Fatal Neurodissemination and SARS-CoV-2 Tropism in K18-hACE2 Mice Is Only Partially Dependent on hACE2 Expression

Abstract

Animal models recapitulating COVID-19 are critical to enhance our understanding of SARS-CoV-2 pathogenesis. Intranasally inoculated transgenic mice expressing human angiotensin-converting enzyme 2 under the cytokeratin 18 promoter (K18-hACE2) represent a lethal model of SARS-CoV-2 infection. We evaluated the clinical and virological dynamics of SARS-CoV-2 using two intranasal doses (10⁴ and 10⁶ PFUs), with a detailed spatiotemporal pathologic analysis of the 10⁶ dose cohort. Despite generally mild-to-moderate pneumonia, clinical decline resulting in euthanasia or death was commonly associated with hypothermia and viral neurodissemination independent of inoculation dose. Neuroinvasion was first observed at 4 days post-infection, initially restricted to the olfactory bulb suggesting axonal transport via the olfactory neuroepithelium as the earliest portal of entry. Absence of viremia suggests neuroinvasion occurs independently of transport across the blood-brain barrier. SARS-CoV-2 tropism was neither restricted to ACE2-expressing cells (e.g., AT1 pneumocytes), nor inclusive of some ACE2-positive cell lineages (e.g., bronchiolar epithelium and brain vasculature). Absence of detectable ACE2 protein expression in neurons but overexpression in neuroepithelium suggest this as the most likely portal of neuroinvasion, with subsequent ACE2 independent lethal neurodissemination. A paucity of epidemiological data and contradicting evidence for neuroinvasion and neurodissemination in humans call into question the translational relevance of this model.

Keywords: comparative pathology; immunohistochemistry; in situ hybridization; in vivo imaging; translational animal model; transmission electron microscopy; viral pathogenesis.

Figure 1

SARS-CoV-2 caused lethal disease in K18-hACE2 mice irrespective of dose (10⁴ vs. 10⁶ PFU). K18-hACE2 mice (n = 50) were inoculated intranasally with either 1 × 10⁴ or 1 × 10⁶ plaque forming units (PFU), with n = 6 additional Sham/PBS negative controls. Body weight (A), clinical signs (B), body temperature (C), and survival (D) were monitored daily in Sham/PBS animals (black) and in infected animals (male, red; female, blue; up to 14 dpi). Mice meeting euthanasia criteria were counted dead the following day. Viral loads (viral RNA genome copy numbers/mg of tissue) or viral titers (infectious virus particles; PFU/mg of tissue) were quantified in the lung and brain (E–G). RNA copies were also examined in the serum (genome copies/mL) either directly on serum (H) or via a re-infectivity assay (I) using Vero E6 cells. The limit of detection is shown with a dashed line. Clinical data (A–D): 10⁴ PFU; male (n = 9), female (n = 6); 10⁶ PFU male (19), female (n = 16); Sham/PBS male (n = 3), female (n = 3). Molecular and virologic data (E–I). 10⁶ RT-PCR: lung 2 dpi (n = 3), 4 dpi (n = 6), 7 dpi (n = 8); brain 2 dpi (n = 3), 4 dpi (n = 5), 7 dpi (n = 7). 10⁶ PFU analysis: lung 2 dpi (n = 8), 4 dpi (n = 8), 7 dpi (n = 5); brain 2 dpi (n = 8), 4 dpi (n = 8), 7 dpi (n = 8). 10⁴ PFU analysis: lung 7 dpi (n = 4); brain 7 dpi (n = 4).10⁶ Serum RT-PCR assay: Sham (n = 6), 2dpi (n = 6), 4 dpi (n = 6), 7 dpi (n = 15). 10⁶ Serum infectivity assay: Sham (n = 5), 2dpi (n = 3), 4 dpi (n = 5), 7 dpi (n = 8). One-way or two-way ANOVA. * p ≤ 0.05, ** p ≤ 0.01, **** p ≤ 0.0001. For A–D, shades of blue and red asterisks compare sham group vs. male group, and sham group vs. female group, respectively, with darker shades designated for 10⁴ PFU, and brighter shades for 10⁶ PFU.

Figure 2

Temporal analysis of SARS-CoV-2 infection in the nasal cavity of K18-hACE2 mice. Histological changes, and viral protein (brown) and RNA (red) distribution and abundance were assessed in non-infected (Sham/PBS: A,D,G,J; n = 3) and infected mice at 2 (B,E,H,K, n = 3) and 4 (C,F,I,L, n = 5) days following intranasal inoculation. At 2 dpi, neutrophilic rhinitis in the rostral and intermediate turbinates (B, arrow) correlated with intraepithelial SARS-CoV-2 protein and RNA (E, inset). Viral protein and RNA were detected in the olfactory neuroepithelium (ONE, K and inset) in the absence of histologic lesions (H). At 4 dpi, only sporadically infected cells were noted in the epithelium lining the nasal turbinates and ONE (F,L, arrow, and insets) in the absence of histologic lesions (C,I). Sham/PBS-infected are depicted in A, D, G,J. H&E, DAB IHC (viral protein), and Fast Red ISH (viral RNA), 200× total magnification. Bar = 100 μm.

