Respiratory Disease following Viral Lung Infection Alters the Murine Gut Microbiota

Abstract

Alterations in the composition of the gut microbiota have profound effects on human health. Consequently, there is great interest in identifying, characterizing, and understanding factors that initiate these changes. Despite their high prevalence, studies have only recently begun to investigate how viral lung infections have an impact on the gut microbiota. There is also considerable interest in whether the gut microbiota could be manipulated during vaccination to improve efficacy. In this highly controlled study, we aimed to establish the effect of viral lung infection on gut microbiota composition and the gut environment using mouse models of common respiratory pathogens respiratory syncytial virus (RSV) and influenza virus. This was then compared to the effect of live attenuated influenza virus (LAIV) vaccination. Both RSV and influenza virus infection resulted in significantly altered gut microbiota diversity, with an increase in Bacteroidetes and a concomitant decrease in Firmicutes phyla abundance. Although the increase in the Bacteroidetes phylum was consistent across several experiments, differences were observed at the family and operational taxonomic unit level. This suggests a change in gut conditions after viral lung infection that favors Bacteroidetes outgrowth but not individual families. No change in gut microbiota composition was observed after LAIV vaccination, suggesting that the driver of gut microbiota change is specific to live viral infection. Viral lung infections also resulted in an increase in fecal lipocalin-2, suggesting low-grade gut inflammation, and colonic Muc5ac levels. Owing to the important role that mucus plays in the gut environment, this may explain the changes in microbiota composition observed. This study demonstrates that the gut microbiota and the gut environment are altered following viral lung infections and that these changes are not observed during vaccination. Whether increased mucin levels and gut inflammation drive, or are a result of, these changes is still to be determined.

Keywords: Bacteroidetes; Firmicutes; Mucin 5ac; gut microbiota; influenza; respiratory syncytial virus infections.

Figure 1

Gut microbiota diversity is altered following viral lung infection. Adult BALB/c mice were intranasally dosed with 2 × 10⁶ PFU/ml respiratory syncytial virus (RSV)-A2, sterile phosphate-buffered saline (PBS) or untreated (naïve). Feces were collected under sterile conditions before infection (D0) and at days 4 (D4) and 7 (D7) after infection. (A) Weight was measured after dosing. (B) Viral load in the lungs and colon (C) was estimated using RSV L gene qPCR at D4 and D7 after infection (limit of detection LD for the assay was 2,800 copies; not detected ND 0 copies/no CT value). (D) Colonic microbiota composition (red) was compared to the fecal microbiota composition (black) of the same mice (shapes represent individual mice). (E) Bacterial load in the feces was estimated using 16S rRNA qPCR. (F) Number of operational taxonomic units (OTUs) in feces before and after infection. (G,H) Alpha diversity of the gut microbiota was analyzed using the phyloseq package in R v3.4.1. (I) Beta diversity of the fecal microbiota was analyzed using non-metric multidimensional scaling (NMDS) on a Brays–Curtis distance matrix. N = 5 mice. (E,H) Colored points represent indicial mice. Two-way repeated measures Analysis Of Variance with Dunnett’s correction was used to test for significant differences in viral and bacterial load. Significant changes in microbiota diversity were tested for using Permutational Multivariate Analysis of Variance. *p ≥ 0.05, **p ≥ 0.01, ***p ≥ 0.001.

Figure 2

The ratio of Bacteroidetes to Firmicutes increases in the gut microbiota following respiratory syncytial virus (RSV) infection. (A) The relative abundance of the Firmicutes and Bacteroidetes phyla before and after RSV infection, phosphate-buffered saline (PBS) dosing and among naïve mice. Points represent the mean of N = 5 mice, ±SEM. (B) The relative abundance of each phylum splits into the most abundant gut microbiota families (only families with >1% total abundance included). (C) Fold change of actual operational taxonomic units (OTU) abundance after RSV infection and PBS dosing (day 7) (D) compared to before (p = 0.01 cutoff for significance). Two-way repeated measures Analysis of Variance with Dunnett’s correction was used to test for significant differences in phyla and family abundance. *p ≥ 0.05, **p ≥ 0.01.

