Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Orlinger, K. K. Articles by Mandl, C. W. Search for Related Content PubMed PubMed Citation Articles by Orlinger, K. K. Articles by Mandl, C. W.

 Previous Article  |  Next Article 

Journal of Virology, December 2006, p. 12197-12208, Vol. 80, No. 24
0022-538X/06/$08.00+0     doi:10.1128/JVI.01540-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Construction and Mutagenesis of an Artificial Bicistronic Tick-Borne Encephalitis Virus Genome Reveals an Essential Function of the Second Transmembrane Region of Protein E in Flavivirus Assembly

Klaus K. Orlinger, Verena M. Hoenninger, Regina M. Kofler, and Christian W. Mandl^*

Clinical Institute of Virology, Medical University of Vienna, Vienna, Austria

Received 19 July 2006/ Accepted 23 September 2006

Top Abstract Introduction Materials and Methods Results Discussion References

 
Flaviviruses have a monopartite positive-stranded RNA genome, which serves as the sole mRNA for protein translation. Cap-dependent translation produces a polyprotein precursor that is co- and posttranslationally processed by proteases to yield the final protein products. In this study, using tick-borne encephalitis virus (TBEV), we constructed an artificial bicistronic flavivirus genome (TBEV-bc) in which the capsid protein and the nonstructural proteins were still encoded in the cap cistron but the coding region for the surface proteins prM and E was moved to a separate translation unit under the control of an internal ribosome entry site element inserted into the 3' noncoding region. Mutant TBEV-bc was shown to produce particles that packaged the bicistronic RNA genome and were infectious for BHK-21 cells and mice. Compared to wild-type controls, however, TBEV-bc was less efficient in both RNA replication and infectious particle formation. We took advantage of the separate expression of the E protein in this system to investigate the role in viral assembly of the second transmembrane region of protein E (E-TM2), a second copy of which was retained in the cap cistron to fulfill its other role as an internal signal sequence in the polyprotein. Deletion analysis and replacement of the entire TBEV E-TM2 region with its counterpart from another flavivirus revealed that this element, apart from its role as a signal sequence, is important for virion formation.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Flaviviruses, i.e., members of the genus Flavivirus, family Flaviviridae, comprise a number of medically important, mostly arthropod-transmitted pathogens, including tick-borne encephalitis virus (TBEV), yellow fever virus (YFV), West Nile virus, the dengue viruses, and Japanese encephalitis virus (28). Flavivirus virions are small enveloped particles that are assembled from only three structural proteins. The capsid (or core) protein C, together with the unsegmented positive-stranded RNA genome, forms the nucleocapsid, which is surrounded by a host-derived membrane into which the two surface proteins M (derived from a larger precursor, prM, by a late proteolytic cleavage event) (39) and E are anchored by carboxy-terminal anchor regions that span the membrane twice in an antiparallel orientation (34, 45, 46).

The transmembrane regions of both prM and E are composed of a tight hairpin structure with two interacting antiparallel transmembrane alpha helical segments called TM1 and TM2. These hairpin structures have been observed by cryoelectron microscopy (45), but their contact sites have not been identified precisely. One puzzling feature of flavivirus envelope proteins is that they do not appear to protrude through the viral membrane into the interior of the virion and therefore do not make substantial contact with the nucleocapsid (45, 46). This implies that the coordination of flavivirus assembly must depend on mechanisms other than specific interactions between the capsid and the envelope proteins. Flaviviruses are formed intracellularly, apparently by budding into the lumen of the endoplasmic reticulum (ER), although budding intermediates are rarely observed (28, 34). The membrane-anchored proteins prM and E are synthesized as heterodimers that are oriented into the lumen of the ER membrane and associate into higher-order structures that eventually form a curved surface lattice with icosahedral symmetry (26, 39, 47). These interactions in the viral envelope appear to be the guiding force for virion budding, but it is still unclear how this is coordinated with engulfment of the nucleocapsid. In fact, budding frequently occurs without the participation of the nucleocapsid, resulting in the formation and release of highly structured but capsidless subviral particles (1, 10, 17, 22-24).

The flavivirus genomic RNA contains a 5' cap structure and serves as the only viral mRNA for protein synthesis (28). The genome encodes all of the viral proteins in a single long open reading frame, whose primary translation product is a polyprotein that requires co- and posttranslational cleavage by a combination of viral and cellular proteases to form the individual structural and nonstructural proteins. A series of internal signal sequences within the polyprotein establishes its correct topology in the ER membrane, where replication and assembly occur (28).

The strategy of using a polyprotein precursor is common among positive-strand RNA viruses, allowing them to optimize their genomes to encode a maximum of genetic information within a minimal genome size. This is necessary because viral RNA-dependent RNA polymerases are error prone and would otherwise introduce too many mutations into each copy of the genome for the virus to thrive (8, 37). The polyprotein strategy also allows multiple usages of particular sequence elements in various contexts and provides for temporal and spatial control of assembly through concerted cleavage events, which sometimes involve functionally important intermediate cleavage products. In the case of flaviviruses, the coordinated processing of the junction between the C- and prM-coding regions by the viral protease and host signalase is believed to be important for retaining the envelope proteins and capsid components at the site of virion assembly since, as mentioned above, flavivirus budding does not appear to involve specific interactions between these elements (27, 29, 40).

While the economy of the polyprotein strategy provides certain advantages for the virus, it can, at the same time, make it more difficult for the researcher interested in studying the role of individual proteins and functional elements in isolation. For example, although previous studies examining assembly of capsidless subviral particles have suggested that the TM2 region of the flavivirus protein E might be dispensable for virion assembly (2, 35), this has not been tested directly in an infectious system because this same region serves simultaneously as an internal signal sequence in the polyprotein, which is required for translocation of the nonstructural protein NS1 and establishing the correct orientation of the remaining nonstructural proteins relative to the ER membrane. Disrupting it would be expected to disrupt polyprotein processing in general. We were therefore interested in finding out whether it is possible to construct a viable artificial flavivirus genome in which the surface proteins (pr)M and E are encoded in a separate cistron and therefore separate from the normal polyprotein precursor. The feasibility of this approach had already been suggested by an earlier study in which single-round infectious particles could be produced using a replicon from which the region encoding prM and E of TBEV had been deleted and was instead provided in trans by a packaging cell line (11). Furthermore, in other studies it was shown that TBEV can be used as a bicistronic (bc) expression vector by inserting an internal ribosome entry site (IRES) from encephalomyocarditis virus (EMCV) and an enhanced green fluorescent protein (EGFP) reporter gene into a nonessential variable region in the 3' noncoding region (3'-NCR) of the genome (11, 14).

Now, we demonstrate that an artificial bicistronic TBEV genome, in which the portion of the viral open reading frame encoding the prM and E proteins had been removed from its normal context and placed under the control of a heterologous IRES element inserted into the 3'-NCR, can be successfully packaged and incorporated into infectious virions that are readily propagated in cell cultures and in mice. The constructs used in this study, however, were significantly impaired in terms of efficiency of RNA replication and virion production. Nevertheless, we were able to use the bicistronic construct to study the functional role of the second transmembrane (TM2) region of the E protein separately from its other role as an internal signal sequence in the polyprotein. The use of sequential C-terminal deletions in this region revealed that TM2, independent of its role in the polyprotein precursor, is important for virion assembly. Furthermore, a construct in which the TM2 sequence was replaced by the corresponding element from another flavivirus, yellow fever virus, was severely impaired, suggesting a requirement for interaction of specific amino acids in TM2 with another viral component, most likely the TM1 element of the same protein.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cells and virus. BHK-21 cells were grown in Eagle's minimal essential medium (Sigma) supplemented with 5% fetal calf serum (FCS), 1% glutamine, and 0.5% neomycin (growth medium) and maintained in Eagle's minimal essential medium supplemented with 1% FCS, 1% glutamine, 0.5% neomycin, and 15 mM HEPES, pH 7.4 (maintenance medium).

