Influenza infection rewires energy metabolism and induces browning features in adipose cells and tissues

Figures

• 

  Fig. 1. Influenza alters adipose tissues’ functions. a Body weight loss (% from initial body weight (BW)) of mock-treated and IAV-infected mice, at day 7 post-infection (7 dpi), n = 33 mock-treated animals and n = 46 IAV-infected animals. Individuals values, as well as means ± SD are shown. ****p < 0.0001. b Adipose tissue (AT) mass (g/g BW) of mock-treated and IAV-infected mice at 7 dpi, n = 25 mock-treated animals and n = 28 IAV-infected animals. *p < 0.05, **p < 0.01. c Real-time quantitative PCR (RT-qPCR) analysis of the expression of lipolytic genes (Atgl, Hsl, Mgl) and lipogenic genes (Glut4, Me2, Fasn) in SCAT and EWAT from mock-treated and IAV-infected mice at 7 dpi. For lipolytic genes: n = 5 mock-treated animals and n = 8 IAV-infected animals. For lipogenic genes: n = 5 mock-treated animals and n = 18 IAV-infected animals. RT-qPCR data were normalized to Eef2 housekeeping gene expression and expressed relative to the expression obtained in the samples from mock-treated mice. *p < 0.05, **p < 0.01, ****p < 0.0001, ns = not significant. d Determination of IL-1β, IL-10, IL-6, and leptin levels (by ELISA) in the supernatants of SCAT and EWAT explants from mock-treated and…

• 

  Fig. 2. Influenza induces subcutaneous adipose tissue browning. a Representative images of hematoxylin/eosin staining of SCAT sections from mock-treated and IAV-infected mice at 7 dpi, n = 7 mock-treated animals and n = 7 IAV-infected animals. Brown-like/beige adipocytes are arrowheaded. b Frequency distributions (%) of adipocyte sizes (areas, μm2) in the SCAT from mock-treated and IAV-infected mice, 7 dpi, n = 6 mock-treated animals and n = 6 IAV-infected animals. Individuals values, as well as means ± SD are shown. *p < 0.05, **p < 0.01. c Western blot of UCP1 in lanes loaded with EWAT, lung, and SCAT protein lysates from mock-treated and IAV-infected mice at 7 dpi. GAPDH was used as an internal loading control. The figure is representative of four independent blots covering a total of: SCAT from n = 5 mock-treated animals (SCAT mock), SCAT from n = 35 IAV-infected animals (SCAT IAV), EWAT from n = 5 IAV-infected animals (EWAT IAV), and lungs from n = 5 IAV-infected animals (Lung IAV). Of note, the uncropped version of the blot is shown on Supplementary Fig. 2d. d Quantification of normalized densitometric UCP1/GAPDH ratios (arbitrary units) of all western blots. **p < 0.01, ****p <…

• 

  Fig. 3. Transcriptomics shows fat-depot-specific response to influenza. SCAT and EWAT from mock-treated and IAV-infected mice were treated for transcriptomic analysis at 7 dpi, n = 4 mock-treated animals and n = 4 IAV-infected animals. a Venn diagrams summarizing the number of distinct and overlapping differentially expressed genes (DEGs) in the SCAT and EWAT from IAV-infected mice vs. the SCAT and EWAT from mock-treated mice. Upregulated (UR) and downregulated (DR) DEGs numbers are shown. b The top five canonical pathways identified by Ingenuity Pathway Analysis (IPA) for the genes upregulated in SCAT and EWAT upon infection (IFN = interferon, IRF = interferon regulatory factor, RA = retinoic acid). The corresponding upstream regulators and their target molecules are shown in Supplementary Table 1. c The top five canonical pathways identified by IPA for the genes downregulated in SCAT and EWAT upon infection (SP = superpathway). The corresponding upstream regulators and their target molecules are shown in Supplementary Table 3. d The top five canonical pathways identified by IPA for the genes downregulated upon infection exclusively in SCAT (TCA cycle = tricarboxylic acid cycle). For…

• 

  Fig. 4. Influenza durably alters the host’s energy metabolism. a Blood levels of glucose (10–12 h fasting, mg/dl), insulin (10–12 h fasting, μg/l) and resistin (fed state, ng/ml) in mock-treated and IAV-infected mice at 7 dpi. For glucose: n = 20 mock-treated animals and n = 30 IAV-infected animals. For insulin and resistin: n = 15 mock-treated animals and n = 15 IAV-infected animals. Individuals values, as well as means ± SD are shown. **p < 0.01, ****p < 0.0001. b Blood levels of glucose, insulin and resistin in mock-treated and IAV-infected mice at week 20 post-infection, n = 8 mock-treated animals and n = 6 IAV-infected animals. Individuals values, as well as means ± SD are shown. *p < 0.05, **p < 0.01. c Body weight change (% initial body weight) of mock-treated and IAV-infected mice fed with standard diet (SD) or with high-fat diet (HFD) from 7 dpi (arrow), n = 5 mock-treated SD-fed animals (mock SD), n = 10 IAV-infected SD-fed animals (IAV SD), n = 10 mock-treated HFD-fed animals (mock HFD), and n = 15 IAV-infected HFD-fed animals (IAV HFD). Data are expressed as mean ± SD. ****p < 0.0001. Corresponding area under the curve (AUC) is shown in Supplementary Fig. 4b. d…

