The Role of Lipid Rafts in the Early Stage of Enterovirus 71 Infection Open Access

Enterovirus 71 (EV71) is a small, non-enveloped, positive-stranded RNA virus which belongs to human enterovirus species A of the Enterovirus genus of the Picornaviridae family. EV71 and coxsackievirus A16 are known as major causative agents of hand-foot-and-mouth disease (HFMD) in young children. EV71 infection can progress to severe neurological disease, including fatal encephalitis, aseptic meningitis and acute flaccid paralysis. Following the initial identification of EV71 in USA in 1969, outbreaks have been reported throughout the world, especially in Asia-Pacific region, including Australia, Malaysia, Japan, Vietnam and China [1]. Until now, no antiviral treatment has been developed to treat EV71 infections and effective vaccines are unavailable. Due to the serious consequences of its infection, much effort is being made to understand the basic mechanisms of EV71 life cycle.

The 7.4-kb EV71 RNA genome encodes a single polyprotein that is cleaved by the host and viral protease into four capsid proteins (VP1, VP2, VP3 and VP4) and seven nonstructural proteins (2A, 2B and 2C, 3A, 3B, 3C and 3D). The capsid protein VP1 is variable and confers distinct antigenic properties. Based on VP1 gene sequence, EV71 can be divided into genotype A, B, and C [2]. The capsid proteins mediate the initiation of infection by binding to a receptor on host membrane. To date, several cell type-specific candidate receptors have been identified, including AnnexinⅡ in HepG2 cells [3], sialylated glycans in DLD-1 intestinal cells [4], human P-selectin glycoprotein ligand-1 (PSGL-1) in Jurkat T cells [5] and scavenger receptor class B, member 2 (SCARB2) in rhabdomyosarcoma (RD) cells [6]. Host endocytic pathways, such as clathrin-mediated endocytosis, caveolae-dependent uptake, macropinocytosis/phagocytosis and cholesterol-dependent endocytosis, are commonly employed to mediate the infectious entry of virus particles into host cells [7]. Although it has been widely accepted that EV71 enters permissive cells via receptor-mediated endocytosis, the detail of virus-host interaction still needs more exploration.

Lipid rafts are membrane microdomains enriched in cholesterol and sphingolipids that accumulate in liquid-ordered, detergent-resistant membrane (DRM) domains [8]. They perform crucial roles in the establishment of cell polarity [9,10] and act as platforms mediating signal transduction and protein trafficking [11,12,13]. A number of raft-associated proteins have been identified, including tyrosine kinase cascades, G protein-coupled receptors, Src family kinases, and cytoskeletal and intermediate filament proteins [14]. The association of proteins with lipid rafts is determined by posttranslational modifications, such as glycophosphatidylinositol (GPI) anchor, cysteine acylation and N-terminal myristic acid tail [15].

Membrane cholesterol is critical for infection by many viruses. Recent studies have demonstrated the important functional roles of lipid rafts in the entry process of human immunodeficiency virus type 1, Ebola virus, hepatitis C virus, Japanese encephalitis virus, simian virus 40, and severe acute respiratory syndrome coronavirus [16,17,18,19]. There were also some evidences that influenza virus, Measles virus and rotavirus utilize lipid rafts for assembly and egress [20,21]. Members of the picornaviruses have been reported to enter cells in a lipid raft-dependent pathway. Internalization of echovirus 1 depends on caveolae and lipid rafts [22], while poliovirus RNA delivery requires cholesterol after endocytosis [23], However, the exact step at which lipid rafts play a role in EV71 infection and the mechanism involved remain uncharacterized. In this study, we demonstrated the critical role of lipid rafts in virus endocytosis and in the activation of PI3K/Akt signaling pathway. Our results provided new insights into the endocytic pathway involved in EV71 internalization, which will further improve our understanding of the underlying mechanism of viral entry.

The EV71 strain FJ08089 was isolated from a confirmed 8-year old boy with HFMD in Fujian Province, China. Human rhabdomyosarcoma (RD) cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (FBS, Invitrogen). The virus was propagated in RD cells and viral titers were determined via viral infectious plaque assays using RD cells.

