To identify the proteins that interact with SARS-CoV E protein, a recombinant SARS-CoV expressing the E protein fused to a tag on its carboxy terminus was engineered as a BAC (pBAC-SARS-CoV-EtagCt) (Fig. 1A). The tag contains the FLAG epitope followed by a TEV protease site, and the HA epitope. Vero E6 cells were transfected with this plasmid or with a parental plasmid encoding the wt SARS-CoV (pBAC-SARS-CoV^FL) that lacks E protein associated tags. Recombinant viruses rSARS-CoV-EtagCt and rSARS-CoV-wt were recovered from both plasmids with high titers and showed the characteristic cytopathic effect induced by SARS-CoV infection (data not shown). The recombinant virus expressing tagged E protein was amplified, cloned by three rounds of plaque isolation, and passaged twice to generate a virus stock.Fig. 1

Generation of a recombinant SARS-CoV expressing a tagged E protein. (A) Scheme of recombinant virus expressing tagged E protein. L, leader sequence; REP, replicase gene; S, spike protein, E, envelope protein; M, membrane protein; N, nucleocapsid protein; pA, poly(A) tail; FLAG, FLAG epitope; TEV, tobacco etch virus protease site; and HA, influenza haemmaglutinin epitope. (B) Vero E6 cells were mock infected (Mock) or infected with the recombinant wild-type (WT), the rSARS-CoV-∆E (∆E) or rSARS-CoV-EtagCt (E-tag) viruses. Viral mRNA expression was analyzed by RT-PCR using the oligonucleotides specific for sgmRNAs of E, M and N genes. (C) Western blot analysis of infected cell lysates using E, FLAG, HA and N protein-specific antibodies followed by peroxidase-labelled goat anti-rabbit or anti-mouse antibodies.

Vero E6 cells were infected with SARS-CoV-EtagCt virus, and the E-tag protein complexes were purified by two affinity chromatography steps as described in the Materials and methods section. As control, the same purification was performed with mock- or SARS-CoV-wt-infected cell extracts. Cell lysates were applied to an anti-HA resin and bound proteins were eluted by proteolytic cleavage of the tag at the TEV protease site. Eluates from the proteolytic cleavage on the anti-HA resin were incubated with an anti-FLAG agarose and proteins bound to matrix were eluted by specific competition with FLAG peptide. To identify proteins that copurified with E-tag, the eluted proteins were separated by SDS-PAGE and visualized by Coomassie blue staining. In the case of purifications from cells infected with SARS-CoV-EtagCt, several protein bands were detected, whereas purifications in parallel using extracts from mock-infected cells or cells infected with the wild-type virus did not reveal any detectable protein band (Fig. 4A). These data indicated that the proteins identified specifically bound to E-tag protein. Some of the bands were excised, digested with trypsin, and subjected to MALDI TOF/TOF mass spectrometry to identify the corresponding proteins by screening against the available NCBI protein database. A set of 12 proteins was reproducibly identified in at least two out of four experiments with significant MASCOT scores (p  > 0.05) (Table 1 ). Among the identified proteins binding to E, the viral proteins nsp3, S and M were identified as E protein interacting partners. In addition, the proteins dynein heavy chain, fatty acid synthase, aminopeptidase puromycin sensitive, phosphofructokinase platelet, tubulin alpha and beta, actin beta, transmembrane protein 43 and lactate dehydrogenase A, were identified as cellular proteins interacting with E protein. This paper is focused on the interaction of E protein with nsp3.Fig. 4

Identification of viral proteins interacting with SARS-CoV E protein. (A) Purification of proteins interacting with E protein using a tandem affinity purification method. Vero E6 extracts from mock-infected cells or cells infected with rSARS-CoV-wt (wt) or rSARS-CoV-EtagCt (E-tag) were used in a double affinity chromatography. Purified proteins were detected by staining gels with Coomassie blue dye. Bands were excised from gels and were identified by mass spectrometry. (B) Coimmunoprecipitation assays of nsp3 and E protein. Extracts from Vero E6 cells infected with rSARS-CoV-wt or rSARS-CoV-∆E were immunoprecipitated with nsp3, E or TGEV N protein-specific antibodies. Immunoprecipitations were analyzed by Western blot using E and nsp3 protein-specific antibodies followed by peroxidase-labelled goat anti-rabbit or anti-mouse antibodies.

