Metabolic Model of the Phytophthora infestans-Tomato Interaction Reveals Metabolic Switches during Host Colonization

ABSTRACT The oomycete pathogen Phytophthora infestanscauses potato and tomato late blight, a disease that is a serious threat to agriculture. P. infestans is a hemibiotrophic pathogen, and during infection, it scavenges nutrients from living host cells for its own proliferation. To date, the nutrient flux from host to pathogenduring infection has hardly been studied, and the interlinked metabolismsof the pathogenand host remain poorly understood. Here, we reconstructed an integrated metabolicmodel of P. infestans and tomato (Solanum lycopersicum) by integrating two previously published models for both species. We used this integrated model to simulate metabolicfluxes from host to pathogenand explored the topology of the model to study the dependencies of the metabolismof P. infestans on that of tomato. This showed, for example, that P. infestans, a thiamineauxotroph, depends on certain metabolicreactions of the tomato thiaminebiosynthesis. We also exploited dual-transcriptome data of a time course of a full late blightinfection cycle on tomato leaves and integrated the expression of metabolicenzymes in the model. This revealed profound changes in pathogen-host metabolismduring infection. As infection progresses, P. infestans performs less de novo synthesisof metabolites and scavenges more metabolites from tomato. This integrated metabolicmodel for the P. infestans-tomato interaction provides a framework to integrate data and generate hypotheses about in planta nutrition of P. infestans throughout its infection cycle. IMPORTANCE Late blightdisease caused by the oomycete pathogen Phytophthora infestansleads to extensive yield losses in tomato and potato cultivation worldwide. To effectively control this pathogen, a thorough understanding of the mechanisms shaping the interaction with its hosts is paramount. While considerable work has focused on exploring host defense mechanisms and identifying P. infestans proteins contributing to virulence and pathogenicity, the nutritional strategies of the pathogenare mostly unresolved. Genome-scale metabolicmodels (GEMs) can be used to simulate metabolicfluxes and help in unravelling the complex nature of metabolism. We integrated a GEM of tomato with a GEM of P. infestans to simulate the metabolicfluxes that occur during infection. This yields insights into the nutrients that P. infestans obtains during different phases of the infection cycle and helps in generating hypotheses about nutrition in planta.