An updated genome-scale metabolic network reconstruction of Pseudomonas aeruginosa PA14 to characterize mucin-driven shifts in bacterial metabolism

Abstract

Mucins are present in mucosal membranes throughout the body and play a key role in the microbe clearance and infection prevention. Understanding the metabolic responses of pathogens to mucins will further enable the development of protective approaches against infections. We update the genome-scale metabolic network reconstruction (GENRE) of one such pathogen, Pseudomonas aeruginosa PA14, through metabolic coverage expansion, format update, extensive annotation addition, and literature-based curation to produce iPau21. We then validate iPau21 through MEMOTE, growth rate, carbon source utilization, and gene essentiality testing to demonstrate its improved quality and predictive capabilities. We then integrate the GENRE with transcriptomic data in order to generate context-specific models of P. aeruginosa metabolism. The contextualized models recapitulated known phenotypes of unaltered growth and a differential utilization of fumarate metabolism, while also revealing an increased utilization of propionate metabolism upon MUC5B exposure. This work serves to validate iPau21 and demonstrate its utility for providing biological insights.

Fig. 1. Characteristics and MEMOTE benchmarking of iPau21.

a Properties of iPau21 as compared to iPau1129. b MEMOTE scores of iPau21, iPau1129, and iML1515, a high-quality reconstruction of E. coli.

Fig. 2. Updated reconstruction of P. aeruginosa enables accurate growth rate, gene essentiality, and carbon source utilization predictions.

a Model doubling time predictions compared to experimental results gathered from literature. Gray bar represents the experimental range. b Model carbon source utilization predictions compared to results gathered from literature.

Fig. 3. Contextualization of updated reconstruction shows shifts in P. aeruginosa metabolism in response to mucins and mucin components.

a NMDS analysis of flux samples (n = 500) from each contextualized model. b Comparison of non-consensus reactions present within models displays subsets of reactions that are shared by groups of contextualized models.

Fig. 4. Network utilization does not correlate with network structure.

The distance between median NMDS coordinates for each pair of networks was calculated as a metric of difference in network utilization, while the Jaccard distance of network reactions for each pair of networks was calculated as a metric of structural difference. Spearman’s correlation shows an insignificant relation between the two metrics (p = 0.92).

Fig. 5. Random forest analysis between ABTGC and MUC5B shows the networks differ most in terms of propionate and fumarate metabolism utilization.

The top seven most discriminating reactions between the two models belong to propionate and fumarate metabolism. MUC5B utilizes these two types of metabolism more highly than the ABTGC model.