Use of genome-scale metabolic reconstructions of avian pathogenic Escherichia coli (APEC) phylogroups for the identification of lineage-specific metabolic pathways

Abstract

Avian pathogenic Escherichia coli (APEC) are a genetically diverse pathotype primarily associated with extra-intestinal infections in birds. APEC lineages are predicted to have unique metabolic capabilities contributing to virulence and survival in the host environment. Here, we present a genome-scale metabolic model for the APEC pathotype based on 114 APEC genome sequences and lineage-specific models for the phylogroups B2, C and G based on a representative isolate for each phylogroup. A total of 1,848 metabolic reactions were predicted in the 114 APEC isolates before gap filling and manual correction. Of these, 89% represented core reactions, whilst the 11% accessory reactions were mostly associated with carbon and nitrogen metabolism. Predictions of auxotrophy were confirmed by inactivation of the conditionally essential lysA and the non-essential potE genes. The APEC metabolic model outperformed the E. coli K-12 iJO1366 model in the Biolog Phenotypic Array platform. Sub-models specific to phylogroups B2, C and G predicted differences in the metabolism of 3-hydroxyphenylacetate (3-HPAA), a phenolic acid derived from the flavonoid quercetin, which is commonly added to poultry feed. Two 3-HPAA-associated reactions/genes distinguished APEC phylogroup C from APEC phylogroups B2 and G, and 3-HPAA supported the growth of APEC phylogroup C in minimal media, but not phylogroups B2 and G. In conclusion, we have constructed genome-scale metabolic models for the three major APEC phylogroups B2, C and G and have identified a metabolic pathway distinguishing phylogroup C APEC. This demonstrates the importance of lineage-and pathotype-specific metabolic models when investigating genetically diverse microbial pathogens.