The interaction between nsp3 and E protein detected with the tandem affinity purification was confirmed by coimmunoprecipitation. To this end, protein complexes formed in Vero E6 cells infected with SARS-CoV-wt, SARS-CoV-∆E, or mock-infected cells were pulled down using antibodies specific for E and nsp3 proteins. Immune precipitates were analyzed by Western blot using the anti-E and anti-nsp3 antibodies (Fig. 4B). Using an E protein-specific antibody, nsp3 was coimmunoprecipitated in samples derived from wild-type infected cells but not in extract derived from mock or SARS-∆E-infected cells. Conversely, using anti-nsp3 antibody to pull-down nsp3, E protein coimmunoprecipitated in extracts from cells infected with SARS-CoV-wt, but not in those from mock-infected cells, or cells infected with SARS-CoV-∆E. In contrast, no specific bands were observed in control experiments where the immunoprecipitations were performed with a monoclonal antibody specific for the N protein of transmissible gastroenteritis virus (TGEV). The amount of E or nsp3 proteins coimmunoprecipitated by antibodies specific for nsp3 or E proteins respectively was relatively low, suggesting that in the experimental conditions used, the binding between these two proteins was relatively weak. Overall, these results indicated that E protein forms complexes with nsp3 protein.

To further confirm the interaction between nsp3 and E protein, a coimmunoprecipitation assay was performed using nsp3-derived fragments expressed in a cell-free coupled transcription–translation system (Fig. 5A). Biotinylated Lys-tRNA was used in translation reactions for subsequent detection of the expressed proteins with streptavidin conjugated to peroxidase (Fig. 5B). Recombinant purified SARS-CoV E protein was added to translation lysates and then E protein was pulled down using the specific monoclonal antibody indicated above. Fragment 1 of nsp3 was coimmunoprecipitated together with E protein, while no bands were detected in a control experiment in which the porcine reproductive and respiratory syndrome virus (PRRSV) glycoprotein 5 (Gp5) was added to translation lysates and pulled down with anti-Gp5 antibody (Fig. 5B). Overall, these results indicated that the interaction between nsp3 and E protein is independent of the presence of other SARS-CoV proteins, and that the interaction site in nsp3 is located in the N-terminal domain of the protein.Fig. 5

Coimmunoprecipitation assays of nsp3 fragments. (A) Scheme of nsp3-derived fragments synthesized in TNT® coupled reticulocyte lysate systems (F1 to F4). The nsp3 domains are shown in the scheme: UB1, ubiquitin-like domain 1; AC, acidic hypervariable domain; ADRP, ADP-ribose-1″-phosphatase; SUD, SARS unique domain; UB2, ubiquitin-like domain 2; PLP, papain-like protease; NAB, group II-specific domain; G2M, group II-specific maker; TM, transmembrane motif; ZF, putative metal-binding region; and Y, Y region. Synthesized fragments were mixed with recombinant E or Gp5 proteins and then were used in immunoprecipitation assays using E or Gp5 protein-specific antibodies. Immunoprecipitations were analyzed by Western blot using E and Gp5 protein-specific antibodies followed by peroxidase-labelled goat anti-rabbit or anti-mouse antibodies. Luciferase (Luc) was used as a control.

To further delimit the domain of nsp3 involved in the interaction with E protein, the N-terminal acidic domain (UB1-AC), the ADRP, and the SUD domains of nsp3 protein were expressed in cell-free systems as described above (Fig. 6A). Purified E or Gp5 proteins were mixed with each of the nsp3 N-terminal biotinylated polypeptides and immunoprecipitated using antibodies against E protein or the unrelated protein Gp5, as a negative control (Fig. 6B). While no bands were detected in a control experiment when Gp5 was used as bait, a band corresponding to UB1-AC domain was observed in samples in which purified E protein was added to the translation mixtures and subsequently immunoprecipitated with E specific antibody (Fig. 6B). Moreover, bands corresponding to ADRP and SUD domains were not coimmunoprecipitated together with E protein (Fig. 6B). This data indicates that nps3 interacts with E protein through the ubiquitin-like domain-1 located in the N-terminus of the protein.Fig. 6

Mapping the nsp3 region required to interact with E protein. (A) Scheme of nsp3-F1-derived fragments synthesized in TNT® coupled reticulocyte lysate systems (UB1-AC, ADRP and SUD). The nsp3 domains are shown in the scheme: UB1, ubiquitin-like domain 1; AC, acidic hypervariable domain; ADRP, ADP-ribose-1″-phosphatase; SUD, SARS unique domain; UB2, ubiquitin-like domain 2; PLP, papain-like protease; NAB, group II-specific domain; G2M, group II-specific marker; TM, transmembrane motif; ZF, putative metal-binding region; and Y, Y region. Synthesized fragments were mixed with recombinant E or Gp5 proteins and then were used in immunoprecipitation assays using E or Gp5 protein-specific antibodies. Immunoprecipitations were analyzed by Western blot using E and Gp5 protein-specific antibodies followed by peroxidase-labelled goat anti-rabbit or anti-mouse antibodies.

