Angiotensin-converting enzyme 2 and COVID-19: patients, comorbidities, and therapies.

On March 11, 2020, the World Health Organization declared coronavirus disease 2019 (COVID-19) a pandemic, and the reality of the situation has finally caught up to the widespread reach of the disease. The presentation of the disease is highly variable, ranging from asymptomatic carriers to critical COVID-19. The availability of angiotensin-converting enzyme 2 (ACE2) receptors may reportedly increase the susceptibility and/or disease progression of COVID-19. Comorbidities and risk factors have also been noted to increase COVID-19 susceptibility. In this paper, we hereby review the evidence pertaining to ACE2's relationship to common comorbidities, risk factors, and therapies associated with the susceptibility and severity of COVID-19. We also highlight gaps of knowledge that require further investigation. The primary comorbidities of respiratory disease, cardiovascular disease, renal disease, diabetes, obesity, and hypertension had strong evidence. The secondary risk factors of age, sex, and race/genetics had limited-to-moderate evidence. The tertiary factors of ACE inhibitors and angiotensin II receptor blockers had limited-to-moderate evidence. Ibuprofen and thiazolidinediones had limited evidence.

INTRODUCTION

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) caused over 3 million infections and over 250,000 deaths worldwide 2 months after the World Health Organization (WHO) declared the virus-induced disease a pandemic. Mathematical models have shown drastic exponential growth in these infections. However, the coronavirus disease 2019 (COVID-19) does not indiscriminately affect the population. Certain comorbidities and risk factors have shown increased susceptibility and disease severity (1, 2).

The average observed case-fatality ratio is around 4% but varies widely (3). Earlier reports suggested that elderly patients with comorbidities, particularly diabetes, cardiovascular disease, and hypertension, suffered greater mortality than younger patients (2, 4–6). Nevertheless, younger patients with no apparent disease also appeared prone to rapid progression, severe/critical disease, and death.

Angiotensin-converting enzyme 2 (ACE2) is an important regulator of the renin-angiotensin system (RAS), and SARS-CoV-2 propagates on to host ACE2 receptors as a vehicle for invasion of human cells (7). ACE2 is found mostly in the endothelium, lungs, heart, kidneys, and intestines, which parallels the detection of SARS-CoV-2 in multiple organs found in silico analysis (7–10). Moreover, certain comorbidities have been shown to increase ACE2 levels, which may increase COVID-19’s severity and susceptibility potential.

This article reviews the relationship among ACE2 levels, common comorbidities, and risk factors in COVID-19 cases. An electronic search was conducted in LitCovid, PubMed, Google Scholar, WHO, and Centers for Disease Control and Prevention (CDC) databases. Search terms included COVID-19, SARS-CoV-2, 2019-nCOV, angiotensin-converting enzyme 2, and ACE2. Manuscripts published were reviewed; relevant references were checked. The information in this review is current as of this article’s writing (November 3, 2020), since new data will likely emerge during the pandemic.

PHYSIOLOGY OF ACE2

ACE2/ANG-1–7/Mas axis is the prominent counterregulator against activated RAS pathophysiology (Fig. 1). SARS-CoV-2 uses the receptor ACE2 for host cell entry and viral replication.

Figure 1.

Schematic representation of renin-angiotensin system (RAS) and homeostatic features. RAS regulates vascular function, blood pressure, and fluid and electrolyte balance. The liver synthesizes and releases angiotensinogen into the circulatory system. Angiotensinogen is then converted to the decapeptide angiotensin I until it reaches the lungs, where angiotensin-converting enzyme (ACE) converts it to the octapeptide angiotensin II (ANG II) (11, 12). ANG II, a powerful vasoconstrictor, has short-term presence in the blood before it is metabolized (13). The proinflammatory effects of ANG II are further mediated by ANG II type I receptor, which stimulates aldosterone secretion from the adrenal medulla and antidiuretic hormone from the posterior pituitary. A key regulator of RAS is ACE2, a monocarboxypeptidase, that metabolizes and inactivates ANG II to the hepapeptide angiotensin 1–7 (ANG-1–7), which, after binding with the G protein-coupled receptor MAS receptor, decreases the vasoconstrictor stimulus (11, 12). ANG-1–7 can also be produced directly via zinc metallopeptidase neprilysin/prolyl endopeptidases or through conversion of angiotensin 1–9 by ACE but with lower efficiency (11, 12). ANG-1–7’s other protective effects include anti-fibrotic, anti-inflammatory, antioxidant and antihypertrophic qualities (14). Knockout mice have shown that reduced ACE2 increases tissue and circulating levels of angiotensin II (13, 15). Additionally, the therapeutic treatment mechanism is shown here: β-blockers inhibit renin and prevent conversion of angiotensinogen to angiotensin I; ACE inhibitor blocks ACE and prevents conversion of angiotensin I to II; angiotensin receptor blockers (ARBs) prevent angiotensin II from binding to its receptor.

SARS-CoV-2 is transmitted via respiratory droplets (>5 μm) or aerosol transmission (<5 μm) (16, 17). Early in infection, the virus targets the nasal and oral mucosa, bronchial epithelial cells, and pneumocytes as they are enriched with ACE2 receptors, the entry points in facilitating SARS-CoV-2’s access to the airways and the body in general (18, 19). On the virion surface, Spike’s glycoprotein (S protein) mediates receptor recognition and membrane fusion onto ACE2’s peptidase domain, together with the type 2 transmembrane serine protease (TMPRSS2), with a high affinity of Kd ∼15 nM (20–22). Viral entry further upregulates ADAM17 protease activity, which downregulates ACE2 by cleaving the receptor from the cell surface (“shedding”), thereby shifting protective ACE2/angiotensin 1–7/Mas axis toward the disease state and accumulation in angiotensin II (ANG II) (12, 14). Recent studies show that SARS-CoV-2 binds with greater affinity than the prior SARS-CoV (23–25) and resiliently binds to ACE2 receptors with a low-species barrier (26). Additionally, current studies suggest antigenic drift variations in the harboring S protein to possibly present with increased affinity to ACE2 docking sites. For example, 614 G variants positively correlated with infectivity and fatality rates (27–29), despite reduced affinity for ACE2 due to faster dissociation rates (30). The continuing evolution of SARS-CoV-2’s durable and adaptable bind with ACE2 receptors could partly explain the transmissibility affecting most nations.

PHYSIOLOGY OF EXTRAPULMONARY MANIFESTATIONS

COVID-19 infection ranges from asymptomatic to critical. Figure 2 highlights key phases of disease progression from presymptomatic/asymptomatic to late phase/severe disease involving direct organ-toxicity, systemic inflammatory response syndrome (cytokine storm), and endothelial damage and thromboinflammation (12, 32).

Figure 2.

Infected individuals may remain asymptomatic up to a week before encountering mild to moderate symptoms of fever, dry cough, sore throat, loss of smell and taste, or head and body aches. Eventually severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), an enveloped, nonsegmented positive-sense RNA virus, infects angiotensin-converting enzyme 2 (ACE2)-expressing type II alveolar epithelial cells in the lower respiratory tract, where 90% of symptomatic individuals have reported pneumonitis (31). Progression to severe/critical involves viral toxicity, disruption of the epithelial-endothelial barrier, complement depositions, and hyperinflammation, commonly requiring ventilation and/or life support for multiorgan injury (31). Severe COVID-19 is defined as dyspnea, respiratory rate ≥30/min,  ≤93%, / <300, and/or lung infiltrates >50% within 24–48 h, compared with critical disease, which further involves respiratory failure, septic shock, and organ dysfunction/failure (5). [Figure reproduced with permission from Matheson and Lehner (31).]

ACE2 and Organ-Toxicity

SARS-CoV-2’s direct tropism onto ACE2 receptors on varying tissue, aside from lung parenchyma, is believed to cause multiorgan injury (7–10). Histopathological reports confirm SARS-CoV-2’s direct organotropism, thus speculate preexisting conditions to aggravate COVID-19’s course (9, 33–38). The mechanism of the virus’s systematic spread, whether lymphatic, hematogenous, or otherwise, remains elusive.

ACE2 and Cytokine Storm

Host cells lyse during rapid viral replication directly producing proinflammatory cytokines such as TNF-α, IL-1β, IL-6, INF-γ, monocyte chemoattractant protein-1, and TGF-β. INF-γ and IL-6 are prominent inflammatory markers linked to the worst outcomes (39–45). Specifically, IL-6’s signaling pathway elevates the following serum inflammatory markers: C-reactive protein (CRP), procalcitonin, erythrocyte sedimentation rate, D-dimer, fibrinogen, lactate dehydrogenase, ferritin, and cardiac stress markers (12, 32, 46). Excessive inflammatory responses may also predispose patients to thrombosis due to platelet activation, endothelial dysfunction, and stasis (47). Thrombosis further exacerbates proinflammatory effects of endothelial injury and phagocytic responses.

In addition, the imbalance of the ACE2/ANG-1–7/Mas axis contributes to systemic inflammation. INF-γ and SARS-CoV-2 downregulate ACE2, resulting in elevated ANG II, which causes pulmonary vasoconstriction, additional inflammation, and oxidative and fibrotic organ damage, ultimately advancing toward acute respiratory distress syndrome (48–50). Studies have demonstrated that the loss or downregulation of ACE2 perpetuates acute lung injury (ALI); in conjunction, rhACE2 improves it (49, 51–53). Conversely, Ziegler et al. (54) found that INF-γ and influenza increased ACE2 in human nasal epithelia and lung tissue, suggesting a finer balance between INF-γ and SARS-CoV-2 regulation of ACE2.

A combination of viral-induced lysis, thrombosis, and dysregulation of ACE2 overwhelms the systemic release of cytokines, resulting in cytokine storm. Emerging studies reveal a positive correlation of type-I INF deficiency to clinical severity (55, 56). Therapeutics in study, e.g., dexamethasone, interferon-α-2b, interferon β-1b, aim to regulate hyperinflammatory responses (57–59).

ACE2 and Thrombosis

Risk of arterial/venous thrombosis, acute coronary syndrome, and stroke in severe/critical patients remains high: 29.4%–50% (60–63). Endothelial dysfunction in COVID-19 patients results from ACE2-mediated binding of SARS-CoV-2, hyperinflammation, and prothrombin upregulation (32, 33, 64). Direct viral toxicity creates microthrombi and fibrinous exudates that dysregulate coagulant pathways (33, 38, 65–67). In conjunction, endothelial injury (i.e., elevated von Willebrand) and endothelialitis facilitate proinflammatory effects by triggering elevated D-dimer, prothrombin time, and activated partial thromboplastin time prolongation, fibrinogen, and complement. Endothelialitis involves neutrophil and macrophage-facilitated cytokine release and neutrophil extracellular trap formation, which further damage the endothelium and activate extrinsic and intrinsic coagulation pathways (32, 63, 68). Moreover, hypoxia-mediated upregulation of hypoxia inducible factor-1 (HIF-1) activates cytokines, tissue factor, and plasminogen activator inhibitor type-1 (63, 67). These manifestations promote a procoagulative state, aggravating Virchow’s triad (32, 63, 68, 69). Heparin and other anticoagulants reduce thrombotic complications, decreasing COVID-19 patient mortality (47, 70, 71).

