicuous, it is difficult to effectively identify the improved strain with the desired phenotype [72]. In the conventional screening methods, individual strains are separated, and their productivities are measured directly by high-performance liquid chromatography, gas chromatography, and microtiter plates. However, the analytical methods are less likely to detect meaningful positive mutants in the library, because the number of variants tested per experiment is generally lower than 104, which is time-consuming and labor-intensive for assessing library spaces [51, 52, 97]. Therefore, high-throughput screening methods are required to alleviate the limitations and cover the entire library size to effectively screen for positive mutants [39]. High-throughput screening of strains with phenotypes such as high production of target metabolites has become possible with the introduction of a screening and selection system based on a genetically encoded biosensor including a riboswitch and transcription factor (Fig. 3a, Table 2) [57]. A synthetic genetic device for high-throughput screening consists of a specific upstream ligand-responsive biosensor and downstream expression module [51]. An antibiotic-resistant gene, essential gene, or gene encoding a fluorescent protein is mainly used as an expression module. When a biosensor specifically binds to a cognate ligand, conformational changes occur in response to the intracellular ligand concentration, and expression of the downstream module is regulated at the transcriptional or translational levels [19, 21, 27, 50]. Fig. 3 Open in new tabDownload slide Genetically encoded biosensors and high-throughput screening methods for metabolite-producing microbial cells. a Genetically encoded biosensors. On-type transcription factor (TF)-based biosensor and riboswitch-based biosensor are illustrated. Ligands (particularly, metabolites in this article) bind specifically to their cognate biosensor. Conformational changes occur upon binding of a specific ligand to the ligand-binding domain (LBD) of TF or riboswitch. Both biosensors regulate the expression of ‘Output’ genes depending on the metabolite concentration. Antibiotic resistance gene, essential gene, and fluorescent protein gene are typically used for the downstream genes in the biosensors. b Selection and screening of the chemical producers from a mutant library. Biosensors confer different levels of fitness under the selection pressure or fluorescence to the mutant strains according to the intracellular metabolite concentration. Growth-coupled selection strategy enriches high producers from the mutant population through serial cultures (upper). High-throughput screening analyzes individual cells and collects positive strains using fluorescence-activated cell sorting (FACS) instruments microfluidics devices (not shown in this figure) (lower) Genetically encoded biosensors for chemicals Product . Detector . Output . Function . Reference . Naringenin Transcription factor/cognate promoter pair Expression of TolC Growth-coupled selection, counter selection [68] Glucaric acid Transcription factor/cognate promoter pair Expression of TolC Growth-coupled selection, counter selection [68] Threonine Transcription factor/cognate promoter pair Expression of eGFP beta-Galactosidase Screening of improved production strain [54] 3-Hydroxypropionic acid Transcription factor/cognate promoter pair Expression of TetA Growth-coupled selection [80] Free fatty acid Transcription factor/cognate promoter pair Expression of TetA Growth-coupled selection [97] Tyrosine Transcription factor/cognate promoter pair Expression of TetA Growth-coupled selection [97] Naringenin Riboswitch Expression of GFP Naringenin sensor [36] Lysine Riboswitch Expression of TetA Growth-coupled selection [99] N-acetyl glucosamine Ribozyme Expression of FCY1 Growth-coupled selection [49] Tryptophan Riboswitch Expression of sGFP Representing the intracellular l-tryptophan concentration of each single cell [38] Theophylline Aptazyme Expression of GFP Screening of improved enzyme [56] Product . Detector . Output . Function . Reference . Naringenin Transcription factor/cognate promoter pair Expression of TolC Growth-coupled selection, counter selection [68] Glucaric acid Transcription factor/cognate promoter pair Expression of TolC Growth-coupled selection, counter selection [68] Threonine Transcription factor/cognate promoter pair Expression of eGFP beta-Galactosidase Screening of improved production strain [54] 3-Hydroxypropionic acid Transcription factor/cognate promoter pair Expression of TetA Growth-coupled selection [80] Free fatty acid Transcription factor/cognate promoter pair Expression of TetA Growth-coupled selection [97] Tyrosine Transcription factor/cognate promoter pair Expression of TetA Growth-coupled selection [97] Naringenin Riboswitch Expression of GFP Naringenin sensor [36] Lysine Riboswitch Expression of TetA Growth-coupled selection [99] N-acetyl glucosamine Ribozyme Expression of FCY1 Growth-coupled selection [49] Tryptophan Riboswitch Expression of sGFP Representing the intracellular l-tryptophan concentration of each single cell [38] Theophylline Aptazyme Expression of GFP Screening of improved enzyme [56] Open in new tab Genetically encoded biosensors for chemicals Product . Detector . Output . Function . Reference . Naringenin Transcription factor/cognate promoter pair Expression of TolC Growth-coupled selection, counter selection [68] Glucaric acid Transcription factor/cognate promoter pair Expression of TolC Growth-coupled selection, counter selection [68] Threonine Transcription factor/cognate promoter pair Expression of eGFP beta-Galactosidase Screening of improved production strain [54] 3-Hydroxypropionic acid Transcription factor/cognate promoter pair Expression of TetA Growth-coupled selection [80] Free fatty acid Transcription factor/cognate promoter pair Expression of TetA