ACE2 knockout hinders SARS-CoV-2 propagation in iPS cell-derived airway and alveolar epithelial cells

Abstract

Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), the causative agent of COVID-19, continues to spread around the world with serious cases and deaths. It has also been suggested that different genetic variants in the human genome affect both the susceptibility to infection and severity of disease in COVID-19 patients. Angiotensin-converting enzyme 2 (ACE2) has been identified as a cell surface receptor for SARS-CoV and SARS-CoV-2 entry into cells. The construction of an experimental model system using human iPS cells would enable further studies of the association between viral characteristics and genetic variants. Airway and alveolar epithelial cells are cell types of the lung that express high levels of ACE2 and are suitable for in vitro infection experiments. Here, we show that human iPS cell-derived airway and alveolar epithelial cells are highly susceptible to viral infection of SARS-CoV-2. Using gene knockout with CRISPR-Cas9 in human iPS cells we demonstrate that ACE2 plays an essential role in the airway and alveolar epithelial cell entry of SARS-CoV-2 in vitro. Replication of SARS-CoV-2 was strongly suppressed in ACE2 knockout (KO) lung cells. Our model system based on human iPS cell-derived lung cells may be applied to understand the molecular biology regulating viral respiratory infection leading to potential therapeutic developments for COVID-19 and the prevention of future pandemics.

Keywords: ACE2; CRISPR-Cas9; SARS-CoV-2; gene editing; gene knockout; iPS cells.

FIGURE 1

Genome editing strategy for ACE2 KO in B2-3 cells. (A) Designs of guide RNAs (gRNAs) used in this study. The PAM sequences are shown by the pink line over bases. Microhomologies (µHs) are boxed in green. The expected deletion is framed by the dotted line. (B) Diagram of ACE2 protein. The positions of each gRNAs are shown above the diagram. (C) Tracking of Indels by Decomposition (TIDE) analysis to identify the gene editing outcomes of bulk populations. (D) Sequences of each allele of the selected clones.

FIGURE 2

Quality control of ACE2 KO iPSC clones. (A) Immunostaining of pluripotency markers, OCT3/4, NANOG, TRA-1-60, and TRA-1-81. Scale bar 100 µm. (B) The image of karyotyping assay of 4 clones (C) Off-target sites were identified by GGGenome and were confirmed by Sanger sequencing to be unedited.

FIGURE 3

Confirmation of iPS cell differentiation and ACE2 KO. (A) RNA-Seq to quantify the expression of lung-related genes in differntiated ACE2 WT and KO iPSCs (Log10 TPM). (B) Western blotting for ACE2 in WT and KO cells. (C) RNA-Seq analysis showing reduced expression of ACE2 mRNA in KO cells. (D) Expression of proteases related to viral infection remain unchanged. The dots in the bar graphs represent biological replicates (N = 3). No statistically significant difference was observed in the expression level of differentiation markers or proteases between differentiated WT and ACE2 KO cells (DESeq2 adjusted p-value >0.05).

FIGURE 4

SARS-CoV-2 infection in the gene edited iPS cell lung model. (A–D) Observation of SARS-CoV-2 viral particles in wild-type airway and alveolar epithelial cells using TEM. The images were taken at 4 dpi (days post-infection) in airway epithelial cells (A,B) and 2 dpi in alveolar epithelial cells (C,D). A and C were imaged at ×5,000 magnification with a 1 µm scale bar, while B and D were further enlarged to a scale equivalent of ×20000 magnification with a 200 nm scale bar. (E) Measurement of virus titers following infection of SARS-CoV-2 strain WK521 into parental and ACE2 KO iPS cell-derived airway and (F) alveolar epithelial cells. p-value from one-way ANOVA with Dunnett’s or Tukey’s multiple comparisons post hoc test is shown in the individual graphs. The dots represent each biological replicate (N = 3).

FIGURE 5

Workflow of this study.