Western subtype TBEV prototypic strain Neudoerfl or its derivative mutant R88 (9) was used as a control in all experiments. The biological properties of the wild-type (wt) strain Neudoerfl, including virulence, have been previously characterized in detail (31), and its complete genomic sequence is known (GenBank accession no. U27495). Mutant R88 carries a single point mutation which renders protein prM resistant to furin cleavage and therefore generates noninfectious immature virus particles but exhibits unaltered RNA replication and particle release (9, 21).

Plasmids and cloning procedures. All plasmids were derivatives of a previously described infectious cDNA clone system, which contains cDNA corresponding to the entire genome of Western subtype TBEV strain Neudoerfl (GenBank accession no. U27495) inserted into the vector pBR322 under the control of a T7 transcription promoter (30). One such derivative, plasmid pTNd/ME, encodes a TBEV replicon RNA lacking the structural proteins (pr)M and E and was described previously (11). The same replicon, but with an IRES-EGFP cassette replacing the variable region of the 3' noncoding region, can be transcribed from plasmid pTNd/ME-EGFP, which was characterized in a previous publication (12). The bc mutants analyzed in this study were constructed by replacing the EGFP gene of plasmid pTNd/ME-EGFP by the genes coding for proteins prM and E.

Various plasmids were generated to code for bc constructs that differed at the C-terminal end of the protein E gene in the second transmembrane region (E-TM2). Plasmid pTNd/bc contained an intact E-TM2, whereas pTNd/bc5, pTNd/bc10, and pTNd/bc24 had this region truncated at the carboxy-terminal end by 5, 10, and 24 codons, respectively. Finally, pTNd/bcYF was made by replacing the TBEV E-TM2 with the homologous region of the YF virus genome. RNAs transcribed from these plasmids and the resulting viral mutants are referred to as bc, bc5, bc10, bc24, and bc YF RNA or virus.

Plasmid pTNd/bc24 was constructed in two cloning steps. First, the prM and E coding region was assembled from two fragments and used to replace the EGFP reporter gene in the commercially available plasmid pIRES2-EGFP (Clontech) to put the prM/E genes under the control of the ECMV IRES contained in this plasmid. The first of these two fragments was generated by PCR and contained a BstXI restriction site at its 5' end, followed by an artificial ATG start codon placed in a sequence environment optimized for translation according to Kozak's rules, namely, that of two further artificial codons for the amino acids Val and Gly, which were then fused to the protein prM coding sequence, including the natural signal sequence that precedes it, starting with residue Asp-100 of the TBEV polyprotein and extending to a unique AgeI restriction site at position 960 of the TBEV genome near the junction of the prM and E protein coding regions. The primer sequences (termed p-BstXI and p-AgeI) used for the construction of this fragment are listed in Table 1. The second fragment was derived by digestion of the previously described plasmid SV-PE472 (2) with enzymes AgeI and NotI and comprised the entire coding region of protein E up to residue 472 (corresponding to the border between transmembrane regions 1 and 2) followed by an artificial stop codon. Then, the newly created IRES-prM/E cassette was excised, taking advantage of unique SacII and NotI sites, and was inserted in place of the IRES-EGFP cassette of plasmid pTNd/ME-EGFP, resulting in plasmid pTNd/bc24.

View this table: [in this window] [in a new window]

TABLE 1. Oligonucleotide primers for mutant construction

The other bc constructs were generated from pTNd/bc24 by swapping fragments coding for the carboxy-terminal region of protein E, taking advantage of unique sites SnaBI (at position 1878 corresponding to the wild-type TBEV genome numbering) and NotI (behind the artificial stop codon). The introduced fragments were generated by PCR and differed from the removed fragment by the presence of part or all of the TBEV E-TM2 or the corresponding YF E-TM2 coding regions followed in each case by an artificial stop codon. The PCR primers that were used are also summarized in Table 1. Sequences coding for partial or complete TBEV E-TM2 were "wobbled" (7) to avoid instability of the plasmids which may arise from homologous recombination with the second copy of the E-TM2 sequence in front of the NS1 gene, where it acts as a signal sequence for this protein (compare the structure of pTNd/ME as described by Gehrke et al.) (11).

As a replication-negative control, mutant NS5 was used as described in previous studies (19, 21). This mutant has a large deletion removing approximately one-fifth of the protein NS5 (the viral replicase) gene and part of the 3'-NCR. As another control, CME, a previously characterized replicon lacking most of the region coding for the structural proteins (11), was used. This replicon is fully competent for RNA replication but cannot produce or export viral particles.

All plasmids were amplified in Escherichia coli strain HB101, and new constructs were verified by sequence analysis of at least all of the regions that were modified or newly inserted in the course of the cloning procedures.

RNA manipulations. RNA was synthesized by in vitro transcription from various plasmids with T7 RNA polymerase (Ambion T7 Megascript transcription kit) as in previous studies (9, 20). Template DNA was degraded by incubation with DNaseI for 15 min at 37°C, and the RNA was purified using an RNeasy Mini kit (QIAGEN). The correct size and integrity of the RNA was checked on agarose gels as described previously (21). RNA concentrations were estimated from band intensities or, as specified in the text, measured spectrophotometrically (21). Equal amounts of RNA were then introduced into BHK-21 cells by electroporation, applying two subsequent pulses (setup values of 1.8 kV, 25 µF, and 200 ) with a Bio-Rad Gene Pulser as in previous studies (9, 12, 20).

Real-time PCR was performed using a pair of primers and a probe specific for a portion of the NS5 gene, as described in detail elsewhere (21).

Protein expression studies. Protein expression of RNA-transfected BHK-21 cells was analyzed qualitatively by immunofluorescence (IF) staining or quantitatively by fluorescence-activated cell sorter (FACS) analysis.

For IF analysis, RNA-transfected cells were seeded in 24-well cluster plates containing microscope coverslips or, in subsequent infection experiments, cells in such plates were inoculated with virus-containing supernatants. Intracellular protein expression was detected by indirect IF assay after fixation and permeabilization of cells using acetone-methanol (1:1). For specific detection of protein E or NS1, monoclonal antibodies directed against E (A5) (16) (dilution 1:50) or NS1 (6E11) (18) (dilution 1:75) were used. In passaging experiments, viral proteins were visualized using a polyclonal rabbit anti-TBEV serum which can be used for both structural and nonstructural protein detection (unpublished data). Staining was performed using a secondary fluorescein-isothiocyanate-conjugated anti-rabbit or anti-mouse antibody (Jackson ImmunoResearch Laboratories) as appropriate.

For FACS analysis, transfected cells were seeded into 25-cm² tissue culture flasks and collected from individual flasks at various time points after detachment with trypsin. Cells were then washed three times with phosphate-buffered saline (PBS) supplemented with 2% bovine serum albumin (BSA) and counted in a Casy 1 TT cell counter (Schärfe Systems). A total of 200,000 cells were fixed with lysing solution and permeabilized using Perm2 solution (Becton Dickinson) (diluted 1:10 with PBS prior to usage). Each incubation was followed by three washing steps with PBS containing 0.5% BSA. Each sample was divided into two halves (each containing 100,000 cells) and reacted with monoclonal antibodies directed against protein E (B2) (32) (dilution 1:2,000) as in a previous study or NS1 (6E11) (dilution 1:500) and then stained with fluorescein isothiocyanate-conjugated anti-mouse antibody (Jackson ImmunoResearch Laboratories).

For quantitative comparison of expression levels, FACS analysis was carried out with a FACSCalibur flow cytometer (Becton Dickinson) (15mW argon laser, 488 nm) with a 530/30 bandpass filter (FL-1). The mutual main cell population was gated, and 10,000 events per sample were analyzed. CellQuestPro software (Becton Dickinson) was used for calculation of FL-1 geometric mean values. To account for differences in RNA transfection efficiencies and virus infectivity, mean values were calculated for the 10% of the cell population that exhibited the brightest fluorescence intensity. Changing the percentage of cells used for calculation (such as the brightest 5% of the cells) yielded slightly different numbers but did not change the proportions between the individual samples.

