Antiviral Peptides Targeting the Helicase Activity of Enterovirus Nonstructural Protein 2C

ABSTRACT

Enteroviruses belong to the genus Enterovirus of the family Picornaviridae and include four human enterovirus groups (EV-A to -D): the epidemic of enteroviruses such as human enterovirus A71 (EV-A71) and coxsackievirus A16 (CVA16) is a threat to global public health. Enteroviral protein 2C is the most conserved nonstructural protein among all enteroviruses and possesses RNA helicase activity that plays pivotal roles during enteroviral life cycles, which makes 2C an attractive target for developing antienterovirus drugs. In this study, we designed a peptide, named 2CL, based on the structure of EV-A71 2C. This peptide effectively impaired the oligomerization of EV-A71 2C protein and inhibited the RNA helicase activities of 2C proteins encoded by EV-A71 and CVA16, both of which belong to EV-A, and showed potent antiviral efficacy against EV-A71 and CVA16 in cells. Moreover, the 2CL treatment elicited a strong in vivo protective efficacy against lethal EV-A71 challenge. In addition, the antiviral strategy of targeting the 2C helicase activity can be applied to inhibit the replication of EV-B. Either 2CL or B-2CL, the peptide redesigned based on the 2CL-corresponding sequence of EV-Bs, could exert effective antiviral activity against two important EV-Bs, coxsackievirus B3 and echovirus 11. Together, our findings demonstrated that targeting the helicase activity of 2C with a rationally designed peptide is an efficient antiviral strategy against enteroviruses, and 2CL and B-2CL show promising clinical potential to be further developed as broad-spectrum antienterovirus drugs.

IMPORTANCE Enteroviruses are a large group of positive-sense single-stranded RNA viruses and include numerous human pathogens, such as enterovirus A71 (EV-A71), coxsackieviruses, and echoviruses. However, no approved EV antiviral drugs are available. Enteroviral 2C is the most conserved nonstructural protein among all enteroviruses and contains the RNA helicase activity critical for the viral life cycle. Herein, according to the structure of EV-A71 2C, we designed a peptide that effectively inhibited the RNA helicase activities of EV-A71- and coxsackievirus A16 (CVA16)-encoded 2C proteins. Moreover, this peptide exerted potent antiviral effects against EV-A71 and CVA16 in cells and elicited therapeutic efficacy against lethal EV-A71 challenge in vivo. Furthermore, we demonstrate that the strategy of targeting the 2C helicase activity can be used for other relevant enteroviruses, including coxsackievirus B3 and echovirus 11. In summary, our findings provide compelling evidence that the designed peptides targeting the helicase activity of 2C could be broad-spectrum antivirals for enteroviruses.

INTRODUCTION

Enteroviruses (EVs) are a large group of positive-sense single-stranded RNA viruses belonging to the genus Enterovirus of the family Picornaviridae, and they include numerous important human pathogens, such as poliovirus, enterovirus A71 (EV-A71), coxsackieviruses, and echoviruses. Enteroviruses are serious threats to human health and cause about 3 billion human infections per year across the globe. These viruses are the causative pathogens for diverse human diseases ranging from moderate illness, such as the common cold, upper respiratory illness, and hand-foot-and-mouth disease (HFMD), to severe or even lethal ones, including aseptic meningitis, encephalitis, myocarditis, neonatal sepsis-like disease, and poliomyelitis (1–7). Thus, far, there is no effective antiviral therapy available for any enterovirus.

Helicases are a class of nucleic acid remodeling proteins that possess ATPase activity and utilize the energy of ATP binding and/or hydrolysis to unwind DNA or RNA duplexes, and they are classified into six superfamilies (SFs), designated SF1 to SF6, on the basis of conserved helicase consensuses (8–11). In particular, RNA helicases are believed to be involved in most ATP-dependent rearrangements of structured RNAs. For RNA viruses, including enteroviruses, their viral RNAs contain numerous highly structured cis-acting elements whose proper folding is pivotal for the functionalities of viral RNAs (12). Besides, during viral RNA replication, the replicative intermediate double-stranded RNA (dsRNA) must be efficiently unwound to allow the recycling of viral RNA templates for multiround production of progeny viral RNAs (12). All these processes require the participation of helicases, while a wide range of RNA viruses have been found to encode their own RNA helicases, such as coronavirus nsp13, flavivirus NS3, norovirus NS3, alphavirus nsP2, alphatetravirus Hel, etc. In addition, some viral proteins, such as Ebola virus VP35 and cypovirus VP5, contain helicase-like or RNA-chaperoning activity that can unwind or destabilize RNA structures dependently or independently of ATP (12–19). Given the pivotal roles of helicases in viral life cycles, helicases are widely considered promising targets for the development of antiviral drugs (20).

