Zika virus infection modulates the metabolomic profile of microglial cells

Abstract

Zika virus (ZIKV) is an emerging arbovirus of the Flaviviridae family. Although infection with ZIKV generally leads to mild disease, its recent emergence in the Americas has been associated with an increase in the development of the Guillain-Barré syndrome in adults, as well as with neurological complications, in particular congenital microcephaly, in new-borns. To date, little information is available on neuroinflammation induced by ZIKV, notably in microglial cells in the context of their metabolic activity, a series of chemical transformations that are essential for their growth, reproduction, structural maintenance and environmental responses. Therefore, in the present study we investigated the metabolomic profile of ZIKV-infected microglia. Microglial cells were exposed to ZIKV at different time points and were analyzed by a Liquid Chromatography-High Resolution mass spectrometry-based metabolomic approach. The results show that ZIKV infection in microglia leads to modulation of the expression of numerous metabolites, including lysophospholipids, particulary Lysophosphatidylcholine, and phospholipids such as Phosphatidylcholine, Phosphatidylserine, Ceramide and Sphingomyelin, and carboxylicic acids as Undecanedioic and Dodecanedioic acid. Some of these metabolites are involved in neuronal differentiation, regulation of apoptosis, virion architecture and viral replication. ZIKV infection was associated with concomitant secretion of inflammatory mediators linked with central nervous system inflammation such as IL-6, TNF-α, IL-1β, iNOS and NO. It also resulted in the upregulation of the expression of the gene encoding CX3CR1, a chemokine receptor known to regulate functional synapse plasticity and signaling between microglial cells. These findings highlight an important role for microglia and their metabolites in the process of neuroinflammation that occurs during ZIKV pathogenesis.

Fig 1. Human microglia cells are permissive to ZIKV infection.

Microglia cells were infected with ZIKV (MOI = 1) and expression of viral RNA was measured at different times post-infection by real-time RT-PCR (A). Viral replication was also determined by plaque assay analysis of culture supernatants of ZIKV-infected cells (B). Experiments were performed three times and errors bars represent standard error of the mean. The t-test was employed to analyze the differences between sets of data. *, P < 0.05.

Fig 2. ZIKV induces the production of inflammatory mediators and CX3CR1 in microglial cells.

CHME5 cells were exposed to ZIKV (MOI 1) and mRNA levels were quantified over time by real-time RT-PCR. Results are expressed as the fold induction of transcripts in ZIKV-infected cells relative to those in Mock-infected cells. Data are representative of three independent experiments, each performed in duplicate (error bars represent SEM). The Wilcoxon-Mann-Whitney test was employed to analyze the differences between sets of data. p-values of < 0.05 *; < 0.01**; < 0.001 *** were considered significant.

Fig 3. Analysis of NO in ZIKV-infected microglial cells.

CHME5 cells were exposed to ZIKV (MOI 1) and expression of iNOS and eNOS was determined by real-time RT-PCR (A). Results are expressed as the fold induction of transcripts in ZIKV-infected cells, relative to those in mock-infected cells. Data are representative of four independent experiments, each performed in duplicate (error bars represent SEM). The Wilcoxon-Mann-Whitney test was used to analyze the differences between sets of data. p-values of < 0.05 *; < 0.01**; < 0.001 *** were considered significant. NO expression (green fluorescence) was determined by immunofluorescence in CHME5 cells infected with ZIKV at 48 hpi (B). As a positive control, cells were treated with a combination of TNF-α and IFN-γ. Staining of cell nuclei is shown as blue fluorescence. NO activity in supernatants of mock-and ZIKV-infected cells was measured using Griess reagent (C). Data are presented as the mean ± standard deviation of five independent experiments. p-values of <0.05 were considered significant (*).

Fig 4. Principal component analysis and orthogonal projection to latent structure regression.

(A) PCA score plot of the LC-HRMS dataset scaled to unit variance. (B) OPLS score plot of LC-HRMS dataset with virus titration level as Y input. Circle diameters are scaled according to virus titer.

Fig 5. Pathway enrichment analysis using all identified features from the LCMS dataset.

(1) Glycerophospholipid metabolism; (2) Alanine, aspartate and glutamate metabolism; (3) Arginine and proline metabolism; (4) Linoleic acid metabolism.

Fig 6. Volcano plot using all identified features with VIP score > 1.

The volcano plot was depicted as a log scaled axes of fold change (x-axis) and p-value from impaired t-test (y-axis). Dashed line delineates metabolite with fold change >1.5 and p-value < 0.05. Metabolite IDs were displayed for marked up regulated (right side) or down regulated compounds (left side). tpi: time post infection.

Fig 7. Significative variation of identified metabolites from 6 to 48 hpi expressed in relative mean peak area.

Bars display standard error mean (SEM), stars indicate significative differences from tow way ANOVA and Sidak post hoc test (alpha = 0.05; * < 0.05; ** < 0.01; *** < 0.001). tpi: time post infection.