RD cells were fixed with 4% paraformaldehyde for 20 min at room temperature and permeabilized with 0.1% Triton X-100 (Sigma). Then cells were stained with anti-EV71 VP1 monoclonal antibody 10F0 (Abcam) at room temperature for 1 h. After being washed with phosphate-buffered saline (PBS) for three times, the cells were reacted with Alexa 488 conjugated anti-mouse antibody (Invitrogen). Nuclei were counterstained with DAPI (Roche).

RNA was isolated from harvested cells using TRIzol reagent (Invitrogen), and RNA was prepared according to the instructions of the manufacturer. RNA from 10⁵ cell equivalents was analyzed by RT-PCR using a one-step reverse transcription-PCR (RT-PCR) system (Applied Biosystems). Genome copy numbers were normalized to GAPDH (glyceraldehyde-3-phosphate dehydrogenase) or β-actin gene copy numbers by using the comparative cycle threshold values determined in parallel [24]. The mRNA levels of 3-hydroxymethylglutaryl coenzyme A reductase (HMGCR) in RD cells infected with EV71 was determined by one-step RT-PCR. The reagent used was from Applied Biosystems inventoried assay (assay identification no. Hs00168352_m1).

RD cells were washed with PBS, and treated for 1 h at 37 °C with varying concentrations of methyl-β-cyclodextrin (MβCD) (Sigma). To replenish cellular cholesterol level, MβCD-treated cells were washed with PBS and then incubated with 150 μg/ml cholesterol (Sigma) for 1 h at 37 °C. After treatment, cells were washed three times with PBS and infected with EV71 for 1 h at 37 °C in serum-free medium. At 24 h post infection, cells were fixed and analyzed by immunofluorenscence assays and quantitative real-time RT-PCR as described above.

RD cells treated with 10 mM MβCD were washed with PBS for three times and then lysed in lysis buffer (1% Triton X-100, 25 mM HEPES (PH 7.4), 150 mM NaCl, 5 mM MgCl₂, 20 μg/ml aprotinin, 10 μg/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride). Lysates were spun at 10,000 g for 5 min to remove debris, and cholesterol levels were quantitated using an Amplex Red cholesterol assay kit (Molecular Probes) according to the manufacturer's instructions. A standard curve using purified cholesterol was generated for each experiment and normalized to the number of cells.

Expression of SCARB2 on the surface of RD cells and the binding of EV71 to RD cells were evaluated by flow cytometry. Untreated or MβCD-treated cells were collected and incubated with or without EV71 for 1 h at 4 °C. After washing three times with PBS, the cells were incubated with anti-SCARB2 antibody (R&D Systems) or anti-EV71 VP1 monoclonal antibody (10F0, Abcam) for 1 h at 4 °C. The cells were washed three times with PBS and incubated with Alexa 488-conjugated anti-goat antibody or Alexa 488-conjugated anti-mouse antibody for 30 min at 4 °C. The stained cells were analyzed by flow cytometry.

RD cells seeded on cover slips were washed twice with PBS and incubated for 45 min at 4 °C with DMEM containing 0.5% bovine serum albumin with or without of Alexa 555-conjugated Cholera toxin B (CTB) (Invitrogen) at 10 μg/ml and with or without purified EV71. After attachment at 4 °C, cells were transferred to 37 °C for 10 min to allow the endocytosis of CTB and EV71. Cells were washed twice with PBS and then subjected to immunofluorescence analysis as described above.

Virus endocytosis experiments were performed as previously described [25]. Briefly, virus attachment was permitted as described above and virus internalization was allowed at 37°C for indicated time points. Cells were washed, fixed and stained as described above with anti-VP1 antibody, followed by Alexa 488 -conjugated secondary antibodies. Nuclei were counterstained with DAPI. Cells were observed using a confocal fluorescence microscope (Zeiss).

