Identification of dihydroorotate dehydrogenase inhibitor, vidofludimus, as a potent and novel inhibitor for influenza virus

Influenza A virus (IAV) infection causes respiratory disease. Recently, infection of IAV H5N1 among mammals are reported in farmed mink. Therefore, to discover antivirals against IAV, we screened a compound library by using the RNA‐dependent RNA polymerase (RdRp) assay system derived from H5N1 IAV including a drug‐resistant PA mutant (I38T) and a viral polymerase activity enhancing PB2 mutant (T271A). Upon screening, we found vidofludimus can be served as a potential inhibitor for IAV. Vidofludimus an orally active inhibitor for dihydroorotate dehydrogenase (DHODH), a key enzyme for the cellular de novo pyrimidine biosynthesis pathway. We found that vidofludimus exerted antiviral activity against wild‐type and drug‐resistant mutant IAV, with effective concentrations (EC50) of 2.10 and 2.11 μM, respectively. The anti‐IAV activity of vidofludimus was canceled by the treatment of uridine or cytidine through pyrimidine salvage synthesis pathway, or orotic acid through pyrimidine de novo synthesis pathway. This indicated that the main target of vidofludimus is DHODH in IAV RdRp expressing cells. We also produced recombinant seasonal IAV H1N1 virion and influenza B virus (IBV) RdRp assay system and confirmed vidofludimus also carried highly antiviral activity against seasonal IAV and IBV. Vidofludimus is a candidate drug for the future threat of IAV H5N1 infection among humans as well as seasonal influenza virus infection.


| INTRODUCTION
Influenza viruses have four types, type A, B, C, and D. Most of the outbreaks and epidemics of influenza are caused by influenza A viruses (IAV) and influenza B viruses (IBV). 1 IAV is an enveloped virus belonging to the family Orthomyxoviridae.IAV contains eight negative-sense single-stranded RNA (vRNA) genome segments and four proteins, including RNA dependent RNA polymerase (RdRp) subunits PB1, PB2, PA, and nucleoprotein (NP), which form ribonucleoprotein complexes (RNPs) (Supporting Information: J Med Virol.2024;96:e29372. wileyonlinelibrary.com/journal/jmv Figure S1). 2,3The RNP of IAV contributes to viral pathogenesis. 4,5fluenza viruses with high polymerase activity showed high pathogenicity. 6,7These pieces of evidence suggest that the RdRp contributed to the pathogenesis of influenza viruses.][10][11][12][13] Although a previous study described the function of the RdRp subunit of IAV, the effects of the RdRp complex and NP in IAV on the host cells and the antiviral compounds have remained largely undetermined. 14RdRp is a target for the development of anti-IAV drug.
Two IAV subtypes, subtypes H5 and H7, possess high pathogenicity with the potential to cause severe disease and high mortality in infected animals (chickens, pheasants, etc.). 15,16Notably, IAV H5N1 emerged in 1997 and led to the mass deaths of many exotic waterfowl species, including tufted ducks, swans, and flamingos, in Hong Kong. 17,18Although the IAV H5N1 is poorly transmissible from avian-human and human-human, 864 cases of human infection with IAV H5N1 were reported from January 2003 -March 2022 in 18 countries and resulting in a 53% case-fatality rate. 19,20In addition, an uncommon mutation (T271A) of the viral RdRp subunit PB2 gene was reported in a previous study and this mutation could enhance the viral polymerase activity in seasonal IAV H1N1. 21cently, infection of avian IAV H5N1 among mammals are reported in mink farm. 22This is the first report of the avian IAV H5N1 endemic among mammals and means that the warning level of IAV H5N1 endemic among humans might be raised by one rank.
Interestingly, mutation (T271A) in PB2 was found only in mink IAV H5N1 but not in avian IAV H5N1. 22The pandemic of IAV H5N1 might be the threat to global public health and economy.Therefore, we tried to test candidate compounds for the anti-IAV H5N1 drug.
