Extended Data Fig. 7: Generation of the AAV-hACE2-transduced mouse model of COVID-19.
From: Bispecific IgG neutralizes SARS-CoV-2 variants and prevents escape in mice

a, Diagram of the AAV-hACE2 plasmid and corresponding AAV vector. b, Western blot analysis detecting hACE2 expression in the lungs of one non-transduced control mouse (ctrl) and 12 mice transduced with two doses of AAV-hACE2 viral particles (5 × 1010 or 1 × 1011 genome copies (GC)). Lung tissue was collected 1, 2 or 4 weeks (w) after transduction. Histone H3 was used as control for quantification (bottom). Quantitative analysis represents normalized data from membrane images (top), and was performed using ImageJ. Representative data from two independent experiments are shown. c, Preparation of concentrated AAV-hACE2. AAV-hACE2 plasmid was co-transfected with pHelper and AAV Rep/Cap 2/9n vectors into 293AAV cells (Methods). To increase viral titres, viral particles from both cell lysate and PEG-precipitated growth medium were ultracentrifuged in a discontinuous iodixanol gradient. The silver-stained SDS–PAGE gel shows 14 consecutive fractions: fractions 1–9 represent enriched AAV fractions used for experiments, and fractions 10–14 are contaminated with proteinaceous cell debris. Iodixanol was chosen as a density gradient medium owing to its low toxicity in vivo and its easy removal by ultrafiltration. M, protein marker; *AAV capsid proteins VP1, VP2, and VP3. Representative data from two independent experiments are shown. d, The amount of AAV particles was estimated by RT–qPCR. The number of genome copies expressed as log was calculated from a standard curve. From one 15-cm2 dish, 75 μl with 2.0 × 1012 genome copies per ml were prepared, which is sufficient for hACE2 humanization of 37 mice. e, Kinetic of lung histopathology in SARS-CoV-2-infected ACE2-humanized mice. H&E-stained sections showed inflammatory infiltrates composed of lymphocytes, macrophages, neutrophils and fibroblasts replacing the alveoli. The size of the affected areas increased over time (area of diffuse alveolar damage: control <5–10%, 2 dpi <10–30%, 5 dpi 20–80% and 8 dpi 50–90%). Alveolar septa were thickened in areas that were close to infiltrates. In samples collected at 5 and 8 dpi, an increased number of activated macrophages with foamy cytoplasm (black arrowheads) was seen. AAV-hACE2-transduced, SARS-CoV-2-uninfected mice were used as control and showed no noticeable pathology. Each image is representative of two separate experiments (n = 3 to 5 mice per group).