In this work, we describe the protocols used in replicon-based and viral enzyme-based assays to screen for inhibitors of Zika virus replication in a high-throughput screening format.
Antiviral drug discovery requires the development of reliable biochemical and cellular assays that can be performed in high-throughput screening (HTS) formats. The flavivirus non-structural (NS) proteins are thought to co-translationally assemble on the endoplasmic reticulum (ER) membranes, forming the replication complex (RC). The NS3 and NS5 are the most studied enzymes of the RC and constitute the main targets for drug development due to their crucial roles in viral genome replication. NS3 protease domain, which requires NS2B as its cofactor, is responsible for the cleavage of the immature viral polyprotein into the mature NS proteins, whereas NS5 RdRp domain is responsible for the RNA replication. Herein, we describe in detail the protocols used in replicon-based screenings and enzymatic assays to test large compound libraries for inhibitors of the Zika virus (ZIKV) replication. Replicons are self-replicating subgenomic systems expressed in mammalian cells, in which the viral structural genes are replaced by a reporter gene. The inhibitory effects of compounds on viral RNA replication can be easily evaluated by measuring the reduction in the reporter protein activity. The replicon-based screenings were performed using a BHK-21 ZIKV replicon cell line expressing Renilla luciferase as a reporter gene. To characterize the specific targets of identified compounds, we established in-vitro fluorescence-based assays for recombinantly expressed NS3 protease and NS5 RdRp. The proteolytic activity of the viral protease was measured by using the fluorogenic peptide substrate Bz-nKRR-AMC, while the NS5 RdRp elongation activity was directly detected by the increase of the fluorescent signal of SYBR Green I during RNA elongation, using the synthetic biotinylated self-priming template 3′UTR-U30 (5'-biotin-U30-ACUGGAGAUCGAUCUCCAGU-3').
The Zika virus (ZIKV) is an emerging arthropod-borne virus member of the genus Flavivirus, which includes the closely related Dengue virus (DENV), Japanese encephalitis virus (JEV) and Yellow Fever virus (YFV), that pose constant threats to public health1. The 2015-16 ZIKV outbreak in the Americas received global attention following its emergence in Brazil due to the association with severe neurological disorders, such as congenital ZIKV-associated microcephaly in newborns2,3 and Guillain-Barré syndrome in adults4. Although the number of infection cases declined throughout the next two years, autochthonous mosquito-borne transmissions of ZIKV were verified in 87 countries and territories in 2019, therefore, evidencing the potential of the virus to re-emerge as an epidemic5. To date, there are no approved vaccines or effective drugs against ZIKV infection.
Antiviral drug discovery requires the development of reliable cellular and biochemical assays that can be performed in high-throughput screening (HTS) formats. Replicon-based screenings and viral enzyme-based assays are two valuable strategies to test small-molecule compounds for inhibitors of ZIKV1. The flavivirus non-structural (NS) proteins are thought to co-translationally assemble on the endoplasmic reticulum (ER) membranes, forming the replication complex (RC)6. The NS3 and NS5 are the most studied enzymes of the RC and constitute the main targets for drug development due to their crucial roles in viral genome replication. NS3 protease domain, which requires NS2B as its cofactor, is responsible for the cleavage of the immature viral polyprotein into the mature NS proteins, whereas NS5 RdRp domain is responsible for the RNA replication6.
Replicons are self-replicating subgenomic systems expressed in mammalian cells, in which the viral structural genes are replaced by a reporter gene. The inhibitory effects of compounds on viral RNA replication can be easily evaluated by measuring the reduction in the reporter protein activity7. Herein, we describe the protocols used for screening inhibitors of the ZIKV replication in a 96-well plate format. The replicon-based assays were performed using a BHK-21 ZIKV Rluc replicon cell line that we have recently developed8. To characterize the specific targets of identified compounds, we established in vitro fluorescence-based assays for recombinantly expressed NS3 protease using the fluorogenic peptide substrate, Bz-nKRR-AMC, whereas for NS5 RdRp we measured the elongation of the synthetic biotinylated self-priming template 3′UTR-U30 (5'-biotin-U30-ACUGGAGAUCGAUCUCCAGU-3'), using the intercalating dye SYBR Green I.
The ZIKV protease (45-96 residues of NS2B cofactor linked to residues 1-177 of NS3 protease domain by a glycine rich linker [G4SG4]) was obtained, as described for YFV9, while the polymerase (276-898 residues of RdRp domain) was cloned and expressed, as detailed in10. Both enzyme sequences were derived from GenBank ALU33341.1. As primary antiviral screenings, compounds are tested at 10 µM and those showing activities ≥ 80% are then evaluated in a dose-dependent manner, resulting in the effective/inhibition (EC50 or IC50) and the cytotoxic (CC50) concentrations. In the context of representative results, the EC50 and CC50 values of NITD008, a known flaviviruses inhibitor11, from replicon-based screenings are shown. For the enzymatic assays, the IC50 values of two compounds from the MMV/DNDi Pandemic Response Box, a library composed of 400 molecules with antibacterial, antifungal and antiviral activities, are shown. The protocols described in this work could be modified to screen for inhibitors of other related flaviviruses.
