Fig 1.
Reciprocal interactions between Acod1 expression and inflammation during influenza A virus infection in mice.
A. Higher pulmonary expression of Acod1, Tnfaip3, and Hmox1 and 2 mRNA in DBA/2J (D2) than in C57BL/6J (B6) mice following IAV infection (2x103 FFU PR8M; reanalysis of published RNAseq data [31]) (n = 3 each strain and time point). *p<0.05; **p<0.01; ***p<0.001 (pairwise t-test with Benjamini-Hochberg multiple testing correction). B. Acod1 mRNA is among the most highly upregulated transcripts in mouse lung within 48 h of IAV infection (RNAseq, n = 3 per time point). C. Lower pulmonary levels of itaconate in IAV-infected Acod1+/- than in Acod1+/+ mice. Itaconate was not detected in the Acod1-/- mice (n = 36 Acod1+/+, 27 Acod1+/-, 4 Acod1-/- mice). *p<0.05; **p<0.01; ***p<0.001 (unpaired t-test). D. Greater weight loss of female Acod1-/- C57BL/6N mice during infection with IAV. Mice had to be euthanized when weight loss exceeded 20%; n = 9 Acod1+/+, 10 Acod1+/-, 9 Acod1-/- mice. *p<0.05; **p<0.01; ***p<0.001 (unpaired t-test) E. Survival data of the same experiment as in C. Median survival after day 7 was significantly lower in Acod1-/- mice (p = 0.018, Mann-Whitney-U test). F. Semi-quantitative histopathological scoring of lungs on days 8/9 (mid), 11 and 14 (late) reveals highest inflammation in Acod1-/- mice; n = 9 Acod1+/+, 10 Acod1+/-, 9 Acod1-/- mice (all female). *p<0.05; **p<0.01; ***p<0.001 (two-way ANOVA, Tukey’s multiple comparisons test) G. H&E stained lung sections showing inflammation severity scores of 0 (left, Acod1+/+ mouse) and 3 (Acod1-/- mouse).
Fig 2.
Differences among human myeloid cells in ACOD1 mRNA induction during IAV infection.
Cells were infected with the indicated IAV strains (MOI = 1) and gene expression determined at the time points post infection (p.i.) indicated on the x-axes. A. Induction of ACOD1 and TNFAIP3 mRNA (RT-qPCR) in M1 and M2 macrophages and PBMCs during infection with IAV (PR8M). B. Induction of ACOD1 and TNFAIP3 (RT-qPCR) in dTHP-1 cells correlates with virulence of three IAV strains (Gi-WT = H1N1(pdm2009) field isolate; G-NS-PR8 = H1N1(pdm2009) reassortant carrying NS segment of PR8; PR8M = H1N1(PR8M). Reference for fold change = uninfected cells at the same point. Mean ±SEM (n = 4). HA = IAV hemagglutinin. *p<0.05; **p<0.01; ***p<0.001 (two-way ANOVA, Tukey’s multiple comparisons test, with reference to M1 macrophages in A, and to Gi-WT in B).
Fig 3.
Effects of itaconate and DI on cellular inflammation due to IAV infection of dTHP-1 cells.
A. Outline of experiment. dTHP-1 cells were treated with itaconate or DI overnight, or left untreated, and then infected with IAV (PR8M, MOI = 1). Dose-response curves of effects of a range of itaconate and DI concentrations on HA and CXCL10 mRNA expression are shown in S3 Fig. B-C. 48 h time course, DI = 0.25 mM (n = 3). mRNA expression (RT-qPCR) of HA (B) and CXCL10 mRNA (C); reference for fold change = uninfected cells at the same time point. *p<0.05; **p<0.01; ***p<0.001 (unpaired t-test). D-I. Global transcriptomic changes (microarray analysis) 12 h p.i. due to itaconate (25 mM) and DI (1 mM) treatment of uninfected and IAV-infected dTHP-1 cells (n = 3). D. Venn diagram showing unique and common differentially expressed genes by itaconate and DI treatment of infected cells. E-I. Volcano plots showing differential mRNA expression in uninfected dTHP-1 cells under itaconate and DI treatment (E-F) and infected dTHP-1 cells with or without itaconate or DI treatment (G-I). The 2–3 most significantly differentially expressed mRNAs are identified by labels. J. Effects of itaconate (25 mM) and DI (1 mM) treatment on levels of inflammation-related polypeptides in dTHP-1 cell supernatants 12 h p.i. Multiplex cytokine/chemokine analysis of supernatants from the experiment shown in D-I. Mean ± SEM (n = 3). *p<0.05; **p<0.01; ***p<0.001 (One-way ANOVA, Tukey’s multiple comparison test).