Baloxavir marboxil has been approved by the U.S. Food and Drug Administration (FDA) as an antiviral drug for influenza.Baloxavir marboxil reduces viral proliferation by inhibiting the cap-dependent endonuclease of the influenza virus. 23On the other hand, recent clinical studies found that IAV carrying an I38T mutation in the viral RdRp subunit PA showed resistance to baloxavir marboxil. 24,25We planned to examine the mutations regarding drug resistance and virulence.For this purpose, we used the IAV H5N1 RdRp assay system previously reported in IAV H1N1. 26,27RdRp assay system in IAV H5N1 has an advantage of no risk of virus leak from the laboratory.
As safety is the biggest reason for discontinuing drug development, we chose the compound library consisting of drugs with clinical trial experience (Phase II Drop library).Drugs in Phase II Drop library cleared Phase I but not Phase II.We can set the new goal to develop the antivirals from the original goal of the drugs, such as antitumor, antihypertension etc. with guaranteed safety from the Phase I trial.
Another advantage using Phase II Drop library is the information of the drug concentration in the blood.
In this study, we found vidofludimus as a potent and novel inhibitor for IAV H5N1 under the criteria of inhibition (>80%) and cytotoxicity (<10%).Vidofludimus is an orally active inhibitor for dihydroorotate dehydrogenase (DHODH) and is under clinical trial for multiple sclerosis. 28DHODH is an enzyme catalyzing the de novo synthesis of pyrimidine.Previous reports showed inhibitors targeting DHODH had broad-spectrum antiviral ability against RNA viruses. 29 The anti-IAV activity of vidofludimus is due to the inhibition of DHODH.We also tested the antiviral activity of vidofludimus against seasonal IAV H1N1 using the RdRp assay and infectious virus, and against IBV using the RdRp assay.Vidofludimus also showed antiviral activity for both seasonal influenza viruses.
In the present study, we found vidofludimus as a potent and novel inhibitor for influenza virus.

| Cells
Cells of the human embryonic kidney 293 line (HEK 293), and Madin-Darby canine kidney cell line (MDCK) were maintained in Dulbecco's Modified Eagle Medium with 10% fetal bovine serum (10% FBS-DMEM) (Thermo Fisher Scientific). 30MDCK cell line was universally used in influenza study. 31Cells were cultured at 37°C in a 5% CO 2 incubator.

| Reagents and compound library
Baloxavir marboxil was purchased from Shionogi.Favipiravir, Following the previously reported approach, we constructed the recombinant plasmids using the cDNA of A/Hong Kong/213/2003 (H5N1). 26,27We artificially synthesized the PB1, PB2, PA, and NP genes of H5N1 (Supporting Information: Table ).Then, we cloned the cDNA of the open reading frames (ORFs) of the PB1, PB2, PA and NP genes into a retroviral vector pCX4bsr.We also used a QuickChange Site-Direct mutagenesis Kit (Agilent) to construct the mtPA I38T to test the baloxavir marboxil-resistance, and the mtPB2 T271A to assess its effect on RdRp activity.To monitor the replication level, we constructed a vRNA containing secreted nanoluciferase (sNLuc).The negative strand of the sNLuc gene was inserted between the 3' and 5' untranslated regions (UTR) of the H5N1 NP gene.Then, we flanked it with the human RNA polymerase I (hPol I) promoter and the mouse PoI I terminator (mPol T) and cloned it into a vector (pUC-GW-amp) to express the vRNA-encoding sNLuc gene. 32To mimic the natural end of vRNA 3'UTR, we inserted a hepatitis D virus (HDV) ribozyme (Rz) between 3' UTR and mPol T. 33 Using the same approach, we also constructed the recombinant plasmids using the cDNA of B/ Yamagata/16/1988 (IBV) (Supporting Information: Table ) (Genewiz).
We used sNLuc as a reporter to monitor the replication/ transcription of the IAV H5N1 and IBV vRNA.Briefly, we cotransfected PB1, PB2 PA, NP and sNLuc plasmids with FuGENE HD (Promega) according to the manufacturer's instructions into HEK293 cells.The sNLuc activities were measured at each time point (0, 24, 48, and 72 h) using a Nano-Glo ® Assay system (Promega) (Supporting Information: Figure S1).
