Denise Tieman*, Mark Taylor*, Nicolas Schauer†, Alisdair R. Fernie†, Andrew D. Hanson*, and Harry J. Klee*‡
*Department of Horticultural Sciences, University of Florida, P.O. Box 110690, Gainesville, FL 32611-0690; and †Max-Planck Institut fur Molekulare ¨Pﬂanzenphysiologie, Am Muhlenberg 1, 14476 Golm-Potsdam, Germany ¨ Communicated by Hans Janos Kende, Michigan State University, East Lansing, MI, March 27, 2006 (received for review February 3, 2006)
An important phenylalanine-derived volatile compound produced by plants is 2-phenylethanol. It is a major contributor to ﬂavor in many foods, including fresh fruits, such as tomato, and an insectattractingscent in roses and many other ﬂowers. Despite the centrality of 2-phenylethanol to ﬂavor and fragrance, the plant genes responsible for its synthesis have not been identiﬁed. Here, we describe a biosynthetic pathway for 2-phenylethanol and other phenylalanine-derived volatiles in tomato fruits and a small family of decarboxylases (LeAADC1A, LeAADC1B, and LeAADC2) that can mediate that pathway’sﬁrst step. These enzymes each catalyze conversion of phenylalanine to phenethylamine and tyrosine to tyramine. Although tyrosine is the preferred substrate in vitro, phenylalanine levels in tomato fruits far exceed those of tyrosine, indicating that phenylalanine is a physiological substrate. Consistent with this view, overexpression of either LeAADC1A or LeAADC2 in transgenic tomato plants resulted infruits with up to 10-fold increased emissions of the products of the pathway, including 2-phenylacetaldehyde, 2-phenylethanol, and 1-nitro-2phenylethane. Further, antisense reduction of LeAADC2 signiﬁcantly reduced emissions of these volatiles. Besides establishing a biosynthetic route, these results show that it is possible to change phenylalanine-based ﬂavor and aroma volatiles in plants bymanipulating expression of a single gene.
metabolic engineering phenylalanine taste
also probably related to their attractiveness to mammals and other seed dispersers (9). Such multiple roles in defense and reproduction suggest that regulation of their synthesis is likely to be critical to the plant. Despite the importance of 2-phenylacetaldehyde and 2-phenylethanol to flavor and aroma, it is notclear how plants synthesize them. The yeast Saccharomyces cerevisiae produces 2-phenylethanol from phenylalanine via phenylpyruvate and 2-phenylacetaldehyde (10). Deuterium-labeling studies in rose (Rosa damascena Mill.) indicated that there might be as many as four pathways of synthesis (10). In addition to the yeast pathway, Watanabe et al. (11) reported synthesis via a phenethylamine2-phenylacetaldehyde route and a trans-cinnamic acid phenyllactate pathway. Because plants contain many aromatic L-amino acid decarboxylases (AADCs) (12), a pathway that begins with phenylalanine decarboxylation is a reasonable assumption. Here we demonstrate that tomato (Solanum lycopersicum), indeed, uses a pathway whose first step is decarboxylation of phenylalanine to phenethylamine. This reaction iscatalyzed by a set of related AADCs. Overexpression of the corresponding genes in transgenic tomato plants led to accumulation of significantly higher levels of 2-phenylacetaldehyde and 2-phenylethanol as well as the related compounds 2-phenylacetonitrile and 1-nitro-2-phenylethane. Results
The Pathway for Synthesis of 2-Phenylethanol in Tomato Fruits. Because 2-phenylethanol comprises a benzenering with a two-carbon side chain, it may be derived from phenylalanine by removal of the carboxyl and amino groups. If the carboxyl group is lost first, the predicted initial reaction product is phenethylamine, whereas, if the amino group is lost first, the product is phenylpyruvate. To determine the preferred tomato pathway, we examined fruits for the presence of possible intermediates....