RNA replication and export. Intracellular RNA replication was monitored by real-time PCR as described in a recent study (21). Briefly, equimolar amounts of purified in vitro-transcribed RNA were introduced into BHK-21 cells by electroporation. To remove noninternalized RNA, cells were washed four times with 20 ml growth medium supplemented with 5% FCS and then seeded into 25-cm² tissue culture flasks. Cells were collected at individual time points and counted, and cytoplasmic RNA was purified from defined numbers of cells and subjected to real-time PCR (PE Applied Biosystems) quantification.

To measure the export of RNA from transfected cells, cells were treated and intracellular RNA was measured in the same way as for RNA replication experiments, but in addition, aliquots of supernatants collected at the same time points were subjected to RNA quantification by real-time PCR. Supernatants were cleared by low-speed centrifugation to remove cell fragments containing nonspecifically attached RNA. Then, viral RNA was prepared using a QIAamp viral RNA Mini kit (QIAGEN) following the protocol provided by the manufacturer. Real-time PCR conditions were the same as for the quantification of intracellular RNA. A standard curve was prepared using a serial 10-fold dilution of an RNA preparation of a known concentration and the same medium and purification steps as for the samples. RNA concentrations determined from the aliquots of cell lysates and supernatants were used to calculate the total amount of intracellular and extracellular RNA present in each culture flask. Specific RNA export was then deduced from these numbers by calculating the percentage of the total RNA (intra- plus extracellular) that was in the extracellular fraction.

Particle analysis. BHK-21 cells were transfected with RNA and seeded into 20 tissue culture flasks (175 cm²) and maintained with growth medium containing 5% FCS, which was replaced by maintenance medium with 1% FCS 18 h after transfection. Supernatants from all of the flasks were collected 96 h posttransfection and precleared by low-speed centrifugation. Particles were pelleted by ultracentrifugation at 44,000 rpm for 2 h at 4°C, resuspended in 1.2 ml TAN buffer (0.05 M triethanolamine, 0.1 M NaCl [pH 8], 0.1% BSA), and applied to a 12 to 50% sucrose gradient for rate zonal fractionation. Sucrose gradients were centrifuged at 38,000 rpm for 3 h at 4°C and then fractionated into 0.6 ml samples with a Biocomp gradient fractionator. The presence of virions in each sample was quantified by real-time PCR using the purification and cycling conditions described above. The total amount of RNA in all of the gradient fractions was calculated, and the percentage of the total in each individual sample was determined.

Infectivity and neurovirulence assays. Infectivity was assessed both in cell culture and in suckling mice. For cell culture experiments, transfected or infected BHK-21 cells were maintained for 6 days. Then, supernatants were harvested and cleared of cell debris by low-speed centrifugation, and 200-µl aliquots were used without intermediate freezing to inoculate fresh cells in 24-well cluster plates. After 90 min, the inoculum was removed, and the cells were washed twice and then kept in maintenance medium. Intracellular expression of viral proteins was visualized 3 days later by indirect IF staining using a polyclonal rabbit serum that reacts with structural and nonstructural TBEV proteins.

To test infectivity in the most sensitive growth system available for TBEV, litters of suckling mice (0 to 1 days old) were inoculated intracranially with approximately 20 µl per mouse of supernatants from RNA-transfected cells. Mice were observed for 28 days for the development of disease symptoms. The presence of virus in the brains of diseased mice was verified by reverse transcription-PCR and sequence analysis. Serum samples were drawn from surviving mice and tested for seroconversion using a previously described antibody enzyme-linked immunosorbent assay (15).

For the comparison of neurovirulence properties, litters of suckling mice were inoculated with equal amounts of virus-containing supernatants and monitored as described above for infectivity determinations except that the inocula were first standardized to contain exactly equal numbers of virions on the basis of the amount of genomic RNA measured by real-time PCR.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Construction of bicistronic TBEV mutants. To investigate whether viable flavivirus particles can be made by providing prM and E from a separate cistron, the plasmid construct pTNd/bc was made by inserting the coding region for prM and E, under the control of the EMCV IRES, into the variable region of the TBEV 3'-NCR, which can be removed or modified without destroying the viability of the virus (31). This construct was analogous to one used previously for expressing EGFP from an IRES in the same position (12). This plasmid was used as a template for making a bicistronic TBEV genome by in vitro transcription. The resulting RNA transcript, called TBEV-bc (Fig. 1A), retained the genes encoding the C protein and all of the nonstructural proteins in the natural "cap cistron" but had the entire coding region for the full-length prM and E proteins moved into the inserted 3' "IRES cistron" instead of in its normal position.

View larger version (13K): [in this window] [in a new window]

FIG. 1. Bicistronic mutants derived from TBEV. (A) Schematic drawing of the wt and the bc TBEV genome (not drawn to scale). For the construction of TBEV-bc, the prM and E coding regions were deleted from their natural position within the long open reading frame and reinserted together with an heterologous EMCV-IRES element in place of the variable region of the 3'-NCR as indicated in the drawing. The transmembrane regions present in the structural proteins are indicated by black bars. Filled triangles indicate copies of the second transmembrane region of protein E (E-TM2), which in the wt genome simultaneously serves as a signal sequence for protein NS1. In the bc mutant, TM2 is present in two copies, once as a signal sequence for NS1 and once as part of the protein E anchor. To prevent genetic instability, the sequence of the second E-TM2 copy was wobbled, and the nucleotide and amino acid sequences of this altered element are shown below with the altered (wobbled) nucleotides of the synonymous codons underlined. (B) The three functional domains of E-TM2 (n, polar region, h, central hydrophobic region, c, domain required for signalase cleavage) are depicted. The amino acid sequence changes in the three truncation mutants and mutant bc YF, in which E-TM2 was replaced by the corresponding region from yellow fever virus, are shown below. Asterisks depict the residues in the YFV sequence that are different from those in the TBEV sequence.

The modified cap cistron now lacked all of the region encoding proteins prM (including its signal sequence, which is also involved in regulating the maturation of protein C) and E except for the second transmembrane region of protein E (E-TM2), which simultaneously serves as the signal sequence for NS1 and is therefore required for proper expression and membrane topology of NS1 as well as the rest of the polyprotein (28). Because this bifunctional sequence is duplicated in the bicistronic construct, the potential for genetic recombination between these two copies, if they were identical, could make the plasmid construct unstable in its bacterial host. Therefore, the codons in the 3' copy of TM2 (in the E protein of the IRES cistron) were "wobbled," i.e., replaced by synonymous codons with nucleotide changes at the third "wobble" position (7) as well as in other positions, where possible (Fig. 1A), to prevent homologous recombination. The plasmid construct was stable in E. coli strain HB101, and the sequence of the entire region corresponding to the bicistronic viral genome was confirmed.

Bicistronic expression of envelope and nonstructural proteins in BHK-21 cells. To test for expression of the envelope proteins from the IRES cistron, full-length, plus-strand genomic RNA from the bicistronic construct was made in vitro from the pTNd/bc plasmid DNA template and expressed in BHK-21 cells. A full-length infectious RNA encoding the natural, unmodified TBEV genome, transcribed from the cDNA plasmid clone pTNd/c (30), was used as a control. Expression of protein E (from the IRES cistron) and NS1 (from the cap cistron) was detected by immunofluorescence assay or FACS analysis using specific monoclonal antibodies against each of these proteins (Fig. 2A and B). Cells transfected with TBEV-bc expressed both E and NS1, demonstrating that both cistrons were translated and that the IRES was functional. However, NS1 levels from the bicistronic construct were approximately half of those of the wild type, whereas E protein expression was 10 to 100 times lower with the bc RNA than with the wild-type RNA at all time points. These data not only suggest a generally reduced efficiency in the bicistronic system but also show that the intracellular ratio of E to NS1 (and thus presumably of both surface proteins, prM and E, to all of the other viral proteins, C and the nonstructural proteins) was lower with the bicistronic genome RNA than with the wild-type genome RNA. This reduced relative level of protein E expression indicates that translation of the 3' IRES cistron is apparently less efficient in this construct than that of the cap cistron. Although IRES elements are generally strong translation initiation sites (3, 4, 13), it has been reported that in the context of bicistronic expression vectors, IRES translation is weaker than that of the Cap cistron (33). Our result now demonstrates that this is also the case in the context of a bicistronic flavivirus genome.