To obtain complementary support for the interaction between nsp3 and E protein, Vero E6 cells infected with SARS-CoV-wt, SARS-CoV-∆E and SARS-CoV-EtagCt were analyzed by confocal immunomicroscopy using HA (to detect E-tag) or nsp3 specific antibodies. All infected cells showing the presence of E protein also stained with the nsp3 specific antibody giving an identical pattern, consistent with a perinuclear distribution (Fig. 7 ). The localization of the tagged E protein was the same as untagged E in infected cells indicating that the tag sequence did not change the E protein localization (Supplementary Fig. 1). These results suggest that nsp3 and E protein colocalize in the cytoplasm of infected cells and that the absence of E protein did not change the subcellular distribution of nsp3.Fig. 7

Colocalization of nsp3 and E proteins. Vero E6 cells grown on glass coverslips were infected with rSARS-CoV-wt, rSARS-CoV-∆E or rSARS-CoV-EtagCt at an moi of 0.5. At 15 h post-infection the cells were fixed with 8% paraformaldehyde. Cells were labelled with nsp3 (red) or HA (green) specific antibodies.

To analyze whether E protein is ubiquitinated, a plasmid expressing E protein was cotransfected with a plasmid expressing a modified ubiquitin (Ub-OK) fused to a His tag into VeroE6 cells. The modified ubiquitin prevents the formation of poly-ubiquitin chains and the subsequently degradation of poly-ubiquitinated proteins by the proteasome. The cells were lysed 24 h post transfection, the proteins were resolved by SDS-PAGE and E protein was detected by Western blotting using an antibody E protein specific (Fig. 8A). An additional band corresponding to monoubiquitinated E protein was detected only when E protein was expressed in presence of Ub-OK (Fig. 8A, lane 3). By contrast, samples from cells expressing E protein (Fig. 8A, lane 2) or Ub-OK (Fig. 8A, lane 1) alone did not show the novel band. The promyelocytic leukemia protein (PML) was expressed in parallel to serve as a positive control of ubiquitination (Fig. 8B). In this case, different lower migrating bands corresponding to different ubiquitination states of the protein were observed. This result indicates that SARS-CoV E protein was ubiquitinated in transfected cells.Fig. 8

Ubiquitination of SARS-CoV E protein. Ubiquitin conjugation to E protein. Vero E6 cells were transiently transfected with the plasmid mixtures shown in the figure. The control plasmid pcDNA was used to equalize the total amount of transfected plasmid in all cases. 24 h after transfection, the cells were lysed by Laemmli lysis buffer followed by boiling for 5 min. The denatured lysates were analyzed by SDS-PAGE and Western blot with an antibody against E protein (A) or HA epitope (B). (C) In vitro ubiquitination of SARS-CoV E protein. Baculovirus purified E protein was incubated with mono-ubiquitin (mUb) or poly-ubiquitin (pUb) in the presence of the all the enzymes of the ubiquitination process. The reactions were resolved by SDS-PAGE and transferred to nitrocellulose. The ubiquitinated (Ub-E) and non-ubiquitinated E protein was detected by Western blot using a specific antibody against E protein.

African green monkey kidney-derived Vero E6 cells and the Huh-7.5.1 clone derived from the human hepatome Huh-7 cells were kindly provided by E. Snijder (University of Leiden, The Netherlands) and F. V. Chisari (Scripps Research Institute, La Jolla, California, USA), respectively. In both cases, cells were cultured in Dulbecco's modified Eagle medium (DMEM, GIBCO, Grand Island, NY, USA) supplemented with 25 mM HEPES and 10% fetal bovine serum (FBS, Biowhittaker, Berviers, Belgium). Virus growth and titrations were performed in Vero E6 cells following standard procedures previously described in detail (DeDiego et al., 2007). All work with infectious virus was performed in biosafety level (BSL) 3 facilities by personal wearing positive-pressure air purifying respirators (3 M HEPA AirMate, Saint Paul, MN).

