Entry of SARS-CoV2 through the basal surface of alveolar endothelial cells – A proposed mechanism mediated by CD147 in Covid-19

Severe Covid-19 disease is associated with endothelial infection, viraemia, and multi-organ dysfunction. The process through which SARS-CoV2 causes severe disease is yet to be determined. Here, we propose that in severe Covid-19 infection, SARS-CoV2 reaches the host bloodstream by infecting endothelial cells through their basal surface. This occurs, independently of ACE2, through CD147, a putative SARS-CoV2 receptor. The pathway proposed here encourages research on the mechanisms mediating endothelial cell infection

with either the alveolar cavity or bloodstream, respectively (Morgenroth and Ebsen, 2008). In vitro studies have demonstrated that human coronavirus infections of polarised epithelial cells result in preferential release of progeny viral particles from the apical surface of these cells (Cong and Ren, 2014;Jia et al., 2005;Pyrc et al., 2010;Sims et al., 2008;Tao et al., 2013;Tseng et al., 2005;Wang et al., 2000). However, a small proportion of all viral release occurs from the basal surface of SARS-or MERS-CoV infected polarised epithelial cells (Tao et al., 2013) (Figure 1: A & B).
The proximity of epithelial and endothelial cells in the alveoli suggests that SARS-CoV2 may infect both types of cells in succession, ultimately resulting in release of virus to the blood. At times, only 100nm of basement membrane separates the basal surfaces of epithelial and endothelial cells in the alveoli (Morgenroth and Ebsen, 2008). In vitro studies demonstrate that SARS-CoV and MERS-CoV, are released from the basal surface of infected polarised epithelial cells (Tao et al., 2013). In vivo, this could mean that infected alveolar epithelial cells release viral particles to the basement membrane they share with endothelial cells. If this is the case then, alveolar endothelial cells might become exposed to viral particles at their basal surface, and thus become infected. Since endothelial cells are in direct contact with the bloodstream, SARS-CoV2 viraemia may occur as a result of viral release by the infected alveolar endothelium.

SARS-CoV2 can infect endothelial cells via ACE2
Such mechanism would depend on the permissiveness of endothelial cells to allow SARS-CoV2 to enter them via their basal membrane. Although, SARS-CoV2 has been shown to infect endothelial cells in-vitro and in-vivo (Monteil et al., 2020;Varga et al., 2020), the cellular surface through which viral entry occurs is uncertain. Monteil et al. (2020) has demonstrated that blood vessel organoids can be infected by SARS-CoV2 and that this infection depends on the ACE2 receptor. However, the path of endothelial entry was not determined. In fact, no experiments to date have studied whether the basal surface of polarised endothelial cells is susceptible to human coronavirus infection.

Subcellular expression of ACE2 determines susceptibility to coronavirus infection
Studies of polarised epithelial cell infection by SARS-CoV may shed light on endothelial susceptibility. In-vitro, polarised epithelial cells are readily infected by SARS-CoV at their apical surface, but are not susceptible to SARS-CoV infection through their base (Jia et al., 2005;Ren et al., 2006;Sims et al., 2008;Tseng et al., 2005). This has been explained by the fact that while ACE2, the receptor for SARS-CoV and SARS-CoV2 (Hoffmann et al., 2020;Li et al., 2003), is expressed in the apical surface of polarised epithelial cells, it is not expressed in the basal surface (Jia et al., 2005;Ren et al., 2006;Tseng et al., 2005). Thus, the susceptibility of a polarised cell to SARS-CoV2 infection may depend on the subcellular location of ACE2 receptor expression.

Subcellular pattern of ACE2 expression in endothelial cells
Endothelial cells are highly polarised (Worzfeld and Schwaninger, 2016) and are known to express ACE2 on their cell surface (Cao et al., 2012;Donoghue Mary et al., 2000). However, the subcellular localisation of ACE2 in endothelial cells has not been characterised.
Nonetheless, it may be hypothesised that since ACE2 mediates the breakdown of the bloodborn angiotensin I and II (Donoghue Mary et al., 2000;Rice et al., 2004;Walker et al., 1976), ACE2 expression may be primarily located at the apical surface of endothelial cells, which faces the bloodstream.