View larger version (17K): [in this window] [in a new window]

FIG. 2. Protein expression and RNA replication of TBEV-bc in BHK-21 cells. (A and B) Expression of protein E (A) or NS1 (B) was determined by immunostaining with specific monoclonal antibodies and FACS analysis from 1 to 5 days after transfection of BHK-21 cells with bc or wild-type RNA. The mean fluorescence intensity of the brightest 10% of each cell population, after subtracting the background value obtained for mock-transfected control cells (Mock), is plotted in arbitrary units (a.u.) on a logarithmic scale. Immunofluorescence pictures of cells 4 days after transfection are shown to the right. Intracellular RNA levels per cell for the bc mutant and its parental replicons ME and ME-EGFP, as well for as a replication-negative control (NS5), were determined at the indicated time points by quantitative PCR. RNA detected 2 h posttransfection corresponds to residual input RNA.

Expression of flavivirus proteins depends on both RNA replication and translation. Translation is required for the formation of replication complexes, and RNA replication increases the number of templates with which translation can occur. We therefore also compared the relative rates of replication of these RNA molecules at two early time points using a recently established intracellular replication assay (21). TBEV-bc was derived from the bicistronic expression vector ME-EGFP, which is a derivative of replicon ME (see Materials and Methods). In Fig. 2C, the intracellular RNA levels of the bc RNA 18 and 24 h posttransfection are shown in comparison to those obtained with the two parental constructs ME and ME-EGFP. Replication of ME-EGFP was approximately twofold less efficient than that of ME, indicating that insertion of the IRES-EGFP cassette somehow interfered with this process. An even stronger impairment was observed with the IRES-prM/E cassette in the bc construct (approximately eightfold reduction compared to that seen with ME), suggesting that IRES-driven translation of a longer cistron may conflict more with the replication process than that of a shorter gene.

Assembly and secretion of virus particles expressed from the bicistronic construct. To investigate whether cells transfected with TBEV-bc RNA were able to synthesize and secrete viral particles containing the packaged bicistronic genome, the amounts of genomic RNA in the transfected cells and in the cell supernatant were measured at different time points using the real-time PCR assay. A replicon lacking all of the viral structural proteins (CME) (11), and therefore unable to make virus particles, was used as a negative control to assess the level of nonspecific release of RNA from dying cells, etc. The monocistronic R88 RNA, which releases wild-type levels of viral particles but is incapable of causing secondary infections (9), was used as a positive control.

In Fig. 3, the number of RNA copies released each day after transfection is shown in panel A, and the ratio of extracellular to total (extracellular and intracellular) RNA is shown in panel B. Secretion of bc RNA was clearly detectable by 2 days after transfection and reached levels well above the background defined by CME, suggesting that packaging and export of this RNA had indeed occurred. However, the proportion of extracellular bc RNA remained almost 10 times lower than that of the wild-type control (Fig. 3B). Since the relative amount of envelope protein produced with the bicistronic construct was shown above to be reduced compared to that seen with the normal monocistronic construct, the observed lower rate of RNA export might be primarily attributable to a smaller pool of envelope protein molecules (prM-E heterodimers) available for virion assembly, resulting in a smaller fraction of the RNA getting packaged. In addition, the fact that prM and E were not provided in their natural gene order, with its concerted cleavage events at the C-prM junction, but instead from a separate cistron probably also contributed to this significantly reduced export efficiency of the viral RNA.

View larger version (12K): [in this window] [in a new window]

FIG. 3. RNA export kinetics of the bc mutant. (A) The concentration of viral RNA in the supernatants of BHK-21 cells transfected with bc RNA (open squares) or monocistronic R88 RNA (filled triangles) was monitored between 2 h and 4 days posttransfection. Values obtained for mock-transfected cells (Mock) were below the cutoff value 102 (dashed line) at all times. (B) The percentage of the total RNA (intra- plus extracellular RNA, both measured by real-time PCR) released from the cells was calculated at each time point using the same RNA samples whose results are shown in panel A as well as a control replicon (CME) from which all of the structural protein genes had been deleted and which was therefore unable to make virus particles (open circles). Mean values from two independent experiments (each consisting of double values) with error bars indicating standard deviations are shown. Values obtained 2 h posttransfection (marked by an asterisk) represent residual input RNA (removed by a subsequent medium change) rather than exported RNA.

To confirm that this extracellular viral RNA had actually been secreted in the form of viral particles, the cell culture supernatants were applied to sucrose density gradients to compare their sedimentation properties. After centrifugation, the gradient was fractionated by upward displacement, and the viral RNA in each fraction was quantified. The sedimentation profile in Fig. 4 confirms that the TBEV-bc RNA was secreted in the form of particles with a sedimentation behavior very similar to that seen with those made from the wild-type construct. These results indicate that viral assembly and packaging occur in this system, albeit with a much lower efficiency than that seen with wild-type virus, despite the changes made to the gene order and translation units.

View larger version (13K): [in this window] [in a new window]

FIG. 4. Sedimentation analysis of particles produced by the bc mutant (open squares) or wt virus (filled triangles). Particles harvested from supernatants of RNA-transfected cells and concentrated by ultracentrifugation were fractionated by rate zonal centrifugation on continuous 12 to 50% sucrose gradients. The amount of viral RNA in each 0.6-ml fraction was determined by quantitative PCR and is plotted as a percentage of the total RNA.

Infectivity of assembled virus particles. To test whether the viral particles synthesized from and carrying the bicistronic genome were infectious, BHK-21 cells were transfected with in vitro-synthesized RNA, and cleared supernatants collected 6 days later from these cells were used to infect fresh BHK-21 cells. This procedure was then repeated several times, each time using supernatant from the previous passage. At each passage, the cells were tested by IF assay for expression of viral proteins.

As shown in Fig. 5A, the supernatants of cells transfected with bc RNA, as with the wild-type control results, contained infectious material that could be further propagated through several passages, indicating that the particles containing the bicistronic genome are viable virions that are capable of entering cells and completing a new cycle of infection at each passage. However, the bc particles had a significantly lower specific infectivity than wt virus particles. The ratio of RNA copies to infectious particles was approximately 10² for the wt control but 10³ to 10⁴ for the bc virus.

View larger version (26K): [in this window] [in a new window]

FIG. 5. Infectivity of bc mutants in cell culture. BHK-21 cells were transfected with wt or bc RNA (or mock transfected) (A) and with various bc mutants (B) as indicated on the left. Supernatants were transferred to fresh cells 6 days later, and this passaging procedure was continued for a second round or several more cycles as indicated by arrows. Infection of cells was detected 3 days after transfection or infection by IF staining using a polyclonal serum that recognizes both structural and nonstructural proteins of TBEV. Results are shown for transfected cells and passages 1 and 2 as indicated at the top.

In these experiments, it was observed that the fluorescence intensity of cells infected with bc virus appeared to increase with increasing passage number, and the specific infectivity (as measured by the ratio of RNA copies to infectious particles) also increased to almost wild-type levels (data not shown), suggesting that adaptive mutations were being sequentially selected. Sequencing of the entire coding regions of both cistrons indicated no mutations at the amino acid level except for changes in protein E that probably increase the affinity for heparan sulfate (data not shown), as has been frequently observed previously with wild-type TBEV (25, 32). Preliminary data indicate, however, that some adaptive mutations may involve the sequence within the inserted IRES. The identity and functional significance of these compensatory mutations are the topics of a currently ongoing investigation.

