Introduction

Enterovirus 71 (EV71), a member of the Picornaviridae family [1], typically causes human hand, foot, and mouth disease (HFMD) with neurological and systemic complications in young children and infants. Infection with the Enterovirus 71 (EV71) can even lead to severe neurological disease, including fatal encephalitis, aseptic meningitis, or acute flaccid paralysis [2]. According to the surveillance report on HFMD in 2014, the incidence of HFMD in China is 203.16/100,000, while the mortality is 18.03/100,000 [3]. Existed EV71 vaccines are often unsuccessful due to the generation of viral mutants and less effective [4, 5]. Identifying novel antiviral drugs is therefore of critical importance.

Innate immunity serves as the host’s first line of defense against pathogen invasion. The retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs) pathway is initiated through the recognition of viral nucleic acid by host pattern recognition receptors. It activates IKK-related kinases that launch transcriptional factors, such as interferon regulatory factor 3 (IRF3), resulting in the expression of type I interferons (IFN-α/IFN-β) [6,7,8,9]. IFN-α and IFN-β subsequently activate downstream signaling pathways that induce a diverse set of IFN-stimulated genes and protect host cells against the invading virus [10]. On the other hand, EV71 infection leads RLRs signaling molecules cleavage [11], thus attenuate the type I IFNs production and response. This impaired immunity is also believed to associate with higher morbidity and mortality of the EV71 virus [12].

In addition to this core pathway, other pathways including the Hippo pathway, which is regulated by the nutrient and physical environment of cells, can serve as a potent regulator of IFN production and antiviral defense [13]. Activation of the Hippo pathway triggers phosphorylation and inactivation of the yes-associated protein (YAP) through degradation or cytoplasmic sequestration by 14-3-3 proteins, abrogating the inhibitory effect of YAP/TAZ on TANK-binding kinase 1 (TBK1). Subsequently, unsuppressed TBK1 activates IRF3-mediated expression of type I IFN genes [14]. Therefore, targeted manipulation of the balance between Hippo signaling and IFN responses could have therapeutic potential in the field of antiviral therapy.

Medicinal plants have been used widely throughout history for the treatment of infections as traditional healing remedies due to their broad therapeutic spectrum and minimal side effects. Natural products remain promising antiviral drug sources [15]. Anemoside B4 (B4) is a natural saponin constituent isolated from the roots of Pulsatilla chinensis (Bunge) Regel. P. chinensis extracts possessed a wide variety of biological activities, such as antitumor [16, 17], immuno-modification, and immunological adjuvant activities [18]. As reported B4 could inhibit the secretion of cytokines in Porcine circoviruses-induced endothelial cells [19] and lipopolysaccharide-induced rat intestinal microvascular endothelial cells [20]. Moreover, our previous research showed that B4 presented anti-inflammatory and immune-modulatory activities in vivo through inhibition of the NF-κB pathway mediated pro-inflammatory response [21].

However, the clear role of B4 on the innate immune response and its ability to inhibit EV71 propagation has not been investigated. In this study, we found that B4 exhibited significant anti-EV71 activities both in vitro and in vivo. It is the first report that B4 significantly inhibited EV71 propagation through upregulating 14-3-3 and activating the type I IFN responses.

Materials and methods

Viruses, drugs, cell lines, and mice

Enterovirus 71 strain SHAPHC695F/SH/CHN/10 (695F) was isolated from a 1.8-year-old patient in Shanghai Public Health Clinical Center in 2010 [22]. Human rhabdomyosarcoma cell line (RD) and African Green Monkey Kidney cell line (Vero) were purchased from ATCC (CCL-136, CCL-81, Manassas, VA, USA). B4 (10.0 g) was isolated in our lab from the roots of P. chinensis (No. 09-04-18-01), and the structure was identified by comparison of its spectroscopic data with those of the reported [23]. The purity of B4 was determined as 99.9% by analytical HPLC with PDA detection. Specific-pathogen-free ICR mice (Charles River Laboratories, Wilmington, MA, USA) were maintained in the animal facility of the Shanghai Public Health Clinical Center. Pups of the same experimental group were housed together in an environment of 50% humidity at 22 °C under a 12-h light/dark cycle. They were kept with their mothers to provide food.

Ethics statement

Specific-pathogen-free suckling ICR mice were carried out under the National Institute of Health Guide for the Care and Use of Laboratory Animals and approved by the Shanghai Public Health Clinical Center Laboratory Animal Welfare and Ethics Committee with the number of 2019-A006-01.

EV71 infection and drugs administration in mice

Two-day-old suckling mice (2.0–2.3 g) were injected with 106 pfu EV71 (strain 695F) and subsequently underwent intraperitoneal (i.p.) injection with 1–5 μL of B4 to achieve the dosage of 200 mg kg−1 of B4 (n = 10), or an equivalent volume of physiological saline was injected in the EV71 group (n = 10). The drug was injected every day for 16 days. Infected mice were monitored daily for signs of morbidity and mortality. The sickness of mice was evaluated using a graded score (0, healthy; 1, slow movement; 2, weakness in hind limbs; 3, paralysis in single limb; 4, paralysis in two limbs; and 5, death). To better illustrate the recovery of the surviving mice, deaths were calculated into the average score only once at the first observation date. All mouse muscle tissues and serum were obtained from the 5th-day post infection. The tissues were further assayed for quantitative reverse transcriptase-PCR (qRT-PCR), histology, and immunohistochemistry analysis.

