The pivotal link between ACE2 deficiency and SARS-CoV-2 infection

Evidence in favor of the essentiality of human cell membrane-bound ACE2 and against soluble ACE2 for SARS-CoV-2 infectivity

Introduction An enzyme that metabolizes angiotensin II (Ang II) and other substrates was discovered in 2000, and was termed angiotensin converting enzyme 2 (ACE2) because of its substantial homology with ACE. Since the beginning of the coronavirus disease 2019 (COVID-19) pandemic, ACE2 has received unprecedented attention because in its membrane-bound form, it is the essential cell-entry receptor for severe acute respiratory syndrome coronavirus 2 (1). ACE2 has a single zinc metalloprotease active site and acts as a monocarboxypeptidase, cleaving a hydrophobic or basic amino acid (aa) from the C-terminus of small peptides. Within the renin angiotensin system (RAS), enzymatic actions of ACE2 are in opposition to those of its homolog ACE, such that ACE forms Ang II, whereas ACE2 fosters its degradation by forming Ang 1–7. This action is important at the tissue level, and seemingly minor in the systemic circulation, where other RAS enzymes are involved in the metabolism of Ang II (2). The ability of ACE2 to downregulate the RAS within the kidney prompted us to design shorter forms of ACE2 as novel renal therapeutics that can pass the glomerular filtration barrier (3). In this article, we discuss the different forms of enzymatically active ACE2 present in the kidney and urine. Kidney ACE2 A membrane-bound full-length (FL) form of ACE2 is abundantly expressed in the kidney, mainly at the apical surface of proximal tubules, but also in glomerular epithelial cells (podocytes), parietal cells in the Bowman's capsule, and in the renal vasculature (4). Within the renal vasculature, ACE2 is present in the tunica media, but absent or weakly expressed in endothelial cells (4). Increased albuminuria and kidney lesions were observed in aged ACE2-deficient mice, and after pharmacological ACE2 inhibition in diabetic mice, suggesting that amplifying ACE2 could have therapeutic potential for kidney disease (5). There are different forms of enzymatically active ACE2 that need to be recognized (Figure 1). The tissue form of ACE2 is bound to the cell membrane as an 805 aa–long FL protein. The main part of FL-ACE2 is present in the extracellular space (ectodomain), encompassing the 740 N-terminal aa and containing the catalytic unit, which determines enzyme activity. FL-ACE2 is anchored to the cell membrane by a short transmembrane domain (aa 741–768) and has a small intracellular C-terminal tail (aa 769–805) with cell-signaling properties (6). The ectodomain of the FL-ACE2, when detached from the plasma membrane, is referred to as soluble ACE2. The term soluble means the protein is not membrane or tissue bound, but rather is freely dissolved in plasma, interstitial fluid, and other body fluids, such as the urine and the cerebrospinal fluid. The soluble ACE2 that is shed mainly by the action of a metalloproteinase ADAM17 is catalytically active, and hydrolyzes Ang II to Ang 1–7 (7,8). The FL form of ACE2 can also circulate bound to the membrane of exosomes (9). This form, although circulating, should not be considered soluble ACE2.Figure 1.: Different forms of ACE2 with preserved enzymatic activity and hypothetical formation of small soluble fragments. Upper panel. In kidney tissue, a full-length (FL) angiotensin converting enzyme 2 (ACE2) of 805 amino acids (aa) is a plasma membrane–bound protein with a large part protruding into the extracellular space (ectodomain ACE2 with 740 aa). A small membrane domain (red) anchors the FL-ACE2 in the plasma membrane and a short C-terminal tail (green) with cell-signaling properties appears on the intracellular site of the kidney cell. The 740 aa–long ectodomain dimerizes and the total size of native FL-ACE2 is 98 kDa ×2. The ectodomain contains the catalytic unit (egg shape, brown) responsible for enzyme activity. Lower panel. In the urine, ACE2 can be found as an FL-ACE2 bound to the membrane of exosomes. The FL-ACE2 is the source of soluble ACE2 fragments, which under denaturing conditions (SDS- PAGE), would appear as large (approximately 98 kDa) and small (approximately 62 kDa) fragments. The large 98 kDa soluble ACE2 can maximally encompass the entire ectodomain (approximately 740 aa) or can be approximately 708 aa–long after cleavage by ADAM17. The intact dimerization-prone neck domain (616–726 aa) is present in the FL-ACE2 and the large 740 aa soluble ACE2 that both contain the entire ectodomain. It is not known whether in the 708 aa soluble ACE2, the neck domain is sufficiently intact to enable dimerization (blue thick line) under native conditions or not (absence of dimerization). The small 62 kDa soluble ACE2 can be proteolytically formed from longer FL or soluble ACE2 species, and does not dimerize due to the absence of the neck domain.ACE2 in the urine of patients with CKD (10) and in animal models of diabetic kidney disease is increased (11). This could suggest the possibility that this increase might be the result of more circulatory ACE2. However, when large amounts of soluble recombinant ectodomain ACE2 (aa 18–740) were infused to nondiabetic and diabetic wild-type (WT) mice (11,12) and global ACE2 knockout (KO) mice (11), the resulting