Growth-coupled selection [97] Tyrosine Transcription factor/cognate promoter pair Expression of TetA Growth-coupled selection [97] Naringenin Riboswitch Expression of GFP Naringenin sensor [36] Lysine Riboswitch Expression of TetA Growth-coupled selection [99] N-acetyl glucosamine Ribozyme Expression of FCY1 Growth-coupled selection [49] Tryptophan Riboswitch Expression of sGFP Representing the intracellular l-tryptophan concentration of each single cell [38] Theophylline Aptazyme Expression of GFP Screening of improved enzyme [56] Product . Detector . Output . Function . Reference . Naringenin Transcription factor/cognate promoter pair Expression of TolC Growth-coupled selection, counter selection [68] Glucaric acid Transcription factor/cognate promoter pair Expression of TolC Growth-coupled selection, counter selection [68] Threonine Transcription factor/cognate promoter pair Expression of eGFP beta-Galactosidase Screening of improved production strain [54] 3-Hydroxypropionic acid Transcription factor/cognate promoter pair Expression of TetA Growth-coupled selection [80] Free fatty acid Transcription factor/cognate promoter pair Expression of TetA Growth-coupled selection [97] Tyrosine Transcription factor/cognate promoter pair Expression of TetA Growth-coupled selection [97] Naringenin Riboswitch Expression of GFP Naringenin sensor [36] Lysine Riboswitch Expression of TetA Growth-coupled selection [99] N-acetyl glucosamine Ribozyme Expression of FCY1 Growth-coupled selection [49] Tryptophan Riboswitch Expression of sGFP Representing the intracellular l-tryptophan concentration of each single cell [38] Theophylline Aptazyme Expression of GFP Screening of improved enzyme [56] Open in new tab First, we summarize recent studies that have used growth-coupled selection to obtain a strain with an improved target metabolite productivity and review recent studies of cell sorting and screening through fluorescence signal detection based on microfluidics in the following section (Fig. 3b). Genetic devices and circuits for artificial screening selection Transcription factor-based selection tools Transcription factors (TFs) have been reported to regulate expression in response to a variety of metabolites including amino acids, flavonoids, and cofactors [60, 68]. Natural TFs-cognate promoter sets were used as detectors, or artificial promoters that responded to the desired metabolites were identified by proteomics analysis [54]. Selection devices for regulating the expression of antibiotics-resistance genes using TFs as a sensor protein that responds to target metabolites have been constructed and applied [24, 77, 98, 101]. A synthetic selection device responding to the 3-hydroxypropionic acid concentration was constructed to control tetA gene expression exploiting 3-HP-responsive-TF, LysR, and the cognate promoter, PC4M. An aldehyde-binding site library of alpha-ketoglutaric semialdehyde dehydrogenase, a 3-HP production pathway enzyme, was constructed and applied for growth-coupled selection. A selected ALDH variant showed a 2.79-fold improvement in the catalytic efficiency and E. coli with an ALDH variant exhibited 25% improved 3-HP productivity compared to the parental strain [80]. Single-cell level quality control devices, population quality control (PopQCs), were developed and applied to prepare strains producing high levels of chemicals as the major population, while the minor population which could not produce sufficient amounts of metabolites was eliminated by antibiotic selection. In this research, the free fatty acid-responsive transcription factor FadR and synthetic promoter PAR and tyrosine-responsive TF TyrR with the tyrosine-activated promoter, PT1 or PT2, were utilized to maintain the quality of the population for a and b production, respectively [97]. By expanding the scope of evolutionary metabolic engineering to the whole biosynthesis process of metabolites, 18 endogenous loci in E. coli were simultaneously mutagenized by multiplex automated genomic engineering within regulatory and coding sequences to construct a mutational library. Expression of the membrane protein TolC was regulated by a natural transcription factor, TtgR, and sensing by the Pttga promoter depends on the naringenin concentration. Toggling positive and negative selection applied to libraries can enrich higher-producing cells and reduce the possibility that cheater cells dominate the population [68]. Modifications of an allosteric transcription factor are based on a computational strategy for altering effector specificity of the protein. In this study, the repressor protein LacI was re-engineered to respond to fucose, gentiobiose, lactitol, and sucralose [85]. In addition to the allostery, the stability of proteins can be used to engineer biosensors. Jester et al. engineered steroid biosensors using dimeric ligand-binding domains, the stability of which can be controlled by the ligand [40]. By fusing each dimeric domain to a DNA-binding domain and a transcriptional activation domain, a full transcriptional activator was able to be reconstituted only in the presence of two ligands that bind and stabilize the dimeric domains. The researchers further engineered the heterodimerization and selectivity of the sensor through directed evolution. Selected mutants were subsequently analyzed by NGS, elucidating the spectrum of mutations that allowed improved sensor performance. Riboswitch-based selection tools The riboswitch is a cis-regulatory RNA element that consists of an aptamer domain that specifically binds small molecules and an expression platform that regulates downstream gene expression by causing a conformational change upon ligand binding to the aptamer [39]. The riboswitches used for directed evolution function mainly at two levels: transcriptional and translational. Forming the rho-independent transcription terminator, which regulates gene expression at the