Histidine at Residue 99 and the Transmembrane Regi
http://www.100md.com
病菌学杂志 2005年第13期
Institute of Biotechnology, Department of Life Science, National Tsing-Hua University, Hsinchu 30013, Taiwan, Republic of China
ABSTRACT
The formation of the flavivirus prM-E complex is an important step for the biogenesis of immature virions, which is followed by a subsequent cleavage of prM to M protein through cellular protease to result in the production and release of mature virions. In this study, the intracellular formation of the prM-E complex of Japanese encephalitis virus was investigated by baculovirus coexpression of prM and E in trans in Sf9 insect cells as analyzed by anti-E antibody immunoprecipitation and sucrose gradient sedimentation analysis. A series of carboxyl-terminally truncated prM mutant baculoviruses was constructed to demonstrate that the truncations of the transmembrane (TM) region resulted in a reduction of the formation of the stable prM-E complex by approximately 40% for the TM1 (at residues 130 to 147 [prM130-147]) truncation and 20% for TM2 (at prM153-167) truncation. Alanine-scanning site-directed mutagenesis on the prM99-103 region indicated that the His99 residue was the critical prM binding element for stable prM-E heterodimeric complex formation. The single amino acid mutation at the His99 residue of prM abolishing the prM-E interaction was not due to reduced expression or different subcellular location of the mutant prM protein involved in prM-E interactions as characterized by pulse-chase labeling and confocal scanning microscopic analysis. Recombinant subviral particles were detected in the Sf9 cell culture supernatants by baculovirus coexpression of prM and E proteins but not with the prM H99A mutant. Sequence alignment analysis was further conducted with different groups of flaviviruses to show that the prM H99 residues are generally conserved. Our findings are the first report to characterize the minimum binding elements of the prM protein that are involved in prM-E interactions of flaviviruses. This information, concerning a molecular framework for the prM protein, is considered to elucidate the structure/function relationship of the prM-E complex synthesis and provide the proper trajectory for flavivirus assembly and maturation.
INTRODUCTION
Japanese encephalitis virus (JEV), of the genus Flavivirus of the family Flaviviridae, is an arthropod-borne virus that causes encephalitis in Asia with a high morbidity and mortality rate (4, 15). The genomic RNAs of all flaviviruses consist of two noncoding regions at the 5' and 3' ends and the coding sequence, which encodes the three structural proteins, i.e., the core (C), the precursor membrane (prM) or the membrane (M), and the envelope (E), and the seven nonstructural proteins, i.e., NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5. During flavivirus assembly, the prM and E proteins interact with each other to generate a prM-E heterodimeric complex in the endoplasmic reticulums (ERs) of the infected cells (16, 17, 18). In the intracellular synthesis of the immature virions, the viral genomic RNA is encapsulated with the C protein and the surface lipid bilayer, into which the prM-E heterodimeric complex is inserted. Next, the immature virions transported through the secretary pathway are converted into mature virions by the cleavage of the prM protein to the M protein by cellular furin or a furin-like protease (8). The prM protein, by interacting with the E protein, can also protect the immature virions against inactivation in the acidic vesicles and promote the appropriate transporting/folding of the stable complex of the E protein (8). In tick-borne encephalitis (TBE) virus, coexpression of prM and E proteins alone without viral genomic RNA and C protein resulted in formation of recombinant subviral particles (RSPs) (3, 17). The lateral interactions between prM and E proteins are a crucial step to stabilize subviral particle formation by flaviviruses in cultured cells (6).
Although the prM/M structure of flavivirus has yet to be determined, the X-ray structures of the E protein have been solved for TBE virus (26) and recently for dengue virus (20, 21). High-resolution images obtained from cryoelectron microscopy of dengue (DEN) and yellow fever (YF) viruses reveal that the prM-E heterodimers are laterally arranged on the surfaces of the immature virions (30, 31). The prM-E interacting complex has also been characterized using immunoprecipitation on the native virions of dengue (27), TBE (7) and yellow fever (24) viruses. Recently, the minimum functional elements of the E protein involved in the prM-E interaction of TBE virus were identified at the stem region close to transmembrane regions at the carboxyl-terminal end of E protein (3).
Our present work provides the first evidence of the location and characteristics of the prM binding elements involved in prM-E heterodimeric interaction of JEV, using a baculovirus expression system. The intracellular formation of the prM-E complexes of JEV was investigated in Sf9 cells by baculovirus coexpression of prM and E genetically in trans, immunoprecipitation by the JEV E protein-specific monoclonal antibody (MAb) E3.3, and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis. A series of carboxyl-terminally truncated prM mutant baculoviruses was constructed to map the minimum binding elements of prM involved in prM-E interactions. The formation of the stable prM-E heterodimeric complex and the release of RSPs were further characterized using sucrose gradient sedimentation, pulse-chase labeling, double-staining confocal microscopy, and transmission electron microscopy (TEM). Sequence alignment analysis of the prM protein was finally conducted with 11 groups of flaviviruses, including JEV and Kunjin, Murray Valley encephalitis, St. Louis encephalitis virus, West Nile, DEN type 1 to 4 (DEN1 to -4), TBE, and YF viruses. This is the first report to characterize the minimum binding elements of prM protein involved in prM-E interactions in flavivirus assembly and maturation.
MATERIALS AND METHODS
Cell lines and viruses. C6/36 cells were grown in Hanks minimal essential medium supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Invitrogen) at 28°C. Spodoptera frugiperda Sf9 cells were grown at 28°C in TNM-FH medium (Sigma) supplemented with 10% FBS. Hybridoma cells producing MAbs E3.3 and 5B1 were grown at 37°C in Iscove's modified Dulbecco's medium (Invitrogen) supplemented with 10% FBS. The JEV strain CH2195LA (Gene Bank accession no. AF221499) was produced in C6/36 cells. The recombinant baculoviruses for plaque selection and propagation in this study were in Sf9 cells.
Construction of recombinant baculoviruses. The genomic RNA of the JEV CH2195LA strain was extracted from the infected C6/36 cell culture supernatants by using the Trizol kit (Invitrogen), followed by the addition of Superscript II and Elongase (Invitrogen) for reverse transcriptase PCR to obtain the cDNAs of the prME-, prM-, and E-coding genes by using the following oligonucleotide primers: (i) for the prME gene, forward primer 5'-ATgcgggATccATgggAAATGAAGCCTCAATG-3' and reverse primer 5'-cgcgAATTCCTAAGCATGCACATTGGTCGC-3'; (ii) for the prM gene, forward primer 5'-ATGCGGGATCCATGGGAAATGAAGCCTCAATG-3' and reverse primer 5'-GGGAATTCCTAGAAACTGTAAGCCGGAGCGACC-3'; (iii) for the E gene, forward primer 5'-CGCGGATCCATGATGCTTGGCAGTAACAACGGTCAACGCG-3' and reverse primer 5'-ATGGGAATTCCTACATTTTTAGCCTGCATTTCAG-3'; (iv) for the carboxyl-terminally truncated prM genes, reverse primers for (a) prM147tr (5'-GGGAATTCCTAACTGCCAAGCATCCAGCC-3'), (b) prM130tr (5'-GCGAATTCCTACCTTATGATCCAGTTCTCAGT-3'), (c) prM126tr (5'-GGGAATTCCTAGTTCTCAGTTTTCATGAGATA-3'), (d) prM108tr (5'-GGGAATTCCTATTTTTTATTCACTAGTGAACT-3'), (e) prM104tr (5'-GGGAATTCCTATAGTGAACTCTCCCCATGTGT-3'), (f) prM99tr (5'-GGGAATTCCTAATGTGTTTGGACCGATACGGA-3'), (g) prM96tr (5'-GGGAATTCCTAGACCGATACGGATCTCCTGCT-3'), and (h) prM92tr (5'-GCGAATTCCTATCTCCTGCTTCGCTTGGAATG-3'); and (v) for the point mutation prM genes, reverse primers for (a) Bac-prM-H99A (5'-GTCCAAACAGCTGGGGAGAGT-3' and 5'-ACTCTCCCCAGCTGTTTGGAC-3'), (b) Bac-prM-G100A (5'-CAAACACATGCCGAGAGT-3' and 5'-ACTCTCGGCATGTGTTTG-3'), (c) Bac-prM-E101A (5'-CATGGGGCCAGTTCACTA-3' and 5'-TAGTGAACTGGCCCCATG-3'), (d) Bac-prM-S102A (5'-CATGGGGAGGCCTCACTA-3' and 5'-TAGTGAGGCCTCCCCATG-3'), and (e) Bac-prM-S103A (5'-GGGGAGAGTGCCCTAGTG-3' and 5'-CACTAGGGCACTCTCCCC-3'). All the forward primers contained a BamHI restriction site, and the reverse primers contained an EcoRI restriction site for ligation to the pBlueBac4 baculovirus vector (Invitrogen). All of the sequences were confirmed by DNA sequence analysis (Mission Biotech Inc., Taipei, Taiwan). To generate the recombinant baculoviruses, the constructed pBlueBac4 plasmids were cotransfected with the linearized baculovirus Bac-N-Blue DNA (Invitrogen) into Sf9 cells treated with Cellfectin (Invitrogen) and incubated for 5 to 7 days. Recombinant baculoviruses were then picked from the blue-staining plaques formed in the infected Sf9 cells overlaid with agarose medium containing 5-bromo-4-chloro-3-indoyl-?-D-galactoside (X-Gal). Three plaque purifications were conducted to obtain the recombinant baculoviruses used for the studies.
