Designing a whole-cell biosensor applicable for S-adenosyl-l-methionine-dependent methyltransferases

This study was undertaken to develop a high-throughput screening strategy using a whole-cell biosensor to enhance methyl-group transfer, a rate-limiting step influenced by intracellular methyl donor availability and methyltransferase efficiency. An l-homocysteine biosensor was designed based on regulatory protein MetR from Escherichia coli, which rapidly reported intracellular l-homocysteine accumulation resulted from S-adenosyl-l-homocysteine (SAH) formation after methyl-group transfer. Using S-adenosyl-l-methionine (SAM) as a methyl donor, this biosensor was applied to caffeic acid 3-O-methyltransferase derived from Arabidopsis thaliana (AtComT). After several rounds of directed evolution, the modified enzyme achieved a 13.8-fold improvement when converting caffeic acid to ferulic acid. The best mutant exhibited a 5.4-fold improvement in catalytic efficiency. Characterization of beneficial mutants showed that improved O-methyltransferase dimerization greatly contributed to enzyme activity. This finding was verified when we switched and compared the N-termini involved in dimerization across different sources. Finally, with tyrosine as a substrate, the evolved AtComT mutant greatly improved ferulic acid biosynthesis, yielding 3448 mg L^-1 with a conversion rate of 88.8%. These results have important implications for high-efficiency O-methyltransferase design, which will greatly benefit the biosynthesis of a wide range of natural products. In addition, the l-homocysteine biosensor has the potential for widespread applications in evaluating the efficiency of SAM-based methyl transfer.