Histology and immunohistochemistry analysis

Routine hematoxylin and eosin (H&E) staining of various mouse skeletal muscles was performed using 3.7% formaldehyde-fixed, paraffin-embedded, and sectioned (4 mm thick) tissues. The sections were deparaffinized, washed in phosphate-buffered saline (PBS, 0.01 M, pH 7.2) 5 min for three times, then heated at 100 °C in a microwave oven 2 min for six times, incubated in 3% H2O2 in deionized water for 10 min to block endogenous peroxides activity, and then washed 5 min with PBS for three times. The sections were incubated overnight at 4 ˚C with anti-EV71 VP-1 monoclonal antibody (MAB979, Millipore, Billerica, MA, USA) and IFN-β antibody (NBP1-77288, Novus Biologicals, Littleton, CO, USA). After washing 5 min with PBS for three times, the appropriate HRP-polymer anti-mouse/rabbit immunoglobulin G was added to the sections and incubated at 37 °C for 20 min. The sections were then washed 5 min with PBS for three times, and the color was developed with DAB (34065, Thermo Scientific, Waltham, MA, USA) for 3–5 min. The nuclei were lightly counterstained with hematoxylin. Rigorous controls (matched tissue sections negative for EV71) were included in parallel to ensure the signals obtained were specific to viral antigen. The concentration of IL-10 and IFN-γ in serum was detected by U-Plex Biomarker Group 1 (ms) Assays (K15069l-1, Meso Scale Diagnostics, Rockville, MD, USA) with Meso Sector S 600 (Meso Scale Diagnostics, Rockville, MD, USA).

Cell cultures and virus infection

RD and Vero cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM, Life Technologies, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, Carlsbad, CA, USA) and 1% penicillin/streptomycin (Life Technologies, Carlsbad, CA, USA). RD cells were used for propagation and plaque titration of EV71. The virus infection was performed at a multiplicity of infectivity of 1 in DMEM medium free of FBS. The EV71 virus stocks were collected from the supernatants of infected cells at 24 h post infection (hpi). Cell viability was measured using a cell counting kit (CCK-8; Dojindo Laboratories, Kumamoto, Japan).

Titration assay

The titers of the virus stocks were determined by a TCID50 assay, and the stocks were aliquoted and stored at −80 °C. As of the TCID50 assay, serially diluted viruses from 10−1 to 10−9 in DMEM were inoculated to RD cells in 96-well plates, and the cells were incubated for seven days at 37 °C. TCID50 were calculated by counting the wells with cytopathic effect in infected RD cells using the formula: log10 TCID50 = L − d × (s − 0.5), where L is the log of the lowest dilution, d is the difference between dilution steps, and s is the sum of the proportion of positive wells.

Quantitative RT-PCR

Total RNA from infected tissues or cells was isolated using the Trizol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instruction. Complementary DNA was generated from 2 μg of the RNA by reverse transcription with oligo-T (dT) primer. qRT-PCR was performed with primers for VP-1, IFN-β, and 14-3-3β (summarized in Table S1 in the Supplementary information) using a CFX96TM real-time system (Bio-Rad, Hercules, CA, USA). The mRNA levels of target genes were normalized to β-Actin or GAPDH levels and were compared by the 2−ΔΔCt method.

ELISA assay for IFN-β

Cell culture supernatants were harvested from uninfected and infected RD cells with or without B4 administration at indicated time points, followed by centrifugation at 1000 × g for 15 min at 4 °C. A human IFN-β ELISA kit (Multisciences, LiankeBio, Shanghai, China) was used following the manufacturer’s protocol.

Stable-isotope labeling by amino acids in cell culture (SILAC) based proteomic profiling

DMEM media were made for SILAC experiments by adding light (12C6) or heavy (13C6) lysine and arginine into the DMEM without L-lysine and L-arginine (Thermo Scientific, Waltham, MA, USA), supplemented with dialyzed FBS (Invitrogen, Carlsbad, CA, USA). RD cells were cultured in these two DMEM media for a week to enable complete labeling.

Cell protein samples were prepared and analyzed as previously described before [24]. In a typical procedure, equal amounts of cell protein samples from light- or heavy-labeled cells were mixed. Then, 100 μg of protein samples were digested by using trypsin, and then the peptide samples were desalted with a C18 solid-phase column, dried in a vacuum centrifugal concentrator, and dissolved in 0.1% formic acid for MS/MS analysis. LC–MS/MS analysis was conducted using Orbitrap mass spectrometer (Thermo Scientific, Waltham, MA, USA). Raw MS data were matched by MaxQuant software (version 1.6.1.1) against a human protein database released from UniProt in July, 2017. The SILAC analysis was performed with three biological replicates.

