Sustained vascular inflammatory effects of SARS-CoV-2 spike protein on human endothelial cells

doi:10.21203/rs.3.rs-5003230/v1

Download PDF

Research Article

Sustained vascular inflammatory effects of SARS-CoV-2 spike protein on human endothelial cells

https://doi.org/10.21203/rs.3.rs-5003230/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection has been associated with systemic inflammation and vascular injury, which contribute to the development of acute respiratory syndrome (ARDS) and the mortality of COVID-19 infection. Moreover, multiorgan complications due to persistent endothelial dysfunction have been suspected as the cause of post-acute sequelae of SARS-CoV-2 infection. Therefore, elucidation of the vascular inflammatory effect of SARS-CoV-2 will increase our understanding of how endothelial cells (EC) contribute to the short- and long-term consequences of SARS-CoV-2 infection. Here, we investigated the interaction of SARS-CoV-2 spike protein with human ECs from aortic (HAoEC) and pulmonary microvascular (HPMC) origins, cultured under physiological flow conditions. We showed that the SARS-CoV-2 spike protein triggers prolonged expression of cell adhesion markers in both ECs, similar to the effect of TNF-α. SARS-CoV-2 spike treatment also led to the release of various chemokines observed in severe COVID-19 patients. Moreover, increased binding of leucocytes to the endothelial surface and a procoagulant state of the endothelium were observed. Transcriptomic profiles of SARS-CoV-2 spike-activated HPMC and HAoEC showed prolonged upregulation of genes and pathways associated with responses to virus, cytokine-mediated signaling, pattern recognition, as well as complement and coagulation pathways. Our findings support experimental and clinical observations of the vascular consequences of SARS-CoV-2 infection and highlight the importance of EC protection as one of the strategies to mitigate the severe effects as well as the possible post-acute complications of COVID-19 disease.

SARS-CoV-2

spike protein

endothelial cells

vascular inflammation

Although severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) primarily infects the respiratory tract, there is evidence that the causative agent of COVID-19 disease directly interacts with the vasculature [1], [2], [3]. The observed thrombotic symptoms in severe COVID-19 patients, with arterial and venous thrombosis contributing to their mortality, highlights the substantial consequences of SARS-CoV-2 on the endothelium [4], [5], [6]. Beyond the lungs, vascular coagulopathy and inflammation leading to cardiovascular and neurological complications have been reported in kidney, heart, and brain [7], [8], [9], [10]. Moreover, preexisting cardiovascular disorders associated with endothelial dysfunction are one of the main comorbidities correlated with severe COVID-19 disease outcomes and deaths [11], [12].

Endothelial cells (ECs) are constantly exposed to plasma components and have a crucial role in maintaining vascular homeostasis, including regulating the coagulation state, as well as local and general inflammation. Although direct infection of SARS-CoV-2 of ECs has been described, findings regarding the susceptibility of ECs for infection and the expression of the receptor angiotensin-converting enzyme 2 (ACE2) and transmembrane protease serine 2 (TMPRSS2) are controversial [13], [14], [15], [16]. Several reports described that SARS-CoV-2 particles, including the spike protein, can interact directly with the endothelium, leading to barrier damage and vascular dysfunction [17], [18], [19]. In the lungs, SARS-CoV-2 spike protein alone can induce infection-like injury and vascular inflammatory responses, including increases in endothelial adhesion molecule expression and recruitment of immune cells [20], [21], [22]. Circulating viral particles in the plasma of COVID-19 patients correlated with disease severity, indicating a high incidence of endothelial exposure, also beyond the respiratory tract [6], [23], [24]. Moreover, spike protein present in the plasma can be taken up and deposited in various organs, suggesting that systemic spread of SARS-CoV-2 is facilitated by ECs [25], [26].

Several studies have documented possible mechanisms through which SARS-CoV-2 causes vascular injury and endothelial dysfunction, including downregulation of ACE2 and disruption of mitochondrial function [6], [22]. Others have described that the vascular damage effect of SARS-CoV-2 spike protein is caused by endothelial glycocalyx disruption and engagement of integrins, leading to the activation of TGF-b signaling [18], [27], [28]. Robles et al. showed that SARS-CoV-2 spike protein can induce the nuclear translocation of NF-kB, leading to expression of procoagulant and proinflammatory responses of ECs [29].

The presence of biomarkers for systemic inflammation and an increased procoagulant state in the plasma are prominently observed in acute COVID-19 infection [1], [4], [30], [31]. Moreover, systemic inflammation was still observed, when viral infection was cleared in COVID-19 patients, potentially contributing to the multi-organ disorders in the post-acute sequelae of SARS-CoV-2 (PASC), also known as “Long COVID” [32], [33], [34], [35]. In this context, persistent endothelial dysfunction has been identified as a significant driver of prolonged non-respiratory symptoms and PASC [36], [37]. Therefore, understanding the extent of vascular inflammation as well as the sustained changes to the endothelial functions caused by SARS-CoV-2 is necessary to get a better grip on the consequences of COVID-19 infection.

In this study, we investigated the interaction of SARS-CoV-2 spike protein with primary human ECs from two anatomical origins: aortic (HAoEC) as well as lung microvascular (HPMC), which were cultured under physiological flow to simulate the vascular environment. We showed that SARS-CoV-2 spike protein elicited prolonged inflammatory responses in both the macro- and microvascular ECs. We also investigated pathways mediating the interaction of SARS-CoV-2 spike protein and cellular dynamic of EC post-interaction, elucidating possible persistent effect of SARS-CoV-2 on the endothelium.

Primary endothelial cells and cell lines

Primary human pulmonary microvascular cells (HPMC, Promocell, C-12281) and primary human aortic endothelial cells (HAoEC, Promocell, C-12271) were obtained from a commercial supplier. For expansion, the cells were grown in a fibronectin-coated flask using complete media (Endothelial growth medium MV2 (Promocell, C-22011), supplemented with heat-inactivated FBS to a total concentration of 10%, 100 IU penicillin, and 100 µg/ml streptomycin). Cells were maintained at 37°C in a humidified incubator supplied with 5% CO2. For each experiment, cells at passage 5 or 6 were used.

Vero-E6 (kindly provided by M. Müller, Charite Berlin) were cultured om Dulbecco’s modified Eagle’s medium-GlutaMAX (DMEM-GlutaMAX, Gibco) supplemented with 10% heat-inactivated FBS, 1x non-essential amino acids (Gibco), 100 IU penicillin and 100 µg/ml streptomycin. Cells were maintained at 37°C in a humidified incubator with 5% CO2.

Calu-3 cells (American Type Culture Collection (ATCC), HTB-55) were propagated on DMEM-GlutaMAX supplemented with 10% heat-inactivated FBS, 100 IU/mL penicillin, 100 µg/ml streptomycin, and 1x non-essential amino acids. Cells were maintained at 37°C in a humidified incubator with 5% CO2.

SARS-CoV-2 viral stock

SARS-CoV-2 (SARS-CoV-2/München-1.1/2020/929) was propagated in Calu-3 cells. Briefly, 24h before infection, Calu-3 cells were seeded in T75 flasks. The day after, they were infected with the virus stock at the MOI of 0.01. 3 days post infection, the supernatant of the infected Calu-3 cells was collected and cleared by centrifugation (500 x g, 5 min). To generate high titer viral stock and to eliminate possible influence of cell culture supernatant to the experiment, the viral particles were isolated using an Intact Virus Precipitation Reagent (Thermofisher, 10720D) according to manufacturers’ protocol. The virus pellet was resuspended in fresh cell culture media, aliquoted, and stored at -80°C prior to use. The titer of the virus stock was determined by TCID50 assay in Vero E6 cells as previously described and calculated according to the Spearman-Kaerber formula [38]. All experiments involving SARS-CoV-2 virus were performed in a biosafety level-3 (BSL-3) laboratory.