The data obtained using BHK-21 cells show that TBEV-bc is apparently an impaired, but still viable and infectious, mutant. To further compare its biological properties with those of wild-type virus, its neurovirulence in newborn mice was investigated. Cell culture supernatants were collected from BHK-21 cells 6 days after transfection and cleared by low-speed centrifugation. RNA was quantified, and an amount of virus corresponding to 40,000 RNA molecules (presumably 40,000 virus particles) was used for each 20 µl mouse injection. Mice were then observed for 28 days after injection.

As shown in Fig. 6, the virus obtained using the bicistronic construct was infectious and virulent in the mice, killing two-thirds of them within 20 days. However, compared to the wild-type control results, the mice that were killed by the bc virus had a much longer mean survival time (12.1 ± 4.6 days versus 4.1 ± 0.3 days for wt virus), and several of the mice survived without seroconverting. This indicates that the bc particles were able to replicate and spread in vivo but were less neurovirulent than wild-type virus.

View larger version (11K): [in this window] [in a new window]

FIG. 6. Neurovirulence of wt virus and mutant bc upon intracranial inoculation of suckling mice. Supernatants of transfected cells were diluted to the same virus concentration (2,000 RNA molecules/µl). Survival of mice was recorded over a 28-day period.

Mutational analysis of the second transmembrane region of protein E. The physical separation of the NS1 and E genes into different cistrons in TBEV-bc provided an experimental system in which the functional importance of the TM2 region of protein E could be investigated by mutation without simultaneously disrupting its other function as the signal sequence for the NS1 protein. To test whether TM2 is still essential for virus particle formation under these conditions, a set of three deletion mutants was constructed (Fig. 1B). One of these, bc24, had all 24 amino acids of TM2 deleted, and the other two, bc5 and bc10, had 5 and 10 amino acids, respectively, deleted from the C-terminal end but retained the N-terminal portion of this transmembrane region.

As shown in Fig. 1B, the TM2 region has the essential features of a cleavable signal sequence (43, 44), including a central hydrophobic "h" region (11 amino acids), which is preceded by a more polar "n" region (7 amino acids) and followed by a "c" region (6 amino acids) required for signalase cleavage. Deletion of the five C-terminal amino acids (bc5) removed most of the c region, including its characteristic helix-breaking glycine residues as well as the alanine and valine residues at positions –1 and –3, respectively, that are important for recognition by signalase. However, this deletion left intact the central h region, the part of the signal sequence that is essential for targeting and membrane insertion (43). Deletion of 10 residues (bc10) resulted, in addition to the loss of the c region, in a truncation of the h region (from 11 to 7 residues) but nevertheless left more than the minimum of 6 residues reported to be essential for functionality (43). In bc24, the entire signal sequence (including all of n, h, and c) was deleted.

In addition to the three deletion mutants, a fourth mutant, bc YF, was constructed in which the entire TM2 region was replaced by the corresponding sequence from a different flavivirus, YFV. As shown in Fig. 1B, the YFV TM2 element is the same length as the one from TBEV, with a high degree of sequence conservation in the n and c regions but only two identical residues in the h region.

RNA was synthesized in vitro from each of these constructs as well as the plasmid encoding the full-length TBEV-bc and used for electroporation of BKH-21 cells. As shown in the left panels of Fig. 5B, cells transfected with each of these RNAs expressed viral proteins with apparently similar levels of efficiency, as judged from their fluorescence intensities. A FACS analysis of protein E and NS1 expression levels, like that presented in Fig. 2A and B, revealed no major differences between TBEV-bc and the deletion mutants either in terms of absolute fluorescence intensity or in the ratio of E to NS1 levels (data not shown). A quantitative PCR assay, like that shown in Fig. 2C, also did not reveal any major differences in RNA replication efficiency among the bicistronic constructs (data not shown). These short deletions in the TM2 region of the E protein therefore did not influence RNA replication or translation of either of the cistrons.

However, when supernatants from these constructs were transferred to fresh cells 6 days after transfection, only the mutant with the shortest deletion (bc5) was able to infect new cells, albeit with relatively low efficiency (Fig. 5B, middle panels). Furthermore, the bc5 virus did not get propagated further by sequential passage of supernatants every 6 days (Fig. 5B, right panels). Supernatants from cells transfected with the larger deletion mutants, bc10 and bc24, were completely unable to infect new cells even in the first passage. These data indicate that the presence of an intact TM2 element in the E protein is important for production of infectious virus particles. Mutant bc YF, which had an intact E-TM2 element but derived from a different but related flavivirus sequence, appeared to produce an extremely low level of infectious material in the first passage only and could not be propagated in further passages. This indicated that the presence of an intact TM2 element was necessary but not sufficient for infectivity and suggested that specific amino acid residues in the central h region, where the YFV E-TM2 is different from the TBEV sequence, are required for assembly of infectious TBEV virions.

The infectivity of these mutants was further investigated by injecting approximately equal volumes of BHK-21 cell supernatant, 6 days after RNA transfection, directly into baby mouse brain. Use of newborn mice has been shown previously to be about 100 times more sensitive than use of the cell culture assay for measuring infectivity of TBEV (31), so this was done to see whether any infectivity at all could be detected with the larger deletion mutants and bc YF. As summarized in Table 2, viral particles made using the TBEV-bc construct with an intact TM2 element killed all of the mice. (The higher mortality of TBEV-bc in this experiment compared to the experiment whose results are shown in Fig. 6 can be explained by an approximately 100 times higher virus inoculation dose based on subsequent RNA measurements). Consistent with the immunofluorescence results, the supernatant from cells transfected with mutant bc5 RNA killed 12 out of 14 mice, confirming that it contained infectious virus particles. In contrast, cell culture supernatants from BHK-21 cells transfected with bc10 or bc24 RNA did not appear to be infectious at all, and all of these mice survived without seroconversion. Interestingly, mutant bc YF, which had shown a very low level of infectivity in cell culture, did not infect any of the mice (no mice were killed and there was no seroconversion in any of the mice). In a second experiment, however, one of the mice injected with bc YF supernatant developed encephalitis and died. Sequence analysis confirmed that this mouse was indeed infected by mutant bc YF and that there were no adaptive mutations within the anchor region of protein E. This result indicated that bc YF indeed has a very low level of infectivity in both cell culture and mice.

View this table: [in this window] [in a new window]

TABLE 2. Infectivity of TBEV mutants in suckling mice

Requirement for TM2 for virus export. The experiments described above indicated that the presence of at least a partially intact TM2 region in protein E is important at some stage in the cycle of virus propagation in both cell culture and mouse brain, although deletions in this region did not seem to affect RNA replication or translation. To test whether the mutants with deletions in TM2 were able to make virus particles, the amount of viral RNA released into the supernatants of BHK-21 cells transfected with TBEV-bc, bc5, bc10, bc24, and bc YF RNA was quantified as described above. To normalize for differences in absolute expression levels, the amount of RNA in the secreted particles was divided by the total amount of RNA in the supernatant and cell lysate together, as was done in the previous secretion experiment. Figure 7 shows the percentage of the total RNA in the extracellular fraction on day 3 after transfection with each RNA. This experiment confirmed that cells transfected with TBEV-bc RNA, encoding an intact E protein, secreted RNA-containing virus particles, with extracellular RNA levels reaching almost 10% of the total. The efficiency of particle secretion was approximately three times lower with bc5, and very little extracellular RNA was observed with the larger deletions, bc10 and bc24, and with bc YF. These data fit very well with the infectivity experiments shown in Fig. 5B and Table 2 and provide evidence that deletion of even the C-terminal five amino acids of TM2 causes a significant impairment of the ability of TBEV to assemble into an infectious virion. This strongly suggests that, in addition to its important role as an internal signal sequence in the polyprotein, the TM2 region, as a functional component of the E protein, is also important for correct assembly of the virion. The strongly impaired functionality of bc YF further demonstrates that substitution of the corresponding TM2 element from another flavivirus is not sufficient for efficient assembly and suggests that interactions of specific amino acids in TM2 with other TBEV-specific elements (most likely TM1) play an essential role.

