key: cord-0789681-bk0ggrin
authors: Froggatt, Heather M.; Heaton, Brook E.; Heaton, Nicholas S.
title: Development of a fluorescence based, high-throughput SARS-CoV-2 3CLpro reporter assay
date: 2020-06-24
journal: bioRxiv
DOI: 10.1101/2020.06.24.169565
sha: 2b30c1de1de9f5b0865ee866723aacb5f6a4d66e
doc_id: 789681
cord_uid: bk0ggrin

In late 2019 a human coronavirus, now known as SARS-CoV-2, emerged, likely from a zoonotic reservoir. This virus causes COVID-19 disease, has infected millions of people, and has led to hundreds of thousands of deaths across the globe. While the best interventions to control and ultimately stop the pandemic are prophylactic vaccines, antiviral therapeutics are important to limit morbidity and mortality in those already infected. At this time, only one FDA approved anti-SARS-CoV-2 antiviral drug, remdesivir, is available and unfortunately, its efficacy appears to be limited. Thus, the identification of new and efficacious antivirals is of highest importance. In order to facilitate rapid drug discovery, flexible, sensitive, and high-throughput screening methods are required. With respect to drug targets, most attention is focused on either the viral RNA-dependent RNA polymerase or the main viral protease, 3CLpro. 3CLpro is an attractive target for antiviral therapeutics as it is essential for processing newly translated viral proteins, and the viral lifecycle cannot be completed without protease activity. In this work, we present a new assay to identify inhibitors of the SARS-CoV-2 main protease, 3CLpro. Our reporter is based on a GFP-derived protein that only fluoresces after cleavage by 3CLpro. This experimentally optimized reporter assay allows for antiviral drug screening in human cell culture at biosafety level-2 (BSL2) with high-throughput compatible protocols. Using this screening approach in combination with existing drug libraries may lead to the rapid identification of novel antivirals to suppress SARS-CoV-2 replication and spread. IMPORTANCE The COVID-19 pandemic has already led to more than 400,000 deaths and innumerable changes to daily life worldwide. Along with development of a vaccine, identification of effective antivirals to treat infected patients is of the highest importance. However, rapid drug discovery requires efficient methods to identify novel compounds that can inhibit the virus. In this work, we present a method for identifying inhibitors of the SARS-CoV-2 main protease, 3CLpro. This reporter-based assay allows for antiviral drug screening in human cell culture at biosafety level-2 (BSL2) with high-throughput compatible sample processing and analysis. This assay may help identify novel antivirals to control the COVID-19 pandemic.

The COVID-19 pandemic has already led to more than 400,000 deaths and innumerable changes 46 to daily life worldwide. Along with development of a vaccine, identification of effective antivirals 47 to treat infected patients is of the highest importance. However, rapid drug discovery requires 48 efficient methods to identify novel compounds that can inhibit the virus. In this work, we present 49 a method for identifying inhibitors of the SARS-CoV-2 main protease, 3CL pro . This reporter-based 50 assay allows for antiviral drug screening in human cell culture at biosafety level-2 (BSL2) with 51 high-throughput compatible sample processing and analysis. This assay may help identify novel 52 antivirals to control the COVID-19 pandemic. 53

In December 2019, a novel human coronavirus (hCoV) was identified in the Hubei Province of 56

