Identification of a pharmacological approach to reduce ACE2 expression and development of an in vitro COVID-19 viral entry model

Abstract

Because of rapid emergence and circulation of the SARS-CoV-2 variants, especially Omicron which shows increased transmissibility and resistant to antibodies, there is an urgent need to develop novel therapeutic drugs to treat COVID-19. In this study we developed an in vitro cellular model to explore the regulation of ACE2 expression and its correlation with ACE2-mediated viral entry. We examined ACE2 expression in a variety of human cell lines, some of which are commonly used to study SARS-CoV-2. Using the developed model, we identified a number of inhibitors which reduced ACE2 protein expression. The greatest reduction of ACE2 expression was observed when CK869, an inhibitor of the actin-related protein 2/3 (ARP2/3) complex, was combined with 5-(N-ethyl-N-isopropyl)-amiloride (EIPA), an inhibitor of sodium-hydrogen exchangers (NHEs), after treatment for 24 h. Using pseudotyped lentivirus expressing the SARS-CoV-2 full-length spike protein, we found that ACE2-dependent viral entry was inhibited in CK869 + EIPA-treated Calu-3 and MDA-MB-468 cells. This study provides an in vitro model that can be used for the screening of novel therapeutic candidates that may be warranted for further pre-clinical and clinical studies on COVID-19 countermeasures.

1. Introduction

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged as a global pandemic at the end of 2019.1, 2, 3 Since the Pfizer-BioNTech COVID-19 vaccine was authorized for emergency use by the FDA on Dec. 11, 2020, a total of three COVID-19 vaccines have been authorized or approved for use in the U.S.A.⁴ However, subsequent emergence of SARS-CoV-2 variants, e.g., Alpha, Beta, Gamma, Delta, and Omicron, gave rise to additional public health concerns.5, 6, 7 Furthermore, it has been reported that waning efficacy of COVID-19 vaccines was most notable against the Omicron variant.⁸ Thus, there is still an urgent need to develop novel and effective therapeutics to treat COVID-19.

Angiotensin-converting enzyme 2 (ACE2) is a dimeric, type 1 membrane protein expressed in a wide variety of human tissues, including lungs, heart, kidneys, and intestines.⁹ The SARS-CoV-2 spike protein binds to ACE2 to facilitate viral entry. The expression pattern of ACE2 suggests that in addition to playing important roles in the regulation of the biological functions in those tissues and organs, ACE2 also serves as the receptor for SARS-CoV-2 to infect other tissues and organs apart from the lungs.^10,11 A correlation has been shown between a high level of ACE2 expression and increased SARS-CoV-2 infection,^12,13 and downregulating ACE2 expression may reduce SARS-CoV-2 infection.^12,14

In this study, in order to establish an in vitro model of SARS-CoV-2 cellular entry, we first examined ACE2 expression in a variety of mammalian cell lines, some of which are commonly used to study SARS-CoV-2. Based on ACE2 expression levels, Calu-3, Vero, and MDA-MB-468 cells were selected for further study to develop the in vitro model. As the second step, a variety of compounds or inhibitors were screened based on ACE2 expression changes, and a treatment combination of CK869, an inhibitor of the actin-related protein 2/3 (ARP2/3) complex plus 5-(N-ethyl-N-isopropyl)-amiloride (EIPA), an inhibitor of sodium-hydrogen exchangers (NHEs) was selected as this combination efficiently decreased ACE2 protein expression in the cell lines tested. Next, we used pseudotyped lentiviruses expressing the SARS-CoV-2 full-length spike protein from the Wuhan-Hu-1 strain, Delta and Omicron variants to study viral entry. Finally, we confirmed that the combination of CK869 plus EIPA inhibited viral entry in MDA-MB-468 and Calu-3 cells.

