Is the T cell is Turned Into a Duplication Mechanism That Continues to Make Copies of the Virus
J Virol. 2014 Oct; 88(19): 11576–11585.
Duplication of the A17L Locus of Vaccinia Virus Provides an Alternate Route to Rifampin Resistance
Karl J. Erlandson
aLaboratory of Viral Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, USA
Catherine A. Cotter
aLaboratory of Viral Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, USA
James C. Charity
aLaboratory of Viral Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, USA
Craig Martens
bResearch Technologies Section, Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Hamilton, Montana, USA
Elizabeth R. Fischer
bResearch Technologies Section, Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Hamilton, Montana, USA
Stacy M. Ricklefs
bResearch Technologies Section, Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Hamilton, Montana, USA
Stephen F. Porcella
bResearch Technologies Section, Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Hamilton, Montana, USA
Bernard Moss
aLaboratory of Viral Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, USA
G. McFadden, Editor
Received 2014 Mar 7; Accepted 2014 Jul 19.
ABSTRACT
Specific gene duplications can enable double-stranded DNA viruses to adapt rapidly to environmental pressures despite the low mutation rate of their high-fidelity DNA polymerases. We report on the rapid positive selection of a novel vaccinia virus genomic duplication mutant in the presence of the assembly inhibitor rifampin. Until now, all known rifampin-resistant vaccinia virus isolates have contained missense mutations in the D13L gene, which encodes a capsid-like scaffold protein required for stabilizing membrane curvature during the early stage of virion assembly. Here we describe a second pathway to rifampin resistance involving A17, a membrane protein that binds and anchors D13 to the immature virion. After one round of selection, a rifampin-resistant virus that contained a genomic duplication in the A17L-A21L region was recovered. The mutant had both C-terminally truncated and full-length A17L open reading frames. Expression of the truncated A17 protein was retained when the virus was passaged in the presence of rifampin but was lost in the absence of the drug, suggesting that the duplication decreased general fitness. Both forms of A17 were bound to the virion membrane and associated with D13. Moreover, insertion of an additional truncated or inducible full-length A17L open reading frame into the genome of the wild-type virus was sufficient to confer rifampin resistance. In summary, this report contains the first evidence of an alternate mechanism for resistance of poxviruses to rifampin, indicates a direct relationship between A17 levels and the resistance phenotype, and provides further evidence of the ability of double-stranded DNA viruses to acquire drug resistance through gene duplication.
IMPORTANCE The present study provides the first evidence of a new mechanism of resistance of a poxvirus to the antiviral drug rifampin. In addition, it affirms the importance of the interaction between the D13 scaffold protein and the A17 membrane protein for assembly of virus particles. Resistance to rifampin was linked to a partial duplication of the gene encoding the A17 protein, similar to the resistance to hydroxyurea enabled by duplication of the gene encoding the small subunit of ribonucleotide reductase and of the K3L gene to allow adaptation to the antiviral action of protein kinase R. Gene duplication may provide a way for poxviruses and other DNA viruses with high-fidelity DNA polymerases to adjust rapidly to changes in the environment.
INTRODUCTION
Poxviruses are large, enveloped, double-stranded DNA viruses that replicate within the cytoplasm (1). Vaccinia virus (VACV), the model poxvirus, is being employed to construct recombinant vaccines, study antiviral mechanisms, and investigate the basic biology of poxviruses, including assembly. Assembly of VACV begins as viral membranes form and curve into a crescent shape, which is stabilized by a honeycomb lattice of capsid-like D13 protein trimers (2,–4, 40). The crescents enlarge to form spheres, termed immature virions (IVs) that, upon processing of the A17 membrane protein and removal of D13, condense into brick-shaped mature virions (MVs) (5). Recombinant viruses with inducible expression of the D13 or A17 protein were used to demonstrate that both are essential for crescent formation (6,–8).
A17 is an integral membrane protein that undergoes posttranslational modifications, including phosphorylation and proteolytic cleavages at the N and C termini (9,–13). The N-terminal region of A17 anchors D13 to the viral membrane during crescent formation, and cleavage of the N terminus of A17 by the I7 protease correlates with the subsequent release of D13 and the transition of IVs to MVs (5, 12, 14). The structure of A17 has yet to be determined, and the topology is uncertain (14,–17).
The antibiotic rifampin (also known as rifampicin) is a well-characterized inhibitor of bacterial DNA-dependent RNA polymerases but has an entirely different antiviral action and has been used as a tool to study the initial stages of VACV morphogenesis (18,–20). In the presence of rifampin, VACV assembly is inhibited prior to the crescent stage with accumulation of irregular-shaped membranes lacking spicules, now known to represent the D13 scaffold. The irregular membranes surround electron-dense regions containing core proteins or their precursors. Under these conditions, D13 accumulates in separate inclusion bodies (21). Rifampin has a narrow pharmacological window of ∼75 to 200 μg/ml for inhibition of VACV replication (20, 22). In addition, the inhibitory effect of rifampin is readily reversible, as IVs and regular-shaped crescents can be seen within 5 min after drug removal, even if protein synthesis inhibitors are added (18). Once formed, however, crescents and IVs are unaffected by the addition of the drug, indicating that rifampin acts before assembly. To date, all known rifampin-resistant VACV isolates encode point mutations that predict an amino acid change within the D13L protein (23,–26). The number of such mutations was greatly increased by random mutagenesis of the D13L gene; a total of 32 mutations of 24 different amino acids were correlated with rifampin resistance (26). Although the mutations are widely dispersed, the majority cluster in three contiguous regions that map to the likely membrane-proximal region of D13 (4, 40).
The strong link between D13 mutations and rifampin resistance and the failure to discover other gene targets suggest that D13 interacts directly with rifampin, although this has not been demonstrated. In the present study, we took an unbiased approach to uncover additional rifampin resistance-producing mutations. Screening for rifampin-resistant virus was carried out after a single passage of unmutagenized virus in the presence of the drug. As expected, the majority of the rifampin-resistant virus isolates had a mutation in the D13 open reading frame (ORF). However, one rifampin-resistant virus had a partial duplication of the A17-to-A21 locus and no mutations within D13, providing the first evidence of an alternate mechanism for the resistance of poxviruses to rifampin. Further studies demonstrated that duplication of A17 was sufficient to confer a rifampin resistance phenotype. Duplication of other VACV or recombinant genes has been shown to provide resistance to hydroxyurea (27) and allow adaptation to the antiviral action of protein kinase R (28, 29). Thus, gene duplication may be a general mechanism for poxviruses and other viruses to overcome environmental stress.