China (1-3). The virus, now known as SARS-CoV-2, causes the transmissible and pathogenic 57 disease COVID-19 (4). COVID-19 has become a global pandemic and infected over 8 million 58 people and caused ~500,000 deaths to date (5). Current efforts to control COVID-19 are largely 59 focused on behavioral modifications such as social distancing and the use of masks (6). These 60 approaches attempt to slow the spread of the virus, but meaningful control of the virus will 61 ultimately be the result of a combination of efficacious vaccines and antiviral therapeutics (7). 62 63 Antiviral therapeutics aim to disrupt the replication cycle and reduce viral load in infected 64 individuals. Therapeutic development efforts have led to a number of candidate antiviral 65 compounds focused mainly on two essential viral enzymes, the RNA-dependent RNA polymerase 66 (RdRp) and the viral proteases. Remdesivir (GS-5734), recently FDA approved as an antiviral for 67 SARS-CoV-2, targets the polymerase to suppress hCoV replication by inducing termination of 68 RNA polymerization (8); however, the benefits of this drug in clinical trials and early use appear 69 limited (9). Another nucleoside analogue, β-D-N 4 -hydroxycytidine (NHC; EIDD-1931), also 70 inhibits SARS-CoV-2 polymerase activity, likely via inducing lethal mutagenesis of the viral 71 genome (10). In addition to the RdRp, the viral proteases, which are critical to liberate individual 72 viral proteins from the polyprotein produced by initial genome translation, present another 73 attractive drug target. For SARS-CoV-2, lopinavir/ritonavir, a protease inhibitor combination, is 74 shown to interact with the main coronavirus protease, known as 3CL pro or M pro (11); however, 75 early clinical trial results with these compounds have shown no significant benefits to SARS-CoV-76 antivirals targeting the SARS-CoV-2 protease, 3CL pro (13-15). At this time, these newly designed 78 compounds are in the early stages of testing. Thus, the discovery of additional effective SARS-79

CoV-2 antiviral drugs remains of high importance. The identification (and subsequent 80 improvement) of novel drugs targeting SARS-CoV-2 will require robust and high-throughput 81 screening approaches. 82

Here, we report the development and validation of a fluorescent reporter optimized to detect 84 SARS-CoV-2 3CL pro activity. This assay is performed in human cell culture and does not require 85 biosafety level 3 (BSL3) containment. Our reporter is based on FlipGFP, which only fluoresces 86 after protease mediated activation (16). We generated and tested three reporter constructs with 87 distinct cleavage target sequences for activation by the SARS-CoV-2 3CL pro . We also show that 88 the reporter with the best signal-to-noise ratio for SARS-CoV-2 is also activatable by other 89 coronavirus 3CL pro proteins across subgroups (beta, alpha, gamma) and host species (human, 90 rodent, bird). Finally, we used this reporter to test the inhibition of the SARS-CoV-2 3CL pro with 91 a known coronavirus 3CL pro inhibitor, GC376 (17), and then validated the correlation between 92 reporter inhibition and inhibition of SARS-CoV-2 viral replication. These experiments together 93 demonstrate the utility of this approach for the identification of novel antiviral drugs that target 94 the SAR-CoV-2 main protease, 3CL pro . 95

Generation of a fluorescent SARS-CoV-2 3CL pro activity reporter 98

In order to develop a fluorescent reporter responsive to the SARS-CoV-2 main protease, we started 99 with the FlipGFP protein (16). FlipGFP is used to detect protease activity by expressing the GFP CoV reporter 1 contains the conserved nsp4-5 cleavage site present in the SARS-CoV and SARS-124

CoV-2 viral polyproteins (31). CoV reporter 2 contains an optimized cleavage sequence for the 125 SARS-CoV 3CL pro (32). CoV reporter 3 contains an optimized sequence shown to be highly 126 FlipGFP splits GFP into b1-9 and b10-11, with b11 held in parallel to b10 by heterodimerized coiled coils E5/K5 and a linker sequence containing a coronavirus cleavage site. The CoV main protease, 3CL pro , cuts at the cleavage site allowing b11 to "flip" anti-parallel to b10, enabling self-assembly of the complete GFP beta-barrel and resulting in detectable fluorescence. The pan-coronavirus 3CL pro consensus sequence, LQ, is in bold. Scale bars are 100µm. B) Microscopy of 293T cells 48 hours post-transfection with each FlipGFP reporter and either the SARS-CoV-2 3CL pro or an influenza viral protein (A/PR8/1834 NP). Green = cleaved FlipGFP, blue = nuclei. C) Quantification of 1B. Data shown as mean ± SD, n=3, statistical analysis relative to NP control. P-values calculated using unpaired, two-tailed Student's t-tests (*p<0.05, **p<0.001).