Metabolic labeling and immunoprecipitation. Subconfluent monolayers of Sf9 cells in six-well plates were infected with recombinant baculoviruses or JEV at a multiplication of infection (MOI) of 5. At 24 h postinfection, the TNM-FH medium was removed and replaced with TNM-FH medium containing 0.01% normal methionine (Met) and 0.02 to 0.10 mCi/ml of [35S]methionine-cysteine. Baculovirus-infected Sf9 cells were treated with a lysis buffer (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 2 mM EDTA, 1% Triton X-100, and 1 mM phenylmethylsulfonyl fluoride) and precleared using protein A-Sepharose (Pharmacia). The purified cell lysates were then reacted with MAb E3.3 for 5 h at 4°C, followed by incubation with a Sepharose slurry for 16 h at 4°C. The immunoprecipitated samples were further separated by centrifugation at 200 x g for 3 min, and the pellet was washed twice with the incubation buffer (63 mM Tris-HCl [pH 6.8]). The precipitated samples were solubilized by heating (95°C for 5 min) with 40 μl of electrophoresis reducing sample buffer (50 mM Tris-HCl [pH 6.8], 2% SDS, 0.1% bromophenol blue, 5% 2-mercatoethanol, and 10% glycerol) and further analyzed by SDS-PAGE (12%) and fluorography.
Cross-linking analysis for intracellular formation of prM-E heterodimer. Sf9 cells plated in 60-mm dishes were infected with recombinant baculoviruses or JEV at an MOI of 5 for 24 h. The maintenance medium was removed, and the cells were labeled with [35S]methionine-cysteine as described above. The virus-infected cells were rinsed with phosphate-buffered saline and scraped into Microfuge tubes. The infected cells were then cross-linked by the addition of dithio-bis(succinimidylpropionate) (DSP) (Pierce) to a final concentration of 1 mM and further incubated for 1 h at 4°C. The reaction was then stopped by adding glycine (to a final concentration of 50 mM), and the mixture was further incubated for 15 min at 4°C. The precipitated samples were analyzed with both reducing sample buffer (50 mM Tris-HCl [pH 6.8], 2% SDS, 5% 2-mercaptoethanol, 0.1% bromophenol blue, and 10% glycerol) and nonreducing sample buffer (50 mM Tris-HCl [pH 6.8], 2% SDS, 0.1% bromophenol blue, and 10% glycerol) and analyzed by 7.5% SDS-PAGE and fluorography.
Sucrose gradient sedimentation analysis for stable prM-E complex. The cell lysates were solubilized in 1% Triton X-100 and then loaded for centrifugation in a 3 to 60% (wt/wt) sucrose gradient made with gradient buffer (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 2 mM EDTA, 0.5% Triton X-100, 1 mM phenylmethylsulfonyl fluoride). The gradients were centrifuged in a Hitachi RPS40ST rotor at 38,000 rpm and 15°C for 22 h, and each fraction was collected and immunoprecipitated with monoclonal antibody E3.3 (anti-JEV E protein) as described above. The immune complexes were analyzed by reducing SDS-PAGE and fluorography.
Pulse-chase analysis for stable prM-E complex. The Sf9 cells infected with recombinant baculoviruses were cultured on six-well subconfluent plates (MOI = 5) in TNM-FH medium for 24 h. The maintenance medium was removed, and then the cells were starved for 30 min in methionine-free medium. The cells were pulse-labeled with [35S]methionine-cysteine (0.25 mCi/ml) for about 30 min and incubated with TNM-FH medium for different chase times (0, 1, 2, 4, and 8 h). The immunoprecipitated cell lysed were resolved by 12% SDS-PAGE and further analyzed by fluorography.
Subcellular localization using confocal laser scanning microscopic analysis. Infected Sf9 cells grown on 15-mm glass coverslips at 24 h postinfection were washed three times with TBS (140 mM NaCl, 25 mM Tris-HCl, pH 6.1), incubated in medium containing the ER-staining dye DiIC13(3) (10 nM; Molecular Probes) for 30 min, and finally fixed with 3.7% paraformaldehyde for 1 h. The cells were then permeabilized with TBS containing 0.1% Triton X-100 for 1 h. Expression of the JEV proteins in these cells was revealed by staining with the anti-E MAb E3.3 or anti-prM MAb 5B1 followed by the addition of goat anti-mouse Alexa Fluor 647 immunoglobulin G (Molecular Probes). The subcellular localizations in these cells were examined in a confocal laser scanning microscope (Zeiss LSM 500) with a 100x 1.3 oil objective.
TEM for RSP visualization. Culture supernatants of baculovirus-infected Sf9 cells (MOI = 2.5) or JEV-infected Sf9 cells at 5 days postinfection were centrifuged, precipitated with 10% polyethylene glycol 8000, and resuspended in 1 ml TN buffer (10 mM Tris-HCl [pH 7.5], 100 mM NaCl). A continuous sucrose gradient (3 to 60%, wt/wt) was applied, and centrifugation was carried out at 4°C and 38,000 rpm for 24 h (RPS40ST rotor and Hitachi 85P-72 ultracentrifuge). The virus particles were collected from 13 tubes of 0.85 ml for each fraction for TEM visualization. A single-droplet negative staining procedure was used in parallel for TEM visualization. Droplets of 30-μl samples were absorbed for 10 min onto Formvar carbon-coated cooper grids (200 mesh; Agar Scientific), washed with distilled water, and then stained with 2% uranyl acetate for 1 min. After air drying, the grids were examined under a Hitachi H-7500 TEM at 100 kV.
RESULTS AND DISCUSSION
Detection of prM-E interaction in baculovirus-infected Sf9 cells. The prME-, prM-, and E-encoding cDNAs of the JEV CH2195LA strain (14, 28, 29) were cloned into the baculoviruses (Bac-prME, Bac-E, and Bac-prM) with a polyhedron promoter (PPH) control (Fig. 1A) to study the prM-E interaction in Sf9 insect cells. Sf9 insect cells were initially infected with the recombinant baculovirus, radiolabeled with [35S]methionine-cysteine medium for 24 h, and then immunoprecipitated using the JEV E-protein specific MAb E3.3. The intracellular formation of prM-E complex in Sf9 insect cells was characterized using (i) a single recombinant baculovirus that carries the precursor prME gene (Bac-prME) and (ii) two separate recombinant baculoviruses that carry the prM and E genes (Bac-prM and Bac-E). The expression cassettes for the precursor prME in cis or the separate prM and E in trans all include the SP1 and SP2 signal peptide sequences derived from the carboxyl-terminal ends of the C and prM proteins (Fig. 1A), as the signal peptide sequences are responsible for stable insertion into the ER membrane to properly modify the prM-E complex formation (1, 2, 19, 24).
Formation of prM-E heterodimers was found in baculovirus expression of the prME precursor protein in cis and the prM and E proteins in trans, revealing two reactive bands, corresponding to prM (21 kDa) and E (56 kDa) in SDS-polyacrylamide gels, detected in Sf9 cells after immunoprecipitation (Fig. 1B). Only the E protein (and not the prM protein) was detected in Sf9 cells infected with Bac-E (Fig. 1B). Simultaneously, Sf9 cells and C6/36 cells (the mosquito cell line commonly used for JEV isolation) were infected with the JEV 2195LA strain at an MOI of 1, and the cell lysates were also found to form the prM-E heterodimers after immunoprecipitation (Fig. 1B). The use of a baculovirus expression system to analyze the intracellular formation of the prM-E complex in this study is appropriate, since Sf9 cells were susceptible to JEV infection (Fig. 1C). This result is in agreements with two recent reports that Sf9 cells are suitable for persistent JEV infection (10, 11). Although our findings in Fig. 1C showed that the replication kinetics of JEV in Sf9 cells was somewhat slower than that in C6/36 cells (i.e., the maximum titer decreased by approximately 30-fold), Sf9 cells can still be suitable to dissect the molecular interactions of prM and E proteins during JEV assembly and maturation.
To provide further evidence of the formation of noncovalent prM-E heterodimers, DSP was used as a cross-linking agent to preserve protein-protein interactions in lysis buffer (containing 1% Triton X-100). DSP is a thiol-cleavable molecule with two amine-reactive groups that is used to preserve the physical interaction between two proteins and has been successfully used for characterizing the coronavirus E and M proteins (5). Our results revealed that DSP-treated Sf9 cells either infected with Bac-prME or coinfected with Bac-prM and Bac-E yielded the formation of the noncovalent prM-E heterodimer (70 to 80 kDa) under nonreducing SDS-PAGE analysis (Fig. 1D). The noncovalent prM-E heterodimers were also present in the DSP-treated samples of Sf9 cells and C6/36 cells infected with the JEV CH2195LA strain (Fig. 1D). The formation of the prM-E heterodimers was effective and stable in the infected cells as demonstrated by sucrose gradient sedimentation and pulse-chase labeling analyses.
Mapping of the prM binding elements involved in prM-E heterodimeric interaction in Sf9 cells. The flavivirus prM protein has two transmembrane regions (TM1 and TM2) close to its C-terminal end, as reported for dengue virus (19) and yellow fever virus (24), so the TM1 and TM2 regions were predicted to be at prM residues 130 to 147 and 153 to 167 of the JEV CH2195LA strain (http://www.cbs.dtu.dk/services/TMHMM). A series of carboxyl-terminally truncated prM baculoviruses was constructed, including Bac-prM147tr (with only the TM2 truncation), Bac-prM130tr (with only the TM1 and TM2 truncations), and Bac-prM126tr, Bac-prM108tr, Bac-prM104, Bac-prM99, Bac-prM96, and Bac-prM92tr for fine mapping of the prM binding elements involved in prM-E heterodimeric interactions (Fig. 2A). Following immunoprecipitation by MAb E3.3, the intensity of the E protein-reactive bands was almost constant, but the intensity of the prM-reactive bands decreased or even became undetectable (Fig. 2B). The reduced binding intensity of TM-truncated prMs (78% for Bac-prM147tr and 43% for Bac-prM130tr) exceeded those governed by further carboxyl-terminal truncations (36% for Bac-prM126tr, 26% for Bac-prM108tr, and 24% for Bac-prM104tr) for the formation of prM-E complex (Fig. 2C). Moreover, the prM protein of the prM-E complex was not detected in Sf9 cells coinfected with Bac-prM99tr, Bac-prM96tr, and Bac-prM92tr (Fig. 2B). These results indicated that the TM regions and prM residues 99 to 103 were also important for the stable formation of the prM-E complex in the coinfected Sf9 cells.
