Administration of aerosolized SARS-CoV-2 to K18-hACE2 mice uncouples respiratory infection from fatal neuroinvasion

Abstract

The development of a tractable small animal model faithfully reproducing human coronavirus disease 2019 pathogenesis would arguably meet a pressing need in biomedical research. Thus far, most investigators have used transgenic mice expressing the human ACE2 in epithelial cells (K18-hACE2 transgenic mice) that are intranasally instilled with a liquid severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) suspension under deep anesthesia. Unfortunately, this experimental approach results in disproportionate high central nervous system infection leading to fatal encephalitis, which is rarely observed in humans and severely limits this model’s usefulness. Here, we describe the use of an inhalation tower system that allows exposure of unanesthetized mice to aerosolized virus under controlled conditions. Aerosol exposure of K18-hACE2 transgenic mice to SARS-CoV-2 resulted in robust viral replication in the respiratory tract, anosmia, and airway obstruction but did not lead to fatal viral neuroinvasion. When compared with intranasal inoculation, aerosol infection resulted in a more pronounced lung pathology including increased immune infiltration, fibrin deposition, and a transcriptional signature comparable to that observed in SARS-CoV-2–infected patients. This model may prove useful for studies of viral transmission, disease pathogenesis (including long-term consequences of SARS-CoV-2 infection), and therapeutic interventions.

Fig. 1.

Intranasal inoculation, but not aerosol exposure, of SARS-CoV-2 leads to fatal neuroinvasion in K18-hACE2 transgenic mice. (A) Illustration of the two modalities used to infect K18-hACE2 mice with SARS-CoV-2. On the left, intranasal injection (IN) is shown. On the right, representation of an unanesthetized mouse placed in the nose-only Allay restrainer on the inhalation chamber is shown. In red, aerosolized virus with a particle size of ~4 μm; in light-blue, primary flow set to 0,5 L/min/port; in grey, mouse breathing outflow (see Methods). (B) Schematic representation of the experimental setup. K18-hACE2 mice were infected with a target dose of 1 × 10⁵ TCID₅₀ of SARS-CoV-2 through intranasal (IN) administration or through aerosol (AR) exposure. Lung, brain, nasal turbinates, olfactory bulbs, and blood were collected and analyzed 3 days and 6 days post infection. (C) Mouse body weight was monitored daily for up to 6 days and is expressed as the percentage of weight relative to the initial weight on day 0. Statistical significance of comparison between IN- (n = 17-29, blue dots) and AR-infected mice (n = 9-23, red dots) is shown. Control mice treated with PBS are also shown (n = 9-11, black dots). Data are represented as mean ± SEM. n indicate a range of mice, some of them have been dropped out from the analysis to be sacrificed for planned experimental end-points. (D) Clinical score was assessed evaluating the piloerection (0-3), posture (0-3), activity level (0-3), eye closure (0-3) and breathing (0-3) (see Methods). Statistical significance of comparison between IN- (n = 16, blue dots) and AR-infected mice (n = 16, red dots) is shown. PBS-treated control mice did not exhibit any clinical signs. (E) Survival curve of IN- (n = 32, blue dots) and AR-infected mice (n = 26, red dots). (F, G) Quantification of SARS-CoV-2 RNA in the brain of IN- (n = 7-11, blue dots) and AR- (n = 6-10, red dots) infected mice as well as of PBS-treated control mice (n = 4, black dots) measured 3 days (F) and 6 days (G) post infection. RNA values are expressed as copy number per ng of total RNA and the limit of detection is indicated as a dotted line. (H, I) Viral titers in the brain were determined 3 (H) and 6 days (I) after infection by median tissue culture infectious dose (TCID₅₀). PBS-treated control mice: n = 2-3, black dots; IN-infected mice: n = 4, blue dots; AR-infected mice: n = 4, red dots. (J) Representative immunohistochemical micrographs of sagittal brain sections from PBS-treated control mice (left), IN- (middle) and AR-infected mice (right) 6 days post infection. N-SARS-CoV-2 positive cells are depicted in brown. Scale bars, 1 mm. (K) Representative confocal immunofluorescence staining for N-CoV-2 (red) in sagittal brain sections of IN-infected mice. Cell nuclei are depicted in blue. White boxes indicate different brain areas: cerebellum (Cer); dentate gyrus (DG); hippocampus (Hipp); corpus callosum (Cc); cerebral cortex (Ctx); thalamus (Tha); striatum (Str). Scale bar, 1 mm. (L) Representative confocal immunofluorescence micrographs of sagittal brain sections from PBS-treated control mice (first panel) and IN-infected mice 6 days post infection. N-CoV-2 is depicted in red, NeuN neural marker in green and cell nuclei in blue. Fields of cerebral cortex (Ctx), hippocampus (Hipp), striatum (Str), thalamus (Tha) and cerebellum (Cer) are shown. Scale bars, 100 μm. (M) Representative confocal immunofluorescence micrographs of cerebral cortex (Ctx) in PBS-treated control mice (left panel) and IN-infected mice (right panel) 6 days post infection. iNOS⁺ cells are depicted in red and cell nuclei in blue. Scale bars, 100 μm. (N) Representative confocal immunofluorescence micrographs of two areas of the brain from IN-infected mice 6 days post infection. Corpus callosum (Cc) of the cerebral cortex, left panel, and dentate gyrus (DG), right panel. N-CoV-2 is depicted in red, GFAP astroglial marker in green and cell nuclei in blue. Scale bars, 100 μm. (O) Representative confocal immunofluorescence micrographs of the cerebral cortex (Ctx) of IN-infected mice at 6 days post infection. N-CoV-2 is depicted in red, Iba1 microglial marker in green and cell nuclei in blue. White box indicates the magnification represented in the right panel. Scale bars represent 100 μm (image) and 25 μm (magnification). (P, Q) Representative confocal immunofluorescence micrographs of the cerebral cortex (Ctx) of PBS-treated control mice (P) and IN-infected mice (Q) 6 days post infection. Left panels show Iba1 microglial marker in green and cell nuclei in blue; right panel shows CD68 marker of microglial activation in white. Scale bars, 10 μm. (R, S) Quantification of SARS-CoV-2 RNA in the olfactory bulbs of IN- (n = 4-12, blue dots) and AR- (n = 4-12, red dots) infected mice as well as of PBS-treated control mice (n = 2-4, black dots) measured 3 days (R) and 6 days (S) post infection. RNA values are expressed as copy number per ng of total RNA and the limit of detection is indicated as a dotted line. Data are expressed as mean ± SEM. Data in (C-I, R, S) are pooled from 2 independent experiments per time point. * p-value < 0.05, ** p-value < 0.01, *** p-value < 0.001; two-way ANOVA followed by Sidak’s multiple comparison test (C, D, comparison between blue and red dots for each time point); Log-rank (Mantel-Cox) test (E); Kruskal-Wallis test (F-I, R-S).