ACE2 and Susceptibility/Severity of COVID-19

ACE2 may mediate unique susceptibility to SARS-CoV-2, but the enzyme’s dysfunction evidently contributes to the severity of COVID-19.

Prior studies on SARS-CoV illustrate significant correlations to ACE2 expression in vitro (72–74). However, evidence between ACE2 expression and SARS-CoV-2 remains new. Limited reports of hACE2 mice models depict increased SARS-CoV-2 viral load, compared with controls (75–77). Rapidly emerging human studies mirror similar sentiments. Early COVID-19 histopathological reports noted increased SARS-CoV-2 expression and ACE2-positive receptors in affected lung, cardiac, and renal tissue (9, 33–38). For instance, increased ACE2-positive endothelial cells were observed in significant endothelial injury, potentially correlating ACE2 expression to increased disease progression (33). Also, ACE2 RNA expression was associated with increased viral load from 430 nasopharyngeal swabs (78) and high admission viral load predicted in-hospital mortality in n = 2,914. Here, the mortality rate was 38.8% in high viral-load, 24.1% in medium viral load, and 15.3% in low viral-load patients (P < 0.001) (79). In conjunction, viral load was an independent predictor of mortality in a large hospitalized cohort, n = 1145—5.2 versus 6.4 mean log10 copies/mL, respectively, in alive versus deceased patients (80).

Furthermore, limited clinical studies depict elevated plasma ANG II and aldosterone levels correlating to COVID-19 severity (81–84). Significant plasma ANG II level elevations were seen in 90.2% of the observed COVID-19 cases, especially in 100% of the critical COVID-19 cases (84); although Henry et al. (85) saw no differences in ANG II regarding disease severity. Liu et al. further delineated markedly increased ANG II levels linearly associated to viral loads and lung injury (82), and multivariate analyses observed aldosterone levels positively associated with severity (83).

Initial data indicate that increased ACE2 receptor availability poses greater severity to COVID-19, including an increased viral load, organ-toxicity, hyperinflammation, and endothelial dysfunction. Later on in the disease progression, ADAM17, together with inflammatory markers, directly downregulate/dysregulate ACE2, which leads to imbalance of ACE2/ANG II toward the disease state of elevated ANG II and aldosterone levels, further worsening conditions (52, 86). Novel therapeutics in study, e.g., camostat mesylate, nanobodies, decoy receptors, aim to disrupt S protein to inhibit viral entry into the host ACE2 (23, 87–89). Early phase 2 investigations of human recombinant soluble ACE2 antibodies (NCT04335136) were seen to reduce SARS-CoV-2 viral loads in infected Vero-E6 cells by a factor of 1,000–5,000 and inhibit viral infections of kidney and vascular organoids, potentially decreasing direct organotropism and disease progression (90).

PRIMARY RISK FACTORS

Primary comorbidities increase COVID-19 susceptibility and severity. Reports have indicated that most COVID-19 patients have more than comorbidities; of these, ACE2 activity has been widely studied in experimental and clinical trials (Fig. 3).

Figure 3.

Respiratory disease: angiotensin-converting enzyme 2 (ACE2) receptors in the lower airways, most prominently in alveolar type II and epithelial cells (7). Although ACE2 in the lungs is lower compared with nasopharyngeal mucosa and other organs, ACE2 receptors are not evenly distributed throughout the lungs, which may be perceived as decreased ACE2 expression in immunohistological stains (19). ACE2 prevents prolonged increased ANG II production, which triggers pulmonary edema and acute respiratory distress syndrome (49). Knockout mice models for ACE2 led to severe lung injury when mice contracted H5N1, but treating knockout mice with rhACE2 decreased injury (91). Cardiovascular disease: ACE2 receptors localized in cardiac myocytes and intramyocardial vessels extending into the aortic intima. Elevated ACE2 metabolizes ANG II, a critical inotrope and growth factor for remodeling the cardiac extracellular matrix. Knockout mice illustrate that ACE2 loss results in early hypertrophy, accelerated myocardial infarction, fibrosis, and dilated cardiomyopathy from oxidative stress, pathologic hypertrophy, increased neutrophilic infiltration, and inflammatory cytokines INF-γ, IL-6, and the chemokine monocyte chemoattractant protein-1 (10, 12, 92, 93). Conversely, overexpression of ACE2/ANG-1–7 significantly reduces deleterious myocardial infarction-induced cardiac remodeling (94, 95). Hypertension: Experimental models have solidified ACE2 as a protector against hypertension, while deficiency exacerbates hypertension, defining the enzyme’s essential role for maintaining healthy blood pressure (96–98). Models further illustrated that rhACE2 prevents hypertension by reducing plasma ANG II while increasing plasma ANG-1–7 levels (99); rhACE2 also has an established record for treating pulmonary arterial hypertension (NCT01597635 and NCT03177603) (100, 101). Renal disease: expressed predominantly in the proximal tubule, endothelial, podocytes, and smooth muscle cells of renal vessels (102, 103). Experimental animal models propose the importance of ACE2 in regulation of renal diseases to prevent injury and fibrosis, e.g., ACE2-deficient mice have been reported to increase age-related glomerulosclerosis (104). Diabetes mellitus: many organs involved in controlling blood sugar are rich in ACE2 (105). Although ACE2’s function here is unknown, it is implicated to cause β-cell proliferation and insulin secretion by decreasing islet fibrosis, possibly reducing type 2 diabetes (T2D) onset (106). Obesity: ACE2 expression was found to be higher in human subcutaneous adipose tissue and human visceral adipose tissue (107). ACE2 expresses potent anti-inflammatory effects in adipose tissue of obese, as seen in T2D mice (108). Gastrointestinal disease: presence of ACE2 was found in intestinal glandular cells, as well as gastric, duodenal and rectal epithelial cells. ACE2 may regulate homeostasis of intestinal amino acids, expression of antimicrobial peptides, and ecology of gut (109–111). Cerebrovascular disease: ACE2 receptors are nonspecifically located in brain tissue but more prominently found in brain vasculature. ACE2 was observed to have beneficial effects on neurogenic blood pressure, stress response, anxiety, cognition, brain injury, and neurogenesis (32, 112). PT, prothrombin time; aPTT, activated partial thromboplastin time. [Figure reproduced with permission from Vabret et al. (46).]

Respiratory Disease

Strong evidence pertains to chronic obstructive pulmonary disease (COPD) and emphysema associated with increased risk of COVID-19 susceptibility and severity (1, 6, 113, 114). Meta-analysis of preexisting respiratory disease between severe and nonsevere COVID-19 patients was odds ratio (OR): 2.46 (1.76–3.44) (6), although mixed evidence is analyzed in asthmatics for risk and severity (1, 115–117). School reopenings are shifting the focus toward safety of young asthmatic individuals (115, 118), with reports of 27% of hospitalized COVID-19 cases in young adults (119). In addition, smoking was associated with increased severity, intensive care unit (ICU) admission, and death in hospitalized COVID-19 patients [OR: 2.0(1.3–3.1) – 2.2(1.3–3.7)] (120–124). Although likely related to severity, evidence to quantify the risk to smokers is unavailable (122).

RAS activity is intrinsically high in the lungs. Histopathological reports acknowledge direct viral toxicity, resulting in atypia and detachment of type II pneumocytes, hyaline membrane formation, interstitial inflammatory response, and endothelial dysfunction (33, 35, 37, 38, 125). A pathophysiological timeline of 65 cases stressed direct lung damage from viral organotropism during the first week but later transitioned to host inflammatory and hypercoagulable responses 10–28 days into the disease phase (125). Reports show mainly bronchitis and pneumonitis in mild/moderate individuals (31) and pneumonia, acute respiratory distress syndrome, and pulmonary embolism in severe/critical patients (125).

ACE2 elevation exists in acute and chronic lung disease to prevent lung injury. A study on the previous SARS-CoV virus revealed it in autopsy specimens from severe SARS patients with ALI elevated ACE2, SARS‐CoV S protein, RNA, and proinflammatory cytokines (126). Chronic lung disease studies have pursued ACE2’s role in COPD, emphysema, asthma, and other ailments.

Recent literature provides overwhelming evidence that COPD upregulates ACE2 and TMPRSS2 expression in the nasal, bronchial and lower airways (127–134). Significant inverse relationships between ACE2 gene expression and predicted forced expiratory volume for 1 s (r = −[0.24 – 0.40]; P < 0.05) were reported (127, 129), and higher ACE2 mRNA and protein levels in lung tissue were seen in moderate to severe COPD (128). Toru et al. (135) discovered significant serum ACE2 level increases in 27 COPD patients. However, reductions in endothelin-1, which downregulate ACE2, were found during exacerbations of COPD, suggesting that ACE2 dysregulation exacerbates disease potentiation (136, 137). Although the mechanisms of ACE2 upregulation in COPD patients is unknown, potential gene regulators related to histone modifications, e.g., HAT, HDAC2, and KDM5B, were correlated to ACE2 expression (131).

Moreover, inflammatory cytokines have proved to be key mediators. Studies have found that TNF‐α concentrations increased in COPD patients’ sputum and plasma (138). TNF‐α, IL‐12, and IL‐17A were found to upregulate ACE2 expression in BEAS‐2B cells derived from lung tissue (134). Interestingly, the preprint of Finney et al. (139) reported that inhaled corticosteroid therapy reduced ACE2 pulmonary expression through suppression of type I interferon in in vitro and in vivo COPD models and thus may reduce COVID-19 susceptibility.

However, mixed and limited evidence supports ACE2’s role in asthma pathophysiology, which parallels the divided consensus on asthma as a significant COVID-19 comorbidity. In fact, patients with COPD had higher frequencies of severe cases than asthmatics (57.1% vs. 4.6%, P < 0.01) (134). Saheb Sharif-Askari et al. (133) found ACE2 and TMPRSS2 lung airway expression to be upregulated slightly, while plasma ACE2 was significantly upregulated in asthmatics. Radzikowska et al. (132) similarly detected elevated TMPRSS2 expression in children and adult asthmatics from available RNA-Seq databases. ACE2, correlated gene signatures were found to be significant in a subset of type 2, low patients with asthma with characteristics resembling known risk factors for severe COVID-19 (140). However, other recent studies concluded no significant differences or reductions in ACE2 expression in the lower airways with allergic sensitization and asthma, suggesting additional factors beyond ACE2 modulation in the response to COVID-19 in individuals with allergies (130, 134, 141). Heterogeneity of asthma endotypes, type 2 (allergic) versus nontype 2 (nonallergic), formulate possible confounders; IgE and blood eosinophils have been shown to lower ACE2 expression, potentially regulating ACE2 in type 2 inflammation (134, 141). Additionally, hypoxia is known to regulate ACE2 expression (142, 143), which could explain significant ACE2 expression in COPD patients and not asthmatics.

