1. Introduction
Enterovirus 71 (EV-A71) is the major etiological agent of hand, foot and mouth disease (HFMD) frequently seen in young children and has also been implicated in severe neurological manifestations, including meningitis, encephalitis, and acute flaccid paralysis [
1]. EV-A71 belongs to the
Picornaviridae family, and consists of a single-stranded RNA of positive polarity. The length of its genome is approximately 7.4 kilobases. This virus encodes a single polyprotein that is proteolytically cleaved to a P1 region consisting of four structural proteins (VP1, VP2, VP3, and VP4), a P2 region consisting of non-structural proteins 2A, 2B, and 2C, and a P3 region consisting of non-structural proteins 3A, 3B, 3C, and 3D. Several precursors exist prior to cleavage, including VP0, 2BC, 3AB, and 3CD. The maturation of an immature viral particle into infectious viral particle involves the RNA-mediated cleavage of VP0 into VP2 and VP4 [
2]. The genomic replication of enteroviruses occurs at cytosolic membranous vesicles. The vesicles observed during late infection resemble the membranous compartments induced by cellular autophagy [
2,
3,
4].
The autophagy pathway is a catabolic lysosome-dependent process that is triggered by various intracellular stimuli including nutrient deprivation, misfolded protein aggregation and infections [
5,
6]. Autophagy begins with the formation of membranes called phagophores, which mainly depends on Beclin-1/VPS34 activity. Phagophores elongate up to the fusion of both membrane extremities, trapping cytosolic contents in newly formed double membrane-containing vesicles, known as autophagosomes [
6]. The presence of microtubule-associated protein 1 light chain 3-II (LC3-II) on the autophagosome membrane is the hallmark of autophagy. During the early activation of autophagy, cytosolic LC3-I is cleaved and conjugated with the lipid phosphatidylethanolamine, converting it to the membrane-bound form, LC3-II [
7,
8]. Autophagosomes can fuse with late endosomes to generate amphisomes, which integrate vacuolar ATPases and contribute to intracellular vesicle acidification [
9,
10]. Autophagosomes or amphisomes can then fuse with lysosomes to generate autolysosomes with single membrane morphology. The cytosolic cargo sequestered within the lumina of autolysosomes, including LC3-II and long-lived proteins such as sequestosome-1 (SQSTM1/p62), are then degraded [
10,
11,
12]. Autophagosome maturation that leads to degradation of cargo by lysosomal proteases is termed autophagic flux [
13]. Autophagosome-lysosome fusion can be achieved through interaction with
N-ethylmaleimide-sensitive factor attachment receptor (SNARE) proteins [
14]. These SNARE proteins are syntaxin-17 (STX17), vesicle-associated membrane protein-8 (VAMP8) and synaptosome-associated protein of 29 kDa (SNAP29). Recently, the LC3-II-PLEKHM1-Rab7 complex was also found to mediate autophagosome-lysosome fusion [
15,
16]. Autophagic flux can be impeded by drugs which inhibit vesicle acidification, including the vacuolar ATPase inhibitor bafilomycin A1 (BAF-A1), as well as lysosomotropic compounds, such as chloroquine (CQ) and ammonium chloride (NH
4Cl) [
17,
18,
19]. The presence of regulatory mechanisms (such as the SNARE-mediated pathways) acting on late stages of autophagy implies that under certain circumstances, the generation of autophagosomes may not necessarily lead to autolysosome formation.
During late infection with poliovirus (PV), another member of the
Picornaviridae family, double-stranded RNA (dsRNA) from the membranous viral RNA replication complex co-localizes with autophagosomes [
20]. Other studies have found that autophagic vesicles are optimally produced in PV-infected cells at 5 h post-infection throughout the cytoplasm [
21]. These vesicles contain both LC3-II and the lysosomal-associated membrane protein 1 (LAMP1), which indicates that the fusion of autophagosomes with lysosomes is not compromised upon PV infection [
21]. In the context of EV-A71 infection, the autophagy machinery which is activated in infected cells contributes to the production of the virus in vitro and in vivo [
22]. However, whether formation of autolysosomes during EV-A71 infection is required and allowed to proceed remains unknown. Furthermore, the non-structural proteins of EV-A71 that could trigger the autophagic machinery have yet to be identified.
