key: cord-1027694-pjnco5qe
authors: Zhou, Lanlan; Huntington, Kelsey; Zhang, Shengliang; Carlsen, Lindsey; So, Eui-Young; Parker, Cassandra; Sahin, Ilyas; Safran, Howard; Kamle, Suchitra; Lee, Chang-Min; Lee, Chun Geun; Elias, Jack A.; Campbell, Kerry S.; Naik, Mandar T.; Atwood, Walter J.; Youssef, Emile; Pachter, Jonathan A.; Navaraj, Arunasalam; Seyhan, Attila A.; Liang, Olin; El-Deiry, Wafik S.
title: Natural Killer cell activation, reduced ACE2, TMPRSS2, cytokines G-CSF, M-CSF and SARS-CoV-2-S pseudovirus infectivity by MEK inhibitor treatment of human cells
date: 2020-09-02
journal: bioRxiv
DOI: 10.1101/2020.08.02.230839
sha: af25a740b7d6c5c5d03ad706c7e7f43b26a2d701
doc_id: 1027694
cord_uid: pjnco5qe

COVID-19 affects vulnerable populations including elderly individuals and patients with cancer. Natural Killer (NK) cells and innate-immune TRAIL suppress transformed and virally-infected cells. ACE2, and TMPRSS2 protease promote SARS-CoV-2 infectivity, while inflammatory cytokines IL-6, or G-CSF worsen COVID-19 severity. We show MEK inhibitors (MEKi) VS-6766, trametinib and selumetinib reduce ACE2 expression in human cells. In some human cells, remdesivir increases ACE2-promoter luciferase-reporter expression, ACE2 mRNA and protein, and ACE2 expression is attenuated by MEKi. In serum-deprived and stimulated cells treated with remdesivir and MEKi we observed correlations between pRB, pERK, and ACE2 expression further supporting role of proliferative state and MAPK pathway in ACE2 regulation. We show elevated cytokines in COVID-19-(+) patient plasma (N=9) versus control (N=11). TMPRSS2, inflammatory cytokines G-CSF, M-CSF, IL-1α, IL-6 and MCP-1 are suppressed by MEKi alone or with remdesivir. We observed MEKi stimulation of NK-cell killing of target-cells, without suppressing TRAIL-mediated cytotoxicity. Pseudotyped SARS-CoV-2 virus with a lentiviral core and SARS-CoV-2 D614 or G614 SPIKE (S) protein on its envelope infected human bronchial epithelial cells, small airway epithelial cells, or lung cancer cells and MEKi suppressed infectivity of the pseudovirus. We show a drug class-effect with MEKi to stimulate NK cells, inhibit inflammatory cytokines and block host-factors for SARS-CoV-2 infection leading also to suppression of SARS-CoV-2-S pseudovirus infection of human cells. MEKi may attenuate SARS-CoV-2 infection to allow immune responses and antiviral agents to control disease progression.

Coronavirus 2 (SARS-CoV-2) infection progresses to a rapidly lethal adult respiratory distress syndrome (ARDS) associated with high mortality especially among the elderly or those with multiple comorbid conditions [1] [2] [3] [4] [5] . Patients with cancer are particularly vulnerable in part due to their weakened immune system and are further at risk due to the immune suppressive effects of chemotherapy [6] [7] [8] . The lethality of SARS-CoV-2, the causative agent for the COVID-19 disease, involves a fulminant cytokine storm with bilateral lung infiltrates observed on chest X-rays and CT scans [9] . It has become clear that COVID-19 disease involves multiple organ systems including pulmonary, neurological, renal, hematological and gastrointestinal systems, among others [10] [11] [12] [13] [14] [15] . The SARS-CoV-2 virus binds to angiotensin converting enzyme 2 (ACE2) receptors and cellular entry is facilitated by TMPRSS2 protease [16] . Current therapeutic approaches include a number of agents such as anti-inflammatory agents that block IL-6, steroids, anti-viral agents, convalescent serum and alpha receptor blockers [17] [18] [19] [20] [21] . There are ongoing approaches for drug discovery and drug repurposing [22, 23] .

Once SARS-Cov-2 enters into cells it triggers a host immune response that leads to pathogenesis and disease progression [24] . A SARS-CoV-2 SPIKE protein variant (D614G) has emerged as the dominant pandemic form with evidence that it increases infectivity of the COVID-19 virus [25] .

