Cytokine Signature Induced by SARS-CoV-2 Spike Protein in a Mouse Model

Abstract

Although COVID-19 has become a major challenge to global health, there are currently no efficacious agents for effective treatment. Cytokine storm syndrome (CSS) can lead to acute respiratory distress syndrome (ARDS), which contributes to most COVID-19 mortalities. Research points to interleukin 6 (IL-6) as a crucial signature of the cytokine storm, and the clinical use of the IL-6 inhibitor tocilizumab shows potential for treatment of COVID-19 patient. In this study, we challenged wild-type and adenovirus-5/human angiotensin-converting enzyme 2-expressing BALB/c mice with a combination of polyinosinic-polycytidylic acid and recombinant SARS-CoV-2 spike-extracellular domain protein. High levels of TNF-α and nearly 100 times increased IL-6 were detected at 6 h, but disappeared by 24 h in bronchoalveolar lavage fluid (BALF) following immunostimulant challenge. Lung injury observed by histopathologic changes and magnetic resonance imaging at 24 h indicated that increased TNF-α and IL-6 may initiate CSS in the lung, resulting in the continual production of inflammatory cytokines. We hypothesize that TNF-α and IL-6 may contribute to the occurrence of CSS in COVID-19. We also investigated multiple monoclonal antibodies (mAbs) and inhibitors for neutralizing the pro-inflammatory phenotype of COVID-19: mAbs against IL-1α, IL-6, TNF-α, and granulocyte-macrophage colony-stimulating factor (GM-CSF), and inhibitors of p38 and JAK partially relieved CSS; mAbs against IL-6, TNF-α, and GM-CSF, and inhibitors of p38, extracellular signal-regulated kinase, and myeloperoxidase somewhat reduced neutrophilic alveolitis in the lung. This novel murine model opens a biologically safe, time-saving avenue for clarifying the mechanism of CSS/ARDS in COVID-19 and developing new therapeutic drugs.

Keywords: COVID-19; SARS-CoV-2; acute respiratory distress syndrome; cytokine storm syndrome; murine model.

Figure 1

SARS-CoV-2 mimic–induced acute lung injury in BALB/c mice. Schematic diagrams of the delivery of polyinosinic-polycytidylic acid (poly[I:C]) + recombinant SARS-CoV-2 spike-extracellular domain protein (ECD) (poly[I:C] + SP) by intratracheal administration to (A) adenovirus 5-human angiotensin-converting enzyme 2 (ad5-hACE2) transgenic and (B) wild-type Balb/c mice. (C) Surface plasmon resonance sensorgrams of transgenic (left panel) and wild-type (right panel) mice. (D) Histologic characteristics of lung injury and interstitial pneumonia induced by poly(I:C) + SP at 6, 24, and 48 h after challenge. Black scale bar = 100 µm; blue scale bar = 50 µm. (E) In vivo small animal magnetic resonance images documenting the pleural effusion induced by poly(I:C) +SP at 6, 24, and 48 h after challenge. Control group mice (Saline) received saline + poly(I:C) challenge and were assessed at 24 h.

Figure 2

Development of angiotensin-converting enzyme 2-humanized (hACE2) mice sensitized to SARS-CoV-2 mimic challenge. (A) Western blot assay of hACE2 expression using antibodies against GAPDH, ACE2 and the DYKDDDDK (FLAG) tag. Five days post virus transduction, mice were challenged with SARS-CoV-2 mimic via three different deliver methods: nasal (I.N.), intratracheal (I.T.), and intravenous (I.V) injection. Inflammation-related cytokines in bronchoalveolar lavage fluid (BAL); α-FLAG and α-GAPDH cropped from different parts of the same gel, but different exposures, α-ACE2 cropped from different gels (B) and neutrophil infiltration (C) in hACE2 mice. (D) Western blot assay in HEK293T cells transduced with adenovirus 5 (Ad5)-hACE2 or Ad5-EGFP at a multiplicity of infection (MOI) of 25 or 50, or transiently transfected with PCDNA3.1 carrying hACE2 or mouse ACE2 (mACE2), using antibodies against β-actin (control), ACE2, and the FLAG tag; α-ACE2 and α-actin cropped from different parts of the same gel, but different exposures, α-FLAG cropped from different gels. (E) Recombinant SARS-CoV-2 spike-extracellular domain protein (rSpike-ECD) was incubated with HEK293T cells transiently transfected with hACE2 or mACE2 for 1 h at room temperature and detected by flow cytometry using allophycocyanin (APC)-labeled anti-DYKDDDDK antibody. Neutrophil infiltration (F) measured by flow cytometry and inflammatory-related cytokine profile (G) after challenge with SARS-CoV-2 mimic (poly[I:C] + SP), poly(I:C) alone, S alone, saline control (each n ≥ 4), via nasal injection at 5 days post virus transduction). *P < 0.05; **P < 0.01; ***P < 0.001, one-way ANOVA with a post hoc Bonferroni test.