In the present study, we show that EV-A71 triggered autolysosome formation to mediate virus replication. The STX17, SNAP29, LC3B, and LAMP1 proteins were all important for the production of infectious EV-A71. The non-structural protein 2BC was identified as the viral protein that triggers formation of autolysosomes via its interaction with STX17 and SNAP29. Taken together, our study results provide new molecular insights into EV-A71 infection, particularly the role of the late maturation stage of autophagy in viral replication.
2. Materials and Methods
2.1. Reagents (Chemicals, Antibodies, and Small Interfering RNA)
BAF-A1, NH4Cl, E-64d protease inhibitor, leupeptin, and pepstatin A were purchased from Merck. Rapamycin, 3-methyladenine (3-MA), and CQ were obtained from Sigma (St. Louis, MO, USA). Monoclonal antibodies targeting LC3B, LAMP1, and β-actin were purchased from Cell Signaling Technology (Danvers, MA, USA), polyclonal antibody targeting GFP fusion protein (anti-GFP) without HRP conjugation, and monoclonal antibodies targeting SQSTM1/p62 and SNAP29 were purchased from Abcam (Cambridge, UK). Monoclonal antibodies targeting STX17 were obtained from Sigma. Anti-EV-A71 monoclonal antibody that targets both VP0 and VP2 (MAB979) structural proteins was purchased from Millipore (Billerica, MA, USA). Anti-EV-A71 polyclonal antibody that targets 2C and 3A was produced by immunizing rabbits with synthetic peptides of the non-structural proteins. Control, STX17, SNAP29, LC3B, and LAMP1 siRNAs (small interfering RNA) were purchased from Santa Cruz Biotechnology (Dallas, TX, USA).
2.2. Cell Lines and Viruses
Human rhabdomyosarcoma (RD; ATCC no. CCL-136) and human embryonic kidney (HEK-293; ATCC no. CRL-1573 and HEK293T; ATCC no. CRL-3216) cells were maintained in complete media containing Dulbecco’s modified Eagle’s medium (DMEM) (HyClone, Logan, UT, USA), while human cervix adenocarcinoma (HeLa; ATCC no. CCL-2) cells were maintained in Eagle’s minimum essential medium (EMEM), supplemented with l-glutamine and 10% fetal bovine serum (FBS) (Gibco, Billings, MT, USA) plus penicillin/streptomycin (200 U/mL), and cultured at 37 °C in a 5% CO2 incubator. RD and HEK-293 cells stably expressing LC3 (HEK-293/LC3) or both LC3 and LAMP1 (RD/LC3/LAMP1) were obtained as described below. All the stable cells were grown under similar conditions as the original cell line. At 80% confluence, cells were trypsinized with 0.25% trypsin (HyClone) and subcultured in the complete medium. RD cells were used for infection experiments, and RD, HEK293 and HEK293T cells were used for transfection experiments. Cell lines were then infected with EV-A71 strains UH1/PM/1997 (GenBank accession number AM396587), 41 (a gift from Tan Eng Lee, Singapore Polytechnic, Singapore, GenBank accession number AF316321), or 41-eGFP at the indicated multiplicity of infection (MOI). Unless otherwise stated, the UH1 strain was used in all experiments. The PV type 1 vaccine strain was obtained from the Diagnostic Virology Laboratory, University Malaya Medical Centre, Kuala Lumpur, Malaysia. After an hour, the unbound viruses were removed from the cells and then cultured with fresh medium supplemented with 2% FBS. Cells were harvested at the indicated times after viral infection.
2.3. Cytotoxicity Analysis
The cytotoxicity of CQ, BAF-A1, and NH4Cl was determined using the CellTiter 96 Aqueous One solution proliferation assay reagent (Promega, Madison, WI, USA). Briefly, these compounds were added at the indicated concentrations to overnight-cultured RD or HEK-293 cells, which were then incubated for the indicated time intervals. Subsequently, 20 µL of the proliferation assay reagent was added to each well of the 96-well plate. The plate was then analyzed at the absorbance of 490 nm after 2 h of incubation at 37 °C.
2.4. Plaque Assay
EV-A71 and PV viral stocks were titrated in RD cells. For the collection of intracellular virus, the infected cells treated with chemicals were washed with phosphate buffered saline (PBS) and collected in 0.5 mL PBS supplemented with 100 µg/mL MgCl2 (Merckmillipore, Billerica, MA, USA) and 100 µg/mL CaCl2 (Merck). The collected cells were then lysed by one cycle of freeze-thawing. After an hour of virus adsorption, the unbound viruses were removed and cells were overlaid with 0.9% carboxymethylcellulose (Sigma) in 2% FBS DMEM. After 72 h, the overlaid medium was removed and the cells were fixed with 3.7% formaldehyde followed by staining with crystal violet.