Enterovirus encodes a polyprotein that is proteolytically cleaved into four structural proteins and seven nonstructural proteins. Moreover, an alternative encoding strategy of harboring a novel open reading frame encoding a short peptide has been recently reported in gut epithelial cells infected with some enteroviruses (21). Among enteroviral nonstructural proteins, protein 2C is the most conserved and has been found to possess ATPase and helicase activities that are indispensable for viral RNA replication (22–25). Besides, EV-A71 2C has also been found to contain ATP-independent RNA chaperoning activity that can destabilize RNA duplexes (24). Based on the conserved consensus motifs, enteroviral 2C is classified as part of the helicase SF3 that functions in a hexameric form (22, 26). Previous studies also revealed that enteroviral 2C is present in homo-oligomeric form that is required for its proper activity (24). In addition, a recent structural study uncovered that the C terminus of EV-A71 2C forms an amphipathic helix that is inserted into a deep pocket formed in another 2C protein, and this interaction mediates and stabilizes the homo-oligomerization of 2C (27). Given the necessity of homo-oligomerization for SF3 helicases like enteroviral 2C, as well as the high conservation of 2C among all enteroviruses, we propose that targeting the deep pocket of enteroviral 2C would abrogate its helicase activity, which could be an ideal strategy to develop broad-spectrum antienterovirus therapy.

In this study, we designed a peptide based on the amino acid sequence of the EV-A71 2C C-terminal domain, designated 2C-ligand (2CL), which may compete with the C-terminal amphipathic helix of 2C for the deep pocket. Our results showed that this peptide effectively impaired the oligomerization of EV-A71 2C, also inhibited the helicase activity of EV-A71- or coxsackievirus A16 (CVA16)-encoded 2C, and showed a potent in vitro antiviral effect on EV-A71 and CVA16, both of which belong to the group A enterovirus (EV-A) and are the two major causative agents of HFMD, which has over 3 million confirmed cases—mainly in children in the Asian-Pacific region. Importantly, 2CL treatment elicited a strong in vivo antiviral activity that protected mice against lethal EV-A71 challenge, indicating the clinical potential of this peptide. Furthermore, we applied this novel strategy of antiviral development (i.e., designing a 2C-targeting peptide) onto two other viruses in group B enterovirus (EV-B), coxsackievirus B3 (CVB3) and echovirus 11 (Echo 11), both of which are important human pathogens. CVB3 is often associated with severe myocarditis and endocarditis in adults, while Echo 11 has been reported to cause clinical syndromes in neonates, including aseptic meningitis, neonatal hepatitis, and myocarditis, with high mortality (6, 7). Our data showed that this 2C-targeting peptide, designated B-2CL, also showed strong antiviral activity against both CVB3 and Echo 11 in infected cells. Our findings provide the first demonstration that targeting the helicase activity of 2C by abrogating its homo-oligomerization is a promising broad-spectrum strategy to develop antienterovirus drugs.

RESULTS

Design of a 2C-targeting peptide.

Based on the structure of EV-A71 2C (27), this protein contains an N-terminal membrane-binding domain, an ATPase domain, a zinc finger domain, and a C-terminal domain formed by a long protruding helix (α6), as illustrated in Fig. 1A and B. Among the 6 α-helices, the helix α6 is protruded and can dock into a deep pocket formed by the ATPase and zinc finger domains of another 2C protein in a “lock and key” shape (Fig. 1C and D). In addition, this α6-pocket interaction mediates the homo-oligomerization of EV-A71 2C, which is indispensable for 2C helicase activity. Therefore, targeting the “lock and key” α6-pocket interaction would be a logical strategy to abrogate the homo-oligomerization and subsequently inhibit the helicase of 2C as well as viral RNA replication. For this purpose, we designed a peptide that is derived from the amino acid sequence of the 2C α6 helix and includes the pocket-binding domain and RNA binding domain (Fig. 1E). This peptide, designated 2CL, should have the potential to compete with the α6 helix of a 2C protein for the deep pocket of another 2C, thereby abrogating 2C homo-oligomerization.

FIG 1.

Design of an EV-A71 2C-targeting peptide. (A) Overview of the key domain of the EV-A71 2C monomer. The C-terminal domain is colored in red. (B) Ribbon model of EV-A71 2C from residues 116 to 329. This model is built based on the published structure of EV-A71 2C (PDB no. 5grb). The α-helix is shown in red, and the β-sheet is shown in blue. The C-terminal helix (α6) is noted in the flank. (C) The intermolecular interaction between two 2C monomers. The C-terminal helix (α6) of one 2C monomer is in red, and another 2C monomer is shown by molecular surface in yellow. The interaction area of two monomers is framed in green. (D) Magnified view of the interaction area of two 2C monomers. The C-terminal portion of one 2C monomer (α6, in red) docks into the pocket formed by ATPase domain and zinc finger domain of another 2C monomer (molecular surface, in yellow). Residues 320 to 328 are shown with a stick model and labeled in brown. The pocket area is shown with a blue arrow. (E, top) The amino acid sequences of the C-terminal domain. α6 is shown in the pattern of the α-helix in red, and β7 is shown as a thick arrow in blue. The RNA binding domain and the pocket-binding domain in α6 are indicated. (Bottom) 2CL is designed based on the α6 of C-terminal domain, including the RNA binding and pocket-binding domains. Moreover, in order to permeate into cell membrane, we added TAT_47–57 as a cell-penetrating peptide (CPP) ahead of the key sequence and a linker (GSG) between CPP and the key sequence.