Additionally, studies consistently correlate tobacco use to an increase in ACE2 and TMPRSS2 gene expressions in bronchial and alveolar epithelial, bronchial alveolar lavage, and protein in blood and lung tissue, which may intensify COVID-19 severity in smokers (127–134, 144). Current smokers had significantly higher gene expression than ex-smokers and nonsmokers (2.77 ± 0.91 vs. 2.00 ± 1.23 vs. 1.78 ± 0.39, respectively; P = 0.024) (128, 129). The additive factor of smoking in COPD patients caused them to have higher ACE2 expression, compared with nonsmoking COPD patients, suggesting that smoking may emulate a drastic modifying factor rather than a pathological disease. Therefore, hypoxia from smoking may contribute to elevated ACE2. These studies suggest that smokers are at an increased risk for COVID-19 susceptibility and severity, although evidence associates smoking only with increased severity.

In sum, ACE2 significantly contributes to lung injury repair in acute and chronic lung disease. This potentially increases COPD patients’ and smokers’ risk and susceptibility to severe/critical COVID-19, although evidence for asthma is less clear.

Cardiovascular Disease

Individuals with serious cardiac conditions, e.g., heart failure, coronary artery disease, cardiomyopathies, have the strongest, most consistent evidence for increased COVID-19 susceptibility and severity (1, 6, 124, 145). Interestingly, a large retrospective study (n = 144,279) from the United Kingdom found no difference between preexisting ischemic heart diseases in COVID-19 (11.4%) and non-COVID-19 (12%) deaths (146).

Mounting evidence from pathological reports observed viral myocarditis from SARS-CoV-2 direct cardiotoxicity as the presumed etiology of primary cardiac injury (9, 147, 148), while hyperinflammation is another mechanism (32, 37, 149). Reports indicate myocardial injury, cardiomyopathy, acute coronary syndrome, cor pulmonale, arrhythmias, and cardiogenic shock (32). In China, acute and chronic cardiovascular damage were seen in nearly 20% of 416 patients, arrhythmias in 44% of ICU patients, and significant higher risk of all-cause mortality in hospitalized patients (150–153). Studies in New York City found that 6% of 4,250 patients have prolonged QTc at admission and atrial arrhythmias were in critical patients but not noncritical ones (17.7% vs. 1.9%) (4, 154). Moreover, echocardiography findings from 69 countries revealed heart damage in over half of COVID-19 patients (155). Postcomplications were noted from cardiovascular magnetic resonance imaging of ongoing cardiac involvement and myocardial inflammation in 78% and 60%, respectively, of German patients independent of comorbidities and severity (156).

Although the pathophysiology for cardiovascular manifestation is probably multifactorial, ACE2 abounds in cardiac tissue, suggesting a mechanism of direct SARS-CoV-2 infestation (157, 158). Individuals with underlying cardiovascular disease have upregulated RAS as a compensatory mechanism to maintain cardiac output mainly through the elevated ANG II pathophysiology (Fig. 3). Therefore, ACE2-generating cardiac diseases may sensitize the myocardium to elevated SARS-CoV-2 entry and viral replication, leading to ACE2 shedding. This would cause the myocardium to lose protective effects of ANG-1–7, causing inflammation, reactive oxygen species, and vasoconstriction for cardiac damage (159, 160).

Clinical and experimental studies have demonstrated relations between cardiovascular disease and ACE2. Elevated serum ACE2 was found within animal models of myocardial infarction, atherosclerotic development, reduced left ventricular ejection fraction, cardiomyopathies, and heart failure (12, 92). Burrell et al. (157) found that myocardial infarction significantly increased ACE2 mRNA in day 3 and 28 postmyocardial infarction for rats, which paralleled explanted failing hearts in human subjects. ACE2 was the most upregulated gene, and a fivefold increase in ACE2 protein was found in hypertrophic cardiomyopathy human cardiac tissue, compared with that of controls (159). Plasma ACE2 activity directly related to persistent atrial fibrillation (AF; 22.8 pmol/min/mL) and paroxysmal AF (16.9 pmol/min/mL), compared with control (13.3 pmol/min/mL) (161). Also, a study of 79 obstructive coronary artery disease patients revealed that they had significantly elevated plasma ACE2, which correlated with increased adverse long-term cardiovascular outcomes (162).

Moreover, numerous studies confirmed ACE2’s elevation in heart failure. Recent studies measured elevated plasma ACE2 concentrations in 1,485 men and 537 women with heart failure and a threefold increase in myocardial ACE2 gene expression in heart-failure patients (163, 164). Another study showed near-doubling of ACE2 activity in acute heart-failure patients, compared with healthy controls: 52.5 pmol/h/ml versus 22.5 pmol/h/ml, respectively. Chronic heart-failure patients also had increased ACE2 levels (33.6 pmol/h/ml) (165). Other studies demonstrate that soluble ACE2 activity positively correlates with heart failure severity (166, 167). Novel drugs targeting ACE2 are being studied in heart failure patients and are shown to attenuate effects of systolic and diastolic dysfunction (168). Ongoing clinical trials are investigating modulation of ACE2/ANG-1–7 balance with rhACE2 (NCT00886353) and cardiac progenitor cells (NCT02348515) (11).

These findings suggest that elevated ACE2 in preexisting cardiovascular disease may increase susceptibility and severity to SARS-CoV-2.

Hypertension

This is a notable risk factor for increased morbidity and mortality from COVID-19 (1, 6, 124), but whether hypertensive individuals are more susceptible to infection is unclear (169). Hypertension prevalence in COVID-19 patients seen in a meta-analysis denotes 21.1% (13.0–27.2) (6), in severe COVID-19 [OR: 2.36(1.46–3.83) – 2.72(1.60–4.64)] (6, 124). Furthermore, a hazard ratio (HR): 1.7–3.05 was reported for mortality in COVID-19 patients with hypertension (44, 45). SARS-CoV-2 predisposition may be due to ACE2 polymorphisms in individuals with hypertension (170).

Experimental models have solidified ACE2 as a protector against hypertension (Fig. 3). Hypertension’s integrative role with other organs affects RAS systemically and locally, which is activated by distinct signals from differing physiologic and pathophysiologic conditions (171, 172). Essential hypertension can be delineated into two groups based on plasma renin activity: those with low activity, and those with normal-to-high activity (173). Studies show that hypertension with high plasma renin activity relates to increased cardiovascular complications and vascular damage (173–175). The deleterious effects may destroy ACE2 mRNA and protein activity in tissue, as hypertensive clinical and experimental models show, primarily in the kidneys and heart and thus depict hypertensive progression decreasing local ACE2 activity (97, 176–180). We believe that organ damage from hypertension causes local destruction of ACE2 within the targeted organ. An autopsy study of 20 patients diagnosed with hypertensive cardiomyopathy or nephropathy revealed decreased local ACE2 expression (177).

However, human and animal models positively correlate plasma ACE2 activity with increased systolic blood pressure (167, 181–187). The Leeds Family study detected significantly higher systolic and diastolic blood pressure, compared with those without detectable plasma ACE2 (182). The study also provided evidence of genetic effects on plasma ACE2, estimating that hereditable factors influenced 65% of variability in ACE2 levels. Meta-analysis revealed possible association of ACE2 G8790A and rs2106809 polymorphisms with essential hypertension risk (188). Furthermore, Li et al. (181) determined higher serum ACE2 concentrations in hypertensive patients than in healthy subjects (170.31 [83.50–707.12]p g/ml vs. 59.28 [39.71–81.81] pg/ml, respectively; P < 0.001). Another study revealed that ACE2 activity was 1.5 times greater in 239 hypertensive patients than in healthy volunteers (167).

We believe that shedding from endothelial cells release the catalytically active ectodomain ACE2 as a soluble form in plasma (12, 186) and that compensatory release of ACE2 from other organs provides appropriate systemic response to regulate homeostatic changes similar to the anti-inflammatory and antifibrotic ones observed in rhACE2 that increased systemic and/or local ACE2 levels (189–192). Other studies have reported no association between plasma ACE2 and hypertension (193, 194). Whether SARS-CoV-2 utilizes plasma ACE2 as an active domain to replicate is unclear, but ACE2 shedding amplifies endothelial dysfunction and hyperinflammation, thus potentially increasing COVID-19 severity.

However, mixed results were noted in the relationship between secondary hypertension and ACE2 (178). Variability in experimental models may relate to acute and chronic models, mechanism of the type of secondary hypertension, and compensatory effects of other organs. RAS and ACE2 activity is notable in neurogenic hypertension, which is characterized by an increase in sympathetic activity and often resistance to drug treatments. The rostral ventrolateral medulla reportedly stimulates sympathetic preganglionic neurons, which activate ACE2 to regulate blood pressure. ACE2’s role downregulates ANG II-mediated presser and fluid retention effects (195). Loss of compensatory activity during neurogenic hypertension was seen in 27 patients with increased ACE2 activity in the cerebrospinal fluid, correlating with systolic blood pressure. Moreover, increased TNF-α was found, suggesting the upregulation of ADAM17 (186).

Experimental studies and pathological reports clearly find a decrease in local tissue ACE2 activity, which may relate to progressive organ damage from hypertension. However, strong evidence suggests associations between hypertension and elevated plasma ACE2 and upregulation of ADAM17, which may increase COVID-19 severity in hypertensive patients. Large clinical studies are needed to clarify ACE2’s role in essential and secondary hypertension, acute and chronic hypertension, and hypertensive disease progression. Determination of plasma renin activity in both hypertensive and nonhypertensive COVID-19 patients may also indicate prognosis and treatment decisions (173).

Renal Disease

Chronic kidney disease (CKD) has strong, consistent evidence for increased COVID-19 susceptibility and severity (1, 196, 197). Meta-analysis of 25 studies indicated CKD to have the most significant relative risk of mortality for COVID-19, 3.25(1.13–9.28) (197). Genetic risks are amplified for infected individuals with APOL1 gene variant, occurring in 14% of Blacks, causing deadly viral-induced collapsing focal segmental glomerulosclerosis (198).

The kidney is reportedly a target for SARS-CoV-2, which replicates there in almost 30% of infected patients (199). Increased renal disease prevalence before admission and acute kidney injury development was seen during hospitalization (200). Histopathological reports from postmortem COVID-19 patients indicated direct virulence in the kidneys (9, 36). Light microscopy observed proximal tubule injury with the loss of brush border, vasculitis, nonisometric vacuolar degeneration, and necrosis, indicating the combined effects of viral toxicity, thrombosis, and cytokine storm (9, 32, 36, 64).

