Model-Guided Metabolic Engineering of Increased 2-Phenylethanol Production in Plants

2-Phenylethanol (2-PE) is a natural aromatic with properties that make it a candidate oxygenate for petroleum-derived gasoline. In plants, biosynthesis of 2-PE competes with the phenylpropanoid pathway for the common precursor phenylalanine. The phenylpropanoid pathway in plants directs significant carbon flux towards lignin biosynthesis, a major biopolymer in plant cell walls that impedes the process of biofuel production. We therefore propose a genetic engineering strategy at the phenylalanine branch point, whereby a portion of the carbon flux towards lignin biosynthesis is diverted towards the production of an economically valuable product, 2-PE. Transgenic Arabidopsis thaliana were generated that overexpress aromatic aldehyde synthase (AAS) in tandem with tomato phenylacetaldehyde reductase (PAR) introducing a pathway to produce 2-PE. To analyze the competition between lignin and 2-PE biosynthesis, excised stems and leaves were exogenously fed with ¹³C₆-ring labeled Phe, and isotopic enrichment of downstream metabolites were quantified in time-course to calculate fluxes. Combining metabolic flux analysis with measurements from in vitro kinetic assays of pathway enzymes revealed that endogenous Phe limits 2-PE production. This prediction was tested by combining the overexpression of PAR/AAS with: (1) the overexpression of a feedback-insensitive 3-deoxy-D-arabino-heptulosonate 7-phosphate (DAHP) synthase known to have hyper-induced phenylalanine biosynthesis, and (2) with the double mutant pal1 pal2 known to have reduced activity of the competing enzyme, phenylalanine ammonia lyase (PAL). Furthermore, to evaluate the effect of subcellular partitioning on the extent of competition between PAL and AAS, the PAR/AAS tandem overexpression construct was fused to chloroplast transit peptides to localize 2-PE biosynthesis in plastids. The high accumulation of plastidial Phe combined with the lack of competition from cytosolic PAL resulted in significantly elevated 2-PE levels validating the predictions derived from kinetic modeling. Combining kinetic modeling with time-course in vivo metabolomics led to successful rational engineering of 2-PE accumulating plants.