Sucrose gradient sedimentation analysis of prM TM region affecting stable prM-E complex formation. To further verify whether the TM regions considerably influence the formation of the stable prM-E complex in Sf9 cells, cell lysates before immunoprecipitation were subjected to sucrose gradient sedimentation. These experiments showed that the Sf9 cells coinfected with Bac-prM formed a stable prM-E heterodimeric complex, mainly at fractions 5 and 6 (Fig. 3A), while the prM-E heterodimeric complex was detected at the same fractions 5 and 6 for Bac-prM147tr and Bac-prM130tr but the intensity of the prM binding protein was lower (Fig. 3B and C). The percentage of dimerization of prM was taken as the prM/E ratio, measured for the stable prM-E complex in each sucrose gradient fractionation, to compare the reduction in intensity of the prM protein from that of the most constant E proteins detected in the sucrose-fractionated samples. The results indicated that most prM-E heterodimers were in fractions 5 to 7 (Fig. 3D). The percentage of prM-E heterodimer formation, which is indicated as the prM/E ratio in Fig. 3D, declined from 100% for Bac-prM (fraction 6) to 80% for Bac-prM147tr (fraction 5) and to 40% for Bac-prM130tr (fraction 5). The truncations of both TM1 and TM2 (Bac-prM130tr) resulted in a reduction of the formation of the heterodimer by approximately 40%, compared to TM2 truncation with a 20% loss. This result suggests that TM1 is more important than TM2 in assisting the formation of the stable prM-E complex in the infected cells.
We also used homology modeling to predict the prM-E complex structure of the JEV CH2195LA strain based on the cryoelectron microscopic results published for the dengue virus prM and E membrane proteins (PDB 1P58) (30). The predicted prM structure contains only part of prM (prM residues 113 to 167 [prM113-167] of CH2195LA) according to the published cryoelectron microscopic structure (30). Our modeling structure indicates that TM1 (prM130-147) is topologically closer than TM2 (prM153-167) to the helix 2 region of TBE virus E protein that was previously reported to be the minimum E binding elements for the prM protein (3). However, we cannot rule out the possibility of the N-terminal portion of the M protein (i.e., prM93-112) being involved in prM-E interaction, as it has been recently speculated to cooperate with the hole between the two monomers to form the homodimer of the E ectodomain (30).
Site-directed mutagenesis studies on prM99-103, which is involved in prM-E interaction. Alanine-scanning site-directed mutagenesis of the prM99-103 region was performed by constructing five more prM mutant baculoviruses (Bac-prM-H99A, Bac-prM-G100A, Bac-prM-E101A, Bac-prM-S102A, and Bac-prM-S103A) to map the amino acid residue(s) that was most critically involved in prM-E interaction (Fig. 4A). The immunoprecipitation results indicated that the prM-reactive band was absent from the Sf9 cells coinfected with Bac-prM-H99A; however, both prM and E protein bands were detectable in the Sf9 cells coinfected with Bac-prM, Bac-prM-G100A, Bac-prM-E101A, Bac-prM-S102A, and prM-S103A (Fig. 4B). Additionally, sucrose gradient sedimentation analysis of the infected cell lysates showed that Sf9 cells coinfected with Bac-prM-H99A had fewer prM-reactive bands in fractions 5 and 6 (Fig. 5B) than those coinfected with Bac-prM, Bac-prM-G100A, Bac-prM-E101A, Bac-prM-S102A, and Bac-prM-S103A (Fig. 5A, C, D, E, and F). Therefore, the His99 residue was demonstrated to be the critical element for the stable prM-E heterodimeric complex formation.
Pulse-chase kinetics, subcellular localization, and RSP formation of the prM-E stable complex affected by the prM His99 mutation. Pulse-chase experiments were performed in a time course from 0 to 8 h of chase to investigate the kinetics of the newly synthesized prM-E complex affected by the single amino acid mutation at prM His99. The newly synthesized prM-E complex was evident at 0 h of chase and slowly decreased in amount to 8 h of chase in the Sf9 cells coinfected with Bac-prM (Fig. 6A). The results agree with those previously obtained from Vero cells infected with dengue virus, which showed that the prM-E complex formed persisted for about 4 to 6 h (27). However, no prM-reactive protein was detectable in the cells coinfected by the prM mutation (Bac-prM-H99A) from 1 to 8 h of chase compared to the newly synthesized E protein (Fig. 6B). To further confirm the expression of the mutant prM in the infected cells, a pig anti-JEV serum was used to immunoprecipitate the prM and E proteins. The pulse-chase results showed that Sf9 cells coinfected with Bac-prM-H99A showed both prM and E proteins, indicating that prM and E are both expressed and that the mutated form of prM is not degraded as shown from 0 to 8 h of chase (Fig. 6C).
The subcellular localization of the prM-E complex in Sf9 cells was further investigated using confocal laser scanning microscopy. The Sf9 cells coinfected with Bac-prM-H99A, along with the controls of Bac-prME and JEV infection, were doubly stained with a prM-specific MAb and DiIC16(3), an ER-specific dye. The double-staining images obtained using confocal laser scanning microscopy showed that the intracellular distributions of prM and the mutant prM (prM H99A) almost overlapped with the ER marker (Fig. 7), suggesting that the loss of prM-E interaction in the prM with the mutation at the His99 residue did not occur because of the difference in their subcellular locations for prM-E interaction in Sf9 cells.
Furthermore, as the formation and release of RSPs by coexpression of prM and E proteins in mammalian cells was extensively reported (9, 12, 13, 22), we evaluated the RSP formation in baculovirus-infected Sf9 cells. Culture supernatants were collected from Sf9 cells infected with the JEV 2195LA strain, coinfected with Bac-prM (BacprM-H99A) and Bac-E. Samples were then collected from the fractions of sucrose gradient sedimentation for TEM visualization. The TEM results after negative staining with uranyl acetate showed that the culture supernatants of JEV-infected Sf9 cells yielded formation of spherical particles as the infectious virions (i.e., 50 nm in diameter) in the sucrose gradient-fractionated samples (Fig. 8A). Baculovirus coexpression of prM and E in trans in Sf9 cells led to the formation and release of smaller RSPs with a diameter of about 30 to 50 nm (Fig. 8B). Coinfection of Sf9 cells with BacPrM-H99A and Bac-E did not result in the formation and release of RSPs in culture supernatants (Fig. 8C). Although all of these experiments demonstrated that the replacement of prM H99 by an Ala strongly reduced prM-E heterodimerization and abolished RSP secretion, we still cannot exclude the possibility that the His-to-Ala mutation may alter prM folding and that misfolded prM would not be able to interact with E because the region of interaction on prM is not properly exposed on the protein.
Sequence alignment analysis of prM99 and TM regions in flaviviruses. To further demonstrate whether the prM99 and two TM regions are important for a general interaction of prM and E proteins for flaviviruses, we analyzed the amino acid sequences of the flavivirus M protein sequences (prM93-167), including four groups of the JEV serocomplex viruses (JEV, Kunjin virus, Murray Valley encephalitis virus, St. Louis encephalitis virus, and West Nile virus) and six groups of other flaviviruses (DEN1, DEN2, DEN3, DEN4, TBE, and YF viruses) (Fig. 9). Multiple amino acid sequence alignment was done by using the Showalign program of the Jemboss software through the Jemboss server (http://bioinfo.life.nthu.edu.tw/bioinfo.htm). The results indicate that the prM H99 residues are all conserved in all the flavivirus groups (Fig. 9). In contrast, the amino acid residues at prM100-103 vary with each individual group: GE/DSS/T for the JEV serocomplex, VGL/MG for the DEN virus serocomplex, AQGE/D for TBE virus, and ENHG for YF virus. Other conserved amino residues were also identified in all groups of flaviviruses, including prM L104, prM W111, prM E125, and prM W127 (Fig. 9). However, the reduced binding intensities of prM to the prM-E complex by the TM-truncated mutants (Bac-prM104, Bac-prM108, Bac-prM126, and Bac-prM130) in Sf9 cells were less significant (Fig. 2B and C), indicating that the conserved residues at prM L104, prM W111, prM E125, and pM W127 in flaviviruses are unlikely to be involved in prM-E interaction in virus assembly. Therefore, the conserved residue at prM H99 may contribute solely to the structural integrity of the prM-E heterodimeric complex interaction in flavivirus assembly. However, the conservation of prM H99 among all the flaviviruses analyzed may not be necessary, in favor of a direct role for this residue being involved in protein-protein interaction, and that may come from some coevolution of the sequences involved in interactions within the prM and E proteins. It is also likely that a mutation in one of the envelope proteins might compensate for other mutations in the other envelope protein. Our present finding of the prM H99 residue being involved in JEV prM-E heterodimeric interactions does not agree with a recent study showing the histidine at M39 (equivalent to prM-131 in Fig. 9) as a critical residue for prM/E heterodimers of DEN2 virus assembly (25).