RD cells were suspended and incubated with EV71 for 2 h at 4 °C. Cells were washed with PBS for three times and then lysed in lysis buffer as described above. After 1 h at 4 °C, insoluble material was pelleted at 16,000 g (25 min, 4 °C) and supernatants were precleared by incubation with protein G-Sepharose (Roche Molecular Biochemicals). The supernatants were then incubated with anti-SCARB2 antibody (R&D Systems) for 1 h at 4 °C, followed by overnight incubation with protein G-Sepharose. Immune complexes were collected by centrifugation, washed three times in lysis buffer, and then analyzed by SDS-PAGE (12% acrylamide) under non-reducing conditions. The proteins were transferred to PVDF membranes and probed with primary antibody, followed by horseradish peroxidase (HRP)-conjugated antibodies (Santa Cruz Biotechnology). Bound antibodies were detected with the ECL Plus Western blotting detection reagents (PerkinElmer Life Sciences).

Akt phosphorylation in EV71-infected cells was detected as previously described [26]. Briefly, RD cells were infected with EV71 at 37 °C. After 20, 40 and 60 min of viral infection, the cell lysates were analyzed by Western blotting with specific antibodies for phospho-Akt or Akt (Cell Signaling Technology).

Specific siRNAs targeting PIK3CG (M-005274-02) were obtained from Dharmacon. As a control, we used the non-targeting siRNA sequence (D001810-10) (Dharmacon). RD cells were transfected with 30 nM siRNA using Lipofectamine RNAiMAX according to the manufacturer's instructions. Subsequent experiments were carried out 72 h after siRNA transfection. Knockdown efficiencies were analysed by Western blotting.

Cells (5×10⁷) were washed twice with ice-cold PBS and lysed on ice for 30 min in 1 ml TNE-buffer [25 mM Tris (pH 7.5), 150 mM NaCl, 5 mM EDTA, and complete protease inhibitor cocktail (Roche)] with 1% Triton X-100. The cell lysates were homogenized and then centrifuged at 4 °C for 5 min at 1000 g to remove insoluble materials. The supernatant was mixed with 1 ml 80% sucrose in TNE buffer, placed at the bottoms of ultracentrifuge tubes and overlaid with 7 ml 30% and 3 ml 5% sucrose in TNE buffer. The cell lysates were ultracentrifuged at 4 °C for 18 h at 36,000 rpm in a SW41 rotor (Beckman). After centrifugation, twelve 1-ml fractions were collected from the top to the bottom and analyzed immediately by Western blotting.

About 20 μl of each sucrose gradient fractions was resolved by SDS-PAGE (12% acrylamide). Immunoblotting was performed using anti-caveolin-1 (Sigma), anti-CD71 (Santa Cruz Biotechnology), anti-SCARB2, anti-EV71 VP1, anti-phospho-Akt and anti-Akt antibodies separately and HRP-conjugated secondary antibodies. Blots were visualized by chemiluminescence.

All statistical analyses were performed using Microsoft Excel software. All graphs represent means ± standard deviations (SDs). P values for all data were determined using the paired t test.

To study the role of cholesterol in EV71 infection, we used MβCD to deplete cell membrane cholesterol. MβCD is a derivative of cyclic oligosaccharides and has a lipophilic property that extracts cholesterol from membranes, resulting in lipid raft disruption [27]. RD cells were treated with the increasing concentrations of MβCD (from 2.5 to 10 mM) for 1 h at 37 °C, washed and then incubated with EV71. After this treatment, the expression of viral protein detected by the anti-VP1 antibody decreased markedly in a dose-dependent manner (Fig. 1A). The pretreatment of increasing concentrations of MβCD decreased more than 85% intracellular EV71 RNA levels (Fig. 1B). To further investigate that the inhibitory effect was due to cholesterol depletion and not due to a pleiotropic effect of MβCD on other membrane components, we supplemented MβCD-depleted cells with exogenous cholesterol in the same experiment. As shown in Fig. 1C, the decrease in VP1 expression was reversed when MβCD-treated RD cells were incubated with exogenous cholesterol prior to infection. Adding cholesterol to MβCD-treated RD cells also restored intracellular viral RNA levels (Fig. 1D). In order to determine whether MβCD treatment depletes cellular cholesterol levels, we quantitated the amount of cellular cholesterol before and after MβCD treatment. MβCD-treated RD cells reduced the total intracellular pools of cholesterol by 65% compared to mock-treated cells (Fig. 1E). These data demonstrate that EV71 infection in RD cells is dependent on lipid rafts.