ACE2 independent infection of endothelial cells via their basal surface
The presence of a coronavirus receptor at the basal cell surface appears to be necessary for viral entry at that site. As described above, SARS-CoV cannot infect the ACE2-deficient basal Since there is evidence of endothelial infection in cases of Covid-19 Varga et al., 2020), alternative receptors to ACE2 should be investigated as candidate mediators of SARS-CoV2 basal surface entry to endothelial cells. CD147 a novel mechanism for SARS-CoV2 entry CD147 has been proposed as an alternative receptor for SARS-CoV2 . CD147 a glycoprotein, previously identified as a putative receptor for SARS-CoV (Chen et al., 2005), has been shown to bind the SARS-CoV2 spike protein. In addition, antibodies against CD147 prevent SARS-CoV2 infection in vitro . Based on this evidence it has been proposed that CD147 may represent a novel invasive route for SARS-CoV2 . Interestingly, a randomised clinical trial using Meplazumab, a humanised antibody against CD147, for Covid-19 has been started and results from a preliminary trial appear promising Tang-Du Hospital, 2020). Cellular pattern of CD147 expression CD147 is found throughout the body including the lungs (Jouneau et al., 2011;The Human Protein Atlas, 2020). It is expressed in the cell membrane of epithelial cells (Deora et al., 2005;Moheimani et al., 2018), T-cells (Hu et al., 2010;Koch et al., 1999) and endothelial cells (Maïssa et al., 2017;Patrizz et al., 2020). Physiologically, CD147 can preferentially localise to the basal domain of polarised cells, as it has a functional role in chaperoning carboxylate transporters to the basal surface of epithelial cells (Deora et al., 2005;Kirk et al., 2000). In addition, in pathological states it can activate matrix metalloproteinases (MMPs) at the basement membrane of epithelial and endothelial cells as a response to injury (El-Asrar et al., 2017;Gabison et al., 2005;Jouneau et al., 2011;Xie et al., 2019).

CD147 in endothelial stress and SARS-CoV2 infection
It is important to note that CD147 expression and function are enhanced by proteins that have been implicated in the pathogenesis of human CoV disease (Sardu et al., 2020;Tanaka et al., 2013), namely cyclophilins and angiotensin II (Pons et al., 2011;Seizer et al., 2014).
Interestingly, cyclophilins and angiotensin II also play roles in the endothelial dysfunction seen hypertension, diabetes and heart disease (Wang et al., 2010;Zhang et al., 2017). Given the apparent relationship between severe Covid-19 disease and the metabolic syndrome diseases (Dutta et al., 2020), as well as its relationship with SARS-CoV viraemia , CD147 represents a strong candidate as a mediator of SARS-CoV2 endothelial cell infection and viraemia in severe Covid-19 disease.

Conclusion
Thus, here we propose a mechanism of alveolar endothelial infection by SARS-CoV2 mediated by CD147 at the basal surface of endothelial cells. This may occur because alveolar epithelial cells release SARS-CoV2 particles from their basal surface to the shared basement membrane. In this mechanism, CD147 mediated SARS-CoV2 entry is promoted by the endothelial cellular stress associated with metabolic syndrome diseases (Figure 1: C).
Future work should help determine the validity of several assumptions on which this hypothetical mechanism is based. The following questions should be answered:  Do infected alveolar epithelial cells release SARS-CoV2 at their basal surface in vivo?

 Can polarised endothelial cells be infected via the basal surface in vitro?
 What is the pattern of subcellular ACE2 expression in endothelial cells?
 Does endothelial cellular stress change the pattern of SARS-CoV2 infectivity?
Tackling these questions may result in finding pharmacological and screening targets that could be used to alleviate the global burden of Covid-19.

Contributors
ADAS carried out the literature search and drafted the paper. ADAS and FCAS reviewed the evidence and formed the theoretical mechanism. FCAS revised the manuscript. ADAS and FCAS have reviewed the paper prior to submission. This manuscript has not been currently accepted or published elsewhere.

Declaration of interests
We declare no competing interests.