marked increase in circulating (serum) ACE2 did not translate into any increase in urinary ACE2 (11,12). This was to be expected because a single chain of ectodomain of soluble ACE2 is approximately 100–110 kDa (3,11), which makes it unlikely to cross the glomerular filtration barrier. Moreover, in a native state, both the soluble and membrane-bound full-length ACE2 dimerize, doubling its monomer size (13). This makes it even more unlikely for such a large molecule to filter through an intact glomerular filtration barrier (12). Our group, moreover, has examined this question using radio-labeled ACE2 ectodomain infused to mice, and has shown there is no renal uptake, which is in contrast with shorter variants of soluble ACE2 that we have bioengineered (Wysocki et al., unpublished 2019). Therefore, increases in serum ACE2 activity as observed in diabetes or other conditions such as in patients with COVID-19, cannot account for elevated urinary ACE2 activity (if the circulating soluble ACE2 indeed comprises the entire ectodomain). Accordingly, when the glomerular filtration barrier is not severely damaged, the source of increased urinary ACE2 must be the kidney itself because the soluble ectodomain ACE2 (that includes 740 N-terminal aa), even if highly elevated in circulation, does not appear in the urine (11,12). In contrast, when the glomerular filtration barrier is highly disrupted, for example, in a mouse model of Alport syndrome, even a large molecule such as soluble ectodomain ACE2 can be recovered in the urine when the circulating levels are increased markedly (12). The next question, therefore, is what form of ACE2 is found in the urine, and where does it come from? Urine ACE2 ACE2 protein measured by ELISA and ACE2 activity is found in small amounts in mouse and human plasma, whereas urine enzymatic ACE2 activity is much higher than in plasma (approximately 5–10 times per volume unit) (11). Studies with mouse and sheep urine moreover showed that urinary ACE2 is capable of Ang II degradation (8,11). In urine of human and mice, two major species of ACE2 can be found. In mouse urine, one of the species of 100–110 kDa is likely the result of shedding from the proximal tubule apical membrane where FL-ACE2 is abundantly expressed (11) (Figure 1). The other ACE2-immunoreactive protein species detectable in mouse urine was of lower molecular size (approximately 75 kDa) (3,7,11). Both ACE2 immunoreactive bands appear enzymatically active (7,11). Of note, the FL-ACE2 and the larger soluble ACE2 (100–110 kDa) might not be easily distinguishable on gel electrophoresis because of their size proximity (11). We showed that the ectodomain of soluble ACE2 of 100–110 kDa can be catabolized by kidney and urine proteases to shorter ACE2 fragments that retain ACE2 enzyme activity (3). The cleavage likely happens from the C-terminal end, which contains a fairly large structural portion (from aa 616 upwards) that has no enzymatic function and can be removed without loss of enzyme activity (3). Part of this structural entity is the so called neck domain (616–726), which is largely responsible for the dimerization of soluble ectodomain (13). A recent study by Nelson et al. (14) reports that ACE2 can be filtered in the urine. On the basis of current views of protein passage across the glomerular filtration, however, only a short form of ACE2 (less than about 75 kDa) could be filtered. In the study of Nelson et al., kidney cross-transplantation between ACE2 deficient and WT mice was carried out to dissect the contribution of ACE2 originating from the kidney versus extrarenal sites to urine ACE2. In an elegant design, four different models were generated: (1) "kidney ACE2KO," with ACE2 deletion confined to the kidney (a kidney transplanted from ACE2KO into WT mouse), (2) "systemic ACE2KO," where ACE2 was deleted from extrarenal tissues (ACE2KO mouse with a kidney transplanted from WT), (3) "total ACE2KO" (ACE2KO mouse with a kidney transplanted from ACE2KO), and (4) "global WT" (WT mouse with a kidney transplanted from WT). When all four types of mice were challenged with a pressor dose of Ang II, increased albuminuria (and cardiac hypertrophy indicative of higher BP) was observed in the total ACE2 KO compared with the other groups. In urine of WT mice, two ACE2 species were detected (sizes of approximately 62 kDa and 98 kDa), which was consistent with previous studies (3,7,11). Intriguingly, in the urine from kidney ACE2KO mice only a shorter form of soluble ACE2 (approximately 62 kDa) was detected and the authors suggest that in response to a pressor dose of Ang II, the 62 kDa urinary ACE2 observed in the "kidney ACE2KO" is being shed from cells outside of the kidney, enters the circulation, and crosses the glomerular filtration barrier, ultimately being excreted in the urine (14). Consistent with this, Ang II has been shown to induce ADAM17 (the protease that can cleave membrane-bound ACE2, forming soluble ACE2) in cardiac tissue with concomitantly increased plasma ACE2 levels (15). Although the cross-transplantation model is useful, it may also permit confounding influence from the interfaces between donor ACE2KO kidney, and the host WT tissues as a source of ACE2 proteins originating from more distal urogenital tissues. Therefore, one might argue that some of the soluble 62 