Figure 3

Polyinosinic-polycytidylic acid (poly[I:C]) + recombinant SARS-CoV-2 spike-extracellular domain protein (SP)-induced cytokine-release storm in the mouse lung. (A) Total cell count in the BAL from poly(I:C) + 5-, 10-, or 15-µg SARS-CoV-2 spike protein-challenged mice; control group, Saline (each n ≥ 3); (B) Cellular composition in BAL. (C) Production of IL-6, IL-1α, and tumor necrosis factor (TNF)α in the bronchoalveolar lavage fluid (BAL) from poly(I:C) + SP-challenged mice. 6 or 24 h after SARS-CoV-2 mimic challenged mice. Control group; Saline (each n ≥ 4); (D) percentage of neutrophils in BAL; (E) Levels of dsDNA in the BAL. Single cells dissociated from the lung tissue of poly(I:C) + SP-challenged mice. (F) IL-6 concentration from cell culture medium. (G) Concentrations of IL-6, IL-1α, and tumor necrosis factor (TNF)α in bronchoalveolar lavage fluid (BALF) after challenge with SARS-CoV-2 mimic (poly[I:C] + SP), poly(I:C) alone, SP alone, FC alone, saline control (each n ≥ 5), (H) Levels of dsDNA in the BALF. *P < 0.05; **P < 0.01; ***P < 0.001, ns (non-significant), p > 0.05, one-way ANOVA with a post hoc Bonferroni test.

Figure 4

Neutralization and blocking of inflammatory cytokines. Concentrations of IL-6, IL-1α, and tumor necrosis factor (TNF)α in bronchoalveolar lavage fluid (BALF) of mice after treatment with the following monoclonal antibodies (mAbs): anti–IL-6 (n ≥ 3) (A), anti–IL-6R (n ≥ 6) (B), anti-TNF-α (n ≥ 5) (C), anti-TNF receptor 2 (TNFR2) (n ≥ 5) (D), anti–IL-1α (n ≥ 6) (E), and anti–granulocyte-macrophage colony-stimulating factor (GM-CSF) (n ≥ 5). (F) Cellular composition of BALF of mice treated with the following mAbs: anti–IL-6 (G), anti–IL-6 receptor (IL-6R) (H), anti-TNF-α (I), anti-TNFR2 (J), anti–IL-1α (K), and anti-GM-CSF (L). (M) Levels of dsDNA in the BALF; *P < 0.05; **P < 0.01, unpaired T test. Poly(I:C) stimulation of IL-6 production in the mouse macrophage cell line RAW264.7 showing the IL-6 concentration in the culture medium (N); **P < 0.01, one-way ANOVA with a post hoc Dunnet test. (O) Histologic characteristics of mouse lung injury and interstitial pneumonia. Black scale bar = 100 µm; blue scale bar = 50 µm.

Figure 5

Blocking of inflammation-related signaling pathways. Effect of treatment with the indicated inhibitors on (A) the concentrations of IL-6, IL-1α, and tumor necrosis factor (TNF)α in mouse bronchoalveolar lavage fluid (BALF) (each n ≥ 5); (B) Cellular composition in BALF; (C) levels of dsDNA in BALF; *P < 0.05, unpaired T test. (D) poly(I:C) stimulation of IL-6 production in the cell culture medium of mouse primary macrophages; *P < 0.05, ns (non-significant), p > 0.05, one-way ANOVA with a post hoc Dunnet test, and (E) histologic characteristics of lung injury and interstitial pneumonia. Black scale bar = 100 µm; blue scale bar = 50 µm.

Figure 6

Graphical abstract of the SARS-CoV-2 mimic and potential therapeutic applications of monoclonal antibodies (mAbs) and inhibitors.