Isolation of human peripheral blood mononuclear cells (PBMCs)

Human blood samples were obtained from consenting volunteers, in accordance with local ethic committee’s approval. PBMCs were isolated from EDTA-anticoagulated whole blood by density centrifugation. Briefly, EDTA blood was diluted 1:1 with PBS. 15 ml of diluted EDTA-blood was transferred into a 50 ml tube containing 10 ml of Ficol-Paque (Merck, GE17-1440-02) and centrifuged at 400 x g for 20 min. The PBMC layer was isolated and washed with PBS. Isolated PBMCs were frozen in 10% DMSO and 90% heat-inactivated FBS and stored at -150°C before use.

Infection of HPMCs and HAoEC with SARS-CoV-2 on static conditions

HPMC and HAoEC were seeded on a chamber slide at a seeding density of 50,000 cells/well 1 day prior to infection. SARS-CoV-2 virus stock was diluted to the desired multiplication of infection (MOI) with complete media and 200 µl of diluted virus stock was used to infect the ECs at MOI of 1. As controls, untreated cells were used. The cells were incubated in a humidified incubator at 37°C and with 5% CO₂ for 48h. Thereafter, the inoculum was removed, and the cells were washed three times with PBS, followed by PFA fixation for immunofluorescence analysis.

Culturing and activation of endothelial cells in a microfluidic system

To culture the ECs in a microfluidic system, 100,000 HPMC and HAoEC in 100 µl of growth media were seeded in a µ-slide VI 0.4 (Ibidi, 80606) overnight. Prior to seeding, the µ-slide was coated with 100 µl of 12.5 µg/ml fibronectin (Merck, FC010) for at least 30 min at 37°C. The day after, each channel was connected to a peristaltic pump (Gilson, Minipuls 3) using sterile silicon tubings. Cascade media, which consists of Medium 200 (Gibco) supplemented with 10% FBS, 1% glutamine, 1%BSA, and 4% dextran (Sigma adlrich, 31390), was used as the flow medium. The laminar shear stress was adjusted to 10 dyn/cm² and maintained for 72h in a humidified incubator at 37°C and with 5% CO₂. The media was refreshed every day.

After 72h of EC culture under flow, HPMC and HAoEC were perfused with 4 ml of cascade media containing 1 µg/ml of recombinant SARS-CoV-2 spike protein (Genscript, Z03481), or 1 ng/ml recombinant human TNF-a (RnD Systems, 210-TA-020) for 24h. Thereafter, the cells were washed with cascade medium by continuous perfusion of 4 ml medium. The medium was then refreshed, and the cells were further cultured under flow for a total of 96h after activation. A schematic overview of the experiment is given on Supplementary Fig. 3.

Immunofluorescence analysis

For immunofluorescence staining, cells grown under static or under microfluidic flow were fixed with 4% PFA for 15 min at RT. The cells were washed with PBS, followed by a blocking step at RT with PBS containing 3% BSA. Next, cells were incubated with primary antibody diluted in the antibody solution (1% BSA, 0.05% Tween 20 in PBS) for 2h at RT, or overnight at 4°C. Subsequently, cells were stained using a goat polyclonal antibody against human VE-Cadherin (RnD systems, AF938), mouse antibody against ICAM1 (Abcam, ab2213), and E-Selectin (Sigma, S9555) to visualize the endothelial junctions and activation markers. To stain for the viral nucleoprotein and the presence of double-stranded RNA, a SARS-CoV-2 cross-reacting rabbit antibody against SARS-CoV Nucleoprotein (Rockland 200-401-A50) and a mouse antibody against double-stranded RNA (Scicons, J2) were used, respectively. Subsequently, cells were incubated with secondary antibody labelled with fluorophores: Donkey anti goat IgG (H + L) conjugated with Alexa Fluor 633 (Invitrogen, A21082), donkey anti mouse IgG (H + L) conjugated with Alexa Fluor 488 (Invitrogen, A32766). All secondary antibodies were diluted in the antibody solution and the incubation was performed for 1.5h at RT. The cells were counterstained with DAPI to visualize nuclei. Cells were imaged using a 20x objective on a Zeiss LSM 980 confocal microscope. Figures were analyzed using ImageJ (Version 2.14.0/1-54f) and assembled using FigureJ package [39], [40]. Brightness and contrast were adjusted identically to the corresponding controls. Quantification of the immunofluorescence signal was done by measuring the area above threshold on six images acquired randomly for each channel.

Chemokines analysis

Perfusion media at 24h and 96h was collected and stored at -20°C prior to analysis. The chemokines released by EC in the microfluidic media were simultaneously measured using the Bio-Plex Pro™ Human Chemokine Panel 40-Plex kit (Biorad Laboratories, 171AK99MR2) according to the manufacturers’ protocol. The fluorescence signal was measured using the Bio-Plex 3D Suspension Array System. Chemokines expression fold was calculated as a ratio between spike- and TNF-α-treated versus mock.

Bulk-RNA sequencing and data analysis

Total cellular RNA from mock-, spike-, and TNF-α-activated HAoEC and HPMC cultures were isolated using the NucleoSpin RNA kit (Macherey Nagel, 740955) according to the manufacturer’s guidelines. The total RNA was quantified with QuantifluorÒ RNA system (Promega, E3310) according to the manufacturer’s protocol. Bulk RNA barcoding and sequencing (BRB-seq)was performed by Alithea Genomics as described previously [41]. Briefly, a total of 200 ng of cellular RNA from four independent biological replicates was used for the generation of BrB-seq libraries, followed by sequencing on an Illumina HiSeq 4000 platform to a depth of approximately 5 million raw reads per sample. Demultiplexing, alignment, and count matrix generation was performed using the BRB-seq pipeline.

The libraries were normalized and log1P transformed using Seurat package (Version 4.3.0). Differential expressed gene analysis was performed with FindMarkers() function in Seurat (test.use="DESeq2") with an adjusted p-value cutoff of 0.05 and absolute logFC cutoff of 0.25. Visualizations of overlapping DEGs amongst samples were performed using UpSetR (version 1.4.0), additionally an online tool provided by VIB/UGent (https://bioinformatics.psb.ugent.be/webtools/Venn/) was used. Pathway enrichment analysis was performed using clusterProfiler (version 4.6.2) with a false discovery cutoff of 0.05. KEGG analysis was performed using pathview (version 1.38.0). Further data analysis and visualization was performed using a variety of additional packages in R.

Microfluidic PBMC adhesion assays

Characterization of PBMC Adhesion was performed as previously described [42]. To trace PBMC adhesion on EC using microscopy technique, frozen PBMCs were thawed, washed with PBS, and labelled with CFSE (Thermofisher, C34554) according to the manufacturer’s protocol. After the activation of the EC in the microfluidic channels as described above, the cells were stained for nuclei by perfusing the cells with media spiked Hoechst 3342 (Tocris, 23491-52-3). The EC in the microfluidic channels were subsequently perfused with 1 million/ml CFSE-labelled PBMCs at a shear stress of 0.2 dyn/cm². Using confocal microscopy (Zeiss LSM 980), images were acquired every 3s for 20 min during PBMC perfusion. PBMC adhesion to the surface of the ECs was defined as the number of cells immobilized to the surface of the ECs for at least 3 frames. Cell counting was done manually for each time-lapse recording.