The pBAC-SARS-CoV-EtagCt plasmid encoding a rSARS-CoV expressing the E gene fused to a tag consisting on the FLAG and the HA epitopes separated by TEV cleavage site, was constructed from a previously generated full-length infectious cDNA clone (plasmid pBAC-SARS-CoV^FL) (Almazan et al., 2006). The tagged E gene was introduced by overlap extension PCR using the plasmid pBAC-SARS-CoV^FL. The oligonucleotides SARS-E-VS (5′-CTCTTCAGGAGTTGCTAATCCAGCAATGG-3′) and SARS-EtagCt-RS (5′-GAGCTCACCCTGAAAATACAAATTCTCCTTATCGTCGTCATCCTTGTAATCGACCAGAAGATCAGGAACTCC-3′), which includes the TEV site and the FLAG epitope followed by a sequence complementary from the nucleotides (nt) 26325 to 26348 of the SARS-CoV genome, were used to generate a PCR product from nt 26018 to 26348 of the SARS-CoV genome. The oligonucleotides SARS-EtagCt-VS (5′-GAGAATTTGTATTTTCAGGGTGAGCTCTACCCATACGATGTTCCAGATTACGCTTAAGATCTTCTGGTCTAAACGAACTAACTATTATTATTATTC-3′), which contains the TEV site and the HA epitope followed by a sequence from nt 26334 to 26372 of the SARS-CoV genome, and SARS-29794-RS (5′-CCAGGTCGGACCGCGAGGAGGTG-3′) were used to generate a PCR product spanning nt 26334 to 29794 of the pBAC-SARS-CoV^FL plasmid. Both overlapping products were used as templates for PCR amplification using primers SARS-E-VS and SARS-29794-RS. The final PCR product was digested with the enzymes BamHI and RsrII and cloned into pBAC-SARS-CoV^FL digested with the same enzymes to generate the plasmid pBAC-SARS-CoV-EtagCt.

Baby hamster kidney (BHK) cells were grown to 90% confluence in 12.5 cm² flasks and were transfected with 6 µg of pBAC-SARS-CoV-Etag-Ct or pBAC-SARS-CoV^FL, as a control, using 18 µg of Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. Recovered viruses were cloned by plaque titration as described in detail (Almazan et al., 2006, DeDiego et al., 2007).

Vero E6 or Huh-7.5.1 cells grown to 90% confluence were infected at a multiplicity of infection (moi) of 0.05 with the viruses rSARS-CoV-wt, rSARS-CoV-∆E or rSARS-CoV-EtagCt. Culture supernatants were collected at different times post-infection, and virus titers were determined as previously described (DeDiego et al., 2007).

Proteins were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to a nitrocellulose membrane by wet immunotransfer and processed for Western blotting. The blots were probed with monoclonal antibodies specific for FLAG tag (dilution 1:1000; Sigma), HA tag (dilution 1:10,000; Sigma) or PRRSV Gp5 protein (dilution 1:1000; INGENASA, Madrid, Spain) or polyclonal antibodies against E (dilution 1:6000; kindly provided by Shen Shuo, Institute of Molecular and Cellular Biology, Singapore), nsp3 (dilution 1:2000; Rockland, Gilbertsville, PA), or N proteins (dilution 1:2000; Imgenex). Horseradish peroxidase-conjugated antibodies against the different species and the Immobilon Western chemiluminescent substrate (Millipore) were used to detect bound antibodies. Chemiluminescence was detected by exposure to Agfa X-ray film.

Recombinant histidine-tagged SARS-CoV E and PRRSV Gp5 proteins were expressed in the baculovirus/insect cell system. Recombinant proteins were purified to near homogeneity by Ni²⁺-NTA affinity chromatography (J. L. Nieto-Torres, M. L. DeDiego, E. Alvarez, and L. Enjuanes, CNB-CSIC, Madrid, Spain, unpublished results).

Purified E protein was the basis to generate the SARS-CoV E protein monoclonal antibody by immunizing BALB/c mice. Positive hybridoma clones were selected by ELISA, immunofluorescence and Western blot (J. L. Nieto-Torres, M. L. DeDiego, E. Alvarez, and L. Enjuanes, CNB-CSIC, Madrid, Spain, unpublished results).