Immunoprecipitation and Immunoblotting analysis

Cells were lysed with 80 μL of RIPA (containing 1% PMSF) (Sigma-Aldrich, St Louis, MO, USA) on ice for 10 min. Cell lysates were centrifuged at 12,000 × g for 10 min, then the concentrations was quantified with the bicinchoninic acid assay BCA (Pierce, Rockford, IL, USA). An equal amount of proteins was used for immunoprecipitation and immunoblotting analysis. For Flag-tag protein immunoprecipitation, anti-Flag M2 affinity gel (Sigma-Aldrich, St Louis, MO, USA) was used and the performance followed the technical procedure. Proteins in the lysates were separated by 8%–10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE). The separated proteins were transferred to PVDF membranes (Millipore, Billerica, MA, USA) and probed with the anti-EV71 VP-1 (MAB979, Millipore, Billerica, MA, USA), anti-EV71 3Cpro (GTX630191, GnenTex, Irvine, CA, USA), anti-IRF3 (D614C, Cell Signaling, Danvers, MA, USA), anti-phospho-IRF3 (Ser396, Cell Signaling, Danvers, MA, USA), anti-IFN-β (EPR12687, Abcam, Waltham, MA, USA), anti-MxA (EPR19967, Abcam, Waltham, MA, USA), anti-FLAG (F3165, Millipore, Billerica, MA, USA), anti-YAP (14074, Cell Signaling, Danvers, MA, USA), anti-14-3-3β (sc-25276, Santa Cruz, Dallas, TX, USA), and anti-β-Actin (Sigma-Aldrich, St Louis, MO, USA) primary antibodies overnight at 4 °C. After incubation with an HRP-conjugated secondary antibody (Cell Signaling, Danvers, MA, USA), the specific proteins were visualized using a chemiluminescent HRP substrate (Millipore, Billerica, MA, USA).

Confocal immunofluorescence

The target cells were fixed with paraformaldehyde for 30 min, washed three times with PBS, permeabilized with 0.2% Triton X-100 buffer, and blocked with PBS containing 1% bovine serum albumin (BSA) for 1 h at room temperature. Then, the cells were incubated with the primary antibody overnight at 4 °C and washed three times with PBS containing 0.01% Tween 20 and 1% BSA. Then, cells were incubated with their respective secondary antibodies for another 1 h. The cells were then washed and stained with DAPI (4083S, Cell Signaling, Danvers, MA, USA) to detect nuclei. Images were captured with confocal laser microscopy (LSM 710, Carl Zeiss, Germany).

Cell transfection

The siRNA targeting human IFN-β, 14-3-3β, or control siRNA with scrambled sequence were purchased from Ribobio (RiboBio, Guangzhou, China). To knockdown IFN-β or 14-3-3β expression, cells were transfected with the respective siRNAs using Lipofectamine 3000 (L3000-015, Invitrogen, Carlsbad, CA, USA). Twenty-four hours after transfection, the medium was changed using DMEM supplemented with B4 or DMEM for 1 h. Cells were harvested for future study after they had been infected with EV71 for 24 h.

Statistical analysis

Comparison of survival was performed with Mantel–Cox test. The 50% inhibitory concentration (IC50) were calculated by software (GraphPad Software, La Jolla, CA, USA). All data were expressed as means ± SEM. The two-tailed Student’s t-test was used to evaluate these data. Differences with P < 0.05 were considered statistically significant.

Results

B4 exhibited antiviral activity against EV71 in RD cells

To address whether B4 exhibit antiviral activity, we incubated RD cells with different concentrations (12.5–200 μM) of B4 prior to infection with EV71. B4 exhibit strong antiviral effect in a dose-dependent manner, evidenced as both the cell viability reached 100% (Fig. 1a) and the cell morphology remained unchanged (Fig. 1b) in EV71-infected cell at B4 dosage of 200 μM. The IC50 of B4 against EV71 was calculated as 24.95 ± 0.05 μM after linear regression in RD cells (Fig. 1c). Importantly, B4 itself did not show any toxicity to RD cells in the tested concentrations (Fig. 1d).

