Quercetin and Its Metabolites Inhibit Recombinant Human Angiotensin-Converting Enzyme 2 (ACE2) Activity

Abstract

Angiotensin-converting enzyme 2 (ACE2) is a host receptor for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Inhibiting the interaction between the envelope spike glycoproteins (S-proteins) of SARS-CoV-2 and ACE2 is a potential antiviral therapeutic approach, but little is known about how dietary compounds interact with ACE2. The objective of this study was to determine if flavonoids and other polyphenols with B-ring 3′,4′-hydroxylation inhibit recombinant human (rh)ACE2 activity. rhACE2 activity was assessed with the fluorogenic substrate Mca-APK(Dnp). Polyphenols reduced rhACE2 activity by 15–66% at 10 μM. Rutin, quercetin-3-O-glucoside, tamarixetin, and 3,4-dihydroxyphenylacetic acid inhibited rhACE2 activity by 42–48%. Quercetin was the most potent rhACE2 inhibitor among the polyphenols tested, with an IC₅₀ of 4.48 μM. Thus, quercetin, its metabolites, and polyphenols with 3′,4′-hydroxylation inhibited rhACE2 activity at physiologically relevant concentrations in vitro.

Introduction

Human angiotensin-converting enzyme 2 (ACE2) is an 805 amino-acid transmembrane protein that contains an extracellular domain with a typical HEMGH metalloproteinase zinc-binding site.^1,2 ACE2 is localized at the lung alveolar epithelial cells, arterial and venous endothelial cells, renal tubular epithelium, and the epithelia of the small intestine.³⁻⁵ This protein acts as monocarboxypeptidase that exclusively cleaves a single C-terminal residue from angiotensin II (Ang II), generating angiotensin-(1-7).⁶ It counterbalances the accumulation of Ang II formed by the action of the angiotensin-converting enzyme (ACE). The anti-inflammatory properties of ACE2 are mediated by the activation of the Mas receptor through the ACE2-Ang-(1-7)-Mas axis⁷ in the renin–angiotensin–aldosterone system (RAAS).⁷⁻⁹ More recently, ACE2 was identified as a receptor for S-proteins of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) to infiltrate host cells.¹⁰⁻¹² Inhibiting the interaction between S protein and the host ACE2 is predicted to inhibit SARS-CoV-2 infection.¹³⁻¹⁵

A preliminary computational molecular docking study identified quercetin, luteolin, and eriodictyol as potential inhibitors of the interaction between S protein and ACE2.¹⁶ It is predicted that polyphenols inhibit the entry of SARS-CoV-2 into cells by restraining the viral S protein–human ACE2 interface; however, detailed studies that describe the interaction of flavonoids with ACE2 are scarce. It is evident from earlier studies that the kaempferol and polyphenol-containing vegetable extracts inhibit rhACE2 activity.¹⁷ Quercetin, luteolin, and eriodictyol are structurally similar flavonoids having 3′,4′-hydroxylated B-rings. Therefore, we hypothesized that flavonoids and phenolic acids with B-ring 3′,4′-hydroxylation would inhibit ACE2 activity (Figure ​Figure11). Since polyphenols are extensively metabolized upon consumption, we further evaluated the potential for quercetin metabolites to also inhibit rhACE2. In this study, we describe the validation of an rhACE2 assay and the extent these polyphenols act as rhACE2 inhibitors.

Open in a separate window

Figure 1

Quercetin and its related glycosides and metabolites, as well as flavonoids with B-ring 3′,4′-dihydroxylation were screened in this study. Quercetin primarily exists as glycosides in nature (R = glucose, galactose, or rutinose). These quercetin-3-O-glycosides are converted to quercetin by the host’s gastrointestinal tract and further metabolized by the gut microbiota and host.

Materials and Methods (Including Safety Information)

