Figures & data
Figure 1. Heterogeneity of mouse influenza resistance is closely associated with gut metabolites.
(a)Experimental setup for IMT experiments. Cecal contents were collected 5 d after GX infection (n = 30); 20 mice were divided into high (n = 10, TH) and low virus titer (n = 10, TL) groups based on the level of viral load in the lungs. The water-soluble metabolites were extracted and transplanted to SPF mice pretreated with an antibiotic solution for 3 d. After 5 d, the mice were infected with 1 × 104 EID50 GX. (b) Viral RNA transcript counts in the lungs of infected SPF mice at 5 dpi. (c) Viral titer in the lungs of infected SPF mice at 5 dpi. (d) survival and (e) weight change curves for the TH donor, TL donor, and NC donor (NC, mock infection using PBS) group after metabolite transplantations (n = 10 per group). (f) Representative images of an H&E staining mouse lung. In another repeat metabolite transplantation experiment, the mice were sacrificed at 7 dpi and the lungs were collected. Survival was assessed with a log-rank (Mantel-Cox) test. (g) Blinded sections were assessed to determine the level of pathological severity. In order to evaluate the overall histological changes, lung tissue sections were scored according to the criteria specified in the panel. The scoring system used was as follows: 0, no pathological change; 1, affected area (≤10%); 2, affected area (<50%, >10%); 3, affected area (≥50%). Weight changes were assessed using a two-way ANOVA. *P < .05, ***P <.001 (TL donor vs NC donor); #P < .05, ##P < .01, and ##P < .001 (TL donor vs TH donor). Error bars show the mean ± SD for each treatment. All experiments were performed at least twice under similar conditions.
Figure 2. The influenza resistance of infected mice is associated with the enrichment of several gut metabolites.
(a) Protocols for Metabolome Sequencing. (b) The PCA of the sample in NC, TH, TL, and QC group (c) OPLS-DA showing differences in metabolite composition between the TH, TL, and NC groups. (d) Volcano plot comparison of NC vs TH, NC vs TL, and TH vs TL. Each point in the volcano plot represents a metabolite, the abscissa represents the fold change of each substance in the group compared to each other (taking the logarithm with base 2), and the ordinate represents the P-value of the Student’s t-test (taking the pair with base 10). The size of the scatter point corresponds with the VIP value of the OPLS-DA model. Significantly upregulated and downregulated metabolites are presented in red and blue, respectively. (e) Hierarchical clustering analysis of TH vs TL. The abscissa represents different experimental groups, the ordinate represents the differential metabolites compared in the group, the color blocks represent the relative expression levels of the metabolites at the corresponding positions, yellow indicates high expression levels, and blue indicates low expression levels. (f) Comparison of the relative content of each metabolite between the TH, TL, and NC groups. Data are presented as a minimum, first quartile, median, third quartile, and maximum. *P < .05, **P < .001.
Figure 3. GlcNAc and adenosine exert anti-influenza effects in vivo and in vitro, respectively.
Caco-2 cells were pre-incubated with different concentrations of metabolites for 12 h, infected with the GX at an MOI of 0.1, and treated with the corresponding concentrations of metabolites. Cell supernatant samples were harvested 12, 24, and 36 h after infection. (a) The nucleoprotein (NP) level was measured in Caco-2 cells treated with different concentrations of metabolites. (b) The effect of metabolites on GX vRNA abundance at 12, 24, and 36 hpi. (c) The virus titer in the cell supernatant were detected by TCID50. (d) Workflow of the mouse experiment. SPF mice received daily oral administration for 1 week with GlcNAc, adenosine, L-arginine, cellobiose, maltose, citraconic acid, malonic acid or Glu-Pro, or the same volume of PBS (200 μL). All mice were intranasally inoculated with 1 × 104 EID50 of H7N9 influenza virus (n = 10 per group). (e) Survival and (f) body weights of the remaining mice per group were monitored daily for 14 d. Statistics for vRNA abundance, virus titer, and weight changes were two-way ANOVA. Survival was assessed with the log-rank (Mantel-Cox) test. #P < .05 (Control vs 0.468 mM); +++P < .001, ++++P < .0001 (Control vs 0.935 mM); *P < .05, **P < .01; ***P < .001 (Control vs 1.87 mM). Error bars show the mean ± SD for each treatment
Figure 4. Oral administration of GlcNAc protects against H7N9 infection in recipient mice.
(a) Histological examination (H&E staining) and (b) immunohistochemistry of mice orally administrated 1000 mg/kg GlcNAc or PBS (200 μL) daily for 1 week and infected with H7N9 influenza virus GX. At 0, 3, and 5 dpi, three randomly selected mice per group were sacrificed. (c)Blinded sections were scored for levels of pathological severity. To evaluate comprehensive histological changes, lung tissue sections were scored based on criteria indicated in the panel. The following scoring system was used: 0, no pathological change; 1, affected area (≤10%); 2, affected area (<50%, >10%); 3, affected area (≥50%). (d) Quantitative analysis of the fluorescence intensity of NP in lung. The fluorescence intensity was calculated by the mean fluorescence intensity. In addition, 30 SPF mice were randomly divided into two groups and treated as described above. Lungs were collected from five randomly selected mice per group at 0, 3, and 5 dpi, and homogenized in PBS (1 mL/lung). (e) NP mRNA levels were measured by qRT-PCR. (f) Virus titers were measured using 10-d-old SPF embryonated chicken eggs. (g) Cytokine concentrations Weight changes, virus titers, and cytokine concentrations were assessed using a two-way ANOVA. *P < .05, ***P < .001, ****P < .0001. Error bars show the mean ± SD for each treatment.