Fig. 1: B4 inhibited EV71 propagation without obvious cytotoxicity.
figure 1

a The anti-EV71 activity of B4 was evaluated by cell viability assays. RD cells were infected with EV71 after treatment with B4 at various concentrations. The viability of the control group (blank cells) was set as 100%; b The morphological change of cytopathy effects in RD cells infected with EV71 after treatment with B4 at different concentrations. Scale bar = 50 μm; c Dose-response influence of B4 on cytopathic effects induced by EV71, and IC50 value was calculated as described in the “Materials and methods”; d Evaluation of the cytotoxicity of B4. RD cells were treated with B4 at various concentrations as indicated. The effects of B4 on cell viability were evaluated by cell viability assay at 24 or 48 h. The viability of blank cells was set as 100%; e RD cells were treated with different concentrations of B4 before EV71 infection. Cell supernatants were harvested at the 12–48 h post infection and the viral titers were determined by TCID50 assays; f Total RNAs of RD cells were prepared from different groups and subjected to qRT-PCR for the VP-1 region of EV71 viral RNA. GAPDH was also analyzed as internal control. Values were expressed as a percentage of RD cells without infection; g RD cells were treated with different concentrations of B4 at 24 h before EV71 infection, and then total cell extracts were subjected to Western blot analysis with anti-VP-1 antibody and anti-3Cpro antibody. β-Actin was also analyzed as loading control. Values were expressed as a percentage of RD cells without treatment. Data are presented as mean ± SD (n = 3). **P < 0.01, ***P < 0.001 vs cells infected EV71 without treatment and ###P < 0.001 vs cells without infection.

Full size image

We also quantified the progeny virus titer in the supernatants after EV71 infection with and without B4 treatment. A clear inhibition of virus replication was observed as there is approximately 2-log lower of progeny virus in the supernatant harvested in B4-treated cells compared to untreated cells (Fig. 1e). Next, the antiviral effects were validated by measuring the intracellular viral genomic RNA copies and viral protein levels. B4, at concentration of 200 μM suppressed EV71-induced VP-1 mRNA transcripts in RD cells by ~50% at 24 hpi (Fig. 1f). Consistently, the levels of intracellular VP-1 and main EV71 protease 3Cpro were significantly inhibited by B4 in a dose-dependent manner (Fig. 1g). Taken together, all these data suggested that B4 exhibited strong antiviral effect in vitro by inhibiting viral replication.

Induction of IFN-β is required for B4 to exert its antiviral effects

The potential antiviral mechanisms of B4 against EV71 infection were investigated using a time-of-addition assay. B4 (200 μM) was supplemented to EV71-infected RD cells according to different protocols described (Fig. 2a). When comparing B4 treatment prior, in the meantime, or post EV71 infection, we observed pre-treatment (1 h incubation) of the cells with B4 gives the best antiviral activity (Fig. 2b). This led us to further investigate whether the antiviral effect of B4 was associated with activation of cellular defense pathway. Indeed, we observed B4 sensitizes the cellular defense response during EV71 infection as shown by significantly upregulated levels of IFN-β and IFN-stimulated genes (MxA) (Fig. 2c), whereas compared to cells that were infected by EV71 alone. The virus infection was confirmed by the expression of VP-1 (Fig. 2c). Moreover, the upregulated expression of IFN-β in cell supernatant of B4-treated cells was confirmed by ELISA (Fig. 2d). These results indicated that B4 not only enhanced the virus-triggered IFN-β induction but also IFN-β response.

Fig. 2: B4 potentiated virus-triggered RLRs pathway activation.
figure 2

a B4 treatment (200 μM) was performed before or after EV71 infection as indicated in the figure; b The viability of RD cells according to different protocols was detected using cell viability assays. The viability of the control group was set as 100%; c Total protein of RD cells with or without B4 treatment after infection in different times was subjected to Western blot for detection of IFN-β, MxA, VP-1 levels. β-Actin was also analyzed as loading control; d The level of IFN-β in cell supernatants harvested from RD cells with or without B4 administration at different infection time; e Total protein of RD cells with or without B4 administration after infection at different times were subjected to Western blot for detection of p-IRF3, IRF3 levels. β-Actin was also analyzed as loading control; f Representative confocal microscope image showing localization of p-IRF3 and VP-1. At 24 h after treatment with B4, RD cells were immune-stained with antibodies against p-IRF3 (red), VP-1 (green), and nucleus was stained with DAPI (blue), Scale bar = 20 μm; g Quantitation of p-IRF3 nuclear translocation. A total of 400 p-IRF3-positive cells from different fields were counted. Data are presented as mean ± SD (n = 3). *P < 0.05, **P < 0.01 and ***P < 0.001 vs cells infected EV71 without treatment and ###P < 0.001 vs cells without infection.

Full size image

Critical steps for signal transduction via IFN-β are TBK1-IKKε-induced phosphorylation of IRF3 at Ser396 and subsequent entry of IRF3 into the nucleus [25]. We therefore assessed the effect of B4 on those molecular events. The phosphorylation of IRF3 was significantly decreased after infection with EV71 24 h, however, EV71-induced phosphorylation of IRF3 was dramatically reversed in B4 stimulated cells (Fig. 2e). Moreover, B4 enhanced translocation of p-IRF3 to the nucleus (Fig. 2f, g), compared to cells only infected with EV71 which showed neglectful staining of nuclear p-IRF3, thus confirming the above results.