Abundant ACE2 in the kidneys (Fig. 3) threatens to be a culprit for risks of susceptibility and severe/critical COVID-19. Overwhelming evidence of experimental and human models has been reported in local tissue reduction in ACE2 mRNA and protein activity in glomerular and tubular pathology in primary glomerulopathy, IgA nephropathy, hypertension, nephrosclerosis, and nephrectomy (177, 179, 201–206). Conversely, Lely et al. (103) detected ACE2 increases in glomeruli and tubules in primary and secondary renal disease and renal transplant patients. However, only eight patients were observed, and the variation in ACE2 expression markers among tissue samples from diabetic nephropathy patients was not considered.

The pathophysiological process of renal disease damages glomerular and tubulointerstitial tissue, causing ACE2 loss locally. However, diabetic nephropathy provides mixed observation within studies. Most experimental and clinical studies find significant decreases in glomerular ACE2 and increases in tubular ACE2 (207–210), although an early animal study showed that ACE2 protein levels decreased ∼30% in glomerular tissue in diabetic kidneys (211). Diabetic nephropathy may progress slower than other renal diseases, partly due to disease process and early aggressive treatments. Thus the damage is seen primarily at the glomerulus. As diabetic nephropathy becomes more severe, however, it may also damage the tubulointerstitium and hence decrease ACE2 in tubular tissue (203, 204).

Although local renal ACE2 production was decreased, plasma ACE2 was significantly elevated, many studies show (15, 92, 183, 184, 187, 212–217). Experiments demonstrated a twofold increase in circulating ACE2 in diabetic nephrotic mice (15). Clinical studies also support elevated circulating ACE2 in diabetic and renal disease patients. Roberts et al. (183) found increased plasma ACE2 in CKD patients. Increased levels were further supported by predialysis CKD stages of three to five patients (n = 1,456), compared with patients on dialysis (n = 546) (212, 213). Furthermore, a study of 859 type 1 diabetes patients found serum ACE2 increases in diabetic individuals with micro- and macro-albuminuria, which negatively correlated with glomerular filtration rates (184). As mentioned, shedding and systemic compensatory production may elevate serum ACE2 during development of renal disease.

In sum, major evidence correlates renal disease, reduced local ACE2, and elevated plasma ACE2.

Endocrinological Disorders

Type II diabetes mellitus (T2D) and/or obesity (body mass index ≥ 30) patients are evidently at high risk of COVID-19 susceptibility and severity (1, 4, 218–222), while evidence for risk in type I diabetes mellitus (T1D) is limited (1, 4, 218, 220). However, recent data in England’s death registry showed that T1D had 3.5 times the risk of COVID-19 in-hospital death versus two times in T2D (218, 223). Studies had strong evidence for the relationship of T2D and severe COVID-19 [OR: 2.75(2.09–3.62)] and mortality [OR: 1.90(1.37–2.64)] (224). Complications include new-onset diabetes, hyperglycemia, preexisting diabetes, euglycemic ketosis, hyperosmolarity, and classic diabetic ketoacidosis (225). These manifestations pose challenges in clinical management with glucose-lowering medications, due to ineffective results in diabetic COVID-19 patients (219).

Many blood sugar controlling organs are rich in ACE2 (105). Aside from diabetic nephropathy modulating ACE2 expression as mentioned, pancreatic islet cells reportedly have elevated enzyme expression and have been postulated as a target for SARS-CoV-2 infection (226–228). Thus pancreatic viral toxicity potentially explains for new-onset diabetes or worsening metabolic control in patients with diabetes (225). Bindom et al. (229) found that islet ACE2 expression increases early in the disease course and decreases with disease progression in T2D mice. A similar trend is noted with β-cells as with renal ACE2 production in renal disease: damage of local tissue from the disease seemingly prevents local ACE2 production, although systemic ACE2 is upregulated.

Based on one of the largest genome-wide studies on T2D to date (n = 898,130), T2D was causally linked to raised ACE2 expression (P = 2.91E-03; Mendelian randomization-inverse‐variance weighted) (230). IL-6 and INS genes were also associated with diabetic patients in 700 lung transcriptome samples (131). INS gene encodes the insulin hormone, and insulin is associated with the NAD-dependent histone deacetylase sirtuin-1, which reportedly regulates ACE2 (142). Animal models in diabetic (STZ induction) mice observed elevated serum ACE2 levels (15, 187). Although evidence for clinical models is limited for circulating ACE2 levels in diabetic patients, Soro-Paavonen et al. (184) noted that ACE2 activity was increased in men with T1D and microalbuminuria, compared with patients without albuminuria or controls: 30.2 ± 1.5 versus 27.0 ± 0.5 versus 25.6 ± 0.8 ngE/ml, respectively; P < 0.05. Furthermore, a preprint of a single center population-based study of 5,457 Icelanders found significant associations of elevated serum ACE2 levels in smokers and obese or diabetic individuals (231). Elevated ACE2 is likely caused from shedding, as seen in increased urinary ACE2 and ADAM17 in 40 T2D patients (232), as well as possible compensatory ACE2 mechanism in vascular and renal function regulation.

One study determined the pathogenesis of increased glucose to directly increase viral load, ACE2, and IL-1β expression in SARS-CoV-2-infected monocytes in a dose-dependent manner. Subsequent treatment with glycolysis inhibitors completely inhibited viral replication in infected monocytes and decreased ACE2 and IL-1B expression (233). In addition, HIF-1 was a strong inducer in glycolysis, also involved in inflammatory response and endothelial dysfunction, suggesting elevated glucose as a catalyst for severe COVID-19 (233).

While pathophysiology in increased risk of worse outcomes in diabetics is most likely multifactorial, obesity has strong associations to T2D and increased risk to COVID-19 susceptibility and severity (1, 42, 221, 222, 234). Of 257 critically ill patients hospitalized in New York City, 36% were diabetic, 46% obese (40). Pooled analysis from 75 studies found that obese individuals were more at risk for COVID-19, 46.0% [OR: 1.46(1.30–1.65)]; for hospitalization, 113% [OR: 2.13(1.74–2.60)]; for ICU admission, 74% [OR: 1.74(1.46–2.08)]; and for mortality, 48% increase in deaths [OR: 1.48(1.22–1.80)] (222).

ACE2 expression was found to be higher in human subcutaneous adipose tissue and human visceral adipose tissue than in human lung tissue, suggesting that adipose tissue may be more vulnerable to COVID-19 (107). Strong evidence in mice models observed increased ACE2 mRNA expression, protein, and circulating ACE2 levels (235–239). Moreover, augmented cardiac ACE2 was detected in lean and obese mice (240). No recent published clinical studies were examined for ACE2 and obesity; thus they need to be for future evidence.

Possible physiology includes viral toxicity on adipocytes. The prior SARS‐CoV was found to use cholesterol to facilitate viral budding following S-protein binding of cellular ACE2 receptors. Depletion of cholesterol in ACE2-expressing cells resulted in markedly reduced viral S-protein binding (241). Viral organotropism may explain elevated aspartate aminotransferase concentrations and poorer prognosis in nonalcoholic fatty liver disease and nonalcoholic steatohepatitis in COVID-19 patients (242). After viral binding, deactivation of ACE2 suggests increased macrophage polarization to proinflammatory response such as TNFα, IL-6, IL-8, leptin, and adiponectin, as seen in epicardial adipose tissue, suggesting aggravations to cytokine storm (32, 242–244).

In sum, elevated glucose levels and increased adipose tissue evidently correlate to increased COVID-19 susceptibility and severity by contributing to both increased viral replication and cytokine production. Limited studies of elevated ACE2 and adipose tissues demand further investigation of correlation between obesity and ACE2 levels.

Other Manifestations

Other comorbidities, e.g., gastrointestinal diseases, cerebrovascular disease, cancer, may affect COVID-19 infection. Although gastrointestinal diseases, e.g., inflammatory bowel disease, irritable bowel syndrome, peptic ulcer disease, gastroparesis, have not been readily studied as risk factors for COVID-19, more than 20% of COVID-19 patients suffer from gastrointestinal symptoms (111). A U.S. multicenter study found that almost two-thirds of patients hospitalized with COVID-19 presented at least one gastrointestinal symptom (anorexia, 34.8%; diarrhea, 33.7%; nausea, 26.4%; vomiting, 15.4%) (245). Studies suggested fecal-oral transmission, as reported in a meta-analysis detecting viral mRNA in feces in 40.5% (27.4–55.1) of patients (246). Pathophysiology of gastrointestinal infiltration in COVID-19, presumably multifactorial, direct viral mediated ACE2 damage, may contribute to gut dysbiosis. More studies are needed to clarify the relationship of SARS-CoV-2 to gastrointestinal pathophysiology.

Mixed evidence exists for patients with cerebrovascular disease to have an increased risk in severity and mortality, with limited evidence of COVID-19 susceptibility (197, 247, 248). Cerebrovascular disease was associated with increased mortality with borderline significance 2.04(1.43,2.91) – 2.16(0.97,4.80) (197, 248). ACE2 expression has been found in the neurovascular system, suggesting that viral aggravation causes nonspecific neurological symptoms in up to 40% of patients, including severer presentations such as acute stroke and acute necrotizing encephalopathy (32). In combination with direct viral toxicity, ACE2 also contributes to endothelial dysfunction and inflammatory responses, causing meningoencephalitis and acute stroke of varying arterial and venous mechanisms. Limited evidence supports a relevant role for ACE2 in several neuropsychiatry conditions, cerebrovascular ischemic and hemorrhagic lesions, and neurodegenerative diseases (249). A mouse model indicated that cerebral ischemic lesions increased regional cerebral and circulating ANG-1–7 at 12 h, compared with control (7.276 ± 0.320 vs. 2.466 ± 0.410 ng/mg, serum; 1.024 ± 0.056 vs. 0.499 ± 0.032, brain; P < 0.05) (250). In addition, ACE2 expression increased in the cortex penumbra in pathological reports of rats after ischemic injuries and smoking (251).

Further investigations are needed for ACE2’s role in gastrointestinal diseases, cerebrovascular diseases, and other comorbidities, e.g., cancer, sickle-cell disease, solid organ transplantation, that present consistent evidence of increased COVID-19 susceptibility and severity (1).

SECONDARY RISK FACTORS

Data from the WHO, CDC, and other health organizations uncover a bias in infections toward elderly males. Trends also indicate that COVID-19 has affected certain races, ethnicities, and countries more than others, suggesting that genetics may contribute to infection.

Age

Much is debated regarding increased susceptibility and age. In the early phase of the outbreak, COVID-19 appeared to occur in older people in most world regions (242, 252). A meta-analysis of 32 studies suggests that susceptibility in children/adolescents is half that of adults (253). However, other data indicate that young adults and children are as susceptible to the disease as older adults (254, 255), possibly for social reasons, making optimal strategy for reopening schools and universities difficult. In fact, significantly greater amounts of viral nucleic acid were detected in children (<5 yr; n = 46) than in adults (18–65 yr; n = 48) (256). While the infection prevalence’s relation to age is unclear, increased risk for severe illness evidently increases with age, making elders a high-risk population (4, 5, 257–259).