kDa ACE2 observed in the urine from "kidney ACE2KO" might, at least in part, originate from the genitourinary tissues of the host WT animals, such as testicle epithelia, which are very rich in ACE2. To substantiate the notion of soluble ACE2 passage through the glomerular filtration barrier, Nelson et al. (14) performed micropuncture studies in WT mice, which revealed the presence of ACE2 protein in the filtration fluid collected from the Bowman's capsule. This would have been indeed a convincing piece of evidence if the fluid had been collected from the Bowman's space of the "kidney ACE2KO" mice. However, because WT mice were used, this does not provide a definite answer to filterability of the soluble ACE2. Of note, peptide alignment showed that the ACE2 recovered in the Bowman's capsule in the micropuncture experiment and that examined from the blood must be of a much larger size than 62 kDa (14). The peptide alignment suggested a size above the putative ADAM17 cleavage site (between the aa Arg708 and Ser709 for human ACE2 [16] and Met706 for mouse ACE2 [7]). The length of such soluble ACE2 protein that extends beyond ADAM17 cleavage site would be consistent with the approximately 98 kDa soluble ACE2 protein. Because parietal epithelial cells of the Bowman's capsule abundantly express ACE2 on the apical surface, shedding of ACE2 and not filtered ACE2 might have accounted for the observed ACE2 protein. In addition, ACE2 is expressed, albeit much less abundantly, in the podocytes (4). The soluble ACE2, in the Bowman's space therefore, might have been generated through shedding from parietal and visceral (podocytes) epithelial cells and not necessarily originate from filtered plasma. Therefore, if indeed such a large ACE2 (approximately 98 kDa) is filtered from the blood (which we think is only possible with substantial disruption of glomerular filtration barrier), one must infer that this larger form is subsequently converted to the smaller 62 kDa ACE2 within the lumen of the nephron by the action of tubular proteases. The formation of this smaller 62 kDa ACE2 fragment could be the result of proteolytic cleavage of the 98 kDa ACE2 species (3) (Figure 1). Indeed, in a previous study when the recombinant ectodomain soluble ACE2 18–740 was exposed to ACE2KO mouse kidney lysates and urine, the 100–110 kDa ectodomain ACE2 was digested to smaller enzymatically active ACE2 fragments of 75 and 60–70 kDa (3). One also cannot exclude a possibility that ADAM17 has more than one cleavage site within mouse ACE2 (7), and could also be involved in this process. This seems plausible because ADAM17 expression has been found to be augmented by Ang II in cultured proximal tubule cells and associated with increased release of ACE2 protein in the culture media as two bands of 90 kDa and 70 kDa (7). We developed shorter forms of recombinant soluble ACE2 that, unlike the large ectodomain soluble ACE2 (aa 18–740), can indeed be filtered into the urine as demonstrated by injecting them intravenously to global ACE2KO mice (3). The intravenously injected recombinant short soluble ACE2 protein was not only recovered in the final urine in an enzymatically active form, but in addition, was found in the kidney, indicating there is a tubular reabsorption of ACE2 that had been filtered into the urine (3). In summary, the urine contains enzymatically active ACE2 in at least three forms: (1) FL-ACE2 present in exosomes originating in the kidney most likely the proximal tubule, (2) soluble ACE2 with a molecular size of about 98 kDa, which is also likely of renal origin, and (3) smaller fragments that may be the result of action of kidney proteases, or possibly originate from the circulation and be filtered (Figure 1). The main form of circulating soluble ACE2, to our knowledge, has not been completely identified, because western blot analysis is not sensitive to detect the low levels in plasma. Recombinant soluble ACE2 proteins of human or murine origin with 740 aa are too large to pass the glomerular filtration barrier under normal conditions (11). To overcome this limitation, we bioengineered shorter forms of ACE2 that pass the glomerular filtration barrier under normal conditions and that are taken up by the kidney parenchyma and capable to foster the formation of Ang 1–7 from Ang II within the kidney (3). The therapeutic potential of soluble ACE2 for the kidney and beyond clearly deserves more attention than it has had during the two decades since its discovery. Disclosures D. Batlle is a founder of Angiotensin Therapeutics Inc.; and reports receiving a grant and consulting fees from AstraZeneca, and consulting fees from Advicenne, all unrelated to this work. D. Batlle and J. Wysocki are coinventors of patents entitled "Active Low Molecular Weight Variants of Angiotensin Converting Enzyme 2," "Active Low Molecular Weight Variants of Angiotensin Converting Enzyme 2 (ACE2) for the treatment of diseases and conditions of the eye," and "Soluble ACE2 Variants and Uses Therefor", which includes the use of prevention and treatment of COVID-19. J. Wysocki reports having a scientific advisory capacity for Angiotensin Therapeutics Inc. Funding This work was supported by a gift from the Joseph and Bessie Feinberg Foundation and National Institutes of Health grant 1R21 AI166940-01 (to D. Batlle).