2. Materials & methods

2.1. Cells

Calu-3, Vero, Caco-2, and BT-20 cells were purchased from the American Type Culture Colelction (ATCC) and were maintained in Minimum Essential Medium (MEM) containing 10% foetal bovine serum (FBS). MDA-MB-468 (ATCC) and BT-474 cells (ATCC) were maintained in RPMI-1640 medium containing 10% FBS. SKBR-3 cells (ATCC) were maintained in DMEM/F12 (1:1) containing 10% FBS. JIMT1 (DSMZ), MCF-7 (ATCC), HT-1080 (ATCC), and HFF (kindly provided by Susan Yamada, NIH, Bethesda) cells were maintained in Dulbecco's modified Eagle Medium (DMEM) containing 10% FBS.

2.2. Ligand, chemical compounds, and inhibitors

All ligands, chemical compounds, and inhibitors used in this manuscript were cell biological grade: EGF (Sigma-Aldrich, cat# E9644), NH₄Cl (Sigma-Aldrich, cat# A0171), NSC23766 (TOCRIS, cat# 2161), Casin (TOCRIS, cat# 5050), ZCL278 (TOCRIS, cat# 4794), ML141 (TOCRIS, cat# 4266), PBP10 (Calbiochem, cat# 529625), Rapamycin (Sigma-Aldrich, cat# S-015), U0126 (Sigma-Aldrich, cat# 19–147), Quercetagetin (Calbiochem, cat# 551590), Afatinib (Selleckchem, cat# S1011), 5-(N,N-Dimethyl)-amiloride hydrochloride (DMA, Sigma-Aldrich, cat# A4562), 5-(N-ethyl-N-isopropyl)-amiloride (EIPA, Cayman Chemical Company, cat# 14406), Wiskostatin (TOCRIS, cat# 4434), 187-1, N-WASP inhibitor (TOCRIS, cat# 2067), CK869 (TOCRIS, cat# 4984), CK666 (TOCRIS, cat# 3950), Cytochalasin D (TOCRIS, cat# 1233), and LY294002 (Sigma-Aldrich, cat# L9908).

2.3. Cell culture on Matrigel

Matrigel matrix (Corning, cat# 354234) preparation was described previously.¹⁵ Briefly, vials of aliquoted Matrigel stored at -20°C were thawed on ice and the liquid Matrigel was immediately applied to 6-well plates (≈600μl per well). Dishes were incubated at 37°C for 30 min until the Matrigel was polymerized. 5 x 10⁶ cells were seeded on top of the polymerized Matrigel in 6-well plates and incubated for 4 days. Cells cultured on 6-well plates were subjected to Western blotting analysis.

2.4. Viral entry assay

Lentiviral particles pseudotyped with the SARS-CoV-2 spike protein were produced in 293T cells by transfection of a lentiviral backbone encoding CMV-Luciferase-IRES-ZsGreen as well as lentiviral helper plasmids and Wuhan-Hu-1, Delta (B.1.617.2), and Omicron (BA.1) spike expression plasmid as previously described.¹⁶ To measure viral entry, 2.0 x 10⁵ Calu-3, Vero and MDA-MB-468 cells were seeded in 96-well plates and incubated overnight at 37°C. Cells were pre-treated with a combination of 50 μM CK869 and 40 μM EIPA or left untreated for 24 h at 37°C. Prior to the addition of lentiviral pseudovirus, cells were pretreated with SARS-CoV-2 pseudovirus infection enhancer (101Bio, cat# CoV2) at a volume of 1/10 of the cell culture media in each well and incubated for 30 min at 37°C. Lentiviral pseudovirus with a titer of approximately 1 x 10⁶ relative luminescence units (RLU)/mL of luciferase activity was then added, and Calu-3, Vero, and MDA-MB-468 cells were incubated for 72 h at 37°C. Cell extracts were harvested, lysed, and luciferase levels were assayed using a luciferase-based assay system (Promega, Madison, WI). The experiment was performed in at least triplicate.