Peptide 2CL effectively inhibited the helicase activity of 2C from EV-A71 or CVA16.

We sought to examine whether peptide 2CL can affect the helicase activity of EV-A71 2C. Because the peptide 2CL corresponding sequence within CVA16 2C is completely identical to that of EV-A71 2C, we also examined the effect of 2CL treatment on the helicase activity of CVA16 2C. To this end, recombinant maltose-binding protein (MBP)–EV-A71 and -CVA16 2C fusion proteins (MBP-2C_EV-A71 and MBP-2C_CVA16, respectively) were eukaryotically expressed via a baculovirus system (Fig. 2A) and then reacted with the dsRNA helix substrate, which was generated by annealing a shorter 5′-hexachloro-fluorescein (HEX)-labeled RNA and a longer nonlabeled RNA (24) in the absence or presence of peptide 2CL at the indicated concentrations. The dsRNA unwinding was visualized by the separation of the substrate strands via gel electrophoresis as previously described (24). Our results showed that 2CL treatment effectively inhibited the unwinding of dsRNA substrate mediated by 2C from either EV-A71 or CVA16 in a dose-dependent manner (Fig. 2B and C), showing that 2CL does have the capacity to suppress the helicase activity of 2C protein in vitro.

FIG 2.

Inhibition of the RNA helicase activity of 2C for EV-A71 or CVA16 by 2CL in vitro. (A) MBP was expressed in a prokaryotic system (Escherichia coli), and MBP–EV-A71 2C and -CVA16 2C fusion proteins (MBP-2C_EV-A71 and MBP-2C_CVA16, respectively) were expressed in a eukaryotic (baculovirus) system. After purification, proteins were analyzed by 10% SDS-PAGE followed by Coomassie brilliant blue staining. (B and C) Increasing concentrations of 2CL (0.025, 0.05, and 0.1 mg/ml) were added to the reaction mixture containing RNA/RNA hybrid helix substrate (R*/R; 0.1 pmol), together with MBP-2C_EV-A71 (0.1 mg/ml; lane 4) (B) or MBP-2C_CVA16 (0.1 mg/ml; lane 4) (C). After incubation, the helix-unwinding activities of 2C proteins in the presence of 2CL were detected via gel electrophoresis followed by scanning on a Typhoon 9200 imager. The reaction mixture without any protein or with MBP alone (lanes 1 and 3) was used as a positive control, and the boiled reaction mixture (lane 2) was used as a negative control. The HEX-labeled strand is indicated by asterisks. (D) MBP-2C_EV-A71 was incubated with 20 μM 2CL or the same column of ddH₂O on ice for 1 h, and then subjected to size exclusion chromatography analysis. The green line indicates the elution profiles of 2C protein incubated with ddH₂O, and the purple line indicates that of 2C protein incubated with 2CL. The green and purple peaks on the right represent the molecular weight of 2C protein in the absence and in the presence of 2CL, respectively. The purple peak on the left represents the molecular weight of 2CL.

Moreover, we performed a size exclusion chromatography assay to examine the effects of 2CL on the 2C oligomerization. As shown in Fig. 2D, the chromatogram indicated the elution profiles of the 2C protein alone and in combination with 2CL. We found that the molecular weight of 2C oligomer (green line) was dramatically altered when incubated with 2CL (purple line), indicating that 2CL treatment indeed affected the oligomerization status of EV-A71 2C.

Peptide 2CL treatment elicited potent antienterovirus activity in cells.

Because the helicase activity of 2C is indispensable for enteroviral replication, we sought to examine whether 2CL has antiviral effects on EV-A71 or CVA16 in infected cells. To allow entry into cells, peptide 2CL was conjugated with the TAT_47–57 peptide, which is a commonly used cell-penetrating peptide (CPP) with good safety records in clinical studies (28). As shown in Fig. 3A, TAT-2CL (we will use the term “2CL” subsequently for simplicity) can efficiently enter rhabdomyosarcoma (RD) cells regardless of EV-A71 infection. Besides, the cytotoxicity of 2CL was examined via Cell Counting kit-8 (CCK-8) assays, and the 50% cytotoxic concentrations (CC₅₀) of 2CL in RD cells were greater than 150 μM (Fig. 3B).

FIG 3.

