Mathematical modeling of the benzenoid network in Petunia hybrida

Emission of volatiles from plant floral and vegetative organs is important for reproductive success. Petunia flowers rhythmically emit a bouquet of benzenoid and phenylpropanoid compounds. The key benzenoid intermediate benzoic acid (BA) is believed to be synthesized from phenylalanine through the shortening of the propyl side-chain by two carbons. Limited knowledge is available about the enzymes involved in the chain-shortening reaction but it has been hypothesized that this reaction can occur via either a β-oxidative or a non-β-oxidative pathway. In vivo stable isotope labeling in combination with computer-assisted metabolic flux analysis (MFA) of the benzenoid network in petunia flowers revealed that both the β-oxidative and non-β-oxidative pathways contribute to the formation of benzenoids, and suggested that benzylbenzoate is an intermediate between L-Phe and BA. To test this hypothesis, transgenic petunia plants were generated in which the expression of benzyl alcohol/phenylethanol benzoyl transferase (BPBT) was down-regulated. Elimination of benzylbenzoate formation led to a reduction of endogenous BA and emitted methylbenzoate and increased benzylalcohol and benzylaldehyde emission, confirming the contribution of benzylbenzoate to BA formation. The combined approach of isotopic labeling, MFA, and targeted genetic manipulation was extended to develop a kinetic model capable of simulating whole network responses to different concentrations of supplied Phe in petunia flowers and capturing flux redistributions caused by genetic manipulations. A novel method for kinetic modeling of plant metabolism was employed to determine kinetic parameters for benzenoid network reactions. The model’s ability to predict network response to genetic perturbations was validated by comparing simulated and observed redistributions of metabolite pool sizes and labeling data from transgenic petunia flowers. Model-determined kinetic parameters were used for metabolic control analysis, which described the distribution of control among network parameters on flux toward volatile emission as well as the differential regulation surrounding a key branch point at Phe. The modeling approach presented here is widely applicable to other plant metabolic networks. A valid kinetic model can simulate dozens of in silico metabolic engineering strategies for the study of plant metabolic networks and their regulation, which can ultimately save labor and expense and be a powerful predictive tool