2.5. Western blotting

The 5 x 10⁵ Calu-3, Vero, and MDA-MB-468 cells were seeded in 6-well plates one day prior to cell treatment, the next day cells were treated with a series of drugs as described in the manuscript. Twenty-four or 48 hours after drug treatment, cells were washed with PBS twice and then lysed with a NP40 lysis buffer on ice for 30 min. After centrifugation, whole cell lysates (WCL) were subjected to Western blot analysis. Western blotting panels shown in the figures are a representative of three independent experiments. Image J software (NIH, Bethesda) was used for densitometry of Western blotting. The following primary antibodies were used for Western Blot analysis: ACE2 (Abcam, cat# ab15348), TMPRSS2 (Abcam, cat# ab92323), EGFR (BD Biosciences, cat# 610016), phospho-EGFR (Y1045) (Cell Signaling Technology, cat# 22371), Actin (Sigma-Aldrich, cat# A1978). For detecting secreted form of ACE2 in cell culture media, 1.5 x 10⁶ of Vero and Calu-3 cells were seeded in 10-cm dishes one day prior to change to serum-free media. The next day the cell culture media containing 10% FBS was changed to FBS-free media, and cells were then cultured for 4 days. After 4 days, the conditioned cell culture media was collected, and cell debris was removed by centrifugation. The cell culture medium was concentrated using Amicon Ultracel 10k centrifugal filters (Millipore, cat# UFC901024) and the 50-times concentrated cell culture media was subjected to Western blot analysis for detection of ACE2 expression (Abcam, cat# ab15348).

2.6. Flow cytometry

Cell surface ACE2 expression level was evaluated using flow cytometry. Briefly, after harvesting cells using 0.05% Trypsin-EDTA (Thermo Fisher Scientific, cat# 25300-054), cells were washed with PBS twice and fixed in 4% paraformaldehyde (PFA) for 30 min. Then, cells were washed with PBS twice and incubated with anti-ACE2 antibody (Thermo Fisher Scientific, cat# MA5-32307) in FACS buffer (1% FBS in PBS) on ice for 1 h. After washing with PBS twice, cells were incubated with FITC-conjugated secondary antibody for 30 min at room temperature (RT). The rabbit IgG was used as isotype control. Subsequently, cells were washed with FACS buffer (1% FBS in PBS) and analyzed using a LSR Fortessa flow cytometer (BD Bioscience, San Jose, CA, USA).

2.7. Statistical analysis

GraphPad Prism was used for statistical studies. Statistical significance was determined by Student's t-test (*, p-value <0.05; **, p-value <0.01; ***, p-value <0.0001). Data is expressed as mean ± SD.

3. Results & discussion

To develop an in vitro model to screen potential drugs that can inhibit ACE2-dependent viral entry into mammalian cells, we examined expression levels of ACE2 and TMPRSS2 in cells using Western blot analysis. As shown in Fig. 1a and b, two ACE2 bands with different molecular weights (120 kDa and 85 kDa) were detected. According to published literature, the high molecular weight version of ACE2 is a glycosylated form, which exists on the cell surface, whereas its lower molecular weight version is the enzymatically deglycosylated form or a secreted form that is not located on the cell surface.¹⁷ Among the cell lines that we have tested, Calu-3 cells (human respiratory epithelial cell line), which are commonly used in COVID-19 studies, expressed the highest level of ACE2 (Fig. 1b), while MDA-MB-468 cells, a triple-negative breast cancer cell line, expressed the highest level of EGFR (Fig. 1a and b). TMPRSS2 is a co-receptor for SARS-CoV-2 and supports ACE2 binding and entry.^18,19 It was detectable in all cell lines (Fig. 1a and b), but levels of TMPRSS2 expression did not correlate with that of ACE2. Furthermore, cell surface ACE2 expression was evaluated in MDA-MB-468, Vero, and Calu-3 cells using flow cytometry analysis. As shown in Fig. 1c, cell surface ACE2 expression level was very similar in Vero and Calu-3 cells, but not detected in MDA-MB-468 cells. The secreted form of ACE2 was also examined in conditioned cell culture media from Vero and Calu-3 cells after the collected conditioned cell culture media was concentrated 50 times using centrifugal filters. Only a 120 kDa band of ACE2, but not a 75 kDa one was detected in Calu-3 cells (Fig. 1d). None of the 120 kDa and 75 kDa bands were detected in Vero cells (Fig. 1d). These results suggest that the lower band of ACE2 shown in Fig. 1a, b and 1d is likely a deglycosylated form of ACE2, but not the secreted form. Based on the results shown in Fig. 1a, b, and 1c, Calu-3, Vero, and MDA-MB-468 cells with different levels of ACE2 expression were selected for further studies.

