Interferon (IFN)-γ promotes monocyte-mediated lung injury during influenza infection

Abstract

Influenza A Virus (IAV) infection triggers an exuberant host response that promotes acute lung injury. However, the host response factors that promote the development of a pathologic inflammatory response to IAV remain incompletely understood. In this study, we identify an interferon (IFN)-γ-regulated subset of monocytes, CCR2⁺ monocytes, as a driver of lung damage during IAV infection. IFN-γ regulates the recruitment and inflammatory phenotype of CCR2⁺ monocytes, and mice deficient in CCR2 (CCR2^−/−) or IFN-γ (IFN-γ^−/−) exhibit reduced lung inflammation, pathology, and disease severity. Adoptive transfer of WT (IFN-γR1^+/+), but not IFN-γR1⁻/⁻ CCR2⁺ monocytes, restore the wild-type (WT)-like pathological phenotype of lung damage in IAV-infected CCR2^−/− mice. CD8⁺ T cells are the main source of IFN-γ in IAV-infected lungs. Collectively, our data highlights the requirement of IFN-γ signaling in the regulation of CCR2⁺ monocyte-mediated lung pathology during IAV infection.

One Sentence Summary:

Interferon-γ mediates acute lung injury by regulating the recruitment and inflammatory phenotype of CCR2⁺ monocytes during IAV pathogenesis.

Graphical Abstract

Introduction

Despite seasonal vaccines, IAV infection continues to pose a significant threat to public health (Bailey et al., 2018; Lafond et al., 2021; Rolfes et al., 2018). Exuberant host response triggered by IAV can cause acute lung injury severe enough to require hospitalization (Gregory and Kobzik, 2015; Herold et al., 2015; Klomp et al., 2021). Additionally, IAV lung pathology promotes secondary bacterial disease, causing significant morbidity and mortality worldwide (Metzger and Sun, 2013; Morris et al., 2017; Rynda-Apple et al., 2015). Despite the wealth of information on IAV and host responses, the contribution of pathologic host response to IAV lung damage remains incompletely understood. Acute injury in IAV-infected lungs is orchestrated by interaction of leukocytes with non-hematopoietic cells, i.e., epithelial cells (Ellis et al., 2015; van de Sandt et al., 2017). Although these interactions are central to the resolution of IAV infection, the magnitude of leukocyte host response determines both the outcome of infection resolution and the degree of immune-mediated lung injury. The relative contribution of leukocyte subsets in IAV lung damage, and the host factors regulating the inflammatory phenotype of leukocytes, is yet to be fully elucidated. To identify the immune cells implicated in influenza lung injury, we performed unbiased single-cell RNA-Seq (scRNA-Seq) of CD45⁺ (leukocytes) cells from lungs of IAV-infected mice that displayed significant weight loss, distinct signs of lung pathology, and increased susceptibility to secondary bacterial infection by Streptococcus pneumoniae (Spn). The scRNA-Seq identified CCR2 expressing (CCR2⁺) monocytes and monocyte-derived cells, such as dendritic cells (DCs) and macrophages, as the most predominant myeloid cells that correlated with acute lung injury.

CCR2⁺ monocytes have been shown to cause acute lung damage in animal models of IAV infection (Coates et al., 2018; Ellis et al., 2015). However, the host response factors that govern the hyper-inflammatory and tissue damage associated phenotype of CCR2⁺ monocytes are not well understood. Interferon (IFN)-γ plays both protective and pathological functions in infection and inflammatory diseases (Califano et al., 2018; Flynn et al., 1993; Liu et al., 2019; Zhang, 2007). The crucial role of IFN-γ has been shown in the polarization of pro-inflammatory M1 macrophage phenotype (Chauhan et al., 2018; Zhao et al., 2019). Recent reports have shown the pathological functions of IFN-γ during IAV infection (Califano et al., 2018; Liu et al., 2019). However, the host response mechanisms implicated in IFN-γ mediated lung pathology during IAV infection remain elusive. Our scRNA-Seq data reveal the significant enrichment of IFN-γ induced pathways in CCR2⁺ monocytes and monocyte-derived dendritic cells. These observations led us to hypothesize that IFN-γ contributes to monocyte-mediated lung injury during IAV infection. Our data demonstrated that CD8⁺ T cell-derived IFN-γ promoted the recruitment and inflammatory phenotype of CCR2⁺ monocytes in IAV-infected lungs. Additionally, CCR2^−/− and IFN-γ^−/− mice exhibited reduced disease severity, lung injury, and enhanced resistance against Spn secondary bacterial infection. Furthermore, the adoptive transfer of IFN-γR1^+/+, but not IFN-γR1^−/− CCR2⁺ monocytes, restored the WT-like lung pathology in IAV-infected CCR2^−/− mice. Our data demonstrate that IFN-γ regulates CCR2⁺ monocyte-mediated lung pathology during IAV infection.

Results

scRNA-Seq identifies CCR2⁺ monocytes and monocyte-derived DCs and macrophages as the most significant leukocytes in IAV-infected lungs.

To interrogate the immune cell landscapes in the lungs of IAV-infected mice, we performed scRNA-Seq from CD45⁺ (leukocytes) cells of un-infected (hereafter mock) and IAV-infected lungs at day 7 post-infection (p.i.). The day 7 time point was selected based on an IAV kinetic experiment that indicated peak weight loss, lung inflammation, tissue-pathology, and viral load at day 7 p.i. (Figure S1A–F). By day 14 p.i., mice regained the lost weight (Figure S1B) and resolved lung injury (Figure S1C–S1E). Therefore, this previously established sublethal model (Rutigliano et al., 2014) allowed us to study the pathologic host response associated with IAV lung injury and secondary Spn infections. Using scRNA-Seq, we identified 15 cell clusters in the lungs of both mock- and IAV-infected mice (Figure 1A). A total of 8 cell types were subsequently validated using established markers for cell identification (Figure S2A). We then compared the distribution of cellular compartments between the mock and IAV-infected groups. The scRNA-Seq data revealed that CCR2⁺ monocytes, monocyte-derived DCs and macrophages, and CD8⁺ T cells significantly accumulated in IAV-infected lungs (Figure 1A). Flow cytometry-based t-distributed stochastic neighbor embedding (t-SNE) analysis supported the scRNA-Seq data showing these cell types as the most significantly recruited cells in IAV-infected lungs (Figure 1B).

Figure 1. CCR2⁺ monocytes and monocyte-derived cells are the most enriched immune cells in IAV-infected lungs.

C57BL/6 (WT) and CCR2^−/− mice were mock-infected with PBS or infected with 250 PFUs of IAV (PR8) intranasally. At days 1, 3, 7, 10, and 14 p.i., mice were euthanized, and lung cells were processed for scRNA-Seq or flow cytometry.

(A) tSNE embedding of CD45⁺ single-cell transcriptomes annotated by cell type (top panel) and bar plot (bottom panel) showing proportional changes by cell type.

