We have previously generated and tested a kinetic mathematical model for essential oil biosynthesis in peppermint [Rios-Estepa et al., 2008]. The inputs for our peppermint model were developmental profiles of gene expression, enzyme concentration, enzyme activity, metabolite levels, and the numbers of the specialized cells producing essential oils as variables, and kinetic properties of enzymes as static parameters. Our model predicted that, when plants were grown at low light intensities, the branch-point enzyme (+)-pulegone reductase was affected by (+)-menthofuran, a stress-induced side product, which acted as a weak competitive inhibitor of (+)-pulegone, the primary substrate for this enzyme. Further experiments with recombinant (+)-pulegone reductase demonstrated that this model prediction was correct, thus indicating the power of iterative approaches integrating mathematical modeling and experimentation.

Here we present a second generation model to describe changes in essential oil profiles under various environmental conditions and in different transgenic lines. Model adjustments include experimentally determined variations in the number of glandular trichomes (specialized essential oil-producing structures on leaf surfaces), the amount of essential oil stored per trichome, the distribution of glandular trichomes at different stages of leaf development, and modulation of gene expression patterns in transgenic lines with modified essential oil composition. The implications of these findings for transgenic approaches aimed at improving essential oil yield and composition are discussed. [Rios-Estepa et al., 2008. PNAS Feb 26;105(8):2818-23]