Microfluidic coagulation assays

Coagulation assays to measure the clotting time after activation of the ECs in the microfluidic system was performed as previously described [42]. Cells were perfused with Hoechst 3342 to visualize the nuclei. Thereafter, the cells were washed with calcium- and magnesium- free PBS and subsequently perfused with human citrated plasma spiked with 15 µg/ml AF488 labeled human fibrinogen (Thermofisher, F13191). The plasma was recalcified with 25mM CaCl₂ immediately before imaging. Perfused channels were imaged every 5s up to 20 min using a time-lapse program on a confocal microscope (Zeiss LSM 980). The coagulation time was determined when complete occlusion of the channel occurred and saturated signal of fibrinogen 488 appeared on the acquisition frame.

Statistical analysis

All data are presented as the mean ± standard deviation (SD). The statistical analysis was performed using GraphPad Prism 10 (version 10.0.2). For multiple comparisons ordinary one-way ANOVA followed by Fisher LSD test were performed. All experiments were independently replicated at least three times.

SARS-CoV-2 spike protein triggered prolonged expression of ICAM1 in aortic and lung microvascular endothelial cells

To characterize the extent of vascular inflammatory effects of SARS-CoV-2 spike protein, we utilized HPMC and HAoEC as representatives of two types of ECs from different anatomical regions. In line with earlier studies, we showed that these ECs cannot be productively infected by active SARS-CoV-2 virus (Supplementary Fig. 1) [16]. Neither nucleoprotein (NP) nor double-stranded RNA (dsRNA) can be detected on HPMC and HAoEC 48h after exposure to SARS-CoV-2 at a multiplicity MOI of 1, in contrast to control cells, Vero E6. Nevertheless, we observed that both HPMC and HAoEC can be activated with the whole SARS-CoV-2 virus 48h after exposure and by its spike protein 24h after treatment, indicated by the increased expression of the cell adhesion molecule ICAM1 (Supplementary Fig. 2). As this result indicated that the whole virus particle is not needed to elicit a vascular response we decided to use SARS-CoV-2 spike protein to serve as a proxy for studying the effect of SARS-CoV-2 on the vascular endothelium.

To better replicate the physiological conditions of the vascular endothelium, we cultured and treated the EC under flow (shear stress = 10 dyn/cm²). We allowed the cells to grow under flow for 72h, after which they were exposed o 1 µg/ml of recombinant SARS-CoV-2 spike protein for 24h. As controls, we included treatment with 1 ng/ml TNF-α. To evaluate the more long-term effect, we stopped the treatment after 24h and further cultured the ECs to 96h post-treatment. A schematic overview of the experimental setup for HPMC and HAoEC under flow is depicted in Supplementary Fig. 3.

We saw that treatment of both HPMC and HAoEC with SARS-CoV-2 spike led to the expression of the cellular adhesion molecules ICAM1 and E-Selectin. We observed a significant induction of ICAM1 (p-value = 0.0001 for HMPC, < 0.0001 for HAoEC) and E-Selectin (p-value = 0.0069 for HPMC, 0.01 for HAoEC) expression at 24h and post-activation with SARS-CoV-2 spike compared to untreated controls (Figs. 1a, b, and d). At 96h post-treatment, we saw that a significant ICAM1 expression can still be detected on both HPMC and HAoEC (p-value = 0.03 and 0.0026, respectively, Figs. 1a, c). The expression of E-Selectin, however, was not detectable anymore at 96h post treatment (Figs. 1a, e). Treatment with TNF-α had a similar effect as SARS-CoV-2 spike on ICAM1 expression after 24h (Fig. 1b) and E-Selectin expression after 24h and 96h (Figs. 1d, e). In contrast, expression of ICAM1 after 96h was higher with TNF-α treatment on HAoEC (Fig. 1c). Taken together, these results suggest that the expression of immune cell adhesion molecules by SARS-CoV-2 activation, especially ICAM1, might continue even after the infection has been cleared.

SARS-CoV-2 spike activation triggers chemokine release

To see the profile of the chemokine releases by human ECs due to SARS-CoV-2 spike protein treatment, we measured the level of chemokines in the perfusion media 24h and 96h after treatment using a Luminex-type assay. For HPMC (Fig. 2a) and HAoEC (Fig. 2b), we observed chemokine changes due to treatment with SARS-CoV-2 spike protein or TNF-α at both timepoints. At 24h, most chemokines were upregulated to various degrees. For both HPMC and HAoEC, several chemokine levels in SARS-CoV-2 spike treated ECs were increased to a comparable level as with TNF-α (Figs. 2a, b). Some cell type- and treatment-specific changes were also observed. Notably, chemokines associated with microbial infection, such as CXCL1 and CXCL2, and proinflammatory chemokine IL-6 were expressed at a considerably higher level with SARS-CoV-2 spike than with TNF-α (Figs. 2a, b). An increased TNF-α release in SARS-CoV-2 treated ECs was observed at 24h. At 96h post-treatment we saw that the overall induction of chemokine expression was reduced. However, some chemokines (for instance CCL2, CCL13, and CXCL8 for both cells, CCL17 for HAoEC) were still detectable at a relatively high level.

SARS-CoV-2 spike triggers a procoagulant state of EC and the attachment of PBMCs to the surface of the EC

Next, we sought to evaluate whether the activation of the human endothelium by SARS-CoV-2 spike protein leads to an increased interaction of the ECs with innate immune cells. We performed leucocyte binding assays by perfusing the activated ECs with fluorescently labeled PBMCs and quantified the number of adhering PBMCs using time-series confocal microscopy imaging. Adhered PBMCs were defined as cells attached for two or more frames in a time-series recording that captured the image every 3s. Using this assay, we saw that treatment with SARS-CoV-2 spike protein induced significant binding of leucocytes to HPMC and HAoEC at 24h (p-value = 0.0035 and 0.023, respectively), and even more so with TNF-a treatment (Figs. 3a, b). At 96h post-treatment, however, we no longer observed PBMC binding on both types of ECs for either treatment (Figs. 3c, d).

Figure 3. SARS-CoV-2 spike activation leads to increased leucocyte adhesion to the surface of the endothelium. After 24h activation with 1 µg/ml SARS-CoV-2 spike protein or 1 ng/ml TNF-a, HPMC and HAoEC were perfused with fluorescently labeled PBMCs for 20 min. Representative images from the time-series recording of PBMC adhesion to HPMC and HAoEC 24h and 96h post-treatment (a). Adhered PBMCs (defined as PBMCs that adhered for more than 3s in the frame) to HPMC and HAoEC per 20 min at 24h (b) and 96h (c) post-treatment. Statistical analysis was done using one-way ANOVA with multiple comparisons.

We then analyzed the influence of SARS-CoV-2 spike activation on clot formation on the EC. After activating the cells with SARS-CoV-2 spike or TNF-α, we perfused the EC with recalcified human citrate plasma spiked with fluorescently labeled fibrinogen and used time-series microscopy imaging to record the clot formation time. As shown in Figs. 4a and b, we observed a significant decrease in clot formation time, indicative of a procoagulant state in HPMC and HAoEC, 24h post-treatment with SARS-CoV-2 spike protein, and also with TNF-α. At 96h post-treatment, the reduction of clot formation time was less pronounced and only statistically significant for HAoEC treated with TNF-α (Figs. 4a, c).

