Model-guided metabolic engineering of 2-phenylethanol in Arabidopsis

2-Phenylethanol (2-PE) is a natural aromatic compound with properties that make it a potential biological oxygenate for petroleum-derived gasoline. In plants, 2-PE biosynthesis competes with the phenylpropanoid pathway for the common precursor, phenylalanine (Phe). The phenylpropanoid pathway directs significant carbon flux towards the formation of lignin, a major biopolymer in plant cell walls that impedes the process of biofuel production. Prior genetic engineering in Arabidopsis used acetaldehyde synthase (AAS) in tandem with phenylacetaldehyde reductase (PAR) to redirect part of the carbon flux from lignin towards 2-PE production as a value-added product. To identify the bottleneck(s) in 2-PE biosynthesis, we established a baseline by generating transgenic Arabidopsis overexpressing AAS and PAR. Next, fluxes to 2-PE and lignin were calculated based on the time course of isotopic enrichment of downstream metabolites after feeding with ¹³C₆-ring labeled Phe, which revealed a limitation in Phe precursor availability. To increase substrate availability in plants, we tested two independent strategies by crossing lines with the highest AAS/PAR expression to: (1) Arabidopsis overexpressing a feedback-insensitive 3-deoxy-D-arabino-heptulosonate 7-phosphate (DAHP) synthase known to increase Phe production, and (2) the pal1/pal2 double mutant known to reduce the activity of the competing enzyme, phenylalanine ammonia lyase (PAL). Although both strategies increased 2-PE production in both Arabidopsis stem and leaves, the second strategy had a higher impact. To identify additional metabolic targets, we performed a metabolic control analysis, which revealed that the plastidial Phe transporter limits flux towards 2-PE formation. To avoid this transport step, the PAR/AAS tandem construct was fused to a sequence encoding a chloroplast signal peptide to target 2-PE biosynthesis to plastids. The direct availability of Phe to AAS in plastids combined with the lack of competition with cytosolic PAL resulted in significantly elevated 2-PE levels. Thus, integrating metabolic control analysis with experimental validation of model predictions establishes a foundation for the rational engineering of 2-PE in plants.