Foot-and-Mouth Disease Virus Nonstructural Protein 2C Interacts with Beclin1, Modulating Virus Replication

Abstract

Foot-and-mouth disease virus (FMDV), the causative agent of foot-and-mouth disease, is an Apthovirus within the Picornaviridae family. Replication of the virus occurs in association with replication complexes that are formed by host cell membrane rearrangements. The largest viral protein in the replication complex, 2C, is thought to have multiple roles during virus replication. However, studies examining the function of FMDV 2C have been rather limited. To better understand the role of 2C in the process of virus replication, we used a yeast two-hybrid approach to identify host proteins that interact with 2C. We report here that cellular Beclin1 is a specific host binding partner for 2C. Beclin1 is a regulator of the autophagy pathway, a metabolic pathway required for efficient FMDV replication. The 2C-Beclin1 interaction was further confirmed by coimmunoprecipitation and confocal microscopy to actually occur in FMDV-infected cells. Overexpression of either Beclin1 or Bcl-2, another important autophagy factor, strongly affects virus yield in cell culture. The fusion of lysosomes to autophagosomes containing viral proteins is not seen during FMDV infection, a process that is stimulated by Beclin1; however, in FMDV-infected cells overexpressing Beclin1 this fusion occurs, suggesting that 2C would bind to Beclin1 to prevent the fusion of lysosomes to autophagosomes, allowing for virus survival. Using reverse genetics, we demonstrate here that modifications to the amino acids in 2C that are critical for interaction with Beclin1 are also critical for virus growth. These results suggest that interaction between FMDV 2C and host protein Beclin1 could be essential for virus replication.

INTRODUCTION

Foot-and-mouth disease virus (FMDV), a single-stranded positive-sense RNA virus, is the causative agent of foot-and-mouth disease (FMD), a highly contagious viral disease of domestic and wild cloven-hoofed animals. Seven serotypes of FMDV exist (A, O, C, Asia, SAT1, SAT2, and SAT3), and recovery from one serotype does not provide immunity against the others (7, 22). The infectious virion is a nonenveloped icosahedron composed of four structural proteins: VP1, VP2, VP3, and VP4. The genome of approximately 8,400 nucleotides has a single open reading frame (ORF) that is translated into a polyprotein, which is processed by the three viral proteases Lpro, 2A, and 3C into the polypeptide products P1 (VP1 to VP4), P2 (2A, 2B, and 2C), and P3 (3A, 3B, 3Cpro, and 3Dpol). Further cleavage of these regions yields 14 mature virus proteins, along with several protein intermediates, that are needed for viral replication (18, 19).

During replication, FMDV forms a replication complex produced by the rearrangement of intracellular membranes into vesicular structures containing viral nonstructural proteins (2, 31). Many other positive-strand RNA viruses also initiate production of replication complexes upon infection of a cell (3, 4, 11, 38, 39). FMDV 2C, a 318-amino-acid protein, is the largest membrane-binding component of the virus RNA replication complex (30). FMDV 2C binds ssRNA nonspecifically, has ATPase activity (44), and is involved in the RNA replication complex (25). 2C prediction studies suggest that an amphipathic helix in its N terminus would be responsible for its ability to bind the intracellular membranes (46). The structure and size of 2C suggests that it plays multiple roles in the process of virus replication, including interactions with several host cellular factors during infection.

In order to better understand the role of FMDV 2C in virus replication, we attempted to identify host cell proteins that interact with 2C utilizing a yeast two-hybrid approach. Our screen identified a host protein, Beclin1, as a binding partner for 2C of FMDV serotypes O1 Campos and A24 Cruzeiro. Beclin1 is a central regulator of the autophagy process that regulates multiple steps of the autophagy pathway (48). Beclin1 is involved in the initiation of the autophagy pathway by marking membranes to form the first double membrane structure in the autophagy pathway, the phagophore (21, 37). Later in the autophagy pathway, Beclin1 functions to mediate autophagosome to lysosome fusion (28, 37). We have previously reported that FMDV 2C colocalized with autophagosome marker LC3 and that downregulation of the autophagy pathway resulted in decreased viral yields, while induction of the autophagy pathway resulted in an increase in virus titer (35). Thus, the cellular autophagy pathway appears to be critical for FMDV replication.

Here we show that interaction between FMDV 2C and cellular Beclin1, initially identified using yeast two-hybrid screening, actually occurs in FMDV-infected cells, as confirmed using coimmunoprecipitation and confocal microscopy. Importantly, modulations of the expression of Beclin1, as well as Bcl-2 (another host protein playing a critical role in the autophagy pathway), can have a negative effect on FMDV replication in cell culture. We also provide evidence that binding of 2C to Beclin1 may block the fusion of FMDV-containing autophagosomes to lysosomes, preventing virus degradation. In addition, identification of areas within 2C interacting with Beclin1 was performed by alanine scanning mutagenesis. These mutations were introduced into an infectious clone of FMDV and were determined to be critical for virus replication, suggesting that the 2C-Beclin1 interaction may play a significant role in virus replication.

MATERIALS AND METHODS

Cell lines, viruses, and plasmids.

Human mammary gland epithelial cells (MCF-10A) were obtained from the American Type Culture Collection (catalogue no. CRL-10317) and maintained in a mixture of Dulbecco minimal essential medium (Life Technologies, Grand Island, NY) and Ham F-12 medium (1:1; Life Technologies) containing 5% heat-inactivated fetal bovine serum (Thermo Scientific, Waltham, MA), 20 ng of epidermal growth factor (Sigma-Aldrich, St. Louis, MO)/ml, 100 ng of cholera toxin (Sigma-Aldrich)/ml, 10 μg of insulin (Sigma-Aldrich)/ml, and 500 ng of hydrocortisone (Sigma-Aldrich)/ml.

FMDV type O1 strain Campos (O1C) was derived from the vesicular fluid of an experimentally infected steer. The virus was grown in baby hamster kidney-21 (BHK-21) cells, and the titer was determined by plaque assay on BHK-21 cells according to standard techniques (35).

Plasmids were purchased from Addgene, i.e., green fluorescent protein (GFP)-Bcl-2 (47) (Addgene plasmid, catalog no. 17999) and PCDNA4-Beclin1-FL (43) (Addgene plasmid, catalog no. 24388), or bought commercially, i.e., phrGFP II-N mammalian expression vector (Agilent Technologies, catalog no. 240145).