MATERIALS AND METHODS
Materials.
Lipofectamine 2000, pcDNA3.1D/V5-His-TOPO, SeeBlue Plus2 protein marker, iBlot membranes, and Phusion polymerase were purchased from Invitrogen (Carlsbad, CA). Isopropyl-β-d-thiogalactoside (IPTG) and cOmplete mini protease tablets were from Roche Diagnostics (Indianapolis, IN), and the In-Fusion HD cloning kit was from Clontech Laboratories (Mountain View, CA). Rifampin, sodium dodecyl sulfate (SDS), and NP-40 were purchased from Sigma-Aldrich (St. Louis, MO). LI-COR (Lincoln, NE) secondary reagents were used for Western blot assays. Oligonucleotides were purchased from Integrated DNA Technologies (Coralville, IA).
Cells and viruses.
BSC-1 cells were maintained in minimum essential medium with Earle's balanced salts containing 8% fetal bovine serum, 2 mM l-glutamine, 100 U of penicillin and 100 μg of streptomycin per ml (Quality Biologicals, Gaithersburg, MD). Wild-type (WT) VACV and recombinant versions including vTF7-3 (30) and vRB12 (31) were derived from the Western Reserve (WR) strain. RifR9 and RifR10 were selected as spontaneous rifampin-resistant mutants in the presence of 100 μg/ml of the drug. vRB12-A17, vRB12-A17T, and vRB12-rev were produced by introducing genes under the control of the A17 natural promoter into pRB21, recombining it into vRB12, and picking large plaques as described previously (31). Recombinant viruses were plaque purified, and the modified regions were sequenced. VACV MVs were purified by sedimentation through a 36% sucrose cushion, followed by banding in a 25 to 40% sucrose density gradient (32). The recombinant viruses used and the corresponding references are listed in Table 1.
TABLE 1
Viruses used in this study
| Virus | Mutation | A17 Promoter | Reference |
|---|---|---|---|
| vTF7-3 | Inserted T7 RNA polymerase gene | NA a | 31 |
| RifR9 | Duplicated A17-21 region | Natural | This study |
| RifR10 | D13 I485L | NA | This study |
| vRB12 | Partial deletion of F13 | NA | 32 |
| vRB12-A17 | Addition of A17 and restoration of F13 | Natural | This study |
| vRB12-A17T | Addition of A17T and restoration of F13 | Natural | This study |
| vRB12-rev | Reversion with restoration of F13 | NA | This study |
| vT7lacOI-A17L | 2nd A17 copy in HA locus | Natural and T7 | 8 |
Construction of plasmids.
For infection/transfection experiments, A17, A17T, and D13iV5 were inserted into the pcDNA3.1D V5-His-TOPO vector under the control of the T7 promoter. The V5 tag was inserted into D13 at amino acid position 302 of D13 by using the In-Fusion HD cloning kit with forward primer 5′-TAACCCTCTCCTCGGTCTCGATTCTACGGATGGTATCGTTTCTATTCAAGATG-3′ and reverse primer 5′-CCGAGGAGAGGGTTAGGGATAGGCTTACCACCATCGCTAACAATAACTAGATC-3′. The plasmids used and the corresponding references are listed in Table 2.
TABLE 2
Plasmids used in this study
| Plasmid | Vector | Protein expressed | Promoter | Reference |
|---|---|---|---|---|
| pRB12-A17 | pRB21 | WT A17 | A17 | This study |
| pRB12-A17T | pRB21 | A17T | A17 | This study |
| pCDNA3.1-A17 | pCDNA3.1D V5-His-TOPO | WT A17 | T7 | This study |
| pCDNA3.1-A17T | pCDNA3.1D V5-His-TOPO | A17T | T7 | This study |
| pCDNA3.1-D13iV5 | pCDNA3.1D V5-His-TOPO | D13iV5 | T7 | This study |
Plaque assay and determination of virus yields.
BSC-1 cell monolayers in 6- or 12-well plates were used for plaque assays. Tenfold dilutions of virus were allowed to adsorb to cells for 1 h at room temperature and aspirated, and the cells were overlaid with medium containing 0.5% methylcellulose. After 24 or 48 h, the cells were stained with crystal violet and plaques were counted.
Western blot analysis.
Whole-cell lysates or antibody-bound proteins were resuspended in lithium dodecyl sulfate sample buffer with reducing agent and Benzonase before electrophoresis on 4 to 12% Novex NuPAGE acrylamide gels in 2-(N-morpholino)ethanesulfonic acid buffer (Invitrogen). Proteins were transferred to nitrocellulose membranes with the iBlot system (Invitrogen). The membranes were blocked with 5% skim milk in Tris-buffered saline containing Tween 20 and incubated with primary antibodies overnight. LI-COR secondary antibodies and the LI-COR system were used to detect proteins.
Instability of A17T expression after multiple passages in the absence of rifampin.
RifR9 was passaged five times in BSC-1 cells at a multiplicity of infection of ∼1 PFU/cell and then five additional times at a multiplicity of infection of ∼0.1 PFU/cell. Western blotting was done essentially as described above. Lysates from ∼1/6 of one well of an uninfected or infected six-well plate were loaded onto a 10% Bis Tris polyacrylamide gel in 1× 2-(N-morpholino)ethanesulfonic acid buffer (Invitrogen), transferred to a nitrocellulose membrane, and probed with rabbit anti-A17-N antibody and mouse anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) antibodies, followed by secondary IRDye anti-rabbit–800 and IRDye anti-mouse–680 antibodies (LI-COR). The protein bands were visualized with an Odyssey imager (LI-COR).
Association of A17 and A17T with V5-D13.