1. Luciferase activity assay
NOTE: Ensure that all procedures involving cell culture are conducted in certified biosafety hoods (see Table of Materials).
2. Cell proliferation-based MTT assay
3. NS2B-NS3 protease activity assay
4. NS5 RdRp elongation assay
NOTE: All materials used in this assay are RNase, DNase and pyrogenase free certified.
All the protocols described herein were stablished in 96-well plates and allows the evaluation of 80 compounds per plate in a primary screening of a single concentration, including the negative and positive controls placed at the first and last column of the plates, respectively. The replicon-based screenings are represented in Figure 1, which includes the RNA construct developed to obtain the BHK-21-RepZIKV_IRES-Neo cell line (Figure 1A), the assays schematic representation (Figure 1B) and the dose-response curves of NITD008 (EC50 of 0.28 µM, CC50 > 10 µM) (Figure 1C). The EC50 and CC50 values of hit compounds are determined as the concentrations required to inhibit 50% of the Rluc activity and cause 50% cytotoxicity, respectively. With respect to the luciferase assay, DMSO 1% is used as a no inhibitor control (0% inhibition) and NITD008 is used as a positive control (100% inhibition), as previously described8.
The NS2B-NS3 protease activity is measured by fluorescence monitoring of AMC released due to the proteolytic activity of the protease (Figure 2A). Aprotinin, a protein that acts as trypsin inhibitor and is already described as an inhibitor of flavivirus proteases13,14,15, was used in this assay as an experimental positive control (IC50 of 0.13 ± 0.02 µM, data not shown). Figure 2B illustrates a dose-response inhibition curve of a molecule targeting the protease activity, the compound MMV1634402 (IC50 of 0.36 ± 0.08 µM). The elongation activity of NS5 RdRp is measured in real time by the increase in fluorescence intensity of SYBR Green I when intercalated with the synthesized dsRNA (Figure 2C). The dose-response inhibition curve of a hit molecule targeting ZIKV RdRp, the compound MMV1782220 (IC50 of 1.9 ± 0.8 µM), is showed in Figure 2D. Since nucleoside analog inhibitors, such as NITD008, are not suitable for enzymatic assays, as phosphates needs to be incorporated intracellularly to the molecule16, we did not use any positive control for NS5 RdRp elongation assay. However, Clofazimine, a commercial antibiotic, which we recently identified as an inhibitor of viral polymerase8, could be used as an experimental control in next assays.
Figure 1: Replicon-based screenings. A) Schematic representation of the ZIKV replicon construct containing a Rluc sequence at the 5' UTR terminus and a Neo gene at the 3' UTR terminus, that we have developed to obtain the BHK-21-RepZIKV-IRES_Neo cell line 8. B) Schematic representation of the luciferase activity assay and cell proliferation-based MTT assay performed to screen for inhibitors of ZIKV replication. C) The dose-response curves (EC50 and CC50) of NITD008. The assay was performed in duplicates. Error bars represent standard deviations. Please click here to view a larger version of this figure.
Figure 2: Viral enzyme-based assays. A) Schematic representation of NS2B-NS3 protease activity assay. B) The dose-response inhibition curve (IC50) of compound MMV1634402. C) Schematic representation of NS5 RdRp RNA polymerase activity assay. D) The dose-response inhibition curve (IC50) of compound MMV1782220. The assays were performed in triplicates. Error bars represent standard deviations. Please click here to view a larger version of this figure.
The protocols described herein could be readily adapted for screenings in a 384 or 1536-well formats. For biochemical and/or cell-based screenings performed in HTS format, the Z' factor value, a statistical parameter, is calculate for each plate to ensure the sensitivity, reproducibility and accuracy of those assays12. A Z' factor value of 0.5 or above is expected for replicon-based screenings while a value of 0.7 or above is expected for the NS3 and NS5 activity assays. For the replicon-based HTS, we have developed the BHK-21-RepZIKV_IRES-Neo cells, a stable cell line harboring a replicative ZIKV replicon containing a Renilla luciferase (Rluc) sequence at the 5' UTR region and a neomycin phosphotransferase (Neo) gene driven by an internal ribosomal entry site (IRES) at the 3´UTR. We retained 38 residues of capsid and 30 residues of envelop genes that are required for the correct initiation of the RNA translation, to maintain comparable replication levels and drug sensitivity between cell passages8. Due to the lack of structural genes, replicons do not produce progeny virions, thus eliminating the risk of laboratory-acquired viral infection17.