For viral replication studies, MCF-10A cells were plated at a density of 1 × 10⁶ per well in a six-well plate (Falcon; Becton Dickinson Labware, Franklin Lakes, NJ). The indicated plasmids were transfected into cells using Fugene (Roche Applied Science, Indianapolis, IN) according to the manufacturer's protocol. After 24 h, the cells were infected with FMDV type O1C at the specified multiplicity of infection (MOI) or mock infected. Virus was allowed to absorb for 1 h and then acid washed, and then fresh medium was added containing 0.5% serum. Samples were taken at the indicated time points.

Antibodies and reagents.

Monoclonal antibody (MAb) 12FB, directed against the FMDV type O1 structural protein VP1, has been previously described (42). MAb 3D10, directed against the FMDV type O1 nonstructural protein 2C, was developed at the Istituto Zooprofilattico Sperimentale della Lombardia e dell Emilia-Romagna, Brescia, Italy. Rabbit antibodies to LC3 (AnaSpec, Fremont, CA), Lamp1 (BD Biosciences, San Jose, CA), and UVRAG1 (Sigma-Aldrich) were used as autophagy markers. Protein disulfide isomerase (PDI; Affinity Bioreagents, Golden, CO) and bafilomycin A1 (Sigma-Aldrich) from Streptomyces griseus, an inhibitor of autophagosome fusion to lysosomes, were prepared as a 0.5 mM stock solution in dimethyl sulfoxide (DMSO) and then diluted in minimal essential medium (MEM) to 0.5 μM.

Infection and transfection of cells for confocal and deconvolution microscopy.

Subconfluent monolayers of MCF-10A cells grown on 12-mm glass coverslips in 24-well tissue culture dishes were transfected with the indicated plasmids. After 24 h, the samples were infected with FMDV O1C at an MOI of 10 50% tissue culture infective dose(s) (TCID₅₀)/cell in MEM (Life Technologies) containing 0.5% heat-inactivated fetal bovine serum, 25 mM HEPES (pH 7.4), and 1% antibiotics. After the 1-h adsorption period, the supernatant was removed, and the cells were rinsed with ice-cold 2-morpholinoethanesulfonic acid (MES)-buffered saline (25 mM MES [pH 5.5], 145 mM NaCl) to remove unabsorbed virus. The cells were washed once with medium before fresh medium was added, followed by incubation at 37°C. At the indicated time points after infection, the cells were fixed with 4% paraformaldehyde (EMS, Hatfield, PA) and analyzed by confocal or deconvolution microscopy. To express GFP-BCL2, Beclin1-FLAG, or GFP protein, monolayers of MCF-10A cells were transfected with 0.5 μg of plasmid DNA using FuGene (Roche, Mannheim, Germany) according to the manufacturer's recommendations. At 19 to 24 h posttransfection, the cells were infected as described above and then fixed with 4% paraformaldehyde (EMS) at the appropriate times.

After fixation, the paraformaldehyde was removed, and the cells were permeabilized with 0.5% Triton X-100 for 5 min at room temperature, followed by incubation in blocking buffer (phosphate-buffered saline [PBS], 5% normal goat serum, 2% bovine serum albumin, 10 mM glycine, 0.01% thimerosal) for 1 h at room temperature. The fixed cells were then incubated with the primary antibodies overnight at 4°C. When double labeling was performed, the cells were incubated with both antibodies together. After three washes with PBS, the cells were incubated with the appropriate secondary antibody, goat anti-rabbit immunoglobulin G (IgG; 1/400; Alexa Fluor 594 or Alex Flour 647; Molecular Probes) or goat anti-mouse isotype-specific IgG (1/400; Alexa Fluor 488 or Alexa Fluor 594; Molecular Probes), for 1 h at room temperature. After this incubation, the coverslips were washed three times with PBS, counterstained with the nuclear stain TOPRO-iodide 642/661 (Molecular Probes) or DAPI (Life Technologies) for 5 min at room temperature, washed as described above, mounted, and examined using either a Leica scanning confocal microscope or a Nikon Eclipse 90i deconvolution microscope. The data were collected utilizing appropriate prepared controls lacking the primary antibodies, as well as anti-FMDV antibodies in uninfected cells, to give the negative background levels and to determine channel crossover settings. The captured images were adjusted for contrast and brightness using Adobe Photoshop software.

Knockdown of Beclin1 using siRNA.

Small interfering RNA (siRNA) SMARTpools consisting of four RNA duplexes targeting Beclin1 (20) and a control siRNA (siRNA Glo) were purchased from Dharmacon (Lafayette, CO). MCF-10A cells were grown to densities of 10⁵ cells per well in 24-well tissue culture dishes in 1 ml of medium without antibiotics, followed by transfection using DharmaFECT (Dharmacon) as recommended by the manufacturer's protocol. At 48 h posttransfection, the cells were infected with FMDV O1C at an MOI of 1 for 1 h at 37°C. After adsorption, the inoculum was removed, and the cells were rinsed with ice-cold MES to remove residual virus particles. The cells were then rinsed with MEM containing 1% fetal bovine serum and 25 mM HEPES (pH 7.4), followed by incubation at 37°C. At the indicated times postinfection, samples were drawn for titer assays using BHK-21 cell monolayers.

Viral replication in the presence of bafilomycin A1.

MCF-10A cells were incubated with bafilomycin A1 for 30 min at 37°C prior to infection. The cells were then infected with FMDV O1C at an MOI of 10 in the presence or absence of bafilomycin A1. At the end of the adsorption period, the supernatant was removed, followed by a single rinse with ice-cold MES-buffered saline to inactivate the unabsorbed virus. The cells were washed once with medium before fresh medium with or without bafilomycin A1 was added. One set of cultures was immediately frozen at −70°C, whereas the other set of plates was incubated for an additional 4 h at 37°C and then moved to −70°C. The plates were then thawed, the cell debris was removed by centrifugation, and virus titers were determined based on the TCID₅₀ on BHK-21 cell monolayers.

Coimmunoprecipitation of FMDV 2C and Beclin1.