Viral RNA synthesis was quantified by Q-RT-PCR. cDNAs were synthesized from 50 ng of total RNA extracted from SARS-CoV-wt, SARS-CoV-∆E or SARS-CoV-EtagCt infected cells using specific reverse sense oligonucleotides to genomic or subgenomic messenger RNA (sgmRNA) of genes M, 6 and N (DeDiego et al., 2007). The High Capacity cDNA Transcription kit (Applied Biosystems) was used in reverse transcription following the manufacturer's instructions. To analyze viral RNA synthesis a SYBR green assay (Applied Biosystems) was used. PCRs were performed using a virus sense primer complementary to the leader sequence, and specific reverse oligonucleotides to genomic or sgmRNAs (DeDiego et al., 2007). Data were acquired with an ABI PRISM 7000 sequence detection system (Applied Biosystems) and analyzed with ABI PRISM 7000 SDS version 1.0 software. Levels of viral RNAs are represented in comparison to reference levels from cells infected with SARS-CoV-wt.

Vero E6 were grown to 90% confluence and infected at an moi of 0.1 with rSARS-CoV-wt or rSARS-Co-EtagCt. After an adsorption period of 1 h the inocula medium was replaced with fresh DMEM containing 10% FBS. At 40 h post-infection the cells were washed twice with ice-cold phosphate-buffered saline (PBS), scrapped off the plates and lysed in a buffer containing 10 mM Tris–HCl pH 8, 150 mM NaCl, 0.5 mM EDTA, 1% IGEPAL CA-630 (Octylphenyl-polyethylene glycol, Sigma) and protease inhibitor cocktail (Roche) and incubated at 4 °C for 20 min. The extracts were clarified by centrifugation at 10,000 ×  g for 15 min. The supernatant was collected and the total protein concentration was determined with Coomassie Plus protein Assay (Pierce) and cell extracts were aliquoted to avoid repeated freeze/thaw cycles.

Cell extracts (4 ml) at a protein concentration of ∼ 5 mg/ml were incubated with 30 µl of Red Anti-HA Affinity Gel (Sigma) overnight at 4 °C in an orbital shaker. Agarose beads were washed 10 times with 10 volumes of wash buffer containing 10 mM Tris–HCl pH 8, 150 mM NaCl, 0.5 mM EDTA, and 0.1% IGEPAL CA-630 and then, protein complexes were eluted from the matrix by incubation with 500 U of AcTEV protease (Invitrogen) for 3 h at room temperature. Eluted complexes were incubated with 30 µl anti-FLAG M2 Affinity Gel (Sigma) overnight at 4 °C and then beads were washed 10 times with 10 volumes of wash buffer. Protein complexes were eluted by competition with FLAG peptide (Sigma) at a final concentration of 1 mg/ml diluted in a buffer containing 10 mM Tris–HCl pH 8, 300 mM NaCl, 0.5 mM EDTA, and 0.1% IGEPAL CA-630. Proteins were precipitated by trichloroacetic acid at a final concentration of 10 % (v/v) and then were resuspended in 30 µl of NuPage Sample buffer (Invitrogen) and incubated 10 min at 65 °C. Purified proteins were loaded into 1.0 mm NuPAGE 4–12% Bis-Tris gels (Invitrogen) and electrophoresis was performed at 100 V using the MES SDS running buffer from the same manufacturer. The gels were washed three times in deionized water, stained with Coomassie blue Simply Blue Safe Stain (Invitrogen), and the protein bands were excised from the gels for their identification by mass spectrometry.

Excised protein bands were in-gel digested with sequencing grade modified porcine trypsin (Promega). Peptides were extracted from gel bands in 0.5% trifluoroacetic acid, dried by speed vacuum centrifugation and resuspended in 4 µl of MALDI solution. A 0.8 µl aliquot of each digestion was deposited and dried onto a 2386-well OptiTOF™ plate (Applied Biosystems) and co-crystallized with 0.8 µl of matrix solution (3 mg/ml CHCA in MALDI solution). Samples were analyzed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry in an ABI 4800 MALDI TOF/TOF mass spectrometer (Applied Biosystems) at the proteomic facility of the National Center of Biotechnology (CNB-CSIC, Madrid, Spain). Data were analyzed using the ABI 4000 series explorer Software v3.6 and searches were performed with the MASCOT software v2.1 (Matrix Science) against the MSDB from the non-redundant NCBI protein database with mass tolerance of 100 ppm. No restrictions were imposed for protein molecular weight, although in some cases taxonomy restrictions for the viral, human or mammalian databases were included.