Transcriptional dynamics of SARS-CoV-2 spike activation of the human endothelium

Given the observed influence of SARS-CoV-2 spike protein on EC activation, we sought to gain further insight into the influence of the spike protein on EC dysfunction by transcriptional analysis. We extracted the total cellular mRNA from mock-, SARS-CoV-2 spike-, and TNF-α-treated HPMC and HAoEC 24h and 96h post-treatment. RNA sequencing was performed using the bulk RNA barcoding and sequencing (BRB-seq) pipeline [41]. 653 unique, differentially expressed genes (DEGs) were identified across samples by performing pairwise comparisons between each treatment and the mock counterpart for each cell type and time point (adjusted p-value < 0.05, absolute logFC > 0.25). For spike- and TNF-α-treated HAoEC 24 hours post-activation, 215 and 352 DEGs were identified, respectively. For HPMC, 320 and 342 DEGs were identified for treatment with spike and TNF-α after 24 h, respectively. DEGs were also identified at 96h, albeit significantly lower (spike-treated HAoEC = 20, TNF-α-treated HAoEC = 4, spike-treated HPMC = 33, TNF-α-treated HPMC = 17), confirming the prolonged effect on EC gene expression profiles. All the DEGs identified in all comparisons are listed in Supplementary table 1.

More DEGs were shared among all treatments at 24h post-treatment than at 96h (Fig. 5a, Supplementary Figs. 4a and b). For instance, 50 upregulated and 8 downregulated DEGs were shared among all conditions at 24h (Supplementary Figs. 4a, b). At 96h, only 1 DEG was shared among conditions. Unique DEGs for HPMC and HAoEC were also identified, highlighting distinct transcriptional responses of the two ECs stemming from different anatomical locations. Hierarchical clustering of all DEGs identified for HPMC and HAoEC also showed a distinct expression of several gene clusters in SARS-CoV-2 treated ECs in comparison to those with TNF-α treatment (Figs. 5b, c).

To examine the pathways involved in the vascular inflammatory effect of the SARS-CoV-2 spike protein, we performed a pathway-enrichment analysis on DEGs of HPMC and HAoEC at 24h and 96h time points, respectively (Fig. 5d-g, Supplementary Figs. 4c and 5). We saw that 24h after treatment with SARS-CoV-2 spike protein, upregulation of cytokine-mediated signaling is observed for HMPC and HAoEC, similarly to the treatment with TNF-α (Figs. 5d and f, Supplementary Fig. 5). Pathways associated with responses to viruses are more enriched for HPMC, while pathways associated with responses to other pathogens were seen in HAoEC. Pattern recognition receptor signaling pathway, toll-like receptor signaling, and NF-kB signaling, were also enriched. Furthermore, pathways associated with cellular development were downregulated for HPMC at 24h (Supplementary Figs. 4c). At 96h, upregulation of pathways associated with virus response were still observed in the spike-treated HPMC (Fig. 5e). In addition, antigen presentation and processing pathways were highly represented, indicating the activation of immunomodulating functions in the ECs. For HAoEC, pathways associated with endothelial apoptosis and responses to virus were among the most significantly upregulated while TGF-b regulation pathways were downregulated (Fig. 5g, Supplementary Fig. 4c). For TNF-α-treated HPMC and HAoEC, pathways associated with antigen processing were upregulated, while pathways associated with endothelial development were downregulated (Supplementary Fig. 5).

SARS-CoV-2 spike treated cells exhibit sustained inflammation and alteration of antigen presentation and coagulation state of the endothelium at a later stage

To have a closer look at the prolonged transcriptomic changes associated with the observed pathological conditions in SARS-CoV-2 spike treatments, we analyzed the DEGs of SARS-CoV-2 treated ECs at 96h with Kyoto Encyclopedia of Genes and Genomes (KEGG) Mapper software. Prolonged expression of individual markers associated with TNF-a signaling for HPMC (Fig. 6a, b) and HAoEC (Supplementary Fig. 6a, b) confirmed a sustained vascular inflammation. MHC class I (HLA-A, HLA-B, and HLA-C) and the transmembrane glycoprotein TAPBP, predominantly mediating the antigen processing and presentation pathway, were also enriched at 96h (Fig. 7a, Supplementary Fig. 7a). Moreover, the upregulation of Cathepsin S (CTSS) indicates the influence on the MHC class II pathway. The influence on the coagulation pathway was marked by the upregulation of serpin, factor VIII, PAI-1, tPA and uPA, as well as the downregulation of anticoagulant factors such as protein S and alpha-2-macroglobulin (A2M) (Fig. 7b, Supplementary Fig. 7b). Complement associated factors such complement factor B, C5, C1 receptor, and MAC were upregulated, whereas a downregulation of C4BP and C5R1 was observed. This suggests that the vascular procoagulant effect of SARS-CoV-2 on the endothelium, although to a lower degree, might last beyond the presence of virus particles.

In this study, we investigated the scope of the vascular inflammatory effect of SARS-CoV-2 spike protein on phenotypic, functional, and transcriptional levels in both lung microvascular and aortic ECs. To provide a more comprehensive insight into the profile of vascular inflammation, we included a comparison with the inflammatory cytokine TNF-α, which is known to induce prolonged activation of EC [43], [44]. Our study revealed that SARS-CoV-2 spike protein triggered prolonged cell adhesion marker expression and chemokine releases, along with increased immune cell binding and formation of a procoagulant state of the ECs. We observed a similar degree of vascular inflammation by SARS-CoV-2 spike protein to that with TNF-a. However, distinct gene activation profiles were found for the viral spike protein. We also showed that on the transcriptome level, SARS-CoV-2 spike resulted in sustained inflammation, changes in antigen presentation and coagulation state of the endothelium. The observed prolonged effects beyond the presence of the spike protein suggests possible long-term consequences of SARS-CoV-2 on the endothelium.

Involvement of ACE2 and vascular infection in mediating vascular inflammation have been described, although these findings are controversial [13], [22], [45], [46]. In the present study, we showed that the observed vascular inflammatory effect of SARS-CoV-2 is not caused by active viral replication and most likely not mediated by ACE2, due to lack of ACE2 expression in both types of ECs used. Therefore, although interaction of SARS-CoV-2 in the respiratory tract is ACE2-dependent, other spike-binding receptors might be involved in mediating the interaction of SARS-CoV-2 with vascular ECs. Several reports have described that SARS-CoV-2 spike engages glycosaminoglycans (GAGs) on the EC glycocalyx and further binds integrins [18], [29], [47], [48]. Others have described neuropilin-1 as a SARS-CoV-2 spike receptor that mediates the SARS-CoV-2 cellular entry [49]. Additionally, TLR recognition might also be important in mediating the vascular inflammatory effect of SARS-CoV-2 on the vasculature [50]. It remains to be elucidated which receptors are involved in activation of the vascular endothelium via SARS-CoV-2 spike protein.