transcription level leading to premature transcription termination and sequestering or releasing the ribosome-binding site (RBS), can control gene expression at the translational level [12]. Likes TFs, numerous RNA-based biosensors are known to exist in nature. Furthermore, natural detector elements were identified by methods such as NGS and Parallel Analysis of RNA Conformations Exposed to Ligand binding (PARCEL) [17, 84]. High-throughput screening of the riboregulator can control transcription termination in response to a specific metabolite concentration by combining the transcription start site sequencing and transcription termination site sequencing method based on NGS [17]. Furthermore, an experimental strategy for identifying RNA aptamers in vitro known as PARCEL in the transcriptomes revealed 58 novel RNA aptamers in both untranslated regions and coding sequences [84]. A natural riboswitch was engineered with specificity for non-natural small molecules and no response to the original ligand [20]. For a suitable RNA-based detector that specifically responds to target biochemicals that do not exist in nature, riboswitch binding to the desired chemical can be artificially prepared using an in vitro selection method known as systematic evolution of ligands by exponential enrichment [36, 37, 100]. The riboswitch-based biosensor was successfully applied for growth-coupled selection [61, 88, 89]. For example, a riboselector as a synthetic RNA device that specifically reacts with l-lysine and l-tryptophan to express the tetracycline resistance gene was constructed. The riboselector conferred a fitness advantage to strains with high performance by expressing tetracycline resistance. Particularly, selection of the ppc promoter library to obtain the high-producing strain of l-lysine resulted in successful enrichment of the top 3 improved strains to 75% of the population in only four rounds of enrichment cycles [99]. In addition to the riboswitch, metabolite-responsive ribozymes have been exploited to select metabolite high producers. For example, a glmS ribozyme, which responds to the glucosamine 6-phosphate concentration, was combined with the 3′-untranslated region of the cytosine deaminase, FCY1, in S. cerevisiae to construct a synthetic suicide riboswitch to screen for an N-acetyl glucosamine-overproducing S. cerevisiae strain. A glutamine–fructose-6-phosphate transaminase mutant library and haloacid dehalogenase-like phosphatases library were efficiently screened with the growth-coupled genetic circuit, resulting in the N-acetyl glucosamine overproducer [49]. High-throughput screening with fluidic systems High-throughput screening, in which positive strains dominate following library selection, has been widely used for directed evolution, as it is possible to exploit selectable markers and appropriate selection pressures without additional equipment. However, cheater cells adapted to resistance under the selection pressure may be selected rather than positive mutants with the desired phenotype, which are not selected. Counterfeit transcription, such as a mutation in a promoter expressing a metabolic sensor or selectable marker in an antibiotic resistance gene, may have growth advantages under the selection pressure regardless of production of the desired metabolite. As a result, cheater cells dominate the population [72]. In addition, to enrich the positive strain, labor-intensive and long-term serial culture is required. Furthermore, artificial selection cannot confirm the target metabolite production ability at the individual cell level in real time [19]. Therefore, genetically encoded biosensors were constructed by connecting expression of the fluorescence markers GFP and RFP to the target metabolite concentration; these biosensors have been applied with various types of analysis equipment such as microfluidics devices and fluorescence-activated cell sorting (FACS) instruments [38, 92]. FACS can be used in combination with genetically encoded biosensors to screen mutant enzymes with increased product selectivity and activity. For example, a caffeine demethylase library containing 106–107 variants was screened in a few hours through FACS using fluorescence from a ribozyme-based biosensor as a readout. A screened variant showed 22- and 33-fold higher selectivity and activity, respectively, compared to the parental strain. Modular assembly of input domain (aptamer) and output domain (ribozyme) of the biosensors can create synthetic RNA switches that will be applied to various metabolic pathways [56]. Furthermore, FACS was also utilized to select variants with increased target metabolite production. For example, FACS was coupled with a biosensor to select threonine high producers [54]. A threonine biosensor was developed by placing a fluorescent protein gene under the control of a threonine-responsive promoter cysJHp. A mutant library containing 2 × 107 mutants was constructed based on an industrial threonine producer strain ThrH(pTZL2). The mutant library was screened using FACS and the threonine biosensor within 1 week and 465 mutants were selected. Forty-four strains sorted from the library showed higher threonine productivities compared to the parental strain, and the highest mutant produced approximately 18% higher levels of threonine [54]. In addition to FACS, a microfluidics system was applied for the screening of high-performance cells. The microfluidic static droplet array can trap a single E. coli cell in a droplet in assigned wells. A tryptophan riboswitch-based biosensor linked to the fluorescence protein was constructed and tryptophan overproduction strains were collected by detecting the output signals with the static droplet array system [38]. Screening that relies on the intracellular concentration of a substance using biosensor neglects the extracellular content. For secreted products, it is difficult to screen for true-positive