The amino acid sequences at the C terminus of the prM protein consist of two TM regions as reported for flaviviruses (19, 24), i.e., TM1 (prM130-147), TM2 (prM153-167), and the nearby connecting segment (four hydrophilic residues and one charged residue). No conserved residues were observed among TM1 and TM2 for all groups of flaviviruses; however, they contain at least 9 to 12 hydrophobic residues to form three or four alpha-helix turns that stably insert in the lipid bilayer. Based on the homology modeling and the cryoelectron microscopic results published for the dengue virus prM and E membrane proteins (PDB 1P58) (30), there are four alpha-helix turns in TM1 (prM132-135, prM136-139, prM140-143, and prM144-147) and three alpha-helix turns in TM2 (prM153-156, prM157-160, and prM161-164) predicted for the JEV CH2195LA strain. The TM1 region of hepatitis C virus envelope glycoprotein E1 has been demonstrated to contain four helix turns by using nuclear magnetic resonance analysis, where the GXXXG motif located in the first helix turn was shown to greatly affect the E1E2 heterodimerization (23). The presence of the GXXXG motif was found in the JEV serocomplex viruses at prM142-146 of TM1 (Fig. 9), and our preliminary results also showed a significant reduction of prM-E heterodimeric complex formation as demonstrated by alanine insertion on the GXXXG motif (data not shown). Although we were able to experimentally delineate that TM1 is more important than TM2 in assisting the prM-E stable complex based on prM C-terminal truncations (Fig. 2A to C), further studies are still needed to dissect the detailed molecular elements in TM and connecting segment regions involved in assisting the prM-E heterodimeric complex formation in flavivirus assembly.
In conclusion, the present study provides evidence that the JEV structural proteins prM and E interact with each other to form a stable heterodimeric complex in Sf9 cells. His99 and the TM regions of the prM protein were identified to be the most important regions involved in the interaction with E protein in JEV assembly and maturation. This information, concerning a molecular framework for the prM protein, is considered to elucidate the structure/function relationship of the prM-E complex synthesis for the immature virions in flavivirus assembly.
ACKNOWLEDGMENTS
This work was supported by the National Science Council grants NSC93-2214-E-007-015 and NSC93-3112-B-007-015 and the National Health Research Institutes, Taiwan, Republic of China.
REFERENCES
Allison, S. L., J. Schalich, K. Stiasny, C. W. Mandl, C. Kunz, and F. X. Heinz. 1995. Oligomeric rearrangement of tick-borne encephalitis virus envelope proteins induced by an acidic pH. J. Virol. 69:695-700.
Allison, S. L., K. Stadler, C. W. Mandl, C. Kunz, and F. X. Heinz. 1995. Synthesis and secretion of recombinant tick-borne encephalitis virus protein E in soluble and particulate form. J. Virol. 69:5816-5820.
Allison, S. L., K. Stiasny, K. Stadler, C. W. Mandl, and F. X. Heinz. 1999. Mapping of functional elements in the stem-anchor region of tick-borne encephalitis virus envelope protein E. J. Virol. 73:5605-5612.
Burke, D. S., and T. P. Monath. 2001. Flaviviruses, p. 1043-1125. In D. M. Knipe and P. Howley (ed.), Fields virology. Lippincott-Raven, Philadelphia, Pa.
Corse, E., and C. E. Machamer. 2003. The cytoplasmic tails of infectious bronchitis virus E and M proteins mediate their interaction. Virology 312:25-34.
Ferlenghi, I., M. Clarke, T. Ruttan, S. L. Allison, J. Schalich, F. X. Heinz, S. C. Harrison, F. A. Rey, and S. D. Fuller. 2001. Molecular organization of a recombinant subviral particle from tick-borne encephalitis virus. Mol. Cell 7:593-602.
Heinz, F. X., K. Stiasny, A. Puschner-Auer, H. Holzmann, S. L. Allison, C. W. Mandl, and C. Kunz. 1994. Structural changes and functional control of the tick-borne encephalitis virus glycoprotein E by the heterodimeric association with protein prM. Virology 198:109-117.
Heinz, F. X., and S. L. Allison. 2003. Structures and mechanisms in flavivirus fusion. Adv Virus Res. 55:231-269.
Hunt, A. R., C. B. Cropp, and G. J. Chang. 2001. A recombinant particulate antigen of Japanese encephalitis virus produced in stably-transformed cells is an effective noninfectious antigen and subunit immunogen. J. Virol. Methods 97:133-149.
Jeong, H. S., J. H. Shin, Y. N. Park, J. Y. Choi, Y. L. Kim, B. G. Kim, S. R. Ryu, S. Y. Baek, S. H. Lee, and S. N. Park. 2003. Development of real-time RT-PCR for evaluation of JEV clearance during purification of HPV type 16 L1 virus-like particles. Biologicals 31:223-229.
Kim, H., S. J. Lee, J. Y. Park, Y. W. Park, H. S. Kim, H. Y. Kang, B. K. Hur, Y. W. Ryu, S. I. Han, and J. S. Kim. 2004. Study on persistent infection of Japanese encephalitis virus Beijing-1 strain in serum-free Sf9 cell cultures. J. Microbiol. 42:25-31.
Kojima, A., A. Yasuda, H. Asanuma, T. Ishikawa, A. Takamizawa, K. Yasui, and T. Kurata. 2003. Stable high-producer cell clone expressing virus-like particles of the Japanese encephalitis virus e protein for a second-generation subunit vaccine. J. Virol. 77:8745-8755.
Konishi, E., A. Fujii, and P. W. Mason. 2001. Generation and characterization of a mammalian cell line continuously expressing Japanese encephalitis virus subviral particles. J. Virol. 75:2204-2212.
Lin, C. W., and S. C. Wu. 2003. A functional epitope determinant on domain III of the Japanese encephalitis virus envelope protein interacted with the neutralizing antibody combining sites. J. Virol. 77:2600-2606.
Lindenbach, B. D., and C. M. Rice. 2001. Flaviviridae: the viruses and their replication, p. 991-1041. In D. M. Knipe and P. Howley (ed.), Fields virology. Lippincott-Raven, Philadelphia, Pa.
Lorenz, I. C., S. L. Allison, F. X. Heinz, and A. Helenius. 2002. Folding and dimerization of tick-borne encephalitis virus envelope proteins prM and E in the endoplasmic reticulum. J. Virol. 76:5480-5491.
Lorenz, I. C., J. Kartenbeck, A. Mezzacasa, S. L. Allison, F. X. Heinz, and A. Helenius. 2003. Intracellular assembly and secretion of recombinant subviral particles from tick-borne encephalitis virus. J. Virol. 77:4370-4382.
Mackenzie, J. M., and E. G. Westaway. 2001. Assembly and maturation of the flavivirus Kunjin virus appear to occur in the rough endoplasmic reticulum and along the secretory pathway, respectively. J. Virol. 75:10787-10799.
Markoff, L., A. Chang, and B. Falgout. 1994. Processing of flavivirus structural glycoproteins: stable membrane insertion of premembrane requires the envelope signal peptide. Virology 204:526-540.
Modis, Y., S. Ogata, D. Clements, and S. C. Harrison. 2003. A ligand-binding pocket in the dengue virus envelope glycoprotein. Proc. Natl. Acad. Sci. USA 100:6986-6991.
Modis, Y., S. Ogata, D. Clements, and S. C. Harrison. 2004. Structure of the dengue virus envelope protein after membrane fusion. Nature 427:313-319.
Mutoh, E., T. Ishikawa, A. Takamizawa, T. Kurata, T. Sata, and A. Kojima. 2004. Japanese encephalitis subunit vaccine composed of virus-like envelope antigen particles purified from serum-free medium of a high-producer J12#26 cell clone. Vaccine 22:2599-2608.
Op De Beeck, A., R. Montserret, S. Duvet, S. Cocquerel, R. Cacan, B. Barberot Le Maire, F. Penin, and J. Dubuisson. 2000. The transmembrane domains of hepatitis C virus envelope glycoproteins E1 and E2 play a major role in heterodimerization. J. Biol. Chem. 275:31428-31437.
Op De Beeck, A., R. Molenkamp, M. Caron, A. Ben Younes, P. Bredenbeek, and J. Dubuisson. 2003. Role of the transmembrane domains of prM and E proteins in the formation of yellow fever virus envelope. J. Virol. 77:813-820.
Pryor, M. J., L. Azzola, P. J. Wright, and A. D. Davidson. 2004. Histidine 39 in the dengue virus type 2 M protein has an important role in virus assembly. J. Gen. Virol. 85:3627-3636.
Rey, F. A., F. X. Heinz, C. Mandl, C. Kunz, and S. C. Harrison. 1995. The envelope glycoprotein from tick-borne encephalitis virus at 2 A resolution. Nature 375:291-298.
Wang, S., R. He, and R. Anderson. 1999. PrM- and cell-binding domains of the dengue virus E protein. J. Virol. 73:2547-2551.
Wu, S. C., W. C. Lian, L. C. Hsu, and M. Y. Liau. 1997. Japanese encephalitis virus antigenic variants with characteristic differences in neutralization resistance and mouse virulence. Virus Res. 51:173-181.
Wu, S. C., and S. C. Lee. 2001. Complete nucleotide sequence and cell-line multiplication pattern of the attenuated variant CH2195LA of Japanese encephalitis virus. Virus Res. 73:91-102.
Zhang, W., P. R. Chipman, J. Corver, P. R. Johnson, Y. Zhang, S. Mukhopadhyay, T. S. Baker, J. H. Strauss, M. G. Rossmann, and R. J. Kuhn. 2003. Visualization of membrane protein domains by cryo-electron microscopy of dengue virus. Nat. Struct. Biol. 10:907-912.
