5.3. Safety and Efficacy Concerns of Chelating Agents
Based on ACE2 mechanism of action, there are alternative ways to inhibit ACE2 (and ACE), for example concentration of cation/anion might also targets of intervention to inhibit the zinc metalloprotease ACE2 (and ACE). In this regard, EDTA, a cation chelator, has been shown to be able to inhibit ACE2 activity [79]. Of interest, EDTA binds with 106- and 102-fold higher affinity to Zn2+ than to Ca2+ and Fe2+, respectively [123], and in plasma, free Ca2+ concentration (1.05–1.30 mmol/L) is “only” about 103-fold higher than free Zn2+ (0.1–2.0 µmol/L), while majority of iron in plasma is bound to transferrin. As a consequence, the iron concentration in plasma of healthy subjects is very low on the order of 10−18M [112]. Among the essential metal ions present in the organisms, the specific affinity of EDTA for systemic Zn2+ in physiologic conditions is highlighted by the evidence that prolonged treatment with CaNa2EDTA results in specific zinc depletion that is believed to mediate the teratogenic effects of the drug [123]. As already mentioned, zinc is an important mediator/messenger involved in several cellular activities and excess of free zinc has been shown to be toxic (see also Box 4); however, zinc deficiency has also detrimental effects on growth, neuronal development, and immunity, and in severe cases its consequences can be deadly [102]. For these reasons, treatments with CaNa2EDTA require a careful monitoring of the patient for the therapeutic effects as well as possible complications, so as to titrate the appropriate dosage.
CaNa2EDTA was approved by FDA in chelation therapy for lowering blood lead levels a long time ago (1953). Then, different (commercially available) iron chelating agents, expected to work in chelating zinc ion as well, were approved by FDA. Chelation therapy comprises intravenous or oral administration of chelating agents that remove metal ions such as zinc from the body. Cells synthesising high amounts of ACE or ACE2 needs of high amounts of bioavailable (free) zinc. Therefore, it is conceivable that plasma levels of free zinc may influence not only ACE and ACE2 activities but also their cellular synthesis. Indeed, both ACE and ACE2 synthesis might be particularly sensitive to reduction of free zinc levels in case of their upregulation. In addition, metal chelating agents, by limiting the availability of free zinc to cells, might have effects on both ACE2/ACE synthesis and conformation (when assembled on the plasma membrane). Indeed, the closed conformer of ACE2 homodimer that is the preferential conformation for virus binding [28], needs the presence of both zinc and substrate/inhibitor in the catalytic site [90], suggesting that zinc chelation (differently from MLN-4760) might also inhibit ACE2-mediated viral entry.
Intriguingly, chloroquine has been shown to specifically enhance zinc uptake and its accumulation/sequestration in the lysosomes [124], raising the possibility that it might work on COVID-19 patients by reducing zinc recycling and zinc functions. As already mentioned, free Zn2+ promotes clot stability by binding to fibrinogen (see Box 4 and [91]) and chloroquine has been shown to have an anti-thrombotic activity by inhibiting platelet activation [125]. To this regard, upon platelet activation the release of Zn2+ store from their secretory granules has been shown to participate to the pro-coagulant activity in platelet-dependent fibrin formation [126], suggesting that an elevated free Zn2+ concentration might occur and contribute to thrombotic predisposition in COVID-19 patients, a phenomenon possibly countered by chloroquine.
For all of the above reasons, cation chelating agents, administered alone or in combination with other therapies, might be effective to counter COVID-19 infection, in particular when, induced by hypoxia, both arms of the RAS are upregulated. A scenario that would deserve an investigation. However, some formulations of metal chelating agents carry a black box warning because they may cause serious and fatal renal toxicity and failure, hepatic toxicity and failure, gastrointestinal haemorrhage, arrhythmias, tetany, hypocalcaemia, hypotension, convulsions, respiratory arrest, and agranulocytosis that can lead to serious infections and death [123]. For the sake of completeness, besides renal toxicity linked to the route of drug excretion, EDTA has been shown to be effective in chronic renal artery diseases [123]. Finally, for its teratogenic effects, the drug is also contraindicated in pregnancy [123]. As a result, treatments with metal chelating agents require close patient monitoring, including laboratory tests of renal and hepatic function, and absolute neutrophil count should be monitored before and during treatment. Alternative ways of RAS pathway inhibition are also described in Box 5.
