key: cord-0257789-u8ukma2i
authors: Hou, Ningke; Peng, Chen; Zhang, Lijing; Zhu, Yuyao; Hu, Qi
title: BRET-based self-cleaving biosensors for SARS-CoV-2 3CLpro Inhibitor Discovery
date: 2021-07-28
journal: bioRxiv
DOI: 10.1101/2021.07.28.454072
sha: 2d1df62bb192d5f829899efffc9bfacdd049a933
doc_id: 257789
cord_uid: u8ukma2i

The 3C-like protease (3CLpro) of SARS-CoV-2 is an attractive drug target for developing antivirals against SARS-CoV-2. A few small molecule inhibitors of 3CLpro are in clinical trials for COVID-19 treatments and more inhibitors are being developed. One limiting factor for 3CLpro inhibitors development is that the cellular activities of such inhibitors have to be evaluated in a Biosafety Level 3 (BSL-3) or BSL-4 laboratory. Here, we design genetically encoded biosensors that can be used in BSL-2 laboratories to set up cell-based assays for 3CLpro inhibitor discovery. The biosensors were constructed by linking a green fluorescent protein (GFP2) to the N-terminus and a Renilla luciferase (RLuc8) to the C-terminus of SARS-CoV-2 3CLpro, with the linkers derived from the cleavage sequences of 3CLpro. After over-expression of the biosensors in HEK293 cells, 3CLpro can be released from GFP2 and RLuc by self-cleavage, resulting in a decrease of the bioluminescence resonance energy transfer (BRET) signal. Using one of these biosensors, pBRET-10, we evaluated the cellular activities of several 3CLpro inhibitors. These inhibitors restored the BRET signal by blocking the proteolysis of pBRET-10, and their relative activities measured using pBRET-10 were consistent with their anti-SARS-CoV-2 activities reported previously. We conclude that the biosensor pBRET-10 is a useful tool for SARS-CoV-2 3CLpro inhibitor discovery. Furthermore, our strategy can be used to design biosensors for other viral proteases that share the same activation mechanism as 3CLpro, such as HIV protease PR and HCV protease NS3. Highlights Sensitive cell-based biosensors for 3CLpro inhibitor discovery in BSL-2 laboratories. The BRET-based self-cleaving biosensors mimic the in vivo autoproteolytic activation of 3CLpro. Similar biosensors can be designed for other self-cleaving proteases, such as HIV protease PR and HCV protease NS3.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causing the global 2 pandemic of COVID-19 poses a great threat to public health (Zhou et al., 2020) . Despite 3 several vaccines have been accessible, effective antivirals are still urgently needed for 4 the treatment of COVID-19 (Grobler et al., 2020) . The RNA genome of SARS-CoV-2 5 encodes two large overlapping polyproteins pp1a and pp1ab, and several structural 6 proteins and accessory proteins (Hartenian et al., 2020) . During virus replication in host 7 cells, pp1a and pp1ab are expressed and then cleaved to generate 16 non-structural 8 proteins (nsps). The cleavages are catalyzed by nsp3 and nsp5 -two proteases included 9 in the 16 nsps. Specifically, the papain-like protease (PLpro) domain of nsp3 cleaves 10 the peptides bonds between nsp1 and 2, nsp2 and 3, and nsp3 and 4; the peptide bonds 11 between other nsps are cleaved by nsp5 (also called 3C-like protease, 3CLpro or the 12 main protease). Inhibition of 3CLpro is an effective strategy to develop antivirals 13 against SARS-CoV-2 (Zumla et al., 2016) . Several 3CLpro inhibitors have been 14 reported, two of which are now in COVID-19 clinical trials (Dai et al., 2020; de Vries 15 et al., 2021; Drayman et al., 2021; Fu et al., 2020; Vandyck and Deval, 16 2021; Zhang et al., 2020) .

Enzymatic assays using purified 3CLpro were frequently used in the initial 18 screening of 3CLpro inhibitors, but to evaluate the cell permeability and cellular 19 activities of the inhibitors, cell-based antiviral assays are necessary. The requirement of 20 Biosafety Level 3 (BSL-3) or BSL-4 laboratories for doing the cell-based anti-SARS-

CoV-2 assays has slowed the development of 3CLpro inhibitors.