Zhang, Y., J. Corver, P. R. Chipman, W. Zhang, S. V. Pletnev, D. Sedlak, T. S. Baker, J. H. Strauss, R. J. Kuhn, and M. G. Rossmann. 2003. Structures of immature flavivirus particles. EMBO J. 22:2604-2613.(Ying-Ju Lin and Suh-Chin )
ABSTRACT
The formation of the flavivirus prM-E complex is an important step for the biogenesis of immature virions, which is followed by a subsequent cleavage of prM to M protein through cellular protease to result in the production and release of mature virions. In this study, the intracellular formation of the prM-E complex of Japanese encephalitis virus was investigated by baculovirus coexpression of prM and E in trans in Sf9 insect cells as analyzed by anti-E antibody immunoprecipitation and sucrose gradient sedimentation analysis. A series of carboxyl-terminally truncated prM mutant baculoviruses was constructed to demonstrate that the truncations of the transmembrane (TM) region resulted in a reduction of the formation of the stable prM-E complex by approximately 40% for the TM1 (at residues 130 to 147 [prM130-147]) truncation and 20% for TM2 (at prM153-167) truncation. Alanine-scanning site-directed mutagenesis on the prM99-103 region indicated that the His99 residue was the critical prM binding element for stable prM-E heterodimeric complex formation. The single amino acid mutation at the His99 residue of prM abolishing the prM-E interaction was not due to reduced expression or different subcellular location of the mutant prM protein involved in prM-E interactions as characterized by pulse-chase labeling and confocal scanning microscopic analysis. Recombinant subviral particles were detected in the Sf9 cell culture supernatants by baculovirus coexpression of prM and E proteins but not with the prM H99A mutant. Sequence alignment analysis was further conducted with different groups of flaviviruses to show that the prM H99 residues are generally conserved. Our findings are the first report to characterize the minimum binding elements of the prM protein that are involved in prM-E interactions of flaviviruses. This information, concerning a molecular framework for the prM protein, is considered to elucidate the structure/function relationship of the prM-E complex synthesis and provide the proper trajectory for flavivirus assembly and maturation.
INTRODUCTION
Japanese encephalitis virus (JEV), of the genus Flavivirus of the family Flaviviridae, is an arthropod-borne virus that causes encephalitis in Asia with a high morbidity and mortality rate (4, 15). The genomic RNAs of all flaviviruses consist of two noncoding regions at the 5' and 3' ends and the coding sequence, which encodes the three structural proteins, i.e., the core (C), the precursor membrane (prM) or the membrane (M), and the envelope (E), and the seven nonstructural proteins, i.e., NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5. During flavivirus assembly, the prM and E proteins interact with each other to generate a prM-E heterodimeric complex in the endoplasmic reticulums (ERs) of the infected cells (16, 17, 18). In the intracellular synthesis of the immature virions, the viral genomic RNA is encapsulated with the C protein and the surface lipid bilayer, into which the prM-E heterodimeric complex is inserted. Next, the immature virions transported through the secretary pathway are converted into mature virions by the cleavage of the prM protein to the M protein by cellular furin or a furin-like protease (8). The prM protein, by interacting with the E protein, can also protect the immature virions against inactivation in the acidic vesicles and promote the appropriate transporting/folding of the stable complex of the E protein (8). In tick-borne encephalitis (TBE) virus, coexpression of prM and E proteins alone without viral genomic RNA and C protein resulted in formation of recombinant subviral particles (RSPs) (3, 17). The lateral interactions between prM and E proteins are a crucial step to stabilize subviral particle formation by flaviviruses in cultured cells (6).
Although the prM/M structure of flavivirus has yet to be determined, the X-ray structures of the E protein have been solved for TBE virus (26) and recently for dengue virus (20, 21). High-resolution images obtained from cryoelectron microscopy of dengue (DEN) and yellow fever (YF) viruses reveal that the prM-E heterodimers are laterally arranged on the surfaces of the immature virions (30, 31). The prM-E interacting complex has also been characterized using immunoprecipitation on the native virions of dengue (27), TBE (7) and yellow fever (24) viruses. Recently, the minimum functional elements of the E protein involved in the prM-E interaction of TBE virus were identified at the stem region close to transmembrane regions at the carboxyl-terminal end of E protein (3).
Our present work provides the first evidence of the location and characteristics of the prM binding elements involved in prM-E heterodimeric interaction of JEV, using a baculovirus expression system. The intracellular formation of the prM-E complexes of JEV was investigated in Sf9 cells by baculovirus coexpression of prM and E genetically in trans, immunoprecipitation by the JEV E protein-specific monoclonal antibody (MAb) E3.3, and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis. A series of carboxyl-terminally truncated prM mutant baculoviruses was constructed to map the minimum binding elements of prM involved in prM-E interactions. The formation of the stable prM-E heterodimeric complex and the release of RSPs were further characterized using sucrose gradient sedimentation, pulse-chase labeling, double-staining confocal microscopy, and transmission electron microscopy (TEM). Sequence alignment analysis of the prM protein was finally conducted with 11 groups of flaviviruses, including JEV and Kunjin, Murray Valley encephalitis, St. Louis encephalitis virus, West Nile, DEN type 1 to 4 (DEN1 to -4), TBE, and YF viruses. This is the first report to characterize the minimum binding elements of prM protein involved in prM-E interactions in flavivirus assembly and maturation.
MATERIALS AND METHODS
Cell lines and viruses. C6/36 cells were grown in Hanks minimal essential medium supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Invitrogen) at 28°C. Spodoptera frugiperda Sf9 cells were grown at 28°C in TNM-FH medium (Sigma) supplemented with 10% FBS. Hybridoma cells producing MAbs E3.3 and 5B1 were grown at 37°C in Iscove's modified Dulbecco's medium (Invitrogen) supplemented with 10% FBS. The JEV strain CH2195LA (Gene Bank accession no. AF221499) was produced in C6/36 cells. The recombinant baculoviruses for plaque selection and propagation in this study were in Sf9 cells.
Construction of recombinant baculoviruses. The genomic RNA of the JEV CH2195LA strain was extracted from the infected C6/36 cell culture supernatants by using the Trizol kit (Invitrogen), followed by the addition of Superscript II and Elongase (Invitrogen) for reverse transcriptase PCR to obtain the cDNAs of the prME-, prM-, and E-coding genes by using the following oligonucleotide primers: (i) for the prME gene, forward primer 5'-ATgcgggATccATgggAAATGAAGCCTCAATG-3' and reverse primer 5'-cgcgAATTCCTAAGCATGCACATTGGTCGC-3'; (ii) for the prM gene, forward primer 5'-ATGCGGGATCCATGGGAAATGAAGCCTCAATG-3' and reverse primer 5'-GGGAATTCCTAGAAACTGTAAGCCGGAGCGACC-3'; (iii) for the E gene, forward primer 5'-CGCGGATCCATGATGCTTGGCAGTAACAACGGTCAACGCG-3' and reverse primer 5'-ATGGGAATTCCTACATTTTTAGCCTGCATTTCAG-3'; (iv) for the carboxyl-terminally truncated prM genes, reverse primers for (a) prM147tr (5'-GGGAATTCCTAACTGCCAAGCATCCAGCC-3'), (b) prM130tr (5'-GCGAATTCCTACCTTATGATCCAGTTCTCAGT-3'), (c) prM126tr (5'-GGGAATTCCTAGTTCTCAGTTTTCATGAGATA-3'), (d) prM108tr (5'-GGGAATTCCTATTTTTTATTCACTAGTGAACT-3'), (e) prM104tr (5'-GGGAATTCCTATAGTGAACTCTCCCCATGTGT-3'), (f) prM99tr (5'-GGGAATTCCTAATGTGTTTGGACCGATACGGA-3'), (g) prM96tr (5'-GGGAATTCCTAGACCGATACGGATCTCCTGCT-3'), and (h) prM92tr (5'-GCGAATTCCTATCTCCTGCTTCGCTTGGAATG-3'); and (v) for the point mutation prM genes, reverse primers for (a) Bac-prM-H99A (5'-GTCCAAACAGCTGGGGAGAGT-3' and 5'-ACTCTCCCCAGCTGTTTGGAC-3'), (b) Bac-prM-G100A (5'-CAAACACATGCCGAGAGT-3' and 5'-ACTCTCGGCATGTGTTTG-3'), (c) Bac-prM-E101A (5'-CATGGGGCCAGTTCACTA-3' and 5'-TAGTGAACTGGCCCCATG-3'), (d) Bac-prM-S102A (5'-CATGGGGAGGCCTCACTA-3' and 5'-TAGTGAGGCCTCCCCATG-3'), and (e) Bac-prM-S103A (5'-GGGGAGAGTGCCCTAGTG-3' and 5'-CACTAGGGCACTCTCCCC-3'). All the forward primers contained a BamHI restriction site, and the reverse primers contained an EcoRI restriction site for ligation to the pBlueBac4 baculovirus vector (Invitrogen). All of the sequences were confirmed by DNA sequence analysis (Mission Biotech Inc., Taipei, Taiwan). To generate the recombinant baculoviruses, the constructed pBlueBac4 plasmids were cotransfected with the linearized baculovirus Bac-N-Blue DNA (Invitrogen) into Sf9 cells treated with Cellfectin (Invitrogen) and incubated for 5 to 7 days. Recombinant baculoviruses were then picked from the blue-staining plaques formed in the infected Sf9 cells overlaid with agarose medium containing 5-bromo-4-chloro-3-indoyl-?-D-galactoside (X-Gal). Three plaque purifications were conducted to obtain the recombinant baculoviruses used for the studies.