Fig. 1.

ACE2 is highly expressed in Calu-3, Vero and MDA-MB-468 cells. (a) The levels of ACE2, TMPRSS2, and EGFR expression were evaluated by Western blotting in whole cell lysate (WCL) of SKBR-3, BT-474, JIMT1, BT-20, MDA-MB-468, MCF-7, HFF, and HT-1080 cells. (b) The levels of ACE2, TMPRSS2, and EGFR expression were evaluated by Western blotting in WCL of Calu-3, Vero, Caco-2, and MDA-MB-468 cells. (c) The levels of cell surface ACE2 were evaluated using flow cytometry analysis in non-permeabilized MDA-MB-468, Vero, and Calu-3 cells. (d) The levels of secreted form of ACE2 were examined by Western blotting in concentrated conditioned cell culture medium (CM) (50x) of Vero and Calu-3 cells.

We recently found that cell extrinsic factors from Matrigel activate epidermal growth factor receptor (EGFR), resulting in primary resistance of HER2-positive breast cancer cells to T-DM1 which is an FDA-approved antibody-drug conjugate (ADC) for the treatment of HER2-positive breast cancers.¹⁵ Because MDA-MB-468 cells express high levels of EGFR (Fig. 1a and b), we examined if ACE2 expression can be affected by cell extrinsic factors. MDA-MB-468 cells were seeded on a Matrigel and cultured for 4 days. ACE2 expression was then examined in the WCLs using Western blot analysis. MDA-MB-468 cells formed spheroid-like clusters on the Matrigel matrix after 2–3 days (data not shown). An 8.2-fold increase in ACE2 expression was observed in MDA-MB-468 cells grown on Matrigel as compared to that of cells cultured on regular tissue culture dishes (2D system), while a 1.4-fold increase in EGFR expression was observed (Fig. 2a). In contrast, a decrease in TMPRSS2 expression was observed in the MDA-MB-468 cells grown on Matrigel (Fig. 2a). An 18.9-fold increase of ACE2 expression was also observed in Vero cells grown on Matrigel (Fig. 2b).

Fig. 2.

ACE2 expression is increased when MDA-MB-468 and Vero cells grow on a Matrigel matrix and decreased after MDA-MB-468 cells are treated by EGF. (a) The levels of ACE2, TMPRSS2 and EGF expression were evaluated by Western blotting in WCL of MDA-MB-468 cells cultured either on 2D (lanes 1, 2, 3) or grown on Matrigel matrix for 4 days. Lanes 1, 2, and 3 shows different WCLs harvested from three different cell densities of MDA-MB-468 cells on 2D. (b) The level of ACE2 expression was evaluated by Western blotting in WCL of Vero cells cultured either on 2D (WCLs harvested from three different cell densities of Vero cells) or grown on Matrigel matrix for 4 days. Lanes 1, 2, and 3 shows different WCLs harvested from three different cell densities of Vero cells on 2D. (c) The levels of ACE2, TMPRSS2, EGFR, and phosphorylated EGFR (Y1045) were evaluated by Western blotting in WCL of MDA-MB-468 cells in the absence or presence of 100 ng/ml EGF for 2 days. It should be noted that actin amount shown in the lower panel was used as the reference to calculate the fold changes in ACE2, TMPRSS2, EGFR, and phospho-EGFR (Y1045).