The host inflammatory response phase of COVID-19 is the phase where patients become critically ill leading to high patient mortality [26] . We sought to better understand and modulate the host immune response to SARS-CoV-2 in order to prevent or reduce disease severity. This includes strategies to inhibit expression of ACE2, the receptor SARS-CoV-2 uses to enter cells.

It is clear that while the host systemic inflammatory response makes patients critically ill, the host innate immune system including natural killer (NK) cells is involved in fighting and eliminating virally-infected cells [27] . Over the last 25 years we have studied this innate immune system pathway that the immune system uses to eliminate transformed and cancer cells as well as virallyinfected cells [28] [29] [30] [31] [32] [33] [34] . Natural killer cells secrete TRAIL which is involved in killing virally-infected as well as transformed cells [35] [36] [37] [38] . Thus, our goal was to better understand and modulate the host immune response to increase the innate immune system early in SARS-CoV-2 infection while reducing the severe inflammation that occurs late in the disease course. We further wanted to understand the impact of current therapeutics used to treat COVID-19 on SARS-CoV-2 infectivity factors, the innate immune system and the cellular inflammatory response.

Prior work has suggested that coronavirus SPIKE protein can through ACE2 activate the MAPK pathway and downstream inflammatory responses [39] . Other data suggested that MAPK regulates ACE2 [40] , and so we investigated the impact of MEK inhibition on ACE2 expression as a strategy to attenuate early SARS-Cov-2 infection. Since remdesivir has been shown to reduce hospitalization [19] and may reduce mortality in patients with severe COVID-19 infection Our results suggest that MEK inhibitors, as a class, suppress host SARS-CoV-2 infectivity factors such as ACE2 and TMPRSS2, and that alone or in combination with remdesivir, there is innate immune system activity along with suppression of inflammatory cytokines and stimulation of Natural Killer cell activity. Our results support the further investigation of MEK inhibitors as a strategy to dampen early SARS-CoV-2 infection to allow host immunity as well as potentially antiviral agents to be more effective.

Based on prior literature that SARS coronavirus SPIKE protein through ACE2 can activate MAPK signaling [39] , we hypothesized that MEK inhibitors (MEKi) may inhibit SARS-CoV-2 cellular effects. We used human tumor cell lines as well as normal human lung cells as a model to test effects of MEKi on ACE2 expression. We initially observed in H1975 human non-small cell lung cancer (NSCLC) cells that at doses below IC50, three different MEKi's suppressed ACE2 protein expression ( Figure 1A ). VS-6766 (5 µM), a small molecule RAF/MEK inhibitor, MEKi Selumetinib (20 µM), or MEKi Trametinib (5 µM) all inhibited expression of ACE2 protein (as detected by PAB13444) with more subtle effects detected by another ACE2 antibody (CS4355) that recognizes glycosylated ACE2. We include results with the two commercially available antibodies we used to demonstrate that these antibodies did not always give concordant results. In this experiment the reduction in ACE2 was clearly demonstrated with the PAB13444 antibody. We observed that the cleaved active SP-domain of TMPRSS2 was increased by chloroquine or hydroxychloroquine and this was potentiated by the MEKi's ( Figure 1A ). Inflammatory cytokine IL-6 was reduced by all 3 MEKi's (RAF/MEKi VS-6766 showed the greatest reduction in this experiment) with no benefit from addition of chloroquine or hydroxychloroquine.

We introduced an ACE2-promoter luciferase-reporter (ACE2-luc) in HCT116 human tumor cells to investigate whether MEKi could inhibit expression of ACE2 from its endogenous promoter.

Because of data that the anti-viral agent remdesivir could reduce hospitalization of COVID-19 infected patients [19] , we investigated the effects of combining remdesivir on ACE2 expression. Surprisingly, we found that in HCT116 cells, ACE2-luc reporter expression was increased by remdesivir, and this was attenuated by the addition of VS-6766 at non-toxic doses ( Figure 1B ).