The CDC reports that 8 of 10 COVID-19 deaths in the U.S. are over age 65 (260). Case fatality rates by age group were observed in Italy and China: 96.5% of total deaths in Italy and 81% of those in China were past age 60 (258); 0.3% of total deaths in Italy and 2.6% of those in China were below age 39. A study of critically ill patients in China found that, compared with survivors, nonsurvivors were older (64.6 ± 11.2 yr vs. 51.9 ± 12.9 yr) and had comorbidities (259). Moreover, in-patient mortality rates in New York City were higher in those older than 65 versus those aged 18–65 (4). Although data suggest that the elderly develop more complications and young adults will less likely become symptomatic (261), a case series highlights the presence of genetic variants in young men with severe COVID-19. Rare putative loss of function TLR7 variants were associated with impaired IFN responses (262).

In earlier reports, ACE2 receptor availability was thought to decrease in the elderly in most tissues, increasing their risk of severe illness. Since ACE2 is a key regulator for inflammatory and immune responses, ACE2 reduction could result from a weakened immune system, increasing COVID-19 infection vulnerability. Experimental models demonstrate significantly lower ACE2 levels in older 24-mo-old male and female mice and aging endothelial cells, suggesting that ACE2 receptors in older adults could be decreased in the respiratory tract, possibly weakening the immune system (263–265), although other animal models depict no significant difference in ACE2 expression in cerebrovascular tissue (266, 267). More recently, RNA-Seq gene profiling in 30 tissues across thousands of individuals found ACE2 expression to significantly decrease with age in Caucasian males, although data remained nonsignificant in other ethnic groups (268). However, other analyses of gene expression profiling repository data found no substantial differences in ACE2 and TMPRSS2 with ages >60 yr versus <60 yr and ≤49 yr versus >49 yr (269, 270). Moreover, a clinical study found no statistical differences in activity of ACE/ACE2 among four age groups ranging from neonates to seniors (271), although the study had limited samples of 17–29 individuals in each group. Recent studies suggest otherwise.

Mouse models revealed that elderly mice had higher ACE2 and TMPRSS2 expression in nasal mucosa than younger ones (272, 273). Similarly, clinical models found significant linear trends in nasal epithelial ACE2 gene expression with advancing age groups and olfactory and gustatory deficits in a cohort of 305 individuals aged 4–60, indicating that children could have milder COVID-19 symptoms (274, 275). Though limited in age, nasopharyngeal, oropharyngeal, and/or blood specimens were collected in 192 children aged 0–22. The study revealed that age did not impact viral load, but older children (>10 yr) had significantly higher ACE2 expression than younger ones. Furthermore, a positively weak correlation (r = 0.20, P = 0.02) between ACE2 expression and age suggests that future studies must clarify a plausible linear relationship (276).

Likewise, Saheb Sharif-Askari et al. (133) observed significantly elevated ACE2 and TMPRSS2 in adults compared with children in nasal and bronchial tissue in multiple transcriptomic datasets. Increased expression of ACE2 and TMPRSS2 in upper and lower airways of adults may contribute to increased severity of infection in adults. Children may hypothetically be protected from severe COVID-19 due to cross-protective antibodies and/or decreased ACE2 receptors to SARS-CoV-2 in the lower airway (277), possibly because their lungs do not fully mature until age 20–25. Moreover, novel findings of ACE2 expression in gastrointestinal tissue moderately correlate with age. Findings have evaluated ACE2 mRNA in duodenal and ileal biopsies, which correlated with age (r = 0.32, P = 0.0099; r = 0.64, P = 0.0099, respectively) (278, 279). The different ACE2-associated amino acid transporter (B0AT1, SIT1) expression at the brush borders in older patients may impact susceptibility to intestinal symptoms and/or increased disease severity (279).

In addition, age differences in serum ACE2 may provide insights for COVID-19. Studies of 118 healthy individuals aged 41–70 and 213 patients with newly diagnosed mild-to-moderate hypertension determined a positive association between age and ACE2 serum activity (280, 281). Moreover, a large cohort of participants (n = 2,051) in a commercial wellness program found significantly higher plasma ACE2 levels in older individuals; age association was more pronounced in women pre- and postmenopausal (282). A recent longitudinal study observed low serum ACE2 in males and females up to age 12 and significantly increased serum in adolescents and young adults, implying that androgen sensitivity in puberty may exacerbate ACE2 production (283).

Although increased circulating ACE2 enzyme offers protection against influenza A (H7N9) virus-induced ALI (284), the apparent paradox for older individuals to have elevated soluble ACE2 levels potentially heightens COVID-19 susceptibility and/or severity in older adults. Although concerns present for direct viral organ-toxicity and ACE2 dysregulation, age-associated frailty involves increased baseline inflammation, called “inflammaging,” which can cause exuberant inflammatory responses in older individuals with high baseline IL-6 and IL-8 (285). For example, the preprint of Baker et al. (286) detected increased ACE2 expression with age in the setting of alveolar damage observed in patients on mechanical ventilation, providing a potential mechanism of exacerbated inflammatory responses for increased COVID-19 mortality in the elderly.

Sex

Despite similar sex distribution of individuals infected with SARS-CoV-2 (male 51%, female 49%), a sex difference is notable in COVID-19 fatality rates: males could be at higher risk than females for contracting severe COVID-19 (6). China’s CDC confirmed increased case fatality rates in males compared with females (2.8% vs. 1.7%, respectively), revealing that 64% of deaths in China have been male (287). Italy had similar trends: male mortality is apparently twice that of females in every age group. Italy’s Public Health Research Agency noted that 59.8% of SARS-CoV-2 cases and 70% of national deaths so far have been male (288). New York City had increased hospitalized males more than females (60.3% vs. 39.7%, respectively; n = 5,700) (4). Moreover, Open SAFELY, an analytics platform covering over 17 million records in England, revealed that males have over one-half the mortality risk of females, HR: 1.59(1.53–1.65) (145). Reduced SARS-CoV-2 infection in females could be attributed to their increased protection from viral infections by an additional X chromosome and varied sex hormones (289). The ACE2 gene, which lies on Xp22.2, is affected by sex hormonal regulations.

Mice models show ACE2 activity to be higher in males than females in specific tissue and serum. Studies mainly revealed elevated local ACE2 activity in male kidneys, compared with female ones (290–293). Conversely, Sampson et al. (294) noted increased ACE2 receptor expression within female kidneys. Likewise, serum ACE2 levels were elevated in male normotensive mice (293, 295). However, levels were contrary in hypertensive mice and comparable to those of hypertensive patients (187, 293, 296), suggesting comorbidity-dependent hormonal modulation on ACE2 gene activity.

Although animal models provided increased evidence supporting elevated ACE2 activity in renal tissues, human studies indicated mixed results in large transcriptome public databases. Bulk RNA-seq profiles determined increased ACE2 expression in male lung tissue, largely in type II pneumocytes, and male plasma (297–299). Other transcriptome studies revealed comparable ACE2 and TMPRSS2 concentrations in multiple tissue (270, 300), while Chen et al. (268) revealed Asian females to have higher ACE2 expression than males out of three studied ethnic groups, suggesting ACE2 regulation through genetic differences. Limited independent human tissue studies and recent plasma studies are nonetheless evident for males exhibiting significantly higher ACE2 expression levels. Male bronchial biopsy and bronchoalveolar lavage generally led to increased expression of ACE2 and ACE2-related genes in smokers and nonsmokers (132, 278). ACE2 concentrations were reported higher in male smokers than in male nonsmokers, implicating gender-specific behavioral differences (299). Furthermore, ACE2 expression in healthy ileal biopsies (n = 154) were found to be 130% greater in men than in women (P = 0.0256) (278).

Recent serum studies indicate strong evidence of increased circulating ACE2 levels in healthy males and males with comorbidities (132, 164, 217, 280, 282, 283, 301). Studies are revealing comorbidities, e.g., endocrine manifestations, hypertension, heart failure, end-stage renal disease, exacerbating increased serum ACE2 in men (164, 217, 282, 301). Sama et al. (164) observed male sex as the strongest predictor of elevated concentrations of ACE2 in control and heart failure cohorts (coefficient = 0.19 and 0.26, respectively; P < 0.001), and Stienen et al. (301) detected lower ACE2 levels in females with preserved ejection fraction heart failure. These effects suggest that gender-specific factors and behaviors may potentiate ACE2 receptors and thus demand investigation.

In contrast, no sex differences were detected for ACE2 in 118 healthy men and women, possibly because ACE2 activity in females was affected by postmenopausal period (281). Kornilov et al. (282) confirmed higher levels of plasma ACE2 in postmenopausal women than in premenopausal women (P = 0.02). A longitudinal study found similar low-serum ACE2 in both sexes until age 12, where ACE2 increased more in boys than in girls, emphasizing possible sex hormonal regulation (283). In fact, mechanisms for sex hormonal regulation of ACE2 and TMPRSS2 remain controversial with limited data.

ACE2 expression may be upregulated in females by estrogen [i.e., 17β‐estradiol (E2)], X chromosome inactivation, or reduced ACE2 methylation, providing increased ACE2 levels to maintain RAS equilibrium (302). Experimental models have observed estrogen upregulating ACE2, AT2R, and MAS expression levels through effects at estrogen receptor-mediated binding at the ACE2 promoter (303, 304). Yet, low androgen levels and elevated E2 levels in females may suppress ACE2 expression, providing a protective factor against COVID-19 (302). The only known stimuli of TMPRSS2 gene transcription are androgens, and hospitalized COVID-19 patients exhibited androgenic alopecia (305). E2-treated NHBE cells expressed lower levels of ACE2 mRNA (306), while ovariectomy (loss of E2) increased ACE2 activity, and orchiectomy (loss of androgens) decreased enzyme activity (292, 307, 308). Further studies correlating sex hormones to ACE2 are needed.

Other factors, e.g., population composition, sex-related comorbidities, immunological responses, must be considered. More males than females appear affected in China and Italy, as mentioned, though males comprise a larger percentage of the Chinese but not Italian population (105.6 males per 100 females, 93 males per 100 females, respectively) (309). Therefore, population composition cannot solely explain the sex discrepancy. Another confounding variable: males are more likely than females to have comorbidities and gender-specific behavior that increase ACE2 levels, e.g., cardiovascular disease, hypertension, obesity, and smoking. For example, obesity was found to be a major risk factor for COVID-19 mortality in men but not women (310). Smoking habits are also disproportionately increased in males more than females, as seen in China (288 million and 12.6 million, respectively) and globally (40% and 9% tobacco use, respectively) (311, 312). Additionally, males with severe COVID-19 reportedly have higher CRP concentration than females, independent of comorbidities and age (313). Males have elevated proinflammatory cytokine responses, but females apparently have robust T-cell activation (305, 314, 315). Elevated cytokine response and poor T-cell response correlate with worse disease outcomes (305, 315). Therefore, sex differences may be less drastic than initially noted, due to the effects of confounding variables with ACE2 levels.

