Infectious Bronchitis Virus 3a Protein Localizes t
http://www.100md.com
病菌学杂志 2005年第10期
Department of Cell Biology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205
ABSTRACT
All coronaviruses possess small open reading frames (ORFs) between structural genes that have been hypothesized to play important roles in pathogenesis. Infectious bronchitis virus (IBV) ORF 3a is one such gene. It is highly conserved among group 3 coronaviruses, suggesting that it has an important function in infection. IBV 3a protein is expressed in infected cells but is not detected in virions. Sequence analysis predicted that IBV 3a was a membrane protein; however, only a fraction behaved like an integral membrane protein. Microscopy and immunoprecipitation studies demonstrated that IBV 3a localized to the cytoplasm in a diffuse pattern as well as in sharp puncta in both infected and transfected cells. These puncta did not overlap cellular organelles or other punctate structures. Confocal microscopy demonstrated that IBV 3a puncta lined up along smooth endoplasmic reticulum (ER) tubules and, in a significant number of instances, were partially surrounded by these tubules. Our results suggest that IBV 3a is partially targeted to a novel domain of the smooth ER.
INTRODUCTION
Members of the family Coronaviridae are positive-sense, single-stranded RNA viruses (36). They recently received much attention due to a newly discovered coronavirus implicated in severe acute respiratory syndrome. Coronaviruses have been divided into three groups based on serological cross-reactivity and sequence analysis. However, despite their division into separate groups, coronaviruses share many common characteristics. For example, coronaviruses from all three groups possess a helical capsid structure (43) and derive their viral envelopes by budding into the endoplasmic reticulum (ER)-Golgi-intermediate compartment (ERGIC) (23). They also share similar replication strategies (36). Once the virus enters the cell, the RNA genome serves as a template for the translation of two polyproteins, 1a and 1ab, containing the viral RNA-dependent RNA polymerase (47). A set of negative-stranded, subgenomic RNAs then are transcribed (36). These negative-stranded RNAs are used to create positive-stranded, subgenomic RNAs. One to three of the 5'-most open reading frames (ORFs) then are translated from these positive-stranded, subgenomic RNAs.
Coronaviruses encode four structural proteins: spike (S), envelope (E), membrane (M), and nucleocapsid (N). The S protein, found in the virus envelope, is responsible for binding of the virus to its receptor and subsequent fusion of the viral envelope with cellular membranes. The E and M proteins are also components of the viral envelope. Although their functions are still under investigation, they are thought to be important for virus budding. Finally, the N protein binds to both the viral genome and the M protein, allowing incorporation of the genome into mature virions (45). In addition to these well-studied structural proteins, coronaviruses possess additional ORFs interspersed throughout their genomes (9). Some of these ORFs are known to produce proteins during viral infection. However, these proteins are largely thought to be involved in pathogenesis rather than virus replication or assembly. For example, ORF 5a in infectious bronchitis virus (IBV), a group 3 coronavirus, was recently found to be nonessential for virus replication in cell cultures (49). Additionally, the deletion of several ORFs (2a, hemagglutinin-esterase, 4a, 4b, and 5a) in the group 2 coronavirus murine hepatitis virus appears to have no effect on virus replication in cell cultures (13). However, these deletions in murine hepatitis virus did appear to reduce virus virulence during infection of the natural host.
One ORF-containing region, region II, lies between the S and E genes and is present in all coronaviruses. IBV region II produces two small proteins, 3a and 3b, during infection (28). These proteins are translated from subgenomic mRNA 3, a functionally polycistronic mRNA, via leaky ribosomal scanning (30). The E structural protein (originally called 3c) is translated from the same mRNA by an internal ribosome entry site. While region II is not well conserved among different coronavirus groups, it is very well conserved within the group 3 coronaviruses. One study which compared the polypeptide sequences of group 3 field isolates from different continents and decades found the similarities of IBV 3a to be 81 to 86.2% and of IBV 3b to be 87.5 to 95.4% (21). Recently, a study demonstrated the emergence of a truncated form of IBV 3b in cell cultures, indicating that IBV 3b may not be essential for virus replication (42). However, the persistent presence and high degree of conservation of IBV 3a and IBV 3b despite unfavorable selective pressure suggest that they are important in some aspect of productive viral infection.
Here we report the characterization of the IBV 3a protein. We found that one pool of the IBV 3a protein is cytoplasmic and that another pool is tightly associated with membranes. The membrane-associated fraction localizes to puncta that line smooth ER membranes, suggesting a potentially novel function for this protein.
MATERIALS AND METHODS
Cells and viruses. Vero and HeLa cells were maintained in high-glucose Dulbecco's modified Eagle's medium (DMEM) (Invitrogen Life Technologies, Carlsbad, CA) supplemented with 5 or 10% fetal calf serum (Atlanta Biologicals, Norcross, GA) and 0.1 mg/ml Normocin (Invivogen, San Diego, CA). The previously described Vero-adapted Beaudette strain of IBV (32) was used in all IBV infection experiments, and titers were determined with Vero cells. A vaccinia virus encoding T7 polymerase (vTF7-3) was obtained from Bernard Moss (National Institutes of Health, Bethesda, MD) (15). HeLa cells stably expressing sec13-green fluorescent protein (GFP) were created by previously described methods (33).
IBV 3a sequence analysis. The IBV 3a sequence (Beaudette strain) has been published (8). Hydropathy analysis was carried out by methods developed by Kyte and Doolittle (25). Prediction of a signal sequence cleavage site in IBV 3a was obtained from the PSORT II website (http://psort.ims.u-tokyo.ac.jp/form2.html) developed by Horton and Nakai (18).
Plasmids. The IBV 3a gene was cloned by reverse transcription-PCR of mRNA isolated from IBV-infected Vero cells. Briefly, primers containing an EcoRI restriction site in front of the 5' coding region of IBV 3a and 3' primers containing a BamHI restriction site behind the stop codon of IBV 3a were used to amplify the IBV 3a gene by reverse transcription-PCR. These two restriction sites then were used to clone IBV 3a into pBluescript SK (pBS) (Stratagene, La Jolla, CA) behind the T7 promoter. Dideoxy sequencing confirmed that the IBV 3a sequence in this construct (called pBS/IBV 3a) matched the known IBV 3a Beaudette strain sequence. A truncated version of IBV 3a (pBS/truncated 3a) that consisted of C23 to the end of the protein was created by PCR amplification. Briefly, a primer containing an EcoRI site followed by a Kozak sequence and an AUG start codon was placed 5' of the codon for C23. The same 3' primer used for making pBS/IBV 3a was used in this reaction. The PCR product then was inserted into plasmid pBS by using the EcoRI and BamHI restriction sites. A version of IBV 3a predicted to be uncleavable (1) was generated by QuikChange mutagenesis (Stratagene, La Jolla, CA) according to the manufacturer's protocol with pBS/IBV 3a as a template, changing Ser22 to an Ile (pBS/3a VLS-VLI).
The EcoRI and NotI restriction sites in pBS/IBV 3a were used to subclone IBV 3a into pcDNA3.1/Myc-His(+)A (Invitrogen) behind the cytomegalovirus promoter, creating pcDNA/IBV 3a. The EcoRI and BamHI restriction sites in pBS/IBV 3a were used to subclone IBV 3a into pEGFP-N1 (BD Biosciences Clontech, San Diego, CA) behind the cytomegalovirus promoter, creating pEGFP-N1/IBV 3a. A C-terminally GFP-tagged version of IBV 3a was created by performing QuikChange mutagenesis on construct pEGFP-N1/IBV 3a, creating pEGFP-N1/IBV 3a-GFP.
A myc-tagged peroxisome thioesterase 1 (PTE-1) construct was a gift from S. J. Gould (Johns Hopkins School of Medicine, Baltimore, MD). A construct containing human red fluorescent protein (RFP)-tagged Sm-like protein 1 (LSm1) was a gift from H. C. Dietz (Johns Hopkins School of Medicine, Baltimore, MD). A plasmid encoding sec13-GFP was received from F. Gorelick (Yale University School of Medicine, New Haven, CT). A plasmid (pCFP-ER) encoding cyan fluorescent protein with the ER localization signal KDEL (CFP-KDEL) was obtained from BD Biosciences Clontech.
Antibodies. A peptide corresponding to the C-terminal 14 amino acids (with an additional cysteine residue) of IBV 3a was used to generate polyclonal antibodies by previously described methods (10). The rabbit polyclonal IBV E and IBV M antibodies used in this study were described previously (10, 32).
Polyclonal rabbit anti-MxA antibodies were a gift from M. McNiven (Mayo Clinic, Rochester, MN). Mouse anti-ERGIC 53 antibodies were a gift from H. P. Hauri (Biocenter of the University of Basel, Basel, Switzerland). Mouse anti-TIA-R antibodies were a gift from P. Anderson (Brigham and Women's Hospital, Boston, MA). Mouse anti-LAMP2 antibodies were a gift from T. August (Johns Hopkins University School of Medicine, Baltimore, MD). Mouse anti-mannose-6-phosphate receptor (MPR) antibodies were a gift from S. Pfeffer (Stanford University School of Medicine, Palo Alto, CA). Mouse anti-gamma-adaptin (AP1) antibodies were obtained from Sigma-Aldrich Co. (St. Louis, MO). Mouse anti-rab9 and anti-GFP antibodies were obtained from Roche Molecular Biochemicals (Indianapolis, IN). Goat anti-BiP antibodies were obtained from Research Diagnostics, Inc. (Flanders, NJ). Rabbit anti-GFP and Alexa-488-conjugated donkey anti-mouse (immunoglobulin G [IgG]) antibodies were obtained from Molecular Probes, Inc. (Eugene, OR). Mouse anti-GM130, mouse anti-p230, and mouse anti-p115 antibodies were obtained from BD Transduction (San Diego, CA). Texas Red-conjugated donkey anti-rabbit (IgG), fluorescein isothiocyanate (FITC)-conjugated anti-rat (IgG), FITC-conjugated goat anti-mouse (IgG), FITC-conjugated donkey anti-rabbit (IgG), Texas Red-conjugated donkey anti-mouse (IgG), and Texas Red-conjugated donkey anti-goat (IgG) antibodies were obtained from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA). Horseradish peroxidase-conjugated donkey anti-rabbit (IgG) antibodies were obtained from Amersham Pharmacia Biotech, Inc., (Piscataway, NJ).
The titers of the IBV M and IBV 3a antibodies were compared by enzyme-linked immunosorbent assays (ELISAs). Stock solutions (1 mg/ml) of IBV 3a and IBV M peptides were diluted to 0.1 μg/150 μl in 0.1 M NaHCO3 (pH 9.6) and absorbed overnight at 4°C in a 96-well plate (Corning Incorporated, Corning, NY). Following removal of the peptide solution, wells were blocked for 2 h at room temperature with 10 mg/ml bovine serum albumin (BSA) (Sigma) and 0.02% Tween 20 in phosphate-buffered saline. Wells were extensively washed, and anti-IBV 3a or anti-IBV M antibodies were added to wells in serial dilutions. Primary antibody incubation was performed for 45 min at room temperature and for an additional 15 min at 37°C. Secondary horseradish peroxidase-conjugated donkey anti-rabbit antibodies were added (following extensive washing) for 30 min at 37°C. BM Blue POD soluble substrate (Boehringer-Mannheim GmbH, Mannheim, Germany) was used according to the manufacturer's protocol, and antibody binding was measured by using a model 3550 microplate reader (Bio-Rad Laboratories, Hercules, CA). The titers (the dilution at 50% the maximal optical density for 0.1 μg peptide) were approximately 1:10,000 for both anti-IBV 3a and anti-IBV M antibodies.
Infection, transient transfection, indirect immunofluorescence microscopy, and confocal microscopy. All cells used in immunofluorescence studies were Vero cells unless otherwise specified. In colocalization studies, cells were plated on coverslips in 35-mm dishes for approximately 1 day before receiving one of three treatments: (i) infection with IBV, (ii) transient transfection, or (iii) infection with vTF7-3 followed by transfection. IBV infection was performed at a multiplicity of infection (MOI) of 1 by allowing virus to absorb to cells for 1 h at 37°C before transferring cells to normal growth medium. IBV-infected cells were stained for immunofluorescence at between 10 and 28 h postinfection (hpi). Transient transfection was performed by adding 2 μg DNA/35-mm dish for approximately 24 h with either LT1 (Mirus Bio Corporation, Madison, WI) or Fugene 6 (Roche) according to the manufacturers' protocols. Infection with vTF7-3 was performed as previously described (10). At 1 hpi, vTF7-3-infected cells were transfected with pBS/IBV 3a by using Lipofectin (Invitrogen) according to the manufacturer's protocol. Cells were processed for indirect immunofluorescence microscopy (described below) at approximately 4 hpi.
