Stenoparib, an inhibitor of cellular poly (ADP-ribose) polymerases (PARPs), blocks in vitro replication of SARS-CoV-2 variants.

We recently published a preliminary assessment of the activity of a poly (ADP-ribose) polymerase (PARP) inhibitor, stenoparib, also known as 2X-121, which inhibits viral replication by affecting pathways of the host. Here we show that stenoparib effectively inhibits a SARS-CoV-2 wild type (BavPat1/2020) strain and four additional variant strains; alpha (B.1.1.7), beta (B.1.351), delta (B.1.617.2) and gamma (P.1) in vitro, with 50% effective concentration (EC50) estimates of 4.1 μM, 8.5 μM, 24.1 μM, 8.2 μM and 13.6 μM, respectively. A separate experiment focusing on a combination of 10 μM stenoparib and 0.5 μM remdesivir, an antiviral drug, resulted in over 80% inhibition of the alpha variant, which is substantially greater than the effect achieved with either drug alone, suggesting at least additive effects from combining the different mechanisms of activity of stenoparib and remdesivir.

Introduction

As of June 2022, the coronavirus disease (COVID-19) pandemic, caused by SARS-CoV-2, has caused over 534 million infections and over 6.3 million deaths worldwide [1]. Although protective vaccines are available, the pandemic continues, and both old and new SARS-CoV-2 variants exhibit varying degrees of disease severity and resistance to vaccination. Additional effective therapeutics are urgently needed. Here we describe the antiviral activity of a small molecule, stenoparib, an inhibitor of mammalian poly (ADP-ribose) polymerases (PARPs). We show that stenoparib effectively inhibits replication of SARS-CoV-2 wild-type (wt) and variant strains in vitro.

SARS-CoV-2 has undergone adaptation and mutation since the beginning of the COVID-19 pandemic, resulting in new variants of the virus. Only two antiviral drugs, remdesivir and molnupiravir, or treatment with monoclonal antibodies, have been approved by the United States Food and Drug Administration as COVID-19 therapies under the Emergency Use Authorization [2, 3]. Remdesivir and molnupiravir are nucleoside analogs that affect the activity of the RNA-dependent RNA polymerase (RdRp). After incorporation into viral RNA, remdesivir stalls the RdRp during elongation [4], while molnupiravir results in the incorporation of mutations that can be functionally deleterious [5].

We recently published a study on the activity of a poly (ADP-ribose) polymerase (PARP) inhibitor, stenoparib, also known as 2X-121, which inhibits viral replication by affecting pathways of the host [6] as opposed to targeting viral replication. ADP-ribosylation (ADPR) pathways may have either anti- or pro-viral properties, and their importance in host-virus interactions is becoming increasingly recognized [7]. Unlike remdesivir, which inhibits viral replication downstream of entry into the cell, stenoparib inhibits virus entry and post-entry processes [6]. Stenoparib inhibited the SARS-CoV-2 USA-WA1/2020 virus and the HCoV-NL63 human seasonal respiratory coronavirus in vitro, exhibiting dose-dependent suppression of virus multiplication and cell-cell spread in cell culture [6]. Stenoparib exhibits a unique dual activity against the PARP1, 2 and PARP5a, 5b (tankyrase 1, 2) enzymes, which are important intermediates in the Wnt/β-catenin immune checkpoint [8, 9]. As a host-targeting therapeutic that does not directly select for resistance in viruses, we hypothesize that stenoparib should be able to inhibit all SARS-CoV-2 variant strains.

Materials and methods

Cell culture

For the ViroSpot reduction assays (performed in The Netherlands), the antiviral activity of stenoparib (catalog # S8419, Selleckchem Chemicals, Radnor, PA, USA) on SARS-CoV-2 was assessed in vitro using Vero E6 Cercopithecus aethiops kidney cells (catalog # CRL-1586, ATCC, Manassas, VA, USA, acquired September 2020) maintained in Dulbecco’s Modified Eagle Medium (DMEM; catalog # BE12-733F, Lonza, Basel, Switzerland) supplemented with 2 mM L-glutamine (catalog # BE17-605E, Lonza, Basel, Switzerland), 100 U/ml penicillin and 100 μg/mL streptomycin (catalog # DE17-602E, Lonza, Basel, Switzerland), and 3% fetal bovine serum (FBS; catalog # FBS-12A, Capricorn Scientific, Ebsdorfergrund, Germany).