BSC-1 cells in 24-well plates were treated with cytosine arabinoside (AraC) for 1 h at 37°C and infected by adsorbing vTF7-3 at a multiplicity of infection of 3 PFU per cell for 1 h at 37°C. After removal of the inoculum, plasmids derived from the pcDNA3.1D/V5-His-TOPO vector expressing A17T, full-length A17, or D13iV5 under the control of the T7 promoter were transfected into BSC-1 cells in the presence of AraC. The cells were harvested after 24 h at 37°C, and cells pelleted for 10 min at 2.5 × g were lysed in 500 μl of TSNS buffer (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 1% NP-40, 0.1% SDS, 1 cOmplete tablet) for 1 h at 4°C. Lysates were cleared by centrifugation at 14,000 × g for 10 min (5). Lysates (0.5 ml) were incubated for 1 h at room temperature with 20 μl of anti-V5 tag monoclonal antibody (MAb)-magnetic beads (MBL International, Woburn, MA). Beads were washed three times with a DynaMag-2 (Invitrogen) in TSNS buffer and one time with phosphate-buffered saline. Bound proteins were eluted with sample buffer containing SDS and subjected to Western blot analysis.
Whole-genome sequencing.
RifR9 and RifR10 mutants were isolated from a WT stock of VACV in the presence of 100 μg/ml rifampin and propagated in the same amount of the drug as described previously (26). Paired-end libraries were prepared by using the Illumina TruSeq DNA protocol and following the manufacturer's instructions. Sequencing was performed on a HiSeq 2000 next-generation sequencer (Illumina) and generated 2 × 95-bp paired-end reads. Reads were trimmed for adaptor and poor-quality sequences and mapped to the VACV genome by using Bowtie2 software (SourceForge.net). Single nucleotide polymorphisms (SNPs) were identified by using the Samtools/Bcftools (SourceForge.net) pipeline. Copy number and duplication and deletion analyses were performed by using CNVnator (33). To identify the precise nucleotide base positions for the duplication and insertion, we mapped the reads to the VACV WR genome by using the fusion mapping option of Tophat2 (34).
Transmission electron microscopy.
BSC-1 cells grown on Thermanox coverslips (Thermo Scientific) were infected at a multiplicity of infection of 3 PFU/cell. After 24 h, cells were fixed with 2.5% glutaraldehyde in 0.1 M sodium cacodylate and postfixed for 1 h with 0.5% osmium tetroxide–0.8% potassium ferricyanide, for 1 h with 1% tannic acid, and for 1 h with 1% uranyl acetate at room temperature. Samples were dehydrated with a graded ethanol series and embedded in Spurr's resin. Thin sections were cut with a Leica UCT ultramicrotome (Leica, Vienna, Austria) and stained with 1% uranyl acetate and Reynolds lead citrate prior to viewing at 120 kV on a FEI Tecnai BT Spirit transmission electron microscope (FEI, Hillsboro, OR). Digital images were acquired with an AMT digital camera system (AMT, Chazy, NY) and processed with Adobe Photoshop CS5 (Adobe Systems Inc., San Jose, CA).
Nucleotide sequence accession number.
The sequencing reads for RifR9 and RifR10 were deposited under accession number SRP041416 in the Sequence Read Archive of the National Center for Biotechnology Information.
RESULTS
Isolation of a gene duplication mutant with rifampin resistance.
Spontaneous rifampin-resistant mutants were isolated after a single round of replication in the presence of 100 μg/ml of rifampin and plaque purified as previously described (26). The D13 locus of DNA from individual virus plaques was sequenced as a first step in screening for alternative mutations. All rifampin-resistant isolates contained missense mutations in the D13L ORF, except for RifR9. RifR9 plaque sizes in BSC-1 cells were compared to those of a coselected rifampin-resistant virus (RifR10) determined to have an I485L mutation in the D13 ORF and to those of the WT virus in the presence or absence of rifampin. In the absence of the drug, the plaque sizes of the three viruses were similar to each other (Fig. 1A). However, both RifR9 and RifR10 formed plaques in the presence of rifampin, whereas no plaques were formed by the WT virus (Fig. 1A). The RifR9 plaques were smaller than the RifR10 plaques, suggesting a lower level of rifampin resistance. After 2 days, RifR9 and RifR10 plaques had mean diameters ± standard deviations of 0.8 ± 0.2 and 1.2 ± 0.3 mm, respectively (Fig. 1B). These results confirmed that RifR9 is resistant to rifampin, though apparently to a lesser extent than RifR10, and suggested the presence of a mutation residing outside the D13L locus.
Plaque sizes of rifampin-resistant mutants. (A) Plaques that formed in 2 days on BSC-1 cell monolayers in six-well plates by rifampin-resistant mutants RifR9 and RifR10 and WT VACV in the presence of 100 μg/ml of rifampin or the dimethyl sulfoxide (DMSO) carrier were visualized by staining with crystal violet and scanned. (B) ImageJ was used to calculate the average size of 50 to 80 plaques for each virus in the presence or absence of rifampin.
To determine the location of the putative mutation in RifR9 conferring rifampin resistance, the virus was amplified, and the genomic DNA was purified. As a control, RifR10, with a known mutation in the D13L ORF, was analyzed in parallel. Illumina next-generation whole-genome sequencing revealed two SNPs that were unlikely to be biologically significant in both of the rifampin-resistant viruses compared to the published sequence of VACV WR (GenBank accession no. {"type":"entrez-nucleotide","attrs":{"text":"AY243312.1","term_id":"29692106","term_text":"AY243312.1"}}AY243312.1). The latter alterations were a deletion of a C residue (position 42198) within a run of 5 C residues in the promoter region of F15L and a G-to-T transversion (position 184876) in a pseudogene. RifR9 contained no alterations in the D13L gene; however, a T-to-A transversion at position 108429, encoding the I485L mutation in D13, was found in RifR10, confirming the results of the sequence screening described above. Surprisingly, no other mutations were found in RifR9. However, there was an increase in the number of sequence reads in the region comprising VACWR137 (A17L), VACVWR138 (A18R), VACVWR139 (A19L), and half of VACVWR140 (A21L) in RifR9 compared to RifR10 (Fig. 2A). Copy number analysis determined that there was a 2× duplication (normalized read depth of 2.1143, P value = 1.11e-14) of this genomic region. The area undergoing duplication and insertion included nucleotides 125,792 to 128,172. Accordingly, A18R and A19L were completely duplicated, while the first copy of A21L was disrupted (the promoter region and much of the 5′ end were deleted in the fusion event), leaving only a single, full-length copy of A21L under the control of its natural promoter (Fig. 2B). The A17L duplication contained the WT promoter and 5′ end. At the fusion site, the 3′ end of A17L was truncated and fused in frame to a short portion of A21L before reaching a stop codon (Fig. 2C). Thus, the truncated A17 protein was predicted to be missing 69 amino acids from the C terminus, which included a predicted transmembrane segment. Primers were designed to amplify the region of duplication and confirmed the insertion (data not shown).
