SARS-CoV-2 S protein activates NLRP3 inflammasome and deregulates coagulation factors in endothelial and immune cells

Abstract

Background: Hyperinflammation, hypercoagulation and endothelial injury are major findings in acute and post-COVID-19. The SARS-CoV-2 S protein has been detected as an isolated element in human tissues reservoirs and is the main product of mRNA COVID-19 vaccines. We investigated whether the S protein alone triggers pro-inflammatory and pro-coagulant responses in primary cultures of two cell types deeply affected by SARS-CoV-2, such are monocytes and endothelial cells.

Methods: In human umbilical vein endothelial cells (HUVEC) and monocytes, the components of NF-κB and the NLRP3 inflammasome system, as well as coagulation regulators, were assessed by qRT-PCR, Western blot, flow cytometry, or indirect immunofluorescence.

Results: S protein activated NF-κB, promoted pro-inflammatory cytokines release, and triggered the priming and activation of the NLRP3 inflammasome system resulting in mature IL-1β formation in both cell types. This was paralleled by enhanced production of coagulation factors such as von Willebrand factor (vWF), factor VIII or tissue factor, that was mediated, at least in part, by IL-1β. Additionally, S protein failed to enhance ADAMTS-13 levels to counteract the pro-coagulant activity of vWF multimers. Monocytes and HUVEC barely expressed angiotensin-converting enzyme-2. Pharmacological approaches and gene silencing showed that TLR4 receptors mediated the effects of S protein in monocytes, but not in HUVEC.

Conclusion: S protein behaves both as a pro-inflammatory and pro-coagulant stimulus in human monocytes and endothelial cells. Interfering with the receptors or signaling pathways evoked by the S protein may help preventing immune and vascular complications driven by such an isolated viral element. Video Abstract.

Keywords: Blood coagulation factors; Endothelial cells; Monocytes; NLRP3 inflammasome; SARS-CoV-2.

Fig. 1

S protein promotes NLRP3 priming and NF-κB activation in primary HUVEC cultures. Human umbilical vein endothelial cells (HUVEC) were treated with viral S protein at 7, 35 and 70 nM and IL-1β at 2.5 ng/mL for 18 h and the protein levels of the NLRP3 inflammasome system components A NLRP3 (n = 5), B pro-IL-1β (n = 4), C pro-casp-1 (n = 5), and D ASC protein levels (n = 3) were analyzed in total cell lysates by Western blot. In addition, NF-κB activation was quantified by means of E phospho-p65 (P-p65) protein levels by Western blot (n = 3) and F visualized by indirect immunofluorescence as the translocation of p65 (green) to cell nuclei countersatained with DAPI (blue). Representative images from confocal microscopy of p65 immunofluorescence staining are shown. Scale bar represents 50 μm (G). For quantification of cells with nuclear localization of p65 at least 200 cells per treatment were counted (n = 4). For Western blots, representative gels are shown on top of the corresponding graphs, with β-actin used as a loading control. Bar graphs represent mean ± SEM. Statistical differences were analyzed by t-test. *p < 0.05 versus unstimulated control

Fig. 2

S protein promotes the activation of the NLRP3 inflammasome in HUVEC. Human umbilical vein endothelial cells (HUVEC) were treated with viral S protein (35 nM), IL-1β (2.5 ng/mL) or the S1 fragment of S protein (35 nM) for 18–24 h. A The formation of toroidal specks corresponding to activated NLRP3 inflammasome were visualized by indirect immunofluorescence against ASC (red) using a confocal microscope. Nuclei were counter-stained with DAPI (blue). Scale bar represents 50 μm B NLRP3 inflammasome activation was quantified by manual blind scoring of 27 radial distributed fields per sample as the number of ASC speck-positive cells. White arrows indicate speck positive cells. Scale bar represents 50 μm. In addition, the cellular protein levels of C active cleaved caspase-1 (casp-1) (n = 6) and D mature IL-1β (n = 4). E IL-1β released to the cell supernatants was quantified by ELISA (n = 5–8). F Gasdermin D (GSDMD) and cleaved GSDMD-NT (n = 8) were determined by Western blot in total cell lysates from HUVEC treated with 7, 35 and 70 nM S protein or 2,5 ng/mL IL-1β. Representative gels are shown on top of the corresponding graphs, with β-actin used as a loading control. Bar graphs represent mean ± SEM. Statistical differences were analyzed by t-test. *p < 0.05 versus control with no S protein

Fig. 3

The activation NLRP3 inflammasome and the release of IL-1β increase coagulation factors mediated by S protein. Human umbilical vein endothelial cells (HUVEC) were treated with viral S protein at 7, 35 and 70 nM or IL-1β at 2.5 ng/mL for 18 h. The levels of von Willebrand Factor (vWF) were determined A by Western blot in total cell lysates (n = 6), and B visualized in HUVEC stimulated with S (35 nM) or IL-1β (2.5 ng/mL) by indirect immunoflorescence (red). Representative images are shown in which vWF can be seen both as granules within the HUVEC cytoplasm and multimeric filaments in the extracellular space. Nuclei were counter-stained with DAPI (blue). Scale bar represents 50 μm. C vWF was also measured by ELISA in cell supernatants (n = 4). In the same experimental conditions, the protein levels of D ADAMTS-13, E FVIII, and F TF were quantified in total cell lysates by Western blot. Representative gels are shown on top of the corresponding graphs, with β-actin used as a loading control. Bar graphs represent mean ± SEM. Statistical differences were analyzed by t-test. *p < 0.05