Although large transcriptome studies are inconsistent across multiple tissues, animal and human models favor elevated ACE2 in males, compared with females. We believe that moderate evidence of increased ACE2 expression relates to sex differences. However, studies directly comparing ACE2 and TMPRSS2 expression by sex and COVID-19 outcomes/severity are needed.

Race, Ethnicity, and Genetics

Current epidemiological data strongly indicate variability of case-fatality rates from as high as 15.1% in U.K., 14.2% in Italy, 3.3% in the U.S., and 2.1% in South Korea (3, 316). Furthermore, increasing evidence suggests that COVID-19 disproportionately affects some racial and ethnic minority groups (317–321). The OpenSAFELY extensive analytic platform indicated HR: 1.62–1.88 (adjusted for age and sex) for Black, South Asian, and mixed ethnicity, compared with white patients; and HR: 1.43–1.48 after adjustment of all included factors (145). Moreover, race/ethnicity data from the Morbidity and Mortality Weekly Report revealed that 33% of hospitalizations were for Blacks, two times the U.S. Black population. Conversely, 45% of those hospitalized were white, less than half of the white population (119). Likewise, the Johns Hopkins University and American Community Survey reported that infection and death rates were more than three times and six times higher, respectively, in Black counties than in white ones in the U.S (322).

Unfortunately, scant data examine the biological mechanisms underlying ethnic differences in COVID-19 susceptibility and severity. Preliminary data of nasal epithelial gene expression identified significantly higher nasal gene expression of TMPRSS2 in Black than in other races and ethnicities, suggesting that ACE2 variations among ethnic groups may partially contribute to the higher burden of the disease among Blacks (274). However, the study fails to present demographics, comorbidities, and smoking history of the 305 participants, which may affect ACE2 expression.

We may question whether genetic variations in ACE2 and TMPRSS2 among racial and ethnic groups exist. Fujikura and Vesaka (323) detected 349 and 551 single nucleotide variations in human ACE2 and TMPRSS2, respectively, in 156,513 individuals. The single nucleotide variations were rare but population specific and deleterious, suggesting susceptibility of different populations to SARS-CoV-2. Relevant ACE2 polymorphisms under review include rs2285666, rs1978124, and rs714205 (324). Also, deleterious variants in ACE2 differed among nine populations in gnomAD, specifically Black and non-Finnish European ones: 39% versus 54%, respectively (325).

Moreover, a recent study noted significantly higher ACE2 expression among Asians than Blacks and whites (326). Analysis of GTEx and other public data examined higher allele frequencies in expression quantitative locis (eQTLs) associated with elevated ACE2 expression in Eastern Asian population tissues; eQTLs were calculated close to 100% in Eastern Asians and >30% higher than other ethnic groups (268, 327). Conversely, Hou et al. (325) found no eQTLs for ACE2 across different populations, but polymorphisms (i.e., p.Val160Met [rs12329760]) were found in TMPRSS2. Furthermore, studies found that Asians and other races express similar levels of genetic polymorphisms of the SARS-CoV-2 entry receptor (328, 329), and transcriptomic datasets of lung tissue revealed no significant ACE2 gene expression disparities between Asians and whites (330). Therefore, genetics consortia report no direct association between SNVs and eQTLs in ACE2 and TMPRSS2.

Other haplotypes comprising ACE1, ABO-locus, 9q34.2, and 3p21.31 are of interest (324, 331). The ACE1 II genotype had a strong negative correlation with SARS-CoV-2 cases and deaths (332). The study noted that the European population had lower ACE1 II genotype frequency and higher prevalence of mortality than the Asian population (332). This may factor into Central Europe experiencing far more COVID-19 cases and deaths than East Asia.

ABO blood groups and COVID-19 incidence and mortality are being studied as well (333). In a meta-analysis of 318 studies, blood group A has a significant partial risk factor [OR: 1.16(1.02–1.33)] for COVID-19 infection, whereas AB has an insignificant but considerable one [OR: 1.25(0.84–1.86)] (334). Blood group O was considered as a possible protective factor [OR: 0.73(0.60–0.88)] against infection (334). Additionally, ABO blood groups versus severe COVID-19 extrapolated similar sentiments with blood group A and blood group O: OR: 1.45(1.2–1.75), OR: 0.65(0.53–0.79), respectively. O blood type carriers may have lower ACE levels and higher regulated IL-6 levels, suggesting increased balance in the ACE/ANG II axis (331). Guillon et al. (335) also observed that anti-A antibodies blocked S-protein/ACE2-dependent adhesion, indicating that anti-A antibodies may block the interaction between the virus and its receptor. In addition, locus 9q34.2 coincides with the ABO blood group locus with an increased OR: 1.32(1.20–1.47) with severe COVID-19.

Another genomic region of interest relates to several genes on chromosome 3 (SLC6A20, LZTFL1, CCR9, FYCO1, CXCR6 and XCR1) (333); 3p21.31 presented an increased OR: 1.77(1.48–2.11) with severe COVID-19. One gene in particular, SLC6A20, encodes for an amino acid transporter that interacts with ACE2, which may alter the binding affinity to the virus. Other genes in cluster affect immunological responses related to chemokine receptors, C-X-C motif chemokine receptor 6, and CC-motif chemokine receptor 9, in response to T-cell differentiation and recruitment.

Researchers have also hypothesized possible genetic predisposition to ACE2 polymorphisms linked to diabetes, stroke, and hypertension (170). A meta-analysis (n = 11,051) provided strong evidence that ACE2 gene polymorphism G8790A had an increased risk factor for essential hypertension across different ethnic populations in female subjects and Han-Chinese male subjects (336). Genetic predispositions associated with increased comorbidities may factor into higher prevalence of heart disease, hypertension, diabetes, and obesity in minority groups, leading to increased COVID-19 susceptibility and severity. Although current but limited studies indicate possible genetic variability in COVID-19 contraction and/or severity, much is debated. Further research is needed to understand the molecular and pathophysiological mechanisms underlying the relationship among genetics, race/ethnic disparities, and COVID-19 infection and severity.

While our review emphasizes physiologic factors, we acknowledge that COVID-19 susceptibility and severity largely result from socioeconomic inequities, which need further study. Inequities in social determinants of health affecting minority groups, e.g., racism, barriers to healthcare access and use, higher representation in occupations with increased chance of exposure, education, income, wealth gaps, crowded housing conditions, are interrelated and influence disease outcome and mortality. These are only a few factors associated with increased rates of COVID-19 cases, hospitalizations, and deaths within minority racial and ethnic communities. Behavioral and environmental factors also increase rates of certain medical conditions, such as CKD, COPD, and T2D, which affect COVID-19 incidence and severity. Easier access to information, affordable testing, and medical care are needed to mitigate inequity and empower individuals. Further research is needed to loosen systemic barriers for appropriate and critical care.

TERTIARY RISK FACTORS

The thought of certain medications increasing risk of SARS-CoV-2 infection remains controversial. ACE inhibitors (ACEi) and angiotensin receptor blockers (ARBs) are particularly criticized for possibly aggravating lung injury in infected patients due to increased ACE2 expression in the body (337). Other concerns are directed toward diabetic medications and ibuprofen (170).

Angiotensin-Converting Enzyme Inhibitors/Angiotensin Receptor Blockers

Many COVID-19 patients have cardiovascular and renal comorbidities that usually require these medications. Both medications were believed to increase ACE2 expression, potentially increasing susceptibility/severity to COVID-19.

ACEi inhibit ACE and ARBs block RAS’s effects, causing angiotensin I accumulation; ACE2 then metabolizes it into angiotensin (33, 92, 107, 112, 118, 212,213, 285, 288) (Fig. 1). ACEi do not inhibit ACE2, because ACE2’s S2′ pocket in the active site is smaller than ACE’s corresponding pocket (338). Most studies were reported in animal models (Table 1), while few were conducted in human tissue. One animal study showed that ACEi increased ACE2 mRNA 1.8-fold and that losartan showed a significant ACE2 mRNA increase (338). Other studies reported similar trends in ACE2 mRNA for enalapril, olmesartan, and valsartan (339, 341–343). Interestingly, ARBs have shown to improve lung recovery in ACE2-downregulated mice infected by the previous SARS-CoV (52). Others have also suggested that ACE2’s compensatory elevation may cause the benefit of ACEi/ARBs in non-COVID-19 viral pneumonia. However, this elevation may not be beneficial in a COVID-19 context, as SARS-CoV-2 uses ACE2 as a vector for infection (337, 345). Thus patients treated with ACEi/ARBs could be at higher risk of severe COVID-19 due to increased S-protein coronavirus binding sites in their lungs (277).

Table 1.

Evidence of the relationship between ACEi/ARBs and ACE2 levels in animal models

Study	ACEi/ARB	Tissue	ACE2 mRNA	ACE2 Serum/Protein
Angiotensin-converting enzyme inhibitor
Ocaranza et al. (339)	Enalapril	Heart	↑	↑
Ferrario et al. (338)	Lisinopril	Heart	↑	↔
Tikellis et al. (10)	Perindopril	Kidney	↓	↓
Lezama-Martinez et al. (340)	Captopril	Aorta	↓	N/A
Hamming et al. (102)	Lisinopril	Kidney	↓	↔
Ferrario et al. (338)	Lisinopril	Kidney	↔	↑
Burrell et al. (157)	Ramipril	Heart	↔	↔
Angiotensin receptor blocker
Ishiyama et al. (341)	Losartan	Heart	↑	N/A
Ishiyama et al. (341)	Olmesartan	Heart	↑	N/A
Whaley-Connell et al. (342)	Valsartan	Kidney	↑	N/A
Takeda et al. (343)	Candesartan	Heart	↑	↑
Ferrario et al. (338)	Losartan	Heart	↑	↑
Kuba et al. (52)	Losartan	Lung	N/A	↑
Ferrario et al. (338)	Losartan	Kidney	↔	↑
Lezama-Martinez et al. (340)	Losartan	Aorta	↓	N/A

Summary of the evidence of the relationship between ACEi/ARBs and ACE2 levels. More comprehensive table reviewed in Kreutz et al. (344; Table 1): effect of renin-angiotensin system blockers on ACE2. ACEi: angiotensin-converting enzyme inhibitor; ARB: angiotensin receptor blocker; ACE2: angiotensin-converting enzyme 2.

Although evidence supports increases in ACE2 by ACEi/ARBs, it remains controversial. Perindopril has shown to reduce plasma and cortical ACE2 activity in both healthy and diabetic mice (10). Similarly, captopril, lisinopril, and losartan decrease ACE2 mRNA in mice (102, 340). In addition, Burrell et al. (157) indicated that ramipril does not affect cardiac ACE2 mRNA and proteins.