IBV-infected and vTF7-3-infected cells were used in studies that assessed colocalization between IBV 3a and the following proteins: ERGIC 53 (41), GM130 (a marker for the cis-Golgi compartment) (35), p115 (a marker for the cis-Golgi compartment and ERGIC) (48), p230 (a marker for the trans-Golgi network) (24), BiP (a marker for the ER) (7), TIA-R (a marker for stress granules) (3, 46), and AP1 (a marker for the trans-Golgi network and late endosomes) (38). IBV-infected cells were used in studies that assessed colocalization between IBV 3a and the following proteins or dyes: rab9 (a marker for late endosomes) (31), MPR (a marker for the trans-Golgi network and late endosomes) (14), MitoTracker Green FM (Molecular Probes, Inc.) (mitochondrial dye), LAMP2 (a marker for lysosomes) (34), IBV E, and Alexa-594-conjugated transferrin (Molecular Probes, Inc.) (a marker for early endosomes) (12). Transient transfection of pEGFP-N1/IBV 3a and pEGFP-N1/IBV 3a-GFP by using Fugene 6 allowed visualization of these proteins in HeLa cells.
Colocalization of IBV 3a with LSm1 (a marker for mRNA processing bodies) (20) was assessed by cotransfection of pEGFP-N1/IBV 3a and an RFP-LSm1 construct into cells by using Fugene 6. Colocalization of IBV 3a with CFP-KDEL was assessed in the same manner. Colocalization of IBV 3a with myc-PTE-1 (a marker for peroxisomes) (22) was assessed by cotransfecting cells with pcDNA/IBV 3a and a myc-PTE-1 construct by using LT1. Colocalization of IBV 3a with sec13-GFP (a marker for ER exit sites) (17) was assessed by transfecting HeLa cells that stably expressed sec13-GFP with pcDNA/IBV 3a. Colocalization of IBV 3a with MxA (a smooth ER resident protein) (2) was assessed by transfecting HeLa cells with pEGFP-N1/IBV 3a by using Fugene 6. At 24 h following transfection, cells were treated with 1,000 U/ml of beta interferon (Sigma) for 24 h to induce MxA expression.
Immunofluorescence studies that assessed colocalization between IBV 3a and the following proteins or dyes were carried out with fixation and permeabilization methods that have been described previously (10): ERGIC 53, GM130, p115, p230, transferrin, BiP, rab9, CFP-KDEL, TIA-R, MitoTracker, IBV E, and myc-PTE-1. Briefly, cells were fixed with 3% paraformaldehyde in phosphate-buffered saline for 20 min at room temperature. Permeabilization with 0.5% Triton X-100 for 3 min at room temperature followed. Cells then were stained with the appropriate primary and secondary antibodies. HeLa cells assessed for IBV 3a and IBV 3a-GFP localizations were fixed and permeabilized in the same manner. To assess antibody specificity, anti-IBV 3a antibodies were incubated in a 1:1 (vol:vol) ratio with either 1 mg/ml of IBV 3a peptide (described above) or distilled H2O (control) for 24 h at 4°C.
Cells assessed for TIA-R staining were treated with 0.5 mM sodium arsenite for 1 h at 37°C immediately before fixation. Cells stained for mitochondria were treated with 100 nM MitoTracker dye (according to the manufacturer's protocol) for 1 h at 37°C immediately before fixation. Transferrin staining was accomplished by incubating cells in DMEM without serum for 1 h at 37°C followed by 10 min of incubation at the same temperature with Alexa-594-conjugated transferrin. Cells then were fixed and permeabilized as described above. For assessment of the effects of disrupting transport from the ER to the Golgi apparatus on IBV 3a localization, IBV-infected cells were treated with 1 μg/ml brefeldin A (Molecular Probes, Inc.) for approximately 1 h immediately before fixation and staining for IBV 3a and GM130. For assessment of the effects of disrupting microtubule assembly on IBV 3a localization, IBV-infected cells were treated for 1 h with 5 μg/ml nocodazole (Molecular Probes, Inc.) before being fixed and stained for IBV 3a and GM130. For assessment of the effects of disrupting actin polymerization on IBV 3a localization, IBV-infected cells were treated for 1 h with 1 μg/ml cytochalasin D (Sigma) before being fixed and stained with Texas Red-conjugated phalloidin (Molecular Probes, Inc.) and anti-IBV 3a antibodies.
Cells assessed for RFP-LSm1 staining were fixed and stained as described above; however, permeabilization was performed with 0.2% Triton X-100 for 20 min at room temperature. Cells assessed for IBV 3a colocalization with AP1, MxA, and MPR were fixed as described above; however, cells were permeabilized with 0.05% saponin (EM Biosciences, Inc., San Diego, CA) for 5 min at room temperature. Subsequent antibody incubation of these cells was performed in the presence of 0.05% saponin, 1% BSA, and 0.01 M glycine in phosphate-buffered saline. Selective permeabilization of the plasma membrane was performed with digitonin as previously described (10). Briefly, cells were permeabilized with 25 μg/ml of digitonin (EM Industries, Inc., Gibbstown, NJ) for 5 min on ice. This permeabilization occurred either before or after fixation as described above.
IBV 3a colocalization with LAMP2 and sec13-GFP was assessed by using methods described by Hammond and Glick (17). Briefly, cells were fixed with methanol for 10 min at –20°C. Cells then were cross-linked by incubation for 30 min at room temperature in a phosphate-buffered solution containing 0.1% n-octyl-?-D-glucopyranoside (Boehringer) (PBSO) and 100 μM bis-sulfosuccinimidyl (Pierce Chemical Co., Rockford, IL). Cells then were incubated for 15 min in PBSO containing 0.1 M ethylenediamene (Sigma) before being stained with the appropriate primary and secondary antibodies in PBSO with 1% BSA.
Indirect immunofluorescence images were taken on an Axioskop microscope (Zeiss, Thornwood, NY) with an attached Sensys charge-coupled device camera (Photometrics, Tucson, AZ). IP Lab imaging software (Signal Analytics, Vienna, VA) was used. Confocal imaging was used to assess the colocalization of IBV 3a with CFP-KDEL and MxA. A spinning-disk Ultraview confocal microscope (Perkin Elmer, Inc., Lexington, MA) with an LSI cooled 12-bit camera and Ultraview software were used. Volocity software (Improvision, Inc., Lexington, MA) was used to create one image from approximately 15 (CFP-KDEL and 6 smaller MxA images), 4 (large MxA images), or 50 (rotation images of MxA inset) confocal sections that were each 0.4 (CFP-KDEL experiments) or 0.2 (MxA experiments) μm thick.
Immunoblotting. Immunoblotting was performed as described previously (10). Briefly, Vero cells were infected with IBV at an MOI of 1 before being lysed in a solution containing 4% sodium dodecyl sulfate (SDS), 30% glycerol, 1% bromophenol blue, 10% 2-mercaptoethanol, and 0.1 M Tris-HCl (pH 8). Samples then were run on a 17.5% polyacrylamide denaturing gel before being transferred to a polyvinylidene difluoride membrane (Millipore Corporation, Billerica, MA). The membrane was blocked for 30 min at room temperature with 5% nonfat milk dissolved in TTBS (150 mM NaCl, 0.05% Tween 20, 10 mM Tris-HCl [pH 7.4]). Primary antibody incubation was done overnight at 4°C in TTBS plus 1% nonfat milk. Following extensive washing, the membrane was incubated with horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature in TTBS plus 5% nonfat milk. Analysis was performed by using enhanced chemiluminescence (ECL) Western blotting detection reagents (Amersham) and a VersaDoc model 5000 imaging system (Bio-Rad) with an attached cooled charge-coupled device AF Nikkor camera (Nikon, Inc., Melville, NY). Quantitation was performed by using Quantity One software (Bio-Rad).
In vitro translation assays. A TNT quick coupled transcription/translation T7 system (Promega Corporation, Madison, WI) was used according to the manufacturer's protocol in the presence or absence of canine pancreatic microsomal membranes (Promega). Briefly, pBS/IBV 3a, pBS/truncated 3a, or pBS/3a VLS-VLI was incubated for 90 min at 30°C in a transcription/translation master mix containing 35S-Redivue-methionine (Amersham). For each plasmid, two reaction mixtures were made, one that contained microsomal membranes and one that did not. Reaction mixtures were suspended in swelling buffer (15 mM NaCl, 1 mM MgCl2, and 10 mM Tris-HCl [pH 7.4]) and then subjected to ultracentrifugation at 136,000 x g for 1 h through a 10% sucrose cushion in swelling buffer. The resulting supernatants and pellets were immunoprecipitated as described previously (32) and analyzed by SDS-polyacrylamide gel electrophoresis (PAGE) and 35S quantitation.
Alkaline carbonate extraction. Alkaline carbonate extraction was performed as described previously (10). Briefly, cells were plated on 6-cm dishes for 24 h prior to transfection or infection. HeLa cells were transfected with either pEGFP-N1/IBV 3a or pEGFP-N1/IBV 3a-GFP as described above. Vero cells were infected with IBV at an MOI of 1 for 31 h. HeLa and Vero cells were labeled for 1 h at 37°C with 35S-methionine-cysteine (Pro-mix; Amersham) and homogenized in a swelling solution containing 8% sucrose, 15 mM NaCl, 1 mM MgCl2, and 10 mM Tris-HCl (pH 7.4) with 60 strokes of a tightly fitting Dounce homogenizer. Following the pelleting of nuclei, microsomal membranes were treated with 0.1 M Na2CO3 (pH 11.5). Treatment with 0.1 M NaCl or a detergent solution containing 0.05 M Tris-HCl (pH 8), 0.0625 M EDTA (pH 8), 0.4% deoxycholic acid, and 1% octylphenylpolyethylene glycol served as a control. Microsomal membranes then were pelleted by ultracentrifugation at 136,000 x g for 1 h. The resulting supernatants and pellets were immunoprecipitated as described previously (32) and analyzed by SDS-PAGE and phosphorimaging. Quantitation of 35S was done by exposing gels to a K-HD imaging screen (Bio-Rad) followed by scanning with personal molecular imager FX (Bio-Rad). Quantity One software (Bio-Rad) was used to quantitate images. The percent total calculation was performed with the following formula, where S is supernatant and P is pellet: [S/(S + P)] x 100 or [P/(S + P)] x 100.
RESULTS
IBV 3a expression in infected and transfected cells. A polyclonal antibody to IBV 3a was generated using a synthetic peptide corresponding to the C-terminal 14 amino acids of IBV 3a. Before use in expression studies, the titer of this antibody was compared to the titers of IBV E and IBV M antibodies (10, 32) with ELISAs. The IBV 3a antibody titer was found to be approximately equal to that of the IBV M antibody (see Materials and Methods). The IBV M antibody titer had been previously shown to be approximately one-third that of the IBV E antibody titer. Vero cells were infected with IBV for the times indicated, harvested, and analyzed by Western blot (Fig. 1A). An approximately 6-kDa band corresponding to IBV 3a was detected in infected cells. This band was absent from uninfected cells, indicating that the anti-IBV 3a antibody was specific. Quantitation indicated that IBV 3a was expressed later in infection than IBV M, and at much lower levels than either IBV E or IBV M. IBV 3a was approximately 13-fold less abundant than IBV M at 24 hpi. However, this quantitation is a rough estimate since antibody titers were determined by ELISAs, not immunoblotting.
The anti-IBV 3a antibody then was used to explore the intracellular localization of IBV 3a. Following infection with IBV for approximately 15 h, IBV 3a and IBV E were visualized by indirect immunofluorescence microscopy (Fig. 1B). IBV 3a showed a distinct localization compared to IBV E, which is targeted to the Golgi complex (10). IBV 3a showed both a diffuse localization and sharp puncta of approximately 500 nm to 1.5 μm in size. Similar localization was observed in cells transiently transfected with a plasmid encoding IBV 3a (Fig. 1C), indicating that the intracellular localization of IBV 3a is independent of other viral proteins. Uninfected or nontransfected cells on the same coverslip served as negative controls for antibody staining (see Fig. 6; also, data not shown). Preincubation of the IBV 3a antibody with the C-terminal IBV 3a peptide eliminated the signal, indicating that the antibody was specific to IBV 3a (Fig. 1C). IBV 3a protein was not present in purified virions released from IBV-infected, 35S-methionine-cysteine-labeled cells (data not shown), indicating that it is not a structural protein. This is consistent with previous findings (29).
