Microbiota-derived acetate protects against respiratory syncytial virus infection through a GPR43-type 1 interferon response

Abstract

Severe respiratory syncytial virus (RSV) infection is a major cause of morbidity and mortality in infants <2 years-old. Here we describe that high-fiber diet protects mice from RSV infection. This effect was dependent on intestinal microbiota and production of acetate. Oral administration of acetate mediated interferon-β (IFN-β) response by increasing expression of interferon-stimulated genes in the lung. These effects were associated with reduction of viral load and pulmonary inflammation in RSV-infected mice. Type 1 IFN signaling via the IFN-1 receptor (IFNAR) was essential for acetate antiviral activity in pulmonary epithelial cell lines and for the acetate protective effect in RSV-infected mice. Activation of Gpr43 in pulmonary epithelial cells reduced virus-induced cytotoxicity and promoted antiviral effects through IFN-β response. The effect of acetate on RSV infection was abolished in Gpr43^-/- mice. Our findings reveal antiviral effects of acetate involving IFN-β in lung epithelial cells and engagement of GPR43 and IFNAR.

Fig. 1

High-fiber diet protects mice against RSV-induced disease. a Female BALB/c mice were fed with control fiber (CF) or high-fiber (HF) diets for four weeks before and during RSV infection. Analyses were performed on day five post infection. b Percentage of weight loss post infection relative to initial body weight (day 0) (n = 9). c RSV viral load detected in lung tissue by real-time PCR (viral copies/g of lung tissue) (n = 9). d Total cell number and differential cell counts in bronchoalveolar lavage fluid (BALF) (n = 9). e Representative images of lung tissue section stained with hematoxylin and eosin (H&E) and its respective inflammation scores (n = 3). Scale bars = 100 μm. f Percentage of CD11c⁺CD86⁺ cells in axillary lymph nodes (n = 6). g, h Female BALB/c mice were fed with control fiber (CF), high-fiber (HF) or low-fiber (LF) diet for four weeks before and during RSV infection. Analyses were performed at day five post infection. g Percentage of weight loss post infection relative to initial body weight (day 0). h Total cell number in BALF (n = 5). i Analysis of the fecal microbiota composition from mice fed with CF and HF diets at the family level (relative abundance). j SCFA quantification in colonic luminal content (mg/g feces) (n = 8). k Spearman linear correlation between viral load and acetate quantification of animals fed with control and high-fiber diet. Data in a–e and j are from two independent experiments. All data are expressed as mean ± SEM and were compared using Kruskal–Wallis. In the other cases, Mann–Whitney was used. *p < 0.05, **p < 0.01, ***p < 0.001. Data in b, c, d, g, h, j are provided as a Source Data file

Fig. 2

Microbiota and acetate are essential to HF diet protection against RSV infection. a BALB/c mice were fed a HF diet for four weeks before and during RSV infection. Three days before infection mice received an antibiotic mix (Abx) in the drinking water. Acetate was administered in drinking water after removal of abx treatment. Analyses were performed on day five post infection. b Percentage of body weight loss post infection relative to initial weight (day 0) (n = 9). c RSV viral load detected in lung tissue by real-time PCR (viral copies/g lung tissue) (n = 9). d Total cell number and differential cell counts in BALF (n = 9). e Representative images of lung tissue section stained with H&E and its respective inflammation scores (n = 3). Scale bars = 100 μm. f Acetate quantification in colonic luminal content (mg/g feces) (n = 5). Data are from two independent experiments. All data are expressed as mean ± SEM. The groups were compared using Kruskal–Wallis, except in Fig. 2f, in which the comparison between groups was done using Mann–Whitney test. *p < 0.05, **p < 0.01, ***p < 0.001. Data in b, c, d, f are provided as a Source Data file

Fig. 3

SCFAs pre-treatment protects mice against RSV infection. a BALB/c mice were given, sodium acetate, sodium propionate or sodium butyrate in sterile drinking water at a final concentration of 200 mM for 3 weeks before and during RSV infection. Analyses were performed on day five post infection. b Percentage body weight loss post infection relative to initial weight (day 0) (n = 5). c RSV viral load detected in lung tissue by real-time PCR (n = 5). d Total cell number in BALF (n = 5). E, Representative hematoxylin and eosin (H&E)-stained lung tissue images and its respective inflammation score. Scale bars = 100 μm. f Immunohistochemical (IHC) staining using an anti-RSV Fusion Protein antibody. Scale bars 200 μm. The results are expressed as mean ± SEM. ND not detected. Statistical significance between the groups was determined by Kruskal–Wallis, except in c, in which Mann–Whitney was used to compare the groups. *p < 0.05, **p < 0.01, ***p < 0.001. Data in b–d are provided as a Source Data file

