Pharmacological characterization of the mechanism of action of R523062, a promising antiviral for enterovirus D68

Abstract

Enterovirus D68 (EV-D68) is a re-emerging virus that causes moderate to severe respiratory diseases in children. In severe cases, EV-D68 infection can lead to neurological complications called acute flaccid myelitis (AFM). There is currently no antiviral or vaccine available for EV-D68. The goal of this study is to delineate the mechanism of action of a promising antiviral drug candidate R523062 that was identified through a phenotypic cytopathic effect (CPE)-based high-throughput screening. R523062 inhibits multiple contemporary EV-D68 strains with single-digit micromolar EC₅₀ values and is less effective against the enterovirus A71 strains. Resistant mutants identified through serial viral passage experiments were mapped to four viral proteins including VP1-G178S, 2A-V112I, 2C-I227L/Q322R, and 3A-V54A. The involvements of VP1-G178S, 2A-V112I, and 3A-V54A mutants in drug resistance were ruled out by the drug time-of-addition experiment, protease enzymatic assay, and the plaque assay with recombinant virus, respectively. In contrast, recombinant virus encoding the 2C-I227L/Q322R double mutants confers significant drug resistance, which is consistent with the result from serial passage experiments. Thermal shift binding assay showed R523062 binds to the wild-type EV-D68 2C, 2C-Q322R, but not 2C-I227L or 2C-I227L/Q322R, confirming 2C as the direct drug target of R523062 and 2C-I227L alone confers drug resistance. The 2C inhibitor R523062 also showed additive antiviral activity with the viral 2A protease inhibitor telaprevir as well as the viral capsid VP1 inhibitor R856932. Collectively, this study identified a promising EV-D68 antiviral drug candidate R523062 with a confirmed mechanism of action by targeting the viral 2C protein.

Graphical Abstract

Enterovirus D68 (EV-D68) has been identified as a possible cause of recent outbreaks of acute flaccid myelitis (AFM), a rare but severe poliomyelitis-like spinal cord syndrome that can result in permanent paralysis and even death.^1–2 EV-D68 is a member of the genus Enterovirus (EV), which belongs to the family Picornaviridae. The EV genus also contains several other significant human pathogens including EV-A71, poliovirus, rhinovirus, and coxsackievirus. EV-D68 mainly infects young children and leads a broad-spectrum of disease manifestation, ranging from mild symptoms including running nose, fever, muscle aches to severe symptoms including wheezing, difficulty breathing, and even permanent paralysis and death.

The EV-D68 virion is about 30 nm in size and surrounded by an icosahedral capsid made of viral capsid proteins VP1-VP4.³ EV-D68 genome is a single-stranded positive-sense RNA, approximately 7.5 kb, containing a long and highly structured 5′ untranslated region (UTR) (~700 base), a single open reading frame and a short 3′ UTR (< 100 base). The long 5’ UTR contains an internal ribosome entry site (IRES), which initiates viral genome translation into a polyprotein. This polyprotein is initially processed by the viral 2A and 3C proteases into three precursor proteins, P1, P2, and P3, which are further cleaved into four structural capsid proteins (VP4, VP2, VP3, and VP1) from P1 region, and seven mature nonstructural proteins (2A, 2B, 2C, 3A, 3B, 3C, and 3D) from P2 and P3 regions.⁴

There is currently no antiviral or vaccine available for EV-D68 and treatment is mainly limited to supportive care to help relieve symptoms.⁵ Since the first major EV-D68 outbreak in the United States in 2014, research efforts started to focus on developing EV-D68 antivirals. Majority of the drug candidates were identified through drug repurposing approach, and a few were discovered through target-based design or screening.⁵ Telaprevir, an FDA-approved drug used for the treatment of hepatitis C virus (HCV) infections, was found to be a potent antiviral against EV-D68 by targeting the 2A^pro enzyme.⁶ Rupintrivir, a protease inhibitor targeting the viral 3C protease, had broad-spectrum antiviral activity against both the historic Fermon strain as well as the contemporary EV-D68 strains with high potency in cell culture.^7–8 Pleconaril, a viral capsid VP1 inhibitor developed to treat common cold from rhinovirus infections, was also effective against EV-D68 infections.^{7, 9} Compound R856932, another viral capsid VP1 binder also displayed potent in vitro antiviral activity against EV-D68 in cell culture.⁹ Fluoxetine, an FDA-approved drug used for the treatment of major depression and anxiety disorders, inhibited EV-D68 replication in cell culture by targeting viral 2C protein.^10–11 However, in a small, non-randomized retrospective study, fluoxetine did not seem to restore muscle strength in AFM patients.^12–13 Moreover, dibucaine, an FDA-approved topical anesthetic drug and its analogs had potent in vitro anti-EV-D68 activity by targeting the viral 2C protein.¹⁴ For more details regarding preclinical development of EV-D68 antivirals, please refer to recent reviews.^{5, 15}

Among the EV-D68 replicative enzymes, the viral 2C protein is one of the most conserved nonstructural proteins with multiple associated functions, rendering it an attractive target for developing direct-acting antivirals.^14–17 The 2C protein is proposed to play important roles in RNA replication,¹⁸ RNA binding, ¹⁹ and membrane rearrangement ^20–21 and encapsidation.²² Moreover, the 2C protein was proposed to function as a helicase, though no experiment was performed to directly confirm the helicase activity.¹⁶

Here we report our discovery of a promising EV-D68 antiviral R523062 and its drug target deconvolution through a consortium of biochemical, biophysical, and reverse genetics techniques. The results collectively suggest viral 2C protein is the direct target of R523062. The combination therapy potential of R523062 was further evaluated with both the viral 2A protease inhibitor telaprevir and the capsid inhibitor R856932. Overall, this study confirmed R523062 as a potent EV-D68 antiviral by targeting the viral 2C protein.