A potential signal sequence cleavage site in IBV 3a is not used, and a pool of IBV 3a associates with membranes. Hydropathy analysis of the IBV 3a sequence indicated a highly hydrophobic, potential membrane-spanning segment at the N terminus (Fig. 2A). Some sequence analysis programs also predicted that the hydrophobic region was a signal sequence, cleavable after S22. To test whether IBV 3a was cleaved at S22, two mutants of IBV 3a were created: a truncated version of IBV 3a corresponding to amino acids 23 to 57 (truncated 3a), and an uncleavable version of IBV 3a in which S22 was mutated to an isoleucine (3a VLS-VLI). Cells were transiently transfected with a plasmid encoding wild-type IBV 3a, truncated 3a, or 3a VLS-VLI. Cells were labeled with 35S-methionine-cysteine and immunoprecipitated with anti-IBV 3a antibodies. Wild-type IBV 3a comigrated with 3a VLS-VLI but not with truncated 3a (Fig. 2B). These results indicated that the potential signal sequence cleavage site in IBV 3a was not used. IBV 3a from infected cells also comigrated with 3a VLS-VLI and not with truncated 3a (data not shown).
Since the hydrophobic segment at the N terminus of IBV 3a was not a cleaved signal sequence, we predicted that this sequence might serve as a membrane anchor. To assess this possibility, microsomal membranes prepared from IBV-infected, 35S-methionine-cysteine-labeled cells were extracted with alkaline carbonate. This treatment induces membrane sheets, releasing lumenal contents and peripheral membrane proteins. Microsomes were treated with Na2CO3 (pH 11.5), 0.1 M NaCl, or detergent as a control. Following these treatments, membranes were pelleted via ultracentrifugation. The resulting supernatants and pellets then were immunoprecipitated with anti-IBV 3a antibodies and analyzed by SDS-PAGE. IBV E, a known transmembrane protein (10), was used as a control. We found that approximately 60% of IBV 3a was cytoplasmic and 40% of IBV 3a was tightly associated with membranes (Fig. 2C). IBV 3a from transiently transfected cells behaved similarly (data not shown).
A pool of IBV 3a associates with membranes in vitro. To confirm the partial membrane association of IBV 3a, in vitro translation in the presence or absence of canine pancreatic microsomal membranes was performed. We used truncated IBV 3a and IBV E as controls. Following in vitro translation, samples were subjected to ultracentrifugation to pellet microsomal membranes. The resulting supernatants and pellets then were immunoprecipitated and analyzed by SDS-PAGE. IBV E was poorly translated in the absence of microsomes, but was mostly in the pellet when microsomes were included. Similarly, there was an increase of IBV 3a in the pellet fraction when microsomal membranes were included during translation (Fig. 3). These data indicated that IBV 3a was capable of associating with membranes. There was no increase of truncated 3a in the pellet with the addition of microsomal membranes. Thus, the N terminus of IBV 3a was responsible for its membrane association. Additionally, although more IBV 3a was found in the pellet when microsomes were present, a significant portion of IBV 3a was still found in the supernatant in these conditions. Thus, these in vitro data confirmed previous alkaline carbonate results from infected cells (Fig. 2C) showing partial membrane association of IBV 3a.
GFP-tagged IBV 3a associates with membranes more efficiently than wild-type IBV 3a. The results indicating that a pool of IBV 3a was cytoplasmic while another pool of IBV 3a was membrane bound were intriguing. The membrane-associated pool appears to enter the secretory pathway (see below). However, the short length of IBV 3a (57 amino acids) may preclude efficient interaction with signal recognition particle for cotranslational translocation into the ER (19). We asked if extending the length of IBV 3a (with a C-terminal GFP tag) increased the efficiency of IBV 3a membrane insertion. To ensure that IBV 3a-GFP puncta represented similar structures to IBV 3a puncta, cells were doubly transfected with IBV 3a-GFP and IBV 3a constructs. The anti-IBV 3a antibody does not bind to IBV 3a-GFP, presumably due to interference of the GFP tag with the C-terminal epitope recognized by the IBV 3a antibodies. Indirect immunofluorescence with anti-IBV 3a and anti-GFP antibodies demonstrated that IBV 3a and IBV 3a-GFP largely colocalize (data not shown). Thus, IBV 3a-GFP puncta were similar to those containing wild-type IBV 3a.
Interestingly, cells transiently transfected with a plasmid encoding GFP-tagged IBV 3a showed reduced diffuse staining while maintaining distinct puncta (Fig. 4A). To determine if this finding reflected more efficient membrane association, alkaline carbonate extraction was performed on Vero cells transiently transfected with a plasmid encoding either wild-type IBV 3a or GFP-tagged IBV 3a. In transiently transfected cells, the percentage of IBV 3a that was tightly associated with membranes was approximately equal to the percentage seen in infected cells. However, GFP-tagged IBV 3a more efficiently associated with membranes than wild-type IBV 3a in both physiological salt and alkaline carbonate conditions (Fig. 4B). These data indicated that the short length of IBV 3a is responsible for its inefficient insertion into membranes.
The diffusely staining pool of IBV 3a is soluble. To determine whether the diffusely stained pool of IBV 3a observed by indirect immunofluorescence was cytoplasmic, IBV-infected Vero cells were permeabilized with digitonin before fixation. Under the conditions used, digitonin selectively permeabilizes the plasma membrane and allows a significant portion of the cytoplasmic contents to be washed from cells (27). BiP was used as a control to confirm that intracellular membranes were not permeabilized. Digitonin permeabilization prior to fixation resulted in a substantial reduction in the diffusely stained pool of IBV 3a (Fig. 5A). This loss did not occur in cells that were fixed before digitonin treatment. BiP was not released with either condition, since it was clearly detected after additional permeabilization with Triton X-100. Thus, the diffusely staining pool of IBV 3a was cytoplasmic. This conclusion was substantiated by subjecting the digitonin-released contents to ultracentrifugation. IBV-infected Vero cells were labeled with 35S-methionine-cysteine before treatment with digitonin. The cellular contents released upon digitonin treatment then were collected and subjected to ultracentrifugation. The released portion of IBV 3a was soluble (Fig. 5B).
IBV 3a topology. Digitonin permeabilization was also used to determine the orientation of the membrane-associated IBV 3a. IBV-infected Vero cells were fixed and permeabilized with either Triton X-100 (which permeabilizes all cell membranes) or digitonin. BiP was again used as a control. The localization pattern of IBV 3a in cells permeabilized with Triton X-100 was similar to the pattern seen in digitonin-permeabilized cells (Fig. 6), indicating that the C terminus of both cytoplasmic and membrane-bound IBV 3a was located in the cytoplasm. This result, combined with the results of the hydopathy analysis, suggests that for the pool of IBV 3a that is membrane associated, the amino terminus is exposed to a lumenal compartment, the N-terminal hydrophobic region crosses the membrane once, and the C terminus is located in the cytoplasm. However, these data do not rule out the possibility that some portion of membrane-associated IBV 3a has the opposite topology, with the C terminus in a lumenal compartment.
IBV 3a puncta localize to the smooth ER and are distinct from other membrane-bound organelles. To determine the exact cellular location of IBV 3a puncta, colocalization immunofluorescence studies with various cellular marker proteins were performed on IBV-infected Vero cells. We first analyzed potential colocalization with punctate organelles. The IBV 3a puncta were distinct from endosomes, lysosomes, peroxisomes, and ER exit sites (Table 1). There was also no colocalization with Golgi membranes or mitochondria. IBV 3a puncta were also distinct from cytoplasmic foci such as stress granules (marked by TIA-R) and mRNA processing bodies (marked by LSm1). Furthermore, the IBV 3a puncta were unaffected by drugs that disrupted transport from the ER to the Golgi apparatus (brefeldin A), actin polymerization (cytochalasin D), and microtubule assembly (nocodazole) (data not shown). However, we did find that the IBV 3a puncta lined up along ER tubules labeled with CFP-KDEL (Fig. 7).
IBV 3a puncta did not appear to colocalize with BiP when analyzed by indirect immunofluorescence. Therefore, we decided to further explore the localization of IBV 3a to ER membranes by testing a marker for smooth ER, MxA (2). This experiment could not be done in infected cells, since we could not detect the MxA protein in Vero cells. In addition, since both anti-IBV 3a and anti-MxA antibodies were prepared in rabbits, we used tagged IBV 3a to allow double labeling. HeLa cells were transfected with a plasmid encoding IBV 3a-GFP 24 h prior to being treated with beta interferon (to induce MxA expression) for 24 h. Cells then were fixed and double-labeled for IBV 3a-GFP and MxA and analyzed by confocal microscopy. Beta interferon treatment did not change the distribution of IBV 3a-GFP puncta. The results indicated that IBV 3a-GFP puncta lined, contacted, or were surrounded by MxA tubules (Fig. 8). Using software that can rotate stacks of confocal images in three dimensions, IBV 3a-GFP puncta were assessed for their proximity to MxA-positive tubules. Analysis of 349 puncta from six different cells showed that 99.1% were proximal (within half the diameter of an IBV 3a-GFP punctum) to MxA-stained structures, suggesting that IBV 3a puncta arise from smooth ER tubules. A gallery of images from four different cells is shown in Fig. 8B.
DISCUSSION
IBV 3a is a highly conserved protein among group 3 coronaviruses (21), suggesting it possesses a critical function during viral infection. Our results demonstrated that IBV 3a is expressed in infected cells (Fig. 1A and B), similar to earlier studies by Liu et al. (28). Our data also demonstrated that IBV 3a is expressed at much lower levels than either IBV E or IBV M during viral infection, that it is expressed later in infection than IBV M, and that it is not incorporated into virions. In both infected and transfected cells, IBV 3a was found to localize diffusely to the cytoplasm as well as to membrane-associated puncta (Fig. 1B). These IBV 3a-containing puncta were distinct from early and late endosomes, lysosomes, peroxisomes, ER exit sites, the Golgi complex, the ERGIC, stress granules, and mRNA processing bodies (Table 1). Ultimately, confocal microscopy demonstrated that IBV 3a localized to smooth ER domains, which are distinct from the rough ER (5). Additionally, IBV 3a in membrane-associated puncta was oriented with its C terminus exposed to the cytoplasm. These results suggest that membrane-associated IBV 3a spans the membrane once, with its N terminus exposed to the ER lumen (Fig. 6).
Our results raise several interesting questions. Why does a pool of IBV 3a localize to the cytoplasm while another pool localizes to smooth ER membranes? One possible explanation is that the short length of IBV 3a (57 amino acids) precludes efficient interaction with signal recognition particle for translocation into ER membranes (19). This idea is supported by our finding that lengthening IBV 3a by the addition of a GFP tag to the C terminus increased the proportion of IBV 3a that was tightly associated with membranes (Fig. 4B). These results might indicate that only small amounts of IBV 3a are needed at smooth ER membranes. Perhaps the energy expended in producing a protein, only a portion of which is properly targeted, is offset by the energetic advantage of a more streamlined genome. Alternatively, these results could indicate that too much IBV 3a localized to smooth ER membranes is detrimental to virus infection; therefore, IBV 3a evolved so that relatively low amounts reach membranes. This later possibility is intriguing in light of alkaline carbonate extraction data for IBV E versus GFP-tagged IBV 3a. Unlike IBV E (Fig. 2C), a known transmembrane protein, the membrane association of GFP-tagged IBV 3a was not complete. Therefore, the 5' sequence of IBV 3a may naturally possess a low affinity for signal recognition particle, indicating that the virus has evolved at least two mechanisms to ensure that only limited amounts of IBV 3a reach smooth ER membranes. However, the previous arguments assume that only membrane-bound IBV 3a possesses a function critical for virus infection. In fact, IBV 3a may have evolved its variable membrane association characteristics if both cytoplasmic and membrane-bound IBV 3a possess important, possibly distinct, functions. Future experiments will address the consequences to the viral life cycle of replacing the wild-type IBV 3a sequence with that of the extended or truncated versions using an infectious clone of IBV (49).
Also of interest is the close proximity of IBV 3a puncta and MxA containing smooth ER tubules (Fig. 8). MxA is a member of the dynamin family of proteins (reviewed in reference 16). This family is known to promote vesicle formation, presumably through the intrinsic GTPase activity of its family members. Similarly, MxA has been shown to tubulate lipids both in vitro and in vivo. However, MxA is unique from other known dynamin family members because its expression is induced by alpha and beta interferons. Consistent with this type of inducible expression, MxA possesses antiviral activity against a number of different viruses that have either RNA genomes or genomes that go through a RNA intermediate. In some of these cases, MxA is thought to function by binding and sequestering essential viral proteins, thus preventing their proper localization (40). The close localization of IBV 3a to MxA-containing membranes suggests that IBV 3a may interact with and potentially abrogate the antiviral activity of MxA. However, the effects of MxA on IBV replication have not been established. Also, recent experiments demonstrated that stable MxA expression in Vero cells (which lack the ability to respond to alpha and beta interferons) does not negatively affect replication of the severe acute respiratory syndrome coronavirus (44). Therefore, the localization of IBV 3a may be unrelated to interaction with MxA. Future biochemical experiments will assess physical interaction of MxA with IBV 3a. Additionally, generation of an infectious clone of IBV lacking functional IBV 3a may be useful in determining functional interactions between IBV 3a and MxA.