(B) Flow cytometry tSNE analysis (top panel) and relative expression change (bottom panel) among immune cell types.

(C) Pseudotime trajectory construction analysis of monocytes, macrophages, and dendritic cells. Top panel: The prediction of cells developmental trajectory with pseudotime. Middle panel: Pseudotime map of the cells reflecting the differentiation trajectories. Lower panel: CCR2 expression changes along the differentiation trajectory. Each dot represents a single cell.

(D-F) The counts of total monocytes (% CD11b⁺ Ly6C⁺ cells per 1×10⁵ cells), DCs (% CD11c⁺ MHCII⁺ cells per 1×10⁵ cells), macrophages (% CD11b⁺ F4/80⁺ cells per 1×10⁵ cells), CD4⁺ T cells (% CD3⁺ CD4⁺ cells per 1×10⁵ cells), CD8⁺ T cells (% CD3⁺ CD8⁺ cells per 1×10⁵ cells), and epithelial cells (% CD45⁻ Epcam⁺ cells per 1×10⁵ cells) in mock and IAV-infected WT and CCR2^−/− mice (Figure S3).

(G) Heat map of GSVA enrichment scores of selected significant pathways in macrophage, monocytes, and dendritic cells.

Standard error bars represent mean ± SEM; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 and ****p ≤ 0.0001 (One-Way ANOVA with Tukey’s post-hoc analysis). All data are representative of 2–3 independent experiments of n=5 per group, unless otherwise stated.

Next, we used scRNA-Seq data to trace cell fate and reconstruct cell lineage direction using the cell trajectory approach (Trapnell et al., 2014). The cells from mock and IAV-infected lungs were ordered along pseudotime (Figure 1C). A vast majority of monocytes, DCs, and macrophages aligned in a continuum along the inflammatory state 1 (Figure 1C; Figure S2B), demonstrating a significant differentiation trajectory of monocytes to DCs and macrophages. CCR2 expression was primarily expressed by monocytes, and the differentiation of monocytes to DCs or macrophages resulted in the downregulation of CCR2 expression (Figure 1C). Because CCR2 can be expressed by a number of non-monocyte immune cells, including CD4⁺ and CD8+ T cells (Connor et al., 2004; Nansen et al., 2000), we investigated the impact of CCR2 deficiency in the recruitment of non-monocyte immune cell types in IAV-infected lungs. Consistent with scRNA-Seq, IAV-infected CCR2^−/− mice (compared to WT) exhibited a significant reduction in total monocytes, DCs and macrophages (Figure 1D). These data support the scRNA-Seq findings that DCs and macrophages derived from CCR2⁺ monocytes. The CCR2^−/− mice did not exhibit any defects in T cell (CD4⁺ and CD8⁺) recruitment into IAV-infected lungs (Figure 1E). A prior report showed that alveolar epithelial cells expressed CCR2 (Christensen et al., 2004). However, our data show that WT and CCR2^−/− mice had comparable epithelial cell (CD45⁻ Epcam⁺) frequencies in both steady-state and during IAV infection (Figure 1F).

Next, we performed Gene Set Variation Analysis (GSVA) to identify functionally enriched pathways in monocytes, DCs, and macrophages in both mock and IAV-infected groups. Compared to the mock, CCR2⁺ monocytes and DCs from IAV-infected mice show significantly upregulated IFN-γ associated pathways (Figure 1G). Collectively, these data highlight a correlation between a predominance of CCR2⁺ monocytes and monocyte-derivative cells in IAV-infected lungs and the development of acute lung injury. It also highlighted that CCR2⁺ monocyte and monocyte-derived DCs exhibited a strong enrichment of IFN-γ response pathways.

IFN-γ regulates the recruitment and inflammatory phenotype of monocytes in IAV model.

Our scRNA-Seq data reveals the enrichment of IFN-γ response pathways in monocytes and monocyte-derived DCs (Figure 1G). Based on these findings, we postulated that IFN-γ signaling may regulate monocyte inflammatory response in IAV-infected lungs. First, we found that CCL2 (the most abundant CCR2 ligand) and IFN-γ protein levels in the BAL peaked at day 7 p.i., which correlated with total (CD11b⁺ Ly6C⁺) and CCR2⁺ monocytes (CD11b⁺ Ly6C⁺ CCR2⁺) in IAV-infected lungs (Figure 2A–C). Furthermore, IFN-γ^−/− mice (compared to WT) showed reduced levels of CCL2 (Figure 2D), total monocytes, and CCR2⁺ monocytes in IAV-infected lungs (Figure 2E). However, IFN-γ^−/− mice did not exhibit any defects in T cell recruitment, as both WT and IFN-γ^−/− mice had similar levels of CD4⁺ or CD8⁺ T cell frequencies in IAV-infected lungs (Figure 2F). Both hematopoietic and non-hematopoietic cells can produce CCL2 and thus facilitate the recruitment of CCR2⁺ monocytes from the bone marrow into the lungs (Maus et al., 2003; Rollins, 2006). To study the role of IFN-γ in the induction of CCL2 in vitro, we stimulated the human bronchial epithelial cells (HBECs) grown in air-liquid interface (ALI) in the presence of recombinant human IFN-γ. The stimulation of HBECs with IFN-γ led to a robust increase in CCL2 expression (Figure 2G). Interestingly, IAV infection, compared to IFN-γ treatment, resulted in a reduced CCL2 secretion in culture supernatants. This suggests that IFN-γ plays a crucial role in eliciting CCL2 secretion in both infected and non-infected epithelial cells. These data show that IFN-γ is crucial to regulating the CCL2 expression and recruitment of CCR2⁺ monocytes into IAV-infected lungs.

Figure 2. IFN-γ promotes the recruitment and inflammatory phenotype of CCR2⁺ monocytes.

WT and IFN-γ^−/− mice were mock-infected with PBS or infected with 250 PFUs of IAV (PR8) intranasally. At day 7 p.i., mice were euthanized, and the BAL fluid and lungs were aseptically collected.

(A) IFN-γ and CCL2 protein levels in BAL, as measured by ELISA.

(B) Total monocyte frequency (% CD11b⁺ Ly6C⁺ cells per 10⁵ cells) in lungs.

(C) CCR2⁺ monocytes (% of total monocytes) in the lungs.

(D) CCL2 protein levels in BAL of WT and IFN-γ^−/− mice at day 7 p.i.

(E) The frequency of total (left) and CCR2⁺ monocytes (right) in the lungs of WT and IFN-γ^−/− mice at day 7 p.i.

(F) Percent T-cells (CD3⁺CD4⁺ and CD3⁺CD8⁺ per 1×10⁵ cells) in the lungs.

(G) Protein levels of CCL2 in PR8-infected and IFN-γ stimulated HBECs grown in ALI.

(H-J) RNA-Seq analysis of total monocytes from mock and PR8 infected WT and IFN-γ^−/− mice.