The COVID-19 pandemic represents an unprecedented threat to global health. Millions of cases have been confirmed around the world, and hundreds of thousands of people have lost their lives. Common symptoms include a fever and persistent cough and COVID-19 patients also often experience an excess of fluid in the lungs, which makes it difficult to breathe. In some cases, this develops into a life-threatening condition whereby the lungs cannot provide the body's vital organs with enough oxygen. The SARS-CoV-2 virus, which causes COVID-19, enters the lining of the lungs via an enzyme called the ACE2 receptor, which is present on the outer surface of the lungs’ cells. The related coronavirus that was responsible for the SARS outbreak in the early 2000s also needs the ACE2 receptor to enter the cells of the lungs. In SARS, the levels of ACE2 in the lung decline during the infection. Studies with mice have previously revealed that a shortage of ACE2 leads to increased levels of a hormone called angiotensin II, which regulates blood pressure. As a result, much attention has turned to the potential link between this hormone system in relation to COVID-19. However, other mouse studies have shown that ACE2 protects against a build-up of fluid in the lungs caused by a different molecule made by the body. This molecule, which is actually a small fragment of a protein, lowers blood pressure and causes fluid to leak out of blood vessels. It belongs to a family of molecules known as kinins, and ACE2 is known to inactivate certain kinins. This led van de Veerdonk et al. to propose that the excess of fluid in the lungs seen in COVID-19 patients may be because kinins are not being neutralized due to the shortage of the ACE2 receptor. This had not been hypothesized before, even though the mechanism could be the same in SARS which has been researched for the past 17 years. If this hypothesis is correct, it would mean that directly inhibiting the receptor for the kinins (or the proteins that they come from) may be the only way to stop fluid leaking into the lungs of COVID-19 patients in the early stage of disease. This hypothesis is unproven, and more work is needed to see if it is clinically relevant. If that work provides a proof of concept, it means that existing treatments and registered drugs could potentially help patients with COVID-19, by preventing the need for mechanical ventilation and saving many lives.