To specifically determine which stage of EV71 infection that was affected by MβCD treatment, RD cells were incubated with MβCD for 1 h prior to the start of infection or at the indicated times after infection. As shown in Fig. 2, EV71 infection was maximally inhibited when MβCD was added before viral entry. Interestingly, addition of MβCD at 30 min or later time points after the incubation of EV71 had little to no effect on VP1 expression (Fig. 2A) or intracellular viral RNA levels (Fig. 2B). Since MβCD only had an antiviral effect when added prior to or at the start of infection, these data suggest that MβCD act at a very early stage in EV71 entry.

To determine if cholesterol biosynthesis is influenced during EV71 infection, RD cells were infected with EV71 and then collected at 1, 12 and 24 h post infection for RNA isolation. The levels of mRNA for HMGCR, an enzyme involved in cholesterol biosynthesis, were measured by qRT-PCR. As shown in Fig. 2C, a significant induction of HMGCR was observed in EV71-infected cells at early time points. At 24 h post infection, HMGCR levels were similar to those in uninfected controls (Fig. 2C), indicating that de novo synthesis of cholesterol may not be required for EV71 RNA replication. Collectively, these data demonstrate the requirement of lipid rafts for EV71 entry.

Lipid rafts are cholesterol-enriched microdomains, where many cellular proteins, including viral receptors, are preferentially localized. Since membrane cholesterol played an important role in early infection stages, the interaction of EV71 with its receptor SCARB2 may occur in membrane microdomains. To test this hypothesis, RD cells were incubated with EV71 for 1 h at 37 °C and then lipid rafts were isolated from RD cells. Total membranes from mock- and virus-infected RD cells were treated with 1% Triton X-100 at 4 °C, fractionated using a sucrose flotation gradient, and the individual fractions assayed for the presence of EV71 VP1 and SCARB2 by immunoblotting. In mock-treated cells, the majority of SCARB2 was found in detergent-insoluble fractions (Fig. 3A). In EV71-infected cells, we found that VP1 protein is located into the raft fractions, the same fractions as its cellular receptor (Fig. 3B). In control experiments, the raft marker caveolin-1 appeared in low density fractions and the non-raft marker transferrin receptor (CD71) marked in the dense fractions (Fig. 3). Thus, the biochemical fractionation analysis suggests the association of EV71 with lipid rafts during viral entry.

Since EV71 VP1 and SCARB2 partitioned in lipid rafts, we assessed if the effect of cholesterol depletion of MβCD on viral entry is due to the inefficient virus binding. Equal amounts of EV71 were added to untreated or MβCD-treated RD cells at 4 °C for 1 h. Unbound virus were removed by extensive wash and the amount of cell-associated viruses were determined by flow cytometry using anti-EV71 VP1 antibody. No obvious shifting of peaks was observed, indicating that viral protein bound comparatively to MβCD-treated or untreated RD cells (Fig. 4A). These results suggest that depletion of cholesterol does not affect virus attachment.

RD cells were pre-incubated with MβCD and infected with EV71, after which flow cytometry analysis, biochemical fractionation and co-immunoprecipitation were performed. As shown in Fig. 4B, cell surface expression of SCARB2 was not affected by cholesterol depletion with MβCD. In MβCD-treated cells, high levels of SCARB2 and EV71 VP1 protein were detected, but these were largely confined to the non-raft fractions as well as caveolin-1 (Fig. 4C). The anti-VP1 antibody recognized the 34-kDa protein in the immunoprecipitated complexes of both untreated and MβCD-treated cells (Fig. 4D, lane 3), while no protein bands were observed using a normal rabbit IgG (Fig. 4D, lane 2) or in the lysates without EV71 infection (Fig. 4D, lane 1). As expected, the presence of SCARB2 in total extract from RD cells was also observed (Fig. 4D, lane 1). Taken together, these results demonstrate that MβCD shifts EV71 and its receptor SCARB2 to non-raft microdomains, but the interaction of EV71 VP1 protein with SCARB2 was not affected by MβCD treatment.

