Enhancing human ACE2 expression in mouse models to improve COVID-19 research

Abstract

Mice are one of the most common biological models for laboratory use. However, wild-type mice are not susceptible to COVID-19 infection due to the low affinity of mouse ACE2, the entry protein for SARS-CoV-2. Although mice with human ACE2 (hACE2) driven by Ace2 promoter reflect its tissue specificity, these animals exhibit low ACE2 expression, potentially limiting their fidelity in mimicking COVID-19 manifestations and their utility in viral studies. Here, we created and compared hACE2 mouse models generated with different strategies. Our findings show that distinct β-globin insertion within hACE2 cassette significantly influences its expression, with downstream placement enhancing transcription. Moreover, optimizing hACE2 codons (opt-hACE2) improves translation efficiency in multiple tissues. Notably, opt-hACE2 mice displayed more active immune responses and severe COVID-19 phenotypes following SARS-CoV-2 challenge compared to other models. Our study demonstrates the dual regulatory role of β-globin element in transgene transcription and suggests that opt-hACE2 mice might serve as valuable tools for SARS-CoV-2 research.

Keywords: COVID‐19; hACE2; mouse model; β‐globin.

Fig. 1

Characterization of different hACE2 mouse models. (A) Schematic diagram showing four different molecular cloning strategies including WPRE, β‐globin/hACE2 (upstream), hACE2/β‐globin (downstream), and opt‐hACE2 for hACE2 cassette insertion at the mAce2 locus. (B) Schematic diagram of different hACE2 mice generated using mES editing combined with tetraploid complementation techniques. (C) Sanger sequencing results at the key positions of the hACE2 cassette, further confirming the appropriate insertion of hACE2 with different strategies (D) Images of the healthy hACE2 mice generated from different strategies.

Fig. 2

hACE2 expression in different tissues from four different hACE2 mouse models. (A) Quantification of mRNA levels of hACE2 in lungs, kidneys, intestines, testes, and brains from wild‐type (n = 6), hACE2‐WPRE (n ≥ 4), β‐globin/hACE2 (n ≥ 2), hACE2/β‐globin (n = 4), and opt‐hACE2 (n ≥ 5) mouse models using RT‐qPCR. Data are shown as relative expression levels compared to Gapdh. Error bars, SEM; Student's t‐test, *P < 0.05, **P < 0.01. (B) Western blot analysis of hACE2 in lungs, kidneys, brains, and intestines from different hACE2 mouse models. * unspecific bands.

Fig. 3

Viral infection results in COVID‐19 phenotypes in all tested hACE2 mouse models. (A) Expression levels of viral genes in the lungs of infected hACE2 mice from mock (n = 3), infected hACE2‐WPRE (n = 5), hACE2/β‐globin (n = 4), and opt‐hACE2 (n = 3) mice. ORF3a, E, and S expression levels are detectable in only 2 out of 3 mock replicates. Error bars, SEM from at least 3 biological replicates. Student's t‐test, *P < 0.05, **P < 0.01. (B) Volcano plots showing transcriptome changes in the lungs between uninfected and infected hACE2 mice from different mouse models. Significantly differentially expressed genes (|log₂(fold change)| > 1, P‐value < 0.05) are highlighted in green (upregulated in control) and red (upregulated in the infected mice). (C) GO analysis of activated genes in the lungs of hACE2 mice from different models upon viral challenging. (D) Heatmap showing the enrichment of GO terms from different infected hACE2 mice. Data were shown as −log₁₀(qValue). (E) Bar plots showing the expression levels of genes involved in immune response pathway in the lungs of infected hACE2‐WPRE (n = 5), hACE2/β‐globin (n = 4), and opt‐hACE2 (n = 3) mice. Data from A and E were shown as normalized expression representing normalized read counts from RNA‐seq data. Error bars, SEM from at least 3 biological replicates. Student's t‐test, *P < 0.05, **P < 0.01.

Fig. 4

Histological changes revealed a more active immune response in opt‐hACE2 mice. Lung tissues of hACE2 mice from different mouse models challenged by SARS‐CoV‐2 at 3 days postinfection were stained with H&E. Lesions are not observed in uninfected mice (A–B, E–F, I–J). In lungs of infected hACE2‐WPRE and hACE2/β‐globin mice, staining with H&E shows thickened alveolar walls and pulmonary hemorrhage (yellow arrows) along with mild inflammatory cell infiltration (C–D, G–H). In opt‐hACE2 mice, lung tissues show thickened alveolar walls, mild pulmonary hemorrhage, and more severe inflammatory cell infiltration after rival infection. Red arrows, thickened alveolar walls with immune cell infiltration; yellow arrows, hemorrhage; green arrows, vascular leakage. Black scale bar = 200 μm, and green scale bar = 50 μm. (M) Quantitative scoring of pathology of lung tissues from mock and infected hACE2‐WPRE (3 vs 3), hACE2/β‐globin (2 vs 4), and opt‐hACE2 (3 vs 2) mice. Error bars indicate the standard error of mean.