Development and characterization of a fully humanized ACE2 mouse model

Abstract

Background: Many humanized angiotensin-converting enzyme 2 (ACE2) mouse models of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection do not replicate human ACE2 protein expression and thus exhibit pathology infrequently observed in humans. To address this limitation, we designed and characterized a fully humanized ACE2 (hACE2) mouse by replacing all exons/introns of the mouse Ace2 locus with human DNA comprising the entire ACE2 gene and an upstream long noncoding RNA (LncRNA).

Results: Compared to the popular Keratin18 ACE2 (KRT18-ACE2, K18) mouse model of SARS-CoV-2 infection, hACE2 mice displayed a similar tissue expression profile of ACE2 as that seen in human tissues. Further, hACE2 mice showed comparable blood pressure, angiotensin II metabolism, and renal cortical transcriptome as wild-type mice. Intranasal infection of K18 mice with the beta variant of SARS-CoV-2 resulted in high viral replication and inflammation of the lung and brain, weight loss, and compassionate euthanasia five days post-infection (PI). Similarly infected hACE2 mice displayed viral replication and inflammation in the lung (but not in brain), sustained weight, and 100% survival up to 12 days PI, with clear evidence of acquired immunity. CRISPR-mediated disruption of the upstream LncRNA caused minimal effects on ACE2 mRNA and protein.

Conclusions: The hACE2 model offers a more accurate approach to studying mechanisms underlying tissue-restricted expression of ACE2, elucidating noncoding sequence variants and an upstream LncRNA, and defining pathways relevant to human disease and associated co-morbidities.

Keywords: ACE2; Angiotensin; CRISPR; Long noncoding RNA; Mouse; SARS-CoV-2.

Fig. 1

Generation and validation of the hACE2 mouse model. a Schematic illustration of the strategy for generating and sequence validating (by CRISPR-LRS) the humanized ACE2 mouse model (hACE2). The human sequence includes the ACE2-DT LncRNA and a truncated dACE2 isoform (not studied here). Red numbering in the bottom panel represents the human and mouse genomic DNA coordinates per each respective genome with CRISPR-LRS informative read numbers shown at the bottom. For a more detailed schematic of the strategy used, see Additional file 2; Fig. S1. Relative expression (to Gapdh) levels of human ACE2 (b) and ACE2-DT (c) in various tissues of the hACE2 mouse model (n = 3 mice per tissue type). Br, brain; He, heart; In, intestine; Ki, kidney; Lu, lung; Te, testis. Bar graphs represent the means ± SEM. Significance levels are indicated as follows: P < 0.05: *, P < 0.01: **. RMCE, recombination-mediated cassette exchange

Fig. 2

Blood pressure and gene expression in hACE2 mice. a Comparison of blood pressure between hACE2 (n = 6) and wild type (WT, n = 5) mice. b Volcano plot showing changes in gene expression from bulk RNA-seq analysis of hACE2 and WT mouse kidney cortex (n = 3 independent kidneys per genotype). c PCA plot of bulk RNA-seq in WT and hACE2 mice (n = 6; three females and three males per genotype). d ACE2 expression in male and female hACE2 mice (n = 3 mice per sex) assessed via Bulk RNA-seq reads (in 1000 s). Bar graphs represent the means ± SEM). Non-significance, ns

Fig. 3

ACE2 protein expression in K18 and hACE2 mice. Western blot analysis and confocal immunofluorescence staining of ACE2 protein expression in a kidney, b brain, c intestine, d lung, e testis, and f heart of K18 and hACE2 mice (n = 3 mice per model). ACE2 protein in g kidney and h intestine of K18 and hACE2 mouse models (n = 3 mice per model) versus corresponding tissues of human origin (n = 2 independent samples of each human tissue)

Fig. 4

Sex differences in ACE2 protein expression in hACE2 mice. Western blot a and quantitation b of ACE2 protein expression in the kidney of hACE2 mouse model, including WT and hemizygous male (Hemi) and heterozygous (Het) and homozygous (Homo) female mice (n = 5 per group). c ACE2 protein expression levels in the kidney of hACE2 mouse model after spaying (females) or castration (males), highlighting the impact of gonadectomy on ACE2 expression. Bar graphs represent the means ± SEM. Significance levels are P < 0.001: ***

Fig. 5

Viral replication in WT, K18, and hACE2 mice. a Schematic representation of the SARS-CoV-2 infection experimental design; Euth, euthanasia. b Kaplan–Meier survival curve for infected WT, K18, and hACE2 mice (n = 5 mice per group). c Body weight change between WT, K18, and hACE2 mouse models after SARS-CoV-2 infection (n = 5 mice per group). d–e Viral genomic copy (VGC) levels in the lungs and brain of K18 and hACE2 mouse models after three and five days of SARS-CoV-2 infection. Viral load was quantified by RT-qPCR. Data are expressed as VGC/μg of RNA (n = 3–5 mice per group). Bars represent the means ± SEM. Significance levels are indicated as follows: P < 0.05: *, P < 0.01: **, P < 0.001: ***, non-significance, ns. f Neutralizing antibody response. Sera collected from infected hACE2 mice 12 days after infection were analyzed for neutralization activity against SARS-CoV-2 Beta variant (B.1.351) via plaque reduction neutralization assay (n = 4 mice). Middle horizontal bar represents the mean ± SEM. g Confocal immunofluorescence staining for SARS-CoV-2 Nucleocapsid protein and double-stranded RNA (dsRNA) in the lung and in the brain tissues of K18, hACE2, and WT mice 3 and 5 days following infection. A representative image is shown for each group (n = 3 mice per group)

Fig. 6

Histopathology in K18 and hACE2 mice following SARS-CoV-2 infection. a H&E staining of lung and brain tissues from K18, hACE2, and WT mice three and five days after SARS-CoV-2 infection (n = 3 mice per group). Black arrows indicate areas of inflammatory infiltration. b Immunostaining for CD45, a marker of immune cell infiltration, in lung and brain tissues of K18, hACE2, and WT mice three and five days after infection (n = 3 mice per group)

Fig. 7

CRISPR-mediated deletion of ACE2-DT LncRNA. a Strategy for generating ACE2-DT knockout mice using two-component CRISPR [49] to delete the first exon of the ACE2-DT LncRNA. b Comparison of ACE2-DT RNA in the kidney of hACE2 and hACE2-DT knockout (hACE2-DT-Del) mice (n = 4 mice per genotype). c ACE2 mRNA levels in the kidney (Ki) and intestine (In) of hACE2 and hACE2-DT-Del mice (n = 5 mice per genotype). d–f ACE2 protein expression in the kidney and intestine of hACE2 and hACE2-DT-Del mice (n = 3 mice per genotype). Bar graphs represent the means ± SEM. Significance levels are P < 0.001: ***. Non-significance, ns