Further investigations were performed to determine whether IFN-β were required for the antiviral effect of B4. Three specific siRNAs for IFN-β (si-IFN-β No. 1#−3#) were designed, and si-IFN-β 3# effectively knocked down endogenous IFN-β expression (Fig. 3a). The results from the viability protection assay indicated that IFN-β knockdown suppressed the antiviral effects of B4, decreased viability of infected cells (Fig. 3b). Meanwhile, there were no differences observed in the expression of viral proteins VP-1 and 3Cpro (Fig. 3c) and viral titers (Fig. 3d) in the infected IFN-β knockdown cells, regardless of B4 treatment or not. In addition, we used Vero cells, which lack functional type I IFN genes [26], to confirm the vital role of IFN-β on the antiviral activity of B4. EV71 replication was not affected in Vero cells (Fig. 3e–g), excepting for the improvement of B4-treated cell viability (Fig. 3e, P < 0.05, compared with EV71-infected cells). These results indicated that B4 exerted its antiviral function through potentiating the RLRs pathway.

Fig. 3: The inhibitory effect of B4 on EV71 propagation required IFN-β.
figure 3

a RD cells were transfected with si-IFN-β or scrambled-siRNA for 48 h, and IFN-β mRNA levels were determined by qRT-PCR. GAPDH was also analyzed as internal control; RD cells were transfected with IFN-β-specific siRNA and control siRNA for 48 h and then infected with EV71 after B4 treatment. b Antiviral activity was determined by the cell viability assay. Cell viability of transfected scrambled-siRNA cells was set 100%. Data are presented as mean ± SD (n = 3). ***P < 0.001; c The levels of VP-1 and 3Cpro were evaluated by immunoblotting with VP-1 and 3Cpro antibodies. β-Actin was also analyzed as loading control. Data are presented as mean ± SD (n = 3). *P < 0.05, no sign., no significant difference; d Cell supernatants were harvested at the 24 h post infection and the viral titers were determined by TCID50 assays. Data are presented as mean ± SD (n = 3). *P < 0.05, no sign., no significant difference; e Vero cells were infected with EV71 after B4 treatment for 24 h. Antiviral activity was determined by the cell viability assay. Cell viability of blank cells was set as 100%; f EV71 production in the supernatants was estimated by TCID50 assays; g Total cell extracts were subjected to Western blot analysis with anti-VP-1 antibody. β-Actin was also analyzed as a loading control. Data are presented as mean ± SD (n = 3). ###P < 0.001 vs cells without infection and *P < 0.05 vs cells infected EV71 without treatment. no sign., no significant difference.

Full size image

Target profiling of B4 by SILAC-based proteomics analysis

We further investigated the molecular mechanism through which B4 augmented type I IFN production. To identify these regulated proteins, we performed SILAC experiments with or without B4 treatment in EV71-infected cells (Fig. 4a). Briefly, proteomes were lysed from the “light” and “heavy” infected cells incubated with PBS or B4, respectively, and light and heavy proteomes were mixed and finally subjected to trypsin digestion. The digested peptides were analyzed by liquid chromatography tandem mass spectrometry to identify proteins that are affected by B4 administration. After applying a cut-off of 1.4 for the averaged SILAC ratio in the experiment, we collectively identified 92 proteins regulated by B4 (Fig. 4b, P < 0.05 and Table 1).

Fig. 4: SILAC-based proteome profiling of the identified proteins in the EV71-infected cells upon B4 administration.
figure 4

a Schedule of SILAC-based proteome profiling; b The volcano plot for the MS identified proteins in RD cells. Each data point indicates the log2fold change (X-axis) with their corresponding −log10P value (Y-axis). The threshold for differential expression (cut-off = fold change >1.4 or <0.7 and P < 0.05) is indicated by dashed black lines. Solid green and red respectively depicted the significantly decreased and increased proteins after B4 supplementation; c Classification landscape of 92 B4-regulated proteins according to the biological process by KEGG analysis. The number of differentially expressed proteins in each category has been shown. Significantly enriched metabolic or signal transduction pathways in differentially expressed proteins were identified in the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://www.genome.jp/kegg); d Western blot validation of the expression levels of 14-3-3β in EV71-infected RD cells treated with/without B4. β-Actin was also analyzed as loading control; e qRT-PCR validation of the expression levels of 14-3-3β in EV71-infected RD cells treated with/without B4. GAPDH was also analyzed as internal control; f Western blot analysis of 14-3-3β in the nuclear and cytoplasm fractions derived from RD cells in different groups. Densitometry analysis 14-3-3β in nuclear and cytoplasm presented relative to the respective controls for the nuclear fraction (laminin B) and the cytoplasmic fraction (β-Actin). Data are presented as mean ± SD (n = 3). ##P < 0.01 vs cells without infection and *P < 0.05, **P < 0.01 vs cells infected EV71 without treatment.

Full size image
Table 1 List of proteins that were differentially expressed in the EV71-infected and B4-treated RD cells.
Full size table

Further, KEGG analysis (Kyoto Encyclopedia of Genes and Genomes of the pathways) associated with these 92 proteins revealed a top-ranking functional cluster of the “Hippo pathway”, which includes six proteins (Fig. 4c and Table 2). Among these proteins, 14-3-3 was reported as the molecular scaffolds which exerts an important regulatory function in the IFN response by modulating YAP cellular localization [13]. Immunoblotting analysis confirmed that a twofold increase of 14-3-3β protein level in B4-treated cells compared with that of the PBS-treated cells (Fig. 4d). B4 treatment could also increase 14-3-3β mRNA level (Fig. 4e). Since 14-3-3 protein family members have been implicated in innate immunity, where they function to sequestration of YAP in the cytoplasm. The enhanced 14-3-3β cytoplasm localization in B4-treated cells (Fig. 4f) suggested that 14-3-3β might affect the intracellular localization of YAP.