The plasmids used to generate the SARS-CoV nsp3 protein fragments were engineered by inserting the PCR products digested with EcoRI/XhoI in the same sites of pcDNA3 (+) plasmid. The corresponding nucleotides (nt) of each fragment in the SARS-CoV genome are the following: i) fragment 1 (F1), nt 2719–4896; ii) fragment 2 (F2), nt 3886–5829 ; iii) fragment 3 (F3), nt 4888–6672; iv) fragment 4 (F4), nt 6397–8484; v) fragment UB1-AC, nt 2719–3055; vi) fragment ADRP, nt 3269–3814 ; and vii) fragment SUD, nt 3815–4896. The in vitro transcription/translation reaction was carried out with the TNT® coupled reticulocyte lysate system (Promega) according to the manufacturer's instructions. Biotinylated Lys-tRNA (Promega) was used in translation reactions for subsequent detection of the expressed proteins with streptavidin conjugated to peroxidase (dilution 1:5000).

The pcDNA-E plasmid used to express the SARS-CoV E protein was engineered by inserting the PCR product digested with EcoRI/XhoI in the same sites of pcDNA3.1 (+) plasmid.

The pcDNA-UB-OK plasmid that expresses a mutated ubiquitin fused to His tag, which is unable to form poly-ubiquitin chains, was kindly provided by Manuel S. Rodriguez (CIC-BIOGUNE, Spain).

The HA-PMLIV plasmid, expressing the PML protein fused to HA epitope, and used as ubiquitination positive control in culture cells, was kindly provided by Jin-Hyun Ahn (Sungkyunkwan University School of Medicine, South Korea).

For immunoprecipitation assays Protein A/G Plate IP Kit (Pierce) was used following the manufacturer's instructions. Briefly, coated plates were incubated with the appropriate antibodies (polyclonal anti-nsp3 or monoclonal anti-E) diluted in the immunoprecipitation buffer (PBS, 1% Surfactant-Amps X-100) for 2 h at room temperature. Then, antigen samples were diluted with 1 volume of immunoprecipitation buffer and were incubated in the coated plates overnight at 4 °C. The wells were washed five times with 200 µl of immunoprecipitation buffer and then immune complexes were eluted with 50 µl of elution buffer. Analysis of precipitate complexes was carried out by SDS-PAGE and Western blotting.

For confocal microscopy, Vero E6 cells grown on glass coverslips were infected with rSARS-CoV, rSARS-CoV-∆E and rSARS-CoV-EtagCt at an moi of 0.5. At 15 h post-infection the growth medium was removed and cells were washed twice with PBS and fixed with 8% paraformaldehyde for 30 min at room temperature. Then, cells were washed twice in PBS, and permeabilized for 10 min with 0.2% Triton X-100 in PBS. All antibody incubations were carried out for 1 h in PBS containing 10% FBS. The immunofluorescence was done with monoclonal antibodies specific for HA tag (dilution 1:1000; Sigma), or polyclonal antibodies against E (dilution 1:2000; kindly provided by Shen Shuo, Institute of Molecular and Cellular Biology, Singapore) and nsp3 (dilution 1:500; Rockland, Gilbertsville, PA). Coverslips were washed three times with PBS between primary and secondary antibody incubations. Alexa 488- or Alexa 594-conjugated antibodies against the different species (dilution 1:500; Molecular Probes) were used as secondary antibodies. Coverslips were mounted in ProLong Gold anti-fade reagent (Invitrogen) and examined on a Leica SP5 confocal microscope (Leica Microsystems).

For a standard reaction, 1 mg of the baculovirus produced SARS-CoV E protein was incubated in a 10 μl reaction including an ATP regenerating system (50 mM Tris–HCl pH 7.6, 5 mM MgCl₂, 2 mM ATP, 10 mM creatine phosphate, 3.5 U/ml of creatine kinase and 0.6 U/ml of inorganic pyrophosphatase), 10 ng ubiquitin, 50 ng human E1 and 500 ng human E2 (Ubch5). Reactions were incubated at 37 °C for 2 h. After terminating the reactions with SDS sample buffer containing mercaptoethanol, reaction products were fractionated by SDS-PAGE. Detection of ubiquitinated and non-ubiquitinated E protein was done by Western blot using an antibody specific for E protein.