In line with previous findings, we showed that spike-treated ECs express high levels of cellular adhesion markers. Increased ICAM1 expression on ECs mediates the recruitment and attachment of leucocytes and neutrophil extracellular trap (NET) formation, as well as a prothrombotic state of the endothelium. Indeed, we showed that treatment with SARS-CoV-2 spike directly triggered leukocyte adhesion and increased the procoagulant state of the ECs. Elevated ICAM1 in plasma, which could be released by the damaged endothelium, is also positively correlated with disease severity as has been observed in COVID-19 patients [51], [52]. Moreover, elevated ICAM1 and other EC adhesion molecules associated with disease severity have been described in chronic cardiovascular diseases including atherosclerosis and coronary heart disease [53], [54]. Our results also show that the expression of ICAM1 seems to persist beyond the presence of SARS-CoV-2 spike, suggesting a state of sustained inflammation of the ECs. Similarly, several studies have shown increased levels of ICAM1 in serum of patients recovered from COVID-19 [35], [55], [56]. The circulating ICAM1, which could originate from damaged endothelium, may contribute to prolonged inflammation even in recovered and no longer infectious COVID-19 patients, indicating the involvement of the endothelium in PASC.

Our results showed similar profiles of chemokine expression due to SARS-CoV-2 spike activation on human EC to those observed in COVID-19 patients [30], [31]. Several studies described that elevated IL-1β, IL-6, IL-8, IL-17 in plasma is associated with disease severity in COVID-19 patients [57], [58], [59], [60], [61]. Therefore, ECs may play a significant role in the production of various inflammatory cytokines and chemokines contributing to the cytokine storm and excessive inflammatory response, exacerbating the disease in severe COVID-19 patients [58], [62]. Expression of IL-1β, CXCL1, CXCL8, and CCL20 could contribute to neutrophil recruitment to the surface of the endothelium, leading to NET-formation and immunothrombosis [60], [61], [63], [64], [65]. Chemokines such as CCL8, CXCL2, and CXCL10 can lead to recruitment of monocytes and macrophages to the activated ECs [66], [67], [68]. The increased expression of CCL2 also contributes to the amplification of monocyte and macrophage activation [69]. Recruitment of immune cells to the surface of spike-activated EC can lead to infiltration of inflammatory cells and further damage to the surrounding tissue, which can happen independently of an active infection and in different anatomical regions. Different chemokine expression levels and dynamics over time between HPMC and HAoEC suggest a possible EC origin-specific response. Our RNAseq analysis further highlights the distinct transcriptomic signatures, and the pathways associated with SARS-CoV-2 spike activation of HPMC and HAoEC. It is therefore necessary to characterize organ-specific vascular responses from organs that are also affected by COVID-19, such as brain and kidneys.

Our RNAseq data further highlight that SARS-CoV-2 spike alone can trigger an array of pathogen-associated responses, induction of robust proinflammatory states, alteration of EC development, and apoptosis, likely associated with the observed thrombo-inflammatory symptoms in COVID-19 patients [70], [71]. Moreover, prolonged expression of genes associated with proinflammatory pathways and apoptosis could induce persistent endothelial dysfunction and damage. The observed prolonged increase of adhesion molecules and antigen presentation could lengthen the recruitment of immune cells and mediate EC interaction with CD8 + and CD4 + T lymphocytes [72], [73]. In addition, the transcriptomic analysis also showed a prolonged disruption of the regulation of the complement- and coagulation cascades, reflecting a possible sustained prothrombotic state and increased cardiovascular complications after COVID-19 infection. It is worth noting that we did not see significant changes in the leucocyte binding and clotting time at a later time point in vitro, which could be due to the limitation in the assay sensitivity in our model. It is, therefore, essential to validate the long-term changes due to SARS-CoV-2 spike activation in EC in a more extensive study, for instance, in animal models or clinical studies involving convalescent COVID-19 patients. In addition, future studies should consider evaluating the consequences of EC activation by SARS-CoV-2 beyond the indicated time points, as well as the vascular inflammatory effects of other SARS-COV-2 spike variants.

In summary, our results provided a detailed and comprehensive characterization of the vascular inflammatory effects of SARS-CoV-2. We showed that the endothelium plays an essential role in determining the outcome of COVID-19 infection, such as vascular inflammation and systemic organ damage during and possibly beyond the acute infection phase. Therapeutic strategies should also consider the extent of SARS-CoV-2 inflammatory effects on the vascular endothelium. Treatments directed to EC protection and prevention of endothelial damage might be essential in the prevention and management of the post-sequelae effect of COVID-19.

Funding

This study was supported by Swiss National Science Foundation grants (CoVasc grant agreement no 198297). L.L. is supported by the Independent Research Fund Denmark Sapere Aude Starting grant (8048-00072A) and the Novo Nordisk Foundation (NNF21OC0071718). Y.L. is supported by the Novo Nordisk Foundation (NNF21OC0072031, NNF21OC0068988) CRISPRnet and Lundbeck Foundation (R396-2022-350).

Competing interests

Authors declare no competing interests.

Author contributions

MG: Conceptualization, methodology, investigation, writing original draft, reviewing and editing the manuscript.

CBB, AM, AD, JS: Methodology, investigation.

LL, YL: Investigation, methodology, supervision.

YD: Conceptualization, supervision, critical review of the manuscript.

RR: Conceptualization, methodology, supervision, reviewing and editing the manuscript.

Data availability statement

Data generated during this study are included in the article. RNAseq data and several basic visualization tools are available online on the Aarhus University repository (https://dreamapp.biomed.au.dk/SARS_EC/).

Acknowledgements

We thank Ronald Dijkman (University of Bern) for providing the virus isolates and cell lines. We thank Melle Holwerda, Gabriele Chiffi, Chiara Stüdle, and Berenice Martinez-Salazar, as well as the Biosafety team of the Institute for Infectious Diseases, University of Bern (Katharina Summermatter, Julia Feldmann, and Monika Gsell-Albert), for the support with the BSL-3 work. We thank the Microscopy Core Facility (MIC) of the University of Bern and Alithea Genomics, Lausanne, for providing essential technology. We also thank the Interregional Blood Donation Center of the Red Cross, Bern, for assistance in the procurement of the human blood samples.