(H) Venn diagram of Differentially Expressed Genes (DEGs) (I) Heatmap shows log 2-fold change of DEG among group comparisons. Relative directionality of DEGs is normalized across rows to represent change in expression levels across models. (J) Canonical pathways identified with a log10 significance (p-value) cutoff of 1.3 are displayed with the Z-score.

(K) Gene expression analysis of SOCS-1, IRF-1, FasL, Stat-1, Granzyme B, and IFN-γ in FACS sorted pooled CCR2⁺ monocytes from PR8-infected WT (n=5) and IFN-γ^−/− (n=5) mice, representative of one experiment. Statistics calculated using Students t-test.

Standard error bars represent mean ± SEM; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.005 and ****p ≤ 0.001. (A-G) Statistics calculated using One-Way ANOVA with Tukey’s post-hoc analysis. All data are representative of 2–3 independent experiments of n=5 per group unless otherwise denoted.

Next, we magnetically purified monocytes, defined as CD11b⁺ Ly6C⁺, from WT and IFN-γ^−/− mice on day 7 p.i.. Because CCR2⁺ monocytes are extremely scarce in the lungs of mock-infected mice, total lung monocytes were used to perform RNA-Seq. Differentially-expressed genes (DEGs) were determined by comparing gene expression in a pairwise manner between the mock, IAV-infected WT, and IAV-infected IFN-γ^−/− mice with a minimum absolute log₂ fold change of 1 and adjusted p-value <0.05. The shift in inflammatory response can be observed at a gene level by the DEGs distinctly affected by the presence (WT-IAV, 469 DEGs) or absence (IFN-γ^−/−-IAV, 1241 DEGs) of IFN-γ compared to the mock (Figure 2H). Canonical pathways enriched within at least two comparisons were incorporated into a pathway network and examined for shared genes between pathways. Many pro-inflammatory pathways that were upregulated via IAV infection in WT mice were repressed in IFN-γ^−/− mice (Figure 2I–2J). Compared to the mock-monocytes, monocytes from IAV-infected WT lungs upregulated the expression of various pro-inflammatory genes, including Stat1, Irf1, GzmB, FasL, Nos2, Socs, and showed enrichment of molecular pathways associated with IFN-γ signaling. Consistently, IFN-γ ablation resulted in a significant downregulation of monocyte inflammatory molecules (Figure 2I). To validate the monocyte (total) RNA-Seq findings, CCR2⁺ monocytes (CD11b⁺ Ly6C⁺ CCR2⁺) were FACS sorted from the lungs of IAV-infected WT and IFN-γ^−/− mice. Consistent with the RNA-Seq findings, CCR2⁺ monocytes from IFN-γ^−/− mice exhibited a significantly reduced expression of key pro-inflammatory genes, such as FasL, GzmB, Irf-1, and Stat1 (Figure 2K). Together, these data demonstrated that IFN-γ promoted the inflammatory phenotype of monocytes during IAV infection.

CCR2^−/− and IFN-γ^−/− mice exhibit reduced lung pathology and increased resistance against secondary Spn disease.

Because IFN-γ regulated the recruitment and inflammatory phenotype of CCR2⁺ monocytes, we compared the disease severity, as determined by weight loss, and lung pathology in WT, CCR2^−/−, and IFN-γ^−/− mice 7 days p.i.. Compared to IAV-infected WT mice, IFN-γ^−/− and CCR2^−/− mice displayed significantly reduced weight loss (Figure 3A). Furthermore, CCR2^−/− and IFN-γ^−/− mice did not exhibit defects in regaining the lost body weight, suggesting that CCR2 or IFN-γ deficiency did not produce long-term defects following IAV recovery (Figure 3A). The H&E-stained lung sections of IAV-infected CCR2^−/− and IFN-γ^−/− mice (compared to WT) exhibited significantly reduced levels of bronchiolitis and perivascular leukocytic cuffing (Figure 3B), as well as reduced vascular damage (Figure 3C). The degree of vascular damage was determined by the vacuolation of endothelial cells and separation from the underlying musculature (Zheng et al., 2020). Next, we compared the lung barrier integrity among WT, IFN-γ^−/−, and CCR2^−/− mice by investigating the expression of cell junction proteins, ZO-1 and E-cadherin, as well as albumin leakage in the BAL. Compared to IAV-infected WT mice, IFN-γ^−/− and CCR2^−/− mice had higher expression of cell junction proteins, correlating with reduced albumin leakage in the BAL (Figure 3E). Consistent with reduced lung injury, IFN-γ^−/− and CCR2^−/− mice (compared to WT) had reduced levels of TNF-α in the BAL (Figure 3F). Notably, IFN-γ or CCR2 deficiency did not impair viral clearance as CCR2^−/− and IFN-γ^−/− mice had similar viral loads to WT on day 7 p.i., as determined by TCID₅₀ and qPCR (Figure 3G, Figure S4). The reduced lung injury in IAV-infected CCR2^−/− and IFN-γ^−/− mice also resulted in enhanced resistance to Spn co-infection, as CCR2^−/− and IFN-γ^−/− mice exhibited reduced Spn bacterial burden in the lung and significantly higher survival rates than those of WT mice (Figure 3H).

Figure 3. IFN-γ^−/− and CCR2^−/− mice display reduced lung pathology and increased resistance against IAV/Spn co-infection.

WT, IFN-γ^−/−, and CCR2^−/− mice were mock-infected or infected with 250 PFUs of IAV (PR8) intranasally. At day 7 p.i., mice were euthanized, and the BAL fluid and lungs were aseptically collected.

(A) Percent weight change of WT, IFN-γ^−/−, and CCR2^−/− mice for up to 21 days after PR8 infection.

(B) H&E images of lung sections of WT, IFN-γ^−/−, and CCR2^−/− (20X).

(C) Lung vascular endothelial damage scores of WT, IFN-γ^−/−, and CCR2^−/− mice at day 7 p.i.

(D) Immunofluorescence images of the tight junction protein ZO-1 and gap junction protein E-cadherin (60X). Quantitation performed using ImageJ.

(E-F) Albumin and TNF-α levels in BAL.

(G) Viral titers in lung homogenates by TCID₅₀.

(H) Bacterial burdens in lung homogenates of WT, IFN-γ^−/−, and CCR2^−/− mice 48 hours after Spn co-infection.

(I) The survival of WT, IFN-γ^−/−, and CCR2^−/− mice monitored for up to 2 weeks after co-infection. n=9–10 per group of one independent experiment. Statistical analysis performed using the Mantel-Cox log-rank test. A p value <0.05 was considered statistically significant.

Standard error bars indicate mean ± SEM; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 and ****p ≤ 0.0001. Statistics calculated using One-Way ANOVA with Tukey’s post-hoc analysis. All data are representative of 2–3 independent experiments of n=5 per group unless otherwise stated.

Adoptive transfer of WT (IFN-γR1^+/+) monocytes restores WT-like lung pathology in CCR2^−/− mice.

