Influential parameters for the analysis of intracellular parasite metabolomics

Metabolomics is increasingly popular for the study of many pathogens. For the malaria parasite, Plasmodium falciparum, both targeted and untargeted metabolite detection has improved our understanding of pathogenesis, host-parasite interactions, and antimalarial drug treatment and resistance. However, purification and analysis procedures for performing metabolomics on intracellular pathogens have not been explored. Here, we investigate the impact of host contamination on the metabolome when preparing samples using standard methods. We purified in vitro grown ring stage intra-erythrocytic P. falciparum parasites for untargeted metabolomics studies; the small size of this developmental stage amplifies the challenges associated with metabolomics studies as the ratio between host and parasite biomass is maximized. Following metabolite identification and data preprocessing, we investigated whether host contributions could be corrected post hoc using various normalization approaches (including double stranded DNA, total protein, or parasite number). We conclude that normalization parameters have large effects on differential abundance analysis and recommend the thoughtful selection of these parameters. However, normalization does not remove the contribution from the parasites extracellular environment (culture media and host erythrocyte). In fact, we found that extra-parasite material is as influential on the metabolome as treatment with a potent antimalarial drug with known metabolic effects (artemisinin). Because of this influence, we could not detect significant changes associated with drug treatment. Instead, we identified metabolites predictive of host and media contamination that can be used to assess sample purification. Our findings provide a basis for development of improved experimental and analytical methods for future metabolomics studies of intracellular organisms.