To set up cell-based 3CLpro assays that can be done in BSL-2 laboratories, several 23 biosensors have been developed. The first is called Flip-GFP that has a 3CLpro cleavage 24 site inserted into GFP; the fluorescence of GFP was decreased by the insertion but could 25 be restored by the 3CLpro-catalyzed cleavage (Froggatt et al., 2020) . Similar biosensors 26 have been developed by other groups (O'Brien et al., 2021; Gerber et al., 2021) . The 27 second is an engineered luciferase having two complementary luciferase fragments linked by a 3CLpro cleavage site; the luminescence was lost by 3CLpro cleavage and 1 restored when the 3CLpro activity was inhibited (Rawson et al., 2021) . The third is a 2 GFP fusion protein having an ER targeting domain linked to the C-terminus of GFP 3 through a 3CLpro cleavage site; 3CLpro-catalyzed cleavage led to a translocation of 4 the GFP from ER to the nucleus which can be quantified using light microscopy 5 (Pahmeier et al., 2021) . The fourth is a biosensor in which a Src myristoylation domain 6 and a HIV-1 Tat-GFP fusion protein are linked to the N-and C-terminus of 3CLpro, 7 respectively, through 3CLpro cleavage sites; expression of this biosensor in HEK 293T 8 cells showed little GFP fluorescence while inhibition of 3CLpro greatly increased the 9 GFP fluorescence, probably because 3CLpro-catalyzed self-cleavage led to degradation 10 of the biosensor (Moghadasi et al., 2020) .

The limitations of these biosensors are that (1) their readouts are highly dependent 12 on the expression level of the biosensors, and for the first three types of biosensors, the 13 expression level of 3CLpro also affects the readouts; (2) the first three types of 14 biosensors require either co-transfection of two plasmids (the biosensor and 3CLpro 15 plasmids) or transfection of 3CLpro into cells stably expressing the biosensors, while 16 the fourth only needs to transfect one plasmid but its sensitivity to 3CLpro inhibitor 17 GC376 was much lower.

In this study, we developed a series of BRET-based biosensors to set up cell-based 19 assays for 3CLpro inhibitor discovery. We linked a green fluorescent protein (GFP2) 20 and a Renilla luciferase (RLuc8) to the N-and C-terminus of SARS-CoV-2 3CLpro, 21 respectively, using 3CLpro cleavage sequences as the linkers (Fig. 1A&B ). The BRET 22 from Rluc8 to GFP2 of the biosensors was disrupted upon self-cleavage catalyzed by 23 3CLpro and can be restored by adding 3CLpro inhibitors. The effect of variance in the 24 biosensor expression level on the readouts was minimized by normalizing the BRET 25 signal with the luminescent signal of RLuc8. 

The gene sequences of RLuc8 and GFP2 are the same as that reported previously (Bery 2 et al., 2018) . The gene sequences of SARS-CoV-2 3CLpro and its cleavage sites are the 3 same as that in the SARS-CoV-2 genome (NC_045512.2). The DNA fragment encoding 4 the biosensor pBRET-1 was synthesized at GENEWIZ (Suzhou, China) and inserted 5 into pcDNA3.1 vector at the site after the FLAG-tag. Then an HA-tag was added to the 6 C-terminus of pBRET-1 by Gibson homologous recombination using primers HA-F and 7 HA-R (Table S1 ). The plasmid of pBRETmut-1 was constructed by introducing the 8 3CLpro C145A mutation into pBRET-1 through site-directed mutagenesis using 9 primers C145A-F and C145A-R (Table S1 ). The plasmids carrying other pBRET 10 biosensors were constructed on the basis of pBRET-1 using Gibson homologous 11 recombination method. The primers were shown in Table S1 . The protein sequences of 12 all the BRET-based self-cleaving biosensors were shown in Table S2 . 

