Myeloid Zfhx3 deficiency protects against hypercapnia-induced suppression of host defense against influenza A virus

Abstract

Hypercapnia, elevation of the partial pressure of CO2 in blood and tissues, is a risk factor for mortality in patients with severe acute and chronic lung diseases. We previously showed that hypercapnia inhibits multiple macrophage and neutrophil antimicrobial functions and that elevated CO2 increases the mortality of bacterial and viral pneumonia in mice. Here, we show that normoxic hypercapnia downregulates innate immune and antiviral gene programs in alveolar macrophages (AMØs). We also show that zinc finger homeobox 3 (Zfhx3) - a mammalian ortholog of zfh2, which mediates hypercapnic immune suppression in Drosophila - is expressed in mouse and human macrophages. Deletion of Zfhx3 in the myeloid lineage blocked the suppressive effect of hypercapnia on immune gene expression in AMØs and decreased viral replication, inflammatory lung injury, and mortality in hypercapnic mice infected with influenza A virus. To our knowledge, our results establish Zfhx3 as the first known mammalian mediator of CO2 effects on immune gene expression and lay the basis for future studies to identify therapeutic targets to interrupt hypercapnic immunosuppression in patients with advanced lung disease.

Keywords: Immunology; Influenza; Innate immunity; Macrophages; Pulmonology.

Figure 1. Zfhx3 is expressed in mouse and human alveolar macrophages.

(A–D) Mouse lung tissue (A), mouse AMØs (B), cultured mouse BMDM (C), and human AMØs (D) were fixed and stained with specific anti-Zfhx3 antibody (magenta). Lung tissue in A was double stained with anti-F4/80 (green) to identify MØs (stars). In A (top panel), alveolar epithelial cells (arrow heads) also stain for Zfhx3. Nuclei were stained with DAPI (blue). Scale bars: 10 μm. (E) Immunoblot of mouse BMDM (top panel) and human THP1 MØs (bottom panel) for Zfhx3 and β-actin.

Figure 2. Hypercapnia alters gene expression in alveolar macrophages, resulting in downregulation of innate immune and antiviral pathways.

AMØs isolated by flow cytometry from the lungs of mice exposed to ambient air or normoxic hypercapnia (10% CO₂/21 % O₂, HC) for 7 days were subjected to RNA-Seq. (A) Pie chart indicating proportion of genes downregulated or upregulated by hypercapnia. (B) Volcano plot showing statistical significance (−log₁₀ [P value]) plotted against log₂ fold change for hypercapnia versus air. Plot indicates significantly upregulated genes (log₂ [fold change] ≥ +0.5, adjusted P <0.05) in red and downregulated genes (log₂ [fold change] ≤ −0.5, adjusted P < 0.05) in blue. (C) K-means clustering of differentially expressed genes is presented as a heatmap of genes downregulated (cluster 1 [C1]) and upregulated (C2) in HC. (D) Bars indicate the top GO biological processes represented by genes downregulated (blue, C1) and upregulated (red, C2) genes in HC.

Figure 3. Zfhx3 deficiency abrogates hypercapnia-induced changes in expression of innate immune and inflammatory pathway genes in alveolar macrophages.

AMØs isolated by flow cytometry from the lungs of Zfhx3^+/+ and Zfhx3^–/– mice exposed to ambient air or normoxic hypercapnia (10% CO₂/21% O₂, HC) for 7 days were subjected to RNA-Seq. (A) K-means clustering of differentially expressed genes is presented as a heatmap of genes downregulated by hypercapnia in AMØs from Zfhx3^+/+ mice but not those from Zfhx3^–/– mice (cluster 1 [C1]) and genes upregulated by hypercapnia in AMØs from Zfhx3^+/+ mice but not those from Zfhx3^–/– mice (cluster 2, C2). (B) Bars indicate the top GO biological processes represented by genes in C1 (red) and C2 (blue).

Figure 4. Myeloid Zfhx3 deficiency protects against hypercapnia-induced increases in lung injury and mortality in IAV-infected mice.

(A–E) Zfhx3^+/+ (Z3⁺) and Zfhx3^–/– (Z3^–) mice were preexposed to normoxic hypercapnia (10% CO₂/21% O₂, HC) for 3 days, or air as control, before being infected intratracheally with IAV (A/WSN/33) (A) at 30 (B and C) or 3 (D and E) pfu per animal; n = 6–10 per group. Lungs from IAV-infected mice harvested 4 dpi were sectioned and stained with H&E, and montage images of whole lung sections (B) were assessed to determine histopathologic scores (HPS) for lung injury (C) analyzed by 1-way ANOVA plus Sidak’s multiple-comparison test; **P < 0.01, ***P < 0.001. Body weight changes over time (D) and Kaplan-Meier plot of survival (E) after infection with 3 pfu IAV were analyzed by log-rank test. **P < 0.05 versus Air Zfhx3^+/+, ***P < 0.05 versus HC Zfhx3^+/+.

Figure 5. Myeloid Zfhx3 deficiency prevents hypercapnia-induced increases in viral replication and suppression of antiviral gene and protein expression in IAV-infected mice.

(A–E) Zfhx3^+/+ and Zfhx3^–/– mice were preexposed to normoxic hypercapnia (10% CO₂/21% O₂, HC) for 3 days, or air as control, before being infected intratracheally with 30 (A, B, D, and E) or 300 (C) pfu IAV (A/WSN/33) per animal; n = 6–10 per group. Viral titers in homogenized lung tissue determined by plaque assay at 4 dpi (A). Expression of viral NS1 protein (magenta) assessed in lung tissue sections from mice sacrificed 4 dpi (B). Ifn1b, Rsad2, and viral ns1 transcript expression in lung tissue sections from mice infected with IAV 300 pfu 1 dpi was detected by RNAscope (C). AMØs from IAV-infected mice were obtained by BAL 1 dpi and cultured under normocapnic (5% CO₂/95% air, NC) or hypercapnic (15% CO₂/21% O₂/64% N₂, HC) conditions for 18 hours, after which viral titers in culture supernatants were determined by plaque assay (D) or assessed for viral NS1 (magenta) and M2 (green) protein expression by immunofluorescence microscopy (E). Nuclei were stained with DAPI (blue) (B, C, and E). In A and D, differences were analyzed by 1-way ANOVA plus Sidak’s multiple-comparison test; *P < 0.05, **P < 0.01. Scale bars: 50 μm (B) and 10 μm (C and E).