Metabolomic Insights into Human Arboviral Infections: Dengue, Chikungunya, and Zika Viruses

Abstract

The global burden of arboviral diseases and the limited success in controlling them calls for innovative methods to understand arbovirus infections. Metabolomics has been applied to detect alterations in host physiology during infection. This approach relies on mass spectrometry or nuclear magnetic resonance spectroscopy to evaluate how perturbations in biological systems alter metabolic pathways, allowing for differentiation of closely related conditions. Because viruses heavily depend on host resources and pathways, they present unique challenges for characterizing metabolic changes. Here, we review the literature on metabolomics of arboviruses and focus on the interpretation of identified molecular features. Metabolomics has revealed biomarkers that differentiate disease states and outcomes, and has shown similarities in metabolic alterations caused by different viruses (e.g., lipid metabolism). Researchers investigating such metabolomic alterations aim to better understand host⁻virus dynamics, identify diagnostically useful molecular features, discern perturbed pathways for therapeutics, and guide further biochemical research. This review focuses on lessons derived from metabolomics studies on samples from arbovirus-infected humans.

Keywords: alphavirus; arbovirus; flavivirus; lipid; metabolism; metabolomics.

Figure 1

Workflow for metabolomic studies. Samples used for metabolomics analyses, most commonly serum or urine, should be carefully collected, stored, and characterized to confirm infection status and group membership. Metabolites are then quenched and extracted for nuclear magnetic resonance (NMR) or mass spectrometry (MS) analyses. Common MS techniques include direct-injection mass spectrometry (DIMS), liquid chromatography–mass spectrometry (LC-MS) and gas chromatography–mass spectrometry (GC-MS). Analytical methods generate data that must be aligned, extracted, corrected, and filtered to obtain molecular features. Metabolite identification and pathway analyses provide insight into the pathways perturbed during a disease state. Statistical analyses, such as principal component analysis (PCA) and partial least squares discriminate analysis (PLS-DA), differentiate and further classify samples and identify potential biomarker candidates. Once pathways and potential biomarkers have been elucidated, secondary verifications, such as targeted analyses, are often performed to confirm and further investigate biological connections, modes of action, or therapeutic potential of identified biomarkers.

Figure 2

Lipid structures. LIPID MAPS naming conventions indicate “carbon atoms”: “double bonds” (location and conformation of double bonds) [80]. Phosphatidylcholine (PC) is composed of two lipid tails esterified to glycerol 3-phosphate with a choline head group. Glycerol’s stereospecific carbons are numbered sn-1, sn-2, and sn-3 [81]. Phosphatidylethanolamine (PE), phosphatidylserine (PS), and phosphatidylinositol (PI) have the same structure, but with alternative head groups. Ceramide (Cer) contains a sphingosine backbone, and a fatty acyl group. Sphingomyelin is Cer with a phosphocholine or phosphoethanolamine head group. The “d” in the name indicates that these have 1,3-dihydroxy sphingosine backbones.

Figure 3

Fatty acid metabolism. Fatty acids are freed from glycerophospholipids by phospholipases producing lysophospholipids (LPL). Fatty acids are then conjugated to coenzyme A (CoA). The fatty acyl is bonded to carnitine for transport to the mitochondrial matrix, where they can undergo β-oxidation generating acetyl-CoA. The acetyl-CoA can enter the tricarboxylic acid (TCA) cycle to generate NADH which will produce adenosine triphosphate (ATP) during oxidative phosphorylation, or the acetyl-CoA can be used to build new fatty acids and glycerophospholipids. The fatty acids can also be used to remodel glycerophospholipids in the Lands’ cycle without being degraded [98].