Table 2 KEGG analysis of differentially expressed proteins in the EV71-infected and B4-treated RD cells.
Full size table

B4 increases 14-3-3β to interact with YAP and affects its capacity to enhance the RLRs pathway activation

It is reported that YAP participates in negative regulation of type I IFN response, so B4 may enhance type I IFN response by promoting the cytoplasmic sequestration of YAP. Intriguingly, confocal immunofluorescence detected endogenous expression of YAP (immunostaining green) with colocalization of 14-3-3β (immunostaining red) to the cytoplasm after stimulation with B4 administration (Fig. 5a). Meanwhile, we detected the interaction of 14-3-3β, YAP, and IRF3 in FLAG-YAP transfected cells. We found that B4 promoted the association of YAP with 14-3-3β (Fig. 5b).

Fig. 5: B4 increased the interaction with 14-3-3β and YAP and 14-3-3β is critical for the antiviral activity of B4.
figure 5

a Representative confocal microscope image showing co-localization of YAP and 14-3-3β in EV71-infected cell with or without B4 administration. RD cells were immune-stained with antibodies against 14-3-3β (red) and YAP (green), Scale bar = 20 μm; b Immunoprecipitation with anti-FLAG. Western blot analysis of the interaction between YAP and 14-3-3β or IRF3. HEK293T cells were transfected with Flag-YAP plasmids for 24 h and then infected with EV71 after B4 treatment. IP was performed with anti-Flag M2 affinity gel, then protein interaction was detected by immunoblot; Input, the whole-cell lysates without immunoprecipitated; c RD cells were transfected with si-14-3-3β or scrambled-siRNA for 48 h, and 14-3-3β mRNA levels were determined by qRT-PCR. GAPDH was also analyzed as internal control; d RD cells were transfected with si-14-3-3β or scrambled-siRNA for 48 h, and 14-3-3β levels were evaluated by immunoblotting with 14-3-3β antibodies. β-Actin was also analyzed as loading control; RD cells were transfected with 14-3-3β specific siRNA and control siRNA for 24 h and then infected with EV71 after B4 treatment. e Antiviral activity was determined by the cell viability assay. Cell viability of transfected scrambled-siRNA cells was set 100%; f Immunoprecipitation with anti-FLAG. Western blot analysis of the interaction between YAP and 14-3-3β or IRF3; g The levels of IFN-β, MxA, and VP-1 were evaluated by immunoblotting with IFN-β, MxA, and VP-1 antibodies. β-Actin was also analyzed as loading control. Data are presented as mean ± SD (n = 3). *P < 0.05, **P < 0.01.

Full size image

We then tested whether the knockdown of 14-3-3 affects B4’s capacity to inhibit EV71 propagation and the RLRs pathway activation. The influence of 14-3-3 on B4-enhanced RLRs pathway activation and antiviral efficacy was analyzed by siRNA knockdown of 14-3-3β. Gene-silencing efficacy was examined by qRT-PCR and Western blot, and si-14-3-3β 1# effectively knocked down endogenous 14-3-3β expression (Fig. 5c, d). 14-3-3β knockdown dramatically inhibited the viability of the infected cells upon B4 stimulation (Fig. 5e). Moreover, the interaction of 14-3-3β and YAP in B4-treated cells was disturbed by 14-3-3β knockdown (Fig. 5f). Meanwhile, knockdown of 14-3-3β increased the association of YAP with IRF3 (Fig. 5f). Consistent with these results, B4-induced IFN-β expression and IFN-β response (MxA expression) were also decreased by 14-3-3β knockdown, and the downregulated antiviral activity was confirmed by the expression of VP-1 (Fig. 5g). These results suggested that 14-3-3 was required for full activation of RLRs signaling and antiviral efficacy of B4.