For the cytotoxicity assays and plaque assays (performed in the United States), the cytotoxicity and antiviral activity of stenoparib on SARS-CoV-2 was assessed in vitro using Vero E6 Cercopithecus aethiops kidney cells (catalog # CRL-1586, ATCC, Manassas, VA, USA, acquired September 2020) maintained in Eagle’s Minimum Essential Medium (EMEM; catalog # 30–2003, ATCC, Manassas, VA, USA) supplemented with 2% or 10% FBS, 100 U/mL penicillin and 100 μg/mL streptomycin (catalog # P0781, Sigma Aldrich, St. Louis, MO, USA), 0.01 M HEPES solution (catalog # H0887, Sigma Aldrich, St. Louis, MO, USA), 1 mM sodium pyruvate (catalog # 11360070, ThermoFisher Scientific, Waltham, MA, USA), and 1× non-essential amino acids solution (catalog # SH3023801, Fisher Scientific, Waltham, MA, USA).

ViroSpot reduction assay

The activities of stenoparib and remdesivir (catalog # 30354, Cayman Chemical Company, Ann Arbor, MI, USA) were assessed in a BSL3 facility against a wt SARS-CoV-2 strain (BavPat1/2020), and four additional SARS-CoV-2 variants of concern; alpha (B.1.1.7), beta (B.1.351) delta (B.1.617.2), and gamma (P1), hereafter referred to using Greek nomenclature (Table 1). We performed ten, serial 2-fold dilutions of compound, mixed with 100 plaque-forming units (pfu) of virus, and added the mixture to 80% confluent Vero E6 cells growing in multi-well plates. The Vero E6 cells were fixed 20 hours (h) after infection and stained with a SARS-CoV/SARS-CoV-2 Nucleocapsid monoclonal antibody (Sino Biological, Wayne, PA, USA) followed by a horseradish peroxidase (HRP)-labeled Goat anti-Mouse IgG (catalog # G-21040, ThermoFisher, Waltham, MA, USA) and TrueBlue Peroxidase Substrate (catalog # 5510–0030, Seracare, Milford, MA, USA). Spots were counted using a CTL ImmunoSpot Image Analyzer (Cleveland, OH, USA) as previously described [10]. Percent inhibition was calculated as percent of the well surface area covered by virus-positive immunostaining, and was normalized by taking the average fraction of the wells that showed positive staining compared to the control wells with the highest positive staining [10]. The EC50 values were approximated and compared in a pair-wise fashion using package ‘drc 3.0–1’ in R version 4.1.1 [11-13]. We calculated means and standard deviations using package ‘matrixStats’ version 0.62.0 and ‘psych’ version 2.2.5 in R version 4.1.1[13-15]. Mean normalized percent inhibition and dose-response curves were plotted using Hill coefficient, minimum, maximum, and EC50 parameter estimates from package ‘drc’ version 3.0–1 and the Hill equation in R version 4.1.1 [11–13, 16].

Table 1

SARS-CoV-2 variants used in ViroSpot reduction and plaque assays to evaluate antiviral properties of stenoparib and remdesivir.

