SARS-CoV-2 Spike Protein Causes Endothelial Vascular Dysfunction


The research findings of a new vitro study by researchers from the University of Texas-Austin-USA using three-dimensional engineered vascular networks has validated the previous claims that SARS-CoV-2 spike protein causes endothelial vascular dysfunction.

 The study findings were published on a preprint server and are currently being peer reviewed.

The ongoing COVID-19 pandemic, caused by the novel coronavirus SARS-CoV-2, has as of the time of writing infected 605 million people, 6.5 million of whom have died1. Although initial focus was placed on the virus’ effects on the lungs, it is now known to affect the endothelium as well, especially in severe infection 2.

Severe COVID-19 is associated with varied vascular pathologies, including microthrombi, complement pathway activation, and endothelial cell sloughing from the vessel wall. This is known as endothelial dysfunction, and presence of endothelial dysfunction is associated with a 5-10-fold increase in mortality rate 3.

Additionally, these mechanisms have led to profound vascular and cardiac morbidity secondary to COVID-19 4,5. COVID-19 mediated endothelial dysfunction occurs as a result of SARS-CoV-2’s binding to the ACE2, a surface protein involved in blood pressure regulation, resulting in downstream signaling that promotes inflammation 2.

This is in contrast to other coronaviruses, since although they are able to infect endothelial cells and promote an inflammatory environment, extensive vascular damage is unique to SARS-CoV-26. To better understand the interactions between SARS-CoV-2 and endothelial cells and to evaluate the effectiveness of proposed treatments, there is a need to create in vitro tissue models of SARS-CoV-2-induced endothelial dysfunction.

Other research groups have created in vitro tissue models for SARS-CoV-2’s effects on the endothelium 3,7. Following exposure to the virus, researchers have observed viral infection of endothelial cells, decreases in expression of cell adhesion molecules leading to disruption of the endothelial barrier, and increased secretion of inflammatory cytokines.

However, existing models for the effects of SARS-CoV-2 on the endothelium were developed using cells cultured in 2D monolayers, which do not model conditions in vivo as accurately compared to 3D models; culturing cells in 2D rather than 3D results in significant changes in proliferation rate, cell morphology, gene expression, and drug responsiveness 8,9.

In addition, endothelial cells cultured in 2D cannot form vascular networks, which is a key feature for identifying healthy and functional cells; this is especially important for COVID-19 modelling, where vascular network disruption is a hallmark of viral infection6.

Previously, our group has utilized human induced pluripotent stem cells (hiPSCs) differentiated to endothelial progenitors (EPs) and encapsulated in collagen hydrogels to study angiogenesis 10-12. Here we utilize this model to demonstrate that hiPSC-EP laden hydrogels treated with SARS-CoV-2 spike protein (CSP) results in significant vascular disruption, both in terms of the number of cells part of the vascular networks as well as vascular connectivity.

We also measured increases in inflammatory cytokine release consistent with that observed in clinical settings as well as a reduction in CSP-induced vascular dysfunction following treatment with the anti-inflammatory drug dexamethasone.

In conclusion, we are able to reproduce COVID-19-induced endothelial dysfunction seen in clinical settings using hiPSC-EP-laden hydrogels treated with CSP.

The results of this research demonstrate that CD34+-hiPSC-EP-laden hydrogels treated with SARS-CoV-2 spike protein can act as a 3D model for COVID-19’s effect on the endothelium. While the effects of COVID-19 on the endothelium have been previously reported, no published research utilizes endothelial cells cultured in 3D or studied the impact of CSP on vasculature.

Compared with 2D culture, 3D models produce cells with significant differences in morphology, proliferation, and drug responsiveness 8,9. Specifically, for COVID-19 modeling, this study shows that CSP can disrupt vascular networks, which is both commonly seen in COVID-19 patients and impossible to replicate in 2D culture.

Most importantly, our model enables study of how CSP affects vasculature formation in a 3D environment. As we previously described, after encapsulation, CD34+-hiPSC-EPs extend and then form capillary-like structures over a 7-day period10-12; CSP treatment results in the loss of these structures, however the extent of these effects depends on the timing of CSP treatment administration during the process of vasculature formation.

Although CSP addition at either day 1 or day 5 resulted in a decrease in network connectivity (percent largest network), CSP addition at a later time point (day 5), once the capillary plexus was formed, results in a significant decrease in the number of vessel-forming cells (branch points and volume fraction).

The differences in the number of vessel-forming cells may be due to cell proliferation following CSP removal. The day 1 condition is cultured for 6 additional days after CSP treatment; this allows the CD34+-hiPSC-Eps to proliferate to a greater extent than the cells in the day 5 condition, which are only cultured for 2 additional days.

Other researchers have shown that SARS-CoV-2 creates an inflammatory environment both through its interactions with ACE2 and through toll-like receptor signaling 14. Because TLRs bind a variety of biological molecules other than viral proteins, in order to determine if CSP-induced endothelial dysfunction is due to ACE2 signaling and therefore specific to CSP, we treated CD34+-hiPSC-EPs-laden hydrogels with the spike protein from Middle Eastern Respiratory Syndrome (MSP).