RESULTS AND DISCUSSION

Compound R523062 inhibited multiple EV-D68 and EV-A71 strains

In order to search novel EV-D68 antivirals, we carried out a cell-based cytopathic effect (CPE) screening assay against EV-D68 US/MO/14–18947 strain in RD cell.⁹ We chose EV-D68 US/MO/14–18947 strain because it is one of the contemporary EV-D68 strains that is associated with AFM.² Compound R523062 was identified as one of the positive hits when screened at 15 μM (92% CPE protection).

To confirm the primary screening assay results, R523062 was re-purchased and titrated against various contemporary EV-D68 and EV-A71 strains. Compound R523062 inhibits all the strains tested in a dose-dependent manner in the CPE assay. The EC₅₀ value against the EV-D68 US/MO/14–18947 (clade B1) strain was 4.1 μM. Compound R523062 displayed similar efficacy against other EV-D68 strains in different clades with EC₅₀ values ranging from 2.3 to 6.4 μM (Table 1). For EV-A71 strains, we observed varying efficacy against EV-A71 strains with EC₅₀ values ranging from 9.6 to 55.7 μM (Table 1), which were less effective than that of EV-D68. R523062 was not cytotoxic in RD cells at the observed antiviral concentration and the CC₅₀ value was 119.6 ± 22.0 μM (Table 1), which gave an selectivity index (SI) ranging from 19 to 52 for various EV-D68 strains (Table 1). Overall, R523062 displayed a broad-range antiviral activity against contemporary EV-D68 strains from different clades with a favorable SI.

Table 1.

Antiviral Potency and Cellular Cytotoxicity of R523062 against EV-D68 and EV-A71 in RD Cells^a

R523062 RD cells: CC50 = 119.6 ± 22.0 (μM)
Virus strains	EC50 (μM)b	SIc
EV-D68 US/MO/14–18947 (clade B1)	4.1 ± 0.8	29
EV-D68 US/MO/14–18949 (clade B1)	2.4 ± 0.3	50
EV-D68 US/IL/14–18952 (clade B2)	4.9 ± 3.5	24
EV-D68 US/IL/14–18956 (clade B2)	2.3 ± 0.5	52
EV-D68 US/KY/14–18953 (clade D)	6.4 ± 0.8	19
EV-A71 Tainan/4643/1998	18.0 ± 4.3	7
EV-A71 MP4	55.7 ± 7.4	2
EV-A71 US/CT/2016–19519	9.6 ± 2.5	12
EV-A71 US/AK/2016–19516	44.5 ± 10.2	3

^aThe antiviral potency EC₅₀ values and the cellular cytotoxicity CC₅₀ values were determined by CPE assay and cytotoxicity assay, respectively. For details, please refer to the methods section.

^bEC₅₀ (μM) = mean ± standard deviation. The values are the mean ± standard deviation from three replicates.

^cSI: Selection Index

Compound R523062 inhibits Viral RNA and Protein Synthesis.

We further confirmed the antiviral effect of R523062 against EV-D68 by real-time quantitative PCR (RT-qPCR), western blot, and immunofluorescence imaging assay. For this, RD cells were infected with EV-D68 US/MO/14–18947 virus at a MOI of 1 either in the presence or the absence of testing compounds (Figure 1). The viral 3C protease inhibitor rupintrivir was included as a positive control. At 9 post-infection (hpi), RD cells were fixed for immune-staining (Figure 1A), or harvested for quantification of viral capsid protein VP1 level and viral RNA level by western blot (Figure 1B) and RT-qPCR (Figure 1C), respectively. It was found that R523062 treatment significantly reduced VP1 protein immnunofluorescence signal (Figure 1A bottom two panels), similarly as rupintrivir treatment (Figure 1A second panel). This result is consistent with the western blot assay results, which showed that R523062 treatment reduced the viral VP1 protein expression level in a dose dependent manner (Figure 1B). Moreover, R523062 greatly reduced viral RNA level when quantified by RT-qPCR (Figure 1C). Taken together, the antiviral activity of R523062 against EV-D68 was confirmed by immunofluorescence assay, western blot, and qRT-PCR assay.

Figure 1.

Compound R523062 reduced the EV-D68 viral RNA and protein levels. (A) R523062 inhibited EV-D68 VP1 protein synthesis in immunofluorescence assay. RD cells were infected with EV-D68 strain US/MO/14–18947 at a MOI of 1, fixed at 9 hpi, and stained with anti-VP1 and DAPI for detecting VP1 and nucleus, respectively. Drug concentrations used were shown in the figure. Rupintrivir was included as a positive control. (B-C) R523062 reduced the EV-D68 protein level and viral RNA levels. RD cells were infected with EV-D68 strain US/MO/14–18947 at a MOI 1 in the presence of various concentrations of testing compounds. At 9 hpi, cells were harvested for viral protein quantification by western blot (B) or for viral RNA quantification by RT-qPCR (C). Asterisks indicate a statistically significant difference in comparison with the DMSO control (one-way analysis of variance analysis by Prism 5, *** P < 0.001).

Selection of EV-D68 Mutants That Are Resistant to Compound R523062 via Serial Viral Passage Experiment.

In order to identify the drug target(s) of R523062, we carried out a serial viral passage experiment with increasing drug selection pressure of R523062 with an attempt to select drug-resistant mutants (Table 2). If resistance were to emerge, sequencing the viral genome of the mutant viruses might be able to pinpoint putative drug target(s). For this, EV-D68 US/MO/14–18947 strain was chosen for the serial viral passage experiment because: 1) this is a clinically relevant EV-D68 strain and has been shown to be associated with AFM, and 2) our laboratory has established the reverse genetics system for this strain, allowing the generation of recombinant viruses to confirm the drug resistance.⁶

Table 2.