Our data show that IFN-γ promotes lung injury by regulating the recruitment and inflammatory phenotype of CCR2⁺ monocytes. The similarities of IAV lung pathology between CCR2^−/− and IFN-γ^−/− mice further supported the relationship between IFN-γ and CCR2⁺ monocytes. To determine whether IFN-γ directly contributed to CCR2⁺ monocyte mediated lung pathology, we performed the adoptive transfer of IFN-γR1^+/+ and IFN-γR1^−/− CCR2⁺ monocytes into IAV-infected CCR2^−/− mice. The CCR2⁺ monocytes (CD11b⁺ Ly6C⁺ CCR2⁺) were FACS sorted from the bone marrow of naïve WT (IFN-γR1^+/+ CCR2⁺) or IFN-γR1^−/− (IFN-γR1^−/− CCR2⁺) mice and stained with CFSE, as described earlier (Si et al., 2010). IFN-γR1⁺ CCR2⁺ or IFN-γR1⁻ CCR2⁺ monocytes were adoptively transferred to CCR2^−/− mice (Figure 4A) 24 hours after IAV-infection. The lung pathology of adoptively transferred CCR2^−/− mice were subsequently analyzed at day 7 p.i. The CCR2^−/− mice that received CCR2⁺ monocytes from WT or IFN-γR1^−/− mice exhibited similar levels of CCR2⁺ monocytes in the lungs (Figure 4B), suggesting that IFN-γR1 deficiency did not affect the recruitment of CCR2⁺ monocytes in IAV-infected lungs. Consistent with data in Figure 3, CCR2^−/− mice (compared to WT) showed reduced levels of lactate dehydrogenase (LDH) and albumin in BAL (Figure 4C–4D). The adoptive transfer of IFN-γR1^+/+ CCR2⁺ monocytes, but not IFN-γR1^−/− CCR2⁺ monocytes, led to a significant increase in LDH (Figure 4C) and albumin (Figure 4D) levels in BAL of IAV-infected CCR2^−/− mice. However, there was no difference in LDH and albumin levels between WT and CCR2^−/− mice recipient of IFN-γR1⁺ CCR2⁺ monocytes (Figure 4C–4D). Consistent with increased LDH and albumin levels in BAL, the H&E-stained lung sections from CCR2^−/− mice that received IFN-γR1^+/+ CCR2⁺ monocytes displayed increased immune cell infiltration and vascular injury (Figure 4E). Overall, these data demonstrated that adoptive transfer of IFN-γR1^+/+ CCR2⁺ monocytes restored WT-like lung pathology in CCR2^−/− mice.

Figure 4. Adoptive transfer of WT (IFN-γR1^+/+) CCR2⁺ monocytes restore WT-like lung pathology in CCR2⁻/⁻ mice.

WT and CCR2^−/− mice were mock-infected or infected with 250 PFUs of IAV (PR8) intranasally.

(A) Schematic representation of the adoptive transfer experiment. CCR2⁺ monocytes (CD11b⁺Ly6C⁺CCR2⁺) were FACS-sorted from naïve WT and IFN-γR1^−/− mice, and one million cells were adoptively transferred (retro-orbitally) into CCR2^−/− mice day 1 p.i.

(B) Counts of CCR2⁺ monocytes (per 1×10⁵ cells) detected in lungs of adoptively transferred mice at day 7 p.i.

(C, D) LDH and albumin levels in BAL.

(E) H&E images of lung sections at 20X (left panel) and lung vascular endothelial damage (right panel).

(F) Intracellular cytokine staining for IFN-γ (left panel) and relative changes (right panel) (compared to mock) in IFN-γ expressing cells in mock- and PR8-infected lung cells at day 7 p.i.

(G) Experimental outline: PR8 infected mice were treated with anti-CD8 depletion antibody (WT-PR8-αCD8) and levels of CCL2, IFN-γ and the numbers of CCR2⁺ monocytes were detected at day 7 p.i.

(H-I) IFN-γ and CCL2 protein levels in BAL.

(J) The frequency of CCR2⁺ monocytes in the lungs.

Statistical analysis performed using One-Way ANOVA. Data are representative of 1 independent experiment of n=5 per group. Standard error bars indicate mean ± SEM; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 and ****p ≤ 0.0001.

To investigate the cellular source of IFN-γ, we performed intracellular cytokine staining in lung cells at day 7 p.i.. Although a number of immune cells such as natural killer (NK) cells, CD4⁺ T cells, CD11b⁺, and CD8⁺ T cells expressed IFN-γ (Figure 4F), CD8⁺ T cells were the most significant expressors of IFN-γ (Figure 4F). Next, we performed antibody-mediated depletion of CD8⁺ T cells (Figure 4G) and determined the protein levels of IFN-γ, CCL2, as well as the frequency of CCR2⁺ monocytes in IAV-infected lungs. The depletion of CD8⁺ T cell resulted in a significant reduction in IFN-γ protein level (Figure 4H), which correlated with reduced protein levels of CCL2 (Figure 4I) and CCR2⁺ monocytes (Figure 4J) in IAV-infected lungs.

Discussion

In this study, we examined the relative distribution of leukocytes in a murine IAV model of acute lung injury using a combined approach of scRNA-Seq and flow cytometry. The major finding of this study is that CD8⁺ T cell-derived IFN-γ regulated the recruitment and inflammatory phenotype of CCR2⁺ monocytes, which exhibited the lineage differentiation trajectory with DCs and macrophages. Together, CCR2⁺ monocytes and monocyte-derivative cells (DCs and macrophages) are the most significant cell populations enriched in IAV-infected lungs. The CCR2^−/− mice or IFN-γ^−/− mice showed significantly reduced disease severity, lung pathology, and enhanced resistance against secondary Spn bacterial infection. Furthermore, the adoptive transfer of IFN-γR1^+/+ (WT) CCR2⁺ monocytes restored the WT-like lung pathology in IAV-infected CCR2^−/− mice, suggesting that despite the differentiation plasticity in CCR2⁺ monocytes, the blockade of the IFN-γ dependent inflammatory phenotype of CCR2⁺ monocytes was sufficient to mitigate the lung damage.

Our scRNA-Seq data reveal that CCR2⁺ monocytes demonstrated a lineage differentiation trajectory with DCs and macrophages. Based on the expression of distinct DEGs, the lineage differentiation revealed two independent states, i.e., early, and late differentiation, with monocytes expressing reduced CCR2 as they differentiate into DCs and macrophages. CCR2⁺ monocytes and monocyte-derived DCs have been shown to cause lung pathology in IAV models (Heung and Hohl, 2019; Lin et al., 2008; Lin et al., 2014). A recent report also showed the accumulation of CCR2⁺ macrophages in a mouse model of SARS-CoV2 infection (Qin et al., 2021). However, host factors that govern the recruitment and inflammatory phenotype of CCR2⁺ monocytes in IAV-infected lungs remain incompletely understood. Our scRNA-Seq data reveal an enrichment of IFN-γ response pathways in CCR2⁺ monocytes and DCs. Prior studies have shown the association of IFN-γ with the development of pathologic M1 macrophage phenotype in infection and inflammation models (Chauhan et al., 2018; Wang et al., 2018; Wang et al., 2019). These observations led us to speculate that IFN-γ contributes to monocyte-mediated lung injury during IAV pathogenesis.

