Don't judge too RAShly: the multifaceted role of the renin-angiotensin system and its therapeutic potential in COVID-19.

to the editor: We read with great interest the recent Letter by Abassi and colleagues (1) highlighting—among others—the emerging role of the renin-angiotensin system (RAS) as both potential mediator as well as therapeutic target in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and subsequent development of COVID-19. The requirement for binding of the SARS-CoV-2 spike proteins to angiotensin-converting enzyme 2 (ACE2) as a prerequisite for host cell infection (7), and the recent observation that patients treated with ACE inhibitors or angiotensin receptor blockers (ARBs) have a worse outcome (6), has fueled the recognition of a putative mechanistic role of the RAS in COVID-19. However, we feel that the conclusions reached in this letter, namely that 1) ACE inhibitors and ARBs may be detrimental in COVID-19 patients by upregulating ACE2 and 2) downregulating expression or inhibiting activity of ACE2 could present a promising strategy to treat these patients may be misleading, and potentially cause more harm than good in patients suffering from COVID-19.

Our argument is based on the following considerations: First, although ARBs have been shown to upregulate ACE2 in some studies (3), the association with higher mortality in COVID-19 patients can largely be explained by the higher prevalence of hypertension (and thus, ARB treatment) in the elderly, which are contemporaneously yet independently also at increased risk for lethal outcome in COVID-19. Since discontinuation of antihypertensive therapy by ARBs poses a very real risk for cardiovascular events in these patients over an unproven advantage in terms of susceptibility to SARS-CoV-2 infection, various national and international agencies have issued advisories against such a strategy (5).

Second, while higher levels of ACE2 in lung epithelial cells may potentially increase the uptake of the SARS-CoV-2 virus, ACE2 is also key in protecting the lung from acute respiratory distress syndrome (ARDS) (6), the main cause for lethality in COVID-19 patients (2, 4). Therefore, interventions targeting ACE2 expression may have to be implemented on a personalized basis: While it may be beneficial to have less ACE2 in noninfected subjects to reduce the likelihood of infection, we would need to sustain or boost ACE2 activity for lung protection once patients are tested positive for SARS-CoV-2. The latter intervention seems particularly appropriate as SARS-CoV infection rapidly downregulates endogenous ACE2 expression (10), thus depriving the lungs from an important endogenous protection mechanism.

Third, the proposed approach using ACE2 inhibitors like MLN4760, which bind with high affinity to the active zinc site of the enzyme, thus emulating the transition state during peptide hydrolysis, may not prove effective to block viral entry as viral binding is not related to the enzymatic activity of ACE2, but to the quantity of ACE2 molecules expressed on the cell surface (which is, however, not affected by ACE2 inhibitors). Worse, this approach could be detrimental, as the reduction in ACE2 activity would again increase the lung’s susceptibility for COVID-19 related ARDS.

While the protective effects of ACE2 in ARDS were initially attributed to a reduced activation of the angiotensin receptor type 1 by angiotensin (Ang) II (8), subsequent work by us and others identified that these protective effects are primarily caused by the cleavage product of ACE2, the heptapeptide Ang-(1–7). Ang-(1–7) exerts important barrier-stabilizing and anti-inflammatory effects which could be harnessed to protect lungs from acute injury in various preclinical animal models (9, 11). These beneficial effects are, at least in part, mediated via binding of Ang-(1–7) to its receptor, the G protein-coupled receptor Mas (9). As such, treatment with Ang-(1–7) may present a particularly promising approach for the treatment of COVID-19-related ARDS, as it 1) would exert barrier-protective and anti-inflammatory effects independent of ACE2 activity or expression status, 2) could thus substitute for loss of ACE2 in SARS-CoV-2 infection, 3) would be more efficient in stimulating Ang-(1–7) receptors than increasing ACE2 activity with no additive risk in terms of facilitating viral entry, and 4) downregulates ACE2 mRNA expression in a classic feedback loop, thus potentially reducing viral entry (own unpublished data). In contrast to Ang-(1–7), which needs to be given intravenously or subcutaneously or has to be formulated with cyclodextrane, nonpeptidic small-molecule activators of Ang-(1–7) receptors such as the Aventis compound AVE0991 (also mentioned by Abassi and colleagues) bear seeming advantages in that they can be administered orally. This advantage is, however, outweighed by a low bioavailability of the substance, which in the past decade ultimately stopped the company’s AVE0991 program. Since the patients likely to benefit most from treatment with Ang-(1–7) receptor agonists are expected to be those already hospitalized with respiratory symptoms or pneumonia yet not ARDS (see below), they will commonly have a venous line through which Ang-(1–7) could be readily infused.

Importantly, the pronounced barrier-protective effects of Ang-(1–7) are ideally suited for the slow clinical progression of COVID-19. Compared with classic ARDS, which develops acutely and as such, does not allow for preventive measures, COVID-19 is characterized by a gradual onset of symptoms that progress over time from flulike symptoms to pneumonia and ultimately, respiratory failure (12). As such, Ang-(1–7) presents in our eyes a promising candidate drug in COVID-19 that could be administered to patients as adjunctive therapy early after the onset of respiratory symptoms before their progression to alveolo-capillary barrier failure, lung edema, and respiratory failure.

DISCLOSURES

T.W. and W.M.K. are the inventors of the 2010 filed patent application “Use of an Ang-(1–7) receptor agonist in acute lung injury.”

AUTHOR CONTRIBUTIONS

T.W. and W.M.K. drafted manuscript; T.W. and W.M.K. edited and revised manuscript; T.W. and W.M.K. approved final version of manuscript.