Although both CSP and MSP associate with TLRs, we did not observe any significant changes in vascular network formation following MSP treatment. This indicates that CSP-induced endothelial dysfunction is due to ACE2 signaling.

In addition to formation of vasculature, CD34+-hiPSC-EPs also mature over the 7 days in culture, which is characterized by decreases in CD34 expression and increases in CD31 and CDH5. This was true for our CD34+-hiPSC-EPs at all time points tested. We also measured expression of ACE2, the receptor onto which CSP binds for viral entry. We expected ACE2 RNA to increase with time, as that would provide one possible explanation for the observed increases in CSP toxicity at later time points.

However, ACE2 expression decreased with time relative to day 0 CD34+-hiPSC-EPs. In addition, we measured changes in gene expression 24 hours after CSP treatment. ACE2 downregulation requires viral entry, which is primarily mediated by the S2 subunit30. Because our experiments only use the S1 subunit, not surprisingly we did not observe any change in ACE2 following CSP treatment. we would

An alternative explanation is that CSP-induced endothelial dysfunction occurs through an ACE2-independent pathway. Previous research has shown that adult endothelial cells have low expression of ACE2 and that CSP-induced endothelial dysfunction can occur through TLR4 signaling 16.

In addition, CD34+ endothelial progenitors have also been shown to express ACE231, so it is possible that the decrease in ACE2 in the CD34+-hiPSC-EPs during the 7-day culture period is a normal part of the cells’ maturation.

In addition to endothelial dysfunction, COVID-19 also results in the release of inflammatory cytokines that results in further endothelial cell death; this is known as a cytokine storm. Some of the major cytokines that play a role in the cytokine storm include IL-1α, IL-1ß, IL-6, and TNFα24-26.

Although endothelial cells do secrete many cytokines, this is often in response to paracrine effects of immune cells. For example, IL-6 and TNFα secretion by endothelial cells is often associated with IL-1 secretion by immune cells 32,33.

Although this model does not contain immune cells, we observed changes in cytokine secretion from endothelial cells following CSP treatment. Specifically, we found that treating CD34+-hiPSC-EPs encapsulated in collagen hydrogels with CSP results in an increase in secretion of IL-8 and CXCL1.

Both proteins are secreted by endothelial cells 22,23 and are present in elevated levels in COVID-1924-26. In fact, they both play a similar role in inducing initial migration of neutrophils 34,35. This suggests that this system can be used to model the early stages of COVID-19 and through coculture with immune cells could be extended to later stages as well.

The corticosteroid dexamethasone is commonly used in clinical settings for treatment of moderate and severe COVID-19 because it reduces the impact of the cytokine storm 27. Dexamethasone reduces cytokine release by binding to the glucocorticoid receptor, which inhibits transcription of multiple inflammatory cytokines elevated in severe COVID-19, such as IL-1, IL-2, IL-6, IL-8, and TNFα27.

Because of this, dexamethasone has been widely used in clinical settings for treating moderate and severe COVID-19. We delivered dexamethasone at concentrations ranging from 2 ng/mL to 8 ng/mL to CD34+-hiPSC-EPs encapsulated in collagen hydrogels 5 days after encapsulation. In moderate to severe COVID-19, 6-8 mg of dexamethasone is given daily, and the concentrations tested correspond to 25%-100% of the peak serum concentration following an 8 mg dose.

We found that a 16.6 ng/mL dose of dexamethasone was the only dose that did not significantly affect vascular network formation. The results of this study were used to test the anti-inflammatory effects of dexamethasone when delivered at the same time as CSP. We treated CD34+-hiPSC-EP-laden hydrogels with either 2 ng/mL dexamethasone alone, 10 μg/mL CSP alone, or both dexamethasone and CSP 5 days after encapsulation for 24 hours.

We verified that CSP treatment alone resulted in a significant decrease in both number of vessel-forming cells and connectivity. When dexamethasone and CSP were delivered simultaneously, the CSP was not able to induce endothelial dysfunction, and no outputs from the computational pipeline were statistically different from dexamethasone alone. As a result, EPs in collagen hydrogels are able to model the effects of candidate drugs as seen in clinical settings.

In conclusion, we have demonstrated that treating induced pluripotent stem cell-derived endothelial progenitors encapsulated in collagen hydrogels with SARS-CoV-2 spike protein is able to replicate endothelial dysfunction seen in clinical settings. Following 24 hours of CSP exposure, we observed both a decrease in the number of cells forming capillary-like vascular networks as well as a decrease in connectivity of the vasculature.

Based on our results this dysfunction is unique to the SARS COV-2 virus and is triggered by release of inflammatory cytokines from the endothelial cells. Same cytokines were measured in patients that had COVID-19 cytokine storm. In addition, we demonstrated that treatment with the corticosteroid dexamethasone is able to block the toxic effects of CSP.

These results demonstrate that we can use this system to model COVID-19’s effect on the endothelium to determine the effectiveness of candidate therapeutics or better understand the impacts of long covid in addition, this system can be utilized more broadly for other diseases that affect the endothelium.


Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.