strains with a system that detects only intracellular concentrations, and obtaining false-positives is very likely. This resulted in a reduced probability of screening of true-positive strains that produce and secrete more product and increased probability of screening of false-positive strains that accumulate more products. To evaluate extracellular concentrations, screening with a droplet compartment is necessary. A single cell or cells required for co-culture were compartmented into a single droplet, and then incubated for fluorescence-based screening to ensure true-positive strain screening including extracellular content. One study conducted library screening of innately fluorescent riboflavin-producing strains. Single-cell FACS and droplet-based FACS screening were conducted to demonstrate that the extracellular concentration represents total production rather than the intracellular concentration. The results showed that screening based on extracellular concentrations was more effective for isolating strains with higher total production [92]. Systems for continuous evolution Natural evolution occurs continuously: genetic diversity is generated, selective pressure is applied, and adapted offspring reproduce. This cycle repeats indefinitely. In contrast, many evolutionary engineering techniques involve frequent intervention at each step of evolution. This requirement for intermittent handling by a researcher limits the efficiency, multiplexity, and scalability of evolutionary engineering. Therefore, researchers have attempted to conduct artificial continuous evolution which operates autonomously with minimal intervention [16]. The most critical issue for establishing continuous evolution would be how to control mutagenesis to generate sufficient genetic diversity at the genetic level while maintaining host cell stability. To address this challenge, there have been several attempts to develop genetic circuits for regulating mutagenesis or mutating only desired genetic loci. First, synthetic genetic circuits enabled controlled mutagenesis in response to the phenotype of interest. In this approach, the expression of mutagenic genes is repressed after achieving the desired phenotype in a cell through the accumulation of sufficient mutations. For example, metabolite-responsive promoters controlled the expression of an error-prone DNA polymerase subunit to evolve mutant strains with improved metabolite productivity such as tyrosine and lycopene [13]. In another example, an engineered riboswitch responsive to changes in pH was utilized to control the expression of an integrase [65]. The pH-dependent expression of the integrase, in turn, controlled the expression of a mutagenic gene. Mutant strains that maintained intracellular pH at a neutral value or in a low pH environment deactivated mutagenic gene expression and stopped further mutagenesis. Next, mutagenesis can be focused to desired genetic loci to minimize any undesired non-specific mutations that may lead to serious growth retardation or even elimination of the host cell. These strategies are explained in Sects. 2.1, 2.2, and 2.3. The target gene(s) was cloned in a specific plasmid or locus with specialized elements in the OrthoRep [70] or using the retrotransposon method [15], respectively. These methods mutated the cloned “cargo” efficiently, but background mutagenesis was maintained at a low level. In contrast, EvolvR [31] enabled targeted mutagenesis at any locus in the genome by designing gRNAs for target genes. However, this method has a relatively short mutagenesis window (~ 350 nucleotides) compared to other methods. Another well-known pioneering strategy is using viral replication to propagate the evolved mutants. The phage-assisted continuous evolution (PACE) [23] system utilized the M13 bacteriophage genome as a vector for target genes. In a continuous bioreactor, E. coli cells are used to mutate the target gene, and mutants that activate the expression of a phage coat protein are packed in an M13 bacteriophage virion. This released virion infects fresh E. coli cells to continue the next round of mutagenesis and selection. Phage-assisted continuous evolution has been utilized to evolve various proteins including TALEN [34], insecticidal toxin [2], aminoacyl-tRNA synthetase [8], RNA polymerase [67], protease [64], antibody [93], and Cas9 [32]. In addition to mutagenesis methods, a continuous culturing apparatus was fabricated to support continuous evolution. Evolutionary engineering requires a continuous culturing system that allows for a multiplexed experiment with highly controlled culture parameters. However, the parameters in batch culture cannot be precisely controlled, while current continuous culture systems are not suitable for parallelized cultures. Recently, a framework for continuous culture known as eVOLVER was reported [94]. This system measures various culture parameters such as optical density or temperature in real time and enables automated control of the parameters. Moreover, a high-throughput evolution study using eVOLVER was conducted for 468 continuous cultures, at a level that cannot be achieved using previous methods. Because of its modularity, small size, and automated nature, multiplexed evolutionary engineering can be performed more easily using this platform. Concluding remarks Evolutionary approaches have enabled engineering of biomolecules and microorganisms for useful applications. Discovery of new biological components and mechanisms have led to the development of advanced technologies for mutagenesis and screening. In this progression, synthetic biology has provided a theoretical and experimental foundation to create synthetic genetic circuits. In addition, microfluidic and millifluidic systems have assisted in cultivation and screening in a high-throughput manner. Furthermore, previously discrete steps in evolutionary engineering can now be