Expression Pattern of the SARS-CoV-2 Entry Genes ACE2 and TMPRSS2 in the Respiratory Tract

Abstract

To address the expression pattern of the SARS-CoV-2 receptor ACE2 and the viral priming protease TMPRSS2 in the respiratory tract, this study investigated RNA sequencing transcriptome profiling of samples of airway and oral mucosa. As shown, ACE2 has medium levels of expression in both small airway epithelium and masticatory mucosa, and high levels of expression in nasal epithelium. The expression of ACE2 is low in mucosal-associated invariant T (MAIT) cells and cannot be detected in alveolar macrophages. TMPRSS2 is highly expressed in small airway epithelium and nasal epithelium and has lower expression in masticatory mucosa. Our results provide the molecular basis that the nasal mucosa is the most susceptible locus in the respiratory tract for SARS-CoV-2 infection and consequently for subsequent droplet transmission and should be the focus for protection against SARS-CoV-2 infection.

Keywords: ACE2; COVID-19; SARS-CoV-2; TMPRSS2.

Figure 1

The expression of SARS-CoV-2 entry genes. (a) Established SARS-CoV-2 entry genes, ACE2 and TMPRSS2, in five different types of samples. The difference between ACE2 and TMPRSS2 in the small airway epithelium is highly significant (p = 1.00 × 10^-12). In the nasal epithelium, the expression of ACE2 is significantly higher than that in small airway epithelium (p = 6.92 × 10⁻⁴), alveolar macrophages (p = 5.02 × 10⁻⁴), MAIT (p = 0.016), and masticatory mucosa (p = 5.48 × 10⁻⁷); the expression of TMPRSS2 is lower than that in small airway epithelium (p = 5.20 × 10⁻³), but higher than in alveolar macrophages (p = 2.98 × 10⁻³), MAIT (p = 0.040), and masticatory mucosa (p = 1.40 × 10⁻⁵). (b) The expression of ACE and possible SARS-CoV-2 alternative entry genes, BSG, DPP4, and FURIN. Y-axis represents normalized FPKM values by the average of relative levels of 6 HKGs. The boxplot produced by the IBM SPSS Statistics Version 23 shows the mean, the first quartile and the third quartile, and the 95% confidence interval (CI). FPKM: fragments per kilobase of transcript per million mapped reads.