Remdesivir also increased ACE2-Luc reporter activity in Calu-6 lung cancer cells at doses that did not reduce cell viability (Supplementary Figure 1A) . The increase in ACE2-Luc reporter activity following remdesivir treatment of HCT116 cells was observed with three different ACE2-Luc 

To extend the ACE2-Luc reporter studies, we evaluated the effects of remdesivir at 1 and 5 µM and VS-6766 (0.01, 0.1, and 1 µM) on ACE2 mRNA expression in HCT116 cells. We found a dose-dependent increase in ACE2 mRNA expression in HCT116 cells treated with remdesivir that was inhibited by addition of VS-6766 ( Figure 1B) . These results confirm that ACE2 mRNA expression was increased in remdesivir-treated HCT116 cells and this increase was attenuated at multiple doses of RAF/MEKi VS-6766.

To further evaluate the effects of MEKi plus remdesivir on ACE2 expression, we evaluated ACE2 protein expression in different human cell lines. We treated H1299, MST0211H, and Calu-3 cells 

Given our interest in using MEKi treatment as a strategy to suppress SARS-CoV-2 infectivity either alone or in the presence of anti-viral agent remdesivir, we tested nontoxic doses of the drugs in the various experiments that are shown. Data in support of the lack of toxicity of the drugs are shown in Supplementary Figures 6-8 .

We tested whether SARS-CoV-2 SPIKE protein could increase MAPK and ERK signaling as was shown previously with SARS coronavirus [39] . We added fragments of recombinant SPIKE protein to human cells in culture and observed an increase in ACE2 expression (as detected by the SC390851 antibody) in Calu-3 cells (Figure 2A ). ACE2 expression was suppressed by MEKi inhibitor treatment of Calu-3 non-small cell lung cancer cells (Figure 2A ). We further observed that in HT-29 colon cancer cells and BEAS-2B human bronchial airway epithelial cells, pERK was increased by SPIKE protein, while the addition of RAF/MEKi VS-6766 inhibited expression of both pERK and ACE2 ( Figure 2B ). Total ERK levels were unchanged in these cells ( Figure 2B ). The recombinant SPIKE fragments did not appear to increase pERK in the experiment in Calu-3 cells ( Figure 2A ).

To further investigate the correlation between MAPK-pERK activation and ACE2 expression we Figure 3A ). In MRC-5 normal human lung fibroblast cells, we observed an increase in pERK following remdesivir treatment under normal culture conditions or following serum stimulation of serum-deprived cells ( Figure 3B ). We used serum deprivation and restimulation as a strategy to modulate MAPK signaling to further explore the effects of remdesivir and MEKi. Both 8 pERK and ACE2 levels were suppressed by treatment of MRC-5 cells with VS-6766 alone or in combination with remdesivir in serum deprived or serum-deprived and subsequently serumstimulated MRC-5 cells ( Figure 3B ). An increase in pERK was observed in H460 and A549 lung cancer cells following remdesivir treatment and this was suppressed by VS-6766 alone or in combination with remdesivir ( Figure 3C , D). In Calu-3 cells, we observed an increase in pERK that was suppressed by addition of VS-6766 either following serum deprivation or serum stimulation of serum-deprived cells ( Figure 3E , right panel) while at later time points ACE2 expression was inhibited by VS-6766 plus remdesivir ( Figure 3E , left panels). Thus, pERK was increased by remdesivir treatment under multiple experimental conditions, and this was correlated with expression of ACE2 ( Figure 3 ). With VS-6766, both pERK and ACE were suppressed, including when VS-6766 was combined with remdesivir ( Figure 3 ).

We performed serum deprivation and re-stimulation to further test correlations between the status of activation of the MAPK pathway, proliferation state, and ACE2 expression. The data shows a correlation between phospho-Rb, and ACE2 expression ( Figure 3F -H). This is in addition to further replication of the effects of remdesivir and MEKi on ACE2 in these experiments where we observe correlations between pERK, pRb, and ACE2 ( Figure 3F ). Serum-starved cells showed a decrease in pRb levels compared to the 10% FBS control, which correlated with a decrease in ACE2 ( Figure 3F , lanes 1 and 5). An increase in pRb/total Rb was observed under serum stimulation, but with no increase in ACE2 ( Figure 3F , compare lanes 5 and 9). VS-6766 and the combination treatment decreased pERK, which under normal conditions correlated with a decrease in pRb relative to total Rb and a reduction in ACE2 levels. Remdesivir increased ACE2 expression under all conditions, most significantly under stimulation ( Figure 3F ). An increase in pRb relative to total Rb was observed upon serum stimulation and this correlated with ACE2 levels ( Figure 3H ). pERK correlation with ACE2 was heterogeneous, with a correlation seen in MCF7 cells but not H1299 or HCT-116 cells ( Figure 3H ). In H460 and HCT116 cells, pRb relative to total Rb decreased upon starvation and increased upon stimulation, which correlated with ACE2 expression ( Figure 3H , right panels). Thus, the results in Figure 3F show the triple correlation between pERK/pRb/ACE2 after MEKi and/or remdesivir treatment, and shows correlation of pRb/ACE2 when evaluating untreated cells before and after starvation. We quantified pRb and show as a fraction of total Rb ( Figure 3G ). Figure 3H similarly shows that under conditions of starvation/stimulation, there is a correlation between pRb and ACE2.

