Characterization of low copy number human angiotensin-convertase 2 (hACE2)-transgenic mice as an improved model of SARS-CoV-2 infection

Coronaviridae are significant human pathogens, as evidenced by several outbreaks of severe respiratory infections in the past 20 years and culminating with the COVID-19 pandemic. Mouse models of COVID-19 have included transgenic expression of the main SARS coronavirus entry receptor on human cells, human angiotensin-converting enzyme 2 (hACE2). However, the original hACE2-Tg mouse strain overexpresses many copies of the transgene, leading to neuropathology not representative of human infection. Aiming to improve physiological relevance, we generated two new lines of hACE2-Tg mice using the original transgene construct expressing hACE2 under the control of the keratin 18 promoter (K18-hACE2). We show that relative to the original strain, which expressed 8 copies of the transgene (8-hACE2-Tg), lines 1 and 2 expressed 1 and 2 copies of the transgene (1-hACE-2-Tg and 2-hACE-2-Tg, respectively). Upon intranasal (i.n.) infection with 103 plaque-forming units (pfu) SARS-CoV-2 WA-1/US, 8-hACE2-Tg mice succumbed to infection by d. 7. 2-hACE2-Tg mice exhibited 31% survival, with less viral replication in the lung and brain when compared to 8-hACE2-Tg mice. Furthermore, SARS-CoV-2 infection in 1-hACE2-Tg mice exhibited no mortality and had no detectable virus in the brain, although they did show clear virus replication in the lung. All three mouse strains analyzed showed SARS-CoV-2-related weight loss that tracked with the mortality rates. 1-hACE2-Tg mice mounted detectable primary and memory T effector cell and antibody responses. We conclude that these strains, particularly 1-hACE2-Tg mice, provide improved models to study hACE2-mediated viral infections.