Cell penetration, cell viability, and anti-EV-A71 activity of 2CL in RD cells. (A) Cells were incubated with 2.5 μM FITC-labeled 2CL in the absence (top) or presence (bottom) of EV-A71 strain H (VR-1432) infection and then analyzed by fluorescence microscopy at 24 hpi. Scale bar, 100 μm. (B) Increasing concentrations of peptide 2CL were added to RD cells for 24 h, and cell viability was determined by CCK-8 assay. (C) Two-fold increasing concentrations of 2CL were added into EV-A71-infected (MOI of 0.1) RD cells, and the cytopathic effect (CPE) of RD cells was observed by microscopy at 24 hpi. (D) Dose-dependent inhibition of EV-A71 infection by 2CL was determined by plaque assay in RD cells. RD cells were seeded in 12-well plates followed by infection of EV-A71 (100 PFU), and then serially 2-fold-diluted 2CL was added. At 4 hpi, cells were subjected to plaque assays. (E) One-step growth curve of EV-A71 in the presence and absence of 2CL. Twenty-four-well plates were seeded with RD cells, followed by infection of EV-A71 (MOI of 0.1), and then 2.5 μM 2CL was added. At 6, 12, and 24 hpi, the supernatants were collected and subjected to plaque assays, respectively. (F) RD cells were infected with EV-A71 (MOI of 0.1), followed by treatment with 2-fold-increasing concentrations of 2CL. At 24 hpi, the cell lysates were harvested and subjected to Western blotting with anti-VP1 antibody. (G and H) Effects of 2CL on EV-A71 RNA accumulation in RD cells. Twenty-four-well plates were seeded with RD cells, followed by infection of EV-A71 (MOI of 0.1), and then serially 2-fold-diluted 2CL (G) or TAT_47–57 (H) was added. At 24 hpi, total cellular RNAs were extracted and subjected to qRT-PCR to examine the viral RNA accumulation of EV-A71. Data are means ± standard errors of the means (SEM). All experiments were independently repeated at least twice in triplicate with reproducible results.

We then assessed the antiviral effect of 2CL by examining the cytopathic effect (CPE) and performing plaque and one-step growth curve assays with EV-A71-infected RD cells, respectively. Our results showed that 2CL treatment had potent antiviral activity against EV-A71 in RD cells in a dose-dependent manner (Fig. 3C to E). This anti-EV-A71 activity of 2CL was also confirmed by determining its effect on the level of EV-A71 coat protein VP1 via immunoblotting (Fig. 3F) or measuring viral RNA accumulation via quantitative reverse transcription-PCR (qRT-PCR), which shows the 50% inhibitory concentration (IC₅₀) of 2CL against EV-A71 in RD cells as 1.35 ± 0.10 μM (Fig. 3G). On the other hand, the negative-control peptide TAT_47–57 showed no antiviral effect on EV-A71 infection in the same concentration range (Fig. 3H). In addition, the low IC₅₀ (∼1.35 μM) and high CC₅₀ (>150 μM) yielded a selectivity index (SI = CC₅₀/IC₅₀) above 111 (Table 1).

TABLE 1.

IC₅₀, CC₅₀, and SI of 2CL in different cell lines

Cells	IC50 (μM)	CC50 (μM)	SI (CC50/IC50)
RD	1.35	>150	>111
Vero	0.66	>150	>227
Huh7	0.41	>300	>731
293T	3.00	>300	>100

We further examined the inhibitory effects of 2CL on EV-A71 in different cell lines, including green monkey kidney Vero cells, human liver Huh7 cells, and human embryonic kidney 293T cells, via measuring viral RNA accumulation. As shown in Fig. 4A to F, 2CL treatment exhibited potent anti-EV-A71 activities in all these cell lines, with IC₅₀s of 0.66 ± 0.06 μM in Vero cells, 0.41 ± 0.08 μM in Huh7 cells, and 3.00 ± 0.57 μM in 293T cells. Moreover, the CC₅₀s of 2CL were >150 μM in Vero cells and >300 μM in Huh7 cells and 293T cells (Fig. 4G to I and Table 1). Therefore, 2CL has SIs ranging from >100 to >730, indicating the good safety and promising clinical potential of 2CL.

FIG 4.

Antiviral activity against EV-A71 of 2CL in different cell lines. (A to F) Effects of increasing concentrations of 2CL (A to C) or TAT_47–57 (D to F) on EV-A71 RNA accumulation in Vero, Huh7, and 293T cells. The indicated cells were seeded in 24-well plates followed by infection of EV-A71 (MOI of 0.1), and then serially 2-fold-diluted 2CL (G) or TAT_47–57 (H) was added. At 24 hpi, total cellular RNAs were extracted and subjected to qRT-PCR for examining the viral RNA accumulation of EV-A71. (G to I) Cell viability of 2CL in Vero, Huh7, and 293T cells. Increasing concentrations of peptide 2CL were added to Vero (G), Huh7 (H), and 293T (I) cells for 24 h, and cell viability was determined by CCK-8 assay. Data are means ± SEM. All experiments were independently repeated at least twice in triplicate with reproducible results.

As EV-A71 strain H (VR-1432), which was used by us in the present study, was a common strain for laboratory research, we also examined the antiviral effect of 2CL on a clinical strain of EV-A71 (strain XY833). Expectedly, 2CL treatment effectively inhibited the infection of EV-A71 strain XY833 at an IC₅₀ of 1.50 ± 0.20 μM in RD cells (Fig. 5A). Moreover, since we have found that 2CL can inhibit the helicase activity of CVA16 2C, we also examined its antiviral effect on CVA16 infection by using a clinically isolated strain of CVA16 (GD09/24). Our results show that 2CL could effectively inhibit CVA16 infection at an IC₅₀ of 2.16 ± 0.17 μM in infected RD cells (Fig. 5B).

FIG 5.