Fig 4.
Effects of endogenous Acod1 (Irg1) gene and timing of compound administration on anti-inflammatory activity of itaconates.
A-C. Effect of endogenous ACOD1. BMDMs from WT or Acod1-/- C57BL/6N mice were preincubated with DI (0.25 mM) for 6 h, infected with IAV (PR8M, MOI = 1) or stimulated with IFN-γ (15 ng/ml) for 2 h, and then subjected to post-infection/stimulation treatment with DI (0.25 mM) for another 12 h as indicated. Uninfected/unstimulated cells were exposed to 6 h treatment only. Expression of Cxcl10 mRNA was determined by RT-qPCR, using β-actin as internal control. A. Outline of experiment. B. CXCL10 mRNA expression after IAV infection. C. CXCL10 mRNA expression after IFN-γ stimulation. D,E. Effect of timing of compound administration. dTHP1 cells were infected with IAV (PR8M, MOI = 1) and subjected either to pre-incubation (24 h) or post-infection treatment (12 h) with itaconate (10, 25 mM) or 4OI (125 μM). Expression of Cxcl10 mRNA was determined by RT-qPCR, using β-actin as internal control. Reference for fold change = uninfected, untreated 14 h. D. Outline of experiment. E. CXCL10 mRNA expression. Mean ± SEM (n = 3). *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001 (two-way ANOVA with Fisher’s LSD test for multiple comparisons in C,D; one-way ANOVA with Dunnett’s multiple comparison test in C).
Fig 5.
Effect of overexpression of ACOD1 and exogenous treatment with DI on IAV-infected A549 cells.
A. A549 cells were transfected with pCMV6-Hu-ACOD1, pCMV6 empty vector, or mock transfected for 48 h and then infected with IAV (PR8M, MOI = 1) for 48 h (n = 3). Expression of the indicated mRNAs was measured by RT-qPCR, itaconate concentrations by LC-MS/MS. *p<0.05; **p<0.01; ***p<0.001 (2-way ANOVA, Tukey’s multiple comparisons test) B. A549 cells were treated with 0.25 mM DI overnight, infected with IAV for 48 h (n = 3), and expression of the indicated mRNAs measured by RT-qPCR. Dose response curves of the effects of various concentrations of itaconate and DI on cell viability (MTT assay) are shown in S2 Fig. Reference for fold change = uninfected, untreated cells at the same time point. Mean ±SEM (n = 3). *p<0.05; **p<0.01; ***p<0.001 (2-way ANOVA, Sidak’s multiple comparisons test).
Fig 6.
Effects of itaconates on phosphorylation of STAT1 and AKT, ACOD1 mRNA expression, and ROS levels.
Treatments (itaconate, 20 mM; DI 0.5 mM; 4OI 125 μM) were applied as indicated to dTHP1 cells infected with IAV (PR8M, MOI = 1), and analyses performed 12 h p.i. A. Itaconate, DI, and 4OI inhibit STAT1 phosphorylation, but only DI and 4OI inhibit AKT phosphorylation (immunoblot for the indicated targets, using β-actin as internal control). B. 4OI exerts the strongest ACOD1 mRNA reduction (RT-qPCR, n = 3). Reference for fold change = uninfected cells 12 h. C. The three itaconates reduce IAV-induced mitochondrial ROS levels to a similar degree (flow cytometry, n = 3). Mean ±SEM *p<0.05; **p<0.01; ***p<0.001 (1-way ANOVA followed by Tukey’s multiple comparison test).
Fig 7.
Anti-inflammatory effect of itaconate and DI on IAV-infected human lung tissue explants.
Human primary lung tissue from patients with emphysema and pulmonary arterial hypertension was incubated overnight with itaconate (25 mM), DI (1 mM), or medium only and was then infected with IAV (PR8M, 2x105 FFU/ml) for 24 h. A. Viral titers in supernatant (focus forming assay). B-D. CXCL10 mRNA (B), and ISG15 (C) mRNA in tissue (RT-qPCR); IP10 concentration in supernatant (EIA) (D). Mean ± SEM (n = 6 donors, 3 tissue pieces per replicate). Reference for fold change = uninfected 24 h. *p<0.05; **p<0.01; ***p<0.001 (Mann-Whitney U test).
Fig 8.
Anti-inflammatory effects of itaconate, DI, and 4OI on IAV-infected human PBMCs.