Then, we cloned the cDNA of the ORFs of all eight genes into a retroviral vector pCX4bsr.Following the previously reported methods, we cocultured HEK293 cells and MDCK cells into a 6-well plate and cotransfected all 16 plasmids with Lipofectamine 2000 (Thermo Fisher Scientific) according to the manufacturer's instructions. 34,35After 48 h posttransfection, the supernatant was harvested, and the virus particle was quantified by focusforming units (FFU/ml) assay using anti-NP antibody for the infection experiments. 36We also construct a vRNA containing sNLuc using H1N1 NP backbone and cloned it into a vector (pUC57-amp) for the monitor in RdRp assay as the same method as shown in Section 2.3.
For the study of infectious virus, MDCK cells were plated into a 6-well plate (3 × 10 5 cells/well).After 24 h, cells were washed twice with phosphate-buffered saline (PBS) and inoculated with a multiplicity of infection (MOI) 0.002 of influenza virus (H1N1) at 37°C for 2 h.The inoculum was removed, and cells were covered with 2 mL of 0.3% FBS-DMEM containing the indicated concentration of compounds.
Samples extracted at 24 h (RdRp) and 48 h (RdRp or infection) after transfection or infection were lysed with 2× sodium dodecyl sulfate (SDS) Laemmli buffer, and cells were treated with ultrasonic apparatus (ELESTEIN ® ).The proteins in the lysates were separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE).Then, we transferred these proteins to Immobilon ® -PVDF transfer membranes by using a semi-dry transfer system (ATTO).
The proteins were detected by antibodies shown in Supporting Information: Table .The density of western blot bands was quantified with beta-actin as an internal reference protein by the software ImageJ 1.54f.

| Antiviral compounds assay
We plated HEK293 cells into a 24-well plate (8 × 10 4 cells/well).The following day, the plasmids were cotransfected into cells.At 24 h posttransfection, the compounds (baloxavir marboxil, favipiravir, and Phase II Drop library) were prepared in 10% FBS-DMEM.At 24 h after treatment, we extracted 20 µL of the supernatant into 1.5 mL tubes and mixed it with sNLuc substrate (Nano-Glo ® 1:1).We kept the samples at room temperature for 3 min.Then, luciferase signals were measured by a GLOMAXTM 20/20 luminometer (Promega).
The luciferase activities of compound-treated cells were contrasted with a negative control (compound-free well, set as 100%) and the relative luciferase signals were calculated.Triplicated data were used for statistical analysis.
The following day, compounds were prepared in 10% FBS-DMEM at adjusted concentrations.Then we replaced the culture medium with a medium containing the compounds.At 24 h posttreatment, we added 10 µL of WST-1 assay reagents (WST-1 cell proliferation assay kit) (TAKARA, Shiga, Japan) to each well of the cell plates and kept them at 37°C for 2 h.The absorbance of each sample was measured by a BIO-RAD Model 550 Microplate Reader (wavelength: 450 nm), and was compared with the absorbance for a negative control (compound-free, set as 100%) and a blank, and relative data were calculated.Triplicated data were used for statistical analysis.

| Statistical analysis
Statistical analysis was performed using GraphPad Prism 10.0 software.Statistically significant differences were determined using one-way analysis of variance (ANOVA).The average and standard deviations of three independent measurements were used for analysis.Data are presented as the means ± standard deviations (SD).Values of p < 0.05 were considered statistically significant.

| Screening of compound library using H5N1 RdRp assay system
The IAV H5N1 virion infects host cells and starts replication/ transcription using the RNP complex. 37In this study, we focused on the viral replication/transcription stages in the viral life cycle of the IAV H5N1.To monitor the viral specific replication/transcription, we constructed a reporter plasmid for the transcription of antisense gene of the sNLuc.The IAV H5N1, PB1, PB2, PA, and NP were cloned into the vector pCX4bsr, which could directly transcribe mRNA in the host cell (The details showed in section of MATERIALS AND METHODS and Supporting Information: Figure S1).
