Microscopic morphomolecular evaluation of transgenic humanized ACE2 murine models of SARS-CoV2
OA Version
Citation
Abstract
INTRODUCTION: SARS-CoV-2 is the causative agent for the ongoing pandemic that was first declared in 2020, taking the lives of almost six million people and disrupting communities worldwide. Although an impressive global effort from the scientific community has yielded multiple vaccines and therapeutics, more research is crucial for continued progress against SARS-CoV-2 and future emerging infectious diseases. Animal models have played a significant role in the development of many advancements throughout the pandemic, and better models are needed for more effective research.
OBJECTIVE: Although many animals have been utilized for SARS-CoV-2 research, a model to recapitulate severe pulmonary disease is still lacking. Routinely utilized models have consisted of non-human primates, Syrian hamsters, and mice. Excluding ethical concerns, non-human primates are expensive and limited in supply, limiting the ability to execute more statistically powerful studies. Pneumonia caused by SARS-CoV-2 in non-human primates is also very mild, with nearly all animals surviving, creating substantial skepticism surrounding the frequency of acute respiratory distress syndrome (ARDS) and diffuse alveolar damage (DAD) occurrence in these animal models. Syrian hamsters are also naturally permissive to SARS-CoV-2 and consistently display the most severe lung pathology of any existing animal model, but the lack of availability of species specific reagents and research tools makes studying this model difficult. Utilization of mouse model does not require development of new research tools, as mice have been classically utilized for preclinical research for decades. This work seeks to characterize and evaluate two human ACE2-expressing transgenic mouse models to provide the scientific community with knowledge on their translational relevance.
METHODS: K18-hACE2 (K18) and Rosa26-hACE2 (Rosa26) mice were infected with SARS-CoV-2 and checked daily for temperature and weight. Plaque assays and qPCR were utilized to determine viral load. Tissues were stained with H&E for histopathological scoring and qualitative analysis. K18-HACE2and Rosa26-hACE2tissues were fluorescently labeled using two different multiplex immunohistochemistry panels. Slides were digitized by a Vectra Polaris™ fluorescent whole slide scanner, unmixed using inFORMTM vxxx and digital analysis was completed using HALO™ vxxx. Statistical analysis was conducted using GraphPad Prism™ 9.0.1.
RESULTS: Both transgenic models succumbed to SARS-CoV-2 infection, with neurodissemination and death/euthanasia corresponding with peak viral loads in both models. hACE2 mRNA and ACE2 protein anatomical distribution and expression levels was similar in both models as determined by RNAscope® ISH and IHC respectively. In brains, hACE2 was expressed sporadically in neurons, but consistently in blood vessels and choroidal epithelium. In the lungs, viral load peaked on day 2 and 4 while lung infiltrate steadily increased throughout the course of infection, peaking on day 7- 8. However, severe lung pathology was not observed in any animals and many of the hallmarks of diffuse alveolar damage were absent, namely the formation of a hyaline membrane, hemorrhage, edema, alveolar fibrin polymerization and neutrophil influx. K18-HACE2 mice showed less lung infiltrate when compared to Rosa26-hACE2mice, which had more T-cell rich infiltrate. No significant difference exists between the two strains in terms of IBA1+ cells and CD11b+ cells in the lungs, though both cell populations increased throughout the course of infection. Both models demonstrated neuroinvasion as early as day 4, but neurodissemination in the Rosa26-hACE2infection was limited to ventral portions of the brain, while the K18-hACE2 showed severe and near global dissemination within the brain aside from cerebellar sparing which was observed in both models. K18-HACE2mice showed a significant decrease in neuronal density and an increase in microglial reactive processes consistent with SARS-CoV-2 induced neuronal loss and microglial reactivity. Together, these findings show that neither hACE2 transgenic mouse model represents a model of severe lung disease for SARS-CoV-2 and that the main determinant of lethality is viral neuroinvasion and neurodissemination. Although hACE2 was under the control of the keratin 18 promoter in both models, the distinct insertion location resulted in distinct clinicopathological outcomes that are not easily explained but bring appreciation to the complexity of the central dogma of biology.
CONCLUSION: Using digital image analysis of immunohistochemistry paired with histopathological scoring and traditional molecular and virological techniques, this study demonstrates that although the transgenic hACE2 mouse models available to researchers result in lethal disease following SARS-CoV-2 infection, death/euthanasia is invariably resulting of neurodissemination with mild pneumonia limiting their translational relevance of mirroring severe COVID-19 in humans..
Description
License
Attribution-ShareAlike 4.0 International