Since we observed that MEKi can inhibit ACE2 expression to potentially reduce SARS-CoV-2 infectivity, we further investigated effects of MEKi as well as remdesivir on Natural Killer (NK) cell activity against target cells. NK cells serve as a natural defense against transformed and virally 

As a pilot study, we set up a screening cytokine array to evaluate cytokine levels in nine COVID- 

To further evaluate the potential use of MEKi to suppress SARS-CoV-2 infectivity and disease severity, we evaluated the impact of VS-6766 on cytokine release by various human cell lines in culture. Using a Luminex-200 multiplexed cytokine array read-out, we observed reduced cytokine release following cell treatment by the MEKi either used alone or in combination with remdesivir ( Figure 5B -E). Cytokines that were suppressed include G-CSF, M-CSF, IL-1RA, MCP-1, IL-1a.

The TNF-related apoptosis-inducing ligand (TRAIL) is involved in killing transformed cells and virally-infected cells, with little inflammatory response. It is therefore important that any therapeutic agent being tested to control COVID-19 be monitored for effects on TRAIL-mediated killing of 

We developed a SARS-CoV-2 pseudovirus model system to investigate further our findings that MEKi may attenuate virus infectivity. We generated a pseudotyped SARS-CoV-2 virus which has a lentiviral core but with the SARS-CoV-2 D614 or G614 spike (S) protein on its envelope and used VSV-G lentivirus as a negative control ( Figure 6A ). Both SPIKE protein variants were expressed by the lentivirus ( Figure 6A ). The results suggest more G614 than D614 S protein was present on each viral particle. In order to establish an experimental model system for SARS-Cov- We performed an additional experiment evaluating the effects of the three MEK inhibitors on the D and G SPIKE variants in pseudovirus infectivity of human small airway epithelial cells (HSAECs).

All MEKi blocked pseudovirus cell entry of HSAECs in a very significant way, while having no effect on the pantropic VSV-G cell entry ( Figure 6H ). Overall cell survival was more than 75%.

The goal of this work was to identify candidate drugs that are available for clinical use that could be tested for their efficacy in suppressing the infectivity of SARS-CoV-2 as well as disease severity in preclinical studies. Our approach was to evaluate candidate small molecules for their effects on SARS-CoV-2 infectivity factors such as ACE2, and TMPRSS2 expression while We envisioned the potential to test these ideas in the clinic in patients with early COVID-19 infection where it may be possible to control the progression and spread of the infection throughout the body. Our hypotheses were that suppression of viral entry into the cell could be achieved through inhibition of ACE2 and TMPRSS2 expression, that blocking viral infection will reduce the spread of SARS-CoV-2 and allow the innate immune system and antivirals such as remdesivir to more effectively suppress viral infection, and that combinations of drugs that reduce ACE2 and TMPRSS2 may be helpful in addressing unexpected effects of remdesivir on ACE2 expression or hydroxychloroquine on active TMPRSS2.

We further explored the relationship between MEK and ACE2 given prior literature that SARS coronavirus SPIKE protein can increase ACE2 and stimulate the MAPK pathway and that MAPK pathway regulates ACE2. We reasoned that if MEKi impact on ACE2 expression, could they be used to attenuate early infection? Moreover, could MEKi be combined with remdesivir to improve its antiviral efficacy? We observed that several MEK inhibitors suppress ACE2 expression at nontoxic doses either alone or in the presence of remdesivir. We observed unexpectedly that under some experimental conditions, remdesivir increases ACE2 promoter activity, mRNA expression and protein expression, and this is suppressed when remdesivir is combined with MEKi such as VS-6766.