Analysis of a genomic duplication in RifR9. (A) Illumina TruSeq paired-end libraries of RifR9 and RifR10 were analyzed. Read counts of a high-coverage area spanning VACWR137 to VACWR140 were found that suggest a genomic duplication. The duplication was confirmed by PCR, and the TopHat program was used to locate the exact nucleotide breakpoint for the duplication within the A17L gene. (B) Diagram of duplicated and adjacent regions in RifR9 showing ORFs. (C) Alignment of WT A17 expressed by RifR10 and truncated A17T expressed by RifR9. Identical amino acids are represented by dots, and missing amino acids are represented by dashes. The predicted amino acid sequence of A17T is identical to that of WT A17 until the end of the third hydrophobic domain at amino acid position 135. The last five amino acids (NEPHT) of A17T result from the fusion with A21 and are followed by a stop codon. The position of the peptide target of the anti-A17-N antibody is shown.
Expression of the truncated A17 protein (A17T).
Although any part of the duplicated region in RifR9 could be responsible for resistance to rifampin, we focused on A17 because previous studies had shown an interaction between D13 and the N terminus of A17 (5, 14), whereas A18 is a DNA-dependent helicase involved in transcription (35, 36) and A19 is involved in a late stage of morphogenesis (37, 38). A17 is a 23-kDa protein that undergoes several modifications, including C- and N-terminal cleavages and phosphorylation. The A17 proteins synthesized in cells infected with the RifR9 and RifR10 mutants were analyzed by Western blotting with an antibody that recognized a peptide just internal to the N-terminal cleavage site (10). A 23-kDa A17 band and a more intense band that migrated faster because of the C-terminal cleavage (12) were detected upon Western blotting of proteins from cells infected with WT VACV (not shown) and RifR10 (Fig. 3A). An additional, faster-migrating band of the size predicted for A17T was detected upon Western blotting of proteins from cells infected with RifR9 (Fig. 3A). The viral A11 protein was probed with antibody as a positive control.
Expression and virion membrane association of A17T. (A) Western blot assays of infected-cell lysates. BSC-1 cells were infected with 1 PFU/cell of RifR9 (lane 1) or RifR10 (lane 2) for 24 h and harvested in SDS gel buffer. Following electrophoresis, Western blot assays were performed with rabbit anti-A17-N antibody, control anti-A11 antibody, and LI-COR secondary antibodies. The upper and lower bands marked A17 represent full-length and C-terminally truncated A17. The A17T and A11 bands are marked. The positions of mass markers in kDa are indicated on the left. (B) Incorporation of A17T into virions. Lysates from cells infected with RifR9 (lane 1) were pelleted through a 36% sucrose cushion, the pellet (lane 2) was suspended, and the purified virus (lane 3) was recovered from a band following centrifugation in a 24 to 40% continuous sucrose gradient. Proteins in the samples were resolved by electrophoresis and Western blot assays were probed with anti-A17-N and A3 (control) antibodies as in panel A. The upper and lower A3 bands represent unprocessed and processed forms. The lower A17 and A17T bands appear as doublets, indicating N-terminal processing. (C) Detergent extraction of proteins from purified virions. Purified RifR9 virions were incubated with 50 mM Tris–150 mM NaCl supplemented with 0.5% NP-40 alone or with 50 mM DTT. Soluble (S) and pellet (P) fractions were obtained by centrifugation and analyzed by gel electrophoresis. Western blot assays were probed with rabbit anti-A17-N, rabbit anti-A3, and mouse anti-D8 antibodies as in panels A and B. The values to the left of each panel represent the positions of mass markers in kDa. Slight mobility differences between A17 and A17T in individual lanes were apparently due to differences between the detergent and DTT.
Next, we analyzed whether A17T expressed by RifR9 was incorporated into virions. Western blot assays were made of fractions following cell lysis, sedimentation through a sucrose cushion, and banding in a sucrose density gradient. Antibody to the A3 protein (p4b), a major component of MVs that is processed during assembly, was used as a positive control. Bands derived from full-length A17 and A17T were detected by antibody to A17 in similar ratios during the purification steps (Fig. 3B). In this gel, the C-terminally truncated form of A17 could be resolved as a doublet because of N-terminal processing. The A17T protein was also resolved as a doublet, suggesting that the N terminus was similarly processed. Since three of the four putative hydrophobic domains of A17 were present in A17T, we investigated the presence of the protein in the membranes of virions. Purified virions were extracted with the detergent NP-40 with or without the reducing agent and then separated into soluble and pellet fractions. A17 was fully extractable only in the presence of NP-40 and dithiothreitol (DTT) (Fig. 3C). Interestingly, A17T was extractable with NP-40 alone, likely because of the absence of cysteine 178 from the C terminus, which has been shown to form intermolecular disulfide bonds between A17 molecules (9, 14, 39).
Overexpression of A17 is sufficient to confer resistance to rifampin.
