ABSTRACT
The global spread of SARS-CoV-2 has made millions ill with COVID-19 and even more from the economic fallout of this pandemic. Our quest to test new therapeutics and vaccines require small animal models that replicate disease phenotypes seen in COVID-19 cases. Rodent models of SARS-CoV-2 infection thus far have shown mild to moderate pulmonary disease; mortality, if any, has been associated with prominent signs of central nervous system (CNS) infection and dysfunction. Here we describe the isolation of SARS-CoV-2 variants with propensity for either pulmonary or CNS infection. Using a wild-type SARS-CoV-2 isolated from a COVID-19 patient, we first found that infection was lethal in transgenic mice expressing the human angiotensin I-converting enzyme 2 (hACE2). Fortuitously, full genome sequencing of SARS-CoV-2 from the brain and lung of these animals showed genetic differences. Likewise, SARS-CoV-2 isolates from brains and lungs of these also showed differences in plaque morphology. Inoculation of these brain and lung SARS-CoV-2 isolates into new batch of hACE2 mice intra-nasally resulted in lethal CNS and pulmonary infection, respectively. Collectively, our study suggests that genetic variants of SARS-CoV-2 could be used to replicate specific features of COVID-19 for the testing of potential vaccines or therapeutics.