View larger version (8K): [in this window] [in a new window]

FIG. 7. Comparison of RNA export capacity of the different bicistronic mutants. Intracellular and extracellular RNA was quantified by real-time PCR 72 h after transfection of BHK-21 cells with RNA, and the calculated percentage of the total RNA in the cell supernatant is shown for each mutant. Values represent means from the results of two independent experiments (each consisting of double values) with error bars indicating standard deviations.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Viruses that use a single cistron to produce a polyprotein precursor have to adapt to a number of constraints imposed by this expression strategy, but the strategy also has certain advantages in terms of regulation of replication and assembly. One consequence of using a polyprotein is that all of the viral proteins are synthesized in equal molar ratios. This implies that these viruses either have adapted to using equal amounts of these proteins or have evolved mechanisms for accumulating necessary components at the right time and place and getting rid of ones that are in excess.

The polyprotein strategy also provides an opportunity for spatial and temporal coordination, since all of the structural and nonstructural proteins are synthesized from the same ribosome, and this would presumably facilitate the formation of replication and assembly complexes. In the case of the flaviviruses, this aspect appears to be especially important considering that budding is driven by the envelope proteins, which do not appear to make specific contact with the nucleocapsid and are capable of being released as empty subviral particles without first incorporating the nucleocapsid (1, 24). This problem is, at least in part, overcome by the fact that the junction between the C and prM proteins contains two sites for proteolytic processing—one for the viral protease and one for cellular signalase. Cleavage of the signalase site is delayed until the upstream site is cleaved by the viral protease (27, 29, 40), and this uncleaved precursor provides a temporary covalent connection between the envelope proteins on one side of the membrane and the capsid protein on the other and prevents premature budding of the envelope components. This has been demonstrated in studies in which this temporal regulation was disturbed by modifying the prM signal sequence to make cleavage by signalase independent of prior cleavage by the viral protease. In this situation, synthesis and release of subviral particles is favored at the expense of virion formation, resulting in a strong decrease in the amount of infectious virus released (27). For this reason, it is somewhat surprising that it is possible to obtain infectious virus particles using the bicistronic genome described here and the trans complementation system described earlier (11), because in these systems this level of coordination would be completely lacking due to the fact that the prM and E proteins are produced separately from the C protein. While these studies demonstrate that the presence of a covalent C-prM precursor is not an absolute requirement for virion assembly, it is nevertheless likely that this lack of temporal and spatial control is at least partially responsible for the low efficiency of TBEV-bc and its derivatives.

The artificial bicistronic construct was deficient in other ways as well. Comparison of the E and NS1 protein ratios in cells transfected with TBEV-bc RNA and the normal monocistronic RNA showed that envelope protein expression efficiency was 10 to 100 times lower when expressed from the foreign IRES in the 3'-NCR. This confirms an earlier observation by Mizuguchi et al. (33) that IRES elements, although usually very efficient in their natural context, are considerably less efficient when placed downstream of a cap cistron in a bicistronic vector as we have done here. The difference in translation efficiency of the cap and IRES cistrons resulted in a disruption of the normal stoichiometry of surface proteins, capsid proteins, and nonstructural proteins that otherwise arises naturally from using a single polyprotein precursor, and this unbalancing of the molar ratios of these components is probably also partly responsible for the lowered efficiency of packaging of RNA into infectious particles by TBEV-bc.

We also observed that the rate of replication of TBEV-bc RNA was low compared to that seen with the monocistronic control. This is most likely due to a disruption of RNA replication caused by the presence of the IRES in the 3'-NCR. This idea is supported by our observation that some adaptive mutations associated with improved growth and a higher specific infectivity in cell culture after multiple passages were in the IRES element itself, and we speculate that binding of ribosomes to the IRES-driven cistron might hinder replication of the viral RNA. It is also interesting that we have not observed any adaptive mutations in the capsid or surface proteins of TBEV-bc over multiple passages, except for commonly observed point mutations in the ectodomain of E protein that are related to increased affinity for heparan sulfate and are very frequently selected when wild-type or mutant TBEVes are grown in BHK-21 cells (25, 32). Moreover, sequence analysis after multiple passages did not provide any evidence for genetic instability or the accumulation of defective genomes as one might have expected for a bicistronic flavivirus genome.

The ability of our artificial bicistronic construct to make infectious particles and to complete multiple replication cycles allowed using it to investigate the function of the C-terminal membrane anchor region of the E protein by mutational analysis without disturbing the membrane topology and processing of the polyprotein encoding the capsid and nonstructural proteins. While earlier studies had indicated that the first transmembrane segment, TM1, is important for the formation of subviral particles and retention of the prM-E complex in the ER (2, 35, 36), it remained unclear whether the second transmembrane segment, TM2, actually has a function other than serving as a signal sequence for the NS1 protein, which immediately follows it in the polyprotein precursor (Fig. 1A) and is normally translocated into the lumen of the ER. Using a transient expression system for expression of TBEV prM and E, it was shown previously that a truncated E protein lacking TM2 was still able to assemble and secrete subviral particles, although these particles did appear to be less stable than ones containing the full-length E protein (2). In another study using a comparable expression system, it was shown that insertion of single alanine residues at various positions in the TM1 and TM2 regions of both prM and E of yellow fever virus severely impaired secretion of subviral particles and that some of these mutations also abolished infectious virus particle production when introduced into the viral genome (35).

The use of high-resolution cryoelectron microscopy made it possible to see the orientation of the transmembrane regions of both prM and E protein in intact mature dengue virus virions and showed that each forms a tight hairpin structure consisting of antiparallel alpha helices that are in contact with each other and are buried in the viral membrane (45). These anchor regions do not protrude into the interior of the virus particle, and there is no direct contact between the anchor elements of prM and E in the mature virion. These structural data suggested that these TM2 regions are not only signal sequences for the next protein in the translation unit but also form half of a defined structural element by which each of these proteins is anchored in the membrane. Whether having these structures in their intact form is a prerequisite for virion assembly or infectivity, however, was still not known.

We have now shown that removal of even the last five amino acids of the TM2 element of the TBEV E protein severely impairs the release of virus particles containing the artificial bicistronic genome and that larger deletions seem to abolish it altogether. This suggests that the double membrane anchor formed by TM1 and TM2 plays an essential role in flavivirus virion assembly. Computer calculations using the "DAS" transmembrane prediction server (www.sbc.su.se/miklos/DAS/) indicated that TM2 truncated by five residues still has a strong propensity for membrane integration, whereas removal of 10 residues more likely forces the remainder of TM2 to remain in the lumen of the virus where it may interfere with capsid interactions. However, mutant bc24, which entirely lacked TM2 and thus cannot extend to the viral interior, was also not infectious. Furthermore, TM2 could not be efficiently replaced by the corresponding functional element from YFV, although this element clearly had the same membrane topology as the natural TBEV-TM2. These observations suggest that TM2 is involved in specific amino acid interactions which are essential for infectivity. Although the most obvious explanation would be that specific TM1-TM2 interactions are required, it is also possible that interactions of TM2 with other components are important at earlier stages of assembly and/or maturation, before the mature virion is formed.

It is interesting that the membrane anchor topology of the flaviviruses differs significantly from that of another member of the family Flaviviridae, hepatitis C virus. Like the flavivirus prM and E proteins, the hepatitis C virus E1 and E2 proteins have two transmembrane segments at their C termini, the second of which serves as a signal sequence for the following protein. However, unlike those of the flavivirus proteins, their TM2 regions undergo a topological reorientation after signalase cleavage, with their C-terminal ends moving from the luminal to the cytoplasmic side of the membrane, and this change is important for E1E2 heterodimer formation, membrane anchoring, and ER retention (6, 42).