Metabolic labeling and immunoprecipitation. Subconfluent monolayers of Sf9 cells in six-well plates were infected with recombinant baculoviruses or JEV at a multiplication of infection (MOI) of 5. At 24 h postinfection, the TNM-FH medium was removed and replaced with TNM-FH medium containing 0.01% normal methionine (Met) and 0.02 to 0.10 mCi/ml of [35S]methionine-cysteine. Baculovirus-infected Sf9 cells were treated with a lysis buffer (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 2 mM EDTA, 1% Triton X-100, and 1 mM phenylmethylsulfonyl fluoride) and precleared using protein A-Sepharose (Pharmacia). The purified cell lysates were then reacted with MAb E3.3 for 5 h at 4°C, followed by incubation with a Sepharose slurry for 16 h at 4°C. The immunoprecipitated samples were further separated by centrifugation at 200 x g for 3 min, and the pellet was washed twice with the incubation buffer (63 mM Tris-HCl [pH 6.8]). The precipitated samples were solubilized by heating (95°C for 5 min) with 40 μl of electrophoresis reducing sample buffer (50 mM Tris-HCl [pH 6.8], 2% SDS, 0.1% bromophenol blue, 5% 2-mercatoethanol, and 10% glycerol) and further analyzed by SDS-PAGE (12%) and fluorography.
Cross-linking analysis for intracellular formation of prM-E heterodimer. Sf9 cells plated in 60-mm dishes were infected with recombinant baculoviruses or JEV at an MOI of 5 for 24 h. The maintenance medium was removed, and the cells were labeled with [35S]methionine-cysteine as described above. The virus-infected cells were rinsed with phosphate-buffered saline and scraped into Microfuge tubes. The infected cells were then cross-linked by the addition of dithio-bis(succinimidylpropionate) (DSP) (Pierce) to a final concentration of 1 mM and further incubated for 1 h at 4°C. The reaction was then stopped by adding glycine (to a final concentration of 50 mM), and the mixture was further incubated for 15 min at 4°C. The precipitated samples were analyzed with both reducing sample buffer (50 mM Tris-HCl [pH 6.8], 2% SDS, 5% 2-mercaptoethanol, 0.1% bromophenol blue, and 10% glycerol) and nonreducing sample buffer (50 mM Tris-HCl [pH 6.8], 2% SDS, 0.1% bromophenol blue, and 10% glycerol) and analyzed by 7.5% SDS-PAGE and fluorography.
Sucrose gradient sedimentation analysis for stable prM-E complex. The cell lysates were solubilized in 1% Triton X-100 and then loaded for centrifugation in a 3 to 60% (wt/wt) sucrose gradient made with gradient buffer (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 2 mM EDTA, 0.5% Triton X-100, 1 mM phenylmethylsulfonyl fluoride). The gradients were centrifuged in a Hitachi RPS40ST rotor at 38,000 rpm and 15°C for 22 h, and each fraction was collected and immunoprecipitated with monoclonal antibody E3.3 (anti-JEV E protein) as described above. The immune complexes were analyzed by reducing SDS-PAGE and fluorography.
Pulse-chase analysis for stable prM-E complex. The Sf9 cells infected with recombinant baculoviruses were cultured on six-well subconfluent plates (MOI = 5) in TNM-FH medium for 24 h. The maintenance medium was removed, and then the cells were starved for 30 min in methionine-free medium. The cells were pulse-labeled with [35S]methionine-cysteine (0.25 mCi/ml) for about 30 min and incubated with TNM-FH medium for different chase times (0, 1, 2, 4, and 8 h). The immunoprecipitated cell lysed were resolved by 12% SDS-PAGE and further analyzed by fluorography.
Subcellular localization using confocal laser scanning microscopic analysis. Infected Sf9 cells grown on 15-mm glass coverslips at 24 h postinfection were washed three times with TBS (140 mM NaCl, 25 mM Tris-HCl, pH 6.1), incubated in medium containing the ER-staining dye DiIC13(3) (10 nM; Molecular Probes) for 30 min, and finally fixed with 3.7% paraformaldehyde for 1 h. The cells were then permeabilized with TBS containing 0.1% Triton X-100 for 1 h. Expression of the JEV proteins in these cells was revealed by staining with the anti-E MAb E3.3 or anti-prM MAb 5B1 followed by the addition of goat anti-mouse Alexa Fluor 647 immunoglobulin G (Molecular Probes). The subcellular localizations in these cells were examined in a confocal laser scanning microscope (Zeiss LSM 500) with a 100x 1.3 oil objective.
TEM for RSP visualization. Culture supernatants of baculovirus-infected Sf9 cells (MOI = 2.5) or JEV-infected Sf9 cells at 5 days postinfection were centrifuged, precipitated with 10% polyethylene glycol 8000, and resuspended in 1 ml TN buffer (10 mM Tris-HCl [pH 7.5], 100 mM NaCl). A continuous sucrose gradient (3 to 60%, wt/wt) was applied, and centrifugation was carried out at 4°C and 38,000 rpm for 24 h (RPS40ST rotor and Hitachi 85P-72 ultracentrifuge). The virus particles were collected from 13 tubes of 0.85 ml for each fraction for TEM visualization. A single-droplet negative staining procedure was used in parallel for TEM visualization. Droplets of 30-μl samples were absorbed for 10 min onto Formvar carbon-coated cooper grids (200 mesh; Agar Scientific), washed with distilled water, and then stained with 2% uranyl acetate for 1 min. After air drying, the grids were examined under a Hitachi H-7500 TEM at 100 kV.
RESULTS AND DISCUSSION
Detection of prM-E interaction in baculovirus-infected Sf9 cells. The prME-, prM-, and E-encoding cDNAs of the JEV CH2195LA strain (14, 28, 29) were cloned into the baculoviruses (Bac-prME, Bac-E, and Bac-prM) with a polyhedron promoter (PPH) control (Fig. 1A) to study the prM-E interaction in Sf9 insect cells. Sf9 insect cells were initially infected with the recombinant baculovirus, radiolabeled with [35S]methionine-cysteine medium for 24 h, and then immunoprecipitated using the JEV E-protein specific MAb E3.3. The intracellular formation of prM-E complex in Sf9 insect cells was characterized using (i) a single recombinant baculovirus that carries the precursor prME gene (Bac-prME) and (ii) two separate recombinant baculoviruses that carry the prM and E genes (Bac-prM and Bac-E). The expression cassettes for the precursor prME in cis or the separate prM and E in trans all include the SP1 and SP2 signal peptide sequences derived from the carboxyl-terminal ends of the C and prM proteins (Fig. 1A), as the signal peptide sequences are responsible for stable insertion into the ER membrane to properly modify the prM-E complex formation (1, 2, 19, 24).
Formation of prM-E heterodimers was found in baculovirus expression of the prME precursor protein in cis and the prM and E proteins in trans, revealing two reactive bands, corresponding to prM (21 kDa) and E (56 kDa) in SDS-polyacrylamide gels, detected in Sf9 cells after immunoprecipitation (Fig. 1B). Only the E protein (and not the prM protein) was detected in Sf9 cells infected with Bac-E (Fig. 1B). Simultaneously, Sf9 cells and C6/36 cells (the mosquito cell line commonly used for JEV isolation) were infected with the JEV 2195LA strain at an MOI of 1, and the cell lysates were also found to form the prM-E heterodimers after immunoprecipitation (Fig. 1B). The use of a baculovirus expression system to analyze the intracellular formation of the prM-E complex in this study is appropriate, since Sf9 cells were susceptible to JEV infection (Fig. 1C). This result is in agreements with two recent reports that Sf9 cells are suitable for persistent JEV infection (10, 11). Although our findings in Fig. 1C showed that the replication kinetics of JEV in Sf9 cells was somewhat slower than that in C6/36 cells (i.e., the maximum titer decreased by approximately 30-fold), Sf9 cells can still be suitable to dissect the molecular interactions of prM and E proteins during JEV assembly and maturation.
To provide further evidence of the formation of noncovalent prM-E heterodimers, DSP was used as a cross-linking agent to preserve protein-protein interactions in lysis buffer (containing 1% Triton X-100). DSP is a thiol-cleavable molecule with two amine-reactive groups that is used to preserve the physical interaction between two proteins and has been successfully used for characterizing the coronavirus E and M proteins (5). Our results revealed that DSP-treated Sf9 cells either infected with Bac-prME or coinfected with Bac-prM and Bac-E yielded the formation of the noncovalent prM-E heterodimer (70 to 80 kDa) under nonreducing SDS-PAGE analysis (Fig. 1D). The noncovalent prM-E heterodimers were also present in the DSP-treated samples of Sf9 cells and C6/36 cells infected with the JEV CH2195LA strain (Fig. 1D). The formation of the prM-E heterodimers was effective and stable in the infected cells as demonstrated by sucrose gradient sedimentation and pulse-chase labeling analyses.
Mapping of the prM binding elements involved in prM-E heterodimeric interaction in Sf9 cells. The flavivirus prM protein has two transmembrane regions (TM1 and TM2) close to its C-terminal end, as reported for dengue virus (19) and yellow fever virus (24), so the TM1 and TM2 regions were predicted to be at prM residues 130 to 147 and 153 to 167 of the JEV CH2195LA strain (http://www.cbs.dtu.dk/services/TMHMM). A series of carboxyl-terminally truncated prM baculoviruses was constructed, including Bac-prM147tr (with only the TM2 truncation), Bac-prM130tr (with only the TM1 and TM2 truncations), and Bac-prM126tr, Bac-prM108tr, Bac-prM104, Bac-prM99, Bac-prM96, and Bac-prM92tr for fine mapping of the prM binding elements involved in prM-E heterodimeric interactions (Fig. 2A). Following immunoprecipitation by MAb E3.3, the intensity of the E protein-reactive bands was almost constant, but the intensity of the prM-reactive bands decreased or even became undetectable (Fig. 2B). The reduced binding intensity of TM-truncated prMs (78% for Bac-prM147tr and 43% for Bac-prM130tr) exceeded those governed by further carboxyl-terminal truncations (36% for Bac-prM126tr, 26% for Bac-prM108tr, and 24% for Bac-prM104tr) for the formation of prM-E complex (Fig. 2C). Moreover, the prM protein of the prM-E complex was not detected in Sf9 cells coinfected with Bac-prM99tr, Bac-prM96tr, and Bac-prM92tr (Fig. 2B). These results indicated that the TM regions and prM residues 99 to 103 were also important for the stable formation of the prM-E complex in the coinfected Sf9 cells.