Serial Viral Passage Experiment of R523062 against EV-D68 US/MO/14–18947 virus.^a

Passage #	Selection pressure (μM)	EC50 (μM)
0	0	4.1 ± 0.8
1	10	N.D.
2	20	>20
3	40	>50

^aEV-D68 US/MO/14–18947 (MOI = 0.1) was treated with compound R523062 at the indicated concentrations (μM); EC₅₀ (μM) = mean ± standard deviation. The values are the mean ± standard deviation from three replicates in CPE assay. N.D.: Not determined.

For the R523062 resistance selection experiment, it was found that resistance emerged at passage 3 (P3), and the EC₅₀ of R523062 against the P3 virus was over 50 μM, compared to the EC₅₀ of 4.1 μM for the P0 wild-type virus (Table 2). In contrast, we did not observe any mutation at P3 in the absence of drug selection pressure. We subsequently sequenced the whole viral genome of R523062-P3 virus and identified five mutations located in four viral proteins, namely G178S in capsid VP1 protein, V112I in protease 2A protein, I227L and Q322R in 2C protein, and V54A in 3A protein.

Drug Time-of-Addition Experiments Indicated That Compound R523062 Did Not Inhibit Viral Entry, But At A Later Stage of Viral Replication.

Since mutants are mapped to four viral proteins VP1, 2A, 2C, and 3A, we next set to confirm or rule out the relevance of each one of them in the R523062 drug resistance. To examine the relevance of viral capsid protein VP1, we first performed the drug time-of-addition experiment. In this experiment, testing compound was added at different stages of a single viral replication cycle at either before, during, or after viral infection, and viruses in the supernatant were harvested at 14 hpi, which approximately equals one cycle of viral replication. The viral titer was subsequently quantified by plaque assay. VP1 inhibitors are expected to exert the antiviral activity when added during viral infection, and lose the antiviral activity when added after viral infection. In contrast, 2A, 2C and 3A inhibitors are expected to maintain antiviral activity even when added after viral infection. Two compounds, pleconaril and dibucaine, were added as controls for the time-of-addition experiment. Pleconaril is a VP1 inhibitor and dibucaine is a 2C inhibitor.

As shown in Figure 2A, pleconaril only exerted its inhibitory activity during attachment and/or entry stage of the virus replication cycle, which is consistent with its inhibition mechanism by binding viral capsid VP1 protein therefor blocking the virus attachment and entry.⁹ Unlike pleconaril, compound R523062 had no effect on viral attachment or entry (Figure 2B, 0 h to 1 h). Compound R523062 completely blocked the production of progeny virus even when added at 5 hpi. When treatment started at 7 hpi or onwards, R523062 gradually lost its antiviral potency (Figure 2B). Dibucaine treatment (Figure 2C) displayed a very similar pattern as compound R523062: no inhibition at attachment stage (0 h to 1 h); complete inhibition at 1 hpi, 3 hpi and 5hpi; gradual loss of potency starting from 7 hpi. One of the proposed 2C functions is its involvement in viral genome replication. Picornavirus genome replication starts at 4 hpi or later,²³ which is consistent with our drug time-of-addition result of the 2C inhibitor dibucaine.

Figure 2.

Time-of-addition experiments by quantify progeny virus yield in cell culture supernatant by plaque assay (A-C) and within the cell by immunofluorescence staining (D). RD cells were infected with EV-D68 US/MO/14– 18947 at a MOI 0.01 at the 0 h time point. At 1 hpi, inoculant virus was removed and washed with PBS buffer; progeny viruses in the cell culture supernatant were harvested at 14 hpi and were quantified with plaque assay. The red arrow represents the period of time that the pleconaril (A), compound R523062 (B) or dibucaine (C), was present in the cell culture. The concentrations used for pleconaril, compound R523062 and dibucaine were 12.5 μM and 30 μM and 30 μM, respectively. Asterisks indicate a statistically significant difference in comparison with the DMSO control (one-way analysis of variance analysis by Prism 5, * P < 0.05; ** P < 0.01; *** P < 0.001). The value is the mean of two independent experiments ± S.D. (D) Time-of-addition experiments by immunostaining VP1 protein inside the host cells. RD cells were infected with EV-D68 US/MO/14–18947 at a MOI 1 at the 0 h time point, and at 1 hpi, inoculant virus was removed and washed with PBS buffer. Pleconaril (left column), Compound R523062 (central column) or dibucaine (right column) was applied at −1, 0, or 2 hpi to 6 hpi; for the 0 h → 1 h condition, compound was only present during the infection stage for 1 h, and it was washed away and fresh media was applied. The concentrations used for pleconaril, compound R523062 and dibucaine were 12.5 μM, 30 μM, and 30 μM respectively. Cells were fixed at 9 hpi and viruses were detected via staining VP1 protein with anti-VP1 antibody. The images were representatives of three independent experiments.

To further validate the results, we performed a second set of time-of-addition experiment by quantifying the intracellular viral VP1 protein level using immunofluorescence imaging. For this experiment, RD cells were infected with EV-D68 US/MO/14–18947 at a MOI of 1, and cells were fixed at 9 hpi and VP1 protein level was quantified via immunofluorescence imaging. Compound R523062, pleconaril or dibucaine was added to the virus-infected cells at the period depicted in Figure 2D. Similar to the above time-of-addition experiment, pleconaril blocked viral entry, but had no effect on viral replication when added after viral infection. In contrast, compound R523062 had no effect on viral entry (0 h to 1 h), but completely blocked viral VP1 protein synthesis when it was added at 2 hpi and 4 hpi. When it was added at 6 hpi, R523062 started to lose the antiviral potency. Once again, dibucaine showed indistinguishable pattern as R523062 in the immunofluorescence imaging-based drug time-of-addition assay, suggesting these two compounds might have a similar mechanism of action by targeting the viral 2C protein. Collectively, these two sets of time-of-addition experiments suggest VP1 is not the drug target, and 2C might be the plausible drug target of R523062.

Compound R523062 Does Not Inhibit 2A Protease (2A^pro) Activity.