Alphaviruses, which are members of a different family of enveloped viruses (Togaviridae), have an envelope protein, E1, that is structurally and functionally related to the flavivirus E protein (38). The membrane anchors of the alphavirus surface proteins, however, span the membrane only once, and the virion is organized in part by direct and specific transmembrane contacts between the nucleocapsid and the E1E2 protein complexes (5, 41). Because the entry functions (receptor binding and membrane fusion) mediated by these proteins are very similar to those of flavivirus envelope proteins, it seems unlikely that the second membrane spanning plays an essential role in these functions. The differences are more likely related to differences in the mechanisms of assembly of these viruses.

Finally, the ability to use artificial bicistronic flavivirus genomes to make viable viruses is of interest for the development of attenuated strains. The TBE-bc construct used in this study showed reduced neurovirulence in mice compared to the wild-type monocistronic construct, but this might have been largely due to generally lower replication efficiency. The expression of E protein from the IRES in the 3'-NCR was rather inefficient with the particular construct used in this study, and the resulting alteration in the normal stoichiometry of viral proteins probably contributed to its reduced virulence in mice. The principle that expression from the second cistron could possibly be fine tuned to produce the desired ratio of envelope proteins (and immunogenic subviral particles) to nonstructural proteins and other components is one that deserves further consideration.

 
We are grateful to Rainer Gehrke for participation in mutant construction and Ursula Sinzinger and Gabriela Perstinger for technical help with FACS analysis. We thank Paul Breit for expert photographic documentation. We are especially grateful to Steve Allison for very helpful discussions and his valuable assistance during preparation of the manuscript.

This project was funded in part by the Austrian "Fonds zur Förderung der wissenschaftlichen Forschung" (FWF grant P17584-B14).

 
* Corresponding author. Mailing address: Clinical Institute of Virology, Medical University of Vienna, Kinderspitalgasse 15, A-1095 Vienna, Austria. Phone: 43-1-40490, ext. 79502. Fax: 43-1-40490-9795. E-mail: christian.mandl{at}meduniwien.ac.at.

Published ahead of print on 11 October 2006.

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. 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.[Abstract]
2
2. 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.[Abstract/Free Full Text]
3
3. Borman, A. M., J. L. Bailly, M. Girard, and K. M. Kean. 1995. Picornavirus internal ribosome entry segments: comparison of translation efficiency and the requirements for optimal internal initiation of translation in vitro. Nucleic Acids Res. 23:3656-3663.[Abstract/Free Full Text]
4
4. Borman, A. M., P. Le Mercier, M. Girard, and K. M. Kean. 1997. Comparison of picornaviral IRES-driven internal initiation of translation in cultured cells of different origins. Nucleic Acids Res. 25:925-932.[Abstract/Free Full Text]
5
5. Cheng, R. H., R. J. Kuhn, N. H. Olson, M. G. Rossmann, H. K. Choi, T. J. Smith, and T. S. Baker. 1995. Nucleocapsid and glycoprotein organization in an enveloped virus. Cell 80:621-630.[CrossRef][Medline]
6
6. Cocquerel, L., A. Op de Beeck, M. Lambot, J. Roussel, D. Delgrange, A. Pillez, C. Wychowski, F. Penin, and J. Dubuisson. 2002. Topological changes in the transmembrane domains of hepatitis C virus envelope glycoproteins. EMBO J. 21:2893-2902.[CrossRef][Medline]
7
7. Crick, F. H. 1966. Codon-anticodon pairing: the wobble hypothesis. J. Mol. Biol. 19:548-555.[Medline]
8
8. Elena, S. F., R. Miralles, J. M. Cuevas, P. E. Turner, and A. Moya. 2000. The two faces of mutation: extinction and adaptation in RNA viruses. IUBMB Life 49:5-9.[CrossRef][Medline]
9
9. Elshuber, S., S. L. Allison, F. X. Heinz, and C. W. Mandl. 2003. Cleavage of protein prM is necessary for infection of BHK-21 cells by tick-borne encephalitis virus. J. Gen. Virol. 84:183-191.[Abstract/Free Full Text]
10
10. Fonseca, B. A., S. Pincus, R. E. Shope, E. Paoletti, and P. W. Mason. 1994. Recombinant vaccinia viruses co-expressing dengue-1 glycoproteins prM and E induce neutralizing antibodies in mice. Vaccine 12:279-285.[CrossRef][Medline]
11
11. Gehrke, R., M. Ecker, S. W. Aberle, S. L. Allison, F. X. Heinz, and C. W. Mandl. 2003. Incorporation of tick-borne encephalitis virus replicons into virus-like particles by a packaging cell line. J. Virol. 77:8924-8933.[Abstract/Free Full Text]
12
12. Gehrke, R., F. X. Heinz, N. L. Davis, and C. W. Mandl. 2005. Heterologous gene expression by infectious and replicon vectors derived from tick-borne encephalitis virus and direct comparison of this flavivirus system with an alphavirus replicon. J. Gen. Virol. 86:1045-1053.[Abstract/Free Full Text]
13
13. Ghattas, I. R., J. R. Sanes, and J. E. Majors. 1991. The encephalomyocarditis virus internal ribosome entry site allows efficient coexpression of two genes from a recombinant provirus in cultured cells and in embryos. Mol. Cell. Biol. 11:5848-5859.[Abstract/Free Full Text]
14
14. Hayasaka, D., K. Yoshii, T. Ueki, T. Iwasaki, and I. Takashima. 2004. Sub-genomic replicons of tick-borne encephalitis virus. Arch. Virol. 149:1245-1256.[CrossRef][Medline]
15
15. Heinz, F. X., R. Berger, W. Tuma, and C. Kunz. 1983. A topological and functional model of epitopes on the structural glycoprotein of tick-borne encephalitis virus defined by monoclonal antibodies. Virology 126:525-537.[Medline]
16
16. Heinz, F. X., W. Tuma, F. Guirakhoo, and C. Kunz. 1986. A model study of the use of monoclonal antibodies in capture enzyme immunoassays for antigen quantification exploiting the epitope map of tick-borne encephalitis virus. J. Biol. Stand. 14:133-141.[CrossRef][Medline]
17
17. 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.[CrossRef][Medline]
18
18. Iacono-Connors, L. C., J. F. Smith, T. G. Ksiazek, C. L. Kelley, and C. S. Schmaljohn. 1996. Characterization of Langat virus antigenic determinants defined by monoclonal antibodies to E, NS1 and preM and identification of a protective, non-neutralizing preM-specific monoclonal antibody. Virus Res. 43:125-136.[CrossRef][Medline]
19
19. Kofler, R. M., J. H. Aberle, S. W. Aberle, S. L. Allison, F. X. Heinz, and C. W. Mandl. 2004. Mimicking live flavivirus immunization with a noninfectious RNA vaccine. Proc. Natl. Acad. Sci. USA 101:1951-1956.[Abstract/Free Full Text]
20
20. Kofler, R. M., F. X. Heinz, and C. W. Mandl. 2002. Capsid protein C of tick-borne encephalitis virus tolerates large internal deletions and is a favorable target for attenuation of virulence. J. Virol. 76:3534-3543.[Abstract/Free Full Text]
21
21. Kofler, R. M., V. M. Hoenninger, C. Thurner, and C. W. Mandl. 2006. Functional analysis of the tick-borne encephalitis virus cyclization elements indicates major differences between mosquito-borne and tick-borne flaviviruses. J. Virol. 80:4099-4113.[Abstract/Free Full Text]
22
22. Konishi, E., and A. Fujii. 2002. Dengue type 2 virus subviral extracellular particles produced by a stably transfected mammalian cell line and their evaluation for a subunit vaccine. Vaccine 20:1058-1067.[CrossRef][Medline]
23
23. 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.[Abstract/Free Full Text]
24
24. Konishi, E., and P. W. Mason. 1993. Proper maturation of the Japanese encephalitis virus envelope glycoprotein requires cosynthesis with the premembrane protein. J. Virol. 67:1672-1675.[Abstract/Free Full Text]
25
25. Kroschewski, H., S. L. Allison, F. X. Heinz, and C. W. Mandl. 2003. Role of heparan sulfate for attachment and entry of tick-borne encephalitis virus. Virology 308:92-100.[CrossRef][Medline]
26
26. Kuhn, R. J., W. Zhang, M. G. Rossmann, S. V. Pletnev, J. Corver, E. Lenches, C. T. Jones, S. Mukhopadhyay, P. R. Chipman, E. G. Strauss, T. S. Baker, and J. H. Strauss. 2002. Structure of dengue virus: implications for flavivirus organization, maturation, and fusion. Cell 108:717-725.[CrossRef][Medline]
27
27. Lee, E., C. E. Stocks, S. M. Amberg, C. M. Rice, and M. Lobigs. 2000. Mutagenesis of the signal sequence of yellow fever virus prM protein: enhancement of signalase cleavage in vitro is lethal for virus production. J. Virol. 74:24-32.[Abstract/Free Full Text]
28
28. Lindenbach, B. D., and C. M. Rice. 2001. Flaviviridae: the viruses and their replication, p. 991-1041. In D. M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
29
29. Lobigs, M., and E. Lee. 2004. Inefficient signalase cleavage promotes efficient nucleocapsid incorporation into budding flavivirus membranes. J. Virol. 78:178-186.[Abstract/Free Full Text]
30
30. Mandl, C. W., M. Ecker, H. Holzmann, C. Kunz, and F. X. Heinz. 1997. Infectious cDNA clones of tick-borne encephalitis virus European subtype prototypic strain Neudoerfl and high virulence strain Hypr. J. Gen. Virol. 78(Pt. 5):1049-1057.[Abstract]
31
31. Mandl, C. W., H. Holzmann, T. Meixner, S. Rauscher, P. F. Stadler, S. L. Allison, and F. X. Heinz. 1998. Spontaneous and engineered deletions in the 3' noncoding region of tick-borne encephalitis virus: construction of highly attenuated mutants of a flavivirus. J. Virol. 72:2132-2140.[Abstract/Free Full Text]
32
32. Mandl, C. W., H. Kroschewski, S. L. Allison, R. Kofler, H. Holzmann, T. Meixner, and F. X. Heinz. 2001. Adaptation of tick-borne encephalitis virus to BHK-21 cells results in the formation of multiple heparan sulfate binding sites in the envelope protein and attenuation in vivo. J. Virol. 75:5627-5637.[Abstract/Free Full Text]
33
33. Mizuguchi, H., Z. Xu, A. Ishii-Watabe, E. Uchida, and T. Hayakawa. 2000. IRES-dependent second gene expression is significantly lower than cap-dependent first gene expression in a bicistronic vector. Mol. Ther. 1:376-382.[CrossRef][Medline]
34
34. Mukhopadhyay, S., R. J. Kuhn, and M. G. Rossmann. 2005. A structural perspective of the flavivirus life cycle. Nat. Rev. Microbiol. 3:13-22.[CrossRef][Medline]
35
35. 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.[Medline]
36
36. Op De Beeck, A., Y. Rouille, M. Caron, S. Duvet, and J. Dubuisson. 2004. The transmembrane domains of the prM and E proteins of yellow fever virus are endoplasmic reticulum localization signals. J. Virol. 78:12591-12602.[Abstract/Free Full Text]
37
37. Pugachev, K. V., F. Guirakhoo, S. W. Ocran, F. Mitchell, M. Parsons, C. Penal, S. Girakhoo, S. O. Pougatcheva, J. Arroyo, D. W. Trent, and T. P. Monath. 2004. High fidelity of yellow fever virus RNA polymerase. J. Virol. 78:1032-1038.[Abstract/Free Full Text]
38
38. 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.[CrossRef][Medline]
39
39. Stadler, K., S. L. Allison, J. Schalich, and F. X. Heinz. 1997. Proteolytic activation of tick-borne encephalitis virus by furin. J. Virol. 71:8475-8481.[Abstract]
40
40. Stocks, C. E., and M. Lobigs. 1998. Signal peptidase cleavage at the flavivirus C-prM junction: dependence on the viral NS2B-3 protease for efficient processing requires determinants in C, the signal peptide, and prM. J. Virol. 72:2141-2149.[Abstract/Free Full Text]
41
41. Vogel, R. H., S. W. Provencher, C. H. von Bonsdorff, M. Adrian, and J. Dubochet. 1986. Envelope structure of Semliki Forest virus reconstructed from cryo-electron micrographs. Nature 320:533-535.[CrossRef][Medline]
42
42. Voisset, C., and J. Dubuisson. 2004. Functional hepatitis C virus envelope glycoproteins. Biol. Cell 96:413-420.[CrossRef][Medline]
43
43. von Heijne, G. 1998. Life and death of a signal peptide. Nature 396:111-113.[CrossRef][Medline]
44
44. von Heijne, G. 1990. The signal peptide. J. Membr. Biol. 115:195-201.[CrossRef][Medline]
45
45. 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.[CrossRef][Medline]
46
46. 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.[CrossRef][Medline]
47
47. Zhang, Y., W. Zhang, S. Ogata, D. Clements, J. H. Strauss, T. S. Baker, R. J. Kuhn, and M. G. Rossmann. 2004. Conformational changes of the flavivirus E glycoprotein. Structure 12:1607-1618.[Medline]