Variant	Pango lineage	Strain	Isolate	Source
wt	-	BavPat1/2020	Germany	European Virus Archive–GLOBAL;
				026V-03883
alpha	B.1.1.7	hCoV-19/USA/CA_CDC_5574/2020	California, USA	BEI Resources;
				NR-54011
beta	B.1.351	hCoV-19/USA/MD-HP01542/2021	Maryland, USA	BEI Resources;
				NR-55282
delta	B.1.617.2	hCoV-19/USA/PHC658/2021	Tennessee, USA	BEI Resources;
				NR-55611
gamma	P.1	Isolate hCoV-19/Japan/TY7-503/2021	Japan	BEI Resources;
				NR-54982

The wt, alpha, beta, delta, and gamma strains were used in the ViroSpot reduction assay. The alpha strain was used in the plaque assays. The delta strain carries an ORF7a deletion which may have the potential to impact virulence [17]. The following reagents were obtained through BEI Resources, NIAID, NIH: SARS-Related Coronavirus 2, Isolate USA/CA_CDC_5574/2020, NR-54011, deposited by the Centers for Disease Control and Prevention; SARS-Related Coronavirus 2, Isolate hCoV-19/USA/MD-HP01542/2021 (Lineage B.1.351), in Homo sapiens Lung Adenocarcinoma (Calu-3) Cells, NR-55282, contributed by Andrew S. Pekosz; SARS-Related Coronavirus 2, Isolate hCoV-19/USA/PHC658/2021 (Lineage B.1.617.2; Delta Variant), NR-55611, contributed by Dr. Richard Webby and Dr. Anami Patel; and SARS-Related Coronavirus 2, Isolate hCoV-19/Japan/TY7-503/2021 (Brazil P.1), NR-54982, contributed by National Institute of Infectious Diseases.

Plaque assay

Vero E6 cells were infected with the SARS-CoV-2 alpha variant in a BSL3 facility using a multiplicity of infection (MOI) of 0.1, and the activity of stenoparib was assessed with and without remdesivir using a plaque reduction assay. Following viral infection and treatment with stenoparib and/or remdesivir, cells were overlaid with low melting point agarose (catalog # 1613112, BioRad, Hercules, CA, USA), fixed at 120 h post infection with 4.0% paraformaldehyde (catalog # AAJ19943K2, Fisher Scientific, Waltham, MA, USA), stained with crystal violet (catalog # V5265, Sigma Aldrich, St. Louis, MO, USA), and plaques were manually counted. We used Welch’s Two Sample t-tests in R version 4.1.1 [13] to perform pairwise t-tests between the untreated wells and wells under all other treatment conditions. Standard deviations and error bars were plotted using packages ‘matrixStats’ and ‘psych’ in in R version 4.1.1 [13-15].

Cytotoxicity assay

Cytotoxicity was measured using the Promega CytoTox 96 Non-Radioactive Cytotoxicity Assay (catalog # G1780, Promega, Madison, WI, USA) in 50 μL reactions and 96-well format (catalog # 161093, Thermo Scientific, Waltham, MA, USA) according to the manufacturer’s protocol. The assay was measured using a BioTeK Synergy HT plate reader (model # 7091000, BioTek, Winooski, VT, USA) set at 490 nm wavelength. Vero E6 cells were assessed for cytotoxicity at 5 days post-treatment. Percent cytotoxicity was calculated by dividing the experimental LDH release at 490 nm wavelength (OD490) by the maximum LDH release control (lysis) and multiplying by 100. Means and standard deviations were calculated and plotted using packages ‘matrixStats’ and ‘psych’ in in R version 4.1.1[13-15].

Results

We explored the activity of stenoparib against a wt SARS-CoV-2 strain and four variant strains alpha, beta, delta, and gamma (Table 1). Inhibition of virus replication by stenoparib was dose-dependent, with 50% effective concentration (EC50) estimates ranging from 4.1 μM to 24.1 μM across the five virus strains tested (Fig 1A). Interestingly, the EC50 estimate for stenoparib and the beta variant (24.1 μM) was 2.8- to 5.8-fold higher than for the wt, alpha, and delta variants, which were 4.1 μM, (p = 1.19×10−4); 8.5 μM, (p = 7.96×10−9); and 8.2 μM, (p = 0.06; S1 and S2 Figs). Similarly, the EC50 estimate for stenoparib against the gamma variant (13.6 μM) was lower than for the beta variant (24.1 μM), however this difference was not statistically significant (p = 0.22; S1 and S2 Figs). An analogous phenomenon was noted for remdesivir. The beta EC50 estimate (9.9 μM) for remdesivir was 2.1- to 4.3-fold higher than for the wt, alpha, delta, and gamma variants; 3.8 μM, (p = 2.76×10−5); 4.7 μM, (p = 2.05×10−11); 2.3 μM, (p = 4.97×10−3); and 4.1 μM (p = 2.32×10−4), respectively (Fig 1B, S1 and S2 Figs). The standard deviation around the percent inhibition estimates was high for the beta variant, however, so the dose-response correlation may not be precisely calculated in this case (Fig 1B). The ViroSpot experiment was conducted using technical quadruplicates, and infected and untreated cells were used as controls.