One of the mutations identified in the serial viral passage experiment was 2A-V112I. The 2A protein is an EV-D68 viral cysteine protease and has been shown to the drug target of telaprevir.⁶ In order to confirm or rule out that compound R523062 targets 2A^pro, we performed protease proteolytic functional assay using the FRET-based substrate Dabcyl-KIRIVNT/GPGFGG-E(Edans).⁶ In this assay, 1 μM 2A^pro was pre-incubated with DMSO, 1 μM telaprevir or 100 μM compound R523062 for 1 hr at 30 °C, then 20 μM of FRET substrate was added to reaction mixture to initiate the proteolytic reaction, and the fluorescence signal was monitored for 2 hrs (Figure 3). It was found that 1 μM telaprevir completely blocked 2A^pro protease activity, while 100 μM compound R523062 had no activity on 2A^pro proteolytic activity. Therefore, it can be concluded that 2A^pro is not the direct drug target of R523062 and 2A-V112I is not relevant to the drug resistance against R523062.

Figure 3.

Compound R523062 did not inhibit EV-D68 2A protease activity. The proteolytic reactions of 1 μM EV-D68 2A^pro protein in the absence of compound (red) or in the presence of 1 μM telaprevir (black) or 100 μM compound R523062 (blue) with 20 μM FRET substrate ⁶ were monitored up to 2 h at 30°C.

Mutation I227L in 2C Protein Confers Drug Resistance against R523062.

After ruling out VP1 and 2A^pro as the drug targets of R523062, we next set to examine the relevance of 2C and 3A as the plausible drug target(s). In total, three mutants were identified from the serial viral passage experiment, two located at the 2C protein I227L, and Q322R, one located at the 3A protein V54A. To elucidate which mutant(s) confers drug resistance against R523062, we used reverse genetics to generate recombinant viruses that harboring the above-mentioned 2C and 3A mutants.^{6, 9} The reverse genetics was based on the EV-D68 US/MO/14–18947 genome, which allows direct comparison of the antiviral potency between the recombinant mutant virus and the wild-type virus. As the R523062-P3 mutant virus was sequenced as a mixture, we were not sure whether the 2C-I227L and 2C-Q322R mutants emerged from one virus or two viruses. As such, we attempted to generate recombinant EV-D68 viruses harboring either the single mutant or the double mutants. However, after several trials with various transformation conditions, we could not recover recombinant virus with the 2C-I227L mutant, indicating this mutant might be lethal for the viral replication. Nevertheless, when 2C-I227L was introduced together with the 2C-Q322R, we successfully recovered the double-mutant recombinant virus. We named these recombinant viruses as rMO WT, rMO 2C-Q322R, rMO 2C-I227L/Q322R, and rMO 3A-V54A (Table 3). The drug sensitivity of these recombinant viruses against compound R523062 was evaluated in both the plaque assay (Figure 4A, Table 3) and cytopathic effect assay (Table 3), and compared with recombinant rMO-WT virus. The introduction of V54A mutation in 3A protein did not alter the antiviral potency of R523062 (Figure 4A, Table 3), suggesting 3A is not the relevant drug target. This leaves viral 2C protein as the only plausible drug target. Introduction of Q322R alone in 2C protein did not significantly increase R523062 EC₅₀ values in both plaque and CPE assays. However, the recombinant double-mutant virus rMO 2C-I227L/Q322R displayed significant increased EC₅₀ values in both assays (Figure 4A, Table 3). The drug resistance of rMO 2C-I227L/Q322R against R523062 was further evaluated in the viral titer reduction assay (Figure 4B). It was found that there was no significant difference (~0.3 log₁₀) between the growth curves of the rMO 2C-I227L/Q322R and the rMO WT, suggesting the double mutant did not affect the replication of the virus. R523062 had no effect on the replication kinetics of rMO 2C-I227L/Q322R at 10 μM, while it completely inhibited the amplification of the rMO WT virus. Collectively, although virus containing 2C-I227L mutation alone was not viable and could not be rescued, it can be concluded that 2C-I227L mutation alone confers R523062 resistance. The 2C-Q322R mutant is a compensatory mutant that renders the 2C-I227L mutant virus viable, and it is not involved in drug resistance.

Table 3:

Drug sensitivity of R523062 against recombinant EV-D68 US/MO/14–18947 viruses.

Recombinant viruses	CPE EC50 (μM)	PRA EC50 (μM)
rMO WT	4.9 ± 0.4	3.4 ± 1.1
rMO 2C-Q322R	6.1 ± 0.6	4.2 ± 1.2
rMO 2C-I227L/Q322R	53.4 ± 1.8	20.5 ± 4.9
rMO 3A-V54A	4.3 ± 0.4	3.1 ± 0.7

CPE: Cytopathic effect assay

PRA: Plaque reduction assay

EC₅₀ (μM) = mean ± standard error. The values are the mean ± standard deviation from three replicates in the CPE assay.

Figure 4.

Drug sensitivity and growth curves of rMO viruses. (A) Representative plaque reduction assay with compound R523062 against recombinant rMO WT, rMO 2C-Q322R, and rMO 2C-I227L/Q322R and rMO 3A-V54A viruses. Approximately 100 pfu/well of recombinant viruses were applied to RD cell monolayers. After washing out the inoculum, 1 to 100 μM of compound R523062 was added to the virus-infected cell culture in the presence of 1.2% Avicel overlay. 3.0 μM telaprevir (Tel), or 1.0 μM compound R856932 was included as positive controls. RD cells were stained with crystal violet 3 days after infection. The images were representatives of three independent experiments. (B) Multicycle growth curves of rMO WT and rMO I227L/Q322R with or without R523062. RD cells were infected with the reconstituted viruses rMO WT or rMO 2C-I227L/Q322R separately at MOI of 0.01 in the presence of DMSO or 10 μM R523062. Amplified viruses from supernatant were harvested 24 hr, 48 hr and 72 hr post infection and viral titers were determined by plaque reduction assay.