PBMCs were isolated from freshly donated human blood and then infected with IAV (PR8M, MOI = 1) for 12 h in the presence or absence of itaconate (10 mM), DI (0.5 mM), or 4OI (25 μM) without preincubation. A. Expression of the indicated mRNAs in PBMCs (RT-qPCR). Reference for fold change = uninfected, untreated 12 h. B. Concentrations of the indicated proteins in PBMCs supernatants (ELISA). Mean ±SEM (7 donors, 3 replicates per donor). *p<0.05; ***p<0.01; ***p<0.001 (1-way ANOVA). C. GO enrichment analysis of effects of itaconate, DI, and 4OI on IAV-infected PBMCs. Microarray analysis of PBMC RNA from four of the donors featured in A-B. RNA was pooled from the three replicates of each donor, resulting in 4 samples per group. GO term analysis was performed on DEGs (p<0.05, FC>|1.5|) and terms with an FDR <0.02 in at least one condition are shown.
Fig 9.
Single-cell transcriptomic responses of PBMCs to IAV infection and DI treatment.
PBMCs were isolated from freshly donated human blood (one donor), and then infected with IAV (PR8M, MOI = 1) for 12 h with or without DI (0.5 mM) in the medium. A,B. UMAP plots identifying 13 cell types (A) that make up the five major cell types (B) studied. The RNA cell markers used are listed in Table 4. C,D. Differential mRNA expression in uninfected, infected and infected/DI-treated PBMCs. E. Differential mRNA expression in monocytes in the indicated paired comparisons. Major transcriptional reprogramming due to IAV infection is evident, which is largely prevented by treatment with DI. F. Differential mRNA expression in CD8+ T cells in the indicated paired comparisons. Reprogramming in the presence of IAV is much less, but a DI treatment effect is evident. G,H. Venn diagrams based on RNAs differentially regulated (FDR <0.05) by both viral infection and DI treatment in each of the 5 major cell types, either irrespective of FC (G) or FC >|2| (H).
Fig 10.
GO enrichment analysis at the single cell level of DI effects on uninfected and IAV-infected PBMCs.
Analysis based on the scRNAseq data shown in Fig 9. Go terms that are enriched/depleted in all cell types by infection and DI treatment are marked with a red asterix in A and C. A. GO enrichment analysis of monocytes. GO terms are listed in reverse alphabetical order from top to bottom, with upright (red) triangles indicating enrichment (upregulation) and downward pointing (blue) triangles depletion (downregulation). B. Venn diagrams illustrating GO terms that are commonly or uniquely regulated (FDR <0.05) in monocytes, CD4 T cells, CD8 T cells, B cells, and NK cells; due to infection alone, DI treatment of infected cells, or DI treatment of uninfected cells. Go terms that are enriched/depleted in all cell types by infection and DI treatment are printed in bold. C. GO term enrichment analysis of CD4 T cells, CD8 T cells, B cells, and NK cells.
Fig 11.
A-C. Itaconates differ greatly in the ability to reduce progeny virions in productive IAV infection. A549 cells were infected with IAV PR8M (MOI = 1) and treated with itaconate (20 mM), DI (0.5 mM) or 4OI (125 μM) as indicated. HA mRNA and viral titers were measured 24 h p.i. A. HA mRNA (RT-qPCR). B,C. A549 cells, titers in culture supernatants (foci-forming assay) 24 h p.i. expressed on log10 (B) or linear (C) scale. D-F. DI treatment increases survival and prevents pulmonary inflammation in the mouse model of IAV infection. 6–8 week-old female C57Bl/6J mice were infected with 1x103 FFU of mouse-adapted IAV (A/PR/8/34) and injected once daily with DI (50 mg/Kg intraperitoneally) for 5 days (n = 9), the first dose being given the day before infection, or PBS (n = 11). Body weight was determined daily and mice were sacrificed 5 d p.i. (PBS mock infection) or 6 d p. i. (IAV infection). D. Survival curve indicating higher survival in the DI-treated group (p = 0.0097, 2-tailed t test). E. Weight loss curves indicating no effect of DI treatment on body weight. F. Pulmonary inflammation scores (H&E stained sections; same score as used in Fig 1) in randomly selected samples (n = 6 per group). DI treatment prevented pulmonary inflammation nearly completely. Representative histological findings are shown in S25 Fig. Viral titers in lung homogenates (7.0x106 vs. 8.9x106 FFU) did not differ significantly between treated and untreated IAV-infected mice (p = 0.8, two-tailed t test, n = 3 randomly selected samples per group).
Table 1.
Comparison of key features of the three itaconates.
Table 2.
Scoring system to assess inflammation in IAV-infected mouse lung.
Table 3.
List of RT-qPCR primers.
Table 4.
mRNA markers used to define PBMC sub-populations in scRNAseq.