To further demonstrate that EV71 entry is mediated by lipid rafts, we examined the internalization of EV71 VP1 and Alexa 555-conjugated cholera toxin B (CTB) by RD cells. CTB has been shown to enter cells by binding to its receptor gangliosides (GM1) in a lipid raft-dependent manner [28]. We permitted EV71 (MOI=0.1) and 10 μg/ml fluorescent CTB to bind to cells at 4 °C for 10 min. Cells were then shifted to 37 °C for 30 min to allow internalization, washed extensively, fixed and analyzed by immunofluorescence microscopy using anti-VP1 antibodies. As shown in Fig. 5A, colocalization of virus and CTB were observed at the plasma membrane of RD cells. This strongly suggests the involvement of lipid rafts for internalization of EV71 virions.

We next examined the effect of MβCD on EV71 internalization. Cells were infected with EV71 (MOI=0.1) at 4 °C for 30 min for virus attachment and then shifted to 37 °C for 30 min to allow internalization. Immunofluorenscence assays were performed to track the entry process and cellular localization of EV71 within infected cells at indicated time points. In control cells, EV71 particles were observed predominantly attached to the plasma membrane of the cells at time 0 (Fig. 5B). At time 30 min, virus particles were clearly observed in an intracellular location (Fig. 5B). No intracellular EV71 particles could be detected in MβCD-treated cells at 30 min p.i., and a few of the viral particles were present at the cell peripheries in the cholesterol-depleted RD cells (Fig. 5B). Na⁺-K⁺-ATPase, a cell membrane marker, was used to demonstrate the retaining or release of viral VP1 protein from the plasma membrane (Fig. 5C). These data demonstrate that cholesterol depletion by MβCD affects EV71 internalization.

Previous reports have shown that EV71 infection at early steps stimulates phosphorylation of PI3K/Akt signaling pathways [26]. To detect the role of PI3K/AKT signaling in viral entry, the effect of PI3K depletion by small interfering RNA (siRNA) treatment on EV71 entry was investigated. The efficiency of siRNA-mediated knockdown of PIK3CG (PI3K gamma complex) was analysed by Western blotting. As shown in Fig. 6A, siRNA specific for PIK3CG reduced PI3K expression. Depletion of PI3K resulted in a strong inhibitory effect on EV71 VP1 expression (Fig. 6B), suggesting PI3K/Akt signaling was involved in EV71 infection. Since lipid rafts act as signaling platforms and initiate a variety of signal transduction processes critical for virus entry and propagation, we investigated whether membrane cholesterol is required for triggering PI3K/Akt signaling during EV71 entry. Indeed, cell lysates from RD cells showed increased phosphorylation levels of Akt as early as 20 min post infection (Fig. 6C). As shown in Fig. 6D, a dramatic decrease in Akt phosphorylation was observed after viral infection in MβCD-treated cells. These results clearly suggest that disruption of lipid rafts immediately after EV71 infection impairs the activation of PI3K/Akt signaling pathways.

Lipid rafts often serve as an entry site for many viruses. Numerous studies have shown that extraction of cholesterol by MβCD inhibits viral infection in the early stages [19,29,30,31,32]. Our results support the notion that lipid rafts on the plasma membrane facilitate entry of EV71. We observed a reduction in the number of EV71-positive cells and the viral titers in cholesterol-depleted cells. This inhibition probably occurs at initial steps of infection, since no significant inhibitory effect was observed when rafts were disrupted following EV71 adsorption. A dramatic induction of cholesterol biosynthesis during early infection lends further support to the critical role of lipid rafts in EV71 entry. Although no significant upregulation of cholesterol biosynthesis was observed after viral entry in our experiments, we cannot rule out the possibility of the presence of intracellular cholesterol sufficient to participate in EV71 replication. It is possible that cellular cholesterol levels were replenished by de novo synthesis upon the removal of the drug, allowing for cholesterol engagement in viral replication, assembly or egress.