Sucrose gradient sedimentation analysis of prM TM region affecting stable prM-E complex formation. To further verify whether the TM regions considerably influence the formation of the stable prM-E complex in Sf9 cells, cell lysates before immunoprecipitation were subjected to sucrose gradient sedimentation. These experiments showed that the Sf9 cells coinfected with Bac-prM formed a stable prM-E heterodimeric complex, mainly at fractions 5 and 6 (Fig. 3A), while the prM-E heterodimeric complex was detected at the same fractions 5 and 6 for Bac-prM147tr and Bac-prM130tr but the intensity of the prM binding protein was lower (Fig. 3B and C). The percentage of dimerization of prM was taken as the prM/E ratio, measured for the stable prM-E complex in each sucrose gradient fractionation, to compare the reduction in intensity of the prM protein from that of the most constant E proteins detected in the sucrose-fractionated samples. The results indicated that most prM-E heterodimers were in fractions 5 to 7 (Fig. 3D). The percentage of prM-E heterodimer formation, which is indicated as the prM/E ratio in Fig. 3D, declined from 100% for Bac-prM (fraction 6) to 80% for Bac-prM147tr (fraction 5) and to 40% for Bac-prM130tr (fraction 5). The truncations of both TM1 and TM2 (Bac-prM130tr) resulted in a reduction of the formation of the heterodimer by approximately 40%, compared to TM2 truncation with a 20% loss. This result suggests that TM1 is more important than TM2 in assisting the formation of the stable prM-E complex in the infected cells.
We also used homology modeling to predict the prM-E complex structure of the JEV CH2195LA strain based on the cryoelectron microscopic results published for the dengue virus prM and E membrane proteins (PDB 1P58) (30). The predicted prM structure contains only part of prM (prM residues 113 to 167 [prM113-167] of CH2195LA) according to the published cryoelectron microscopic structure (30). Our modeling structure indicates that TM1 (prM130-147) is topologically closer than TM2 (prM153-167) to the helix 2 region of TBE virus E protein that was previously reported to be the minimum E binding elements for the prM protein (3). However, we cannot rule out the possibility of the N-terminal portion of the M protein (i.e., prM93-112) being involved in prM-E interaction, as it has been recently speculated to cooperate with the hole between the two monomers to form the homodimer of the E ectodomain (30).
Site-directed mutagenesis studies on prM99-103, which is involved in prM-E interaction. Alanine-scanning site-directed mutagenesis of the prM99-103 region was performed by constructing five more prM mutant baculoviruses (Bac-prM-H99A, Bac-prM-G100A, Bac-prM-E101A, Bac-prM-S102A, and Bac-prM-S103A) to map the amino acid residue(s) that was most critically involved in prM-E interaction (Fig. 4A). The immunoprecipitation results indicated that the prM-reactive band was absent from the Sf9 cells coinfected with Bac-prM-H99A; however, both prM and E protein bands were detectable in the Sf9 cells coinfected with Bac-prM, Bac-prM-G100A, Bac-prM-E101A, Bac-prM-S102A, and prM-S103A (Fig. 4B). Additionally, sucrose gradient sedimentation analysis of the infected cell lysates showed that Sf9 cells coinfected with Bac-prM-H99A had fewer prM-reactive bands in fractions 5 and 6 (Fig. 5B) than those coinfected with Bac-prM, Bac-prM-G100A, Bac-prM-E101A, Bac-prM-S102A, and Bac-prM-S103A (Fig. 5A, C, D, E, and F). Therefore, the His99 residue was demonstrated to be the critical element for the stable prM-E heterodimeric complex formation.
Pulse-chase kinetics, subcellular localization, and RSP formation of the prM-E stable complex affected by the prM His99 mutation. Pulse-chase experiments were performed in a time course from 0 to 8 h of chase to investigate the kinetics of the newly synthesized prM-E complex affected by the single amino acid mutation at prM His99. The newly synthesized prM-E complex was evident at 0 h of chase and slowly decreased in amount to 8 h of chase in the Sf9 cells coinfected with Bac-prM (Fig. 6A). The results agree with those previously obtained from Vero cells infected with dengue virus, which showed that the prM-E complex formed persisted for about 4 to 6 h (27). However, no prM-reactive protein was detectable in the cells coinfected by the prM mutation (Bac-prM-H99A) from 1 to 8 h of chase compared to the newly synthesized E protein (Fig. 6B). To further confirm the expression of the mutant prM in the infected cells, a pig anti-JEV serum was used to immunoprecipitate the prM and E proteins. The pulse-chase results showed that Sf9 cells coinfected with Bac-prM-H99A showed both prM and E proteins, indicating that prM and E are both expressed and that the mutated form of prM is not degraded as shown from 0 to 8 h of chase (Fig. 6C).
The subcellular localization of the prM-E complex in Sf9 cells was further investigated using confocal laser scanning microscopy. The Sf9 cells coinfected with Bac-prM-H99A, along with the controls of Bac-prME and JEV infection, were doubly stained with a prM-specific MAb and DiIC16(3), an ER-specific dye. The double-staining images obtained using confocal laser scanning microscopy showed that the intracellular distributions of prM and the mutant prM (prM H99A) almost overlapped with the ER marker (Fig. 7), suggesting that the loss of prM-E interaction in the prM with the mutation at the His99 residue did not occur because of the difference in their subcellular locations for prM-E interaction in Sf9 cells.
Furthermore, as the formation and release of RSPs by coexpression of prM and E proteins in mammalian cells was extensively reported (9, 12, 13, 22), we evaluated the RSP formation in baculovirus-infected Sf9 cells. Culture supernatants were collected from Sf9 cells infected with the JEV 2195LA strain, coinfected with Bac-prM (BacprM-H99A) and Bac-E. Samples were then collected from the fractions of sucrose gradient sedimentation for TEM visualization. The TEM results after negative staining with uranyl acetate showed that the culture supernatants of JEV-infected Sf9 cells yielded formation of spherical particles as the infectious virions (i.e., 50 nm in diameter) in the sucrose gradient-fractionated samples (Fig. 8A). Baculovirus coexpression of prM and E in trans in Sf9 cells led to the formation and release of smaller RSPs with a diameter of about 30 to 50 nm (Fig. 8B). Coinfection of Sf9 cells with BacPrM-H99A and Bac-E did not result in the formation and release of RSPs in culture supernatants (Fig. 8C). Although all of these experiments demonstrated that the replacement of prM H99 by an Ala strongly reduced prM-E heterodimerization and abolished RSP secretion, we still cannot exclude the possibility that the His-to-Ala mutation may alter prM folding and that misfolded prM would not be able to interact with E because the region of interaction on prM is not properly exposed on the protein.
Sequence alignment analysis of prM99 and TM regions in flaviviruses. To further demonstrate whether the prM99 and two TM regions are important for a general interaction of prM and E proteins for flaviviruses, we analyzed the amino acid sequences of the flavivirus M protein sequences (prM93-167), including four groups of the JEV serocomplex viruses (JEV, Kunjin virus, Murray Valley encephalitis virus, St. Louis encephalitis virus, and West Nile virus) and six groups of other flaviviruses (DEN1, DEN2, DEN3, DEN4, TBE, and YF viruses) (Fig. 9). Multiple amino acid sequence alignment was done by using the Showalign program of the Jemboss software through the Jemboss server (http://bioinfo.life.nthu.edu.tw/bioinfo.htm). The results indicate that the prM H99 residues are all conserved in all the flavivirus groups (Fig. 9). In contrast, the amino acid residues at prM100-103 vary with each individual group: GE/DSS/T for the JEV serocomplex, VGL/MG for the DEN virus serocomplex, AQGE/D for TBE virus, and ENHG for YF virus. Other conserved amino residues were also identified in all groups of flaviviruses, including prM L104, prM W111, prM E125, and prM W127 (Fig. 9). However, the reduced binding intensities of prM to the prM-E complex by the TM-truncated mutants (Bac-prM104, Bac-prM108, Bac-prM126, and Bac-prM130) in Sf9 cells were less significant (Fig. 2B and C), indicating that the conserved residues at prM L104, prM W111, prM E125, and pM W127 in flaviviruses are unlikely to be involved in prM-E interaction in virus assembly. Therefore, the conserved residue at prM H99 may contribute solely to the structural integrity of the prM-E heterodimeric complex interaction in flavivirus assembly. However, the conservation of prM H99 among all the flaviviruses analyzed may not be necessary, in favor of a direct role for this residue being involved in protein-protein interaction, and that may come from some coevolution of the sequences involved in interactions within the prM and E proteins. It is also likely that a mutation in one of the envelope proteins might compensate for other mutations in the other envelope protein. Our present finding of the prM H99 residue being involved in JEV prM-E heterodimeric interactions does not agree with a recent study showing the histidine at M39 (equivalent to prM-131 in Fig. 9) as a critical residue for prM/E heterodimers of DEN2 virus assembly (25).