cleaved by many CoV family members (24). As a negative control, we also generated a construct 127 harboring the Tobacco etch virus (TEV) protease cleavage site. 128 129 Our goal was to identify a construct with minimal background fluorescence while still being 130 efficiently cleaved by SARS-CoV-2 3CL pro , allowing strong fluorescence for detection via 131 microscopy, plate reader, or flow cytometry. To test our 3CL pro reporters, we co-transfected each 132

reporter with a SARS-CoV-2 3CL pro expression plasmid. At 48 hours post-transfection, we could 133 detect GFP-positive cells with each of the three CoV reporters transfected with in the SARS-CoV-134 2 3CL pro (Fig. 1B) . In contrast, with transfection of a negative control, nucleoprotein protein from 135 an H1N1 influenza virus (A/PR8/1934 NP), we did not detect any signal above background levels 136 of fluorescence. Further, the reporter containing the TEV cleavage site was not activated by SARS-137

CoV-2 3CL pro (Fig. 1B) CoV 3CL pro proteins are reasonably conserved across coronavirus groups ( Fig. 2A) compared to the control influenza nucleoprotein with CoV reporter 3 (Fig. 2B) . Quantification 152 with a plate reader demonstrated that SARS-CoV (Beta, human) and avian infectious bronchitis 153 (IBV-Gamma, avian) resulted in similar levels of fluorescence to SARS-CoV-2 (Beta, human) 154 (Fig. 2C) . Murine hepatitis virus (MHV-Beta, murine) and human coronavirus 229E (HCoV-155 229E-Alpha, human) were less compatible with CoV reporter 3, while still producing 12-and 80-156 fold changes in fluorescence, respectively, over background (Fig. 2C) . These experiments show 157 our FlipGFP 3CL pro reporter is generally compatible with many CoV 3CL pro proteins across 158 coronavirus groups and host species, potentially enabling protease inhibitor screening for a variety 159

of CoVs in addition to SARS-CoV-2. To develop an assay for protease inhibitor screening using our CoV reporter 3, we first needed to 164 optimize the experimental conditions. We performed a transfection timecourse with SARS-CoV-165 2 3CL pro to determine an early, appropriate timepoint for sample collection (Fig. 3A) . At 12 hours 166 post-transfection, only a few GFP fluorescing cells are visible and fluorescent signal is just above 167 background (Fig. 3B) . At 24 hours post-transfection, green cells are visible without appreciable 168 background signaling (Fig. 3B) . At 48 hours post-infection, most cells produce a high GFP signal, 169 with some background fluorescence detectable (Fig. 3B) . We therefore selected the 24 hour post-170 transfection timepoint. To increase the sensitivity of our assay, we titrated the level of SARS-CoV-171 2 3CL pro transfected with CoV reporter 3; our goal was to maximize activation of the reporter 172 while minimizing the amount of protease available in the cell. We transfected cells with five ratios 173 of reporter-to-protease: 1:1, 1:0.8, 1:0.4, 1:0.2, and 1:0. 24 hours post-transfection, we observed Data shown as mean ± SD, n=3, statistical analysis relative to NP control. P-values calculated using unpaired, two-tailed Student's t-tests (*p<0.05, **p<0.001). C) Quantification of 293T cells 24 hours post-transfection with CoV reporter 3 and SARS-CoV-2 3CL pro , with decreasing levels of 3CL pro . Data shown as mean ± SD, n=3, statistical analysis relative to 1:1 ratio reporterto-protease. P-values calculated using unpaired, two-tailed Student's t-tests (*p<0.05, **p<0.001). D) In black: quantification of 293T cells 24 hours post-transfection with CoV reporter 3 and SARS-CoV-2 3CL pro and treated with the pan-coronavirus protease inhibitor, GC376. Data shown as mean ± SD with nonlinear fit curve, n=3. In gray: cell viability was calculated relative to untransduced, vehicle-only (DMSO) samples. Data shown as mean ± SD, n=3. E) In black: RT-qPCR of VeroE6 cells 24 hours post-infection with SARS-CoV-2 (MOI 0.01) and treatment with the pan-coronavirus protease inhibitor, GC376. Data shown as mean ± SD with nonlinear fit curve, n=4. In gray: cell viability was calculated relative to un-transduced, vehicleonly (DMSO) samples. Data shown as mean ± SD, n=3. significant decreases in reporter activation at reporter-to-protease ratios 1:0.4 and 1:0.2 (Fig. 3C) . 176 However, a 1:0.8 reporter-to-protease ratio resulted in no significant loss of fluorescence compared 177 to a 1:1 ratio (Fig. 3C) . Based on these experiments together, we selected a 1:0.8 reporter-to-178 protease ratio for transfection and a 24-hour post-transfection end point as the optimal conditions 179 for our protease inhibitor assay using the FlipGFP 3CL pro reporter, CoV reporter 3. 180 181 Finally, we wanted to verify that our assay could detect drug inhibition of the SARS-CoV-2 3CL pro 182 with a known inhibitor. Therefore, we selected a recognized pan-coronavirus 3CLpro inhibitor, 183 GC376, to test our assay (17). Four concentrations of GC376, that did not significantly impact cell 184 viability compared to vehicle alone, were applied to cells at the time of transfection with CoV 185 reporter 3 and SARS-CoV-2 3CL pro . As expected, reporter activity levels were maintained at the 186 lower protease inhibitor concentrations, while fluorescence was reduced at the higher 187 concentrations of GC376 (Fig. 3D) . Thus, our assay successfully detected inhibition of SARS-188