The genomic duplication in RifR9 included A18R, A19L, and part of A21, as well as A17T (Fig. 2). Therefore, we needed to determine whether overexpression of A17 was sufficient to induce resistance to rifampin. Advantage was taken of a recombinant VACV that contains a WT A17L ORF regulated by an inducible T7 promoter, in addition to the endogenous A17 ORF under the control of its natural promoter (8). Synthesis of A17 increased at IPTG concentrations between 3.9 and 62.5 μM (Fig. 4A). Concurrently, there was an ∼1.5-log increase in the yield of VACV in the presence of rifampin, though it did not reach the yields obtained without the drug (Fig. 4B). Overexpression of A17 gradually reduced virus yields in the absence of rifampin, suggesting a fitness cost. The decreased virus yield was not due to IPTG itself, which has been shown to have no effect on the yield of the WT virus at concentrations of up to 500 μM (8). Plaque size in the presence of rifampin was also IPTG dependent (Fig. 4C). In addition, we tested a recombinant VACV that contained both a WT A19L ORF and an inducible A19 ORF (37). However, enhanced rifampin resistance was not detected when A19 was overexpressed in the presence of IPTG (data not shown).
Overexpression of A17 provides resistance to rifampin. (A) Effect of IPTG on A17 expression. BSC-1 cells were infected with recombinant VACV containing two WT A17L ORFs, one expressed from its natural promoter and another expressed from an inducible T7 promoter, and 100 μg/ml of rifampin or dimethyl sulfoxide (DMSO) carrier was added. A17 expression above WT levels was induced with the indicated amounts of IPTG for 24 h. Cleared lysates were analyzed by Western blotting with a rabbit anti-A17-N antibody. The values to the left represent the positions of mass markers in kDa. (B) Effects of IPTG on virus yields. Virus titers of BSC-1 cells without rifampin were determined in three independent experiments. (C) Effect of IPTG on plaque formation. BSC-1 cell monolayers were infected with the A17-inducible virus as described in panel A, in the presence or absence of IPTG and presence or absence of 100 μg/ml rifampin. After 48 h, plaques were visualized by staining with crystal violet. (D) Recombinant viruses expressing second copies of A17 exhibited resistance to rifampin. Recombinant VACVs were constructed to express a second copy of WT A17 (+A17) or of A17T (+A17T) under the control of the natural A17 promoter to mimic the expected expression level of the RifR9 virus. In three independent experiments, cells were infected at 0.1 PFU/cell with +A17, +A17T, WT, or RifR9 virus in the presence of rifampin or DMSO carrier for 24 h. Cells were harvested, and virus titers were determined in the absence of rifampin.
Since overexpression of full-length A17 regulated by the T7 promoter increased rifampin resistance, we next determined whether the introduction of the full-length A17 or the A17T ORF, under the control of the natural A17 promoter, into a WT virus could also induce rifampin resistance. The additional ORF was inserted into the F12/F13 locus without disrupting any genes by the small-to-large plaque selection procedure previously described (31). Knocking in of either full-length or truncated A17 induced rifampin resistance similar to that of RifR9 (Fig. 4D), indicating that the partial duplication of A17 present in RifR9 is sufficient for the rifampin resistance phenotype.
Effects of A17 overexpression on virion assembly.
Transmission electron microscopy was used to visualize the effects of overexpression of A17 and A17T on crescent formation in the presence of rifampin. In cells infected with WT virus, curved crescent membranes and spherical IVs (Fig. 5B), as well as brick-shaped MVs (not shown), formed in the absence of rifampin, whereas none of these forms appeared in the presence of rifampin (Fig. 5A). Instead, there were large masses of dense viroplasm delimited by irregular smooth membranes lacking the D13 scaffold as described previously (18,–20). In contrast, typical crescents and IVs formed in cells infected with RifR10, expressing the I485L rifampin-resistant form of D13, in the presence (Fig. 5C) or the absence (Fig. 5D) of rifampin. Although RifR9 exhibited a WT phenotype in the absence of rifampin (Fig. 5F), both normal and aberrant virus structures were observed in the presence of the drug (Fig. 5E). This intermediate phenotype was distinct from that of the WT virus and consistent with the lower rifampin resistance of RifR9 than RifR10 (Fig. 1). As expected, the A17-inducible virus was sensitive to rifampin in the absence of IPTG (Fig. 5G). However, in the presence of IPTG, the A17-inducible virus exhibited full rifampin resistance because of the large amounts of A17 produced (Fig. 5H).
Crescent and IV formation by WT and rifampin-resistant VACVs. BSC-1 cells treated with 100 μg rifampin/ml (A, C, E, G, H) or dimethyl sulfoxide (DMSO) carrier (B, D, F) were infected at 3 PFU/cell with WT VACV (A, B), RifR10 (C, D), RifR9 (E, F), or A17-inducible virus in the presence of rifampin without IPTG (G) and with IPTG (H). After 24 h, cells were fixed and viewed by transmission electron microscopy. Arrows indicate irregular membranes lacking scaffold; arrowheads indicate crescent or IV membrane with scaffold. Scale bars are 0.25 μm.
Binding of A17T to D13.
Previous studies showed that D13 binds A17 through the N terminus of the latter (5, 14). We presumed, therefore, that A17T, which has a C-terminal truncation, would also bind D13. To investigate binding, we infected cells with recombinant VACV expressing the bacteriophage T7 RNA polymerase in the presence of AraC to prevent the expression of endogenous A17 and other viral postreplicative proteins. A plasmid encoding D13 with an internal V5 tag (D13iV5) designed to locate the epitope on the membrane-distal side of D13, together with a plasmid encoding full-length A17 or A17T regulated by the bacteriophage T7 promoter, were transfected into the infected cells. At 24 h after transfection, D13iV5 was captured with anti-V5 magnetic beads along with A17 or A17T, indicating that both forms of A17 could individually interact with D13 (Fig. 6).
Association of A17 and A17T with D13. BSC-1 cells were infected with a recombinant VACV expressing bacteriophage T7 RNA polymerase regulated by an early promoter in the presence of AraC. The cells were then transfected with plasmids expressing internally V5-tagged D13 (D13iV5), WT A17 (A17), or A17T regulated by a T7 promoter. After 24 h, the cells were lysed in buffer containing 1% NP-40 and 0.1% SDS and clarified by centrifugation. The supernatants were analyzed directly by Western blotting (Input) or after binding to and elution from an anti-V5 peptide antibody conjugated to magnetic beads (IP). The Western blot assays were probed with antibodies to V5 and A17. The positions of molecular mass markers (values are in kDa) are shown on the left. The asterisk denotes the IgG light chain that eluted from the beads; a faint band representing the heavy chain is present just below the D13iV5 band.
