Intriguingly, the binding affinity of ACE2 to SARS-CoV-2 binding domain has been reported to be either equal to or 10- to 20-fold higher than ACE2 binding to SARS-CoV [27,32]. The affinity of spike protein for ACE2 has been shown to correlate with the severity of disease [24]. Although SARS-CoV produces more severe respiratory symptoms than NL63-CoV does, both viral receptor binding domains bind to ACE2 with similar affinity [78], indicating that SARS development is not related to the strength of binding affinity and depends on other mechanisms. In contrast to SARS-CoV, NL63-CoV did not induce ADAM17/TACE-mediated both ACE2 shedding and TNF-α secretion [17], suggesting that increased cleavability of ACE2 receptor and possibly of S1-S2 boundary may be crucial for disease severity. Indeed, the spike glycoprotein of SARS-CoV-2 but not of SARS-CoV, contains a furin-like cleavage site at the S1-S2 boundary which indicates an increased cleavability [13]. Of note, TNF-α and IL-1β were shown to both be upregulated in SARS-CoV-2 [2] and induce viral-independent ACE2 shedding from epithelial airway cells [16]. Moreover, viral-independent ACE2 surface release from epithelial cells was not only inducible by cytokines (e.g., TNF-α and IL-1β) but also constitutively and spontaneously produced when ACE2 surface expression was upregulated [16], for example upon IL-1β stimulation [56]. This suggests that systemic release of proinflammatory cytokines such as TNF-α and IL-1β can mediate an increase of sACE2 and its systemic activity. Of note, activation of ADAM17/TACE metalloprotease was induced by SARS-CoV and necessary for efficient infection (and TNF-α secretion) [17]. Intriguingly, both SARS-CoV infection and concomitant TNF-α secretion were significantly attenuated not only by knock-down of ADAM17/TACE expression by siRNA but also by deletion of the ACE2 cytoplasmic tail which is responsible for ADAM17/TACE activation [17]. Altogether these observations suggest the possibility that SARS-CoV may induce a positive feedback loop leading to surface expression and shedding of both ACE2 and TNF-α. Indeed, upon spike protein binding to ACE2, downstream pathway activation can sustain positive feedback loops at different levels (Figure 2):(1) SARS-CoV can induce IL-1β and TNF-α systemic secretion that can mediate viral-independent surface membrane ACE2 upregulation and shedding. Of note, ACE2 shedding, on one hand, protects from viral infection but, on the other hand, increases circulating/systemic active sACE2, leading to its downstream pathway activation.
(2) Hypoxia in combination or not with hypercapnia can upregulate the activity of both arms of the renin–angiotensin system by inducing renin, ACE and ACE2 synthesis, which can increase expression of Ang I, Ang II, Ang (1–7), Ang (1–9), Ang (1–5) and the inactive metabolite bradykinin (1–7), but also membrane bound ACE2, finally giving more chances to SARS-CoV-2 entry.
(3) ACE2 can induce vasodilatative hypotensive effects by Ang II catabolism. Hypotension can induce again renin and ACE upregulation finally providing further Ang II, as a ACE2 substrate for further Ang (1–7) production.
(4) Ang (1–7) antiproliferative and apoptotic effects, possibly in part through IL-10, may mediate eosinopaenia and lymphopaenia that, on one hand, reduce inflammatory responses but, on the other hand, impair immune system ability to counter virus infection, finally predisposing the organism to further infections. Ang (1–7) immunosuppressive activity, mediated or not by IL-10, may also support the reduced ability to generate an effective immunization to SARS-CoV-2 infection.
(5) Ang (1–7)/MasR pathway can sustain ACE2 synthesis even in the presence of elevated concentrations of Ang II (such as in hypoxia), by inhibiting Ang II/AT1R-mediated down-modulation of ACE2 activity (see Figure 1).
(6) Ang (1–7)/MasR pathway can produce cardiac dysfunction and lung alteration leading to systemic hypoxia, which, in turn, upregulates the activity of both arms of the RAS.
(7) Although the ACE2 catalytic efficiency is 400-fold lower with Ang I than with Ang II [79], high concentration of circulating ACE2 may be able to produce significant increase of Ang (1–9) that, by binding AT2 receptors, can produce arteriolar microvascular thrombosis and local hypoxic conditions finally inducing local upregulation both arms of the RAS.