key: cord-0854239-n4g9odh7
authors: Tao, Sijia; Zandi, Keivan; Bassit, Leda; Ong, Yee Tsuey; Verma, Kiran; Liu, Peng; Downs-Bowen, Jessica A.; McBrayer, Tamara; LeCher, Julia C.; Kohler, James J.; Tedbury, Philip R.; Kim, Baek; Amblard, Franck; Sarafianos, Stefan G.; Schinazi, Raymond F.
title: Comparison of anti-SARS-CoV-2 activity and intracellular metabolism of remdesivir and its parent nucleoside
date: 2021-08-12
journal: Current Research in Pharmacology and Drug Discovery
DOI: 10.1016/j.crphar.2021.100045
sha: 485ce4ce263da3ffcf5d3fe1dd26a947260b32ff
doc_id: 854239
cord_uid: n4g9odh7

Remdesivir, a monophosphate prodrug of nucleoside analog GS-441524, is widely used for the treatment of moderate to severe COVID-19. It has been suggested to use GS-441524 instead of remdesivir in the clinic and in new inhalation formulations. Thus, we compared the anti-SARS-CoV-2 activity of remdesivir and GS-441524 in Vero E6, Vero CCL-81, Calu-3, Caco-2 cells, and anti-HCoV-OC43 activity in Huh-7 cells. We also compared the cellular pharmacology of these two compounds in Vero E6, Vero CCL-81, Calu-3, Caco-2, Huh-7, 293T, BHK-21, 3T3 and human airway epithelial (HAE) cells. Overall, remdesivir exhibited greater potency and superior intracellular metabolism than GS-441524 except in Vero E6 and Vero CCL-81 cells.

Coronavirus disease 2019 , caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was declared a pandemic by the World Health Organization on March 11, 2020 (Cucinotta and Vanelli, 2020) . Ever since the initial outbreak in December 2019 in Wuhan, China, the global scientific community has been actively developing and testing therapeutic agents and vaccines against SARS-CoV-2 (Ciotti, et al., 2020; Nitulescu et al., 2020; Asselah et al., 2020) . Remdesivir was initially developed by Gilead Sciences as an anti-Ebola agent (Warren et al., 2016) . In record time, remdesivir was found to have anti-SARS-CoV-2 efficacy in vitro , and clinical trials were initiated and conducted to evaluate its safety and efficacy in patients (Eastman, et al., 2020) . On May 1, 2020, remdesivir was issued FDA Emergency Use Authorization for COVID-19 patients hospitalized with severe disease for its potential to shorten the time to recovery (FDA, 2020a) and on October 22, 2020, FDA approved remdesivir for intravenous administration in adult and pediatric patients (> 12 years old, > 40 kg) for treatment of COVID-19 requiring hospitalization (FDA, 2020b) . The drug does not appear to reduce mortality in patients with severe symptoms (Dyer, 2020; Beigel et al., 2020) and there is some controversy regarding its value for the treatment of COVID-19 by the World Health Organization (WHO) (Dyer, 2020) . It is worth noting that, unlike the Adaptive COVID-19 Treatment Trial-1 (ACTT-1) study which showed significant reduction in the duration of patient hospitalization, the WHO study was not performed under rigorous placebo controlled double-blind study conditions. J o u r n a l P r e -p r o o f As a McGuigan-type adenosine monophosphate prodrug, remdesivir is hydrolyzed to the corresponding monophosphate (GS-441524-MP) and then intracellularly phosphorylated to its active metabolite GS-441524-triphosphate (GS-443902) (Warren et al., 2016; (Fig. 1) . It has been reported that remdesivir exhibits a short plasma half-life of about 0.4 h in monkeys (Warren et al., 2016) , 1 h in humans (Humeniuk et al., 2020) , and that GS-441524 is the main metabolite. Interestingly, GS-441524 showed comparable efficacy to remdesivir against SARS-CoV-2 in Vero E6 African green monkey kidney cells and Calu3 2B4 human lung epithelial cells .

However, as GS-441524 has an easier synthetic route, potentially better safety profile, and may not require intravenous administration (Warren et al., 2016) , some researchers in the field are advocating for the alternate use of GS-441524 for patients with COVID-19 (Yan and Muller, 2020) . Thus, a side-by-side evaluation of these two compounds was warranted to determine advantages and differences between remdesivir and GS-441524.