MCF-10A cells were grown to 90% confluence and then infected at an MOI of 10 or mock infected. The cells were lysed at 2.5 h postinfection using protease inhibitors (Protea Biosciences, Morgantown, WV) and radioimmunoprecipitation assay (RIPA) buffer (Teknova, Hollister, CA). The protein lysate was used for immunoprecipitation using protein G-beads (Sigma-Aldrich) coupled to an MAb directed against Beclin1 (H-300). The cell lysate was incubated with anti-Beclin1 antibody for 2.5 h and then incubated with antibody and beads overnight at 4°C. The beads were washed five times using RIPA buffer, and then protein elutes were collected for each sample and examined by Western blot probing for anti-2C (3D10).

Development of the cDNA library.

A cDNA expression library was constructed (Clontech, Mountain View, CA) using tissues susceptible to FMDV infection (dorsal soft palate, interdigital skin, middle tongue epithelium, and middle anterior lung) from a healthy, uninfected bovine. Total RNA was extracted using an RNeasy extraction kit (Qiagen, Valencia, CA). Contaminant genomic DNA was removed by DNase treatment using Turbo DNA-free (Ambion, Austin, TX). After DNase treatment, genomic DNA contamination of RNA stocks was assessed by real-time PCR amplification targeting the bovine β-actin gene. RNA quality was assessed using RNA nanochips on an Agilent Bioanalyzer 2100. Cellular proteins were expressed as GAL4-AD fusion proteins, while FMDV 2C was expressed as GAL 4-BD fusion proteins.

Library screening.

The GAL4-based yeast two-hybrid system provides a transcriptional assay for detection of protein-protein interactions (10, 14). The bait protein, FMDV strain O1C 2C protein, was expressed with an N terminus fusion to the GAL4-binding domain (BD). Full-length 2C protein (amino acids 1082 to 1399 of the FMDV polyprotein) was used for screening and for full-length mutant protein construction. The previously described bovine cDNA library of proteins fused to GAL4-AD were used as prey. The reporter genes used here are histidine and adenine for growth selection. The bovine library used contains more than 3 × 10⁶ independent cDNA clones. For screening, the yeast strain AH109 (Clontech) carrying 2C protein was transformed with library plasmid DNA and selected on plates lacking tryptophan, leucine, histidine, and adenine. Tryptophan and leucine are used for plasmid selection, and histidine and adenine are used for the identification of positive interacting library fusions. Once identified, the positive library plasmids were recovered in Escherichia coli and sequenced to identify the cellular interacting protein. Sequence analysis also determined whether the library proteins (cellular) were in frame with the activation domain. To eliminate false-positive interactions, all library-activation domain fusion proteins were retransformed into strains carrying the 2C-binding domain fusion protein, as well as into strains carrying Lam-binding domain fusion. Lam is human lamin C, commonly used as a negative control in the yeast two-hybrid system since lamin C does not form complexes or interact with most other proteins. The Beclin1 recovered from the library contained amino acids 1 to 211 of the bovine Beclin1 (NCBI reference sequence NP_001028799.1) amino terminus fused to GAL4 activation domain.

Site-directed mutagenesis.

Full-length pO1Ca (5) or 2C-BD was used as a template in which amino acids were substituted with alanine, introduced by site-directed mutagenesis using a QuikChange XL site-directed mutagenesis kit (Stratagene, Cedar Creek, TX) performed according to the manufacturer's instructions, where the full-length plasmid was amplified by PCR, digested with DpnI to leave only the newly amplified plasmid, transformed into XL10-Gold ultracompetent cells, and grown on Terrific broth plates containing ampicillin. Positive colonies were grown for plasmid purification using a Qiagen maxiprep kit. The full-length pO1Ca was sequenced to verify that only the desired mutation was present in the plasmid. The primers were designed using the Stratagene primer mutagenesis program, which limited us to a maximum of seven amino acid changes and was the basis for deciding on the regions to be mutated. Primers were designed using the manufacturer's primer design program (https://www.genomics.agilent.com/collectionsubpage.aspx?pagetype=tool&subpagetype=toolqcpd&pageid=15).

Construction of mutant FMDV viruses.

Plasmid pO1Ca or its mutant version was linearized at the EcoRV site following the poly(A) tract and used as a template for RNA synthesis using the MegaScript T7 kit (Ambion) according to the manufacturer's protocols. BHK-21 cells were transfected with these synthetic RNAs by electroporation (Electrocell Manipulator 600; BTX, San Diego, CA) as previously described (5, 29). Briefly, 0.5 ml of BHK-21 cells at a concentration of 1.5 × 10⁷ cells/ml in PBS was mixed with 10 μg of RNA in a 4-mm-gap BTX cuvette. The cells were then pulsed once at 330 V, infinite resistance, and a capacitance of 1,000 μF; the cells were then diluted in cell growth medium and allowed to attach to a T-25 flask. After 4 h, the medium was removed, fresh medium was added, and the cultures were incubated at 37°C for up to 24 h.

The supernatants from transfected cells were passaged in LF-BK αVβ6 cells until a cytopathic effect appeared. After successive passages in these cells, virus stocks were prepared, and the viral genome was completely sequenced using a Prism 3730xl automated DNA sequencer (Applied Biosystems), as previously described (5).

RESULTS

FMDV nonstructural protein 2C is highly conserved among different serotypes.

Nonstructural FMDV 2C is a 318-amino-acid protein that is essential for virus replication (7). Comparison of the amino acid sequence of 2C from multiple serotypes of FMDV revealed a high degree of similarity, with >85% of the amino acids being identical in all reported isolates from all seven serotypes (Fig. 1). In addition, 2C is made up of 72% invariant residues, or residues that are 100% conserved among all reported isolates. These invariant residues include the proposed ATP/GTP binding domain of 2C (amino acid residues at positions 110 to 116, 160 to 163, and 243 to 246) (8). This high degree of similarity among isolates suggests that 2C has evolved to have essential functions during infection with FMDV 2C.

Fig 1.

Multiple sequence alignment using Bioedit software was performed using a representative sequence for each of serotypes of FMDV. The sequences used were from serotypes O, Asia, C, SAT1, SAT2, SAT3, and A with the respective GenBank accession numbers AJ320488.1, NP_937964, NC_002554, NC_011451, NC_003992, NC_011452, and AY593768.1.

