Recombinant human plasma gelsolin reverses increased permeability of the blood-brain barrier induced by the spike protein of the SARS-CoV-2 virus

Abstract

Background: Plasma gelsolin (pGSN) is an important part of the blood actin buffer that prevents negative consequences of possible F-actin deposition in the microcirculation and has various functions during host immune response. Recent reports reveal that severe COVID-19 correlates with reduced levels of pGSN. Therefore, using an in vitro system, we investigated whether pGSN could attenuate increased permeability of the blood-brain barrier (BBB) during its exposure to the portion of the SARS-CoV-2 spike protein containing the receptor binding domain (S1 subunit).

Materials and methods: Two- and three-dimensional models of the human BBB were constructed using the human cerebral microvascular endothelial cell line hCMEC/D3 and exposed to physiologically relevant shear stress to mimic perfusion in the central nervous system (CNS). Trans-endothelial electrical resistance (TEER) as well as immunostaining and Western blotting of tight junction (TJ) proteins assessed barrier integrity in the presence of the SARS-CoV-2 spike protein and pGSN. The IncuCyte Live Imaging system evaluated the motility of the endothelial cells. Magnetic bead-based ELISA was used to determine cytokine secretion. Additionally, quantitative real-time PCR (qRT-PCR) revealed gene expression of proteins from signaling pathways that are associated with the immune response.

Results: pGSN reversed S1-induced BBB permeability in both 2D and 3D BBB models in the presence of shear stress. BBB models exposed to pGSN also exhibited attenuated pro-inflammatory signaling pathways (PI3K, AKT, MAPK, NF-κB), reduced cytokine secretion (IL-6, IL-8, TNF-α), and increased expression of proteins that form intercellular TJ (ZO-1, occludin, claudin-5).

Conclusion: Due to its anti-inflammatory and protective effects on the brain endothelium, pGSN has the potential to be an alternative therapeutic target for patients with severe SARS-CoV-2 infection, especially those suffering neurological complications of COVID-19.

Keywords: Blood–brain barrier; COVID-19; Microfluidics; Plasma gelsolin (pGSN); SARS-CoV-2; Tissue engineering.

Fig. 1

Plasma gelsolin (pGSN) significantly reduces disruption of the blood–brain barrier caused by the SARS-CoV-2 Spike protein S1 subunit in the 2D Transwell permeability assay. Confluent monolayer of human cerebral microvascular endothelial cells (hCMEC/D3) seeded on a Transwell semi-permeable membrane exposed to tested compounds added to the upper chamber (blood) (Panel Ai). The functional state of the cells as a barrier was evaluated with transendothelial electrical resistance (TEER) measurement and Dextran-FITC permeability assay (Panel Aii). The dextran-FITC intensity was measured in the lower chamber (brain) that migrated from the upper chamber (blood) in time (Panels B, D). Change in TEER was measured using Epithelial Voltohmeter EVOM2 (Panels C and D). The data represent the mean ± SEM of four (n = 4, with 2 inserts used per condition each time) independent experiments. * and ^ indicate statistical significance at p ≤ 0.05 compared to CT and S1, respectively, by one-way ANOVA and Tukey post hoc test

Fig. 2

3D blood–brain barrier model. A Schematic of a microfluidic device with the location of inlet/outlet ports and ports to fill the hydrogel reservoir. B Picture of the PDMS device. Scale bar, 1 cm. C Permeability testing with 4 kDa Dextran-FITC configuration. D TEER measurements

Fig. 3

Plasma gelsolin reverses the destructive effect of SARS-CoV-2 Spike protein S1 subunit on blood–brain barrier function in the 3D flow model. Confocal images of hCMEC/D3 cells (panel A), tight junction protein ZO-1 (red), and nuclei (blue). Permeability coefficient measured from dextran experiments for endothelial channels exposed to S1 and S1 + pGSN (Panel B). TEER measurement results (Panel C). Images demonstrating the measurement of vessel permeability using 4 kDa FITC-dextran (green) and the effects of the S1 and S1 + pGSN (Panel D). Barrier permeability and TEER measurement were performed after 4 h of perfusion with 10 nM of S1 and S1 + pGSN. The data represent the mean ± SEM of four independent experiments (n = 4). * and ^ indicate statistical significance at p ≤ 0.05 compared to CT and S1, respectively, by one-way ANOVA and Tukey post hoc test

Fig. 4

Plasma gelsolin improves migration of hCMEC/D3 upon SARS-CoV-2 Spike protein S1 subunit treatment in wound-healing assay. Images of endothelial cells in a wound healing setting (Panel A), the yellow color indicates wound width, which is quantitatively shown in Panel B. Western blot quantitative analysis of VEGRF2 expression in hCMEC/D3 cells after 24 h stimulation with pGSN, S1, and S1 + pGSN (Panel C). The data represent the mean ± SEM of four independent experiments (n = 4). * and ^ indicate statistical significance at p ≤ 0.05 compared to CT (100%) and S1, respectively, by one-way ANOVA and Tukey post hoc test

Fig. 5

Secretion of inflammatory mediators by hCMEC/D3 cell line stimulated with SARS-CoV-2 Spike protein S1 subunit and S1 + pGSN after 6 and 24 h. Expression of IL-2 (A, G), IL-6 (B, H), IL-8 (D, I), TNF-α (D, J), INF-γ (E, K), and GM-CSF (F, L). Protein expression after 6 h of treatment is presented on Panels A–F, and expression after 24 h is on Panels G–L. Alternation of inflammatory response was monitored using a magnetic bead-based assay. The data represent the mean ± SEM of three independent experiments (n = 3). * and ^indicate statistical significance at p ≤ 0.05 compared to CT and S1, respectively, by one-way ANOVA and Tukey post hoc test

Fig. 6

Spike protein S1 subunit selectively decreases cell junction proteins in endothelial cells, forming a blood–brain barrier. Western blot bands (Panel A) and protein expression fold change (Panel B). The data represent the mean ± SEM of four independent experiments (n = 4). * and ^indicate statistical significance at p ≤ 0.05 compared to CT (100%) and S1, respectively, by one-way ANOVA and Tukey post hoc test

Fig. 7

Plasma gelsolin (pGSN) inhibits NF-κB activation by the SARS-CoV-2 Spike protein S1 subunit in hCMEC/D3 cells. Panel A shows a Log₂ fold change heat map for genes involved in VEGF signaling and activation of blood–brain barrier endothelial cells upon 6 h stimulation with pGSN [250 µg/mL], S1 [10 nM] and S1 [10 nM] + pGSN [250 µg/mL]. Log₂ fold change was calculated based on delta Ct values compared to the control samples. Warmer colors imply increased expression, while cold reflects decreased expression. The assay was performed twice in quadruplicate; data within frames shows Log2FC for every tested gene. Panel B shows the schematic representation of signaling pathways triggered by the SARS-CoV-2 Spike protein S1 subunit in hCMEC/D3 cells. (1) Spike protein interacts with the given receptor on a cell membrane, which (2) activates the catalytic effect of PI3K on PIP2, (3) enzymatically transforming it to PIP3, which is possibly inhibited by plasma gelsolin given its direct binding to PIP2. (4) PIP3 binds to AKT, promoting its phosphorylation and activation. (5, 6) Activated AKT regulates transcriptional activity of MAPK kinases and NF-κB by inducing phosphorylation and degradation of inhibitor of κB (IκB). (7) MAPK initiates the downstream induction of NF-κB and its translocation (8) to the nucleus. (9, 10) NF-κB, after activation, triggers the transcription of various genes and thereby regulates inflammation