While the pathologic role of IFN-γ has been reported in IAV models (Califano et al., 2018; Liu et al., 2021), there is a lack of data on the relationship of IFN-γ and monocyte-mediated lung injury during IAV pathogenesis. Our data provides evidence demonstrating the significant role of IFN-γ in the recruitment of CCR2⁺ monocytes in the lung via regulating CCL2 response. Furthermore, our data show that IFN-γ deficiency attenuates the monocyte inflammatory response by modulating the expression of a number of inflammatory and cytotoxic genes associated with cellular apoptosis and tissue pathology. Notably, our data show that similar to cytotoxic CD8⁺ T cells, IFN-γ proficient (WT) monocytes express genes associated with cell cytotoxicity, inflammation, and tissue pathology, i.e., granzymes, FasL, and Nos-2 (Cartland et al., 2019; De Trez et al., 2009; Suwanpradid et al., 2017; Trapani and Smyth, 2002). Because monocytes and monocyte-derived myeloid cells establish crosstalk with non-hematopoietic cells to resolve IAV (Brazil et al., 2019), the attenuated monocyte inflammatory response produced by IFN-γ-deficiency likely leads to reduced lung inflammation and injury; resulting in ameliorated barrier integrity in IAV-infected lungs. Our data strongly support this notion, as CCR2 and IFN-γ-deficient mice exhibited reduced lung pathology and improved lung barrier integrity. Notably, despite CCR2 or IFN-γ deficiency, the residual inflammation in the lungs of CCR2^−/− or IFN-γ^−/− mice was sufficient to resolve IAV, suggesting that CCR2 or IFN-γ deficiency had a beneficial effect on the host by limiting tissue damage without impairing viral clearance. However, our findings are limited to the primary IAV model, and the impact of CCR2 or IFN-γ deficiency in secondary protection against IAV re-infection needs to be ascertained. Because airway and lung injury constitute a significant risk factor for the permissiveness of secondary bacterial infections (Ambigapathy et al., 2019; McCullers, 2014), the reduced lung pathology in IAV-infected CCR2^−/− or IFN-γ^−/− mice enhanced resistance against secondary bacterial infection by Spn. However, partial protection observed against secondary Spn disease in CCR2^−/− or IFN-γ^−/− mice necessitates further evaluation of whether the Spn serotype-specific differences contribute to differential protection or susceptibility in the context of prior IAV infection and lung pathology.

The adoptive transfer of IFN-γR1⁺ (WT) CCR2⁺ monocytes restored WT-like lung injury in IAV-infected CCR2^−/− mice, suggesting a direct involvement of IFN-γ receptor signaling in promoting pathologic phenotype of CCR2⁺ monocytes. The pathologic function of IFN-γ requires identifying the cellular source(s) of IFN-γ in IAV-infected lungs. While a number of immune cells were associated with intracellular IFN-γ expression, CD8⁺ T cells served as the most significant source of IFN-γ. Consistent with prior reports (Graham et al., 1993; Nicol et al., 2019), IFN-γ deficiency did not impact the lung CD8⁺ T cell frequency, suggesting that IFN-γ was not crucial to regulating CD8⁺ T cell responses during IAV pathogenesis. Our findings reveal significant cooperativity between CD8⁺ T cells and CCR2⁺ monocytes, with CD8⁺ T cells regulating the recruitment and inflammatory phenotype of CCR2⁺ monocytes via IFN-γ expression.

Limitations of the Study

In this study, we have identified that CD8⁺ T cell derived IFN-γ acts as a significant contributor to host response mediated lung injury during IAV infection. We identified the pathologic function of IFN-γ is mediated via regulating the recruitment and tissue damage associated phenotype of CCR2⁺ inflammatory monocytes. Furthermore, we show that IFN-γ is dispensable to IAV control in a mouse model of primary IAV infection. Nevertheless, the role of IFN-γ in IAV infection warrants further investigation. The limitations of this study is, while CD8⁺ T cells have been shown to cause lung pathology in IAV models, (Duan and Thomas, 2016; Moskophidis and Kioussis, 1998; Myers et al., 2021), mice with impaired CD8⁺ T cell response (deficient in class I major histocompatibility complex-restricted T cells) fail to resolve IAV and exhibit increased mortality (Bender et al., 1992). These findings suggest that CD8⁺ T cells constitute a pivotal anti-viral framework in the lung. However, the dispensability of IFN-γ in IAV infection resolution and unperturbed CD8⁺ T cell recruitment in IFN-γ deficient mice presents a likely scenario of modulating CD8⁺ T cell-mediated lung inflammation for translational utility via inhibiting IFN-γ induced pathologic function of CCR2⁺ monocytes without impairing the viral clearance. Furthermore, the impact of IFN-γ deficiency in long-term IAV-specific CD8⁺ T cell memory generation needs to be ascertained. These questions necessitate further investigations.

STAR Methods

RESOURCE AVAILABILITY

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Dr. Nadeem Khan (mkhan@dental.ufl.edu)

Materials Availability

This study did not generate new unique reagents.

Data and Code Availability

The accession number for the raw and processed sequencing data reported in this paper is GEO: GSE189407. This paper does not report original code. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Animals, cells, and microbial strains.

Wild-type (WT) C57BL/6J, CCR2^−/−, IFN-γ^−/−, and IFN-γR1^−/− mice (6–8 wk old) were purchased from Jackson Laboratory (Bar Harbor, ME) and bred in-house. Ethical approval was granted by the University of North Dakota Animal Care and Use Committee (IACUC) to perform this study (protocol #1808-8). Age- and sex-matched male and female mice were included in the study. At all times, mice received food and water ad libitum. Influenza A Virus (H1N1 A/Puerto Rico/8/1934 or PR8) was purchased from Charles River, Norwich, CT, and a plaque assay (Gaush and Smith, 1968) was performed to determine the plaque forming units (PFUs) for IAV infection. Streptococcus pneumoniae (Spn) serotype 6A strain BG7322 was obtained from the Rochester General Hospital Research Institute (RGHRI) as previously reported (Khan et al., 2017).

IAV and IAV/Spn co-infection models.

For IAV infection, mice were anesthetized (4%v/v isoflurane/oxygen) and intranasally administered with 100 μl of sterile PBS or 250 PFUs of PR8 in 100 μl PBS. Mice were monitored daily for weight loss and bronchiolar lavage (BAL) fluid was collected via the intratracheal administration of 1 mL of ice-cold PBS. The BAL fluid was centrifuged at 800 × g for 10 minutes, and the supernatant was preserved at −20°C for further analysis. The lungs were perfused with sterile PBS, aseptically removed, and processed accordingly for downstream applications. For co-infection studies, mice were intranasally inoculated with 200 CFUs of Spn serotype 6A in 50 μl of sterile PBS 7 days after PR8 infection and euthanized 48 h after Spn co-infection. To determine Spn burden, homogenized lungs were serially diluted in sterile PBS and plated onto Triple Soy Agar II medium plates containing 5% of sheep blood (BD, Franklin Lakes, NJ) and incubated at 37°C overnight. Bacterial CFUs were enumerated and expressed as CFU/ml of homogenized lungs.