FMDV nonstructural 2C protein interacts with the bovine host protein Beclin1.

A yeast two-hybrid system (15) was used to identify host cellular proteins that interact with FMDV 2C protein. An N-terminal fusion of the Gal4 protein DNA binding domain (BD) with FMDV 2C protein from FMDV O1 Campos (FMDV O1C) was used as “bait.” For “prey,” we used a custom cDNA library that was derived from RNA extracted from FMDV-susceptible bovine tissues (dorsal soft palate, interdigital skin, middle tongue epithelium, and middle anterior lung), expressed as N-terminal fusion of the Gal4 activation domain (AD). More than 1 × 10⁷ independent yeast colonies were screened from the library containing more than 3 × 10⁶ independent clones, representing a >3-fold screening of the library. Putative protein interactions were selected after cotransforming yeast strain AH109 with BD-2C fusion and cDNA library-AD fusion. Colonies were selected for growth in defined media lacking amino acids leucine, tryptophan, histidine, and nucleobase adenine. Plasmids were recovered from positive colonies and sequenced. In-frame ORFs were retested for specificity of the observed protein interaction with the 2C protein. To address the issue of false-positive interaction with FMDV 2C protein, human lamin C protein expressed as a fusion with BD (BD-LAM) was used as a negative control in AH109 yeast cotransfected with the positive colonies from the 2C library screen. One specific protein binding partner for 2C, Beclin1, was selected for further study due to the involvement of Beclin1 in the autophagy pathway (Fig. 2A). The plasmid that was recovered from the yeast two-hybrid screen containing Beclin1 was found to be truncated, containing amino acids 1 to 211 of bovine Beclin1 (NCBI reference sequence NP_001028799.1). However, this truncated portion of Beclin1 still contained the BH3 domain involved in binding Bcl-2 (amino acids 114 to 123) (6, 13) and part of the CCD domain known to bind to UVRAG (amino acids 144 to 269), suggesting the possibility that 2C could bind Beclin1 in a similar manner as Bcl2 or UVRAG.

Fig 2.

Protein-protein interaction of FMDV 2C with bovine Beclin1 in the yeast two-hybrid system (A), coimmunoprecipitation (B), confocal microscopy (C), and deconvolution microscopy (D). (A) Yeast strain AH109 was transformed with either GAL4-binding domain (BD) fused to FMDV 2C (2C-BD) or as a negative control human lamin C (Lam-BD). These strains were then transformed with GAL4 activation domain (AD) fused to Beclin1 (Beclin1-AD) or T antigen (Tag-AD) as indicated above each lane. Spots of strains expressing the indicated constructs containing 2 × 10⁶ yeast cells were spotted on selective media for protein-protein interaction in the yeast two-hybrid system, either SD−Ade/His/Leu/Trp plates (−ALTH) or and nonselective SD−Leu/Trp (−TL) for plasmid maintenance only. (B) Western blot probing for FMDV 2C (band is ∼40 kDa). Input cell lysate was from mock-infected (lane 1) or FMDV-infected (lane 2) cell lysates. Coimmunoprecipitation of FMDV 2C from mock-infected (lane 3) or FMDV O1C-infected (lane 4) cell lysates was performed using an antibody specific to Beclin1. (C and D) Analysis of the distribution of Beclin1 and FMDV 2C protein in MCF 10A cells. Cells were infected with FMDV O1C and processed by immunofluorescence staining as described in Materials and Methods. FMDV 2C was detected with mouse MAb (3D10) and visualized with Alexa Fluor 594 (red). Beclin1 was detected with monoclonal antibody (H-300) and visualized with Alexa Fluor 488 (green). Yellow indicates colocalization of the Alexa Flour 594 and 288 in the merged image. (E) Analysis of Beclin1 expression in MCF-10A cells. Cells were infected with FMDV O1C and processed for Western blotting at the indicated time points. Beclin1 was detected using MAb (H-300).

To confirm that the interaction identified using the two-hybrid system in yeast occurs during FMDV infection of host cells, coimmunoprecipitation experiments were performed using MAbs specifically recognizing both proteins. Human epithelial cell line MCF-10A was infected (MOI = 10) with FMDV O1C, and samples were harvested at 2 h postinfection (hpi), the time point when 2C is beginning to accumulate in MCF-10A cells as determined by Western blotting (data not shown). MCF-10A cell lysates were collected from infected or mock-infected cells and immunoprecipitated with an anti-Beclin1 MAb, H-300 (Santa Cruz Biotechnology, Santa Cruz, CA), followed by a Western blot with an FMDV 2C-specific MAb, 3D10 (Izler, Brescia, Italy). A single band at the expected molecular mass of FMDV 2C (40 kDa) was clearly obtained, indicating that during FMDV infection, 2C coimmunoprecipitates with Beclin1, confirming the previous yeast two-hybrid results, suggesting a 2C-Beclin1 interaction (Fig. 2B). Attempts to perform reverse coimmunoprecipitation, immunoprecipitating 2C and detecting Beclin1 on the blot, were inconclusive since the presence Beclin1 was masked by the immunoglobulin heavy chain.

The localization of 2C and Beclin1 during infection was assessed using double-label immunofluorescence and confocal microscopy in cells infected with FMDV. MCF-10A cells were infected (MOI = 10) or mock infected with FMDV O1C. The cells were fixed on glass coverslips at 30-min intervals for up to 4 h after infection and stained with MAbs that exhibit specific fluorescence for FMDV 2C (3D10) or Beclin1 (H-300). The results indicated that a clear colocalization of FMDV 2C and Beclin1 proteins occurred at between 2 and 2.5 hpi, with both of the proteins displaying a small punctuated distribution pattern (Fig. 2C), supporting the hypothesis that the interaction between these two proteins occurs during viral infection. However, at other time points examined, no clear colocalization occurred, suggesting that the FMDV 2C and Beclin1 protein interaction occurs only during the early phase of infection.

Overexpression of the autophagy protein Beclin1 results in decreased FMDV replication.

To understand how Beclin1 may affect FMDV replication, we measured fluctuations in the intracellular levels of Beclin1 and changes in virus yield from cells infected with FMDV. MCF-10A cells were infected (MOI = 10) with FMDV O1C, and cell lysates were collected every 30 min during the course of infection from 1 to 4 hpi. Samples were tested by Western blotting to assess the levels of Beclin1 present in the cell lysate. The results demonstrated that levels of Beclin1 remain unaltered during the course of infection with FMDV (Fig. 2D), ruling out the possibility that 2C may promote the degradation of Beclin1.