It is important to note and emphasize that there is variability and heterogeneity in the extent of In pursuit of a therapeutic agent that could attenuate cytokine storm while reducing viral infectivity and boosting NK cells activity, we found that VS-6766 decreases G-CSF and other cytokines.

These cytokines of interest were increased in COVID-19-(+) patient plasma samples in our study.

The combination of remdesivir and VS-6766 was not associated with increased cytokine expression at nontoxic doses of the drugs. The MEKi plus remdesivir drug combinations do not block NK-mediated cell killing and in fact the MEKi stimulate NK killing activity towards target cells.

Moreover, the drug combinations do not inhibit TRAIL-mediated killing of target cells. The observed stimulation of NK cell killing of target cells by MEK inhibitors is a novel finding that is relevant and important not only in the context of the current work focused on COVID-19 but also for cancer therapy mechanisms. Additional experiments will need to further evaluate the role of NK cells in anti-tumor efficacy in vivo, for example by using immune-depletion approaches [30] .

Our results support the idea that MEK inhibitors as a drug class may suppress COVID-19 infectivity factors while allowing (or in some cases boosting) NK-mediated (but not T-cell mediated) infection that may be of use as a therapeutic approach in patients with early or less severe COVID-19 disease.

COVID-19 (+) human plasma samples were received from the Lifespan Brown COVID-19 biobank at Rhode Island Hospital (Providence, Rhode Island). All patient samples were deidentified but with available clinical information as described in the manuscript. The IRB study protocol "Pilot Study Evaluating Cytokine Profiles in COVID-19 Patient Samples" did not meet the definition of human subjects research by either the Brown University or the Rhode Island Hospital IRBs.

Normal, healthy, COVID-19 (-) samples were commercially available form Lee BioSolutions (991-58-PS-1, Lee BioSolutions, Maryland Heights, Missouri). All samples were thawed and centrifuged to remove cellular debris immediately before the assay was ran. FBS. All cell lines were incubated at 37 °C in humidified atmosphere containing 5% CO2. Media containing 1% serum was used for serum starvation experiments. Cell lines were authenticated and tested to ensure the cultures were free of mycoplasma infection.

Cells were plated at 3.5 × 10 4 cells in a 48 well plate (Thermo Fisher Scientific, Waltham, MA) in CM and incubated at 37°C with 5% CO2. At 24 h after plating, almost all the tumor cells were adherent to the bottom of the flask and then the CM was completely replaced. Subsequently, the culture supernatants were collected after another 48 hr of incubation, centrifuged to remove cellular debris, and then frozen at -80°C until the measurement of cytokines was performed.

Human ACE2-Luc promoter constructs containing 1119, 252, and 202 base pairs of the ACE2 promoter linked to firefly luciferase were obtained from Addgene. Human tumor cell lines were transfected with each ACE2-promoter luciferase-reporter using lipofectamine 2000 as described in the protocol (Invitrogen, USA). At 24 hours after the transfection, the cells were trypsinized and seeded into 96-well black plates at a density of 3 × 10 4 cells/well. The next day, cells on the 96well plate were treated with drugs for 24 hours. The luciferase activity was evaluated by bioluminescence imaging in cells using the IVIS imaging system (Xenogen, Alameda, CA). DMSO treatment was used as a negative control in each screened plate. The bioluminescence in each treatment was normalized to DMSO treatment.

Cells were plated at a density of 3 x 10 3 cells per well of a 96-well plate. Cell viability was assessed using the CellTiter Glo assay (Promega). Cells were mixed with 25 µl of CellTiter-Glo reagents in 100 µl of culture volume, and bioluminescence imaging was measured using the Xenogen IVIS imager.

Immunoblotting for proteins was performed using the following antibodies: Cell Signaling 

Total RNA was isolated from cells using the RNeasy Mini Kit (Qiagen). Reverse transcription was performed with random primers using the SuperScript II First-Strand Synthesis System 

Cells were plated at a density of 5 x 10 5 per well of a 6-well plate. After 16 hours of incubation at 37 degrees Celsius in 5% CO2, cells were treated with 5 µM remdesivir (RDV), VS-6766, or the combination. After 24 hours, cells were treated with TRAIL (50 ng/mL) for an additional 4 hr.

