Mouse model of SARS-CoV-2 reveals inflammatory role of type I interferon signaling

Abstract

Severe acute respiratory syndrome-coronavirus 2 (SARS-Cov-2) has caused over 13,000,000 cases of coronavirus disease (COVID-19) with a significant fatality rate. Laboratory mice have been the stalwart of therapeutic and vaccine development; however, they do not support infection by SARS-CoV-2 due to the virus's inability to use the mouse orthologue of its human entry receptor angiotensin-converting enzyme 2 (hACE2). While hACE2 transgenic mice support infection and pathogenesis, these mice are currently limited in availability and are restricted to a single genetic background. Here we report the development of a mouse model of SARS-CoV-2 based on adeno-associated virus (AAV)-mediated expression of hACE2. These mice support viral replication and exhibit pathological findings found in COVID-19 patients. Moreover, we show that type I interferons do not control SARS-CoV-2 replication in vivo but are significant drivers of pathological responses. Thus, the AAV-hACE2 mouse model enables rapid deployment for in-depth analysis following robust SARS-CoV-2 infection with authentic patient-derived virus in mice of diverse genetic backgrounds.

Graphical abstract

Figure 1.

AAV-hACE2 transduction allows for productive SARS-CoV-2 infection in vivo. (a) Schematic of experimental plans. C57BL/6J mice were transduced intratracheally with an AAV coding for hACE2 (AAV-hACE2) or control (AAV-GFP or PBS) and infected with SARS-CoV-2 2 wk after. Lung and blood samples were collected at days 2, 4, 7, and 14 for analysis. (b) Viral RNA from lung homogenates were measured using quantitative PCR against SARS-CoV-2. (c) Viral titers from lung homogenates were performed by plaque assay on VeroE6 cells (AAV-hACE2 values noted as mean ± SEM from three independent experiments; n = 7 at 2 and 14 DPI; n = 6 at 4 and 7 DPI. Control values noted as mean ± SEM from two independent experiments; n = 4 at 2, 7, and 14 DPI; n = 3 at 4 DPI). (d) Frozen lung tissue was stained for SARS-CoV-2 N protein (red) and epithelial cells (EpCAM, green). (e) Fixed lung tissue was paraffin-embedded and stained with H&E. Magnified panels highlight leukocyte infiltration and perivenular inflammation. (f) Images from panel e were scored by a pulmonary pathologist for perivenular inflammation (n = 2). (g) At 2 DPI, single-cell suspensions of lung were analyzed by flow cytometry. Data are shown as frequency of CD45⁺ cells (monocyte-derived macrophages, Ly6Chi monocytes, and neutrophils), frequency of parent cells (CD44⁺CD69⁺CD4⁺ T cells, CD44⁺CD69⁺CD8⁺ T cells, and CD69⁺ NK cells), or mean fluorescence intensity of CD64 (Ly6Chi monocytes; AAV-hACE2 and control values noted as mean ± SEM from two independent experiments, n = 4). (h) Serum antibodies were measured against spike protein using an ELISA. (i) Day 7 and 14 sera from panel h were used to perform a PRNT on VeroE6 cells incubated with SARS-CoV-2 (AAV-hACE2 noted as mean ± SEM from two independent experiments, n = 4; control value, n = 1). P values were calculated by two-tailed unpaired Student’s t test. *, P < 0.05; **, P < 0.01; ***, P < 0.005. Scale bars, 100 µm.

Figure S1.

AAV-hACE2 infection makes WT mice susceptible to SARS-CoV-2. (a) Immunofluorescence staining of ACE2 in mice transduced with AAV-hACE2, 20 d after transduction. (b) Flow cytometry gating strategy for Fig. 1 g; Fig. 3, g–i; Fig S1 c; and Fig S3, a–c. (c) Representative flow cytometry plots for Fig. 1 g T cells and NK cells. (d) Representative flow cytometry plots for Fig. 1 g myeloid cells. (e) Different macrophage populations in the lungs of SARS-CoV-2–infected control (AAV-GFP or Mock) or AAV-hACE2 mice 2 DPI by flow cytometry. Data are pooled from two independent experiments, n = 4 mice per group, and are represented as mean ± SEM. P values were calculated by two-tailed unpaired Student’s t test. **, P < 0.01. (f) Sera of mice were collected 7 DPI with SARS-CoV-2, and limiting dilutions were made to measure reactivity against S1 protein of SARS-CoV-2 using ELISA. (g) Sera of mice were collected 14 DPI with SARS-CoV-2, and limiting dilutions were made to measure reactivity against S1 protein of SARS-CoV-2 using ELISA. Scale bar, 100 µm. FSC-A, forward scatter area; FSC-H, forward scatter height; FSC-W, forward scatter width; SSC-A, side scatter area; SSC-H, side scatter height; SSC-W, side scatter width.

Figure S2.