In order to assess the role of Beclin1 in FMDV replication, we attempted to manipulate the levels of Beclin1 in infected cells. Specific siRNA targeting Beclin1 was transfected into MCF-10A cells infected (MOI = 0.01) with FMDV O1C. Beclin1 expression decreased >80% in the treated cells, as determined by Western blotting (Fig. 3A) and as quantified using ImageJ software (Fig. 3B) (obtained from the National Center for Biotechnology Information) using the recommended procedure to calculate relative density compared to a control treated sample (siGlow). The decrease of Beclin1 expression in the Beclin1-specific siRNA-treated cells did not affect viral yields compared to siGlow (as a control)-treated cells (Fig. 3C). This result may indicate that decreased translation of Beclin1 does not affect virus replication or, alternatively, that remaining low levels of Beclin1 are still enough to allow FMDV replication.

Fig 3.

Effect of Beclin1-specific siRNA treatment on FMDV replication. MCF-10A cells were transfected with four RNA duplexes targeted Beclin1 or with a control siRNA, siGlo, for 48 h at 37°C as described in Materials and Methods. (A) Whole-cell lysates were collected 48 h after siRNA treatment and analyzed by Western blotting with a Beclin1-specific antibody (band is ∼65 kDa). (B) Quantification of Beclin1 bands in Western blot showed in panel A using ImageJ analysis software. (C) Viral yields from FMDV infections in MCF-10A cells treated with siRNA to reduce the intracellular concentration of Beclin1. After transfection, triplicate plates were infected with FMDV O1C (MOI = 1). Titers were determined in BHK-21 cells and expressed as the log₁₀ TCID₅₀/ml.

The effect of overexpression of Beclin1 on FMDV replication was also assessed. MCF-10A cells were transfected with a plasmid utilizing a cytomegalovirus (CMV) promoter that overexpresses Beclin1-FLAG (pBeclin1-Flag; Addgene, plasmid no. 24388) or GFP (control plasmid) (vitality hrGFP; Agilent, catalog no. 240145). Overexpression of Beclin1 in transfected cells was assessed at 19 h posttransfection by Western blotting (Fig. 4A). Furthermore, MCF-10A cells overexpressing Beclin1-FLAG or GFP were infected (MOI = 0.1) with FMDV O1C, and virus present in the supernatant was measured hourly between 0 and 5 hpi and at 24 hpi. A considerable reduction (∼2 logs) in virus titer was observed in cells overexpressing Beclin1-FLAG compared to cells overexpressing GFP and containing endogenous levels of Beclin1 (Fig. 4B). In addition, no differences were found when virus titers at the intracellular and extracellular compartment were compared (Fig. 4C). These results indicate that overexpression of Beclin1 causes a decrease in virus yield.

Fig 4.

Viral yields from FMDV infections in MCF-10A cells overexpressing Beclin1-Flag. MCF-10A cells were transfected either with a plasmid encoding Beclin1-Flag (pBeclin-Flag) or GFP (pGFP), as a control as described in Materials and Methods. (A) Western blot showing endogenous intracellular levels of Beclin1, as well as overexpressed levels of Beclin1-Flag in MCF-10A cells transfected either with pBeclin-Flag or pGFP. As a loading control, the detection of the levels of intracellular GADPH (40 kDa) was performed. (B) After transfection, triplicate plates were infected with FMDV O1C (MOI = 0.1). Titers were determined in BHK-21 cells and are expressed as the log₁₀ TCID₅₀/ml. (C) Triplicate plates were infected with FMDV O1C (MOI = 0.1). Extracellular and intracellular samples were taken at 24 hpi. Titers were determined in BHK-21 cells and are expressed as the log₁₀ TCID₅₀/ml.

Effect of Beclin1 overexpression on viral VP1 and cellular UVRAG.

UVRAG (UV radiation resistance-associated gene) binds the coiled-coil region of Beclin1 getting incorporated into autophagosomes and is later involved in autophagosome maturation to autophagolysosomes (12, 28), a process known to initiate the degradation of the material inside the autophagosome. It is possible that FMDV 2C prevents the incorporation of UVRAG into autophagosomes, which would later prevent UVRAG-mediated fusion of autophagosomes to lysosomes, thus preventing viral protein degradation. To examine this hypothesis, the colocalization of FMDV structural protein VP1 with cellular UVRAG was tested using FMDV-infected (MOI = 10) MCF-10A cells. No colocalization of VP1 and UVRAG was observed (Fig. 5A) at 4 hpi. However, when the same experiment was performed in cells overexpressing Beclin1, autophagosomes containing FMDV VP1 were seen to localize with UVRAG. MCF-10A cells were transfected with pBeclin1-FLAG and 19 h later infected (MOI = 10) with FMDV O1C. Four hours later, the colocalization of UVRAG and VP1 was clearly observed (Fig. 5B). Thus, the overexpression of Beclin1 would promote the incorporation of UVRAG into autophagosomes containing viral proteins, facilitating the process of autophagosome-lysosome fusion and provoking the degradation of virus particles.

Fig 5.

Analysis of distribution of FMDV VP1 and UVRAG in FMDV-infected MCF-10A cells. Mock-transfected cells (A) or cells transfected with Beclin1-Flag (B) were infected (MOI = 10) 24 h posttransfection with FMDV O1C. These preparations were processed for immunofluorescence staining as described in Materials and Methods, along with uninfected control cells (C). UVRAG was detected with rabbit polyclonal antiserum (Sigma-Aldrich, catalog no. U7508) and visualized with Alexa Fluor 488 (green). FMDV VP1 was detected with a MAb (10GA) and visualized with Alexa Flour 594 (red). Yellow indicates colocalization of the Alexa Flour 594 and 288 in the merged image.