• 

  Fig. 5. Viral protein-expressing cells are detected in adipose tissues. a Viral RNA levels (expressed as cycle threshold (Ct) values) in the lungs, SCAT, EWAT, and pancreas at 7 dpi. Lungs and pancreas from n = 10 IAV-infected animals, and SCAT and EWAT from n = 25 IAV-infected animals. The dotted line indicates pancreas’ mean Ct value. Individuals values, as well as means ± SD are shown. ****p < 0.0001. b Viral RNA levels (expressed as Ct values) in the lungs, SCAT, and EWAT from mock-treated and IAV-infected mice, at 0, 2, 4, 7, 14, and 28 dpi, n = 5 mock-treated animals and n = 8 IAV-infected animals per time point. The dotted line indicates day 0 of infection. Data are expressed as mean ± SD. *p < 0.05, **p < 0.01. c Fractionation procedure used to separate stromal vascular fraction (SVF) cells from mature adipocytes, of the adipose tissues. Immune cells (characterized as CD45+ cells) and preadipocytes (characterized as CD45− CD31− CD29+ CD34+ Sca-1+ cells) from the SVF were then sorted by flow cytometry (see gating strategy Supplementary Fig. 5b). d A selection of sorted immune cells and preadipocytes expressing the viral protein hemagglutinin (green) observed by immunofluorescence…

• 

  Fig. 6. Influenza induces brown-like adipogenesis in preadipocytes. a Viral RNA levels (expressed as Ct values) in IAV-infected preadipocytes (3T3-L1 cells) (MOI of 0.01, 0.25, 0.5, 1, or 2) at 6, 24, and 48 h post-infection (hpi), n = 4 biologically independent samples per condition. The dotted line indicates the 6 hpi/0.01 MOI mean Ct value. ****p < 0.0001. b Representative transmission electron microscopy sections of mock-treated and IAV-infected preadipocytes (MOI of 5, 24 hpi), n = 3 biologically independent samples per condition. Classical nuclear (yellow arrowheads) and nucleolar (yellow dashed frame) fingerprints associated with IAV infection are shown. A detailed view of empty viral particles at budding regions is presented. N = nucleus. c Relative mRNA expression (RT-qPCR) of the antiviral innate-immune-related genes Tlr3, RigI, Mda5, Mx1, and Vip in mock-treated and IAV-infected preadipocytes (MOI of 1, 24 hpi). For Tlr3 and Vip: n = 5 biologically independent samples per condition. For RigI, Mda5, and Mx1: n = 7 biologically independent mock-treated preadipocyte samples and n = 18 biologically independent IAV-infected preadipocyte samples. **p < 0.01, ****p < 0.0001, ns = not…

• 

  Fig. 7. Influenza induces metabolic reprogramming of preadipocytes. Murine preadipocytes (3T3-L1 cells) and adipocytes (differentiated from 3T3-L1 cells) were infected at a MOI of 1, and samples were prepared for transcriptomic analysis at 24 hpi, n = 4 biologically independent samples per condition. a Venn diagrams showing the numbers and intersections of the differentially expressed genes (DEGs), either upregulated or downregulated, in IAV-infected vs. mock-treated preadipocytes and adipocytes in the whole-genome expression profiling microarray data set. Top canonical pathways and related upstream regulators are listed in Supplementary Table 4 (genes commonly upregulated upon infection in preadipocytes and adipocytes), Supplementary Table 5 (genes only upregulated upon infection in adipocytes), Supplementary Table 6 (genes only upregulated upon infection in preadipocytes), and Supplementary Table 7 (genes only downregulated upon infection in preadipocytes). b Scheme of the serine-glycine-one-carbon (SGOC) metabolic pathway. Genes coding for cytosolic and mitochondrial SGOC enzymes (in red) that were upregulated in preadipocytes after IAV infection are framed. c Relative mRNA expression…

• 

  Fig. 8. Metabolic reprogramming associated with influenza infection in subcutaneous adipose tissue (in vivo) and in preadipocytes (in vitro). a Transcriptome analysis of the subcutaneous (SCAT) and the visceral (EWAT) adipose tissues from mock-treated and IAV-infected mice revealed activation of the STAT1-dependent-IFN-mediated signaling pathway and the concomitant inhibition of the cholesterol biosynthesis pathways in both SCAT and EWAT. IAV infection was associated with rewired energy metabolism only in the SCAT, as reflected by activation of the sirtuin signaling pathway and inhibition of glycolysis, OXPHOS, and TCA cycle. This was associated with a white-to-brown phenotypic change of the SCAT. b In vitro IAV infection of preadipocytes and adipocytes led to activation of the IFN signaling pathways in both cell types. In preadipocytes only, IAV infection rewired cell metabolism through activation of the SGOC metabolic pathway. This was associated with the induction of a brown-like/beige genetic program in preadipocytes. OXPHOS = oxidative phosphorylation, ETC = electron transport chain, SGOC metabolism = serine-glycine-one-carbon metabolism. Activated and inhibited pathways are…