To infect host cells, the virus initially binds to receptor molecules present on the cell surface. Cholesterol-enriched membrane microdomains may provide a convenient platform to concentrate the receptors on host cell membrane [18]. Our data indicated that the interaction of viral capsid protein with its receptor occurs preferentially at lipid rafts. A number of non-enveloped viruses recognize their receptors in membrane microdomains to access endocytic pathways, thereby facilitating viral entry. Decay-accelerating factor (DAF or CD55), a GPI-anchored membrane glycoprotein, functions as a receptor by many enteroviruses including enterovirus 11 [33], enterovirus 70 [34], hemagglutinating echoviruses [35], and coxsackievirus A21 [36]. Components of the receptor complex for group A rotavirus all localize to cholesterol-rich microdomains prior to or as a result of virus attachment, thus favoring binding and internalization of rotavirus particles [30,37]. A decrease in productive virus entry may reflect a reduction in virus binding to cells. We then investigated the effect of cholesterol extraction on virus attachment. No significant decrease in the cell surface expression of SCARB2 and the amount of bound virus were observed between MβCD-treated and control untreated RD cells, indicating that cholesterol depletion of cell membrane does not perturb the attachment of virus, but alter viral entry at post-binding steps. We also showed that extraction of cholesterol by MβCD excluded SCARB2 from raft domains. Cholesterol levels are important for maintaining membrane fluidity, and its removal can reduce lateral diffusion within the cell membrane. This reduction in fluidity could perhaps affect migration of SCARB2 within the membranes, which may impair a chain of events that could participate in EV71 pathogenesis triggered by the interaction of viral capsid protein with SCARB2 located in cholesterol-enriched microdomains.

Using confocal microscopy, we found that EV71 particles failed to be internalized in cholesterol-depleted cells. Our results do not dispute the role of clathrin in EV71 endocytosis [38,39], but suggest that lipid rafts are also required for EV71 internalization. Many viruses are known to enter cells via multiple pathways, such as simian virus 40 (caveolar and cholesterol-dependent pathways) [40], adenovirus (clathrin-mediated endocytosis and macropinocytosis) [41], coxsackievirus A9 (β2-microglobulin, dynamin and Arf6-mediated internalization) [42], and echovirus 1 (caveolar and dynamin-2 dependent pathways) [22]. Further experiments directed to analyze the endocytic pathways involved in EV71 internalization as well as the role of other proteins in this process are being performed in our laboratory.

Lipid rafts act as functional platforms for multiple signaling and trafficking processes. A number of viruses, including enveloped and non-enveloped viruses, hijack lipid raft machinery to gain access to various signaling elements in the membrane cholesterol for viral entry or the later stages of infection [43]. In picornaviruses infection, the importance of PI3K/Akt pathways in maintaining host cell survival and stimulating anti-apoptotic pathways during early infection has been established [26,44,45]. Indeed, our results showed that EV71 activated PI3K/Akt at the early stages of infection. Recently, it has been demonstrated that the silencing of the PIK3CG and PIK3C2G, genes involved in PI3K/Akt signaling pathways, caused a reduction in viral infection [38]. It seemed that the activation of PI3K/Akt signaling pathway was lipid raft-dependent, as disruption of lipid rafts prior to EV71 infection resulted in significant decline in Akt phosphorylation. Therefore, we suggest the critical role of lipid rafts in initiation of signaling cascades during early EV71 infection.

In conclusion, our data demonstrate the importance of lipid rafts for EV71 endocytosis as well as in triggering signaling pathways at the early steps of life cycle. These results further our understanding of the mechanisms of EV71 entry and provide a new target for antiviral potential.

There are no conflicts of interest to disclose for all authors.

We are grateful to You-Heng Wei (Fudan University, Shanghai, China) for excellent technical assistance. This work was funded by Natural Science Foundation of China (31370196, 81221061 and 81302812), Military S&T Project (AWS11C001), National S&T Major Project for Infectious Diseases Control (2012ZX10004801-002-005).

Y.-Z. Zhu, D.-G. Wu and H. Ren contributed equally to this work.