If IBV 3a does not interact with MxA, then IBV 3a may be involved with another function of smooth ER membrane domains. An interesting possibility is suggested by recent data from Ng and Liu (37), who suggested that IBV appeared to anchor its replication machinery to ER membranes. If virus replication were found to occur at vesicles derived from smooth ER membranes, then IBV 3a might participate in the genesis or functioning of these complexes. Alternatively, IBV 3a might take part in other functions known to occur at smooth ER membranes, such as lipid metabolism (6, 11), atypical protein trafficking events (4, 26, 39), or an as-yet-undiscovered process occurring at these membranes. Analysis in the context of an IBV 3a-null infectious clone will likely distinguish among these possibilities and help to elucidate the specific function of IBV 3a in virus infection. Additionally, IBV 3a may be useful in shedding light on the dynamics of smooth ER membranes.
ACKNOWLEDGMENTS
This work was supported by National Institutes of Health grant GM64647.
We thank Emily Corse for the generation of IBV 3a antibodies and plasmid pBS/IBV 3a, Stuart Hicks for the HeLa cell line stably expressing sec13-GFP, and all members of the Machamer Laboratory for helpful comments and stimulating discussions. We also thank M. McNiven (Mayo Clinic, Rochester, MN) for the generous gift of anti-MxA antibodies.
REFERENCES
Abraham, D. J., and A. J. Leo. 1987. Extension of the fragment method to calculate amino acid zwitterion and side chain partition coefficients. Proteins 2:130-152.
Accola, M. A., B. Huang, A. Al Masri, and M. A. McNiven. 2002. The antiviral dynamin family member, MxA, tubulates lipids and localizes to the smooth endoplasmic reticulum. J. Biol. Chem. 277:21829-21835.
Anderson, P., and N. Kedersha. 2002. Stressful initiations. J. Cell Sci. 115:3227-3234.
Benlimame, N., P. U. Le, and I. R. Nabi. 1998. Localization of autocrine motility factor receptor to caveolae and clathrin-independent internalization of its ligand to smooth endoplasmic reticulum. Mol. Biol. Cell 9:1773-1786.
Benlimame, N., D. Simard, and I. R. Nabi. 1995. Autocrine motility factor receptor is a marker for a distinct membranous tubular organelle. J. Cell Biol. 129:459-471.
Bishop, W. R., and R. M. Bell. 1988. Assembly of phospholipids into cellular membranes: biosynthesis, transmembrane movement and intracellular translocation. Annu. Rev. Cell Biol. 4:579-610.
Bole, D. G., R. Dowin, M. Doriaux, and J. D. Jamieson. 1989. Immunocytochemical localization of BiP to the rough endoplasmic reticulum: evidence for protein sorting by selective retention. J. Histochem. Cytochem. 37:1817-1823.
Boursnell, M. E., M. M. Binns, and T. D. Brown. 1985. Sequencing of coronavirus IBV genomic RNA: three open reading frames in the 5' ‘unique’ region of mRNA D. J. Gen. Virol. 66:2253-2258.
Brown, T. D. K., and I. Brierley. 1995. The coronavirus nonstructural proteins, p. 191-217. In S. G. Siddell (ed.), The Coronaviridae. Plenum Press, New York, N.Y.
Corse, E., and C. E. Machamer. 2000. Infectious bronchitis virus E protein is targeted to the Golgi complex and directs release of virus-like particles. J. Virol. 74:4319-4326.
Daleke, D. L., and J. V. Lyles. 2000. Identification and purification of aminophospholipid flippases. Biochim. Biophys. Acta 1486:108-127.
Daro, E., P. van der Sluijs, T. Galli, and I. Mellman. 1996. Rab4 and cellubrevin define different early endosome populations on the pathway of transferrin receptor recycling. Proc. Natl. Acad. Sci. USA 93:9559-9564.
de Haan, C. A., P. S. Masters, X. Shen, S. Weiss, and P. J. Rottier. 2002. The group-specific murine coronavirus genes are not essential, but their deletion, by reverse genetics, is attenuating in the natural host. Virology 296:177-189.
Dintzis, S. M., V. E. Velculescu, and S. R. Pfeffer. 1994. Receptor extracellular domains may contain trafficking information. Studies of the 300-kDa mannose 6-phosphate receptor. J. Biol. Chem. 269:12159-12166.
Fuerst, T. R., E. G. Niles, F. W. Studier, and B. Moss. 1986. Eukaryotic transient-expression system based on recombinant vaccinia virus that synthesizes bacteriophage T7 RNA polymerase. Proc. Natl. Acad. Sci. USA 83:8122-8126.
Haller, O., and G. Kochs. 2002. Interferon-induced mx proteins: dynamin-like GTPases with antiviral activity. Traffic 3:710-717.
Hammond, A. T., and B. S. Glick. 2000. Dynamics of transitional endoplasmic reticulum sites in vertebrate cells. Mol. Biol. Cell 11:3013-3030.
Horton, P., and K. Nakai. 1997. Better prediction of protein cellular localization sites with the k nearest neighbors classifier. Proc. Int. Conf. Intell. Syst. Mol. Biol. 5:147-152.
Ibrahimi, I. M., D. Cutler, D. Stueber, and H. Bujard. 1986. Determinants for protein translocation across mammalian endoplasmic reticulum. Membrane insertion of truncated and full-length prelysozyme molecules. Eur. J. Biochem. 155:571-576.
Ingelfinger, D., D. J. Arndt-Jovin, R. Luhrmann, and T. Achsel. 2002. The human LSm1-7 proteins colocalize with the mRNA-degrading enzymes Dcp1/2 and Xrnl in distinct cytoplasmic foci. RNA 8:1489-1501.
Jia, W., and S. A. Naqi. 1997. Sequence analysis of gene 3, gene 4 and gene 5 of avian infectious bronchitis virus strain CU-T2. Gene 189:189-193.
Jones, J. M., K. Nau, M. T. Geraghty, R. Erdmann, and S. J. Gould. 1999. Identification of peroxisomal acyl-CoA thioesterases in yeast and humans. J. Biol. Chem. 274:9216-9223.
Klumperman, J., J. K. Locker, A. Meijer, M. C. Horzinek, H. J. Geuze, and P. J. Rottier. 1994. Coronavirus M proteins accumulate in the Golgi complex beyond the site of virion budding. J. Virol. 68:6523-6534.
Kooy, J., B. H. Toh, J. M. Pettitt, R. Erlich, and P. A. Gleeson. 1992. Human autoantibodies as reagents to conserved Golgi components. Characterization of a peripheral, 230-kDa compartment-specific Golgi protein. J. Biol. Chem. 267:20255-20263.
Kyte, J., and R. F. Doolittle. 1982. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157:105-132.
Le, P. U., and I. R. Nabi. 2003. Distinct caveolae-mediated endocytic pathways target the Golgi apparatus and the endoplasmic reticulum. J. Cell Sci. 116:1059-1071.
Lineberger, B., D. D. Dawicki, K. C. Agarwal, N. Kessimian, and M. Steiner. 1989. Permeabilization of platelets: an investigation of biochemical, ultrastructural and functional aspects. Biochim. Biophys. Acta 1012:36-45.
Liu, D. X., D. Cavanagh, P. Green, and S. C. Inglis. 1991. A polycistronic mRNA specified by the coronavirus infectious bronchitis virus. Virology 184:531-544.
Liu, D. X., and S. C. Inglis. 1991. Association of the infectious bronchitis virus 3c protein with the virion envelope. Virology 185:911-917.
Liu, D. X., and S. C. Inglis. 1992. Internal entry of ribosomes on a tricistronic mRNA encoded by infectious bronchitis virus. J. Virol. 66:6143-6154.
Lombardi, D., T. Soldati, M. A. Riederer, Y. Goda, M. Zerial, and S. R. Pfeffer. 1993. Rab9 functions in transport between late endosomes and the trans Golgi network. EMBO J. 12:677-682.
Machamer, C. E., and J. K. Rose. 1987. A specific transmembrane domain of a coronavirus E1 glycoprotein is required for its retention in the Golgi region. J. Cell Biol. 105:1205-1214.
Mancini, M., C. E. Machamer, S. Roy, D. W. Nicholson, N. A. Thornberry, L. A. Casciola-Rosen, and A. Rosen. 2000. Caspase-2 is localized at the Golgi complex and cleaves golgin-160 during apoptosis. J. Cell Biol. 149:603-612.
Mane, S. M., L. Marzella, D. F. Bainton, V. K. Holt, Y. Cha, J. E. Hildreth, and J. T. August. 1989. Purification and characterization of human lysosomal membrane glycoproteins. Arch. Biochem. Biophys. 268:360-378.
Nakamura, N., C. Rabouille, R. Watson, T. Nilsson, N. Hui, P. Slusarewicz, T. E. Kreis, and G. Warren. 1995. Characterization of a cis-Golgi matrix protein, GM130. J. Cell Biol. 131:1715-1726.
Navas-Martin, S. R., and S. Weiss. 2004. Coronavirus replication and pathogenesis: implications for the recent outbreak of severe acute respiratory syndrome (SARS), and the challenge for vaccine development. J. Neurovirol. 10:75-85.
Ng, L. F., and D. X. Liu. 2002. Membrane association and dimerization of a cysteine-rich, 16-kilodalton polypeptide released from the C-terminal region of the coronavirus infectious bronchitis virus 1a polyprotein. J. Virol. 76:6257-6267.
Pearse, B. M., and M. S. Robinson. 1990. Clathrin, adaptors, and sorting. Annu. Rev. Cell Biol. 6:151-171.
Pelkmans, L., J. Kartenbeck, and A. Helenius. 2001. Caveolar endocytosis of simian virus 40 reveals a new two-step vesicular-transport pathway to the ER. Nat. Cell Biol. 3:473-483.
Reichelt, M., S. Stertz, J. Krijnse-Locker, O. Haller, and G. Kochs. 2004. Missorting of LaCrosse virus nucleocapsid protein by the interferon-induced MxA GTPase involves smooth ER membranes. Traffic 5:772-784.
Schweizer, A., J. A. Fransen, T. Bachi, L. Ginsel, and H. P. Hauri. 1988. Identification, by a monoclonal antibody, of a 53-kD protein associated with a tubulo-vesicular compartment at the cis-side of the Golgi apparatus. J. Cell Biol. 107:1643-1653.
Shen, S., Z. L. Wen, and D. X. Liu. 2003. Emergence of a coronavirus infectious bronchitis virus mutant with a truncated 3b gene: functional characterization of the 3b protein in pathogenesis and replication. Virology 311:16-27.
Siddell, S. G. 1995. The Coronaviridae: an introduction, p. 1-10. In S. G. Siddell (ed.), The Coronaviridae. Plenum Press, New York, N.Y.
Spiegel, M., A. Pichlmair, E. Muhlberger, O. Haller, and F. Weber. 2004. The antiviral effect of interferon-beta against SARS-coronavirus is not mediated by MxA protein. J. Clin. Virol. 30:211-213.
Sturman, L. S., K. V. Holmes, and J. Behnke. 1980. Isolation of coronavirus envelope glycoproteins and interaction with the viral nucleocapsid. J. Virol. 33:449-462.
Taupin, J. L., Q. Tian, N. Kedersha, M. Robertson, and P. Anderson. 1995. The RNA-binding protein TIAR is translocated from the nucleus to the cytoplasm during Fas-mediated apoptotic cell death. Proc. Natl. Acad. Sci. USA 92:1629-1633.
Van der Most, R. G., and W. J. Spaan. 1995. Coronavirus replication, transcription, and RNA recombination, p. 11-31. In S. G. Siddell (ed.), The Coronaviridae. Plenum Press, New York, N.Y.
Waters, M. G., D. O. Clary, and J. E. Rothman. 1992. A novel 115-kD peripheral membrane protein is required for intercisternal transport in the Golgi stack. J. Cell Biol. 118:1015-1026.