In this study, we compared the anti-coronavirus activity of remdesivir and GS-441524 in Vero E6, Vero CCL-81, Calu-3, Caco-2 and Huh-7 cells. Furthermore, we compared the cellular pharmacology of these two compounds in Vero E6, Vero CCL-81, Calu-3, Caco-2, Huh-7, 293T, BHK-21, 3T3 and human airway epithelial (HAE) cells. Our results in cellular models provide a reference set of data for a comparison between remdesivir and GS-441524.

J o u r n a l P r e -p r o o f 

Remdesivir and GS-441524 were synthesized in house with a purity higher than 98%.

GS-443902 was purchased from MedChem Express (Monmouth Junction, NJ). All other reagents were the highest quality available from Thermo Fisher Scientific (Waltham, MA).

The African monkey kidney Vero E6 (ATCC  CRL-1586 TM ) and Vero (ATCC  CCL- USA-WA1/2020) and HCoV-OC43 was obtained from ATCC (Manassas, VA). Cytotoxicity was evaluated using MTS method in Vero CCL-81, Calu-3, Caco-2, and Huh-7 cells as previously described. (Zandi et al., 2020) .

Vero E6, Vero CCL-81, Calu-3, Caco-2, Huh-7, 293T, BHK-21, 3T3 and HAE cells were seeded at 1 x 10 6 cells per well in 12-well plates or 0.15 x 10 6 cells per well in 24well plates (HAE only) and incubated with 10 µM of each compound at 37 °C with 5% CO2. After 4 h, cells were washed twice by ice-cold PBS and resuspended in 70% icecold methanol overnight at -20 °C. The supernatants were then dried and reconstituted with HPLC mobile phase and then subjected to LC-MS analysis (Tao et al., 2019) . Levels of the phosphorylated metabolites 5'-mono-, di-, and triphosphate (GS-441524-MP, GS-441524-DP, and GS-443902) were measured. GS-441524-MP and GS-441524-DP were semi-quantified using the calibration curve for GS-443902.

Vero E6 and Vero CCL-81 cells are interferon deficient and do not secrete interferon alpha or beta when infected by viruses (Desmyter et al., 1968) . Hence, these two cell lines are commonly used as in vitro models for virus research. As they are also highly susceptible to SARS-CoV-2 infection , they have been used broadly to evaluate anti-SARS-CoV-2 compounds in vitro Zandi J o u r n a l P r e -p r o o f et al., 2020; Jeon et al., 2020; Fu et al., 200) . We compared the two compounds in both Vero E6 and Vero CCL-81 cells to see if there was any difference in these two cell lines in terms of activity. Calu-3 cells are used widely as an in vitro respiratory model for the SARS-CoV-2 infection Ko et al., 2021; Hoffmann et al., 2020a) .

In this study we also included Caco-2 cells since clinical reports indicate that individuals with COVID-19 can develop gastrointestinal symptoms (such as diarrhea) either alone or with respiratory symptoms (D'Amico et al., 2020) . Indeed, recent studies indicated that SARS-CoV-2 can infect the gastrointestinal tract because of high expression of the angiotensin-converting enzyme 2 (ACE2) receptor and serine protease TMPRSS2 that activates the virus to a fusogenic state in enterocytes (Hoffmann et al., 2020b) . Both of these host factors that are required for efficient SARS-CoV-2 infection are robustly expressed in Caco-2 cells (D'Amico et al., 2020) . The median effective concentration (EC50) and the concentration with 90% inhibitory effect (EC90) against in vitro replication of SARS-CoV-2 in Vero E6, Vero CCL-81, Calu-3, Caco-3 cells are shown in Table 1 . In both Vero E6 and Vero CCL-81 cells, remdesivir showed almost same potency as GS-441524, which is in agreement with previous reports ; and for each compound, similar potency were observed in Vero E6 and Vero CCL-81 cells. In Calu-3 cells, the EC50 value of GS-441524 was about 2 times higher than remdesivir, while interestingly, GS-441524 was about 80 times less potent than remdesivir in Caco-2 cells.