Fig 1

Dose-response curves for stenoparib and remdesivir on wild type SARS-CoV-2 and four variant SARS-CoV-2 strains.

Normalized percent inhibition (Y-axis) of wt SARS-CoV-2 and variant strains alpha, beta, delta, and gamma in Vero E6 cells treated with (A) stenoparib or (B) remdesivir at the indicated concentrations (X-axis). Error bars are the standard deviation.

In an additional set of experiments, we focused on the SARS-CoV-2 alpha variant. Vero E6 cells were infected with the alpha variant using a multiplicity of infection of 0.1, and the activity of stenoparib was assessed with and without remdesivir using the plaque reduction assay. Plaques are areas of dead or destroyed cells and appear as small, clear regions in an infected cell monolayer after fixation and staining with crystal violet. We combined 2.5, 5.0, and 10 μM doses of stenoparib with the previously reported EC50 of remdesivir (0.5 μM) [6]. The three control types were 1) infected and untreated cells, 2) uninfected and untreated cells, and 3) infected cells treated with a combination of camostat mesylate and aloxistatin (E64d) (combination hereafter referred to as Camostat-E64d, or ‘CE’), which are protease inhibitors that prevent Spike (S) protein cleavage and virus entry into the cell [12].

As shown in Fig 2A and 2B, neither 10 μM stenoparib nor 0.5 μM remdesivir achieved greater than a 50% reduction in plaquing efficiency compared to the infected, untreated cells. When combined, however, plaque formation was reduced by over 80% compared to infected, untreated cells. This reduction was superior to what was achievable with stenoparib (p = 3.03×10−5) or remdesivir (p = 1.99×10−4) alone at these doses. Notably, cytotoxicity remained near baseline levels (Fig 2C). The cytotoxicity experiments were run in triplicate and cells were treated with CE, dimethylsulfoxide (DMSO), lysis buffer, or left untreated (no compound) as controls. When combined, the activity of two or more drugs with different mechanisms of activity may synergize, with the potential benefit of reducing individual doses of each drug and minimizing undesirable side effects in the patient.

Fig 2

Activity of stenoparib and remdesivir on the alpha SARS-CoV-2 variant and cytotoxicity of stenoparib and remdesivir in Vero E6 cells.

(A) Virus plaque assays. Vero E6 cells were infected for 2 hours with SARS-CoV-2 alpha variant, washed, and overlaid with low-melting temp agarose/MEM with or without stenoparib and/or remdesivir at the concentrations indicated. The cell monolayers were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet prior to manual counting of plaques. The experiment was performed using technical triplicates and error bars represent the standard deviation of three independent experiments. For pairwise comparisons to untreated cells, * indicates p-values less than 0.05 and ** indicates p-values less than 0.01. (B) Photographs of plaque assay plate wells after crystal violet staining. Labels indicate the concentrations of stenoparib or remdesivir. Abbreviations: S = stenoparib, R = remdesivir at the concentrations indicated (i.e. 2.5 / 0.5 = 2.5 μM stenoparib, 0.5 μM remdesivir). (C) Cytotoxicity measurements performed by the lactate dehydrogenase release assay (Promega Cytotox 96) 48 hours after infection. Inhibitors stenoparib or remdesivir were used at the indicated concentrations (X-axis). The assays were performed according to the manufacturer’s recommendations. Results from drug treatments were normalized to the fraction of maximum cytotoxicity (% Cytotoxicity; Y-axis) achieved by detergent lysis of cells. The samples were processed in triplicate and error bars represent the standard deviation of five independent experiments.