Youn, S., J. L. Leibowitz, and E. W. Collisson. 2005. In vitro assembled, recombinant infectious bronchitis viruses demonstrate that the 5a open reading frame is not essential for replication. Virology 332:206-215.(Amanda R. Pendleton and C)
ABSTRACT
All coronaviruses possess small open reading frames (ORFs) between structural genes that have been hypothesized to play important roles in pathogenesis. Infectious bronchitis virus (IBV) ORF 3a is one such gene. It is highly conserved among group 3 coronaviruses, suggesting that it has an important function in infection. IBV 3a protein is expressed in infected cells but is not detected in virions. Sequence analysis predicted that IBV 3a was a membrane protein; however, only a fraction behaved like an integral membrane protein. Microscopy and immunoprecipitation studies demonstrated that IBV 3a localized to the cytoplasm in a diffuse pattern as well as in sharp puncta in both infected and transfected cells. These puncta did not overlap cellular organelles or other punctate structures. Confocal microscopy demonstrated that IBV 3a puncta lined up along smooth endoplasmic reticulum (ER) tubules and, in a significant number of instances, were partially surrounded by these tubules. Our results suggest that IBV 3a is partially targeted to a novel domain of the smooth ER.
INTRODUCTION
Members of the family Coronaviridae are positive-sense, single-stranded RNA viruses (36). They recently received much attention due to a newly discovered coronavirus implicated in severe acute respiratory syndrome. Coronaviruses have been divided into three groups based on serological cross-reactivity and sequence analysis. However, despite their division into separate groups, coronaviruses share many common characteristics. For example, coronaviruses from all three groups possess a helical capsid structure (43) and derive their viral envelopes by budding into the endoplasmic reticulum (ER)-Golgi-intermediate compartment (ERGIC) (23). They also share similar replication strategies (36). Once the virus enters the cell, the RNA genome serves as a template for the translation of two polyproteins, 1a and 1ab, containing the viral RNA-dependent RNA polymerase (47). A set of negative-stranded, subgenomic RNAs then are transcribed (36). These negative-stranded RNAs are used to create positive-stranded, subgenomic RNAs. One to three of the 5'-most open reading frames (ORFs) then are translated from these positive-stranded, subgenomic RNAs.
Coronaviruses encode four structural proteins: spike (S), envelope (E), membrane (M), and nucleocapsid (N). The S protein, found in the virus envelope, is responsible for binding of the virus to its receptor and subsequent fusion of the viral envelope with cellular membranes. The E and M proteins are also components of the viral envelope. Although their functions are still under investigation, they are thought to be important for virus budding. Finally, the N protein binds to both the viral genome and the M protein, allowing incorporation of the genome into mature virions (45). In addition to these well-studied structural proteins, coronaviruses possess additional ORFs interspersed throughout their genomes (9). Some of these ORFs are known to produce proteins during viral infection. However, these proteins are largely thought to be involved in pathogenesis rather than virus replication or assembly. For example, ORF 5a in infectious bronchitis virus (IBV), a group 3 coronavirus, was recently found to be nonessential for virus replication in cell cultures (49). Additionally, the deletion of several ORFs (2a, hemagglutinin-esterase, 4a, 4b, and 5a) in the group 2 coronavirus murine hepatitis virus appears to have no effect on virus replication in cell cultures (13). However, these deletions in murine hepatitis virus did appear to reduce virus virulence during infection of the natural host.
One ORF-containing region, region II, lies between the S and E genes and is present in all coronaviruses. IBV region II produces two small proteins, 3a and 3b, during infection (28). These proteins are translated from subgenomic mRNA 3, a functionally polycistronic mRNA, via leaky ribosomal scanning (30). The E structural protein (originally called 3c) is translated from the same mRNA by an internal ribosome entry site. While region II is not well conserved among different coronavirus groups, it is very well conserved within the group 3 coronaviruses. One study which compared the polypeptide sequences of group 3 field isolates from different continents and decades found the similarities of IBV 3a to be 81 to 86.2% and of IBV 3b to be 87.5 to 95.4% (21). Recently, a study demonstrated the emergence of a truncated form of IBV 3b in cell cultures, indicating that IBV 3b may not be essential for virus replication (42). However, the persistent presence and high degree of conservation of IBV 3a and IBV 3b despite unfavorable selective pressure suggest that they are important in some aspect of productive viral infection.
Here we report the characterization of the IBV 3a protein. We found that one pool of the IBV 3a protein is cytoplasmic and that another pool is tightly associated with membranes. The membrane-associated fraction localizes to puncta that line smooth ER membranes, suggesting a potentially novel function for this protein.
MATERIALS AND METHODS
Cells and viruses. Vero and HeLa cells were maintained in high-glucose Dulbecco's modified Eagle's medium (DMEM) (Invitrogen Life Technologies, Carlsbad, CA) supplemented with 5 or 10% fetal calf serum (Atlanta Biologicals, Norcross, GA) and 0.1 mg/ml Normocin (Invivogen, San Diego, CA). The previously described Vero-adapted Beaudette strain of IBV (32) was used in all IBV infection experiments, and titers were determined with Vero cells. A vaccinia virus encoding T7 polymerase (vTF7-3) was obtained from Bernard Moss (National Institutes of Health, Bethesda, MD) (15). HeLa cells stably expressing sec13-green fluorescent protein (GFP) were created by previously described methods (33).
IBV 3a sequence analysis. The IBV 3a sequence (Beaudette strain) has been published (8). Hydropathy analysis was carried out by methods developed by Kyte and Doolittle (25). Prediction of a signal sequence cleavage site in IBV 3a was obtained from the PSORT II website (http://psort.ims.u-tokyo.ac.jp/form2.html) developed by Horton and Nakai (18).
Plasmids. The IBV 3a gene was cloned by reverse transcription-PCR of mRNA isolated from IBV-infected Vero cells. Briefly, primers containing an EcoRI restriction site in front of the 5' coding region of IBV 3a and 3' primers containing a BamHI restriction site behind the stop codon of IBV 3a were used to amplify the IBV 3a gene by reverse transcription-PCR. These two restriction sites then were used to clone IBV 3a into pBluescript SK (pBS) (Stratagene, La Jolla, CA) behind the T7 promoter. Dideoxy sequencing confirmed that the IBV 3a sequence in this construct (called pBS/IBV 3a) matched the known IBV 3a Beaudette strain sequence. A truncated version of IBV 3a (pBS/truncated 3a) that consisted of C23 to the end of the protein was created by PCR amplification. Briefly, a primer containing an EcoRI site followed by a Kozak sequence and an AUG start codon was placed 5' of the codon for C23. The same 3' primer used for making pBS/IBV 3a was used in this reaction. The PCR product then was inserted into plasmid pBS by using the EcoRI and BamHI restriction sites. A version of IBV 3a predicted to be uncleavable (1) was generated by QuikChange mutagenesis (Stratagene, La Jolla, CA) according to the manufacturer's protocol with pBS/IBV 3a as a template, changing Ser22 to an Ile (pBS/3a VLS-VLI).
The EcoRI and NotI restriction sites in pBS/IBV 3a were used to subclone IBV 3a into pcDNA3.1/Myc-His(+)A (Invitrogen) behind the cytomegalovirus promoter, creating pcDNA/IBV 3a. The EcoRI and BamHI restriction sites in pBS/IBV 3a were used to subclone IBV 3a into pEGFP-N1 (BD Biosciences Clontech, San Diego, CA) behind the cytomegalovirus promoter, creating pEGFP-N1/IBV 3a. A C-terminally GFP-tagged version of IBV 3a was created by performing QuikChange mutagenesis on construct pEGFP-N1/IBV 3a, creating pEGFP-N1/IBV 3a-GFP.
A myc-tagged peroxisome thioesterase 1 (PTE-1) construct was a gift from S. J. Gould (Johns Hopkins School of Medicine, Baltimore, MD). A construct containing human red fluorescent protein (RFP)-tagged Sm-like protein 1 (LSm1) was a gift from H. C. Dietz (Johns Hopkins School of Medicine, Baltimore, MD). A plasmid encoding sec13-GFP was received from F. Gorelick (Yale University School of Medicine, New Haven, CT). A plasmid (pCFP-ER) encoding cyan fluorescent protein with the ER localization signal KDEL (CFP-KDEL) was obtained from BD Biosciences Clontech.
Antibodies. A peptide corresponding to the C-terminal 14 amino acids (with an additional cysteine residue) of IBV 3a was used to generate polyclonal antibodies by previously described methods (10). The rabbit polyclonal IBV E and IBV M antibodies used in this study were described previously (10, 32).
Polyclonal rabbit anti-MxA antibodies were a gift from M. McNiven (Mayo Clinic, Rochester, MN). Mouse anti-ERGIC 53 antibodies were a gift from H. P. Hauri (Biocenter of the University of Basel, Basel, Switzerland). Mouse anti-TIA-R antibodies were a gift from P. Anderson (Brigham and Women's Hospital, Boston, MA). Mouse anti-LAMP2 antibodies were a gift from T. August (Johns Hopkins University School of Medicine, Baltimore, MD). Mouse anti-mannose-6-phosphate receptor (MPR) antibodies were a gift from S. Pfeffer (Stanford University School of Medicine, Palo Alto, CA). Mouse anti-gamma-adaptin (AP1) antibodies were obtained from Sigma-Aldrich Co. (St. Louis, MO). Mouse anti-rab9 and anti-GFP antibodies were obtained from Roche Molecular Biochemicals (Indianapolis, IN). Goat anti-BiP antibodies were obtained from Research Diagnostics, Inc. (Flanders, NJ). Rabbit anti-GFP and Alexa-488-conjugated donkey anti-mouse (immunoglobulin G [IgG]) antibodies were obtained from Molecular Probes, Inc. (Eugene, OR). Mouse anti-GM130, mouse anti-p230, and mouse anti-p115 antibodies were obtained from BD Transduction (San Diego, CA). Texas Red-conjugated donkey anti-rabbit (IgG), fluorescein isothiocyanate (FITC)-conjugated anti-rat (IgG), FITC-conjugated goat anti-mouse (IgG), FITC-conjugated donkey anti-rabbit (IgG), Texas Red-conjugated donkey anti-mouse (IgG), and Texas Red-conjugated donkey anti-goat (IgG) antibodies were obtained from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA). Horseradish peroxidase-conjugated donkey anti-rabbit (IgG) antibodies were obtained from Amersham Pharmacia Biotech, Inc., (Piscataway, NJ).
The titers of the IBV M and IBV 3a antibodies were compared by enzyme-linked immunosorbent assays (ELISAs). Stock solutions (1 mg/ml) of IBV 3a and IBV M peptides were diluted to 0.1 μg/150 μl in 0.1 M NaHCO3 (pH 9.6) and absorbed overnight at 4°C in a 96-well plate (Corning Incorporated, Corning, NY). Following removal of the peptide solution, wells were blocked for 2 h at room temperature with 10 mg/ml bovine serum albumin (BSA) (Sigma) and 0.02% Tween 20 in phosphate-buffered saline. Wells were extensively washed, and anti-IBV 3a or anti-IBV M antibodies were added to wells in serial dilutions. Primary antibody incubation was performed for 45 min at room temperature and for an additional 15 min at 37°C. Secondary horseradish peroxidase-conjugated donkey anti-rabbit antibodies were added (following extensive washing) for 30 min at 37°C. BM Blue POD soluble substrate (Boehringer-Mannheim GmbH, Mannheim, Germany) was used according to the manufacturer's protocol, and antibody binding was measured by using a model 3550 microplate reader (Bio-Rad Laboratories, Hercules, CA). The titers (the dilution at 50% the maximal optical density for 0.1 μg peptide) were approximately 1:10,000 for both anti-IBV 3a and anti-IBV M antibodies.
Infection, transient transfection, indirect immunofluorescence microscopy, and confocal microscopy. All cells used in immunofluorescence studies were Vero cells unless otherwise specified. In colocalization studies, cells were plated on coverslips in 35-mm dishes for approximately 1 day before receiving one of three treatments: (i) infection with IBV, (ii) transient transfection, or (iii) infection with vTF7-3 followed by transfection. IBV infection was performed at a multiplicity of infection (MOI) of 1 by allowing virus to absorb to cells for 1 h at 37°C before transferring cells to normal growth medium. IBV-infected cells were stained for immunofluorescence at between 10 and 28 h postinfection (hpi). Transient transfection was performed by adding 2 μg DNA/35-mm dish for approximately 24 h with either LT1 (Mirus Bio Corporation, Madison, WI) or Fugene 6 (Roche) according to the manufacturers' protocols. Infection with vTF7-3 was performed as previously described (10). At 1 hpi, vTF7-3-infected cells were transfected with pBS/IBV 3a by using Lipofectin (Invitrogen) according to the manufacturer's protocol. Cells were processed for indirect immunofluorescence microscopy (described below) at approximately 4 hpi.