Fig. 4

Acetate treatment protects mice against RSV infection. a BALB/c mice were simultaneously infected with RSV and treated with 200 mM acetate in the drinking water. Analyses were performed on day five post infection. b Percentage body weight loss post infection relative to initial weight (day 0) (n = 9). c RSV viral load was measured 24, 48, 72, and 120 h after infection detected in lung by real-time PCR (viral copies/g lung tissue) (24, 48, and 72 h, n = 5; 120 h, n = 9). ND = not detected. d Total cell number and differential cell counts in BALF (n = 9). e Representative images of lung tissue stained with H&E and its respective inflammation scores (n = 3). Scale bars = 100 μm. f Release of TNF-α and IL-10 in BALF measured at different times after infection (n = 4). g Percentage CD4 + IL-4 + T cells in lung and its representative FACS profile (n = 6). h Percentage of IFNγ ⁺ , IL-17a ⁺ and FoxP3  CD4 T cells in the lung. The representative FACS profiles are shown in Supplementary Fig.3 A and B. i, j Female BALB/c mice were infected with RSV (10⁷ PFU/ml) and 24 h after infection started the treatment with acetate through the intranasal route (10 mM). Treatment was performed daily up to 5 days after infection. i Percentage body weight loss post infection relative to initial weight (day 0) (n = 5). j Total cell number in BALF (n = 5). All data are expressed as mean ± SEM. Data in b, c, d are from two independent experiments. Statistical significance between the groups was determined by Kruskal–Wallis, except in g, h, j, in which Mann–Whitney was used to compare the groups. *p < 0.05, **p < 0.01, ***p < 0.001. Data in b, c, d, g, i, j are provided as a Source Data file

Fig. 5

Acetate treatment protects against RSV infection in the absence of T cells. a–d Female Rag1 knockout mice (background C57BL/6) were simultaneously infected with RSV and treated with 200 mM acetate in drinking water. Analyses were performed on day five post infection. a Percentage body weight loss post infection relative to initial weight (day 0) (n = 8). Ctrl = untreated and uninfected mice. b RSV viral load detected in lung tissue by real-time PCR (viral copies/g lung tissue) (n = 8). c Total cell number and d differential cell counts in BALF (c, n = 8; d, n = 5). All data are expressed as mean ± SEM. Data are from two independent. Statistical significance between the groups was determined using Kruskal–Wallis test except in b, in which Mann–Whitney was used. *p < 0.05, **p < 0.01, ***p < 0.001. Data in a–d are provided as a Source Data file

Fig. 6

Acetate treatment mediated IFN-β response in pulmonary cells lines reducing viral load. a–d A549 cells were pre-treated with 260 µM acetate for 24 h and infected with RSV for 96 h. a Percent of PI (propidium iodide) positive cells detected by flow cytometry. b RSV RNA levels detected using real-time PCR (2^-ΔCt). c Quantification of RSV PFUs. d Representative images of viral titration assay indicating viral plaques. Lysis plate titration was performed using an anti-RSV antibody. e IFN-β protein levels in supernatants of A549 cells pre-treated with 260 µM acetate for 24 h and infected with RSV for further 24 h. f Cellular viability of A549 WT and A549 Ifnar1-/- assessed using PI by flow cytometry. g RSV RNA levels detected using real-time PCR (2^-ΔCt analysis). h Quantification of RSV by PFU assay. i Quantification of translocated NF-kB p65 subunit in A549 cells pre-treated with 260 µM acetate 24 h and infected for 2 h (n = 2 experiments in sextuplicates). j Fluorescence images of NF-kB p65 (green—A) and cell nuclei using DAPI (blue—B). Panel C shows data co-localization (merge). Scale bars = 20 µm. k Western blot analysis of NF-kB p65 in cytoplasmic and nuclear lysates. Cytoplasmic protein bands were normalized by β-actin and nuclear protein bands were normalized by PCNA. L-N, A549 cells were pre-treated 1 h with 2 µM BAY 11–7085. Cells were then treated with 260 µM acetate for 24 h and infected. l Cell viability was evaluated by MTT assay. m RSV RNA levels detected using real-time PCR. n IFN-β protein levels detected by ELISA in the culture supernatant 24 h after infection. All data are expressed as mean ± SEM. Data in a, f, h, m are quadruplicate mean from 3 experiments. Data in b, g, l, m are sextuplicate mean from two experiments. * Significant difference relative to ctrl; # significant difference relative to RSV. Statistical significance was determined with Kruskal–Wallis, except in b, c, g, h, in which Mann–Whitney. *p < 0.05, **p < 0.01, ***p < 0.001. Data in a, b, c, e, f, g, h, i, k, m are provided as a Source Data file