Antiviral activity of 2CL to EV-A71 and CVA16 clinical strain. (A) Dose-dependent inhibition of clinical isolated EV-A71 strain (XY833) (MOI of 0.1) by 2CL was determined by CCK-8 assay. (B) Twenty-four-well plates were seeded with RD cells, followed by infection of the clinically isolated CVA16 strain (GD09/24) (MOI of 0.1), and then serially 2-fold-diluted 2CL was added. At 24 hpi, total cellular RNAs were extracted and subjected to qRT-PCR to examine the viral RNA accumulation of CVA16. Data are means ± SEM. All experiments were independently repeated at least twice in triplicate with reproducible results.

Viral infections trigger cellular intrinsic immune defenses, including type I interferon (IFN-I) and cytokine pathways. To rule out the possibility that the antiviral effect of 2CL in infected cells is attributed to induction of the IFN-I and/or cytokine pathway, we assessed the effect of 2CL treatment on the levels of expression of the downstream products of these two pathways, including IFN-α, IFN-β, IFN-stimulated gene 56 (ISG56), interleukin-6 (IL-6), and CXCL10, in cells in the presence or absence of EV-A71 infection (Fig. 6A and B). Our data showed that 2CL treatment had little effect on the induction of IFN-I or cytokine pathway downstream products.

FIG 6.

The 2CL treatment does not affect the IFN-I and cytokine pathways. (A and B) RD cells were infected with (A) or without (B) EV-A71 (MOI of 0.1), and then 2.5 μM 2CL was added. At 24 hpi, total RNAs were extracted, and the mRNA levels of the factors in the IFN-I and cytokine pathways as indicated were measured via qRT-PCR. Data are means ± SEM. All experiments were independently repeated at least twice in triplicate with reproducible results.

Taken together, our findings show that targeting the helicase activity of 2C can be a promising antienterovirus strategy, and peptide 2CL shows considerable clinical potential to be further developed as an antiviral against EV-A71 and CVA16, the two major pathogens of HFMD.

In vivo protective efficacy of 2CL against lethal EV-A71 challenge.

As HFMD and HFMD-associated severe neurological diseases mainly occur in children under 5 years old, the newborn mouse is normally used as an in vivo murine model for EV-A71 infection (29, 30). To assess the in vivo antiviral efficacy of 2CL, we treated 1-day-old ICR mice with peptide 2CL at a dose of 20 mg/kg or with vehicle via intraperitoneal (i.p.) injection after a 10⁷-PFU EV-A71 challenge, followed by 2CL treatment twice a day for 7 days. As shown in Fig. 7A, the vehicle-treated group showed 50% mortality at 10 days postinfection (dpi). In contrast, all the mice treated with 2CL survived the viral challenge (Fig. 7A). Moreover, the clinical symptoms of the vehicle-treated group after EV-A71 challenge increased rapidly 4 dpi, whereas the 2CL-treated infected group had no significant difference from the noninfected group (Fig. 7B). Together, our results indicate that 2CL has potent therapeutic efficacy against EV-A71 infection in vivo.

FIG 7.

Protective activity of 2CL against EV-A71 challenge in newborn ICR mice. (A) Survival of EV-A71-challenged newborn ICR mice. One-day-old ICR mice were infected with 10⁷ PFU EV-A71 (strain H, VR-1432) and injected with 2CL (n = 9) at 20 mg/kg or vehicle at the same volume (n = 10) as a control; all injections were i.p. Mock, newborn mice with no infection and treatment (n = 8). Survival of ICR mice was observed and recorded for 21 days. (B) Clinical scores were recorded for three different groups in panel A. The log rank (Mantel-Cox) test was used for comparison of the survival rates of the mice, and a two-sided unpaired t test with Welch’s correction was used to compare their clinical symptoms (scores). All experiments were repeated at least twice with reproducible results.

Targeting 2C can be a broad-spectrum strategy to develop antivirals against different enteroviruses.

After determining that the 2C-targeting peptide 2CL can effectively inhibit the two group A enteroviruses (EV-As) EV-A71 and CVA16, it is intriguing to examine whether this new antiviral strategy can be extended to other enteroviruses, particularly the group B enteroviruses (EV-Bs), which include numerous important human pathogens, such as CVB3 and Echo 11.

To this end, we compared the corresponding amino acid sequences of 2CL in EV-As with those in EV-Bs and found that their sequences are slightly different in a few residues (Fig. 8A). Using the same rationale, we designed a new peptide, designated peptide B-2CL (i.e., group B 2C-ligand), to mimic the features of the α6 helix of EV-B 2C (Fig. 8B). We then treated CVB3- or Echo 11-infected RD cells with peptide B-2CL. Our data showed that B-2CL treatment effectively inhibited the replication of CVB3 or Echo 11, with an IC₅₀ of 2.29 ± 0.51 or 0.38 ± 0.12 μM, respectively (Fig. 8C and D), while similar to 2CL, this peptide showed little cytotoxicity at a concentration much higher than its IC₅₀ (Fig. 8E). Together, our results show that B-2CL has potent antiviral activity against both CVB3 and Echo 11, the two important EV-Bs.

FIG 8.