Fig. 4

NF-κB pathway and NLRP3 inflammasome components in S protein stimulated monocytes. Enriched monocytes from healthy donor were cultured under control conditions or stimulated with S protein 15 nM (S15). A NF-κB mRNA expression analysis by qPCR in enriched monocytes cultured for 16 h (n = 12). B Western blot protein values of phosphorylated NF-κB p65 particle (pp65) relative to β-actin protein values from lysates of enriched monocytes cultured for 16 hours (n = 3). C mRNA expression analysis by qPCR of TNF-α (n = 12) and IL-6 (n = 9) in enriched monocytes cultured for 3 hours. D IL-6 concentration in supernatant from enriched monocytes cultured for 16 hours (n = 12). E Left: CD14⁺ gated-cells normalized mean fluorescent intensity (MFI) of NLRP3 (n = 20); Right: normalized NLRP3 to β-actin protein ratio determined by Western blot from enriched monocyte’s lysates cultured for 16 hours (n = 7). F CD14⁺ gated cells normalized mean fluorescent intensity (MFI) of ASC (n = 12). G Normalized amount of active caspase positive (Casp-1⁺) CD14⁺ cells analyzed by flow cytometry from enriched monocytes cultured for 16 hours (n = 20). H Supernatant IL-1β concentration measured by CBA (n = 12). I Normalized TF to β-actin protein ratio determined by Western blot from enriched monocytes lysates cultured for 16 hours (n = 5). Differences were analyzed using Wilcoxon’s paired test. J Enriched monocytes were stimulated with S protein or control, treated or not with MCC950 and cultured for 16 h. Left: CD14⁺ gated-cells normalized mean fluorescent intensity (MFI) of NLRP3 (n = 12); center: normalized amount of Casp-1⁺ CD14⁺ cells analyzed by flow cytometry (n = 15); right: supernatant IL-1β concentration measured by CBA (n = 6). Differences were analyzed by repeated measures ANOVA and Tukey’s multiple comparison test. All data are represented as mean ± Standard Error of the Mean (SEM). Only statistically significant differences are stated: *p < 0.05, **p < 0.01., ***p < 0.001 and ****p < 0.0001

Fig. 5

Effect of S protein stimulation over monocytes treated with TLR4 inhibitor TAK242. (A-I) Enriched monocytes from healthy donor were cultured for 16 h and were stimulated or not with S protein (S15) and treated or not with TLR4 inhibitor (TAK242). A NF-κB mRNA expression analysis by qPCR. (n = 9). B Western blot protein values of phosphorylated NF-κB p65 particle (pp65) relative to β-actin protein values (n = 3). C mRNA expression analysis by qPCR of TNF-α (n = 10) and IL-6 (n = 7). D IL-6 concentration in supernatant from enriched monocytes (n = 10). E Left: CD14⁺ gated-cells normalized mean fluorescent intensity (MFI) of NLRP3 (n = 20); right: normalized NLRP3 to β-actin protein ratio determined by Western blot (n = 7). F CD14⁺ gated cells normalized mean fluorescent intensity (MFI) of ASC (n = 12). G Normalized amount of active caspase positive (Casp-1⁺) CD14⁺ cells analyzed by flow cytometry (=20). H Supernatant IL-1β concentration measured by CBA (n = 12). I Normalized TF to β-actin protein ratio determined by Western blot (n = 5). Differences were analyzed by repeated measures ANOVA and Tukey’s multiple comparison test. J-K Enriched monocytes from healthy donor were transfected in absence (control) or presence of TLR4 siRNA (siTLR4) and cultured for 16 h with S15 or not. J Normalized amount of CD14⁺ cells expressing surface TLR4 (n = 7). K Left: CD14⁺ gated-cells normalized mean fluorescent intensity (MFI) of NLRP3 (n = 7): right: normalized amount of active casp-1⁺ CD14⁺ cells analyzed by flow cytometry (n = 7). Differences were analyzed using Wilcoxon’s paired test. All data are represented as mean ± SEM. Only statistically significant differences are stated: *p < 0.05, **p < 0.01, and ***p < 0.001

Fig. 6

Diagram of cellular pathways activated by SARS-CoV-2 S protein and the targets for pharmacological interference. In human monocytes and HUVEC, S protein as an isolated element activates NF-κB, promotes the release of pro-inflammatory cytokines and triggers the priming and activation of the NLRP3 inflammasome system, leading to the formation and release of mature IL-1β. This cytokine can in turn act on IL-1R, thus fueling and amplifying an auto-inflammatory loop. Moreover, S protein enhances the cellular content of factors involved in coagulation processes, including vWF, FVIII and TF. In human endothelial cells, S protein fails to over-express the protease ADAMS-13 in order to counteract the hypercoagulation capacity of vWF multimers. While these effects of S protein are mediated by TLR4 in human monocytes, the receptors involved in HUVEC remain to be elucidated. Drugs such as anakinra, TAK242 (resotorvid), MCC950, or anti-coagulant drugs could interfere with the deleterious pro-inflammatory and pro-coagulant actions of S protein