Fig. 7

Acetate protects against RSV infection in an IFNAR-dependent manner. Female BALB/c mice were treated with 200 mM acetate in drinking water and infected with RSV (10⁷ PFU/mL) for 24, 48, 72, and 120 h. a Ifnb1 gene expression in the lung detected at different time points after infection (2-ΔCt analysis in real-time PCR) (n = 5). b IFN-β protein detected in the BALF or lung homogenates c at different time points after infection (n = 5). d Oas1 and Isg15 gene expression in the lung (2-ΔCt analysis in real-time PCR) (n = 5). e Pulmonary epithelial cells from female naïve BALB/c mice were treated with acetate (260 µM) for 24 h and then infected with RSV (10⁴ PFU/mL) for a further 24 h to detect IFN-β production (by ELISA) and Ifnb1 expression (fold change compared to untreated/uninfected control) (n = quadruplicate mean of 4 animals). f Ifnb1 expression of CD45⁺CD326^- and CD45^-CD326⁺ sorted cells from lung of mice acetate-treated and RSV-infected for 72 h (n = 5). g IFN-β production by alveolar macrophages of mouse untreated or treated with acetate in drinking water (200 mM) for 5 days and infected ex vivo with RSV (10⁴ PFU/ml) for 24 h (n = 6). Gate strategy, pre-sorting and post-sorting panel are shown in Supplementary Fig. 9. h–l Wild-type and Ifnar-/- mice were simultaneously infected with RSV and treated with 200 mM acetate in drinking water. Analyses were performed on day 5 post infection. h Percentage body weight loss post infection relative to initial weight (day 0) (n = 8). i RSV viral load detected in lung tissue by real-time PCR (viral copies/g lung tissue) (n = 8). j Viral titer in the lung (PFUs/g lung tissue) (n = 5). k Inflammation score of lung histology (n = 3). l Immunohistochemistry staining of RSV in lung tissue sections. All data are expressed as mean ± SEM. Data are from two independent experiments. Statistical significance between groups was determined with Kruskal–Wallis, except in e, in which Mann–Whitney was used. *p < 0.05, **p < 0.01, ***p < 0.001. Data in a, b, c, d, f, h, i, j are provided as a Source Data file

Fig. 8

Specific activation of GPR43 protects against in vitro RSV infection. A549 cells were treated with 260 µM acetate or a GPR43-synthetic agonist (10 µM 4-CMTB) for 24 h and then infected with RSV (10⁴ PFU/ml). Four days later response parameters were analyzed. a cell viability was accessed by MTT assay. b Percent of cell death was determined by PI staining. c RSV RNA levels were detected using real-time PCR (2-ΔCt analysis). d Concentrations of IFN-β in the cell supernatant 12 h after infection were detected by ELISA. All data are expressed as mean ± SEM. Data in a and b are quadruplicate mean of two experiments. Data in c and d are sextuplicate mean of one experiment. Statistical significance was determined with Krukal-wallis. **p < 0.01, ***p < 0.001. Data in b–d are provided as a Source Data file

Fig. 9

GPR43 is necessary for acetate protection of RSV-induced disease. a–g Gpr43 (G-protein coupled receptor 43) knockout mice (background in C57BL/6) and their controls (WT) were infected with RSV and treated with 200 mM acetate in drinking water. Analyses were performed on day five post infection. a Percentage weight loss post infection relative to original weight (day 0) (WT, n = 6; Gpr43−/−, n = 9). b RSV viral load detected in lung tissue by real-time PCR (viral copies/g of lung tissue) (WT, n = 6; Gpr43−/−, n = 9). c Total cell number and differential cell counting in BALF(WT, n = 6; Gpr43−/−, n = 8). d Immunohistochemistry staining of RSV in lung tissue sections. e mRNA expression of Ifnb1 in the lung (2-ΔCt analysis) (WT, n = 6; Gpr43−/−, n = 8). f mRNA expression of Oas1 and Isg15 genes in the lung (2-ΔCt analysis) (WT, n = 6; Gpr43−/−, n = 8). g IFN-β protein levels detected in the BALF or lung homogenate (WT, n = 4; Gpr43−/−, n = 6). h, i Pulmonary epithelial cells from female C57BL/6 wild-type and Gpr43−/− mice were treated with 260 µM acetate for 24 h and then infected with RSV (10⁴ PFU/mL) for 24 h or 4 days. h cell viability evaluated by MTT assay. Control untreated/uninfected is showed as the dotted line. i IFN-β levels detected in supernatant of primary mouse lung cells culture (n = quadruplicate mean of 4 animals). j, k, Female C57BL6 wild-type and Gpr43−/− mice were infected with RSV (10⁷ PFU/ml) for 24 h and then treated with acetate through the intranasal route (10 mM). After 24 h the lung was collected and processed for detection of IFNβ production by ELISA. All data are expressed as mean ± SEM. Data in a–g are from two independent experiments. Statistical significance between the groups was determined with Kruskal–Wallis. *p < 0.05, **p < 0.01, ***p < 0.001. Data in a, b, c, f, g, i are provided as a Source Data file