Antiviral activity of peptides against CVB3 and Echo 11. (A) The amino acid sequence alignment of enteroviral C-terminal domain of 2C proteins was performed by Clustal Omega (http://www.ebi.ac.uk/Tools/msa/clustalo/). GenBank accession numbers are shown in parentheses for the following viruses: EV-A71 (AEF32490.1), CVA16 (AVX33601.1), CVA4 (QGW48255.1), CVA6 (ANT96618.1), CVA8 (AKR53677.1), CVA10 (ANT96626.1), CVB3 (U57056.1), and Echo 11 (MN817130.1). The regions of 2CL are indicated. (B) Amino acid sequence alignment of B-2CL and 2CL. The different sequences between 2CL and B-2CL is marked in blue. (C and D) Effects of increasing concentrations of B-2CL on CVB3 (C) or Echo 11 (D) RNA accumulation in RD cells. (E) Increasing concentrations of peptide B-2CL were added to RD cells for 24 h, and cell viability was determined by CCK-8 assay. (F and G) Effects of increasing concentrations of 2CL on CVB3 (F) or Echo 11 (G) RNA accumulation in RD cells. Data are means ± SEM. All experiments were independently performed in triplicate and at least repeated twice with reproducible results.

Because protein 2C is the most conserved nonstructural protein of enteroviruses and the corresponding sequences of 2CL and B-2CL are homologous, we also tested whether 2CL can also exert any antiviral activity against these EV-Bs. To this end, we treated CVB3- or Echo 11-infected RD cells with peptide 2CL. Interestingly, peptide 2CL showed similar antiviral efficacy against both viruses (Fig. 8F and G).

Together, our results clearly show that targeting the helicase activity of 2C can be a promising strategy to develop broad-spectrum antiviral therapy against diverse enteroviruses, while peptides 2CL and B-2CL have considerable clinical potential to be further developed.

DISCUSSION

Although there is a great demand for antivirals against enteroviruses, which cause a wide range of diseases and affect about 3 billion persons per year, there is no antienterovirus therapy available so far. Due to the indispensable role of helicases in the life cycles of diverse viruses, helicases have long been recognized as attractive targets for developing antivirals. Nonstructural protein 2C is a conserved SF3 helicase encoded by enteroviruses. Previous studies have reported a number of small-molecule compounds that targeted enteroviral 2C and showed in vitro antiviral activity; however, few compounds showed in vivo and broad-spectrum antienterovirus efficacy (25, 31–35).

In this study, we reported an effective antienterovirus strategy by targeting enteroviral nonstructural protein 2C. We showed that peptide 2CL can impair the oligomerization of EV-A71 2C and effectively inhibit the helicase activity of 2C, encoded by either EV-A71 or CVA16, and also exerts potent antiviral efficacy against EV-A71 and CV-A16 in cells. More intriguingly, peptide 2CL showed effective in vivo antiviral activity by protecting mice from lethal EV-A71 challenge. It is worth mentioning that EV-A71 and CVA16 are the two prevalent causative pathogens for HFMD, which is also caused by other members of EV-A, including CVA4, CVA6, CVA8, and CVA10, while the sequence analyses of 2C proteins encoded by different EV-As show that the corresponding sequences of 2CL are highly conserved (mostly identical) in these 2C proteins (Fig. 8A). Therefore, peptide 2CL could be a broad-spectrum antiviral for a variety of EV-A-caused diseases, such as HFMD and hemorrhagic conjunctivitis (caused by CVA24 or EV-A70).

Interestingly, our study showed that either 2CL or B-2CL, the peptide designed based on the 2CL-corresponding sequence of EV-Bs, exerted effective antiviral activity against two important EV-Bs, CVB3 and Echo 11, which cause severe myocarditis and endocarditis in adults and aseptic meningitis, neonatal hepatitis, and myocarditis in neonates, respectively (6, 7). Since 2CL or B-2CL is designed to impair the “lock and key” structure of the homo-oligomerization interface of 2C molecules, our results imply that the oligomerization is indeed fundamental to 2C helicase activity and enteroviral RNA replication. Besides, given that 2C is a multifunctional protein, it is possible that disruption of 2C oligomerization may also affect its other functions, in addition to unwinding viral dsRNAs.

In conclusion, our study provides compelling evidence that targeting the helicase activity of 2C with a rationally designed peptide showed potent antiviral efficacies against a number of important enteroviruses, including EV-A71, CVA16, CVB3, and Echo 11. In addition, future studies should be performed to assess the antiviral efficacies of peptides 2CL and B-2CL against other enteroviruses and picornaviruses and further optimize these peptides to enhance their antiviral activity, stability, and bioavailability. Considering that 2C is the most conserved nonstructural protein of enteroviruses and rhinoviruses, these peptides show promising clinical potential to be further developed as broad-spectrum antiviral drugs against all enteroviruses, rhinoviruses, and even other picornaviruses.

MATERIALS AND METHODS

Cells and viruses.

The RD, Vero, Huh7, and 293T cell lines were obtained from ATCC commercially and cultured at 37°C with 5% CO₂ in a humidified atmosphere with a supply of Dulbecco’s modified Eagle’s medium (DMEM; Gibco) containing 10% fetal bovine serum (FBS; Gibco), 100 μg/ml streptomycin, and 100 U/ml penicillin (HyClone).