Determination of rhACE2 Enzyme Activity

The catalytic activity of rhACE2 was monitored using the substrate Mca-APK(Dnp), in which Mca fluorescence is quenched by Dnp until cleavage at Pro-Lys¹⁸ separates them. Fluorescence intensity was measured in black 96-well, optical polymer base plates (Thermo Scientific Nunc, Rochester, NY) at 320 nm excitation and 405 nm emission, with 180 rpm continuous shaking using a fluorescence plate reader (Varioskan Flash, Thermo Scientific, Vantaa, Finland). The assay buffer used in the rhACE2 activity assay was made in-house, having 0.05 M 2-morpholinoethane-sulfonic acid (MES), 0.3 M NaCl, and 10 μM ZnCl₂, pH 6.8. It was stored at 4 °C when not in use. Polyphenol standards (quercetin, quercetin-3-O-glucoside, quercetin-3-O-galactoside, rutin, tamarixetin, isorhamnetin, (±)-eriodictyol, (−)-epicatechin, luteolin, and 3,4-dihydroxyphenylacetic acid) and nicotianamine were first dissolved in methanol with a concentration of 2 mM as stock solutions. A methanol vehicle affected assay kinetics, so to conduct experiments in the absence of methanol, aliquots of these stock solutions were evaporated at room temperature and then reconstituted in the assay buffer. The rhACE2 enzyme, the Mca-APK(Dnp) substrate, and DX600 stock solutions were further diluted in the assay buffer to working solutions. For accuracy, all of the reagents were prewarmed to assay temperatures for 10 min before addition to the microplate.

To assess rhACE2 activity and optimize assay conditions, rhACE2 was serially diluted in assay buffer with a starting concentration of 250 ng/mL. In a 100 μL total reaction volume, 40 μL of the enzyme solution was mixed with 60 μL of the Mca-APK(Dnp) substrate (11.25 μM final concentration). The mixture was incubated at 37 °C for 60 min. Background controls consisted of 40 μL of assay buffer with 60 μL of the substrate.

For inhibition assays, 20 μL of the inhibitor was coincubated with 40 μL of rhACE2 (50 ng/mL final concentration), and then 40 μL of the substrate (final concentration 11.25 μM) was added for 10 min, and the fluorescence was measured at 2 min intervals at 37 °C. For quercetin, activity was fitted as a variable slope [inhibitor]-normalized response model in GraphPad Prism 7.0 to determine the IC₅₀ value. The mixture of 50 ng/mL rhACE2 and 11.25 μM substrate without quercetin exhibited the maximum rhACE2 activity. Inhibitors were screened by coincubating 50 ng/mL rhACE2 enzyme with 11.25 μM substrate with 10 μM of compounds or 1 μM of DX600 at 37 °C for 10 min, and the fluorescence was determined at 2 min intervals. Inhibitors were included in background control samples. The inhibition (%) was calculated by comparing the fluorescence with the maximum fluorescence as

where A is the fluorescence in the presence of the inhibitor, B is the background control (inhibitor + substrate) fluorescence, C is the fluorescence of the uninhibited rhACE2 control, and D is the background substrate control fluorescence.

Results

Optimization of the rhACE2 Activity Assay

An rhACE2 protein dose response was established from 3.13 to 250 ng/mL with 11.25 μM substrate at 37 °C (Figure ​Figure22). After 10 min of incubation with the substrate, enzymatic activity declined (Figure ​Figure22A). At 10 min of incubation, the amount of the product formed was linear from 3.13 to 250 ng/mL rhACE2 (Figure ​Figure22B). Thus, a 10 min incubation period was used for subsequent experiments.

Open in a separate window

Figure 2

rhACE2 activity toward the substrate Mca-APK(Dnp). (A) Time-dependent increase of the Mca product from the 11.25 μM substrate by rhACE2 at 3.13–250 ng/mL at 37 °C. (B) Dose response of rhACE2 on product formation with the 11.25 μM substrate for 10 min at 37 °C. Linear regression was R = 0.9901, P < 0.0001. Data were generated from the three independent experiments with triplicates and expressed as means ± SDs.

The intra- and inter-assay reproducibility were determined to test the robustness of the assay (Table 1). The intra-assay coefficient of variance (CV) for RFU with 25, 50, and 100 ng/mL of rhACE2 and the 11.25 μM substrate at 37 °C was 6.04–7.88%, whereas the inter-assay CV was 2.52–9.96%.

Table 1

Intra- and Inter-Assay Variation of rhACE2 Activity^a

	intra-assay RFUb			inter-assay RFU
rhACE2 concentration (ng/mL)	mean	SD	CV (%)	mean	SD	CV (%)
100	3.88	0.31	7.88	3.97	0.10	2.52
50	1.61	0.10	6.54	1.62	0.04	2.50
25	0.68	0.04	6.04	0.71	0.07	9.96

Open in a separate window

^aActivity of rhACE2 was determined at 37 °C for 10 min with the 11.25 μM substrate.

^bRelative fluorescence unit (RFU) values were generated from subtracting the substrate blank from the readings for the corresponding wells. Data are from n = 3 experiments with triplicate samples.