merged, allowing for continuous directed evolution without intervention by researchers. Even with this significant progress, some challenges remain. The size of genetic cargo that can be mutated should be larger than the current size limit (~ 20 kb) to evolve large genetic constructs such as biosynthetic gene clusters for secondary metabolites. Universal methods that can be applied to both prokaryotes and eukaryotes are needed for a wide variety of applications. For example, microbial consortium consisting of prokaryotic and eukaryotic cells can be engineered as a whole using such universal method. Finally, evolved genes, pathways, and strains from initial rounds of continuous evolution can be recombined with each other during the evolution process to provide an improved mutant pool for the next round of evolution which will eventually facilitate the identification of superior mutants in a limited timeframe. Acknowledgements This work was supported by the C1 Gas Refinery Program [Grant Number NRF-2018M3D3A1A01055754] and the Global Research Laboratory Program [Grant Number NRF-2016K1A1A2912829] funded by the Ministry of Science and ICT of the Republic of Korea through the National Research Foundation of Korea and by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea [Grant Number 20174030201600]. Compliance with ethical standards Conflict of interest The authors declare no conflict of interests. References 1. Arnold FH Directed evolution: bringing new chemistry to life Angew Chem Int Ed 2018 57 4143 4148 Google Scholar Crossref Search ADS WorldCat 2. Badran AH , Guzov VM, Huai Qet al. Continuous evolution of Bacillus thuringiensis toxins overcomes insect resistance Nature 2016 533 58 63 4865400 Google Scholar Crossref Search ADS PubMed WorldCat 3. Badran AH , Liu DR Development of potent in vivo mutagenesis plasmids with broad mutational spectra Nat Commun 2015 6 8425 4633624 Google Scholar Crossref Search ADS PubMed WorldCat 4. Badran AH , Liu DR In vivo continuous directed evolution Curr Opin Chem Biol 2015 24 1 10 Google Scholar Crossref Search ADS PubMed WorldCat 5. Betteridge AL , Sumby KM, Sundstrom JFet al. Application of directed evolution to develop ethanol tolerant Oenococcus oeni for more efficient malolactic fermentation Appl Microbiol Biotechnol 2018 102 921 932 Google Scholar Crossref Search ADS PubMed WorldCat 6. Bjork SM , Joensson HN Microfluidics for cell factory and bioprocess development Curr Opin Biotechnol 2018 55 95 102 Google Scholar Crossref Search ADS PubMed WorldCat 7. Boiteux S , Guillet M Abasic sites in DNA: repair and biological consequences in Saccharomyces cerevisiae DNA Repair 2004 3 1 12 Google Scholar Crossref Search ADS PubMed WorldCat 8. Bryson DI , Fan C, Guo L-Tet al. Continuous directed evolution of aminoacyl-tRNA synthetases Nat Chem Biol 2017 13 1253 1260 5724969 Google Scholar Crossref Search ADS PubMed WorldCat 9. Bull JJ , Sanjuán R, Wilke CO Theory of lethal mutagenesis for viruses J Virol 2007 81 2930 2939 1865999 Google Scholar Crossref Search ADS PubMed WorldCat 10. Camps M , Naukkarinen J, Johnson BP, Loeb LA Targeted gene evolution in Escherichia coli using a highly error-prone DNA polymerase I Proc Natl Acad Sci USA 2003 100 9727 9732 Google Scholar Crossref Search ADS PubMed WorldCat 11. Carter JLL , Bekhouche M, Noiriel Aet al. Directed evolution of a formate dehydrogenase for increased tolerance to ionic liquids reveals a new site for increasing the stability ChemBioChem 2014 15 2710 2718 Google Scholar Crossref Search ADS PubMed WorldCat 12. Chang AL , Wolf JJ, Smolke CD Synthetic RNA switches as a tool for temporal and spatial control over gene expression Curr Opin Biotechnol 2012 23 679 688 3354030 Google Scholar Crossref Search ADS PubMed WorldCat 13. Chou HH , Keasling JD Programming adaptive control to evolve increased metabolite production Nat Commun 2013 4 2595 Google Scholar Crossref Search ADS PubMed WorldCat 14. Cobb RE , Chao R, Zhao H Directed evolution: past, present and future AIChE J 2013 59 1432 1440 4344831 Google Scholar Crossref Search ADS PubMed WorldCat 15. Crook N , Abatemarco J, Sun Jet al. In vivo continuous evolution of genes and pathways in yeast Nat Commun 2016 7 13051 5071640 Google Scholar Crossref Search ADS PubMed WorldCat 16. d’Oelsnitz S , Ellington A Continuous directed evolution for strain and protein engineering Curr Opin Biotechnol 2018 53 158 163 Google Scholar Crossref Search ADS PubMed WorldCat 17. Dar D , Shamir M, Mellin JRet al. Term-seq reveals abundant ribo-regulation of antibiotics resistance in bacteria Science 2016 352 aad9822 5756622 Google Scholar Crossref Search ADS PubMed WorldCat 18. Dietrich JA , McKee AE, Keasling JD High-throughput metabolic engineering: advances in small-molecule screening and selection Annu Rev Biochem 2010 79 563 590 Google Scholar Crossref Search ADS PubMed WorldCat 19. Dietrich JA , Shis DL, Alikhani A, Keasling JD Transcription factor-based screens and synthetic selections for microbial small-molecule biosynthesis ACS Synth Biol 2013 2 47 58 Google Scholar Crossref Search ADS PubMed WorldCat 20. Dixon N , Duncan JN, Geerlings Tet al. Reengineering orthogonally selective riboswitches Proc Natl Acad Sci USA 2010 107 2830 2835 Google Scholar Crossref Search ADS PubMed WorldCat 21. Dougherty MJ , Arnold FH Directed evolution: new parts and optimized function Curr Opin Biotechnol 2009 20 486 491 2775421 Google Scholar Crossref Search ADS PubMed WorldCat 22. Dragosits M , Mattanovich D Adaptive laboratory evolution—principles and applications for biotechnology Microb Cell Fact 2013 12 64 3716822 Google Scholar Crossref Search ADS PubMed WorldCat 23. Esvelt KM , Carlson JC, Liu DR A system for the continuous directed evolution of biomolecules Nature 2011 472 499 503 3084352 Google Scholar Crossref Search ADS PubMed WorldCat 24. Farmer WR , Liao JC Improving lycopene production in Escherichia coli by engineering