EV-A71 strain H (VR-1432) was obtained commercially from ATCC, and XY833, a clinical EV-A71 strain, was provided by Hubei Province Center for Disease Control and Prevention (Hubei, China). CVA16 (GD09/24) was generously provide by Cheng-Feng Qin (Beijing, China). CVB3 was generously provide by Zhaohua Zhong (Haerbin, China). Echovirus 11 was provided by the Guangzhou Women and Children Medical Center.

Peptides.

Peptides 2CL (YGRKKRRQRRRGSGREYNNRSAIGNTIEALFQ), TAT_47–57 with a linker (YGRKKRRQRRRGSG) and fluorescein isothiocyanate (FITC)-labeled 2CL and B-2CL (YGRKKRRQRRRGSGREYNHRHSVGATLEALFQ), as well as 2CL with no TAT_47–57 (REYNNRSAIGNTIEALFQ), were synthesized by GenScript Biochem with a purity of 95%. All of these peptides were acetylated at the N termini, amidated at the C termini, and dissolved in ultrapure water at a concentration of 2 mM as stock peptides for the in vitro experiment.

Mice.

Institute of Cancer Research (ICR) mice were purchased from the Beijing Vital River Laboratory Animal Technology Co. (Beijing, China). All experiments were performed according to protocols approved by the Institutional Animal Care and Use Committee of the Wuhan Institute of Virology.

Structural modeling analysis of EV-A71 2C.

The structure of EV-A71 2C was illustrated in the SWISS-PDB viewer based on the published data (PDB no. 5grb) (27), with the secondary structure view and surface view of 2C protein. The interactions of amino acids are shown on the surface of α6, and the pocket of another monomer is indicated by blue arrows in Fig. 1.

Cellular fluorescent imaging.

Thirty-five-millimeter dishes were seeded with RD cells for 24 h and treated with 2.5 μM FITC-labeled 2CL with or without the infection of EV-A71 an hour before. After 24 h, cells were fixed by 4% paraformaldehyde for 20 min at 25°C, and then the cells were observed with a fluorescence microscope.

Quantitative reverse transcription-PCR.

Quantitative reverse transcription-PCR (qRT-PCR) was performed after RNA extraction (36). Total RNAs were extracted using a Foregene Total RNA kit according to the manufacturer’s instructions. Then qRT-PCR was performed using the specific primers for EV-A71, the indicated IFN-I and cytokine pathway products in Fig. 6, and the universal primers of enterovirus for CVB3 and Echo 11. The primer list is shown in Table 2.

TABLE 2.

The qRT-PCR primers used in this study

Primera	Sequence (5′ to 3′)
EV-A71-F	GGCCATTTATGTGGGTAACTTTAGA
EV-A71-R	CGGGCAATCGTGTCACAAC
Human β-actin-F	AGAGCTACGAGCTGCCTGAC
Human β-actin-R	AGCACTGTGTTGGCGTACAG
CVA16 F	ATCCAGTAAGGATCCCAGACT
CVA16 R	GATTTGCATAGTGGAGAGCAG
Monkey β-actin-F	CACACAGGGGAGGTGATAGC
Monkey β-actin-R	GCACTTTTATTCAACTGGTCTCA
Human-IFNα-F	AGAATCTCTCCTTTCTCCTG
Human-IFNα-R	TCTGACAACCTCCCAGGCAC
Human-IFNβ-F	TTGTTGAGAACCTCCTGGCT
Human-IFNβ-R	TGACTATGGTCCAGGCACAG
Human-ISG56-F	TTGATGACGATGAAATGCCTGA
Human-ISG56-R	CAGGTCACCAGACTCCTCAC
Human-CXCL10-F	AGTGGCATTCAAGGAGTACC
Human-CXCL10-R	GCAATGATCTCAACACGTG

^aAll primers shown were used for qRT-PCR. Forward primers end in “F,” and reverse primers end in “R.”

Assays for antiviral activity.

We use different methods to measure the antiviral activity of 2CL, including plaque assay, qRT-PCR, CCK-8 reagent (Yeasen), and a one-step growth curve assay. The plaque assay is the direct method for detecting a specific number of viruses. Briefly, 12-well plates were seeded with RD cells, followed by infection with approximately 100 PFU EV-A71, and 1 h later, serially 2-fold-diluted 2CL was added to the infected cells with incubation time of 4 h. Then the supernatants were replaced by fresh DMEM containing 2% low-melting-point agarose, 6% FBS, 200 μg/ml streptomycin, and 200 U/ml penicillin (HyClone). The plaques were incubated for 2 days postinfection (dpi) until they were evident. The cells were stained and fixed with phosphate-buffered saline (PBS) containing 1% crystal violet and 4% formaldehyde at 4°C for 2 h. After the agarose was washed away, the plaque reduction was calculated.

qRT-PCR was performed after infection and treatment. Briefly, 24-well plates were seeded with cells followed by infection with EV-A71 at a multiplicity of infection (MOI) of 0.1. One hour later, the diluted peptides were added onto infected cells. At 24 h postinfection (hpi), cells were harvest and RNAs were extracted to perform qRT-PCR with the specific primers for EV-A71 and universal primers of enterovirus for CVB3 and Echo 11. The primer list is shown in Table 2.