Quercetin Inhibits rhACE2 Activity

The reaction rate of 50 ng/mL rhACE2 in the presence of 5–100 μM quercetin was time- and concentration-dependent at 37 °C (Figure ​Figure33A). In contrast, the reaction rate of rhACE2 was consistent at 0.100 RFU/min (P > 0.05). After 2.5 min, 100 μM quercetin reduced rhACE2 activity to 0.004 ± 0.029 RFU/min (P < 0.0001 vs control), while 10 μM quercetin reduced activity to 0.048 ± 0.022 RFU/min (P = 0.0007 vs control). Incubation with 5 μM quercetin reduced rhACE2 activity from 2.5 to 8.5 min, but at 10.5 min, the difference became nonsignificant relative to the control (P > 0.05). The impact of time on rhACE2 inhibition was greatest for 5 μM quercetin (Figure ​Figure33B). To further describe the time dependency of inhibition, the quercetin IC₅₀ value was determined at 2.5 and 10.5 min (Figure ​Figure33C,D). The rhACE2 IC₅₀ of quercetin increased from approximately 4.48 μM after 2.5 min to 29.5 μM at 10.5 min.

Open in a separate window

Figure 3

Quercetin inhibits rhACE2 activity. (A) rhACE2 velocity (ΔRFU/min) and (B) rhACE2 inhibition (% of control) in the absence or presence of 5–100 μM quercetin incubation with 50 ng/mL rhACE2, 11.25 μM substrate, for up to 10.5 min at 37 °C. For (A), rate differences were analyzed by two-way ANOVA, followed by Tukey’s multiple-comparison test. Incubation time accounted for 5.92% of total variation (P = 0.0002), whereas the quercetin concentration accounts for 78.2% with P < 0.0001, however their interaction was not significant (P > 0.05). Within each time interval, bars bearing different letters indicate significant differences (P < 0.05). For (B), the % inhibition was determined relative to the untreated control at each time point. Differences were assessed by one-way ANOVA, followed by Tukey’s multiple-comparison test among different quercetin concentrations at each time interval, *P < 0.005 compared to the control. For (A) and (B), the data were obtained by three independent experiments with triplicates and shown as means ± SDs. (C, D) Analysis of rhACE2 inhibition by quercetin concentration at 2.5 min (C) and 10.5 min (D). Data were fitted using a variable slope [inhibitor]-normalized response model in GraphPad Prism 7. Data are represented by three independent experiments with triplicates, and IC₅₀ values are means ± SEMs.

Inhibition of rhACE2 was dependent on temperature (Figure ​Figure44). rhACE2 activity increased with temperature (Figure ​Figure44A). The amount of Mca formed at 10 min was 2.2-fold when temperature increased from 25 to 37 °C. At each temperature, 10 μM quercetin inhibited rhACE2 activity relative to the control (Figure ​Figure44B).

Open in a separate window

Figure 4

Temperature affects rhACE2 activity. (A) Temperature dependence on the time course of product formation with or without 10 μM quercetin. (B) Effect of temperature on product formation after 10 min of incubation with or without 10 μM quercetin. Assays were performed at 50 ng/mL rhACE2 enzyme concentration with 11.25 μM substrate. Statistical analysis was performed by one-way ANOVA, followed with Tukey’s multiple-comparison testing, ***P < 0.0005 and ****P < 0.0001. Differences between the treated and untreated samples are denoted by upper and lowercase letters, respectively. Data are means ± SDs and are of three experiments performed with triplicate samples.

Kinetics of Enzyme Inhibition

A kinetic study was performed at increasing substrate concentrations to determine how rhACE2 activity is linked to catalytic efficiency. At the optimal reaction conditions (Figure ​Figure22A), activity over a 10 min incubation period was linear over varying protein concentrations, indicating that the reaction rate was stable over this period. The Michaelis–Menten constant K_m and maximum velocity V_max were determined by coincubation of 0.7–90 μM of the substrate and 50 ng/mL rhACE2 in the presence of quercetin (10 μM), rutin (10 μM), or DX600 (1 μM), as a positive control (Figure ​Figure55, Table 2). Overall, these compounds decreased the affinity of rhACE2 to the substrate (e.g., higher K_m values compared to the rhACE2 control) and lowered the velocity (with the exception of rutin), resulting in a decreased catalytic efficiency (K_cat/K_m). DX600 reduced the maximum initial velocity (3.49 ± 0.66 pmol/min) by 74% compared with the ACE2 control (13.3 ± 0.5 pmol/min) and weakened the affinity of rhACE2 for the substrate with a 5.8-fold increased K_m value and a reduced K_cat/K_m value. A similar trend was also observed with quercetin, as it had 28% lower V_max (9.56 ± 0.36 pmol/min) and 1.7-fold increased K_m compared to the control. For rutin, although the K_m value was lower than rhACE2 alone, the maximum velocity was higher than the rhACE2 control. Thus, the K_cat/K_m value for rutin was not similar to quercetin (Table 2).