metabolic control Nat Biotechnol 2000 18 533 537 Google Scholar Crossref Search ADS PubMed WorldCat 25. Farzadfard F , Lu TK Genomically encoded analog memory with precise in vivo DNA writing in living cell populations Science 2014 346 1256272 4266475 Google Scholar Crossref Search ADS PubMed WorldCat 26. Finney-Manchester SP , Maheshri N Harnessing mutagenic homologous recombination for targeted mutagenesis in vivo by TaGTEAM Nucl Acids Res 2013 41 e99 Google Scholar Crossref Search ADS PubMed WorldCat 27. Frommer WB , Davidson MW, Campbell RE Genetically encoded biosensors based on engineered fluorescent proteins Chem Soc Rev 2009 38 2833 2841 3000468 Google Scholar Crossref Search ADS PubMed WorldCat 28. Garst AD , Bassalo MC, Pines Get al. Genome-wide mapping of mutations at single-nucleotide resolution for protein, metabolic and genome engineering Nat Biotechnol 2017 35 48 55 Google Scholar Crossref Search ADS PubMed WorldCat 29. Glassner BJ , Rasmussen LJ, Najarian MTet al. Generation of a strong mutator phenotype in yeast by imbalanced base excision repair Proc Natl Acad Sci USA 1998 95 9997 10002 Google Scholar Crossref Search ADS PubMed WorldCat 30. Greener A , Callahan M, Jerpseth B An efficient random mutagenesis technique using an E. coli mutator strain Mol Biotechnol 1997 7 189 195 Google Scholar Crossref Search ADS PubMed WorldCat 31. Halperin SO , Tou CJ, Wong EBet al. CRISPR-guided DNA polymerases enable diversification of all nucleotides in a tunable window Nature 2018 560 248 252 Google Scholar Crossref Search ADS PubMed WorldCat 32. Hu JH , Miller SM, Geurts MHet al. Evolved Cas9 variants with broad PAM compatibility and high DNA specificity Nature 2018 556 57 63 5951633 Google Scholar Crossref Search ADS PubMed WorldCat 33. Huang G , Zhong Z, Miersch Set al. Construction of synthetic phage displayed Fab library with tailored diversity J Vis Exp 2018 135 e57357 Google Scholar OpenURL Placeholder Text WorldCat 34. Hubbard BP , Badran AH, Zuris JAet al. Continuous directed evolution of DNA-binding proteins to improve TALEN specificity Nat Methods 2015 12 939 942 4589463 Google Scholar Crossref Search ADS PubMed WorldCat 35. Jakočiūnas T , Pedersen LE, Lis AVet al. CasPER, a method for directed evolution in genomic contexts using mutagenesis and CRISPR/Cas9 Metab Eng 2018 48 288 296 Google Scholar Crossref Search ADS PubMed WorldCat 36. Jang S , Jang S, Xiu Yet al. Development of artificial riboswitches for monitoring of naringenin in vivo ACS Synth Biol 2017 6 2077 2085 Google Scholar Crossref Search ADS PubMed WorldCat 37. Jang S , Jang S, Yang Jet al. RNA-based dynamic genetic controllers: development strategies and applications Curr Opin Biotechnol 2018 53 1 11 Google Scholar Crossref Search ADS PubMed WorldCat 38. Jang S , Lee B, Jeong H-Het al. On-chip analysis, indexing and screening for chemical producing bacteria in a microfluidic static droplet array Lab Chip 2016 16 1909 1916 Google Scholar Crossref Search ADS PubMed WorldCat 39. Jang S , Yang J, Seo SW, Jung GY Riboselector: riboswitch-based synthetic selection device to expedite evolution of metabolite-producing microorganisms Methods Enzymol 2015 550 341 362 Google Scholar Crossref Search ADS PubMed WorldCat 40. Jester BW , Tinberg CE, Rich MSet al. Engineered biosensors from dimeric ligand-binding domains ACS Synth Biol 2018 7 2457 2467 10.1021/acssynbio.8b00242 Google Scholar Crossref Search ADS PubMed WorldCat 41. Johnson TJ , Halfmann C, Zahler JDet al. Increasing the tolerance of filamentous cyanobacteria to next-generation biofuels via directed evolution Algal Res 2016 18 250 256 Google Scholar Crossref Search ADS WorldCat 42. Ju M-S , Min S-W, Lee SMet al. A synthetic library for rapid isolation of humanized single-domain antibodies Biotechnol Bioprocess Eng 2017 22 239 247 Google Scholar Crossref Search ADS WorldCat 43. Kaminski TS , Scheler O, Garstecki P Droplet microfluidics for microbiology: techniques, applications and challenges Lab Chip 2016 16 2168 2187 Google Scholar Crossref Search ADS PubMed WorldCat 44. Kan SBJ , Lewis RD, Chen K, Arnold FH Directed evolution of cytochrome c for carbon-silicon bond formation: bringing silicon to life Science 2016 354 1048 1051 5243118 Google Scholar Crossref Search ADS PubMed WorldCat 45. Kellenberger CA , Sales-Lee J, Pan Yet al. A minimalist biosensor: quantitation of cyclic di-GMP using the conformational change of a riboswitch aptamer RNA Biol 2015 12 1189 1197 4829349 Google Scholar Crossref Search ADS PubMed WorldCat 46. Kingsbury DT , Helinski DR DNA polymerase as a requirement for the maintenance of the bacterial plasmid colicinogenic factor E1 Biochem Biophys Res Commun 1970 41 1538 1544 Google Scholar Crossref Search ADS PubMed WorldCat 47. Korman TP , Sahachartsiri B, Charbonneau DMet al. Dieselzymes: development of a stable and methanol tolerant lipase for biodiesel production by directed evolution Biotechnol Biofuels 2013 6 70 3670234 Google Scholar Crossref Search ADS PubMed WorldCat 48. LaCroix RA , Sandberg TE, O’Brien EJet al. Use of adaptive laboratory evolution to discover key mutations enabling rapid growth of Escherichia coli K-12 MG1655 on glucose minimal medium Appl Environ Microbiol 2015 81 17 30 10.1128/AEM.02246-14 Google Scholar Crossref Search ADS PubMed WorldCat 49. Lee S-W , Oh M-K A synthetic suicide riboswitch for the high-throughput screening of metabolite production in Saccharomyces cerevisiae Metab Eng 2015 28 143 150 Google Scholar Crossref Search ADS PubMed WorldCat 50. Li S , Si T, Wang M, Zhao H Development of a synthetic malonyl-CoA sensor in Saccharomyces cerevisiae for intracellular metabolite monitoring and genetic screening ACS Synth Biol 2015 4 1308 1315 Google Scholar Crossref Search ADS PubMed WorldCat 51. Lim HG , Jang S, Jang Set al. Design and optimization of genetically encoded biosensors for high-throughput screening of chemicals Curr Opin Biotechnol 