CoV-3 3CL pro by the protease inhibitor GC376. However, it is also important to verify that 189 inhibition of our reporter is strongly correlated with inhibition of the SARS-CoV-2 virus. We 190 infected VeroE6 cells with SARS-CoV-2 at an MOI of 0.01 before applying protease inhibitor at 191 the same four concentrations as tested with the protease reporter. 24 hours post infection, we 192 collected RNA and performed RT-qPCR to detect SARS-CoV-2 viral RNA; similar to the reporter, 193 viral RNA levels were suppressed in a dose-dependent manner (Fig. 3E) . Our observed inhibition 194 of the virus is consistent with reports of inhibition of SARS-CoV-2 by GC376 in the literature (22, 195 23) . All together, these experiments demonstrate feasibility of using our FlipGFP CoV 3CL pro 196 reporter assay to identify protease-targeting inhibitors of SARS-CoV-2. 197

Our goal for this study was to develop a cell-based assay to screen for novel SARS-CoV-2 antiviral 200 drugs at BSL2; to our knowledge, no such assay optimized for SARS-CoV-2 currently exists. 201 Therefore, we generated a reporter requiring a coronavirus protease, 3CL pro , for activation of a 202 GFP-fluorescent signal. We showed this reporter is responsive to the SARS-CoV-2 3CL pro , in 203 addition to many different coronavirus 3CL pro proteins. After optimizing screening conditions, we 204 demonstrated that our reporter was sensitive to treatment with a known coronavirus protease 205 inhibitor, GC376. These experiments illustrate the utility of our approach to identify, and 206 subsequently optimize, novel protease inhibitors of SARS-CoV-2. 207 208

To meet the demands of virus research during the SARS-CoV-2 pandemic, reporter assays need 209 to be flexible and high-throughout. Our reporter is activated with expression of a single CoV 210 protein, 3CL pro , allowing for SARS-CoV-2 drug testing at BSL2. Additionally, the reporter is 211 compatible with many CoV 3CL pro proteins, supporting rapid testing of inhibitors against a variety 212 of coronaviruses, present or future, and without synthesis of protease substrates or purification of 213 viral proteins (13, 17, 22-28). Further, as our assay is performed in living cells, our system enables 214 the discovery of protease inhibitors while simultaneously evaluating effects on cellular viability. 215

Our assay is scalable, and the analysis requires only a basic fluorescent plate reader, supporting 216 high-throughput screening. In addition to applications in drug discovery pipelines, this assay could 217 be deployed to determine targets of antivirals identified via viral screening. Isolate USA-WA1/2020, NR-52281. Biocontainment work was performed in the Duke Regional 288 Biocontainment Laboratory, which received partial support for construction from the National 289

Institutes of Health, National Institute of Allergy and Infectious Diseases (UC6-AI058607). We 290 would like to thank Dr. Clare Smith for help establishing SARS-CoV-2 viral infection assays at 291 BSL3 and Laura Froggatt for designing the FlipGFP diagram. 292