Figure 3

Respiratory syncytial virus (RSV) infection—associated changes in the gut microbiota are not due to cage effect or passed on to cage mates. BALB/c mice were intranasally infected with 2 × 10⁶ PFU/ml RSV-A2, dosed with phosphate-buffered saline (PBS) or untreated, animals were either housed (A) separately by treatment regime or (B) co-housed with mice receiving a different treatment. (A,B) Weight was measured after treatment. (C) Feces were collected before (D0) and after infection (D7), and beta diversity of fecal microbiota assessed. (D) Relative abundance in microbiota at the phyla and (E) family levels. (F) BALB/c mice infected with RSV and housed separately were allowed to recover their lost weight. (G) Further fecal samples were taken at D14, D21, and D28, and relative abundance at phyla (H) and family levels assessed. N = 5, points represent the mean ± SEM. Two-way repeated measures Analysis Of Variance with Dunnett’s correction for multiple comparisons was used to test for significant weight loss and changes in phyla and family abundance. *p ≥ 0.05. Non-metric multidimensional scaling (NMDS) on a Brays–Curtis distance matrix was used to visualize diversity, and significant changes in diversity were analyzed using Permutational Multivariate Analysis of Variance, comparing D0–D7 for all groups and D0 between groups.

Figure 4

The gut microbiota is altered following influenza virus infection, but not live attenuated influenza virus (LAIV) vaccine. BALB/c mice were intranasally infected with 4 × 10⁵ PFU/ml A/Eng/195/09 influenza virus, intranasally immunized with 1 × 10⁶ PFU/ml LAIV or intranasally dosed with phosphate-buffered saline (PBS). Vaccinated mice were challenged with 4 × 10⁵ PFU/ml A/Eng/195/09 influenza virus 3 weeks later to establish that this dose of LAIV was protective against infection. Feces were collected before (D0) and after primary infection/immunization (D7). (A) Weight loss was recorded for 7 days after primary infection/immunization and for an additional 7 days after immunized mice were challenged. (B) Beta diversity of the fecal microbiota before and after influenza virus infection, LAIV immunization, and PBS dosing. (C) The relative abundance of the Firmicutes and Bacteroidetes phyla. (D) The relative abundance of Bacteroidetes and Firmicutes splits into the most abundant families (>1% total abundance) before and after influenza virus infection, LAIV immunization, and PBS dosing. N = 5, points represent the mean ± SEM. Two-way repeated measures analysis of variance with Dunnett’s correction for multiple comparisons was used to test for significant weight loss and changes in phyla and family abundance. *p ≥ 0.05, **p ≥ 0.01, ***p ≥ 0.001. Non-metric multidimensional-scaling (NMDS) on a Brays–Curtis distance matrix was used to visualize diversity, and significant changes in diversity were analyzed using Permutational Multivariate Analysis of Variance comparing D0–D7.

Figure 5

Respiratory infection results in increased colonic Mucin 5ac (Muc5ac) and low-grade gut inflammation. (A) Peribronchiolar, perivascular, and interstitial airway inflammation was assessed in a blinded manner for the upper and lower airways and combined together to give a combined inflammatory score. (B,C) Colonic inflammation was scored by counting and measuring the number (B) and size (C) of lymphoid aggregates in the colonic epithelium. (D) Low-grade gut inflammation was assessed by measuring fecal lipocalin-2 levels. (E) IL-17, IL-13, and IFN-y cytokine levels were measured in the bronchoalveolar lavage fluid (airways) and colonic lavage fluid (colon) after respiratory syncytial virus (RSV) infection or phosphate-buffered saline (PBS) dosing. (F) Muc5ac levels were measured in the airway and colon after RSV infection, H1N1 infection, and PBS dosing. N = 5–10 mice/group ± SEM, two-way analysis of variance. *p ≥ 0.05, **p ≥ 0.01, ***p ≥ 0.001.