Since EGF-dependent signaling causes EGFR activation and downregulation,²⁰ we next examined if EGF treatment can affect ACE2 expression in MDA-MB-468 cells. Cells were cultured in 2D dishes in the absence or presence of 100 ng/ml EGF for 2 days. As shown in Fig. 2c, EGF ligand treatment enhanced EGFR phosphorylation (the second panel from the bottom) and decreased EGFR expression. Interestingly, the EGF-mediated downregulation of EGFR was accompanied by a 67.5% decrease in ACE2 expression compared with that of non-EGF-treated cells (Fig. 2c). Taken together, the results shown in Fig. 2a and c suggest that extrinsic factors and EGF-mediated signaling can alter ACE2 expression in MDA-MB-468 cells.

We then tested a variety of compounds and inhibitors, which are either directly or indirectly involved in the downstream effectors of growth factor receptors (e.g., EGFR and cytoskeleton rearrangement), to further determine if they can regulate ACE2 expression in MDA-MB-468 cells treated with EGF (Fig. 3a, b, 3c). The compounds and inhibitors used in this study included 20 mM ammonium chloride (NH₄Cl, autophagy inhibitor), 50 μM NSC23766 (a Rac1 GTPase inhibitor), 5 μM Casin and 20 μM ZCL278 (Cdc42 GTPase inhibitors), 20 μM ML141 (a Rac1/Cdc42 inhibitor), 10 μM PBP10 (FPR2 antagonist), 1 μg/ml rapamycin (a mTORC1 inhibitor), 10 μM U0126 (a MAPK kinase inhibitor), 50 μM quercetagetin (a flavonol that inhibits proto-oncogene serine/threonine-protein kinases, Pim-1), 200 μM afatinib (a kinase inhibitor of HER2 and EGFR), 20 μM DMA (5-(N,N-Dimethyl)-amiloride hydrochloride) and 40 μM EIPA (5-(N-ethyl-N-isopropyl)-amiloride) (inhibitors of the Na⁺/H⁺ exchanger (NHE)), 5 μM wiskostatin and 5 μM 187-1 (N-WASP inhibitors), 50 μM CK666 and 50 μM CK869 (Arp2/3 inhibitors), 10 μM cytochalasin D (an inhibitor for actin polymerization), and 10 μM LY294002 (an inhibitor of PI3K). While ACE2 reduction was observed in a range of 30% to 50% in the indicated cells treated with EGF alone, a 74% reduction of ACE2 expression was observed in PBP10 + EGF-treated cells, a 73% reduction in quercetagetin + EGF-treated cells, a 57% reduction in 187-1 + EGF-treated cells, and a 58% reduction in CK869 + EGF-treated cells (Fig. 3a, b, 3c). The results from this pilot screen suggested that PBP10, quercetagetin, 187-1, and CK869 are promising candidates to downregulate ACE2 expression in cells. The reduction of ACE2 expression was also detectable when Vero cells were treated with those compounds or inhibitors with or without EGF (Fig. 3d, e and 3f).

Fig. 3.

Chemical compounds and inhibitors can downregulate ACE2 expression in MDA-MB-468 and Vero cells. The level of ACE2 expression was evaluated by Western blotting in WCLs of MDA-MB-468 cells (a, b, c) and Vero cells (d, e, f) after the cells were treated with indicated compounds and inhibitors for 24 h in the absence or presence of EGF. It should be noted that actin amount shown in the lower panel was used as the reference to calculate the fold changes in ACE2 and TMPRSS2.