2018 54 18 25 Google Scholar Crossref Search ADS PubMed WorldCat 52. Lin J-L , Wagner JM, Alper HS Enabling tools for high-throughput detection of metabolites: metabolic engineering and directed evolution applications Biotechnol Adv 2017 35 950 970 Google Scholar Crossref Search ADS PubMed WorldCat 53. Liu J-W , Hadler KS, Schenk G, Ollis D Using directed evolution to improve the solubility of the C-terminal domain of Escherichia coli aminopeptidase P. Implications for metal binding and protein stability FEBS J 2007 274 4742 4751 Google Scholar Crossref Search ADS PubMed WorldCat 54. Liu Y , Li Q, Zheng Pet al. Developing a high-throughput screening method for threonine overproduction based on an artificial promoter Microb Cell Fact 2015 14 121 4546291 Google Scholar Crossref Search ADS PubMed WorldCat 55. Maas WK , Wang C, Lima Tet al. Multicopy single-stranded DNAs with mismatched base pairs are mutagenic in Escherichia coli Mol Microbiol 1994 14 437 441 Google Scholar Crossref Search ADS PubMed WorldCat 56. Michener JK , Smolke CD High-throughput enzyme evolution in Saccharomyces cerevisiae using a synthetic RNA switch Metab Eng 2012 14 306 316 Google Scholar Crossref Search ADS PubMed WorldCat 57. Michener JK , Thodey K, Liang JC, Smolke CD Applications of genetically-encoded biosensors for the construction and control of biosynthetic pathways Metab Eng 2012 14 212 222 Google Scholar Crossref Search ADS PubMed WorldCat 58. Moore CL , Papa LJ, Shoulders MD A processive protein chimera introduces mutations across defined DNA regions in vivo J Am Chem Soc 2018 140 11560 11564 6166643 Google Scholar Crossref Search ADS PubMed WorldCat 59. Mundhada H , Seoane JM, Schneider Ket al. Increased production of l-serine in Escherichia coli through adaptive laboratory evolution Metab Eng 2017 39 141 150 10.1016/j.ymben.2016.11.008 Google Scholar Crossref Search ADS PubMed WorldCat 60. Nguyen NH , Kim J-R, Park S Application of transcription factor-based 3-hydroxypropionic acid biosensor Biotechnol Bioprocess Eng 2018 23 564 572 Google Scholar Crossref Search ADS WorldCat 61. Nomura Y , Yokobayashi Y Reengineering a natural riboswitch by dual genetic selection J Am Chem Soc 2007 129 13814 13815 Google Scholar Crossref Search ADS PubMed WorldCat 62. O’Brien EJ , Utrilla J, Palsson BO Quantification and classification of E. coli proteome utilization and unused protein costs across environments PLoS Comput Biol 2016 12 e1004998 4924638 Google Scholar Crossref Search ADS PubMed WorldCat 63. Packer MS , Liu DR Methods for the directed evolution of proteins Nat Rev Genet 2015 16 379 394 Google Scholar Crossref Search ADS PubMed WorldCat 64. Packer MS , Rees HA, Liu DR Phage-assisted continuous evolution of proteases with altered substrate specificity Nat Commun 2017 8 956 5643515 Google Scholar Crossref Search ADS PubMed WorldCat 65. Pham HL , Wong A, Chua Net al. Engineering a riboswitch-based genetic platform for the self-directed evolution of acid-tolerant phenotypes Nat Commun 2017 8 411 5583362 Google Scholar Crossref Search ADS PubMed WorldCat 66. Portnoy VA , Bezdan D, Zengler K Adaptive laboratory evolution—harnessing the power of biology for metabolic engineering Curr Opin Biotechnol 2011 22 590 594 Google Scholar Crossref Search ADS PubMed WorldCat 67. Pu J , Zinkus-Boltz J, Dickinson BC Evolution of a split RNA polymerase as a versatile biosensor platform Nat Chem Biol 2017 13 432 438 5823606 Google Scholar Crossref Search ADS PubMed WorldCat 68. Raman S , Rogers JK, Taylor ND, Church GM Evolution-guided optimization of biosynthetic pathways Proc Natl Acad Sci USA 2014 111 17803 17808 Google Scholar Crossref Search ADS PubMed WorldCat 69. Ravikumar A , Arrieta A, Liu CC An orthogonal DNA replication system in yeast Nat Chem Biol 2014 10 175 177 Google Scholar Crossref Search ADS PubMed WorldCat 70. Ravikumar A , Arzumanyan GA, Obadi MKAet al. Scalable, continuous evolution of genes at mutation rates above genomic error thresholds Cell 2018 175 1946 1957 Google Scholar Crossref Search ADS PubMed WorldCat 71. Ravn U , Gueneau F, Baerlocher Let al. By-passing in vitro screening—next generation sequencing technologies applied to antibody display and in silico candidate selection Nucl Acids Res 2010 38 e193 Google Scholar Crossref Search ADS PubMed WorldCat 72. Rogers JK , Taylor ND, Church GM Biosensor-based engineering of biosynthetic pathways Curr Opin Biotechnol 2016 42 84 91 Google Scholar Crossref Search ADS PubMed WorldCat 73. Romero PA , Arnold FH Exploring protein fitness landscapes by directed evolution Nat Rev Mol Cell Biol 2009 10 866 876 10.1038/nrm2805 2997618 Google Scholar Crossref Search ADS PubMed WorldCat 74. Ronda C , Pedersen LE, Sommer MOA, Nielsen AT CRMAGE: CRISPR optimized MAGE recombineering Sci Rep 2016 6 19452 4726160 Google Scholar Crossref Search ADS PubMed WorldCat 75. Rong M , He B, McAllister WT, Durbin RK Promoter specificity determinants of T7 RNA polymerase Proc Natl Acad Sci USA 1998 95 515 519 Google Scholar Crossref Search ADS PubMed WorldCat 76. Rouet P , Smih F, Jasin M Introduction of double-strand breaks into the genome of mouse cells by expression of a rare-cutting endonuclease Mol Cell Biol 1994 14 8096 8106 359348 Google Scholar Crossref Search ADS PubMed WorldCat 77. Scalcinati G , Partow S, Siewers Vet al. Combined metabolic engineering of precursor and co-factor supply to increase α-santalene production by Saccharomyces cerevisiae Microb Cell Fact 2012 11 117 3527295 Google Scholar Crossref Search ADS PubMed WorldCat 78. Selifonova O , Valle F, Schellenberger V Rapid evolution of novel traits in microorganisms Appl Environ Microbiol 2001 67 3645 3649 93066 Google Scholar Crossref Search ADS PubMed WorldCat 79. Seo J-H , Min W-K, Lee S-Get al. To the final goal: can we predict and suggest mutations for protein to develop desired phenotype? Biotechnol Bioprocess Eng 2018 23 134 143 Google Scholar Crossref