In sum, a review of the available literature revealed the use of ARBs to potentially elevate ACE2 mRNA and proteins in experimental models, while ACEi has inconsistent changes (344). However, the Human Lung Tissue Expression Quantitative Trait Loci Study, which analyzed the gene expression of ACE2, TMPRSS2, and ADAM17, observed no alterations in gene expression by ARBs but a reduction in ACE2 and TMPRSS2 expressions by ACEi (346). The study suggests that long-term ACEi use may downregulate lung ACE2 expression by reducing substrate (e.g., ANG II) availability, thereby reducing risk of SARS-CoV-2 infection. The significance of these findings is unclear. Clinical studies are needed to elucidate the relationship of ACEi/ARBs to ACE2.

Ending ACEi/ARB use in COVID-19 patients has been debated. However, observational-cohort studies denoted no statistical significance in samples of 173 and 634 patients on ACEi/ARBs, β-blockers or other anti-hypertensives to critical COVID-19’s severity and outcomes (347, 348). Population-based case-control studies in Italy (n = 6,272) and Spain (n = 1,139) reached similar conclusions (349, 350). A systematic review confirmed scant evidence for the use of ACEi/ARBs, prophylactically or therapeutically, against COVID-19 (351). In fact, the Council on Hypertension urges patients to “continue treatment with their usual antihypertensive therapy because there is no clinical or scientific evidence to suggest that treatment with ACEi/ARBs should be discontinued” (352). Also, medication changes would require patients to visit a pharmacy or obtain bloodwork, which can increase COVID-19 exposure. Since ACEi/ARBs are widely prescribed for cardiovascular and renal disease, more attention should be paid to COVID-19 patients and close contacts using ACEi/ARBs (353). β-Blockers also inhibit renin production, but their relation to ACE2 regulation and COVID-19 is unknown. Also, no evidence shows that calcium channel blockers increase ACE2 activity, so these could be alternative treatments for these patients (170). Additional pharmacological studies of the relation of ACE2 levels and COVID-19 are needed.

Thiazolidinediones and Ibuprofen

Other common medications, such as thiazolidinediones and ibuprofen, have been brought into question. Thiazolidinediones are popular diabetes treatments. These medications act on peroxisome proliferator-activated receptor-γ, which subsequently acts on RAS. One study found that pioglitazone significantly increases ACE2 protein expression in the liver, adipose tissue, and skeletal muscle, compared with the high-fat diet group (354). Yet, evidence for this is limited.

Furthermore, the French health minister warned patients to use acetaminophen over nonsteroidal anti-inflammatory drugs based on observing serious side effects from ibuprofen. A study of the link between ibuprofen and cardiac fibrosis in diabetic rats revealed an association between ibuprofen and elevated ACE2 (355). However, no strong, consistent evidence shows that ibuprofen increases ACE2. The WHO and European Medicines Agency do not condemn ibuprofen, because no clinical or population-based data on this topic are available (356).

CONCLUSIONS

ACE2 conducts extensive vascular and organ protection in significantly underlying risk factors for COVID-19. Table 2 summarizes the primary, secondary, and tertiary factors and ACE2, as discussed above. Men aged >60 with underlying comorbidities are reportedly more vulnerable to COVID-19 susceptibility and severity. Whether patients’ demographics, therapies, or other factors influence this is unclear, but COVID-19 appears related to ACE2’s role.

Table 2.

Evidence of the relationship between ACE2 and primary, secondary, and tertiary risk factors in COVID-19 patients