To monitor the self-cleavage of pBRET-1 and to evaluate its sensitivity to 3CLpro 24 inhibitor GC376, HEK 293T cells were seeded into 6-well cell culture plates at about 40% confluence, and after 24 hours the cells were transfected with plasmids carrying 1 pBRET-1 (5 μg/well) using PEI as the transfection reagent. After transfection, 3CLpro 2 inhibitors in DMSO were added into the cell culture to reach the indicated working 3 concentrations. The final DMSO concentration in the cell culture was 0.5%. After 4 additional 24 hours, the cells were harvested and equally divided into two parts: one 5 part for western blot analysis, and the other for BRET assay. To set up high-throughput BRET assay for optimizing pBRET-1 and evaluating the 2 activities of different 3CLpro inhibitors, HEK 293T cells were seed into a 96-well clear-3 bottom white plate (Corning, 3610) at about 40% confluence, and after 24 hours the 4 cells were transfected with plasmids carrying the biosensors (0.4 μg/well) using PEI as 5 the transfection reagent. Then 3CLpro inhibitors in DMSO were added into the cell 6 culture to reach the indicated working concentrations. Twenty-four hours later, the 7 BRET ratios were measured using same protocols as described above. The maximal distance for BRET is about 10 nm (Bacart et al., 2008) . According to a 11 crystal structure of SARS-CoV-2 3CLpro (PDB code: 6Y2E), the distance between the 12 N-and C-terminus of 3CLpro is 22.9 Å . To construct our first 13 biosensor -pBRET-1, we linked GFP2 to the N-terminus of SARS-CoV-2 3CLpro 14 using the cleavage sequence between nsp4 and 3CLpro as the linker, and linked RLuc8 15 to the C-terminus of 3CLpro using the cleavage sequence between 3CLpro and nsp6 as 16 the linker; furthermore, we added a FLAG tag before GFP2 and a HA tag after RLuc8 17 (Fig. 1A) . We also constructed pBRETmut-1, in which the catalytic residue C145 of 18 3CLpro was mutated to alanine. (Fig. 1C) . Adding GC376, a reported inhibitor of 3CLpro (Fu et al., 2020), 22 to the cell culture increased the BRET ratio of pBRET-1 in a concentration-dependent 23 manner. These results indicate that the BRET ratio of pBRET-1 is negatively associated 24 with the protease activity of 3CLpro. 25 We also monitored the self-cleavage of pBRET-1 in HEK 293T cells using Western blot. We first used anti-HA antibody to detect the self-cleavage products (Fig. 1D) . For 1 pBRET-1 in the absence of GC376, only RLuc was detected; as the concentration of 2 GC376 increased, the band of the 3CLpro-RLuc8 fragment and that of the full-length 3 GFP2-3CLpro-Rluc8 fusion protein appeared. We also used anti-FLAG antibody to 4 detect the self-cleavage products (Fig. 1E) . Interestingly, as the concentration of GC376 5 increased, only the bands of GFP2 and the full-length GFP2-3CLpro-Rluc8 fusion 6 protein were detected, but no band of the GFP2-3CLpro fragment. These results suggest 7 that the cleavage at the N-terminus of 3CLpro occurred before the cleavage at the C-8 terminus, which is consistent with the maturation process of 3CLpro reported 9 previously (Li et al., 2010) . There are eleven 3CLpro cleavage sites in the polyproteins pp1a and pp1ab of SARS-12 CoV-2 (Mody et al., 2021) . The efficiencies of 3CLpro to cleave these sites are different. 13 We presume that modulating the self-cleaving efficiency of pBRET-1 by changing the 14 cleavage sequence between GFP2 and 3CLpro and that between 3CLpro and RLuc8 15 may increase the sensitivity of the biosensor to 3CLpro inhibitors. As shown in Fig. 2,   16 nine pBRET biosensors were designed and their sensitivities to GC376 were tested.

Among them, pBRET-10 showed the highest sensitivity, with an EC50 value of 2.72 μM 18 for GC376; in contrast, the EC50 value measured using pBRET-1 was 11.60 µM (Fig.   19 2C). 

As pBRET-10 has the highest sensitivity, we measured the EC50 values of three other 22 3CLpro inhibitors (Boceprevir, compounds 11a and 13b) using pBRET-10 (Table 1) 23 (Dai et al., 2020; Fu et al., 2020; Zhang et al., 2020) . Compound 11a showed an activity 24 slightly lower than GC376, while the activities of Boceprevir and compound 13b were 25 an order of magnitude lower than that of GC376. The EC50 values of GC376 and 26 Boceprevir are comparable to that measured using cell-based anti-SARS-CoV-2 assays (Fu et al., 2020) . But for compounds 11a and 13b, the EC50 values from our 1 measurement were about 9 times the reported values from anti-SARS-CoV-2 assays 2 (Dai et al., 2020; Zhang et al., 2020) . 

We have developed a class of BRET-based self-cleaving biosensors that can be used in 5 BSL-2 laboratories to set up cell-based assays for 3CLpro inhibitor discovery. One of 6 them, pBRET-10, showed comparable sensitivity to cell-based antiviral assays. Self-7 cleavage catalyzed by 3CLpro in these biosensors mimics the activation process of 

HCV NS3-4A Serine Protease

2021. Viruses 13, 173. Suthar