L.-A. Teuwen, V. Geldhof, A. Pasut, and P. Carmeliet, “COVID-19: the vasculature unleashed,” Nat. Rev. Immunol., vol. 20, no. 7, pp. 389–391, Jul. 2020, doi: 10.1038/s41577-020-0343-0.
B. S. Kadosh et al., “COVID-19 and the Heart and Vasculature: Novel Approaches to Reduce Virus-Induced Inflammation in Patients With Cardiovascular Disease,” Arterioscler. Thromb. Vasc. Biol., vol. 40, no. 9, pp. 2045–2053, Sep. 2020, doi: 10.1161/ATVBAHA.120.314513.
M. K. Chung et al., “COVID-19 and Cardiovascular Disease: From Bench to Bedside,” Circ. Res., vol. 128, no. 8, pp. 1214–1236, Apr. 2021, doi: 10.1161/CIRCRESAHA.121.317997.
M. Ackermann et al., “Pulmonary Vascular Endothelialitis, Thrombosis, and Angiogenesis in Covid-19,” N. Engl. J. Med., vol. 383, no. 2, pp. 120–128, Jul. 2020, doi: 10.1056/NEJMoa2015432.
B. T. Bradley et al., “Histopathology and ultrastructural findings of fatal COVID-19 infections in Washington State: a case series,” The Lancet, vol. 396, no. 10247, pp. 320–332, Aug. 2020, doi: 10.1016/S0140-6736(20)31305-2.
L. Perico, A. Benigni, and G. Remuzzi, “SARS-CoV-2 and the spike protein in endotheliopathy,” Trends Microbiol., vol. 32, no. 1, pp. 53–67, Jan. 2024, doi: 10.1016/j.tim.2023.06.004.
J. Wenzel and M. Schwaninger, “How COVID-19 affects microvessels in the brain,” Brain, vol. 145, no. 7, pp. 2242–2244, Jul. 2022, doi: 10.1093/brain/awac211.
P. Hansrivijit et al., “Incidence of Acute Kidney Injury and Its Association with Mortality in Patients with Covid-19: A Meta-Analysis,” J. Investig. Med., vol. 68, no. 7, pp. 1261–1270, Oct. 2020, doi: 10.1136/jim-2020-001407.
Z. Varga et al., “Endothelial cell infection and endotheliitis in COVID-19,” The Lancet, vol. 395, no. 10234, pp. 1417–1418, May 2020, doi: 10.1016/S0140-6736(20)30937-5.
T. P. Buzhdygan et al., “The SARS-CoV-2 spike protein alters barrier function in 2D static and 3D microfluidic in-vitro models of the human blood–brain barrier,” Neurobiol. Dis., vol. 146, p. 105131, Dec. 2020, doi: 10.1016/j.nbd.2020.105131.
W. Guan, W. Liang, J. He, and N. Zhong, “Cardiovascular comorbidity and its impact on patients with COVID-19,” Eur. Respir. J., vol. 55, no. 6, p. 2001227, Jun. 2020, doi: 10.1183/13993003.01227-2020.
C. J. M. Vrints, K. A. Krychtiuk, E. M. Van Craenenbroeck, V. F. Segers, S. Price, and H. Heidbuchel, “Endothelialitis plays a central role in the pathophysiology of severe COVID-19 and its cardiovascular complications,” Acta Cardiol., vol. 76, no. 2, pp. 109–124, Mar. 2021, doi: 10.1080/00015385.2020.1846921.
L. Schimmel et al., “Endothelial cells are not productively infected by SARS-CoV‐2,” Clin. Transl. Immunol., vol. 10, no. 10, p. e1350, Jan. 2021, doi: 10.1002/cti2.1350.
R.-C. Yang et al., “SARS-CoV-2 productively infects human brain microvascular endothelial cells,” J. Neuroinflammation, vol. 19, no. 1, p. 149, Dec. 2022, doi: 10.1186/s12974-022-02514-x.
T. Klouda et al., “Interferon-alpha or -beta facilitates SARS-CoV-2 pulmonary vascular infection by inducing ACE2,” Angiogenesis, vol. 25, no. 2, pp. 225–240, May 2022, doi: 10.1007/s10456-021-09823-4.
I. R. McCracken et al., “Lack of Evidence of Angiotensin-Converting Enzyme 2 Expression and Replicative Infection by SARS-CoV-2 in Human Endothelial Cells,” Circulation, vol. 143, no. 8, pp. 865–868, Feb. 2021, doi: 10.1161/CIRCULATIONAHA.120.052824.
A. V. Letarov, V. V. Babenko, and E. E. Kulikov, “Free SARS-CoV-2 Spike Protein S1 Particles May Play a Role in the Pathogenesis of COVID-19 Infection,” Biochem. Mosc., vol. 86, no. 3, pp. 257–261, Mar. 2021, doi: 10.1134/S0006297921030032.
S. B. Biering et al., “SARS-CoV-2 Spike triggers barrier dysfunction and vascular leak via integrins and TGF-β signaling,” Nat. Commun., vol. 13, no. 1, p. 7630, Dec. 2022, doi: 10.1038/s41467-022-34910-5.
G. J. Nuovo et al., “Endothelial cell damage is the central part of COVID-19 and a mouse model induced by injection of the S1 subunit of the spike protein,” Ann. Diagn. Pathol., vol. 51, p. 151682, Apr. 2021, doi: 10.1016/j.anndiagpath.2020.151682.
R. M. L. Colunga Biancatelli, P. A. Solopov, E. R. Sharlow, J. S. Lazo, P. E. Marik, and J. D. Catravas, “The SARS-CoV-2 spike protein subunit S1 induces COVID-19-like acute lung injury in Κ18-hACE2 transgenic mice and barrier dysfunction in human endothelial cells,” Am. J. Physiol.-Lung Cell. Mol. Physiol., vol. 321, no. 2, pp. L477–L484, Aug. 2021, doi: 10.1152/ajplung.00223.2021.
Z. Qin et al., “Endothelial cell infection and dysfunction, immune activation in severe COVID-19,” Theranostics, vol. 11, no. 16, pp. 8076–8091, 2021, doi: 10.7150/thno.61810.
Y. Lei et al., “SARS-CoV-2 Spike Protein Impairs Endothelial Function via Downregulation of ACE 2,” Circ. Res., vol. 128, no. 9, pp. 1323–1326, Apr. 2021, doi: 10.1161/CIRCRESAHA.121.318902.
J. Fajnzylber et al., “SARS-CoV-2 viral load is associated with increased disease severity and mortality,” Nat. Commun., vol. 11, no. 1, p. 5493, Oct. 2020, doi: 10.1038/s41467-020-19057-5.
M. De Michele et al., “Evidence of SARS-CoV-2 spike protein on retrieved thrombi from COVID-19 patients,” J. Hematol. Oncol.J Hematol Oncol, vol. 15, no. 1, p. 108, Aug. 2022, doi: 10.1186/s13045-022-01329-w.
M. Brady et al., “Spike protein multiorgan tropism suppressed by antibodies targeting SARS-CoV-2,” Commun. Biol., vol. 4, no. 1, p. 1318, Nov. 2021, doi: 10.1038/s42003-021-02856-x.
G. Y. Oudit, K. Wang, A. Viveiros, M. J. Kellner, and J. M. Penninger, “Angiotensin-converting enzyme 2—at the heart of the COVID-19 pandemic,” Cell, vol. 186, no. 5, pp. 906–922, Mar. 2023, doi: 10.1016/j.cell.2023.01.039.
P. Simons et al., “Integrin activation is an essential component of SARS-CoV-2 infection,” Sci. Rep., vol. 11, no. 1, p. 20398, Oct. 2021, doi: 10.1038/s41598-021-99893-7.
D. Nader and S. W. Kerrigan, “Molecular Cross-Talk between Integrins and Cadherins Leads to a Loss of Vascular Barrier Integrity during SARS-CoV-2 Infection,” Viruses, vol. 14, no. 5, p. 891, Apr. 2022, doi: 10.3390/v14050891.
J. P. Robles, M. Zamora, E. Adan-Castro, L. Siqueiros-Marquez, G. Martinez de la Escalera, and C. Clapp, “The spike protein of SARS-CoV-2 induces endothelial inflammation through integrin α5β1 and NF-κB signaling,” J. Biol. Chem., vol. 298, no. 3, p. 101695, Mar. 2022, doi: 10.1016/j.jbc.2022.101695.
B. A. Khalil, N. M. Elemam, and A. A. Maghazachi, “Chemokines and chemokine receptors during COVID-19 infection,” Comput. Struct. Biotechnol. J., vol. 19, pp. 976–988, 2021, doi: 10.1016/j.csbj.2021.01.034.
S. Hue et al., “Uncontrolled Innate and Impaired Adaptive Immune Responses in Patients with COVID-19 Acute Respiratory Distress Syndrome,” Am. J. Respir. Crit. Care Med., vol. 202, no. 11, pp. 1509–1519, Dec. 2020, doi: 10.1164/rccm.202005-1885OC.
S. Mehandru and M. Merad, “Pathological sequelae of long-haul COVID,” Nat. Immunol., vol. 23, no. 2, pp. 194–202, Feb. 2022, doi: 10.1038/s41590-021-01104-y.
M. Merad, C. A. Blish, F. Sallusto, and A. Iwasaki, “The immunology and immunopathology of COVID-19,” Science, vol. 375, no. 6585, pp. 1122–1127, Mar. 2022, doi: 10.1126/science.abm8108.
A. Nalbandian et al., “Post-acute COVID-19 syndrome,” Nat. Med., vol. 27, no. 4, pp. 601–615, Apr. 2021, doi: 10.1038/s41591-021-01283-z.
T. A. Smith-Norowitz, J. Loeffler, Y. M. Norowitz, and S. Kohlhoff, “Intracellular Adhesion Molecule-1 (ICAM-1) Levels in Convalescent COVID-19 Serum: A Case Report,” Ann. Clin. Lab. Sci., vol. 51, no. 5, pp. 730–734, Sep. 2021.
M. Haffke et al., “Endothelial dysfunction and altered endothelial biomarkers in patients with post-COVID-19 syndrome and chronic fatigue syndrome (ME/CFS),” J. Transl. Med., vol. 20, no. 1, p. 138, Dec. 2022, doi: 10.1186/s12967-022-03346-2.