Recombinant rMO 2C-I227L/Q322R Showed Cross-Resistance against Known 2C Inhibitors.

To check whether the rMO 2C-I227L/Q322R virus confers cross-resistance against other known 2C inhibitors, we tested the antiviral potency of guanidine, dibucaine, and an optimized dibucaine analog 12a,¹⁴ against recombinant rMO 2C-I227L/Q322R virus. The rMO WT virus was included as a reference (Table 4). All three inhibitors R523062, guanidine, and dibucaine displayed reduced potency against rMO 2C-I227L/Q322R virus compared to the rMO WT virus, suggesting these compounds might target the same region of EV-D68 2C protein. In comparison, the rMO 2C-I227L/Q322R virus showed similar antiviral activity as the rMO WT virus against viral 2A protease inhibitor telaprevir and the viral VP1 inhibitor pleconaril, suggesting the 2C-I227L/Q322R did not have cross resistance against EV-D68 antivirals that do not target the 2C protein.

Table 4:

Drug sensitivity of known EV-D68 antivirals against EV-D68 WT and EV-D68 2C-I227L/Q322R viruses

Recombinant viruses	R523062 EC50 (μM)	Guanidine EC50 (μM)	Dibucaine EC50 (μM)	12a EC50 (μM)	Telaprevir EC50 (μM)	Pleconaril EC50 (μM)
rMO WT	4.98 ± 0.35	114 ± 19	2.09 ± 1.25	0.08 ± 0.03	0.51 ± 0.02	0.03 ± 0.00
rMO 2C-I227L/Q322R	53.4 ± 1.8	606 ± 189	8.13 ± 2.07	0.25 ± 0.07	0.53 ± 0.03	0.03 ± 0.00

EC₅₀ is determined via cytopathic effect (CPE) assay

EC₅₀ (μM) = mean ± standard error. The values are the mean ± standard error from three replicates.

Compound R523062 Directly Binds to EV-68 2C WT Protein but not the 2C-I227L Mutant.

To further confirm that compound R523062 directly targets the viral 2C protein, we performed thermal shift assay (TSA) with purified recombinant WT and mutant 2C proteins from EV-D68 US/MO/14–18947 strain covering residues 40 to 330. In the thermal shift assay, the increased melting temperature (T_m) indicates that the binding of a ligand stabilizes the target protein. Pleconaril, a known viral VP1 capsid inhibitor, was included as a negative control. Three known 2C inhibitors including dibucaine, guanidine, and 12a, were included as positive controls. For the EV-D68 2C WT, it was found that pleconaril had no effect on the 2C stability when tested at 50 μM (Figure 5, Table 5). This was expected as pleconaril is a known VP1 inhibitor and does not target the 2C protein. R523062, dibucaine, guanidine, and 12a shifted the T_m by 1.16, 2.43, 1.35, and 6.71 °C, respectively, at 50 μM (Figure 5A, Table 5), suggesting all these compounds bind directly to 2C protein. For the EV-D68 2C-I227L mutant, no significant ΔT_m was observed for R523062 or the other three positive control compounds dibucaine, guanidine, and 12a (ΔT_m < 1°C) (Figure 5B, Table 5). This result suggest the I227L mutant confers cross drug resistance against these 2C inhibitors. For the EV-D68 2C-Q322R mutant, it was noted that this mutant protein was less stable than the WT protein (T_m of 37.97 vs of 43.35 °C). Nevertheless, R523062, dibucaine, guanidine, and 12a all bind to this mutant with ΔT_m of 1.11, 4.31, 1.42, and 6.91 °C, respectively (Figure 5C, Table 5). The thermal shift pattern of the Q322R mutant is similar to that of WT upon drug binding, indicating this mutant does not confer drug resistance. For the EV-D68 2C-I227L/Q322R double mutant, no significant binding was observed for R523062 or the three positive control compounds (ΔT_m < 1°C), indicating drug resistance (Figure 5D, Table 5). Taken together, the thermal shift assay results indicate that: 1) R523062 binds to 2C protein and prevents it from thermal denaturation, similarly to dibucaine, guanidine, and 12a; 2) the 2C-I227L mutant alone or in combination with the Q322R mutant confer cross drug resistance against R523062 as well as other 2C inhibitors; 3) the 2C-Q322R mutant has no effect on 2C inhibitor binding. Since the recombinant rMO 2C-I227L could not be generated, the thermal shift assay results represented herein provided direct evidence that the 2C-I227L mutant alone confers drug resistance against R523062 and other known 2C inhibitors.

Figure 5.

Binding of R523062 and control 2C inhibitors to EV-D68 2C WT and mutant proteins using differential scanning fluorimetry (DSF). 4 μM of EV-D68 2C WT (A) or I227L (B) or Q322R (C) or I227L/Q322R (D) was incubated with DMSO (red), or 50 μM R523062 (black), or 50 μM Dibucaine (blue), or 1 mM Guanidine (magenta), or 50 μM 12a (green), raw data were plotted using Prism (v5) software to determine the T_m.

Table 5:

Thermal shift assay of EV-D68 2C WT and mutants with and without testing compounds.

Compound	EV-D68 2C WT Tm/ΔTm (°C)	EV-D68 2C-I227L Tm/ΔTm (°C)	EV-D68 2C-Q322R Tm/ΔTm (°C)	EV-D68 2C-I227L/Q322R Tm/ΔTm (°C)
DMSO	43.35/0	45.63/0	37.97/0	45.49/0
50 μM Pleconaril	43.57/−0.13	N.T.	N.T.	N.T.
50 μM R523062	44.51/1.16	45.38/−0.25	39.08/1.11	46.02/0.53
50 μM Dibucaine	45.78/2.43	46.29/0.66	42.28/4.31	45.86/0.37
1 mM Guanidine	44.7/1.35	45.81/0.18	39.39/1.42	45.83/0.34
50 μM 12a	50.06/6.71	46.24/0.61	44.88/6.91	45.94/0.45

N.T. = not tested; the assay conditions are the same as shown in Figure 5.

