Fig. 3

GPX8 regulates lipid metabolism in ccRCC A-B, Heat map (A) and metabolic pathway enrichment analysis (B) of significantly different metabolite levels from untargeted metabolomics comparing GPX8-KO vs. WT Caki1 (n = 5). Red asterisks indicate metabolites related to the glycerophospholipid pathway (A). C-D, Volcano plot (C) and pathway analysis of downregulated genes (KEGG pathway) (D) from RNA-seq data for GPX8-KO vs. WT Caki1. Lipid metabolism-related pathways are in red box (D). Criteria: |Log2(fold change)|≥ 2 with P-value ≤ 0.05. E, GSEA analysis of GPX8 correlation with lipid metabolism pathways: regulation of FAO (from Gene ontology) and lysophospholipid pathway (from Pathway Interaction Database). F, Representative pictures (left) of neutral lipid BODIPY 493/503 staining from WT vs. GPX8-KO Caki1 and shGPX8 786O cells with or without doxycycline (100 ng/mL). Quantitation of the lipid droplet (right) (n = 3): The number of lipid droplets per cell was quantified as detailed in the Methods section. G, Representative pictures of BODIPY 500/510 C1, C12 staining for WT and GPX8-KO Caki1 cells for lipid uptake as measured by fluorescent intensity with flow cytometry (right). H-I, FA de novo synthesis (CH3ω) and triacylglycerol synthesis (esterified-glycerol) from U¹³C-glucose with NMR (H), and bar graphs for their relative levels normalized by total protein level (BCA) comparing WT vs. GPX8-KO Caki1 cells (left) and shGPX8 786O cells with or without doxycycline (100 ng/mL) for 3 days (right) (I). Data presented in panels (F) and (I) are means ± SD (n ≥ 3). P-value was calculated by unpaired t-test