IBV-infected and vTF7-3-infected cells were used in studies that assessed colocalization between IBV 3a and the following proteins: ERGIC 53 (41), GM130 (a marker for the cis-Golgi compartment) (35), p115 (a marker for the cis-Golgi compartment and ERGIC) (48), p230 (a marker for the trans-Golgi network) (24), BiP (a marker for the ER) (7), TIA-R (a marker for stress granules) (3, 46), and AP1 (a marker for the trans-Golgi network and late endosomes) (38). IBV-infected cells were used in studies that assessed colocalization between IBV 3a and the following proteins or dyes: rab9 (a marker for late endosomes) (31), MPR (a marker for the trans-Golgi network and late endosomes) (14), MitoTracker Green FM (Molecular Probes, Inc.) (mitochondrial dye), LAMP2 (a marker for lysosomes) (34), IBV E, and Alexa-594-conjugated transferrin (Molecular Probes, Inc.) (a marker for early endosomes) (12). Transient transfection of pEGFP-N1/IBV 3a and pEGFP-N1/IBV 3a-GFP by using Fugene 6 allowed visualization of these proteins in HeLa cells.
Colocalization of IBV 3a with LSm1 (a marker for mRNA processing bodies) (20) was assessed by cotransfection of pEGFP-N1/IBV 3a and an RFP-LSm1 construct into cells by using Fugene 6. Colocalization of IBV 3a with CFP-KDEL was assessed in the same manner. Colocalization of IBV 3a with myc-PTE-1 (a marker for peroxisomes) (22) was assessed by cotransfecting cells with pcDNA/IBV 3a and a myc-PTE-1 construct by using LT1. Colocalization of IBV 3a with sec13-GFP (a marker for ER exit sites) (17) was assessed by transfecting HeLa cells that stably expressed sec13-GFP with pcDNA/IBV 3a. Colocalization of IBV 3a with MxA (a smooth ER resident protein) (2) was assessed by transfecting HeLa cells with pEGFP-N1/IBV 3a by using Fugene 6. At 24 h following transfection, cells were treated with 1,000 U/ml of beta interferon (Sigma) for 24 h to induce MxA expression.
Immunofluorescence studies that assessed colocalization between IBV 3a and the following proteins or dyes were carried out with fixation and permeabilization methods that have been described previously (10): ERGIC 53, GM130, p115, p230, transferrin, BiP, rab9, CFP-KDEL, TIA-R, MitoTracker, IBV E, and myc-PTE-1. Briefly, cells were fixed with 3% paraformaldehyde in phosphate-buffered saline for 20 min at room temperature. Permeabilization with 0.5% Triton X-100 for 3 min at room temperature followed. Cells then were stained with the appropriate primary and secondary antibodies. HeLa cells assessed for IBV 3a and IBV 3a-GFP localizations were fixed and permeabilized in the same manner. To assess antibody specificity, anti-IBV 3a antibodies were incubated in a 1:1 (vol:vol) ratio with either 1 mg/ml of IBV 3a peptide (described above) or distilled H2O (control) for 24 h at 4°C.
Cells assessed for TIA-R staining were treated with 0.5 mM sodium arsenite for 1 h at 37°C immediately before fixation. Cells stained for mitochondria were treated with 100 nM MitoTracker dye (according to the manufacturer's protocol) for 1 h at 37°C immediately before fixation. Transferrin staining was accomplished by incubating cells in DMEM without serum for 1 h at 37°C followed by 10 min of incubation at the same temperature with Alexa-594-conjugated transferrin. Cells then were fixed and permeabilized as described above. For assessment of the effects of disrupting transport from the ER to the Golgi apparatus on IBV 3a localization, IBV-infected cells were treated with 1 μg/ml brefeldin A (Molecular Probes, Inc.) for approximately 1 h immediately before fixation and staining for IBV 3a and GM130. For assessment of the effects of disrupting microtubule assembly on IBV 3a localization, IBV-infected cells were treated for 1 h with 5 μg/ml nocodazole (Molecular Probes, Inc.) before being fixed and stained for IBV 3a and GM130. For assessment of the effects of disrupting actin polymerization on IBV 3a localization, IBV-infected cells were treated for 1 h with 1 μg/ml cytochalasin D (Sigma) before being fixed and stained with Texas Red-conjugated phalloidin (Molecular Probes, Inc.) and anti-IBV 3a antibodies.
Cells assessed for RFP-LSm1 staining were fixed and stained as described above; however, permeabilization was performed with 0.2% Triton X-100 for 20 min at room temperature. Cells assessed for IBV 3a colocalization with AP1, MxA, and MPR were fixed as described above; however, cells were permeabilized with 0.05% saponin (EM Biosciences, Inc., San Diego, CA) for 5 min at room temperature. Subsequent antibody incubation of these cells was performed in the presence of 0.05% saponin, 1% BSA, and 0.01 M glycine in phosphate-buffered saline. Selective permeabilization of the plasma membrane was performed with digitonin as previously described (10). Briefly, cells were permeabilized with 25 μg/ml of digitonin (EM Industries, Inc., Gibbstown, NJ) for 5 min on ice. This permeabilization occurred either before or after fixation as described above.
IBV 3a colocalization with LAMP2 and sec13-GFP was assessed by using methods described by Hammond and Glick (17). Briefly, cells were fixed with methanol for 10 min at –20°C. Cells then were cross-linked by incubation for 30 min at room temperature in a phosphate-buffered solution containing 0.1% n-octyl-?-D-glucopyranoside (Boehringer) (PBSO) and 100 μM bis-sulfosuccinimidyl (Pierce Chemical Co., Rockford, IL). Cells then were incubated for 15 min in PBSO containing 0.1 M ethylenediamene (Sigma) before being stained with the appropriate primary and secondary antibodies in PBSO with 1% BSA.
Indirect immunofluorescence images were taken on an Axioskop microscope (Zeiss, Thornwood, NY) with an attached Sensys charge-coupled device camera (Photometrics, Tucson, AZ). IP Lab imaging software (Signal Analytics, Vienna, VA) was used. Confocal imaging was used to assess the colocalization of IBV 3a with CFP-KDEL and MxA. A spinning-disk Ultraview confocal microscope (Perkin Elmer, Inc., Lexington, MA) with an LSI cooled 12-bit camera and Ultraview software were used. Volocity software (Improvision, Inc., Lexington, MA) was used to create one image from approximately 15 (CFP-KDEL and 6 smaller MxA images), 4 (large MxA images), or 50 (rotation images of MxA inset) confocal sections that were each 0.4 (CFP-KDEL experiments) or 0.2 (MxA experiments) μm thick.
Immunoblotting. Immunoblotting was performed as described previously (10). Briefly, Vero cells were infected with IBV at an MOI of 1 before being lysed in a solution containing 4% sodium dodecyl sulfate (SDS), 30% glycerol, 1% bromophenol blue, 10% 2-mercaptoethanol, and 0.1 M Tris-HCl (pH 8). Samples then were run on a 17.5% polyacrylamide denaturing gel before being transferred to a polyvinylidene difluoride membrane (Millipore Corporation, Billerica, MA). The membrane was blocked for 30 min at room temperature with 5% nonfat milk dissolved in TTBS (150 mM NaCl, 0.05% Tween 20, 10 mM Tris-HCl [pH 7.4]). Primary antibody incubation was done overnight at 4°C in TTBS plus 1% nonfat milk. Following extensive washing, the membrane was incubated with horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature in TTBS plus 5% nonfat milk. Analysis was performed by using enhanced chemiluminescence (ECL) Western blotting detection reagents (Amersham) and a VersaDoc model 5000 imaging system (Bio-Rad) with an attached cooled charge-coupled device AF Nikkor camera (Nikon, Inc., Melville, NY). Quantitation was performed by using Quantity One software (Bio-Rad).
In vitro translation assays. A TNT quick coupled transcription/translation T7 system (Promega Corporation, Madison, WI) was used according to the manufacturer's protocol in the presence or absence of canine pancreatic microsomal membranes (Promega). Briefly, pBS/IBV 3a, pBS/truncated 3a, or pBS/3a VLS-VLI was incubated for 90 min at 30°C in a transcription/translation master mix containing 35S-Redivue-methionine (Amersham). For each plasmid, two reaction mixtures were made, one that contained microsomal membranes and one that did not. Reaction mixtures were suspended in swelling buffer (15 mM NaCl, 1 mM MgCl2, and 10 mM Tris-HCl [pH 7.4]) and then subjected to ultracentrifugation at 136,000 x g for 1 h through a 10% sucrose cushion in swelling buffer. The resulting supernatants and pellets were immunoprecipitated as described previously (32) and analyzed by SDS-polyacrylamide gel electrophoresis (PAGE) and 35S quantitation.
Alkaline carbonate extraction. Alkaline carbonate extraction was performed as described previously (10). Briefly, cells were plated on 6-cm dishes for 24 h prior to transfection or infection. HeLa cells were transfected with either pEGFP-N1/IBV 3a or pEGFP-N1/IBV 3a-GFP as described above. Vero cells were infected with IBV at an MOI of 1 for 31 h. HeLa and Vero cells were labeled for 1 h at 37°C with 35S-methionine-cysteine (Pro-mix; Amersham) and homogenized in a swelling solution containing 8% sucrose, 15 mM NaCl, 1 mM MgCl2, and 10 mM Tris-HCl (pH 7.4) with 60 strokes of a tightly fitting Dounce homogenizer. Following the pelleting of nuclei, microsomal membranes were treated with 0.1 M Na2CO3 (pH 11.5). Treatment with 0.1 M NaCl or a detergent solution containing 0.05 M Tris-HCl (pH 8), 0.0625 M EDTA (pH 8), 0.4% deoxycholic acid, and 1% octylphenylpolyethylene glycol served as a control. Microsomal membranes then were pelleted by ultracentrifugation at 136,000 x g for 1 h. The resulting supernatants and pellets were immunoprecipitated as described previously (32) and analyzed by SDS-PAGE and phosphorimaging. Quantitation of 35S was done by exposing gels to a K-HD imaging screen (Bio-Rad) followed by scanning with personal molecular imager FX (Bio-Rad). Quantity One software (Bio-Rad) was used to quantitate images. The percent total calculation was performed with the following formula, where S is supernatant and P is pellet: [S/(S + P)] x 100 or [P/(S + P)] x 100.
RESULTS
IBV 3a expression in infected and transfected cells. A polyclonal antibody to IBV 3a was generated using a synthetic peptide corresponding to the C-terminal 14 amino acids of IBV 3a. Before use in expression studies, the titer of this antibody was compared to the titers of IBV E and IBV M antibodies (10, 32) with ELISAs. The IBV 3a antibody titer was found to be approximately equal to that of the IBV M antibody (see Materials and Methods). The IBV M antibody titer had been previously shown to be approximately one-third that of the IBV E antibody titer. Vero cells were infected with IBV for the times indicated, harvested, and analyzed by Western blot (Fig. 1A). An approximately 6-kDa band corresponding to IBV 3a was detected in infected cells. This band was absent from uninfected cells, indicating that the anti-IBV 3a antibody was specific. Quantitation indicated that IBV 3a was expressed later in infection than IBV M, and at much lower levels than either IBV E or IBV M. IBV 3a was approximately 13-fold less abundant than IBV M at 24 hpi. However, this quantitation is a rough estimate since antibody titers were determined by ELISAs, not immunoblotting.
The anti-IBV 3a antibody then was used to explore the intracellular localization of IBV 3a. Following infection with IBV for approximately 15 h, IBV 3a and IBV E were visualized by indirect immunofluorescence microscopy (Fig. 1B). IBV 3a showed a distinct localization compared to IBV E, which is targeted to the Golgi complex (10). IBV 3a showed both a diffuse localization and sharp puncta of approximately 500 nm to 1.5 μm in size. Similar localization was observed in cells transiently transfected with a plasmid encoding IBV 3a (Fig. 1C), indicating that the intracellular localization of IBV 3a is independent of other viral proteins. Uninfected or nontransfected cells on the same coverslip served as negative controls for antibody staining (see Fig. 6; also, data not shown). Preincubation of the IBV 3a antibody with the C-terminal IBV 3a peptide eliminated the signal, indicating that the antibody was specific to IBV 3a (Fig. 1C). IBV 3a protein was not present in purified virions released from IBV-infected, 35S-methionine-cysteine-labeled cells (data not shown), indicating that it is not a structural protein. This is consistent with previous findings (29).