To exclude the false positives compounds, we examined these compounds twice using the same approach.Our criteria for the drug selection were inhibition (>80%) and cytotoxicity (<10%).
Vidofludimus could reduce the luciferase signal by 80% and keep high cell viability.For the further evaluation of vidofludimus, we determined its antiviral activity and toxicity by the 50% effective concentration (EC 50 ) and 50% cytotoxic concentration (CC 50 ) (Figure 1D,E).EC 50 and CC 50 of vidofludimus were 2.10 µM and 137.68 µM, respectively.
These results indicate that vidofludimus is a potentially effective antiviral drug for IAV H5N1.

| Evaluation of two mutations in the H5N1 RdRp subunit on viral polymerase activity and antiviral activity of the compounds
In 2018, the compound baloxavir marboxil was approved by the FDA to treat influenza.Nevertheless, A previous study reported that some IAV strains-namely, H1N1 and H3N2, both with mtPA I38Texhibited resistance to baloxavir marboxil. 38Therefore, we next constructed plasmids with I38T mutant in RdRp subunit PA of IAV H5N1 and examined whether this mutation leads to drug resistance in IAV H5N1.Our results showed the EC 50 of baloxavir marboxil was largely elevated in mutation-type IAV H5N1 (wtEC 50 vs mtEC 50 = 9.84 nM: 651 nM) (Supporting Information: Figure S2A and S2B).We also evaluated whether this mutation affected vidofludimus.Remarkably, the EC 50 of vidofludimus was unaffected in mutation-type IAV H5N1 (Figure 2A).
Following the recent report, an IAV H5N1 strain with an uncommon mutation in the viral RdRp subunit PB2 T271A exhibited high transmission ability among farmed minks in Europe. 22Therefore, we also constructed this mutant PB2 (mtPB2 T271A) and assessed its effect on viral polymerase activity and inhibitors.For the future clinical application of vidofludimus, we evaluated the effect of each mutation or the combination of two mutations on the drug at clinically achievable blood concentration (C max ).We used approved drugs, baloxavir marboxil and favipiravir, as the positive controls. 39The C max values of baloxavir marboxil, favipiravir, and vidofludimus were 0.18 µM, 230 µM, and 30 µM, respectively. 28  We further examined that the antiviral activity of vidofludimus is due to the inhibition of the enzymatic activity of DHODH.The plasmids-transfected HEK239 cells were treated with vidofludimus in combination with a gradually diluted concentration of L-Hydroorotic acid (L-Dihydroorotic acid, DHO) or orotic acid (OA).DHO and OA are substrate and product of DHODH, respectively.The addition of DHO unaffected the antiviral activity of vidofludimus even using high concentrations (Figure 3D).In contrast, addition of OA reversed the antiviral effect of vidofludimus significantly in a dose-dependent manner (Figure 3E).
Taken together, these results suggest that the anti-IAV H5N1 activity of vidofludimus is due to the inhibition of DHODH through blocking the production of OA from DHO in the de novo synthesis of pyrimidines.

| Vidofludimus inhibits natural influenza virus infection in vitro
To confirm, if the RdRp assay system correctly reflects the natural IAV infection, we constructed recombinant IAV particle by reverse genetics following the previously reported protocol (Supporting Information: Figure S3A). 34,35We produced seasonal influenza virus (H1N1) belonging to influenza virus type A same as H5N1.
Firstly, we evaluate the capacity of inhibitors to protect cell viability from CPE by IAV infection.We infect IAV to MDCK cells with MOI 0.002, then cells were treated with baloxavir marboxil, favipiravir and vidofludimus using gradually diluted concentration at a 96-well plate.After 72 h postinfection, CBB staining showed both baloxavir marboxil and vidofludimus (1.2 and 2.5 µM) treated cells kept cell viability similar to Mock (Figure 4A).WST-1 assay also showed the same result (2.5 µM) (Supporting Information: Figure S3B).We also extract proteins and RNAs from cells at 48 h postinfection to evaluate viral RNA replication/transcription. The western blot analysis showed baloxavir marboxil and vidofludimus largely reduced viral protein synthesis (Figure 4B) and qPCR analysis (30 cycles) also showed the viral RNA synthesis was prevented (Figure 4C).