2.5. Construction of Plasmids
2B, 2C, 2BC, 3A, and 3AB genes of EV-A71 were chemically synthesized from GenScript and cloned into the BamHI and EcoRI sites of pEGFP-N1 vector (Clontech, Mountain View, CA, USA) to generate GFP fusion proteins. The STX17 gene was cloned into pGBKT7 (Clontech) and pCherry (Addgene, Cambridge, MA, USA).) vectors while the 2BC gene of EV-A71 was cloned into the pACT2 (Clontech) and pDEST27 (Thermo, Waltham, MA, USA) vectors. The pmRFP-LC3 and LAMP1-YFP vectors were obtained from Addgene. The tandem tagRFP-eGFP-LC3 vector was purchased from Thermo. The EV-A71 41-eGFP infectious cDNA clone was constructed by cloning the full-length genome of the virus with eGFP into pCR-XL-TOPO (Invitrogen, Carlsbad, CA, USA) as previously described [
23].
2.6. DNA and siRNA Plasmid Deliveries
For DNA transfection, RD, HEK-293, HEK293/LC3 and RD/LC3/LAMP1 cells were grown in 6-well plates and seeded at a density of 3 × 105 cells per well 24 h prior to transfection. The cells were transfected using Lipofectamine LTX with a total of 2.5 µg of DNA plasmid unless otherwise stated. For transduction of DNA vector, RD cells were infected with 100 MOI of baculovirus bearing the construct for 72 h. For transient knockdown of gene expression, 40 nM of siRNA construct (unless otherwise stated) was transfected using Lipofectamine 2000 into the host cells for 48 h.
2.7. Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis and Western Blotting
Cells were lysed in radioimmunoprecipitation assay (RIPA) buffer containing 50 mM Tris pH 8, 150 mM NaCl2, 0.1% SDS, 0.5% sodium deoxycholate, 1% Triton X-100 and protease (Sigma) and halt phosphatase inhibitors cocktail (Thermo). Lysed cells were electrophoresed by either 8% or 15% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a 0.2 μM pore size polyvinylidene fluoride membrane (Millipore). The membrane was incubated with antigen pretreatment solution (Thermo) for 10 min prior to blocking with 5% bovine serum albumin (BSA, Merck) in 0.1% Tween-20 Tris buffered saline (TBS) for 30 min. This was followed by the incubation of membrane with the indicated primary antibody diluted in primary antibody diluent (Thermo) for 1 h. The membrane was washed twice with 0.1% Tween-20 in TBS. Subsequently, the membrane was incubated with secondary antibody (HRP or IRDye conjugated) diluted in 2% BSA TBS 0.1% Tween-20. Finally, the membrane was washed twice before chemiluminescent development with Clarity Western ECL Substrate (Bio-Rad, Hercules, CA, USA) or without any substrate development. Images were captured with either BioSpectrum Imaging System (UVP, Cambridge, UK) or Odyssey SA Infrared Imaging System (Licor, Lincoln, NE, USA). The Western blot bands were quantified with Image Studio software.
2.8. Yeast Two-Hybrid Screening
A total of 47 autophagy-associated cDNAs were cloned into pGBKT7 from pDONR vector and transferred into the prey strain, while the 2BC gene of EV-A71 was cloned into pACT2 from pDONR vector and transferred into yeast bait strain AH109. Yeast cells were mated and subsequently plated on a selective medium lacking both leucine and histidine to determine the interaction-dependent transactivation of the
HIS3 reporter gene [
24].
2.9. Co-Affinity Purification (Co-AP)
Each expression vector (1.5 µg) was transfected into HEK293T cells for 48 h. After cell lysis, Glutathione (GST)-sepharose 4B beads (GE Healthcare, Chalfont St Giles, UK) were used for the co-affinity purification (co-AP) [
24]. GST-tagged 2BC and FLAG-tagged STX17 were detected using anti-GST and anti-FLAG monoclonal antibodies, respectively.