We next examined kinetics of ACE2 expression in Vero cells after exposure to different concentrations of cytochalasin D. ACE2 expression was decreased in a dose-dependent manner (Fig. 4a). Moreover, a greater reduction in ACE2 expression (50%) was observed when Vero and Calu-3 cells were co-treated with CK869 plus EIPA as compared to a single inhibitor treatment (Fig. 3, Fig. 4c). Then, cytotoxicity of CK869 plus EIPA was evaluated in Calu-3 cells. From the results shown in Fig. 4d, the 50% cytotoxic concentration (CC₅₀) was 150 ± 3.0 μM CK869 + 120.3 ± 2.4 μM EIPA. We next tested whether the ACE2-mediated viral entry was inhibited in cells treated with 50 μM CK869 + 40 μM EIPA using pseudotyped lentivirus expressing the SARS-CoV-2 full-length spike protein from the Wuhan-Hu-1, Delta, and Omicron variants. Calu-3 and MDA-MB-468 cells were selected for this experiment because luciferase activity was not detectable in Vero cells (data not shown). As shown in Fig. 4e, luciferase activity was dramatically diminished in CK869 + EIPA-treated Calu-3 cells compared with mock-treated Calu-3 cells. Interestingly, even though cell surface ACE2 was not detected in MDA-MB-468 cells (Fig. 1c), similar levels of luciferase activity to those in Wuhan-Hu-1, Delta and Omicron-infected Calu-3 cells were detected in MDA-MB-468 cells, and these were significantly diminished in CK869 plus EIPA-treated cells (Fig. 4f). These results showed at least two possible mechanisms for SARS-CoV-2 viral entry. One is that the reduction of ACE2 expression correlates with the inhibition of the ACE2-mediated entry of pseudotyped lentivirus expressing the SARS-CoV-2 full-length spike protein into cells and the other is that the entry of pseudotyped lentivirus may occur independently of ACE2.

Fig. 4.

A combination of CK869 and EIPA reduces ACE2 expression in Calu-3 and inhibits ACE2-mediated viral entry in Calu-3 and MDA-MB-468 cells. (a) The levels of ACE expression were evaluated by Western blotting in WCLs of Vero cells after the cells were treated with different concentrations of cytochalasin D for 24 h. (b) Kinetics of ACE2 expression in cytochalasin D-treated Vero cells were obtained from results of Fig. 4a. (c) The level of ACE2 expression was evaluated by Western blotting in WCLs of Calu-3 cells after the cells were treated with indicated compounds and inhibitors for 24 h. Noted that actin amount shown in the lower panel was used as the reference to calculate the fold changes in ACE2. (d) Cytotoxicity of CK869 plus EIPA in Calu-3 cells. (e) Luciferase activity is proportional to the number of Calu-3 cells infected with pseudotyped lentivirus particles expressing SARS-CoV-2 full-length spike protein from Wuhan-Hu-1, Delta and Omicron variants. Luciferase expression (RLU) was quantified. ***: p < 0.0001. Viral entry assay using Wuhan-Hu-1, Delta, and Omicron variants shown in this figure are a representative of four and two independent experiments, respectively. (f) Luciferase activity is proportional to the number of MDA-MB-468 cells infected with pseudotyped lentivirus particles expressing SARS-CoV-2 full-length spike protein from Wuhan-Hu-1, Delta and Omicron variants. Luciferase expression (RLU) was quantified. ***: p < 0.0001. Viral entry assay using Wuhan-Hu-1, Delta, and Omicron variants shown in this figure are a representative of four and two independent experiments, respectively.

The ACE2 gene and protein expression is controlled by many factors, including both extrinsic and intrinsic cellular factors.21, 22, 23, 24 In this study, we have defined a pharmacological approach to reduce ACE2 expression levels in a variety of mammalian cell lines which should facilitate the discovery of novel drugs capable of either blocking or reducing ACE2-mediated viral entry. These may help to alleviate the multi-organ complications and severity of COVID-19.