---

Journal of Virology, December 2006, p. 12197-12208, Vol. 80, No. 24
0022-538X/06/$08.00+0     doi:10.1128/JVI.01540-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Mansfield, K. L., Johnson, N., Phipps, L. P., Stephenson, J. R., Fooks, A. R., Solomon, T. (2009). Tick-borne encephalitis virus - a review of an emerging zoonosis. J. Gen. Virol. 90: 1781-1794 [Abstract] [Full Text]  
• Sultana, H., Foellmer, H. G., Neelakanta, G., Oliphant, T., Engle, M., Ledizet, M., Krishnan, M. N., Bonafe, N., Anthony, K. G., Marasco, W. A., Kaplan, P., Montgomery, R. R., Diamond, M. S., Koski, R. A., Fikrig, E. (2009). Fusion Loop Peptide of the West Nile Virus Envelope Protein Is Essential for Pathogenesis and Is Recognized by a Therapeutic Cross-Reactive Human Monoclonal Antibody. J. Immunol. 183: 650-660 [Abstract] [Full Text]  
• Schlick, P., Taucher, C., Schittl, B., Tran, J. L., Kofler, R. M., Schueler, W., von Gabain, A., Meinke, A., Mandl, C. W. (2009). Helices {alpha}2 and {alpha}3 of West Nile Virus Capsid Protein Are Dispensable for Assembly of Infectious Virions. J. Virol. 83: 5581-5591 [Abstract] [Full Text]  
• Schrauf, S., Schlick, P., Skern, T., Mandl, C. W. (2008). Functional Analysis of Potential Carboxy-Terminal Cleavage Sites of Tick-Borne Encephalitis Virus Capsid Protein. J. Virol. 82: 2218-2229 [Abstract] [Full Text]  
• Orlinger, K. K., Kofler, R. M., Heinz, F. X., Hoenninger, V. M., Mandl, C. W. (2007). Selection and Analysis of Mutations in an Encephalomyocarditis Virus Internal Ribosome Entry Site That Improve the Efficiency of a Bicistronic Flavivirus Construct. J. Virol. 81: 12619-12629 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Orlinger, K. K. Articles by Mandl, C. W. Search for Related Content PubMed PubMed Citation Articles by Orlinger, K. K. Articles by Mandl, C. W.