2.10. Co-Immunoprecipitation (Co-IP)
RD cells were washed twice with PBS pH 7.4 to remove traces of media. Cells were lysed with freshly prepared RIPA buffer supplemented with protease and Halt phosphatase inhibitors. Lysates were vortexed overnight at 4 °C. Following centrifugation, the clarified lysates were incubated with primary antibodies or control IgG bound Protein G Dynabeads (Invitrogen) for at least 3 h at 4 °C with end-over-end rotation. The antibody-bound magnetic beads were then washed 5 times with PBS. Binding partners of target proteins were eluted by incubating the beads with RIPA buffer containing SDS loading buffer for 10 min at 95 °C. The eluted proteins were separated on 15% SDS-PAGE and transferred on polyvinylidene fluoride (PVDF) membrane. Specific monoclonal antibodies were used to identify the binding partners of target proteins.
2.11. TaqMan Quantitative Real-Time PCR Assay (TaqMan qPCR)
Forward primer 5′-GAGCTCTATAGGAGATAGTGTGAGTAGGG-3′, reverse primer 5′-ATGACTGCTCACCTGCGTGTT-3′, and TaqMan probe 5′-6-carboxyfluorescein (FAM)-ACTTACCCA/ZEN/GGCCCTGCCAGCTCC-Iowa Black FQ-3′ were used to quantify EV-A71 as previously described [
25]. Intracellular viral RNA from infected cells was extracted by using QIAmp viral RNA Mini Kits (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The TaqMan real-time reverse transcription (RT)-PCR assay was performed using the StepOne Plus Real Time System (ABI, Foster City, CA, USA) with the TaqMan Fast Virus 1-step master mix (ABI). The synthesis of cDNA from RNA began with reverse transcription for 5 min at 50 °C, followed by amplification for 40 cycles at 95 °C for 3 s and 60 °C for 30 s.
2.12. Immunofluorescence Assay
Infected and transfected cell lines were fixed with 4% paraformaldehyde prepared in PBS for 20 min at room temperature and permeabilized with 0.25% Triton X-100 (Sigma) for 10 min. The cells were then blocked with Image-iT FX signal enhancer (Invitrogen) for 1 h. Primary and subsequently secondary antibodies were incubated in the cells for 1 h, respectively. To visualize nuclei, the cells were counterstained with 4′,6-diamidino-2-phenylindole (DAPI, Sigma) for 5 min. The slides were washed 3 times with PBS and covered with coverslips containing ProLong Gold antifade reagent (Invitrogen). The images were then analyzed using a Leica TCS SP5 confocal microscope and its associated software (Leica Microsystems, Wetzlar, Germany). Co-localization was analyzed using ImageJ software with Manders coefficient plug-in. Pearson’s correlation coefficients were calculated and reported as Rr values.
2.13. Transmission Electron Microscopy
Infected and transfected cell lines were fixed with 2.5% glutaraldehyde in 0.1 M sodium phosphate buffer (pH 7.4) for 1 h at room temperature. The cells were harvested and fixed with fresh 2.5% glutaraldehyde for at least 4 h at 4 °C, prior to post-fixation in 2% osmium tetroxide. The cells were then dehydrated with sequential washes in 50%, 70%, 90%, 95%, and 100% ethanol and embedded in resin. The areas that consist of cells were block mounted and thinly sliced before grid staining and analyzing using a HT7700 (Hitachi, Tokyo, Japan) or 1400 EX (Jeol, Tokyo, Japan) transmission electron microscope.
2.14. Statistical Analysis
The data presented are the means ± standard deviations (SD) obtained from three independent experiments. Error bars were used to represent the SD. Comparisons between groups were determined by either unpaired Student’s t-test, two-way, or one-way ANOVA using GraphPad Prism (version 6.0, GraphPad Software, San Diego, CA, USA). A p value of <0.05 was considered statistically significant.
4. Discussion
We have shown here that EV-A71 triggers autolysosome formation in RD cells. Inhibition of autophagosome-lysosome fusion impaired the replication and production of EV-A71 at the step of virus RNA replication. The induction of autophagic flux is also not strain-specific. Parallel to our findings, hepatitis C virus (HCV) requires the autophagy machinery to facilitate viral translation and replication [
31]. Infection of host cells with coxsackievirus B3 requires only the early stage of autophagosome formation, and not the late maturation stage of autophagosomes, for productive replication [
32]. For PV infection, autophagosome maturation does not affect translation, but inhibits virus RNA replication and promotes maturation of viral particles [
29]. In both poliovirus and EV-A71 infections, the inhibition of autophagosome maturation reduced the viral titres, indicating that this process is important in the life cycles of these closely related enteroviruses. A recent study showed that Beclin-1, class III PI3K Vps34,
N-glycanase (NGLY1) and the valosin-containing protein (VCP), which function in autophagosome formation, also facilitate the replication of EV-A71 [
33]. Therefore, EV-A71 could utilize multiple steps in the autophagic machinery to benefit its replication.