S. Charfeddine et al., “Long COVID 19 Syndrome: Is It Related to Microcirculation and Endothelial Dysfunction? Insights From TUN-EndCOV Study,” Front. Cardiovasc. Med., vol. 8, p. 745758, Nov. 2021, doi: 10.3389/fcvm.2021.745758.
J. C. Hierholzer and R. A. Killington, “Virus isolation and quantitation,” in Virology Methods Manual, Elsevier, 1996, pp. 25–46. doi: 10.1016/B978-012465330-6/50003-8.
J. Schindelin et al., “Fiji: an open-source platform for biological-image analysis,” Nat. Methods, vol. 9, no. 7, pp. 676–682, Jun. 2012, doi: 10.1038/nmeth.2019.
J. Mutterer and E. Zinck, “Quick-and-clean article figures with FigureJ,” J. Microsc., vol. 252, no. 1, pp. 89–91, Oct. 2013, doi: 10.1111/jmi.12069.
D. Alpern et al., “BRB-seq: ultra-affordable high-throughput transcriptomics enabled by bulk RNA barcoding and sequencing,” Genome Biol., vol. 20, no. 1, p. 71, Apr. 2019, doi: 10.1186/s13059-019-1671-x.
A. Milusev, A. Despont, J. Shaw, R. Rieben, and N. Sorvillo, “Inflammatory stimuli induce shedding of heparan sulfate from arterial but not venous porcine endothelial cells leading to differential proinflammatory and procoagulant responses,” Sci. Rep., vol. 13, no. 1, p. 4483, Mar. 2023, doi: 10.1038/s41598-023-31396-z.
P. Baluk et al., “TNF-α drives remodeling of blood vessels and lymphatics in sustained airway inflammation in mice,” J. Clin. Invest., p. JCI37626, Sep. 2009, doi: 10.1172/JCI37626.
J. S. Pober, “Endothelial activation: intracellular signaling pathways,” Arthritis Res., vol. 4, no. Suppl 3, p. S109, 2002, doi: 10.1186/ar576.
A. C. Montezano et al., “SARS-CoV-2 spike protein induces endothelial inflammation via ACE2 independently of viral replication,” Sci. Rep., vol. 13, no. 1, p. 14086, Aug. 2023, doi: 10.1038/s41598-023-41115-3.
J. Nascimento Conde, W. R. Schutt, E. E. Gorbunova, and E. R. Mackow, “Recombinant ACE2 Expression Is Required for SARS-CoV-2 To Infect Primary Human Endothelial Cells and Induce Inflammatory and Procoagulative Responses,” mBio, vol. 11, no. 6, pp. e03185-20, Dec. 2020, doi: 10.1128/mBio.03185-20.
E. G. Norris, X. S. Pan, and D. C. Hocking, “Receptor-binding domain of SARS-CoV-2 is a functional αv-integrin agonist,” J. Biol. Chem., vol. 299, no. 3, p. 102922, Mar. 2023, doi: 10.1016/j.jbc.2023.102922.
B. Martínez-Salazar et al., “COVID-19 and the Vasculature: Current Aspects and Long-Term Consequences,” Front. Cell Dev. Biol., vol. 10, p. 824851, Feb. 2022, doi: 10.3389/fcell.2022.824851.
J. L. Daly et al., “Neuropilin-1 is a host factor for SARS-CoV-2 infection,” Science, vol. 370, no. 6518, pp. 861–865, Nov. 2020, doi: 10.1126/science.abd3072.
S. Khan, M. S. Shafiei, C. Longoria, J. W. Schoggins, R. C. Savani, and H. Zaki, “SARS-CoV-2 spike protein induces inflammation via TLR2-dependent activation of the NF-κB pathway,” eLife, vol. 10, p. e68563, Dec. 2021, doi: 10.7554/eLife.68563.
M. Tong et al., “Elevated Expression of Serum Endothelial Cell Adhesion Molecules in COVID-19 Patients,” J. Infect. Dis., vol. 222, no. 6, pp. 894–898, Aug. 2020, doi: 10.1093/infdis/jiaa349.
S. Kaur et al., “Elevated plasma ICAM1 levels predict 28-day mortality in cirrhotic patients with COVID-19 or bacterial sepsis,” JHEP Rep., vol. 3, no. 4, p. 100303, Aug. 2021, doi: 10.1016/j.jhepr.2021.100303.
A. D. Blann and C. N. McCollum, “Circulating Endothelial Cell/Leukocyte Adhesion Molecules in Atherosclerosis,” Thromb. Haemost., vol. 72, no. 01, pp. 151–154, 1994, doi: 10.1055/s-0038-1648827.
S.-J. Hwang et al., “Circulating Adhesion Molecules VCAM-1, ICAM-1, and E-selectin in Carotid Atherosclerosis and Incident Coronary Heart Disease Cases: The Atherosclerosis Risk In Communities (ARIC) Study,” Circulation, vol. 96, no. 12, pp. 4219–4225, Dec. 1997, doi: 10.1161/01.CIR.96.12.4219.
F. W. Chioh et al., “Convalescent COVID-19 patients are susceptible to endothelial dysfunction due to persistent immune activation,” eLife, vol. 10, p. e64909, Mar. 2021, doi: 10.7554/eLife.64909.
B. E. Fan et al., “Hypercoagulability, endotheliopathy, and inflammation approximating 1 year after recovery: Assessing the long-term outcomes in COVID ‐19 patients,” Am. J. Hematol., vol. 97, no. 7, pp. 915–923, Jul. 2022, doi: 10.1002/ajh.26575.
D. Blanco-Melo et al., “Imbalanced Host Response to SARS-CoV-2 Drives Development of COVID-19,” Cell, vol. 181, no. 5, pp. 1036–1045.e9, May 2020, doi: 10.1016/j.cell.2020.04.026.
C. Huang et al., “Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China,” The Lancet, vol. 395, no. 10223, pp. 497–506, Feb. 2020, doi: 10.1016/S0140-6736(20)30183-5.
D. M. Del Valle et al., “An inflammatory cytokine signature predicts COVID-19 severity and survival,” Nat. Med., vol. 26, no. 10, pp. 1636–1643, Oct. 2020, doi: 10.1038/s41591-020-1051-9.
S. Li et al., “Clinical and pathological investigation of patients with severe COVID-19,” JCI Insight, vol. 5, no. 12, p. e138070, Jun. 2020, doi: 10.1172/jci.insight.138070.
D. Wang et al., “Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus–Infected Pneumonia in Wuhan, China,” JAMA, vol. 323, no. 11, p. 1061, Mar. 2020, doi: 10.1001/jama.2020.1585.
F. Coperchini, L. Chiovato, L. Croce, F. Magri, and M. Rotondi, “The cytokine storm in COVID-19: An overview of the involvement of the chemokine/chemokine-receptor system,” Cytokine Growth Factor Rev., vol. 53, pp. 25–32, Jun. 2020, doi: 10.1016/j.cytogfr.2020.05.003.
F. P. Veras et al., “SARS-CoV-2–triggered neutrophil extracellular traps mediate COVID-19 pathology,” J. Exp. Med., vol. 217, no. 12, p. e20201129, Dec. 2020, doi: 10.1084/jem.20201129.
E. A. Middleton et al., “Neutrophil extracellular traps contribute to immunothrombosis in COVID-19 acute respiratory distress syndrome,” Blood, vol. 136, no. 10, pp. 1169–1179, Sep. 2020, doi: 10.1182/blood.2020007008.
K. M. Henriquez, M. S. Hayney, Y. Xie, Z. Zhang, and B. Barrett, “Association of interleukin-8 and neutrophils with nasal symptom severity during acute respiratory infection,” J. Med. Virol., vol. 87, no. 2, pp. 330–337, Feb. 2015, doi: 10.1002/jmv.24042.
M. Liao et al., “Single-cell landscape of bronchoalveolar immune cells in patients with COVID-19,” Nat. Med., vol. 26, no. 6, pp. 842–844, Jun. 2020, doi: 10.1038/s41591-020-0901-9.
X. Ren et al., “COVID-19 immune features revealed by a large-scale single-cell transcriptome atlas,” Cell, vol. 184, no. 7, pp. 1895–1913.e19, Apr. 2021, doi: 10.1016/j.cell.2021.01.053.
L. He, Q. Zhang, Y. Zhang, Y. Fan, F. Yuan, and S. Li, “Single-cell analysis reveals cell communication triggered by macrophages associated with the reduction and exhaustion of CD8 + T cells in COVID-19,” Cell Commun. Signal., vol. 19, no. 1, p. 73, Dec. 2021, doi: 10.1186/s12964-021-00754-7.
R. Channappanavar et al., “Dysregulated Type I Interferon and Inflammatory Monocyte-Macrophage Responses Cause Lethal Pneumonia in SARS-CoV-Infected Mice,” Cell Host Microbe, vol. 19, no. 2, pp. 181–193, Feb. 2016, doi: 10.1016/j.chom.2016.01.007.
R. Bhayana et al., “Abdominal Imaging Findings in COVID-19: Preliminary Observations,” Radiology, vol. 297, no. 1, pp. E207–E215, Oct. 2020, doi: 10.1148/radiol.2020201908.
N. Lushina, J. S. Kuo, and H. A. Shaikh, “Pulmonary, Cerebral, and Renal Thromboembolic Disease in a Patient with COVID-19,” Radiology, vol. 296, no. 3, pp. E181–E183, Sep. 2020, doi: 10.1148/radiol.2020201623.
W. Huang et al., “Lymphocyte Subset Counts in COVID -19 Patients: A Meta‐Analysis,” Cytometry A, vol. 97, no. 8, pp. 772–776, Aug. 2020, doi: 10.1002/cyto.a.24172.
Y. Peng et al., “Broad and strong memory CD4 + and CD8 + T cells induced by SARS-CoV-2 in UK convalescent individuals following COVID-19,” Nat. Immunol., vol. 21, no. 11, pp. 1336–1345, Nov. 2020, doi: 10.1038/s41590-020-0782-6.

