RESUMEN
Rosmarinic acid as a polyphenolic compound has great values in the pharmaceutical, cosmetic, and food industries. To achieve efficient biosynthesis of rosmarinic acid, the major obstacles such as imbalanced metabolic flux among branching pathways and substrate promiscuity of pathway enzymes should be eliminated. Here, a rosmarinic acid producing Saccharomyces cerevisiae strain was constructed by introducing codon optimized d-lactate dehydrogenase gene mutant (OD-LDHY52A), 4-coumarate CoA ligase gene (OPc4CL2), and rosmarinic acid synthase gene (OMoRAS) into a previously constructed caffeic acid hyper-producer. To identify the metabolic bottleneck, the substrate specificity of OPc4CL2 and OMoRAS was figured out by bioconversion experiments and HPLC-MS/MS analysis. Subsequently, the byproducts formation was alleviated by removing prephenate dehydratase and tuning down the expression level of OPc4CL2. The final strain YRA113-15B produced 208 mg/L rosmarinic acid in a shake-flask culture (a 63-fold improvement over the initial strain), which was the highest rosmarinic acid titer by engineered microbial cells reported to date. This work provides a promising platform for fermentative production of rosmarinic acid and offers a strategy to overcome the intrapathway competition.