Prompt Antiviral Action of Pulmonary CD8+ TRM Cells Is Mediated by Rapid IFN-γ Induction and Its Downstream ISGs in the Lung

Figure 1. Mouse infection models demonstrated the cross-protectivity of T-cell memory response to virus infection. (A–C) Exploration of harmless H9N2 virus as a live vaccine against pathogenic H7N9 virus. The sequential intranasal infections of naïve mice with H9N2 and H7N9 viruses were performed according to a 4-week schedule (A). Following H7N9 challenge, the mice were daily monitored for weight changes with the control (ctrl) group representing the animal group only receiving PBS (B) and survival (C). (D–F) An experimental approach combining adoptive transfer and sequential virus infections to demonstrate cross protectivity of memory CD8+ T cell via cross-recognizing a single epitope. The experimental scheme was illustrated in (D). In brief, mice were adoptively transferred with 5 ×104 naïve P14 CD8+ T cells, then i.n. infected with sublethal dose of PR8 or PR8-gp33 as the prime, and 30 days later challenged with LCMV Armstrong at 2 × 105 of plaque‐forming units (PFU). Shown in (E) was lung viral load determination at day 5 post LCMV Armstrong challenge, and in (F) were representative images of H&E-stained lung sections sampled at the indicated time points post challenge. Scale bar…

Figure 2. Lung CD8+ TRM cells have high potential to afford cross-protein against virus infection. (A, B) Mice primed with PBS or H9N2 infection were challenged with a lethal dose of H7N9 virus following the schedule described in Figure 1A , except that intraperitoneal injection of anti-CD8a or control antibody was applied at 1 day prior to H7N9 challenge as well as day 1 and day 3 after the challenge (A). H7N9 infection-induced weight changes were daily monitored (B). (C) Mice received transfusion of serum from H9N2- or H7N9-infected mice and were subsequently challenged with a lethal dose of H7N9. The survival rate was determined daily after the challenge. (D–F) Mice were seeded with P14 cells by adoptive transfer and infected with PR8-gp33 or PR8 virus. Thirty days later, the abundance of P14 cells in the bronchoalveolar lavage (BAL) and spleen of resulting memory mice were analyzed by flow cytometry, with representative flow cytometry plots shown in (D), and pooled absolute and percent (among total CD8+ population) enumeration data shown in (E). The BAL fluids were harvested from memory mice (n = 5) and i.n. transferred into naïve mice, which were then challenged with i.n. infection of…

Figure 3. Activated lung TRM cells rapidly upregulate IFN-γ, associated with induction of ISG expression inside themselves and in the lung. Shown in (A) is the experimental scheme. Naïve mice received P14 cells by adoptive transfer and were then infected with a sublethal dose of PR8-gp33 virus or control PR8 virus, 30 days later challenged with LCMV Armstrong. The lung and BAL fluid were harvested from animals at 0 h, 3 h, 6 h, 12 h, 24 h, and 48 h after the challenge. P14 cells were isolated from the BAL fluid through flow cytometry. Microarray analyses were performed on RNAs pooled from three lungs or P14 cells isolated from three BAL samples. The revealed dynamic expression pattern of different IFNs and downstream ISGs were respectively shown as the upper and lower panel for lung CD8+ TRM cells (B) and lungs from the two treatment groups (C, D). Shown in (E) is the heatmap constructed by the ratio of mRNA values obtained in (C, D).

Figure 4. Validation of microarray data by RT-PCR and RNAscope in situ hybridization. (A, B) Animals were treated following the same scheme shown in Figure 3 , and lungs were collected at the indicated time points after LCMV Armstrong challenge for quantitative RT-PCR analysis of mRNA expression of four representative ISGs (A) and lung IFN-γ (B). (C, D) Dynamics of lung IFN-γ mRNA determined by RNAscope in situ hybridization, with representative lung sections shown in (C) and quantification data shown in (D). Data are representative of two or three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; two-way ANOVA (A, B, D).

Figure 5. IFN-γ plays important roles in mediating the protective activity of lung CD8+ TRM via contribution to induction of downstream ISGs and reduction of inflammation signals. (A, B) Naïve mice were subjected to P14 seeding and sequential infection with PR8-gp33/PR8 and LCMV Armstrong the same as described in Figure 3A . BAL was sampled with 1 ml of PBS at 12 h, 24 h, and 48 h after LCMV Armstrong challenge; the protein levels of IFN-γ and two type I interferons, and those of two representative proinflammatory cytokines, IL-6 and MCP-1, were assessed by cytometric bead array (CBA), respectively, shown in (A, B). (C–H) Naïve mice were seeded with P14 cells, subjected to sublethal PR8-gp33 infection for memory induction, and then challenged with LCMV Armstrong while receiving an anti-IFN-γ IgG or an isotype-matched control IgG or intranasally at both 1 day before and 6 h after the challenge (C). At various time points post challenge, lung and BAL were collected and analyzed respectively for mRNA expression of seven ISGs by quantitative RT-PCR (D, E), and protein expression of IFN-α and IFN-β (F), or two representative proinflammatory cytokines, IL-6 and MCP-1 by CBA assay (G). (H) The…