The Drug-Resistant rMO 2C-I227L/Q322R Virus is Less Fit Than rMO WT Virus in RD Cell Culture.

We compared the fitness of replication of the drug-resistant rMO 2C-I227L/Q322R virus with the rMO WT virus by a co-cultivation competition assay in RD cells ^{9, 24}. We started the co-cultivation by mixing the rMO 2C-I227L/Q322R virus with the rMO WT virus in a 100:1 ratio, and infected the RD cells with this virus mixture at a MOI of 0.01. Two days after infection, culture media supernatant was harvested and viruses were quantified with plaque assay, and 28,000 PFU progeny viruses were used for the next round of infection. We sequenced 2C protein coding region of the viruses at each passage and estimated the percentage of WT and mutant rMO 2C-I227L/Q322R virus by measuring the height of the nucleotide sequence electropherogram peaks at positions 227 and 322. As the passage number increases, we observed a gradual decrease of both the I227L and Q322R mutants, and a corresponding increase of the WT residues (Figure 6). It is interesting to note that residues 227 and 322 co-evolve, suggesting they might work together to maintain the normal functions of the 2C protein. The viral mixture appeared to reach an equilibrium at passage 5 with a ratio of 3:2 between WT and rMO 2C-I227L/Q322R virus. This result suggests that rMO 2C-I227L/Q322R has a slightly reduced fitness of replication than the WT virus.

Figure 6.

Competition growth assay to assess the replication fitness of rMO 2C-I227L/Q322R virus. 280 PFU rMO WT and 28,000 PFU rMO 2C-I227L/Q322R viruses (ratio = 1/100) were used to infect RD cells in a T25 flask. Two days after infection, culture media supernatant was harvested and viruses were quantified with plaque assay, and 28,000 PFU progeny viruses were used for the next round of infection. The 2C coding sequence of viruses from each passage was determined. Codon ATA is for Ile, and condon TTA is for Leu; Codon CAG is for Gln, and codon CGG is for Arg. (A) Electropherogram traces of 2C protein coding region at positions 227 and 322 from viruses at each passage. (B) The percentage of each virus was estimated by averaging the height of the nucleotide sequence electropherogram peaks at positions 227 and 322.

Compound R523062 Displays Additive Effect with 2A Protease Inhibitor Telaprevir as well as the Capsid VP1 Binder R856932 in Combination Therapy.

Combination therapy is one of the strategies to combat the drug resistance. While virus can quickly develop resistance to a single antiviral, it might take longer to develop resistance to two or more drugs simultaneously. Combination therapy has been actively explored in influenza, HIV, and chronic HBV and HCV infection treatment.^25–26 Here we explore the combination treatment potentials of R523062 with either the EV-D68 2A protease inhibitor telaprevir ⁶ or the capsid VP1 binder R856932⁹ in cell culture setting. Figure 7 depicts a plot of combination indices (CIs) versus the EC₅₀ values of compounds at different combination ratios. The red line indicates the additive effect, and the right upper area above the red line indicates antagonism, and the left bottom area below the line indicates synergy.²⁷ In both combination scenarios, the combination indices at all the combination ratios fell on the red line (Figure 6), indicating that R523062 exerts additive anti-viral effect with both the 2A protease inhibitor telaprevir, and the VP1 capsid binder R856932.

Figure 7.

Combination therapy of R523062 with telaprevir (A) or R856932 (B). The graph is a plot of combination indices (CIs) versus the EC₅₀ values of compounds at different combination ratios. The red line shows the additive effect, and above the line indicates the antagonism, and below the line indicates the synergy.²⁷ Data are mean ± SD of two independent experiments.

CONCLUSION

The neurotropic effect of EV-D68 virus and the biennial pattern of EV-D68 outbreaks strongly underscores the immediate need for the design and development of potent EV-D68 antivirals. For this, we recently carried out a high throughput screening using the EV-D68 induced cytopathic effect assay to identify potential drug candidates.⁹ Here we report our discovery of a promising hit R523062 and its drug target deconvolution. R523062 displayed a broad-range of antiviral activity against EV-D68 viruses from different clades with an acceptable selectivity index. The antiviral effect of R523062 was less potent against EV-A71 than EV-D68, suggesting that the 2C proteins between these two viruses are not highly conserved. Given the promising antiviral activity of R523062 against EV-D68, we were interested in identifying its drug target(s). Using a standard approach of evolving drug resistance through the serial viral passage experiment, we identified five mutants in four viral proteins in R523062 drug-resistant viruses. We subsequently ruled out the relevance of VP1-G178S and 2A-V112I in the phenotypical drug resistance through the drug time-of-addition assay and protease assay, respectively. The relevance of the 3A-V54A in drug resistance was ruled out by the plaque assay using the rMO 3A-V54A virus. Eventually, the drug target of R523062 was narrowed down to the viral 2C protein, and the 2C-I227L was shown to be the predominant drug-resistant mutant from the plaque assays using rMO 2C-I227L/Q322R and rMO Q322R viruses. Using thermal shift binding assay, we confirmed the direct binding between R523062 and EV-D68 2C WT protein. No binding was detected for the 2C-I227L mutant or the 2C-I227L/Q322R mutant, while the 2C-Q322R mutant showed similar binding patterns as the 2C WT, suggesting 2C-I227L alone confers drug resistance. Next, we aimed to dissect the fitness of replication of the drug-resistant rMO 2C-I227L/Q322R virus using the competition growth assay. It was found that rMO 2C-I227L/Q322R had a slightly reduced yet comparable fitness of replication compared to WT rMO. The 2C-I227L/Q322R double mutant also caused cross-resistance against other structurally distinct 2C inhibitors. As such, it is important to evaluate the antiviral potency of future 2C inhibitors against the rMO 2C-I227L/Q322R virus in order to avoid cross-resistance. Finally, combination therapy experiments showed that the antiviral effect of R523062 is additive with both the 2A^pro inhibitor telaprevir and the viral VP1 capsid inhibitor R856932.