Search ADS WorldCat 80. Seok JY , Yang J, Choi SJet al. Directed evolution of the 3-hydroxypropionic acid production pathway by engineering aldehyde dehydrogenase using a synthetic selection device Metab Eng 2018 47 113 120 Google Scholar Crossref Search ADS PubMed WorldCat 81. Simon AJ , Morrow BR, Ellington AD Retroelement-based genome editing and evolution ACS Synth Biol 2018 7 2600 2611 Google Scholar Crossref Search ADS PubMed WorldCat 82. Stemmer WPC Molecular breeding of genes, pathways and genomes by DNA shuffling Sci World J 2002 2 130 131 Google Scholar Crossref Search ADS WorldCat 83. Storici F , Durham CL, Gordenin DA, Resnick MA Chromosomal site-specific double-strand breaks are efficiently targeted for repair by oligonucleotides in yeast Proc Natl Acad Sci USA 2003 100 14994 14999 Google Scholar Crossref Search ADS PubMed WorldCat 84. Tapsin S , Sun M, Shen Yet al. Genome-wide identification of natural RNA aptamers in prokaryotes and eukaryotes Nat Commun 2018 9 1289 5876405 Google Scholar Crossref Search ADS PubMed WorldCat 85. Taylor ND , Garruss AS, Moretti Ret al. Engineering an allosteric transcription factor to respond to new ligands Nat Methods 2016 13 177 183 Google Scholar Crossref Search ADS PubMed WorldCat 86. Tizei PAG , Csibra E, Torres L, Pinheiro VB Selection platforms for directed evolution in synthetic biology Biochem Soc Trans 2016 44 1165 1175 4984445 Google Scholar Crossref Search ADS PubMed WorldCat 87. Tokunaga M , Wada N, Hishinuma F Expression and identification of immunity determinants on linear DNA killer plasmids pGKLl and pGKL2 in Kluyveromyces lactis Nucl Acids Res 1987 15 1031 1046 Google Scholar Crossref Search ADS PubMed WorldCat 88. Topp S , Gallivan JP Random walks to synthetic riboswitches—a high-throughput selection based on cell motility ChemBioChem 2008 9 210 213 Google Scholar Crossref Search ADS PubMed WorldCat 89. Topp S , Gallivan JP Emerging applications of riboswitches in chemical biology ACS Chem Biol 2010 5 139 148 2811087 Google Scholar Crossref Search ADS PubMed WorldCat 90. Tyo KE , Alper HS, Stephanopoulos GN Expanding the metabolic engineering toolbox: more options to engineer cells Trends Biotechnol 2007 25 132 137 Google Scholar Crossref Search ADS PubMed WorldCat 91. Vaughan TJ , Williams AJ, Pritchard Ket al. Human antibodies with sub-nanomolar affinities isolated from a large non-immunized phage display library Nat Biotechnol 1996 14 309 314 Google Scholar Crossref Search ADS PubMed WorldCat 92. Wagner JM , Liu L, Yuan S-Fet al. A comparative analysis of single cell and droplet-based FACS for improving production phenotypes: riboflavin overproduction in Yarrowia lipolytica Metab Eng 2018 47 346 356 Google Scholar Crossref Search ADS PubMed WorldCat 93. Wang T , Badran AH, Huang TP, Liu DR Continuous directed evolution of proteins with improved soluble expression Nat Chem Biol 2018 14 972 980 6143403 Google Scholar Crossref Search ADS PubMed WorldCat 94. Wong BG , Mancuso CP, Kiriakov Set al. Precise, automated control of conditions for high-throughput growth of yeast and bacteria with eVOLVER Nat Biotechnol 2018 36 614 623 6035058 Google Scholar Crossref Search ADS PubMed WorldCat 95. Wrenbeck EE , Faber MS, Whitehead TA Deep sequencing methods for protein engineering and design Curr Opin Struct Biol 2017 45 36 44 Google Scholar Crossref Search ADS PubMed WorldCat 96. Xiao W , Samson L In vivo evidence for endogenous DNA alkylation damage as a source of spontaneous mutation in eukaryotic cells Proc Natl Acad Sci USA 1993 90 2117 2121 Google Scholar Crossref Search ADS PubMed WorldCat 97. Xiao Y , Bowen CH, Liu D, Zhang F Exploiting nongenetic cell-to-cell variation for enhanced biosynthesis Nat Chem Biol 2016 12 339 344 Google Scholar Crossref Search ADS PubMed WorldCat 98. Xu P , Li L, Zhang Fet al. Improving fatty acids production by engineering dynamic pathway regulation and metabolic control Proc Natl Acad Sci USA 2014 111 11299 11304 Google Scholar Crossref Search ADS PubMed WorldCat 99. Yang J , Seo SW, Jang Set al. Synthetic RNA devices to expedite the evolution of metabolite-producing microbes Nat Commun 2013 4 1413 Google Scholar Crossref Search ADS PubMed WorldCat 100. Yang W , Lai L Computational design of ligand-binding proteins Curr Opin Struct Biol 2017 45 67 73 Google Scholar Crossref Search ADS PubMed WorldCat 101. Zhang F , Carothers JM, Keasling JD Design of a dynamic sensor-regulator system for production of chemicals and fuels derived from fatty acids Nat Biotechnol 2012 30 354 359 Google Scholar Crossref Search ADS PubMed WorldCat 102. Zhao H , Arnold FH Combinatorial protein design: strategies for screening protein libraries Curr Opin Struct Biol 1997 7 480 485 Google Scholar Crossref Search ADS PubMed WorldCat 103. Zhao H , Arnold FH Optimization of DNA shuffling for high fidelity recombination Nucl Acids Res 1997 25 1307 1308 Google Scholar Crossref Search ADS PubMed WorldCat 104. Zhao H , Giver L, Shao Zet al. Molecular evolution by staggered extension process (StEP) in vitro recombination Nat Biotechnol 1998 16 258 261 Google Scholar Crossref Search ADS PubMed WorldCat 105. Zheng X , Xing X-H, Zhang C Targeted mutagenesis: a sniper-like diversity generator in microbial engineering Synth Syst Biotechnol 2017 2 75 86 5636951 Google Scholar Crossref Search ADS PubMed WorldCat Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. © Society for Industrial Microbiology 2019 This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) © Society for Industrial Microbiology 2019
TI - Tools and systems for evolutionary engineering of biomolecules and microorganisms
JO - Journal of Industrial Microbiology and Biotechnology
DO - 10.1007/s10295-019-02191-5
DA - 2019-10-01
UR - https://www.deepdyve.com/lp/oxford-university-press/tools-and-systems-for-evolutionary-engineering-of-biomolecules-and-Pp0AZwvram
SP - 1313
EP - 1326
VL - 46
IS - 9-10
DP - DeepDyve
ER -