Loss of A17T expression upon repeated passages of RifR9 without selection.
Until now, we propagated RifR9 in the presence of rifampin, which provided continuous selection for maintenance of the gene duplication. Although the plaque size (Fig. 1) and virus yield (Fig. 4D) of RifR9 were similar to those of the WT virus in the absence of the drug, this did not preclude a small but significant fitness cost that could be linked to overexpression of A17 or A17T (Fig. 4B). In the experiment depicted in Fig. 7, RifR9 was passaged multiple times in the absence of the drug with the idea that mutants no longer expressing A17T might arise and compete effectively with the parent RIFR9 virus. In the first five passages, a multiplicity of infection of 1 PFU/cell was used. Western blot analysis indicated a modest decrease in A17T expression relative to that of full-length A17. For the next five passages, the multiplicity of infection was reduced to 0.1 PFU/cell in order to facilitate the expansion of the fittest virus. After five more passages under these conditions, little or no expression of A17T was detected (Fig. 7A). Virus yield experiments indicated that the loss of A17T expression correlated with loss of rifampin resistance (Fig. 7B). The difference in replication in the presence of rifampin between the passage 10 virus and the RifR9 virus was highly significant (P = 0.009). These data further confirm the relationship of A17T overexpression and rifampin resistance.
Loss of A17T expression during multiple passages of RifR9 in the absence of selection. RifR9 was passaged five times in BSC-1 cells at a multiplicity of infection of ∼1 PFU/cell and then five additional times at a multiplicity of infection of ∼0.1 PFU/cell in the absence of rifampin. Expression of full-length A17 and A17T by viruses from successive passages 2 (P2), 4 (P4), 6 (P6), 8 (P8), and 10 (P10) was determined by probing Western blot assays with rabbit anti-A17-N antibody, followed by secondary IRDye anti-rabbit–800 (green). Equal loading was confirmed by probing with a mouse MAb to GAPDH, followed by IRDye anti-mouse–680 (red). The positions and masses in kDa of electrophoretic markers are shown on the left.
DISCUSSION
D13 is a capsid-like protein that forms an external lattice that provides curvature to the membrane of crescents and IVs (2,–4, 40). When expression of D13 is repressed or the drug rifampin is added, irregular viral membranes are formed and assembly of virus particles is blocked. All previously known VACV mutations that confer resistance to rifampin were found to be missense mutations in the D13L ORF, leading to the still unproven idea that the drug interacts with D13 directly to inhibit VACV assembly (23,–26). This report describes a spontaneous rifampin-resistant VACV mutant with an unaltered D13 ORF. Surprisingly, whole-genome sequencing did not reveal a point mutation that could account for rifampin resistance. However, a doubling of reads in the region encompassing most of A17L, A18R, and A19L and part of A21L indicated a duplication event. Although duplication of any of the above genes could be involved in rifampin resistance, we focused on A17L because the A17 protein was known to interact with D13 (5, 14). The duplicated A17 protein was called A17T because it was missing 69 predicted amino acids at the C terminus, which is required for intermolecular disulfide bond formation, C-terminal proteolytic processing, phosphorylation, and the transition of IVs to MVs but not for binding of D13 and formation of crescents and IV structures (5, 14). We showed that A17T was incorporated into the membrane of virus particles and retained the ability to bind D13. A17T was extracted from viral membranes with a detergent in the absence of a reducing agent, consistent with the loss of a cysteine residue near the C terminus (9, 14, 39). Importantly, insertion into the WT virus of the gene for a second full-length A17 protein or the truncated version was sufficient to confer rifampin resistance. The inability of C-terminally truncated A17 to function in the conversion of IVs to MVs (5, 14) suggests that a high copy number of A17T could have a dominant negative effect, possibly explaining why mutants with multiple A17T copies were not selected during further propagation in the presence of the drug (data not shown). Although the N and C termini of A17 appear to have different functions, tandem expression of the membrane-bound fragments lacking the N and C termini could not complement the assembly defect caused by repression of WT A17 expression (unpublished data). It seems likely that with further screening we would also find spontaneous mutants with full-length A17, though overexpression of full-length A17 also comes with a fitness cost.
In the absence of rifampin, the yield of RifR9 was similar to that of the WT virus in a single-cycle growth experiment. Nevertheless, a virus not expressing A17T arose and became dominant during multiple passages without the drug, indicating that A17T expression carries a small but significant fitness cost. This fitness cost was also demonstrated in a single-round infection by inducibly increasing the expression of full-length A17. How overexpression of A17 interferes with virus replication was not determined. However, this may be a general phenomenon, since loss of K3L duplications was found when the selection pressure was removed (28).
We considered the following observations in constructing a model to explain why overexpression of the N-terminal region of A17, as well as mutations in D13, can contribute to resistance to rifampin. (i) A17 is associated with the assembly of viral membranes (8), (ii) D13 interacts with the N-terminal region of A17 (5, 14), (iii) point mutations in D13 that confer rifampin resistance map to its presumed membrane-proximal region (4, 40), (iv) high concentrations of rifampin prevent the association of D13 with nascent membranes (21), (v) rifampin inhibition of virus assembly is highly reversible (18), and (vi) rifampin cannot dissociate D13 from viral membranes (19). We suggest a model to account for rifampin resistance based on the assumptions that D13 binds rifampin reversibly and A17 irreversibly and that D13 bound to rifampin cannot simultaneously bind A17 (Fig. 8). Consequently, mutations in D13 that diminish binding to rifampin decrease reaction 1 or increase reaction 2, allowing free D13 to interact irreversibly with A17 and form crescents (reaction 3). However, increasing A17 would also favor reaction 3, resulting in an alternative mechanism of rifampin resistance.
Model depicting rifampin resistance induced by A17 overexpression. The model makes the following assumptions. (i) D13 reversibly binds rifampin (Rif) (arrows 1 and 2) or irreversibly binds A17 (arrow 3), and (ii) D13 bound to rifampin cannot bind A17 (arrow 4). The colon indicates an interaction between molecules. The curved line under A17:D13 represents a crescent membrane. Overexpression of A17 favors binding of D13 to A17 in the presence of rifampin.