China Novel Coronavirus 296 Investigating and Research Team. 2020. A Novel Coronavirus from Patients with 297 Pneumonia in China

Early Transmission Dynamics in Wuhan, China, of 303 Novel Coronavirus-Infected Pneumonia

A familial cluster of pneumonia 307 associated with the 2019 novel coronavirus indicating person-to-person transmission: a 308 study of a family cluster

The species Severe acute respiratory syndrome-related coronavirus: classifying 310 2019-nCoV and naming it SARS-CoV-2

COVID-19 Systematic 313 Urgent Review Group Effort (SURGE) study authors DK

319 Physical distancing, face masks, and eye protection to prevent person-to-person 320 transmission of SARS-CoV-2 and COVID-19: a systematic review and meta-analysis

The Current and Future State of 323 Vaccines, Antivirals and Gene Therapies Against Emerging Coronaviruses

Baric 328 RS. 2017. Broad-spectrum antiviral GS-5734 inhibits both epidemic and zoonotic 329 coronaviruses

An orally bioavailable broad-spectrum antiviral inhibits SARS-CoV-2 in human 341 airway epithelial cell cultures and multiple coronaviruses in mice

Why Are Lopinavir and Ritonavir Effective 344 against the Newly Emerged Coronavirus 2019? Atomistic Insights into the Inhibitory 345 Mechanisms

A 352 Trial of Lopinavir-Ritonavir in Adults Hospitalized with Severe Covid-19

Structure-based design of antiviral drug 357 candidates targeting the SARS-CoV-2 main protease

Hilgenfeld 359 R. 2020. Crystal structure of SARS-CoV-2 main protease provides a basis for design of 360 improved α-ketoamide inhibitors

Structure of 364 Mpro from SARS-CoV-2 and discovery of its inhibitors

Designing a Green Fluorogenic 367 Protease Reporter by Flipping a Beta Strand of GFP for Imaging Apoptosis in Animals

Broad-Spectrum Antivirals against 3C or 3C-Like Proteases of 371 Picornaviruses, Noroviruses, and Coronaviruses

Genome Composition and 374 Divergence of the Novel Coronavirus (2019-nCoV) Originating in China

Coronavirus 3CLpro proteinase cleavage 377 sites: Possible relevance to SARS virus pathology

379 Identification of novel proteolytically inactive mutations in coronavirus 3C-like protease 380 using a combined approach

Assessing Activity and Inhibition of 382 Middle East Respiratory Syndrome Coronavirus Papain-Like and 3C-Like Proteases 383 Using Luciferase-Based Biosensors

Boceprevir, GC-376, and calpain inhibitors II

CoV-2 viral replication by targeting the viral main protease

Feline coronavirus drug inhibits the main protease of SARS-CoV-2 and blocks virus 390 replication

Profiling of Substrate Specificities 392 of 3C-Like Proteases from Group 1, 2a, 2b, and 3 Coronaviruses

Coronaviruses resistant to a 3C-like protease inhibitor are 395 attenuated for replication and pathogenesis, revealing a low genetic barrier but high fitness 396 cost of resistance

Screening Identifies Inhibitors of the SARS Coronavirus Main 399 Proteinase in the first step of the formation of the crucial replication-transcription 400 complex. The activity of 3CL pro , so named for its similarity to 3C proteinases of 401 Picornaviridae

Reversal of the Progression of Fatal Coronavirus 404 Infection in Cats by a Broad-Spectrum Coronavirus Protease Inhibitor

Inhibition of SARS-CoV 3CL protease by 407 flavonoids

Printed in Great Britain Conservation of substrate specificities among coronavirus main proteasesJournal of General Virology

Purification, and Substrate Specificity of Severe Acute Respiratory 412 Syndrome Coronavirus 3C-like Proteinase

Prediction of the SARS-CoV-2 (2019-nCoV) 3C-414 like protease (3CLpro) structure: virtual screening reveals velpatasvir, ledipasvir, and 415 other drug repurposing candidates

The substrate specificity of 417 SARS coronavirus 3C-like proteinase

Chimeric exchange of coronavirus nsp5 proteases (3CLpro) identifies common 420 and divergent regulatory determinants of protease activity

NCBI viral genomes resource