Second, CPE experiment showed vidofludimus reduced CPE in a dose-dependent manner and that EC 50 might be between 1.2 μM and 0.6 μM (Figure 4A).The western blot analysis showed similar results to the CPE analysis (Figure 4D).We further performed a qPCR experiment using a gradient dilution of the drug starting from 1.5 μM of vidofludimus (Figure 4E).The EC 50 of vidofludimus was calculated as 0.7 μM.We also evaluated the effect of vidofludimus to prevent viral particle production.The cells were treated with 2.5 μM of vidofludimus after infection and the virus particles in supernatant were quantified at 24, 48, and 72 h posttreatment by FFU assay (Supporting Information: Figure S3C).
The cells were treated with various concentration of vidofludimus for 72 h after infection and the viral RNAs in the supernatant were assessed by RT-qPCR analysis (Supporting Information: Figure S3D).Viral production was inhibited by vidofludimus treatment in a dose-dependent manner.
Finally, to evaluate if the RdRp assay system correctly reflects the natural IAV infection, we construct the monitoring plasmid for sNLuc vRNA using A/WSN/1933 (H1N1)/NP vRNA backbone, same as the H5N1 assay system.Briefly, we cotransfected pCX4bsr/ WSN33/PB1, PB2, PA, NP and pUC57/WSN33/sNLuc into HEK293 cells and the culture mediums were changed to fresh mediums with the indicated concentration of compounds after 24 h posttransfection.On the following day, the luciferase activity of the samples was measured (Supporting Information: Figure S3E).In the baloxavir marboxil and vidofludimus treated groups, the luciferase signals were significantly reduced compared with the compound-free group.This result is consistent with that of natural IAV H1N1 infection experiments.
Taken together, these results indicate that vidofludimus is a candidate antiviral compound to combat seasonal IAV infection.
Furthermore, we confirmed RdRp assay system could correctly reflect the natural IAV (H1N1) infection.Based on this result, we could hypothesize that H5N1 RdRp assay system might have the relevance with the IAV H5N1 natural infection in drug evaluation.

| DISCUSSION
In the present study, we constructed a plasmid-mediated cell-based assay system for IAV H5N1 vRNA replication/transcription inhibitors based on prior studies. 26,27Using this RdRp assay system, we found a potential antiviral, vidofludimus, for influenza virus wild-type and mutation-type infection.
As the step of replication/transcription in viral lifecycle is the main target in the development of antivirals, we focused on the RdRp of IAV for the screening of drugs.A previous study also demonstrated that the viral RdRp subunits PA and PB1 are associated with a high pathogenicity. 4Several inhibitors such as oseltamivir or baloxavir marboxil could be used to treat highly pathogenic (HP) IAV infections.However, some IAV with drug-resistant mutation may cause problems in the future.Therefore, a convenient assay system for the screening of anti-IAV H5N1 compounds are needed.The plasmid-mediated assay system for influenza virus is well-established and widely used.As the plasmid-mediated assay system expresses each viral polymerase subunit using a single plasmid, the construction of the plasmid with several drug-resistant mutations could be easily developed.
A previous study reported that multiple viral mutations induce resistance to current drugs for seasonal influenza (e.g., baloxavir marboxil and favipiravir). 42However, the mutation(s) inducing resistance to drugs for IAV H5N1 have remained unclear.In the present study, we used the most common baloxavir marboxilresistance mutation PA I38T in IAV H1N1 and H3N2 strains for the treatment of favipiravir.This may have been due to the high concentration used.
