Novel enzymes are essential to establish green and sustainable catalysis for environment-friendly, safe, and cost-effective bioprocesses. In the last two decades, well-studied techniques such as directed evolution and gene mining from metagenome libraries have been developed to create new industrial enzymes. These methods commonly require large genetic libraries and effective selection methods for the desired catalytic activities, while the latter mainly burdens the development of economically feasible bioprocesses. In this regard, I challenged to reduce the screening burdens by developing a single-cell assay system that consists of genetically-encoded enzyme sensors and fluorescence-activated cell sorting (FACS). Efficacy of the developed technology was probed by four cases: i), a phenol-responsive biosensor was designed and developed for directed evolution of new phenol-lyase from similar protein scaffold; ii), the phenol biosensor with improved sensitivity was applied for the mining of cellulase gene from metagenome; iii), a cellobiose-responsive biosensor was designed for the screening of exo-type cellulase from various metagenomes; iv), a cellobiose-inducible transcriptional regulator, CelR was characterized by determining crystal structure and studying mutations for further application of biosensors.
To create new phenol-lyase activity, saturation mutagenesis of target sites was carried out using a 40% homologous indole-lyase. All target sites were in proximate to the active site. A phenol-sensitive genetic sensor, namely Genetic Enzyme Screening System (GESS) and FACS-based screening was applied to isolate phenol-lyase positive clones from the library of indole-lyase. GESS is a transcriptional regulator based-genetic circuit using a DmpR which responds to phenolic derivatives and subsequently activate the expression of a reporter, green fluorescent protein. GESS-based screening of libraries resulted in the evolutionary creation of the final variant harboring D137P, F304D, V394L, and I396R mutations. The variant showed a significant level of phenol-lyase activity, while completely losing the original indole-lyase activity.
To extend the applicability of GESS, the wild-type DmpR regulator was replaced with a mutant DmpR (E135K) and the new GESS was characterized with p-nitrophenol (pNP)-mediated transcriptional activation which is not observed in original GESS. After optimizing the screening protocol, a single-cell based high-throughput screening platform responding to pNP was successfully developed and applied to screen cellulase from metagenome library with pNP-substrate, p-nitrophenyl β-D-cellotrioside. After flow cytometry sorting and further CMC (carboxymethyl cellulose) assay, two clones showing cellulolytic activity were successfully isolated from the metagenome library.
To develop a GESS-based high-throughput screening method specific for exo-type cellulase, a whole-cell biosensor detecting cellulolytic activity was created using a cellobiose-inducible transcriptional regulator, CelR from Thermobifida fusca. This new GESS was named as cellobiose-detectible genetic enzyme screening system (CBGESS) that recognizes cellobiose and activate transcription of its downstream gfp reporter gene. The fluorescence intensity of CBGESS was directly proportional to the concentration of cellobiose. The capability for in vivo quantification was demonstrated through analyzing the cellulolytic activity with two cellulosic substrates, CMC and p-nitrophenyl β-D-cellobioside in cellulase-expressing Escherichia coli. In addition, CBGESS easily sensed crystalline cellulolytic activity of commercial Celluclast 1.5L when Avicel was used as a crystalline substrate. Because exo-type cellulases are very difficult to be found through conventional screening methods, CBGESS would be a powerful tool for screening of exo-type cellulases from genomes and environments.
To design a novel GESS by engineering CBGESS, the first structure of CelR in complex with cellobiose was determined for the first time. CelR consists of two domains (N-terminal DNA binding domain and C-terminal regulatory domain) and exists as a homodimer with each cellobiose sandwiched between the interface of N-terminal and C-terminal subdomain of regulatory domain. Leu59 lies on the putative hinge region as “Leucine Lever” that is crucial for transcriptional activation. A distinctive α4-helix of CelR mediates the ligand binding signal to transcriptional activation and contributes to structure stabilization. From mutation studies, we found that Tyr84 and Gln301 in the active site of CelR recognizes cellobiose as ligand.
Overall, these results showed that the genetically-encoded biosensors using transcriptional regulators can be powerful toolkits for enzyme engineering and metagenome mining. Therefore, I believe, my study contributes to provide new solutions for synthetic biology and biotechnology related with new enzyme developments.