No competing interests reported.

Supplementaryfigure1.tif
Supplementary figure 1. No productive infection of SARS-CoV-2 in human endothelium. Human pulmonary microvascular endothelial cells (HPMC) and human aortic endothelial cells (HAoEC) were grown under static condition on a chamber slide and infected with purified SARS-CoV-2 virus stock at a MOI of 1 for 48h. As a positive control, Vero-E6 was included. Fixed cells were stained for SARS-CoV-2 Nucleoprotein (red), double-stranded RNA (green), and VE-cadherin (yellow) for endothelial cell markers. Scale bar = 50 µm.
Supplementaryfigure2.tif
Supplementary figure 2. SARS-CoV-2 and SARS-CoV-2 spike protein can activate human endothelium. SARS-CoV-2 infected (upper panel) and SARS-CoV-2 spike treated (lower panel) HPMC and HAoEC were stained with ICAM1 (green) to visualize the expression of adhesion molecules, indicating the activated state of the endothelium. Scale bar = 50 µm.
Supplementaryfigure3.tif
Supplementary figure 3. Schematic overview of the experimental setup.
Supplementaryfigure4.tif
Supplementary figure 4. Upregulated and downregulated differentially expressed genes (DEGs) of activated endothelial cells. Upset plot depicting the shared upregulated (a) and downregulated (b) DEGs across samples. Bar plots of the most enriched biological processes by downregulated DEGs in the SARS-CoV-2 spike treated ECs (c).
Supplementaryfigure5.tif
Supplementary figure 5. Transcriptional changes in TNF-atreated endothelial cells. Barplots of top enriched biological process by upregulated (a) and downregulated (b) DEGs in the TNF-atreated ECs.
Supplementaryfigure6.tif
Supplementary figure 6. A segment of the KEGG pathway showing the enrichment of the TNF signaling pathways by DE genes associated with SARS-CoV-2 spike-activated HAoEC at 24h (a) and 96h (b).
Supplementaryfigure7.tif
Supplementary figure 7. Enrichment of antigen processing and presentation (a) as well as the complement and coagulation pathway (b) in the spike-treated HAoEC at 96h.
Supplementarytable1.xlsx
Supplementary table 1. List of DEGs identified in HPMC and HAoEC.

Download PDF

Editorial decision: Revision requested
11 Oct, 2024
Reviews received at journal
23 Sep, 2024
Reviews received at journal
18 Sep, 2024
Reviewers agreed at journal
09 Sep, 2024
Reviewers agreed at journal
04 Sep, 2024
Reviewers agreed at journal
04 Sep, 2024
Reviewers agreed at journal
04 Sep, 2024
Reviewers invited by journal
04 Sep, 2024
Editor assigned by journal
03 Sep, 2024
Submission checks completed at journal
03 Sep, 2024
First submitted to journal
30 Aug, 2024

You are reading this latest preprint version

Sustained vascular inflammatory effects of SARS-CoV-2 spike protein on human endothelial cells

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Primary endothelial cells and cell lines

SARS-CoV-2 viral stock

Isolation of human peripheral blood mononuclear cells (PBMCs)

Infection of HPMCs and HAoEC with SARS-CoV-2 on static conditions

Culturing and activation of endothelial cells in a microfluidic system

Immunofluorescence analysis

Chemokines analysis

Bulk-RNA sequencing and data analysis

Microfluidic PBMC adhesion assays

Microfluidic coagulation assays

Statistical analysis

Results

SARS-CoV-2 spike protein triggered prolonged expression of ICAM1 in aortic and lung microvascular endothelial cells

SARS-CoV-2 spike activation triggers chemokine release

Transcriptional dynamics of SARS-CoV-2 spike activation of the human endothelium

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1