Recently unique mutation was reported in PB2 of IAV H5N1 in farmed mink.T271A in PB2 was detected only in mink IAV H5N1 but not in avian IAV H5N1. 22This mutant was also reported in the pandemic of hog IAV H1N1 in 2009. 21So we constructed a plasmid containing the mutation mtPB2 T271 and evaluated its effects on viral polymerase activity and inhibitors.We found that the mtPB2 T271A mutation enhanced IAV H5N1 RdRp activity.This might be a key point for understanding how the avian HP IAV (H5N1) caused an outbreak among mammal minks.
In the case of vidofludimus, the main antiviral mechanism is the inhibition of DHODH.We confirmed that inhibitory activity was through blocking the oxidation of DHO to OA by DHODH in the de novo synthesis of pyrimidines (Figure 3).4][45] All of these DHODH inhibitors significantly reduced sNluc signals (Supporting Information: Figure S4B).Interestingly, inhibitory activity of the BAY2402234 (IC 50 : 1.2 nM) against DHODH was 100-fold stronger than that of vidofludimus (IC 50 : 134 nM).Against our expectation, the antiviral activity of BAY2402234 was lower than that of vidofludimus (Supporting Information: Figure S4B).This result suggests that the anti-IAV activity of vidofludimus might also include other pathways.
To clarify the mechanism of anti-IAV activity of vidofludimus, further exploration is needed.
To further test the correlation of the antiviral activity of vidofludimus between natural IAV infection and RdRp assay system, we produced infectious influenza virus (H1N1) belonging to IAV same as H5N1.We also constructed H1N1 strain based RdRp assays system and compared it with the infection system.As we expected, similar results were observed between natural infection and RdRp assay system (Figure 4 and Supporting information Figure S3E).
Based on this result, we could hypothesize that the H5N1 RdRp assay system might reflect the results of antiviral activities evaluated by the natural infection of IAV H5N1.To directly prove this hypothesis, we will plan to collaborate with other laboratories with BSL3 facility in the near future.
Previous reports showed inhibitors targeting DHODH had broad-spectrum antiviral activity against RNA viruses. 29Here, we also evaluate the antiviral activity of vidofludimus against another type of influenza virus.We constructed the recombinant plasmids using the cDNA of B/Yamagata/16/1988 (IBV) for RdRp assay.
The result showed vidofludimus also carried highly antiviral activity to IBV (Supporting information Figure S5).The antiviral activity of vidofludimus shown in this study for IAV H1N1, H5N1 and IBV might also be employed for the other types/subtypes of influenza viruses and other RNA viruses.We also plan to determine that in the future.
In conclusion, we found vidofludimus as a potent antiviral to influenza virus.

2. 8 |
Cytopathic effect (CPE) assay MDCK cells were plated into 96-well plate (7 × 10 3 cells/well).After 24 h, the cells were washed twice with PBS and inoculated with a MOI 0.002 of influenza virus at 37°C for 2 h.The inoculum was removed, and cells were covered with 100 µL of 0.3% FBS-DMEM containing the indicated concentration of compounds in the presence of 1 µg/mL TPCK-trypsin (T1426, Sigma-Aldrich).After 72 h of incubation, the cells were washed 3 times with PBS and stained by 0.6% Coomassie brilliant blue (CBB).

2. 9 |
RNA isolation, reverse transcription (RT), and quantitative PCR (qPCR) MDCK cells in 6-well plate were harvested after 48 h (cells) or 72 h (supernatant) postinfection with or without compounds and RNAs were extracted using RNeasy Mini kit (QIAGEN) for cells or using ISOGEN-LS (NIPPON GENE) for supernatant according to the manufacturer's protocol.Total RNA (2 µg) for cells or the viral RNA of the supernatant (0.25 mL) was used to perform RT by M-MLV Reverse Transcriptase (Thermo Fisher Scientific) with Random hexamers primers (Thermo Fisher Scientific).The viral RNA loads in infected cells were determined by qPCR with THUNDERBIRD SYBR qPCR Mix (TOYO-BO) on Applied Biosystems StepOne Real-Time PCR Systems according to the manufacture's protocol.PCR reactants were electrophoresed by 3% Tris-EDTA agarose gels.The relative RNA expression of cells was calculated using the 2^− ΔΔCt method with GAPDH as an internal reference gene.The viral RNA of supernatant was calculated using the 2^− ΔCt method.The primer sequences used for analysis are shown in Supporting Information: Table.