B4 exhibited excellent antiviral efficacy in EV71-infected suckling mice

As B4 showed superior antiviral efficacy in vitro, to determine the efficacy of B4 in vivo, we used an EV71-infected suckling mice model for investigation (Fig. 6a). In this model, without drug administration, EV71 infection-related symptoms appeared on day 5 post infection and this was characterized mostly by weakness in the hind limbs. In later days, the condition of EV71-infected mice further deteriorated into hind limb or front limb paralysis and reached a peak at day 6–10 post-infection (Fig. 6b). On the other hand, B4 administration significantly alleviated these symptoms, as observed clinical scores were markedly lower than that in the EV71 group (Fig. 6b, P < 0.01 Wilcoxon rank test). As for the mortality of challenged mice, the group of B4 injection at 200 mg ·kg−1 was almost completely protected (95.0% survival, Fig. 6c) whereas the EV71 group yielded a 55.6% survival at the end of the observation (P < 0.01, Mantel–Cox test). Histological observation revealed massive myofibril fracture and myocyte disruption, necrotizing myositis with inflammatory infiltrates in the limb muscles, and intercostal spaces even along the spine in the EV71 group (Fig. 6d, see black arrows). B4 administration significantly improved the integrity of limb muscle structure, although slight damages were still observed (Fig. 6d, see black asterisks). Furthermore, systemic inflammatory response syndrome caused by virus infection is a typical consequence that may result in sepsis. It was reported that EV71-infected mice exhibited a subtle and transient elevation in IL-10 and IFN-γ, which was associated with the pathogenesis of HMFD [27]. Interestingly, the anti-inflammatory cytokine IL-10 was significantly increased (Fig. 6e), with the decline of IFN-γ (Fig. 6f) in the B4-treated mice.

Fig. 6: B4 treatment improved survival and inhibited viral propagation in EV71-infected suckling mice model.
figure 6

a Schedule of EV71-infected suckling mice model; Clinical scores (b) and survival curve (c) of 2-day-old ICR mice i.p. inoculated with EV71 and treated with physiological saline (n = 10 mice), B4 (200 mg · kg−1, n = 10 mice). Deaths were calculated into an average score only once at the first observed date; d Skeletal muscle samples were collected on day 5 post infection and subjected to H&E staining as described in the material and method section. Scale bar = 20 μm; the levels of IL-10 (e) and IFN-γ (f) in mouse serum were detected by MSD assay; total RNAs of skeletal muscle were prepared from different groups and subjected to qRT-PCR for the VP-1 region of EV71 viral RNA (g) and IFN-β (h). β-Actin was also analyzed as internal control; skeletal muscle samples were collected on day 5 post infection and subjected to immunohistochemistry (i) as described in the Materials and methods section. Scale bar = 20 μm. Data are presented as mean ± SD (n = 10). #P < 0.05, ##P < 0.01, ###P < 0.001 vs Normal control group; *P < 0.05, **P < 0.01, ***P < 0.001 vs physiological saline infection group.

Full size image

At last, in order to confirm the activation of type I IFN response in B4-treated mice, the expression of IFN-β was detected in EV71-infected suckling mice model. The qRT-PCR results showed a significant increase in IFN-β mRNA (Fig. 6h) and dramatic decline in viral RNA when B4 was administered (Fig. 6g). Immunohistochemistry staining of VP-1 and IFN-β was employed to evaluate the extent of viral inhibition and type I IFN induction caused by B4. Skeletal muscle tissues exhibited positive staining, as intensive and widespread signal was detected, indicating IFN-β expression (Fig. 6i, see black asterisks) in the B4-treated group. Meanwhile, VP-1 expression was largely suppressed by B4 in the muscle (Fig. 6i, see black arrows). These results revealed that B4 exhibited excellent antiviral efficacy and  type I IFN induction efficacy in vivo.

Discussion

There is currently no effective antiviral drug available to treat patients infected by EV71. Although the EV71 vaccine protected EV71-associated HFMD [28], another important question that viral RNA gives rise to frequent mutations in the newly synthesized viral genome, possibly restricting the use of vaccines [29]. Therefore, searching for effective antiviral drugs that target the host immune response is an advisable strategy. According to the theory of traditional Chinese medicine, HFMD is mainly caused by an accumulation of damp-heat and toxicity in the body [30], and its treatment may involve the usage of heat-clearing and detoxifying medicines [31]. Accordingly, many medicinal plant extracts and natural products with heat-clearing and detoxifying activities are worthy to be investigated. In addition to direct antiviral potencies, some medicinal plant extracts and natural products also exert anti-inflammatory effects to suppress the excessive inflammatory response caused by viral infection or regulate the immune system to resist viral infection. P. chinensis is a medicinal plant with a long history of use for heat-clearing and detoxification, used for the treatment of intestinal diseases, such as amebic dysentery [32]. A previous study showed B4, the major saponin component of P. chinensis, exhibited anti-inflammatory and immune-modulatory activities [21]. Here, we evaluated the antiviral activities of B4 against EV71 in cell cultures. B4 was a potential antiviral agent, with an IC50 value of 24.95 ± 0.05 μM on the EV71 replication in RD cells. Moreover, B4 could also effectively inhibit viral titer and viral protein in a dose-dependent manner.