We further validated the higher potency of remdesivir versus GS-441524 in Caco-2 cells using icSARS-CoV-2-mNG inhibition assay (Fig. 2) (Xie et al., 2020) . Resultant mean EC50 of remdesivir from three independent experiments was 0.018 μM while EC50 J o u r n a l P r e -p r o o f of GS-441524 was 1.3 μM, 70-fold lower compared to remdesivir. These data independently confirmed the antiviral results in Caco-2 cells determined by q-RT PCR assay (Table 1) . It has been established that SARS-CoV-2 infects the liver and directly contributes to the liver injury in patients with COVID-19 . Therefore, we also used the human hepatocarcinoma Huh-7 cells to evaluate compounds against closely related virus HCoV-1 strain OC43. Remdesivir was active against this virus with an EC50 of 0.01 µM while GS-441524 was markedly less active with an EC50 of 4.1 µM. Of note, Huh-7 cells have lower susceptibility to SARS-CoV-2 infection than Vero E6 cells, and therefore are not always selected as an in vitro model for anti-SARS-CoV-2 assays .

Cytotoxicity was evaluated using MTS method in Vero CCL-81, Calu-3, and Caco-2 cells, and additionally Huh-7 cells, a cell line we use to routinely predict liver toxicity (Zandi et al., 2020) . No relevant toxicity was detected for either compound in Vero CCL-81, Calu-3 and Caco-2 cells. Interestingly, remdesivir was quite toxic in Huh-7 cells (Table 1) .

Notably, liver injury has been reported in remdesivir-treated COVID-19 patients (Zampino et al., 2020) . Fortunately, remdesivir is given intravenously bypassing first-pass metabolism in human liver and is only given for 10 days or less. 

In addition to assessing the potency of the two compounds in Vero E6, Vero CCL-81, Calu-3, Caco-2 and Huh-7 cells, we compared their intracellular metabolism in these five , and positively correlates with the higher level of GS-443902 formed in these cells from remdesivir. In

Caco-2 cells, the levels of GS-443902 formed from remdesivir were 67-fold higher (p < J o u r n a l P r e -p r o o f 0.001, t-test) than that from GS-441524 (Fig. 3) . Anti-SARS-CoV-2 activity of remdesivir in Caco-2 cells is usually measured by analyzing cytopathic effect (EC50 ranging from 0.1 to 0.8 µM (Meyer et al., 2020; Ellinger et al., 2021; Bojkova et al., 2020) ) or by assessing virus expression using RT-qPCR (EC50 = 46 nM) (Krüger et al., 2021) . In our hands, we observed a correlation between the higher potency of remdesivir vs GS-441524 (EC50 of 1 nM vs 80 nM, respectively, or 0.018 µM vs 1.3 µM by icSARS-CoV-2-mNG inhibition assay) in Caco-2 cells and the high levels of active NTP (308 vs 4.6 pmol/million cells, respectively) formed in this cell line. Interestingly, the levels of active NTP formed with GS-441524 only marginally fluctuate from cell line to cell line, which seemed to correlate with the relatively similar antiviral activity in Vero E6, Calu-3, and Caco-2 cells ( Table 1) .

On the other hand, remdesivir produced significantly more GS-443902 in Calu-3 and

Caco-2 cells than in Vero E6 cells (6-and 37-fold more, p < 0.001, t-test), a trend that follows the potency observed for that compound in the three different cell lines. We hypothesize that this difference is most likely due to a variation in the level of enzymes required to cleave remdesivir's prodrug moiety from one cell line to the other.

This difference in anti-HCoV-OC43 potency between remdesivir and GS-441524 (0.01 µM vs 4.1 µM) correlated with the levels of active NTP formed, since incubation of Huh-7 cells with 10 µM remdesivir for 4 h generated more than 120-fold higher level of metabolite GS-443902 than incubation with 10 µM GS-441524. Notably, remdesivir was quite toxic in Huh-7 cells (Table 1) , probably due to the very high intracellular levels (about 1,160 pmol/million cells) of active NTP formed (Fig. 3) .