Figure 3

Temporal qualitative and quantitative analysis of SARS-CoV-2 pneumonia in K18-hACE2 mice. Lung tissues from non-infected mice (PBS/Sham inoculated: (A,B) and from infected mice at 2 dpi (C,D), 4 dpi (E,F), 7 dpi (G,H) and 14 dpi (n = 2; I,J) following intranasal inoculation were analyzed. Subgross histological images of the lungs and corresponding pneumonia classifiers for each timepoint are depicted in panel (K) (green = normal; yellow = pneumonia). Mild-to-moderate interstitial pneumonia was evident starting at 2 dpi with frequently reactive blood vessels (D, arrow). At 7 dpi, alveolar type 2 (AT2) cell hyperplasia was observed (H, arrows). Residual mild-to-moderate pneumonia was observed in the two male survivors at 14 dpi from the survival curve, with rare sporadic lymphoid aggregates (J-arrows). H&E, 50× (A,C,E,G,I; bar = 500 μm), 200× (B,D,F,H,J; bar = 100 μm) and 1× (K) total magnification. Pneumonia classifier: PBS/Sham (n = 2), 4 dpi (n = 5), 7 dpi (n = 8), 14 dpi (n = 2). One-way ANOVA; ns, non-significant.

Figure 4

SARS-CoV-2 tropism following intranasal inoculation in K18-hACE2. (A–C) At 4 dpi, SARS-CoV-2 (yellow) tropism for RAGE+ alveolar type 1 (AT1, magenta) and scattered SPC+ alveolar type 2 (AT2, red) pneumocytes (B,C, arrowheads, and arrows, respectively) but not for CD31+ endothelial cells. (B,C) represent higher magnification images of the white hashed boxes from (A), respectively. At 6 dpi-terminal disease (D–F), interpretation of transmission electron microscopy images illustrated virus particles (VPs) bound by vesicles in AT1 (E) and AT2 (F) cells. AT1 contained abundant caveolae. Another unique feature observed in AT1 cells was the presence of cubic membranes (CuM). AT2 pneumocytes were characterized by presence of lamellar bodies (LB). (E,F) represent higher magnification images of the yellow hashed boxes from (D). (G,H) Viral particles or viral induced membrane alterations were not identified in ciliated or non-ciliated club (Cl) bronchiolar epithelial cells. (H) represents a higher magnification of the yellow hashed box in (G). Multiplex fluorescent IHC, 100× (A; bar = 100 μm) and 200× (B,C; bar = 50 μm) total magnification. TEM, bar = 2 μm (D), 100 nm (E–H), and 3 μm (G). A, alveolar lumen; BM, basement membrane; C, capillary; Cav, caveolae; Ci, ciliated epithelium; Cl, club epithelium; CuM, cubic membranes; DMVs, double-membrane vesicles; END, endothelium; VPs, viral particles.

Figure 5

Temporal immunoprofiling of the pulmonary host inflammatory response to SARS-CoV-2. (A–H) Quantification and 4-plex fluorescent IHC targeting SARS-CoV-2 Spike (A,E–H), and macrophage Iba-1+ (B,E–H), CD8+ (C,E–H) and CD19+ cell (D,E–H) infiltration in the lung of PBS/Sham inoculated mice and in SARS-CoV-2 inoculated mice (2 dpi (n = 3), 4 dpi (n = 4), 7 dpi (n = 5) and 14 dpi (n = 2). In inoculated mice, SARS-CoV-2 Spike peaked between 4–7 dpi (A,F,G). Iba-1+ macrophages (red) increased significantly peaking at 7 dpi (B,G), along with a lower infiltration of CD8+ T lymphocytes-magenta that peaked between 4–7 dpi (C,F,G) while Sham/PBS mice had low residual inflammatory cells (E). CD19+ B cells arranged in aggregates were only evident in the two male survivors euthanized at 14 dpi (D,H). Insets depict immune cell phenotyping outputs that were applied across the whole slide image. Sham/PBS-infected mice (n = 3) were used as baseline controls for quantitative analysis. Multiplex fluorescent IHC (E–H): 100× and 400× (insets) total magnification, Bar = 100μm. (I) Neutralizing activity of serum isolated from a naïve/non-infected K18-hACE2 (purple) and from the two male14 dpi survivors (survivor 1 and 2, red and black, respectively). An anti-SARS-CoV-2 Spike RBD antibody (anti-RBD, blue) was used as a positive control. Serum was serially diluted by 2-fold. One-way ANOVA. ** p ≤ 0.01; **** p ≤ 0.0001.

