Drug Target Identification based on the construction of a Genome-scale metabolic model for the human pathogen Candida parapsilosis

Abstract

Candida parapsilosis has seen one of the most significant rises in incidence among pathogenic Candida spp., often taking second place only to C. albicans 1. Addingto this increased incidence is the rise in resistance to first line antifungals and lack ofadequate alternative therapeutics, not only for C. parapsilosis but throughout the genus 2,3. Genome Scale Metabolic Models (GSMMs) have risen as a powerful in silico tool for the understanding of pathogenesis due to their systems view of metabolism and, above all, drug target predictive capacity 46. In this study the first validated GSMM for C. parapsilosis was constructed iDC1003 comprising 1003 genes, 1804 reactions and 1278 metabolites, across four compartments and an intercompartment. In silico growth parameters as well as predicted utilisation of several metabolites as sole carbon or nitrogen sources were experimentally validated, with iDC1003 showing realiably predictive accuracy. Finally, iDC1003 was exploited as a platform for the prediction of 147 essential enzymes in mimicked host conditions, with 56 also predicted as essential in C. albicans and C. glabrata. These promising drug targets include, besides those already used as targets of clinically used antifungals, others that seem to be entirely new and worth further scrutiny. The obtained results strengthen the position of GSMMs as promising platforms for drug target discovery and as guiding tools for designing novel effective antifungal therapies.