Box 5  Alternative ways of RAS pathway inhibition.
Of interest, a small cationic inhibitor of ACE2 (but not of ACE) has been detected in plasma samples [116]. The endogenous inhibitor might play a compensatory fine-control (within a threshold limit) for normal fluctuation of ACE2 protein in plasma and it has been hypothesized to be a basic AA or a small basic peptide able to compete with ACE2 substrates [116]. I have already mentioned that AA can chelate zinc forming labile zinc complexes of amino acids (see Box 4). It is therefore possible that a basic AA or a small basic peptide capable to specifically accommodate into the catalytic site of ACE2, but into that of ACE, could inhibit ACE2 activity. Indeed, despite the similarities, ACE and ACE2 possess different substrate specificity that depends on the smaller ACE2 binding pocket compared to that of ACE [90]. This specific aspect might determine the specificity for a specific basic amino acid or molecule that is able to accommodate into the catalytic pocket of ACE2 and bind to zinc. Among the possible endogenous ACE2 inhibitors, agmatine, decarboxylated arginine, has a chemical structure that resembles that of an ACE2 inhibitor, NAAE [127]. NAAE is a small molecule that had demonstrated an anti-SARS-CoV activity, by acting on both ACE2 catalytic activity and ACE2 binding domain for spike protein of SARS-CoV [127]. Unfortunately, NAAE has never been used in vivo. Indeed, NAAE is a weak ACE2 inhibitor, it is in fact more than a thousand-fold less potent than MLN-4760; however, if agmatine will be proven to have ACE2 inhibitory activity, it might be helpful to prevent the trigger of the positive feedback loops in the first mild phases of the disease. Indeed, it has an important role in down-regulating NO synthesis reducing NO overproduction by different mechanisms [128]. Of note, NOS pathway has been shown to be upregulated by both Ang (1–7)/MasR and Ang (1–9)/AT2 receptor pathways that are downstream ACE2 activity [38,39,41,43]. Moreover, agmatine has a regulated plasma concentration in the range of 20-80 ng/mL and the use of dietary agmatine has been shown to be safe and effective in reducing neuropathic pain [129]. Moreover, agmatine sulfate is regularly taken as a bodybuilding supplement.
Among natural metal-chelating agents, phytates and folic acid are two chelating agents from vegetables that might reduce zinc intestinal absorption and possibly its systemic concentration (see Box 4 and [111]). Similarly, nicotianamine that is extracted from plants (soybean) is a low-molecular weight metal chelator with high affinity for divalent metal cations. Nicotianamine has been shown to inhibit the activity of both zinc metalloproteases, ACE2 and ACE [130]. In addition, zeolites might also be effective in reducing free zinc availability. Zeolites are a group of aluminosilicate minerals with crystalline microporous structure that are originated from volcanic rocks. Their molecular structure generates cavities that allow high absorbency capacity for a wide range of charged elements such as water, heavy metals, cations and many toxins. Clinoptilolite is one of the zeolites that has been widely studied in veterinary and human medicine. The increased usage of clinoptilolite-based products in vivo has stimulated several investigations on its safety and its positive medical effects related to human health (see [131,132]).
Finally, soluble and catalytically inactive forms of ACE2 have been shown to be potent inhibitors of SARS-CoV infection products [19,21]. An approach that could be pursued (in combination with other therapies) to inhibit SARS-CoV-2 entry. Indeed, soluble forms of ACE2 are expected to protect from viral infection and a similar strategy using a recombinant form of human ACE2 has been proposed not only in COVID-19, but also in ARDS and PAH [108,133,134]. However, it is possible that catalytic active form of ACE2 might favour adverse effects in these specific pathological conditions. To this regard, a clinical trial using recombinant hACE2 protein has been recently started (ClinicalTrials.gov number, NCT04287686) in COVID-19 patients and pilot clinical trials of rhACE2 in ARDS and PAH started in 2017 and 2018, respectively [108,134]; unfortunately, no conclusive results are available yet.