CD8 T cell depletion.

Mice were intraperitoneally (i.p.) injected with 200 μg of anti-CD8⁺ T cell monoclonal antibody (clone 2.43, BioXCell, Lebanon, NH) every other day, starting 24 h prior to IAV infection. The CD8⁺ T cell depletion efficiency was confirmed by flow cytometry.

ALI cell culture.

Human bronchial epithelial cells (HBECs) were grown in air-liquid interface as described earlier (Jiang et al., 2018). Briefly, 0.2 mL of HBECs were seeded in type IV collagen-coated permeable transwell supports (6.5 mm or 24 mm, 0.4-μm polyester membrane, Corning, NY) at a density of 1×10⁵ cells/ml. Then, 0.5 mL of fresh differentiation medium (Stem Cell Technologies, Vancouver, BC) was added to the lower chamber. After 48 h, medium was removed from the apical chamber and cells were maintained under ALI conditions for ten days. Then, HBECs were infected for 24 h with PR8 (2:1 MOI) and/or treated for 24 h with IFN-γ (20μg/mL). The levels of CCL2 were measured in supernatant using the LEGENDplex cytokine bead array (Biolegend, San Diego, CA) per manufacturer’s instructions.

METHOD DETAILS

IAV titrations.

For TCID₅₀, Madin-Darby Canine Kidney (MDCK) cells were seeded in 96-well plate at a density of 1.5×10⁵ cells/ml in 100ul of complete DMEM and incubated for 16–18 h at 37°C in 5% C_O2 incubator. After reaching 90–100% confluency, the medium was removed, and 10-fold serial dilutions of lung homogenate (partial lobe of right lung) supernatants were made in infection medium and subsequently transferred to cell plates. After the observation of cytopathic effects (within 3–6 days after infection), infection medium was removed, and cells were rinsed with pre-warmed PBS. Subsequently, 100 μl of 10% paraformaldehyde in PBS per well was added for 10 min at room temperature. After removal of fixative, cells were washed with PBS and 100 μl of 1% Gram Hacker’s crystal violet was added for 10 min at room temperature. Crystal violet was removed, plates were rinsed with PBS once, tapped on counter to remove excess, and TCID₅₀ was calculated using the Spearman-Karber method, as described earlier (Benton et al., 2001). For PCR quantitation of IAV load, viral RNA was extracted from the lung homogenates of mock and PR8-infected mice. Viral RNA was quantitated using qPCR targeting Influenza A Matrix (M) gene as previously described (Khan et al., 2017; Piralla et al., 2013). The data were expressed as the IAV (PR8) copy number/μL of viral RNA.

Flow cytometry, intracellular staining, and cytokine analysis.

The lungs were aseptically excised and digested with 5 mL of media containing collagenase (25 U/ml) and DNase (0.5 mg/mL) for 30 minutes and passed through 70 μm cell strainers. Red Blood Cells (RBCs) were lysed using ammonium–potassium–chloride (ACK) lysis buffer (Life Technologies, Carlsbad, CA), and 1×10⁶ cells were stained with live-dead (Ghost Dye) (Tonbo Biosciences, San Diego, CA) following manufacturer’s recommendation. The surface staining was performed for 30 min at room temperature using anti-mouse antibodies against CD45 (30-F11), CD3 (145–2C11), CD4 (RM4–5), CD8 (53–6.7), CD11b (M1/70), Ly6C (HK1.4), Ly6G (1A8), NK1.1 (PK136), IA/I-E (M5/114.15.2), EP-CAM (G8.8), CD11c (N418), F4/80 (BM8), and CCR2 (SA203G11); all from BioLegend (San Diego, CA). For T-Distributed Stochastic Neighbor Embedding (tSNE) analysis, samples were down sampled to 15,000 events, concatenated, and tSNE was generated using FlowJO™ (BD) (Toghi Eshghi et al., 2019). Cell populations were gated according to strategy presented in Figure S3A. For intracellular cytokine staining (ICS), cells were permeabilized and stained with the BD Cytofix/Cytoperm™ solution following the manufacturer’s recommendations. A BD FACSymphony A3 flow cytometer was used to acquire 100,000 events, and the data were analyzed using FlowJo (BD).

For multiplex cytokine analysis, the lungs were homogenized in tissue protein extraction reagent and centrifuged at 10,000 × g for 10 min, and the supernatants were collected. The cytokine analysis was performed in BAL fluids and lung homogenate using the LEGENDplex Mouse Multianalyte Inflammation Panel (BioLegend), following the manufacturer’s recommendation. Samples were acquired on a BD FACSymphony A3 flow cytometer, and data were analyzed using LEGENDplex V8.0 Data Analysis Software (BioLegend).

Lung Histology.

After perfusion, the left lobe was fixed in 10% neutral buffered (pH 7.4) formalin for 24 h at room temperature. The lung tissues were embedded in paraffin, sliced into 5 μm sections to reveal the maximum longitudinal view of the main intrapulmonary bronchus of the left lobe, and stained with hematoxylin and eosin (H&E). The H&E-stained slides were coded and evaluated by three independent pathologists blinded to the experimental groups. Each tissue section was scored based on a scale of 0–4 with increments of 0.5 (15), with 0 as no inflammation and 4 as highest degree of tissue infiltration of immune cells. Vascular damage was scored on a range of 0–4 based on vacuolation of endothelial cells and separation from underlying musculature. Representative histological images were acquired using a NanoZoomer 2.0-HT Brightfield Fluorescence Slide Scanning System (Hamamatsu Photonics, Japan) and analyzed using NDP.view2 Viewing Software (Hamamatsu). The inflammation and vascular damage scores were performed by a blinded observer.

Immunofluorescence.

Lung sections were deparaffinized, rehydrated, and antigen retrieval was performed in sodium-citrate buffer pH 6.0. Non-specific bindings were blocked in 5% normal goat serum, followed by the overnight incubation with primary antibodies against E-cadherin (1:200, Cell Signaling) and ZO-1 (1:200, Thermo Fisher, Waltham, MA). Subsequently, the lung sections were incubated at RT for 30 minutes with goat anti-rabbit Alexa Fluor-488 (AF) conjugated secondary antibody (Thermo Fisher) and mounted with ProLong Diamond Antifade Mountant with 4′,6-diamidino-2-phenylindole (DAPI). Sections were visualized by TIRF microscopy on an Olympus microscope with a Hamamatsu RCA flash 4.0 camera and quantified using FIJI.

Albumin ELISA and LDH Assay.

