Angiotensin-converting enzyme 2 (ACE2) was identified as a receptor for the spike (S) protein of SARS-CoV, finally facilitating viral entry into target cells [13]. ACE2 is abundantly expressed in airway epithelial cells and vascular endothelial cells, and it is believed to play a crucial role in the mechanism of acute lung injury induced by SARS-CoV [14]. The ability of spike-Fc protein treatment (3h) to downregulate ACE2 protein expression has been shown in an in vitro system (cell lines) and also in vivo in lung cells of mice [14,15], suggesting that ACE2 pathway may be down-modulated during infection. However, ACE2 is constitutively expressed and released from the apical cell surface of human airway epithelia into airway surface liquid [16]. Of note, its surface down-modulation upon spike protein challenge has been shown to be due to ACE2 shedding mediated by activation of extracellular ADAM17/TACE metalloprotease, which concomitantly induces shedding/production of TNFα [17,18]. Interestingly, ACE2 shedding is enhanced not only by binding with spike protein [17,18], but also by IL-1β and TNFα inflammatory cytokines [16], cytokines that are secreted at relatively high concentration in COVID-19 patients [2]. Moreover, soluble (s)ACE2 (induced or not by virus binding) released from human airway epithelia has been demonstrated to retain both its enzymatic activity and its binding ability for spike viral protein, finally reducing spike protein-mediated viral entry into target cells [16,17]. Therefore, the interaction of ACE2 with spike protein of SARS-CoV induces a cellular “protective” ACE2 shedding feedback response that initially limits viral entry. Nevertheless, ADAM17/TACE-mediated ACE2 shedding or ACE2 enzymatic activity have been shown to intriguingly correlate positively with viral infection and disease complications [17,19,20]. In contrast, HNL63-CoV, which similarly binds to ACE2 through its spike protein, infects ACE2-bearing cells and mainly induces the common cold, leading to neither ACE2 shedding nor SARS [17]. Moreover, catalytically inactive forms of sACE2 can potently inhibit SARS-CoV infection [19,21], suggesting that events downstream of ACE2 shedding and/or its enzymatic activity may indirectly and subsequently favour viral infection and/or disease complications. To this regard, sACE2 was also associated with myocardial pathological conditions [22] and cardiovascular complications including hypotension (known to enhance both renin and angiotensin I, the substrate of ACE/ACE2) and tachycardia were common in SARS-CoV patients [23]. Since spike protein has been shown to not inhibit ACE2 enzymatic activity that is retained by sACE2-spike protein complex [16,17,24] and, in general, sACE2 maintains its enzymatic activity, we cannot consider its higher circulating expression a mere disease biomarker. Indeed, ACE2 shedding might repress its local function but it certainly enhances its circulating/systemic activity. Interestingly, a recent integrative bioinformatics analysis shows that the gene expression of ACE2 in human bronchial cells infected with SARS-CoV is dramatically increased 24h after infection and remained at a high level for at least 2 days, suggesting that ACE2 may be involved in a positive feedback loop post-infection [25]. In the same report, it has been shown that ACE2 expression level in bronchial epithelium obtained by brushing from asthmatic and normal subjects was similar, suggesting that respiratory epithelial cells of healthy subjects and asthmatic patients have similar ability to bind to SARS-CoV-2 through ACE2. Of note, ACE2 was also identified as the receptor for the novel spike protein of SARS-CoV-2 [13]. Although the role of ACE2 in the pathogenesis of SARS-CoV-2 and in inducing lung injury is still unknown, ACE2 behaves similarly to original SARS-CoV [13] (Box 1).