A potential signal sequence cleavage site in IBV 3a is not used, and a pool of IBV 3a associates with membranes. Hydropathy analysis of the IBV 3a sequence indicated a highly hydrophobic, potential membrane-spanning segment at the N terminus (Fig. 2A). Some sequence analysis programs also predicted that the hydrophobic region was a signal sequence, cleavable after S22. To test whether IBV 3a was cleaved at S22, two mutants of IBV 3a were created: a truncated version of IBV 3a corresponding to amino acids 23 to 57 (truncated 3a), and an uncleavable version of IBV 3a in which S22 was mutated to an isoleucine (3a VLS-VLI). Cells were transiently transfected with a plasmid encoding wild-type IBV 3a, truncated 3a, or 3a VLS-VLI. Cells were labeled with 35S-methionine-cysteine and immunoprecipitated with anti-IBV 3a antibodies. Wild-type IBV 3a comigrated with 3a VLS-VLI but not with truncated 3a (Fig. 2B). These results indicated that the potential signal sequence cleavage site in IBV 3a was not used. IBV 3a from infected cells also comigrated with 3a VLS-VLI and not with truncated 3a (data not shown).
Since the hydrophobic segment at the N terminus of IBV 3a was not a cleaved signal sequence, we predicted that this sequence might serve as a membrane anchor. To assess this possibility, microsomal membranes prepared from IBV-infected, 35S-methionine-cysteine-labeled cells were extracted with alkaline carbonate. This treatment induces membrane sheets, releasing lumenal contents and peripheral membrane proteins. Microsomes were treated with Na2CO3 (pH 11.5), 0.1 M NaCl, or detergent as a control. Following these treatments, membranes were pelleted via ultracentrifugation. The resulting supernatants and pellets then were immunoprecipitated with anti-IBV 3a antibodies and analyzed by SDS-PAGE. IBV E, a known transmembrane protein (10), was used as a control. We found that approximately 60% of IBV 3a was cytoplasmic and 40% of IBV 3a was tightly associated with membranes (Fig. 2C). IBV 3a from transiently transfected cells behaved similarly (data not shown).
A pool of IBV 3a associates with membranes in vitro. To confirm the partial membrane association of IBV 3a, in vitro translation in the presence or absence of canine pancreatic microsomal membranes was performed. We used truncated IBV 3a and IBV E as controls. Following in vitro translation, samples were subjected to ultracentrifugation to pellet microsomal membranes. The resulting supernatants and pellets then were immunoprecipitated and analyzed by SDS-PAGE. IBV E was poorly translated in the absence of microsomes, but was mostly in the pellet when microsomes were included. Similarly, there was an increase of IBV 3a in the pellet fraction when microsomal membranes were included during translation (Fig. 3). These data indicated that IBV 3a was capable of associating with membranes. There was no increase of truncated 3a in the pellet with the addition of microsomal membranes. Thus, the N terminus of IBV 3a was responsible for its membrane association. Additionally, although more IBV 3a was found in the pellet when microsomes were present, a significant portion of IBV 3a was still found in the supernatant in these conditions. Thus, these in vitro data confirmed previous alkaline carbonate results from infected cells (Fig. 2C) showing partial membrane association of IBV 3a.
GFP-tagged IBV 3a associates with membranes more efficiently than wild-type IBV 3a. The results indicating that a pool of IBV 3a was cytoplasmic while another pool of IBV 3a was membrane bound were intriguing. The membrane-associated pool appears to enter the secretory pathway (see below). However, the short length of IBV 3a (57 amino acids) may preclude efficient interaction with signal recognition particle for cotranslational translocation into the ER (19). We asked if extending the length of IBV 3a (with a C-terminal GFP tag) increased the efficiency of IBV 3a membrane insertion. To ensure that IBV 3a-GFP puncta represented similar structures to IBV 3a puncta, cells were doubly transfected with IBV 3a-GFP and IBV 3a constructs. The anti-IBV 3a antibody does not bind to IBV 3a-GFP, presumably due to interference of the GFP tag with the C-terminal epitope recognized by the IBV 3a antibodies. Indirect immunofluorescence with anti-IBV 3a and anti-GFP antibodies demonstrated that IBV 3a and IBV 3a-GFP largely colocalize (data not shown). Thus, IBV 3a-GFP puncta were similar to those containing wild-type IBV 3a.
Interestingly, cells transiently transfected with a plasmid encoding GFP-tagged IBV 3a showed reduced diffuse staining while maintaining distinct puncta (Fig. 4A). To determine if this finding reflected more efficient membrane association, alkaline carbonate extraction was performed on Vero cells transiently transfected with a plasmid encoding either wild-type IBV 3a or GFP-tagged IBV 3a. In transiently transfected cells, the percentage of IBV 3a that was tightly associated with membranes was approximately equal to the percentage seen in infected cells. However, GFP-tagged IBV 3a more efficiently associated with membranes than wild-type IBV 3a in both physiological salt and alkaline carbonate conditions (Fig. 4B). These data indicated that the short length of IBV 3a is responsible for its inefficient insertion into membranes.
The diffusely staining pool of IBV 3a is soluble. To determine whether the diffusely stained pool of IBV 3a observed by indirect immunofluorescence was cytoplasmic, IBV-infected Vero cells were permeabilized with digitonin before fixation. Under the conditions used, digitonin selectively permeabilizes the plasma membrane and allows a significant portion of the cytoplasmic contents to be washed from cells (27). BiP was used as a control to confirm that intracellular membranes were not permeabilized. Digitonin permeabilization prior to fixation resulted in a substantial reduction in the diffusely stained pool of IBV 3a (Fig. 5A). This loss did not occur in cells that were fixed before digitonin treatment. BiP was not released with either condition, since it was clearly detected after additional permeabilization with Triton X-100. Thus, the diffusely staining pool of IBV 3a was cytoplasmic. This conclusion was substantiated by subjecting the digitonin-released contents to ultracentrifugation. IBV-infected Vero cells were labeled with 35S-methionine-cysteine before treatment with digitonin. The cellular contents released upon digitonin treatment then were collected and subjected to ultracentrifugation. The released portion of IBV 3a was soluble (Fig. 5B).
IBV 3a topology. Digitonin permeabilization was also used to determine the orientation of the membrane-associated IBV 3a. IBV-infected Vero cells were fixed and permeabilized with either Triton X-100 (which permeabilizes all cell membranes) or digitonin. BiP was again used as a control. The localization pattern of IBV 3a in cells permeabilized with Triton X-100 was similar to the pattern seen in digitonin-permeabilized cells (Fig. 6), indicating that the C terminus of both cytoplasmic and membrane-bound IBV 3a was located in the cytoplasm. This result, combined with the results of the hydopathy analysis, suggests that for the pool of IBV 3a that is membrane associated, the amino terminus is exposed to a lumenal compartment, the N-terminal hydrophobic region crosses the membrane once, and the C terminus is located in the cytoplasm. However, these data do not rule out the possibility that some portion of membrane-associated IBV 3a has the opposite topology, with the C terminus in a lumenal compartment.
IBV 3a puncta localize to the smooth ER and are distinct from other membrane-bound organelles. To determine the exact cellular location of IBV 3a puncta, colocalization immunofluorescence studies with various cellular marker proteins were performed on IBV-infected Vero cells. We first analyzed potential colocalization with punctate organelles. The IBV 3a puncta were distinct from endosomes, lysosomes, peroxisomes, and ER exit sites (Table 1). There was also no colocalization with Golgi membranes or mitochondria. IBV 3a puncta were also distinct from cytoplasmic foci such as stress granules (marked by TIA-R) and mRNA processing bodies (marked by LSm1). Furthermore, the IBV 3a puncta were unaffected by drugs that disrupted transport from the ER to the Golgi apparatus (brefeldin A), actin polymerization (cytochalasin D), and microtubule assembly (nocodazole) (data not shown). However, we did find that the IBV 3a puncta lined up along ER tubules labeled with CFP-KDEL (Fig. 7).
IBV 3a puncta did not appear to colocalize with BiP when analyzed by indirect immunofluorescence. Therefore, we decided to further explore the localization of IBV 3a to ER membranes by testing a marker for smooth ER, MxA (2). This experiment could not be done in infected cells, since we could not detect the MxA protein in Vero cells. In addition, since both anti-IBV 3a and anti-MxA antibodies were prepared in rabbits, we used tagged IBV 3a to allow double labeling. HeLa cells were transfected with a plasmid encoding IBV 3a-GFP 24 h prior to being treated with beta interferon (to induce MxA expression) for 24 h. Cells then were fixed and double-labeled for IBV 3a-GFP and MxA and analyzed by confocal microscopy. Beta interferon treatment did not change the distribution of IBV 3a-GFP puncta. The results indicated that IBV 3a-GFP puncta lined, contacted, or were surrounded by MxA tubules (Fig. 8). Using software that can rotate stacks of confocal images in three dimensions, IBV 3a-GFP puncta were assessed for their proximity to MxA-positive tubules. Analysis of 349 puncta from six different cells showed that 99.1% were proximal (within half the diameter of an IBV 3a-GFP punctum) to MxA-stained structures, suggesting that IBV 3a puncta arise from smooth ER tubules. A gallery of images from four different cells is shown in Fig. 8B.
DISCUSSION
IBV 3a is a highly conserved protein among group 3 coronaviruses (21), suggesting it possesses a critical function during viral infection. Our results demonstrated that IBV 3a is expressed in infected cells (Fig. 1A and B), similar to earlier studies by Liu et al. (28). Our data also demonstrated that IBV 3a is expressed at much lower levels than either IBV E or IBV M during viral infection, that it is expressed later in infection than IBV M, and that it is not incorporated into virions. In both infected and transfected cells, IBV 3a was found to localize diffusely to the cytoplasm as well as to membrane-associated puncta (Fig. 1B). These IBV 3a-containing puncta were distinct from early and late endosomes, lysosomes, peroxisomes, ER exit sites, the Golgi complex, the ERGIC, stress granules, and mRNA processing bodies (Table 1). Ultimately, confocal microscopy demonstrated that IBV 3a localized to smooth ER domains, which are distinct from the rough ER (5). Additionally, IBV 3a in membrane-associated puncta was oriented with its C terminus exposed to the cytoplasm. These results suggest that membrane-associated IBV 3a spans the membrane once, with its N terminus exposed to the ER lumen (Fig. 6).
Our results raise several interesting questions. Why does a pool of IBV 3a localize to the cytoplasm while another pool localizes to smooth ER membranes? One possible explanation is that the short length of IBV 3a (57 amino acids) precludes efficient interaction with signal recognition particle for translocation into ER membranes (19). This idea is supported by our finding that lengthening IBV 3a by the addition of a GFP tag to the C terminus increased the proportion of IBV 3a that was tightly associated with membranes (Fig. 4B). These results might indicate that only small amounts of IBV 3a are needed at smooth ER membranes. Perhaps the energy expended in producing a protein, only a portion of which is properly targeted, is offset by the energetic advantage of a more streamlined genome. Alternatively, these results could indicate that too much IBV 3a localized to smooth ER membranes is detrimental to virus infection; therefore, IBV 3a evolved so that relatively low amounts reach membranes. This later possibility is intriguing in light of alkaline carbonate extraction data for IBV E versus GFP-tagged IBV 3a. Unlike IBV E (Fig. 2C), a known transmembrane protein, the membrane association of GFP-tagged IBV 3a was not complete. Therefore, the 5' sequence of IBV 3a may naturally possess a low affinity for signal recognition particle, indicating that the virus has evolved at least two mechanisms to ensure that only limited amounts of IBV 3a reach smooth ER membranes. However, the previous arguments assume that only membrane-bound IBV 3a possesses a function critical for virus infection. In fact, IBV 3a may have evolved its variable membrane association characteristics if both cytoplasmic and membrane-bound IBV 3a possess important, possibly distinct, functions. Future experiments will address the consequences to the viral life cycle of replacing the wild-type IBV 3a sequence with that of the extended or truncated versions using an infectious clone of IBV (49).
Also of interest is the close proximity of IBV 3a puncta and MxA containing smooth ER tubules (Fig. 8). MxA is a member of the dynamin family of proteins (reviewed in reference 16). This family is known to promote vesicle formation, presumably through the intrinsic GTPase activity of its family members. Similarly, MxA has been shown to tubulate lipids both in vitro and in vivo. However, MxA is unique from other known dynamin family members because its expression is induced by alpha and beta interferons. Consistent with this type of inducible expression, MxA possesses antiviral activity against a number of different viruses that have either RNA genomes or genomes that go through a RNA intermediate. In some of these cases, MxA is thought to function by binding and sequestering essential viral proteins, thus preventing their proper localization (40). The close localization of IBV 3a to MxA-containing membranes suggests that IBV 3a may interact with and potentially abrogate the antiviral activity of MxA. However, the effects of MxA on IBV replication have not been established. Also, recent experiments demonstrated that stable MxA expression in Vero cells (which lack the ability to respond to alpha and beta interferons) does not negatively affect replication of the severe acute respiratory syndrome coronavirus (44). Therefore, the localization of IBV 3a may be unrelated to interaction with MxA. Future biochemical experiments will assess physical interaction of MxA with IBV 3a. Additionally, generation of an infectious clone of IBV lacking functional IBV 3a may be useful in determining functional interactions between IBV 3a and MxA.
