26Zika virus (ZIKV), an arbovirus of global concern, remodels intracellular membranes to form 27 replication sites. How ZIKV dysregulates lipid networks to allow this, and consequences for 28 disease, is poorly understood. Here, we performed comprehensive lipidomics to create a lipid 29 network map during ZIKV infection. We found that ZIKV significantly alters host lipid 30 composition, with the most striking changes seen within subclasses of sphingolipids. Ectopic 31 expression of ZIKV NS4B protein resulted in similar changes, demonstrating a role for NS4B in 32 modulating sphingolipid pathways. Disruption of sphingolipid biosynthesis in various cell types, 33including human neural progenitor cells, blocked ZIKV infection. Additionally, the sphingolipid 34 ceramide redistributes to ZIKV replication sites and increasing ceramide levels by multiple 35 pathways sensitizes cells to ZIKV infection. Thus, we identify a sphingolipid metabolic network 36 with a critical role in ZIKV replication and show that ceramide flux is a key mediator of ZIKV 37 infection.
Results
112
ZIKV alters the lipid landscape of host cells 113To understand how cellular lipid metabolism is altered by ZIKV infection, we carried out 114 a global lipidomic survey of the model Huh7 human hepatic carcinoma cell line 20 infected for 24 115 or 48 hours with Asian lineage ZIKV strain FSS13025 (Fig. 1a). Lipids extracted from 116 populations of mock and infected cells (n = 5 biological replicates per condition) were analyzed 117 with electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) (Supplementary Data 118 1 and Supplementary Fig. 1a-f). We identified 340 lipid species spanning the phospholipid, 119 sphingolipid, glycerolipid, and sterol classes for which pairwise comparisons of normalized 120 abundance between mock and infected cells could be made (Supplementary Data 1). Of these, 121 80 species (23.5%) showed significant changes in abundance by 24 hours post-infection (hpi) 122 and 172 species (50.6%) were significantly altered by 48 hpi (P < 0.05, ANOVA or g-test) 123( Supplementary Data 1). Principal component analysis (PCA) of these observations confirmed 124 that infection status (mock or ZIKV) and timepoint (24 or 48 hpi) accounted most of these 125 changes, with changes in lipid composition between mock and infected cells increasing over 126 time (Fig. 1b). 127Next, we examined how ZIKV-induced changes in host lipid composition broke down by 128 subclass and species (Fig. 1c, d). A map of the pairwise correlations of all 340 species at 48 hpi 129 ( Supplementary Fig. 2a) revealed that lipid subclasses largely fell into two groups of species 130 that were either enriched or depleted in abundance ( Supplementary Fig. 2b), suggesting that 131 individual metabolic pathways are up-or down-regulated to create a specific lipid milieu around 132 the events of the viral replication cycle. Supporting this, many of the trends we observed were 133 consistent with earlier reports of functional roles for lipids during flavivirus infection. In ...