The amino acid sequences at the C terminus of the prM protein consist of two TM regions as reported for flaviviruses (19, 24), i.e., TM1 (prM130-147), TM2 (prM153-167), and the nearby connecting segment (four hydrophilic residues and one charged residue). No conserved residues were observed among TM1 and TM2 for all groups of flaviviruses; however, they contain at least 9 to 12 hydrophobic residues to form three or four alpha-helix turns that stably insert in the lipid bilayer. Based on the homology modeling and the cryoelectron microscopic results published for the dengue virus prM and E membrane proteins (PDB 1P58) (30), there are four alpha-helix turns in TM1 (prM132-135, prM136-139, prM140-143, and prM144-147) and three alpha-helix turns in TM2 (prM153-156, prM157-160, and prM161-164) predicted for the JEV CH2195LA strain. The TM1 region of hepatitis C virus envelope glycoprotein E1 has been demonstrated to contain four helix turns by using nuclear magnetic resonance analysis, where the GXXXG motif located in the first helix turn was shown to greatly affect the E1E2 heterodimerization (23). The presence of the GXXXG motif was found in the JEV serocomplex viruses at prM142-146 of TM1 (Fig. 9), and our preliminary results also showed a significant reduction of prM-E heterodimeric complex formation as demonstrated by alanine insertion on the GXXXG motif (data not shown). Although we were able to experimentally delineate that TM1 is more important than TM2 in assisting the prM-E stable complex based on prM C-terminal truncations (Fig. 2A to C), further studies are still needed to dissect the detailed molecular elements in TM and connecting segment regions involved in assisting the prM-E heterodimeric complex formation in flavivirus assembly.
In conclusion, the present study provides evidence that the JEV structural proteins prM and E interact with each other to form a stable heterodimeric complex in Sf9 cells. His99 and the TM regions of the prM protein were identified to be the most important regions involved in the interaction with E protein in JEV assembly and maturation. This information, concerning a molecular framework for the prM protein, is considered to elucidate the structure/function relationship of the prM-E complex synthesis for the immature virions in flavivirus assembly.
ACKNOWLEDGMENTS
This work was supported by the National Science Council grants NSC93-2214-E-007-015 and NSC93-3112-B-007-015 and the National Health Research Institutes, Taiwan, Republic of China.
REFERENCES
Allison, S. L., J. Schalich, K. Stiasny, C. W. Mandl, C. Kunz, and F. X. Heinz. 1995. Oligomeric rearrangement of tick-borne encephalitis virus envelope proteins induced by an acidic pH. J. Virol. 69:695-700.
Allison, S. L., K. Stadler, C. W. Mandl, C. Kunz, and F. X. Heinz. 1995. Synthesis and secretion of recombinant tick-borne encephalitis virus protein E in soluble and particulate form. J. Virol. 69:5816-5820.
Allison, S. L., K. Stiasny, K. Stadler, C. W. Mandl, and F. X. Heinz. 1999. Mapping of functional elements in the stem-anchor region of tick-borne encephalitis virus envelope protein E. J. Virol. 73:5605-5612.
Burke, D. S., and T. P. Monath. 2001. Flaviviruses, p. 1043-1125. In D. M. Knipe and P. Howley (ed.), Fields virology. Lippincott-Raven, Philadelphia, Pa.
Corse, E., and C. E. Machamer. 2003. The cytoplasmic tails of infectious bronchitis virus E and M proteins mediate their interaction. Virology 312:25-34.
Ferlenghi, I., M. Clarke, T. Ruttan, S. L. Allison, J. Schalich, F. X. Heinz, S. C. Harrison, F. A. Rey, and S. D. Fuller. 2001. Molecular organization of a recombinant subviral particle from tick-borne encephalitis virus. Mol. Cell 7:593-602.
Heinz, F. X., K. Stiasny, A. Puschner-Auer, H. Holzmann, S. L. Allison, C. W. Mandl, and C. Kunz. 1994. Structural changes and functional control of the tick-borne encephalitis virus glycoprotein E by the heterodimeric association with protein prM. Virology 198:109-117.
Heinz, F. X., and S. L. Allison. 2003. Structures and mechanisms in flavivirus fusion. Adv Virus Res. 55:231-269.
Hunt, A. R., C. B. Cropp, and G. J. Chang. 2001. A recombinant particulate antigen of Japanese encephalitis virus produced in stably-transformed cells is an effective noninfectious antigen and subunit immunogen. J. Virol. Methods 97:133-149.
Jeong, H. S., J. H. Shin, Y. N. Park, J. Y. Choi, Y. L. Kim, B. G. Kim, S. R. Ryu, S. Y. Baek, S. H. Lee, and S. N. Park. 2003. Development of real-time RT-PCR for evaluation of JEV clearance during purification of HPV type 16 L1 virus-like particles. Biologicals 31:223-229.
Kim, H., S. J. Lee, J. Y. Park, Y. W. Park, H. S. Kim, H. Y. Kang, B. K. Hur, Y. W. Ryu, S. I. Han, and J. S. Kim. 2004. Study on persistent infection of Japanese encephalitis virus Beijing-1 strain in serum-free Sf9 cell cultures. J. Microbiol. 42:25-31.
Kojima, A., A. Yasuda, H. Asanuma, T. Ishikawa, A. Takamizawa, K. Yasui, and T. Kurata. 2003. Stable high-producer cell clone expressing virus-like particles of the Japanese encephalitis virus e protein for a second-generation subunit vaccine. J. Virol. 77:8745-8755.
Konishi, E., A. Fujii, and P. W. Mason. 2001. Generation and characterization of a mammalian cell line continuously expressing Japanese encephalitis virus subviral particles. J. Virol. 75:2204-2212.
Lin, C. W., and S. C. Wu. 2003. A functional epitope determinant on domain III of the Japanese encephalitis virus envelope protein interacted with the neutralizing antibody combining sites. J. Virol. 77:2600-2606.
Lindenbach, B. D., and C. M. Rice. 2001. Flaviviridae: the viruses and their replication, p. 991-1041. In D. M. Knipe and P. Howley (ed.), Fields virology. Lippincott-Raven, Philadelphia, Pa.
Lorenz, I. C., S. L. Allison, F. X. Heinz, and A. Helenius. 2002. Folding and dimerization of tick-borne encephalitis virus envelope proteins prM and E in the endoplasmic reticulum. J. Virol. 76:5480-5491.
Lorenz, I. C., J. Kartenbeck, A. Mezzacasa, S. L. Allison, F. X. Heinz, and A. Helenius. 2003. Intracellular assembly and secretion of recombinant subviral particles from tick-borne encephalitis virus. J. Virol. 77:4370-4382.
Mackenzie, J. M., and E. G. Westaway. 2001. Assembly and maturation of the flavivirus Kunjin virus appear to occur in the rough endoplasmic reticulum and along the secretory pathway, respectively. J. Virol. 75:10787-10799.
Markoff, L., A. Chang, and B. Falgout. 1994. Processing of flavivirus structural glycoproteins: stable membrane insertion of premembrane requires the envelope signal peptide. Virology 204:526-540.
Modis, Y., S. Ogata, D. Clements, and S. C. Harrison. 2003. A ligand-binding pocket in the dengue virus envelope glycoprotein. Proc. Natl. Acad. Sci. USA 100:6986-6991.
Modis, Y., S. Ogata, D. Clements, and S. C. Harrison. 2004. Structure of the dengue virus envelope protein after membrane fusion. Nature 427:313-319.
Mutoh, E., T. Ishikawa, A. Takamizawa, T. Kurata, T. Sata, and A. Kojima. 2004. Japanese encephalitis subunit vaccine composed of virus-like envelope antigen particles purified from serum-free medium of a high-producer J12#26 cell clone. Vaccine 22:2599-2608.
Op De Beeck, A., R. Montserret, S. Duvet, S. Cocquerel, R. Cacan, B. Barberot Le Maire, F. Penin, and J. Dubuisson. 2000. The transmembrane domains of hepatitis C virus envelope glycoproteins E1 and E2 play a major role in heterodimerization. J. Biol. Chem. 275:31428-31437.
Op De Beeck, A., R. Molenkamp, M. Caron, A. Ben Younes, P. Bredenbeek, and J. Dubuisson. 2003. Role of the transmembrane domains of prM and E proteins in the formation of yellow fever virus envelope. J. Virol. 77:813-820.
Pryor, M. J., L. Azzola, P. J. Wright, and A. D. Davidson. 2004. Histidine 39 in the dengue virus type 2 M protein has an important role in virus assembly. J. Gen. Virol. 85:3627-3636.
Rey, F. A., F. X. Heinz, C. Mandl, C. Kunz, and S. C. Harrison. 1995. The envelope glycoprotein from tick-borne encephalitis virus at 2 A resolution. Nature 375:291-298.
Wang, S., R. He, and R. Anderson. 1999. PrM- and cell-binding domains of the dengue virus E protein. J. Virol. 73:2547-2551.
Wu, S. C., W. C. Lian, L. C. Hsu, and M. Y. Liau. 1997. Japanese encephalitis virus antigenic variants with characteristic differences in neutralization resistance and mouse virulence. Virus Res. 51:173-181.
Wu, S. C., and S. C. Lee. 2001. Complete nucleotide sequence and cell-line multiplication pattern of the attenuated variant CH2195LA of Japanese encephalitis virus. Virus Res. 73:91-102.
Zhang, W., P. R. Chipman, J. Corver, P. R. Johnson, Y. Zhang, S. Mukhopadhyay, T. S. Baker, J. H. Strauss, M. G. Rossmann, and R. J. Kuhn. 2003. Visualization of membrane protein domains by cryo-electron microscopy of dengue virus. Nat. Struct. Biol. 10:907-912.
Zhang, Y., J. Corver, P. R. Chipman, W. Zhang, S. V. Pletnev, D. Sedlak, T. S. Baker, J. H. Strauss, R. J. Kuhn, and M. G. Rossmann. 2003. Structures of immature flavivirus particles. EMBO J. 22:2604-2613.(Ying-Ju Lin and Suh-Chin )