If IBV 3a does not interact with MxA, then IBV 3a may be involved with another function of smooth ER membrane domains. An interesting possibility is suggested by recent data from Ng and Liu (37), who suggested that IBV appeared to anchor its replication machinery to ER membranes. If virus replication were found to occur at vesicles derived from smooth ER membranes, then IBV 3a might participate in the genesis or functioning of these complexes. Alternatively, IBV 3a might take part in other functions known to occur at smooth ER membranes, such as lipid metabolism (6, 11), atypical protein trafficking events (4, 26, 39), or an as-yet-undiscovered process occurring at these membranes. Analysis in the context of an IBV 3a-null infectious clone will likely distinguish among these possibilities and help to elucidate the specific function of IBV 3a in virus infection. Additionally, IBV 3a may be useful in shedding light on the dynamics of smooth ER membranes.
ACKNOWLEDGMENTS
This work was supported by National Institutes of Health grant GM64647.
We thank Emily Corse for the generation of IBV 3a antibodies and plasmid pBS/IBV 3a, Stuart Hicks for the HeLa cell line stably expressing sec13-GFP, and all members of the Machamer Laboratory for helpful comments and stimulating discussions. We also thank M. McNiven (Mayo Clinic, Rochester, MN) for the generous gift of anti-MxA antibodies.
REFERENCES
Abraham, D. J., and A. J. Leo. 1987. Extension of the fragment method to calculate amino acid zwitterion and side chain partition coefficients. Proteins 2:130-152.
Accola, M. A., B. Huang, A. Al Masri, and M. A. McNiven. 2002. The antiviral dynamin family member, MxA, tubulates lipids and localizes to the smooth endoplasmic reticulum. J. Biol. Chem. 277:21829-21835.
Anderson, P., and N. Kedersha. 2002. Stressful initiations. J. Cell Sci. 115:3227-3234.
Benlimame, N., P. U. Le, and I. R. Nabi. 1998. Localization of autocrine motility factor receptor to caveolae and clathrin-independent internalization of its ligand to smooth endoplasmic reticulum. Mol. Biol. Cell 9:1773-1786.
Benlimame, N., D. Simard, and I. R. Nabi. 1995. Autocrine motility factor receptor is a marker for a distinct membranous tubular organelle. J. Cell Biol. 129:459-471.
Bishop, W. R., and R. M. Bell. 1988. Assembly of phospholipids into cellular membranes: biosynthesis, transmembrane movement and intracellular translocation. Annu. Rev. Cell Biol. 4:579-610.
Bole, D. G., R. Dowin, M. Doriaux, and J. D. Jamieson. 1989. Immunocytochemical localization of BiP to the rough endoplasmic reticulum: evidence for protein sorting by selective retention. J. Histochem. Cytochem. 37:1817-1823.
Boursnell, M. E., M. M. Binns, and T. D. Brown. 1985. Sequencing of coronavirus IBV genomic RNA: three open reading frames in the 5' ‘unique’ region of mRNA D. J. Gen. Virol. 66:2253-2258.
Brown, T. D. K., and I. Brierley. 1995. The coronavirus nonstructural proteins, p. 191-217. In S. G. Siddell (ed.), The Coronaviridae. Plenum Press, New York, N.Y.
Corse, E., and C. E. Machamer. 2000. Infectious bronchitis virus E protein is targeted to the Golgi complex and directs release of virus-like particles. J. Virol. 74:4319-4326.
Daleke, D. L., and J. V. Lyles. 2000. Identification and purification of aminophospholipid flippases. Biochim. Biophys. Acta 1486:108-127.
Daro, E., P. van der Sluijs, T. Galli, and I. Mellman. 1996. Rab4 and cellubrevin define different early endosome populations on the pathway of transferrin receptor recycling. Proc. Natl. Acad. Sci. USA 93:9559-9564.
de Haan, C. A., P. S. Masters, X. Shen, S. Weiss, and P. J. Rottier. 2002. The group-specific murine coronavirus genes are not essential, but their deletion, by reverse genetics, is attenuating in the natural host. Virology 296:177-189.
Dintzis, S. M., V. E. Velculescu, and S. R. Pfeffer. 1994. Receptor extracellular domains may contain trafficking information. Studies of the 300-kDa mannose 6-phosphate receptor. J. Biol. Chem. 269:12159-12166.
Fuerst, T. R., E. G. Niles, F. W. Studier, and B. Moss. 1986. Eukaryotic transient-expression system based on recombinant vaccinia virus that synthesizes bacteriophage T7 RNA polymerase. Proc. Natl. Acad. Sci. USA 83:8122-8126.
Haller, O., and G. Kochs. 2002. Interferon-induced mx proteins: dynamin-like GTPases with antiviral activity. Traffic 3:710-717.
Hammond, A. T., and B. S. Glick. 2000. Dynamics of transitional endoplasmic reticulum sites in vertebrate cells. Mol. Biol. Cell 11:3013-3030.
Horton, P., and K. Nakai. 1997. Better prediction of protein cellular localization sites with the k nearest neighbors classifier. Proc. Int. Conf. Intell. Syst. Mol. Biol. 5:147-152.
Ibrahimi, I. M., D. Cutler, D. Stueber, and H. Bujard. 1986. Determinants for protein translocation across mammalian endoplasmic reticulum. Membrane insertion of truncated and full-length prelysozyme molecules. Eur. J. Biochem. 155:571-576.
Ingelfinger, D., D. J. Arndt-Jovin, R. Luhrmann, and T. Achsel. 2002. The human LSm1-7 proteins colocalize with the mRNA-degrading enzymes Dcp1/2 and Xrnl in distinct cytoplasmic foci. RNA 8:1489-1501.
Jia, W., and S. A. Naqi. 1997. Sequence analysis of gene 3, gene 4 and gene 5 of avian infectious bronchitis virus strain CU-T2. Gene 189:189-193.
Jones, J. M., K. Nau, M. T. Geraghty, R. Erdmann, and S. J. Gould. 1999. Identification of peroxisomal acyl-CoA thioesterases in yeast and humans. J. Biol. Chem. 274:9216-9223.
Klumperman, J., J. K. Locker, A. Meijer, M. C. Horzinek, H. J. Geuze, and P. J. Rottier. 1994. Coronavirus M proteins accumulate in the Golgi complex beyond the site of virion budding. J. Virol. 68:6523-6534.
Kooy, J., B. H. Toh, J. M. Pettitt, R. Erlich, and P. A. Gleeson. 1992. Human autoantibodies as reagents to conserved Golgi components. Characterization of a peripheral, 230-kDa compartment-specific Golgi protein. J. Biol. Chem. 267:20255-20263.
Kyte, J., and R. F. Doolittle. 1982. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157:105-132.
Le, P. U., and I. R. Nabi. 2003. Distinct caveolae-mediated endocytic pathways target the Golgi apparatus and the endoplasmic reticulum. J. Cell Sci. 116:1059-1071.
Lineberger, B., D. D. Dawicki, K. C. Agarwal, N. Kessimian, and M. Steiner. 1989. Permeabilization of platelets: an investigation of biochemical, ultrastructural and functional aspects. Biochim. Biophys. Acta 1012:36-45.
Liu, D. X., D. Cavanagh, P. Green, and S. C. Inglis. 1991. A polycistronic mRNA specified by the coronavirus infectious bronchitis virus. Virology 184:531-544.
Liu, D. X., and S. C. Inglis. 1991. Association of the infectious bronchitis virus 3c protein with the virion envelope. Virology 185:911-917.
Liu, D. X., and S. C. Inglis. 1992. Internal entry of ribosomes on a tricistronic mRNA encoded by infectious bronchitis virus. J. Virol. 66:6143-6154.
Lombardi, D., T. Soldati, M. A. Riederer, Y. Goda, M. Zerial, and S. R. Pfeffer. 1993. Rab9 functions in transport between late endosomes and the trans Golgi network. EMBO J. 12:677-682.
Machamer, C. E., and J. K. Rose. 1987. A specific transmembrane domain of a coronavirus E1 glycoprotein is required for its retention in the Golgi region. J. Cell Biol. 105:1205-1214.
Mancini, M., C. E. Machamer, S. Roy, D. W. Nicholson, N. A. Thornberry, L. A. Casciola-Rosen, and A. Rosen. 2000. Caspase-2 is localized at the Golgi complex and cleaves golgin-160 during apoptosis. J. Cell Biol. 149:603-612.
Mane, S. M., L. Marzella, D. F. Bainton, V. K. Holt, Y. Cha, J. E. Hildreth, and J. T. August. 1989. Purification and characterization of human lysosomal membrane glycoproteins. Arch. Biochem. Biophys. 268:360-378.
Nakamura, N., C. Rabouille, R. Watson, T. Nilsson, N. Hui, P. Slusarewicz, T. E. Kreis, and G. Warren. 1995. Characterization of a cis-Golgi matrix protein, GM130. J. Cell Biol. 131:1715-1726.
Navas-Martin, S. R., and S. Weiss. 2004. Coronavirus replication and pathogenesis: implications for the recent outbreak of severe acute respiratory syndrome (SARS), and the challenge for vaccine development. J. Neurovirol. 10:75-85.
Ng, L. F., and D. X. Liu. 2002. Membrane association and dimerization of a cysteine-rich, 16-kilodalton polypeptide released from the C-terminal region of the coronavirus infectious bronchitis virus 1a polyprotein. J. Virol. 76:6257-6267.
Pearse, B. M., and M. S. Robinson. 1990. Clathrin, adaptors, and sorting. Annu. Rev. Cell Biol. 6:151-171.
Pelkmans, L., J. Kartenbeck, and A. Helenius. 2001. Caveolar endocytosis of simian virus 40 reveals a new two-step vesicular-transport pathway to the ER. Nat. Cell Biol. 3:473-483.
Reichelt, M., S. Stertz, J. Krijnse-Locker, O. Haller, and G. Kochs. 2004. Missorting of LaCrosse virus nucleocapsid protein by the interferon-induced MxA GTPase involves smooth ER membranes. Traffic 5:772-784.
Schweizer, A., J. A. Fransen, T. Bachi, L. Ginsel, and H. P. Hauri. 1988. Identification, by a monoclonal antibody, of a 53-kD protein associated with a tubulo-vesicular compartment at the cis-side of the Golgi apparatus. J. Cell Biol. 107:1643-1653.
Shen, S., Z. L. Wen, and D. X. Liu. 2003. Emergence of a coronavirus infectious bronchitis virus mutant with a truncated 3b gene: functional characterization of the 3b protein in pathogenesis and replication. Virology 311:16-27.
Siddell, S. G. 1995. The Coronaviridae: an introduction, p. 1-10. In S. G. Siddell (ed.), The Coronaviridae. Plenum Press, New York, N.Y.
Spiegel, M., A. Pichlmair, E. Muhlberger, O. Haller, and F. Weber. 2004. The antiviral effect of interferon-beta against SARS-coronavirus is not mediated by MxA protein. J. Clin. Virol. 30:211-213.
Sturman, L. S., K. V. Holmes, and J. Behnke. 1980. Isolation of coronavirus envelope glycoproteins and interaction with the viral nucleocapsid. J. Virol. 33:449-462.
Taupin, J. L., Q. Tian, N. Kedersha, M. Robertson, and P. Anderson. 1995. The RNA-binding protein TIAR is translocated from the nucleus to the cytoplasm during Fas-mediated apoptotic cell death. Proc. Natl. Acad. Sci. USA 92:1629-1633.
Van der Most, R. G., and W. J. Spaan. 1995. Coronavirus replication, transcription, and RNA recombination, p. 11-31. In S. G. Siddell (ed.), The Coronaviridae. Plenum Press, New York, N.Y.
Waters, M. G., D. O. Clary, and J. E. Rothman. 1992. A novel 115-kD peripheral membrane protein is required for intercisternal transport in the Golgi stack. J. Cell Biol. 118:1015-1026.
Youn, S., J. L. Leibowitz, and E. W. Collisson. 2005. In vitro assembled, recombinant infectious bronchitis viruses demonstrate that the 5a open reading frame is not essential for replication. Virology 332:206-215.(Amanda R. Pendleton and C)