Albumin concentrations in BAL fluid samples were measured using a murine specific albumin ELISA kit (ALPCO Diagnostics, Salem, NH) according to the manufacturer’s instructions. BAL fluid samples were diluted 1:10,000, and the total albumin was measured at 450 nm with a Synergy HT (Bio-Teck, Vermont, CA). An LDH Assay Kit (Abcam, Cambridge, UK) was used to measure LDH activity, following manufacturer’s instructions.

qPCR.

Total RNA was extracted from lung tissues using the RNeasy Plus Mini Kit (QIAGEN, Valencia, CA). cDNA was synthesized using SensiFast cDNA Synthesis Kit (Bioline, Taunton, MA), and PowerUP SYBR Green Master Mix (Life Technologies) was used for PCR template. Changes in specific gene expression were normalized to housekeeping gene (β-actin) from each sample, and fold differences were calculated using the 2(−ΔΔCt) values.

Adoptive transfer.

For adoptive transfer experiments, bone marrow was isolated from naïve WT and IFN-γR1^−/− mice. A single-cell suspension was prepared, and CCR2⁺ monocytes (CD11b⁺Ly6C⁺CCR2⁺) were stained with live/dead (Ghost Dye™) (Tonbo Biosciences), CD11b (M1/70), Ly6C (HK1.4), and CCR2 (SA203G11, BioLegend) and FACS sorted using BD FACSAria™. The FACS sorted cells were stained with CellTrace CFSE Cell Proliferation Kit (Thermofisher) following the manufacturer’s recommendations. CCR2⁺ monocytes (1×10⁶ cells/mouse) from WT or IFN-γR1^−/− mice were retro-orbitally injected into sex- and age-matched CCR2^−/− mice at 1-day p.i. infection. Mice were euthanized at day 7 p.i.

scRNA-Seq.

Mock or IAV-infected mice (250 PFU_S) were euthanized at 7 days p.i. and whole lungs were perfused with cold PBS. Lungs were aseptically removed and placed in 1 mL complete media (10% FBS, 1% Pen/Strep in RPMI). A single-cell suspension was prepared as noted above. For CD45⁺ cell isolation, mock and IAV-infected lungs (n=5/group) were pooled, stained for CD45 (30-F11, BioLegend) for 1 h at RT, and sorted using the BD FACSAria™ into 40% FBS/RPMI. After sorting, 2 ×10⁶ cells were washed and resuspended in medium containing 10% DMSO, 20% FBS and RPMI. Cells were slowly frozen in Mr. Frosty and shipped to SingulOmics (NYC, NY).

CellRanger (v5.0) was used to aligned scRNA-Seq data to the mm10 mouse genome and quantify the gene expression. Then, the output files were read into Seurat v4 (Stuart et al., 2019) and cells with low quality were further excluded from the downstream analysis based on filtering by the following criteria (nFeature_RNA < 200 & nFeature_RNA > 7000 & percent.mt > 0.75). Dimension reductions include principal component analysis (PCA), clustering, and tSNE were then performed on the integration data. Differentially expressed genes were identified for each cluster with FindAllMarkers and cell types were assigned by using the CellMarker and PanglaoDB database as the reference, with manual correction. A single cell trajectory and Monocle states were constructed by using Monocle2 (Trapnell et al., 2014). To assign the Gene Ontology (GO) biological activity for individual cells, we applied the Gene Set Variation Analysis (GSVA) analyses with the GSVA package (Hanzelmann et al., 2013). Wilcoxon test was used to identify the significant pathways between different types of cells.

RNA-Seq.

Mock or IAV-infected mice were euthanized at 7 days p.i. and lungs were perfused with cold PBS as described above. A single-cell suspension as prepared and cells were pooled, and monocyte isolation was performed using the Monocyte Isolation Kit (Miltenyi) as per the manufacturer’s protocol. The purity (CD11b⁺Ly6C⁺) and viability were confirmed by flow cytometry using monocyte putative markers, CD11b⁺ and Ly6C⁺. The purity and the viability of isolated monocytes were > 90%. Total RNA was isolated from purified monocytes using the RNeasy Plus Mini Kit (QIAGEN, Valencia, CA). Then, PolyA RNA-Seq libraries were prepared using a NEBNext Ultra II RNA-Seq library kit as per manufacturer’s instructions. Prior to RNA-Seq, the samples were subjected to quality control (QC) on a Bioanalyzer 2100 RNA Nano to determine the RNA quantity (samples with a RIN ≥7 were considered to have passed the QC step). Samples were run on one NovaSeq 6000 lane. Preliminary QC analysis of fastq files was performed with FastQC v0.11.8 (McDermott and Frenkel, 2001). Adapters were trimmed using Trimmomatic v0.39 (Bolger et al., 2014), and reads were aligned to a mouse genome (mm10) with HISAT2 (Kim et al., 2015). FeatureCounts v1.4.6 (Liao et al., 2014) was used to count reads mapped to individual genes. Differentially expressed genes (DEGs) between groups were identified using the DESeq2 v1.24.0 (Love et al., 2014) with a significance cutoff of minimum absolute log₂ fold change of 1 and adjusted p-value <0.05. Enrichment Analysis (GSEA) for Kyoto Encyclopedia of Genes and Genomes (KEGG) were performed by using richR package (https://github.com/hurlab/richR) and the adjust p value < 0.05 was chosen as the cutoff value to select significant KEGG.

QUANTIFICATION AND STATISTICAL ANALYSIS.

The data are presented as mean ± SEM. Comparisons between two groups were made using Students t-test. Experiments containing multiple groups were analyzed using a One-Way ANOVA followed by a Tukey’s post-hoc for comparisons made between groups. Statistical analysis was performed using Prism v9.3.1 (GraphPad Software, La Jolla, CA). Values representing number of replicates or samples in each group are listed in the corresponding figure legend. Survival curves were analyzed using the Mantel-Cox log-rank test. For all tests, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Key Resources Table.