The structural simplicity of R523062 and its confirmed mechanism of action render it a promising drug candidate for further development. It is expected that structure-activity relationship studies will yield more potent and selective analogs with favorable pharmacokinetic properties. It is also possible to develop analogs that do not have cross-resistance with R523062. Such studies are ongoing and will be reported when available. It was noted that an N-benzyl-N-phenylthiophene-2-carboxamide 5a, which shares structural similarity with R523062, had potent antiviral activity against the EV-A71 virus in cell culture.²⁸ However, the mechanism of action of 5a remains elusive. Nevertheless, continuous optimization based on this scaffold might yield broad-spectrum antivirals that are active against both EV-A71 and EV-D68.

METHODS

Chemicals, Cells, and Virus.

R523062 and R856932 were purchased from Sigma-Aldrich with Cat # of R523062 and R856932, respectively. Pleconaril was purchased from Sigma-Aldrich (Cat # SML0307). Telaprevir was purchased from Achemblock (Cat # F-4593). Rupintrivir was purchased from Sigma-Aldrich (Cat # PZ0315). Human rhabdomyosarcoma (RD; ATCC CCL-136) was maintained at 37 °C in a 5% CO₂ atmosphere and cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin−streptomycin antibiotics. The following reagents were obtained through BEI Resources, NIAID, NIH: Human Enterovirus D68, US/MO/14–18949, NR-49130; Enterovirus D68, US/MO/14– 18947, NR-49129; Enterovirus D68, US/IL/14–18952, NR49131; Enterovirus D68, US/KY/14–18953, NR-49132; Enterovirus D68, US/IL/14–18956, NR-49133; Human Enterovirus A71 (EV-71), Tainan/4643/1998, NR-471; Human Enterovirus A71 (EV-71), MP4, NR-472. Enteroviruses A71 (EV-71) US/CT/2016–19519 and US/AK/2016–19516 strains were obtained from Dr. William Nix at the Centers for Disease Control and Prevention under a material transfer agreement. All viruses were amplified in RD cells prior to infection assays.

Cytopathic Effect Assay (CPE) and Plaque Reduction Assay (PRA).

The high-throughput screening assay condition was previously described.⁹ The detailed procedures for these two assays were described in our previous publications.^{6, 9, 14} Briefly, for EV-D68 virus infections, RD cells were washed with PBS saline containing Mg²⁺ and Ca²⁺ and infected with virus diluted in DMEM with 2% FBS and 30 mM MgCl₂. Virus-infected cells were incubated for at least 1 h at 33°C in a 5% CO₂ atmosphere, followed by addition of compound as well as 1% penicillin-streptomycin, then incubated for 3 days. The CC₅₀ was measured similarly but in the absence of viral infection. For cytopathic effect (CPE) assays, cells were stained with 66 μg/ml neutral red for 2 h, and neutral red uptake was measured at an absorbance at 540 nm using a Multiskan FC microplate photometer (Thermo Fisher Scientific). The EC₅₀ and CC₅₀ values were calculated from best-fit dose-response curves using GraphPad Prism software. For plaque reduction assays, a 1.2% Avicel microcrystalline cellulose (FMC BioPolymer, Philadelphia, PA) overlay in DMEM media supplemented with 2% FBS and 30 mM MgCl₂ was used, and the cells were stained after 3 days at 33 °C as previously described.⁹ For EV-A71 virus infection, the procedures are identical as EV-D68 virus, except that 30 mM MgCl₂ was omitted in all the media and viruses were infected and incubated at 37°C instead of 33 °C.

Western Blot, RNA Extraction and RT-qPCR.

The detailed procedures for these assays were described previously.^{6, 9} Briefly, the RD cells infected with EV-D68 strain US/ MO/14–18947 at a MOI of 1 were treated with indicated compound for 9 h. Then, total proteins were extracted from the cells using RAPI lysis buffer. Equal amounts of total proteins were separated by SDS-PAGE, then transferred to PVDF membrane. Target protein was detected using protein-specific antibody and corresponding horseradish peroxidase (HRP)-conjugated secondary antibody. Total RNA was extracted using TRIzol reagents. After digestion of genomic DNA with RQ1 RNase-free DNase, 1.5 μg of total RNA was used to synthesize the first strand of cDNA of viral RNA and host mRNA using SuperScript III reverse transcriptase. Target gene was amplified on a QuantStudio 5 Real-Time PCR System using FastStart Universal SYBR Green Master and virus-specific primers.

Immunostaining.

RD cells were infected with EV-D68 strain US/MO/14–18947 with a MOI of 1. At 9 hpi, cells were fixed with 4% formaldehyde for 10 min, followed by permeabilization with 0.2% Triton X-100 for another 10 min. After blocking with 10% bovine serum, cells were stained with rabbit anti-VP1 antibody and followed by staining with anti-rabbit immunoglobulin secondary antibody conjugated to Alexa Fluor 488. The nuclei were stained with 300 nM DAPI after secondary antibody incubation. Fluorescent images were acquired using a ZOE fluorescent Cell Imager (Bio-Rad).

Time-of-Addition Assay.

The detailed procedure was described previously.⁹ Briefly, approximately 90% confluent RD cells were treated with compound R523062, pleconaril or dibucaine at 1 h before infection (−1 hpi), at the time of infection (0 hpi), or at 1−7 hpi with EV-D68 US/MO/14–18947 with a MOI of 0.01. The supernatant was harvested at 14 hpi, and the progeny viruses in the cell culture supernatants were quantified via plaque assay. For quantification of the progeny virus inside the RD cells via immunostaining, a MOI of 1 was used to infect the RD cells, and the cells were fixed at 9 hpi, and immunostaining was carried as described in immunostaining section.