Western blots evaluating cleaved caspase 8 were performed using Cell Signaling antibody (#9746). Ran was probed with BD Biosciences antibody (#610341) as a loading control.

We used a lentiviral packaging system to produce a replication incompetent SARS-CoV-2 pseudovirus. Briefly, HEK293T cells at 75% confluency were co-transfected with the backbone 

The statistical significance of differences between groups was determined using the Student t test.

The minimal level of significance was P < 0.05. Following symbols * and ** represent, P < 0.05 and P < 0.01, respectively.

Gallagher Table 5 . 

Clinical Characteristics of Coronavirus Disease 2019 in China

Clinical Characteristics of Covid-19 in

Epidemiology of Covid-19 in a Long-Term Care Facility

Comorbidity and its impact on 1590 patients with COVID-19 in China: a nationwide analysis

novel coronavirus (severe acute respiratory syndrome coronavirus 2) pandemic: A joint american college of academic international medicine-world academic council of emergency medicine multidisciplinary COVID-19 working group consensus paper

Clinical impact of COVID-19 on patients with cancer (CCC19): a cohort study. The Lancet

COVID-19 mortality in patients with cancer on chemotherapy or other anticancer treatments: a prospective cohort study. The Lancet

Clinical characteristics, outcomes, and risk factors for mortality in patients with cancer and COVID-19 in Hubei, China: a multicentre, retrospective, cohort study. The Lancet Oncology

Covid-19 in Critically Ill Patients in the Seattle Region -Case Series

Mini organs reveal how the coronavirus ravages the body

COVID-19 and Multiorgan Response. Current Problems in Cardiology

Neurological associations of COVID-19. The Lancet Neurology

Neurological and neuropsychiatric complications of COVID-19 in 153 patients: a UK-wide surveillance study. The Lancet Psychiatry

Management of acute kidney injury in patients with COVID-19. The Lancet Respiratory Medicine

COVID-19 update: Covid-19-associated coagulopathy

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

Tocilizumab in patients with severe COVID-19: a retrospective cohort study. The Lancet Rheumatology

Remdesivir for the Treatment of Covid-19 -Preliminary Report

A Randomized Trial of Convalescent Plasma for COVID-19-Potentially Hopeful Signals

Targeting the catecholamine-cytokine axis to prevent SARS-CoV-2 cytokine storm syndrome. medRxiv

Discovery of SARS-CoV-2 antiviral drugs through large-scale compound repurposing

Proteomics of SARS-CoV-2-infected host cells reveals therapy targets

The trinity of COVID-19: immunity, inflammation and intervention

Tracking Changes in SARS-CoV-2 Spike: Evidence that D614G Increases Infectivity of the COVID-19 Virus

Potential Immunotherapeutic Targets For Hypoxia Due to COVI-FLU. Shock

Natural killer cell specificity for viral infections

KILLER/DR5 is a DNA damage-inducible p53-regulated death receptor gene

Targeting apoptosis in cancer therapy

Dose intensification of TRAIL-inducing ONC201 inhibits metastasis and promotes intratumoral NK cell recruitment

AMG-232 sensitizes high MDM2-expressing tumor cells to T-cellmediated killing

Reduction of TRAIL-induced Mcl-1 and cIAP2 by c-Myc or sorafenib sensitizes resistant human cancer cells to TRAIL-induced death

TRAIL-R deficiency in mice promotes susceptibility to chronic inflammation and tumorigenesis

TRAIL pathway targeting therapeutics

Roles of natural killer cells in antiviral immunity

NK cells: A double edge sword against SARS-CoV-2. Advances in Biological Regulation

Tumor Necrosis Factor-Related Apoptosis-Inducing Ligand (Trail) Contributes to Interferon γ-Dependent Natural Killer Cell Protection from Tumor Metastasis

Antiviral response by natural killer cells through TRAIL gene induction by IFN-α/β

Upregulation of the Chemokine (C-C Motif) Ligand 2 via a Severe Acute Respiratory Syndrome Coronavirus Spike-ACE2 Signaling Pathway

The work was supported by a Brown University COVID- 19