To further examine that overexpression of Beclin1 associates with the fusion of autophagosomes that contain viral proteins with lysosomes, we examined the localization of FMDV structural protein VP1 with cellular lysosome-associated membrane protein 1 (LAMP1), a late lysosome marker. VP1-LAMP1 colocalization was first tested in MCF-10A cells infected with FMDV (MOI = 10). No colocalization of VP1 and LAMP1 was observed in these cells at 4 hpi (Fig. 6B). However, when the same experiment was performed in cells overexpressing Beclin1, FMDV VP1 was seen to localize with LAMP1 (Fig. 6A). MCF-10A cells were transfected with pBeclin1-FLAG and 19 h later infected (MOI = 10) with FMDV O1C. Four hours later, the colocalization of LAMP1 and VP1 was clearly observed (Fig. 6A). Thus, overexpression of Beclin1 associates with the fusion of autophagosomes containing viral proteins, with lysosomes perhaps causing virus degradation and the concomitant reduction in viral yield.

Fig 6.

Analysis of distribution of FMDV VP1 and Lamp1 in FMDV-infected MCF-10A cells. (A) Cells transfected with Beclin1-Flag or (B) Mock transfected cells were infected (MOI = 10) 24 h posttransfection with FMDV O1C. These preparations were processed for immunofluorescence staining as described in Materials and Methods, along with (C) uninfected control cells. Lamp1 was detected with mouse MAb (BD Biosciences 555798) and visualized with Alexa Fluor 488 (green). FMDV VP1 was detected with a MAb (10GA) and visualized with Alexa Flour 594 (red). Yellow indicates colocalization of the Alexa Flour 594 and 288 in the merged image.

To provide additional confirmation that autophagosome fusion to lysosomes is not required during FMDV infection, we used bafilomycin A1, a specific inhibitor of vacuolar H⁺-ATPase, which blocks UVRAG-induced autophagosome fusion to lysosomes (28). To determine the effect of bafilomycin A1 on FMDV replication, MCF-10A cells were pretreated for 30 min at 37°C with 0.5 μM bafilomycin A1. The medium was removed, and the cells were infected with FMDV O1C (MOI = 1) for 1 h, acid washed, and then incubated with or without bafilomycin A1. Samples were taken at 1 and 5 hpi. The results (Fig. 7) showed that treatment with bafilomycin A1 had no effect on virus replication, indicating that autophagosome fusion to lysosomes is not required for FMDV replication.

Fig 7.

Effect of bafilomycin A1 on FMDV replication in MCF-10A cells. Cells were pretreated with bafilomycin A1 at a concentration of 0.5 μM for 30 min and then infected with FMDV O1Ca (MOI = 10) for 1 h at 37°C. The cells were then treated or mock treated with bafilomycin A1. Samples were taken at 1 h and at 5 and 24 h after infection. Titers were determined in BHK-21 cells and are expressed as the log₁₀ TCID₅₀/ml.

Overexpression of the autophagy protein Bcl-2 blocks FMDV replication.

Beclin1-dependent autophagy is known to be partially regulated by endogenous Bcl-2. Bcl-2 binds Beclin1, tethering it to the endoplasmic reticulum (9, 37). Overexpression of Bcl-2 inhibits the dissociation of Beclin1 from the endoplasmic reticulum, preventing the progression of Beclin1-induced autophagy (37). To determine the affect of overexpression of Bcl-2 on FMDV replication, Bcl-2 was overexpressed using a plasmid encoding GFP-Bcl-2 (pGFP-Bcl-2) under the CMV promoter (47); a plasmid encoding GFP alone, phrGFP II-N (Stratagene, catalog no. 240145), was used as a negative control. MCF-10A cells were transfected with either plasmid and, 24 h later, a high rate of GFP-Bcl-2 or GFP expression was observed by immunofluorescence (Fig. 8A). Transfected cells were then infected (MOI = 0.1) with FMDV O1C, and the virus yield assessed at 0, 5, and 24 hpi. A decrease (∼1.5 to 2 log₁₀) in virus titers in cell cultures overexpressing Bcl-2 (Fig. 8B) was measured.

Fig 8.

Viral yields in FMDV-infected MCF-10A cells overexpressing Bcl-2. MCF-10A cells were mock transfected or transfected with a plasmid encoding Bcl-2-GFP (pBcl-2-GFP) or GFP (pGFP) as a control. (A) Detection of Bcl-2 overexpression in Bcl-2-GFP transfected (left panel) or mock transfected (right panel) MCF-10 cells by fluorescence microscopy. (B) After transfection, triplicate plates were infected (MOI = 0.1 TCID₅₀/cell) with FMDV O1C. Titers were determined in BHK-21 cells and expressed as the log₁₀ TCID₅₀/ml for the extracellular virus. (C) Analysis of distribution of Beclin1 and PDI in GFP-Bcl-2-expressing MCF10A cells. Cells were processed for immunofluorescence staining as described in Materials and Methods. Bcl-2 was visualized by GFP (Green). Beclin1 was detected with a MAb (H-300) and visualized with Alexa Flour 647 (Light blue). PDI was detected with an MAb (Affinity Bioreagents MA3-018) and visualized with Alexa Flour 488 (red).

Evidence supporting that Beclin1 is tethered to the endoplasmic reticulum in MCA-10A cells expressing Bcl-2GFP was obtained by demonstrating the colocalization of Bcl-2, Beclin1, and PDI, a marker for the endoplasmic reticulum (Fig. 8C). Therefore, tethering of Beclin1 to the endoplasmic reticulum, which could prevent the progression of Beclin1-induced autophagy, may be the reason that the overexpression of Bcl-2 is detrimental virus replication.

To examine whether the overexpression of Bcl-2 may impede viral entry into the autophagy pathway, we assessed the colocalization of FMDV VP1 and cellular LC3 (a microtubule-associated protein and well-known autophagosome marker) in infected cells overexpressing Bcl-2. MCF-10A cells transfected with GFP-Bcl-2 and 19 h later infected (MOI = 10) with FMDV O1C exhibited colocalization of FMDV VP1 and LC3 at 4 hpi (Fig. 9). This result suggests that a decreased FMDV yield in cells overexpressing Bcl-2 must be mediated by an unknown mechanism not affecting the entry of FMDV into the autophagy pathway.

Fig 9.