In summary, the present study provides the first evidence of an alternative mechanism of rifampin resistance by poxviruses. In addition, it affirms the importance of the interaction between the D13 scaffold protein and the A17 membrane protein for the assembly of virus particles, which previously had been demonstrated only by immunoprecipitation. Furthermore, the expansion of the A17L region of the VACV genome to overcome the antiviral activity of rifampin resembles the resistance to hydroxyurea enabled by duplication of the gene encoding the small subunit of ribonucleotide reductase (27) and of the gene for K3L to allow adaptation to the antiviral action of protein kinase R (28), except that higher copy numbers were attained in the other cases. Gene duplication may provide a way for poxviruses and other DNA viruses with high-fidelity DNA polymerases to adjust rapidly to changes in the environment.
ACKNOWLEDGMENTS
We thank P. S. Satheshkumar, Z. Yang, and J. Laliberte for helpful discussions.
This research was supported by the Division of Intramural Research, National Institute of Allergy and Infectious Diseases. K.J.E. received support from a PRAT fellowship provided by the National Institute of General Medical Sciences.
Footnotes
Published ahead of print 30 July 2014
REFERENCES
1. Moss B. 2013. Poxviridae, p 2129–2159 In Knipe DM, Howley PM. (ed), Fields virology, sixth ed, vol 2 Lippincott Williams & Wilkins, Philadelphia, PA [Google Scholar]
2. Heuser J. 2005. Deep-etch EM reveals that the early poxvirus envelope is a single membrane bilayer stabilized by a geodetic "honeycomb" surface coat. J. Cell Biol. 169:269–283. 10.1083/jcb.200412169 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
3. Szajner P, Weisberg AS, Lebowitz J, Heuser J, Moss B. 2005. External scaffold of spherical immature poxvirus particles is made of protein trimers, forming a honeycomb lattice. J. Cell Biol. 170:971–981. 10.1083/jcb.200504026 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
4. Hyun JK, Accurso C, Hijnen M, Schult P, Pettikiriarachchi A, Mitra AK, Coulibaly F. 2011. Membrane remodeling by the double-barrel scaffolding protein of poxvirus. PLoS Pathog. 7(9):e1002239. 10.1371/journal.ppat.1002239 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
5. Bisht H, Weisberg AS, Szajner P, Moss B. 2009. Assembly and disassembly of the capsid-like external scaffold of immature virions during vaccinia virus morphogenesis. J. Virol. 83:9140–9150. 10.1128/JVI.00875-09 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
6. Zhang Y, Moss B. 1992. Immature viral envelope formation is interrupted at the same stage by lac operator-mediated repression of the vaccinia virus D13L gene and by the drug rifampicin. Virology 187:643–653. 10.1016/0042-6822(92)90467-4 [PubMed] [CrossRef] [Google Scholar]
7. RodrÃguez D, Esteban M, RodrÃguez JR. 1995. Vaccinia virus A17L gene product is essential for an early step in virion morphogenesis. J. Virol. 69:4640–4648 [PMC free article] [PubMed] [Google Scholar]
8. Wolffe EJ, Moore DM, Peters PJ, Moss B. 1996. Vaccinia virus A17L open reading frame encodes an essential component of nascent viral membranes that is required to initiate morphogenesis. J. Virol. 70:2797–2808 [PMC free article] [PubMed] [Google Scholar]
9. Rodriguez D, Rodriguez JR, Esteban M. 1993. The vaccinia virus 14-kilodalton fusion protein forms a stable complex with the processed protein encoded by the vaccinia virus A17L gene. J. Virol. 67:3435–3440 [PMC free article] [PubMed] [Google Scholar]
10. Betakova T, Wolffe EJ, Moss B. 1999. Regulation of vaccinia virus morphogenesis: phosphorylation of the A14L and A17L membrane proteins and C-terminal truncation of the A17L protein are dependent on the F10L protein kinase. J. Virol. 73:3534–3543 [PMC free article] [PubMed] [Google Scholar]
11. Derrien M, Punjabi A, Khanna R, Grubisha O, Traktman P. 1999. Tyrosine phosphorylation of A17 during vaccinia virus infection: Involvement of the H1 phosphatase and the F10 kinase. J. Virol. 73:7287–7296 [PMC free article] [PubMed] [Google Scholar]
12. Ansarah-Sobrinho C, Moss B. 2004. Role of the I7 protein in proteolytic processing of vaccinia virus membrane and core components. J. Virol. 78:6335–6343. 10.1128/JVI.78.12.6335-6343.2004 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
13. Matson J, Chou W, Ngo T, Gershon PD. 2014. Static and dynamic protein phosphorylation in the vaccinia virion. Virology 452:310–323. 10.1016/j.virol.2014.01.012 [PubMed] [CrossRef] [Google Scholar]
14. Unger B, Mercer J, Boyle KA, Traktman P. 2013. Biogenesis of the vaccinia virus membrane: genetic and ultrastructural analysis of the contributions of the A14 and A17 proteins. J. Virol. 87:1083–1097. 10.1128/JVI.02529-12 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
15. Krijnse-Locker J, Schleich S, Rodriguez D, Goud B, Snijder EJ, Griffiths G. 1996. The role of a 21-kDa viral membrane protein in the assembly of vaccinia virus from the intermediate compartment. J. Biol. Chem. 271:14950–14958. 10.1074/jbc.271.25.14950 [PubMed] [CrossRef] [Google Scholar]
16. Betakova T, Wolffe EJ, Moss B. 1999. Membrane topology of the vaccinia virus A17L envelope protein. Virology 261:347–356. 10.1006/viro.1999.9870 [PubMed] [CrossRef] [Google Scholar]
17. Maruri-Avidal L, Weisberg AS, Moss B. 2013. Direct formation of vaccinia viral membranes from the endoplasmic reticulum in the absence of the newly characterized L2-interacting A30.5 protein. J. Virol. 87:12313–12326. 10.1128/JVI.02137-13 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