,40,41  HEK293 cells were transfected with wild type RdRp or mutation type RdRp (PA I38T or PB2 T271A, or a combination of two mutations) expressing plasmids and treated with each drugs.The sNLuc value in untreated cells transfected with wild-type RdRp expressing plasmids was used as 100% (Figure2B).Interestingly, the RdRp with mtPB2 T271A significantly enhanced viral polymerase activity by about 1.3-fold (p = 0.03) compared to the wild-type.Baloxavir marboxil and vidofludimus showed high antiviral activity against RdRp with mtPB2 T271A.It is noteworthy that vidofludimus could maintain high antiviral activity even when the RdRp carried both mutations (Figure2B).

3. 3 |
Vidofludimus prevents IAV H5N1 replication/ transcription by inhibition of DHODHVidofludimus is an orally active inhibitor of DHODH, therefore, the activity of vidofludimus will lead to the depletion of the intracellular pyrimidine pool.There are two main pathways for the pyrimidine biosynthesis in mammalian cells.The de novo pathway was through a multistep process starting from glutamine and produced an intermediate product dihydroorotate, then the dihydroorotate will be converted to orotate by DHODH at the inner mitochondrial membrane, and finally produced uridine triphosphate (UTP) or F I G U R E 1 RdRp-dependent screening assay for IAV H5N1 vRNA replication/transcription inhibitors.MedChemExpress (MCE) Phase II drop compound library was tested using RdRp assay system.HEK293 cells were treated with each compound for 24 h.(A) Inhibitory activity of the compounds.HEK293 cells were transfected with RdRp expressing plasmids (PB1, PB2, PA, NP) and reporter plasmid (sNLuc).RdRp levels of the compound treated cells were evaluated according to the relative sNLuc activity, when the sNLuc level in cells without compound is taken as 100%.(B) Cell viability of the treated cells.Cell viability in treated cells with the compound was evaluated by WST-1 assay.The absorbance level in cells without compound is taken as 100%.(C) The structure of vidofludimus.(D and E) The antiviral activity and cytotoxicity of vidofludimus for IAV H5N1 wild type were assessed by EC 50 and CC 50 values, respectively.The data were shown as means ± SD. cytidine triphosphate (CTP).The other pathway is the salvage pathway uses the extracellular nucleoside pool, or intracellular nucleic acid degradation product through uridine/cytidine kinase (UCK) and finally produces UTP or CTP (Figure 3A).To verify whether the pyrimidine biosynthesis pathway is implicated in the anti-IAV activity of vidofludimus, we investigated the affection of the addition of exogenous pyrimidine (UTP or CTP) ribonucleosides, uridine or cytidine, to the antiviral activity of vidofludimus.We cotransfected PB1, PB2, PA, NP, and sNLuc plasmids of IAV H5N1 into HEK293 cells, and treated cells with 10 μM of vidofludimus using mediums that contain a gradually diluted concentration of uridine or cytidine after 24 h posttransfection.Uridine or cytidine themselves had no effect on RdRp activity.The sNLuc signals in the cells with 10 μM of vidofludimus were restored by the addition of uridine or cytidine in a dose-dependent manner (Figure 3B,C).