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Rat Brilliant Violet 711™ anti-mouse CD4 Antibody clone RM4-5	BioLegend	Cat#100557
Hamster PerCP/Cyanine5.5 anti-mouse CD11c Antibody clone N418	BioLegend	Cat#117328
Rat Alexa Fluor® 700 anti-mouse I-A/I-E Antibody clone M5/114.15.2	BioLegend	Cat#107622
Rat Brilliant Violet 650™ anti-mouse F4/80 Antibody clone BM8	BioLegend	Cat#123149
Rat APC anti-mouse CD326 (Ep-CAM) Antibody clone G8.8	BioLegend	Cat#118214
E-Cadherin (24E10) Rabbit mAb #3195	Cell Signaling Technology	Cat#3195S
Rabbit ZO-1 Polyclonal Antibody (ZMD.437)	ThermoFisher Scientific	Cat#40-2300
Mouse PE/Cyanine5 anti-mouse NK-1.1 Antibody clone PK136	BioLegend	Cat#108716
Rat InVivoPlus anti-mouse CD8α clone 2.43	BioXCell	Cat#BP0061
Rat APC/Cyanine7 anti-mouse/human CD11b Antibody clone M1/70	BioLegend	Cat#101225
Rat Brilliant Violet 711™ anti-mouse Ly-6C Antibody clone HK1.4	BioLegend	Cat#128037
Rat PE anti-mouse IFN-γ Antibody clone XMG1.2	BioLegend	Cat#505808
Rat PE/Cyanine7 anti-mouse CD192 (CCR2) Antibody clone SA203G11	BioLegend	Cat#150611
Goat anti-Rabbit IgG (H+L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 488	ThermoFisher	Cat#A-11008
Rat FITC anti-mouse Ly-6G Antibody clone 1A8	BioLegend	Cat#127605
Rat Pacific Blue™ anti-mouse CD45 Antibody clone 30-F11	BioLegend	Cat#103126
Ghost Dye™ Violet 510	TONBO Biosciences	Cat#13-0870-T500
Hamster APC/Cyanine7 anti-mouse CD3ε Antibody clone 145-2C11	BioLegend	Cat#100330
Rat PE/Cyanine7 anti-mouse CD8a Antibody clone 53-6.7	BioLegend	Cat#100722
Bacterial and Virus Strains
H1N1 A/Puerto Rico/8/1934 or PR8	Charles River	Material#10100374 Liquid
Streptococcus pneumoniae serotype 6A strain BG7322	Rochester General Hospital Research Institute
Chemicals, Peptides, and Recombinant Proteins
ProLong™ Diamond Antifade Mountant	ThermoFisher Scientific	Cat#P36961
Paraformaldehyde	Millipore Sigma	Cat#158127
RNEasy Plus Mini Kit	QIAGEN	Cat#74134
Gibco ACK Lysing Buffer	Life Technologies	Cat#A1049201
SensiFast cDNA Synthesis Kit	Bioline	Cat#BIO-65054
Recombinant Human IFN-gamma Protein	R&D Systems	Cat#285-IF-100
PowerUP SYBR Green Master Mix	Life Technologies	Cat#A25741
CellTrace™ CFSE Cell Proliferation Kit	ThermoFisher	Cat#C34554
T-PER™ Tissue Protein Extraction Reagent	Thermo Scientific	Cat#78510
PneumaCult™-Ex Plus Medium	Stemcell Technologies	Cat#05040
Critical Commercial Assays
BD Cytofix/Cytoperm™	BD	Cat#554714
LEGENDplex™ Mouse Inflammation Panel (13-plex) with V-bottom Plate	LegendPlex	Cat#740446
LEGENDplex™ Mouse Proinflammatory Chemokine Panel (13-plex) with V-bottom Plate	BioLegend	Cat#740451
Mouse Albumin ELISA	ALPCO	Cat#41-ALBMS-E01
LDH Assay Kit (Cytotoxicity)	Abcam	Cat#ab65393
Deposited Data
Raw and analyzed data	This paper	GSE189407
Experimental Models: Cell Lines
Primary Bronchial/Tracheal Epithelial Cells; Normal, Human	ATCC	Cat#PCS-300-01
Experimental Models: Organisms/Strains
B6.129S4-Ccr2tm1Ifc/J	The Jackson Laboratory	Strain#004999
B6.129S7-Ifngr1tm1Agt/J Mouse 6–8 weeks of age	The Jackson Laboratory	Strain#003288
C57BL/6J (wild-type) Mouse 6–8 weeks of age	The Jackson Laboratory	Strain#000664
B6.129S7-Ifngtm1Ts/J	The Jackson Laboratory	Strain#002287
Oligonucleotides
SOCS-1-F (5′- CAAGATGTTGAGGAACGAGTTCA -3′)	IDT	N/A
IRF-1-F (5′- GTTGTGCCATGAACTCCCTG -3′)	IDT	N/A
FasL-F (5′- TCCGTGAGTTCACCAACCAAA -3′)	IDT	N/A
STAT-1-F (5′- GCTGCCTATGATGTCTCGTTT -3′)	IDT	N/A
Granzyme B-F (5′- CCACTCTCGACCCTACATGG -3′)	IDT	N/A
IFN-g-F (5′- AACGCTACACACTGCATCTTGG -3′)	IDT	N/A
B-actin-F (F: 5′- TCCGGCACTACCGAGTTATC-3′)	IDT	N/A
Software and Algorithms
FIJI	Open Source	https://imagej.net/software/fiji/
Olympus TIRF Microscope (445nm, 491nm, 514nm, and 561nm) with a Hammamatsu RCA flash 4.0 camera	Olympus	https://www.olympus-lifescience.com/en/landing/olympus-microscopes/?gclid=Cj0KCQiAxc6PBhCEARIsAH8Hff26jFB7afXCWzdsaDp18NQsRRxaf4MBiJLwnL7XbZvjlkJDiLg1CTwaAjYUEALw_wcB
FACSAria Cell Sorter	BD
FACSymphony A3 Flow Cytometer (488nm/ 405nm/ 561nm/ 637nm lasers)	BD	https://www.bdbiosciences.com/en-us/products/instruments/flow-cytometers/research-cell-analyzers/bd-facsymphony-a3
FlowJo™	Becton, Dickinson and Company	https://www.flowjo.com/solutions/flowjo/downloads
LEGENDplex V8.0 Data Analysis Software	BioLegend	https://www.biolegend.com/en-us/legendplex?gclid=Cj0KCQiAxc6PBhCEARIsAH8Hff2qBjYyag-4hfhaZffCO-HU0DihmjDZLhiIVs5-8wpIDZygQHMqX_IaAuCzEALw_wcB
NanoZoomer 2.0-HT Brightfield Fluorescence Slide Scanning System	Hamamatsu	https://nanozoomer.hamamatsu.com/jp/en/scanner/search.html
GraphPad Prism version 9.3.1	GraphPad	https://www.graphpad.com/scientific-software/prism/
NDP.view2 Viewing software U12388-01	Hamamatsu	https://nanozoomer.hamamatsu.com/jp/en/software/search/U12388-01/index.html
CellRanger version 5.0	10x Genomics	https://support.10xgenomics.com/single-cell-gene-expression/software/overview/welcome
Seurat version 4.0	Stuart et al., 2019	https://satijalab.org/seurat/install.html
Monocle	Trapnell et al., 2014	http://cole-trapnell-lab.github.io/monocle-release/
GSVA	Hanzelmann et al., 2013	https://www.bioconductor.org/packages/release/bioc/html/GSVA.html
FastQC	McDermott and Frenkel, 2001	https://www.bioinformatics.babraham.ac.uk/projects/fastqc/
Trimmomatic version 0.39	Bolger et al., 2014	http://www.usadellab.org/cms/?page=trimmomatic
HISAT2	Kim et al., 2015	http://daehwankimlab.github.io/hisat2/
FeatureCounts version 1.4.6	Liao et al., 2014	http://subread.sourceforge.net/
DESeq2 version 1.24.0	Love et al., 2014	https://bioconductor.org/packages/release/bioc/html/DESeq2.html
richR	Open Source	https://github.com/hurlab/richR