18. Moss B, Rosenblum EN, Katz E, Grimley PM. 1969. Rifampicin: a specific inhibitor of vaccinia virus assembly. Nature 224:1280–1284. 10.1038/2241280a0 [PubMed] [CrossRef] [Google Scholar]
19. Grimley PM, Rosenblum EN, Mims SJ, Moss B. 1970. Interruption by rifampin of an early stage in vaccinia virus morphogenesis: accumulation of membranes which are precursors of virus envelopes. J. Virol. 6:519–533 [PMC free article] [PubMed] [Google Scholar]
20. Nagayama A, Pogo BG, Dales S. 1970. Biogenesis of vaccinia: separation of early stages from maturation by means of rifampicin. Virology 40:1039–1051. 10.1016/0042-6822(70)90150-9 [PubMed] [CrossRef] [Google Scholar]
21. Sodeik B, Griffiths G, Ericsson M, Moss B, Doms RW. 1994. Assembly of vaccinia virus: effects of rifampin on the intracellular distribution of viral protein p65. J. Virol. 68:1103–1114 [PMC free article] [PubMed] [Google Scholar]
22. Grimley PM, Moss B. 1971. Similar effect of rifampin and other rifamycin derivatives on vaccinia virus morphogenesis. J. Virol. 8:225–231 [PMC free article] [PubMed] [Google Scholar]
23. Tartaglia J, Piccini A, Paoletti E. 1986. Vaccinia virus rifampicin resistance locus specifies a late 63,000 Da gene product. Virology 150:45–54. 10.1016/0042-6822(86)90264-3 [PubMed] [CrossRef] [Google Scholar]
24. Baldick CJ, Moss B. 1987. Resistance of vaccinia virus to rifampicin conferred by a single nucleotide substitution near the predicted NH2 terminus of a gene encoding an Mr 62,000 polypeptide. Virology 156:138–145. 10.1016/0042-6822(87)90444-2 [PubMed] [CrossRef] [Google Scholar]
25. McNulty-Kowalczyk A, Paoletti E. 1993. Mutations in ORF D13L and other genetic loci alter the rifampicin phenotype of vaccinia virus. Virology 194:638–646. 10.1006/viro.1993.1303 [PubMed] [CrossRef] [Google Scholar]
26. Charity JC, Katz E, Moss B. 2007. Amino acid substitutions at multiple sites within the vaccinia virus D13 scaffold protein confer resistance to rifampicin. Virology 359:227–232. 10.1016/j.virol.2006.09.031 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
27. Slabaugh MB, Roseman NA, Mathews CK. 1989. Amplification of the ribonucleotide reductase small subunit gene: analysis of novel joints and the mechanism of gene duplication in vaccinia virus. Nucleic Acids Res. 17:7073–7088. 10.1093/nar/17.17.7073 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
28. Elde NC, Child SJ, Eickbush MT, Kitzman JO, Rogers KS, Shendure J, Geballe AP, Malik HS. 2012. Poxviruses deploy genomic accordions to adapt rapidly against host antiviral defenses. Cell 150:831–841. 10.1016/j.cell.2012.05.049 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
29. Brennan G, Kitzman JO, Rothenburg S, Shendure J, Geballe AP. 2014. Adaptive gene amplification as an intermediate step in the expansion of virus host range. PLoS Pathog. 10(3):e1004002. 10.1371/journal.ppat.1004002 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
30. Fuerst TR, Niles EG, Studier FW, Moss B. 1986. Eukaryotic transient-expression system based on recombinant vaccinia virus that synthesizes bacteriophage T7 RNA polymerase. Proc. Natl. Acad. Sci. U. S. A. 83:8122–8126. 10.1073/pnas.83.21.8122 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
31. Blasco R, Moss B. 1995. Selection of recombinant vaccinia viruses on the basis of plaque formation. Gene 158:157–162. 10.1016/0378-1119(95)00149-Z [PubMed] [CrossRef] [Google Scholar]
32. Earl PL, Cooper N, Wyatt LS, Moss B, Carroll MW. 2001. Preparation of cell cultures and vaccinia virus stocks. Curr. Protoc. Mol. Biol. Chapter 16:Unit16.16. 10.1002/0471142727.mb1616s43 [PubMed] [CrossRef] [Google Scholar]
33. Abyzov A, Urban AE, Snyder M, Gerstein M. 2011. CNVnator: an approach to discover, genotype, and characterize typical and atypical CNVs from family and population genome sequencing. Genome Res. 21:974–984. 10.1101/gr.114876.110 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
34. Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R, Salzberg SL. 2013. TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol. 14:R36. 10.1186/gb-2013-14-4-r36 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
35. Simpson DA, Condit RC. 1995. Vaccinia virus gene A18R encodes an essential DNA helicase. J. Virol. 69:6131–6139 [PMC free article] [PubMed] [Google Scholar]
36. Lackner CA, Condit RC. 2000. Vaccinia virus gene A18R DNA helicase is a transcript release factor. J. Biol. Chem. 275:1485–1494. 10.1074/jbc.275.2.1485 [PubMed] [CrossRef] [Google Scholar]
37. Satheshkumar PS, Weisberg AS, Moss B. 2013. Vaccinia virus A19 protein participates in the transformation of spherical immature particles to barrel shaped infectious virions. J. Virol. 87:10700–10709. 10.1128/JVI.01258-13 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
38. Satheshkumar PS, Olano LR, Hammer CH, Zhao M, Moss B. 2013. Interactions of the vaccinia virus A19 protein. J. Virol. 87:10710–10720. 10.1128/JVI.01261-13 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
39. Betakova T, Moss B. 2000. Disulfide bonds and membrane topology of the vaccinia virus A17L envelope protein. J. Virol. 74:2438–2442. 10.1128/JVI.74.5.2438-2442.2000 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
40. Bahar MW, Graham SC, Stuart DI, Grimes JM. 2011. Insights into the evolution of a complex virus from the crystal structure of vaccinia virus D13. Structure 19:1011–1020. 10.1016/j.str.2011.03.023 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
Articles from Journal of Virology are provided here courtesy of American Society for Microbiology (ASM)
Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4178789/