Fig. 2.. Aerosol exposure of K18-hACE2 transgenic mice to SARS-CoV-2 leads to efficient respiratory infection, anosmia, and fibrin deposition in the lung.

(A-B) Quantification of SARS-CoV-2 RNA in the nasal turbinates of PBS-treated control mice (n = 3, black dots) and of IN- (n = 4-7, blue dots) and AR- (n = 4-7, red dots) infected mice 3 days (A) and 6 days (B) post infection. RNA values are expressed as copy number per ng of total RNA and the limit of detection is indicated as a dotted line. (C) Illustration showing social scent-discrimination test. Male mice were free to investigate for 5 min two different tubes containing their own cage bedding or female cage bedding placed at two opposite corners of a clean cage. (D, E) Time that males spent sniffing their own male scent or female scent (D) and the sum of the two times (E) is expressed as time sniffing (seconds). Analyses were performed 3 or 6 days post IN- (n = 3-6, blue dots) or AR-infection (n = 3-5, red dots). As control, PBS-treated mice are shown (n = 8, black dots). In D, comparison between female and male time in each group of mice. n.d., the sniffing time could not be determined since mice were completely lethargic. (F) Preference index for male mice was calculated as (female time – male time)/ (total sniffing time female + male), as an indicator of the time spent sniffing preferred (female) or non-preferred (male) scents. n.d.: the sniffing time could not be determined since mice were completely lethargic. (G, H) Quantification of SARS-CoV-2 RNA in the lungs of IN- (n = 7-11, blue dots) and AR- (n = 7-10, red dots) infected mice as well as of PBS-treated control mice (n = 4, black dots) measured 3 days (G) and 6 days (H) post infection. RNA values are expressed as copy number per ng of total RNA and the limit of detection is indicated as a dotted line. (I, J) Viral titers in the lungs were determined 3 (I) and 6 days (J) after infection by median tissue culture infectious dose (TCID₅₀). PBS-treated control mice: n = 2-3, black dots; IN-infected mice: n = 4-7, blue dots; AR-infected mice: n = 4-10, red dots. (K) Representative immunohistochemical (left) and confocal immunofluorescence (right) micrographs of lung sections from PBS-treated control mice (top), IN- (middle) and AR-infected mice (bottom) at 6 days post infection. N-CoV-2 positive cells are depicted in brown (left panels) or in red (right panels). Cell nuclei are depicted in blue (right panels). Scale bars, 30 μm. (L) Pulmonary function was assessed by whole-body plethysmography performed 3 and 5 days post IN- (n = 6-14, blue dots) and AR-infection (n = 7-14, red dots). As control, PBS-treated mice were evaluated (n = 6, black dots). Frequency (left) and Rpef (right) parameters are shown. Calculated respiratory values were averaged over a 15 min-data collection period. (M) Representative aggregometry curves induced by arachidonic acid (AA) on platelet-rich plasma from PBS-treated control mice (left), IN- (middle) and AR-infected mice (right) 6 days post infection. Platelet aggregation was measured by light transmission aggregometry for 5 min and is expressed as % aggregation. (N) Representative immunohistochemical micrographs of lung sections from PBS-treated control mice (left), IN- (middle) and AR-infected mice (right) at 6 days post infection. Fibrin deposition is shown in brown. Scale bars, 30 μm. (O) Quantification of the size of fibrin deposits (μm²). n = 4. Data are expressed as mean ± SEM and are pooled from 2 independent experiments per time point. * p-value < 0.05, ** p-value < 0.01, *** p-value < 0.001; Kruskal-Wallis test (A, B, E-J); two-way ANOVA followed by Sidak’s multiple comparison test (L); Mann-Whitney U-test two-tailed (D, O). In D, statistical analysis was performed comparing female and male sniffing time within the same experimental group of mice.

