ZIP8-Mediated Intestinal Dysbiosis Impairs Pulmonary Host Defense against Bacterial Pneumonia

Abstract

Pneumococcal pneumonia is a leading cause of morbidity and mortality worldwide. An increased susceptibility is due, in part, to compromised immune function. Zinc is required for proper immune function, and an insufficient dietary intake increases the risk of pneumonia. Our group was the first to reveal that the Zn transporter, ZIP8, is required for host defense. Furthermore, the gut microbiota that is essential for lung immunity is adversely impacted by a commonly occurring defective ZIP8 allele in humans. Taken together, we hypothesized that loss of the ZIP8 function would lead to intestinal dysbiosis and impaired host defense against pneumonia. To test this, we utilized a novel myeloid-specific Zip8KO mouse model in our studies. The comparison of the cecal microbial composition of wild-type and Zip8KO mice revealed significant differences in microbial community structure. Most strikingly, upon a S. pneumoniae lung infection, mice recolonized with Zip8KO-derived microbiota exhibited an increase in weight loss, bacterial dissemination, and lung inflammation compared to mice recolonized with WT microbiota. For the first time, we reveal the critical role of myeloid-specific ZIP8 on the maintenance of the gut microbiome structure, and that loss of ZIP8 leads to intestinal dysbiosis and impaired host defense in the lung. Given the high incidence of dietary Zn deficiency and the ZIP8 variant allele in the human population, additional investigation is warranted to improve surveillance and treatment strategies.

Keywords: gut-lung axis; host defense; microbiome; pneumonia; zinc; zinc transporter.

Figure 1

ZIP8 loss alters the intestinal microbial community and inferred functional capacity. 16S rRNA gene sequencing of WT and Zip8KO mice cecal microbial community. (A) Beta diversity of WT and Zip8KO mice, as determined by distance-based redundancy analysis (dbRDA) on sample-wise Bray–Curtis dissimilarity distances. (B) Differentially abundant ASVs as determined by DESeq2 using negative binomial generalized linear models for each taxa and Wald test for significances. (C) PCA plot of the inferred functional capacity of the cecal microbial communities from WT and Zip8KO mice. (D) Differentially abundant functional pathways, as determined by STAMP. n = 5/group.

Figure 2

Zip8 loss decreases intestinal butyrate levels. Short chain fatty acid analysis of cecal content from WT and Zip8KO mice. Levels of cecal (A) butyric acid, (B) propionic acid, (C) acetic acid, (D) valeric acid, (E) 2-methylbutyric acid, (F) isobutyric acid, and (G) isovaleric acid. n = 6–9/group.

Figure 3

Recolonization of antibiotic cleansed mice results in maintenance of donor community structure. PCA plot of the cecal microbial communities from donor and recipient mice recolonized with the cecal microbial community from WT and Zip8KO mice, via STAMP. Beta diversity statistics were calculated on sample-wise Bray–Curtis dissimilarity distances. n = 5/group.

Figure 4

ZIP8-associated dysbiosis impairs pulmonary host defense. Mice recolonized with the cecal microbial community from WT and Zip8KO mice were infected with S. pneumoniae, and pulmonary host defense was assessed. (A) Weight change in recolonized mice post infection. (B) Representative histological staining of lung tissue, and (C) quantified inflammatory score. (D) Representative TUNEL staining of lung tissue and (E) quantitative mean pixel intensity. S. pneumoniae burden in the (F) lungs and (G) spleen. Bars represent the mean ± SEM and dots represent individual mice. p values are indicated in the figure and were determined by one-way ANOVA with Sidak’s multiple comparison test. n = 10/group, 5/experimental replicates.

Figure 5

ZIP8-associated dysbiosis increases pulmonary immune cell numbers. Mice recolonized with the cecal microbial community from WT and Zip8KO mice were infected with S. pneumoniae, and the number of BAL immune cells were assessed. (A) Total BAL leukocytes in recolonized mice post-infection. (B) Total BAL neutrophils in recolonized mice post-infection. (C) Total BAL protein in recolonized mice post-infection. Total BAL (D) macrophages and (E) lymphocytes in recolonized mice post-infection. Bars represent the mean ± SEM and dots represent individual mice. p values are indicated in the figure and were determined by one-way ANOVA with Sidak’s multiple comparison test. n = 10/group, 5/experimental replicate.

Figure 6

ZIP8-associated dysbiosis increases pulmonary inflammation. Mice recolonized with the cecal microbial community from WT and Zip8KO mice were infected with S. pneumoniae, and the number of pulmonary cytokine levels were assessed. (A) IL-6 levels in the BAL of recolonized mice post-infection. (B) CXCL1 levels in the BAL of recolonized mice post-infection. (C) TNF-α levels in the BAL of recolonized mice post-infection. (D) IFN-γ levels in the BAL of recolonized mice post-infection. Bars represent the mean ± SEM and dots represent individual mice. p values are indicated in the figure and were determined by one-way ANOVA with Sidak’s multiple comparison test. n = 10/group, 5/experimental replicate.

Figure 7

Zip8-associated dysbiosis alters the percentage of pulmonary immune cells. Mice recolonized with the cecal microbial community from WT and Zip8KO mice were infected with S. pneumoniae, and the percentage of pulmonary immune cells were assessed. Percentage of pulmonary (A) neutrophils, (B) dendritic cells, (C) alveolar macrophages, (D) inflammatory macrophages, (E) anti-inflammatory macrophages, (F) CD4+ T-cells, (G) CD8+ T-cells, and (H) NK cells. Bars represent the mean ± SEM and dots represent individual mice. p values are indicated in the figure and were determined by one-way ANOVA with Sidak’s multiple comparison test. n = 10/group, 5/experimental replicates.

Figure 8

ZIP8-associated dysbiosis increases pulmonary and intestinal epithelial damage post S. pneumoniae infection. Mice recolonized with the cecal microbial community from WT and Zip8KO mice were infected with S. pneumoniae and the levels of circulating SPD-1 and IFABP were assessed. Circulating levels of (A) SPD-1 and (B) IFABP in mice recolonized with the cecal microbial community from WT and Zip8KO mice. Bars represent the mean ± SEM and dots represent individual mice. p values are indicated in the figure and were determined by one-way ANOVA with Sidak’s multiple comparison test. n = 6–12/group, 5–3/experimental replicates.

Figure 9

ZIP8-associated microbial products suppress macrophage cytokine production. Microbial products derived from the cecal microbial community from WT and Zip8KO mice were co-cultured with mouse MH-S macrophages for 6 or 24 h and macrophage viability and cytokine production was assessed. (A) Cytotoxicity of microbial products on MH-S cells. (B) IL-6 levels in the MH-S supernatant following co-culture with microbial products. (C) TNF-α levels in the MH-S supernatant following co-culture with microbial products. Bars represent the mean ± SEM and dots represent individual mice. p values are indicated in the figure and were determined by one-way ANOVA with Sidak’s multiple comparison test. n = 6/group, 3/experimental replicates.