We first took advantage of our previous findings that protein levels of cell surface receptors such as HER2 and EGFR can be regulated by cell extrinsic factors from the Matrigel system that activates EGF-mediated receptor activation and downregulation.¹⁵ Surprisingly, we found that Matrigel can modulate ACE2 expression in cells, which suggests that EGF-coupled signaling may also play a role in regulation of ACE2 protein expression on the cell surface. Similar to EGFR, ACE2 expressio was also downregulated when cells were treated with EGF, which correlated with EGFR activation and downregulation. Our findings provide an important clue that compounds or inhibitors that are functionally directly or indirectly linked to the EGFR-couple signaling and/or other important biological activity such as cytoskeleton rearrangement may modulate ACE2 protein expression. However, further study is needed on how ACE2 protein expression is regulated by extrinsic factors from the Matrigel matrix and EGF-mediated signaling. Several drug candidates, i.e., 187-1, quercetagetin, EIPA and K869, were evaluated using our model system for their capability to reduce ACE2 protein expression. The combination of two inhibitors, i.e., CK869 + EIPA, exhibited high potency in reducing ACE2 protein levels in Calu-3 cells than did either agent alone. This finding indicates that this combination potently inhibits ACE2-mediated SARS-CoV-2 viral entry into Calu-3 cells. Calu-3 cells were originally isolated from the lung epithelial tissue of a male patient with lung adenocarcinoma.²⁵ Since Calu-3 cells naturally express relatively high levels of ACE2, these cells are therefore a highly relevant in vitro system for studying ACE2-mediated viral entry. Interestingly in our study, we found that pseudotyped lentivirus particles bearing the SARS-CoV-2 spike were unable to infect Vero cells that had similar levels of ACE2 expression to that in Calu-3 cells. These results suggest that the expression of ACE2 alone may not be sufficient to mediate SARS-CoV-2 cellular entry or that in Vero cells there is a mechanism, yet unidentified, that might inhibit SARS-CoV-2 cellular entry. Recently, ACE2-independent viral entry mechanisms have been suggested.26, 27, 28, 29 We found that pseudotyped lentivirus particles bearing the SARS-CoV-2 spike can infect MDA-MB-468 cells that do not express ACE2 on the cell surface. This finding suggests that SARS-CoV-2 can infect ACE2-negative cells via ACE2-independent mechanism(s). The combination treatment of CK869 plus EIPA blocks viral entry in both Calu-3 and MDA-MB-468 cells, indicating that this combination treatment has an impact on both ACE2-mediated and ACE2-independent SARS-CoV-2 cellular entry pathways.

4. Conclusions

The results from our study provide a proof of concept that the ACE2 protein expression levels can be downregulated by a number of inhibitory compounds that target EGF-mediated signaling and that the downregulation of ACE2 expression reduces ACE2-mediated viral entry into human cells. This study also found that SARS-CoV-2 may infect human cells via an ACE2-independent pathway. Taken together, our results suggest that further pre-clinical and clinical studies are warranted to develop drugs that could reduce ACE2 protein expression and inhibit ACE2-mediated and ACE2-independent viral entry to treat COVID-19.

Author contributions

Conceptualization, Y.E. and W.J.W.; methodology, Y.E., B.H., N.I., H.P., and K.T.; software, Y.E., B.H., and N.I.; validation, Y.E. and W.J.W; formal analysis, Y.E., and W.J.W.; investigation, Y.E. and W.J.W.; resources, W.J.W.; data curation, Y.E. and W.J.W.; writing-original draft, Y.E. and W.J.W.; editing, Y.E., B.H., N.I., N.M., H.P., R.P.D., and W.J.W.; visualization, Y.E. and W.J.W.; supervision, W.J.W.; project administration, W.J.W.; funding acquisition, Y.E. and W.J.W. All authors have read and agreed to the published version of the manuscript.

Disclaimer

This article reflects the views of the authors and should not be construed to represent the U.S. FDA's views or policies.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.