(
A) Venn diagrams describing the relationship between differentially expressed genes (DEG), metabolic genes (GO:0008152), and lipid metabolism genes (GO:0006629) in COVID-19 patient sample groups including epithelial cells isolated by bronchoalveolar lavage (lavage) and post-mortem lung biopsies (autopsy), as well as primary small airway epithelial cells (alveoli) and primary bronchial epithelial cells (bronchial) infected with SARS-CoV-2. (
B) Sunburst graphs showing the coverage of composite metabolic terms (
Levy et al., 2016) on general metabolic response induced by SARS-CoV-2 infection. Lipid and mitochondrial metabolism dominate the transcriptional metabolic signature of infection across all four sample groups. (
C) Heat map of metabolic genes (
Figure 1D) across four sample groups. Red and green boxes are up and downregulated by infection, respectively.
# Indicates small sample size. (
D) Metabolic analysis of SARS-CoV-2 and mock-infected primary bronchial epithelial cells confirms a 50% increase (n=6, p<0.001) in lactate production 48 hr post-infection. (
E) The ratio of lactate production to glucose uptake (glycolytic index) in SARS-CoV-2 and mock-infected primary cells. Index increases from 1.0 to 1.7 out of 2.0 indicating a transition to glycolysis (i.e
. Warburg effect: n=6, p<0.01). (
F) Schematic depicting ER stress pathways superimposed with pathway-associated genes. Red and green boxes are up and downregulated by infection, respectively. * marks differentially regulated genes (n=3, FDR <0.05). Red and green arrows schematically note interactions based on the transcriptional response. XBP1S is the IRE1 spliced form of XBP1. (
G) Heat map of ER stress pathway-associated genes (
Figure 1G) across four sample groups. Red and green boxes are up and downregulated by infection, respectively. * p<0.05, ** p<0.01, *** p<0.001 in a two-sided heteroscedastic student’s t-test against control.
# indicates a small sample size. Bar = 20 µm. Error bars indicate S.E.M.