In addition to Vero E6, Vero CCL-81, Calu-3, Caco-2 and Huh-7 cells, we also evaluated the intracellular metabolism of remdesivir and GS-441524 in 293T, BHK-21, 3T3 and HAE cells. 293T cell line is one of the most commonly used cell lines for in vitro SARS-CoV-2 studies because of its high expression of ACE2 and TMPRSS2 (Hoffmann et al., 2020b; Kumar et al., 2021) . BHK-21 and 3T3 cells are also essential cell lines used for vaccine production (Kumar et al., 2021) . Besides, BHK-21 cells support the replication of HCoV-OC43 (Shen et al., 2016) . 3T3 cells had previously been modified for SARS-CoV-2 studies by transfection with a SARS-CoV-2 spike and ACE2 glycoprotein remdesivir over 100-fold more (p < 0.001, t-test) than that from GS-441524. (Fig. 3) .

Finally, primary HAE cells have been used widely by researchers to study the SARS-CoV-2 Pizzorno et al., 2020; Hattori et al., 2021; Mulay et al. 2021 , Do et al., 2021 . Researchers compared the anti-SARS-CoV-2 activity of remdesivir and GS-441524 in HAE models and showed that remdesivir was more potent than GS-441524 in human tracheal airway cultures (Do et al., 2021) . In HAE cells at 4 h, remdesivir generated about 20-fold more (p < 0.001, t-test) GS-443902 than the parent nucleoside GS-441524 (Fig. 3) , and these levels positively correlated with the higher potency. It has been reported that inhalation formulations are being developed for pulmonary administration (Sahakijpijarn et at., 2020; Sahakijpijarn et al., 2021) and clinical trials underway. With the higher potency against SARS-CoV-2 in HAE cells and superior cellular metabolism, remdesivir could be a better option than GS-441524 in the development of the inhalable formulation.

Overall, the large differences in active NTP formation after incubation with remdesivir or GS-441524 in different cell lines probably results from the fact that remdesivir is a nucleoside phosphoramidate prodrug that can be directly metabolized intracellularly to GS-441524 monophosphate, therefore bypassing the often-limiting monophosphorylation J o u r n a l P r e -p r o o f step (Varga et al., 2016) . It is also worth noting that, if only passive permeation is considered, remdesivir, with an estimated LogP of 3.2 (ChemDraw Professional 16.0 -PerkinElmer), is more likely to effectively cross the hydrophobic phospholipid bilayer than the parent nucleoside GS-441524 (estimated LogP = -1.65) and therefore display higher potency.

It is worth noting that during the preparation of this manuscript, Li et al. reported that GS-441524 effectively inhibits SARS-CoV-2 infection in a mouse model .

Their reported in vitro data seems consistent with ours except that we observed a far greater difference in EC50 between GS-441524 and remdesivir in Caco-2 cells (80-fold vs 6-fold). It is not surprising to see some in vivo efficacy in a mouse model (transduced BALB/c mice with adenovirus associated virus vector expressing hACE2) with dosing at 25 mg/kg per day (ip for 8 days) started one day before infection. Based on that study, and without a side-by-side mouse experiment with remdesivir, no definitive conclusion can be made regarding a potential advantage of GS-441524 over remdesivir in mice.

In summary, we systematically evaluated the anti-coronavirus activity of remdesivir and its parent nucleoside GS-441524 in Vero E6, Vero CCL-81, Calu-3, Caco-2, and Huh-7 cells, and intracellular pharmacology of these two compounds in Vero E6, Vero CCL-81, Calu-3, Caco-2, Huh-7, 293T, BHK-21, 3T3 and HAE cells. Except for Vero E6 and Vero CCL-81 cells, remdesivir demonstrated greater antiviral potency and markedly higher phosphorylation efficiency than GS-441524 in all cells, especially in Caco-2 and J o u r n a l P r e -p r o o f Huh-7 cells. These results show that the relative potency of the nucleoside prodrug and the parent nucleoside is cell-type dependent, emphasizing the fact that cell model selection is highly important for evaluation of antiviral activity, especially with nucleoside analogs (Mumtaz et al., 2017) . Ideally, the best nucleoside analogs are those that are active in most susceptible relevant cell culture systems. Because of the broader phosphorylation profile and higher potency of remdesivir compared to GS-441524 in the several cell systems we studied, we anticipate that remdesivir will likely continue to be a better clinical option than GS-441524.