It has been accepted that host immune responses resulting from virus infection play a vital role in clearing the infection. Activating or restoring the IFN system in an organism is critical for the initiation of host immune responses [12, 33]. Therefore, an IFN-inducing agent, which could also stimulate the response of type I IFNs, should be a high-priority target for antiviral drug development. In this study, we measured the expression of IFN-β and the downstream antiviral effectors IFN-stimulated genes, such as MxA, upon B4 administration. As reported in the previous study [34], the host cells could sense EV71 infection and stimulate IFN-β production. However, the levels of downstream IFN-stimulated gene MxA could not be successfully induced and even suppressed by EV71. Interestingly, we found that B4 could enhance the induction of IFN-β and robust the IFN-β response in the EV71-infected cells. IRF3 is a well-characterized signaling mediator/transcription factor that is essential for RLRs pathway activation. Activated IRF3 dimerizes and enters the nucleus to regulate both type I IFN and IFN-stimulated genes [35]. Consistent with the results of previous studies [36], IRF3 activation was not observed in the cells that were infected with EV71. Moreover, p-IRF3 significantly decreased in 24 h of EV71 infection. However, B4 treatment could dramatically enhance the phosphorylation and translocation of IRF3. To the best of our knowledge, this is the first report on B4 as a potent inducer of IFN-β.

As IFN-β plays essential roles in innate immunity, we further examined the antiviral action of B4 upon IFN-β knockdown or deletion. We found that IFN-β knockdown suppressed the B4-induced improvement of EV71-infected cells viability and reduction of viral titer. Meanwhile, there was no difference in expression of viral protein VP-1 and 3Cpro between IFN-β-knockdown cells and Scrambled-knockdown cells with B4 treatment. To confirm the essential role of IFN-β, we selected the Vero cell line to detect the antiviral efficacy of B4, because the gene loci encoding IFN-α/β are missing from the genomic DNA of Vero cells [37]. Indeed, B4 hardly exhibited antiviral efficacy in Vero cells. These results suggested that activation of host immune responses might be the antiviral mechanism of B4.

To gain more insight into the mechanisms behind the B4-regulated IFN-β induction, we analyzed the changes of infected cells’ proteomes upon B4 treatment. SLIAC-based quantitative proteomics approach was used, 92 proteins were identified, and most of the proteins were found to be involved in critical biological processes, such as Hippo signaling pathway, focal adhesion, and mRNA surveillance pathway. Meanwhile, we found 6 proteins involved in the Hippo signaling pathway, that is PATJ (gene ID:10207), 14-3-3β (gene ID: 7529), 14-3-3ζ (gene ID: 7534), PPP1CA (gene ID: 5499), PPP1CB (gene ID: 5500), PPP1CC (gene ID: 5501). Interestingly, the previous study showed that 14-3-3 molecular scaffolds display widespread functions in diverse cellular processes including signal transduction, cell apoptosis, and type I IFN responses [38]. Previous studies showed that 14-3-3β functions as the regulatory switch of RIG-I-mediated inflammatory signal transduction [39], therefore, we focused on 14-3-3β for follow-up research, and confirmed the upregulated expression in the cytoplasm of B4-treated cells. Since 14-3-3 usually binds to phosphorylated serine/threonine (pS/T) residues, thereby influencing interactions of target proteins either by enabling or blocking the access to other proteins [40]. As reported 14-3-3 proteins could bind to YAP/TAZ transcriptional coactivators, preventing inhibition of IRF3 activation by cytoplasmic sequestration of YAP [13]. Our results inferred that B4 could significantly promote 14-3-3β bind to YAP, and blocking the interaction of YAP and IRF3. Confocal immunofluorescence confirmed the colocalization of 14-3-3β and YAP in the cytoplasm of the infected cells after B4 treatment. In addition, the antiviral efficacy of B4 and innate immunity response were overall reduced in 14-3-3β-knockdown cells. We propose that 14-3-3β upregulation by B4 serves to cytoplasmic sequestration of YAP, and disrupt the downstream innate immune signaling retardation. Governance of 14-3-3 by B4 offers an effective approach for control of the innate immune response in infected cells. In addition, other studies showed that 14-3-3ζ interacted with and inhibited apoptotic effectors, such as Bad, caspase-2, and Bim, to promote cell survival [41]. These results could explain B4 slightly improved the survival of EV71-infected Vero cells and IFN-β/14-3-3β-knockdown cells.

Moreover, we employed the suckling mouse model [42] to evaluate the anti-EV71 effect of B4 in vivo. Encouragingly, the administration of B4 could strongly suppress EV71 replication and result in the improvement of muscle inflammation and limb activity. Meanwhile, B4 could also inhibit the expressions of HFMD biomarkers IL-10 and IFN-γ [43], attenuating complications of EV71 infection. To this end, activation of innate immunity pathway was confirmed by the expression of IFN-β in B4-treated mice.

Conclusion

Together with our current observations, these studies implicate B4, as a natural type I IFN signaling enhancer, can strongly inhibit EV71 propagation. By upregulating 14-3-3, B4 can enhance 14-3-3 interaction with YAP and activate type I IFN response (Fig. 7). Considering the low-toxic characteristics, B4 may serve as a potential lead compound for anti-EV71 drug development. Possible applications of B4 for other viral infections are presently being explored.

Fig. 7: Schematic diagram of the possible anti-EV71 mechanism of B4.
figure 7

Overview of B4 inhibits viral propagation through upregulating 14-3-3 and activating type I IFN response.

Full size image