Letter by Weir Regarding Article, "Angiotensin-Converting Enzyme 2: SARS-CoV-2 Receptor and Regulator of the Renin-Angiotensin System: Celebrating the 20th Anniversary of the Discovery of ACE2".

Background and AimsSevere acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has led to dramatic loss of lives due to COVID-19. Individuals with chronic conditions, including patients with kidney failure and/or kidney transplants, are affected more substantially due to multiple comorbidities and altered immune system. The first step of this infection process is the binding of SARS-CoV-2 with angiotensin-converting enzyme 2 (ACE-2) receptor, followed by its priming by transmembrane protease serine 2 (TMPRSS2). We hypothesized that the expressions of ACE-2 and TMPRSS2 are increased in microvasculature, in addition to increased circulating soluble ACE-2 levels in patients with end stage kidney disease (ESKD); i.e. prerequisites to explain to why ESKD patients are susceptible to SARS-CoV-2 infection. Therefore, we assessed if there is a difference in the expression of ACE-2 and TMPRSS2 receptors in the resistance artery and subcutaneous adipose tissue, alongside circulating soluble ACE-2 levels in ESKD patients versus controls.MethodA total of 210 participants were enrolled, representing 80 ESKD and 73 healthy controls for soluble ACE-2 analysis, and 31 ESKD and 26 healthy controls for isolated subcutaneous vasculature bioassay. Immunofluorescence techniques were performed for the detection and evaluation of ACE-2 and TMPRSS2 in isolated subcutaneous resistance artery (200-300 µm of internal âCE€) and adipose tissue. Soluble ACE-2 protein concentration was detected using commercially available ELISA kits.ResultsSoluble ACE-2 levels were significantly higher in ESKD (3.8 ng/mL, IQR 2.4-5.5, n=80) vs control groups (2.7 ng/mL, IQR 2.1-3.7 ng/mL, n=73). There was no difference in soluble ACE-2 between females and males in either group. Soluble ACE-2 was positively correlated with IL-6 (rho=0.257, p=0.02, n=80), while it was negatively correlated with cholesterol (rho= -0.248, p=0.02, n=78) in ESKD patients. The expression of ACE-2 receptor was observed both on endothelium and vascular smooth muscle cells (VSMCs) in arteries from both groups. The expression was higher in ESKD patients (19.1%, n=23) vs controls (15.4%, n=15). Patients with ESKD on ACE-inhibitor/angiotensin receptor blocker treatment showed higher expression of ACE-2 vs. non-treatment group (treatment: 20.2%, n=12 vs non-treatment: 12.8%, n=11) in resistance artery. In subcutaneous adipose tissue the ACE-2 staining was not statistically different among the groups (ESKD: 2.9%, n=10 vs controls: 3.6%, n=10). In addition, TMPRSS2 was expressed both on endothelium and VSMCs in resistance artery, however there was no difference in the expression (ESKD: 8.4%, n=23 vs controls: 10.2%, n=15) between the groups.ConclusionSoluble ACE-2 levels and ACE-2 receptor expression in the vasculature were higher in patients with ESKD as compared to controls. The ACE-2 receptor is present both in the endothelium and VSMCs from arteries in peripheral microcirculation. This supports the suggestion that the uremic milieu induces an optimal environment for SARS-CoV-2 entrance in microcirculation with following consequence on the vasculature during the COVID-19. Similarly, TMPRSS2 expression was observed in vessels from both groups, while increased expression of ACE-2 receptor was observed in those ESKD patients receiving ACE-inhibitor/angiotensin receptor blocker treatment. Further studies are warranted to assess possible sex differences in the target receptor expressions with further elaboration on specific treatment regime(s) for different comorbidities present in patients with ESKD.