COVID-19: discovery, diagnostics and drug development

ACTT-1 Study Group Members, 2020. Remdesivir for the treatment of Covid-19 -final report

SARS-CoV-2 and SARS-CoV differ in their cell tropism and drug sensitivity profiles

The COVID-19 pandemic

WHO declares COVID-19 a pandemic

Diarrhea during COVID-19 infection: pathogenesis, epidemiology, prevention, and management

Defectiveness of interferon production and of rubella virus interference in a line of African green monkey kidney cells (Vero)

A robust SARS-CoV-2 replication model in primary human epithelial cells at the air liquid interface to assess antiviral agents

Covid-19: remdesivir has little or no impact on survival, WHO trial shows

Remdesivir: A review of its discovery and development leading to emergency use authorization for treatment of COVID-19

A SARS-CoV-2 cytopathicity dataset generated by high-content screening of a large drug repurposing collection

Remdesivir EUA letter of authorization

FDA approves first treatment for COVID-19

Both boceprevir and GC376 efficaciously inhibit SARS-CoV-2 by targeting its main protease

A small molecule compound with an indole moiety inhibits the main protease of SARS-CoV-2 and blocks virus replication

Chloroquine does not inhibit infection of human lung cells with SARS-CoV-2

SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor

Safety, tolerability, and pharmacokinetics of remdesivir, an antiviral for treatment of COVID-19, in healthy subjects

Identification of antiviral drug candidates against SARS-CoV-2 from FDA-approved drugs

Comparative analysis of antiviral efficacy of FDA-approved drugs against SARS-CoV-2 in human lung cells

Drug inhibition of SARS-CoV-2 replication in human pluripotent stem cell-derived intestinal organoids

Clinically relevant cell culture models and their significance in isolation, pathogenesis, vaccine development, repurposing and screening of new drugs for SARS-CoV-2: a systematic review

Remdesivir metabolite GS-441524 effectively inhibits SARS-CoV-2 infection in mice models

Key metabolic enzymes involved in remdesivir activation in human lung cells

Lack of antiviral activity of darunavir against SARS-CoV-2

SARS-CoV-2 infection of primary human lung epithelium for COVID-19 modeling and drug discovery

Cell-line dependent antiviral activity of sofosbuvir against Zika virus

Comprehensive analysis of drugs to treat SARS-CoV-2 infection: mechanistic insights into current COVID19 therapies

N-Glycan Modification in Covid-19 Pathophysiology: In vitro structural changes with limited functional effects

Characterization and treatment of SARS-CoV-2 in nasal and bronchial human airway epithelia

Remdesivir inhibits SARS-CoV-2 in human lung cells and chimeric SARS-CoV expressing the SARS-CoV-2 RNA polymerase in mice

Development of remdesivir as a dry powder for inhalation by thin film freezing

In vivo pharmacokinetic study of remdesivir dry powder for inhalation in hamsters

An orally bioavailable broad-spectrum antiviral inhibits SARS-CoV-2 in human airway epithelial cell cultures and multiple coronaviruses in mice

Safe and sensitive antiviral screening platform based on recombinant human coronavirus OC43 expressing the luciferase reporter gene

Intracellular metabolism and potential cardiotoxicity of a β-D-2'-C-methyl-2,6-diaminopurine ribonucleoside phosphoramidate that inhibits hepatitis C virus replication

Intracellular metabolism of nucleoside/nucleotide analogues: a bottleneck to reach active drugs on HIV reverse transcriptase

Remdesivir and chloroquine effectively inhibit the recently emerged novel coronavirus (2019-nCoV) in vitro

SARS-CoV-2 infection of the liver directly contributes to hepatic impairment in patients with COVID-19

An infectious cNDA clone of SARS-CoV-2

Advantages of the parent nucleoside GS-441524 over remdesivir for Covid-19 treatment

Repurposing nucleoside analogs for human coronaviruses

Liver injury in remdesivir-treated COVID-19 patients

Acknowledgements: This work was supported by NIH grants R01-AI-141327 (BK and RFS), and in part by the Emory Center for AIDS Research P30-AI-050409 (RFS) and R01-AI-121315 (SGS) grants. The Nahmias-Schinazi Chair fund is also acknowledged (SGS). We thank World Reference Center for Emerging Viruses and Arboviruses (WRCEVA) and Dr. Pei-Yong Shi from the University of Texas Medical Branch for providing icSARS-CoV-2-mNG. This paper is dedicated to the memory of our friend and colleague Dr. John C. Martin whose leadership in the field of antiviral agents led to the development of remdesivir for Ebola and more recently for COVID-19.