Figure 5. Oral administration of GlcNAc enhances NK cell responses before and after infection in mice.
SPF mice received daily oral administration for 1 week with GlcNAc (500 mg/kg) or PBS (200 μL). All mice were intranasally inoculated with 1 × 104 EID50 of GX influenza virus. Lymphocytes from local draining mediastinal peripheral blood (a, b), lung (c, d), or spleen (e, f) were analyzed by flow cytometry. (a) Representative flow cytometry plots with gating strategy and summary graphs showing the proportion of NK cells in lymphocytes and total number of NK cell in peripheral blood post infection. (b) Frequency of IFN-γ and CD107a surface expression in gated CD3-CD49B+ NK cells. (c) As in (a), but showing the lymphocytes from the lung. (d) As in (b), but showing the NK cells from the lung. (e) As in (c), but showing the lymphocytes from the spleen. (d) As in (f), but showing the NK cells from the spleen. The proportion and number of NK cell and frequency of IFN-γ and CD107a surface expression in gated CD3-CD49b+. Flow cytometry data were assessed using a two-way ANOVA. *P < .05, **P < .01. Error bars show the mean ± SD for each treatment.
Figure 6. GlcNAc improves host defense against H7N9 infection by regulating NK cells.
The PBS and GlcNAc groups (n = 20 per group) were orally administered PBS or 1000 mg/kg GlcNAc for 1 week and infected with H7N9 influenza virus GX. At 0, 1, 3, and 5 dpi, five mice were randomly selected from each group and euthanized. NK cells were isolated from peripheral blood(a), lung(b), and spleen(c) and their activities were measured by a calcein-release-assay. (d) Experimental setup for NK-cell adoptive transfer. SPF mice were divided into PBS and GlcNAc groups and treated as described above. NK cells were purified from the lungs of mice in the two groups and then 200 µL PBS containing 1 × 106 NK cells were injected i.v. into recipient mice (n = 10 per group) via the tail vein. The recipient mice in the control group were injected with an equivalent volume of PBS. All recipient mice were intranasally inoculated with 1 × 104 EID50 H7N9 influenza virus. (e) Survival rate and (f) body weight of infected mice. SPF mice were divided into PBS and GlcNAc groups and treated as described above. (g) After treatment with PBS or anti-asialo GM1 antiserum, NK cells in peripheral blood and the lung were examined by flow cytometry after one day. On day 0, mice were infected with 1 × 104 EID50 GX influenza virus. (h) Survival and (i) body weight of mice per group were monitored daily for 14 d. Survival was assessed with a log-rank (Mantel-Cox) test. NK cell activity assay and body weight was assessed using a two-way ANOVA. Error bars show the mean ± SD for each treatment. All experiments were performed at least twice under similar conditions.
Figure 7. Upregulation of specific microbiota increased GlcNAc production in the gut.
Cecal content was sampled for the TH and TL groups for metagenomic analysis. (a) PCoA of the weighted UniFrac distances between the TH and TL group at the species level. The relative abundance of the (b) top 10 phyla and (c) top 30 genera in the TH and TL groups. (d) Hierarchically clustered heatmap analysis at the genus level; the relative values for bacterial genera correspond with color intensity. (e) The taxa with different abundances between the TH (blue bars) and TL group (yellow bars). The length of the histogram represents the influence size of the different species (LDA score > 2.0). (f)KEGG pathway enrichment analysis revealed the enzymes regulating GlcNAc production in differentially expressed microbiota. Quantitative determination of GlcNAc in cecal contents by UHPLC – MRM-MS in vitro (g) and in vivo (h). Data were assessed using a Wilcoxon – Mann–Whitney test. Statistics for GlcNAc determination in vitro using t-test and in vivo was one-way ANOVA. *P < .05, **P < .01, ***P < .001, ****P < .001. Error bars show the mean ± SD for each treatment
Supplemental material
Supplemental Material
Download Zip (19 MB)Data availability statement
The datasets generated in the current study were deposited to the NCBISRA database under the BioProject accession PRJNA861413. The metabolomics datasets generated and analyzed during the current study are available in the MetaboLights repository, using accession number MTBLS5410, http://www.ebi.ac.uk/metabolights/MTBLS5410. Raw data for UHPLC – MRM-MS is publicly available upon acceptance at: https://doi.org/10.6084/m9.figshare.20377545.v1 and https://doi.org/10.6084/m9.figshare.24042927.v1. All other relevant data are available from the corresponding authors upon reasonable requests, after signing a data access agreement. Correspondence and requests for materials should be addressed to Meilin Jin or Qiang Zhang.