ABSTRACT
Background: There is an ongoing controversy as to whether SARS CoV-2 can infect the kidney parenchyma directly. To date, the presence of SARS CoV-2 in the kidney has been described mainly post-mortem in autopsy studies of patients who died of or with COVID-19, but this has not been examined in an experimental model where the timing of SARS-CoV-2 infection can be defined. We used transgenic mice expressing human ACE2 (k18hACE2) susceptible to lethal SARS-CoV-2 infection to study this issue directly on kidney tissue taken at defined time points and using lung tissue as positive control. Method(s): Transgenic k18hACE2 mice were inoculated with 3x104 PFU SARSCoV-2 in a BSL-3 facility. Kidneys and lungs were removed from the animals sacrificed on days 5 to 7 and used for histology (PAS-staining), immunofluorescence (IF) of the S1 spike protein of SARS-CoV-2 and measurement of viral load by plaque assay. Kidney samples were additionally evaluated by IF using kidney injury markers NGAL and KIM-1. Result(s): Kidney tissue stained using an anti-S1-spike antibody showed negative results in all samples (n=15). By plaque assay, viral titers were also not detectable in any of the kidneys. By contrast, lungs from infected mice showed strong staining for the S1 spike protein in 13 of 14 cases and this was associated with positive viral titers in all lung samples. Despite severe lung disease, only mild and variable kidney damage was observed by histopathology. Positive staining for NGAL in the proximal tubules was consistently seen, while KIM-1 staining was rarely positive. Conclusion(s): In a transgenic mouse model with lethal SARS-CoV-2 infection and severe lung but mild kidney disease there is no evidence of S1 spike protein in the kidney, which is consistent with lack of detection of replicating virus by plaque assay.