Open in a separate window

Figure 5

Plot of the rate of Mca product formation (Δpmol/min) as a function of substrate concentration (μM) in a steady-state reaction condition. rhACE2 enzyme (50 ng/mL) was incubated with various concentrations (0.70, 1.41, 2.81, 5.63, 11.25, 22.50, 45, and 90 μM) of the substrate in the presence of 10 μM quercetin, 10 μM rutin, or 1 μM DX600. The absence of quercetin/rutin/DX600 was regarded as the control. The initial velocities of each treatment were obtained by fitting the fluorescence intensity value versus time (0–10 min) data with linear regression. The plot of velocity versus substrate concentration was fitted to the Michaelis–Menten equation to determine the apparent K_m, V_max, and K_cat constant (listed in Table 2). Fit for control: R² = 0.974; rutin: R² = 0.986; quercetin: R² = 0.987; and DX600: R² = 0.949. The fitted curves were different from each other (P < 0.0001). Results are from three independent experiments with triplicates. Data present means ± SDs.

Table 2

Steady-State Kinetic Parameters of 10 μM Rutin, 10 μM Quercetin, and 1 μM DX600 with rhACE2^a

treatment	Vmax (pmol × min–1)b	Km (μM)b	Kcat (s–1)c	Kcat/Km (μM–1 s–1)
rhACE2	13.3 ± 0.5	13.2 ± 1.6	3.78 ± 0.15	0.29
rhACE2 + rutin	15.1 ± 0.6	21.6 ± 2.1	4.28 ± 0.16	0.20
rhACE2 + quercetin	9.56 ± 0.36	22.9 ± 2.2	2.71 ± 0.10	0.12
rhACE2 + DX600	3.49 ± 0.57	76.5 ± 21.9	0.989 ± 0.156	0.013

Open in a separate window

^a50 ng/mL rhACE2 enzyme was incubated with various concentrations (0.70, 1.41, 2.81, 5.63, 11.25, 22.50, 45, and 90 μM) of the substrate in the presence of 10 μM quercetin, 10 μM rutin, or 1 μM DX600. Results represent triplicates from three independent experiments. The V_max,K_m, and K_cat values were expressed as means ± SEMs.

^bK_m and V_max were determined by fitting the velocities to the Michaelis–Menten relationship by nonlinear regression.

^cTurnover number K_cat was calculated from the V_max value (K_cat = V_max/[E]), using a molecular mass of 85 kDa for rhACE2.

Discussion

Angiotensin-converting enzymes, ACE and ACE2, share considerable homology with 41.8% sequence identity at the catalytic domain, and both belong to the M2 family of metallopeptidases with HEMGH zinc-binding motifs as an amino-terminal catalytic domain.^9,20−22 Despite these similarities, their structures enable different substrate specificity and peptidase activity in the RAAS.^8,9,23,24 Structural and functional studies have revealed that flavonoids inhibit ACE because of the double bond between C2 and C3 on the C-ring; the ketone group of C4 at the C-ring; and the 3′,4′-catechol group in the B-ring.²⁵⁻²⁷ Although the ACE inhibition mechanism by flavonoids has not yet been fully understood, the catechol group in the B-ring may exert a charge–charge interaction with the Zn²⁺ ion in the ACE active site.²⁶ Specifically, luteolin exhibits the highest capacity to inhibit ACE activity among 17 flavonoids with an IC₅₀ value of 23 μM, followed by quercetin with an IC₅₀ value of 43 μM.²⁶

Under optimized assay conditions, quercetin inhibited rhACE2 activity by reducing its affinity for the Mca-APK(Dnp) substrate and led to a lower catalytic efficiency (K_cat/K_m). In addition, quercetin had mixed rhACE2 inhibition as it reduced V_max and increased K_m. Structurally related polyphenols also inhibited rhACE2. Quercetin metabolites (isorhamnetin, tamarixetin, 3,4-dihydroxyphenylacetic acid, and quercetin-3-glucuronide), rutinosides and glycosides of quercetin, and other flavonoids (luteolin, (±)-eriodictyol, and (−)-epicatechin) also inhibited rhACE2. Notably, polyphenols were less potent inhibitors than DX600. However, flavonoids still decreased K_m and K_cat/K_m values, indicating interaction with the ACE2 active site. Further analysis by computational simulation, X-ray crystallography, or NMR needs to be employed to define the specific interactions of flavonoids with 3′,4′-dihydroxylation and ACE2.