Analysis of distribution of LC3 and FMDV VP1 in GFP-Bcl-2-expressing (A) and mock-transfected (B) MCF10A cells infected with FMDC O1C (MOI = 10) 24 h posttransfection with FMDV O1C. These preparations were processed for immunofluorescence staining as described in Materials and Methods, along with uninfected control cells (C). Bcl-2 was visualized by using GFP (green). FMDV VP1 was detected with a MAb (10GA) and visualized with Alexa Flour 488 (red). LC3 was detected with rabbit polyclonal antiserum (catalog no. 53341; Anaspec, Fremont, CA) and visualized with Alexa Fluor 647 (light blue). Pink indicates colocalization of the Alexa Flour 488 and Alexa Flour 647 in the merged image.

To determine whether overexpression of Bcl-2 had any effect on autophagosome-lysosome fusion (as the overexpression of Beclin1 has), the colocalization of FMDV VP1 and UVRAG in infected cells overexpressing Bcl-2 was assessed. The same cell preparations used to determine FMDV VP1 and LC3 colocalization were used to analyze colocalization between FMDV VP1 and UVRAG. No colocalization was observed between these two proteins (data not shown) in FMDV-infected cells overexpressing Bcl-2, suggesting that the mechanism of viral inhibition in cells overexpressing Bcl-2 should be different than that triggered in cells overexpressing Beclin1.

Identification of the Beclin1 binding site on FMDV 2C.

To determine the binding site(s) for Beclin1 present in 2C, an alanine scanning mutagenesis approach was used. We used site-directed mutagenesis to construct a set of 46 mutant 2C proteins containing sequential stretches of seven amino acids where the native amino acid residues were substituted by alanine residues (Fig. 10A). These mutated 2C proteins were assessed for their ability to bind Beclin1 utilizing the yeast two-hybrid system. 2C proteins containing mutations in areas 15, 16, 17, 19, 20, 29, 33 to 40, and 42 were unable to bind Beclin1 (Fig. 10A). To ensure that all 2C alanine mutants were still able to be expressed in the yeast two-hybrid system, protein MCM7-AD (another bovine host protein that was detected as a binding partner for 2C) was used as an internal control. MCM7-AD was able to interact with all 2C mutants lacking Beclin1 binding (Fig. 10C), thereby demonstrating that mutating these areas specifically interrupted the binding between Beclin1 and 2C.

Fig 10.

Scheme showing FMDV 2C alanine mutants used in the present study. (A) Each alanine 2C name mutant is followed, in parentheses, by the amino acid residues mutated for that mutant. All indicated residues were mutated to an alanine. The highlighted 2C mutants represent mutations that resulted in lack of binding of 2C to Beclin1 in the yeast two-hybrid system. (B) Subdivisions (labeled A, B, and C) of the original alanine scanning mutants 16, 17, and 19. Highlighted mutants were found not to bind Beclin1 in the yeast two-hybrid system. (C) Yeast strain AH109 was transformed with either GAL4-binding domain (BD) fused to FMDV 2C (2C-BD), the indicated FMDV 2C mutation, or as a negative control human lamin C (Lam-BD). These strains were then transformed with GAL4 activation domain (AD) fused to Beclin1 (Beclin1-AD) or MCM7 (MCM7-AD) as indicated above. Strains expressing the indicated constructs containing 2 × 10⁶ yeast cells were spotted onto selective media to evaluate protein-protein interaction in the yeast two-hybrid system, using either SD−Ade/His/Leu/Trp plates (−ALTH) or else nonselective SD−Leu/Trp (−TL) for plasmid maintenance only.

Reverse genetics were used to assess the effect of 2C mutations identified as critical in mediating the interaction between 2C and Beclin1. Infectious clones of FMDV O1C (5) contained areas of 2C harboring the same alanine substitutions for 2C-15, 2C-16, 2C-17, 2C-19, 2C-20, and 2C-36 selected from areas shown to alter 2C-Beclin1 reactivity in the yeast two-hybrid model that were sequentially introduced into the infectious clone construct. These infectious clone constructs were then used to produce the corresponding RNAs by in vitro transcription, which were then used in cell transfections to produce their respective FMDV progeny. Although transfection with parental FMDV O1C RNA produced viable virus progeny, mutations in 2C-15, 2C-16, 2C-17, 2C-19, 2C-20, and 2C-36 consistently resulted in nonviable virus. A more detailed mapping of the area of interaction between Beclin1 and 2C was performed to exclude the possibility of detrimental effects from unidentified causes by reducing the number of amino acids mutated while still disrupting the Beclin1-2C interaction. Regions in mutants 2C-16, 2C-17, and 2C-19 were further subdivided into three separate subareas that were individually assessed for their reactivities with Beclin1 in the yeast two-hybrid system. This procedure resulted in several smaller areas (comprising two to three residues each) responsible for the Beclin1-2C interaction: 16A, 16B, 17A, 17C, and 19A, while maintaining the MCM7-2C interaction (Fig. 10B). Interestingly, none of the infectious clones harboring these mutated areas was able to produce viable progeny, suggesting that amino acid residues critical to FMDV 2C-Beclin1 interaction are also critical for virus replication. These results suggest that the interaction between 2C and Beclin1 is essential to the process of virus replication. However, we cannot rule out the possibility that these mutations in 2C have other effects by changing the structure of 2C or by inhibiting 2C to interact with other proteins.

DISCUSSION

Viruses have developed complex mechanisms to manipulate normal cellular pathways to facilitate replication and to evade host defense mechanisms. To do so, viruses often interact with cellular proteins to modify their function, thus modifying natural cellular pathways. We report here that FMDV nonstructural protein 2C binds to a central regulator of autophagy pathway, Beclin1. Beclin1 plays dual roles in the autophagy pathway: involvement at the initiation step of autophagosome formation, and later, at autophagosome fusion to the lysosome. Several studies have suggested that the autophagy pathway can function as an antiviral pathway by degrading viruses or as a proviral pathway, helping viruses replicate or exit the cell (17, 32, 40). In this report, a yeast two-hybrid model was used to show that Beclin1 is a specific protein binding partner for viral 2C. In concordance with this result, we also demonstrated the occurrence of coimmunoprecipitation of 2C with Beclin1 and colocalization of 2C with Beclin1 in cells infected with FMDV.

