Neutrophil proteases are protective against SARS-CoV-2 by degrading the spike protein and dampening virus-mediated inflammation

Abstract

Studies on severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2) have highlighted the crucial role of host proteases for viral replication and the immune response. The serine proteases furin and TMPRSS2 and lysosomal cysteine proteases facilitate viral entry by limited proteolytic processing of the spike (S) protein. While neutrophils are recruited to the lungs during COVID-19 pneumonia, little is known about the role of the neutrophil serine proteases (NSPs) cathepsin G (CatG), elastase (NE), and proteinase 3 (PR3) on SARS-CoV-2 entry and replication. Furthermore, the current paradigm is that NSPs may contribute to the pathogenesis of severe COVID-19. Here, we show that these proteases cleaved the S protein at multiple sites and abrogated viral entry and replication in vitro. In mouse models, CatG significantly inhibited viral replication in the lung. Importantly, lung inflammation and pathology were increased in mice deficient in NE and/or CatG. These results reveal that NSPs contribute to innate defenses against SARS-CoV-2 infection via proteolytic inactivation of the S protein and that NE and CatG limit lung inflammation in vivo. We conclude that therapeutic interventions aiming to reduce the activity of NSPs may interfere with viral clearance and inflammation in COVID-19 patients.

Keywords: COVID-19; Inflammation; Neutrophils; Proteases; Serpins.

Figure 1. Neutrophil serine proteases degrade S and inhibit SARS-CoV-2 entry.

(A and B) SDS-PAGE of recombinant trimeric S(614G) (A) and S(MA10) (B) incubated with purified human CatG, NE, and PR3 at indicated concentrations. Arrowheads indicate S protein and asterisks cleaved fragments. (C and D) Immunoblots of recombinant trimeric S(614G) and S(MA10) incubated with mouse neutrophil lysates with or without α1AT and probed with anti-S1 (C) or anti-S2 (D) subunit antibodies. (E) Titers of VSV*ΔG-S_Δ21 incubated with NSPs. (F and G) Titers of SARS-CoV-2^614G (F) and SARS-CoV-2^MA10 (G) incubated with NSPs. Data in A–D representative of at least 3 independent experiments. Data in E–G were analyzed by 1-way ANOVA with Dunnett’s multiple-comparison test, comparing the NSP-treated group to the PBS control (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 2. CatG^–/– mice have increased viral replication in the lungs and in the nose compared with WT mice.

WT, CatG^–/–, NE^–/–, PR3^–/–, and NE.CatG^–/– mice were inoculated intranasally with 1 × 10⁴ TCID₅₀ SARS-CoV-2^MA10. (A) Relative body weight change overtime after infection. Data were analyzed by 2-way ANOVA with Dunett’s multiple-comparison test, comparing the knockout groups to the WT group. Data were pooled from 6 independent experiments (n = 12–31 mice/group). *P < 0.05; color of asterisk corresponds to the genotype. (B) Viral copy numbers were determined by RT-qPCR in oropharyngeal swabs or tissue homogenates. (C) Viral titers in lung homogenates. (B and C) Data were log₁₀ transformed and analyzed by 1-way ANOVA with Dunnett’s multiple-comparison test comparing the knockout groups to the WT group at each time point. Data pooled from 7 independent experiments (2 dpi, n = 16–19/group; 4 dpi, n = 10–17/group). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 3. NE and CatG regulate levels of cytokines and chemokines in the lungs.

(A and B) WT, CatG^–/–, NE^–/–, PR3^–/–, and NE.CatG^–/– mice were inoculated intranasally with 1 × 10⁴ TCID₅₀ SARS-CoV-2^MA10. Cytokines (A) and chemokines (B) were measured in lung homogenates on the indicated dpi. Data were analyzed by 1-way ANOVA with Dunnett’s multiple-comparison test comparing the groups of knockout mice to the WT group at each time point. Data pooled from 7 independent experiments (2 dpi, n = 8–11/group; 4dpi, n = 10–19/group). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 4. Increased acute pathological lesions in NE.CatG^–/– mice after SARS-CoV-2 infection.

(A) Representative histopathological lesions in the lungs of mice inoculated with 1 × 10⁴ TCID₅₀ SARS-CoV-2^MA10 at the indicated dpi. The black dots indicate consolidated areas corresponding to interstitial pneumonia, and the arrowheads indicate peribronchiolar and perivascular mononuclear infiltrates. H&E staining. Scale bars: 1 mm. (B) Lung inflammation score was analyzed by 1-way ANOVA with Dunnett’s multiple-comparison test comparing the knockout groups to the WT group. *P < 0.05. Data were pooled from 7 independent experiments (2 dpi, n = 8–11/group; 4 dpi, n = 10–19/group).