Figure 6

Invasion of SARS-CoV-2 into the central nervous system. (A) Sagittal sections of the head of non-infected (Sham/PBS, top panel) and infected (4 and 7 dpi, middle and bottom panel, respectively) were analyzed for viral protein and RNA distribution. At 4 dpi (middle panel), SARS-CoV-2 infected neurons within the mitral layer of the olfactory bulb (1, arrow) as well as small clusters of neuronal bodies within the cerebral cortex (2, SARS-CoV-2 RNA in inset). At 7 dpi (bottom panel), SARS-CoV-2 protein was widespread along the mitral layer of the olfactory bulb (1) and throughout the central nervous system (2, SARS-CoV-2 RNA in inset) with exception of the cerebellum. EPL, external plexiform layer; GCL, granular cell layer; GL, glomerular layer; ML, mitral layer. DAB (viral protein) and Fast Red (viral RNA). 7.5× (bar = 2.5 mm) and 200× (bar = 100 μm) total magnification. On the right of each panel, pictures labelled 1 and 2 are 266× total magnification insets represented by the hashed squares labeled in the lower (7.5×) magnification images. (B) Representative three-dimensional profile view (right side) of a K18-hACE2 mouse following inoculation with a SARS-CoV-2 NL virus (10⁶ PFU). NanoLuc bioluminescent signal was detected and quantified at 6 dpi following fluorofurimazine injection (Sub-cutaneous) using the InVivoPLOT (InVivoAx) system and an IVIS Spectrum (PerkinElmer) optical imaging instrument. Location of the lungs and brain are indicated.

Figure 7

SARS-CoV-2-associated neuronal morphological changes, neuronal antigen abundance and glial response. Morphological changes in the brain were noted as early as 6 dpi (A,B) and were characterized by variable spongiosis (B, arrowheads, and top inset) with neuronal degeneration and necrosis (B, bottom inset and arrows) involving multiple areas within the cerebral cortex and elsewhere. (C–E) Quantification of 3-plex fluorescent IHC targeting SARS-CoV-2 Spike protein, astrocytes (GFAP) and microglia (Iba-1) in the brain of Sham/PBS mice and in inoculated mice (2, 4, 7 and 14 dpi). The amount of viral protein rapidly and markedly increased by 7 dpi, along with an intense astrocytic and microglial response (C–E). H&E, 100×, bar = 200 μm. Multiplex IHC, 200× total magnification, bar = 100 μm. One-way ANOVA; * p ≤ 0.05.

Figure 8

Ultrastructural features of SARS-CoV-2-infected neurons. (A) Neurons (Neu) contain abundant intracytoplasmic viral particles and various replication-associated intracytoplasmic membranous structures. Necrotic neurons are diffusely electron dense (NNeu). (B) Higher magnification of the squared area on (A) depicting double-membrane vesicles (DMVs) and spherules (DMSs). (C) Cubic membranes (CMs) are also noted among DMVs and DMSs. (D) Higher magnification of the squared area on (A). Mature viral particles are indicated as VPs. (E,F) VPs are associated with membranous structures related to viral replication (DMVs, DMSs and CMs) and with the rough endoplasmic reticulum (RER). (G) Necrotic neurons (NNeu) are diffusely electron dense and contain numerous cytoplasmic vacuoles. (H) Higher magnification of squared area on (G). Note the high cytoplasmic electron-density. Scale bars = 100 nm.

Figure 9

Distribution of ACE2 in lungs, nasal cavity, brain, and olfactory bulb of wild-type C57BL/6J and uninfected and SARS-CoV-2 infected K18-hACE2 mice. Lung (A–C), nasal (rostral/intermediate turbinates [R/I]) and olfactory epithelium (ONE) (D–F), olfactory bulb (G–I) and brain (J–L) from non-infected C57BL/6J, and from non-infected and infected K18-hACE2 mice (7 dpi). K18-hACE2 mice were analyzed via immunohistochemistry using a cross-reactive anti-ACE2 antibody. In the lungs (A–C), ACE2 expression (brown) was mostly restricted to the apical membrane of bronchiolar epithelial cells with scattered positive AT2 cells (inset arrows). Nasal (rostral/intermediate turbinates [R/I]) and olfactory epithelium (ONE) were devoid of ACE2 in C57BL/6J mice (D), but expression was enhanced in K18-hACE2 mice with intense apical expression (E,F). ACE2 expression within the olfactory bulb (G–I) and the brain (J–L) was restricted to capillary endothelium with no neuronal expression. DAB, 200× total magnification. Bar = 100 μm.

Figure 10

Expression and distribution of hACE2 mRNA in the brain and lungs of non-transgenic wild-type C57BL/6J and K18-hACE2 transgenic mice via RNAscope^® ISH. (A–C) hACE2 lung expression. While no expression of hACE2 was noted in the lungs of wild-type C57BL/6J mice (A), hACE2 was expressed in the bronchiolar epithelium (arrowheads) and sporadic AT2 cells (arrows) in transgenic K18-hACE2 mice (B,C), which correlated with immunohistochemical findings. (D–F) hACE2 brain expression. hACE2 was not expressed in the Cerebrum of C57BL/6J mice (D) but in clusters of neurons within the cerebrum (E) and hippocampus (F). Fast Red, 400× total magnification. Bar = 50 μm.