Manipulation of the autophagy pathway has been shown to occur by a wide range of viruses (17) and, interestingly, all of them produce their effect by binding to Beclin1. Three proteins in different herpesviruses have been shown to inhibit the formation of autophagosomes: herpes simplex virus 1 (HSV-1) protein ICP35.5 (26), Kaposi's sarcoma herpesvirus protein orf16 (Bcl-2 homologue) (45), and murine gamma herpesvirus 68 protein M11 (Bcl-2 homologue) (23, 41). All three proteins antagonize Beclin1 to prevent its incorporation into autophagy complexes, possibly by anchoring Beclin1 to the endoplasmic reticulum, resulting in the suppression of the initial steps of autophagy. As an example, in HSV-1, deletion of the binding domain in ICP35.5 that binds Beclin1 causes reduced neurovirulence in mouse models. The neurovirulence in mouse models can be regained by using knockout mice with a defect in the autophagy pathway (26). RNA viruses utilize the cellular autophagy pathway differently than DNA viruses, using it to benefit their replication strategy by stabilizing autophagosomes and preventing autophagosome-lysosome fusion. Infection by poliovirus, HCV, or FMDV results in an accumulation of autophagosomes, benefiting viral replication through mechanisms not well understood (1, 20, 35). In the case of influenza virus A, autophagosomes are stabilized by protein M2, which binds Beclin1 to prevent the fusion of autophagosomes to lysosomes (16). In a similar manner, human immunodeficiency virus (HIV) stabilizes autophagosomes by binding of viral protein nef to Beclin1 (24). Blocking autophagosome fusion to lysosomes prevents RNA virus degradation by the lysosomes (17).

In a previous study we demonstrated that FMDV triggers the autophagy machinery, enhancing viral replication (35). Furthermore, chemical stimulation or inhibition of the autophagy process directly correlated with an increase or decrease in virus production. To understand how Beclin1 plays a role during FMDV replication, we analyzed the effect of overexpression of Beclin1 (Fig. 4B). The overexpression of Beclin1 resulted in a severe decrease in viral yields. It is possible that Beclin1overexpression favors UVRAG-mediated incorporation of viral proteins into autophagosomes and later the fusion of these phagosomes to lysosomes, as evidenced by the sequential colocalization of viral VP1 with UVRAG (Fig. 5B) and LAMP1 (Fig. 6A). Both events are absent during the normal process of FMDV replication (Fig. 5 and 6B). It is possible that during viral infection 2C binding to Beclin1 would prevent autophagosome incorporation of UVRAG by blocking the UVRAG binding site in Beclin1, thereby inhibiting UVRAG-mediated autophagosome maturation. These results also suggest that FMDV 2C may function in a similar form as HIV nef (24) or influenza virus A M2 (16), by binding to Beclin1 to help the virus escape the autophagy-induced protein degradation pathway.

To determine whether autophagosome maturation is necessary for FMDV replication, we tested the effect of bafilomycin A1, an inhibitor of autophagosome-lysosome fusion, on FMDV replication. Our results clearly support the fact that autophagosome-lysosome fusion is not required for FMDV replication (Fig. 7) and determined that there was no difference in viral yield, suggesting that fusion with lysosomes is not a required process for FMDV replication.

It is well known that Beclin1 is partially regulated by Bcl-2 (9, 37). Therefore, we evaluated the effect of overexpression of Bcl-2 in MCF-10A cells infected with FMDV. As shown in Fig. 8B, the overexpression of Bcl-2 is detrimental for virus replication. Interestingly, these cells still showed colocalization of 2C viral protein with LC3 (Fig. 9), an autophagy hallmark, but not with UVRAG. This result suggests the virus was still able to enter the autophagy pathway (perhaps in a Beclin1-independent way) even though the autophagosome-lysosome fusion did not occur. These observations suggest that overexpressing Bcl-2 should trigger a mechanism for disrupting FMDV replication mediated by a different pathway than that induced by overexpression of Beclin1. The difference in the inhibitory mechanisms mediated by the overexpression of Beclin1 and Bcl-2 suggests the possible existence of multiple mechanisms allowing FMDV to interact with the autophagy process in favor of viral replication.

Interestingly, Beclin1 as it was identified in the yeast two-hybrid screen contained only amino acids 1 to 211 of bovine Beclin1 (NCBI reference sequence NP_001028799.1), indicating that the Beclin1 protein recovered in the yeast two-hybrid experiment was a truncated version containing areas responsible for binding Bcl-2 and UVRAG: the BH3 domain (amino acids 114 to 123) (6, 13) and part of the CCD domain (amino acids 144 to 269) (27). Therefore, it is possible that FMDV 2C competes with UVRAG or Bcl-2 for binding to Beclin1 and that, to prevent autophagosome-lysosome fusion and subsequent virus destruction, FMDV 2C blocks the UVRAG binding site in Beclin1. However, additional studies mapping the binding domain in Beclin1 would be needed to determine whether this is true.

Although it appears that the 2C-Beclin1 interaction is necessary for virus replication and that the autophagy pathway is used for replication by FMDV, the precise mechanism still needs to be elucidated for how FMDV enters the autophagy pathway and manipulates the autophagosome to facilitate its replication and why in previous studies the chemical induction of autophagosomes increased viral yield (35). Previous studies have shown that FMDV enters the cell via endosomes (33); however, there was no colocalization between MPR (mannose 6-phosphate receptor), an endosomal marker, and FMDV proteins (34), suggesting that the mechanism of FMDV entering the autophagy pathway is not due to endosome fusion to the autophagosome, as was shown with dengue virus (36). Therefore, FMDV could enter the autophagy pathway by different means that are independent of both endosome fusion to the autophagosome and of Beclin1-induced autophagy. Further studies are required to explore this possibility.

The results reported here identify, for the first time, a cellular host protein, Beclin1 that interacts with a viral protein of FMDV, 2C. This interaction appears to be critical for virus growth since mutated FMDV genomes harboring the 2C mutations that disrupted the interaction between 2C and Beclin1 resulted in viruses completely unable to replicate in cell cultures (Fig. 10). Importantly, 2C-Beclin1 interaction appears to modulate the autophagy pathway by preventing virus destruction in the final steps of autophagy. This presents new possibilities of exploration for FMDV pathogenesis, and further work still needs to be done to explore other cellular proteins that may play a role during FMDV infection. Further understanding host protein-viral protein relationships will encourage the design of novel therapeutic strategies that disrupt viral-host interactions.