Myeloid cell infiltration in SARS-CoV-2–infected mice. (a) Flow cytometry gating strategy for Fig. 1 g; Fig. 3, d–f; Fig S1, b and e; and Fig. S2, b–d. (b) Relative percentage of different cell populations in indicated knockout mice infected with SARS-CoV-2 2 DPI (WT [light gray] nontransduced, noninfected values from single experiment, n = 2; WT [dark gray] AAV-hACE2 and not infected values from two independent experiments, n = 4; WT [blue] AAV-hACE2 and infected values from two independent experiments, n = 4; IFNAR^−/− values from three independent experiments, n = 6; and IRF3/7^−/− values from two independent experiments, n = 5. (c) Representative flow cytometry plots for Fig. 3, d and e. (d) Relative percentage of myeloid cell populations in WT mice transduced with AAV-hACE2 at 2, 4, and 7 d after SARS-CoV-2 infection (WT [light gray] nontransduced, noninfected values from single experiment, n = 2; WT [dark gray] AAV-hACE2 and not infected values from two independent experiments, n = 4; WT [blue] AAV-hACE2 and infected values from two independent experiments, n = 4, at each time point. FSC-A, forward scatter area; FSC-H, forward scatter height; FSC-W, forward scatter width; SSC-A, side scatter area; SSC-H, side scatter height; SSC-W, side scatter width.

Figure S3.

T cell and NK cell infiltration in SARS-CoV-2–infected mice. (a) Representative flow cytometry plots for Fig. 3, g–i. (b) Relative percentage of different lymphoid cell populations in different knockout mice infected with SARS-CoV-2 2 DPI (WT [light gray] nontransduced, noninfected values from single experiment, n = 2; WT [dark gray] AAV-hACE2 and not infected values from two independent experiments, n = 4; WT [blue] AAV-hACE2 and infected values from two independent experiments, n = 4; IFNAR^−/− values from three independent experiments, n = 6; and IRF3/7^−/− values from two independent experiments, n = 5). (c) Relative percentage of lymphoid cell populations in WT mice transduced with AAV-hACE2 at 2, 4, and 7 d after SARS-CoV-2 infection (WT [light gray] nontransduced, noninfected values from single experiment, n = 2; WT [dark gray] AAV-hACE2 and not infected values from two independent experiments, n = 4; WT [blue] AAV-hACE2 and infected values from two independent experiments, n = 4, at each time point).

Figure 2.

AAV-hACE2 mice infected with SARS-CoV-2 show similar IFN signatures as COVID-19 patients. (a) Volcano plot showing differential expression (DE) of genes from whole lungs of mice infected with SARS-CoV-2 with and without AAV-hACE2 at 2 DPI. Gray indicates significantly differentially up-regulated genes, and blue indicates subsets of genes that are known ISGs. (b) Significantly up-regulated genes were put into Interferome (https://www.interferome.org) to identify how many genes are stimulated by type I, type II, or type III IFNs. (c) Gene Ontology Enrichment Analysis was performed on significantly up-regulated genes to identify enriched cellular processes. (d) Up-regulated gene list from human samples (Blanco-Melo et al., 2020) was graphed (left panel) and used to perform hierarchical clustering for differentially expressed genes from lungs of AAV-hACE2 SARS-CoV-2–infected mice (blue, mouse ISGs; gray, other genes). (e) Subset of differentially expressed genes in panel a that were significantly up-regulated in lungs of COVID-19 patients (from Blanco-Melo et al., 2020).

Figure 3.

AAV-hACE2 mice infected with SARS-CoV-2 exhibit type I IFN-dependent immune cell infiltration. C57BL/6J (WT), IFNAR knockout, and IRF3/7 double knockout mice were transduced intratracheally with an AAV encoding hACE2 (AAV-hACE2) and infected with SARS-CoV-2 2 wk later. (a) Viral titers in the lungs of mice were measured using quantitative PCR against SARS-CoV-2. (b) Lung homogenates titered on VeroE6 cells (control and WT were from the same experiment as Fig. 1; IFNAR^−/− values from three independent experiments, n = 6 at 2, 4, and 7 DPI, and n = 4 at 14 DPI; IRF3/7^−/− values from two independent experiments, n = 5 at 2 and 7 DPI, and n = 4 at 4 and 14 DPI). (c) Heat map of top 100 up-regulated genes in SARS-CoV-2–infected C57BL/6J (WT) mice transduced with AAV-hACE2 versus SARS-CoV-2–infected AAV-GFP–transduced mice. Relative expression of these genes in noninfected/nontransduced mice and IFNAR^−/− and IRF3/7^−/− AAV-hACE2–transduced and SARS-CoV-2–infected mice. (d–i) At 2 DPI, lungs of mice were made into single-cell suspensions for flow cytometry (WT [light gray] nontransduced, noninfected values from single experiment, n = 2; WT [dark gray] AAV-hACE2 and not infected values from two independent experiments, n = 4; WT [blue] AAV-hACE2 and infected values from two independent experiments, n = 4; IFNAR^−/− values from three independent experiments, n = 6; and IRF3/7^−/− values from two independent experiments, n = 5). P values were calculated by one-way ANOVA with Tukey’s multiple comparison test. *, P < 0.05; **, P < 0.01; ***, P < 0.005; ****, P < 0.001. ns, not significant.