Fig. 3.. Histopathological changes, immune response, and transcriptional signatures in the lungs of infected mice.

(A) Representative hematoxylin/eosin (H&E) micrographs of lung sections from PBS-treated control mice (left), IN- (middle) and AR-infected mice (right) 6 days post infection. Scale bars, 50 μm. (B-E) Absolute number of total cells (B, D) and of single cell populations (C, E) recovered from lung homogenates (B, C) and bronchoalveolar lavage (BAL) (D, E) of PBS-treated control mice (n = 4-6, black dots), IN- (n = 3-7, blue dots) and AR-infected mice (n = 3-7, red dots) analyzed 6 days post infection. CD8⁺ T cells (Live, CD45⁺, CD8⁺); CD4⁺ T cells (Live, CD45⁺, CD4⁺); B cells (Live, CD45⁺, CD8^-, CD4^-, B220⁺, CD19⁺); AM, alveolar macrophages (Live, CD45⁺, CD8^-, CD4^-, Ly6g^-, CD11b^-, F4/80⁺, SiglecF^hi); Mono, monocytes (Live, CD45⁺, CD8^-, CD4^-, Ly6g^-, SiglecF^-, CD11b⁺, Ly-6c⁺); Eosino, eosinophils (Live, CD45⁺, CD8^-, CD4^-, Ly6g^-, CD11b⁺, SiglecF^int); Neu, neutrophils (Live, CD45⁺, CD8^-, CD4^-, CD11b⁺, Ly6g⁺). (F) Principal Component Analysis (PCA) of RNA-seq expression values from the lungs of PBS-treated, control (n = 2, black), IN- (n = 3, blue) and AR-infected (n = 4, red) mice. Percentages indicate the variance explained by each PC. (G) Volcano plot of RNA-Seq results. The X-axis represents the Log2 Fold-Change of Differentially Expressed Genes (DEG) comparing AR- to IN-infection, the Y-axis the -Log10(FDR). Genes significantly up-regulated in AR- relative to IN- (|logFC| > 1 and adjusted P value < 0.05, horizontal and vertical dashed line) belonging to the “Adaptive immune system” pathway from BioPlanet 2019 (55) are highlighted in violet. (H) GSEA of three gene sets described in (18), comparing the transcriptome of IN- and AR-infected mice, presented as the running enrichment score for the gene set as the analysis 'walks down' the ranked list of genes (reflective of the degree to which the gene set is over-represented at the top or bottom of the ranked list of genes) (top), the position of the gene-set members (black vertical lines) in the ranked list of genes (middle) and the value of the ranking metric (bottom). (I) Heatmaps of genes (one per row) belonging to the three signatures as in (H), expressed in logarithmic normalized read counts. Each column represents an individual sample. (J, K) Absolute number of CD8⁺ (J) and CD4⁺ (K) T cells producing IFN-γ, TNF-α or both and expressing CD44 and Granzyme-B (Gz-B) in the lungs of PBS-treated control mice (n = 4, black dots), IN- (n = 5, blue dots) and AR-infected mice (n = 5, red dots) 6 days after infection. Data are expressed as mean ± SEM. Data in (B-E, J, K) are pooled from 2 independent experiments. * p-value < 0.05, ** p-value < 0.01; Kruskal-Wallis test (B-E, J, K).