The colorimetric viral infection assay using the CCK-8 reagent was performed as follows. Ninety-six-well plates were seeded with RD cells, followed by infection with EV-A71 at an MOI of 0.1 for 1 h. Then the infected cells were treated with the diluted peptides for about 24 hpi until an evident cytopathic effect (CPE) could be observed, at which point CCK-8 reagent was added into the well for 2 h. Finally, the optical density at 450 nm (OD₄₅₀) was measured using a microplate reader, and the data were analyzed.

For the one-step growth curve assay, 24-well plates were seeded with RD cells followed by infection with EV-A71 at an MOI of 0.1. One hour later, cells were washed, and 2.5 μM 2CL was added to the infected cells. At 6, 12, and 24 hpi, the supernatants were collected and subjected to plaque assay, respectively.

Cell viability test.

CCK-8 reagent was used to evaluate the cytotoxicity of 2CL in RD, Vero, Huh7, and 293T cells. Briefly, 96-well plates were seeded with cells and incubated with 2-fold-increasing concentrations of 2CL. After incubation for 24 h at 37°C, CCK-8 solution was added to the well for 2 h, followed by measurement of the value with a microplate reader.

Western blotting.

Cells were harvested using lysis buffer containing 150 mM NaCl, 50 mM Tris-HCl (pH 7.4), 0.25% deoxycholate, and a protease inhibitor cocktail and lysed for 30 min, followed by 10% SDS-PAGE. Then Western blotting was performed as previously described (37). Anti-VP1 EV-A71 (Abnova) antibody was obtained commercially and used in Western blotting.

Plasmids, protein expression, and purification.

pFastBac HTB-MBP and pFastBac HTB-MBP-2C were generated as previously described (24). In addition, the methods for expression and purification of MBP and MBP-2C were performed as described previously (12, 24, 38). Briefly, Sf9 cells were infected with the recombinant baculoviruses. Then the cells were harvested, resuspended, and lysed by sonication after 72 hpi, followed by centrifugation at 11,000 × g for 30 min to remove debris. The proteins in the supernatant were purified according to the manufacturer’s protocol for amylase affinity chromatography (NEB). Next, Amicon Ultra-15 filters (Millipore) were used to concentrate the protein. Proteins were then quantified and stored at −80°C. Finally, 10% SDS-PAGE were performed to separate proteins, and the gel was stained with Coomassie blue.

Nucleic acid helix unwinding assay.

The assay of the standard helix destabilizing was described previously (12, 24). Briefly, 0.1 pmol of HEX-labeled helix, the indicated concentration of 2CL (0.025, 0.05, and 0.1 mg/ml), and 0.1 mg/ml of 2C protein were added to a mixture of 2.5 mM MgCl₂, 50 mM HEPES-KOH (pH 7.5), 0.01% bovine serum albumin (BSA), 2 mM dithiothreitol (DTT), 20 U RNasin (Promega), and 5 mM ATP. The mixture was incubated for 60 min at 37°C, and the reaction was terminated using proteinase K (1 μg/μl) and 5× loading buffer containing 1% SDS, 100 mM Tris-HCl, bromophenol blue, and 50% glycerol, (pH 7.5). Then the mixtures were electrophoresed on 12% native PAGE gels and scanned using a Typhoon 9200 imager.

Exclusion chromatography assay.

The affinity-purified EV-A71 2C protein was first concentrated by tangential flow filtration to 1 mg/ml using Amicon Ultra centrifugal filters (Merck). Before being loaded onto the column, 2C was incubated with 20 μM 2CL or the same column of double-distilled water (ddH₂O) for 1 h on ice. For size exclusion chromatography, samples were preequilibrated with buffer containing 50 mM HEPES-KOH (pH 8.5) and then loaded onto the Superdex 200 increase 10/300 GL column (GE Healthcare). Programmed chromatography was performed with BioLogic DuoFlow system (Bio-Rad) at a flow rate of 1 ml/min. Automatic zero-baseline action was performed at the 5th minute to zero with the UV signal. Peak analysis was performed using the ASTRA software package (BioLogic Chromatography Systems).

Antiviral activity of 2CL in ICR mice.

All of the 27 1-day-old ICR mice were randomly distributed into three groups, the vehicle groups had 10 mice, and the therapeutic groups had 9 mice, respectively, and then the mice were challenged i.p. with 10⁷ PFU EV-A71. There were 8 mice in the mock-infected group, and they were injected with the same volume of PBS instead of EV-A71. At 1 h postinfection, the newborn mice were treated with 20 mg/kg 2CL or vehicle intraperitoneally for the first time. Then 2CL was injected twice a day for consecutive 7 days. In addition to mortality, clinical symptoms were monitored every day for 21 days. Clinical symptoms were evaluated by clinical score, as described previously (29, 30): 0 is heathy, 1 is slow with hunchbacked movement, 2 is weakness in one limb, 3 is paralysis in one limb, 4 is paralysis in two limbs, and 5 is death.