Quercetin is mainly distributed in foods as glycosides and rutinosides. It is metabolized by deglycosylation, glucuronidation, sulfation, methylation, or further catabolism by gut microbiota to phenolics such as 3,4-dihydroxyphenylacetic acid. In humans, quercetin intake results in C_max values of ≤10 μM.²⁸ Microbial quercetin metabolites are more abundant, presumably because other flavonoids are also metabolized to these phenolic acids. For example, fasting serum levels of 3,4-dihydroxyphenylacetic acid have been reported as high as ∼100 μM.²⁹ ACE2 is primarily a membrane-bound enzyme located in the heart, lungs, vasculature, kidneys, and oral and digestive tracts.³⁰ In rats, quercetin and its methyl metabolites accumulate to nmol/g (dw) concentrations in the lung, kidney, heart, and muscle after consumption of 1% quercetin for 11 weeks.³¹ After a single dose of [2-¹⁴C]quercetin-4′-glucoside to rats, 71% of the dose was recovered in the GI tract at 6 h, which later declined to 3.4% by 24 h.³² Thus, ACE2 inhibition is plausible in the oral cavity and digestive tracts.

Increased flavonoid consumption is associated with reduced mortality from cardiovascular diseases in older women.³³ Meta-analyses of randomized controlled trials have associated the intake of >500 mg quercetin per day with reduced blood pressure.^34,35 Yet the precise antihypertensive mechanism of quercetin is still not clear,³⁶⁻³⁸ partly due to the variation of administration dosage, experimental duration, type of experiment design, and participant population.³⁹In vitro studies have demonstrated that flavonoid-rich foods are capable of inhibiting ACE activity.⁴⁰⁻⁴² Also, an acute hypertensive rat model supports the antihypertensive potential of quercetin through reduced plasma ACE activity.⁴³ In contrast, ACE inhibition by flavonoids has not been observed in other human and animal studies.⁴⁴⁻⁴⁷ On the bias of homology between ACE and ACE2, it would be expected that inhibition of ACE2 by quercetin would be independent of its ability to modulate blood pressure. Further mechanistic studies are needed to define the impact of polyphenol ACE2 inhibition on the RAAS.

Inhibition of ACE2 may be undesirable because functional ACE2 inhibits inflammation by reducing activation of the angiotensin II type 1 receptor pathway.¹⁵ SARS-CoV-2 uses ACE2 as a receptor to enter cells, and the resulting proteolysis of ACE2 contributes to lung damage.^15,48 While disrupting S protein and ACE2 interactions might prevent SARS-CoV-2 entry to cells, inhibiting ACE2 activity could be detrimental to infection recovery. Therefore, it is important to clarify if polyphenol–ACE2 interactions inhibit S-protein binding and ACE2 activity in tissues. If both occur, it will be important to understand if the antioxidant and anti-inflammatory activity of polyphenols through other pathways (e.g., inhibition of nuclear factor κB and activation of nuclear factor erythroid 2-related factor 2) would negate the proinflammatory aspects of ACE2 inhibition during SARS-CoV-2 infection.

In summary, polyphenols with 3′,4′-dihydroxylation inhibit rhACE2 activity in vitro. Among the polyphenols tested, quercetin was the most effective rhACE2 inhibitor, and several of its known metabolites also function as inhibitors. In the context of flavonoid metabolism, it is plausible that dietary polyphenol intake could inhibit ACE2, particularly in the digestive tract. Given these findings, it is urgent to further investigate the functional effects of polyphenols on ACE2 in vivo.

Acknowledgments

Glossary

Notes

This work was supported by funding to B.W.B. as the Fritz Friday Chair of Vegetable Processing Research. X.L. was supported by the Guangzhou Elites Scholarship Council.

Notes

The authors declare no competing financial interest.

References