Lipidomic Analysis Reveals Serum Alteration of Plasmalogens in Patients Infected With ZIKA Virus

Abstract

Zika virus (ZIKV) is an arthropod-borne virus (arbovirus) in the Flavivirus genus of the Flaviviridae family. Since the large outbreaks in French Polynesia in 2013-2014 and in Brazil in 2015, ZIKV has been considered a new public health threat. Similar to other related flavivirus, ZIKV is associated with mild and self-limiting symptoms such as rash, pruritus, prostration, headache, arthralgia, myalgia, conjunctivitis, lower back pain and, when present, a short-term low grade fever. In addition, ZIKV has been implicated in neurological complications such as neonatal microcephaly and Guillain-Barré syndrome in adults. Herein, serum lipidomic analysis was used to identify possible alterations in lipid metabolism triggered by ZIKV infection. Patients who presented virus-like symptoms such as fever, arthralgia, headache, exanthema, myalgia and pruritus were selected as the control group. Our study reveals increased levels of several phosphatidylethanolamine (PE) lipid species in the serum of ZIKV patients, the majority of them plasmenyl-phosphatidylethanolamine (pPE) (or plasmalogens) linked to polyunsaturated fatty acids. Constituting up to 20% of total phospholipids in humans, plasmalogens linked to polyunsaturated fatty acids are particularly enriched in neural membranes of the brain. The biosynthesis of plasmalogens requires functional peroxisomes, which are important sites for viral replication, including ZIKV. Thus, increased levels of plasmalogens in serum of ZIKV infected subjects suggest a link between ZIKV life cycle and peroxisomes. Our data provide important insights into specific host cellular lipids that are likely associated with ZIKV replication and may serve as platform for antiviral strategy against ZIKV.

Keywords: Zika virus; arbovirus; phosphatidylethanolamine; plasmalogens; serum lipidomics.

FIGURE 1

Global lipidomics analysis of serum of ZIKV patients. Number of identified lipid species in ZIKV and control groups by untargeted LC-MS analysis (A) and sparse partial least squares discriminant analysis (sPLS-DA) of ZIKV infection (ZKV) and controls (CTL) (B). 1G-cer, Glycosylceramide; Acyl-OH, Hydroxy-acyl; Cer, Ceramide; FFA, Free fatty acid; LysoPC, Lyso-phosphatidylcholine; oPC, Plasmenyl-phosphatidylcholine; PC, Phosphatidylcholine; PDME, Dimethylethanolamine; PE, Phosphatidylethanolamine; PI, Phosphatidylinositol; pPC, Plasmanyl-phosphatidylcholine; pPE, Plasmenyl-phosphatidylethanolamine; SM, Sphingomyelin; CE, Cholesteryl ester; Ch, Cholesterol; ADG, Alkyldiacylglycerol and TAG, Triacylglycerol.

FIGURE 2

Volcano plot (A) of the lipid molecular species altered in ZIKV patients vs. control group. The log₂ fold change (FC) of ZIKV vs. CTL group was plotted against the -log₁₀ p-value. Statistical significance was evaluated by t-test (p-value < 0.05) and FC threshold was set to higher than 1.5. Dots in red and blue represent significantly altered pPE and PE species, respectively. Total PE and pPE levels in ZIKV-infected patients (B). Data are expressed as mean ± standard error (n = 12 for each group). Comparison between the groups was performed using t-test. A p-value < 0.05 was considered statistically significant. Total PE and pPE levels were calculated as the sum of all lipid species in each of the lipid classes. Concentration values are given in μg lipid/μL of serum.

FIGURE 3

Serum concentration of molecular species of PE and pPE in ZIKV infected (ZKV) and controls (CTL) patients. Concentration values are given in μg lipid/μL of serum and data represented as mean ± standard error (n = 12 for each group). Comparison between the groups was performed using t-test (p-value < 0.05) and significantly altered species are marked with ^∗. Identified lipid species were annotated by a shorthand nomenclature corresponding to the level of detail attainable by the analysis (e.g., PE 16:0/20:4). The symbol “/” denotes constituting fatty acids moieties of the lipid species, and not their sn-1 and sn-2 position on the glycerol-backbone.

FIGURE 4

Receiver operating characteristic (ROC) curves and serum concentration of the top 2 pPE lipid molecular species significantly increased in ZIKV-infected patients. pPE p18:0/22:6 (A) pPE p16:0/22:5 (B) using AUC > 0.9 with a 95% interval confidence for comparing ZIKV patients and control groups. The ROC curve analysis of potential biomarkers was calculated with Youden’s index. ROC analysis was performed without log₁₀ transformation.