EV-D68 2A protease expression and proteolytic reaction.

The expression of EV-D68 2A protease was described previously.⁶ Proteolytic reaction was carried out as follows: 1 μM 2A protein in reaction buffer (50 mM Tris pH7.0, 150 mM NaCl, and 10% glycerol and 2 mM DTT) was incubated with either DMSO, 1 μM telaprevir, or 100 μM compound R523062 at 30 °C for 1 hr, then the proteolytic reaction was triggered by adding 20 uM FRET-1 substrate.⁶ The reaction progress was monitored for 2 hr at 30 °C in a Cytation 5 imaging reader (Thermo Fisher Scientific) with filters for excitation at 360/40 nm and emission at 460/40 nm.

EV-D68 2C (40–330) WT and mutant proteins expression and differential scanning fluorimetry (DSF).

The DNA fragment coding for EV-D68 US/MO/14–18947 2C (amino acid residues 40 to 330) was ordered from GenScript (Piscataway, NJ) and inserted into the pET28a (+)-TEV vector with E. coli codon optimization. The three mutants I227L, Q322R and I227L/Q322R of EV-D68 2C (40–330) were obtained by using QuikChange XL Site-Directed Mutagenesis Kit (Agilent, Santa Clara). The recombinant plasmids were transformed into E. coli Rosetta 2(DE3) competent cells and bacterial cultures were grown in LB containing 50 μg/mL of kanamycin and 34 μg/mL of chloramphenicol at 37 °C, and expression of the target proteins were induced at an optical density (A600) of 0.6–0.8 by the addition of isopropyl β-D-1-thiogalactopyranoside (IPTG) to 0.5 mM. Growth of bacterial cultures were continued at 18 °C for 12–16 h post-induction, and the bacteria were harvested by centrifugation (6000 g, 10 min, 4 °C) and re-suspended in lysis buffer (20 mM Hepes [pH 7.5], 300 mM NaCl, 4 mg/ml lysozyme, 1 mM phenylmethylsulfonyl fluoride [PMSF], 0.01 mg/ml DNase I), and lysed by sonication for 30 mins. The cell debris were removed by centrifugation at 17,000 g for 1 hr. The supernatant was incubated with nickel-nitrilotriacetic acid (Ni-NTA) resin for over 2 h at 4 °C and all proteins were purified to greater than 98% homogeneity followed by dialysis (two times) against a buffer containing 20 mM Hepes (pH 7.5), 300 mM NaCl and 1 mM DTT. Purified proteins were fast frozen in liquid nitrogen and stored at −80 °C freezer. The binding of R523062 on EV-D68 2C proteins was monitored by differential scanning fluorimetry (DSF) using a Thermal Fisher QuantStudio™ 5 Real-Time PCR System. TSA plates were prepared by mixing EV-D68 2C proteins (final concentration of 4 μM) with compounds and incubated at 30 °C for 1 hr. 1× SYPRO orange (Thermal Fisher) were added and the fluorescence of the plates were taken under a temperature gradient ranging from 20 to 95 °C (incremental steps of 0.05 °C/s). The melting temperature (T_m) was calculated as the mid-log of the transition phase from the native to the denatured protein using a Boltzmann model in Protein Thermal Shift Software v1.3. ΔT_m was calculated by subtracting reference melting temperature of proteins in the presence of DMSO from the T_m in the presence of compound. Raw DSF data were exported and curve fitting was performed using the Boltzmann sigmoidal equation in Prism (v5) software.

Serial Passage Experiment and Viral Genome Sequencing.

Serial passage experiment and viral genome sequencing were performed as previously described.^{6, 9} Briefly, serial passages were carried out in the presence of increasing concentration of compound R523062 or DMSO as shown in Table 2. The RD cells were infected with EV-D68 US/MO/14–18947 at a MOI of 0.1, and the virus was harvested after approximately 3 days when significant cytopathic effect was observed. Virus titer in the cell culture supernatant was quantified by a plaque assay. The passage P3 virus drug sensitivity EC₅₀ value was determined via CPE assay. The genome sequences were determined by first purifying viral RNA using a QIAamp viral RNA mini kit, followed by reverse transcription using a SuperScript III first strand reverse transcriptase (Invitrogen) with an oligo(dT) primer and PCR amplification. The whole viral genome was sequenced via 14 sequencing reactions by Eton Bioscience, Inc. The sequencing primers were reported before.⁹

Reverse Genetics of EV-D68 Virus.

A plasmid-based reverse genetic system for EV-D68 US/MO/14–18947 was generated in a pHH21 vector as described Pan et al. with modifications.^{9, 29} The mutations were introduced via site-directed mutagenesis with Agilent Technologies QuikChange II XL kit according to manufacturer protocol and confirmed by sequencing.

Virus Growth Competition Assay.

This assay was carried out similarly as previously described.⁹ To compare the relative viral fitness of rMO 2C-I227L/Q322R virus with rMO WT virus, rMO WT virus was mixed with rMO 2C-I227L/Q322R virus at a 1:100 ratio. Every 2 days after infection, culture media supernatant was harvested and viruses were quantified with plaque assay, and 28,000 PFU progeny viruses were used for the next round of infection. After 5 passages, 2C coding sequences of each passage viruses were determined by reverse transcription-PCR, followed by sequencing reactions as described in the serial passage experiment and virus genome sequencing section. The percentage of each virus was estimated by measuring the height of the nucleotide sequence electropherogram peak.^{9, 24}

Combination therapy experiments of R523062 with either telaprevir or compound R856932.

Compound R523062 was combined with teleprevir or compound R856932 at combination ratio 8:1, 4:1, 2:1, 1:1, 1:2, 1:4, and 1:8. A combination indices (CIs) plot depicts the EC₅₀ values of each compound at different combination ratios. The red line indicates the additive effect, and above the line indicates the antagonism, and below the line indicates the synergy.³⁰