Discussion

According to a recent report, the beta variant appears to exhibit a significantly reduced eclipse period (length of time between initial infection and the production of virus by a cell) and more rapid replication in vitro compared to the alpha variant [17]. This could explain the relatively higher stenoparib and remdesivir EC50 estimates for the beta variant. In addition to numerous changes to the S protein, the beta variant carries some unique mutations in the N protein and in the Nsp3 polyprotein (ORF1a) that are not found in the other variants [18], and may be involved in viral degradation pathways including ADPR following infection [19]. Differential susceptibility to ADPR may be related to the altered replication kinetics of the beta variant, although verifying this requires further investigation.

In this study, the activity of remdesivir and stenoparib against SARS-CoV-2 was examined in Vero E6 African green monkey kidney cells, which are highly permissive to infection with SARS-CoV-2 [20]. Our previous stenoparib study utilized Vero E6 and Calu-3 human lung adenocarcinoma cells, which were less sensitive than Vero E6 to the toxic effects of stenoparib at higher concentrations and the extended incubation times (120 h) required for the plaque assay [6]. Host cell lines vary in their susceptibility to infection and in their response to treatment with antivirals. Additional studies using other host cell lineages, including primary respiratory epithelial cells, should be instructive regarding the efficacy of PARP inhibitors as host-targeting antiviral therapeutics.

Besides the known roles for the PARP1 and PARP2 enzymes in DNA repair [21, 22], members of the PARP family have numerous additional functions [23]. The 18 known human PARPs appear to differentially affect viral replication; some exhibit proviral activity and others exhibit antiviral roles [7, 24–28]. In coronavirus, the N protein is mono (ADP) ribosylated (MARylated) during infection. Stability of the N protein is critical for viral genome replication, translation, packaging, and modulation of the host cell cycle [29]. The N protein is the only ADPR target definitively identified in coronavirus thus far [24], and the prevalence of this modification across multiple coronavirus families suggests an important role in virus stability [24]. Despite this, the role of N protein MARylation, and of ADPR in general with regards to coronavirus stability, is not well characterized. Possibly ADPR has a role in regulating virus genome structure, like ADPR of histones or the adenovirus core proteins [30, 31]. While remdesivir and molnupiravir inhibit viral RNA replication, stenoparib likely acts through multiple targets. A combination of stenoparib and remdesivir or molnupiravir may be highly effective at inhibiting SARS-family coronaviruses including SARS-CoV-2.

Here, we showed that stenoparib is broadly inhibitory to SARS CoV-2, including emerging variants of concern. We expect these results to apply for additional, emergent variants as well, and recent preliminary data with stenoparib and the omicron variant supports this. A host-targeting therapeutic like stenoparib could be a significant benefit for COVID-19 patients as a standalone therapy, or as part of a combinatorial COVID-19 treatment strategy with an antiviral drug such as remdesivir or molnupiravir. Combinations of two or more drugs may lead to synergism through different mechanisms of action, which has the potential benefit of reducing individual doses of each drug and minimizing undesirable side effects.

Stenoparib and remdesivir EC50 estimates.

Stenoparib (gold shading) and remdesivir (blue shading) EC50 estimates for Vero E6 cells infected with SARS-CoV-2 wt, alpha, beta, delta, or gamma variants.

(TIF)

Click here for additional data file.

P-values for SARS-CoV-2 variant pairwise comparisons of stenoparib and remdesivir EC50 estimates.

Stenoparib (above the diagonal, gold shading) and remdesivir (below the diagonal, blue shading) EC50 pairwise comparisons for SARS-CoV-2 wt, alpha, beta, delta, or gamma variants.

(TIF)

Click here for additional data file.

Data from ViroSpot assays, plaque assays, and cytotoxicity assays.

(XLSX)

Click here for additional data file.