The antiviral assays using the ZIKV replicon cells consists in the luciferase activity and the cell proliferation-based MTT (cytotoxicity) assays performed in parallel. This is necessary to exclude false-positive hits, comprising molecules that interfere directly with the reporter protein expression and/or activity and those that adversely affect cell health7. Replicon systems allows the discovery of molecules that inhibit RNA replication but not those required for viral entry and virion assembly/release. Alternatively, replicons can be packaged to produce virus replicon particles (VRPs) by providing the structural proteins in trans17. The resulting single-round infectious particles (SRIPs) are infectious, but progeny virus cannot propagate as the package genome lacks structural genes. Therefore, VRPs could be used to test for inhibitors of viral entry/replication by measuring the levels of the reporter protein7.
In addition to the replicon-based screenings, we also detailed herein the protocols used in viral enzyme-based assays for the recombinant NS3 protease and NS5 RdRp. The proteolytic activity of the viral protease was measured by using the fluorogenic peptide substrate Bz-nKRR-AMC, which contains the ZIKV protease recognition and cleavage sequence coupled with the fluorescent tag 7-amine-4-methylcoumarin (AMC). Due to the protease activity, the fluorescent tag is released and the reaction rate is directly measured by monitoring the fluorescence in a spectrophotometer18,19. This assay is highly sensible, relatively cheap, quick and suitable for screening of large compound libraries20,21. The major drawback is the possible quenching between tested compounds and the fluorophore that can lead to false-positive hits. However, this issue could be addressed by an additional fluorescence measurement in the presence of AMC. Also, compounds showing emission or absorption in the same wavelength of the fluorophore cannot be evaluated by this method18,20.
Regarding the NS5 RdRp, its elongation activity is directly detected by the increase of the fluorescent signal of SYBR Green I during the elongation of a self-priming 3'UTR-U30 template. The protocol was adapted from assays with intercalating dyes such as Pico Green and SYTO 9 that have been widely used to evaluate compounds for different viral polymerases22,23,24,25,26. Even though we have used a self-priming biotinylated template27 in the assay, other templates, such as poliU, can be used as well25. The main disadvantage of this method is the high number of false-positive hits that interact with the dye, either by interfering with the fluorescence or by decreasing the dsRNA intercalation28. Therefore, hit compounds need to be validated with counter-assays such as biophysics methods or by comparing the SYBR™ Green I fluorescence in dsRNA with and without the compound29. Nevertheless, the easy implementation, direct measurement and affordability are key points to the use of fluorescence-based methods as HTS platforms, in comparison to radio-labeled or coupled assays that are difficult to implement in medium/large scale campaigns27,30,31.
The authors have nothing to disclose.
This work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), CEPID grant 2013/07600-3 to GO, grant 2018/05130-3 to RSF and 2016/19712-9 to ASG, and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (grant 88887.516153/2020-00) to ASG. We would like to gratefully thank the Medicine for Malaria Ventures (MMV, www.mmv.org) and the Drugs for Neglected Diseases initiative (DNDi, www.dndi.org) for their support, design of the Pandemic Response Box and supplying the compounds.
5'-biotin-U30- ACUGGAGAUCGAUCUCCAGU -3' | Dharmacon | – | 100 ng |
96-well cell culture plates | KASVI | K12-096 | |
96-well PCR Microplate | KASVI | K4-9610 | |
96-well White Flat Bottom Polystyrene High Bind Microplate | Corning | 3922 | |
AMC (7-amine-4-methylcoumarine) | SIGMA-Aldrich | 257370 | 100 mg |
Aprotinin from bovine lung | SIGMA-Aldrich | A1153 | 10 mg |
ATP | JenaBioscience | NU-1010-1G | 1 g |
Bz-nKRR-AMC | International Peptides | – | 5 mg |
Class II Biohazard Safety Cabinet | ESCO | ||
Diethyl pyrocarbonate | SIGMA-Aldrich | D5758 | 25 mL |
DMSO (Dimethyl sulfoxide) | SIGMA-Aldrich | 472301 | 1 L |
Dulbecco’s Modified Eagle Medium | GIBCO | 3760091 | |
Fetal Bovine Serum | GIBCO | 12657-029 | 500 mL |
G418 | SIGMA-Aldrich | A1720 | Disulfate salt |
Glycerol | SIGMA-Aldrich | G5516 | 1 L |
HERACELL VIOS 160i CO2 incubator | Thermo Scientific | ||
MnCl2 tetrahydrate | SIGMA-Aldrich | 203734 | 25 g |
MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) | Invitrogen | M6494 | |
NITD008 ≥98% (HPLC) | Sigma-Aldrich | SML2409 | 5 mg |
qPCR system Mx3000P | Agilent | ||
Renilla luciferase Assay System | PROMEGA | E2810 | |
SpectraMax Gemini EM Fluorescence Reader | Molecular Devices | ||
SpectraMax i3 Multi-Mode Detection Platform | Molecular Devices | ||
SpectraMax Plus 384 Absorbance Microplate Reader | Molecular Devices | ||
SYBR Green I | Invitrogen | S7563 | 500 µl |
Triton X-100 | SIGMA-Aldrich | X100 | 500 mL |
Trizma base | SIGMA-Aldrich | T1503 | 1 kg |
Trypsin-EDTA Solution 1X | SIGMA-Aldrich | 59417-C | 100 mL |