Risk Factors	Susceptibility	Severity	ACE2 Relationship	Experimental Models	Human Models
Primary
Respiratory disease	• Strong evidence: COPD, emphysema, smokers• Mixed evidence: asthma	• Strong evidence: COPD and emphysema• Mixed evidence: asthma• Limited evidence: smokers	• Strong: COPD, emphysema, smoking• Limited: asthma		• Increased ACE2: autopsy specimens from severe SARS patients with ALI had elevated ACE2, SARS‐CoV S protein, RNA, and proinflammatory cytokines (126)• Increased ACE2: COPD upregulates ACE2 and TMPRSS2 expression in the nasal, bronchial and lower airways (127–134)• Increased ACE2: significant inverse relationship between ACE2 gene expression and FEV1% (r = −[0.24–0.40]; P < 0.05). (127, 129)• Increased ACE2: in moderate/severe COPD lung tissue (128)• Increased ACE2: significant serum ACE2 level increases in 27 COPD patients (135)• Increased ACE2/TMPRSS2: slightly upregulated ACE2 and TMPRSS2 lung airway expression in asthmatics, while plasma ACE2 was significantly upregulated in asthmatics (133)• Increased TMPRSS2: elevated in children and adult asthmatics from available RNA-Seq databases (132)• Increased ACE2: ACE2-correlated gene signatures were found to be significant in a subset of type 2-low patients with asthma with characteristics resembling known risk factors for severe COVID-19 (140)• Increased ACE2/TMPRSS2: increase in ACE2 and TMPRSS2 gene expressions in bronchial and alveolar epithelial, bronchial alveolar lavage, and protein in lung tissue and blood (127–134, 144)• Increased ACE2: current smokers had significantly higher gene expression than ex-smokers and nonsmokers (2.77 ± 0.91 vs. 2.00 ± 1.23 vs. 1.78 ± 0.39, respectively; P = 0.024) (128, 129)• No difference in ACE2: no significant difference in ACE2 expression in the lower airways with allergic sensitization and asthma (130, 134, 141)
Cardiovascular Disease	• Strong evidence: heart failure, coronary artery disease, or cardiomyopathies (6, 124, 145, 337, 357)	• Strong evidence: heart failure, coronary artery disease, or cardiomyopathies	• Strong	• Increased ACE2: elevated serum ACE2 myocardial infarction, atherosclerotic development, reduced left ventricular ejection fraction, cardiomyopathies, and heart failure (12, 92)• Increased ACE2: elevated ACE2 mRNA days 3 and 28 postmyocardial infarction (157)	• Increased ACE2: ACE2 gene was most upregulated and a fivefold increase in ACE2 protein in hypertrophic cardiomyopathy human cardiac tissue, compared with that of controls (159)• Increased ACE2: plasma ACE2 activity directly related to persistent AF (22.8 pmol/min/mL) and paroxysmal AF (16.9 pmol/min/mL), compared with control (13.3 pmol/min/mL) (161)• Increased ACE2: elevated plasma ACE2 concentrations in 1,485 men and 537 women with heart failure and a threefold increase in myocardial ACE2 gene expression in patients with heart failure (163, 164)Increased ACE2: elevated ACE2 serum in acute heart failure (52.5 pmol/h/ml) and chronic heart failure (33.6 pmol/h/ml), compared with healthy controls (22.5 pmol/h/ml) (165)• Increased ACE2: elevated ACE2 mRNA in explanted failing hearts (157)• Increased ACE2: in patients with heart failure and correlated with disease severity (166, 167)• Increased ACE2: elevated plasma ACE2 in 79 patients with coronary artery disease and correlated with adverse long-term outcomes (162)
Hypertension	• Limited evidence: essential and secondary hypertension	• Strong evidence: essential and secondary hypertension	• Strong: essential hypertension• Limited: secondary hypertension	• Increased ACE2: positive plasma ACE2 correlation with increased systolic blood pressure (187)• Decreased ACE2: decreased local ACE2 mRNA and protein activity in tissue, primarily in the kidneys and heart, thus depict hypertensive progression to decrease local ACE2 activity (97, 176–180)	• Increased ACE2: positive plasma ACE2 correlation with increased systolic blood pressure (167, 181–187)• Increased ACE2: plasma ACE2 1.5 times greater in 239 hypertensive patients (167)• Increased ACE2: patients with detectable plasma ACE2 had significantly higher systolic and diastolic blood pressure, compared with those without (182)• Increased ACE2: elevated serum ACE2 concentrations in hypertensive patients, compared with healthy subjects (170.31 [83.50–707.12] pg/ml vs. 59.28 [39.71–81.81] pg/ml, respectively; P < 0.001) (181)• Mixed ACE2 results: noted in the relationship between secondary hypertension and ACE2 (178)• Decreased ACE2: decreased local ACE2 mRNA and protein activity in tissue, primarily in the kidneys and heart, thus depict hypertensive progression to decrease local ACE2 activity (97, 176–180)
Renal Disease	• Strong evidence: chronic kidney disease	• Strong evidence: chronic kidney disease	• Strong	• Increased ACE2: (twofold) in diabetic nephrotic mice (15)• Decreased ACE2: local tissue reduction in ACE2 mRNA and protein activity in glomerular and tubular tissue in primary glomerulopathy, IgA nephropathy, hypertension, nephrosclerosis, and nephrectomy (177, 179, 201, 205)	• Increased ACE2: elevated plasma ACE2 in CKD stages 3-5 predialysis patients, n = 1,456 (212, 213)• Increased ACE2: local ACE2 increases in glomeruli and tubules in primary and secondary renal disease, as well as renal transplant patients (103)• Increased ACE2: significantly elevated plasma ACE2 (15, 92, 183, 184, 187, 212–217)• Increased ACE2: elevated serum ACE2 in type 1 diabetes, n = 859, with micro- and macro-albuminuria; negatively correlated with glomerular filtration rate (184)• Decreased ACE2: local tissue reduction in ACE2 mRNA and protein activity in glomerular and tubular tissue in primary glomerulopathy, IgA nephropathy, hypertension, nephrosclerosis, and nephrectomy (177, 202–204, 206)
Diabetes Mellitus	• Strong evidence: type 2 diabetes• Limited evidence: type 1 diabetes	• Strong evidence: type 2 diabetes• Limited evidence: type 1 diabetes	• Strong	• Increased ACE2: twofold increase serum ACE2 in diabetic nephrotic mice (15)• Increased ACE2: islet expression early in the disease and decreased with disease progression in type 2 diabetic mice (229)• Increased ACE2: elevated serum ACE2 observed in diabetic mice (15, 187)• Increased ACE2: increased glucose to directly increase viral load, ACE2, and IL-1β expression in SARS-CoV-2 infected monocytes in a dose-dependent manner.• Decreased ACE2: decreased local ACE2 by ∼30% in diabetic kidneys (211)	• Increased ACE2: type 2 diabetes was causally linked to raised ACE2 expression (P = 2.91E-03; MR-IVW) (230)• Increased ACE2: elevated serum ACE2 in Type 1 diabetes, n = 859, with micro- and macro-albuminuria; negatively correlated with glomerular filtration rate (184)• Increased ACE2: elevated expression in the pancreas of healthy subjects (n = 74); slightly higher than in the lungs (227)• Increased ACE2: 5,457 Icelanders found significant associations of elevated serum ACE2 levels in smokers and in obese or diabetic individuals (231)• Increased ACE2: increased urinary ACE2 and ADAM17 in 40 T2D patients (232)
Obesity	• Strong evidence: BMI >30 kg/m2	• Strong evidence: BMI >30 kg/m2	• Strong	• Increased ACE2: seen in obese, type 2 diabetic mice (108)• Increased ACE2: elevated ACE2 mRNA expression, protein, and circulating ACE2 levels (235–239)• Increased ACE2: elevated local ACE2 in augmented cardiomyopathy in lean/obese mice (240)	• Increased ACE2: ACE2 expression was found to be higher in human subcutaneous adipose tissue and human visceral adipose tissue than in human lung tissue (107)
Gastrointestinal Disease	• Limited evidence: Inflammatory bowel disease, irritable bowel syndrome, peptic ulcer disease, gastroparesis	• Limited evidence: Inflammatory bowel disease, irritable bowel syndrome, peptic ulcer disease, gastroparesis	• Limited		• Increased TMPRSS2: Crohn’s disease ileum was 70% higher (P < 0.05) than in controls. Ulcerative colitis ileum was 30% higher (P < 0.05) than in controls (278)• Increased ACE2: Crohn’s disease colonic was 30% higher (P < 0.05) than in controls. Ulcerative colitis colonic was 70% higher (P < 0.05) than in controls (278)• No difference ACE2: ulcerative colitis ileum did not differ compared with controls (278)• No difference TMPRSS2: Crohn’s disease colonic did not differ compared with controls (278)• Decreased ACE2: Crohn’s disease ileum was 60% lower (P < 0.05) than in controls (278)
Cerebrovascular Disease	• Limited evidence	• Mixed evidence	• Limited	• Increased ACE2: in mice with cerebral ischemic lesions, resulting in a significant increase in regional cerebral and circulating ANG-1-7 at 12 h, compared with control (7.276 ± 0.320 vs. 2.466 ± 0.410 ng/mg, serum; 1.024 ± 0.056 vs. 0.499 ± 0.032, brain; P < 0.05) (250)• Increased ACE2: elevated ACE2 expression in the cortex penumbra in pathological reports of rats after ischemic injuries and smoking (251)	• Limited ACE2 evidence: in several neuropsychiatry conditions, cerebrovascular ischemic and hemorrhagic lesions, and neurodegenerative diseases (249)
Secondary
Age	• Mixed evidence: Prevalence of infection is related to age	• Strong evidence: Increased risk for severe illness increases with age, making elders a high-risk population (4, 5, 257–259)	• Moderate	• Increased ACE2/TMPRSS2: elderly mice had higher expression of ACE2 and TMPRSS2 in nasal mucosa, compared with younger mice (272, 273)• No difference ACE2: no significant difference in ACE2 expression in cerebrovascular tissue and age in older mice (266, 267)• Decreased ACE2: significantly lower ACE2 levels in older 24-mo-old male and female mice and aging endothelial cells (263–265)	• Positive ACE2 correlation: with age, n = 118, 41-70 yr (281)• Positive ACE2 gene expression correlation: With age groups; n = 305, 4-60 yr (274)• Increased ACE2: Significant linear trends were found in nasal epithelial ACE2 gene expression with advancing age groups in a cohort of 305 individuals aged 4-60 (274, 275)• Increased ACE2: positively weak correlation (r = 0.20, P = 0.02) between ACE2 expression and age (276)• Increased ACE2: significantly elevated ACE2 and TMPRSS2 in adults, compared with children in nasal and bronchial tissue in multiple transcriptomic datasets (133)• Increased ACE2: ACE mRNA in duodenal and ileal biopsies moderately correlated with age (r = 0.32, P = 0.0099; r = 0.64, P = 0.0099, respectively) (278, 279)• Increased ACE2: positive association between age and ACE2 serum activity in 118 healthy and 213 mild-moderate hypertensive patients (280, 281)• Increased ACE2: large cohort of participants (n = 2,051) in a commercial wellness program found significantly higher plasma ACE2 levels in older individuals; age association was more pronounced in women pre- and postmenopausal (282)• Increased ACE2: significantly increased serum in adolescents/young adulthood (283)• No difference in ACE2: between four age groups, neonates to >65 yr (271)• No difference in ACE2/TMPRSS2: no significant differences in ACE2 and TMPRSS2 gene expression with ages >60 yr vs. <60 yr and ages ≤49 yr vs. ages >49 yr (269, 270)• Decreased ACE2: ACE2 expression to significantly decrease with age in Caucasian males, although data remained nonsignificant in other ethnic groups (268)
Sex	• Limited evidence	• Strong evidence: Males at higher risk than females for contracting severe COVID-19	• Moderate	• Increased ACE2: elevated local ACE2 in mice male kidneys (290–293)• Increased ACE2: serum ACE2 levels were found elevated in male normotensive mice (293, 295)• Decreased local ACE2: in mice male kidneys (294)• Estrogen upregulates: ACE2, AT2R and MAS expression levels through effects at estrogen receptor-mediated binding at the ACE2 promoter (303, 304)• Estrogen downregulates: E2-treated NHBE cells expressed lower levels of ACE2 mRNA (306), while ovariectomy (loss of E2) increased ACE2 activity, and orchiectomy (loss of androgens) decreased enzyme activity (292, 307, 308)	• Increased ACE2: increased ACE2 expression in male lung tissue, largely in type II pneumocytes, and male plasma (297–299)• Increased ACE2: ACE2 expression in healthy ileal biopsies (n = 154) were found to be 130% greater in men than in women (P = 0.0256) (278)• Increased ACE2: increased circulating ACE2 levels in both healthy males or males with comorbidities (132, 164, 217, 280, 282, 283, 301)• Increased ACE2: male sex as the strongest predictor of elevated concentrations of ACE2 in control and heart failure cohorts (coefficient = 0.19 and 0.26, respectively; P < 0.001) (164)• Increased ACE2: increased ACE2 levels in males with preserved ejection fraction heart failure (164, 301)• Increased ACE2: ACE2 increased more in boys than in girls, emphasizing possible sex hormonal regulation (283)• No difference in ACE2: no sex differences for ACE2 in 118 healthy men and women (281)• No difference in ACE2: gene expression in gender in transcriptome databases (268, 270, 300, 328, 330)• Decreased ACE2: Asian males to have lower ACE2 expression than females out of three studied ethnic groups (268)
Race / Ethnicity / Genetics	• Limited evidence	• Strong evidence: ethnic/racial minorities at higher risk for severe COVID-19. In addition, significant national differences in case-fatality ratios	• Limited		• Increased TMPRSS2: significantly higher nasal gene expression of TMPRSS2 in Blacks than in other races and ethnicities (274)• Increased ACE2: deleterious variants in ACE2 differ among nine populations in gnomAD, specifically in African Americans and non-Finnish European population: 39% vs. 54%, respectively (325)• Increased ACE2: ACE2 expression is significantly higher among Asians than African Americans and Caucasians (326)• Increased eQTLs: eQTLs associated with elevated ACE2 expression in tissues of Eastern Asian population: close to 100% in Eastern Asians and >30% higher than other ethnic groups (268, 327)• No difference in eQTLs: no difference in eQTLs any eQTLs for ACE2 across different populations, though polymorphisms (i.e., p.Val160Met [rs12329760]) were found in TMPRSS2 (325)• No difference in genetic polymorphisms: Asians and other races express similar levels of genetic polymorphisms of the SARS-CoV-2 entry receptor (328, 329)• No difference in ACE2: expression in Asians compared with other races and no unique genetic polymorphisms (328)• No difference in ACE2: gene expressions in lung tissue between Asians and Caucasians in transcriptome databases (330)• Decreased ACE1 II: European population had lower ACE1 II genotype frequency and a higher prevalence of and mortality than the Asian population (332)• Decreased ACE2: O blood type carriers may have lower ACE levels and higher regulated IL-6 levels, suggesting increased balance in the ACE/ANG II axis (331)
Tertiary
ACEi/ARBs	• Strong evidence: for no significant differences in ACEi/ARBs use in non-COVID-19 and COVID-19 patients	• Strong evidence: for no significant differences in ACEi/ARBs use in non-COVID-19 and COVID-19 patients	• Moderate/ limited	• Increased ACE2 mRNA: in mice treated with losartan, olmesartan, valsartan, candesartan, enalapril, and lisinopril (338, 339, 341–343)• Increased ACE2 protein: In mice treated with losartan, candesartan, enalapril, and lisinopril (52, 338, 339, 343)• Decreased/no change in ACE2 mRNA: in mice treated with lisinopril, losartan, perindopril, captopril, losartan, lisinopril, and ramipril (10, 102, 157, 338, 340)• Decreased/no change in ACE2 protein: in mice treated with lisinopril, perindopril, and ramipril (10, 102, 157, 338)	• Decreased ACE2 expression: lung eQTL study showed the possibility that long-term ACEi use downregulates lung ACE2 expression by reducing substrate availability (346)• Decreased TMPRSS2 expression: lung eQTL study showed ACEi deceased TMPRSS2 expression (346)• No difference in ACE2 expression: lung eQTL study showed that ARBs did not alter ACE2 gene expression (346)
Thiazolidinediones			• Limited	• Increased local ACE2: in the liver, adipose tissue, and skeletal muscle when treated with Pioglitazone (354)
Ibuprofen			• Limited	• Increased ACE2: When on ibuprofen in diabetic rats with cardiac fibrosis (355)

Summary of the evidence of the relationship between ACE2 levels and primary, secondary, and tertiary risk factors. ACE2, angiotensin-converting enzyme 2; COVID-19, coronavirus disease; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; ACEi, angiotensin converting enzyme inhibitor; ARBs, angiotensin II receptor blockers; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; MR-IVW, Mendelian randomization-inverse‐variance weighted is correctly written out as inverse variance weighted; BMI, body mass index; TMPRSS2, type 2 transmembrane serine protease; AF, atrial fibrillation; eQTL, expression quantitative locis. ARBs had moderate evidence for elevated ACE2, while ACEi had weak evidence.

As ACE2 provides a pathway for SARS-CoV-2 invasion, increasing studies are investigating the underlying imbalance in the ACE2/ANG-1-7/Mas axis. Dysregulation of ACE2 during COVID-19 may come at a striking cost of direct organ-toxicity, cytokine storm, and endothelial dysfunction. Previous studies have noted that risk factors elevate plasma ACE2 activity. Whether increased plasma ACE2 activity reflects increased synthesis from tissue, ACE2 mRNA, or increased shedding of tissue ACE2 remains to be determined.

Nevertheless, recent studies are uncovering the correlation of ACE2 activity, viral load, and severity of the disease with the preliminary consensus of 1) increased SARS-CoV-2 expression and ACE2 activity is found in direct viral organ-toxicity in severe/critical COVID-19; and 2) elevated viral load relates to increased severity. However, ACE2 activity and susceptibility of COVID-19 are yet to be determined. Further clinical studies are needed to understand ACE2’s relationship to susceptibility and severity of the disease. Novel therapies against SARS-CoV-2 will undoubtedly explore its association with ACE2.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

G.P. prepared figures; G.P., P.P.F., and A.R.H. drafted manuscript; G.P., P.P.F., A.R.H. and A.E.A. edited and revised manuscript; G.P., P.P.F., A.R.H. and A.E.A. approved final version of manuscript.