Metabolome-wide association study of peripheral parasitemia in Plasmodium vivax malaria

Abstract

Background: Plasmodium vivax is one of the leading causes of malaria worldwide. Infections with this parasite cause diverse clinical manifestations, and recent studies revealed that infections with P. vivax can result in severe and fatal disease. Despite these facts, biological traits of the host response and parasite metabolism during P. vivax malaria are still largely underexplored. Parasitemia is clearly related to progression and severity of malaria caused by P. falciparum, however the effects of parasitemia during infections with P. vivax are not well understood.

Results: We conducted an exploratory study using a high-resolution metabolomics platform that uncovered significant associations between parasitemia levels and plasma metabolites from 150 patients with P. vivax malaria. Most plasma metabolites were inversely associated with higher levels of parasitemia. Top predicted metabolites are implicated into pathways of heme and lipid metabolism, which include biliverdin, bilirubin, palmitoylcarnitine, stearoylcarnitine, phosphocholine, glycerophosphocholine, oleic acid and omega-carboxy-trinor-leukotriene B4.

Conclusions: The abundance of several plasma metabolites varies according to the levels of parasitemia in patients with P. vivax malaria. Moreover, our data suggest that the host response and/or parasite survival might be affected by metabolites involved in the degradation of heme and metabolism of several lipids. Importantly, these data highlight metabolic pathways that may serve as targets for the development of new antimalarial compounds.

Keywords: Glycerophospholipid; Heme; Host-pathogen interactions; Metabolomics.

Figure 1

Significant associations between plasma metabolites and P. vivax parasitemia. (A) m/z features selected with univariate analysis of variance (ANOVA). (B) m/z features selected with linear regression adjusted by age and gender. Significant features were identified with a FDR < 0.05 and colored in red. Dashed lines represent a P-value of 0.05.

Figure 2

Dynamics of the abundance of metabolite features associated with P. vivax parasitemia. (A) One-way hierarchical clustering based on the intensity of highly significant metabolite features selected by ANOVA (FDR < 0.001, 50 m/z features). (B) One-way hierarchical clustering based on the intensity of highly significant metabolite features selected by liner regression models (FDR < 0.001, 103 m/z features). The yellow to red scale indicates lower to higher intensity levels based on a Z-score.

Figure 3

Distribution of top predicted metabolites in patients with low, moderate and high levels of P. vivax parasitemia. Additional statistics were performed with Kruskal-Wallis followed by Dunn’s pairwise multiple comparisons procedure; mean values, standard deviation (SD) and significance levels are shown (*, P < 0.05 and **, P < 0.01).