The maturation of autophagosomes requires v-ATPases, and these decrease the pH of the vesicles. The acidic environment in the vesicles is crucial for the activation of hydrolases and proteases, which are involved in the degradation of incoming cargo [
13]. Several viruses have developed unique mechanisms to thrive in these acidic compartments. PV triggers autophagosome maturation and this process mediates the cleavage maturation of VP0 into VP4 and VP2 structural proteins. However, inhibition of lysosomal proteases did not facilitate the production of PV infectious progeny [
29]. We found that three inhibitors of vesicle acidification (CQ, BAF-A1, and NH
4Cl) inhibited the expression level of viral proteins, viral RNA replication, and viral titres albeit at different sensitivities. Similar to PV, treatment with inhibitors of lysosomal proteases did not influence the production of infectious EV-A71.
The 2BC viral protein of PV triggers LC3 conversion, and in combination with 3A viral protein induces autolysosome accumulation [
21,
26]. The 2C non-structural protein of coxsackievirus A16 (CV-A16) triggers the formation of autophagosomes [
34]. In our study, we found that the 2BC non-structural protein of EV-A71 is capable of inducing autolysosome accumulation in transfected cells. However, the specific functional domains of 2BC that facilitate the accumulation of autolysosomes are yet to be identified. In support of our findings, a recent study has shown that the expression of neither 2B, 2C, 3A, 3B, 3AB, 3C nor 3D individually can trigger autophagy, revealing the potential unknown role of other precursor proteins in this machinery [
35]. Although we observed that 2BC plays a role in the accumulation of autolysosomes, other viral proteins may synergistically contribute to this process during EV-A71 infection.
The recent identification of STX17 and SNAP29 as SNAREs involved in the formation of autolysosomes led us to study their involvement in EV-A71 infection. STX17 localizes to the autophagosome membrane while SNAP29 is the adaptor protein that links STX17 to lysosomal SNARE [
14,
15,
16]. Additionally, the homotypic fusion and protein sorting (HOPS) is a tethering factor that has an identical function to SNAP29, mediating fusion between STX17 and the lysosomal transmembrane protein [
36]. A recent study has found that HCV inhibits autolysosome formation by reducing the expression of STX17, which increases the production of infectious HCV viral particles [
37]. In contrast, we found that silencing of STX17, SNAP29, LC3B, and LAMP1 inhibited autophagic flux and reduced the production of infectious EV-A71. The presence of low virus titres rather than complete inhibition may be attributed to an alternative mechanism of autophagosome-lysosome fusion via the LC3-PLEKHM1-RAB7 machinery [
15,
16]. We further found that 2BC non-structural protein of EV-A71 specifically interacts with STX17 and SNAP29, and this reaffirms the importance of STX17 and SNAP29 in EV-A71 infection. Yeast-2 hybrid and co-affinity purification assays further show that 2BC interacts with STX17. The findings from co-immunoprecipitation assays further indicate that 2BC binds to STX17 as well as SNAP29 in RD cells to regulate autolysosome formation. Importantly, the immature structural protein of EV-A71, VP0 also interacts with SNAP29. This interaction between VP0 and SNAP29 opens the possibility for further investigation into its role in EV-A71 infection.
As autophagy also regulates innate immune responses [
38], the interplay between EV-A71, autophagy and innate immunity should be further explored. Many autophagy proteins have now been shown to have non-autophagic roles. Both ATG13 and FIP200, components of the ULK complex that regulate upstream of the autophagy pathway via mTOR, have no beneficial effect on EV-A71 replication in U20S cells, suggesting that some autophagy-related proteins have unconventional functions in specific cell lines [
39].
In conclusion, we have shown in this study that EV-A71 infection triggers autolysosome accumulation in vitro. The formation of autolysosomes enhances the expression of viral proteins, viral replication, and production of infectious EV-A71 during the virus RNA replication step. This study opens the possibility for the development of novel antivirals that specifically target 2BC to inhibit formation of autolysosomes during EV-A71 infection.