Influence of drug-resistant mutant (mtPA I38T) and polymerase activity enhancing mutant (mtPB2 T271A) on the antiviral activity of compounds at clinically achievable blood concentrations.(A) The antiviral activity of vidofludimus was evaluated for IAV H5N1 mutation type (PA I38T) by determining its EC 50 values.The data were shown as means ± SD. (B) HEK 293 cells were transfected with RdRp expressing plasmids (PB1, PB2, PA, NP, mtPA I38T, and mtPB2 T271A) and reporter plasmid (sNLuc).Then, cells were treated with baloxavir marboxil, favipiravir, and vidofludimus at their C max .The sNLuc activities were determined at 24 h postcompounds treatment.In each group (wild type [gray]; mtPA I38T [green]; mPB T271A [blue]; mtPAI38T and mtPB2 T271A [orange]), the sNluc activity of the drug treated cells was compared with that of the untreated cells.The data were shown as means ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.F I G U R E 3 Vidofludimus prevents replication/transcription of IAV H5N1 via the inhibition of DHODH activity in plasmids-transfected cells.(A) The pyrimidine biosynthesis pathway.(B and C) The effect of uridine or cytidine for the anti-IAV H5N1 RdRp activity of vidofludimus.HEK293 cells were transfected with IAV H5N1 RdRp expressing plasmids and treated with 10 μM vidofludimus in the presence of gradually diluted concentrations of uridine or cytidine for 24 h after transfection.The RdRp activity was evaluated using sNLuc as an indicator.The data were shown as means ± SD, **p < 0.01, ***p < 0.001, ****p < 0.0001 compared with each treated by vidofludimus alone group.(D and E) The effect of L-hydroorotic acid or orotic acid for the anti-IAV H5N1 RdRp activity of vidofludimus.HEK293 cells were transfected with IAV H5N1 RdRp expressing plasmids and treated with 10 μM vidofludimus in the presence of gradually diluted concentrations of L-hydroorotic acid or orotic acid for 24 h after transfection.The sNluc activity of the drug treated groups was compared with each treated by vidofludimus alone group.The data were shown as means ± SD. *p < 0.05, ***p < 0.001.F I G U R E 4 Vidofludimus showed highly antiviral activity on seasonal IAV H1N1 infection in vitro.(A) MDCK cells were infected with IAV and treated with baloxavir marboxil, favipiravir, and vidofludimus at gradually diluted concentrations.After 72 h posttreatment, cells were stained by CBB reagent.(B) MDCK cells were infected with IAV and treated with baloxavir marboxil, favipiravir, and vidofludimus at a concentration of 2.5 μM.Cells were extracted at 48 h posttreatment and subjected to the western blot analysis.The relative band-intensities of each lane were calculated with an internal reference protein beta-actin, and compared with compound-free infection control (100%).(C) MDCK cells were infected with IAV and treated with baloxavir marboxil, favipiravir, and vidofludimus at a concentration of 2.5 μM.Intracellular RNA was extracted and purified at 48 h posttreatment.The RNA levels were determined by RT-qPCR.The PCR reactants were also electrophoresed by 3% Tris-EDTA agarose gels.The data were shown as means ± SD. (D) MDCK cells were infected with IAV and treated with vidofludimus by a gradient dilution.Cells were extracted at 48 h posttreatment and subjected to the western blot analysis.The relative band-intensities of each lane were calculated with an internal reference protein beta-actin, and compared with compound-free infection control (100%).(E) MDCK cells were infected with IAV and treated with vidofludimus by a gradient dilution (from 1.5 µM).Intracellular RNA was extracted and purified at 48 h posttreatment.The RNA levels were determined by RT-qPCR.The data were shown as means ± SD.
Conceptualized and designed the study; resources; writing-review and editing; project administration.Jiazhou Li: Designed the study; validation; performed experiments and curated the data; methodology; writing-original draft.Midori Takeda: Methodology; resources; validation; writing-review and editing.Mikiko Imahatakenaka: Performed experiments.All authors were accountable for all aspects of the work.
assay system.Vidofludimus could inhibit replication/transcription both of IAV H5N1 wild-type and mutation-type at clinically achievable blood concentrations.Compared with vidofludimus, the antiviral activity of baloxavir marboxil for mutation-type was drastically reduced (C max /wt EC 50 : 18.37 vs. C max /mt EC 50 : 0.28) (Table1).Vidofludimus maintains its antiviral activity for IAV H5N1 with mtPA I38T as well as wild type (C max /wt EC 50 : 14.38 vs. C max /mt EC 50 : 14.31).To evaluate the effect of drug resistant mutation (I38T)