Very low reporting rate of connective tissue diseases among coronavirus disease 2019 (Covid-19) patients and the renin-angiotensin system – An overlooked association?

Human recombinant soluble ACE2 in severe COVID-19

Antihypertensive drugs and risk of COVID-19?

Angiotensin-converting Enzyme 2–containing Small Extracellular Vesicles and Exomeres Bind the Severe Acute Respiratory Syndrome Coronavirus 2 Spike Protein

SARS-CoV-2 infection and ACE2 inhibition.

COVID-19, vaccines and deficiency of ACE2 and other angiotensinases. Closing the loop on the "Spike effect"

Probiotics and COVID-19

Angiotensin converting enzyme-2 (ACE-2) is the cell-surface receptor enabling viral uptake of corona virus 2019 (SARS-CoV-2), thus ACE-2 is a first step towards COVID-19 disease. ACE-2 is a metalloenzyme located primarily on the apical surface, and serves as the entry point also for other coronaviruses, including HCoV-NL63 and SARS-CoV. Throughout evolution, ACE-2 precedes renin, suggesting that ACE-2's role changed over time.1.

To investigate changes in the angiotensin converting enzyme 2 (ACE2) and angiotensin (1-7) [Ang (1-7)] and to explore the role of ACE2-Ang (1-7)-Mas receptor axis in hypertension with heart failure with preserved ejection fraction (HFPEF). Methods: A total of 70 patients with primary hypertension and preserved left ventricular ejection fraction (LVEF>50%) were recruited and patients were divided into a hypertension group (HBP) and a heart failure with preserved ejection fraction group (HFpEF) according to the diagnostic criteria of HFpEF. Thirty-five healthy participants were selected randomly as a control group. Enzyme linked immunosorbent assays (ELISA) method was used to detect concentration of Ang (1-7), ACE2, angiotensin II (Ang II), brain natriuretic peptide (BNP) in plasma. Male Sprague- Dawley (SD) rats was randomly divided into 2 groups: An HFpEF group (n=16) and a sham group (n=8). Rats (n=8) in the AAC group were given Ang (1-7) [0.5 mg/(kg.d), intraperitoneally] for 6 weeks, and the rest were given equal dose normal saline. Then all the rats were killed, and the hearts were taken out for hematoxylineosin (HE) staining. The protein expressions of angiotensin converting enzyme (ACE), ACE2, and Mas receptor were detected by Western blot. Results: The BNP and Ang II were significantly increased in the HBP group and the HFpEF group compared with the control group (P<0.01). There were not significantly different in levels of ACE2 and Ang (1-7) between the HBP group and control group (P>0.05), whereas those levels were significantly increased in the HFpEF group compared with the HBP group and control group (P<0.01). HE staining showed obvious hypertrophy of myocardial cell in the AAC group compared with the sham group. Hypertrophy of myocardial cell in the AAC+Ang (1-7) group was significantly higher than that in the AAC group. Expressions of ACE, ACE2, and Mas receptor proteins were significantly higher in the AAC group than those in the sham group (P<0.05), while the expressions of ACE2 and Mas receptor proteins in the AAC+Ang (1-7) group were significantly higher than those in the AAC group (P<0.05). There was no significant difference in the ACE protein expression between groups (P>0.05). Conclusion: ACE2 and Ang (1-7) are important predictive factors for the severity of heart failure and myocardial remodeling of HFpEF with hypertension; ACE2-Ang (1-7)-Mas receptor axis may play a protective role in preventing myocardial remodeling in HFpEF with hypertension.

he prevalence of coronavirus disease 2019 (COVID-19) has posed a great threat to people's health worldwide, bringing a great challenges to the public healthcare systems. A recent study has confirmed that severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) uses severe acute respiratory syndrome coronavirus (SARS-CoV) receptor angiotensin-converting enzyme 2 (ACE2) for host cell entry. 1 ACE2 expression was previously found to correlate with susceptibility to SARS-CoV infection in vitro.